Magnum Opus of Sanskrit

Source: pdf

01. Introduction – Importance of Pāṇini

The अष्टाध्यायी of पाणिनि (5th century B.C.) is the most refined and sophisticated grammar of all the ancient and modern languages of the world 12345. Accepting its importance even the western scholars have called it in express terms ‘one of the greatest monuments of human intelligence’ 6, ‘a wonderful specimen of human intelligence’ 7, ‘a notable manifestation of human intelligence’ 8, ‘one of the greatest productions of the human mind’ 9 and ‘an important invention of human intelligence’ 1011. The algebraic formulation 12 of Pāṇini’s rules was not appreciated by the first western scholars. They regarded the work as abstruse 13, ambiguous and in the highest degree obscure 14 or artificial 15; its each aphorism more dark and mystic than the darkest and most mystical of oracles 16, pregnant with endless progeny of interpretations and commentaries sometimes as obscure as the original 17; a system with a network of mysticism 18; and the order of the sūtras as illogical and impracticable for any one to learn Sanskrit by its means 19. But the western critique was muted and eventually turned into praise when modern schools of linguistics developed sophisticated notation systems of their own 202122232425262728. Today some scholars go upto the extent of saying that ‘the Aṣṭādhyāyī is not a Sanskrit grammar, it is a work on general linguistics’29.

Acquaintance with the Pāṇinian analysis of root and suffixes and his recognition of ablaut — though only indirect via Charles Wilkins’ (c.1750-1836) Sanskrit Grammar (1808) — inspired Franz Bopp (1791-1867) and others to develop the imposing structure of Indo-European comparative and historical linguistics 30. Otto Böhtlingk (1815-1904) was so much impressed by Pāṇinian method that he wrote the grammar of the Turkic Yakut language of Siberia 31, in which the working out of phonology, morphology, and syntax is detailed and complete in a very Pāṇini-like fashion, and which Edward Sapir used repeatedly to praise as a model of excellent method 32. The Russian scholar E. E. Obermiller (1901-1936), who was profoundly impressed by, and full of admiration for the greatest Indian linguist, knew by heart almost every sūtra and cherished for some time a scheme of writing the grammar of the Russian language according to the grammatical sūtra method of Pāṇini 33.

02. Additions to Pāṇini

In view of the importance and usefulness of Pāṇini’s monumental work, attempts have been made almost right from the time of Pāṇini to revise and improve it. The famous Paribhāṣāsūcana of Vyāḍi, a contemporary of Pāṇini 3435, contains also some supplementary rules called vārttikas 36. The great commentary called Mahābhāṣya of Patañjali (c.150 BC) preserves about 4280 vārttikas 3738 of his predecessors Kātyāyana (c.350 BC), Bhāradvājīyas, Saunāgas, Ślokavārttikakāra etc. The two grammatical saints, Kātyāyana and Patañjali, scrutinised every doubtful word of the sūtras 39. In addition to the vārttikas of his predecessors, Patañjali gave some extra supplementary rules called iṣṭis. In the 3983 sūtras 40 (excluding the 14 śivasūtras) of the Aṣṭādhyāyī, 270 sūtras deal with Vedic morphology and 334 with the accent. In the Kielhorn edition of Mahābhāṣya, 1713 sūtras have been discussed by Patañjali — 1245 with Kātyāyana’s vārttikas and the remaining 468 with śloka-vārttikas and bhāṣya or only with bhāṣya; 1153 sūtras have been cited, 532 without discussion; 1760 sūtras have been given in the footnotes by Kielhorn, out of which 516 sūtras are neither discussed nor cited by Patañjali; in case of 1222 sūtras, Kielhorn has not indicated in the footnote 41. Later in the commentary called Kāśikā, Jayāditya and Vāmana (6th century AD), gave 917 vārttikas 42 (excluding those having explanatory nature in the Mahābhāṣya), out of which 78 vārttikas 43 are extra, not found in the Mahābhāṣya, and 138 iṣṭis 44.

03. Revised Editions of Pāṇini

3.1 Cāndra-vyākaraṇa

— the first great revised edition of Pāṇini45

Candragomin (c.5th century AD) wrote his grammar called after him as Cāndra-Vyākaraṇa, embodying all the suggestions and corrections of Kātyāyana and Patañjali. Dr K C Chatterji calls it the first great revised edition of Pāṇini 45. This grammar had both laukika and Vedic parts in which the Vedic morphology was covered in ch.VII and the accent rules in ch.VIII — this is evident from its commentary 46. But at present only the first six chapters have come down to us. Many other systems of grammar came up in the first millenium of the Christian era but only the laukika portion was given a place in them.

3.2 Sarasvatīkaṇṭhābharaṇa

— the second great revised edition of Pāṇini 47

In 11th century AD Bhojadeva wrote a monumental grammatical work called Sarasvatīkaṇṭhābharaṇa 48 in eight chapters, each chapter in turn containing four quarters or pādas like the Aṣṭādhyāyī of Pāṇini. The Sarasvatīkaṇṭhābharaṇa (SKĀ) is the biggest grammar of the sūtra style containing 6432 sutras (see TABLE-I in section 5) in all. It is the only extant complete revised edition of Pāṇini in which the rules of accent and the Vedic morphology are also taken into account. It embodies all the suggestions and corrections of Vyāḍi, Kātyāyana, Patañjali, Jayāditya, Vāmana and the post-Pāṇinian grammarians such as Śarvavarman, the author of Kātantra grammar, Candragomin and (Jain) Śākaṭāyana. The vārttikas, iṣṭis, uṇādi-sūtras, gaṇa-pāṭha, gaṇa-sūtras, paribhāṣās and phiṭ-sūtras have been incorporated by Bhoja in the body of the sūtra-pāṭha itself. Thus, the SKĀ can safely be placed next to the Aṣṭādhyāyī of Pāṇini 49. The work itself was not known, except through citations, till about the first quarter of the 20th century 50 (S. K. Belvalkar does not make any mention of this work in his 1915 edition of Systems of Sanskrit Grammar), when manuscripts of the work were discovered, all in Mālābār 51.

In the SKĀ, unlike Aṣṭādhyāyī, all the saṁjñā sūtras are kept in the first pāda of the first chapter, while the second pāda contains all the paribhāṣā-sūtras. The uṇādi-sūtras are placed in the proper order in the kr̥t-prakaraṇa itself and form the first three pādas of ch. II. The first two pādas of ch. VIII contain the sūtras related to the Vedic morphology, while the accent rules of both laukika and Vedic Sanskrit are placed in the last two pādas. The phiṭ-sūtras are incorporated in the grammar of Bhoja in the svara-prakaraṇa itself under the sūtras VIII.3.109-196. All the words of gaṇa-pāṭha are enumerated in the form of sūtras only in the respective places.

The strategy behind the preparation by Bhojadeva of the complete revised edition of Pāṇini is expressed in a very brief and compact form in one sentence by the Prastāvanā writer, K S Mahādeva Śāstrī, thus:

“लोकवेदोभयानुग्राहकत्वाद् आर्षत्वाद् बहुभिर् आदृतत्वाद् व्याख्या-शतैर् उपबृंहितत्वाच् च सर्वतोमुखीं प्रतिष्ठाम् आस्थाय प्रावर्तमाने ऽस्मिन् पाणिनीये तन्त्रे
वार्त्तिक-गणपाठादि-सापेक्षतया ऽध्येतॄणाम् अतीव क्लेशजालं पश्यन्न्
आचार्य-देशीयो भोजदेवः बहु-ग्रन्थालोडनम् अन्तरा एकेनैव ग्रन्थेनाधीतेन कृत्स्नस्यापि व्याकरण-शास्त्रस्य लघुनोपायेन प्रतिपत्ताव् अभ्युपायं चिन्तयन्
प्रायेण पाणिनिम् एवानुरुन्धानः
तत्र तत्र चान्द्र-कातन्त्रादि-गतान् अप्य् अर्थान्
यावद्-अपेक्षं सञ्चित्य
वार्त्तिक-गणोणादि-परिभाषापाठ-फिट्-सूत्रादिकं निखिलम् अपि संगृह्य
पाणिन्य्-अन्-अनुशिष्टानां तत्-तत्-समय-समुच्चितानां महाकविभिर् अन्यैश् च साधुतया प्रयुज्यमानानां शब्दानाम् अप्य् अनुशासनं विदधत्
सरस्वतीकण्ठाभरणाख्यं ग्रन्थ-रत्नं विनिर्ममे ।” 52+++(5)+++

Bhojadeva appears to have made great efforts towards making grammar easy for the readers. This point is expressed by Mahādeva Śāstrī thus:

“इहहि पाणिनीयं तन्त्रम् अधिजिगांसमानानां महान् क्लेशः सम्पद्यते । तथाहि -
क्वचिद् ज्ञापकेन, क्वचिद् योगविभागेन, कुत्रचिद् व्याख्यानेन, क्वचन पदानाम् अनुकर्षेण, क्वचिच् च मण्डूकप्लुत्या, क्वचिच् च परिगणनेन, एकत्र भाष्यार्थ-परिशीलनेन, इतरत्र वार्त्तिक-परिचिन्तनेन, अपरत्र च गण-पाठादि-शीलनेन ते तेऽर्थाः साधनीया दृश्यन्ते ।
तानेतान् क्लेशान् परिजिहीर्षुर् भोजदेवः गण-पठितान् शब्दान्, परिभाषाः, ज्ञापक-भाष्येष्टि-वार्त्तिक-व्याख्यागम्यान् अर्थांश् च
सूत्र-रूपेण ग्रन्थ-शरीर एव समायोजयत् ।
क्वचित् तत्र तत्र विप्रकीर्णानां सूत्राणाम् एकत्रैव समावेशेन,
क्वचिच् च प्रकरणादिविनिमयेन,
सूत्राणां प्रदेश-विनिमयेन, पौर्वापर्य-विपरिणामादिना च
महती सौकर्य-सम्पत्तिः सम्पादिता ।”53+++(5)+++

Many sūtras in Pāṇini grammar are interpreted through jñāpaka. This is especially the case regarding the optional dīrghatva of aṣṭan, when a vibhakti follows (this point has been discussed by Mahādeva Śāstrī in his Prasatāvanā, p.10). Patañjali rejects some of the Kātyāyana vārttikas on the ground that the same are indicated by the wordings of Pāṇini’s other sūtras. Bhoja has included such implied meanings of jñāpakas in his sūtras itself. Two examples are presented here.

Consider first P.2.3.13 चतुर्थी सम्प्रदाने. The Kātyāyana vārttika no.1 says: चतुर्थी-विधाने तादर्थ्ये उपसंख्यानम्. The Bhāṣyakāra says: “चतुर्थी-विधाने तादर्थ्ये उपसंख्यानं कर्तव्यम् । यूपाय दारु । कुण्डलाय हिरण्यम् ॥ … … तत् तर्हि वक्तव्यम् । न वक्तव्यम् । आचार्यप्रवृत्तिर्ज्ञापयति भवत्यर्थ-शब्देन योगे चतुर्थीति यदयं चतुर्थी तदर्थार्थ (२.१.३६) इति चतुर्थ्यन्तस्यार्थ-शब्देन सह समासं शास्ति॥” Bhoja includes this implied meaning of jñāpaka when along with the sūtra चतुर्थी सम्प्रदाने (2.3.13), he also gives तादर्थ्ये (3.1.234).
On the sūtra सार्वधातुके यक् (P.3.1.67), vārttika 3 says: भावकर्मणोर्यग्विधाने कर्मकर्तर्युपसंख्यानं कर्तव्यम्. On this the Bhāṣyakāra remarks: “भावकर्मणोर्यग्विधाने कर्मकर्तर्युपसंख्यानं कर्तव्यम् । पच्यते स्वयमेव । पठ्यते स्वयमेव । … अथवाचार्य-प्रवृत्तिर्ज्ञापयति भवति कर्मकर्तरि यग् इति यदयं न दुह-स्नु-नमां यक्-चिणौ (३.१.८९) इति यक्-चिणोः प्रतिषेधं शास्ति ।” Bhoja includes the implied meaning of this jñāpaka and gives the sūtra कर्मकर्तरि च (१.३.१०४) after the sūtra सार्वधातुके यक् (३.१.६७).

A list of such sūtras along with corresponding Pāṇini sūtras is given in TABLE-I. A good discussion of all the peculiarities of Bhoja’s SKĀ can be found in the Prastāvanā (in Sanskrit) of K S Mahādeva Śāstrī , Vol. IV of Trivandrum Edition.

TABLE-I : स॰कं॰ सूत्रs based on ज्ञापक सूत्रs in पाणिनि

स॰कं॰ सूत्र (Madras edition) पाणिनि सूत्र (+vtt.) ज्ञापक सूत्र in पाणिनि
1.3.104 3.1.67 v.3 3.1.89
2.4.113-114 3.3.90 v.1-2 3.2.117
3.1.234 2.3.13 v.1 2.1.36
4.1.21 4.1.87 v.2 6.3.34
6.4.53 7.2.84 6.1.172; 7.1.21
7.1.53 3.1.30 v.1 न कम्यमिचमाम् (ग॰सू॰)
7.1.91 7.3.45 v.1 5.4.39
7.2.31, 33 1.1.56 v.21 8.2.35
7.3.63 8.2.7 v.2 8.2.68

04. Commentaries on SKĀ

There are three incomplete commentaries on SKĀ – the first one is the Hr̥dayahāriṇī by Daṇḍanātha Nārāyaṇa Bhaṭṭa (either contemporary of Bhoja or 12th c. AD) and extends only up to ch.VII. Not only the commentary, but also some sūtras are missing in between (see TABLE-IV & V in the next section). The other commentary called Puruṣakāra is by Kr̥ṣṇalīlāśukamuni (c.12th century AD), the author of the commentary Puruṣakāra on the dhātupāṭha called Daivam. M. Krishnamacharya in his ‘History of Classical Sanskrit Literature’ mentions his name twice 54 and says that in Kr̥ṣṇalīlāśuka’s commentary on Bhoja’s grammatical treatise, Sarasvatīkaṇṭhābharaṇa, Pāṇini’s verses are quoted freely as illustrations. ‘A Catalogue of Skt MSS’, edited by T Gaṇapati Śāstrī (1921), Vol. VI, mentions at p.6, No.35 the commentary Puruṣakāra (sūtra-vyākhyā-rūpaò) of size 2200 granthas. Probably this commentary is related to Bhoja’s SKĀ. The New Catalogus Catalogurum, Vol IX, mentions Puruṣakāra as the commentary on Daivam only. The third commentary in Gujarati on ch. VIII by Dr N M Kansara has been written only recently (see the last para in the next section).

05. Editions of SKĀ

The complete sūtra-pāṭha of SKĀ, edited by T R Chintamani, was published in the Madras University Sanskrit Series No.11 in 1937 with a Foreword by C Kunhan Raja and Preface by the editor himself giving a brief sketch of the life and works of Bhojadeva. The Madras edition is based on the following MSS (The details of the MSS have not been given in any of the published works on SKĀ. Based on the Descriptive Catalogues on Skt MSS in different libraries, these details have been compiled by the present author and are separately given at the end of this article) — (i) R. No. 3279 (ii) R. No. 4179 (iii) No. 698 (iv) a transcript of the SKĀ sūtras, prepared from the original palm leaf Ms (v) a fragment of the commentary of Daṇḍanātha, the last two supplied to the editor by his friend M Ramakrishnakavi. The Madras edition suffers from a lot of errors, mostly misprints. Some of the errors have arisen due to limitations of the editor in understanding Bhoja sūtras. This edition contains 6432 sūtras in all. The pāda-wise break-up is shown in TABLE-II.

TABLE-II: Number of sūtras in the Madras Edition of SKĀ

Ch. –> 1 2 3 4 5 6 7 8
पाद 1 212 348 304 211 181 175 140 173
पाद 2 135 255 161 149 224 180 151 149
पाद 3 231 191 148 269 156 183 150 263
पाद 4 277 283 135 207 189 192 178 232
Total-> 855 1077 748 836 750 730 619 817

The available Hr̥dayahāriṇī commentary of Daṇḍanātha has been published upto first six chapters in the Trivandrum Sanskrit Series Nos. 117, 127, 140 and 154, edited by K Sāmbaśiva Śāstrī, Vol. I (ch. I, 1935), Vol. II (ch.II, 1937), Vol. III (ch. III-IV, 1938) and V A Rāmaswāmi Śāstrī, Vol. IV (ch. V-VI, 1948). The Trivandrum edition is based on the following MSS — (i) No. 806 (ii) No. 817 (iii) No. 557 (iv) R. No. 3279 (v) the Ms containing the sūtra-pāṭha alone obtained from Brahmasri Venkitarāma Śāstrī, Shenkotta. The commentary on ch. VII is still unpublished. There is considerable variance in readings of the sūtras in the two editions. Upto ch. VI, the number of sūtras in the Trivandrum edition is 5019, as against 4996 in the Madras edition. The pāda-wise break-up is shown in TABLE-III.

TABLE-III: Number of sūtras in the Trivandrum Edition of SKĀ

Ch. –> 1 2 3 4 5 6
पाद 1 211 348 307 212 182 175
पाद 2 134 255 164 152 227 185
पाद 3 230 190 150 272 157 181
पाद 4 278 285 135 208 189 192
Total –> 853 1078 756 844 755 733

There is a difference also in the number of pratyāhāra-sūtras in the two editions. The Madras edition has 14, i.e. same as in the Aṣṭādhyāyī. But the Trivandrum edition has only 13, as in Cāndra-vyākaraṇa. The two sūtras ‘ह य व र ट् । ल ण् ।’ are combined into one as ‘ह य व र ल ण् ।’ With ṭakāra only one pratyāhāra viz. aṭ is used in grammar. The aṭ pratyāhāra does not occur in any Bhoja-sūtras. Therefore, the reading without the ṭakāra viz. ‘ह य व र ल ण् ।’ appears to be the correct one. The aṭ occurs in four Pāṇini sūtras: आतोऽटि नित्यम् (८.३.३), दीर्घादटि समानपादे (८.३.९), अट्-कुप्वाङ्-नुम्व्यवायेऽपि (८.४.२) and शश्छोऽटि (८.४.६३). The Bhoja-sūtras corresponding to P.8.3.9, 8.4.2 and 8.4.63 are: दीर्घादटि समानपादे (८.२.१२८), चु-टु-तु-ल-शर्-व्यवायेऽपि (७.४.१४५) mentioning the letters the intervention of which prevents ṇatva and चयश्शश्छोऽमि (७.४.१७६), which is based on the Kātyāyana’s vārttika: छत्वममि तच्छ्मश्रुणा तच्छलोकेनेति प्रयोजनम्. Bhoja has rejected the Pāṇini sūtra आतोऽटि नित्यम्, following the statement found in the Kāśikāvr̥tti : केचिदनुस्वारमधीयते, which indicates that the anunāsikatva is not nitya.

The Gujarati commentary of Dr Narayan M Kansara on ch. VIII has been published by Rashtriya Veda Vidya Pratishthan and Motilal Banarsidass, New Delhi, 1992. The editor deserves praise for bringing out the so far unavailable commentary on this chapter. This edition is based on the Madras edition only. No available MSS have been consulted. The variations from the Ms are not supplied. The Madras edition itself does not give the original readings of many sūtras. That is why the Gujarati edition also suffers from inaccuracy in case of many sūtras. In case of a few sūtras, the vr̥tti requires to be rewritten. Moreover, many correct sūtras of the Madras edition have been blunderingly modified in this edition. For example, the Bhoja-sūtra हियासामः (८.३.११) makes provision for the udāttatva of हि (the substitute for सि in लोट्), यास् (= यासुट्, the आगम in case of लिङ्) and आम् of अनडुह्. In the Gujarati edition this sūtra has been changed to हिमारण्याभ्याम्.

06. Review of the first three chapters by Louis Renou (1957)

Renou has given an excellent review of the first three chapters of the Trivandrum edition of SKĀ in his ‘Le Sarasvatīkaṇṭhābharaṇa’, pp. 121-127 in Vol. III of ‘Études Vediques et Pāṇineennes’ (in French). According to him, the SKĀ is much more than a revised edition of Pāṇini (beaucoup plus qu’une edition “revue” de P.). First of all, it has incorporated in the frame of sūtras all the positive teachings (tout l’enseignement positif) of the old vārttikas, which are more or less preserved in the Kāśikā : casting aside the vārttikas which are of technical, scholastic and explanatory character. Secondly, the new additions in the form of इति वक्तव्यम्, इत्युपसंख्यानम् take the form of sūtra. Every effort has been made to remove the limitations of Pāṇini, which makes this manual that the Sanskrit tradition has bequeathed to us, more complete 55. He adds in the footnote that it contains some rare

TABLE-IV: Sūtras without Daṇḍanātha commentary

--------------------------------------------------- Ch.पाद सूत्र सं॰ सूत्र सं॰ in Total in TSS Madras Edition --------------------------------------------------- 1 1 10 10 1

3 6-17 6-18 12(13) --------------------------------------------------- 2 4 84a, 98a, 98b, 83, 98, 99, 199 4 200

3 1 20, 27-41, 19, 26-40, 220-275 73(72) 223-279

2 16-20, 80, 16-20, 80, 83-114 41(38) 83-117

3 102-113, 142 102-113, 140 13

4 24, 12# 24, 12# 1 + 1 --------------------------------------------------- 4 2 46 45a 1

3 224-226, 205#, 223a-223c (fn), 3 + 2 248# 205a, 245 --------------------------------------------------- 5 2 77-96, 76# 77-93, 76# 20(17)+1 4 161-167 161-167 7 --------------------------------------------------- 6 2 153, 154-155, 151a(fn), 16(14) 164-176 151b(fn), 160-171 3 4, 26, 127, 3a, 25-27, 128, 16(19) 150-162 151-164 4 39-52, 72#, 39-52, 72#, 112# 14 + 2 112# --------------------------------------------------- **7 1 52, 101 Same 2 2 90, 105, 106, Same 3 + 1 25# 4 66, 67, Same 13 158-167, 173 ---------------------------------------------------

The symbol # after the sītra number indicates that only one or two words of commentary are available. The letters a & b following a sītra number indicate that the sītra is not available in the corresponding edition but if also followed by (fn), indicate that the sītra is available in the foot note.

Vedic examples (naturally in the commentary only) – I.1.118 रक्षामाकिर्नो अघशंस ईशत (ऋ॰ ६.७५.१०); 119 अधैनं वृका रभसासो अद्युः (ऋ॰ १०.९५.१४) ; विश्वकर्मा विमना आद् विहायाः (यजु॰ १७.२६) etc 56. Regarding paribhāṣā-sūtras he says, not less interesting is the 2nd pāda which condenses all the paribhāṣās taken from Pāṇini as well as vārttikas and bhāṣya. There are some paribhāṣās which are not attested by Sīradeva, e.g. आदिष्टादचः पूर्वस्य (49). One also finds the maxims of Bhāṣya which have not been brought under special treatment, such as नानिष्टार्था शास्त्रप्रवृत्तिः (123). A few others make appearance unprecedented, such as the last but one, अभिधानलक्षणाः कृत्-तद्धित-समासाः (133) which summarises a point of view often expressed by the commentators 57.

Commenting on the series of exceptional sūtras in between III.2.62-68, 73-88, related to composition (samāsa), he says, “This shows the considerable part of novelties, which the grammar of Bhoja provides. Novelties in the sense of relative term, because the vr̥tti of

TABLE-V: Sūtras with incomplete Daṇḍanātha commentary (*)

Ch. पाद सूत्र सं॰ (TSS) सूत्र सं॰ (Madras)
1 1 113m, 128m 113, 128
1 2 9e 9
1 3 18i 19
2 1 277m 277
2 3 95m 96
2 4 61m, 98e, 98bi, 59a, 97, 99, 183, 282
3 1 61e, 62i, 115m, 120m, 212m, 280i 60, 61, 114, 119,209, 276
3 2 6m, 10i, 41m, 43m,82e, 118i, 125m,150i 6, 10, 41, 43, 82,115, 122, 147
3 3 11i, 101e, 114i,141e, 143im 11,101,114,139,141
3 4 1m, 7m, 10me, 11e,14m,16m,18ie, 19i,20e, 21ie,23m,25m,54e, 55i same as in TSS
4 1 7i, 64e, 65i, 66i,193i, 197i 7, 64, 65, 66, 193,196a(fn)
4 2 47i 45b
4 3 49e,50i,55i,121m,247e 49, 50, 55-56, 122,244
4 4 9e,26m,154e,155i 9, 26, 154, 154a

--------------------------------------------------- 5 1 88e, 89i, 133m 88, 88a, 131

2 97i, 108e, 112m, 94, 105, 109, 164, 167m, 174i, 178i 171(misprinted 271), 175

3 149e, 150i 148, 149

4 10m, 160e, 168i 10, 160, 168 --------------------------------------------------- 6 1 13e,13a(fn)i,27m, 12, 13, 27, 45, 92, 45i(ह्वः),92m,109m, 109, 172 172im

2 20m, 72e, 73i, 20, 72, 73, 74, 74e(possibly),75i, 74a(fn), 112, 151, 113m, 152e, 178e 173

3 9m, 11m, 19m, 22m, 8, 10, 18, 21, 33, 32m(possibly),64m, 65,76,77,103, 105, 75i,76m,102m,104e, 106, 113, 126, 127, 105i, 112m, 129, 165 125e(misprinted 122), 126i, 128m, 163i

4 38e, 113i 38, 113 --------------------------------------------------- **7 1 8m, 46i, 56m, 75m, Same 83e, 84i, 123m

2 14m,97m,138e,139i Same

3 4m, 27m,44i,84m, Same 149m

4 70i,157e,168i,175m, Same 177m ---------------------------------------------------

(*) Letters i, m & e following the sutra numbers indicate respectively the initial, middle & end portion of the commentary as missing.

** - Available only in manuscript without sītra numbers, sītra numbers indicated are taken from Madras edition.

Candragomin, much earlier to the SKĀ, gives in the long commentary on the sūtra चार्थे (II.2.48) most of the examples cited in the (Daṇḍanātha)vr̥tti or those resulting from the teachings of the sūtras of Bhoja.” 58

07. Article of Robert Birwé (1964)

An excellent article entitled ‘Nārāyaṇa Daṇḍanātha’s Commentary on Rules III.2.106-121 of Bhoja’s Sarasvatīkaṇṭhābharaṇa’ of Robert Birwé has appeared in the Journal of American Oriental Society, Vol. 84, 1964, pp.150-162. The article is mainly meant for providing an alternative to the missing commentary on many rules of the sūtras under the above title. Nārāyaṇa Bhaṭṭa quotes in his प्रक्रियासर्वस्व, a commentary on Pāṇini’s grammar, numerous rules from Bhoja’s SKĀ, among them the rules Bh III.2.106-121. Besides Nārāyaṇa Bhaṭṭa’s commentary there is another one – the Kāśikā-manuscript No. 2440 of the India Office Library (cf. J Eggeling, Catalogue of the Skt MSS in the Library of India Office, Vol. II, p.159, No. 991, 992 = Ind. Off. Lib. Nos. 2440, 2441. The Ms, in Devanāgarī characters, has been written between 1630-1632 AD) on Rule P II.1.72: मयूरव्यंसकादयश्च. It is a unique fact that this Ms comments a Pāṇinian sūtra by quoting rules from a non-Pāṇinian grammar and a commentary thereon. Since this commentary is, moreover, partly missing in the Trivandrum edition of Bhoja, it deserves to be published.

First quoting, on pp. 153-155, the vr̥tti of the above Ms, Dr Birwé discusses in section 11, p.155, the possible author of this Ms. In his opinion it must be Daṇḍanātha. In section 12, pp.155-158, he has attempted to edit and reconstruct the missing portions of Daṇḍanātha’s vr̥tti on Bhoja’s rules III.2.106-121. In sections 13-17, pp.159-162, he has discussed the date of Daṇḍanātha. With the help of a comparative study of the Daṇḍanātha’s commentary on Bh.I.3.196 and the Hemacandra’s auto-commentary called Br̥hadvr̥tti on Haima sūtra V.1.52, Birwé concludes that Hemacandra must have copied Daṇḍanātha’s commentary. With this he leaves a scope of further research – a systematic search for parallel passages in the works of Hemacandra and Daṇḍanātha. According to Dr. Birwé, the year 1100 A.D. is the upper limit of Daṇḍanātha’s date (p. 161).

08. Articles of Dr N M Kansara (1989, 1990)

The article entitled ‘The Vaidika Vyākaraṇa of Bhojadeva’ by Dr Kansara was published in the Journal of the Oriental Institute, Baroda, Vol. 38, Nos.3-4,1989, pp.309-313. In this paper he has given an outline of the contents of ch.VIII of Bhoja’s grammar based on the Madras edition. At the end he mentions the possible scope of research work regarding search for the correct readings of the sūtras of this chapter with the help of the parallel sūtras in Pāṇini.

A second article entitled ‘Emendations Essential to the Vedic Grammar of Bhojadeva’ by the same author was presented to the AIOC, 35th Session, Hardwar, 1990, pp.43-53. In most of the sūtras discussed by him, he has modified the Bhoja sūtras by adding many words for the sake of anuvr̥tti. In his Gujarati edition of SKĀ, he has emended many more sūtras by adding certain words to them. In my opinion such additions, except in very few cases, are unwarranted. Consider, for example, the Bhoja sūtras ल्यप् क्त्वा वा (VIII.2.67) and प्लुत उदात्तः (VIII.4.225). In the Gujarati edition they are changed to समासेऽनञ्पूर्वे ल्यपः क्त्वा वा (VIII.2.68) and वाक्यस्य टेः प्लुत उदात्तः (VIII.4.228). The sūtra ल्यपः क्त्वा वा simply means that in case of Vedic literature the suffix क्त्वा is optionally applicable to the case where the suffix ल्यप् has been prescribed in the laukika section. Here the context does not demand where the suffix ल्यप् is used. This provision has already been made in the laukika part under the sūtra भाविनि गत्युपपदसमासे क्त्वो ल्यप् (VI.4.6). The Bhoja sūtra VIII.4.225 makes provision for udāttatva to pluta. This sūtra does not require any statement regarding which अच् is pluta. This part is already covered under the sūtra वाक्यस्य टेः प्लुतः (VII.3.131). Thus modifying the Bhoja sūtra प्लुत उदात्तः to वाक्यस्य टेः प्लुत उदात्तः is not only unnecessary but also wrong.

09. Ph.D. Thesis on SKĀ by K Neelakaṇṭham (1989)

A Ph.D. thesis entitled ‘A Comparative Study of Pāṇini’s Aṣṭādhyāyī and Bhoja’s Sarasvatīkaṇṭhābharṇam’ by K Neelakaṇṭham was submitted to Osmania University, Hyderabad, in 1989. For the comparative study, the order of Aṣṭādhyāyī has been followed except in saṁjñā and paribhāṣā sūtras. This thesis of 6+718+iv pages gives all the novelties in SKĀ with exact number of vārttikas included as sūtras, added and deleted words in gaṇas and added and deleted uṇādi suffixes. Thus it is a good contribution to the study of SKĀ. However, it must be noted that except in a few cases, no attempt has been made to correct the corrupted sūtras of SKĀ. Comparison of the two grammars has been done only for the selected sūtras and not for all the sūtras of Pāṇini. Moreover, in case of many sūtras of Pāṇini which contain the accent part also, such as चतुरनडुहोरामुदात्तः (P VII.1.98), he has given the parallel sūtras from SKĀ only for the morphological part, completely neglecting the accent part of SKĀ. For P.VII.1.98, the parallel sūtra in SKĀ is हियासामः (VIII.3.11), as mentioned earlier. At p.323, he says “SKĀ has सायं-चिरं-प्राह्णे-प्रगेऽव्ययेभ्यष्ट्युः (4.3.109) as against सायं-चिरं-प्राह्णे-प्रगेऽव्ययेभ्यष्ट्यु-ट्युलौ तुट् च (4.3.23) of Aṣṭ. Bhoja following CG omits the affix ट्युल्.” He does not look in SKĀ for the svara-siddhi of the affix ट्यु corresponding to Pāṇini’s ट्युल्. In SKĀ, it will suffice to provide an optional लित्-स्वर to ट्यु. This gives us a clue for correcting the highly corrupted sūtra रद्यौ वा [?] 8.3.71 in the Madras edition of SKĀ. It must be ट्यौ वा. The Gujarati edition gives रथोर्वा (8.3.74), which is obviously not correct.

Many conclusions indicated in the thesis are wrong. For example, at p.39, Dr Neelakaṇṭham says, “षष्ठी स्थाने योगा (1.1.49) of P. is not found in SKĀ…. Bhoja does not seek the help of this sūtra and only implies this rule in the sūtra like इको यणचि (6.1.77)”. Actually based on the Bhāṣya on P.1.1.49, Bhoja has accepted the second alternative to this sūtra, viz. निर्दिश्यमानस्यादेशाः (I.2.38) where षष्ठ्या comes as anuvr̥tti from the sūtra I.2.30. On p.661 he says that ‘the P.VIII.4.66-68 are not found in SKĀ’. But actually corresponding to P.VIII.4.66-67, we have the Bhoja sūtras उदात्तादनुदात्तः and उदात्त-स्वरित-परश्च वा यणः (VIII.4.229-230). The latter sūtra is wrongly printed in the Madras edition. In the Ms also यणः has been wrongly included in sūtra no. 230. Actually it forms part of the next sūtra which is supposed to be parallel to P.VIII.2.4. At p.149, he says that ‘the अनुदात्तौ सुप्पितौ (P.III.1.3) is not found in SKĀ’. This topic has been discussed in detail in the Februry 1999 issue of वेदवाणी by the present author.

10. Ph.D. Thesis on SKĀ by V G Śāstrī (1989)

A Ph.D. thesis entitled ‘Bhojadevakr̥ta Sarasvatīkaṇṭhābharaṇa (Ek Samīkṣātmak Adhyayan)’ by V G Śāstrī was submitted to Gujarat University, Ahmedabad, in 1989. It has been published by Parimal Publications, Delhi, 1996. The main contribution of this thesis lies in its first chapter, which contains the discussion on the life and works of Bhojadeva. Out of the total 287 pp., 120 pp. have been allotted to the description of contributions of other grammarians! As per the title and his Foreword, this thesis is supposed to be the critical study of SKĀ, especially on ch.VIII. One hardly finds in the published thesis any significant contribution to the study of SKĀ. What has been expressed by Dr Chintamani in the tabular form in Bhoja-Pāṇini Concordance in Appendix-I, has been expressed in oft-repeated words in ch.6 of this thesis. As per Dr N M Kansara’s information, all the corrections in the sūtra-pāṭha with the help of parallel sūtras in Pāṇini, done by V G Śāstrī (but not compiled separately in the thesis), have been incorporated in the Gujarati edition.

11. SKĀ cited as Authority in Various Works

The great depositaries of learning such as Hemacandra, Kṣīrasvāmin, Vardhamāna, Mallinātha Sūri, Devarāja Yajvan, Sāyaṇācārya, Nārāyaṇa Bhaṭṭa and Bhaṭṭoji Dīkṣita have cited in their works the SKĀ as authority. This topic has been described in detail by the present author in the May 1995 issue of वेदवाणी. Some details not supplied in this issue are given below.

Vardhamāna has quoted Bhojadeva by name as a lexicographer, grammarian and commentator in the following kārikas of गणरत्नमहोदधि (edited by Julius Eggeling, 1879; Reprint MLBD, New Delhi, 1963):
2(2) 7(2) 13 22 25
27 28 29 31 39
40 45 46 49 52
64 80 82 85 91(2)
93 94 95(2) 96(3) 101
105 107 109 112 113
114 123(2) 131 135 139
140(2) 145 146 157 164(2)
170 175 182 186 191
197 202(2) 203 207 209(2)
211 212(3) 214 215 216
219(4) 222 228(2) 229 230
233(2) 240 243 245 249
261 268 272(2) 286 288
302 304 314(2) 315 317
326 330 342 348 351
352 353 359 392(4) 403(2)
404 405 406 409 414(2)
417 419 420 423 426
433 434 435 437(2) 438
446 449 450 452 457(2)

Total 1 (in कारिका 2)
+ 129 (in the commentary)

Devarāja Yajvan quotes Bhoja in his निघण्टु-निर्वचनम् (edited by Sudyumnācārya and published by Ram Lall Kapoor Trust, Bahalgarh, 1998) at the following pages:
By name:
3 13 14 19 20
24 26 31 39(2) 49(2)
52 55 67 83 92
102 117 122 123 127
128(2) 129 131 135 137
138 139 150 178(2) 207
210 216

Total 36 times

SKĀ sūtras quoted without mentioning by name:
41 144 152 154 157
179 183 186 188(2) 189
201 202 204 206(2) 208
209 210(2) 211 214 216
230 237 244 246 274
276 279 281 282 284(3)
285(2) 288 290 291(2) 301
302 308 310 311(2) 322

--- Total 48 times

दण्डनाथवृत्ति (by name) : 152 198 202

On page 210 under the commentary on ‘दुरोणे’ the Bhoja-sūtra (2.2.184) is mentioned along with its vr̥tti. But this vr̥tti is found otherwise in Trivandrum Edition.
Scholars like Bhimsen Śāstrī and Hariścandramaṇi Tripāṭhī have made good use of the SKĀ and its commentary in their theses entitled ‘न्यास-पर्यालोचन’ (in Hindi) and ‘निपातार्थनिर्णयः’ respectively, published by Bhaimi Prakashan, New Delhi, 1979 and Sampurnanand Sanskrit Vishvavidyalaya, 1991.

12. Ākṣepas against SKĀ

Certain modern scholars like K V Abhyankar and Nemicandra Śāstrī have put up ākṣepas against this monumental work. K V Abhyankar says in ‘A Dictionary of Sanskrit Grammar’ –

“ By the anxiety of the author to bring together, the necessary portions of the gaṇa-pāṭha and the paribhāṣās, which the author has included in his eight chapters, the book instead of being easy to understand, has lost the element of brevity and become tedious for reading. Hence it is not studied widely.”

-- It appears that reading the Pāṇinian grammar scattered in different books like the Aṣṭādhyāyī with Kāśikā and Mahābhāṣya, gaṇa-pāṭha, Uṇādi-pāṭha, phiṭ-sūtras and paribhāṣās with their commentaries and sub-commentaries was not tedious for him !

In “Ācārya Hemacandra aur unkā śabdānuśāsan’(1963, in Hindi), pp.101-102, Nemicandra Śāstrī remarks :

(i) The Bhoja grammar is heavily loaded with paribhāṣā. It can be said in express terms that the above-mentioned grammar cannot be understood without the knowledge of Pāṇinian grammar. Only an expert in Pāṇini can understand it well. Regarding paribhāṣā it appears essential that the knowledge of the Pāṇini grammar be gained first…. Bhojarāja has collected all the paribhāṣā-sūtras found in Paribhāṣenduśekhara. Due to this an initial complexity appears in this work.

With the above remarks Nemicandra Śāstrī has only strengthened the statement that the SKĀ is nothing but the great revised edition of Pāṇini and supports also the view of Robert Birwé who says in his article mentioned earlier: “I doubt very much that the understanding of the SKĀ is impossible without the vr̥tti, as maintained by K. Sāmbaśiva Śāstrī. There is, I am convinced of it, no serious difficulty, not to speak of impossibility, to understand it for anyone fully conversant with Pāṇini’s Aṣṭādhyāyī and the main works of his followers”(p.152). It is always preferable to have all the information implied by the sūtrakāra at one place. Nemicandra Śāstrī has expressed the view that Hemacandra has kept his grammar free from the burden of paribhāṣā. But in order to fully understand the grammar of Hemacandra, Hemahaṁsagaṇi has compiled 141 paribhāṣās, the largest number of paribhāṣā in Sanskrit grammar ! Out of these, 57 paribhāṣās have been cited by Hemacandra himself in his auto-commentary Br̥hadvr̥tti.

(ii) Bhojarāja’s treatment of the feminine affixes is very complicated.

  • Nemicandra Śāstrī has compared the SKĀ with the brevity of Haima grammar in which the accent part has been neglected. Naturally the affixes ङीष्, ङीप् and ङीन् are changed to a single affix ङी. That cannot happen in case of a work like the Aṣṭādhyāyī and SKĀ where accent is also taken into account. He shows happiness over the fact that Hema has only one sūtra ‘अजादेः’ regarding the feminine affix, whereas Bhoja has constructed 13 sūtras for the same affix टाप्, III.4.2-14. It is to be noted that Bhoja had to construct so many sūtras as exceptions to the affix ङीष्.

13. Articles on SKĀ by the Present Author

Ten articles on SKĀ by the present author have been published so far in Vedavāṇī (July 1993 - June 2000), mostly dealing with the correction of some sūtras from ch. VIII whose readings are found incorrect in both the editions. This has been possible with the help of the “Concordance of Nine Sanskrit Grammars” prepared by the author especially for this purpose. In this Concordance, Pāṇini sūtra and vārttika numbers have been given in the first column in serial order and the corresponding sūtra numbers of the remaining eight grammars – Bhoja, Kātantra, Cāndra, Jainendra (mahāvr̥tti), Jainendra (laghuvr̥tti or śabdārṇava-candrikā), Śākaṭāyana, Haima and Malayagiri – in the following columns.

14. Details of the MSS on the Sarasvatīkaṇṭhābharaṇa (grammar)

I. K. Sāmbaśiva Śāstrī (1938):

“A Descriptive Catalogue of the Skt MSS in H. H. the Maharajah’s Palace Library, Trivandrum", Vol III - Vedānta, Mīmāṁsā, Vyākaraṇa, Nyāya and Jyotiṣa
(1) No. 806 Sarasvatīkaṇṭhābharaṇam
Palm leaf, 11" x 1½", 6 leaves, 11 lines per page, 40 letters per line, Grantha script, and size - 150 granthas.
The Ms contains only sūtra-pāṭha and concludes with the 2nd pāda in the 1st chapter.
(2) No. 817 Sarasvatīkaṇṭhābharaṇam (savr̥ttikam)
Palm leaf, 22⅜" x 1⅞", 330 leaves, 9-11 lines per page, 88 letters per line, Malayalam script, size - 18000 granthas.
The Ms contains the Hr̥dayahāriṇī commentary of Daṇḍanātha Nārāyaṇa Bhaṭṭa upto the end of ch. VII.

II. K. Sāmbaśiva Śāstrī (1939):

“A Descriptive Catalogue of the Skt MSS in the Curator’s Office Library, Trivandrum”, Vol III – Vedānta, Mīmāṁsā and Vyākaraṇa.
(3) No. 556 Sarasvatīkaṇṭhābharaṇam
Paper, 13" x 8", 112 pages, 24 lines per page, and 24 letters per line, Devanāgarī script, and size – 2000 granthas.
The Ms contains only sūtra-pāṭha. Complete.
(4) No. 557 Sarasvatīkaṇṭhābharaṇam (savr̥ttikam)
Palm leaf, 11⅜" x 1⅜", 64 leaves, 7-8 lines per page, 28-36 letters per line, Malayalam script, size – 960 granthas.
The Ms contains the Hr̥dayahāriṇī commentary of Daṇḍanātha Nārāyaṇa Bhaṭṭa. It begins on a portion of the 1st pāda of the 2nd ch. and breaks off on the 4th pāda of the same chapter. The last 18 leaves are blank.
(5) No. 558 Sarasvatīkaṇṭhābharaṇam (savr̥ttikam)
Paper, 13¼" x 8⅜", 1067 pages, 14 lines per page, 20 letters per line, Devanāgarī script, size – 9000 granthas.
The Ms contains the Hr̥dayahāriṇī commentary of Daṇḍanātha.. It extends from the beginning of ch. V. upto the end of ch. VII.

III. S. Kuppuswami Śāstrī (1928):

“A Triennial Catalogue of MSS collected during the Triennium 1919-20 to 1921-22 for the Govt Oriental MSS Library, Madras”, Vol IV, Part I, Sanskrit B.
(6) R. No. 3279 Sarasvatīkaṇṭhābharaṇam
Paper, 10⅝" x 9¼", 62 foll., 20 lines per page, 28-33 letters per line, Devanāgarī script. Good. Transcribed in 1920-21 from a Ms of M R Ry K C Valiyarāja of Kottakal, Malabar District. Foll. 61b and 62 are left blank.
The Ms contains only sūtra-pāṭha. Complete.
(7) R. No. 4179 Sarasvatīkaṇṭhābharaṇa-vyākhyā – Hr̥dayahāriṇī
Paper, 10¾" x 9¾", 298 foll., 20 lines per page, Devanāgarī script. Good. Transcribed in 1922-23 from a Ms of Mahāmahopādhyāya Nārāyaṇa Nambudiripad Avargal, Kunnanukulam, Cochin State. Fol. 297 b contains the name of the owner of the original Ms. Fol. 298 is left blank.
The Ms contains ch. I to III complete and breaks off in ch. IV.

IV. V. Krishnamacharya (1947):

“A Descriptive Catalogue of Skt MSS in the Adyar Library, Madras”, Vol VI – Grammar, Prosody and Lexicography.
(8) No. 697 (38.I.7) Sarasvatīkaṇṭhābharaṇam – Bhojadevakr̥tam
Paper, 8½" x 6⅝", 170 foll., 14 lines per page, Devanāgarī script. Good. Good writing.
The Ms contains only sūtra-pāṭha. Complete in 8 chapters. It is a copy of R. No. 3279 prepared by K. Ramachandra Sharma, Adyar.
(9) No. 698 (39.I.8) Sarasvatīkaṇṭhābharaṇa-vyākhyā – Hr̥dayahāriṇī
Paper, 8" x 6½", 1247 foll., 14 lines per page, Devanāgarī script. New. Good. Good writing.
Bound in 7 volumes or 8 parts on the whole, the volume 3 containing 2 parts. The pagination in parts 1 to 3 is continuous as 1 to 1043 and in parts 4 to 8, as 1 to 1449. The Ms contains the commentary up to the end of ch. VII.

REMARKS:

  1. The Revised Catalogue of the Palace Granthappura (Library), Trivandrum, edited by K Sāmbaśiva Śāstrī (1929) at p.35 mentions palm leaf MSS Nos. 806 and 817. It appears that these same MSS are described in the Descriptive Catalogue of 1938. However, in the case of No. 806 the character is mentioned to be Malayalam and no. of granthas 175. In the case of No. 817, the no. of granthas is mentioned as 20000.
  2. The Annual Report on the Administration of the Dept for the Publication of Skt MSS for 1104 ME, Travancore State, mentions on p.16, No.62 Sarasvatīkaṇṭhābharaṇa-vr̥tti of Daṇḍanātha. This appears to be same as described under No. 557 earlier. The Ms belongs to Brahmadattan Nambudiripad, Plakkattiri, Koodalloramana, Pattambi. However, the size is mentioned to be 1200 granthas.
  3. A Catalogue of Skt MSS, edited by Gaṇapati Śāstrī (1923), Trivandrum, Vol III, p.9, No.53 mentions a Ms of Sarasvatīkaṇṭhābharaṇa (size 2200 granthas) which was prepared from the Ms obtained from the Govt Oriental MSS Library, Madras. The latter appears to be R. No. 3279, which was copied to Ms bearing No. 556 mentioned earlier.

Acknowledgement

The author expresses his deep gratitude to Prof E R Rama Bai, Head of the Sanskrit Dept, University of Madras, for making available the manuscript numbers on SKĀ from different libraries. The author is thankful to Dr S D Laddu, Curator, Bhandarkar Oriental Research Institute, Pune, for his suggestion to write to the Editor, New Catalogus Catalogurum, University of Madras, in regard to the details of MSS on SKĀ. The author is also thankful to Dr P G Lalye, who made available for reference a copy of the unpublished thesis of Dr K Neelakaṇṭham. Dr Narayan Kansara, Ex-Director of Maharshi Academy of Vedic Sciences, Ahmedabad, deserves special thanks, who provided the required information on SKĀ through correspondence from time to time.
This paper is dedicated to late Mahāmahopādhyāya Pt. Yudhiṣṭthir Mīmāṁsak who has been a great inspiration to the author regarding svādhyāya (self-study) and who created interest in research on SKĀ, the result of which is the state-of-art in this field, presented for the first time, especially the details of MSS, which would be of great help to the future researchers on SKĀ.

NOTES

* The abridged version of this paper was submitted to the 40th Session of AIOC.

Published in : SAMBODHI, Vol. XXV, December 2002, pp.91-120; Publisher: L.D. Institute of Indology, Ahmedabad.

  1. “संसार भर में किसी भी इतर प्राचीन अथवा अर्वाचीन भाषा का ऐसा परिष्कृत व्याकरण आज तक नहीं बना ।” – पं॰ युधिष्ठिर मीमांसक, संस्कृत व्याकरणशास्त्र का इतिहास, 4th edition, Ram Lall Kapoor Trust, Vol .I, 1984, p 193, lines 13-15 (in Hindi).

    “1. This general conception of the history of India ia apparently mentioned as implying the opinion of the author about the social milieu (p.107) in which nothing but obscure magic could possibly originate, an opinion fully shared by Prof. Keith. It would be interesting to know the opinion of both these authors about the milieu in which the grammar of Pāṇini, this one of the greatest productions of the human mind, originated !” --- pp.26-27 in Fedor Ippolitovich Stcherbatsky, “The Conception of Buddhist Nirvāṇa”, Leningrad: Publishing Office of the Academy of Sciences of the USSR., 1927; VI+246 pp. ↩︎

  2. Commenting on Pāṇini’s extraordinary work and about one hundred and fifty grammarians and annotators who followed in the footsteps of the great Father of Sanskrit grammar, Monier Williams (1819-1899) remarks — “It cannot be wondered ….. that the science of Sanskrit grammar should have been refined and elaborated by the Hindus to a degree wholly unknown in the other languages of the world", A Practical Grammar of the Sanskrit Language, Third Edition, The Clarendon Press, Oxford, 1864, p xii. (Reprint The Chowkhamba Sanskrit Series, Vol.XXI, 1962). ↩︎

  3. “For no language of the past have we a record comparable to Pāṇini’s record of his mother-tongue, nor is it likely that any language spoken today will be so perfectly recorded.” — Review by Leonard Bloomfield of Bruno Liebich’s “Konkordanz Panini-Candra”, Language: Journal of the Linguistic Society of America, Vol.V, 1929, p.274. ↩︎

  4. Commenting on the works of Pāṇini and his successors T Burrow remarks — “The importance of the grammarians in the history of Sanskrit is unequalled anywhere in the world. Also the accuracy of their linguistic analysis is unequalled until comparatively modern times. The whole of the classical literature of Sanskrit is written in a form of language which is regulated to the last detail by the work of Pāṇini and his successors”, The Sanskrit Language, Faber and Faber, London, 1977, p 47. ↩︎

  5. Commenting on Pāṇini and his predecessors A B Keith remarks — “… in this field Pāṇini, or more correctly his predecessors, achieved very remarkable results, as in the postulate of Guṇa and Vr̥ddhi changes, of forms with long ॠ vowel, roots in ai, masj as the original of majj, dive, s as the ending of inflexions. The analysis of forms is normally carried out with great acumen; …. In comparison with the work of Greek Grammarians Pāṇini is on a totally different plane in this regard”, A History of Sanskrit Literature, The Clarendon Press, Oxford, 1928, pp 424-425. [ “Although Pāṇini’s work has a history behind it, it is the achievement of one man”, Bloomfield, op.cit., p.274, para 2.] ↩︎

  6. The Father of modern linguistics, Leonard Bloomfield, in his book ‘Language’, which is considered the Bible of modern linguistics, remarks — “This grammar (Pāṇini’s Aṣṭādhyāyī) which dates from somewhere round 350 to 250 BC is one of the greatest monuments of human intelligence. …. No other language to this day has been so perfectly described”, Language, Henry Holt and Company, New York, 1933, p 11. Cf also Language, Vol.V, 1929, p 268, para 2 and p 274, para 2. ↩︎

  7. “पाणिनिचे व्याकरण म्हणजे मानवी बुद्धिचा एक आश्चर्यजनक नमूना आहे असे मोनियर-विलियम्सने काढलेले उद्गार प्रसिद्धच आहे”, K V Abhyankar, व्याकरण महाभाष्य, प्रस्तावना खण्ड, Part VII, Deccan Education Society, Pune, 1954, p 153, fn 1 (in Marāṭhī). ↩︎

  8. K V Abhyankar and J M Shukla, A Dictionary of Sanskrit Grammar, Oriental Institute Baroda, 1986, p 286. ↩︎

  9. “Pāṇini’s book is something more than a mere grammar. It has been described by the Soviet Professor Th. Stcherbatsky of Leningrad, as one of the greatest productions of the human mind”, Jawaharlal Nehru, The Discovery of India, Asia Publishing House, Bombay, 1972, p 115 (Reprint of 1961 edition).

    Original citation under fn 9:
    On the topic of “THE POSITION OF THE LATER SCHOOLS OF THE HÅNAYANA”, Prof. Stcherbatsky writes:
    M. de la Valee Poussin insists that in order to escape obscurity we must construct an outline (“un schema d’ensemble”) of the history of Buddhism, that this outline must harmonize with the general conception we have about the history of ancient India[1], and that questions of detail become at once settled, if they find their place in this historical outline (p.XX). ↩︎

  10. “संसार के व्याकरणों में पाणिनि का व्याकरण चोटी का है । उसकी वर्णशुद्धता, भाषा का धात्वन्वय सिद्धान्त और प्रयोगविधियाँ अद्वितीय एवं अपूर्व हैं … … यह मानव मस्तिष्क का अत्यन्त महत्त्वपूर्ण आविष्कार है ।” – W W Hunter’s statement translated in Hindi and quoted in युधिष्ठिर मीमांसक, op.cit., p.224.

    Original citation under fn 10:
    On the section “Pāṇini” under “Sanskrit Grammar”, W.W. Hunter writes:
    “The grammar of Pāṇini stands supreme among the grammars of the world, alike for its precision of statement, and for its thorough analysis of the roots of the language and of the formative principles of words. By employing an algebraic terminology it attains a sharp succinctness unrivalled in brevity, but at times enigmatical. It arranges, in logical harmony, the whole phenomena which the Sanskrit language presents, and stands forth as one of the most splendid achievements of human invention and industry.” - pp.100-101 in W.W.Hunter, “THE INDIAN EMPIRE: ITS PEOPLE, HISTORY AND PRODUCTS”, Second Edition; London: Truebner & Co., Ludgate Hill, 1886, xxx+747 pp. ↩︎

  11. “सर विलियम हंटरने सुद्धा पाणिनीय अष्टकाला मानवी बुद्धिचा महत्त्वपूर्ण आविष्कार असे म्हटले आहे” – K V Abhyankar, op.cit., p.153, fn.1. ↩︎

  12. Hermut Scharfe, Grammatical Literature, Otto Harrassowitz, Wiesbaden, 1977, p 112, para 2. In the footnote he remarks —“Often mislabeled in the past as ‘mnemotechnical devices’ ”. ↩︎

  13. “ … the study of the more abstruse work of the first great grammarian, Pāṇini”, Monier Williams, op.cit., p xi. ↩︎

  14. Even H T Colebrooke (1765-1836), the profoundest Sanskrit scholar of his day, imbued with a predilection for every thing Indian,* remarks on Pāṇini’s work — “The studied brevity of the Pāṇinīya sūtras renders them in the highest degree obscure; even with the knowledge of the key to their interpretation, the student finds them ambiguous. In the application of them, when understood, he discovers many seeming contradictions; and, with every exertion of practised memory, he must experience the utmost difficulty in combining rules dispersed in apparent confusion through different portions of Pāṇini’s eight lectures”, Miscellaneous Essays, Higginbotham and Co., Madras, 1872, Vol.II, pp 6-7. Further on p.11 he adds — “The outline of Pāṇini’s arrangement is simple, but numerous exceptions and frequent digressions have involved it in much seeming confusion…. The apparent simplicity of the design vanishes in the perplexity of the structure. The endless pursuit of exceptions and limitations so disjoins the general precepts, that the reader cannot keep in view their intended connexion and mutual relation. He wanders in an intricate maze, and the clew of the labyrinth is continually slipping from his hand.” [* Monier Williams, op.cit., p.xiii. It was the same Colebrooke, who, before tasting the beauties of Sanskrit literature, believed that there was nothing worth learning in this land of hot sun. According to him, Charles Wilkins was “Sanskrit mad”, “ ‘Asiatic Miscellany’ was a repository of nonsense”, and “ ‘the Institute of Akbar’ a dunghill in which, perhaps, a pearl or two might be found”, Eminent Orientalists, G A Natesan & Co., Madras, 1922 (Reprint Asian Educational Services, New Delhi, 1991), p.49]. ↩︎

  15. Not only (Pāṇini’s) grammar but also the Sanskrit language according to Monier Williams is artificial. He remarks – “At the best, a grammar is regarded by an European as a necessary evil, only to be tolerated because unavoidable. Especially must it be so in the case of a language confessedly more copious, more elaborate and artificial, than any other language of the world, living or dead”, op.cit., p. ix. On p.xiii he says about the grammar — “…constructed a complicated machinery of signs, symbols and indicatory letters.” ↩︎

  16. Monier Williams, op.cit., p.xiii, lines 4-6. ↩︎

  17. Ibid, p.xiii, lines 6-8. ↩︎

  18. Ibid, p.xiii, lines 8-9. ↩︎

  19. Pāṇini’s Grammar is written in cryptic sūtra form and its illogical order renders it impracticable for anyone to learn Sanskrit by its means”, Alfred Master, Jones and Pāṇini, JAOS, Vol.76, 1956, p.187. Cf also A B Keith, op.cit., p.424. ↩︎

  20. Scharfe, op.cit., p.112. Cf also p.115, para 2, where he elaborates this point —“It is a sad observation that we did not learn more from Pāṇini than we did, that we recognized the value and the spirit of his ‘artificial’ and ‘abstruse’ formulations only when we had independently constructed comparable systems. The Indian New Logic (navya nyāya) had the same fate: only after Western mathematicians had developed a formal logic of their own and after this knowledge had reached a few Indologists, did the attitude towards the navya nyāya school change from ridicule to respect. A striking example of how we only understand what we already know is the frequent translation of varṇa as ‘letter’ by F. Kielhorn and others who followed the Western grammatical tradition at least in their choice of words, while the linguistically inclined O. v. Böhtlingk at the same time correctly used ‘Laut’ (e.g. in his translation of I.3.9 and in the index under varṇa).” ↩︎

  21. “It was…. the linguistics of the India of more than two millenia ago that was the direct germinal origin of the linguistics of the Western world of today” — M B Emeneau, India and Linguistics, JAOS, Vol.75, 1955, p.145, col.1, para 1. ↩︎

  22. “The native and Medieval Greek and Latin phonology is immature and inept compared with the Hindu phonetic, phonemic, and morphophonemic analysis. One result of the difference is that on numerous points we can only guess how Latin and Greek were pronounced, while we are almost one hundred percent sure of the pronunciation of Sanskrit in Pāṇini’s time”, ibid, p.147, col.1, last para. ↩︎

  23. “Indo-European comparative grammar had (and has) at its service only one complete description of a language, the grammar of Pāṇini. For all other Indo-European languages it had only the traditional grammars of Greek and Latin, wofully incomplete and unsystematic.”, Bloomfield, Language, Vol.V, 1929, p.270, para 1. ↩︎

  24. “The Hindu achievements in morphological description were of a kind that we have only just begun to rival in modern Western descriptive linguistics and that we have not yet bettered.”, Emeneau, op.cit., p.147, col.2, para 2. ↩︎

  25. “ …. Benfey, Whitney, and all others who described Sanskrit for the West. They tended to graft on to their virtual translations and rearrangements of Pāṇini the traditional European method of description by means of paradigms, in which morphological units are essentially unidentified, similarities of the morphophonemics type are not analysed, and the description proceeds in terms of whole words – a graft which tends to obscure the excellences of the Pāṇinean system”, ibid, p.149, col.2, para 1. ↩︎

  26. John Brough remarks — “It is well known that the discovery of Sanskrit by the West at the end of the 18th century* provided the operative stimulus for the development of the comparative study of the Indo-European languages. It has also been recognized that the Pāṇinean analysis of Sanskrit into a system of roots, stems, and suffixes pointed the way to the method which has prevailed in Indo-European studies to the present day. It is true that roots and suffixes were not entirely new concepts to Europe, but it remains doubtful whether the method would have been applied with such thoroughness if it had not been for Pāṇini’s example. It is customary to add at this point the deprecatory remark that Pāṇini was, of course, aided in his analysis by the extraordinary clarity of structure of the Sanskrit language; but we are apt to overlook the possibility that this structure might not have seemed so clear and obvious to us if Pāṇini had not analysed it for us. ..we in the West have acknowledged a debt to Pàṇini in the matter of formal analysis”, Theories of General Linguistics in the Sanskrit Grammarians, Transactions of the (American) Philological Society, 1951, p.27.

    (i) Father Heinrich Roth (1620-1668) was the first European to write a grammar of the Sanskrit language in 17th century itself. He was a Jesuit missionary to the Moghul court in Agra. In Agra, Roth learnt Sanskrit and was able to discuss with Brahmins in the language. He realised the importance of Sanskrit and wrote a grammar between 1660-1662. The grammar is descriptive and shows great pedagogical skill – the explanations are given in Latin. The grammar is based on Pāṇini. The grammar was later highly appreciated by Max Müller. The manuscript was taken to Rome, but never printed, although several scholars and even the Austrian Emperor wanted to have it published…. Roth not only studied Sanskrit but was also well-versed in Sanskrit literature and Indian philosophy…. The grammar and manuscripts of Father H. Roth were published in a facsimile edition in Leiden, 1988.”, Valentina Stache-Rosen, German Indologists, 2nd revised edition, Max Müller Bhavan, New Delhi, 1990, pp.1-2.
    (ii) The other early European scholars to write Sanskrit grammar are — Johann Ernest Hanxleden (1681-1732) and Jean-Francois Pons (1688-1752?). See Filliozat (Ref. under footnote 29), Introduction, p.41, para 2. For details see J. C. Muller, “Recherches sur les premières grammaires manuscrites du sanskrit”, Bulletin d’études indiennes, No 3, Paris, 1985, pp. 125–144.
    (iii) Theodore Benfey (1809-1881) in his ‘Geschichte der Sprachwissenschaft und orientalischen Philologie in Deutschland’ (History of Linguistics and Oriental Philology in Germany), 1869, p.222 tells about the Italian scholar Filippo Sassetti (c.1585) – “Schön im 16. Jahrhundert schrieb Sassetti, daß das Italienische vieles mit dem Sanskrit gemeinsam habe, Zahlwörter und andere Wörter” (As early as 16th century Sassetti wrote that Italian has much in common with Sanskrit regarding the numerals and other words) and at p.336 onwards — “Im Jahre 1725 verglich Benjamin Schultze, Missionar in Tranquebar, die Zahlwörter des Sanskrit mit den lateinschen bis 40” (in the year 1725, Benjamin Schultze, a missionary in Tranquebar, compared the numerals of Sanskrit with those of Latin upto 40). The original German quotations are taken from Ernst Windisch’(1844-1918) “Geschichte der Sanskrit Philologie und Altertumskunde” (History of Sanskrit Philology and Ancient Culture), p.24, fn 1. Cf also Alfred Master, Jones and Pāṇini, op.cit., p.186, col.1. ↩︎

  27. “Pāṇini and the Mahābhāṣya are documents of the first importance for the historical interpretation of old India and of the Sanskrit language, as well as for the history of linguistic science in general”, Paul Thieme, Pāṇini and the Pāṇinīyas, JAOS, Vol.76, 1956, p.23. ↩︎

  28. Regarding importance of Pāṇini, Louis Renou refers to Pavolini, Asiatica, 3, No.1 (1938), vide Jakob Wackernagel’s Altindische Grammatik (in German), Vol.I, 2nd edition, Vandenhoeck and Ruprecht, Göttingen, 1957, Introduction générale (in French), p.113, fn.513. I have not had access to the journal Asiatica mentioned by him. ↩︎

  29. “… l’Aṣṭādhyāyī, n’est pas une grammaire sanskrite, c’est une oeuvre de linguistique générale’, Pierre-Sylvain Filliozat, GRAMMAIRE SANSKRITE PÂËINÉENNE, Picard, Paris, 1988, Introduction, p.12, lines 40-41 (in French). Cf also the following lines of the article “Les notions de verbe et de substantif dans l’Ecole pāṇinéenne” by the same author, where he expresses his views more lucidly — “ L’étude de Pāṇini est un chapitre important de l’histoire des sciences. La grammaire dans l’Inde ancienne n’est pas une simple discipline d’érudition, c’est une science. Elle se présente elle-même comme une linguistique générale et non comme une grammaire d’une langue particuliére en une période donnée. Pāṇini et ses successeurs traitent du langage en général en traitant du Sanskrit. Ils considérent que le Sanskrit est le langage parfait, unique, et que toute autre langue en est une corruption. Il n’y a donc pas chez eux de considérations historiques et comparatives.” (The study of Pāṇini is an important chapter in the history of science. The grammar of ancient India is not a simple discipline of erudition, it is a science. It presents itself as general linguistics and not as a grammar of a particular language of a given period. Pāṇini and his successors treat the language in general while dealing with Sanskrit. They think that Sanskrit is a unique and perfect language and that all the other languages are a corruption. There are, therefore, no historical or comparative considerations in them), Bulletin d’études indiennes, No 1, Paris, 1983, p.81, para 1. ↩︎

  30. Scharfe, op.cit., p.115, para 2. The title of the Wilkins’ grammatical work is ‘A Grammar of the Sanskrita Language’ (1808) [Reprint Ajay Book Service, 704 Chandni Mahal, Darya Ganj, New Delhi, First Indian Edition, 1983]. ↩︎

  31. Die Sprache der Jakuten (A. Th. Von Middendorff, Reise in den äussersten Norden und Osten Siberiens, Bd.III, St. Petersburg, Kais. Ak. D. Wissen., 1851) – Emeneau, op.cit., p.150, fn 18. It was the first scientific grammar of the hitherto not at all studied Yakut language. Böhtlingk, member of the Russian Academy of Sciences, initiated the civilian written form of the Yakut language. While doing research on Yakut language, he devised the civilian Yakut alphabet for the first time, against the parallel and already existing missionary alphabet. The publications of religious literature by the missionaries in Yakut language from 1819 onwards were not accessible to the Yakut people, as the masses were illiterate. The school education started only in the last few decades of the 19th century (p.23 in the Ref . mentioned below). In a letter to Böhtlingk, the Yakut born Russian A Y Uvarovskiy (1800-c.1860), teacher and great inspiration to the former regarding research on the Yakut language, warmly greeted his initiation of the creation of written Yakut language: “Iz-za otsutstviya pis’mennosti yakutskiy yazik schitaetsya mertvim yazikom. Nedaleko to vremya, kogda Vi ozhivite ego. Nedaleko to vremya, kogda Vi polucite pokhvali ot visokoobrazovannikh lyudey I beskonecnuyu blagodarnoct’ ot yakutskogo naroda. Gryaduschee pokolenie yakutov polnoct’yu ispol’zuet Vase tvorenie I visoko podnimet Vase imya…. Eto budet vozdayaniye ot nikh, eto budet nagrada dlya Vas.“ (Due to the absence of the written form, the Yakut languge is considered to be a dead language. The time is not far away, when you will revive it. The time is not far away, when you will receive appreciation from highly educated people and the endless thanks and gratitude from the Yakut people. The coming generation of Yakut will fully utilize your work and raise your name high…… It will be a requital from them and a reward for you.”), Ocerk istorii yakuctskoy sovetskoy literaturi, izdatel’stvo akademii nauk SSSR, Moskva, 1955, str. 24 (Essay on the History of Yakut Soviet Literature, Publishing House of the Academy of Sciences, USSR, Moscow, 1955, p.24). - in Russian. As expected by Uvarovskiy, Böhtlingk’s work excited great interest in the scientific circle, which had important after-effects in the study and development of the Yakut language. ↩︎

  32. Emeneau, op.cit., p.150, col.1, para 2. ↩︎

  33. “Obituary Notice: Dr E E Obermiller”, by Th. Stcherbatsky, IHQ, Vol.12, 1936, p.380. Cf also Scharfe, op.cit., p.115, fn 119. ↩︎

  34. Pt Yudhiṣṭhir Mīmāṁsak assigns same date to Pāṇini and Vyāḍi, op.cit., p.193 and 298. He considers Vyāḍi to be the maternal uncle of Pāṇini. – cf p.301, line 5. ↩︎

  35. “He (i.e. Vyāḍi) is believed to have been a relative and contemporary of Pāṇini”, A Dictionary of Sanskrit Grammar, op.cit., p.378. ↩︎

  36. “As Vyāḍi has used a number of Pāṇini’s rules in deciding his Paribhāṣās and, as he has actually quoted a few vārttikas he appears to have flourished a few decades after Pāṇini, when there was an addition of a few vārttikas only for the sūtrapāṭha of Pāṇini”, K V Abhyankar, Paribhāṣāsaṁgraha, B.O.R.I., Poona, 1967, Introduction, p.12. ↩︎

  37. “सम्पूर्ण भाष्य में 4280 वार्त्तिक हैं । इनमें 3870 वार्त्तिक कात्यायन के हैं । 410 वचन अन्य आचार्यों के एवं स्वयं भाष्यकार के हैं”, वेदपति मिश्र, “व्याकरण-वार्त्तिक : एक समीक्षात्मक अध्ययन”, पृथिवी प्रकाशन, वाराणसी, 1970, आमुख, p.7. ↩︎

  38. “Nāgeśa appears to have divided vārttikas into two classes as shown by his definition ‘सूत्रेऽनुक्त-दुरुक्त-चिन्ताकरत्वं वार्त्तिकम्’. If this definition be followed, many of the vārttikas given in the Mahābhāṣya as explained and commented upon the sūtras will not strictly be termed as vārttikas, and their total number which is given as exceeding 5000, will be reduced to about 1400 or so. There are some manuscript copies which give this reduced number, and it may be said that only these vārttikas were written by Kātyāyana while the others were added by learned grammarians after Kātyāyana. In the Mbh. there are seen more than 5000 statementüs of the type of vārttikas out of which Dr Kielhorn has marked about 4200 as vārttikas”, A Dictionary of Sanskrit Grammar, op.cit., pp.247-248. ↩︎

  39. Goldstücker, “Pāṇini: His Place in Sanskrit Literature”, London and Berlin, 1861 (Reprint, The Chowkhamba Sanskrit Series Office, Varanasi, 1965), p.57. ↩︎

  40. The number of total sūtras varies from edition to edition. Ram Lall Kapoor Trust, Bahalgarh, edition contains 3964 sūtras, while Kāśikā published from the same place contains 3981; the Vr̥ndāvan edition (Reprint अनीता आर्ष प्रकाशन, पानीपत, 1990), contains 3973; Krishnadas Academy, Varanasi edition, edited by Satyanarayan Shastri Khanduri, contains 3983, which is same as in Böhtlingk’s edition (various reprints are available, e.g. MLBD, Delhi, and Rinsen Book Company, Kyoto, 1977); the one edited by C. शंकरराम शास्त्री (Reprint Sharda Publishing House, Delhi, 1994) contains 3981 sūtras. The number of sūtras in the last one and that in Kāśikā mentioned above is, although, same, there is difference in the number of sūtras in ch.4, pāda 1 and ch.8, pāda 4. The number of sūtras preserved in Siddhānta-kaumudī, published by MLBD, is 3978. The number mentioned in the main body of the article is taken from Böhtlingk’s edition. ↩︎

  41. Robert Birwé, Studien zu Adhyàya III der Aṣṭādhyāyī Pāṇinis (in German), Otto Harrassowitz, Wiesbaden, 1966, Table at pp.156-157. ↩︎

  42. रघुवीर वेदालंकार, “काशिका का समालोचनात्मक अध्ययन”, Nag Publishers, Delhi, 1977, परिशिष्ट-7, pp.359-387. The list gives 913 vārttikas plus additional four under 191A, 413A, 418A and 711A. ↩︎

  43. Ibid परिशिष्ट-8, pp.388-390. ↩︎

  44. Ibid, परिशिष्ट-10, pp.394-398. ↩︎

  45. K C Chatterji, Cāndravyākaraṇa of Candragomin, Part-I (ch.1-3), Deccan College Post-Graduate and Research Institute, Poona, 1953, PREFACE, p.v, lines 1-2. ↩︎

  46. E.g. IV.3.93. “स्वरविशेषं तु स्वराध्याये वक्ष्यामः” and I.1.145. “स्वरविशेषमष्टमे वक्ष्यामः” – Chatterji, op.cit., Part II (ch.4-6), 1961, p.90 and Part I, p.51. ↩︎

  47. Chatterji, op.cit., Part I, p.v, lines 28-29. ↩︎

  48. This work is not to be confounded with another work of Bhoja named likewise Sarasvatīkaṇṭhābharaṇa. It deals with poetics. ↩︎

  49. Madras Edition of SKĀ, FOREWORD, p.vii. ↩︎

  50. Ibid ↩︎

  51. Ibid ↩︎

  52. Trivandrum edition of SKĀ, Vol.IV, Prastāvanā, p.6, para 3. ↩︎

  53. Ibid, pp.6-7. ↩︎

  54. “In Kr̥ṣṇalīlāśuka’s commentary on Bhoja’s grammatical treatise, Sarasvatīkaṇṭhābharaṇa Pāṇini’s verses are quoted freely as illustrations” and “In the field of grammar and philosophy, his proficiency was over as great as in the field of poetry. He commented on the Aṣṭādhyāyī of Bhoja (known as Sarasvatīkaṇṭhābharaṇa) and there quotes several verses of Pāṇini.”, M Krishnamachariyar, History of Classical Sanskrit Literature, MLBD, New Delhi, p.85 and pp.335-336 respectively. ↩︎

  55. “ ..le S(arasvatī)K(aṇṭh)Ā(bharaṇa) est beaucoup plus qu’une édition “revue” de P. D’abord, il a incorporé dans la trame des sū. tous l’enseignement positif émané des vieux v(ār)tt(ika), celui-là même qui, à peu de choses près, nous a été conservé dans la K(āśikā): laissant de côté, comme fait cette dernière, les vtt. de caractère technique, scolastique, argumentatoire. …. Ensuite, des additifs nouveaux – semblables aux axiomes introduits ailleurs par iti vaktavyam, ity upasaṁkhyānam – prennent rang de sū. …. Il y a eu là tous un effort pour sortir des limites pāṇinéennes, ….et qui fait de ce manuel le plus complet que nous ait légué la tradition sanskrite.”, p.121. ↩︎

  56. fn.2, pp.122-123. ↩︎

  57. p.123. ↩︎

  58. “Ceci montre le part assez considérable de nouveautés qu’apporte la grammaire de Bhoja. Nouveautés au sens relatif du terme, car il est aisé de voir que la vr̥tti de Candragomin, oeuvre de l’auteur même des sū. et bien antérieure au SKĀ, donnait déjà, dans le long commentaire qu’elle fabrique sur le sū. II.2,48, la plupart des examples que cite la vr̥tti de Bhoja ou qui résultent de l’enseignement des sū.”, p.125. ↩︎