Pramāṇa · Search · About · API

An Account of the Different Existing Systems of Sanskrit Grammar c7: The Chāndra School

An Account of the Different Existing Systems of Sanskrit Grammar c7

Language
English

The Chāndra School

c7.1

42. The Chāndra School.—The earliest reference to the Chāndra school of grammarians occurs in Bhartṛihari's Vākyapadīya (see p. 41 above), while one of the latest is perhaps that of Mallinātha, who quotes a rule of his in his commentary on Kālidāsa's Meghadūta, stanza 25 (नीचै-राख्यं गिरिमधिवसेः).² Mallinātha, however, does not appear to have had a direct access to the Cāndra vyākaraṇa, seeing that Mss. of the work have been extremely rare, none of the various ‘Searches for Sanskrit manuscripts’ instituted by Government having been able to bring to light any works of the school except a fragment brought by Dr. Bühler from Kāśmir in 1875, and a complete copy of the Cāndra vyākaraṇa written in the Nepalese year 476 (i. e. 1356 A. D.) brought by Haraprasāda Shastrī from Nepal.¹ However, by the labours of Dr. Bruno Liebich, the whole system has now been recovered in the original or Tibetan translation. The same scholar has also published the Cāndra vyākaraṇa (Leipzig 1902). The account of the system given below is mostly based on his writings.

c7.2

² In the passage cited Mallinātha says that while Pāṇini allows only the form विश्रम Chandra allows विश्राम also. As a matter of fact Chandra allows only one form (Chandra sūtra vi. 1.42); it is Śakaṭāyana and Hemachandra who allow both the forms, which are indiscriminately used in classical Sanskrit. Presumably, therefore, Mallinātha either had access to a work of the Chāndra school not known to us, or more probably he meant by Chandra Hema-chandra, unless the whole is a positive mistake. I owe this note to Mr. Krishṇaji Govinda Oka, editor of the Kshīrataraṅgiṇī.

c7.3

¹ See Nachrichten der Goettinger Gesellschaft 1895, pp. 272-321.

c7.4

43. The date of Chandragomin.—Chandra, or more accurately, Chandragomin must have lived at least some time before the authors of the Kāśikā, which has borrowed, always without acknowledgment, such sūtras of Chandra as have no parallel either in Pāṇini or in Kātyāyana. This gives us 650 A. D. as the lower limit for Chandragomin. The upper limit is supplied by a vṛtti on the Cāndra sūtras, most probably the work of Chandragomin himself,² which gives the sentence अजयद्गुप्तो(? Ms. जतो or जमो) हूणाच् as an illustration of the use of the imperfect to express an event which occurred within the lifetime of the speaker. This victory over the Hūṇas can refer either to their temporary defeat by Skandagupta soon after 465 A.D., or (less likely) to their final expulsion by Yaśodharma³ in 544 A. D. This gives us 470 as the approximate date for Chandragomin. This result is further confirmed by the fact that Vasurāta the preceptor of Bhartṛhari acknowledged Chandrācārya (Chandragomin) as his master.¹ Chandragomin must have lived therefore at least two generations before the author of the Vākyapadīya. All accounts agree in stating that Chandragomin was a Bauddha. He was one of the laity, and is not to be confused with Chandradāsa who belonged to the order.²

c7.5

² See Dr. Liebich's paper "Das Datum Chandragomin's und Kalidasa's", p. 3.

c7.6

³ Who, however, was not a Gupta.

c7.7

¹ See Vākyapadīya Kāṇḍa ii, stanzas 489-90 and com. thereon.

c7.8

² Liebich, ibidem, p. 10-11; Kern: Manual of Buddhism, pp. 129, 130; also Ind. Ant, xv. pp. 183-184.

c7.9

44. Nature of Chandragomin's work.—Chandragomin's grammar was meant as an improvement on that of Pāṇini, Kātyāyana, and Patañjali, mainly in the way of greater brevity and precision. Accordingly he has omitted, for obvious reasons, the Pāṇinīya rules about Vedic accent and grammar, although he includes some Vedic roots in his Dhātupāṭha. He has lessened the number of pratyāhāra-sūtras by one (fusing हयवरट् and लण् into हयवरलण्), omitted some of the Pāṇinīya pratyāhāras and coined others. In many cases, the rules of Pāṇini are recast simply for the sake of securing facility of pronunciation.³ The really original contributions of Chandragomin amount to about 35 sūtras and these have been incorporated in the Kāśikā. In all these cases Kaiyyaṭa has the remark अपाणनीयः सूत्रेषु पाठः. The total number of the Cāndra sūtras is about 3100 as against 4000 of Pāṇini. The work consists of six chapters of four pādas each, the matter of Pāṇini's first two chapters being scattered all through.

c7.10

³ For Pāṇini's अनेकाल् शित् सर्वस्य (i. 1. 55) Chandra reads शित्-नेकाल् सर्वस्य (i. 1.12).

c7.11

The object of Chandragomin was to rearrange the grammatical material with the object of bringing together all the rules that deal with the same phonetic or grammatical operations as well as the same part of speech.' The Cāndra terminology with slight changes is that of Pāṇini. The mode of presenting the subject is also artificial, after the fashion of Pāṇini. The grammar goes by the nickname of असंज्ञक, perhaps because the संज्ञाs are not here treated of separately, but probably because wherever in his sūtra Pāṇini has used the word संज्ञा Chandragomin uses the word नामन्.¹

c7.12

¹ Compare Cāndra sūtras i. 2.30, i. 3. 77, ii. 2. 14, &c. with Pāṇini's iii. 2. 46, iii. 3. 174, ii. 1. 21 &c. A few cases do occur, however, where Chandra permits the use of the word: e. g. Chandra i. 1. 123=Pāṇini iii. 1. 112.

c7.13

45. Accessory treatises of the Cāndra grammar.—In addition to the sūtras in six adhyāyas Chandragomin has put together an Uṇādi list in three parts, a Dhātupāṭha in ten sections (both published by Dr. Liebich), as also Liṅgakārikās or Liṅgānuśāsana, Gaṇapāṭha, Upasargavṛtti, and Varṇa sūtras. The Uṇādis differ from those belonging to the Pāṇinīya school principally in their mode of presentation, the suffixes being here arranged according to their final letter. In a few cases Chandra also derives the words differently. The Dhātupāṭha, as we saw (p. 52, above), is referred to by Kṣīrasvāmin and was subsequently incorporated in the Kātantra system. The Liṅgānuśāsana is referred to by Vāmanācārya, Ujjaladatta, and Rāyamukuṭa (see above, p. 53). As to the Gaṇapāṭha no separate work of the kind has yet been discovered, but we must assume the existence of such a work as we find it embodied in the sūtra-vṛtti, just as the Kāśikā has done with regard to the Pāṇinīya Gaṇapāṭha. The Upasargavṛtti is found in Tibetan version only, and explains the meaning and use of about twenty upasargas. Finally, the Varṇasūtra (Ms. no. 289 of 1875-76 in the Deccan College collection) is a very short treatise² corresponding to the Pāṇinīya Śikṣā and gives in about 40 sūtras the स्थान and प्रयत्न of वर्णs. No work on Paribhāshās in connection with the Chāndra school has come down to us.

c7.14

² I take this occasion to publish the work entire on the basis

c7.15

Besides the above grammatical works Chandragomin is credited with the authorship of a religious poem called Śishyalekhā, and a drama called Lokānanda, neither probably of much consequence.

c7.16

46. Later history of the Chāndra school.—We have already alluded to Chandragomin's own vritti on his grammar. Fragments from it extending from about v. 1. 13 to v. 1. 176 are still extant. This vritti was later incorporated in a commentary by Dharmadāsa, a complete Ms. of which exists in the Library of the Mahārāja of Nepal.

c7.17

It is undoubted that there must have been written numerous commentaries on the Chāndra Vyākarana during the palmy days of Buddhistic literature; and they must have been very popular, seeing that a good many of them have been translated and freely circulated in Tibet at least since 1000 A. D., if not earlier, when Sthiramati, one of the translators of most of the Chāndra texts in the Tibetan language, probably lived. Some of these works had also gone to Ceylon along with other Buddhistic texts. However, at present, in addition to the works above mentioned, only a few more—about fifteen—are known to exist, mostly in Tibetan translations.¹ Such of the Sanskrit Mss. as we know of, come all from Nepal.

c7.18

¹ For a list of these see Ind. Ant. xxv, pp. 103 and following.

c7.19

Having once enjoyed such a vast circulation, the almost total disappearance of the system from India requires explanation. We can account for this fact, firstly, on the ground of its want of originality, such of the original matter as there was—and it was not much—being already incorporated in the Pāṇinīya school through the Kāśikā. Mainly however we must look to the cause of its disappearance in its non-secular character. Being the work of a Buddhist for the Buddhistic community, it shared the fate of Buddhism, and having obtained vogue for a few centuries it gradually ceased to be cared for, its aid being invoked in later times only for the sake of justifying an otherwise unjustifiable word, or for pointing out and rejecting such of its rules as went counter to the established system of grammar. The Grammar, we are told, is still extensively studied in Tibet.

c7.20

In Ceylon its fate was different. Being a Buddhistic country we expect the Cāndra system to be diligently studied there. As a matter of fact, the current Sanskrit grammar in Ceylon belongs to the Cāndra school, but we shall look in vain for any original Mss. either of the Cāndra-sūtras or of commentaries thereon.

c7.21

The reason is that about 1200 A. D. a Ceylonese Buddhistic priest, Kāśyapa by name, wrote a popular recast of the Cāndra grammar called Bālāvabodha. It corresponds to Varadarāja's Laghu-kaumudī in treatment and subject-matter. The work was so popular in Ceylon that it quite superseded the original Cāndra text, with the result that all other Cāndra works have disappeared in course of time, just as the works of the pre-Pāṇinīya grammarians did after the advent of Pāṇini.

c7.22

Under these circumstances, it is quite impossible to pursue any farther the history of the Cāndra school of grammarians in India.

Cites (12)