3. Sanskrit Usage (Study)

Sankaracarya commenting on the first verse of the Isopanisad: isavasyam idfam sarvam yat kinca jagatyam jagat I tena tyaktena bhunjitha ma grdhah kasyavid dhanam || offers two alternative interpretations of the fourth quadrant: ma grdhah kasyavid dhanam. He says: ma grdhah grdhim, akanksam, ma karsir dhanavisaya kasyasvid dhanam kasyacit parasya svasya va dhanam ma kanksir. ity arthahi svid ity anarthako nipatahi athava ma grdhahi kasmati kasyasvid dhanam ity aksepartho, na kasyacid dhanam asti yad grdhyetal M Obviously the Acarya regards the √grdh as transitive for he says in the first instance kasyasvid dhanam kasyacit parasya svasya va dhanam ma kanksih. Here he looks upon this quadrant va as one single sentence. In the second instance he splits it up into two clauses: ma grdhah, dhanam kasyasvit. Even then he persists in viewing the root as transitive for he remarks: na kasyacid dhanam asti yad grdhyeta. The verse under reference is alluded to in the Mahabharata: kasyedam iti kasya svam iti vedavacas tatha'. Evidently the Mahabharatakara splits the Upanisadic sentence in the same way as Sankaracarya does. But this splitting by itself would clearly point to the intransitive nature of the root, precluding dhanam from standing as an object to grdhah. It is the Acarya's obsession with the transitive nature of the root that llection, New Delhi. Digitized by $3 Foundation USA J

Sanskrit Usage 31 leads him to supply yad grdhyeta even in the second interpretation. Against this view of Sankaracarya we have a mass of evidence which goes to show that √grdh is almost invariably used intransitively. The impression that √grdh is transitive has probably originated from the meaning assigned to it in the Paniniya chatupatha which reads: grdhu abhikanksayam. As it 3 stands it would mean that the senses of grah is to covet' for " grdhyati would be paraphrased by abhikanksati which would require normally an object. That the root is intransitive is further evidenced by the use of √lubh as intransitive. The Dhatupatha gives the meaning of √lubh as gardhya, covetousness. the nature cap of this root is not at all in dispute. Unfortunately we have very little use of this root as a finite verb (in tinanta form) though we have ample use of it in the verbal derivative forms such as grdhi, gardha, gardhana, grdhnu etc. in classical Sanskrit literature. The Vedas and the epics abound with the use of the root as a finite verb. The following few illustrations would suffice to prove the point: i. yasyagrdhad vedane vajy aksahi2 ii. niramino ripavo 'nnesu jagrdhuhi³ iii. ma grdho no ajavisur iv. durnama tatra ma grdhati V. te patnisv eva gandharva gardhisyantil vi. yada grdhyet parabhutau nrsamsahi" vii. paravittesu grdhyatahi 8 7 viii. esa dharmah paramo yat svakena raja tusyen na parasvesu grdhyeti' ix. anityam yauvanam rupam jivitam ratnasancayahi aisvaryam priyasamvaso grdhyet tatra na panditahi¹º X. anyonyam abhigarjanto gosu grddhahi" xi. grahane dharmarajasya bharadvajo'pi grdhyati¹² xii. grdhyed esu na panditah113 xiii. snatanuliptagatre'pi yasmin grdhyanti maksikah 114 Bhattoji Diksita in his Siddhantakaumudi takes √trp of the IV conjugation to be transitive as well, leaning on Bhatti's use of it as a transitive verb: pitfn atarpsit. The Sanskrit usage, howevery negates its positively. Not only is by Foundation USA is this root of the IV

conjugation invariably intransitive, but also the root of the V conjugation is likewise as can be seen from the following examples: i. anukamam tarpayetham indravaruna raya a 15 ii. indra somasya varunasya trpnuhil 16 iii. hiranyavarna atrpam yadavahi" iv. trpyantu hotra madhvah18 v. vag devi jusana somasya trpyatu!19 - 300 vi. na hi trpta'smi kamanam jyestha mam anuman-yatam 20 vii. aharahar nayamano gam asvam purusam pasumi vaivasvalo na trpyati suraya iva durmadi I viii. cirasya drstva dasarha rajanah sarva eva tau ! amrtasyeva natrpyan preksamana janardanam 1122 ix. atrptis cannasyal 23 x. nayunjano bhaksyabhojyasya trpyeti 24 xi. kamanam avitrptas tvam srnjayeha marisyasil 2 xii. gavyasya trpta mamsasyar²26 xiii. nagnis trpyati kasthanam napaganam mahodadhihi 27 xiv. avitrptah sugandhasya samantad vyacarad vanami XV. atrpto'smy adya kamanam²⁹ xvi. madbananam tu vegena hatanam tu ranajirel adya trpyantu mamsadah...30 xvii. apam hi trptaya na varidharajsvaduh sugandhih svadate tusaral³¹ xviii. athava sreyasi kena trpyater³2 Grammar lays down the rule that the object takes the Genitive when in construction with a verbal derivative 33 In contravention of this, the usage sanctions only the Accusative. And this has the tacit approval of Panini himself for he uses the Accusative in construction with arham, a verbal derivative, in the sutra: tadarham,34 e.g., i. na parityagam arheyam matsakasad arindamas ii. prthivirajyam arho 'yam nahngarajyam naresvarah136 iii. tasmat pravaranam purvam arhah partho dhananjayah iv. arhas tvam asi dharmajna rajasuyam mahakratum 138 v. tada visargam arhah syur itidam dhatrsasanam 19 vi. naivarhah paitrkam riktham fi

Sanskrit Usage 33 According to the Dhatupatha √vad (with or without sam) is Parasmaipadin but according to usage with sam it is invariably Atmanepadin. The Bhasyakara is merely upholding the usage when he employs the Atmanepada termination in the expression: vacikasadikau na samvadete. 41 We have in the Rgveda as also in the Nirukta the Atmanepada use of sam+vad: uta svaya samvade tat, 42 indra tvam marudbhih samvadasva. 43 devasunindrena prahita panibhir asuraih samuda ity akhyanam 44 The Brhaddevata also reads: sukte preti tu nadyas ca visvamitrah samudire. 45 The Chandogya Upanisad too has it in atha hagnayah samudire tapto brahmacari kusalam nah paryacarit. 46 So do have Kausitaki Brahmana Upanisad 47 and the Brhadaranyaka Upanisad 48. i. mrsa kila vai mam samvadistha brahma te bravaniti ii. sa hovacajatasatrur ma maitasmin samvadistha iti Elsewhere also we come across, not infrequently, this use of the Atmanepada. Thus: i. kumaram jatam samvadanta upa vai susrusatel 49 ii. deva vai brahma samavadanta po iii. sa vai na sarveneva samvadeta ...na vai deva sarveneva samvadante³¹ As required by grammar anurakta and anuvrata should have either the Locative or the Genitive of the person or the thing to whom or to which one is devoted. And we have this use. But the Accusative preponderates, though it has no explicit grammatical sanction, for example: i. satyakih krtavarma ca narayanam anuvratau 52 ii. rajano rajaputras ca dhrtarastram anuvratah$3 iii. abhyagacchad adinatma damayantim anuvratahi 54 iv. anviksikim tarkavidyam anurakto nirarthikam! 55 v. alabhyam anuraktavan kim ayam atmanarijanami 56 vi. api vrsalam anuraktah prakrtayah?57 vii. mahodadhim ivaksobhyam aham ramam anuvrata 58 viii. rajyac cyutam asiddhartham ramam parimitayusam kair gunair anuraktasi p The Kasikakara specifically limits the use of rajasvala and karmuka in the sense of 'a lady in menses' and 'a bow' edc

respectively. In the sutra tad asyasty asminn iti matup 60 he says that iti is valid in all the following sutras ordaining the possessive suffixes. And this iti limits the option of the speaker. Thus to convey the sense rajo'smin grame vidyate we cannot say rajasvalo gramah. Now this is contradicted by usage: i. sarve vidhvastakavacas tavaka yudhi nirjitah 1 rajasvala bhrsodvigna vikssamana diso dasa 161 ii. gadaya bhimasenena bhinnakumbhan rajasvalani dhavamanan apasyama kunjaran parvatopaman 1162 iii. pankadigdhan rajasvalan 63 iv. rajasvalam anityam ca bhutavasam imam tyajet 11 64 In the case of karmuka the Kasikakara says under Pan. karmana ukan65 karmukam dhanuhi dhanuso 'nyatra na bhav-ati anabhidhanat. This is not supported by usage. We have the use of karmuka in the sense of 'effective' in the Caraka Samhita: i. na tu kevalam gunaprabhavad eva karmukani bhavanti (dravyani) 166 ii. bahuso gulikah karyah karmukas syus tato 'dhikam thº¹ II Panini's Regard for it In his sutra 'avadyapanyavarya garhyapanitavyanirodhesu'68 'Panini reads varya and says that though not regularly derived, it has to be accepted as correct, provided it is used in the feminine, in the sense 'to be freely chosen'. Normally the form in this sense would be varya derived from √vrn with nyat. He restricts the use of varya to the feminine, for such is the usage. But Bhatti for whom usage is not very important (for he draws for his forms more upon his imagination than upon the standard literature) has no hesitation in using varya in the masculine Sugrivo nama varyo sau. Again in framing the sutra padasvairibahyapaksyesu ca 69, J Panini uses bahya (outer) in the feminine, implying that the form grhya evolved from √grah with the suffix kyap has to be restricted 500

Sanskrit Usage 35 to the feminine, for such is the usage. This limits our option. We can say nagaragrhya sena but not nagaragrhyam sainyam. As a grammarian, the Acarya would have stopped at evolving the correct form grhya without defining its precise use, padam vyakaranam smrtam, Vyakarana is analysis of word forms. Panini lays down certain rules for the evolution of certain forms. He analyses them and sets his seal of approval on them, and then proceeds to recognise certain other forms not directly covered by his rules, because they are in vogue amongst the cultured (sistas). He asks his students to look for such forms in literature, viz., in the speech of the sistas and take them as approved by grammar, for they have the sanction of usage. While urging his students to do so, he uses expressions like anyesv api drsyate, anyebho pi drsyate or bahulam or prayah, etc. 72 By his sutras 'saptamyam janer dah70', 'pancamyam ajatau?\','upasarge ca samjnayam 2, 'anau ca karmani.73 Panini prescribes the use of the suffix da after √jan and sets down forms like mandurajah, buddhijah, prajah, pumanujah as the appropriately evolved usable forms, but he is conscious that these rules of his leave a large majority of forms apparently with da uncovered. He acknowledges the existence of such forms in his sutra 'anyesv api drsyate'74 and asks his students to pick up such forms in literature and accept them as grammatically correct. Thus forms like ajah (na jayata iti), dvijah (dvir jata iti), brahmanajah (dharmah), abhijah (kesah) are all correct, though there is no explicit sanction in grammar, for they are sanctioned by usage. Not only that, usage goes far beyond. By the use of api in the sutra, Panini suggests that da may come after any other root than jan even in a sense other than that of agency, just as in parikha (parikhayate parikhanyata iti). Here we have da in the sense of the Accusative. Trju In the sutras 'bharajabhasadhurvidyutorjiprgravastuvah kvip'75 Panini mentions several roots such as bhraj, bhas as taking kvip a Tacchilika suffix (signifying habitual action, etc.). Immediately after he reads anyebhyo pi drsyate. Why does he use drsyate? The Vrttikara answers: drsigrahanam vidhyantaro16

pasangrahartham, kvacid dirghah, kvacid dvirvacanam, kvacid asamprasaranam. tatha cahakvibvacipracchyayatastukataprujusrinam dirgho Samprasaranam ca, drsyate is there to ensure forms got from other roots and to include other unnamed grammatical operations involved in the process of the evolution of the usable forms from the roots. Thus there is nothing wrong with the forms evolved from these roots, if some of them have lengthening, or the samprasarana, or the absence of it, or reduplication as in ayatastu, vidyut, katapruh, vak, jabdaprat, juh not permissible in grammar. Panini reads 'itarabhyo 'pi drsyante 76.' He observes that the svarthika suffixes tasil, tral etc. go beyond their proper sphere and are found used even after bases ending in other case-endings than the fifth or the seventh. It is to be noted that tasil and tral come after kim, sarvanama and bahu. While sa bhavan is perfectly normal tato bhavan, tatra bhavan are quite abnormal. Similarly while tam bhavantam is quite reasonable, tato bhavantam, tatra bhavantam has no justification. To a student of grammar who has not yet had access to literature, these expressions sound strange. But usage is all powerful and has to be accepted. In the sutras 'pancamyas tasil' and 'saptamyas tral 78, Panini lays down that bases such as sarvanama, kim and bahu take tasil after their pancamyanta forms and tral after their saptamyanta forms without any additional sense. By his sutra itarabhyo 'pi drsyante 79 he extends their use to bases with other case-endings. Thus along with tato bhavatah, tatra bhavati we can, if we like, say tato bhavan, tatra bhavan, tato bhavantam, tato bhavata, tatra bhavata so on and so forth. On the face of it, this use of tatah and tatra seems to be ill-adjusted and entirely unjustifiable. Yet Panini does not discard this. Under bahulam used a number of times in his sutras Panini covers such forms where there is scope for his rule and yet they do not apply or apply optionally, while his rules would be otherwise obligatory or admit of certain changes or modification which his rules do not ordain. Nowhere does he reject such forms

Sanskrit Usage 37 as ungrammatical. We illustrate: ramo jamadagnyah, arjunah kartaviryah have no compounding, though the two make a fit case for one. This is because Panini uses bahulam in his sutra 'visesanam visesyena bahulam 80. Usage restrains our option. In krsnasarpah and lohitasalih we have compounds which are nitya, not analysable into their components. Panini prescribes the krtya suffixes in the sense of bhava and karma, predominantly in the sense of karana and adhikarana. But he extends their use to senses even specified to be observed in actual use in literature. This he does by his sutra 'krtyaluto bahulam 81. This lends grammatical sanction to multitude of forms in krtya and lyut. In 'rogakhyayam nvul bahulam 82 Panini tells us that names of diseases are formed by the addition of nvul. He does not specify the base or bases for this suffix. Here too usage is the only guide. Hence, he reads bahulam in the sutra. Thus while we have formations in vul like pracchardika (vomitting), pravahika (dysentery), vicarcika (itching), we have siror 'rti (headache) in kvin. What Panini achieves by the use of bahulam he achieves by the use of prayena also. In his sutra 'pumsi samjnayam ghah prayena 83, he observes that in the speech of the sisfas there are t forms with gha in the sense of karana and adhikarana signifying a noun in the masculine. Here too he does not specify the bases. They are to be picked up by an extensive study of literature. Thus we have ghatah, jayah (asvah), smarah (kamah), tvacah along with the more common dantacchadah, akarah, alayah. In another place Panini reads 'sese 84 which is both an Adhikara-sutra and a Vidhi-sutra. This is how it is understood and interpreted by the commentators. As a Vidhi-sutra it lays down that wherever in the speech of the cultured formations with an n are seen in senses other than apatyadi noticed in the foregoing sutras and jatadi in the sutras subsequent to it, they are to be deemed correct, although there are no pertinent grammatical injunctions. By a single stroke of the pen, Panini gives sanction forms as sravanah to a number of such forms as sravahah sabdah (sound grasped

by the ear), aulukhalo yavakah (lac pounded in a mortar), asvo rathah (a chariot drawn by horses), etc. noted by the Vrttikara. Besides, there are so many others, kaunapah (eater of a carcase), n. smartah (dictated by the Smrti), caksusah (an object of sight), vaitarani (crosses with the help of charity) and brahmanah (one who knows Brahman). There could not be a more convincing evidence of Panini's regard for usage than his framing of the sutra 'udicam mano vyatihare 85. The people of the north used an expression apamitya (apamaya) yacate which is palpably wrong and in no way justifiable. Ktva is used here in the absence of paurvakalya which grammar does not permit. Both the Vrttikara and the Kaumudikara agree in declaring that the sutra is to provide for the use of ktva in the case of apurvakala (absence of priority of action). It is only after a Brahmana has got something by begging that he seeks to barter it for something else. Naturally, therefore, yacitva pamayate is what is reasonably expected. Panini accepts apamaya yacate out of sheer regard for usage. But for it, he would have rejected it forthwith. There is another equally strong evidence of Panini's regard for usage. In framing the sutra 'tadarham 86, he uses tad in conformity with the practice before his time. He is so carried away by it that he does not observe his own rule 'kartrkarmanoh krti 87 which enjoins Genitive. How do we know that this was the usage in the case of krdanta from arha? There is widespread evidence in the literature of the sistas to this effect. Let us quote: 1. na parityagam arheyam matsakasad arindama 88 2. indratvam arho rajayam tapasety anucintya vai³9 3. vrthamaranam arhas tvam vrtha dya na bhavisyasi⁹0 4. viddhyausanasi bhadram te na tvam arhos "smi bhamini" 5. arhah purur idam rajyam yah sutah priyakrt tava⁹2 6. arhas tvam asi dharmajna rajasuyam mahakratum⁹3 7. tvam eva rajasardula samrad arho mahakratum 8. tasmat pravaranam purvam arhah partho dhananjayah⁹ 9. sudhanvan na tvam arho si maya saha samasanam⁹ 10. sakhanagaram arhas tu sahasrapatir uttamah⁹ 3

Sanskrit Usage 39 11. arhas tvam asi kalyanam varsneyam srnu yat param⁹8 12. tada visargam arhah syur itidam dhatrsasanam⁹ 13. lokan arho yan aham tan vidhatsva¹00 14. gatram hi te lohitacandanarham kasayasamslesam anarham etat 101 15. naivarhah paitrkam rktham 102 That there are here and there departures from this established usage does not affect its prevalence. By sanctioning the form aparasparah by his sutra 'aparasparah kriyasatatye 103 Panini accepts aparasparah as a kriyavisesana, an adverbial adjunct, implying continuity of action. He confirms his belief in the supremacy of usage. As a grammarian through and through, he would be the last person to permit the use of a kriyavisesana in its Nominative. He would have all kriyavisesanas used in the neuter singular of the Accusative without distinction. In fact, this use of aparaspara is a remnant of the Vedic usage in respect of the kriyavisesana. We have numerous instances of it in the Vedas. They are too wellknown to the Vedic scholar to be listed here. A quotation or two would suffice. The Satapatha Brahmana reads purvo rajno bhivadati.1 The Gopatha reads vayam purve svar esyamo vayam purve. 105 By his sutra 'prsodaradini yathopadistam 106 Panini says that all so-called irregular formations such as prsodara with unaccounted for elision, augmentation and substitution are to be recognised as correct when met with in the speech, oral or written, of the sistas. prsodara (the abdomen of a spotted antelope), c. p. though an irregular formation made up of prsat and udara with the dropping of t, has to be accepted for it is there in the speech of the sistas. balahaka, a cloud, has to be accepted in preference ap to varivahaka. Indeed the regular varivahaka cannot replace balahaka. It cannot yield the sense of a cloud. It can only signify a water carrier. Similarly jimuta is used for a cloud which obviously stands for the fuller jivanamuta but jivanamuta would not convey the sense of a cloud in which sense the word jimuta has got to be used. It can stand for a cast of water, a water -jar. Such is usage and Panini has unstinted regard for it.

REFERENCES 1. Asvamedhikaparva, 32.16. 27 2. Rgveda, X. 34.4 3. ibid., II. 23.16. 4. Atharvaveda, XI. 3.21. 5. ibid., VIII. 6.1 6. Satapatha Brahmana, 3.9. 3-20. 7. MBH., Udyogaparva, 29.30. 8. ibid., 72.18. 9. ibid., Vanaparva IV. 7. 10. ibid., II. 47. 11. ibid., Virataparva 32.3. 12. ibid., Dronaparval III.14. 13. ibid., Striparva, II.25. 27 14. Caraka, Indriyasthana, V. 15. 15. Rgveda, I.17.3 16. ibid., II.16.6. 17. Atharvaveda., III.13.6. n J J 2 18. Sukla Yajurveda, 7.15. 19. ibid., 8.37. 20. MBH, Adiparva 125.25. n 27 21. Mahabhasya, 2.2.29. 22. MBH, Udyoga/94.51 arvan, 23. Apastamba Dharmasutra, 2.1.3 24. MBH., Udyogaparva 29.6. 25. ibid., Dronaparval 55.36. 26. ibid., Karnaparvga, 44.27. 27. ibid., Anusasanaparva/38.25. 28. Vamanapurana, 21.32. 29. Bhagavatapurana, IX.18.37. 30. Ramayana, VI.57.18-19. 31. Naisadhacarita, III.93. 32. Sisupalavadha, I.29. 33. Pan. II. 3.65. 34. V.1.117 35. Ramayana. 1.53.2. 36. MBH., Adiparva 19.12. 212 n an n n 37. ibid., Udyogaparva! 7.17. m

38. ibid., 7, 13.32. 39. ibid., Santiparva, 267.15. 40. Manavadharmasastra, 9.144. 41. II.25.1. 42. Rgveda, VII.86.2 43. I.170.5. 44. XI.25.1 45. IV.99. 46. IV.10.4. 47. 4.18. n Sanskrit Usage 41 48. II.1.9. 49. Aitareya Brahmana, 3.2 50. Maitrayani Samhita, IV.1.1, Kathaka Samhita, 30.10. 51. Satapatha Brahmana, III.1.1.10. 52. MBH., Adiparva, 63.5. 53. ibid., Vanaparva 35.30. 54. ibid., 54.27. 55. ibid, Santiparva, 180.47. 56. Mudraraksasa, VI.16. 57. ibid., Act I, after verse 20. 58. Ramayana, III.47.33. 59. ibid., III.49.14-15. 60. 5.2.94 61. MBH., Dronaparva 32.3. 62. ibid., Salyaparva, 24.30. J^5 | 63. ibid., Sanitparva, 11.7. 64. Manusmrti, VI.76. 65. 5.1.103 66. Sutrasthana, 26.12. 67. Cikitsasthana, 5.84. 68. 3.1.102 69. 3.1.199 70. 3.2.97 71. 3.2.98 72. 3.2.99 73. 3.2.100 74. 3.2.10 75. 3.2.177 76. 5.3.14 th 22 27 22 22. 22

. 5.3.7 78. 5.3.10 79. 5.3.14 80. 2.15.7 81. 3.3.13 82. 3.3.108 83. 3.3.118 84. 4.2.92 85. 3.4.19 86. 5.11.107 87. 2.3.65 88. Ramayana, 1.53.12 89. Mahabharata, Adiparva 63(4) n/x 90. ibid., 156-26 91. ibid., 85-31 92. ibid., 85.31 93. ibid., Sabhaparva 13.32 n 94. ibid., 33.23 95. ibid., Udhyogaparvas 7.17 n 96. ibid., 35.15 97. ibid., Santiparva(87-8 th 98. ibid., 210.13 100. ibid., Anusasanaparva 71.17 aa 99. ibid., 267.15 101. Buddhacarita, 10.24 102. Manusmrti, 9.144 103. 6.1.144 104. Satapatha Brahmana, 3.1.4.14 105. Gopatha Brahmana, purva, 4.23 106. 6.3.109 purva, cap