UDBHATA'S COMMENTARY

ON THE

KĀVYĀLAMKĀRA OF BHĀMAHA

ISTITUTO ITALIANO

PER IL MEDIO ED ESTREMO ORIENTE

SERIE ORIENTALE ROMA

SOTTO LA DIREZIONE

DI

GIUSEPPE TUCCI

VOLUME XAVII

LA REDAZIONE DELLA SERIF È CURATA DAL PROF ANTONIO GARGANO

ROWA I w r o 1962

DEPARTMENT OF ARCHAEOLOGY OF PAKISTAN and ISTITUTO ITALIANO PER IL MEDIO ED ESTREMO ORIENTE

LITERARY AND HISTORICAI DOCUMENTS FROM PAKISTAN

1I

RANIERO GNOLI

UDBHATA'S COMMENTARY

ON THE

KĀVYĀLAMKĀRA OF BHĀMAHA

WITH AN APPENDIX BY MARGHERITA TATICCHI INCLUDING SOME FRAGMENTS OF KĀLIDĀSA'S RAGHUVAMSA

POD

ROMA ISTITUTO ITALIANO PER IL MEDIO ED ESTREMO ORIENTE 1962

TUTTI I DIRITTI RISERVATI

A BHAVAN (BOMBAY-7 141521 BH LIBRAR

Printed in Ftaly - Stampato in Itala.

PREFACE

As the Introduction gives all that I have at present to say on these fragments, there now remains only the welcome duty of acknowledging the help received from various sources, Prof. Raghavan, Prof. Renou, Dr. Dave and Dr. Pandeya have been kind cnough as to read the proofs of some fragments and to aid me by many useful suggestions. But my thanks are due above all to Prof. T. Venkatacharya. He has been good enough as to read the proofs of all the fragments and to his vast Anowledge of Alamkarasastra and Grammar I owe many valuable remarks and the identification of some passages quoted mn the text

Rome, March 1962

APPARATUS

The critical apparatus has the following:

( > pointed brackets indicate the editor's addition or emendation.

Italics indicate uncertain readings.

× a cross indicates a lost aksara.

. a dot indicates the loss of part of an aksara. ] [ square brackets mark lost beginnings and ends of lines.

three dots indicate the loss of a number of superior or inferior lines in the leaf

The recto and the verso of the leaves has been marked by the letters a and 6 respectively. Obviously, in the fragments incertae sedis this distinetion is merely conven- tional. Critical notes do not refer to the actual lines of fragments, but to the serial numbers of lines given in margin.

CONTENTS

PREFACE VI

APPARATUS VINI

REFERENCES AND ABBREVIATIONS X

INTRODUCTION

1 The Manuscript XIII

2 Autorship etc XYI

§ 3 Generalities XIXVII

§ 4 The Text of Bhamaba XIXIX

ADDENDA XLIII

FACSTMILES between pp XLIII and 1

TEXT (a) Fragmenta Incertae Sedis 51

APPENDIX 81

REFERENCES AND ABBREVIATIONS

Alamkaracudamani = see HC (Hemacandra). AS = The Alamkarasartasta by Ruyyaka (the Vrtti perhaps by Mankhaka); ed. by G. Dvived1, KM, 35, Bombay 1939 (with the commentary of Jayaratha); and by K. Sambasiva Sastri, TSS, XL, Trvandrum 1926 (with the commen- tary of Samudrabandha). Bhamaha = Edd. ete. of the Kdt yalamkara quoted to p. XXXIX, q v. Bodhicaryatatara = Bodhicaryaratara of Santideva, with the commentary (Panjika) by Prajfakaramati; ed. by L. de la Vallee Poussin, Calcutta 1901. BSPS = Bombay Sanshrit and Prakrit Series. Candisataka == the Candisatala by Bana, ed. by G. P. Quacken- bos, The Sanshrit Poems of Mayura, etc., Columbia Uni- versity Prees, New York 1917. DhA = Dhtanyaloka of Anandavardhana with the commen- tary Locana of Ablunavagupta, KM, 25, Bombay 1891 (4th ed. 1935); KSS, 135, Benares 1910. The first chapter with the sub-commentary Koumudi of Uttuugodaya ed. by S. Kuppuswami Sastr, Madras 1914. See alo the ed. with an elaborate English exposition by B. Bhattacaryn, Calcutta 1956; and the translation by K. Krishnamoorthy, Poona 1955. EW' = East and W'est, Rome. COS = Garhtcad's Ortental Series, Baroda. Harsacarsta = Harsacarita of Bana, NSP, Bombay 1916. HC = Hemacandra, the Jain author of the Karyanusasana with the commentary Alamharecudamani and the sub- commentary Tireka, ed. by Rasiklal C. Parikh, Bombay 1937 Also, KM. 71, 2nd ed., NSP Bombay 1934.

[ x ]

. REFERENCES AND ABBREVIATIONS

HSP = The Histary of Sanskrit Poettes, by P. V Kane, 3rd. ed., Bombay 1951. Indische Spruche = Indische Spruche by Otto Bohtlingk, St. Petersburg, 1870-1873 IHQ = Indian Historical Quarterly JOR = Journal of Oriental Research, Madras. JRAS = Journal of the Royal Asiatic Society Koumudi = The Kaumudi Commentary to the Dhranyaloka by Uttungodaya See Dhranyaloka. KASS = Kavyālamkārasārasamgraha, qv. Kavyalamkarasarasamgraha = by Udbhata; ed by Narayana Daso Banhatti, BSPS, LXXIX, Bombay 1925 (with the commentary Laghuirtt of Induraja); and by K. S Rama- swamı Sastri Siromanı, GOS, LV, Baroda 1931 (with the commentary Vart of Tilaka) Kāvyaprakāsa = The Kāvyaprakāća of Mammata, with the commentary (Samheta) by Manikyacandra, ed. by R. Shama Sastry, Mysore 1922 See also Anandasrama Skt. Series, 89, ed V Abhyankar, Poona 1921 Kıratarjuniya = Kiratarjuniya by Bharavi (with the commen- tary of Malhnatha), NSP, 1942. KSS = Kashı Sanskrit Series (Haridās Sanskrıt Granthamāla). Locana = See DhA (Dhvanyaloka) KM = The Kavyamala seres, pubhshed by the Nirnaya Sagar Press, Bombay, actually two series respectively numbered 1-14 for short works and 1-95 for Jonger M The recension of the Raghuvamsa of Kaldasa according to Malhnatha MaBha = The Mahabharata for the first time critically edi- ted by V S Sukthankar, Poona 1933 sqq. MBh = The Mahabhasya of Patanjalı, NSP, Bombay N = The Raghuvamsa of Kaldasa, ed by G R Nandargıkar, 3rd. ed Bombay 1897 NSP = The Nirnaya Sagar Press, Bombay Nyayamanari - The Nyayamanjari by Jayantabhatta, KSS, 106, Benares 1936 P = Pamin1. { xI ]

REFERENCES AND ABBREVIATIONS

PV = The Pramanavarttika of Dharmakirt with a commen tary of Manorathanandın, ed by Rahula Sankrtyayana Appendix to JBORS, voll XXIV-XXVI, 1938-1940 The first chapter with the suto-commentary ed by R Gnoh, SOR, Roma 1960 Ramayana = The Ramayana, critically edited by G H Bhatt, Baroda 1960 Ratnaral The Ratnaval by Harsa, NSP, Bombay 1938 Siddi = Vynaptimatratasiddhi, La Siddhs de Huan-isang, iradunte et annotee par L de La Vallee Poussin, Paris 1928 SOR = Serte Orientale Roma SP = History of Sanskrit Poetics, by S K De, 2nd ed , Cal cutta 1960 Srngaraprakasa = The Srngaraprahasa of Bhoja, ed by G R Josyer (the first eight chapters only), Mysore 1955 ŚrP = The Śpngaraprakasa, q v , V Raghavan, Bhoja's Śrngara Prakaśa, Bombay 1940 Subhasitarah = Edited as of Vallabhadeva by P Peterson and

ŚV Durga Prasada, Bombay 1886 = The Stsupalavadha of Magha, NSP, 1940 TSS = Trtandrum Sanskrit Series UV = The Udyanatrth, a valuable sanskrit modern commen tary to the Kavyalamkara of Bhamaha by D T Tatacha rya Siromam (see p xxxIx, n 2) Vamana = The Katyalamkarasutra with the Vrtti, by Vamana, NSP, Bombay 1953 Vweka = See Hemacandra Vyaktwweka = The Iyaktwvieka of Mahimabhatta, ed with a commentary of Ruyyaka, KSS, 121, Benares 1936 ZDMG = Zeuschrift der Deutschen Morgenlandischen Gesell schaft

INTRODUCTION

THE MANUSCRIPT

Two years ago there have come to hght in the vicinity of Kafirkoth a certam number of fragmentary leaves and mnor fragments (many of them extremely mmnute and contaming only a few aksaras) of a manuscript mn birch- bark, written in sarada characters The period to which ths MS belongs is probably the 9th-11th cent These fragments have been kindly sent by Dr F. A Khan, Director General of the Archaeological Depariment of Pakistan to Prof G Tucc for restoration -- as they were extremely brittle-to be undertaken by the Institute of Book Pathology (Istituto di Patologia del Libro) On my first exammning them after they were restored, I reco- gmzed two different works, that is a commentary to Bha- maha's Kavyalamkara, whch 1s, in all probability, the Udbhata one which was lost, and the Raghuvamsa of Kal- dasa The fragments contaiing the Raghuvamsa, tran- scribed by Miss Margherta Taticchi, have been described and edited by her as an appendix to this volume This is the earliest manuscript known so far of the work, it seems So far as the comment to Bhamaba is concerned, we are not in a position to say anything as to the number of leaves which orginally made up this manuscript. The [ xIII ]

RANIERO CNOLI

only leaves where, on the left hand margin, we still find the mark of numeration, are the 4th an the 47th which cor- respond, in the Bhamaha text, to the comment on stanzas I, 9-12 (Fr. 6), II, 69-71 (Tr. 24). The size of leaves was originally about 32 x 9 cm. The right hand side of all the leaves - about one half - had been destroyed by fire and hence lost. The number of lines per page is 8 and in one leaf only, 9. Each line must have contained about 40 to 60 aksaras. The script, as we have already mentioned, is sarada of the 9th-11th cent. A more precise dating is difficult, the Sarada being a kind of writing which under- went only slight modifications. Certain fragments are writ- ten on a face only. Apparently, not all the leaves are written by the same hand. The ink of the first chapter is rather weak and the writing is smaller and less regular than that of most of the second chapter. The writing of the third chapter is comparatively smaller than that of the preced- ing ones, and in turn is at variance with that in chapters V and VI which is more rounded aud bolder. The manuscript is rather uncorrect. The signs for numbers are the same as those occurring in the MS of the Sanghabhedatastu - being a part of the Vmaya of the Mulasarvastivadin - from Gilgit, also written in sarada script, which is now being worked out for publcation in the Is.M.E.O. Margin- al additions are few. Frequent on the other hand, the interhnear ones and aksaras perhaps corrected by a second hand. The poor state of preservation of a large part of these fragments did not always permit me to read them with sufficient certainty, especially where the meaning was not clear to me. Moreover, the sarada script is one which leads to frequent confusion in certain groups of signs, e.g., dhya and vya; nta rta. and tta; nva and ndha, tu and

[ xIv ]

Udbhata's Vivarana, etc

bhu, stha and sva, ddha and da, nna and rto, dr and sr, rda and nda, and occasional confusion in others, eg, hta and rta, ta and bha, ga and sa, pa and dha, rna and la, ete The consonants b and v are denoted by the same sign The tisk of reconstruction of these fragments, which have come to us in no specific order whatever, was very long and all the more difficult masmuch as under the effects of fire and time, certam pages had become superim posed, and, as it were, merged mn each other In some cases it was necessary to divide the page of bhurjapatra m half through its thickness (in the meantime, all the fragments had been laid between two very fine layers of cellulose to prevent them from dismntegrating entirely) to see whether it was not really a double or triple page, as had repeatedly occurred with certai fragments Lastly, I am fully aware of the precariousness and uncertamty lying in several pomts of this work which have however had a counterpart in the satisfaction of having done my best in the five or six months I have spent mn the company of Udbhata and Bhamaha I have spared no efforts in trying to make the tmy fragments tally - many of the present fragments are made up of about ten lesser ones - with the help of sunhght in the summer, and of a powerful reflector during the winter, in order to rescue from fire damage and time the greatest possibie amount of aksaras In the pomts where the manuscript was hardest to read, or the reading uncertain or utterly hopeless, I have had no hesitation in merely transenbing the apparent (or assu- med) outhnes of the characters, without any attempt whatsoever at integration or at mterpretation, in such cases impossible, or doubtful at best Such passages are transcribed mn italcs [ xv ]

RANIERO GNOLI

§ 2. AUTORSHIP, ETC.

As already intimated, the work contained in this MS is a commentary on Bhamaha's Kavyalamhara, to be iden- tified, to my mind, with the lost vivarana of Udbhata who, so far as we know, was Bhamaha's only commenta- tor. Indeed, not only was Udbhata Bhamaha's only com- mentator, but some passages of the present commentary seem to have been the source of some lmes of Hemacan- dra's Alamkaracudamant and Vieka1. The great Jaina scholar had, as it is Anown, a certain tendency to copy and paraphrase works of earlicr authors, and in the afore- mentioned passages also, it is likely that he based himself on a well-known and authoritative work, such as the com- mentary of Udbhata, rather than on some obscure comment, of which no trace has come down to us 2). The few frag-

1. Cp Fr 22 b, 11 4, 5, 8, 23 a, 1 5, 40 a, 1 8 See also btlow, pp XXXDI, XXXVII The character and the sources of the Kapyalamkara with its two commentanes Alamkaracadamant and Vireka have bren disenssed at lenght by Sivaprasad Bhattacarya in his paper Hemacandra and the Elerenth Century Kashmir Poett custs, JAS, XXIII/1957, pp 117-129 As we shall see, one of these sources, which, of course, the learned author of thuis article could not known was Ud bhata's Vivarano ) The identity of this commentary with that of Udbbata has been doub- ted by V Raghavan It is worthwhile, in this connection, to quote what he said on the occasion of The Twenty first All-India Oriental Conference mn his presidential address delvered in Snnagar during the 14th, 15th, and 16th October 1961 (see the Address by V Raghavan, pp 17-18), (each graphical pe cnhanty of the onginal has been farthfully reproduced) "This is - he said - the only commentary on Bhamaha fonnd anywhere and from our available Alankara literature we know there was only one com- mentary on Bhamaha and that by Udbhata of Kashmair One is therefore m chned to take this commentary as Udbhata's and Gnoh clauns that this is *without any doubt the lost ritarana of Udbhata' [m a private commumcation addressed by me to Prof Raghavan] If it is so, ths would be perhaps

[xVI ]

Udbhata's Vivarana, etc

ments of this commentary from which a complete meaning can be made out are in addition in perfect agreement with the theories and the language of Udbhata, such as they appear in later tradition and m his minor work, the Kavya- lamkarasārasamgraha Let me call the attention of the reader on some particular pomts

one of the foremost discoveries of our times But having examined all the fragments I find it is not posyible to assert that it is Udbhata s There are some clear tests to ident fy Udbhata s Bhamahavivarana and as these frag ments are meagre we can apply only one of these tests In one of the frag ments, numbered 4 [the present no 10] by Gnoli, we have Bhamaha I 9 śabdaś chandobhidhanarthah itihāsasrayah kathai loko yuktıh kalas cetı mantavyah kavyahetavahll This i9 a crucial verse so far as Udbhata's commentary is concerned Stu dents of the Dhvanyaloka know that when elucidating at the beginning of the work the text bhaktam ahus tam anye and its orm Abhinava gupta quotes from Udbbata s Vivarana on the verse given above and Abhi nayagupta says here that to explain why Bhamaha after saying sabd ab first said agan abhidhana in abhidhanarthan Udbhata inter prets the second word abhidhana not as word but abhidhavya para, the significatory capacity of a word which Udbhata adds is of two kinds Mukhya and Gunourit This is a very vital context as Udbhata shows here his acceptance in Kavya of a Vrth other than the Mukhya Unfortuna tely in the fragments discovered i the Swat valley the gloss on th s verse does not offer this wellknown comment of Udbhata Instead, the commentary In the fragment says the oppos te ie taking abhidhana normally as sabda abhidhanartha sabdanam arthah which i a serious obstacle to Gnolis proposed identification But if Udbhata had first given the straight reaning and then given as an alternative or better interpre tation what Abhmavagupta has quoted and if mn the broken up portion of the leaf we have lost this second interpretation we may then at least tenta tively take this as the Bhamohourarana of Udbhata To my mind the fact that in these fragments the words cited by Abhi navagupta are lack ng is not conclusive since, as Prof Raghavan observes they could well belong to the missing part of the leaf Literal quotations from Udbhata a commentary are, as we have remarked extremely few and it 1s far from surprising if none of them occurs in these fragments On the other hand notwhistanding the contrary opimion of V Raghavan the remarns of thus fragment clearly support as we shall see, Udbhata s views

{ xvn ]

RANIERO GNOLI

a) The fragmentary comment to the stanza sabdas chando 'bhidhanarthah, etc (1, 9, see Fr 10), which, according to V. Raghavan, is, as we have seen, a serious obstacle to the attribution of this commentary to Udbhata, to my mind, is rather in support of it According to this commentary, Bhamaha's words abhidhanarthah require to be explamed as sabdanam arthah1 and sabdanam abhidhanam. Words, in other terms, have a meaning and, of course, between these two there must be a connection, sambandha The pur- port of a bhidhana is just that of ensuring this conne- ction Words and meanings unconnected are not perceived and there must therefore be some capacity, implicit m the words, which connects them with their respective senses This is, according to the commentator, the reason why the word abhidhana has been mentioned m the stanza The words quoted by Ablnavagupta - ś a b d a - nam abhidbanam abhidhavyaparo mu- Khyo gunavrttis ca - occurred probably in the missing part of line 4, as an explanation of a bhidhan a The meaning of abhıdhana is abhidhavya- para, ie, power or capacity of denotation, and, in this power two different aspects or functions may be di- stingushed, namely, mukhyavrtti and guna-

1. In other words sabdanam arthah is to be considered as a comment of artha only and not of abhidhanarthab as it seemed to Prof Reghavan (above, p XvI) The meaning of artha, accord ing to the commentator is in this case, the well attested one of abhi- dheya In a similar way artha w explaiped as vastu in the Fr 50 a 1 5 See in this conaretion the Amarakofa av artho 'bhidheya- ratvastuprayojananıvrttışu The term abbidhana was explained on the other hand m sebdanam abhidhanam (ep ifra)

[ XVIII ]

Udbhața's Vivarana, etc

vrttr, viz a primary and a secondary function This supposition is in fact supported by the subsequent hines In them, indeed, the mukhyavrtti or primary capacity of a word is exemplified by the sentence par vatād eti devadattah, and the secondary or figurative power by the well-known verse eti jivan- tam ānando naram varşagatād apı". Whereas in the first mnstance the verb eti mantains its pnmary value, in the second case it is used m a figuratrve sense, since the root i in its primary sense (in gatau) cannot be construed with such abstract things as a n an da The third example is somewhat more complex We know that, according to Panim, the application of the suffix Atva requires that the two actions involved have the same agent and that the action indicated by the root to whch ktva is added temporally precede the other ") This rule, generally observed, sometimes is however appa- rently violated This is the case with a stanza borrowed from the Kiratarjuniya, III, 21, which runs thus 3) nırıkşya samrambhanırastadhaıryam rādheyam aradhıtajamadagnyam | asamstuteşu prasabham bhayequ jayeta mrtyor apı pakşapatah || How to explain this irregularity> According to Mahi- mabhatta and Mallmnatha", we here have stmply a case of

1. Ramayona Sundarakapda V 34 (iee \I 126) D P IIÍ 4 21 samanakartrkayo purvakāle ") This is I terally translated " Having seen the son of Ridbt who had propitiated the dencendant of Jamadagnt and bad pushed back the firmness (of the enemirs) because of anger incl pation to the unfamubar fears would ruddenly be born even of Death bimself *) Cp the )yakteireka p 28 Mallnatha whule comment ng on thi stanza observrs atra janikriyåpekpayā samšnakartrka

[ xIx ]

RANIERO CNOLI

constructio ad sensum. The word mrtyoh is to be taken as a subjective genitive of pakşapata. The last and not jayeta is, as a matter of fact, the action sub- sequent to that of seeing, just as if the poet had said niriksya .... mrtyuh pakşe patati. In this case, mrtyu is obviously the subject of niriksya and of patatil. Udbhata takes a quite different view. First of all, it is impossible, he says, to maintain that the action sub-

tvābhāve 'pi pakşapātalrıyāpekşayā tatsambhavan nırıkşyeti lyapnırdesahl ") Pnor to Mallinatha, this stanza, as we have smd, is cited by Mahimabhatta also. This is one of the instances adduced by him in support of the theory according to which action alone is the fandament of the use of nouns for objects. In this case, indeed, the actions preceding that indicated by the root to which ktva is added, are expressed by two nouns, namely bhaya and pakşa- pata. To the objection that action can be proper to sentient things only and not to insentient things - as, f.i., a jar, or, as in this case, bhaya and paksapata -, he answers that this is not true and action can be pro- per to inanimate berngs also. Ths answer is based, as bis commentator Ruyyaka, explicitly remarks, on the tenets of Kashmir Shivaism, according to which mnammate things also are really consisting of consciousness and liberty (pp. 33-34). According to Mahimabhatta the gunsvrtti does not exist. A word can have one fnction only, that as the literal one The so-called secondary or meta- phorcal function supposes necessarily two or more words - therefore, a parti- cular context - from which it 1s inferred. In other terms, it resolves itself in a celation of sadhya an sadhana and is then to be considered as a form of anumana. The Vyaktieka deserves more attention than it has been given. Malimabhatta's critics to the doctrine of dhyani are sometimes very penetrating and intelligent. I share completely the opmion of Mr. Kane (HSP, p. 243) that "his work is one of the masterpieces of the Alamkara Laterature and deserves to be saved from the unmerited oblvion in which it has fallen. His work contains brilhiant arguments and exhibits great erndition, logical acumen, fastidious critieism and deep insight ". Re- cently, the views of Malimabhatta have been expounded by K. C. Pandey in lus bool on Indian Aesthetics (Banaras, 1950), pp. 270 sqq , and in the article Kashmir Saita Tendencies of Mahima Bhatta, Bharatiya Vidya, XI/1950, pp 187-94. see also the paper hy B Bhattacharya, Suggestion tersus Inference In Sanskrut Aesthetics (Indian Culture, Vol. XIII, No. 1)

[ xx ]

RANIERO GNOLI

These words seem to be a paraphrasis of Udbhata's text They occur, indeed, in a similar form, mn II 7 and 8 of our fragment

Lıne 7 kartrbhedān nāstı ktvāpratyayah | atha pa- kşapatanakriyato X X X Xkrıyax x xtvam tada (about 15-20 aksaras lost) Line 8 ghatate | sasthyabhıhıtasya kartuh paksapā- takrıyayaıva sambandha (ıs y a(te ) na) mırıksana(Lrı- yaya) (about 15-20 aksaras lost)

With respect to these lines the text of Bhoja is somewhat abbreviated and perhaps mutilated After the first alterna- tive, ending with the words ktyapratyayah and before the words abhidhavyaparas tu vi- ghatate the sense would require, mdeed, yadi ca pakşapatakriyato niriksanakrıyay ab pūrvatvam tadā ktvāpratyayah suta- rām upapadyate | abhıdhāvyāpāras tu vighatate | or something like that Let us now see Udbhata's text The words preceding 1 7, at the end of l 6, must have been sımilar or identical to the first alternative mn the Bhoja's text The sen- tence atha pakşapatanakrıyāto x x x x kriyax x xtvam tada, missing in the Bhoja's text, can be conjecturally restored as atha paksapa- tanakrıyato (nırikşana) krıya(yah pūr- va)tvam tada, etc, or something similar In the ending of lne 8, was perhaps contamed Panini's state- ment as ret forth by Bhoja in support of his opinion This discussion, no doubt, may appear to the modern reader somewhat abstrase and pedantic, especially regard [xxu ]

Udbhata's Vivarana, etc.

ing the third example. Udbhata's interpretation cannot be shared in any way and all the reasons are on the side of Mallinatha. The importance of this passage, however, consists in the mention of the transferred meaning. In Indian thought, the origins of a secondary or transferred meaning are, as a matter of fact, very ancient and pro- bably to be traced to some ritual speculations. As to certain texts, one perceived that a litteral interpretation could be scarcely satisfying. Their words, therefore, must have another significance, transferred or metaphorical 1. This conception, in its turn, was inherited and refined by various thinkers and became a common property of Indian philosophy and linguistics 2. Udbhata by introducing into poetry the secondary function of words - which is, as Bhoja says, the very life of vakrokti, that is, the curved or oblique diction, peculiar to poetic language - let open the door to the con- ception of a third potency of language") - the vyañja-

1. See, f 1., the Sabarabhasya, 1, 4, 23 sqq , III, 2, 1, etc. The difference between the primary and the secondary sigmificance has been discussed at lenght by Aumarila, in his comment (Tantravarttika) on the afore-mentioned work, q v. *) Vide the Mohabhasya, I, 1, 15 (but ef also VIII, I, 12) and the Vakya- padtya, II, 252 sqq, (ef., in this connection, also D. Seyfort Ruegg, Contribu- tions a I'Histoire de la Philosophie linguistque indienne, Paris 1959, pp 27-38)- The distinction between a primary and secondary significance is of course refuted by the buddbist schools, according to whch all language is metaphorieal (see, f. 1, Vasubandhu, Trunfika, 1 and Sthiramati's commentary; Siddht, I, pp 84-89, Dharmakirti, Pramānavārtika, III, 35-37, 150-153) a) ŚrPr, VII, ? abhidheyavinabhutapratitir lakpane+ tı yā | saıgā vidagdhavakrotijīvitam vrttir işyateg This stanza has been discussed by V Raghavan, SrPr. pp. 134-138 I am not fully convinced that this floka is truly by Bhoja, as V. Raghavan seems to be- heve (p 137, n 1) The tad uktam, which precedes it, makes us pause, all the more that the preceding stanza, likewise antroduced by tad aha, is borrowed from Dharmakirh, PV, III, 37 The first hmne, abhidheyavi- nabhuta, etc, is, anyhow, taken from Kumarila Vide below, p. XXXIV.

[ XXIII ]

RANIERO GNOLI

navrttr -, different both from the primary and the trans- ferred sense of words, which characterizes true poetry 1. The ideal progress of this conception is quite logical and starts, as it were from the very awareness of the strument ality and transitivity of ordmarv language, the nature of which has been understood and analysed in India more than elsewhere "People ", Dharmakirti says, " do not have such a delusive fancy, as to stay in a state of misery, without fixing conventions and using words And what then ? Only that activity which produces some fruits is worthy to be assumed, for every fruitless activity deserves to be abandoned. Therefore, he who apples words to something, logi- cally does that in view of some fruit And this fruit is twofold, namely characterized by grving up or obtaming that which we do or do not desire Then, after having known that which we do or do not desire, we apply words or are attentive to their application, with the very view of determining or diverting one's own or another's activity towards that which is or is not a means for the desired things Otherwise, Jan- guage deserves to be neglected " 2)

The poetical sense could not be compatible with the practical one. Poetry - it was percerved soon enough - 18 characterized by a different lingustical attitude, which, it is true, makes use of the same words of practical speech,

") On the origtn of the theory of suggestion see the penetrat ng article of K Kunjunn Raja The Theory of Suggestion in Indian Semantics Adyar Library Bullenin Vol XIX part 1 2 pp 20-26 1) Dharmaktrts Pramanararttıkam I comm to stanza 92 See below I xxx n.

[ xxn ]

Udbhata's Vrvarana, etc

but for a different purpose In poetry, words are not sudd- enly replaced by their sense, do not exhaust themselves in fruit and comprehension, but remain, so to speak, alive before us, in our knowledge, without dying out in the kno- wable "The expected fruit of the poetical words ", Abhı- navagupta says, "is not a cognition ordered to some practical action, as, for instanee, brmnging or taking away something, but a cogmtion, which is the source of a repose on the cogni- tion itself, and which, as such, does not consume itself mn the intended thing, but rests on the very intention"1) The poetical speech, at the very expense of the practical value enrches itself with various proceedings - rhymes, figures, inversions - conceived as useless in a purely func- tional language, but essential in the poetical one One of the most important elements of these speech figures, diffe- rent from the modes of practical language dominated by a direct way of expression, was seen in the secondary function of words - already exammned and discussed in the schools of ntualistics and linguistics -, which was considered as the very life of poetical language, mn antithesis with the practical one Lastly, a scrutiny more penetrating yet foun- ded on this conception, revealed that the secondary function does not necessarely imply poetry Actually all language 1s metaphorical 2) The source of poetry must be then ano- ther sense or value that words assume, altogether different from the primary and the secondary one This new sense - the poetical sense - irreducible, as it is, to the literary one, cannot however do without it, but is, as it were,

) Locana p 442 (Benares Ed ) # Cfr above p xxIIt n 2 Also Locana p 282 (Madras Ed ) evam anaya lakşanaya pançavıdhaya vışvam eva vyaptam| etc [ xxv ]

Udbhaļa's Vivaraņa, etc. that, for them. it was comprised in the province of g un a - vrtti, and not considered as an independent potency of words. In short, Udbhata and Anandhavardhana along with his commentator Abhinavagupta are the starting and the final point of this trajectory. In some observations about practical and poetic language, one of the most sensi- tive critics of our times, Paul Valery, ideally connects him- self to Anandavardhana. "La poesie ", he says,

"est un art du langage. Le langage, cependant, est une creation de la pratique. Remarquons d'abord que toute communication entre les hommes n'a quelque certitude que dans la pratique, et par la vérification que nous donne la pratique. Je vous demande du feu. Vous me donnez du feu: vous m'avez compris. Mais, en me demandant du feu, vous avez pu pronon- cer ces quelques mots sans importance, avec un cer- tain ton, et dans un certain timbre de voix - avec une certaine inflexion et une certaine lenteur ou une certaine precipitation que j'ai pu remarquer. J'ai compris vos paroles, puisque, sans même y penser, je vous ai tendu ce que vous demandiez, ce peu de feu.

ximity, ete. [according to a stanza attnbuted to Bhartrmitra, the cause of Jakşana arc five, viz, abhidheyena sambandha, sadrsya, semavava, vaiparitya, and kriyayoga] Where this relation ia similarity there is vakrokti. Take for instance unmimila kama- lam sarasinam kairavam ca nimimila muhūrtat | Here unmimila and nimimila primanly signify 'opened and shut the eyelids'. Thi opening and shutting of the eyelids is incongruos in the lotos and hence through the simuarity of the lotus to the eye, they are taken in their secondary sense of blossoming and closing Thus poetic charm consists in the implication of some similarity in secondary sigmification . This implica tion whuich gives a pecuhar charm to the verse naturally falls within the sphere of dhvans or suggestion "

{ XXVII ]

RANIERO GNOLI

supported by it1). A truly poetical word or expression is that which cannot be replaced by other words, without losing its value. Poetry knows no synonyms 2. The fa- ther of this new conception, Anandavardhana, was fully aware of his indebtness to Udbhata. " He also, he says - and the reference, according to Abhinavagupta, is to Udbhata and Vamana -, who pointed out the secondary usage of words in poetry slightly touched the fringes of the doctrine of suggestion, though he was not defining it "3). Udbhata and Vamana, according to Anandavardhana, did not deny the esistence of dhvan1. Implicitely they admitted it. The only point of difference rests on the fact

) DhA, I. 9-12 -. 9 Just as a man interested in percerving objects (in the dark) direets hus efforts towards securing the flame of a larp since at 1s a means to realise his end, so also does one who is uluimately interested in the suggested meaning first evince interest in the conventional meaning 10. Just as the purport of a sentence is grasped only through the sense of individual words, the knowledge of that sense is got at only through the medium of Expressed sense. 11. Though by its own power the word-import conveys the sentence-import, just as it escapes notice once its purpose is served, - 12 so also that Suggested menning flashes saddenly across the truth-perceiving minds of cultured erities when they are indifferent towards the conventional meaning (tr by K. Krishna- moorthy, op eut " By the way. I am strongly convinced that both the kūrka s and the vrtt of the Dhranyaloka are of a same author, viz Anandavardhans *) DAA. I. 15 " DhA. I, 1 gupavşttyā kavyeşu vyavahåram darin- yata dhvanımārgo manakapreto' pi na lakyyate. Here I bavr followrd, with some shght modifications, the translation of K. KRisH- NAMOORTHY. Anandarardhana's Dhranyalokah or Theory of Suggestion in Poetry. translated into English wtth Notes. Poona, 1955, p 3 According to Abhinavagunta, the autborities whom Anandavardbana had in mind, arr here Udbhata and Vamana Udbhata's text was discusied at length in the preceding lines Vamana's textis sadrsyad laksana vakroktib, te. scondary signification having umilarity as its bans is vakrokts Irte here the words of A Sankaran. Some Aspeets of Literaty Crilcum in Sanskrit or The Theorses of Rasa and Phrans, Madras 1929, pp 62 " When the primaty sense of a word in a sentence is incongruon, we take it in ts secondary signification on the basis of errtain relation such as pro-

( XXVI )

Udbhata's Vivarana, etc

mes en possession d'exprimer sous d'autres formes l'idee que le discours avait composee en nous L'acte du langage accompli nous a rendus maîtres du pomt central qui commande la multiplcite des expressions posszbles d'une idee acquise En somme, le sens, qui est la tendance a une substitution mentale uniforme, unque, resolutoire, est l'objet, la loi, la limite d'exi- stence de la prose pure Tout autre est la fonction de la poesie Tandis que le fond unique est exigible de la prose, c'est ici la forme unique qui ordonne et survit C'est le son, c'est le rythme, ce sont les rapprochements physiques des mots, leurs effets d'induction ou leurs mnfluences mutuelles qui domment, aux depens de leur propnete de se consom- mer en un sens defin et certamn Il faut donc que dans un poeme le sens ne pusse l'emporter sur la forme et la detruire sans retour, c'est au contraire le retour, la forme conservee, ou plutot exactement reproduite com- me umque et necessaire expression de l'etat ou de la pensee qu'elle vient d'engendrer au leeteur, qui est le ressort de la puissance poetique Un beau vers renait indefiniment de ses cendres, il redevient, - comme l'effet de son effet, - cause harmonique de soi-meme" 1)

Let us now listen to some lmes of the great com- mentator of the Dhvanyaloka, Abhinavagupta " Aesthetical esperience takes place, as everyone can notice, by virtue, as it were, of the squeezing out of the poetical word Persons aesthetically sensible, indeed, read and taste many times over the same poem In

1. Varete Commentaires de Charmes p 1510 (ed ent ) [xxxx]

RANIERO GNOLI

contradiction to practical means of perception, that, their task being accomplished, are no more of any use and must then be abandoned1), a poem, indeed, does not lose its value after it has been comprehen- ded. The words, in poetry, must have therefore an additional power, that of suggestion, and just for this reason the transition from the conventional meaning to the poetic one is unnoticeable 2). What some people say, namely that a phrase would then come to have many different meanings, is due solely to their ignorance. A sentence - they say referr- ing to ordinary strumental language - which has heen pronounced once and the meaning of which has already

1. This is a quotation from the Vakyapadıya, II, 38 Apart from Bhartrhari, the practical purpose of language has been discussed at lenght and with an admirable penetration by Dharmakirti in his Promanavarttika, especially I, 92 agg Dharmakirti's work was well known to Abhinavagupta, who cites it frequently The afore-mentioned stanza I, 92 i9 eited twice in the Locano, pp. 444 and 542 (Benares ed ) ") When we read a poem we become, as it were, simultancously aware of the rasas, viz the sentiments not practically experienced but aesthetically contemplated, that it suggests. The temporal sequence between the cogmitions of the expressed and the suggested sense would be noticeable only in case the suggested sense were opposed to the expressed one or smilar to it, that is, on the same footing See DhA, III, 33 " But this temporal sequence in the two fanction of sounds cannot be noticed when sentiments [1 e , rasas] are suggested because sentiments are neither opposed to the expressed sense nor appear as similar to the nther senses, they are not capable of being conveyed by anght else and all their accescories work together with lightning-quckness " (transl cited above) The concept of alakyyakrama and the necesaty of admitting it has been expounded at lenght by Anandavardhana himself, Dh.A, IIL, 33 I cate here some words of P Valery (Varsete, Premiere Leçon du Cours de Poelique, ed ct , p 1356) " J'expliquerai un jour comment cette alteration se marque dans le language des poetes, et qu'il y a un langage poetique dans lequel les mots ne sont plus le mots de l'usage pratique et ubre Ils ne s'associent plus selon les memes attractions il sont charges de deux valeurs simultanement en- gagee et d importance equivalente leur son et leur effet psychique instantane". [xxx ]

Udbhata's Yivarana, etc been perceived by force of convention, cannot lead one to perceive two different meanings The subject, indeed, cannot remember, at the same time, several mutually contrasting conventions, and if, on the other hand, these conventions are not contrasting, the mean- ing of the sentence remains, then, one Nor it can be admitted that the different meanings are perceived one after another, because the words, after they have made one meaning perceptible and have thus ceased to be efficacious, have no longer any power to render per- ceptible any other meamng And even if the phrase is pronounced a second time, the meaning remams invariably the same, the convention and the context being the same Should someone object that a sen- tence can lead one to percerve another meaning, inde- pendentely of the one perceived through convention and context, it can be rephed that, then, there is no lon- ger any fixed relation between word and meaning, and that one falls thus into the counter-sense, described in the stanza "Therefore, what reason can one adduce for the fact that, on hearing the phrase ' He who desrres Heaven, must offer the fire-sacrifice', one does not percerve the meaning ' he must eat dog-flesh ' ? 1)"' Moreover, there would be no hint to the number of possible meanings and there would exist a general state of uncertainty The fact of admitting that a sentence can have several meanings is thus a fallacy " The case of poetical words is however different Here, indeed, the aesthetical expression, etc , once percerved, tends to become itself an object of aesthetic experience

1. This 1s a stanza from the Pramanaiarttaka I 318

[ xxxI ]

RANIERO GOLİ

and therefore one has no ulterior applcation of con- ventions Aesthetic cognition is not, mn fact, the same as the forms of perception proper to a didactic work, namely 'I am commanded to do this', 'I want to do tlus', and ' I have done what I had to do' Such forms of perception tend, in fact, to an extrinsic end, successive to them mn time, and are thus of an ordinary, practical nature In aesthetic experience, what hap pens is, instead, the birth of the aesthetic tasting of the artistic expression Such an experience, just as a flower born of magic, has, as its essence, solely the present, it is correlated neither with what came before nor with what comes after Tlus experience is there- fore different both from ordinary experience and from the relgious one 1)

Apart from some modern intutions, of whch Paul Valery is perhaps the most penetrating and bnllant mnter- preter, in order to find something similar in the western linguistical evegesis, one must turn to the conception - in the West also connected with the holy scripture - of a sensus historicus vel lteralis, different from the sensus spi ntuals, qui, however, super litteralem fundatur et eum supponit 2) Some passages of the scripture, if literally taken, are absurd and meaningless They must therefore have another sense Every word of the scriptures has, as a pomt of fact, a ludden or spiritual sense 3) The mam

1. Locana I 21 (ed of Benares pp 158 160) *) St Thomas Summa Theologica I 1 10 #) On the seriptural exeges s in the West one may now ronsult the beau tiful book by H de Lubac Histoire et Esprit L intell gence de i Ecriture d apres Ongene (Paris 1950) This concept on has been successively examined by the

[ XXXII ]

Udbhata's Vivarana, ete difference between India and the West is based on the fact that with us this conception - which goes back to the Alexandrine school and specially to Origenes - has remamed restricted to the theological speculation Had it been developed m a literary direction, we would have had a sort of counter-part to the conception of Anandavardhana Notwithstanding the undemable differences, these two con ceptions have, however, as a common foundation, the intul tion that both poetical language, and, in another sense, the religious one, do not exhaust themselves in their tran sitive value, but, using the very words of Paul Valery, survive to comprehension But alam ativistarena and let us turn to the gunavrtti b) The secondary function of words appears in an other passage also, namely mn the definition of the rūpaka figure", which has been explained by Hemacandra in his Vweka, by the words 2

upameyasabdo dharmuvācı dvitıyenopamānasabdena tathabhutenanupapadyamānasāmanadhıkaranyas ta- syaıvopamanapadasya svablidheyavmnabhutagunavr- tttam nıyamay atı | tatas ca sāmanadhikaranyam apy upapadyate | ata eva ca bhede' py abhedapratıtır ıyam napanyayah |

These very words, whch, by the way, closely recall the defimtion of the same figure in the Kavyalamkarasa

same author in the three volumes Exegese-Medievale Les Quatres Sent de l Ecri lure Paris 1959 61 1) See the Alamkaracudaman 45 idam (sci the gunavrit ete) vakşyamaņasya rupakalamkarasya bijum | 2) Vwveka p 350 [ xXxIII ]

RANIERO GNOLI

rasamgraha1), occur verbatim in the Fr. 16, which has been, no doubt, the source of Hemacandra's citation. The only exception are the words tatas ca samanādhika- ranyam apy upapadyate, that look Lke an addition by Hemacandra himself. The expression s va- bhidheyavinabhuta, applied to the secondary function, appears to be a reminiscence of a stanza by Kumārila 2):

abhidheyāvinābhūtapratītir lakşanesyate | laksyamanaguņair yogad vrtter ista tu gaunata #

This stanza is quoted by Abhinavagupta and Mammata also 3). Strictly speaking, the expression abhidhe- yavinabhuta, used by Kumarila, does not refer to the secondary function but to indication (laksana), which is held by Kumarila to be a separate function 1). Like

1. KASS, Il śrutyā sambandhavırahad yat padena padantaram | gunavrttipradhanena yuyate rupa- kam tn tat || The term anupapadyamanapadantarasam. bandhah, used by Induraja (p 10, ]l. 17-18, Tilaka says anupapa- dyamsnarthantarasambandhah, p. 8, 19), 13 quite interesting. Indeed it seems to echo the anupapadya (supply anupepadya- manasamanadhıkaranyas, as we read in the Vireka) occurnng in the present commentary. *) Tontracarthha, I, 4, 23, *) Locana, 1, 18, Kāiyaprakāsa, chap. II. 4) For the convemience of the reader, I quote here the very words of Kumarla as translated by G. Jha (The Tantrararttiko, translated into English by G. Jha, Caleutta 1905, p 440) "Is there, then, any difference between Indication and Secondary St- gmfication? Certainly, there is, because when a word sigmfies sometluing, not entirely disconnected from that which is the directly expresaed meaning, then we have what is called Indication, whereas when the word uigmfies something (even thongh it be in no way connected with the expressed meaning), through the

[ xxxIv]

Udbhața's Vivarana, etc, Mammata, Udbhata also seems to have considered the secondary function and indication as two variants of the sa- me function, generically called by the term of gunavrtti, in accordance with the view of several later rhetoricians, for whom the distinction between laksana and gauni vrtti, mantained by Kumarila, even when noticed, must have appeared more technical and pedantic than real ". c) Fr. 17, a, 1. 6 likewise supports Udbhata's author- ship. Comparison, as it is known, may be brought out even without particles as yatha, iva, etc. In this case we have, using Bhamaha's words, a samasabhi- hita (Bhamaha, II, 32) comparison. The word sa- masa, according to this commentary, is a synonym of samksepa (samāsasabdas cātra sam-

simulanty of the qualities indicated (by the expressed meaning of the word), then this sigmification is called ' Gauna ' or Secondary. Thus then, just as the word, directly expressing the Class, indicates the Individual whch is always present in the Class, in the same manner, it 1s by mere Indication, that the words 'lance ', elevated shed ' and ' horse ' sigmify the persous connected with them, (And it is not this sort of sigmifi- cation that we are dealing with) What we are dealing with is the case of such sentences as 'The Student is Fire', where we are not cognzant of any sort of invariable concomitance of the Student with the Fire, what we are cogni- zant of in this sentence is that (1), the word ' Fire ' denotes the Class . Fire ' (2). this Class indicates the qualities of Fire, such as the pecuhar colour, bri- ghtness, &c., and (3) the presence of these qualities in the student gives rise, through similarity, to the idea of his being Fire itaelf ". See also DhA, IIf, comm to at 33 (transl cit , p. 107) "There are two forms of indication - (1) Usage of a word to convey a second meaning because of some reason and in such a way that the primary meaning is entirely concealed as in the example - ' The student as a veritable fire'. (2) Usage of a word in such a way that its primary meaning also is conveyed as in the example ' There is a hamlet in the Ganges'" *) The nature of lakeana and gaupi vrttr has been clearly expounded and discussed by B. Bhattacharya in bis commented edition of the Dhvanyaloka (Dhvanyaloka of Anandavardhana [Uddyota I], Caleutta 1956) pp, 17 sqq

[ xxxv ]

RANIERO GOLI

Lsepaparyayah). Now this same interpretation - samkşepābhıhıtāpy esā sāmyavācaka- vicyuteh - occurs in the KASS also d) Another interesting fragment is no 27. According to Udbhata, the śles a figure " (called by hım ślista) is alamkarantarotpattihetu. This view is not shared hy later rhetors For them, mndeed, it is not ślesa that produces an image (pratibha) of other alamkaras, but some other alamkara that produces the Image of slesa. The scanty remains of Fr 27 support Udbhata's opinion. In it, indeed, we read that, by virtue of antarapratıbha (unfortunately the beginning of the hne is lost; should we supply (rth) antara- pratibha?) an image (pratibha) of some other alamkara or sentence is produced2). The opimion of Mammata and others was, as we have seen, just the opposite e) As it is known, Udbhata improved Bhamaha's de fimtion of aprastutaprasamsa, by stating expl- citily that it must be prastutarthanubandhi- ni, otherwise, as Induraja and Hemacandra say, this figure of speech would resemble, as it were, to the uncon- nected ravings of a madman, and there would be not even

') This is a general statement The views of Udbhata concerming the lesa figure have been detailed and compared with those of Mammata etc, by K P Trivedi in his edition of Ekaval (BSS No 63, pp 622-628) See algo KASS (ed cit ) notes pp 112 118 HSP, 2th part, pp 196-201 SP, II, pp 56 85 163 232-4 *) The defimition of flesa by Udbhata in hs KASS (IV 50-51) very close to that runs as follows eksprayatnoccharyanam tac chayam caiva bibhratamtsvarıtadı ganair bhinnair bandhah ślıştam ihocyate || alamkarāntaragatām pratıbham janayat padaih | dvividharr arthasab doktıvıdıştam tat pratıyatām

[ xxxVI ]

Udbhata's Vivarana, etc the smell of an alamkara As a matter of fact, - adds Hemacandra - who, having to describe a mountam, would describe the nature of the sea? The fact that these very words - tatha hi parvatopavarnanāyām kah samudrasvarūpam upavarnayet-qli te in accordance with Udbhata's conception, occur both mn the Vaveka and mn the present commentary (Fr 40a,1 8), are an additional proof that we are faced with the original com mentary of Udbhata In this same fragment the explica- tton of the words adhikara, occurring mn Bhamaha's definition, by upavarnanāvasara is a well-known one and occurs both mn Induraja's commentary and m the Viveka This has been, no doubt, the source from which It has been borrowed

§ 3 GENERALITIES

As expected in a work of this kind, the stanzas quoted as udaharana are found in great number in the commentary Of some of them I have not succeeded in tiacing the source I hope that others will be luckier than I Wherever I have found it possible, I have mdicated the metres used Numerous, in VI ch, the grammatical quo- tations The commentary of Udbhata, whose true title 1s unk- nown and which is known as the vivarana, was until now only known to us through a few quotations by Abhinavagupta and a few rhetoricians after lm " In

") A number of passages have been collected with as usual d l gence by P V Kane in hs History of Sanskrit Poettes (Bombay 19>1 pp 1"6-T) Literal quotations from Udbhatas commentary are very rare The frt of them occurs in the Lorana p 32 (Benares ed 1940) bhamahenok

RANIERO GNOLI

this work the great Kashmir rhetorician discussed a num- ber of subjects not treated by him in the shorter Alamka- rasamgraha1). Although the htteral passages are rare, discussions on one or another opmion set forth by Udbhata are, nonetheless, more frequent "). However, this is hardly the place for a disquisition on the pecuhar theories of Ud-

tam sabdas chandobhidhanartha ity abhidhanasya sabdad bhedam vyakhyhtum bhattodbhato habhape śabdānām abhıdhānm abhidhavyāpāro mukhyo gu- navrttis ceti Another one is that of Hemacandra (KA. p 17) 1ha tubhayeşam (scil guna-s and alamkāre-s) samavāyena sthatir sty abhidhays tasmad gaddarskapravshena guņalamkārabheda itı bhamahavivarape yad bhat. rodbhajo `bhyadhat tan nirastam Udbhata, as 19 known, identified guna-s and alamkara-s, against the entire later tradition (the nature of and matual relations between the gupa-s and the slamkara-s have been discussed by V Raghavan in Ch XVI of his Bhoja's Spngara Prakasa (Bom bay, 1940) Udbhata s opiion on the subject is discussed on pp 301-5 A third short quotation may be seen in the commentary by Samudrabandha on Alamkarasarıasva (Trıvandrum, 1915, p 89) Udbhatena ca kavyā. lamkaravıvştau satkavıtvavırahıtāya vidagdhatāyā asthaıryasyāsobhanasyn ca pratipadanāya nıdar sanadvayam ıtı vadata ka śrır ıty asya srır asthire- ty artho 'bhihitah Some other small quotations (but it is doubtful whether hteral or not) have been cited by P \ Kant, op cu , pp 126-127. ") The Alamkarasarasamgraha has been issued by the Nirnayasagara Press (Bombay, 1915) with a commentary called Laghurri by Pratihara Induraja Another edition with the same commentary, introduction and notes, is that of the Bombay Sanskrit Series (1925) In 1932 this work was edited by K S Ramaswami Sastri Srromam in the Gaekwad s Orental Senes of Baroda along with the commentary of Tilaka #) Although the commentary on Bhamaha is perhaps Udbhata s most important work it is probably not the only source from which the knowledge of his ideas shown by later rhetoricians has been drawn A rather important work from which Abhinavagupta has left us some passages was a commentary (whether complete or partial, it is unknown) on Natyasastra Udbhata's opi mons on the Vrttis taken by Abhmavagupta from that work, have been discussed by V Raghavan in his article The Vrttis (JOR 1932/VI pp 347-370 1933/VII, pp 33-52 91-112) Another work by Udbhata which is now lost was the Kumarasambhaua a poem homonymous with the one by Kaldasa From this poem he borrows in his Alamharasarasamgraha the examples of alamkara-s

[ XXXVIII ]

Udbhata's Vivarana, etc.

bhata and the intricacies of the Alamkarasastra. That is a task I gladly leave to others more competent- than myself in this field 1 In the course of time the fate of Udbhata's work was the same as that of so many other works of ancient and medieval India, that 1s, oblivion. The Kashmiri master's text was no longer copied and studied, and people passively continued to repeat the opinions expressed by previous rhetoricians on it and its content.

§ 4. THE TEXT OF BHAMAHA.

This discovery is nevertheless important not only with regard to Udbhata's lost commentary, but also for the knowledge of Bhamaha's own text which this manuscript enables us to correct m some places 2). The Kāvyālamkara,

i) Apart from the above-cited work by Kane, we may refer, to mention only the most important works, to the History of Sanskrit Poetics by S K De (Calcutta, 1925, 2nd ed Calcutta 1960), Vol II, pp 41-94, and various works and articles by V Raghavan see especially, mn addition to the above- cited Srngara Prakasa, Some Concepts of the Alamkarafastra (Adyar, 1924) ") The Kavyalamkaro has been publshed for the first time in 1909 by Mr K. P Trivedi as an appendix (VIII) to his edition of the Pratapnrudraya- sobhusana in the Bombay Sanskrit Series The edition is based on Madras MS no. 12920 (Cat XII, p 8675) In 1927 P. V Naganatha Sastry publuahed the text based on four MSS with an English translation, notes, etc (Tamjore, 1927) Then the work of Bhamaha was published in 1928 by Batuk Nath Sarma and Baladeva Upadhyays in the Kash Sansknt Sene So far as I know, the last editions of the Kavyalamkara are that by D T. Tatacharya Siromanı, who has also written a Sanskrit commentary on it, called UdyanG Vrt (Tiruvadı, 1934), and that by C Sankara Roma Sastri (with English Translation and notes on Pariccedas 1 to 3). Mylapore, Madras 1956. The edition by Bathuk Nath Sarma and Baladeva Upadhyaye was violently critrcized by K B Pathak in his artiele Dharmakirh and Bhamaha (ABORI, XII/1930, pp 372-395) The readings proposed by Pathak, however, are not always better than those of the edition he eriticizes Equally inaeceptable are his conclusions regarding the dates of Bhamaha which. according to him, YEXIX ]

RANIERO GNOLI

like so many other works, was preserved only in southern Indir, and even there only in a very few manuscripts, probably deried, as noted by Messrs Batuk Nath Sarma and Baladeva Upadhyaya m therr edition of Bhamaha, from a single archetype 1) There are a good many ob scure pomts m Bhamaha's text, nor are the corrections and new readings suggested by D T Tatacharya Siromam in his valuable Sanskrit commentary, Udyana Vrtti, on the Kavyalaml ara always fehcitous Some of the readings of this manuscript are undoubtedly the original ones I may refer to the following evamples a) in stanza I, 1, the reading s arv a m is confir med against the lectio facihor sarvam mentioned (but on what authority>) by T Siromam, UV, p l pra- namya sarvam ity api pathah tada sar- vajnam sarvam ity anvayah2) b) The reading of stanza I, 9, as we found it in the Kaumudi, a commentary by Uttungodaya on the locana of Abhinavagupta, p 67 (Madras, 1944) itihasā- srayah kathah |loko yuktıh kalas ceti mantavyãh kavyahetavah l), agrees with that accepted in our MS The reading of B, kavyayair vası(kāvyayair amı') and the conjectural resto

came after Dharmakirtı In fact logical diseuss ons of Bhamaha are not based on Dharmakirtı but as demonstrated by G useppe Tucer in lis article Bhamaha and Dinnaga (The Indian Antiquary LIY/1930 pp 142-47) on the work of Dinnaga Bhamaha s flourishing therefore is to b set within the interval of time between D nnaga and Dharmakirtı and no later Op eu p 70 It seems to be quite probable that originally there was only one manuseript available from which all transeripts have been made from time to time It might have been the one which was depos ted n the Maharajah s Library Trivandrum #) As to the word sarvam and its meaning ef P V Kane HSP Pp 77 sqq see also S K De SP p 48

Udbhata's Vivarana, ete ration by T Sıromam to kavyavaikharı are uncorrect c) In stanza II, 70, we must rend hetur, as m the MS, mnstead of setur m B The reading hetur for setur has already been suggested by T Siromanı, UV d) The reading of V, 18, remains doubtful, the leaf here bemg damaged and the aksaras mutilated In the MS we read yathasuci sunu straini The reading of the two aksarns ci su is not certun, but stra- in1 is perfectly legible and confirmed by the subsequent commentary beginmng with the word stri Thus the conjectural restoration, trini, suggested by T Siromanı (UV) and K B Pathak (op eut . p 392) is uncorrect e) The correct readmng of the verse V, 25 is ita dvayaıkānugatıvyīvrtti laksmasādhu- ta In other words, the reason in order to be vald, must possess three characteristics, of which two are positive and one is negative The two positive characteristics are its residence in the subject (paksadharmata) and its presence in all or in some of the affirmative mstances (sapakse sattva), the negative one is its absence in all the negative mnstanees (vipakge'sattva)" f) The reading of verse V, 56, is, according to the MS, sadhyasadhinayor vrttir ukta ta. tra hi nesyate Other mteresting vinants of Bhamiha's text may be found in I, 29, where our MS gives jatah katham.

1. Tlis is the doctrine of D anaga an i of the sehool of Buddhist logies In ths connection see G Tucei Bhamaha and D Anata (quoted abave) and the Ayoyamukha of D andgn (IImdelberk 1930) Lastly on the h story of ths theory see by the same sutbor te volume Fr-Dindga Buddhist Texts on Logie (Baroda 1929)

RANIERO GNOLI

cana instead of jatah katham janah. In I, 33, the first two aksaras of the word a smakavam- ś a d are lost and the other four kav a (m?) syad1 We find another varia lectio in I, 42, where we read t a - trabhramaraharitacakrao, mstead of tatha bhramaraharitacakrao In verse II, 68, we read aksepam iti for aksepa iti, and m II, 69, iyad evastu vanyena, instead of iyad evastv ato 'nyena fuklamsukany In verse II, 83 we read tatha asann angesv instead of tatha suklamsukani syur an- gesv, etc In verse V, 20, we read sti ūpe usnas candrama Although the metre here is defective and the correct reading would not be candrama but ksa- pakara, as we read m the printed text, the reading upe is mteresting, and although it is impossible for us to decide either for rüpe or kūpe (but küpe seems the better), both of which are graphnally possible, we see, in any case, that the reading rupam mn the printed text is erroneous Besides, the reading k u pa was already attested in some MS, having been cited mn a note by B

[ XLII ]

ADDENDA

Fr. 9, b, 1. I. The words namas tungax{ seem to be a citation of Harya- carita, I, I. The entire stanza runs thus namas tungasira- şcumbicandracāmaracārave | trailokyanagarā. rambhamulastambhāya eambhave | Possibly, the ecpression rambheti () 7) refers to this same stanza, pada c. Fr. 11, b, I 7. Possibly the words vitaku, etc., are from some prakrit verse. Fr. 19, 1 6. Possibly here the text was oraktanaya(na)m atifaya- drtrka(ntha)vigraham Uncertain metre, Fr. 21 b, I 5. The reading yabhisusubhe i equally possible || Fr. 23, b, 1 7. The words yopayam, etc, look fike some stanza which I have not succeeded to trace The metre, too, 19 not clear to me, and perhaps the text itself is somewhat vitiated Fr. 23, b, 1l. 7-8 Apart from this, I do not know of other citations from the Candtfatako in alamkara-texts, with the exception of Bhoja's Saraspatikanthabharana, where stt 40, 49, 66 are cited. Fr. 34, a, 1. 3. The stanza he hamsas occurs, with some variants, in the Sarngadharapoddhat 807 (anon ) The text runs as followe bho hamsaa tavad ambhoruhakuhararajoranjıtangab sahelam hamsibhib padmakhande madhuramadhukararavaramye ramadhvam yavan nāyam duranto haragalagaralavyālajalālınīla- pronmilanmeghamalamalinasakaladiamandalo 'bhyeti kalab I Fr. 3d, a, 1. 2. The stanza gayantu, ete, occurs in the Sarngadhara- poddhat 1247 Here it ie attributed to Srivalmikikav In the MS the reading gayantı (as in the Subhasttaralt) is equally possible, the leaf being here sersously damaged F, 37, b, ll 6 sqq With the words dhyanamalls beginns apparently a stanza, the third pada of which is jhaokarab prathamam grhe grhajane, etc The metre is the Sardulavikridita. The fourth pada was in the missing part of l. 7 and # (at the end and the beginning respectively) Being it so, the few and hardly legibly words in 1 8, up to rty, seem to belong to another udaharana, The ap- parent addition after kalika ( 7) of the word gandha, written in the lower margin, is not consistent with metre, and not to be ac- cepted in the text, Fr. 39 a, 1. 2. The metre of the stanza occuring in ths hne was perhaps the Haripi. Fr. 42. Possibly this fragment belongs to IV, 20 and is to be put after Fr. 44.

[ XLIII ]

FACSIMILES

(Reproduced in the original size except where otherwise stated)

Fr Ia

Fr 1b

Fr 10 a

Fr. 10 b.

ستا

Fr 11 a and o (reduced to approximately two thirds the s ze of the orig nal)

Fr 14a

Fr 14b

Fr 21 a

Fr 21 b

Fr 22 a

Fr 22 b

Fr 26 a

Fr 26 b

Fr 29 a

Fr. 29 b

Fr 34a

Fr 34 b

Fr 35 a and b (reduced to approx mately twu th rds of the < ze of the or g ml)

Fr 37 @

Fr 37 b

Fr. 39 a.

Fr 39b

Fr 40 @

Fr 40b

Fr $3 a

Fr 43 b

Fr 41 a

Fr 44 b

Fr 46 @

Fr 16 ₺

Fr 53 a and b (reduce I to pprox n itely two third the e ze of the or g nal)

TEXT

Fr. I

1 ]xjjyotirbhāsitāl ka[ 2 Jraḥ kriyāyām cal 3 ] l| om pranamya sarvam [I, 1]l [ 5 4 ] ājanada x 5 lx e i[ .

10 4 ]x tra ity [ 5 ]mokşeşu vaicakşanyam [I, 2]] [ 6 Jcanāt | na hy anyarthala[ 7 ]xsyasya prayojanava[ 8 Jmokşeşu vaicakaa |[I, 2] [ 15

Fr. 2

Jx kavyan [ Jm tathā i[ 20

3 The voealic part of the syllable I i uncertain, anyway, it is not o 12 The aksara kaa is aubscript 14 For sya the reading dya 1s equally possible.

[3]

RANIERO GNOLI

b

Jh phala y [ ] m acayanıl 5 .

Fr. 3

)purux x x[ 10 bjajjyotsna[

. 1x y h [ 15 ]tattvam Lintu[

Fr 4

a

20 ly[ Jnā 1I

6 Fr 3 - Ths and the following minor fragments up to No. 7 seemingly part of the same leaf as Fr 1

[4]

Udbhata's Vivaraņa, etc.

]tıyal

Fr. 5

Jt | evamx [

b

Jovyaval

Fr. 6

Jvirel

Jx tki i[

[5]

RANIERO GNOLI

Fr. 7

]ka 5

10 Fr. 8

1 ]dharmādişu mukhyeşv apil 2 Jvyakhyanam freyah praf 3 ]prayojanena x 1.I

15

6 Įd. vyakhyayat.[ 7 J atkavch kavyani ku[ 8 lkimcid atrasambaddham .e[

Fr. 9

20 I Jtha | na danena na man [

21 These are the Erst words of a stansa qaeted by Bhoja (Sroggrepre

ns sAtbyron as sevayd t an thetreoa an brdydo tab sarvatha kim iva striyabI Thas Gloknis cited by Bhoja at #R ersmple of as abhidhiyamandrtha, which is neither vidhi bot

[6]

Udbhaļa's Vivarana, etc 2 Jvāca sabdayogmno ya[ 3 ]bhidhavyapirah kvacid f 4 Jsabdasastrair iti[ 5 Jc (ch)abdasastram s(ak)re chrutva y babul 6 ]sastrasravanim antarena [ 7 lpatıkāvyakaranasya | a[ 8 ]s [

b Jtatrapt uktah namas tungax[ 2 Jvara X ranad vigamo x[ 10 3 ]x t gataX Xtra susatrum h [ 4 Jte rthesu satsuddhi dvi[ 5 Jparipustas tu | a y hā[ 6 ] yamayim asesam jyotsnal 7 ]x rambheti slokah atra[ 15 8 Jrthah na catredim vaktavyas [

Fr 10

1 X X x xparıjnane prayatıtavyam ity alam atıprasan- gena | xtra sax sul 20

nişedha (avidhinıedharupa) (but hee p 271 also) This same stanza occurs with some variants in the Hutopudesa also na danena na manena nārjavena na sevaya / na sastrena na sastrena vişamib sarvatha striyah 1 See Buhtlingk, In duche Spruche 3283 14 The aksara m a is doubtful read ja" 16 The reading seems to be certain but the text ie bere apparently corrup ted, the remaihs of the aksara subsequent to vaktavya suggest s 19 Fr 10 - This important fragment has heen discussed in the Introd pp XVII sqq 19 prayatrtavyam corrected from MS prayatıkavyam

[7]

RANIERO GNOLI

2 hāsāśrayāņ kathā Į loko yuktiļ kalās ceti mantavyāh kāvyahetavah [[I, 9]] | esām kāvyox [ 3 x x xabhidhanarthah sabdanam arthab yaur r Xsya etad evam ukta x atra pada[ 5 4 yujyamānatvāc chabdārthā nāsambanmo labhyata ity abhidhanagrahana X X ixi[ 5 parvatad eti devadattab eti jivantam anando naram (var)sasatad apiti guņavrttix x[ 6 mırastadhaıryam rādheyam ārādhitajāmadagnyam | asa- 10 m(stuteşu) prasabham (bhayesu) jāyeta[ 7 kartrbhedan nāstı ktvapratyayah atha paksapatanakri- yato x x x xkryax x X tvam tada[ 8 ghatate şasthyabhihitasya karttuh paksapātakrıyayaiva sambandha (1ş)ya(te | na) nirīksana(kriyaya) [

15 1 asambhavinām Xta iyanıbandhaparyalocanena matany avagamya[ . ksanam kāvyax x xgin x xsanām | tenasobhanam 2 kāvyam x x x x x xpradhan [

As lo the reading kavyahetavab see the lntrod, p XI. $ This is apparently a stnbal mstake for abhidhonarthab. B Reading doubtfnl after yair, the akparas being partly lost. Appa- rently the akparas subsequent to yai are hke to rgranejhasya (ne') S The word osambanino la to be considered as a senbal mistake for uambandhino Apparently labhyats is a mistake for In- bhysnta. 7 eti corrected from MS ati. Thu verse is a well known adage whose carlieat occurrance ia in Valmila's Ramayano (Sundarakanda, V, 34, b) The first half of the stanza runs thus kalyani bate gatheyam lao. kıkI psatıbhätı mål D This ts e quotation from Bharavi's A:ralaruniya, III, 21 The entire Htansa runs as follows nirikşye samrambhansrastadhairyam

Babham bhayesu jayeta mrtyor apı pakeapatabl 18 This refem to I, 10

[8]

Udbhata's Vivarana, ete

3 Xm avady x x x (vi)laksmana h kavyena dussute- neva mndyate I, 11] | prax x xn k Xm api pal 4 nā na kāvy X dussuteneva kuputreņeva nındyate ka Xnax 1[ 5 dandanaya va | kukavıtvam punas sākşān mrtım s āhur manīşmnah [1, 12] | kāvya[ 6 XX xtena ca kavır mryate sarveşam utpadyanta ram evax x{ 7 X Xmarana(m) pandıtāh pravadantı | asādhuni hı k [ 8 x xLāvyambandhana evadara kartavyah sa cax X 10 padärtha[

Fr II

a

1 hātah kathamcana || I |I, 29] al 2 Jna hı jāto nayakah prāpta[ 15

3 ]tha ya dhiraprasanto na(ya)ka t khy- äyık [ 4 ]gambhīryagunasyo(tka)reena | yasyaX[ 5 # xd xx[ 6 jhal 20

7 ]thã mḽ

3 The ugn like dka at the begmning of the line, just before pa, 1s a numerical symbol, namely 4 7 Thus is a common mustake for mtiyate, see Fr 51 a 1 3 also kryate for kriyate 14 As to the reading kathamcana see the Introd p XLII 15 Here the leaf has been shrivelled and twisted owing to the action of fire and the space for the syllable ya and in the line that follows for eka, is apparently missing See HC, pp 432-33 [9]

RANIERO GNOLI

2 Jhakrx [ 3 ]kaya[ 4 ]xm atmamıx Xya X X vaksipta[ 5 5 Įdayādhāritaprašānto ga[ 6 ]xra pītkarasabdacakı[ 7 Juitaku X haladhattah[ 8 ]x sam ekavākyan tv ekala[

10 Fr. 12

a

1 ]vaidarbham tad apy evam evapi[ 2 ] vrttayah kāvyasya | komal [ 3 ]kap komalaih u[ 15 4 ]tena vaidanābandha 1* 5 Jyana naiva cal 6 ]caryante | te[

b

3 ] ādınā pra[ 20 4 ] y uhatānurūpaf 5 ]devata agnifikhe[

8 Reading uncertain The ill-formaed akgara after ku looks like d bhd, should we think to a senbal mistake for to? The two akparms after Ia sug gest patta (after which one sees the posutle remains of a vwarga) howe- ver the readmg dhatta or dhadr is equslly posuble Meaning' 15 Reading uncertain a mistake for vaidarbha' The two mulse+ quent akparas can be read batva. tatra or tandhn sbo bhe-ta and tva-ndha being couely confuwd

[10]

Udbhata's Vivarana, etc. 6 Jkhyenābhisambaddho na co[ Ipam eşa yo uttara[ 8 Jbkrāņām vā yogyatayopamā[

Fr. 13

5 1 123 Įthak | gatānu I, 32 [ Idhyaprabhedā y [ 3 lyam jamad [

10

6 ]d etatf ]darbhax x i i[ 8 ]d anapetanyal

Fr. 14 15

I yam idam etat tuf 2 thakkaraņam uktam de[ 3 X x vaidarbham I, 32? iti | prl 4 XX X X Xh prabhedam x [ 20 5 x xkava(m?)syadi vaidarbham iti ka(thyate) } [[I, 33] 6 x x x xh kintv asmākam daivarbhe ye[

3 The reading tranam is equally possible. 22 The word daivarbhe is an obvious seribal mstake for vaj. darbhe.

[11]

RANIERO GNOLI

7 x xkamam tathāstu prāyena samjūie [[I, 33]] [ 8 x x dirūpāsraymām kāvyasya[

1 x xpra ix dhriyāt | tasmat[ 2 x xtatredam aha |i apustartha [I, 34]] [ 3 x xt | yas cabhidhasyamanah etad vi[ 4 x x x xlankrtih a[ 5 x x x xbandham tathā[ 6 x x x érutìxcajál

10 7 lam | gaudiyam api [I, 35]| 8 adaır avāntarava[

Fr. 15

15 3 Jseşena pal 4 ] r ryopadasannidhil 5 ]klıstam vyavahitam [[I, 40]] dosa iti[ 6 ]sambandhisannidhane[ 7 Jharanam pratilomy [ 20 8 Jsambandho durasthasyapi ta[

b

1 ]risamāptes tathā ca ja[ 2 lyady upamadosa[

4 The two akgaras after pra are almost entirely lost and illegible. 15 The leaf 1s broken just above the syllable n a. Had it been surmonnted with the sign of the vowel o, the reading sepenopa (sepenopama", ete ) would be equally possble 19 The akpara ti is by no means clear, more resembung t1.

[12]

Udbhata's Vivarana, ete 3 Jm apı hı catvāry upamāļ 4 Jranīyah sa ca pamcal 5 lvā | tatrādye i[ 6 Jnex x x[ 5

Fr 16

133 Kkva)cıl laksanamatram kvacid udaharan Jtadanyartham nanu naiva sabd Xn y nt [ Jasvagonyadayas sabdas sa(dha)vo visayan- 10 tara 1tâ[ 4 Jpre py aprayogat | a a yadosam y xpa o[ 5 lsau punas sabdarthasambandhapravrt. tâv [ 6 ]hrur [I, 40] ity etac yamanam upasargaX X X X> 15 tvaxkt pexI 7 ]x | prag(ukta>)nam ca mpatattvam upasargo X X x x x x x vIta[ Jx ham x x ix prasiddharthapax 1 xX X X X x Xh anya[ 20

1 ]doşatā | avacakavyakhyanartham aha[ 2 ]a)t humapaho vahnh tasyamttras toyax x x x rn x x{

4 For pe pethaps po i to be read the next two aksaras are almost entirely lout and sllegible D So to be read? or naivam? 10 The source of the words asvagony" - is Vakyopadiya I. 149 szvagoņyādayab labdab sadhavo viçsyantace | ni. mittabhedat cervatre sādbutvam ca vyavastbıtaml 23 Thit is a comment to I 4! [13]

RANIERO CNOLI

7 1 yanarthakah etat tritayam api punaruktā- bhasam yatha | tathabharanabharanax xyuf 8 ]dvayor ekārthas svaravyañjanasamudāya- os (sva)yam punaranarthavairūpau ca tatrāpi punaruktāl

Fr. 20

a

1 ]sandaştakadınamantarakaranam x x x x x prabhed x ks[ 2 Jdhuna sadhuna tena rajata r(ajata bhrta j ) sa-

10 hıtam sahıtam kartum |II, l1p [ 3 Jtarat tacchux x sitasai[

6 ]t y dosaxi 7 ]ksatah antenadryamanex XXXX y s uxt [ 15 8 ]k bhūsrtivānsangam gangam natra kix x x x xyai- kayā vah bhaya[

Fr. 21

1 ] tat | prathamadvitīyapādābhyāsodāharana[

The akgars ra before ckarapam is subscupt 14 For keatah perhaps krtah also can be read 15 In the MS ovarisangam has been corrected from varinangam The anusvara of gangam is apparently marked as to be deleted the subsequent akeara, first written as tra, and then corrected into trya, has also been marked as to be deleted and substituted with na The metre of the stanza was perhaps the Vasantatilaka 17 Fr 21 - Tluis i a commentary to II, 9, etc

[16 ]

Udbhata's Vivarana, etc 2 ]bhyasodaharana(m) r va rdhe y dh punar apy anu- sangalam praX X samikayāsurase x ya[ 3 ]bhyasodaharanam | ratnali bhimate _ stu va u radbrais sadvesān analasamāna axto _ sadve- sān vuo [ 5 4 Jna x dhıpasāvitarenukarenavah į karıvaraih kakubhas sabhayais srıtāh tad avalokya cane u[ 5 lyasobluduraloke kopadhamaranad r bh Xya- sobhıduralok kop [ 6 ]x x xlamkrte vividhamārge tava vrsa- 10 bhocitadharesum [ 7 ]tmano bhrtiprthuladhamno | saty apı pibatt dha[ 8 ]padabhyasodaharanam | paramaxn [ 9 ]purusena rajata sobhamanena raja[ 15

1 ]vatam trisikham asrnkhalam dadhat sa[ 2 ]so x jala x x visarpibhis sasanas rana x na X ca[ 3 ]drsyante yatha il yamalam ahur Xja- gatpray [ 20

1 These are apparently the remainings of a itanza in Drutavilambita metre In MS instead of aangalam was first written aamoalam The anusvara of a a m was then marked as to be deleted and the ayllable s a substituted with u ga (subscript) 4 Uncertain reading the akgaras mate (but te ws doubtful) are subseript The metre was perhaps the Praharginı C This is apparently part of a stanza in Drutavilambita metre 8 The syllable du has been corrected in the MS from ta Uncertain metre 11 Uncertain metre 12 For odhamno odhamne also can be read 15 Thas refers to II 11 17 Apparently this is a part of a stanza in the Rucita metre

[17]

RANIERO GNOLI

3 ]mati | apārthadoşūd asyāvišeș. X X X X X rūdham vācye rthe II, 41]I Jc X | ata eva na prahelikāšāvasya- prayogo na vihanyal 5 5 Įvaļ( | )tatrābhramarahārītacakravākašukā- dayab (I, 42]| jalabhyanmax [ 6 ] artham hi dūtasam Xsaxnn ceteme doşa- visesa x x[ 7 ] i| athāco jalabhyatprabhrtay. X X vya 10 ktavãl 8 ]n kam ihanena yuktimatof

Fr. 17

15 ]dprax x x xvya i[ Jd x xlabhyan kata[

20 Jbda ex x pasabdasya Jxérāx xd v [

5 The readng tatrao is apparently a mistake for t atha, which seems to be the correct reading; cp the Introd, p XLN. 7 The akyara subsequent to the syllable s am 15 ill-formed and I have not been able to read it O The reading is clear but the text is here somewhat vitiated, it can be restorated as avaco jalobhrtprabhyta', ctc.

[14]

Udbhața's Vivaranā, etc Fr r8

]s daXrthad X[

Itsanta iti y [ 10

Fr 19 One face only 123 Jpal jpūrva r t bhūyastvax Xn vinyāsa x xhrdaya[ ]r ity aha } latiyam apy anuprasam ihecchanty 15 apare yathā | drştumn drştısukhām dbehi candras can- dramukhoditah (|II, 81 [ 4 ]nam vinyas bh x y rthah drytım drştisukham dhehi candras candramukha ity udāha(ra)nam | he can- dramukhe[ 20 5 ] o vā yasyārthabhedas tadā yamakālankāro rthabhedas cet punaruktabh isah nartad el 6 ] ā asaxyathā | nararudhraraktanaya x mitifa- yasıtıka x vigraham vande x na[

21 Read yamakalankararthanam" But traces of the vowel o alove the finalr of alankara are vinble and the akyara mubsrquent to rtha is almgost entirely lost and sliegible [ 1s ]

RANIERO GNOLİ

4 ]x krta X vivesa dūrāvadātabhavanes khanda- maulo 5 Jdanum kuam isa x x x x yā | sandhy a yatısusubhe prakasayāpreta x xru[ 5 6 ]sya yah pax vasyanakrtavi X X Xpmrımeyah atha kathamcit parcchettum ta y xh tadasya[ 7 ] | munin api haranty ete ramaniyesu samgata [[II, 13]] | tatra yulpā[ 8 Juarana X ndhavatam x x x yukpadasyodaharana[

10 Fr 22

1 ]t matat na yamakalaksanam prāg uktam pa risamapta 1x x x x x xkala Xra[ 2 ]sabdam ojasvitı ta x nayasuvrtti ox X Xmala

15 yamakaX x xh kara[ 3 ] ix x x x x x k pox Xyan na taX X X X X sarvo na bhavatity aha )| nanadha [II, 19][ 4 Įdüsya x x x x x x x yu x padapadarthasam- bandhā[

20 5 dhã[ ]rama) sarmanacyuttotare yamakam mbad

1 Read *sura° for °dura? Apparently this ia part of a stanza in the Vasantatilaka metre 3 The reading kuama seems clear is it a word from some stanza in Prakrıt? 9 The reading of the akgarasxndha is uncertain The first of them partly effaced resembles ga or va the second one can be read ddh a or nva also these aksaras be ng eas ly confused 14 After vrttr three akparas nearly lost read o)asvi? 17 Doubtful reading these three aksaras being almost entirely effaced by the fire

[18]

Udbhata's Vivarana, etc.

6 Jx x cyavitaramasarmācyutottare /| yamakavyapade- śınī prahelka[ 7 ]x x dūşyante | atyantagudharthataya svanimitta(m?) nānadhatvarthagambhira II, 19 | [ 8 ]sādharanam ıtthambhūtam kāvyam s acāru pratyastamitalaghuprayatnataral

b

1 ]sastravat | utsavah sudhiyām eva hanta durmedhaso hatab || |[II, 20]] yadı kāvya X ari[ 2 ]vatsalāta na sukumarama(tīnam) kāvyabh yo- 10 gānāmm atra dhīsabdena vā[ 3 ]tvam upameyasya rūpyate gunānām samatāņı drştvā rūpakam nāma tad viduh || || II, 21 | [ 4 Ksa)bdo dharmivaci dvitīye- nopamanasabdena tathabhatenanupapa[ 15 5 ]svabludheyavina(hhatagu)nav rttr(tam niya- maya)t(i | )ata eva ca bhede py abhedapratitir iyam napanyayahI 6 ] adayantı sam x X ubhay(anuy a)yıropakasya ca ko viseşa itı tad atyantam asambaddham na he sam [ 20 7 ]veguh X x x x tax x x xreşu khadgeşu ko va bhavatam murarh x X 1 rupakopagamāt u[

1 The two first aksaras are demaged and hardly legible, the thurd one can be read sya or dya, these two compounds being rasily confused Anyhow, it is impossible, it seems ta read here hy udita as in the prin ted text (II, 19) but posuibly there words form part of the commentary and bot of Bhamaha's text D The aktara dur has been correeted from dvar in the margin 10 The first three akparas of this line are damaged and the sendog is doubtful 14 As to this pamaage, ep HIC, Fsreka, p 350 and the Introd, p xxxut sqq 21 This os part of a stanzs in the Indravajra metre

[19]

RANIERO GNOLI

](mayura)vyamsakaditvad evasabdalopi samāso 8 rūpakatve | vyaghradid. x x x x x xpi samaso[

Fr. 23

,1 (upamanam a)prasiddham upameyam iti Lecit etac tāyuktam candramukhadinam anyonyam upaf 2 pāndunā | netrānandena candrena māhendrī dig alam- kṛtā || tena prākaraņika[ 3 sāmyam ca yat sahrdayanam hrdayaharam tad iha 10 grhyate lankāre prakaranāntarāsa[

1 The words mayūravyamsakādıtvad refer to P. II. 1,72 mayuravyamsakadayas ca. They have been quoted by He- macandra, Vireka, p. 351. 4 Here the leaf is broken. See however HIC, Vireka, p. 351. vy aghra- didvāreņevasabdalopī samāso luptopamāyam driy. ate. Cf. also P. 4. 1, 49 The three aksaras read as prasiddham are lost, with the excep- tion of the snferior part. As to the meaning, see IIemacandra, Vireka, p 341. who seems to paraphrase this very passage tat katham ucyate pra- sddham upamånam sprasiddham upameyam ity afa- okyaha, ete. This is a confutation of the view, according to wluch u pa- mana, ie., the standard of comparison, must be necessarily prasiddha and upameyam, viz, the subject of companson, aprasiddha. Asa mat- ter of faet, the unique ehterion is the volstion of the poet. Thit view is exem. plifed with the verse tatab kumudanathena kaminiganda- păşduņăjnetrănandenn candreça mbendri dig aln- mkrtal (Mahabharata, Dro. P, 181 46, Chitrasala Press ed., 1931). Here candra, which w almost always a standard of comparuon, is upam- eya. viz. the subject of comparison, and it w compared to the cheeks of a beautiful womeo, which are usually a subject of comparwon. After anyo- nyem, if my rending is corrret, the text was here upamanopameyatvst or something like that. 10 The akjarm subseguent to prakarep arr nearly lost and the rrad- ing difficnlt. Should we rend prakaranantarksambhavat?

(20 ]

Udbhata's Vivarana, etc

4 (vise)sanam etac ca na yuktam mukhakamalādısu- pamndyu y prasangat ( na hi tatraX ha[ 5 x uktax tad ucyate viruddheneti II, 30 tena pu- rusa iva purusa iti saty (apı) purusadvayasya pu- rușal 5 6 X x x x ) sarvasāmyam hı bhāvānām na sambhavatı tadartha X x surye gunaleseneti II, 30 yatheral 7 x x] yatharX x ā tac cetr sloka(ś) cāyam asambad- dhah yady ux x x x parigananaparas tadapy [ 8 xna sabdayor viśesanam nısphalam sarva eva hi 10 bhava bhavantarebhyo vya(tire)kiņah f na hy atra yatheval

b

1 upa X vacakaparıgananaparam etan na bhavati kmntv etad atra vidhuya x y hy upamaval 15 2 uprayoge yat tat samyam tad ubhayanuyayitvena pratīyate śrutyā lax x īd eva dataļ 3 Xıtı y x x x x x samanadhıkaranyat tatraiva pra- tıyate x tra vaiyadhıkaranyatXnyaya[

The words iva purusa are subscript this passage has been freely quoted by HC p 339 Cp the Introd p xW 7 Here surye seems clear but the text is perhaps vit ated read turye (te in the fourth pada)? 10 The syllable hi is subscript 15 I am not aure that the akgara dhiy a after atra vio really belong to tlus leaf After them there are perhaps two or three akgaras ent rrly lost The akara preceding hyu is partly lost and the remains of it suggest possibly bhyo (but nyo is afo possible) 17 The akgara subsequent to frutya is almost entirely lost and my reading as la Is by no means certain After it there are two or three akşaras entirely lost The remains of the next alpara suggest as to the vowel part, 1 the consonant part was perhaps kp or nke [21]

RANIERO GNOLI

4 śata | yathevaśabdayor abhidhāne vrttam darsaya- ti | iyam tavac chanda X yopamokta yatra punar i[ 5 bdabhyām sama abhihita para | yatha Lamalapatrā- kşī śaśankavadaneti ca || [[II, 32|] yatheva[

5 6 masādeh pratīyeta tadā sāpy upamāvaseyā samāsa- śabdas catra samkşepaparyayah adhyaye[ 7 yopayam dadaty arthibhyo uttatara na carthino vasaty antavari srakşyati tān apetya (tath) ā humkare[ 8 xx x x x xx x srji nijālaktakabhrāntibhājah skan- 10 dhe vindhyadribuddhya ni[

Fr. 24

5 ]xty xrū.y.[ 15 6 Jdiv x xity aha | vatinapi kriya ||II, 33] [ 7 ]yasamyam ucyate avyayibhavad iva v [ 8 ]nyasena prativastupamocyate [II, 34]] f

1 Read abhidhānavrttam? The stroke of the vowel e has been apparently marked as to be deleted The word dadaty has been corrected in the MS from vadaty. 7 The werd vasats has been corrected in the MS from vasaty. S In the MS for srakeyatr there was originally wntten pra- kşyatı. D This is the stanza No. 2 of the Candifataka by Bana. It begins by the word humkare in the preceding line. Instead of the reading "bhajab, oceurring here, we have in the ptinted text "bhajt. The entıre stanza runs thus humkare nyakkrtodanvatı mahati jıte fiâjıtair npuraaya flışyacchrngeksate 'pi kpa- radssrjı nijalaktakabbrantbhajı | #kandhe vind- hyādıbaddhyā nskaçati mabtpasyabito 'sun ahārpid

rotu.

[22 ]

Udbhaja's Vivarana, etc

b

1 ] stūpamā nāmopamā ) aļ 2 ]s) āmyam ubhayatvānuyāyitvena fakyate | 3 ]xram etat || vaksyamanodaharanany [ 4 Įsvādupākaphalā II, 36] S

Fr 25

]Xvādıśabdex[ ] aparyaya XI[

b

Jha X }ña[ 15 ] || sādhu sādhārana || II, 351 [

Fr 26

1 Jsvarūpākhyapaneccha[ 1234 20 2 Jnantarıtam iyam prax 1[ ]x xjalim bibhranam axlorucip s [ 4 Jstaro vrtha || JI, 38]j samanyagunās sarvabhavanam ha[

10 This fr is part of the same leaf as the preceding one 28 For vrtha we bave in the transmitted text mudha

[23]

RANIERO GNOLI

5 Jparo heyopadeyopehsx[ 6 Jmānalaksana[ 7

2 Jn [ 3 ]daharanam laks [ 4 mãn (II, 40] [ Jnodıtāh { ) sodāharanalaks-

5 ]s sendrayudho me(gha) Ivavabhase l

10 [[II, 41]] sa marutax [ 6 Jnkhah tam sanbhasam candrady(utim> salılam savıla(sam) [ Jdhanurmatram va tatpragr[ 18 7 ]m āha | sakracāpagra ([II, 42 ] [

15 Fr. 27

1 xxxxxxxxxxx tra cocyate |i sarvam sar- vena sarupyam nastı bhavasya kasyacıt | ||II, 43p [ 2 x x x tatpratikrt bh a | vapa sax X yos tu prakr- tyanupamanat ( atra hi[

Here begins the examination of the seven defeets (dora) of upama (II 39-65) The first of themis hinats or deliciency of upemint. egemphfied by the atanza so marutakampitepttavaoi ete 11 Here the text was appareatly abjah saakhah sankhab beng an explanation of the word abjah, oceurring in the stenra 17 There is not enough Irft of the akgara before 1ra to make its reading posmble above it there is a omall crois which is usually a mark of correction After co we have in the MS the akgara no. which i markrd, however. at to be drleted

[ 24 ]

Udbhata's Vivarana, etc. . LIBRAR 3 v.grahab atrocyate || it(i) duşanapanavap rem ka-vi (sy?)eti atra vikalpatva x pra X [ 4 x x pameyaviseşanabhota x x x x x x x x - tavya.ix Xtradyam paksam adhikrtyaha[ 5 X X X X X tare X x X x X X X X X cakora 5 upamāyā[ 6 etad uktam bhavati hx x x x x x xşa x X x xmāya XXx yam vindhya iti mukhal 7 drstā x x x x x x x x x x x x x x nanam adyu(ti | yat kimcit) kantisamanyac chasi |II, 44] [ 10 8 x xXXXXXX XXXX 1xte | ta- tha hy x xdanorthayama u[

I kimca kā(vyāni neyāmı la)ksanena mahātmanām | drs- tam va sarvasarupyam rajamittre yathoditam || [[II, 15 45l al ryayab puxxh y xxx a ilakşanam it1 | tam eva lakşa x xdh n x x Xnena śradhayā[ 3 rūpyam ubhayānanuyā x x dršya X (rā)jamittrākhye kā- vye yathodāharanam | sūryāmsusammī(litalocaneșu 20 dinesu padmāmlanirmadeşu | sādhvyah svageheşv iva bhartrhīnah)

1 The aksaras subsequent to wt are damaged and partly logt, dusa. (n a) seems clear,t he next aksara looks like pa (but sa or da are prrhaps equally posstble), n a is clearly written, the two next aksaras are hardly legi- bie Eventually we have, clearly visible, rem, which must be some senbal mistake for ram (oparam?) 2 The correct reading 1s perhaps vikalpedvaya, but the akpara after tva (or dva) 18 partly effaced and illegible 19 After onanuy& there is a loss of two or three akparas. Perhaps the text was here ubhayananuya(yitvena) drsya(te). etc {25]

RANIERO GNOLI

4 kekā vmeśuś śikhmam mukhesu | [[II, 46]| atrādıtya- kıranāmlatī X yanatā dmatvam padmapātraf 5 xs sādhayo dhāratāmātram atrobhayānuyāyı | etāv x xn kavaya upamānān [ 5 6 x doşah |j asambhavodāharanam āha } mspetur āsyād va tasya dīptāh sara dhanurmandala |II, 47] [ 7 śarā vārıdhārā iveti bhinnakramevasabdamīlanam vā yathopasthitevasabdabhsambandhanam vā | xx[ 8 xxxxxxx x x x xcandropamānopameyabhāvavā- 10 calam kımeid astītı prathamal

Fr. 28

1 ]darsayıtum aha | [ 2 Jsambhavalakşanadosa ucyal 15 3 ]vīta sasalakşmanah | f[II, 49]]

b

6 Jatra lınga[ 20 7 ]x varakhyapanaparam ucyate[ 8 lyah etat tritayam apy udahara[

After onila the leaf has suffered by the fire and the remaina of the three next akgaras are too scanty to make the reading certamn 5 The two vertical strokes marking the punctuation have been added below the line

[26]

Udbhata's Vivarana, etc

Fr 29

a

1 Jıvasabdaprayogānyathānupa- (pa)ttyānumeyam sādrsyam ast(i) [ 2 Jvidhānam apı doșatvam tathā- 5 nya(na)m ı(ty) ucyate | kalapuruşaka[ 3 Hix x bh y xtbharad babhau bharte vayos x x x xgambbirya x sarena[ 4 Imürtijļ kausumbharagarucirasphuradamsu- (kāntā | vi)bhrājase makaraketanam a[ 10 5 ] o cax x xkabhedah yaś capsarovibhrama- mandanānam sampadayıtrīf 6 ladhya bhriyate pi tu dhatumattaiva teno- pameyasya kakax x | upamā[

1 Fr 29 - Thas is a diseusson on the seven dopas which vitiate the figure of upama Sce II, 39 aqq 7 The words bharad babhau ete form part of a stanza Source untraced The akyara preceding va yo s is uncertain it can be read kte alao 9 This is Rotnapalt 20 The entire stanza runs as follows pratys. gramajjanavıšcşavıvıktamurtıb kausumbharāgarn* cirasphuradamiukantā | vibhrājase makaraketanam arcayantı bålapravalavitapaprabhavā lateva fIn the MS vibhrajase has been corrected from vibhrajate|I am not aure that the fragment with the words murti really belongs to ths leaf, and accordingly in this and in the following lines I have bracketed the words con tained in it 11 The words yaś capsaroº etc are a quotation from KS I 4 The entire stanza runs as follows yad capsarovibhramamands hanam sampadayıtrım sikharasr bibharti | balaha kacchedavıbhaktaragam akalasamdhyam iva dhi tumattåml I8 Supply (akālasa)ndhya bhriyate, ete The syllable pi has been corrected mn MS from d ba

[27]

RANIERO GNOLI

7 Jsatyamevibhaktivipariņāmavākyabhedākāń- kşayogyatasannidhih prax[ 8 Jlata yatha bhrajate tatha (tvam) bhrajase | yathanyah kascid akālasamdhyām bibharti tathā

5 himavān x[

b

1 Kyathapagaviga)hya tatha tvam asy avigāhya iti sambandhanam na kenacin nivarayitum sakya x[ 2 ]bandhah tac cārukāvyam yatra vibhakti- 10 vipariņamadibhir abhisambaddhas tan na carunapy xI 3 Jbhiprayah ity X X cāru kāvyam na bhavatiti doșa eta iti tatra hinatāyām a[ 4 ]r udahriyatex X X XtvamXt m sankhavattvam copameyaviseşanam naiva sambadhyata iti[ 15 5 Jxte punar yax x x x x (li)ngavacobhedayoX x viparināmasyāsraye X yatra[

1 Above the aksara me there are two vertical strokes, that posssbly belong to the superior line, which, in ths part of the leaf, 1s lost. 7 The words (yathapagaviga)hya, etc., refer to II, 53. 1º The aksaras eta, clearly written and legible, are apparently due to a scrıbal mustake for eşa. 13 The word dankhavattvam possibly refers to the stanza II, 58 (sa pitavāsah pragrhītasarngo manojnabhimam vapur ipa krspab | 6atabradendrayudhavan nisayam sam. srjyamanah fasineva meghah !), cited by Bhamaha as an instance of apamanadhikatvam The fault in this stanza consists in the fact that bere the mention of the moon is exceeding, the iankha having not been noticed as upameya. Sce Mammata, X, 142 . atro- pameyasya fankhader anirdese safino grahanam atıficyata ity adhıkapadatvam. 15 After the akgara y o the leaf is very damaged for a space covering two or three syllables, the vowel of the first character seems to have been 1, The erpression lingavacobhedan occurs in II, 52. 16 The two characters suhsequent to o fra ye are uncertan, apparently the akgara I have read as ya looks more like ye l

[28]

Udbhața's Vivarana, etc. 6 Ju ity apare t x xļmūrtyāvakay įx lingax yetyubhayanuyayix Xna doşah yatha candra iva mu[ 7 Jļrāj īļx xyādişu bāhulyena punna- pumsakayo x kyavišeşaḥ yathā 8 Jkifcid upadarditam adhikaviparya- yah punah aprotsāhyasāļ 5

Fr. 30

A

]tāyā upamax[ 10 Jpan na dhal

b

15 ]xrājjax r l Jeva bhamahal

Fr. 31

20

1 Jkrtacarucandra[

1 As to the word {murtyavakay ), see the face a, 1 9. (note) 2 After onuyayı about two characters are lost in a break and stain, posmbly onuyayitvene is to be suppbed. 7 This and the three following minor fragments apparently belong to the Bame leaf as Fr 27 [29]

RANIERO CNOLI

7 ly [ 8 ]r sa sampratix x x xdi [

Fr 32

a

]sabdābhisambaddhapa x [

b

J(na)pumsakānıvad [

Fr 33

Jv [ Jnc ttax[ ]ma[

b

Jme[

[30]

Udbhata's Vivarana, etc

Fr 34

12 Jhnatayam iva gatix x ul Jdhıkyam upamananam [II, 61] n1- badhnantı kavayah yatha gayantu kinnaraganah saha{ 5 Jyasas te | tatha be ham- sas tavad ambhoruhakapisarajorusitangāh sa[ 4 mālāmalınasakaladı[ ]vyā(laja)lahmilapronmilanmegha-

5 ]pamānaviseşanānı ivaśabdapra- 10 yoganyathanupapattya xnus u[ 6 Jha | aty ukta upamabhedo vaksyate caparah punah( | )upamāder ala [[II, 65] [ 7 Itah a upamābhedo lakşyate tenarvam vyākhyātā upamābhe[ 15 8 ]pamayāh ka.ya- pārthādı | sampratı kim kryata ity ā(ha) x x i[

5 The words gayantu kinnaraganab etc up to yains te in the next line belong to a stanza quoted in the Subhasitavalt of Vallabhadeva 2494 (anon ) The stanza runs thus gaymntu kimnaragannb anhn kımnarıbhib šrırāmacandra šıkhareņu himācalasyal iesendukundadalabalamroalanalintharaharaharaha- sasıtam yafas te | The last part of this stanza with the vanant kar- pura for nihara 19 quoted by Vamana IV 2 18 6 As to the stanza he bamsas ete see the Addends p xLttr 7 The word orajo has been corrected in the MS from obhajo 14 The first four akgoray of this I ne are mutilated from the remains of them I think the text was here tah(?) kajcid upamabhedo ete 16 The word tena is bere repeated twice and the first one is marked as to be deleted 15 The ayllable bhe after upa ta is almost illegible the vocal e part deems to be formed by the vertrcsl stroke of the long a andt he aiga of i I6 Apparently in the MS first was written m ah and then correeted on a secunda manu into y&h 17 Read kunyåparthadı? [31]

RANIERD GNOLI

1 ]d(i)ty artha Xpunah praXmyo vacano nuprasade i[ 2 ]tıreko vıbhāvanā | samāsārısayoktī ca sadala- nkrtayo parah [[11, 66] f

5 3 ]ddista esam ca prabhedasamdarsanartham aha | vaksyamano(ktivisa)[ [II, 67]] [ 4 Jssal vakşyamānavısayataya ca dvividhab śesās tv alankārāh ekarūpa[ 5 ]tı vyaktam arthantaranyasam spaștam eva

10 vaksyatı granthakārah tathātisayoktir a e[ 6 Jtad uktam bhavati yathayam aksepo dvi- prabhedo visayax x de x mibaddhaf 7 ]tsaya | aksepam itr tam santas sansanti dvıvıdham yatha || [II, 68]] yatra kāvya abhipretā[

15 8 ]traksepo lankarah X x dvividhah[

Fr. 35

1 ya)d evastu vanyena kim uktenapryena te i [II, 69] bhavantam yadā na pasya[

2 The aksara subsequent to pra is partly effaced and ill-formed, the vowel part (o) of the akgara m y a (m y o) is perhaps marked as to be deleted Should we read prathamya° or prathamyao? 3 The word samisao hes been corrected from samayao 6 The words vakeyamanoktio arc a scribal mustake far va-

7 The two akgaras ssa and Iva seem ciear 1º The reading of this line after vigaya is very doubtful The akgara B1 serms to have been marked as to be deleted and the ngn of the $ is apparently missing 13 As to the reading aksepam see the Introd p XLI! 18 See, for the reading owtu vanyena, the Introd, p xLt

[32]

Udbhata's Vivarana, etc

2 na te ıty atrānabhıdhānam eva mām apasyanti mrı- yate vasyam ıyam iti [ 3 n na tavoddhatıh ( f ) ko vā hetur alam sındhor vıkā- rakaranam prati || II, 70]| va x[ 4 dhıtsayaı tadvıdhānam pratısedhayatīva ko vā hetur ityadına | athavatrax x xhax x x y [ 5

5 n ıtı vısesah pratiyate t arthāntaranyāsam vyākhyā- tum aha | upanyasanam a |II, 71] [ 6 kte yat präkaramıkasyārthasyopanyasanam prāhara- nıkarthabhısambaddham tatrarthantarasyasa[ 10 s samarthamyah | samarthakas tv aprastutah nanu ca upanyasanam anyasya yadarthasyanye x X tasya[ 8 x xpūrvottaropanyāsabhedāt | tatra yadā samarthaka- sya pūrvopanyāsa[

b 15

1 x x(ragını nalıne) lakşm(ım) dıvaso nıdadhātı dina- karaprabhavam | anapeksı(tagunadosah paropakārah satām vyasanam ( ) mkam dınakaravyāpārahs tv aprākaranıkah yadā pu- nah prakaranıkam upanyasya tasyarthantara[ 20 3 tallakşanad udıtād uktāt prastutād anyasyarthasya ya (u)panyasana[ 4 kām bhīmani viviAsor na tava vyatha | sādhu vāsa- dhu vagāmi pumsam atmaiva sansatı II, 72 [ m aha | yasmad agammni subhasubhe pums im atmar- 25 vāntahkaranam ern kathayatı | atra purug [

found in C also see the Introd p XLI 10 Thia stanza 1s from Harpacarita 22 23 Correeted in the MS from vivakpor [33]

RANIERO GNOLI

6 x xtmamulhyārthe punar ātmasabda ātmana ātma kathayatıx x x xha sareroddhoratsa[ 7 sayıtum āha | hısabdenāpı hetvarthaprathanād ukta sıddhaye | ayam arthantaranyasas sutarām vyajyate

5 [II, 73] 8 na bhāvasya sıddhır bhavatı tadartham ayam arthan- tarany asas sutaram vyajyate [[II, 73] [

Fr 36

a hatas tarur ayam coccaih pa [[II, 80]] [ 1234 10 Jkyārthaparısamāptı[ 3 Jtmasama x x I 4 Im upal

15 b

6 Jatrāpy udāharaņaį 7 ]soktısvarūpavı(se)sana[ 8 ]vagamo 1[

20 Fr 37

1 ] āntx pratyayah spaştam punar idam udaha- ranam samāsokteh upodhal

23 The word upodha is the beg nming of a famous stanza fre queatly quoted as an exemple of samasokts (ef f ex DhA p 109 AS p 109) The entire stanza rons as follows opodharagena vilolatarakam

[34]

Udbhata's Vivarana, etc.

2 lpuro pi mohād galitam na ralșita(m) | atra sasirajanivyasanapreyah prax x x sahasu x ta[ 3 Jpraveso x bhimato sya granthakrtah tenāsya granthakārasya(bh)iprayena vyatirek [ 4 Jidam punar atrarthatattva(m) | yeşam alankara- 5 nām bhangyantarena nibandhanasambhava X tesam sa[ 5 ]xnya x x x x (la)nkāravisayasya sakalasyāpa- hāras syāt | ucca xstavyakhya[ 6 ]manyante tisayoktin tām alankāratayā yathā [[II, 81] lokātikrantago! 10 7 Jyoktiķ syād etatparhārartham nimittata iti | etad uktam bhavati nimi(ttam ki)mecid āsritya ya[ 8 Jriņyā candrabhāsa tirohitāb | anvamīyanta bhrngaliva(ca sapta) cchadadrumah || JII, 82]) a[

15

parijñāyamānāļ bņūgālīvācā madhukara- pankt(ayab) tenānvamīyantā[ 2 Jeyuta syat phaninam iva | tatha suklāmsukany

tathā grhītam dascna atkåmukham | yatha samastam tımırāmsukam taya puro' pi ragad galitam na lak. gitam This stanza is traditionally asenbed to Panini Many verdes of this poet, most unlkely to be identified with the great grammaran, have been collected by P Peterson, JRAS, 1891, pp. 313-316, The identity of this poet with the grammanan has been maintained by R Puschel, ZDMG, XXXIX pp. 95-8, 313-316, and recently by K Upadhyaya, IHQ, XIII, p. 167. The date of the 9th cent., ascribed to him by L. Renou (La Durghatovrt de Sarana- deva, Paris, 1940, vol. I, p 73) requires to be brought back by one century at least. The varant puro'pi mohad galitam na rakertam is interesting. Possibly this was the orginal reading, later on modified, owing to some need of symmetry, into the more difficult puro'pi ragad ga. litam na takartam. which became the lectio universally accepted. 7 Read tac for ue? The akyara is ill-formed and by no means clear.

[35 ]

RANIERO GNOLI

asann angesv ambhasr yosıtām || [[II, 83]] ambhasam yadı[ 3 Jsuklamsukanı X X tāntariyām bhaveyuh | ayam atra X yah eva ta yosıtah ko sa[ 5 4 ]na kenacıd apıxsaseti || bahutaraprayogatvam atısayokteh khyapayıtum aha | ity evamādır u[ 5 ]dharmātısayayogatah ( || > [[II, 84]] atısayo- ktır nıbaddha kavıbhis tatra tatra sarvaiva sātiśa- yoktır ıtı ([II, 84]] yatha[ 10 6 lyaśah patetām | tadopamīyeta tamālanīlam āmuktamuktalatam asya vaksah į| tathā dhyanāmallı[ 7 Jntanavasūtacūtakalıkāma r hitasra ibhih jhankarah prathamam grhe grhajane[ 8 ljax ep nkalanaru x Xtam nana. 15 jatarayannalah || ity araryon[

Fr 38

]x xevam tarlu suv ü[

1 As to the reading asann see the Iutrod p XLII 3 The danda ia subscript 9 sarvaıva satısayoktı MS sarveivatıāayoktıs tu AB || The stanza introduced by yatha is borrowed from the SV by Magha (III, 8) it is quoted by Vam, IV 3 10 The two first padas run as follows ubhau yadı vyomnı prthakpravahay aka- fagangapayasah patetam | Cp HC also p 347 11 muktalatam muktaphalatam MS The akara pha is obv ously due to some scnbal mistake 12 Above the syllable ka in kalika there is a sign of reference, concermng the addition in the inferior margin of the word gundha 14 The reading nana is uncertain perhaps nanu (the second na [or n a] is added in the margin)

[36]

Udbhata's Vivarana, etc ]pamānena yat tatvam upame [[III, 14t f ]kın tad ity āha upamānenopal Jrasthitih X s X X 1 1[

b 5

]mānopameyal Jnta ity evamādı svala Jpūrvi vācakasaktıbhedād bh [ Jvam vyayam anyatha[ 10 Jyady apy e[

Fr 39

a 1 ] īya tada osțhyo vakāro yadā tu vyālam sar- pam vina va sametya y [ 15 2 ]satrughnas tvam Asitau janalasthiteh itt suca- ritais suktibhra y [ 3 Jdısabdadvayavıvakşayam laghuprayatnataro lankarah nasyalo[ 4 ldrst Xpāyāt kadambakusumastabaka iva mu- 20 khanivāsu X [ 5 ] y yo ntodittam iti | tenox X Xvisnye dhatsa e[

1 The words ya tada osthyo have been corrected in the MS from yade osthyo This is part of the commentary to III 18 18 The aksara e is somewhat ill formed and ancertarn but the read ng of tru (or bhru) seems sure 17 Under the akgaras oritars there are traces of two subacript akparas the firt of them elearly written can be rend as kr the iccond one is an certain possibly p a [37 ]

RANIERO GNOLI

6 ]dvayi gatih x x x y tpratyayah X tra vax[ 7 ]s u[ 8 Jsleab[

b

Jso yad gil 1234 5 1 Jlankārānām prākaraņika[ 3 ]sampraty arthax x x x x x slokadvaya yayā[ 4 Jıva | ratnavattvād agādhatvāt sva(maryādāvi)- langhanat | [[III, 19-20] [ 10 5 ]pratipādakabhāvah kintu tābhyām prākaraņi- kāprākaranikaļ 6 ļr t į yatra tu šabdānām atyantasarūpāņām api dantyausthyalaghupra[ 7 ] āntarapratibhā | tayālankārantare vākyantare 15 va pratibhotpadyate x [ 8 ngi dadhati sy [ Igenoddīpayantī smara | kancīdāma nitambasa-

Fr. 40

20 1 Judbhavandhadbhavarupay h nana y[ 2 ]s bhāvam mrduvikasatkamalākaran kiņcil 3 Jyadapi dvayor bahtnām cāprākaranika(tvam)[

19 The word .arupan&m was first wntten as saropandm and then corrected into saropanam. 14 Perhaps (rth) antarapratibha; as to the meaning, sce the Introd , p. XXXVI 16 This 19 a stanza in the Sardulavilridita metre.

[38]

Udbhata's Vivarana, etc

4 day [ ]s samaghavate | hrtam

5 laksanāt tulyakāryakriyā |[III, 27] [ 1x t g X atra ca

6 jkārād apetasyetyādı (III, 29]f { adhi- 5 kāra upavarņanāvasarah kathamcı 7 ]kso va apratyekam aprastutaprasam- seti samjnā | stutigrah x x laksi[ 8 ] | tathā hi sa X topavarnana(yām ka)s samudrasvarupam upavarnayet | upavarna[ 10

Jlankrıyākā y sya sahrdaya(hrda)yā- varjanalakşaņāsahāprastutapraśamsā | aļ 2 ] x visesabhāvād vā ex tāvad dvividhah sārūpyād vāprastutapraśamsā[ 15 3 ]sa(m)sāvat granthakāras tu yadvi- (ses)abhıdhanaprastutam samanyaprastuta[

6 As to the term upavarnanavasarab, cp the Introd , p XXXVI 8 Should we restorate stutigra(hagopa)lakoz(ta)* The words tathahi ete occur in the fiveka p 358 ep the Introd, p xXXVII The word sa x topavarnanayam is apparently s eerbal mi atake for parvatopavarnanayam, as we read in the Viveke and the sense requires After upavarna the leaf is broken, but, in all probabihty, the text was here (as in the Vwekd) upavarnane va fi şfavışarhanam avadyam bhavı 19 Here the text is vitiated i seems The words of Hemacandra (} weka, 359), santareņa sabdavyapareņa gocarikrıyamapah prakaraņıko yo rthas tena sahaprastutasya karya- kārapabhavadau sambandhe satı sahrdayahrdayā varjakam alamkārarupatvam etasyab, look Ike a para phrasis + perhaps somewhat amplfied - of this hne 17 The vocalc part of the syllable ny(in samanya) is doubtful more resembling e or o in the stanza pripitapranayı avadu kale pazıņatam babuļvină purupakareņa phalam paiyata [39]

RANIERO GNOLI

4 bhavati[ Kkāl)e pariņatam vyādhikāri yan na

5 ]dhānyopavarņanāyām sakalasa- msāra[

5 6 Jinu bbakappiam jan na niaighari[ 7 Jś t x panaparebhyo prastutapraśamsā | yatrāpi sārūpyax xt x[ 8 ltamārīcavadhe rāmasauryopavarnanaya keśarihari- navyapara[

10 Fr. 41

2 Ipārtha(m) kāvyam apa- gatartham mirarthax x y [

fakbinaml, cited by Bhamaha as an instance of aprastutapra famsa and to which this passage probahly refers (cp. the next line), there is a perception of the samanya from the visesa (ep. Induraja, p. 36) 3 The sense of this line must have been that, just like the fruts of trees are produced irrespective of the absence or presence of puruyakara, in the same manner there is no relation of cause and effect between puruņakara and all happenings in the world. Everything is depending on destiny. From what we can argue from the scanty remarns, the text was here phalasya daivapradhānyopavarnanāyam sakalasamsārasya daivapradhanyam gamyate, or something uke that. See Indurāja, p. 36. 5 This is part of a stanza quoted by Ruyyaka also, AS, 133 The enture stanza runs as follows tannatthi kimpi pahano (but here we have (pa)tnu) pakappiam (here bbakappiam) jam na pial- gharanie j anavara agamaņasīlasa kālapahiasea pā- hijjam. 8 Between the akparas ta and ma there is an akpara sabscript, appa- rently ntho 10 Fr 41. - Tlis is a commentary to the first stanza of chap. IV. This fragment (especially the face a) 1s hardly legible, the akgaras being almost completely discoloured

[40 ]

Udbhata'a Vivarana, etc 3 taksyat padalal Jadhantam punas yax x

4 Jkavarnānāmidănisastra Xntıyal 5 Ju y x xkrtya var- 5 pasa(mgha)ta ity [ 6 ]kşāņām paraspa- ram | mrāka [[IV, 4] [ Jkatra padarth sarvatra x x sar[ 8 ]x x mã | x xhuh ye[ 10

b

1 ]da x samsarga x x x x x x x x x x h padärtha- tān teşām aneka[ 2 ]xm aha | buddhan tux x x x x x xtve pl pratıksanam ||IV, 5]| 1 yatah kra[ 15 3 I y te | na cāsau vaca- sam sambhavatiti e[ 4 ]varnanubhavavastha[ 5 J(dhīr a)ntyasabdavısayā vrttasabdāhita [[IV, 6]] [ 20 6 jva smrtıbhāsīnı vākyam ısyantı punah[ 7 ]kşavıśesah atra dūșanāva- saram darśa[ 8 ] ādūşaņasahasropamıpātaka- 25 dalıtaf

14 Strangely enough between the words X m aha | buddhau tu and da xsamsarga in the previous line, there is, in this part of the leaf, an intermediate hne in which the words ucearyante te can be read

[41 ]

RANIERO GNOLI

Fr. 42

]guņā[

5 JLrtātmanam samskrtamal ltatkathadau yadi m [ ] udāharanam x x[

b

10 . Jtatkrte yu t [ savıparyasaka[ ] ād anupetam mārga[

15 Fr. 43

]x x punaruktam de[ ]x yadı | 1hax xI 20 Jbhidhasyata it [IV, 13g [ ]x | bhayaso [[IV, 14]I [

15 Read sad?

[ 42 ]

Udbhata's Vivarana, etc.

. Jviruddhyata 1[ Jx x asthi[ ]dainyāga x X stīvala[ ljam vyakhyatum aha yal

Fr. 44

1 ] iyāh rājānaļ puraś cırakālapratyāsatti- 10 kax[ 2 Jıti į apakrasya vyākhyātum āha | yathopa- desam (IV, 201 [ 3 ] athāsamkhyam masam rtum samvatsa- ra xrņāni visa[ 15 4 ]vipūrvasya vā kab karaņavyax X sabdo tra nipata X sraya X [ 5 ]x x xnau dadh X X XXy au kiritenda[ 6 Jx x x pakramasye x X x hara X varnayanti sux x x drsto[ 20 7 Jsistakrama x x x pa)darthanam sabdabhıdhānavela X utavakramās s [

8 Fr. 44 - This 1s a discussion on two flaws or dopas, viz apakrama and sabdahina. The nature, etc , of dosas has been discussed at lenght by V. Ra. ghavan, SrP. pp 216-257. 19 The word apakrasya is apparently due to some seribal mistake for apakramam or apakramasya svarupam- 14 In the MS, the character ptu is ill-formed, more resembling ntu or something lke that 18 Before the word sabdo there is in the MS . space left blank of sbout two akparas

[43]

RANIERO GNOLI

8 ]x sya mvrttatve smaranakramo baliyan ity an Xsa x x tatra smrti[

1 ]x s tvam paścā x ghanıśāsu smrtır it na kımcid 5 atra viruddham | atha puna[ 2 gena | sabdahinasam(da)rsanartham aha | su- trakrtpādakareştetyādı [[IV, 22]| | sūtrakrt [ 3 Jkhyanavakyam ucyate X rtam ca bahulyad evam uktam | x xkrtām pa[ 10 4 ]x xjñatve purusasvabhavavasād eşu sā- dhusabdeşv apy x x x teşām aprayoga[ 5 ]bho yad vrttam asmi(n) grhe | saubha- gyavyayaśanka[ 6 }h srasasau X pathanaśapyamXvamśa- 15 ta[ 7 Jktam sūtrakrtpādakāreștaprayogād yo nyatha bhavet [cf IV, 22] 8 ]x vyākaranam anıx X X X saıva ya[

Tr 45

20 a

1 ] apı na vatsalatām upai[

D The two akgaras after uktam are almost entirely lost the vowel of the first one was pezhaps e or o the second one was a composed akşars, the lower part being like to e 12 These words are from a stanza cited by the Subhasstavalt also 1592 The entire stanza runs as follows kale nılabalahake satadıtı pritiprade barhınam ascaryam kathayamı vah puta bho yad vrttam asmin grhe | saubhagyavyayasanka. yaikabhavane kantapriyabhyam aho manibhyam bata ratrım eva sakalam cirnam pravasivratam | Here the grammatical prayogs concerny the word cirna 19 Fr 45 - One face of this fragment (the present face b) 1s a comment

[44]

Udbhata's Vivarana, etc.

2 J āvatseśam praty avyabhi[ 3 ]vahanti naubhıb krīnutā[ 4 Jsalile vahantī naubhiļ krīnuļ 5 ] it cāham anugrhīt [ 6 1 im i[ 5

b

4 ]matrarthapratiyat [V, 1]] [ 5 Jkim it sugatam 1 x x[ 6 ]x pratity aha prayena [[V, 2] [ 10 7 ]margapravartanartham s [ 8 ] mātrapradarsanam prayo

Fr. 46

15 34 ]smābhır vyākhyātā vi[ ]taddharmo pi na setsyatiti V, 15] etad uktam na tya[ 5 lyati yatra pramanantarapra 1x xnnedha i[ 6 ]rsayitum aha svasiddhantaviruddhatvāt IV, 17] [ 20 7 Jxram prajanānās s asamana x dhama[

to V, 1 sqy I am doubtful whether this face is e comment on the last lines of the 4th ch or to the 5th ch After the akgara nau therejs a danda, which is due apparently to some seribal mistake 8 This is a mistake for tanmatrarthapratitaye, it teems

[45 ]

RANIERO GNOLI

8 ]rtham tasmat sarvagamavirodhini yatha suci.sunu- straiņi | strīsa[

1 (n)yapramāņapratikșe[ ]xh pramānapravāha işyata eveti sāmā- 5 2 ]p tam etad anyatrā- smabhih prasiddha[ 3 ]pratyakşabādhinī da- rsayitum aha (prat)y(a) [[V, 20]] [ 10 4 mā iti [V, 20] { hetulaksa[ Jsti upa ușnas candra-

5 ssamudāyas sādhyadharmas cātra[ ]m āha dharmadharmi-

6 Jsiddha ityadi 15 dharmiņi[

Fr. 47

a

1 (pa) kşanupadana ityadi V, 23] | yady atra para-

20 pakşaprabāņ· xkriyete tatra da sif

1 Hete the MS has auffered by the fire and the reading is by no means certain, after the aksars fu, which is clearly wntten, there is the loss of an akşara (the vowel part, i, still visible), the remains of the subsequent akyara suggest su, and the next akpara is seemingly rtu (both the ngns of the r and of the u are clearly visible) Probably the text is here vitiated, the cor- rect reading being apparently that of B, yatha sucin tanuh straini (for the meaning see, f ex . the Bodhicaryavatara, IX 8 and the comm. of Praj- nakaramatı) Cp. the Introd also, p XLI. 11 The consonant part of the akpara subsequent to sti is damaged and uncertain (r or k'), the vowel part i5 u I think the correet reading is here kupa (quoted as a variant in B) and not rupam, as we read in B. The word eandrama looks like a scnbal mistake for the reading of B, kşa- pakarab, required by the metre Cp. the Introd, also, p. XLI.

{ 46 ]

Udbbata's Vivarana, ete 2 mıtyah sabdah krtakatvāt ayam ćabdādwyaktx x x pratyanyatarasıddha! 3 rşto yo dharmas tadanugamena yas sadrśah paksas samapaksas tatra ca yas san ity ukter iti | [ 4 X tatra ca jātah pratyākhyānān na kaścit kenacıt sadrśas tat katham ucyate sadhya(dharma) nugamat [[V, 24]][ 5

5 X Xpacaryata ||V, 24] itı sāmanyam iha samvrtişad dlu nabhipreta(m) yay sadhyal 6 x yc y y sadharanadıs sapaksa eka ıva ekārtha- krıyakarity evopacaryata itı vyā[ 10 7 XX | sapakşena yo vısadršas sādhyadharmānāfra- ya(h) sa vipakşas t x xvyatrxtu hetus s [ B x x (da)rsayıtum aba ıtı dvayakānugatıvyavrtti lakşmasadhu(tā) ĮV, 25] ) iti sa(py e)ka bhangyāx[

15

1 X x Xvrttuuşpatti te lakşmaņo lakşanasya sādhutā tad uktam bhavatı bharatāx[ 2 x xkalakşanam punar yatra sādhyābhāvo hetvabhā- vah kh(yapı)ta ıtı || drştāntam lakşayıtu[

$ The akgara after pakya is somewhat ill formed, more resembl ng ip o su, and my reuding ta i conjectural 7 Rend anmvrtiand this is an sllumon to the wellknown baddbut view, according to wh ch skmanya u a mental construction only aud has not a real subs stence Seef ex PI III 3 arthakriykyamertham yut tad atra paramartharat | nnynt samvrtioet pro

O The reading v&dbareudis (for ukdharaoadis) is donbt ful, here the Jeaf is very damaged and the partly +faeed akyuras looh more hke saghnpadis than vadharensdis But the meaning" 12 The vwarga ta m mnng in the MS. ti As to the reading see the Introd., p xti

[47]

RANIERO GNOLI

3 xbhute trtīya sādhyasādhanadharmayogī yas sıddha (1)ty arthavisesasadhyadharmadrstānta[ 4 x x x y drstantah drstabhasah punas sādhyasādha- nayob pratyekam punas samudāyax[ 5 5 sya darsayıtum aha sadhyena linganugatir [[V, 27] ıtyādı yatra dhūmas tatrāgnır yathā maha(nasādau) [ 6 saldadav itı evam dvidhanyair drstantalaksanam uk- tam dūsanānı darsayitum āha dū [IV, 28n[ 7 nesyata itı prastave dusanalaksanaX sa X X X X Xntı 10 tat Lim ival 8 (pra)trjnadım dustam yas ca pratix x x x y te ta- trax Xh p [

Fr 48

a

15 ]dusaņabhasa itr IV, 29] [ Jyas āt saprapanca[

1 The read ng of the akparas bhute trtiya (read trtiyam?) u doubtful In the place of these syllables there had been superimposed u fragment belonging to another sheet (Ist Ine fuklamsukani 2nd ) na kenaeid apa sec the present fr 375 Il. 3 and 4) as soon as the latter has been detached the underlying syllables have crumbled and I have barely had time to read them Apparently these words belong to another part of the commentary I The akpare subsequent to siddba is almost entirely effaced and the leaf senously damaged The I ne may be completed yatre nagnir na tatra dhumo yatha (L 4) s aliladav. etc D The words neyyata iti prastave are a reference to IV. 2 11 The two akgaras subsequent to dustans are uncertan

[48]

Udbhata's Vivarana, etc.

Jyathā[ Jstrıx xya darsay [

5

Fr. 49

4 ]kāvya[ 5 ]pādıstan yathabhibitam it vyarthal 10 6 Jtayjnaih kavyaprabodhanartham[ 7 ]kāsrayam kāvyam āgamās tattvasarıx ([V, 33]l [ 8 Jkāvyam itı nātra vise[

1 ] yâna x x x x neneda[ 15 2 ]tz | hetunā drețāntenānanyal 3 ] avataratı tenena pra i[ 4 ] yadısadharmyasamadya jaf 5 Jna Arta itı | anya x vana[

Fr 50 20

1

2 X X X tab traya ime dosah ye punar ajbana X sam- sayajaanaviparyayas e t [

12 Apparently a mistske for tattvndnmsi' [49] 7

RANIERO GNOLI

3 vasyānutpādyatvāt tatkaranavırodhah esām u(dā)ha- rana(m i)ty aha ( | ) Lasā hara kasa hara [[V, 53]] [ 4 tıtvam yatha sarārmnām evam kururadīnam apa y naika x x X kārānām raktanayan [ 5 5Lavyadrştantalaksanartham aha | uktasyarthasyetyadı [[V, 55] upanyastasya vastunah pratibimbapra[ 6 mänam evāstv ıty asyopamāne ntarbhavamanyanā na X vah praha apariharah na hetvanabhidhanakr [[V, 56] [ 7 yamiv acipray ogas ca taddrstantenasti it nopamayām 10 xntarbhavo syasankanīyah vastumi[ 8 yanopamanupravesah yatha tanur iyam kva vilo- canahāriniX X ga X gavacata[

1 X am apy asyantarbhavo nasti | upamanopame- 15 yabhavadıvatx X X syapy atrapi[ 2 sadhanayor vrttir ukta tatra hi nesyata iti [V, 56[] tatra hy p x x y sarvax X Xuyapyas t y [ 3 t punar upamanalaksanam sarvatrasty eva kim punah karanam sax sadhyasadhano X x x upamā[ 20 4 1 | etad uktam bhavati upamane Kutsitasadhy asādhana[ 5 sayayogadıtsur iveti tada klıstam karyax x x şahX X mānam udāharanam tadā[ 6 vrddhanusitatvad ita |V, 57| hetuh purvakas t X Xya Xtivrddhax x Artayuge bhavax[

2 Owing to a seribal mistake, the words kasa hara have been repe ated twice 7 Corrected from "pemånam' Should wr restorate "pamano"? 11 The words tanur iyam kva, etc are part of a stanza in the Drutavilambita metre 20 The akgara subsequent to ne is doubtful it looks hke tu or kta

[ 50 ]

Udbhata's Vivarana, etc. 7 ]ddhāguna āx x ·y samyag darsayi- tum upakramel 8 Jvyakti[ 41521 Fr. 51

5 1 (sva)rānām vyanjanānām ca dvivi[ 2 J m pravrttasya grahasabdasya[ 3 lyam ratnam kryata ity aha[ 4 ]x harsaya bhavati y [ 5 Ipravahadarsanad veti[ 10 6

4 ]x k śabdaśabdāpr { 15 5 J(sa)mudāyah katham arthavan syāt | krama[ 6 ļkadeśakalatvāt pad [ 7 lkrto vagantum | dārūni bha[ 8 Jıtyadi | yasmad varnanam ta[

Fr. 52 20

A

1 lny evamvidhas sabdaprayogo neşyate

[[VI, 28] [ Lidrk tarhi gabdah prayojya ity aha | kra(magatam)

4 Fr. 51 - This i a commentary to VI. 1 4qq S This u apparently a mistake for kriyata See Fr 10 b. 1 6 18 The word arthavan as apparently a scribal mistake for artha- van. This is a commentary to VI, 9, it seems

[51 ]

RANIERO GNOLI

2 ]xh kramas tenāgatam sabdam prayu- ijita Lramaparyuanam ca dvividham salsad tā[ 3 Jksy ca sa eva kavipravāha anidampra- thama(taya) pravrtteh yasya punah kasyacı[ 5 4 ]vajnatvenasmrtivipary ay asmrtiyam bhara X x xtadanurodhena sabda[ 5 Jrutısukham prayuujita yady apı śrutıka- stax x x x xjıhladadayas [ 6 ]x mukhesv atyantadarakhyapanartham idam 10 uktam srutisukham ity x xmatıyax [ 7 ]tapratitikaranasathyam tadapi khstadoseX X x x sabdaśra X [ 8 Jrne pi vakyapadapūranartham prayunjana drśya x x x x x nartha 1

15 b

1 Jlankaram svaravyanjena ca ~ tı | ayam stutya x x x x x xlanbarebhya[ 2 In ke tyantam caravah ke ca na carava ili darsay ann aha siddho yas copa [[VI, 29] [ 20 3 ]x yah mrjer ajad(au) samkrame vibhā- sā vrddhır i y tı ityādayah( | )ana[

8 (a)jıhladadayas cf I 53 13 The vowel part of the aksara rn is lost 16 After svaravyanjena the text is apparently corrupted and the meamng of the honzontal stroke after ca is by no means clear to me Perhaps the loss of an aksara? Poesibly this is a corrupt reading from s va ravyanjanacaruta (or °caruteti) 20 The words mrjer a (corrected by me from MS mr)ena) Ja dan samkrame vibhaea are a quotation from Kasika I 1, 5 cf also MBh I p 127 (Virnayasagar ed ) 21 The remains of the alsara subsequent to vrddhir i suggest vva o dhya (the vowel perhaps is long) but this is apparently due to some scnbal mistake and the original text was surely ie yata

[52]

Udbhata's Vivarana, etc 4 ]yogah karyah na tu yogavıbhāgajam |VI, 29]| ıtı yogavıbhāgade fistais si[ 5 ] iyānyatarasyām ıty atra yogavıbhā- go dvıtiyārthah | vmā vātam vmā varşam vil 6 ]vrddhipaksam |VI, 31] ityadmna sam- 5 krame guņavrddhıpratı(sedha)visaye mrjch svarādau[ 7 ]patanto baspabındava it: ||VI, 31 | | "sarúpanam ekasesa ekavıbhaktāv " ity [ 8 Jaś ca bhavānī ca davo varunaś ca varunāni varunānitı tad āha || sa |VI, 32] [ 10

Fr 53

1 lyabha x xkratibhatena na yujya- te prayojyo yathāsamgī(t)1[ 2 lyatha x x x xna | nyasafrantho yujikrita- 15 dahkantavoja x pāsa[

For yogavibhagadı siştais? or yogavıbhågådeh fietars? 3 The words iyanyatarasyam are from P, II, 3, 32 The entsre sutra is prthagvindnanabhis trtryanyatarasyåm # The wordg vina vatai, etc, are from a stanza cited in the Kanka II, 332 the entire stanza is as follow vind våtam vinå varşam vidyutprapatanam vina | vina hastikrtān dosan kenemau patstan drumaui 8 The words sarūpānām ekaiepa ekavibhaktan are a quotation from P I 2,61 D The word dayo is here e corrupted reading for bhavau 10 Instead of varunanits the sense requires here varugavitr 13 The reading krnatibhavena i doubtful the akyaras being partly effaced corrupted text" l6 See P, 3 3 107 oyasasrantho yuc After yuc the text 1 apparently vitiated 16 See VI, 50 updseneti ca yucam nityamisch pre. yojayet

[53 ]

RANIERO GNOLI

3 ] x x maņantāt x x iti nīpratyayah < ! ) yatha paurandarix Xtena raktam i[ 4 Jkardamat thag itix x (pra)tyayau tāvat krtau yatha māhārajano lākşikab ( ) ) kumudaį 5 5 Jākşikas sā(stri)kah < | ) "tasmai hitam" iti prakarane sarvāņ nasya vādanam i[ 6 ]" (prthvādi)bhya i(mani)jv "eti pratyayo bhimatah yatha pațima (I)aghimāx [ 7 lyasyetideti lopavyaticeti lopavyatirikter i[ 10 8 Jrinitambasvayam iti | matupprakarane X X . adayas tathal

b 1 ] | mali mayi "adabhyastad" iti jher adadesas sammatab ( | ) ru.i[ 15 2 ]xh prayoga istas tatha "videh satur va- sur " ity anena sutrena ghahpra

2 Cp P, 4, 2, I. tena raktam rågåt. 3 See P, IV, 2,2 lakşārocanātakalakardamāt įhak. 4 Ser P, 4, 2, 2 and 4, 2, 35. However, the leaf 1s here broken and the remains of the third akpara of maharajano auggest more ra than ra. 4 As to the word kumuda, see P. 4, 2, 87. The words tasmai hitam are a quotation from P. S. 1, S. 7 Tlus 19 x citation from P, 5, 1, 122. 9 Tlus line is corrupted, it seems. 10 The words ]rinitambasvayam are apparently a corruption from VI, 55 janudaghni sarin narintambadvayasam sarab, on which verse see P, 5, 2,37 pramane dvayasajdaghna- Jmätracah. 10 The words matupprakarane refer to P, 5, 2,91 tad asy. asty asminn itl matup The two akaaras subsequent to praks- rane ate almost entirely lost, perhaps, one may conjecturally supply Jyotsnâdayas (see P, 5, 2, 114) 13 The words adabhyastat are a quotation from P, 7, 1. 4. In- stead of adadesas the MS has adadesas, which reading ia obniously due to a senhal mistake Inatead of ru. perhaps can be read ta also. 15 The words videh fatur vasuh are a citation from P, 7, 1, 36

[ 54}

Udbhata's Vivaraņa, etc

3 ]x kah samprat sadhavo pi ye na pra- yoktavyas tā(n) darsayıtum aha[ 4 Ju vanam (vanaya)locana it: [VI, 60] ] naıkatraukārabhuyastvam [VI, 61]] ity upalaksanapaf 5 ] antāt saptamy ekavacanam uktvā ( [ > "jhayo 5 ho nyatarasyam" ıtı sāvarnyākrtvāha va[ 6 ]paddhatısabdah pra(yu)jyata upalaksanaparam cai- tat tad x x yo pi prayogof 7 ]tam na sakyate uttamamadhyamadhamais sādhusabdaır vıbhajya asesan kathayıtum | x[ 10 8 ]darenety aha | vidyanam ityadı [VI, 63| | etad uktam bhavaii | bahnpa[

P, 8, 4, 62 5 The words jhayo ho nyatarasyam are a quotation from

Į 55 ]

FRAGMENTA INCERTAE SEDIS

Fr 54

Jayam atra vākyārthah ( ) ) mtyasya saktau[ Jvyal

]vırodhād ıtı | sampratı kāvyanyāyaĮ 5 CT

Jya am[

b

10 ]x X isyata atac ca vastumarthavica[ ] a ca | ata evedam aha asaram lakşatenya[ J avāśrax X idha[

Fr 55 15

Jpanalamblutasya[

4 Fr incertae sed u grafically it belongs to 6th Ch it Leems 11 The words vastomarthavicat are clear but apparently the tezt as here vit ated 19 The words auaram lakente (to be read lakentot) nya" look I ke a citat on but I have not been able to find them out ne ther in Bhamnha nor elsewbere [59]

RANIERO GNOLI"

Fr. 56

]prāg [

b

lya samasa iva pal

Fr 57

a Jvyatirikte sastraśray [

] y punah pra i u[

Fr. 58

] o deva[

Jevam x [

[ 60 ]

Udbhafa's Vivarana, etc

Fr 59

lyoh pra[

b

]darśı[

Fr 60

]tıya[

]nã 1[ ]prakāx [

Fr 61

Įtah gatas sahakārāį

[61]

RANIERO GNOLI

Fr. 62

]t sarvatrax[ Jpadasya[

Fr. 63

a

Jgada[ ]śveh.o[

b

Jpanā[ Jvate ! [

Fr. 64 ]b phullah kusu[ ]hā[

Fr. 65

Jma xy.[ ]vamādi[

[62]

Udbhata's Vivarana, etc ..

Fr. 66

]kriyāĮ

Fr. 67 5

]t a.i[ Jvan. v.[ 10

b

]na prake.i[

Fr. 68 15

] ālankāra.r.[ .

Fr. 69

20

]x ·yadrecha[ 21 This i probably part of the comm. to VI, 21, 22. [63 ]

RANIERO GNOLİ

]samgrahah syat( | )tadīdam āha f ]X vāh sūrasenāday-E ]m iyatta paricchedo[ ]vicāram rakșamāf 5

]h pra[ ]x visaya if[ 10 Jādityapakşıvisese[ Įsyata iti tadvisayo yal ] m lakanam asti tatra laļ

Fr. 70

15

1 Jnuyāyi tatra devadattasabdā iĮ 2 ] eyākarana[

b

20 8 lya bibhyad divasa x kastka i[

Fr. 71

lpāsānavišesa[

1 tadidam obvioysly a scmbal mistake for tad idam.

{ 64 ]

Udbhața's Vivarana, etc

]xh pratidıx [ ]ta[

Fr 72

a

Jkırātapas [ 2 Jdhamanda x h bha[ 3 ]sa[

6 7 Jrısunym ahsa[ 8 ]tıkānkșıtam da]

Fr 73

] ımā uryasthe[ ]x ghavisesana!

Tr 74

lxy y l ] ajanyany asta x [ [65 ]

Udbhata's Vivarana, etc.

Fr. 77

Jnaxku[

b

ly [ Jdaxu athel 10 .

Fr. 78

Jpakşanaf

Fr. 79 15

Ipanpātu[ Jpamati dral

12 The other face illegible 18 The other face lost [67 ]

RANIERO GNOLI

Fr. 80

. Jrūpan s.[ Jeva vedasu[

b

]m eval ] iparidi[ ]h <pra)siddhaya[

Fr. 82

. l.y [ ]x sahi[ Jne[

. ]vassiva]

[ 68 ]

Udbhaļa's Vivaraņa, etc

Fr. 83

1 ]siddham | atha pu[ 2 Jxdixf

b 5

Jtatra smrti

Fr. 84

Isadh(u) saml 10

Fr 85

Jv n [ Jvacat 15

20

8 The other face illegible

RANIERO GNOLI

Fr. 86

. Jndhesm [

b

Jagham[

Fr. 87

] apam utsrtya[

b

Jnıvartayal ]tah || x i[

Fr. 88

Jy udāharaļ

[70 ]

Udbhata's Vivarana, etc

] ıx yā[

Fr 89

1 ilata p { 1 el

b

Jtaragahot

Fr 90

One face only

1 ]ca tathax x[

Fr 92

( 71 )

RANIERO GYOLI

] ānyatal

b

5 ]sesaf

Fr 93

a

Ipanopal 10 8 Jsã x xhrf

1 ]x tādvasa x [

Fr 94

15 I ]ccala[ 2 ]x tum i[ 3 Jddham / tasmax xf

5 ]bahupra ura[ 20 6 Iprav rttibhaja[ 7

8 ]sambandhah Xt [ 9 Jmam ity aprax!

21 Before and after the akpara a there ate eppatently two veztical strokes

[72]

Udbhata's Vivarana, etc

5 ]xnyapāgas u[ 6 ]yogādıśı[ 7 ]ryasa[

Fr. 95

123 Jnga x krtya[ ] yāyate[

b

6 Jvarnabheda[ 7 ]xnranayo avasyam caita[ 8 ] y nya y xt dartha slesax s [

Fr 96

Jvire[ Jkrl Jpavistāxd dohal /kalana prak tu ti i[ Jyam | kimarthal Jkal

b

Jmam[

[73 ]

RANIERO GNOLI

]dham | sādha[ Jdāś kākalīsamjñāį ]t | yo pi manyate kal ]gax X syapi[ Jsyast [

Fr. 97

One face only I ] a madhyānta udātta X ca yix cārutvam | a[ 2 ]x xpra x ddhāya m [ Imbhix xv. lalitad X tejamsi{ 3 Jasam x x japapradānavisayatāyām tu su[ 4 Ipanayanti ghanah punah krto bhyah a[ 5 ljaladha ati[ 6 Jntamanvādisa[ 7 [ixsyex[

Fr. 98

Jv s yysx y l ]rthalaksanam asrayadvayam alamkare py [ ] ox prasiddho vedah yathā purușanām of

] y x xbdantara y nul Jsā cārutvakrtā pratyāsattyām śr { Inām el

[74 ]

Udbhața's Vivaraņa, etc

Fr 99

Jlav tāvala[ Jkşanaśaśva[

5

b

]padhārane | s [

Fr 100 10

Jtmankaśabdaś [

15

Jdarsal

Fr 101 20

)pātrārthapraf

3 The akyar dhå maght be read as va niso [ 75 )

RANIERO GNOLI

Jtad ahax [

Fr I02

]dauham x [

Jviśvel

Fr 103

a ]yt y rthel ]sa samkepel

b

Hlohah kal [ Jantāc ca tādr[

[76]

Udbhata's Vivarana, etc.

Fr. 104

Jvatāra i[

6

In nipex[

Fr. 105

Jtah jya 1[

Jdsdal

Tr. 106

One face only

Jul Jna te pu Jaity aparel

(77]

RANIERO GNOLI

Fr 107

]x panavar [ JlamX o a[

basahena[

Fr. 108

a

Irmanı satsa[

]dvitiyam dars [

Fr. 109

Jdrumadarsa[

[78]

Udbhața's Vivaraņa, ctc.

]ştavişay- [

Fr. 1I0

]x ganam nīya[

b

Jxmānata i[

(79]

APPENDIX

FRAGMENTS OF THE RAGHUVAMSA BY M. TATICCHI

In addition to the fragments of the Commentary to Bhamaha described and cdited by Prof Gnol, there are other fragments belonging to the Raghuvamsa of Kaldasa The latter are written as the former on birch-bark and in sarada characters, both sets apparently belong to the same epoch, namely the 10th-12th century The only diffe- rence lies mn the writing, which mn the Raghuramsa frag- ments is more accurate and regular, the MS is more cor- rect than is the case for the commentary to Bhamaha The size of the sheets was origmnally of about cm 30 The sheets are burned on the right side The aksaras contamed in each line numbered about 50 The lines for each sheet number 8 The interest of this MS, which is apparently, as far as we know, the most ancient of the Raghuvamsa, hes mn the numerous variants that it presents, in comparison with the text commented upon by Mallmatha The edi tion of these fragments might therefore bring a contribu- tion of no slght importance to the history of the text of the Raghuramsa A critical edition of this work, it 1s well-known, is still a desideratum, and the best edition of it so far is always the one by Nandargikar1 While awaiting for other studies and editions of this poem - promised by V Raghavan and A Scharpe - it

1. The Raghuvamsa of Kalsdasa touh the Commentary by Mallsnatha edited bɣ R Nandargikar Bombay 1896 The edition we have had to band and to which refer the numbers of the cited pages is the third one of 1897

[ 83 ]

M TATICCII has therefore seemed to us that the best way was that of publishing the simple transcription of these fragments; for the convenience of the reader, only mdicating the points in which the text differs from the text of Mall- natha, and the variants noticed or not noticed in the edi- tion of Nandargikar, without any attempt whatsoever at a critical reconstruction. Any critical exammation, on the grounds of present- day studies and awaiting the above-mentioned works, would have been not only ahead of time, but even outside the sphere of our own competence Statements such as Harı Chand's, according to whom " les variantes qui se rencontrent sont depourvues d'importance et s'expliquent facilement par les mfidelites inevitables d'une transmission orale Il ne saurait en aucune manière être question de recensions ""), are most likely to be reappraised. It may well be that a recension specifically Kashmir of the Ro- ghur amsa may have existed, as is the case for other works by Kaldāsa

1. Kaldasa et lart poftique de l'Inde, Paris 1917, p 237.

FACSIMILES

(reduced to approximately twe thrds the sze of the onginal)

Fr la

Fr 1b

Fr 7 a Fr 7b

Fr 11 a Fr I1b

Fr. 15 g,

Fr. 15 b

Fr Ia

[M, III, 43-53]

I lganasprsa raghus svarena dhirena ni(va)rta- (yan)n ıva || makhāmsabhājāmį 2 ]vartase || trilokanathena sata makhadvisas tvaya 5 (mya)mya nanu divyaca(kşusa)[ 3 Jm pratimoktum arhası | pathas sucer desayıtara X XX malimasām ādadate na[ 4 Jmasa ratham (savisma)yah pracakrame ca pra- tıvaktum uttaram || yad attha rāja[ 10 5 Jthaikah (purn)sottamas smr x(mah)esvaras try- ambaka eva nāparah( | )tathā vi[ 6 ]haritah( | ) alam prayatnena tavatra ma mdhāh padam padavyām sagara(sya sam)tateh[ 7 Jjitya raghu(m) krti bhavan || sa evam ultva 15 maghavantam unmukhah 8 mayena patrına hrdı ksıto gotrablud apy amarsanab[

[M, III, 54-64] 20

1 Jh< | ) papāv anāsvādıtapūrvam āsugah kutūha- leneva manusyasonıtam i/ l 2 ]khāna sāyakam || jahāra cānyena mayuralaksa- nam sarena śakrasya[

5 aata sada M see N p 82 17 kşito kşato M this reading is not noticed by N 23 mayuralakaanam mayurapattrink M ths teading 1s not noticed by N [87 ]

M. TATICCHI

3 jvısabhīmadarsanaih( | ) babhūva yuddham tu- mulam jayaisinor adhomukhair[ 4 ]vas svatas cyutam vahnim ivadbhir ambudaḥ l tataḥ prakosțhād dharicandanā(nk)i[ 5 5 ]vivrddhamatsarab pranāsanaya prabalasya vid- visah( | )mahidhra[ 6 Jmesamatrad avadhuya ca vyatham sahotthitas sainikaharsamsvanarh[ 7 Jsangam adrisv api saravattaya na me tvada- 10 nyena vil 8 ]dilipasūnuh pratisamharann isum priyamvadah pratyavadat sureśva[

Fr. 2 a [M, III, 65 sqq .; IV, 1]

15 I lsradīksātanutadya me guruh krator asese x xlena yujyatãm[ 2 ldhiyatam < || ) tatheti Lax x x x x xn raX X x x tam matalısāra[ 3 Įs śāsanahārmā x x ( | ) x rāmrsan ha(rsaca)-

20 lena panina tadiyam (ang)amx x x x xnkitam [ [ 4 Jatha visayavyavrttatma yatha[ 5 ]tam if Jghuvamse hax x x trtiya(h)[ 6 Jsusru(vu)sam patim | [

4 prakoşthad prakosthe M, see N. p 87 7 ca vyatham tadvyatham M, sec N, p. 88 11 dilipasunuh narendrasunuh M, see N. p 89 15 The reading of the three akearas subsequent to dike a w hy no means clear The reading of M, "prayatah sa madguruh. is not sup- ported by the remains of the akgaras, which rather suggest tanu or some thing hke so 15 me guruh madguruh M 21 After the word atha the syllable sa has been omitted in the MS out of a seribal mstake 23 The words suśru(vu)eam patim belong to a stanza not represented in M, see however N, p 92

[ 88 ]

Fragments of the Raghuvamea

[M, IV, 10-21] 2 ]śasta[ K(e)vabhava x ksa x X x xvad uttarah I [ ]sa(m) x x xgunah || nave tasmi(n ma)hipāle 5 şarvam navam ıvā[ 5 Kmeghar)s savitns tasya cobhayoh | vardhiX vo dıx bha x pratāpa[ 6 ]punda(rīkā)tapatras tam vilasatkasacāmarah[ Jit samarasā dvax ( I| )x x şremisu tārāsu kumu- 10 dvatsu[ 8 Jdghātam sāh x x x x x sah | prasasādoda[

Fr 34 [M, IV, 50-66]

1 Jdadus tasmai yasas svam iva samcitam j sa 15 mırvisya yathakamam tatesv almnacandanau | stanāv lva disas tal ]dınyas srastamsukam alamghayat || tasyānīkair vışarpadbhır aparāntajayodyataıh | ramegūtsārito pyā[ 3 Jrabānānām ayatnapatavāsatām || abhyabhūyata 20 vāhānām caratam (gatra)smntaih | varmabhi pavaļ 4 Kkar)ınam petuh pumnāgebhyas ılımukbāh || av akāsam kılodanvān rāmayābhyarthitu dadau( | ) aparāl

3 The word sasta is not represented in M F The sotazas 13-14 M are not represented in the MS as to the stanza I5 M, different from the version accepted by M see N p 97 D vilasato vikasuto M cp N 98 lrameque ramastro° M ep N p 110

[89 ]

M. TATICCRI

5 ]m eva tatroccair jayastambham cakara sab i pārasīkams tato jetum pratasthe sthalavartmanā | i[ 6 ·]lātapam ivābjānām akalajaladodayab || samgrā- mas tumulas tasya pārasīkaśva[ 5 7 ]smasrulair mahim | tastara saraghavyaptaih sa kşaudrapa(ța)lair iva || apanītasirastrānāh[ 8 Jjayaśriyam | āstīrnājinaratnāsu drākşāvalaya- bhūmișu || tatah[

b

10 [M, IV, 67-80]

1 Jrassarān | maharnavam ivaurvāgnih pravivesot- tarāpatham || vinītādhvaśramā[ 2 Jvikramam | kapolapātanādesi babhūva raghuce- şțitam || kāmbojās samare vīryam tasya 15 3 narāsayaļ( | ) vivisus tam visām nātham udanvantam ivapagāh( || ) tato gaurīgurum sailam ā[ 4 ]se py asambhramam || guhagatanam simhanam parivrtyāvalokitam( | ) bhūrjeșu marmarībhū[ 5 ]chāyāsv adhyāsya sainikāh | drsado vāsitotsan- 20 gā nisannamrganabhibhiņ || saralāsakta[

4 parasikasva" pascattyair asva. M. this reading has not been quoted by N. 7 jayasrıyam ojayaśramam M, this resding has not been quoted by N. 12 The stanza preceding vinitadhvasramas (M, 67) 1 not re- presented in M, see N, p. 115. 13 °patana% opatala" M, see N. p. 115. 14 viryam sodhum M Appareatly the lectio of the MS was vi- ryam tasya sodhum for sodhum tasya viryam, this va- riant has not been quoted by N 16 The reading vivisus tam visamgatham udanvan- tam ivapagah has been noticed hy M also, see N, p 116 17 guhagatanam guhasayanam M, see N, p. 117.

[90]

Fragments of the Raghuvamsa 6 ] āseşu dānārdraır gandabhıttıvighațtanaih || ga- javarsma kırātebhy ah sasamsur devadaravah || vimarda[ 7 Ksa)rair utsay asamketan sa krtva karadan krti ļ jayodāhıranam bihvor gāpayāmasa kimnaraih || pa[ 8 ] | tatraksobhyam ya(so)rasım mvesyavaruroha sah( | )paulisty atultasyadrer adanana iva friyam I [ 5

Fr 4 a [M, IV, 81-88, V, 1-4]

]drumaih || na prasehe sa ruddharkam anabhra- mayadurdmam || rathavartmarajo py asya kuta eva 10 patakınım <# > 2 ]h || kamarupesvaras tasya hemapıthadhidevatām | ratnapuspopaharena cchayām ana(rea) pādayoh ||

3 ]sa vışvajıtam ajahre kratum sarvasvadakşmnam 15 ) ādanam (h)1 visargaya satām (vā)rimu(cā)m iva || sa[ 4 Kv(ra)hotsukāvarodhan rājanyan svapuramivrt- taye numene || te rekhadhvajakuk(i)satapatracihnam[ 5 ]gauram ! l digvyayo nāma caturthah sargah # || tam adhvare viśval 20

1 The reading was perhaps tannivanegu etc the words preced ing the stanza gajavarşma kiratebbyah (M 73) are not repre Bented in M they have not been noticed by N As to the reading v1 marda ete in the sabeequent stanza ep N p 119 3 karadan krtı vjratotsavân M the readng has not been noticed by N 4 kinnaraih k nnaran M this reading ie not not ced by N 6 adanana ths is a scr bal m stake for adadhana or adadana driyam hriyam M wee N p 120 10 anabhramayadurdınam adhårāvarşadurdinam M ths readmg has not been noticed by N lo åjahre kratum arebhe yajnam M the readng kra tum for yajr am is not not ced by N

[9]]

M. TATICCHI

6 Js(1)4y ah || sa mrnmaye vitahranmayatvat patre nıdhāyā[ 7 ]vad vidhjuas tapodhanam manadhanagray ayi j krtajalh krtyavid i[ 8 Isam aptam caitan(y )am (u)g(r)ad i(va d)i(ks)i- (tena)[

[M, V, 6-16]

1 Kp)r(a)mukha(ih) prayatnah X xdhitanam suta 10 xxxx(1 ) ka[ 2 Jtadankasayyacyutanabhdlana kaccı(n)x x xx x (gh)a prasutih < [I ) 3 ]m vas tirthajalam kaccit || nivarapikadx x x x x x(m)rsyate janapal 15 4 Inā tvam samyag vmiyanumato grhīya | kālo x x x x xtum dvitiyam sarvopaka[ 5 ja sasıtur atmana va prapto sı sam(bh)āvayı- tum vax x< || ) ity arghyapatranumitavyayas(ya)[ 6 ]kutsam | sarvatra no vartam avehi rajan nathe 20 kutas tvayy aśubham prajanām( | ) sūrye tapaty āva- ran [ 7 yātısése | vyatītakālam tv aham abhyupetas tvām arthibhavad iti me visadah || sarira[

4 In the MS the two last padas of the stanza V 3 (M) have been inverted 5 catanyam ugrad iva dikeitena lokena car

11 onabhilana (corrected m a secunda manus mnto on a bhilina) apparently a scrbal mistake for onabhinala 19 The two akgaras preceding to sarvatra, ete (M V. 13) are clearly kutsam and not sipyah as we read in M in the MS there was here s different reading or perhaps some other stanza e2vyatıtakâlam vyatıtakalas M

[92 ]

Fragments of the Raghuvamda 8 Jtih stambena nivara ivavasistah || sthane bhavan ekanaradhıpas sann akımcanatvam makhajam[

Fr 54 [M, VI, 28-39] 1 ]v(1)n(ai)v a harah || msargabhinnaspadam ekasam- 5 stham asmin dvayam sris cal 2 Kca)ksu y Xtı janyam avadat kumāri nasau na kamyo (na) ca veda samyag drastum na sa bh(z)nnaru! 3 ](vı)seşadrsyam mndum navotthanam ivenduma- tyah || avantinatho yam u(dagra)bahur visalava[ 10 4 Jr vajıbbır utthitam | kurvant samanta X X(ma)- ninam prabha[ 5 ]sparsasukham bhunkte || (an)e(na yuna) saha parthivena rambhoru[ 6 J(r)asu || tasminn apix x x X X X X X tapasam- 15 sosıtasatrupank(e) | X x x x nottamasaukumāryā[ 7 Kvi)dhatur jagada bhuyas su(datım sunanda || ) sangrāmam[ 8 Jmakālam eva pradurbhavams cāpadharah pura stāt < | > anta[ 20

[M, VI, 40-51]

1 Jkarāgrhe mrjitavasave x x x x x X sitama pra- sār

9 ondumatyah ondumatyal M ths read ng is not noticed by N 13 As to the reading sparsasukhans bhudkte see N pp 168-9 15 tasminn apı taaminn abhi" M this reading is not noticed by N [ 93 ]

M TATICCHI

2 ]h (pra)mrştam || ayodhane krşnagatim saXya- m axpya X X triyakalaratrim | [ 3 livapramtambaka(nc)im < | > prasadajalair jala- Xn xxxvām yadı prekşitum astı ka[

5 4 Jlo nahnyah || sī surasenadhipa x x x nam udd X x äntaragı 5 ]th v (esa) X v un Xm asritya paraspare- na | naisargiko py u x Xje virodhas siddhasramam sānta[

10 6 Jharmyāgrasamrūdhatrnānkureșu (tejo) visahy- am rpumandıreșu | yasyavarodhastanacandan inam praksalanā[ 7 ]bhatı || tratena tarksyat kıla kalyena manim vısrştam yamu(nau)kasa yah( | ) vaksahsthalavyā[ 15 8 ]lottarapuspasayye |i vrndāvane caitrarathād anune nırvısyatam (su)ndarı yauvana(srī)h |) adhyasya cämbhah[

Fr 6 a

[M, VI, 52-63]

20 1 ]tamanojnanabbs sa vyatyagad anya(vadhur bha)- vitri | mahīdharam[ 2 ]sedusım sādıtasatrupaksam bālām abālendu- (mu)hīm babhage || [ 3 ]h ndrah || y x xrekhe subhujo bhujabhyam 25 bibhartı yas capabhrtam pu[

9 In the MS the two last padas of the stanza VI 46 (M) have been laverted 13 tratena trastena M see N p 175

[ 94 ]

Fragments of the Raghuvamsa 4 ]tatapūgamalī | mandradhvanıtyajıtayamaturyah prabodhayaty arnava eva suptam[ 5 ]va marudbhih | pralo X X x x x x x nnatabhrūr vidarbharajavaraja tayai[ 6 Jācarapūt X x vamsadīpam guddhantaraksya 5 jagade kumarī || pandyo[ 7 Jvadrirajah || vindhyasya samstambhayita ma- hadrer mssesapītojjhitasmdhu x x | )x tyaśva[ 8 ]ndralokāvajayaya srstah( | > purā janasthāna- vimardasamki samdhaya lankadhipatıh x x x | ane[ 10

[M, VI, 63-74]

1 lpatnī bhava daksmasyāh || tāmbūlavallīparınad- dhapūgāsv elālatāli[ 2 Jrirayastıh( | ) anyonyasobhaparnv r(ddhaye v)am 15 yogas tadıttoyadayor ıvāstu || [ 3 ]X || (sa)mcarini dix x x x X au yam yam vyatiyaya patımvara sa( | >[ 4 ]x yurabandhocchvasitair nunoda || tam[ 5 Jv(esita)cittavrttim induprabham indumatim 20 aveksya | [ 6 ]k(ut)sthasabdam yata (u)nnateccha(h) slaghyam dadhaty uttarakosalendrah || mahendram asthaya ma- hoksa[

1 As to the reading tayapugamal ete ser N, p 178 3 As to the readig pialo ote ie N po 178-Q 5 This is M VI 45 with the vanant puta for suddha Thi reading has not been noticed by N D sretab drptab M see N, p IB1 [95]

M TATICCHI

7 Jlanavislatham yas samghattayann angadam an- gadena | upeyuşas svam api[ 8 ]d (e)k(o)nasatakratutve sakrabhyasuyavmvrt- taye yah l|

5 Fr 7 a [M, VI, 76-86]

1 2 ]lam |] asau kum[ 3 ja navena gunais ca tais tair vinayapradhanaih ({) tvamt 4 ]vamāsa drstya padmasraja samvaranarthayera[ 5 ]yam parıhasapurvam sakhyam sa[ 6 Jyāmāsa sırodharāyām kanthe[ 7 ]varenyah || sasınam upagateyam kaumudı me- 15 ghamuktam[ 8 ]vruh || pramuditavarapaksam ekatas tatksıX X t mandala[

6

[M, VII, 2-10]

30 1 ]mano năma sașthas sargah (I [ 2 ]san prthıvıksıto pı jagmur vibhātagrahamanda- bhāsah[

11 Here the text 19 d fferent from that accepted by M ie dretya prasadamalaya kumaram pratyagrahıt samvara nasrajeva (VI 80b) This reading is not noticed by N 13 °yamasa sirodharayam sam M this reading is not noticed by N oyamasa yathaprade

[96]

Fragments of the Raghuvamsa

3 kșitipā[ Jkutstham uddisya camatsaro (p)i XXX tena

4 ]ritoşnam || tatas ta(da)lokanasatva[ 5 ]dv(e)stanavāntamālyab | banddhum na sambha- vita[ 5 6 ]ktakānkām padavīm tatāna || vilocanam dakși[ 7 Kprasthanabh)i(nna)m (na babandha n)i(v)im / nabhipravișțā[ 8 ]éesā || [

Fr. 8 a 10

[M, X, 81-86; XI, 1]

4 ]babhūvatu[ 5 Ino jabrur nidāghānte śyā[ 6 ]māsus te gurum guruvatsalāh( | )[ 7 ]gair upayaih | harr iva yugadirghair dorbhir 15 amsais tadiyaih pati[ 8 Įdasamaņ sargaņ || kaušikena sa kila ksitīșvaro rāmaf

[M, XI, 1-5] 20

1 Jmiksyate || krcchralabdham api labdhavarna- hhak tam di[ 2 ]māya puramārgasatkriy(ām)[ 3 Jtor namrayor upari baspabi[ 4 Jpitur nayana[ 25

8 tndalokanasatva tadālokanatatpa. M, cp. N. p- 192.

[97] 10

M. TATICCHI

Fr. 9 a [M, XI, 17 sqq.] 4 Jntramekhalām ( | ) tām vi[ 5 ]viştapatrayaparajayasthiram ravanasriį 5 6 ]ratam agamad antakasya tat || ramamanmatha- sarena tādit [ 7 ] || nairrtaghnam atha mantravan muneh prapad astram avadānatoșitāt | E 8 stapovanam prapya dāsarathir āttakārmukah < | ) 10 vigrahenal

[M, XI, 22-28] 1 Junmanah prathamajanmacestitany asmarann api babhūva rāghavaḥ || āsasāda[ 15 2 Im x povanam || tattra di(s)i(tam rsim) rara- ksatur vighnato[ ljivaprthubhiḥ pradūsitām < { ) sambhramo bhava[ 4 ]d ambare grdhrapaksapavane[ 20 5 ]stram ugrajaval

Fr. 10 a [M, XI, 35-42] 3 ](artha)Lamasa[ 4 ]pi vañcanām manah || yū[

3 The verse ending with the words oratam agamad anta. kasya tat is, according to M, VI, IB Our MS fust reads the 19th verse and then the 18th See N. p 332 10 After the stanza begmmng with the words nairriaghnam atha (M, XI, 21), we have in our MS the stanza M, XI, 13, see N, p 333

[98 ]

Fragments of the Raghuvamsa

5 lpus sisoh parthivah prathita[ 6 ]karma duskaram | tatra naha(m anu)mantum utsahe moghavrtti ka[ 7 Jnatvaco bhujan svan vidhuya dhig iti prata- sthire | pra(tyuvā)ca tam rsir nisamya[ 5 8 Jrav iva || attham (ap)tavacanat sa paurusam kākapa(ksa)kadhare pi x ghave( | )éra[

[M. XI, 43-51]

I ]vagān kārmukabhiharanaya maithilah || taijaX 10 sya dhaX X pravrtta X toyada[ 2 ]mrganusarinam yena banam asr(jad vr(sa- dhvajaḥ || ātatajyam aka[ 3 K(smara)b ( I ) bhajyamānam atımātrakarsaņāt tatsvanena gaganaspr(sa)[ 15 4 ]rāghavāya ta[ 5 ]parigra X X x tām ku[ 6 ]m || ta[

Fr. II a [M, XI, 54-64] 20

Igatau svamitrayā tau kusa[ 12 ]sannibhah( || >evam atmara[

6 ittham evam M, see N. p 340 15 tatsvaneno garanaspriå tena vajraperupas- vanam M, see N. p 342 21 °gatau ivamitrayă tau orajau varaujasan tau M, this reading is not noticed by N 92 evam itmao evem itta. M, this rending (owing to a vcri- bal mistake") is not noticed by N [99]

M TATICCHI

3 ]vartmam dhvajatarupramathmnal 4 Jmitasya bhogmno bhogave[ 5 ]r avalobanaksamah < i| ) bhaskaras ca di 6 lpavanadı vaıkrtam kaıprasāntyam[

5 7 ]vāhmīmukhe | yah pra[ 8

[M, XI, 65-75]

1 1j(i)yata (ghrna) tato mahi( | )

10 2 ]dhabhavena manyuna rajavams[ 3 ]t(e) ca darune ) hrdyam asya bhayadāyıl 4 ]h samdadhe drsam udagratārakam( Il ) 5 Im apakaravam me tan mha[ 6 ]mıtapurvam aksınoh( | ) yan m[

15 7 ltı me sa sampratı vyastavrttı[ 8 ]m udyatah || ksatrıyantaka[

Fr 12 a

[M, XI, 76-85]

1 ] r(aı)śvara(m) dhanur abhājıl

20 2 ]dhanuh ( { > uişthatu praX Xm evam apy aham!

1 vartmanı vartmasu M see N, p 346 4 kşıpradānıyom prekşya sāntim M sce N. p 349 14 akeinoh aksanoh M see N. p 352 14 yan tan M this reading is not noticed by N 20 As to the reading of the verst preceding tisthatu ete (M, XI, 77) see N pp 353-4

[ 100 ]

Fragments of the Raghuvamda

3 ]dakathmāngulır vrtha badhyatam abhayayaca- nifjalıh || evam u[ 4 ]ta samartham uttaram || pūrvajanmidhanusa samagatah so timatralaghu[ 5 Jmıtaıkakotma karmukam ca balmadhiropitam 5 | prīpa varņavıkrtım ca bhā[ 6 Jparıhīnatejasau | pasyatı sma janatā dınatyaye pārvanau sasıdıvāka[ 7 ]m asugam vyajahara harasunusammbhah || na prahartum alam asmi m[ 10 8 ] || pratyuvāca tam rir na tattvatas tvām na vex puxsam pura[

[M, XI, 86-93]

I Jsudhām sasā(ga)rām | ahi[ 15 2 Jme | pīdayışyatı na mām khi(lī)krtā svarga[ 3 jto pi so bhavat svargamargaparigho duratya- yah || ra[ 4 Jtir eva sobhate | rajasatvam avadhuya ma- trıkam pıtrya[ 20 5 Jmy aham avighnam astu te devakaryam upapa- dayısyatah[ 6 Jr eva jatam ( | ) tasyabhavat ksa(nasu)cah[ 7 Jīs x kalpah < | > purama[

1 "dakathına" (ie )yavımardakathinac) jyānı ghatakathina M see N p 354 5 As to the read ng of this stanza see N p 355 The reading *mitai° 1s clear 19 sobbate kırtaye M see N p 358 19 matrikam for matrkam is obviously due to a ucrbal mistake

[101]

M. TATICCHI

Fr. 13 a [M, XII, I-14]

1 Jreir ivoșasi || tam karņamūla(m a)gatya rame[ 2 h(l)adayamcakre kuk(y)evod(y)ana x dapan < ) 5 3 Jrtrā tatsamsrutau varau | udvavā[ 4 Jām rudan ramah prān mahīm pratyapadya[ 5 Jjanah || sa sītāla[ 6 ]gamātrena suddhilā(bham ama)nyata i { 7 Kkyrt(a)y(o) matribandhunivasinam | [ 10 8 Jśriy(o) py asīt parānmukhah || Xsainyas cānvagā[

[M, XII, 15-24]

1 Jo( | ) laksmyā nimantraya(mcakre) tam anuc- chistasampada |I { 15 ]nab pituḥ < ] ) yayāce pādu(k)e pascāt kartum rā(j>ya[ 3 ļrajyatrsnaparā(nmu)khab < [ ) mātub pāpasya bharata(b)[ 4 Jbhavastambhitacchāyam a(śrita)s sa vanaspa- 20 tìm | kadal 5 Jeihnesu paurobhagyam iv(a)x x x xn asthad isi[ 6 ]manam punah < | ) as(a)x x <xxx x x xkū- tastha[ 25 7 ]ahixt [ Jm | x -y[

9 matri" is obviously due to some seribal mistake for mitro.

[ 102]

Fragments of the Raghuvamsa

Fr 14 @ [M, XII, 86-99]

1 ]pus suradvisām || anyonyadar sanaprāptavi[ 2 Kh)y ayath ipurvam matrvamsa iva sthitah | jetaram[ 3 5 ]tasamgamasamsm | rucakhanadhikakrodhas sa- (ra)m (sa)vyetare (bhuje) || ra[ 4 Jor vak(y)am as(tr)a(m astr)e(na) x ghnatoh || anyonyajayasam[ 5 Jkrtapratıkrtaprītais tayor muktam surasuraih 10 ( 1> pal 6 Jlim aksipat || ragha(vo ra)tham apraptam tam aśăm ca sural 7 ]stram priyasokasalyam x X X sadham tad vyomnı dasadha[ I5 8 Ipanktim ajatavranave[

[M, XII, 101-104, XIII, 1-4]

Ītasya (ś)ırāmsı patitāny apı < / ) mano nātivi- saśvāsa[ 20 2 lya | avinyamitaratne murdhm paulastyasatroh surabhi[ 3 ]nāmāmkaravanasaracıtaketuyastım ū(r)dhvam[ 4 Jya şrıyam varınah ( | > ravisutasahitena tena[

• ayathapurvam ayathapurvo M see N, p 389

reading is not noticed by N 23 °garacitaº farankıta. M this reading also i not noti ced by N

[ 103 ]

M TATICCHI

5 Jathatmanas sabdagunam (guna)joah padamį 6 lyad vibhaktam matsetuna <phen)lam al 7 ]mite turamge | tad artham urvi(m a)vadara[ 8 Jvasum | ahindhanam vahnım asau bibharti pra-

5 hlada(nam) jyotir ajany ane(na >

Fr 15 a [M, XIII, 46-55]

]sthita suputresv iva padapesu || x x x x xnda[ 12 Jesa prasannastimitapravaha sard vidurantara- 10 bhav atanvi | ma[ 3 Jlas tava pravalopacayena yasya | karņārpitenā[ 4 ]ksam | vanam tapahsadhanam etad attrer avi- skrtodagratapahpral 5 ]trisrotasam tryambalamaulmalam || dhirasanair 15 dhyanaju[ 6 ]sakhmno pi || tvayā purastād upayacito yas so yam vatah syamal 7 Jklamseva rātryā kundasragmdıvaramālayeva krttır hare!

20 8 Jdıseva į gavāksakālāgurudhūmarājyā harmya sta[

11 tava pravalopacayena yasya|karoarpitenaº, etc pravalam adaya sugandhi yasya etc M see N. p 415 13 °dagratapaho odagratara" M ace N p 416 14 dhirasanair virasanair M this reading is not noticed by N 18 The stanzas XIII, 54-57 M are not represented in our MS, and, sab sequently to stanza 53 there are the four stanzas beginning with the words tamisraya suklaniseva ete, noticed by N pp 418-9 IS ratrya bhinna N p 418

[104 ]

Fragments of the Raghuvamda

[M, XIII, 56-66]

1 Įgocarānām sreniva kīdambavıhangapanktyā l mtāntāsu X spha x kā X yogā[ 2 ] || samudrapatnyor jalasannipate putatmanam atra kı[ 5

3 Idam tad yasmın maya maulımanım vhaya | jatāsu baddhāsv aru[ 4 Jbrahmam sarah kāraņam āptavāco buddher ıvavyaktam udaharanti I! [ 10 5 ]rīkrtan ( yām saıkatotsangasukhocitānam prå- jyaıh pa[ 6 Jrājnā sarayur viyuktā | dūre pi santam śisıra- mlar(r mam) tarangahastau u[ 7 ]tyudgato mam bharatas sasamnyah || atra (6r- 15 yam) paritasangarāya[ 8 ]padatth pascad avastha[

Fr 16

One face only [M, XVI, 14-29] 20

123 ]ştinıvasabhamgan mrdangasabdapa[ ]rgam avapnuvantr | upantavanira[ ]manuşım tam yatha gurus te parama[

18 dure pa santam dure vasantam M ef N.p 421 lo atra addha M this reading is not poticed by N 21 The stanzas XVI 15 20 are not represented in the MS 22 avpnuvantı anapnavent: M this readmng is no noticed by N [105 ]

M TATICCHI

4 ]babhuva || tad ad(bhu)tam samsadı rātrivrttam prā[ 5 lyatrānukūle ham sāvarodhah | anudruto[ 6 ]sena ra(tho)daragrha prayane tasyabhal 5 7 ]n ıva nīyamānah || tasya pray[ 8 lya pascat puro nive[

Fr 17 a [M, XVI, 21-29[

1 Jmādalā x x praf 10 2 Igrhesu drstvā[ 3 Jmātma mūrtim tatheti tasyāh pra[ 4 ]tar dvijebhyo nrpatis sasamsa[ 5 Jvayur ivabhravrndath[ 6 ]marājadhanī |I tenāį 15 7 Jthininam pidam a[ 8 ljanti || sāyatra[

[M, XVI, 30-39]

1 ]turamgamān(ām | re)n(uh) pra[ 20 2 Jhavirava baddhax x x x guhamukhant | [ 3 basetubandhāt pratipa[ 4 IpurvaxxxxxxxxxxxXXX Xv1(gra)- hanam f suf 5 kusas sax x x x x x xn vitatadhva-

25 rānăm yū[

D Apparently these words belong to a verse not noticed by N 10 °sthequ "grhans M wee N p 502

[106 ]

Fragments of the Raghuvamsa 6 In < | ) tam klantasamyam kularājaX X X tyu- JJaga x pavanānta[ 7 Jjāmı mvesayamasa bali ba(l)anı || tām silpı samghah prabhul 8 JpitamiX X ( i| > tatah x x x x x sūpaharam punah parardhyapratımā[ 5

Fr 18 a

[M, XVII, 41-54]

1 ]prajās tad guruna nadyo nabhaseva vivar- dhitāh | [ 10 2 ]pra x x payan || vayorupavıbhū[ 3 Jbhyas sa navo py asid drdbamula iva drumab (I [ 4 Jlāpı svabhāvatah ( | ) mıkaşe hemalekheva śrır āsi[ 5 K(na) tasya mandale rajno nyastapramdhididhi- 15 teh | a[ 6 Jvikalpaparanmukhah | mantrah pra[ 7 ]pasarpaır jajāgāra yathakalam svapan napi[ 8 Ina jatupacito pi sah < | ) vrddhau nadr(mu)khe[

20 [M, XVII, 55-68]

1 Jkāmam prakrtıvaırāgyam satyam samayıtum kşamāh | [ 2 jh prarthi davanalah | na dharmam arthal

6 punab (rrading not noticed by N) pareb M 18 hemalekheva hemarekheva M tee N. pp 542-3 22 satyam sadyab thu rending is not noticed by N

[ 107 ]

M. TATICCHI

3 Jna madhyamasaktini mitrāņix x x xnyatab ! [ 4 jīmūtaś cā[ Jm iti tasyarthasamgrahah ( | ) ambugarbho hi

5 ]şv iva sravantīşu vaneşūpavaneșv iva | [ 5 6 ]varņair api şadamsabhak || khanibhih suşu[ 7 Jkhavix mah( | )babhūva vinix x x x n I 8 Jsa tasyāh phalam anaśe< R X

Fr. 19 a

[M, XVIII, 10-17]

10 1 Jpadāvasānaņ devādi nāmaļ Ipürvas tayor atmasame ciro[ 3 ]sīd dvişatām apīştaḥ < | ) sakrd vivignā[ 4 Jd yuvapy anarthyairvarvyasanair vihi[ 5 ]paralokayatrām yatary arī[ 15 6 ]pakșo pi si[

b

[M, XVIII, 24-30]

3 Jaradhya <vyil 4 ]s sa jate hiranyanābhe ta[ 20 5 Jni lipsuḥ ( | ) rājanam āj[ 6 Jsuto bhun netrotsavas soma i[ ]prasutam įį tasmin kulāpiį 8 Jrusevanena spastakrtiņ pa[

13 The akpara rva after rthyai is obviously due to some seribal mistake. 14 yatary jetary M, see N, p. 561.

[108]

Fragments of the Raghuvama

Fr. 20 a [M, XVIII, 52-53; XIX, 1-7]

I Jmadhu vanitanam netranirvesapeyam manasi- jatarupuspam ragabandha[ 2 ]pratikrtix x x x x x x x x tabhyas sama(dh)- s ika(tara)rūpāh śuddhasaļ 3 ļkanyāḥ || kramo nāmāștādasas sargabĮ 4 mah pascime vayasix x xvasi || tatra tīrtha- sali[ 5 ]h < || > labdhapa(lana)vidhau na tatsutah khedam 10 āpa gurunā[ 6 ]scana svayam a(va)rtayat sa(mab | ) sannive- sya saciveşv ata(b)[ 7 Jm adhikardhir u(ttara)b purvam ut(sava)m apo- had utsavah || i[ I5 8 Jkāh prajāḥ || Xravād yad api jātu mantriņām darśal

[M, XIX, 8-17]

I Jnujīvinaņ kox x x x kharāgarūșitam |/ bhejire[ 20 2 ]dirghikah ( | ) gudhamohanagrhas tadambubhis sa vyagah[ 3 ]bhayann arpitaprakrtikantıbhir mukhaih < || ) ghrānakāntamadhu[ 4 abhile[ ]satirekamadagandhinam rahas tena dattam 25

3 onirvesapeyam onirvesaniyam M, sce N. p. 576. 25 omadagandhinam omadakaranam M, sce N. p. 581.

[109 ]

M. TATICCHI

5 Jte tasya ninyatur asunyatam ubhe | vallaki ca hrdayamgama[ 6 ]bh)inayatilanghinih parsvavartișu gurux xjja[ 7 ]jivad amaralakesvarau || tasya savaranadrstaX x kāmyavastusu[ 8 Jlīkisalayāgratarjanam bhrūvibhangakuțilam ca vīkșitam | xkhala[

[110]

PRINTED IN ROME BY AZIENDE TIPOGRAFICHE EREDI DOTT G BARDI SALITA DE CRESCENZI 16 1962