365bet娱乐

VARGA V. P. 62, LL 2-4-( Karikas 52, 53). The order of enumeration followed here is that of Bhamaha. Bhamaha has- apahnurti visesokti virodham tulyayogitam | aprastutaprasamsam ca vyajastutinidarsane || upamarupakam canyadupameyopamamapi | sahoktiparivrtti va sasamdehamananvayam || kavyalamkara -- 3 . 2-3 . Udbhata has no and Bhamaha fails to recognise ; otherwise the alankaras and their order of enumera-

Notes. 119 tion are the same in both. But if we consult Dandin's Kavyadarsa we shall find that his order differs materially from the order of these two Kashmirians. We give below the lists of alankaras in order from the works of these three authors. Bhamaha. Udbhata. Dandin (Kavyadarsa, udatta ii, 6-7.) udatta udatta slista apahnuti apahnuti bisesokti apahnuti visesokti virodha virodha slista visesokti tulyayogita tulyayogita tulyayogita virodha aprastutaprasamsa aprastutaprasamsa aprastutaprasamsa byajastuti vyajastuti vyajastuti nidarsana nidarsana nidarsana upamarupaka upameyopama sahokki upameyopama sahokti sahokti samkara parivrtti parivrtti asih 000000000 samkirna (samkara ) ... parivrtti These lists clearly indicate that Dandin followed a tradition which was distinct from the Kashmirian tradition upheld by Bhamaha and Udbhata. However if we compare the whole lists of alankaras of these three authors we find that Dandin's list agrees very closely with the list of Bhamaha and Udbhata, so much so that we can almost say that the order followed in the treatment of alankaras is the same in the works of the three authors. It is also curious to see that no other writer in alankara literature substantially follows this order. Vamana has his own distinct order. Rudrata arranges alankaras

according to an intrinsically different principle of classification; his treatment has very little in common with other writers. Mammata arranges his 61 alankaras on a somewhat scientific basis and most writers after him more or less follow his grouping. As between Bhamaha, Udbhata and Dandin, we have pointed out that Dandin's order of enumeration differs in a few important points from that of the other two. Dandin places at the very beginning of his list. His list has in order svabhavokti, upama, rupaka, dipaka, and avrtti ; while the Kashmirians have rupaka, dipaka and upama at the beginning of their treatment. They have no avrtti, and svabhavokti stands thirteenth in their order. Dandin puts in, and suksma lava after ; while the Kashmirians do not recognise these three alankaras at all. Some further difference in the order occurs after 3 which we have specially referred to in the beginning of this discussion. (Vide the lists quoted above). 66 P. 62, L. 6-7 - apahnutih | apahnutirabhista ca | ( Karika 54 ). That cherished figure is apahnuti which includes upama (similarity) in a slight degree. Wise men compose it by concealing the object belonging to the theme. (r=- nikartha = upameya ). " Udbhata's definition of is practically the same as that of Bhamaha. Bhamaha has bhutarthapahnavena . ( bhamahalamkara - 3 .20 ) vague description of apahnuti, viz. kavyadarsa - 2 . 304 . He does not seem to hold that form ; hence the denial of instead of But Dandin has quite a apahnutirapahnutya kimcidanyarthadarsanam | is necessary to for the sake of establishing its own excellence is not at all wanted by him and his examples also point to the same fact. His example is- na pacesuh smarastasya sahasram patrinamiti || kavyadarsa -- 9 . 304 . The real as recognised by all other writers is e. g. tatvapahnava rupaka by him. apahnuti

Vamana's 掳 Notes. naitanmukhamidam padmam na netre bhramaravimau | etani kesaranyeva naita dantacisastava || mukhaditvam nivatyaivam padmaditvena rupanat | udbhasitagunotkarsa tattvapahnavarupakam || kavyadarsah - 2 94-95 . 121 definition is samena vastunanyapala popahnutih | kavya- . Here though he does not clearly state whether the apahnava is to be made of upameya or upamana, yet his example shows that he means the apahnava of upameya and of nothing else. All other writers such as Mammata, Ruyyaka, Rudrata and those who came after them follow the definition of Bhamaha and Udbhata, which has (1) upamanopameyabhava and ( 2 ) the suppression of upameya as its two integral elements; and Dandin seems to be all alone with his ideas about apahnuti . P. 62, LL. 9, 10-The MS. presents two blanks here. It is evident that some such words as those that are put into brackets must be originally existing there. P. 62, LL. 21, 22 -- visesoktih | yatsamagrayepi saktinam 22--16:11 (Karika 56) 1" Setting forth the non-production of effect even when the powers (causes) are all present, with a view to establish some peculiarity, is called ff. P.63, LL. 6, 7 99 (Karika 56) The varieties off are here described. "In poetry the beautiful structure of it (an) is observed to have two varieties, the one with the reason why the effect was not produced expressed, and the other with the reason not expressed. " We have taken laksya (the object) where the laksana ( of visesokti ) is seen to be fulfilled i. e. a poem, and have translated it accordingly. Another interpretation is also possible, viz. the thing to be impressed,' i. e. the ; then the construction will be laksye (phalanutpattau ) nimittadarsanena etc. The second variety viz. by 1 is illustrated first. P. 63, LL. 10, 11-maharddhini grhe janma | rupa = surupa, lavanya here. Here janma, rupam and vayah are the causes which are generally seen to produce sukhaprapti . But the phala-sukhaprapti is here said to be not forthcoming. The fc that is sought to be conveyed 16 [K.s.s.]

122 Kavyalankara sara sangraha. here is the wonder that springs from the gas having failed to produce sukha . Here the reason (nimitta ) why sukha was not produced is not given therefore this is adrstanimitta visesoktih ittham visamsthulam P. 63, LI. 19. 23-..... A hasty man always does his work very hastily. But here even a hasty man is described as not doing the action of asking in a hasty manner. satvaratva is the cause ; me vacah nodeti is the denial of effect. vicestitam visamsthulam drstva is the reason why his speech does not come out. Though the cause viz. exists, yet the effect does not take place; hence a sort of wonder is produced which is the fa sought to be conveyed. Or better, the may be taken to be the extremeness of visamsthulatva . Parvati's action was so strange (visamsthula ) that even the impatient Sankara was stunned to silence. fig= cross, perverse, whimsical, strange. the Much need not be said about Udbhata's fan. His definition of the figure appears to be the undeveloped form of the definition of Mammata which runs thus: qu karanesu phalavacah | kavyaprakasa - 10, pr . 658 . Here akhandesu karanesu is a paraphrase of Udbhata's samagraye'pi saktinam and phalavacah is same as phalanutpattibandhanam . We have always pointed out with emphasis the fact that although Mammata's wording. does not always agree with that of Udbhata, his ideas and his treatment of alankaras in most cases bear a close resemblance with the contents of Udbhata's work. His definitions of alankaras always borrow their ideas from Udbhata. Udbhata on his own part is indebted to Bhamaha in much the same way as Mammata is indebted to the former. But while admitting the general indebtedness of Udbhata to Bhamaha we must not omit to notice the great stride of advance effected by Udbhata over the comparatively primitive structure of Bhamaha. Bhamaha's ideas are, in many places, crude, vague and unsuited to the advanced notions that were in vogue in the time of Udbhata; therefore Udbhata polishes or improves upon them. Sometimes he puts his own wording in the place of that of Bhamaha. The figure is an instance in point. Bhamaha's definition and example are-

66 Notes. ekadesasya vigame ya gunantarasamstutih (or samsthitih ) | visesaprathanayasau visesoktirmata yatha || sa ekatrini jayati jaganti kusumayudhah | haratapi tanum yasya sambhuna na hrtam balam || bhamahalamkara - 3 . 22-23 . 123 When along with the withdrawal of one quality the existence of another quality is sought to be established for the purpose of conveying excellence, the figure is visesokti according to Bhamaha. It will be seen that this definition is of a different type from that of Udbhata and others. Bhamaha does not require ( 1 ) the presence of karyakaranabhava and (2) the non-occurrence of phala or karya as the important elements of the figure. It is true that his illustration is an excellent example of fa even from the modern point of view and is quoted by Mammata as acintyanimitta visesokti . But the view-point of Bhamaha and the modern Alankarikas is different. Bhamaha seems to regard and bala as two qualities of kusumayudha ( madana ) of which the former is suppressed and the latter extolled; while the modern Alankarikas would regard balaharana ( which is here denied, as the result (phala ) of the cause (karana ) tanuharana . It will thus be clear how Udbhata has abandoned Bhamaha and has created a new definition of which has proved acceptable to all who came after him. P. 63, LL. 25, 26- virodhah | gunasya va .... ( Karika 57 ) | "The expression of a thing different and contradictory to the quality or action [of the varnyavastu ] for the purpose of establishing the excellence [of it] is called virodha . " , Here 'gunasya va kriyaya va seems incomplete. Hence Induraja supplies, by the reptition of the word - dvirvasabdasyopattatvat dravyasya . The vivrtikara supplies jati and dravya ) - vasabdo'navaklaptim dyotayati tena jatidravye api grhyete . But the word viruddhanyakriyavacah seems to have proved a hard nut for the commentators to crack. Induraja's explanation amounts to this- viruddha ya anya kriya (1 utpadanam, padarthah, kavipratibhaya nirmitah abhinavah padarthah ) tatpradhanam vacah, i. e. imaginary creation by the poet of a thing contradictory (to the varnyavastu ) etc. The vivrtikara has anyasya viruddhasyeva kriya svapratibhaya (ta) karanam which is rather

unintelligible. It amounts to what Induraja has given as his explanation. It seems better to take in the ordinary sense viz. 'action'. This word should include under it guna and dravya by laksana . The expanded definition will then become- gunasya va kriyaya va ( dravyasya va ) viruddha - anya - ( guna - dravya ) -kriya - vacah tam virodham viduh etc, which is quite correct, intelligible and suitable to the example given. 10-:: ... P. 64, L. 10 - bhavatyah kkayamakarah | In this example the two opposite guna s viz. komalata and tapah patava reside in the same person (parvati ) at the same time. Bhamaha's definition of fis Bt M gunasya va kriyaya va viruddhanyakriyavacah | ya visesabhidhanaya virodham tam vidurbudhah || bhamahalamkara - 3 .24 . Rudrata's definition very clearly confirms the interpretation of viruddhanyakriyavacah ( Karika 57) that we have offered above. He has :- yasmindravyadinam parasparam sarvatha viruddhanam | ekatravasthanam samakalam bhavati sa virodhah || 0 kavyalamkara - 9 .30 . dravyadinam dravyagunajatikriyanam In the above example the two opposite guna s viz. komalata and tapahpatava reside in the same thing (parvati ) at the same time. All these writers, of course, have no idea about the real nature of fat such as is defined by Mammata, viz. virodhah so'virodhe'pi viruddhatvena yadvacah . They are not at all aware of the notion that the f (contradiction) between things should not be real but an apparent one. Real contradiction according to Mammata is a fault in poetry. Hence it should not exist at all. Thus all the examples of the authors like Bhamaha, Udbhata and others do not fall under Mammata's virodha . The present example of Udbhata -bhavatyah kvayamakarah kedam tapasi patavam is an example of visamalamkara (first variety ) according to Mammata, kacidyadativaidharmyana sleso ghatana- miyat - ka . pra . 10 . p. 719. Vamana in his kavyalamkarasutra gives the definition of virodha according to Mammata's standpoint. He has viruddhabhasatvam

Notes. 125 faza:. But the examples given by Vamana fall under the of Mammata and later writers. The underlying reason of this vagueness-and inaccuracy from the modern point of view-in the ideas of the older writers such as Bhamaha, Udbhata, Vamana and others is in the fact that they failed to discriminate between different types of contradiction or antagonism of sense to be found in poetry. Later writers from the age of Mammata mention alankaras such as virodha, asamgati, visama, and e each of which contains a distinct type of contradiction. All the alankaras, viz, virodha, asamgati, visama and adhika of modern authors contain some kind of contradiction and the old authors, not being very minute in their descrimina. tion, called all these different types of contradiction by one and the same name fa. Hence their definitions are rather vague and their examples also do not stand the modern test of criticism and sometimes do not even satisfy the conditions laid down by their own definitions. P. 64, LL. 20-21 - tulyayogita | upamanopameyokti ...I (Karika 58.) The prose order of words is : upamanopameyoktisunyaih aprastutaih prastavabhagbhirva [ sabdaih ] samyabhidhayi [ yat ] vacah [sa ] tulyayogita | ""The expression conveying similarity between things that are both or between things that are both c, the things themselves not being intended either as upamana or upameya, is tulyayogita . " The things between which similarity is expressed must all be either or ; naturally, therefore, there cannot be any upamanopameyabhava between them. For upamana is a thing which is always aprakrta, while upameya is that which is always prakrta . P. 64, L. 25-- purvabhedasyodaharanam | purvabheda is that which is indicated by aprastutairvacah in the Karika. svadangamardavam drastuh etc. I Here malati, sasabhrllekha and kadali are all aprastuta, kathorata is the sadharanadharma between them. Though Udbhata, like Mammata, does not say in his definition that is necessary for the formation of e, yet he states it in his examples; for really there is no other way of conveying similarity between things which are all either or aprastuta .

P. 65, LL 6-8 - dvitiyabhedah yogapatto jatajalam ...... | 1 Here similarity between yogapatta, taravitvaka and mrgajina is shown by the sadharanadharma sariranaucitya ( unfittedness for the delicate frame of Parvati). 6 In Udbhata has approached very near towards Mammata and has left Bhamaha and Dandin far in the distance. Mammata has niyatanam sakrddharmah sa punastulyayogita (ka . pra . 10 p. 642 ) as his definition of tulyayogita . Udbhata does not expressly say that one and only one of the things that are similar must be stated, for samya in samyabhidhayi cannot be taken to mean. It can only mean similarity or commonness and not common quality.' Yet it will be seen that in there is no other way of expressing similarity except by the statement of sadharanadharma . There is no upamanopameyabhava, no vacakasabda, nothing of the kind. The statement of, therefore, comes as a matter of course though not included in Udbhata's definition. the rule, established by Mammata, that only one should be stated is absent in Udbhata. Bhamaha's tulyayogita is :- nyunasyapi visistena gunasamyavivaksaya | tulyakarya kriyayogadityukta tulyayogita || Dandin has kavyalamkara - 3 .26 . vivaksitagunotkrstairyatsamikrtya kasyacit | kirtana stutinindartha sa mata tulyayogita || kavyadarsa - 2 .330 . But Both the definitions are very similar though differently worded. There is no restriction that both the parts must be either prakrta or aprakrta . On the other hand it seems clear that one part of the comparison is to be and the other to be aprastuta . The tulyayogita of these writers, therefore, does not very much differ from Modern authors like Mammata would certainly include it under . In one of the parties between which similarity is sought to be expressed is prastuta and the other is aprastuta . The prastuta thing is the upameya and the aprastuta is the upamana . In tulyayogita both or all the parties are either prastuta or aprastuta and there

Notes. 127 is no upamanopameyabhava existing between them. This is the point of difference between dipaka and tulyayogita, otherwise they are all similar. Udbhata and following him Mammata have discerned this point of difference while Bhamaha and Dandin have failed to do so. Udbhata's definition of dipaka is- adimadhyantavisayah pradhanyetarayoginah | antargatopama dharma yatra taddipakam viduh || ( Karika 14 ) pradhanyetarayoginah = prastuta prastutasambandhinah Mammata also clearly brings out this difference. His dipaka is sakrdvrttistu dharmasya prakrta- prakrtatmanam and his tulyayogita is niyatanam sakrddharmah sa punastulyayogita . niyatanam is explained as prakaranikanameva aprakaranikanameva va . P. 65, LL. 12-15 - aprastutaprasamsa | adhikaradapetasya ... (Karika 59 ) | " The commendatory statement of an object removed from the principal theme of description but conveying with it a sense of the principal theme ( prastuta ), is called aprastutaprasamsa ." P. 65, LL. 16-17 - na caivamapi ... prakhyata | If the prastuta is driven away and the is elaborately described in its place, then it will only have the value of the ravings of a madman who does not even understand what he has to describe. But such is not the case in aprastutaprasamsa yatah sa ... nubandhiniti | For in aprastutaprasamsa the aprastuta which is stated brings with it by suggestion the principal theme of description ( prastuta ). 17-1=9露a. P. 65, L. 17 - svajanyena sambandhena . Some kind of relation exists between the aprastuta and the prastuta ; and it is owing to this relation (sambandha ) the prastutartha comes in by implication. The sambandha is either karyakaranasambandha, samanyavisesasambandha Or upamanopameyasambandha . P. 65, LL. 20-21 - yanti svadehesu jara ... 1 "The rustic charms of inaccessible regions consisting of an abundance of fruits and flowers wither in their places not having found any enjoyer. All this is aprastuta and conveys the a sense that Parvati's beauty will die away not having found any enjoyer in this condition of ascetic life. 99 Bhamaha has as his definition of aprastutaprasamsa- adhikaradapetasya vastunonyasya ya stutih | aprastutaprasamseti sa caivam kathyate yatha || bhamahalamkara - 2 . 28 .

128 Kavyalaikara sara sangraha. Udbhata has improved this happily by putting prastutartha- nubandhini . Bhamaha has not that word in his definition but he means it; otherwise his definition would only apply to the ravings of a madman as Induraja says above. Neither Bhamaha nor Udbhata try to distinguish the different varieties of aprastutaprasamsa Ruyyaka in his alamkara sarvasva writes down distinct varieties: (1) samanyavisesabhave, (2) karyakaranabhave, and (3) sarupye i. e. upamanopameyabhave . Mammata has a similar but still more elaborate and minute treatment of aprastuta- prasamsa . ( Vide kavyaprakasa 10 . pp. 618-627) The modern commentator fabrings in his commentary these divisions of recognised by modern authors. He says :-- (1) esa ca ( i. o yanti svadehesu etc. ) svarupe prastute svarupasyabhidhane | (2) paramamitranarakasuravadhuryadi sarnginam nihatya nihsokah karosi tattvameva slaghya iti visese prastute kvacitsamanyabhidhanam, yatha- 'suhrdvadhu baspajalapramarjanam karoti vairapratiyatanena yah | sa eva pujyah sa puman sa nitiman sujivitam tasya sa bhajanam sriyah || ( 3 ) yah kascidraja sa tvamapatantameva pasyatiti kvacitsamanye prakrte visesa- tyabhidha, yatha--- 'airavanam sprsati mantrayate marudbhirvajram paramrsati pasyati yodhasarthan | merostadani visamikurute mahendrastvacchankaya nisi na yati narendra nidram || ' (4) prasthanavuddhirmaya tyakteti karye kvacitprastute karanasyoktih, yatha - yatah kim na milanti ' tyadi | (5) yatrapravrttatvat jnatvaiva vismrtasukadimoksaih kalatraih saha tvadarayah palayya gata iti karane prastute kvacitkaryasya vacanam, yatha- rajanrajasuta na pathayati mamiti | All this is after the manner of Mammata and Ruyyaka Mammata lays down five varieties of aprastuta- prasamsah- (1) karye prastute karanasya vacah (2) nimitte prastute karyasya vacah, (3) samanye prastute visesasya vacah, (4) visese prastute samanyasya vacah and (5) tulye prastute tulyasya vacah . Thus it will be seen that the passage above quoted from fa exactly follows Mammata. aprastuta Dandin has a very different sort of aprastutaprasamsa, viz. the praise of any da thing with a view to condemn the prastuta Bhojaraja who belongs to the same school of Alankarikas as Dandin has also the same sort of aprastutaprasamsa, /

Notes. 129 But the generality of alankara writers does not seem to have accepted Dandin's and Dandin with his follower Bhoja seems to be all alone with his rather queer ideas about this alankara. 99 P. 66, LL. 1-3 - vyajastutih | sabdasaktisvabhavena ...! - "Where censure is meant by the [direct] potentiality of words, but where praise is really desired to be principally understood, the figure is vyajastuti . saktih - arthapratyayanonmukhyam = the tendency to express the meaning svabhavah = niyatarthanisthatvam = the natural disposition (of the) to indicate a settled meaning. is the natural tendency of a word to express a settled meaning i. e. the process of a word. P. 66, L. 6 - vakyarthibhutatva ; avantaravakyartha | See the words occurring in the commentary and notes on c, P. 48. P. 66, LL. 13-15 - dhigvadopahatatvam na tatsvatmaparyavasitam | tatsvatma =itself, i. e. the fear itself. The condemnation meant by the word f does not end in itself. Though the word fa means condemnation outwardly its true aim is different, It really suggests the unrivalled beauty of Parvati. P. 66, LL. 11-12-... 1 This is not a very good example of vyajastuti, for the censure of Parvati's lavanya can be taken to be a real one. Parvati's lavanya is really cen surable if it hinders her union with a fit husband. For the sole aim of quis to obtain a fit companion for one's life. This kind of meaning indicating the reality of lingers in our mind when we read the verse. Such a thing is not at all desirable in. The examples. vyajastuti . hitva tvamuparodhavandhya etc. and he helajitabodhisatva etc. given by Mammata (vide . . 9掳 pp. 670-71) are very appropriate and beautiful. In them the outward and are absolutely unreal. Mammata's definition of vyajastutirmukhe ninda stutirva rudhiranyatha | 66 is- kavyaprakasa - 10, p. 670. It means: censure or praise at first sight, the final apprehension (:) being the reverse of it, is called vyajastuti 鈥� ." Thus it will be seen that Mammata, and fol- 17 [K.S.S.] 9.9

lowing him other modern authors, recognise a twofold: (1) where is apparent and is real, and (2) where is apparent and is real. Udbhata recognises only the first variety of these two and frames his definition and example accordigly. Mammata gives a twofold derivation of the word to suit his two varieties vyajarupa vyajena va stutih Uabhata seems to admit only gyajena stutih vyajastutih 06 P. 66, LL. 20-21-gz:... (Karika 61). When an impossible relation between two things or a possible one causes comparison (upamanopameyabhava ) between the two things to be understood, then the figure is called vidarsana ." P. 66, LL. 22- 24 - yatra padarthanam tatra vidarsana | (1) When the expressed relation between two things being impossible. suggests between these two things then the figure is T. (2) Also when the possible connection between two things brings in 3 as a support to itself the figure is vidarsana . visistasyarthasya P. 66, L. 24. - visistasyarthasyo ... padarsanat | Induraja explains the word fac according to its etymology. Now the word as the name of this alankara is very peculiar and uncommon. All writers on alankara except Udbhata have uniformly used the word and nobody even hints there is a variant fa in use somewhere. Our original MS. invariably writes face, and Induraja also derives the word with the 3c and not with by the expression etc. This circumstance precludes any possibility of a mistake at the hands of the scribe and we can feel certain that Udbhata's text as received by Induraja contained the title vidarsana and not nidarsana . But the Madras MS. containing the commentary has the name everywhere. This certainly raises some difficulty, but it cannot deter us from adopting fas the true title originally intended by Udbhata. Induraja is a very old commentator and he must have had a reliable text in his possession to comment upon. His declared. authority is on the side of the name . While the vidarsana .

Notes. 131 Madras MS. is not always trustworthy and the commentator is silent on the point. Bhamaha's definition of vidarsana or nidarsana is : kriyayaiva visistasya tadarthasyopadarsanat | jneya nidarsana nama yathevavatibhirvina || bhamahalamkara, 3, 32 . The definition is not quite clear in meaning. Still it is obvious that its sense is different from that of the definition of Udbhata. Thus Udbhata has here abandoned. Bhamaha and has created a new definition of his own, which was adopted by Mammata in later times. abhavanvastu- sambandha upamaparikalpakah | kavyaprakasa ; 10 . p. 613. P. 66, LL. 26-27 - vinocitena patya ca | Here the visobhata ( pallidity) of vibhavari ( night) is said to be borne by kamini which is impossible. For the sobha of night is inherently different from the of a lady. Thus the connection of vibhavari and kamini expressed through visobhata is impossible. A lady cannot hear the fac of night. This impossibility brings in the comparison : yatha vidhuvandhyavibhavari visobhatam pravibharti tatha ucitena patya vina kamini visobhatam prabibharti | This is the first variety illustrated. Example of the second variety is not given by Udbhata. Induraja therefore brings in a verse of Bhamaha to illustrate it. The vivrtikara 's comment runs thus : nanu bhinnayorvisobhatayoraikya- madhyavasitam iti bhede abheda ityevamatmikeyamatisayoktih | visobhatakhyo guna eko na tu svarupabhinna sambandhibhedattu bhidyate | svarupabhinnayoraikyatve'tisayoktih | yatha- svarupabhinnasambandhibhedattu 'snigdhasyamalakantiliptaviyata ' iti | atra hi vyaptilepanayoh svarupabhedaikyamadhyava- sitam | dvitiyasyodaharanamiha nastityanyata idam deyam | ayam mandadyutiriti | 10-11 P. 67, LL. 10 - 11-- ayam mandadyutirbhasvan eto. | bhamahalamkara, 3 . 33 . Bhamaha has no varieties of nidarsana like Udbhata. He gives this single verse as the example of nidarsana . Induraja enters into a rather stiff discussion on this verse. As usual he has brought in the science of grammar to dilate upon. His discussion has two parts : (1) tatra prathamodaya ... vasayatiti yatha ; here the discussion is grammationl and (2) tena ca prayojya ... vidarsanaya bhedah ; this explains how the

alankara vidarsana is formed in the example ayam mandadyutih etc. We take the second first. P. 67, LL. 19 - 23 - tena ca prayojya ... dvitiyo vidarsanaya bhedah | Here Indnraja makes the e. shows how the example fits in with the definition of the alankara. In this example the connection between srimantah and bhasvan is possible and real, not impossible as in the first variety. For the rich people are really made to understand by the sun that prosperity is followed by adversity. This relation between srimantah and bhasvan is called prayojyaprayojakabhava . srimantah is the (that which is directed or caused) and I is the (subject of the causal action). Now to establish the cogency of this prayojyaprayojakabhava sambandha comparison comes in. The sun (1) causes the rich people to understand that rise terminates in a fall just as his own rise terminates in setting. Thus the sun teaches the thing which is similar to, and is easily deduced from, his own conduct, viz. rising and setting. Similarity between the sun's movement and the instruction derived by the rich is quite essential here; otherwise no instruction can be derived from the sun who cannot directly impart any instruction by speaking etc. If the sun were to instruct some thing which is not similar to his own conduct, then his instruction would have proved useless and the prayojya prayojakabhava sambandha between the sun and the rich would have failed. But the similarity between the sun's conduct, and the instruction imparted give support to the prayojyaprayojakabhava . Then the cHT (which is not expressed but implied) helps the (possible) here. Hence this is the second variety of vidarsana . P. 67, LL. 12 - 19 - tatra prathamodaya ... vasayatiti yatha | Here Induraja explains how the really exists (bhavan ) between srimantah and bhasvan . bhasvan cannot impart any instruction as he cannot speak or make signs which will be understood by men. Thus it would appear at first sight that the connection of srimantah and bhasvan expressed through is impossible. But it will now be proved that the prayojyaprayojaka sambandha between sramintah and bhasvan is real and consistent,

Notes. 133 66 66 9 In a causal form there are always two subjects and two actions. One is the primary ( prayojya ) action which in this case is understanding". The other is the causal ( prayojaka ) action which is causing to understand" in this case. One subject is the prayojyakarta . He is the object of the causal action but subject of the primary action. The other is the hetukarta or prayojakakarta who causes another to do something. Here bhasvan who makes to understand ( bodhayan ) is the hetukarta and srimantah who understand ( budhyanti ) is the prayojyakarta . Now hetukartrtva is of two kinds one is direct and the other is in the form of samarthacarana ; eg (1) devadatto vasayati and (2) bhiksa vasayati . (1) devadatta causes some one to live, Here devadatta directly says usyatamatra and makes the person stay. Thus becomes directly the (producer) of basatikriya . (2) Alms cause some man to stay (in a place ). Here alms do not directly say to the man "usyatam " etc. He gets plenty of alms in the place and therefore he is induced to stay there. Thus alms behaves in such a way as to make the nian stay. This is called tat - ( prayojana ) - samarthacarana . In our present case also there is no direct causation and bhasvan is tatsamarthacaranavan hetukarta . The sun does not directly instruct the rich people just as a teacher instructs his pupils. The sun only sets and the rich people derive their instruction from the sun's setting. Hence samarthacarana . 66 " P. 67, L. 18 - presanadhyesanayorabhavat | presana is ordering and adhyesana is requesting. These actions are present in directly causal subjects; they are absent in samarthacarana . For, the samarthacaranavan hetukarta is not capable of giving direct order by speech or otherwise. The Mahabhashya of Patanjali contains a very clear exposition on this point of which is the basis of Induraja's discussion. It has : -- yavadryannimittam karanam tavaddheturiti | kim prayojanam | bhiksadisu darsanat | bhiksadisu hi nij drsyate bhiksa vasayati karisogniradhyapayati iti | kim punah karanam | paribhasike hetau na sidhyati | evam manyate cetanavata etadbhavati presanamadhyesanam ceti | bhiksascacetanah | naisa dosah | navasyam sa eva vasa prayojayati ya ahosyatamiti braviti | kim tarhi | tusnimapyasino yastatsamartha- nyacarati sopi vasam prayojayati | yatha karisognirvivate ekante samprajvalito adhyayana prayojayati | - mahabhasya | a . 3 | pada 1 | ahnika 2 | su . 26 |

Thus although the sun is not the direct af it is still the samarthacaranavan hetukarta and the prayojyaprayojakasambandha is existent and hence possible. The comments on the example very succinctly thus. bhasvato'tisambandha upapadyate | sa ca samameva | he srimantah yusmakamudayah patayeti upamaparyavasayi kila ravim drstva srimanta eva pratipadyante | ravistu tatsamartha- caranena prayojakah | P. 67, L. 24-:is the blending together of two figures. It is not an alankara by itself, and as such its treatment should in the fitness of things be deferred till all other alankaras are treated. Later writers like Ruyyaka and Mammata have done the same thing. Even Dandin gives the last place in his treatment of alankaras. Bhamaha too has enumerated (he has no ) at the end of all other alankaras. samkara according to Udbhata is of four types (1) samdehasamkara ; (2) sabdartha vartyalamkarasamkara ; (3) ekasabdabhidhanasamkara and (4) angangibhava samkara . P. 68, LL. 2-3-samdehasamkarah | anekalakriyollekhe ... (Karika 62) "[A sort of] occurs when more than one alankaras seem to exist; yet all of them cannot exist at a time; and there is neither any reason for nor any reason against accepting any of those alankaras [in preference to others]". P.68, LL. 4-20 - anekasyalamkarasyollekhe ... ityasankayoktam tatra na Induraja according to his usual habit takes and as the primary epithet, and accounts for all other epithets one by one. If possibility of many alankaras ( anekalam kriyollekha ) is laid down as the sole condition then in yhzq|hadeiht also there exist many alankaras; e. g. murarinirgata etc. Here gangeva cakradhara is upama and narakapratipanthini contains slesa narakapratipanthini = (1) The destroyer of narakasura ; (2) enemy (antidote) of hell. Thus the verse contains two alankaras and hence would claim to be To avoid the possibility of such examples becoming the words are put. The alankaras must be such as cannot exist at one time. If one is accepted the other must vanish. In murarinirgata etc. slesa and upasa can exist side by side.

Notes. 135 Therefore it is not samdehasamkara, nanu yadyanekalamkaro . ...... pramanabhavastatra samdeha eva | Even these two conditions do not sufficiently describe samdehasamkara . Two alankaras exist and both of them cannot exist at one time-these two conditions are fulfilled. But there is some badhakapramana for one and sadhakapramana for the other, Therefore the former is abandoned and the latter is accepted. Such an example will also fall under samdehasamkara . To exclude such examples, the condition viz. ekasya ca etc. is put . 'When no reasons either for or against any alankara, exist and both alankaras are equally possible,' then only it becomes samdehasamkara . The sadhaka and badhaka pramana s must all be absent. If sadhaka and badhaka are both existent then that alankara having sadhakapramana will be accepted and that having badhakapramana will be rejected. If only sadhaka is present the alankara having the sadhaka- pramana will be accepted. If only badhaka is present the alankara having that will be rejected and the other will be accepted. In all these cases there will be no at all, one alankara being totally rejected and the other accepted. Thus when all sadhaka s and all badhaka s are absent there is a possibility of samdehasamkara . Illustrations of all these cases having sadhaka or badhaka pramana s are given in kavyaprakasa - 10, pp. 761- 764 which may be consulted with advantage. maya badhakam | na hi Our modern commentator also cites examples in which the alankara is determined by the existence of sadhaka Or badhaka pramana s. It will be to our purpose to take down the passage here. 'arigirisu haristvam ' ityadau tu rupake, 'kulisabhrdiva deva tvam sampannacalanam ' ityadau copamayamasti sadhakam pramanam | 'smaranti jyotsnayah sasimukhi cakorastava drsi ' iti rupake badhakam | tadbhave hi 'pibanti jyotsnam iti syat | ' rajanarayanam laksmistvamalingati nirbharam ' | sadrsah priyabhiralingayate | The last example is drawn from kavyaprakasa . Further on he continues the discussion after the manner of kavyaprakasa and tries to assign a reason why Udbhata used the word aneka instead of ubhaya in the Karika bhanekala- kriyollekhe etc.. Thus he says-' nayananandadayindorbimbametatprasidati ' iti mukhyasyadhyavasanatkimatisayoktih, kimetaditi mukham nirdisya indusamaropanadrupakam, kim mukhanairmalya prastave'prastutaprasamsa, kim dipaka, kim tulyayogita, kim pradosavarnane visesanasamyatsamasoktih, kim madanoddipakah kalo vartate iti paryayoktam, iti bahu- lamkarasamdeho bhavati | ityubhaya sabdamapasyanekagrahanam | |

P. 69, LL. 11-12-zagian 1 This is an example of according to Udbhata; for in a there is a possibility in his opinion of two alankaras upama and rupaka . The ecmpound may be solved either as vara eva induh or vara 7. This is according to Udbhata ( and Induraja also ). According to later canons of criticism this would not at all be an example of samdehasamkara . upama and not rupaka will be the alankara here. The verse is addressed to Parvati and we ought to have as the word principally meant. If we accept rupaka ( vara eva induh ) the anvaya culminates on indu and not on vara . By upama ( varah induriva ) we can have the word vara the prominent member in the compound. Therefore 34 is to be accepted and not. The verb words in the sentence are consistent with and not with 5. Thus there is no possibility here of at all. Udbhata seems to have no idea of this line of thinking which is displayed in the works of Ruyyaka, Mammata and other later writers. P. 69, L. 22 and all other f...(Karika 63 a) 1 When alankaras pertaining to words and those pertaining to sense appear (separately) in one sentence (or verse as the case may be) it becomes [a sort of] ." This is the samsrsti of later writers. Ruyyaka says: tesam ( alamkaranam ) tilatandulanyayena samsrstih | niraksiranyayena samkarah | - alamkara sarvasva, pp. 192 and 197. is of three kinds: (1) of two (of course occurring separately); (2) of two sand (3) of one sabdalamkara and the other arthalamkara . This third kind is the sabdarthavartyalamkarasamkara of Udbhata. The other two kinds only are (q. v. p. 78 of the text.) in his opinion. P. 70, LL. 2-3 - ittham sthitirvarartha cet ... | Here anuprasa is the sabdalamkara and arthantaranyasa is the arthalamkara . arthantaranyasa (q. v. p. 34 of the text.) of Udbhata only require; statement of samanya and visesa is not required by him. Here rupena te etc. is the samarthaka of ma krtha vyarthamarthitam . But according to the definition of Mammata and other modern writers this figure will not be arthantaranyasa . For both the statements are visesa . (particular) statement; for it is etc. is a addressed to Parvati and pertains to Parvati's case only.

Notes. 137 Had it been rupavatya yuva sarvah padabaddho hi kimkarah it would have been a good example of arthantaranyasa, for it would have referred to beautiful ladies in general and not to a particular lady. 86 P. 70, L. 12- ekavakyamsapravesaddabhidhiyate ... (Karika 63 b) | is also formed when two alankaras come together in a part of a sentence." vakyamsa is a word or a group of words that form a part of a sentence. P. 70, LL, 16-17 - maivamevassva ( stha ) sacchayavarnika carukarnika | Here the alankaras 3c and are present and both are expressed by the one word iva . The upama is clear in ambhojiniva citrastha . slesa occurs in sacchayavarnika and carukarnika . varna means the pink colour of the lotus and the fair complexion of Parvati. Also is the central cup of the lotus, and means the ear of Parvati. This exists together with the because the word is present. If a is removed the 3 is lost and the double meaning of carukarnika and sacchayavarnika is also lost. Thus the two alankaras and have entered into one word iva . The far has a very serious difference of opinion as regards the nature of the ekasabdabhidhanasamkara of Udbhata. While Induraja says that ekasabdabhidhanasamkara is a mixture of two arthalamkara s, the vivrtikara insists that this samkara occurs when & sabdalamkara and an arthalamkara combine in one vakyamsa . He severely criticizes Induraja and explains the stanza maivamevassva etc. very differently. The whole of his comment on the stanza is worth quoting :- carukarnau carukarnika ca yasyah | sacchayetyadau padadvaye slesanuprasayoh sthitih | yattvekavakyamsetyadina arthalamkarasamkaro vyakhyatastadasat | caturthe samkare samsrstau vasyantarbhavat | udaharane ca slesopamayoryatsamkarye vyakhyatam tadapyupeksyam | slesasya nirvisayatapatteh | ' yena dhvastamanobhavena ' iti asyodaharana- miticedanudbhatamiti cet | The vivrtikara is here acute and piercing. On the whole it seems to us that the is in the right if we look to the wording of Udbhata. P. 71, LL. 2-3- anugrahyanugrahaka or angangibhavasamkara | parasparopakarena ... (Karika 64 ) | " When the alankaras are not independent of each other but exist by helping each other to develop, then that also is called samkara ." 18 [x. s. s.]

P. 71, LL. 5-8 - upakaryopyupakaraka parasparopakareneti | The alankara which helps another to develop clearly does some. And the alankara which is thus helped indirectly does by being the receptacle of the help of the other. Thus both are helping each other. upakara . P. 71, LL. 10-11:... Here there are three outstanding figures. hareneva tvaya is the upama . anangikrta means (1) not accpted, despised (1); and (2) deprived of body (asariratvamapaditah ). Hence it contains slesa dhastaryadiva na muncati is (hetu ) utpreksa, the cause ( dhastarthat ) of the action of releasing being a fanciful one. Here the is based on the first meaning of . It is only through boldness that a despised person does not leave the despiser. Thus here the slesa in anangikrta helps the utpreksa and utpreksa also lends additional charm to the slesa . The upama, hareneva tvaya, is also based on the slesa in anangikrta ; for anangikrta becomes the slista sadharanadharma . Parvati has abandoned (3) cupid; and has hara deprived cupid of his body (). But according to Udbhata is predominant wherever it occurs; other alankaras are to be thrown to the back ground. Thus according to him is the principal figure here and and are its accessories. P. 71, LL. 17 - 23 - evamayam ... catvarah samkarabhedah | Induraja's recapitulation of all the divisions of described before. It is to be noted that the names given to the four varieties of are of Induraja's invention. The Karikas do not contain any names of varieties. The ideas of Dandin and Bhamaha about are not at all so elaborate and systematic as those of Udbhata, Dandin uses the words and rather promiscuously. Bhamaha has only the name Exactly what kind of he intends to put forth is not quite clear, either from his definition or from his examples. He has - ghara vibhusa samsrstirbahale karayogatah | racita ratnamalava sa caivamudita yatha || - bhamahalamkara, 3, 48,

Notes, $139 From the expression it seems that he does not want to include the samkara of modern writers in his samsrsti, Dandin has :- nanalamkarasamsrstih samkirne tu nigadyate | kavyalamkara, 2, 359 . He has two varieties:- angangibhavavasthanam sarvesam samakaksata | ityalamkarasamsrsterlaksaniya dvayi gatih || - kavyadarsa, 2, 360 . It will be clear from this that Udbhata is greatly advanced in ideas about beyond Bhamaha and Dandin. P. 71, LL. 25-26 - upameyopama | anyonyameva yatra ... (Karika 65) "When comparison (upamanopameyata ) is reciprocal, the object being to remove the possibility of all other comparisons, [ the figure] is called upameyopama . " upamanopameyata = the relation between upamana and upameya ; comparison. P. 72, LL. 2-11 - nanu ca prakaranikam ... paksantarahanau tatparyat | upameya is that which being a party to the comparison is prakaranika ( belonging to the theme ). Now in upameyopama the real upameya is made upamana and the real upamana is made upameya . Thus the prakaranikatva of upameya will not be experienced in that case. Therefore the auothor says paksantaraha nigam . The chief object cizarzilaniq. of upameyopama is to exclude the possibility of other upamana s. The real upamanatva and upameyatva of the respective things is not violated by reversing the upamanopameyabhava between them. By saying varam visam bhaksaya ma casya grhe bhuktva, one does not mean that the person addressed should eat poison. The chief intention is only to dissuade him from dining at a particular man's house. In the same way the reversion of upamanopa- meyabhava is not important here, but the sense that there is no the two things ( upamana and upameya ) is important. Thus in kamalamiva mukham mukhamiva kamalam, the sense that there is no third thing comparable to and is important and not the upamanopameyabhava . third thing comparable 20 The vivrtikara has : - upamanantarasadbhavarupa paksantarabhavamavagamayitum yadupa mayopamevopamanakriyate sa upameyenopama ityupameyopama | anyonyasabdopadanaccatra vakyarthadvayavartya yamalamkarah | ananvayastvekavakyarthavarti | J

140 Kavyalankara sara-samgraha, Bhamaha's definition is- upamanopameyatvam yatra paryayato bhavet | upameyopamam nama bruvate tam yathoditam || bhamahalamkara - 3, 36 . He probably uses the word paryaya in the sense of parivrtti . Dandin calls this figure and includes it under the varieties of . Udbhata's definition of this alankara is a good one, that of Mammata being too elliptical viz. viparyasa upameyopama tayoh | ka . pra . 10, p. 583. P. 72, LL. 12-13 - siramsi pankajaniva ... | Here kara and cakra are made upameya and upamana by turn. The real upameya is kara, the sadharanadharma is vegotpatana ( quick annihilation ), the implication is that there is no third thing comparable to and in destroying the enemies. This verse and the following upto verse 5, Varga VI are relative clauses connecting themselves with the principal clause : etc. (verse VI. 5. ). In this verse it happens that kara and cakra are both prakaranika for the prowess of fag's and both is described, and both therefore can be upameya s. Thus there is no difficulty here. But in a case like kamalamiva mukham mukhamiva kamalan where one thing is prakaranika and the other aprakaranika, the prakaranika (viz. chere) is always to be regarded as zcc according to Induraja. Though it is expressed as 34 qla in gafya kamalam still by its real nature it is upameya and not upamana . P. 72, LL. 21-22-sahoktih | tulyakale kriye yatra ... 1 - (Karika 66) "When two actions taking place at the same time but depending upon two different things are described by one word (pada ), [ the figure ] is sahoti . " P. 72, L. 24-P. 73, L. 2-: etc....... caiva dipaka iti nativyaptih | Induraja says: In the example of dipaka viz. samjahara saratkalah etc. (g. v. p. 15) also two actions concerning two different things are spoken of by one word sajahara . kumumasriyah and sukhasampadah are both objects of the verb . That is, the two different actions belonging to kusumasriyah and to sukhasampadah are indicated by the one word samjahara .

Notes. 141 samjahara . Thus this example of dipaka also would come under sahoti . But such a possibilily is averted by the word tulyakale . . The two actions must be taking place simultaneously, and from the construction of the alankara the simultaneous nature of the actions must be impressed on the reader's mind. This is absent in dipaka . In samjahara saratkalah etc. the destruction of kusumasriyah and sukhasampadah is not done at the same moment or occasion; first the were destroyed and consequently were done away with. On this point vivrtikara remarks - na catra dipakatetyaha tulyeti | tatra hi ekavisrantakriya anusangenopakaraka sambandhameti | iha tu yugapadubhayagatatvena pratipadyate | P. 73, LL. 2-9 - sahadina casabdena ... vakrabhaniteh sadbhavat | Now this tulyakalata is expressed in two ways. (1) devadattayajnadattau saha Here both the actions (of dining) are equally predominant and both have their respective subjects directly connected with the verb (2) devadatto yajnadattena saha bhunkte | Here devadatta is directly connected with the action of dining ( bhojana- kriya ) as devadatta is in the nominative case. But the connection of yajnadatta with bhojanakriya is through the word saha . This second sort of way of expressing tulyakalata is resorted to in sahokti ; for this is included in a (indirect way of speaking). And must always be a necessary element in alankaras. The vivrtikara similarly ramarks - alamkara prastavaccatra caitrena saha maitro bhunka ityatra sahoktirna tu caitramaitrau saha bhunjate ityatra | Here the two P. 73, LL. 12-13-... actions are mrtyoh manorathavaptikarana and dyujanasya manorathavaptikarana . They are expressed by the same words aptamanorathah cakre . Now these are two words and the definition says padenaikena kathyete . But the purport of the definition is to exclude two verbs separately things. And we have expressing the actions concerning t no such two verbs here. Only the verb expresses the action concerning and. On the other hand the mrtyu ghujana . answer is that the action is a general one and without any particularizing attribute, it is quite incapable of conveying the sense. Therefore the particular manorathavapti is embraced by the general verb and the whole then appears as one word and not two. The fa also remarks here- aptamanorathacaka iti padadvayenapyabhidhane sadharanatvenoktih sthitaiva | sadharanartho hyeka-

sabdo vyakhyata iti kecit | ekena padenaikakriyakarakagataikakriyabhidhayina sahartha- paryalocanayarthaparakriyaksepina dve kriye kathyete sa sahoktiriti tu yuktam | 67) P. 73, LL. 25-26 - parivrttih | samanyunavisisthaih ... | ( Karika "Exchange of something with its equal, inferior or superior having an unacceptable, a positively disadvantageous and an acceptable nature respectively is called parivrtti . 99 The order in is changed in the translation. arthasvabhava means 'of acceptable nature.' anarthasvabhava has been made to yield two meanings: (1) of indifferent and hence unacceptable nature; and (2) of positively harmful nature. P. 74, L. 2-That with which our thing is exchanged, i. e. the thing which we take. The thing which we give for some other thing (parivartana karaka ). P. 74, LL. 3-14 - - tatra yasyah samo'rthah upadeyatvadarthasvabhavata | The parivrtti in which equal things are exchanged is anarthasvabhava i. e. of not an acceptable nature. means here abhavah, equals there is no desirableness. The in which a superior thing is given for an inferior thing is also anarthasvabhava . Here anartha means arthasya pratipaksah the opposite of desirable i. e. harmful. Thus made to yield two meanings. The six meanings of are well known viz, (1) tatsadrsyam (2) tadanyatvam (3) tadatpatvam (4) virodhita | (5) aprasastya - (6) mabhavasca nanarthah sat prakirtitah || Of these the sixth and the fourth are respectively made use of here. The cin which an inferior thing is given and a superior thing is recieved is . . of a desirable nature. % absence of desirableness. In an exchange with P. 74, LL. 16-17-... Example of Here uro datva conveys utsaha by laksana . urodana is the means to display valour ; and it implies utsaha . Here uras and yasas are regarded as equal to each other and hence The word is connectsed with c05: in verse 5, Varga VI. Here some minor jewels are offered to the milky sea and the great jewel is taken in return. A paltry thing is given and a valuable thing is taken. Hence it is nyunaparivrtti ; for the parivrtti s P. 74, LL. 23-24.

Notes, 143 are named occording to the thing that is given. Of course this is of an acceptable nature. P. 75, LL. 2-3... A great thing is such as c is given and a bad attribute such as accepted. Therefore it is fa. This is really of a despicable nature. P. 75, LL. 5-7 - - abhidheyapeksaya ... parigrhitatvat | Induraja here says that this example etc. is like the preceding one only in its outward sense. purport we shall see that it is not Looking to its at all. For (shortness) was voluntarily accepted by Vishnu (in to Gods. Vishnu has return recieved the as a means for giving not first given cc to Gods and in da from them. Therefore this is in its real nature not an exchange, and not also. Now the alankarain itself is not constituted as Induraja imagines. does not contain a real exchange. at all. An imaginary description of exchange where there is no real exchange at all is Any real exchange. such as krinanti yatra muktabhirbadaranyapi balikah cannot become charming and hence cannot be termed alankara even. In yo balau vyapta etc. as well as in netroragabala etc. and uro datvamararinam etc. there is no real exchange but only a poetical description of exchange. rasagangadhara . This unreal nature of the exchange in is explicitly pointed out by Jagannatha in his . He says:- esu (i. o. parivrttibhedesu ) danadanavyavaharah kavikalpita eva | yatra vastavastatra nale- karah | yatha - ' krinantipravikacalocanah samantanmuktabhirbadaraphalani yatra balah | rasagangadhara, p. 482. Bhamaha's qff is quite of a different nature. It is this- visistasya yadadanamanyapohena vastunah | arthantaranyasavati parivrttirasau yatha || pradaya vittamarthibhyah sa yasodhanamadita | satam visvajaninanamidamaskhalitavratam || bhamahalamkara - 3, 40-41 .

144 Kavyalamkara sara-sangraha. We do not know what has made him link with parivrtti . Probably his parivrtti does not include in it arthantaranyasa ; he is only giving in this place an example of parivrtti + arthantaranyasa and not only of parivrtti . To agree with this we have also arthantaranyasavati parivrttih yatha . arthantaranyasavati should not be included in the definition. It should refer to the example. Udbhata's definition of fis deficient in one respect. He repeats the word in the definition. Thus by putting a word. (viz. cada) derived from the same root in the definition he leaves the word unexplained. gia More advanced writers like Ruyyaka and Mammata have avoided this inaccuracy. They use the word instead of af. Thus Mammata's definition is: parivrttirvinimayo yo'rthanam syatsamasamaih | Similarly Ruyyaka has:ste kavyaprakasa - 10 . p. 674, samanyunadhikanam samadhikanyunairvinimayah parivrttih | alamkara sarvasva -- p. 152.