*

'' J.tt * l

'^4. 'J 1

*-" a-j:

PIFAC I ^I ,.. ^ ^ That.t^jW^J^sira qf ^barata is the most important single jtext on the performing ^ ta the Indip tradition, p^haps not be disputed. Nevertheless, it isa regrettable fact that this text ha^ not iecei^^ as mci critical attentif at the handsof mod^scholar, especiJly^histor^^ls it deservesi One reason for thi^ is that the text bristles with techmcSaits which can only be tmdnstood properly with the help of traditional e*pJa>|ations. Unfortunately, of the seve^ ancient commences qn the NtyaSstra the only; <aie Which survives apfrt ,i|oin the fra^entary^ Manbh^a ofJ^yadpT^a,^oj5e.or^ess fiiy K the Abhjnavabhrafff Abhiiiav3gupta^Tlie coinmtary of.AbhinavgupJa, too,.li not beea fiiUy. ths^-task , translated qr. e^p^ted. ^ ^ Much^of the ^ ffort"in i, 2 has -il been. ^ diiycted towar^ givin^ an account of dramatur^.. The Mp^a^st^a chaptos on music^ in particular, rem^ largely unattended! g ^
y

The present work tkesnp ft translation and critical explanation the 28th Qiaptn of the Ntyasstra along with Abhinava's commentary on iL This cliapter contains the principal part of.th musicolo^ xSf Bharata and Abhinava's commoitary is o^ sce guide to it". Tlfe coimentaty, thus, isof spinal importance for the miderstanding of the classical foundaliorf of Indi^ music. The difficulties of the task are as great as itsi importance. The text of AbhinayabhratTis corrupt and defective in som places*. An example is providedby the'ver^ which Abbinava quotes about calculating the nasia and udcBsta bf the tanas. The text as given here does not enable us to teach the formula in question. The pauity of mnuscript resources hastnade t^ task of.its editors extremely difficult. At places the construing of the text manifestly calls for some emandation..Otherwise too. Abbinava uses highly tchnical language and sometimes alludes to philosophical principles. It is only the combined rsouices of the under* cf. V. Raghavan, Abhindvgupta and His Works, pp. 171 ff., Varaasi, 1980.

standing of Sanskrit, musicology, philosophy and critical histoiy which can unlockthetreasurehouseoftheAbhinavabhrat. It isforthe scholarly reader to judge how far the present author has succeeded in this attempt Iwouldalso crave his mdulg^cfor^tiieinevitableshortcomingsofafirst effort. * I must express my gratefuhiess to the UGC for enabling me to complete this work as part of my project as Research Scientist. I am also grateful to the authorities of the Institute of Art History, Conservatiln &

CONTENTS
INTRODUCTION

^useologyoftheNationaMus^,Nev^DeIhiWiirefambasedtGC
Research Scientist. It is impossible for nie to acknowldge in detaU my mdebteAiess to tiie numerous scholar from whose works and'advice t have profited but I have made detailed klluson to the work which Ihaveactually utilized.

NTYASSTRA AND ITS COMMENTARY

Wor^Ja'to express my gfti'tude towards my parents Mrs. Sudlia Pande and Prof. G,C. Pande withoih whose inteU^timl tod moral support US work would not have seen the light of print. I also tiiank my moerin-lawMrs.rajanganaPandeforaUMerencourgemeii'Afie(dImust ^my husb^, Sri D.P. Pande, and my son Ansuman Who'" have always been so midtstandmg and helpful' am Vo b&lden to my pubhsho- Sn Rakesh Jiwari for bringing out"tins volutae.

NOTES

SELECT BIBLIOGRAPHY

^ e . . . . . , . Srikrishna JanamashtWi, V.S. 2()54,

Delhi. ,

INDEX

A n u p aPande .1
"

Introductior

Abhinavgupta is generally believed to bave lived in Kashmir in the second half of the 10th and first half of the 11th centiify A.D. This is mferted from the fact that bis Kramastotra was composed in the year 66^ of the Sajptarsi yeafwhicb is said to have begun 25 years after the beginning' of the Kali era. This would correspond to A.D. 990-1. His Jvarapratyabhijn-brhaRvimari gives the date of its composition S" the year 4115 of the KalT era^ which corresponils'to A.D. 1014. If we, suppose that be Was twenty-fivewhe his literary activities begaEa and that; he continued for ateW more yeafs affr the Brhii,bis dates would rang betwn c. 965 to 1025''A.IXr The line of Parvaragupta aiid later of the Loharas ruled Kashmir at that time. Didd exercised power till A.D. 1003 and th came Sangrmarja \yho foimiled the Lohara dynasty.The invasion of Mahmud Ghaznavi occurred duringthe lifetnne of Abbinava but although the Sbhis of Udabbn^ura fell to the Turkish invader, the Kingdom of Sangrmarja escaped destruction. The family of Ab^iinaya bad come tof Kashmir afeot two centuries earlia* v^hen the famous scholar Atrigupta Who belonged to the Antarvedi was brought there bjr te grt Kashmiri ruler Lalitiditya MuktapTito. Abbinavagupta himself describes bis ancestry in his Partrimiik vykhy and the Tantraloka. In the Imeage of Atrigupta lived Varahagupta whose son Narasirnhagupta was known as Cukbulaka popularly. He was the fath of Abbinava. Vimala was the name of his mother. Abbinavagupta was reputed tobe a precocious student at school. His father introduced bim to the mysteries of grammar. He studied the Saiva Agamas from'the son of Bhtirja and Laksmanagupta, Natyasastra and

ABHINAVABHARATI - INTRODUCTION. TRANSIATION & COMMENTARY 4 ABHINAVABHARA TI

literary criticism from Bhatta Tauta and Induraja, Tantra from / Sambhunatha. In fact, he wandered outside Kashmir also in search of leaming and studied at the feet of many masters. He studied heterodox philosophies also - ristikrhatabauddhdi - which is a point of considerable importance.'^ As. his mother died while he was still a child and his father later renounced the world. Abbinava was at first drawn strongly to the pursuit of literature and the fine arts but was subsequently drawn to the devotion / of Siva and ultimately became a famous Master and Adept. He did not marry and had no wife or child.^ We do not have many biographical details of Abbinava but his numerous works and the references to him in the works of others give some idea of his personality and achievements.He was a versatile scholar, poet, critic and musician, saint and philosopher. He collected and expounded theSaiva Agamic traditions of Kashmir, gave them a systematic philosophical form, revived and elucidati the Natyasflstra tradition of the performing arts, dance, drama apd music, devploped the doctrines of Dhvani and Rasa in the light of the philosophy of-Kashmir Saivism and thus laid the foundations of a truly originalfadian aesthetics. His Tantraloka, Isvarapratyabhijna-vimarsim, Dhvanyaloka and Abhinavabhrau remain pereimially admired and universal classics. It has been pointed out that thete. is a wonderful penportrait of> Abbinavagupta in some ancient verses of his^ pupil Madhurija Yogin.^ There, Abbinavagupta is described as seated on a golden seat i in a vine-grove, (drfor/na) inside a crystalline pavilion adorned by, pictures, perfumed by flower-garlands, incense and sandal paste and iL luminated by lamps, constantly resounding .with music and dance arid surrounded by bands of YogiriK and Siddhas. At his feet sat his disciples Ksemaraja and others attentively writing down his words. On the two sides stood two Dts bearing in their hands a jar of Sivarasa, betel-box, citron and blue lotus. His eyes were tremulous with ecstasy, a clear tilaka of ashes marked his forehead, rudraksa adorned his ears, his hair was tied with a garland, and he had a flowing beard. He had a rosy hue, his neck was besmeared with Yaksa-panka, his sacred thread was long and loose, he wore a white silk cloth and was seated in the yogic posture called nra.

His right hand rested on his knee and carried a rosary, his left hand played on the rida-mn. He was vaily the incarnation of lord Snkantha in Kasmiira. This pen-picture highlights the hnage of Abbinavagupta as a Tntrika and Yogi, teacher and artist. The chronological order of theworks of Abbinava has been discussed by several scholars. His encyclopaedic Tantraloka appears to a relatively earlier work which has been refrared to in his Dhvanyaloka-locanaas well as Isvarapratyabhijm-vimarim.The Brhaf-vimarim was composed in 1014 AJ). and the Isvarapratyabhijn-vimarim followed it. Abhinavabhrati refers to the Dhvanyaloka-locana^ Thus,' the first phase of his writings appears to be of T^tric works like Sritantraloka and Tantrasara. The great philpsophical works came towards the end. The aesthetic works could be of the same age or earlier. Whatever the precise chronological position of Locana or Abhinavabharati, there is no doubt that they presuppose some of the major philosophical ideas of Kamira Saivism. The notions of Sabda,Nda and Ntya, Dhvani and Rasa acquire in Abbinavagupta a characteristic depth on account of their suggestive reverbertions within the grand philosophical universe he helped to systematise and elaborate. In fact, it could be said without exaggeration that Indian philosophical thinking reached its highest peak in the writings of Abbinava. His practical and theoretical interest in the arts led him to lay down the abiding foundations of a truly Indian aesthetic. Kasnra had been the home of leammg and philosophy smce at least the Gupta age. According to one traditioii the ecumenical council convened by Kaniska' was held in JCundxilavana vihara in Kasmira. In any case, Buddhist schools flourished there. This is attested by archaeological evidence as well as by the evidraice of travellers like Hsuan Chwang and Ou-Kong.^ Kasnra lay on some of the routes joining India to Central Asia, Tibet and China.Students and pilgrims, Buddhist and Brahmanical, Indian and foreign, gathered there to study from celebrated Masters. There was, as a result,^much intCTaction of thought, which in any case was taking place on a wider scale and theintellectual life of Kasmira was not isolated. Vednta had presumably influced MaKyana and was in turn influenced by it as is shown by the example of Gaudapada.'^ Bhartrhari influenced the orthodox and the heterodox alike.' ' Snkhya-yoga and Sapistivda

ABHINAVABHARATI

ABHINAVABHARATI-INTRODUCTION. TRANSLATION & COMMENTARY

I t ' ' -

have many points of common intdrest. Nyya an<i Buddhist logic helped each other by mutual criticism.'^ Tahtricism was a common tendency shared by the Buddhists and the Saivas and Saktas alike. ^ Kasrmra Saivism owed its origin not only to the multiplestreamed Agamic-Tntric traditions but also to a mixed philosophical heritage der^ed from Snkhya-yoga, Nyya, Vednta and Buddhism. The dualistic Agamas could be connected witi the Pupatas who had a dualistici theistic orientation and were allied with the Nyaya-Vaiesika. The Snkhya-yoga with its theory of manifestation and essences influraiced Sarvstivada as well as the Agamas. Mahyna and Vednta influenced the understanding of non-dualistic Agamas. A diversity of Tmitric as well as philosopiiical traoiuons, thus, underlies Kasnra Saivism. The beginnings of Saivism have been traced back to Mohjodaro.'^ Rudra was an important Vedic deity who has been surmised by some scholars to have been apotropaic.'^ The Svetvataropanisadgives us the first formulation of Sdiva philosophy in which the coimection with Snkhya-yoga is clear.''The MahbHrata maitions the Pupata as one of the five main schools current then.'^ Archaeological evidence of Siva worship surfaces in theSukga-Stavhana period and is fully supported by literary evidence.'^ Thus, PataSjali mentions the worship of Siva as 20 Bhagavan. Lakua the traditional founder of the Piupata sect has been placed about the same time, a supposition which could be consistent with the evidence of the Mathura Pillar Inscription of the time of Candiagupta 11.^' It may be recalled that several ancient authorities refer to a fourfold division of the Saivas or Mhevaras viz., Saiva, Palpata, KrunikaSiddhntins, and Kplikas?^ Of these not much is known about the last two. The Pupatas werean ancient sect The Vednta^tras mention the dualistic-theistic theories of the Pasupatas?^ Psupatastras, Kaundinya's Pancrtha-bh^a on them and BhasarvajSa's GanaKrik remain the main sources of Pasupata beliefs and practices.^ The Saiva Siddhanta of the south was also dualistic but relied on an Agamic tradition which ultimately formed the basis of the composition of Tamilian saints in the 7th century and after. Like the Pupatas, the
/ /

Saiva-siddhantins believe in the absolute freedoin of 5iva and the dependence of the Jva, but their relationship which ultimately may attain to sayujya is to be distinguished from bheda, abheda and bhedbheda. It is through grace and worship that theJTva may reach this ultimatestage. The Iwenty-five tattvas of the SrMiya are elaborated to thirty-six ,in'tis system.^^ In contrast with these schools,,the school of Kasmiira Saivism which Abbinavagupta espoused was non-dualistic,; According to a tradition quoted by h'm there were three original varitids f the T^tra named after Rudra, Siva and Bhakavi^ representing the pomts of view xjf bheda, bhedbheda and abheda?^ There are supposed to have been eighteen dualistic Raudra Agamas of which different lists with a similar core are found in different sources. Sinnlarly there are lists of ten Saiva Agamas and sixty-four Bhairava gamas. The actual number of surviving Agamic works is very large but which of them may be regarded as ancient and authentic is still a matter for historical and critical research. Surviving Tantras like Mrgendra or Svacchanda are, however, held in high regard. / Abbinavagupta's Sritantraloka is a voluminous and encyclopaedic but clear and systranatic expression of the tantric lore. The words' 'Tantra' and "Agama ' have been variously understood. Literally, Agma is tradition but it is usual to regard it as more or less an esoteric and ritualistic tradition of spiritual knowledge, distinct from the Vedas, but claiming authority as the words of thesupreme deity. Although the Vedas are also called Agama and so are the Buddhist-and Jaina canonical traditions, this more restricted use of the word to signify the / / ^ tradition of Tantric texts especially of the Saivas and the Sktas is common. These traditions believe that spiritual knowledge is transmitted by God through the Word which in its essence is nothing but the self-affirming power of consciousness Abbinava avers that there is really only one Agama 29 Ritualism, too, has many levels in Tantra, culminating in the self/ realization of consciousness. The founder of the Saiva tradition in the Kali age is said to have been Snkantha who is nothing but another name of Siva although some scholars'regard himiis a historical figure. At his behest
f

ABHINAVABHARATI

ABHIN-iVABHRAii - INTRODUCTION, TRANSLATION & COMMENTARY

three siddhas descended on eatth. They were called Tryambaka, Amardaka and Sfihtha. They founded the Advaita, Dvaita and Dvaiiadvaita schools of Saiva Agamas. In the line of Tryambaka, the nineteenth was Somnanda who was a historical figure and was a great grand teacher of Abbinavagupta. Since the succession from Tryambaka was through his daughter, this school was know as Ardhatryambaka.^^ It has been suggested that if Somnanda belonged to the 9th centiiry, Tryambaka who preceded him by 19 generations should have belonged to the 4th cMitury A.D., which would be the date for the mtroduction of the Saiva Agamas in Kasmlra.^^ As a graieration of 25 years m vidy-sampradya is ovarconservative, this introduction could have taken place a century earlier. If Snkantha is to be regarded as a historical figure how early heshould be placed cannot be determined. This Ardhatryambaka tradition of non-dualistic Saivism in Kasnra has also been called the Fourth School {Tiiryakhya). It was appareny connected with the Tahtrika Kula or Kaula tradition because Abhinava's Kaula teachia- Sambhumtha belonged to the spiritual lineage of Somnanda through Siimatinia.^^ Now it is interesting to note that ie founder of the Fourth Tradition is described as Macchanda or Mina who arose in Kmarpa.^^ Thus Somnanda apparently an heir to two distinct but interconnected traditions, an Agamic one descending from Snkantha through Tryambaka, another more esoteric from Kamaipa tiirough Macchanda who is a well-known legendary name in the tradition of the Siddhas and could not have beai very far removed in time from Somnanda. Presumably Somnanda's fourti ancestor Sangamditya brought with him the Kmaipa Tradition to Kamira. Somnanda, ie author of Sivadrsti, is regarde^ as the archphilosopher of the Pratyabhijn branch of non-duaUstic 5aiva philosophy of Kasnra. A closely allied branch of tiie same philosophical system was called the Spanda branch which was founded by Vasugupta, a siddha whol was the contemporary of Avanti varman.^^ Vasugupta is said to havebeen divinly inspired to discover the Sivasutras inscribed on a rock in MaHdevagiri?^ Kallata developed the system further. Abbinavagupta was heir to the Pratyabhijn school of Somnanda through Utpala and Lalsmanagupta, and to the Spanda branch through

KallaU, Mukula and Bhattenduraja. He had been initiated in the Kaula tradition by Sambhunatha and learned the Krama system from Laksmanagupta and Bhutiraja. Kamlra Saivism as a non-dualistic system of theory and practice, thus, comprised several branches with subtle distinctions. These branches are called nayas or perspectives of undo-standing and acng.-Kula and Krama, Spanda and Pratyabhijn indicate the different nayas current within the Saiva tradition to which Abbinavagupta belonged. The expression Trika is sometime used for the last two as a unified philosophical system and spiritual way. Trika is also called Sadardha and is explained as the uity of Para, Apara and _ * / / Parapara, or of Nara, Sakti and Siva. The knowledge of trika is the same as Pratyabhijn. The real distinction between these different nayas is one of approach or choice of Upya. Kramanaya, also called Knaya or Mahanaya emphasizes Saktopaya and its ritual was centered in psychic practices. Kula-naya emphasizes Smbhavopya and disparages ritual. Trika has a broader perspective and neither enjoins nor rejects ritual.^^
f / I

The philosophy of Kasnra Saivism as developed by Abbinavagupta constitutes a monumental synthesis of diverse earlier traditions. This is not to run down its integral urty, originality or consistency but to emphasize its comprehensive richness. It would be a mistake to think of it as an isolated Kasmira phenomenon, for Kasnra was then an important centre for visiting scholars and pilgrims not only from all over India but from Central Asia and even the Far East. The Trika conception of reality as a non-dual consciousness or universal self creating the world out of itself by its own free will as a semblance of duality, of man as essentially divine, and of the summum bonum of man as the realization of this innate divirtity, are squarely a linear continuation of the spiritual monism of the Upanisads?^ If all the diversity of the world is the expression of the original unity of consciousness, if being is nothing but consciousness, how is the appearance of insentient diversity to be explained, is an inevitable question which requires to be answered. To regard phenomenal diversity as 'empirically real but transcendtfntally ideal', to use a Kantian expression, implies that there is a transcendental illusion at work, that creation is nothing except an illusion projected by the Supreme Magician. This illusion or My is

IO
t

ABHINAVABHIU

ABHINAVABHRATI- INTRODUCTION. TRANSLATION & COMMENTARY

11

an accq)ted principle in Vednta as well as Saivism. There was, however, a strong Vedantc tradition which believed in the reality of creation and insisted that My or Avidya only misled man into believing in the independent and ultimate reality of the created world. Against this, the tradition of Gau^pada and Sankara insisted on the total unreality of creation. There is no doubt that in this Gaudapada had been greatly indebted to the illusionism of MahlySia.^^ The reason for this was the realization that consciousness as the transcendental unity of experience must be timeless and changeless and hence free front real action or creation. This insight into the radical separation of consciousness and activity goes back to the Sakkhya which holds the Purusa to be etanal and attributed creation to Prakrti. The attempt of the Brahmaparinmavdins to thinkof Brahman as the Purusa with the powers of Prakrti was rejected by Gauckpada who declared the wiiole world to be an illusion superimposed on eternal consciousness.^' Sankarcrya, however, finds a place for sagxma Brahman and creation in his system vth its twin points of view, Vyvahrika and Pranirthika. Bdaryana had interpreted Vednta as a theistic monism, distinguishing it from the dualism of Stokhya and the illusionistic nihilism of the Buddhists.^ Sankara maintains this distinction but his emphasis on the transcend^tally illusory character of the world appears to militate against the commonly accepted notions of theistic creationism.
/

not constituted by its insentient and finite externality but its spirituality which is free and infinite and is not given at the common empirical level, else everyone would have been freed by Anupya ! As the'force of Karman, My and Avidy decline, the nature of teality is revealed by reflection, introspection and inner spontaneity. It is only the self-realized Salva who sees thai tiie world is noiing bt Siva just as it is the self-realized Vedntin who sees the world as Brahman. Sarvam Khalvidam Brahman. The process of self-realization m Saivism consists of the four Upyas to which parallels exist in Vednta. For Sankara duality is an unreal appearance of non-dual reality. Similarly to construe the etanity of consciousness in Sankara as inertness, is to confuse consciousness with some insentient object. For Sankara, God or saguna Brahman is of the nature of consciousness and endowed with creativity. In Saivism action is reduced ultimately to self-consciousness or tmaparmara or Vimara or Svatantrya. In Sankara, too, consciousness is self-consciousness. The question is, how to conceive the freedom of infiniteself-consciousness? Sankara conceives it as ineffable transcendence, Saivismconceives it as the manifestation of infinite appearances in the modes of space and time. It is, however, strange that Abbinavagupta does not make any reference to Sankara. Sankara, on the other hand, shows his tacit approval of the basic Agamic Principle of the spontaneity of consciousness in his Daksirimrti stotra of which the authenticity is shown by the fact that Surevara hascomposed the MnasoUsa-vartika on it. As Sankara issaid to have visited Kasrriira and his dates are not too far removed from those of Vasugupta, his non-mention m the Kasmira school is enigmatic. The indebtedness of Kamira Saivism io Sokhya is obvious since they take over the twenty-five tattvas of the latter bodily and adding eleven more to them make their own thirty-sk tattvas. The categories of Sakhya represent the most widely accepted categories of cosmologicalcum-anthropological analysis in ancient Indian thought. The analysis of tiie physical world into five types of matter was ahnost universally accepted, except for the rejection of akaia as material in Buddhist thought. The five tanmtras represent the ultimate sense-data. Unlike the realistic Nyaya-Vaiiesika, the Shkhya regards the tanmtras not as dependent

It is for this reason that the non-dualism of Sankara has been distinguished from that of Abbinavagupta and the continuity between / Upanisadic Vednta and Agamic Saivism has been obscured. It has been ' . ' said that Sankara regards the world as unreal while Saivism regards it as real. And, again, that while Sankara regards consciousness as inert and / passive, Saivism regards it as essentially active and creative. Such a differentiation, however, overstates the aaual distinction. Sankara does not regard the world as absolutely unreal, a mere nihil, tuccha or sUnya. He regards the world as an appearance conditioned by Ignorance but grounded in reality, accepting a theory of the gradation of reality into empirical and transcendental, relative and absolute. Nor does Saivism regard the world as absolutely real. It regards the world as a semblance or bhsa in which the nature of its reality is that the essence of the world is
/

12

ABHINAVABHARA TI

ABHINAVABHARATI-INTRODUCTION, TRANSLATION & COMMENTARY

13

qudities Of matenal elements but as their causal matrices. Along with the
sensoiy apparatus consisting of fvey Wtr-a^, five tomendnyos

manoi which constitute their subjective comiterpart, the tanmtras aie themselves derived from Ahamkara. The analysis of experience mto sense-data and sensoiy faculties is derived from obvious perception and simple inference. Smce tiie manas is able to receive tiie impressions of 4e various, senses it is asshnilated to tiiem. That the bimtas are derived from tiie_.a. and tiiat tiie whole sensory apparatus consisting of tiie tanmatras, tiie indnyas and tiie mams is derived hoxtv Ahakmra are two major metaphysical propositions which give Shkhya aud PrefabAyna a distinct subjectivist slant where tiie way to cosmological Aderme f7 philosophical antiiropology. The introspective process of Yoga becomes tiie process of reaching tiie source of human existence as well as of the world. The derivation oAham^a from Buddhi and of Buddhi from Prakrti IS accepted m Pratyabhijna as weU as in Shkhya but in tiie forma prakrti SSnki'^l classical San^ya, though npt so necessary in pre-classical Sinkhya.'^^ Besides the Pratyf^jna Prakrti is not independent but dependent on Aghora or ^anta. Tlus, while Snkhya at first emphasizes subjectivism by its ^nvation of dl tiie elements of experience-cmn-reality from Ahakkra and tiien emphasizes ideahsm by deriving Ahakkra from Buddhi it seeks to reverse ^ by deriving Buddhi itself from one universal natural pmciple which is regarded as insentient. Thus in Shkhya whtie tiie denvation of physical reality from cosmic Intelhgence represente a metaphysical idealism which is in harmony witii a tiieory of "Inmate Nature tiioroughly reverses tiie process except tiiat Prakrti itself is supposed to work for tiie bondage and tiie liberation of tiie Purusa. Nature^ bemg IS bemg for tiie spirit In Pratyabhifri, Nature clearly ceases to be an ulmate source but becomes an agency of tiie Creative Spirit of tiie natural world {-Anania) which is different for each soul.'^ Purusa is no longer an ultmiately mdividuated and unchangeable principle/but tiie Supreme Dmmty in a limited mode. The Shkhyan duality of subject and object IS sublated m favour of a self-conscious principle capable of objectifying itself tiirough a self-assumed limitation.

For Saivism the empirical objects are not manifestations and transformations of a matrix of insentient objectivity, but semblances - bhsas - produced by the free self-limitation of primal consciousness. As consciousness delimits itself in stages by the threefold mala viz., aitava, niyJya and karma, the universe of many different subjects and objects gets projected. The common empirical subject who identifies itself with the physical body, senses and mind is called Sakala Pramt, i.e., the subject with limited capabilities. When the subject is merely aware f nothingness oxiwtya, he isthe unya-pramt or Pralayakala. These two belong to the impure world of My. When it is aware of itself as pure consciousness without its inherent power being manifest,,it is ailed Vijnnkala or Vijrinakevala. Such a subject is above My but below Pure Knovyrledge (uddha vidya). Beyond this stage are the four universal subjects with fidi powers."^^ The twenty-five tattvas of Shkhya belong to the order of impure creatio, auddhdhv. The responsibility for this lies with My which operates with Five Limitors or Kancukas viz., Kala, Vidya, Rga, Niyati, and Kla. Of these Kala is the basis of the other four. Limited by these Kancukas, the individualized self ot anu perceives and acts in a world of alien objects arising from the Prakni corresponding to it and operated byAiuinta^ The pure order or Suddhadhva consists of the universal subject with unobscured powers, i.e., they are merely 'moments' of the supreme consciousness in a timeless order. Parama Siva is the name given to the supreme Being Who is both immanent and transcendent His'nature is free and infiniteself-consciusrieSs in which Praksaand Vimaria afe'uhited. Siva and Sakti represnt its two moments of the predominance of fraksa and Viniarsa respectively. With the distinction of aham and idam without the sense of real differentiation, three distinct moments arise viz., Sadsiva, Isvara^ and Sadvidy. In the first Aham predominates and objectivity or Idam is apprehended in an unclear or asphuta manner. Isvara consciousness has the form 'This {idam) am I {aham)'. Here objectivity becomes clear. In the Sadvidy the two sides are equally balanced.'^'
t _

14 ABH'NA VABHARA T ABHINAVABHARATI-INTRODUCTION, TRANSLATION & COMMENTARY 15

adds to those akeady recognized in VtA

^^ich iaivisw,

pure and impure creation rests on the fa^thaTin the f gap between/'.a^,andP7;a;:i,thenon-dualsdr^^r''"^ ness IS unhindered while in the latter th.r. consciousbetween being and knowine knn j ^ apparnt discontinuity
acting are delimited io VMious wa""

developed out of the Vedhgas and the Upavedas and which included the sciences of music, drama, sculpture and architecture. It recapitulates the traditional notion of the sacred and invisible or transcendental (adrara) value of art forms and activities and at the same time formulates the notion of art as entertainment (jonjana) and enjoyment (rasd).^^ In the post-Bharata ageof development of thearts, there was not only a proliferation of forms and techniques but their systematization in different Sastras - ritya, ilpa, sarigta and Ayya as well. Although the master concepts of Purusartha, pratibh, anukni, laksana, pranina, ipa and rasawere seen as univMsally relevant, they werestill not emphasized in their universality to constitute a general aesthetic. This was basically because of the fact that the incommensurable diversity of the media for the different arts acts as a limit to those whose perception remains concoitrated on thecharacteristic forms and practices of thedifferent arts. Thus acting (abhinaya), movement (gari), sound {dhvani) and words {abda) aire the media of dance, drama, music and poetry. The theorist analyses these in terms of form {laksana) and measure {praimna), excellence {guna, alankra) and defect {dosa),purpose and satisfaction, but in so doing generally ends with the characteristic diversity of the media. Music hassound and form but no meaning or inherent feeling. Poetry has word and meaning but no acting. Drama arises from acting or hnitative action. Dance is essentially rhythmic movement which is visually pierceived. The satisfaction which music gives to the ear, dance to the eyes or poetry to the intellect how are they to be compared ? The break-througlim this situation was reached by the development of the theory of dhvani in poetry which brought out theinherently dramatic character of poetry and hence enabled the concept of rasa to cover both poetry and drama. Abhinavagupta's great commentaries, Locana and Abhinava-bhratt, established this once for all. The philosophical interpretation which Abbinava gave of rasa as samvid-viranti or camatMra underplayed the essentiality of the specific roles of different media and techniques in the different arts and converted rasa into a universal aesthetic category comparable to Beauty The emphasis on Beauty suggests something objective and hence promotes the danger of seeking it exclusively in specific art forms. Rasa, on the other hand, clearly

whichtSsttx -Shed dow.e ^eSr " ^"

uc teaUsm or a e-sided spiritai

the universe within itself even as le needs is to recognized ^11 taneously TheworldisbothldeaandWm^Tco"'^

contains to it spon-

isnothingbut self-expression, bssnothingbutretumt?^!^^^ and Its method the recomitinn f ti, consciousness,^

ideas in aJ^im^dTll fte'^S OT conceived as the primary artist endo architect follow the paiadiemafi
being his creation.5i Human art imitate Th

of aesthetic divinity which is divine. The sculptor and

myths, dance presents divine and'^c^TrCt^^ ^ relate itself to the radiant effluence of the s f ^ Soma. This is not to say t^IfCsac^fl unconnected with popular and folk forms ^ ^'^''3w^ere A^iyasas/ra of Bharata may be said tor ctv ''onions of art The
the sacred and popular notions of the Ve^candWa^*''^'^'"'^

UIC V eaic and Janapada ages which had

16

ABHINAVABHARA F

ABBINAVABHARATI-INTRODUCTION. TRANSUTON& COMMENTARY

17

emph^izes subjectivity of art experience. Bhattanyaka had already brought out the universal character of this subjectivity.^-^ TTie distinction of from any merely psychological experience is clear in Abbinava where rasa ,s transcendental, tire return of consciousness to its own innate and umversal but mmiediate ecstatic nature. It is only the springboard to tins return t^at ,s provided by tiie well-formed media of the different aits. The visible movement-forms of dance serve to communicate rbythmic motion (tala-layasritam), i.e., tile pulsation of Pmna^^ as an evocation and since Pram is tire first manifestation of samvi^^ its inward movement becomes tiie door for its ma-paramara which is of tiie nat^e of camat^ra. Similarly tiie musical notes are tiie manifestations of N-ada which is tiie primal form of Vimarsa?^ Words, too are tiie expressions of Nda orprimal sound when articulatedand tiieirsignicatwn m poetry is not of tiie nature of information but of expressing roia atya, too is not of tiie nature of imitating external objects of nature but the mward recognition {anu^avasayd) of tiie consciousness intuited tiirough tiie representations of persons and situations.^^ Thus tiie media used by the different arts achieve tiieir ultimateeffect not tiiough specific causal activity producing sensations, emotions or ^mg mformation, but by helping consciousness to return to itself. Whetiier it is music or dance, poetry or drama, tiieir first effect is to attract and focus tiie nlmd and tiius reduce distiaction and dullness."Their next effect is to induce a generalized consciousness which is distant from tiie actual ego-subject or actual objects given innature.' Inpoetry and drama there is^apprehension of tiie essential nature of feelings tiirough images In the apprehension of tiiese suggested or evoked feelings. onsdousQess nolongersubordiriated'toaction or reaction as in tiie behavioural world The w of die vorld, instead, is a content reflective of consciousness So tiie last effect of art is to lead consciousness to a deepening and lucid intuition of itself. Consciousness creates tiie world, even as

the 17th caitury when the Vaimava acaryas reared a new metaphysical structure for understanding Bhak as rasa. The NtyaSstra of Bharata was the inspiration for many works of a commentarial or topical nature. Kohal^, Dattila and Tumburu were celebrated authorities on theatre, music and dance, already known in the age of Bharata. While the work of Dattila survives, the other two are known only from references principally in iheAbhinavabhratias'mee are most of the other famous authors and commentators on iVS preceding / Abbinava. Rhula, Raghunatha, Adhvahara, Jayadcva, Bhatta Sankara, Bhatta Yantra, Ktidhara, Mtrgupta and Sn Harsa are known as authors / , on theatrics and its divisions. Lolla^ Udbhata, Sankuka, Bhattanayaka, Bhatta Tauta and Nimyadeva are known as famous commentators on NS preceding Abhihava. Except for Nahyadeva the oths are known only from references to their views.
I / '* /

Itself.

^^st do^. And the perenal fimcton of art is to lead consciousness back from tiie. world to itself." Such a tiieory of rasa and by implication of art was made possible by tiie philosophical geruus of Abbinavagupta. It was not matched again till

Owing to the loss of earCT literature as also owing to its inhent excellence, AbhinavabhraR remains a work of singular importance. Abbinavagupta draws attention to theparallelismbetweai the 36 chapters of the NS and the 36 Tattvas of Saiva Philosophy. The first chapter corresponds to the Lord as Earth because that is the foundation for the seed of the cosmic tree. Here Bharata gives the traditional view about the nature and origin of NtyaSstra. Abbinavagupta interprets Imitation in a philosophical way aanuvyavasya or introspective reflection. He defines Ntya as"svdana-mpa-samvedana-samvedyam vastu rasa-svabhvam iti", the intuitive experience qf rasa. The second chapter dealing with the construction of the theatre and the stage is paralleled by the element of water asrepresented by the ocean, for life (samsra) is itself like a drama where the seed and growth need a pervasive and supportive influence. The third chapter dealing v/ith Ranga-Pj is paralleled by the element of fire which mediates between men and gods. The fourth chapter dealing with Tndava is appropriately associated with wind, the ever-moving element. The fifth chapter called Prva-ranga-vidhna is paralleled by ether which like the Prva-rahga provides space for the world-theatre. The sixth or rasdhyya shouldcorrespond to gandha-tanmtra but the relevant verse of Abhinavabhra is missing. The seventh or Bhvavyanjaka corresponds to rasa-tanmtra.The non-available comment on,the 8th chapter
/

ABHINAVABHARA TI - INTRODUCTION. TRANSLATION & COMMENTARY

19

18

ABHINAVABHARATI

mination by space, form and causality, or niyati,and tranporality, or Kla but also produces a distinctive objective world, vedya pratfi. Kala invests man with ImutedfreedcMn and creativity and makes him the determinate subject of an objective world which is highly individualized. This would be paradigmatic of the himian art world but for the fact that the actual or sakala pramata lacks the detachmrait necessary for aesthetic experience and natural to him to the extent he becomes enlightened. "Thus in hearing sweet music or feeling the touch of sandal etc., when insensitive subjectivity disappears (mdhyasthya vigama) there arises a toobbing in the heart Qirdaye spandamiiai) which is the expression of-Anandiakti" When the mind is immrased in music it is able to feel the pulsation of divine bliss, the bliss mtrinsic to the freedom of self-consciousness. In the process of manifestation Praka becomes the Vcya, the reflected universe of meanings, objects and forms. Vimara or freedom becomes the Vcaka, the expressive power which projects them and also apperceives and appropriates them ultimately withinthe unity ofself-consciousness.' Absolute consciousness itself becomes the Bindu or the focus of subjectivity in so for as it appears limited by objects. In so for as Bindu desires to mamfest its self-appropriation of the objects, it becomes sound or word. As the vital throb ih aU living beings (jTva-kala) it is Nda which is the same as Brahma-iakti or Par Vk, which iss near Brahman as the luminosity of the gem is near to the gem. Four stages of Nda have been defined - Para, Pasyanii, Madhyam and Vaikha?^ "ITiese successive stages of the Expressive Power lead to inarticulate musical sound as well as articulate sounds used in language. Musical notes are identified with the ^gross payanf (sthla pacanti). Its beauty is due tb the indivisibility of its form {avibhgaikarupatvam mdhuryam).The notes of instmmental music are identified as the gross aspect of madhyani which is less intuitive and inward than Payanti?^ Thus the metaphysical basis of musical theory is threefold. Musical notes arise from Nda, their appreciation is a fimction of Vimara, their production at the human level is subject to Kal. Vimaria as the appropriation of the object in the autonomy of consciousness is the common

entitled uphga vidhna should have spoken of mpa-tanniatra which would have beai appropriate as the chapter speaks, inter alia, of 36 types of glances. The 9th chapter or AngiKdhyya is the counterpart of sparsatanmtra. The 10th chapter or cnvidKna corresponds to sabda tannitra. The commentary on the 11th or mandaldhyya praises the Lord's powCT of making, the 12th or gatipracara provides the occasion for invoking Motion. At the beginning of the 13th chapter Abbinava piays to Siva as Vrsahka. From the 11th to the 13th the appropriate references to Pyu, Upastha and Padaare passed over. The 14th issaid to correspond to the Pmndriya. The 15th dealing with metres is appropriately held to be paralleled by Vk. The 16th dealing with laksana and alahkara provides the occasion to recollect Siva as the illuminer. The 17th corresponds to the 17th tattva, i.e., the sense of taste. The 18th dealing with the dasampas appropriately brings the mention of the eye, the 19th concOTied with the sakdhis the mention of the sense of touch, the 20th of the ear. The 21st chapt dealing with hrya is appropriately dedicated to the Mind. In the 22nd, Ahakkra is connected with sninybhinaya, the 23rd recalls Buddhi which like a hetaira shows any and every form. The 24th concerned with the characteristic Prakrtis of men and womai and their threefold gradation corresponds to Prakrti. The 25th chapta: concerned with diverse acting (citrbhinaya) raninds Abbinava of the Purusa-tattva. fTiapters 26th-30th are connected with the five Kmcukas viz., rga, vidy, kal, niyati and kla. The succeeding chapters 31st to 36th correspond to My, Sadvidy, Isvara, Sadsiva, Sakti and Siva. Thus the 28th chapt dealing with Jti-mmic and taken up here for tran!latioTi and explanation corresponds to the Kal-tattva. Ka^ is the primary product of Mya and is the first of the veils or Kancukas of the individualized spirit or Am.The universal self, as it were, fails to see its Own true nature as absolute fi:eedom or svtantrya. This non-seemg or akhyti is mva-mala. This delimited self or Am, oblivious of its true nature, becomes subject to My and acquires a newand limited idoitity. It bepomes capable of acting in a limited way. This is Kal, a restricted freedom; or limited, w/ont/ya, of which the other four Kancukas are products "Kala he kicit kartrtvam sute svalinganad anoh". It not only produces derminate knowledge or vidya, attachment or raga,self-deter-

20

ABHINAVABHARA TI

ABHINA VABHARA TI - INTRODUCTION. TRANSIA TION & COMMENTARY

21

pmciple in aU art appreciation in which the experience of wh objecvely presented is transformed into elf./. ^ camatkra, mandaOI rasa. Rasa \hus he samvidsive principle of .tecs 2^ l<mbam-nbmva comssponding to m,w T T ^

12.

Apart from the concept of tattvas and dharmas, the discussion of time in the two systems invites comparison, vide Stcherbatslgr, Central Concepts of Buddhism,G.C. Pande, Studies in the Origins of Buddhism, pp.549Cf. D.N. Shastri Critique of Indian Realism. Ttus Uddbotakara, the author of the celebrated Vartika on the Nyyasutras is called a Pasupatacarya. Cf. Marshall, Mohenjodaro and the Indus Civilization. B.K. Ghosh in Vedic Age, p.207. Cf. G.C. Pande, Origins of Budhism, p.307. Cf. Susmita Pande, The Birth ofBhakti in the Indian Religions. V.S. Agrawal, Bharatiya Kala K Itihsa Mahabhasya on Panini, V. 2.76 Cf. V.S. Pathak, History of S aiva Cults in Northern India; Select Inscriptions, ed. D.C. Sircar, Vol. I. Cf. Bhamaon Vednta-stras,2.2.37, Susmita Pande, op. cit Vednta-sutras, 2.2.31-A1. Fsupatastram (Trivandrum, 1940). e.g., Mrgendrgama (Pondicheny, 1962), Rauravgama, Vol.1 Sivajmnabodham (Benares. 1908). TA, Vol.1, pp.35 ff. Cf. G.N. Kaviraj, Tntrika Sahitya ry4.Vol.Vol.Vm.p.3657 Ib., p.3659.'Eka evgama scyatn Vibhun sarvadarsiri Darsito yah pravrtte ca nivme capathisthitah" Vide TA, 36th hnika, Svapii Lakshmana Ju, Kashmir Saivism, (Delhi, 1988), pp.87-95. Soomanda, Siva-drsti (Varanasi, 1986), pp.296-98. K.C. Pandey, op. cit., p.l37 Ib. p.547. Ib. p.545. Kalhana, Rjatrangir. Cf. Jaideva Singh, Sivastras, pp.xvi-xvii. R.C. Dwivedi, Trkadarsanam (1992), pp.10-11, Lakshmana Jee op cit, pp.132-37. For diffwences between Trika and Advaita Vednta Cf. Ib. pp.103-7. i'iW VidhuseMiaraBhattacharya, Gaudapdiyam Agamasstram. For Sakara's predecessors Cf. G.C. Pande, Sakkarcrya, Chap.6. Cf. G.C. Pande, Foundations of Indian Culture, Vol.1 Cf. G.C. Pande, Sakkarcrya, pp.351-55.

13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

mediated in the different arts.

differeny

REFERENCES

2.

fF(=ffFF,KashmirSansknSi "Ax navatitamesminVatsarentye yugaihse Tithi^asi-jaladhisthe mrgaslrsvasne.' M tthi stands for 15, sasi for one and jaladhi for four anH a t vomato^oii we have 4115 as the date Th-v fo and as Pandey, op. cit., p. 8. " ^ K-C. ParaM^ikS (kss). pp.279-80. Wafote (Delhi. 1987 VI vrn PP'^70I-2./^afarapVi7(cf S C Rnv cv. / uoI.VIII, a957irhaniT^ Culture ofKasmir Tamrloka, VIII. pp 3704-5 K.C. Pandey. o;,.c/f.,p.li,r^ Vol.1,pp.28-55 'PP-^'W-S, lb. Lc., also lb. VUl, pp.3702-4. qxiotedby KC. Pandy.op. cit.,p738 AbhinavabharaH L Tefen tn -it. ,
b. .he

3.

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

" J7

7.

S-.

IO'

^ong. see S.C. Roy, op. at. p.40 . G.C. Pande,Ufe and nought of Sakkarcrya, (Delhi,1994), pp.150

22

ABHINAVABHARATI

1NAVABHARATI

INTRODUCTION. TRANSLATION & COMMENTARY

23

42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64.

Cf. Encyclopaedia of Indian Philosophy, Vol.FV, G.C.Pande, Foun4ation^_. of Indian Culture, Vol.1 TA, Vol.IV, p.1689 -The lord Aghor creates the impura adhv in order to ptovide for the anu-jivas thirsting after bhoga. K.C. Pandey, op. cit., p.378. Cf. Lakshmana Jee, op. c.,pp. 47 ff. TA, Vol.IV, pp.1768 ff, Sivasstravimarsim,ad sivastras III.3. (ed. & tr. Jaideva Siagh, pp.132-33) Cf. Paranirthasra (Srinagar, 1916), pp. 40-44. Cf. lb. pp.35-40. K.C. Pandey has attempted a detailed characterization of the system - vide his work cited above, pp. 320-55. Cf. Paranirthasra, p.30. e.g., Isa - Kavir mam pabhh svayambhh ythtathyato 'rthn vyaddhc chsvaitbhyah satmbhyah. Vide the author's Historical i Cultural Study of the Ntyasstra of Bharata. Cf. Coomaraswamy, The Dance of Siva, pp.35 ff. Abhinavabharbti on chapter VI of the Ntyasstra contains the classical description of thetheories of thesecommentators of Bharata - Vide R Gnoh, The Aesthetic Experience According to Abhinavagupta (Varanasi, 1985). lb. p.14. Cf. Abanindra Nath Tagore, Silpa-sadanga on Prana-cchanda. Cf. "prk samvitprneparinata",TA quoted K.C. Pandey, op. cit., p. 885. This developed into Ndabrahmavda in Sangitaratnkara. Abhinavabharati (Parimala), Vol.1, pp.3,35-36. Cf Gnoli, op. cit., pp.64-65. Bhattenyaka was celebrated for this theory of Aesthetic distance - Cf. GnoU, op. cit.,pp. XX-XXm. Cf. K.C. Pandey, Comparative Aesthetics, Vol.1, pp.558 ff. Cf. Gnoli, op. cit., pp.XXIII, fri. Cf. Dhvanyloka (Benares, 1940), p.498 "Apare Kvya'-samSr Kavir ekah prajpatih". The creative faculty was known as Pratibh which Anadavardhana defines as inventive Prajri and which is identified with visarga inSaiva metaphysics - Gnoh, op. cit.,p.LII, also Cf. G.N. Kaviraj, 'The Doctrine of Pratibh in Indian Philosophy ABORl. So Abhinavagupta in AB, VoI.I, p.4, "sva-hrdayyatana-satatoditapratibhbhidhna-para-vg-devatnugraho-tthita-vicitr-plUrmrthanirmna-sakti-slinah Prapateriva Kma-janita-jagatah." He further quotes Bhattanyaka Nomas strailokya-himianakavaye sambhave yatah /
/

Pratiksanam jaganntyaPrayoga-rasiko janah// (Ib. p.5) TA. Vol.V, p.1946"Prakse khalu visrntim visvam srayati cet tatah/ Nny kcidapekssya Krta-krtyasya sarvatah //" Samvid-visrnti, Camatkra, nanda. Rasa, these are all treated by Abbinava as identical, e.g., TA, Vol.11, p.553. Vide K.C. Pandey, Abhinavagupta, pp.185 ff7/4. Vol.IV, p.1770. TA. Vol.n, p.552 "Tath hi madhure gite sparse va candandike// Mdhyasthya-vigame yasya hrdaye spandamnat / Ananda-saktih saivokt yatah sahrdayo janah //" This beccmes possible because the being of things is constituted by their vedyat. On the theory of the Four Stages of Vk,see K.C.Vaasy,Abhinavagupta, pp.625 ff. TA, Vol.11, pp.578 ff - The vocally produced sound of musical notes is described as sthl pasyan, the sound of musical instrument as sthl madhyam "Tatra y svara-sandarbha-subhag nda-mpirH S sthla Khalu pasyan varndyapravibhgatah / Avibhgaikarjipatvam mdhuryam saktirucyate//" Here the sweetness or beauty of musical notes is ascribed to the continuity of their vibrations. Cf. K.C. Pandey, Comparative Aesthetics,Vol. I, pp.563 ff. About the nature of rida and its relation to Bindu and the ultumate principle, there are many interpretations -see K.C. Pandey. op. cit., p.574 ff.

TRANSLATION OF THE TEXT

Ntyasstra and its Commentaiy

(Chapter-XXyni)

f;I; Ci

Text of theNS : Now we sitali describe tie rules about mscl^ aents :
I

Verse-1 : "The instruments shall be known as fourfold viz., tu a), bound (avanaddhd), solid (ghqna) and hollow {subira), along |'*plh their characteristic marks" Commentaiy ; Beyond this, is (the section) concerning Music. I bow |0 him, who is (Crescent) moon-omamented, whos&expre^ion as kalap 14iat from which the Middle Note derives its beauty and the g^ut of six^ *2 ttmdas originates in the order of the srutis.

The arrangement of the musical instruments was mMitioned at the ieod of flie last chapter. Hence for connecting with it, the verse read in the I sixth chapter is repeated h&T&-'stringed tc.' It may be objected, that, at that place (6.10) wht is said is "AccomI pamment, nots, musical instruments, singing and the stage, these con^ stitute'the^hole set". (So vi^y should the discussion of musical instr\ Bients precee that of notes ? The answer is thus). Hie stringed instruments prevent those doubts, such as about the actOTS in the role of Rama etc., from rising in the mind which would tend to obstruct (the spectator's) absorption (in the spectacle), through (such tunes as) 'the lord enters the forest'^ etc. Rhythmic timing (ila), too, bee^ the performance, notes and tempo evMi and smooth, like a regulating string and thus keeps the spectator's heart ent\vined, maintaining the beat-pattCTn {smya) tied to melody and thus functioning in the interest of rasa.Stringed and reed instrumoits are used in the production of notes. They bestow on the notes their ultimate value which is melodiousness. The notes produced by the singers may have hundreds of faults owiiig to

28

ABHINAVABHARATI

TNSUTIONOF THE TEXT: NATYA^STRA AND T$ COMMENTARY

29

harshness, but the notes of strings and flutes have a natural sweetness. In so for as there is a deficiency in the sweetness of the voice, it needs to be compensated by superimposed sweetness to be acquired from practice. As the notes are tied to the stringed instruments thae is no apprdiension of these being lower or higher. Where the harp is proply tuned, even an untrained person may obtain the correct notes. This is not so in a flute. Hence the stringed instruments have primacy. That is why the Masters have indicated the (voice- producing) hiunan (body as the) lute by coupling them thus," the wooden harp and the bodily harp". Hence, the harp is commonly regarded as an image or reflection.^ Here again, the notes of the harp are mentioned first - "the notes,, being of the haip and the body, have a dual location." In gandharva, too, the rule is that, the result accrues to the performCT, according to which the primacy of the body in the context of the result is shown. In singing, again, the priniacy is of producing a pleasing effect for the audience, by which the Sage (i.e., Bharata) has indicated the instrumental character of the siiiging. In the ila-portion again, being introduced through the ghana and avanaddha, the chapter on dhruva is used for the words of the song (padmsa). By it (pada ?) does not become the natural resultant of the essence of gandharva. ^ As traditionally stated, the 'song arises from the Samans', hence the song (as a structure of notes) originates from the Samaveda. Being produced from Brahmagita^ the gita {"gtaka ?) has not been described as Saman. Hence the definition of Modes (tis) is given first. So the strings are first mentioned. The notes are to be obtained there as stated. "Of the notes" etc. Their separate mention is without purpose, smce in the chapter on Kaku (Intonation) called the list of obstructions, they have been mentioned earlier. What more ? 'There are four types of instruments', from this it is clear that there is no inconsistency in holding that the nature of instruments is to be described. 'Characterized by the mark', this shows, that, since other types of instruments are excluded they have not been defined even when available. Hence, it should be understood thus- they are all included within these, sometimes as their parts, or else, if they are different, as their accessories. For example, the ctuk etc, are only parts of the percussion instruments like pata etc. except in the context of dombi. Similarly, in the jvlpatta, phalakavda etc., used in the ntya. Since clearly explicit

svara-var^ are not available in the tuning of the percussion instruments (mrjanmarga), all follow some part of the four types mentioned here or are derived from them. ^
Verse-2 : "The taut (ffltd) shall be understood as made of strings, the bound {avanaddha) is the drum (pau^kara). The solid (ghana) is to be kno\vn i|s la and the flute (vamia) is called subira or hollow". Commentaiy ; Taut, 'stretched mde of strings' etc., reveals thg

fourfoldness mentioned earlio*. The 'covered' {avanaddha - percussion iostrumfents) artf bound with leathei. TTie 'solid' (ghana) are made of hard solidity (mrii-kthinya) btit are not useful in the production of distinct notes {viista-vanui) relevant to ras and bhva. They are used for keeping time or maintaining the bet - pattern (smya) and are consequently called tld or time-mako- (beat-maker). ^ Where the reading is 'ghanam susiram' the neuter eliding isrelative to (i.e., on the assumption that they qualify) todya. They are also uSed With the genders of the specific qualiflands when referring to than. (An) objection (is raised). Sincethereeds produce notes, why are they -here mdntioned immediately after ? TTie answer is - In the secon on notes, the harp alone is regarded as pre-minfent Since the dxan'{vanaddha) has the capabiiity of producing 'mstrumental notes' (dhtu) and wordless or meaningless" melodies' (uska),and since thegftona are used to measure them (i.e., keeping time), the reeds {susira) aie included in the string (tata) like the harp. Then we have ii flute. The fourfold instrunients (todya) lend subsidiary colgur (uparmijiM). The acting needs to be made entertaining (uparanfamya), hence the two (i.e.,the orchestra and acting) should form a single ordered set (i.e., should be in harmony with each other). Although the instruments are fourfold, they are classified according to the primacy of svara oxtla into two viz., stringed and percussion instruments. Susira and ghana are successively their accessories. Acting implies the group of characters. Thus the three groups (=actors on the stage, the singers and instrumentalists) are designated by the word kutapa^ that which protects the kuta or sound, or that which enlivens or brightens theWge.

30 ABHINAVABHARATI TRANSIATION OF THE TEXT : NATYA^ASTRA AND ITS COMMENTARY 31

.1, '' m drama should be under-stoodas threefold viz., strmged instruments, percussion instruments and the players".

tanets (talikas) are for maintaining harmony (samya) in vocal and instrumental music. They are used only in the orchestra {kutapa).
VerSfc-6/: "The 'group' (Jcutapd) used in the performance of the play consists of the noble, base and middling characters and is drawn from different regions".

Commentary : This is said in "their (employment, but AB understands it as'combmaton) is threefold". Of diese four kinds of mstruments depending on drama, that is, includin|the set of actors in the play, there is a threefold special combination or coUection since the ms^ents themselves make two sets andtheclassofactorsparticipating m the drama constitute the third. That is said by. 'the other consists of the playere . Although, bemg constituted by stretching, binding and hollowness (e mstmments) have a mixed character (and hence constitute a miscellaneous group), stiU, since tiiey are treated as one. tiie designation another is us^ (for tiie actors). The primacy of tiie notes is tiirough tiie stretc^g (of tiie stimg), not from tiie drum,where percussion produces (onJy)aharmonyofechoingva/7m(varrz5Vra;,whichisnottiiecase m the strings (which produce clearly sounding notes). The same shaU be said m other cases.
Verse 4-5 : "The orchestra should be organized (kutapa sa) of the smger along with his wife or colleagues, the harpist, the

Commentary : Now 'he' (i.e., Bharata).describes the ritya-kutapa by "uttanidhama-madhyambhih". That is, the troupe of stage pMsons (patra - samuhah), fit to play tiie role of%igh' (noble charactCTs) etc.'Tu' expresses a distinction from what has gone before. The groups (kufapa) in the stringed and percussion instruments have a fixed place as will be mentioned in the chapto: on Drums (puskara). As for the theatrical band or group (ritya-kutapa), its location should be without obstacles in the interval (avyavadhi) as its proximity is useful, haace there is no restriction on its-location. So-'he' (Bharata) says 'rih deseti', which according to others gives a direction for specific use (i.e., specially in the context of location as variable; pratyupayoga).
Verse-7 ; Thus vocal music, instrumental music and stage performance (ritya), (although) located (i.e., performed) in different persons, should be used by the producers (riafya-yoktjibh^) in the likeness of a moving fire-brand' (Le. giving the illusion of a continuity)".

Meplayerandtheflute-p,ayer.Theplayersonrrfaga.p^^^^^ dardara constitute the other percussion orchestra ". Commentary : The assemljlage or orchestra should be constituted The IS ^id by 'tiie orchestra shall be organized' (tiius). Pangraha is (tiie ^ ^ o' colleagues) witi, tiie brass mia (symbols) in tiie hands, vipaiicl (nine stringed lute) whichdoesnothavetiiecompletestrings(oftiietiireeoctaves),and which IS played by tiie kona (plectrum), as well as tiie v75 witii twenty-one stogs. (And '{ca) indicates tiie inclusion of otiier kinds of vws Tat^iva cq (=fuithemiore) is intended to mclude tiie Wa' or liollow (reeds). e percussion group is meiitioned as 'mrdahgika ' etc. Mrdahgas are tiie puspas (tile triple dhmi); pacava has strings inside and has tiie ^a^ of the h^mka. Dardura has tiie shape of a huge pot. The words tathaiva ca (fiirtiiermore) include marola, karata etc. The bias cas-

Commentary : Are,nbt these three groups mutually indepoidait ? No. So 'he' (Bharata) says, 'Evani gnam ca vdyamceti'. That is, all these three bands orgroups (sanrnhl^sie to be made one (elbKva), one alone is not to be used. By saying 'the singer, the drummer (mardangika), the noble character (uttamaptra) ' many persons are indicated. Anticipating the objection that vocl music etc., (beliglo diffaent persons), 'he' (BharataXsays, "ofwhich the locationsaresdiverse." Singing etc., are not said to be indqioident of their prfatmeis. It has berai said (text missing) with care,^ it has to be accomplished. Hence, careful effort is needed here. Since the basis is diverse, that is since they are diverse p^ormances perceived by differentsenses, hence their, unity is'*to be accomplished, so that they may be the objects of a uqifled perception in the audienc'e/spectators.Hie light particlesof the flre brand

32

ABHOUVABBRA

TRANSLATION OF THE TEXT: NATYAASTRA AND ITS COMMENTARY

33

(aita) do not belong to different places simultaneously, but, their continuity IS ensured by the quiclmess of movement So in thestage-pnxiucon many actions are to be brought together in hannony. Hence the expression -almd-cakrapratimam-}^ in the likess of a moving fir^ brand.

(Bharata). There, as we have already said, it is vocal and instriimental music which is prmintot.^^ What is more, this being sO, why should the harmony of vocal and instrumental music be mentioned'in the chapt^ on 'success' {Siddhyadhyya). The earlier mode of explanation would be appropriate. The fimess of the drama for staging, when already helped by the acssories (pafanjaka) being distinguishable is taken up lat^. The commentator {kra) says. If song comes at the begitming, instruments being secondary come aft^'wards in the< middle in order, when the tryaira and caturara^^ are determined in accordance with&e ritya [should it not be "not iii accordance with ri^a but ?], but in accordance with song. Thus, song, ven thou^ it is an embdlishmrat (uparemjaka, what adds to the entertainmmt), it doe so while giving a d^inite order to the pedbrmance of the play. Hmce, it is approjaiate ot the rahga (Le., 17 purvaranga). But we are unable to understand this. Nothing can proceed without dqiending on the fi^ya.-Song, too, nee^ to be nployed acrding to the ritya, and the determination of tala (fryara etc.) foUows the directions in the chapto- on Gaiti (pipbably Gta). Hiis should suffice.
Verse-8 : "The which is stated to be created by (the music of) the string (Le., v^), depends on different instruments, and consists of svara (the seven notes of the octave) tola (time measure) and pada (words of the son^, that is to be undentpod as Gandharva (music)".

^ It may be objected that this has already been stated in the chapter on Actmg m General' (Samanyabhinaya). Tme, but that relates to acting h^e mutuaUy connected song and ihsfrumental music constitute tiie' subject As to why tiiis divison into three grops, it is tiie dramatic action on stage (rSiya) w^ch ^ to be spported (by otiier elements).!^ general acting' (scamnyabhuiya, unity is brought abot by tiie force of actmg itself. There is no dispute here. Hie structure consi^g of tiie melodic riiovement of notes (svara-gat) mutuaUy joined togetiier has to ^ made similar to tiie moving fire-bftnd.^^ Altiiough belongitJg to e^, flute and tiie die praformance on tiie instimnents (vadyayndhi) IS to be unified-l-^ Hence die division faito diree groups is rightiy spoketfof. ^ (Tlie) objection (raised is), wh die diiee divisions are unified die OTcc^ of the performance would be charaterizd by tiiefa- harmonious blending m its course (pryoga-samyaktvaX wiich is why tiieir faults are considered obstructive of success. Hence die nature of success shold be mamtiiied-after th topic of vocal music is takenup (geydhH^r). (The) ^wer (IS), but whi it is said tiiat success results ftooi song, musical instruments and dr^ wh is princ^ and what is subsidiary could be a matter of doubt Hence, before the exposition of die natui of success, tte constuents of performance should be known. Hence, it is mentioned tiiatiepetformancehastobemadeentertaining(/wr<w7aanbf;. The exposition of succs clearly presupposes mhksivad) die constituents of what would help die p^formance. There is anotiier view.> What is mdicated here,is, tiiatsuccess my be gamed from a perfortnance consisting of dialogues of die ten dramatic forms even witiiout singig and musical instruments. But tiiis is not correct, because, diat would make tiie performance (hty) mcomplete while .1 IS tiie complete nature of die m^ya tiiat is intended by tiie sage'

Commentary : To, describe the nature of string instfum^ts as a principal topic 'hie' (Bharata) lays a foundation by ^ying "yattu tantkrtam proktam "thus (what has been motioned as created by stings" etc.). The Commitator (tiiSkara) says, that, th varieties of vTn5, vipaci etc. are not to be counted as ofya. This is not logicaL All are included graierally in stringed instruments' (tantkrtam) ^d the rest ^ similar. This is accqited in our school. What is based on the performance of differrait instruments' i.e.,die instrumental performance follows the forms of gandharva as accessories. Haice, those fonns of gandharva remain primary. Theidea is ibaigandharya is the measure, standard, process and container in which it is submerged. As for ordar, gandharva,is distin-

34

ABOINAVABBARAT

TRANSLATION OF THE TEXT: NATYA^STRA AND ITS COMMENTARY

35

guished by notes, rhydun, and text,.and die notes etc., are thus ordered. The order is maintained in notes, rhythm and text. Thus the sioger along with his companion (parigraha) is to be counted within the strings (tala-madhye) . It is for this reason that the text says 'string-produced', i.e., with the strings as principal. Gandharva means, that of which authority is to be foimd in the Gndharva-sstra.
Verse-9 ' "It is called Gandharva because it is exceedingly dear to the gods and also pleasing, and is of the Gandharv^". Verse-lO : " Its (i.e. Gandharva's) source (yoni) is gna (Le., Smagna),the rn or harp as also the flute, I shall now speak of the rules arising from their notes".

"Of Gndharvas" in the text indicates the performers. Thus, the singer gets the result or reward, by attaining to a deep immersion in consciousness (samvit) like the appreciation of a fragrance-[gowdAatvi foigandhavaf\. So it has been said that iVan/ayowT/iJ performed even once in accordance with prescription, purifies the slayer of a brhmana. Thus, the result accruing to the performer is primary. It may be objected that, just as in gna, even here 'pleasing' shpuld primarily beof the audience,^^ and hence the connection in the text shOuld be betwem Garidharvas and pleasing. And moreovCT, since what is heard enhances pleasure, being tmelimelessly and since gndharva has visible and invisible fniits and is called gandharva because it is performed by them principally, it follows that the text should be construed to mean 'desired by the Gods and pleasing to the gandharvas'. Hence that which enhances pleasure in a special measure is gna, this should be supplied or added to the meaning bf the text. Since gna is used only for the sake of giving pleasure it camiot, therefore, be identified with (thei gndharva). This also eliminates the error of subsuming it under the result as it spontaneously follows from the beginningless pleasme of the gods [navomahphalasyandeh' is apparoitiy a mistake for na va mahphalasyndeh]. This discrimination with referraice to gndharva will be demonstrated by us in the Dhruvdhyya with respect to dance. Otherwise, if it were to be dernonstrated hre, it might confuse those who are not convMsant with the; technical modes of description. "Exceedingly desired by thegods"; Gndharva is that which supports or carries speech and also that which is of the gndharvas. As has been said byc/ya Vikhila 'in the ancient past, it is the speech that had beai lost from heaven from the gods.' TTiis illustration explains the present also. The etymology of "enhancing pleasure" (ptivardhana) has aheady beai indicated with reference to Nrada. The tikimra explains that 'ga' is to be understood for geya or song, dha is to be understood as produced from the speech or instructions of the Creator , as designation for the sentence, [perhaps it should read, veti vkyasya sanjneti for vkyasya sajeti] where va stands for the sentence, ra stands for musical instru,ment or playing with kaku (iriovements and modulations of the voice).
Possibly 'dhtu pravdajam'.

Commentaiy: It has been said, that gita is derived from Saman. Here Sama/w are thecal^ of causes. GndharvacomesfromSman, and, Gna comes from Gndharva.To say that since gna and gandharva are both constituted by notes etc., gna is not included in gndharva, is strange. Why could not the opposite be the case or why should they not be regarded as identical? To ronove such doubts, the text proceeds "exceedingly desired by the gods" thus. This indicates that it is p'ennial ]flditvam intheprintedtext appears amistakefor anditvam]. How would the gods abandon what they like? That it gives pleasure by pleasing the gods, shows that it produces a transcendent (adrsta) result. Now gods are the lords of the soises, the mind, soisations etc. These senses etc., function w^en impacted or vibrated and are (like) the musical instruments of the gods. By offering the extnnal objects such as sounds etc., and through their ttanscidaice(in pure apperception) one achieves a transcendent sacrifice (c^arthami^am). And this sacrifice at will iicch-yajanam) is also, indicative of.the attainment of supreme inw^d beatific coiisciousness (parasamvit). In this way is illustrated the attainment of the fhiit of mancipation, since the experiraice thus attained, approximates the blissful state proper to emancipation."Thus, this is a sacrifice of the gods which istranscendent and independentof wealthietc., as it has been said that Siva is more pleased by gandharva thn by the ancient ascetic pratices etc. ( or, by the recitation of the Purmts and ascetic practices, or by devotion to Puranas - Purnayogdibhyah).
/

36

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATTA^STRA AND ITS COMMENTARY

37

According to the fikkra the verse has the same sense as this etymology. Here there is no need to do more as it has already been explained and that is why our teachr has remarked-one could ask from where mdeed has gndharva come up here? Since gna has its origin in Saman, how has it ignharva?) been mentioned first? What is more, since gana is undear consideration, that alone needs to be defined, not gndharva To remove this doubt, we have the verse {'asya yonir bhavedgnam' QiS. 28,10) "The source of gndharva isgnc?^, vTn and va/ia. We shall mention their method arising from the notes"). Of this, 'asya gndharvasya' i.e. of gndharva.Song organa is the name applied tothe Tis or songs which are the matrix of Sman. Vna means the audumban vin used in Mahavrata., Vamsa is in accordance with the tradition of teachers like NSrada etc. These constitute the source of gndharva. Moreover, since the parts of the kutapa were to be mentioned, this description of gna,'^na, vamsa etc., has been taken up.^^ That is the origin of gndharva 'in the context of what is intended to be described'. This is left unexpressed. That is to say, gandharva will be described here,because it is used in th kutapa. Some others say, that, the intention of the verse is to make out the source of gndharva to be the songs sung by singers in the folk tradition (lokapravha) within the brahmagita^^ The commentator-('ifcrij holds that gna is primary while the lute and flute are secondary. The commutator has brought out this judgment about primary and secondary by his great effort to expound the text, 'they sing here rathantara namely gyatn'. This effort is of no use in the present context, besides, the root meaning 'to sing' has many meanings and may be used metaphorically. (The effort thus) is like the chewing of sand. Where it is held that rathantara is excluded, then it would be through ie use of a particular scheme of notes on the other hand, if it is only the song (giti) which is to be understood, then the relationship between the common and the particular, between thewords which make up the text and the notes which qualify them** would be like 'the resting on itself of the sky'^. Even if the twin characteristics of expressing the note and residing in the locus of the song are accepted there is no difficulty, nor does anything relevant . ,,. *** emerge m this context.
*

As to what is the use of moitioning gndharva at the'beginning of gna, the idea is to connect it with what has been mentioned earlier. Thus, the directions for the notes (svara-vidhi) connected with the emotions (bhvdi) etc., will be givoi along with the gandharva mode, qot any other.,
Verse-11 : "Gandharva should be understood as threefold consisting of svara, tala and pada. I will mention the definition and function of each of the three". Verse-12 : "The notes have t\^o bases: the body (i.e., the human throat) and the n^ l shall now describe the rules pf these (i.e., vocal and instrumental music) along wi& theif characteristics." Commentary : Well, ^at are the other processes (apart from svara-vidhi)"! (This is explained by saying) "gandharva is to be imderstood as three fold." "Three fold' means that, which is prescribed or constituted out of the three, svara etc. {svara^ tla &,pada). The ancits believed that the word vidh or mode refers to that which depaids on another (i.e.,to an accident or propaty of a substance); but the word vidM here, does not have the meaning of form or type here (prakra, the idea is, that, svara, tla'arid pada are not three prakras of gndharva, they are its three sources). If it v^ere so, we should use gndharva separately for each, but that is not so, that is why crya Viskhila says"gndharva is the union (samvya)'of svara, pada and tla". The,use of the word tma (svaratlapadtmakam) here shows their mixture. Perhaps pada indicates cause or attention (avadhna). This is logical because avadhna is not a separate part like svara, tla and pada to be counted in the list The combination of the different elements is part of the metiiod of use (itikartavyata), it Ls not determined by their relative prominence. The order of their mention shows their relative prominence. Hence notes are primary, tla helps them through smya since tla being derived from the root 'ia/a ', in thesense of establishinga foundation mkes the meaning clear. Still farther is the word pada, which, by its literal meaning suggests the basis of support (dfirat). So has been^id by crya Dattila "the combination of notes is located in the pada and is well measured by
** "talo namatma" should preferably read "talo manatma"

**

'nirpni' in the printed text seems to be a mistake.

'svarniti' does not constnie..It could be 'svarnmiti'.

*** "mandasya nalatve pi' is also cintya^ possibly, "padasya gumtvepi."

38

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYA^STRA AND TTS COMMENTARY

39

tala."'^'' (TTiis hierarchy of) pada and tla is not similar in gna, as compared to that in gndharva*. But with respect to notes alone there is much use of the stringed instrumraits (which apparently provided a common basis to both the systems with respect to the notes). Avadhna, being a kind of yoga or concentration is not relevant here (i.e., in gna). In the use of parivartakas on the percussion instruments in the purvarahga, success is attained only by pleasing the gods (possibly through avadhna)?^ That is why it has been said, "If the singer does not attain to the final station through yoga, evai sn, he becomes the companion of Rudra and lives happily with him". (This is to show the adrsta phala of avadhana). This is said by some. That end, not being attainable by words, cannot be gained without avadhna. Laksana means, their nature in terms of four rutis (i.e., whether the note measures 2,3 or 4 srutis) etc. Karmameans the activity which orders the notes in the form of graha, ama etc., so that the different jatis arise from them. Thus, although, the kutapa is very much present on the stage, the musical happenings or process there, consists of the rise and fall of notes on the v7n5, the voice and the flute. It is through the impact of the breath, fast or slow, that notes are produced in the flute as in the voice. In the lute, as in a mirror left and right are reversed, so high and low are reversed. For this reason, it has been said thai the notes have a double locus, as they arise from vim and the body. The notes-of the lute are maitioned first as they are more miportant. How that is to be made, has been shown. Their characterizations are to follow the gandharva. Now, the text mentions the varieties of the notes of the lute. This mention of them collectively is for thesake of quick understanding. What is mentioned as a list here, consists of the notes takai together and separately. Dravl Vm is produced out Of wood and is of nature of the goddess Vk. Hence there is no conception. "That sound is bom out of to the lost speech, "thus has ya Viskhila shown. As its basis even the wooden structure is called vTn. Similarly in the body also, the Muse, in the form of the voi is called VTTI. Mrchari-^^
'tena is appatently 'te na'.

Va^ 13-14 "Notes (svani), the two mosical scales (gramau), mrcchahs, &nas, stfdtas, vrtts, uka, 'sdhnpfa (notes), var^a, aiamkra, dKtus, snais,yaiis (oi* tUetis made ofthe notes), this collection is alwayssaid tobe in the wooden lute^.e., this ratire group is saidto belong to lnstrumital music) Verse-15 : "Svara, grama, alamkra, vartfa, stKna,ti and the (two) sdhrana (notes), this collection is (available) in the bodily lute (sarin vli) i.e., these are the constituents of vocal music)".

Commentaiy : Mrcchari is the same in the srin and the dran. Tana,on the other hand, although possible in the Srin should not be used there, because it is not conducive to ease that is, it is inconvenient For the" sake of practite it is used, but success in it arises only irough the exact number of notes of the vTn. Thus, even when one is not able to produce the notes from the voice only, by, having the distinction of notes comprehraided by the hear he i? able to produce the notes of the vTn [For hrdayparigrhitah svaraviesah should be read hrdayaparigrfiasvaraviesah], 'Kmisarfe of the dAms or the basis which support the iiwfoi etc. The use of dhtiduska is pleasant on the vT/i alone. [For prayogasya read pmyogi^^ca] Thus, the notes of the voice are the support (upavya) for the dhtu. That will be described as produced from the alankra. Srutis are useful only on the vTn, because they are based on the tightening and loosening of the strings i.e., they are derived from tuning of the \m?^ Thus will b described how to attain the given number or the standard sruti on the vTn. [Part of the line is lost]. That is why wm is useful for the practiceof the difl^eroit varieties of notes, wherein, beauty is reflected and that goes over to the pada consisting of words also ['VInenam' seems a mistake].
Verse 16-17 : "Vyanjanas (consonants), svaras (vowels), varias (syllables), sandhis (euphonic junctions), vibhakti (case-endings), riman (nouns) khyta (verbs), vpasarga (prefixes), mpata (particles), taddhita (secondaiy suffixes), cchanda (metre) and aUmIcra (figures of speech) should be undo^tood as relating to the verbal themes of music {paddi. That pada is to be understood as two fdd: composed {mbaddhd) and improvised {anibaddha, Le., composed of meaningless syllables)."^

40

ABHINAVABHARATI
TRANSLATION OF THE TEXT : NATYA^ASTRA AND ITS COMMENTARY

41

Verses 18-20:"Dhtw, v^a,nifkimna vili$^a,praesa,samy^iia, pavarta along with vastu, nitm, prakaran^ og^, vidri^ yat, laya^ avt^ava, margad^damrgq ai(H^with pt}^, fliese twenty-cme shouldbe known by thewise as (cmtained)inl.This indeed, K the cfdlection (cmistitatin^^luftorva andwill be esqxMinded in detail" Comnlentaiy : Thus chaxacteiised, the note is measured by t a l a , with this intention the pada is maitioned. The topic of tola, is here separated (the elements of pada are moitioned). Consonants and vowels consisting of ac and hai are to be undostood here. The use of notes may be understood in teims of the pada because the note is a property of the syllable. Others, however, say thatthe consopants hereare indicativeandrefto particles siich asyAonim etc.*^ The sounds are those which are seoi in thetis like the RaktagndM^^and in the pniks The vowels are those in theprofyAra'ac'. are groupsofvowels and consonants. Sandhis are for the functional and verbal affixes (sup aad tin). Upsargas are pra etc. Nipata is .should beavoided being harsh. Othas are to be understood through the division of the materiaL AlmKras ate upam, vesara etc. Uncomposed means, in prose or for a diffament purpose. Opposed to that is composed (nibaddham). Since the distinction Of consonants etc., has been mentimed what is recalled here,is, vocalacting (vcikbhinaya) what is omitted such as vibhva etc, will be mraitioned suitably. [There is some mistake in 'Kula va smaratyam]. With referoice to the Niagara CH-section of the elements of gandharva are listed as dftravetc. Andwiththis,thethreelistedaiejoinedtogethCT as gandharvasamgraha. Here, by the word prakarana is designated the madrakas^ etc., of which the details have beengiven. TTiis is cc^tituted by definition and examination. It is implied in the text that the reader should apply his own resourceful intelligoice; because* the OTder of enumeration has not been followedin thedefinitions etc. Thus, the uthor would be defining snui aftra- grama, the two sdKra^ and tis after stfinaS. This is not the order of enumalion, wiiich in any case has not beeafollowed. There, the purpose is in listing, not providing an occasion
'fvaram/enti'should be 'svarastaUna'.

for definition. The ikakara has tried to explain why the-ordefof svara etc., is not followed, by saying, that it is for the ease of understanding and simple prescription, which has not-served^^uch purpose and invites criticism and is incoherent (The point of the tikKra has not been clearly explained). Even so, the notes are presupposed by the two gramas, and, on them depend mrcchans and tanas. This is the logical order. As to the reason adduced for mentioning the sthnas after them, the sthma is not the'basis of distinguishing the svaras in their essence, but only a basis for designating them as tara etc. [svarapohakari is perhaps svampopakri, unless, apoha is taken technically!. If the note is produced in that isthna, it is so called. But then, since by this, the note does not acquire any essential nature outside tara, mandra and madhya, whether there would be anything left to be called its essential description, would be xmcertain after thesuska,sadharana has been mentioned. (Its explanation is missing and not clear). It is not true that kakali and antar are used only in iuska, because they are also used in th'eyfisin which nimda and gtlhra are used sparingly. Jti-sadharana is believed to be a coimnon point of meeting among thejatis.In thesuska, where is the occasion for a similarity of form. Svaras presuppose the Srutis, why are they ien mentioned without any occasion (i.e., why are they then mentioned later)? Having raised this objection, the commentator explains, that, this is only with reference to the body i.e., vocal music, not of the wooden lute. This is absurd. This, in fact, means that the Sruti has no use m the dravl But if h is objected that they have been mentioned in order to divide the scale, then they should have been mentioned only there. So, there is no point in trying to justify the order.
Verse-21 : "ThenthenoteS-:$adja(sa)iriiabhairi),gndHra(ga) as well as madhyama (ma), pancama (pa), dhaivata (dha), as well as th seventh (note) nidada (ni)."

Commentaiy : Now, intending to speak of the notes, the text begins 'now the svaras' etc. What is implied is, that having been listed they are now being described. The word svara is derived from the root svr, which has the meaning 'tosound or to afflict', or from svara m the sense oksepa i.e., blaming or attracting. Hence, it has been said that the svaras are so called, because thy afflict the mental state constituted by the perception

42

ABHINAVABHARAJJ

TRANSLATION OF THE TEXT : NATYA^ASTRA AND ITS COMMENTARY

43

of sound {sabdasvabhvacittavrtii) by making it abandon its habitual state of self-centered indifference, and at the same time, on account of their excessive chami superimpose and affirm their own content.'^^ Thus, they are called svara. Etymology has also been given in term of the letters. The svaras are so called because in their own varieties of ti, rga and bhs (melodic structures derived from ja(is), they shine forth by themselves. Now, 'he* (i.e., Bharata) describes the immediate and fixed order of jvaras by mentioning rsabha etc., the wqrd occurring immediately after this (referring to gndhra may refer to antara gndhra).But there is no such difference with referenceto tsabha. This is indicated by rsabha. -The difference comes about, somehow, on account of the notes Mkali and antara. Hence the particle 'caiva' (moreover) is added to rsabha and dhaivata. This is the sdhrana note and will be explained when it is described. Some say that the note or svara is that, which gives a specific form to its basic sruti {tasysrayabimiyh sruteruparcmjakah) and has the property of being smooth and pleasing at a particular position of the srutis (srutisthn) within a given and fixed interval, produced by a light touch of breath."^^ Others say, that, t^e sadja iruti is called svara. Some say it isa collection or samUha (apparentlyof srutis).We our-selves, hold, that svara is the smooth and sweet sound constituted by resonance and produced by the sound arising from impact upon a srutisthn.*^ Nrada etc., have given the following etymology of the names ; "Sadja is so called since it arises from the six places, namely, nose, throat, chest, palate, tongue and teeth". (1.57). This has not been accepted by the Sage (Bharata) because of its irrelevance. In the mn, even when that is not the case {tatfibhvbhvo 'pi) sadja etc., are obtained there. As to the explanation offered by some, namely, that v7n is an imge and hence can offer the image of the notes sung by the human voice perceptible by the same senses, that lacls cogency. What is meant by saying that vTn isan image of the body? Vina is certainly not perceived as the likeness of the body imaged in the mirror; nor does it provide a locus which may reflect the body like the mirror; nor is the v7n5 an image of the note of the human voice as the two are not perceived by the same sense, since the image of the sound, which is of the nature of an echo, is not there transmitted as in the sky (=ether).'^' Nor is the note of the human voice

seen to produce an image in the note of the v7n5 since the notes are not perceived as two. Even in th absence of a note sung by the human voice, the note of the lute can be produced, hence the note produced bythe hum^ voice is not an image of the note of the lute, the two being produced by separate efforts. Thus, the note is of the nature of resonance, produced iiiunediately after the sound, arising from the impact of the three fingers, subject to the regular pressing of the strings. Nor are there the six origins like the nose etc. Hence, it is false to speak of imaging simply from the reversal of the order of high and low. The image like character has been stated, simply because the notes can be raised or lowered in tuning in accordance with the notes of the humanvoice. By mentioning the number seven, it is shown that kkail and antara svaras are not separate notes. "There are seven notes", this is owing to the mode of production. Others say, that, in the beginning Brahma articulated the seven syllables (varnas) in tatsaviturvarenyam tlffough the notes ni, go, ri, dha, ma, pa, sa.That is why there are seven notes. We shall settle this later. Here, some have offered another explanation of the objection. The objection was that the srutis should haveberai maitioned first because the svaras are manifested in the order of the Srutis. The answer that is given to this, is, that this would be so if the Srutis called dhavni and nda were to be themselves perceived as svaras at definite intervals. But that is not so, because even high or low Srutis appear dependent on the svaras. As Bhattamatrgupta has said "the whole detail of theiraiis arises along with svara spontaneously. It assumes the form of a nectar - like essence for the ear by depending on the svara.^^ "It is only by the svarahood of the rati that what is the pancama in thesadja grma recognised bythe fourth ruti, that becomes the pancama in the madhyama grma when it has only the form of the third Sruti [Caturathasvarampa in the text should be caturthaSrutimpa. Similarly, Triyah Srutimpah ^ould be triyaSrutifpa. The meaning then would be, that the pcmcama, characterised by the form of the fourth Sruti from mdhyama in the sadja-grma, comes to have the form of the third Sruti from madhyama in the madhyama grma]. The third and fourth cannot be the same in the two grmas. The same will be the principle about dhaivata. And how would kkalland amara be nisda and gndhra.

44

ABHINAVABHMJ

TRANSLATION OF THE TEXT: NATYA^ASTRA AND ITS COMMENTARY

45

Nradiyasiks says that the cuckoo sings the pancama m spring.^^ How does that happen? Because no fixed srutis are located in the nose, throat etc. If they were so located then they would all be svaras. If the svaras were to manifest themselves by depending on the order of the srutis, then their mutual dependence would be circular, because the point from which the order is to be determined would itself remain undetermined. Hence, the srutis are described only to divide the grma. In the discussion of the satnvdis etc., in the rgas where they occur, the svaras do not have any fixed height etc, and in this, the luie resembles the body. The srutis have been menitioned after the svaras to show this (that the srutis have no independently determinate places in the lute or the body). On the basis of the svaras themselves, the crya Viskhila mentions the srutis after the gramas. Since there is no subdivision of the grmas in the Smaveda, hence in the Nradiyasiks which is relevant to the Samaveda, their mention (of sruti) is only for the sake of indicating the higher or IOWCT pitch. 'Yah smagnam'. The five srutis described as dipta, yat etc., are said to be prodticed by the special performanceofseparatenotesetc.Somesay;sincethesrMii isnotin itself a svara, but the svaras are manifested by the order of the snais, the sruti should have naturally been mentioned earlier. Here it may be said, if the word sruti signifies some particular spot which the breath strikes then by the rule " they should be understood to be sixty six." then all these would be (text lost). Some call such a spot^vara. Of this, smoothness or roughness is not a property. This is the rule about the breath and its impact. But if the sruti is not intended to be the product of the impact of the breath on a particular spot, and svara is the property of musical charm belonging to it, then there also, the illogicality of pancama and dhaivata remains. The characteristic of asound arising from the impact of thefourth sthana cannot be of that arising from the third sthna, because their characteristics re fixed, becaus'e the property of one subject (dharnii) cannot become the property of another. This also disposes of Hkaan antara. That the cuckoo sings the pancama would be here even more illogical.
'niyata sruttmpa abhvat ' should be 'niyatasrutimpbhavca \

There may be an objection, if there is no such sthna^*^ then the propejty of the sound produced from that sthana should be the propoty of that substrate,^^ since there is no reason for its absence. This objection is worthless. After the impact has produced the sound and after this, another sound is produced and this is characterised by resonante. This secondary sound or resonance has a naturally fluent and charming form. When a rope is struck or alternately a string, or when a stone is struck or brass, the presence or absence of this nsical porperty may be clearly noticed. Thus, this element of svara isvarm), although it is relative to a conjoint whole, is, nevertheless in practice seen to be different from the sound arising from the conjoint whole (samudya). [Here svarma or resonance is apparently distinguished from the primary sound of impact]. For this reason, the resonance produced by the soimd from the conjunction at the fourth sthna is the same as that produced by the sound produced from the third sthna. Only, being produced from a higher sthna or position it is recognized as higher. Just as the mandra sdja is recognised in (in relation to) madhya and tra sadja, similarly, the same resonance is produced by the sound of the cuckoo or by the sound produced by the contact of the fingo: with the strings or the impact of the maUaka and the striker. It may be objected, that, since the sound is produced by a different cause, the resonance should also belong to a different class. But this objection should not be made, because there is no such rule about causation. Thus, even though fire may be produced by a variety of causes, such as iron, crystal, wood, lightning etc., even thai, tjie smoke arismg from it is of the same class. Further, in the case of two banana trees, arising from a seed or a bulbous root, the same effect is seen despite the diffrence of causes.This may also be seen amongst scorpions etc.^^ Besides, there is a subtle difference, viz., smokiness (of the fire) although, the smokinss is the same, similarly here also." One can distinguish the note of the wh fropi that of the mallaka although the note may be the same sa^.That is why, the striking by non-singers does not produce resonance, [gtr of the text should apparently be agtr and ghth should be ghta.] For that reason, although they equally produce sound they are called imperfect. The

46

ABHINAVABHAMT

TRANSIATION OF THE TEXT: NATYA^ASTRA AND ITS COMMENTARY

47

intensity of the resonance corresponds to that of the sound produced by the impact, and it is correspondingly high, sharp, harmonious or the opposite. For one whd believes that the sounds produced by the impact of the air on sixty stlnas, produce (m tum) the sound consisting of the musical note, of which the essence is anusvara or resonance,^' For him^ the bodily mstrument is the finest when it is perfect. It gives a concrete form to gradation in which there is (high) and low. [In the text, sabdah are co-ordinated withjanayet which is a plain mistake]. By this declaration of the identity of causes [the .meaning actually is, identity of effect despite difference of causes] the charge of circularity is also rebutted.^ If the srtUis are wholly contributory to the svaras in the subdivision of the two gramas, then, bemg constructs and attached to the grmas and usefijl for them, they are mentioned afterwards. [Vibhgepi could give . better sense than vibhagopi].The effort in vocal singing is not mdependent of tiie order of sQunds. The order is noticed in loud singing, altiiough, it is rapid in its course. That is not so, the order is sensed in non-loud singing- '(The next sentene is partiy brokai and does not make sense). The musician ttamed in the notes learns ascent (roh) and descent (avroha) of notes. Then, tiirough tiie manipulation of tiie strings during practice he comes to be acquainted witii die places where tiie sounds are produced so tiiat tiiey are capable of generating svaras. This he leams just as he leams the vadi and the samvcS. So it has been said- "By the difference of the sthanas of the upward or downward pressure on the stiings, a new audible property is noticed. "So, crya Dattila has also said, tiiat a single sruti known at an atomically minute position or sthna may still be noticed, and thus by the word dhavni, he has described this minute position (anusthna) "ITiat sound which is perceived as sadja in ie sad/a grama." (1.22)
The srutis are clerly illustrated on the vTn. Keeping in mind that tiie sadja grma, and thejmd^yama grma, have tiie commencement of tiie order (i.e., octave) with tiie pancama note, the srutis are mdicated later' on. If, by a certain effort a certam place (sthana) is stmck, and ft'om that

SO can there be between notes.^' For the clear division of notes there must

be in the middle a position to be avoided. Hence, is obtained a svara with two irutis. If there is a difference of two positions, tiien, a note with three rutis is obtained. If thereis a difference of three positions, then a note of four srutis is obtained. Beyond that, with an intervalof four or more on account of excessive effort, there is discordance in the note {vaisvarya)^ Hence, there camot be notes with five srutis etc, Hence, raising by two srutis is for the two rutis notes. That is, for ga and ni alone. Increase by one ruti has not been mentioned. Because the confusion of sthnas wll lead to confusion of notes. Four ruti note, threeruti note, two ruti note, m this order we reach the desired ruti [kramemnartharuti is apparentiy a mistake for kramemrtharuti ]. Then again, four ruti note, three ruti note,- (and two ruti note?). This is the sadja grama, where sadja has primacy consisting of sa, ri, ga, ma, pa, dha, ni. Madhyama-grma consists of ma, pa, dha, ni etc. The order is caturuti, triruti and dviruti. Thus, in the same breast(= singer) the seven notes in a single vocal register,are arfanged high and low.The initial note(sa) is complete because it consists of four rutis (i.e., the maximum number of rutis a note can have). In the next two notes the number of rutis is eliminated successively by one. After reachmg the minimum number of rutis of a note, madhyama is then formed agam with a maximum of four rutis. We, thus get the four notes sa ri ga ma. In diesame octave, in the upper region, we have pa dha ni sa. Here, sadja alone touches the second octave. This same arrangement occurs in aU the three sthnas, viz., in the chest (mandra sthnd), the throat (madhyasthna) and head (trasthna). The position of aiiia and its samvdi is accorded only to notes which are similar (i.e., which have a similar number of rutis).Hence in reality, there are only three notes sa, ri, ga or pa, dha, ni (i.e., theii complementarles). Madhyama is the unchanging middle note. The caturuti note, being high is called udtta, the dviruti note is anudtta since it is lower. Triruti, being in tiie middle and a synthesis, is called svarita. That is why, the Vaidikas u^ the tremolo only in the svarila. Here also, in the alamkras, tremolo is used for only triruti notes. Notes above and below that may be touched plausibly, but one cannot go above the cqtusruti nor below the dviruti: Hence, the tremolo cannot be used there-[faimpathatn kampah should be katham kampah].^^

striking a sound is produced, and from that, another sound of the nature of resonance, tiien if the immediate next place is struck, there would be confijsion.* Between two notes, just as tiiere can be a positional confusion.
'adhyavahita' seems to be possibly 'avyavahita'.

48

ABHINAVABHARATI

TRANSLATION OF THE TEXT: NATYASASTRA AND ITS COMMENTARY

49

Verse-22 : Vdi, samvdi, vivd and anuvdi (thus), these (these notes) should be understood as fourfold by the employers of gna (gnayoktribhih this includes both the organisers and performers). Prose passage between verses 22 and 23 : That (note) when used as arnsa (in a jati or melodic structure), is then also vadi.Those two (notes) which have the difference or interval of nine and thirteen srutis, they are the samvadis (in harmonic relationship) of each other. Thus as, sadja-pmcama, rsabha-dhaivata, gndhranisda and sadja-mc^hyama in the sadja-grama. In the madhyama-grma also, these (obtain). Instead of the sadja-pancama, here (there is) the samvada of rsabha-pacama. Here is a verse (as regards this). Vrse -23 : "In the madhyam grama there is the samvada of pancama and of rfabHa. n the fadja-gmma, indeed, there is the samvada ofga^a and of pancama". Prose passage between verses 23 and 24 : Those (notes) which have two sruti interval, they are vivadi, just as rsabha and gandlmra, dhaivata and nifda. The vdi, samvdiand vivdl having been established, the rest are anuvadis. [In the sadjagrama, the anuvadis are termed_thus-of sadja (the anuvadis) are gandiera, dhaivata and nisda-, of rsabha are madhyama^ pcmcama and nisda\ madhyama, pancama and dhaivataare also (the anuvadis)of gndfira; of madhyama are dhaivata, pancama and nisda of pancama and of dhaivata (the anuvadis) are sadja, madhyama and pancama. In the madhyama-grma also, of madhyama (the anuvadis) are dhaivata,nisda, rsabha, sadja and gndhra-, of pancama (the anuvdis) are dhaivata, nisda, rsabha and gndhra-, of dhaivata are sadja, rsabha and gndhra-, of nida are sadja and rsabha-, and (the anuvdis) of sadja are rsabha and gndhra.] Because it 'voices' (rather, unfolds thejati or rga) it is (known, as) vdi, because it 'speaks or voices' in concurrence or harmony (with the vdi nptej it is (known as) samvdi, because of discordance (with the vdi), it is vivdl, because of following (the vdi), it is known as anuvdi. If these notes are (a little) low or high this (i.e., slight'discordance) this may be dpe to faults in the (instrument's) string, tying place, beam or one's (own) sense facultji. iTie rules of the notes of the fourfold classification are thus.

There are two grmas (musical scales) sadja grma and madhyama grma. Here, twenty two srutis (i.e., musical microfones) dependent (on the grmas) are demonstrated by the svara-mandala or octave of notes.
Commentary : To know the common combination of notes, the author proceeds to show their fourfold character, or, the particle eva is

intended to indicate the logical position in performance (of the vdi). The word 'ca' indicates the production of the samvdi when .touched by that. The word'eva' indicates the regular production of the samvdi. The word 'dtha', meaning 'thereafter' indicates the continuation of the samvadi and the vdi. 'Ca' indicates disregard for vivdi. Api shows that anuvTdepends upon the characteristics of the samvdi, that is why it is mentioned at the end. Thus, anuvdi bears the 'anupalapana' (is compatible with the frequent use oO of the vdiand samvdi, and it is not itself used much [Anupalapana now seems a mistake for anulapana]. It may be objected, that, the characteristics of the rgas in practice, namely graha etc., alpatva etc at the end, the varieties of the use of the grahas need to be mentioned. That may be, but graha, nyasa and apanysa are covered by voiffitself. samnyasa and apanyasa will be mentioned in connection with the suddha jatis. It has been said that sdavita, audavita, and alpatava are differer\t from vivdi and anuvdi. Bahutva is through the samvdi.Tara and mandra are relative to the nysa. This is not correct. This may be so in the suddha jtis, but since this is absent elsewhere, how can this be correct.^^ Anuvadi must certainly be mentioned everywhere as the lirriit. In that case, should not graha need to be mentioned even more? Without that, no melodic passage can proceed. The answer is, this is being mentioned as useful in common combinations. In the tis, the manifestation of bh etc., is produced by only this much without the diversity of graha, apanyasa etc. Thus, at some place, when the svara is being repeatedly articulated, the samvdi follows it. The awMviir imitates it, the vivadi is used occasionally, then the form of the rga is clearly expressed, for example "i dhap p, sa p sa ni ri p sa ni s sa, ma sa sa pa pa sa dhdni.ni fi<ni ni sa ri. "Here sadja is the vadi, (gndhra) and nisda are the iamvdi, madhyama etc., are anuvdi, and
anupalapana'cx)u\<\ix 'anapalapana'pi 'anulapana'.

50

ABHINAVABHARA TI

TRANSLATION OF THE TEXT ; NATYASASTRA AND ITS COMMENTARY

51

thus the form of the Mlava Kaisika is expressed^ This is the heart of performing. The addition of graha etc., only brings in some additional peculiarity or excellence. In gandharva,it is for the sake of some particular invisible result. This we shall explain later. Then the author defines the vdT. Vdi is known in performance by its vivid shining out. It is also frequently articulated and indicates the determmation of tra and mandra.Others say the amsa is a synonym. Still others, to prove it, read it as "That (note) when it touches the amsa". This is not coirect. Dattila etc., 'say that amsa is the vdi. It would be stated there, that, a separate definition of amsa is not necessary' The usage is with reference to the idea (bhva). Then 'he' (Bharata) gives the definition of samvdi. Those which have an interval of nine and thirteen srutis. ITiis is naming the svaras. Others say eight or nine are called the group of nine or eight. Similarly, the interval of thirteen means where there re twelve srutis in between. They are called samvdis,but the Updhyyas says antara does not mean interval (antarla). It means nature. Hence, the references to the note, of which thenature consists of nine srutis. Similarly, that of which the nature consists of thirteen rutis. Such notes are mutually samvadis. Just as, three octaves consist of twenty two srutis from sadja to sa^'a, including the three srutis which follow it and the antara sruti. Hie same mode of speech is employed here and so about having nine srutis, thirteen srutis. It may be objected that, thus, notes may have intervals of more than four srutis. What is more, from sadja to nisda an interval of eighteen srutis would be used. This should not be said. When so many srutis are gained with the impact of air touching their locations and producing fractional resonances (ndma) these are utilized.' Hence, thae is the rule of the number of srutis in the notes. Hce if there is a* continuous hearing of 'parts' of the sound, then the note is he^d as deformed or discordant.' Hence, rsabha consists of threesrutis. It is not the third sruti. Acaryas like Viskhila have shown, "whai the performer gives close attention to the ultimate Statioh of the srutis which can be experienced only inwardly just as in the cae of japa, then he attains to special invisible results." So, Bhattatauta has said, the self-experienceable sruti is svara (svasamvedya). In singing, the audience evay where, has a glimpse of

one's own nature (svampa). So it has' been said "while they are being heard, they are apprehended as if profound." The svaru is not (perceived or constituted) by part?. Nor is it a collection,since there is no simultaneity (in the parts). Even though there is succession, on account of continuity, caused by quick happening, there is a sense of simultaneity* and the continuing mental impression produced by the successive rutis upto the last ruti-sthna is the svara ensemble [anye appears a mistake for antye]. Let us retutH to the te*t.'' Here, the form of the note has been mentioned as nine or thirteen srutis, relative to the srutis to be jumped or accepted. They are said to be mutually samvdi, because with the same sruti interval they sound together (samvadanat). In the lute, the practice, is, that, this can be done by pressingimother finger, whilp the place of sadja etc., is being plucked. That is why thesamvdi should never be excluded. The name ^ according to the the meaning, and it is relative to the svaras and not to the srutis as mutually samvdi. Thus, there is no samvadi between madhyama and nisda, even though the difference is of thirteen and nine, that is why in the Sac^amadhyam joti where the madhyama is the amsa, the sadava (or hextoic form) produced by the elimination of nisda is not excluded.'^ In the madhyma grma, there is no samvada of rsabha and dhaivata. Hence, in Kaiiia where dhaivata is the amsa, the sa^va by (elimination of) rsabha is not excluded. It may be objected, th^ince ie meaning (of samvada) is obtained by ie name itself (safna), why^give the definition. No, in that case, even outside sadja and madhyama gramas there may be samvada of sa and ma, sa and dha and in the use of the sdhranas between ni and s, and ga and ma there would be no fault. Hence this is ie reason for (enumerating) the samvaditva in botii ie gramas. To indicate this very purpose, the author himself states thatin the sadja grma the samvdis are sa ma, sa pa, ri dha, ga ni. Elsewhere ri dha, pa is eliminated in favour of ri,dhaP It may be objected, that, if this is a complete enumeration, what then. With reference to thfe text 'thus etc.', (the enumeration) may be for including another illustration (of samvditva) similar to what is already weU established. But tiiat is not so it has been said, iat tiie sorvadva in boi (gramas) has been
here. So

should bti

asbhnvikf'td

52

ABBINAVABBARATT

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

shown here. This effort has to be made in gndharva, so that the order of thesamvarfismaynotbelost.(Ing5nd/wrva)evenwhen/Sa/randa/iia/-a notes are used, the relative pitch distance of the sadja and madhyama should not be reduced by a sruti. Hence it is used very little (in gandharva). Where thete is {alpatva of) ni and ga, there the sdhrxina notes are used. So (in gandharva), the use of sdhrana notes tends to enter into (antaragamana) sadja, mid madhyama. In {dhruva) gna, thesdhram notes are used freely in sancra. As there is no samvdi of such notes, therefore the anulapana of samvdi is obtained through the ainsa becausJ of its Very nature. Otherwise theother samvdis may be disturbed.'"* This is why the designation of sadja etc., is given. Even when savais the amsa, even then auduvita may take placfc with pancama being omitted (i;e., this may happen iii gna). However, in the case of SadjMadhyani, (there is no reduction of sruti in ghdharva - the text is missing here and this has" been^urmised). That is when the dots are used in the notation to indicate the sadharna notes, the sounding of to and ma takes place in the nisdd string. Where there is (alpatva of) ni and ga, there the sadhfanas are used. There {alptva or use of sdhranas) is like the antaragamana of sadjaand madhyama. Then, m singing, when on account of the sdhram, there is movement {sacra)of the srutas [h^d notes ?] they axesamvdi with only some. And if that is so, then, there is anulapana of the samvadi arid aksa acquires its characteristics And thus the other sakvdis may disappear. Thus is the designation sa4ja etc.; even when the ^ja is the amsa, we have the auduvita on account of the elimmation of pancama. Hence not sadja madhyama. (text lost). In the use of the dot or bindu, those which have the same place or articulation on the nis^i tantrl, they have [discordance (?)]. 'Thus etc'., gives the illustrations and also the enumerations. As a resuh in the movement of srutisa strange concordance may be seen Now we define the vivdis. Ga ni have twosrutis^ on account of tljeir nature bemg hidden {antarhitasvampa)?^ Hence, gndhra and nisda,' with two srutis ate the vivadis for aU the other notes. It is in relation to the other notes that the vivaditva has been mentioned, cm account of this characteristic difference, ^abha and gndhra, dhaivata and nisda, these four are mentioned in the text because they are near.' Others say.

that, the vivaditva of two srutis is with reference to the examples which can be found. This is not correct. Eveiy note that is vivdi has to have similar, dissimilar and neutral notes. Thus, when the sadja is amsa, ma pa are satnvdis, ri dha are anuvadis. In the rsabha, (as amsa) dha ma, tii ga, andpa ma sa?^ In gndhra, nisda{vdi), there is no vivdi, because nisda is samvadi, and the other five are anuvadis. Similarly, should one follow elsewhere. Where vci is the sovereign {svanii)' samvdi is like the minister {amatya) who follows, vivadiis like the enemy, hence infrequent, anuvadi is like attendants. [Yogavdi should be yo'nuvadi ]. Now the author defmes the anuvdis. This is'quite clear when 'he' gives the etymology of the names saying 'vdi' is from souhding etc. It may be objected that in the vJn, because of the interruption of the order of the srutis sa, ri dha there is irrfegular division of the samvadis etc. Hence the text says "Etesam" etc. This is about the samvdi etc., of the notes. Thus, between sadja and madyama which are regarded as samvdi, there could be equality or excess in relation to the intervals of thirteen and nine. For the vivadialso, the discordance is in relation to relevant intervals. Anuvditva also continues to fit even if there is an excess or defect. Thus, when fsabha is* the amsa, a note with a defective sruti may become the anuvdi of sadja, gndhra with an additional sruti may abandon the vivditva. If samvadi etc., were based on the sadja, then vaditva would also be incoherent, because vditva is the preeminence among the samvdi etc. Hence, excess and defect, do not destroy (text missing). On account of defect in the cause, a thin string though tuned still becomes too taut or toq slack. On account of humidity and dryness, the tying f the strings also becomes loose and acquires other faults m tuning. The beam or dqnda also Ijecomes bent or warped, and in the same way, the vocal singing may also be sometimes discordant. This is only an illustration of the defects which may occur in the instruments and the fingers etc. Thus, the notes have been described. Now, a bare note,may be useful in the empirical or transcendent context somewhere, but the notes are relevant'' in performance, only, as part of a group. Such a group of notes
* 'svarila bharii'.

54

ABHINAVABHARATI

TRANSLATION OF THE TEXT: NATYASASTRA AND ITS COMMENTARY

55

is called a grama. The grama is two-fold on account of the difference of pancama. GndKra-grma is thus rejected.'^ Madhyama bemg imperishable, preeminent and fixed, "and sadja being its samvdi is also prominent but pancama is not the samvdi of madhyama. Sometimes it has equal srutis (text missing). Acarya Dattila has given the reason for naming the gramas after sadja and mad/iyama, because the murcchanas the first, second etc., are obtained through the preeminence of the sadja.There are as many mrcchans as there are notes in the sadja and madhyama grmas jmd therefore, there are these two grmas.* But this is trivial, bause,designating the mrcchar^ as first etc., is of no use anywhere. It is merely a matter of traditional convention. In case their significance was to be based on countmg, there would be the defect of circularity.'^ 'Dependent here' in the text, (referring to the twenty two srutis) mens they are the causes through which the nature of tl^e grma is reached,
Verse-24 : "The rule (of sruti) in the adja grma should be three, two and foor, fourand three and two only and the commencing four". Verse 25-26 : should be of foursrutis,is considered as of three sruts and gndhra of two srutis, madhyama of four srutis and pancama should be similar (i.e., of four srutis), dhaivata is known as of three srutis, nida is of two srutis; such should be the rule in ihe fadja-grma". Prose passage between verse 26 and 27 : In the madhyama grma,

pancama should indeed be made lower by one sruti. In this way, the difference which occurs in pancama when it is raised or lowered by a sruti and when consequential slackness or tension (of strings) occours will mdicate a typical or standard {pranina) sruti. We shall expo^d the system of these {srutis). For instance, two vTns with strings, tying adjustments, beam and thesuccession of notes {mrcchand),all of similar measure and in the sadja grma should be made (ready). One of these shuld be tuned in the madhyama grma by lowering pancama (by one
* "literally, "The raimber of murcchanas is the same as of the nsXes sadja atid
madhyanixrLhst:wQ gramas."

sruti). The same {cala vn) by the raising of one ruti of pancama would be turned to the sadja grma. Thus would be known the diffemce of one ruti. If there is a further lowering on the {cala) wln so that its gndhra and nisda would correspond to the rsabh and dhaivata of the {acala) mm., there would be difference of two rutis. A further lowering' would lead to the correspondence of dhaivata and rsabha in one with pancama and sadja in the other owing to a difference of three rutis. Still further lowering would lead to the correspondence of pancama, madhyama and sadja in one with madhyama, gndhra and nisda, there bemg a difference of four rutis. By this dmonstration of rutis, the twenty-two rutis in the two grmas should be understood [dvau grmikyau should be dvaigrmikyah\. Commentary : It may be objected that owing to the difference of octave, ther should be sixty-six rutis, why are only twenty-two mentioned? The answer is, the gamut of notes consisting of seven notes, is* produced by.only twenty-two, that is why in another octave, the gamut of notes {svara-mandala)is mentioned as a cycle {cakrd)^^ This circle has been demonstrated by sages like Agirasa, Ksyapa etc. Sruti, indeed, means a distinct, new audible impression produced by a minimal sound. It may be objected that the ruti is a portion of tim&, because even \he caturutika (note) persists for a limited tune in the alamkras like bindu etc., and even the dvirutika when eniployed in the sthyl varna is used for much time, since like the sound of the bell and the conch it persists for much time. This is not so. We have already said that ruti is not a constituent or part of sound. Well, if tiiere-are twenty-tworutis in both (the grmas) then what is the distmction? With this in mind, the text mentions the order of rutis, 'in the two grmas.' ' VidhV is the division of rutis. Now 'he' (Bharata) mentions the employment {viniyoga) of rutis, "sadja will be of four rM/is etc.' That the names in the division are shown as well known. It is indicted that the dhaivata is never dropped in the sadja grama. Now, in s gandharva,crya Viskhila etc., have said that attention should be given to the last ruti which gives clarity to the note, or which
'antyayam " = ''antyyam'

56

ABHINAVABHARATI

TRANSIA TION OF THE TEXT : NA TYALASTRA AND ITS COMMENTARY

57

manifests the note. Thus "srutis belong to the interval of the svara" or ' 'which lead to another note" or "Belongmg to the svaras, and dependent on seven", or "Others belonging to the interval". But in gana,the purpose of this sub-division is what pleases the ear. Just as Mtrgupta hassaid "As the note decreases, the listener experiences within himself the order minutely. When it is being heard the musical sound is single nd charming". Hence, where gna is primarily intended, catussruti etc., is the common statement. But in Kn, they acc^t the gndharva paksa also. That to which belong three or four iratisby altOTiative support or jointly. [This appareny gives the meaning of trisruti md catussruti efymologically].^^ Now 'he' desi^bes the madhyama grma. Whai the pancamays of four srutis, madhyama and pancama are two samvadis of the sadja like two ministers. Although the madhyama is samvdi, the reasoij for calling it so is its being like samvit [covenant ?/consciousness ?]. When the pancama is deficient by one sruti, ien iere is only one samvdi of madhyama as also of sadya. Moreover (text missing). There isanothra (reason) for the predominance of madhyama. Without it being made(predominant), the esigaaon-madhyamgrma cannot be given regularly. Why should not several catussrutikas then-be predominant ? On account of fullness. Only two notes are such. The nature of pancama is to be catussrutika. The catussrutika nature of Kka and antara is an accident, not essential. Their permanence is similar in th two cases. Hence there are only two grms.^^ Now in the third sruti also of madhyama, (probably Aiadhyama grma) is produced a sound (jida) which hasthat kind of resonance and expression as would not be available in the fourth sruti of sadJH grma. Ii is lower only because the samsthna is lower. That is noticeable. From this, through the lower measure (apamnal) of sruti it becomes pancama. It has ben said that there is a tiremolo (kampa) of the srutL Thus has been said "the veda sruti is almost like a ripe kapittha, which is facing a slight breeze. Yatisruti appears as if trembling or the sruti is immersed in its own marvellous charm." [unfortunately, sruti and svdbhutarganisthah are not in concordance. TKe
'Krtayam' ?

meaning of ya/i-iruii is also not clear] Hence, the use of a/pa nisada and gndhra (text lost) and of kkaliand antara.Of tiie triruti note, there may be kampita, kuharita, or recita.^^ The vivaditva of ni and ga is owing to the continuation (anuvrtti ) of a different melodiousness (raktyantara). Thus the definition of madhyama grma is, that, here pancama is deficient by one sruti. This pancama is prminent and is never excluded from the madhyama grma. Others says that in the two grmas, dhaivata and pancama cannot be excluded because the movement or comprehension igati) of the madhyama depends on this. Now the question may be, what is this sruti"] The text: 'thus the lowering or raising by one sruti ' creates an interval by laxity or tighten^ ing. That measure is sruti. Laxity (mrdavet) means relaxation of the string. The opposite of that is tension. The raising (utcrsa) of the ruti, that is audible sound signifies higher pitch (tivrat),lowering (apakarsa) means lower pitch {mandata). On this account relaxation and tension, both have been mentioned as above. Thus, on account of relaxation and tension, which are the cause of higher and lower pitch, there is an interval which is perceived distinctively and is the measure or detemynant (pramana =niscayaka) of the ruti. That is to say, that measure, by which whether deCreasing.or increasing in terms of the accentuation or lowering of the pitch, a new sound, distinct from the earlier one is noticed, that is sruti. Although, the raising or lowering could also be infinitesimal (parmnutah), that distinction of sound cannot thereby be apprehended (i.e., by the ear). [Hence, that (minute) interval with which the pitch of the sound is raised or lowered and which is the minutest that can be apprehended by the ear that is the standard measure or (pramana)ruti.] Where the rise or fall of pitch cannot be perceived that is a constant sruti (eka sruti). The paramcrya has said "In addressing from afar, the pitch is at a constant ruti." Objection : In the vinas of gourd etc., the traces of the distinction of sruiis of that kind are not seen to be regular.^ Hence, why hasthe division of srutis been said to be regular, since that cannot be determined on the mm. The reply is negative. There, too, it is definite and it is false to speak of irregularity. As to the opinion that there is a confusion in the distances in which the beam of the mn is divided, it (remains a fact, that the sruti)

ss

ABHINAVABHRA

TRANSLATION OF THE TEXT: NATYAkASTRA AND ITS COMMENTARY

59

is noticed with regularity.There, also, Vmanahas (opined?) that the sruti is defined by a particular division marked by an external character {upalaksam) out of the mfinite divisions perceptible to man \purusa synecdochically] (text missmg). That may not be so, but what is the contradiction. It is only the noticeable difference that is accepted in the form ofirai/. That is why perception or graham is used asan adjective for the increaseof pitch (utkarsa).By the context, the lowering of the pitch is also intended. When there are two vTns, side by side and the string of one vTm is higher in pitch, then th^e is a higher pitch in its note, while the string of the other vim shows a lower pitch in its note. While the lower pitch of one is bemg noticed, tiie higho: pitch of tiie otiier vim becomes noticeable. Similarly, tiirugh relaxation also, the higher and lower pitches of the two vims may be shown and remove any doubt Both relaxation and tension are mentioned. Else,'he' should have spoken "tiirough tiie tension of higher pitch and relaxation". In speakmg of pranina it is shown tiiat sruti is not a minute part of tmie {lakala), nor is it a portion of sotind (hdmsa), nor is it duration (yu), nor a location (sthna) nor a mode of action (karana). Hoice, tiie meanmg is, tiiat sruti is a single sound which shows tiie distinctions of higher and lower (i.e., which belongs to an ascending or descending scale) as is to be described later on. Thus, having mentioned tiie nature of one sruti, tiie number of srutis is determined according to demtons. To show this, 'he' begins 'now their demonstration' (nidarsanam). By 'he' mpanc that tiife srutis are numerous. Demonsti-ation or nidarsana means, a metiiod by which tiiey can be perceived definitely. Thus, at first sight, tiie distinction between two charmmg notes {raktayaohT) with higher and lower pitches, becomes as clear as tiie distinction between tiie granine and the nongenuine, (or thedistinction between two notesof highand low pitch, when the notes are charming and naturally refined or otiierwise can be seoi at first sight) But, the desire fo see their mutual difference leads to attention and effort, and through its force, having perceived one, cme perceives anotiiCT, tiien the first, tiien again tiie second, and through tiie force of tiiis concentration there is a clear determination and conception of the difference as to \lvhich is to be called superior or higher (abhyadhika)F In the iJresrat instance also, when tiie two strings are higherand lower, whai

one of them is pulled, the, resultant sound is perceived. In order to determine the mutual difference of one sound fipm another, it is repeated, as the motion of a-swing through repeated pulling and the consequent force of attention the final conception that will arise will be clear and apprehend the peculiarity therein. This is called demonstration. Now there, the first tuning fixes the immovable (dhruv) vina properly, the second, the measure of the sruti, the third shows the four rutis, the fourth six srutis, the fifth twelve srutis. This is the meaning of the five tunings (sran). Now we explain thfe'text. Measare, (pramna) means length and breadth. Some say it refers to the similarity of the strings in number and thickness etc. Hence, the word catura-dahda{l) has been repeated. Hence, this analysis (yigraha) of.the compound of the two tias which have similar measure in the string, beam and mrcchari. In the case of murcchana, the standard number alone has to be considered. On account of the similarity of the beam and string etc., the'note is not differentiated in sometneasure, so that in both, the two may appear as if one. Notes have a disthictivccharacter. If any one characteristic (note) is being apprehended, then one vina is to be tuned there. This mri is not to be moved (dhruVakasthrilya = reference v75) and its tunmg is also not to be changed (acala sbrana). In another v7a, the string for pancama should be relaxed by one sruti, then it becomes madhyama grama. Later, all the notes madhyama etc., should be lowered by one sruti. Then it. becomes sadja grma. Thus, in the reference v75 and the movable v7/ja, when notes are sounded, the excess or defect of one cuti may be noticed by direct experience, witii tiie help of attention. Hence, tiie author says that the fifth sruti becomes lowpred -from all other notes, that is to say, "tht sruti is'directly perceptibler with its.peculiar quality. Altiiough, even witii an infinitesimal (patamnumtr,a) changp of location, there does occur a change in sound, which may be pb^erved by,an observation appropriate to the yogins. Nevertheless, the distinction which is desig-, nated by die word sruti and is demonstrated here, is, for tiie sake of the knowledge of the number of srutis and is obtained from four srutis. Again, 'similarly may lower'- In the movable \na, beginning from the pancama one should lower all the notes by one sruti. Thus, in the

60

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND irS COMMENTARY

61 -o

movable v7/i, gndhra and nisda are lowered to the same position as dhaivata and rsabha in the dhruva vTna because irt the reference v7n5 all the" nofes ar Kighr "by wo srutis, Thus by the reduction of dhaivata, {vailaksya = loss of position) there is an equation of gndhra and nisda with the dhaivata and rsabha of the other vina. This is called sruti. Hence, the usage about sruti, which is noticeable [laksyo might be lak^e, meaning, in laksya samglta, i.e., music as actually obtained in practice] that is, which may be seen in practice, is not without foundations. For example, while deficiency and excess being (text lost). Hene it is called n/ on accunt of the special employment. It is not so in the earlier ones (?) Thus is the nature of srutis known. The two srutis which belong to each .of the two notes - gndhra and nisda are clearly perceived by recognizing their mixture with dhaivata and rsabha. Hence, the srutis become directly perceptible. Then again, when there is a lowering of one sruti of all-notes in the movable vTn, then the dhaivata and rsabha of this movable vin become equated with pancama and sadja respectively of the dhruva v7/i, because in the dhruva v7n5 the notes are higher by three srutis. Thus, in each,of the two notes, three srutis are clearly obtained and as a result one obtains six srutis. Three (notes) obtained by the reduction of all notes by one sruti [dhruvannym should be cakmr^am or it may be construed as meaning, with reference to the dhruva v7n5 which would not be very grammatical] namely,;7acma, madhyama and sadja will be respectively similar to madhyamd, gndhra ar^ nisda of the i//iruvv7/i, because of that vin airthe notfes are higher by four srutis. Thus, in each of the three notes, four srutis are deifisfrated, andhnce twelve srutis are perceived. Thus, four, six and twelve together make twenty-two (srutis)** as may be seen. Ga ni are two srutis higher thn rsabha and dhaivata. Hence, lowering by twcy srutis, will tend to their overiapping. Ri amd dha are two sruti more than sd and pa. Dha (apparently sa) ma, and pa are four srutis more than ni, ga ma. Hence, of those srutis (text missing) should be introducerf into the next note. By saying that,'that is more than two iruiw', three
* * ''punvdisu naivamiti "upalabhyo ' is obviously 'upalambho

reasons for introducing special notes is explained.^As 'he' s^ys "bS' lbis demonstration of srutis' (text missing).^ This Is d^^srutika and, thus, the nature of sruti is obtained* Hence, 'two in two' means th^^^are in the three srutis. Thus, the measure of the sruti in -(he dhruv vin is oq clarified. The forms pf two sruti notes, three sruti notes, four srutis notes and the measure of a sruti become clear by the chart where alone tney 'can be clearly perceived. Some show it as a straight line graph ({aAaa prastara) of twenty two lines. Other use a circular graph (mandaidprastara). This consists of five horizontal and six vertical lines and counting both ends (of each lines) they together make twenty tw'pints.^

' _ J } 'V ! 'l Verses 27-28 : In the begining should be Uttaramandra^ iii&n Rajarii and Uttaryat, the fourth is Suddhasadj and the-fifth Matsanki;tit, the sixth is Asvqkranta and the seventh Abhirudgata.Thts^ seven mrcchans should be known as belonging to ie s'^^a-rm^.
There are fourteen murcchanas belonging to the two grariiasOf these (mrcchans) sadja, nisda, dhaivata, pcana', mqdhyma, gndhra, rsabha are the iiitial notes in due order. ; ^ fibio

__

Verses 29-30 : The Uttaramandr (murcchana should haveits.ijt^l note) in $adja, the Abhirudgata in rsabha, the AsvkrniintrKqi^ has its initial note) indeed ingndhra,the Matsankfi mtnqdhyaimff^p / Suddhasadj should have its intial note) in pancama, the Vtiaray{i^,ix^ dhaivata and the Rcgam (should hav it) in n^adia^ Jheseja^Cithe mrcchans of the ^adja-gmma. Now (the nmrcchaf^\ n p^g^^^ma'

vsc\,i^ri!i;r.voi

Verse-31 : Sauvln,Harinsv sind Kalopana^,uddjha^f^dfiy as well as Mrgi, Pauran, and Hr$yak; these seven ni^^ch^pas ^iuf^li be known be as belonging to the madhyama-gr^q. , ^ La Prose passage between verses 31-32 1h^SQ\t}\fldliyflifia.,

gndhra, rsabha, sadja,nisda, dhaivata,pat(!ama,ShQ'xt^Xi9t^k^ in due order. The Sauvlfl (mrcchari begins)'.!with i(nEt.), Haritisv with gndhra, Kalopanai (begins^ 'mthvfflhha,. Sud; dhamndhyani (mrcchari begins) with sddjafiiM&r^m^vnp^, Pauravi with dhaivata and Hrsyak (begins) with pacama. Thus, these fourteen mrccharis with an orderly succession-of-notes-afe.tfa/;?r/a
. .IX. . . ^ .7 % 1 '^ . t II Ir

62

ABHINAVABBARAJ

TRANSIATION OF THE TEXT : NATYAkiiSTRA AND ITS COMMENTARY

63

or complete (i.e.,with sevennotes), renderedhexatonic or pratatonic(i.e., with six or five notes), rendered with the (two) auxiliary notes (i.e., antara gndhra and MkaU-nisadd). And also Verses-32 : A sequence of seven notes is known as a mrcchan. Tlje ones rendered as hexatonic or pentatonic with six or five notes are tanas. And the ones with sdKrai}a notes which are adorned by the kka {two sruti nida) and endowed with antara svara (two sruti gndhra) are mrcchans of the two grmas. Commentary ; Thus, the two gramas, and the measure of the rules governing the srutis ia the svaras as well as the evidence fOT the existoice of the sruti which are usefiil to the two grmas have beem stated, and as mrcchan follows next, it is now decribed. The essential form of mrcchari consists of sevennotes. These are indicatd by th orderly and successive descent from above which is being listened to.'' Murcchana (is derived) xjm a root which is listed in the sense of rismg up or swellmg.'^ Hence, the author will later on specify by saying in sum, ordered notes (are murcchaiiS). This means, that thare will be no mrcchan'v/ben there is no order. Thus, from sadja the.ascMit up to rtisad,^^ from dhaivata up to pcmcama, from pcmcama up to madhyama, from ihadhyama up to gndhra, from gandhara up to rsabha from rsabha up to sadja. Sadja is at the mterval of thevsecofld sptaka, touched by the higher octave [how it touches the,fora saptaka, is not clear]. Objection ; There are then seven mrcchars. Answer - True, by ^ lowering pcincama by one sruti we get the first seven murcchanas. Hence 'he' says,that, there are fourteai mrcchahs belonging to the two grmas^* By the lowering of the pancama, the system of samvadi and anuvdi, as well as the systems of notes to he excluded or retained bcome quite different, where by, much diffemce is caused. There, from madhyama to ga, from ga to rsabhaffrom rsabha to sa, from sa to nisda, from nisda to dha,from dha topa,pa to na [obviously ma].These are seven. Here the mention of names Ls a part of the Vedic ritual, so it has been shown. Thus, it is heard in the Vedic texts, he should sing three

self composed gths* by Uttaramandr** "T^os& wives \patyo should be patnyo] will sing to you by palikas. There areseven in the madhyama grma. Foreach one of them, there are four varieties of the nmrcchahs. There are sevennots inthe sompSrna form, six in the ^dava Sdava may be draiyed, thus, w^CTe ihe performance is favoured by six notes.The performance produced from those sk notes is called sadava. By the use of five notes there is auduvika. Udu means star. Where they move, that is, uduva or sky. Haice the numbCT five.-From them are known sevMi. Because the fifth is maAfc/iia , hence thatnumber. On their account, thenotes arecalled auduvita.Where they are used, that pCTformance is auduvita. By its conjxmctio, the mrcchan is also that, or auduvi is the number of those nine. What has that numba: (is auduvita.) The fifth, kia m the middle is caUed uduva. Thus, also is the usage found. Perhaps, these are in the auduva (or they have attamed ormoved into theauduva) This may also be anetymology if the elision of a vowel {svaralopa) along with the openness of a part are also accepted. For this reason, appearing as identical, it is called mrcchari. But with the conjunction of the kkali and antara notes, which use the nida and gndhra sparingly, it (i.e., the mrcchari) is tMmed sdhrana. To summarise tto, two t^kas are used here. Verse 32-33 : "The seven notes in order are called mrcchan. The &nas are of six and five notes based on 4ava and au^avika." "The mrcchahs of the two grmas, adorned by kakaB and containing antara svaras, are sdhraiidqis." Comment : Seven means complete. Kramayuktatassss where there is ah order, an ascent or descent of notes. Tna means, that by which the of performance are elaborated and given a charming diversity.

* **

misprinted 'gadha'. cf atapatha Brhmam 13.4. 3. 5'uttaram andrm uJghmntisrah svayam

sambhrth gtK gyati."

*** This text is not traceable, though the word PtuUtka occurs in the Poniniya
Gampatha. **** \ipcSica is read for pancama better sense would be obtained "As there are

"lop^ lopysvara-vyavasth^'* should be"lopyalopyasvaravyavasthif

five mahbhs".

64

ABHINAVABHARA 77

TRANSLATION OF THE TEXT : NAJYAHASTRA AND ITS COMMENTARY

65

_ Having thus collected three fornis. 'he* mentions the fourth form as s^hara^as. Those^oJ/iara^aredifferentfromthenormal (vikrta) but boundless (manta). He (i.e., Bharata) defines''adorned with kkali,' otncdmihantarasvara' and concludes by ' mrccharis of ie grmas'.'
Prose Passage between verses 33-34 : A murcchana can be ac-

comphshed in two ways; there in thesafygrma,^rfAra having been merged m dhaivata by the raising by two srutis, there is a change ih the murcchana sngrma.Because of this, madhyama etc., (madhyama etc., notes of the sa^a-grama) according to serial order obtain nisdatva. In the same way, m madhyama gram, due to relaxation of dhaivata [of two srutis It becomes dvirutika gndhra], thore is a two-fold differratiation Due to there being an mterval of the same number of srutis, there is a difference of names. In the madhyama-grama thereisa differrace of four srutis pancama and dhaivata. Due to that, gndhra bemg raised (by two srutis), it becomes of four srutis. And the lemainmg also madhyama,pancama,dhaivata,nisada, sadja and rsabha (ofih&sadjagrm) due to equal sm/mtervals become nisda, sadja, rsabha, gndhra, madhyama and pancama (of madhyama grma). The exposition of th antara (antara gandhara and kkali nisda) has also been spoken of in ruti demonstration. De^ndent on the mrcchan, there are eighty four nas. The hexatonic (tanas) are forty-nine and pentatonic are thirty-five. Of the hexatomc, flwre is a seven-fold variety. Four are devid of sadja, rsabha, hisada OTpancama in the sadja grma. hi themadhyama-grma thereare toee devoid of sadja, rsabha or gandhara. hi this way, these hexatonics bemg^worked m aU the murcchanas (of both the gramas) wl give rise to forty-nme tanas. Of thepentatonic, they are five types only. Characteristic of the sadja-grama, devoid of sadja-pacama, devoid of rsabhapancama, and^ devoid of nisada-gandhara, thus three (types) In the m^hyama-grarna, devoid of gndhra-nisda, devoid of rsabhadhaivata^us two (types), hi this way, these pentatonics being worked in all die murcchanas (of the two gramas) there are (a total of) thkty-five (pentatonic tanas). There are two ways of working the tnas m the string (mstmments) - by pravea and by nigraha. The pravea (or assimilation) is

made, by sharpening or raismg the preceding [adhara, not madhurd] (note), or softemng or lowering the succeeding (uttara). Nigraha is non-touchmg (i.e., omitting the note). The mdication of the mrcchari is done by the madhyama note (of the vTn) because of the mdestnictible (aria ) or constant nature of madhyama. Hence the nigraha and paryagraha (do not relate) to madhyama. The variety of tnas and mrccharis, thus (arising), provides entertainment to the listener (i.e., the audience) as well as the performer. The use of the mrcchari and tna is also for the attainment of the voice-register (sthna-prpti). There, voiceregisters are three-fold. Their description h^ been given m the rules regardmg the intonation (kaku). Commentary : Now for instructing the lute player (vainika), the author says, the mrcchari is accomplished m two ways.^^ Mrcchari, here, refers to the two formsof sadja-grmiki mrcchari and similarly of madhyama grmiki.By diverse raising and lowering of a note unusual mrccharis can be obtained in the madhyama grma as in the sadjagrama. How this may be done is shown, when, in the sadja grma the gndhra is raised by two rutis,that is, when the gndhra string is raised by two rutis, it becomes the caturutika dhaivata of the, madhyama grma.The notes pa, ma, dha, ni, sa become the notes sa, ma, ga. If madhyama becomes dvirutika, then it becomes nisda, pancama becomes rsabha bemg of three rutis. In gndtmra it becomes rsabha. The notes become nut, pa, sa (probably dha), ni, sa, ri, sa (probably gal) The reason, here, given is because of equal ruti intervals. There is only a difference of name. The difference is of names in thesvara and thegrama, but the equidistance of the nti intervals is then clarified. (The tnas) bemg depradent on the mrccharis are particular states of the mrcchan. Now, if we take the seven mrcchans of six notes each, by deducting from them, one of these foufsa, ri, ni pa, then we shall have twenty-eight tnas. By deducting one of the three sa, ri, ga, we have twenty-one madhya mrccharis (i.e., madhyagrmiki mrccharis). In all, we have forty-nine sdava tnas. The audavita tnas are twenty one in the sadja grma and fourteen elsewhere, being together thirty five. Taken together we have eighty four.^'In the vm, when there istnakriy, then, the string is used by avoiding or skipping the note. If, there is

66

ABHINAVABHARA.37

TJXA^S'T F 'THE TEXT : NAJYA&STRA AND ITS COMMENTARY

67

bsaymglhattoeamifc^of,S^^/^"*^ "
Tnakriy means activity or method tnr fnr^ ^ pared to rsabha the otfi^r p l (producmg a) tana. Comtuming into raaL. That LSI ^ into nisda. Whichever is the IfroneCT be included and dissolved. Messing nieans Uttaramadr, changed should

eliminai. Hrc sadja has to be'\he Sfecond, fourth note from madhya (middle twofoith madhya-dvicaturtHa).No tna can be obtained. Sohas Dattila said "^voi after&na has beiso comprehended, theexpert would count the avimSi, i.e., the pexmanoit note, and detemine that so many are the mrcchans.' t Objection : In the first and seventh murcchanas, [apparently of the sadja gramas if] sadja is eliminated, then ri, ga, ma pa dha, ni.as th (resultant) fonn (common to both the 1st and 7th mrcchans) is difficult to know sqparaiely. True, there is no diffo-ence in writing, but it is not so (in practice). Thus, aft- the mrcchari sa, ri, ga, [ma should be added hitte] pa, dha, m, the nmrcchah will be ni, sa, ri, ga, ma. pa, dha, ni [ni should be omitted here] is constructed beginnmg with the earlier nisd, but the cannot be the reentry of the note aft- the octave. Then, sadja being eliminated, thae is no distinction of pratham and saptanii. In that case, the ordedsuccession mentionedby Dattilcrya will not be carried out Besides, tte tmcchans will become largely tnas. But when after usmg sa, ri, go, ma, pa, dha, ni,the second mrcchan begins with nisda in the mandrastqttaka below the sadja as ni, sa, ri, ga ma, pa, dha, thrai from dhmvata below tusada, then from pcmcama below that, till one reaches ^abha; then mandra saptaka being successively transited or the middle saptaka being deleted, the notes sadja, rsabha etc., upto nisada having been heard inthe tatidrasaptaka, it is known as the seventh lupt mrcchan. Iq the six notes belonging to the middle octave, the first in ordCT, beginning from the tra saptaka and descending in the order of.... (missing) whoi this is-the ordear sa, ga, ri, ma, pa, ni (?) [seems to be a mistake for so, ri, ga, ijia.pa, dha, ni]. Havmg used it, below sadja is the place of nisda,dhivgta, pcicama, madhyama, gndhra, rsabha in this order. Then althou|jh, anotheroctave is not touched, still, at the fjxd place of sadja from the place above it, rsabha having been used and ri, go, ma, pa, dha, ni, having beai heard, the first mrcchan withjthe lupt sad is well known. When the place of sadja has been occupied by rsabha, then the possibility of sadja below that does not exist, ri, ga, ma,'pa, dha, ni, havingeeai heard, one knows that tra sadja has to be here and since it is not being heard, this seventh is the lupm sadjq mrcchari.^* Thus, in both the alternatives, the attinment f the place determined by succs-

"e
Madhyama, here, is used for th "madhyama svara.

of

.e wd..., oc.,

paryagraha. By that, some intend

^ ^akriya^ and taraknya is

the kriy (production of notes) is possiWe'^r^l"^'"'^ octavenot otherwise TMpr/jA ^ ^oce to the middle no. obutai.^ Za^ T^."^''''''^- C-"-

U M

" " 'o-tio-, W m

"av.. That is what I ^t. articulated by the throat is the miHHi throughtheeliminationofc^/i then how canTe LI fe f t " ^at ismore, when -the,.^y.^.ma

obiecUo^ a.e ., sa..

68

^ABHlNAVAHAJf^77

TRANSIATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

69

sive order is quite clear. Nigraha inimadhyama grama is attanw^ from diminishing of the pancama by one sruti. Thus 'rsabha hy,pancama" [= Thus witfrthis/^aicama there would be paryagrahojui of rsabhdi}^^ Where there be paryagraham on accouqt of lack of samrda, that paryagraha will be (of ?) sa^'a grama. There and at both placs^wjorfhyamasvara is avimslor mdestnictible. Proceedmg from the madhyama, parwama is reached, but in nigraha in madhyama grama (one iruti of pancama) is lost hi this way (having obtamed to) paryagraha in mqdhyama grama in the sadja grma it is (referring to pancama) neither toq low, nor too high and is indestructible [i.e., pa is stable in sa^a grma and does not lose a sruti\. The movonent from madhyama (topacama) is three srutis (m the madhyama grma). By stoppmg (th^e) it (i.e., one sruti) is lost. Objection : There may be mdestructbility of rsabha also. Hence, he' says during nigraha (when ^dja is omitted), smce nida has to be taken, that which is of three irutis is dhaivata. As has been said by Dattilacya.

The answer is (no) because it is sen that in;a/is,'the notes (of the

mrccharis) other than those mentioned in the tnas are not dropped. Thus; sadja madhyam is a jti in the sadjagrma. It has a s4ava (form) with gndMra.^^ Now in the tnas of sadja grma, gndhra has not been counted (for droppmg)'because only sa, ri, ni, pa are mentioned as suitable for being dropped. Since the tnas have been enumerated, then why has it' been said that pacama cannot be dropped in madhyama grma, dhmvata in the sadja grma, and madhyama in both. Now, the ktatnas have beai shownearlier to be fiVe thousand and thirty three. Their prastara (pennutational expansicm), nasta and uddista [nosia is finding the tna asthe n'th permutation, where 'n' is specified; addica is finding the value of 'n' where the tna is given], and samkhy (total
number, of permutations) have been maitioned by the Masters in detail. For knowing the na^a and wddtoa and the ntimber, the formula has been described. Now the verses for the knowledge of the nasta and uddista. First about uS^a "Beginning with the last place in the origmal order, as many places should be left out as areoccupied by each relatively preceding (elanent). The places below those occupied by the original numbers placed above**, are indicated again, and its penultimate etc., should be declared. Thus should be known the number of uddista. For the nasta, it is known by the figure placed above the original first number. It is as many as are foun^ m ^e lower place, if there is only one braket. If there are many brackets thai to know the samkhy These rules are not shown here, the Sage assuming them has made a general statement thajt thq variety of tnas and mrccharis is for pleasmg the audience and this performa. As to the objection- where is the use of * 'losta' appears a mistake for kostha.

Pancama in the madhy-grmaand dhaivata in the sadja grma afe to be regarded as imperishable, and madhyama is everywhwe regarded as imperishable."^''
Objection : How can ri,being a consonant be the imitation of svam? Tentative counter objection- well, what is the form of sa etc.? Objector - that is not the mtial of raaAa. Counter objection - then sa should not become the imtial o sadja ? Answer. It is merely a symbol (samketanitram), whether h is ri or r, there is no fault either way. Others say that the view of Dattila etc., is not accepted by the Sage. They say, that for 'him' the madhyama alone is not to be eliminated and hence it is called madhyama. Objection : Well mrccharis are not like jati, graha* and bhs, useftil b performance, but tnas are useful in the orchestra.
The text has jajjati which is apparently a for

** 'paristhimh' - uparisthitaih ?

jH.

*** Those versesformulate some Combinatorialfonula whicli cannot be made out since the verses are not free ftom conuptioa The formula is clearly knowij for metrics vide Vmaramkara. Vide uuiva-Vsam, pp. 101-10, Darbhanga, 1959.

70

ABHINAVABHARATI

TRANSLATION OF THE TEXT: NATYASASTRA AND TTS COMMENTARY

71

murcchanas and tanas, it has been saidihat the fonn of the tna is for the sake of the mrcchari* As to the statement, h is for pleasing the poformers, the idea is that the listeners know the tradition. This is being said, although m this tradition thereis no use of thenmrccharis,(i.e., they are not sung per se), even so, they have a clear use m the Sman (perhaps as musical scale]. So it has ben shown. "He sings three songs by Uttaramandra.' ' Thus of thetnas (which are, agam, not especially usefid m the sarlrJ vim). "The first would belong to agnistoma, the second to Vajapeyika". By mentionmg these names, the use of vases of Sman and useful in sacrifices has been shown in the VSy Purana etc. By the agni^omic sman, Siva is praised, and thereby the paformer gets the adma result. This is heard m e Vedas. Thus " hi the sacrificial assembly, havmg heard the agnistoma sman,one is freed from grave sms and wms' the worid". "One who recites the oration of Daksa by the Suddha-sadja as well as one who listens to it everyday, morning and evening, both of them will go to the world of Rudra." (i.e.,tieuse omrcchis in Sman smgmg was probably left as legacy to gandharva). Thus, the use has been shown, whereby, thej)erformersseciuedthehappmessofa<i^. Through the use of the rks,gaths and the soman, Nraciya SUim and the Pur^ have been demonstrated the sphere of notes upto the tnas only. The variety of the mrcHans has been shown by subsidiary rules, o that the audience may be "pleased and thus to this end the performance may be jomed (by the performer). This is beingsaid m the Jtis (viz., m gandharva) only a fixed number dava and auduvika (tanas) are enumaated. (But in gana) is seen thef free dropping of notes except madhyama (i.e., m Ae rendering of tnas) in order that the wonderous variety-of rga and bhasa which pleases the ear may b (established).** As to the objection - why are the litairias not mentinned, the fact of their being fdita i.e., being in irregular order is not in itself of any importance. They have the general characteristic of giving pleasure. Only the rle of the enumerated eighty four (i.e., the regular 84 tnas of gndharva) have the fixed function of pleasing particular ifevatSs. They may not thus have a
* Th Parimala ed, reads nafor ca.The iSutose of tofta would thai be ottertaimnent. Vnmulayitum is an evident mistalce.

use, (referring to iaita tna which are not used for pleasijig gods) but m a general way their use is certam. As to why they have not been counte^d, the answer is, how are they to be counted when they are infinite ? As has been said, 'Infinite is the diversity of compositions in music' [GOS ed. places it withm Bharata's text, which is improbable. The source of the quotation is not traceable to Mgha II.3, as mentioned by the editor]. By mutually mixing the full notes and sdava,auduvika etc., even Brhaspati canot enumerate (so many of them). Perhaps, then what is the purpose ? No, when the performer is proper (i.e., skilled), even that produces pleasure for the audience. That this is correct is shown by saying that the purpose of mrcchari and tna (is the attainment of {sitions) (sthnaprptyartham). If mrcchari and the iayhaymg been taken out from the rk, gths and soman are arranged properly for the sake of attainmg these positions (stHanas) which giv special charm (text missmg) in the movement m the sadja sthiha if there is freedom in articulation, the employment of the notes becomes particularly charming to the audience. The performers, also hearing it feel pleased. By attaining the character of ruti (text missmg) There is a proper culmination in the grma, rga, bhs etc. When the notes of the full mrcchari are rendered with ease in a successive or non successive order (i.e., laitatna order), musical relish follows at once. When it is not full thereis avoidance m the rendering of rga and tarn. In each grma rga etc., all the mrcchans and fitatnas may be used in many and diverse ways. It follows that there is no limitation This has been said - for the sake of attaining the sthnas. Then what is this sthna ? The answer is, sthna is the triple (octave) as mentioned earher quite clearly. The mention of the triplicity through the prescription of kku is the same as the sixty-six divisions.Briefly, thevarna, amga and alrhkra are used m three ways.
Text : Now to narrate (the rules of) the sadhra. Sdhraita means ^e interval of notes, How ? the difference which exists between two (the overlapping), that is sdhratfa. For instanceVerse-34 : "In the shade one feels cold, but sweats staying under the sun; neither has spring come and nor is winter(flly) over". Thus (is) overlapping in time or the transition of a season (kla

72

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NAJYAkASTRA AND ITS COMMENTARY

73

sdKraifata). There are two types of sadharai}a,ti-sadharatfa and svar sdhraa. Svar^ sadharana are the kakaU and antara notes. There the nidada which is raised by two srutis is known as Kkai. In the same way (i.e., by the raising of two srutis), gndhra is known as antara-svara. That collection of jatis which have the same amsa and are midifferentiated (i.e., similar) is known as jti sdhrana. The characteristic is known according to each ama. Svara sadharana is of two types and of both the grmas. How ? Sdhrai}a here (in this context) is (the name of) a particular or special svara or note, because of this it is called a4ja sdhrana {asdhra^a appears a mistake for sdhran[. In the same way in madhyama also {madhyama sdhraifa in madhyama-grma). Because of the subtle nature of itsemployment, a second name ^katika\ thus, has sprung up. Because it is sweet and indistinct, it is termed as kka, because it is distorted (i.e., displaced from its proper sruti) it cannot be an amsa, according to the sayings of reliable people or sources, it is not different from the seven (notes) but is nida only. Just as the pungent flavour is known as one amongst the six rasas or flavours in the same way is nifda termed as kka, and gndhra is known as antara svara. And here are two (verses). Verse-35 : The antara svara should always be used in conjunction with the rof (ascending vora), never with the avaroKi or descending. That employment too, should be particularly limited." Verse-36 ; "If the descending or avaroM (varna) is being used in a limited or proUflc (way), then the antara svara destroys the essence of the/at as well as sruti." Commentaiy ; Thus having explained the appropriate positions of the mrcchans and tnas, the author proceeds to explain the fourth division called sdhranakrta. Produced in the interval is antara,moved from its own position and on to another position. It is still charmmg and not discordant (visvara). That condition is sadharana. Its nature is sdhranya. Thus.yii sdhram notes are those which are i^oduced in
* " Basically nothing but a distorted form of saline or sour.

these intervals and arecommon to thesubdivision of the j'tis. This is their general nature. 'Why' i.e., is there ^y empirical or popular example ? Hence the answer, just as "sitting in the shade one feels cold,but sweats in the sun, neither hasspring come, nor winter ended. This is (the idea of) sdhrana as a period of time. "The idea is that, since one feels cold sitting in the shade, this shows that spring has fully come*. That ,there is perspiration in the sun shows that there is a lapse from winter time. This is what is said here. What is excluded is not having come. Perfection or fullness is not asserted. Together this indicates that it is time in between the two meanings of being destroyed or being full. Such a time is sdhrana. Similarly, the svara sdhram is midway between two

notes.''"
This division is shown by saying there are two sdhranas : Jti sdhram and svara-sdhrana. Hence when some say that there is a tna-ruti sdhram, that carmot be accepted. In the sadja grma when there is sdava and auduyjita, nisda is dropped. Gndhra is dropped in the auduvita. The opposite takes place m the madhyatm grma smce, by not depending on its peculiarities, the principle of omission operates. If there was tna sdhrana, then this would not mean anything There is no such thing as ruti sdhram, because apart from nisda and gndhra, no other (note) is accentuated in ruti from the condition of droppmg. He (that is Bharata) explains the first, 'there the sdhraim'. He defines them m order. Kka is the name of the nisda which is raised by two rutis.He mentions the difference in designation. The differentnamp; are (only) for practice. In reality, both are of the nature of kkahecavse of being sharper. As 'he' would say "the name kka is given on account of being sweet." Both are antara svaras. As has been said "Sdhram means being an antara svara"."' Now 'he' defines the other (that is) Jtisdhraria. Those j'tis which have the same arna employed without distmction and in which the inner path(aniaramarga) characterised by the repetition (anulapam) of vdyama, is alike (thatisjti sdhrarm). Now, there may b a doubt depending on what is this jti sdhrarjal It
* Periiaps the opposite is meant that the spring has not fully come. PHr^tva should perhaps be apr^tva.

74

ABHINAVABHABATI

TRANSLATION OF THE TEXT : NATYA^STRA AND ITS COMMENTARY

75

is not that there is an identity of the twoyaiis m that portion.* Th answer is by the coinherence of the tis and the recognition of the characteristics relating to the otttscis. The common set of the ten features such as gruhu, nysa etc., furnish the reason on account of which even when the amsa of the tis is not distinct or different [ainse' vicitre should be amse avicitre], their distinct characteristics can be jmown.''^ As 'he' says "Nysa, and antara mrga* serv to distinguish.''^ 'Tu* here shows difference. The idea is that there is no identity on that account. Some explain that ti sdhram is the knowledge of the characteristics of the parts called graha etc.""^ This is explained in the section "Svara sdhram is two fold" etc. Heallows the use of kakaU and antara mboth the grmas and also prescribes a new name, where nisda and gndhra are used sparingly. The two notes in the sadja grama, sadja sadharana and in madhyama grama, madhyama sadharana. This appears trivial, since it has already been said that mrcchan is produced by svara sdhrana, what more is said by this ? Others, in order to show the regulations in the two gramas say, in the sadja grama, nisada is kakaU, because it makes the sadava. In the auvita, gandhara is antara svara. Hence, thereis sadjasadharana.In madhyama gramait isthe opposite. By this also, something is intoided. As tosaymg to whomkaisika belongs, now, if itbelongs to Ickaliand antara, even thai, on its belongmg to nisda and gndhra in the satva and auduvita forms in distinct gramas according to the proper way, such (distinction) isnot seen in case of either the Sadjakaisila with all the notes or KaisiBwhich has been made dava and auduvita by dropping rsabha and dhaivata-}^^ Even of Suddha and Bhinna Gaudamlava and Sadja Kaisila, the production from Kaisila ti (textmissing) the conventionshave been demonstrated by the sage Kasyapa etc. Therefore, this is the way in which my teacher Utpala-devapada thinks. Now, here, three", two, four, this is the order in which the natureof characteristic positions has been mentioned (missing) The ultimaie natural form has beai mentioned. That sadja is catussrutika, and so on upto to nisada as dvisrutika,(is the representation
* 'bhgo' would be a mistake for 'bhge'. Else the translatoa would be. "thus there is no division(bhgo) there between the jatts but only a unique and distinctive identity,

of thek natural ultimate forms) just as there is a form of living'beings which have aU their limbs in full, and, they may have ano&er fofm which is not so. Although, there are two forms, there is a pure form, just a the dog may have form without a tail or two tails (i.e., a viibt0,'(shfailarl>0 'h account of the discordance (vivaditva) of gndhra and ni^da whn they are commenced, vikni is mentioned. By their vikrti (derangement) sadja and madhyama are also altered. The derangement of pancama is already shown by mentioning it as trisrutika. In this way one can speak of the vikrti of dhaivata and rsabha also.'" Now, 'he' demonstrates this"Svara

sdharam is two-fold, belonging to the two grmas,'. The characterisation given of kakaU and antara in the two grmas and called svara sdhrana is of two kinds. He (Bharata) asks here, why? If the two kinds are due tothe differenceof Kkali and antara,then they have aheady been
mentioned, why repeat this? If it is for another reason, thai that it should be mentioned. There is another form and 'he' mentions that "Svara sdhrana here is a specific note and for this reason is called sadja sdhrana." When nisda is raised by one sruti and rsabha also, then nisada becomes trisrutika, sadja dvirutika, but rsabha is caturutka. Then the sadja on which depend the two (notes) nisda and rsabha, becomes sadja sdhrana. When gandhara is raised by one ruti and pancama of the madhyama grama takes on oneruti of madhyama,but dhaivata is of owinitis, there is a loss (of rutis) belonging to the madhyama grma. Then, since, madhyama is the central note and depended upon (by ga and pa), we now have (madhyama as) madhyama sdhrana. ' .The objection raised is, that, nisada and gndhra have been said to be sadhardm when raised by two rutis. Hpw, then, is the raising by one sruti mentioned now?. The answer is, that the form of the note is quite distinct from its earher form, that is used in svara sdhrana. By this, it is shown that the peculiarity and strngeness of notes, owing tohigher and lower pitch, have been regulated in the gndtiarva.^^^ For empirical piuposes (drsta siddhi), the one ruti character of notes has been demonstrated. Such varied peculiarities are. plainly seen in the lak^a music of raga, bhasa etc. Thus the elder Kasyapa says-

not bhga.

76

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

77

"In the rga, bhss, cme may use in all ways the no^of four, three, two or one srutis, by the addition of kka and antara. The sevrai notes

sadja etc., and the raised pancama and further, separate from dhaivata, the notes called kka and antara and sadja, madhyama and gndhra, these four are to be used in a ways in the tis. Thus fifteen notes in all
are to be used."'^'' In tending to show its impossibility in practice (periiaps asambhava should be sambhava), 'he' givesthe definition"It is called kaisika because of the rarity or subtlety of usage". Tu is used for determination, that is to say, of this only, oi sadja, madhyama sdharam. It can be achieved only through subtle and marvellous skill. Kaisika means the use of kaisila form, where the performance is dfelicate and charming. Or, kaisika is derived from the root kas.Or, from the subtlety of the point of a pata (?). This is a separate effort, that Sadja KaisiE is produced from Sadja

Sruti sdhrana is the fact that all svaras* have an indefinite number of srutis. The objection is, why is nisda called :fca/r ? The answer is, because it is sweet A little sweet (here) means sharpness. That which has ** T I. it, is called kakali or the T is on account of the gauraai-group. In the reading named l^ka-satijnaka, nisda itself has almost reached sadja, like a dog without a tail, hence by being vikna it does not attain prdhnya and hence is not ama. The objection is, let it not be out of the seven, but anothra- nisda which is vikrta and that is not sadja. What is the reason? The answer is "Due to the instruction of those who know" ca indicates the possibiUty of another reason. Originally, when the form of nisda is' apprehended in tie mind, it is determined as such and is in the first place called as such. Here, 'he' (Bharata) gives an illustration, that the taste of salt is also when sharpened acidic. It is said by Canakya (?) etc., to be madhura rasa (!}*** In moving from madhyama to rsabha, it is used in the interval andhence called antqra svara. Sdhrana having been described, therest are now mentioned. In the, sdlmraiia, when aftra resting on the madhyama, which js the ascending^ note roh) from tte antara svara,if wishing to move to rsabha one uses the antara svara, then it is p-opraly joined in iruti. Its employment is
always for a limited function. (It is) never (to be used) as the first in; descent Hence, when the antara svara has berai used rsabha is not-

Gndhri. Since it depends upon a portion of the madhyama grma, dhaivata becomes catussrutika. The occurrence of catussrutika rsabha and dhaivata should be weak. Nisada and gndhra have to be as usual. It will be stated later on "dhaivata and rsabha should be weakened. "Again, the pancama of the sadja grma has to be of four srutis. In the use of the kaisika, through non-conjunction with the movement of the sadja sdhrana, sadja kaisiE and kaisiE also belong to madhyama grma. Hence, dhaivata is of four srutis. ^abha, also is of four srutis. The weakness of rsabha here will be mentioned. >\^ere rsabha and dhaivata are held fit to be dropped, the reference to them is of four srutis.^^^ All the jtis,e\cep.DhaivaandArsablaie'm\hoiA^dhram. Hence, owing to the introduction of sadja grma and the requirement of having four srutis, pancama bas to be (givrai) there. Hence, sadja (and) madhyama sdhrana have the alternate name of kaisika and have their own place. By generalising, this applies even to the grma rgas like Kaisika, Bhinna-Kaisika etc. This is what has been called by crya
Malanga as gfma-sdharaiux.

dvarohi***^
That is not to be employed. When a madhyania is used which is joiid to -an antara svara^ ien ascenfalone has to be made. Having used the antara svara aftra madhyama, thrai to madhylima of kewhre, (My a^et is permissible. Nv^ 'Be* (the author) strengthens what had been said, namly, onlyascent, tevra descrait ("If th antara svara is used as a descraiding. note, \diethCT if the use is much or little it will surely destryt/i, rga and Sruti"): Jti mdicates adrsta and rgamans pleasure or rakti,which-is a visible or empirical result Srutii hrae shows that it leads to what ,is not pleasing (yairasya). Antara svara hrae includes kka al^.'^

**
* This can refer cjnly togna.

J Pnim4.I.4I.
, --.r

*patgra could be vlgra* tips of the hair.

*** Th text appeals defective,, **** The meaning of naya is not clear.

78

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND FTS COMMENTARY

79

Text : Now we shall speak of the Jatis Verse-37 :"Three jtis should beknown ascharacterised by vara sadharana, viz. Madhyama, Pdcann as also ^adjamadhyanm. Verse-38 : The ainsas in these (jtis) should be (respectively) known as adja, madhyama and pacama. In case of pacama, it is to be applied as an alternative to an extreitiely weak note. Verse-39 : Jtis are eighteen, this has been thus said by Brahma in the past On those, indeed, will I throw light, along with their divisions of graha, amsa etc. ^ Verse-40-41 : Dependent on the sa4ja-grama (the jtis) are $4, Arsabh, Dhaiva, Nifdim, Sadjodicyava, ^adjakaisUa as also ^adjamadhy.

will be manifested, is their subdivision into graha, amsa, tra, mandra, dava, auduvika etc^., in the order of their characteristics. This is the meaning. The division of srutis, belonging to the sadja grma is workec out and the jatis which rely upon them. Similarly, others depend on the madhyama grama. 'Threetis,' include these (i.e., sdmrana) svaras. Because, in them nisda and gndhra, produce dava and audavita andf hence they are vakra in the first place, still, on account of their debilitation (of ga and ni) they are distorted or vikrta and become l^kali or antara. Hence, svara sdhram is used here on account of being specially laboured. Thus, kkaUis used in Suddha-sadja. Having shown this, (since it is held) that on account of being weak notes they are not used as amsa, it may be asked here, what is their amsa, which is the content of sadharana! He. (i.e., Bharata) answers, "their amsas are sadja, madhyama and pacama. According to their nature, they are weak and pacama is used as an alternative." There are seven amsas in Sadja-madhyani (ti). In this (iii) when nida and gndhra (as fiill two sruti notes) are amsas, then sdhram (i.e., antara ga and kkalirii) canmt be^sed. They (i.e., the sdhram svaras) are used in a special way only with sa, ma and pa (as amsas). The same is true of Madhyan. That avoids two srutis, and has five amsas. In the PcmcanH, rsabha and pcmcama are the, ainsas. In this there may, be the use of kkali and antara (only) with pacama (i.e., as amsa). Hence it is said, that on account of debilitation, there is-revers^ in,Pacanu. Only debilitation is to be m^e. HenCe, in the place of grulhra, which causes dava, when there has to be a weak antra note, it^hpuld be made weakerstill. Similarly, in the place of nisda causing auduvita, the use is of a weak kka, which has to be made

Verses 41-43 l'Hereafter, I will speak of the ones dependent on the madhyama grma (which should be) known as Gndhn, Madhyama, (jndhrodkyav, Pticam, Raktagndhn, Gndhrapancaim, Madhyamodyam, as well as Natidayani,^rmravl,ndhn, and Kaisila. Vrse-44 : There are threejatis where the learned enjoined the use of the sadharana note, (these are) Madhyam, ^adjamadhy and Pacanu. Verse-45 : The ainsas of these (/5/is which permit the use svara dhraiya) should be knownas madhyama and pancama. In the case of Pacanu it is to be applied as an alternative to an extremely " weak.note." Commentary : Having used sadja and kakaU, using the .same, or another ^ote in ascent (text missing) why has this been said? In this qontext, it-is said 'Now we shall describe the jatis'. (They are the) seven modes,(rTft") consisting of suddha, bhinna, gauda, rga, sdhram,bhs and vibfs produced for the sake of visible achievement (drs siddhi) and for thelsce of helping the appreciation of bhva and rasa. Because these arise from tem'they ar"called jatis.Thsb eighteen jatis have been spoken by Brahma in the beginning. Because it is- a-cred tradition iptgam), therefore it is unalterable.'^^ I will desribe them in their division into graha,amsa etc. They are already given-iif the heart. ' What

weakr.'^^ Picse passage' between verses 45-46 : Svara jtis are suddha or pi^ andvOgia or modified. TheSodi/uu in theare?4ji, DhuivaS and NiditS. In the madhyama grama (they) are Gndhn, Madhyama and Pacanu. These are witt complete (seven i.e., not deficient) notes, and their eunsa, graha, nysa and apanysa are according fo the svara witib which e jti was named.

80

ABHINAVABHARA 77

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

81

The mjtas are characterised by tlie distortion of one, two or

many ofthesecharacteristcs(thelOytf/ata9)exceptfor the ya.

These (modifedyat) are known as Of these (i,e., x^SuddKs) the rule about the nyasa was, that it was to be in h^^mandr r lower octave, but as regards the there was no such rule. The Vikris ansmg from mutual contact are eleven. Due to interactioii (between ' theyats) eleven (/atts) are caused. NamlyVerse-46 : Suddhas and are indeed bom of samavaya or collection (of the 10 jati iakgatfos like graha, nyma etc.) Then again, thenon-pureorthemodified(arisingoutofmutualcombinaon)are eleven. Versed? : Thereafter,now, I shaUspeak here, in due order and with pro^r brevity of jatis which are caused (orarise) by the notes and ainsas of these (the suddha)tis. ^ Commentary : Now 'he' (Bharata) divides the jatis, e., Wajatis or those name4 after svaras, which 'are pure, and (th others are) dtraed iVMS): Their divisionjnto the grmas is then mentioned. Th are defiped as' Arsabhl, Dhaivaiiand Nisadinl in sadja gra^, Gndfin, Madhyam, PaJicaml in madhyama grama\ There is no deficiency m the notesthey useand havetheir own svara,ainsa, gfaha,

of contact; the distorted produce elevai through mutual contact. Thus, 'he' (Bharata) mentions an ry (verse) to summarise this "Suddha and Vikn jtis are bom of combination (samavaya)." The etymofogy of jti here is because they are bom of the set of ruti,svara, graha etc., through combination. Suddha and Vikrta, this is the division. 'Eva' showis that the / vikrtatva is of the Suddhs alone. They are not any separate jtis. 'Then ! again', the Suddhas cause the Vikrtas and by them are caused the eleven. n^eword 'again' isformakingadistinction(of the seven from theeleven). Only eleven, not more; 'Tu' is for negation. Among them, separated from the others, they should be described by their names. They are not to be / described as Suddhs. Being derived from matraial substance an object is named after the cause (but they are not to be named like this).^^ 'Now this' for conjunction, connects with thentinuing part of the causes.Thus the contact (samsarga) jti of sadja and madhyam is called Sadja madhyama. Let others he jtis, i.e., to say inthe aina, rga, vibh,de, nirga etc. There areeighteen nirganysas becausethey areable tocreate a notion of commonness on account of the continuity of the principal antaramdrga, nysa etc. Others say that jtis are so called because th^ are bom from the schrane belonging to arna, grma etc. (This is after reading gata for ata. Other wise it wotd be- from the hundredschranes of etc.}.The commentator (ihKa) says that ti is so called becauSie it is the source of the birth of all the rga etc. One may iask, from which Vikrta does \diich jati arise? Witly this qU^on in mind and also wishing to explain more, the author affimis 'those which are boTR from them'. Those two being produced it is also proceed. Or, it is produced in that, thus it may be either pacami tatpuru^ or karmadhraya (Aitober (Privation wiiere part of the sentence is missing). Among the notes, graha, apanysa etc, the propo* form will be mentioned. Thus, by grasping thesvara among the ama, theaias are obtained as a dividend, because of their prominence, they are mentioned separately. In the caseof deficiencies consisting of sadavaand au4uvika,'h.e' says that'he' woulddescribe their modeand form inorder, disregarding the list of names or content.

nyasaanda;,ays.lbeWimsdonothavethesecharacteristics They
are distmguished the distortion of one. two or several of these characteristics, amely,^nia,5/^o, apanyasa etc., but excluding yya. Thus there-may be sadaya-vik^a, grah-vikr aitiSa-vikn and vikrta in all three, graha, amsa and dpanyasa. . the divisionofgraha andapahysaiJo^ is thedivfeion Suddha and mri. The answer is given in the text in the rule

aboutHyasaalso: Amongst theseytwhicharepure, the

(I.e., the nivosa svara which gives the name to theyor,), which is mandra by rule is unregulated in the yikrias-By describing the nysa and an, faramar^a (missing) by the shdowiUke fomi-evn though distorted (vikrta) that form is partfy manifested.. ^ The objectiOT raised is, that tIii dOT-bes the division of the seven Oatis), but eighten have been lisied. The answer is, thafeleven are bora

82

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

83

Prose passage between verses 47- : $a4fanudhyan should be kno^ as produced [myrm should be nimita] fhmi (the combination of theyato) and Madhyama, ^a4jakaisi!a from ^/and GndhSri-, ?adfodlcym from Gandhan,a4ja and DhaivaS-y GndharotScyav ftom' ^ad, Gndhan, Madhyama and DhaivaS-, MadhyamawScyav ftom DJmvaS, Pancaml, Madhyama and Gndhn^ RaktagndHn fitm Madhyaim, Paiicanu and N^idamS-, AndhnfTom (ndhn '^^^-,Nandyat^frfmiP<mcanu,ArfabmandGndlri;Krni^

Verse-56-57 : In the madhyama-grma, Kantmra^ Gndhrapancann and Madhyamodlcyava should be known as with full (seven) notes. And again, Gndhrodtcyav, Andhi and NandayanS of the madhyama-gi^ma are known as hexatonic by the Wise. In this way are the.^ic of the two gramas known by the Wise. Verse-58 :Hereafter (I) will speak of how to determine the ainsas of these {tis).The hexatonic form of^adjamadhyani is notdesirable on the seventh (note. Le., nidada) beit^ itsjniling) amsa. Verse-59 : In the same manner, it (the .hexatonic form in $a4jamadhyanm) is not desirable on gndhra (being the amsa), due to the rule of sainvadydU/pa, Le., the sathvadt note cannot be omitted. (Because of this) the f^va form of Gndhri, Raktagndhn and KaisiE is notpossible when there is pacama (as the ruling amsa). Verse-60 : In $a4 with^ndhra as amsa,the hexatonic rendei^ ing should be prevented. In the $a4jodicyav, on dhaivata being the amsa there is no salava rendering. Verse-61-63 : Because of the rule that the samv^ cannot be omitted, Aese (being the rulingomsos) prevent hexatonic rendering. In Gndhanand Raktagndhriwhen faifja, madhyama, pmcama and nigadd (are the amsas then) there is no pentatonic rendering. In a4janiadhyam, two ainsas, gndhra and nida; fabha inPacamr and dhaivata in Eaiila (prevent atufuvita). In this way these twelve {amsas) here always prevent the pentatonic rendering (of certaintis).
Commentaiy : He (Bbaiata) mentions the derivations first. 'From SSi/rand Ma4hyam etc.', 'excluding DhaivaSandJrja/iT, ending with 'KaisiJa'. The first three are in sad grama. This is obvious from sad being prefixed to them. Even if grcdia etc., are mixed, there may not be a confusipn of gramas. The Kaisika, which is grama sdHrana is used there frequently. The rest is clear. He (Bharata) concludes, 'Not mutually produced', i.e., produced mutually by contact. Since it may be objected that the characteristics in Sad madhyam are separately notified, so 'he* says, 'with separate characteristics,' What is the use of that ? Answer : those svaras and the division of graha etc., are to be construed through their being relied upon.

tromNifada^,A,^abKianP<mcam-,Gndhn^mcanutv^Pmcam and Ondha-, Kaisila from omitting DhaivaS aa A/yaft/ff(Le, Kaisiki is producedbythecombinatiohofthe5(p,G&aflhiMt^^ and Naifa^tis).
Verse-48 ; Theseyat should indeed be known as produced by mutual contact (Even though)dependenton the masvn(;flto)of the two gramas, they are endowed with distinct characteristics. Verse-49 : Fouk-ya/is are known asalways heptatonic by&e Wise, four are known as hexatonic and ten are remembered as pentatonic. Veree-50 : Madhyamodkya^, ^a^akaisUa, KmOm-n as also ndharapancanu are with fedi (L&, seven) notes, Verse-5 :$4, Andh, Nandayanaa Gndfiroincyav, these should be known as heratonic. The pentatonic are ten. Verse-S2: NaifiS,i^abhi,Dhaivaff,$a4jamadhyam5and $ad/odicyavaiithese five (pentatonic) should be remember^ as of the sadja grama.

. Rakta-ginhSn, Madhyani, ^mcaim and Kaisa, these five are dependent on the madhyama-gmma.
Verse-S4 : Those which are known as of seven notes (the heptatonic/Sto) and those which are remembered as ofsixnotes (hexatonitf jatis) may sometimes be (rendered as) hexatonic aruT pentatonic (respectvefy^.
Verse-SS : In the fa^j-grma, ^akaisilu shonld be known as

jati)

heptatonic. Eleven t&oagh hexatonic, should be understood as heptatonic m the context of gandhara (bemg the ruling aiis<r of this

84

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

85

the force of definition gives 'what belong to the two gramas three in sadja grma and eight in the others. Or, it may be so called because their origin is in the jatis of the two gramas or, out of contact samsarga, mutually (these Vikrta jatis) being dependent on the svaras nd named after them acquire separate characteristics and become derivative in / _ relation to the Suddhas, but parental in relation to the jatis born of samsarga [tyatvam should be correcteffto janyatvam]}^^ Now wishing to mention the division into purna, sadava and auduvika through the mode of deficiency (i.e., dropping of notes)^ 'he' describes the prnas and then the rule by saying, there ate 'four'Thus, purna j'tis are four 'Madhyamao^cyav' etc. Then 'he' mentions the' sdavas, Sdji etc. Apart from this, there remain only tenj, consisting of five notes. So 'he' shows that, 'of the paricasvaras' ten are to be understood.- " He (Bharata) describes the division of the pacasvaras. 'Nisdirn; Arsabf etc.' 'Dependent on the sadja' means, dependent on the sadja grma. Now 'he' answers the enquiry about madhyama-gmm. 'Those which are of-seven notes etc.' Those with seven notes are the*four Madhyamodicyav etc, as per rule. Hence, they are never olessXi.e., than' seven) notes. Those four beginning with sadja they becofaies sdayd and auduva etc, * Now 'he' describes the nature of prnas and sadavas according to the nature Of their grcnas. 'In the sadja grma etc.' ^Through thpaddition of gndhra' i.e., to say, through Ae use of gndhra On accuilt'of its predominance as amsa, SdjT, even though named on the sadja (and even though it consist of six'notes) is to be regarded as puma. Nisd mflceS them sdava. When gndhra is the"amia then nisda being its samvdi cannot be dropped and is in the sense of 'also'. 'v' is to b constued after the word purn. TTie three Ktmrain eto.^are of rhadhyama grma and of all full notes. GaridhrodlCyav etc., are three sdavas. They are. I <55 ^ sometimes auduvika 'Assigning the amsas' That by which the arnsa notes are detenhihed. This note (is amia) in this'jaii.'There are six amia notes. This will be

discussed in the text on nulana (?) That will be described l^r. In the reading ariga vikalpanam', vikalpana means statement of the ahga, i.e., sadavaj>r audt{vika form. With this amsa that (form sadava or aui^va) 'In the seventh'- the meaning is that when nisada or does not exist. gndhra is amsa, then six notes are not desired. The reason is, that, since the samvadi is not dropped, when there is nisda (as amsa), there will be no sdava, because it is- the samvdi of gndhra.^^ Of these three, Gndhri etc., thehexatonic form is effected by the (dropping of) rsabha. That (i.e., ri) is the samvadi of pcmcama in madhyama grma. When that (i.e., pcmcama) is the amsa, then there is no dava. In Sadjodicyav with dhaiPata asthe (ruling) mia there is no hexatonicrendering because it (dha) is (samvadi) with rsabha. In sadjagrama it (i.e., n) is the samvadi of dftaivai.'^ He concludes together 'these seven' ni, ga, pa, pa, pa. ga, dha, these are the ainscL These (being the amsas) cause asadava sutessively in Sadjcmuulhyam etc., and in Sadjodlcyava etc.^^^ Now,, 'he' speaks about ther cansas which bar auduvita. In GruJhn and Raktagndhn it.is the seventh (note) i.e., nisada (which prevents auduvita). Of Sadjamadhyam, the two, gndhra and nisda being the (ruling) amsas there isTio pentatonic rendering. He (Bharata) concludes 'thus there are twelve', (amsas which prevent auduvita of certain/fis)- eight [sa, ma, pa,nieach] in Gndhri aad Raktagndhn^^^,tv/o [ga,n] inSadjamadhyarri, and two, i.e., in Pacanu [rsabha] and Kaisila [dhaivata]. Now, sometimes there is a reason for not dropping of the amsa, the (rule of) samvadi. In its absence such a description is given as the reason. So much is the number (asankhya ? ). There are sixty amsas . There are mne amsas in the four jtis which are always plirm (i.e., cannot be rendered either sdava orauduva). Since these are asadava (with nine amsas) there are (as remainder only) fifty-foi^ (amsas). From these (i.e., out of-these fiftyrfour), there are seven (amsas which cause jtis to be) asadavas. Thus, with four prnas and sdavas, there are (remaining) forty-seven amsas.^'^^ Among the other fourteen (which are not purna jtis), four are not auduvika. In connection with them, twelve amsas will
* * could bea mistake for Later it is given as 'trisastHx vide note 40.

l'io.

.,

140

88

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYA&STRA AND TTS COMMENTARY

89

Now the grahas Verse-67 : The grahas of thejSiis are known as simflar to amsas, by whictthe song commences, that is ernia or alternatively the graha. Now amsa there Verse-68 :Amsa (is thatnote in themusical composition) m which lies the chaim of the rga and from which the charm proceeds. It determines the mandraand ira octaves andit is themost prolificnote. (Variant reading - It is found proUfically in the combination many notes. The others which are strong are the samvadiand anuva^

also).
Veise-69 : That which is the determinant of graha, apanyasa, vtoasaaiidai.n)'5saBdj5s, thathichi repeatedand toUowrf, that , marked bj ten characteriscs. The mowment m the higher octave is up to five notes. ^ Verse-70 ; The movement in the higher octave here (in the jats) should be known as up to the fourth svara ornote from the amia (note). Or up to the fifth note, beyond wlch it is not desirable. Then again, the Wise should also count the omitted note m this

'Sodava and auduvita' - this is wdiere they take place (i.e., in;aiis). This character is npt universal. He (Bharata) shows\his by 'thus'. They (referring to sodavo and auduvita) have the meanmg f arising in the mannpa- mdicated. The otiier eight are without exception. Thus vnthout tiie first and tiie middle bhvas* how can tiiere be. aii order? Hence graha, apanyasa and nysa. Musical appreciation or rakti cannot be plausible in terms of the relationship of primary and secondary. Hence dmia is predominant and its samvadi is like its minister. Its samvadi (muchness) of that is in relation to rarity of usage. The rarity of the notes belongs to them when they are dropped on account of vibhva, etc. In tiie mere notes of the middle saptaka prescribed in tiie indication of the mrcchari being used monotonously there is no hmer enjoyment (rakti)}'" Hence, the use of tara and mandra. And hraice, the eight features are universal. Where some note is dropped in the jati, tiiis dropping is anothra characteristic.
Thus the ten having been determined, (the text) goes on to characterise graha (literally the second half of the sentence does not construe). 'But the grahas', 'of all the jtis', grahas are connected with them. Together like ornothey become sixty-three. What is this graha? The text answers- It is another name for that (for oihSa apparentiy). How ? The answer is 'that, by which the song is to begin.' That by which the performance of the jti istaken on, that is graha. Hence, it is another name for arnia when conjomed with another property. Hesays that 'constructed by graha' By the property of graha, thus constructed, it is made as if of two natures. It becomes graha, not simply by this predominance. G'raKa is produced by that capacity. Then why has it been said so explicity. So 'he' says 'constructed by graha'. Sometimes ama does Aot become, graha. For example, such is pancama in Nandayanti, gndhra is its graha. Hence graha must be mentioned sepaarately.'''^^ It is not that-this principlehere, has many illustrations (i.e., mti gna). In the music as actually practiced (lak^e) they generally tend to be one, however,, there, is a purpose for givmg a separate definition for graha. In grma-ragas etc., the recognition of the graftaiva (of a note) apart from the amia is.
* ** "Prathatna - madhyamanyapi bhavannd vin". The text is obscure, hi fact, it does not ccmstrue. Text obscure and defective

ruleof the higher octave.The movementinthelower octave isof three types, at is) up to amia (note), up to the nyasa and up to the apany^a - The (movement in the) lower octave isnot beyond the amia smd two of the nysa (i.e., up to the nysa note and also one more note beyond
the njisa). _ ^ _ VeRe-71 ; (For mstance) if gandhara isnamed as the nyasa, then (the

avaroha movanent hwe would be) uptaj^abha.

Verses-66-71 : Commentary : Well, it has been said 'in ihe jatis'^ What ien is this m f The answer is svaras themselves when they have a distinct atr^gement and when they produce musical enjoyment (raM) ad also invisible good (adrsta) in the other world and visible here, then they ar called/ati. W is this order or arrangement (sannfveia) ? So it is said the characteristics of yoii are ten, 'graha and amia.' Tbe dual hrae, suggests that often as a rule ainsa alone js to be graho-'lra, mandra notes'. Mutual relativity is shown here.

90

ABHINAVABHARA

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

91

(not?) possible.* For example, when the fonn (i.e., composition) is sung in A/ava-/faisito,allofthem(ail theseforms)donothavesaci;aasgrafta regul^ly.^^' Simiiarly one should construe elsewhere. The text describes aa. 'In which the raga occurs'. Which bemg present, raga or rakti (i.e., charm) arises and (because of which) the jam receive their form, just s, when the head is there, a person may be recognised. Thus, it is. that ^ara (on. which ihe jati is based and) wluch isusedmuch.'Fromwhich'.herethe>acm is usedowmg to omission of lyap It has been said that where samvadiand anuvadi 2xe sodded with emphasis, such as a note in the lower octave, there, the samvadi and the flnuvdTfunction like aksa even without being amsa. (A) doubt (anses). Will not the ji be from the vivadis ? On that account, for considermg the Umit in the tra sthana, 'he' will describe the positions m tara and mandra from ie paHcama eie}'" That which occi^frequenUy com-

Similarly, it Has to be thus considered in other places also. Hence, this is not the proper explanation. What would it be thai ? Listen. Wehave already gathered that the aina is to be produced by mrcchan, from the madhyama sapiaka. From this, it follows, that all the notes in that (madhya) saptaka are tobe used, disregarding (here such) considerations (as) of iara and mandra. With respect as to how many notes within the octave produce worldly and other worldly good, what is formulated, is, only the limits of ienotes vibrating inthe head andheart (irasyaurasyas-

h*

varanisth)}^'
Objection : When thrae is already the middle saptaka,should all the notes of the tra saptaka be used after that ? The answer is, not always. If sadja is the ama, then the tra saptaka should be upto ri, ga, ma (pa), if one is capable of doing so (i.e., stretching the voice so high). But even if one is capable, one should not go beyond that But taking a lower note, too, B not an error. This is shown by the word para. When rpibha is the aihia, the notes are to be taken upto dhaivata (fix)m) the rsabha of the tra saptaka. When gndhra is the aina then the seven notes are to be taka ^ding with ntsda. In madhyama, pancama, dhaivata, nisda, mthese (being ^e AN^A) the notes are to be taken with nisada at the end. ID these five arnas, the vdole of the tara saptaka is to be taken. If the capacity (range of the voice) is medium, [madhyamah is possibly madhyama] then the foha is only upto four notes. But in the Nandayan, the extreme limit (is reached) in thetra.It is explained there thus, that "the movement in the ira saptaka never goes beyond (iSra) sadja". Hence, it is not correct to explain, as others havedone, that the.text 'upto tiie fourth note' refers to Nandayan^'^ The Imrika 'ainttragatim vidyt ' has this menig. It says that, when the middle saptaka is the source of the mrcchari, then depending on the ama, the movement in the ira or aroha is to be upto the note in the tra saptaka, flot beyond.it. Aina has been earlier defined as the leader oFth tra and mandra. This has not been properly considered by those comhiitators. In rsabha and gndhra, dhaivata and riisada are not to be employed, since they lack rakti, they cannot be svaras.' It is not correct to say, that if one does not (go upto tradha and ni),one does not have the power (to stretch the voice so high).

pared to aU oier notes in ie whole song (i.e., iat which itself determines ie five forms of graha etc.. or its samvadi, anuvadi foTws,
butneverav/v5dr.i^^ Grato etc., are ie content to be determmed by iat (i e aia determines graha etc.). That which colours oier notes even as ii ceases, just as one may cover what has its face turned the o&er way That a/kio, being ie principal, has'ten charactensUcs . Allie characteristics togeier suffice to produce tiie character of amsa. Hence. ie first characteristic has been mentioned separately from iat

ainsa.
Now. 'he' describes tiie tra "ie movement of tiie tara is up to five notes". Objection, if tiie rule is,1hat one has to ascerd from tiie ama svara along tiie order of ascent (atoha krama) as far as tiie fiftii from tiiat, then it will greatly contradict practice {Msya virodha). F' Sadjl, from tiip sadja ainia^me would need to ascend upto tiie fiftii from tiiat vey note. So it is seenm practice in So# dha dha m sa m dha pa ni dha pa, tiius. fii ArsabM'ihe ascent would be upto dhaiyata. But, it is seen upto nisada, noidha, ri. ni, ma. go, ma. In Ni^vati.aroha is seen upto tiie pcmcama pa, pa, ri, ga thus.
cf. Panmal ed.

92

ABHINAVABHARATI

TRANSLATION OF THE TEXT: NATYASASTRA AND TTS COMMENTARY

93

As to what others have explained - from the ainka svara four or five notes of the ira saptaka be used, for example^whensad;a is amsa, sa, ri, ga, ma, pa; in rsabha ri, ga, ma, pa, dha; in ^2ndAaro - ga, ma, pa, dha, ni; in madhyama ga, ma, pa, dha, ni, and similarly > pancama dhaivata and nisda. It is these that are established by practice (laksya). Thus in Nisdava, where nisada is aHia, tara sadja is seen toough the arliiaasthemeansMtisunnecessarytosaymore-Theexplanationasgiven brings out the laksam and should be respected. 'N5a/iparamii',oneisfreetostopearlierinthetraapafai.Vftei', the idea is! that, this is with respect to iht jtis. This idea does not hold ahontgrma-rgas etc.'Budhmriti','he' is saying, that, those who re^d anote as due to be omitted, they SU have to countthe_omitted note.-mus VisSkhilc^a "The nsl svara is counted in the tara vidhi". ^ Tra means that which causes the voice to cross forcefully. Now 'he' mentions the mandra. 'Tridh mandragatiriti'. The root mad has the sense of rejoicing. Mandra is delightful {lalim) because it is low (amra i.e.. bends down). That(note) which is theamsasvaram the mc^hya saptaka, upto that (very note) in i&,mandra saptka should be taken.x,r upto'the n^sa svara, or beyond that, meanipgthe note that is beyond the nysa svara. The useofparashowstheadmissibihty of one lower note being taken.-This gives a certain latitude. Objection-.Inthatcase,how can d/iaivaiabe the limitof the mandra in ffid/T? Why should not the limit be upto (mandra) sadja"] (The answer is. that) whUe it is admissible to use the frill mandra sapafaitiptosad;a;onemayponallygo(only)uptod/aivaa.Thatsa^;a. which is graha and nysa (ih SdyT). is used from the mandra^ toQ madhya (to the tara- this is missig in the text). The rule whiqh speaks of mandra does not mean that ira (sa) is excluded, because m S4}i, sadja is heard in aU -three octaves% .They read this Krika hcr^ mandrastvaitparo risRtL Thus, m the Nisadava. where ntsada is the ama and graha svara in^the madhya saptahi, the range is not seen upto mandra nisda qply. but upto the nyas<i svara. (Infact. here) Aere are two limits in th^m^ndra, upto nyasa and the note beyond the nyasa. Now the
Tlie printed text is ambiguous.
% t

illustration and the lak^a corresponding to it. Therule isthat in the laksam of gndhra\as nysa),the mandrais reachedby being (onenote) beyond the nysa. So why repeat? Thre is reason which will be mentioned later on. 'The rsabha may be omitted in the mandra'. So 'he' shows that this option is with respect to the limitations laid down. Objection, the Sage puts forwardstatements as indications, then why this mention of gndharva ^ossibl^ gndhra)! But this has already berai said before, so there is no harm. Now there are twenty one nysas. Nysa is the concludingnote of an aitga (a specific unit or division of the melodic composition). In the same way, apanyasa is (used) in the middle of an aitga (i.e., on semi-completion). There are fifty-six nysas. Namely V?rse-72 ; Nysa is (to be used) on the completion of an aitga and they aretwenty-one in number.The apanyasasare fifty-sixin number and they are used in the middle of an aitga. Sannyasa and yinyasa arealso in the middleof an aiigaonly. Since they are not fixed, they are prolific. ' Verse-73 : Thus the nysa nte was sed iathe Middle of the firsf vida (a sub-division of the melodic structure): Having avoided the vivdT,'one may speak of sannysa. Since anywhere it has been kept (vihyst) at the conclusiotf of a pada, it is called vinysa. The reduction (alpatva) of notes is-efFected by laitghana (gliding over note sas to leave it unemphsized) and by anabhyasa (avoiding Its repetftfon)! Thus, there is noif-repetition of noiies wliich are, used wQifti"he antaramdrga [i.e., the caan] of a gita or melodici structure, and of those which are responsible fbr its hxatonic oi^ pentatonic rendeiing and which are not arhias. (As desired) in each , due to lahghatm (gliding) aaanbhysa (non-repetition) thereis alpatva (reductioii)*bi'2>a^'il'a (m^ificatin). ThusVerse-74 ; In the a^Hitva and bahutva, th strength and weakness ti is always of two types.
r > I

should be ascertained by the notesof the (specific) jas. This alpatva of the

94

ABHINAVABHRATI

TRANSLAnON OF THE TEXT : NAJYASASTRA AND TTS COMMENTARY

95

Verse-75 : Those which express the identity of ihe jatis are, the strong notes {balasthna) manifest in the movement (of notes) in the ama, in the a^poivo which manifests itself in the weak (notes), nyasa and antaramrga [the typical movement or expression of ajaftl.

Commentary ; Now 'he' mentions the number of nyasas and their laksams. 'Thare are twenty-one nysas', two in Sadjamadhyam, three in Kaiiila and sixteoi in the rest. Nysa is doived thus, that which is used m the performance, so that when the body of the;aii has to be completed, it is the note where this ccanpletion is don [i.e., it is the concluding note of a jti composition]. Now what is the diffence betweai nysa and apanysa 1 Nyasa is at the end of the ahga, while apanysa is in its midde, that is, it is a subsidiary ending or avntarasanipti. Thus, Dattilacya saj^ ndn is in the middle. Thus, in the sadjagrma, there are seven apanyasas of the Sadjamadhyam. In the Sadjodicyav, there are two. Of the five, thereare three for each, thus tha:e are twenty-four. The rest are m the madhycOna grma. This voy nahire of apAysa is de^bed m the anump called vitnavrtta. Thus- 'There are twenty-one nysas in the anga-samap and there are fifty-six apanysas in ahgamadhya.' Now, although it had tot been named earlir, still asjnrsupposed in the definition of aisa, 'he' defmes mcidentally- 'Samyasa and vinysa are within ama It means that they are withm the vida. As to what is the number of these irregulars, the text says 'th^ ^ many, because of irregularity.' Ihe meaning is, like nysaand apanyasa, there is a ml for their employment m eachyri Now 'he' describes the sannyasa. 'Tatra prathama vidri madhye so' bhidhtavyah'. It is a pote, ^ch, not being the vivmE of the ama is used at the id of the first vidn. Such a note is sannysa. Being the nysa, that is neat (sanpabhta, i.e., samipabfmto nysah sanysah - such a meaning is sugg^ted). 'Krtva padvasne vinysah'.if & note is thesamvdi or anuvadi of ihe ama occasionally [kvapi, not A^i] and is placed at the end of the pada wMch has the form of the vidn {vidbK^pasyapadasya) that is vinyasa. Thus having defined the two, smiysa and vinysa, 'he' defines the ^atva, which belongs to the context of amia. He does it.by a 'two fold

division. Lahghana means ptxxeding to anoth^ note while touchmg but not resting. In this, the note becomes important for itself. It is wrong to say ibsS. lahghana is sUght avoidance (varjanam = isatvarjanam). In that case, while speaking of alpatva,there would be no sq)arate alpatva on account of the rule of sadava and auduvita. In the Nandayanti, rsabha is samvdiof the ama pancamaand hoice is not fit for langhana, as thesamvdshould not be canitted. Since there is objection in rendering ithexatcmic andpmtatonic inthis way, i.e., by omission of samvdT, hence 'he' divides alpatva into lahghana and abhysa.^'^^ 'Tatra sdvaudavita rndmiti". In the prnavastha (i.e., in the heptatonic - rendering ofthejti), the lopya svara (is not) totally omitted, (but in fact) there that note is frequently glided OVCT (Umghana). Occasionally, there is also non-repetitioQ or anabhyasa.Anabhyasa, is used with those notes which are not pryyma such as nisada and rsabha in the Sdji, when they are used in theantaramrga. Sometimes in the melody, there is also lahghana of them, excepting the position of graha and samvadi. 'Antaranirganasya (ntrgasya)'.'YathajatitV (this is not in the present text). Sometimes a note, though not atha, is still not a^a. Thus in the Krmram, gndhra is much used in the antaramrga m harmony with all the notes (sarvasvarasahgati).This will be mentioned later in the text 'gndhra e^ecially moves every where'. Now 'he' moitions bahutva and defmes it by saying that, alpa is that which is weak (abala) and in contrast, the strong note is frequfcnt (bahutva) as may be understood by implication. Hence, it is, said 'Jtisvaraistu caitod".. (Verse-74).The meaning is, that, like the alpatva belonging to theyati, the bahutva also, is two-fold, and that is shown by the notes of theyii (Jtisvard). That is to say, by the alternate amas (paryymas) and notes which are not samvdis. Thus, bahutva is two-fold,.due to alahghana (non-skipping) and abhysa (repetition). Thus, showing the use of alpatva and bahutva, 'he' goes on to say 'sahcjrna. vyaktikrakah' (verse-75). That, which is the arna by sancra i.e., by paryya or alternation has its sphere of application in certain jtis, of which the nysa along with the antaramrga reveals its (i.e., the jtis') nature. Nototherwise. That is where lies its real use. 'Alapatva iti'. The lpatva operates on notes which are to be omitted (lopya) and on non-amia (ahama),and in theyaiis weakened

96

^ABHINAVABHARATI

TRANSLATION OF THE TEXT: NATYASASTRA ANDTTS COMMENTARY

97

by such notes, {i.e., which have many anaihsas or permit omission of notes) sounds the alpatva,not otherwise. 'Anye' (not m the printed text). Others explam this verse as the definition of antarmarg. They beheve, that the form or nature of the antaramarga cannot be prop^ly delmeatfid without alptva and bahutva. Considering this as th intention they expound this verse as defining the antaramarga. How? Because antaramrga manifests the;"5ris. It is characterised by its dependence u^n alpatva and bahutva. When the anuvadi svara is used and when tiie amsa svara is predommant in a selective group of notes (da/a) and when tiie sakvacTi notes are forceful, tiien, on accomit of tiie alpatva.^ -by skipping over tiie notes vi^ch are weak, tiiat which makes a nyasa, tii^ is antaramarga.^^ Otiiers declare tiie meaning of tiie verse tiius - tiiis;aii isinherentlyweak,tiiiS7istrong.Such(adistinction)maybemiderst^d from tiie rtyasa and antaramarga in which is i^erent tiie bahutva which is made to operate on alternate amias (paryayamsoi). The hexatonic.rendering is constituted by six notes. This (hexatomc or pentotonic rendering) is of fourteen types. Thus, by tiie aforesaid, tiie afnas specified in tiie tis (i.e., tiie a^as which permit hexatomc rendering) are forty-seven. Verse-76 : Five notesconstitute au4uvita or pentatonicrendering, and should be known as ten-fold by the expert performers. The aforesaid characteristic of this has been said to be of thirty types (i.e., there are a total of thirty amsas which when used, permit pentatonic rendeiing). Verse-77 :The rendering (of melodies) is,ofsix notes or hexatonic and also of five notes or pentatonic. The rendering (of a melody) can abo be with four notes as indeed in the avorio dhruvs here. Commentary : Altiiough s5dflva has been mentioned before, 'he' speaks of it again to aUay tiie contradiction arismg from tiie listing of tiie ten'characteristics 'sadava consists of six notes'. There ar^ fourteen ways of tiie tis There are fourteen presoiptions f the tis. Four bemg always imrka, tiiey are not counted.;Of forty-even types (pr^a)'^ There are a total number of sixty-tkee amia^.'which are not dropped. There are seven exceptions. So 'he' says 't^ama-pral^ramiti'. In tiie

tis there are cotain ama svaras. They have a certain form (prakrd) which is by the transgression of the notes or apavada vidhi, Vtch has been mentioned. Then auduva is moitoned, consisting of five dotes and of ten types. In all thetoitis,thexe.are thirty prakras of thesixty-three, four are nitya-samprna, and hoice their nine amias are dropped. Four are nitya-sadava. Their twelve are dropped. Thus, forty-tworemain. Here twelve are to be takm and thus thirty remain. Hence, it is said, that the 1 defiiij^on has Ijeen moitioned before. Now, apart from this, there is something in the dhruvs, by saying which, it is denied in the gndharva 'Verse-77'. (Next two lines are incomplete, and hoice not clear). In gndharva, it is nec^sary. In the dhruva-gana used in the natya, even four notes are used. It is of three types. This is shown by the use of the word also. Where ? In the avah^ dhruvs, i.e., in these which have many long syllables and are used in karma (rasa). Verse-78 : AD thetis of the two gramas should always be known to have (a total of) sixty-threeamias. And m the same way their grahas too. Now I will speak of the amias and grahas. There (in the tis) Vaise-79 : P<mcainab theoiitia asalso the graha of MadhyamotEcyav, Nandayan as also GndhrapancamL Verse-80 : Dhmvata anj^abha should be known as the amias of DhaivaS. Pancama and rsabha should be known 9s the graha and amia of Pacan. Verse-81 : ad and madhyama should belcnown as the grah and aihia of Gndhrodicyav. Dhaivata, /fabha and nida of

rsablS.
Verse-82 : Gndhra, nidada and rsabha, these three are known s the grahas and amias of Nisadira. Verse-83 : $a4Ja, pmcamaand gandhra, these three only should be known as the amias and grahas of adjakaiiii. Verse-84 : $a4ja, and madhyama, nifda as well as dhaivata are known as the grahas and amias of the ti adjodicyavaS. Verse-85 : Pacama, r$abha, nida and dhaivata have been declared by the Wise as the amias and grahas of Krmran.

98

ABHINAVABBAJTA 77

^^JIONOFTrnTEn-.^TYASiSTJiAANDmcOMMENTAKr

99

Verse-86 : Gndhara,/^abha,pancama and nifoda thesefour ate the amias and grahas otndhn, Verse-87 : a^ja, jjabha, madkyama, pancama along with dhaivata should be known as the grahas and ainsas o Madhyama. Verse-88 : Nisada, fa^ja, gndhra, madhyamaasweUaspmcanta should be known as the grahas and eanias of Gndhn and. Raktag-

Vei^92 : The .classification of all the jatis is in groups of three e W^. know these to be seven groups, with the (athia) notes mcreasmg successively (one by one). Vense-93 ; (These groups should be of) one (aia) note (ti ^np) two (amia) notes (ti group), (group of)'three (atkia) note (/te)^foDr (amia) note (tis),four (jtis) of five (athia) notes (each) one Qati) each with six and seven (amia) notes. Commentary ; Now the division bf theyO'a/iiia^ which are mom than forty-seven is declared to merit a close examination, and hence 'he' summarises the prmcipal classification of number of all the jtis. The coUectionisofthreeyar.OnwhatprincipIearethecoIIectionstobemade . e answ is, that, they are to be made on the basis of tiie increasing ""Til of th^amias is tiie bdsis on ^ch tiie different classes or gam of jtis are to be made. The number o amias vary from one to seven. ^ to ^ seven ganas, each of tiiree jtis, tiien be twenty-oneyar. But tiiat is not so, since only eighteenyi have been mtioned by Brahma. ^e ^wer is, tiiat, tiiere, is a rule about making an exception of tiie nuinbCT tiuee m tiie ganas. Thus tiiere are four ganas having accJtis tech. Thus, wegettwelveyat. In tiiese, theo^iiosrange fromone to four. Now, tiie exception is, tiiat tiiere is a gam of four jtis when tiie ahiias are five. This gives sixteen yat. When tiie a^ias are six or seven tiie g^contam only oeyarieach.'Eachoneyr/isalledagana because ofi^Tuleofvyapadeiivadbhva.'^ Thus, tiiere are pigbteen jtis based on&e numberof amiosvoras. Thereare sixty-tiuee a,iiias. They are tiie granai also. Verse-94 : ThecompositonofthegraAasand amiasofthese Qatis) has been menUoned by me.Now,IwUl againspeak(oftheseyafe)^th the combmation of their (the Jtis) nysa and apanysa ^^95-Excludmgtbe(notes)nifddand,ryahhh,^sbouId have amias. Gndhra as wdl as paiicama should be the apanyasas here.

ndhri
Verse 89:^4fihas its grahas and omios^tiiroa^(or by)dhaivata, gndhra, fa4ja, madhyama and ptmeamtf. It is distorted by the combination of (different) nots if sainv or vivM is made Uie graha then it is distorted]. Verse-90 : Leaving aside /yabha, all the six notes should be remembered as the grahas and amias of KaiiiH. a4jamadhyam should be known as having all seven notes as gmhas and amsas. Verse-91 : These sixty-three should be known as the amias of all theyot. The grahas of these are always to be the same as the amsas. Commentary : Now 'he' mentions the division (of amias) for each yii inorderto describe the number ofthegroAos andamia:^ 'veise-78'. Since the (total) number (of aihsas):is to be reckoned, having disregarded the order of the ^rma division (of jtis) 'he' begins by moitioning the one amia ij^tisJ upto sevoi aihia {tisy. 'Madhyamodicyav' etc: Now three (tis) having one ainsa each. Three (tis) have two (amias each). Three (/iis) have three(amios each).Three (/oris) h^ve four (amsas each). Four (jt) have seven (ainsas). Thus (a total oO sixty-three (ainias). Ordy that wWch is amia may alone be graha. In Nandayanti too the amsa note (pa) is the graha (also) Howevt-, an alternative (opimon) is admitted, gndhra too, on account of its characteristcs as graha may be seen (actingas graha).^^ The Sage hasaffirmed title grohatva of the amia, hence it is predominant Whensome other note, Samvdi or vivadietc., is madie thegraha, then theyare deformed (vikrta). So 'he' says, they are deformed or vikna on account of combination of notes. But this deformation of the initial note (grahavikra) exists for all the jtis.

loo

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND TTS COMMENTARY

101

Verse-96 : The nyasa here (in should be fo^ja and the I seventh (note, i.e.n^da) may be omitted (m thehexatonic rendering). The simcara or movement (in the antaramarga of thisjo^O is between !4Ja and gndhra, and between fo^ja and dhaivata. Verse-97 : In the hexatonic rendering (of this jti), nifa and j-abha should be made weak {(dpau). Gndhra, indeed, should be made prolific here (in this jti) by the Performers. Verse-98 : In Arfobfu (Jti) the amsas an rsabha, nifda as well as dhaivata. These are only the apanyasas aad/yabha is to be remembered as the nysa. Here, it (this jti) is rendered hexatonic by omitting fa4ja and pmcama respectively. Verse-99 : In DhaivaR (jti), the nysa is dhaiyata, the cotisas are j^abha and dhaivata. The apanysas here are dhaivata, fabha and
madhyama:

Verse-107 : Its cpmyasas are dhaivata and a4ja.The movement of the amsa notes {sa, ma, dha, tti) with each other was desirable. Verse-108 : Those proficient in the Gmdharva Veda, render it hexatonic by onttagffabha. The pentatonic rendering there, is, by omitting/rancaina andjfobha. Verse-109 :^a^a,jjabhq and gndhra should be made strong. The prolificuse of gndhra in the lower octave is to be accomplished. Vetse-llft : All.(sevea notes) are the amit inJ^a^jamadhyam and thesaiearethe tpai^asas.^a^ja and tadkyamaare to be made the nysas by thepetfonnets. Verse-ni : The pentantonic rendering is to be accomplished by excluding gandhara and nidada. The hitatonic rendering here, is, strived for bythe excluding tn^oda. Verse-l2 : The movement {smicrd) of aD notes is dsirable (in thisyafO- These should be known as the seven jtis dependent on the
fa4ja grma. ' Commentary : He (Bharata) concludes, that, now, 'I will maition the elemojts including nysaand apanysa of the ten laksams'. The idea is, that, amia and graha have be^ mmtioned, and so also nysa and iqtanysa. Now, there are otha: elements wliich fall within the ten characteristics, suchas sadava, auduvita etc., which have nothem mentioned brfore,andneed to be mentioned for eachyiL Nysa, apmysasia.Ayoga, they, togetho-, constitute a dvanda-sansa here. He want; to explain the ideas of graha and amia again, because that wiU make understanding otherthmgs easier. Geoerally,oflier technical elanaitsdep^id on the atnia ^ diaiacteristics of the ja/i also become clearer at one place. All this attention to jati is useful for dt^a and adrsta puiiposes. Otherssay, that the line only means that the characteristics of ata, grcdia, etc., are to be ivi. He defines Sdji. Sa, ga, ma, pa, dha are amias.Two apmy^as are gapa. Nysa is sa. Wh! nida is omitted sdava is formed. There is a mutual sanearaof saga ttodsdha}^'^ GndAara being very frequent is

Verse-lOO :The petatonic form is rendered by omhting^a^a and pacama.The hexatonic rendering is declared to be without pacama. Verse-101 : (However when this jti is heptat(mic or pnta then) }a4ja and pancama should be used only in the ascent, although they were otherwise notes on which lahghana applied. Nidada aaj^ahha (being attisas) are strong notes as also gndhra. Verse-1021 In MfditS, tgda is bje ana along with gndhra as well as.i?aAa.-These are only the apmysas and the sevoith (note) or n^da is the njsa here. Verse-103 \ Its hexatonic and pentatonic renderg shall be done in the same way as DhaivaS (jti). The (application oQ lahghana wd strong notes there, too, are similarUrBhaivaS. Verse-104 : The amias of $a4jakmii13 are gt^ja, gndhra and pacama, Th^ apmysas here are ;sa{^a, pacama and the seventh note (i.e. nifd). Verse-105 : Gandiera, here (in this jti). is the nysa. No note is - to be omitted here (there was no auduva,i^ava rendering of thisya). The weak notes here were dhmvata, {madhyamd) wA fsabha. Verse-106 :^a^ja and madhyatna, nida as wellas dhaivata, these are the athias of a4joScyaVa. The nysa is madhyama.

it

* The next sentcniK isdefective and (^scuie.

02

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

03

vdi. {^abha) is vivdi, (hence) rsabha is not an amsa, Nisda is lopya. Amsas are ten. They are suddha and vikna. Five are in purna and four are in sadava. On gndhra being tlie amsa there is no hexatonic rendering.'^ The amsas (gd,in this case) are known clearly by the omitted notes {ni, which is-a samvadi of ga). Nourished by that, are the jatyamsas in the svara portion. prt from that and the srita portion, in the nigada stutis and smidhems (l) the gndfirva phala (i.e., adrsta phala) is clear in boih.'^ As has beerfsaid.^'even once properly employed, ieNandayani purifies'. This has been said to make known its preeminence, hi thejoka (among the people), the jati gna is known through the vartika marga, through the smnya tala (i.e., the basic tala) caccatputa etc;, desregarding any special ma. Where there is no alternative to caccatputa etc., and the beats {kaipta) are to be sounded according to the rule of four {catuskaiS), that is not well known among the folk {lake nap^asidhah). This is so, because svarabhaga is prominent there. The forms in sanskrta and prkrta, spoken by Brahma are well known. But in the gitdca etc., tla-bhga is predominant. As has been said, 'those who perform vardhanina, they go to the world of Siva'. This is said to point out its excellence. Now trotik, gitaka [obviously different from the gandharva form of gitaka], etc., are different from the pure forms produced from the jtyamsas and are well known in the order of the grma-rgas. Principally, with the j'tyamsakas, ie tala by giti, raga etc., sanskrta podas and clear attention, all thege form the mixed (forms of) gndharva. Then in that, gitaka, vrdhanwna etc., were manifested of yore [Herejori gana is disting- uished from gitaka; grma-rgas are distinguished from both].
He defines ArsabHi. The amsas are ri, dha, ni, and apanyasas are ri, dha, ni. Nysa is rsabha. Sdava form comes by omitting sadja.Auduvita comes from losing sadja and pacama. There is concordance or sahgati of sa dha and ri ga. Some say that pcmcama may be ski^d (lahghana). In the full form, sadja, gndhra and pcmcama are infrequent (alpatva). In the auduvita, a and ma are infrequent (alpatva). Tala is caccatputa. It is employed usually in th Naiskrniidhruv^^^ Now Dhaivati. The amsas are rsabha, dhaivata. In the vikrta state apanysas are ri, dha, ma. Nyasa is dhaivata. Sdava comes from omitting pancama. In the full form, pa and sa are to be employed in the

aroKi varria or ascent. By 'lopyatvaf is signified that, they can be glided

over. But a^ain (by being glided over) they gain strength. There are seven

amsakas. Suddha and vikrta, two in the full form and in .sdava and \ . auduvita Ni^dava. The amsas are ni ga ri. These are also the apanysas. Nysa is ni. Sadava is like Dhaivati. With pancama omitted there is sdava, Auduvita is with sa-pa lopa. Sa, pa are to be skipped (lahgharOya) in avarohana, aie non-amsas, and have a multiple "dependence. Ten amsas-three suddhas, three vikrta^, purrui, sadava and auduva.
" Now Sajakaisila. Sa, ga, pa are amsas. Apcmysas are sa, pa, ni. Nysa is gcL It is always ^ samptir^ forni. Madhyama and rsabha are weak. Since their weakness is aheady clear, the mention for the second time is for the emphasis of this feet. Ni, dh are more frequent. It has no suddhatva [it should perhaps be asuddhiatva], hence it has three purm

amsakas. KVfSadjocReva^. Ainsas are sa, ma,ni, dha. Apanysa are dha, sa. Nysa is mcL There is mutual sdhgaii of athsas. Sativa is through ri lopa. Auduva is through ri-pa lopa ^abha is samvdi of dhaivata in the sadja grma, where it (i.e., dha) is an indispensable note. Sa, ri, ga are strong. That sadja isstrong is clear from it being arnsa.Rsabha is very infrequent, but to check this grulhra is frequent in the mandra sthna. Arnsakas are 17^ eleven, four are purna, three sadava, auduvita are four. There is no suddha. About Kaisika, the explanation has been given earlier. By combining with it, there is the jti Kaisiki. About luEcya, it is said, since there
are northern performers frequently for it (i.e., it is popular in the North) so it is called uiScya. 'Vanti' means gacchanti. In udicya, i.e., the north also, it (the regional influence) is seen. A gita (musical composition) is often named through regional names because of its frequency or popularity (in those particular regions). Thus, Takkarga, Mlavapcmcama, Gaudi, Malavi, Kmbhoetc. (sjiow regional influence through their names).

Sadjamadhycmm. All are athsas in the Sadjcmiadhyarti. And the (same) seven are apanysas. Nysas are sa and ma. Sdava is through ni lopa. Others hold that auduvita is through ni, ga, lopa. When the other note are amsas, then ni is infrequent.''^ The sancra is according to

TRANSUTION OF THE TEXT: NATYASASTRA AND ITS COMMENTARY ABHINAVABHARATI 104

105

gndhra lahghamya here (this meant that even when the jari was

one's discretion.'"'^ It has seventeen ainakas. Seven are pu^ five

4am md five & sadja grama.

He concludes ttot these sevenya depend o.

Vet-113 : Hereafler, I 01 spedL of Ihe dependent (,W. Ex

five amsas. Verse-114 ; ?a^ja and pacama are said to be the two apanyasas

rendered as purna, ga, was particularly weak). Verse-123 : Madhyamocyav is with seven notes (i.e., it does not have hexatonic and pentatonic forms), i'acamo is the sole amia.The rest should be made shnilar to GndharoMcyava. Verse-124 : In P<!icam (jat), the two a/bsas are r?abha and pacama. It(pacama)alongwithri(wafe/ta)andn^ada arethe
and pacama is the nyasa.

Gndrashouldbethenjsa,andthefa^vaorhexatomcrenden g
is without.iyafeft. Veree-llS : The pentatonic rendering should be by the exclusion of ivi^-ni'^Thesetwc,shouldbe^ided over and (in the heptatonic form), the movement shonld ways be fro
to dftoivato. Thus,themovementofthenotesofGmiftan with

Verse-125 : The hexatonic and pentatonic rendering should be similar to Madhyam (Jti). The weak notes are fa(&a, gandhara and
madhyama.

Veree-126 ; Here (m this jati) a movemoit should be made between pmcama and nabha. A movement, though less fi-equent should also be
made from nda to gondftant _ ~

the nysa and amsa have been said.

^ ^ , ,j i, Venie-116 : The characteristics of Raktagandhan should be rememberedasthesameasthatofG5ndtn:Dvatoisstrongheie,

Vase-127:Now,iwcamaisknownastheamsaof Gndharapancana

0).Themovementmtheni (higher octave) was never to reach beyond

th

(even though) it is a Zopya svara and should be weak.

VeKe-117 : There is movement {saiicara) of gandhara and ga4ja (but) by excluding .i?afcfta. Madhyama is to be made the apanyasa

this jao. The nysa is gMm Mid it (this ah.ys rei^ heptatonic ^). Tbe sakcara dHUd b. (like that) of Pancam, and

ira (saptaka) sadja. Verse-128 : l^fabha and pmcama are known as the apmyasas (of

here t Ven-118. .nd madhyam, shd ta k. a th. m r GnmroScyca. There la no p.tatoni. nmdering hen., and the hexatonie rendering is by the exctasionrfif_ Verse-U9 : The aniaramaTga, nyasa and apanyasa so e similar t?W-^."ltisnot.end.rediththepentatn.c torn. Vers.-120 ; The asas of Madhyani (Jffl) are by exctadntg n<mra and nida (I.e., they areso, ft w ')These are the only apanyasas andthe only nyasa is mMyama. Vee-121 : The pentatonie rendering is to be accontptehrf by .beom.ionofg5andif5a.Tbehexato.iedengshonld
be done by (dropping gandfeara).

Gndhn (jtis). Verse-129 : Pacama,rfabha, gndhra and nifoda these four are the amsas of Andh. These are only the apanyasas. Vetse-130 : Gndhra shonld be the nysa (hew). Th. hexatonic dering is by dipping a^a. The. is a sancra or movement
between gandhara zxiAi^abha. Verse-131 : A movement of dha and ni should be made and there is an orderly movement upto tiie nysa (m tiie order of the amios). There is droppmg of ^adja here (for hexatonic rendering), and there is never 3ny penttonic rendering* Verse-132 ; The nysa, apanysa and amia of Nandayan m due order, are always gndhra, madhyama and pacama. Verse-133 : The hexatonic rraidering is by droppmg sa4ja and its iahghana. Its smcaraa or movement (of antammrga) should not be

Ve.se-122 = Here (in tbis/aO, prolific " t" "'''f " nd oJC<.The performer, or th. Directors shold.iysmak.

106

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

07

like Andhi. There, movanent ot tic rfObha in the mandra maj be

sloped {Icmghanam).

skipping' means, that, iitthe purtivastK one* proceeds from rsabha to

dhaivataJ^^
Thciniovement Of-all other notes,(is with respect'to amsa or, nysa,) 'he' says. Others explain this by saying, 'it is the sphere of fhe amsa'ar the nysa bf the svaras'; In this, there are thirteen atnsas, suddha and vikna, five punta, four sdava. In the fifth, there is an exception. In the auduvita, when gndhara alone is the ainsa, then it is accepted. Now Raktagndhri. Five athsas, sa ga ma pa ni. ^dava is by ri lopa. Auduvita is by ri-dha lopa. What is then the difference ? He (Bharata) says, that, dhaivata is powerful here, (although) it' should b weak, by being lopya. And rsabha have being skipped in moving^from sa to ga (and back), the two notes are brought togther and there is the combinatidh {melana of sa and ga). Apanyasa is madhyama.^^'^ There are twelve amsakas of this. Four are sativa, in pancama also there is an apavada. Three are auduvita. But in madhyama and sad -there is an

Verse-i34 : The movement m the tra or higher octave should never be beyond the (tara) fo^a.Gndhra should be made the graa and nysa always. Verse-135 : The amsas of Kmiran are to be remembered as fgabha as well as pancama, dhaivata and nigada. These only are the
apanysas.

Verse-136 : Pacama should be the nysa, and no svara or note is omitted (i.e., it is always pTirna) here. The special movement is of gndhra, whichshould move every where (i.e.,be associated with the other notes). Verse-37 : Kaisiki ios all (the notes) as amsas, except rfabha. These only are the apanyasas.Gandhara and the seventh note, {nigada) are the nyasas. Verse-138 : On dhaivata and nigada bemg the (ruling) ainsas (of &s jti), pancama as nysa is desirable, ^gabha, too, sometimes may be made the apanyasa. Verse-139 : The hexatomc rendering is by droppmg/yoAfca and the pentatonic rendering, here is by the exclusion of dhaivata and rgabha Nigada and pancama are strong (notes). Verse-140 : Rgabha, here, is a weak note and lahghana should particulariy apply to iL The rolesimilar to ainsa is done by others (other notes ihaargabha ?) m the accomplishment of the hexatonic form. The movement here (m this ti) should be shnilar to SadjamadhyamafitL Verse-141 : In this way should the Wise know ibetis along with their ten characterfetics. In whichrasa is eachj to be used, (this) I will speak of (please) understand. ,
Commentary i Now comniences another (i.e., madhyatna gratna)

apavada of them. Gndhrodtcyav. There are two amsas, sa and ma. Sadava 'is through rsabha: Similarly auduvita.^^^ Another in Sadjodicyav. Antaramrga is the mutual sancra of the two arnsa svaras. Apanysas are sa and dha. Amsa also is frequent by sadja. Rsabha, although lopya is infrequent (rarely used) in prnvasth. In the' mandra sthana, -gandhara is also much used. There are four athsas-Vf/o punta, two sadUva.^^^
Now Madhyanm. Amsas are sa ri ma pa dha. The same are

apanysas. Nyasa is ma. Sdava is through ga lopa. Auduvita on gd ni lopd. Sa and ma are much used s paryymsas. Since it is mentioned again it should be considered as emphasized. Gndhra being lopya is alpa. It is thus found {alpa) inthe prrmvasth also. It has eight amiakas, five suddhas, vikrtas as many as dava and auduvita.
Now Madhyamodicyav. Pacama is arnsa. It is always punta. The procedure is, as is in the case of Gndhraodicyava. Apanyasa are sa, dha. Bhulya with ma, ga Nysa is ma, one amsaka and punta. Now Pancfinu. Two ainsas are n and pa. Apanyasas are ri and ni.

'Atah paramitV Now here of Gndhan. Sa ma ga pa ni are ainsas. Sa and pa only are apanys. Gajs nyasa, Sdava in by omission of ri. Auduvita is with omission of ri, dha, Pancama is alopya i.e., not omitted in this (i.e., madhyama) grama. 'Their

'

' Yso
,

Nysa is pa.

108

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND TTS COMMENTARY

109

It is implied that wha ga is omitted, thae is sadava and when ga and m are omitted it is auduva as in the case of Madhyni. Due to sadja ' 181 and madhyma being agamias and gandhara being lopya their alpatva is proved, the motion is for (showing) it to be even more weak. There is a mutual sangati of ri-ma. In the prmvasth there is a movement from gndKra to nida but it is limited.^^ Six amsakas are uddha, two punm,two dava and one auduvita, and on rsabha (as) amie there is an apavada of dhaivata. Now GndKrapancami. Amia is pa. Apanyasa is ri-pa. Nysa is ga. It is always purna. Movement is from rsabha to dhaivata. This is the saneara of Gndhri It has been said that, 'in the PcmcanH, there is sahgati of madhyama and rsabha and movement from nida to gndhra.' Both of these {sahgatis) are to be executed p.this {jti) also^^^. It has only one amsaka. Now Arulh. Amias are ri, ga, pa, ni,. The same are apanysas. Ga is Nysa. ^dava raidering is on the lopa of sa. Tho-e is no auduvita. 18^ There is sangati of ri and ga. 'Nyasa of nisda and dhaivata in orderly movement'. That which is the principal amia, from that a movement should be made. As has beai said, 'moving to nysa is in the order of the (enumeration of) amia'. Others say, the arnsa svaras are to be rendered in the sameorder in which they are read, uptothe apanysa svarcu So here are eight amsakas, four plirm and (four) dava. Now Nandayan. Nysa and graha are pcL Apanysa are sa and ma. Pa is amia, and is also the apanysa.Sdava is through sadja, which can be omitted or it can be glided over. Since it is mentioned again, it means it may be used (but), very rarely. 'The scmcarana (movanait) should not be like Andh. Andh is bom of Gandl and ArsabHi., Nandayanti is (bom) from (the Jtis) Pancanii, Arsabiu and Gandha This is propounded in the section on sarhsargaja (mixtures). Here, the pronouncement that, where there is similarity and comndon origin, the sahgati should be like Andh, is restrained by the principle that where there is a similarity of origin, the tis should have dissimilar sondati. Rsabha in the mandra is to be skipped. In the tragati, sadja is not'to be exceeded. The top seven notes (i.e., tra saptaka) are not to be touched \pras^h

should be spr^avyah]. Others say the sadja itself never goes up and, thus, tra is not to be touched. The tra (sa)-is optional. There is^onepnrhifl, one dava, thus there are two. Now Krmram. The amias ri, pa, dha, ni. The same are the apanysas. Pa is nysa.It (thisti) isalways prria. Others (say thateven t RS though purria) mm-amias notes are used less. Gndhara has a special sahgati. As Viakhilcrya has said, since the anarhias are prolific, one should particularly make a movement from all (the notes) to gndhra. (It has) four amias and is purna. Now Kaiiila. The amias are sa, ga, ma,pa, dha,ni. The same are th apanysas. Nysas-ese^a, ni. Whoi dhaivata and nisda are paryymia (optional), then pancama is nysa. 'Some tim even rsabha' (is apanysq). It is not so when it (i.e., the jti) has omitted (notes), but is so optionally in the heptatonic form (full octave), '^abha is weak', because it can sometimes be omitted. On its lopa there is dava. Auduvita is through lopa of ri dha.Msdaand pacama aie more powerful inrelation to any one of the paryymias. As has been said for Sadjamadhyam,(so also for Kaiiila) that there is free sancra when dhaivata is the svafmia. Since here (thisti) has six amias (not sev as Sadjamadhyani,hence comparatively) scmcra is a little curtailed. Some read the following Ary (ver^) to indicate the origin of the grma-rgas bom of the jtis. "As for the mixed musical piece which shows a variety of forms; it is delimited by the requiranents o thejti^ It shows, however, a wonderfiil variety of the sequences of alarnkras bom bf mixture and produces grt jtertainment". That which & mixed by being joined with the ten laksarws of the tyamias, that is miira. And geyti means that wiiich Has the characteristics of grtna-rga etc. Thus, if the amia is related to some jatyamiaka and the nysa is of anotho" and 1 RA the apanyasa of still another, then there is muh variety. Such a miira geya tends lo'belong to some jti and this is regulated by Matg, Nandikaiyapa, Y^lika etc., on the basis of jnuch elaboration of forms {fpalMhulya), (i.e., to say mixed fotms are attributed to thatti^ whose forms are most mmierous in thon). This is a mixed drink.?' It does not follow any order. So 'he' says in this much triixed ti, when, there is mixture or sankara,then numerous strange alahkras are produced. How

no

ABHINAVABHARATI

TRANSLATION OF THE TEXT : NATYASASTRA AND ITS COMMENTARY

111

are they to be respected ? The answeris, through their being very pleasing. Their idea is that its use is not too much in vocal etc. Dattilacarya also says, when there is sahkara, then jati is indicated by rpabhulya i.e., frequency of form. Thus Bhinnasadja {rage}) is similar to Sadjodicyavaii (jati ), because the forms frequent in it are different from those in Nandayanti. Dhaivata is amsa, madhyania is nysa. Rsabha, pancama arc absent. Hence the laksam of gram-ragas is spoken by the Sage. The authors, Kaiyapa etc., are also well known. But the totally ignorant cheat the world. So that there may be clear demarcation we briefly read the laksanas below in order. [Grama-rangaS arise from jtis through sahkara apparently. They have numerous alamkras. They are praised for their rdjana, which is primary in dhruv-gna, not in gandharva. Their jati is determined by nipa-bhulya. An attempt is made to demarcate the grama-rgas]. The two (grma-rgas) named Sadjagrma and Madhyamagrma arise from the combination of Sd-Sadjamadhyarn^^^ and Madhyamodicy^^^ respectively. They have ma as nysa.

Rga Suddhasdava is produced from Madhyam (jati) and has rm as amsa and nysa.^^" (Rga) Pcmcama has pa as amsa and nysa, weak ga-ni and is bom from the Madhyam and Pdcarri jtis.Kaisikamadhya is produced from the jtis Dhaivai^"^ and Sadja-Madhyani, sa is amia and ma is nysa. Suddhasdhrita [there is no such rga as Sdharita, hence, <it should be Suddhasadharita] is obtained- from Sadjamadhyam (jti). It hs sa'amsa, ma is nysa, ga and ni are alpa. Kaisika (rga) is obtained by (the mixture of) Kaisiki and Krmravl (jtis). It has,sa amsa, pa as nysa and is purna.^^'^
Thse seven belong to the uddha giti.^^' Bhinna-sadja is obtained hotti.Sadjodicyava, has dha as amsa, ma as nyasa and pa and ri are omitted. i^^.fi/iiVinatona''' [Bhinnatgah shoul he, Bhinnatnah] has pa as amif; ma as nysa and is obtainedfrom (the mixture of) Pancanii and Madhyania (jati)i Rga Kaisikarnadhyama^^^ should be knpwn with sa amia, ma nysa and.weakga.and-jiL It is obtained from Sadjqmadhyma(jiit (Rga) BhinnapacaM .has^haasmsa,pa as nysa, weak n.aiid^ises,fron> the Pancaim and',Madhyarri jtis}^'^ Bhinnakaisika is bom iTom (jtis)^KrmfavJin<^ Sadjamadhy.am}!P^Sa. is amsa, pa is

nysa. These are regarded as the five types of Bhinna gJtis^"^ Gaudapancarm is bom of (j'tis) Dhaivati and $adjamadhy, pa and ni are omitted, dha is amsQ, nia is nysa. Kaisikamadf}yam has sa as amsa, ma nysa, pa is omitted. It arises from the Dhaivati and Madhyam tis^'^^ (?) (R^Xisi^kaisxisa as amsa and pa (as) nysa. ifs (bora) from (jtisy JSWZESS ctncf Sadjamadhya?'^ Ail these three (i.e., Gaudapantama, Gauda Kaisikamadhyarmand Gauda Kaisika)are of the * ^ 9ns Gada giti Sauvira arises from ^djamadhya (ti). Takka has sa as amsa, many types of (note ?) combinations, sa as nysa and arises from Sadjamadhya?^ Sauvira omits ga and ni. Mlavavesara arises from PcmcanH and Madhyam (?). Arnsa and nysa are pa (? not clear).^' Ga and ni are omitted. Vesarasdavahas ma as amsa and nys. It arises from Sadjamadhy and ga, ni are omitted.^^ (The sourceyat of Botta are not clear).^^ It has pa as arnsa and arises from the ri (i.e., ArsabHi) and dha (i.e., Dhaivati) jtis.Takkakaisika has dha as arrisa and nysa, weak ni ga, and arises from ma (i.e., Madhyarria) and dha (i.e., Dhaivati ) j'tis.^ Mlavakaisika which arises from Kaisikihas the mie of sa as amsa and nyscP'^^ There are eight types of Vesar gitis or styles?^^ Bhammampdcama has sa as amsa, ma as nysa, we^k ga ni, and arises from Sadjamadhyam (/aft")^''* Sadjamadhyam and Naimdi are the source (tis) of RpasdHra^ '^ which has sa as arnsa and ma as nysa. Gndhrapcmc'ama (raga) is bom of the gandhara (i.e., Gndhan and Raktagndhn)tis?^^ It has ga as arnsa and nysa. The three sruti note is weak (here).^'' Revagupta has ri as amsa and ma as nysa. It is bom of ArsabHi (j'tif^^ and is devoid of sadja?^^ Two sruti notes are weak (text missing). Sadjakaisika^^" arises of Kaisiki and has * ' * rr . "^221 sa as amsa, Sakaparicama is bom of (jatis) Sdjiand Dhaivati , it has sa as amsa and nyasa and ni and ga are weak.^^^ Thus, it should be remembered that, Sdhram (etc.), are the nine [should be five] gitis (or modes of singing). These two- bh and Vibhasa gitis being dependent on grma-rga, according to received opinion, are (included as) fomiing _ _ 223 the seven gitis for use in the dhruva-gana. It is for the summing up (scmgraha) that this Arya has been read by
some. But it is not to be found in our manuscripts. It is not that without

112

ABHimVABBASATi

the Ary all this would remain imsummarized, for the wdiole has already been accepted in the (treatment) of the principal notesof themelody- types (jatya^kas). There, too, the Sage has permitted a variety that may be useful may be utiUzed?^ Saying 'the onployment of the four notes in the avakrsta dhniys^^ is to be done here,' He has implied that the variety of alamkras is to be classified later.^ Summing up the sense of the chapt, 'he' iatroduces a new subject. ''Thus these (are to be known as the tis with toi laksanas)". That is to say, in these^ the ten laksams^ are never absoit In which rasa they are to be used, with wMchainsa, nysa etc.,that is(part of the charartfrof these) tis, 'As is-being stated by me,' shows self disparagonoit. The idea i, the word of the authoralone shouldnothe regardedas pramna, since the Sage Kaiyapa etc., have already so formulated it. Abhinavagupta,'whose affliction (heat) has beai removed by resorting to the lotus-feet of the (God) with the cold-rayed (moon) as hiscrown, has thus discussed the secret chapter of the tis.

NOTES

Bhaata memtions here,.a four-fold classification of musical .instruments viz., tata (cprdophones), avanaddha (membranophones), ghana (idiophones) and susira (areophones). Abhinavagupta is known as the greatest master of Khmra ' r* Saivism designated Pratyabhijna dariana or Trika darsana. It was a philosophy of monistic theism or liavardvaita. It conceives ultimatereality asconsciousness, andconsciousness itself as the unity of praMa and vimara. That is to say, that consciousness not merely apperceives, but also determines its content. Consciousness is not simply passive, it is active and creative. This active or creative aspect is called vimara and manifests itself at many levels in the process of creation. Kal has many meanings. It stands for a part or digit tM- time-fraction, expressive power, esp of sound, limited power of the am, suprane power (= amttara kal). Although in Abhinava's order / / of idaitifying the NS chapters with Saiva tattvas, Kala here should stand for the kemcukaso nailed, itactually seems to stand for the expressive power of Vk or Nda pf which Bhavabhti

2.
*

says "Vandemahi ca tm vcam awrtm tmanah kalm". It

thus indicates the basic creative power pf consciousness. This power expresses itself first, in the form of Nda or sound which has four sta^ges Par, Payati, Mdhyatm, and Vaikhap. In its original form, sound is a rhythmic but unstruck vibration in anptines? otsunya. In its suprrahe form, ASda is indistinguishable from consciousness, but in its lower form it is the audible sound. In itsmusical aspect, ithas asubtle andgross aspectcalled iruti and svara respectively. The middle point of the svaras is the madhyama svara,regarded as imchanging. It may be recalled that ma is a well known name of Siva andso Abhinavagupta, the

114

ABHINAVABHASA 77

NOTES

115

philosopher, poet, musician, imagines that the unchangeability and beauty of madhyama arises from its connection with Siva through Kal and Nda. The verse, thus, says that from the form of Siva as Kala arises Nda, of which srutis and svaras are aspects, and in the sequence of svaras, madhyama shines by its beauty. It may also be noted that the gndharvapadas appear to have been mostly in praise of Siva and that madhyama had a special importance in this system of music. 3. Music helps the spectators to move away from their soise of eveiyday reality into the spectacle. Psychologically, it creates a partial disassociation of the mind. Abhinva, here, mentions two types of vTns \az., drann wn and the gatra na. The dram v7n is the actual wooden hite and the gtra nm is the human throat He refers to the dorow v5n5 as a pratibimba, obviously of ibe gtra vna or human voice. Thus, the gtra itm is of primary iinpotrance. This, he reiterates by saying that in gndharva, the performo" reaps the merit, hence the importance of thehuman voice-'Gndharve'pi (hi) prayoktuh phalan

associated with the gTtakas,not with thejatis,and, were probably a tla structure basically, but the Kapla as described as above is clearly a svara structure associated with the jatis." Lath, M., Astudy of Dattilam p-139. The gJtakas WCTC major tlastructures of gndharva. Jtis were svara or melodic structures analogous to' our present day

rgas.
7. Abbinava he speaks of 'vdyntarsy', i.e., of instrumoits other than the ones mentioned - i.e. tla etc. What could these be ? Why have they not been maitioned ? The reply is 'tasrnllaksanam na knamityha', i.e., 'their characterisation hasnot beentraditionally made, it is said.' Now, this is an int^estingrefCTence. We hear of instruments other than the four aforesaid, which have not been mentioned by the / Sstras. The tanptation is irresistable, that these may have been instruments current among the people, that they constituted the folk instruments of the times. Since they wae current in society. Abbinava must have felt impelled to refer to them. Since, however, they may have hem rudimentary and not described in traditional scriptures, he did not feel it necessary, to give a separate and detailed description of them. He just says that they can be included in either of the categories-stringed, percussion or wind. This inclusion may be either as an ahga or limb or as an auxiliary (upayogi). Of the first kind, he gives examples, but defective published textmakes it difficult tomake out the names of the instruments. 8.

4.

tadanu^rena sanraprdhnyam phale darsitam'.

In the same paragraph, he indicates that tata-todya or stringed instruments are connected with svaras and hence have been mentioned first. 5. The three elements of gandharva were svara,tla and pada. The first two were primary and pada was useful only in so fcH" as it formed the base. This tendency was perhaps inherited from Saman singing. The Sman singers did not attach much importance to the intelligibility of the hymns, bt distorted the words freely. Svara arid tala wCTe the essence of gndharva and from their proper practice accrued transcendoital' merit or adr^

phala.
6. The composition of brahma-gtas hasTjeen ascribed to Brahma by ancient texts. Dr. Lath explains these, as well as some others viz.,^ Gth, FniK and Kaplas and some minorgndarva forms to be of sacerdotal character. "The brahma-gita has been

Tata meant stringed instruments, chiefly the vin. Susira meant areophonic instruments, chiefly the vamsa or flute. Other subsidiary instruments of this category were the ndi, tnava, saikha etc.Percussion instruments have beemclassified into two basic categories by Bharat-avanadd/ia and ghana. The former
are membranophonic instrmrients or skin vibrators in which the sound waves are due to the vibrations of a stretched skin or jftembrane when struck. The latter or ghana are idiophonic instruments or self - vibrators, i.e., instruments of solid sub-

116

ABHINAVABBAAH stance, which havea sonority of theirhave own, wiiich isemitted in waves when they are struck.

NOTES

NY As Abhinavagupta explains, though the song, acting etc., are p^ceived through different senses and consist of divergent actions, care has to be taken to unite them into one intelligible form. Though music is pCTceived by the ears, mime by the eyes, the mind is the unifying factor that prarceives them as one intelligible whole. The kanel of the idea is that the grops of vocalists, instrumentalists andstage-performers (actors) anddanters ar^e to work in unison, harmony and continuity.

Tla in gndharva music was played on ghana instruments, whichconsisted of cymbals, graerally of btonze. Because of the hardness of its form it could iiot play a varirty of varnas used for expressing rasa and Ava (\diich were the essence of dhruv gJtas). Hence, it was not useful fw dhrum gna. Tla or ghana wasonly used forestablishingsmyaingmftarvamusic. Smya indicated neutral 'balance' or 'equipoise' betweai Sa and svara structures. In another place (AB <m NS 31, 1), Abbinava says"ghana vdyacan produceonly asingle monotonous sound. Unvaried in nature, and, is used only to keep the measure in the tala by sounding at the proper intavals"-'Tatv hanyata iti ghanah Kathihataikampah tata iva namanonama(nma)ruin(na)mai (ai) thilydiyogbh(va)d(iksaravaicitryam svaravaicitryam cnuruddhyamnanmrempayog kmsycdlikdintcyate'. It could not be used for expressiveplaying, as inavanaddhn instnimpaitg THfa in dhrum gna was played on avanaddha instnmients. These
membranpphonic drums were capable of expressing a wide variety of vornos or 'bok'. Thesedrums, through the pliancyand resulting inflection^ of the taut skin on whic they are played, produce a numb of no^ t different pitches as well as a wide variety of diffi^nt expressive sounds,. 9. 10. 11. The word 'kutapa' indicates an.opera of vocalists, instrumentalists and act(xs. There is a yariant reading 'ttridesasansrayah' or 'sainritah' for 'samudbhavah'. Does desa leiet to regimOT spot? The.idea is that singing dq)ends on voice, instrumental TTHTOV arises ftom different types of instruments, acting dq)aids on the movemmt etc., of actors. These difiicrent jdimomeoa arising frwn different sourcesand in craistant movrao^t have toappear . like one continuous whole "whidi produces the niprffRir>iti of harm(my and form, just as the rapid whirling of a fire-brand produces the illusion of a fiery circle.

12. 13; 14. 15.

That is, song and instrumaital music areaccessories to the action of the play. That is, the difffCTent notes should appear continuous.

Vdya hwe seems to cover the body (gtra) also.

Abbinava first raises an objection, saying that many people thhik, that plays can be shnply read and that tho-e is no need for orchestra and music; ThCTeafter, he negates this doubt by saying that, in this way plays are mcomplete. Song and mstrumental music are necessary. The varios tala structures of gandharva music were classified into two basic groups, viz., the trayara (i.e. those structured on triple grouping) and caturara or those structured on duple arrangement (NS 31,7). The two formula names for the caturara and trayasra tlas were caccatputa and ccaputa (ibid. 31, 8). These names do not have any significance in themselves, but when separate, split up into single syllables of guru, laghu and pluta, they give the basic patterns of beats of the trayara and caturara talas in the ekakal tempo. Evoi now, one may note the triple groupmg or trayara in tlas like ddra (6 beats), ekatala (12 beats) and caturara grouping in talas like Keherava (8 beats) and tritla (16 beats)

16.

17. 18.

Ranga refers to the stage, as well as the diversely standardized opaiing or preamble of the performance.
Abhinavagupta interprets atyarthami^am not as 'excessively desired', butas 'tianscendentally sacrificed'. The offaing in the

118

ABHINAVABHFAT

NOTES

sacrifice consists of sense objects sucbas musical sounds, spectacles etc.. When something is offwed through the fire, it is sublated. Nevertheless, there is a transcdenb effect, both for Gods as well as the sacrifices Similarly, here, with the offermg of sounds and spectacles it is likmed to a sacrifice producing transcendent effects. This apparoitly follows the characterisation of dance by Klids^ as a 'visual sacrifice' - 'devhmi-

22.

danmananti' munayah Mlavikgnimitratn. 1,4.

19.

sntafn

kraturn

cksusam'

The word gndharva scanetmes stood for music, in genoal, but also had the techincal soise of a particular system of music. It is i lliis scuac, Uiat the word invariably occurs in the NS and its commentary the AB. Gna, too, may mean singing, in gaieral, but here, it has the sense of a particular type of music - dhruvgna or theatrical music. Gndharva was ritualistic music, dear to the gods. Its performance was difficuh and needed concentration- 'avadhna' {Dattilam 3). Hoice, thede^ hmnersion and concaitration in gndharva music is likoied to one being totally imm^sed and subsumed by a fragrance. Well performed gndharva music resulted in adi^a phala or transcoidental merit, hmce Abbinava says that the 'result accming to the performer is pimary'. On the other hand, gna had a drsta-phala, to produce rasa, thereby giving pleasure to the audience. Thus, here the primacy was with the audiaice. Hence, as Abbinava, further says that grui should not be idaitified with gndharva.

20.

Geya is the relevant group of words which are to be sung. This

lyric is produced by the human throat. The instrumentsonly help to enhance the beauty of the lyric and are thus subsidiary. (See Acarya Brihaspati Ntyas^ra-Atthaisy Adhyya, p.11, pub. Brihaspati Publications, New Delhi, 1986).

21.

M.M. Ghosh has taken gtra here as a variant reading. He thus, interprets verse 10 to mean that "the source is the himian voice etc."

Here, Abhinavagupta tries to trace the origins of gndharva music. Earlier, he had said, that from Sman came gndharva. The gndharva form was apparently intermediate between later Sman forms and the gna form of singing described in the NS. NS 28,10 gives the genesis of gndharva as gna, the vT5 and the vamia. This, in itself, would be quite cryptic, but Abhivana comes to our rscue. Gna, here, he explains are the songs of Sman, i.e., Smagna. The other two influences were the vaia or flute and the v7n. The v7n, here, is the audumban v7n used in the Mahvrata ceremony. This ceremony pertained to the preparation and offering of Soma and constituteda festive, albeit holy occasion. Maidens bearing pitchers did a circulardance, the foot-movement of which was in concordance with the rhythm of songs known as gths. On this occasion, the ladies played the audumban v7n. Hence, even though the matrix of gndharva music was the formal, ritualistic Sman singing, yet,the mention by Abbinava of the audumban used in Mahvrata ceremony is significant andconnotes"folk influence. In fact, percussionmusic too, it seems, is the offspring of the popular music of the Vedic times. Percussion music was, then, incorporated into the gndharva forms, whaein, it was developed and elaborated into complex tla stmctures. It may be noted that Sman chanting shows noevidence of tla.Thus, though Smanmusic must have given the basic structure (the rigid, traditional, ritualistic aspect) to gndharva,folkmusic too, musthave helped its development. Acarya Brihaspati {op. cit.,pp 13-14) opines that the audumban vin was the precursca" of the Mattakokil (21 stringed) v7n. What, however, is not acceptable, is that Acarya Brihaspati takes the vama referred to by Abbinava, as the vamia of Mataga muni. In fact, he (Brihaspati) says that the genesis of gndharva was 'Smagtti, audumban nn' and the 'varnias of munis like Mtaga etc. The time of Mataga is approximately the 7th cea A.D. This was the time of the efflorescence of dhrum-gna and Mataga himself was the author of Brhadde, a text on deii or rga music. It can hardly be, that his vamia contributed to the

^20

ABHINAVABHARATI

NOTES

121

genesis of gandharva, which was akeady in vogue about 1st cen BC - IsTCen AD, the approximate date of the NS. 23. The mention of a popular tradition of songs within the brahmagta is highly mtriguing. Actually, we must remember that rites and festivals must have been concurrent in the remote past as they are now, with the result that there would be a kind of cultural counterpoint in the folk singing of the festival to the ritualistic and formal singing of the Soman. It seems, that brahmagita was coimected both with Sman and folk traditions. It may have belonged to the fonner and also inspired the latter to creative development which helped the rise of gndharva. 24. A distinction is apparently being made out between the melody which is tamed Sman, and the words of the song which are sung in that particular melody or time. The commoitator apparently held that the word - combination was primary, thus making instruments secondary. Abbinava, on the othCT hanH^ seems to argue, that melody and words qualify each other. Sri M.M. Ghosh has interpreted 'gndharvam trividham' as three kinds of gandharva (p.3., footnote). "The three- kinds of gndharva seem to be three kinds of musical performance in which individual notes, beating time, or songs respectively play their principal or only part". Actually, svara (notes), tla (Aythm and time-measure) and pada (words of the song) are the three constituent elements of gndharva. Gndharva is a synthesis of these three. 26. Refer to note 19 for avadhana. See also Lath, M., Op. cit. p. 5, also pp. 192-93.5vara, tla and pada in their totality constituted gandharva, and its performance could, not be done without concentration - avadhna. Thus, avadhna, was a part of gndharva, and not separate from these three elanents. Abbinava remarks that in defining gndharva, it is not sufficient to state that it is a synthesis of svara, tla and pada, because this implies an equaUty of status for all three elements.. The fact is that, a hierarchy exists; svara is the dominant elemrat, tla 29. 30. 31.,

follows it next: its function being chiefly to give an equipoise and equilibrium (smya) to svara, pada comes at the bottom of the list. Bh^ta hasnot expliciy set forth this hierarchy,though according to Abbinava, he has imphed it by listing the three elements of gndharva serially as svara, tlaand pada. Toprove his point. Abbinava quotes Dattila's definition of gndharva, which, besides implying a togetherness of svara, pada and tla, also suggsts a hierarchical relationship between the three. Even now, in classical music, it is thesvara and tla which are priihary, the words of the composition are firequently distorted and serve only as a basis for the exposition of the rga. In contrast to gndharva, in the dhrum gna, pada is the most important, since it the pada which would draw attention to, as also enhance the dramatic situation. 28.

25.

Dhruv gna was used in the theatre and avadhana was not really relevant to it. However, the purvarahga or tlie preamble of the play consisted of some rahgapj and had elements of gndharva. Here, possibly, elemraits of the complex gndharva tla were played on thedrums, which needed intaise concentration or avadhna.Bharata does not specifically defineparivarta,
but he mentions the term in certain places in the sense of repetition. Perhaps, it was comparable to the' present day avar-

tanas of.tla ?
i.e.,the function of designating a certain note as graha, amia, etc.,- all these which go to make a melodic structure of jti. Much of the text is missing here. Bharata, in his list of topics relating to svara, mentions only seven items as relating to the rTrT vina or the human body as the source of vocal music. Tna has not been included in it. Abbinava explains this by saying that, even though, mna is possible in vocal music, it is inconvenient. It may be ud only in vocal {wactice. Even then, one has to tate the help of the vm. which is capable of producing the exact number of notes required. That is, even when one is not able to produce the notes

27.

122

ABHINAVABHARAH

NOTES

123

from the voice, one may, by having the distincticm of notes comprehoided by the heart, be able to produce the exact notes of the tona on the vna. It seems, that though tana was possible in the sa,it was actually more popular on the v&w. It may be noticed that the word tna derives from &e root 'iqn' which means 'to stretch' or 'pull'; perhaps this referenqe to stretching or pulling of strings indicates instrumental playing as the source of tnas. 32.

'

could be discerned, for it was not conceived in terms of any mathematical ratio. TTius, the rutis were held to be audible and practically useful, particularly in the case of instiimftntal music nd more so. when vinos of th harp type were used. NevotheleSs, no strict mathematical measures were relied upon. The tuning itself depended on the tension of strings rather than the control of their available length.

Suska refers to meaningless words like jhantt-m, digle, digidigi

etc. These were used for fulfilhnent of the metre, but could also be set to metre independaitly too. TTius, these WCTC called 'bahirgta' or 'nirgiia\ Bahirgita was a gta with meaningless words. According to Abbinava (AB on NS 31,323), bahirgita refeired to upohana, which meant the preludeto a melody. These nonsensical syllables were rraidered both in vocal and instrumental music. However, they were more popular in the latter, and were know as dhtuiuska. On the v7n5, meaningless syllables like jhanttim, jagatiya formed the basis of making different strokes {AB on NS 34,33).

34.

Pada is one of the constituent elemoitsof gndharva. It consists

of the words of the melody. Bhi^ta, and following him Abbinava, ennumate all the linguistic elements which relate to pada. Thus, they speak of grammar and syllables, coiisonant, vowels, halanta syllables, prefixes, suffixes, sdndhis, vibhaktis etc.

35.

The various tla suctures of gndharva were constituted by basically eight types of beats, four unsounded and the other four sounded nes {NS 31,30-31). The former consisted of vpa, niskrma, viksepa and pravea, the latter wCTe amy, tla, dhruv and sannipta. In gndharva, these beats wore denoted by some typical gestures of hands and palms and one could not deviate from these prescribed movements. Thus, these gestures seem to have had some ritualistic significance. Avpa was the folding of fingers with pahn facing upwards {uttrihgulasahkoca ibid 31,30). Niskrma was the adhogata pahn with fingers extended prasararmt (ibid. 31, 33). Casting of this hand to the right was

33.

A ruti was a cognisable musical microtone; however, it was not sung or plajred per se. A ruti was a musical interval; it functioned as a unit of measurement of svara or note, and a svara was supposedto consistof a specific numberof rutis - two, three o;^,four. It may be noted, that, Bharata does not mention ruti in connection with the nn vTn, but instead relates it to the dram. Abbinava thinks, that rutis are useful for relaxation and tightening of strings in tuning. Agin "the nais are clearly illustrated on the vTna" {AB on NS. 28, p.l3). '"Die rutis are described only to divide the grma-" (ibid.p. 12). TTie purpose of rutis was to subdivide the octave into twenty-two micro-intervals and, thus, it could be of help in instumental music. It is significant, that the most fundamental aspect of rutis viz., the pranina hiti of the standardized ruti is described in the 'context of the tuning process of the vTn. In fact, that is how it

viksepa {Lc.) Drawing in of the fingers stretched during viksepa pravea (ibid. 31,34). Of the sounded beats, amy was the beat sounded with the right hand, tla with the left hand and
sonmpofa with both the hands (iWd 31 36-37). OArav has been described by .Sorngadeva, as a movanent of the hands after tLe snapping of fingras {SR 5,9 Adyar ed). 36. Just as present day tala ismeasured by nitrs (foreg., ektla* 12 mtrs, teentl^l mtfs etc.), the basic time-umt of the ancient tlas was, what was termed kal. However, one kal

ABHINAVABHARAH

NOTES

125

measured five nimesos, and not one nimesa (the time ta Iren to uttCT a shbrt sanskrit vowel), as the presait day nitr. One unit of two kalas in the dvikal mode was termed pdamrga,^and four sach pdairirgas made a largo: tala unit called mir-this was quite different from the metric nitr (equal to one nimesa) spoken of above. The gTtakas were the major tola structures of gandharva, and prakarana was the generic name for the seven gltakas. The torn vastu though used proUfically by Bharata, is, howere, not defined by him. A vastu was one of the main constituaits with which the structure of the gJtakas was built. It consisted of nitrs (not the metric one, but the larger tola unit). Accordmg to Dattila (Dattilam 139), it consisted of sk, three or one and a half such mtr units. Abbinava gives a two-fold meaning of ahga, as tJhga and varnhga. The fomier, as described by Bharata, are of three types, viz., vivada, cafoi and vma. Bharata says thatthese were formed within the vastus and avayavas (limbs) of the gJtakas (ibid. 32,190-91). Bharata defines vidri as the consummation of a pada or varna. Abbinava terms vidn as a gitakhanda, i.e., a musical segment, and the division seems to have been made on the basis of tla. Bharata speaks of three tempos (laya), druta or fast, madhya pr middling and vilambita or sl^w. These corresponded to the three mrgas; citra mrga (druta laya), vrtta nirga was in madhya laya and the daksina mrga related to vilambita laya.

samapam, avapni and uparipani. If the laya of the tala and that
of the melody played or sung were in harmony, it was termed samapani. i.e., their beats synchronised with each other. If the melodic stresses preceded the tala beats, it was termed avapiti, and if they followed the f/a-beats, it was teimed uparipni. 41. What Abbinava means, is,that one pointof view is that vyanjana (cotisonant) indicated meaninglesssyllables, theusKksara like jhanpm etc. But, since earlier (see note 32), the uska has berai understood as basically connected with instumental music, this meaning would be out of place. It would be more appropriate to connect it with the grammar of the pada.

42. 43. 44.

Raktagndhn wasa madhyama-grmiljti with sa-ma movement dominant. See note 6. ,

Prakarana wasa general name for the seven gitakas, also called saptafpa.They wCTe majorrZa-structures of gandharva music. Madraka was the name of one of the seven major Sa-compositions. Both are related to the taldhyya.
The gandharva system of music admitted of only seven notes. Apart from these seven, the only others that were permissible were,faiZr nisadaand antara-gndhra. Both had the nature of being slightly sharpoied notes, being two rutishigher than their regular intervals. As they were modified froms of the regular gndhra and nisada, they did not have the same status as the seven pure or avikrta notes' and w^e subsidiary to the latter. Hence, these were known as the vikrta notes or the two sdhrana notes.

45.

Yati was theregulation of speedor tempo of words of varnas, i.e., syllables of both vocal and instrumental melody. This was regulated by three types of laya. The one which had the same tempo in the begining, middle and end, was san. Srotogai, like the river current, began with a slow speed and gradually accelerated to a fast one. Gopucch was the reverse of srotogai, commencing as a/ast laya and concluding in a slow one. Pni related to the synchronisation of tala, and the tola stresses
of the vocal or instrumental melody. TTiis was of three kinds -

KZikai nisada and antara gndhra could not beemployed as ama in any jti, since they were only modified notes. Moreover, the sdhrana notes were to be used only'in some specific jtis and there, too, in a limited fashion; Madhyani, Pancanii and the Sadjamadhyam were the only three Jtis in which the svara-sdhrana could be used. However, these two svaras were used proUfically in dhruva gna.

^20

ABHINAVABHARA

NOTES

127

46.

The idea is, that on hearing, the miad is absorbed in sound and assumes its from. Normally, this state of sound perception does not affect the emotional neutrality of the mind. Musical notes destroy this neutrality anddraw the attention of the mind to their own specific beauty. Here, the svara is defined as arising from a transformation of a srutL First, we have n interval consisting of a fixed number of rutis, vthin this interval, att fixed point, the sruti is magnified through a slight force of breath, and the resultant note is smooth and pleasing. The essence of svara, is, not just its smoothness or sweetness, but the fact that it is characterisd by resonance. In fact, this is what ruti lacked and hence it could not be sung or played per se. The rutis have fixed places; \diai some out of them are struck by an hnpact and a particularly pleasing and continuous resonance is produced then we have a svara. Svara, thus, is of the nature of resonance {armraimtma),which has continuity and sweetness and is produced from the magnification of specific rutis at the end of particular intervals. Cf; Lath, M, A study of Dattilam p. 210, wiere snigdhatva and raktatava are not distinguished. Snigdhatva seems to indicate the unbrokoiness and continuity of resonance, whereas, raktatva seons to refer to ils accoustically pleasing effect Dr. Lath opines that the resonance of the svara is to be undrastood within the law of harmonics. Abbinava criticises the doctrine which regards the vTn as an image or reflection of the human-body or voice. His argimient is that, (1) the v7n5 is not n image at all. A mirror shows the likoiess of the body reflected in it VJn is not percived as such a reflected likoiess. (2)Nor is thevin pCTceived Ukea reflecting mirror m wdtich a likeness may be reflected. (3) Nor is the vTn a reflection or echo of the note produced by a human voice because the two are perceived by different senses; the v7n is se^ and the note is heard. If the meaning is that the note of the v7n is not Utei^ly an echo of the note produced by the humanvqice, then the reason given should be interpreted to mean that

the two are perceived by two different acts of saisatn. (4) This second interpretation mentioned in (3) is given as an additional argument by saying that thesame sound does not travel in space. I The percepticm of sound was a subject of debate in Ancient Tnifian philosophicalSchools. Thetwo theoriesgenerally current were tamed (a) ci-tarahganyaya and (b) kadamabogolakan^ya. In both the theories, the original sound is the cause of further sounds, and thus a senes is produced of which the proximate membo: is perceived. The nature of the series is disputed. For some, it is like a soies of waves. For others, it is Uke the diffa-ent concentric rings of a kadamba flower. Tho-e is a further difference of opinion on the subject. Some Uke the Vaiserikas and Buddhists believed sound to be impermanoit and momoitary, while the M5na/sakas, Vaiykaranas, and Vedntins beUeve sound to be' eternal and consequentiy that it is only air-mpvemoits which are produced, through which sound is mahifested. In denying the translocation of sound in the ether. Abbinava, perhaps, has in mind its real indivisibility and pahaps the idea that when an echo is heard, it is the original sound itself that is heard. Hence hearing the human voice andhearing the v7n correspond to diffemt acts of perception with two different objects. 50. Bharata, and following him Abbinava, hold svara to be primary. The other school of thought, led by Dattila, believed ruti to be the basis of svara. Only certain specific rutis out of the twenty two, which weremusically relevant attained the status of a svara, thus making the svaras dependent on the rutis.

47.

48.

49. '

51.

'Pu^asdKrane kle kokil vakti pcmcamam.' Nradiyd Siksa. 1,5,4.

Then again j / _

'Urasah irasah kanxKdutihitah pancatnah svarah' ibid., 1,5,6.

'

Here, the identification of the svara, is in terms of some detenninate pitch, rather than relatively to therutis; because the

ABHINA VABHARA fi

NOTES

129

srutis; cannot be connected either to the seasons or locations of

the body. The reference to cuckoo i the spring is an attempt to indicate the absolute pitch of the pancama. This'procedure would be irrelevant if the pancama was to be dfmed in a fixed order of srutis, and the sruiis were available independently. 52.

55.

The idea apparently is, that the property of musical charmshould belong to the soimd produced by theimpact of the breath on fixed places of the body. Tradition believes that scorpions arise from cow-dung. Abhinavagupta is arguing, that, the true musical property belongs, not to the sound consisting of mere phj^ical vibration, but rather to a kind of resonance. The physical soimds seem to manifest musical property of which it is the immediate locus or form. The physical sound isproduced, while the musical property of resonance is manifested. Themusical property has a necessary psychological aspect. It can be recognised by the mind as an ideal form, even when thesounds manifesting it are different. In essdice, Abhinava's concq)tion of svara, may be expressed as 'dhvani-vyahgya-dharmaviiesa', and readily reminds one of the grammatical notion of sphota. In fact. Abbinava declares the svara to be resonance and compares it with anusvara. Musical property, thus becomes a supervenient ideal quality. It may be compared with poetic dhvani which was accepted for non-verbal media also, vide Dhvanyloka. The charge was, that irutis and svaras presuppose each other. Without a fixed starting point in svara, we carmot count the irutis, and without coimting the irutis, we cannot determine the svaras. The answer seems to be, that, svaras are recognisable per-se, and the srutis serve to subdivide the octave and measure the intCTvals. The meaning is, that sounds produced from two immediate positions will not lead to two clearly distinguishable notes; i.e., to say, if a iruti follows another immediately and the former represents'a note, then the latter cannot be a new and distinct note. In gndharva music, svaras or notes are of three types - those which consist of four irutis, those which have three irutis and those which have two irutis. No svara can have more than four srutis or lss than two.

56. 57.

Sruti is a microtone, i.e., a minimal segment of an audible musical sound 'dhvaniri/dasamjmtah ^varah'. Svara is a musical note characterised by resonance {anuranana, ramno) and
harmonics. ^

'Samavede grmavibfmgbhvt ', - Abhinav makes an important statement here. Does this mean that the concept of srutis was not really applicable tor Sman but arose in the context of gandharva, the genesis of which was intimately cormected 'iJvith thfe wooden lute? Sman miSic was primarily vocal singing, hd" it would bedifficult to fix the twenty-two srutis m the throat'Dram gtramn ca Uve nne gnajtisu/ Smaigtrami tu tasyh srunuta laksanam// Gtrann tu s prokt yasym gyanti Sqmagh/ Svaravyanjanasamyukt angulyahgustkaranjit '// Mradiya Siks 1, 6, 1-2.
* There was possibly an artificial division of the octave. With' the development of instrumental music, perhaps the need for srutis arose ? How were thfey to be tuned accurately and the notes fixed ? The relationship, of the seven notes to "the twenty-two srutis is clearly analogus to the relationship of the diametekand the circumference, a ratio which is now-knw as n. Perhaps, the semi-circular form of the harp-shaped vn suggested this kind of relationship. There are five iruti-jatis viz., Dipt, Ayat, Karun, Mrdu and Madhy. The twenty-two. srwiis are classified into these five 58.

59.

53.

' 60.

......J9t.is.
54. This apparently means that toe the irutis. are'no such fixed sthnas for

ABHINAVABHARATI

NOTES
*

131

The concept of grama is rather difficult for the modem mind to comprehend, for the notion is long since extinct. The gandharva seven-note octave had a basic two - fold division on the basis of somewhat differing number of srutis contained by certain notes. These two divisions were the sadja-grma and the madhyama-

note or a dvirutika note, if either are raised or lowered by even a single ruti, thai it will create discordance, since a five ruti or a single ruti nte is not possible. In anotho- place (AB on NS, 28, 27), Abbinava discusses the tremolo of the trirutika pacamd of the madhyama-grma. He speaks of three types of tronolos - kampita, kuharita and

grma. The sadja grma began with the sadja note and the arrangement of svaras and mtis was thus : sodja rsabha gndhra madhyama pacama dhaivata nisada --

recita
63.

4 ;irutis 3 " 2
4

"
"

4
3

"
"

Vdi, samvdi, vivdiand anuvdi were the four terms for four diffrait kinds of notes to be found in jati singing of gndharva music. Bharata equates the vdi with the atha- the predominant note in a jtL Bharata says, that, notes which have an interval of nine or thirteen rutis between themare mutually samvdior that
they Have a natural harmony. Th meaning of vivdi in the context of present day Indian music is that note which is omitted in a certain rga, or, that which brings about discordance. The concept of vivdi m gndharva seems to have beai different. As regards vivadi, Bharata says, that, "those which have two ruti intervals are termed vivdis such as ri and ga, dha and nL Thus, particular notes have not been Singled out and described as .vivdis to particular jtis. lstead, two pairs of notes, ri and ga, dha and ni are described as vivdis to each other, ga being at a two ruti interval ft'om ri, and so also ni from ga. The concept of anuvadi seans to be, that wiiich is not vdi or sarnmcR, but also not vivdt.

"

In madhyama - grma, pcmcama was lowered by one sruti, becoming triirutika. Conseqently, dhaivata gained <Me ruti becoming catussrutija. The srutis of the rest of the notes were the same. The madhyama-grama,howevCT, commeilced with th madhyama note, and its ruti arrangement was thus ;

ma pa dK ni sa ri ga

4 rutis 3 4 2 4 3 " " " " " 64.

2' "

Abhinavagupta makes an interesting statement here-that only three ruti notes are to be used for tremolo. The explanation seems to be, that, while in oscillation during-the tremolo or kampana, even if, perchance, the note was to touch a sruti above or below it, it will still not create vaisvarya as notes of two, three or four srutis are legitimate. But, in the tremolo of a catusrutika

. 65.

Here, Bharata enuniierates different elements of a jti, Nysa is the concluding noteof the melodic structure. Apanysa occurred at the end of smaller parts, within the melodic structur. Each,jti had specific notes pscribed as nysa or apanysa for them. Sannysa was the concluding, note of the first vidn (a sqbdivision of the melodic structure). There were eighteen jatis which were sub-divided -into twoSuddh and Vikn. Suddh j'tis were those which wae named after the seven svaras,and that Very note after which theyaii was

132

NOTES
ABHINAVABHARAH

133

"Samvadi svaras are those between which are 8 or 12 irutis" (SR.

learned as its amsa, graha, nysa and apanysa. It also had the nysa svara regularly m the mandra, and did ot have notes dropped from it. 66. 1, 3,48-49). Abbinava, however, quoting his teacher, attempts to clarify the two points of view. He says that antara here does not mean interval, but nature or form (i.e., of the svara). Thus, the contradiction is resolved. The two views are but one and the same, and are only two different ways of expressing the same 69. thing. Abbinava had earlier said that the maximum span of iruti intervals between two notes consists of four irutis. An objection is raised - that if one moves from (madhya) sadja to madhya nisda, we shall have an interval of eighteen irutis. Abbinava argues that this does not contradict the basic principle, because in covering such long mtervals, the breath necessarily touches intermediate stations. In other words, while from one note to the other adjacent one, there can be a clear staccato jump, for larger intervals intermediate stations are touched and passed over. Abbinava is arguing, in effect, that to reach one note from another, the voice must jump fixed intervals of 4, 3 or 2 irutis neither more nor less. Thus, if rsabha has to be articulated after sadja, there can only be one jump of 4 irutis, because there are no intervening notes of 2 of 3 irutis. These intervals, thus, become like msical quanta. 71. The svara is a partless nd integral unity, not a compound of simultaneous or successive parts. In the irutis,there is imdoubtedly a succession of umts. But on account of the quick process of transition, their succession appears continuous and the last iruti where the svara is reached, acquires its character from the continuous growing impresion produced by the successive irutis on the mind. In this sense, the svara may also be regarded as a whole or an nsemble, which is reached through a successive and orderd process but whithin which no parts can be distinguished, Thus, while the passage from one note to another constitutes a quantum leap in terms of musical intervals, in another sense, it is constituted by a continuous wave- like passage, where, only the last effect can be self-consciously apprehended.

Mlavakaisika is a grma rga. The df\rum songs were set to musical forms such as grma-ragas, rga, bhs, vibhs etc. Abbinava explains, that, these musical forms were derived from the melodic structures of gndharva i.e., the jtis. These grmargas were bom through combming the various elements of various yaiis. Mlavakaisika had Kaisiki as its source jti, sa as amsa and nysa, and, was sung in the vesar giti (style).
Bharata equates the vdi with the amsa, i.e., the predominant note in a. jti - tatra yo yadamsah sa tadvdi (NS. 28, p.15). As stated by Abbinava, Dattila, too, regards vdi arid amsa as synonyms - Yo 'tyantabahulo yatra vdi vmsasca tatra sah (Dattilam. 18). Mataga (Brhaddesi. p. 13) and Smgadeva (SR. 1,3,50; also ibid. 1,3,47), giving an analogy for the vcR, have called it the ruler among the other notes. Simhabhpla (Sudhkara comm. on SR 1, 3,47) and Kallintha (Comm. on 5R 1. p. 183) also term vdi to be the most recurring note and synonymous with amsa. Kallintha explains, that, vcR was the main amsa of a jti. The remaming amsa notes were the paryayamsas (ibid p. 190). He also says, that, any amsa notes of a jti could be made vdi and graha alternately (ibid p. 186). Accoding to Bharata, the notes which havean interval of nine or thirteen srutis between them are mutually sainvcE, or, that they have a maturai harmony (NS. 28. p. 5). He enumerates the pairs of samvadis in both the grmas. Tise are, namely, sa-pa, ri-dha, ga-ni and sa-ma samvda in tlie sadja grma. Io the madhyama-grama, the sa-pa samvda does not obtain and is replaced by ri-pa samvda.Now, Bharata, Dattila (Dattilam. 18) and others have spoken f nine-thirteen sruti interval samvda, i.e., dha is located on the thirteenth iruti from ri, ni is on the thirteenth sruti from ga and so on. But the actual interval existing between these notes is eight or twelve. Perhaps this is why Samgadeva, at a later period upholds the other point of view

67.

70.

68.

134

ABHINAVABHARA fi

NOTES

135

72.

Here, Abbinava says that there is no samvada between madhyama and nisada even though the,interval is of nine and thirteen rutis. He gives the example of the Sadja-mqdhyama jati, where in its sadava or hexatonic form, even though madhyama is the amsa, there is lopa of nisada. Bharata does not raise the question at all.jrhe answer is, however, given by Abbinava. He says that samvada vll accrue whoi two notes are formed with an equal number of srutis-samasrutika (besides, of course, the fact that there should be an interval of nine and thirteen iraiis between them). Now, madhyama has four srutis and nisada has three, so there wl be no smvda. Sirnilarly, in the madhyama-gmma, where dhaivata becomes catussrutika, no samvada will obtain with ^abha wiiich is trisrutika. Abbinava herecites the example of Kaisika jati, where, in its sadava fofm, the. elimination of rsabha even with dhaivata as amsa is not an exception. What is probably meant, isthat samvada or ri-dha, sa-pa are eliminated in favour of ri-pa in the madhyama grama The idea seems to be, that, in gmdharva, the use of kka and antara is so slight and transient that sadja and madhyama (the unchanging note) should over-all appear unchaged. Again, there are obviously some problems relatmg to samvdaras regards the sadharana notes. Hence, they are rarely used in gndharva In the dhruv gna, sdhrana notes were used freely in the extension of the rnelody. This could upset the classic scheme of the samvdis. In such a case, reliance was placed on the ama as a basic stabilizing factor.

76.

This passage of Abbinava has been understood by some to mean, that, the gndhra and nisada are vivadis to all notes, which is not a correct interpretation. Themeaping of vivdi, in the context of present day Indian music is, that note which is omitted in a certain rga, or, that which brings about discordance. The concept of vivadi in gndharva seems to have been different. As regards vivdi,Bharata says, those which have tworuti intervals are termed vivdis such as rsabha and gndhra, dhaivata and / nisada (NS 28, p. 15). Thus, particular notes have not been singled out and described as vivdis to particular jtis. Instead, two pairs of notes, rsabha and gndhra, dhaivata and nisda are described as vivdis to each other; gndhra being at a two ruti interval from rsabha, and so also nisda from dhaivata It should perhaps run like this - With rsabha (as ama) dha pa, (insted of dha ma) ni ga, ma sa (instead of pa, ma sa). It would then mean that with ri atha, dha pa are samvadis (of ri), ni-ga vivdi and tna sa anuvci. It seems, that, in an earlier period there were three gramas current viz., sadja grma, madhyama grma and gndhra grma Nrada in his NracSyaiks (1, 2, 7), mentions the gndhra grma, although, he says that it does not exist in this world and is foimd only in the world of gods. This shows, that, ven by Nrada'stime,\he gndhra grma had already become extinct. TTiat it was definitly extinct by Bharata's time is proved by the fact that he does not even mention it. Why are thereonly two grmas? Why arenot there other grmas according to the other notes too ?This is thebasic question which Abbinava is trying to answCT here. He first quotes Dattila by saying, that, the answer is, that this is because the mrccharis are named by the sadja and madhyama grmas - saddi mrcchari and madhyattidi mrcchari. But he himself rejects this as not valid, asone could easily havea mrcchari begmning from rsabha ending in sa, or a mrcchari beginning from ga ending in ri, and there would be thus the defect of circularity. He answers it by saying that madhyama is imperishable, preeminent

77.

73. 74.

78.

74(a).

79.

74(b).

What is probably meant, is, that whoi kkaU-nisda is used, it is soused \zXsadja can also be heardon the same string. In other words the distinction between them is kept wholly elusive in

gndharva.
75. This probably refers to the fact tiat apart from the 2 rutis specified in nisda and gndhra, thra-e are 2 other ruis latent in them. These hidden rutis are manifested, only when kkali and antara svaras are suiig.

136

ABHINAVABHARATI

NOTES

137

and fixed, and sadja being its samvdi is also prominent. What he means is, that, sadja and madhyama are the only two fiill i.e., caturutka notes and pemanent. Hie dvihitika and trirutika notes are not strong enough to goierate a new grma; Abbinava points out that pancama also has equal rutis, i.e., it is also caturutka. But, since it becomes a trisrutika note in the madhyama grma, it is not a stable note. Sadja and madhyama are the only two notes which have fiill rutis and do not change their nature (i.e., loose any ruti) at all. Hence the raison dtre for sadja and madhyama grmas: This, however, would not apply to gandhra grma if gandhra is trirutika. 80. If the rutis of all three octaves (twenty-two in each octave) are taken, then there would be sixty-six rutis. lowevCT, the gamut of seven notes are produced only by twenty-two, so only twraity-two rutis have been mentioned here. Distinction is here made between the apprehension of rutis as atomic units and the svaras as the real units. Also a difference is made between the relationship of rutis to svaras, in gna and gandharva. In madhyama grma, only the sa-ma samvda obtains and not the sa-pa. Again, Abbinava repeats, why twQ gramas only ? Although pancama is catusrutika in the sadja grma, it has a variant in the madhyama grma, where it is not caturutika. That is why pancama has two forms. Hence if is not a filled up (purna) note m that sense. So also, gndhra and nisda which have their variant forms in the kka and antara notes. If veda ruti is taken as fourth ruti, and yati ruti as third, then the meaning would be, that, the four ruti note does not waver, just as a ripe kapittha fruit. The four ruti ote is a complete note, unable to take on more rutis and is like the kapittha fruit which cannot ripen any more, as, it is fully ripe. Just as, when there is a slight breeze the fruit bends in that direction, so also the four ruti'Tiote (i.e.,1t bends or lends its ruti to any preceding or succeeding note).

84.

Abbinava, here, refers to the lamkras, kampita, kuharita and recita. These seem to be alamkras of three mti notes, i.e., ri and dha (and perhaps pa in madhyama grma). In chapter 29, he says - 'recitakampitakuharstu' irovaksahkanthanivistasya

tririiteh svarasya kamparp ityabhijrineripi daritumaaky iti na vilikhita iti (AB on NS 29,21-22). Recita, kampita, kuharita, resulted from a,subtle quivearing of the three ruti note in thehead (tra) chest j^mandra) and throat (madhya). These alamkras were so subtle, that, though they could be
recognised in a melody, thy could not be expressed through words. Bharata, however, describes kampita as "a quiver of the time duration of three foiZs" {Kampitafn tu kaltrayam NS 29.43). Abbinava, commenting on this says, that kal could not be taken as ruti here (as was opined by some) but idstead was to be taken as a measure of time - kaltra na rutih api tu kalklah(AB oaNS 29,43). Perhaps it means alamkras, which consisted of a quivo- of ruti notes lasteda time duration of three

81.

kalsl
85. Bharata, thus, explains the pramna rutL Pancama, in the sadja grama consists of four rutis, iri die madhyama grma it is lowered by one ruti. It is this relative difference of one mti that Bharata terms as pramna ruti, and it was not conceived of in terms of any mathematical ratio. As Abbinava says here, "The difference which occurs iri pancama, when it is raised or lowered by a ^uti and when consequent slackness or tension (of string) occurs, that mdicates a standard (pramm) ruti." This is illustrated by the procedure of comparing two vTns, first tuned to sadjagrama and then differentiated, so as to yield the requisite ruti diffCTences. Perhaps the demonstration of the standard ruti can be better demonstrated on the harp type of vin rather than the lute type of vTn made of gourd etc. The difference of one ruti (that one pramna ruti) between the two notes viz. caturutika pancama of sadja grma and tri-

82.

83.

86.

87.

138

ABHINAVABHARAfi

NOTES

139

rtutika pcmcama of madhyama grama,struck on two differeaitly tuned but otherwise idoitical (in string, beam etc.) v7^.
two indentically constructed vmas are taken. Both are tuiied to the seven notes of the sadja grma. One is termed the dhruva viiui - this is the 'iimnovable' or constant viro which i& not touched!. The pcmcama of the other vTno, called th cala viri, is slightly lowered, so that it sounds just a little lower than the pmcama of the dhruv -ita, this lowo-ed pitch is tuned to the pancama of th madhyama grma. This is three rutis from madhyama, and is thusdistinct from the sadjagrmika pcmcama wliich was four rutis from madhyama. Tliis gave the measure of one ruti.Thej^with this lowered pcmcama as the focal point, all the other strings of the cala vim are also lowCTed so that they return to the sadja grama but at a pitch of one ruti lower than the dhruv nn. . 88. Abl^vagupta seems to imply clearly, the diffwence betweai laksam and lak^a in music. Laksam is the formal aspect as prescribed in the stras. Laksya is the music actually practised. The dhruva and the cala vvs have already been described. Further lowering of the cala who in relation to the dhrum im, revealed the pramm or measure of the intrvals of two ruti, three ruti or fourruti The tuning procedure was thus. ACT all the notes of the cala vTn had bei lowd to the measure of one ruti, the gndhra and nisda of the cala m, now, only slightiy higher than the rsabha and dhaivata of the dhruva vim were so lowered that they reached the same pitch as the rsabha and dhaivataof the dhrum vttm.The restof the v7n5 strings were again lowo'ed to conform with the new psitions of ri dha This' second sran or tuning gave the magnitude of the two niti intCTval. In the third sram or tuning, the trirutika notes, rsabha and dhaivata of the cala m, (which were now only a ruti higher than the sa and pa of the dhrum v&!).wre so lowa-ed, so as to become one Adth the ^ja and pcmcama of the dhrum v7n5. This demonstrated the magnitude of three sruti intervals in . three steps of a niti each. The rest of the strings of the cala m

were again lowaedto conform to this new positionof ri ^d dha In the ultimate sran, the three svaras pa, ma and sa (all caturutika svaras) of the cala vl5 (which were already three ruti low than these same notes on dhrum vTn) were furtho: lowrared and made one with the notes ma, ga and ni respectively of the dhruv vTn. TTiis demonstrated, the fourirw

intimai. 90.
Bharata does not mition any. graph. The Brhadde is the first work where such graphs are described and (kawn, though they rnay have existed earliw. Danda Prastara - This could be some thing like this.

sa

I I I I I I I I I I I I pa !I I I dhaI Im I I 11
ri ga
I

Aia Sadja-grama
I

ma

pa

dha ni sa Madhyama-grma

I -I

I |.| I

ri

ga

89.

Mandala - Prastara Sadja-grma 3 X2. ni. 5.1 r- 10

11

Madhyama-grma f SCL li'

11\ % 3 ^ S" fe

p-7 > 8 1 . 10 'CUK.11 17

16 15

M
18

W: vi, iqv181

17 91.

is

15

13

h*. 14' 13

The mrccharis are in ascending order. Why Abbinava should refer to them as in orderly and successive descent is not clear. Perhaps he means that each new mrcchari begins from a successively lower note, and hoice there is a devrait.

140

ABHINAVABHARATI

NOTES

141

92.

The etymology of the word mrcchan is given here: The word mrcchari has been derived from the root rriurcch which has two mpjning?; : moha or loss of conciousness and samucchrya vidch meansto swell orrise. It is the lattCT meaning whichseems to ajJply ho-e. Mataga, while defining mrcchari states that mrcchan is the basis on which rga is built or erected. .

arrangement of samvdis, anuvdis, notes which were permitted omission as well as in the order of mrccharis. 95. Though Bharata does not mention mrccharis in connection with the ari vTn, strangely enough. Abbinava does. He tries to justify the role of mrccharis in singing by pointing out thenuse in the singing of Sman. So it has been shown "he sings three songs by the Uttaramandr" (AB. p. 30). The same has been repeated here, as also the singing by Pmliks (thfe appears to be some Vedic mrcchari though its identification has not yet been made). He also says that jati Arsabf wassimg in the mrcchari beginning with pancama (AB. p. 55). Thus, the concept of a mrcchari bears the strongest esemblanc to scale, seven in eachgrama, each commencing from a diffrait note. Indeed, in the ancient musical system, with its rigidly fixed scheme of determined ruti intervals between the notes, tha:e could be no other method of obtaining a Variety of scales. 96. Abbinava, here, it seenls, is trying to give the etymology of the word auduva. The wu^uduva', he says, means the sky, in which-move about the stars 'udu'.TTie earth, water, fire, air and sky these are the pcmca-mahbhtas or the five great elements. Since the sky (uduva) is thp^fifth elonoit here^ the mrccharis derived from five nots are termed auduva and the rendering of five notes is auduvita^The explanation is rather far-fetched and he seems to be trying to somehow attach some metaphysical significance to the word auduva and aiuiuvita. It seems that there were four classes of murcchanas; with full seven notes, with six notes, with five notes and with auxiliary notes. But this seems to be contradicted by the fact that soon after, Bharata says that mrcchari is an orderly sequence of seven notes. Were the mrccharis, then, rigid heptatonic structures ? What in that case would these four classes of murcchanas be ? Could mrccharis be rend^ed hexatonic and pentatonic too ? One view it seems, was that, the mrccharis rendered hexatonic or pentatonic were tut another foim of tnas (cf

93. 94.

Nisdddhaivatantam should be inserted in-between i.e., from nisda to dhaivata. This seons to be missing.
Ancient Indian music recognised two gramas, on the basis of two different arrangements of sruti intovals according to the seven notes of an octave. Each of these gramas could result in Seyen mrcchans, which, wa:ethe seven svarasof an octave in a serially ascoiding order; each new mrcchari beginning on a new and successively lower note. These murcchanas were numbaed serially and each had a distinct denomination.

Sadjagrmii mrccharis : ri 1. Uttararrmndr - sa sa - ni 2. Rajar - dha ni 3. Uttaryat dha 4. Suddhasadj - pa 5. Matsafikrt - mn pa - ga ma 6. Avkrnt ga 1. Abhirudgat - ri

ga ri sa ni dha pa ma

ma ga ri sa ni dha pa

pa rria ga ri sa ni dha

dha pa ma ga ri sa ni ri' sa rii dha pa ma ga

ni dha pa ma ga ri sa ga risa ni dha pa ma

MadhyamagramiH mrcchanas: sa - nia pa dha ni 1. Sauvl ni dha pa ma ga 2. HarinsJ ga ma pa dha 3. Kalopanat - ri ga ma pa 4. Suddha-madhyarri-sa ri ga m ri sa - ni 5. Mrgl ga ri sa - dha ni 6. Pauran ri sa - pa dha ni 7. Hr^syak
94{a).

97.

Due to the lowering of pancama by one sruti in the madhyama grma, a lot of differoice arises in th two grmas as regards the

ABHINAVABHARA

NOTES

143

Bharata, NS. 28, 32). Samgadeva, too, says - Tnah s^.i mrcchamh sdavuduynhh {SR (Adyared), Svamdhyy; p. 115). The otho" view is that mrcchans w of four typeshq)atonic, hexatonic, poitatonic and with auxiliary potes. Simhabhpla gives the view of Dattila and Mataga on this {ibid. p. 114). Abbinava, here, also expounds such a view. Acary| Brihaspati opines that the sdhramkrt (with auxilia^ i.e.^feaZr and antara notes) mrccharis were used not in the jti singing but in the roidering of rgas (Ntyasstra - 28th adhyya svarwdhyya. p.- 60). A mrcchan can be accoqiplished in two ways. If in the sadja grma, gndH-a is raised by two srutis and considered as dhaivata of the madhyama grama, the rest of the notes get automatically adjusted to^ti intervals of the madhyama-grma and, thus, we can obtain iudfj/ia mrccharis of the madhyamagrma. Similarly, by lwraxng of dhaivata by two srutis iti the rrmdhyama-grma and considering it as gndhra of the sadja grma, thesrai intervabwiU get.so adjusted so as to correspond with the notes of the sadjagrma. Bharata declares tnas to be d^ndent on the mrccharis and gives their numbo^ as 84 - tatra mrccharisritsinscaturaslti / _ (NS 28. p. 27). Abhinava explains tanas as particular states of mrccharis. There were seven ways of rendering tnas as hexatonic - by dropping 4 notes i.e., sa, ri, pa and ra in the sadja grma and 3 notes" so, ri, ga in the madhyama grma. For example, the first (i.e. Uttaramandr) mrchan (sa ri ga ma pa dha ni) of the sadja grma would work out thus 1. 2. 3. 4.
-

'

ly, in the madhyama grma by the lopa of sa, ri and ga, each one of the seven madhyama grma mrccharis could be rendered in three ways, giving' twenty-one hexatonic madhyama grmil mrccharis. Ulis , would give forty-nine hexatonic tnas in both the grmas. Pentatonic tnas could be rendered in five ways. There were three ways in the sdja grma - by omission f sadja and'pacarrui, by omission of rsabha and pacama, and by omission of gndhr and jpsda. In madhyama grma the two ways, of rendaing were by omission, of gndfmraand nisdaand by that of rsabha-dhaivata.There were twty-one pentatonic tnas in the sadja grma and fourteen in the madhyarita grma giving a sum of thirty-five. TTius, fortynine hextonic tnas and thirty-five pentatonic tnas in both the grmas gavea sum total of eight-four tartas. It may be noted that the rule of samvditva govemed the omission of notes in the pentatonic tnas. 100.

Tnakri^ was the method of playing tanas on the nn, when certain notes were to be dropped. Tnakriy could be executed in two ways - by pravea and nigraha. Nigraha is not-touching. Pravea is the sharpoiing of the preceding note or the softening of the succeeding note. Nigraha is clear enough. It means that whoi a note is dropped, it is simply avoided. Explaining pravea,
Abhinava says that when a note, saywas to be dropped, then (by tightening of the string) the note could be-raised and rendered as rsabha. Alternately, in the Uttaramandr rrircchari (th sadjagrmii mrcchari) begining with (^adja), whoi sa was tobe dropped(the strihg couldbe) lowered and tuned to nl Whthor the note'was roiderd highn or lower depended on w^chevar note happened to be stroniger in that particular jti, and thus further strength was imparted to the already strong note. The basic idea was that in pravea, the omissible note was not avoided while playing, but assimilated int its neighbouring note, whether higher or lowr as required by the exigenci of the melodic structure.

ri
-

ga ga ga ga

sa ga sa

ri ri

ma ma ma ma

pa pa
-

dha ni dha ni dha ni dha

-

pa

In the same way, each one of the seven sadja grma mrcchari could be rendered in four ways, so that there would betwenty-eight hexatonic sadja^ grmilmrcchans. Similar-

ABHINAVABHARAH 101. The mention of tiie notes of tiie dravl v7/i is also for obtaining the nnsungnotes thatare present in thesnrilta vT/i. In the latter, the notes are not fixed by strings. They are latent, and, are manifested only when the musician sings - hence 'unsung'. Before singing, the musician first fixes tiie pitch of his voice with the help of some musical instrument. Some appear to have regarded ie flute as the basis of determining the srutis and the relative position of notes in the grmas, since they apparently thought that in the case of the lute, notes may be defective on account of some instability in the tying of the strings to the pegs of the vTn. The text of Bharata says mrcchan is to be demonstrated from the notes of the middle octave, because they are permanoit in nigraha and paryagraha. Abhinava suggests that when all the positions are either not fully reached, or some positions are reached but not all, the middle octave, in any case, is fully available. Ifence, that remains the basis for demonstration of

145

beginning with sa will begin from the sa of the middle octcave. Each new mrcchari will begin on a successively lower note. This is the lupt sadja mrcchari. Apparently, this refers .to the Samvda of pancama (of the madhyama grma) with rsabha. This new samvda is made possible by nigraha. In madhyama grma, pancama was never to be dropped. This is, perhaps, because the focal point and nucleus of the madhyama grma was the trirutika pcmcama. The reason for never dropping dhaivata in the sadja grma is a little more difficult to comprehend. Again, peihaps this was to distinguish the sadja grma from the madhyama grma. The caturutika pa and trirutika'dha distinguished the sadjagrma ruti arrangment from that of madhyama grma (with trirutika pa and caturutika dha). Either pa or dha whould have to be indispensable to preserve the identity of the grma. Thus, pa was indispoisable in the madhyama grma and dha (though it could have been pa again) in the sadja grma. As regards the indispensability of madhyama, Kallinatha gives two views. The first was simply a reiteration of tradition that madhyama was indispensable because Bharata and others have said that it is never to be dropped in tnas on renderingyaiis pentatonic or hexatonic. But this seems arbitrary. The other view seems more logical. It held that madhyama is the cpntral note dividing the octave in tho two parts :sa, ri,ga and pa,dha, ni. Madhyama isnever to be dropped because of its solitary and focal position (Kalnidhi on SR 1,4, 6-8). One may compar the same idea implied by Abhinava on p.l4. Kumbha, the author of SahgJtarja has used an interesting term "trika" (group of three) in this context and has pointed out that ma is the central note between the lower and higher tetracbords (trikas) viz., sa ri ga and pa dha ni. For this reason it is an indispensable note {SahgJtarja, 2, Ij 1,235). In the sdava form grulhra is not dropped, only ni is. In the auduvita form both ga, ni are dropped.

102.

103.

mrcchans.
104. Abhinava, here, is discussing the sdavita of a mrcchari in the sadja grma which is to be accomplished with the elimmation of sadja. But a problem arises here. In this sdava form, how is the first to be differentiated from the seventh mrcchari ? In the prm forms, they will be 1) sa ri ga ma pa dha ni 7) ri ga ma pa dha ni sa. When sa is eliminated, both will remain as ri ga ma pa dha nl To this, Abhinava says that when sa is eliminated from the seventii mrcchari,the middle saptaka being depleted the notes sadja, rsabha etc., bemg heard in the mandra saptaka, i.e., this mrcchari will be entirely in the mandra saptaka.This h& calls the seventh lupt mrccftari. Similarly, the first mrcchari with the sa jeliminated^ will belong only to the madhyama saptaka. (What Abhinava means by saying that the sadja eliminatedhere is that of the tra saptaka is not in the least clear. Thisseems to be a mistake for the madhya-saptaka sadja). It has clearly been indicated that the mrcchari will begin from madhya saptaka i.e., the first mrcchari of the sadja grma.

146

ABHINAVABHARATI

NOTES

147

108.

The prastara of seven notes would have 5040 permutations as litatnas if all the notes are used at the same time. These were arranged in a fixed order in numbered rows so that given the permutation one could find the row and vice versa. As has already been stated, seven note mrccharis on being rendered hexatonic or pentatonic were termed tnas. However, here too, there were rules regarding thedropping of a note or pair of notes in either grma, and, as a result only a total of eightyfour hexatonic and pentatonic tnas were permissible in gndharva. In gna, the tnas were limited by no such restrictions. Besides tna, gna utilised haatanas which werebasically permutations of two or more notes in all possible arrays. Abhinava says " in gndharva there are fourteen mrccharis and eighty-four tnas.\ngna thtte is a variety of litatnas depending on the utilisation of (as few as) two notes upto all the (seven) notes. Thus there would be 2,6,24,720 and 5040" {AB on NS. 33,1). The sevoi notes of an octave could be permitted and combined in 5040 possible ways - two notes gave rise to only 2 combinations three notes gave 6 combinations, four notes to 24, five notes to 120 and six notes to 720 combinations. All these possible combinations were permissible in gna. Bharata has not described litatnas. However, Dattila {Dattilam. 38-39) says that liptna was a mrcchan, but with this basic differencethat unlikethe tnrcchan its note sequence was in disarray. A ktaina when rendered with all the seven notes was called punta laitatna, when rendered with lessthan seven, i.e., with hxatonic or pmtatonic structure, it was temied aprna ktatna. The latter could be rendered with four notes or less. Dattila gives the mmiber of litatnas as 5033. These Uttatanas pertained only to gna and it is only in the context of rgas that Abhinava discusses them. The regular 84 tnas of gndharva were for pleasing the gods - i.e., for transc^dental merit. The litatnas were "infinite" and tbeir purpose was to "prodijce pleasure for the audience" -.e., the dr^ purpose or rakti.

110.

109.

Abhinava gives the analogy of seasons to explain the svara sdhrana. There is a time when winter is not fully ovo: but spring has not fiiUy arrived. This is in-between season, between twomajor seasons, winter andspring i.e., it neither has the bitmg cold of winter, nor has it matured into the warmth of spring. It shares the characteristics of both winter and summer, so also the sdhranasvaras.They are 'inbetween' notes,having borrowed two rutis tiey have become sUghtly sharp, but have not npened to the status of the full fledged regular seven notes of the octave. The gndharva systn of music admitted of only seven notes. Apart form these seven, the only others that wore permissible were kka nisda and antara gndhra. Both had the nature of being sghy sharpened not^, being two rutis higher than their regular intervals. As they were modified forms of the regular gndhra and nisda, they did not have the same status as the seven pure notes or avikrta notes and were subsidiary to the latter. Abhinava pomts out that the two terms kakali and antara were only a matter of convention- As a matter of fact,either could be tmned kka or antara. These two notes were also coUectively know as svara sadharana, What Abhinava is trying to make clear is, that even when two (or more) yfliis are grouped as 50d/i5rana by way of having common aias or vdT etc., yet they do not become identical. They may have some shnilar features, but there are other charect^c elements of;aIi too, which serve to distinguishtiiem firom each other. This sentence quoted.from tiie text in tiie commentary is not
found in the Baroda edition of die text.

111.

112.

113.
114.

Jti Sdhrana is because of some similarity in some portions of two or more jatis due to simUarity of tiie ainsa and graha. Dissmilarity of tiie jatis continues on account of the feature of which tiie most impotant are nysa and antaramrga, Hrejri sdhrana is sought to be defined . in terms of graidilaksanaparijmna. Since this is precisely what (stin-

148

ABHINAVABHARA 77

NOTES

149

guishes iixe, jati sadharana it can be emphasized only by neglecting these laksanas. 115. Perhapswhat Abbinavaisreferringt0,isitbatinthesaf/grwa, murcchanas are rendered dava {nas) by dropping the note ni?daaxau(^vitahyhoinUga.lximadhyamagrtna,ixis the opposite - i.e. sadava by droppmg ga only and auduvita by both ga hi The idea is that the texts gives to svara s-dhrana the name kaiika also. Now, which note is this ? If it is kkali or antara, then it will belong to nida and gndhra and will have appropriate forms in accordance with the change in grma and extent of thescale - sa^va, auduva etc. But in either case, it does not appear to be disconible in Sadja Kaii or KaiiiE as explained in the commentary. Sadja Kaisila is a purna ti, it does not have sadava auduva forms at all. KaiiiEhas ri dha dropped in hexatonic and pentatonic forms. Abhinavagupta quotes his teacher's teacher Utpaladeva who held that the svaras,had 2 kinds of forms - prkrta and 'vikna, natural and defonned. Any of the notes may acquire deformation or vikrti. This idea is not clearly mentioned in the text. It seems to fore-shadow the medieval practice, only the distortion h^e presu^sed the system of grmas and mrcchari and not a fixed tonic with varying tlmp. In pages 33-34, Abhmava has been trying to drive home two pomts viz. (i) what is sadja and madhyama sdhrana (il) what IS kaUikal One type of svara sdhrana i.e., kakali nisda and antara gandhra with two rutis raised has already been mentioned. But if the same was meant why should it be repeated? To this, Abhinava answerstotanothertypeof'sdArana'ismeant here, viz., sadja sadhrana and madhyama sdlmrana. When the first ruti o sadja is brarowed hy nisda and its(i.e., sadja's) last ruti is borrowed by nabha, i that state, sadja is termed as sadja sadharana. Kaiika means 'fine as a hair' and denotes subdety. This trirutika nisda is now termed kaiika nisada. Thus, m the state of sadja sdhrana; sadja is of two irutis.

rsabha four and nc/a of threeiruiis. Sinjilarly, whengnd/ira takes the first ruti of madhyama and pmcama its last ruti, in such a state madja/na is termed as Tnacfyamasd/ira^. Such a gndhra raised by one ruti is termed sdhcram gndhqra. 119. The objection raised here is,svara - sdhrana has akeady been mentioned s the raising of ga ni by 2 rutis, then what does this new svara sdhrana mean? The answer is, that, in the.context of gndharva music there is only one svara sdhrana - kkali nisada and antara gndhra. Gndharva music admitted only two vikna notes- ga and ni raised by two rutis collectively termed svara sadharana. The othr svara sdlrana, however, Potains only to gam system of music. Here, a variety of notes were used, and kkali nida and antara gndhra were differet from the svara sdtmraim mentioed hrae. According to Abhinava the notes used in the raga, bh, did not followtberigidrules of gni/Aarva. Theraisingnd lowering of pitch did not in popular practice follow the quantum of two rutis, but varied by any number. The fifteen notes, thus, used were as follows. Sa^a grama nisada trirutika rsabha sadja Madhyama grama gndhra pancama dhaivata caturutika dvirutika trirutika tritutika caturutika .

116.

120.

117.

118.

The 7 pyre notes + these 6 notes + 2 notes - {kkali, antara)= 15 notes. Thus, it is shown that in actual singing of the period of Abhinavagupta, notes were not merely used in thestandard form of the octave, but variations or viAra for each one of them were recognised. 121. Two kinds of sadharana had been mentioned svara sdhrana and jti sadhraim. Jti sdhrana was described in terms of

150

ABMNAVABBABAV
similarity

151

of tis. In this very context, however, there is also a mention of sadja sdhrana and madhyama - sdArana, which seems to refer to the raising and lowering of the notes of the two gramas in accordance witha crtain actual but highly skilled and difficult style of music. The altered notes in this style are alsb called kaisika and the style is kaisiki. Abhinava mentions the alteration of notes in the 2 grmas in this comietion. This apparently links up with the quotation from the elder Ksyapa. It is, thus, incorrect to think that the only vikrta notes used in Indian music were gndhra, nisada and pancama. Here rsabha and dhaivata ar0 also raised by one sruti each. The \diole concqpt of sadhrana is the concept of the iteration of notes from their standard gndarva forms to accohiodate the actual practice in rga, bhs etc. In the jtis of gndharva, only a restricted useof sdhrana was permitted. But how, the concept was generahsed and elaborated. This reflects the evolution of Indian music from the time of Bharata to Abhinavagupta. The use of antara gmdhra in ascent should be thus viz., ma ga ri ga ma or ma ga ma. ^ the same way kakali ni^da in ascent should be used thus:'sanidh^nisa' or 'sa ni sa'. Even this uSe should be in a limited way. These notes should not be used in descent. The seven modes of singing pertain to gna system of musit. Hence he says that these produce bhva and rasa and thus pleasure or dma pTiala. The grma-rgas are bom from jtis which have traditionally been known to be eighteen. Jtis are said to be presoit in the heart, ^at the sstrakra does is to propound an ordered system of definitions. We can, thus, notice that Abhinavaindicates the 3 poles of the creative process of the musical triangle. The apex is m the heart, the instinctive, intuitive apprehension; then there is the pole of popular practice or lak^a and finally the traditional conception or laksana.
This

music. "Thereareonly three;/aiis- viz.,AfadAyamX PacamTand Sadja-madhy which re coimected with the use of the sdhraim svaras. The amas in these tisare respectively sadja, madhyama and pacama. In case of pcmcama it is to be applied as an alternative to an extremely weaknote." {NS 28,37- 38; 28, 44-45).
/

122.

123.

124.

Bharata's language is cryptic and it is Abhinava who elucidates. The ama in a ti was its dominant note and can be equated with the vdisvara. Many tis had not one but several amas. These multiple amas were termed paryymas or altemate amas. Only one of the multiple amas could obviously be the dominant atna at a time. All these three tis had multiple amas. MadhyanorMadhyamha five amas, sa, ri, ma, pa, dha; Pancarm had two amas ri and pa and Sadja-madhy or Sadja madhyam had.all its seven notes as potential amas. Abhinava explains by saying that in these three jtis only when sa, maotpa was the acting ama could the sdhrana svaras be used. In the sevMi ama ti Sadja- madhyam, antara ga and kkali ni could not be used when the full fledged two ruti notes gndhra and nisdawere th acting amas. They could be used only whai sa, ma or pa were the ruling amas. The same injunction accmed in the case of Madhyani which had five possible amas including sa, ma and pa. Pancanii had two possible amas ri and paand the sdhrana svaras could be used only when pa was the aiha. This jti was rendered sdava by the lopa or dropping of the note ga and auduva by the lopa of both ga and nL Abhinava suggests that when this ti yras rendered sdava by omitting ga and with pa as ama, thra in place of ga,which was thus raidered weak (a dropped note was often not totally omitted but rendered weak), an exceedingly weak antara gndhra was to be used. Similarly in the auduva form of this jtL kkaS nisda in place of the regular two ruti nisda, too, was used as a voy feeble nte. The Vikrtas, although, derived from the Suddhs, are not $ _ described as the Suddhas,but described by their own names. This

. 125.

use of the sdhrana svaras - antara gandhra and kakali nisda was greatly restricted in the jti smging of gndharva

152

ABHINAVABHABAI7

NOTES

153

isudikematerial substances derived from their made of earth called ahhMi-ware, but it is not like this. 127. This is a very curious derivation. The vigraha vakya is totally at variance with the types of compounds mentioned, a sure sign of the bad state of the text.
128.
Should not oparaspara be oarospara?

7. Gndhri, Arsabf

Andh Krmrw Gndhrapancann KaiiE.

/

8. Arsabh, Pacanii, GandM Nandayan

9. Arsabh, Pacanii 10. Gdndh Pacan 11. ^dji, Gndhri, Madhyam, Pacan, NaisdL

129.

The idea seems to bethat dvaigramikya cl^cter may bedefinitional by virtue of the use of svaras, grahas. etc. from the two gramas^ EighteoiJiaiis have been enumerted by Bh^ta,1 belonging to, xndjn grTirr.a and the rest to mc^yamagrand. Thgse eightew Jtis were further subdivided into two - Suddh and Vikm, Suddha jatis were those which were i^ed ft the even svaras, and that vray note after whichthe/oti was named was ks amsd, graha, nyasa and apanyasa. It also had thcnyosa svara regularly in the mandrd and did not have note^dropped from it. When two ormore characteristics oftheSttcftajaiwere altered exCeptfortiieny5sa(tiioughitcouldbesonletimesintiiema<i/iya sthanaiao),it was termed aVikrtatL These we bom tiirough mutual combination(samsarga) of ihe uddha jatis. These were eleven in number and tiieir names and origin have been enumerated by Bharata. ParentJSlis
iSuddh)

130.

__ Since these had some characteristics of the Suddha altered in them, they were termed Vikn", since they were bom of combination, they were also Sarhsargaja. See also my A Historical and Cultural Study oftheN^astra of Bharata; fh 141,pp. 219-21. 131. 132. Perhaps sadjasvaratvena is a mistake for satsvaratvena. This would be closer to the text of Bharata. Bharata speaks of four jatis which always had sevoi notes, i.e., they were prna jtis-, four were hexatonic -Q/dava) and ten pentaton'c i.e., atiduva. Madhyamodicyav, Sadjakaiil, Krmran and Gndhrapancarm had all seven notes, ^djl, Andh, Nandayan and GndhroiEcyav we hexatonic. The poitatonic jtis of the s<^ja grma were Naisdi, Arrabbi, Dhaiva, Sadja-madhyarn and Sadjodicyavaii; the madhyamagrmil paitatonic jtis were Gndhn, Raktagandh, Madhyam, PacamJ and Kaiil3. However, Bharata also adds that those that we hexatonic could sometimes be rendered as pentatonic and vice-vCTsa. Here Abhinava discusses the amias which did not permit sdavita and auduvita. A genal rale that governed these jtis, was that inthe rendering of sdvita mdduduvita of these jatis, the samvdis could not be dropped. Hce the jtis had to be rendwed hexatonic or pentatonic in such a way so as not to effect the samvditva. The three jtis indicated by Gdndi etc., are Gandha, Raktagandha and Kaiil, all belongmg to tiie madhyamagrma. Here a sammda existed between ri ^d pa. In these yaiis, the hexatonic rendraing was effected by the dropping of rsabha.

DerfvedJaris
{SainsargajVikrm

1. ^jTand Madhyam 2. Gndhn, Sd 3. :d}l, Gndhri, Dhaivati

4. SdjT, Gndfn,

Sadja-madhyama Sadja Kaia3 Sadjo(icyava Gndhrodicyav Madhyamodicyavi Raktagndhn

133. 134.

Madhyam, Dhaivati. 5. Gndhri, Pmcan, Madhyam, Dhaiva.

6. Gndhan, Pancaml,

135.

Saptami (Naisdi)

154

ABEniAVABHA^V

NOTS

155

However, whenpa (which was indispoisable in the mdh^magrm) was the amsa, ri could not be dropped and then sadava 136 rraidering was not pnutted. Sadjodicyava belonged to the sadjagrama where a jsavda existed between ri and dha. In sjati, sadava was effected by the dropping of ri.Tisti had four amias viz. sa, ma, ni, dha. When dhaivafa (which was indispensible in the sadja grma) was the ruling n^a, ri could not be dropped because of the rule of sfnvadcC. He, i.e., Bharata sums up the sevai amas which prevent hexatonic rendering in certainyoiis. These are ni, ga, pa, pa, pa, ga, dha. In case of either ni or ga being the (ruling) amia, the jati Sadja-madhyama cannot berendered hexatonic. If pcmcatM^ is the dominant^imia in the threeylis Gndhri, Rakfagndhan and KaisiG thrai these jatis cannot be rendered sadava. If gndhrifis Jhe ama in Sad/T and dhaivata in Sadjodicyava, then both lhese j'tis cannpt be rendered sdava. No independent rationale can be diScemed here, i.e., for ba^g auduvita with all four amas (s, ma, pa, hi>in these two jatis. It seems simply a matta of prescription in the Sstras. Possibly, Abhinava means the recounting of the numbo" of amsas which do not permit sdava and auduidta. Abhinava probably means sixty-three. That this is what he has in mind isobvious by the next sentence. By subtracting mne amas from the total number of arnsas he arrives at the number fiftyfour. The four ever puma j'tis were Madhyamodicyaya (1 amsa), Sadjakaiil (3 ' ai^as), Krmravl (4 arhas) and GndhrapaUcaml (I aina), thus giving a total of 9 o/nsas-jn Sadjamadhyam (ga or ni), Sdji,(ga), Kaiil (pa), Gndhan (pa), Raktagndhn (pa) and Sadjodicyav (dha). These sevoi amas prevented the sdavita rendering of these yis. Thus subtracting 16 (=9+7) from 63 amas, we are left with 47 amias. It is these remaining forty-seven amiaswhich when used permit the j'tis to be rendered as sdava.

142.

TheevCT prm jtis have Sdxvsi jtis have Atiduva jtis have Jftsnamed 7 notes have

63 amas 47 30 7 " " "

147 amas With this total of hundred and forty-;seven arhas both the gramas can be obtained. See also Brhaspati ibid., pp. 100-101. 143. Probably what is meant is grma rga (not,grma and rgd) belonging to the popular dhruv gna system of singing. But as to what 'may happen' is not clear. Probably it may refer to the violation of rules regarding soavit and auduvita (\v1iich was the context earCT). Z)Arv5g5na wasnotgovonedbysuchrigid t inles. This passage is quite cryptic and does not give clear reading. In the ltt part, however, Abhinava seems to refer again to the twelve amas which prvoit auduvita viz., eight in Gndfmfi and Raktagndhn (sa, ma, pa, ni ech), two (ga, ni) in Sadjamadhyam and two in Pacan (ri) and Kaiil (dha) respectively. Bharata clearly refers to the indispensability of only the-madhyama-(NS. 28,33; 28,65). He does not say this about the other notes..But according to Abhinava therewas anothra opinion also and in this context he (AB. NS 28,34) quotes Dattila (Dattilam. 20), Viikhilcrya and others. These theoreticians held the opinion that dhaivata was mdispensable in the sadja grma, pmicama in madhyama grma and madhyama in both the gramas. They do not accq>t the opinion of Bharata that it was only madhyama which was indispraisable. As said, a peculiar feature of gndharvawas the indispraisability of madhyama in both the grmas. It seons to have been borrowed iiom smagna. Bharata speaks of this as a common feature of gndharva and soman music."Madhyama is the chief

137.

144.

138.

139. 140.

145.

141

146.

ABHINAVABHARATI

NOTES

157

of all notes and is tomed as mdispensable - so it has been said in the rules of gandharva and soman" (28,65). The reason given by Abhinava, for its importance m gandharva and soman is that it holds a position of equilibrium. Madhyama is the central note dividing the octave into two parts: So, ri, go and pa,dha ni,the IOWCT three being consonants to the upper tinee. So to pa, ri to dha and go to ni Between these two divisions, madhyama,the nucleus, stands alone and has no note left as its samvadi. Hence it is not to be dropped due to its solitary and cmtral focal position. In this context (the indispeaisability of madhyama) Bharata uses the word gandharva kalpo. Abhinava interprets the term gandharva kalpo to mean not only gandharva but also grama rogos wiiich were those forms of dhruva gna that were closest to the jtis.These i.e., grma rgas were bom directiy of the jtis in contrast to the bhs, vibhs etc., which were bom not of gndharva forms but out of grama rgas. Kallintha explaining the significance of the name grama roga and its relation to 'grma',connects this form directly with the j'tis, "even though the grama rgas are not directiy bom of the grmas but only by way of the j'tis, yet they are less removed from the grmas than such forms as bhss, rgas etc. Hence they are called gramo rgas {Kalnidhi on SR 2,1-8-14). Hence in the grma-rgas, too, as in the jatis, madhyama is said to be an indispensable note. However, it did not have this position in other dhrum forms such as bhs etc. 147. The word 'rakti' has been translated by Dr. Mukund Lath as 'musical charm', whichdoes not seem to do justice to its subjective, experiential implications. Graha was the note used at tlje commencing of a melody: that the graha was the intial noteof a melody isan accepted fact The controversial aspect, howeVer, is its relationship with the amsa. Is graha only similar to amsa, or is it in fact, identical with it ? From the passage of Abhinava it is clear that though invariably

graha and ama were the same, there weresome exceptions too, as in Nandayan Jti, hence each had then- independent status too. Graha, being the intial note had a limited role, whereas ama was a much larger concept It was the note which determined the form of a melodic structure and was the dominant note in it The grahas specified for a jti are exactiy the same as then amas, exc^t for Nandayanti 149. 'Sadgrahh' here does not make, sense, hice the alternative reading 'sodjogrohah' is suggested. Sadja is graha, ama and nysa oi Mlavokaiika (Brftaddei, 346). Abhinava also says this while referring to grmargas at the end of the chapter. The meaiung, then, would be that in the various forms m which Mlovakaiika issung, all do not regularly have sadja as graha. (Hoice the need to define graha and ama sqiarately). In the jti-gna the graha and ama are invariably thesamefexcept for Nandayan. HOWCVCT, this was not always so in the grmargas, as is evidoit fixim the example of Mlavokaiiko as given by Abhinava. This means that, aiho is the note which governs the movemoits in the tra and mandra octaves. Ama is the most prolific note as conq)ared to all others and, hence, it is the dominant note. 'Svasvarpeksa^' here, seems a mistakes for 'sarvosvorpe^ay'. In relaticm to ama, otha notes are established such as sammdi, anuviS etc., and on it depoid the five factors graha, apanysa, vinysa, sannysa and nysa. The ama is, however, in no way related to tho vivdi note. Even if some of thenotes (not samvdi,anifv<i, but others apart fixim than) do not look in the same direction as the atha (i.e., are not associated with it), yet it may still impart charm to than, by a kind of covered withdrawl. In verses 68-70 Bharata enumerates these ten charactoistics of the ama. These are (1) Ama is that note (of the musical composition) in whibh lies the charm and from which is

150. 151.

1^2. , nVi-

153.

148.

154.

158

ABHINAVABBATTA

NOTES

159

goierated the aesthetic fonn of thecon^rasition; (2) it detenniaes the range of the mandra-, (3) also the range of the ira; (4) it is the most prolific note; it determines the (5) gr<ia, (6) apanyasa, (7) vinysa,(8) sannysa, (9) nysa; (10) it is the note which the others follow. 155. What Abhinava is trying to say, is, that this rule of ascoit upto fifth of ainsa, potains only to notes vibrating in the head and heart, i.e. the tm and mandra. If it is applied to the madhya saptaka,(i.e., ascoA up to the fifth note of the amsa in the middle octave), thaiit would greay contradict practice. Ia .the madhya sapika'allnotes are-to be taken and the restrictive rule is to be applied only in th lower and higher octaves. Here Abhinava comments on the movanent of the tata. Hesays that if Sadjawas the atna, th^ starting from the sadja itself one moved upto the fifth note, i.e., sa, ri, ga, ma, pa. This was if one could stretch the voice so high. Thoe was no fault in going upto a note lower, but it was never to be beyond the fifth note. With waAa as aina, one could move upto tra dhaivata, and with gndhr as ainsa upto tra tusada. With ma, pa, dha, ni, the maximum pomissiblei movement was the nisada of the higher octave. "ITiis meant that theatitra was never to be takea The ti Natulay'anR was an exception to the rule of the tra movanent. In the Nandayan, the piovemait in the higher octave was confined only up to th first tara note, i.e., the ira sadja. In NS 28,24, Bharata singles out two pairs of notes ri-ga, an cia-ni as being -imdis. It is po-haps because of this vivaditva that Abiinav talks of lck pf rat ha^e ? How would the tara movemait b calculated in tis whih WOB hettnic and ptiatctaic ad which onttted notes ? The rule was, thai; as regards sachtis,wliile detemiining the upper limit of the highear octave,,the omitted note wa? also tp be counted. These rules, about the tra limit pertained only tothe jatisof gruiharva and not to grma rgas.

was the aina note itself in the mandra saptaka. Another option for the lower limit was the nysa svara. Yet another altenative for die limit of the Iowa: movement was the note beyond the nysa, for example, if gndhra was die nysa, one could move down up to the rsabha. 158(b) Here, the unedited text reads 'larighanabhyayoh visam; the 'edited text has 'latighahbhyyoh visam; die proposed readmg is 'langhahnabhysayoh' where vtsam does not make sense and so is omitted. , 159. Alpatva was of a note which was weak, or used rarely pr omitted. The alpatva of a note in any jti could be effected in wo ways, viz., by lahghana and by anabhyasa. Langhana was effected by gliding ovCT a note in such a mannr so as to leave it unemphasised, and, there by, rendering it subordinte to the emphasised note. Anabhysa was die avoiding of the repetition of die note. What WS the difference of context ? When did the alpatva operate through larighatta and when by atiabhyasa ?"Abhinava has an- answer to this. In the pnvasth of ^e ti (where no note was to be dropped), thsat alpatva was effected ofl the weak note dirotgh Umghana, Anabhysa operated oq notes diat were , hot potential ainas (paryayama). He gives the example of^e jti .where, anabhysa operated on the anamas, nisada ad rsafete. However, it could npt opoate on the graha and i 60: sarnvfiT nots. Bakiti>a was thphSlific use of note, lilis could be effcted in Hwo ways,.s^zj by abf^asa and by qlaitghcaui. The former meant fi-equetitreptitionofthenote. yt/agAaaifteant,notslappingovranotewiiiterendering thejit-sdtifira. i The strong nots in /i, were, obviously diea/ka, die vif and die samvfldT. Apart frot diese inherently strong.notes, die other notes on which bahutva operated wa-e die paryayamas (alternative ainas) and notes which even though not samvadi were strong. Bharata has not maitioned the tam paryayainas, but Abhinava has.

156.

157.

158.

158(a). There were three pptions for the extait ofmovement in the Iowa octave. One possible liiriitior the movement in the Iowa*octave

160

ABHINAVABHARAH

NOTES

161

160(a). Some of the text is missing here, but it seems that Abhinava is tryiiig to define the antaramarga. The antaramrga manifests mti. Where the amsa note is predominant in a group {dala) of selective notes, the samvdis, too, are sbong, the anuvadi notes follow, alpatva operates on weak nots "and bahutva on strong notes; such a play and interplay of notes is called the antaramarga of ibsjati and helps to unfold of manifest the nature or form of the jtL 161. 162. 163. , See note 132. See notes 140,141,142 and 144. ^ rendering with four notes was only possible Q dhruva gna, but was iiot permissible in gndharva music. Only heptatonic, hexatonic and pentatonic renderings were possible in gndharva music In all iht jatis, graha and amsa were the same note (iV5.28,67). Nandayan, however, seemed theonly exception. In vMse-79 of this chapter Bharata gives pancama both as the graha and amsa. However, in another place (verse-134) he says that the gndhra should always be made the graha and amsa. Mataga {Brhaddesi. 275), and, quoting him Samgadeva {SR. 1,7,107108), too, explicitly state both the options, i.e., (i) pa as amsa and graha (ii) ga as graha and pa as amsa - "Nandayanpin pacamo'mso gndhrastugrahahsmrtahl kaiscittupacamah probo graho'sm gtavedibhih // Abhinava {AB on NS. 28.67) mentions gndhra as graha, although h^e. {ibid. 28,79) he gives it as an option to bfc accepted. Dattila {Dattifam,%5) however, gives only gndhra as graha. Jtis Group of one amsa jtis
1. Madhyamo<Scyav
2.

Jtis Grotip of 2 amia jtis' 4. Dhaiva

5. Pdcan
6. Gndhrodicyav

Total np. of ainas in each group.

Group of 3 aihia jtis I. ArsabHi Z.Nisadim 9. SadjakaiiiM Group of 4 amia j'tis 10. Sadjo<Ecyavatt II.Kmiraw 12. Andh Group of 5 amia jtis 13. Madhyani J4. Gandlw 15. Rahgndh 16. ^d 6amiajti 17. Kaiiil 7 aihia j'ati

12

164.

20

l8. Sadjamadhyam
Total = 63 ainas 166. Vyapadea means designation. Vytyjqdeinfs one that is designated. Vyapadefvadfwva en&\Ats one to treat the object in accordance with its designation. Thus, &jati is here called a gana becouse it is treated as belonging to a designated gana having a particular number of amas. Bharata {NS 28, 96-97) prescribes a sancra or movement of sadja and gndhra as also sadja and dhaivata in the/aii Sdju The commentary on the Brhaddei (251) also says the same -

165.

Total no. of amias in each group.

Nondayan

167.

3.Gndhrapancan

ABHINAVABHARA TI

NOTES

163

'sadjagndhrayoh sadjadhaivatayosca samgatih, gcmdhro' tiveEpditvt parsparagamamm ca samgatih: Kallinatha, the commentator of the Sahgitaratrkara defines this movement more specifically. He says "the sahgati, here, should be of sa-ga and sa-dha". Here sadja has to be associated with gndhra and dhaivata, both of which happen to be remoyed from it by one note (thus : sa {ri) ga, and dha (ni) sa). This association (of notes) should be aesthetically channing. The sahgati should be thus: either sa ga sa ga sa dha sa or ga sa ga sa dha sa - 'sagayoh sadhayoscatra samgatiriti. Atra sadjasya gndhrenaikntaritena tdremiva dhaivatena ca yathrakti sambandhah sagasagasadhaseti gasagasadhaseti va kryah: KaBnidhi on SR, 1,7,61. The jti ^dhas five aihss viz, sa, ga, ma, pa, dha. It had 2

forms; viz, the heptatonic and hexatonic and as suh would have a total of 10 amsas. ^d was rendered hexatonic'by nhiitting the note nisada. However, when ^dhsA gndhra as amsa, it could not be rendered hexatonic,-because m s a samvadicovM. hot be omitted. Conseqently ^cold be rendwed hexatonic. with only 4 amsas (not five) viz, sa, ma, pa, dha. Hence, in practice there were only a total of 9 (not 10) aimas in both the sudd^ and vikrta forms. The Kmi on BrhaieM, 251. also says the same'Sd- (dvi) vidh nityasarhprm sdav eti....->:...,.Suddh vikrtsca panca prnscatvtah davh ._ g^tire*ms&. sadavpavadt, Suddham pari^ajya cdfurvid^ ^ vikn boddhavy. ,,

etc. Kala was the basic unit of time - measure in ancient tlas. (For further details see notes 35 and 36). The beats could be sounded and unsounded. Pta literally meant a soimded beat. The basic strcture^ of tla were classified into two basic groups, a dupleanangoncaitandatripleaTrangement. Thus, caccatputah and cacaputahwere the two basic fa-prototypes.Bharata terms them yqnivadya {NS 31,7), because they were the baseon which were constructed other tolas. For further details of these two see note 16. Bharata speaks of three layas or tempos viz., druta or fast, madhya or middling and vilambita (slow). These corresponded to the three nirgas; citra nirga {druta laya), vrtta or vartika nirga was in madhya laya or medium tonpo and the daksim niprga related to vilambita laya or slow tempo. In the three nirgas, tha:e were thi^ different styles of applying the kolas. Thus, in the citra mrga, the style of applying the kal was ekakal, i.e., for eg., in the caccatputah tla it would beS S S S - Hete the mode of single hil was used. In 1 1 1 1 Ocal) the vartika, the dvikal mode was used Pribhsic matm - (see note 36)

"fot"eg ;

SS
2.

SS 3

- .dvikaJ caccatputah

p a d a . m r g a s (see ^ ^ note 36) In the daksina marg^the ca^ka&,TD.ot was used.

Pribh^ nitr

Just as the recitation of Vedic mantras ise as farm\4 {tiigada stuti) or while reciting mantras during tfitf kindmg of sacrificial fire {smidhen), \here is an 'hiyisible or tran^ndent effect {dmaphala), similarly also.thv pitriiTf'grdhrva music taDs adrsta phal. . ' ''<* H ',< , Some technical terms pertaining o'itoa Bav-boi mtioned here, viz., kal, kaipta, 'ccatpid,'tiitski, firiika mrga

SSSS SSSS SSSS SSSS for eg ; 12 3' p d a margas

catuskal caccatputah

The point \diich Abhinava is trying t make, is, that in the singing of this Jti (probably only, to which he has referred to), it is sung in cacca^uta tla.Th^d is sung in the pancapni tola {Brhadde Vrtti. p.70;'SR 1;7,62) which was a

164

ABHINA VABHARA 77

NOTES

165

I
I

j j
'

different mode of tala, the popular practice was to render it in the vartika marga, which meant medium tonpo. The ila, too, ordinary caccatputahtla. He goes on the say that (theoretically) the real application should have been the catuskam mode (i.e., daksina mrga), however, it is not popular m actual practice. The catuskal mode would have been an extremely vi/amtoa form r slow tempo, perhaps tiiat is why tiie vartika marga or medium tempo was prefmed by people by tiie time of Abhinava. 171. f^'Aan'amusicconsistedo70singingndwasdistinguished from dhruva-gana which consisted of grama ragas.The jti gna Itself had 3 parts: tiiepoifa (words),svara (notes) and tla (time measure). The wara&Aaga consisting of svara andsaihskrta and prai^a pada or words was well known. However, the tla part O gandharva music was exti^emely complicated The tiiree basic - caccatputah, ccaputah and pacapn helped m buUding lie structure of large, elaborate and complex tla forms such as tiie vardhamanaka, seven gakasetc., which have no paraUel in present day tla. Vardhaninaka was made up of a combmation of (to/aconstituents such as) sritas (NS, 31,69).

pp.259-269. Trotika, mentioned here, seems a type oideM composition, set to a de metre. Sorngadeva refras to a Topkaprabandha. SR,2,4,244-45 172.
Dhruva gna was ancient stage music. Five types of dhru^ have been enumerated by Bharata viz. praveila, ksepif, prsdU, antar and niskrmil (NS 32, 310-315) The dhrums suggested acte and moods of different characters m a play, and this was suggested by tiie contente of the songs, s well as their metre; language, tempo and tla. Particular types of dhruvs were to be raidered on particular junctures. The entiance and exit-of characters was indicated through the prveil and naiskrmiki dhruvs respectively. If there was a sudden disturbance in tiie prevailing rasa of tiie scene, by tiie imposition of a new element, thrai the ksepil dhrum was employed. The prevailing rasa which had been disturbed was once again purified and stabilised by the prsd dhruva. The antar dhruva was sung to cover up a fault or mistake by tiie actor during the actual enacting of a play. Dhaivata is the samvdi of rsabha in the sadja-grma. Here, in tiie sadja-grma, dha is an mdispensable note, i.e., it can never be dropped. Hence mthe Sadjodicyava jti,there in noiexatonic roidering with dha as amia. Consequentiy, there are only three hexatonic renderings- witii sa, ma or ui as ama. Saigdeva, too, says - 'dhaivata mie na davam' (SR. 1,7,82).

j I

j to/a-structures
j

173.

Itconsistedoffoursriiasviz.,fen^a,%ntora,mii/iyama andyye^Ao. The different saritas diff^ed as regards tiie words andlaya. Bharata(iWd.31,156)saystiiattiie vardAamKofoiwas so named smce it increased constantiy as regards kals (from 9 to 17 to 33 to 65) words and laya (from ekakal to dvikal to catuskam - tiiough in effect tiie speed is decreasing). The vardhaninaka was tiie tala forni whicti accompanied tiie tondavadanceperformedduringtiieprvarigaoriepreamble of tiie play. The gitaka was a complex tla - cum-melody structure. Ite main importance, however, lay in ite complex and elaborate^ifa. The^seven gJtakas were: madrakti, ullopyaka, aparantka,prakafi, ovenaka,rovindaka and uttara. They could' be rendered in tiie ekakal, dvikal and catuskal modes.
For further details of vardhaninka and gaka see my /I
Historical and Cultural Study of the Ntyasstra of Bharata.

174.

Samgdeva explains that ni is alpa or rare, except when ga is the amsa, or when it (ni) is vadi. niralpo' msdgdrte vditm vin' (ibid 1,7,86) The Vrtti on BrhaddeSi 251 (p.72) elucidates that the sancra in tiiisyoii is according to one's own discretion, as long as it did not violate the rules of the grma - 'grmvirodhena yathestam sancra' The Dattilam, 73., also says the same.This is probably explained by the fact, that, since there were as many as seven amas in thisyaii, it could be sung in a multiplicity of ways. -, Bharata (NS 28,115) and Abhinava both refer to a typical movement in the Gndhn jti 'from rsabha to dhaivata' ('rsab-

175.

176.

168

ABHINAVABHARAH

NOTES

169

Gndhn had a ri-dha movement Pcmcami had two scmcaras -a primary one between ma and ri and a secondary one from ni t ga. Gndhrapancaml included all these movonents. 183. Andhrihzd a typical movement of rirga. Thoe also seons to be a sahgati of ni and dha. This is not made clear by Bharata. However Malanga and Samgadeva state it clearly. rigayomidhayostath // sangatih also S.R. 1,7,105. 184. The text as printed suggests an opposite sense , but leads to a rontrr.diction. The text, therefore needs to be ammoided so that 'rcdtulya is read as 'raktitulya'. This ammendment has beai adopted by Dr. Mukund Lath, but he has not specifically justified it {A Study of Dattilam. p.290) 'anamM ihbhuyah' {anamsas are less) should peiiiaps be 'anamsa ihablmyah (anamsas are prolific). In fact, in the next sentence, Abhinava quotes ViSkhilcrya,sajng that the latter held that anamsas were prolific in this jati. Mataga also says the same - 'Bahavo' ntaramrgatvdanamh parilrtith' / , * * (Brhaddesi, 268), Samgadeva repeats exactly the same (5.1,7,101). In fact, Kallinatha, the commutator of SR,raises a pertinent question in this context : If both the amas and non-omias were to bedominant in this jati,how thai were they to be distinguished from each other ? He answacs by saying that, the amas in the antaramrga w emphasized in the stlm-varm (sthyitvena) and the non-amsas were emphasized in the smca-varna (scmcritven). 186. At the end of this chapter, Abhinava quotes a long passage in which grma-rgasare traced totheir parent jatis.Hie Vrttikara of Brhadde also says that grma-rgas are bom of j'tis and he 'ascribes this statement to Bharata. "j'tisambhttvdgrmargnmiti" - Vrtti on Brhaddesi., 321. Kallinatha, too, says the same (Kalanidhi on S..-2,1,8-14). ./ Brhaddesi 172-173.

In the context of explaining the significa'nce bf thfe term grma of th word grma-rga, he says, that, grma-rgas are not bqm of.grmas but of j'tis, yet, compared't forms such as bhss, rgas etc., they are less removed'from grmas, hence the name ^rma-rgo "Grmayorjtivyavadhnenotpannrimapi bhsrgdyapeksay vyavadMrilpatvdetesin grmargatvavyapadeah" (Kal on SR.l.c.) Abhinava, t6o, says "grmaragas, because grma means a collection of j'tis and connected with that are the excellences of entertainment and raktyatiaya" - (Ata eva hyete grmarg ityuktah grihohij'tisanihastasya sambandhino raktyatiaya iti - AB on NS.29,S). Abhinava includes grma-rgas in the forms that were close to gntf/iarva (gndharvakalpa NS.28,65). In fact, the Vrtti on Brhadde 364, quotes a passage of Kaiyapa which ascribes to tbogrma-r^osthe same ten laksanas which characterise the j'tis namely, ama, nysa, sdava, auduva, alpatva, bahutva, graha, apanysa, mandra, tra. Yet, even though close to the j'tis, the grma-rgas did not belong to the gandharva system, but, to the popular, fi-eer system of music termed gna; their ultimate purpose was not adrsta or transcendental merit (as in gandharva), but to produce pleasure and entertainment. AU the grma-rgas used the antara-gndhara and kakali nisda proUfically. 187. The mixed drink has been mentioned elsewhere too (chapter.32). Since mirageya, i.e., grma-rgas were bom through combining various elements of various j'tis, they were analogus to panaka (a popular mixed drink of the period). However, just as the pnaka, though made by a mixture of spices had a flavour of its own, similarly, the mixed forms, though bom of j'tis, were new and ind^endent musical fomis, with characteristics pecuUar to them and different from the forms they were derived from. However, the parent jti could always be traced by discerning theyaiioc jtiswhose structure dominated theraga form.

185.