Why rasa theory had to come into being, the four historical schools that argued out how it actually works, the full technical apparatus of vibhāva-anubhāva-sthāyibhāva, the evolution of kāvya from mahākāvya to the bhakti kṛti, and a working analysis of how Tyāgarāja, Muttusvāmi Dīkṣitar, Śyāmā Śāstri, Annamācārya, Bhadrācala Rāmadāsu, and Śaṅkarācārya each used this exact machinery.
A theory of rasa is not a decorative overlay added to an already-functioning dramatic and poetic practice. It exists because a specific practical problem has no obvious solution without it: a play or a poem reliably makes an entire hall of strangers feel the same specific emotion at the same moment, despite each spectator carrying a different personal history, different actual griefs, different actual loves — and the emotion produced is not the spectator's own grief or love re-triggered by association, but something that arises, is savoured, and subsides cleanly within the two or three hours of the performance, leaving no residue once the play ends. Bharata's tradition needed to explain three things simultaneously: how a fictional, publicly known-to-be-false situation on stage produces a real felt state in a spectator; why that felt state is pleasurable even when its content is grief, terror, or disgust, none of which are pleasurable in ordinary life; and why the felt state is shared uniformly across a socially and psychologically heterogeneous audience rather than varying idiosyncratically the way ordinary emotional reaction to real events does. No existing framework available to the tradition — neither ordinary psychology of emotion (bhāva as it occurs in life), nor a theory of imitation alone, nor a theory of poetic ornament (alaṅkāra) alone — answers all three questions at once. Rasa theory is the specific answer constructed to close exactly this gap, and its subsequent four-school history, covered in this Part, is four successive attempts to locate precisely where and how the transformation from ordinary bhāva into extraordinary, shared, self-limiting rasa actually takes place.
The problem is compounded by a further fact the tradition took seriously from the outset: rasa cannot be the srota's own sthāyibhāva (dominant emotional disposition) triggered directly, because a spectator who has never experienced romantic union, and one who has experienced it many times, both experience śṛṅgāra rasa watching the same scene, and experience it as comparably intense and comparably pleasurable. If rasa were simply the spectator's private memory reactivated, the two spectators' experiences should differ sharply in both kind and intensity, since one has the relevant memory-material to draw on and the other does not. That they do not differ in this way is the single empirical observation the entire apparatus that follows — vibhāva, anubhāva, sāttvika bhāva, vyabhicāribhāva, sthāyibhāva, sādhāraṇīkaraṇa, camatkāra — is built to account for.
↑ back to contentsNāṭyaśāstra chapter six states the sūtra on which the entire theory rests:
विभावानुभावव्यभिचारिसंयोगाद् रसनिष्पत्तिः। Nāṭyaśāstra 6.31 (numbering varies by recension)
Rasa arises (niṣpatti, literally "accomplishment" or "cooking-to-completion," a word deliberately chosen over a simpler word for "occurs," since niṣpatti carries the specific sense of a process brought to ripened completion, as food is cooked) from the saṃyoga (conjunction, coming-together) of three named components: vibhāva (the situational determinants that give rise to the emotion — who and what is present, and under what circumstances), anubhāva (the deliberate, visible physical responses an actor performs to communicate the emotion outward), and vyabhicāribhāva (the transient, subsidiary emotional states that colour and complicate the dominant emotion as it develops). Bharata's commentators immediately note what the sūtra does not name explicitly but presupposes throughout the surrounding chapter: a fourth component, sthāyibhāva, the single dominant, stable emotional disposition (rati, utsāha, śoka, and so on) that the entire conjunction is in service of bringing to ripened, savourable completion. The sūtra names the three visible, constructed ingredients an actor and playwright supply; sthāyibhāva is the substance those ingredients are cooking. A fifth component, sāttvika bhāva, the involuntary physical symptoms of genuine emotional absorption (horripilation, tears, trembling), is treated by Bharata as a special sub-class of anubhāva rather than a separate category, though later theorists, and this document following them, treat it separately because its involuntary character raises questions the deliberate anubhāvas do not.
The sūtra's grammatical construction is itself part of its content, a point developed at length in companion documents on Sanskrit grammar and Bharata's use of the causal instrumental. What concerns this section is only the substantive claim: rasa is not a fourth, independently supplied ingredient added alongside the other three, but the name for what results when the other three are correctly combined and brought to completion around an already-present sthāyibhāva. Removing any one of the three named components — situating the emotion without a visible response, or producing a response with no situational cause, or producing a stable dominant emotion with no transient complicating states around it — does not produce a weaker rasa; the commentarial tradition is emphatic that it produces no rasa at all, only bhāva, ordinary unripened emotion, which a spectator may recognize intellectually as depicted but does not experience as the specific extraordinary savour rasa names.
↑ back to contentsVibhāva is subdivided into two functionally distinct kinds. Ālambana-vibhāva (the supporting or foundational determinant) is the person or object the emotion is directed toward — the beloved in śṛṅgāra, the enemy in raudra, the corpse in bībhatsa. Uddīpana-vibhāva (the excitant determinant) is everything surrounding the ālambana that intensifies the emotion already directed at it without itself being the object of the emotion — moonlight, a garden in bloom, a cuckoo's call in śṛṅgāra; a burning cremation ground, vultures, or a moonless night in bībhatsa or bhayānaka. The distinction matters because it specifies a testable structural claim: removing the ālambana removes the rasa entirely, since there is no longer an object for the emotion to attach to, while removing or varying the uddīpana changes only the rasa's intensity or specific flavour, not its basic identity — the same lovers under moonlight versus under an ordinary afternoon sky remain in śṛṅgāra rasa in both cases, at different intensities, whereas removing one lover from the scene changes the rasa's category outright, typically toward vipralambha (separation) śṛṅgāra or toward karuṇa.
A further distinction Abhinavagupta insists on, developed fully in the school-history sections below, is that the vibhāva of dramatic art is deliberately generalized (sāmānya) rather than particular (viśeṣa): the ālambana on stage is not "this specific historical woman, Śakuntalā, whom the actor happens to be portraying," presented as a report of a real, particular person's real circumstance, but a generalized romantic situation any spectator can occupy the emotional position of, regardless of whether their own particular beloved resembles the actress at all. This generalization is not a simplification imposed on a richer particular reality; it is, on Abhinavagupta's account, the specific mechanism that makes rasa universally accessible to a heterogeneous audience rather than accessible only to spectators whose particular circumstances happen to match the depicted particulars — a claim examined in full in the sādhāraṇīkaraṇa discussion below.
↑ back to contentsAnubhāva is the actor's consequent, deliberate physical enactment of the emotion the vibhāva has situated — the glance, the gait, the specific hand gesture (hasta) and eye movement (dṛṣṭi) catalogued exhaustively across the Nāṭyaśāstra's āṅgika abhinaya chapters. The term's etymology is precise: anu-bhāva, that which follows from and makes known (bhāvayati) the bhāva that causes it. Anubhāva is deliberately controllable and trained — an actor rehearses and repeats a specific glance for śṛṅgāra until it can be produced reliably on cue, night after night, which is precisely what distinguishes it from the sāttvika bhāvas discussed next. Anubhāva's function in the rasa-sūtra is communicative rather than merely expressive: its purpose is to make the actor's, and by extension the character's, internal state legible to a spectator seated at a physical distance, under stage lighting, without access to any information beyond what is visibly performed — a constraint that explains why the Nāṭyaśāstra catalogues anubhāva with such granularity (thirteen head positions, thirty-six eye states, twenty-four single-hand gestures, thirteen standing postures), since a communicative code intended to be legible at a distance to an audience with no other access to the actor's internal state needs to be far more standardized and far more finely differentiated than the comparatively loose, idiosyncratic expressive repertoire any individual naturally produces in ordinary, un-staged emotional life.
↑ back to contentsWhere anubhāva is trained and deliberately produced, sāttvika bhāva names the eight involuntary physiological responses Bharata holds cannot be produced by will alone, and whose presence is therefore treated as certifying that an actor's absorption in the depicted emotion has become genuine rather than merely technically correct: स्तम्भ (stambha, paralysis/freezing), स्वेद (sveda, sweating), रोमाञ्च (romāñca, horripilation), स्वरभेद (svarabheda, voice-breaking), वेपथु (vepathu, trembling), वैवर्ण्य (vaivarṇya, change of complexion/pallor), अश्रु (aśru, tears), and प्रलय (pralaya, fainting/loss of consciousness). The traditional position is not that these responses are literally impossible to fake — a sufficiently skilled actor can produce a passable simulated tremor — but that the tradition builds its pedagogy of sāttvika around the premise that a spectator can perceptually distinguish a merely technically correct anubhāva from one accompanied by genuine sāttvika absorption, and that this distinction is precisely what separates an adequate performance from one capable of producing full rasa-niṣpatti in the audience. The eight sāttvika bhāvas therefore function within the theory as the specific, named diagnostic by which "the actor is technically executing correct abhinaya" is distinguished from "the actor has actually arrived, even momentarily, at the emotional state being portrayed" — a distinction with no equivalent granularity in any comparably early theory of acting outside this tradition.
↑ back to contentsVyabhicāribhāva (also called saṃcāribhāva, "that which moves along with" the dominant emotion) names the thirty-three subsidiary, transient emotional states that arise, colour, and subside within the span of a single scene while a single sthāyibhāva remains dominant throughout: nirveda (despondency), glāni (weakness), śaṅkā (apprehension), asūyā (envy), mada (intoxication), śrama (fatigue), ālasya (indolence), dainya (wretchedness), cintā (anxiety), moha (delusion), smṛti (recollection), dhṛti (contentment), vrīḍā (shame), capalatā (fickleness), harṣa (joy), āvega (agitation), jaḍatā (stupor), garva (pride), viṣāda (dejection), autsukya (impatience/longing), nidrā (drowsiness), apasmāra (epilepsy/ possession), supta (sleep), vibodha (waking), amarṣa (indignation), avahittha (dissimulation), ugratā (ferocity), mati (deliberation), vyādhi (sickness), unmāda (madness), maraṇa (death), trāsa (fright), and vitarka (conjecture). The functional distinction this list encodes against sthāyibhāva is durational and structural rather than merely intensity-based: a sthāyibhāva persists as the governing emotional thread of an entire act or entire play, absorbing and being coloured by whichever vyabhicāribhāvas arise and subside within it, while no single vyabhicāribhāva persists across the whole; a scene of śṛṅgāra rasa may move through harṣa, then śaṅkā, then autsukya, in sequence, while rati (the sthāyibhāva of śṛṅgāra) remains the unbroken thread the entire time. The number thirty-three is not arbitrary padding; it is presented as an exhaustive catalogue of every named transient state the tradition holds a dramatic scene needs, since a playwright reaching for an emotional colour outside this list is, on the theory's own terms, reaching for something the system does not recognize as a distinct dramatically usable state.
↑ back to contentsBharata names eight sthāyibhāva-rasa pairs; śānta, and its corresponding sthāyibhāva śama, is added by later theorists (the addition is already defended at length in Abhinavagupta's Abhinavabhāratī and is universally accepted by the time of the developed theory), bringing the canonical count to nine — the navarasa. Each pairing also carries a conventional presiding deity and colour association, catalogued for performative and iconographic use:
| Rasa | Sthāyibhāva | Colour | Presiding deity | Core vibhāva domain |
|---|---|---|---|---|
| Śṛṅgāra | Rati (love) | Śyāma / dark blue-green | Viṣṇu | Union and separation of lovers |
| Hāsya | Hāsa (mirth) | Śveta / white | Pramatha (Śiva's attendant) | Incongruity, deformity, mimicry |
| Karuṇa | Śoka (grief) | Kapota / dove-grey | Yama | Loss, separation, calamity |
| Raudra | Krodha (anger) | Rakta / red | Rudra | Insult, betrayal, injustice |
| Vīra | Utsāha (energy/determination) | Gaura / pale gold | Indra | Combat, generosity, righteousness |
| Bhayānaka | Bhaya (fear) | Kṛṣṇa / black | Kāla | Threat, the unknown, isolation |
| Bībhatsa | Jugupsā (disgust) | Nīla / blue | Mahākāla | Decay, filth, the repugnant |
| Adbhuta | Vismaya (wonder) | Pīta / yellow | Brahmā | The marvellous, the divine, the unprecedented |
| Śānta | Śama (tranquility) | Śveta / white (pure) | Nārāyaṇa / the self | Renunciation, self-knowledge, cessation of desire |
Two structural rules govern the table beyond the individual pairings. First, sthāyibhāva-rasa correspondence is theorized as convertible only in one direction inside a given performance: a sthāyibhāva can ripen into rasa given the correct vibhāva-anubhāva-vyabhicāribhāva conjunction, but a rasa cannot revert to being merely one spectator's private bhāva without breaking the aesthetic frame entirely — which is precisely the failure mode Bharata's tradition calls rasābhāsa (a spoiled or misdirected rasa), discussed further in the composer-application sections below. Second, the later Kashmiri aestheticians, Abhinavagupta chief among them, hold śānta to be architecturally prior to the other eight rather than merely the ninth addition to a completed list — since śama, tranquil selfhood undisturbed by any of the other eight dominant emotions, is the ground state every other rasa arises from and eventually subsides back into once the performance ends, giving śānta a foundational role in aesthetic theory that directly parallels its role in the darśana traditions this document's companion Śāstrex Vāk series treats at length.
↑ back to contentsBharata's tradition names the specific failure mode of the rasa-sūtra's mechanism rasābhāsa, "the semblance of rasa" — a case in which the vibhāva-anubhāva-vyabhicāribhāva apparatus is assembled and a felt response results, but the response fails to meet one or more of the conditions Part I identified as rasa theory's founding requirements, most commonly the requirement of propriety (aucitya) between the ālambana chosen and the sthāyibhāva directed at it. The standard classical example is Rāvaṇa's śṛṅgāra toward Sītā in the Rāmāyaṇa itself: the situational apparatus of śṛṅgāra — ālambana, uddīpana, the full vyabhicāri palette of longing and frustration — is fully present in Rāvaṇa's own experience, yet the tradition holds that a spectator does not, and on aucitya grounds should not, relish this as ordinary śṛṅgāra rasa, because the ālambana (another man's unwilling wife) violates the propriety condition rasa's universalizing mechanism depends on: a rasa experience a spectator would be ashamed to actually inhabit, rather than merely observe at the generalized remove sādhāraṇīkaraṇa provides, is graded not as full rasa but as its semblance. Rasābhāsa is not a minor footnote to the theory; it is the specific concept that keeps rasa theory from collapsing into a claim that any emotionally arousing conjunction whatsoever counts as legitimate rasa, and it is the theoretical resource later composers draw on, explicitly or implicitly, whenever they need to distinguish devotional śṛṅgāra of the Annamācārya-Tyāgarāja kind from an accusation, occasionally raised by less sympathetic readers, that madhura-bhakti poetry is merely erotic poetry given a devotional gloss: the tradition's answer is that the ālambana (the Lord) and the aucitya of the relationship (the soul's rightful, non-illicit longing for its own source) are precisely what keep madhura bhakti's śṛṅgāra from sliding into rasābhāsa, where Rāvaṇa's parallel case does not meet the same aucitya condition.
↑ back to contentsBhaṭṭa Lollaṭa (circa ninth century, known only through fragments preserved in later commentators' refutations of him, chiefly Abhinavagupta's Abhinavabhāratī) proposes the earliest surviving systematic account of where rasa is located: utpattivāda, the doctrine of production. On this view, the sthāyibhāva named in the rasa-sūtra genuinely belongs, in the first instance, to the character being portrayed (Rāma, say, in his rati toward Sītā) — a real, if fictional, particular emotion intensified by the vibhāva-anubhāva-vyabhicāribhāva conjunction into rasa. The actor, through sustained identification with the role (anusandhāna, "thorough application" or "identification"), comes to actually possess this same intensified sthāyibhāva during performance, and the audience, perceiving the actor's genuinely intensified emotion, comes to share in it derivatively. Rasa, on this account, is a real emotion, actually produced (utpanna), first in the character, transferred to the actor through disciplined identification, and from the actor to the spectator through perception.
Lollaṭa's account was subjected to sustained and largely successful attack precisely because of the three requirements the opening section of this Part named. If rasa is a real, actually-produced emotion, located first in a fictional character and transferred by identification to an actor, then a spectator watching Rāma grieve for a genuinely dead Daśaratha ought to feel actual bereavement of the specific, particular kind associated with a father's death — yet spectators demonstrably do not leave the theatre in mourning, nor do they attempt to console the actor playing Rāma as they would console an actually bereaved person. Lollaṭa's account, in other words, explains how an emotion could be produced and transmitted, but not why the transmitted emotion is experienced as pleasurable, self-limiting, and universally shared rather than as ordinary grief accidentally caught by contagion from watching someone else's real distress — the specific gap the next school addresses directly.
↑ back to contentsŚrī Śaṅkuka's response, anumitivāda (the doctrine of inference), relocates rasa away from any claim that a real emotion is produced anywhere in the chain, and toward a claim about how a spectator cognizes a represented emotion. Śaṅkuka's central technical move is his account of the actor's ontological status: the actor playing Rāma is neither identical to Rāma (the spectator knows perfectly well the man on stage is an actor, not the historical or legendary Rāma), nor entirely different from Rāma in the spectator's perception (if the actor were perceived as simply "a random man performing arbitrary gestures," no rasa would arise at all) — the actor is perceived as rāma-sadṛśa, "similar to Rāma," through a specific cognitive operation Śaṅkuka names using the analogy of citra-turaga-nyāya, "the maxim of the painted horse": a viewer looking at a painted horse does not believe a real horse is present, nor dismiss the painting as containing no horse at all, but apprehends "horse-ness" through the painting via a distinct cognitive act of resemblance-apprehension, neither straightforward perception of a real horse nor straightforward non-perception. Rasa, on this account, is the spectator's inferred cognition (anumiti) of the sthāyibhāva as belonging to the imitated character, arrived at via the actor's skillfully executed vibhāva-anubhāva combination functioning as the inferential grounds (liṅga) from which the audience infers (anumāna) the underlying emotion, exactly as smoke functions as inferential grounds from which fire is inferred in the standard Nyāya example.
Anumitivāda solves Lollaṭa's central problem — the spectator is not catching a real transmitted emotion, hence no actual bereavement follows watching Rāma's grief — but generates a new and, to later critics, equally serious problem: inference (anumāna) is, in every other domain Nyāya epistemology treats it, a cool, intellectual, propositional cognition ("there is fire on the hill, because there is smoke"), not an experience carrying felt pleasure or emotional colour at all. If rasa is simply an inferential judgment that "Rāma is grieving," structurally identical to the cognition "there is fire on the hill," Śaṅkuka's account fails to explain why watching a rasa-laden performance is experienced as delightful, absorbing, and qualitatively unlike drawing a logical conclusion from evidence — the specific gap Bhaṭṭa Nāyaka's account addresses next.
↑ back to contentsBhaṭṭa Nāyaka (also known largely through Abhinavagupta's citations and refutations, since his own text is lost) introduces the concept that ultimately settles the dispute, even though his own overall theory is itself later revised: sādhāraṇīkaraṇa, "generalization" or "universalization." Nāyaka's account proceeds through two named operative functions of poetic language beyond the ordinary denotative (abhidhā) and figurative (lakṣaṇā) functions treated in kāvya theory generally: bhāvakatva ("the power of generalizing," or "actualization-power"), by which the poetic vibhāva-anubhāva-vyabhicāribhāva conjunction strips the depicted sthāyibhāva of its particular, biographically specific attachments — this is not Rāma's grief for his specific, historically particular father, nor any one spectator's grief for their own specific father, but grief-as-such, generalized and detached from any single knower's biographical particulars — and bhojakatva or bhogakṛttva ("the power of enabling relishing"), by which this generalized sthāyibhāva, once stripped of particularity, becomes available to be tasted or relished (bhukti, hence bhuktivāda, "the doctrine of relishing") by every spectator uniformly, since a generalized emotion belongs to no one spectator's particular biography more than another's and is therefore equally available to all. Nāyaka reaches for a further analogy from Mīmāṃsā ritual theory to describe this relishing: rasa-experience is compared to the transcendent, non-worldly delight (alaukika-camatkāra) associated with realizing a Vedic injunction's meaning (bhāvanā) — neither ordinary worldly pleasure nor mere intellectual cognition, but a third, distinct category of experience.
This is the decisive move for the whole subsequent history of the theory, because sādhāraṇīkaraṇa directly answers the empirical observation the opening section of this Part named as rasa theory's founding problem: a spectator with no personal experience of romantic union and a spectator with extensive such experience feel comparably intense śṛṅgāra rasa watching the identical scene, because what each is relishing is not their own particular biographical memory reactivated, but a generalized rati that neither spectator's particular history is more or less entitled to access than the other's. Nāyaka's own overall architecture — the specific claim that bhāvakatva and bhojakatva are functions distinct from and irreducible to abhidhā, lakṣaṇā, and vyañjanā — is rejected by Abhinavagupta on the grounds treated in the next section, but the sādhāraṇīkaraṇa concept itself survives Nāyaka's own theory intact and becomes, in a revised form, the load- bearing mechanism of Abhinavagupta's own final account.
↑ back to contentsAbhinavagupta's synthesis, developed across his Abhinavabhāratī commentary on the Nāṭyaśāstra and his Locana commentary on Ānandavardhana's Dhvanyāloka, retains sādhāraṇīkaraṇa as the correct answer to how a heterogeneous audience shares one experience, but relocates the mechanism producing that generalization into the dhvani-theoretic apparatus this document's companion volume on Ānandavardhana treats in full: rasa is not inferred (against Śaṅkuka), nor actually produced and transmitted (against Lollaṭa), nor the output of two freestanding functions separate from ordinary poetic suggestion (against Nāyaka's specific architecture), but is itself the supreme, terminal case of vyañjanā — poetic suggestion — operating at maximum intensity: rasa-dhvani. Rasa is never directly stated (vācya) by any dramatic or poetic text; it is entirely suggested, through the correctly assembled vibhāva-anubhāva-vyabhicāribhāva conjunction, to a spectator whose sthāyibhāva is already present as a latent disposition (vāsanā) universal to all human minds, requiring only the correct suggestive stimulus to be awakened into conscious relishing. Sādhāraṇīkaraṇa, on this revised account, is not a separate operative power alongside abhidhā and lakṣaṇā as Nāyaka held, but simply what happens automatically whenever suggestion (vyañjanā) rather than direct statement (abhidhā) is the operative channel — a suggested meaning, unlike a directly stated one, is never anchored to a specific named particular in the way a direct statement is, and is for that structural reason already general rather than particular the moment it is grasped through suggestion rather than assertion.
Abhinavagupta names the resulting experience camatkāra — a wondering, astonished relishing, structurally likened, following Nāyaka's earlier Mīmāṃsā-derived analogy but developed with considerably more precision, to brahmāsvāda-sadṛśa, "resembling the tasting of brahman": not identical to liberative self-knowledge, since rasa-experience is temporary and dependent on external stimulus while brahmāsvāda is neither, but sharing brahmāsvāda's specific qualitative signature of being a pure, undivided, self-luminous relishing untouched by the ordinary subject-object, self-other fragmentation that characterizes worldly pleasure and pain alike. This is the specific philosophical payoff of locating rasa in vyañjanā rather than in inference or transmission: a suggested meaning is grasped by the mind (pratyabhijñā-adjacent recognition, drawing on Abhinavagupta's own Śaiva non-dual commitments developed at length elsewhere in his corpus) in a single, unmediated flash rather than through the sequential subject-predicate structure that ordinary inference and ordinary perception both require, and it is exactly this unmediated, flash-like quality that both distinguishes rasa from Śaṅkuka's cool inferential cognition and explains its structural resemblance to non-dual self-recognition. The four-school history, read end to end, is therefore not simply an accumulation of refinements arriving at a finally correct answer by trial and error, but a genuine convergence: each school inherits and is forced to solve the specific unresolved problem the previous school's account left open, ending in an account that ties dramatic and poetic theory directly into the tradition's own most developed epistemology of unmediated cognition.
↑ back to contentsBefore dhvani theory reorganized poetics around suggestion, an earlier stratum of kāvyaśāstra, running from Bhāmaha and Daṇḍin through Vāmana, analyzed a poem's aesthetic success through three more surface- level categories that remain load-bearing throughout the later tradition, including in every composer examined in Part V. Guṇa (poetic quality or virtue) names properties of style such as mādhurya (sweetness, associated with śṛṅgāra and karuṇa, short compact words and soft consonant clusters), ojas (vigour, associated with vīra and raudra, long compounds and hard consonant clusters), and prasāda (clarity, a transparency of sense regardless of subject matter). Rīti (regional style or diction) names the historically recognized compositional idioms — Vaidarbhī (associated with mādhurya and compact phrasing), Gauḍī (associated with ojas and compound density), Pāñcālī (an intermediate style) — each a recognizable, nameable combination of guṇas a poet could deliberately select to suit the rasa being composed for. Alaṅkāra (figuration) supplies the specific rhetorical devices — upamā (simile, an explicit comparison marked by a comparative particle), rūpaka (metaphor, identification without a comparative marker), utprekṣā (poetic fancy, imaginatively conceiving one thing as another), vyatireka (distinction, in which a compared object is asserted to exceed its standard of comparison), and dozens more catalogued exhaustively by Daṇḍin, Bhāmaha, and Ānandavardhana's own predecessors — through which vibhāva and uddīpana are actually rendered in words. These three categories do not compete with dhvani theory's later suggestion-based reorganization; Ānandavardhana's own explicit position is that guṇa, rīti, and alaṅkāra remain fully valid and necessary, but are properly subordinate (aṅga) to dhvani, functioning as the surface-level craft through which suggested meaning is delivered rather than as the poem's own primary aesthetic achievement. This subordination is the exact structural point Śaṅkarācārya's corpus, examined in section twenty-five, depends on: Saundaryalaharī's nakha-śikha description of the Devī is composed with full command of guṇa, rīti, and alaṅkāra at the surface level, while its actual doctrinal payload operates one level down, at dhvani's suggested level — precisely the layered relationship the earlier alaṅkāra-focused theorists' own subordination thesis predicts.
↑ back to contentsThe Rāmāyaṇa's own Bāla Kāṇḍa supplies the tradition's principal account of where kāvya, poetry as a formal art distinct from ordinary utterance, actually originates. Vālmīki, witnessing a hunter kill one of a mating pair of krauñca birds, hears the surviving bird's cry of grief and, without premeditation or deliberate technical composition, produces the first metrically perfect śloka:
मा निषाद प्रतिष्ठां त्वमगमः शाश्वतीः समाः।
यत्क्रौञ्चमिथुनादेकम् अवधीः काममोहितम्॥ Rāmāyaṇa, Bāla Kāṇḍa, sarga 2
The tradition's own gloss on the moment turns on a paronomasia treated as doctrinal rather than incidental wordplay: śoka (grief) becomes śloka (verse). The substantive claim embedded in the account is precise and worth separating from the poetic charm of the story itself: metrical, rule-governed form is not something subsequently and deliberately applied to raw, formless grief by a craftsman working after the emotional event has passed; it is what sufficiently intense grief becomes, spontaneously and without deliberation, when it arises in a mind already thoroughly saturated in Chandas — the metrical science this document's companion volume on Nāda-Vāk-Śāstra treats in full. The account is therefore doing exactly the same explanatory work for kāvya that the rasa-sūtra does for nāṭya: both locate the origin of formal aesthetic structure in a witnessed emotional event crystallizing into ordered sound, rather than in a craftsman's separate, subsequent technical invention layered onto an emotion that already existed complete in itself before the form arrived.
This has a direct consequence for how kāvyotpatti and rasautpatti relate to each other as accounts, rather than remaining two unconnected origin myths standing side by side. If grief only becomes śloka in a mind already saturated in metrical form, the capacity for a witnessed particular event (a hunter's arrow, a single bird's cry) to crystallize into universal, generalized, shareable verse — verse any reader encountering the Rāmāyaṇa for the first time can relish as karuṇa rasa without having witnessed any bird being killed themselves — is already, in miniature, the same sādhāraṇīkaraṇa operation Bhaṭṭa Nāyaka and Abhinavagupta later formalize at the level of dramatic theory generally. Vālmīki's krauñca-vadha verse is not merely the traditional example of where kāvya began; on this reading it is the tradition's own founding demonstration case of sādhāraṇīkaraṇa itself, centuries before the concept received its technical name.
↑ back to contentsNāṭyaśāstra chapter one supplies nāṭya's parallel origin account, and states its own motivating problem with unusual explicitness rather than leaving it implicit in a single poetic image. In an age already in decline, marked by the text's own diagnosis of rising kāma (desire), lobha (greed), krodha (anger), and mātsarya (envy) across the four yugas, the gods petition Brahmā for a new mode of instruction, one accessible to all four varṇas without exception, including śūdras, who stood barred from direct study of the four Vedas proper. Brahmā's response is the composition of a fifth Veda, the Nāṭyaveda, formed by extraction from each of the existing four in turn: pāṭhya (recited text) from the Ṛgveda, gīta (song) from the Sāmaveda, abhinaya (enactment) from the Yajurveda, and rasa itself from the Atharvaveda — explicitly so that the instruction embedded in nāṭya could reach an audience through what can be seen and felt, rather than only through Sanskrit recitation competence the source Vedas structurally required and the excluded audience structurally lacked.
This is worth reading as a genuine design specification rather than only a mythic charter, because it names a precise standard nāṭya is committed to meeting: sarvavarṇika, belonging without exception to every social stratum, meaning no spectator's birth-category or formal educational access should determine whether the emotional-aesthetic instruction nāṭya delivers reaches them. Read alongside kāvyotpatti, the two accounts divide a single underlying claim about aesthetic origin along two different axes rather than stating two unrelated theories: kāvyotpatti addresses the individual, private moment at which witnessed emotion first crystallizes into structured, generalizable form inside a single composer's mind; nāṭyotpatti addresses the collective, institutional problem of how that crystallized form is then delivered to an audience whose access to any single register or medium of instruction cannot be assumed uniform. Kāvya without nāṭya's delivery-mandate risks remaining a private crystallization accessible only to readers who already share the composer's Sanskrit derivational literacy; nāṭya's sarvavarṇika mandate without kāvya's krauñca-vadha mechanism would specify an accessible delivery vehicle with nothing structurally proven capable of carrying genuine aesthetic weight through it. The bhakti-kīrtana tradition examined later in this document is best understood as the historical point at which both requirements are finally met inside a single, small-scale, non-dramatic form — a claim developed fully in the sections on the kṛti below.
↑ back to contentsDaṇḍin's Kāvyādarśa, the earliest surviving systematic treatise on poetics as a formal discipline distinct from grammar and dramaturgy proper, establishes the foundational division kāvya theory works with throughout its subsequent history: śravya kāvya ("kāvya meant to be heard" — that is, read or recited, without staged visual enactment) against dṛśya kāvya ("kāvya meant to be seen" — nāṭya proper, requiring staged bodily and visual enactment through the four abhinaya channels this document's companion volume treats at length). The division is not a hierarchy in which one form is a lesser or incomplete version of the other; both are held capable of producing full rasa-niṣpatti, differing only in which of the sarvavarṇika mandate's access-channels each relies on — śravya kāvya reaching its audience through sound and recited meaning alone, dṛśya kāvya adding the further channels of visible bodily enactment, costume, and stagecraft.
↑ back to contentsWithin śravya kāvya, Daṇḍin and subsequent theorists distinguish forms chiefly by scale and structural completeness. Mahākāvya (the great, sustained narrative poem — Kālidāsa's Raghuvaṃśa and Kumārasambhava are the standard exemplars) is required to meet a specific, checkable set of formal criteria: division into sargas (cantos), a dhīrodātta nāyaka (a noble, self-possessed hero drawn from itihāsa or purāṇa rather than freely invented), inclusion of conventional descriptive set-pieces (city, ocean, mountain, season, sunrise, sunset, love-union, separation, battle), and a single dominant rasa (usually śṛṅgāra or vīra) sustained across the whole while other rasas appear subordinately within it. Khaṇḍakāvya ("fragment-poem") relaxes mahākāvya's scale requirement, treating a single episode or a shorter continuous narrative without the full descriptive apparatus mahākāvya mandates — Kālidāsa's Meghadūta, a single sustained message-poem rather than a multi-canto epic, is the standard example. Muktaka ("detached, freed") relaxes narrative continuity itself: a single self-contained verse, complete in its own right and requiring no surrounding context to deliver its full rasa, collected in anthologies of subhāṣita (well-said verses) rather than embedded in a continuous story. The historical trajectory across these three forms is one of progressive compression — mahākāvya's full descriptive and narrative apparatus is progressively stripped away across khaṇḍakāvya and muktaka, leaving, at the muktaka end of the spectrum, the minimal unit capable of independently producing rasa- niṣpatti without any surrounding narrative scaffold at all. This compression trajectory is the direct formal ancestor of the still-further-compressed devotional lyric examined in the sections that follow.
↑ back to contentsGadyakāvya (prose poetry, chiefly the ākhyāyikā and kathā forms, Bāṇa's Kādambarī and Harṣacarita the standard exemplars) and campū (a mixed prose-and- verse form combining gadya's narrative flexibility with the metrical density verse alone can carry) extend kāvya's formal range without altering the basic śravya/dṛśya division. Running alongside this narrative lineage, and structurally more important for the bhakti tradition examined below, is the parallel and much older lineage of stotra and stava — direct address and praise-poetry to a deity, structurally closer to muktaka's compressed, self-contained verse-unit than to mahākāvya's sustained narrative, but distinguished from ordinary muktaka by its addressee: not a generalized ālambana-vibhāva in Abhinavagupta's technical sense, but the specific deity being addressed, named, and praised directly, in the second person, often by the poet's own first-person voice rather than through a third-person narrated character. This first-person, directly addressed structure is the specific formal innovation the bhakti movement inherits and builds its entire subsequent lyric architecture upon, and it marks a genuine departure from the third-person narrated ālambana-vibhāva structure mahākāvya and dṛśya kāvya alike are built around: in a stotra, the poet is not depicting someone else's rati or utsāha for an audience to relish at one remove through sādhāraṇīkaraṇa, but declaring their own bhakti directly to its object, collapsing the distance between depicted character and addressing poet that the earlier dramatic and narrative forms had structurally required.
↑ back to contentsThe specific formal step from stotra's direct praise-address toward the padam (the Telugu devotional lyric form Annamācārya's corpus is built from, examined in full below) reintroduces something stotra's direct first-person address had set aside: a depicted relational situation, but now with the deity cast not as the addressee of direct praise but as the nāyaka (hero/beloved) of an explicitly śṛṅgāra-structured relationship, and the devotee's soul (jīva) cast as the nāyikā (heroine) — the specific literary-theological structure the tradition calls madhura bhakti or the nāyaka-nāyikā bhāva, with clear precedent in the Rādhā-Kṛṣṇa literature that develops in close historical parallel. This is not a lapse from stotra's directness back into mahākāvya's depicted-character distance; it is a deliberate redeployment of śṛṅgāra rasa's entire technical apparatus — ālambana and uddīpana vibhāva, viraha (separation) and sambhoga (union) as the two structural poles, the full vyabhicāribhāva palette of longing, jealousy, and reconciliation — toward a referent (the individual soul's relationship to the divine) that ordinary śṛṅgāra kāvya was never built to address. The formal payoff of this redeployment, examined at length in the Annamācārya section below, is that it lets rasa theory's entire universalizing, sādhāraṇīkaraṇa-based machinery — already proven, across the school-history covered in Part II, to make a generalized emotion equally available to every listener regardless of their private biography — do the specific theological work of making bhakti itself, not merely romantic love, universally accessible and universally relishable, exactly as the sarvavarṇika mandate originally demanded.
↑ back to contentsThe Carnatic kṛti — the tripartite pallavi (opening refrain), anupallavi (secondary section elaborating the pallavi's theme at a higher register), and caraṇa (one or more verses developing the theme further, typically returning to the pallavi's melodic material) — is best read as the historical point at which several previously separate lineages this Part has traced finally converge into one small-scale performative unit. From muktaka, the kṛti inherits self-containment: a single kṛti, unlike a mahākāvya canto, requires no surrounding narrative context to deliver full rasa-niṣpatti on its own. From stotra, it inherits direct first-person address to a named deity. From the padam and the nāyaka-nāyikā bhāva, it inherits, where the composer chooses to use it, the full relational and śṛṅgāra-structured apparatus for expressing bhakti rather than only declarative praise. And from nāṭyotpatti's own founding fifth-Veda account, it inherits — quite literally, not merely by loose analogy — the identical four-source integration Brahmā's origin-composition is said to perform: pāṭhya (the kṛti's sāhitya, its composed Telugu, Sanskrit, or Manipravalam text) corresponds to the recited-text source; gīta (the kṛti's rāga-bound melodic setting) corresponds directly; abhinaya's role is carried, in the kṛti's typical performance context of individual or small-ensemble rendering rather than staged drama, by the singer's own vocal and gestural nuance (bhāva-abhivyakti) rather than full four-channel dramatic abhinaya; and rasa remains rasa, now aimed in the overwhelming majority of the Carnatic kṛti repertoire at bhakti rather than at the eight worldly rasas mahākāvya and dṛśya kāvya more typically deploy. The kṛti is, in this precise sense, not simply "devotional music" as a separate genre standing apart from the kāvya and nāṭya theory traced through this document, but the specific small-scale form in which that entire theoretical apparatus — genesis account, rasa mechanics, sādhāraṇīkaraṇa, and the sarvavarṇika delivery mandate together — finds its most portable and individually performable expression, requiring neither a staged dramatic production nor a sustained multi-canto narrative to deliver complete rasa-niṣpatti to a single listener in a few minutes of performance.
↑ back to contentsThe kṛti's melodic and rhythmic setting is not a neutral carrier applied after the sāhitya's rasa is already complete; rāga and tāla carry independent rasa-association inherited directly from the gāndharva material this document's companion volume treats in full — the conventional association of specific rāgas with specific emotional registers (a rāga such as Śubhapantuvarāḷi conventionally associated with karuṇa and bhakti-laden pathos, a rāga such as Hamsadhvani conventionally associated with a bright, invocatory adbhuta-adjacent quality suited to opening or auspicious kṛtis) functions as a further, musically carried uddīpana-vibhāva operating alongside, and reinforcing, whatever uddīpana the sāhitya's own imagery supplies. A composer's choice of rāga for a given kṛti is therefore best read as a rasa- theoretic decision in the fullest technical sense — an uddīpana selection precisely analogous to a mahākāvya poet's choice of vasantatilakā or mandākrāntā meter for a given passage, discussed in the companion Chandas material — rather than as an independent musical decision made separately from, and only loosely coordinated with, the text's already-complete emotional content.
↑ back to contentsŚṛṅgāra rasa, the single most technically elaborated of the nine, is itself internally divided into two structurally distinct conditions the tradition treats almost as separate compositional problems requiring different vibhāva configurations: sambhoga-śṛṅgāra (union — the ālambana physically present, the vyabhicāribhāvas skewing toward harṣa, capalatā, and garva) and vipralambha-śṛṅgāra (separation — the ālambana absent, remembered, or withheld, the vyabhicāribhāvas skewing toward autsukya, śaṅkā, glāni, cintā, and smṛti). The two conditions are not graded, with union treated as the fuller or more complete form and separation as a lesser, deficient version of it; classical theory holds vipralambha capable of a sustained intensity and length sambhoga cannot match, precisely because absence, unlike presence, permits indefinite elaboration of longing without the scene resolving. Kālidāsa's Meghadūta, discussed in Part IV as the standard khaṇḍakāvya exemplar, is built entirely on sustained vipralambha with no sambhoga scene at all. This division is the specific technical resource the entire madhura-bhakti lineage examined in Part V depends on: because the devotee's ordinary condition relative to the divine is one of separation rather than constant union, vipralambha- śṛṅgāra, not sambhoga, is structurally the dominant compositional mode available to Annamācārya's virahiṇī padams and to Tyāgarāja's "Nagumōmu" alike, and the sustained, unresolved intensity classical theory already attributes to vipralambha over sambhoga is precisely what gives devotional longing-poetry its characteristic ability to be elaborated at length without the scene needing to resolve into an actual depicted union, which would in most cases be theologically as well as dramaturgically premature.
↑ back to contentsAnnamācārya (fifteenth century, Tallapāka lineage, over twelve thousand padams reportedly composed for Veṅkaṭeśvara of Tirumala) is the historical figure whose corpus most fully realizes the nāyaka-nāyikā bhāva described in the previous Part. A substantial portion of his śṛṅgāra-padams describe, with full technical command of ālambana and uddīpana vibhāva, a directly erotic relationship between Veṅkaṭeśvara and Padmāvatī (Alamēlumaṅga) — union, adornment, playful quarrel, reconciliation — composed with the same descriptive density a court poet would apply to any royal nāyaka-nāyikā pair. Read only at this literal level, these padams are straightforward śṛṅgāra kāvya with a divine couple as ālambana. The doctrinal weight of Annamācārya's corpus, however, sits in a second class of padams, addressed by the poet's own voice directly to Veṅkaṭeśvara, in which the poet-devotee occupies the nāyikā's position himself — cast as virahiṇī (the woman suffering separation), pleading, complaining, and longing for the Lord exactly as a nāyikā would for an absent nāyaka. This double register is the padam form's central technical achievement: the same apparatus of ālambana-vibhāva (the Lord's beauty, ornament, and specific iconographic attributes catalogued with the same granularity āhārya abhinaya theory demands), uddīpana-vibhāva (season, temple ritual timing, festival occasion), and the full vipralambha-śṛṅgāra vyabhicāribhāva palette (autsukya, śaṅkā, vrīḍā, glāni) is redeployed from describing an external divine couple to enacting the individual devotee's own interior relationship to the divine — precisely the lakṣaṇā-to-vyañjanā shift this document's companion volume on dhvani theory treats as Ānandavardhana's central technical distinction, here applied at the scale of an entire devotional corpus rather than a single verse.
↑ back to contentsKāñcarla Gōpaṇṇa, known as Bhadrācala Rāmadāsu (seventeenth century), composed his kīrtanas to Rāma predominantly in dāsya bhāva — the devotional stance of servant to master — rather than Annamācārya's madhura register, and his corpus supplies a structurally distinct test case for the theory precisely because of one biographical fact the tradition holds to be historically documented rather than legendary embellishment: a substantial portion of his kīrtanas were composed while imprisoned, following his diversion of temple-collection revenue toward constructing the Bhadrācalam Rāma temple, and address Rāma directly with the specific grievance of a devotee-servant whose master has, from the devotee's own perspective, delayed rescuing him. Kīrtanas such as "Ēmani Migula" work through karuṇa-coloured dāsya — reproach mixed with unwavering loyalty, structurally close to a servant's aggrieved but still-faithful address to a delayed master rather than to śṛṅgāra's longing register. What Rāmadāsu's corpus adds to the theoretical picture developed in this document is a genuine limit case for the ālambana-vibhāva distinction drawn in section three: where Annamācārya's padams present a generalized, iconographically catalogued ālambana available to sādhāraṇīkaraṇa in the fullest sense, Rāmadāsu's specific imprisonment gives several of his kīrtanas an anubhāva with a real, historically particular referent behind the conventional dāsya vocabulary — the poet's actual chains are, by persistent tradition, said to be the literal referent behind verses that read, on their surface, as conventional devotional complaint. This does not break sādhāraṇīkaraṇa for a listener without knowledge of the biography, who relishes the generalized dāsya-karuṇa rasa exactly as theory predicts; but it supplies, for a listener who does know the biography, a second, simultaneously available layer in which the same words carry a particular, non-generalized historical weight as well — a doubling the theory's own vyañjanā apparatus readily accommodates, since suggested meaning is never restricted to a single determinate content for every possible listener.
↑ back to contentsTyāgarāja's roughly seven-hundred-plus surviving kṛtis, together with his two operatic kīrtana works (Prahlāda Bhakti Vijayaṃ and the Nauka Caritramu), are the single largest working demonstration in the Carnatic tradition of the navarasa table's full range deployed for bhakti ends rather than confined to a single devotional register throughout. "Endaro Mahānubhāvulu" works predominantly in a bhakti-inflected adbhuta-śānta register, its ālambana-vibhāva a catalogue of the great devotees themselves rather than the deity directly, its anubhāva the composer's own bowing homage stated explicitly in the text. "Nagumōmu Gaṇaleni" works in vipralambha-śṛṅgāra structurally continuous with Annamācārya's madhura register — Rāma cast as the withholding nāyaka whose smile the poet-nāyikā longs for — while "Kaddanuvāriki" works in a register closer to raudra-inflected vīra, the poet directly and pointedly rebuking those who deny Rāma's supremacy, garva and amarṣa functioning as the dominant vyabhicāribhāvas rather than the gentler autsukya of the madhura kṛtis. The Prahlāda Bhakti Vijayaṃ, as a sustained kīrtana-nāṭaka rather than an isolated kṛti, restores full dṛśya-kāvya narrative scope — Hiraṇyakaśipu's raudra, Prahlāda's śānta-bhakti fused register, and the climactic Narasiṃha episode's bhayānaka-adbhuta fusion, all sustained across a continuous narrative in the manner of a khaṇḍakāvya rather than compressed into a single self-contained kṛti. Read across the corpus as a whole, Tyāgarāja's practice is best described as treating the navarasa table not as a menu from which bhakti kīrtana must select only the gentler, śānta-adjacent entries, but as the complete instrument bhakti composition is entitled to use in full, exactly as dṛśya-kāvya's dramatic tradition always held the navarasa table applied to worldly subject matter.
↑ back to contentsDīkṣitar's corpus stands in deliberate contrast to Tyāgarāja's on two structural axes at once. First, register: Dīkṣitar's dominant rasa is overwhelmingly śānta rather than the fuller navarasa range, his kṛtis characteristically slow-paced (vilamba-kāla dominant), meditative, and structured around sustained contemplation of a deity's iconographic and doctrinal attributes rather than around narrative incident or dramatic complaint — a register closely continuous with śānta's architecturally prior status in the navarasa table discussed in section seven. Second, and more directly connected to this document's companion material on Sanskrit grammar: Dīkṣitar's sāhitya is composed almost entirely in dense, technically precise Sanskrit, frequently deploying compound structures (long tatpuruṣa and bahuvrīhi samāsa of the kind treated at length in the companion volume's grammar sections) that pack a single kṛti line with an entire catalogue of a deity's names, attributes, and doctrinal correspondences — the Kamalāmbā Navāvaraṇa kṛtis, mapping the nine āvaraṇas (enclosures) of the Śrī Cakra directly onto nine successive kṛtis addressed to Kamalāmbā, are the clearest single example, each kṛti's sāhitya functioning simultaneously as devotional address and as a versified doctrinal exposition of Śrīvidyā cakra-symbolism precise enough to serve as a teaching text in its own right. This double function — kṛti as bhakti-vehicle and kṛti as doctrinally exact teaching text simultaneously — is the specific achievement Dīkṣitar's corpus adds to the theoretical picture: where Annamācārya's padams redeploy śṛṅgāra's apparatus toward bhakti, Dīkṣitar's kṛtis redeploy Sanskrit's own derivational precision, examined at length in this document's companion grammar volume as vaikharī's domain, directly into the kṛti form, making the kṛti carry doctrinal content with the same load-bearing exactness a Sanskrit śāstra treatise carries it, rather than only carrying generalized devotional feeling.
↑ back to contentsŚyāmā Śāstri's comparatively smaller surviving corpus, concentrated overwhelmingly on Kāmākṣī and other forms of the Devī, works in a bhāva the navarasa table's standard eight-plus-one list does not name directly but the broader bhakti tradition recognizes as a distinct and legitimate devotional register: vātsalya, the tender protective affection ordinarily directed by a parent toward a child, here directed by the devotee toward the Devī but with the customary hierarchy inverted — the devotee approaches the Mother not primarily as a servant approaches a master (Rāmadāsu's dāsya) nor as a lover approaches a beloved (Annamācārya's madhura), but as a child approaches a mother, dependent, trusting, and permitted a specific register of complaint and cajoling a servant could not use toward a master. His svarajatis, a form combining sustained melodic elaboration with devotional text more extensively than the standard kṛti structure, and kṛtis such as those addressed to Kāmākṣī in the rāga Ānandabhairavi, work this vātsalya register through vyabhicāribhāvas closer to a child's — dainya (helplessness), autsukya directed at the mother's attention rather than a lover's presence, and a characteristic plaintive quality the tradition associates specifically with his musical idiom. This bhāva's absence from Bharata's original eight-plus- śānta table is itself instructive rather than a gap needing correction: vātsalya, along with several other bhāvas the bhakti tradition develops (sākhya, friendship-devotion, is a further example), belongs to a devotional taxonomy — most fully systematized later in the Gauḍīya Vaiṣṇava analysis of bhakti-rasa — that extends Bharata's originally worldly-dramatic navarasa table into a parallel devotional register, applying the identical sādhāraṇīkaraṇa and vibhāva-anubhāva-vyabhicāribhāva mechanics this document has traced throughout to relational stances the original dramatic theory, built for depicting worldly human relationships, had no specific occasion to name.
↑ back to contentsŚaṅkarācārya's own poetic corpus supplies a final, structurally distinct case: a body of work in which kāvya's alaṅkāra apparatus — upamā (simile), rūpaka (metaphor proper, identification rather than comparison), and dhvani's suggestion mechanism together — is deployed not primarily to produce worldly or even devotional rasa as an end in itself, but as a vehicle for transmitting non-dual (advaita) doctrinal content that direct, literal statement (abhidhā) communicates only awkwardly or discursively. The Saundaryalaharī works this method most elaborately: its first forty-one verses (the Ānandalaharī portion) describe the Devī's form and the Śrīcakra's structure using the full descriptive apparatus of śṛṅgāra-adjacent kāvya — ornament, limb-by-limb description in the conventional style of nakha-śikha (toe-to-crown) description found throughout secular kāvya — redirected entirely toward Śrīvidyā tantric doctrine, so that a verse describing the Devī's waist or eyes is simultaneously, at the vyañjanā level, encoding a specific cakra-correspondence or a specific stage of upāsanā practice, exactly as Dīkṣitar's kṛtis later encode Śrīvidyā doctrine directly into devotional sāhitya. The Nirvāṇa Ṣaṭkam works the opposite method: rather than positive metaphor built up through descriptive accumulation, it proceeds entirely through negation (neti neti-structured rūpaka), a sustained sequence of "I am not this, not this" claims — not mind, not intellect, not body, not caste, not fear, not doubt — building toward the positive identification with pure consciousness (cidānanda-rūpaḥ śivo'ham) precisely through the accumulated weight of what has been systematically excluded, a rhetorical structure alaṅkāraśāstra recognizes as a form of vyatireka (distinction-by-negation) deployed here for doctrinal rather than descriptive ends. Bhaja Govindam supplies the deliberate contrast case within Śaṅkarācārya's own corpus: composed in comparatively direct, non-metaphoric admonitory language addressed bluntly to an old scholar absorbed in grammatical rule-mongering rather than in liberative practice, it largely dispenses with dhvani's suggestive indirection in favour of straightforward upadeśa (direct instruction), demonstrating within a single author's output that the tradition treats metaphor-laden suggestion and direct statement as two available registers to be chosen deliberately according to the specific communicative task at hand, rather than treating suggestive kāvya as invariably superior to plain statement in every doctrinal context.
↑ back to contentsBhoja of Dhārā (eleventh century), in the Śṛṅgāraprakāśa and the earlier Sarasvatīkaṇṭhābharaṇa, advances a position that the four-school history in Part II does not accommodate at all, because it rejects the shared premise every school in that history takes for granted: that rasa is a genus with nine (or eight) co-equal species. Bhoja's ekarasavāda holds that śṛṅgāra alone is rasa in the full, primary sense, and that the remaining eight are not co-equal species of one genus but derivative modifications of śṛṅgāra itself, arising when śṛṅgāra's underlying substrate is inflected by a specific obstruction or intensification. Bhoja's argument proceeds from a redefinition of the substrate all rasa is grounded in: not Bharata's eight-plus-one discrete sthāyibhāvas taken as independent psychological primitives, but a single underlying ahaṃkāra (self-regard, ego-sense, used here in a technical rather than pejorative sense, closer to "the self's fundamental orientation toward its own flourishing and expansion") that Bhoja holds is identical in kind across all nine conventional rasas and differs only in the specific object and obstruction-pattern it takes. Raudra, on this account, is śṛṅgāra's ahaṃkāra directed at an object that has injured what one loves; karuṇa is the same ahaṃkāra confronting the loss of what one loves; bībhatsa is the same ahaṃkāra recoiling from what threatens what one loves; even śānta, the rasa Abhinavagupta holds architecturally prior to the other eight, is on Bhoja's account not prior at all but a further inflection of the identical ahaṃkāra turned toward the self's own liberation as its object of attachment. Bhoja supports this reductive move with an extensive taxonomy internal to śṛṅgāra itself — he subdivides it into śṛṅgāra proper (union and separation between two qualified lovers) and further specialized categories governing improper or partial attachment — arguing that this internal richness is sufficient to generate, by systematic modification, everything the other eight rasas separately name.
The force of Bhoja's position for the wider theory this document has traced is not that it displaces Abhinavagupta's synthesis — subsequent alaṅkāraśāstra, including Jagannātha Paṇḍita discussed below, treats ekarasavāda as a minority position rather than the settled view — but that it exposes a genuine unresolved question the four-school history in Part II passes over by treating the navarasa list as a fixed empirical inventory rather than a claim requiring its own justification: why nine, and why these nine, rather than some other partition of the space of possible emotional dispositions into rasa-eligible categories. Bhoja's reductive answer, that there is really only one rasa-eligible substrate and the conventional nine are its surface variants, at minimum forces later theorists to state explicitly, rather than simply assume, what principle fixes the navarasa list's boundaries and prevents either Bhoja's reduction to one or an equally possible proliferation to twenty or thirty. This is also the specific theoretical resource a listener could invoke to explain why devotional composition returns so persistently, across Annamācārya's, Tyāgarāja's, and the wider bhakti-kīrtana corpus, to śṛṅgāra's apparatus even when the ostensible content is bhakti rather than romantic love examined in the ordinary sense: on Bhoja's account this is not merely a contingent literary preference among composers but a direct consequence of śṛṅgāra occupying a structurally privileged, generative position relative to every other rasa a composer might otherwise reach for.
↑ back to contentsKuntaka, a near-contemporary of Abhinavagupta writing from within the same broader Kashmiri milieu, proposes in the Vakroktijīvita a unifying principle for poetic excellence that neither reduces to dhvani's suggestion-based architecture nor to the older guṇa-rīti-alaṅkāra apparatus Part II·2 surveys, but competes directly with dhvani for the position of poetry's single defining criterion: vakrokti, "oblique" or "deviant" expression, meaning any manner of saying a thing that departs, through the poet's own kavi-vyāpāra (creative agency, or "poetic skill-in-action"), from the plain, unremarkable way the same content could otherwise have been stated. Kuntaka catalogues vakrokti across six distinct levels of linguistic organization at which this deviation can occur — varṇa-vakratā (obliquity at the level of individual phoneme choice and sound-patterning), pada-pūrvārdha-vakratā and pada-parārdha-vakratā (obliquity in a word's root-selection and its suffixation respectively), vākya-vakratā (obliquity at the level of the full sentence's construction), prakaraṇa-vakratā (obliquity in how an episode within a larger work is handled), and prabandha-vakratā (obliquity at the level of the entire composition's overall design) — arguing that genuine poetic excellence is identifiable, at any and every one of these six levels simultaneously, by the presence of a departure from the plainest available expression that is nonetheless not arbitrary distortion but a deliberate, skill-bearing choice a discerning reader recognizes as more striking, more apt, or more delightful than the unmarked alternative would have been.
Kuntaka's explicit target is dhvani theory's claim that suggestion (vyañjanā) is poetry's single defining operation, with everything else, including the guṇa-rīti-alaṅkāra apparatus, properly subordinate to it; Kuntaka's counter-claim is that vakrokti, not vyañjanā, is the genuinely unifying principle, since a directly stated (vācya) sentence can itself already be markedly, strikingly oblique in its diction or construction without carrying any additional suggested sense beyond what is stated, and Kuntaka holds such sentences are still recognizably good poetry on grounds dhvani theory alone does not capture. The historical resolution, reached by later theorists rather than by either school conceding outright, treats vakrokti and dhvani as operating at different levels rather than as strict competitors: vakrokti describes the expressive texture through which content, whether directly stated or suggested, is rendered strikingly rather than flatly, while dhvani describes what content, stated or suggested, is actually being communicated — meaning a poet's choice of vakrokti at Kuntaka's six levels is properly one further resource, alongside the guṇa-rīti-alaṅkāra apparatus already discussed, for delivering the dhvani-level suggested meaning with maximal aesthetic force, rather than a genuinely rival account of what poetic excellence ultimately consists in. This resolution is directly visible in Dīkṣitar's compound-dense Sanskrit examined in Part V: his characteristic samāsa constructions are simultaneously vakrokti at Kuntaka's pada- and vākya-levels (a strikingly compressed, non-obvious way of saying what a plainer sentence could have said) and the vehicle for dhvani-level suggested doctrinal content, the two operating jointly rather than in competition within a single kṛti line.
↑ back to contentsJagannātha Paṇḍita (seventeenth century, Mughal court), the last major systematic theorist in the classical alaṅkāraśāstra lineage, opens the Rasagaṅgādhara by replacing the definition of kāvya every prior theorist from Bhāmaha onward had worked with some variant of — kāvya as śabda and artha bound by guṇa, alaṅkāra, and an absence of doṣa (fault) — with a single, deliberately more economical criterion: ramaṇīyārthapratipādakaḥ śabdaḥ kāvyam, "kāvya is that language which conveys a delightful (ramaṇīya) sense." Jagannātha's stated reason for this replacement is that every earlier list-based definition (requiring guṇa, requiring alaṅkāra, requiring absence of doṣa as separately checked conditions) both admits counterexamples — verses lacking a formally catalogued alaṅkāra that are nonetheless universally regarded as excellent poetry — and fails to explain why the presence of the listed properties matters in the first place, whereas defining kāvya directly by the delight (ramaṇīyatā, closely allied to but not simply identical with Abhinavagupta's camatkāra) its sense produces in a qualified reader (sahṛdaya, "one whose heart is attuned") both handles the counterexamples and states the actual underlying criterion the older list-based properties were only ever imperfect proxies for. Jagannātha devotes extensive sections to defending this definition against objections from the entire prior tradition, arguing point by point that guṇa, alaṅkāra, and absence of doṣa are neither necessary nor sufficient conditions for ramaṇīyatā taken individually, but are typically, though not invariably, present together in cases where ramaṇīyatā is achieved, which is why the earlier tradition mistook a frequent correlate for the defining criterion itself.
Jagannātha's treatment of rasa proper largely accepts Abhinavagupta's camatkāra-based account inherited through the Dhvanyāloka-Locana lineage, but sharpens one specific point the earlier schools left comparatively loose: the precise relationship between śābdabodha (ordinary verbal cognition of a sentence's literal sense) and rasa-camatkāra's arising. Jagannātha argues that rasa cognition, while dependent on and arising immediately consequent to śābdabodha, is not itself a species of śābdabodha at all, since śābdabodha is propositional in structure (subject, predicate, relation) in a way rasa-camatkāra, on the school's shared commitment inherited from Abhinavagupta's account of unmediated suggestion, is not. This sharpening matters directly for how a sahṛdaya-reader is expected to engage a text like the Kamalāmbā Navāvaraṇa kṛtis examined in Part V: Jagannātha's framework predicts, correctly on the internal logic of the theory, that a listener can fully grasp Dīkṣitar's dense Sanskrit śābdabodha — parsing every samāsa correctly, following every doctrinal cross-reference to the Śrīcakra's āvaraṇas — while still not experiencing rasa-camatkāra at all, if the further, non-propositional condition of sahṛdayatā and aucitya- governed reception is absent; conversely, a listener with comparatively less technical command of the Sanskrit but a rightly prepared sahṛdaya disposition may experience camatkāra a more technically expert but less rasika listener does not, since on Jagannātha's account the two cognitions, propositional understanding and rasa-relishing, are related but genuinely distinct achievements.
↑ back to contentsNo sustained kāvya or nāṭya work, once it extends beyond muktaka's single self-contained verse, confines itself to a single rasa throughout its full length, and the theory accordingly requires a mechanism governing how multiple rasas coexist within one composition without producing an incoherent, directionless aesthetic experience. The governing concept is aṅgāṅgibhāva, the relation of subordinate part (aṅga) to principal whole (aṅgin): a sustained work is held to a single dominant rasa (aṅgi-rasa) that governs the composition's overall aesthetic direction and toward which every other rasa appearing within it is subordinated, functioning either to intensify the dominant rasa by contrast (raudra intensifying vīra in a battle narrative whose overall aṅgi-rasa is vīra) or to provide necessary narrative variation without displacing the aṅgi-rasa's ultimate governance (a comic, hāsya-toned subplot subordinated to an aṅgi-rasa of śṛṅgāra in a romantic mahākāvya). The specific rule this relation is subject to, stated explicitly by Viśvanātha in the Sāhityadarpaṇa and widely accepted subsequently, is that a subordinate rasa may appear with its full technical apparatus of vibhāva, anubhāva, and vyabhicāribhāva intact, but must not be allowed to reach the same degree of sustained elaboration and climactic emphasis the aṅgi-rasa itself receives, since a subordinate rasa elaborated to equal or greater length than the governing rasa would, on the theory's own terms, effectively become a second aṅgi-rasa competing with the first, producing the directionless, incoherent effect the aṅgāṅgibhāva rule exists specifically to prevent.
Tyāgarāja's Prahlāda Bhakti Vijayaṃ, examined in Part V as a sustained kīrtana-nāṭaka rather than an isolated kṛti, is a direct working instance of this rule rather than merely an incidental case of rasa variety: its aṅgi-rasa is bhakti-inflected śānta, sustained across the entire work through Prahlāda's own unwavering devotion regardless of the surrounding narrative's shifting content, while Hiraṇyakaśipu's raudra and the Narasiṃha episode's bhayānaka-adbhuta fusion, however vividly elaborated in their own scenes, remain subordinated aṅga-rasas whose specific dramaturgical function is to intensify, by direct contrast, a listener's relishing of Prahlāda's śānta-bhakti once the narrative returns to it — the raudra and bhayānaka episodes exist, on this reading, not as independent aesthetic destinations in their own right but as structurally necessary contrast-material subordinated to the work's single governing rasa, exactly as Viśvanātha's aṅgāṅgibhāva rule specifies must be the case in any well-formed sustained composition.
↑ back to contentsThe term sāttvika, applied in section five to the eight involuntary physiological responses (stambha, sveda, romāñca, svarabheda, vepathu, vaivarṇya, aśru, pralaya), is not a free-standing coinage but a direct technical borrowing from Sāṃkhya's guṇa-traya, the three constituent strands — sattva (lucidity, transparency, the capacity to reveal), rajas (activity, agitation, the capacity to move), and tamas (inertia, concealment, the capacity to obstruct) — that Sāṃkhya holds combine in every configuration of prakṛti, including the psycho-physical apparatus (antaḥkaraṇa) through which any experience whatsoever is registered. The specific reason Bharata's tradition names these eight responses sāttvika rather than, for instance, rājasa or tāmasa is that sattva's defining Sāṃkhya function is precisely revelatory transparency — sattva is the guṇa through which the antaḥkaraṇa becomes a clear, undistorting medium capable of registering and manifesting an internal state rather than obstructing or agitating its expression — and the eight sāttvika bhāvas are exactly the bodily responses the tradition holds cannot be produced by deliberate, effortful will (which would implicate rajas, the guṇa of directed activity) but arise only when the mind's sattva-component is sufficiently dominant and unobstructed to let an actual, fully absorbed internal state show through the body directly, without the mediating distortion ordinary deliberate action introduces. This is the specific philosophical grounding behind the diagnostic function assigned to sāttvika bhāva in section five: an actor whose sāttvika responses appear is, on this account, not merely performing well but is in a psycho-physical condition in which sattva has, at least momentarily, achieved sufficient dominance over rajas and tamas for the portrayed emotion to manifest with the same unobstructed transparency Sāṃkhya attributes to sattva generally, which is exactly why the tradition treats these eight responses, and not any deliberately produced anubhāva, as certifying genuine rather than merely technical absorption.
This grounding also explains a specific detail of the sāttvika list's own internal structure that would otherwise look arbitrary: the list includes responses associated with sthāyibhāvas across the full emotional register, from joy-adjacent romāñca to grief-adjacent aśru to terror-adjacent stambha and vaivarṇya, rather than being restricted to sattva-associated positive emotional states alone. This is consistent with Sāṃkhya's own position that sattva is a formal, content-neutral capacity for revelation and clarity, not itself a positive emotional valence — sattva reveals whatever content is present in the antaḥkaraṇa at a given moment, whether that content is rati, śoka, bhaya, or krodha, with equal transparency, which is precisely why sāttvika bhāva theory holds that all nine rasas, not only the more conventionally pleasant ones, are equally capable of producing the full eight-fold involuntary response set in a sufficiently absorbed actor, a claim directly consistent with the sāṃkhya-guṇa architecture the term sāttvika is borrowed from and would be considerably harder to motivate without that borrowing made explicit.
↑ back to contentsKṣemendra (eleventh century, Kashmir), in the Aucityavicāracarcā, proposes a criterion that section 7·2's treatment of rasābhāsa already invokes locally but that Kṣemendra elevates into a fully general theory covering every dimension of poetic composition, not the ālambana-sthāyibhāva relation alone: aucitya, propriety or fittingness, understood as the requirement that every compositional element — word choice, meter, alaṅkāra, characterization, and the vibhāva-anubhāva- vyabhicāribhāva conjunction itself — be fitted precisely to every other element and to the work's overall rasa, with any single element out of proportion or out of keeping with its surroundings sufficient to spoil (produce ābhāsa in) an otherwise correctly assembled composition regardless of how excellent each individual element is when considered in isolation. Kṣemendra's specific and influential thesis is that aucitya, not guṇa, alaṅkāra, or even rasa taken as an independently sufficient criterion, is the single property whose presence or absence actually determines whether a work succeeds or fails as poetry, on the grounds that a work can possess excellent guṇa, skillfully deployed alaṅkāra, and a technically correct rasa-conjunction at the level of each individual component, and still fail as a whole if the components are not properly fitted to one another — an ill-fitted excess of ojas-guṇa in a tender vipralambha-śṛṅgāra passage, for instance, or an alaṅkāra correctly executed by the rules governing that specific figure but jarringly out of register with the surrounding rasa, both count on Kṣemendra's account as aucitya-failures capable of spoiling the passage even though no individual rule of guṇa, alaṅkāra, or vibhāva-anubhāva construction has been technically violated.
Kṣemendra catalogues aucitya's application across more than two dozen distinct compositional dimensions — propriety of word to meaning, of meter to rasa, of ornament to context, of a character's speech to that character's social station and situation (a direct convergence with the register-assignment material this document's companion volume treats under Bharata's sarvavarṇika dramaturgy), of description to narrative pace, and, closest to the concerns of this Part, of vibhāva and vyabhicāribhāva selection to the specific sthāyibhāva being elaborated. This last application is the specific resource behind the propriety-based resolution of rasābhāsa developed in section 7·2: Rāvaṇa's śṛṅgāra fails not because śṛṅgāra's formal apparatus is technically misassembled, but, on Kṣemendra's more general criterion, because the ālambana selected is improper (anaucitya) relative to the ethical and dramaturgical context the epic as a whole establishes, exactly the kind of whole-composition fittingness-failure Kṣemendra's theory names as the single most general diagnostic available across all of kāvyaśāstra, applicable equally to a single misplaced epithet and to an entire misjudged emotional register.
↑ back to contentsA question left implicit throughout the four-school history of Part II, but treated with explicit technical care by Abhinavagupta and his successors, is how rasa-cognition should be classified within the standard pramāṇa taxonomy (pratyakṣa, perception; anumāna, inference; śabda, verbal testimony; and the further pramāṇas variously admitted by different darśanas) that the wider Indian philosophical tradition uses to classify every other form of valid cognition. Śaṅkuka's anumitivāda, examined in section nine, is precisely the position that rasa-cognition is a species of anumāna, and its rejection on the specific grounds that inference is propositional and affectively neutral in a way rasa-experience demonstrably is not, already implies rasa-cognition cannot simply be filed under an existing pramāṇa without remainder. Abhinavagupta's own position, developed most explicitly in his treatment of camatkāra's precise temporal structure, holds that rasa-cognition fails to meet the defining criteria of pratyakṣa as well: ordinary sensory perception is held, across virtually every Indian epistemological school, to register a determinate external object standing apart from the perceiving subject, whereas camatkāra's own defining phenomenal character, on Abhinavagupta's repeated insistence, is precisely the momentary dissolution of the ordinary subject-object distinction perception otherwise requires, the same structural feature that motivates his comparison of rasa-experience to brahmāsvāda discussed in section eleven. Rasa-cognition is therefore argued to require its own distinct classification, generally glossed in the later tradition as alaukika-pratyakṣa ("extraordinary perception," borrowing and repurposing a category Nyāya itself uses for other specialized cases such as yogic perception) or left as an explicitly sui generis category the standard pramāṇa list was never built to accommodate, since it was built to classify cognitions of an external world the perceiving subject remains distinct from throughout, not cognitions whose defining achievement is the temporary suspension of exactly that distinctness.
The specific technical dispute this classification question generates concerns yaugapadya, simultaneity: does rasa-camatkāra arise instantaneously, in a single undivided moment, once the vibhāva-anubhāva-vyabhicāribhāva conjunction is complete, or does it arise sequentially, building incrementally as each vyabhicāribhāva is registered in turn across the unfolding scene. Abhinavagupta's considered position holds that camatkāra itself, the culminating relishing, is strictly instantaneous and undivided when it occurs — consistent with its structural resemblance to non-dual recognition, which by definition admits no internal sequence — while the vibhāva-anubhāva-vyabhicāribhāva material that occasions it is necessarily sequential, since a dramatic scene unfolds in real time and no single instant contains the entire vyabhicāribhāva sequence a sthāyibhāva passes through on its way to ripened rasa. The relationship between the two is therefore not itself sequential in the ordinary causal sense of one discrete event producing a second discrete event some interval later, but is better described, on Abhinavagupta's own preferred formulation, as the sequential material serving as the occasion (nimitta) for an instantaneous recognition that, once it occurs, occurs all at once rather than building up in degrees proportional to how much of the preceding sequential material has been registered — which is the specific technical reason later theorists resist describing rasa-experience as growing gradually more intense across a performance in the way ordinary sensory stimulation intensifies with exposure, holding instead that what changes across a performance is the accumulating sequential material, while camatkāra itself, whenever and however many times it occurs within that performance, is each time a complete, undivided event rather than a partial installment of a larger cumulative one.
↑ back to contentsThe sapta svara did not begin as seven notes. Samagāna, the melodic rendering of the Sāmaveda, is built on a system that precedes the seven-note scale by a considerable historical distance and operates instead on a three-tone accentual base already familiar from ordinary Vedic recitation: udātta (raised pitch), anudātta (unraised, lower pitch), and svarita (a falling compound pitch, traditionally analyzed as udātta immediately followed by anudātta within a single syllable, and further subdivided by later phoneticians into jātya-svarita, kṣaipra-svarita, and other refined varieties). This three-tone system is not itself musical in the sense the developed rāga tradition later understands music; it is a pitch-accent system governing correct Vedic recitation, whose misplacement, as the indraśatru case discussed in this document's companion Śikṣā material shows, could invert a mantra's meaning outright. Samagāna's specific historical contribution is the discovery that this three-tone accentual base, when stretched, ornamented, and given sustained melodic extension for chanting rather than spoken recitation, naturally differentiates into a larger number of distinguishable pitch-levels — the Sāmavidhāna Brāhmaṇa and the later Nāradīya Śikṣā both describe a progressive expansion from the three Ṛgvedic accents toward a seven-level melodic scale used specifically in sāman chanting, traditionally named not sa-ri-ga-ma-pa-dha-ni but numerically, as the first (prathama), second (dvitīya), and so on down to the seventh svara, counted in a descending rather than ascending order from the udātta-associated highest pitch. This numerical, non-onomatopoetic naming convention is itself evidence that the seven-note system originates as a direct, incremental refinement of the three-tone accentual system rather than as an independently invented musical scale later retrofitted onto Vedic chant; the names come later, once the system is repurposed for gāndharva (secular and theatrical) music, and are treated in the following section.
↑ back to contentsThe Nāradīya Śikṣā, the specific Śikṣā text most concerned with sāman-chanting technique rather than Ṛgvedic recitation alone, supplies the tradition's foundational physiological account of where the seven svaras are produced in the body, an account the later gāndharva and nāṭya material inherits wholesale rather than re-deriving independently. The text locates the origin of breath-driven sound (prāṇa-borne nāda) at the nābhi (navel), from which breath rises through the hṛdaya (heart/chest region), the kaṇṭha (throat), the śiras (head), and is finally shaped into differentiated pitch at the point of articulation in the mouth (mukha) — a four-stage vertical model of sound-production the tradition treats as continuous with, rather than separate from, Śikṣā's general treatment of sthāna (place of articulation) for phoneme production discussed in this document's companion grammar volume. The specific technical claim this model supports is that svara-sthāna, the location assigned to each of the seven notes, is not an arbitrary convention fixed by an authority's decree but a mapping of each note onto a specific point along this vertical breath-path, with lower svaras (the numerically later, lower-pitched sāman tones) associated with sound generated closer to the nābhi and higher svaras associated with sound generated and resonated closer to the śiras — the same vertical logic later gāndharva theory inherits when it assigns each of the seven classical svaras a specific bodily point of resonance and a specific animal-cry association, treated in the section that follows.
This physiological grounding is what licenses the tradition's insistence, discussed in this document's companion volume on Śikṣā's three axes, that svara is not merely an abstract musical value comparable to a pitch on a modern staff but a psycho-physical event with a determinate location and generative mechanism in the reciter's own body — a claim with direct pedagogical consequence, since it implies correct svara production is trained by cultivating the breath's correct rise through the four named stations rather than by imitating a target pitch from the throat or mouth alone, without engaging the deeper nābhi-and-hṛdaya breath-support the Nāradīya Śikṣā holds is the actual origin-point of the sound being shaped.
↑ back to contentsThe seven classical svara names — षड्ज, ऋषभ, गान्धार, मध्यम, पञ्चम, धैवत, निषाद (ṣaḍja, ṛṣabha, gāndhāra, madhyama, pañcama, dhaivata, niṣāda) — replace Samagāna's numerical counting once the system is adopted into gāndharva theory, and each name carries an etymology the tradition holds to be motivated rather than arbitrary, in keeping with the derivational transparency this document's companion grammar volume treats as a general feature of Sanskrit technical vocabulary. Ṣaḍja ("born of six") is traditionally explained as the note produced by the simultaneous coordinated action of six sound-sources — variously enumerated across sources as the nose, throat, chest (uras), palate, tongue, and teeth — making it, on this account, the foundational note from which the others are subsequently derived rather than one note among seven co-equal peers, a status directly parallel to sattva's foundational role among the three guṇas discussed elsewhere in this document. Ṛṣabha ("bull") and the remaining five names are explained through the specific animal or natural sound each note's characteristic timbre was traditionally held to resemble: ṛṣabha to a bull's bellow, gāndhāra to a goat's bleat, madhyama to a heron's or krauñca-bird's cry (a detail worth noting alongside the krauñca-vadha origin of kāvya discussed elsewhere in this document, though the tradition does not treat the two krauñca references as directly connected), pañcama to a cuckoo's call, dhaivata to a horse's neigh, and niṣāda to an elephant's trumpet. This animal-cry etymological layer is presented by the tradition not as fanciful decoration but as a genuine mnemonic and pedagogical device: a student learning to produce a given svara's correct timbre by ear, prior to any formal instrumental reference pitch being available, is instructed to reproduce the specific quality of the named animal's cry as the practical target, making the etymology functionally load-bearing in oral pedagogy rather than merely decorative nomenclature layered on afterward.
↑ back to contentsThe Nāṭyaśāstra's gāndharva chapters (adhyāyas 28 through 33 in the standard recension) perform the specific historical work of taking the physiologically grounded seven-svara system inherited from Śikṣā and Samagāna and embedding it within a considerably finer-grained microtonal framework: the twenty-two śruti division of the octave discussed in this document's companion volume, with each of the seven svara assigned a specific consecutive-śruti count (traditionally 4-3-2-4-4-3-2 across ṣaḍja through niṣāda) rather than treated as seven equally-spaced points. Bharata's formalization is historically significant for a reason beyond the microtonal precision itself: it is the point at which the svara system, previously described mainly through the vertical nābhi-to-śiras production model and the animal-cry naming convention, receives its first systematic acoustic measurement procedure, described in the same chapters through the two-vīṇā comparative-tuning method — two identical stringed instruments tuned in unison, one then progressively retuned by successive quarter-tone reductions on a fixed string while the other retains the reference tuning, with the resulting audible difference between the two instruments used to empirically demarcate individual śruti-intervals rather than assuming them as an unverified theoretical postulate. This is the earliest systematic empirical procedure the tradition records for verifying microtonal claims about svara-sthāna rather than relying on the animal-cry mnemonic or the nābhi-śiras physiological model alone, and it marks the specific historical transition from svara treated primarily as a feature of correct Vedic recitation and chant toward svara treated as an autonomous object of systematic acoustic and musicological study in its own right, a transition presupposed by every later development this Part traces.
↑ back to contentsThe actual graphic notation used to record svara across this history undergoes its own distinct evolution, separate from but running parallel to the conceptual evolution traced in the preceding sections. Vedic manuscripts mark udātta, anudātta, and svarita through a small, closed set of diacritic marks placed above or below the akṣara being accented — a vertical stroke above the syllable for svarita in one common convention, a horizontal or subscript stroke for anudātta, udātta frequently left unmarked as the default — a notation adequate to the three-tone accentual system but not designed to record seven or twenty-two finer distinctions. Bharata's own gāndharva chapters, by contrast, do not use a comparable graphic notation at all; the twenty-two-śruti system is transmitted through the two-vīṇā verification procedure and through oral pedagogy rather than through any surviving graphic svara-script contemporary with the Nāṭyaśāstra itself, which is itself informative: the tradition's most microtonally precise historical claim (the twenty-two-śruti division) belongs to the period with the least developed graphic notation, meaning the precision was sustained through the two-vīṇā method and guru-śiṣya transmission specifically, not through any notational record a later reader could independently verify against.
Matanga's Bṛhaddeśī (circa seventh to ninth century, the exact dating contested) marks the next major stage, and is historically significant for introducing the term rāga itself into systematic musicological use, defined there as a specific, named combination of svara and varṇa (melodic movement-pattern) capable of delighting the minds of listeners — the historical bridge document, discussed in this document's companion volume, between the older jāti-mūrcchanā system and the fully developed rāga system Sāraṅgadeva later systematizes. Matanga's text also expands the notational vocabulary used to describe melodic movement itself, distinguishing sthāyī (a note sustained without movement), āroha and avaroha (ascending and descending movement), and sañcāra (characteristic wandering movement patterns specific to a given rāga) — a vocabulary of melodic motion-description that supplies the specific conceptual apparatus the following section's transfer into bodily movement-notation depends on, since a term such as sthāyī, describing a note held motionless at a fixed pitch-point, is structurally the direct musical analogue of sthāna, a bodily position held motionless at a fixed spatial point, in the karaṇa system discussed later in this Part.
↑ back to contentsSāraṅgadeva's Saṅgīta Ratnākara (thirteenth century) performs the final major pre-modern act of systematization this Part traces, consolidating the twenty-two-śruti framework, Matanga's rāga concept, and the jāti-mūrcchanā material discussed in this document's companion volume into a single comprehensive treatise that both historical strands, Hindustani and Carnatic, trace their subsequent theoretical development back to. Of direct relevance to notational evolution specifically, the Ratnākara's treatment of svara-sthāna variation — the recognition that a single svara-name (ṛṣabha, gāndhāra, dhaivata, niṣāda) can occupy more than one distinct śruti-position depending on the specific rāga or jāti being rendered — supplies the direct historical ancestor of the later Carnatic system's sixteen named svarasthāna positions (śuddha ṛṣabha, catuḥśruti ṛṣabha, ṣaṭśruti ṛṣabha; śuddha gāndhāra, sādhāraṇa gāndhāra, antara gāndhāra; śuddha madhyama, prati madhyama; pañcama fixed and invariant; śuddha dhaivata, catuḥśruti dhaivata, ṣaṭśruti dhaivata; śuddha niṣāda, kaiśikī niṣāda, kākalī niṣāda), each name specifying not only which of the seven svara-functions is being rendered but exactly which of several possible śruti-positions within that function is intended — a notational precision considerably finer than the seven-letter sa-ri-ga-ma-pa-dha-ni system alone can express, and one that could only be reached once Bharata's twenty- two-śruti division, Matanga's rāga-vocabulary, and Sāraṅgadeva's own systematic cataloguing of jāti-to-rāga transition had each, in turn, already been established.
↑ back to contentsVeṅkaṭamakhin's Caturdaṇḍiprakāśikā (seventeenth century) supplies Carnatic music's final and still- current systematization, reducing the entire accumulated history traced through this Part — three Vedic accents, seven sāman tones, seven named gāndharva svaras, twenty-two śruti-positions, sixteen named svarasthānas — into a single closed combinatorial system: seventy-two melakartā (parent scales), generated by systematically permuting the sixteen svarasthāna names across the four variable svara-positions (ṛṣabha/gāndhāra treated as a linked pair of three positions each, and dhaivata/niṣāda similarly linked, madhyama offered in two variants, pañcama and ṣaḍja held invariant), such that every possible seven-note scale admissible under the system's own combinatorial rules receives exactly one designated place in an exhaustive, non-overlapping catalogue, each melakartā further assigned a mnemonic name (through the kaṭapayādi numerical-syllable code, a separate notational device converting Sanskrit consonants into digits) that directly encodes the scale's own serial number within the seventy-two-item list. This is structurally the identical combinatorial-design method the Māheśvara Sūtras apply to the phoneme inventory and Piṅgala's Chandaḥśāstra applies to metrical patterns, discussed at length in this document's companion grammar and metrics volumes: a closed primitive inventory (here, the sixteen svarasthāna names), combined under an explicit permutation rule, generating an exhaustively enumerated and individually named output set — the same underlying design instinct, expressed a third time, now applied to the space of possible seven-note scales rather than to phonemes or metrical feet, and supplying the direct proof that the tradition's habitual combinatorial-cataloguing method, examined throughout this document's treatment of Chandas and karaṇa, is not confined to grammar and metrics but is a general methodological signature recurring across every domain the tradition subjects to systematic formalization.
↑ back to contentsThe specific technical term this Part's history has repeatedly used to name a svara's location — sthāna — is the identical technical term the Nāṭyaśāstra's āṅgika abhinaya chapters use to name a fixed bodily stance or standing position, and the parallel is not treated by this document as a loose or merely poetic resemblance between two unrelated technical vocabularies that happen to share a word. A svara-sthāna, in every stage of the evolution traced above, names a fixed, nameable point along a continuum (the nābhi-to-śiras vertical breath-path in the earliest physiological account, a specific śruti-position within the twenty-two-fold octave in Bharata's formalization) that a trained performer must be able to locate precisely and return to reliably on demand; a bodily sthāna, in the identical technical sense, names a fixed, nameable point within the body's own possible configuration-space (samapāda, vaiśākha, maṇḍala, and the remaining named standing positions the Nāṭyaśāstra catalogues) that a trained dancer must likewise be able to locate precisely and return to reliably on demand. Both usages depend on the same underlying conceptual move discussed in this document's companion grammar volume in connection with pratyāhāra and chunking: a continuous space of possible variation (continuous pitch, continuous bodily posture) is discretized into a closed, named inventory of stable reference points, each reference point functioning as a stable unit a combinatorial system can subsequently be built from — svara- sthāna feeding the rāga and melakartā systems traced above, bodily sthāna feeding the karaṇa system this Part's closing section addresses directly. The tradition's own willingness to use one word, sthāna, across both domains without troubling to coin a separate term for the bodily case is itself evidence that the architects of this material understood the two discretization projects, acoustic and kinetic, as one and the same underlying method applied twice, rather than as two independently invented systems that happen to share incidental vocabulary.
↑ back to contentsThe Nāṭyaśāstra's fourth adhyāya catalogues the specific combinatorial output this Part's entire history has been building toward: one hundred and eight named karaṇa, each a fixed, reproducible unit of dance movement generated by combining one named bodily sthāna (standing position), one named cārī (leg-and-foot movement pattern, itself subdivided into bhaumī, ground-based, and ākāśikī, aerial, categories), and one named nṛtta-hasta (a hand gesture used specifically for its rhythmic and decorative movement-quality rather than for semantic denotation, distinguishing it from the meaning-bearing hastas used in abhinaya proper) into a single, indivisible, named compound unit — structurally identical in design logic to a melakartā's combination of named svarasthāna positions into a single named seven-note scale, and to a Sanskrit compound's combination of named morphological elements into a single derivationally transparent word, discussed at length in this document's companion grammar volume. Bharata explicitly links successive karaṇas into larger named sequences (aṅgahāra), each aṅgahāra built from a specified ordered string of individual karaṇas exactly as a metrical vṛtta, discussed in this document's companion Chandas material, is built from a specified ordered string of laghu-guru syllable-weights — the same generative, rule-governed sequencing logic recurring a further time, now at the level of connected dance-phrases rather than individual movement-units or individual syllables.
The historical claim this Part has been assembling section by section can now be stated directly: the 108 karaṇas are not an independently invented dance-vocabulary that happens to appear in the same treatise as the gāndharva chapters on svara and śruti; they are the terminal application, to the domain of bodily movement, of a single combinatorial-formalization method whose entire developmental history — three Vedic accents differentiating into seven sāman tones, the Nāradīya Śikṣā's vertical nābhi-to-śiras production model, Bharata's own twenty-two-śruti empirical measurement procedure, Matanga's motion-vocabulary, Sāraṅgadeva's sixteen svarasthāna names, and Veṅkaṭamakhin's seventy-two melakartā reduction — this Part has traced from its earliest Vedic root to its most refined classical elaboration. The seers and ācāryas responsible for this material did not first perfect a science of sound and only subsequently, and separately, invent an unrelated science of movement; the karaṇa system's own explicit borrowing of the term sthāna, its combinatorial design identical in structure to the melakartā system, and its sequencing into aṅgahāra exactly as svara sequences into rāga-defining sañcāra, together indicate that a single sustained intellectual tradition, having already discovered that continuous acoustic variation could be discretized, named, verified, and recombined into an exhaustive and systematically generated catalogue, subsequently recognized that the identical discretize-name-verify-recombine method applied equally well to continuous bodily movement, and built the 108 karaṇas, and the temple sculptural programs at Cidambaram's Naṭarāja temple and the Bṛhadīśvara temple at Tañjāvūr that later preserve them in stone, as that method's direct and deliberate extension from the ear to the eye.
↑ back to contentsRasautpatti, kāvyotpatti, and nāṭyotpatti are, on the account this document has traced, three names for stages of a single continuous problem and its solution, not three independent theories that happen to share vocabulary. The problem, stated in Part I, is why a witnessed particular emotional event reliably produces a shared, pleasurable, self-limiting experience in a heterogeneous audience rather than an idiosyncratic private reaction varying by biography. Kāvyotpatti's krauñca-vadha account answers the question of origin — where the crystallization from raw emotion into shareable form first occurs, in a mind already saturated in metrical and grammatical structure. Nāṭyotpatti's fifth-Veda account answers the question of delivery — why that crystallized form must be engineered to reach every social stratum without exception, rather than only those already possessing the composer's own formal literacy. And the four-school history traced in Part II — Lollaṭa's transmitted emotion, Śaṅkuka's inferred resemblance, Nāyaka's generalized relishing, Abhinavagupta's suggested camatkāra — answers the mechanical question of how the crystallized, delivered form actually produces its shared effect once it reaches an audience, arriving at sādhāraṇīkaraṇa via vyañjanā as the specific, tested mechanism.
The kāvya-form lineage traced in Part IV — mahākāvya's full narrative apparatus progressively compressed through khaṇḍakāvya and muktaka, stotra's direct address grafted onto that compression, the padam's nāyaka-nāyikā redeployment of śṛṅgāra's technical machinery toward bhakti, and the kṛti's final integration of pāṭhya-gīta-abhinaya-rasa into one small, individually performable unit — is the specific historical process by which nāṭyotpatti's sarvavarṇika delivery-mandate and kāvyotpatti's crystallization-mechanism were finally married into a form requiring neither a royal court's patronage for a multi-canto epic nor a full dramatic troupe's staged production to reach a listener, but only a single voice and a few minutes. The six composers examined in Part V are not six unrelated case studies appended to a completed theory, but six demonstrations that the theory's full apparatus — the navarasa table, the ālambana/uddīpana distinction, the sāttvika/vyabhicāri machinery, sādhāraṇīkaraṇa, and dhvani's suggestive redirection of literal imagery toward doctrinal content — remained a live, working compositional toolkit many centuries after Bharata and Abhinavagupta, deployed with full technical command by composers who inherited it not as antiquarian theory but as the working grammar of how to make devotion itself, in Nāṭyaśāstra's own founding phrase, sarvavarṇika: reaching every listener, regardless of birth, training, or private biography, through what can be heard and felt.
↑ back to contents