5. FROM TREATISE ON DRAMA OF BHARATA

FROM TREATISE ON DRAMA OF BHARATA

(6.1-7) After the procedures for the stage preliminaries had been recounted to them, the great sages all addressed Bharata once again: "Answer five questions for us. Authorities on drama list various dramatic rasas. Can you tell us what it is that makes them rasas, 'tastes'? Why are the emotions (bhava) so called, and can you tell us what they bring into being (bhavayanti)? What exactly is meant by 'catalogue,' 'epitome,' and 'definition'?" When Bharata the sage heard the sages' words, he made the following reply regarding the analysis of rasas and emotions. "I shall explain it to you in full, holy men, as well as catalogue, epitome, and definition. It is altogether impossible to plumb the depths of drama. Why? Because the theoretical knowledges are numerous and the practical skills incalculable. The depths of even one ocean of knowledge cannot be plumbed accurately, let alone all the others. (6.8) That said, I will present a catalogue of the topics of drama, beginning with the rasas and emotions, which briefly offers the sense of the aphorisms and their commentary,12 yet allows for larger inferences.13 .... (6.15-23) The erotic, comic, tragic, violent, heroic, fearful, macabre, and fantastic are the eight dramatic rasas. These are the eight that were enunciated by the great Druhin.14 Now I shall tell you about the emotions: the stable, the transitory, and the emotions generated by one's 'sensitivity.' The stable emotions are desire, amusement, grief, anger, determination, fear, revulsion, and amazement. The transitory emotions are despair, fatigue, disquiet, resentment, intoxication, exhaustion, torpor, despondency, anxiety, confusion, remembrance, satisfaction, shame, recklessness, joy, agitation, numbness, pride, depression, longing, sleepiness, possession, dreaming, waking, vindictiveness, dissimulation, ferocity, sagacity, sickness, madness, dying, fright, perplexity. The eight sensitivities are paralysis, perspiration, horripilation, a broken voice, trembling, pallor, weeping, and fainting. The four registers of acting that pertain to drama are the physical, verbal, psychophysical, and costuming.15 .... (6.31) Such is the compendium of drama in brief aphorisms. From this point on, I provide an analysis based on the aphorisms and commentary. (p. 1.266) We will explain the rasas first, for no part of the enterprise16 will succeed without attention to rasa. Rasa arises from the conjunction of factors, reactions, and transitory emotions.17 [281] What would be an analogy? Just as taste arises from the conjunction of various condiments, spices, and substances, so rasa arises from the presentation of various factors and emotions. That is to say, just as physical tastes, that of lassi, for instance, or other such drink, are produced by substances such as brown sugar, plus condiments and spices, so [282] the stable emotion, in the presence of the various factors and emotions, turns into rasa. Here one might ask: What does 'rasa' actually mean? Our answer is that rasa is so called because it is something savored. And how can rasa be said to be 'savored'? Just as discerning people relish tastes when eating food prepared with various condiments18 and in doing so find pleasure, so discerning viewers relish the stable emotions when they are manifested19 by the acting out of various transitory emotions and reactions [283] and accompanied by the other acting

registers (the verbal, physical, and psychophysical), and they find pleasure in doing so. This explains why we call them 'dramatic rasas,' or tastes. On this matter there are two traditional verses:

(6.32-33) Just as connoisseurs eat and savor their fare when prepared with many condiments and substances, so the learned fully savor in their heart the stable emotions when conjoined with the factors, transitory emotions, and reactions.20 That is why they are called dramatic rasas, or "tastes."

[286] On this point the question has been raised whether [1] rasas are produced by the emotions and other aesthetic elements,21 or [2] emotions by rasas. And some have held [3] that rasas and emotions are mutually constitutive. The last two positions are false; we can readily observe that rasas are produced by the emotions and other elements and not the reverse. On these three positions there are several verses:22

(6.34-38) [1] "Emotions" (bhavas) are so called by the creators of drama because when conjoined23 with the various registers of acting they bring into being (bhavayanti) the rasas. Just as various substances bring a new flavor24 into existence, so the emotions with the aid of the registers of acting bring the rasas into being. [3] There is no rasa without the emotions and other aesthetic elements, and no emotions without rasa. Their production is mutually effected in the course of acting. Just as the conjunction of condiments and spices makes food savory,25 so the emotions and rasas bring each other into being. [2] Just as a tree grows from a seed, and flower and fruit from a tree, so rasas are a root, and the emotions and other elements are all determined from them.

[288.3] Accordingly, we shall now explain the rasas: their coming into being, their specific colors, superintending deities, and finer points. [289] Among the rasas, four are generative causes: the erotic, the violent, the heroic, and the macabre. To clarify: (6.39-45) From the erotic arises the comic, from the violent the tragic, from the heroic the fantastic, and from the macabre the fearful.26 The comic may be described as an imitation of the erotic, and the tragic rasa may be understood as an effect of the violent. The fantastic is declared to be an effect of the heroic, and the fearful should be understood to come about at the sight of the macabre.27 Their colors: the erotic is blue-black, the comic is white, the tragic is gray, and the violent red; the heroic is golden, the fearful black, the macabre blue, and the fantastic yellow. Their superintending deities: Vishnu is presiding deity for the erotic, Shiva's goblins for the comic, Rudra for the violent, Yama for the tragic, Mahakala for the macabre, Kala for the fearful, Indra for the heroic, and Brahma for the fantastic. [293.12] We shall now explain the definitions and give the finer points of the various rasas when conjoined with reactions, factors, and transitory emotions, and we shall show how a stable emotion becomes rasa. [294] To begin with, the erotic has its origin in the stable emotion of desire, and has the nature of radiant attire.28 [295] In real life anything that is pure, immaculate, radiant, or beautiful is compared to love; a person dressed in radiant clothing is thus said to be "lovely." Just as men's names derive from their paternal lineage, maternal family, or professional function, and come to be established through the usage of competent speakers of the language, so the rasas and emotions and other elements of the drama derive their names from their function and are established through competent usage. Thus, the erotic rasa is established by virtue of its function

and is so named because it has the nature of "lovely," radiant attire. Its cause is a woman and a man; its characters are a young couple of high status. [297] It has two states, the erotic enjoyed and the erotic thwarted. The erotic enjoyed, for its part, comes about from stimulant factors such as the enjoyment of the appropriate season, flower garlands, scented creams, jewelry, a boon companion or beloved object, a beautiful home; the experience of strolling in a garden, or hearing about or observing it; sporting in the water; mimicry of the beloved. [299] The erotic is to be acted out by such reactions29 as the skillful play of the eyes, movements of the eyebrows and sidelong glances, and gentle and pleasing bodily motions and verbal utterances. [300] All transitory emotions can be used except for torpor, ferocity, and revulsion. The erotic thwarted, however, is to be acted out by despair, fatigue, disquiet, resentment, exhaustion, anxiety, longing, sleeping, dreaming, waking, sickness, madness, possession, numbness, death, and other reactions.30 [302] Here one might ask: If the erotic has its origin in the stable emotion of desire, how is it subject to transitory emotions typically pertaining to the tragic? Here is our answer: We earlier stated that the erotic is actualized in two ways, as enjoyed and thwarted, [303] and the authors of treatises on erotics state that it has ten stages,31 which we shall describe in the chapter on acting in general.32 That said, the tragic rasa, arising as it does from such foundational factors as a curse or misfortune befalling a loved one, the loss of one's wealth, death, or imprisonment, is a state of hopelessness; the erotic thwarted, by contrast, is a state of hopefulness, arising as it does from longing and anxiety.33 [304] Thus, the tragic and the erotic thwarted are entirely different; at the same time, the erotic is indeed subject to all the transitory emotions.34 With respect to this: (6.46) When a man is joined with a woman, feeling mostly pleasant things, and availing himself of the appropriate season, flower garlands, and so on, it is designated the erotic rasa.35 [305] Here are two verses in arya meter that embody the meaning of the aphorism on the erotic:

(6.47-48) The erotic rasa arises from the appropriate season, flower garlands, jewelry, a boon companion, music, engagement with poetry, pastimes like strolling in a garden. It is to be acted out by clarity of eye and face, smiles, sweet language, satisfaction and joy,36 and gentle movements of the limbs.37

.... (6.77-82) The erotic rasa has three different forms, depending on whether it is in essence a matter of language, action, or costume, and the same holds for the comic and the violent.38 The tragic is likewise held to be threefold, depending on whether it is produced by a violation of morality, occasioned by the loss of one's wealth, or brought about by grief.39 Brahma has proclaimed that the heroic is similarly threefold: the heroic in munificence, in morality, and in war. The fearful likewise has three forms: it can be simulated, or stem from a criminal act, or be anything a timid person finds frightening.40 The macabre can be pure or impure: the former is disturbing, and is brought about by the sight of blood and the like; the latter is disgusting, and is brought about by the sight of excrement, maggots, and so on. The fantastic is twofold as well, being either heavenly, that is to say, produced by the sight of something divine; or blissful, that is, produced by joy.41

(6.83) These are what are known as the eight rasas, defined according to their appropriate definitions. I shall now present the definition of emotions. (p. 1.336) We shall now explain emotions. The author asks: Why are emotions (bhava) so called?

Is it because they come into being (bhavanti), or because they bring something else into being (bhavayanti)? The answer is that they are so called because they bring into being the meaning of the literary work42 when these are given verbal, physical, and affective form .... Here are verses to this effect:43

(7.1-3) An emotion is something brought about by the foundational and stimulant factors, and apprehended by means of the reactions in concert with the verbal, physical, and psychophysical registers of acting. "Emotion" (bhava) is also so called because it serves to "bring into being" (bhavayan)44 the poet's inner emotion (bhava), by means of the four registers of acting: verbal, physical, psychophysical, and makeup. "Emotions" (bhāvas) are so called by the creators of drama.45

[342.8] These emotions include the eight stable emotions, thirty-three transitory emotions, and eight psychophysical responses. These forty-nine emotions should be understood to be the causes of the manifestation46 of literary rasa. The rasas arise from them in combination with their common properties.47 Here is a pertinent verse: (7.7) The use48 of subject matter with which the heart concurs is the source of rasa, and it can engulf the body entirely, the way fire engulfs dry wood. [343] Here one might ask: If the rasas arise from the forty-nine emotions connected with the subject matter of a literary work, when these emotions are manifested by the factors and the physical reactions in combination with their common properties, why is it only the stable emotions that are said to become rasas? Our answer is as follows: Just as human beings, despite the fact that they have common physical traits, similar hands and feet and torsos and frames, and common major and minor limbs, will become, some of them, kings by reason of their family, character, wisdom, actions, skills, or cleverness, whereas others of lesser intelligence will become their servants, so the factors, physical reactions, and transitory emotions are subservient to the stable emotions. And the stable emotions, since the many other elements serve them, become the master; the other emotions, being subordinates like the men in the analogy, serve them in a subordinate manner; the transitory emotions are their retinue. Take as an example a king: although he has an entourage of many men, he alone acquires the title and no one else, however great. In the same way, the stable emotion alone, though surrounded by the factors, reactions, and transitory emotions, is awarded the name of rasa. Here is a traditional verse on the matter: (7.8) Just like a king among men or a teacher among students, the stable emotion is preeminent among all emotions. [344.3] We have in fact already given the definition of the emotions when identified as rasa. We shall now give their general definitions as emotions,49 starting with the stable emotions.50

[349.11] We shall now explain the transitory emotions. These are called vyabhicāris, from the preverbs vi and abhi and the verbal root car in the sense of "come": they "come (caranti) variously (vi[vidham]) into proximity (abhi[mukhyena])." When endowed in performance with the verbal, physical, and psychophysical registers of acting, they "lead" the rasas. In what sense [350] can they be said to lead? This is a just a conventional expression, the same way we say the sun "leads" the day or an asterism, without doing so by using arms or shoulders .... The transitory emotions lead the rasas just as the sun does the day or an asterism.51

.... (7.93cd) I shall now discuss the psychophysical responses in due order.

[373.3] Here one might ask whether the acting out of the other emotions is supposed to happen without psychic sensitivity, such that only these responses should be called psychophysical. My answer is as follows. Psychic sensitivity as defined here is something that arises from the mind; it is said to be the mind in a state of heightened awareness, since the psychic sensitivity arises when the mind is thus aware. The particular nature of each of its different emotions, such as horripilation, weeping, pallor, and the like, cannot be brought about when the actor's mind is elsewhere. Given that the drama imitates the nature of the world, the role of psychic sensitivity is essential. An example: the dramatic emotions that are effects of pleasure or pain should be so purified by psychic sensitivity that they are identical.52 In the case of pain, for example, which consists of weeping and the like, how can it be acted out by someone not feeling pain; or pleasure, which consists of joy and the like, by someone not feeling pleasure? It is precisely thanks to the actor's psychic sensitivity that he is able, even when not feeling pleasure or pain,53 to display horripilation or weeping, and it is for this reason that these emotions have been explained as being psychophysical." 54

1.2 RASA AS A FIGURE OF SPEECH IN NARRATIVE POETRY: THE EARLY VIEWS

Ornament of Poetry of Bhamaha (c. 650)

From the time of the Treatise on Drama-assuming the existence around the third century of a work on dramaturgy that bears some relationship to the text of The Six Thousand we now have-some four centuries elapsed until rasa was addressed in Sanskrit literary theory. And when this occurred, in Bhamaha's Ornament of Poetry, rasa took a form radically different from what Bharata's account would have led us to expect. The Ornament of Poetry is our first extant work on poetics in Sanskrit. Scholarly controversy long clouded Bhamaha's date and relationship with the poetician and poet Dandin (see the next section), with whose Looking Glass of Poetry the Ornament bears a complex relationship. But the question of priority at least has finally been sorted out. Bhamaha is to be dated no later than the middle of the seventh century, and must be viewed as the predecessor of Dandin (c. 700).55 Bhamaha does cite several forerunners in the field of poetics, but all their works have vanished, and we have no idea what, if anything, they may have thought about rasa. We also have no exegeses of his work (save for some fragments of a commentary by Udbhata, c. 800), and his compressed style can make his intention difficult to grasp. In fact, it is often Dandin's fuller exposition or reworking that offers compelling (if not always decisive) illumination. I describe the Ornament as a treatise on "poetics" in order to distinguish Bhamaha's work fundamentally from the domain of dramaturgy. Bhamaha is concerned exclusively with the nature of literary language in kāvya, poetry, which was (usually) clearly differentiated from nātya, drama.56 Whereas Indian thinkers did use the term kāvya also to refer to a higher-order category of language use-what we would call belles lettres or literature as such-their subdivision of kāvya into literature meant to be heard (or read) and literature meant to be seen marked out two very different domains of expressive language use. And rasa as first theorized for literature in performance was emotion the spectator could see. Precisely when and how the concept of rasa was extended from drama to narrative poetry, from the seen to the heard (and read), and with what consequences for aesthetic theory, is something we need to puzzle out from our textual sources. With the sole exception of Rudra Bhatta,57 no thinker ever announced he was extrapolating from the one domain to the other, or described the reasons for doing so, or addressed head on the difficulties in adapting a performative scheme to a narrative form. In Bhamaha's work, in fact, the transition appears to be just beginning.

Not only does the Ornament not address rasa as a separate topic, it does not yet even recognize rasa as a separate analytical category. Instead, given Bhamaha's overall concern with figures of speech and how they "ornament" language and render it literary rather than documentary, he conceives of rasa, or more precisely, verbal expressions of emotions, as tropes, that is, particularized types of expressive language use. Three that he defines would long be discussed in relation to rasa: the "rasa-laden statement"; the "affectionate utterance"; the "haughty declaration." A fourth figure, which he characterizes as the expression of happy coincidence, will later be redefined as an "emotion trope" too; so also, to a degree, the "noble," which for some authorities has a component of emotion, though for others, a purely material dimension (I supply translations of these in brackets).58 It is symptomatic of the inchoate stage of the understanding of emotion in literature-where the analytical focus remains entirely on literary form, not reader response-that Bhamaha should present these emotion tropes in a list of twenty-three other very heterogeneous figures of sense. Some may be statements where affect is in some sense present, if less obviously so (the "disingenuous remark," for example), but all the rest are familiar figures of speech such as intentional contradiction, false praise, or subtypes of simile. Besides his rather obscure listing, and in accordance with his spare style, Bhamaha only briefly defines the figures and exemplifies them with very abbreviated quotations from works (in most cases, lost works) of Sanskrit literature. These diverse materials are not easy to understand at first glance. The categories themselves will be redefined over the coming centuries, so much so that by the eleventh, Bhamaha's views were considered almost incoherent.59 Even the particulars can be hard to grasp, though here Dandin's fuller account goes some way in clarifying Bhamaha's intention. For the "rasa-laden statement," for example, Dandin's illustrative verse offers a perfect exemplification of Bhamaha's definition: it marks the moment in a lost work about the romantic king Udayana when his queen Vasavadatta's disguise as a mendicant is removed and she is revealed to be alive, returned as it were to life. At this moment the erotic rasa, as Bhamaha has it, is indeed "clearly shown."60 Notwithstanding Bhamaha's abbreviated characterization, the figure has nothing to do with disguise as such, but rather with a transition in the narrative where the rasa can actually achieve full development. This is also manifestly the case with Dandin's next example, the violent rasa, which again shows the moment of climax.61 A "rasa-laden statement" was thus thought of as prompting the reader to focus on that specific moment in the development of a rasa when it finally and actually achieves "full development," as Dandin's exposition shows. Similarly, Bhamaha's two other related figures, while not directly linked with rasa, do flag moments of emotional climax-of joy (in the affectionate utterance and arrogance in the haughty declaration-whatever else may be going on in the overall development of the rasa. While devoid of the expressive "indirection" that characterizes metaphor, metonymy, and the other figures of sense, these emotion tropes nonetheless remain charged uses of language and hence are capable of being seen as alankāra, literally "enhancements" of speech.

FROM ORNAMENT OF POETRY OF BHAMAHA

(3.1) Scholars declare the following also to be figures: affectionate utterances; rasa-laden statements; haughty declarations; expressions of disingenousness, of happy coincidence, or of nobility.62

(3.5) An example of an affectionate utterance is what Vidura says to Krishna when he comes to his house:

"The joy I experience today by your coming to my house, Govinda, I will have in the future only if you return."63

(3.6) A rasa-laden statement is one where a rasa such as the erotic is clearly shown, for example: "The queen appeared, no longer disguised as a religious mendicant."64

(3.7) An example of haughty declaration is when Karna rebuked the snake-arrow meant for Arjuna on its return: "Dear Shalya, does Karna shoot an arrow twice?"65 [(3.8-9)66 The disingenuous is where what is said is not exactly what is meant, as when Krishna addresses Shishupala in the "Theft of the Jewel": "We do not eat, at home or abroad, food not first eaten by learned Brahmans." This is said, in fact, in order to avoid Shishupala's poisoning him.67 (3.10) An example of happy coincidence is when in the "King's Friend," Narada appears to the Kshatriya women as they are going to appease Parashurama.68 (3.11) An example of nobility is when Rama, powerful though he was, in accordance with his father's word humbly abandoned the kingship and betook himself to the forest. (3. 12) Others interpret the figure differently, namely, when something is presented as richly appointed. (3.13) An example: Chanakya going at night to the pleasure house of Nanda recognized that it was covered with moonstones only by the drops of water that adhered.69]

LOOKING GLASS OF POETRY OF DANDIN (C. 700) WITH THE COMMENTARIES RATNA'S GLORY OF RATNASHRIJNANA (C. 950) AND GUARDING THE TRADITION OF VADIJANGHALA (C. 950)

In contrast to Bhamaha, we have a reasonably clear sense of the time and place of the critic and poet Dandin, for he states in one of his works that he was attached to the Pallava court in south India at the end of the seventh century, and that his great-great- grandfather had originally come there from farther to the northwest, in today's Karnataka (which might explain Dandin's predilection for the literary style of Vidarbha, the northern Deccan). In addition to his literary-theoretical treatise, he wrote a celebrated (though now only fragmentarily preserved) prose poem, The Story of the Beautiful Woman of Avanti (Avantisundarīkathā), a portion of which, the Adventures of the Ten Princes (Daśakumāracarita), circulated independently, and also a now-lost poem where "two tales are joined"-probably the two epics-in simultaneous narration through extended double entendre (dvisandhānakāvya). The Looking Glass of Poetry is one of the most influential works in the global history of poetics, probably second only to Aristotle's treatise in breadth of impact. It was translated into a number of South Asian languages and exerted influence on literatures as distant as Recent Style Chinese poetry of the late Tang dynasty and seventeenth-century Tibetan poetry. Like Bhamaha, though far more expansively, Dandin deals mainly with figures of speech in poetry. Unlike Bhamaha, Dandin attracted several commentaries that have been preserved; among the most important are two composed in the mid-tenth century, one by a Buddhist monk from Shri Lanka, Ratnashrijnana (also responsible for translating the Looking Glass into Sinhala), the other by an scholar at the court of Krishna III of the Rashtrakutas, whose actual name has been displaced by his sobriquet, "Gazelle Among Disputants."70 The first thing that strikes the reader of the Looking Glass is the profusion of literary features to which the term "rasa" is applied. Dandin uses it in three largely unrelated senses: first, as a general term for the "tasteful," or sweet, poetic style; second, as a term referring to any "sophisticated" turn of phrase, both usages being part of a broader literary theory of poetic language; third, as a technical term for the various affective dimensions of literary expression we have seen in Bhamaha. Dandin introduces the first two senses in connection with his discussion of what in Sanskrit are called gunas, or "language qualities," which generally refer to linguistic aspects of words (phonological, syntactical, or lexical, such as, respectively, soft or harsh phonemes, the degree of compounding, etymologically simple or complex lexemes) or their content. "Sound rasa," as the commentator Ratnashrijnana calls the first quality, comprises a sort of phonemic echoing across words (similar to, but not fully patterned like, alliteration or assonance); "content rasa," on the other hand, refers to sophistication in expression.71 Both these senses of rasa will be discarded in later tradition.72

Rasa as the affective content of literary discourse for Dandin is embodied in the same three figures of speech as for Bhamaha: "affectionate utterance," "rasa-laden statement," and "haughty declaration." While only the second is specifically connected with rasa, all three self-evidently comprise expressions of emotion, and this is clearly indicated by the definitions Dandin offers for them, which emphasize heightened feeling.73 (Again, I include a fourth in brackets, the "noble," which, as in Bhamaha and later in Udbhata, has, on one interpretation, a component of affect.) Dandin's treatment of the "rasa-laden statement" in particular demonstrates his understanding that rasa in the sense of aesthetic emotion refers principally (perhaps even exclusively) to the intensified emotion of the character. All his illustrations exemplify this: they are the character's direct discourse, and never external description. Rasa as "the stable emotion alone ... once intensified by the causes, reactions, and the rest" is precisely what Abhinavagupta some three centuries later will later identify, in discussing the views of Bhatta Lollata, as "the position of the most ancient" authorities, on which he cites Dandin's own words.74 It is also manifestly the view of Dandin's commentator Ratnashrijnana: an emotion's "achieving the highest stage" of intensification means that it is made manifest by means of verbal reactions, since emotions are internal states and otherwise unknowable to others (a clear and simple explanation of "enhancement" of the sort we do not find elsewhere).75 Accordingly, like other aspects of meaning, rasa and its related emotion tropes were thought of by Dandin as a purely internal, even formal, feature of the work, and as yet had nothing to do with reader response (even for Ratna, the only response that might count is that of the poets themselves, 1.51; 1.52). Like Bhamaha before him and other poeticians to follow, Dandin had no category other than figuration under which to theorize the phenomenon of rasa in poetry. And in fact, explaining how affectionate utterance and the rest could or could not be thought of as figures of speech (even if the Sanskrit term alankāra were taken, with Bhoja, to mean something like "factor of beauty" rather than trope),76 would exercise the best literary minds of the following two centuries. Dandin himself seems untroubled by any conceptual tensions. He may be sensitive to the awkward multivalence of the term "rasa," taking pains to clarify that the second figure, the rasa-laden, relates to the eight dramatic rasas, and thus can be connected only vaguely with the quality of phonemic or semantic "sweetness" (2.290), but he otherwise makes no note of any category difficulties. For Dandin as for Bhamaha, rasa was native to another (i.e., dramaturgical) discourse, where it was more fully explained; in poetry, by contrast, the only category available for its analysis was the figure of speech. It would take several more centuries for it to be conceived of as the dominant feature of the literary work, in fact, the singular feature that distinguished literature from all other forms of language use.

FROM LOOKING GLASS OF POETRY OF DANDIN (D), WITH RATNA'S GLORY OF RATNASHRIJNANA (R) AND GUARDING THE TRADITION OF VADIJANGHALA (V)

(R) The author now addresses the language quality called "sweetness."

(D) (1.51) A poem is defined as "sweet" when it has rasa. Rasa is found in both the language and the subject matter, and insightful people become intoxicated by it like bees by honey.

(R) By "insightful people" is meant poets who understand rasa .... It is they who become intoxicated by a poem filled with rasa.77 (V) By specifying that this is rasa in the words of a poem, the author is distinguishing the language quality from what will later be defined as the ornament called the "rasa-laden" statement (2.278). For the latter is concerned with the erotic and other rasas, whereas what is meant here is the sweetness of the very words themselves.

(D) (1.52) When one word is experienced as similar to another by reason of this or that sound, we have what is called "proximity of words," i.e., such as are comparable in form and the like. This

conveys rasa, and can be combined with alliteration.

(R) Words that have similar sounds and that are not placed far apart from each other produce rasa in poetry, because that is sweet. This is "word rasa," known as "proximity of similar words" or "sound similarity," and is much prized by southerners. Alliteration is word rasa too and is likewise prized by southerners .... Thus word rasa is shown to be of two sorts ... proximity and alliteration, which can be used together but need not be. A poem lacking both, however, will lack rasa, and poets will not savor it. r that sound" refers to such th (V) "This or that sound" refers to such things as phonemic quality, accent, or syllabic weight.

....

(R) The author now proceeds to explain "content rasa":

(D) (1.62-64) Although any ornament can endow a meaning with rasa, the most important factor is sophistication. "I want you, girl, why don't you want me?" This sort of locution is vulgar, and produces poetry devoid of rasa. "The god of love, that wretch, shows me no pity, but happily he bears you no grudge, my sloe-eyed girl": a locution like this, on the other hand, which is sophisticated, conveys rasa.

....

(R) The author now defines three ornaments: affectionate utterance, rasa-laden statement, and haughty declaration:

(D) (2.273) An affectionate utterance is an expression of heightened affection;78 a rasa-laden statement is one beautified by rasa; in a haughty declaration the speaker's ego is prominent. All three, when they reach their intensified state,79 become ornaments.

(R) An expression of "heightened," or intensified, affection ... is an "affectionate utterance." A statement that is "beautified" by reason of the erotic or other rasa to be communicated is called "rasa-laden." A declaration in which "ego," or haughtiness, is conspicuous is "haughty." In this group of three ornaments, "intensification" is said to be present to the highest degree-that is, the expression of emotion is maximal, that of rasa unalloyed, that of haughtiness powerful. This sense is already expressed by the comparative suffix (used in the first case) and the possessive suffixes (used in the latter two), which can denote "to a high degree."80 The added term "in their intensified state" is, hence, intentionally redundant, but it is used to make the point absolutely clear. The author offers an example of an affectionate utterance:

(D) (2.274-275) "The joy I experience today .... "81 This is how Vidura spoke, and properly so, since such contentment can be gotten from no other source. And Krishna, who can be won over by nothing more than devotion, was overjoyed.82 (2.276-277) "The moon, the sun, the wind, the earth, the sky, the sacrificer, fire, and water-who are we to bypass all your forms, Lord, and see you in your very person?"83 This expression of joy, felt by King Ratavarman84 when the deity appeared right before his eyes, is to be understood as the affectionate utterance ornament.85

(R) The author now illustrates and explains the rasa-laden ornament:86

(D) (2.278) "The woman I thought was dead and hoped to rejoin by taking my own life- here she stands, my Avanti! How could I, while still alive, have gotten this woman back?"87

(R) This would seem to exemplify joy, which has already been given. To clarify this the author now explains the distinction between the affectionate utterance and the rasa-laden statement.

(D) (2.279) What was shown earlier was joy, whereas in the poem just given we have sexual desire that has become the erotic rasa thanks to the full development of its form. This type of statement is called rasa-laden.

(R) In the affectionate utterance ornament previously described, the joy or contentment was shown to be directed toward a male.88 The subsequent illustration is an example of desire or passion for a female, a particular emotion that is the source of the erotic rasa and is produced by the requisite foundational and stimulant factors and made known by the requisite reactions-as the author says, "that has become the erotic rasa," i.e., that has been transmuted into the form of that particular rasa. How exactly? "Thanks to the full development of its form." Here "form" means its nature, its own particular type of mental formation, whereas "full development" refers to a state of being stimulated or highly strengthened.89 ... What the author means is this: desire that remains unintensified, existing through the inner principle90 and having the nature of a particular mental formation, will be unmanifest. In this state it is called an emotion, and that is why "emotion" is described as a purely mental transformation (and hence unmanifest). This same emotion when stimulated becomes manifest through some "reaction," understood as the verbal91 or other register of acting. At that point it turns into what is called rasa. This is why it has been said that it is the emotions, once made manifest, that receive the name rasa. Further detail may be had from the treatises devoted to the topic. The same reasoning can be understood to apply to the other rasas the author is about to discuss, the violent and so on, which have the form of particular transformations of the emotions anger and so on.

(D) (2.280) "The one who dragged Draupadi by the hair before my very eyes-here he is, the vile Duhshasana! And now that I have caught him, he will not live a moment longer."92

(R) The author points out the rasa here:

(D) (2.281) In this poem, the anger of Bhima as he watches his enemy achieves the highest stage and becomes the violent rasa. This ornament is called a rasa-laden statement.

(R) Here "achieving the highest stage," or the high point of intensification, means reaching the state of manifestation through the reactions.

(D) (2.282) "Without having conquered land and sea, offered many sacrifices, and given wealth to suppliants, how could I call myself a king?"

(R) He now explains which of the several rasas this is:

(D) (2.283) In this poem, the emotion of determination is intensified and thereby comes to exist as the heroic rasa, with the power of making the words a rasa-laden statement.

(R) It is "intensified," that is, it reaches the highest stage, insofar as it is made manifest.

(D) (2.284) "The queen, whose tender body was pained by a bed of flowers-how can she lay herself down on a blazing funeral pyre?"

(R) The author now clarifies which of the many rasas this is:

(D) (2.285) In this poem, it is the tragic rasa that, once it has attained prominence, comes to be considered an ornament. The same holds for the remaining rasas, the macabre, the comic, the fantastic, and the fearful.

(R) Here, in the manner already discussed, the "tragic"-that is to say, the particular type of emotion called grief that provides the basis for a particular type of narrative dealing with tragic matters93-when it reaches "prominence," i.e., comes to manifestation,94 transforms into the tragic rasa, and is held by knowledgeable people to be an "ornament" or adornment to the literary text.

(D) (2.286) "Drinking the blood of your enemies time and again from cupped hands, the ghouls garlanded with entrails go dancing with headless corpses."

(R) Here the emotion called disgust is generated by the requisite foundational factor and made manifest by a physical reaction of the corresponding form. As it takes on the nature of a rasa, the macabre, which has the form of something loathsome, is considered an ornament.

(D) (2. 287) "My friend, you'd better move your blouse to cover this fresh line of scratches on your breast- aren't you supposed to be angry with him still?"95

(R) The sight of the immodest marks is the foundational factor, which brings about the emotion called amusement, and this in its intensified form becomes the comic rasa.

(D) (2.288) "On the trees in the garden of the gods, the sprouts are silken cloths, the flowers necklaces and other jewelry, the branches mansions-how marvelous!"

(R) Here the emotion called amazement, rendered preeminently clear, becomes the rasa of the fantastic.

(D) (2.289) "This is the thunderbolt of Indra that bears fire on its blades, and just thinking of it makes the demons' wives miscarry."

(R) Here the emotion called fear, which mutates into the form of the fearful rasa, is said to function as an ornament. Since the rasa-laden statement would seem to have already been addressed under the heading of the language quality sweetness, it might be unclear why it is being discussed again. To this point the author now speaks:

(D) (2.290) Earlier it was pointed out that in the case of sweetness, rasa is the source of sophistication in discourse. Here, by contrast, we are discussing the nature of the rasa-laden ornament of language that is connected with the eight rasas.

(R) By "discourse" is meant the content conveyed by the poem ... and by "rasa" is meant what has been called the rasa of sense. This was indicated earlier, on the occasion of distinguishing between the sound and sense aspects that are both common to sweetness, when the author stated, "Although any ornament can endow a meaning with rasa, the most important factor is sophistication."96 He now exemplifies and explains haughty declaration:

(D) (2.291-292) "Do not fear I will take vengeance for your crimes: my sword would never fall upon an enemy in retreat." Here an enemy seized in battle has been freed by a prideful man. This and other similar kinds of expression are known as "haughty declaration."

(R) Another example:97

"I would sacrifice my life in the fire of battle should some powerful enemy appear before me. But were I ever to be so cowardly-I, Ravana !- as to give back Sita, I would cease to draw breath."98

[(D) (2.298) According to experts the "elevated" ornament is when supreme grandeur of either sentiment or material wealth is shown. (2.299) "Raghava was incapable to disobeying his father's command, a man who had no qualms about the heavy task of decapitating Ravana."99 (2.300) "Only with difficulty could Hanuman recognize the lord of Lanka since a hundred images of him were reflected on the jeweled palace walls."]

1.3 RASA AS A FIGURE OF SPEECH: THE LATE VIEW

Essential Compendium of the Ornament of Poetry of Udbhata (c. 800) with the Brief Elucidation of Pratiharenduraja (c. 900) and the Exegesis of Tilaka (c. 1100)

On the testimony of the River of Kings, the mid-twelfth-century history of Kashmir, Udbhata was the chief court scholar of King Jayapida of Kashmir (c. 800). Later evidence, preserved in a commentary on a text on logic, suggests that Udbhata (or an Udbhata) was a very independent-minded grammarian, even a contrarian, as well as a materialist (carvāka), and a renegade (dhūrta) materialist at that.100 The extant works-The Essential Compendium, fragments of his commentary on the poetics of Bhamaha, vague references to followers of his commentary (Audbhatas) on the Treatise on Drama, along with a few quotations from an unnamed work attributed to Udbhata by Pratiharenduraja in his commentary on the present work-are inconclusive on the question of

identity. Would a materialist have written a court epic on the marriage of the god Shiva and Parvati, as our Udbhata did, borrowing large selections of the work for use as examples in The Essential Compendium? Would a Shaiva sectarian have composed one of the great expressions in Sanskrit of the universal solidarity of humankind?101 Why not, after all? In the absence of counterevidence, the contrarian grammarian, renegade materialist, and radical poetician can be taken to be one and the same person. As for the commentators on his extant work, the older, Pratiharenduraja, tells us that he came from the Konkan on the west coast and was a student of Mukula Bhatta, an important language philosopher in Kashmir, which would date him to 900. Tilaka, who knew and followed Induraja closely, was the father of the poetician Ruyyaka (c. 1150), and thus lived around 1100.102 We have no other works from either scholar. Udbhata's treatment of aesthetic emotion marks the final-and by now contradiction-riddled-stage where the conceptual framework of the older rhetorical analysis that had contained it, so far as narrative poetry is concerned, was stressed to the breaking point. And Udbhata nearly broke it.103 He radically redefines the earlier notions of Bhamaha and Dandin, formulated when dramaturgical rasa theory had yet to be fully adapted to poetry, and masterfully assimilates them to what was to become-or instead, here for the first time has become-the classical typology, something not unexpected from the first known commentator on the Treatise on Drama. Thus, the ornament known as the "rasa-laden" statement, earlier viewed as heightened or climactic emotion, now explicitly104 becomes the "full realization" of rasa, with the complete panoply of aesthetic elements (the foundational and stimulant factors, and the rest). "Affectionate utterance" is more clearly defined as the "intimation" of an emotion, where a rasa is not fully developed.105 "Haughty declaration" is transformed into an utterance marked by social impropriety, and hence reconceived as the "semblance of rasa"-the very idea, and the technical term, appearing here for the first time (4.7). Finally, an ornament that Bhamaha and Dandin had not even thought of as pertaining to the realm of affect, "happy coincidence" (samāhita), is redefined by Udbhata as "quiescence," where an existing rasa is resolved and no other takes its place, a category complemented by three others (the emergence of a rasa, the conjuncture of two rasas, the oscillation between rasas) in the works of later writers.106 Yet despite this close approximation to the dramaturgical model, Udbhata continues, like Bhamaha and Dandin before him, to categorize all these forms of expression as figures of speech like the tropes based on semantic indirection, such as double entendre (ślișta), and contentual figures, such as the "elevated" (udātta), where "some richly appointed object" ornaments a poem, "or the deed of a great being when intended as a subordinate feature and not as the main narrative itself" (4.8).107 By the early tenth century Induraja was registering concern over the tensions here-and at the same time marking how far the conceptual terrain had shifted over the course of the preceding century-when he wrote, "Whether the rasas and the emotions, given that they are the source of the highest literary beauty, can really be 'ornaments' of literature or are rather its very 'life force' will not be a subject for consideration here lest it unduly lengthen the book."108 In one of his most notable, and most controversial, innovations, Udbhata includes among the features required to indicate a rasa's presence in a poem the use of the "proper term" for an emotion (and judging from some extant literature, contemporary poets and playwrights may have felt the same way).109 "A proper term," Induraja explains, "is the actual lexeme, 'desire,' for example. This gives us to understand the presence of the emotion because it refers to it." Udbhata's analysis would be critiqued within a generation or two by Shri Shankuka, and what for Udbhata was a component in the creation of rasa became for later scholars a literary flaw. It does not require much reflection to see that announcing "This is erotic" does not per se make a scene erotic (and can in fact be counterproductive). Then again, if an emotion's "proper term" is more broadly conceived, the issue becomes more complicated and avoidance more difficult. Saying "I love you" hardly negates the aesthetic effect of a love scene, whereas for some transitory emotions there is no way to indicate their presence except by naming them (as Induraja points out). The real complexities here will become apparent in Abhinavagupta's later rethinking of the question.110 In intellectual-historical terms, the commentary of Pratiharenduraja is especially valuable in giving us a sense of the wider understanding of rasa at the start of the tenth century. In fact, it provides the first systematic account we have since Bharata of the components of rasa theory. And little in his account deviates from Bharata, though uncertainties mark his exposition at times

(transitory emotions, for example, are rather redundantly said to intimate transitory emotions, and according to his definition of "reactions," they would seem to comprise what the tradition has usually called the registers of "acting," see KASS 4.2, below). His thought, as in that of his near contemporary at the other side of India, Ratnashrijnana, continues to be dominated by the idea that rasa is nothing other than an intensified state of the stable emotion, and that the emotion in question is located in the character: when the quiescence of rasa is discussed, for example, it is Shiva who is said to have reactions to his erotic rasa (on 4.7). Induraja, accordingly, like Udbhata and his predecessors, concentrates his analysis of rasa on its features that are immanent in the text rather than on the phenomenon of reader response, though he does cite one verse (on 4.3-4) that clearly derives from the new reception theory.111 The history of the discourse was reaching a stage where the line between seeing rasa as constructed in the text and seeing it as cognized by the reader was becoming ever more blurry. There are some additional innovations in Induraja's thought. We encounter here for the first time the idea that the stable emotions are not understood "in their unique individuality, but rather in a general form" (sāmānyarūpa), which, while not the theory of "commonization" (sādhāraņīkarana) that Bhatta Nayaka would introduce around the same time (and of which Tilaka was fully aware), has no antecedent in Udbhata's work itself, and Induraja's use of the term may reflect something of the new thinking about rasa taking place in his time. Rasa is now also intimately linked with a didactic component, since it is a "means of attaining the four ends of man" (4.3-4), something that would be more expansively described by both Abhinavagupta and Bhoja. One last point of interest pertains to Udbhata's redefinition of the affectionate trope, which only "intimates" emotion rather than displays it clearly, as in the rasa-laden figure. Induraja designates this "emotion poetry," a new term that will later become standard, if rare in actual usage. This refers to a poem or portion of a literary text where emotion is thought of as present but not developed-or, more properly, unable to be developed-to the point of becoming a rasa. For all later authorities, this arrested condition results from the fact that the object of the emotion is not of the sort required for the rasa to develop. Desire, for example, the emotion that leads to the full development of the erotic rasa, must by definition be sexual desire between a young man and young woman of high status. Yet one can feel desire for a child, a parent, or a god, and literature that addresses such feeling, where the rasa cannot be fully developed because the desire is not of the sort that can be normally consummated, will remain "emotion poetry" (so Tilaka). For Induraja, the rasa has not achieved "full enhancement" because of the treatment of the aesthetic elements, not because of the nature of the object of emotion (in this case, an animal adopted as Parvati's child). The dispute would be adjudicated in the coming decades.

FROM ESSENTIAL COMPENDIUM OF THE ORNAMENT OF POETRY OF UDBHATA (U) WITH THE BRIEF ELUCIDATION OF PRATIHARENDURAJA (P) AND THE EXEGESIS OF TILAKA (T)

(U) (4.1) Some scholars say that the following are also ornaments: the affectionate; the rasa-laden; the impulsive; the disingenuous statement; the quiescent; the elevated, which is twofold; and double entendre.112

(P) First of all, the affectionate:

(U) (4.2) When a poem is composed so as to intimate the emotions such as desire and so on by means of the reactions and the rest, scholars define it as the "affectionate" ornament.

(T) The desire here is the sort directed toward a deity, a guru, a king, and so on.113 When the desire directed toward the beloved is indicated, the author will call the ornament the "rasa-laden." (P) There are three types of emotion: stable emotions, transitory emotions, and psychophysical responses .... 114 All told these

number fifty, and there are four things that "intimate" them:115 the reactions, the factors, the transitory emotions, and their own proper terms. With respect to these four, the reactions are themselves of four sorts:116 physical, verbal, psychophysical, and costuming. Physical refers to hand gestures and other such forms of acting; verbal to the use of intonation and so on; psychophysical to paralysis and the rest; costuming to headgear, dress, and so on. This fourfold set of reactions functions as effects that give us to understand the presence of the emotions that are their causes. [48] Factors are of two sorts, foundational and stimulant. A foundational factor is the substratum117 that enables a stable emotion to arise: Sita, for example, is a foundational factor in the case of Rama. A stimulant factor is that by virtue of which the stable emotions become intensified: the appropriate season, for example; flower garlands; scented creams and the like in the case of the erotic rasa. These two sorts of factors function as causes of desire and the other stable emotions, and as such they give us to understand the presence of the emotions, which are their effects, just as the arrival of dense dark clouds gives us to understand the coming of rain. In the same way that a given effect, once clearly comprehended, is invariably related to its cause, so a given cause, once clearly comprehended, is invariably related to its effect, as is generally demonstrated in everyday experience. Accordingly, the factors, insofar as they are causes, give us to understand the existence of the stable emotions. (T) (p. 33) An emotion does not become the life force of literature merely insofar as it is inferred, but rather when it is understood as being general118 and thereby comes within the domain of the act of tasting. This applies to rasa as well. That is why experts have said that rasa, emotion, the semblances of rasas and emotions, and cessation are all "implied." What Udbhata has said so far implicitly concurs on this point. If we examine the real nature of things, rasas, emotions, and so on are not actually inferred at all, since factors, reactions, and transitory emotions are not primary causes, effects, and auxiliary causes. If they were, the Sutra on Rasa would have said that rasa arises from the conjunction of primary causes, effects, and auxiliary causes. To examine how this is the case, however, would expand the size of my book, so I will not dilate upon it here.119 (P) (p. 48) "Transitory emotion," for its part, is a particular state of a stable emotion, and comprises despair and the rest. And because it is a concomitant, the transitory emotion leads us to assume the presence of the stable emotion, the way one wheel of a cart leads us to assume the existence of the others. Finally, "proper term" means the actual lexeme, "desire," for example, which gives us to understand the presence of the emotion because it refers to it. Although of course we do not observe that such words as "desire" and so on are a sufficient condition for really understanding an emotion like desire-since that is something whose nature is a unique particular120 and thus accessible only through immediate experience121-we nonetheless do find that they in part, like the physical reactions, can become a condition for understanding the emotions. After all, just as the reactions and so on do not give us to understand the presence of the stable emotions in their unique individuality, but rather in a general form,122 so do their proper terms. Enough said. These four-the reactions, factors, transitory emotions, and proper terms123-are thus the causes for understanding the emotions. To quote Bhatta Udbhata: "The emotions have four different elements."124 Now, when a literary work is composed by means of things "intimating" the fifty emotions, desire and all the rest-that is, things that lead us to assume their presence not in their sheer individuality but in their general state-by means of the reactions and the other three elements, either aggregated or disaggregated125 as appropriate, we have the "affectionate" ornament. The word for "affectionate," preyah, a comparative of the adjective "beloved," priya, expresses the foundational factor of desire, whereby desire itself is indirectly connoted; and since all fifty emotions are invariably listed along with desire,126 those in addition to desire can hereby be understood to be implied.127 Thus, the affectionate is metonymy for emotion poetry in general. Given that the poem is the thing to be ornamented, the emotions here function as ornaments. An example of the affectionate:

(U) Feeling a yearning no different from affection for a son, Parvati drew him into her embrace and began to comfort him.128

(P) Drawing to the breast is the physical [49] register of acting; comforting or consoling the verbal register; the baby animal

(the antecedent of "him") is the foundational factor. The longing awakened by her affection is the transitory emotion, since her "yearning," or desire, insofar as it is said not to differ from affection for a son, must have some connection with a species of longing. "Yearning" itself is a proper term. Thus, by these four features, the physical reactions129 and so on, we are enabled to infer130 the presence of the emotion of desire in the form of parental affection. Other verses could similarly be adduced for the other emotions as well. The rasa-laden:

(U) (4.3-4) The rasa-laden is where the coming into being of a rasa, the erotic and the rest, is clearly shown, with its basis in five features: its proper term, stable emotions, transitory emotions, factors, and registers of acting. The erotic, comic, tragic, violent, heroic, fearful, macabre, fantastic, and peaceful are the nine rasas that have been handed down for drama.131

(P) The rasas, which are nine in number, consist in the full enhancement of the nine stable emotions, desire and the rest, in such a way that each becomes, as appropriate,132 the means of attaining the four ends of man, or the grounds for avoiding their opposite. Insofar as they are savored in this enhanced form133 they are referred to by the technical term "rasa," or taste, connected as it is with a specific kind of savoring. By contrast, since there is no such savoring of the transitory emotions, despair and the rest, we do not apply to them the technical term "rasa," connected as the word is with that specific condition of reference. If one intends merely to express savoring pure and simple, there is no contradiction in using the term in reference to the transitory emotions, any more than in reference to the physical tastes, sweet, sour, and the rest. To quote the remark Rudrata makes after listing the rasas, the erotic and so on: "They are called 'rasas' because they are tasted [rasana]."134 To quote again:135 "The special sort of 'savorable' consciousness by which the four ends of man and their opposites are understood as things to be attained and avoided,136 respectively, is held to be rasa."137 Here the word "sort" indicates that the use of the technical term "rasa" in the case of the erotic and so on is dependent on a special kind of savoring. (T) (p. 33) These rasas are such insofar as they enable us to achieve the four ends of man, and avoid their opposite. (P) These nine rasas, the erotic and so on, come to be understood through five different components. To quote Udbhata again, "The rasas have five different elements."138 One of these elements is the proper terms that are expressive of the rasas-for example, in the case of the erotic, the word "erotic" itself. (T) (p. 34) These proper terms have the capacity to express the rasas only generally, not in their individual manifestation. Otherwise the very verse cited above, "The erotic, comic, tragic," would provide us as complete a comprehension of rasa as drama itself does.

(P) Next, the nine stable emotions, which can be called the material cause of the rasas. [50] Then the transitory emotions, despair and so on, which are particular states of the rasas. Then the two types of factors, which are the instrumental cause of the rasas (in one case, a woman, for example; in the other, the right season, garlands, scented creams, and so on). Last, the four registers of acting, the physical and the rest, which are the effects of rasa. (T) The term "acting" in the verse signifies the reactions. (P) These five features, the proper term and the rest, whether aggregated or disaggregated, function as the basis of the rasas; the poem that, as a consequence,139 has the capacity to exhibit the rasas in "clear" form is the ornament "rasa-laden." Rasas, after all, are ornaments to the poem.140 (T) The factors must be directly expressed in the poem, whereas the transitory emotions can be implied. The stable emotion, however, can only be understood by implication, for rasa is nothing but the stable emotion so understood.141 If it could be directly expressed, the physical and other registers of acting meant to function as inferential signs for it would have no function-and a poem would as a result be rather silly.142

(P) Here is an example:

(U) As Shiva contemplated the full extent of Parvati's virtues his passion, laden with fantasies, became overpowering. His body grew slick with sweat and the hair so stiffened it looked like the filaments massed in the calyx of a kadamba bud, while his face was ornamented with a look now pregnant with longing, now immobilized by anxiety, now languid with delight.

(P) What is being presented here is the erotic thwarted, of the craving sort,143 with reference to the lord. Its proper term is "passion ... became overpowering," with the stable emotion being revealed by the word "passion" in the same verse, for the erotic rasa is in essence the enhancement of desire, and "desire" in this discourse is the passion of two young people for each other. Thus, since it expresses a particular type of desire, the word "passion" here may be taken to function as the proper term for the stable emotion. The transitory emotions are longing, anxiety, and joy (representing "delight"), and they are revealed by their proper terms. Perspiration and horripilation, two psychophysical responses, are expressed by their proper terms; they are at the same time transitory emotions, since, like despair and the rest, they are particular states of the stable emotions, and hence transitory. The foundational factor is referred to in the phrase "As Shiva contemplated," since the goddess becomes a factor insofar as contemplated144 as a being endowed with various virtues. As for the forms of acting, that which concerns glancing is referred to by the word "look." Thus, in these verses the erotic rasa (of the craving type) is manifested145 by the five elements, proper terms, and so on. Other examples could be cited for the other rasas. Now, whether the rasas and the emotions, given that they are the source of the highest literary beauty, are really "ornaments" of literature or its very life force will not be a subject for consideration here lest it unduly lengthen the book. Similarly, I will not analyze the nature of rasa and emotion because it would require a great deal of comment and in any case is not directly germane. (p. 51) Now the impulsive:146

(U) (4.5) When rasas or emotions are presented as contrary to propriety because they are caused by craving or anger or the like, we have what is called the impulsive.

(P) Sometimes rasas and emotions are presented in accordance with our moral sensibilities, sometimes in conflict with them. In the former case, the ornaments are referred to as the emotional and the rasa-laden. Where, however, a work presents rasas and emotions that contain a conflict with those sensibilities and with the acceptable behavior rooted in them, the poetry is designated as "impulsive." Rasas and emotions are contrary to propriety when they are presented as being caused by passion, hatred, or delusion. The ornament is therefore called "impulsive," because of the presence of impulsiveness or force as prompted by the agent's own misjudgments. An example:

(U) Shiva's craving increased to such a degree that he was ready to quit the path of the good and take Himalaya's daughter by force.

(P) Here the text describes the taking by force of an unmarried maiden on the part of the lord-though he immeasurably excels all worldly beings-something that is caused by his powerful passion and is contrary to our established moral principles. Hence the impetuosity. His "craving increased" is the "proper term" for the erotic, since the erotic rasa, given that it is an enhancement of desire, essentially consists in the increase of craving. Furthermore, the word "craving" comprised in the phrase is the proper term for the desire that is the stable emotion of the erotic. "Himalaya's daughter" is the foundational factor; "by force" expresses the transitory emotion agitation; "ready to quit the path of the good" the transitory emotion confusion; "ready to ... take" the bodily

reaction. Thus by means of the five elements, the proper term and the rest, the erotic of the impulsive variety is indicated. The impulsive can be similarly exemplified for the other rasas and emotions.147

.... (P) (p. 52) Next, the quiescent:

(U) (4.7) The quiescent is when the existence148 of a rasa, emotion, or the semblances of rasas and emotions, is presented as quiescent, with a complete absence of the reactions of any other rasa or emotion.

(P) So far it has been stated that there are two sorts of rasas and emotions, depending on whether or not they are in conflict with moral standards. When the words "rasa" and "emotion" are used all by themselves, the intention is to refer to the second type; the word "semblance" is added when there is a conflict. When the "existence"-that is, the connection with their human substratum-of a rasa or emotion, or the semblance of rasa or emotion, is presented as quiescent, we have the ornament known as quiescent (samahita), for we then have a resolution (samādhāna) or reconciliation (samādhi) or avoidance of rasas. One might ask whether the cessation of an antecedent rasa (since the existence of some rasa is being avoided in the poem) would not of necessity lead to the presentation of some subsequent rasa, so that one of these ornaments, the rasa-laden and so on, would wind up intruding. It is to address this concern that the author adds "with a complete absence of the reactions of any other rasa or emotion." ... [53] Where the force of the predisposition149 occasioned by an earlier rasa is such that, even if the rasa has ceased, another form of rasa fails to display itself, or if it should, it disappears by virtue of some intervening matter, we have the quiescent ornament. An example:

(U) His loving eyes, the movements of his glances, the play of his brows, his clear complexion, the horripilation150 in his body slick with sweat- all this the lord calmed, as well as his limbs aflame with the fever of love, and he approached the Mountain's daughter with a salutation.

(P) "Calmed" means restored to their natural state .... The lord's reactions to the erotic rasa-the longing in his eyes and so on- are made to disappear by the transitory emotion termed "dissimulation," that is, covering his face. The phrase "with a salutation" indicates this concealing of the face.151

(U) (4.8) The elevated is either some richly appointed object or the deed of a great being when intended as a subordinate feature and not as the main narrative itself.152

(P) "Riches" are treasure such as gold or other wealth. A richly appointed object adorns the poem's subject matter. (P) An example of the first.

And Shiva said, "On Himalaya pearls are produced by boar, bamboo, and elephants,153 and even Shabara women can use them at will to adorn themselves. One of its peaks, Gandhamadana, scrapes the sky with its summits made of precious stones:154 sapphire, beryl, and rubies. Its tablelands are made of solid gold, and vast Emerald Mountain forms one of its foothills. And when at the end of a cosmic age Earth falls into the underworld, it does not fall along with her-no, it is then that its full extent is finally revealed."

(P) (p. 54) Here, masses of precious objects have been presented in the narrative, and this is what makes it elevated. It ornaments the poem by suggesting these objects' supernatural effects. The presentation of a "richly appointed object" forms only one type of elevated figure; the other is an act of a noble-minded being seeking to achieve what is good and avoid what is bad. It would be imprudent to assert that when such a narrative is being presented we have a case of the rasa-laden figure simply because it gives us a sense of the erotic or other rasa, for the noble-minded act here is a subsidiary sentence meaning, insofar as it is subordinate to some other thing. For obviously here the behavior of some great person, since it is subordinate to some other meaning, cannot be the main purport of the sentence. We have the rasa-laden figure, by contrast, only where the rasas are understood as the main purport: in such a case they are being savored in the form of an enhanced stable emotion (itself the means for attaining one of the four ends of man, or avoiding their opposite), since that is the point upon which the sentence comes to rest. How then could we have here even a hint of that figure? (T) (p. 37) With the words "subordinate feature" the author shows that the elevated is not in essence the rasa-laden figure. For here the act of the great being has a secondary, not a primary, role.155 (P) Here follows an example of the second:

(U) "The great Himalaya-so stable it stood unshaken even when the stout shoulder of the Primal Boar struck it repeatedly- has a daughter, and this blessed lady is she."

(P) Here the principal sentence meaning is Himalaya's stability. The act of Lord Vishnu in the Boar incarnation when engaged in saving the universe, which does give us a sense of the heroic rasa, is a subsidiary sentence, and hence intended as subordinate.

1.4 RASA AS THE CHARACTER'S EMOTION, AND HOW WE KNOW IT

*Commentary on the Treatise on Drama, of Bhatta Lollata (c. 825)

Lamentably meager though the remaining fragments of his work may be, with Bhatta Lollata we can perceive the true commencement of the extraordinarily intense investigation into literary emotion that would make the next three centuries in India the most fertile in the history of aesthetics anywhere before European modernity. This commencement was no doubt the result of a rediscovery of, or at least reengagement with, Bharata's Treatise on Drama in Kashmir in the early ninth century, a work that raised, in a productive way, as many questions about rasa as it answered. Although Bhatta Lollata is cited before Udbhata in what would appear to be a chronological listing in a verse in a fourteenth- century musicology treatise ("The commentators on the treatise of Bharata were Lollata, Udbhata, Shankuka, Abhinavagupta, and Kirtidhara"),156 the selection below from Abhinavagupta's New Dramatic Art makes clear that Bhatta Lollata opposed some positions of "Udbhata's followers," so he is likely to be later than 800; and since he himself is the direct object of critique by Shri Shankuka, whom we can reasonably place around 850 (see next selection), putting Bhatta Lollata early in the first half of the ninth century would make sense of all our data. Although a dozen short citations on technical questions of dramaturgy are preserved from his commentary on the Treatise, what we know about Bhatta Lollata's view of rasa is largely restricted to the brief passages presented below. But two verses from what appears to be another work of his on literary criticism, entitled the Exegesis of Rasa (Rasavivarana), are cited by the late twelfth-century scholar Hemachandra when discussing complex figures of sound:

Such figures as this have no purpose other than displaying the poet's skill, and I will not bother with defining their subtypes. They are really an excrescence on the literary work since they do not serve the purpose of teaching any of the ends of man. Good poets, after all, compose poetry to turn tender minds toward those ends, and a poem where the rasa is completely

impeded by sound figures requiring special effort to make sense of is hardly a happy means to such a goal. And in actual fact, descriptions of rivers, mountains, oceans, and so on really just destroy the rasa; how much more so an altogether impenetrable poem. Thus Lolata [sic]: "Exertions in describing rivers, mountains, oceans, cities, horses, towns, and the like that have no purpose other than displaying a poet's skills are not approved by serious scholars for inclusion in literary works. All the variety of sound figures, 'twinned' forms (yamaka) and the rest ... completely impede the rasa, and if not simply a poet's egotistical ostentation, are a result of his blindly following convention."157

Even from this brief citation, Bhatta Lollata can be seen as a forerunner of Anandavardhana in holding that rasa must constitute the core of the literary experience, so much so that anything not contributing to rasa, let alone detracting from it, must be eliminated. A twelfth-century commentator preserves a historical memory of the importance of Bhatta Lollata's views by declaring that if "Lollata, mountainlike himself, could not plumb the depths of the ocean of rasa," then who else could?158 Bhatta Lollata's interpretation of Bharata's Sutra on Rasa marks the starting point ofwhat would evolve into the standard historical narrative of the development of rasa theory, one that would endure basically unchanged to the end of the seventeenth century. Bhatta Lollata argues first that there is one crucial thing left unstated in the Sutra on Rasa, namely, the place in the whole process of the stable emotion itself. It is when the aesthetic elements are "conjoined" with the stable emotion, he asserts, that rasa arises, because rasa is in fact nothing but the stable emotion in a state of being "strengthened" or "enhanced" by these elements. Moreover, the stable emotion in question is that of the character and the character alone; it is decidedly not that of the spectator, about whom Bhatta Lollata, to judge from our fragments, is silent and apparently indifferent. The same holds for the reactions: these are responses in the main character to his own stable emotion, not those produced by his rasa in another character, let alone the reactions of the viewer/reader, as later thinkers would maintain.159 For Bhatta Lollata, "reactions" are the sorts of physical responses discussed in the Treatise itself, as in the case of the erotic rasa: "The erotic is to be acted out by reactions such as the skillful play of the eyes, movements of the eyebrows, and sidelong glances."160 Hence, it is in the character that rasa exists "in the literal sense," and in the actor only figuratively-but in no sense in the spectator, an interpretation whose reality later commentators would confirm even as they sought to refute it.161 In fact, Bhatta Lollata's understanding of the ontology of rasa accords fully with the position of Bharata himself, for whom rasa is the intensified stable emotion and simply and naturally "arises" from the conjuncture of the aesthetic elements.162 This view would remain dominant for many subsequent writers, including Kuntaka at the end of the tenth century, and even Bhanudatta as late as the early fifteenth.

FROM *COMMENTARY ON THE TREATISE ON DRAMA, OF BHATTA LOLLATA

Restatements of Bhatta Lollata's doctrine

(#1a, Abhinavagupta)163 Bhatta Lollata and others, first of all, have interpreted the Sutra on Rasa of the Treatise on Drama ("Rasa arises from the conjunction of factors, reactions, and transitory emotions") to mean: rasa arises when there is a "conjunction of factors, etc.," i.e., a conjunction between these aesthetic elements and the stable emotion. With respect to the elements, the factors164 are the cause that generates a mental state, namely, the particular stable emotion itself. The reactions meant in this analysis are not those produced by rasa, since they could not then be counted as causes of rasa as they are here; 165 they are rather reactions to the stable emotion.166 As for the transitory emotions, although they are themselves mental states and therefore should not be able to coexist with a stable emotion,167 nonetheless what is meant here by "stable emotion" is simply the dominant

predisposition.168 It is just as in Bharata's analogy of the mixed drink: among the various condiments,169 spices, and substances a certain one acts as the dominant "perfuming" element, 170 and hence is like the stable emotion, whereas other ingredients appear intermittently, and hence are like transitory emotions. Therefore, it is the stable emotion alone, once strengthened by the aesthetic elements, that constitutes rasa; in the unstrengthened state it remains a stable emotion.171 Rasa in the primary sense of the term172 exists in the character, Rama for instance; it exists in the actor only by force of his complete identification with the part.173 (#2a, Abhinavagupta)174 Bhatta Lollata argued that, although rasas were potentially infinite in number, it was the opinion of experts that only those listed by Bharata were capable of portrayal on the stage.175 (#3a, Abhinavagupta)176 According to the followers of Udbhata ... were the actor really to undergo the experience of rasa and emotion he would be overcome by them during a death scene, say, and utterly fail to keep the rhythm of the dance .... Bhatta Lollata rejects this view of Udbhata's followers. For one thing, it is perfectly possible for the actor to experience rasas and emotions as well, by way of the stimulation of his own predispositions. For another, he would be able to keep the rhythm by virtue of his complete identification177 with the part. (#4a, Ruyyaka)178 In Bhatta Lollata's view, the "conjunction" referred to in the Sutra on Rasa is a threefold relationship: object and means of production; object and means of knowledge; object and means of enhancement. These correspond, respectively, to the factors (both foundational and stimulant), the reactions, and the transitory emotions. And on this view, accordingly, the word "arises" in the Sutra is explained in three different ways as well. Thus, rasa in the primary sense of the term is the stable emotion coming into being in the character and present in him in an enhanced form. In an unenhanced state it remains the stable emotion.179 In the actor rasa exists only in a figurative sense.

*Commentary on the Treatise on Drama, of Shri Shankuka (c. 850)

Like the work of Bhatta Lollata-who inaugurates an era of not only reengagement with the Treatise on Drama but also regrettable textual loss-all the writings of Shri Shankuka have disappeared save for a few fragments of commentary and poems. This marks what is probably the most grievous loss in the history of rasa discourse after Bhatta Nayaka. There is no reason to doubt that the literary theorist we are concerned with was the same as the poet mentioned by the twelfth-century chronicler Kalhana as the author of the court poem Triumph of the World (Bhuvanabhyudaya). According to the River of Kings, in 850/1 during the reign of King Ajitapida, a terrible battle took place between two royal factions, and "the poet Shankuka, a veritable moon to swell the ocean of learned minds, composed a poem about this battle called Triumph of the World."180 A dozen or so verses of Shankuka are preserved in an important fifteenth-century anthology, though these tell us nothing specific about the poet; one verse cited in another (dated 1363) is ascribed to "Shankuka, son of Mayura," though whether this is the Mayura who composed the Hundred Poems to the Sun (Sūryaśataka), we have no way of determining.181 A late twelfth- or early thirteenth-century dramaturgical work refers to Shri Shankuka as a minister who was also a dramatist and remembers at least one of his plays, a romantic comedy called The Earring of the Many-Colored Lotus (Citrotpalāvalambitaka); it also refers to a dramaturgical argument of his not known from elsewhere.182 That the text of his commentary on the Treatise on Drama was still available then

seems impossible to believe; neither Mammata (c. 1050) nor Hemachandra (d. 1172), to judge from their interpretations, knew anything more about it than what they had read in Abhinavagupta.183 At all events, with a likely date of about 850, we can be sure that Shri Shankuka preceded Anandavardhana (patronized by the subsequent king, Avantivarman, whose rule commenced in 855/6), of whom he appears from the surviving materials to know nothing. An older argument that Shri Shankuka may have been a Buddhist is worth renewed consideration.184 While each individual piece of evidence may not be all that strong, aggregated they carry some force. Shri Shankuka quotes a verse from the work of Dharmakirti, the great Buddhist philosopher of the seventh century, though other later scholars (such as Mahima Bhatta) who were not Buddhists quote him too. The honorific "Shri" might (though does not necessarily) suggest Buddhist affiliation, but he would have had to be a recent convert if he was indeed the son of the poet Mayura. The later critic Bhatta Tota is able to challenge one of Shri Shankuka's interpretations on the (implicit) grounds that it would force him to accept the category of the universal (sāmānya or jāti) and hence to fall victim to the fallacy of defending a position at variance with his core beliefs (apasiddhanta)-and it was only the Buddhists who rejected universals. He accords a central place in his theory to logic in general and inference in particular-one of the only two means of valid knowledge accepted by Buddhists-so much so that when he brings the buzzsaw of inferential reasoning down on Bhatta Lollata, very little is left. But then again, other, non-Buddhist thinkers were logicians in ninth- and tenth-century Kashmir, among them two of the greatest (Bhasarvajna and Bhatta Jayanta). Last, Abhinava appears to attribute to Shri Shankuka a new understanding of the tragic rasa, as general compassion rather than grief for the loss of a loved one, which fits with developments in Mahayana Buddhism.185 As in the case of Bhatta Lollata, Shri Shankuka's commentary on the Treatise on Drama has vanished except for the quotations preserved by Abhinavagupta, who, when he is not simply referencing his interpretations on technical questions of the theater or variant readings of the text,186 opposes him at every point. But Shri Shankuka's ideas are not easily dismissed. His important distinction between referential and expressive language, whereby he critiques Udbhata's new doctrine of the place of the "proper terms" in the creation of rasa, would be picked up by Anandavardhana in his own way and restated by Abhinava even as he dismisses it.187 His key argument in aesthetics, while apparently primarily an epistemological one (how rasa is apprehended), is actually ontological (what rasa is): because we cannot directly perceive emotion, we must infer it, and the content of such inferring, as in all inference, obviously cannot be the real thing itself. For precisely this reason Shri Shankuka understands rasa as an imitation (anukarana, anukrti, literally an "after-making") in the actor of the stable emotion in the main character. And all the aesthetic elements-the factors, reactions, and so on-can therefore be configured as components of an inferential process whereby this emotion in the character comes to be known. Yet the kind of knowledge involved in such aesthetic inference is unlike any other. In cognizing an entity we typically reach one of four possible conclusions: that we have cognized the real entity; that the entity we have cognized is proven to be false after we first thought it to be real; that the cognition is uncertain-it may or may not be the entity we think; or that we have only cognized something similar to what we thought we had cognized. None of these possibilities pertains to aesthetic inference. Instead, as Shri Shankuka says, introducing an analogy that would be repeated down the ages and that tells us as much about the aesthetic objectives of Indian art as about Shri Shankuka's theory, our experience of the character in a play is like our experience of a painted figure, a horse for example, in a painting: we do not say the painting is like a horse, or posit any other of the three cognitive relationships; we simply acknowledge, "This is a horse." Rasa for Shri Shankuka would seem then to be a theory of perfect mimesis, where the viewer is not equating an image with a thing but simply seeing the thing itself so long as the play lasts or the painting is visible (although imperfect mimesis-absence of perspective, say-could arguably produce a reality effect as well). As the literary critic Kuntaka was to put it a century and a half after Shri Shankuka: "This is in the first place to postulate a comparability: between a poem and a painting, a poem's techniques and a painting's, and a poet and painter, because in both cases the principal objective is to reference the actual nature of a thing."188 Of course literature may be said to create what it only imitates (or is thought to imitate), and imitation, as Indian art shows, is no selfsame thing. Yet this was not the grounds on which the theory would be attacked. It was rather the notion of imitation itself,

which was viewed as more a phenomenological than an aesthetic problem-whether and what precisely the actor is imitating, not whether the poet is imitating the world-and the critique of imitation on the part of Bhatta Tota in the following century would ensure for Shri Shankuka what Shri Shankuka himself had ensured for Bhatta Lollata: that his work would be consigned (almost) to that his work oblivion.

FROM *COMMENTARY ON THE TREATISE ON DRAMA, OF SHRI SHANKUKA

Restatements of Shri Shankuka's doctrine

(#1a, Abhinavagupta)189 Shri Shankuka rejects the view of Bhatta Lollata190 for seven reasons: [1] Logically, we cannot have any awareness of a stable emotion prior to its connection with the aesthetic elements, since those elements are the inferential signs required for such an awareness. [We cannot, after all, have an awareness of fire on a mountain without first perceiving smoke, KA.] [2] It would turn out that the stable emotions would have to have been mentioned prior to the rasa. [If in the definition of rasa as understood by Bhatta Lollata (where it is the stable emotion enhanced by the aesthetic elements that is rasa) the stable emotion were in fact the subject and rasa the predicate, the former should have been mentioned first by Bharata, KLV.]191 [That is to say, if Bharata had thought it was the stable emotion itself that turned into rasa as a result of a conjunction-having first been brought into being by the factors, indicated by the reactions, and enhanced by the transitory emotions-then it would have been appropriate to list and define the stable emotions first, which he has not done. The rasas are in fact listed and defined first, KA; and the stable emotions only later, KLV.] [3] If rasa were simply the stable emotion enhanced, no purpose would be served in providing as Bharata does another explanation when the stable emotion is supposedly augmented so as to become a rasa. [Why, that is, does the sage first describe the factors and reactions of the rasas, and once again describe precisely the same factors and reactions of the stable emotions? KA; if rasa were simply the stable emotion enhanced, there would be no point in setting out the factors and reactions again, KLV.]192 [For when defining the rasas one after the other, Bharata will say something like the following: "Next, the heroic rasa. It is embodied in a character of high status, and consists of determination. It arises from such factors as lack of confusion, intentness, leadership, discipline, strength, valor, power, fortitude, and might." And when later on he refers again to the stable emotion he says, "The stable emotion called determination is embodied in a character of high status. It arises from such factors as power, courage, and the absence of despondency."193 These statements are identical in meaning, but when rasa is defined the factors and so on are discussed at length, and only sparsely when the stable emotions are reiterated. Moreover, it would be useless to explain the cause of a thing's arising and to repeat it when the thing is enhanced, KA.] [4] [If it is claimed that the stable emotion is the unenhanced state and rasa the enhanced state, KA], there would turn out to be an infinite number of stable emotions given the infinite degrees of enhancement to which each of them is subject, from dull to duller to dullest, and so on, 194 [as well as innumerable rasas from intense, to more intense, to most intense, KA, KLV].

[Were "rasa" the name reserved for the single point when the stable emotion reaches the state of full strengthening, we would have several problems, KA.] [5] It would be impossible for there to be the six types of comic rasa that Bharata identified.195 [6] [Furthermore, Bharata states that desire has ten stages,16 with each later one relatively more intense than the previous. Here too, given the possibility that each has its own relatively greater degree of intensity, KA] it would turn out that we would have innumerable erotic rasas and stable emotions of desire in each of the ten stages. [7] [It would also turn out that rasas and emotions would always progressively intensify, KLP.] [While Bhatta Lollata has argued that the stable emotion of the initial stage197 gradually turns into rasa when it is strengthened, KA], what we find actually occurring is the opposite. Grief, for example, is powerful at first but gradually lessens rather than strengthens. This is also the case with anger, determination, and desire: [once these stable emotions arise, each from its own cause, KA] we find them diminishing over time with the loss of indignation, resolve, and satisfaction, respectively.198 Therefore, [the Sutra on Rasa has to be interpreted differently, and as follows. The "conjunction" referred to there is an inferential relationship among the aesthetic elements, KA]. On the strength of inferential signs-the causes known as the factors; the effects consisting of the reactions; the auxiliary causes,199 namely, the transitory emotions-which, though factitious, since they are acquired by human effort,200 are not recognized as such, we apprehend as existing in the actor a stable emotion that is an imitation of the stable emotion in the main character, Rama, say, and precisely because it is an imitation it is designated by the special term "rasa."201 A distinct comprehension202 of the foundational and stimulant factors is gained203 on the strength of the literary narrative itself; the reactions are something the actor is trained to produce; the transitory emotions [267] are gotten by inference from their own associated factitious reactions. But we have no way of apprehending the stable emotions, not even from the literary narrative. The proper terms for them, "desire," "grief," and so on, simply render these things referents, insofar as they denote them; they do not make us understand them as if they were "verbal acting," or expression.204 For referential language as such is not at the same time expressive; it is the medium whereby expression is achieved, just as the limbs of the body are not in and of themselves expressive but the medium whereby expression in acting is achieved. Accordingly, in verses such as the following,205

Although my grief is distended, profound, endless, and vast, it is siphoned off by my anger, like the ocean's water by the submarine fire.206

or

He was so paralyzed by grief that his counselors raised a hue and cry, and fearing that his heart might burst, they begged him to try to weep.207

grief is not being "acted out" or "expressed," but simply denoted. In the following poem, however,

As she drew my portrait a stream of teardrops came falling that made my body appear to perspire at the touch of her hand.208

the sentence not only denotes its proper sense, it "acts out," or expresses, King Udayana's stable

emotion (here, pleasurable desire) and does not just speak of it. For the "verbal register of acting," or verbal expression, is the power enabling us to understand something, and is above and beyond the mere saying of the thing. For precisely this reason [the fact that a stable emotion cannot be understood directly from the literary narrative, KLV] the sage does not mention "stable emotion" in the Sutra on Rasa itself, not even in a separate case form.209 [It is only when conjoined with the aesthetic elements, and by no means prior to that point, that the stable emotion can be apprehended, and only as an imitation, KLV.] Accordingly, rasa must be an imitation of a stable emotion,210 and hence it makes sense for the sage to say later that it "arises from" the stable emotion, or "consists of" it.211 We certainly have evidence212 that real effects (such as rasa) can arise from knowledge of something unreal (like the content of an inference based on an imitation). To quote: "Two men run toward two sparkling lights, one a gem and one a lamp, both thinking the lights a gem. Although there is no difference regarding their false knowledge, there is a difference regarding their real effects."213 In this inferential process none of the following notions arises in us: that [1] the actor is actually the happy Rama;214 or that [2] the actor is not in fact Rama and not really happy when we had first thought him to be the happy Rama;215 or that [3] he may or may not be Rama; or that [4] he is similar to Rama. Instead, the aesthetic apprehension we have is different from all four, [1] a true apprehension,216 [2] a false one, [3] a doubt, and [4] a similitude. It is, instead, an apprehension that can be analogized to that of a painted picture of a horse, and has the form: "Here is the happy Rama."217 To quote:218 "There is no appearance of doubt, or indeed of truth or falsehood-we have the thought, 'This is him,' and not 'This is actually him.' We encounter no antithetical ideas,219 and so nothing makes us aware of the conflation.220 It is an experience we actually undergo, and what logical argument can confute such empirical evidence?" (#2a, Abhinavagupta)221 Shri Shankuka held that, in the course of a dramatic performance, while relishing the rasa in the actor, a viewer apprehends the stable emotion in the character, whereas in the actual world depicted in the drama222 a dramatis persona223 causes the rasa to come into being.224 The second position, that emotion and the other aesthetic elements precede rasa, accords with the training intended by the teachers of drama. This is why there is in fact a third option.225 [(#2b, Hemachandra)226 Therefore, emotion does not precede rasa, but just the reverse. When the sage states, "Whether rasa precedes emotion or emotion rasa is a function of the nature of the case: in the course of a dramatic performance, while relishing the rasa in the actor, viewers apprehend the stable emotion in the character," he is affirming the first option. In the actual world depicted in the drama, however, it is as a result of the character's first seeing "emotion"227 that its essential form, namely rasa, arises.] (#3a, Mammata)228 Shri Shankuka's position is as follows. The stable emotion is inferred to exist in the actor,229 whom we grasp by a mode of comprehension different from all four normal forms of apprehension .... This inference arises from a "conjunction"-that is, an inferential relationship-of three elements: (1) a cause, which is designated by the term "foundational and stimulant factor"; (2) an effect, which is designated by the term "reaction"; (3) an auxiliary cause, which is designated by the term "transitory emotion." The first is distinctly comprehended from the literary narration

itself, such as in the following verse, where we have a foundational factor for the erotic rasa enjoyed:

Here she comes into view, a stream of ambrosia upon my limbs, a salve applied to my eyes, my heart's desire incarnate, my life breath's mistress.

Or in the next verse, where we have a stimulant factor in the erotic rasa thwarted:

By a cruel fate I have been parted from her, that woman with large darting eyes- and at the same time the season has come of dense, racing clouds.230

The other two elements are manifested by the actor himself by revealing the effects of each, something he is able to accomplish thanks to training and practice.231 Although these aesthetic elements are factitious, they are not recognized as such; and although the stable emotion is grasped by inference, it is different from any other thing we infer insofar as it is something tasted, thanks to the beauty of the aesthetic event [-it is like our mouth puckering on seeing someone relishing a lemon, KA]. It enters our imagination232 as the stable emotion233 of the character, and although it is completely nonexistent in the actor himself, it is something that can be relished by the audience members by way of their own predispositions. The inferred stable emotion thus relished is rasa.234

On the Tragic Rasa

(#4a, Abhinavagupta)235 The term karuņā (with -a) refers in everyday life to the sense of compassion.236 It receives the technical designation of karuna (with -a), or the tragic rasa, when the spectators apprehend the presence of grief in the actor by means of inferential signs.237

On the Psychophysical Element

(#5a, Abhinavagupta)238 The following verses of Bharata,

Special effort must be taken in the case of psychophysical acting, since drama is founded upon it .... The psychophysical, which is connected with a particular emotion, is unmanifest, but it can be made known by its various properties- horripilation, tears, etc .- and so endowed with the rasa appropriate to the locus.239

have been interpreted by Shri Shankuka and others as follows: Why does the psychophysical require such serious "effort"? Here is the answer: The psychophysical in a particular character, Rama for example, is "connected with an emotion," that is, engendered by an intense state of concentration, and is productive of such psychophysical responses as horripilation. This inner psychophysical element in the drama is "unmanifest," or invisible, and can be "known" only by

way of those things that are properties of it, such as horripilation, since they prompt us to infer it as their cause (there would be no causality240 only if those properties were ever to come into existence in the absence of pleasure, pain, and so on). Moreover, it is then "endowed" with the rasa that happens to be the principal one in relation to the emotion under discussion, and can be known, as pleasure or pain, only with substantial effort by means of the rasa in the character. The horripilation and so on are the effects of this element, and in the absence of this or that outcome, how could a performance pertaining to the psychophysical element from which those effects derive ever succeed without a substantial effort? Such is the gist of the matter.241

1.5 THERE ARE NO RULES FOR THE NUMBER OF RASAS

Ornament of Poetry of Rudrata (c. 850) with Notes of Namisadhu (1068)

With respect to the development of the discourse on poetics, Rudrata must certainly be placed later than Udbhata (c. 800); at the same time, he knows nothing of Anandavardhana (c. 875) and is cited, as we have seen, by Pratiharenduraja around 900. A date of around 850 thus seems reasonable. For some scholars his name (and that of his father, Vamuka) has suggested an origin in Kashmir. Namisadhu, his one surviving commentator (we know earlier ones existed) was a Jain monk who composed his commentary in 1068.242 Although the Ornament of Poetry deals mainly with figures of speech, rasa occupies a more prominent position in it than in any previous work on poetics. In addition to defining rasa, Rudrata illustrates all the varieties (with poems of his own composition) and their subdivisions, with a focus on the erotic, along with details on the characteristics of the leading man and woman that would eventually form a genre of literary criticism in its own right (the "types of hero/heroine" treatise). If there is little discussion about what rasa is or how it works, we do find some noteworthy innovations, as well as continuities. Rudrata is the first scholar of poetics for whom rasa is no longer counted among the figures of sense. Instead, the topic is treated in a separate section of the book, which suggests to the admittedly post-Anandavardhana commentator Namisadhu that rasa has a more essential role to play in the literary work: unlike ornaments that enhance the exterior, the body of the poem, rasa is an "innate" quality. And for the first time (unless the verses ascribed to Bhatta Lollata, above, by Hemachandra are authentic), a clear linkage is found to exist between literature and the ends of man-a theory of didactic aesthetics, which would be restated by Pratiharenduraja in the generation after Rudrata and fully elaborated by Abhinavagupta and Bhoja. As for the rasas themselves, Rudrata's list includes the peaceful as well as the "affectionate," replacing the general "affectionate utterance" trope found in Bhamaha and Dandin with the specific feeling of parental, especially maternal, love (already hinted at by Udbhata). Perhaps Rudrata's most important contribution is the independence he shows in challenging Bharata's traditional list: since any emotion can be "tasted," the number of rasas is, he says, in principle limitless. This is precisely the argument Bhoja will make with far greater force a century and a half later. Last, if Namisadhu is correct in representing rasa in this account as an "enhancement" of the stable emotion, there is little to suggest that Rudrata is not still following Bhatta Lollata, for whom rasa is the stable emotion in the character under conditions of full development, and for whom, accordingly, the formal constitution of aesthetic emotion, rather than its reception by the reader, continues to constitute the focal point of the analysis.

FROM ORNAMENT OF POETRY OF RUDRATA (R) WITH NOTES OF NAMISADHU (N)

(N) The author has already described the benefit to a poet from composing a poem. But what of the benefit to the listener? To

this he now speaks:

(R) (12.1-2) To be sure, literature produces, quickly and sweetly, an understanding of the four ends of man for sensitive people,243 who shy away from learned treatises, dry as dust as they are.244 Therefore, one must take all possible care to endow a literary text245 with rasas. Otherwise such people will recoil from it as from the treatises.

(N) If this were the case, one might object that literature would then only be for the sensitive,246 not for those who lack sensitivity, and would therefore not be available to all people. But this objection does not hold. We are only describing a way to ensure the attention of sensitive people, not to prohibit the attention of the insensitive; they may certainly attend to literature as well. One might also ask why the rasas were not discussed among the ornaments. We would answer: Words and meanings make up the body of a poem. Figures such as "crooked expression" and "direct description"247 are factitious ornaments, like bracelets and earrings. The rasas, however, are innate qualities, like beauty, and accordingly require that a separate chapter be started for them. What are these rasas? He now sets them forth:

(R) (12.3) The erotic, heroic, tragic, macabre, fearful, fantastic, comic, violent, peaceful, and affectionate-such are all the rasas.

(N) The straightforward meaning here is not all there is to it.248 Note that the erotic is placed first in order to indicate its primacy. Moreover, the word "such" means "of this sort," and accordingly other emotions of this sort-sagacity,249 despair, paralysis250-are all to be understood to be rasas. But how is it that despair and the like become rasas? He now speaks to this:

(R) (12.4) Insofar as the teachers have identified certain emotions as "rasas" because they can be tasted (rasanād), the way sweetness, sourness, and the like can be tasted, other emotions such as despair can be rasas as well, since they too can by all means be tasted.

(N) The "teachers" are Bharata and others, who said that the stable emotions are rasas because they are tasted, that is, savored, just like something sweet or sour, which becomes a rasa only by virtue of being savored. To quote, "Just as connoisseurs eat and savor their fare .... "251 In response to the likely objection that taste may be possible only in the case of the stable emotions, the author responds that "other emotions such as despair," i.e., the transitory emotions, and for this reason they too have to be understood to be rasas. Someone who believes that these cannot achieve enhancement will hold that they remain simply emotions. But the author's idea is that there is no mental state that cannot achieve enhancement and become rasa. That Bharata used the technical name only in reference to eight or nine rasas252 was owing to the fact that they are, for the most part, the ones that attract sensitive viewers.253 Now the definition of the erotic:

(R) (12.5-6) The behavior, consisting of desire, of a man and woman who are mutually attracted is the erotic rasa. It is of two sorts, the erotic enjoyed and the erotic thwarted. The former is when the couple are together, the latter when they are parted. The erotic is again twofold in being either hidden or open .... 254

1.6 RASA CANNOT BE EXPRESSED OR IMPLIED, ONLY "MANIFESTED"

Light on Implicature of Anandavardhana (c. 875)

With the literary theory of Anandavardhana we enter a new world of critical sophistication. In his Light on Implicature, rasa becomes the central phenomenon of literariness for poetic as well as dramatic forms. Theoretically, rasa now defines literature; pragmatically, it is now the element to which all other features of the literary text are subordinated. Once again the chronicler Kalhana helps us locate our thinker historically, identifying him as one of the scholars who won fame during the reign of King Avantivarman of Kashmir (855/6-883).255 Aside from the Light on Implicature, only one other work of his is extant, the devotional Hundred Poems to the Goddess (Deviśataka), which ironically embodies little of the literary theory for which he became famous. He refers in the Light to other compositions, but none has survived. In one sense it is impossible to make a selection of passages on rasa from Ananda's work, since as a whole it is about rasa. Although according to his theory, not all literature is concerned with the communication of rasa, when rasa is present it becomes the central organizing component of the work, and all other aspects must be brought into harmony with it: "The first domain of action, for a good poet, is rasa." From the school of scriptural hermeneutics Ananda appropriated the notion that discourse has a "unity of purpose," and it is rasa, rather than figure of speech or language quality or anything else, that constitutes literature's unifying purpose.256 Rasa here, finally and explicitly, has a fully developed character: it applies equally to poetry and to drama; the entire panoply of its components-the "semblance" of rasa, for example, and the notion of its "cessation"-is known, if hardly developed;257 works as a whole are shown for the first time to have a single rasa; Ananda argues in the case of the Mahābhārata, for example, that the dominant rasa is the peaceful. All that being said, however, rasa itself has an undertheorized presence in Ananda's treatise, astonishingly so given its conceptual primacy. He never tells us what rasa actually is or why it should be central to the literary work; he never comments on, or even cites, the celebrated Sutra on Rasa. More important, he never addresses the question of how the reader knows or experiences it. We are justified in inferring from this loud silence that for Ananda, none of these questions mattered, and that the phenomenon of rasa was basically unproblematic. He conceived of it as his predecessors had; the great insights that would transform aesthetic theory were a generation away. For all his remarkable insight into how literary meaning is engendered, Ananda plainly had no interest in how it was experienced, and in this he conformed to the formalistic analysis that he inherited. What was important for this analysis was not so much where rasa is or who knows or experiences it-these questions were to be made central only later-but how it comes to be where it is. Though he never names Bhatta Lollata or Shri Shankuka, Ananda was aware of the disputes associated with them, in particular the processes by which rasa is produced in the literary text, for these are the disputes he sought to resolve: how "language qualities" or figures of sound or sense (Dhà chapter 2) and "texture," or degree of compounding (chapter 3.5) relate to rasa; by what processes rasa is produced at the level of the literary work as a whole (3.10-14); what obstructs or contradicts rasa (3.18-20; here he offers an innovative discussion, which would be fleshed out and more thoroughly systematized by Mammata). What Ananda wants to understand are thus the textual processes by which rasa is made manifest. Rasa holds no interest as a theoretical problem, or as subjective phenomenon-the question of aesthetic reception -however much it is the viewer/reader who makes the judgments about the successful or unsuccessful manifestation of rasa, and to whom in the final analysis rasa is "manifested" or "implied" or "suggested."258 To all appearances, rasa for Ananda is something that exists in the text and indeed, in the main character (or in the author, insofar as he becomes a character when the narrative voice of a lyrical poem is his own). It is something to be "revealed" by the text and so presumably is located in it; the writer "composes" it and the narrative elements "possess" it. When discussing the relationship of linguistic "texture" to the suggestion of rasa, Ananda observes that the rasa must be "appropriate to the speaker and to what he is saying." "The 'speaker,'" he adds, "can be either the authorial voice or someone presented by the poet, and the latter can be either endowed with rasa and stable emotions or devoid of them." The one time Ananda actually mentions the locus of rasa, he identifies it as the character, either the protagonist or the antagonist;259 it is "the hero's erotic rasa," for example, that "has reached the highest pitch of enhancement." Clearly, as Ananda's most recent translators point out, he is using rasa "in its old

sense of a particularly vivid emotion" rather than in its emergent sense of "aesthetic delight."260 All he is concerned to explain is the mechanism by which the text produces rasa, and here the idea of "implicature" takes on critical importance. For Ananda, literature as such is constituted by "implicature," or meaning without saying.261 He got the idea of this new language function from a particular genre of lyric poetry (in Prakrit) where the main point of the narrative is only implied, not openly stated. This insight led to another, that implicature could apply to figures of sense as well, since these are sometimes directly expressed but sometimes only suggested; and to rasa itself, which thus forms a third domain of implicature to complement those of narrative and figure. There is no standard language function, primary or secondary, that can actually produce the sense of emotion in literature. It cannot be expressed literally; Ananda firmly sets aside Udbhata's old idea about the use of the "proper terms" for the rasas. It cannot be expressed figuratively, through metaphor, metonymy, or another trope. A new function has therefore to be postulated: emotion in literature can only be communicated by implicature-or, to use Ananda's other technical term, can only be "manifested"-via the combined activity of the aesthetic elements as each generates its own meaning. Implicature is the "soul" of poetry, and rasa is the "soul" of implicature.262 Again it is the textual processes behind the production of implicature, not its subjective reception, that matters. Ananda had no intention of proposing a new understanding of rasa as such, only a new model of its mode of textual realization coupled with a more thoroughgoing argument about the organizing force it exerts on the literary work. His conceptual universe is that of his immediate predecessors; rather than rejecting Bhatta Lollata's ontological view that rasa actually comes into being in the character, Ananda makes it entirely a function of language, and he does the same with Shri Shankuka's epistemological view that rasa is something inferred on the basis of an imitation in the actor of the character's stable emotion. His silence about his predecessors notwithstanding, it seems likely that his new explanation may have been influenced by Shri Shankuka's distinction- to the degree we can understand it from the meager fragments available-between referential and expressive language.263 The new theory of implicature is complicated by multiple varieties and subvarieties, which we can pass over here. What is critical to register for the intellectual history of rasa discourse is that Ananda is positing a new linguistic modality (śabdavrtti) for generating rasa in the text. And the most startling development in the coming centuries will see this turned into a psychological modality (cittavrtti) for producing rasa in the viewer/reader, something that, given the actual arguments of the Light on Implicature, would likely have left the author mystified. If the idea of implicature transforms the entire future history of rasa discourse, the future history will, in turn, transform the idea itself. One last small but telling innovation of Ananda's concerns the old notion of the three (or four) "emotion tropes," the rasa- laden and the like. For anyone who had absorbed the understanding of rasa made available by the new commentarial work on the Treatise on Drama produced in the course of the ninth century, the very idea of rasa as just another among the host of figures familiar from the rhetorical tradition can no longer have made much sense (we have seen the strains in Udbhata's model). In Ananda's account, the "rasa-laden" refers to the case where a rasa, the tragic, for instance, is present but subordinate to the main thrust of the poem, the poet's flattery of a king, for instance.264 In the example given in the selection below, the tragic rasa only serves to enhance-to "ornament"-the eulogy of the king, and therefore it has a secondary, not primary status, which belongs to the eulogy (it is "rasa-laden," not rasa). Although Ananda clearly captures something real in the cases he cites, Kuntaka in the next century will attack his views, claiming that the expression of emotion remains dominant in the poems adduced, and that in general rasa can never be subordinate.

FROM LIGHT ON IMPLICATURE OF ANANDAVARDHANA

(p. 78) The third type of implicature concerns rasa. Here rasa is suggested by the expressed meaning. [80] This is something completely different from expressed meaning because rasa cannot be the object of the primary function of the words. For it to be "expressed" would mean one of two

things: being directly made known by the proper terms for rasa ("the erotic" and so on), or being indirectly made known, by way of communicating the aesthetic elements.265 With respect to the first option,266 we would wind up with the absurdity that we could not apprehend a rasa unless its proper term were made known. [81] But rasas are not always made known by their proper terms, and where they are, we in fact apprehend them only indirectly, by their aesthetic elements being communicated; and there the proper term would only serve to restate the rasa, not produce it. That is why we do not find ourselves apprehending rasa in other sorts of contexts: [83] thus, in "poetry"267 that merely employs the word "erotic" without communicating the aesthetic elements, we do not have the least apprehension of rasa. Because of these positive and negative concomitances-that we apprehend rasa from the particular aesthetic elements all by themselves, without use of their proper terms, whereas we do not apprehend it from the mere use of their proper terms-rasas can only be said to be suggested by the expressed meaning .... (1.5) (p. 84) Precisely this signification is the soul of literature. And this is why long ago the first poet's grief at the separation of the pair of cranes was shown to become verse.268 [87] A literary text may be beautified by a whole range of words and significations, but it is this suggested meaning that constitutes its essence. [88] This is why the grief (soka) of the first poet Valmiki that arose at the cry of the crane-when its beloved was slain and it was overcome by despair at being forever separated-was shown to be transformed into verse (śloka). [89] For grief is the stable emotion of the tragic rasa; [90] although other kinds of implied meaning are found to exist, they are all meant to be included in the implicature of rasa and emotion, since that is the most important of all .... (2.1-2). Basic implicature is of two kinds: the literal meaning is not intended at all; the literal meaning is intended but is subordinated to some other meaning. The second kind itself has two varieties, depending on whether or not the temporal succession between the literal and implied meanings is registered by the reader. [175] With respect to the two varieties of the second kind of implicature, (2.3) Rasa (along with emotion, semblance of rasa or emotion, cessation of either, and so on) comes to light without the temporal succession being registered; and when it does so as the predominant element of the literary text,269 it is distinguished as the very soul of implicature. [180] Rasa comes to light at seemingly the same moment as the literal meaning. And when it does so as the predominant element, it becomes the very soul of implicature. The author goes on to show that the domain of implicature of the sort where the succession is not registered is to be differentiated from the figure of speech known as the "rasa-laden": (2.4) The domain of implicature is where the signifier, the signified, and the various elements that lend them beauty are all directed toward rasa. [190] We apply the designation "implicature" to a literary text in which the words, their meanings, figures, and language qualities, are all used in subordination to the main meaning-rasa, emotion, semblance of rasa or emotion, cessation of rasa or emotion-and are all themselves mutually differentiated from implicature. (2.5) [191] But in a literary text where some other270 subject matter is predominant and rasa subordinate, rasa will function as a figure of speech. Such is my view. Although earlier scholars have argued for a domain of operation of a figure of speech called the

"rasa-laden," my position is that the domain in which rasa is thought to act as a figure is simply where the rasa is subordinate in the text, with some other subject matter showing itself as predominant.271 Take, for example, a verse of royal flattery: here the affectionate utterance272 itself [193] constitutes the subject matter, whereas the rasas are seen to be subordinate. When rasa thus does function as a figure of speech, it can be either pure or mixed with other rasas or figures. Here is an example of the former:273

"Don't laugh,274 I'll never let you go again! It's been so long since I've seen you. Why do you leave so often, cruel man, what's driving us apart?" This is how your enemies' wives speak in their sleep, embracing their husbands tight, until they awake-and burst into bitter tears at their empty arms.

In this poem, the tragic rasa, which is here unmixed with anything else, is subordinate,275 and it is this that clearly makes it a figure "laden with rasa."276 ... [197] Where rasa itself constitutes the subject matter, however, in what way could it possibly be a figure of speech? For it goes without saying that a figure is by definition a source of beauty for something else, and cannot in and of itself beautify itself. Here is a précis of the matter:

What makes a figure of speech a figure of speech is the fact that it functions in a context where the main point is a rasa.

[198] Accordingly, where rasa constitutes the main subject matter it can never be assigned to the domain of figuration, but must instead be a species of implicature, where simile and all the rest function as figures for it. Rasa functions as a figure only where something other than rasa constitutes the principal subject matter and rasa contributes to the beauty. Thus, implicature, on the one hand, and simile and the like as well as the "rasa-laden" figure, on the other, have entirely distinct spheres of operation. Now, one might object that where rasa functions as a figure of speech, the subject must be a sentient being.277 [200] But that would mean that all other figures, simile and the rest, would have no scope, or virtually no scope, to operate, since wherever the actions of insentient things are the subject, we can always find some way to construe them with actions of sentient beings.278 One might rejoin that, regardless of such a construal, a text with insentient things as subject is beyond the scope of operation of the "rasa-laden" figure. But if that were true, vast numbers of works of literature, storehouses of rasa though they are presumed to be, would be condemned to lacking rasa [203] entirely. Take for example the following verse:

These waves her frowns, these startled birds her belt, trailing foam like a wrap fallen loose in agitation, meandering along a broken course while brooding on my lapses ... surely this unforgiving woman has been transformed into a raging river.279

Although here an insentient entity, the river, may be the main subject, we most certainly construe it with the action of a sentient entity. If, however, one were to admit that where we do have such a construal, we can have rasa functioning as a figure of speech, then we would have virtually no scope for the other figures, simile and the rest, to operate, since there is no action of an insentient entity that cannot be construed with that of a sentient one, at least as a stimulant

factor. [204] In short, rasa can be a figure of speech only where it is subordinate; where it is dominant, it must by all means be what the figures ornament, not a figure itself-and that is the very soul of implicature.

(p. 328) Now, while everyone knows that implicature where the temporal succession from the literal to the implied meanings is not registered-that is, the implicature of rasa-also appears at the level of the work as a whole, as in the case of the Mahabharata or the Rāmāyana, precisely how this comes about needs to be explained .... 280 The literary work as a whole can manifest rasa (and emotion, and so on), and the first factor in its capacity to do so is presenting the narrative in a way beautified by "propriety" in all the aesthetic elements-in other words, when the foundational factor and the rest are appropriate to the rasa meant to be conveyed. [330] The foundational and stimulant factors are familiar to all. As for propriety with respect to the stable emotions, this derives from the proper type of character. Characters can be differentiated as human or divine and so on,281 and among the former, as having high, middling, or low status. When the stable emotion is presented as unmixed with any other and suitably harmonizing with character, the emotion can be said to have propriety. If, on the other hand, an emotion (determination, say, or whatever it might be) that should belong exclusively to a human were presented as being located in a divine being, or the reverse, it would be inappropriate. In the same way, if in describing a king, who is a mere human, actions such as leaping over the seven seas were ascribed to him, they would of necessity lack rasa, however cleverly presented, because of their impropriety. Now, it might be objected that legend tells of kings like Satavahana who visited the realm of the Naga serpents,282 and other things of that sort, so what impropriety could there be in describing extraordinary feats on the part of kings capable of sustaining the entire world? But that would be incorrect. We are not saying that it is improper to describe such feats in the case of kings, only that when a fictional story is composed with reference to a being that is merely human, nothing appropriate to a divinity should be ascribed to him. In the case of a story concerning god-men, there is of course nothing whatever amiss in ascribing behavior as is appropriate to both, as for example in the story of the Pandavas. With respect to the exploits of Satavahana and the like that we hear of in legends, their description appears suitable to the degree that the exploits are restricted to those related in the legends; presenting anything about them in excess of this would be improper. Here is the essence of the matter:

The one thing that can impair rasa is impropriety. Composing Composing with customary propriety-that is rasa's deep secret.

[331] This is why in Bharata's Treatise on Drama a strict rule is given that drama should use well- known plots and a well-known noble hero, for the poet will never be confused about what is and is not appropriate for him. The poet who creates a play with a fictional plot, by contrast, risks erring when describing a hero's nature that precisely because of its unfamiliarity might seem to be inappropriate.283 ... Hence, in the case of desire no less than determination, propriety with respect to the character must be observed-and this applies to all the other the stable emotions, amazement and the rest.

The fact that in this context, great poets may have proceeded in their works without regard for such propriety is simply to be judged a fault, even if, as already observed, it may not always be registered because of their artistry.284 As for propriety of the reactions, this is something already made known in Bharata's Treatise and elsewhere. [334] In short, a poet must not only follow the norms enunciated in the Treatise and examine the works of great poets but also focus his mind and make the greatest effort to avoid any lapse in the propriety of the aesthetic elements. When we say that a factual or fictional narrative becomes capable of manifesting rasa when it embodies propriety, we mean that a poet should accept only such narratives-and no others among all the various tales of history and the like, however rasa- laden they may be-that are proper in all their aesthetic elements; and that he must take far more care with a newly invented fictional narrative than with a factual one. For if he stumbles from inattention there, he will expose himself to a serious charge of ignorance. Here is a verse that summarizes the matter:

A fictional narrative must be handled in such a way that it appears replete with rasa throughout.

And we have already shown that the means to this end is conformity with propriety in the use of the aesthetic elements. Moreover,

In the case of narratives like the Rāmāyana that are famous for a given rasa, any subjective change that might introduce a conflict of rasas should be avoided.

In a word, one should avoid subjective changes in such narratives. To quote: "One must never overstep the limits of the story, however minimally."285 That said, if one must make such a change, it must not introduce a conflict of rasas. A second factor in the capacity of the work as a whole for manifesting rasa lies in eliminating a plot component not in harmony with the rasa in one way or another and replacing it with a narrative element, even a fictive one, more appropriate to the desired rasa; examples may be found in the works of Kalidasa, Sarvasena's Vishnu's Victory (Harivijaya), or my own Arjuna's Deeds (Arjunacarita). As we have said, in composing a poem, a poet must devote himself heart and soul to rasa, and hence, if he notices some component in the plot not in harmony with the rasa, he should go so far as to expunge it and feel free to introduce another story element in harmony with the rasa. For no purpose is served when a poet simply reproduces the storyline, something already accomplished by the historical account itself .... [341] Another ingredient in the capacity of the work as a whole for manifesting rasa is, first, when the rasa is stimulated and moderated as occasion demands [342] (as for example in the drama Ratnavali), and, second, when the dominant rasa is called to mind again whenever it threatens to disappear (as in the drama The Ascetic King of Vatsa). The final ingredient for the manifestation of rasa in the work as a whole, be it a drama or whatever, is the measured use of figures of speech, however great may be the poet's talent for them. For a talented poet can sometimes become so involved in composing figures that he produces his composition without any regard to the position of rasa. I offer this advice because we actually do find poets so filled with the rasa286 of creating

figures that they completely disregard the basic rasa of their compositions.

.... (p. 360) The author, having explained how rasas are essentially "manifested," now asks how rasas can come into conflict with one another: (3.17) In composing the rasas, whether of a whole composition or a single poem, the thoughtful poet will take care to avoid any conflict between them. Otherwise he will not succeed in filling a single verse with rasa. With respect to the rasas whose conflict with one another a poet must take care to avoid, the author has the following to say: (3.18-19) [1] The use of aesthetic elements connected with a conflicting rasa; [2] discoursing at length on something unrelated to the rasa, even if that can somehow be made to construe with the rasa; [3] suddenly discontinuing or suddenly revealing the rasa; [4] continuously stimulating a rasa even after it has attained full enhancement; [5] inappropriate behavior: all this can contradict a rasa. [1] One source of conflict lies in making use of aesthetic elements connected with one rasa that conflicts with another rasa already in play. For the use of [1a] the foundational factor of a conflicting rasa, an example would be when acknowledged factors of the peaceful rasa are first described and directly thereafter is the description of a factor of the erotic. For the use of [1b] the transitory emotion of a conflicting rasa, an example would be a lover's talking about the transience of life and other sources of dispassion when trying to assuage his beloved's anger at his unfaithfulness. For the use of [1c] the reaction of a conflicting rasa, an example would be a poet's describing reactions associated with the violent rasa for a lover who is beside himself with anger when his beloved fails to relent in her jealous anger toward him. [362] [2] Another thing that will disturb the rasa is when a poet discourses at length on something unrelated, even if that can somehow be made to construe with the rasa; as, for example, when he goes on endlessly describing mountains or whatever or gets lost in composing "twinned" figures and the like when he is in the midst of describing a hero in a state of the erotic thwarted.287 [3] Another cause of disturbance is when the rasa is suddenly [363] discontinued or suddenly revealed. An example of an inopportune cessation would be when the hero's erotic rasa288 has reached the highest pitch of enhancement thanks to a heroine with whom he yearns to be united- and it is clear that they are mutually attracted-and then the poet shows complete indifference to the activity that would unite them and arbitrarily begins to describe some wholly unrelated activity. Inopportunely revealing a rasa occurs when, for example, in the midst of an apocalyptic battle where heroes are dying289 the poet starts describing, for no good reason, the erotic rasa of some godlike hero who has not earlier been shown to be in the least subject to it. In the latter case it would not do to argue that the hero of the story may have been driven mad by fate, since after all the creation of rasa is rightly the poet's principal raison d'être;290 plot description is simply a means to that end, as we said earlier.291 And hence, were a poet to be preoccupied with merely describing the plot, he would wind up with a composition where the rasa and emotions have no clear hierarchical ordering, and he would thereby be liable to misstep in the ways noted. It is for this reason I have taken the trouble I have here, because it seemed right to give poets the gist of how rasa is manifested, [364] and not from an obsession with communicating every single dimension of implicature. [4] Rasa will also be disturbed if it is continuously stimulated even after it has attained full

enhancement. For once a rasa has been fully enhanced through the complement of its aesthetic elements and then come to be experienced, it will seem like a faded flower if the poet recurs to it again and again. [5] The last disturbance is inappropriate behavior, as for example if the heroine were to speak to the hero, without any obliqueness whatever, of her desire for sexual intercourse .... 292 Good poets must accordingly strive to avoid such rasa conflicts as these and others along the same lines that they can come up with on their own. Here is a verse that gives the essence of the matter:

Rasa should be a good poet's chief object of concern, and he must avoid all error in creating it. A composition lacking rasa brings infamy on a poet-indeed, he ceases to be a poet at all, and his name will be lost to posterity.

(3.29) (p. 397) A good poet must be particularly attentive in the case of the erotic rasa. For if he commits an error here, it will be immediately noticed. The erotic rasa is far more delicate than any other, and therefore a poet must be particularly attentive or careful about it. If he makes an error, he becomes an object of scorn to receptive readers. This is because the erotic is necessarily an object of every living being's experience,293 and insofar as it is accordingly more pleasing than any other rasa, it is preeminent among them .... [404] No one thinks that the rasas are nothing more than the factors, the reactions, and the transient emotions. Accordingly, the apprehension of the aesthetic elements and the apprehension of the rasa that depends upon those elements are classified as cause and effect respectively. A temporal succession must of course exist between these two, but it is not discernible because of its brevity. And so we have said that the rasas are in the category of things "manifested" with no succession being registered .... [497]294 For mature writers it is unseemly even to try to compose literature if they do not give precedence to rasa; whereas if they do, there is nothing that does not profit by being included as a component of the intended rasa. Nothing, not even an insentient thing, cannot be a component of the rasa, whether by strictly functioning as a foundational or stimulant factor appropriate to the rasa or by being shown to behave like an sentient being. Here [498] are three apposite verses:

In the boundless world of literature the author is sole creator god: the whole universe takes on whatever form he wills. If the poet is filled with passion, the whole world of his poem will consist of rasa; if not, it will be completely devoid of it.295 A good poet can treat insentient things as sentient, and the reverse, just as he pleases, since he has complete autonomy in his literary work.

Therefore, anything can become a component of the intended rasa at the will of the poet who gives precedence to rasa heart and soul, and can enhance the beauty of the composition in which it is included.

1.7 RASA CANNOT BE A FIGURE OF SPEECH; IT IS WHAT FIGURES OF SPEECH ORNAMENT

The Vital Force of Literary Language of Kuntaka (c. 975)

Perhaps the only work in the Sanskrit tradition that can be likened to what today we would regard as literary criticism-sustained interpretation of the literary text, often from the vantage point of the work as a whole-Kuntaka's Vital Force of Literary Language was composed in Kashmir near the end of the tenth century. As often, we can determine the author's date only on the basis of relative chronology: whom he quotes (e.g., Rajasekhara, c. 920) and who quotes (or rather, seems to allude to) him (Abhinavagupta, c. 1000). Almost from the start-and almost to the end-participants in the tradition of discourse on the literary sought to identify some singular feature around which to organize their understanding of literariness.29% For Bhamaha and Dandin, this core feature was figuration; for the early ninth-century poetician Vamana it was "style"; for Anandavardhana, "implicature"; for Kshemendra in the early eleventh century, "propriety." For Kuntaka it was what he called vakrokti, literally "crooked utterance," where the term vakra refers not, as earlier, to indirect expression as exhibited especially in figurative language, but rather to everything from phonemic quality to the meaning of the work as a whole that renders expression notable; something like "striking usage" would be a fair translation. Accordingly, in contrast to Ananda, rasa has a reduced role to play in Kuntaka's work, and in his understanding of that phenomenon he harkens back (without naming him directly) to the founding formulation of Bhatta Lollata. For Kuntaka, rasa is in the first instance the "enhancement" of the stable emotion of the leading character through the combined force of the aesthetic elements. The examples he provides in the selections offered here show the process whereby, in the first case, the erotic thwarted and in the second case, the tragic come to be enhanced in the male character by a range of stimulant factors. Kuntaka's acceptance of the old formalist theorem, moreover, led him to embed his discussion of rasa in an analysis of "the subject matter" (varnanīya). This he divides into sentient and insentient entities, the former further subdivided into two groups, "primarily sentient" beings-deities, demigods, and humans-and "not primarily sentient" beings-animals. Self-evidently, rasa can be placed within such an analytical framework only because Kuntaka conceives of it as a phenomenon contingent on the nature of the receptacle: rasa pertains only to higher beings ("the activity of the primarily sentient beings ... consists essentially in their rasa"); in lower beings there can be only a "semblance of rasa," precisely because they were not thought to possess the cognitive- emotional apparatus to feel it. The complex relationship of rasa and consciousness, whether human, divine, or animal, will occupy thinkers in the following centuries (especially Vidyanatha and Singabhupala); the very verse Kuntaka uses to contrast the "impulses" of animals to the stable emotions of higher-order beings, namely the opening verse of Kalidasa's Shakuntala, will be the first adduced by Abhinavagupta in his presentation of his own theory of rasa, a distinction explained by the latter's focus on aesthetic reception.297 All these points about rasa are largely received opinion; where Kuntaka directs new critical attention is toward the old notion of the "emotion tropes." We have seen that much of the complicated history of rasa theory's conceptual transference from drama to poetry is concentrated here. First categorized as just three figures of speech among countless others (rasa-laden expression, affectionate utterance, and haughty declaration), they were eventually rethought by Udbhata as components of a more fully realized, even dramaturgical, theory of poetry (the full realization of rasa, the intimation of an emotion, and the semblance of rasa, respectively, with the quiescence of rasa added as a fourth). But for all that, they remained figures for Udbhata, no different from simile or metaphor; even Anandavardhana made conceptual space for rasa functioning as a figure. Kuntaka is the first to deny that rasa can ever be a figural "ornament" to a literary text, directly and forcefully critiquing both Udbhata and Ananda. And he does so at a level of detail and with a degree of tenacity impossible to convey in a brief selection. Despite the overall analytic framework that keeps Kuntaka's attention focused on the form of the literary work, he does here and there make reference to audience response. Although he nowhere mentions Bhatta Nayaka (see chapter 3.1), the notion that rasa above all concerns the aesthetic response of the viewer/reader had been fully naturalized by Kuntaka's date. Hence elements of a reception theory of rasa ("rasas produce the rapture of aesthetic relishing in any sentient creature"; "the reader ... whose heart ... overflows with rasa") coexist with Kuntaka's more formalistic focus, suggesting the discursive durability of the older paradigm.

Kuntaka's work has come down to us in somewhat fragmentary form, and without commentary. His often florid style adds to the challenge of following his argument. But if it is a struggle to understand this careful and deeply insightful reader of literature, it is decidedly a worthwhile one.

FROM THE VITAL FORCE OF LITERARY LANGUAGE OF KUNTAKA

(3.7) (p. 138) The "primary" category is made beautiful by the enhancement of their unaffected desire and the like; the other becomes adorned when reference is made to the impulses appropriate to their particular species. The nature of the "primary," that is, principally sentient,298 category of beings-gods, antigods, and the like-becomes fit subject matter for poets, or in other words, enters the field of their proper literary creativity, when they are "made beautiful by the enhancement of their unaffected desire and the like." Desire and so on are the stable emotions, which are "unaffected" when they are free from any constraint, that is, beautiful thanks to their naturalness.299 They are said to be "enhanced" when they come to exist as the erotic and the other rasas, according to the axiom that rasa is nothing but the stable emotion. This makes the category of beings "beautiful," that is, enchanting. An example in the case of the erotic thwarted is the lament of the mad Pururavas in the fourth act of the Urvasi Won by Force:

She might be angry and hiding herself by magic-but she would not stay angry so long. She might have flown up to heaven-but her heart would still be soft on me. Even the foes of the gods could not abduct her from my presence-and yet she is completely lost to view. How can fate ever have arranged this?300

[139] Here the king is almost beside himself with misery from having lost his beloved, and is confounded by his repeated failure to recover her. He tries to imagine some reason for being unable to catch sight of her, carefully making reference to each option from moment to moment, in the very first instance calling to mind her natural gentleness, and then dismissing each argument301 by reflecting on what she would be expected to do. Given the impossibility of any sequence of events302 that might lead to her recovery, he becomes utterly distraught with the certainty of hopelessness. This brings the rasa to the point of full enhancement. And this is further stimulated by later statements, such as the following:

Had that shapely woman touched the earth with her feet where the sandy ground of the forest is wet from rain, I would surely see a trail of footprints, with traces of lac, with a deeper depression at the heel from the weight of her limbs.303

Here his hopefulness ("Perhaps she once touched the ground with her feet")304 leads him to imagine he might recover her. But given that a trail of footprints would have been visible, beautified by red lac and with her figure imprinted deeply at the heel from the weight of her hips (since the sandy ground of the forest had been softened by the rain), his certitude of hopelessness grows all the stronger, and occasions the mad lamentation that marks the following sentences.

An example of the tragic rasa is the lamentation of the king of Vatsa in the second act of The Ascetic King of Vatsa:305

You only glanced at the riverside house and wandered with forlorn look from the pleasure house, heaving with sighs and casting quick glances at the walkway of kesara vines. Why now come to my side, little son, and ply me with your fawning? It was your cruel mother who left you-and left me, too-and departed for the most distant land.306

Here the entire complement of foundational and stimulant factors that are the cause of the rasa's enhancement [140] has been fully developed by the poet. And the statement of the court fool that introduces this verse, "What a sad mistake for the fawn, the queen's adopted child, to be following you," has been employed for the same purpose, enabling the riverside house, the baby deer, and the rest to become stimulant factors of the tragic rasa all the more effectively. The poet has precisely the same intention when he presents the statement of the king (after the minister Rumanvat, hearing the parrot crying for the queen, says, "This is rubbing salt in a wound!"), which is addressed to the dead queen:

Your boon companion, who once touched that cheek of yours with his claw as he tried to peck the ruby earring from your ear, thinking it a pomegranate seed ... this poor parrot is now crying out in utter agony. And how, my queen, can you refuse to respond to him at all?307

Here the poet presents the parrot as the spoiled favorite of the queen with the intention of communicating her affection for him. The use of the word "that" referring to the cheek is meant to point toward its extreme delicacy, which the king himself had had the experience of enjoying. In this way, the tragic rasa, given that its very existence depends on such stimulant factors, is brought to the highest pitch of beauty. It is because these two rasas, the erotic thwarted and the tragic, are extremely delicate that we have drawn our examples from them. One can easily come up with illustrations of the other rasas. By contrast, the nature of the second, or secondary, category-those sentient beings that are animals, such as lions and so on-becomes fit subject matter for poets only when reference is made to the urges appropriate to their particular species.308

.... [143.18] Thus, the "body" of the narrative, which can consist of either an entity's being endowed with rasa or its innate beauty309-the former depending on whether its rasa is predominant, as in the primary category of beings; the latter, on whether its particular nature is predominant, as in the secondary category-must be "that which is ornamented," and cannot itself be an ornament. It has already been explained why the nature of a thing cannot be an ornament. The author now refutes the view of other poeticians who hold that when the subject of the narrative is the activity of the primarily sentient beings, which consists essentially in their rasa, the subject can indeed function as an ornament. (3.11) [144] The "rasa-laden" cannot be an ornament, first, because it constitutes the very thing we are apprehending in a poem, and there is nothing above and beyond it; and second, because no coherent relationship exists between the literal meaning of the word itself

("rasa-laden") and the putative meaning ("the rasa-laden ornament").310 If it is the rasa itself, the erotic and so on, that constitutes the subject matter and hence the "thing to be ornamented," then something else must constitute the ornament. Conversely, if rasa is said to be an ornament itself on the grounds that it gives pleasure to connoisseurs, then something above and beyond it must be proposed as the thing to be ornamented. The ancient poeticians did not give the least thought to this simple distinction when they defined and illustrated the "rasa-laden" ornament .... 311 [145.23] Nor can the example they offer-given that an example's only salvation lies in the definition it is meant to exemplify-be considered an option all on its own. [146] For in the poem adduced, "The woman I thought was dead and hoped / to rejoin by taking my own life,"312 there is no separate thing to be apprehended beyond the mental state itself that constitutes the narrative content, which is nothing other than the enhancement of the stable emotion of desire. Hence, we can only conclude it is that which is to be ornamented. Now, one thinker has introduced into the definition the qualification that the "proper term" provides one of the five basic components of rasa.313 The idea that rasa should be based on its proper term, however, is to our way of thinking completely bizarre. We might ask this learned scholar, who ponders so deeply the essence of rasa and pretends to know its ultimate truth: What is it actually that you are claiming is "based on its proper term," the rasa itself or the "rasa-laden" ornament? On the first alternative, the word "rasa," if we think of it being based in its proper term, means literally "to be tasted," and thus, when referring to the erotic and so on, rasa would necessarily be tasted by anyone who knew the word's meaning-which would amount to saying that the different rasas produce the rapture of aesthetic relishing in any sentient creature as soon as they are denoted by their proper terms and come within earshot. By the same reasoning, any object, when denoted by its proper term-like "candy"-would provide the full experience of savoring the thing, and so anyone who craved any pleasure would, without any effort, just by the word's merest denotation, be proffered by this noble sir314 pleasures equal to a king's ransom. Thank you very much indeed! It cannot be the second alternative, either, where it is supposed to be the "rasa-laden" ornament that is based on its own term, since that word in fact expresses rasa and, as we just saw, it cannot have that as its basis, let alone something entirely unconnected. So at the very outset the idea that the rasa-laden can be an ornament is refuted .... [149.19] Then again, it has been asserted that the rasa-laden figure may be understood as something providing a special beauty and brilliance to the poem.315 Thus one especially learned scholar has declared:

But in a literary text where some other subject matter is predominant and rasa subordinate, rasa functions as a figure of speech. Such is my view.316

That is, in the case where some other subject matter may, on the grounds of its prominence in the poem, be distinguished as the thing to be ornamented, the erotic or other rasa will become an ornament, since it is being presented as something subordinate to that subject matter. In this sort of context, it is claimed,317 a subsidiary element is adorning a principal one in such a way that the poet's own sentiment is what is made manifest, and for that reason, the differentiation of rasa as ornament becomes patently manifest .... [151.2]318 For example, our learned scholar cites the verse, "Don't laugh, I'll never let you go again!"319 and argues that the main subject matter lies in its core

meaning, the poet's flattery: "Here, Your Majesty, you see the kind of suffering, reaching the highest pitch of tragedy, that your enemies' wives undergo when, with their husbands slain by your hand, they are overwhelmed with sorrow and defenseless." The tragic rasa, for its part, is presented as something subordinate to that meaning, whereas the erotic (the variety known as the erotic thwarted through the lover's absence) has no reality whatever.320 Being communicated by a sentence-an array of syntactically coherent elements-and appearing clearly as auxiliary to something else, the rasa, he concludes, here must be called ornamentation. Now, it is certainly not the case that the definition of rasa cannot apply here: that would be so only if the full complement of its causal elements, including rasa's foundational factor, were absent. But the foundational object321 is clearly present. Even less is there any literary fault in the introduction of two conflicting rasas, the erotic and the tragic: although the two rasas do exist in the poem in actual fact, there is no conflict between them in our cognitive experience because they are not both simultaneously vying for primacy.322 Thus, the erotic too has the wonderful capacity here to delight sensitive readers, since there is no proof that only the tragic is at issue in the poem. The erotic thwarted, [152] elicited by its foundational factor (contained in the wives' words) that is its causal element, is found during their dreaming state; the same thing logically applies to the tragic rasa during the waking state. Hence both rasas are possible, and thus it seems hardly reasonable to argue that it is the tragic that should dominate from the very first instant. It hardly defies logic that the king's enemies and their wives should have become separated,323 having fled hither and yon, terrified by the fierce power of the great man who is the object of the poet's flattery. Even if it were clear that this is instead a case of the tragic rasa, we could understand the meaning as follows: the wives of your enemies, their minds overwhelmed by the calamity that has befallen them, finally attain reunion with their husbands in their dreams; they begin to converse as usual with their husbands about their experiences together; when they chance to wake up, they reflect on the difference between things before and after waking, their hearts are lacerated by the contrast in things brought on by those reflections, and they begin to weep. Hence, the erotic would be said either to enhance the tragic or to take on the traits of the transitory emotions appropriate to the tragic.324 How then could one argue that there is no trace of the independent functioning of the erotic rasa here? It is also problematic to argue that the affectionate utterance made to the king is predominant in the poem, whereas the tragic rasa, as something subordinate to it, must be ornamentation. First, in this example as well as the earlier one,325 what is clearly presented as the primary subject matter is the complex transformations of the tragic rasa. Second, it is not reasonable326 that the tragic should be directly expressed-on the analogy with other ornaments such as circumlocution or allegory where, though these are not separate from the expressed sense, something else is indirectly expressed327-precisely because, being a rasa, it is always supposed to be "manifested." Nor [153] could it be classed in the domain of what is called "subordinate implication," because the rasa, as something manifested-here in the form of the tragic-must always appear as predominant. And last, the affectionate utterance and the rasa cannot both be "manifested," because then the relationship of principal and subsidiary supposed to subsist between them would be logically untenable. We have only sought to show to the degree possible the various unacceptable options with

respect to the "rasa-laden" ornament; at all events, the analogy with normal figures of speech simply does not hold.

[156.10] Having refuted the notion that the "emotional" utterance can be an ornament, on the grounds that it constitutes part of the narrative content, the author goes on to offer the same refutation for the remaining figures, since they too constitute narrative content .... [156.18] The ancients tried to offer definitions and examples to show that these are ornaments, but being based on false reasoning, their argument is inadmissible. Let us examine both the definition and the example of the "impulsive" declaration: [157] "When rasas or emotions are presented as contrary to propriety .... "328 [157.12] What sort of impropriety can permit the presentation of rasas based on it to become an ornament? Not only will a rasa apprehended as a violation of propriety fail to become fully "enhanced," but indeed, ugliness will gain entry. To quote: "The one thing that can impair rasa is impropriety."329 Now, perhaps what is intended in the definition is not any real impropriety but, on the contrary, something in perfect harmony with the subject matter with reference to the rasa-which itself is being manifested by aesthetic elements that are indeed proper, not inapposite to the subject matter, and not objectionable. The logic behind such an argument is offered elsewhere in the definition: "because [those rasas or emotions] are caused by craving or anger or the like." But this would not be a very compelling justification either. A rasa or emotion may be presented [158] as touched by impropriety only when the rasa has been fully enhanced by propriety, and then only in reference to a lesser being, as a result of his lust or the like, and certainly not in the context cited.330 Only when its natural charm is imparted by the beauty coruscating with the innate artistry of the poet and enlivened by the full complement of aesthetic elements can even a fully proper rasa convey its deepest beauty to the reader, who is transported by the play of the fantastic and whose heart then overflows with rasa the way a moonstone is liquefied by moonbeams. How then can a rasa that is wilting under impropriety, its very existence compromised by the suspicion that craving or the like is involved, ever attain the high status of ornament? Thus, in the aforementioned context, great poets have found a way to invest the poem with rasa, on a par with other subject matter, and have done so with utter brilliance, for example:

Shiva himself passed those days only with difficulty, longing for reunion with the Mountain's daughter. How could such emotions not disturb a human, who is far less self-controlled, when they touch the lord of all?331

We have examined this in detail since it has come up in the course of our exposition. But it might be objected that we are disputing over matters of concern to the learned experts in the principles of Bharata's Treatise on Drama, namely whether or not the blessed lord, who has been cited in the example poem, should be brought into the narrative as an object of a "semblance of rasa"; deciding about propriety or impropriety, by contrast, is entirely beside the matter. Well, we are happy to concede the point to the good sirs. That said, however, what is narratively central in the poem is that particular mental state of the lord, and hence it cannot escape being the "thing to be ornamented," and not an ornament itself.332