As preoccupied as he must have been with ruling a large kingdom in what is now north central India, Bhoja nonetheless found time to patronize the arts, compose some widely admired poems, build temples and dams, and pen scholarly treatises on a range of topics. These include metaphysics (the Tattvaprakāśa or Light on Fundamental Principles) and aesthetics. The contributions to aesthetics are the Necklace for the Goddess of Language (Sarasvatīkaṇṭhābharaṇālaṅkāra), completed around 1025 CE, and the Light on Passion (Śṛṅgāraprakāśa), completed around 1050 CE, five years before Bhoja’s death (Bhoja 1979, Bhoja 1998). Chapter 5 of the Necklace, which treats rasa, is short, but it comes down to us with contemporary commentary. The Light numbers thirty-six chapters totaling two thousand pages (see Raghavan 1978: 8–11 for a synopsis). The theory of rasa is presented in the Introduction and Chapter 11 of the Light and Chapter 5 of the Necklace, all translated by Sheldon Pollock in his Rasa Reader (Bhoja 2016a and Bhoja 2016b; see also Raghavan 1976, Pollock 1998).

Pollock puts a question that orients us on Bhoja’s project. Why would a busy monarch make it a priority to write the summa poetica that is the Light (Pollock 1998: 140; see also Cox 2012)? Presumably, rasa theory inherits its importance from the importance of rasa and hence the beauty to which rasa contributes. Pollock proposes, in effect, that the Light seeks to answer what has come to be called the “normative question” about aesthetic value.

Rasa contributes to a work’s beauty (Bhoja 2016a: 115, 2026b: 123). Moreover, as the rasa tradition assumes, an item’s beauty is its power to produce pleasure. All this given, two questions arise (Lopes 2018: ch. 2).

One is a demarcation question. Since not all pleasures are aesthetic pleasures, what conditions are necessary and sufficient to demarcate aesthetic pleasures from such pleasures as we take in a massage, in leveling up to the Diamond League, in the fitness to purpose of a tool or policy, or witnessing an act of altruism? None of these is a pleasure in beauty.

A second question is the normative question. Setting aside what demarcates beauty from other goods, one might wonder why the fact that a play is beautiful gives anyone reason to watch it (or act in it, produce it, review it, or write it in the first place).

An answer is ready to hand if Eilidh joins other rasa theorists in espousing aesthetic hedonism.

Should Sonam question that, he would deserve an incredulous stare. However, aesthetic hedonists need not stop at a hedonic answer to the normative question. Many have not, including some rasa theorists (Lopes 2019 and Lawson and Lopes 2024 on Bhattacharyya 2011; Reich 2021 on Abhinavagupta).

Among them is Bhoja. He states two answers to the normative question. First, commenting on the Necklace, Bhaṭṭa Narasiṃha confirms that what “really matters” is that an artwork “delight the hearts of a receptive audience” (2016a: 115–116). Second, Bhoja repeatedly admits that works with rasa offer guidance with respect to the four constitutive ends of human life (e.g., 2016b: 120; see also Pollock 2001: 216–222). The guidance is moral. Pollock suggests a third answer to the normative question. The Light aims to

“Good readers,” he adds, “make good subjects” (1998: 141). Presumably, there is reason for us to be good subjects. So, we get hedonic, moral, and political answers to the normative question.

Granting that all three answers are correct is not entirely satisfactory. In the first place, one might wonder why there are these three answers to the normative question. How do they hang together? Might some deeper answer reveal why Sonam has triple hedonic, moral, and political reasons to check out the movie? Compounding our dissatisfaction, there is a gap between some of the three answers and the content of (r). According to (r), rasa is fundamentally the artistic expression of love. Moreover, Bhoja writes that “passion [love] alone is the means to fulfilling the four ends of man” (2016b: 123). How is the fact that a work artistically expresses love a moral (or political) reason to engage with it?

Suppose that (r) explains why rasa is a route to self-realization. Perhaps (s) supplies a fourth answer to the normative question. Sonam should check out the movie because it promotes self-realization. As we shall see in the final section, this fourth answer to the question implies and unifies the first three answers. Thus the argument for (r) and (s) is that they answer the normative question. Before getting there, a case must first be made that (r) explains (s).

The Light on Passion is no cakewalk. Its arguments are intricate, central claims are sometimes advanced without elaboration, and the main ideas are sufficiently unfamiliar (even to rasa wonks) that discipline is needed simply to take Bhoja at his word. By viewing it through the lens of the normative question, we begin to get the dots arranged and ready to be connected.

The thesis that rasa is fundamentally the artistic expression of love combines three ingredients: fundamentality, love, and artistic expression. Start with artistic expression. Bhoja assumes without argument that rasa is expressed by characters in narrative artworks.

The assumption is not that rasa is to be found only in characters. To begin with, Bhoja acknowledges that “a literary text is ‘rasa-laden’ if it is capable of expressing rasas” (2016b: 130). Texts express rasa at the level of “individual passages” or “whole compositions” (Bhoja 2016b: 127). Nevertheless, at either level, rasa is to be attributed to some character (Bhoja 2016a: 115). What about a work’s creator and audience? Bhoja also grants that poets and playwrights can express emotions through their work: “if the poet is filled with [love], the whole world of his poem will consist of rasa” (2016a: 116). Likewise, narrative artworks evoke emotional experiences in receptive members of the audience. Commenting on the Necklace, Bhaṭṭa Narasiṃha writes that rasa is “experienced by the audience when communicated through an appropriate literary composition” (2016a: 116). Indeed, “the whole point” might be for it to arouse its audience emotionally (2016a: 115).

The appearance of rasa in the character explains its appearance in the work, the artist, and the audience. A poet expresses what they feel by composing rasa-laden verse, the verse is rasa-laden because it depicts characters who express rasa, and audience affect is a response to what characters express through rasa-laden verse.

What is it for characters to express rasa? In company with other rasa theorists, Bhoja focuses on how rasa is expressed in poetry and drama through language (Cox 2023). By using language in special ways, literary works depict characters as expressing rasa, and audiences experience what they do by grasping that same language. Bhoja pays exceptionally keen attention to the mechanical details, often relying on carefully chosen examples (e.g., 2016b: 116). Pollock observes that Bhoja’s m.o. is “to constantly redirect attention back toward the text, toward the literary process itself and the production of literary communication” (2016: 113–114).

One might think, along with his commentator, Bhaṭṭa Narasiṃha, that Bhoja regards rasa as a strictly artistic phenomenon. The claim would be that,

An interesting passage speaks against (1). In distinguishing characters from the (fictional) beings that they represent, Bhoja writes that Rāma, the character, expresses rasa by representing Rāma, a person, as expressing rasa. The character’s rasa-laden speech echoes the person’s, and the person’s speech springs from “his very rasa” (Bhoja 2016b: 130). Both people and characters express rasa. Apparently,

This claim is inconsistent with (1).

The crux is whether rasa can be expressed only in narrative art. Bhoja frequently describes rasa as an intensified state (udayātiśaya or atiśaya), and it stands to reason that the intensity is achieved partly through artistic means. That speaks in favor of (1), but not decisively. One might reply on behalf of (2) that the intensity achieved in art through certain special techniques can be achieved by other means in everyday life. The intense expression of a character in an opera who sings through her final minutes on earth is achieved in another way when the singer actually dies.

For present purposes, we need not choose between (1) and (2). In order to interpret (r), we should concede the distinction between affective states of living creatures and affective states of characters, which must be expressed by artistic means. We should also concede that living creatures either express rasa or something akin to it. (Bhoja certainly agrees, as we shall see below.) Taking these two concessions on board, the difference between (1) and (2) may be treated as terminological. One may use “rasa” to name the artistic phenomenon, taking care to signal its relation to a more general psychological phenomenon. Or one may use “rasa” to name the general psychological phenomenon of which artistic rasa is a special case. Since it is important to interpret (r) while keeping the focus on characters, “rasa” is used below accordance with (1). That is, “rasa” will denote the specifically artistic phenomenon. (2) remains true when interpreted as saying that there is a broader psychological phenomenon of which rasa is a special (artistic) case.

To recap, the first ingredient of (r) is the claim that rasa is expressed by characters, elements of narrative artworks that represent sentient beings. Works are rasa-laden and can evoke affective experiences in audiences only because they house characters who express rasa. This section sidesteps an exegetical conundrum about whether Bhoja takes rasa to be an essentially artistic phenomenon. What matters is that there is an artistic phenomenon continuous with a broader psychological phenomenon.

Rasa theorists of all stripes share a standard model of emotion. Bhoja accepts the model but uses it to make a case for a famously iconoclastic claim. As he writes in the sixth of twelve kārikās (thesis statements in verse) that open the Light, there is exactly one fundamental rasa (2016b: 119). Call this claim “fundamentality.” Going further, (r) identifies the fundamental rasa with love, but this identification should be treated separately. The arguments for fundamentality do not imply that the fundamental rasa is love.

The standard model of emotion (bhāva) parses out several factors in emotional response (e.g., Bhoja 2016b: 130–131). First, Celeste has a psychological disposition (vāsanā, saṃskāra) to affective response in the presence of danger. Second, a lion has just crossed her path. This fact is the foundational factor (ālambanavibhāva) in her response. Third, she realizes that she has wandered away from her companions. Since this is not by itself dangerous, it is not a foundational factor that triggers her affective disposition. Rather, it amplifies her response; it is a stimulant factor (uddīpanavibhāva). Fourth, her feeling mixes with and gets nuance from various transitory emotions (vyabhicāribhāva), such as anxiety and despair. Fifth, she betrays her state through some psychological and physical reactions (anubhāva), such as trembling, crying out for help, and attempting to flee. In all this, the feeling of fear (bhaya) itself is a stable emotion (sthāyibhāva), because its functional role is focal, tying the other factors into a unified package. The stable emotion is just what someone is disposed to feel in the presence of foundational factors amplified by stimulant factors, as betrayed in psycho-physical reactions, accompanied by transitory emotions.

Characters in narrative artworks express rasa only when a sufficient combination of these factors is depicted by literary or dramatic means. Let Celeste be the elephant who is set upon and killed by a python and a lion in The Auspicious Flower (Kalyāṇasaugandhikam). In that scene, she expresses the fearful rasa (bhayānaka) partly because the scene uses theatrical techniques to depict: the lion, her solitude, her will to live, her cries and struggles to escape, and various emotions that she feels ancillary to her fear (see https://www.youtube.com/watch?v=mTdIaQCkKjY starting at 16:20).

In parsing the components of emotional response, the standard model of emotion does not imply a specific number of stable emotions or rasas. Nonetheless, most rasa theorists accepted a canonical menu of stable emotions, each having a counterpart rasa that is expressed or aroused by artistic means. The Nāṭyaśāstra entertained eight canonical rasas: the erotic or romantic (śṛṅgāra), comic (hāsya), violent (raudra), sorrowful (karuṇa), macabre (bībhatsa), fearful (bhayānaka), wondrous (adbhuta), and heroic (vīra). Later theorists expanded the canon. Broadly speaking, they sought to identify affective states that play a significant role in human life, especially in pursuit of what they took to be its four constitutive ends (Pollock 2012: 195–196). The stable emotions motivate the pursuit, with the transitory emotions coming on stage to play helping roles.

Bhoja adopts the standard model, but he spurns the claim that there is a menu of stable emotions to which the rasas correspond (see also Rudrata 2016: 86). The Light’s seventh kārikā leaves his dissent in no doubt: the canonical menu of rasas,

Put more politely, there are no grounds for privileging a canonical list of rasas.

One might read this as a modest proposal to expand the list of rasas, adding a handful to their number, as other rasa theorists did. However, the Light states an argument that undercuts any attempt to canonize some emotions as stable and the rest as transitory.

Chapter 11 of the Light argues that the canonically stable and transitory emotions are functionally interchangeable: a function performed by any of them can be performed by any of the others. Bhoja writes that, “depending on the circumstances, all can be stable or transitory emotions or even… psychophysical responses, because they all derive from the mind” (2016b: 125–126). For example, jealousy is canonically a transitory emotion that accompanies the stable emotion of love. In Othello, however, it is a stable emotion. He is disposed to feel it in the presence of Iago’s portrayal of Desdemona (the foundational factor), and it is amplified by some stimulant factors (her beauty and his precarious social standing). It is betrayed in psycho-physical reactions (including some memorable speeches) and it is accompanied by transitory emotions, especially, in Othello’s case, love and self-doubt. In addition to jealousy, Bhoja mentions other canonically transitory emotions that, in the right circumstances, can be focal, hence stable: shame, anxiety, joy, pride, satisfaction, fatigue, and sagacity. On the flip side, canonically stable emotions can be transient: anxiety can be stable in a circumstance where it is heightened by moments of fear, and Othello’s jealousy is tinged with love. Generalizing, any emotion can play the focal role of a stable emotion, in some circumstance, and any emotion can play the ancillary role of transient emotion, in some circumstance.

In part, Bhoja is shifting the burden of proof. So long as stability and transience are relativized to circumstances, those wishing to defend a limited menu of canonical rasas make good on two claims. Firstly, for each canonically transitory emotion, there is no diegetic circumstance where it can play a focal role. Secondly, for each canonically stable emotion, there is no diegetic circumstance where its role is ancillary. That is a very tall order. Delivering on it might seem possible only under the sway of rather rigid genre conventions that restrict the allowable diegetic circumstances.

More importantly, though, Bhoja is evoking a theory of mind, as the final clause in the passage quoted above hints (“because they all derive from the mind”). A limited menu of canonical stable emotions would suggest a mental architecture that Bhoja rejects. According to Bhoja, emotions are responses to circumstances, and the correct cognitive architecture is not one that rules out a focal role for some emotions and mandates a transient role for the rest, no matter what the circumstance. The discussion that immediately follows the argument against the limited menu of canonical rasas begins to articulate Bhoja’s conception of the architecture of the mind.

Note that we seem at the present juncture to be far from fundamentality. Either there are no rasas, because each would correspond to a canonical stable emotion (and there are no canonical stable emotions), or there are countless rasas, each corresponding to an emotion that is focal or stable in some diegetic circumstance. How, then, can there be exactly one rasa, as fundamentality states? How can we square the commitment to fundamentality in kārikā 6 with the rejection of limited menus of canonical rasas in kārikā 7 (2016b: 119–120)?

Bhoja draws a distinction. On the one hand, there is a single fundamental rasa; on the other hand, there are the countless particularized rasas, each corresponding to an emotion that is focal or stable in some diegetic circumstance. Call the fundamental rasa “f-rasa.” According to kārikā 10, f-rasa is not a character’s occurrent state of mind: it “transcends the plane of production [of feeling]” (2016b: 120). By contrast, stable emotions “are not themselves rasa but rather precisely emotions,” and “the name rasa is applied to them in a secondary sense” (Bhoja 2016b: 125). Let an m-rasa be any emotion that meets two conditions in a given diegetic circumstance: first, it is focal or stable in the circumstance; second, it indirectly expresses or manifests f-rasa. As Bhoja explains, emotions “become rasa [i.e., m-rasa] from their conjunction with rasa [i.e., f-rasa]; in other words, they come to possess a property common to rasa itself [i.e., f-rasa]” (2016b: 137). In this way, f-rasa is “brought to manifestation” (abhivyajyate) by means of m-rasa (Bhoja 2016b: 126; see also 120, 125, 126, 135, 137 138–139).

The picture is this. F-rasa is not an occurrent state that arises in response to worldly circumstances; it is not an emotion on the standard model that states the factors responsible for emotional response. Rather, it is fundamental in the sense that it grounds experiences elicited by those factors: it grounds experiences of m-rasa. As a result, f-rasa is expressed only through expressions of m-rasa. The m-rasas directly express circumstantially stable emotions, and they indirectly express f-rasa. Celeste expresses f-rasa by expressing fear if and only if her expression of fear is the fearful rasa. A character expresses a stable emotion that is not an m-rasa when the emotion is not an indirect expression of f-rasa.

By distinguishing f-rasa from m-rasa, Bhoja squares the commitment to fundamentality in kārikā 6 with the rejection of limited menus of canonical rasas in kārikā 7. Strictly speaking, there is one fundamental rasa, f-rasa, just as (r) and kārikā 6 state. At the same time, there are countless m-rasas, as we learn from kārikā 7 and the argument against canonical menus of rasas.

For this move to succeed, a case needs to be made for a mental architecture where experiences, such as the fearful m-rasa, are explained by appeal to the factors in the standard model in conjunction with an appeal to a more fundamental f-rasa.

In a powerful stretch of close reasoning, Bhoja makes the case by arguing that f-rasa is not reducible to stable emotion (2016b: 126). Here is a reconstruction of the argument. If f-rasa is reducible to a stable emotion, then each m-rasa is also reducible to a stable emotion. Moreover, any character can express any stable emotion—nothing prevents an author from having a character express fear, anger, love, joy, and the like. So, if each m-rasa is reducible to a stable emotion, then any character can express any m-rasa— nothing prevents an author from having a character express any m-rasa. However, we observe that while any character can express any stable emotion, only some characters can express m-rasa. (More on this claim below.) Therefore, the m-rasas are not reducible to stable emotions. It follows in turn that f-rasa is not reducible to a stable emotion. With this result, the question arises what explains the expression of m-rasa. The explanans might be a “common” feature of mind that grounds affective experience in any creature, or it might be a “special capacity” of mind found in only some creatures. If the explanans is a common feature, then any character can express m-rasa. Since only some characters can express m-rasa, we conclude that the explanans is a special capacity. Call it “f-rasa.” Experiences of m-rasa cannot be explained by appeal to the standard model alone. Appeal must also be made to a more fundamental special affective capacity, namely f-rasa.

Since fundamentality is an iconoclastic claim, one would expect Bhoja to try to head off objections from orthodox theorists. They might question, in particular, the observation that only some characters express m-rasa, so that the capacity to express m-rasa is variable. In the paragraph immediately following the stretch of close reasoning, Bhoja cleverly defends the observation by invoking a widely accepted categorization of types of characters (2016b: 126–127; see also 134–135, 139 and 2016a: 117). Rasa theorists accepted that protagonists, antagonists, and bit characters differ in their expression of emotion. Bhoja adapts the idea. Protagonists express f-rasa by expressing an m-rasa that befits the foundational factor—Rāma’s love for Sītā, for example. Antagonists express a “semblance” of f-rasa (f-rasābhāsa) by expressing a “semblance” of m-rasa (m-rasābhāsa) that does not befit the foundational factor—Rāvaṇa’s misplaced love for Sītā. Bit characters do not express f-rasa, m-rasa, or their semblances: they express ordinary emotion as needed only to advance and spice up the story.

With this, Bhoja concludes the argument in Chapter 11 for fundamentality, the claim that there is one fundamental rasa.

One ingredient of (r) remains, namely the claim that f-rasa is love (śṛṅgāra or prema). Bhoja recruits additional resources from his theory of mind in order to answer the question, what is love, if f-rasa counts as love?

In answering this question, Bhoja is at no loss for words, but what he says seems to come out of nowhere. For guidance, we can look to three points of interpretive context. Two come from the argument for fundamentality. First, love must be a piece of mental architecture that varies from character to character; it is a “special capacity.” It is not a necessary basis for having any experience at all, hence not a psychological invariant. Second, the mental architecture must explain why love is manifest in the expression of myriad m-rasas. A third point of interpretive context is obvious. The conception of love that figures in (r) must deserve its name. That is, f-rasa must informatively resemble erotic or romantic love. Recognizing this, Bhoja favors examples where the m-rasa is erotic or romantic love—even though f-rasa can be expressed through any m-rasa. In sum, we should read Bhoja as appealing to a phenomenon that (1) varies interpersonally, (2) finds expression in any m-rasa, and (3) echoes the nature of romantic or erotic love.

The Necklace opens with a striking declaration:

A strict four-way identity, f-rasa = love = ego = sense of self, is rather too strong, however. The Light adds nuance, notably in kārikās 3, 4, 8, and 10 (2016b: 119–120). To begin with, f-rasa is the expression of love by artistic means. Love turns out to be the more general psychological phenomenon of which f-rasa is a special case. Moreover, neither love nor f-rasa is simply ego (ahaṅkāra) or simply sense of self (abhimāna, mānamaya). Each is a special quality or capacity of ego and sense of self, where ego is not the same as sense of self.

Standard translations of “ahaṅkāra” as “ego” leave out what is evoked by a more literal translation as “I-making” or “self-making.” In the tradition to which Bhoja belongs, ahaṅkāra is said to be a creative cry of “I” (Van Buitenen 1957: 19). So, love or f-rasa is a special capacity for self-making.

Self-making is not the same as a sense of self, which consists in an awareness of oneself as a distinct personality with unique traits (Bhoja 2016a: 116, Bhoja 2016b: 119–120, 123). However, the two aspects of self are mutually reciprocal. One makes oneself by having experiences through which one is aware of oneself. So, love or f-rasa is a special capacity for self-making and special quality of self-awareness.

Kārikā 4 of the Light identifies the special capacity and quality that constitutes love: love is a preponderance of sattva (see also Bhoja 2016a: 117, Bhoja 2016b: 123, 125, 128). When sattva dominates, the process of self-making is love, and the products of sattva supply content to loving self-awareness. For Bhoja, f-rasa is an intensely savored love, hence intense self-making and an intense sense of self (2016b: 120, 138). Sattva is a special capacity or quality that provides for intense self-making and self-awareness.

What, then, is sattva? Here we can leverage the three points of interpretive context sketched above. We ask how sattva figures in a mental architecture on which love (1) varies interpersonally, (2) finds expression in any m-rasa, and (3) echoes the nature of romantic or erotic love.

Start with (1). Bhoja outs himself as a member of the sāṃkhya tradition, albeit not, he admits, “one hundred percent” (2016b: 122). The admission invites readers to import the sāṃkhya conception of sattva, which he explicitly adopts in kārikās 51 to 54 of his Light on Fundamental Principles (1971: 288–290; see also Pollock 1998: n. 48). In sāṃkhya, sattva is one of three guṇas. The guṇas are tendencies of mind that jointly explain interpersonal variation in psychological traits (Burley 2007: 101–103). Sattva is reflection, sensitivity, attention, pleasure. Rajas is activity, striving, pain. Tamas is stability, conformity, indifference. All three guṇas are a matter of degree, each more or less present in every person. Without a mixture of all three, there can be no personality at all, and the relative influence of each determines the tenor of one’s personality. Someone in whom rajas dominates, an ahaṅkāra rājasa, is above all enterprising and transformative. The ahaṅkāra tāmasa is mostly conformist: they tend to go with the flow. The ahaṅkāra sāttvika, whose mode of self-making is predominantly sensitive and attentive, enjoys richer sense of themselves as an individual than do those who go with the flow and those who rush about, busily transforming the world. To the extent that it involves sattva, which is a matter of degree and varies from person to person, the capacity for love meets requirement (1).

Recall how Bhoja thinks about variation in f-rasa amongst characters. A protagonist expresses f-rasa by expressing m-rasa; bit characters do not. They express ordinary emotions. The explanation for this variation will be that only in protagonists does sattva dominate the other guṇas. Bit characters are busy about such practical business as moves the plot along, or they contribute to the action by conforming to expectations. In them, rajas or tamas dominates. What about antagonists? Rāvaṇa’s misplaced love for Sītā is not genuine romantic m-rasa; it is a semblance of m-rasa. The reason is that the foundational factor of Rāvaṇa’s feeling is a fantasy of Sītā, which we can now chalk up to a shortage of sattva. By expressing a semblance of m-rasa, Rāvaṇa expresses a semblance of f-rasa, f-rasābhāsa. He experiences himself as an individual, but the experience is distorted, for he cannot own up to his fantasy of Sītā.

Variations in sattva among protagonists (who express f-rasa), antagonists (who express f-rasābhāsa), and bit characters (who express no rasa at all) suggest that sattva is bi-dimensional. Measured one way, sattva is acuity; measured another way, it is accuracy. Protagonists attend more sensitively to small differences: their thought is more fine grained. At the same time, they tend to get things right. Antagonists also attend to small differences, but they routinely fail to get things right. Bit characters, in whom rajas or tamas dominates, get along with generalizations, accurate or not.

We asked how sattva figures in a mental architecture on which love (1) varies interpersonally, (2) finds expression in any m-rasa, and (3) echoes the nature of romantic or erotic love. Addressing (1) has shed little light on why sattva provides for richer self-creation and self-awareness. For that, turn to (3). Love and f-rasa must echo the nature of romantic or erotic love.

Sensitive attention contrasts with conformity to accepted ways of seeing and with engagement in practical affairs. A character whose personality reflects a predominance of sattva accurately attends to worldly specificities, refusing to conform to gross categorizations, taking a stance that is more reflective than practical. Bhoja takes love to be “in essence, awareness.” (2016b: 123). Is it a stretch to liken such a personality with that of a lover? A lover attends to the beloved. She does not see him as this or that kind of person; she will not surrender to gross generalizations. Nor does she size him up as someone upon or through whom she can act to accomplish her goals. Instead, she attends to him accurately and acutely, getting him right in all his particularity (e.g., Bagley 2015).

As every love song attests, a lover’s attention does not leave her untouched. In sāṃkhya, to be a creature with any traits of personality at all is just to experience the world. Self-making is coming to experience the world, and a sense of self is an awareness of that experience as one’s own. A lover, being sāttvika, experiences the world in its particularity, has particularized experiences, and thereby acquires an individualized psychology, of which she is aware. She is aware of herself as the unique being who sees him in all his specificity.

If this is right, then f-rasa is the artistic expression of love because a character expresses love through an awareness of herself as attending to the world and its inhabitants in all their particularity, not as instances of generalities or as opportunities for action. For her to be aware of herself in this way is for her to be aware of herself as a unique individual who sees the world in that way. The proposal is that f-rasa is the artistic expression of love because it is a configuration of the self dominated by sensitive attention to particularity, and love involves, above all, a sensitive attention to particularity.

Finally, to (2). We must explain, by appeal to sattva, why the expression of any m-rasa expresses f-rasa and hence love. Two questions demand answers.

First, why must f-rasa be expressed through myriad m-rasas? The answer is that, when sattva predominates, self-making tunes into worldly particularity. Even the subtlest differences in the world make a difference to self-making and the sense of self—that is, to the self as it experiences itself representing the world. Moreover, in sāṃkhya, the world is taken to be experienced not only through the five senses but also affectively (Larson 1979: 200). Each circumstance is one to perceive and one to react to in feeling. As a result, a sense of self is at least partly a sense of one’s affectively representing worldly particularity. A character’s sense of self, when shaped by acutely sensitive attention to particularity, includes experiences that express the myriad m-rasas. F-rasa is expressed through the myriad m-rasas because it is artistically expressed love and love is a configuration of the self dominated by acutely sensitive attention to particularity. (It turns out that Bhoja’s appeal to love in (r) implies fundamentality.)

Second, why must f-rasa must be expressed through negative or unpleasant m-rasas? Traditionally, these take up half the list of eight canonical rasas—sorrow, anger, fear, and disgust. Bhoja spurns the canon, but he embraces the claim that f-rasa or love is expressed “even in the presence of pain and the like” (2016b: 125). He observes that “people are said to ‘love sex’” and also to “‘love quarreling’ or anger or joking” (2016b: 124). Love is not all sweetness and light: it can be expressed through anger and quarreling. Readers nowadays cringe at his references to pain administered in love-making; however, he is not being kinky. He well knew that pain and suffering are everywhere—even, he hints, in joking. A conception of love as expressed at the dark end of the affective register needs theoretical vindication.

Consider a range of affective responses to the cycle of violence and suffering in which Israelis and Palestinians are locked. The ahaṅkāra tāmasa is liable to bring what is happening under a generalization (e.g., “hopeless”) on which they may carry on as before. The ahaṅkāra rājasa more likely perceives a call to action. The ahaṅkāra sāttvika is prone to take in the suffering in all its particularity, as one of many ways the world can be. Their affective response is one of love just because it is one of acutely and accurately sensitive attention.

In sāṃkhya, one is what one experiences. As result, personal individuality is a function of worldly variety, so that to screen out suffering is to place a limit on individuality. The point is not overly self-regarding. The challenge of love is to fold everything into its embrace, in the sense that it must see everything for what it is, in its particularity. In a modest and more limited way, this is a feature of romantic love too.

Love implicates a special capacity and quality of the self epitomized by the ahaṅkāra sāttvika, in whom acutely and accurately sensitive attention to particularity dominates. Substituting in the reductive account of love, (r) now reads as follows:

The m-rasas are artistic expressions of occurrent, world-directed emotional states grounded in sensitively attentive (acute and accurate) self-making and self-awareness.

Bhoja’s conception of love implies fundamentality, and the Light states an argument for fundamentality, but the case remains to be made for (r), especially the identification of f-rasa with love. The case for (r) is that it explains (s) and thereby answers the normative question. Why is the fact that The Royal Tenenbaums is thick with rasa a reason for Sonam to watch it? The answer is (s), rasa (that is, f-rasa) is a route to his self-realization. Caution is needed here. To answer the normative question, (s) must be a reason for Sonam to watch the Tenenbaums. Presumably, the movie serves as a route to Sonam’s self-realization because it depicts a character as expressing f-rasa.

Not all characters qualify: most are bit roles. However, the protagonist, Royal, expresses f-rasa by expressing various m-rasas, which shift from narcissistic self-love to shame to paternal love, as circumstances change and he matures. When the curtain rises, he is a man of action who uses a cartoonish image of his family to rationalize his putting them out of mind as he pursues his business ventures. When these fail and he faces homelessness, he tries to reconnect with them, hoping to maneuver them into supporting him. They repel his advances, however. As the plot develops, he comes to see them more clearly. Once the credits are ready to roll, rajas and tamas no longer dominate: Royal attends with far more sensitive attention to the details of the lives of those he has come to care about. He is no longer a kind of clown, a caricature of an individual; he is a person with a distinctive character all his own. Moreover, in becoming that person, he helps his family members to see and appreciate themselves for who they are. They also benefit from an uptick in sattva.

(s) must say that Royal’s expressing f-rasa is a route to Sonam’s self-realization. Since Sonam is no literary character, he cannot express f-rasa. Let “self-realization” denote the general psychological process (one not confined to art) wherein sensitively attentive self-making yields a particularized sense of self. Royal gains in self-realization through his expression of f-rasa, and the proposal is that so does Sonam. That is, Sonam gains in self-realization through Royal’s expression of f-rasa. He follows a route to self-realization that somehow detours through the expression of f-rasa by Royal. How might that look?

An obvious answer is that Sonam identifies with Royal. Royal expresses f-rasa, and Sonam exercises sensitive attention by seeing through Royal’s eyes. Since Royal expresses f-rasa by expressing various m-rasas, Sonam feels the emotions that are their real-world analogues (feeling pride instead of the proud m-rasa).

This style of answer cannot work for Bhoja. Here is a problem that animated some tenth-century rasa theorists (unbeknownst to Bhoja—Pollock 1998: 134–136). The stock examples are portrayals of Rāma’s love for Sītā. Someone who identifies with Rāma cannot feel romantic attachment to Sītā. Sītā is a goddess, after all, and partner to Rāma, a god! Some theorists therefore proposed that characters are not determinate individuals; they are generalized types. Someone watching a story about Rāma and Sītā identifies not with Rāma but with lovers-in-general and imagines having feelings not about Sītā but about beloveds-in-general. Bhoja must reject this proposal. If some characters are portrayed as distinct individuals, as (r) entails, and if audience members identify with those characters, then audience members have highly inappropriate feelings. So, if audience members do not have the highly inappropriate feelings and yet they identify with the characters, then the characters are not portrayed as distinct individuals: (r) is false. Bhoja must reply, in defense of (r), that audience members do not have the highly inappropriate feelings, so they do not identify with characters.

Recall Celeste, who is set upon and killed by predators in The Auspicious Flower. She expresses the fearful m-rasa, and audiences respond strongly. However, nothing in (r) or (s) requires that their response be one of fear. Indeed, in the play, Bhīma has witnessed Celeste’s ordeal, and the scene concludes with him naming a rasa. He pronounces the scene “wondrous” (adbhuta). We are not Bhīma, of course. We are more likely to feel sadness, pity, and despair. Still, we do not feel fear.

Audiences can and do respond emotionally when reading poems and watching plays, but nothing in (r) or (s) implies that they feel the emotion that corresponds to the m-rasa that a character expresses. All that is required is a rational connection between the audience’s response and the m-rasa that the character expresses. Celeste expresses the fearful m-rasa, and those who witness fear have reason to feel pity in response, so audiences who witness the fearful m-rasa have reason to feel pity in response. Often enough, audiences do in fact respond as they have reason to respond. Likewise, Royal expresses the proud m-rasa. His pride has a foundational factor, a circumstance, which he comes to apprehend through sensitive attention. Suppose that he comes to see his children as decent people bravely confronting their problems. Since Royal’s children are not Sonam’s, Sonam is not warranted in feeling pride. Instead, he has reason to take satisfaction in their courage. The movie helps him to feel that by inducing him to see the children as having features that Royal sees them as having.

So much for m-rasa. How is Royal’s expression of f-rasa a route to self-realization for Sonam? In the Tenenbaums, Royal gradually acquires the capacity for sensitive attention that typifies love. He comes to appreciate others in his life in all their many particularities, not as an examples of this or that kind of person and not as creatures through which he can act to accomplish his goals. By regarding others as he does, he becomes more fully an individual: his mind inherits a newfound determinacy from the newfound determinacy of his apprehension of others and their circumstances. Watching the movie, Sonam also exercises and thereby acquires a capacity for sensitive attention: the movie entices him to follow along with Royal and to appreciate the characters in all their particularities. In so appreciating them, he becomes more fully an individual, for his mind inherits a newfound determinacy from the determinacy of his apprehension of (fictional) others and their (fictional) circumstances. Sonam does not express f-rasa, but f-rasa is a special (literary) case of the broader psychological phenomenon of sensitive attention, which plays a role in Sonam’s self-making and sense of self.

In short, (s) states an answer the normative question. Sonam should watch the movie because Royal’s expressing f-rasa is a route to his, Sonam’s, self-realization. Its being a route to his self-realization is a reason for him to watch it.

As a bonus, this answer to the normative question gives Bhoja a potential advantage over what is nowadays regarded as the dominant tradition in rasa theory. We saw that some solve the problem of audience responses to Rāma’s love for Sītā by construing characters as types or generalizations. Bhoja must reject this solution because (r) requires the distinct individuality of protagonists and antagonists. With this in mind, consider an extension of the normative question.

Sonam knows that Wes Anderson and Owen Wilson were riffing on Louis Malle’s movie and that Royal Tenenbaum is modeled on Alain Leroy. Eilidh can turn to Bhoja for an answer to Sonam: acute sensitivity to the smallest differences in personality and circumstances matters to the project of self-realization. That is why we do not stop at reading one poem or seeing one movie. By contrast, the dominant tradition, on which characters are types or generalizations, has its work cut out for it in explaining why Sonam should check out the Tenenbaums. The whole point of types or generalizations is that exposure to one instance gets you all you need to get.

So far, a case has been made that (r) and (s) state an answer to the normative question, but we saw in the opening section that Bhoja accepts the hedonic, moral, and political answers too. It would lend weight to (r) and (s) were they to imply and unify the hedonic, moral, and political answers to the normative question. Might (r) and (s) state the answer to the normative question?

Their implying the hedonic answer is straightforward. Self-realization is by definition a process that produces a particularized sense of self through sensitively attentive self-making. In sāṃkhya, sattva is a tendency to pleasure, and thought in which sattva dominates is pleasurable. There will be pleasure for Sonam in the process of self-realization that the Tenenbaums invites.

The moral and political answers to the normative question enrich Bhoja’s rasa theory (Pollock 2001: 208–211). At the same time, Bhoja explicitly prioritizes (r) and (s). He comments that “what we’ve striven to communicate” in the Light is that “passion [love] alone is the means of fulfilling the four ends of man” (2016b: 123). That is a strong claim. What resources can Bhoja draw upon in its defense?

Recall that f-rasa varies among characters. Protagonists express f-rasa by expressing an m-rasa that befits the foundational factor—Rāma’s love for Sītā, for example. Their sensitive attention is mostly acute and accurate. Antagonists express only a semblance of f-rasa by expressing a semblance of m-rasa that ill fits the foundational factor—Rāvaṇa’s misplaced love for (or obsession with) Sītā. Their sensitive attention is mostly acute but not mostly accurate. It is mostly a detailed fantasy. Yet, to act well, morally speaking, one must exercise sensitive attention so as to accurately assess the morally relevant features of situations. Rāma’s love for Sītā is seated in his seeing her as she is, but Rāvaṇa profoundly mistakes who she is, with the result that he profoundly mistakes himself too. Members of the audience need not identify with either character: they only need to exercise acutely and accurately sensitive attention in feeling some occurrent emotion—disgust or outrage, in response to Rāvaṇa. Here, disgust and outrage are moral emotions. Moreover, if Rāvaṇa is morally depraved, he is also a big time political disruptor. (s) implies that audiences sensitively attuned to political reality have reason watch the play, reacting to him as a political threat, thereby entrenching the affective side of their political sensibilities.

A defense of the strong claim can also draw from Bhoja’s deepest vision. The self-making (ahaṅkāra) that Bhoja promotes is precisely what the Buddhist philosophers advised us to vanquish. For them, the moral project is one of unselfing. For Bhoja, the moral project is one of selfing. However, Bhoja’s ideal moral agent is no egoist. That creature is, at best, the ahaṅkāra rājasa. The ideal ahaṅkāra sāttvika is a maximally determinate self because their experience accurately encompasses the world in all its nuance. Acutely and accurately sensitive attention is selfless in the sense that matters for morality.

Viewed at a distance and from certain angles, the main themes of Bhoja’s aesthetics resonate with those found in other traditions. For example, since sensitive attention (sattva) opposes practical cognition (rajas) and complacent expectation (tamas), the aesthetic zone is disinterested and free of rules, much as Kantians insist. Kantians also insist that all aesthetic response is grounded in a fundamental capacity (for the free play of imagination and understanding). The parallels are there to be drawn, but Bhoja’s aesthetics has an extraordinary flavor of its own. The fundamental aesthetic capacity is love, the impulse that makes us who we are in relation to others.