While it is well known that larger works contain a variety of rasas, a poet who seeks the excellence [of his works] will make just one of them predominant …

[But there is] no obstruction to a single rasa by its being mixed with others … Readers with a ready sense of discrimination, who are attentive and intelligent, will rather take a higher degree of pleasure in such a work.³⁴

To which Abhinavagupta adds:

If the rasa that has been taken in hand extends throughout the whole plot and is fitted for predominance by this extensiveness, its predominance will not be harmed by the introduction, by the filling in, of other rasas brought in by the needs of the plot and running through only limited sections of the narrative. Rather than being injured, the predominance of the rasa which appears as an abiding factor throughout the plot will be strengthened. In other words, the subsidiary rasas, although they attain a degree of charm by being fully developed each at its own stage by its own set of vibhāvas and the like, still do not attain such a charm that our apprehension will rest on them; rather, it will be carried on to some further delight.³⁵

A song in an Indian film is an interlude which exists outside of story-logic and story-time, but within the emotional palette of the film; its function is to provide subsidiary rasas that will strengthen the predominant rasa of the whole. The first Indian talkie, Alam Ara (The ornament of the world, 1931), featured seven songs. The newspaper advertisements touted an “all-Star-Cast Production” that was “All Talking/Singing/Dancing.”

This is why the Aristotelian unities of British and American films seemed so alien to me when I watched them as a child. But this emotional monotone was also — implicitly — modern and grownup, as opposed to the premodern and childish sentiment-mixing of our own movies. So, self-consciously serious filmmakers in India have tended to eschew the traditional forms beloved of commercial cinema, and have signaled their noble artistic and political intentions by hewing to conventions native to more “developed” countries.

Anandavardhana gives an example of mixing rasas from the Mahabharata; a wife searches for the body of her warrior-husband upon a bloody battlefield, and finds his severed arm:

This is the hand that took off my girdle,

that fondled my full breasts,

that caressed my navel, my thighs, my loins,

and loosened my skirt.³⁶

Here, the stable emotion of grief is made sharper and more profound by the tasted memory of the erotic. And this provides, for the reader, the savoring of karuna-rasa, pathos.

When I inflict butcheries on the characters in my fictions, I sometimes think about how strange it is that we can savor, even, the horror of battlefields on which entire races die. This is monstrous. We are monstrous.

But savoring is a form of knowledge. And what is most delicious on my palate is a medley of tastes that come together to reveal a dominant rasa. Longer works of fiction that insist on a monotony of emotion always seem awkward to me, incomplete, even if they are elegantly written. But it is not just that the sweet tastes sweetest when placed next to the salty. If savoring is a form of knowledge, then a complexity of affect affords the most to know. I am given pause, I linger, I relish, and I am brought to chamatkara—wonder, self-expansion, awe. When I love a book, a film, a poem, a sentence in a novel, when I am absolutely ravished by it, I always find that my delight is overdetermined, has “more determining factors than the minimum necessary” (as the OED puts it). Where does that “(final) feeling” come from — from the plot, the pace, the words themselves, all those fading memories of the peripheral characters, from the undertones of emotion that I hardly remember? From all of those, at once. From knowing all of those, together.

The theorists of rasa-dhvani gave me a way to think about writer, text, and sahrdaya. I also gained from them a way to think about literary convention — if in poetry “the savouring … arises like a magical flower, having its essence at that very moment, and not connected with earlier or later times,” and also “the feelings of delight, sorrow, etc., [produced by the representation] deep within our spirit have only one function, to vary it, and the representation’s function is to awaken them,” then the claims made for one particular set of conventions — often rather ambitiously called “realism”—are not only epistemically questionable, they are just irrelevant. There are many ways to manifest dhvani, I told my realist writer-friends. Choose the ones that work for you and your sahrdaya, and leave off with the proselytizing and pronouncements of your virtuous artistic rigor, of your deeper connection to what-really-is.

All this was satisfying enough, but dhvani—or at least resonance, reverberation — was crucial to the structure of the novel I was writing. The book’s shape followed what contemporary literary theorists call a “ring composition,” in which the ending of the narrative somehow joins up with the beginning, forming a circle. A ring composition is often used as a frame, within which further rings are embedded. Elements within a ring often reflect back on each other to form a chiastic structure, A-B-B-A, or A-B-C-B-A. Often, language or tropes or events are repeated, each time somehow changed. “In ring composition repetitions are markers of structure,” the anthropologist Mary Douglas writes.³⁷ Ring composition is a structure used all over the world, in narratives as varied as the Bible and mediaeval Chinese novels, she tells us, “so it is a worldwide method of writing.”³⁸ She adds that “ring composition is extremely difficult for Westerners to recognize. To me this is mysterious.”³⁹

In India, ring composition is a standard architecture, found prominently in epics like the Ramayana and the Mahabharata, in poems, in the Rig Veda, and in Panini’s grammar.⁴⁰ When I was writing my first book, I had never heard the phrase “ring composition,” but the method and its specific implications and techniques came readily to hand because — of course — I had seen and heard it everywhere. What I wanted within the nested circles or chakras of my novel was a mutual interaction between various elements in the structure. That is, for one chapter to act as the transformed reflection of another, for a nested story to act as an echo for another story nested within itself, and so on. Each of these connections would — I hoped — act as a vibration, a spanda, and all of them would come together in a reverberation, a dhvani—perhaps not quite in the sense that Anandavardhana used the word, but a dhvani nevertheless — a hum that would be alive and full and endless.

I didn’t exactly plan this architecture, sketch it all out before I began. I knew the general outline, and groped and felt my way into the specifics. I didn’t plan it because I couldn’t have; the unfolding of the story, all the stories, comes from the tension between intention and discovery. There are unbelievably delicious moments when you feel the pieces falling into place, when you find harmonies and felicities and symmetries that you can’t remember constructing, and at those times you cannot help becoming a mystic, believing that you are after all a little bit of a kavi, a seer of some sort.