Here the speaker is a woman, and what is being suggested is a vastu, a narrative element: the woman wants to keep the wandering monk away from a trysting place where she meets with her lover. So what we have here is vastu-dhvani, through which the poet can suggest things, facts, situations, prohibitions, injunctions. In this poem, the denotative meaning is exactly the opposite of what she really wants, what she is really doing; what the reader grasps is beyond abhidha and lakshana. Here is another famous example of vastu-dhvani:
Mother-in-law sleeps down there, and I here.
Look while day remains, O traveler.
Do not, blind in the night,
lie down in my bed.42
The presumed speaker is a lovelorn woman, a familiar type within the conventions; this married woman is extending—or suggesting—an invitation.
Dhvani can also suggest a figure of speech, an alamkara. So:
O lady with tremulous long eyes,
as your face
completely fills the directions
with the radiance of its beauty,
the ocean now remains calm,
absolutely still.
And so I know it is nothing
but an insentient mass of water.43
The ocean is not stirred now—as it was a moment ago, when the moon rose—therefore it must be truly insentient. The eyes are tremulous—perhaps from momentary jealousy?—but now the smiling face “fills the directions.” What is being suggested here is a metaphor, the beautiful face as the moon. This is the second variety of suggestion, alamkara-dhvani.
But the most powerful dhvani in poetry, the poet’s most desired effect, is the suggestion of rasa. Rasa is a term which, until Anandavardhana, had mostly been used in dramaturgical texts. At some time between 200 BCE and 200 CE, a perhaps-apocryphal sage named Bharata is said to have written the Natyashastra (Treatise on the drama). The Natyashastra is a theater professional’s handbook: it includes chapters on what a playhouse should look like; on different gaits; the use of local dialects; costumes and makeup; on the factors that lead to the success of a dramatic performance. The sixth and seventh chapters famously analyze the nature of aesthetic pleasure, rasa. According to Bharata:
People who eat prepared food mixed with different condiments and sauces, etc., if they are sensitive, enjoy the different tastes and then feel pleasure (or satisfaction); likewise, sensitive spectators, after enjoying the various emotions expressed by the actors through words, gestures and feelings feel pleasure, etc. This (final) feeling by the spectators is here explained as (various) rasa-s of natya [drama].44
So rasa—the word literally means “taste” or “juice”—is not emotion (bhava); it is the aestheticized satisfaction or “sentiment” of tasting artificially induced emotions. Generations of thinkers developed the notion of rasa along with a notion of the ideal viewer; the locus of rasa was this viewer, not the actors or the stage. Rasa is what the drama produces in the sahrdaya, the sophisticated “same-hearted” connoisseur who is the playwright’s necessary counterpart. The sahrdaya—because of education, experience, and temperament—is able to experience rasa precisely because he or she does not identify in a personal, egoistic way with the tragedy on the stage. The naive spectator who ascribes some sort of reality to what is happening on the stage and identifies personally with the emotions of the characters is incapable of rasa, which is an impersonal, disinterested pleasure. One might say that a certain psychical distance is necessary for rasa to be experienced. Rasa is sublime.
On stage, the characters and situation and the patterning of events make up the determinants or catalysts, the vibhavas; the actors portray the consequent outward manifestations (anubhavas) such as speech, bodily posture, and involuntary reactions (such as trembling). In response to the actors’ depictions of momentary situations like a waiting lover’s anticipation or doubt, the spectator experiences fleeting emotional states (vyabhicaribhavas); and all of these various feelings come together—like condiments and sauces—to allow the viewer access to a stable emotion, a dominant mood, a sthayibhava, such as grief. Note that this stable emotion is in the viewer; it is a “permanent emotional state” that is ever present in all human beings as a potential, a latent trace. The actors cannot act out a sthayibhava—it really doesn’t matter what the actors feel or don’t feel; the purpose of their craft is to allow the sahrdaya access to his or her own stable emotions.
And, very importantly, the stable emotion is not the rasa. If the actors portray a scene that allows access to the stable emotion of grief, shoka, what the viewer relishes is the rasa of karuna, pathos. The rasa is in the tasting of grief, in the relishing of grief, in the reflective cognizing of grief. If the actors portray desire, rati, the viewer relishes sringara, the rasa of the erotic. The pleasure of rasa comes from the meta-experience of experiencing oneself experience the stable emotions.
Bharata names eight rasas in all; so, in addition to karuna and sringara, the viewer enjoys hasya, the comic; raudra, the wrathful; vira, the heroic; bhayanaka, the terrible; bibhatsa, the disgusting; and adhbuta, the wonderful. A ninth rasa was added by later theoreticians: shanta, the peaceful, which arises from the stable emotion of vairagya, detachment, dispassion, and which is manifested especially by epics like the Mahabharata, which place the specificity of human striving and passion against the vastness of time.
According to Anandavardhana, poetic language can manifest dhvani through the operation of vyanjana, suggestion, and so offer the reader an opportunity to taste rasa. For instance:
While the divine sage was speaking
to Pārvatī’s father,
she, eyes downcast,
counted the petals
of her toy lotus.45
The “divine sage” has come to propose to Himavata, the king of the Himalayas, that he marry his young daughter Parvati to the god Shiva. Parvati is in love with Shiva, and has performed great austerities to win him, but in this moment, Anandavardhana says, “the counting of the petals of the lotus subordinates itself and without the help of any verbal operation reveals another matter in the form of a transient state of mind (vyabhicāribhāva) [of the emotion love, namely shyness].”46 In the perception of the reader, the shyness of Parvati suggests desire, love. The reader therefore relishes the rasa of the erotic, sringara, evoked through dhvani.
We can see that the operations of suggestion are heavily dependent on context (as in “The sun has set”). Whereas the denotative and connotative meanings of a phrase or a text are limited, vyanjana is infinite and can never be exhausted. Consider the following verse, in which the speaker is Rama, the hero of the Ramayana; the season is the monsoon, when lovers come together. Rama has lost his kingdom through court intrigue, has been banished to the wilderness, has faced many dangers, and his beloved wife has been kidnapped by the evil Ravana.
Clouds smear the sky,
flocks of cranes tremble
across their viscous blue-black beauty.
The winds sprinkle rain, the peacocks call
their soft cries of joy.
Let all this be, as it likes. I am Rāma.
I am hard-hearted, I can endure all.
But Vaidehī? How will she survive?
Ah alas, my goddess, be strong!47
Anandavardhana writes, “In this verse, the [suggestive] word is ‘Rama.’ By this word we understand Rama as developed into various suggested qualities, not simply as the possessor of the name.”48 That is, the reader already knows that the speaker is Rama; no information is added here by the use of the name. But the name evokes, for the reader, all the tragedy and suffering that this man has already experienced and will experience in the future. In a flash, this sudden explosion of light illuminates the past and the future.
A century later, the greatest exponent of rasa-dhvani theory, Abhinavagupta, wrote about this verse:
The suggestions of other properties … are endless; for example, his banishment from the kingdom, etc. And since these suggestions are coun
tless, they cannot be conveyed [simultaneously] by means of the denotative functions of words. Even if these innumerable suggested properties were to be conveyed [by denotation] one by one, since they will not be had in one single act of cognition, they will not be the source of a wondrous aesthetic experience and hence they will not give rise to a great beauty. But if these properties are suggested, they will assume countless forms (kiṃkiṃ. rūpaṃ na sahate) because in the suggestion their separateness will not be clearly perceived. In this way they will become the source of a strikingly beautiful aesthetic pleasure that is analogous to the flavor of a wonderful drink, or cake, or sweet confection [where the individual ingredients cannot be separately tasted but yet add to the flavor of the final product]. For it has been said already [by Anandavardhana] that a word which is suggestive reveals a beauty “which cannot be conveyed by another form of expression.”49
Rasa-dhvani can operate at the level of a word, a sentence, or an entire work. According to Anandavardhana, rasa is “an object on which no words can operate directly,” and therefore dhvani is the only way to manifest rasa.50 So Anandavardhana might have said to an aspiring writer: suggest, don’t tell. Dhvani is literally “reverberation,” and is often compared to the “sounding of a bell” or “a needle falling through a pile of lotus leaves.” If we hear the phrase, “A village on the Ganga” (in Sanskrit, gangayam ghoshah, literally “a village in the Ganga”) we understand that the village cannot literally rest on the water, and that we are talking about a village located on the banks of the Ganga. This would be an example of figurative speech. But Anandavardhana would argue that this phrase about the holy river also carries a suggestion—or causes in the listener the manifestation—of coolness, sanctity. This underlying, affective meaning does not emerge from the denotative or figurative aspects of the phrase. This kind of suggestion functions in all speech, is present even in mundane language, but poets knowingly and intentionally concentrate vyanjana to construct a coherent, sustained engagement within the reader, and thus to manifest dhvani, and hence rasa. Patrick Colm Hogan writes:
Rasadhvani, the “truest” form of dhvani, it is an experience—along the lines of what we would call “a moment of tenderness” or “a pang of sadness.” It is, in short, an experience of rasa … [These rasas] are evoked through the clouds of non-denumerable, non-substitutable, non-propositional suggestions which surround these texts.51
The very sounds and rhythms of language—which preexist meaning—contribute to our experience of rasa. Abhinavagupta says that when we hear poetic language
without waiting for our understanding of the expressed meaning, [the stylistic qualities] set about building up the rasas, giving us a foretaste (āsvāda) of them. This is as much to say that as the rasas are suggested by style (saṅghaṭanā), the ground is laid for the relishing of a rasa at the very beginning of the appropriate style before our understanding of the meaning has come into play; and that it is on this account that the rasa, even at the later moment, after we have understood the expressed meaning and when the rasa has assumed its full flavour, does not appear to have arisen later [than our understanding].52
Anandavardhana observes, “When ornamented by even one from among the varieties of dhvani, speech acquires a fresh colour, even though it follows a subject matter that has been treated by poets of the past.”53 Since the properties manifested by dhvani are countless, “poetical material … finds no limit …”
Not even Vacaspati [the god of speech] in a thousand efforts could exhaust it, any more than he could exhaust the nature of the universe.
For just as the nature of the universe, although it has manifested this marvellous proliferation of matter through the succession of past ages, cannot be said now to be worn out and unable to create anything new, just so is the situation in poetry, which, although it has been worked over by the minds of countless poets, is not thereby weakened, but increases with ever new artistic abilities.54
Anandavardhana accepted that there may be poetic texts in which the suggested meaning isn’t the dominant pleasure, or even present at all; he rather disdainfully refers to the latter as chitra kavya, picture poetry, flashy poetry: “Poetry which lacks rasa or an emotion (bhāva) as its final meaning, which is composed only by relying on novelties of literal sense and expression, and which gives the appearance of a picture, is citra …” Poetry that “gives the appearance of a picture” refers to very difficult pictorial arrangements in verse, similar to visual pattern poetry and topiary verses in the West—Sanskrit writers wrote stanzas in which interlocking syllables, if connected by drawn lines, revealed the shapes of drums, swords, wheels, and so on.55 This chitrabandha is probably where the more general term chitra kavya originates. Anandavardhana continues, “[This poetry also includes] verbal citra, such as difficult arrangements, yamakas (echo alliterations), and the like. Semantic citra … may be exemplified by poetic fancy (utprekṣā) and such figures … It is not real poetry, for it is an imitation of poetry.”56
Now that Anandavardhana has shown us how vyanjana works in poetry to produce dhvani and rasa, he tells us, “Now that instruction is being offered to modern poets in the true principles of poetry, while citra may be much used in the efforts of beginners who are seeking practise, it is established for mature poets that dhvani alone is poetry.”57
So rasa is what I felt that afternoon I discovered Hemingway at our kitchen table in Bombay, when the bleak undertow of his stories, roiling with unspoken emotion, flung me into an exaltation, a state of delight. Hemingway’s famous taut rhythms, the stripped simplicity of his diction, those repetitions of sound that he meticulously builds into his prose, all these enhance the iceberg-sized dhvani of what he leaves unsaid. Every word, every pause, every hesitation makes the dhvani of a story.
Flannery O’Connor writes:
The meaning of a story has to be embodied in it, has to be made concrete in it. A story is a way to say something that can’t be said any other way, and it takes every word in the story to say what the meaning is. You tell a story because a statement would be inadequate. When anybody asks what a story is about, the only proper thing is to tell him to read the story. The meaning of fiction is not abstract meaning but experienced meaning.58
The only way to explain to you what I experienced when I first read Hemingway is to tell you to read those stories. And even then, you will read different stories. We may read the same texts, but the dhvani that manifests within you will be unique. Your beauty will be your own. If you reread a story that you read ten years ago, its dhvani within you will be new. Poetry’s beauty is infinite.
6 THE BEAUTY OF CODE
This is what ugly code looks like:
This is a dependency diagram—a graphic representation of interdependence or coupling (the black lines) between software components (the gray dots) within a program. A high degree of interdependence means that changing one component inside the program could lead to cascading changes in all the other connected components, and in turn to changes in their dependencies, and so on. Programs with this kind of structure are brittle, and hard to understand and fix. This dependency program was submitted anonymously to TheDailyWTF.com, where working programmers share “Curious Perversions in Information Technology” they find as they work. The exhibits at TheDailyWTF are often embodiments of stupidity, of miasmic dumbness perpetrated by the squadrons of sub-Mort programmers putting together the software that runs businesses across the globe. But, as often, high-flying “enterprise architects” and consultants put together systems that produce dependency diagrams that look like this renowned TheDailyWTF exhibit. A user commented, “I found something just like that blocking the drain once.”
If the knot of tangled hair in figure 6.1 provokes disgust, what kind of code garners admiration? In the anthology Beautiful Code, the contribution from the creator of the popular programming language Ruby, Yukihiro “Matz” Matsumoto, is an essay titled “Treating Code as an Essay.” Matz writes:
Judging the a
ttributes of computer code is not simply a matter of aesthetics. Instead, computer programs are judged according to how well they execute their intended tasks. In other words, “beautiful code” is not an abstract virtue that exists independent of its programmers’ efforts. Rather, beautiful code is really meant to help the programmer be happy and productive. This is the metric I use to evaluate the beauty of a program.1
He goes on to list the virtues of good code: brevity; reusability (“never write the same thing twice”); familiarity (Ruby is an “extremely conservative programming language” that does not use “innovative control structures” but “sticks to traditional control structures programmers are familiar with”); simplicity; flexibility (which Matz defines as “freedom from enforcement of tools,” so programmers aren’t forced to work in a certain way by the tools or languages they use); and, finally, balance: “No element by itself will ensure a beautiful program. When balanced together and kept in mind from the very beginning, each element will work harmoniously with the others to create beautiful code.”2
So, beautiful code is lucid, it is easy to read and understand; its organization, its shape, its architecture reveals intent as much as its declarative syntax does. Each small part is coherent, singular in its purpose, and although all these small sections fit together like the pieces of a complex mosaic, they come apart easily when one element needs to be changed or replaced. All this leads to the happiness of the programmer, who must understand it, change it, extend it. This longing for architectural coherence leads to comparisons of code with music, which is often described as the most mathematical of the arts. There is, in fact, an anecdotal but fairly generalized belief among American programmers that there is a high correlation between coding and music-making, that many coders are musicians. A similar claim is made about mathematicians and music. These connections seem culturally encoded to me, specific to America—I’ve never heard of Indian programmers or mathematicians having a special affinity for music, apart from some being passionate listeners. Still, the code-and-music analogy is illuminating in that both practices prize harmonious pattern-making and abhor cacophony, a loss of clarity and structure. The snarl in the dependency diagram (figure 6.1) may strike the civilian as a pretty picture, with its swirl of lines and punctuating sparks of gray; to the programmer, it is an abomination because it speaks of incoherence, incomprehensibility, unpredictability, sticky seams of connection that prevent swift diagnosis and make excision and replacement all but impossible.