Dancing About Architecture

DISCUSSED: Snobbish Unattributed Truisms, Dorm Freshmen circa the Mid-1980s, Sheet Music, Linda McCartney’s Grotesque Backup Vocals, Personal Associations with the Word “Carrot,” Goering’s Listening Habits, Pumps for Inflating the Soul, A Girl Named Esther, Seascapes and Moonlight, Milan Kundera’s Bach Fantasy

Dancing About Architecture

Arthur Phillips
Facebook icon Share via Facebook Twitter icon Share via Twitter


I just published a novel about music. Early in the process of writing it, I was warned by a similarly music-obsessive friend that “writing about music is like dancing about architecture.”[1] Since that first somewhat menacing reminder, I’ve heard the line frequently.

At first blush, the claim is a smugly dismissive one: verbal descriptions of music are doomed to be pointlessly, perhaps even ridiculously, inferior to actual music. As a reader, I resisted this idea; it just felt false, though I couldn’t quite say why. But as a writer, this assertion paralyzed me: I didn’t want to waste two or three years trying to produce something that could not be produced.[2] I tried to put aside the line’s foundational snobbery (“My music is too ineffable for your inky art”), and then, reassuringly, it seemed like nothing more than a truism: words are words and music is music. And perfume is perfume; paintings are paintings; facial features are facial features. Yet writers are never counseled against attempting to evoke paintings or smells or faces or feelings or buildings or the nonmelodic sounds of jackhammers, thunder, or snoring. What was so elusive about music that it couldn’t be captured by words?


I decided to consult my bookshelf to see how those who came before me had tackled the problem of writing about music.

The most common—and least risky—means of injecting music into fiction is to treat the music as a prop, an inaudible ornament, in order to set a scene, establish a social milieu, earn a little plausibility. For example: “That year, all the freshmen in my dorm owned Bowie’s Changes One album and Squeeze’s Singles: 45’s and Under.” Or: “While she peeled potatoes and waited to hear news of the attack at Normandy, the radio played ‘Don’t Sit under the Apple Tree.’” Or: “She went out dancing that night and met him at a new club; it was the kind of place where Norwegian DJs sampled Astrud Gilberto and Messiaen over Ibiza beats.”

Despite hand-wringing that contemporary fiction is brand obsessed, this trick is hardly a recent innovation. In Emma, Jane Austen has a character dropping sheet music onto the piano (the brand of which is duly noted) and mentioning John Baptist Cramer by name, because you just know what kind of girls listened to him.

And that, of course, is the price of this sort of pop-referencing: it dates the book. At first the reference dates the story efficiently (if all the frosh have Singles: 45’s and Under, if they own and play full albums at all, it must be the mid-1980s), but later the reference dates the book cruelly: who the hell is John Baptist Cramer, and what, if anything, does it mean that Mr. Churchill suggests playing him? Author, what your Amy Winehouse joke gains you in twenty-first-century nods of recognition, you will pay for in twenty-second-century footnotes.[3] And, at its worst, the reference implies a nervous author, grubbing for street cred. An unnecessary but mildly recherché mention of a band at a party (“Elvis Costello’s ‘Alison’ came on”) can seem as if the writer just wants us to know we are in cool hands.

At its best, the technique can produce a hint of atmosphere. But the author is betting on a shared musical vocabulary, an occasionally dangerous wager. Some readers will have that song and its cultural meaning ready at hand. Others will be willing to go look up the reference. Others will turn the page, deaf to the implications. And some, feeling excluded, will close the book forever. In Infinite Jest, David Foster Wallace both utilizes and parodies the use of music references during a scene where criminals torture their victim to a specially prepared CD of Wings, engineered to play only Linda McCartney’s grotesque backup vocals. This continues to be my favorite passage in the novel, but I could easily have been one of the masses of one-star Amazon grumblers, certain the author was solipsistically interested only in his own amusement.

Specific music references can also be used to establish character. This is a detective-fiction standby: “He pulled his bourbon from the bottom desk drawer and put Coleman Hawkins on the record player.” If you happen to know the music, you gain an additional sense of the character’s mood and personality. Even if you don’t, you can read into the name Coleman Hawkins character traits such as “bluesy” and “unpretentious.” Also the fact that the character listens to music alone might imply to you that your gumshoe is of the heart-bruised, heartsick variety. (It’s a trick, of course, if this detail makes you feel any sort of sympathy for the character: Goering liked to listen to music alone, too.)

That variability of meaning is an essential point. A perfectly shared vocabulary between writer and reader does not exist, which is one of the great wonders of reading: a book leads a thousand different lives in the heads of a thousand different readers. I write carrot; one reader smells their beloved mother’s cooking; the next reader gags. Musical references are even trickier, because even two different readers who love the same music will likely have different responses to the same reference. And it is worth pointing out that to some people—some very intelligent people, like Vladimir Nabokov—music is all unintelligible noise.


Early in my search for support for the idea that writing about music was not a waste of time, I turned to a favorite writer whom I remembered for his extensive descriptions of music. But I misremembered, in a way.

Milan Kundera does not assume shared musical knowledge with his reader, does not rely on descriptions of music to produce desired reactions, does not allow the reader to extrapolate a character based on theme music, and tightens, as best he can, all interpretive loopholes so that the reader is associating with the music reference only what Kundera wants him to associate with it. Contrary to my memory, Kundera does not try to make you hear music. He uses music extensively, but the music arrives fully interpreted for the reader, almost silenced; the character is not yours to help create, because the music is not yours to feel:

Laura loved music sincerely and deeply; I recognize the precise significance of her love for Mahler. Mahler is the last great European composer who still appeals, naïvely and directly, to homo sentimentalis.

And yet, as this passage illustrates, it’s impossible to close all interpretive loopholes. If you should happen to disagree with Kundera’s interpretation of Mahler, then the book begins to blur and tremble and possibly disintegrate in your hands. If you do not believe Mahler is naïve and sentimental, then you are left arguing with a presumably infallible narrator-author (“I” has no other role in the novel; it’s just Kundera) over the nature of his character (whom he has explicitly invented in the narration to serve his purposes in telling you a story). He seems to want a reader who knows Mahler’s work, but is willing to surrender all past subjective reactions to it. While you are a visitor in Kundera’s kingdom, Mahler’s music will be played but mediated through the king’s thoughts.

Kundera, a trained musician, employs music coldly, intellectually (almost resentfully, I might say), with no expectation or desire that you will feel something (something cheap?) by having the illusion of hearing music in his work. That said, music and his love for it are very closely tied to his prose. He includes bits of scores on the page; he has said that he structures his novels to mirror classical music forms; he arranges chapter length to simulate musical tempo changes.

In The Unbearable Lightness of Being, he places music in his characters’ heads and mouths:

Tomas shrugged his shoulders and said, “Es muss sein. Es muss sein.”…. This allusion to Beethoven was actually Tomas’s first step back to Tereza.

That Beethoven score is written out for you on the page, with the lyrics; you are welcome to “hear” the allusion now, but only as Kundera’s illustration: accept the role music plays in the interior life of these people.

Kundera also uses music to comment on theme and to buttress his philosophies. For example, he returns often to the idea of music as a narcotic, a religion substitute, as here, in Immortality:

Music: a pump for inflating the soul. Hypertrophic souls turned into huge balloons rise to the ceiling of the concert hall and jostle each other in unbelievable congestion.

He also uses it as a multi-surfaced reflector to describe different characters’ contradictory views in the same musical moment. Franz, a music lover, asks Sabina:

“Don’t you like music?” […]

“No,” said Sabina, and then added, “though in a different era…” She was thinking of the days of Johann Sebastian Bach, when music was like a rose blooming on a boundless snow-covered plain of silence.

Once music was overwhelming, Sabina reminds us, because of its rarity and evanescence. Here it isn’t music that’s being evoked, but the feelings evoked by its absence.

Kundera writes about music’s importance, its philosophic implications, but doesn’t try to put the sound into words. He writes musically and seems to want his writing to move and affect and challenge the reader as music does, but he doesn’t want music to do it. For him, literature about music is something entirely different than music. It seems to be something like, um, dancing about architecture.


I found other writers, though, who use music with more expectation that a reader might experience, via a personal association, the sensation of hearing it.

In William Gaddis’s The Recognitions, Esther is surprised to see her long-estranged husband arrive uninvited at a party:

The music was The Great Elopement: a chill horn raised her, twisted her up and exalted her for a moment; and then she was let go, and lowered evenly on strings. —He’s here, Esther said, her flush already failing. —He’s come… here.

I don’t know The Great Elopement. But the words chill horn had a potent and instantaneous impact on me. Branford Marsalis’s saxophone solo in the Sting song “I Was Brought to My Senses” has a very specific physical effect on me, repeatedly, something akin to an actual chill. As I read this passage the first time (and as I re-read it now), the words chill horn reminded me of Marsalis’s horn, and by the time I reached raised her, I was recalling my physical reaction to that different piece of music. The images of raising up, twisting up, exalting, and lowering apply to me, evoke in me a specific sense-memory, and attach my memories to this character, Esther, as if, for a moment, I felt Esther’s reaction to her piece of music at the party. This particular writing about music produced in me an instantaneous emotion—provoked by music—which I then laid over this fictional construction called “Esther.”

Of course, I “heard” some other piece of music entirely. If, by chance, I had known The Great Elopement, and had similar associations to it, I imagine that I would feel even closer to Esther, would create her even more clearly in my mind. Kundera would explain what I needed to know; Gaddis didn’t. As a result, I was lucky enough to have the experience of Esther coming to “life” (in a way Kundera might not approve of) and, in Kundera’s books, of having the equally pleasurable but opposite experience of engaging with the author’s musical mind on the (silent) page.


Kundera’s careful control of interpretation lessens the risk of a reader simply not “getting it.” A reader’s frustration at not hearing the music, at not feeling the reference, at being left behind, is an inherent risk in musical (or any other) reference-making. But an author can also use that frustration for a powerful aesthetic effect.

In James Joyce’s story “The Dead,” music’s power (neither virtuous nor evil—just power) is expressed especially well precisely because the reader can’t hear it.

The song seemed to be in the old Irish tonality and the singer seemed uncertain both of his words and of his voice. The voice, made plaintive by distance and by the singer’s hoarseness, faintly illuminated the cadence of the air with words expressing grief.

Like Kundera, Joyce makes almost no effort to describe the music, and, unlike Gaddis or Kundera, doesn’t even name the song until later, long after this vague description above. But that delay and denial turn out to be very much to his purpose.

The protagonist’s wife has been greatly affected by the song:

At last she turned towards them and Gabriel saw that there was colour on her cheeks and that her eyes were shining. A sudden tide of joy went leaping out of his heart.

“Mr. D’Arcy,” she said, “what is the name of that song you were singing?”

“It’s called ‘The Lass of Aughrim.’”

As it turns out, the tune would have been well known to his original readership, but it required, at least in my case, some YouTubery. Even if the reader now recognizes, however, the song, its performance (which so touched the woman Gretta) is long past; we’re already downstairs, on our way out of the party, concentrating on other details and action. The moment when a reader might have felt a horn’s chill has passed. But Gretta is still moved by it, because it carries vast and significant connotations about her past (and only hers), while the reader (like Gabriel) has only recently been given a title, and the experience of watching another person have a musical experience they cannot share.

Is there a difference between not knowing The Great Elopement and not knowing “The Lass of Aughrim”? I think there is. Unlike the Gaddis citation, Joyce’s description of emotion is separated by some distance from the music that inspired it. The delayed mystery of the song’s title distances me from Gretta: something is happening to her because of music I don’t know. Gaddis, however, cited the music and described its effect on the character simultaneously, while Joyce described an emotional state, triggered by distant music (distant in the space of the house and over the space of the pages). Gaddis would, I believe, like me to feel what Esther is feeling. Joyce would, I believe, like me to notice from a distance what Gretta is feeling, to recall it not as something that I have felt, but as something that I have seen someone else feel.

The story’s heartbreaking conclusion is that the song has profound meaning to Gretta but none at all to Gabriel, and I find the story so wrenching (even in re-reading) precisely because Joyce has placed me firmly in the shoes of the character who has heard nothing special.


There is nothing more open to an individual reader’s unique perspective than fictional music—meaning invented music that exists only in a book.

Critics have suggested that Marcel Proust’s composer Vinteuil from In Search of Lost Time was based on Saint-Saëns, Franck, Fauré, or Debussy. Proust knew some of these men, even wrote fan letters to some of them. More to the point, Proust wrote other fiction in which he did name an actual composer, just as Jane Austen did. But in Swann’s Way, he commissions a fictional creation instead. This is not trivial.

When, at a party, Swann hears the andante movement of Vinteuil’s Sonata for Violin and Piano, it overpowers him. As in Kundera’s Bach fantasy, Swann lives before commonplace sound reproduction, and so, having heard the sonata only once, he staggers about in a frenzy of desire to hear it again, or even to learn its name, a quest that takes a full year:

When he returned home he felt the need of it: he was like a man into whose life a woman he has seen for a moment passing by has brought the image of a new beauty which deepens his own sensibility, although he does not even know her name or whether he will ever see her again.

He spends months looking for this music, because he feels, among other sensations, revived by it, even morally revived:

… as though the music had had upon the moral barrenness from which he was suffering a sort of re-creative influence, he was conscious once again of the desire and almost the strength to consecrate his life.

But to go back to the first time: Swann doesn’t know the piece or its name. We watch him hear it for the very first time. This is Proust, so it takes several pages to describe all the emotional effects on Swann of this experience. First, the music evokes poetic images in Swann’s mind: seascapes, moonlight. Then, Proust proceeds to the sounds’ effect on Swann’s spirit:

… without being able to distinguish any clear outline, or to give a name to what was pleasing him, suddenly enraptured, he had tried to grasp the phrase or harmony—he did not know which—that had just been played and that had opened and expanded his soul….

It is interesting to note that Swann is not musical. He doesn’t know the terminology of what he’s hearing, nor can he even hum or whistle the sounds that so affect him.

And that ignorance (and our ignorance: this is not Debussy or Fauré, this is only and explicitly Vinteuil) allows Proust to point out that Swann

received so confused an impression, one of those that are none the less the only purely musical impressions, limited in their extent, entirely original, and irreducible to any other kind… vanishing in an instant. [Italics mine]

Pure musical appreciation is chaos, wordless. And, a fascinating observation, music (especially music we don’t yet “know”) moves faster than our own experience of it:

… the notes themselves have vanished before these sensations have developed sufficiently to escape submersion under those which the succeeding or even simultaneous notes have already begun to awaken in us.

If music moves us beyond our ability to put words to it, moves us faster than we can assimilate its effect on us, then I was going to surrender and confess that even Proust believes this writing to be “merely” dancing about architecture. But then he noted this: we have memory to help us rebuild fleeting and wordless feeling:

Scarcely had the exquisite sensation which Swann had experienced died away, before his memory had furnished him with an immediate transcript, sketchy, it is true, and provisional, which he had been able to glance at while the piece continued, so that, when the same impression suddenly returned, it was no longer impossible to grasp… he had before him something that was no longer pure music, but rather design, architecture, thought, and which allowed the actual music to be recalled.

Music has become like architecture (!), and now it can be recalled and described.

Proust gives us little sense of what the sonata “sounds like” (it probably sounds like a violin-piano sonata by Debussy), but when I read these pages in silence and with patience, I have the sensation of hearing this fictional music. This is a result of Proust’s creative work: something musical in my own remembered (architecturally reconstructed) experience enters the auditory imagination, and, under the heat of Proust’s language, the memory is melted down again into that wordless sensation of “pure music,” experienced and heard again as Swann hears it, only then to be refrozen into words, memory, theory. And yet—here’s the paradox—to melt my memory down again, to reexperience (within limits, with diminishing returns, as with a madeleine dunked in linden tea) that shocking and emotionally chaotic, wordless sensation of the first time (so like love), can be achieved by reading words. Proust, via prose, is able to trick my brain into secreting a familiar mixture of intoxicants and confusion; he’s reproduced on paper and in my mind something like what happens when I hear music. But for fear of a legitimate accusation of pretension, I might almost say that I am literally learning to like Vinteuil’s sonata.

The passage is, obviously, just words, but something about the effort to harness music specifically has produced a readerly high (at least in me) that seems not far from my musical highs.

Or not. These pages can leave the next reader entirely deaf. (I do wonder what effect it might have on a literally deaf reader. Interestingly, this volume, Swann’s Way, was the only one of In Search of Lost Time’s seven books that the boastingly unmusical Nabokov endorsed.) Some readers will, like me, insanely think they hear Vinteuil, some will imagine they hear Debussy, some will recall how they felt when they heard Debussy, some will hear nothing yet love the words and descriptions nevertheless. So is the result merely “inspired by music,” like a dance about architecture? Are its pleasures independent of music except as the nominal topic? Does writing about music have to be music to be worthy of existence? There is nothing inherently laughable or false about a dance about architecture. But you wouldn’t want to forsake buildings to live in the dance, nor would I give up music in exchange for my favorite writers’ descriptions of it. Yet there is a pleasure to be found in that dance, in that writing, that both invokes the subject and creates something new from it, and, when you return to the architecture, to the music, when you wander the halls or turn on your stereo, you carry with you now some new wisdom and love for the house, for the song, that you gained elsewhere.

More Reads

Analects of the Influence of Artaud

Rick Moody

Close Encounters of the Metafictional Kind

Garrett McDonough

The American Vicarious

Jonathan Lethem