How Music Works

by David Byrne

  After some years of a more or less traditional songwriting process—words and music completed by one person, or finished words by one set to music by another—Talking Heads evolved a kind of collaborative music-writing system based on collective improvisations. Sometimes these jams would happen in a rehearsal loft—the song “Life During Wartime” began as a one-chord jam with no lyrics based on a riff I’d brought in, which was wedded to a second chord that became the chorus. Sometimes these improvisations and jams wouldn’t happen until we were in a recording studio. In such instances, the writing and recording were simultaneous. Jazz players, of course, respond fluently to one another by improvising in their live performances and in their recordings. We, however, were fairly minimal about what we would contribute. The aim of our improvising, probably inspired by our R&B heroes, was for each person to find a part, a riff, or even just a freaky honking accent, and then stick with it, repeating it over and over. So by improvisation I don’t mean long meandering guitar solos. Quite the opposite. Ours were more about hunting and pecking with the aim of “finding” short, sonic, modular pieces. These pieces were intended to interlock with whatever was already there, so the period of actual improvisation would be short. It would end as soon as a satisfactory segment was found. Then we would shape those accumulated results into something resembling a song structure.

  In this system, one person’s response to another’s contribution could shift the whole piece in a radically different direction—harmonically, texturally, or rhythmically. Pleasantly unexpected surprises would occur, but just as often they could seem like rude and arrogant impositions that missed the significance and integrity of the preexisting material. The guitarist Robert Fripp added a part to the Talking Heads song “I Zimbra,” overdubbing a weird harmonic ostinato that he played through the whole song. The whole song! Initially that destroyed the song, and seemed like someone was being willfully perverse. But, as it turned out, when used sparingly it added an little psychedelic swirl to our Afro-pop groove, which put everything in a new perspective. Is this disruption and destruction a risk worth taking? Did the piece just get ruined, or did it really need to get radically rethought in order to go somewhere new and exciting? You can’t be too precious in this process. For us, this method resulted in music in which the authorship was to some extent shared among a whole group of people, though I still usually wrote the vocal melody and eventually the words. The musical bed was, in these instances, very much collaborative.

  NOTATION AND COMMUNICATION

  There are not a lot of languages for describing and passing on music outside of traditional notation—and even that method, though almost universally accepted, sacrifices a lot. The same piece of written music can sound completely different depending on who plays it. If Mozart could have described in notation exactly how he intended every aspect of his compositions to sound, there would be no need for multiple interpretations. When musicians play together and record, they come up with terms—real and invented—to try to communicate musical nuance. Funkier, more legato, more holes and spaces, less pretty, spikier, simpler, pushed hard, more laid back— I’ve said all of those things when trying to describe a musical direction or the feel I was looking for. Some composers resort to metaphors and analogies. You could use food, sex, texture, or visual metaphors; I’ve heard that Joni Mitchell described the kind of playing she wanted by naming colors. Then there’s the shorthand of referring to other recordings, as Talking Heads did. So, interpreting a written score, reading music notation, is itself a form of collaboration. The performer is remaking and in some ways rewriting the piece every time he plays it. The vagueness and ambiguities of notation allow for this, and it’s not an entirely bad thing. A lot of music stays relevant thanks to the opportunities for liberal interpretation by new artists.

  To encourage this kind of collaboration, to make the interpretive aspect more overt, some composers have written their pieces as graphic scores. This is a way of granting a generous degree of freedom in the interpretation of their work, while simultaneously suggesting and delimiting the organization, shape, and texture of their pieces across time. Below is one example, a graphic score by the composer Iannis Xenakis.A

  This approach isn’t as crazy as it might seem. While these scores don’t specify which notes to play, they do suggest higher or lower pitches as the lines wander up and down, and they visually express how the players are to relate to one another. This type of score views music as a set of organizing principles rather than a strict hierarchy—the latter viewpoint usually ends up with melody at the top of the pile. It’s an alternative to the privileged position melody is usually given—it’s about texture, patterns, and interrelationships.

  Courtesy of Bibliothèque nationale de France

  Robert Farris Thompson, a professor of art at Yale, pointed out that once you let yourself see things this way, lots of things become “musical scores”—although they might never have been intended to be played. He argued that in a lot of African weaving, one can sense a rhythm. The repetition in these fabrics doesn’t consist of a simple looping of mirror images and patterns; rather, modular parts recombine, shift position, and interact over and over with one another, aligning in different ways over time, recombinant. They are scores for a funky minimalist symphony. This musical metaphor implies a kind of collaboration as well. While each color module in a quilt or textile is essential, no one part defines the whole the way we might define many Western compositions by their dominant melody. Western compositions can often be picked out—the melodies, at least—with one finger on a piano. How would one pick out the “score” below in that way? There’s no dominant motif or top line, though that doesn’t stop it from having a distinct identity. It’s a neural network, a personality, a city, the Internet.

  Below, on the left, is an African textile.B No surprise that later versions of these patterns, like the one pictured on the right, originated in the New World.C There are musical breaks, fugues and stanzas, inversions and recapitulations here. It’s not that crazy to believe that some part of the vast African musical sensibility was carried across the oceans and reconstructed using visual means—that these fabrics functioned as a structural mnemonic aid. Perhaps they functioned as metaphors for how music could be organized, which is also a lesson that can be applied to other parts of life. I’m not suggesting that musicians sat down and “played” a quilt, but some of the organizing sensibility might have been kept alive and transmitted by such means.

  Kente Prestige Cloth, on display at The British Museum, London

  Kente Prestige Cloth, on display at The British Museum, London

  If music can be regarded as an organizing principle—and in this case one that places equal weight on melody, rhythm, texture, and harmony—then we start to see metaphors everywhere we look. All kinds of natural phenomena are “musical.” And I don’t mean they make sounds, but rather that they organize themselves, and patterns become evident. Forms and themes arise, express themselves, repeat, mutate, and then become submerged again. The daily street ballet that Jane Jacobs wrote about, and the hustle and bustle of an outdoor market, are each a kind of music. Stars, bugs, running water, the chaotic tangle of vegetation. Musicians playing together find a kind of symbiotic relationship between one another and an interplay between their parts, so that the interlocking and interweaving create a sonic fabric.

  How does this work? Let me share a few very different examples.

  STANDING ON THE SHOULDERS OF GIANTS

  A recent record of mine, Everything That Happens Will Happen Today, was pretty typical, as far as the collaborative process goes. Brian Eno, whom I hadn’t worked with in more than twenty-five years, had a slew of largely instrumental tracks on his shelf that seemed to want to become songs rather than ambient tracks or film scores, but he was unhappy with his own attempts at completing them. He didn’t have much to lose by passing them to me—they were just gathering dust (although I was told one did get passed to Coldplay), so unles
s I did something horrendous, which we agreed he could veto, it was a win-win situation.

  As might be obvious by now, most contemporary collaborations, at least the ones I do, don’t take place face-to-face anymore. They are the result of digital music files being shuttled back and forth via email or other Internetbased file-transfer formats. Does something get lost when the live aspect of collaboration disappears? Simple miscommunication can certainly spiral out of control without the subtle signals we send through our facial expressions and body language. And the encouragement, coaching, hyping, and prodding that tends to happen in person—“Why not try this?” or “That’s great, but what if you play it on a different instrument?”—may not happen, and certainly not as spontaneously.

  That said, there are big advantages to the new protocol. If I can use a pingpong analogy, with Internet exchanges one can wait overnight or longer to return the serve, planning out what addition might work best, with no pressure to come up with something brilliant on the spot. The breathing space is a luxury you don’t have when your collaborator is looking over your shoulder.

  From Eno’s studio in London, I was sent stereo mixes of his musical ideas, to which I added my vocal melodies and (eventually) lyrics without altering his music beds in the least. Sometimes this made for some odd lyrical structures. On the song “The River,” Brian had a portion of what became a verse repeat quite a few times—as if the song had gotten stuck, and couldn’t move on. I accepted the challenge to write without straightening this peculiarity out—I knew that if it could be made to work, this unexpected variation in structure might prevent the song from being too predictable. It worked, and it added a kind of tension, as it delayed the musical resolution that came at the ends of verses. But just as often I slightly restructured his songs to bring them closer to a traditional form—repeating a section to create a place where a second verse could go, or nominating a “larger”-sounding section as a chorus, and then I might copy that, too, in order to make it recur again later in the song. However, I never even thought about requesting any substantial musical changes in the tracks, like key changes, or changes in groove or instrumentation. The unwritten rule in these remote collaborations is, for me, “Leave the other person’s stuff alone as much as you possibly can.” You work with what you’re given, and don’t try to imagine it as something other than what it is. Accepting that half the creative decision-making has already been done has the effect of bypassing a lot of endless branching—not to mention a lot of waffling and worrying. I didn’t ever have to think about what direction to take musically—that train had already left the station, and my job was to see where it wanted to go. This restriction on creative freedom turns out, as usual, to be a great blessing. Complete freedom is as much curse as boon; freedom within strict and well-defined confines is, to me, ideal.

  I listened to Eno’s instrumental tracks on and off, trying to get a sense of the story the music was trying to tell. These tracks weren’t ambient, as one might have expected from him, and I sensed that song structures might emerge with just a little coaxing. “Emergence” is a popular term these days, but it almost perfectly evokes how musicians and songwriters cultivate the latent potential of a humble musical kernel. That’s why writers and musicians often say they feel only partially responsible for the creation of the works they’ve nurtured. They claim that the song, painting, dance piece, or words they’re working on “tells” them what kind of thing it wants to become. But when that thing that is speaking to you originated with someone else, it’s sometimes even more of a puzzle. Does it necessarily speak the same language as you? Is it sincere? Could their version be ironic instead? Is that clunky section supposed to be funny, or should you try to “fix” it? Do they want it to remain as beautiful and pretty as it seems, or would a little grit help?

  Well, I didn’t exactly know at first what to make of Eno’s tracks. Maybe I had some trepidation working in the shadow of Bush of Ghosts, which after thirty years had amassed kind of a weighty reputation. I knew we couldn’t let ourselves do a Bush of Ghosts II. Music history is as much an influence on composition as anything else. After living with the tracks for almost a year, I eventually wrote Eno back. I told him the music inspired a sort of folk-electronic-gospel feeling, and suggested that my words and tunes might reflect this, and did that direction seem okay? Brian had discovered his love of gospel music years ago, and as he eventually wrote in the liner notes to Everything That Happens:

  “Surrender to His Will,” by Reverend Maceo Woods and the Christian Tabernacle Choir was the first gospel song I ever really responded to. I heard it on a distant South American radio station whilst in Compass Point, Nassau, working with Talking Heads on the album More Songs About Buildings and Food. Spending time with them, and becoming aware of their musical interests, opened my ears to genres and styles I hadn’t really noticed up to that point, including gospel. So, it’s fitting that the circle should close with this record.

  As a foreigner in New York, where I ended up shortly after recording More Songs, I was surprised by how little attention Americans gave to their own great indigenous music. It was even slightly uncool, as though the endorsement of gospel necessarily implied support of its associated religious framework. Thanks to Reverend Woods however, I began to see gospel music as conveying the act of surrender more than the act of worship; and this, of course, intrigued me, and has informed my music ever since. Perhaps it’s the reason I use modes and chords that are easy to follow and harmonize with. I want music to be inviting, to offer the listener a place inside it.

  Though my trajectory as I described it to him was vague, Eno seemed fine with it, so I attacked the first song, which I think he had given the working title “And Suddenly.” I’d just finished reading Dave Eggers’s book What Is the What, which is about a young man named Valentino Achak Deng and his hallucinatory and horrific journey from his destroyed village in South Sudan to Atlanta, Georgia, and beyond. Valentino’s story was harrowing but also beautiful, uplifting, and at times even funny. I think I may have been under the spell of his story when I sat down in front of my microphone. The result was “One Fine Day.” I sang a few harmonies in the choruses to make it sound fuller, and emailed the result to Eno.

  We were both thrilled: what the song—the whole album, really—was to become was fully articulated here, in this first piece. The words I had gravitated toward indeed had some Biblical allusions (that would be the gospel connection I’d mentioned), but nothing too overt. We agreed to continue with the project.

  I realized that the harmonic foundations of some of the tracks Eno had sent were simple, much like traditional folk, country, or old-school gospel songs before those styles evolved to become as sophisticated as some are today. Brian’s chord structures were, in their apparent musical plainness, unlike anything I would have chosen myself. My music-geek side wouldn’t have allowed me to write a song with essentially just three major chords in it, not anymore—I thought I was supposed to have outgrown that. However, the fact that this almost naïve directness was someone else’s idea meant I could excuse myself—I could blame someone else, which made it okay. This pushed me in a new (old) direction, which, of course, was a good thing.

  The lyrical challenge was more emotional than technical: how to respond to these harmonically “simple” (though texturally complex) foundations and write heartfelt words without drawing on the clichés that such chords and structures might bring to mind? I was surprised that the results that began to emerge were often hopeful and positive, even though some lyrics describe exploding cars, war, and similarly ominous scenarios.

  There were some remnants of our previous work in these songs—no surprise there—but something new emerged as well. Where did this new sanguine and heartening tone come from, particularly in those troubled times? Every day, as the songs were emerging, I continued to be appalled by the cynical maneuvers of Dick Cheney, Donald Rumsfeld, Karl Rove, Tony Blair, and all the rest, as well as the disappointingly complia
nt manner in which they were reported by the media. By then, McCain was running for president and his minders had picked Sarah Palin to be his running mate—a move that was taken surprisingly seriously. A black man was running against them—a man who wrote inspiring speeches and held out a tiny bit of hope for some of us, though I think all politicians possess some amount of poison in their system. This was the political context in which I wrote these songs, and I found that my response was similar to that expressed on my previous solo records—hope and humanity as a force to counter cynicism and greed.

  Some of the lyrics and the plaintive melodies I came up with were a response to what I sensed was already there, hinted at, but buried deep in Eno’s music. I wanted to find a reason not to be cynical—to have some faith even when nothing around me seemed to justify it. Writing and singing seemed to be an attempt at a kind of musical self-healing.

  DREAMWORLD

  Red Hot, the AIDS charity organization founded in 1989, produces a series of benefit records in which they initiate collaborations between disparate musicians. Although he’s not Portuguese, it was suggested in 1999 that the Brazilian composer and singer Caetano Veloso and I collaborate on a song for their Red Hot + Lisbon collection. I’m a huge fan of Veloso’s, and we’d met a few times, so the idea of working together wasn’t too insane. I happened to have a song in progress on which I was using a percussion loop taken from one of his songs—an aid in the writing process that I would typically replace with real musicians somewhere down the line. Though some composers appear to be able to write over forms they hear in their heads, I find that when the rhythms I’m writing over are audible and a little complex, when they swing a bit, then actually hearing them keeps me on the rails as far as the metric of potential melodic vocal lines. That I’d been writing over a loop from one of Veloso’s songs meant that in a sense we’d already started collaborating and it made Red Hot’s invitation seem fortuitous.