How Music Works

by David Byrne

  So far, this was all very much like the process Eno and I had used in the Bush of Ghosts sessions. But we weren’t going to use found vocals this time, so when these sections had been created, Eno or I would go into the studio and sing to them, improvising a wordless melody. Often it took a few tries before arriving at a suitable melody for a verse and a different one for the chorus. Sometimes even harmonies, also wordless, were added, to give the impression of a rousing chorus. We then did rough mixes of these “songs”—including these gibberish vocals—for everyone to listen to, while I took them home to write actual words. We agreed to reconvene in New York after I’d finished my lyric-writing assignment. On a previous record, the song “I Zimbra” actually got left as gibberish—or rather, the gibberish early-twentieth-century Dada poet Hugo Ball had written seemed a perfect fit. But now I wanted to find real words to substitute, which was going to be tricky.

  How did this recording process affect the music? A lot. For starters, the constant loop-like grooves on these tracks and the unchanging bass line that carried through many of the songs meant that tricky structural devices like meter changes or half measures were unlikely to occur. Complex chord changes of the sort one might hear in ordinary pop songs, bossa novas, or standards were very unlikely, too (usually there were no chord changes at all). Such devices are often employed to keep a song interesting, so we had basically abandoned the rules we had previously accepted for determining structure and arrangement. While punk rock was celebrated for needing only three chords, we had now stripped that down to one. This fairly strict self-limitation might seem perverse, as it restricted the kinds of top melodies one might write, and a melody that doesn’t change keys has a hard row to hoe—it runs the risk of being overly repetitive and boring. But using only one chord has its advantages, too: more emphasis gets placed on the groove. Even if a given song wasn’t particularly aggressive rhythmically, the groove tended to feel insistent, and you noticed it more. This made the tracks feel more trance-like, somewhat transcendent, ecstatic even—more akin to African music or Gospel or disco, though the way we played placed us well outside those traditions. Not only were these tracks groove-centered, they were also very much about texture. The changes from one section to another were sometimes driven more by textural variation than by melody or harmony—more like minimal classical music or some traditional forms of music around the world than the rock and pop traditions we came out of.

  I’m exaggerating a little here—there were certainly other kinds of pop music in North America that worked this way. Many James Brown songs, Hamilton Bohannon, and some Mississippi blues guys basically groove, twist, and elaborate around one chord. We knew and loved much of that music too. We had, it seemed, taken the long way around to arrive at a place that, structurally at least, we should have been slightly familiar with already. Like that T.S. Eliot quote about arriving where we started and knowing it for the first time, we were essentially reinventing something we already knew, something right there in our own backyard. But of course, in the reinvention process we got much of it “wrong,” and, for example, the version of funk we ended up with was skewed, herky-jerky, and somewhat robotic. The result was something with that familiar structure, but now made of strange and different parts.

  MUSIC WRITES THE WORDS

  Significantly, rhythm and texture are the two most difficult aspects of music to express in conventional Western musical notation. These qualities, some of the most resonant and important in contemporary popular music, and in some ways the most “African,” were excluded from, or maybe simply outside of, the system by which music was traditionally taught, passed on, notated, discussed, criticized, and—very important—copyrighted. The copyright of a musical composition is based on the top-line melody, the specific harmonies that support it, and, in the case of a song or opera, the lyrics. There is no acknowledgement of groove, sound, texture, or arrangement—all of which are features of the recorded music of our era that we listeners have come to savor and identify as integral to an artist’s work. This failure would lead to some conflicts down the road. The drummer on James Brown’s tune “Funky Drummer,” Clyde Stubblefield, claimed that he was due some percentage of the money received by Brown and his publishers when that song (the drum break especially) was sampled by countless acts in recent years. Legally, Stubblefield’s contribution fell outside of what one would traditionally call “composition,” but realistically those were his drum breaks that everyone wanted. Determining and attributing these contributions is complex. One could argue that it was Brown who suggested that Stubblefield take his famous break, and similarly, that if it wasn’t on a James Brown record, no one would ever have heard it. Stubblefield has suggested that he should be compensated, but the issue has not been resolved.

  I felt that the melodies and the lyrics I was going to write for Remain in Light had to respond to all these new (for us) musical qualities. One might even say that the recording process, because it privileged trance-like and transcendent music, was about to affect the words that I would gravitate toward. The gently ecstatic nature of the tracks meant that angsty personal lyrics like the ones I’d written previously might not be the best match, so I had to find some new lyrical approach. I filled page after page with phrases that matched the melodic lines of the verses and choruses, hoping that some of them might complement the feelings the music generated.

  On the following page is a small sample of some of the phrases I jotted down in this manner for the song that eventually became “Once in a Lifetime.”F Judging by the fact that on this particular page there are actual stanzas and couplets, I suspect I was pretty far along in the process—earlier pages would have included anything, absolutely anything, that fit the meter and syllabic form, and might not have included any rhymes or be on a trajectory toward a subject. What’s with the color coding? I believe that the red texts were the culled favorites. I still often write in very much the same way today, but without the color coding and minuscule handwriting.

  Courtesy of David Byrne

  I tried not to censor the potential lyrics I wrote down. Sometimes I would sing the melodic fragments over and over, trying random lyric phrases, and I could sense when one syllable was more appropriate than another. I began to notice, for example, that the choice of a hard consonant instead of a soft one implied something, something emotional. A consonant wasn’t merely a formal decision, it felt different. Vowels, too, had emotional resonances—a soft ooh and a pinched nasal aah have very different associations. I felt I had to adhere to whatever syllables seemed to fit the existing melody best, so I’d listen to the gibberish vocals respectfully and let those be my guide. If I seemed to instinctively gravitate to an aahh on the gibberish vocal mix, I’d try to stick to that sound in the lyrics I was writing. More restrictions, but okay.

  In keeping with the rapturous nature of some of the tracks, I was also drawing lyrical inspiration from the radio preachers I’d been listening to and that we’d used on the Bush of Ghosts record. At that time, American radio was a cauldron of impassioned voices—live preachers, talk-show hosts, and salesmen. The radio was shouting at you, pleading with you, and seducing you. You could also hear great salsa singers, as well as gospel being broadcast straight from the churches. I can only imagine Eno’s reaction, coming from a country with four fairly restrained radio stations! I don’t listen to the radio much anymore, though. There is still variety on some stations, but it’s mostly been homogenized, like so many other parts of our culture.

  I wrote the verses of “Once in a Lifetime” at my home on East Seventh Street in the East Village. I started by taking on the character of a radio preacher I’d heard on one of my cassettes. There was a serious use of anaphora—employing the same phrase to begin each sentence. It’s a common device that preachers use, and it brings their speechifying one step closer to poetry and song. One or two fragments that I used—the repetition of the phrase “You may find yourself,” for example—were straight lifts from the radio pr
eacher, but from there I’d improvise and change the focus from a Christian message to, well, I wasn’t sure at first what I was getting at. The preacher was focusing on the lack of spirituality in material striving; he may have begun by telling the listener that there was nothing wrong with living in a shotgun house, a house in which all the rooms are in a line so straight, you could shoot a shotgun straight through (you see them a lot in New Orleans). So the mention of the beautiful house, the beautiful wife, and the trappings of an ideal life situation would have been a natural segue for me. I’d get myself worked up, pacing back and forth, breathing in synch with the preacher, phrases would come into my head and I’d jot them down as quickly as possible. I maybe went off topic once or twice.

  The lyrics for another song, “The Great Curve,” were inspired by Robert Farris Thompson’s writings about African spirituality, and the feminine goddesses that survive today in remnants like Mother Nature or Yansan and Oshun in Afro-Atlantic cultures. My lyrics, though they didn’t all start out addressing that subject, began to approach and circle around that idea, so I tried to urge them closer, and to reject ones that seemed to go in a different direction. It was tough work—I had to reinvent myself as a lyricist to match those tracks. I wasn’t writing about my own anxieties anymore—I had to leave much of that behind. Not every track we recorded became a finished song. There were some really wonderful tracks that I just couldn’t find words to fit. But we had ample music, so when we went back into the studio I could sing the words I’d written, get everyone’s reaction, and if one set of lyrics worked, we’d record a properly sung vocal and then move on.

  We spent maybe two weeks in New York recording vocals and some additional overdubs—Adrian Belew’s amazing guitar solos, some trumpets and percussion. It was all fairly exciting. But when the record finally came out in 1980, radio stations balked. I suspect it may have been the videos on MTV that introduced that record to many people. MTV had just launched, and they were starving for content; they’d play pretty much any decent material they were handed. Not too many people had cable TV back then, so MTV had no hesitation about playing the same videos over and over. Hard to believe, but at the time, if you made almost any halfway interesting video you could possibly have it up and running on cable TV almost instantly. For me it was a godsend—a way to reincorporate my art-school roots into the music side of things. The video for “Once in a Lifetime” and the one for “Burning Down the House” were made fairly cheaply, and they both went into pretty heavy rotation. Years later I gave MTV an animated video for a song by a Brazilian singer, Jorge Ben (“Umabarauma”), and even that got played.

  Arranging songs by switching tracks on and off like we did for Remain in Light (and on the two records that followed) was very much what hip-hop artists were doing at the same time: looping a groove to make an underlying bed and then making sections by nominating other sounds and parts to go in designated places. Samplers didn’t really exist yet, so even the hip-hop artists were looping somewhat manually by playing break sections of records over and over and then layering other sounds and vocals on top of the breaks.

  I went a bit further texturally with this process on The Catherine Wheel, a score I wrote for choreographer Twyla Tharp in 1981. On The Catherine Wheel, I worked with some musicians outside my band—ney player Richard Horowitz, Adrian Belew, who came back after touring with us, drummer Yogi Horton, and Bernie Worrell, who added some keyboards. I got to pay back writer and drummer John Miller Chernoff, whose book on African drumming had inspired both Eno and me, when I invited him to play some African drum patterns on a prepared upright piano. The strings inside the piano were muffled so the sound didn’t ring out, and the result was pitched thuds. It worked.

  TWO RECORDS AT ONCE

  We did another record, Speaking in Tongues, that continued with this idea of using improvised initial riffs and gibberish vocals as a guide for lyric writing. That record turned out to be the most commercially successful so far. After completing a tour that was filmed (Stop Making Sense), I got it into my head to direct a movie. Talking Heads, as a band, was fairly popular by then, and I figured if the movie had some songs scattered throughout, it would help me get financing. I was right, but it still took a long time to get the funding and production in place, so in the meantime, I wrote some songs that could be used for a new Talking Heads record. I decided to write them ahead of the recording sessions, in what now seemed to us the old-fashioned way—playing a guitar and singing along. Sometimes I wrote the songs by using two boomboxes. I recorded the guitar chords on one machine, then played that back and sang along, recording the result of that live “overdubbing” on a second machine. Other times I used a Tascam, an attaché-sized 4-track recorder that used ordinary audiocassettes played at high speed. The quality wasn’t great, but as a writing tool and a way to make demos for the band, it was sufficient.

  We decided to economize. After rehearsing the new songs, we recorded them relatively quickly and conventionally in New York. We were fairly comfortable in the studio by now; much of the alienation and terror had worn off. When we had finished the overdubs and the singing on that batch of songs, we let Eric Thorngren start mixing them while we went into the adjacent live room and began rehearsing another album’s worth of material that I’d written for the movie. These would be recorded with my “guide vocals,” but in most cases it would be the actors who would later sing the songs, replacing my voice singing over our tracks.

  This was all very workmanlike, and by the time Little Creatures, the first of these records, came out, I was already in Texas preparing to shoot the movie, True Stories. I took the multitrack tapes of our backing tracks for the movie songs to the Dallas set and added some local Texas flavor—fiddle and pedal steel on some songs, Norteño accordion on another, a gospel choir on yet another. The actors from the movie came into a local studio in Dallas and sang as well.

  They didn’t sing live when they were being filmed—they lip-synched, as I’d been doing in music videos for a few years. Lip-synching was an old Hollywood musical technique that allowed the audio part of performances to be more consistent, as everyone mimes to pre-existing recordings. It also enabled the camera and other departments to plan, down to the second, how long or involved each shot might be, since they can time each bit of the audio recording. The camera might plan a dolly shot for a particular lyric, and by using these recordings they’d know how long they had, down to the second. One might sacrifice some spontaneity and happy vocal accidents with this approach—it’s not entirely without its downside—but it also meant that all the takes would match, something that’s hard to do in a performance unless there are multiple cameras on a shoot. Recently I was involved in a film by the Italian director Paolo Sorrentino, who wanted a performance by my band, which consisted of me, a rhythm section, and six string players. Rather than have us lip-synch, he wanted to film it live, so we’d be performing and recording at the same time as the cameras ran. This approach is more authentic— the sound is actually us playing and singing—but it’s possible that artifacts of the documentation of a live performance (the less than pristine sound, for example) might be distracting to the viewing audience. In the screening I saw, the sound mixer had “helped” the reaction of the audience of extras later by adding in the applause and shouting. And it worked! I believed the hype as much as a potential movie audience presumably would. The scene was definitely more exciting; even knowing how all the parts were put together, I’m as easily manipulated as anyone else.

  PARIS, AN AFRICAN CITY

  After Little Creatures and True Stories, Talking Heads wanted to return to the more collaborative writing approach we’d used before—but with some adjustments. Instead of going into the studio with absolutely no songs, as we’d done on Remain in Light and some other records, we decided that we’d improvise some grooves and riffs ahead of time and choose the best of them as the foundation for studio recordings. I recorded some of these rehearsal jam-fragments on audio cassettes,
and by cueing up specific moments I managed to string together a few sequences that could be used to make a song structure. We’d learn how to play a bit of fragment A, then we’d move on to a bit of riff B, then go back to A, and then on to C. In this way we already had some song framework—the material for what might become verse and chorus sections. Still no words or top-line melodies, though. I’d do those later, as I had before.

  Over the years I’d been to a number of Paris clubs to hear music with the late Jean-François Bizot, who had a magazine called Actuel that I admired. We’d see Cuban or African bands or singers, and we’d eat at African restaurants. The African Diaspora was turning Paris into a hub that featured some of the best African music in the world—many of the best musicians had moved there, or spent much of their time there. I proposed to Talking Heads that we record in Paris to take advantage of what seemed to me like a special moment and to work with some of those musicians. Not to pretend to be an African band, but to see if something new—a third thing—could emerge. It helped that we already had basic parts and structures to play—a minimal foundation, but one that could be built on.

  We worked at Studio Davout, a former movie theater out on the Périphérique. The room was immense, unlike most New York studios, and we were going to use digital recorders, which made us feel that our record was going to sound sparkly fresh. As an ensemble, we were all playing at the same time, and there was enough distance between us that we could hear and see each other but still have some acoustic separation.G Our new producer, Steve Lillywhite, enjoyed having some of that comforting isolation between instruments.