How Music Works

by David Byrne

  The local musicians, among them guitarist Yves N’Djock, percussionist Abdou M’Boup, and keyboard player Wally Badarou, were great. They were professionals, very much in demand locally. They could adapt to styles outside of the traditions they had grown up with, and their response to our music was also entirely one of enthusiastic adaptation and accommodation.

  There was a new development in my writing for this record. Though I knew it might present problems when the time came to write the melodies and, even worse, the lyrics, I either bravely or foolishly decided to put together non-repeating sequences based on our instrumental sections. The passages I proposed we play in sequence would be similar to one another in certain ways, but they’d also keep changing and evolving as the song went on. Your typical pop song has a verse section, then a chorus section that is often bigger and might contain the “hook,” then back to the verse arrangement, and the whole thing repeats again. There are variations on this structure, but it’s fairly pervasive—even opera arias repeat sections this way. But what if each section, rather than being identical, were instead a stepping stone, a variation and elaboration leading to another similar but slightly different section, and there were no clear repetitions? I liked this idea. It proposed a song structure that was more like a conversation or a narrative. As a listener, you’d be on familiar ground and accompanied by familiar faces, but the landscape and settings would keep changing.

  Courtesy of David Byrne

  One new song, “Cool Water,” maintained a repetitive rhythm, but the key changed over and over until the very end, when it settled on a big G major chord. Other songs, such as “The Democratic Circus,” also proceeded through a series of similar but distinct sections. By the end, you were somewhere very different from where you’d started, but each step along the way was gradual and logical. Not every song worked like this, but I was curious to see if I could gently break the routine of slipping into familiar song structures without things sounding “difficult.” Sometimes the usual verse-chorus verse-chorus-bridge pattern could seem a little predictable—and, as I’d learned, your attention can wander when you know what’s coming.

  I improvised vocal melodies over the recorded tracks, just like I had in the past. We made rough mixes and then took a break, as before, while I sequestered myself and wrote words to match these “vocals.” I remember coming up with the words for the song “(Nothing But) Flowers” while driving around suburban Minneapolis. My wife at the time was working on a theater project there, and the only gear I needed to write lyrics was a cassette player to play the tracks for inspiration, another small one to record my lyric ideas, and a pad of paper to write them down on. I could work anywhere that I wasn’t going to be bothered—anywhere no one would hear me singing little fragments over and over, trying different words out.

  It wasn’t surprising that while driving around the suburbs, not all that far from the Mall of America, I began to imagine a scenario in which the economy had changed and the malls and housing developments had all begun to crumble and devolve to a prior state. The twist was that this scenario allowed me to also frame the song as a nostalgic look at vanishing sprawl, a phenomena I hadn’t thought that I was terribly sentimental about. It was obviously ironic in intent, but it also allowed me to express a love and affection for aspects of my culture that I had previously professed to loathe.

  NEW YORK, THE SECRET LATIN CITY

  For the Talking Heads record that was eventually titled Naked, I brought in Angel Fernandez to arrange Latin horns for “Mr. Jones;” I had also recently recorded a duet called “Loco de Amor” with my idol, the Queen of Latin music, Celia CruzH—a kind of salsa-reggae song for a movie soundtrack. But my love for Latin music hadn’t quite been quenched. I was still grooving to those records, particularly the older ones. At home and on the road I played them on a boombox and danced around to them in hotel rooms or rented apartments. I didn’t know the right steps, but no one was watching.

  I decided in 1988 that I would try to make a pan-Latin record, to dive into that world using a batch of songs I’d written as a foundation. I’d gotten into the habit of visiting Latin clubs and continued to immerse myself in the old records—it was all part of the history of my city, New York, so why not partake of it? Some songs had words and vocal melodies already, and others were instrumental tracks with verse and chorus sections in place. Jon Fausty, who’d recorded so many classic New York salsa records, joined Steve Lillywhite and me, and we decided to ask some experts how best to rhythmically and musically approach and develop my demos. Fausty brought in Milton Cardona and José Mangual Jr., two amazing percussionists from the New York salsa scene, to listen to my musical sketches and recommend appropriate approaches to rhythms and arrangements. I knew I wanted to include grooves drawn from a wide swath of South America—a cumbia rhythm from Colombia and a samba from Brazil, as well as the classic son montuno and cha-cha grooves that were the base of the Afro-Cuban base of New York salsa. I was being pretty ambitious. Latin musicians generally tend to specialize in one or another of these styles; salseros don’t usually play sambas, just like blues-rock guitarists don’t often play speed metal. But we recruited players from all around the New York metropolitan area, where pretty much every kind of musician from the New World could be found, and in this way we began the advance work.

  Courtesy of David Byrne

  We organized a series of recording sessions to lay down the rhythmic and harmonic beds for these songs—we wouldn’t worry about brass or strings or other arranged parts yet. Usually in the rhythm section there would be three percussionists working side by side, plus Andy González laying down a bass part on an electric upright and Paquito Pastor on piano. We were recording on a digital reel-to-reel recorder, like we had on Naked, though in retrospect this might not have been the best idea. The promoters of this new technology advertised a more accurate and pristine sound, but, as with the early generation of CDs, I don’t think the technology was up to the job yet. Those recordings have a slightly brittle quality that we might have convinced ourselves was actually crystalline clarity. The excitement of using new technology also inspired everyone to believe that what we were doing was important, of the moment, and top quality, just in case anyone had any doubts.

  As usual, I often improvised melodies to tracks that didn’t yet have any melodies or words. The horn and string arrangements would be recorded after I had these vocal melodies in place. Hernandez and other arrangers would react to my wordless vocal melodies. Their horns and strings would fill in the gaps around the vocals and leave a musical space for them to be heard.

  My singing style was inspired to change again, just as it had when we recorded Remain in Light. The vaguely melancholy melodies over the syncopated grooves—typical of Latin music—was attractive as an emotionally liberating combination. That the melodies and often the lyrics could be tinged with sadness, but the buoyant music acted as a counterweight—a sign of hope and an expression of life going on amid life’s calamities. The vocal melodies and lyrics often hinted at the tragic nature of existence, while the rhythms and music said, “Wait—life is wonderful, sexy, sensual, and one must persevere, and maybe even find some joy.” When it was time to record the vocals, I began to sing in the studio with as much of that feeling as I could muster, which maybe wasn’t all that much, given my background, but it was the beginning of another big change for me. My daughter was born around that time, so perhaps my evolving a more open singing approach might have reflected this big change in my life.

  With the help of Fernandez, I put together a live band, a fourteen-piece orchestra, including Brazilian singer Margareth Menezes, for a world tour. Near the end, in the middle of a South American leg, there was a bit of trouble. Some of the percussionists left (pushing everyone to play in myriad styles had its limits), and they were replaced by Oscar Salas, a great Cuban drummer from Miami. He knew all the grooves from the various regions, so it wasn’t as crazy a swap as it might have seemed. I realized that by
adding a kit drummer, I could possibly make some music that fused the muscle of funk and other styles with the lilt and swing of Latin grooves.

  I didn’t forget that insight, and for the next record, Uh-Oh, I continued working with Salas, and I brought in George Porter Jr, the incredible bass player from the New Orleans band the Meters, to work with us. Much Latin music has a framework referred to as the clave (the key), which sometimes isn’t even played or audibly articulated by any one instrument. (What a beautiful concept that is: the most important part is invisible!) The clave divides the measures into a three-beat and a two-beat pattern—for rockers it’s like a Bo Diddly beat, or the Buddy Holly song “Not Fade Away.” (Rock and roll didn’t just come from country and blues mixing; there was Latin flavor in there, too!) All the other parts, even the horns and the vocals, acknowledge the clave pattern and play with awareness of it, even if it isn’t always audible.

  I heard an undercurrent of clave in the New Orleans funk that the Meters made famous, which is not surprising given that city’s waves of immigration from Cuba and Haiti. I thought George might help me find a way to create a hybrid out of Latin grooves and the funk he was accustomed to. By now I was a little more familiar and comfortable with some of those grooves, and I felt I could let things move into uncharted and undefined territory. The rhythms didn’t have to be restricted to only one genre. I didn’t feel the need to have Milton or José determine how a song would go on this project; I more or less knew when I was writing the songs what the rhythms would be.

  I asked the Brazilian musician Tom Zé to do the arrangement for one of these songs, “Something Ain’t Right.” The groove was based on an ijexa rhythm, which is usually played on cowbells and is often associated with Candomblé, the Afro-Brazilian religion. The groove is featured in a number of songs by artists from Salvador, in Brazil’s Bahia region, so I knew Tom would be familiar with it. He did some wonderful horn arrangements, but then he surprised us all by pulling out some Bic pens, minus the ink cartridges, and passing them out to the horn players. Each plastic penholder had a thin little piece of plastic taped to the end, which functioned like a reed on a saxophone or clarinet. He had an arrangement all worked out for these guys to play the pens in one section of the song. It wasn’t just noise, either. He had them play a hocket pattern, where each player plays just one note, quickly, and by deciding which players played what, and when, an intricate pattern resulted. It was brilliant. only Tom Zé would have had the nerve to ask these New York session guys to play Bic pens.

  BACK TO THE BEGINNING

  In 1993, I wanted to write some songs that were more stripped down, and to foreground their emotional content. I sensed that if the horns and strings and multiple percussionists were stripped away a bit, then what I was singing about might communicate more directly. Maybe I had been getting carried away with the window dressing. This emphasis on a thoroughly personal kind of writing was a big change for me, and it was possibly as much in response to a recent death in the family as it was musical evolution.

  I wanted to jettison everything, to start from scratch. I had heard Lucinda Williams and my friend Terry Allen, and I wanted to write songs that seemed to come from the heart as much as theirs did. Writing from experience went against the grain for me, but I wanted to let the lyric content dictate the music a little more. I had the concept, though not the same instrumentation, of traditional jazz combos in mind. Musically I might have been inspired by the recent rise of the improvisation scene around the old Knitting Factory on Houston Street. I liked the idea of a small ensemble that listened to and played off each other and whatever the lead instrumentalist or vocalist was doing. So I wrote the songs, and instead of going straight into a recording studio, I put together a small band and performed them live, in small clubs.

  The songs and their arrangements began to gel as they were tested in front of live audiences. The plan was to more or less record the band playing live in the studio relatively soon, as I had an image in my mind of classic small-ensemble jazz recordings, with all the musicians more or less in a circle in the middle of a studio—a recreation of a club stage or bandstand. A situation in which everyone could hear and see everyone else, very old school. I hoped that after having done some live shows, the band would know the arrangements and would play each song as if it were second nature, an old friend.

  It didn’t work out that way. A band member was fired, producers came and went, the whole plan fell to pieces. But the core group—the rhythm section of Todd Turkisher and Paul Socolow—survived. The fairly stripped down and bared-soul aspect of that David Byrne record managed to allow me an escape from the musical cage I’d made for myself. I’d just recorded and toured with two very large, Latin-inspired bands, and as much as I loved that experience, I could tell I was being branded as the rocker who had abandoned the cause. This new record did feel like a fresh start, even though it was born out of death and thwarted plans.

  THE STUDIO COMES HOME

  By the late nineties, new audio technology had emerged that allowed musicians to make professional-caliber recordings in their home studios. I bought a little mixing board and a couple of DA-88 machines, which used Hi-8 video cassettes to record eight tracks of good-quality digital audio. Other companies came out with other machines. The ADAT devices used super-VHS videocassettes, which were cheaper than the Hi-8 cassettes, so more musicians adopted these for their home studios. They often synched the recorders to midi devices too, so as tape rolled, the sampler or other devices could be instructed to play pre-determined notes or drum samples along with the recorded tracks. Cheap Atari computers sometimes entered the picture as well. They had software that allowed you to make visual representations of these midi sequences which would then be used to trigger beats (typically), samples, and synthesizers. A whole song arrangement could be created without actually playing anything, and without the need for recording tape. With this gear one could see that the need for an expensive recording studio was beginning to become superfluous.

  TECH TALK

  These days I work in a home studio, which I’ve carved out of the larger room you can see on the bottom of this page.I

  Tidy, eh? Embarrassing, actually.J

  There is no professional sound-absorbent baffling here as there might be in a professional recording studio, but the floors of this former industrial building are concrete, so sound doesn’t really escape and bother the neighbors. I put industrial carpet down, and one wall is covered with a kind of sound-absorbent sheetrock, so I’ve taken some precautions to prevent sound leaking out. Unwanted sound coming in can be an issue too, but unless a truck backfires outside or an ambulance goes by, I’ve found that it is perfectly adequate, at least for recording vocals and guitars. There’s no space in my room for drums or anything like that, but for writing, playing one instrument, programming, and singing, it’s fine. There’s a good tube microphone, another mic for a little old guitar amp, and a nice preamp and compressor to massage the mic signals before they become ones and zeros. That’s pretty much all you need to find out where a song wants to go and, I’ve found, even enough to record real vocals and some guitar parts.

  Serial numbers and security codes for software are pinned to the wall, along with a Tammy Wynette poster. The computer is tucked under the desk. It’s a mess, but amazingly, this is how we make records now.

  Home-studio recordings can now sound as good as the big-name studios, and the lower pressure (and less expensive) vibe in a home environment is often more conducive to creativity. Home recordings can be used for more than just demos. This idea is somewhat revolutionary as far as recording and composing music goes, and the repercussions of these baby steps will be huge further down the road.

  Courtesy of David Byrne

  Courtesy of David Byrne

  By 1996, I’d written some new songs, which arrived in a wide variety of styles—maybe because I wasn’t writing for a specific band anymore. It seemed to me that the songs would best be interpreted by ei
ther different musical groups or by a single group pretending to be a variety of groups. For most of this record, I chose the former, deciding to record the material by inviting a lot of musicians and producers I liked to perform and record specific songs. Judicious casting of these collaborators, who would also be creative producers, helped me get the variety I felt the songs were asking for. I worked with Morcheeba at their studio in London, with the Black Cat Orchestra in Seattle, with Devo at their studio in LA, with Joe Galdo in Miami, with Hahn Rowe at my New York apartment, and with Camus Mare Celli and Andres Levin at a brownstone in Brooklyn. I racked up lots of frequent-flyer miles in the process, but most of the time I recorded economically, as each group had their own home-style studio.

  The record, Feelings, was necessarily done piecemeal, a few songs at a time, rather than in a concentrated burst. This too was new to me. The record evolved incrementally, and I had time to think about where it was going (or where it could go) as bits of each song and eventually the variety of song styles became audible. This more leisurely approach opened up the risk of my becoming indecisive, because I’d now have the option to postpone decisions regarding arrangements or which vocal take was the good one. However, I hoped that by this point in my life I’d internalized a fairly rigorous decision-making process and that I wouldn’t leave too many options dangling. Although it was technically possible to pile up tracks and delay making most decisions, I knew that I did have some intuitive sense of where a song wanted to go, so I would make a commitment quickly whenever possible.