How Music Works

by David Byrne

  Recording a performance with a band and singer all playing together at the same time in the same room was by this time becoming a rarity. An incredible array of options opened up as a result, but some organic interplay between the musicians disappeared, and the sound of music changed. Some musicians who played well in live situations couldn’t adapt to the fashion for each player to be isolated. They couldn’t hear their bandmates and, as a result, often didn’t play very well. I also hated this situation myself at first, but after many years I began to adapt. I’m not proud that I got used to this arrangement, but I did. I persisted in believing that what the microphone and the tape machine played back was more or less exactly what we had played, despite the fact that often what I heard on tape didn’t sound like what I had just heard in the room. I assumed, as Edison claimed, that technology was neutral, though now I know it isn’t. My ears were telling me one thing and conventional wisdom another. Which was I to believe? Other musicians and producers blossomed in this world, and took advantage of the opportunities that allowed them to construct elaborate simulations of bands and orchestras, but I felt like it wasn’t working for me at all.

  As pianist Glenn Gould pointed out, recording technology put part of the creative process in the hands of the producer, the tape editor, and the recording engineer:

  It would be impossible for the listener to establish at which point the authority of the performer gave way to that of the producer and the tape editor, just as even the most observant cinema-goer cannot ever be sure whether a particular sequence of shots derives from circumstances occasioned by the actor’s performance.1

  One could argue that these technicians were as responsible for how records came to sound as the composers or performers were. In effect, the authorship of a recording, and of music in general, was being spread around, dispersed. It became harder and harder to know who did what, or whose decisions were affecting the music we were hearing. Though music copyright and publishing still reflect an older, more traditional view of composition, these creative technicians demanded (and often received) an ever-larger piece of the monetary pie. Often they got a larger percentage than the individual performers.

  Just as theater is an actor and writer’s medium, and cinema is a director’s medium, recorded music often came to be a producer’s medium, in which they could sometimes out-auteur the artists they were recording.

  Earlier I mentioned that this recording philosophy meant idealizing the isolation of each instrument. This made eminent sense for vocals, which would inevitably have to be mixed so that they could be heard clearly above the din of the instruments that were often acoustically louder. However, there were risks in doing this. If there was a lot of drum “leakage”—if the sound of the drummer could be heard on the vocal mic—then every time you made the singer louder to hear a lyric better, the drums would inevitably get louder too. But isolation, which was the solution to this problem, often meant that we later found it hard to play together live. A band would go from being a tight unit to a chaotic mess.

  Yet this divide-and-isolate approach still had much to recommend it: mistakes on individual instrumental tracks could be fixed later, effects could be added to one instrument at the mixing and balancing stage without affecting another, and the relative volume of each instrument could be determined in mixing so that, for example, the brass parts could be made quieter during the sung verses.

  As we discovered with Talking Heads, simply sticking up a mic and assuming that it would capture the truth of a performance didn’t necessarily result in something that felt like a band, even though it was, objectively, capturing exactly what you’d played. I have gone out into the world with a good-quality recorder, and it’s fascinating to listen to the results of recording common places whose sound you’re super familiar with. The true audio chaos of a place is evident in a recording. Though I use the word true, and though this might be what enters our ears, it’s not what we “hear” when our brains process that sound. As far as our brains are concerned, what is “true” is often wrong.

  Sound mixers in movies work much like recording engineers do. They isolate the sound of the actor’s voice as much as possible, and then recreate the ambience of the environment later, adding birds, crickets, restaurant chatter, or whatever else is needed. What’s fascinating is that even those “natural” sounds are not “true” documents of an environment. For example, the sound of one solitary cricket mixed quietly into a scene can evoke a complete outdoor nighttime ambience in the mind of the audience, even though a true nighttime ambience would likely include distant traffic, wind, a faraway plane, a couple of barking dogs, and, yes, a whole bunch of crickets. All of this would be, if heard all together, perceived as sonic chaos, and wouldn’t convey the tranquil nighttime mood that was intended. Our brains organize the sound we hear in the same way our eyes selectively see. Recreating what something sounds like, what a sound or sonic environment feels like, in a music recording or a film, has become an art that some people are better at than others.

  Some musicians and producers ignored this isolate/deconstruct/reconstruct dogma. The Cowboy Junkies made their first record using one mic to record the whole band, and Steve Earl did something similar with a bluegrass band. In 2012, I participated in some recordings on which some brass players were seated next to one another (which meant their playing wasn’t isolated) while others were in completely isolated booths. Furthermore, the guitar and vocal tracks had been recorded beforehand in an apartment with no sound isolation at all. Nowadays the dogma isn’t adhered to quite as strictly; it’s one of a number of approaches one can take, and sometimes many approaches coexist on the same recording.

  By the time Talking Heads recorded our second record, More Songs About Buildings and Food, we had become friends with Brian Eno. We liked his music, so we asked ourselves, “Why not have him be our producer? At least he gets what we’re about, and we’re already friends.” Friendship, common interests, and sensibilities seemed more important to us, especially based on our previous experience, than whatever technical skills or hit-making track record he might have possessed.

  Seeing that we had turned into a pretty tight live band, Eno suggested we play live in the studio, without all of the typical sonic isolation. His semiblasphemous idea seemed worth a try, despite the risk of it resulting in a muddy recording. Removing all that isolating stuff, or at least most of it, was like being able to breathe again. The result sounded—surprise!—more like us, like the way we were used to hearing ourselves on stage, and so our playing was more inspired (or so we felt). Removing some of the sound-absorbent stuff meant that you could hear the faint sound of guitars in the background of the drum tracks, and maybe an electric piano might be audible in the background of the bass track. There wasn’t absolute separation of instruments, but we felt that the comfort factor more than compensated for this technical challenge. Eno even believed that the two mics he placed in the center of the room might be all that was needed to capture the entire band. That proved not to be entirely the case, though the sound of those mics did come in handy as a kind of supplementary ambience.

  Eno suggested we try some fairly unconventional vocal recording approaches also. Sometimes I sang along to all the music after it had been recorded, mic in my hand, often right in the control room with the speakers blasting, like a big live sound system positioned on either side of me. This wasn’t a kosher recording technique back then. The sound of the band coming through the speakers behind me could be heard on my vocal tracks, and typically producers would insist that singers go into silent padded booths. That isolation works for some, but it requires a leap of faith, as even the greatest headphone mix doesn’t sound as good as singing in front of a live band or blasting speakers. You have to imagine while you’re singing in the booth that the not-so-exciting band you’re hearing over the headphones is one day going to sound better than it does. I wondered how those professional singers did it. I wasn’t going to acquire those skills instantly, so
, to his credit, Eno’s suggestion was a good workaround. I’ve heard this is what Bono does, too.

  Most of the songs we recorded we had been performing live for a while, though a couple—“Found a Job” and “The Big Country”—were brand new, written for that record. In subtle ways those songs were shaped in the studio, though I’d initially written them as something I could play solo on my guitar. This process of learning new songs just before recording them would become commonplace for us. It was practical, but it meant that the songs often didn’t get as “broken in” before recording as they would have after a series of live performances.

  More Songs About Buildings and Food took three weeks to make. I think 77 might have taken two weeks, including mixing. That meant that the recording costs advanced (i.e., loaned) to us by the record company were low enough that, even with modest sales, we were able to pay them back relatively quickly. It also meant that the record company itself would probably make a profit on the record, ensuring (we hoped) a continuing relationship between them and the band.

  Did we think about those costs, and did they affect the way we recorded the music? Do music-business finances, especially those dictated by recording technology, in some ways determine what music sounds like? Absolutely. Limited finances acted as a set of creative restrictions, which, for us, was generally a good thing. Decisions about whether an overdubbed part worked or was well played had to be made relatively quickly. Indecision cost time, which we knew was constrained by our budget. The question of whether to rearrange a song or try it in a manner different from how we played it live often never came up—these were non-decisions that saved time and money. Using few outside players, who would have had to be paid, was another cost-saving move. These factors make certain records sound the way they do; they are not purely the result of musically inspired directives. Financial strictures don’t determine a melody, the lyrics, or harmony, but they do affect how a record is recorded, and therefore what it ultimately sounds like.

  For Fear of Music, the next Talking Heads album, we worked with Eno again. We decided to take our comfort level one step further: we recorded all the basic tracks (the four of us playing together, without me singing) in the loft where we rehearsed, which was also where Chris Frantz and Tina Weymouth lived. We hired one of those mobile recording studios that are used to record live concerts and sporting events, and parked it outside on the street, stringing the various cables through the window.A, B

  I’m not sure why, but it sounded far better than the messy lo-fi sound one would expect from recording outside of the pristine studio environment. We were finally beginning to capture what we sounded like live! Stepping outside of the acoustically isolated recording-studio environment wasn’t, it turned out, as catastrophic as it was made to sound. Hmmm. Maybe those rules of recording weren’t as true as we thought.

  This wasn’t the only way our recording process had changed. We also recorded some basic tracks for songs that I hadn’t written words to yet, and later we altered and supplemented some of the sounds of the instruments well after they had been recorded; some now-familiar arrangements were actually created after the performance. Recording songs that were “incomplete” was risky for me. I wondered, could I really create a song by singing over prerecorded riffs and chord changes? In some cases I did so quite successfully (“Life During Wartime”), but other tracks never got finished (“Dancing for Money”). But succeeding even half the time meant that this was indeed a feasible working method. I knew that in the future this meant I might be able to write words that responded to and were guided by prerecorded music, instead of getting all bent out of shape trying to get the music to fit—not just melodically, but sonically and texturally—prewritten lyrics. Others do this all the time now. St. Vincent (Annie Clark) went into the studio to do her second record, Actor, with no complete “songs” written; she had only musical fragments. You wouldn’t know this immediately from listening to her records, but it can be evident when others write their vocal melody over a sample of someone else’s song. In those instances, you know that the music existed years before the new song was written. The authorship is dislocated in such cases, both temporally and spatially. There can be no illusion that the music and lyrics were conceived together. Sometimes the writer of the music and the writer of the lyrics never meet.

  Courtesy of Record Plant Remote

  Courtesy of Record Plant Remote

  Eno was the one who encouraged us to mess with the sounds after they were recorded. He’d done a little of that on our previous record, but now the gloves were off. The furthest we went was on the song “Drugs.” We had initially recorded a fairly straightforward backing track that seemed a bit conventional, so we began to mute some of the instruments, sometimes just silencing specific notes. This made some parts open up; there were more gaps, more air. I had a recording of koalas that I’d made while we were on tour in Australia (they mainly grunt and snort, in contrast to their cutesy appearance), and that got added in here and there. The grunts worked like indeterminate animal answers and echoes to my singing. More sounds went on—an arc-like melody created using an echo machine, and then a guitar solo at the end that was made by selecting fragments from a number of improvised solos. Finally, I sang the song after jogging in the studio, because for some reason I wanted to sound out of breath. Of course, I was singing the same words and melody as I had been on the earlier, straighter, version of the song, but now to a vastly altered musical track—a fact that also affected how I sang. The song, as it was released, was an arrangement of sounds that one would never have come up with in rehearsal or by sitting writing with a guitar. It could only have been created in a studio. As Eno observed at the time, the recording studio was now a compositional tool.C

  IMAGINARY FIELD RECORDINGS

  (MY LIFE IN THE BUSH OF GHOSTS) 1980

  Around this time, Eno and I were listening to a lot of the beautiful recordings released on the French Ocora label. Some of these were typical field recordings, similar to the ones Alan Lomax had collected in the United States: they were often made with one mic and a portable recorder and featured African guys playing rocks, Pygmies singing their intricate hocketing songs, and Muslim calls to prayer. Other Ocara releases were recorded in more controlled situations: in concert halls, churches, temples, and maybe even a few studios. There were Georgian choirs, classical Indian musicians, Iranian traditional music, and North African griots, or praise singers.

  Ocara ecords treated a huge variety of music as if it were the equal of Western classical or art music. The recordings were given respect, thoughtful presentation, and technical attention that was all too rare for non-Western music. I had grown up on Folkways’s Nonesuch field recordings and the stuff Lomax had done for the Library of Congress, but the production values on the Ocora releases were on a whole other level. Eno and I realized that music from elsewhere didn’t need to sound distant, scratchy, or “primitive.” These recordings were as well produced as any contemporary recording in any genre. You were made to feel, for example, that this music wasn’t a ghostly remnant from some lost culture, soon to be relegated to the almost forgotten past. It was vital, and it was happening right now. To us there was strange beauty there, deep passion, and the compositions often operated by rules and structures that were radically different from what we were used to. As a result, our limited ideas of what constituted music were exploded forever. These recordings opened up myriad ways that music could be made and organized. There were many musical universes out there, and we had been blinkered by confining ourselves to only one.

  Photo by Hugh Brown

  By the end of 1979, Talking Heads had just completed what for us was a long world tour for Fear of Music. It was the first time we’d received offers to play in many places (the south island of New Zealand comes to mind), and we accepted almost every one of them. After we got back I took some time off to recuperate. I began spending time with Eno and Jon Hassell, who was beginning to develop and elucidate his “fo
urth world” concept, and we enthusiastically exchanged cassettes and vinyl we had found. Most of the music we were excited about came from outside the English-language pop-music axis. At the time, there was no way to find out about this kind of music except by word of mouth. There was no Internet yet, and there were almost no books about the pop music, not to mention folk or classical music, that was flourishing outside the English-or European-language zones.

  I seem to remember one day when Jon played some Milton Nascimento records, which I didn’t get at the time—it took years until I did. Brian and I shared a fascination with African pop-music, although aside from Fela Kuti we could find little information about any of the artists whose records we’d stumbled across. Pop music in many other countries, we discovered, came in wildly different flavors, and it wasn’t yet a suitable subject for ethnomusicologists. The Lebanese-Egyptian singer Farid al-Atrache was a big favorite, as was duduk player Djivan Gasparyan. I had some cassettes of Balkan brass bands and Ghanaian pop groups. We passed around our records, lugging vinyl from coast to coast, apartment to apartment.