Book Review: How Music Works by David Byrne

I chose to pick up this book when I saw my friend Tyler’s review:

He gave it 5/5. I’ll give it 3/5. There were parts I found fascinating and there were sections that were a trudge to get through.

Here’s a list of the more interesting nuggets…


The first Edison recorders were completely mechanical devices (there was nothing electrical about them). The recording was etched into a wax record. All the elements of these machines existed back in the days of the ancient Greeks. The first recordings that Edison distributed didn’t have the names or photos of the artists, but they did have a photo of Edison himself. He went around the nation holding demonstrations to show how accurate his machines were: on stage there was a live performer and a machine with a recording of the same artist, and he would swap back-n-forth to see if the audience could tell the difference. Byrne can’t imagine that the audience couldn’t tell the difference because the quality back then was crap. He suspects that Edison trained the singers to distort their voice.

In the early recording studios all the artists had to huddle around one mic. The studio had hired hands who were responsible for pushing and pulling the performers into position when, say, it was time to switch from singing to a sax solo.

The old recording technology had a hard time with short/low/loud sounds, such as drums. These sounds would cause very sudden deep or wide grooves in the records, which caused skips during playback. So studios had drums placed way in the back, away from the mic, to soften the sound. Sometimes they threw blankets over the base drums and sometimes they fully replaced them with tubas or other instruments. Musicians in other cities would later hear these recordings and didn’t always realize that the tubas were there due to technical limitations — they thought it was a new style.

The early phonographs were like YouTube, everyone was swapping home-made audio recordings. The next generation of phonographs didn’t have this capability. They did playback better, but they didn’t support home-made recordings. Byrne suspects that this was due to the lobby of the emerging recording industry. It wasn’t until the audio cassette that people regained the ability to record at home again.

The old 78 and 45 records had a limited capacity: about 3.5 minutes per side. Artists began to write songs that fit within that timespan. This improved with the introduction of LP (long play) records. They could accomodate 20-24 minutes of music, the exact length was a function of the music. Loud and low sounds used up more space, so a record with lots of low/loud sounds was shorter.

It used to be that when you thought of a song, you thought of the live performance, and the recording was a secondary version. Today of course that’s flipped — when people think of a song, they think of the recording, and the live performance is a secondary version.

Vibrato used to be frowned upon by musicians. It was like cheating. But that changed with recordings. Artists found that they could be heard more easily if they added some vibrato, and the wobble would cover up mistakes (which would otherwise be there permanently).

“Bing Crosby, the singer who had mastered an innovative use of microphones, was getting tired of having to do his very successful radio show live every day. Bing wanted to spend more time playing golf, but because his shows had to be done live, his time on the links was limited . Crosby realized that by using these new machines to record his shows, he could conceivably tape a couple of shows in one day and then play golf while the shows were being broadcast. No one would know the shows weren’t live.”

Electric guitars were tricky to develop. If you simply slap a microphone onto an accoustic guitar, you get terrible feedback. Various people tried to solve this problem including Rickenbacker and Les Paul. Les Paul’s early guitar was nicknamed “the log” because that’s pretty much how it looked (he removed the hollow chamber). By the way, Les Paul also gets credit for inventing the first multitrack recording.

The walkman was a big deal in many ways. Socially, it marked the first time that you could experience music by yourself. To some people this was like drinking alone. “It was antisocial and psychologically dangerous. It was described as self-stimulation!”

Phone companies had a deep vested interest in figuring out what makes voice recognizable. Why? Because they wanted their existing phone line infrastructure to support as many simultaneous calls as possible. The figured if they could decompose a recorded voice, strip out all the unnecessary bits, and transmit just what’s left, then they could save a bunch of phone line capacity and therefore support more simultaneous calls. Their research lead to digital-based audio technology (and eventually, the CD).

“It was rumored that the length of the CD was determined by the duration of Beethoven’s Ninth Symphony, because that was Norio Ohga’s favorite piece of music, and he was the president of Sony at that time.”

Since the mid-nineties, nearly all music has been recorded using computers. As a result, music now tends to have tempos and rhythms that don’t change.

Recording studios made it a priority to remove as much ambient sound as possible by using sound-insulated rooms. Byrne doesn’t agree with this philosophy; in his view the acoustic properties of the room where a song is played are a part of the song. Also, studios tend to isolate each instrument from the others, the reasoning being that it gives you more flexibility later, to make one instrument louder and another quieter. Byrne had a hard time with this setup at first; playing in isolation took the feeling out of the song. And Byrne is not alone, e.g. the Cowboy Junkies recorded their first album with a single mic for the whole band.


Revenue from CD sales peaked in 1999, and then tanked (see image). Wow.


Recording an album is similar to buying a house: there are huge up-front expenses, so you need to take a loan and pay it back. In the case of an album, the loan is money that the record company fronts the artist in order to pay for studio time, CD production, distribution, marketing, etc. Banks would never give a loan to a kid with a guitar.

Some artists fall into a bad cycle of debt. They sign a contract, get a big loan and use it both for recording and for spending money. Then their record sales don’t go as planned, and the only outlet is to sign another contract and hope the next album does better.

The industry has changed. Record companies don’t give big advances as often as they used to. But some artists still get them. Not long before the book was written, Madonna signed a deal with Live Nation and got a $120M advance.

Byrne gives a detailed breakdown of the costs for his Grown Backwards record (2004):
* He got an advance of $225,000
* Recording costs were $218,000 (which is on the high side, he had a lot of studio musicians)
* 38% – musician fees
* 24% – studio fees
* 17% – engineer mixing fees
* 7% – arranger / copyist fees
* 4% – AFM duties (?)
* 3% – travel
* 3% – net income from advance
* 1% – meals and entertainment
* 1% – equipment, supplies, rentals, repairs
* 1% – freight
* 1% – rehearsal space

How much money does the artist get from each record? As an example, take a CD in a store that retails for $18. Of that, about $8 goes to the retailer, which leaves $10. A typical royalty for artists is 14%, which means $1.40 for each CD sold. If the artist not only performed the songs but also wrote them, he would get 9.1 cents per song for the first 10 songs, so another 91 cents for the album. Byrne believes that artists in the 1960’s noticed that you can earn a lot more if you write your own songs instead of recording other people’s songs, which was the norm. This lead to an explosion of creativity.

Back to the Grown Backwards album, if Byrne wrote every song on this record (he didn’t) he would earn about $2.30 per album, and he would need to sell 100,000 records to pay back his original advance. In reality he ended up selling 127,000 albums, 53,000 digital singles, and 8,000 digital albums. Total revenues were $276,000 and he netted about $58,000. Byrne notes that this money rolled in over a 6-year period, so it’s definitely not a great money-maker. He’s lucky that he’s had other successes and he notes that he always works on multiple projects at once.

Byrne gives a rougher breakdown for another album: Everything That Matters. For this album the recording costs were much lower: $59,850. But, on this album he and Brian Eno decided to follow a DIY approach — they paid for a website, servers, design, promotion, and manufacturing. All told, their costs were $315,000. The album sold 160,000 copies, but this time he and Brian kept a much larger share of each sale, so he netted $324,500. Not bad!

It’s getting harder to sell lots of records…
2006: 35 albums sold more than a million copies
2007: 27 albums
2008: 22 albums
2009: 12 albums
2010: 10 albums
Put another way: of the 97,751 albums released in 2009, only 2.1% sold more than 5,000 copies.

Byrne doesn’t license his songs for commercials, but he does license them for TV and film, and he sees more money from these contracts than from record sales.

In the early 80’s MTV was young and was starving for content. If you had any kind of half-decent video, you would get played regularly. Talking Heads benefited from this — their video for “Once in a Lifetime” got plenty of coverage. In later years the record companies decided that MTV should pay them a fee for the content. As a result, MTV started playing fewer music videos.

Retail stores like Target, Walmart, and Best Buy have a lot of power. They enforce “price ceilings”, meaning they force the record labels to sell them albums at a lower price than they do to smaller shops (which is why small record stores go out of business). Also, they charge record labels money to have CDs featured in a more prominent places in the store (every position is paid for).

Online stores like iTunes, Amazon, and eMusic keep 30% of each track sold. In Europe, people don’t buy digital music as much as in the US.

Copyright for a musical composition is given for a song’s top-line melody, for specific harmonies that support it, and for the lyrics. But the copyright doesn’t include the instrumentation and, notably, does not include the drums. James Brown had a song called “Funky Drummer” that featured a drumbreak by Clyde Stubblefield. That drumbeat has since been sampled by lots of hip hop artists, and Clyde has argued that he deserves some of the revenue from these samples, but legally the copyright belongs to Brown.


Artists create music that fits the space where it will be played. In Europe music was often played in churches, and churches have a deep echo, so composers avoided percussive sounds and tended to favor slow chord progressions. By contrast, music in Africa was played outdoors and had to be loud enough to be heard over the sound of dancing and socializing, so musicians there tended to pick drums.

This sort of thing happens in nature too. Birds that live on the forest floor develop lower-pitched sounds that don’t bounce or become distorted by the ground as much as higher-pitch sounds would.

“The first public concert was in London in 1672. It was organized by a composer and violinist named John Banister shortly after he was fired from the royal band. The price was one shilling, and the audience could make requests.”

Audiences for classical music weren’t always quiet – they used to chat and eat during performances. It was only around 1900 that this be-quiet expectation came to be. Composers took advantage of this and made music that had very quiet passeges — they could count on them to be heard.

Jazz improv began as a way to prolong a song so as to give dancers more time to dance to the same groove. A similar thing happened with with hip hop.

“Playing for the door” is where the venue charges a small fee at the door and this money goes to the band. The venue makes moeny from the bar. This was a new concept in the 1970’s, in venues like CBGB in NYC, where The Talking Heads got their start. It was also unique for CBGB to allow bands to play their own material.

Byrne has heard of situations where bands pay the venue to be able to perform. He calls this “perverted”.


Before recording devices came around, music was something we DID. People played piano at home, sang in churches, or saw live performances. Children in schools were taught to make music. After the advent of recording devices, that all changed. There are still amature musicians, but not as many. Schools focus on teaching music appreciation.

John Philip Sousa was really concerned about this shift. He predicted that record players would kill amature musicianship in America. He noted that when people go to a dance with live music, the performers need to take a break once in a while, and this break is a time to socialize. By contrast, record players are tireless — they don’t need breaks.

The shift to music appreciation also included a subtle twist: schools taught kids to appreciate the RIGHT music: classical. Granted, there have been a few studies that showed a decrease in crime when classical music is played at 7-Elevens, the London Underground, etc. Some people point at this as evidence that classical music is somehow better. Byrne clearly disagrees. He notes that a lot of public money is spent to assist public programs for classical music, but no public money is spent on pop music. Granted, pop music can stand on its own two feet. Still, he argues: “Why not invest in the future of music, instead of building fortresses to preserve its past?”


If you look at a piano, every octave has 12 notes (counting the white and black keys). Pretty much every culture today uses scales that use some of these 12 notes. Western music uses 7 notes (diatonic scale). Chinese classical music uses 5. This is interesting because the human ear can recognize many other tones in each of these octaves, but humans around the world have settled on these few notes in these regular intervals.

These choices started a very long time ago. The earliest evidence we have of cave-man making music is flutes from about 45,000 years ago. These flutes also appear to have been designed with the diatonic scale!

There’s an interesting study made by the Purves lab at Duke University:
They recorded 10-20 second clips of people speaking English and other languages. They then removed all culture-specific sounds out, which basically left just the vowels, no consonents. They then analyzed what kinds of tones are used in speech, and they discovered that even when talking, we all pretty much use the same musical scales (“Seventy percent…were bang on musical intervals”)


“In the West, the presumption of a causal link between the author and performer is strong. For instance, it’s assumed that I write lyrics (and the accompanying music) for songs because I have something I need to express. And it’s assumed that everything one utters or sings (or even plays) emerges from some autobiographical impulse. Even if I choose to sing someone else’s song, it’s assumed that the song was, when it was written, autobiographical for them, and I am both acknowledging that fact and at the same time implying that it’s applicable to my own biography. Nonsense! It doesn’t matter whether or not something actually happened to the writer— or to the person interpreting the song. On the contrary , it is the music and the lyrics that trigger the emotions within us, rather than the other way around. We don’t make music— it makes us. Which is maybe the point of this whole book.”

Byrne often creates a song like this:
1. Create a rough melody
2. Jam with the band, evolve the melody, record it
3. Record vocals over this melody but using jibberish (still no real lyrics)
4. Go back and write lyrics to replace this jibberish, and often try to match the general sounds (ooh’s vs aah’s)


David Byrne tried all kinds of clothing styles in his career. At one point when he was busking he wore a suit and a fedora, and had an unkempt beard. A young kid asked him if he’s “one of those people who didn’t ride in cars”. He abandoned this suit-look for a while, but then he brought it back in an exagerated form during the Stop Making Sense tour ( He says the inspiration came from Japanese Kabuki theater, and from the old adage that “on stage everything needs to be bigger”.

Byrne notes that American rock audiences generally believe that the only way to be “authentic” on stage is to be pseudo-naturalistic (no theatrics allowed), and he generally rejects this. He also spent a lot of time on the details of his stage performances. For example: he would have all metal objects on stage (e.g. cymbal stands) painted black so they wouldn’t steal attention from the performers. And he took advice from a lot of stage performers. For example, he learned that it’s important to let your audience know that you’re going to do something big before you do it (but to do it subtly).

Byrne also did some weird experimental art/music. He once had a performance were he shaved his beard on stage while his partners played accordion and held up cue cards written in Russian.

And he dabbled in other “artistic endeavors”, such as creating questionnaires that he would pass out or send by mail (not many came back completed).


In the 70’s Byrne went to a huge music festival in the UK (to listen, not to perform). It was a multi-day affair so he would pass out from time to time. At one point he woke up and noticed that Led Zeppelin was playing. He flipped over and went back to sleep.

At one point he visited Virgin Records and was given an opportunity to see some early videos of Sex Pistols appearances. He thought they were hilarious: “It was almost a parody of a rock-and-roll band; they couldn’t play, they could barely even stand up.”

Rahsaan Roland Kirk, a jazz saxophonist, would sometimes give out bumps of cocain on little spoons to the folks up front.

Talking about the New York art/music scene he says: “Anything that sounds or looks beautiful would seem to that crowd to be merely pretty, shallow, and therefore deeply suspect— morally suspect even, I found out. Noise, for them, is deep; beauty shallow.”

“Some argue that it was the homegrown Indian cinema that forced that country’s citizens to learn a common language.”

“In my opinion, realness and soul lie in the music itself, not in the scratches and pops of old records.”

About the song “Drugs” he says: “I sang the song after jogging in the studio, because for some reason I wanted to sound out of breath.”

At one point Byrne and his friends talked about making a documentary-style recording featuring the music of an imaginary culture. They never did it, and they later learned that a German composer, Holger Czukay, made his own “Ethnic Forgeries” series.

Byrne asked a Brazilian artist, Tom Ze, to help with a song called “Something Aint Right”. Tom had the percussion section playing Bic pens.

“The online music magazine Pitchfork once wrote that I would collaborate with anyone for a bag of Doritos. 1 This wasn’t intended as a compliment— though, to be honest, it’s not that far from the truth.”

Byrne worked with DJ Fatboy Slim on a musical about the first lady of the Philippines, Imelda Marcos.

“Oliver Sachs wrote about a brain-damaged man who discovered that he could sing his way through his mundane daily routines, and only by doing so could he remember how to complete simple tasks like getting dressed.”

In 2005 Byrne turned a NYC building into an instrument. He brought in a pipe organ and hooked it up with wires to various motors and pipes around the building. He then let people come in and play the building.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s