Teaching dynamics and loudness

When I cover dynamics and loudness in music theory class, I only spend a small part of the time talking about forte/piano, crescendo/diminuendo and so on. Once you have the Italian translations, those terms are self-explanatory. They are also frustratingly subjective, and they refer only to unamplified acoustic music. To understand dynamics in the present day, you need to understand decibels, perceptual loudness, and what “dynamics” mean in the era of recorded, amplified and electronic music.

First of all, let’s talk about the classical terms a bit. When you see that classical music uses “piano” to mean “quiet,” you might naturally wonder what the relationship is to the instrument. The piano’s inventor, Bartolomeo Cristofori, originally called it “un cimbalo di cipresso di piano e forte,” meaning “a keyboard of cypress with soft and loud”. This got variously abbreviated as fortepiano, pianoforte, and finally, its current name. This was all in contrast to the piano’s predecessor, the harpsichord, which can only play at one volume.

Wind instruments are capable of remarkably subtle dynamics. Listen to “Black and Tan Fantasy” by Duke Ellington for lots of examples of dynamic modulation. The melody at the beginning is a simple one, but it’s rich with dynamic, pitch and timbre inflection. And listen to the beginning of the trumpet solo at 1:00–it’s one note held out for almost four bars, varying only in dynamics and timbre.

It’s also fun to try to reconstruct the amplitude curve of the long notes in the melody of Miles Davis’ “Sid’s Ahead.”

Next, consider the nuanced and colorful dynamics you can impart on an electric guitar using a volume pedal. Listen to “Steel Guitar Rag” by Bob Wills and the Texas Playboys: right at the beginning of the track, the pedal steel fades in from silence.

Like all electronic and amplified instruments, the electric guitar separates performance intensity from actual loudness. If you want your acoustic guitar to be louder, you have to exert more force on it. However, if you want your electric guitar to be louder, you just turn it up. Pete Townsend’s windmills are for show. Most of the time, heavy metal shredders are barely even touching the strings.

We hear screaming and distortion as “loud,” because for all of human history until very recently, the only way to produce a distorted timbre was to physically scream, or to blow too hard on your wind instrument. You can hear John Coltrane screaming on the saxophone a lot in the last few years of his life.

However, in recordings, you can scream or distort at any volume you want. For example, listen to “Give It Up Or Turnit a-Loose” by James Brown.

James Brown screams his head off over the course of the song, but he’s not even one of the top five loudest elements in the mix. (To my ears, the loudest elements are drums, horns, guitar, bass, congas, in that order.) You might think JB is the “loudest” element of the track because he’s giving the most intense performance, whereas the rest of the band is playing energetically but with calm focus. But in funk recordings, the dominant elements of the mix are the bass and drums.

For a strong contrast, listen to “Let It Go” from Frozen. There, the singer Idina Manzel is both the most intense performer and the actually loudest element of the mix, aside from the orchestra in a few spots. There are drums in this track too, but they are deeply buried.

Musical theater balance sounds weird and lopsided after listening to a lot of James Brown. And JB sounds weird and lopsided after listening to lots of musicals.

Another useful point of comparison: “Simply Beautiful” by Al Green. He sounds like he’s singing very quietly, because in the booth, he obviously was.

But because he’s singing into a hyper-sensitive ribbon mic, and because everything else in the track is mixed so low, Al Green’s voice on the recording is actually putting out a dramatically higher sound pressure level from the speakers than James Brown’s is at the same volume setting. That is counterintuitive!

It is difficult to explain how decibels work without talking about logarithms. (Music students tend to be allergic to math.) The main thing to understand is that going up ten decibels doesn’t mean that the sound gets ten percent louder; an increase of ten decibels means that you are multiplying the sound pressure by ten. This results in the music feeling about twice as loud. It’s confusing.

Finally, let’s compare two strategies for producing recorded music in a noise-polluted world. First, there is the “sonic maximalism” approach, as heard in “Call Me Maybe” by Carly Rae Jepsen.

In a current pop song like this one, everything is layered, compressed and limited to within an inch of its life. Every possible nook and cranny of the frequency spectrum is occupied. Even the pizzicato strings in the opening seconds of the track are compressed to the point of being noticeably distorted. This all sounds perfectly normal to my students, because it’s standard practice in pop. But it sounds strikingly weird to my Generation-X ears.

So that’s one extreme: overpower background noise with brutally maximized caterpillar waveforms. The other approach is the one that hip-hop producers have been using the past ten or twenty years, which I call brutal minimalism. My example here is “A Milli” by Lil Wayne.

The elements of the track are punishingly loud, but there are few of them, and in the case of the drums, they’re punctuated by lots and lots of empty space. There’s empty space in the frequency spectrum, too: the mids are yawningly empty. I demonstrate that by playing the track back loud and speaking over it in my normal tone of voice; the students can hear me fine.

You can practice dynamics concepts by taking a piece of music that’s typically quiet and creating a loud version of it, or by taking a piece of music that’s typically loud and creating a quiet version of it. When I assign this in classes, students submit lots of heavy metal versions of lullabies and holiday carols, and lots of ukulele versions of punk songs. “I Wanna Be Sedated” by the Ramones is a surprisingly effective ballad. And “Hush, Little Baby” played back on a detuned pulse wave synth is a whole mood, as the kids say.

So: what even are dynamics in the contemporary world? They are whatever you want them to be! You can be quiet yet loud like Al Green, loud yet quiet like James Brown, or loud yet somehow even louder like Carly Rae Jepsen, or alternatingly loud and dead silent like Lil Wayne. Perceptual loudness is as much about timbre and density as it as about anything else. It’s a strange and surreal sound world we’re all living in.