Songs vs Grooves

Anne Danielsen’s book Presence and Pleasure: The Funk Grooves of James Brown and Parliament is one of my favorite works of musicology. In the book, Danielsen distinguishes between songs and grooves. “Yesterday” by the Beatles is a song. “The Payback” by James Brown is a groove.

In structural terms, a groove is a small musical cell that repeats indefinitely. A song is a hierarchical organization of smaller cells that form a linear sequence with a beginning, middle and end. The lack of large-scale structure in a groove makes it effortlessly malleable and extensible. Want to make it thirty seconds longer? No problem. Want to make it thirty minutes longer? No problem. Songs are not so flexible. If you wanted to make “Yesterday” longer, would you… make up more verses? Repeat the bridge again?

Songs are amenable to being covered and arranged. Grooves are amenable to being remixed and sampled. The art of dance remixing is the art of turning songs into grooves. Or, more accurately, it’s the art of finding the groove that was always latent within the song.

You can easily notate songs. You can’t easily notate grooves. I mean, you can write them out, but you won’t be able to represent the most salient aspects, the microtiming variations and so on. As a consequence, copyrighting a groove is much harder than copyrighting a song. You can copyright a particular recording of a groove, but there is no copyright protection for the abstract idea of the groove independent of a recording, the way that there is for the abstract idea of a song.

In classical music terms, you can understand a song to be a “piece” or a “work,” and some educators and musicologists insist on referring to songs that way. However, the work-concept is a poor fit for the concept of a groove. You can’t really separate the groove of “The Payback” as an abstraction from the specific way that James Brown’s band plays it.

American popular music is a blend of songs and grooves. Understanding the difference between them is crucial to understanding our musical culture. To grossly oversimplify: the song-ier parts mostly descend from Europe, and the groove-ier parts mostly descend from Africa.

Here’s an iconic track that is often referred to as a song, but is really a groove:

There is a fragment of a song here in the early lines: “I know a place, ain’t nobody cryin’,” etc. But after a few lines, Mavis Staples starts just improvising while the rest of the Staple Singers repeat the line, “I’ll take you there,” and from there the track settles into an open-ended groove, without any particular structure but deep in the pocket.

Early rock was a groove genre. For example, “Bo Diddley” is a pure G Mixolydian groove.

Rock as practiced by the Beatles and subsequent white groups has used groove as a substrate for songs. I like grooves, so I find rock songs to be frequently annoying, because too many of them break up or disrupt the groove in order to be more “interesting.” For me, rock is at its best when it stays close to its Black roots and lets the groove open up. Consider the Beatles’ “Hey Jude.” It’s a lovely song, but the part that really grabs my heart and doesn’t let go is the ending, when it transforms into an epic groove.

Rock songs might be repetitive, but they still form linear structures that are intelligible within the goal-directed and developmental framework of Western art music. The modern band movement has shown that classically-trained music teachers are willing to embrace rock songs as performance repertoire. Rap, on the other hand, is having a tougher time finding acceptance. When Ben Shapiro’s music theorist father who went to music school says that rap is “not music,” he isn’t just reacting to the (supposed) lack of melody. He’s also reacting to the groove structure and his inability to make sense of it.

Anne Danielsen thinks that white people don’t know how to listen to grooves:

[T]he focus on goal-directedness and developmental forms within white Western musical traditions also expresses its dominant listening strategy. Not only can the music be said to be of a teleological kind, but even the listening to it is in a certain way teleological: one expects that something will happen, certainly regarding the aspects most often involved in shaping such large-scale time spans, namely melody, harmony, and musical form. When a funk tune like “The Payback” encounters such a listening habitus, the incongruity between the mode of production and the mode of reception may hinder the engagement in the groove, because there is simply nothing else to the tune (p. 191).

In other words, if you listen to a groove expecting large-scale structure, then you miss out on all the rich structure present within the repeating cells. This kind of listening requires practice. I was listening to a Babatunde Olatunji recording in college, and my roommate complained that he didn’t like it because there was no “beat” and no “melody.” It was nothing but drumming and singing! He meant that he didn’t like the absence of linear structure.

The best way to learn how to listen to grooves is to dance to them. Pop and dance music are groove vehicles. Very often when people think they like a pop song, it’s because they really like the groove underneath it. The charts have always been full of songs with silly or meaningless lyrics that people (rightly) love because they have strong grooves.

The KLF explain this concept in their guide to writing a number one hit:

The club D.J. (like his forerunner the dance band leader of the thirties, forties and fifties) realises that the most important thing is keeping the dance floor full and the thing that keeps the dancers dancing now (as it was then) is the music with its underpinning groove factor. Singing throughout has always just provided a distraction from the main event – what is happening on the dance floor and not on the stage.

Music that’s meant to be listened to alone can be lyrics-oriented and built around a vocal sound. But social music needs to center on the beat.

The idea that grooves are less “intellectual” or substantive than songs has a long and racist history, and it’s an idea that persists right on up through to the present. Every year I ask music tech students to make dance music using loops, and the instrumental performance majors always give their tracks sarcastic names: “Cheezy Club Track,” “Dumb House Thing,” and so on. There’s this intense anxiety about daring to take dance music as seriously as they take their “real” music. The entire concept of the “guilty pleasure” in music is based on the anxiety that white people feel when we dare to enjoy a groove.

Speaking of music teaching: the song/groove distinction has major implications for music theory, because harmony works differently in grooves than it does in songs. When music theorists say they can analyze pop music in terms of Western European harmonic theory, they mean they can analyze songs that way. For grooves, you need something different. Philip Tagg’s theory of loops is the best explanation of groove harmony that I know of.

We also need a better theory of blues harmony to make sense of groove, since blues and groove go together so often. This is what Tony Bolden argues in his book Groove Theory: The Blues Foundation of Funk. Western musicology has given a lot of attention to the harmonic structure of twelve-bar blues, but hasn’t thought enough about it as a groove structure.

The I, IV and V chords are present in a standard blues tune, but they aren’t structural cornerstones the way they are in European music. Old Delta blues singers are especially free with the chord changes. They move them around, anticipate or delay them, and sometimes omit them entirely. The blues scale further undermines the structural importance of harmony, because it ignores the underlying chords completely. This is why the blues scale is such a reliable crutch for inexperienced improvisers. It’s a universal solvent! It fits every possible chord progression, as long as it’s in a groove context.

“Boogie Chillen” by John Lee Hooker embodies the groove structure of the blues. It has no chord changes, unless you count the harmonized blues scale in the solo section, which I don’t. It’s no coincidence that Hooker spends half the song rapping. The rhythms of this music might be different from hip-hop, but the groove structure is the same.

When jazz musicians play the blues, they do often bring in ideas from European harmony: they insert functional cadences, and outline the chords as they go by. But it’s important to realize that jazz soloists and composers don’t have to do any of this; they are also free to just play the blues. Count Basie and Duke Ellington each wrote many iconic blues tunes with melodies that pointedly ignore the chord changes.

Jazz certainly has its song-y aspect as well, and this aspect is fascinatingly in tension with the groove aspect. College-level jazz pedagogy focuses almost entirely on the song aspect, which is unfortunate, but also understandable. You can write exam questions about reharmonizing a rhythm changes bridge. You can’t write them about Charles Mingus’ microtiming.

The really interesting cases are the ones where groove-based harmony interacts and conflicts with song-based harmony, as in “China Cat Sunflower” by the Grateful Dead. A real theory of groove harmony is going to have to consider time (musical and literal.) If you are in G, but you sit on a D7 groove long enough, you “forget” about the G and start feeling the key center to be D Mixolydian instead. I’ve never seen this idea addressed in a formal theory text, but it’s central to the way I understand key centers in grooves.

Not only would a theory of groove harmony make better sense of funk and blues and half of rock; it might also retroactively make better sense of certain pieces of Western classical music too. For example, in Chopin’s Nocturne Op. 9 No. 1, there’s a long passage on an extended Db7 chord. A Schenkerian named Jacob Gran argues that the chord is acting as a V7 chord for a fleeting Gb much later in the piece. I am unconvinced. Let’s call it what it is: a straightforward Db Mixolydian groove.

My favorite canonical classical work is the Bach Chaconne, because underneath its surface complexities, it’s fundamentally an extended dance groove. I would love to hear Baroque dances played with actual dance rhythms. This is why I keep adding beats to them.

In my experience, grooves are harder to play well than songs. If you have a good song, the interest of its structure can easily conceal weak musicianship. In a groove, there is nowhere to hide. Anne Danielsen again:

[T]he relation between the pure form of a groove, with its repetitive structure and cuts and breaks, and funk’s main musical “challenge,” namely the subtle perfection of the basic two-bar pattern, is not accidental. However, this also means that, as a musical form, the pure groove hides no weaknesses. When all attention is on the groove and almost nothing else is happening, even small mistakes become distinct. The pattern has to be kept steady and relaxed all the way through, and this requires quite extraordinary musical skills (p. 190).

I hope that in learning to understand groove-based music better, we also start giving its practitioners the respect they deserve. Calling down the mothership is fun, but it is also an essential emotional survival technique.