In preparation for making a bunch of new YouTube videos, I have been thinking about Anne Danielsen’s distinction between songs and grooves. It’s a useful scheme for thinking about pop, but it doesn’t cover everything in Western music. We need a third category for linear through-composed music. So here’s my proposal: all of the music in our culture falls into three big overlapping categories: pieces, songs, and grooves.
- A piece is linear: a series of non-repeating events that occur in a specific order.
- A groove is circular: a short cell that repeats an indefinite number of times, without any larger-scale structure.
- A song is in between: a linear arrangement of circular elements.
The categories are not perfectly distinct. Think more in terms of a continuum. On one extreme, you have total circularity, an infinite loop of a breakbeat or drum machine pattern. On the other extreme, you have total linearity, a serialist composition without any repetition at all. All Western music lies somewhere on this continuum. (All other music probably does too, but I don’t know enough about everyone else’s culture to be able to speak confidently about it.)
The Piece category includes most European classical compositions, especially as we move closer to the present and away from Baroque dance forms. There is no large-scale repetition and little medium-scale repetition. Pieces align well with the Western classical work-concept. They are often compatible with Schenkerian analysis. Examples include:
The Song category includes most pop, rock and country tunes. Individual sections or phrases are usually grooves, and these sections are organized into a linear structure with a beginning, middle and end. Songs might align with the work-concept, or they might not. Examples include:
- “Kiss” by Prince
- “Wellerman“
- “Children of Production” by Parliament
- “Greensleeves“
- “Eleanor Rigby” by the Beatles
The Groove category includes most funk, rap and dance tracks. These may have nominal sections, but they are structurally identical except for top-line melody and arrangement/timbre. Harmony is static and often blues-based. Melodies are short and circular, if they are present at all. Grooves align poorly or not at all with the Western work-concept. Examples include:
- “Oye Como Va” by Santana
- “Green Onions” by Booker T & the MGs
- “Sonnymoon for Two” by Sonny Rollins
- “Don’t See Us” by The Roots
- “Get Ur Freak On” by Missy Elliott
- “Chain of Fools” by Aretha Franklin
Here are some intriguing edge cases.
- “Keep On Truckin’” by Eddie Kendricks is a collection of excellent grooves organized into a nominal song structure.
- Bach’s Chaconne and Passacaglia are pieces, but they rest on groove foundations.
- “My Favorite Things” is a song in The Sound of Music, but John Coltrane turned it into a groove.
- In a Silent Way by Miles Davis is a set of killer grooves edited together into a large-scale piece. I can’t think of too many other examples of this idea!
- Taken together, do the solos in a jazz tune comprise a groove, or a piece? Opinions differ.
Beyond my academic interest in organizing knowledge, there are practical consequences for music educators in these categories. Each category has its own optimal pedagogical, analytical and performance methods. Traditional music education approaches everything as if it’s a piece. That sometimes works with songs and sometimes doesn’t, and it hardly ever works with grooves. You can notate pieces, you can sort of notate songs, and you can’t really notate grooves. Notation is usually the best way to learn a piece and usually the worst way to learn a groove. You rarely hear Western common-practice tonality in a groove, and you will often hear blues tonality. And because grooves are historically most prevalent in the traditions of the African diaspora, our systematic neglect of them in music education is symptomatic of the field’s structural whiteness.
Update: I created a new version of my graphic showing the most common recording methods for pieces, songs and grooves.
Pieces are most clearly represented by notated scores. You can record them, and even learn them by ear, but the “real” composition is on the page, and any given performance is derivative of the Platonic ideal of the composition. Grooves are most clearly represented by recordings. Notation can only represent the bare skeleton; the specific performative nuances are essential to the music. Songs are in between: they have more structure that you can represent on the page, but there is still essential information that can best be represented by recordings. This is why written representations of songs are usually just memory aids rather than attempts at complete documentation.