Jason Yust on tonality and racism

What is a music theory curriculum supposed to include?

I haven't done any culture war material lately, but Jason Yust recently published an article in the Journal of Music Theory with the title "Tonality and Racism", and I couldn't not respond. The arguments in the paper are relevant to my teaching life in NYU's new and wonderful pop theory and aural skills sequence. These classes are vastly more inclusive and culturally relevant than the extremely white and Eurocentric classical theory sequence that I went through as an NYU grad student. However, we still have some work to do.

In casual language, my NYU students use "music theory" to mean "Western European tonal theory and its accompanying notation and symbolic language." So when they say "[Pop musician] doesn't know music theory", they mean, "[Pop musician] doesn't read notation" or "[Pop musician] doesn't know the conventions of tonal harmony." If I push back, students will quickly self-correct and say that any abstract thinking about music counts as music theory, and that [Pop musician] certainly is thinking theoretically. But I'm concerned about this reflexive usage.

We will get into the specifics of Yust's argument in a minute, but first I want us to understand who we are talking about. This is not just the woke mob gone amok. Yust is a major figure in music theory who has published a ton of scholarship on the Western European classical canon. His book Organized Time: Rhythm, Tonality and Form uses graph theory concepts to show how musical time interacts with harmony and voice leading in canonical works. Here are a few illustrations from the book; the top row of diagrams shows networks representing rhythmic events, and the bottom row shows networks representing tonal events.

The book is hard work! The mathematical terminology is daunting, and I'm not very familiar with the repertoire, so I am moving through it all very slowly. However, whenever I am able to internalize something from it, I find it enlightening.

So, those are Yust's credentials. Next we need to understand what "tonality" is. This is surprisingly difficult for such a foundational theory concept. The usual explanation is to say that tonality is the interaction of a tonic or key center, diatonic scales built on that tonic, triads built from the notes in those scales, voice leading between notes in the triads, and the functional harmonies that arise from voice leading. There is also a sense of different keys interacting with each other in an analogous way to the individual tones within a single key. Tonality is a feature of the Western European musical tradition that the concept was originally intended to describe. The question is whether tonality is a feature of other forms of music as well.

You might wonder why you should care what music is considered tonal and what isn't. The distinction isn't just for musicologists and theorists; it has practical consequences for institutions and the people working and studying within them. At NYU, theory and aural skills classes are divided into three categories: tonal, popular, and non-Western. I am not very familiar with the content of the non-Western classes or the philosophical issues underlying them, so I won't be talking about them here. But my teaching life is very much concerned with the difference between "tonal" and "popular" music theory.

Everyone agrees that Mozart is tonal, but what about James Brown or Missy Elliott? It depends how you define "tonal". If the music has a central tone, with other tones in varying relationships with it, does that make it tonal? That describes almost all Anglo-American pop, more or less. Do those other tones have to form a hierarchy? That still describes most pop. Do the tones interact to create processes of tension and resolution? That describes some pop styles but not others. Do there have to be chords? Songs in the top 40 mostly use chords, but lots of rap and dance tracks don't. Do the chords have to "function"? With each passing decade, fewer and fewer pop songs have "functional" harmony in the Mozart sense. And finally, is tonality a set of features found within the music itself or within the imagination and expectations of the listener? Because listener expectations change as the culture changes, and that can happen fast. I have noticed some substantial differences between my students' harmonic expectations and my own.

Yust argues that the word "tonality" is too culturally and historically specific to be applied outside of its Western European canonical-era context, and wants us to use other, more specific terms instead. For example, very often when we say "tonality", we mean the Western major-minor key system. (That is what NYU teaches in its tonal theory classes.) But Western Europe also has its medieval modal system, its atonal systems, and various folk systems. The United States inherited all the European systems, and developed its own blues system and related jazz and rock systems and more. Other world cultures have a huge variety of other systems. Rather than ask whether music is tonal or not, it would be more constructive to ask whether (or to what extent) it can be described using the European major-minor key system. Mainstream pop mostly can. Jazz usually can but not always. Rock sometimes can and sometimes can't. Blues usually can't. This is a better question than asking about tonality, because it helps us know more specifically which analytical tools we can bring to bear on these musics.

Yust isn't just arguing that "tonality" is imprecise; he wants us to reconsider the word because of its ugly origins. He cites 19th century European theorists like Alexandre-­Étienne Choron and François-­Joseph Fétis, who used tonality as an organizing system for a linear, teleogical theory of music history. Their idea was that tonality represents the culmination of a long growth process, from the primitive music of the "savages" up to the harmonically complex music of "civilized" people.

This history is entangled with a major intellectual project of nineteenth­ century Europe, the pseudoscien­tific defense of colonialism and white supremacy. Fétis adopted one of the main explanatory formulas of white­ supremacist thinkers, a teleological evolution nar­rative in which non­-Europeans occupied the earlier stages in a process that cul­minated with contemporaneous white Europeans. Though largely forgotten by theorists for much of the twentieth century, Fétis’s racist ideas are perpetuated by the conceptual architecture of tonality and the institutional structures built on it (p. 60).

Fétis believed that for most people in the world, music is the "primitive satisfaction of an instinctive, sentimental or traditional need", but that only "modern" (19th century) Europeans have raised it to an art form.

The inhabitants of Europe and those of the colonies founded by them have, in gen­eral, the necessary aptitude for grasping the tonal relations [rapports de tonalité] of certain series of sounds; an aptitude which develops by the habit of hearing music and which is perfected by study, because the law of progress is inherent in the nature of this race. It is through it that they possess the ability to sing in tonal accuracy and to vary the forms of their songs. Savage populations also have the physiological organization by which we perceive the sensation of sound and which allows us to grasp the relationships between sounds, so as not to confuse the intonations and to be aware of their differences, but these sentimental and intellectual operations take place in them within narrower limits, owing to the inferiority of their cerebral conformation. Like the peoples of other races, they also have the memory of sounds and the faculty of reproducing them by the singing voice as they do by the spoken voice, but always imperfectly. Hence it is that their songs are only composed of a small number of determined sounds, which rarely rise above four, and that the steril­ity of their imagination does not allow them to vary the successions; beyond, finally, the remarkable monotony of the songs of all the savage peoples of the earth, par­ticularly those who are cannibals. There is no doubt that the primitive race whose remains have been found in the cave of Chauvaux, on the banks of the Meuse . . . , and whose cerebral conformation was analogous to that of certain tribes of Oceania, has sung in the same formulas as these, and there is also no doubt that if, in a few cen­turies, there are still savage tribes that will not have been modified by contact with white people, their songs will still be what they are today; for, among these unfortu­nate races, there is no progress possible by intuition (Fétis 1869; Yust's translation).

So... that's not great. Aside from his obvious racism, Fétis also has a misguided belief that musical evolution has a direction and a goal, from low to high, from simple to complex. Yust points out that while we may have backed off from the overt racism, we continue to share the unspoken assumption that music evolves toward the goal of greater complexity or sophistication. You can see it in music history curricula that tell a linear story of evolution from church modes to tonality to extended tonality to atonality. Historians of jazz and rock have tended to uncritically echo this evolutionary narrative. It has been very difficult for me to break myself of this mental habit, to think of Louis Armstrong as simply different from Miles Davis, rather than as a "primitive" version of Miles Davis.

The Journal of Music Theory includes several responses to Yust from prominent music theorists, and several of them had the same reaction that I did: maybe tonality as a concept has a racist origin, but does that mean we need to stop using the word? I mean, what idea from Western Europe in the 19th century wasn't racist? Who does it harm if we use "tonality" as a value-neutral descriptor of a particular style from a particular era? Maybe we could just use it to mean "a system of pitches organized around a central tone" and then specify whether we are talking about European classical tonality or blues tonality or bebop tonality or whatever. There are plenty of music theory terms that I find more problematic, one of which appears in Yust's article, in his discussion of whether “Seven Bridges Road” by the Eagles is tonal.

The song uses a standard classic rock progression, D-C-G-D in D Mixolydian mode. Yust points out that just because these chords don't "function" the way that tonal music is "supposed" to, it would be silly to say that the song is therefore nontonal. Instead, we could say that the song uses rock tonality, of which Mixolydian mode is an important component. The thing that jumped out at me is when Yust describes the Eagles song as using "retrogressive" chord movement, as opposed to the "authentic" chord movement you would find in Mozart. What Yust means is that the Eagles' chord movements go flatward around the circle of fifths instead of sharpward like they are "supposed" to.

Yust doesn't mean to say that the Eagles' chord progression is "inauthentic" in the usual sense of the word; quite the opposite. He's using a standard music theory term. But that standard term is so unnecessarily value-laden. Why not just say sharpward and flatward, or clockwise and counterclockwise? I'm similarly bothered by "authentic" cadences. How about "strong leading tone resolutions" or something?

Beyond his argument about semantics, Yust's article makes a profound point about rock harmony. He points out that rock songwriters don't have to use "functional" tonality, because they have different musical tools for establishing the tonic.

Rock music is overwhelmingly a repertoire of songs, and is dominated by chord loops—relatively short repeating harmonic progressions. Lyrics provide clear regular phrase organization, making harmonic phrase­ ending conventions unnecessary. Chord loops automatically generate harmonic expectations, making universal laws of harmonic succession unnecessary for that purpose. Overall, these features lead to harmonic freedom: composers can use whatever harmonic progressions suit other goals, such as supporting a melodic line or differentiating one song from another using harmony. Clear phrasing is enough to define tonal centers—for instance, making it clear that “Seven Bridges Road” is in D major—without the need for functional harmony (p. 80).

There's an interesting bit of Western European music history in here too. Before the conventions of the major-minor key system settled into place in the 1600s, European composers defined tonics using methods more like rock songs: lyrics, rhythm and phrase structures. This music can sound excitingly weird.

It became conventional for European composers to place certain voice-leading gestures at points of structural closure, and then over time, listeners started to hear the voice leading itself as creating the sense of closure. So now listeners who are enculturated in European canonical music hear an authentic cadence like the one at the top of this post as inherently creating a sense of closure. However, it isn't the notes that are creating this sense, it's the way that the notes engage with the listener's enculturation. My theory students, who don't listen to any European classical music and not much functional tonal music generally have quite different expectations.

Anyway, Yust does have a point that words matter. At NYU, the people teaching the tonal theory classes are working hard to be more inclusive, for example by making sure that not all of the examples are from dead white male composers. But if you are teaching the theory of the European canon, that makes inclusion an uphill battle. You could solve the problem by broadening the repertoire beyond the European canon. The tonal theory folks are bringing in lots of songs by, for example, Stevie Wonder. However, that means carefully choosing Stevie Wonder songs that are the most European-sounding in their harmony. You can easily explain "You Are The Sunshine of My Life" in tonal theory terms, but you will have a harder time with "Superstition." The broader the repertoire gets, the more you have to stretch the definition of tonality to accommodate it.

Over on the pop side, we are skimming through everything on the tonal syllabus, plus doing pop-specific material like the blues. So my students have to deal with figured bass and jazz chord symbols, perfect authentic cadences and blues harmony, "You Are The Sunshine of My Life" and "Superstition." It's a lot! It's reasonable to expect pop students to know their tonal theory, but maybe we should split the blues off and make it its own semester. This all depends on what we think the goals of a music theory sequence should be, who it should serve, and what we expect a university music student to know. These are difficult questions, but I am glad we are asking them.

Update: check out this epic response post from Wenatchee the Hatchet