Is there a way to teach counterpoint that is less of a drag?

The other night, the family and I went to a Handel’s Messiah singalong at Lincoln Center. This is an enjoyable holiday tradition where professional singers, conductors and musicians perform the Messiah, and the audience sings along for all the choral parts. You have to bring your own score, and this attracts a very specific kind of person, which my wife is proud to be. The Hallelujah chorus is the famous part but “All We Like Sheep” is the real musical high point.

I’m not enough of a sight singer to be able to keep up with any of the complicated parts, but it’s definitely fun to be in a big crowd doing all that polyphony, and it’s not too difficult to find chord roots to add to the mix.

Handel’s music is very beautiful if you are in the right frame of mind. The thing is, unless I am physically at Lincoln Center in the middle of the crowd, I am not often in that frame of mind. Handel is hard for me to connect with the rest of the time. My students don’t feel much enthusiasm for him either. When we listen to 18th century music together, the general consensus is that it’s too tidy, it describes a world where everything makes sense, where good triumphs over evil, where life is a narrative with a meaning. This is, to put it mildly, not reflective of my or my students’ inner life. I do know plenty of people who love Baroque music, including my parents. However, my guess is that present-day Americans are attracted to Baroque music as a portrayal of a different and better world, not as a portrayal of the world that we ourselves live in. I sometimes feel the pull of that clockwork universe too, but not often.

My inability to emotionally connect to Western European aristocratic culture explains why I had such a bad time studying music theory in formal settings. In the times and places when I was studying it, the only thing on the menu was a formalized version of the rule set underlying Baroque music and the styles immediately following from it, and I did not enjoy studying that at all.

One of the main things you do in the standard Western tonal theory sequence is SATB part writing. The letters stand for Soprano, Alto, Tenor, Bass, representing the four voice parts in a choir: women with high voices, women with lower voices, men with higher voices, and men with lower voices. Your job in part-writing is to create a melody for each of the four voice parts that outline some given chord sequence. Below I show a standard SATB part-writing problem from a music theory web site. The chords are Cm, Fm, G, Cm, with a rhythm of one half note per chord.

Here’s a correct solution—click the image for the MuseScore file.

What makes this a correct solution? Well, at a minimum, I had to get the correct notes for each chord: C, E-flat and G for Cm; F, A-flat and C for Fm; and so on. But you can’t just spell each chord and be done. The challenge is to organize those notes so that each voice part is singing a linear melody that doesn’t jump around too much. So in my solution, the bass part might move by fourths or fifths, but the other parts never move any interval wider than a third. This is called voice leading, and it’s essential if you want your movement from one chord to the next to sound smooth rather than jagged.

This might actually sound kind of fun, if you have a certain kind of personality. (I do have that kind of personality.) However, voice leading is not enough for SATB chorales. There are many more rules about how the notes are allowed to move. The most notorious of these is the prohibition against parallel fifths. Let’s say that in the first chord, the tenor is singing C, and the soprano is singing G, so they’re a fifth apart. On the second chord, you are not allowed to have the tenor go up to F while the soprano goes up to C, because then they will still be a fifth apart. In Baroque chorales, every line is supposed to function as an independent melody. Parallel fifths sound like a single thickened line, not like multiple lines in counterpoint. There are many more rules of that kind, and obeying those rules is what makes SATB chorale exercises so hard, and for me, so tedious.

If you’re a rock musician, you are already outraged reading this, because, forbidding parallel fifths? Rock is nothing but parallel fifths! They’re called power chords! They sound great! Responsible teachers of traditional Western tonal theory will say, yes, power chords sound good in that context, but in this class we are learning to work within the conventions of this specific historical idiom. Less responsible teachers just say, parallel fifths are bad, end of story. This is why rock musicians tend not to like music theory class.

So why do we make so many theory students write counterpoint in strict 18th century style? Some people love the music that results from strict application of that style. I think it sounds stiff and corny, but not everyone does. Others will say, okay, fine, Baroque chorales are very distant from the music that people are listening to outside of class, but there’s pedagogical value in learning to work within their constraints. So the question is: what are the costs and benefits of making everyone learn the specific constraints of Baroque chorales?

Let’s go back to the SATB exercise. My solution above is correct, but I think it sounds boring and clunky. If I wanted to voice-lead through this chord progression in a way that sounds good to me, here’s how I would do it. Again, click the image for the MuseScore file.

Now we’re talking! If you are doing four-voice counterpoint with triads, you have to double notes. But if you use seventh or added-sixth chords, then you don’t have to double anything. And if you leave out the fifths, you can include ninths. I’m especially pleased with the rising chromatic line in the soprano. Are there parallel fifths in here? Who cares?

I didn’t study music theory in college because I took one look at the curriculum and thought, no thank you. By the time I got to grad school, I had spent 15 years playing rock and jazz. I was ready to suck it up and learn SATB but I had a terrible time, because my musical instincts were always wrong-footing me: “incorrect” counterpoint sounds good, and “correct” counterpoint sounds bad. I could only get through these exercises by not using my ears at all, and instead by approaching them as a kind of Sudoku puzzle. It was not ideal. I’ve spoken to many pop musicians who had the same experience.

But Ethan. We can’t just let students write whatever they want. They don’t know what they want, and even if they do, they don’t know how to get there. You can’t expect kids in Theory II to know seventh and ninth chords. Though, why not? If we’re expecting them to understand figured bass and augmented sixth chords, then seventh and ninth chords are no big deal, especially if learning about them gets you to the sounds in your head.

Strict 18th century style is a finite and bounded rule set. It’s so finite and bounded that computers have been able to write chorales for decades. But does learning the rule set prepare you for composing, arranging and analysis? Well, it depends what music you are going to be composing, arranging and analyzing. I read a collection of articles from 2021 that address this question by Chelsea Burns, William O’Hara, Marcelle Pierson, Katherine Pukinskis, Peter Smucker, and William Van Geest. In the introduction, the authors say:

Four-part chorale-style writing and analysis are so central to music theory pedagogy that it is difficult to imagine a curriculum without them. Yet, as much as music theory instructors rely on chorales to teach general principles of music, chorales are specific in structure, and their origins lie in a specifically German religious tradition. This raises some important questions: are chorales effective for the purposes to which we put them? What pedagogical possibilities do they obscure? What can our reliance on this genre tell us about North American music theory and its commitments or anxieties? In sum: are we right to depend on SATB writing as we do?

It depends on the purposes, right? If you want people to be able to write like Bach, it makes sense to practice writing like Bach. As American music theory pedagogy slowly decanonizes and aspires to greater inclusion, however, then it’s harder to justify SATB. One argument for its centrality is that, okay, maybe Bach’s chorale writing doesn’t underlie everything that a person would want to learn, but it’s still the foundation of the building. So trust us, hip-hop beatmakers and jazz improvisers and singer-songwriters, the relevance of what you’re studying in the theory core will become apparent. Another argument is to say, we have this whole teaching infrastructure built around this rule set, and switching it out is going to be a huge amount of work, and besides, what would we even replace it with?

Let’s put a pin in that question for a minute and look at Marcelle Pierson’s article. She distinguishes between the concept of voice-leading generally and the conventions of Bach chorales specifically:

I take the most general definition of voice leading to be parsing simultaneities into separate “voices” and thinking about how those voices move both individually and in relation to one another. This broad concept would apply equally to a number of different traditions; one can discuss, say, the parallel fourths and fifths of heavy metal chords or the characteristic voicings of traditional jazz piano accompaniment. When we discuss voice leading within the undergraduate theory curriculum, however, the definition becomes far more specific. The Grove Dictionary of Music is representative of this specificity: it collapses voice leading into the article for part-writing, putting voice leading in brackets after the subject entry to indicate these terms are interchangeable, and furthermore insists “each line must . . . have a melodic shape as well as a rhythmic life of its own” in its first sentence. It then defines a number of the voice-leading rules familiar from music theory textbooks as exemplary of “good part-writing.” This definition reflects the reality that we almost never discuss voice leading in a value-neutral context. Instead, we teach our students “good” voice leading—the kind of note-against-note movement that will result in something stylistically akin to Bach’s music, and particularly his chorale harmonizations.

If you view Bach’s music as the foundation of everything worth learning, then writing Bach-style chorales is a self-evidently effective way to learn music theory. American music academics did feel that Bach was the foundation of everything worth learning for a long time, and many still do, if only implicitly. There are plenty of arguments about how Bach is universal. Pierson refers to a David Huron article from 2001 that explains Bach-style voice leading in perceptual principles like “tonal fusion”(the tendency to hear octaves and fifths as single sounds, not stacks of distinct pitches) and “pitch proximity” (the tendency to hear proximate pitches as continuous lines).

Huron’s reasoning relies on cognitive processes that are natural or innate and thus argues implicitly for the inherent superiority of Bach-style voice leading, elevating it to something more fundamental and less contingent than a style. Huron is careful to note throughout the essay that one might pursue different goals than individuated voices, and he also notes several perceptual principles that are not capitalized upon in the Bach-style voice leading precepts. The essay, however, lacks any kind of comparative element; the possibility of goals other than individuated voices within a tonal context is acknowledged, but not pursued. It assumes that there is a “natural” way to listen to music, when in fact the meaning of musical structure is (at least in large part) culturally determined. One must question a methodology that leads to the conclusion that something as complicated and specific as Bach-style voice leading is in any way natural or innate; my experience has been that it takes many hours of intensive training to get students to begin to hear parallel fifths in four-voice textures, indicating that this skill is far from inborn.

I learned how to hear those things too, but it was not easy or natural, even though I grew up in a classical-music-listening household. It’s just not the way music works in my listening, playing and producing life.

Outside of arguments about cognitive processes, one argument for Bach-style voice leading points toward its ability to orient students within a particular culture. This argument proceeds along the lines that it is important for students to be able to create stylistically appropriate music in a particular genre (any genre really, but we happen to focus on common practice classical music), thus justifying our practice of model composition.

Yes, we just so happen to focus on that music. There are good practical reasons to focus on it! The European canon is systematic and well documented and it’s something you can build a curriculum around. But against the practical benefits, we have to weigh the fact that it’s the music of an alien culture.

This came home to me vividly last month when I went to Portugal for a seminar with some Western European conservatory faculty. For these folks, there is no question that they should be teaching the Western European classical canon. It’s their music, and they are serene about its central place in their teaching. In the US, it’s more complicated. American institutions historically considered the European canon to be “our” music as well, or at least, the music that we were supposed to aspire to. That reflected American culture more broadly, where the European canon had a substantial footprint, and not just among academics or intellectuals. Regular people routinely listened to opera. However, that time is long in the past. So now what are American institutions supposed to do?

Some music schools and educators see their job as protecting and preserving the European canon regardless of what’s happening in the culture generally. Others want to prepare students to perform, create and understand a wider variety of music, but they think that studying the canon is still the best foundation for learning. This second attitude is becoming the predominant one. So, how well is that working?

One unassailable application of Bach-style voice leading is toward a replication of the complexities of Bach chorales, and more broadly, I have found that students immersed in Bach-style voice leading can apply these skills fairly directly to arrangements for a cappella and church singing groups. The applications of Bach-style voice leading can reach beyond these limited examples, but not without adjustment. It can be a useful launchpad for the study and creation of works in Baroque-style counterpoint, such as Bach-style fugues and inventions, although one must discuss rhythmic figuration and form before students can create these kinds of pieces on their own.

Okay, great. That seems like a niche application, though.

One can also follow Bach-style voice leading in many common-practice piano sonatas by composers like Mozart, Haydn, and Beethoven, although one must account for a slower harmonic rhythm, different rhythmic figuration, and the disappearance and reappearance of voices at different points. These are the most common applications in the traditional theory sequence, and it is worth noting the lack of demographic diversity in this paragraph thus far.

You might say that.

There have been robust attempts to find examples of Bach-style voice leading outside of this demographic, usually extending to isolated works written by white women and sometimes people of color. These projects improve representation in the music theory classroom—a valuable contribution—but they do little to move outside of the framework of valuation exemplified by the collapse of music theory pedagogy into voice leading in general and Bach-style voice leading in particular. They risk substituting tokenism for real, structural diversity work.

This is ideologically problematic, but also practically problematic. If you want to learn about the music of John Coltrane or Aretha Franklin or Kendrick Lamar, there are an awful lot of steps between Bach chorales and where you are trying to get to.

In Western pop music, especially music outside of progressive rock, 7 is often lowered; tendency tones become unruly and, in the case of sevenths, often devoid of tension; blues scales introduce their own logic of tendency and relation between melody and harmony; power chords and piano arrangements delight in parallel motion; inverted chords are a rare occurrence; etc. The set of adjustments becomes even more dire, of course, once we venture outside of triadic music altogether.

Right, and maybe students want to learn about how things work in blues and pop! They don’t just know it automatically. You can figure out the broad strokes of an R&B song easily, but the broad strokes are not adequate. The details matter, and they are complex, and you can’t just generalize out from Bach to get to them.

I have deliberately presented this set of adjustments in a provocative way, in that it assumes that Bach-style voice leading is the origin point of voice leading, if not its enduring gold standard. While this is not necessarily the way that music theory instructors conceive of its status, it would be easy for undergraduate students to get the sense that it is, given the emphasis on it in textbooks and tonal theory curricula writ large. Framed as such, music outside the common-practice canon presents itself as badly behaved, or, worse, as less valuable.

I definitely had professors use the language of “good” or “correct” to describe Bach-style voice leading, and “bad” or “incorrect” to describe everything else. A low point was when a theory prof used the intro to “A Hard Day’s Night” by the Beatles as an example of bad voice leading, and we were all like, oh yeah, I love this song. If she was trying to get us to hear that intro as sounding bad, she failed; hearing that recording made the whole room light up.

is the amount of time we spend on an intimate knowledge of inversions, or the melodic minor scale, or cadential closure, commensurate with the importance of these concepts in our students’ musical lives? Is this the musical knowledge that will most help them continue with a life in music, professionally or otherwise?

Well, it depends which students we’re talking about. This semester, I taught a graduate seminar that was mostly classical instrumental performance majors. They only plan to play and teach the European canon, so they naturally only want to study the European canon. Meanwhile, I also taught two undergraduate classes that were full of songwriters, beatmakers, audio engineers, pop and theater singers, and jazz and rock musicians. There were a few classical music people too, but not many.

NYU undergrads are so broadly hostile to the European canon that I find myself in the unexpected position of advocating for it in class. “Guys, Bach is actually way cooler and more relevant to your goals than you realize.” “You should go to Carnegie Hall once in a while even if you think it’s boring.” I have developed some good strategies for getting them interested. During one class session, I played a recording of a Debussy prelude while making a point about rhythmic phrasing.

One of the beatmakers told me that he went home and made a track from samples of this piece. That felt like a victory.

In the US theory pedagogy literature, there is a lot of disagreement about how to teach because it’s lying on top of unresolved questions about what to teach. The current consensus seems to be, well, we have to teach the canon, but we also need to bring in other musics. So the question is, to what extent is studying the canon an effective way to learn about non-canonical music? The moderate progressives say, maybe Bach chorales are not a universal skeleton key, but of course we should still center them, we just need to make space alongside them for other stuff. But what other stuff, and how much of it? Do we bring in rock, hip-hop, gamelan, West African drumming, raga, maqam? There’s no consensus. The wild-eyed radicals like me think that Bach chorales should be available to study, but as an elective. Unfortunately, the radicals also don’t agree about what should be in the core, or whether there should even be a core.

Without a consensus about goals, you can’t have a meaningful conversation about methods. Should we require SATB part-writing for all music majors? Well, what’s our goal? If the goal is to teach the European canon, then of course we should. If the goal is to teach beatmaking and songwriting, then of course we shouldn’t. We can’t decide until we agree on our goal. In my department, the goal is to teach the canon *and* beatmaking, which is a noble aspiration, but very hard to put into practice. So we have to decide how much SATB the beatmakers need, and how much beatmaking the aspiring symphony cellists need. My pop-oriented undergrads are clear: they do not see any possibility of performing or teaching classical music, so whatever we teach them, it shouldn’t be that. But the classically-oriented grad students are equally clear: they only expect to perform and teach classical, so learning beatmaking is a waste of their time.

My dislike of the canon softened once I made it out of the graduate theory sequence. I do think there’s a way to teach that music to people like me that isn’t so soul-crushing. We could treat it more like learning Latin: a historical specialty, something that could help anyone who is interested but not as the sole gateway to other knowledge. I like Robert Fink’s framing of the canon as an “ancient alien power source”. There is much in there of use, even to the beatmakers. But it’s necessary to approach it as alien, not as “our” music.

If I was in charge of everything, I could come up with a theory and aural skills sequence that I think makes perfect sense, but it would reflect my own biases and mostly focus on jazz, R&B, and 80s and 90s hip-hop. I would do a terrible job introducing the European canon (except, I guess, as a sample library.) I would teach counterpoint and voice leading from the pop and jazz perspective: “Mostly use chord tones on the strong beats, unless you’re going for an effect; try not to have voices jump around too much, unless you like how that sounds; you’re never more than a half-step from the truth; trust your ears and have fun out there.” It would all be based on vibes. I think everyone would have a great time, but I would skip a lot of material that other people value. I ultimately believe that there doesn’t need to be any single core set of skills that every college music major needs to have. But I also recognize my belief to be a fringe one.