This year, in addition to teaching my first NYU pop music theory class, I also taught two semesters of pop aural skills. If you didn’t go through a university music program, you may not know what aural skills class involves. Traditionally, you identify chords and intervals by ear, practice sight-singing, and do dictation (meaning, you listen to a piece of music a few times and then notate it.) Pop music is mainly aural, so naturally, aural skills are important. But what should a pop aural skills class look like? Pop musicians definitely need to be able to identify things by ear, but reading and writing notation are less important for them than learning from recordings and improvising.
Before I go into detail about what I have been doing in class, I want to give some context on my own experience of aural skills class as an NYU grad student, back before the pop sequence existed. In a word, it was bad, the worst couple of classes I took in grad school. I came in having played twenty years of rock, jazz and country guitar, and having produced a lot of electronic music in DAWs. I was both massively over and underprepared for classical aural skills. I could easily distinguish a half diminished seventh chord from a fully diminished seventh chord by ear, but I could not sight-sing at all. I still can’t! I have never had to do it outside of a classroom. If you read this blog, you know how much I like transcribing music, but I can’t do it on a limited number of hearings without a guitar or keyboard, my brain just doesn’t work that way. One of the other aural skills teachers was telling me how two thirds of his students failed one of his dictation quizzes, and I told him that I would have failed it too.
It wasn’t that aural skills class was too hard; half the time it was too easy. The problem was that the whole thing felt unrelated to the skills that I had or that I needed to develop. Failing to sight-sing in a roomful of other students was humiliating, and it didn’t make me want to go practice, it made me want to just go major in something else. The whole thing felt like being drilled on Latin grammar when I didn’t know Latin in the first place and was just trying to learn to write better English. Like, sure, if time and money were no object, I could see how learning Latin would benefit me in the abstract, but time and money were very much an object. Not only did I not improve my aural skills in those classes, I left feeling intense resentment toward the instructors. Which is unfair to them; they did a perfectly fine job with the material! The problem was the material. You can see why I would be so glad of the opportunity to give students like me a better aural skills experience.
As in the theory classes, NYU’s aural skills faculty work from a shared syllabus, but we have wide discretion in how we approach the material. Here’s the list of topics.
- Strategies for sight-reading (I told the class straight up that I have nothing to offer for this except just do it a lot, I guess)
- Chord dictation & shuttles
- Large-scale listening & form
- Sus, added, and power chords
- Extended chords
- Pentatonic collections
- Modes
- I, IV, V
- 12-bar blues
- Descending 4ths & 5ths
- Modal harmony
- Other common schemata
- Basslines & chord inversions
- Review of phrase structure
- AABA and strophic
- Embellishing tones
- Suspensions & motives
- Chromatic embellishing tones
- Blues melody
- Improvising countermelodies
- V7/V
- Other secondary dominants and ii-V-I’s
- Modulation ID
- Modulation dictation
- Tritone substitution
- Embellishing chords
- Descending 5th and 3rd sequences
- Other sequences
The students do weekly Auralia exercises, identifying intervals and chords and such. It’s a mixed bag. It’s great that it grades itself automatically, because that saves me and the other instructors a lot of tedious work. However, it’s a frustrating user experience. When it tests your singing, it’s not smart enough to recognize that you might be singing an octave higher or lower than it wants you to, or that your timing might be a little behind or ahead. So students were always complaining that they were being marked wrong for singing exercises that they were doing right. Also, Auralia’s notation entry system for dictation exercises is primitive, making an already anxiety-producing experience even worse. As I develop more and better assignments of my own, I am going to be pruning the Auralia exercises significantly.
I couldn’t skip sight-reading entirely, much as I might have wanted to. I brought in some original songs I wrote, using a concept that I’m developing with Heather Fortune. We created a bunch of short, repeated melodic grooves and then created progressively more simplified versions of them. You start playing or singing the simplest version, just quarter notes, and repeat it until you have it down. Then you move to the next, slightly more complicated version, and then the next one, until you are singing the groove in all its syncopated complexity. We sang these over looped breakbeats like the Funky Drummer.
I also had the students sing some of my music theory songs. I was hesitant to bring these things in, but I shouldn’t have been, they went over very positively. Also, the students really liked that I had written original music for them to use, so I will be less hesitant going forward.
I did not do any in-class dictation. Instead, I gave take-home transcription projects. Students could listen to the source recordings as many times as they wanted, and I encouraged them to use whatever technological tools they needed: they could slow the audio down, loop sections, use EQ to remove or boost basslines, and so on. I demonstrated these techniques in class using Ableton Live, but for students’ own use I recommended Transcribe!, because it’s much cheaper and easier to use for this specific purpose. This approach meant that I could give much more difficult transcription challenges than I could in a traditional dictation task. More on this below.
I was writing about this on Bluesky, and someone responded: “I suppose the days of working it out with a pencil and paper, during class, are decades gone?” No! Aural skills teachers most certainly still do pencil-and-paper dictation in class, including most of my NYU colleagues. I don’t think it works for pop music, though. It’s too hard for some kids and too easy for others and is totally remote from real musical life. When do you ever have to write down a melody after hearing it exactly five times? Also, the traditional method limits the kind of music you can use. You can’t give a class a Meters groove full of 16th note syncopation and expect them to write it out after a few listens, but you can certainly expect them to get it if they have unlimited time. Also, you can require that they transcribe every instrumental layer, not just the top-line melody or the chords.
I have had other music teachers say, okay, but in real life, sometimes you do need to be able to learn a tune very quickly. I know this, I have learned songs in real time on the stage in front of audiences. But I didn’t develop that ability by doing stressful and artificial dictation exercises, I got there by doing a lot of slow and thorough transcription and aural learning. Speed comes from expertise, not the other way around.
My favorite way to help students internalize theory concepts is to use them for improvisation. I loop sections of songs using Ableton and the students take turns scat-singing over them. Sometimes I will pause and point out an interesting idea or approach, or give support to a student who is struggling, but usually I try to just keep the flow going nonverbally. I have been asked what everyone is doing while waiting for their turn to sing, and the answer is, I don’t care. If they are scrolling Instagram, that’s fine, just hearing what is going on around them has value. The kids are understandably reluctant to scat in front of their peers early on, so I break the ice by doing it myself, including goofy improvised lyrics: “This is the thirrrrd of the chorrrrd, check out how it moves to the root of the neeeext choooord.” Once the ice is broken, the improv sessions tend to feel great.
Aside from the weekly assignments and in-class activities, the pop aural skills syllabus also includes four quizzes, a midterm and a final project. We are on our own to develop those. As in the theory class, I don’t give any exams, because I think that they are counterproductive for learning. Instead, I do them as take-home transcription projects.
Nina Simone transcription
Transcribe the vocal melody of the first verse of Nina Simone’s recording of “Nobody’s Fault But Mine” (1969). Add chord symbols.
This melody has some bent pitches and lots of rhythmic ambiguity, so any transcription into Western notation is going to require editorial judgment. I love this assignment, though I might move it later in the semester next time and start where the pitches are more discrete.
Solfège karaoke
Using Takadimi syllables, record yourself singing along with:
- the horn section break at 1:22 in “Children of Production” by Parliament (1976)
- the first four measures of the bassline to “Forget Me Nots” by Patrice Rushen (1982)
Using do-based solfège syllables, record yourself singing along with:
- the first two measures of the bassline to “Chameleon” by Herbie Hancock (1973)
- the head (composed melody) to “Work Song” by Nat Adderley (1960) (from 0:00 to 0:24).
Our department chair is strongly committed to solfège and takadimi, and I can understand why, they are very useful learning tools for people who get exposed to them early enough. Unfortunately, I am not one of those people. I didn’t learn solfège properly until I was 35 years old, and hadn’t even heard of takadimi until I was 45. I have never used either of these things outside of a classroom; I just think in terms of interval sizes, scale degrees, and sixteenth note subdivisions. Students who come into my classes knowing solfège already are happy to use it, but the ones who haven’t used it before find it to be an obstacle to understanding rather than an aid. Only a few students come in having seen takadimi before, and hardly anyone is happy about it. I will continue to do my best, but it’s a pain point. Anyway, this project is a great set of transcription exercises whether or not the solfège aspect benefits anyone.
Midterm transcription
Transcribe eight measures of the song of your choice. Include all of the layers you can hear: vocals, guitar, keyboards, bass, drums, strings, horns, and so on. Make sure to include chord symbols.
It’s possible to cheat on this assignment by picking a well-known song, finding a transcription online, and copying it. However, few if any students did this. Most of them were happy to be analyzing something of their own choice and threw their whole heart into it. I’ll discuss this in more depth below.
Meters transcription
Transcribe measures six and seven of “Hey Pocky a-Way” by the Meters (1975). Include all instrumental layers: piano, guitar, bass, drum kit, cowbell and tambourine.
This was a struggle for most students, both because the rhythms are hard and because detangling all the layers is a big challenge. I think it was a richly valuable experience, and plan to do lots more projects like it.
Rakim transcription
Transcribe the first sixteen bars of Rakim’s first verse on “Follow the Leader” by Eric B and Rakim (1988). You will need to find a way to represent both the rhythms and the (approximate) pitches.
This was another very challenging assignment, but the kids rose to the occasion admirably. Next semester I will spend more class time discussing and modeling how one might represent the pitches of a rap flow.
Final transcription
Transcribe sixteen measures of the song of your choice. Include every audible vocal and instrumental layer. Be sure to write chord symbols.
This is the same deal as the midterm, only longer. Some representative song choices by last semester’s class include “Washing Machine Heart” by Mitski, “What Was I Made For” by Billie Eilish, “Shut Down” by BLACKPINK, “Crazy” by Gnarls Barkley, “Something” by the Beatles, “How Deep Is Your Love” by the Bee Gees, “Thank You” by Sly and the Family Stone, and “I’m Sprung” by T-Pain. These were labor-intensive for me to grade, because I was unfamiliar with most of the songs they chose, but it was well worth it: there’s nothing more self-motivating than working with actual music that you have an emotional connection with.
Like I said above, the problem with letting the students choose their own transcription projects is that they could try to take advantage of me and choose some well-known song, find an existing chart, and copy it. Fortunately, it’s easy to tell when they do this, because online charts are usually terrible. Issues of academic integrity aside, students need to know that if they are going to learn or perform a pop song, they have to study the recording. Officially published sheet music is full of errors and omissions. These charts are produced by bored Juilliard students for extra cash, and they cut corners whenever they can: they omit intros and endings and instrumental figures, they simplify syncopations, they express dense arrangements with corny piano reductions, and they write hilariously wrong chord symbols because they don’t know pop conventions. Amateur charts are much worse. I don’t mind if people use those things as starting points, but I expect accurate basslines and drum parts from transcriptions and the kids have to use their ears for those. Anyway, abuse has been rare so far.
The biggest challenge in teaching this class is that students come in with widely varying prior knowledge and ability. Some of them sight-read and some of them barely read at all. Some of them can identify a complex chord effortlessly, and some struggle to distinguish major from minor. So everything I do has to have a low floor and a high ceiling. For the transcription projects, I allow students to redo them as many times as they need to until they get them right. I care less about assessing their current abilities than I do about helping them grow, and learn the habits that will enable to keep growing after the class is over.