Fifths

Science blog ·

Tonic, dominant, subdominant: hearing chord function before chord name

A trained harmonic ear hears chords as functions in a key, not as letter names. How to build that ear, starting with three categories.

chord-function harmony progressions roman-numerals ear-training

Ask a beginner to identify a chord by ear and they will usually try to identify its root note — they listen for the bass, label it C or G or A, and then try to figure out major versus minor. This is a perfectly reasonable strategy and it is also a slow one. It does not generalize well across keys, and it does not match how trained musicians actually hear harmony in real time.

A trained ear hears chords as functions in a key, not as letter names. The first thing it identifies is not “this chord is G major” but “this chord is the V” — the dominant in whatever key has been established. Chord quality (major, minor) and chord root note (C, G, A) are downstream details. Function comes first.

This is more than an academic distinction. Functional hearing is what makes a working musician able to learn songs by ear in any key, transcribe progressions on the fly, and improvise sensibly without a chart. It is also, on the available evidence, more teachable than the letter-name approach, because it leverages the tonal hierarchy the ear is already learning anyway.

Three categories cover most of tonal music

Western tonal harmony, in its skeletal form, can be reduced to three functional categories: tonic (T), subdominant (S, or “predominant”), and dominant (D). Almost any chord progression in pop, jazz, classical, folk, gospel, or country music can be understood as motion between these three functions [1].

  • Tonic (T): home. The I chord, and to a useful approximation also the iii and vi chords (which share two of three notes with I).
  • Subdominant (S): away from home, but not yet building tension. The IV chord, and approximately also the ii (which shares two notes with IV).
  • Dominant (D): tension that wants to resolve. The V chord, and approximately also the vii° (which is V with the root removed).

The classic motion of tonal music is T → S → D → T. Twinkle Twinkle Little Star, Hey Jude, Stand by Me, Don’t Stop Believin’, and tens of thousands of other songs are functionally this progression with embellishments.

If you can hear those three categories — home, away, tension — you can hear most of what tonal music is doing. Quality (major vs. minor) and exact chord identity become labels you assign on top of an already-correct functional analysis.

Why function is easier than identity

Two reasons.

1. Functions are perceptual primitives; identities are derived. The tonal hierarchy that the ear extracts from cadences and tonic establishment (see our article on cadence-first practice) directly tracks function — it tells the listener which pitches are stable and which are unstable, which chords feel like home and which feel like motion [2]. Chord identity (“this is G major”) is a separate cognitive operation that requires translating the heard chord through the established key. Function is what the ear reports; identity is what the analyst computes.

2. Function is invariant across keys; identity changes. A V chord sounds like V whether the key is C, F, or Bb. The G major triad sounds like V in C major, I in G major, and IV in D major — the chord is the same, but its function (and therefore its emotional and structural role) is completely different. A learner trained on chord identity has to relearn the entire vocabulary in every new key. A learner trained on function carries one vocabulary across all keys.

This is the same argument the functional-vs-interval article makes for scale degrees, applied one level up. The tonal hierarchy operates on chords as well as on individual pitches, and the ear that hears function is doing what the cognitive system is built to do.

A staged approach to building the functional ear

A research-supported progression — broadly compatible with how the Berklee functional aural-skills track and modern aural-skills textbooks structure it [3] — looks like this:

Stage 1: Tonic vs. not-tonic. Play a cadence. Then play a single chord. Was that the tonic, or something else? This is the lowest-load possible chord question and it bootstraps the entire functional ear. Fifths’ Tonality segment uses the same structure for individual notes; the chord version is the natural extension.

Stage 2: T vs. D. The two strongest functions, with the strongest functional identity contrast. Tonic feels resolved; dominant pulls toward tonic. Listeners typically distinguish these two within a session or two.

Stage 3: T vs. S vs. D. Add the subdominant. Now the listener is tracking three categories. Many learners initially confuse S with T (both feel “stable-ish” relative to the dominant). The fix is more contextual practice — hear S → D → T enough times and the S takes on its own functional identity as “the place that wants to go to D.”

Stage 4: Bass-line root recognition. Once function is solid, the next layer is hearing the scale degree of the chord’s bass — this is what tells you whether you’re hearing I or iii, IV or ii, V or vii°. Bass roots of 1, 4, and 5 cover the lion’s share of common tonal harmony; 2, 3, 6 are the next tier; 7 is rare in common-practice harmony.

Stage 5: Quality on top of function. Major or minor, then sevenths and extensions. By this stage the functional category has already been identified, so quality is a small disambiguation rather than the entire question.

Stage 6: Common progressions as units. I-V-vi-IV. ii-V-I. I-IV-V-I. vi-IV-I-V. These are the chunks of harmonic working memory — the ear that can hear “this is a ii-V-I” as one event holds far more music in working memory than the ear that hears “ii, then V, then I” as three separate events. (Compare the chunking discussion in our melodic-dictation article.)

Why the bass line is the secret weapon

A practical observation that working transcribers consistently report: the bass line carries most of the harmonic information. If you can hear the bass move 1 → 4 → 5 → 1, you have correctly identified a I-IV-V-I progression in nearly every case, regardless of voicing or instrumentation. The upper voices add color but rarely change the function.

This is also a working-memory win. Tracking one voice (the bass) is dramatically easier than tracking three or four (full voicings). For self-directed practice, transcribing the bass line first and then filling in the chord qualities is a faster and more reliable workflow than trying to identify the full chord on first pass [4].

Starting tomorrow

If you are training your harmonic ear without a structured app:

  1. Find a song you know well — Let It Be and Stand by Me are both nearly pure I-V-vi-IV. Hey Jude is a clear I-V-IV-I motion in the verse.
  2. Play along on a single instrument (piano, guitar). Hum the bass line. Identify each chord’s function in the key — home, away, tension.
  3. Once the function is reliable, add quality (major/minor) and finally the specific Roman numeral.
  4. Move to a song in a different key and verify that the functional progression is the same even though the chord names are not. (Stand by Me in A vs. C vs. F — the chords differ; the I-vi-IV-V is invariant.)

This is the same mental operation that the trained ear is doing automatically when it hears music. Building it deliberately, over a few weeks of consistent practice, replaces the slow letter-name approach with the fast functional one. The change is permanent and it generalizes across keys — which is to say, across all music — for free.

References

  1. Tymoczko, D. (2011). A Geometry of Music: Harmony and Counterpoint in the Extended Common Practice. Oxford University Press. The functional-categories framework (T, S/PD, D) is well-established in modern Schenkerian and Riemannian theory; see also Caplin, W. E. (1998). Classical Form. Oxford University Press. ↩︎

  2. Krumhansl, C. L. (1990). Cognitive Foundations of Musical Pitch. Oxford University Press. On chord-level tonal hierarchy: Bharucha, J. J., & Krumhansl, C. L. (1983). The representation of harmonic structure in music: Hierarchies of stability as a function of context. Cognition, 13(1), 63–102. ↩︎

  3. For the holistic-aural-skills approach to functional harmony: Cleland, K. D., & Dobrea-Grindahl, M. (2021). Developing Musicianship through Aural Skills (2nd ed.). Routledge. https://www.routledge.com/9780367030773. See also Berklee Online’s Harmonic Ear Training: Recognizing Chord Progressions course outline at https://online.berklee.edu/courses/harmonic-ear-training-recognizing-chord-progressions. ↩︎

  4. This is the approach taken by the Integrated Aural Skills curriculum: https://intmus.github.io/intas19-20/09-introduction-to-harmonic-dictation/dictation4.html (“Root Movement in Chord Progressions”). The pedagogical argument is consistent with the working-memory literature on chunking: bass-only listening is a lower-load sub-task that serves as a scaffold for full harmonic dictation. ↩︎

← Back to all articles