Bass-up vs melody-down: two ways to internalize a chord progression by ear

A musician learning a tune by ear has to do two things: hear the melody and hear the changes. The melody is usually easy — it is what the listener naturally tracks. The changes are the hard part. The conventional wisdom in most ear-training pedagogy — and in the Berklee functional tradition explicitly (see Singing the changes) — is to start from the bass and work upward. But there is a second valid tradition that does the opposite: start from the melody and infer the changes from where the melody resolves.

The two approaches produce differently-shaped musicians. They suit different learners. They have different stuck points. They are also, importantly, both supported by serious pedagogical lineages and converge in the limit. This post lays out the trade-off, drawing on the Sheila Jordan / Jay Clayton bass-up modal tradition (briefly introduced in The four lineages of jazz ear training) and on the Berklee functional / classical-aural-skills tradition that takes the melody-down approach.

The bass-up approach

The bass-up tradition (most strongly associated in jazz with Sheila Jordan, Jay Clayton, and the modern jazz vocal programs at Manhattan School and the New School) treats the bass note as the cognitive anchor for harmony.

The pedagogy:

First, hear and sing the bass note of every chord in the progression. Just the bass — no chord quality yet. Get the bassline locked in as the structural skeleton.
Add the chord quality. Major or minor third? Once the bassline is solid, the third defines the chord’s emotional color.
Add the seventh. This determines harmonic motion (whether the chord wants to resolve down a fifth or sit modally).
Add upper extensions and color tones (9, 11, 13, alterations). These are the “flavor” — added last because they don’t change the chord’s identity.
Add the melody. The melody is treated as superimposed on a now-fully-understood harmonic foundation.

The cognitive theory behind this: harmony is built on the bass, both in performance (the bassline carries the chord roots) and in chord-symbol notation (the chord symbol always names the bass note first). Building auditory understanding from the bass mirrors how the music is structured.

What this approach is good at: modal jazz, slow ballads, anything where the harmony is the primary musical content. Bass-up listeners are particularly strong at hearing modal interchange, harmonic substitution, and bass-driven music (gospel, R&B, Latin). They tend to have unusually accurate basslines when transcribing.

Where bass-up listeners get stuck: fast bebop with active basslines, where the bass moves so quickly that anchoring to it becomes overwhelming. Melodic detail can suffer — bass-up players sometimes lose track of where a soloist is in the changes because their attention is on the bass.

The melody-down approach

The melody-down tradition (associated with classical aural-skills programs, the Karpinski sight-singing tradition, and the way most musicians actually approach pop tunes the first time they hear them) treats the melody as the cognitive anchor and infers the harmony from how the melody resolves.

The pedagogy:

First, internalize the melody. Sing it, hum it, get it locked in as a recognizable tune in a known key.
Identify the cadence points. Where does the melody land on a tonic-feeling note? On a dominant-feeling note? Each cadence point implies the chord underneath.
Fill in the chord between cadence points. Knowing that one phrase ends on the tonic and the next begins on the dominant, infer the chord motion that connects them — usually a ii-V-I or some variant.
Refine the chord quality. Major or minor mode? Modal interchange? Once the harmonic skeleton is in place, the quality details fill in.
Add the bassline last. The bassline is treated as a confirmation of the inferred chord roots.

The cognitive theory behind this: melody is what the listener actually attends to in real-world listening. Building harmonic understanding from the melody outward mirrors how non-musicians naturally process music, and it leverages the listener’s preexisting melodic memory.

What this approach is good at: standards repertoire, melodic-driven music (Tin Pan Alley, Brazilian, folk-influenced jazz), and anything where the listener already partially knows the melody. Melody-down listeners tend to have unusually accurate phrase-structure recognition and are good at spotting compositional choices.

Where melody-down listeners get stuck: modal jazz where the melody is sparse and the harmony is the primary content (Coltrane’s Impressions, much of Wayne Shorter’s writing). Modal interchange and substitution can be invisible to a melody-down listener who is inferring chords from “what the melody does over them.” Modern jazz where the melody implies harmonies that are not the actual chords is genuinely confusing.

Why both traditions are valid

The two approaches are not in competition; they are different cognitive routes to the same destination. The empirical evidence supports this.

Pomerleau-Turcotte and colleagues’ study of conservatory students found that multiple aural skills track together, including melodic dictation, harmonic dictation, and sight-singing — implying a shared underlying competence rather than separate specialized skills ^[1]. A student who develops via the bass-up route eventually becomes good at melodic listening too; a student who develops via the melody-down route eventually becomes good at harmonic listening too. The starting point shapes the early curriculum and the early stuck points; the long-term competence is the same.

The Berklee functional tradition, despite being most closely associated with the bass-up approach, explicitly trains both — the four-semester Ear Training sequence covers melodic dictation alongside harmonic dictation, and Roberta Radley’s Real Easy Ear Training Book approaches chord identification by decomposing the listening task (bassline first, then 3rd, then function) rather than by insisting that one element is more fundamental than another ^[2]. The bass-first ordering is pragmatic, not ideological.

How to know which approach suits you

Three diagnostic questions.

What do you naturally remember from a song you’ve just heard? If you remember the bass first (“dum-dum, da-dum-dum”), you’re naturally bass-up. If you remember the melody first (“la-la, la-la-la”), you’re naturally melody-down. Most people are melody-down by default; bass-up is more common among bassists, drummers, and electronic-music producers.

What are you trying to learn? If you are trying to play modal jazz, bass-up will probably suit you better — the harmony is the content. If you are trying to play standards repertoire, melody-down will probably suit you better — the melody is the content.

What instrument do you play? Singers and melodic instrumentalists (saxophone, trumpet, violin) tend to default to melody-down. Bassists and chord instrumentalists (piano, guitar, vibes) tend to default to bass-up. Either can be developed; the default tells you which one will take less work.

The strongest pedagogical position, in our view, is to start with whichever approach is your default and explicitly cross-train into the other one over time. A bass-up listener who has never learned to track melody will plateau on standards; a melody-down listener who has never learned to hear bass will plateau on modal jazz. Both gaps close with deliberate cross-training.

How to cross-train

For a melody-down listener developing bass-up skills:

Listen to recordings with the bass foregrounded. Walking-bass-driven jazz (Ray Brown, Paul Chambers, Christian McBride) is ideal. Sing along with the bass instead of the melody.
Transcribe basslines. A walking-bass transcription of a standard is one of the most efficient bass-up training exercises available.
Practice singing bass-up arpeggios. Berklee functional drills (sing root → 3rd → 7th of each chord — see Singing the changes) explicitly train the bass-up skill in vocal form.

For a bass-up listener developing melody-down skills:

Sing standards from memory without instrumental accompaniment. Just the melody. Force the cognitive machinery to anchor on the melodic line.
Identify modulations from melodic cues. When does the melody move outside the home key? Where does it suggest a new tonic?
Practice melodic dictation. Karpinski’s sight-singing curriculum, the broader classical aural-skills tradition, and the existing Fifths Voice section all train this directly.

The bass-up vs melody-down trade-off is one of the more important pedagogical decisions in jazz ear-training, and one of the few that pedagogues genuinely disagree about. The pragmatic answer — start with your default, cross-train into the other — is supported by both lineages and by the broader cognitive-science evidence that shared underlying competences cross over with practice. The deeper answer is that there is no answer, just two equally valid roads up the same mountain.

References

Pomerleau-Turcotte, J., Moreno Sala, M. T., Dubé, F., & Vachon, F. (2022). Experiential and Cognitive Predictors of Sight-Singing Performance in Music Higher Education. Journal of Research in Music Education, 70(3). https://doi.org/10.1177/00224294211049425. PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC9242514/. The finding that multi-part dictation skill is the strongest single predictor of overall aural-skills outcomes supports the claim that the underlying competence is shared across what surface differences in pedagogy treat as separate skills. ↩︎
Radley, R. (2009). The Real Easy Ear Training Book. Sher Music Co. https://www.shermusic.com/188321761X.php. Radley’s decomposition approach (listen for one element at a time — bassline, third, in/out of key) is pragmatically bass-up but explicitly designed to combine with melodic listening rather than displace it. ↩︎