Scale Degrees and Tonal Function: What Your Ear Actually Hears

Degrees are positions, functions are jobs. Both were measured before they were taught.

Short answer: a scale degree is the numbered position of a note inside a key: 1 is the tonic, 5 is the dominant, 7 is the leading tone. Tonal function is the job a chord or a degree does in a phrase, and Western tonal music uses three of them: tonic, subdominant (often called predominant) and dominant. The reason this matters to your ear rather than to your paperwork is that listeners rate the seven degrees in a stable, repeatable order of fit, even listeners with no theory training at all. Your ear is already keeping score. Naming the degrees just gives you access to a measurement you were making anyway.

The seven positions, and what they are called

Pick any major key. Number the notes of its scale 1 to 7 and you have the scale degrees. Each number carries a traditional name, and in movable-do solfège each carries a syllable as well. The three labels describe the same thing at different levels of abstraction: a position, a role, and a singable handle.

The fourth column is not a theorist's opinion. It is measured data, and it is the reason this article exists.

The evidence: the probe-tone experiment

In 1979 Krumhansl and Shepard ran a deceptively simple experiment. Play a listener an incomplete major scale, then play one more tone, and ask how well that tone completes the sequence. Repeat for all twelve chromatic tones. Krumhansl and Kessler extended this in 1982 with fuller contexts (scales, chords, cadences) in major and minor keys, and the ratings settled into a stable profile: the tonic clearly highest, the dominant next, then the mediant, then the rest of the diatonic notes, and the five non-scale tones well below all of them.

Two things about this result are worth sitting with. First, the profile is a hierarchy rather than a binary. The ear does not simply sort notes into "in the key" and "out of the key"; it grades them. Second, the profile is remarkably consistent across listeners, which is why it later became the basis of computational key-finding. Temperley's 1999 reassessment of the Krumhansl and Schmuckler key-finding algorithm showed that a machine can identify the key of a passage reasonably well using little more than these rating profiles correlated against the notes actually present.

You already learned this without being taught

The obvious objection is that the listeners in these studies were musically experienced, so the hierarchy might be a trained artefact. It is not. Tillmann, Bharucha and Bigand showed in 2000 that a self-organising network exposed only to ordinary Western musical input develops the tonal hierarchy and the relationships between keys on its own, without anyone labelling anything. Bigand and Poulin-Charronnat reviewed the wider literature in 2006 and concluded that listeners without formal training show sophisticated sensitivity to tonal structure on a long list of tasks.

What formal training adds is mostly the vocabulary and the deliberate access. Untrained listeners feel that a phrase ending on 7 is unfinished. Trained listeners can say why, and can choose to end there.

Function: three jobs, not seven

Degrees describe single notes. Function describes what a harmony is doing, and common-practice music groups harmonies into three categories: tonic, subdominant and dominant. Open Music Theory characterises each function by the scale degrees that tend to be present: tonic by degrees 1, 3, 5, 6 and 7; subdominant by 1, 2, 3, 4 and 6; dominant by 2, 4, 5, 6 and 7. The ordinary phrase shape is tonic, then subdominant, then dominant, then tonic again, and most of what sounds like harmonic motion to a listener is movement between those three states.

The useful part for ear training is that the two most function-defining degrees are the two least stable ones. Degree 7 pulls up to 1 and degree 4 pulls down to 3. If you can hear those two tendencies, you can usually hear where a phrase is in its cycle, even without naming a single chord.

Context does not just help you name the note. It helps you hear it.

Here is the finding that turns this from theory into practice. Marmel, Tillmann and Dowling reported in 2008 that tonal expectation changes pitch perception itself: listeners judged the pitch of a target tone more precisely when the preceding context made that tone strongly expected. Graves and Oxenham found in 2017 that a familiar tonal context improves the accuracy of pitch interval perception in the same way.

In other words, a note heard inside a key is heard more sharply than the same note heard in isolation. That is an argument for practising with a key established rather than against a silent background. Sound a drone on the tonic, or play a quick cadence first, then do the work. You are not making it easier by cheating; you are giving your auditory system the context it evolved to use.

A practice routine that uses the hierarchy

Start with a drone on the tonic and sing 1, then 5, then 1 again. That interval is the spine of the key. Add 3 next, because 3 carries the major or minor colour. Only then fill in 2, 4, 6 and 7, and practise each against the drone rather than in a scale run, because a scale lets you count your way up rather than hear the position.

Then reverse the exercise. Have someone play a random degree over the drone and name it before you sing it. Naming before singing is the part that transfers, because it forces recognition rather than reproduction. Expect 1, 5 and 3 to be easy and 2 and 6 to be the last to stabilise, which is exactly the order the probe-tone profile predicts.

What our games do here, honestly

All eleven Hearkle games are monophonic. They play one tone at a time, so none of them presents a key, a chord or a harmonic context, and we do not have a scale-degree game. What they do train is the layer underneath scale-degree hearing: how small a pitch difference you can resolve, whether you can track and reproduce the size of an interval, and how long a pitch survives in memory. Those are the raw abilities that a scale-degree exercise then puts to work. If you want the harmonic layer today, you need a drone or a keyboard, and this guide is the map for what to do with it.

Keep reading

• Movable do vs fixed do: the two ways of attaching syllables to degrees, and what each one costs.
• Musical intervals explained: the distances between degrees, named and counted.
• Interval ear training: how to practise the distances once you can hear the positions.
• What pitch is: the perceptual dimension all of this is built on.

Sources: Krumhansl and Shepard (1979), Quantification of the hierarchy of tonal functions within a diatonic context, Journal of Experimental Psychology: Human Perception and Performance 5(4); Krumhansl and Kessler (1982), Tracing the dynamic changes in perceived tonal organization in a spatial representation of musical keys, Psychological Review 89(4); Tillmann, Bharucha and Bigand (2000), Implicit learning of tonality: a self-organizing approach, Psychological Review 107(4); Bigand and Poulin-Charronnat (2006), Are we “experienced listeners”?, Cognition 100(1); Marmel, Tillmann and Dowling (2008), Tonal expectations influence pitch perception, Perception & Psychophysics 70(5); Graves and Oxenham (2017), Familiar tonal context improves accuracy of pitch interval perception, Frontiers in Psychology 8; Temperley (1999), What’s key for key? The Krumhansl-Schmuckler key-finding algorithm reconsidered, Music Perception 17(1); Scales and scale degrees and Harmonic functions, Open Music Theory; Major scales, scale degrees, and key signatures, Open Music Theory (VIVA); Degree and Tonality, Grove Music Online, Oxford University Press.