A scale is a choice of which seven pitches out of twelve are allowed. Everything else — the key signature, the circle of fifths, the three minors — is bookkeeping for that one decision, and all of it is derivable rather than memorised.
Most explanations of scales start by running one up and down, which makes them look like exercises. They are not exercises. A scale is a selection: out of the twelve pitches available in an octave, it says which seven are in play. That is the useful definition, because it is the one that explains why a scale constrains melody and harmony at the same time, and why a note outside it sounds like an event rather than a mistake.
Everything else in this lesson — key signatures, the circle of fifths, the three different minors — is bookkeeping for that one decision. And all of it is derivable. There is very little here worth memorising once you know where it comes from.
Every major scale, in every key, follows the same sequence of steps:
tone – tone – semitone – tone – tone – tone – semitone
Start on any of the twelve pitches, apply that pattern, and you have that note's major scale. There are not twelve separate things to learn; there is one pattern and twelve places to start it. If a step in your result does not match the pattern, you raise or lower the note until it does — and the notes you had to alter are exactly the sharps or flats that key requires.
Start on E and write the next seven letter names in order, because a scale must use each letter exactly once: E F G A B C D E. Now walk the pattern and check each step against a keyboard.
So E major is E F♯ G♯ A B C♯ D♯ — four sharps. Nothing was looked up and nothing was memorised; the pattern produced the answer.
That last sharp is written D-sharp rather than E-flat, even though they are the same key on a piano. The reason is the rule we started with: a seven-note scale uses each letter name exactly once. E is already spoken for by the tonic, so the seventh degree has to be a D of some kind.
This is the whole explanation for every awkward spelling you will ever meet, including double sharps. A scale spelled with one of each letter can be read at a glance, because each degree sits on its own line or space and the eye tracks it as a straight climb. A scale spelled with two Es and no D would put two consecutive degrees in the same position with different accidentals, and it becomes genuinely hard to read at speed. Legibility is the reason, and it is a better one than it sounds.
Applying that pattern to a piece would mean marking the same sharps hundreds of times. Instead, they are collected into a key signature at the start of every staff, and they apply for the whole piece unless something overrides them locally.
Key signatures are written in a fixed order. Sharps go F C G D A E B. Flats go B E A D G C F. Plenty of mnemonics exist for these and they work fine, but the order is not arbitrary and knowing why beats knowing a rhyme.
Look at the sharps: F to C is a fifth. C to G is a fifth. G to D, D to A, A to E, E to B — every step is a fifth up. The order of flats is the same list read backwards, which means every step is a fifth down (or, identically, a fourth up).
This is not a coincidence about how signatures are written. It is a consequence of how the keys themselves are built: move a key up by a fifth and its scale gains exactly one sharp; move down by a fifth and it gains exactly one flat. C major has none. G major, a fifth above, has one — F-sharp. D major, another fifth up, has two — F-sharp and C-sharp, in that order, because the first sharp never leaves.
Lay that out as a loop and you have the circle of fifths, which is nothing more mystical than the twelve keys arranged by how many accidentals they carry.
The two ends meet, which is why three pairs of keys are enharmonic: B and C-flat are the same pitches spelled differently, as are F-sharp and G-flat, and C-sharp and D-flat. Composers choose between them on legibility — five sharps is friendlier to read than seven flats.
The circle also explains why keys a fifth apart sound closely related: their scales share six of seven notes. A key change up a fifth moves one note. A key change to the opposite side of the circle moves nearly everything, which is exactly why it feels like a cliff edge.
Two derivations get you the key without a lookup table.
Sharps: the last sharp is the seventh degree. Sharps are added in fifths, and the sharp that lands last is always the one a semitone below the tonic. So take the final sharp and go up a semitone. A signature ending in D-sharp is E major — which matches the scale we just built.
Flats: the second-to-last flat is the tonic. A signature of B-flat, E-flat and A-flat has E-flat as its second-to-last, so it is E-flat major. The one case this does not cover is a single flat, where there is no second-to-last — that one is F major, and it is the only thing here genuinely worth memorising.
Both derivations give the major key. Every signature also serves a minor key, which is where we go next.
The first degree of a scale is the tonic, and it functions as home: the note a melody sounds finished on, and the chord a progression resolves to. That is not a mystical property of the pitch itself — the same frequency is the tonic in one song and a passing note in another.
Two things create the effect. One is statistical: the tonic and its chord occur more often and in more structurally important places, and the ear learns the pattern within a few bars. The other is mechanical: the seventh degree sits a semitone below the tonic, and a semitone is the smallest step in the system, so it feels like an unresolved leaning. That is why it is called the leading note. Take the semitone away and the pull largely goes with it — which is the entire reason the minor scale needs a second version of itself.
The natural minor scale follows the pattern tone – semitone – tone – tone – semitone – tone – tone. Play the white keys from A to A and you have it.
Which is the clue to the most useful fact about minor keys. A natural minor scale contains exactly the same notes as the major scale a minor third above it — A minor and C major are the same seven pitches, and the only difference is which one is treated as home. Keys related this way are relative: they share a key signature, and the minor starts on the sixth degree of the major. G major and E minor share one sharp. E-flat major and C minor share three flats.
This creates the ambiguity you meet when reading a score: a signature of one sharp is either G major or E minor, and the signature alone cannot tell you which. What tells you is the leading note, and that requires knowing about the second minor scale.
Natural minor has no semitone below its tonic — the seventh degree sits a whole tone away. The pull home is weak, and worse, the chord built on the fifth degree comes out minor, which strips the strongest cadence in tonal music of its force.
The fix is blunt: raise the seventh degree by a semitone. That restores the leading note, makes the fifth-degree chord major, and gives the cadence its pull back. The result is the harmonic minor scale, and the name says exactly what the alteration is for — it is a harmonic repair, not a melodic preference.
It has a side effect. Raising the seventh while leaving the sixth alone opens a gap of three semitones between them — an augmented second, larger than any step in a major scale. That interval is the characteristic colour of harmonic minor, and it is why the scale reads as exotic to ears raised only on major and natural minor: the same gap appears in modes used widely across the Middle East, North Africa and South Asia, where it is not exotic at all, merely ordinary.
Singers found that augmented second awkward to pitch, so the melodic minor raises the sixth degree as well to close it up. In its classical form it does this only on the way up: ascending, the sixth and seventh are both raised; descending, both revert to natural minor. The logic is that on the way down there is nothing to lead toward, so the alterations have no job to do.
You will meet a second convention in jazz, where the raised sixth and seventh are kept in both directions. Neither is wrong; they are different repertoires using the same name for slightly different things, which is worth knowing before an argument starts over it.
One more relationship is worth naming. Parallel keys share a tonic rather than a signature — C major and C minor, three flats apart. Relative keys sound alike and feel different; parallel keys feel alike and sound different, which is why a shift from a major to its parallel minor is one of the most reliably affecting moves in songwriting.
An accidental written into the music applies for the rest of that bar, at that pitch, and then the bar line resets it. It is not a one-note instruction, and it is not permanent. A natural sign cancels a sharp or flat from either the key signature or an earlier accidental in the same bar.
You will often see a courtesy accidental in the following bar restating what the key signature already says. That is not redundancy for its own sake — it is a copyist deciding that a reader arriving at bar ninety at speed should not have to reason about it.
Take the seven notes of a major scale and treat a different one as home each time, and you get seven scales built from the same set of pitches with seven different characters. Those are the modes, and starting on the sixth degree — which produces the natural minor — is one of them. There is a full lesson on them later in this specialization; for now, the thing worth carrying forward is that the pattern of tones and semitones is what defines a scale, not the pitches it happens to contain.
Most sessions have no score. What they have is a chord chart, or a guitarist who knows the shapes, and the key still has to be established because the tuning software and the sample library both want to be told.
Three methods, in order of reliability.
There is one instrument-specific trap here, and it is the exact analogue of the transposing-instrument problem from the first lesson. A guitarist with a capo is playing familiar shapes at a shifted pitch, and will usually name the shapes rather than the sounding key. Shapes in G with a capo on the third fret sound in B-flat — one semitone per fret. So the guitarist says G, the keyboard player says B-flat, and, once again, both are right.
The working move is the same as before: ask what it sounds in, not what it is played in. And if you are printing a chart for anyone who is not holding that guitar, print the concert key.
Two things, and both are practical.
The first is diagnosis. When a tuning-corrected vocal keeps landing wrong on one word, or a guitar part fights the keyboard in one bar, knowing the key lets you say which note is the problem instead of describing a feeling. A note outside the scale is either a deliberate colour or an error, and those get fixed differently.
The second is more concrete and catches people out. The tonic bass note lands on one specific frequency, over and over, on the strong beats, for the whole song. In E, the open low E on a bass is about 41 Hz and the octave above it about 82 Hz. In A, those become 55 and 110. If the room has a mode near that frequency, or the cabinet has a resonance there, or the mix bus has a build-up there, the key of the song decides whether you hear it — and it will be excited on every downbeat rather than at random.
That is the reason a bass part can sound even in one song and lumpy in the next with nothing changed on the desk. It is also the reason a narrow cut that seems oddly specific on one track is often exactly right: it is sitting on the note the song keeps coming home to.
A key signature is not decoration. It is the list of notes that are allowed, written once at the front instead of nine hundred times through the piece.
Students pay for getting unstuck: ask a concept question, routing issue, DAW confusion, or mix decision tied to this lesson.
Free readers can learn from the public Q&A archive. Paid students can ask their own lesson-specific questions and get mentor replies.