Most of my students struggle with key signatures. This is understandable! Like the rest of the Western notation system, key signatures are based on a big assumption: that all of the notes will be within one of the twelve major keys, or within some scale that can be derived from a major scale (most often, the natural minor scale). This assumption makes an awkward fit with the music that the kids are making and listening to. Read on!
Several students have asked me if there is some shortcut or mnemonic for memorizing the key signatures. The answer is, yes, there are many, but I’ve never found them to be helpful. The only thing that worked for me was to learn, write and improvise a lot of music in every major and minor key until they were as familiar as the layout of my apartment. My method was slow, but effective.
You can think of music notation as a graphical representation of the white keys on the piano. The white keys play the seven notes in the C major scale, repeated across octaves. C major is Western European music’s “default setting.” To play the other major scales, you will need to replace one or more white-key notes with one or more black-key notes. Each black key confusingly has two names: A-sharp/B-flat, C-sharp/D-flat, D-sharp/E-flat, F-sharp/G-flat, and G-sharp/A-flat. This is a holdover from historical tuning systems in which A-sharp and B-flat were slightly different notes.
In notation, you change a white-key note to its neighboring black-key note using a symbol called an accidental. You write a sharp next to the note if you want to raise it up to the closest black-key note, and you write a flat next to the note if you want to lower it to the closest black-key note. (But what about F-flat and C-flat? Consult a music theory textbook for this kind of exotica.)
If a note is going to stay sharp or flat all the way through the piece, you can write its accidental at the very beginning of the score, and then that note will be sharp or flat every time it occurs. This collection of “global” accidentals is called the key signature. The key signature in the image at the top of this post is telling you that every time you see a B, you should play B-flat, and every time you see an E, you should play E-flat.
You might reasonably ask, if this is what key signatures are for, why do you even need them? Couldn’t you just write accidentals next to all of your notes one at a time? The answer is, sure you could, and historically, some people have. JS Bach wrote his famous Chaconne in D minor without a key signature. He wrote in the individual B-flats as they arose. That’s fine for D minor, but in other keys, writing all the accidentals individually would be too much work.
Here’s a little tune I wrote in E major, shown in two ways: with accidentals on individual notes, and then with a key signature. You can see how the one-at-a-time method would get old fast.
Aside from convenience, the main reason to use key signatures is that they communicate to the performer what key the music is in. If I see a key signature with four sharps, then I’m going to mentally prepare myself to play E major. If there’s no key signature, I don’t know what to expect, and that’s just more cognitive load for me. At this point, you might reasonably ask why you can’t just write “E major” at the top of the page. In jazz, you sometimes do write the key verbally at the top of the page, especially if the tune is in an unusual mode or scale. But that’s not the conventional way to do things.
The sharps or flats that make up a key signature appear in a specific order. In the following section, I’ll explain how that order works for major keys. In classical music, major keys and major scales are the same thing, so that makes things conceptually easy. In more current music, the situation is more complicated, but we’ll get to that later.
Major key signatures
There’s an elegant and simple way to understand which sharps and flats you use to make the major scales. It involves the circle of fifths, which is one of those music theory things that you should just memorize, because it’s useful, and it comes up a lot. Here’s the circle:
Think of each of these notes as representing a major key. The key of C major uses no sharps or flats (it’s all white keys.) As you go clockwise around the circle from C, you add sharps to make those keys. As you go counterclockwise around the circle from C, you add flats to make those keys.
When you get to the bottom of the circle, you have a choice to make. You can think of that key as being F-sharp, which you write with sharps, or as being G-flat, which you write with flats. In the olden days, F-sharp and G-flat were two different keys using two slightly different-sounding sets of pitches. However, in the era of equal temperament, F-sharp and G-flat sound exactly the same, and they are perfectly interchangeable. So why do we bother maintaining both sets of names? It’s hard to change tradition.
This diagram shows you which notes you have to make sharp or flat to construct each major key signature. The notes are cumulative as you go around.
First, let’s think about the sharp keys.
- To make the key of G, you change F to F-sharp.
- To make the key of D, you change F to F-sharp, and C to C-sharp.
- To make the key of A, you change F to F-sharp, C to C-sharp, and G to G-sharp.
And so on. There’s a pattern here: the last sharp in the key signature is the leading tone (scale degree seven) of the key.
Now, let’s think about the flat keys.
- To make the key of F, you change B to B-flat.
- To make the key of B-flat, you change B to B-flat, and E to E-flat.
- For the key of E-flat, you change B to B-flat, E to E-flat, and A to A-flat.
And so on. There’s a pattern here too: the last flat in the key signature is the subdominant (scale degree four) of the key.
Let’s think a little more about what it means to move from one key to another around the circle of fifths. Here’s another version of the diagram, showing each key center with its dominant chord (the purple ones).
In classical music, you establish a key by playing that key’s dominant chord, followed by its tonic chord. To establish the key of C, you play a G7 chord followed by a C chord. To establish the key of E, you play a B7 chord followed by an E chord. To establish the key of A-flat, you play an Eb7 chord followed by an Ab chord. If you want to change keys in a smooth and logical-sounding way, a good method is to play the dominant chord in your new key and then resolve to its tonic. You can get from any key to any other key that way and it will sound okay, but it sounds especially good (to Western people) if you move your key centers counterclockwise around the circle of fifths.
To move one step counterclockwise around the circle, you take the seventh scale degree of your present key and flatten it. To move from C major to F major, you change B to B-flat. In classical music, you’d most likely put that B-flat on top of a C major triad, which changes it into a C7 chord. In other words, you transform the tonic chord in C into the dominant chord in F major. Then you resolve to an F chord. Try it, and enjoy the smooth sound of functional tonal harmony.
If you then wanted to move from F major to B-flat major, you would change E to E-flat. You could put that E-flat on top of an F major triad to change it to an F7, the V7 chord in Bb major, and then resolve to Bb itself. And so it goes.
This kind of counterclockwise circle-of-fifths key movement happens all the time in real-world music, and it sounds great.
It also sounds good to move your key centers clockwise around the circle of fifths. To do so, you need to raise the fourth scale degree of your current key to the excitingly dissonant sharp fourth. This sharp fourth becomes the leading tone in the new key. If you want to move from C major to G major, you raise F to F-sharp. You can include that F-sharp in a D7 chord, the V7 chord in G major.
To get from G major to D major, you change the 4^ of G major to #4^, that is, raise C to C-sharp. You’d probably put that C-sharp in an A7 chord, the V7 chord in D major. To get from D major to A major, you raise G to G-sharp and put it in an E7 chord, and so on.
Moving your roots this way is extremely common in canonical classical music, and is not uncommon in older pop and country styles. However, you typically don’t go multiple steps clockwise around the circle in sequence.
Take a trip around the circle of fifths, first counterclockwise, then clockwise:
Minor key signatures
Minor key signatures are harder to understand than major keys, because they don’t really get their “own” key signatures. Instead, each minor key shares the same key signature as its relative major. So A minor has the same key signature as C major; E minor has the same key signature as G major; B minor has the same key signature as D major; and so on.
The logic behind this convention is that each natural minor scale contains the same pitches as its relative major scale, so you’re using the same sharps and flats.
To make minor keys, you add your sharps and flats around the circle of fifths the same way you did for the major keys. In sharp-side minor keys, the last sharp in the key signature is on scale degree two. In flat-side minor keys, the last flat in the key signature is on scale degree six. If you are going to use the harmonic or melodic minor scales, you will need to manually stick accidentals on the individual sixth and seventh scale degrees as necessary.
Just like with the major keys, you often want to move through the minor keys around the circle of fifths. Moving counterclockwise is common, and is relatively easy to understand: you just change your tonic chord to the dominant on the same root, and then resolve it down a fourth.
To get from A minor to D minor, you just change the Am chord to A7, the V7 chord in D minor. To get from D minor to G minor, you change the Dm chord to D7, the V7 chord in G minor. And so it goes.
I can’t think of any clever mnemonic for going clockwise through the minor keys, probably because it’s not something I find myself doing very often. You’re taking your ii chord and making it dominant, then resolving it up a fourth. I don’t know, maybe that’s helpful?
Hear the minor keys around the circle of fifths, first counterclockwise, then clockwise:
Diatonic modes
The key signature system comes from a time and place when it was assumed that everything was either in a major key, or in a minor key that mostly used natural minor. That system worked well throughout Western Europe for many hundreds of years. But tastes change, and we Western people in the present find the diatonic keys to be too plain-vanilla-sounding. Pop music of the past hundred-ish years has been increasingly likely to use modes and blues to produce notes that are outside the keys. Jazz and “art” music use all kinds of exotic and non-standard scales. All of this makes using standard key signatures harder and more annoying.
Let’s say you want to write something in one of the diatonic modes: Mixolydian, or Dorian, or Phrygian or whatever. The standard way to do key signatures for the modes is similar to the way you do them for natural minor. You figure out which major scale your mode is derived from, and you use the key signature for that major key. So let’s say you’re Miles Davis, and you’re writing a tune in D Dorian mode.
D Dorian mode uses the same pitches as C major, so you’d write a key signature with no flats or sharps. Now let’s say the bridge of your tune is in E-flat Dorian mode. That scale uses the same pitches as D-flat major, so you’d change the key signature to include B-flat, E-flat, A-flat, D-flat, and G-flat. Then when your tune goes back to D Dorian, you’d switch back to the no-flats key signature for C major.
In classical music, this system for modes is fine, because the modes don’t really exist outside the context of the major/minor key system. When Beethoven wrote the third movement of his string quartet no. 15 in “F Lydian,” he meant that it was really in C major, but that it hung around the note F a lot. So it made sense to write the key signature as C major. But that might give you the idea that everything in F Lydian is “really” in C major. When you listen to “Possibly Maybe” by Björk, most of the tune is in A Lydian mode. It would be conventional to notate it using the key signature for E major, but then you might get the idea that the tune is “really” in E major, and it isn’t.
For another example, John Coltrane’s beautiful tune “India” is in G Mixolydian mode. The classical convention is to say, well, that means it’s “really” in C major, so you should use the key signature for C major. But “India” is obviously not in C major; the bass never moves off G for the tune’s entire fourteen minute duration. So a key signature suggesting C major is illogical.
I think it would make more sense to notate modal tunes using the key signature for the parallel major or minor key, not the relative one. You could write your D Dorian mode tunes using the key signature for D minor, your G Mixolydian mode tunes with the key signature for G major, your A Lydian mode tunes with the key signature for A major, and so on. You’d have to correct a lot of individual accidentals by hand, but it would communicate more clearly to the player what key they’re really in.
In rock and pop music, you frequently combine a parallel major scale and Mixolydian mode (like C major and C Mixolydian), or a parallel natural minor scale and Dorian mode (like C natural minor and C Dorian). In those cases, it would be nice to not to have to put key signature changes all over the place. But the kinds of pop music that spend the most time outside the diatonic key system are usually not notated at all to begin with, so there hasn’t been so much urgency to get this system fixed.
The blues
Talk about a music that exists outside the Western European major/minor system! The usual way to do key signatures for the blues is to use the major key with the same root. So if you’re writing a tune in E blues, you use the key signature for E major, and then you’ll just have to write lots of accidentals on individual notes to make the various blues scales. This is extremely cumbersome! The good news is that you rarely notate blues music anyway, so it probably won’t come up.
Atonality and exotica
What if you want to write using something more unusual? What if you’re writing with a diminished or whole-tone scale, or a non-Western scale, or some idiosyncratic collection of pitches that you invented? There are two typical approaches. You can try to find some major or minor key that’s similar, and then use accidentals on individual notes as needed. Alternatively, if there is no meaningful comparison to any of the Western keys, you can just have no key signature at all, and use accidentals on individual notes all the way through.
A few people have preferred to solve this problem by inventing custom key signatures. For example, Béla Bartók liked to write in Lydian dominant mode. In C, that’s the pitches C, D, E, F-sharp, G, A, and B-flat. The standard method would be to use the key signature for C major, and then put sharps on all the Fs and flats on all the Bs. Free thinker that he was, Bartók preferred to write a key signature with a B-flat and an F-sharp. (Coltrane sometimes used that method too.) Making up your own custom key signatures certainly makes writing in exotic scales easier. Unfortunately, it makes your charts harder to read for people who aren’t used to your system. Conventions can be annoying, but they exist for a reason.