How to Choose a Key Signature

One of the frequent questions I get from clients regards the proper use of key signatures. When we take music lessons growing up, we’re taught that the purpose of a key signature is to avoid having to write all the sharps and flats in the music – that we can just put those sharps and flats at the beginning of the staff, and then we don’t have to repeatedly clutter up the music with accidentals (sharps and flats).

While this is true, a key signature also serves a second, more crucial function than saving ink and clutter: it indicates to the performer the “key center” that the music is in. In almost all music, there’s a repeated harmonic tension and release, tension and release, tension and release throughout the song. It’s this harmonic give and take, this tension and release that makes the music exciting to the listener’s ear. It creates the musical drama. The “key center” of a song is the place that the music tends to want to move towards that creates the most satisfying conclusion of the harmonic tension. This is why for most songs, the quickest way to figure out its key center is to look at the last chord. Most of the time, the last chord of a song is the most satisfying resolution of the harmonic tension – it’s a way to say to the listener, “The tension is resolved, the song has a satisfying ending.” A key signature should reflect the music’s key center. It helps the performer know where the music is going and what it’s ultimately moving towards.

Most of the time, these two purposes of key signatures (eliminating accidentals and indicating the key center) are not at odds with each other – they overlap nicely. The question is what to do in the odd circumstance when a song isn’t in a minor or major key, but rather, is modal.

For instance, you can imagine a piece of music where the key center is clearly C-major. It wants to resolve towards this harmony, it keeps returning there throughout the piece, and the music sounds most completed and at ease when it reaches this harmony. However, this same piece of music doesn’t have a single B-natural in the music – they’re all B-flats, as if the music were in F-major (one flat). What’s the best way to write this? If I notate the key signature as one flat, the performer will get the mistaken impression that the music is moving toward an F-major or D-minor key center, but if I leave the key signature as no sharps or flats, I’ll have to write in the accidentals throughout the score. What’s the solution?

If you’re familiar with jazz harmony, you know that what I’ve described is a song in C- mixolydian, and there’s a bit of a debate as to how it should be notated. It’s my personal opinion that since 99.9% of performers are only thinking about major or minor keys and other modes aren’t even considered, the best solution is to use the key signature that most closely matches the major or minor key center, and write in the accidentals manually throughout.

While there might be some jazz theorists who argue this is technically incorrect, it’s my experience that this cuts down on confusion among performers and decreases the number of mistakes made during sight reading, since most musicians are only thinking in the major-minor paradigm. Avoid confusion and use the key signature that most closely reflects a major or minor key center.