I laughed when I just wrote the title of this blog post because it reminds me of a few axioms that I live by. You have heard me say them before but indulge me:
Those that know the least know it the loudest.
Those that know all the answers don’t know all the questions.
And here is a new one:
It is not easy to get a dogmatic answer from a wise person.
I laughed because this topic today is an example of one of those things where it is sort of wise to avoid dogmatic terms like “correctly.” You might think that it is easy to know whether to use F# or Gb in your music but it really is not always so clearcut. Someone out there may correct me and say this is all written out in some guidebook that all the music publishers swear to but I am not familiar with that book if it exists. I would not be surprised if every publisher consistently tried to follow a set of rules in regards to this topic but I would also not be surprised if there was some variance between publishers.
So, with that in mind, I am not going to be dogmatic today. I am however going to give you the rules I try to follow in the music I write out (though I am sure I break my own rules sometimes). The problem is in real life, some of the rules will contradict others so you have to make choices. Like I said, it is complicated. However, here we go:
- Make things as easy on the reader as possible. All other things being equal, a typical pianist is going to prefer seeing F# rather than Gb and Db rather than C#. There are rules below that will trump this rule but in the end of the day, the overriding principle should be to make the music as easy to play as possible. It serves no purpose to needlessly complicate it.
- Default to the notes of the key. If you are in the key of G, obviously you should not write Gb rather than F#. Even if you are in a tough key like Gb, you really should write Cb rather than B natural even though Cb is a bit harder on pianists to read.
- Spell chords by their appropriate letters. In harmony, chord tones are all named based on their distance from the root by letter. In a Bb7 chord, you are going to have a B, D, F and A because a dominant chord contains 4 notes all a third apart. Therefore, your accidentals need to reflect that. Don’t write A# rather than Bb or G# rather than Ab.
- When picking roots, consider function. If you are in the key of C and you have an F# chord, it is likely to be a diminished chord moving up a half step to G. It is following a functional rule of harmony where #IV likes to move to V. As a result, you should write F#dim rather than Gbdim. Or let’s say that you are in the key of C and you write a ii-V progression that ends on Db7. The chord in front of Db7 should be written Abm7 rather than G#m7 to respect what is happening functionally.
Confused yet? Now you know why I am not being very dogmatic. Consider these examples.
What happens if you build a diminished 7th on C? Should you write the last note as Bbb (following rule #3) or as A (following rule #1)? I don’t know what various publishers might do but I am going to going with rule #1 there and write A. In fact, diminished chords are one situation where I regularly let rule #1 trump the others.
What if I am in the key of C and see a diminished chord written on Eb/D#. Am I going to write that as Ebdim7 or D#dim7? Eb is easier to read but function has to be considered. I would actually look at the next chord which is almost certain to be a II or III chord. If it is a II chord, I am going to use Ebdim7 but if it is III, I will use D#dim7. (By the way, if I wrote Ebdim7, yes, I would write F# in the chord rather than Gb.)
Let’s look at something written. Here are the first few bars of the arrangement of “A Child of the King” I just published last week.
In the first red circle, I used Cb rather than B natural which seems to contract rule #1. However, I was thinking rule #3 because in my mind at least, this is a iv chord in first inversion. The iv is built on Ab so it should have a C in it rather than a B. Therefore, I went with Cb. The same thing happens in the first red circle on the second line.
In the second red circle on the second line, the chord on beat 2 is a C7 secondary dominant (V/ii) which contains a C, E, G and Bb. Therefore, I chose to use E natural rather than Fb. That it is easier to read is just a bonus.
The chord in beat 1 is a bit more interesting. If the C7 is a V/ii, the Gm7(b5) you see here is a ii/ii which leads to V/ii which leads to ii. When the target of a secondary dominant is minor, it is very customary to make the ii half diminished as you see here and add a b9 to the dominant. A G chord has a D in it rather than a C which is why I used Db there rather than C# and in the following C7 chord, it still needs to be Db rather than C# because the 9th of a C chord is D, making the b9th Db.
In a nutshell, what you will usually see me do is apply these rules in a reverse hierarchy. In other words, as you have seen here, rule #3 trumps #1. Here is why I take that approach: my ultimate goal is that the people that read my music understand it. I would rather spell chords correctly to make it clear what I am doing harmonically rather than just trying to make my music as easy to read as possible. In the long run, I think that approach make help people with their reading. I know that analyzing harmony certainly helps me read a lot better.
So that is what I do. However, I am not saying it is right or “correct.” If you have different thoughts, let me know. I would be curious to hear the approach that others use.