Diatonic harmony is boring. Random dissonance is boring too. How do you make your music less predictable, but in a logical-sounding way? Especially if you want your harmony to sound “jazzy”? One reliable technique is to use secondary dominants. The idea is to treat each chord in a key as the temporary center of its own key, and precede it with its own V7 chord. This diagram shows all twelve possible key centers on the inner ring, and each one’s V7 chord on the outer ring:
The idea here is that if you pick any dominant chord on the outer ring, it will sound good to resolve it to its neighbor on the inner ring.
Here’s how to use secondary dominants. First, write a chord progression that’s entirely diatonic to some key. In C major, that means you should choose from C, Dm, Em, F, G, Am, and Bdim (that last one is rare outside classical music.) For example, you might write C, Am, Dm, G. This is pleasant and uninteresting. But now, think of each chord as a key center, and insert each key center’s V7 chord, like so: C, E7, Am, A7, Dm, D7, G. Much better!
Why does this sound so good? The boring progression is comprised entirely of notes from within the C major scale. These notes all work together, but they are overly familiar, and they don’t have much potential for surprise or drama. But now look at the secondary dominants, the chords in red. Each one contains a note from outside the key of C. In E7, it’s G-sharp; in A7, it’s C-sharp; and in D7, it’s F-sharp. In each case, the outside note is the third of the secondary dominant chord, and is also the leading tone of the target key. Each of these notes is quite dissonant out of context, but they retroactively make sense when you hear them resolve up a half step to the root of the temporary key.
You can also use secondary dominants for chords that are outside the key, and this is where the real fun begins. The diagram below shows some possibilities for doing this. The green and blue chords are native to C major, the brown chords are commonly used ones borrowed from the parallel key of C minor, and the grey chords are more rarely used. The outer ring shows the dominant chord for each inner chord.
The inner ring chords will all sound good more or less in any order and any combination. You can precede any inner-ring chord with its corresponding outer-ring chord. This should give you enough dissonance to keep things interesting without having your harmony completely descend into chaos.
I have barely scratched the surface of the possibilities of secondary dominants. To really explore this concept fully, you’ll need to consult a good theory text. I recommend Something Borrowed, Something Blue by Andy Jaffe. But this post should be enough to get you started. Above all, trust your ears and have fun.
Impressive!
This is a simple and effective principle! Thanks for sharing it.