I found this chord progression on my guitar last night :

C - Csus4 - Gsus4 - G - Gm7add(#5) - Am7add(#5) - Asus4 - A

It sounds pretty nice to me but when I look at it theoretically I just can't see why. What is especially bugging me is the G -> Gm -> Am part. Is there some kind of keys changes or modulation happening ? Is there a way to explain why this feels so right ? (if voicing of those chords is relevant I could edit them in)


Although unusual in terms of 18th-century common-practice music theory, this and any connection of harmonies doesn't need some sort of justification. As a musical concern, all that matters is you like it and have ideas about how you'd like to use it.

However, I think there's one aspect that could be a partial explanation of why the progression works for you: common tones. Between every one of the chords—including the ones that are most outside of a single key—share one or more common tones with its predecessor and successor. This is a common technique in a lot of late 19th- and early 20th century European music, and is sometimes referred to as "linear harmony." There are many passages in, for example, Liszt, of successions of chords where some notes are held while other pitches move by a step. It's possible that this is a place where your specific voicing might be significant, because, in a linear harmony passage, the more that common tones are held in the same places and the stepwise connections are truly stepwise (rather than jumping around in sevenths and ninths), the more that a listener can follow the thread.

For example, the succession of chords

Cmin – B – G – Bmin

is certainly not standard in a common practice sense, but can be, especially in smooth voicings, and quite clear connection. The move from Cmin to B holds a note common (the Eb/D#) while the other two notes slide down a half step (C to B, G to F#). Neo-Riemannian theorists call this connection P', and it's fairly common. The move from B to G is called a chromatic mediant relation—two triads of the same quality whose roots are a third apart—which always will have one common tone, one chromatic alteration and one stepwise change. The final move from G to Bmin has two common tones (B and D) and one half step move (G to F#). The Neo-Riemannians call that one L.

Your progression is more sophisticated than that, and whether any particular example appeals to you personally is subjective, but I think the proliferation of common tones throughout you example is at least part of the sauce that makes it stick together.

  • 1
    That P' is also SLIDE, correct?
    – Richard
    Jun 6 '17 at 11:32
  • Yes! That's Lewin's term I believe. I find P' to be a bit more useful, because it's shorter and it makes an explicit connection to the fact that it's the inverse of the Parallel operation, but SLIDE is well-known too. Jun 6 '17 at 11:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.