I've been learning about modes, and one thing I was interested in at first was to change modes while improvising (to get to know them better technically as well as to have a better understanding of what their flavour is). For instance, I improvised in minor pentatonic, and used its ambiguity with the relative minor modes to try to explore their differences. This lead me to question how to properly change modes in a convincing way, and even how to properly change key.

I don't know anything about the theory of modulation, but I wanted to try and experiment to figure out how it could be done.

I first picked two keys at random : E aeolian and G# aeolian.

I basically did what I think is a common-chord modulation. My idea was to pick the common notes between the two keys, and use them to transition from one key to the other. Here, the common notes are B E F#, which are a Bsus4 triad, so I used that chord to transition between the two keys. Going to E aeolian I let Bsus4 resolve to Bm, and going to G# aeolian I let Bsus4 resolve to B.

I used this very simple progression (on a loop) and improvised in E minor pentatonic/E aeolian, going to G# minor pentatonic/G# aeolian and back, to keep it as simple as I could.

Em Am | Em Am | Bsus4 B | G#m C#m | G#m C#m | Bsus4 Bm

My observation is that I managed to change key in a way that sounded pretty good (at least to me) when the change was from E to G#, but not the other way around, I found it much more difficult to sound even decent.

  • I thought it could be because Bm triad is B D F# and F# is not in Em pentatonic, while the B triad is B D# F# which are all in G# pentatonic. So I changed F# to the closest note in Em pentatonic, G, which means I'm left with a G triad. So I tried this :

    Em Am | Em Am | Bsus4 B | G#m C#m | G#m C#m | Bsus4 G

    But the chord themselves didn't sound that good (even if I tried to play G in first inversion to keep B at the bass, to have some sense of resolution for the suspended chord), and when I tried to improvise the problem remained.

  • I thought it could be because of the resolution sus4 -> minor which isn't as appealing as sus4 -> major, so I tried this progression :

    Em Am | Em Am | Bsus4 B | G#m C#m | G#m C#m | Bsus2 Bm

    Bsus2 isn't part of both keys, but at least the resolution Bsus2 -> Bm sounds better. But alas, improvising a melody that sounded good when going from G# aeolian to E aeolian was still a challenge.

Coming back to my initial progression

Em Am | Em Am | Bsus4 B | G#m C#m | G#m C#m | Bsus4 Bm

I don't get how this doesn't work. I thought it could also be that B doesn't have the same function in both keys, but it happens to be that B is V in E aeolian and III in G# aeolian. Functionnally it should work, even if I'm only mostly going from Em penta to G# penta and back, shouldn't it ? Even if that's the issue, I've tried to change B to G, which is III in E aeolian (see first point above) but it doesn't work !

Since I've never tried modulating while improvising before, I doubt it could be my improvising itself that is the issue. If it is, why is it so easy for me to change from E to G# but not the other way around ? And even if it's just my ear that is not accustomed to this change, why is there a difference really ?

I'm also thinking that maybe from E to G# your going up a major third while from G# to E you're going up a sixth, but I'm not sure I understand why it could be that that makes that much difference. Obviously E and G# are as far away from each other as G# and E are in the circle of fifths.

Is it maybe because of the relative functions of the common notes in both keys ? In E aeolian, E F# B are I, II, VI, and in G# aeolian they are II, VI, VII. I'm not sure were to go from there if that's the reason.

TL;DR : Using a common-chord modulation from E aeolian to G# aeolian and improvising, going from E to G# sounds good pretty easily, but not the other way around. Why is that ? I'm using the same method to come up with the chords, and I've never improvised in such a modulation context before, so there is no apparent reason why one change should be easier than the other.

  • Would you mind clarifying some of your chord qualities? Those G♯ and C♯ chords are all minor, correct?
    – Richard
    Aug 8, 2018 at 17:05
  • Yes indeed, mistake on my part !
    – Sylvain J
    Aug 8, 2018 at 17:25
  • (original post edited)
    – Sylvain J
    Aug 8, 2018 at 17:39
  • I think the same problems will be there even if you modulate from E major to G# major, and use diatonic notes. When modulating, usually, the ultimate chord before the change is V of the new key. Going from E to G# (majors) will need D#(7), while returning will need B(7). Same problem, I think. That lack of leading note, as Richard spotted, is a bigger problem.
    – Tim
    Aug 9, 2018 at 11:43

1 Answer 1


One possible fix might be found in this sentence:

it happens to be that B is V in E aeolian

I wonder if you're conflating V as a dominant major triad with the triad built on the fifth scale degree. Because the triad built on the fifth scale degree is B minor (v), which doesn't lead back to an E tonic as smoothly as a B-major triad (V) would.

Technically, this B-major chord doesn't fit your stipulation that you want this chord to be diatonic in the subsequent key (in this case, E Aeolian). But when I play your progression, the D♮ in that chord sticks out to me as a "wrong" note. And culturally, we're really accustomed to hearing a B-major triad move to an E tonic. This means that, just by playing a D♯ in that chord, you really strengthen the resolution to E and make clear to your listeners that you're looping back to the beginning.

And this doesn't interrupt your improvisation either; you can continue to use G♯ Aeolian above this chord if you like. It is, after all, the relative key/scale of this B-major chord!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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