I'm writing a song in 4/4 in A major, and my melody ends first on the 7th degree leading tone (G#) and then the second time on the 6th degree (F#). These are supposed to be unstable tones, yet if I use supporting chords that have these notes (yes, I know some are not from A major), the tones don't seem like they need to go anywhere and feel resolved. It got me thinking that if these tones can feel resolved depending on the chords I use to accompany my melody then really there is no such thing as stable or unstable tones, and it just depends on the chords I use? Is this right? Here is my song so far.
mm. 1-8 without an A major chord

New harmony. mm. 1-8 with resolution on A major chord


2 Answers 2


Not certain at all what you're trying to ask, but here goes.

Taking the scale notes of the major scale. Supposing the bar in question is a I bar, harmony wise, 1,3,and 5 are generally considered as stable notes. Why? Because they match the underlying harmony. 2 and 4 are often considered as unstable, which is where the suspension and retardation (sus 4, 'sus'2) which usually wants to return to 3, is relevant.

6 is stable, and that refers to the relative min where 1 and 3 are the same anyway. 7 is the controversial one. In jazz, it's deemed to fit so well that just about any I bar can have it in the chord as well - Imaj7. In other situations, as it's the leading note, it pulls strongly towards the 1, so is seen to be a bad fit.

All that on the I chord. Not going any further with this explanantion, but using a different chord to base on, as pieces will contain multiple chords, but looking at, say, V, those 'stable/unstable' notes will obviously not work in the same way. A 1 note in a V bar will sound like a 4 note in the I bar. Unstable.

So there cannot and is not a list which will say 'in key Z, these are stable notes'. How can there be?

  • aren't 1,3 and 5 stable and 2,4,6 and 7 supposed to be unstable regardless of the chord? or does it refer to the chord scale degrees not the major scale degrees?
    – user35708
    Feb 20, 2021 at 15:29
  • I am, of course, referring to the original scale degrees. I did say scale notes. It won't make sense otherwise! And 'supposed to' means very little here.
    – Tim
    Feb 20, 2021 at 16:06

Stability and instability are highly dependent on context. If the only context is the A major scale, the G#, the leading tone, will be highly unstable. However, embed that note in an E major chord, it is much more stable.

This is part of the essential foundation of functional harmony. For a piece firmly establish in A major (as shown above, yours is not), a lone G# will stand out as needing to resolve to A. That same G# will be more stable within the E chord, but the E chord itself will be unstable -- in part because of the presence of the G# relative to the larger context. This is what makes a half cadence what it is: locally stable, allowing for the effect of a cadence, but globally unstable, requiring an eventual resolution.

  • As in my answer, that leading tone very often features in a I chord, in jazz particularly. As a single melody note it's teetering! A half cadence is an imperfect one, I think?
    – Tim
    Feb 20, 2021 at 16:43
  • @Tim Yeah, British vs. American, I think. However, not to be confused with the "imperfect authentic cadence", when the highest note of the I chord is not the root or when one or both of the V and I chords is inverted.
    – Aaron
    Feb 20, 2021 at 17:25
  • We don't do IAC per se over here! Imperfect is I>V, which sounds like your half-cadence, neither of which terms depict what happened too well...
    – Tim
    Feb 20, 2021 at 17:31
  • Well, this being 'Murica, our way is the right way. But neither really applies to the OP composition, agreed.
    – Aaron
    Feb 20, 2021 at 17:53
  • 1
    I'm saying there's no strong cadence that firmly establishes A in one's ear. The harmony — whether implied or actual — is what determines the key, and the harmony in this case is ambiguous.
    – Aaron
    Feb 22, 2021 at 2:38

Your Answer

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