I'm trying to analyse the harmony of a piece I've been given and I'm having trouble rationalising its use of non-diatonic chords.

For context, I figured I should show the chord progression of a phrase from this piece. The piece is in C major and the phrase is:
| C . . . | Eø7 . A7 . | Dm7 . . . | Fm7 . G7 . |
| C . Bb7 . | A7b9 . . . | Fm7 . . . | G7 . . . |

I attempted a Roman numeral analysis below:
| I . . . | viiø/IV . V7/ii . | ii7 . . . | iv7 . V7 . |
| I . V7/iii . | V7b9/ii . . . | iv7 . . . | V7 . . . |
(I'm also not certain that my Roman numeral for A7b9 is properly notated, feedback there would be much appreciated)

I'm concerned specifically with the viiø/IV -> V7/ii progression. I've learned that secondary leading tones are meant to resolve to a minor or major chord where the root is a half-step above, and as far as I can find, deceptive resolution works only with secondary dominants, not with leading tones. Is there an explanation someone could provide, or better, a resource that may help illuminate what's going on here?

  • 3
    I would view the Eø7 as iiø/ii because it looks like a minor iiø - V - i progression. And I wouldn’t think the Bb7 - A7b9 part are secondary chords, just a modal walk down from my POV May 29 at 14:49
  • @ToddWilcox Thank you! That demystifies a lot for me.
    – reyspawne
    May 29 at 15:00
  • 1
    Also, remember that things aren't really "meant to be" one way or another in music. Theory is like a description of common trends, not a rule book. Especially in such a chromatic progression which I assume must be Romantic or later. The reason chords like these are linked together is because they share common tones or they are close in terms of voice leading. That's how they're rationalized, before you even consider the harmony. May 29 at 16:05
  • To add to the comment by @ToddWilcox (Todd, you should just elaborate a bit and make that an answer), in jazz and pop music the m7b5 or half diminished chord rarely functions as a traditional vii. Instead it is used mostly as a ii chord in ii-V-i or ii-V-I to either tonic chords or as secondary dominants, in this case ii/ii-V/ii-ii. The root, instead of functioning as a leading tone provides the first note of the descending 5ths bass motion of the 2-5. May 29 at 17:47
  • 1
    And to add to @JohnBelzaguy's add, jazz and pop don't follow the strictures of classical harmony. Functional analysis was not designed to apply to them.
    – Aaron
    May 29 at 17:54

2 Answers 2


As Todd and others suggest, Eø-A7-Dm is a iiø-V-i progression in Dm.

I don't see anything wrong in labeling A7 as V7/ii. However, the preceding Bb7 seems to be a tritone substitution of E7 (note, both chords share the notes Bb, and Ab, which is enharmonically equivalent to G#), so that Bb chord would be a bII7/VI.

Then, Fm is a subdominant chord borrowed from a minor key, and in this progression it substitutes the Dm subdominant chord. So the whole sequence: Bb7-A7-Fm7-G7-C can be seen as originating from diatonic Em-Am-Dm-G7-C, with alterations added.

The two examples above show two very typical harmonic building blocks for jazz tunes: II-V-I, and II-V. Sometimes it even goes III-VI-II-V-I, where "I" might not be the actual tonic of the tune, but a temporary modulation. It may not even be in any reasonable relation to the main key of the tune, so using roman numerals to notate secondary functions might not be the most practical.

The chords progress in descending fifths; minor chords might be replaced with dominant chords and become secondary dominants (very common for vi to become V7/ii); dominant chords may be substituted with dominant chords a tritone away resulting in a chromatic downwards motion. These are the first things look for when analyzing the harmony.


The root progressions follow two common root movements with quite a bit of "decoration" involved. The first is a cycle of fifths (almost) E-A-D-G-C with diminished or seventh (or both) chords rather than major or minor. The second phrase is a bit more irregular; the C-Bb-Ab is a step-wise walk downward finishing with a ii-V (or IV-V, not much difference functionally) that would lead nicely back to the first part. One can usually move roots around the circle of fifths with any flavor of chord attached. Linear root movement (or linear bass movement with roots moving by thirds or fifths) leads to smooth harmonies too.

Both are rather common in Common Practice Harmony and both usually make voice-leading easy.

There are probably other analyses possible.

  • Note, there is A7, not Ab7 in the second half May 30 at 6:16
  • You're right. My second phrase analysis isn't correct. It's a bit more irregular. I'll look at it, but I don't think I'll see anything except Tonic-SomeChords(bVII7-VI7)-PreDominant-Dominant. The bass melody (C-Bb-A-F-G) seems to be the driving force. Without playing the pattern in various voicings, I'd say it sounds fine though.
    – ttw
    May 30 at 12:52

Your Answer

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

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