I watched the video

where the educator showed a (diatonically ?) reharmonized version of the minor key pop song 'Shape of You' where the original harmonies were:

Dm | Gm | Bflat | C

One of the altered versions was then:

F major | F major/A |Bflat | C

(actually he goes on and changes more and more of the harmony...)

While I understand that the first F being the relative major to Dm should be expected to work well, I don't see why this should be the case with substituting F/A for Gm.

Is there an explanation why changing the bass note to the third of F (respectively the fifth of Dm) generates a sound that is (at least for my ears) somehow alike to Gm? It seems surprising to me that changing the bass note of the relative major should somehow make this more "subdominantic"

  • This is unanswerable unless you post a link to the video. There is nothing to say about "changing Gm to F/A" unless you know something about the context - e.g. what is the melody line here?
    – user19146
    Jul 9, 2018 at 10:35
  • 2
    sorry, first post in the music.se, with my expertise lying in a whole other direction of science the apropriate amount of background to give is still a little gamble when I ask about music. The song is Ed Sheeran's Shape of You, the video is by Adam Neely, I'll edit the post to include a link to the video.
    – Chen Huang
    Jul 9, 2018 at 10:38
  • 2
    There many be harmonic ambiguity in the melody that allows for more than one harmonization. Or, if it sounds cool do it. The Fmaj over the Gm produces a G-11. The A in the bass is the 9th. It all works but the 11th is an unusual choice w/o a motive. If it goes with the melody it will work. If you judge the new chord rel to the original it may seem odd, but if you write the melody and harmonize it from scratch you may not make the same choices as the original artist.
    – user50691
    Jul 9, 2018 at 12:30

2 Answers 2


Think of this as a two-bar substitution rather than as two separate substitutions for individual chords. He isn't really thinking about the old chords when coming up with the new chords. For that reason, this is better described as a reharmonization than as a chord substitution.

The constraint he has placed on himself are:

  • rewrite the first two measures
  • transform the progression so it sounds like it's in FMaj (the relative major to the original key, Dmin)

So in essence, he took the old progression:

| Dm | Gm | B♭ | C |

and changed it to:

| ? | ? | B♭ | C |

then said "this is in the key of FMaj," and remembered that a very common progression is:

| FMaj | FMaj/A | B♭Maj | C |

While this sort of I-IV-V progression is extremely common, it's not the only way to approach a IV chord. Other ways to reharm the first two bars in F might have been:

| F | Dm | B♭ | C |

| F | Cm F7 | B♭ | C |

| Gm C7 | Cm F7 | B♭ | C |

We could keep adding chords and complexity (e.g., | F | Dm D♭m Cm F7 | B♭ | C | with the D♭m Cm F7 occurring on the 'and' of 1, the 'and' of 2, & the 'and' of 3), but if you try these, you'll notice that some don't work as well with the melody. And some might deviate too far from the song's original feel. Those sorts of factors are also used when creating a reharmonization.

  • while I daren't judge if this is "correct" or not it opened a new angle in which to look at the problem which actually is enough for me at the moment! Thanks! That is not to say other thoughts on this were no longer appreciated
    – Chen Huang
    Jul 9, 2018 at 12:58
  • I love the question and think it's a great fit for the site! Something I should add is that reharmonizations often change big aspects of a song, including the chord functions and the melody. The Dirty Loops version of Baby is a dramatic example of how chord functions can change. In the video you linked to, the first example talks about changing the melody to better fit the new chords.
    – jdjazz
    Jul 9, 2018 at 13:04

There are many explanations as to why Adam Neely's harmonization (subjectively) works, so here's a few.

I think one of the reasons that this harmonization worked is that by starting on the F major, Neely basically just gave the next chord the function of leading into the B♭ chord. The F/A, being inverted, doesn't sound too different from the F chord, but the A in the bass would lead nicely into the B♭ of the B♭ chord. So the F/A sounds different and makes cool voice leading.

It's also harmonically similar to the iii chord (Am=A-C-E), and one common and useful resolution of the iii chord is to the IV chord, which so happens to be B♭. One could think of the first two bars as having tonic function, and Neely simply replaces what would be a (subjectively) boring I chord with what sounds like a iii chord, but the important thing is that it sets up the IV chord.

Another explanation is that there's a theory of chord substituting that says that any set of chords with the same function are easily substituted. The F/A has essentially subdominant function (even though F is a tonic chord, when inverted it doesn't feel as resolved, or "all the way home", as the F chord in root position), and so does the Gm, and both lead nicely into B♭, so they can be interchanged, giving the particular harmonization Neely derives. Note that this theory I mentioned isn't a universal, hard-and-fast rule so much as a guideline some reharmonizers subscribe to. So, both have the same function, so they're the same.

Two more bonus ideas: The melody over the second chord (A-G-F-F-G) would support most strongly the tonic function chords, since the F and A are emphasized and are part of the tonic chord, so one would probably not want to replace the Gm with anything drastically departing from the I chord; fortunately (or rather intentionally), everything I just described also describes the F/A. And of course, I'm going to get the obligatory "Music is free, and rules are meant to be broken, if it sounds good go ahead" that always shows up in the comments out of the way right here. I agree with it, but since the question isn't debating the power that music theory has to influence artistic decisions, I don't think questions like this really merit that comment.

Adam Neely is awesome.


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.