Chord Progression in Merry-Go-Round Of Life

Question

In Joe Hisaishi's Merry-Go-Round Of Life, I'm trying to understand harmonically what is happening from measures 15 - 22. Below is my chord analysis in this G minor piece:

Measure 15: Bb/F [III]

Measure 16: Gmsus2/E [iv/ v]?

Measure 17: (A b2 b6 7) [V/v]???

Measure 18: Dm [v]

Measure 19: F/C [V/III]

Measure 20: Bb maj7 [III]

Measure 21: Am7 add4 [v/V]

Measure 22: Dsus4 [V]

I'm particularly interested how measures 16 and 17 work. For measure 16, The E in the bass works, since it's in the key of D minor (measure 18) and it is a half step down from the F in the bass in measure 15? It's hard to wrap my head around measure 17. In one train of thought, the F natural, Dflat (enharmonic to Csharp) and G natural are in the key of E minor (measure 18), but aren't in the key of A major, yet this seems like some type of A major chord? How do I analyze this chord/measure?

Answer 1

1. As a first step, there is an error in the arrangement. There should not be a Bb in the chord at m. 15. It should just be a Dm triad, which can be clearly heard in the original piece (link timed to m. 15).

2. The next step is to recognize that the analysis from m. 15 to m. 20 is more simply given in D minor: i.e., the "key of v"

3. Also note that the Db in m. 17 is better understood as C#.

4. Finally, many of the notes are not chord tones but rather accented neighbor or passing tones. That makes the analysis much simpler.

With all of that in mind, here is a reduction of the score with analysis.

This clarifies measures 16 and 17. They are part of a standard ii-V-i progression in D minor. The key to understanding this is to interpret the melody A in m. 16 as an accented upper neighbor (i.e., not a chord tone), and, similarly, the two melodic Fs in m. 17 are passing tones (the first being accented).