I am currently looking at this tune but I'm a bit confused by a part of the progression.

It seems to be using chords built mostly from the C# harmonic minor scale.

So, my current understanding tells me we could use:

  • C#min
  • D#min
  • EMaj
  • F#min
  • G#Maj
  • AMaj
  • CMin

Or extending versions of these (eg G#7)

However, at one point, it goes:

C#m, A7, A13, D/F#, Ddim.

Can anyone help me understand what is going on here?

I thought perhaps the A7 or A13 would be followed by a Dmin (secondary dominant) but is doesn't seem to be the case.

Why do the D/F# and Ddim work?

Any advice appreciated.


  • 4
    Your chord list for Harmonic minor diatonic chords is Wrong. Three of the chords contain non diatonic notes
    – Fergus
    Commented Feb 20, 2014 at 20:54
  • I think this song is Beethoven’s moonlight Sonata played backwards. Yoko was playing it on piano or something and John had the idea to reverse it and incorporate it into a song.
    – Mps
    Commented Feb 3, 2021 at 3:02

5 Answers 5


You found the one unexpected chord.

See here: http://bostonglobe.com/ideas/2012/07/07/when-computers-listen-music-what-they-hear/hzdqdfgsIgEPiWPRe66U8J/story.html

Using Music21, which was designed by Michael Cuthbert and his MIT colleague Christopher Ariza, Harvard physics doctoral student Douglas Mason analyzed Beatles songs, running more than 100 of them under the microscope and discovering that the majority of them were built around one—and only one—highly unexpected chord. “You expect C to appear in a song in the key of C, but you wouldn’t expect a chord that almost never appears in a song in C, like E flat, because it’s really out of key. But the Beatles did stuff like that all the time,” said Mason. “They’ll have a song in major but they’ll bring in a chord from the minor key, and that chord will act as an anchor for the whole song.

  • 1
    Really interesting. Any ideas on why it "works"?
    – user28
    Commented Nov 2, 2012 at 16:52
  • 3
    @Read. I once read that in ancient China they would make a temple with all columns identical and one upside down. The idea is that having a clear symmetry slightly broken is more pleasant than perfect symmetry. In the same way, it is pleasant to have chaos or apparent chaos in which a symmetry is barely seen. It is a general aesthetic principle that works in all forms of art. Commented Jun 14, 2013 at 13:05
  • 3
    There is nothing unexpected about that chord. Google Borrowed chords. The bIII has been borrowed from the parallel minor key for centuries.
    – Fergus
    Commented Feb 20, 2014 at 20:18
  • 3
    Unexpected like finding a Jalapeño pepper in your hot fudge sundae, but not unexpected to someone who eats chicken with Mexican chocolate mole sauce on top. Commented Feb 25, 2014 at 4:05

However, at one point, it goes:

C#m, A7, A13, D/F#, Ddim.

Can anyone help me understand what is going on here?

The A7 seems to be functioning as a tritone substitute for D#7 which would pull to the G#7 dominant of the C#m key.

Try playing these variations:

  • C#m A7 G#7 C#m

  • C#m D#7 G#7 C#m

  • C#m A#o7 G#7 C#m

A7, D#7 and A#o7 all share the G-C# tritone. Approaching the dominant of a minor key with a diminished harmony (the third example) is common in classical music.

But in this tune, it never completes toward the G#. Instead we are given a signal that it's going in a different direction: the harmony lingers on A, following A7 with A13, perhaps to help shake off our memory of A7's remote connection to C#m, and that dominant becomes a pivot to go to D instead.

It doesn't really matter whether it is a major or minor D at that point; the A13 could go to either.

Why the D works is that it provides a path back to the C#m key again.

What is possibly going on here is that the D functions as a subdominant of G#m A, whose relative minor key is F#m. This D subdominant of A functions with respect to F#m similarly to how F is related to Am in the key of C.[2014-2-19 edit] The diminished chord then works as a substitute for C#7 or E7, the dominants of F#m and its relative major A, respectively. It shares three notes with both of these.

Compare these:

  • D Ddim C#m F#m (like F Fdim Em Am in C major! Notes of Fdim are in the Phrygian mode of A harmonic/melodic minor)

  • D E7 C#m F#m (like F G7 Em Am)

  • D C#7 C#m F#m (like F E7 Em Am)

But we don't have a F#m. It is C# that is targeted, seemingly as the root of the Phrygian mode of F#m/AMaj. So D is actually IIb: something like the Phrygian II or Neapolitan.

Compare to the very bold Neapolitan cadence (IIb-V-I).

  • D G# C#m (The Neapolitan D should typically be in a 6th inversion)

This is like F B Em.

The Ddim choice is interesting. It creates a very soft transition compared to the bolder choices of E7 or C#7 because it stays almost entirely in that Phrygian scale.

So the hypothesis is that the path from the D rooted tonality back to C# via the Ddim is something like the Phrygian-based Neapolitan IIb-V-I device, but with IIb dim substituting for the V.

From the top:

C#m, A7, A13, D/F#, Ddim, C#m

Im (IIb7:G#) (V13:D) (IIb:C#) (IIbdim:C#) Im

From the C#m to the A7 is like the familiar Am to F in C major, but with F altered to a dominant. This A7 behaves as a tritone substitute that wants to pull half a step down to G#, but instead it lingers, and turns into a secondary dominant pulling to the D root, and then we have that aforementioned path back to C#.


This is one of my favorite Beatles pieces. I think it has the best harmony in a POP song, along with God Only Knows by Beach Boys, and Julia in the White Album. I'll give it a try with simplified chords. I am not an expert, but this is how I see it:

Verse is functionally some kind of A minors mixture:

    Am F7 E7 F(7) Am F9 (asks for a E7)


A B C D (Eb-E) F (G-G#)

I think it's based in D-Eb-E bass-pivot-tone over A, and this G/G# mutations are only chromatically changed in order to provide some harmony colour to that bass, using traditionally rules eventually no matter the key. IT sounds to me a bit like a chromatic bassline James Bond theme or Is there anybody out there by Pink Floyd.

At a given moment so, we can have the feeling of natural minor(E-G), [harmonic diminished(Eb-G#), ] harmonic minor(E-G#), [natural minor(E-G), ] melodic minor (Eb-G)

It has the flavour of the typical blues-funky progression:


Am7 D7 F7 E7 

For the second part, I think the thing is that root modulates subtly half step forward, so it's now Bb, the rest of the central structure remains almost the same that's why it fits

A->Bb B C D (Eb-E) F (G-G#)

and again harmonized bass is played following the descending over from E-Eb with D-(C#)-C (like ending of Blackbird phrase), so after a C# (Bb-C#) we can move the root from Bb to A again and so:

  • If we move half step back C# we got an Am (A-C) chord that sets Am key again
  • If we move half step forward C# we got a D (D-A) chord that allows stablishing A as A MAJOR key, it works in my ears as a piccardy third

Furthermore A major contains E7, which is the final chord, and works as a V7 pivot for coming back to A minor

A -G#-A
.....B- C
E -E -E


Bb Bbm/Bº (D E7)


The explanation of the A chord being a tritone sub on the way to a G# Dominant, but instead going elsewhere, does not speak to me.

Just for one thing a tritone sub would seem to me to be always full of tensions to resolve. But this A chord appears quite at rest to me until it introduces its own dominant extensions and moves on.

We know that being in a major key is almost like being in it's relative minor at the same time. In the same way, Being in a minor key is almost like being in its relative major at the same time. The Beatles constantly used these mini "modulations" within a key's RELATIVE key centers. If we are in C# minor, we are kind of in the key of E major at the same time.

In the first section we hit the ii V chords in minor, which could lead us back to C#m. Instead however, we use the G# (V) as a leading tone chord taking us to the A chord AS THE IV CHORD of E our relative major. And the A chord SOUNDS like a subdomininant chord. Until it turns into a domininant chord in the next measures, which, as a secondary dominant leads predictably to D.

D to Ddim is certainly an unusual use of the diminished chord. I think D diminished is best seen here as an almost complete, altered G# Dominant leading directly back to C#m. Note that it is played as a D diminished triad, not a full diminished 7.

If it were played as a full diminished 7, that would add the b3rd of G# so that's not optimal. As a G# Dominant then, what we have spells 1 b5 6 BUT, a grace note is played before the 6 which is the b7 of G#.

So now you have a G#7(b5,13) chord with no 3rd which gives you a hanging effect. The grace note however also provides a chromatic walkdown leading to the 3rd of C#m which helps define the resolve.


In functional harmony terms, the Ddim chord is a common-tone diminished chord1 expanding the Dmaj harmony. Ordinarily, this would progress back to Dmaj.

But now consider the Dmaj chord as the Neapolitan chord2 leading to C♯min. In this context, the Ddim serves as a passing chord connecting the Neapolitan (D) with its tonic (C♯min). The A♭ in the Ddim is reinterpreted as G♯ in C♯min.

To summarize:

  1. Dmaj is a pivot chord,3 serving as IV in the preceding Amaj context and ♭II of the succeeding C♯min.
  2. Dmaj shifts to Ddim, forming half of a standard common-tone diminished sequence, but also serving as a chromatic passing motion toward ...
  3. C♯min, which serves both as cadence point and beginning of the new verse.

1 See A chord progression from Leavitt: how to analyze the diminished chord for an explanation of common-tone diminished chords.

2 See What is a Neapolitan 6th? for an explanation of the Neapolitan chords.

3 See https://en.wikipedia.org/wiki/Common_chord_(music) for discussion of pivot chords.

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.