I was playing the chords from Pyramania by The Alan Parsons Project. In the verse from 0:17 the chords are E, Bm7, E, Bm7, C#, A#, G, B. I get the first four chords, E and Bm7 fit in the same scale. But I don't understand how they can go from there to the following C# and the other chords. Please help? I want to understand how to write these genius progressions.

Here is a link to the song

6 Answers 6


As you said, the first four chords can be understood as chords from E mixolydian. Note that from then on the chords follow a downward movement in minor thirds (at least enharmonically): E => C# => A#/Bb => G and from there to B, the V of E. The downward movement in minor thirds is equivalent to going from a minor scale to its relative major scale (and that's also the way I hear it). Note that C# minor is the relative minor scale of E major; so by moving from E major / C# minor to C# major you effectively change from minor to major. This same 'trick' is done 3 times in a row.

  • The parallel scale/key of E maj. is Emin. Do you mean the relative ? Move from E to C#m = relative, then across to C# = parallel.I wonder if you can have a 'relative/parallel' change !!
    – Tim
    Commented Jul 9, 2014 at 7:29
  • @Tim: you're right, it's a language confusion (these terms are used differently in different languages). I'll fix it, thanks!
    – Matt L.
    Commented Jul 9, 2014 at 7:31
  • Thank you for the answer. I think I get it, but not totally. It still seems completely random to change from a minor chord to the parallel major chord, or am I seeing this wrong?
    – teije99zz
    Commented Jul 10, 2014 at 12:42
  • 1
    @teije99zz: No, moving from minor to major or the other way around is very common, in classical music as well as in popular music.
    – Matt L.
    Commented Jul 10, 2014 at 14:01

"Music Theory" — or more specifically, "Common Practice" or "Tonal" Theory — isn't designed with popular music in mind, and popular music frequently isn't constructed with music theory in mind.

View #1: Voice leading

In the case of this particular chord progression, I find it productive to consider it in terms of voice leading. That is, imagine sitting down at a piano and starting with a certain chord — E major in this case — and then just changing one or two notes at a time. Doing so can give a very coherent chord progression, theory aside. In this case, the fact that it eventually arrives at B major — the dominant chord relative to E major — gives an overall key-of-E-major sound, even though none of the other chords are native to that key.

Here is the chord progression itself, notated in E major to emphasize the "alien" nature of most of the chords, indicated by the need for accidentals.

X: 1
T: "Pyramania" chord progression
K: E
M: none
L: 1/1
"E"[EGB] | "Bm7"[=DFAB] | "C#"[C^EGc] | "A#"[^^C^E^A^^c] | "G"[=D=GB=d] | "B"[DFBd] ||

Here is the chord progression rewritten voice-by-voice without regard for key signature. Notice how each voice moves stepwise. (I've substituted enharmonically equivalent chords in some cases for ease of reading.)

X: 1
T: "Pyramania" voice leading
K: none
M: none
L: 1/1
[V:V1] "E"B | "Bm7"B | "Db"_d | "Bb"=d | "G"d | "B"^d ||
[V:V2] ^G | A | _A | _B | =B | B ||
[V:V3] E | ^F | =F | F | G | ^F ||
[V:V4] E | D | _D | =D | D | ^D ||

View #2: Root movement

Chord progressions in popular music, and jazz as well, are often constructed around the root movement of the chords, allowing that motion to define the overall sound, while allowing the chord qualities themselves to provide color.

Through that lens, the chords rooted on E and B are easily explained as serving tonic and dominant functions.

Note now that the roots of the other chords — Db, Bb, G — all come from an Edim7 chord (E G Bb Db).

Having stripped the chord progression down just to root movement, the progression can be explained in terms of more conventional theory.

  • E Bm7 E Bm7 now can be viewed as a prolonged tonic chord.
  • Db Bb G can be seen as a prolonged common-tone diminished chord.
  • B is the dominant chord.

Thus we have Tonic — CTdim7 — Dominant: a very reasonable "common practice" progression, but tied to common practice only in terms of roots of chords, not in terms of the expected chord qualities.


E, Bm7, E, Bm7, C#, A#, G, B

E Bm7 that's a root progression by fifth, ordinary, just as you said. Eventually the B chord returns to E, and the melody dwells on E. We could say it's nominally in E, but with all the chromaticism, it isn't strictly E major.

The progression repeats so write that out for clarity...

E Bm7 E Bm7 C# A# G B | E...

The progression is quirky. What the writer was thinking isn't necessarily how you could analyze it, but...

C# A# and G B both are pairs of chromatic mediants which have a particular colorful sound. Except for the B chord, those chords aren't in E so the question is: where did the come from? Sometimes a chromatic mediant can be explained as a borrowed chord, but that doesn't really make sense in this case, only the G could be called a borrowed chord, G being the bIII chord in the parallel E minor key.

Some other explanation is wanted. Sometimes a progression can have a dominant chord inserted into it. For example a plain I IV V could have a secondary dominant added I IV V6/V V. With that understanding you can then remove a dominant from a progression for analysis purposes to look for some other basic pattern.

If you take the B out of C# A# G B | E... then you get...

C# A# G | E...

...where all of the root are moving by descending minor thirds (allowing for enharmonic re-spelling.) Actually, if you do the re-spelling it's easier to see all of the relations are chromatic mediants...

Db Bb G | E...

...it's interesting that when the B is put back in place that both G | E and G B |... are chromatic mediant changes.

So, you could say all those quirky chord changes are chromatic mediants sequenced around roots descending in minor thirds. That might sound like a mouthful of technical mumbo-jumbo, but IMO if you play around with sequences of chromatic mediant changes it becomes obvious. Just play any of the various chromatic mediant change types, repeat that change type a few times, and you get roots descending by all minor thirds or major thirds.

You could flesh out such a sequence by making one of the chords a "target/home/tonic" - whatever name you care to use - you could make a two chord vamp on the target, or insert it's dominant somewhere. For example...

Take the basic chromatic mediant sequence G B D#(Eb), make G the target, vamp it with a F chord, and also insert it's dominant D...

|: G F7 G G :|: B Eb/G D D :| G

...something like that (I tried a slash chord for the heck of it.) It sort of has the same quirky feel of the Alan Parson's Project progression.

Was the writer of that song thinking this way? I have no idea.


Chromatic Mediant!

Take any two major triads with roots a major or minor third apart. They will always share one note between them. They have a very distinct, fresh, fantastical sound. The last four chords are all chromatic mediants.

C# to A#. Common tone is E# A# to G. Common tone is D (spelled C double sharp in the A# chord) G to B. Common tone is B.

(They're not new; they can be found plentifully in Beethoven and indeed much before.)


This relationship of chords in minor third distance was already used in Renaissance music: it is called in German QUERSTAND and in English FALSE RELATION


like Aaron mentions the music theory and harmonic analysis for the common practice period doesn't fit to explain all Pop music and neither the early music of Renaissance and before...


Chords can 'go together well' in many ways. Here are some of them.

If they are all constructed from notes of the same scale. (Beginners sometimes think this is the only one. But there's LOTS more possibilities!)

If both chords have one or more notes in common.

If a tension in the first chord is resolved, the surrounding notes can be pretty well up for grabs. Voice leading, to put it another way.

If both chords are the same shape. If they're adjacent - one step up or down - all the better. Cm9, Dm9, Em9 ... Roots a minor 3rd apart also work well. C - E♭ - G♭ ... This could 'explain' a lot of the song in question.

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.