I have a song with a part that I am wanting to play arpeggios over with the guitar. But just using the actual chord tones sounds boring to me. So I have started with an arpeggio pattern on the first chord that I am then repeating.

What I am doing is this:

The song is in E minor. On the part that starts with the G7 chord I start playing the arpeggio: A B D F# (and then up to the higher A and then back to F# and D, so just using these 4 notes). It sounds good. And then I start finding out what it is that I am doing there.

So I look at the scale chart for the E natural minor. The third chord is Gmaj7 and that is the closest to the G7 chord that I am getting in this chart. I am in Ionian mode (it says). And I am playing the 2nd, 3rd, 5th and 7th degrees of this mode in my arpeggio. Okay, so far so good.

Now I am applying this same step to the other chords.

Next chord is E-7, which is Aeolian mode, and again I play the 2, the 3, the 5 and the 7 of this mode, resulting in the notes F#, G, B and D. This sounds all fine and dandy.

But now my question.

The next chord I play is B-7. Applying the same procedure, I am now in Phrygian mode (or the fifth degree of the E natural minor scale.) The second degree is "C" but it sounds absolutely terrible! It is better to use C# instead. I play C#, D, F# and A, and it sounds cool over B-7.

So I have to break with my formula, and I am fine with doing so, but I would love to know what is going on on a more theoretical level to understand what is happening here.

Appreciate any insight you might have!

  • Even if there's a C in your scale chart there's no need to stick to C for your whole piece.
    – PiedPiper
    Dec 3, 2020 at 10:40
  • 1
    Don't be trapped by the 'rules'. It makes more sense to use ears rather than what you think are 'rules'. Guess why 'rules' is written like this!
    – Tim
    Dec 3, 2020 at 10:58
  • Don't get trapped by rules, I get that, but I am interested to know why C# sounds awesome on B-7, and C does not. Even though it belongs to the scale and C# does not. I am interested in knowing what rules I am breaking, haha, if you get my drift!
    – Jelmar
    Dec 3, 2020 at 11:52
  • 1
    I don't understand how your A B D F# gets close to what goes over a G7 chord. G7=GBDF. 2 out of 4 notes hardly constitutes chord, and the other 2 aren't in that chord.
    – Tim
    Dec 3, 2020 at 12:39
  • 1
    Funny enough, your A B D F# is more or less Bm7, not G7; that already makes a switch to B7 feel a little "ad hoc". Also, the second degree of B would be C#, not C (must be a whole tone up from B). Dec 3, 2020 at 12:52

2 Answers 2


This is what I understand you have in terms of chords and pitches...

G7 + A B D F# = G(?)9

Em7 + F# G B D = Em9

Bm7 + C# (not C) D F# A = Bm9

E G B D F# A C# E = E dorian

(The stuff you mention about scale and mode and chord charts sounds like the "chord/scale system" which I think is not helpful for understanding music theory/tonality/harmony, because it teach about keys.)

The first thing I notice is your choice of pitches in the arpeggio guitar part extend the chords to ninth chords. With A, F#, C# you add a major ninth to the set of pitches for each chord. The other tones are already in the chord part.

The quality of the seventh on the G chord is ambiguous, because there is F natural in G7 and F# in the second part. It's hard to say much more about that chord without know other details. It seems to be a non-functional dominant seventh chord. That means it doesn't resolve in a dominant to tonic way. That isn't a problem, but if you want to say more about it, more detail is need. For example, the voicing and voice leading could be important. If it was like this...

enter image description here

You might say about the cross relationship between F/F#:

  • the dissonance is mitigated by burying one of the tones in an inner voice
  • the F natural moves by smooth voice leading
  • the dissonance is places into a fairly high register (lower starts sounding muddy or clanging)
  • F# in top voice and B in the bass outlines a perfect fifth in the outer voices mitigating the sense of F# as a dissonance

That's just my example. An analysis would depend on what exactly you are doing. But those are some of the factors you would look at: inner/outer voices, voice leading, register, etc.

Your avoidance of C natural is important. It ensures the Bm7 is extended to a minor ninth chord - with a major ninth - which is the standard minor 9th chord, and it also adds the crucial pitch to make E dorian. In an indirect way the absence of C natural supports the notion that the G7 is non-functional. C would be it's tonic, but you exclude it.

  • Hi Michael, a great answer, thank you. It sheds a lot of light on what is happening. Not being burdened with a lot of theoretical knowledge, I sometimes struggle to create things that are fitting, yet INTERESTING. I find it is often not good advice to 'use my ears' as they often resolve things in a conventional way. I really have to work at it, and sometimes think it's interesting to find procedures. I now have some things that I can use to dig in more. I didn't realize I was using the 'chord/scale system', but that's the trouble: these youtube teachers never reveal their sources!
    – Jelmar
    Dec 3, 2020 at 19:41

Not directly answering your question, but pieces don't 'change modes' for each chord played. If four bars in key G were G, Am, C, D7, then all of that would be G Ionian. It seems you would consider it G Ionian, A Dorian, C Lydian, D Mixolydian. Modes don't work like that.

There's probably where you get confused. ALL the modes from a parent key use exactly the same set of notes. That's one of the points of their existence. And pieces are generally in one mode throughout.

When you get to minor keys, there isn't just the one exclusive. Between them, there are more than the 7 notes we find in major keys. In fact, there are 9 that are included in all minor scales/keys. So thinking like you do, changing mode for each bar/chord, any 'rules' get screwed up anyhow!

Use whatever notes you think sound appropriate. Sounding good = probably is good. Someone will then come up with a good reason (a 'rule') why. But that should be after the fact, not before it.

  • Thanks for this, Tim. I understand that the piece does not change modally, and that it is one key. But it is also my understanding that each of the chords in, say the key of G, can have a mode associated with it. And I have a chart that says so. All I am actually doing is extracting the 2nd, the 3rd and the 5 and 7th note from each of the modal scales associated with each chord degree of the G major (or E minor scale). It is just a procedure. I am not interested why it is right or wrong, but only in why it works in some cases and not in others. So actually, yes, I am trying to find the rule.
    – Jelmar
    Dec 3, 2020 at 12:51
  • 1
    Use either E harmonic or E melodic minor notes, and you'll find that the first has C, the other has C#. Thinking in Bm, the supertonic (2nd) note is C#.
    – Tim
    Dec 3, 2020 at 13:01
  • 1
    @Jelmar Throw all those charts away, they're getting in your way.
    – PiedPiper
    Dec 3, 2020 at 13:02
  • @PiedPiper haha, yeah I get that is what it looks like! But I jut got them and the're real handy to figure out these arps!
    – Jelmar
    Dec 3, 2020 at 14:13
  • @Tim, these melodic and harmonic scales, that's probably where I can dig in more. Thanks!
    – Jelmar
    Dec 3, 2020 at 14:16

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.