I've had a melody for a while and I've slowly added some chords to it that seem to work with what I hear in my head, but I'm struggling to find a chord or two for it. Part of the problem is that the melody goes minor partway through but starts and finishes in major. Can anyone offer a suggestion for the chords in the middle?


  • I added the picture for you to improve the formatting, feel free to roll back the changes if you feel so.
    – Neil Meyer
    Oct 23, 2016 at 12:52

3 Answers 3


Whilst I agree broadly with Neil's suggestions, I sat down at a keyboard for 10 minutes and tried to approach the question at face value. Here's an idea:

| C - G - | Am - G - | F - Am - | D - - - | G - - - | Cm - - - | Am7 - - - | D - - - |

This feels like it should resolve to G, after the final D. I was voicing the Am7 as a C - G directly below the melody, and a A in the bass. Omitting the E avoided any clash with the D# in the melody.

The first D# in bar 6 is probably an Eb, given the Cm, but the second (bar 7) would make more sense as an D#, unless you decide to analyse my Am7 as some flavour of Adim.

I made a recording to illustrate the idea:

I've taken a bit of liberty with the timing, but hopefully it's a helpful illustration, at least.

  • Thanks! I'll play with the chord progression and see how it fits. I'm not opposed to modifying the melody if it has the same feeling. Its supposed to be kinda lonely but adventurous. Think lone space explorer type feel. Oct 24, 2016 at 0:23
  • @Rod I've added a recording to illustrate the idea.
    – endorph
    Oct 24, 2016 at 11:56
  • I like it. Its much closer to what's in my head. I'll try working with it this weekend. Oct 25, 2016 at 11:20

I have a different take on this. I don't mind the melody - it has direction, and it is possible to add a bass with a fair sense of direction. A certain amount of broken chord writing is fine in an instrumental melody. I've made a small change to the rhythm in m. 3 to clarify the phrasing a bit. I'm assuming flute and piano here. I won't claim a great deal of artistic merit for the accompaniment I'm presenting - it's workmanlike, and it's basically to give some idea of what can be done.

The thing here is that the theme modulates, possibly transiently. I see this as moving from C to G major to B♭ major, with the C major being introductory (as it was very easy to slip in D7 as V/V, and reinterpret ms. 2 & 3 as measures leading up to D7 in m. 4 taken as V of G major).

enter image description here

It would be correct to assume that I see the theme as being in an unsettled G major/minor (note the addition of a key signature), with references to the relative major of G minor (i.e., B♭ major) - note that the most prominent notes of the melody are G and D, even in the cadence to B♭. It would also be correct to assume that I don't see the theme as anywhere close to ending here - if I were working on this, I'd undoubtedly firm up the sense of G (while letting it continue to make suggestions of G minor/B♭ major), then take the piece fully into D major for a second theme. Because the there is a fair bit of subdominant emphasis in this opening, I would probably work through B minor to get to the second tonal area in D so as to counterbalance the subdominant-leaning harmonies, and I might even touch on B major (as a kind of "super-dominant") once I got there.

In short, I see this as the kind of theme with which you start to make a sonata.

Now, a few things to take into account when you write something like this:

  • Voice leading matters. You'll note that the accompaniment uses non-harmonic tones quite as much as the melody does, even though the style of the accompaniment is fairly chordal. (The 2nd inversion chord on C in the first half of m. 1, for instance, is as much a pair of lower auxiliary notes as it is a chord.) This allows you the flexibility to de-emphasise certain root movements and emphasise others. Note that the first bar here doesn't really have strong progressions in C, which allows G to pull everything into its orbit. Note that m. 3 has a deceptive cadence in C that is very easy to reinterpret as leading strongly into the dominant of G - it also creates continuity across the bar line.
  • As mentioned above, the very shape of the lines will have an effect on the tonality. Your melody strongly emphasises C, D and G through high notes, low notes, metrically-stressed notes and agogic accents. These are going to very naturally come across as melodic degrees 4, 5 and 1, which is why the full cadence on B♭ in my sketch comes across as hanging unresolved, which brings us to...
  • There is considerably more than one way to skin a cat. Getting to B♭ and harmonising ms. 6-8 in B♭ was a piece of cake. A deceptive cadence at the end to G or D would also have been easy - it depends on where you want to take the theme, and how quickly you want to get there. If I were working on it, the theme would extend at least 24 bars before moving away from G, and probably more. Your mileage may vary.

Now, I don't have a problem with the rhythm of the melody, given that other things can be done to articulate the theme, such as rhythmic differentiation in the accompaniment, texture, tessitura, dynamics, harmonic rhythm, root movements, etc., and even the accompaniment's rhythm can be fairly constant if the rest is in play (e.g., most of late Fauré - he was sneaky good that way).

The point here is that, when you are working with an open-ended melody of slightly extended tonality of this sort, you don't really want to think about setting it in terms of chords only - everything has to work together. By all means, learn textbook harmony and counterpoint - doing exercises in these firms up your ability to make notes do what you want them to do - but bear in mind that these are only facets of what goes into a composition, that is, different planes of the same gem.

In the meantime, while you're learning textbook theory, keep trying out things like this that are at the limits of your grasp (or maybe even a bit beyond), and pound away at them until they sound right, and until you have an understanding of why they sound right. This will take years: there are still times when I go back to complete sketches after shelving them for several years, because my understanding grew enough to tackle a challenge I couldn't handle previously. What a textbook can teach you is how common problems have been handled in the past; what it can't teach you is how to make everything work together to give form and direction to an idea that is a little (or a lot) off the beaten track. You have to come to your own understanding to accomplish that.

  • Wow! That's incredible. I appreciate the advice as well. I used to study piano but I haven't touched it in 20 years and my theory is rusty. I tried building the chords from the melody lines and it didn't make sense (which I see echoed in other comments here). I am thinking of this as an introduction to a longer piece but I was thinking of brass and timpani more than piano. Think Star Trek TNG. I'll keep looking at it. Oct 24, 2016 at 9:50
  • Piano was just tidier and quicker for an example like this. Harmonising the melody was tricky, but not that difficult. Ms. 2-5 actually form quite a strong, fairly conventional progression (allowing for a dual, modulatory interpretation of C & G). I could have done something as strong in m.1, but G minor/B♭ major don't fit well in a context of C major: a stronger C major would have left things adrift. By starting with an E minor harmony moving to a deceptive cadence in A minor in m.3, we have an overall skeleton of vi - ii - V - i through to m. 5, and the first 3 bars give G a modal tinge.
    – user16935
    Oct 24, 2016 at 11:32

The key of this piece is somewhat ambiguous. It starts out looking like D minor even though it does not have d minor key signature.

In the second bar, we have what looks like a, A minor chord that can function as a dominant but for some reason the leading tone is left unraised. It can be construed as the natural minor form as it is descending but for a dominant chord, you really need a raised leading tone to give clarity to the dominant function.

Then in bar three you have what looks like some sort of chord revolving around a C note, which is again not raised and does not resolve properly.

In bar 4 you for some reason now have a D Major chord.

In bar 5 You have what looks like another D chord but then you for some reason jump from a non-chordal note to a chordal note, something which makes the choice of chords unclear.

Only you and God himself really knows what chord you are in, in bar 6 you jump from a D sharp to a F and also from a C to a E. I really have no idea what is happening there.

Lastly, in bar seven you at least end on something resembling a tonic.


  1. Provide a proper harmonic basis for your melody before you right anything. For a newbie just choose your basic 1/4/5 chords, you don't need to reinvent the wheel, just choose a solid base.
  2. Jump less and only to and from chordal notes. You jump way too often and much too large intervals. Also, try and have counter movements when you jump.
  3. Do something about the rhythm. The rhythm does not evolve in any meaningful way, it just drones on, in the same manner, all the way trough. There is no sequence or any theme that is built on, it is just the same rhythm all the way.
  • The D# perhaps should be Eb, to fit with C minor. 'Lastly in bar seven' - maybe bar eight?
    – Tim
    Oct 23, 2016 at 14:31
  • To be honest, while I know some musical theory, I do not write music. That much is blatantly obvious. The melody has just evolved over time as I pick at a keyboard. The same for the rhythm. I think I understand your comments and I'll definitely look into them. For reference, here is what I somewhat hear in my head: Oct 23, 2016 at 16:36

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.