Can the melody determine the key of an ambiguous chord progression

Question

I am writing a song in A major. The chords in the verse are A, C#m, F#m, F#m. Now it is clear to me the key of this part of the song is A/F#m but then in the chorus the C#m chord is switched with the F#m chord making the progression go A, F#m, C#m, C#m. The song also ends like this (on this C#m chord) and to my ears it seems like the song could end on this chord and sound like it has resolved here but what just happened? The song is supposed to be in A/F#m.

When I look to the melody, I am not using the D or D# note meaning the melody doesn't use any of the notes that could determine the key of the song but the melody of the vocal part does end on the C# note on the C#chord so this is where I ask myself. Does the melody in this case, and the fact that it ends on a C# note define the key of the chorus as being C#m instead of F#m in an otherwise ambiguous setting?

Adding the music notation to the chorus of the song to give everyone more insight. Does the melody ending on the C# mean the tonality of that part of the song is in C#m?

Answer 1

We should be careful to distinguish between tonality, key, tonic, key signature, etc.

I like Tymoczko's definition of tonality which is something like: a system of consistent harmony using a limited collection of tones with one central goal tone (my summary.) A major and C# Aeolian are tonalities sharing the same collection of tones, but with different goal tones (tonics) and using different harmonic systems. A major is labelled a key, while C# Aeolian is labeled a mode. Both share the same key signature in staff notation.

Dominant harmony is what really defines a key. It's an aspect of the harmonic consistency of the major/minor system of keys. The dominant of A major is the E major chord or the full E7 dominant seventh chord.

These chords...

A, C#m, F#m, F#m or A, F#m, C#m, C#m

...don't use the dominant so a key isn't defined.

If you consider A to be the tonic, then you would call C#m (iii chord) and F#m (vi chord) the modal chords. Tones used...

     vi    I
     /|\  /|\
    / | \/ | \
   /  | /\ |  \
^6   ^1   ^3   ^5   ^7
             \  |  /
              \ | /
               \|/
               iii

This leaves out ^2 and ^4 so you can't have the ii, IV, V, or viio chords. In terms of functional harmony this excludes the most important subdominants and dominants. Instead of getting a key this will probably create a modal harmony feeling.

Compare that to the intersection of the primary tonal chords I, IV, V...

       V
      /|\
     / |  I
    /  | /|\
   /   |/ | \
  /    |  |  \
 /   / |  |   \
^7 ^1 ^2 ^3 ^4 ^5 ^6
     \      |     /
      \     |    /
       \    |   /
        \   |  /
         \  | / 
          \ |/
           IV

...those three chords give the full diatonic set of tones.

When I look to the melody, I am not using the D or D# note meaning the melody doesn't use any of the notes that could determine the key of the song...

D is the critical tone, but it alone won't determine the key. Rather it would clarify the key signature. Depending on D or D# the key signatures could be A, E, F#m, or C#m. None of those key signatures will define a key without the presence of the appropriate dominant harmony.

For a key of A major we really need the leading tone ^7 - which is present - paired with tone(s) unambiguously not of the iii chord. ^5 and ^7 will give you a partial dominant chord. But, given this three chord palette it will probably just sound like an incomplete iii chord. Either tone ^2 or ^4 with the leading tone would make an unambiguous dominant.

...but the melody of the vocal part does end on the C# note on the C#chord...ends on a C# note define the key of the chorus as being C#m instead of F#m

Not without a dominant.

If you had the D#, you might call this in C# Aeolian.

Defining a key is a matter of harmony even if it is implied harmony from a melodic line. If the harmony is ambiguous, I don't think anything about the melody will change that. Really, melody is part of the harmony it isn't separate. If the melody somehow clarified the harmony, then there isn't actually any ambiguity.

If you are wondering about a key signature for notation, just use three sharps.

Answer 2

The melody could certainly help define this ambiguous question. ;) But I'll first answer the subject line.

Q: Can a melody determine the key of an ambiguous chord progression

A: A melody can, if not determine, at least very strongly suggest an entire chord progression. And a key. However, the key is ultimately in the ear of the listener.

Q: Does the melody in this case, and the fact that it ends on a C# note define the key of the chorus as being C#m instead of F#m in an otherwise ambiguous setting?

A: It can do that. However, you didn't provide the melody, so we cannot tell our opinions. You said that it feels to you like C# is the tonal center, and you're the composer, so it would be entirely OK for you to declare that the end is in the key of C# minor.

Ask yourself, what would happen if you declared C#m as the key for the part. If someone improvised a solo with D# notes in it, thinking that that's what the composer intended since he said the part is in C# minor, would that be OK for you? Do you care? For simply reading the music it shouldn't make a big difference whether you write it with three or four sharps, since you don't use D or D# notes, and it starts with three sharps and you said it's a bit ambiguous anyway. A key signature change might be just confusing if the change isn't very significant, and if the changed notes aren't even used.

Would it be socially or professionally or otherwise unacceptable not to know the key for sure? How about, do it with different feels each time? If you feel you're in a mood for four sharps, go for it.

Edit: now having a piece of the melody, I think it's just ambiguous. I tried playing the two-bar melody with chords, and whether the C#m felt like home depended on how much I had hope for getting a D chord or a B chord. After playing a D major once, the probability for A or F#m being the tonic lasted for a few repetitions and gradually faded away. After playing a B major once, it was pretty clear that it's in C#m - until playing a D again. I was also able to just make myself wish for a D or a B chord and thus changing the perception of what's going on. Sort of like imagining another rhythm over a simple beat to make it sound more interesting. The melody and the chords can lend themselves both ways.

Here are a few different chord progressions for the melody, starting with the original. https://vocaroo.com/i/s0MaGpJVL2Xp If you restart the clip after hearing one of the others, the original chords and melody get a different interpretation depending on what you heard last.

So: a strong melody can help in suggesting a key, but this particular one doesn't feel very strong in that regard.

Answer 3

I agree with @piiperi, the melody can definitely help a lot in determining or strongly suggesting the key. I also like their suggestion to narrow things down by thinking about whether it would make sense for someone else to improvise using either a D or D#.

That said, to me the harmonic rhythm plays a large part in determining the key even without the melody. Meaning, it matters which chord comes first, and which chords you keep coming back to. Do your verse chords and chorus chords cycle at all? If your verse is A, C#m, F#m, F#m and your chorus is A, F#m, C#m, C#m, and both sequences cycle, what stands out the most to me is that they're both going to keep coming back to A at the beginning of each cycle. So it feels like both sections are in A/F#m even though the chorus ends on C#m. At least, that's how I'd hear these progressions in my head without any further context. But it's also entirely possible to write the chorus in a way that establishes C#m at the end.

Also, it's totally OK for a song not to resolve on the one. So it's quite alright to consider the key as being in A/F# and still end on C# in both the melody and harmony. I'd probably write the key signature in A unless there's a strong reason to imply the D# for improvisation or chord voicings.

But also, you're the composer, you get to do what you want! Sometimes it's less about what's "correct" and more about what you'd most like the reader to infer from what you've written.