In his fabulous book on Music in Disney's Animated Features (2017), James Bohn explains what makes Dumbo's psychedelic dream so musically unsettling. After describing the melody's rhythmic instability, Bohn turns his attention to the harmony:
"The most emphasized tone in the melody is actually scale degree six, not the tonic. There are tonal, agogic, and dynamic accents on both the high dominant as well as the leading tone to the dominant (sharp scale degree four). While they are displaced by an octave, together these three prominent melodic tones form a chromatic cluster. It is only the last note of the melody that firmly establishes the tonic." (page 98)
Bohn assumes, quite correctly, that this passage is in the key of A minor. In fact, when I googled the sheet music, every single edition that I found harmonized the melody in A minor, as shown in the sheet music below. (I've added scale degree numbers above the melody so you can clearly see Bohn's "three prominent melodic tones" - #4, 5, and 6)
But here's the thing. If you ignore the accompaniment, and just consider the melody by itself... doesn't it seem like it could be in E phrygian, rather than A minor?
Phrygian is like minor, but with one crucial difference: scale degree 2 is lowered, creating an eerie half-step between scale degrees 1 and 2 and turning the dominant V chord into some sort of mutilated diminished chord. For these reasons, the phrygian mode is often associated with creepiness and evil. (It's often used in the score for Nightmare before Christmas, for example.)
So if we go back to "Elephants on Parade," which is DEFINITELY creepy and evil, it's so easy to see this melody as being in E phrygian. Measures 1-3 emphasize the half-step between scale degrees 1 and 2, which is precisely what makes phrygian unique. Measures 4-5 play with the tension between the tonic (1) and the raised leading tone (#7). Measure 6 repeats the tonic 3 times before cascading down an arpeggio to scale degree 2, which leads straight back into a repeat of the opening alternation of 1 and 2.
The only moment that DOESN'T seem like E phrygian is the final measure, which is clearly A minor. But even Bohn agrees that the only clear moment of A minor in this tune is the very last measure.
SO, I thought, what would this sound like if I harmonized it in E phrygian, rather than A minor?
BEHOLD, friends! Here it is, my re-harmonization of "Pink Elephants on Parade" in E phrygian. Again, I've added the scale degrees above each note of the melody. Be sure, as well, to listen to the audio file under the score:
So, what do you think?
Does it sound better to your ears in A minor or E phrygian? Let me know, either way! And be sure to share this post with your friends, so we can all nerd out together. :-)
What makes the underwater background music in Little Mermaid sound so magical? It's a combination of many factors, including the harmony, melody, rhythm, articulation, instrumentation, and formal structure.
One of these factors is its use of the Lydian mode, which is often associated with wonder, magic, and dreams. Part of Lydian's charm comes from the fact that, unlike major and minor, it has major chords on both I and II. It also has a tritone between 1 and 4 which has often been used to great effect (think of the Simpsons theme or "Maria" from West Side Story).
But in the case of the Little Mermaid, things are a little more complicated. The ostinato shown above can be thought of as I-II7 in Bb Lydian, but it can also be thought of as IV-V7 in F major. If you think of it as F major, then it's as if the music is wavering around the dominant without resolving. In the song "Part of Your World," this makes sense as she's spending the first part of the song talking about how unsatisfied she is. But then, when she finally puts a name to her dream – "I wanna be where the people are" – that wavering IV-V7 finally resolves to I in F major. Ah, resolution...
So the use of Lydian here is a doubly whammy. On one hand, it already comes loaded with connotations of wonder and dreams. And on the other, it serves as a dominant prolongation of the relative major, refusing to resolve until Ariel finally puts a name to her dreams.
Music theory is not about rules! It's about conventions!
And sometimes, those conventions aren't the best way to do things.
Take the opening of "Do You Want to Build a Snowman" from Frozen. The "correct" notation in 4/4, shown above, completely blurs the meter, the counterpoint, the rhythm, and even the genre. What's more, it's hard to play! (Catch that left hand Eb on the last 16th note of beat 1!)
But when we re-beam it to fit the three unequal beats of 8/8 rather than the more conventional 4 equal beats of 4/4, a whole galaxy of details springs to life.
Why does any of this matter? Well, this passage is not just dramatic but also a huge part of both setting up the film's narrative and establishing Anna's personality.
This song comes after that heart-wrenching scene where the troll king erases Anna's memory, to spare her the trauma of her near-death experience. As the scene comes to an end, a confused Anna watches as her sister completely shuns her by locking herself up in her room. The musical background fades into a soft, slow, descending melody, orchestrated very sparsely, a perfect depiction of the loss, abandonment, confusion, loneliness, etc. felt in this scene by both sisters.
And this lonely music moves immediately into a fast, upbeat tango as a now-older Anna races to her sister's door to invite her to play together. What a dramatic contrast! It highlights how playful, giddy, and carefree Anna has become, and makes the tragedy of her memory loss and abandonment all the more poignant.
Sure, you don't need to know any theory to feel this emotional contrast between one scene and the next. But music theory -- including a sensible, if unconventional, notation -- helps us understand that contrast on a much more nuanced level, which means we can also feel it in a more nuanced way. And it also makes it easier to perform!
Sam Zerin is a PhD student in musicology at New York University and a former lecturer in music theory at NYU, Brown University, and the Borough of Manhattan Community College. He also runs Social Media Music Theory (@SocialMediaMus1)