Wednesday, August 17, 2016

Alan Walker - Sing Me To Sleep (Analysis)

After breaking the song down into its form, I realized that the song is both simple and complex. Before I explain this simple but complex idea, let me go through my analysis.

FORM & INSTRUMENTATION - The Breadth and Depth

I have named the arrangement as such: Intro - Verse 1 - Verse 2 - Pre-Chorus - Chorus - Verse 3 - Pre-Chorus - Chorus - Bridge - Outro. It is key to note that:
  1. The song is in 4 4 time
  2. There is no real introduction - the song jumps into the first verse after an almost 1 bar ascending sound effect. The ending is similarly abrupt with just 1 bar of descending sound effect.
  3. Every pattern (A/B/C/D) is 4 bars long
  4. Every main section of the song is 8 bars long, formed by repeating 1 pattern twice
    1. i.e. Verse 1 = A + A, Verse 2 = B + B etc.
  5. Verse 1's pattern A | A is repeated in Verse 3
  6. However, Verse 2's pattern is not repeated (unique)
  7. Pre-Chorus and Chorus ( C | C | D | D ) appears twice in the song as typical of modern pop music
  8. While I call the second last section the 'Bridge', it is really a repeat of the Chorus ( D | D pattern) with a variation of Verse 2's lyrics/melody layered on top of it
As we can see, the simplicity of Sing Me To Sleep lies in points 3 and 4. The patterns are no longer than 4 bars long, plus every section is simply exactly 2 times of a selected pattern. On the first listen, there are clearly two parts (Intro-Chorus + Verse 3-Outro), with the latter seeming like a repeat of the former.

However, it is complex in that instead of having only 2 patterns for just verses and choruses (A/B), I managed to identify what seems like 4 patterns. Pattern B is a standalone and appears by itself only once, although its elements are borrowed in the Bridge, forming Pattern D.

The complexity of Sing Me To Sleep is even more evident as I go on to talk about the instrumentation (depth complexity). I think I have a headache from trying to hear all the different sounds over 2 days, especially since I am not familiar with identifying instruments and the effects in electronic music masks the true sound of instruments. The ones I have most problems with was the Synth/FX versus the muted Guitars, Snare Drum versus Percussion, so pardon me if I mixed their roles up, or heard something that wasn't actually there!
Some explanation of terms:
  • The Pluck is what Walker described as a ''plucky melodic, almost percussion-ish sound'' created with ''chopped off vocals''. It occurs at the beginning of each A-phase, coming in with and at the same pitch as the first note of the piano. It's most obviously heard in Verse 1. It may have occurred throughout the rest of the song but it was too difficult to catch.
  • The Soul Lead is the bouncey sound (*) that comes between "Sing * me * to sleep * now * Sing * me * to sleep * ...".
  • The Percussion in the Verses and Pre-Choruses were clicking-like sounds (like going tick-tock-tick-tock) right in the background, playing the rhythm a fraction of a second after the synth/guitars 
  • A new percussion seems to be added in the bridge, sounds like cymbals (hear it between the voice "A-a-*cymbals*-nytime")

I love the choice of instruments (the flute!), the Soul Lead, and the melody. It gives the electronic music a more pleasing and melodic feel.

Something about 'the piano' I identified
  • The piano is clearly heard at the start (playing a tune) and during the pre-chorus (playing the chords).
  • In the remaining sections where I marked the piano as "Pad/Fills, I'm referring to the very low/bass sound playing the chords... It is possibly another instrument but I've identified it as the piano in here.
What else I like about the song is that it starts really 'calmly' and the build up was done well. As mentioned, each section repeats 1 pattern. In order to keep the song interesting, the build doesn't come only in between sections (i.e. adding instruments when going from Verse 1 to 2), but during sections. One example is in Verse 2, where the bass drums are added only in the second B phase.

At first glance, some might argue the section I named 'Pre-Chorus' should be the chorus since it has the most instruments thus enabling the fullest sound. From this, I learnt that that's not always the case and that the choice of instruments matters. The change in lead from a 'softer' voice to the synth causes the sound to be much fuller in what I call the 'Chorus'. Also, as an electronic music piece, I find it suitable to refer to the instrumental section as the chorus, because voice should not be focal in electronic music.


MELODIC CONSTRUCTION


The Call and Response Technique
One call and response I identified was the flute and guitars (or is it synth) 'answering' to the voice:

Flutes 1 (high): E - F# - G - F# - E - D - C
Guitars 1         : E - E - B - E - F# - G - C

Flutes 2 (low) : C - B - C - B - C - B 
Guitars 2         : E - E - F# - E - F# - G
- What's interesting here is that while the flutes low melody seems to complete the idea, the layer of guitar melody seems to suggest a continuation (think what's this increasing pitch of notes leading to ?)

Guitars 3        : G - G - G - G - F# - D

In Verse 2:
Anywhere, anytime
I would do anything (flutes/guitars) for you
Anything for you (flutes/guitars)
Yesterday got away
Melodies stuck inside (flutes/guitars) your head
A song in every breath
And Verse 3:
Remember me now, time cannot erase (flutes/guitars)
I can hear your whispers in my mind (flutes/guitars)
I've become what you cannot embrace (flutes/guitars)
Our memory will be my lullaby (guitars)

Another call and response, not too sure if this can be considered one, is simply the Soul Lead coming in between 'sing me to sleep', thus 'answering' to each word sung. Hmm... Well even if it isn't, at least it's the hook of the song!


The Hook
"Sing me to sleep" sung with the soul lead in between each word. The soul lead is a signature sound in Walker's music - also heard in Faded and Spectre, and here it is aptly incorporated into the hook.


Other Notable Points
At the end of the Verse 2 ("A song in every breath"), all instruments go out except for the voice: this is a a good way to signal the end of a section, and also create tension/interest before jumping into the pre-chorus build up?


HARMONY


Finally, the harmony (no guarantees, I tried my best...). The song starts in a minor tone (E minor?) and transitions into its relative major sound in the Pre-Chorus, Chorus and Bridge (G major). 

Here is a simple breakdown of the chord progressions (recall: each section is 8 bars long):

Notation: numbers in each cell indicate how many bars each chord lasts


With that... I've come to the end of my analysis! Phew! *wipes sweat off everything* The question now is, how do I apply these elements?! The part I am lacking is actually coming up with a tune... :( wish me luck!


References:
1. Original Song: https://www.youtube.com/watch?v=2i2khp_npdE
2. Instrumental: https://www.youtube.com/watch?v=x6hWKp95Kp8
3. Lyrics: http://www.azlyrics.com/lyrics/alanwalker/singmetosleep.html
4. Behind-the-scenes: https://www.youtube.com/watch?v=VLoqggFdVU8 (really helped in identifying unique sounds)

1 comment:

Unknown said...

Your analysis is very detailed, but to be honest, it's more than the song deserves. A song form of A B C D isn't really complex, it's just another example of the song factory approach. Each section was probably written by a different person. There's not a lot of relationship between each section. Complex is developing an idea, something we call "economy of material". Consider Beethoven's 5th symphony, based entirely on one 4 note idea. Of course, pop music isn't Beethoven, but still the best pop songs have usually tried to develop an idea, rather than just cobble together a bunch of them. Sorry for the rant.

You're right, there's a lot of production in this. Rather than worry too much about what each layer is, just understand that you're going to keep adding layers until you can't anymore.