Writing A Song From Scratch: Part Two

So we return to writing a song from scratch!

And before we get any further, apologies for the terrible wind noise on the video. I was seduced by the architecture and the grandeur of the opening to the Suez Canal, and forgot to take into consideration the influence that a fresh sea breeze would have on our plucky little mobile phone mic. 

In the first installment we began writing a song based on a randomly selected chord progression: III > III > II > VI in 3/4 time. In the last video, I experimented with a variety of approaches to this, trying out some different tempos and keys. The last of these was in DADGAD tuning, and as I’ve been writing in the this tuning a lot recently, I opted to continue the experiment. Or perhaps it would be more truthful to say I couldn’t be bothered to retune my guitar . . .

The songwriting process for the next stage could take a number of different routes;

1. Using the chord progression we have, we could begin exploring some melody ideas for the vocal part. The obvious way to do that would be to write some lyrics to fit the chords. An alternative would be singing a melody using nonsense sounds, allowing the writer to find a strong melody before forcing it to conform to a set of lyrics.
2. As a fingerstyle guitarist, a different track could be to start developing the arrangement, moving it beyond a bland chug through the chord changes to something for intricate and interesting.
3. While there are many songs which bravely stay on a single chord progression, my own predilection (especially with a song which is mainly going to be played solo) is to expand the palette, and write some different progressions for other parts of the song; verse, chorus, bridge etc.

In my opinion, there’s no right or wrong direction here – though the route you choose at the beginning influences the finished product. However, I would encourage caution regarding option 2. In my own experience, I’ve noticed that songs which have no part to be sung often die premature deaths. On many occasions I’ve put a lot of time into developing complex fingerstyle arrangements only to grow bored and forget about them. When there’s something to sing, even if it’s only to the simplest chord progression, the song feels much more like it’s coming to life, motivating you to keep working on it.

For this post I choose the third option, to go ‘widescreen’ and expand the chord sequence to what might be the progression for the entire song. The initial randomly selected chord sequence is composed entirely of minor chords. This causes it to feel claustrophobic, suggesting it would make a good sequence for the verse of the song, simply because in the chorus we can offer something which opens the shutters and shines light into the stuffy place we’ve made. The easiest way to do this is to return to the I chord (in this key Bb).

When creating new sections to songs, the obvious change is to the notes played. However, I heard some interesting advice recently, related from Brian Eno to Nick Mulvey. Eno suggested playing the expected chords, but at unexpected lengths. We can apply this idea to the chorus – begin the chorus on I, stay there long enough to invoke a change in the listeners mood, then continue the progression by running through the same chord sequence as verse but for different measures. We dash through the III and II chords with half a bar each, then linger longer on the VI.

For the sake of further variety, let us throw in a bridge as well. The chorus has expanded upon the world painted by the verse, so a countermove in the bridge could be to do the opposite – an extremely simple two chord sequence; two bars each. A movement like this constrains less when it comes to composing melodies above the chords – here I can take my vocal in many different directions, whereas during the chorus the more complex progression leads me down a much more defined path.

Throughout most of the song, I’ve been taking advantage of the ambiguity of the DADGAD tuning to refuse to commit to major or minor, especially on the VI chord, which should be a G minor. In the bridge I emphasized the flat third of the chord, bringing the VI back in to the minor fold with more conviction.

I should also point out that my references to particular chords here is misleading. I’m thinking in guitar terms in relation to the shapes I’m making, but of course the guitalele is five steps higher in pitch. So in reality, we’re in the key of D#, playing a Cminor chord, in the tuning of GDGFDG, which doesn’t quite trip off the tongue in the same way as DADGAD. 

Of course, it remains to be seen whether these chord progressions will support melodies worth getting excited about. This will be the next step, and might force us to backtrack, and make further alterations to the underlying music. Writing a song is rarely a linear exercise. However, in the next part of the series, we’ll hope to make a forward step and begin adding some lyrics – or at the very least some melody – to the song.