It’s a kind of “chicken or the egg” question, I suppose: Which comes first, the lyrics or the music? True — they could both come at the same time, I suppose, but I have never been so lucky.
For me, which comes first seems to be mostly a matter of fortune or happenstance or perhaps my emotional need at the moment. To my knowledge, I have never consciously set out to follow either path.
The beginning of a song’s life just kind of happens. Something clicks and two phrases or images cohere to express something I am struggling to understand and articulate, and we’re off. Or something clicks and two guitar chords I am toying around with suddenly resonate in a way I haven’t heard before and please me sonically and aesthetically, and again, we’re off.
Strings of words or guitar sounds follow, each adding piecemeal onto that initial kernel of meaning or sound. Eventually the expected structures of the pop song begin to assert themselves over the evolving draft and the series of words or sounds start organizing themselves toward what we know as verses, chorus, and bridge.
Sometimes this all goes very quickly, and sometimes not. And sometimes the words and the music go through all sorts of revisions before settling into some more stable form. Performance, though, endlessly modifies any song that is still alive, so the writing is never truly finished.
For some artists, however, which comes first is a very conscious, deliberate choice. I once saw Paul Simon interviewed on 60 Minutes when Graceland had just come out, and he said that his process is to have the music for a song almost completely finished, take a recording of that music to his beach house in the Hamptons, and then play the song over and over again -- while bouncing a red rubber ball against a wall in his house -- until the lyrics present themselves.
In How Music Works, Davie Byrne talks about this process of music-then-words in very detailed terms. When writing the lyrics to go with the already completed musical tracks for “Once in a Lifetime,” he says, “I tried not to censor the potential lyrics I wrote down. Sometimes I would sing the melodic fragments over and over, trying random phrases, and I could tell when one syllable was more appropriate than another . . . I felt I had to adhere to whatever to whatever syllables seem to fit the existing melody best, so I would listen to the gibberish lyrics respectfully and let those vibe my guide.”
For me, when the music comes first, as I mentioned, it comes as a set of chord changes that eventually wend their way toward the formal structures of a pop song. The next cycle of composing involves finding a melody that seems to fit those chord changes but that also responds to the expectations of verses, choruses, and bridge (such as having a simple and strong melodic line in the chorus to provide the singalong hook, etc.) That melody is worked out by playing the chord changes over and over, singing la-las and doot-doos over those looping chords, sounds that are intentionally devoid of any lyrical meaning, until the chords and melody seem right for each other.
And by the time the chords, melody, and song structures are coming together, this amalgam of sounds has started to suggest something about what the emotional content of the song might be, about where the lyrical content might want to go.
Fitting the lyrics to the music is a complicated affair, one best left to a future blog post. In the video below, though, you can see and hear an example of these first two steps in my version of “music first”— chord changes, then nonsense melody — as they have emerged for a song I am currently working on.