Syncopation or Rubato?

‹-- PreviousNext --›

carolekingI have commented before in passing on the way that popular melodies from the 1960s onwards look - on paper – more rhythmically complex than tunes from the earlier parts of the 20th century. They appear to have a lot more syncopation, and they seem rarely to cadence exactly on the beat – instead either anticipating or delaying over the barline.

The reason for this is not, however, because popular music has particularly become more complex over time (you get examples of both simple and complex melodies throughout the 20th century), but because the processes of musical production have changed.

Before the 1960s (ish), the primary text – the thing that defined ‘the song’ – was the sheet music. Composers and lyricists would collaborate to produce something that was written down, and would be sold in the form of the written text. Performers would record the songs too, of course, and the records would be sold as a separate product, but you’d be aware you were buying a performance, a rendition, rather than a ‘song’.

Well into the 1960s, indeed, you could find multiple versions of the same song performed by different performers in the charts at the same time.

But somewhere along the line, you started to get instances where the primary text was the recording rather than the score. Popular musicians started to learn repertoire by listening to recordings, and started composing directly into performance. This was particularly the case with groups that performed primarily their own material of course: for the Beatles and the Beach Boys, it is the recording that is the definitive version of the song.

So, any sheet music you see of this kind of material is actually a transcription after the fact. And this mode of production also slops across into instances where the songwriter might indeed have actually written the tune down in the first place. If it eventually gets published in written form (usually only in the wake of a top-selling recording), the tune you see won’t be what the song-writer wrote, but a transcription of what the singer who popularised it sang.

A nice example of this is ‘You Got a Friend’, written by Carole King. I’ve seen two published versions of this, one branded as sung by her, one as sung by James Taylor, and whilst the pitch content is largely the same across the two, the rhythms are quite different.

So, what’s going on when you see all this apparent rhythmic complexity in the sheet music is that the person doing the transcription is writing down all the little pushes and pulls that the performer introduces into their delivery as vocal styling, and turning it into part of the ‘text’. Imagine that the only sheet music for ‘Summertime’ in existence was a transcription of Sarah Vaughan’s performance – we’d find that the music of Gershwin was just as ‘complex’ in this case.

Now the relevance of this for the arranger is how we therefore treat these melodies. We are generally brought up with a respect for the composer’s text, and a sense that when we arrange a tune, we shouldn’t be messing with it too much. But it’s clear from the production process of commercial sheet music that what we have here is already a version of the composer’s text, filtered through the mannerisms and expressive habits of a particular performer. And what looks like a load of push-beats and rhythmic cleverness is actually just the reified traces of a performer’s assertion of musical freedom.

Now, we may wish to retain some of these, since the chances are both singers and audiences know and love the recording the music was transcribed from. But it is just plain mean to try and make the people who are going to sing your arrangement read all of those rhythms. There is also a kind of irony in turning one performer’s expression of musical ownership into a literal text which will box subsequent performers in.

Better, in my view, to straighten many of these rhythms out. The task is to reverse-engineer what the song-writer probably would have written from the performer’s interpretation of it. And then arrange it in a way that gives the new singers a bit of space to put their own musical imagination to work.

Hi Liz,
Although I'm not an arranger, this post has really set me thinking.

I remember a very interesting example at a rehearsal some years ago. We had a coach in for the day and we were practising "Everythings Coming Up Roses". The problem was that the coach kept getting us to sing a particular section of the song in the way she remembers it being sung by a particular artist, whereas the MD had a different artist in mind, who sang the same section differently. They couldn't agree and had to agree to disagree. Not surprisingly the version that has stayed with the chorus was the MDs.

If the manuscript had been written specifically either way, it might have stopped the debate, but it was interesting for the chorus to sing it both ways.

Hi Donna,
Nice example, thanks. I should imagine trying to write it down as Ethel Merman performed it would be quite a task!

And, yes, ultimately the MD will always prevail.

Archive by date

Syndicate content