Music Composers Unite!
So here I am pondering on a piano concerto. It's high time I started working on one, especially since chances of a commission and performance are plenty.
And here I am running into trouble, before even launching on my first bar:
My melodic, form, architectural, harmonic ideas all stem from irregular beats/pulses... (think of Messiaen a little). With no time signatures on the bars, since every bar has a different, complicated time signature. For solo piano this is doable (albeit difficult to perform, but it's a concerto, so I won't hold back... :D), because the beams help give a clear idea on what's going on. I've attached a couple of examples from solo piano, and piano + vln + vlc, which are very clear on what's going on.
The problem, though, is that for an orchestral setting this seems to be impossible to notate: If we assume that the piano plays continuously for 20 bars... And it keeps changing pulse every beat or so, so it's completely irregular and seemingly random. The flute above the piano (for example), has a rest for the first 15 bars. Normally the part would include a multi-measure rest of 15 bars and then show the bar with the music, or perhaps with some cue notes somewhere... But in this case, with every bar having a different time signature, which in fact is hidden, cause it's idiotic to have a time signature shown as 17/16, the poor flautist will have an impossible task: To follow what the heck is going on and actually play his cue in the right time.
So... how can this be solved? Assuming that I won't change the music itself. Any ideas? Any experienced people with such an issue, cause I've never run into that before...
Your sketches are very interesting, Nikolas, this looks like it's going to be a real challenge for both the listeners and the performers, so good. There are different approaches you could take. One would be to simply adopt a basic meter of, say, 4/4 throughout, or at least for long stretches, and rely on beams and accents to clarify your intended phrasing. Another would be to score it as you've already done and instead of rests for parts that aren't playing, just leave those areas blank in the score. Instead of expecting the instrumentalists to count rests, place as many cues as possible in each part. And of course you would need to rely on a really good conductor to keep it all together. There are a great many scores already written that deal with this sort of problem, by e.g., Stockhausen, Penderecki, Berio, Carter, etc., so you'll probably be able to get some ideas from studying some of those. Good luck, this looks very promising.
This is a never ending struggle...
My experience so far:
- it is undoubtedly better to have the time signature showed in every measure than to have it change without notifying anyone (check Stravinsky)
- bars don't equal pulse, BUT...
There's quite a few buts, actually. Technically you can simplify the signature without messing up the performer's interpretation, there's plenty of ways to help that (beams across bars is the most obvious one). However, if it's used excessively, it will cause A LOT of confusion. A rather extreme example is something I tried like a year ago - I basically forced a very messed up, Strav-like string quartet piece into 4/4. It was incredibly easy to read - playing, on the other hand, was a real pain simply because what sounded like the measure's downbeat didn't look like one. I suppose you could experiment a little, taking extra care as to not make your performers' heads explode. If that doesn't work, though, you can always stick to one signature per measure method. I daresay 13/8 isn't so much of a problem when there's a conscious conductor to help :)
Other alternatives? I can think of bar-free, piano centered notation with other players' entrances (measured or not) controlled exclusively by the conductor. There should be some other possilibities, too - hard to judge without seeing the music first, I suppose.
Victor and Greg, Thank you both...
I do find that making the time signature 4/4 throughout would be a disaster in this case, since the music would simply bypass any indication of 4/4 in the score... so I think I'm forced to use time signatures, albeit outside the score, in large numbers, in order to avoid confusion (think of Stockhausen, etc), and also do NOT use multi-measure rests in the parts, while using cues extensively...
Since I am to do a 2 piano version of the concerto, I think I'll leave that in a different state! ;)
Again thanks for all your help and ideas and feedback. Much obliged!
That makes a lot of sense, Nikolas. Most conductors today are used to seeing continual meter changes, ever since the Rite of Spring which has now become a repertory standard in many orchestras. And most instrumentalists would rather see lots of cues than have to count measures anyhow.