Can someone make sense of this piece? It's written as if to purposefully discourage performers:

  1. The piece is in 3/8 when it should've been in 4/8; every measure except maybe three* is a 4-tuplet and every note lines up to a 4/8 beat. Keep in mind this is over 30* pages.

  2. The piece is in 3 staves and the staff of each measure appears randomly scrambled. Given "L" for the left hand and "R" for the right hand sometimes the staves are arranged (top to bottom): LLR, RLL, LRL, RRL, LRR, RLR, often front-to-back. And when hands change staves within the measure, the arrangement isn't always optimal.

  3. There are repeat symbols between the measures (which by the way are much more visible than the actual measure bars)! Their great frequency makes absolute time-counting difficult (1-and-2-and-3-and-4-and, 1-and-2-and-3-and-4-and, ...) requiring instead relative time-counting based on the duration of the note (1-2-3, 1-2, 1, 1-2-3, ...), which I was taught to avoid. Instead, the repeated measures could have been expanded. Imagine counting a piece in which random but frequent measures have pickups of varying duration.

I have the universal edition. I've heavily annotated it, but it was time consuming.


I seem to have misplaced the pages after 30 (which ends on a double barline + repeat). There are actually 49 pages (performance duration: ~2 hours) and nearly all the 3/8 measures which can be in 4/8 are like this:

a 3/8 measure that could be in 4/8

All the 3/8 measures which can't be in 4/8 look like this. There is only one staff and one voice, ie easy tuplet material:

a 3/8 measure that can't be in 4/8 another 3/8 measure that can't be in 4/8

While there are more than three, the list is short considering it's across 49 pages:

  • page 16: 1
  • page 17: 3
  • page 18: 1
  • page 20: 1
  • page 21: 2
  • page 22: 3
  • page 23: 7
  • page 24: 1
  • page 25: 2
  • page 30: 2
  • page 31: 7
  • page 32: 4

total: 34

  • Compared with much contemporary keyboard music, there's nothing at all problematical to read here. – user19146 Aug 21 '17 at 4:27
  • I've a piece by Iannis Xenakis and I agree that Feldman's music lends itself to standard notation. From what I've heard, the universal edition "transcribes" Feldman's custom notation to one more traditional. Yet if this is true, why is the score unnecessarily obfuscated. – user19087 Aug 21 '17 at 5:29
  • (From a performance perspective. Are there theoretical underpinnings or was the guy just being difficult?) – user19087 Aug 21 '17 at 5:36
  • 1
    Sorry, but your complaints are mostly factually incorrect. There are a lot more than "maybe three" bars not in 3/8 time, and composers have been writing "repeat symbols between the measures" for hundreds of years. (There are plenty in JS Bach - and you can even find them in hymn books, which are not exactly full of avant garde notation). You are entitled to your opinions, but you don't give much evidence to support them. Some contemporaries of Feldman wrote 300-page scores for solo keyboard (single pieces lasting several hours) without any time signatures or bar lines in the entire score! – user19146 Aug 21 '17 at 23:59
  • 1
    Feldman's rhythm here is actually very simple. It's just 8 pulses (16th notes) in one hand against 6 in the other, with the 6 divided into 3 pairs, and the 8 often divided into some permutation of 3+3+2. Unfortunately, conventional notation can't handle that in a straightforward way - hence the overdose of tuplets. – user19146 Aug 22 '17 at 0:08