(Don't suppose you could explain that a bit?  Not sure I understand your argument.)

What I mean is that it's not just a matter of how clearly the score presrents itself, but also how clear the composer's own envisioning is. If his way of writing down the rhythms totally misrepresents them, that suggests that he isn't imagining that aspect in a very precise way. I certainly found that adopting strict proportional notation gave me a more exact understanding of my own use of rhythms and time.

Rather than "time" as a global or large-scale phenomenon, maybe this notation has more to do with local, gestural information (as has already been intimated)–with the time of two bars, for instance, rather than the time of an entire page.  By rejecting proportional spacing, Ferneyhough seems to be trying to force the player to consider agogic time–and, necessarily, take meter, which is hierarchical in a way that time-space notation alone isn't, more seriously at the level of the phrase/gesture.

I wouldn't necessarily think he would assume that the working-out would equate with a more thorough approach, necessarily, but the non-proportional notation does indeed require a working-out (to determine the relationship b/t local rhythm, meter, and pulse).  And, still playing devil's advocate, it would seem to me that in a proportional notation, it would be a lot easier to, from the very beginning and throughout the learning process, just get the gist of the rhythms (as in, "it's a bit after the third beat"), whereas once one has done the legwork/math (which, you're right, responsible performers do even with proportional notation, of course!), I would imagine they'd have a better understanding of the tuplet pulses and relationships.

As for the second half of your comment, what is it about proportional notation that is necessarily more 'precise' than non-proportional notation?

But the question still seems to be this:  how would the situation change, from a performer's perspective, if Ferneyhough (or I or qt or you, CD) put in those hash marks into the score (or even just the parts) in the first place? 

Just thinking out loud ... does this approach (which several of my colleagues use) not somehow codify/condone/institutionalize the sort of generalized "that attack comes slightly after the third beat" performance approach that, in essence, makes the score into some sort of graphical/proportaional/time-space notation in the end anyhow?



Btw, if he's reading here, James Weeks ought to chime in.  He taught me an exceptionally useful tool -- rather than simply putting in the typical metrical slash marks, he marks in his (conducting) scores the exact decimal point location of attacks.  So, for example, w/ some sort of nested tuplet, rather than merely seeing that an attack comes roughly after the third beat, he'll mark in his score that the attack comes at beat 3.15 (or whatever).  It seemed a very clever solution, to me.

More like three decades, I believe. Yes, of course I know that, and have read his deliberations on the subject. Maybe I should just ask him next year in Darmstadt rather than speculating on his intentions or virtues. Don't worry, I know he's quite a clever fellow...


Not the same objections, because there's obviously less mystery about where the 3rd beat of a 4/4 bar in a Beethoven sonata is located...but I don't much like it either.

OK: my thoughts on proportional notation:

FOR: It makes it significantly easier to get a rough idea of the rhythmic patterning, also to have some sort of overall guideline when performing. Now, in terms of what Aaron mentions about "that attack comes slightly after the third beat", I don't see that is always a problem. In highly pointillistic music, involving single pitches/sounds isolated from others, this approach might be problematic, but in many cases we are dealing not so much with attack as duration. Supposing you have a periodic tuplet group spread over a certain time-period, and it is notated so that you so one attack just after the third beat, one just before the fourth, one a little less than half way after the fourth and so on. One can simply place them in these rough positions, and adjust until they sound even. And in the majority of cases I've found that to be by far the best approach; having the beats facilitates the process (in most non-proportional music, including something like Ferneyhough Opus Contra Naturam, I do write in the beats in most of the score, usually with straightforward house and triangle symbols). Trying to conceive it in terms of isolated attacks: well, it takes an enormous amount of time and effort, and is anyone really sure that the products of the learning process in this respect really 'work' audibly at the time of performance anyhow?

AGAINST: The only real argument against proportional notation I can see is in some music where this will either lead to certain groups of notes being scrunched up together, and as such hard to read, or alternatively things have to be spread too widely over pages, causing more page-turns than is desirable.

Question for various composers and performers here: do you prefer to have proportional notation structured according to metrical units (e.g. quavers/eighth-notes, semiquavers/sixteenth-notes) then to be executed relative to the metronome mark, or to have the proportionality consistent in terms of absolute time? When there are many shifts between different metronome markings, I'd prefer the latter.

EDIT: Thinking about it, bearing in mind Colin Holter's comments about agogic approaches to time, in a more obviously gestural music, I can see why the metrical unit would be preferable to absolute time.

Well thank you...I couldn't really stay away from this subject now, could I?