Re: triplet question

[Edward Sanford Sutton, III]

On Tuesday May 25 2004 10:14, Thorkil Wolvendans wrote:
> Hi Ed & everyone!
>
> At 19:17 24-5-04, you wrote:
> >I haven't ever seen, or heard of, triplets going over a bar line (except
> > when I got Lilypond to do it on my school assignment by mistake).
>
> I attached an example from the book "Music Notation in the 20th Century" by
> Kurt Stone. It gives a nice example of a quintuplet (okay, it's not a
> triplet, but this was the first example I could find) crossing a bar line,
> taken from Elliot Carter's "Symphony of Three Orchestras".
> It was not an uncommon habit for the Post-World War 2 -generation of
> composers to use barline-crossing tuplets. Boulez, Berio, Xenakis, etc.
> used it quite frequently, actually. But I believe even Webern has used them
> in a couple of compositions.

Thank you for the example. Seeing that came as quite a suprise; it also seemed 
the author shows the tied note with the tuplet bracket like I thought it 
might be if it did occur, but suggests the notation that leads to incorrect 
counts. As for the other examples within music, do you have any specific 
pieces to suggest I look at?

> >Changing time signatures is probably what should be done if you are
> >writing music and
> >it has gotten to that point.
>
> Maybe, but... what if you want to start a triplet on the second half of the
> last beat and don't want to loose the 4/4 feel?

Isn't that exactly what a triplet there would be trying to do?

> For instance:
>
> %START EX. 1%
>      \notes \relative c'' {
>      \time 4/4
>      c4 e8 e g4 d | c4 e8 e g4. \times 2/3 { f8 a b} c4. r2
>      }
> %END EX. 1%
>
> One could change time here to \time 9/8, next bar to \time 7/8 and then
> return to \time 4/4. This certainly shouldn't give any (major) problem if
> it's a solo piece, but I think, if it would be a piece for let's say,
> violin & orchestra, the timeshifting (and most importantly: rythm-shifting)
> would be very confusing for the rest of the ensemble, because (considering
> the fact that we want to keep the 4/4 feel) they have to start on the
> weak-weak-beat of the 9/8-measure, which has to function as a preferenced
> Strong-beat and the actual Strong-beat (the first beat in the 7/8-measure)
> has to function as a Weak-beat again.
>
> I think it's easiest for all parties (soloist and/or ensemble) to use a
> barline-crossing tuplet.
> Only, how should one achieve it in Lily? I'd say by faking it, or not?

I think that would be best accomplished by manually splitting the note into 2 
smaller notes that tie over the bar as your example shows (but doesn't 
recommend). Otherwise just put it in as ant normal tuplet and it should come 
out as your example piece demonstrated. I do not know if Lilypond would give 
you any spacing problems or not. In any case I think this would be worth 
looking at in closer detail if it hadn't been already, but as I mentioned 
before, I haven't heard of good old hand engraved examples from well known 
and followed publishers.

> Regards,
> Thorkil

-- 
I didn't like the play, but I saw it under adverse conditions.  The
curtain was up.