NOTATIO

I am not happy with the slur in my entry above and continue to try to find a better shape, so far without success. OCTO's and Knut's seems best to me. I think that OCTO's second solution best conveys the physical action of playing such passage legato, but the bracket might be removed:
A little investigation has shown that the issue of where to put the number in such a passage is an interesting one and I may start another thread about it.

It just struck me that it is the unusual situation that creates the difficulty with the slur. Normally, the slur would go on to the next note, which might lead to a solution like:
It is difficult to do unmusical things in both playing and notating.

I think the usefulness of discussing this measure in isolation is limited because the best placement of both the slur and the number depends on the context and styles used elsewhere in the score. Nevertheless, I don't think placing the slur below is acceptable in this case, because it visually neglects the notes in the upper staff. The placement of the number on the beam side is fine according to literature, but I prefer to have it above the notes since the phrase ends in the upper register of the right hand. If the rest of the piece uses ordinary square tuplet brackets, there is no way to confuse the slur for a bracket.

John Ruggero wrote:It just struck me that it is the unusual situation that creates the difficulty with the slur. Normally, the slur would go on to the next note, which might lead to a solution like:

It is difficult to do unmusical things in both playing and notating.

If the chord immediately following the tuplet is included, I would draw the slur as in the first measure below. I'd place the number above, but have placed it on the beam side in the example for comparison.
Even if you use a wavy slur, there is no need for it to cut through the lower stems. The obstruction should be as minimal as possible, as in the second measure below.

Knut, why is placing the tuplet number under, as in my example with brackets, unacceptable? IF we assume that the tuplet is in the RH than it is perfectly in accordance with the rule.

OCTO wrote:Knut, why is placing the tuplet number under, as in my example with brackets, unacceptable? IF we assume that the tuplet is in the RH than it is perfectly in accordance with the rule.

My objection was to the slur, not the tuplet number or bracket. I would personally prefer having the number above for the reasons stated above, however.

Knut, I think I prefer my S-shaped solution for the new slur situation and accept OCTOs 2nd solution to the old one because of the clarity with which they convey the connecting of the main line G-A-B-Bb. The F#-C# fifth seems decorative to me, and, for this reason, it doesn't bother me to exclude it a little from the slur. Many keyboard players would actually hold the F-A third through the F#-C# a bit longer than a usual legato overlap in an unconscious attempt to connect the F-A to the G-B.

However, from the visual point of view, your two solutions are certainly superior and beautifully accomplished as always.

Your point about context is so true. All it would take to avoid any misunderstanding of your original solution would be a successive repetition or similar pattern with the same slurring but without a triplet marking.

Your observation about the positioning of the triplet number relative to the ending note is also an important one that I hope to explore in another thread.

John Ruggero wrote:Knut, I think I prefer my S-shaped solution to the new slur situation and accept OCTOs 2nd solution to the old one because of the clarity with which they convey the connecting of the main line G-A-B-Bb. The F#-C# fifth seems decorative to me, and, for this reason, it doesn't bother me to exclude it a little from the slur. Many keyboard players would actually hold the F-A third through the F#-C# a bit longer than a usual legato overlap in an unconscious attempt to connect the F-A to the G-B.

I checked with Gould, just to be sure, and indeed she confirms that placing the slur below in a case like this, however convenient, is not acceptable for the reason I've already mentioned.

While I do not share your preference for a wavy slur in this situation, this solution might be acceptable, provided it is drawn in the regular way, above the beam.

John Ruggero wrote:However, from the visual point of view, your two solutions are certainly superior and beautifully accomplished as always.

Thank you very much! While you may prefer different solutions, I think beauty and clarity go hand in hand in this situation, as in so many others.

Edit: I should also mention that I would regard the C#-F# as part of the line and not as a decorative addition to it, and I therefore see no reason not to include it in the slur.

I guess we can agree to differ on this one, Knut. For me, all three solutions are fine, but mean different things interpretively. I would never exclude any notation out-of-hand even if the venerable Gould disapproves.

I do want to clarify that by "decorative" I meant that the F#-C# fifth was not part of the LH lines but yet another set of two voices that are also collapsing on the final second from above. For that reason, I can imagine an interpretation that brings out the lower voices as the leading lines and the upper ones as a kind of accompaniment.

However, all of this is beside the point I was trying to make: that the puzzle is posing an unlikely scenario and that is creating notational difficulties. But I never should have chosen an unusual slurring to illustrate this point. Your first one makes it so much better, i.e. that it is so much easier to deal with this situation when the slur leads on to its natural destination.

I've changed my mind, Knut. Your first one is definitely best because the upper lines, even if considered independently, should be included in the legato, which was the point you were making. Sorry it took me a while. I hereby disavow the S-shaped slur.