Can't change direction of partial beam when beaming across a rest

• May 2, 2015 - 00:08

In the following excerpt from a late 19th-century score, the partial beam on the 16th-note G natural extends to the left of the stem, which makes sense because this note forms a rhythmic unit with the preceding 16th rest (together they make up the middle 8th of a triplet of 8th notes).


I can't seem to get this same result in MuseScore. The closest I can come is this:


I've looked everywhere I can think of to find a way to change the direction of the partial beam. Is there something I'm missing or is this currently not something the user has any control over?

Attachment Size
beam_orig.png 118.3 KB
beam_ms.png 16.12 KB


Direction of partial beams is not something that can be controlled currently. See also #58061: Beam direction of uneven triplets wrong. This is actually the same case: the sixteenth in question is actually exactly on the "&" of the beat and thus should normally point right by that rule. But this case, like the one in that issue, is indeed one in which any given editor might subjectively choose to do otherwise, and makes a good candidate for an override - or even a change in the default, if a truly definitive / authoritative source can be found to justify it.

In reply to by Marc Sabatella

I have to say I've never noticed any difference between the beaming of triplets and the beaming of triple meters, and I think there are good theoretical and practical reasons why there shouldn't be a difference. I'm not familiar with the beam direction rule you mention, but I don't think it should apply to triplets. The idea that a duple division of the beat should have any impact on the notation of a triplet seems peculiar to me, because the whole point of the triplet is to replace the default two-way division with a three-way one.

I'd be in favor of changing the default. Sadly I can't point you to a definitive source at the moment. At the very least, though, I think the user should be able to override MuseScore's default.

In reply to by ghicks

I do understand that line of reasoning. But it's also at odds with your first statement - that there are good reasons why there shouldn't be a difference between beaming rules for triplets versus non-tripelts. The rules as they exist are based (in part) on *position in time*, and that's precisely why the beaming is the way it is. What you are actually asking for is that triplets *are* treated differently - not based on actual position in time as normal notes are, but on some different sort of rule that doesn't apply to non-tuplets. Again, not saying this is unreasonable, but it does mean you are asking for tuplets to be more *different*, not more *similar*, to ordinary notes. And the trick is defining this new set of rules.

In reply to by Marc Sabatella

I'm not sure I understand what you mean when you talk about "position in time". Is this Gould's terminology? I haven't read her book and this may handicap me in discussion of concepts drawn from it.

To the extent that I do understand what you're saying (or perhaps just think I do), I suspect our disagreement has to do with the question of whether the beat and subdivisions thereof imposed by the time signature should have any effect on the notation of relationships within the triplet. You think they should; I don't. I think the whole point of a triplet is to replace the default duple division with a triple division, and that once this is done, it's the triple divison that should rule.

Take a standard siciliana rhythm as in the example below. The 16th note occurs on the & of 2, in a weak position, and the notation reflects this fact.


My contention would be that when the exact same rhythm occurs in a triplet, the 16th note is still in a weak position with regard to the triplet, and this should be reflected in both performance and notation. The fact that this note happens to occur in a relatively strong position with regard to the underlying duple meter may be of interest to the performer (especially if he or she needs to coordinate the triplet with duple rhythms occurring simultaneously in other voices), but it should affect only *when* the 16th note is played, not *how* it's played (e.g. with what degree of stress).

The current MuseScore notation, on the oher hand, implies a much more prominent role for that 16th note than it deserves. In fact, it makes it look like you haven't really divided the quarter note into three equal parts (i.e., replaced two eighth notes with three), but rather you've retained the duple division of the quarter note, but subdivided each eighth note in three:


I would guess conservatively that 99% of the time this is not what the composer intended.

Attachment Size
six-eight.png 16.22 KB
triplet.png 16.62 KB

In reply to by ghicks

As I said, I do understand this line of thinking, and when applied to simple triplets, it's not even hard to see how one could construct a rule one could implement. "Position in time" means exactly what it says - if something is on the "&" of 1, it is on the "&" of 1, period, regardless of whether it is a sixteenth or an eighth, the second note in the measure of the 17th, or whether or not it is part of a tuplet. Nowhere in Gould's description of the rules does she make any allowances for any differences here. Maybe she would have if had been thorough, who knows.

So again, alll I can say here is, I do understand why in the case of a simple triplet, it might seem that one might subjectively want to beam as you have shown. I'd still like to see an authoritative source give an actual rule to follow - one that would also have tuplets that are *not* just simple triplets.

And personally, as a player, I do still prefer the current beaming. I like to know what beat notes are on - this is far more important information as an aid to reading to me than subtleties in which note might or might not be stressed. But again, I *do* get why sometimes someone might subjectively want to see it differently.

As you can see in the issue report, it is being considered. It would still help if someone could find a definitive authoritative source to justify the change.

In reply to by Marc Sabatella

I'm guessing that Gould's text doesn't actually contain the phrase "whether or not it is part of a tuplet", because if it did then you would quote it and it would determine the matter (as long as Gould's word is to be considered final; I don't know anything about her so I don't know if that's reasonable). In the absence of that qualification, then what you take out of her position pretty much depends on what you bring to it. I for one would claim that in the context of the triplet there *is* no "&" of 1; it's been superseded by the triple division imposed by the triplet. (As I said, I can understand how finding the exact temporal midpoint of the triplet might be a convenience to the performer in determining how the notes line up - although I think most performers could work this out without the assistance of beam direction - but [IMHO] if you actually play or sing that 16th note as if it's on the "&" of 1 then you're doing it wrong.)

In any event, I think the very least least MuseScore could do is allow the user to manually change the direction of the partial beam - particularly when there's controversy over whether the default behavior is correct.

In reply to by ghicks

As I said, Gould makes no mention of tuplets in this context whatsoever, so it's impossible to guess how she would feel about them.

Once again, I understand the argument you are making. Hopefully you also understand mine. It's just two different ways of thinking about it, with difference preferences depending on one's goal. I'd still like to see an authoritative reference before committing to making a change, but in any event, it seems like it is already being worked on, and may ultimately get changed even without an authoritative source if enough circumstantial evidence is found. And yes, as a future enhancement, I've already said many times it would be nice to make this something you can flip.

Do you still have an unanswered question? Please log in first to post your question.