This blog post is written by Alexander Plötz, a music engraver, editor, proof reader and more based in Dresden, Germany. Read on to learn more about using Sibelius’s third keypad layout to exert fine control over beams that continue over rests.
This tutorial originated from a discussion over at the Sibelius help forum on how to achieve beam notations like, for example, the ones used in Bela Bartók’s fifth string quartet:
Although there are three global options available regarding beams over rests (see Notations Tab > Beams > To and From Rests / Over Rests / Stemlets), those are often, well, just too global for sophisticated notations (like with the quartet above), and more often than not such details will have to be edited by hand. This is where a deeper understanding of Sibelius’s handling of beaming properties in connection with rests proves very useful for avoiding the frustration that comes with a strictly-trial-and-error approach when using the third keypad.
(Note: for the following the mentioned global options are assumed to be switched off. Also, all the examples below cover tuplet cases, but the process is the same for non-tuplet notations.)
The most important thing to keep in mind is that Sibelius will set all rests to “no beam” by default (which seems reasonable, given the fact that rests do not have beams). So with a notation like the one in the upper staff of the following example it helps to imagine some kind of “virtual” beaming like it appears in the lower staff: a lot of single notes with flags (notes that stand for rests appear in cue note size).
To change these defaults, use the 7, 8 and 9 keys on the third keypad, manipulating the rests just as if they were notes set to “no beam”. With this simple guideline, beaming over rests becomes as straightforward as beaming actual note groups.
Be aware, however, that any such beam must contain at least one actual note (eighth note or smaller value); there is no such thing as a “beam over just rests”. Also, just as it is not possible to beam an eighth note to a quarter note or any even longer value (as, by definition, those note values have no flags that could be extended into beams), the same goes for beaming to rests of such longer values.
The next example shows several beaming constellations along with, between the two staves, the appropriate keypad setting. As can be seen with the penultimate note, even secondary beams work like they would with notes instead of rests.
A useful thing to know is that beams over rests will, by default, behave as if the rests are noteheads on the middle line. This means that the angle of a beam will be slanted just as if all the rests were notes on the middle line. By moving rests up or down, the result is similar to if such a middle line note would be moved up or down, to the point that a beam might stay unaffected by moving a rest, since the “virtual” notes follow the usual beam angle rules; if there is a note protruding towards the beam and across the line between the outer notes of a group, the beam will stay horizontal (as with the beam at the beginning of the second quintuplet in the example below).
There are two other noteworthy peculiarities: while a note will move in intervals of a half space, a rest will move by a whole space, but with roughly the same effect on the beam angle; also, while a note, when moved away far enough from the majority of the other notes of a beam group, may cause the beam direction to flip, this won’t happen when moving a rest.
Though it is good to know about this behavior in order to understand why Sibelius’s default output of beam-over-rest angles is like it is (that is to say: not at all like an engraver would draw them), moving rests around as a means of beam angle adjustment is hardly an optimal solution. Therefore, a bit of manual tweaking of a beam’s two position handles is often needed by clicking on the beam and adjusting the handle, as follows:
- When you drag the left-hand end, you alter the height of both ends of the beam.
- When you drag the right-hand end, you alter the angle.
- Instead of dragging with the mouse, you can type the up arrow or down arrow. Ctrl+up/down arrow or Cmd+up/down arrow moves the beam by 0.25 spaces.
- You can, alternatively, open the Inspector and edit the Y values there.
The example below shows which values had to be changed to get the beams horizontal. Extra tip: when adjusting both ends of a beam, keep in mind that the vertical position of a beam’s right-hand end is dependent on the vertical position of the left-hand end of that beam. So always set the position of a beam’s left-hand end first.
Beware: such edits of beam handles will be easily reverted by accident when using the Reset Position feature at a later point; therefore they belong into the classic category of engraving tasks that should be done as late as possible. This goes especially for passages of double-stemmed writing, where rests are frequently moved to help distinguish the separate voices: if this is done after editing of the beams’ Y values, there is a good chance that some work will have to be redone, as the beam angle will change with the position of a rest. If you plan to add more voices to a passage, which, by engraving custom, will displace all of a bar’s rests, it is even more important to wait with fine-tuning the beam handle positions until all your notes are complete.
 This was actually a point of debate in the original forum discussion, as said Bartók quartet employs eighth beams pointing towards a quarter note several times, if not too consistently. While it works in context of that particular piece, it still has to be considered as quite some liberty taken by the composer. So the fact that you would have to fake this kind of notation in Sibelius is not a shortcoming of the software: orthographically, Sibelius is spot-on in this case.
 Okay, not completely true… strangely, rests can have all the same beaming settings from the third keypad (including feathered beams, by the way), but they are lacking the Flip fractional beam option in the Inspector. So with the last quintuplet in the example, it is very easily possible to have a 3+2 beaming, but 2+3 would have to be faked; a rare example of Sibelius being not so spot-on.