Engraving question: curved bracket

I don’t find how to write triplet with bracket curved: is this obsolete in engraving rules?
best regards

Dorico doesn’t support this automatically, I’m afraid. It is certainly less commonly used in contemporary music engraving practice than it used to be.

I find that with less advanced players and even with some more advanced players, the curved bracket for triplets is confusing. They think that it’s a slur and will play accordingly and will stop and puzzle things out if the triplet is all on the same pitch. I think the modern widespread practice of using the square bracket broken with the 3 in the middle is much more clear and presents no additional problems in the playing. And over 3 8th-notes beamed together the bracket isn’t necessary at all for most players. Only when there are quarter notes or rests and 8th-notes combined in the 8th-note triplet is the bracket truly necessary.


Sometimes musicians like ambiguity, believe it or not…

Usually they’re called composers.


The use of slurs for tuplets originates in hand-written music where there needs to be a way to distinguish the tuplet number from other numbers like fingering and curved lines are quicker to draw than square brackets. There was also a preference at that time for placing the number on the note head rather than the beamed side as being a more visible.

If one prefers the older system, a simple solution would be to create tuplet number characters that have a small slurs that don’t extend much beyond the number over or under the number. In this way they are distinguished from legato slurs:
Tuplet slur

Having never seen this before (or at least, not remembering seeing this before…), I can confirm that I would have read the op’s examples as triplets that were supposed to be slurred, but were poorly engraved.

John’s examples are better, but I would have still second guessed myself, thinking the slurs were just poorly rendered by a computer program.

Hi, Romanos. The tuplet slur was used throughout the 18th and 19th century and well into the 20th. The bracket probably came in at the beginning of the 20th century. (If only there were a reference book that discussed such things.) If you saw the small triplet slur in context, with regular legato slurs all around it, you would have no doubt about its meaning.

Yes, but Dorico uses only one glyph for a tuplet number that will be placed either above or below, in case you add the small arc above, when flipping the tuplet down, you won’t have the arc below the tuplet number, it will stay above it.

The solution would be coding and tell the program what to do…

Elaine Gould’s book “Behind Bars” doesn’t discuss when the change occurs but she does state “The numeral should be encompassed by a square bracket, and not the curved arc of older editions, as this looks like a slur.” Kurt Stone’s book “Music Notation in the Twentieth Century” was written in 1980 and it says that the use of curved lines for tuplet brackets is wrong. Ted Ross’s book published by Hansen House in 1970 called “The Art Of Music Engraving & Processing” says that either the square bracket or the curved line is used. My own observations of published music (the series of books I use in my private music lessons was published in the 1970s and uses curved lines for triplets) are that the change occurred in the 1970s and that after 1980 or so all the newly published music that I have seen uses square brackets to indicate tuplets.


Thanks, NorFonts. As you point out, the program needs to help if done automatically. Ideally there would be glyphs for above and below and also at three different angles.
I guess I was thinking of a workaround in which the appropriate number-slur glyph was placed as an expression. Finale actually has an option for a slur instead of a bracket, but the slur is constrained to the length of the note group, so I simply draw in every slur.

That is helpful information, dhbailey. That was my general impression. A. Arnstein used brackets for triplets in the 1960’s and probably earlier. I just found a Russian edition of Bartok’s op. 1 in which there are gapped tuplet brackets. It could be earlier than 1950. The first edition used slurs. https://s9.imslp.org/files/imglnks/usimg/c/ca/IMSLP00855-Bartok_-Op.1-Rhapsody(1904).pdf

1 Like

Those are all great sources! I think the change was likely pretty standard a decade earlier at least.

1961 - Laszlo Boehm’s book Modern Music Notation (Schirmer Books) states on page 54: “According to the old method, the number indicating the division of time value (2, 3, 4, etc.) was close to the note-head, and the notes or figure were put under a slur. This just added to the ambiguity of possible interpretation of the slur.” The fact that he calls it the “old method” in 1961 likely means that the change had occurred some time prior to that, in Schirmer publications at least.

1963 - Anthony Donato’s Preparing Music Manuscript has the following on page 19: “The practice of enclosing these numerals with a small slur or bracket is no longer followed except in those cases where there might be some doubt about the note grouping. … The preference today is for a bracket, partly because it cannot be confused with legato or bowing slur.”

1967 - Carl Rosenthal’s great little book Practical Guide to Music Notation only calls for straight brackets and says on page 31 that the “traditional small curve or larger phrase-like marking above or below the number can be misread as a phrasing mark (slur) and should therefore be avoided.”


Thanks to all this information and knowledge shared. I had started this post because I wanted to re-write a transcription of a piano accompaniment in which the pianist uses both hands to create a mood that I have notated by hand in the style:

I did several tests in Dorico and I felt like I had never really seen this kind of result (also because of the spacing between the 2nd and 3rd triplet) and being a pretty good sight reader, there is something that bothers me:

I thought the brackets were used for anything other than the 8th triplet. but you have convinced me of my mistake and maybe my method of writing this in Dorico is not good

You could also put the right hand on the top staff and left hand on the bottom staff with the beam centered between the two. Then one or two tuplet numbers where the triplets start and then nothing if the triplets continue on. A bracket wouldn’t be necessary since all three notes would be beamed together.

Of the choices in your example, I prefer the second and fourth, fifth and sixth, but not for the whole measure. One or two is sufficient.

I also found a lot of examples in John Adams “Hallelujah Junction” for two pianos. And I opted for another way to proceed (but longer and which will be my next question in another post) in Dorico, which finally gives: