trill line rules

As I understand it, the standard engraving rules for trill wavy line extensions are:

If the trill is on a single note, no wavy line is needed.

If the trill is on a set of tied notes, the wavy line extends to the last note in the set of tied notes – but not past it. E.g., if the last note in the set of tied notes is a half note, the wavy line extends to the right edge of the notehead, but does not extend for the notes duration.

I don’t mind having Dorico make me choose, trill by trill, which ones have wavy lines, but when wavy lines ARE used, it would be great if I don’t end up having to slide each of the endings back to the left.

If I am correct about standard rules, can this, at least, be a made into a default option – with wavy lines showing up only when ties are involved and extending only to the right edge of the last notehead?

David Froom

You don’t need to slide the wavy lines back to the left: you can set the appropriate property in the Properties panel to hide the wavy line.

When I have tied notes that are trilled, I always want the wavy line. I don’t want to hide these. But Dorico seems to put it in for the length of the note’s duration, which, as I understand it, violates standard engraving rules. I’d like, at least as an option, to have the wavy line stop at the right edge of the last note in the set of tied/trilled notes.

That notation would confuse me, as a keyboard player. I would assume you wanted the trill to stop at the end of the wavy line (and I might wonder why the notes were tied together) - but that doesn’t seem to be what you think the notation means.

I am also a keyboard player.
I don’t think I’m crazy, but I remember clearly that one of the engraving books I studied specified trill wavy lines the way I describe them. I found one such instance online:

If the trill extends over two or more notes, the wavy line continues to the end of the notehead of the last note affected by the trill
-Alfred Essential Dictionary of Music Notation

Does anyone have Ted Ross handy? Or the Schirmer Engraving booklet? I can get both of those early next week, but not until then.

What is clear, from my internet search, is that many folks (and many old engraved scores) have wavy lines representing the full durational length of the trill – so Dorico by default does this correctly in the eyes of many.

I don’t know how the Alfred definition would notate “trill3” (from Beethoven Op 106, edition published in Vienna c. 1920). Since the fugue theme starts with the two trilled notes (bass G and E) it’s fairly clear the trill on G is only meant to last for the first quarter note.

The way I read the Alfred definition, that means there should be no wavy line at all, therefore there is no way to show the trill is cut short. Unless “the end of the notehead…” is supposed to mean “the end of the duration of the note…”, which is not the same thing IMO. Or, unless Alfred implies that every full-length trill must have a wavy line, which certainly doesn’t make sense for short trills in the baroque period, unless you like wasting horizontal space!
Trill1 and Trill2 show the difference depending on where the line stops.

The Alfred definition would have no wavy line if the trill is on a single note. I admit that this would be tricky and ambiguous in that Beethoven example. There always will be such problems, ones that demand a unique solution. And, in the Beethoven example, out of context, it isn’t completely clear that the trill should stop.

There seems to be no universal agreement, although I do admit that most scores I’ve been looking at (all engraved before mid-20th century) does this Dorico’s way. Kurt Stone (20th century notation) suggests wavy lines at the duration of the trill, plus a vertical hash to indicate precisely where the trill stops. Haven’t seen that used anywhere outside of his book, but I’m guessing some people do that.

I’m not advocating or arguing personally for one or the other in terms of clarity. I KNOW I read some years ago about doing it “my” way (Alfred’s way) in a highly couple of reputable sources – so reputable that I changed my way of using wavy lines. I’ll track down those sources. Assuming I’m representing at least part of the musical engraving world on this, it would be nice if Dorico had an option to set it to behave the way I suggested.

To quote Gould who seems to be the modern authority:

“The trill line–a shaded, wavy line, placed directly after the trill sign–is used to indicate the extent of a trill. The trill line is optional for a single note-value, but must be used with tied notes. The trill line extends for the entire duration of the note, and not merely as far as the last written note. [if] in the middle of a bar, the trill line continues right up to the following note-head or its accidental. … When ta trill is to terminate before the end of a note, divide the note-value to show the exact length of the trill. A tie indicates that the note following on from the trill is not re-articulated. To indicate that the trill finishes before the next note, not with the next note, terminate the trill line with a vertical notch. Place the cut-off point before an accidental for the following note; at the end of a bar, place it on the barline.”

FWIW in her examples, if a trill line is connected to a whole note, she does extend the line to the end of the bar to indicate that the trill continues. The only time I’ve ever seen trills without lines on extended notes is in baroque music. I’ve only noticed trill symbols without lines on notes of 2 beats or less (ie- half notes).

Take a look at Brahms Op 118 no.5. In both Peters and Breitkopf editions (see IMSLP), most trills on single notes (halfs and wholes) are without wavy lines. The wavy lines are reserved for trills in subsidiary voices.

I’ll post my sources when I find them. I have the Schirmer engraving manual (not with me at the moment), which I got from a friend who worked at Peters in NYC. I’ll check Ted Ross’s “The Art of Music Engraving and Processing” also (that’s in my college’s library). Ross was published in 1970, and was considered by almost all the folks I know in the copying business as the “bible.”

Also, FWIW, I have had a lot of performances of my music, and the trills have never gone wrong.

I haven’t seen Gould yet, but I hear it is fabulous. Perhaps she should be regarded as supplanting Ross.

You should definitely get it (although it is pricey!). I simply love it. It has been worth every penny. I think her examples are interesting-- I definitely hear you in terms of single notes. Her examples sometimes have other notes tied to them from the previous bar. I would respect just the symbol on a single note although I certainly don’t think it will cause any problems to add the line if you think it would warrant it or add clarity. I think that’s really her M.O. in terms of the lines: they are unambiguous. No guessing how long, just do it for the length of the line displayed. My only regret with trill lines is that for longer passages the score really can become cluttered quickly. :unamused:

OK, I found my copy of G. Schirmer’s “Manual of Style and Usage.” Besides being the house style for G Schirmer throughout at least the last half of the 20th century, it was used by other publishers (as I said above, I got a copy of it from a friend in the NYC office of Peters). It was to apply to computer programs, hand copying, and engraving. From what I can tell, the copy I have was updated in the late 80s or early 90s (Finale and Score were both capable of doing acceptable work by the mid 80s). As far as I know, this style manual is/was not publicly available. It was (still is?) distributed to copyists and engravers.

As with many style manuals, there are things I like and don’t like. Their recommendation for tuplet brackets, for example, specifies that the rh side of the bracket extend only to the edge of the notehead, no matter the rhythm, so 1/8 - 1/4 under a tuplet bracket would NOT cover the rhythmic duration (that style seems to have been supplanted generally). They also suggest that septuplets be normally regarded as 7:8, while it seems more typical these days for them to be 7:4.

Their advice on most things is very fine. And, as a concise 32 page style manual, it makes an excellent and quick read. It often resorts to comments like “an experienced engraver will know how to handle” any number of complex situations, so it doesn’t try to cover everything, but instead seems to encourage people to consult with the “silverbacks” if questions arise.

With regard to trills, their style is this:

Trills should be indicated with the traditional tr above the note. When a note with a trill is tied, a wavy extension line is added [they give an example]. Notice that the extension line only extends to the second notehead; it does not extend for the full duration of the second note. Do not use the extension line if the trilled note is not tied, regardless of its duration [I’m skipping the explanation of the differences between trills and tremolos] …In contemporary music, do not use an accidental above the trill to indicate the ancillary note. Instead, use a small note in parentheses [they give an example]. When a trill is tied onto another line [that is, over a system break], use parentheses: (tr). The extension line is not necessary, unless the note is tied for additional measures.

I follow this practice, more or less. I DO use tr with a sharp or flat above it; I leave it to players to assume that tr without a sharp or flat will be to a natural, but I add the natural if there could be any ambiguity. I use a bit of wavy line, not (tr), to indicate trills continuing over a line break, so that if I change line breaks, I have a minimum of adjustment to do.

As commented on by Romanos401, wavy lines can add to score clutter. I like Schirmer’s style suggestion both because it is unambiguous and is as uncluttered as possible. As I said, I have never had performers handle my trills incorrectly. When a wavy line extends to and completely covers a tied note, it is clear that the trill continues, especially if tr without a wavy line is the default for notes that aren’t continued with ties. Granted, there will be complex and unique situations (like the Beethoven sonata example in a previous post) that require specific solutions.

What I like VERY much about Dorico is that it seems to wish, through user-controllable default settings, to accommodate nearly every standard style choice. Different folks like different looks – which is why in written American English, we see differences between MLA and Chicago style manuals. Gould, I am assuming, is Faber house style. I am hoping only that, for my way of wanting to use wavy lines, that Dorico could allow me (and others who use this style) to have it set this way.

David Froom

Does anyone know if you can set a default so entering a tr will result in the line NOT being shown automatically?
I cannot seem to find a setting for this.

You can control the default behaviour in Engraving Options / Ornaments / Extension line for trills.

I have a copy of the Schirmer manual of style on my desk somewhere at work, so I will take a look at this for myself.

Thanks, Daniel!

Thanks András for some reason I could not find it and it was faster doing it manually until I found out.

One final post from me on this: I checked Ted Ross’s excellent book, “The Art of Music Engraving and Processing.” He recommends no wavy line for single, untied notes. Then says: "Trills extending over several tied notes have tr over the first note and a wavy sign (same as arpeggio and glissando) extending over the remaining notes…The wavy line extends and ends above the last affected note and not the duration of the last note.

Then he gives two examples, the “correct” one having the wavy line end vertically with the right edge of a whole note, and the “incorrect” one with the wavy line filling a measure with a whole note, showing the trill’s full duration.

Gould’s overriding principles seem to be: tradition, readability/clarity, lack of clutter, modern practice, with each of those taking precedent depending on the situation. Typically, she acknowledges competing standard practices that she wants to avoid. I wonder why she doesn’t do so in the case of trills. To me, the extra length of wavy line adds clutter without adding clarity. If tr by itself with no wavy line means to trill the duration of the note, why wouldn’t a wavy line that covers the final note in a set of tied notes also mean to trill the duration of that note?

But I know this is a matter of taste and style, and my age may be showing.


The Gould book is basically the Faber House style, not the rule for trill line terminus. The Theodore Ross book is a better reference for the proper trill line terminus.