When I started writing my Scherzo in D, I initially thought of having the time signature be 3/4 since that is a very typical time signature for a scherzo, especially earlier scherzos that are closer to Minuet and Trio. I wrote my entire Scherzo section without even thinking about the time signature. I focused on melody and harmony when writing the Scherzo section of my piece. Only now has this dilemma on the time signature came to me. This is partly due to some people giving feedback that my piece is in the wrong time signature and that it should instead be in 2/4. And now I'm getting feedback saying that I can't use just a single time signature for my piece.

That, the feedback I am getting saying that I can't use just 1 time signature and that I have to instead use both 2/4 and 3/4 is really confusing me. Here is the piece in the original 3/4:

enter image description here enter image description here

As you can see, everything fits neatly into 3/4, at least with the 3/8 anacrusis. But then, this user Teodor Lontos(who gives detailed comments on my scores) says that the 1 quarter 4 sixteenths figure fits better into 2/4 than 3/4 and that my pickup, my anacrusis, should only be the first eighth note. So far I have no issues. So I change the time signature to 2/4 and save it as a separate file from the original 3/4. Now, I'm noticing a mess. Some of that, I can fix by deleting and readding notes, ties and the key signature change. But, even after that initial cleaning step, some mess still remains, all having to do with one of 2 things, those being:

  • Accent(like for example, the Neopolitan chord moves from being on a strong beat to being on a weak beat)

  • Ties, specifically, ties across the measure(for example, the augmented sixth chord moves the cadential 64 and root position tonic to be tied across the measure, leaving a quarter rest at the end of the Scherzo section)

Here is what I'm talking about:

enter image description here enter image description here

As you might be able to tell, quite a few bars that were neat in the 3/4 version(like for example the triplet bar) are messy in the 2/4 version. Also, in Musescore, it is impossible to have a volta start in the middle of the measure so either I separate it into 2 1/4 measures to get the same playback as in the 3/4 version or have the second ending be abrupt.

So then I tell him about these messy areas in the 2/4 version and he says:

That is because the first bars, you wrote in 2/4 and then later you wrote in 3/4. The correct way to notate it would account for this change in accent.

I then asked him how and he told me that my scherzo can't be in a single time signature, that the first 8 bars fit into 2/4 nicely, but by the ninth bar I need a 5/4(or 3/4 + 2/4) and then the rest of the first section should be in 2/4. This got me confused. My original 3/4 seemed neat, much neater than the 2/4 version. And I don't want to confuse the pianist with multiple time signatures if I don't have to. I personally, usually see time signature changes corresponding exactly to tempo changes. But, maybe that is because I'm used to playing Beethoven. I myself might get confused if I see 2/4 and then a few bars later 3/4, especially if the changes are not at a consistent interval or at important moments of the piece.

Anyway, do I really need to use multiple time signatures for my scherzo? Or is it just fine notated in 3/4 as I originally wrote it?

  • I didn't go right through everything but that first score in 3/4 - the D crochet right before the first barline actually feels like the downbeat. You sight singing the first 8 measures or so - it just doesn't feel like 3/4 to me. Of course you can notate it that way (you can notate it in 7/16 if you want) but it doesn't feel like a 3/4 melofy. The accents seem to fall in the wrong places. Unless that is the effect you are after. The 2/4 flows a lot more naturally.
    – danmcb
    Nov 13, 2019 at 20:15
  • But there are places where the 2/4 places unnatural accents or where the 2/4 makes a previously accented note unaccented. An example of the latter is in measure 24 of the 3/4 version with its accented beat Neopolitan chord. The 2/4 places that Neopolitan onto the weak beat. An example of placing unnatural accents is in measure 33 of the 3/4 version, with the eighth notes leading into the half cadence. The 2/4 places those eighth notes onto beat 1, placing an accent on notes that wouldn't be accented. And there are several other examples of the 2/4 changing accents to not fit the melody.
    – Caters
    Nov 13, 2019 at 20:37

3 Answers 3


Based on your description, even without looking at the music, it seems like you really need two time signatures. Looking at the music, I can also see very clearly that some things work better in 3/4 and others in 2/4. The only way it would seem to work in 3/4 alone would be if you were trying to play with expectations, ie, if you want the standard accents of the meter to be contradictory to the implied accents of the actual music, which could be fun.

As others have suggested, you can write it in whichever time signature you think is best, including ones that don't really have a great argument for them, but you have to consider the context for which you are writing.

If you are writing for yourself and yourself alone (no one else is performing it or grading it), then you have free reign and only need to do what you prefer.

If you're writing this for someone else to play, you want to consider two things: 1. What will be the easiest to read?; 2. What will be the best at conveying your intentions to the performer? These two things are often the same but not always. You may choose to write something in a way that's a little less easy to read but do a better job of expressing your musical ideas, such as a syncopated, staccato rhythm being expressed as either dotted notes with staccato markings or short notes with rests in between. It may be easier to read the short notes with rests but it doesn't indicate your desired phrasing as well.

If you're writing this for a class or private instruction, you need to be prepared to argue for your choice in meter, ie, why did you go with one over the other or both, how does this choice affect the performer that has to read it, etc. In the eyes of academia, there are basically right and wrong choices for how to go about a lot of things, particularly before you get to 20th century music and beyond, but most professors will accept something non-conventional if you're able to put together a really good argument for it.

For myself, when I write things out, I often struggle between writing things based on theory vs. what will be easier to read. I'm big on theory, so I prefer to right things out in such a way that spells out the theory behind a given section but that may ultimately be harder to read. This comes up a decent amount when you're writing out things like altered chords, where chord theory would make you want to name everything relative to the chord tone or alteration, instead of naming it based on what will be easiest to read. In my mind, I would better understand the intention of a piece of music through seeing things spelled out based on the theory but I also understand that using an enharmonic equivalent for a diatonic note is pretty jarring for someone reading the piece.

So you really just need to try to consider who will actually be reading this music and how your choice of meter will impact their interpretation of that music.


This is a rather opinion based comment (and thus not necessarily appropriate for this board) but I think the 3/4 reads easier. (There are scherzos in 2/4 and 4/4 though.) I would go with whichever time signature you (the composer) thinks represents the basic pulse of the music. There can be short sections with another pulse but don't need signature changes. Long sections (like a trio) may have a different signature.


I wonder if you called your piece a scherzo because you were hearing your music in a fast 3/4 or if you chose a fast 3/4 because you wanted to compose a scherzo. :) To me, it does not seem secondary to establish this.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.