[Bevo]

I'm writing a Piano Sonata in A-flat in the Classical Era style, and I'm working on the development section right now, which has always been a struggle of mine. I started in B-flat, briefly went to B-flat minor, then back to B-flat (I do a lot of modal modulations), then went to G (via the swift progression of B dim, C, C# dim, D, and then G). I then am going to perform another modal modulation to G minor so I can move back to the relative major of B-flat and work my way back to the home key. Now obviously there isn't a more distant key to A-flat than G, so there are A LOT of accidentals used. And I know there weren't too many instances in the Classical Era where there were this distant of a modulations, but in instances like this should I use an actual change of key signature in the middle of the piece? I've just never seen Mozart or Haydn do this before. And of course I might even go back and stop the modulation at C major then go to C minor, instead of going to G then G minor, but did Classical Era do this in the middle of a development section? Thanks in advance.

 

[Rik1]

I am sure I have played Classical symphonies (first movements) where there was a key signature change marked. I think the first movement of Haydn's symphony 83 changes key signature from 2 flats (G minor) to 1 sharp (G major) for the final recapitulation which he write in the major rather than the minor mode in the exposition. True, it's not a key signature change for the development section, but it does show composers did change key signature if they felt it made more practical sense. It would have been odd for Haydn not to change the key signature in this instance otherwise you'd have the piece ending in G major, but with a G minor key signature. It doesn't look neat.

I think it's really about practicality from the performers points of view. As a performer if the key change is only for a brief period (a remote key in Classical is likely to be only for a brief section) then a new key signature isn't worth it and might be irritating to see such a thing for just a few bars. However, if the key change is for a considerable length and is a significant section of the work then a key signature change might be appropriate. This is because as a performer, seeing lots and lots of accidentals on the page is more tiring when it isn't necessary. But bear in mind that a key signature change tends to indicate that this is a significant new section of the work.

 

[hreichgott]

A flat is the Neapolitan of G, and Beethoven loved a good modulation to the Neapolitan. Preferably when the Neapolitan comes on quite suddenly. There is one from G to A flat, as it happens, at the beginning of the development section of Op. 2 no. 2 which I am memorizing at present so it's at the top of my brain right now.
I can't right now think of a modulation to the Neapolitan in the place you're suggesting -- entering the recap that way, where the Neapolitan turns out to be the home key -- but someone else here probably will.
Regarding change of key signature: some composers do, for humanitarian reasons, but typically only if there is going to be a long extended section in the new key. I can definitely think of examples in sonatas of Mozart and Schubert. The F minor variations of Haydn change key signature every time they switch from minor to major. None of the sections I can think of is shorter than 16 measures.

 

[hreichgott]

I need to correct myself, that modulation in Op 2 no 2 is from C to A flat (Neapolitan of the dominant) -- it goes from a G7 chord within C major to A flat major.

I did come across a shorter change of key signature, in the Beethoven bagatelle Op. 126 no. 5, changing from G major to C major for 8 measures. Might be interesting to try to figure out why he bothers changing key signature, instead of just writing a few F naturals.