Diabelli Variations

[Spotted Horses]

I have 17 recordings of the Diabelli Variations. I've never had an issue with listening all the way through. Traditionally, my reference was Serkin père but I guess I should do an actual comparison at some point. (On listening to all 17 versions of the Menuetto, the one on this occasion I most wanted to hear the other 32 variations of was András Schiff's Brodmann fortepiano recording, which seemed to best exemplify the qualities I mentioned above.)

Schiff sounds like a good place to start. My issue with the Diabelli's is generic. I don't do well with large collections of miniatures. I see people reporting listening to the Goldberg Variations, or the WTC, all the way through and I am amazed.

[Florestan]

I see people reporting listening to the Goldberg Variations, or the WTC, all the way through and I am amazed.

Says Wikipedia:

In 1921, at age 17, [Rudolf Serkin] made his Berlin debut performing in [Adolf] Busch's ensemble as the keyboard soloist in the Brandenburg Concerto No. 5. At the end of the concert, Busch told Serkin to play an encore to the enthusiastic audience. Serkin later reported that he asked Busch, "What shall I play?" and Busch "as a joke" told him to play the Goldberg Variations "and I took him seriously. When I finished there were only four people left: Adolf Busch, Artur Schnabel, Alfred Einstein and me."

[Mandryka]

Schiff sounds like a good place to start. My issue with the Diabelli's is generic. I don't do well with large collections of miniatures. I see people reporting listening to the Goldberg Variations, or the WTC, all the way through and I am amazed.

In the Bach variations there's an inbuilt caesura midway with the French overture. I've seen it performed with the artist going offstage for a breather at that point. I'm not sure if there's a similar point in the Beethoven.

[Brian]

Schiff sounds like a good place to start. My issue with the Diabelli's is generic. I don't do well with large collections of miniatures. I see people reporting listening to the Goldberg Variations, or the WTC, all the way through and I am amazed.

I have a lot of sympathy with this. A 15-minute variation movement is great. Even up to 25-30 minutes (Eroica, Enigma). After that, you really need a clear narrative arc or emotional progression. Some of those Reicha variation sets where he's just showing off how clever he is are really hard to tolerate.

I can do the Diabellis about once every six months. I usually favor faster, more classical versions but really enjoyed Uchida's. For the Goldbergs, I like the version Busoni did where he trimmed it down and sliced out repeats for non-boring concert performance.

[Spotted Horses]

In the Bach variations there's an inbuilt caesura midway with the French overture. I've seen it performed with the artist going offstage for a breather at that point. I'm not sure if there's a similar point in the Beethoven.

I am more sensitive to the small scale structure of the Goldbergs. invention-showy piece-canon, invention-showy piece-canon. I usually do one or two cycles (three or six variations) to follow listening to a longer piece.

[Jo498]

Goldbergs with all repeats and slowish tempi leading to 80 min. or more duration can stretch one's patience, so I think it can be advisable to skip repeats. I usually have no problems with (almost) repeat free Goldbergs of ~40-50 min length and neither with the usually 45-50 min (depending on tempo and repeats, I have one by Anda below 40 min and the excentric Ugorski takes 61 and Anderszewski 63 min) duration of the Diabellis. It's not much longer than many symphonies or longish sonatas (like Schubert's last or LvB op.106).
There are several suggestions for "arcs" in the Diabellis, some posted further above in this thread. I think one can also listen to parts instead of the whole but I'd much rather do this with the GBV (like the first half) than with the Diabellis.

[Mandryka]

https://orpheusinstituut.be/en/publications/diabelli-machines8


As everyone knows, Beethoven wrote some variations in 1819, and then left the project for a few years. The 1819 version -- if it can be called that -- is on record here on Track 9 disc 1

https://open.spotify.com/album/17QYKgs6GiFfZ05MY91Z79

It's very stylishly and elegantly and expressively and intimately played by Jan Michaels, a new name for me, but I intend to explore some of his other recordings.

https://www.michielsjan.be/

I can't help but think that The Diabelli Variations is like AoF in a way, in that the composer put the project aside for a considerable amount of time before returning to it. In the case of AoF I always feel that the best music came later, and recordings which just cover the earlier fugues leave out my favourite music.

I don't know if I feel like that about the Diabelli Variations.

[prémont]

In the case of AoF I always feel that the best music came later, and recordings which just cover the earlier fugues leave out my favourite music.

I always felt that the manuscript version only has musicological interest compared to the later printed version, even if the printed version barely represents Bach's final thoughts concerning the work.

I don't know if I feel like that about the Diabelli Variations.

In contrast to you I love the most of Beethoven's music, but there are a few of his works which I never have got on terms with. Among these are the Diabelli variations, Missa Solemnis and Fidelio.

[Jo498]

Unlike the AoF which might technically be unfinished or other pieces with alternative versions, I don't really see a point using a supposed intermediate/workshop stage of a finished and published work like the Diabelli variations. (I am not sure if I see such a point in case of the AoF either, admittedly, but I don't know about the stages of that work to have an opinion. It's a collection anyway, so it hardly matters if there are 12 or 16 or whatever pieces or if there are smaller subcollections or the unfinished fugue a single separate piece...)
I have never seen or heard of anyone playing the Waldstein sonata with the Andante favori instead of the Introduzione or attaching the Kreutzer finale to the earlier sonata it was originally intended for.

[Mandryka]

Unlike the AoF which might technically be unfinished or other pieces with alternative versions, I don't really see a point using a supposed intermediate/workshop stage of a finished and published work like the Diabelli variations. (I am not sure if I see such a point in case of the AoF either, admittedly, but I don't know about the stages of that work to have an opinion. It's a collection anyway, so it hardly matters if there are 12 or 16 or whatever pieces or if there are smaller subcollections or the unfinished fugue a single separate piece...)
I have never seen or heard of anyone playing the Waldstein sonata with the Andante favori instead of the Introduzione or attaching the Kreutzer finale to the earlier sonata it was originally intended for.

Here's the essay which accompanies the release, Joe. I think it's really interesting, but then I'm that way inclined, you may not be.

https://drive.google.com/file/d/1utTnYnpMMypRaUFxwd3is5oN5E_YYRLv/view?usp=sharing

The bottom line is that they argue that the 1823 and the 1819 versions represent two very different conceptions of what the piece is "about", reflecting a shift in Beethoven's own thinking about the relation of his music to previous musics.  The paper argues for it, maybe not at Ph.D level, but it's more than journalism.

[Leo K.]

Here's the essay which accompanies the release, Joe. I think it's really interesting, but then I'm that way inclined, you may not be.

https://drive.google.com/file/d/1utTnYnpMMypRaUFxwd3is5oN5E_YYRLv/view?usp=sharing

The bottom line is that they argue that the 1823 and the 1819 versions represent two very fundamentally different conceptions of what the piece is "about", reflecting a shift in Beethoven's own thinking about the relation of his music to previous musics.  The paper argues for it, maybe not at Ph.D level, but it's more than journalism.

Thank you for this very interesting discussion and article!

[Mandryka]

You can hear the audio recording here - I like Bart Vanhecke's contribution.

https://orpheusinstitute.bandcamp.com/album/diabelli-machines8

[Spotted Horses]

Goldbergs with all repeats and slowish tempi leading to 80 min. or more duration can stretch one's patience, so I think it can be advisable to skip repeats. I usually have no problems with (almost) repeat free Goldbergs of ~40-50 min length and neither with the usually 45-50 min (depending on tempo and repeats, I have one by Anda below 40 min and the excentric Ugorski takes 61 and Anderszewski 63 min) duration of the Diabellis. It's not much longer than many symphonies or longish sonatas (like Schubert's last or LvB op.106).
There are several suggestions for "arcs" in the Diabellis, some posted further above in this thread. I think one can also listen to parts instead of the whole but I'd much rather do this with the GBV (like the first half) than with the Diabellis.

The Goldberg Variations was, as I understand it, written for personal study and domestic performance, not as a concert experience. Listening to the entire thing at a go is inconceivable to me. I like the fact that it is organized and sets of three and I normally listen three or six at a time. And I like all of the repeats taken, and I like it when the performer takes advantage of this to play the second differently on repeat.

[vers la flamme]

Sounds like I need to hear the Uchida Diabellis. A work I've never got on with. I have Arrau on Philips and, somewhere, Barenboim on Erato.

[Holden]

And now there's this recording. It appeared in my FB feed and is Arrau from 1952. I really like it.

https://www.youtube.com/watch?v=qRFdwZzKccg

[Mandryka]

And now there's this recording. It appeared in my FB feed and is Arrau from 1952. I really like it.

https://www.youtube.com/watch?v=qRFdwZzKccg

Shame about the sound, I guess it's bad master tapes.

[Jo498]

Does anyone know why Arrau did not re-record the Diabellis when he recorded some other variations and all piano sonatas in the mid-late 60s (or whenever exactly the "classic" Philips recording was done)?

[Mandryka]

Some discussion here about the 1953, which I'd forgotten about. Tom Deacon managed Arrau for Philips, and thought highly of his musicianship, so I trust what he said to me there.

https://groups.google.com/g/rec.music.classical.recordings/c/5RXPDLsKpmk/m/mcCTd6gEu4QJ

[Jo498]

I have not re-rechecked, I only remember the sound of that early 1950s recordings to be pretty bad. That greyish/silver box (sonatas+ variations, no concertos) I have has one clearly botched job that concerns the disc with the variations op.34,35 WoO 80 where IIRC the beginning chord of the Eroica variations is clipped. I don't recall any problems with the sonatas. But the SQ of the Diabellis seems to me to be the fault of the source material.

[Mandryka]

I have not re-rechecked, I only remember the sound of that early 1950s recordings to be pretty bad. That greyish/silver box (sonatas+ variations, no concertos) I have has one clearly botched job that concerns the disc with the variations op.34,35 WoO 80 where IIRC the beginning chord of the Eroica variations is clipped. I don't recall any problems with the sonatas. But the SQ of the Diabellis seems to me to be the fault of the source material.

And annoyingly Tom in that discussion seems confident that there are very good master tapes in a safe in Hamburg, just no one can be bothered to get them out. He was a VP for Philips.