Beethoven's Hammerklavier - too fast or too slow?

[Keith_W]

The above video is a discussion on Beethoven's Hammerklavier sonata. Beethoven indicated crotchet = 138, which nobody plays because it is too fast. The video includes a link to a study of 35 pianists which shows that the fastest version is 131. But new research shows that at the time, they used a "whole beat" metronome, which counts two beats as one - meaning crotchet = 69. Which also means that all recorded versions are not slow enough! Even the slowest version (Gould at crotchet = 80) is too fast, and I find Gould to be ponderously slow.

Having listened to fast versions of the Hammerklavier (Schnabel) and slow versions (Barenboim), I think that the faster versions better capture the rhythm and excitement of the Hammerklavier. What do you think?

Here are two videos illustrating the extremes of tempo. This is the Schnabel, at 131. Schnabel was a rather short man with small, pudgy hands and thick fingers. He is physically incapable of playing massive chords at that speed, and his playing is full of mistakes (wrong notes, simplified chords). Have a listen:

And this is the Gould, at 80:

Which sounds more correct to you?

 

[BJL]

I'll take Schnabel any day of the week over just about anyone else. Another very fine (and fast) performance is Paul Badura-Skoda, played on an historic (1815) instrument. I like it almost as much as Schnabel. No idea if it is available at this time. My copy is an LP on the French Astrée label. I find Gould's performance to be boring. The first movement is marked Allegro, Gould sounds like Beethoven on barbiturates.

 

[Keith_W]

I'll take Schnabel any day of the week over just about anyone else. Another very fine (and fast) performance is Paul Badura-Skoda, played on an historic (1815) instrument. I like it almost as much as Schnabel. No idea if it is available at this time. My copy is an LP on the French Astrée label. I find Gould's performance to be boring. The first movement is marked Allegro, Gould sounds like Beethoven on barbiturates.

I agree, but academic opinion thinks that Gould's tempo is closer to Beethoven's intention than Schnabel. If you think Gould is too slow, and according to academic opinion, not slow enough (!!!), do you have a justification for your preference besides "I think it sounds better"? I am hoping for some musicians / musicologists to chime in and give some justification for Beethoven's either incredibly slow or incredibly fast tempo, depending on whether you think the metronome should be "whole beat" (i.e. half value) or not.

As for Badura-Skoda, according to the video (time stamp 5:13), he comes in at crotchet = 113. That seems to be a slightly above average choice in terms of tempo. Is this the Badura-Skoda version you are talking about? Played on an 1824 Graf piano.

 

[GaryH]

I think this is much more likely:

Academic research paper:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0243616

That would put the correct tempo for the Hammerklavier at 126, as per Gieseking (1949), which sounds about right:

Schnabel's is clearly too fast, and the many mistakes obvious due to this too high tempo.

 

[TimF]

I'm glad you brought this up. I now want to listen through to the Igor Levit performance that I see is at about 36 minutes. Maybe it needn't take your entire Saturday to get through the Hammerklavier. I have a few pianists I favor for the Beethoven piano sonatas. I tried to find a word to express how dear this piece is and what a great treasure it is but it didn't come to mind.

 

[BJL]

I agree, but academic opinion thinks that Gould's tempo is closer to Beethoven's intention than Schnabel. If you think Gould is too slow, and according to academic opinion, not slow enough (!!!), do you have a justification for your preference besides "I think it sounds better"? I am hoping for some musicians / musicologists to chime in and give some justification for Beethoven's either incredibly slow or incredibly fast tempo, depending on whether you think the metronome should be "whole beat" (i.e. half value) or not.

Given the lineage of Schnabel's teachers, I would give his interpretation greater weight than pure academic reconstruction.

 

[BJL]

As for Badura-Skoda, according to the video (time stamp 5:13), he comes in at crotchet = 113. That seems to be a slightly above average choice in terms of tempo. Is this the Badura-Skoda version you are talking about? Played on an 1824 Graf piano.

Yes,, 1824 Graf, and sounds like the same performance. The recording with the 1815 instrument was Waldstein, my bad. And thank you for bringing up this topic. It's been a while since I've listened to the Beethoven sonatas, and thanks to your post I'm spending a few happy hours this week!

 

[Keith_W]

And thanks to you, I am listening to Badura-Skoda this week My wife wonders why I need such a nice hi-fi system, she said "all you do is listen to old recordings!".

This is after a few weeks of Mahler, which always makes me feel depressed even though there is no reason in my life to be depressed! As Nietzsche said, if you gaze into the abyss, the abyss gazes back at you. And when we are talking about Mahler, that's definitely true.

 

[Keith_W]

I think this is much more likely:

View attachment 324544

Academic research paper:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0243616

That would put the correct tempo for the Hammerklavier at 126, as per Gieseking (1949), which sounds about right:

Schnabel's is clearly too fast, and the many mistakes obvious due to this too high tempo.

I think your post ends this thread, for me at least.

 

[MRC01]

The Brendel & Ashkenazy performances I have are faster than Gould, slower than Badura-Skoda. Ashkenazy takes it just a bit faster than Brendel, which works better to my ears.
I do like that Badura-Skoda performance, though it's faster than I am accustomed to hearing.

 

[Robin L]

Just got the Warner Brothers reissue of Beethoven's complete piano sonatas, as performed by Artur Schnabel. As far as I can tell, this box has, overall, the best fidelity of all the transfers, so far. I have owned the Seraphim and German EMI complete sets on LP with a few additional earlier dubs on RCA and Angel, and the first EMI and the Pearl sets on CD. I still have the Naxos transfers, which are a little bit noisier and a little bit brighter than the Warner Brothers transfers but not really by all that much. I've been listening to these recordings for over 50 years.

No doubt about it, Schnabel makes hash out of the outer movements of the Hammerklavier. I don't think the issue is solely the tempo, but also his preparation for the recording. I have heard fast performances of the work, and it seems to me that the metronome indication is for the initial passages of the work, though some flexibility should be expected. However, as we shall see, those that can, do.

Peter Serkin recorded an excellent historically informed performance on an original instrument. Pretty close to the metronome indication, with the transparency of the fortepiano's sound making the textures less of a muddle:

Friederic Gulda shaves off nearly a minute and a half in his performance of the opening allegro of the Hammerklavier compared to Schnabel's recording, here in a live and unedited performance. However, the clarity of his passagework and his overall conception of the movement is miles ahead of Schnabel. Perhaps this is how the movement should sound?

 

[orangejello]

Just got the Warner Brothers reissue of Beethoven's complete piano sonatas, as performed by Artur Schnabel. As far as I can tell, this box has, overall, the best fidelity of all the transfers, so far. I have owned the Seraphim and German EMI complete sets on LP with a few additional earlier dubs on RCA and Angel, and the first EMI and the Pearl sets on CD. I still have the Naxos transfers, which are a little bit noisier and a little bit brighter than the Warner Brothers transfers but not really by all that much. I've been listening to these recordings for over 50 years.

No doubt about it, Schnabel makes hash out of the outer movements of the Hammerklavier. I don't think the issue is solely the tempo, but also his preparation for the recording. I have heard fast performances of the work, and it seems to me that the metronome indication is for the initial passages of the work, though some flexibility should be expected. However, as we shall see, those that can, do.

Peter Serkin recorded an excellent historically informed performance on an original instrument. Pretty close to the metronome indication, with the transparency of the fortepiano's sound making the textures less of a muddle:

Friederic Gulda shaves off nearly a minute and a half in his performance of the opening allegro of the Hammerklavier compared to Schnabel's recording, here in a live and unedited performance. However, the clarity of his passagework and his overall conception of the movement is miles ahead of Schnabel. Perhaps this is how the movement should sound?

Thanks for sharing the Gulda video. Really enjoyed it. Richard Goode makes me hear more in the music than almost any other pianist, Ivan Moravec excepted. Here is his Hammerklavier Goode Opus 129. His articulation is just a notch above Gulda for me.

 

[raif71]

 

[orangejello]

View attachment 332434

I truly hated that movie.

 

[Robin L]

Thanks for sharing the Gulda video. Really enjoyed it. Richard Goode makes me hear more in the music than almost any other pianist, Ivan Moravec excepted. Here is his Hammerklavier Goode Opus 129. His articulation is just a notch above Gulda for me.

No question that Richard Goode is a contentious musician, though I note that his tempo in the opening allegro is well under Gulda's. Glenn Gould could be all over the place in his performances of Beethoven, though this performance is quite the marvel. This is Beethoven's last piece for piano, excepting possibly "Rage Over a Lost Penny", and the 6th Bagatelle of Op.126 is performed at a remarkably slow tempo here. It starts at 16.52.

On the other hand, this performance of the final movement of the Op. 10, #2 sonata in F major suggests Gould could have taken on the metronome indication of the Hammerklavier if he really wanted to. I guess it all boils down to how his meds were working on any given day: