Close, but incomplete Beethoven Sonatas

[Jo498]

Despite Richter and Serkin my favorite Beethoven half-cycle is Gelber/Denon, I listened to most of it in the last two weeks and liked it enormously.
I listed which sonatas he recorded for Denon in #27. The most glaring gaps are opp.106, 109, 110. Gelber also clearly had favorites because he re-recorded all the sonatas he had done for EMI in the late 1960s and 1970s (opp.13, 27/1+2, 28, 31/3, 57, 81a, 90), just like Serkin did with his 1940s/50s and 70s recordings, so no gaps can be closed.

[George]

It's a bit surprising that a pianist like Serkin who was so strongly associated with the Austro-German repertoire apparently had less than half of the sonatas in his longterm repertoire, and when there were plans to make a complete recording for the anniversary 1970 he couldn't (or didn't want to) learn the remainder fast enough. Apparently, his core was basically the "named" sonatas opp.13, 27/2, 53, 57, 81a and op.78 and the late sonatas.

A bit strange for me is also that the sonatas Serkin did record in 1970 (apparently in the course of this abandoned project) are mostly not among the more frequently played pieces, e.g. the least popular of op.10 (2) and op.31(1), op.22 or op.27/1. I wonder if this was him going by some schedule originally planned for a complete recording or by preference. Admittedly, I find these interpretations not particularly noteworthy for the most part, i.e. they are not going to make these sonatas into favorites or become favorite recordings them.
It's a pity that Serkin didn't get around to record sonatas like op.10,1+3, op.31/2, op.90 where I could imagine him being more interesting than in some of the ones just mentioned.

While I often wish Richter and Serkin (and Moravec) recorded more, I have come to believe that what makes their performances special is that the pianists chose works that they loved. That could explain why it was attempting to record the complete Beethoven sonatas that Serkin started to turn up some duds. Could also be an age thing, for I find Serkin's early Beethoven to better his later recordings.

[Jo498]

I was precisely wondering if the choice of sonatas Serkin recorded but apparently not really added to his repertoire in ca. 1970 was driven by personal preference or by some plan for a complete set that never came to fruition. It's not mainly that the recordings are not as good as his earlier ones, it's that it's mostly sonatas I both personally don't much care for and that mostly seem not among the popular ones, so I really doubt that op.10/2 or 27/1 were particular favorites of Serkin, maybe it was just that he started with them because he wanted to do all and these were easier to play or whatever.

OTOH Serkin did have at least one really odd favorite, namely the Fantasy op.77... and there are other pianists (like Michelangeli) who also picked mostly sonatas I don't much care for...

[VonStupp]

While I often wish Richter and Serkin (and Moravec) recorded more, I have come to believe that what makes their performances special is that the pianists chose works that they loved. That could explain why it was attempting to record the complete Beethoven sonatas that Serkin started to turn up some duds. Could also be an age thing, for I find Serkin's early Beethoven to better his later recordings.

This is why I prefer Beethoven sonata cycles (and the SQ cycles too) that are culled together from over a decade or so. It seems less of filling a quota to meet the bottom line by recording over a weekend or two, and more of an attempt to gain insight through living with the music over time. Today, though, meeting the bottom line may be more important since it seems the artists themselves foot the bill anymore.

Regardless, taking that kind of time leaves the chance of not finishing a cycle.

[Jo498]

We don't always know (although of course it is often known or could be researched) how long artists have "lived" with the works before recording them.

There is something to be said for a set of works recorded within a brief period of time to give a consistent picture (Gulda/Amadeo was done within a few weeks, I believe), Schnabel or Arrau (Philips) were done within around 5-6? years but they also show a consistent picture. Similarly for the Gelber/Denon (ca. 1987-92)

Pollini's is an example for a cycle that took almost 4 decades and accordingly development/differences which can also be interesting.

With Richter we have the passage of time, inconsistency of (live) recordings and other factors.

Of course, inconsistency or diversity is not only a question of time. Gould's Beethoven was recorded over 15 years or so but the inconsistency might be more because he was a bit crazy than due to passage of years.

[VonStupp]

We don't always know (although of course it is often known or could be researched) how long artists have "lived" with the works before recording them.

There is something to be said for a set of works recorded within a brief period of time to give a consistent picture (Gulda/Amadeo was done within a few weeks, I believe), Schnabel or Arrau (Philips) were done within around 5-6? years but they also show a consistent picture. Similarly for the Gelber/Denon (ca. 1987-92)

Pollini's is an example for a cycle that took almost 4 decades and accordingly development/differences which can also be interesting.

With Richter we have the passage of time, inconsistency of (live) recordings and other factors.

Of course, inconsistency or diversity is not only a question of time. Gould's Beethoven was recorded over 15 years or so but the inconsistency might be more because he was a bit crazy than due to passage of years.

True, and I like your breadth of examples.

I personally like hearing a growth of vision and the experience of performance over time than I necessarily do consistency. This could certainly give a ragged amount of highs and lows depending on the span of time, the quirks of personality and personal health, and recording technologies and situations, so I completely understand the want for a cohesive insight over a cycle, complete or not; but for me it is not essential. Maybe what I really want is the development and diversity you speak of, and performances over time have more variegated possibilities of that aspect.

On the other hand, I don't know how many cycles Barenboim has given us up to now, but have we been given further insight from him over endless complete cycles? Perhaps not a discussion for this thread.