Thanks for listening, Joachim, and for your interesting comments.
I am particularly impressed with the sheer energy and vitality of the first movement. Your approach reminds me of Schnabel (I don't know if he recorded this particular sonata though) in putting an overall conception and emotive output first.
Thanks for the flattering comparison.
Yes, Schnabel did record this sonata (I believe it's on Youtube), and he does many wonderfully creative things with sound on it. Overall, perhaps, his Schubert is not my favorite, though, since sometimes it seems lacking in rhythmic tightness (a lot of overall tempo vacillation). I suppose if I had any model for this music, it would be Richter. I love his performance of this piece, though my overall conception of it is quite different. In any event, I think you've hit the nail on the head by citing an old pianist like Schnabel (who I guess was one of the first to set Schubert sonatas to disc), because when I listen to modern people (e.g., Uchida, Schiff, Lupu, Andsnes) play this music, I don't hear an interpretation at all. I hear, as Horowitz once said, them practicing it for the 100th time.
It means details are occasionally sacrificed but makes the music much more interesting
Yes, there are so many details here, it can drive one crazy.
Not that every detail isn't important, too; those things tend to be easier to improve when one has lived with the music longer, as I probably need to do with this one.
At the same time I had a strange feeling that the small phrases did not always connect and that the longer lines were not kept together. Somehow this impedes the flow. Reading the discussion I think rainer put the finger on it: you hold the chords of the phrase endings too long. I do understand that you want to create a contrast with the outer movements and for the first minutes it did not really bother me, but doing this throughout some overall coherence is lost. Whether that is an acceptable tradeoff in order to get a more dreamy quality is a matter of taste. Certainly this is a valid experiment. If you go against normal performance practice of a well-known piece you should expect some criticism - but if you are convinced this is what you want it to sound like that's what you should do!
Interesting observation, and I agree with you completely. While I do think Chris and rainer had valid points here, I was indeed trying to elicit a dreamier atmosphere from this one (to contrast against the power and steel of the first movement in particular). But I also acknowledge that it was probably too much and believe that your comment about the lines makes very good sense. I think it's often better to experiment, as you say, and then rein in later, rather than the other way around.
Thanks again for your comments,