I'm working on M0zart's last sonata and have a question which maybe someone with a better score (or better knowledge) can answer.

Bar 141 of the Allegro is insanely difficult to play without a hiccup. I'he heard some pianists (including Uchida) leaving out the low D# in the LH, or in any case it very much sounded like they did. It sure makes that hellish bar a lot easier. Would this be accepted practice, or is it maybe written like that somewhere in an Urtext ?

Better knowledge? Me? Certainly not. Better score? Don't know, but mine is published by G Henle and is undated, but it contains a foreword dated 1963 by Otto von Irmer. The edition claims to be based on (mostly facsimiles of) autographs, contemporaneous copies, first editions, and other early impressions. Edited by Walther Lampe who has provided suggested fingerings.

For the bar in question there are no footnotes to discuss any potential problem, and the low D# is certainly present.

In case it helps, the suggested LH fingerings for the whole passage are:
Bar 138: 4 2 for the first two notes (F#, B) and 3 for the last note (B).
Bar 139: 1 2 for the first two notes (D, F#), and 1 again for the next D.
Bar 140: 1 for the first note (G), and 3 1 4 for the last three notes (E, G, E).
Bar 141: 2 for the first note (G), 1 for the next G, and 1 for the last note (D natural).
Bar 142: 1 5 3 2 4 for the last five notes (A, E, A, C#, A).
Bar 143: 2 for the first note (C#), 3 for the first F#, 1 for the D.

For bar 141 this seems to imply 4 for the problem low D#, and more particularly an identical pattern of 1 2 3 for both of the two descending G-F#-E sequences which are an octave apart. The idea no doubt is that the hand will feel more sure of where it's leaping to if it has to do basically the same thing where it's going as where it has just been.

For completeness, the suggested RH fingerings are:
Bar 138: 1 for the second last note (the third B).
Bar 139: 4 for the first F#, then 1 2 for the D C#, 2 5 for the A# D, 3 for the last F#.
Bar 140: 1 3 for the last two notes (E, G).
Bar 141: 2 5 3 for the D# G F#, and 4 for the C#.

I'm only the messenger, I haven't actually played it. It's too fast for me!

If Uchida does omit the D#, then it must be OK for you too, but then maybe you should also omit the A# in bar 139, for consistency.

I'd be inclined to say that bar should be manageable, D# and all, if you don't insist on legato all the time. (I'm always after students for that, too.) My edition (Peters) doesn't even have phrasing slurs in that section. It would make complete sense to finish the phrase on the D#, whether you're using legato or portato or whatever, then move in the air over to the high G to start a new phrase there. The movement through the air, far from being a "hiccup," can helpfully emphasize the start of a new phrase in the LH esp. if it begins in the middle of a RH phrase.

It might work nicely if you took a similar little breath right after the A# two measures earlier, even though that measure doesn't present any technical obstacles to legato.

Chris,
I think that bar 139 is not difficult at all whereas 141 presents a challange. However, the way that I would approach it really neutralizes the difficulty - if you can do it that is. I would play the 1st D# of the measure (RH) with the thumb of my LH (together with 4th finger playing the F# in the bass) thereby allowing me a 16th rest to reach above for the next high note. This high note G#, I would play with 5, together with my LH 5th finger playing the bass E. Both of the next notes (RH and LH) I would play with 4th finger: that's right, crossing the LH 4th over the 5th (that takes care of the D# in question). Then I play BOTH of the next notes (RH and LH) with just the RH (5&1). Then the LH takes over again with the next lower note (F#, 2nd finger). The RH can continue from the E to the A of next measure with fingers 5,4,3,2,1.

This is much more difficult to describe with tesx than to execute. With the above approach there is hardly any difficulty to it. IMO.

Give it a whirl. I hope it works for you.

Thanks all for the replies. I have no problem with fingering elsewhere (though it's real tricky in places), it is just that darn LH D# in bar 141.

Whatever the 'solution', yes it would make sense to treat bar 139 the same even though there's no technical need for this.

@Eddy: I did already try out a solution like you suggest but found it impossible to bring off in tempo and keep it smooth. I've no doubt you can do this but I am not that good.

I did see a 13-year old Asian pianist playing it perfectly as written, so I may yet try. But yeah, if great pianists can cheat a little, why can't I

I am a bit envious to kids and teens that plays pieces that are hard, well it's hard for me, so easily. Anyway, I'm not yet into Mozart's pieces but will do someday.

I have two editions: one is the Peters and the other Koenemann Urtext. The fingering suggested in te Peters for bar 139 (4 on a sharp followed by 2 for the d seems to suggest relasing the a sharp and not playing legato. On bar 141 there is no fingering, but, by analogy, this is what should be done.

I realise I am wrong and know nothing, but Mozart would not, by any streach of the imagination, have wanted legato because such a thing would have been impossible in his time on the existing pianos. It is only in our times that musis is written for instruments that are yet to be invented that can perform it.