K. 545 1st movement trills, Mozart

[gaer]

Teaching this trill is very frustrating to me. First movement, K. 545, Sonata in C. Measure 37.

I have no problem playing it, but I also have no idea how I do it, and that causes teaching problems.

So far, I've done this, with students:

Step one: Play RH AGAGF# F#, thinking of the whole thing is a 5 note tuplet. I have students play only the C in the LH, first beat, then resume 16th notes on beat two.

Second step: keep the RH as even as possible, playing the 5 notes a bit faster than normal 16ths (just trying to "rough it"), playing the first two notes with the first in the LH, then the next three with the remaining 16ths. This is an obvious distortion making the first 16th double length, but at least I can put the emphasis on getting the RH even.

Third step: "And then a miracle occurs".

I've played this measure countless times on an electronic keyboard, full speed, then slowed it down to see what happens. My LH is even, even when slowed down. The RH is not really even at all, with the first notes compessed a bit, and in slo-mo it sounds very sloppy, but it sounds right to me full speed and also matches what I heard from other pianists I admire.

I've also recorded it, with the above "cheat", making the first 16th in the LH double length, in slo-mo, then speeded it up to the proper tempo (the one I want for myself for performance), and the cheat, while still noticeable, is not outrageous because the beats themselves remain spaced correctly, though the "lingering" on the first 16th is obvious to me and would be to any of you.

I also have no idea how to teach the long trill close to the end of the exposition and end of the recapitulation. I start out with exactly 16 notes, which does not sound horrible for people who are learning, but after recording myself full speed and slowing it down to find out what actually happens, I play between 26 and 28 notes in the RH, including the sufix. It varies. The usual solution is shown as 32 32nds notes in the RH, which sounds like a buzz-saw to me, too fast, inhuman. I can only get 32 in by playing down tempo, then speeding up the playback, and it sounds like a machine. Again, I have NO idea how I learned to play "free trills" above an Alberti bass. I remember not being able to do it, and I remember coming back to the problem later and being able to do it.

This is no solution for a student.

Any suggestions?

Gary

[pianistimo]

i'm no expert, but i know what i personally like.  starting with the trill at the end on measure 72, i would not start with the upper note.  this is why:  there is a sixteenth note run that ends on C in the previous measure.  to make everything 'flow like oil'  i would continue the trill as a part of the run (excepting speeding it slightly).  in other words, i would start the trill on d and make it triplets against two sixteenth notes.  ded to G F, and then ede to B F.  another book i have for young beginners DOES start it from the upper note.  maybe it's what the student can handle.  you're 'supposed' to start from the upper note (if the preceeding note is not immediately above), but in many other places in this piece, you trill from the note because of that.  so, if your student can handle remembering to do a trill one way in one place and another in another - maybe from the upper note would be better for serious performance (just in case people are picky).

for measure 37, i'm not sure that i fully understand the question.  i would play the lh C and rh run in sixteenths on the beat that they should follow.  when you teach students about 'fudging' this and that to get 'artistic' sounds at too early of an age, you risk them taking too many liberties, imo.  i just count it, one - e - and -a , two- e- and - a, etc.
and there would be no pauses or drama to it.  the only thing i would have students practice is beat two, where you have to lift the lh and give a rest.  and also at beat four with the rh.  often students just hold it all the way through.

for the trills at the beginning, i would also start on the note.  ( you didn't ask about this) but for others i would simply (m 4) play a triplet of fgf, and then efe.  it makes it a simple enough trill for a beginning student, and follows the idea of not trilling from the upper note if the note preceeding is only one note above.

hope this sounds right to others as well.   

[gaer]

i'm no expert, but i know what i personally like.  starting with the trill at the end on measure 72, i would not start with the upper note.

Remember, I'm talking about teaching this to people who are not advanced students. Some teachers would refuse to do this, but I like to teach something, get the basics down, then return to it every three to six months, trying to relearn it much faster and bring it to a higher level of performance.

Let me skip around the whole question of starting a trill on the principal note or the upper auxliary. Those of us who are experienced can do it either way, depending on many factors.

Starting on the upper note uses 16 notes in both hands. I'm assuming we are going to play the sufix at the same speed, at this time.

But if you use triplets, you have two problems. First, you can't use all triplets. You are going to end up with 23 or 25 notes. You could still start slower, two 16ths at first, following with all triplets. But now you have a two against three problem in a measured trill. To me, measuring a trill that way, eventually quickly, is much harder than just letting the trill rip.

Instead, the first time I go through this with a student who is not advanced, which is a different story from working with someone advanced), I will stick with 16 notes in the RH because it is a "no brainer".

But if I'm working with someone who is a hard worker and wants to go a bit deeper, I will introduce the idea of just letting the trill go, freely, changing the LH:

1. Just block chords, one to each beat.
2. G, BDF, G, BDF, G, BDF, G, BDF, CE
3. GB, DF, GB, DF, GB, DF, GB, DF, CE
4. GD, BF, GD, BF, GD, BF, GD, BF, CE

By step four, I'm going to stress ending on 24 (CE). But I won't spend a great deal of time doing this, and I have a separate excercise written out showing how to do these steps in the LH.

I will also isolated the LH patterns, as written, and a free trill, going on for several seconds, to work on fluency.

Getting the hands together is the rough step. As I remember, I think I felt totally spastic when I first tried it, with the RH stuttering and the LH uneven, but as I continued to work on trills in many pieces that had the same problem, eventually it just fell into place for me and seems to do so for most students.

I want to stress that when I perform this at full speed, in a way that sounds clean to me, when I slow it down (using midi to do so digitally), the RH does not come together with the LH except purely by accident. It sounds totally wrong. Full speed, it sounds right, and it is the accenting of the LH (and evenness) against the speed of the trill that gives the illusion that it is all coordinated.

Let me leave it at that. I appreciate your answer very much, but apparently no one else is interested. I posted my almost a week ago. I had assumed that this would be a tricky spot for other teachers to teach, but I guess I was wrong.

Gary

[pianistimo]

i agree with the idea of using sixteenth notes to trill with when starting out on this piece.  it sounds kind of boring to more advanced students.  and the triplets are to two sixteenth notes and not four, so they do work out evenly without effort.  it's just that if you start out with the main note, it works out perfectly at the end with dcd going to C.  being that it's slightly faster (six notes per beat - instead of the four sixteenth notes) it sounds more 'trillish.' 

one thing that i've wondered though, is that you really do yourself a service to start with the upper auxillary in most trills because the third finger is naturally longer and stronger and gives a sort of accent (naturally) and can maintain the speed of the trill, while starting on a weaker second finger gives you more instability.

has anyone allowed a student to play  (at m 70) a d for half the beat or the whole first beat and then go to triplets starting from the upper note?

[pianistimo]

just so you know i'm original in my thinking, look a  measure 58.  mozart, himself, starts this trill on the lower note.  so IF one decides to start on the lower note at measure 70, it's not THAT terrible.  it's a beginning piece so, learning to trill is probably what he's trying to teach anyway!  the note that is most important is the note that is part of the chord.  and, the d is part of the B chord.  when you get to the end of the trill (using triplets against two sixteenths) the fingering is important.  the last triplet should be 213 on dcd ( i think) to set your hand up to play the arpeggios.

[gaer]

i agree with the idea of using sixteenth notes to trill with when starting out on this piece.  it sounds kind of boring to more advanced students. 

I would not teach such a simplifaction to an advanced students. I would simply listen for an unmeasured trill with a clean finish (sufix). It makes no technical difference to me in ease or difficulty as a player if I start the long trills at the end of the exposition and recapitulation on the principle note or the auxiliary. I've heard it performed both ways.

I wish I could get more opinions from other members about what they play and teach. For instance, I would like to know what editions other people like to work from.

Let's skip the whole discussion of which note the trills should start on, for the moment, since it is really an more advanced problem.

This next part has me confused:

and the triplets are to two sixteenth notes and not four, so they do work out evenly without effort.  it's just that if you start out with the main note, it works out perfectly at the end with dcd going to C.

That's where I'm lost. There are 16 16th notes in the LH. If you start the last trill on E, in the RH, you get:

EDEDED, EDEDED, EDEDED, EDEDCD, C

24 notes. If you start on D, you can't use triplets throughout the measure. It doesn't work.

You could, of course, do this:

*DE DED, EDEDED, EDEDED, EDEDCD, C

Simply play to regular 16ths with the first two LH 16ths, then "acclerate" to triplets.

But that point I made is that 24 notes, to me, sounds too slow, and 32 much too fast. This was what I was trying to explain. I normally play 28. It is impossible to practice doing this. When you play an unmeasured trill, the notes just "fall where they fall". This is why I start teaching students who begin unmeasured trills to play block chords until they have developed sufficient coordination in the RH. Then I work on fluency in the LH pattern (Alberti bass), and I have to work on both trills and Alberti bass in other pieces before returning to this work, the second time in review, to add the more advanced trill.

being that it's slightly faster (six notes per beat - instead of the four sixteenth notes) it sounds more 'trillish' 

But this involves three against two, which is again another problem that has to be worked on separately. Here is generally what happens when you listen to students play this movement. (I'm talking about students I hear who are not studying with me, but I have no doubt mine do the same thing when I'm not around to give them dark looks.)

They go racing through most of it, much too fast, then slow down when they hit this trill and pretty much butcher the measure. To me it is at least one of the two or three most difficult technical problems in the movement, and until you have some kind of solution, which involves addressing this problem first, the movement is inevitably going to break down at least twice (expo and recap), four times with repeats. And it's like announcing to the world: "I don't know how to play a long trill when something a little complicated is happening in the LH."

I'll answer your next post soon, but I would certainly be nice to get input from other people.

Gaer