Piano World Home Page Forums Our Most Popular Forums Pianist Corner Rubato in Chopin's Barcarolle

I've just listened to two recordings of the Barcarolle, one by Krystian Zimerman and one by Vladimir Feltsman. In my opinion, both of them use too much rubato.

Rubato is OK in many Chopin pieces, of course. But the barcarolle form is based on folk songs sung by Venetian gondoliers, and in my opinion, depends on the rhythmic back-and-forth of the 6/8 time: one...two...one...two.... I can understand a speeding up and slowing down of the tempo, but in this particular piece, rubato within a single bar just doesn't sound right to me. It sounds over-indulgent and almost jerky.

Any opinions?

I think virtually everything Zimerman plays is as close to perfection as possible, and that he is one of the greatest pianists in history. I listened to the first two minutes of his Barcarolle performance and didn't hear anything I would call extreme rubato. Since he won the Chopin competition one could expect him to have a particularly good conception of how Chopin should be played.

How much rubato you like is a personal decision but I happen to like Zimerman's ideas. The right amount of rubato can make the music breathe and come alive. Too much can make it syrupy and saccharine. If you prefer less rubato, make sure you are not reducing the expression to the point of sounding mechanical. What is too much or too little? That's up to you.

I was told that rubato should be primarily in the melody hand, (usually the right), and the hand carrying the beat should stay as stable as possible. If I remember correctly, this idea comes from Chopin. Don't forget, when you steal time by slowing down, you should balance it by speeding up elsewhere. Do what you like but be sure you listen to a recording of yourself to verify that it sounds right.

I'm working on the Barcarolle now too. Is there anything you find particularly challenging?

I don't think Chopin was trying to compete with Venetian gondoliers but instead, tried to capture the style while still being palatable as a solo piano piece. That is not to say that Chopin played with as much rubato as Zimerman or Feltsman, but instrumental pieces don't necessarily mimic their inspirations exactly. I like a lot of rubato with my Chopin. If you don't, there are plenty of renditions available that would be more to your taste. I'm not fond of the barcarolle however it's played.

Heresy!

Is this feeling based on a concept or idea that you have or is it based on an ideal version that you've heard played by someone else? Maybe try Pollini if you haven't heard that one yet. His version is probably quite a contrast in the rubato department. Here's the first that popped up for me and it appears to be live from the 80's:
http://youtu.be/eo6UETnU8yQ

Definitely minimal rubato in that one. Just for fun, here's another Pollini version from who knows when that is quite different:
http://youtu.be/e-4vMJ9c0RQ

Played much slower and with more delicacy. Do either of these address your concerns?

I respectfully disagree.

I found the double trills challenging. And a couple of places where the right hand plays a trill with a melody line underneath, especially because the trill has to be played by the weaker fingers, such as 4-5, or possibly 3-4 or 3-5.

Nice post. I'm sorry you don't like the Barcarolle.

I think the one I prefer - at least in respect of the rubato - is my own rendition. Obviously I'm no Pollini or Zimerman or Feltsman, etc., in terms of virtuosity or interpretation.

But in this case I like a fairly steady rhythm within the bar, in the sections with the main theme. Gradual speeding up (stretto) and slowing down is fine. But I don't like jerkiness within the bar.

I've heard that said before, and it makes absolutely no sense to me. Because I "stretch" a phrase's tempo at a pronounced cadence, I have to speed up elsewhere to compensate?

Regards,

Chopin's barcarolle is a journey across the sea. It starts with the grand departure, followed by a constant underlying feel of the ship's gentle and sometimes turbulent rocking, all of which leads to a glorious and triumphant arrival to your destination.

I agree. It's basically a myth, and I see no logical explanation for it.

The idea that the left hand should be steady while the right hand does the rubato is also a myth IMO. If one actually tried to do this in most cases the two hands would soon be completely out of synch. What actually happens is that the left hand follows any rubato done in the right hand.

Yes, that's right. I heard this from Daniel Barenboim in his Beethoven sonata DVD masterclass.

I respect much of what Barenboim says and does, but, out of context, this comment still makes no sense. It implies a totally arbitrary - not to say unmusical - and erratic interpretation of a musical work. As if - as I implied before - when I slow down the equivalent of a few seconds of time at a pertinent cadenza, I, somewhere along the way have to "rush" to make up those few seconds on the imaginary clock that is ticking the predetermined length of time in which I must play the work.

Nonsense, Mr. Barenboim!

Regards,

I agree. It's not like we have to catch up to the metronome.





Interesting. Do you mean that the RH and LH should at times not match up? Or just that the rubato should be based from what the RH is doing?

Personally, I think the LH should drive the piece and melody should float on top. Otherwise, the RH is pulling the LH along, which for me just doesn't work. In that regard, I would say that while I would follow what the RH is doing as to where I want to rubato (obviously), I would certainly be keeping a strong sense of the LH pulse while 'rubatizing'.

I may have a bit of a unique perspective here as a dance accompanist. I regularly use dance-form classical pieces for accompanying. For class exercises one must have a rock-solid tempo in the LH. The RH can use some rubato as long as the beat remains obvious. EVERY PIECE, from baroque to post-romantic, sounds at least a little distorted when playing this way. These pieces were meant for listening, not for being danced to, and while one should pay homage to the dance rhythm one need not be a slave to it, unless one is playing for actual dancers. And even those actual dancers, when onstage doing real choreography instead of class exercises, use rubato, fermatas, extra rests and tempo changes.

Rubato comes from the Italian (and Latin) verb "Rubare" - meaning literally steal, as said above.

Yes..although I don't think it's a matter of "Having To" as much as it just naturally coming out that way as a reaction to the time taken. My bet is that you do it yourself and hear it being done this way all the time without realizing it, because it is so natural in practice.

Edit: Bruce and Pianolover: Although it's slight, I recommend you listen to Rubinstein at exactly 0:31 in this video:

https://www.youtube.com/watch?v=YGRO05WcNDk

Can you hear how he takes time between the opening F to G, and then slightly (but surely) makes an accelerando towards the END of the group of 6 16ht notes?
It's nothing forced or planned out...but one gets a sense of ebb and flow, push and pull. And this is Rubinstein - one of the most rhythmically tame of all pianists. Listen to how much more Rachmaninoff exaggerates this concept in his recording of the same work:

https://www.youtube.com/watch?v=kj3CHx3TDzw

It was reported by his contemporaries that Chopin's sense of underlying pulse was regular and strict, and that he always keep it going, regardless of any rubato applied over it. And part of how he did it was by compensating for any tempo changes in one direction with an equal amount in the other, so that ultimately it was as if there had been none at all.

So that idea is maybe not so outlandish after all.

But I would guess that it depended somewhat on the piece. I'm thinking he may not have done that with his mazurkas, for example, since he played them with such markedly distorted beats that they came out as having four beats to a measure instead of three. But who knows, maybe even in the mazurkas, any extra rubato beyond the already dance rhythm may have followed that "all time stolen must be given back" principle.

I don't think finding a few examples of something proves anything. I'm sure one could find examples where the pianists sped up and then sped up more.

The simplest example is when might naturally ritard at then end of a phrase. This doesn't mean one has to suddenly speed up afterwards.

Of course, in any piece with considerable rubato, there will be many examples of slowing down and speeding up but I think this has absolutely nothing to do with the idea that "stolen" time must be given back. It's just how the pianist feels the rubato...the natural ebb and flow. I'm sure the pianist is not thinking "I just slowed down so I better speed up soon to give back my stolen time".

Notice also in the Rubinstein example how whenever he slows down or speeds up in the right hand, the left hand completely follows what the right hand is doing so that the notes that are supposed to be played together are played that way.

The only time I think it might be reasonable to keep the LH steady while the RH "rubatos" is when there is some lengthy fioritura passage in the RH. Some examples could be the notorious fioritura passage in the D flat Nocturne, some passages in the Berceuse, or fioritura passages in the Nocturne in F# major. But these types of passages are the exception.