Should you trust the score in polyrythms?

[gn622]

So i noticed that notes in sections containing polyrythms are arranged in unaligned way so that they don't fall directly with a note from the other hand, but should you play it exactly as its written or do you have to decode it musically?

[iansinclair]

Say again?  May I assume that you are talking about true polyrhythms -- say six against four or eight, or such other pleasantries?  Although three against two is a polyrhythm of sorts as well, never mind some of the ornamentation in romantic composers (such as there is a nifty 11 against 3 in the Chopin Opus Posthumous nocturne).

Assuming so, you play it as it is actually written; there is no decoding involved.  Suppose that there are six notes in the right hand in a given measure; you play six evenly spaced notes.  There are four notes in the left hand in the same measure.  You play four evenly spaced notes.  (We are assuming here that all the notes have the same value -- say eighth notes)  Now it will happen that the third left hand note will coincide in time with the fourth left hand note.  If the publisher has printed it that way, wonderful; if not, it doesn't matter.  Simply divide the measure, or section of a measure, into the required number of sub pulses, evenly, in that hand or voice and off you go.

There are occasions when this is almost impossible (that 11 against 3 I mentioned, for instance) and you will need to compromise (for that one, I play 4 + 4 + 3 right hand against 3 even left hand, for instance).

This help?

[iansinclair]

I should add to my previous note -- there is this little matter of accent.  If it is a true polyrhythm, such as 6 against 4, or something of the sort, there is likely to be a definite intended pulse or beat subdivision to the measure; say your 6 against 4 is in a piece in common time (4 beats to the measure).  The sense of beat would normally be present, and should remain present.  A nice (and not obvious) example might be nine against six.  This might be written in three four time -- you would have three beats, each taking three notes of the nice, and two notes of the six.  It might also have the time marker of six eight -- which would conventionally have two beats only.  You would have, quite happily, two beats each of three notes in the six side -- but against that you would have only one beat of nine notes, spanning the entire measure evenly.

Then you have the little matter of the ornaments (the one I was actually thinking of is 8 against 3, and I take it 3 + 3 +2 as to time).  In these, the entire ornament (in this case all 8 notes) is played as one flowing line, with no accentation at all in the length of it.  A bit, perhaps, on the very first note.  The temptation in the cited example is a slight accent in the ornament to match the notes in the left hand (the three).  Don't do it!

[lojay]

Assuming so, you play it as it is actually written; there is no decoding involved.  Suppose that there are six notes in the right hand in a given measure; you play six evenly spaced notes.  There are four notes in the left hand in the same measure.  You play four evenly spaced notes. 

I guess I've been doing it all wrong.

I always assumed how you play the polyrhythm depends on the composer/style.  For composers such as Prokofiev or Stravinsky, I would play as written (if it's 6 vs 5, do 6 vs 5).

With Chopin, unless the rhythm is 2 vs 3 or 3 vs 4, I usually take a lot of liberties (and I thought this was assumed).  I've been studying Chopin's 4th Ballade and there is a section where the left hand is playing arpeggios with 6 notes and the right hand plays 8 notes, 7 notes, 10 notes, etc. I basically subdivided the notes into 3 vs 2 and 4 vs 2 and in tempo it sound great.  Looks like I'm going to have to go back and actually spend time on this. FML

[j_menz]

I might add to what's been said that in some later pieces there's a jazzier approach to rhythm, so one (or more) of the rhythms might be swung. In that case, you can't trust the score much at all.

In Chopin, there is also the issue of rubato, which complicates matters. Usually the main rhythm (usually the bass) keeps strict time and rubato is only applied to the "tune" - which is where you encounter those big odd numbers of notes.

The general principle is that if each rhythm works on it's own, they work together. So use your ears ahead of relying on minute deviations of notes on paper.

[iansinclair]

I might add to what's been said that in some later pieces there's a jazzier approach to rhythm, so one (or more) of the rhythms might be swung. In that case, you can't trust the score much at all.

In Chopin, there is also the issue of rubato, which complicates matters. Usually the main rhythm (usually the bass) keeps strict time and rubato is only applied to the "tune" - which is where you encounter those big odd numbers of notes.

The general principle is that if each rhythm works on it's own, they work together. So use your ears ahead of relying on minute deviations of notes on paper.

Quite so.  The ears are the guide; if it sounds right, artistically, to you, it probably is and I wouldn't get too wound up about having been doing it all wrong!  One does take liberties with Chopin -- particularly with those big odd numbers of notes.

I'd just like to repeat my previous comment, though, and your 4th Ballade is an ideal example.  Be careful of accent.  You want the arpeggios to have a definite pulse to them -- 1 2 3.  But the right hand ornamentation (there really isn't a better word for it) does NOT have a subdivided pulse; just one slight accent on the first note of the group of 8 or 7 or whatever.  The tendency will surely be to do the 8 against 6 as 3 + 3 + 2, and in terms of time that is almost impossible to avoid -- but it will sound "correct" so long as there is no feeling of accent within the 8.

[lojay]

I subdivided the right hand only to figure out a general outline of where I want the notes in the left and right hand to align.  In actual play, I don't think I actually align most of these notes.  I mean, I could have learned it as written, but it would have taken considerably more effort to learn (like seriously 10-15 times longer).

Also, for the most part I don't like to put a slight accent on the first note of the right hand.  I do it maybe for 3-4 of the phrases.  Generally I try to incorporate dynamics like:

p < <  <    <         >    >  > > p

My solution was kind of "cheating", but after I learned the notes, to avoid accenting I practiced the right hand and left hand only in tempo a few times to emphasize the phrasing.  I got like this entire section in a short sitting, but I guess I did cheat.

My current pulse (if you can even call it that) for the left hand is actually 1 - 2 - 3 - 4 - 5 - 6 - etc. corresponding to each note in the left hand.  I'm more worried about bringing out the phrasing (granted the left hand is more or less in perfect time - I think I do place a slight pause in between every 6 notes in the left hand though).  I think it sounds way better and I also really like this contrast, but maybe my thinking is completely wrong (I'm not formally trained so...).  To be completely honest I didn't spend much time on this section, I just basically spent a brief time learning the notes.  I thought I had this section down, look what you guys did!

Should I emphasize the phrasing ultimately creating a contrast or should I emphasize the pulse from the previous sections?  I don't think it's possible to do both.

[nyiregyhazi]

There are occasions when this is almost impossible (that 11 against 3 I mentioned, for instance) and you will need to compromise (for that one, I play 4 + 4 + 3 right hand against 3 even left hand, for instance).

really? Why? While I agree with the rest of your post, I have to go as far as to saw this is simply incorrect, unless I'm misunderstanding you. It's not necessary to subdivide it in any way, if you feel a flow. Turn it into three points of landing together and the whole purpose is defeated. You don't have to get it perfectly metronomic, but the thing that really matters is that if the mathematical version would have no precise moments of togetherness, you absolutely should not have any two notes coinciding precisely together. Beyond that, there is plenty of room for interpretation. The only thing that I'd regard as invalid is turning a clear instruction for freedom into three squarely measured groups with exactly coordinated arrivals. It misses the whole purpose of writing 11 notes. Even if you're perhaps not directly advising such coordinated landings (?) encouraging someone who is inexperienced with these rhythms to subdivide will encourage them to want to coordinate the hands where they need to learn to liberate them from each other.


Am I misunderstanding, or are you actually advising literal coordination of notes to simplify the rhythm? I have to say that I'm overwhelmingly skeptical that could be wise in the Ballade. You can advise that there shouldn't be accents in the middle, but merely to have three coordinated landings is in itself to create an impression of squareness and accent. Only avoiding instances where both hands coincide can you produce a sense of a single group that flows horizontally and without accentuated divisions. To simplify doesn't do the music justice. You need neither do it literally nor cheat in overly simplistic groupings- you merely need voices that are only linked to each other in the moments where they are supposed to land together.

[outin]

I cannot learn anything by just playing both hands separately and then put them together. I need to introduce points where the hands are coordinated. In the cases of these 11 to something, it's more often about getting certain notes not to play together in the hands than trying to align them together. So I do first make a mental draft which notes in the busyer hand go between the notes in the other hand (I might even mark the score). It is clumsy at first, but after I play for a while the draft just disappears and the music takes over. Then it gets more flowing for both hands. This is how my teacher adviced me to do and it has worked fine.

[gn622]

I started this thread because of bar 170 in Chopin's first ballade, in my score, it seems like you have to play b and d right before the g in the left hand, but apparently it is the opposite?

And please forgive me but i got really confused with your posts i know little music theory if any 

[dima_76557]

So i noticed that notes in sections containing polyrythms are arranged in unaligned way so that they don't fall directly with a note from the other hand, but should you play it exactly as its written or do you have to decode it musically?

In simple words: in some music, counting doesn't help much. You just have to have perfect independence of the hands (very underestimated feature in piano playing!) and FEEL how the flow of the music governs the execution of each and both. As soon as you have that, the problems solve themselves.
P.S.: Very often, listening to a YouTube clip by some good artist gives you a lot more than fretting about it yourself. Just saying.
P.S.2: Funny as that sounds, but even in music where all the sounds contact each other (coincide) all the time, you still need perfect independence of the hands to get musically satisfying results!

[j_menz]

bar 170 in Chopin's first ballade

Not one of the scores on IMSLP has bar numbering, and I'm not counting 170 bars out. Can you describe where you mean in some other way?

[lojay]

@j_menz:

The op is talking about the part when the left hand is playing arpeggiated figures much like in the 2nd Scherzo and there is a 5 vs 6 polyrhythm.

I started this thread because of bar 170 in Chopin's first ballade, in my score, it seems like you have to play b and d right before the g in the left hand, but apparently it is the opposite?

Yes, it is the opposite.

[gn622]

https://www.free-scores.com/download-sheet-music.php?pdf=3344

It's in page 7, bar 15.

[johnmar78]

https://www.free-scores.com/download-sheet-music.php?pdf=3344

It's in page 7, bar 15.

Gno, just saw it, its 5/6(LH). remember, DO NOT SCALE...this means you play the first RH notes with first LH, and the rest 4 rh remaining notes has to be played evenly out of rest LH, if not musically, the 6/5 carried out thro more than one bar, just remember in every new bar when you see 6/5. The first rh palyed together with the LH, the rest 4 palyed evenly in between other 4 LH notes...

[iansinclair]

really? Why? While I agree with the rest of your post, I have to go as far as to saw this is simply incorrect, unless I'm misunderstanding you. It's not necessary to subdivide it in any way, if you feel a flow. Turn it into three points of landing together and the whole purpose is defeated. You don't have to get it perfectly metronomic, but the thing that really matters is that if the mathematical version would have no precise moments of togetherness, you absolutely should not have any two notes coinciding precisely together. Beyond that, there is plenty of room for interpretation. The only thing that I'd regard as invalid is turning a clear instruction for freedom into three squarely measured groups with exactly coordinated arrivals. It misses the whole purpose of writing 11 notes. Even if you're perhaps not directly advising such coordinated landings (?) encouraging someone who is inexperienced with these rhythms to subdivide will encourage them to want to coordinate the hands where they need to learn to liberate them from each other.


Am I misunderstanding, or are you actually advising literal coordination of notes to simplify the rhythm? I have to say that I'm overwhelmingly skeptical that could be wise in the Ballade. You can advise that there shouldn't be accents in the middle, but merely to have three coordinated landings is in itself to create an impression of squareness and accent. Only avoiding instances where both hands coincide can you produce a sense of a single group that flows horizontally and without accentuated divisions. To simplify doesn't do the music justice. You need neither do it literally nor cheat in overly simplistic groupings- you merely need voices that are only linked to each other in the moments where they are supposed to land together.

No -- you are right, and I agree with you in principle.  In practice, however, it is almost impossible to not have certain notes coincide.  Note: ALMOST impossible.  Obviously, you try not to.  Dima actually says it best -- one wants the hands to be perfectly independent, and that is the ultimate objective, of course.  However, I'm not good enough to do that consistently...

[j_menz]

https://www.free-scores.com/download-sheet-music.php?pdf=3344

It's in page 7, bar 15.

Thanks.

Apart from the 5/6 polyrhythm here, there's the added complication of the grace note leading to the  next note in the RH. It seems to me that that sounds better if the main RH note lines up with the first note of the sextuplet bass. That means you have to fit six notes effectively against six notes, so the question is why Chopin wrote it the way he did. He clearly intends something other than a straight line-up.

IMO, you need to have a steady pulse in the bass -  even sextuplets, and a feel of 12345 - da DAH in the treble. Not a matter of mathematics at all - a matter of feel. The RH needs to be freeform to give that feel. Try it on its own and make it sound the way you like best, then fit the bass in so that the start of each sextuplet lines up with the first note in the treble and the note after the grace note, and then fit the sixths in evenly.

[nyiregyhazi]

No -- you are right, and I agree with you in principle.  In practice, however, it is almost impossible to not have certain notes coincide.  Note: ALMOST impossible.  Obviously, you try not to.  Dima actually says it best -- one wants the hands to be perfectly independent, and that is the ultimate objective, of course.  However, I'm not good enough to do that consistently...

Really, it shouldn't be difficult. Outin's approach is best, if this is a problem. Don't concern yourself with exact timing but just make damned sure that the hand which has the fewest notes get it's notes somewhere in between the two notes that it should mathematically arrive between. Don't necessarily even concern yourself especially greatly with which note it's mathematically closest to, but just get a feel for what it's coming between. Then it's not truly 100% independent, but you have some form of association between hands that is not in any way associated to an undesirable wish for them to coincide. It doesn't just feel lost, when you fit the longer notes inside a flow of shorter ones.

The only problem with this approach is that it really isn't much good for rhythms where both hands have many notes. It's fine for something like 11 and 3, as long as you take care to try to put a physical flow back into the three (rather than have sense of stop start, as you wait for the next note). However, it won't get you anywhere much with 5/6 in the Ballade. By this point, you have to be able to execute two independent flows- but doing that exercise elsewhere frees a pianist of being limited to wanting to land notes together.  

[nyiregyhazi]

Thanks.

Apart from the 5/6 polyrhythm here, there's the added complication of the grace note leading to the  next note in the RH. It seems to me that that sounds better if the main RH note lines up with the first note of the sextuplet bass. That means you have to fit six notes effectively against six notes, so the question is why Chopin wrote it the way he did. He clearly intends something other than a straight line-up.

I initially did something like this, but I can't agree with the analysis about which note should come with the bass. Grace notes are normally played on the beat in Chopin. He marked this in many students scores of various passages. This can initially seem weird in this passage, compared to what we think we hear from other pianists playing the passage. As listeners, we will never perceive it like that, so it seems odd when we have to observe the grace note as a point of arrival on a beat. However, when executed this way, we have to consider that the listener will not hear as we do, but will still tend to hear the main melody note as if it represents the beat. It's simply delayed in a manner a little like Paderewski's famous right after left playing. In my opinion, this slight ambiguity as to the beat location is the whole idea. Landing the upper note with the bass kills the effect as surely as redistributing a group of 8 to fit with left hand triplets does. The five needs to sound expressively stretched compared to a 6. It's a mistake to tear through the last few notes in order to squeeze the grace note in needlessly quickly, before the next bass.  It would also make very little sense if Chopin really meant to fit 6 notes in the time of 6 but used such a bizarre notation. A more natural interpretation is that it's just a standard group of five plus an arrival at the next bar in which the primary melodic tone is slightly delayed by a typical on-the-beat grace note (that is executed expressively and off-the beat, rather than as a concrete on the beat arrival, after frantically squeezing all the notes into a time limit)

Personally, I never had a problem with the 5/6 but the nasty one is where there are three different rhythmic elements a few bars later, just before it resolves to E flat major. I only recently noticed that I'd be spreading some of the right hand notes out into the wrong place. This is very hard to do on independent flow, as the right hand somehow has to produce two different flows at the same time (while the left does another).

[lojay]

I also do what nyiregyhazi does - I play the grace note on the beat as opposed to what j_menz suggests; however, what j_menz suggests could work.

[nyiregyhazi]

I also do what nyiregyhazi does - I play the grace note on the beat as opposed to what j_menz suggests; however, what j_menz suggests could work.

Are there any particular recordings of people executing it this way? For me, it would simply kill the bel canto of the passage to throw out the feel of an expressive appoggiatura, in favour of three simplistically square landings. It takes all the colour out of the sound and gives a sense of hard accent, even when the player does their utmost to avoid them. I wouldn't say that it would be impossible for a very sensitive pianist to make it work, but personally I'd always be feeling cheated out of the expressive tenuto nature of a true five, plus the rhythmic freedom between bass and arrival note, no matter how well the pianist executed it.

[j_menz]

Are there any particular recordings of people executing it this way?

Horowitz, I believe does. Arrau uses your approach. Zimmerman appears to do something in between (grace note before, main note after, but both only slightly so). Somewhat de gustibus, I suppose, but I do like the Arrau.

I had never intended any more than two "hard landings", the first of which is inevitable anyway. The Zimmerman approach avoids any more entirely, the Arrau needs careful touch to avoid the grace note providing another. The Zimmerman performance doesn't, I think, make the most of the expressiveness of the apoggiatura, but I'm not sure that that's entirely the result of the approach per se.

I understand there is some debate about the extent that Chopin intended the "on the beat" to be a fixed rule, though I claim no immersion in it and certainly don't want to pursue it.

[nyiregyhazi]

Horowitz, I believe does. Arrau uses your approach. Zimmerman appears to do something in between (grace note before, main note after, but both only slightly so). Somewhat de gustibus, I suppose, but I do like the Arrau.

I had never intended any more than two "hard landings", the first of which is inevitable anyway. The Zimmerman approach avoids any more entirely, the Arrau needs careful touch to avoid the grace note providing another. The Zimmerman performance doesn't, I think, make the most of the expressiveness of the apoggiatura, but I'm not sure that that's entirely the result of the approach per se.

I understand there is some debate about the extent that Chopin intended the "on the beat" to be a fixed rule, though I claim no immersion in it and certainly don't want to pursue it.

I had a listen to Horowitz and I think what he does approximates to that- but I don't hear him landing the high note exactly with the bass. With his sensitivity to tone, there's a slight blurring of the exact rhythmic arrival- which prevents the hard square landing. However, I'd never advise this to an amateur pianist- as when they think of landing two notes together they really tend to go all out about landing together in the most literal way. I'd say that success of this still depends on some blurring of the boundaries, rather than truly conceiving a specific alignment of two notes. Also, much as I like Horowitz, I didn't like the loss of the notated expressive stretching. His rushing of the 5 loses that aspect. Horowitz was usually a master of the kind of floating tenuto quality that a proper 5 can create- but instead it's just a blur of wasted notes, here.

By the hard landings, I mean the three moments where the high notes would arrive with the bass, by the way. For a pianist with musicianship, it really shouldn't be difficult to avoid a hard accent on the grace note, if they have any sense of phrase. Clearly the high note is the peak and the true moment of arrival and it would be hard to lose that. That makes it problematic if it lands precisely with the bass, whereas a more gentle appoggiatura with the bass followed by an additional note in freer time will never be in danger of creating the same kind of square and percussive arrival.

I don't usually subscribe to rules, but I've never seen any evidence for a single instance where Chopin asked a student to do grace notes as being before the beat. Given that doing them with the beat is a more expressive practise than squeezing them in before, I'd always favour it unless there was an overwhelmingly good reason to depart from it.

[j_menz]

By the hard landings, I mean the three moments where the high notes would arrive with the bass, by the way.

That's what I took it to mean and meant as well. I take your point about these often being a little "triumphant/grasping on for dear life" in the hands of less experienced pianists - not the effect I meant at all. 

I wonder, after this, whether my conception of "on the beat" may incorporate a certain fluidity. If so, I will need to watch how I speak.

On the matter of Chopin himself, I believe the argument runs that he indicated it on the scores in particular places implying that it was OK to do otherwise where not so annotated. Not sure I buy that, but so be it. 

[nyiregyhazi]

That's what I took it to mean and meant as well. I take your point about these often being a little "triumphant/grasping on for dear life" in the hands of less experienced pianists - not the effect I meant at all.  

I wonder, after this, whether my conception of "on the beat" may incorporate a certain fluidity. If so, I will need to watch how I speak.

Yeah- what I've noticed in teaching is that where I perceive something as being basically together, a student tends to do a whole different level of togetherness that is square and brash. I'm actually doing a mild and unconscious (well, I've since become a little more conscious of these timings after observing what students do asked to align moments) dislocation to smooth it over, much of the time. In order to get students to attempt the effect, they usually have to consciously feel two different events in time, before going on to think of them as essentially being a coordinated to each other (but without necessarily literally landing together). It's actually very closely tied to the manner in which hands never literally meet within the middle of a cross-rhythm.