Collapsing knuckles

[pianoman53]

I've seen many successful pianist let their knuckles collapse, but people here seem to see it as the worst possible thing there ever was, and that people who uses that technique might be possessed by the devil.
Why is that?

[nyiregyhazi]

I've seen many successful pianist let their knuckles collapse, but people here seem to see it as the worst possible thing there ever was, and that people who uses that technique might be possessed by the devil.
Why is that?

It's simple physics. Would you rather someone place a heavy rucksack on your back when squatting/limbo dancing or while being close to upright? The angles demand larger and more complex muscular actions to support any arm pressure at all. A collapsed knuckle that doesn't fall all the way into a palm cluster is usually a very stiff one and the wrist is usually locked too. Some pianists can play this way, but it's very complex and takes great strength and finesse.

Conversely, if you are allowing the knuckles to droop during depression, you are wasting speed and energy. If the distance between knuckle and fingertip is getting smaller due to knuckle collapse, the knuckle is therefore travelling faster than both the fingertip and the key. The largest speed created is not reaching the key, so energy is being wasted, which will not affect the hammer but which will instead be spiralling down towards larger impact at the keybed after sound. Again, this introduces more complexity and a need for more intense muscle contractions (and harder collisions) to achieve the same volume of tone. It is inherently chaotic when something gives way and makes the results overwhelmingly less predictable. However, the answer is not stiffening the knuckles (as nothing but perfectly stiff knuckles can avoid collapse) but movement in a useful direction (ie which creates MORE distance between knuckle and fingertip, thus focusing the largest speeds directly into the key itself, rather than into compressing the hand into a tight position).

I've written a post all about these issues on my blog.

PS. A collapsed knuckle which expands slightly during key movement is not chaotic in the same way as one that gives way during depression. Equally, a knuckle that collapses AFTER depression is less harmful to control than one that collapses during depression (although it will make the next note harder to control if it comes soon after). Don't think that all collapse is fine merely because you sometimes see it in great players. The most successful pianists do things in ways that are only subtly different to watch, but which make a world of difference compared to a pianist who simply never learned how to open their arch properly. Collapse is riddled with limitations unless a pianist is at least fully capable of avoiding it. Katsaris who often lets his arch droop can be seen demonstrating scales to a student on a masterclass with one of the most powerful opened out arches you'll see.

[pianoman53]

It's simple physics. Would you rather someone place a heavy rucksack on your back when squatting/limbo dancing or while being close to upright? The angles demand larger and more complex muscular actions to support any arm pressure at all. A collapsed knuckle that doesn't fall all the way into a palm cluster is usually a very stiff one and the wrist is usually locked too. Some pianists can play this way, but it's very complex and takes great strength and finesse.

Conversely, if you are allowing the knuckles to droop during depression, you are wasting speed and energy. If the distance between knuckle and fingertip is getting smaller due to knuckle collapse, the knuckle is therefore travelling faster than both the fingertip and the key. The largest speed created is not reaching the key, so energy is being wasted, which will not affect the hammer but which will instead be spiralling down towards larger impact at the keybed after sound. Again, this introduces more complexity and a need for more intense muscle contractions (and harder collisions) to achieve the same volume of tone. It is inherently chaotic when something gives way and makes the results overwhelmingly less predictable. However, the answer is not stiffening the knuckles (as nothing but perfectly stiff knuckles can avoid collapse) but movement in a useful direction (ie which creates MORE distance between knuckle and fingertip, thus focusing the largest speeds directly into the key itself, rather than into compressing the hand into a tight position).

I've written a post all about these issues on my blog.

PS. A collapsed knuckle which expands slightly during key movement is not chaotic in the same way as one that gives way during depression. Equally, a knuckle that collapses AFTER depression is less harmful to control than one that collapses during depression (although it will make the next note harder to control if it comes soon after). Don't think that all collapse is fine merely because you sometimes see it in great players. The most successful pianists do things in ways that are only subtly different to watch, but which make a world of difference compared to a pianist who simply never learned how to open their arch properly. Collapse is riddled with limitations unless a pianist is at least fully capable of avoiding it. Katsaris who often lets his arch droop can be seen demonstrating scales to a student on a masterclass with one of the most powerful opened out arches you'll see.

Those "Would you rather" arguments aren't anything I really fall for. To have a heavy bag is nothing like releasing tension in the knuckles.
I also find it much easier to create a slow but strong attack, and especially effective in chords and long melodies, such as Rachmaninov op 16/1.

[falala]

I've seen many successful pianist let their knuckles collapse

really?

I'm not sure I believe you. There are pianists who play with relatively flat fingers that surprise those of us brought up in the very curved Hanonite tradition. But even then, the finger still describes a straight line at the most extreme, and more normally still a very slight curve. I've never seen, live or in video, a professional pianist playing with individual knuckles actually collapsing - ie, bending the other way from the normal curve of the finger, the way the knuckles of some small children do.

In fact I find it almost impossible to make my knuckles do this even if I try. I've often wondered how much of it is a question of training, and how much a developmental thing that children just grow out of as their joints get more stable.

Can you maybe link to any videos of pianists playing this way? I'm intrigued to see whether it actually happens.

[dima_76557]

@ pianoman53

I was given the following easy and logical explanation:
 
Muscles and joints are
- most efficient near the middle of their range of motion
and
- least efficient near the extremes.

Collapse is a negative extreme and is therefore best avoided or kept to a minimum. The opposite extreme (deliberately locking joints to prevent collapse) is also something to be avoided or kept to a minimum.

Does that make sense?

[nyiregyhazi]

Those "Would you rather" arguments aren't anything I really fall for. To have a heavy bag is nothing like releasing tension in the knuckles.
I also find it much easier to create a slow but strong attack, and especially effective in chords and long melodies, such as Rachmaninov op 16/1.

Mechanics is not something you "fall for". It's just objective fact. The closer the angle of a joint is to running parallel to an oncoming force, the lower the force required to stabilise that joint. Would you rather hold a long pole vertically or horizontally against vertical gravity? This is just basic mechanical fact, not metaphor or opinion. Look at the angles in your finger. A drooping knuckle requires larger forces, thanks to the exact same issues of parallel vs perpendicular angles. A drooping knuckle means the situation is more like holding a pole horizontally than vertically. If you dispute this, you should really clarify what possible reasoning could exempt this situation from basic fundaments of newtonian mechanics.

Most importantly, the consequences of these background issues are abundantly visible in practise. Take your finger close to vertical a la rubinstein and tap a table top. Just the faintest lengthening pushes the knuckle back up and stops collapse outright. It's such a small force, many people don't even consciously perceive it. Do the same with an extremely curved finger and the knuckle visibly falls down when you drop on to it unless you work the muscles very hard. You'd have to be suffering rigor mortis not to perceive the muscular work required to stop collapse, in this case. There's no significant difference whether it's the arm bearing down through a finger or a rucksack bearing down upon legs. Aligned joints are easily stabilised with less force. The more you introduce notable angles into the joints, the larger the workload becomes in preventing things dropping further still- with legs and fingers alike.

As I said before, if you're going from collapsed to less collapsed, it doesn't waste speed or energy like when knuckle closes in in fingertip during movement. So it's far from impossible that you can make it work in some situations, if you're going AWAY from a mere starter point rather than actually collapsing as the keys travels. However, you do have to work the harder from these angles, the same as at the bottom of a squat vs standing with only a slight knee bend. The important thing is to ask yourself whether you can control rapid passages with evenness and precision. If so, then play as you wish. If not, I wouldn't invite issues of chaos theory or subject your muscles to the larger forces involved in preventing collapse. The easiest way is to start somewhere non-collapsed in the middle of range and progress to being even less collapsed still. Again, it's not metaphor when I say that it's the same mechanics involved in getting out a squat. The most collapsed position requires the most effort to either get out of or to maintain against oncoming force. Any professor of mechanics will confirm that as absolute fact.

PS you phrase it as releasing tension, but that's a very optimistic viewpoint. When knuckles fall down during depression in loud playing, they send a hell of a lot of useless energy spiralling down towards the keybed. That means you have to then fight against this energy by introducing precisely timed muscular braking in the opposite direction, or keep holding generic tensions to limit movability outright. It's overwhelmingly optimistic to look merely at relaxation in the knuckles rather than at the whole picture of efforts and resulting impacts.

[liszt85]

@Nyir: 1000 words with maybe 100 making sense is nowhere near as effective as 100 words with all 100 making sense. I know this is going to fall on deaf ears but if you would accept the advice, it would make life much more enjoyable for both you and someone listening to you.

[nyiregyhazi]

@Nyir: 1000 words with maybe 100 making sense is nowhere near as effective as 100 words with all 100 making sense. I know this is going to fall on deaf ears but if you would accept the advice, it would make life much more enjoyable for both you and someone listening to you.

Well, as the OP wasn't willing to accept the shorter comparison to a squat I've put the more detailed version out there for consideration. If the OP just wasn't interested in the actual answer to the question he posed that's up to him. But to give the short version, the answer is as simple as the reason why it's harder to squat than to be standing and why it's harder to hold a long pole horizontally than vertically. The explanation is the same in each. if someone isn't prepared to listen to the facts at play, that's up to them. If they are, my post contains both the most basic relevant issues demonstrated in a practical form and supporting reasoning.

[pianoman53]

Well, as the OP wasn't willing to accept the shorter comparison to a squat I've put the more detailed version out there for consideration. If the OP just wasn't interested in the actual answer to the question he posed that's up to him. But to give the short version, the answer is as simple as the reason why it's harder to squat than to be standing and why it's harder to hold a long pole horizontally than vertically. The explanation is the same in each. if someone isn't prepared to listen to the facts at play, that's up to them. If they are, my post contains both the most basic relevant issues demonstrated in a practical form and supporting reasoning.

I never said I didn't accept your so called shorter version. All I said was that arguments such as "Would you rather do this than that?" is not very convincing for me. Dima explained it, I understand and I will see if it changes my playing. My current teacher, who is price winner of the Tchaikowsky, very often plays with somewhat collapsed knuckles. Another teacher from Russia, that I played for a few years ago, used the same technique. Their wrist is slightly higher, but the hand gets as close to relaxed as I can imagine.

[Bob]

Who?  

I haven't seen anyone, and it is seen as something evil.

Joints, not knuckles.

Arch = strong, controlled.
broken/fallen joint, less control, less strength.

[pianoman53]

yes, the knuckles. That part that connects the fingers with the hand. Knuckles, no? Obviously it's not completely dead, but it's really not an arch.

[nyiregyhazi]

I never said I didn't accept your so called shorter version. All I said was that arguments such as "Would you rather do this than that?" is not very convincing for me. Dima explained it, I understand and I will see if it changes my playing. My current teacher, who is price winner of the Tchaikowsky, very often plays with somewhat collapsed knuckles. Another teacher from Russia, that I played for a few years ago, used the same technique. Their wrist is slightly higher, but the hand gets as close to relaxed as I can imagine.

That's why I took the time to go beyond the "would you rather" argument and illustrate the mechanical truth that is common between the situations. People can either bury their heads in the sand and hope to transcend mechanics, or they stop to consider these objective reasons why some positions demand larger forces and efforts than others. I'm not telling anyone to read my posts. I'm giving them an option of reading the objective issues behind the question, plus practical illustrations. If they're not interested in the actual answer to the question, they can can feel free to ignore it.

I had a teacher who played with lowered knuckles much of the time myself. I didn't succeed when I was attempting (as opposed to doing) what he did though. I was dreadful at controlling simple finger work with precision. I needed to learn to open my arch properly and stop drooping during key motion. That's why I took the time to clarify why collapsING knuckles are a very different issue to knuckle that merely start collapsED but act in the very opposite direction. If their's seems relaxed that's because they're so good at generating useful movement. If it weren't, it would be tensing like crazy, the same as when less capable pianists get this difficult technique wrong. People who think collapsING knuckles are fine invariably fail. I watched a film of an Edna Golandsky student on youtube who was struggling horribly to avoid constant holes in their phrases due to this poor habit.

With respect to dima_ogorodnikov, her version is more of a soundbite/call for trust than an explanation, as such. It's fine as long as someone is willing to trust it, but it doesn't really explain what the "middle" actually is or WHY it might be more efficient to avoid extremes. It just asks you to trust what is a highly subjective interpretation of efficiency. I'd add to that version that the middle is not a singular point but a huge range. Also, if anything, the general tendency is that the more collapsed you start, the harder you must work the muscles to achieve a result. The only exception comes if you've already gone so far from collapse that you have used up all of your scope to move in the finger before putting the key down. When there's no movement left, obviously you can't do much but collapse. For that reason, you generally want to leave room for at least a trace of movement in the hand, before starting to move the key.

As long as I'm not actively straining against my range of movement in the ligaments, my personal feeling is that you get extremely good efficency indeed by coming very close to the limits of movement. Look in Alan Fraser's books and look at rubinstein, nyiregyhazi, Curzon, Argerich and many others and you will see movement being taken to the absolute full. The fingers fully lengthen out in the small joints and the finger goes very close to vertical. Although I actually agree in many  ways with Dima's idea for many areas in playing, i don't either agree that's it's most efficient to avoid using BOTH outer ranges of movement or that the sweet spot is necessarily anywhere close to the middle of the range. I'd merely agree that starting from an extreme collapsed position is usually very hard work for the muscles and that you shouldn't be straining forcefully against your ligaments at the other extreme end of motion.

[nyiregyhazi]

See the above film for the harm of collapse in an otherwise very talented youngster. Superficially, he does plenty of arch stuff, but it starts up and collapses down as the key moves, because there's not enough movement in the hand. So despite many qualities elsewhere, there are amateurish holes in his sound. It's literally an issue of chaos theory. When you collapse, a miniscule change in one variable has a big effect on another (here the level of sound). notice the poorly articulated opening triplets and the chromatic runs early on in the right. CollapsING knuckles are ruinous. He's completely disconnected between fingers and arm and the impact on his control of sound is overwhelming- in phrases and passage work alike. Note how inconsistent the note lengths are due to the lack of connect between hand and arm caused by the constant drooping. Really not a good advert for passive fingers or drooping the knuckles- rather than knowing how to actively open them outwards through key depression.

[pianoman53]

Well, that sort of argument aren't very convincing either, I'm afraid. Dima makes his point clear, and that's it. Your examples are simply of people who didn't make it with that technique. Since you were one of them, you seem to be very against it, instead of seeing that it works for some, but not all.
Anyway, I got the reasons why it might not be a great idea for everyone. Thanks all!

[nyiregyhazi]

Well, that sort of argument aren't very convincing either, I'm afraid. Dima makes his point clear, and that's it. Your examples are simply of people who didn't make it with that technique. Since you were one of them, you seem to be very against it, instead of seeing that it works for some, but not all.
Anyway, I got the reasons why it might not be a great idea for everyone. Thanks all!

Who does it work for? Numerous posters asked for examples. Seeing as you've alluded to success stories on numerous occasions, you really ought to say who you're referring to- so we can look for ourselves how they move. As I said, the primary issue in mechanics is whether they go FROM a collapsed position towards a less collapsed one or INTO an increasingly collapsed position while the keys are going down. You should cite your success stories so we can look at what they are actually doing. Many falsely claim Horowitz collapsed, but he never let his knuckles droop while he moved keys. He just played from a slightly flatter starter position(that was neither collapsed in or collapsing).

Also I didn't say I'm against it outright. I said that due to the fact that it demands vastly greater physical effort due to mechanical issues, it would be self-destructive to see it as a good way to play- unless difficult repertoire is already being executed with true ease. Anyone who doesn't play at least moderately difficult repertoire with easy control right now is actively encouraging that situation to remain in status quo when they sanction collapse. Even for those who manage it well, it's still a lot harder on the muscles (just like waddling around in a permanent squat works the muscles harder than getting yourself stood up). A lot of mediocre concert pianists can be seen to collapse a little, but I'd like to know who the truly great ones are that supposedly do it often.

[pianoman53]

There is one, for example. It's not like he stays in the "collapsed state" for very long at once, but he clearly does use it. My teacher doesn't have any videos on youtube, and I wouldn't really like to show you that anyway.

Here is another one. She, just like me, uses it in chords. It seem to work for her, no?

[nyiregyhazi]

There is one, for example. It's not like he stays in the "collapsed state" for very long at once, but he clearly does use it. My teacher doesn't have any videos on youtube, and I wouldn't really like to show you that anyway.

Here is another one. She, just like me, uses it in chords. It seem to work for her, no?

? Collapsed? It's extremely mild and extremely rare in the bits of the trio I just looked at. The vast majority of the time his arch is clearly expanding. It doesn't often start collapsed as such, but very clearly in the middle of the range. As his arch is usually expanding very obviously rather than diminishing in space, I wouldn't call that either collapsed or collapsing more than a fraction of the time. The few occasions where he flops down are on softer single notes, where they are least likely to cause impact.

In the film of the woman I didn't see any places where the arch sunk further as the keys went down. If her arch were actually collapsING (rather than merely starting from a mildly sunk starter position) it wouldn't work at all. I find it overwhelmingly more easy to start with an open hand, but with female pianists with smaller hands that's not always possible in big chords. She manages to work well from the position, but I suspect that she's done plenty of work on at least trying to open her hand for that to work as it does- with no additional collapse during movement. You need a lot of strength in the finger action to avoid droop and impact like that. The problem is that when you just say collapse is fine, you're no longer trying to ensure the right movement are happening, so very few people get those results unless they are used to trying to avoid collapse completely. It's just a mildly sunk starter position followed by no collapse.

[dima_76557]

With respect to dima_ogorodnikov, her version is more of a soundbite/call for trust than an explanation, as such. It's fine as long as someone is willing to trust it, but it doesn't really explain what the "middle" actually is or WHY it might be more efficient to avoid extremes. It just asks you to trust what is a highly subjective interpretation of efficiency.

It is not merely subjective though. There is research in the field of functional anatomy that backs up my quoted explanation. If you are really interested, I could try and locate some documents and/or online sources in the field when I have the time. In the case of both collapsING and collapsEd KNUCKLES (at the hand bridge), for example, the extensors of the fingers and the palmar interossei muscles are put near their extreme range of motion, which cannot really be a healthy thing if it is the default way of playing the instrument.
P.S.: The "middle" is a hand hanging down along the side. It is slightly "cupped", even when "relaxed" because of the natural muscle tonus.

Not so scientific, but still very interesting for the average layman is this Swedish article about piano ergonomics: https://www.artist-musikerhalsan.se/en/musician-ergonomics/35-piano-ergonomics

If we are not very interested in anatomy and ergonomics, then it is good to know that collapsing or not collapsing the structure has consequences for tone too as Alan Fraser shows in his many lectures and student practice sessions on YouTube:

videos?flow=grid&view=0

[outin]

Obviously for people with smaller hands the knuckles will be collapsed for big chords and even octaves. It will just cause harmful tension to try to avoid that (I used to try). But joints collapsing because of the finger movement is not quite the same thing.

[pianoman53]

So what is it Ilja Scheps is doing then? Why isn't that collapsing? It takes like 1 minute, and then I've seen it 5 times or so.


But never mind, I got the answer that I wanted. No need to discuss it anymore, Thank you.

[dima_76557]

So what is it Ilja Scheps is doing then? Why isn't that collapsing? It takes like 1 minute, and then I've seen it 5 times or so.

The keyword is "function". Both Ilja and his daughter Olga (in the next clip) keep their fingers moveable within inevitable extended positions (dictated by their handsize). That is also basically the purpose of healthy "stretching" exercises: you don't force a larger stretch through power exercises to make your hand bigger (such exercises actually DECREASE your reach!), but you find "relaxation", "moveability", comfort, freedom within a position of extension. As soon as you find that freedom within that extension, your hands grow so to speak and you may even be able to avoid "collapse" where it used to be inevitable.
P.S.: As we can see from your clips, both Ilja and Olga go back to "normal" again as soon as the music allows for it.

[anima55]

As I see it, there could be two reasons for playing with collapsed knuckles.

1.  The knuckles collapse because the player has insufficient control over their movement
2.  The knuckles are collapsed deliberately for a particular reason

Obviously, collapsing knuckles caused by lack of control is not to be admired in anyone's technique.  If some choose to collapse the knuckles believing it to be helpful in acquiring a certain sound, touch or whatever, then so be it.  Personally speaking though, I can't really see why collapsing the knuckles would be helpful in any situation so I would avoid it.

[nyiregyhazi]

It is not merely subjective though. There is research in the field of functional anatomy that backs up my quoted explanation. If you are really interested, I could try and locate some documents and/or online sources in the field when I have the time. In the case of both collapsING and collapsEd KNUCKLES (at the hand bridge), for example, the extensors of the fingers and the palmar interossei muscles are put near their extreme range of motion, which cannot really be a healthy thing if it is the default way of playing the instrument.
P.S.: The "middle" is a hand hanging down along the side. It is slightly "cupped", even when "relaxed" because of the natural muscle tonus.

Not so scientific, but still very interesting for the average layman is this Swedish article about piano ergonomics: https://www.artist-musikerhalsan.se/en/musician-ergonomics/35-piano-ergonomics

If we are not very interested in anatomy and ergonomics, then it is good to know that collapsing or not collapsing the structure has consequences for tone too as Alan Fraser shows in his many lectures and student practice sessions on YouTube:

videos?flow=grid&view=0

Ah, fair enough. I just couldn't agree that is a direct explanation for the particular question though - given that collapse can be horrifically damaging within the middle of the range of motion (regularly seen within that golandsky pupil) and that movements can be very successful in numerous great artists at the extreme limits of a lengthening finger. For me, the primary issue is the same mechanism by which it's easier to hold a pole upright than horizontally, plus the issue of how speed is not passed on if collapse occurs mid descent (instead  sending useless energy at the keybed).

It does make sense that joints would have areas where they operate best, but I don't think think this goes very far in explaining either the numerous success stories that are nowhere near the middle of range or in explaining disasters in the midrange. For people with problems, I always teach extreme finger stands. That way, they naturally exaggerate and become equally capable of settling into a midrange too. . When  aiming for the middle, as soon as they stop concentrating they tend to lose the movement again and go back to sagging.

[nyiregyhazi]

The keyword is "function". Both Ilja and his daughter Olga (in the next clip) keep their fingers moveable within inevitable extended positions (dictated by their handsize). That is also basically the purpose of healthy "stretching" exercises: you don't force a larger stretch through power exercises to make your hand bigger (such exercises actually DECREASE your reach!), but you find "relaxation", "moveability", comfort, freedom within a position of extension. As soon as you find that freedom within that extension, your hands grow so to speak and you may even be able to avoid "collapse" where it used to be inevitable.
P.S.: As we can see from your clips, both Ilja and Olga go back to "normal" again as soon as the music allows for it.

I agree but I think it's important to add that the movability is into trying to expand out, not into being squashed into further collapse. What amateurs with limp floppy hands get wrong is to allow further collapse by being too passive. The good players move in the opposite direction to collapse enough not to collapse any further. If the first pianist in the films collapsed with any really notable regularity, his hand would be bobbing in the fast notes. Instead, the fingers are always growing out from the knuckles, except in less than 1 percent of cases. It's not just movability but the instinct for movement in the productive direction, as opposed to passive movement in the collapse direction.

[falala]

There is one, for example. It's not like he stays in the "collapsed state" for very long at once, but he clearly does use it.

Ah yes I see what you mean. OK.

Dude certainly sits a long way from the piano, too.

https://www.youtube.com/watch?v=7MWuy5ioEJU

Here is another one. She, just like me, uses it in chords. It seem to work for her, no?

I can't really see it there. Can you specify a particular place in the video?