Sight Reading- Not stopping to correct mistakes?

[nickadams]

Every sight reading advice I read recommends not to stop or slow the tempo when one makes mistakes- "just keep going" they always say. But whenever I make a mistake, I can't "just keep going" because I end up messing up the entire rest of the piece. Like if my mistake is going up a 3rd when I should have gone up a 4th then if I "just keep going" then every note I play from then on will be one note too high.

I don't get how you are supposed to repair your mistake without stopping/slowing down? My brain just doesn't work that fast 


Any advice?

[nyiregyhazi]

Every sight reading advice I read recommends not to stop or slow the tempo when one makes mistakes- "just keep going" they always say. But whenever I make a mistake, I can't "just keep going" because I end up messing up the entire rest of the piece. Like if my mistake is going up a 3rd when I should have gone up a 4th then if I "just keep going" then every note I play from then on will be one note too high.

I don't get how you are supposed to repair your mistake without stopping/slowing down? My brain just doesn't work that fast  


Any advice?

Practise pieces slowly with one finger (including with your eyes closed sometimes). You need to improve your perception of the keyboard- independently of physical movement. Also, try saying every letter out loud. Intervallic reading is good, but you must also be able to perceive every note as a letter without a moments thought (and associate it to the corresponding key on the piano). It's not enough to only think of it in terms of how one finger leads to the next. You need a more abstracted sense of both what keys you are playing and where they are in relation to you- not merely where would have been in relation to the previous finger (had it been played correctly).

[keypeg]

The idea of sight reading should be qualified more.  For one thing, a person's particular goals for sight reading should be thought through.  There is this generic thing which is meant for somebody who will accompany soloists and has to be able to haul out any score, look at it, and play something instantly that the soloist or a choir can play/sing with.  But as students is that always what we are aiming for?

As an experiment that might not go anywhere - what kinds of things do you think you are trying to achieve - or how do you see sight reading for yourself right now?

[j_menz]

Practise pieces slowly with one finger (including with your eyes closed sometimes). You need to improve your perception of the keyboard- independently of physical movement. Also, try saying every letter out loud. Intervallic reading is good, but you must also be able to perceive every note as a letter without a moments thought (and associate it to the corresponding key on the piano). It's not enough to only think of it in terms of how one finger leads to the next. You need a more abstracted sense of both what keys you are playing and where they are in relation to you- not merely where would have been in relation to the previous finger (had it been played correctly).

I generally agree with this. The only exception I would take is that you should associate a written note with a sound and a key on the keyboard directly, not via the mechanism of associating it with a letter.

[nyiregyhazi]

I generally agree with this. The only exception I would take is that you should associate a written note with a sound and a key on the keyboard directly, not via the mechanism of associating it with a letter.

I wouldn't disagree with the principle- but I think association with the letter serves to aid the transition to a more direct process of association. If you know the keys well by letter and the notes (on the page) well by letter, it effectively becomes the same thing. Saying the letter is not about a letter, but about triggering the direct association to the piano key which has that letter  (guaranteeing that no mental step has been skipped in favour of a mental shortcut). If the poster struggles not to notice when everything becomes a note too high, say, it suggests insufficient familiarity and excessive dependence on intervals. The easiest way to trigger a stronger association between notation and the piano key is to say the letter- not for the sake of saying the letter but for the sake of causing direct thought about the key which is linked to that letter and written note.

In short, I think it's too easy to be mentally lazy when only associating to the key- and to fail to process key information. As soon as you say letters, no information is failing to be processed- which can actually make the mental linking all the more strong. It's basically a way to check that adequate thought is involved- which allows the associations to start becoming flawlessly hard-wired, rather than vaguely present. Even the most advanced players can learn from this. Can you say every letter in turn throughout a Chopin study (from memory- without either reading the score or playing at a keyboard)? It's perhaps the ultimate test of whether mental visualisation of the notes is truly vivid, or merely some of the way there. If I do the same without saying the letters of imagined notes, my brain immediately gets lazy and stops visualising to the same degree of precision and certainty. Holes slips by. However, say the letters and no hole goes unnoticed.

[j_menz]

Can you say every letter in turn throughout a Chopin study (from memory- without either reading the score or playing at a keyboard)? It's perhaps the ultimate test of whether mental visualisation of the notes is truly vivid, or merely some of the way there.

I can't. Though that would appear to have everything to do with memorisation, and nothing to do with sight reading. For sight reading, I associate the notes on the score with positions on the keyboard, and most importantly with sounds. The thinking about the letters would slow me way down, though may be a useful interim step to learn to read more fluently for the OP.

[hfmadopter]

I'm about in the same boat as you on sight reading, in fact maybe worse as it takes a lot of time to learn a piece to begin with, then I can sight read it. I've found it far better that once learned to go on and memorize the piece ( it's just a throw back to how I was taught), once this is done I can play from memory and it's always smoother.

Of course that does no good, if as mentioned, you need to read the music instantly. I also dislike reading and playing at the same time, I feel free playing from memory and that's when I can start to get full expression into the piece.

Additionally, I agree with learning the notes by sight both from the keyboard to the music and vice versa. Staying focused is important, I lose focus far too often, once I start to get a piece working mistakes creep in as I maybe daydream a bit.

If sight reading is important to you there must be studies on it. I think I saw a study book offered about this in the Gold Membership section.

[timothy42b]

Every sight reading advice I read recommends not to stop or slow the tempo when one makes mistakes- "just keep going" they always say. But whenever I make a mistake, I can't "just keep going" because I end up messing up the entire rest of the piece. Like if my mistake is going up a 3rd when I should have gone up a 4th then if I "just keep going" then every note I play from then on will be one note too high.

You're at the "movement creation" stage of sight reading and you want to eventually get to the "pattern retrieval" mode that the fluent readers use.  Mostly. 

But for now.

There is an easy answer to your immediate problem.  Works for some, not for others.  Try it and see.  (It may not work if you are trying to sightread pieces too difficult for your current level, but it probably will work if your keyboard geometry is deficient, as other posters have suggested). 

You must NOT stop and you MUST not slow down, BUT you don't have to proceed. 

Keep the pulse absolutely steady (use a metronome) and repeat the mistake as many times a s necessary.  Say you're playing along quarter notes in 4/4 and you hit a wrong note or chord.  You hit the F instead of the E, in the example you mentioned.  Play that F four times in strict rhythm while you prepare to correct it, play the E correctly on the right beat and move on.  Don't let ANYTHING interrupt the flow or slow the beat. 

[roseamelia]

That is what my teacher said is to keep going to stop mistakes.

[nyiregyhazi]

I can't. Though that would appear to have everything to do with memorisation, and nothing to do with sight reading. For sight reading, I associate the notes on the score with positions on the keyboard, and most importantly with sounds. The thinking about the letters would slow me way down, though may be a useful interim step to learn to read more fluently for the OP.

Yeah- the Chopin study example is nothing do with sight-reading. It's just an example of how much better the observational senses are forced to engage, when you say every individual letter. There's nothing wrong with associating more directly (which is obviously the ultimate goal)- but coming back to letters is a way of ensuring that direct associations are necessarily being made, rather than frequently missed due to sloppy thinking.

[j_menz]

a way of ensuring that direct associations are necessarily being made

I'd have thought that hitting the right notes was all the proof one needed here.

[nyiregyhazi]

You're at the "movement creation" stage of sight reading and you want to eventually get to the "pattern retrieval" mode that the fluent readers use.  Mostly. 

But for now.

There is an easy answer to your immediate problem.  Works for some, not for others.  Try it and see.  (It may not work if you are trying to sightread pieces too difficult for your current level, but it probably will work if your keyboard geometry is deficient, as other posters have suggested). 

You must NOT stop and you MUST not slow down, BUT you don't have to proceed. 

Keep the pulse absolutely steady (use a metronome) and repeat the mistake as many times a s necessary.  Say you're playing along quarter notes in 4/4 and you hit a wrong note or chord.  You hit the F instead of the E, in the example you mentioned.  Play that F four times in strict rhythm while you prepare to correct it, play the E correctly on the right beat and move on.  Don't let ANYTHING interrupt the flow or slow the beat. 

I agree in many ways- but I don't believe in drawing a final an ultimate rule. One of the most effective things a poor sightreader can do is to switch between contrasting modes. One useful mode is to forget rhythm altogether- but never to rush. Every note is to be played long and every physical connection is to be felt most acutely before the key is even sounded. Often, you might tap a key without striking it (or releasing the previous key). You're feeling your way around and ensuring that there is 100% to every note before you play it. The sensory feedback that comes from this is most acute and will carry into a return to rhythmic playing.

The problem with being adamant about always doing things in tight rhythm is that in fast tempos it will regularly cause panic and wild approximation. In slower playing, the arm will often tend to be "stab" too much at beats, rather than maintain a steady flow of simply movement. There are many ways in which it completely interferes with the sensory foundations that are required to play in rhythm with flow and comfortable confidence. Sometimes stepping outside of your ultimate goal is the most important way of all to find what is missing. Once you have an acute feel for physical connection of intervals, you can return to rhythm with confidence and assurance.

[nyiregyhazi]

I'd have thought that hitting the right notes was all the proof one needed here.

Not at all. Can you play any piece you know with assurance using just one finger? Perhaps, but you'd be among few if you can. Physical memory is frequently developed despite omission of key information, that merely falls by the wayside. Are the notes being found by feel for interval or by thought of absolute pitch- or by both? Note what he said about playing one note wrong and then all the others being one too high, say. Most people get too lost in a singular mindset. An advanced reader is equally proficient in both. It's not a matter of EITHER reading an absolute pitch OR finding an interval. For an advanced player, each serves to doubly confirm the other. If you go too much by interval, there's a possibility you're associating nothing but a distance between two fingers. That's why the letters are so valuable. They force you to build upon the skill to read individual notes as absolutes. It's literally impossible to overlook this absolute side of things, if you have to say the letters. It's not about the letters but the thinking that it necessarily forces to occur.

In short, a player with good absolute reading skills and keyboard geography is never thrown by getting an interval wrong. Alarm bells instantly ring and absolute reading instigates an instant correction. If it carries on going wrong,l the player is too lost in intervallic reading that is not backed up by adequate absolute reading.

[j_menz]

Not at all. Can you play any piece you know with assurance using just one finger? Perhaps, but you'd be among few if you can. Physical memory is frequently developed despite omission of key information, that merely falls by the wayside. Are the notes being found by feel for interval or by thought of absolute pitch- or by both? Note what he said about playing one note wrong and then all the others being one too high, say. Most people get too lost in a singular mindset. An advanced reader is equally proficient in both. It's not a matter of EITHER reading an absolute pitch OR finding an interval. For an advanced player, each serves to doubly confirm the other. If you go too much by interval, there's a possibility you're associating nothing but a distance between two fingers. That's why the letters are so valuable. They force you to build upon the skill to read individual notes as absolutes.

I agree that the OP is only reading intervals, not absolute notes and that a combination of the two is required for proficient reading.  Anything that helps achieve this is useful.

I have no idea what you mean by playing something with one finger? How do I play a fugue that way, or a chord even? And what value would it have in any case?  I only play by reading, and don't memorize btw.

[nyiregyhazi]

I agree that the OP is only reading intervals, not absolute notes and that a combination of the two is required for proficient reading.  Anything that helps achieve this is useful.

I have no idea what you mean by playing something with one finger? How do I play a fugue that way, or a chord even? And what value would it have in any case?  I only play by reading, and don't memorize btw.

Try it for a Chopin study. See how much more acutely you perceive the intervals, compared to when you simply slip between one finger and another finger. It's particularly valuable for memorising, as it shows you actually know the construction, rather than merely what it feels like to perform a movement. It develops true understanding of the music- not merely feel for movements.

Even in chords there is value to this. Do you just feel how to space out the fingers, or do you truly perceive the chord as a single mentally grasped entity? A pianist who really knows the music should be able to describe every note in any chord, without recourse to physical memory. I'm not claiming that I can- but when you can't you have learned mere movements and not the true musical construction. Even when reading from the score- are you merely visualising the movements that are required, or are you visualising the fine details within the actual musical make-up? Again, I have no doubt that all good sight readers do both. They can either abstract the notes from the page at sight, or they can imagine the physical means of executing them. They are not lost in just one option.

I could sight read various pieces flawlessly at first sight (note wise), without perceiving even half of the intervals mentally. My hands would play the right notes, but I wouldn't notice every individual interval in each harmony. Playing notes accurately never suggests that everything there is for a good reader to perceive has been perceived.

[j_menz]

Try it for a Chopin study. See how much more acutely you perceive the intervals, compared to when you simply slip between one finger and another finger. It's particularly valuable for memorising, as it shows you actually know the construction, rather than merely what it feels like to perform a movement. It develops true understanding of the music- not merely feel for movements.

Even in chords there is value to this. Do you just feel how to space out the fingers, or do you truly perceive the chord as a single mentally grasped entity? A pianist who really knows the music should be able to describe every note in any chord, without recourse to physical memory. I'm not claiming that I can- but when you can't you have learned mere movements and not the true musical construction. Even when reading from the score- are you merely visualising the movements that are required, or are you visualising the fine details within the actual musical make-up? Again, I have no doubt that all good sight readers do both. They can either abstract the notes from the page at sight, or they can imagine the physical means of executing them. They are not lost in just one option.

I could sight read various pieces flawlessly at first sight (note wise), without perceiving even half of the intervals mentally. My hands would play the right notes, but I wouldn't notice every individual interval in each harmony. Playing notes accurately never suggests that everything there is for a good reader to perceive has been perceived.

I find it odd that you have not mentioned the sound in any of this. When I play (unless I am working on some technical difficulty) what I am conscious of, indeed the only thing I am conscious of,  is the sound, the music.

I don't believe that my understanding of the music is in any way enhanced by being able to label the various components of it. It's as if you are suggesting my appreciation of Shakespeare was simply a matter of being able to spell the words.

[nyiregyhazi]

I find it odd that you have not mentioned the sound in any of this. When I play (unless I am working on some technical difficulty) what I am conscious of, indeed the only thing I am conscious of,  is the sound, the music.

I don't believe that my understanding of the music is in any way enhanced by being able to label the various components of it. It's as if you are suggesting my appreciation of Shakespeare was simply a matter of being able to spell the words.

The sound just goes without saying, for me. Would you practise this on a silent keyboard? Of course the idea is to associate to sound. What I'm talking about is other important associations that are less obvious and which frequently get overlooked. Who forgets that music is sound?

To be honest, if you're not memorising it's a big issue- with regard to these techniques. Would you say an actor truly knows a play if he always has to have the page in front of him to remember any of his lines? I come from a similar position. There are pieces which I could learn in minutes, if I can keep the score. However, take it away and I simply don't know the construction enough to play by memory. It might take at least a week to get it there- even for something easy. What my eyes see goes straight to my fingers with little effort. However, there's an abundance of information that I need to force myself to perceive, if I'm to go beyond simply being able to play the notes from the page and actually know them instead. To go beyond simply being able to follow instructions when they are present, it's important to have as many tools as possible.

Even if you don't want to play by memory, there is much value in bringing your holes in perception to the fore-front- in order to realise how much can be missed, despite reading off a score with perfectly good accuracy. When you have to memorise something (and check the quality of memory with visualisation and one finger practise from memory etc), fewer of these issues can slip by unnoticed.

[j_menz]

N: We clearly have different understandings/approaches to what it means to comprehend a piece of music, which may well be an interesting discussion for another place.

We do, however, appear to agree generally re the approach the OP should take regarding his difficulties, and perhaps should leave this thread for others who wish to contribute on that matter.

[p2u_]

I know a 12-year old girl here in Moscow who can really sight-read most anything at reasonable tempo without mistakes, even on a silent keyboard (!). She says she does NOT "read" the notes. She does NOT realize their names, she does NOT realize the intervals. She just "translates" the signs on the paper directly into 1) positions on the keyboard and 2) instructions to move (choice of fingers and hand positions). She says aural feedback (hearing the sound result) actually tends to keep her back. Reading ahead (planning action) is paramount.
P.S.: I myself am a lousy sight-reader, but that is merely a psychological problem. All skills you would expect for this to work are more than up to par.

Paul

[timothy42b]

I know a 12-year old girl here in Moscow who can really sight-read most anything at reasonable tempo without mistakes, even on a silent keyboard (!). She says she does NOT "read" the notes. She does NOT realize their names, she does NOT realize the intervals. She just "translates" the signs on the paper directly into 1) positions on the keyboard and 2) instructions to move (choice of fingers and hand positions).

The correspondence between what she thinks she does and what she really does may approach zero, particularly for a 12 year old. 

Impressive skills though. 

[p2u_]

The correspondence between what she thinks she does and what she really does may approach zero, particularly for a 12 year old.

I've seen her do it myself with books she had just brought from the library...

Impressive skills though.

She's a prodigy, of course; virtually self-taught. I like learning from those. When I told her I had trouble with it, she was very surprised. "Too busy trying to look good, huh?" That was her razor sharp analysis. And you know, in my case she's probably right. She also told me she learned her skills on the sofa, not near the instrument. What surprised me most was her opinion that it has absolutely nothing to do with the ears. "Ears", she says, "are to check if you played correctly, which is stupid, because it's already too late."  

Paul

[timothy42b]

I've seen her do it myself with books she had just brought from the library...

I'm not doubting she does it, I'm doubting she knows for sure how she does it.

She just "translates" the signs on the paper directly into 1) positions on the keyboard and 2) instructions to move (choice of fingers and hand positions).

I'll buy 1) but doubt 2). 

[fleetfingers]

The idea about letter names for the notes is simply to label them. When I think of a C, for example, I do not think about the letter C as it functions in the alphabet. I picture the note on the piano that sits to the left of the two black keys and has a straight left side. And whatever finger I choose can directly find its way there, without looking. I could call it by any name, it doesn't matter. Using letters makes sense, though, because everyone already knows the order.

I love what the 12-year-old girl said about the ears! Someone with perfect pitch perhaps can associate each note with a sound. I do not, but I can hear in my head how the notes will sound generally when played. For example, I am familiar with the sounds of the intervals, the chords, the arpeggios, etc. So, in that way I believe we can use our ears when sightreading . . . but, as the prodigy said, only to check.

[fleetfingers]

I don't get how you are supposed to repair your mistake without stopping/slowing down? My brain just doesn't work that fast  

The whole idea behind not stopping/slowing down is that you should NOT fix mistakes, but keep playing in time despite them. One of the reasons it makes you a better sight reader is that it trains you to look ahead and be ready for what's coming next, instead of getting stuck somewhere and losing the flow. The skill you are supposed to develop is to stay on track, even when you slip up or miss some notes.

BTW, I had a student who was TOO well trained with this approach, so that she was in the habit of making a lot of mistakes. I asked her to forget about the time and to pause whenever she needed to get her fingers in the right place before she played the note(s). She needed to take her time and focus on accuracy. Remember that there are different practice methods to achieve different ends.

[goldentone]

I believe the conceptual path in becoming a good sightreader is to sightread at the tempo at which you can reasonably accurately play the piece.  And to do that, for young people at least, you must sightread pieces at the right level.  I think there's a balance in that dictum and not worrying about mistakes, because by practicing you'll improve.  

[nyiregyhazi]

The idea about letter names for the notes is simply to label them. When I think of a C, for example, I do not think about the letter C as it functions in the alphabet. I picture the note on the piano that sits to the left of the two black keys and has a straight left side. And whatever finger I choose can directly find its way there, without looking. I could call it by any name, it doesn't matter. Using letters makes sense, though, because everyone already knows the order.

Although this is all true, you're simply missing the point of what I'm saying. Yes, you can do it with any form of labelling or simply with none- merely by knowing which key is signified by which symbol. However, the point about letters is that they reinforce associations and force the conscious mind to observe considerably more points of view- which serve to back each other up and make the mental picturing all the more clear.

Not long ago, I didn't think about letters in the slightest. However, more recently I've come to realise that although I could accurately read notes off with no problem, I was not processing adequate information. I was playing things accurately, yet not really knowing what I was doing. Although the letter itself is nothing more than a labelling, by bring that labelling to the forefront of the mind, you consciously observe substantially more detail than you might ever bring in without carrying out the process. It's not about the letters but about the new thought processes which are triggered by thinking about them. It makes me much more consciously aware of how the music operates. Otherwise, I scarcely even know what my fingers are doing, or how the hell they get on with finding the right notes. This makes transferring from reading to memory a big problem. My ability to read things off well with little conscious thought is in many ways a hindrance. I need to counter it with much more conscious awareness. Ironically, this goes on to make my carefree reading even easier, when I willfully allow myself to return to that mode. The more foundation you develop for detailed conscious analysis, the easier it is to just let go when sight-reading. If you're permanently let go in the same old way, it's easy to get stuck in your current ability.

In particular, I think that pianists who are not advanced sightreaders are typically way short on this. For a piece where I've never thought about letters, it's never more than a split-second for me to identify any note. For poor sight-readers, I often have to wait some 5 or 10 seconds before they can tell me the letter name of note (even if they played it correctly). This is something that usually falls woefully short, but even as a pretty decent reader, I get considerable benefits out of the focus that is caused by forcing myself to go through the process of individually focussing on each separate note, so as to observe what is going on. It's too easy to just get on with playing the notes accurately by sight, without actually having a clue what you're really doing or how the music works. For more advanced readers, there's actually a point where you realise that simply playing the notes accurately is not good enough practise. You can still be missing out on countless styles of thinking that would be required to succeed with bigger challenges. To keep developing, you need to keep finding new associations and ways of looking at a score- beyond what has become hard-wired as your usual way.

[keypeg]

I'll write more when there is an answer to my question.  Before that there is no point.

[nyiregyhazi]

I'll write more when there is an answer to my question.  Before that there is no point.

Personally, I don't see that the differentiation matters. Development of sightreading is always based on seeing and then doing with confidence and acute physical awareness of the association between what has been read and what movement will correspond. Letting go and being willing to skip things but press on is an essential part of a rounded toolbox, but it doesn't contribute anything much to long term progress. It's just a thing you have to be capable of doing when emergency should arise. The foundations of true progress comes from establishing ever greater associations between the visual and the physical, with extreme clarity and certainty. The stronger these are, the easier it is to get back on track after an error. When a person struggles to keep going, my personal opinion is that these foundations are generally not strong enough. Some people need the reminder, but in most cases it's futile to tell someone to keep going. The reading skills are typically just not strong enough for them to leap back into place, no matter how strong the will to press on. The more they force themselves to skip on, the less they develop the real foundations of assurance- because they are constantly scrabbling around in wild panic rather than forming associations with any clarity.

[keypeg]

It does matter.  Many different things are being lumped under one name, and people also don't know what they are chasing, or why they are chasing it.

[nyiregyhazi]

It does matter.  Many different things are being lumped under one name, and people also don't know what they are chasing, or why they are chasing it.

My point is that if foundation stones are missing, it's of no use to simply practise pressing on blind. The foundation of all sight-reading is to read details quickly and effectively and to be able to find the corresponding movements. If this is deficient, all the desire in the world to press on cannot generate long-term progress (because all the approximations and wild guesses prevent the opportunity for firm associations between notation and movements to be cemented). It can only make you better at getting beyond all the screw-ups within your present ability level- which is not where serious progress can be found. Unfortunately, the modern approach to sight reading almost always seems geared towards the pressing on side of things. Without developing foundations (to later start to let go of in emergencies) however, there's really little scope for progress down these lines.

In my ideal world, the concept of pressing onwards whatever happens wouldn't even come into sightreading training, until a student is already at the stage of playing simple things with total confidence and assurance as a matter of routine. For my money, all categories of sightreading skills must originate from such foundations- or the pressing ahead at all costs approach makes for a most unpleasantly uncontrolled and frantic ride, that offers relatively little scope for notable ongoing advancement.

[timothy42b]

It does matter.  Many different things are being lumped under one name, and people also don't know what they are chasing, or why they are chasing it.

Trouble is, it isn't possible for a beginner to answer that question.  They aren't ready to understand it. 

Beginners struggle to master a piece, and when they finish they think the next piece will be easier.  Usually they are shocked and disappointed to find it is not.  There is a difference between the generic skill of playing any piece, and the specific skill to that next piece.  Beginners tend to think that ratio is 90/10, but it's more like 10/90.

So it is with sight reading, on piano (not so much so on monophonic instruments). 

Beginners think there is a generic skill, that of processing visual information and translating it into muscle motions.  And there is.  This is the fundamental that n talks about, and it is important. 

But IMO it is the 10, not the 90, and that is why beginners learn sightreading skills so slowly. 

And that is also why I emphasize not stopping, pressing on without fail.  The halting stumbling habit, once learned, is rarely if ever completely unlearned. 

[nyiregyhazi]

Trouble is, it isn't possible for a beginner to answer that question.  They aren't ready to understand it.  

Beginners struggle to master a piece, and when they finish they think the next piece will be easier.  Usually they are shocked and disappointed to find it is not.  There is a difference between the generic skill of playing any piece, and the specific skill to that next piece.  Beginners tend to think that ratio is 90/10, but it's more like 10/90.

So it is with sight reading, on piano (not so much so on monophonic instruments).  

Beginners think there is a generic skill, that of processing visual information and translating it into muscle motions.  And there is.  This is the fundamental that n talks about, and it is important.  

But IMO it is the 10, not the 90, and that is why beginners learn sightreading skills so slowly.  
 
And that is also why I emphasize not stopping, pressing on without fail.  The halting stumbling habit, once learned, is rarely if ever completely unlearned.

What exactly is the 90 then? Guesswork? I cannot see how you can denigrate foundation skills that govern ALL issues of accuracy to such a low level. Guesses and recoveries should always be the exception. Certainty and confidence should always be the rule. If not, we're merely talking about survival- not remotely good quality sight reading. Personally I'd say that 90% is understanding how to turn an instruction into a suitable movement. 10% is knowing how to get out of limited numbers of emergency scrapes without stopping.

I agree that the stumbling thing is a bad habit- but that's why I believe certainty must be stage one. If you don't feel ready for the next note, you shouldn't be playing it. It's not about forcing yourself to guess through a string of inevitable mistakes, but learning good enough skills to avoid them almost altogether. Only later in the learning process is it okay to encourage a stumble on through after an inaccurate guess. Earlier on, the real question is why the hell something basic was allowed to go wrong in the first place. Too much pressure due to rhythmic commitment (impying the tempo was too fast)? Too little thought? Whatever, anything that is less than certain actively screws up the foundation ability to translate an instruction straight into a movement. Only certainty and confidence develop foundation skills- not erratic stabs in the dark that are inadequately mentally prepared. Nothing but that certainty can rationally explain good, accurate sightreading- so the beginners have it exactly right. If they're not developing on this path, they're just not building strong enough associations.

[mcdiddy1]

Sight reading is a pretty complex task for musicians. You have to have a number of musical skills automated to be able to play correctly. One of the most common mistakes people tend to make about sight reading is to janine sight reading is mearly reading note to note. I think saying letter names is a way of addressing the symptom of poor pattern recognition but not an indicated of everything that needs to happen in sight reading. In order to sight read you have to develope a tactical awareness of the keyboard and acquire a decent level of playing technique. You should be able to identify notes in different octaves, positions, be able to play without looking down, etc. it is difficult to learn to sight read while learning basic technique. I think this is the stage the original poster needs to get past in order be a strong reader. When you get a good sense of the keyboard and experience you will not make mistakes like starting on an incorrect key.

The other steps include identifying they written symbols with sounds. Just by viewing a sheet of music, good sight readers have a sense of what the music should be just from experience and time on the instrument.  Experience sight reader most importantly have an extensive visual and tactics memory bank . Most rhythm and tonal patterns they have already incountered. It is very similar to reading in which you do not read every word of every sentence but you recognize words based on syntax, basically what is around the word to derive meaning. For example sight readers look at music from a variety of angles from an interval aspect, chords, scales, notes etc. for example when I sight read I may read a chord as a D major chord with a c in the middle and have a tactile awareness of what shape my hand needs to form to play it correctly. Or I may see a rhythm pattern and recognize how it needs to sound based on past experiences.

I think general advice like looking a head and to continue without stopping is helpful for learning the process of practicing sight reading but does not address the skills the sight reader needs to have. The best advice is to build you skills at recognizing rhythm and tonal patterns at the appropriate level. Basically you should be reading through a large amount of easy music with rhythm patterns you know or can easily figure out. If you find a book filled with popular tunes that are easily recognizable to you then you will have the advantage of know what the piece should should like and be able to learn rhythm patters and identify mistakes. I also think you should build foundational skills to be able to effectively perform on the instrument. If you start small, you can quickly develop a memory bank, especially if you do it everyday for a year or so.

[fleetfingers]

Although this is all true, you're simply missing the point of what I'm saying. Yes, you can do it with any form of labelling or simply with none- merely by knowing which key is signified by which symbol. However, the point about letters is that they reinforce associations and force the conscious mind to observe considerably more points of view- which serve to back each other up and make the mental picturing all the more clear.

I am sorry that I did not express myself very well, but I don't think I missed your point; I agree with it. In fact, I was going to quote you and type "+1", because you worded it so well, but I decided to write my own thing, and obviously didn't do a very good job. 

I was meaning to counter what another poster said about letter names being unnecessary. I was making a point that it's not the letter itself that matters, but the fact that there is a name to associate the note with. It's OK to think of the key or the sound it makes, but the key and/or sound should have a name.

[tchaikovsky_lover]

What I learned to do IS to slow down to get the notes right until you get the feel of how the notes are placed.  Just don't slow down during exams or tests.  They will take marks away from that.

[fleetfingers]

Also, regarding the point about not stopping to correct mistakes: I was only clarifying to the OP the reason people tell you to do that. If you are put on the spot to accompany another musician, choir, or congregation, that is exactly what you have to do. There may be mistakes, but you go on and get as much of it correct as you can. It truly is a skill that should be practiced if you want to be good at it. Also, the fingering does not always come out the way it would if you had time to prepare; so, you have to be able to quickly adjust and move your hands and fingers in ways that may not be standard to get the notes and stay with the music.

Timothy is right that stuttering to fix every mistake, or to repeat the correct notes when you get them right before moving on, is a huge bad habit. I know, because I used to do that. There does need to be training in playing with a steady flow, even in the beginning. But, as someone else already mentioned, the sight reading pieces for practice should be at an appropriate level so as to ensure good flow/tempo, as well as getting the notes correct.

[p2u_]

The idea of sight reading should be qualified more.  For one thing, a person's particular goals for sight reading should be thought through.

OK, keypeg, I see that nobody is addressing your questions, so I'll give it a try.

Personally, my only argument to become better at sightreading would be: become familiar with a lot more repertoire without necessarily needing to study it in depth. I feel that the other arguments that are usually used by "the system" in favor of developing this skill sound like mere goals of the system itself; the student is not even asked whether (s)he likes it or not, whether (s)he will need that skill or not. If you're bad at it, you're history, even if in other respects you may be a wonderful musician. Is it any wonder they make mistakes they cannot seem to be able to correct?

As the prodigy girl I mentioned earlier says: "They're just too busy trying to look good". Besides, they may also have a psychological problem accepting the poor sound results. It is even conceivable that the eyes as such work incorrectly even in general reading skills (not necessarily musical; just reading a text aloud) - very small eye span, too many wrong fixations, eyes going back when mistakes occur, etc. If the "cure" excludes this kind of problems, then the results will still be poor, no matter how long and how hard the "patient" works according to the instructions given.

P.S.: If he were still alive, I would be hard-pressed to tell someone like the great pianist Lazar Berman (who was a rather poor sight-reader) he has to do what is suggested above because he probably:
* doesn't know the keyboard topography
* doesn't know the names of the notes
* generally can't decide for himself what fingers and handpositions to use
* is incapable of listening to what's going on right now and to what's coming.

Paul

[p2u_]

If sight reading is important to you there must be studies on it.

There has been done quite a lot of research in the field of sight-reading. I don't know what I am allowed to give links to, but a general summary will lead anyone interested in the matter to find what they need to improve in themselves.

Provided the "subject" is able to read notes and has a reasonable knowledge of the keyboard topography (through scales, formulas, idioms, etc.), one may conclude the following (in random order):
1) sight-reading without auditory feedback leads to only slightly more mistakes than normal feedback;
2) omitting visual access to the keys completely results in markedly poorer performance. For good planning one needs to look at the keys sometimes;
3) working memory (the ability to store info in the short term memory to work with it) is paramount;
4) preview of two beats leads to a slower tempo, larger variability in note durations, and errors. Subjects performed better with previews between two and four beats or, ideally, with previews to the end of the next bar. Training preview (and probably also working memory) away from the instrument seems to be the logical approach if one wants to develop the skill quickly, the more so because pianists and organists have to zigzag through several staffs;
5) quality of interpretation (not merely pressing the right keys, but communicate something): we form meaningful units that are influenced by our previous knowledge and expectations. The better we are acquainted with the style and idiom of the piece, the better the results will be;
6) One can never sight-read beyond the level of rehearsed performance, which is an understatement, of course;
7) Training helps, but one has to pinpoint the real problems, which may be different with different subjects.

Paul

[theodore]

When sight reading with both hands, and upon making a mistake, say on beat one, I stop and then count out (silently) the remainder of the measure.  I immediately repeat the measure where I made the mistake, continue on with the next measure.  In this sight reading method, the flow never stops.

Mistaken measures are repeated until they are (or almost) error free.  The steady beat keeps on going.  This method also helps a great deal in the memorization process.  It is best done in slow tempos.

[timothy42b]

6) One can never sight-read beyond the level of rehearsed performance, which is an understatement, of course;

While that seems obvious, there is a subtlety here that is often missed. 

The "level of rehearsed performance" is not a general skill level, but a specific one.

Being able to play a level 4 piece in a given style or genre does not generalize to all, and sightreading is similarly specific.  Someone who sight reads hymns at speed will stumble at a big band piece OF THE SAME difficulty, and vice versa. 

That is why the so-called fundamentals are necessary but not sufficient. 

[keypeg]

I asked the question of the OP, because only when we know what he is after - which starts with him knowing what he is after - can his question be answered in a sensible way.  But in this whole thread, the person asking the question has not popped up once.  Did anyone notice?

I see it this way: People start piano and they read that they are supposed to "learn to sight read".  They also see the common formula: choose music which is below your level, don't stop to correct mistakes, etc.  They do it because they have read they are supposed to do it.  They haven't figured out what skills they need, where they are at, what the purpose or goals of this thing is.

Sight reading of the kind where you take an unfamiliar piece and play it close to tempo not stopping for mistakes is a specialized skill.  It is needed when accompanying a soloist or a choir.  We don't need this skill when starting out.  The thing that is useful is to be able to slowly play through a new piece so that you can find out what it is about, and maybe get a sense of where the difficult parts are going to be.

The skill that we do need to acquire in the beginning is the ability to read music.  That means being able to recognize the notes on both staffs, and associate them with the piano keys.  A sense of intervals and recognition of common chords is probably good, and then there is the physical act of tracking the notes as you play them.

If you recognize that you need the skills that I just mentioned, then you are no longer chasing a generic "we're supposed to practice sight reading, and we do this by playing music below our level a tempo and not stopping if we make a mistake."  You look at where you're at, what you need to learn, and how to get there.  Which was the reason for my question.  The OP needs to know where he is at and what he is after.

I ran into the question of sight reading when I joined forums after getting a piano after decades.   I considered what I was reading and decided this was not for me.  I needed to get a handle on notes on the page and notes on the keyboard.  I had a certain amount of theory and a sense of patterns in music to carry me.  I did not chase the sight reading path because it did not seem to match what I needed to learn first.

Now I do have a rather solid connection to notes etc.  I do go after sight reading now.  I don't think it would have been a good idea to skip notes and allow wrong notes when I started and was trying to get solid in my connection to written notes and the keyboard.

Btw, I DO think that a beginner can get at such goals.  In fact, if you are self-teaching then you have to know what you are aiming for, or else you risk going in circles.

[keypeg]

P2u, thank you for taking the time to try to answer my question.

McDiddy, you seem to be saying the same thing I am, from another angle.

[pianoplayjl]

As I understand, Sight reading is when you play a piece you haven't seen before straight away. So, using the definition,  I don't exactly sight read a piece because when I play it the first time, I scan through the piece for about 1 minute and play on my laps before playing. Whenever I get stuck or make a mistake I just simply skip 2 or 3 notes. That I know is a bad habit but it comes in handy during exams. So my only advice is just scan through the piece and look for accidentals. Play it on your fingers and count in your head. You can drop this process after you have improved sight reading.

JL

[nyiregyhazi]

While that seems obvious, there is a subtlety here that is often missed.  

The "level of rehearsed performance" is not a general skill level, but a specific one.

Being able to play a level 4 piece in a given style or genre does not generalize to all, and sightreading is similarly specific.  Someone who sight reads hymns at speed will stumble at a big band piece OF THE SAME difficulty, and vice versa.  

That is why the so-called fundamentals are necessary but not sufficient.  

This makes no logical sense. They are simply based on different fundamentals. You state that someone who reads hyms well "will stumble" in a big band piece? While I presume you did not mean this literally, it's worth emphasising the fact that this is not something that ought to be assumed for a moment. A person with rounded fundamentals reads both well. A person who only does one well is lacking in fundamental skills, despite possessing ones in certain areas. Good fundamentals apply widely- not just to certain styles. If you can only sightread particular styles, it's because of lacking core skills.

All the example shows is how broadly fundamentals must be developed, in order to achieve wide-ranging success. Provided that I count, I have no more problem sight-reading jazzy pieces than classical ones. Do I regularly play such pieces? No. They're just based on the same fundamental skills of reading.

[nyiregyhazi]

I asked the question of the OP, because only when we know what he is after - which starts with him knowing what he is after - can his question be answered in a sensible way.  But in this whole thread, the person asking the question has not popped up once.  Did anyone notice?

The question doesn't need to be asked. All long term sightreading progress comes from development of fundamentals. The poster said that he struggles not to keep playing everything a note out after an error. That shows beyond any doubt he needs to focus above all on these fundamental reading issues- before he worries about forcing himself to press on. The latter would be putting the cart before the horse and could actively hinder his progress.

[p2u_]

Provided that I count, I have no more problem sight-reading jazzy pieces than classical ones. Do I regularly play such pieces? No. They're just based on the same fundamental skills of reading.

Correct me if I'm wrong, but is it possible that you and timothy42b are talking about different levels of performance, different end results? I agree that what you say is the case if we speak about just hitting the right notes in time, but what about interpretation of such pieces? I can hardly imagine that it would be enough for jazzy pieces to sound convincingly by just counting. One needs a feel for the style too, which is not as fundamental as it seems. The worse the expectations (in terms of sound result) -> the worse the result all over; even the elementary skills may be affected (a psychological problem).

Paul

[nyiregyhazi]

Correct me if I'm wrong, but is it possible that you and timothy42b are talking about different levels of performance, different end results? I agree that what you say is the case if we speak about just hitting the right notes in time, but what about interpretation of such pieces? I can hardly imagine that it would be enough for jazzy pieces to sound convincingly by just counting. One needs a feel for the style too, which is not as fundamental as it seems. The worse the expectations (in terms of sound result) -> the worse the result all over; even the elementary skills may be affected (a psychological problem).

Paul

I didn't get the impression that he was talking about depth of interpretation. If so, we're talking about issues that are not specific to sight-reading, but rather to interpretation in general. Obviously you want to sightread with the best sense of style possible, but if we're talking about feel for "swing" say, it's no easier for a good reader whether you've learned the notes of a piece or are playing it by sight. Either you have it or you don't. I don't personally view that as an issue of sightreading. I wouldn't complain that someone was a poor sightreader if they played a Chopin nocturne unmusically but accurately at first. I'd just assume that they're either a poor musician (or that they require a lot of time to get beyond mere note spinning).

[timothy42b]

Correct me if I'm wrong, but is it possible that you and timothy42b are talking about different levels of performance, different end results?

I can't be sure, but I do not believe this is the case.

For a long time I shared n's belief that sightreading was mostly skill based.

In recent years, through observation and discussion with musicians more advanced than I, I have changed my mind.  I have come to think that the skill based portion is far less important than the memory retrieval function (after the pure beginner stage of course.) 

N refuses to consider that possibility, but then long experience has shown discussions with him to be futile.  It is indeed fortunate that he is already so learned, or imperviousness to new ideas might become a problem.  Observant readers will have noticed by now that I never reply to him directly. 

[edger]

Every sight reading advice I read recommends not to stop or slow the tempo when one makes mistakes- "just keep going" they always say.

You haven't read "Piano guided sight reading" by Leonhard Deutsch.  He advocates (when practicing sight reading) to slow the tempo down in difficult phrases to make sure you get the notes right... Then speed back up again for easier parts...

[nyiregyhazi]

I can't be sure, but I do not believe this is the case.

For a long time I shared n's belief that sightreading was mostly skill based.

In recent years, through observation and discussion with musicians more advanced than I, I have changed my mind.  I have come to think that the skill based portion is far less important than the memory retrieval function (after the pure beginner stage of course.)  

N refuses to consider that possibility, but then long experience has shown discussions with him to be futile.  It is indeed fortunate that he is already so learned, or imperviousness to new ideas might become a problem.  Observant readers will have noticed by now that I never reply to him directly.  

The fact that you fail to address any of my rational points is why I am in no hurry to change my mind.

I used to come from the position you hold. I didn't know how I did it- and assumed that I deduced notes from recognising harmonies etc. However, the more I've thought rationally about it, the more improbable the type of explanation I used to believe becomes. I recognise harmonies and patterns BECAUSE I can process fine detail quickly and with accuracy. The whole thing is in reverse to what I had formerly believed. What are you suggesting enables precise execution- if not precise reading and precise ability for the hands to reflect it?

Statistically, it's just too improbable that good sightreaders can have such a high hit-rate unless they effortlessly process substantial levels of detail. Other than the fact you can recognise a particular unit by identifying its constituent parts, musical writing is nothing like words. There are limited orders for letters- allowing a lot of logical assumptions. However, any simple chord can also have any additional chromatic note in it. Any simple chord can also be spelled God knows how many ways. Good readers pick up on the details- or they could not possibly play with accuracy. Sadly most are so good as to have minimal awareness of what detail they have to process to view things as they do- and hence have false perceptions of seeing harmony first and working backwards. For a person who does not effortlessly process the details, this is simply an impossibility. You cannot reliably work by harmony unless detail can be processed- otherwise there will be regular and very significant errors.

You say it's not about these foundation issues- so what are you saying it is about? To play with 90% accuracy, you'd have to assume that approximately 90% of what is written is being processed effectively. How can you think that foundation skills only account for 10% of a player who plays with 90% accuracy (or preferably far better, in any good reader)? What explains the level of accuracy- if not solid foundation skills?