Sunday, July 20, 2014

My first recording session

My good friend Jonas Pettersson, a composer and incredible musician who was also my first music instructor, allowed me to record some piano pieces in his apartment where he has artist's piano and excellent equipment.  In addition, he wanted to show me some of the basics of how a recording studio operates.

I have about 10ish pieces floating around in my memory at this point, and I played 5 of them with between one and a few takes each over a period of an hour.  I was astonished at what a difference the piano and equipment makes to the quality of sound.  Here are the 5 pieces over an 11 minute track.

Prelude in C Major, Dmitri Shostakovich
Prelude and Fugue in G Major, J.S. Bach (Well-Tempered Clavier Volume 2)
Prelude in E Major, J.S. Bach (Well-Tempered Clavier Volume 2)
Prelude in A Major, Dmitri Shostakovich
Prelude Op. 10 No.4, Alexander Scriabin

I no longer feel bad at all about my sound on inferior pianos, and actually I can more clearly see the flaws in my playing on a piano like this and know I can do a far better job.  I just need to focus on polishing a piece, play on this kind of setup more often, and a little more technique to handle tricky areas is always a necessity and something I know just takes time.

I'm convinced the chief difficulty in producing a good tone is the disconnect between the sound you are trying to create and what you actually create.  If an inferior piano, or a flub of your technique, or both, disconnect you from the sound you are creating, then you lose the musical sense of the piece and it's difficult to recover.  On the other hand, if you get on a roll with your technique, and the barriers between what you are trying to create and what's happening are removed, all of a sudden you can really make beautiful music.  With unpolished pieces, not enough technique, and inferior pianos, that disconnection is almost impossible to avoid in any attempted interpretation, except by incredible luck.

First Piano Composition

I was at my tutor desk on a Monday during this Summer of 2014, and no students came that day.  I had walked to school that morning (4 miles) and before the library opened, I thought I would try to just sketch a piano piece from scratch.  I had made a few other minor attempts to compose in the past, but this felt like my first genuine attempt to conceive something completely original and completely in my mind, and sketch it on staff paper.

I have grown a love of drawing music by hand on staff paper.  I've always been fond of the idea of being able to take staff music anywhere and compose things in my mind, then sketch them.  This piece would be the attempted fulfillment of that dream.

The main work began at the tutor desk.  In about an hour I was finished.  I went to the piano afterwards, and I was excited to hear that not only was it beautiful and interesting, but it was more or less exactly how I imagined it and wanted it to sound.  All I did was add fingering, detail markings (articulations, dynamics, etc.) and add an ending.

I recorded the piece on my digital piano but I much prefer the sonority of even the inferior grands at my school.  Here is the sketch:

And here is a digital recording of the piece that doesn't quite capture what I want, but is good enough:

The piece was a bit more of an exercise in composition and creativity rather than something guaranteed to please the ears.  Not that I don't think it sounds pleasant and interesting of course!  In case there are interested readers, I will describe my thought process for composing it.

Because you have to start somewhere, I picked a harmony out of the blue as the first molecule.  In my mind imagined a chord, and it turned out to be a dominant 7 chord with the 9th added.  I decided upon a basic structure of repeating chords in the right hand with a simple moving bass line underneath.  This kind of sonority is very much inspired by the music I am listening to at the moment, especially Dmitri Shostakovich's book of 24 Preludes and Fugues.  I decided to begin with a thin texture, and then fill out the chord on beats 3 and 4 to create variety and interest.  I decided the piece would begin 5 to 1, as pieces often do - it just sounds good.  As the piece continued I knew I needed a new idea.  I'm not sure why I thought of a resolving suspension in the upper register (9 to 8), except that it occurred to me and I knew it would sound beautiful.  The next few bars were simply preserving symmetry and unity - a thinning of the texture and a 7-8 suspension.

I decided my next goal would be to modulate, and I was struck by the idea of modulating down a whole-step.  In order to do this, I wanted to start with the IV chord in the key of D (G) which is the dominant in the key of C.  I wanted to follow a more unusual path though and drew upon inspiration from my first music instructor, composer, and good friend Jonas Pettersson.  I thought of all the common tones between the two scales and realized I could preserve an element of unity from the beginning of the piece (the dominant 7th with the 9th added) - the only tone that needed to change was an F# to an F natural.  This change is smooth, and yet sudden enough to be surprising.  The next element, a descending step-wise in the upper register, is an extension of what came before and something that I imagined would just sound lovely.  Again, the next few bars are simply symmetrical to preserve unity - I feel like that by now the ear would crave something more and it can certainly be improved.  The idea for the final cadence was that the hands would trade the notes D and C, and the last chord would be in the key between the two - C# Major.  That for me was just a random clever idea I had.  I don't remember ever seeing that or hearing about it before - it just seemed kind of cool and I think it worked out well.

Anyway, thanks for reading and/or listening!

Tuesday, December 17, 2013

Absolute Pitch: Inquiry, and Methods for Learning It

It's time for another edition of my Piano Whatever Blog.  Today's subject is the contentious and misunderstood topic of "Perfect Pitch."

Absolute Pitch:
Inquiry, and Methods for Learning It

I. Introduction

In the Fall Semester of 2012, I became interested in being able to identify what note I am hearing.  As a motive, I thought it would be pretty awesome to be able to do this, obviously.  At one point, my mother thought I had this talent, as I was good at playing the piano by ear as a young child.  One of my tricks was reconstructing pieces I had heard from memory by slowly finding the right notes of something that I had heard.  However, that was a long time ago, and I hadn't had musical training since then until very recently as an adult.  I wasn't convinced I had this talent at all because if you played a note, I had no idea what it was and had no idea how I was supposed to find out.  I suspected that I had what they called 'Relative Pitch' instead.  If there was a note blaring somewhere, with time and some guesses, I could find the note on the piano because I could tell the difference between the notes.  It is also true my pitch memory was fairly keen, and to a very limited extent I could do this from memory.  Of course, I didn't actually know what Relative Pitch was, and so I went off in search of answers to these and other questions.  Over the semester I got involved in reading various materials online, both scientific and amateur, about the phenomenon of Perfect Pitch and what the research says about it, the experiences of people who have it, and the various attempts people have made to acquire it.  I also made the attempt to learn the skill myself by using computer software and lots and lots of mental practice and drilling.

II. Resources

When I first looked into the subject of Perfect Pitch on the internet, it was like beholding a giant ocean stretching from horizon to horizon from the shore.  I dived in, swimming from one island of insight to the next (yay, lame extended metaphors.)  All of it was very interesting to read of course, but there seemed to be a great deal of conflict in concepts and definitions of what Perfect Pitch actually is.  The first reference I examined was an article that was already present in a piano technique book I was reading, which is C.C. Chang's Fundamentals of Piano Practice.  This excellent manual can be found here:

This large reference manual was a labor of love by Mr. Chang and is full of wonderful information about practicing at the piano.  I have enormous respect for this project, and it should give you a good idea of its scope that it includes information about everything from Absolute Pitch to piano tuning.

In his article about Perfect Pitch, he makes the following claims:

  1. Perfect Pitch doesn't really exist.  Instead, there is Absolute Pitch and Relative Pitch.
  2. Absolute pitch is being able to identify an exact note (at the piano for example), whereas Relative Pitch is being able to identify that note with a few tries.
  3. Absolute Pitch can be acquired later in life if one already has relative pitch, but it is difficult and fragile and requires constant mental practice to maintain.
  4. Potentially, all individuals with early musical training should be able to develop Absolute Pitch automatically.
  5. Fundamentally, Absolute Pitch is a learned trait.

These claims were the starting point for my research into this subject.  Particularly, I was interested in the definitions he was using because they pointed the way to greater clarity on the subject.  Of course, Wikipedia has a entry about the subject:

The Wikipedia article asserts that Absolute Pitch can not truly be acquired later in life, but 'psuedo-absolute pitch' can be with a great deal of time, motivation, and effort.  This seems to be more or less what C.C. Chang is claiming, with a slight definition difference.  Perfect Pitch seems to be a more old-fashioned term and is tied more to the belief that nature and not nurture decides if you possess it.  The big question of course is what the real difference between Absolute Pitch and 'pseudo-absolute pitch' happens to be.  What is really going on there?  I continued searching.

The following study, conducted by the UC San Francisco, is attempting to determine the genetic basis for Absolute Pitch:

They assert that Perfect Pitch (which seems to be more or less synonymous with Absolute Pitch in the literature) cannot be acquired later in life as there are no documented cases of an adult developing it, and also that it most likely has a genetic basis due to their distribution graph which shows that individuals either have it, or don't have it, with very few lying somewhere in between the distribution.

Thus, they argue that the scientific evidence argues strongly in favor that Absolute Pitch is largely nature, not nurture, especially because there are individuals with early musical training that still do not develop Absolute Pitch.  This stands against the more optimistic claims of C.C. Chang.  It also is more in line with the traditional view of 'Perfect Pitch', as either you have it, or you don't.

In addition to looking at the theory and ideas surrounding the phenomenon, there were some excellent blogs and forum threads I had read about individuals with varying experiences attempting to acquire this ability when they did not have it.  There were failures, successes, and cases in between.    There was one blog in particular that was readable, fascinating, and documented an individuals research and personal investigations over a long span of time.  Unfortunately it escapes me and I can't dig it up right now.  However I do remember it led me to this website:

This is one of the best sources for Absolute Pitch ear training - in the research and theory sections, there are an enormous number of references.  This websites asserts optimistically that Absolute Pitch can be learned as an adult and has excellent software developed around a number of theories relating to 'chroma' or sound quality.

Another highly developed software training application exists here:

This game is fun to play and challenging - the only downside is that once it gets to higher levels where accuracy as high as 95% is required, it feels like it comes down a lot to luck and grinding it out than progressively learning.  However if you've reached those levels than you are more than good to go when it comes to functionally having the skill.

III. Personal Investigations

This was the literature and theory I read and the games I played over the Fall and Spring semesters while I investigated acquiring this skill.  I will complete this article with my own experiences and opinions on how best to acquire Absolute Pitch as an adult.

First, I will state the results of my training.  For all intents and purposes, I now have Absolute Pitch.  The skills I have developed over two semesters are for all intents and purposes functionally equivalent to 'Perfect Pitch'.  Where they are best labeled 'psuedo-absolute pitch' or 'Perfect Pitch', I think, is more of an academic debate rather than a reality of the usefulness of the skill.

When a note is played, I can nearly instantly (within 3 seconds) identify what it is with an accuracy somewhere between 90%-100%, and with a small amount of time to check and cross reference my near-instant response, the accuracy of my answer will be just under 100%.   I always make the very occasional mistake.  The more practice I am in, the more rare mistakes are.  As I think about and work with music less (such as over the Summer of 2013), they start to wane and deteriorate, although so far I haven't seen anything approaching a reversal of the intense drilling and training I undertook.  Mistakes are so rare at this point that I never worry about it, and I have never been off by more than 1 semitone since the end of last semester.

When multiple notes are played, I can slowly pick them all out one by one and identify them, using knowledge of theory Relative Pitch skills to assist me.  This means that in practice, soon I will be able to identify entire harmonies and therefore chord qualities and inversions being played, all in the correct key.  My current limit of what I can handle is around three to four notes played at once - more than that and it is unlikely I will be able to distinguish them all at my current level of training.  Because of my extensive knowledge now of theory and relative pitch ear training, I can use these skills to fill an enormous number of gaps.  I'm sure that in time, I will be able to arrange and transcribe ensemble pieces with a very high degree of accuracy in the correct key and even play my instrument the piano in this fashion, although my technique and familiarity with the instrument will limit me for quite some time.  In the meantime, I can sing any melody or play it on the piano if it is simple, by ear, in the correct key.

Compared with Fall 2012, this is an enormous amount of progress to say the least.  In the Fall of 2012, if you played a note, I would give you a blank stare.  I could not identify notes, let alone sing them back to you in the right key.  I had difficulties in my singing group, the ECC Chorale, matching the right pitches.  It wasn't so much that I was singing the wrong notes.  Rather, I had difficulties hearing myself, singing the right parts, and not getting confused, suggesting that at least I had some degree of Relative Pitch.  I could still eventually find the note on the piano if I remembered the piece and reconstruct it on the piano slowly, which I could do as a child as well.  So it seems I came into this learning experience with some degree of musical ability already: a relatively strong and accurate pitch memory.  This will be the starting point for describing my experiences in learning Absolute Pitch, and also it's basis.

It all began in my theory class where we were first learning Ear Training.  The teacher explained that we would have to identify an interval of notes.  Someone in the class asked in bewilderment if we had to know what note it was.  The teacher replied that what he was describing was known as Perfect Pitch, and that is not required for the class because it is not something everyone has.  Instead, we will be given the note.  He also went on to explain that by using a device called an 'incipit', which is a melodic idea or tune that illustrates the interval, we can learn to identify or sing intervals without knowing the notes involved.

At the time, I happened to be obsessed with this gorgeous piece of music: - Op.87 Prelude & Fugue No.7 A major, perf. by Keith Jarrett

I had listened to it about fifty times because it is so lovely.  The subject of the Fugue (beginning around 1:10) begins on an A Major Triad.  I had heard it so many times, I could hear the glorious A Major subject in my head fairly clearly.  That's when an idea dawned on me.  If what I remembered was fairly accurate, couldn't I use this an incipit for the note A?  

Thus, I learned my first note.  Of course, the success of this idea rested on whether or not I could clearly remember and perceive the piece in the key that I first heard it.  In this case, it seems I could for the most part - perhaps because I had heard it enough times, or that my pitch memory was already strong.  At first, I was concerned with just singing the note or thinking about it, and then comparing the result with the right note on my instrument.  The largest error I remembered experiencing was no more than a whole tone (rare), or semi-tone (somewhat frequently thinking G# or Bb, for example).  Heartened by the consistent accuracy, I went on to see if I could hear the note and identify it around me.

This next step proved to be much, much harder.  In fact, the immediate and brutal failure was very disheartening and discouraging to the point that I became convinced for awhile that I merely had decent Relative Pitch and was kidding myself about the ability to acquire anything more.  While attempting to listen for the note, I was consistently thrown off, thinking that the note I was hearing was A when it was not.  I also noticed that as I listened to the piece less and less, my memory of the note became somewhat less precise.  I could still hold on to it but since the entire basis of this trick was my memory, I speculated that this situation was temporary and too difficult to sustain over any long period.  I was stuck for about a month with no progress and struggling to hold on to the one note I had.

Nevertheless, as with any obsession of mine, I continued on, if only for kicks.  I purchased a software to train absolute pitch that threw notes at you in quick succession while you had to identify what they are by clicking them on a piano.  I was terrible at the game, but I had some success at nailing A when it did come up.  I was restricted to within one octave - C4 to B4.  I had to set the software to play the notes very slowly so I had time to make the necessary 'calculations'.  It also had a mode where it would play notes so that you could learn them.  This mode didn't work for me at all - I would have it play a note over and over for me and I would try to 'memorize' it or 'learn it' so that I could identify it later, and it was hopeless.  It was fleeting - I would lose it quickly.

Despite these difficulties, I pressed on with the greater challenge of learning more incipits for more notes!  I was struck by the idea of using the Well-Tempered Clavier which had a piano piece in every key in my instrument (I play piano), and thought I would learn all of the themes.  This didn't work so well - a lot of the themes were not necessarily singable or memorable, plus the amount of themes and tunes you would have to learn seemed overwhelming.  I learned an important lesson about this method - in order for it to work, you are better off using melodies that you already know very well, and are very, very tuneful and easy to remember.  A lot of the subjects are complicated, don't always begin on the tonic, and the tonic is obscured by other harmonies.  The fact that this Shostakovich piece rings like a bell on A (and, as I learned later, A440, the tuning note), was a happy coincidence and this is why it worked so well for this purpose.

Bit by bit, I assembled an arsenal of tunes for every key.  I remember that for the note Bb, there was this choral piece our Chorale was working on - the Pergolesi Magnificat in B-Flat Major.  The tuneful theme, beginning on B-Flat, was looping around in my head endlessly.  However, later on in the piece, the tune is stated in the dominant key, and other keys.  This resulted in hilarious errors.  I remember at one point I was going to boldly sing a B-Flat out loud.  Instead, much to my dismay I sung the right tune, but in the wrong key - F Major, the dominant.  I was also very flat.  It's hard to be more far off than that in terms of the actual note!  But another lesson was learned - you have to make sure you are remembering the right part of the song.  Later on, when my skills improved, I noticed how many other students would remember tunes, and, thinking that they were way off, were actually more correct than they could ever give themselves credit for - they were right on the money, but just in the wrong part of the song!  Our chorale director is always trying to remember works in key but is not always successful - but I don't know if she realizes how often she is 100% accurate but is either singing another voice part or is singing a part of the piece where the theme has modulated.

After learning lessons about what tunes worked and what didn't, I collected a motley list of tunes for every note.  Circa late Fall 2012, it looked something like this:

C: Prelude and Fugue in C (Bach, Shostakovich), Invention No. 1 in C Major (Bach)
C#: Fugue in C# Minor WTC 1, Fantasia Impromptu (Chopin, 5 to 1)
D: Piano Concerto No. 3 (Rachmaninoff), Invention No. 4 in D Minor (Bach), Cum Sancto Spiritu (Vivaldi)
Eb/D#: Prelude in Eb, WTC 1 (Bach), Don't You Worry 'bout a Thing (Stevie Wonder)
E: Prelude in E WTC 2 (Bach), Menuet in E (Haydn), Shenandoah (Mack Wilberg)
F: Piano Concerto No. 19 in F (Mozart), Invention No. 8 in F (Bach)
F#: O Vos Omnes 1st ending (Luis Salazar)
G: Prelude in G, WTC 2 (Bach), 5th Symphony (Beethoven)
Ab/G#: Concerto for Solo Piano (Alkan), Fantasia Impromptu (Chopin), Agnes Aria (S.J. Pettersson)
A: Fugue in A Major (Shostakovich), Fugue in D Major (Shostakovich, 5 to 1)
Bb: Prelude in B Flat WTC 1 (Bach), Magnificat (Pergolesi)
B: O Vos Omnes (Luis Salazar), Spasinye Solidelal (Tchesnokov), Mass in B Minor (Bach)

However, I noticed that as old songs became less clear in my memory, I needed to replace them with new repertoire that I was actually working on.  For rare keys though, this posed a challenge.  I came up with the idea of burning a CD of these tunes and listening to them frequently to keep my 'brain tuned'.  I still use this idea, and every once in awhile I make a new CD reflecting my current mental practice music.  I need to listen to it every so often to stay on target.

What about my pitch memory itself, the lynch pin holding everything together?  It was difficult to assess it's strength relative to before these experiments, as I had not really been a musical person or particularly involved in music until late as an adult, and not to this extent until that very semester.  I did notice however that to a small degree, my memory itself improved and became more stable with practice.  Before these tests and early on in my self-training, I would remember things fairly accurately, being off sometimes by a semi-tone or less occasionally a whole tone, or sometimes remembering the right material but in the wrong part of the song.  By fall 2014, I would be remembering music in the right key nearly all the time, the right part of the material, and with the only error being one semi-tone off and even then rarely so.

With the first piece of training software and my arsenal of tunes, I began bit by bit to succeed in identifying notes that I was hearing.  The basic procedure is this - I would hear the note, mirror it internally in my mind, and see if it related to any one of the tunes I had learned.  If it did, and I could clearly remember the tune, by knowing what note the song began on, I could then assert that the note I was hearing was that note.  This process at first was naturally quite slow and subject to confusion and error.  Often I would resort to heuristics (short cuts) to make more educated guesses if I wasn't sure of my answer.  I worked on developing a series of cross-reference checks in my mind - using tunes I was more sure of and interval comparison for example, or using the context of the music I was hearing to help narrow down the options (for example, it's less likely to hear this note in this key, because it's chromatic - etc.).  I found that the more notes you know, in a sense it becomes easier to identify them because you know more about what a note is NOT, as well as what it is.  Also, I learned that in order to progress, you had to just be confident and go ahead and give a note your best guess and then do your best to learn what error you made, if any, and then how to improve on your next guess.  Every time I heard a note anywhere in the music building, I would set my mind to work 'calculating' what it was.  I had a few music student friends test me on the piano, exclaiming that 'perfection' was my goal.  One told me that I would probably ever get it perfect, but they all noted my progress and seemed both interested and impressed by it.

Starting with just one octave and lots and lots of drilling, I reached the point where I could identify any note from C4 to B4 with a note being played at about one note per 4-5 seconds, with about 80-90% accuracy.  Pushing myself, I expanded to two octaves, then three, then the whole piano, then having the notes being played faster and faster.  While I was doing this, I participated in the UC San Francisco study and played various other games on the internet.  At first, I failed to qualify as having Absolute Pitch in the UCSF study, but after more training and a few more attempts, I finally made it.  When I found ProLobe and AP Avenue (linked above), I was able to push my skills even further.  I discovered that the only way to stretch your skills was to challenge yourself with something much harder.  It seemed similar in nature to visualization in chess.  At first, you start off with the ability to visualize a certain number of moves ahead.  In order to improve, you have to push yourself to see just a little bit further, which mentally is quite taxing.  Also, you have to constantly reinforce what you can do over and over in addition to this stretching.  By playing a game that included the whole piano with muddy low notes, ringing high notes, and multiple notes being played at once, identifying just one note being played on my own instrument became a trivial thing.

So now I sit with a very useful skill that can no doubt be developed even further.  I am happy with my progress so far.  Once I reached a certain level around the end of Summer 2013, I lost interested in pushing it even further for the time being, because I had already invested dozens if not hundreds of hours drilling these skills.  For the entirety of Fall 2013, my skills were functionally equivalent to AP and everyone seemed to notice this.  Just being involved in music and working with it in my head (which I do as a music student anyway) was more than sufficient for keeping it fresh, and of course I would keep listening to my CD of tunes every so often.

I am presenting this story to anyone else who may have the same inkling that I did to engage in this nerdy quest.  I have come to the end of this journey with a few broad conclusions:

The method that worked for me is what I believe is referred to broadly as Melody Triggers, and I've seen it discussed elsewhere on the internet as a learning method that some have tried and used with success, others without.  The idea is that a note that you hear triggers a melody in your mind which is tied to a certain note, tipping you off as to what note it is.  It is dependent on having a good long-term pitch memory and relative pitch skills already, as well as the concentration and focus to work with music in your mind.

I believe there are four broad levels of pitch awareness:
    • Tone Deaf - This represents the inability or difficulty of telling the difference between pitch classes.  In other words, someone who hears and E has trouble telling it from a D, even with them played within a short time window.  Most likely their long-term pitch and short-term pitch memory are weak and they are struggle to hear or analyze music in their mind.  There are a variety of levels within this category.  Being totally tone deaf is probably quite rare, whereas the kind of tone deafness just below Relative Pitch is more common.  I've noticed that some people possess only certain kinds of tone deafness - in other words, they are able to sing a note correctly, but unable to hear it or mirror it internally.  Or perhaps they can hear it or mirror it internally, but struggle to hear themselves singing it.  Others may know all of the scale degrees but mix them up in their mind, so have perfect intonation but sing the wrong part or the wrong note for the context.
    • Relative Pitch - This represents the level of many untrained individuals who are habitual music listeners or have had a significant exposure, as well as your average musical and musically trained individual.  Relative Pitch means you are able to distinguish between all 12 tones in relationship to one another, particularly in the Major and Minor scales.  The short-term pitch memory of these individuals is often quite accurate and functionally equivalent to perfect, however their long-term pitch memory is often less reliable.  Like those with Absolute Pitch, individuals with relative pitch possess a degree of internal music making and mental practice, although they are usually not confident that they are in the correct key!
    • Absolute Pitch - This represents the level of individuals who are either very musical innately (fortunately), have had a high degree of musical exposure and training while young, or both.  Functionally, it means having a pitch memory and recognition that is accurate enough to perceive music internally and externally to the key in which it was originally heard.  In other words, it mostly represents a highly accurate pitch memory.  Absolute Pitch possessors are individuals with great pitch memories or good ones with Relative Pitch who have undergone training, and they use this memory to work with notes.  Absolute Pitch is characterized by a high degree of mental practice and internal music making - in fact, it may be a prerequisite for most individuals who possess this skill to continue mental practice or else their skill will wane and deteriorate.
    • Perfect Pitch - It appears some special individuals have even more extraordinary and solid reinforcement of their perception of musical tones.  It is unclear to what degree this is the result of early musical training during the critical early years of life or the product of unusual musical abilities, but their perception of tones is as much linguistically based as memory based and can therefore be even more accurate and confident in their answer than with a skill based purely on memory.  Also, a condition exists called Synesthesia whereby certain individuals can associate colors, sensations, and other experiences with aural stimulus such as pitches.  This can give them 100% accuracy and confidence in their pitch awareness, and someone with this condition perhaps constitutes the classic or old-fashioned case of 'Perfect Pitch'.    It is also double-sided, as individuals with this trait report irritation and even physiological reactions to music that is out of tune.
It is my conclusion that many, if not most, individuals with Relative Pitch have the ability to develop functionally equivalent Absolute Pitch or "psuedo-absolute pitch" with a great deal of practice and training, and the motivation to sustain it with constant practice.  However, I cannot be confident that all individuals with RP would be able to do this.  It is merely my hope and belief, substantiated in part by my own experience, but far from proven.  In my own case I was clearly working with a good pitch memory already.  Perhaps individuals with very unreliable long-term pitch memories may not already have the tools to use the Melody Triggers method, but this does not preclude them possibly training their pitch memory to the point that they could (see below.)  For individuals with Relative Pitch and above average or good long-term pitch memories that aren't quite all the way there, I believe the way is open for them to acquire AP with these methods.  For individuals who have been very pitch deprived and / or are tone deaf, it may be possible for them to acquire Relative Pitch skills, but whether they can take the next step and acquire Absolute Pitch is highly uncertain, especially as adults.  Possible, perhaps, for the exceptionally dedicated and obsessed?

As all I am saying so far applies to adult learners, the news is even more optimistic for those who are young or young adults.  While the difficulty of such studies as an adult are undeniable, the younger one is, the more plastic and 'sponge-like' your brain is generally, especially within the age range of 0-12 years old.  I would speculate along with C.C. Chang that the vast majority of individuals with vigorous musical training at this age should be able to develop Absolute Pitch.  This is founded on the following observation he has made, which is chromatic scale is a human invention - there is nothing innate about it.  It must be learned.  To the criticism that there are individuals with early musical training who do not develop it, I am now very curious to know what methods were used and what theory those methods were informed by.  Teaching a human being to recognize musical tones as a kind of language, and strengthening their musical memory through constant exposure to perfectly-tuned instruments, are the methods called for.  What methods are currently being used?  Are they proper, efficient, and effective?  Are they standardized?  These questions demand answers.  Perhaps a scientifically informed Absolute Pitch course for young children could be developed and taught universally in studies to further test whether this ability is really off limits to anyone.

I would go further and say that second, while to some extent the strength of one's pitch memory (as with any memory) is surely genetically determined, this is true for all memory.  It's also true that all memories that we know of can be improved with practice and training,  even as an adult.  Why should pitch memory be any different?  The rest of pitch memory is surely determined by the degree to one's exposure to music (and therefore pitches) over one's life - i.e., nature.  The higher degree of exposure, the more 'in tune' your brain and pitch memory will be.  It should really be that simple, and to assert otherwise I think is a more extraordinary claim.  Therefore, because the particular method I use is founded on memory and memory training, it should be accessible to everyone as a possible avenue, although of course success for every individual is not guaranteed, especially as an adult learner.  If one's long-term pitch memory is not consistently reliable enough, then the key would be to train this memory - perhaps separate focused methods are called for here, supplementing the ones I've already discussed that worked for myself.

Third, all human beings are born with the capacity to recognize intonation (a critical component of comprehending language) and with some degree of pitch memory (everyone gets a tune stuck in their head, show me one who doesn't!) - we all have the hardware that makes these methods theoretically possible.  If we can all learn a language and learn how to interpret meaning from the inflection of one's voice, we should all be able to associate a letter with a particular pitch sensation because language already does this.  This suggests why speakers of tone languages have vastly higher incidences of perfect pitch.  As a reminder, adults are more than capable of learning a new language - even a tone language. 

Fourth, there is a large prejudice and misconception in our society about the nature of Perfect Pitch and it being a nature and not nurture trait, which only hinders the confidence and the persistence of those who seek to acquire it.  I argue that the evidence to show it is entirely innate is highly dubious - there are self-reported instances everywhere of individuals acquiring it, including myself.  I have a critique in particular to level at the UCSF study based on my experience.  It is worth putting in bold text for its importance.

Because a very reliable pitch memory, instead of just an average one, is required for functionally equivalent Absolute Pitch, this is the reason I speculate why the distribution curve in the UCSF is so one-sided.  It's not that there aren't many gradients on the road to Absolute Pitch - it's just that if your pitch memory is around one whole tone off for example, it's not quite good enough for the functional skill, even though it's still way ahead of someone who is tone deaf.  In the UCSF survey you will both appear to have around the same level of competence at pitch awareness, when this is clearly not the case in reality.  Arguably, an individual who is only around a whole-tone off with some training could have the accuracy necessary for functionally equivalent Absolute Pitch. It may be that there are no recorded instances of 'true Absolute Pitch' being developed in an adult, yet 'psuedo-absolute pitch' is allowed for and recognized.  Perhaps 'true Absolute Pitch' in this case is referring more to Synethesia which was discussed earlier as an unusual condition and not a requirement for AP.  I struggle to see a difference otherwise, and if they are functionally equivalent, I'm sure no one who makes use of their functionally equivalent skill would care what it is called.  Far greater clarity in the real difference between 'psuedo AP' and 'real AP' is required on the part of the literature and I hope my experience and other self-reported success stories can help clarify the true nature of Absolute Pitch.

In the end, there is one final point to be made, and that is the nature of learning such things as an adult versus a child.  It is indeed true that once we are already an adult, these things become much more challenging, and success cannot be guaranteed.  However, for anyone who is already musically inclined and involved in music, I would not hesitate in giving it a try if you are motivated and willing!  I would also recommend the method of Melody Triggers if you are already good at distinguishing pitches, but you might want to start with Chroma methods first if you have difficulty distinguishing between the 7 diatonic and 5 chromatic tones of a scale.  

If you are an adult you have a challenge ahead of you, which is why I believe all children should be taught the musical alphabet at an early age.  Early exposure is critical to learning anything but persistence at any age will be rewarded. 

Tuesday, April 9, 2013

Stuff About Scales

How to Use Scales

The Purpose of Scales, and How to Practice Them:

Scales are complicated exercises that should be practiced a variety of ways to ensure success.  To become very fluent with scales at high speeds, it isn't enough to repeat them over and over with no purpose in mind.  It also goes without saying that this way of practicing them is soul-crushingly boring and tiresome.  There are many difficulties in scales that need to be practiced in isolation before you can combine them together and play the whole scale.  This is similar in analogy to a piano piece, which can't be learned from start to finish but instead has to be broken down into many smaller and more easily learned parts - then combined.  The reason for this in scales and in anything else is that it is very difficult or impossible to focus on improving more than one task at a time, which is what one would have to do if repeating the scale from start to finish.  Even when it is possible in some cases to focus on many things at once, and even if it sometimes gets results to do so,  it is always much less efficient, and therefore inferior to breaking the tasks down to practice one at time, because our time to practice piano is not unlimited.  The good news is scales are a great tool to practice all aspects of music, and it's with these purposes that we should create our philosophy of how to play scales.

Here are the important components of the scale exercise:

0) Learning the Notes

1)      Rhythmic accuracy and perfect counting (Most Fundamental)
2)      Technique
a) Comfort, relaxation, confidence
b) Accuracy
c) Hand/Finger Placement
d) Key motions
   i) Epicycles
   ii) Thumb Under
   iii) Hand shape
3)      Articulation accuracy (legato, staccato, etc.)
4)      Dynamic accuracy
5)      Tonal Control

Each of these components should be practiced in isolation, and gradually combined together.  The way to do this is to play a scale at a very comfortable slow speed - as low as necessary to play well.  Sometimes this means not playing too slowly, because at too slow of a speed your concentration can be easily disrupted.  You may not be very comfortable playing scales at any speed, so pick the best speed you can.  It doesn't matter how slow or fast it is, but should probably be quite slow.  Next, you will need to work very hard on becoming independent from the metronome for awhile.  Rhythmic accuracy, next to playing the right notes, is fundamental and necessary before learning how to improve any other part of your scales (or music for that matter).  Once you are working on other areas of your scales, you must continue to strive for rhythmic precision and come back to working on it as needed.

You need to be using your ear to work on the other aspects of the scale and this only becomes possible once you can ignore the metronome.  The goal of exercises for #1 on this list is to achieve complete independence from the metronome.   Once this is accomplished, pick an exercise to do in a scale and practice it with the goal of achieving perfection in that particular area.  Concentrate only on the goal of the exercise and try to 'file' other things away into your subconscious.  At first, you will be working on some singular aspect while 'filing' away rhythmic accuracy.  If it is too difficult to do, then continue practicing #1 (rhythmic accuracy) until it is automatic.  Once you can practice rhythm and any other aspect, it is not really necessary to rigidly combine things in some particular order - you can try practicing any two combinations (for example, perfect dynamics and articulation) as long as your rhythm is precise.  Then you can try practicing more than two aspects at once, until eventually you try to do them all.  It is extremely challenging and will take an enormous amount of time before you can perfect every aspect of playing a scale at any speed. 

It is not necessary to practice in every key all of the time - in fact, the exercise of the scale itself is what is important, not what particular key you are playing in.  Try to play a different scale every day, or play groups of scales.  Every scale is essentially unique, but at the same time not that different from the others.  The skills you build in one scale will transfer to every other scale.  The point of this approach is to be fun and flexible - you can practice any scale and any aspects you want.  Maybe D# melodic minor will be your scale today, or perhaps D Major?  Just make sure to be precise rhythmically - if you have problems here, solve them first.   Hopefully now scales can be more fun and interesting to practice.  In time you will be able to play any scale and yes, eventually you will have to practice them all.

Before starting any exercises, find your 'comfort speed'.  Find the speed you can play at as comfortably as possible, and write this speed down.  This may vary slightly between scales but don't worry too much - as long as your scales are close to each other in speed it's fine for them to vary a bit- keep a table log of each scale.  Alternatively, you can pick the lowest speed.  You will stay at this speed for all of these exercises until you can successfully combine all of them together and play a flawless scale.  Then increase the metronome by 1bpm and repeat the exercises.  You may find that the amount effort required to surmount the initial comfort speed is colossal, but the increase of 1bpm is almost non-existent.  You may also find that later down the road, for some reason moving a few bpm up from where you are seems very difficult, because you've hit a speed wall for your current technique, and new motions need to be discovered.  All of this is quite normal - piano progress is never linear. 

The most important final note is to give no thought or care whatsoever to the beats per minute you are currently at or planning to arrive at.  You should stay happily where you are until you have achieved total mastery at that speed - a beautiful clean scale at a low speed is worth a thousand sloppy scales at a high speed.


Important note: When I use the term "spurts" in this article, it just means quick repetitions at varying speeds (constant for each spurt, but subject to increase or decrease at your desire).  In this case it would be acceptable to exceed the metronome for these little exercises - after all, you want these difficult areas to be super solid and you want to expose the issues that come up playing them fast.  Aside from these mini-exercises, when playing entire two-handed exercises stay at your comfort speed and enforce rigorous rhythmic accuracy.

0) Learning the Notes

Scales should be learned very well hands separately, and then put hands together very slowly, one note at a time.  This is one of the first very challenging obstacles for a beginning piano player because the hands must become decoupled - they are performing radically different motions together, crossing over and under at different times and assuming different shapes.  To get on the right track, it's a good idea to read the 'Finger/Hand Placement' and 'Key Motions' sections and to roughly aim for these approaches but do not worry at all about perfecting them straight away.  This article for the most part assumes you are an intermediate player who has learned their scales, but wants a better way to practice them (my own situation really.)

1) Rhythmic Accuracy and Perfect Counting:


Rhythmic accuracy is simple to comprehend on its own, but many of us don't understand how to achieve it and find the metronome troublesome and counter-productive to use.  The truth is we will all find it nearly impossible to develop a correct sense of rhythm without working on it in isolation, and it will still take a lot of work.  It can take years to develop a very solid sense of this, so it's important to do it the right way - the metronome is a terrible thing if you waste your time listening to it while practicing other things.  Listening to a metronome and repeating something over and over again until it reaches that clicking speed is a terrible and soul-crushing way to do scales and all other music, and also horrifically backward.  The goal of these exercises is to make rhythm a part of your subconscious thinking so you can eventually get rid of the metronome altogether, just using it to set an initial tempo and check how you are doing without it.  Once it's gone you can then file it away into your subconscious focus on improving the rest of the music.  Free from that incessant clicking or beeping, rhythm will instead become the lovable (and now wonderfully accurate) groove of music that it's meant to be.


1)      Set the metronome, and play the scale counting loudly out loud.  First count quarters only. Then add smaller subdivisions - eighths, and then sixteenths.  Then try longer notes - count only half notes, then whole notes.  A four octave two-handed scale with sixteenth notes as the smallest division will have 14 quarter notes.  You can count it as a time signature with 14 beats per measure, break it up into smaller groups, or just count on one for the whole scale.  Eventually move towards counting softer and softer.
2)      The next step is counting in your head.  Choose quarter notes as your starting subdivision as they are easiest.  Remember to concentrate only on rhythm during these exercises.  Don't purposely be sloppy in your playing, but do not worry about very good technique, proper dynamics, etc. because when you start to focus on other things you will have a very hard time focusing on counting rhythm in your head.  If you want, use some kind of external rhythm counter (foot tapping, gentle swaying, etc.) as well as internal counting but the goal in the end is to jettison this and create an internal pulse or groove.  Feel the groove, but when performing you don't want to be distracting from the music by clicking your teeth or something.  To increase the difficulty, go to eighth notes and sixteenth notes, and then try half notes and whole notes.
3)      It can't hurt to try other rhythmic variations; for example, you can practice playing scales in different subdivisions (like triplets) or different time signatures.  Scales are in fact just the tool for this practice, but these would be less about scales themselves and more about rhythm in general.
4)      Gradually get rid of the metronome.  Start by turning the volume down bit by bit, until it's barely audible.  Then stop using it, trying instead to rely on that internal pulse or groove, only checking occasionally to see how you are doing.  If you have a digital piano, you can record MIDI tracks of your practice which will give you exact feedback of how rhythmically precise you are without a metronome.

2a) Technique: Comfort, Relaxation, Confidence


Comfort is more than just being relaxed at the piano and finding your playing to be easy.  Comfort is also a state of mind - it's playing something while feeling very confident in what you are doing because it is not straining you.  For many players, even ones who can play challenging pieces, this state of mind is foreign to them.  They are never 'truly' comfortable, and always on edge about mistakes and always building tension throughout a piece.  The only way to get better at this is to do it on its own.  Total confidence in scales will come when you have mastered every aspect of them, but you can go a long way practicing comfort, relaxation, and confidence by doing so in isolation.


1)      Breathing exercise - Play your scale, and concentrate on breathing as much as you need to feel calm, keeping every single aspect of your body relaxed except for the tiniest amount of tension necessary to play the notes.  Take deep breaths through your nose as much as possible and through your diaphragm, not shallow breaths with your chest.  There is a lot of information about proper breathing techniques in Yoga manuals and other places and it all applies here.  Try to practice taking deep breaths in opportune periods of time - such as when the rest of the technique allows for it.  Don't force deep breaths in areas of the scale where they aren't comfortable.   Try to feel a positive mindset or mind when you are playing - that you are having fun and that what you are playing is taking effort, but it's not worrying you or making you tense.  Instead, think about how fun and enjoyable it is to play the piano while also feeling calm, relaxed, oxygenated!

2)      Posture exercise - Play your scale and pay attention to your entire body.  Discover that perfect posture that works for you - keeps you looking upright to the audience (difficult to find for some with back problems like me) but also gives you as little pain or tension (or none of you can) anywhere in your body.  Also find that perfect distance and bench height (distance is where you can access the whole piano comfortably, height is where your arms are level with the keys and don't have to raise up or down).  Discover all of this - discover your piano playing pose.  Once you find it, play your scale over and over and pay attention to when you are shifting around or becoming uncomfortable, and try to revert to your comfortable zone.  A lot of people never find this before they dive into all sorts of piano stuff and pay the price later. 

If you can't do these exercises at the speed you are playing then slow down.  Don't consciously worry very much about other aspects of scales while practicing these things, but don't get too sloppy either of course.  It's also a good idea to make an exception to the usual principle of not combining exercises: it's never a bad idea to practice breathing and relaxation while doing any other exercise!  However, you will find that until you can do it in isolation, you will probably not be able to pay any attention to it while concentrating on making every note exactly as loud as the next (for example)!

2b) Accuracy


When we are playing anything in piano, it's easy to get a bit sloppy and compensate for it, instead of be rigorously accurate.  For example, all but one major scale and one minor scale include black keys.  Striking these black keys must be precise because they are smaller than the other keys.  Actually you should be precise on white keys also - aim for the center of the note, and with the motions to make accuracy as sure as possible.  If your motions are naturally getting your fingers into trouble, it's time to work on them.  Striking those black keys requires a degree of precision you may not appreciate until you are in the 80bpm range of speed and suddenly you are flubbing black keys left and right.

Remember, the more precise you are early on, the more precise you will be later if you continue to work on it - it will carry indefinitely into future speeds until you are playing 160bpm striking every single key dead center.  It also means that if for whatever reason you are not up to par (exhaustion, nervousness, etc.) your usual precision will mean that any deviations these random factors create will be greatly nullified. If you are off by a little bit, you still won't be flubbing any notes because of how precise your accuracy is.


1)      Play your scale, and concentrate on placing your fingers as accurately as possible.  Play as slow as you need in order to hit every note comfortably and accurately in such a way that with practice you feel precise enough not to worry about flubs.

2c) Hand/Finger Placement


Closely related to the above topic of accuracy is hand/finger placement.  Black keys are often struck with a curved finger, but this need not always be the case despite what some teachers want you to believe.  Playing black keys flatter offers some more ease but it can be awkward when you are playing surrounding white keys in a curved fashion.  Whenever you are playing white and black keys, use the 'grey' area of the piano - the area between being all the way into the black keys and all the way down in the white keys.  Experiment with different combinations of flat and curved fingers and different placements of the hand in or out of the keys.  An important truth to realize is that there is no one right way to play in all situations, but there are right and wrong ways in every particular situation.  It may seem like there are a gazillion situations but they can all be reduced to a finite number of placements, hand positions, and finger curvatures - it's all about what you are playing and your goal should be to find the combination and motion that is the most comfortable and produces the sound you want.  To some extent, scales are just difficult for human hands to play in a lot of ways - it seems we can never find perfect solutions for our hands to effortlessly roll up and down the keys from white notes to black notes with our uneven fingers.  But this is your chance to practice these little groups of notes on their own without having to worry about playing the entire scale perfectly.


1)      Experiment - for any one group of notes in a scale that is troublesome, practice different hand and finger placements to find a comfortable solution, and then practice your solution spurts.
2)      White Keys and Black Keys - Practice small groups of notes in spurts and focus on evenness.
3)      Black Keys - Pick a section of black keys in a row (such as the key of F# major or D# Minor) and play them with very flat fingers in spurts aiming for perfect evenness.
4)      For White Keys - Play an entire C Major scale with curved fingers.  Now try flat finger positions - both are appropriate for groups of white keys, depending on the situation.

2d) Key Motions (epicycles, thumb under, hand shape)


People who practice scales quickly figure out that not all motions are equal in difficulty.  In fact, there seem to be just a couple of motions that cause almost all of the trouble.  The first is the thumb crossing under going up in the right hand, and the thumb crossing under going down in the left hand.  The next two are the fingers crossing over the thumb - L.H. going up, and R.H. going down.  Included in this discussion are all of the small variations of these themes in other scales.  It stands to reason then that we should give these the appropriate amount of concern, yet so many people just repeat scales over and over practicing these motions only a few times on each go, while these weakest links prevent them from speeding up the whole thing.  Meanwhile the whole thing is probably ready to be played faster!  Don't get sucked into this trap - these motions require focused practice.

There are big 'debates' to be had about which motions are 'optimal' but anyone can see that basic or intuitive motions often don't get the job done; repeating them over and over causes a lot of strain and never gets that fast.  The most important thing to realize is our imperfect hands on this strange contraption will not be playing scales like typing on a keyboard - imperfection in form will be necessary to create perfect sound!  What I mean by that is that our hands will look 'funny' or 'asymmetrical' and that's exactly what is needed.  Some people might think that all scales are played the same, or with the same motions.  In reality, every single scale is unique in the motions it requires and the shape of the hand.  I don't even think it's particularly useful to group them into fingering patterns, etc.  Of course, although being unique, the differences between them are not enormous either, so you don't have to worry about devoting a year to mastering each scale's particularities.  In fact, skill in one scale will transfer to every other scale perfectly well, even though they are completely unique!  How incredible the human mind is!


So what basics apply to pretty much all scales?  The first is epicycles - in order for a finger to cross over another finger, the hand should rotate to help out.  This will require turning your wrist in the appropriate direction.  This may result in your elbow and forearm tiling out or in a bit - this is necessary.  Epicycles should be practiced in isolation and the goal is to make them comfortable and to do what is necessary but not more.  Most people use these motions to at least some extent subconsciously, but since they aren't aware of what they are doing, they don't stop to consider exactly what it is they are doing and why, and whether they are doing it too much or to little.  The goal of an epicycle should be that the rotation of the wrist allows the motion of the scale to stay fluid and uninterrupted.  The point of this technique is that you can continue to go up and down without interruption by making it one fluid motion instead of stopping, starting, and lots and LOTS of strain that will happen otherwise.  For example, in the right hand going down in the C Major scale, as soon as the thumb reaches C4 and your 4th finger wants to cross over, you are aiming for a motion that starts really before your thumb hits C and ends after your 4th finger hits B3.  In one fluid motion, your wrist prepares to rotate sometime earlier (around E or D perhaps) and then goes around your thumb to allow your 4th finger to strike B3 in one motion.  This requires quite a turn on the part of your wrist and can be feel quite strenuous at first but this is why you have to practice it in isolation - if practiced enough with comfort and relaxation as the goal, along with economy of motion, freedom and looseness of the wrist should be achieved to the point that this motion causes no undue strain.  If this motion is bothersome, attempting to play a scale is hopeless.  Of course, it may never be totally effortless either, but your goal is to get as close as possible.

Thumb Under

Next is thumb under motions.  Having to cross your thumbs under your hand is the notorious bane of all scales and greatly restricts what humans can accomplish.  There are some who argue that thumb crossings at high speeds to achieve legato is unnecessary or even impossible, and that a lose wrist and a quick break in legato is preferable.  The debate is not that important however because in the end, these so-called different methods really converge on the same thing - a free wrist.  When playing legato, your wrist must learn to be free to rotate and stay rotated to some extent.  If your hand is not positioned at a comfortable angle, the thumb under motion is just too slow and straining to ever achieve comfort and high speeds.  By tilting your hand, your thumb has less travel distance and requires less motion of those slow muscles to go where it needs to go.  Also, it becomes possible (just as with epicycles) for the thumb under motion to begin and end before it actually happens, creating one fluid uninterrupted motion.  It's also important to focus on the thumb itself - it is said that having a straight thumb and playing with the knuckle is ideal.  Of course, this isn't how it works for everyone, but if your thumb is pointing awkward directions or is very tense take a moment to find out what works for you.

Hand Shape

Now it's time to discuss hand shaping.  When playing anything on the piano, an important goal to achieve is what is commonly called "quiet hands".  This just means that your hands are not freaking out from note to note or chord to chord with lots of unnecessary motions.  Most importantly, they are 'relaxed' while achieving this economy of motion.  In scales, how much motion of your hands is really necessary to play the scale?  If we understand that the same notes are being repeated over and over, it seems like we really need maybe only one hand shape, plus the motions we already discussed.  It's possible to divide the scale up further perhaps, but by and large I feel that only one primary shape is required.  This doesn't mean that you keep your hand perfectly still and tense to make it look as quiet as possible!  It just means your hands should largely be static and relaxed while the shape of your hand is already set for the work that needs doing.  An easy example is the C# Major scale.  In this scale, your thumbs play white keys while the rest play black keys - possibly the most harmonious arrangement of all the scales!  Your hands can maintain a steady shape throughout this scale and the fingers need only do a small amount of work to play the scale.  This is probably the ideal scale I would use to practice hand shape and quiet hands.  More than any other scale, this scale gets closest to that 'perfect' form of typing that many imagine should be used to play scales.  In fact, placing your hands perfectly even and spaced like on a keyboard is almost the perfect motion for this scale!

Other scales can get much harder however and the law of imperfect form to achieve perfect sound applies instead.  Tension will build up quickly trying to keep one hand shape.  But the goal is that there is going to be a shape of the hand that requires the least amount of extra motion to do all of the necessary motions required for that scale.  When you watch a very good pianist play a rapid scale, you notice that their hand adopts a shape and rockets up and down the keyboard alternating between thumb under motions for R.H. or epicycles for L.H., and vice versa.  If you've tried to play scales very fast and had any success, you notice that this is just what your hand naturally seems to want to do.  There is just not time for anything but the bare minimum of motion you need - your hand will naturally want to form the optimal shape that requires the least extra motion so that your fingers and wrist rotations can get the rest done.  Like everything else, this must be practiced in isolation, and no faster than you can do correctly (except when using exploratory spurts).


1)      Play your scale and focus on epicycles and thumb under motions.  When going up with a two-handed scale, your left hand will be doing epicycles while your right hand is rotated to accommodate the thumb under motion.  It will be the reverse going down.  
2)      Try practicing just going down from the top of the scale, in isolation.  The reason for this is a lot of people get stuck practicing scales from the start over and over, and this sometimes means they get more practice going up the keyboard than down.  Also, different issues crop up when you have to lean over to the higher register compared with when you are in your comfort zone (where pianists typically start their scales.)  
3)      Pick a particular troublesome epicycle or thumb under in a scale and practice it in spurts, in isolation.
4)      When practicing epicycles and thumb under motions, try to adopt a rigid hand shape and imagine that they are 'anchors' that stay nearly motionless while your fingers do the playing.  If the position is uncomfortable or builds tension, adjust it until it feels more relaxed and natural.  Don't tense up your hand to keep it rigid - if it's an economic shape it should want to form that way to reduce effort!

2e) Articulation Accuracy


Scales are in general played legato, but it's easy to get sloppy and play them too legato.  Also, other articulation patterns are called for in music and scales are a great tool to practice them all.


1)      Play your scale and focus on being as cleanly legato as possible.
2)      Reverse articulation - to improve your legato, try playing a scale staccatisimo to get used to a clean break.
3)      Practice a whole scale with a tenuto or accent on each note.  Reversing the articulation you are desiring to produce counter-intuitively will help you learn it.
4)      Note that for things like accents and tenutos, you will want to get your entire body involved for a good tone (especially your wrists).  Feel free to practice finger accents (playing loudly from the knuckle, Hanon style) but this isn't the only thing you should practice.

3) Dynamic Accuracy


After having mastered and combined all of the previous aspects for hours and hours, there is still more work to be done.  Without dynamic control over notes or scales in general, we can't hope to ever play musically.  However, having reached this step and having gone onto learn it, we are taking our first steps into the wonderful world of musical interpretation - the point of all this work!  There are really two aspects to dynamics - the touch and feel (tactile sense) and what you are hearing (aural sense).  Perfect control of dynamics is achieved when these senses are in harmony.  Often, most players have very good senses of one or both in isolation, but have enormous trouble combining them.  A good place to start is to play everything at the same level of volume.  The first step in mastering dynamic accuracy is to play every note equally loud (mf).  Although we should be cautious about playing everything loudly to mask our incompetence, aiming for no dynamic variation at all is actually just as challenging as putting in consciously-directed and musical dynamic variation.  This is because they are really the same thing - consciously controlling and paying attention to the dynamics.  Because of the differences in finger strength, and all of the technical motions required to perform a scale, this is an extremely difficult exercise to really do right.  Most people give up at trying to perfect this and just accept a lower standard.  It's all up to you in the end what kind of quality of playing you want to achieve.  You don't have to be a God of dynamics to make things sound good but, don't you want to be?  So play that scale and listen - listen very carefully.  Are you really playing everything at the same level?  I would go with mf as your dynamic level because it will throw out the idea of smashing everything working.  If you hear even the slightest variation, work on it.  If you are unable to do this or some of the other above aspects suffers, you are either not ready or playing too fast.  Many, many hours and of patient work in this area will be rewarded. 


1)      Play a scale with every note being mezzo-forte.
2)      Play a scale with every note at some other dynamic level (I'd go with p and pp next, be careful with f)
3)      Play a scale with a gradual crescendo going upwards and a decrescendo going downwards.
4)      Play a scale with multiple crescendos and decrescendos, starting with just two going up and one going down.
5)      If you have a digital piano, MIDI recordings can be used to see how loud you are playing every note!  This will give you feedback on both your ear and your touch sense.  You may find that you aren't as accurate as you want but more importantly, this can be used as a tool to perhaps achieve ungodly levels of dynamic precision.

4) Tonal Control


Tone is the endgame of technique.  Having mastered rhythm, fingerings, articulations, and dynamics, you now realize that you have a wealth of sounds on the piano you can create at will.  You have realized that subtle changes of finger placement alter touch, subtle increases of tension in your playing in some areas adds a different tone, and subtle fluctuations in rhythm create an infinity of possibilities in phrasing.  Combining these skills with knowledge of the damper pedal and soft pedal, and suddenly the piano is instrument of incredible musical potential.  There are limits to what kind of sounds the piano can produce, but those limits are far higher than you probably realize.


1)      Experiment - try playing scales with different touches, varying articulations, and paying attention to and learning the subtly different sounds of each.  Create a musical inventory of these sounds.
2)      Try using different amounts of pedal and soft pedal as you play a scale.  See what sound a particular approach creates without pedal, with 1/4 pedal, with 3/4 pedal, and with full pedal.  Try adding different levels of soft pedal as well.
3)      Introduce tightly controlled but audible rubato in your scales.  Focus switching back to keeping rigid rhythm at will.
4)      Use scales now as a tool to explore a whole world of music!