Idiomatic Gestures, the Economy of Motion and Jazz Piano Improvisation

Sept. 21/18 edit: Based on feedback from students, understanding the idea of good fingering, hand positions and anchors may be best demonstrated first using the chromatic scale. I’ve also come to think that playing the chromatic scale with good fingering is foundational for understanding idiomatic hand gestures in general. So, I’ve added a short section below (before the scale exercises) that outlines the process of discovering the anchors in the chromatic scale and using those anchors as the basis for designing exercises.

Before I get into the exercises, I’d like to share a few (incomplete) thoughts and observations that I’ve had over the last few years.

(Thanks to Scott Suttie, John MacLeod, and Trevor Giancola for the inspiring conversations while I was writing this!)

Playing the piano part for Rhapsody in Blue sounds harder than it is.  The most technically sounding sections are very idiomatic to playing the piano.  Just as we can analyze and simplify the music into chucks of logical harmonic progressions, pianists can also analyze and simplify the music into chucks of logical hand positions and gestures.

This sequence of hand positions may vary slightly between pianists (using the 3rd finger instead of the 2nd for example), but generally, we can say that there will be an agreement among pianists about the most physically optimal way to play Rhapsody in Blue, or any other written piece of music.  The more technically demanding a piece is, the more important that pianists draw upon idiomatic gestures and an economy of motion during a performance, lest they run out of energy or hurt themselves.

From another angle, some music may not be ‘optimally’ written for piano in the first place.  Perhaps it was written by a violin player, or by someone exclusively using harmonic theory, with no regard for idiomatic gestures. This would be akin to writing music for trombone in B major.  No doubt there are players who would perform it perfectly, but at the cost of less optimal economy of motion (especially compared to performing it in Bb major).

Performers are constantly navigating this evolving relationship between their instrument, technical abilities, and the music they want to play.  Unfortunately for pianists, these idiomatic gestures aren’t always explicit when reading music.  Even if the score has fingerings, those fingerings don’t always encourage the player to visualize, analyze, and simplify the music in this way.  So, part of learning a piece of music is developing your own ‘interpretations’ of what these idiomatic gestures are.  Considering the number of ways you can physically play a single passage on the piano, inexperienced pianists will often create strange fingerings, odd hand positions, and bad habits. These can lead to an uncomfortable relationship with their instrument, poor performances, and eventually injury.

For improvising jazz pianists, the issue of idiomatic gestures is more complicated.

When jazz musicians play together, they’re drawing from a common body of harmonic, melodic, and rhythmic language.  I suspect that because this language is shared among so many different instruments with different physical gestures, the music is often communicated through music theory and an aural tradition.  After all, piano fingerings and hand positions are not relevant to playing the saxophone (and vice versa). Chords, be-bop scales, swing, jazz standards, playing together, using your ear, and having musical conversations are relevant to ALL instruments.

It’s encouraged that everyone learns Charlie Parker solos, no matter what instrument they play.  It’s encouraged that you learn tunes in all twelve keys.  D Dorian is a common scale to use when improvising over Dm7.  But don’t just play the scale, listen to how it sounds in relation to the harmony. Or better yet, before you play, try to hear a melody first, then play it.

In this aural tradition, all semitones are equal. However, physically, that’s not the case. On the piano, E-F feel (and look) different than D#-E.  It’s much easier to play A-Bb on the trombone than it is to play Bb-B.  One of my favourite examples of this is shown on the piano, with this melodic pattern.

The top example (in Gb Major) is very easy to play with good fingering.  The bottom example (in G Major), despite being an exact transposition, has a higher learning curve. In fact, it took me four tries to find the most ‘optimal’ fingering for myself. And even with this fingering, it took a while to internalize the hand positions.

The point is this: transposing a melody, chord, or harmonic sequence will require different physical gestures, some more idiomatic than others.  Sometimes this is a compromise: sacrificing economy of motion to learn a shared language.

This relates to an issue concerning what inspires improvisation.

It occurred to me that when I play the piano, my improvising is being inspired by many different streams, not just theoretical knowledge, what I’m hearing in the moment, or what my fellow musicians are playing.  Sometimes I’ll play something because it physically feels good to play. Or because it visually creates an interesting pattern of white and black notes.  Or because it’s going to be difficult and I want to take a risk.

All of these streams of inspiration are working, fighting, and dancing together until ultimately resolved in the act of playing.  If this is the case, some streams may be more dominant than others.  Personally, I’m more likely to practice permutations of fingering patterns than melodic patterns.  There’s cross over between the two, sometimes one informing the other, but generally, I’m a physical learner and enjoy analyzing and simplifying piano music into chunks of fingerings and hand positions.

This includes transcriptions of improvised solos.  The most technically proficient jazz pianists most definitely use idiomatic gestures and have a developed economy of motion.  When I look at an Oscar Peterson transcription, I primarily see shapes and patterns of white and black notes and imagine how Oscar’s hands fit on these patterns.  It’s possible that Oscar’s improvisations were being conceived this way in the first place, as well as being complemented with ideas of harmonic and melodic patterns.

In fact, I know they were. Because in the act of playing, you can’t separate theoretical content from physical gestures.

But transcriptions of improvised solos almost exclusively focus on harmonic, melodic, and rhythmic content (sometimes they don’t even include the left-hand!). Rarely do they include an analysis of possible fingering and hand positions.  Once again, jazz pianists are forced to develop their own ‘interpretations’ of what these fingerings and hand positions are, often leading to strange fingerings, odd hand positions and bad habits.

The following exercises are the beginning of a personal exploration of practicing jazz improvisatory language in conjunction with physical gestures at the piano.  They’re treated as equal partners.

Exercises, patterns, and permutations that are conceptualized from harmonic/theoretical ideas will include particular fingerings and hand positions. Similarly, exercises that are conceptualized from physical gestures will fit stylistically within the jazz music paradigm.

I once heard about a jazz piano player who practiced all scales with the same fingering. So, try playing the Db Major scale with the fingering you’d normally use for C Major.

I respect the value in this, but this is not the approach I’m taking here. I’m interested in creating exercises and music that feels good and natural to play on the piano.  These exercises strive for a solid economy of motion.

A few obligatory notes:

Obviously, these exercises are for piano and keyboard players.  As these exercises develop, they may include theoretical material that would be interesting to other instrumentalists.  The actual methods of developing the exercises may be applicable to their practice also.  But primarily, the most value will be gained by pianists and keyboardists.

The piano has a long tradition,and there are hundreds of exercise books formalizing idiomatic gestures on the piano.  I’m familiar with a few of them and they’re somewhat relevant (Rafael Joseffy’s School of Advanced Piano Playing, Hanon exercises, and Moszkowski’s Scales & Double Notes). But as far as I know, these books aren’t made for improvising jazz pianists, so their methods aren’t wholly relevant.  I admit though that I’m mostly naïve to the field of idiomatic gestures and piano pedagogy. So, if you’re aware of other books, research and pianists that explore these ideas, please share!

These aren’t musical exercises, they’re physical exercises. For that reason, they’re fairly maximalist, often encouraging a steady stream of 8th notes, with little rhythmic variation.  The idea is to become proficient at an exercise and unlock a particular potential in your playing.  With musical discretion, it’s easier to take all those notes out than put them in.

I can’t emphasize enough how much physical repetition is required to incorporate this into your playing.  This is grunt work that follows a simple two-step loop:

  • Repeat until it’s too easy
  • Make it slightly more difficult

Personally, I know I’m making progress when I’m away from the piano but can still visualize the notes and hand positions as if I’m playing them.  This usually takes a few cycles of practice and rest to achieve this.  With regular practice, I’ll notice it coming out naturally in my improvisations after about three months.

These exercises are primarily relevant to my own practice.  I won’t claim that my fingerings and hand positions are the most optimal for all pianists.  Actually, I’m sure better fingerings will be discovered days, weeks, and months after I publish!  So please, share your own!

Lastly, as I mentioned above, this is the beginning of a personal exploration.  I have no idea where all of this will take me, but I’m going to go with the flow, and document it as best as I can.

When it comes to fingering the chromatic scale, this example is common among my piano students:

This fingering is superior:

The hand positions not only optimize playing the chromatic scale itself, but they’re foundational for understanding idiomatic hand gestures in general and can be applied to playing almost anything that includes stepwise motion.

From C, the right hand fingering can be simplified as a sequence of 4-note and 3-note groupings starting on the thumb (4-3-4-3-3-4-4). When ascending the scale, the placement of the thumb looks, feels and acts as a kind of anchor for the entire hand. These groupings and anchors could be visualized like this:

Or this:

When descending in the right hand, the fingering is identical, but because cross over points are now on the 4th and 3rd fingers, hand positions and anchor points are perceived from those fingers:

This relationship is inverted in the left hand. From C#, when ascending the scale, the placement of the 4th and 3rd fingers anchor the hand. When descending, it’s the thumb:

When you look closer, you’ll notice that fingerings and hand positions in the second octave are different from the first octave. In the right hand, the 3rd finger plays D, but in the second octave, the thumb plays D:

This opens up an opportunity to practice the chromatic scale in an interesting way.  A major theme to these exercises is that physical gestures at the piano are treated as equal partners with jazz improvisatory language.  Often, harmonic and theoretical ideas dictate the rules and goals of certain exercises.  We physically conform to these goals and learn the gestures needed to play them.

But what if that relationship was inverted?  Rather than learning to play sequences of semitones ascending and descending, why not also learn to play sequences of hand positions ascending and descending?

If we look at all of the thumb placements in the right hand, they fall on C, E, G, B, D, F, and A:

If we order these thumb placements in the same manner as a scale (C-D-E-F-G-A-B), we get this sequence:

We could also play this “scale” descending (C-B-A-G-F-E-D):

Similarly, when descending the chromatic scale, all of the 3rd and 4th finger placements fall on C, Ab, E, Db, Bb, Gb and Eb.

If we order these in the same manner as a scale (C-Bb-Ab-Gb-E-Eb-Db), we get this:

And this:

Despite what they look like, these are actually very easy to play.  For further practice, I’d recommend finding these anchor points in the left hand and experimenting with different orderings of anchor points (besides stepwise motion).  The point here is understanding this process of breaking down scales into hand positions and anchors.

Let’s look at another scale:

Here’s a scale in the key of C Major:

To start, these exercises are played with constant 8th notes (or 16th notes).  Stable chord notes are played on stable beats (similar to practicing be-bop scales). The idea is that with good fingering, the frequencies of chord tones don’t just SOUND stable and consonant, but they FEEL stable and consonant too.

The stable tones make a C6 chord.  Each chord tone is approached chromatically below which all form a B6 chord (This scale can also work over an Am7). For the most part, chord tones are approached from the chromatic tone below.  This scale is for ascending motion.  I’ll introduce a complementary scale for descending later.

With this stream of 8th notes, it’s encouraged that we’re always swingin’.  8th notes are actually quarter + eighth triplets, and the 3rd triplet is accented.  To work on this, put the metronome on 240.  Each ‘click’ is a triplet.  At first, exaggerate that accent on the 3rd triplet.

When ascending the scale by step, there are two fingerings that feel good in the right hand. With thumbs on C and G, and with thumbs on E and A.

Putting the thumb on B can feel right for a moment, but my thumb with naturally switch to E and A.

Left-hand fingering:

One observation I made is that when ascending by step in the right-hand, the thumb is used for cross-overs and my attention is focused on its placement to optimize range of motion.  However, in my left-hand, when ascending by step, that cross-over point is in the 4thfinger. When playing in unison, the hand positions in my right hand are very different than my left hand.

Hand positions in the right hand:

Hand positions in the left hand:

(As we’ll see later, this relationship is inverted when descending. The anchor points are in the 4thfinger in the right-hand, and the thumb in the left-hand)

This is a relationship that I’ve taken for granted and haven’t consciously observed until now.  Playing in unison ‘splits’ my attention in a way, as I focus on the different cross-over/anchor points in my right and left hands.  For example, playing this is easier than playing in unison because the crossover points line up:

Most pianists, myself included, develop this relationship at a young age from practicing scales, hands together, ascending and descending.  It’s actually an interesting hand independence problem that I’ve never observed or deliberately practiced in a jazz, improvisatory context.  Now that I’ve observed it, I may revisit it later.

Moving on….

As seen above, this scale can be observed as having four different anchor points in each hand.  In the right-hand, the thumb is placed on C, E, G and A.  Each of these points have four notes, and can be played 1-2-3-4.

When practicing these kinds of patterns, I often simplify and visualize them in ‘hand clusters:’

In the left-hand, the anchor points are on D#, F#, G#, and B and can be played 4-3-2-1. Keep in mind that in the figure below, the anchor points appear on strong beats.  I did this to better show the hand positions visually.  When practicing, chord tones are played on the downbeat.

Similarly, I visualize these in ‘hand clusters:’

Already we have an interesting variation.  We can create more with different sequences of anchor points. Here are a few:

These anchor points are always spaced four notes apart.  Another possible variation could look something like this:

The fingering above suggests two hand positions: one group of two fingers, and one group of four fingers. After playing with it, I started using the following fingering, which feels more like one group of six:

Up to this point, having the 2nd finger on a stable tone and the 3rd finger on a leading tone is new.  To help maneuverability throughout the scale, it could be useful to isolate this relationship and practice this fingering:

Let’s looks at some left-hand patterns.  We’ll start with creating difference sequences of left-hand anchor points:

As with the right-hand, these anchor points are always spaced four notes apart.  Another possible variation could look something like this, with anchor points spaced six notes apart:

Again, to help maneuverability throughout the scale, it could be useful to isolate the new crossover with the 2nd finger and practice this variation:

Playing the left and right hands together create some interesting fingering problems. This is mainly because crossover and anchor points are different between the two hands (as mentioned above). If we take a closer look at what’s happening theoretically, we can isolate the fingering problem and construct more interesting exercises.

We constructed these two exercises using a certain sequence of anchor points in the right and left hands:

They follow a similar logic of ascending stepwise motion, followed by a descending skip to an anchor point.  But because we’re committed to playing stable tones on stable beats, those descending skips occur at different points in the meter.

In the right hand, skips always land on chord tones, on stable beats:

In the left hand, skips always land on non-chord tones, on weak beats:

Let’s look at the right hand first.

There are two new fingering patterns that can be used to play the above pattern and help with maneuverability through the scale.  The first is using the 2nd finger to lead into the anchor point on the thumb.  The second is using the 5th finger on a chord tone just before skipping down to a non-chord tone.

For practice, it could be helpful to isolate these fingerings:

This opens up a number of variations that include this logic:  if the thumb is on a chord tone, the player can ascend-skip to a non-chord tone and follow the 4-5 finger pattern.  If the 5th finger is on a chord tone, the player can descend-skip to a non-chord tone and follow the 2-1 finger pattern:

Now that the patterns are getting more interesting, so are the ‘hand clusters.’ From the first two examples above:

The highlighted sections in the third and fourth examples above represent subtle fingering exceptions.  The fingering sequence established at the beginning of the pattern could be continued, but I also found that my thumb naturally set on B and created a 3-note cluster (B-C-D#) (1-2-3-).

I also experimented with fingering patterns that had the thumb on all non-chord tones, which include black keys.  For as long as I can remember, I’ve followed the general principle that thumbs don’t play black keys, especially for stepwise motion and small skips.  For large descending skips however, I think exceptions can be made:

In the left-hand, we encounter similar fingering problems.  We know from observing previous right-hand exercises that skips landed on stable beats:

Once again, there are two new fingering patterns that can be used to play the above pattern. The first is using the 5th finger on the stable beat after a descending skip.  Up to this point, stable tones have only been played by the 3rd and 1st fingers. The second is using the pattern 3-2-1 to prepare for the descending skip.

Depending on the length of the skip, I sometimes find myself comfortable using my 4thfinger on the stable tone.  Here are a few variations:

All of these fingerings and hand positions should give a good amount of maneuverability with both hands.  There’s also enough variation to improvise with.

I think this is a good introduction to the kind of approach I’ll be taking.  Of course, there are many pathways I’d like to take for future articles.  They include:

  • A scale for descending motion
  • Applying 3-note groupings to these scales
  • Playing the scale in different keys
  • Incorporating left-hand alternatives (chords, bass lines, comping etc.)
  • Connecting with chords changes (and tunes)

Though I’m focusing on this scale at the moment, I’d be interesting to apply this method to other jazz scales (be-bop scales) and to transcriptions.

Eventually, I’d also like to include video demonstrations with these articles. That depends on getting the right video and audio setup in my studio, which will probably take a few months.

Thanks for reading!

Watch for future updates on these exercises here

Leave a Reply