One of the things I love about playing a piece over many years is that after I've grown into it, it grows with me.

Many people think of musicians' work as endless mechanical repetition. In reality, mechanical work is fatal to music-making. We learn what we do, not what we think or hope we're doing, nor what we intend to do later. If we play with no emotion, we learn to play with no emotion. Calling some of the playing "practice" and some "performance" doesn't change the fact.

The exact emotional tone and progression of a piece take time and thought to work out. All the contributing elements—notes, dynamics, tempos, rhythms, articulation, sonority—have implications. Is a contrast between connected and disconnected notes important to the meaning, as in the second movement of this sonata? Then one must honor the contrast wherever it occurs. One may connect notes by finger-action or by using the pedal; but the pedal also changes the sonority. If the pedaled sonority goes against one's sense of the piece, then one must use finger-connections. But what if that's not possible in a particular case?

Or take dynamics. The short musical idea at the opening of the sonata is marked "very loud" (fortissimo). At its return late in the movement, it is just "loud" (forte), but a few bars later, "very loud" again. It reappears in the middle of the fourth movement with a stunningly different emotion, a reflective flavor realized partly through a "medium-loud" (mezzoforte) dynamic. And when the same musical idea, slightly elaborated, ends the entire sonata, it's "ultra-loud" (fortississimo). Much thought and work is needed to scale these dynamic relations accurately over a 26-minute span while still subordinating them to a fluent and convincing flow of feeling.

Tempo relations must be scaled, too. The middle section of the third movement is marked to be faster than the opening of the movement, but if this middle section is hurried, it misses a certain luxuriating quality which I think it should have. Of course it's easy enough just to play the opening slower yet, and indeed Prokofiev calls for the beginning to be at "slowest waltz tempo". But the waltz must live and breathe, not slouch like the end of a dance marathon.

All the various elements of a performance—dynamics, tempi, articulation, sonority, and more—must relate to each other and to the emotion which is the goal. The inter-relationships are so complicated that, in the great works, analysis can only begin the working-out process. I think this is what the great pianist Artur Schnabel referred to when he said that he was interested only in playing "those works which could never be played well enough."

 
Conscious analysis can help begin the interpretive work, but to continue it, I try to hear with the composer's ears, to feel what he felt. I dream the piece intentionally; and I experiment, too, imposing my passing moods on it. I play it when I'm happy, sad, angry, depressed, tense, relaxed, joyous, despairing, sentimental, impatient, loving or hating, superficial or intense. I put my feeling of the moment into every note. Mostly, this teaches me what does not work, but sometimes there is a place where the feeling fits. Then I've learned something forever.

But even such total mobilization of myself is not enough. After a point, to hear with fresh ears and to experience the work as emotional communication, I must perform for an audience. After many performances and two or three re-learnings of the piece, it's mine; I have grown into it.

Then it starts growing with me. Where the initial learning is dominated by the conscious, this process is a settling-in that only the unconscious can do, quietly and by itself. Sometimes I think that the whole point of the first problem-solving attack on a work is simply to draw the attention of one's unconscious to the important issues.

The growing-with process is difficult to detail, just as it's hard to remember the steps by which you came to know someone intimately, or by which a book read ten times came to seem straightforward, when at first it appeared complex.

Growing-with benefits from one's general musical maturing, too. In the ten years since I first performed this piece, my playing has changed, and the sonata with it. My normal tone of voice at the keyboard is softer than it was; my touch, less percussive. Dynamic range is wider, and shading more flexible.

Hearing Prokofiev's own playing for the first time recently, on a record of some of his other works, started me thinking, too, about what really creates convincing playing. Prokofiev plays as though he not only wrote the music but invented the piano, which he will now take apart and re-assemble for your amazement and edification. He makes almost all mere pianists sound like cliché-mongers.

 
The slow, wonderful growing-with process is familiar to me now from many pieces over many years. So I was startled, when I began practicing the Prokofiev for this concert, to find that it had changed radically in the eight months since I last performed it. It came out of my hands more fluent, more emotional, more subtle; and with a "going" quality I've been trying to achieve for a long time. All this happened without my being aware of any particular growth in myself or my playing.

Is something developing without my knowing it? Did the impact of Prokofiev's playing make the difference? Or is it that I intentionally did not re-study the work from the ground up, but took for granted and enjoyed my command of it?

Whatever the cause, this sudden change after ten years of living with the sonata is a new experience. It feels like an important message from deep inside.

—James Boyk