A Multi-Level Approach to More Secure Memorization

  • PDF: https://www.jstor.org/stable/41225253

Memorizing is essential for the solo pianist, thanks to a long tradition dating back to the nineteenth century. Requiring solo piano recitals to be performed from memory, however, is not without controversy. While memorizing comes naturally to some musicians, others struggle to at least some extent. Opponents of memorized performance argue that the mere fear of a memory slip can impede musicality. In extreme cases, a memory slip can psychologically paralyze a performer. How should a struggling pianist deal with unreliable memory? And what assistance should the piano teacher offer? Memorization is a skill not often taught, yet much discussed in professional journals, books, and DVDs. (See the annotated bibliography at the conclusion of this essay.) Some of these sources emphasize a specific approach, for example, assuming a semi-conscious Zen-like state, focusing on a single musical parameter (such as form or harmony), or relying on psychological processes.1 Most authors, however, endorse a more general multi-sensory approach, engaging four main memory types: tactile or kinesthetic, visual, aural, and analytical. In this article, I offer a strategy for securing memory that engages the physical, visual, aural, and intellectual, while integrating theory and performance. The effectiveness of this approach lies in its assigning different memory types specific responsibilities in the memorization process. I will explain my own strategy, drawing on concepts from linear analysis, provide a brief example of its application and report on progress made by my own students.

Left on their own, students too often overly rely on kinesthetic memory. This often involves entering into a mesmerized state, shutting off the logical-rational part of the brain. Although effective in the practice room, it can prove fatal in the concert hall. Under the pressure of a memorized performance, the brain “goes into overdrive,” with each memory type on heightened alert, insisting on fully participating in the task at hand. Memory types that have not been “programmed” in practice can only participate in a misdirected capacity, instilling fear and insecurity in the pianist. They taunt the performer: “I’ll bet you don’t know the next note!” While there is a general consensus that kinesthetic memory is essential, it is, by itself, unreliable. All memory types must be involved in the memory process.

In order to assign memory types unique roles, we must acknowledge that we process and perform music at many levels. It seems only logical that we should memorize unique levels differently. All memory types can be assigned a principal level at which to work and a specific job to do at that level. In assigning levels, I draw on Heinrich Schenker’s theory of tonality, which is rooted in performance. He disseminated his ideas in private piano lessons rather than theory classrooms. In his day he was known as a piano teacher, critic, and composer rather than the iconic theorist we think of today. His book, The Art of Performance, is based on the interconnectedness of theory and performance. In it he states, “. . . A superficial acquaintance with the work of art is insufficient. What is essential is a thorough knowledge of all laws of composition. Having enabled the composer to create, these laws, in a different way, will enable the performer to recreate the composition” [emphases mine].2 Understanding Schenker’s laws of tonality even superficially A Multi-Level Approach to More Secure Memorization allows one to manage the role of each memory type by setting up a division of labor based on the way music inherently works.

Schenker recognized there are multiple levels in music; that is, music is structured hierarchically. He identified three main levels:

  • The foreground includes the musical surface, beginning with note-to-note connections.
  • The middleground reduces out chords of lesser structural importance. Performers think in terms of middleground connections when they play or sing “through a phrase” or “to a goal.”
  • The background is an important, more abstract level, guiding the overall coherence of a piece.

There are additional structural levels between each adjacent pair. I will focus on levels of which performers are immediately aware: the surface foreground, the deeper foreground, which reduces out non-chord tones, and the lower middleground, which reduces out embellishing (non-functional) chords.

Schenker also recognized that tonal structure is recursive. The interdependency of harmony and melody create a limited number of voice leading rules that are shared by all levels. According to Schenker, tonal progressions at all structural levels are based on one of three voice leading prototypes. The two most common are presented in Figure 1, a and b. (Of course these prototypes may be made incomplete by temporarily concluding on the dominant harmony.) Prototypes are fleshed out by a limited number of voice leading transformations to arrive at the musical surface. For the pianist, the benefit of recognizing prototypes and transformations is the ability to memorize in musical gestures rather than individual notes. (Discussion of voice leading transformations follows shortly.)

Figure 1. Schenker’s more common Tonal Prototypes.

dickinson fig1


I recommend that my students begin the process of memorization at the phrase level by writing out rhythmic reductions using notes of voice-leading prototypes and representative notes from transformations. They then commit these reductions to visual memory. Placing transformation labels between reduced notes cues analytic memory to provide connections between visually stored pitches and the musical surface. Kinesthetic memory fills in surface pitches, while aural memory recalls surface rhythms, dynamics, and articulation. Eventually, reductions may be further condensed with visual memory responsible for higher level prototypes and transformations, analytic memory filling in larger-spanning transformations, and aural memories anticipating larger-level connections.

Familiarity with voice-leading transformations is crucial to reducing the musical surface to a limited number of pitches that can be memorized visually. Transformation types are few in number and usually not difficult to recognize. In his book Explaining Tonality, Matthew Brown categorizes Schenker’s voice leading transformations into four groups.3 I will draw from two of these. The first group he calls “horizontalizing transformations” because they involve melodies that are made up of horizontalized (arpeggiated) chords. Example 1, the opening of Mendelssohn’s Scherzo in E Minor, illustrates repetition, one of the simplest ways to extend a tone or chord in time. When a pitch is repeated, there is no need to visualize each printed note. Rather visualize the first pitch and rely on aural memory for the repetition pattern. A reduction printed below the passage in this and other examples represents a model for visual memorization. Example 2, the opening of Beethoven’s Piano Sonata in E-Flat, Op. 81a, provides an example of register shift, merely repeating material in a different octave.4 In such an occurrence, there is no need to remember specific notes of both appearances. In Example 3, arpeggiation underlies a passage from the first movement of Beethoven’s “Moonlight” Sonata, Op. 27, No. 2. Here it is easy to store underlying blocked chords in analytic memory and rely on kinesthetic and aural memories for the surface detail. Example 4, the opening of Schubert’s Impromptu in B-Flat, Op. 142, No. 3, illustrates unfolding.

Example 1. Repetition in Mendelssohn, Op. 16, No. 2, mm. 1-4.



Example 2. Register Shift in Beethoven, Op. 81a (I), mm. 1-5.



Example 3. Arpeggiation in Beethoven, Op. 27, No. 2 (I), mm. 32-40.


Example 4. Unfolding in Schubert, Op. 142, No. 3, mm. 1-2.

dickinson fig1


This transformation allows melodic lines to be reduced to progressions of harmonic intervals. In this excerpt, the reduction condenses the compound melodies in soprano and bass into a four-voice block texture. Example 5, the opening of Beethoven’s Piano Sonata in A-Flat, Op. 110, begins with several instances of voice exchange. This transformation


Example 5. Voice Exchange in Beethoven, Op. 110 (I), mm. 1-5.





occurs when pairs of outer voices are switched. One need visualize only the first pair. Example 6 returns to Beethoven’s Piano Sonata, Op. 81a. The melody in the second measure initiates an arpeggiation of a C Minor triad, which extends through m. 4. Circled notes “reach over” notes of the triad, each overshooting the next chord tone by a step, then quickly “correcting” itself. In this instance of the transformation reaching over, one need visualize only the underlying arpeggiation.


Example 6. Reaching Over in Beethoven, Op. 81a (I), mm. 1-4.



Brown’s next group, “filling-in transformations,” occur when the notes of arpeggiated chords are filled-in by step. Example 7 contains an excerpt from Mozart’s Piano Sonata in A, K. 331. In measures 1-2, the transformation neighboring motion produces a step between repeated tones in the first half of mm. 1 and 2 (arpeggiation completes the measures). One need visualize only downbeats of these measures. In Example 8, the same excerpt is further reduced, using the transformation linear progression. Linear progressions simply fill in the space between any two triadic intervals—here between melodic thirds in mm. 1-3 and 3-4. This transformation often occurs in two voices simmultaneously in the form of parallel tenths, as in this passage. By relying on analytic memory to recall transformations, one need visualize only the basic prototype. In Example 9, from Chopin’s Prelude in Db, Op. 28, No. 15, a Db Major triad is arpeggiated into an inner voice Ab in m. 1. The Ab then ascends through harmonized passing tones Bb and C to the Db in m.2, beat 4 in an instance of motion from an inner voice. A blocked chord is a sufficient visual cue. (The related transformation motion to an inner voice results from descending stepwise motion to a lower voice.)


Example 7. Neighboring Motion in Mozart, Piano Sonata K. 331 (I), mm. 1-4.




Example 8. Linear Progressions in Mozart, K. 331 (I), mm. 1-4.




Example 9. Motion from an Inner Voice in Chopin, Op. 28, No. 15, mm. 1-4.



In addition to familiarity with these transformations, the solo pianist would do well to be familiar with strict resolutions of dissonant intervals for further memory security. Resolution of the tritone (both as diminished fifth and augmented fourth) and the minor seventh are shown in Figure 2. One should also be comfortable with the concept of polyphonic, or compound melody.

Figure 2. Strict Resolution of Dissonant Intervals.

dickinson fig2

I recommend a six-step process for my students when learning a piece for memorized performance. The first half of the Menuet in G Minor from the Notebook for Anna Magdelena Bach, serves as an illustration (Example 10).

Example 10. Menuet from the Notebook for Anna Magdalena Bach with Reductions.














































STEP 1: Get to know the piece with the score. In this step, the pianist is accomplishing three things: learning the piece, preparing a reduction, and memorizing the musical surface kinesthetically and aurally. In learning the piece, thoroughly acquaint yourself with pitches, rhythms, articulations, expressive indications, dynamics, phrasing, pedaling, interpretation, etc. Work out technical challenges and break down the piece into formal sections, key areas, and phrases. The Menuet example consists of regular four-measure phrases making up the first half of a binary form, concluding with a modulation to the major mediant. (Example 10, levels a and d reproduce the score with levels b and e supplying initial reductions. Level c further reduces the first phrase.) In preparing a work for reduction, search for prototypes and transformations in each phrase. Circle notes and label chords of the prototype. The clearest prototype in Example 10 occurs in the fourth phrase, mm. 13-17. Here the stepwise descent of a fifth F—Eb—D—C—Bb coincides with the concluding perfect authentic cadence in B-flat. Notice transformations between notes of the prototype and label each. While the first and third phrases of the Menuet feature linear progressions (LIN), arpeggiations (ARP), and register shifts (REG SHIFT), the second and fourth phrases make use of neighboring motion (N) and motion to/from and inner voice (MIV).

By the sheer amount of repetition required to get a piece “into your fingers,” you are kinesthetically memorizing fingering, hand placement, stretches, shifts, and lines moving between hands at a subconscious level. Get the piece “into your ear” by aurally memorizing melodies and rhythms, both of which you should be able to sing with indicated dynamics and articulation.

STEP 2: Make a reduction of the piece/passage. On staff paper, write out a reduction of each phrase using notes from the prototype and transformations. Adjust the rhythmic value of each note in the reduction so that it extends until the next note. Do not feel obligated to work linearly. If you find a particular passage challenging, leave a blank space and fill it in later. Place transformation labels between notes of the reduction. Keep in mind that you will probably need to make a reduction of your initial reduction since it may contain more detail than you can memorize visually. (Example 10c illustrates a secondary reduction of 10b.) Replace lower level pitches with transformation labels and commit to analytic memory.

STEP 3: Reading the reduction, play the piece according to the score. Rely on your fingers and ears to fill in the surface. Away from the piano, hear the piece according to the score, looking only at the reduction. If you forget notes, rhythms, fingerings, articulations, etc., return to the score immediately.

STEP 4: Memorize the reduction. You may find it helpful to write it out away from the piano.

STEP 5: Play the piece from memory as it is written, visualizing the reduction.

STEP 6: Forget everything above! In performance, visual and analytic memories are best served as emergency back-ups as you devote your mental energies to expression, emotion, communication and interpretation.

This memorization system is not for everyone, as memorizing is a very personal endeavor. I use this process for entire pieces, particularly dense Bach fugues. Others may find it helpful for only problematic sections. In using this system with my own piano students, I encourage them to adapt it to their unique learning styles. The following four cases illustrate this point.

Seth was secure in his memorization of a Two-Part Invention by J. S. Bach, with the exception of a four measure passage two-thirds through the work. During his lesson, we studied the voice leading in the passage and wrote out a rhythmic reduction. By the next week’s lesson, he had memorized the reduction and was able to explain transformations occurring within the passage. He then performed the passage securely by memory.

Dan was at a loss as how to approach his first attempt at memorization, a short binary dance movement. I explained to him the concepts of polyphonic melody, prototypes, and transformations and helped him find occurrences in his piece. Primarily a visual thinker, he found the visualization aspect most helpful in organizing a task that at first seemed impossible.

My approach was unsuccessful with Lizzie, a self-described kinesthetic learner. I presented my strategy to her with the intention of not only providing her with an aid for memory, but encouraging her to play with a greater sense of continuity. While my strategy served the latter purpose, she admitted that she prefers to rely on tactile memory alone for the former. Her memorized performances only occasionally suffer from inconsistencies.

Rachel, who had never suffered from memory problems, reached a crisis point while working on a Chopin prelude. We discussed nested prototypes and transformations in one-third of the prelude and wrote out a reduction using colored pencils to distinguish different transformations. She was able to regain her confidence and performed the piece on her jury with no memory problems.

By combining four memory types and defining the role of each, memorizing can be made more secure. Analytic and visual memories are responsible for lower middleground prototypes and transformations. Kinesthetic memory is responsible for filling in surface pitch detail. Aural memory recalls surface rhythm, dynamics, and articulation, and reinforces pitch. This approach is actually intuitive, drawing from simple theoretical concepts that can be used at any ability level. In addition to offering a method of more secure memorization, this levels-based approach can serve to inform one’s interpretation by encouraging identification of musical goals and promoting deeper listening. In the process, students naturally come to make a connection between analysis and performance.



Proponents for Memorized Performance:

Bernstein, Seymour. With Your Own Two Hands. New York: Schirmer, 1985.

Argues that a work cannot be fully known unless it is memorized.

Johnson, Rebecca Grooms. “What’s New in Pedagogy Research? Memorized Versus Nonmemorized Performances.” American Music Teacher 49, no. 4 (2000): 84.

Reports on research confirming that audiences are more responsive to memorized performances.

Critics of Memorized Performances

Brandfonbrener, Alice G.“Memorization: A Learned Skill or an Inborn Talent?” Medical Problems of Performing Artists 16, no. 3 (2001): 83-84.

Montparker, Carol. “Carillon.” Clavier 40, no. 10 (2001): 47.

Both Brandfonbrener and Montparker argue that memorization is a skill separate from musicianship.

Memorization Approaches Emphasizing A Specialty Approach:

Douthit, James and Amy Stanley. “Reinventing the Inventions.” American Music Teacher 53, no. 1 (2003): 37-40.

Douthit and Stanley recommend analysis as an aid in memorization.

Elson, Margaret. Passionate Practice: The Musician’s Guide to Learning, Memorizing, and Performing. Oakland, CA: Regent Press, 2002.

A practical manual presenting several “meditative” exercises in preparation for memorized performance.

Imreh, Gabriela. “Understanding and Developing Musical Memory: The Views of a Concert Pianist and Cognitive Psychologist.” American Music Teacher 46, no. 3 (1996-97): 20-24, 67.

Draws on psychological principles in her approach to memorization.

Reubart, Dale. Anxiety and Musical Performance. New York: Da Capo, 1985.

Reubart reports that pianist Anton Kuerti feels that a successful performance is one in which the pianist transcends the details of the music with the music taking on a will of its own.

Shafer, Timothy. “Forming Your Teaching to the Teaching of Form.” American Music Teacher 53, no. 6 (2004): 18-20.

As the title implies, Shafer endorses an emphasis on form in memorizing.

Shockley, Rebecca Payne. Mapping Music: For Faster Learning and Secure Memory. 2nd ed. Middleton, WI: A-R Editions, 2001.

An interesting and valuable approach emphasizing visual (non-notational) “mapping.”

Turner, Myra Brooks. “Following Harmonic Progressions, Thumbprints of Great Composers.” Clavier 41, no. 10 (2002): 6-9.

Turner focuses on harmonic progressions as a guide to memorizing.

Memorization Approaches Emphasizing A Multi-Sensory Approach:

Bryant, Celia Mae. “A Clavier Classic: Memorizing—A Science.” Clavier 38, no. 8 (1999): 28-32.

This reprint from a 1963 Clavier article suggests using “analytical, aural, tactile and visual aides” to improve memory.

Cooperstock, Andrew. “Reviews: Videos—Stewart Gordon: ‘Memorization in Piano Performance.’” American Music Teacher 54, no. 5 ( 2005): 97-98.

In a video-recorded lecture, Gordon, chair of keyboard studies at USC, recommends a multi-sensory approach.

Greer, Amy. “Rattling the Cage.” American Music Teacher 53, no. 5 (2004): 32-35.

Greer advocates preparing a “mindful performance” by resisting the temptation to perform on “autopilot.”

Kopfstein-Penk, Alicia.“Memorization for String Players.” American String Teacher 47, no. 4 (1997): 51, 53-57.

This author of several articles on memorizing emphasizes using mental and muscle memories to reinforce each other; suggests steps for memorizing using five types of memory.

Street, Eric. “Bridging the Gap Between Sight Reading and Memorizing.” The American Music Teacher 36, no. 6 (1987): 32-33.

Lists practical ideas for memorizing using different memory types.

Studies Relating to Schenkerian Analysis:

Brown, Matthew. Explaining Tonality and Beyond. Rochester: University of Rochester Press, 2005.

Explores the scientific viability of Schenker’s theory of tonality through Schenker’s transformation of counterpoint and functional tonality in the interest of theoretical accuracy.

Cadwallader, Allen and David Gagné. Analysis of Tonal Music: A Schenkerian Approach, 3rd ed. New York: Oxford University Press, 1998.

A comprehensive introduction to linear analysis.

Salzer, Felix and Carl Schachter. Counterpoint in Composition. New York: McGraw-Hill, 1969.

An excellent discussion of the relationship between horizontal and vertical aspects of tonal composition. Includes many rhythmic reductions.

Schenker, Heinrich. The Art of Performance. Edited by Heribert Esser and translated by Irene Schreier. Oxford: Oxford University Press, 2004.

Schenker discusses all aspects of performance, including technical issues and the relationship of analysis and performance.



1Please consult the annotated bibliography for representative sources of each approach.

2Schenker, The Art of Performance.

3Brown, Explaining Tonality.

4Whereas Brown limits his definition of register transfer to “octave leaps . . . projecting a tone from one register to another” (77), I have extended the use of this term to include entire passages repeated in a higher or lower octave.

Read 3325 times

Last modified on Monday, 01/10/2018

Stefanie Dickinson

Stefanie Dickinson is Assistant Professor of Music Theory at the University of Central Arkansas. She holds degrees in piano performance from the University of Georgia (BM) and Auburn University (MM) and in music theory from Northwestern University (MM) and the Eastman School of Music (PhD). Her primary areas of research include the music of Liszt’s late experimental period, issues in analysis and performance, and music theory pedagogy. She has presented her work at regional and national meetings of the Society for Music Theory and The College Music Society, and at international meetings of CMS, the Dutch—Flemish Society for Music Theory, the 12th Biennial International Conference on Nineteenth-Century Music, the International Conference on Music and Gesture, and the First National Symposium of Musical Analytics in Shanghai. Her articles can be found in GAMUT, College Music Symposium, and Liszt 2000: The Great Hungarian and European Master at the Threshold of the 21st Century, published by the Hungarian Liszt Society in honor of the millennial anniversary of the state of Hungary.

Go to top