Analysis for Performance: Teaching a Method for Practical Application

October 2, 2006

One of the more difficult tasks facing the college music theory teacher is linking analysis with performance. Too often students do not understand the connection between the analytical techniques they learn in theory classes and the decisions they make as performers. Even after several semesters of core theory, some students continue to view the subject as dry, arcane—and most unfortunate of all—irrelevant to them as performers. Such students believe that performance decisions should be based on intuition alone. In this article I will describe a college course at my institution for senior undergraduate music students that intertwines both their analytical and performing skills. In this course we teach a method for interpretation based on analysis of the score.1


Preliminary Discussion

To steer our students toward thinking about the relationship between analysis and performance, we begin the course by formulating some basic definitions. We aim to arrive at a general consensus as to the differences between analysis versus theory, and between score versus performance. Students concur that an analysis is an explanation of the structure of a specific piece and various relationships therein. An analysis is a selective view of the piece, wherein the analyst chooses certain details and relations to emphasize. A good analysis reveals what is unique about the piece. In contrast, the term theory is understood to be an explication of general attributes of a body of music or a concept, such as functional tonality. The score functions as the transcription of the composer's original idea or concept, while a performance represents the performer's understanding of the score. Although both the score and the performer attempt to convey the composer's original idea, they are necessarily inexact renditions. The score cannot always capture every nuance of the composer's concept; consequently, the performer cannot always realize exactly the concept contained in the score. Many variables are left to the performer's individual interpretation, variables that can change from one performance to the next.

This understanding leads us to the seminal question of the course. On what basis should performers form their interpretive decisions: on their own musical intuition, on the composer's indications in the score, or on a thorough analysis of the piece? Students will usually acknowledge that they need a combination of all three, but are unsure as to how specifically an analysis can be helpful. It is the goal of this course to show that a thorough analytical investigation of the score will provide ample guidance for developing an interpretation.


Aspects of the Score

While we agree that taste, temperament, intuition, and inspiration all play a role in a performer's approach to interpretation, the composer's intentions may frequently be gleaned from clues in the score. Thus, in the first few classes we focus on the specific aspects of the score itself, determining which are fixed and which are variable, or open to interpretation. The fixed aspects comprise the notated pitches, the durational values, and the tempo, when specified by a metronome marking. The variable aspects include general tempo indications, dynamics, articulation, phrasing, and rubato. The degrees to which these aspects are open to interpretation (exact tempo, the loudness level of a note, the true length of a staccato note) cannot be defined with precision. Nonetheless, the score itself usually provides the performer with the clues necessary to establish the appropriate range of variability.

We encourage students to consider carefully other factors as well. This includes the textural relationship between or among the performing elements. Is the music polyphonic or imitative? For a monophonic texture, does it embed a compound melody? For a homophonic texture, are the melody and accompaniment interrelated? The performer should also consider the implications if a text is involved. When the text and music complement one another, the character of the music is generally quite clear. In other cases, a disconnect may obtain between the text and the music, making interpretive decisions much less straightforward for the performer.

Once we have identified those aspects of the score that are open to interpretation, we begin to chart the steps that will transform a literal playing of the score into a musical and convincing performance. We stress to our students that we will not construct hard and fast rules that will inhibit them, nor will we suggest that there is only one "correct" approach. Our goal is to provide students with the fundamentals on which to make interpretive decisions that acknowledge the details of the score while allowing them to follow their artistic inclinations. The performer who has learned to analyze the score for its interpretive clues will apply appropriate nuances and inflections with confidence, avoiding mechanical executions on the one hand and exaggerated renditions on the other.



The variable aspects of the score are each addressed in turn, beginning with tempo. In the absence of metronome markings, the performer needs to ponder several factors in establishing an appropriate tempo. If the composer gives a general marking, does it refer more to the mood of the piece or its speed? The convention has been to let the fastest notes determine the tempo taken by the performer. For example, if the tempo is too fast, the performer may lose control of the more technically challenging passages. Another risk is that the precision of certain rhythmic figures may be sacrificed; for example, dotted rhythms may begin to sound more like triplets. Faster tempos can easily accommodate broken chords, simple decorations of a pitch, or scalar passages. On the other hand, slower tempos are preferred for non-chordal groupings and non-scalar passages, faster harmonic rhythms, and contrapuntal figures and textures. Range and color of the performing medium also influence the choice of tempo. Even articulation must be considered: a clearly defined articulation (or enunciation for singers) will give the perception of greater speed than one that is less meticulous.

Finally, tempo departures also have a role to play in interpretation; namely, they serve as transitional adjustments between two established tempos. These departures include accelerando, stringendo, ritardando, ritenuto, allargando, among others. Such tempo adjustments are very effective in delineating formal structures and changes of mood or character.



We encourage our students to annotate their scores with symbols denoting their myriad interpretive decisions. Although the Western notational system contains symbols and terms for tempo changes, dynamics, and articulation, it lacks any established symbols for the slight adjustments to tempo that fall under the category of rubato. For this course of study we borrow symbols used by David Blum in his book on Pablo Casals.2 As shown in Example 1a, Blum uses a forward-pointing arrow to indicate "compression," which he defines as a slight quickening or pushing-ahead of short notes. According to Blum, Casals identified a number of situations calling for compression, including dotted rhythms, anapest and dactylic rhythms, and extended series of notes that belong together in a single wave of movement.


Example 1a.



In order to keep symbols consistent, we advise students to use a backward-pointing arrow to indicate a slight holding-back of notes, as shown in Example 1b.


Example 1b.



Such lengthening of notes is often necessary to compensate for a preceding compression, or to indicate the ending of a phrase or unit. In addition, we use the conventional tenuto marking to indicate an agogic accent, or a slight lingering on a note, as in Example 1c.


Example 1c.



We also adopt Blum's use of single slash marks to indicate minute breaks in sound. These breaks, such as those illustrated in Example 1d, help to isolate significant notes.


Example 1d.



In conjunction with our discussion of tempo, we engage in a variety of comparative listening exercises. In one such activity, we play two or three contrasting professional recordings of the same piece of music, while students mark the scores with the rubato symbols described above to reflect the tempo adjustments made by the performers. We then discuss how differences in articulation, texture, and tempo deviations affect their perceptions of the performers' tempo choices.3



The next variable aspect of the score we address is articulation, widely understood as the connection and separation of notes. Students compile a list of the various types of articulation, both general and specific to individual instruments. We examine how changes in articulation can bring about changes in the character of a passage, even though the actual pitches remain the same. We explore the topic of shaping individual notes, observing that the quality of release is almost as important as the quality of the attack. Other issues for discussion include differentiating phrasing slurs from legato slurs, achieving legato on non-sustaining instruments, and the graded levels of separation from non-legato, to portato, to staccato. Moreover, in the absence of any articulation marks, the performer may examine the intervals within the melodic line to suggest the appropriate articulation: legato for stepwise motion, slurs for mid-size intervals, and non-legato for large leaps.

Within the area of articulation, accents receive special consideration. In determining the execution of an accent the performer should factor in the style and period of a work, as well as the specific instrument and the range involved. Even in the absence of an accent marking, certain notes and chords will stand out from the surrounding context for a number of reasons, as listed by Hans Lampl in his chapter on this topic: dynamic, agogic (length), metric (strong-weak), harmonic (non-chord tones), pitch (high), embellishment (ornaments), color, and phrase (beginnings and endings).4



The area of dynamics constitutes one of the most significant variables in the context of interpretation. Performers must constantly assess the effect created from their choice of dynamics, from subtle to extreme, and from gradual to sudden. They must consider how the range of dynamic levels will reflect the general character, style, and historical period of a given work. Other factors to take into account include the acoustics of the hall and the dynamic range of the voice and/or instrument.

As a starting point, we advise students to choose dynamics based on the contour of the melodic line. We follow the generally accepted association of crescendo for ascending melodic lines and diminuendo for descending lines, with the qualification that exceptions to this practice are often desirable. Other situations calling for dynamic expression abound, such as differentiating repeated notes and/or patterns (units), intensifying sustained notes, leading gestures to a climax, and so on. Consequently, we have our students follow a three-fold procedure for assigning interpretive dynamics. First, they study the score for circumstances warranting dynamic shading; second, they add dynamic markings at appropriate points in the score; and third, they must justify their choices by adding annotations next to each marking. We use the following abbreviations for this purpose: C (for contour), CE (contour exception), NR (note repetition), UR (unit repetition), SEQ (sequential repetition), L (long note), and EHP (expressive high point).


Example 2.




In determining how the harmonic vocabulary of the score may influence a performer's interpretation, students must first understand the functions and relationships of individual chords, chord classes, key areas, and the various cadences. We introduce qualitative descriptors to the students' harmonic lexicon, designating modulations as "harmonic turning points" and tonicizations as "harmonic shifts." We label isolated chromatic chords as "color chords" (such as borrowed chords, Neapolitan chords, augmented sixth chords, and so on). Finally, we seek out unexpected or unconventional chord successions and resolutions. Students learn that these chords and events contribute increased tension, dissonance, and color to the music, thereby demanding special attention from the performer. Our students begin their analytic work by assigning Roman numerals and identifying all non-chord tones. Then they study their analyses to locate these harmonic entities, and then determine how best to bring out their function through interpretive nuances.



We contend that students must have a solid understanding of the music's formal structure, from the shorter-range aspects of cadence identification and phrase relationships, to the longer-range considerations of overall compositional design, such as binary form, sonata form, theme and variations, and so on. The performer should also be clear that the variable aspects of the score may be made to either confirm or hinder the perception of the music's formal structure in its various aspects.


Elements for Review

As a prerequisite for success in this endeavor students must be competent in analytical skills. Depending on the constituency of the class in any given semester, it is often advisable to review briefly various topics concurrently with the discussions of interpretive techniques. These topics may include any of the following areas:

  • Cadential formulae: perfect and imperfect authentic, plagal, half, deceptive, Phrygian half
  • Key relationships: relative, closely related, parallel, distant, mediant
  • Non-harmonic tones: passing tone, neighbor tone, escape tone, appoggiatura, incomplete neighbor, changing tone, suspension, retardation, anticipation, pedal point
  • Principles of melodic writing: contour, climax notes, melodic patterns (scalar, chord outlines), presence of non-chord tones, motivic processes
  • Motivic processes: repetition, transposition, sequence, inversion, intervallic alteration, retrograde, fragmentation, extension, and any of these in combination
  • Chromatic harmony: secondary functions, modal mixture, augmented sixth chords, Neapolitan chords, altered dominants, extended tonicizations, chord mutation
  • Modulatory techniques: common chord (diatonic pivot), chromatic, enharmonic, common tone, sequential, direct
  • Aspects of form: phrases, phrase segments, unit repetition, sequences, phrase extensions and interpolations, small and large forms (period, binary, ternary, sonata, rondo, fugue)

If it is determined that certain topics warrant review, we schedule them in small amounts over the first few weeks along with other activities, rather than all at once at the beginning of term in advance of the primary curriculum. In this way we can more easily incorporate the review as part of our interpretive discussions.


Course Activities

The course of study is not lecture-based; rather, it involves a variety of activities, including class discussion, comparative listening, writing essays, analysis, in-class performances, and presentations. In addition, students complete two take-home exams and a major paper. In the first few weeks of the term we focus on diatonic, non-modulating music. Performance decisions are based mostly on moment-by-moment events. We move into chromatic and modulatory music after the first few weeks. At this point, performance decisions are based not only on surface events, but also on underlying structures and compositional designs.

The pieces for study are drawn from the students' own repertoire. At the beginning of the semester students select a few pieces from the common practice period from their studio repertoire. The instructor then organizes these pieces in various categories according to instrumentation, texture, harmonic vocabulary, form, manner of development, modulation, and so on. These pieces will form the core musical examples for the semester, and be revisited several times for different purposes.

Active listening is an integral component from the beginning of the course. In virtually every class students listen to professional recordings, with a variety of related exercises. Even in the very first class meeting the students can participate in an exercise in analytical listening: they listen to a recording of a piece while following a score that contains very little or no expressive markings.5 The students take note of the expressive effects made by the performer, and then try to determine why the artist made those choices. As we progress through the course, the activities expand to include analyzing the score for its melodic, harmonic, rhythmic, motivic, formal, and textural content, and then annotating the score according to the performer's interpretive decisions using symbols introduced in class. An extension of this procedure would include comparing two or three recorded performances of the same piece by different artists, as described above.

As we discuss each of the variable aspects of the score, we guide students in forming and justifying their own interpretive decisions. Students also listen to each other perform in class periodically throughout the term. In a typical situation the student performers first present their interpretation of their piece and explain each decision. Immediately after the performance, the students discuss the degree to which the individual's performance was a successful rendition of their stated interpretive decisions, and whether or not they concur with the interpretation.

Other activities include reading scholarly articles on current thought in the literature (see the select bibliography at the end of this article). Students also complete two take-home exams that present them with pieces to analyze and from which to formulate performance decisions. They are expected to grapple with the interpretive issues that have been covered in class. Finally, in the last weeks of the course, students give a major presentation based on a piece from their own repertoire. The presentation is in three parts: students first give an analytical overview, then discuss their performance decisions, and then they perform the piece. The students then convert this presentation into a major paper due at the end of term.


Procedure for Interpretive Analysis

Once we have moved on to include chromatic and modulatory music in our course, the students are ready to do more detailed work in interpretation. The list below gives the procedure we generally follow. The order of steps and the degree of attention paid each of them may vary, depending on the piece. These steps include:

  • Play or listen to the piece / excerpt
  • Establish the relationship between / among the performing elements
  • Determine the relationships and hierarchies among the phrases and cadences
  • Construct a diagram of the form
  • Complete a Roman numeral and non-chord tone analysis
  • Label all color chords, harmonic shifts, and turning points
  • Trace the motivic processes throughout the piece
  • Locate opportunities for rubato, transitional tempo changes, articulation details, dynamic shadings, and other interpretive aspects
  • Assign interpretive nuances to the score with appropriate markings and symbols


An Analytical Application

One of the many pieces we study is Chopin's Mazurka, Op. 7, No. 2. For the initial listening experience, we hear the piece played in as literal a manner as possible, with a concerted effort to avoid any expressive nuances and inflections. We then analyze the score, following the first five steps of the interpretive procedure outlined above. The students generally make the following preliminary observations: first, the melody is played by the pianist's right hand, while the left hand supplies the chordal accompaniment. The students decide that the left hand part will need to be softer than the right hand, since its role is to support the melody. Second, the class observes that the excerpt comprises two phrases, both ending in perfect authentic cadences, with the first phrase tonicizing the relative major key. They conclude that this excerpt forms a modulating parallel period. This realization will assist the performer in interpreting the phrases in a question-answer relationship.

Although the score contains a few editorial markings, students must make additional annotations indicating their interpretive decisions based on their analyses. Example 3 shows one student's analysis of the piece. First, the student has constructed a form diagram beneath the score. Next, the student provides a complete harmonic analysis of the piece. By labeling the chords with Roman numerals, the student can more readily perceive the harmonic functions of tonic prolongation and dominant preparation. Other significant results emerge from this analysis; the student can now easily locate the color chords, harmonic shifts, and turning points in the music. The student has also identified all of the non-chord tones, and has decided to emphasize the accented dissonances by placing tenuto markings on the suspensions and appoggiaturas. Furthermore, the student has supplied dynamic markings along with abbreviations justifying their addition, based on considerations of melodic contour. Finally, the student has added arrows to indicate points of rubato, since the rhythmic figurations at these points are judged to be suitable candidates for temporal elasticity.


Example 3.




Barra's Graphing Symbol

During the second unit of the course we probe means of projecting longer melodic lines and underlying structure. Donald Barra developed an analytical symbol that lends itself very well to this purpose.6 As illustrated in Example 4, this symbol comprises three parts: an arch, a vertical line, and an inverted arch. The arch reflects the growth phase of increasing energy toward a climactic focal point of highest intensity, the vertical line represents that climactic point, and the inverted arch that follows represents a concluding phase of release or relaxation. The three components make the symbol a flexible, useful tool, since in any given excerpt the growth and release phases may vary in length, and the climactic focal point may occur in the middle, near the beginning or near the end. Another powerful feature of this symbol is its ability to reflect hierarchical structures. That is, these internal phases may also be combined to form higher-level phases of motion. We encourage our students to use Barra's graphing technique to indicate the growth, climax, and release phases of phrase segments, phrases, and entire sections; they are then in a better position to form interpretive decisions that bring out the dynamic nature of the music.


Example 4.





Aspects of Schenkerian Analysis

The time frame for this course's curriculum does not permit delving into Schenkerian analysis in great detail or depth. Nonetheless, we find it beneficial to acquaint students with some basic concepts that are immediately accessible. Our approach is driven by Schenker's contention that tonal music prolongs a tonal harmony in numerous ways and over several levels. We begin with the Schenkerian concept of the Ursatz, the fundamental harmonic structure comprising the tonic chord, tonic prolongation by some means, and an authentic cadence of dominant-tonic. We discuss Schenker's Urlinie, the underlying melodic framework comprising a stepwise descent over the course of the piece to a position of rest at scale degree 1. Students learn to distinguish the three levels of foreground (the actual musical score), background (the Ursatz), and one or more middlegrounds (varying levels of analytical detail). Students practice differentiating among the levels through standard symbols used in Schenkerian analytical notation: stemmed white notes, stemmed black notes, unstemmed black notes, beams, and slurs.7

We employ the following procedure in conducting our simplified Schenkerian analyses:

  • Listen to a performance or a recording of the piece.
  • Provide a complete harmonic analysis, and then bracket all Roman numerals that seem to belong together (in terms of passing or neighbor motions in the bass, for example).
  • On a new sheet of staff paper, notate stemless noteheads for the right hand, omitting repeated notes. Then write out the bass notes of the left hand, omitting upper chord tones.
  • Stem the most important left hand notes (I and V) and slur the remaining notes.
  • Stem the important right hand notes, with one stem in the right hand corresponding with one in the left hand (to provide harmonic support). Those that aren't vertically aligned should be connected with a diagonal line.
  • Determine the first tone (headtone) for the fundamental line, and then locate the remaining tones of the fundamental line (the Ursatz), making certain that each note fits over the prevailing harmony in the left hand.
  • Ensure that all notes have been taken into account, as part of a motion leading to or away from a significant gesture or event.

Example 5 shows a sample analysis that we would do together in class.


Example 5.



The Schenkerian perspective allows students to observe how each note functions at the musical surface while also understanding how a long-range melodic line may travel across phrases and periods. This view allows for more organic unity versus the segmented approach of more form-oriented perspectives. It also allows for a more complex, rich, diverse understanding of phrases and gestures. The excerpt in Example 5, for instance, divides formally into two phrases, clearly delineating an antecedent-consequent relationship of a contrasting period. The Schenkerian graph, however, reveals a cohesive, melodic gesture that extends over the two halves, enveloping them as part of one unified whole. As with their usage of the Barra analytical symbol, students benefit from their Schenkerian analysis by constructing interpretations that take into account both surface events and deeper levels of structure. It allows them to discover the possibility that some gestures in the foreground are embedded in the middle-ground as well. Moreover, students find their memorization skills are greatly improved by the careful study and deeper awareness resulting from their Schenkerian analyses.



In this course students learn to weigh many parameters when working on a piece: their own philosophical approach to performance; a range of contexts for stylistic considerations, historical practices, current thoughts and trends; the fixed and variable aspects of the score; the composer's stated and unstated intentions; and the attributes and limitations of the instruments, artists, and performance venues, among other issues. On evaluation forms at the end of the term, students comment that they find this course challenging, satisfying, and extremely valuable to their musical development. They state that the ability to make informed interpretive decisions on their own is far more artistically fulfilling and rewarding to most students than being limited to imitating their mentors. In sum, this course has proven to be highly effective in educating our future performers and teachers on how analysis complements, and truly enhances, performance.


Select References

Barra, Donald. The Dynamic Performance: A Performer's Guide to Musical Expression and Interpretation. Englewood Cliffs, N.J.: Prentice-Hall, 1983.

Blum, David. Casals and the Art of Interpretation. Berkeley: University of California Press, 1977.

Cone, Edward T. Musical Form and Musical Performance. New York: W.W. Norton, 1968.

Cook, Nicholas. A Guide to Musical Analysis. London: J.M. Dent & Sons, 1987.

Cooper, Grosvenor W. and Leonard B. Meyer. The Rhythmic Structure of Music. Chicago: University of Chicago Press, 1960.

Forte, Allen and Steven E. Gilbert. Introduction to Schenkerian Analysis. New York: W.W. Norton, 1982.

Lampl, Hans. Turning Notes into Music: An Introduction to Musical Interpretation. Lanham, MD: Scarecrow Press, 1996.

LaRue, Jan. Guidelines for Style Analysis, 2nd ed. Warren, MI: Harmonie Park Press, 1992.

Lester, Joel. Bach's Works for Solo Violin. New York: Oxford University Press, 1999.

Meyer, Leonard B. Emotion and Meaning in Music. Chicago: University of Chicago Press, 1956.

Neumeyer, David and Susan Tepping. A Guide to Schenkerian Analysis. Englewood Cliffs, N.J.: Prentice-Hall, 1992.

Rothstein, William. "Analysis and the Act of Performance." In The Practice of Performance: Studies in Musical Interpretation, ed. by John Rink, 217-40. Cambridge: Cambridge University Press, 1995.

Salzer, Felix and Carl Schachter. Counterpoint in Composition. New York: Columbia University Press, 1969.

Schenker, Heinrich. Free Composition, trans. and ed. Ernst Oster. New York: Longman, 1979.

. The Art of Performance. Edited by Heribert Esser and translated by Irene Schreier Scott. New York: Oxford University Press, 2000.

Schmalfeldt, Janet. "On the Relation of Analysis to Performance: Beethoven's Bagatelles, Op. 126, Nos. 2 and 5." Journal of Music Theory 29/1 (1985): 1-31.

Stein, Deborah, ed. Engaging Music: Essays in Music Analysis. New York: Oxford University Press, 2005.

1We do not require a textbook for this course. Instead, we draw on material from the sources cited in the list of references at the end of this article, primarily those by Lampl, Turning Notes into Music; Blum, Casals and the Art of Interpretation; and Barra, The Dynamic Performance.

2Blum, Casals and the Art of Interpretation, 69-100.

3Ideally, the instructor would pair recordings of important works representing the vocal or instrumental forces of the class members. To illustrate, some of the pairings we have used include: (1) "Die Wetterfahne" from Schubert's Die Winterreise, D. 911, with recordings by Hakan Hagegard, baritone and Thomas Schuback, piano (RCA Red Seal Digital, ARCI-4523, 1982) compared with Hans Hotter, bass and Gerald Moore, piano (Seraphim, S 60025, 1966); (2) the exposition from the first movement of Beethoven's Symphony No. 8 in F, Op. 39, with recordings by the London Classical Players with Roger Norrington, conductor (EMI, CDC 476982, 1987) compared with the Chicago Symphony Orchestra with Sir Georg Solti, conductor (CSP-9, CS-6926, 1975); and (3) the first movement of Brahms's Clarinet Sonata in F minor, Op. 120, No. 1, with recordings by David Shifrin, clarinet and Carol Rosenberger, piano (Delos 3025, 1988) compared with Charles Stier, clarinet and William Bloomquist, piano (Elan CD 2224, 1989).

4Lampl, Turning Notes into Music, 79.

5One piece that works well in this context is Schubert's "Nacht und Träume," Op. 43, No. 2, D. 827. There are numerous recordings available: we use the performance by Dietrich Fischer-Dieskau, baritone and Gerald Moore, piano (Deutsche Grammophon Catalog #363502).

6Barra, The Dynamic Performance, passim.

7There are numerous resources the interested reader may consult for further information about Schenkerian graphing symbols, including Forte and Gilbert, Introduction to Schenkerian Analysis; Neumeyer and Tepping, A Guide to Schenkerian Analysis; Salzer and Schachter, Counterpoint in Composition; and Schenker, Free Composition.

7769 Last modified on October 4, 2018