The Craft of Tonal Counterpoint
The Craft of Tonal Counterpoint
The Craft of Tonal Counterpoint
Second Edition
Thomas Benjamin
Published in 2003 by
Routledge
29 West 35th Street
New York, NY 10001
www.routledge-ny.com
All rights reserved. No part of this book may be reprinted or reproduced or utilized in
any form or by any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or retrieval system, without permission in
writing from the publishers.
10 9 8 7 6 5 4 3 2
p. cm.
MT55.B446 2003
781.2’86—dc21
2003012527
Contents
Acknowledgments
Preface to the Second Edition
Notes to the Instructor
Notes to the Student
Introduction
Chapter 1 Line and Other Elements of Style
Shape; Tonal Framework; Range and Tessitura
Structural Pitches
Filling In (Motivic) Figures
Phrase and Cadence
Cadence Figures
Meter and Rhythm
Melodic Intervals; Details of Line
Scales; Chromaticism; Tendency Tones
Nonharmonic (Nonchord) Tones
Means of Coherence
Compound Line
Suggestions for Melodic Writing
Melodic Writing Checklist
Chapter 2 Nonimitative Two-Voice Writing
General Observations
Specific Details of Voice Relationships
A Note on Compound Meter
Contrapuntal Analysis Checklist
Essentials of Two-Voice Counterpoint
Chapter 3 Chromaticism in Two Voices
Nonfunctional (Nonessential, Decorative, Melodic) Chromaticism
Functional (Essential, Harmonic) Chromaticism
Harmonies Related to Chromatic Lines
The Neapolitan Triad
Augmented-Sixth Chords
Chapter 4 Composition of Binary Dance Forms
Chapter 5 Double (Invertible) Counterpoint
Double Counterpoint at the Octave or Fifteenth
Double Counterpoint at the Twelfth
Double Counterpoint at the Tenth
Chapter 6 Imitation; Canon
Imitation
Canon
Variants in the Imitative Process
Chapter 7 The Two-Voice Invention
The Exposition: Theme and Countertheme
The Episode
The Invention as a Whole
Analysis of a Complete Invention
Chapter 8 Three-Voice Counterpoint I: Texture, Rhythm, Harmony
Texture and Rhythm
Range and Spacing
Relative Motion
Harmony
Cadential Figures
Nonharmonic Tones
Chapter 9 Three-Voice Counterpoint II: Chromaticism, Triple
Counterpoint, Canon
Chromaticism
Cross-Relation
Triple (Triple Invertible) Counterpoint
Accompanied Two-Voice Canon
Canon in Three Voices
Chapter 10 Fugue I
The Subject
The Answer
The Exposition
Chapter 11 Fugue II
Overall Structure
The Episode
Middle Entries
The Counterexposition
Augmentation and Diminution
Inversion
Stretto
Pedal Point
The Ending Section
Analysis of a Complete Fugue
Chapter 12 Four-Voice Counterpoint
Texture and Rhythm
Harmony
Four-Voice Fugue
Other Fugal Variants
Chapter 13 Variation Forms
The Passacaglia
The Chaconne
The Goldberg Variations (Clavierübung, Part IV)
Chapter 14 Cantus Firmus Procedure: The Chorale Prelude
The Ornamented Chorale Harmonization
Cantus with Motivic Counterpoints
Canonic Treatment of Cantus and/or Accompanying Parts
Chorale Prelude Involving “Vorimitation” (Preimitation)
Chorale Prelude with “Obbligato” Melody
Conclusion
Appendix 1: Harmony
Appendix 2: Composing for the Organ
Glossary
Bibliography
Anthology
Index
This book is gratefully dedicated to my supportive and witty colleagues in
theory and composition at the University of Houston’s School of Music, the
Peabody Conservatory of the fohns Hopkins University, and to a few special
friends who have been of more help to me over the years than they will
probably ever know, first among them my dear wife and best friend, Carol.
Acknowledgements
In the roughly seventeen years since the first edition of this book appeared, a
great deal has changed in the field of music theory, and several intelligent and
interesting new textbooks for the study of tonal counterpoint have appeared.
At the same time, the continuing popularity of this text, as well as the need to
update its explanations and exercises, has encouraged the author to undertake
a revision.
My original reasons for writing this book still seem valid. Of the several
other texts in this field currently available, all seem, while in many ways
estimable, problematic as regards their pedagogy. Some strike me as
insufficient in the ordering, rigor, or comprehensiveness of their exercises;
crucial conceptual or skills-developing steps may be omitted. Other texts
seem unfocused and diffuse in terms of musical style: works by vastly
disparate composers are thrown together in a way that seems likely to confuse
the student. Still others appear to be aimed at the already expert theorist of
music, and are so tied to specific metatheoretical concepts and language that
many students will be sure to find them mystifying.
The danger of such approaches is to complicate technical and stylistic
matters by wrapping them in theoretical constructs, reinforcing the tendency
of most music students to assume that “music theory” is an abstruse and
mystical discipline divorced from their own experience of music—in other
words, that “music theory” is an undertaking entirely separate from musical
practice, and from their own lives and needs as practicing musicians.
The attempt in the current edition is to construct a book that the intelligent
student can read and understand, with some help from the instructor. It
focuses on the work of Johann Sebastian Bach, and aims to be practical and
useful as a study of both style and technique through performance, listening,
analysis, and writing. It reflects current thinking about tonal music in its
emphasis on structural harmony and line, structural-pitch reduction, linear-
intervallic patterning, and so on, but presents these concepts in what I hope is
a practical and immediately applicable way, as free as possible from abstract
theorizing, and always focused on the music itself.
Those students and instructors who wish more background information on
Bach’s life and output and other issues of music history, theory, and literature
are referred throughout the text to appropriate resources.
Notes to the Instructor
This book seeks to provide a great deal of fine music for performance,
listening, analysis, and model making; straightforward explanations of
technical and aesthetic matters; and a multiplicity of exercises in listening,
analysis, and writing that are both technically sound and musically satisfying.
The approach taken here is style-specific; the examples are drawn from the
instrumental music of Johann Sebastian Bach, especially his keyboard music.
But the techniques covered transfer easily to other media within the Bach
style, at the instructor’s option. It is, of course, also true that composers of
Bach’s time considered performing media to be to some extent
interchangeable. The emphasis on the music of one composer allows this text
to be used not only as a study of contrapuntal technique but also as a model of
style analysis that is procedurally transferable to the music of other composers
and historical periods. The instructor is free to focus as much as the course
goals allow on matters of style in addition to the more purely technical
aspects (to the degree to which style and technique can in fact be
distinguished from each other).
This text is based on a study of musical practice, and the student is directed
in the analytical exercises to deal with the music itself. It is clear that any
meaningful study of music must be solidly based on experience in listening,
performing, analyzing, and writing; it is urged that each class period include
performances—recorded or live—of the music under study, and a discussion
of what has been heard, including reference to issues of music history and
literature, and performance practice. Some instructors may even wish to
combine this text with a survey of the music of Bach, and listening
assignments and tests would be most appropriate. As we all know, technical
studies tend to be bloodless and abstract unless constantly connected with
musical experiences; such connections, made explicit by the instructor, will
tend to validate and clarify these studies.
On a more specific note, this book makes no use of a strict species
approach. A multiplicity of analytic and written exercises provides ample
practice in technique within a realistic musical context. In the interests of ease
of reading, modern clefs are used throughout. The explanatory material is
written as simply and practically as possible. Considerable generalization as
regards both style and technique is inevitable, though, and I hope that in the
search for directness and comprehensibility no outright errors have been
made. It should be made clear to the student, in any case, that this book
presents a picture of Bach’s most typical practices (and, by extension, of the
other masters of tonal counterpoint), and that there are exceptions to these
practices. The language chosen is intended to be as positive as possible and
avoids burdening teacher and student with a multiplicity of “rules” and
prohibitions beyond those that seem unavoidable. It is urged that the word
rules be avoided, and that such terms as practices or typical procedures be
used.
To be properly prepared for this study, the student should be somewhat
conversant with common practice theory, including basic four-voice part-
writing principles and practices.1 It will save class time if students know
figured bass symbology, chord nomenclature, the basics of voice leading
(including doubling), and principles of harmonic progression. According to
the reports of Johann Philipp Kirnberger and others, Bach in his own teaching
started his students with four-voice harmonic studies, using figured bass
realization as the basis for written exercises. The sense of integration of
vertical and horizontal aspects gained from such studies will prove invaluable
as a background to contrapuntal study.
The instructor need not feel locked into the order of presentation given
here, though chapters 1 and 2 should be taken in order. (It is important to note
also that certain terms and abbreviations mentioned in the introduction and
chapter 1 will appear in later chapters.) As time permits, the chapters on
chromaticism, canon, chorale prelude, and passacaglia may be shuffled or
omitted to suit the instructor’s needs and the course goals. The instructor
should not feel limited by the music presented in the anthology; it can be
supplemented by other works of his or her own choosing.
Following are a few practical suggestions, based on my classroom
experience and in no way intended as prescriptive, for presentation of the
material.
1. Listening and performance. As suggested earlier, it is most important
that each class session include some listening examples and if possible
live performances of the works under study. This can lead to discussion
of such fruitful issues as performance practice, music history, cultural
history, and so on.
2. Background information. Although this text by design does not deal with
the historical or biographical background, the instructor may choose to
present such information in the interests of establishing a style-historic
context; he or she may assign information gathering in the form of class
reports by students. Such reports might include such topics as the life
and works of Bach and his major contemporaries; a survey of the output
of Bach; cultural and political currents in the German Baroque; national
influences on Bach; the history and development of various musical
genres and forms; and so on. A basic bibliography begins on page 295; it
is intended as a starting place for such background studies. Appropriate
articles in The New Grove Dictionary of Music and Musicians (see
bibliography) are cited throughout.
3. Listening list. It is suggested that a listening list of Bach’s major works
be distributed to students, and that they be made responsible for a
listening (recognition) familiarity with the works on it, perhaps by way
of periodic listening quizzes. All typical genres and media should be
represented.
4. Classroom process suggestions. In class it is always a good procedure to
vary the format, avoiding too much lecturing and ensuring that each
student is actively engaged in the learning process. This can be
accomplished by having students perform, discuss, and critique each
other’s written and analytic work; having them work at the board
individually or in small teams; and assigning brief in-class writing
exercises critiqued on the spot by the instructor. A very helpful
classroom activity involves the solving of a problem— from the current
or upcoming assignment—by the instructor, with student input. This
modeling of the compositional process will prove invaluable for the
student and will improve both the efficiency and the quality of his or her
own work. Critiquing of individual student work in each class period,
either with the class seated around the keyboard or with the aid of an
onscreen projection, will be of great value.
5. Possible omissions and reorderings. There are far more exercises here
than any class is likely to have time for, so the instructor will need to be
selective in assignments. Such selection will be based on the background
and abilities of the class and the aims of the course. If the focus of the
course is on analysis, it will be possible to use the Directed Study section
of each chapter, along with the exercises in analysis found at the end of
each chapter, in conjunction with this volume’s anthology. Such classes
will omit the written exercises. In classes where the writing of
counterpoint is the goal, a sampling of all the exercise types within each
chapter is recommended. As far as possible, the written exercises in each
chapter should be done in the order in which they are given.
It is also possible to omit the Directed Study and/or Sample Analysis
sections, if that better fits a given instructor’s style, as the Discussion
sections contain all the essential information. Or, the Directed Study
sections may be taken up at the end of each chapter.
Further, some slight reordering of the chapters will be possible.
Chapters 13 and 14 can be taken up earlier in the course of study if
desired (following chapters 9 or 11) or they can be omitted. Chapters 3
and 4 may be postponed or omitted if necessary, depending on the course
goals and the time available.
6. Student writing. The musicality of all student writing should be
emphasized in addition to its technical proficiency; emphasis on this
aspect will pay dividends in terms of student commitment to quality
work. Technique, style, and “playability” should be stressed in
discussing written work, and all work should be played and discussed in
class, when time permits.
7. Linear reductions. This book depends to some extent on linear reductive
principles, but it is not an orthodox Schenkerian text. The approach taken
here focuses on the central issues of shape and direction through the
analysis and composing-out of the structural-pitch framework,
employing an informal system of reduction. The instructor is, of course,
free to use whatever terminology and analytic system he or she prefers.
8. General suggestions for analysis. It is important to stress the general
musical principles operating in the music to be studied and not merely
the individual details of the technique. The fundamental musical laws
operating behind the surface should be pointed out at every opportunity,
including all the sources of shape; the organic nature of musical logic;
the mutual interactions of all elements, and such basic dualities as
continuity/articulation, tension/relaxation, unity/variety, and so on.
Reference to Jan LaRue’s Guidelines for Style Analysis (see
bibliography), as appropriate, will be useful in assuring that the student
is aware of organic processes and not just discrete events.
9. Definitions. A glossary of musical terms, also new to this edition, is
given, starting on page 290. This is intended as a practical reference tool
only, and in no way substitutes for the comprehensive treatment of
technical matters found in the text. The table of contents and index
provide reference to detailed discussions of technique. The glossary
includes cross-references to many standard alternative terms.
10. Ear training and improvisation. Beyond analysis and writing, many
exercises in this text may be used for ear training, sight-reading or sight-
singing, and improvisation. It is always wise to work as much by ear as
possible, in order to draw students away from excessive dependence on
musical notation, and improve their aural acuity and musicianship.
For instance, the error-detection exercises found throughout the text
may be done by ear, using the piano. Solutions to the sequence-writing
exercises in chapters 1 and 2 can be improvised vocally or
instrumentally, by the class or by individual students, as can the melody-
writing exercises at the end of chapter 1. The exercises concluding
chapter 2 will prove particularly useful for sight-singing, or
improvisation at the keyboard or on individual instruments.
All exercises that can be easily sung should be, using solfège or any
system of the instructor’s choosing.
11. Harmony. Classes needing any review of tonal harmony, including chord
functions, figured bass symbols, and nonharmonic tone definitions are
referred to appendix 1. Even those students somewhat conversant with
common practice harmony will benefit from a class session devoted to
review. Several central points made there should be emphasized in all
discussions of harmony:
What can you expect to gain from a course in the analysis and writing of
counterpoint? First, such a study should make you a more accomplished and
discriminating musician, whether as listener, performer, or composer. It can
teach you what to listen for in a passacaglia, for example, or what to bring out
in performing an invention, or how to compose a convincing fugue. It can
help you to understand compositional technique and can enhance your own
creative abilities. There is an excitement in composing good music that is
hard to match in any other musical activity, even when this music is written in
a borrowed style. Contrapuntal studies are essential for theorists, composers,
and keyboard players (especially harpsichordists and organists), and for any
serious student of music who wishes to refine his or her musicianship.1
The study of the music of Johann Sebastian Bach (1685-17502) is an
excellent way to learn how to grasp the underlying principles of all good
counterpoint. The musical principles and processes as applied by Bach are
common to music in general. For instance, the means by which a composer
keeps one voice distinct from the others around it are essentially the same for,
say, Bach, Béla Bartók, and Orlando di Lasso.
Additionally, this book can be read as a model of how to approach and
understand the music of any composer; it can be used, in other words, as a
model for style analysis. There is little about musical style that is mysterious.
Any style can be learned with study and application. One learns a composer’s
music best through maximum exposure to it by listening, performing,
analyzing, and writing. And through such exposure one often reaches an
appreciation of the intellectual discipline that any creative art demands, and of
the fact that in great music we find a perfect mixture of intuition, spirit, and
intellect. In the case of a composer as great as Bach, this combination of craft
and genius is especially awe inspiring. There is no more contradiction
between technique and instinct for a composer than there is for a performer.
The technique must be in perfect working order to seive as a vehicle for the
musical instinct.
Finally, as a reason to study counterpoint, there is simply the satisfaction of
being creative, of producing something that was not there before. You will be
asked in this book to write music that is not only technically “correct” but
also musically satisfying and as close as possible to the style of Bach. This
will require familiarity with the sound of Bach’s music. It is not possible to get
to know a composer’s music simply by memorizing a lot of rules. You need to
immerse yourself in the music, as listener and/or performer. Play and/or listen
to all the music assigned for analysis, and always play your own written work,
both while working on it and after it is completed.3
Notes
1. For more background on counterpoint in general, see The New Grove Dictionary of Music and
Musicians, “Counterpoint,” vol. 4.
2. A good introduction to Bach’s life and works may be found in Tide New Grove Dictionary of Music
and Musicians, “Bach, Johann Sebastian,” vol. 1.
3. It is excellent ear and mind training to work away from the keyboard, going to it only to check your
work.
Introduction
Almost all music is to some degree contrapuntal. Even music that is usually
studied for its harmonic content is often equally linear in conception and
effect. The distinction usually made between harmony (the chordal or vertical
aspect) and counterpoint (the linear aspect, the ways in which independent
voices interrelate) is a pedagogical convention not supported by actual
musical practice. In most polyphonic music, one can say only that there are
both horizontal and vertical controls present (as well as many other kinds of
controlling elements). Thus, we will concern ourselves in this book with both
chord and line, and with how they influence each other.
This study is rooted in musical practice, not in abstractions, and is based on
the instrumental music of Johann Sebastian Bach, with emphasis on the
keyboard music. Why Bach? Because in his music we find an ideal
coordination of strong, directional harmonic progression and energetic,
interdependent lines, all of this suffused with greatness of spirit and largeness
of musical conception. His mastery of harmony, melodic and motivic
processes, and extended contrapuntal composition is of an extremely high
order, yet his astounding technique is always at the service of musical ends,
never an end in itself. Only in the greatest performers and composers do we
sense this perfect unity and balance of technique and expression. Bach
represents a culmination of musical trends in the Baroque era, a composer
who pulls together in his work at least three separate national styles, and
brings to their highest point all the contrapuntal forms he inherited. He
exerted a tremendous influence on later composers. Wolfgang Amadeus
Mozart exclaimed, on hearing a Bach motet for the first time, “Now, there is
something one can learn from!” Ludwig van Beethoven said; “Not Bach
[brook] but Meer [sea] should be his name”; and Robert Schumann advised
the young musician, “Let the Well-Tempered Clavier be your daily bread.
Then you will certainly become a solid musician.”
Directed Study
In the Directed Study sections of this book you will be asked to perform,
listen to, and think carefully about a variety of musical excerpts. Though you
will be focusing on one aspect of the music at a time, still it is important to be
aware of the ways in which the musical elements work together to produce an
effect. For instance, a climactic effect may well be achieved through
intensification of several aspects: a rising line, lengthening phrase structure,
increasingly chromatic or dissonant harmony, more active rhythm, and
thickening texture. Music in which all the elements work together toward the
same result is said to exhibit continuity.1 The music of the great composers
always displays a high degree of concinnity.
A Note on Graphic Analysis
The system of graphic analysis used here requires a brief introduction. The
approach has been suggested by but is not an orthodox version of the linear
reductive system developed by Heinrich Schenker (1868—1935).2 Schenker
realized that the conventional “analytic” tools of his time were merely
descriptive of discrete events, taking into account neither the fact that music
has shape and direction nor that some tonal events are more important
structurally than others. His system grew out of an attempt to show
graphically the fundamental pitch structure of a work. His writing suggests
that a musical work consists of layers (Schichten) that can be peeled away,
onionlike, to reveal a skeletal inner structure, the Ursatz. The Ursatz forms a
background or skeleton for the work, over which we hear the middleground
and foreground (surface) levels of activity. Schenker’s mature work posits a
small repertoire of fundamental structural lines in common practice music
(including that of Bach) that descend stepwise during the work from some
member of the tonic triad to the tonic note. According to Schenker, it can be
shown that many works represent a working out in time of the descent from,
say, the mediant to the tonic note, 3—(2)—1. All other notes in the melody
serve to ornament, expand, prolong, and connect between the pitches of this
fundamental melody (Urlinie). Such ornamentations consist mainly of triad
arpeggiations, passing tones, and neighbors.
This book employs the following symbols:
In making your own graphic reductions, keep the following points in mind:
The following excerpts and their graphic reductions are taken from chapters 2
and 4. These reductions may be discussed in class, if the instructor wishes, as
they are intended to exemplify and clarify the analytic concepts discussed
above. In this discussion, alternative reductions may be considered. Several
more elaborate graphic reductions are to be found throughout the text.
Ex. I-1 French Suite III, Menuet
Linear Intervallic Patterns (LLPs). One of the most powerful and practical
of all Schenker’s contributions to musical understanding was the observation
that between pairs of voices, especially outer voices in sequences, there is a
pattern of harmonic intervals that seems the main organizing principle in such
passages. While this was not his exclusive discovery, he codified the principal
patterns and pointed out their extensive use in tonal music.
The very common 10-10 pattern (equivalent to 3-3 or 17-17) in example 1,
mm. 5-8, has already been pointed out. Many other such patterns will be
noted as they occur in the music. It will be very helpful in your own analyzing
and writing to keep LIPs in mind at all times, and indeed to keep in mind all
aspects of patterning, which is central to the music of Bach and his
contemporaries.
Here are some of the most common linear intervallic patterns:
The LIPs are typically ornamented with motivic figures:
Ex. I-3
Notes
1. This useful term, adopted from the vocabulary of rhetoric by Jan LaRue in Guidelines for Style
Analysis (see bibliography), may be defined as a property of close agreement (in both logic and
style) among all the elements of an artwork.
2. Those wishing more information on Schenker and his theories are directed to The Forte/Gilbert and
Parks books (see bibliography) and the following articles in The New Grove Dictionary of Music
and Musicians: “Analysis, III” (vol. 1) and “Schenker, Heinrich” (vol. 16).
Chapter 1
Line and Other Elements of Style
Here are a number of melodies drawn from the instrumental works of Bach.
Perform each one a number of times.1
There is often one note outside the framework, related as a neighbor note to
the high or low pitch, as for example:
Ex. 1-14
Exercises
1. Perform and critique the newly composed “melodies” on p. 34 in
terms of shape and tonal framework.
2. Identify the range and tonal framework in examples 1-1 to 1-12.
Which scale degrees are involved in these frameworks? Identify
the tessitura (this is often a somewhat subjective judgment).
3. Identify the climactic moments in melodies selected from examples
1-1 to 1-12. Discuss how these are achieved (approached,
emphasized, left). Represent the contour of each of these melodies
with a line drawing, and compare the contours of several melodies.
4. Perform in class other lines by Bach, perhaps from works you
already know or are preparing for performance, and discuss them
in terms of shape and framework, as well as performance-practice
issues.
Structural Pitches
Ex. 1-16
These are not by any means the only filling-in figures available in the style.
The choice of a particular figure will depend on factors of harmony,
counterpoint, scale, and the overall motivic content of any given work. This
will be discussed later.
Notice that each of these figures is an expansion and/or ornamentation of
neighboring or passing processes. As suggested earlier, successive structural
pitches will be connected by arpeggiated or scalar filling-in figures. Single
structural pitches will often be prolonged by neighboring motion.
Exercise
Analyze the lines at the beginning of this chapter in terms of structural
pitches. The individual voices of the polyphonic works found later in
this book may, at the discretion of the instructor, also be so analyzed. At
least two structural levels should be shown (primary and secondary
pitches). Some melodies will be clearer than others in terms of structural
line. After each reduction is made, the filling-in figures should be
written down and discussed. Be aware of all types of prolongation and
filling in: neighbors, scales, and arpeggios. Note also how the principal
implied harmonies are prolonged and clarified through the primary
structural pitches. Some lines involve octave transfer, or may be
compound (see pp. 30-32).
Phrase and Cadence
Sample Analysis
Ex. 1-18 Orchestral Suite in C Major, Gavotte I4
Discussion
All levels of formal structure in this music reveal the influence of their
ultimate origin in dance. As humans are bipedal, many aspects of time in
music (rhythm, meter, phrase, and finally form) tend to occur in twos and
multiples of two: a gesture and a balancing or opposing gesture; action and
reaction. Dance movements, such as the gavotte in example 1-18, will tend
clearly to show their historical dance antecedents. More abstract works,
especially those based on imitation, will tend to be less regular at all levels of
formal structure.
While the underlying phrase structure is basically regular, its effect is often
less obvious on the musical surface than in more consistently homophonie
music. Phrase is a controlling element in the melody in example 1-19. Notice
in it the variety of rhythmic values, the distinct articulations every four
measures, and its essential vocality (as against the instrumental flavor of the
Bach excerpts).
Ex. 1-19 Franz Joseph Haydn, Symphony No. 104, First Movement
Compared to this music, that of Bach is usually less obviously built around
the phrase and is more continuous (pulsatile). It is far less likely to use the
phrase as the main unit (module) for development, employing instead the
spinning out (Fortspinnung is the German term) of motivic fragments by
means of various manipulations: repetition, inversion, sequence, and so on.
See, for instance, examples 1-3, 1-4, 1-9, 1-10, and 1-11.
Cadence Figures
Cadences (breathing places that articulate the ends of phrases) are principal
articulating and form-clarifying devices of this music. In any music,
insufficient use of cadential effects may lead to an inarticulate or too
continuous result. Overuse of cadences will result in a discontinuous effect.
Bach’s music is beautifully balanced between these extremes. Bach is careful
to continue motion over most internal cadences, weakening them just enough
for continuity. Only at main cadences is the motion allowed to stop.
Principal cadential points in this style are emphasized by:
All the cadential melodic figures you have heard so far have been variants
of the two basic cadential shapes in tonal music:
Ex. 1-20
Both of these archetypal shapes approach the goal note by step. The word
cadence comes from the Latin cadere, “to fall,” and in fact the cadential
gesture in most melodies is a falling one. In this style, the goal note is
normally placed on a strong beat, though it may also follow an on the-beat
dissonance such as a suspension, passing tone or appoggiatura, especially in
slow movements. Some characteristic ornamentations in the Bach style of
these fundamental cadences are seen in example 1-21. These ornament an
authentic cadence in G major, with structural pitches on scale degrees 2-1 or
7-8.
Ex. 1-21
Other cadence figures will be found, but will on examination prove simply
to be variants of the above figurations.6
Exercises
1. Investigate phrase structure and cadence figurations in the poorly
composed melodies on p. 34. Critique these aspects.
2. Analyze other music, as assigned by the instructor, from the
anthology, focusing on phrase structure and cadences. Be
especially aware of any irregular phrases, the use of motion to
cover internal cadences, and the use of ties and evaded cadences to
make the phrasing more flexible and continuous. Analyze cadences
by frequency, placement, type, and melodic figures.
3. Ornament the following fundamental cadences in the Bach style,
using mixed note values, including eighths and sixteenths. All are
authentic unless otherwise marked. Use as your models cadences
from works in this book, or other Bach pieces you know.
Meter and Rhythm
Sample Analysis
Exercises
1. Critique the lines on p. 34 in terms of meter and rhythm.
2. Perform and analyze works from the anthology, as assigned by the
instructor, being aware of factors of both regularity and flexibility
in meter and rhythm.
3. Critique the following phrase in terms of the Bach style:
Ex. 1-24
Leaps are treated with care in this style. All melodic intervals up to the
octave are used, but with some restrictions. The m7 is normally found as an
ascending interval; the M7 is extremely rare. Intervals larger than a fifth often
imply the presence of a compound line (see p. 30.) The P8 is most often used
to adjust register (avoiding extremes of range), to accommodate a change of
direction in line, or to change spacing or avoid voice-crossing in a
contrapuntal context. Leaps are most often found from the strong part of a
beat to the weak, or from strong to weak beats (except in the bass voice
which, when treated as harmonic support, leaps more freely than the others).
They most often occur between tones of the same chord. The larger the leap,
the more compelling the tendency to follow it by motion, most often stepwise,
in the opposite direction. Sixths and sevenths almost always resolve by
opposite motion, unless involved in a compound line. Thirds are often not
balanced, but larger intervals usually are. Leaps into melodic or harmonic
tendency tones are most often resolved by opposing motion. If a line moves in
one direction by a mixture of leaps and steps, the leaps usually precede the
steps.
Ex. 1-26
Ex. 1-27
Repeated-note figures are used with care, as they can easily interrupt the
sense of direction. When used, they are treated consistently as part of the
motivic fabric or they may be found as part of the cadential anticipation
figure.
Ex.1-28
Augmented and diminished intervals are treated with great care, as they are
dissonant and require specific resolutions. Briefly stated, diminished intervals
resolve inward, augmented outward. The most common are the A4, d5, d7,
and A2. They are nearly always found in association with dominant harmony
(V7 or vii°7). The A4 and d5 are used quite frequently and are carefully
resolved (either immediately or very soon thereafter). The A2 is much rarer
and is normally associated with vii°7, descending in quick note values (see p.
20).
Ex. 1-29
Not only are direct dissonant intervals resolved, but stepwise or arpeggiated
lines implying these intervals are treated with the same restrictions.
Ex. 1-30
Exercises
1. Perform and critique the “melodies” on p. 34 in terms of the
incidence and treatment of leaps and stepwise figures.
2. Critique the following melodic fragments.
3. Identify and resolve the following intervals and lines. These may
be done vocally, as ear-training exercises, as long as a key-context
is established first, for each exercise.
Scales; Chromaticism; Tendency Tones
The raised submediant (A♮) is used to avoid the A2 between A♭ and B♮ (m.
4).
The subtonic (B♭) is used to lead away (down) from the tonic (mm. 5, 6).
The submediant (A♭) leads down to the dominant note (mm. 7-8).
Note the descending tonic-dominant structural line (c2-g1) formed by the
downbeats of mm. 5-8. The natural (or descending melodic) minor form
is used here: C-B♭-A♭-G (8̂ 7̂ 6̂ 5̂).
Discussion
In the Bach style (and in tonal music generally), the minor scale is used in
specific, restricted ways, depending on factors of line and harmony.
1. The leading tone is used to lead up to the tonic note, especially when the
underlying harmony is dominant or tonic. It may also be used as part of a
scale leading down from tonic to submediant, when dominant harmony
is implied.
2. The subtonic usually leads down from tonic to submediant, as in the
natural minor scale, when the underlying harmony is not dominant. The
subtonic is almost never used as a lower neighbor to the tonic (except,
rarely, with subdominant harmony).
3. The raised submediant is used either to lead up to the leading tone (as in
ascending melodic minor), or as part of a descending line following the
leading tone (avoiding the A2), associated with dominant harmony. It is
almost never used as an upper neighbor to the dominant note.
4. The submediant (as in natural minor) leads down to the dominant note. It
may follow the leading tone (creating an A2) or the subtonic, or may
occur as an upper neighbor to the dominant.
5. The A2 between the leading tone and the submediant (as in harmonic
minor) is occasionally used, normally descending in quick notes, and
always with dominant harmony.
Ex. 1-32
Ex. 1-33
The melodic minor is used most often with tonic harmony, the harmonic or
ascending melodic minor with dominant, and the natural minor with the other
diatonic triads, thus:
CHORD SCALE FORM
i melodic
V, V7 harmonic or ascending melodic
vii°7 harmonic
ii°, III, iv, VI natural
Ex. 1-34
There are no such problems with the major scale, as it contains no variable
degrees. Any accidentals will imply either modulation, altered chords such as
secondary dominants, or nonharmonic tones, such as an occasional chromatic
lower neighbor.
No scales other than major and minor are in common use in Bach.7 Thus,
use of the pentatonic scale is very rare, and whole-tone scale effects are
uncommon except as incidental to the ascending melodic minor.
Ex. 1-35
1. Altered notes usually resolve in the direction of their alteration; that is,
raised notes resolve up, lowered notes resolve down. See example 1-31
for several typical examples.
2. Altered notes may be either chord tones or, less commonly, nonchord
tones (such as the f♯2 in m. 6 and the e2 in m. 7 of example 1-31, both of
which are lower neighbors).
3. In his highly expressive, slow music, Bach often makes considerable use
of both functional and nonfunctional chromaticism. This will be
discussed separately later, in chapters 3 and 9.
Ex. 1-36
Ex. 1-37
Exercises
1. In the Error-Detectlon exercises, p. 34, identify all the errors of
scale, accidental, and tendency tone resolution.
2. Perform music from the anthology, as assigned by the instructor,
and analyze the use of the minor scale degree variants, the function
of any chromatic notes, and the resolution of melodic tendency
tones.
3. Critique the following lines in terms of the treatment of the minor
and other scales, and the resolution of tendency tones. Suggest
ways of correcting the problems you find. These may be done by
ear, away from the notation.
1. Passing tones (p.t.) are freely used. They fill in leaps of thirds and
fourths between chord tones, by step. In this style, passing tones tend to
be short, diatonic, and more often unaccented than accented (though the
accented descending passing tone occurs often). They may freely ascend
or descend. Two successive passing tones in a scalar figure form a
common idiom.
2. Neighbor (Auxiliary) tones (n.t.) are also quite common, the lower
neighbor more so. Like passing tones, these are normally short,
unaccented, and diatonic (though the chromatic lower neighbor is
possible).
3. Anticipations (ant.) are typical only at cadences.
4. Suspensions (susp.) are widely used and are a principal expressive
device. They will be treated in detail in chapter 2.
5. Appoggiaturas (app.) are quite rare in this style, being more
characteristic of later music. They are almost always diatonic and rarely
last for more than one beat. The unaccented appoggiatura (called by
some theorists a cambiata) is referred to here as an incomplete neighbor
tone (i.n.t.).
6. Escape tones (e.t.) are fairly rare in this music. They are typically very
short and are always diatonic.
7. Pedal point is a special device which will be taken up later. It will suffice
for now to point out that pedal effects are largely restricted to the
dominant and tonic notes, whether in a bass line or as part of a
compound line. Pedal point is debatably nonharmonic, as it is typically
the upper voices rather than the pedal that are heard as nonharmonic. In
any case, the pedal note always begins and ends as a consonance.
Ex. 1-38
It is important to note that passing and neighboring tones may exist either
on the surface (decorative or foreground) level of structure, or, as we have
seen in the graphic analyses, on the middleground level. Review the analysis
of example 1-31 to see typical passing and neighboring tones on the
middleground level.
Exercises
1. Perform the following melody, and locate and identify the
nonharmonic tones.
2. Define, in terms of approach, resolution, length, and metrical
placement the following nonharmonic tones, as used in the Bach
style:
passing tone
neighbor
escape tone
anticipation
appoggiatura
suspension
Sample Analysis
Note that the localized structural pitches of “d” are the same as for m. 1.
Ex. 1-41
Observe also how, in m. 9, the “b” motif gives rise to a typical cadential
figure. In m. 10 we find both fragmentation and alternation of “d” and “b′.”
The “e” motif may be derived from “a′” but is more likely heard as merely
related informally by rhythm. In purely rhythmic terms, there are really only
two ideas-, groups of eighths and groups of sixteenths.
Transformation or development processes include inversion (“a′” is an
inversion of a”), sequence (mm. 1-3, 6-8, 10-11, 12-13, fragmentation (m.
10), and extension (m. 5, beats 1 and 2, extends “a”; m. 8, beats 3 and 4,
extends “d” and returns to “c”). You may also have noticed the highly
organized descending structural pitch lines in the above melody and may wish
to prepare a graphic analysis of them.
Discussion
Bach’s is a highly coherent style on every level, and perhaps most clearly on
the motivic one. Movements are “spun out” of the opening materials. As was
mentioned on p. 11, the unit or module out of which this music grows is more
characteristically the motif than the phrase. Motifs often consist of three- to
five-note patterns, identifiable by their rhythm and their shape or interval
content. Growth or development processes applied to these motivic units, or
cells, include repetition; alternation of two different motifs; return after
intervening material; sequences (see below for detail); fragmentation,
inversion, diminution or augmentation of rhythm; retrogression; and
combinations of these devices. Some of these are demonstrated below.
Ex. 1-42
The excerpt in example 1-43 begins with a pedal sequence in which the
upper notes are sequential (rising by step), and the lower note (F♯) is a
dominant pedal. A true sequence begins in m. 2, with a two-beat unit,
transposed down by step. The unit is heard twice, ending with the note (d2, m.
4) that would have begun the next unit were the sequence to have continued,
thus ending the sequence very smoothly. The f♯2 in m. 2 is both the expected
subsequent note of sequence 1 and the initial note of sequence 2, a subtle
interlocking of the two sequences. Note the clear pattern formed by the initial
note of each unit, a scalar expansion of the note d2.
Ex. 1-44
Here, as is normally the case, the initial note of each unit is heard as a
structural pitch.
Ex. 1-45
Symbology:
The beginning of each unit is shown as г
The first note of each unit is circled, to indicate transposition: Օ
The point at which the sequence breaks is shown as ⅂.
Ex. 1-47
Comments
Measures 1-4 constitute a pseudosequence in that mm. 2 and 4 are nearly
sequential, but mm. 1 and 3 are not.
A sequence begins at m. 5. It employs a two-measure unit with two
different figures (very typical of Bach), heard a total of two times and
transposed down by step.
Another sequence begins at m. 9. If the preceding sequence were to have
continued, its next unit would have begun on the note c2 in m. 9, which is in
fact the first note of the second sequence. This interlocking of sequences is
highly typical.
The second sequence has a one-measure unit, transposed down by step.
This halving of the length of the second unit is also characteristic, and
contributes greatly to the sense of drive or acceleration through the phrase.
This unit is heard twice and again comes to the note that would have begun
the third iteration (the a♭1 at the beginning of m. 11).
Measures 14-15 have a pedal sequence, in which one implied voice is
sequential while the other voice is a pedal point.
The first note of each sequential unit in mm. 5-11 forms a clearly directed
descending stepwise structural line, emphasizing the new tonic note, E♭.
Exercises
1. Analyze all the sequences in the melodies beginning this chapter,
as shown in example 1-46.
2. Analyze in music from the anthology, as assigned by the instructor,
the motivic content and the processes of unification and variation
employed, including a detailed analysis of the sequences.
3. Exercises 2-4 on pp. 34-35 may be done, in whole or part, at this
time.
4. Bring into class, and perform, works of Bach; lead a discussion of
motivic unity; find any instances of repetition, fragmentation,
extension, inversion, and sequence.
Compound Line
Sample Analyses
Discussion
Baroque instrumental music is characterized in part by the frequent use of a
compound line. A great deal of keyboard music makes use of this device, as
does much of the music written for such solo instruments as flute, violin, or
cello, giving these the ability to play implied counterpoint. Sometimes the
number of implied voices is clear; often it is not, or the voices are heard as
sharing some of the same pitches. (The final d in m. 7 of ex. 1-48 may be
such a note.) Each voice is musically satisfying, directional, and coherent.
There are two basic types of compound line sequences, as shown above. In
type 1 (mm. 1-4), both voices move by the same transposition interval. In
type 2 (mm. 5-6), one voice is a pedal point (and the sequence thus a pedal
sequence).
Ex. 1-49
Further, the two voices may share the same motivic material, or may not, as
shown in example 1-50.
Ex. 1-50
Exercise
1. Analyze more instances of compound line, as found in the
examples beginning this chapter (examples 1-9 to 1-12 all contain
some use of compound line), and also from the anthology, as
directed by the instructor.
Suggestions for Melodic Writing
Rhythm
Scale
Scale
only typical scales used
idiomatic use of minor scale
all tendency tones resolved normally
idiomatic use of chromaticism (if any)
tendency tones not emphasized in line (especially leading tone)
whole-tone, pentatonic, or modal scale effects avoided
Line
clear structural pitch outline
typical stylistic figures employed for filling in between structural pitches
or prolonging them
clear phrase structure
scale passages left by step
idiomatic use of compound line
repeated notes used only if motivic
balanced use of leaps, steps
clear and balanced contour
larger leaps balanced by opposing melodic motion
well-placed climax and clear approach to cadences
typical cadential idioms, approached smoothly and set up beforehand
no lines that are too disjunct, steep, abrupt, or flat
successive leaps in one direction form chordal outlines
all dissonant intervals and chromatic notes resolved
sequences employed idiomatically
consistent and restricted motivic material
strong beat usually approached by step
Melodic Writing Checklist
1. Be sure you fully understand and can apply all the points on the
preceding Suggestions for Melodic Writing list before beginning the
exercises below.
2. As you write, analyze all the following aspects:
3. Remember that you are writing music, not “theory exercises.” Your work
should be musically satisfying, even though it will be rather simple at the
early stages of your study. Play all your work to make sure it is correctly
notated and sounds as you intended. Work away from an instrument,
“singing” the line internally in your head; then check the result at an
instrument.
A. Given unit
A. Given framework:
7. Spin out melodies of about 8-16 measures from the given motivic
material, using instruments available in the class. End in a strong
cadence. Analyze fully.
8. Write melodies in the style of Bach for instruments available in class,
using the following formats (which may be slightly varied as needed).
These formats may also be used for improvisation.
Notes
1. The instructor is reminded that throughout the discussions in chapter 1, classes needing review of
common practice theory—including harmonic functions, nonharmonic tones, cadences, modulation,
and figured bass symbols—should refer to the materials on pp. 283 ff. Some instructors may wish to
reserve discussion of those melodies employing extensive compound line (particularly exs. 1-9 and
1-11) for the section in this chapter dealing with that technique, pp. 30-32.
3. The term figure is used here as synonymous with melodic or motivic pattern. It should not be
confused with the more specific use of the term to denote particular expressive melodic devices such
as the Seufzer or “sigh motif.” See The New Grove Dictionary of Music and Musicians, “Affections,
Doctrine of the” (vol. 1), “Figures, Doctrine of Musical” (vol. 6), “Motif” (vol. 12), and “Rhetoric
and Music, III” (vol. 15).
4. Only the outer voices (oboe I and continuo) are given here. The continuo part is given only for
clarification of the cadences: it is not suggested that the contrapuntal relationship between the voices
be discussed at this point.
5. A period consists of two successive phrases, usually of equal length, the second one completing the
musical idea begun by the first. The cadence ending the first phrase will be to some degree
inconclusive (HC or IAC); the second cadence will be more final, usually a PAC. Cadence
definitions are given in Appendix 1.
6. See also The New Grove Dictionary of Music and Musicians, “Cadence,” vol. 3.
7. There are instances in which Bach uses an incomplete key signature for a work in a minor key,
implying a vestigal sense of modality (most often a Dorian mode). In such works, though,
accidentals obscure the feeling of modality. In other isolated examples Bach, in the course of
harmonizing a modal chorale, will employ a somewhat modal harmonization (this often happens
with Phrygian chorale melodies).
8. Classes needing to review the definitions of nonharmonic tones and any other aspects of harmony at
this stage of study are again referred to the review materials on pp. 283 ff.
9. Sequence can be briefly defined as the repetition of a melodic pattern at a new pitch level.
Chapter 2
Nonimitative Two-Voice Writing
Perform the following two-voice music of Bach, repeating each passage until
you are familiar with it. Then play each voice by itself to gain an appreciation
of the individual lines. These excerpts can be performed, as Bach intended, on
a keyboard instrument (preferably harpsichord), or on two separate
instruments such as flute (or violin or oboe) and cello (or bassoon)—
whatever is available in the class.
Ex. 2-1 French Suite II, Menuet
Directed Study
Based on what you hear and see in examples 2-1 through 2-5, what would you
say it is, in general terms, that makes this good counterpoint?1 Are the voices
individually satisfying? Do they make sense as independent musical lines?
Are they always of equal interest and importance, or are there moments when
one voice seems supportive? If so, under what conditions does this happen? In
what ways do the voices relate, in terms of the musical materials they share?
In what ways, generally speaking, are they kept distinct from each other?
Discussion
1. Integrity of the individual voices. Each voice is a satisfying line with the
characteristics of melody discussed in chapter 1. Each is coherent
motivically, well-shaped, and clear in meter and harmony.
2. Equality or near equality of both voices. There are times at which one
voice will dominate because of its rhythmic activity, but there is little
feeling overall that the lower voice is merely harmonic or figurative
support. This will, of course, vary to some degree with the movement
under discussion.
We are dealing now with nonimitative dance movements, which will
tend to be somewhat more homophonic than imitative works. Compare,
for instance, examples 2-4 and 2-5 above in terms of the function and
importance of the lower voice.
3. Consistency of materials between the voices. Even when the same
materials are not literally shared, as they would be in imitative works,
still both voices will share the same general thematic content, especially
when both are of equal importance. For instance, in example 2-1, the
voices both have some of the same musical ideas, but they also have
their own independent motifs. One might understand the motivic content
of mm. 1-4 as follows:
Ex. 2-6
Directed Study
Look at how, in the music beginning this chapter, the general contours of the
voices relate to each other. Note the overall melodic shapes within each four-
and eight-bar phrase, each two-bar group, and finally within each bar. Prepare
contour graphs of the voices, as suggested on p. 44.
Next, consider these relationships in more detail. Count from beat to beat,
and also from the end of each divided beat to the beginning of the next, the
incidence of the four possible relative-motion types: contrary, oblique,
similar, and parallel (see example 2-8). Use the format suggested in example
2-7, or one given by the instructor. Prepare a summary of your findings. What
generalizations can you make concerning the use and relative frequency of the
four types? Focus especially on parallel motion. Which intervals are used
consecutively? Which are not?
Notice that the shapes of the two lines contain both similar and contrary
elements. The voices will usually exhibit some degree of independence of
shape, as contours that are too consistently similar will not make for effective
counterpoint.
In example 2-7 the beat-to-beat relative motions have been shown for mm.
1-11, symbolized as c (contrary), s (similar), o (oblique), and p (parallel).
These four types can be shown in a simplified form as follows2:
Ex. 2-8
1. Both voices in a two-voice texture will rarely leap in similar motion into
a P5 or P8. This is known as direct fifths or direct octaves, and is heard
as detracting from the independence of the voices, due perhaps to the
acoustically “open” sound of the P5 and P8.
2. Both voices will occasionally move by similar motion into a P8 or P5 if
the upper voice moves by step. This is especially typical at cadences.
Ex. 2-9
It is in the matter of parallel motion that Bach exercised the most care. His
practice can be reduced to the following observations:
1. Parallel (consecutive) thirds and sixths (and their compounds, tenths and
thirteenths) are freely used, though it is unusual to find a total of more
than five or six in succession, as this detracts seriously from
independence. The second idiom below is avoided, unless the implied
tritone can be resolved.
Ex. 2-10
Ex. 2-11
3. Parallel perfect fifths are not found in his music, nor are fifths by
contrary motion.4 Further, fifths on successive strong beats (or strong
parts of beats) are not used, though fifths falling on successive weak
beats (or parts) are possible.
Ex. 2-13
Ex. 2-14
Ex. 2-15
Exercises
1. Perform and analyze excerpts from two voice works found in the
anthology, pp. 301 ff. Analyze the directional and intervallic
relationships between voices. Be especially alert to any parallel
intervals; how are they treated?
2. Write in two voices, on two staves (bass and treble), brief, isolated
examples showing:
direct fifths
direct octaves
parallel dissonances
parallel and contrary fifths
parallel and contrary octaves
Incorrectly treated unequal fifths
Rhythmic Relationship
Directed Study
In the musical excerpts beginning this chapter, observe how the voices relate
rhythmically to one another. Are both voices equally active? Note where they
move in the same values; for how many beats does this continue? When they
do not move in the same value, what is the usual ratio between their values?
Is the subdivision unit (eighth or sixteenth note) normally present in at least
one voice? Is it often present simultaneously in both voices? Do the voices
share some of the same rhythmic motifs? Is there any feeling of acceleration
in rhythmic activity through the phrase?
Discussion
Exercises
Perform and prepare outlines of the rhythms of several selected two-
voice excerpts, as assigned by the instructor. The following suggested
format shows a rhythmic outline of example 2-1, mm. 1-3.
Then perform these outlines in class, conducting the meter and intoning
the rhythms on a neutral syllable such as “ta.” Perform each voice
separately, then both voices together. Discuss what you observe.
Spacing; Crossing; Overlapping; Range
Directed Study
In the examples beginning this chapter and in music selected from the
anthology, note the distance the voices generally are from each other. What is
the widest interval you observe? Do they often remain widely separated (say,
more than two octaves distance)? Are the voices kept distinct from each other
by registrar placement? Are their registers allowed to become and remain
very close (say, within an octave) for long? Do you observe any crossing of
voices? What are the highest and lowest pitches used? What is the range,
generally speaking, of each voice?
Discussion
Bach is very careful to keep the voices in separate registers. Wide spacings
predominate in his two-voice keyboard writing. The voices are often two to
three octaves apart, though they rarely remain very widely spaced for more
than a measure or two at a time, tending to come back together by contrary
motion. The extreme registers are used to set up a “registral tension” that is
released through the answering use of the middle registers (example 2-3, mm.
1-4; example 2-5, mm. 1-3). On the other hand, the voices rarely share a
register for long, as this makes it difficult for the ear to distinguish them. Both
voices will occasionally rise together into their highest registers for a
climactic effect (example 2-2, mm. 3-6). Voice crossing in two voices is very
rare, as it obscures the voices. Likewise, overlapping is to be avoided at this
stage.5
Ex. 2-16
Ex. 2-17
The upper voice may occasionally go slightly lower, and the lower voice
slightly higher, but not for sustained passages. The outer limits (Great C to f3)
are those of most of Bach’s keyboards and should be strictly observed.
Obviously, when writing for specific instruments available in the class, their
particular ranges will have to be observed.
Directed Study
Make an analysis of the harmonic intervals between the voices in some of the
excerpts above, as selected by the instructor. Note first the intervals on the
beginning of each beat, then those within the beat (if there are subdivisions
present). Which intervals predominate? Which are not found, or are found
only rarely? Of these rarer intervals, note carefully where they fall metrically:
are they in short or long values, weak or strong metrically? Where are the
perfect intervals (PI, P5, P8, and their compounds) placed? How are dissonant
intervals (A and d intervals; seconds, sevenths, P4) treated in terms of length,
strength, approach, resolution? Can you analyze these dissonances as typical
nonharmonic tones?
Sample Analysis
The most used essential intervals are the major and minor thirds and sixths
(and their compounds), the so-called imperfect consonances.
The perfect consonances (P5, P8, PI) are found at the beginning and
cadence, and on weak beats or parts of beats.6
The circled tones are those that might be heard as nonessential
(nonharmonic) tones, depending on one’s perception of the underlying
harmonic structure.
Note the large number of weak-beat dissonances (seconds, sevenths,
fourths). Passing tones predominate.7
The only essential diminished interval (the d5 in m. 2) resolves normally
(to a third, m. 3, beat 1).
Discussion
Ex. 2-19
To review:
These strict resolutions are subject to the usual variations of delay and
transfer discussed on p. 22.
Some comments on detail follow.
The d and A intervals, as well as the M2 and m7 are often associated with
dominant harmony (V7 or vii°7) and may thus be essential intervals (that is,
involving chord tones). Note in example 2-19 that all the dissonant intervals
(except the P4) are part of a dominant seventh chord built on G.
In two voice writing, the A2 and d7 are rarely used as harmonic intervals;
they resolve to the P4 and P5, respectively.
Ex. 2-20
Ex. 2-21
Directed Study
Analyze the harmony in selected excerpts from the beginning of this chapter
and from selected two-voice works in the anthology (excepting works with a
great deal of chromaticism).10 Be aware of such matters as cadences,
harmonic rhythm, chord vocabulary and function, chord inversions, clarity of
harmonic structure (is it always clear what the harmonies are?), use of altered
chords, modulation (where; to what keys; by what means?), and doubling and
nonharmonic tones. At this point, appendix 1 (p. 283), and the introduction (p.
xix) may be reviewed.
Ex. 2-23
Discussion
A review of harmony and related elements is given on pp. 283 ff. This
material must be mastered. A few additional comments are appropriate at this
point.
Ex. 2-24
As you have seen, a third (tenth) usually implies the root and third of a
triad. In two voices, this is the most common sound, especially on the
beat. A third from chord third to chord fifth is also found quite often. The
P5 from root to fifth is rarely used on a strong beat or fraction, as the
sound of the “open” fifth is relatively weak. A unison or octave will
usually imply a doubled chord root, found most often at beginnings and
at cadences. Thus, the sonorities that may imply a C-major triad are as
follows:
Ex. 2-25
Ex. 2-26
5. Inversions.
A. First inversion. You have observed in the music under study that
first inversions are used quite often.11 First inversion is necessary
for (indeed, may be said to arise out of) strong bass lines, and has
the additional advantage of “lightening” the harmony and
increasing forward momentum. Overuse of root position results in
disjunct and awkward bass lines and overstable harmony, lacking in
momentum. While root position chords tend to occupy the strong
beats, first inversion is widely used elsewhere (that is, weak beats
and fractions), and in mid-phrase even on strong beats. Phrases tend
to begin and end with root position triads for stability. Any triad
may be used in first inversion, but the most usual are I6, V6, ii6, and
vii°6. In other words, vi and iii are quite rare in this style. Some
typical idioms follow:
Ex. 2-27
In the idioms above, V may substitute for V6, and ii for ii6.
Note that in the majority of these idioms the bass in the first
inversion chord resolves by step (often up) into the root or third of
the following chord. Observe the strongly linear character of these
bass lines.
Some theorists doubt that the “ chord” can even be said to exist
in this music, especially in thin textures. This is expressed through
the use of parentheses around the analytic symbols for these chords.
Parentheses are often used when linear chords (those arising from
nonharmonic tones) are implied. In such cases, the notes implying
the can often be appropriately analyzed as nonharmonic tones, at
the instructor’s discretion.12
Ex. 2-29
Ex. 2-30
Even though Bach uses the altered chords shown on the chart on p. 283,
it is best at this stage of study to employ them only sparingly in one’s
writing, limiting usage to an occasional secondary dominant chord,
resolved normally.
7. Nonharmonic tones (more detail). In your harmonic analysis of the
excerpts beginning this chapter, you have noticed the types of
nonharmonic tones discussed in chapter 1 (pp. 23-24). A few further
points should now be made:
A. Not all nonharmonic tones are dissonant, nor are all dissonances
nonharmonic,13 as is shown in the following sketch:
Ex. 2-31
Ex. 2-32
Ex. 2-33
Ex. 2-34
Ex. 2-35
B. Suspension chains are fairly common in Bach. These are normally part
of a sequential passage, and thus should not be carried on too long. Note
the descending stepwise structure, characteristic of passages employing
suspensions.
Ex. 2-36
Ex. 2-37
D. Ties. It will be recalled that ties are commonly used in a ratio of 1:1, 2:1,
3:1, or 4:1. Thus, the note of preparation will be at least as long as the
suspension. In few cases is the preparation shorter than an eighth note.
Ex. 2-38
E. The upward-resolving suspension (or “retardation,” as some term it) is
rare in Bach, reserved mainly for very slow, expressive movements. It
becomes very common later in the Viennese classical style.
Ex. 2-39
A Note on Compound Meter
Most of the music examined thus far has been in simple meters, with
durational ratios between voices of 1:1, 2:1, and 4:1. There are no new
technical observations concerning 3:1 counterpoint. A few brief examples will
suffice, shown as elaborations of 1:1 models. In the examples below, note the
use of passing, neighboring, and (more rarely) arpeggiated motion to prolong
or connect between the model pitches. These elaborations are, of course, only
some of those possible in the style; the choice of a particular figure will as
always depend on factors of motif, harmony, genre, tempo, and affect.
Motivic unity and consistency within a given movement continue to be of
primary importance.
Ex. 2-40
Exercises
1. Analyze two-voice works selected from the anthology, focusing on
harmony and nonharmonic tones, as suggested on pp. 49 ff. Be
sure to attempt two-voice structural pitch reductions, as these are
revealing of underlying linear organization. Also analyze the
structural harmony, cadences, and all instances of patterning
(linear-intervallic patterns and sequence).
2. Determine which triads and seventh chords could be implied by the
interval d1-f1.
3. In the key of B♭, which note is least likely to be doubled? Which
note in G minor? Which notes on both keys are most likely to be
doubled?
4. Write authentic and half cadence bass lines (of three or four notes)
in the keys of D, c, Bl and e. Use figured bass symbols and analyze
the implied harmonies. Show scale degrees.
5. Write bass lines in the keys of d and F; showing eight typical first
inversion usages in each key. Use figured bass symbols and show
scale degrees.
6. Embellish example 2-37 with eighth and sixteenth notes, showing
several typical ornamentations of the suspensions. Work for a clear
sequence and limited motivic content.
7. Compose a brief two-voice example in a simple texture, showing
the 4-3, 7-6, and 2-3 suspensions, with typically ornamented
resolutions.
8. Find and discuss the many technical errors in the following
example.
Relation of Voices
number of independent voices
motivic relationship of voices
function of bass line: motivic? imitative?
rhythmic relation of voices
directional relation of voices; overall contours and localized relations
harmonic intervals formed between voices; which occur on strong beats?
any instances of linear-intervallic patterns?
invertible counterpoint
textural changes
treatment of dissonant harmonic intervals
imitation: pitch and time interval between entrances. Length of imitation.
Harmony
chord vocabulary and placement in phrase, including inversions; analyze
structural harmony
harmonic rhythm; does it change anywhere?
cadence types and placement
nonharmonic tone usage—types, frequency, and placement in meter
modulations: where, by what means and to which tonal areas?
Special Devices
pedal point
stretto
inversion
retrogression
augmentation
diminution
Essentials or Two-Voice Counterpoint
1. Rhythmic relationships—1:1, 2:1, 4:1, and (in compound meters) 3:1
and 3:2 are to be used.
2. Motion relationships—available are parallel, similar, oblique, contrary. A
mix of all types is typical. Parallels—imperfect consonances only; no
parallel perfect consonances or dissonances. Avoid extended use of
parallel thirds or sixths.
3. Harmonic intervals—use mostly imperfect consonances. Perfect
consonances occur mainly at beginnings and cadences, or on weak beats,
using strong scale degrees (1, 4, 5). Treat all dissonances correctly in the
usual nonharmonic tone idioms. Resolve all d and A intervals. The P4 is
a dissonance. Make use of the LIPs, where possible.
4. Harmony—must be clear, functional, with regular (patterned) harmonic
rhythm. Be aware of scale-degree/harmonic formulae, especially in the
bass. Use inversions idiomatically, and strongly directed bass lines.
Detail Reminders
1. Avoid voice crossing for now.
2. Avoid consecutive dissonances, except for a weak passing tone followed
by a strong passing tone descending.
3. No parallel, contrary, or direct fifths or octaves. No unequal fifths.
4. Nonharmonic tones cannot be used to correct parallels.
5. Doubling: freely double strong scale degrees; generally avoid doubling
tendency tones (leading tone, chord sevenths, altered tones, chord
thirds); line sometimes takes precedence over doubling.
6. Never choose a note for harmonic reasons only.
7. Stay focused on line. Use typical melodic figurations, including cadence
figures; work for motivic unity, shape, pattern, and continuity.
8. Always work from, and be aware of, a structural pitch framework
(skeleton), both for shape of line and strength of counterpoint.
9. You are writing music, not just “theory exercises.” Rhythmic flexibility,
coherence and expression are very much to the point.
10. Check all work at a keyboard, or two instruments.
11. Analyze as you work:
shape
coherence
consistency
pattern
flow (continuity)
Cumulative Exercises
Play the following examples and locate and discuss all the musical, stylistic,
and technical errors in them. There are many errors of line, harmony, rhythm,
counterpoint, and style.
PROCESS DEMONSTRATIONS. It will be helpful before proceeding to
the exercises to read and play through the following process demonstrations.
Stay, in all your waiting, within the range of Bach’s keyboards, as found in
The Well-Tempered Clavier (C→c3).
Given Figured Bass
1. Work out chord implications of figured bass. Spell the chords.
2. Supply structural pitches, based on the chord tones available, working
for strong vertical intervals and a good shape. This is the middleground
framework.
3. Fill in between structural pitches, working for regular rhythmic flow,
using consistent motivic figures and a strong cadence. This forms the
rhythmic/motivic foreground.
4. Check your work, analyzing all chords, nonharmonic tones, vertical
intervals, motives, dissonance treatment, scale-degree patterns in the
bass, sequences, and LIPs.
Given Upper Voice
1. Work out the chordal and nonharmonic tone implications and cadence
type. Give alternate possible chord choices.
2. Write a bass line resulting from your chord choices, working for typical
progressions, strong harmonic intervals, a strong line, typical use of
inversions and nonharmonic tones, and a good contrapuntal relationship
to the given voice.
3. Check your work, analyzing all chords, nonharmonic tones, and vertical
intervals.
Add a new voice above the given voice, mainly in eighth and some
sixteenth notes, using a variety of typical nonharmonic tones, including
suspensions. Use sequences and LIPs as appropriate.
Note: Some of the following basses are unfigured. You should first
determine where inversions are appropriate, based on the scale-degree
patterns in the bass, and then work out the harmonic implications.
Again, solutions to these basses may be improvised at the keyboard, or
using other instruments.
To the lines below, add new bass lines mainly in quarter and eighth notes.
Work for clear harmony, steady harmonic rhythm, strongly directed lines,
typical use of inversions and bass scale-degree formulas. Use sequence and
LIPs where appropriate. Analyze.
To exercises 19 and 20 add new bass lines. Review the instruction before
exercise 11 first. Analyze fully.
Compose two-voice counterpoint based on the following harmonic models.
Analyze fully.
2. These can be shown from the beginning of each beat to the next, or from the end of each divided
beat to the beginning of the next. Both facets should be explored. In the sample here, they are shown
only from the end of each beat into the next beat.
3. Obviously, in 2:1 counterpoint (in which one voice moves twice as fast as the other) there will be
constant oblique motion within the beat. This should simply be ignored in favor of analyzing the
motion from one beat into the next.
4. Whatever the reasons for the virtually universal proscription of consecutive (parallel) fifths and
octaves, the fact is that under these conditions the voices seem to lose their sense of independence.
An occasional exception to the prohibition of parallel fifths can be found in Bach, as a result of
simultaneous nonharmonic tones. Such parallel motions invariably involve short, unaccented tones.
5. In crossing, the two voices exchange places, the higher voice becoming the lower. In overlapping,
one crosses where the other has just been on the preceding beat.
6. Observe, incidentally, the direct fifths (very rare) in m. 7, between beats 1 and 2.
7. There are, of course, questions as to which notes are perceived as nonharmonic. For instance, the d
in the lower voice, m. 1, could be heard as a passing tone, within tonic harmony, making the d2 in
the right hand also nonharmonic, though they are consonant with each other. Whether or not a note
is heard as belonging to the underlying harmony is largely a function of the length and metric
placement of that note, in relationship to the speed of the harmonic rhythm.
8. The conditions under which the P4 becomes a consonant interval (in writing for more than two
voices) will be discussed later.
9. The fact that we are focusing on vertical intervals at this point in our study should not make us lose
sight of the fact that these intervals arise from the combination of lines, which continue to be of
primary importance.
10. For practical reasons, this text uses Jean-Philippe Rameau’s concept of chordal identity, and of
roman-numeral functional analysis. The instructor may wish to discuss at this point Bach’s approach
to teaching, which stressed a more purely intervallic concept—that of figured bass. In two-voice
textures, the underlying harmony is implicit rather than explicit, and results from the intervallic
relations between the voices. The use of roman-numeral analysis should not be taken to imply that
this music is purely harmonic in orientation, or that harmony must precede line in the compositional
process.
11. This text posits for practical reasons the identity of chords in inversion. The instructor may wish to
point out that inversions can often be demonstrated to arise from linear motion in the bass. It should
also be stressed that root position chords are far more stable and have much stronger structural
significance—especially the structural harmonies, T, PD, and D.
12. The larger issue raised implicitly of whether some “chords” exist at all as entities, especially in two-
voice music, is more philosophical than practical; this text makes the assumption, for pedagogical
reasons, that they do. Many of the “chords” in a tonal work, other than the strongest structural T, D,
and PD harmonies, may be understood as arising from line. Such an understanding will be helpful in
focusing the student on the principal harmonic goals of a work, as well as on the primacy of line.
13. Some theorists use the term essential dissonance to refer to a dissonant note that is a chord member,
such as a chord seventh or ninth.
Chapter 3
Chromaticism in Two Voices
To this point the music with which we have been dealing has been largely
diatonic. It is appropriate now to take up the subject of chromaticism, since
the music of Bach is more characteristically chromatic than that of his
contemporaries.1
Before dealing with Bach’s specific chromatic usages, a few general
observations are in order.
Two factors are involved in correct chromatic spelling: direction of
resolution and harmonic background. Generally speaking, notes resolve in the
direction of their inflection: upward-inflected notes resolve up, and
downward-inflected notes resolve down.
Ex. 3-1
Ex. 3-2
Ex. 3-3
The most typical case occurs when the cross-relation involves the use of
the two forms of the melodic minor scale, one form in each voice.
Ex. 3-4
Ex. 3-5
As with all the other musical elements, chromatically altered notes are used
purposefully and consistently. That is, they usually function as part of the
motivic/thematic fabric, rather than as incidental details. Further, they most
often occur within an essentially diatonic structural framework. It is possible
to distinguish two basic categories of usage: nonfunctional and functional
chromaticism.
Nonfunctional (Nonessential, Decorative,
Melodic) Chromaticism
Bach’s music makes sparing use of chromatically altered decorative tones,
usually restricting their use to brief, unaccented passing and neighboring
figures. Dissonances, as we have seen, tend to be brief and diatonic. A few
altered decorative tones are to be found in the examples below, and will be
pointed out as they occur. example 1-31 (p. 19) may be reviewed at this time.
Functional (Essential, Harmonic)
Chromaticism
Bach’s use of chromaticism most often involves functional chords, usually
secondary dominants.
Ex. 3-7
You will have observed that these passages are developmental music,
operate by sequence, and are given harmonic interest by the use of secondary
dominants. example 3-9 also functions on the same basis, following a
sequential pattern around the circle of fifths, again with a developmental
purpose.4 In such passages there is normally a very strong structural bass line.
Ex. 3-9 Kleine Präludien (from the Notebook for W. F. Bach), No. 2
The following reduction of example 3-9 (mm. 6-17) reveals a very simple
and highly directional pitch structure. Note the passing motions, as well as the
parallel tenths apparent in the skeleton graph.
(Note the barely avoided parallel fifths between bass and middle voice in
the fourth and fifth measures of the example.)
Exercises
1. Analyze several more chromatic passages from the anthology, as
found on p. 324 ff. Discuss nonessential chromaticism, functional
chromaticism, harmonic sequence, and the use of structural bass
lines.
2. Spell the following chords, from the root up.
D: V7/V, vii°7/iii, V7/IV, vii°7/ii, V/vi, vii°/vi;
d: V7/iv, V7/lII, V7/VII, vii°7/iv, vii°7/VI.
3. Substitute for the given diatonic chord two secondary dominant
chords with the same function. Consult the chart on p. 283, if
necessary.
2. For background on cross-relations, see The New Grove Dictionary of Music and Musicians, “False
Relations,” vol. 6.
3. It is again assumed that the student is conversant with the theory and practice of secondary
dominants, the Neapolitan chord, and the augmented-sixth chords. If not, a brief discussion in class
should suffice by way of introduction, perhaps with reference to appendix 1 (p. 283). Scale-degree
alterations as they relate to secondary dominants should be discussed.
4. Such progressions nearly always move up by fourth (or down by fifth), rarely in the other direction
around the circle.
5. Further discussion and examples of this technique are to be found on pp. 171 ff and 260 ff. The term
tetrachord is used broadly here to include any stepwise filling in of the descending tonic-to-
dominant leap, not only the diatonic four-note version. Consideration of this material may be
reserved until the discussion of passacaglia and chaconne in chapter 13, or chapter 13 may be taken
up in a preliminary way at this time.
1. Overall form, showing main sections, cadence points, and phrase and
period structure, using a format similar to this one:
Ex. 4-3
2. An instructor preferring to speak of “temporary tonicizations of the tonic triad members D and F♯”
should feel free to do so, as this text does not wish to require any particular view of modulation.
3. Definitions and further details are to be found in The New Grove Dictionary of Music and
Musicians, under each suite movement, and in Gauldin (see bibliography).
Chapter 5
Double (Invertible) Counterpoint
Directed Study
Play and discuss the excerpts above. Compare position I of each to position II.
How does II relate to I? What is it about the vertical (harmonic) intervals used
in I that makes II equally successful? Generalize about what you find here in
terms of harmonic intervals as well as rhythmic, directional, and motivic
relations between the voices.
Discussion
The above excerpts are written to sound equally satisfactory with either of the
two voices as the upper, that is, in the versionsBA andAB. These passages are
written in double or invertible counterpoint.1 The technique of double
counterpoint is widely used by composers. Without it, it would be far more
difficult to compose extended contrapuntal works, as one would constantly
have to invent new accompanying material. The use of double counterpoint
ensures a consistent thematic content and a contrapuntal relationship not
achievable by other means. Composers often invent themes in pairs, with one
theme intended to accompany the other at each occurrence. We will study this
relationship later (as that of subject and countersubject). The use of double
counterpoint also ensures that a theme will not always have to appear in the
same voice, but can be freely interchanged between voices. Each voice must
be capable of functioning as a convincing bass line and must as always be a
satisfying melodic line in itself. In any given work, the two versions
(positions) may follow one another immediately, as they do in the music
above, or the second (voice-exchanged) version may be placed at some later
point in the work, usually transposed to some other key.
Bach employs three types of double counterpoint, which we will examine
in order of importance.
Double Counterpoint at the Octave or
Fifteenth
By far the most common type of double counterpoint is shown in examples 5-
1 and 5-2, where the voices exchange positions, moving by a total of one,
two, or three octaves. This is known as double counterpoint “at the octave” or
“at the fifteenth.” (The number of steps in two octaves adds up to 15, not 16.)
It is also sometimes called “natural double counterpoint,” in distinction to that
at other intervals. In example 5-3, an original pair of voices (a) has been
exchanged at the octave (b) and at two octaves or the fifteenth (c). In practice,
either voice may be transposed by one or two octaves to cross the other, so
that either version of (b) or (c) is possible, depending on considerations of
register or shaping of lines.
Ex. 5-3
You will observe in example 5-3 that, under conditions of voice exchange
at the octave or fifteenth, most of the harmonic intervals retain their character.
That is, imperfect consonances remain imperfect (thirds become sixths, and
vice versa), dissonances remain dissonant (seconds become sevenths, and vice
versa), and the octave and unison exchange, as do diminished and augmented
intervals (A4 becomes d5; d7 becomes A2).
In neither position (the original or the exchanged version) are the two
voices allowed to cross, as this would obscure the double counterpoint, and
the independence of each voice. For this reason it will often be necessary to
exchange the voices at the fifteenth (or even at three octaves), rather than the
octave. It is also wise to keep the range of each voice within an octave.
The only interval that takes on a different character is the perfect fifth,
which becomes a perfect fourth and thus changes from a consonance to a
dissonance. Therefore, in the original version, perfect fifths must be treated as
if they were dissonant—in other words, in typical nonharmonic tone idioms.
Ex. 5-5
Ex. 5-6
The 7-6 becomes 2-3 (and vice versa); but the 4-3 becomes 5-6 which,
while acceptable, is less effective since the perfect fifth is not a dissonant
interval.
Below is a table of intervals for double counterpoint at the octave. Observe
that each interval-pair adds up to nine.
Double Counterpoint at the Twelfth
Here the lower voice of the original is transposed up an octave (or two) and
the upper voice down a perfect fifth, for a total of twelve steps, as in example
5-7, position II. Or the upper voice may be transposed down an octave (or
two) and the lower voice up a perfect fifth for the same intervallic result
(example 5-7, position III). Double counterpoint at the twelfth is typically rich
in thirds and tenths, since these two imperfect consonances are inversionally
complementary at the twelfth. Perfect consonances remain perfect (1→12,
5→8, and the reverse), and most dissonances remain dissonant (2→11, 4→9,
and the reverse). The problem interval here is the sixth, which becomes a
seventh and must therefore be treated as if it were dissonant. Observe in
example 5-7 that the sixths are treated as passing tones in the original position
(I) so that they become properly treated dissonances (sevenths) in positions II
and III. Also note that the 11-10 (equal to 4-3) suspension at the end becomes
a 2-3 suspension, indicating the invertibility of both the 4-3 and the 2-3
suspensions at the twelfth.
As with double counterpoint at the octave, the voices are not allowed to
cross each other in either the original or the exchanged positions, and the
individual voice ranges are often kept within one octave.
Ex. 5-7
In example 5-8 the voices have exchanged places in the second version,
with the upper voice transposed down an octave and the lower voice up a
perfect fifth, for a total of twelve steps. Note that each voice sounds equally
good as the lower voice, and that each voice has considerable independence
of motivic content as well as integrity and shape. The individual lines to be
used in double counterpoint must of course display all the features of good
linear writing that we have discussed. Here again, thirds and tenths
interchange, all dissonances are handled normally in both versions, and the
sixths in position I are treated as if they were suspensions (ornamentally
resolved on the next beat), so that they become correctly treated sevenths in
position II.
Ex. 5-8 Well-Tempered Clavier I, Fugue No. 2, mm. 5-6
In example 5-9 the upper voice has been transposed down an octave, and
the lower voice up a fifth (plus two octaves, to maintain registrai
independence from the other voice, and avoid crossing), for double
counterpoint at the twelfth. This example demonstrates well the
interchangeability of fifths and octaves.
Ex. 5-9 The Art of Fugue, Fugue No. 9, mm. 59-62 (outer voices only)
In example 5-11 the upper voice has been transposed down an octave, and
the lower up a third, crossing to form invertible counterpoint at the tenth
(octave plus third equals tenth). Observe in this excerpt (which has been
simplified for demonstration) the preponderance of tenths, fifths, and octaves
in position I, which become unisons, sixths, and thirds in the inverted form
(position II). Note also the considerable use of contrary and oblique motion,
and the complete absence of parallel motion, arising from the limitations of
this technique.
General Comments
The following four points should be kept in mind when writing double
counterpoint.
Ex. 5-12
Ex. 5-13
5. Add new voices to the lines given above (in exercises 3 and 4),
using double counterpoint at the twelfth and tenth. Analyze the
harmonic implications and vertical intervals.
Notes
1. Double will be the preferred term here, as invertible often seems confusing because of its
association with melodic inversion. Invertible counterpoint in three voices will be referred to as
“triple invertible counterpoint.” For further information on double counterpoint, consult The New
Grove Dictionary of Music and Musicians, “Invertible Counterpoint,” vol. 9.
Chapter 6
Imitation; Canon
Directed Study
Perform and discuss the excerpts above, focusing on the thematic relationship
between the voices. They share the same material, by the process we call
imitation, in which the two voices begin the theme at different times, and
often on different pitches. When analyzing imitation, we need to focus in turn
on several different aspects of the music:
Discussion
Ex. 6-6
A. The voices can end together (as in example 6-5), the canon
continuing up to the cadence.
B. The leader can continue to the end, the follower breaking off before
the end and continuing with free (nonimitative) material to the
cadence. Or the leader may break off, the follower continuing its
imitation to the end.
C. Both can break off before the end, finishing with nonimitative
figuration, as in several of the Goldberg Variations canons.
D. The canon can continue indefinitely, if it is so written (using repeat
signs or a verbal instruction to that effect). This is an “infinite” or
“circular” canon. One of the canons in A Musical Offering is of this
type.
E. The voices can simply trail off as each finishes the theme, so that
the canon ends with one voice alone. Vocal canons (rounds) often
end this way.
1. Compose the leader up to the point at which the follower enters, keeping
in mind harmonic clarity and simplicity, shape, motivic unity, and
rhythmic drive.
2. Bring in the follower in an appropriate place in terms of time and pitch
interval.
3. Continue the leader against the follower, keeping in mind the precepts of
good two-voice counterpoint as regards vertical intervals, clear harmony,
idiomatic nonharmonic tones, and rhythmic interplay. This line must
represent a smooth, logical continuation of the opening of the leader.
4. Continue the follower exactly as you did the leader in example 3, above.
Here, though, you may begin to make subtle adjustments (by accidental
only, not note) to accomplish a modulation and/or avoid tonal monotony.
See example 6-8, m. 5, lower voice c ♮.
5. Meanwhile, the leader will continue with its counterpoint, reinforcing
any new tonal direction taken by the follower. The canon may be
continued in this way for as long as the materials suggest. Be sure to
return to the tonic key before the end.
Ex. 6-8
Ex. 6-9
One further point about imitation (it will be covered in more detail under
fugue, pp. 188 ff): the imitations we have been studying have mostly been
real (see examples 6-1, 6-2, 6-4, 6-7, and 6-8)—that is, the melodic intervals
are the same number-value for leader and follower. In most typical cases of
tonal imitation, a tonic-to-dominant leap at or near the beginning of the leader
is answered by a dominant-to-tonic leap at the comparable place in the
follower, and vice versa. There are other aspects of tonal imitation, but this
statement covers most cases. In canons real imitation is preferred. Changes of
accidentals only, often needed for harmonic (scalar) reasons, do not constitute
tonal imitation. See example 6-4, m. 3, beat 1.
Ex. 6-11
Variants in the Imitative Process
While it is not the intention of this text to be exhaustive in its coverage of all
possible forms and procedures, still it will be of interest to show examples of
several special types of imitation, as these will become useful later as devices
for development within larger forms, especially in fugal composition. Any of
these devices can be carried out consistently through a work, making it a
canon, or one may appear as an incidental contrapuntal detail in the course of
a larger work.
In imitation by melodic inversion (motion contrario, or contrary motion),
the follower is a melodic inversion of the leader. example 6-12 imitates by
contrary motion at the fourth below, at three beats. There is a free, supporting
bass.
example 6-13 is a canon by contrary motion at the fifth above at two beats,
with a free (non-imitative) bass line. In canons at the fifth above (or fourth
below) the follower often enters with dominant harmony.
Ex. 6-13 Goldberg Variations, Variation No. 15
Ex. 6-14
The excerpt below (example 6-16), which Bach describes as a “canone per
augmentationem in motu contrario,” is for only two voices, with no freely
supporting voice.
Exercises
1. Briefly define the terms imitation, canon, stretto, contrary motion,
real and tonal imitation, augmentation canon, double canon,
retrograde canon, and round.
2. Make a step-by-step list of the process for writing a canon.
3. Analyze several examples of strict canonic imitation, as selected
from the literature by your instructor. The following are the major
collections of canons by Bach:
2. A segmental approach is suggested here as a practical and effective way of learning to compose
canons. The danger of this process is that the student may lose sight of the shape and continuity of
the canonic melody, and these aspects may need periodically to be brought to his or her attention.
The aim is to develop the ability to construct a melody that will form good counterpoint against
itself. To this end it will be wise to play all the way through each canonic voice of the examples in
this chapter and discuss its rhythmic, intervallic, and harmonic construction as these aspects relate to
its suitability as a canonic theme.
3. More information on the history and literature of canon may be found in The New Grove Dictionary
of Music and Musicians, “Canon,” vol. 3.
Chapter 7
The Two-Voice Invention
The fifteen Two-Part (that is, Two-Voice) Inventions of Bach stand among the
very few major collections of two-voice contrapuntal works. Originally
intended as keyboard and compositional studies for his son Wilhelm
Friedemann Bach (1710-84), the Inventions are a definitive catalog of devices
for the making of a great deal out of very little musical material.1 In some
ways, two-voice imitative writing is the most difficult compositional
discipline, as one must make the harmony clear and satisfying with a minimal
texture, and without the use of any “filler” material. It is perhaps in their
works in thinner textures that the technique and imagination of the great
composers come through most clearly.
Bach’s inventions (his so-called Three-Part Inventions we will call
Sinfonías) are the most efficient music: texturally lean, balanced between the
voices, and compact. There are no superfluous notes and no purely
accompanimental patterns. As with fugue, it is not accurate to speak of
“form” in an invention. It is a procedure rather than a form, arising out of the
principles of imitation and motivic manipulation, as well as certain principles
of tonality. Because of this, we will need to be careful not to generalize too
much from what we find in any given invention. There are, however, some
generalizations one can safely make about all inventions, and that is the
purpose of this chapter.
The Exposition: Theme and Countertheme
Below are given the expositions of several inventions. Play them, and discuss
them as directed.
Directed Study
Analyze these expositions in the following terms:
1. Identify the theme (T) and countertheme (CT). There may or may not be
a consistent CT (the counterpoint heard against the T). In what ways is
the T-CT relationship an example of good counterpoint?
2. Analyze the imitation intervals (both pitch and time) between the voices.
For how long is the imitation carried on strictly?
3. Analyze the harmony and nonharmonic tones. Is the tonic made clear? Is
the harmony simple and functional? In what ways is the CT used to
clarify the harmonic implications of the T?
4. What motivic ideas seem to comprise each T? Each CT?
5. Based on what you hear and see in these excerpts, what would you say
are the characteristics of a good T? A good CT?
6. Graph the form of each exposition. Here as a sample is a graph of No. 1.
T CT T CT
T T
Discussion
The first section of an invention, in which the thematic material is presented,
is called the exposition. It consists of from two to four statements of the
theme, often with a countertheme in the other voice.2 As you have seen, there
are various possible layouts for an exposition, in fact even more than appear
above. The type of invention we will be concentrating on is typified by
Inventions 1, 3, and 4, in which the T is heard first in the upper voice, alone,
and imitated immediately by the lower voice at the octave, while the upper
voice continues with the (optional) CT.3 There are several variants of this
process, though. For instance, in Invention No. 9 both T and CT appear
simultaneously, in double counterpoint, exchanging voices in mm. 5-7.
Numbers 5, 6, and 9 are of this type. In several of the inventions the lower
voice is used at the opening to establish key and downbeat, though in No. 13
this voice has a version of the fragmentary CT (m. 1, beats 1-2). And two of
the inventions, 2 and 8, have extended expositions in canon.
While there are several ways, then, of organizing an exposition, some
generalizations can be made.
1. The T is a relatively short musical entity (usually 2-6 beats; never longer
than four measures), with all the features of a workable musical theme:
2. The CT, when there is a consistent one, exhibits all the features of good
counterpoint in this style:
Inventions with relatively long T’s (Invention No. 2, for instance) tend to
use a type A exposition. Only with a very short T (two to four beats) will the
“extra” entries of types B and C not seem redundant. One way of avoiding the
feeling of redundancy inherent in type C is to transpose the T and CT to the
dominant, as is done, for example, in Invention No. 1 (m. 2). This will work
especially well if the T ends on the dominant note.
Here is an analysis of the exposition of Invention No. 4.
Ex. 7-7
Exercises
1. Analyze and graph several more expositions as suggested on p.
125. Several inventions can be found in the anthology.
2. Compose themes based on these chord formats, based on the
observations above:
e: i | iv V | i or: T | PD D | T
2. T written:
4. Exposition composed:
Points to note are clear harmony, strong vertical intervals (tenths and
sixths), typical nonharmonic tones, clear shaping, 2:1 rhythms, T and CT
eliding smoothly, motivic material suitable for manipulation later.
The Episode
It is in the episodes that a composer’s inventiveness and technique are most
evident. Bach’s inventions are a great compendium of ways in which thematic
material can be manipulated and combined.
The purpose of an episode is developmental and modulatory. It subjects the
thematic material of the invention to processes of transformation and
combination, exploring the developmental potential of the theme. It serves the
function of a modulatory passage, leading away from the tonic key that
unified the exposition tonally. Tonal works tend to modulate; the ear quickly
becomes fatigued with one key area, and it is necessary to settle at least
momentarily in subsidiary keys.
The first episode follows smoothly after the exposition, without pause, such
that the ear may not realize for the first few beats that an episode has begun,
especially as the thematic content of the exposition is continued here. The last
beat of the exposition is the first beat of episode I. The first episode will last,
depending on the meter and length of the theme, for four to twelve measures,
and will end in a strong PAC in the relative key (in minor key inventions) or
dominant key (in major key inventions). The first episode is often written in
double counterpoint, and may be reused in voice-exchanged form as a later
episode. It is usually directed downward by step or third, by sequence.
A variety of processes is used in the episodes, including:
This episode makes use of repetition (mm. 5-6 are repeated as mm. 7-8), a
device not often employed in this music, as it can easily produce a static
effect. It is used here in conjunction with a pedal note in the lower voice,
which serves to establish the new key, A major, by reiteration. Note the
imitation between m. 9, upper voice, and m. 10, lower voice, and note further
that the two voices are written in double counterpoint. (The voice-exchanged
version is used later.)
This episode makes use of an idea that, while not exactly that of the main
theme, still seems closely related by interval and shape—and, of course,
rhythm. This new idea, introduced in m. 3, becomes almost a second theme
for this work. Again, both voices are imitative, sequentially organized, and
controlled by the circle of fifths. The upper voice has a sequence unit of four
beats, transposed down by step in mm. 3-4. In mm. 5-6, it is halved by
fragmentation into a two-beat unit, again transposed down by step. The lower
voice imitates at two beats, a fifth (twelfth) below, then produces its own
fragmentary version of the theme in m. 5, sequentially treated, and cadences
in m. 6. The effect of the shortening of the sequential unit in m. 5 is to give a
sense of excitement and drive toward the cadence, and a stretto-like effect.
Incidentally, you may have noticed that the initial sequences of most works
tend to be transposed down. Sequences toward the ends of inventions (and
fugues), especially a final sequence, will tend to be transposed upward,
building toward a climax.
Prepare a linear reduction of this passage and note the clear descending
stepwise structures and (slightly hidden) 10-10 pattern.
A more extended stretto is to be found in one of the episodes of Invention
No. 14.
Ex. 7-12 Invention No. 14, mm. 12-13
This is a stretto at one beat at the fifteenth, progressing partly around the
circle of fifths (roots: C-F-B♭-E♭).
For a superb model of extended canon as the basis for an invention, see
Invention No. 2, on p. 320 of the anthology.
You will have noticed the strong sense of harmonic/linear direction these
episodes exhibit as they drive downward from one strong beat to the next and
finally into the PAC that ends the first episode. As is typical when sequence is
involved, these sections tend to be organized around very highly directed
structural pitch schemes, and often operate on the harmonic basis of the circle
of fifths. The circle also forms the basis for the sequences in example 7-11
(mm. 3-4, roots: A-D-G-C; mm. 5-6, roots: A-D-G-C-F). The structural pitch
lines of example 7-11 are quite clear:
Ex. 7-13
Exercises
1. Analyze several more Invention episodes, as found in the
anthology on pp. 319 ff. Discuss them in detail, as suggested on pp.
123 ff.
2. Based on your observations of Bach’s processes, and on the
demonstration above, write a number of motivic manipulations, in
one voice only, applied to:
Cadences
The cadence figures used may or may not relate in an obvious way to the
theme. Further, within a given invention they may all be similar melodically,
or they may not. For example, in Invention No. 1, the first two cadences
resemble each other, but the last is quite different. No two are literally the
same, nor is any necessarily drawn from the theme in any obvious way,
beyond very general similarities of figure or rhythm. Cadence idioms or
formulas, in other words, are usually more style-defining than distinctive to a
particular work. Rarely are they an important unifying factor within any given
piece.
Ex. 7-15
In Invention No. 3, however, all five cadences (mm. 11-12, 23-24, 37-38,
52-53, and 57-58) have the same figuration.
Authentic cadences will be placed quite regularly within a work—in a brief
invention about every six to eight measures, in a longer one roughly every
twelve to sixteen—since the sections in a larger invention are proportionally
longer. There may be a deceptive cadence placed near the end, replacing an
expected PAC at or shortly after a climactic point, in the tonic key.
Halfcadences are found infrequently in the inventions, and indeed are far less
common in Bach’s music than authentic cadences.
Episodes
In general, the later episodes will tend to be longer, more complex in
technique and more tension-producing, tending more than the earlier ones to
such devices as imitation, stretto, and upward-moving sequence. If episode I
is written in double counterpoint, it may well reappear as episode III or IV,
with the voices in their exchanged positions, and even (as in Invention No. 1)
by contrary motion (melodic inversion).
Middle Entries
The theme is likely to return literally, in some closely related key, at least
twice in the course of the invention. These returns (middle entries) are
distinguished from episodes by the fact that they contain the entire theme (not
just fragments and/or manipulations of it), though the ear cannot always
distinguish in context between a middle entry and the beginning of an
episode. The music immediately following the end of the first episode is
usually a middle entry in the dominant or relative key, with the T heard first in
the lower voice. For typical middle entries, see Invention No. 1, mm. 7-8; No.
6, after the double bar; No. 8, mm. 12-13; No. 13, mm. 6-7. Middle entries
usually contain just two entries of the theme, although one entry is sufficient.
The two entries may be imitative at the octave, with the lower voice leading.
Sectional Elisions
It is important in writing to assure that the sections are elided and continuous.
The juncture between the exposition and episode I in particular is
accomplished by elision. There are no “dead” or thematically empty beats
here, and cadences are immediately followed by important thematic material.
There are climactic moments around mm. 34—37 and 42-45, with
textural/harmonic/rhythmic tension concentrated around mm. 46-50, and a
rapid tonic octave descent following the DC in m. 49.
A structural-pitch graph of mm. 7-18 follows. Play this reduction and then
the music again.
Ex. 7-22
Exercises
Note: It is suggested that these exercises be done in the given order, and
that each stage of work be checked and approved by the instructor. A
good deal of one-on-one work may be required. It is also suggested that
all written work fit within the range of Bach’s keyboards, or exceed that
only slightly.
3. It would be possible to take as our model either the extended canonic type or the type opening with
extended double counterpoint. The approach typified by Inventions 1, 3, and 4 has been selected as
best approximating the conditions found in fugal writing, and thus as the most useful preparation for
it.
4. Because of their brevity, these ending sections are sometimes termed codettas.
Chapter 8
Three-Voice Counterpoint I: Texture,
Rhythm, Harmony
Perform and analyze the following excerpts as suggested on pp. 147 ff.
Perform the voices separately, then together. These may be performed on
keyboard or on three instruments, or even vocally, as ranges permit.
Directed Study
Analyze and discuss the above excerpts, including texture, rhythm, harmony
and counterpoint.
TEXTURE. Are the voices equally important and active? Does any voice
seem to be harmonic “filler”? Are the voices at times paired with each other?
Which pairings seem most common?
Consider spacing. What is the widest interval between adjacent voices?
What seem to be the typical spacings? Do you find any voice-crossing? Are
the voices always playable by two hands?
What would you say about the range of each voice? Do the voices in
general have a wider or narrower range than in two-voice writing?
Do the voices share the same motivic material? Is there any imitation? If
there is, analyze it in the usual ways.
RHYTHM. Make a rhythmic chart of several excerpts, as below. Generalize
about what you find. Here is a sample, based on example 8-1, mm. 1-3.
Ex. 8-13
Are there moments when all the voices move in the same values? For how
long? Do they ever move together in the fastest available value?
Are the meter and pulse always clear? How is this accomplished?
Investigate the use of rests, ties, and syncopes. What rhythmic and textural
functions do they appear to have?
HARMONY AND COUNTERPOINT. Analyze the harmonic intervals in
several selected excerpts, between each pair of voices. Are there any new
essential intervals? Did you find any linear intervallic patterns in two or all
three voices?
What types of chords do you find, in what positions and with what
doublings and/or omissions of chord tones? Analyze five or six excerpts in
detail.
Does there appear to be any difference in harmonic vocabulary between
two- and three-voice counterpoint? If so, what chords are new here?
What nonharmonic tones do you find? Are there any new idioms? Are
there examples of simultaneous nonharmonic tones? Consider in detail the
use of suspensions, including what the nonsuspending voices do at the points
of suspension and resolution. Are there places where two voices suspend
simultaneously?
Consider the contrapuntal relations between each pair of voices. Which
intervals are used in parallel motion, and for how long? Which are not? Do
you find any instances of direct or unequal fifths, or direct octaves? If so,
which pairs of voices are involved?
Note what the three voices do, melodically, at cadence points. Write down
in musical notation those cadential melodic idioms that appear to be typical,
including scale-degree numbers.
GENERALIZATIONS. Based on this sample, what would you say are the
principal features of good three-voice counterpoint as regards texture, rhythm,
harmony, and any other aspects of contrapuntal relationship? Make a list in
class.
Discussion
Three-voice writing is in many ways the ideal texture for counterpoint. There
is, compared to two-voice composition, an added richness and explicitness of
harmony, more rhythmic variety, and many more possibilities for contrapuntal
combination. Composers at the time of Bach show a great fondness for
counterpoint in three voices, which can be said to be an even more
fundamental texture than four voices. Mastery of three-voice writing leads
easily to mastery of four; there are no basic techniques to be learned in thicker
textures that cannot be learned as well in three voices. And many four-voice
works contain substantial passages of three-voice writing.
There are also some dangers here for the student. One may notice a
tendency to think only harmonically, to lose sight of the primacy of line, to let
the inner voice degenerate into aimless filler, or the lowest voice into purely
harmonic support. New problems of doubling, rhythm, and dissonance
treatment may occur. But these problems are overcome with application, and
the rewards of mastery of three-voice composition are great.
Texture and Rhythm
The excerpts at the beginning of this chapter have been selected to show a
variety of possible textures, and rhythmic and motivic distributions. Some
feature roughly equal voices, in terms of motivic content and rhythmic
activity (examples 8-8, 8-9, 8-12). Five of them pair the upper voices against
a rather less active bass (examples 8-2, 8-3, 8-5, 8-6, 8-10). Two exhibit a
somewhat predominant upper voice (examples 8-1, 8-4), and one an active
bass against paired upper voices (example 8-7). You will have noticed,
though, that these textures are likely to change in the course of just a few
measures. Flexibility of texture is a characteristic of much three-voice writing
in this style. It is possible, through rests, to reduce the texture to two voices
for a few beats at a time. Reduction to one voice is very rare. It is also
possible to thicken the texture in writing for keyboard, for example, at the end
of a movement. example 8-11 shows a typically free keyboard texture,
varying from two to four voices, though it is essentially a three-voice work.
This is a good example of Freistimmigkeit—literally, “free-voicedness.” Your
writing, at this stage, should in general be strict about the number of voices
employed, limiting textural freedom to an occasional passage in which the
texture reduces to one voice (sometimes doubled in octaves), often just before
a final cadence; or thickens to four, five, or even six voices at the final
cadence. Both options are shown in example 8-14, from a work that is
preponderantly in two-voice texture. All your writing should be playable in
two hands, so that the middle voice will always be within an octave of one of
the outer voices.
Textures in which two voices are equally active, with the third voice (often
the bass) moving in different note values are common, but it is rare to find
extended voice pairing unless it characterizes the texture of an entire work, as
in example 8-5. Brief upper-voice pairings (that is, trio-sonata texture), such
as those found in examples 8-3, 8-6 (see especially mm. 5-7 for a typical
texture), 8-7, and 8-9 (mm. 1, 2, and 4) are very common, especially in fugal
episodes, where the upper voices may be in canon. The kind of pairing in
which two voices, often the upper two, move in parallel thirds or sixths in the
same rhythmic values is contrapuntally weak, and should not be continued for
long unless one intends a homophonic effect (see example 8-5). In example 8-
16, observe that brief homophonic passages (mm. 1-2) may alternate with
those that are more contrapuntally distinct (mm. 3-4, 6-8), providing textural
variety and effectively bringing the contrapuntal passages into relief.
1. There are very few passages in which all three voices will move
simultaneously in the same values, as this quickly becomes homophonic
in effect. It does not occur at all in the examples heading this chapter,
and almost never occurs in the shortest available value. Two-voice
homophony, as is noted above, is confined to brief passages, unless it
characterizes an entire movement.
2. A great variety of rhythmic relationships is available, and the fastest
motion is often shared in alternation between two or among all three
voices. Here are three rhythmic reductions taken from the excerpts
beginning this chapter.
Ex. 8-17
These passages, and others like them, should be studied from this point of
view, as they represent an ideal state of rhythmic independence.1 Be aware in
such passages that the shortest note value is passed among the voices, and that
on any given beat there will be two or three different durations in the three
voices. Note especially the use of short rests and ties to break up the
regularity of pulse in each individual line, avoiding the square or plodding
rhythm that would result from each voice attacking the beginning of each
beat. Groups of eighths or sixteenths often start after a short rest or tie.
Ex. 8-18
These devices give a needed flexibility to the rhythm and should be used
freely, as long as the sense of motion is maintained by at least one voice.
In this music the pulse is consistently maintained, and the resultant rhythm
(which some theorists call the combined rhythm or macrorhythm) is very
regular, as can be seen in example 8-17.
Brief rests can be used freely in a line, longer rests more sparingly. A voice
should come to a point of melodic completion before a longer rest. Thus, a
long rest should never be preceded by a note requiring resolution. One should
not hesitate to employ brief rests. They are the windows and doors of music,
letting in air and light.
Ties are normally found in only one voice at a time, though two may tie
simultaneously as long as the third voice continues to move steadily. The
combination of regular overall momentum (resultant rhythm or macrorhythm)
with the individual rhythmic flexibility of each voice is an essential ingredient
of all good counterpoint, and should be kept in mind while composing.
Exercises
1. Critique the rhythmic aspects of the error-correction exercise 3 on
p. 163.
2. Perform, by conducting and intoning the rhythms on a neutral
syllable such as “ta,” several three-voice movements from the
anthology. Discuss the metric and rhythmic aspects of these works.
Is the meter clear from the rhythms only, or is harmonic rhythm
also a factor?
Range and Spacing
The individual voices will often be more restricted in three-voice writing than
in two, so the use of melodic material involving wide-ranging arpeggiation
should be limited. Brief voice crossings or overlaps are acceptable, but are
rare. Be aware that crossing the middle and lowest voices will result in the
creation of a new bass line, and chord inversions must then be calculated from
this new line.2 While spacings of more than an octave, especially between
middle and lowest voice, are found, it is not effective to sustain such passages
too long, as this will tend to isolate one of the voices. The spacing is
common, the spacing less so, as the result quickly becomes muddy, or
isolates the upper voice. An even distribution is the ideal texture for
counterpoint:
In keyboard writing, the middle voice typically “migrates” between the
hands, inhabiting both alto and tenor registers. Be sure your writing is
playable in two hands, and conforms to the range of Bach’s keyboards (at the
discretion of the instructor).
Relative Motion
The restrictions on parallel motion still hold true for writing in three voices.
In all such writing, each of the three voice pairs must be checked in the same
way that one checks between the voices in two-voice counterpoint. The
following technical details should be kept firmly in mind:
Exercises
1. Critique the contrapuntal aspects of exercise 3, p. 163.
2. Critique the exercise below in terms of spacing, rhythm, and
counterpoint. There are about twenty errors.
Harmony
As the harmony at this stage is fuller and the vocabulary slightly richer, a few
additional comments on harmony are appropriate.3
VOCABULARY. In your analyses of the excerpts beginning this chapter, you
will have noticed that, once the nonharmonic tones have been taken into
account, the majority of vertical sonorities are major and minor triads in root
position or first inversion. The diminished triad is usually found in first
inversion (typically in the progressions vii°6-I or vii°6-I6). Augmented triads,
except those occurring momentarily as the result of nonharmonic tone
activity, are extremely
Seventh chords are slightly more common in three voices than in two, as it
is easier to express them with a greater number of voices. Both dominant (V7,
vii°7) and nondominant sevenths are used (ii7, etc.), though not as often as are
triads. As in two-voice texture, sequences involving either secondary
dominant or nondominant sevenths are found, especially in episodic passages.
DOUBLING, VOICING, AND COMPLETENESS. With major and minor
triads, the following voicings and doublings are found, in descending order of
frequency:
complete triad
two roots, one third
one root, two thirds
one third, two fifths
two thirds, one fifth
three roots (at the end, or briefly on weak beats or weak fractions of
beats)
One cannot successfully omit root or seventh for more than a very brief time,
as the chord then changes identity.
It should be reiterated here that the less common voicings and doublings
occur mainly on weak beats or parts of beats, and that considerations of line
take precedence over doubling. The following chart (example 8-20) thus
applies most accurately to sonorities found on strong beats, or the strong half
of the beat.
Ex. 8-20
Ex. 8-21
Ex. 8-22
Ninth chords are extremely rare in three-voice writing. The note sounding
the ninth can always be understood as nonharmonic.
The perfect fourth between the upper notes of the first inversion major or
minor triad is treated as consonant, because of the fact that both these notes
are consonant with the bass note. This is the so-called consonant fourth.
Parallel perfect fourths are possible in writing with more than two voices, in
successive first-inversion triads. Such triads normally move by step, most
often descending. Bach rarely uses more than three such triads in succession,
as the effect tends toward homophony.
Ex. 8-23
Ex. 8-24
3. The passing is also a brief, weak-beat chord, used with a stepwise bass
line to fill in between two different positions of the same chord. Again,
the notes involved may simply be analyzed as nonharmonic tones.
Ex. 8-25
4. The arpeggiated bass line is hardly a true inversion but merely the
result of arpeggiation in the bass. Like the neighboring , this idiom is
fairly rare in the Bach style.
Ex. 8-26
Exercises
1. Discuss the harmonic problems with exercise 3 on p. 163.
2. Discuss the following exercise in terms of chord choice, doubling,
chord completeness, use of inversion, and resolutions. This can
also be done by ear, playing the exercise very slowly. There are, as
usual, a great many errors of all kinds.
3. Analyze and resolve the following isolated chords. Analyze the
dissonant intervals in each sonority, and be sure to resolve them
normally. These can also be done as ear-training exercises, with
each voice of a chord sung and resolved.
Ex. 8-28
Nonharmonic Tones
All the nonharmonic tones employed in the style are, of course, available.
They are used carefully so as not to obscure the underlying harmony nor to
create harsh effects, which would cause unwanted harmonic tension. As
always, the nonharmonic tones in each voice are calculated against the
underlying harmony. Two technical points should be made:
Ex. 8-29
Cumulative Exercises
In completing the written exercises below, it is advisable to avoid extensive
chromaticism, rigorous imitation, stretto, and triple counterpoint, as these will
be studied in chapter 9.
The instructor may wish to specify textural and tempo models from the
anthology—for instance, a suite movement or prelude.
Process Demonstration
A. Cadential frameworks
2. In music employing double bass or organ pedal with 16′ pitch, the tenor may only appear to cross
the bass.
4. As suggested earlier, this text assumes the identity of chords in inversion. At the same time,
“inversion” can often be understood (and heard) to arise from nonharmonic tone activity in the
upper voices or the bass. Such inverted sonorities are of far less structural significance and stability
than chords in root position, especially the primary triads (I, IV, V).
5. This is a point at which students weak in theoretical training may need a brief review of the spelling
and resolution of seventh chords, which is best accomplished by drill on bass scale-degree idioms
6. Again, the choice of whether to analyze ’s as such, or to understand them as the result of
simultaneous nonharmonic activity, is up to the instructor. They are bracketed in the analyses,
indicating a sonority of linear origin, or may, when cadential, be shown as , or simply V with
nonharmonic tones in the upper voices.
Chapter 9
Three-Voice Counterpoint II:
Chromaticism, Triple Counterpoint,
Canon
Chromaticism
There are no specific techniques of chromaticism to be learned now that have
not been discussed earlier, yet the addition of the third voice provides both a
clarifying and a complicating element.1 The opening of Sinfonia No. 9 is
given in example 9-1, with a partial harmonic analysis. Perform and analyze
it, with attention to structural lines and harmony, as well as details of harmony
and nonharmonic tones. Issues of performance practice and affect may also be
discussed, as appropriate.
The application of roman numeral symbols to this music should not be
understood to imply that harmonic forces have necessarily generated it, or
have more importance than line. The point is that the lines can be understood
to imply harmonies that are susceptible of “functional” analysis. The
reduction graph in example 9-3 is intended to clarify the linear orientation of
the music. It would be wise to make a detailed reduction in class, showing the
linear structure of all three voices, which is highly directed and clear.
Ex. 9-1 Sinfonia No. 9
Ex. 9-2
Ex. 9-7
Cross-Relation
The matter of cross-relation occasionally arises in chromatic passages in Bach
and should be dealt with briefly again. Cross-relation almost always results
from the use in proximity of the two different forms of the melodic minor
scale in two different voices.
On beat 2 the b♭ in the alto voice descends to a, and the B♮ in the tenor
ascends to c# their normal resolving tendencies within the key of D minor.
The two pitches are heard in very close proximity, creating a cross-relation.
Simultaneous cross-relation, in which both pitches are heard at the same time,
is very rare in Bach.
Ex. 9-9
Note in such passages that the cross-relation is brief, that at least one note
of the relation is nonharmonic (usually a passing tone or neighbor), and that
both resolve normally.
Exercises
1. Analyze the portions of examples 9-1 and 9-4 that have not yet
been analyzed. Distinguish between structural and embellishing
harmony, and prepare detailed linear reduction graphs. Also
discuss those parts that have been analyzed in the text, and discuss
alternative analyses if any seem possible.
2. Analyze chromatic passages from the anthology, including their
keys, cadences, chords, and nonharmonic tones. Include a
reduction of the structural pitches, especially in the bass line, and
identify structural harmonies.
3. Articulate the frameworks below, using mixed note values and
three roughly equal voices. Work for flow and motivic coherence.
Analyze in detail. Provide figured bass symbols, and note the
structural harmonies.
The instructor may wish, in exercises 3-5, to specify
instrumentation and texture. Any of these exercises may follow a
specific model from the anthology, such as a sinfonia, prelude, or
suite movement. They may also be used for improvisation at the
keyboard.
4. Work out the figured and unfigured basses below in three voices.
Analyze the harmonic implications first. The upper voices should
be equally active, in mixed note values. Work for motivic unity
through informal (motivic) imitation. Analyze fully, including any
instances of patterning.
5. Alter the following framework with secondary dominants, then
articulate. Analyze.
It is unlikely that all six positions will be exploited in any given fugue, but
their availability will make its composition a good deal simpler. Fugal
episodes are also often written in triple counterpoint (just as invention
episodes are often written in double), and are therefore capable of being
reused later in the fugue. It is common to find double counterpoint plus one
free (noninvertible) voice in fugue expositions and episodes.
The following example is from a prelude rather than a fugue, but the
principle is exactly the same.
Ex. 9-9a Well-Tempered Clavier I, Prelude No. 19 (mm. 1-6)
Ex. 9-10
Ex. 9-11
Variation I—canon at the octave between the two upper voices, with the
chorale melody in the bass (pedal voice) in long notes
Variation II—canon at the fifth between the upper voices; bass as in
Variation I
Variation III—canon at the seventh between the lower voices; upper
voice has a free obbligato
Variation IV—in four voices; canon by augmentation at the octave
between tenor and soprano voices; alto is free, though motivic; bass as in
Variations I and II
Variation V—“L’altra sorte del Canone al rovescio”; a series of canons at
a variety of intervals, between varying pairs of voices; the voices not
involved in the canon at any particular time have supporting material;
this movement is Freistimmig, and includes stretto treatment of the
phrases of the chorale melody
Play the upper voices as a pair, and note the intervallic or rhythmic
weaknesses, if any. Then add the bass and observe how it fills in the harmony,
provides complementary rhythmic activity, and, in mm. 6-8, drives toward the
cadence.
A canon at the third (or sixth or octave) normally involves tonic harmony
(or vi) at the point where the follower enters. Therefore the leader often
implies dominant harmony immediately before, as is the case in example 9-
12. While the canonic voices are quite satisfying as a pair, the addition of the
bass improves the effect markedly. For instance:
It can be seen from this brief example that the presence of a free, supportive
bass allows intervallic and rhythmic effects that in an unaccompanied canon
would be weak. The bass in this example shares informally in the motivic
material of the upper voices, resulting in a texture of three more or less equal
voices. The bass in other accompanied canons may be less integral with the
canon than in example 9-12, moving either in faster or slower note values
than the canonic voices.
Fugal episodes often contain briefer examples of upper-voice canons with
noncanonic bass. example 9-13 is a three-voice episode from a four-voice
fugue.
Play the canonic voices, then add the bass and discuss its effect on the
musical result. Also analyze the sequential aspect of the music and the
underlying harmonic and linear structures. Prepare a reduction of the lines,
noting the descending stepwise structure in the upper voices, clarified by the
sequence and typically associated with the circle-of-fifths pattern. Note also
the large-scale 7-10, 7-10 LIPs in mm. 46-52.
This is a canon at the fifth below, at two measures (eight beats), with
noncanonic bass. The bass has the function of filling in and clarifying the
harmony, providing rhythmic impetus and reinforcing the motivic content.
This passage is also sequentially organized; there is a long sequence unit of
four measures, starting in the top voice in m. 45, beat 3, and heard again
starting in m. 49, beat 3, transposed down by step. Since the middle voice is
canonic, it is automatically also sequential. The harmony is organized around
the circle of fifths, with chord roots clarified by the long notes in the bass (G-
C-F-B♭), driving toward the upcoming return of the tonic (E♭). This passage
is highly typical of one type of fugal episode, being
developmental
modulatory
sequential as concerns lines, harmony and LIPs
canonic
constructed on the circle of fifths
In mm. 22-24 we find a canon between the upper voices (the first two notes
of the leader, b1 and c2, have been left off but are supplied by the middle
voice from the end of m. 21 into the beginning of m. 22). The canon is at the
fifth below at two beats, with nonimitative bass. It is also a sequence, with a
four-beat unit, transposed down by step, built on the circle of fifths (made
explicit in the circled chord roots in the bass, C-F-B♭-E♭-A♭). Since this is
sequential, there is automatically an associated LIP, in this case a series of
alternating 8-10 and 10-8 patterns between the bass and the upper voices.
From these last two examples we can readily see that a canon at the fifth
below (or fourth above) is ideally suited to circle-of-fifths sequences.
Exercises
1. Perform and analyze in detail some of the canons from the
Goldberg Variations, as found in the anthology, pp. 339ff. Time
permitting, it would be possible at this point to take up chapter 13,
in which the Goldberg Variations are discussed.
2. Continue as strict canons several of the exercises from chapter 6,
and add a noncanonic bass voice, filling in the harmony and
rhythm and sharing in the motivic content.
3. To one of the canons composed in chapter 6, exercise 7 (p. 122),
add a free bass voice.
4. Write upper-voice canons over the basses given below. Keep the
imitation going as long as possible, breaking off just before the
cadence. Try every possible pitch interval of imitation (unison to
octave), using the time intervals of one and two measures only.
You may attempt a canon by contrary motion over either or both of
these basses, modifying the bass slightly if necessary. It is quite
permissible to modify the follower slightly by the use of
accidentals, in the interests of harmonic variety and function.
Exercises
Continue the canonic openings below in three voices. You may attempt
to imitate the shorter ones, especially Nos. 1 and 4, by contrary motion;
No. 2 is specifically to be imitated in this way, as indicated. Continue
these for at least eight more measures, breaking off the canon smoothly
just before a cadence. The symbol ↓ indicates the entrance of follower 1.
Analyze the imitation and the harmony, including nonharmonic tones.
Notes
1. It may be useful at this time to review pp. 77 ff. and pp. 283 ff.
3. It may be useful now for the class to analyze the aria from the Goldberg Variations, to be found in
the anthology (p. 336), while focusing on the bass line and harmony.
5. There are inconsistencies in the ordering of the canons between the manuscript and the earliest
printed versions.
Chapter 10
Fugue I
The composition of fugues in three and four voices is the goal of most studies
in tonal counterpoint. In the fugue one applies all the techniques one has
learned to a complete and coherent musical entity employing various devices
of exposition and development. In composing fugues, we will need to learn
very little beyond what we already know of contrapuntal techniques, and will
concentrate on building larger structures with these techniques. Composing
fugues is an excellent way to learn the essentials of musical organization. All
the fundamentals of musical structure are here: statement (exposition),
departure (contrast, development, manipulation), and return; large-scale tonal
organization; balance, proportion, and shape; and the musical expression of
such basic aesthetic dualities as unity/variety, continuity/articulation,
departure/arrival, and tension/release. Fugue is a confirmation that in music it
is process, not “form,” that matters most.
The fugues of Bach are indisputably the greatest body of fugal writing. For
him, writing fugues is a natural way of composing. We never sense that in his
work the fugal process is a hindrance or limitation; it is simply an appropriate
means of musical expression. His fugues show a tremendous variety of
character and mood. They can seem solemn, jolly, introspective, or dramatic;
in fact, the whole range of musical expression is contained in them. They also
exhibit a great variety of lengths, textures, processes and shapes, as we will
discover.
Directed Study
Perform and listen to the three-voice fugues from the anthology and others by
Bach which you may already know. Be aware of the extent to which these are
monothematic works, arising out of and permeated by the thematic material
from the exposition (the opening section). Note, too, that each fugue has its
own character and mood based on the nature of the subject (the theme). Be
aware of the seamlessness and flow of the music, and its sense of overall
shape. It would be useful at this time to investigate the history, development,
and literature of fugue.1
The Subject
The main theme of a fugue is called the subject. It is analogous to the theme
of an invention. Following are several fugue subjects from Bach. Play or sing
these in class, and discuss them as suggested below.
Ex. 10-1 Well-Tempered Clavier I, Fugue No. 1
Directed Study
Fugue subjects, like invention themes, come in a variety of lengths and types,
but all possess some of the same characteristics. Consider each subject in
these terms:
1. Is the tonic key clearly emphasized at the opening? On what scale degree
does it begin and end? Does it appear to modulate?
2. How long is it?
3. Is there any feeling of cadence at the end? What type of cadence?
4. Are there one or two (or more) distinctive rhythmic or melodic ideas?
5. Is the implicit harmony clear and functional? Is the harmonic rhythm
fairly steady?
6. Does it have an overall shape? Can you distinguish a clear structural-
pitch line in terms of specific scale degrees?
7. Does it appear to be in two sections?
8. Does it contain sequences?
9. Where is the first note placed metrically?
10. Try each subject in inversion, augmentation, and diminution. Which
sound good in which versions?
Discussion
A good fugue subject will exhibit the features of a good invention theme, just
as a fugue is in many ways no more than an elaborate invention. There is, in
fact, no clear distinction to be made between an invention theme and a subject
(which we will call S) other than to say that most subjects are longer; are
more complex in rhythmic, melodic, and harmonic structure; may modulate;
may be sequential; and may be in two distinct sections. Subjects vary greatly
in length, from one to eight measures or even more (in some organ fugues).
Most are two to four measures, and we will concentrate on these.
The following features characterize most fugue subjects:
1. The tonic key and chord are clearly established at or near the beginning.
Important tonic triad notes are emphasized; weak scale degrees,
especially the leading tone, are not. Most subjects begin on scale degree
one or five, and end on one, three, or five. Some begin on the leading
tone, but only as a short anacrusis. Each comes to a cadential point at the
end, either an IAC or PAC implication, or very occasionally an HC. The
harmony is usually diatonic, except with a chromatic S (see H, K, and L
in example 10-1), when secondary dominants may be implied. The
harmonic rhythm is quite regular, and the harmonic progression is
functional and strong.
2. The cadence is placed, as always, on a strong beat.
3. There will be one or two, or even three, distinctive features that will
enable the subject to be heard clearly in complex passages, and that will
suggest to the experienced composer specific manipulations and
contrapuntal combinations. Some subjects are specifically designed to
work in stretto, inversion, augmentation, or dimunition.
4. As with any good line, the shape will be clear, and there may well be a
clear structural-pitch line (see especially A, F, G, I, and K in example 10-
1). The structural-pitch line often falls from 5̂ to 3̂ (as in 5̂ 4̂ 3̂ or 5̂ 6̂ 4̂ 3̂).
Many subjects in minor feature a diminished-seventh outline or leap,
from 7 (the leading tone) up to 6̂, or 6̂ down to 7̂.
5. Many longer subjects are in two distinct sections, a “head” and a “tail”
(see E, FI, and perhaps G and J in example 10-1). These may modulate
to the dominant, and often contain several distinct motivic ideas.
6. The range will normally be kept within an octave, with the usual tonal
framework formed by tonic triad notes. The more voices, the narrower
the range of the subject, avoiding crossings and other complications.
7. Subjects may begin on a strong beat, or after a brief rest. In the latter
case, the meter may not be fully clear for several beats, or even until the
entrance of the second voice.
8. A good subject will have a strong sense of character and individuality. In
Bach each fugue is sui generis, a thing in itself, in terms of subject
material, overall structure, and expression (affect).
Exercises
1. Perform and analyze additional fugue subjects from the anthology,
as suggested by the instructor, focusing on the subject and its
recurrences through the fugue.
2. Critique the following subjects, in stylistic and technical terms.
Discussion
In example 10-2, number one, the answer literally transposes the subject up
a fifth. The notes of the A are all notes of the scale of the dominant key. This
literally transposed response is called a real answer. Observe that each scale
degree of the S is answered by the corresponding scale degree (in the
dominant key) in the A.
Ex. 10-3
Ex. 10-4
Briefly put, in a tonal A a strong dominant pitch at or near the head of the
S is answered by the tonic pitch in the tonic key at the corresponding point in
the A and, because the A is at the fifth, tonic notes are automatically answered
by dominant (again, thinking in the original key).
After the tonal adjustment has been made, the A continues as if it had been
real. The readjustment to a real A is made as soon as possible after the tonally
adjusted note or notes.
Ex. 10-5
The reasons for the tonal A are to be found in the nature of tonality itself,
that is, the necessity for keeping within the tonal framework to preserve the
integrity of the tonic key and to avoid modulating endlessly around the circle
of fifths. When making the tonal adjustment, it is crucial to think in terms of
scale degrees in the tonic key.
Ex. 10-6
A tonal A will fit harmonically with the end of the S when it enters in the
exposition, and will not require a modulatory link after the S (as will be
discussed later). That is, the tonal answer enters with tonic harmony in the
first key.
Ex. 10-7
Ex. 10-8
Ex. 10-9
One consideration in making the tonal A is that it preserves the identity and
integrity of the S as much as possible and create no awkwardness of line or
harmony. Strong scale degrees (especially tonic and dominant) in the S
should, if possible, be answered by the corresponding strong degrees in the A.
Tendency tones should also be answered by tendency tones when possible, as
in example 10-10 (except in the situation explained in example 10-9). After
the tonal adjustment, the A should imply the same set of harmonies (in the
dominant key, of course) that the S does.
Ex. 10-10
In some cases Bach obviously felt so strongly about the identity of the S
that he gave a real A to an S that would normally have required a tonal one. A
well-known instance is found in the G-minor organ fugue (example 10-11).
Ex. 10-11
Ex. 10-12
Ex. 10-13
To summarize:
1. The S is answered at the fifth above (or fourth below), in the dominant
key.
2. If the S has a prominent dominant note at or near its beginning, it
normally requires a tonal A, in which that dominant note is answered by
the tonic note; after this adjustment, the A returns as early as possible to
its real A form.
3. A modulating S also requires a tonal A.
It would be possible to spend a great deal more time on what Sir Donald
Francis Tovey called the “vexatious minutiae” of the answer,3 but the above
discussion covers the vast majority of cases one is likely to encounter. For
those wishing a more encyclopedic view, the sources cited on p. 210 are more
than adequate.
Exercises
1. Investigate the answers to the fugues in the anthology, including
those in more than three voices. Write out each S with the A below
it for comparison and discuss any tonal adjustments. Analyze both
harmonically.
2. Write answers to the following subjects. Compare and discuss in
class. Some should be worked out together at the board. Where
variant answers seem feasible, these should be discussed. All
except the last eight are by Bach.
3. Write A’s to the S’s you wrote in exercises 3, 4, and 5, on p. 192.
The Exposition
The exposition has the purpose of setting out all the thematic material for the
fugue.4 Most fugues are monothematic compositions, having as their premise
the drawing of an entire work out of its initial materials. Thus, the clarity and
conviction of the exposition are crucial to the success of the fugue. This will
necessitate not only the invention of a memorable subject but also the
construction of convincing counterpoints in the other voices and of suitable
linking and bridging material.
It would be appropriate to discuss first the layout of a three-voice
exposition. The order of voice entries with S and A is fairly standardized, with
Bach’s two favorite successions being:
The second entry is always the A, on the dominant; the third entry is nearly
always the S (on tonic) and is an octave away from the first entry. Further, in
the exposition each following voice comes in on the same beat of the
measure, or a comparable beat (strong or weak) as did the first voice; in triple
meters, this will mean the same beat. In quadruple meters, if the first voice
enters on beat 1, the second or third may enter on 1 or 3; if the first entry is on
2, subsequent entries will be on 2 or 4. In Bach, beats 1 and 3 appear to have
virtually equal weight and accentuation.
There are other possible orders of entry (such as 3, 2, 1 or 2, 3, 1) but they
are rarely used by Bach. The effect of the bass coming in last in the exposition
seems to have been highly prized. In any case, the first two entries are always
in adjacent voices.
There may be an extra entry at the end of the exposition. This will usually
be the A (but see The Well-Tempered Clavier I, Fugue No. 6, voice 1, mm. 8-
9) and is often found in the voice that began the exposition:
In these cases, the redundant A will usually be an octave from the second
entry, just as the two entrances of the S are an octave apart (see The Well-
Tempered Clavierl, Fugue No. 21, voice 1, mm. 13-17). In few cases does any
voice have two successive entries of the S or A, as this would place too much
thematic weight on this voice (for exceptions, see The Well-Tempered
Clavierl, Fugue No. 8, voice 3, mm. 12-14; The Well-Tempered Clavier I,
Fugue No. 19, voice 3, mm. 6-7).
With very short subjects, there will often be an effect of stretto in the
exposition (see The Well-Tempered Clavierl, Fugue No. 22; The Well-
Tempered Clavier 11, Fugue No. 3).
Before going into the other details of the exposition, let us look at an entire
exposition.
Directed Study
Play each voice through and note that it is a living, organic line, not just a
succession of discrete events (subject, link, countersubject, and so on). Then
play the three voices together.
The exposition in example 10-15 can be graphed as follows:
The Link
One of the reasons for the tonal A is that it may allow the A to begin with
tonic harmony. This happens in the two expositions above. Fugue No. 1 from
The Well-Tempered Clavier I has a real answer that allows the same treatment.
Note here that the first clear sense of dominant key is not reached until m. 3,
with the f♯1. In such cases, the brief modulatory link following the S and
eliding into the CS is not needed.7 These links come out of the end of the S in
a natural, almost imperceptible way, so that the ear may be unsure when the S
has ended and when the link gives way to the CS. In fact, what initially
appears to be a link may later be revealed to be part of the CS. Such
smoothness of connection and flexibility of thematic function are important
characteristics of a good fugue. A short link is sometimes used even if it is not
needed to effect a modulation to the dominant, simply to get the line to the
note on which the CS starts (see example 10-15, voice 2, mm. 4-5). The link,
if needed, must be constructed with care so as not to sound artificial or
awkward. Links can be found in many fugues, for instance The Well-
Tempered Clavier I, Fugues Nos. 3 and 7; The Well-Tempered Clavierll,
Fugues Nos. 12 and 15. Subjects beginning on scale degrees 1 or 5 and
ending on 3 often require a link, as will a modulatory S (example 10-18).
Ex. 10-18 Well-Tempered Clavier I, Fugue No. 7
The Countersubject
Directed Study
Study the CS1 in The Well-Tempered Clavier I, Fugue No. 2 (example 10-16).
Play it by itself, then with the A, mm. 3-5. In what ways does this relationship
exhibit the features of good two-voice counterpoint? Think about all aspects:
shape, relative motion, motif, harmonic intervals, rhythm, harmony. Next,
investigate the CS-to-A relationship as found in the three- and four-voice
fugues in the anthology, as directed by the instructor.
Discussion
The CS-to-S (CS-to-A) relationship must exhibit the features of good two-
voice counterpoint. The majority of the three-voice fugues in The Well-
Tempered Clavier have at least one quite consistent CS. The four-voice fugues
less typically use consistent countermaterial. Notice in both expositions above
(examples 10-15 and 10-16) that the CS is a well-shaped, motivically
consistent line which complements the rhythm and shape of the A and is in
double counterpoint with it (the intervals are all imperfect consonances or
properly treated dissonances). A good CS has these features:
The CS will have to be slightly adjusted to fit with both the S and the A if
the A has been tonally adjusted, as shown in example 10-19.
Directed Study
In the exposition in example 10-20, play the CS first by itself, then together
with the A (mm. 3-4); note how successfully their rhythms “dovetail.” Then
play mm. 6-7, first playing CS2 by itself, then combined with the other two
voices. Note its simplicity and the fact that it does not complicate the
harmony or rhythm but fills out and clarifies, while providing a complement.
The Bridge
Directed Study
Analyze the bridges of the three fugue expositions above (examples 10-15,
10-16, and 10-20). What materials do they use? How long are they? Are they
sequential? What LIPs do you find? Are there clear structural-pitch patterns?
What appears to be their harmonic and thematic purpose? Play each
exposition through, first one voice at a time, then together.
Discussion
to provide an efficient modulation back to the tonic key, so that the third
voice can enter with the S
to provide an additional measure or so before the third entry to avoid
excessive predictability and regularity in the entries
to provide preliminary confirmation and development of motifs, often by
fragmentation and sequence
If the S ends on V or modulates to the dominant, the A will then end on I,
so that there will be no necessity for a bridge remodulating to tonic; yet Bach
often provides one anyway as a brief episodic relief before the third entry.
One probable reason why Bach needed a bridge in the E1» fugue (example
10-20) is that if he had brought the third entry in as early as possible (m. 4,
beat 3), a 6 4 chord would have resulted.
A brief discussion of the three bridges will be helpful.
In the fugue in F major (example 10-15) a two-measure bridge (half as long
as the S) is provided, written in double counterpoint, modulating from C
major back to F. The upcoming key is introduced as early as possible (m. 8,
voice 2, B♭). The material in both voices is taken from both the S (m. 1) and
the CS or link (mm. 4-5, voice 2).
The fugue in C minor (example 10-16) has a somewhat more elaborate
bridge, as long as the S (two measures). It is sequential, in two-beat units
transposed up by step. The material is taken from the S (by fragmentation)
and CS (by fragmentation and inversion). The structural pitches in voice 1
(E♭2, F2, g2) are related to the main structural pitches of the S (g1, f1, E♭1).
Note also the structural parallel tenths and sixths between voices.
Ex. 10-21
The structural pitches in this fugue are easy to hear, and function both on a
surface thematic level and on deeper structural levels (which we will
investigate later).
In the fugue in E♭ (example 10-20) we find a six-beat bridge (as long as the
S), organized sequentially. As E♭ major has already been reached in m. 3, this
codetta does not have the function of returning to tonic, but is used to provide
relief and to allow the third voice to enter on beat one. Voice 2 develops
sequentially the material of the link. Note again the strong structural lines,
and observe how voice 2 in mm. 4-5 drives downward, setting up the entrance
of the S (B♭) in m. 6.
Ex. 10-22
Exercises
1. Perform and analyze several more fugue expositions from the
anthology, or from other Bach fugues you know, focusing on the
link (if any), the CS (or free counterpoints to the S and A), and the
bridge. Note that there may or may not be consistent CS material,
and that the CS may be slightly altered without losing its identity.
2. To selected subjects below add a link (if needed), A, and one CS in
double counterpoint in the other upper voice. Do not add a bridge
or the third voice yet.
You should compose at least four expositions, two major and two
minor. At least one should be on a subject of your own
composition, as approved by the instructor.
Notes
1. In The New Grove Dictionary of Music and Musicians, “Fugue” (vol. 7), an excellent introduction
to the history and development of fugue. The books by Horsley, Mann, and Oldroyd, listed in the
bibliography, will be of particular interest.
2. Theoretical and historical questions raised by the fugal answer are beyond the modest scope of this
text to treat exhaustively. Among the several books dealing in detail with these questions are those
by Horsley, Mann, and Naldin, (see bibliography). See also The New Grove Dictionary of Music and
Muscicians, “Answer” (vol. 1).
4. This is not true for the double fugue, which will be discussed later.
6. An alternative view might understand CS1 as beginning on beat 3 of m. 3, with beats 1 and 2 as a
link.
8. This set of relationships is clearly defined in such fugues as The Well-Tempered Clavier I, Fugue
No.2, and the fugue from the Passacaglia and Fugue in C Minor.
Chapter 11
Fugue II
Overall Structure
The overall scheme of a Bach fugue, after the exposition, is variable. As with
the invention, the exposition is fairly standardized, but the rest of the fugue,
being largely developmental, may exhibit any number of different layouts. All
we can accurately say about the plan of a Bach fugue is that there will be an
exposition, followed by episodes and/or middle entries and/or strettos and/or
other manipulations of the main thematic material, modulating through two or
more closely related keys, and returning to the tonic key. Still, it is useful to
generalize about the order of events in most fugues, with particular emphasis
on the fugues in The Well-Tempered Clavier.
The overall structure of most fugues is clarified by the placement of strong
internal cadences. Many shorter fugues have one clear internal cadence,
placed roughly midway through the fugue, dividing it into two balanced
sections, similar to a two-section invention. Other fugues have three fairly
clear sections, with two strong internal cadences, comparable to the three-
section invention. Tonal events may not always coincide with thematic events,
unlike most homophonic forms, in which tonality and theme are indivisible.
Since most fugues are monothematic, thematic contrast cannot be used to
clarify form; thus, the difficulty of deciding on the “form” of a fugue. Another
problem for the analyst and performer is that internal cadences in the Bach
style are often obscured or covered by motion, so that their potential for
defining form may be weakened.
Even with these issues of definition, fugues are somewhat sectional. The
sections may be defined by:
cadences
textural and/or registral changes
modulations
clear entrances of the subject or answer
the distinctive use of such devices as stretto, inversion, and pedal point
These outlines are for the general guidance of the student and should not be
taken rigidly as implying that a fugue must exhibit one of these patterns of
events. Further, a good fugue is almost a seamless musical entity, and the
sections will not be as distinct as the charts above may suggest. The key
schemes given are also to be understood generally, as any closely related key
is available to Bach, at any time. There are even some fugues that never
clearly modulate. The cadences are also variable, but there will normally be at
least one strong internal cadence in a fugue (usually in the relative or
dominant key), as a point of textural and harmonic arrival or relaxation, as
well as tonal and formal clarification.
It is best that the first few fugues one composes be based on specific
procedural/schematic models drawn from Bach. This modeling procedure will
be set out in the exercises concluding this chapter.
The Episode
Directed Study
examples 11-1, 11-3, and 11-4 give three episodes, with middle entries, from
fugues by Bach. Play and discuss them as suggested here.
1. Compare these episodes to the expositions from which they are drawn on
pp. 204, 203, and 207. From what principal thematic material—subject
(S) or countersubject (CS)—are the motifs taken? By what processes are
the motifs altered? By what processes are they developed? Is there any
new motivic material?
2. Discuss the overlappings between episode and middle entry. Is it always
clear exactly where a sequence starts? Where it ends? Where an episode
starts and ends? Do episodes and middle entries (see pp. 216 ff for
discussion) ever overlap?
3. Analyze these passages harmonically and in terms of structural pitches,
especially in the bass voice. What patterns of linear and harmonic
organization do you find? Any LIPs? Sequences?
Ex. 11-1 Well-Tempered Clavier I, Fugue No. 2
Ex. 11-2
Play the reduction in example 11-2 and then play again the episode based
on this strong and simple model. Note the descending scalar shapes outlining
perfect fifths and the tonic (E♭) octave, and the descending filled-in third (E♭,
F, G) an important thematic and structural line in this fugue. This underlying
simplicity of middleground shape is essential as an underpinning to the
foreground intricacy of works like this. Lacking such a simple structure, the
music would not be likely to cohere in a convincing, shapely way. In writing
your own episodes, be aware of the necessity for such directional pitch
structures, controlled by sequential patterns of harmony, line and LIPs. Note
the “nested” LIPs in example 11-2: the outer voices form a 10-8 pattern, the
middle and bass voices form an 8-10 pattern (in alternation with the outer
voice pattern), and the upper two voices a 3-6 pattern.
Next, play and study example 11-3, with particular attention to all aspects
of patterning.
Ex. 11-3 Well-Tempered Clavier I, Fugue No. 11
Ex. 11-5B
Ex. 11-6
Discussion
Episodes usually operate sequentially on highly directed structural-pitch
frameworks, in a way that expositions and middle entries may not. These
inner lines, as we have observed before, often descend by step through a
perfect fifth or octave, setting up the upcoming tonic, while the harmony
typically moves fully or partly around the circle of fifths. All the prominent
motivic material derives from the S and/or CS, most often by processes of
fragmentation and/or inversion, and may be treated imitatively, either between
the upper voices (examples 11-1 and 11-3) or between any pair of voices
(example 11-4). The upper voice canon with supporting bass is an effective
texture. Three-voice canon is difficult to sustain and is rare as an episodic
device. Nonrigorous imitation (example 11-3, mm. 30-34, upper voices), in
which two voices share the same material but not in strict canon, is common.
A well-constructed episode overlaps with the end of the preceding and the
beginning of the following music. And the first note or two of the first
sequence unit may be omitted, due to this overlapping. These interlockings
are very important to an effective episode. One voice may break out of the
end of a sequence only to begin another immediately, as we saw in example
11-4, mm. 9-10, voice 3 (lowest voice). This latter passage is particularly
characteristic: the ending two-beat figure of a four-beat unit is used as the
basis of the following two-beat unit. This halving of the length of the
subsequent sequence unit drives the music forward effectively toward the
cadence. The opposite, a doubling of the length of the unit, is almost never
found.
Episodes vary in length according to the length of the fugue and of the S.
Four measures is a common length in the three-voice fugues. The longer
episodes, when examined, turn out to be two episodes joined smoothly in the
middle, or containing one middle entry, as in example 11-4, which can be
graphed as follows:
Ep. 1 ME Ep. 2
7–11 11–12 12–17
Further, episode 2 is broken into two sequential passages, mm. 12-14 and 15-
17.
Episode 1 is usually simpler in its texture and devices, and shorter, than
later episodes, though it may use stretto if the S is so designed. It may end
with a PAC (as in an invention) or, more often, it may elide into the first ME.
It is usually better on aesthetic grounds to avoid the cadence here. This
episode will normally be followed by the first ME unless there is to be a
counterexposition (which is discussed on pp. 219 ff) Later episodes tend to be
more complex in texture and technique, often including the use of two-voice
canon or three-voice stretto. If there are three episodes, Episode 3 may be the
contrapuntal inversion of Episode 1, assuming the former had been written in
triple counterpoint. It should be pointed out here that a few fugues lack any
episodes, specifically those with an S designed to be treated in stretto, the so-
called stretto fugue (for instance, The Well-Tempered Clavier I, Fugue No. 1).
Some stretto fugues, though, do contain episodes, as may be seen in The Well-
Tempered Clavier I, Fugue No. 3-
Middle Entries
Any entry of S or A after the exposition is a middle entry (ME), including, by
some definitions, those that are a part of the final section of the fugue,
following the return of the tonic key. To be a true ME it must contain at least
one complete version of the S or A, in any closely related key. The presence
of the complete S distinguishes an ME from an episode, which is based on
thematic fragments. The first ME, which usually follows Episode 1, includes
one or two entries of the S (or A). This first ME may be in any closely related
key, though it is most often in the dominant or relative key.
The later MEs may be somewhat longer, with two or three entries of S or
A, and are in a key different from ME1. When the fugue is in a major key,
these entries will often be in subdominant, relative, or supertonic key; when
minor, in subdominant, submediant, or relative key. These key schemes are by
no means rigid. The later MEs often contain stretto, if the S is so designed. If
the exposition is in triple counterpoint, the MEs may, in total or in part, be
voice-exchanged versions of it, though it is unlikely that any ME will be as
long as the exposition. The Well-Tempered Clavier I, Fugue No. 2, uses such a
scheme.
Demonstration of Episodes
Ex. 11-7
Exercises
1. Perform and study several more fugue episodes and middle entries,
as selected from the anthology. It is important to analyze a number
of episodes before attempting to write your own. Note all aspects
of linear and harmonic patterning.
2. To the expositions composed in exercise 5, chapter 10 (p. 209), add
episodes and one or two middle entries. Be careful to let the first
episode flow smoothly out of the end of the exposition and into the
first ME. Attempt to use overlap to obscure the “joints” between
these sections. Play each line as you write it, being attentive to its
overall integrity, continuity, and shape. Analyze fully, including
motivic content, harmony, structural pitches, sequences and LIPs.
Reference back to the layout graphs on p. 212 may be helpful at
this point.
The Counterexposition
Some fugues contain a counterexposition (CE). This is a section, usually
following Episode 1, that presents the S and A in the original keys (tonic and
dominant), normally with the same counterpoints as in the first exposition.
The main difference is that the order of voice entries is not the same as in the
first. Following are some fugues with CEs:
1. The Well-Tempered Clavier I, Fugue No. 11, mm. 17-29. In the first
exposition the order of entries is 2-1-3; in the CE it is 1-2-3. See example
11-8.
2. The Well-Tempered Clavier I, Fugue No. 1, mm. 7-10 (or perhaps m. 12).
As this is a stretto-fugue, the CE is by stretto and sounds episodic.
3. The Well-Tempered Clavier II, Fugue No. 9, mm. 9-12. This is again a
CE by stretto. Whether such sections, because of their developmental
nature, should be thought of as CEs or episodes is an open question. In
one view, the presence of the complete (not fragmentary) S would
identify this as a CE.
4. The Well-Tempered Clavier 11, Fugue No. 17, mm. 13-24. In the first
exposition, the order of entries is 2—1—3—4; in the CE it is 4-2-3-1.
Ex. 11-9
This is in the stile antico (old manner), a restrained, somewhat vocal style
dating back to the Renaissance ricercar,1 historically one of the forerunners of
the fugue. It is used in diminution toward the end of the fugue, in stretto,
achieving a climactic effect through increased rhythmic activity and drive
(probably the principle reason for using diminution).
Ex. 11-10 Well-Tempered Clavier II, Fugue No. 9 (four voices)
Comments on Stretto
Subjects have to be specifically designed to be effective in stretto. A subject
you intend to use in stretto should, before you begin any other work on the
fugue, be imitated at a variety of close time intervals to test its suitability for
stretto.
Following is a demonstration of stretto, using the subject from p. 218.
Ex. 11-16
Exercises
1. Perform and analyze any of the strettos from the anthology that
have not yet been analyzed. Note the time and pitch intervals used,
the length of the S at each entrance, and any adjustments to the S.
What is the function of and material in the non-stretto voices?
2. Try the shorter subjects on pp. 199 ff and 209 in stretto, as
suggested in the demonstration on p. 225. Remember that slight
adjustments are permitted as long as the head is intact, that time
and pitch intervals may vary, and that S and A may be used
interchangeably. Try every possible pitch interval, up to the octave,
at one, two, three, and four beats. Attempt both two-voice stretto
with a nonimitative third voice, and three-voice stretto. Try stretto
by contrary motion.
3. At what other imitative intervals, in stretto, can you make the
demonstration subject (p. 225) work? Try both three-voice stretto
and two-voice stretto with a free third voice. Keep these going for
four to eight measures. Slight adjustments to the S and A are
permissible, as is stretto imitation by contrary motion.
The ending shown in example 11-19 is a coda based on tonic pedal with
final stretto entries of the S over it. Note the E♮ at the end of m. 29, not a
Picardy third but implying a secondary dominant of iv. Such a move toward
predominant harmony is characteristic, nearing the ends of works of most
tonal composers. The tonic pedal is arrived at—in this case, as above—
through a PAC. Bach must have felt that an ending at this point (m. 24) would
have been abrupt, and a coda was needed to stabilize the tonic harmony
through prolongation. In fugues in major keys, the introduction of lowered-7̂
(the subtonic note) pushes the harmony toward subdominant, nearing the end;
this is the function of the B♭’s in mm. 24-26. In fugues in minor keys, the
analogous note is raised-3̂, as in example 11-18, m. 29.
Ex. 11-19 Well-Tempered Clavier I, Fugue No. 1
The final section of the fugue in example 11-20 also contains a final
version of the S, with a textural climax and a brief dominant pedal, which
serves simply to slow the harmonic rhythm at the cadence. The bass line
presents a very strong cadential formula in the last three measures (1̂, 4̂, 5̂,
and 1̂), giving a solid harmonic underpinning to the end.
Ex. 11-20 Well-Tempered Clavier I, Fugue No. 12
You will have noticed the harmony heard over the tonic pedals in examples
11-18 and 11-19. In each case it progresses from tonic to a secondary
dominant of the subdominant, to subdominant, to dominant, to tonic—in
effect, a prolongation of the strongest authentic cadence harmonic formula in
this style, with a broadening of the harmonic rhythm that leads strongly to the
ending. This is a typical harmonic gesture at the end of a fugue. The
movement toward IV (iv in minor) and, in general, the subdominant side of
the circle of fifths characterizes the endings of many tonal works, just as
movement toward the dominant side tends to typify the opening section.
Observe also how the raised mediant degree in the V7/iv prepares the ear for
the Picardy third at the end.
One final example of pedal point is given below.
Ex. 11-22
It is typical that the S does not occur in the voice that began the fugue
(voice 2), but in another voice (voice 1), for variety. This fugue is written in
triple counterpoint throughout, and the S order of voices used here, CS 1, has
not yet been heard in the fugue, so that this is not a literal CS 2 recapitulation
of the exposition. Nor will one often find a repetition of the entire exposition
at this point, as this would seem redundant. Many fugues return to tonic
without an unambiguous return to the S, and thus lack a recapitulation.
The final section will probably contain a thickening of the texture through
increased rhythmic activity, and an intensification of the harmony. It may well
involve two- or three-voice stretto, especially in fugues rich in stretto (see The
Well-Tempered Clavier I, Fugue No. 16, mm. 28-30; The Well-Tempered
Clavier II, Fugue No. 5, mm. 43-end).
Pedal points, as discussed above, may well be employed. Tonic pedal at the
end is common; dominant pedal preceding it is somewhat less so. If the tonic
pedal is prolonged for four measures or so, it will have the effect of a coda,
especially if introduced by a strong cadence (PAC or DC) and if the S is heard
over it. A final stretto over tonic pedal is also a possibility. It is very important
that this coda be fully integrated with the rest of the fugue, in terms of motivic
material, texture, and harmonic vocabulary; it may otherwise feel “tacked
on,” and thus ineffective.
There may be a final very close stretto, perhaps with an improvisatory,
toccata-like flourish just before the final cadence, as in example 11-23.
Here again, the closing section is set off by a PAC (m. 23) followed by a
sequential, imitative passage based on the head of the S, and another
homophonic sequence built on its tail (mm. 25-26). These dramatic,
homophonic codas are not particularly common in Bach (they are much more
so in the work of George Frideric Handel).
There may be a dramatic pause, using a fermata chord or rest, which sets
off the closing section, as in example 11-25.
Ex. 11-28
Exercises
1. Transform selected subjects from p. 209 by the processes
suggested below, then treat the original form and the varied forms
as the material for:
Ex. 11-30
Exercises
1. Analyze several more three-voice fugues from the anthology.
Identify all the thematic material, episodes, middle entries,
sequences, strettos, and all manipulations of the thematic material.
Do a formal graph as on p. 212, and a structural-pitch graph. Show
all modulations, keys, and cadences, and the harmonic organization
of all episodes. Show structural harmony, and all aspects of
patterning, including bass-line root patterns, sequences, and LIPs.
2. Then select one fugue as your model, and prepare a detailed bar-
by-bar graph of it. A graph of mm. 1-12 of The Well-Tempered
Clavier I, Fugue No. 2, is shown as a model.
2. It is possible in stretto to cut the S off after the first few notes, as long as this truncation is obscured
by stretto entries in the other voices; we could call this pseudostretto. See also The New Grove
Dictionary of Music and Musicians, “Stretto,” vol. 18, and the books by Horsley, Mann, and
Oldroyd cited in the bibliography.
3. To characterize pedal point as a type of nonharmonic tone, as is often done, may be misleading. Its
effect is rather to sustain (prolong) a single structural harmonic root under foreground changes in
harmony in the upper voices.
4. Some theorists avoid the term recapitulation as conflicting with its more common application in
later music, preferring the term return.
5. It would be a very useful exercise for the class to make its own reductions of this and several other
Bach fugues, either individually or as a group. The linear/harmonic reductions in the books by
Gauldin and Parks (see bibliography) are also worthy of attention.
6. Most classes will not have time to do all of these exercises, but the general order of analysis and
focused exercises followed by composition should be observed.
Chapter 12
Four-Voice Counterpoint
Perform the fugue expositions given below, and analyze the four-voice
writing in them, focusing not on the specifically fugal aspects but on the more
general features of counterpoint, as suggested on pp. 247 ff.
Are there homorhythmic passages in all voices? For how long are such
effects sustained? Is the pulse always heard in at least one voice? Look at the
use of rests, longer values, and ties. Compared to three-voice works, are they
more or less commonly used here?
Perform these expositions in class, conducting and intoning the rhythm of
each voice (but not the pitches) on a neutral syllable such as ta, to appreciate
both rhythmic independence and complementarity.
HARMONY AND COUNTERPOINT. Analyze the harmony, including
chord types (qualities), inversion, completeness, and doubling. Generalize
about what you find. Circle the roman numerals of the structural chords.
Analyze the nonharmonic tones. Are there any new idioms? What
simultaneous nonharmonic tone idioms do you find? What successive
nonharmonic tones? What do the nonsuspending voices do to accompany
suspensions?
Play each pair of voices. There will be six pairs (voices 1-2, 1-3, 1-4; 2-3, 2
—4; 3-4). Do you find any parallel perfect consonances? Direct fifths or
octaves? Unequal fifths? How many imperfect consonances are on average
found in succession?
Texture and Rhythm
Counterpoint in four voices is no different in technique from that in three, nor
appreciably more difficult to write. The primary distinctions concern texture
and rhythm.
You may have noticed that, due to the rests and longer note values, several
brief passages above are, in effect, in three-voice texture. A glance through
the four-voice fugues of Bach, especially those for keyboard instruments, will
reveal a good deal of three- and even two-voice texture, in particular in the
episodes. The fullest texture will occur at the end of the exposition, in the
later episodes, in stretto passages, and at the end. In most four-voice
counterpoint, the voices are treated as equals. Generally speaking, they are
equally active, and there is rarely a feeling of upper-voice domination, inner-
voice “filler,” or bass lines that are merely harmonic support. The use of short
rests, ties, and longer note values in one or two voices at a time accomplishes
several ends: it avoids textural thickness and rhythmic squareness; it allows
the voices to phrase individually; it avoids contrapuntal problems; and it gives
the voices a feeling of rhythmic independence. Such passages lighten or
“ventilate” the texture, clarify the counterpoint, and make the fuller passages
more impressive by contrast. There will be few, if any, moments during which
all the voices are moving simultaneously, as this effect quickly becomes
homophonic. Even such a passage as in example 12-3, mm. 6ff., in which all
the voices primarily move in eighth notes, is still sufficiently varied by the
use of longer notes and rests. Such passages as the following are particularly
fine from the point of view of rhythmic variety and complementarity.
Ex.12-6 (ex. 12-4, mm. 5-7)
Note in these rhythmic reductions that no two voices are allowed to move
for more than two or three beats in the same values, that the various values
are shared equally among the voices, and that the shorter values often follow
a tie or short rest. At any given moment, any voice may be inactive due to
rests, ties, or long notes; but such passages are typically brief, unless the voice
involved is allowed to rest for an entire episode to lighten the texture. In such
cases, the resting voice must have come to a natural point of rest (on a stable
note) and will almost always return with important thematic material.
There is a certain amount of voice pairing in four-voice texture. It is
important that these voices vary, in terms of the specific pairs involved. It
would be a mistake, for instance, to pair voices 1 and 2 throughout. example
12-3 is especially instructive in this regard; notice how the eighth notes
involve constantly shifting voice pairings. Such pairings, when they do occur,
are relatively brief and usually occur in the fastest values. Any parallel motion
is of course limited to brief passages of imperfect consonances.
A texture in which a short motif in faster values is passed freely between
the voices (informal or motivic imitation) is a particularly effective one. See,
for instance, example 12-2, m. 3, or example 12-3, passim.
The spacings in four-voice texture are, of necessity, somewhat closer than
in thinner textures. Bach was limited by the range of his keyboards so that the
thicker the texture, the narrower the range of each voice and the closer the
spacings. Adjacent voices are rarely more than an octave apart,1 though the
relation of voices 3 and 4 may occasionally exceed this. As usual, the wider
spacings will generally sound best at the bottom of the texture—that is,
between tenor and bass. There may be occasional voice crossings, especially
between the upper voices, though these will not continue for many beats. The
crossing of voices 3 and 4 is fairly rare and creates a new bass line against
which chord inversions will then have to be measured.
Students at this stage of writing often fall into the trap of harmonic thinking
and chorale-like textures. To avoid homophonic effects, one should always
play or sing the individual lines when writing them, and never allow oneself
to make a note choice on purely harmonic grounds.
Harmony
The harmonic language is no different from that in three voices, but of course
the harmony will be fuller and the chords more explicit. Complete seventh
chords are now possible, although they often occur with the fifth or (more
rarely) the third omitted and the root doubled. Seventh chords of all the usual
types are somewhat more common in four voices than in three. A brief
overview of doubling practices in four voices may be helpful at this stage.
Most Usual Least Usual
Chord Type
Doubling Doubling
These are general suggestions for doubling; they are not prescriptive, as
integrity of line takes precedence over doubling. A more useful way to think
of doubling may be in terms of notes not normally doubled, which is to say
chord thirds (especially in major triads and dominant seventh chords) and
tendency tones (chord sevenths and ninths, leading tones and accidentals, or
either note of an augmented or diminished interval). But even a tendency tone
may be doubled, providing this doubling is brief and metrically weak.
As in three-voice texture, triads without thirds are extremely rare; complete
major and minor triads in root position and first inversion predominate;
seventh chords are used in all positions, and diminished triads are usually
found in first inversion. The “chord” is of course a dissonant sonority and
must be treated in the usual ways (see chapter 8).
There are no new nonharmonic tone idioms, though care should be
exercised with regard to their use, since multiple nonharmonic tones can
obscure the underlying harmony. The usual dissonances are found, used in the
ways we have studied. Bear in mind that the faster-moving notes tend to be
consonant with each other, even though one or both may dissonate against the
other voices. Simultaneous suspensions, neighbors, or passing tones in two
voices are usable, but such effects quickly become homophonic and are best
not overused. There will be virtually no cases where three voices are
simultaneously dissonant against the fourth voice, except under conditions of
pedal point (for example, V7 or vii°7 over tonic pedal just before a final
cadence, as in The Well-Tempered Clavier I, Fugue No. 2, mm. 30-31).
Successive nonharmonic tones, either in one or two voices, are found, as long
as both are treated correctly in relation to the other voices and the underlying
harmony is not obscured.
As in three voices, parallel fifths and octaves are not used (even if by
contrary motion), and direct fifths and octaves are not found between the
outer voices, though they may occur between other pairs of voices, as may
unequal fifths. In other words, each pair of voices still operates according to
the principles of two-voice counterpoint, with the exceptions just noted. And
the outer voices behave exactly as in two-voice writing. Each pair must be
played and checked for counterpoint.
Keep in mind the usefulness of contrary motion between moving voices
(especially the outer voices) and the fact that brief rests may occasionally be
employed to avoid contrapuntal problems. There will be few conditions under
which three voices will move in parallel motion, as this will often create
problems of independence and unusable parallelisms, and in virtually no case
can all four voices move simultaneously in the same direction. Recall that
nonharmonic tone activity cannot be used to avoid unacceptable parallelisms.
Exercises
1. Analyze several more four-voice passages from the anthology,
from both fugal and nonfugal works. Do not focus on the imitative
aspects at this point, but rather on the other details of the
counterpoint and harmony, including structural lines and structural
harmony, scale-degree formulas (especially in the bass), sequences
and LIPs, as well as rhythm and texture.
2. Locate the errors in the following examples and suggest possible
corrections.
Suggestions:
1. Work three to four beats ahead in one voice at a time, then catch the
other voices up, adjusting as you write.
2. Play or sing each voice as you write it, and again after it is completed, to
check shape, flow, and coherence.
3. Never write a note for purely harmonic reasons; it must make linear
sense.
4. Always play and analyze your work.
Exercises
Fill in the indicated missing voices in the following excerpts by
Bach. Be attentive to motivic coherence and balance of rhythmic
activity. Play and check your lines and counterpoint with care (check
3. each of the six pairs of voices for problems of parallel or direct
intervals and unresolved dissonances). Analyze the harmony before
you write, and analyze all the non-harmonic tones when you have
finished.2
Work out the following figured and unfigured basses in four
contrapuntal voices, using a mixture of note values, including
4. sixteenths. Work for evenness of both motivic distribution and
rhythmic activity. The given bass voice may be slightly elaborated
for rhythmic and motivic interest. Check carefully and analyze fully.
Use the following chord formats as the basis for counterpoint in four
voices. Employ a texture of four equal voices and a limited motivic
5.
content. Check your counterpoint; analyze fully. See the
demonstration below for process.
Demonstration
2. Articulate.
3. Analyze:
Exercise
Articulate the following four-voice frameworks in any meter and
tempo. Each should be at least eight measures long. Tonic or
dominant harmony may be prolonged as that seems appropriate. Use
6. contrapuntal texture and motivic imitation. Provide figured bass,
then analyze the harmony. Analyze fully when the work is completed
and check your lines and counterpoint with care. See the
demonstration below.
Demonstration
Exercise
Analyze a complete four-voice (nonfugal) suite movement or prelude
7. and use it as a model for a comparable original work. Include in your
model the main cadence points, key scheme, and harmonic structure.
Passacaglia (chapter 13) or chorale prelude (chapter 14) may be taken
up at this point, before or instead of the study of four-voice fugue.
Four-Voice Fugue
Analyze the fugal excerpts beginning this chapter. Identify S, A, CS (if any),
bridge (if any), and locate the end of the exposition. What is the order of
voice entries in each? Do you notice any procedural differences from three-
voice fugue expositions? Graph each excerpt as shown on p. 256.
Discussion
While four-voice fugue is common in the choral and ensemble music of Bach,
as well as in his organ music, his other music exhibits a slight preference for
three-voice fugues. The Well-Tempered Clavier, for example, has one fugue
for two voices (The Well-Tempered Clavier I, Fugue No. 10), two for five
voices (The Well-Tempered Clavier I, Fugue Nos. 4 and 22), nineteen for four
voices, and twenty-six for three voices. Four-voice fugues are not very
different procedurally from those with three voices. There are some
differences: fugues in four or five voices tend to be weightier and graver in
character; their subjects are often shorter, slower moving, and narrower in
range (see the subjects in examples 12-1 to 12-4); they are also often slightly
more homophonic in texture, with more voice crossing and more typically
chordal endings. Four-voice fugues are not necessarily longer than those in
three; this is more a matter of the length of the subject and the nature of the
working-out processes.
Procedurally, four-voice fugues tend to avoid the more complex
contrapuntal treatments. Passages in three- or four-voice canon or stretto are
fairly rare and usually brief. The countersubject (CS) material is often less
consistent, and while double counterpoint is common, triple and quadruple
are not. Because of the ineffectiveness of unrelieved textural thickness, the
episodes are often reduced to three voices. There may even be brief two-voice
passages, for instance, at the beginning of a stretto section. As mentioned
above, when a voice drops out for more than a few beats, it must have come
to a point of completion, on a stable note; and when it reenters, it will
normally have important material, such as the subject (S) or answer (A).
Structurally, the four-voice fugue exhibits no schemes not found in three
voices, though counter-expositions are rare, and the bridge when present is
more variable in placement.
Since there is one more voice to bring into the exposition, we will focus on
this section. Following is a graph of the exposition of The Well-Tempered
Clavier II, Fugue No. 7 (example 12-1).
Here we find the entries S-A-S-Bridge-S, along with two quite consistent
CSs. The bridge, when placed between entrances 3 and 4, gives the effect of a
three-voice fugue with one episode, followed by a final entrance of the S,
almost as if there were to be a counterexposition. The Well-Tempered Clavier
I, Fugue No. 14 also has a variant exposition.
Exercises
1. Play and analyze several four-voice fugues from the anthology.
Analyze these in detail, showing their thematic structure and
overall layout. Identify all the thematic material, episodes, middle
entries, strettos, sequences, and so on. Do a formal graph, as on p.
212, and a structural-pitch graph. Show the harmonic scheme,
including all keys and cadences, and the harmonic plan of the
episodes. A detailed, bar-by-bar analysis, as suggested on p. 236,
may also be helpful.
2. Compose a four-voice fugue on an original S, according to the
processes suggested under three-voice fugue, pp. 238 ff. It is a
good practice to compose models of stretto passages (it is
suggested that you use a subject that will permit at least two-voice
stretto), the exposition (with at least one consistent CS), several
episodes, and perhaps also the ending section, with pedal points.
This should be done in stages, with the instructor checking and
approving all stages. Each stage should be fully analyzed for all
aspects.
Other Fugal Variants
Fugato
This term, which simply means “fugue-like,” refers to a fugal section within a
larger nonfugal work. Such passages occur frequently in the development
sections of instrumental works by the Viennese classical and romantic
composers.
Fughetta
Double Fugue
A fugue with two more or less equal subjects is a double fugue. These
subjects will appear either together in the exposition (as if they were S and
CS), and subsequently through the fugue, or with two separate expositions
and combined toward the end of the fugue. The two subjects must be written
in double counterpoint.
In the first category, the second subject (S2) is heard as a normal though
prominent CS in the exposition. It is associated throughout with SI, and may
be used as the basis for episodes. It often assumes increasing importance as
the fugue progresses. The Well-Tempered Clavier I, Fugue No. 12 (in the
anthology) is a good example of this procedure.
A subcategory of the first category has the two subjects appearing together
from the very beginning (usually m. 1). They are always associated, in double
counterpoint, in the course of the fugue, and either or both may form the basis
for episodes. If one of these subjects seems distinctly less important, it may
better be understood as a CS. This type does not appear in The Well-Tempered
Clavier, but it may be found in some organ fugues, such as the fugue from the
Toccata in E Minor.
In the second category, one finds in effect two expositions in succession,
the subjects of which are later combined in a climactic way toward the end of
the fugue. S2 often seems, when first heard, to be merely a new CS (as it is
normally heard against counterpoints derived from SI), but it then receives an
exposition-like series of imitations and is finally combined with SI. The Well-
Tempered Clavier II, Fugues Nos. 4 and 18 have such a scheme, as does
Fugue No. 9 from The Art of Fugue. Again, the association of S2 with motivic
material from SI may obscure the identity of the second exposition and the
importance of S2.
Triple Fugue
There are two principal options in triple fugue:
Exercises
1. Analyze any double fugue (such as the one on p. 344) and use it as
a procedural model. Your own fugue must have two equally strong
and interesting subjects, of somewhat independent character and
rhythm, written in the best possible double counterpoint.
2. Analyze The Well-Tempered Clavier, Fugue No. 4 (p. 350) and use
it as the model for a comparable triple fugue. You should compose
three equally strong subjects that have distinctive individual
character and rhythmic identity (each should move in somewhat
different note values from the others, as in the model fugue), and
that form solid triple counterpoint, as this will be the contrapuntal
model for the entire work. The analogous measures of this fugue
are mm. 49-51 (or 52-54 or 59-61). Write this in stages, as assigned
by the instructor, with each stage checked and approved. Analyze
fully.
Notes
1. Of course, in keyboard music the span of the hand imposes spacing limits on adjacent voices, as
well as on the range and intervals of each individual voice. The music is very well calculated to “fit
under the hand.”
2. The additive process in these exercises is analogous to one Bach himself employed occasionally
when reworking an earlier composition; for example, his addition of a fourth voice to the slow
movement of the Trio Sonata for organ, BWV 527, which became the slow movement of the Triple
Concerto, BWV 1044.
Chapter 13
Variation Forms
Ex. 13-1
The simplest diatonic version is the oldest, and the basis for all subsequent
versions.
The aesthetic dangers in a process based on the manifold repetition of the
bass line are obvious:
squareness and discontinuity of form
repetitive phrase structure and harmony
too much tonic harmony and too many authentic cadences
a homophonic tendency in the process itself
aimlessness (no sense of overall structure)
melodically eliding the upper voices over the phrase break, and making
sure the upper-voice phrase endings do not always coincide with the bass
line cadences (Henry Purcell was the great master of this technique)
covering the phrase break by motion and suspensions
choosing an eight-measure in preference to a four-measure pattern
varying the harmony as much as possible, given the limitations imposed
by the implications of the bass
using as much motivic imitation and rhythmic independence as possible
to achieve the effect of a contrapuntal texture
avoiding authentic cadences and tonic harmony by suspending notes into
the tonic triad, by substituting VI6, or for I, or by choosing a
pattern that places the dominant rather than the tonic harmony on the
first strong beat (see the Henry Purcell and Dietrich
Buxtehude versions, examples 13-1, C and D)
thinking through the entire work as a single shape (or series of shapes),
building throughout in terms of textural density, rhythmic activity,
harmonic complexity, use of extreme registers and intricacy of
counterpoint
allowing the bass line to move to another voice, to be ornamented, or to
be transposed to one or two other keys
Most of these devices may be seen in example 13-2, and may be observed
in the Passacaglia in C Minor, pp. 376 ff.
There may be an overall tonal plan in larger passacaglias, especially those
for organ. The organ is the most appropriate instrument for this form, as the
bass, with its long notes, can easily be taken by the pedals. Longer organ
passacaglias may be organized in a broad symmetrical design such as the
following, taken from Buxtehude2:
Section Repetition of Bass Pattern Key
1 7 tonic
2 7 relative
3 7 dominant
4 7 tonic
In this scheme there will be a textural and rhythmic intensification within
each section, and through the set. There may also be brief modulating
interludes between the sections. Within each section, variations will be paired
or grouped together by the use of a motif or rhythmic figure.
Bach must have been aware of both the dangers and the opportunities in the
passacaglia. In the extremely affecting “Crucifixus” from the B-Minor Mass
(example 13-2) he uses a chromatic version of the tetrachord, and elides the
cadences (mm. 9, 13) by suspending voices and by bringing in the next voice
on the next weak beat. The harmony is intensified from one variation to the
next (for instance, in m. 5, VI6 substitutes for I, avoiding the expected
cadence; vii°7/iv inm. 6 substitutes for the v6 of m. 2; the in m. 7 replaces
the iv6 of m. 3), and becomes more and more expressive and dissonant as the
movement progresses. The use throughout of suspensions also contributes an
important element of tension and continuity. The voices have their own
motivic material, independent of the bass line, which they share in a process
of motivic imitation. The entire movement (to be found on p. 394) builds
from beginning to end, with the textural density increasing, the counterpoint
increasing in complexity and the harmony in intensity, and the soprano rising
to a high point (e2) near the end. There are three internal cadences, which
provide momentary textural and harmonic relief, and at the end there is a
modulation to G major, via a root position German augmented-sixth chord.
The overall shape could be graphed as:
This graph shows a main climactic section, with three preceding but
subsidiary upward curves that mirror its shape. The principal climactic
moment surrounds m. 42, close to the end; this late climax makes for a
dramatic contour.
Ex. 13-2 Mass in B Minor, “Crucifixus”
Ex. 13-4
The variations are organized not around the melody of the aria, but around
the harmonies and bass line. The first eight measures of the bass, a standard
bass line for the time, are given here, with the harmonization from the aria.
Ex. 13-5
Exercise
1. Use one of the following unfigured basses, as assigned by the
instructor, as the ground bass for a passacaglia. Before you start,
write figured bass symbols and/or roman numerals below the bass
lines, giving every stylistically possible harmonization for each
bass note. As you compose, keep in mind the necessity for motivic
unification, rhythmic continuity and overall shape.
1. Bass is given.
2. Add appropriate figures or roman numerals.
3. Write the upper voice, working for a clear shape and strong
outer-voice framework.
4. Part-write the inner voices, working for clear harmony and
smooth voice-leading.
5. Articulate the four-voice skeleton, working for flow and
motivic unity.
Process Demonstration
Exercises
Compose a passacaglia in the minor mode for organ or for
instruments available in class, on a newly composed bass.3 One of
2. the following two formats for the whole work may be used, at the
discretion of the instructor. Analyze the completed work fully, and
perform.
Section Number of Bass Repetitions Key
A. 1 4 tonic
2 4 relative
3 4 tonic
B. 1 4 tonic
2 4 relative
3 4 dominant
4 4 tonic
For students adept at the keyboard, the improvisation of passacaglias
and chaconnes will be an interesting and challenging exercise. Bach
and his great contemporaries were, to all reports, superb improvisors.
The improvisation of variations over a given bass is far less daunting
3.
than the playing, ex tempore, of more rigorous genres such as the
fugue. Group improvisations, jazz-like, over a given harmonic
framework, will be entertaining for classes and excellent ear and
musicianship training in the style.
Notes
1. For further historical background, see The New Grove Dictionary of Music and Musicians,
“Chaconne” (vol. 4), “Ground” (vol. 7), “Ostinato” (vol. 14), and “Passacaglia” (vol. 14).
2. Bach’s Passacaglia in C Minor has no such key scheme, as it does not modulate. The “Crucifixus”
from the B-Minor Mass likewise does not modulate (until the end).
3. A brief introduction to writing for the organ may be found in the Appendix. A demonstration of
organ-playing techniques will be helpful and interesting, if an instrument is available.
Chapter 14
Cantus Firmus Procedure: The
Chorale Prelude
The first two phrases of the original chorale tune on which this prelude is
based are:
In example 14-1, the chorale melody in the first phrase is placed by itself
on a separate manual, so that it can be given an organ registration (color) to
distinguish it from the voices around it.
For a complete example of this type of chorale prelude, see the anthology,
p. 383.
Cantus with Motivic Counterpoints
There are two subcategories here. In Bach’s usage one is not always clearly
distinct from the other.
In the first type the chorale tune is accompanied by free motivic
counterpoints not clearly related to the cantus itself. The voices may relate
motivically to each other, or they may not. The following examples are
selected mainly from the Orgelbüchlein, which is rich in this type of
procedure.3
example 14-2, a relatively simple chorale prelude for manuals only, uses a
figural accompaniment unrelated to the tune, which is found in voice 1.
Ex. 14-3 Christ lag in Todesbanden
example 14-6 employs motivic imitation based on the first two phrases of
the chorale melody.
In many chorale preludes of this type, the relationship of the counterpoints
to the cantus is well hidden; the process of motivic derivation in these works
will require close study.
Canonic Treatment of Cantus and/or
Accompanying Parts
The Canonic Variations on “Vom Himmel hoch” have already been discussed.
The double canon in “In dulci jubilo” has also been mentioned (p. 119). The
following excerpt contains a canon between voices 1 and 4 at the octave, at
one measure, on the chorale tune. The inner voices have free, though
motivically related, counterpoints. In such works the chorale tune may be
involved in the canon or in the other voices, or (as in ‘In dulci jubilo”) both
may be canonically treated.
In example 14-8 the accompanying voices are based on the first three notes
of the cantus and treat it in a manner similar to a two-voice invention prior to
its entrance in the pedal (at 4′) in m. 6.
In example 14-9 the two upper voices have an extensive imitation based on
an ornamented version of the chorale tune, supported by a nonimitative
(though motivically related) bass. The melody appears in the pedal, with a 4′
stop.
In the second type, the imitative voices are based successively on each
phrase (or only its head) of the melody before it enters in long notes. As these
works are lengthy, none is shown here, but one (Jesus Christus unser Heiland)
is given in the anthology (p. 387). This category is often referred to as a
chorale motet or chorale ricercar. The motet (vocal) and ricercar
(instrumental) were Renaissance forms consisting of successive imitative
sections (“points of imitation”), each section based on a new phrase of the
melody, rather like a series of fugatos, and are the principal forerunners of the
fugue.
Chorale Prelude with “Obbligato” Melody:
(“Ritornello” Procedure)
Here, a clearcut and independent theme, usually regular in phrase structure
and even periodic, precedes the entrance of the cantus. This theme, which
may or may not be related motivically to the cantus, recurs throughout in the
soprano voice in a rondo-like manner, either whole or in part, between and
against successive phrases of the chorale melody, which is usually placed in
the tenor or pedal. A complete restatement of the theme follows the last
phrase of the chorale. Perhaps the best-known example of this type is Bach’s
setting of “Wachet auf, ruft uns die Stimme,” given complete in the
anthology, on p. 391.
Exercises
1. Analyze several of the chorale preludes in the anthology, starting
with the simpler varieties.4 Consider the motivic relationship (if
any) of the accompanying voices to the chorale melody; such
relationships are often obscured by filling in, inversion, and/or
ornamentation. Analyze the implied harmony of the melody first,
or Bach’s own harmonization of the chorale (as found in the 371
Chorales), and compare this to the harmonization used in the
prelude. Consider the relations between the accompanying voices:
are they unrelated, motivically related, or imitative?
2. Analyze the Bach chorales below in terms of keys, cadences,
chords, nonharmonic tones, and motifs in the melody (the soprano
voice).
Du Friendensfürst, Herr Jesu Christ
These excerpts may be written for three or four instruments in the class,
or, if possible, for the organ, using two manuals and pedals. The cantus
firmus should not be highly ornamented.
Process Demonstration
Given chorale phrase:
Exercises
Select your most successful result from exercise 3 above, and
continue it through to the end in the same style. Especially if you
4.
choose to compose the Vorimitation or obbligato type, you may wish
to refer again to one of the models in the anthology. Analyze fully.
Harmonize one of the chorale melodies below in the Bach style.
After it has been checked for correctness of harmony and
counterpoint, use it as the basis for a chorale prelude of any type, as
suggested in exercise 3. Analyze the melody carefully for motivic
5. content and commit it to memory before starting to write. There may
be three or four voices, and it may be written for instruments from
the class or for organ, as specified by the instructor. You may choose
to ornament the cantus slightly. The cantus may appear in soprano,
tenor, or bass (pedal).
For keyboard performers, especially organists and harpsichordists, the
exercise of improvising chorale preludes will be excellent training in
harmonization, style awareness, ear training, and musicianship.
Wer nur den lieben Gott
2. Incidentally, the fermata signs in the chorales are not necessarily intended to indicate that the
affected note be held, but to show phrase endings.
3. The Orgelbüchlein is a collection of forty-six chorale settings dating mainly from the years 1713-
17.
4. Those wishing to compose any of these projects for the organ are referred to appendix 2.
Conclusion
We return, at the end of our study, to some thoughts expressed at its outset.
Your intensive investigation of Bach’s style and technique should be a lasting
source of pleasure and instruction to you. It will already have made you a
better musician—a more understanding performer and listener, a more skilled
composer—and will have reinforced your admiration for the intellect and
spirit of a great human being. In addition, you will have gained a renewed
respect for your own creativity, musical intuition, and ability to master a
complex craft. I hope you will carry these possessions with you always.
Appendix 1:
Harmony
One of the principal reasons for the solidity of Bach’s music is its strong,
functional harmonic basis.1 Chords are placed, and resolve, in predictable
ways (that is, to ears at all familiar with tonal music). Bach’s harmony, while
rather more chromatic than that of his contemporaries, is still mainly diatonic.
This book focuses primarily on his diatonic style, since the more
contrapuntally complex a texture is, the simpler the harmony usually
becomes.2 The chart below is intended simply as a review and for quick
reference.
*The symbol / indicates a secondary dominant chord, so that the symbol /iii
reads “any secondary dominant of iii.”
General Comments on Chord Progression in
Bach
Common progression (down the chart, chord-to-chord) predominates in Bach.
The tonic chord may progress to any other chord.
Other progressions are found often but rarely two in a row. In other words,
a less common progression is normally followed by a common one. Of the
less common progressions, some of the most characteristic are iii-IV, vi-V, V-
vi, and ii—vi.
Since in this music, especially in the thinner textures, it is often difficult to
distinguish ii from IV, and V from V7 or vii°7, the symbols T (tonic), D
(dominant), and PD (predominant) may prove very useful.
The chords toward the bottom of the table above are found most often.
Some simpler works use only tonic, dominant, and predominant chords. The
higher one goes in the chart, the rarer the chords, in most works. The chords
near the bottom of the chart are also most frequently prolonged, and function
as departure and goal harmonies, (structural harmonies), especially I and V.
Chords other than I and V may be thought of as linear (contrapuntal,
decorative), rather than structural, in function.
Works usually begin on tonic harmony and always end on it. The feeling of
a tonal center is rarely suspended. Tonic harmony is most often placed on a
strong beat, at the beginning and/or end of a phrase.
The note forming the chord seventh is often short and metrically weak,
casting some doubt as to whether these notes are best considered chord tones
or nonharmonic tones in such instances.
Chord progressions (chord functions) are largely the same for both major
and minor modes, although there are a few special cases in minor, involving
the descending tetrachord or harmonic sequences.
Harmonic rhythm is usually patterned and regular. The faster the tempo, the
slower the harmonic rhythm will tend to be, avoiding a feeling of nervousness
in the harmony. Chords usually change from a weak into a strong beat, and
thus the bass line moves over the barline (unless the bass is suspended). Thus,
harmonic rhythm supports meter. The harmonic rhythm may accelerate in
mid-phrase; at phrase endings it may either speed up or (less often) slow
down.
Cadences (breathing or resting places, phrase endings, harmonic arrival
points) are a principal means of clarifying form in this music. The harmonic
goal of the cadence arrives on a strong beat (the exception is at a half
cadence, where the V may be preceded by a (I ) with a suspension- or
appoggiatura-like effect. Following is a brief overview of the four available
cadence types in this style.
2. Half (semi-) cadence (HC): an internal (medial) cadence type only, with
tonic or subdominant (or much more rarely vi or a secondary dominant
of V) moving to dominant. The goal chord is almost always a root
position V.
3. Deceptive cadence (DC): replaces an expected PAC, usually very near
the end of a work, with the expected tonic chord being replaced by a vi
or, more rarely, a IV6 or V /V, with 5̂ → 6̂ in the bass.
4. Plagal cadence (PC): the so-called Amen cadence, with subdominant
moving to tonic harmony. These may occur at the ends of relatively
substantial works (such as a fugue) following a strong PAC.
The most common modulations are to the relative and dominant keys. Note
that all closely related keys are diatonic chords in the main key, and that all
differ from the main key by a signature difference of no more than one
accidental.
Modulation is most often accomplished by common (pivot) chord. The first
key is clearly tonicized, and after the common chord the new key is carefully
tonicized by harmony, line and scale before its cadence. The new key is
nearly always established by a PAC. Lacking such a cadence, there may be a
question to the ear whether a modulation has occurred. The common chord in
Bach is diatonic in both keys and is typically a predominant chord in the new
key, followed immediately by the first dominant chord in that key. A typical
common chord modulation could be graphed thus:
a. Passing tones fill in by step between chord tones. Some theorists refer to
an accented passing tone as an appoggiatura.
b. Neighbor tones (auxiliaries) embellish a repeated chord tone by moving
away and back by step.
c. Escape tones are approached by step and resolved by leap, usually
downward. They are often cadential, as in the last example.
d. Anticipations occur immediately before an expected chord tone. They
are most often used at cadences, anticipating the tonic note.
e. Appoggiaturas are approached by leap and resolved by step, usually
downward. They are accented and often relatively long.
f. Incomplete neighbor tones are like brief, unaccented appoggiaturas,
approached by leap and resolved by step. They are sometimes termed
cambiatas.
g. Suspensions are prepared on the same pitch and resolved by step, usually
downward. They are placed on a strong beat. The suspension idiom
involves three facets: preparation, suspension, and resolution by step.
Ex. A-1
Figured Bass Symbols
Figured bass symbols were used by Bach and his contemporaries to indicate
intervals above a given bass line. These symbols were converted into
harmonies by the continuo (harpsichord or organ) player, improvising with
the right hand on the intervals indicated by the symbols, while the left hand
played the bass line. Bach composed—and taught—in terms of these interval
combinations, not of the roman-numeral chordal system, which postdates
him.
The arabic figured bass numerals show which intervals (and their
compounds) are to be played above each bass note. They do not specify
spacing, register, doubling, or texture. These are the standard symbols:
root position: 7
first inversion:
second inversion:
third inversion:
Any interval above the bass note can be shown in the figures, including
nonharmonic tones, have chosen to indicate only chord tones in my use of the
figures, with the exception of the
Notes
1. In introducing this material the instructor may wish to discuss the fact that Bach thought—and
taught— in terms of figured bass symbols, not roman numerals. Because of the problems a purely
figured-bass approach could create for students trained in roman-numeral thinking (and implicitly in
Jean-Philippe Rameau’s chordal-identity concepts), this text suggests using the chord-functional
approach. An instructor preferring the figured bass approach will find this text compatible.
The sections on harmony, modulations, and nonharmonic tones are primarily intended as a brief
review, but even well-prepared classes will benefit from going through them. See Notes to the
Instructor, p. xiii, and the entries “Harmony” (vol. 8) and “Modulation” (vol. 12) in The New Grove
Dictionary of Music and Musicians.
2. There are, of course, some fascinating exceptions, especially in the fugues and canons with
chromatic subjects.
3. This section discusses nonharmonic tones in general. Bach’s characteristic nonharmonic tone usages
are taken up on pp. 23 ff.
Appendix 2:
Composing for the Organ
For students wishing to write any of their composition projects for the organ
(especially the projects involving fugue, passacaglia and chorale prelude), the
following brief comments may prove helpful.
Bach’s pedalboard has a range of from C to d1 (or f1); the two or three
manuals range from C to d3 (or c3). The pedals are ideal for sustained notes.1
Rapid scalar patterns are difficult for the pedals. Leaping patterns that
alternate the feet are more practicable (see the following example). Pianists
will have to remind themselves that there is no damper pedal on the organ or
harpsichord. A key must be held down to sustain that note; therefore, pianistic
figurations such as wide arpeggios are ineffective, and are in any case not
characteristic of the rigorously contrapuntal side of the Bach style. An ideal
texture for organ writing, and one in which Bach excels, is the trio-sonata
texture of two equal upper voices and a linear bass in three-voice
counterpoint. Four-voice writing is also typical, but keep in mind that if two
voices are played on one manual, the notes must fit within the compass of the
hand (in other words, must “lie under the hand”) to be performable. The best
course is to study several characteristic passages from the organ music, as
found in the anthology.
Observe in the following passage from the Passacaglia in C Minor the three
equal and linear upper voices, and the sustained notes and alternating-feet
patterns in the pedal part.
Notes
1. It should be understood that the bass voice, when played by the pedals, will normally be doubled at
the lower octave by the addition of a 16′ stop. To avoid this effect, one would have to specify 8′
(sounding at the written pitch) or 4′ (sounding an octave higher, as in some chorale preludes where
the chorale melody is in the pedal).
Glossary
This glossary gives very brief definitions of principal technical terms used in
this text, and is based on the usages found in Bach. It is not intended to cover
all possible cases, just the most common ones. The reader is referred to the
index page references for fuller coverage of each topic, as well as to the
resources cited in the bibliography.
Affect (Affekt, in German) Expressive character of a given work or
movement, inherent in its tempo, mode, intervals, gestures, genre, and
musical figures. Baroque movements are highly unified in affect and musical
figure, and each figure embodies an affective or rhetorical meaning. These
concepts were variously classified in the seventeenth and eighteenth centuries
into the Doctrines of the Affections (Affektenlehre), and of the Figures
(Figurenlehre), but were never firmly codified nor universally agreed upon.
Answer (response) In a fugue exposition, the second statement of the subject,
usually transposed to the dominant key.
Augmentation Statement of a musical idea (such as a fugue subject) in
slower note values, usually in a 2:1 ratio, compared to the initial statement.
Basso continuo See figured bass.
Basso ostinato See ground bass.
Bridge Transitional passage, as in a fugue exposition—typically of two to
four measures and usually following the end of the answer—that remodulates
to the tonic key preceding the next entrance of the subject.
Cadence Resting or breathing place in music, marking the end of a phrase,
section, or composition; in tonal music cadences are often classified by the
final two harmonies at the point of cadence.
Canon Composition or extended passage in which the voices imitate each
other strictly. See imitation.
Canon cancrizans (crab canon, or retrograde canon) Canon in which one
voice imitates the other by retrogression (backward from end to beginning).
Canon per motu contrario (canon by contrary motion, canon by
inversion) Canon in which one voice imitates by melodic inversion.
Canon per augmentationem Canon in which one voice imitates by rhythmic
augmentation, usually in a 2:1 or 4:1 ratio.
Cantus firmus Preexisting melody, such as a Lutheran chorale, used as the
basis of a contrapuntal work, such as a chorale prelude.
Chaconne Variations based on a repeated chord progression, most often I-IV-
V-I and variants. Closely related to the passacaglia, and often
indistinguishable from it, though chaconnes are more usually in major mode.
May be sectional or continuous.
Chorale prelude Composition based on a preexisting chorale melody, most
often heard in the upper or tenor voice, or in the organ pedals; for organ or
(more rarely) chorus. See cantus firmus.
Codetta Brief (two- to six-measure) ending section of an invention or fugue.
Composite rhythm (macrorhythm or resultant rhythm) Overall pulse, the
combined effect of the rhythms of all voices.
Compound line Single musical line that gives the effect of embodying
several voices, most often through registral differentiation of these voices.
Counterexposition In a fugue, a second exposition, in tonic and dominant
keys, that follows the first exposition.
Countersubject In a fugue, a consistent counterpoint heard against the
subject and answer in the exposition and often throughout the fugue.
Cross-relation (false relation) Relationship between a scale degree in one
voice closely followed by a chromatically altered version of that pitch in
another voice; these alterations most often involve degrees 6 and 7 in minor.
Diminution Statement of a musical idea (such as a fugue subject) in quicker
note values, usually in a 1:2 or 1:4 ratio, compared to the initial statement.
Direct (hidden, covered) fifths and octaves Voice-leading situation in which
two voices move by similar motion into a perfect fifth or octave. This effect is
generally avoided in Bach, especially when both voices leap, particularly in
two-voice (and outer-voice) writing.
Double canon Procedure in which two different canons occur at the same
time.
Double (invertible) counterpoint Relationship between two voices such that
the counterpoint is equally effective when either voice is placed above (or
below) the other.
Double fugue Fugue with two distinct subjects, which may be presented
together in the exposition or may receive separate expositions, and are later
combined.
Episode Transitional and/or development passage (as in an invention or
fugue), typically modulatory and sequentially organized.
Exposition Opening section of a work (such as an invention or fugue),
containing the principal musical materials of that work; in a fugue, this will
include the initial statements of subject and answer in tonic or dominant keys
in all voices. In many fugues it also includes a brief episode, called a bridge.
Figures See affect.
Figured bass (thoroughbass) Bass line under or over which arabic figures
have been placed to indicate intervals above the bass notes, and by
implication the harmonies. In Baroque music, the indicated harmonies would
be filled in (realized) by the keyboard player; figured bass is often employed
in music theory instruction, and was so used by Bach.
Follower Second voice to enter in an imitative work or passage.
Fugato Developmental passage in a larger work (as in development sections
in sonata form) written in an imitative style similar to a fugue but not carried
out as rigorously.
Fughetta Brief fugue, usually with only one episode and one modulation at
most.
Fugue Imitation-based contrapuntal genre, in which the entire work is
developed from its initial materials (subject and countersubject), as found in
the exposition. Developmental episodes and middle entries may alternate,
unless the work is a stretto fugue or double fugue. There will usually be
modulations to closely related keys.
Ground bass (basso ostinato) Repeated bass-line pattern (and thus the
harmonic patterns implied by it) used as the basis of a set of variations. See
chaconne and passacaglia.
Harmonic rhythm The speed of and rhythmic patterns formed by chord
changes. Harmonic rhythm in tonal music is usually regular and supports the
feeling of meter.
Hemiola As applied to musical rhythm, the ratio of 3:2 between note values
either successively within one voice or simultaneously between two. This
effect is often found in Baroque music in certain dances such as the minuet,
approaching cadences.
Hidden fifths See direct fifths.
Imitation Relationship between voices such that voices enter successively
using the same melodic material. The first voice, or leader establishes the
musical materials, which are then replicated in the other voices or followers.
In analyzing imitative passages or works, the time and pitch intervals between
successive voice entries are noted, as well as the length for which the
imitation is carried out in each voice.
Invention Imitation-based two-voice contrapuntal genre in which the entire
work is developed from the initial idea or theme, and sometimes from its
counterpoints (counterthemes). Very like a two-voiced fugue in procedure and
overall layout, with developmental, modulatory episodes and recurring entries
of the theme.
Invertible counterpoint See double counterpoint. Three or four voices may
also be written in invertible counterpoint (thus, triple or quadruple invertible),
such that many or all possible repositionings (top to bottom) of the voices will
be effective.
Leader The first voice to enter in an imitative passage or work.
Linear-intervallic pattern (LIP) Recurrent pattern of intervals formed
between pairs of voices, most often in sequential passages.
Macrorhythm See composite rhythm.
Middle entry Complete entry of the subject or theme in a fugue or invention,
anywhere after the end of the exposition. Such entries will typically be in
closely related keys, and may use melodic inversion or (rarely) retrogression,
augmentation or diminution.
Motive (motif) The smallest unit of melodic and/or rhythmic thematic
material, used as the basis for development as a composition unfolds. Motives
in Bach are often very short, always shorter than a phrase and often of three to
four notes or rhythmic values. They are spun out in the course of a
composition by the processes of repetition, fragmentation, extension,
sequence, inversion and so on.
Neapolitan chord Harmonic structure with predominant function; a major
triad built on the lowered-second scale degree. In Bach, used mainly in minor
mode, as a cadential-preparation harmony.
Ostinato Regularly recurrent pattern, used as the basis of a composition. See
ground bass.
Parallel (consecutive) fifths and octaves Voice-leading situation in which
perfect fifths or octaves are found in succession between any pair of voices.
This effect is universally avoided in tonal (and much pretonal) music, as it
detracts from linear independence. The prohibition on parallel octaves is not
to be confused with the consistent doubling of a melody or bass line for
acoustical emphasis. See direct fifths and unequal fifths.
Passacaglia Variations based on a basso ostinato or ground bass pattern,
usually involving a descending tetrachord from tonic to dominant scale
degrees. The bass line may be diatonic or chromatic, and the work will tend to
be seamless rather than sectional. Typically minor mode and triple meter,
often slow and expressive in affect. See chaconne.
Pedal point Sustained pitch, normally a tonic or dominant note in the bass,
over which the surface harmony changes. Pedal is a principal means of
prolonging and stabilizing underlying harmony. Dominant pedal is most often
found in the middle or nearing the end of compositions such as fugues, in
places where an increase in tension is desired; tonic pedal may begin or end
works, to clarify and stabilize the key center.
Period (periodic) structure Phrase-level formal unit consisting of two
successive phrases of equal length, in a question-answer (or antecedent-
consequent) relationship. The medial (middle or internal) cadence (usually
half cadence or imperfect authentic cadence) is less final sounding than the
final (usually a perfect authentic cadence). Periods are often classified in
terms of the melodic relationship between the two phrases.
Prolongation Emphasis on a particular scale degree or harmony (see
structural harmony and structural pitches). Prolongation can be accomplished
in a variety of ways, through various linear ornamentations (passing,
neighboring or arpeggiated), pedal point, reiteration and so on. Principal
prolonged pitches form the fundamental melodic line (Urlinie) and underlying
harmonic structure.
Real answer Imitative answer, as in a fugue, in which the exact interval
structure of the leader is preserved in the follower, transposed to a different
degree of the scale, most often the dominant degree (and thus temporarily to
the dominant key). Compare tonal answer.
Reduction, graphic (linear reduction, Schenkerian analysis) System of
analysis showing in musical notation and related symbology the principal
melodic and harmonic events in a tonal work. Based on the work of Heinrich
Schenker and his followers, this way of thinking about musical structure
focuses on the overall shaping of a work’s principal lines and harmonies,
within the context of tonality (though the methodology has been applied to
other repertoires). See prolongation, structural harmony, structural pitches.
Response See answer.
Ritornello Returning section, as in a rondo structure, that stabilizes a work.
Ritornello form is found in some chorale preludes.
Sequence Melodic or motivic pattern transposed successively to different
scale degrees; such patterning, a principal means of unification, usually
affects all voices simultaneously.
Stretto Imitation, as in a fugue, in which successive imitative entries occur at
shorter time intervals than those found earlier in the same work (for example,
in the exposition). A common device for building tension in an imitative
work.
Stretto fugue Fugue based on extensive use of stretto, which may replace
sequential episodes. Stretto entrances near the end may be at closer time
intervals, to increase tension and drive.
Structural harmony Principal harmony, underlying more localized surface
harmonic changes, and thus stabilizing and clarifying tonal centers. The
principal structural or prolonged harmonies in tonal music are tonic and
dominant and to a lesser extent predominant, and are often expressed as
prolonged pitches in the bass.
Structural pitches Principal scale degrees in a line, points of departure and
arrival, giving a musical line its sense of shape and direction. In tonal music,
notes of the tonic triad, emphasized by metric placement, length and other
means of prolongation, may be heard as more fundamental than other
(decorative) pitches. In Schenkerian terms, the principal structural pitches
form a composition’s fundamental line or Urlinie, which in this view always
descends from an upper note of the tonic triad to the tonic pitch.
Table canon (Tafelkanon) Two-voice canon in which one performer reads
from one side of the music and the other from the opposite side (with the
music, by implication, placed between them on a table). Thus, a canon by
retrograde inversion.
Theme In an invention, the principal musical idea, as given first in the upper
voice in the exposition. Analogous to a fugue subject.
Tonal answer Imitative answer, as in a fugue, in which a prominent dominant
pitch at or near the head of the subject is replaced by the tonic pitch (in the
principal key) at the corresponding place in the answer. Thus, a tonic-to-
dominant leap in the subject is replaced by a dominant-to-tonic leap in the
answer, and vice versa, preserving the tonal center.
Tonal framework (tonality frame) Principal highest and lowest pitches of a
melody that emphasize its tonal center. The tonal framework normally
involves pitches of the tonic triad.
Triple fugue A fugue with three distinct subjects. Compare double fugue.
Unequal fifths In voice-leading practice, a diminished fifth between two
voices followed by a perfect fifth; this is not usually found in two-voice
writing, or between outer voices of a thicker texture. A diminished fifth may
follow a perfect fifth as long as the diminished fifth resolves normally, but
again this is very rare in two voices.
Bibliography
David, Hans T., and Arthur Mendel. The Bach Reader. New York: W.W.
Norton, 1966.
Forte, Alien, and Steven E. Gilbert. Introduction to Schenkerian Analysis.
New York: W.W. Norton, 1982.
Gauldin, Robert. A Practical Approach to Eighteenth-Century Counterpoint.
Englewood Cliffs, N.J.: Prentice-Hall, 1988.
Horsley, Imogen. Fugue. New York: The Free Press, 1966.
Kennan, Kent. Counterpoint. Englewood Cliffs, N.J.: Prentice-Hall, 1987.
LaRue, Jan. Guidelines for Style Analysis. New York: W.W. Norton, 1970.
Mann, Alfred. The Study of Fugue. New York: W.W. Norton, 1965.
Naldin, Charles. Fugal Answer. New York: Oxford University Press, 1969.
Oldroyd, George. The Technique and Spirit of Fugue. New York: Oxford
University Press, 1948.
Parks, Richard. Eighteenth-Century Counterpoint and Tonal Structure.
Englewood Cliffs, N.J.: Prentice-Hall, 1984.
Randel, Don, ed. The New Harvard Dictionary of Music. Cambridge, Mass.:
Harvard University Press, 1986.
Sadie, Stanley, ed. The New Grove Dictionary of Music and Musicians. 29
vols. London: Macmillan, 1980.
Anthology
A
Accompanied canon. See Canon, accompanied
general, 57-58
Analysis
Answer, 290
Augmentation, 290
No. 8, 259
Chorale preludes
Christ Lag in Todesbanden, 175, 271 Christus, der ist mein Leben, 281
Durch Adam’s Fall ist ganz verderbt, 272 Gottes Sohn ist kommen, 214
No. 2, 144
No. 7, 111
No. 9, 181
No. 11, 83
Menuet I, 4
Passacaglia in C Minor for Organ, 210n. 8, 261, 263, 266, 268n. 2, 289
No. 2, 29
No. 3, 79
No. 5, 80
Sonata in G Minor, 13
second movement, 3
Bourrée, 30
Suites, English
No. 3: Gavotte, 41
No. 5: Passepied I, 2
Suites, French
No. 2: Courante, 43
Suites, Orchestral
No. 1, 108
Fugue No. 1, 188, 193, 205, 217, 219, 223, 224, 226, 228, 245, 256, 257
Fugue No. 2, 101, 179, 184, 204, 206, 210n. 8, 213, 217, 228, 230, 233-38, 242, 250
Binary form
in fugues, 211-12
in inventions, 135
Bridge
deceptive, 230-31
accompanied, 181-84
rounds, 120
stretto in, 181
Chorale
counterpoint, 272-74
motet, 276
Chord functions, xv, xvi, 55, 69, 76n. 10, 236, 263-65, 283-84
Chords
diatonic, 285
secondary dominants (see also Secondary dominants, Seventh chords), 78-81, 82, 174-75, 182, 186,
284
seventh, 55, 76n. 13, 154, 286 (see also Seventh chords)
structural, 247
nonharmonic, 78
spelling, 77
Coda
in inventions, 141n. 4
Coherence
motivic, 25-30
Contrary motion (see also Inversion), 44, 115-16, 118, 155, 223, 226, 240, 250
Counterpoint, 31, 61
Countersubject
Countertheme
melodic, 5
Dissonant intervals
Double counterpoint (see also invertible counterpoint), 97-106, 106n. 1, 112, 129-30, 135, 140, 141n. 3,
291
Doubling
formal, 137
sectional, 137
Entrances
order of in fugue, 192, 202, 208, 211, 213, 216, 217, 219, 236
Episode(s), 291
Expostion(s)
Figures, 291
filling-in, 16
neighboring, 15
repeated-note, 17
Form
in dance-suites, 89-96
in fugues, 211-12
in passacaglias, 260-63
general, 188
answer in, 192-94, 197-98, 202, 203-205, 206, 210n. 2, 211, 215, 223-24, 226, 232-33, 236, 255,
256-57
bridge in, 201, 204, 207-209, 210n. 7, 236-37, 243, 255, 256-57, 290
countersubject in, 177, 201, 203, 205, 206-207, 215, 217, 230-31, 233, 236, 243, 255, 256, 258, 291
episodes in, 178, 181-83, 212, 214-17, 220, 236-37, 241, 256
exposition in, 178, 188, 192, 196, 201-209, 203, 211, 216, 220, 231-32, 236, 243, 244, 247-48, 255,
256-57, 258
four-voice, 255-57
middle entries in, 178, 211, 212, 213, 217-18, 225, 236, 257, 292
organ, 258
stretto in, 191, 202, 211, 217-18, 219-20, 222, 223-27, 231-33, 237, 240-41, 242, 243n. 2, 248, 256,
257, 293
subject in, 188, 190-91, 193-94, 196-97, 202-205, 206, 211, 217, 221, 222, 223-28, 230-33, 236, 238,
241-42, 255, 256-57, 258
Harmony (see also Cadences; Chromaticism; Doubling; Figured bass symbols; Inversion), 283-88,
288n. 1
chorale, 270
general, 20
by augmentation, 117-19
by diminution, 119
motivic, 261-62
real, 114
tonal, 114
Interval(s), 285
consonant, 50-51
structural, 16
vertical, 76n. 9
Invention(s), 292
two-voice, 123-41
generalizations, 98
in inventions, 134
Keys
major, 285
Linear Intervallic Patterns (LIPs), xvi, xx, xxii-xxiv, 66, 130, 163, 184, 212, 214-16, 219, 242, 250, 292
Link
Melodic line
range of, 6
rhythm in, 14
sequence in, 28
shape of, 5
Meter
compound, 61-62
in dance-suite movements, 95
in melodic line, 13
Middle entry
Modulation, 288n. 1
chromatic, 286
Motion
relative, 153-54
Motivic
counterpoints, 270-74
chromatic, 82
in line, 24
Organ
Pedal(s)
points, 24, 149, 211, 227-31, 242-43, 243n. 3, 250, 257, 293
Perfect fifths or octaves, 45, 79, 99, 100, 153, 179-80, 214, 216
Progressions
chord, 284
harmonic, 283-85
R
Range
melodic, 5-6
Recapitulation
Relative motion
Repetition
Rhythm
chromatic, 7, 19-23
Second inversion of triads (see also Triads), 57, 157, 249, 284, 287
Secondary dominants (see also Altered chords; Chords; Seventh chords), 78-81, 82, 111, 171-72, 186,
229
Sequence(s), 293
circle-of-fifths, 57-58, 78, 80-81, 131-32, 139-40, 183-84, 216, 229, 237
pseudo-, 29-30
Seventh chords
doubling, 156
in two voices, 55
secondary dominant, 81
Spacing
Structural pitches
Subject
modulating, 190-91
Suspension(s), 286
Syncopation, 14
T
Table canon (Tafelcanon), 120, 294
harmonic, 22
melodic, 20-21
Texture
Theme
Three-voice counterpoint
Tonal framework
Triads, 287
diminished, 250
tonic, 261
framework, 68-73
forms, 260-68
Voice crossing, 48, 65, 76n. 5, 100, 101, 102, 111, 129