Singing in Style

Singing in Style
This page intentionally left blank

Singing in Style
A Guide to Vocal Performance Practices
Martha Elliott
Yale University Press
New Haven and London

Copyright © 2006 by Yale University.
All rights reserved.
This book may not be reproduced, in whole or in part, including illustrations,
in any form (beyond that copying permitted by Sections 107 and 108 of the U.S.
Copyright Law and except by reviewers for the public press), without written
permission from the publishers.
Set in Adobe Garamond and Stone Sans types by The Composing Room of
Michigan, Inc.
Printed in the United States of America.
The Library of Congress has cataloged the hardcover edition as follows:
Elliott, Martha.
Singing in style : a guide to vocal performance practices / Martha Elliott,
p. cm.
Includes bibliographical references and index.
ISBN 0-300-10932-6 (cloth : alk. paper)
i. Singing—Interpretation (Phrasing, dynamics, etc.) 2. Performance practice
(Music) 3. Vocal music—History and criticism. I. Title.
MT892.E45 2006
783'.043'09—dc22
2005009701
ISBN 978-0-300-13632-6 (pbk.: alk. paper)
A catalogue record for this book is available from the British Library.
The paper in this book meets the guidelines for permanence and durability
of the Committee on Production Guidelines for Book Longevity of the
Council on Library Resources.
ίο 9 8 7 6 5 4 3 2
To Michael and Emily
This page intentionally left blank
Contents
Acknowledgments, xi
Introduction, i
1 The Early Baroque, 5

Baroque Notation, 6
Scores and Editions, 7
Thoroughbass, 9
Accompaniment Instruments, n
Pitch, 13
Vibrato, 14
Words and Music, 18
Historical Pronunciation, 19
Italy, 19
Voice Types and Registers, 20
Ornamentation, 21
England, 32
Voice Types and Ranges, 33
Ornamentation, 35
France, 41
Dance Forms, 43
viii Contents
Notes inégales, 44
Ornamentation, 45
Diction, 49
2 The Late Baroque, 53

Scores and Notation, 54
Modern Editions, 55
Tempo, 56
Articulation, 58
Rhythmic Alteration, 61
Legato, 62
Vibrato and Breath, 64
Pitch, 65
Period Instruments, 66
Recitative, 66
Italian Influence, 69
Ornamenting Arias, 71
Handel's Messiah, 77
Germany, 78
Ornamentation, 79
J. S. Bach and His Singers, 84
France, 85
Secular Cantatas, 86
Ornamentation, 87
3 The Classical Era, 92

Classical Instruments, 95
Articulation and Legato, 97
Tempo and Rhythm, 98
Mozart on Singers, 106
Vibrato and Portamento, 107
Recitative, 108
Ornamentation, no
4 Italian Bel Canto, 126

Articulation and Phrasing, 129
Tempo, Rhythmic Flexibility, and Rubato, 132
Voice Types, 134
Technique, 136
Vibrato, 137
Portamento, 139
Contents ix
Ornamentation, 143
Cadenzas and Improvised Florid Embellishment, 149
5 German Lieder, 160

Editions, 162
Transpositions, 164
German Diction and Technique, 166
Articulation and Expression Marks, 169
Embellishment, 171
Notated Ornaments, 173
Vibrato and Portamento, 181
The Piano, 186
Tempo and Rhythm, 187
6 French Melodies, 194

Editions, Transpositions, and Voice Types, 196
French Taste and Aesthetics, 199
Ornamentation, 201
The Performer's New Role, 202
Words and Music, 202
Diction and Articulation, 203
Portamento and Vibrato, 209
Rhythm and Tempo, 212
7 Second Viennese School, 222

Publications and Editions, 224
Performance and Analysis, 226
The Society for Private Musical Performance, 228
Music and Text, 229
Pierrot lunaire, 233
Sprechstimme in Other Works, 240
Dynamics and Articulation, 243
Tempo and Rubato, 246
Learning This Music, 248
8 Early Twentieth-Century Nationalism, 251

Russia, 252
Musorgsky and Olenina, 253
Rimsky-Korsakov, Rachmaninov, and Chaliapin, 258
Prokofiev, 260
Shostakovich, 263
Editions and Language, 265
Spain, 266
Falla, 267
χ Contents
Rodrigo, 271
Editions, 272
England, 272
Britten and Pears, 273
Editions, 278
United States, 278
Ives, 279
Copland, 280
Barber, 281
Editions, 284
9 Working with Living Composers, 286

Voice Types, 287
Manuscripts, Revisions, and Published Scores, 290
Notation, 293
Special Vocal Effects, 294
Other Special Instructions or Requirements, 296
Learning Difficult Rhythms and Pitches, 297
Vibrato, 299
Text and Music, 300
Composers as Performers, 302
Using Recordings, 304
Conclusions: Finding the Composers' Intentions, 305
Notes, 307
For Further Reading, 327
Index, 347
Acknowledgments
This book came into being when I was searching for a publisher for
my Princeton University senior thesis on the three versions of Bee-
thoven's opera Fidelio. Friend, colleague, and Beethoven scholar Scott
Burnham introduced me to Harry Haskell, then music editor at Yale
University Press. Harry was not interested in publishing my thesis,
but he did think I was the right person to work on a project he had in
mind about vocal performance practice. I was excited and eager and
had no idea what I was getting myself into or how much I would have
to learn about research, writing, and the technical aspects of assem-
bling a book like this. I also had no idea how fascinating and reward-
ing the work would be. Throughout the long process of developing
and writing this book, Harry has been an unfailing source of wise
counsel and generous support, and I am deeply grateful to him for
both the opportunity and the help he has given me.
I am also grateful to everyone in the Mendel Music Library at
Princeton University, especially Paula Matthews, Daniel Boomhower,
and Dan Gallagher, for all their ongoing advice in matters large and
small. Other members of the Princeton University Music Department
xi
χίί Acknowledgments
offered help on many fronts, and I am most thankful to Scott Burnham for
moral support; to Mary Roberts for technical help; to Peter Westergaard, Steve
Mackey, and Simon Morrison for research contributions; and particularly to
Wendy Heller, who reviewed the manuscript with a fine-tooth comb from the
perspectives of both a musicologist and a singer. She also introduced me to her
friend Michael Burden, an invaluable source of information. Caryl Emerson,
chairman of Princetons Department of Slavic Languages and Literatures, also
offered generous help, including the tape of Maria Olenina d'Alheim's master
classes and the translating services of Ivan Eubanks, for which I am deeply ap-
preciative. I must also offer thanks to the many musical colleagues who either
read the manuscript and offered suggestions or shared their own performance
stories; these colleagues included Drew Minter, Phyllis Bryn-Julson, Paul
Sperry, Benita Valente, Mary Nessinger, David Ronis, Matthew Lau, Cheryl
Bensman, Peggy Kampmeier, and Sarah Freiberg. Sarah Freiberg (my cousin)
also merits special appreciation for reading a first draft and providing invalu-
able writing suggestions. Tremendous thanks go to Bob Wagner for his beauti-
ful music examples. Lauren Shapiro and her editorial team at Yale University
Press have been a joy to work with and were especially generous in helping me
navigate my first publishing effort.
To my friends and relatives, who have heard me talking about this project as
it has developed, I offer gratitude for your patience and support, especially to
Terry Desser, Mildred Freiberg, Barbara Westergaard, Anne Gray, Mary Jean
Link, and Candus Hedburg. Unbounded thanks also go to my many students,
who over the years have provided a fascinating laboratory in which to experi-
ment with historical approaches to vocal technique and style. Of course I
couldn't have completed this book without the support of my family, parents,
and beloved daughter, Emily. To my husband, Michael, who talked through
every idea, read every sentence, and patiently waited when I got lost in my com-
puter; who is my love, my friend, my teacher, and my favorite performing
companion: I can't thank you enough.
Introduction
This book introduces the issues of historical performance practice as

they relate to singers and vocal repertoire. It also investigates the ele-
ments that contribute to the style in which an individual work of mu-
sic, from a specific period and by a particular composer, should be
sung. Most important, it explores the working relationships between
singers and composers from the seventeenth century to the present
and traces how their collaboration shaped changing trends in style.
The early music movement and the accompanying explosion of inter-
est in historically informed performances have produced a tremen-
dous amount of information on performance practice. Most of it is
geared toward instrumentalists, though a great deal of information ex-
ists about singing and vocal repertoire. This book presents the perti-
nent scholarly information about vocal performance practice from the
early Baroque through the twentieth century so that singers can make
informed stylistic choices about such topics as ornamentation, articu-
lation, use of vibrato, working with period instruments, and language
issues, to name just a few common concerns.
Each chapter deals with a specific historical period and introduces
1
2 Introduction
some background information about the singers and composers, the vocal
repertoire, and the stylistic conventions of that period. Specific repertoire ex-
amples will be discussed as well, showing how to use the music itself as a con-
text for making stylistic choices. Information on performance practice is not
just a list of rules. Too many frustrating contradictions and unanswerable ques-
tions remain within all the specific instructions and directions in the historical
treatises and their modern interpretations. Often the music itself, when under-
stood within a general stylistic context, can help us make appropriate choices
about tempo, phrasing, articulation, and ornamentation.
Since I don't include exhaustive treatments of these issues, source lists for
further study are arranged by topic at the end of the book in the "For Further
Reading" section. References include major primary sources such as treatises,
letters and memoirs of composers and performers, and commentary from
other writers and critics of the time. Secondary sources include historical and
analytical writings and books and articles on performance practice. You can
read this book in sequence from beginning to end or look up a particular topic
such as improvisation in the Classical era or tempo and rhythm in German
lieder. If you do read the chapters in order, you will be able to follow evolving
trends relating to notation, ornamentation, articulation, legato, rubato, tempo,
vibrato, portamento, language, and national character, as well as who dictated
the fashions: the singers or the composers. A short conclusion at the end of each
chapter summarizes the important issues introduced for that period.
This book concentrates primarily on songs, sacred music, and chamber mu-
sic rather than on opera. Opera necessitates dealing with issues of vocal type
and casting preferences that are outside the scope of this book. It is important,
however, for singers to take into account their own vocal type, as well as the nat-
ural strengths and weaknesses of their instrument, when choosing repertoire.
In considering the specific works and their historic contexts in the following
chapters, singers should evaluate whether their voices are appropriate for an
early seventeenth-century lute song, an eighteenth-century French cantata for
haute-contre, or Schoenberg's Pierrot lunaire. They must decide whether they
can learn and incorporate the necessary techniques for a given genre or style
into their vocal production. Ideally, a singer would be infinitely flexible and
able to reinvent his or her voice to suit any situation, but that is an unrealistic
expectation. I hope that the information in this book will help singers make
sensible choices about repertoire that is appropriate for their individual instru-
ments and their natural abilities.
I want to add a cautionary word about reading the primary sources. Through-
Introduction 3
out the history of music, the discussion of singers and singing seems to have in-
spired heightened passions and heated debate among both performers and ad-
mirers. But the language we must use to talk about singing—in a voice lesson,
at a rehearsal, or in a concert review—is subjective and imprecise at best. Even
new developments in scientific technology for vocal pedagogy may only com-
plicate the problem of communicating with language about something that has
to do with subtle internal sensations. By the same token, the response a singer
can inspire in a listener is equally ephemeral and must be described with lan-
guage that is metaphoric and poetic. When considering a historic text, which
has been translated into modern English from another language from another
century, we must sift through a vast number of general elements relating to the \
culture and the historic context in order to grasp the sense of the meaning be-
fore we can even attempt to contemplate how it relates specifically to singing.
In other words, it is hard enough to communicate about singers and singing
within our own language and cultural references. The historic documents, even
when interpreted sensitively by modern scholars, can only give us a shadowy
impression of what singers in the past sounded like, or how they sang.
The information in this book is not meant to transform modern singers into
authentic recreations of historic personalities, nor is it meant exclusively for the
"early music singer." Rather it is meant to provide tools for the wide [variety of
situations in which singers may find themselves. The repertoire included in his-
torically informed performances now reaches toward the early twentieth cen-
tury. In recent years both taste in singers and ideas about appropriate voices for
certain repertoires have changed dramatically as well. Even given a specific area
of the repertoire, the choices in performance approach are many: Handel's Mes-
siah or the Bach Passions, for example, can be presented with a full modern or-
chestra, large chorus, and "opera singers" as soloists or, on the other extreme,
with period instruments and a smaller complement of singers who are "early
music specialists," or with any kind of combination of elements in between.
Even within the early music world, the quest for authenticity, which ¡was a pri-
mary goal ten or twenty years ago, has lost some of its intensity today. Both mu-
sicologists and performers are deciding that it is not possible, nor in many cases
desirable, to re-create the exact conditions of a first performance. This then
leaves performers with a wealth of historical information to choose from in or-
der to create modern performances that are vivid, energized, interesting, and
compelling for todays audiences.
Really, the only way to experience truly "authentic" music making is to col-
laborate with a composer on the premiere of a new work. The last chapter of the
4 Introduction
book explores how performance issues are put into stylistic practice today. Even
in our world of computer-based musical scores with specialized notation and
elaborate performance instructions, decisions about an infinite array of nu-
ances involving phrasing, dynamics, diction, and articulation cannot possibly
be fully indicated in a score. In a new work, even decisions about style that
could be notated may be made between the composer and performer in a re-
hearsal or in a conversation on the phone but not make it into the final version
of the score. The singer who performs the premiere or records the work may
add personal, idiosyncratic nuances that then define its style for the future. The
composer may make a suggestion to a performer that clarifies or is even con-
trary to the markings in a published score, or he or she may disapprove of the
personal interpretations a singer adds and prefer the way a different singer han-
dles the music in another performance situation.
All these interactions between composer and performer will be seen repeat-
edly as we consider the issues of historical performance practice. When you ac-
tually experience these types of interactions in the twenty-first century, then it
is easy to appreciate the same kinds of issues when you read about them taking
place in an earlier century. By working on new music you will not only con-
tribute to the growth of musical culture in the present and produce musical
documents that musicologists will scrutinize in the future, but you will also
gain a new understanding of the process of creating a music and a style of the
past, while working toward the timeless goal of realizing the composers inten-
tions.
Chapter 1 The Early Baroque
This chapter presents an overview of seventeenth-century vocal music

and practice in Italy, England, and France. (Germany will be dis-
cussed in chapter 2.) In addition to considering issues of composi-
tional development over the course of the hundred years in question,
it will also review differences in national style. It will explore the im-
precise nature of Baroque notation, the changing relationship of text
and music, pitch in different geographical areas, instrumental accom-
paniment, and the development of figured bass. Difficult questions
without easy answers must also be considered: What did the singers
sound like? What was their approach to technique? What kind of or-
namentation did they use and where, and how did that ornamenta-
tion change over the course of the century?
Much general advice is given in the many treatises from this period,
but these comments were directed at students, teachers, and profes-
sional musicians who were already familiar with the prevailing con-
ventions and styles of the times. Specific situations were not often dis-
cussed, and decisions about tempo, articulation, and ornamentation
were usually left up to the individual performer, who was expected to
5
6 The Early Baroque
know how to apply the general instructions. Today we must do the best we can
to make sense of this information. Yet even then, Bénigne de Bacilly, whose
treatise Remarques curieuses sur Гап de bien chanter (1668) was one of the first
and most important discussions of French vocal style, understood the difficult
position of the student singer:
We are saying, therefore, that there is a general vocal method which can be learned.
But the specifics (which are the application of this method to a particular air; to this
word and to that syllable) are matters of such subtlety that often good taste is the
only guiding rule. For this reason, it is necessary (among other virtues) to refer your-
self to the judgment of a wise person in these matters. Also, it is just as often neces-
sary to refer to the thousand different nuances of the vocal art, and also to those
singers who have had more experience. In this way, the student will learn from those
who have performed longer . . . since it is obvious that singing is not always learned
• through a knowledge of its rules.1
Even though we cannot imitate the singers of the seventeenth century, we can
learn from the varying approaches of the singers of our time who have special-
ized in this repertoire and developed their own understandings of the instruc-
tions from the past. The more time you spend working on specific musical ex-
amples, the more familiar you will become with a particular corner of the
repertoire, and you will find that the music itself holds many answers to ques-
tions of performance practice and style.
Baroque Notation
By our modern standards, Baroque notation may seem quite imprecise. In the
seventeenth century, composers and performers did not expect notation to be
self-sufficient or to convey every detail in a work. Composers did not believe
they had to nótate the music exactly as it should sound or even to include all the
information the performers needed, such as tempo, dynamics, instrumenta-
tion, or consistent rhythms and ornaments. Often composers collaborated
closely with performers or participated in performances themselves. If a com-
poser were involved in the rehearsal process, he could tell the performers ex-
actly what he wanted. If not, performers felt knowledgeable enough about the
stylistic conventions of the time to make their own choices about tempo, dy-
namics, instrumentation, and other variables. All performers, including sing-
ers, were also expected to know harmony and counterpoint in order to add em-
bellishments and alter the rhythms in appropriate places. If a singer were also a
composer, as many were, including Giulio Caccini, Barbara Strozzi, Michel
The Early Baroque 7
Lambert, and Henry Lawes, they didn't need to write these things into the
score. Thankfully, some scores with added ornamentation do survive, and thus
we can derive some ideas about the embellishments then used. Yet specifically
notated ornament indications vary widely from country to country, composer
to composer, and even within a single work. This factor will be discussed more
fully in the subsequent sections on ornamentation.
An early Baroque score offers only the bare bones of the music, needing the
experience and judgment of the performers, combined with their inspiration
and spontaneity, to bring it to life. The same is true of a jazz standard, which in-
vites performers to add improvisation, rhythmic alteration, and melodic em-
bellishment in a variety of tempos and moods. It was only in the late nineteenth
and early twentieth centuries that composers began to insist their music be per-
formed exactly as it was written. Consequently, they began to include more
precise performance instructions in the score. As a result of this relatively recent
practice, we are accustomed to treating any score with reverence in order to
serve the wishes of the composer. Most Baroque scores, however, don't begin to
include all the performance information we need, or to fully indicate what the
composer wanted or expected. We must take the responsibility for making nu-
merous musical decisions and adding our own personal touches.
Scores and Editions
The first thing to do when learning a piece of Baroque music is to evaluate the
score being used. A good place to start is to find a facsimile of an early print or
manuscript. More facsimile editions are being published now, and it is also pos-
sible to gain access to original manuscripts in university libraries, special collec-
tions, or on microfilm. When you look at an unedited score, however, you must
be prepared to do a good deal of mental editing for yourself.
For example, example i.i shows a 1623 print of Arianna's Lament from Clau-
dio Monteverdi's L'Arianna (1608), the only surviving excerpt from this opera.
When looking at this score, you must read the vocal part in soprano clef (mid-
dle С on the bottom line), move the text to fit the syllables with the correct
notes, take into account the old-style print for the letter "s," and occasionally
realign the bass notes with the vocal line. Since there is no key signature, you
must decide how to interpret the accidentals that are provided, taking into ac-
count harmonic and melodic considerations. You must also not become dis-
tracted or misled by various ink splotches. You may notice, however, as you
compare this to the familiar version in G. Schirmer's Twenty-four Italian Songs
and Arias, first published in 1894, that the aria is in D minor instead of F minor
8 The Early Baroque
Example i.i. Monteverdi: "Lasciatemi moriré" from L'Arianna (1608), from a 1623
manuscript printed in Venice, reproduced in 26 Italian Songs and Arias, ed. John Glenn
Patón, Alfred, 1991,
in the medium-high edition or С minor in the medium-low edition, bar lines

appear m different places, and the music varies in a number of crucial spots. In
the opening measure, the bass line stays on an A against the B-flat in the vocal
line to create an anguished dissonance, which has been softened by the change
of harmony in the Schirmer version. During the second "lasciatemi" (m. 2), the
The Early Baroque 9
vocal line rises a whole step instead of two half steps, as in the Schirmer score.
There are also differences in rhythms and text underlay at "in cosi dura sorte, in
cosi gran martire" (mm. 6-7). There are even wrong pitches in the seven-
teenth-century print for the penultimate "moriré" (m. 8). If you look at the last
"lasciatemi" you can also see a confusing notation of both a flat and a sharp
sign, which might explain the chromatic vocal line found in the Schirmer
score.
Working from old manuscripts and prints takes some getting used to, and
you need to be familiar with sixteenth- or seventeenth-century time signatures,
as well as different clefs. Virtually all vocal music from the seventeenth and
early eighteenth centuries is notated using С clefs (including soprano, alto,
tenor, and bass), which either indicate which voice part should sing or contrive
to avoid ledger lines by arranging the music all on the staff. Some scores are less
problematic to read than others, and with experience it becomes easier to work
with older notational practices. Since much seventeenth-century vocal music
does not exist in modern editions, an ability to read earlier musical notation
will give you access to a much broader repertoire. These skills are not difficult to
acquire and can be invaluable tools for finding the true spirit of the music.
Robert Donington presents a helpful discussion in his Baroque Music: Style and
Performance.2
In the nineteenth century, editors who prepared collections of early Baroque
vocal music tried to make the scores easier to read. They also corrected incon-
sistencies and made decisions about tempo, dynamics, phrasing, ornamenta-
tion, instrumentation, and accompaniment. As a result, these scores can con-
tain too much information that, if followed strictly, will produce a distinctly
nineteenth-century interpretation. It then becomes the performer s task to sort
through the notes and markings and decide what is editorial and can be left out
versus what is part of the original score and necessary to keep. A good compro-
mise is to find a modern edition with clear and legible notation accompanied
by scholarly commentary and a minimum of added editorial markings. We are
fortunate that many more such editions of seventeenth-century repertoire are
becoming available.
Thoroughbass
Most solo vocal music of the early Baroque consists of the vocal line with a bass
line accompaniment. Early seventeenth-century songs were often transcrip-
tions of choral works, with the counterpoint of the other vocal parts condensed
into an accompaniment part for lute or other instrument. John Dowland wrote
10 The Early Baroque
lute songs that used a form of notation called tablature to indicate fingerings to
facilitate the contrapuntal accompaniment texture. As the seventeenth century
progressed, however, the contrapuntal nature of the accompaniment was sim-
plified into a single bass line from which chords were to be constructed. This
accompaniment, known as thoroughbass or basso continuo, gave support to
the vocal part while allowing the singer greater flexibility. (For more informa-
tion, see "Chapter i: Accompaniment Instruments and Figured Bass" in "For
Further Reading.") Occasionally a number or a sharp or flat would be written
below or above a bass note to indicate the harmonies. This practice was basi-
cally a shorthand method for telling the instrumentalist what notes to play. In
the preface to his Le nuove musiche (1602), Giulio Caccini gave detailed in-
structions regarding how to execute his "new music": "I have been accustomed,
in all the pieces that have come from my pen, to indicate with numbers over the
bass part the thirds and sixths—major when there is a sharp, minor when a
flat—and likewise when sevenths and other dissonances are to be made in the
inner voices as an accompaniment."3 Later in the century more "figures" were
added to the bass line as the music became harmonically more complex, hence
the term "figured bass."
Thus it was not only the singer who was expected to improvise ornaments
and embellishments. The accompanist had to invent and compose a complete
part from the bass line and whatever figures may have been indicated. Com-
posers had conflicting attitudes about whether to include figures or not. In his
Concern ecclesiastici (1610) Giovanni Piccioni commented that "I have not cho-
sen to put any sort of accidentals, such as sharps, flats, and figures, over the
notes as many do, because to such organists as are not expert, they are a source
of confusion rather than otherwise, while to those who know, and to compe-
tent men, such accidentals are not necessary, since they play them correctly by
ear and by art."4
The skills required to "play correctly by ear and by art" include a thorough
knowledge of harmony and voice leading, as well as a familiarity with the ap-
propriate style of embellishment and ornamentation. One of the most plenti-
ful areas of information on performance practice is the literature on lute play-
ing and thoroughbass realization. The general approach to reading basso
continuo is not that different from reading guitar chord charts printed in scores
of Broadway or pop songs: these are printed above the vocal line or below the
bass line of the piano part. "Gy," for example, specifies the bass note and the
chord that goes with it. The player decides how to get from one chord to an-
other and what rhythm and embellishments to add. Performers of Baroque
The Early Baroque ΊΊ
music must similarly learn how to navigate harmonic progressions. This can be
fairly simple in the early i6oos, but it becomes more complex for later music.
While it is certainly not unusual for modern folk or pop singers to accom-
pany themselves on a keyboard or guitar, singers today who are interested in
early music may not have the ambition to learn how to accompany themselves
on the lute. (It is crucial, however, for singers to be thoroughly familiar with the
bass line of Baroque music and to understand how it interacts with their vocal
line.) One alternative is to find a colleague who has experience reading from a
figured bass. The other possibility is to find a modern edition of the music with
a good realized accompaniment. This may be harder than it sounds, as Howard
Mayer Brown explains in his New Grove Dictionary of Music and Musicians
(1980) article on editing: "If the editor provides a stylistically appropriate real-
ization it may seem idiosyncratic and subjective and inappropriate for certain
performance situations. If he leaves it plain it may be insufficient for less expe-
rienced performers. If he provides a fully fleshed out harmonic accompaniment
it is cumbersome and distracting for experienced performers. If he provides the
bare minimum to indicate the harmony he is called unimaginative arid insensi-
tive to style."5 The problems facing modern performers with regard to realizing
a thoroughbass are thus not that different from those that faced composers and
performers of the seventeenth century. Ultimately, the realization of a figured-
bass accompaniment should suit the particular performers and performance
situation.
Accompaniment Instruments
Another element of choice is the instrument that provides the thoroughbass ac-
companiment. Sometimes a composer would specify a particular instrument in
the score. More often the instrumentation, even in larger ensembles, was not
indicated. Certain instruments were more popular and fashionable at different
times. The success of John Dowland's The First Booke of Songs andAyres, pub-
lished in 1597, inspired many more English publications of similar lute-song
books through the 16205. In France the nobility considered lute playing a re-
quired social skill, and it remained the preferred solo and accompaniment in-
strument well into the 16405. Many song publications, however, suggested a
choice of instrumentation, as in Dowland's 1603 The Third and Last Booke of
Songs or Aires. Newly Composed to Sing to the Lutey Orpharion or Viols. For the
amateur player with less experience at providing a complex chordál accompa-
niment, it may have been easier to play the bass line alone, though it was possi-
ble to play some chords on the viol. The decision of what instrument to use in
a given situation could also have been based on what players and instruments
were available. Roger North, a contemporary of Purcell's who wrote a great deal
about the music of his time, describes accompanying his brother's singing: "and
it being necessary to the sound of a voice, whatever it is, to have an instrument
to accompany, and I being well habituated to the viol and the fingering, I used
to touch the principall notes as well as I could, and by degrees to putt in cords,
and at last to full harmony, as the instrument would afford."6
As thoroughbass began to replace tablature during the early decades of the
century, performers had a wider variety of bass instruments to choose from, ei-
ther plucked or bowed. The most widely preferred instrument for song accom-
paniment was the theorbo, also known as a chitarrone, which was a large lute
with an extension to its pegbox, or a second head, to accommodate more low
bass strings. Caccini recommended singing with a theorbo in Le nuove musiche.
The single instrument, especially when played by the singer, allowed an inti-
mate and flexible accompaniment for spontaneous ornamentation and an ex-
pressive rendering of the text.
In sacred music the organ was used. The character and capabilities of differ-
ent organs existing in the seventeenth century varied widely from region to re-
gion and even from church to church. In the opera house, combinations of in-
struments were often used to provide a richer, more varied, and dramatic
texture. In his LOrfeo (1607) Monteverdi called for three theorbos, harps, two
harpsichords, organ, and regal, a small portable reed organ. At the Paris Opera,
Lully used a grand choeur for the large instrumental sections and a petit choeur
including bass viols, theorbos, and a clavichord to accompany the solo airs. The
harpsichord became more popular for vocal music later in the century, and
singers often used it to accompany themselves. Roger North, however, warns
beginners not to sing with harpsichords or other loud instruments "for they by
imitation will carry off a tincture of the instrument."7 The practice of combin-
ing a sustaining instrument (viol, cello, bassoon, and the like) to double the
bass line with a chordal instrument (lute, theorbo, harpsichord, organ) to pro-
vide the harmony is more commonly seen in later Baroque music. Its presence
in music of the seventeenth century depends upon the particular repertoire and
performance situation.
The modern performer has numerous accompaniment alternatives when
singing early Baroque secular songs: collaboration with an experienced lutenist
or viol player, or finding transcriptions of lute songs for modern guitar. Other
options include working with a harpsichordist who knows how to play from a
thoroughbass, or finding a suitable realized accompaniment. For sacred works
The Early Baroque 13
an organ is certainly appropriate, though you must decide whether to work

with a reproduction of a Baroque instrument or a modern organ and then
make the necessary choices about suitable registration. To sing with a modern
piano is, of course, an option, but one that creates a very different sound and
texture from what the composer expected to hear.
Pitch
The flexible and inconsistent nature of pitch in the early Baroque is an ex-
tremely complex issue that has an enormous impact on singers. To transpose a
work even a half step in either direction can make a tremendous difference in
the technical and expressive qualities of a voice. Yet the difficulties in determin-
ing at exactly what pitch certain music was performed has caused considerable
debate among musicologists, leaving the modern performer with still more
choices to make.
In the early seventeenth century, organs in different churches and towns
were tuned to different pitches depending on their individual construction and
pipe configuration. Sometimes the length of the pipes was affected by the rela-
tive wealth of a church: shorter pipes cost less money and produced a higher
sound. The temperature inside a church and its organ pipes would also affect
the pitch, which could vary as much as a whole tone depending on the weather.
Tuning or repairing an organ could also affect its pitch upward, and over time
some organs became so high that they had to have new low pipes added to read-
just their pitch downward. Organs in different countries varied in pitch as
much as a fourth. As the vocal ranges of singers were more predictable than the
length of the organ pipes, complicated systems of transposition were needed to
accommodate choirs or solo voices who sounded better in lower keys. Lutes
were constructed in all different sizes, with no standard pitch for a given instru-
ment. Wind and brass instruments which survive from the period ppvide still
more clues as to the widely varying pitch standards in different areas. The prob-
lems which arose when instruments tuned at different pitch levels had to play
together necessitated complex transpositions and extra crooks fori wind and
brass instruments. Singers had to adjust to the tuning of accompanying instru-
ments, but were also encouraged to choose songs and transpositions which
would show off their voice comfortably, without strain in the low or high range.
The standard modern pitch of a' = 440 cycles per second was established in
1939 at an international conference in London, but the tradition of using a' =
415 as a standard for Baroque instruments came into practice much more re-
cently, mostly as a convenient solution to the complex problems involved in es-
tablishing historical pitch levels. For the modern singer, the decision to sing
certain repertoire at a lower pitch level may allow one to achieve a different
sound quality or to execute certain ornaments more easily. The particular prob-
lems associated with the French repertoire for haute-contre (high tenor) or the
English repertoire for countertenor, for example, may be seen in a different
light with regard to the pitch levels at which this music was originally sung. By
the same token, pitch levels for low voices must also be evaluated. We are ac-
customed to singing in a much higher range than most seventeenth-century
music demands, so the actual height of the notes at modern pitch is not neces-
sarily a problem. But the technical ramifications of singing in a lower pitch
range and the physical sensations affecting diction, breath support, and general
muscular tension or relaxation can contribute significantly to the overall sound
and style a singer can create.
Vibrato
The use of vibrato in the early Baroque is another controversial issue that has
particular importance for modern singers. Writers of the time had conflicting
opinions and terminology regarding the use of what we would call vocal vi-
brato:
The tremolo, that is, the trembling voice, is the true gate to enter the passages and to
become proficient in the gorgia.. . . The tremolo should be short and beautiful, for if
it is long and forceful, it tires and bores. (Zacconi, едб)8
A singer must have a pleasantly vibrating voice (not, however, as some are trained to
do in schools, but with particular moderation) . . . he must be able to maintain a
steady long tone. (Praetorius, i6i9)9
Fermo or the maintenance of a steady voice, is required on all notes, except where a
trillo or ardireis applied. It is regarded as a refinement mainly because the trémulo is
a defect. . . . Elderly singers feature the trémulo, but not as an artifice. Rather it creeps
in by itself, as they no longer are able to hold their voices steady. (Bernhard, 1649)10
The quality of a singer s cadence is a gift of nature, and yet it can be acquired or at
least corrected and perfected through good training and good exercise. Therefore,
there are many people who have an acceptable voice without having a cadence at all.
Others have it, but it is too slow for certain places where the tremblementought to be
compact and compressed; others have cadences that are too fast or sometimes too
coarse, a quality which is commonly called chevrotante. (Bacilly, 1668)x 1 [Caswell
translates cadence as "vibrato," and chevrotante as "wobbly, tremulous, bleating."]
The greatest elegance of the finest voices is the prolation of a clear plain sound. And
I may add, that in voice or instrument it is the most difficult part to performe. (Roger
North, ca. 1695)12
Let him learn to hold out the notes without a shrillness... or trembling. . . the trou-
ble in holding it out, he will get a habit, and not be able to fix it, and will become sub-
ject to a fluttering in the manner of all those that sing in a very bad taste. (Tosi,
174)i3
Most writers and singers today agree that vibrato is a natural part of healthy
singing. The controversy surrounding its use in Baroque music is usually a
question of degree. It is clear from the above quotes that some vibrato was used
with cautious moderation, probably more as an ornament than as an ongoing
presence. The natural vibrato heard in singers was sometimes admired and im-
itated by instrumentalists: organs included a trembling "vox humana" stop, and
string players were instructed to sweeten a note with a certain trembling caused
by the movement of the fingers on the strings. Women were often described as
having vibrato more often than men, and young singers probably had less vi-
brato than the more mature ones. Yet vibrato should not be confused or inter-
fere with the specific vocal ornaments described by the terms trenzólo, trillo,
trill, and tremblement. Unfortunately the use of these terms by seventeenth-
century writers is inconsistent, causing much disagreement among modern
writers and interpreters. It is important to distinguish between a note that is
affected by an intensity fluctuation, caused by a slight change in air pressure re-
sulting in a narrow vibrato, and a note whose actual pitch is altered by a specific
ornament. (This distinction is discussed in more detail in the sections on orna-
mentation.) Most of the writers cited above specifically warned against pitch
fluctuation, especially on long notes. If vibrato was present in the seventeenth
century, it was probably small and shimmering and did not alter the pitch of a
note to a discernible degree.
In John Dowlands "I saw my Lady weepe" (ex. 1.2), the opening phrase de-
mands several long-held notes, which should be maintained, according to the
advice in the quotes above, with a steady tone. In reviewing the harmony of the
song, we see a number of dissonances between the voice part and the accompa-
niment: the repeated As in the tenor part of the accompaniment against the В
in the voice part in m. 3; the G-sharp and B in the bass and alto parts of the ac-
companiment against the С in the voice in m. 4; the A in the bass against the В
in the voice in m. 5, and later in that measure the B in the alto part of the
accompaniment against the A in the voice; and finally the 4-3 suspension in
Example 1.2. Dowland: "I saw my Lady weepe," mm. 1-8, from The Second Booke of Songs
orAyres (1600), reproduced in An Anthology of Elizabethan Lute Songs, Madrigals, and
Rounds, ed. Noah Greenberg, W. H. Auden, and Chester Kallman, Norton, 1955.
m. 6. Here is a situation where a tone with little or no vibrato can be used as an

expressive device to highlight the delicate anguish of the song by tuning against
the dissonances. Varying the type of vibrato can intensify the emotion of the
music and text.
Vocal vibrato is caused by air pressure flowing through the vocal tract. The
size and speed of the vibrato are determined by the quality of air pressure as it
relates to the degree of tension in the vocal apparatus. In the twentieth century,
singers had to learn to sing louder in order to fill larger concert halls and opera
houses, carrying over larger orchestras and producing more penetrating high
notes. This more forceful, consistent flow of air pressure and more muscular,
athletic approach to tone production have led to the modern "opeiatic" vi-
brato, which can alter the pitch of a note anywhere from a half step to a major
third and can go both above and below the principal tone of the note. In the
case of the Dowland example, a sung tone with a fairly wide and continuous vi-
brato would not be able to inflect the different levels of dissonance against the
harmony. In more florid music of the period, a consistent vibrato of more than
a quarter tone might blur the delicate distinctions between trills, trillos, and
rapid passaggi. Yet regulating vibrato for early music is not just a question of
"straightening" the tone. The modern singer should think about adjusting the
pressure of the airflow that supports the tone, thereby reducing the pressure in
the throat, rather than merely removing the vibrato by tightening the throat.
In the seventeenth century, performance spaces differed greatly from what
we are used to for public music today. While music was composed to fill large
churches and grand palace halls, singers in the seventeenth century often per-
formed in small, intimate settings with instruments that made soft and gentle
sounds. As discussed above, the range and pitch level of most of the music were
quite moderate, and singers could use a more relaxed and flexible flow of air to
support and sustain their tone. They were often cautioned against shouting or
forcing their voices beyond their natural limits. Even the great churches, where
voices were praised for their ability to fill large spaces with sound, would not re-
quire the kind of volume or effort expected of opera singers today. A sweet, soft,
and beautiful sound was prized:
ι
For singing does not arise out of bellowing... for no song is embellished by roaring
and screaming;... the higher a voice rises the quieter and lovelier should the note be
sung. (Finck, 1556)14
One sings in one way in churches and public chapels and in another way in private
rooms. In [the former] one sings in a full voice, but with discretion, nevertheless . . .
and in private rooms one sings with a lower and gentler voice, without any shouting.
(Zarlino,i558)15
1
I heard a beautiful voice, powerful and sustained, and when he sings on the stage he
will make himself heard in every corner very well and without strain. (Monteverdi,
i6io)16
Let him take care, however, that the higher the notes, the more it is necessary to
touch them with softness, to avoid screaming. (Tosi, 1723)17
Monteverdi's sense of "powerful and sustained" was quite different from

that of a modern singer who knows what demands the music of Verdi, Puc-
cini, or Wagner place on the voice. Monteverdi's singers would not have been
able to conceive of the limits to which we have stretched our vocal abilities. Yet
modern singers who are steeped in nineteenth-century repertoire and accus-
tomed to the twentieth-century development of continuous vibrato cannot
help but bring that perspective with them when they approach earlier music.
While current taste in singers for early music seems to have shifted from the
straight, white sound favored ten or twenty years ago to voices with more nat-
ural vibrato, vibrato must be used in a manner appropriate for seventeenth-
century music. Singers must take into consideration the harmony of the mu-
sic, the articulation of ornaments, and the ideal of gentle and unforced tone
production.
Words and Music
Vocal music of the sixteenth century was primarily polyphonic. Its many vocal
lines intertwining in complex counterpoint often made the text difficult to un-
derstand, and in many cases the rhythmic setting of the words was far less im-
portant than the voice leading of the polyphony. Singers also added dazzling
ornamental passaggi to the already complex texture, obscuring the words even
further. By the end of the sixteenth century a reaction against this practice
brought about the development of solo song in Italy, with England, France,
and other parts of Europe following close behind. The new monody had a sim-
ple accompaniment and a solo vocal line whose main objectives were to make
the words intelligible and to convey the sense or "affect" of the text. In addition
to delighting the ear with sweet, soft sounds, a singer's primary goal was to
communicate the passions of the text and to move the affections of the listener:
Music is naught but speech, with rhythm and tone coming after. . . since they could
not move the mind without the words being understood. (Caccini, 1601)18
Let every singer conform his voice to the words, that as much as he can he make the
Concent sad when the words are sad; and merry when they are merry. (Dowland,
1609)19
It would also be necessary to have a single guiding precept; that is, that they should
tend to speak in singing and not, like this one, sing in speaking. (Monteverdi, i6i6)20
One of the great perfections of song consists of good pronunciation of the words,
and rendering them so distinctly that the auditors do not lose a singlé syllable.
(Mersenne, i6^6)21
This drastic reaction against the overly florid singing style of the late Renais-
sance led composers and theorists at the beginning of the seventeenth century
to issue grave warnings to singers lest their embellishments interfere with the
communication of the words and the drama. As the new style of declamatory
song became more established, however, increased ornamentation returned,
and vocal display regained prominence. By the end of the century the impor-
tance of musical and vocal issues had once again overshadowed the significance
of the text.
Historical Pronunciation
In re-creating historically informed performances, historical pronunciation is

not often included, yet there are a number of sources available for the interested
reader. (See "Chapter i: Historical Pronunciation" in "For Further Reading.")
The difference, for example, between todays "BBC" English and Restoration
pronunciation is significant, enough perhaps to make a Restoration text unin-
telligible to modern English-speaking audiences. Depending on the situation,
some modern performers choose intelligibility over authenticity, yet many lev-
els of compromise can be achieved. While Italian pronunciation (nodincluding
the many dialects) is much the same today as it was four hundred years ago, is-
sues related to seventeenth-century French pronunciation can be incorporated
into modern diction and will be discussed further below.22
ITALY
Italy was the home of the Florentine Camerata, whose study and investigation
of the ideals of ancient Greek theater inspired the birth of the "new music" or
stile recitativo of the seventeenth century. The Italian seconda prattica, as the
new style was also called, led the way for the development of solo song through-
out the rest of Europe. Traveling Italian musicians and foreigners who studied
in Italy spread the word and the style, which was widely emulated and imitated.
Italy also had the most singable of all languages and thus a predisposition for
bel canto. Rodolfo Celletti, in A History of Bel Canto, points out that the term it-
self was unknown in the seventeenth and eighteenth centuries. He suggests that
the original goal of bel canto was to create a beautiful fantasy world, with fairy-
tale operatic settings, unusual vocal timbres from high-voiced castrati, and del-
icate, virtuosic vocal display as a sort of antithesis to the commonplace vulgari-
ties of daily life. Today we associate bel canto with beautiful singing in general
and particularly with the famous Italian schools of vocal training that began in
the early Baroque. But the conservatories of Naples and other Italian cities were
not the only places where singers were trained. Many learned from family
members who passed on the traditions from generation to generation. Singers
could be associated with a church or a court, or could merely sing at private
gatherings or for their own enjoyment. Most Italian Baroque singers were com-
plete musicians, skilled on an instrument as well as in composition and theory.
Voice Types and Registers
High voices were the most favored in the early Baroque. While women singers
achieved great fame in opera and concert, castrati became popular in churches
where women were forbidden, and by the end of the seventeenth century they
dominated the stage. Tenors were also favored; Giulio Caccini was himself a
renowned tenor, and Monteverdi composed prominent tenor roles in his op-
eras. Much of the advice in many of the early treatises is aimed at high voices:
Caccini for tenors, and Pier Francesco Tosi, a castrato at the height of his
singing career in the late i6oos, for sopranos. While the bass voice was never as
popular, there is evidence of virtuosic low male singers as well as some castrati
with lower alto or mezzo ranges.
The use of different vocal registers and the importance of properly combin-
ing or blending them are discussed at length in many sources. Earlier vocal mu-
sic had a modest range of not much more than an octave. By the end of the sev-
enteenth century, however, music that now spanned almost two octaves
required singers to expand their ranges and manage the use of different regis-
ters. The chest voice (voce di petto) was referred to as the "natural voice," while
the head voice (voce di testa) was called the "feigned voice" or "falsetto." Al-
though male falsettists were common in church choirs, even in the age of the
castrati, the use of falsetto in male voices inspired many differing reactions.
Caccini preferred the singer to "sing with a full, natural voice, avoiding
falsetto," so as not to have a noticeable break in the sound of the voice or a tense
forced quality in the high range.23 Tosi was in his seventies when he wrote his
Observations on the Florid Song m 1723. He formed his opinions based on the
singing style of the late i6oos and recommended uniting the two registers "that
they may not be distinguished." Tosi goes on to remark that "Among the
Women, one hears sometimes a soprano entirely di petto, but among the male
sex it would be a great rarity."24 Modern singers who are used to blending the
different registers of their voice into one smooth-sounding whole might not
think this particular issue is especially relevant to them. What we must con-
sider, however, are the warnings against screaming and the preference for gen-
tle, soft higher notes. Women might feel free to add a little chest voice to very
low notes, and both men and women might want to blend a little more head
voice into the higher notes. The specific range of a particular work and the
pitch at which you choose to sing it would also affect decisions about this issue.
Ornamentation
Ornamentation in the seventeenth century falls into two general categories:

graces and diminutions. Graces are the smaller ornaments that don't signifi-
cantly alter the contours of the melody but rather adorn it with trills, trillos,
gruppiy esclamazioniy and messe di voce. Diminutions or divisions, on the other
hand, divide larger note values into smaller portions by filling them in with
more rapid notes, often necessitating the recomposition of the melody. Because
a white whole note or half note would thus be transformed into a series of black
eighth, sixteenth, or even thirty-second notes, this was also called "coloring"
the notes, or coloratura^ In the sixteenth century, performing improvised
diminutions was as much a display of contrapuntal composing ability as of vo-
cal technique and agility, and many of the treatises of the period suggested pat-
terns for certain situations. In the seventeenth century, the new importance of
the words caused restrictions to be placed on where and how a singer could im-
provise ornaments, so as not to disturb the communication of the text. The or-
naments were also intended to enhance the text and add heightened expressiv-
ity to a singer's performance.
DISPOSITIONE AND DIVISIONS
The ability of a singer to execute diminutions with clear accuracy and dazzling
speed was often called dispositione di voce. Many seventeenth-century writers
described articulating the notes with the throat, or gorgia, and while this notion
may sound dangerous and unhealthy to modern singers, it is a special tech-
nique that allows great speed and clarity. It is not the same thing as diaphrag-
matic articulation, and it is difficult to achieve with a heavy tone or large vi-
brato. Tosi suggests that it is easier "to make one sing soft than loud," even
though divisions must be sung both piano ana forte.26 Both Zacconi and Mon-
teverdi stress the importance of combining chest and throat for a good disposi-
tione. Modern scholars interpret Monteverdi's recommendations to mean a
legato flow of breath with a well-articulated throat. If the breath is not smooth,
then the throat sounds harsh; if the throat is too smooth, then the notes are not
27
articulated enough.
Singers were criticized if their passaggi were too detached with aspirated
"h's," which sounded like laughing, a howling wolf, or "a little animal bleating
because it has lost its mother."28 They were also warned not to use their tongues
or lips to help with the vigorous articulation of the notes. It was understood
that some singers had a more natural dispositione, while others had to work
harder to achieve the desired agility and clarity by patient practicing: singing
divisions slowly at first and then gradually increasing the speed. Some singers
whose voices were not naturally flexible were told to avoid divisions entirely if
they could not do them well. It was important to keep an evenness of time and
motion (Tosi), or "tempo and measure" (Zacconi), and it was also important to
add passaggi at appropriate times and places, and not too often. "The diminu-
tions would become tiresome when the ear is saturated with them," cautioned
Giovanni Gamillo Maffei,29 while Zacconi observed that "that singer will al-
ways be praised who, with a few ornaments makes them at the right mo-
ment."30 This "right moment" was often at a cadence. In his Compendium
musicesof 1552, Adrian Petit Coclico offered a number of suggestions for orna-
menting cadential sequences (see ex. 1.з).31
Many writers prescribed diminutions only on long syllables, only on certain
vowels (preferably "a" or "o"), only on the penultimate syllable of a cadence, or
only where it would add appropriate intensity to the meaning of the text. Well-
known countertenor Drew Minter has offered some suggestions for cadential
32
ornaments, illustrated in example ι.φ
THE TRILL
Of the smaller type of graces, or accenti, the trill was considered the most im-
portant. In his Le nuove musiche, Caccini describes a trill on one note: in order
to rearticulate the pitch the singer must "re-strike each note with the throat."33
This device was also known as a trillo and sometimes referred to as a tremolo
(which could easily be confused with vibrato, depending upon the context),
and it was used by Monteverdi and other composers of the early seventeenth
century.
Students were instructed to practice slowly and to gradually increase speed
until their trills were fast and cleanly articulated. Scholars disagree about
whether the gradually increasing note values shown by Caccini in example 1.5
34
indicate how to perform the figure or merely how to practice it. It is generally
Example 1.3. Coclico: cadential ornaments from Compendium musices (1552), reproduced in
Carol MacClintock, Readings in the History of Music in Performance, Indiana University
Press, 1979, pp. 32-33.
Example 1.4. Italian cadential ornaments by Drew Minter: final measures of "Disprezzata
regina" from act ι of Monteverdi's L'incoronazione di Poppea.
Example 1.5. Caccini: from Le nuove musiche (1602), reproduced in Source Readings in
Music History: The Baroque Era, ed. Oliver Strunk, Norton, 1950, p. 24.
understood today that the ornament is not strictly measured but flexible and
expressive. It can include a change in speed, either increasing or decreasing, de-
pending on the specific musical and dramatic situation. Julianne Biaird, in a
November 2002 master class at Princeton University, advised students to con-
sider two different types of ornaments on one note depending on the dramatic
context: fast, using the throat, for a lighter, happier situation; and slower, using
the diaphragm, for a more anguished or sobbing effect.
The figure in which two pitches alternate either a whole or a half step was
originally called a gruppo. Later in the century, as the trillo fell out of use, this
figure became known as the shake or trill. It, too, was articulated in the throat
so that the two pitches could be clearly heard. Both the trillo and the gruppo
could be used wherever a grace was needed, at the discretion of the singer. Most
often this would be at cadences, as long as it was not used so frequbntly that
every phrase became tiresomely similar. It is thus important for singers of
Baroque music to be able to make clear distinctions between notes that vibrate
with vibrato and those that are rearticulated either on one pitch (as ä trillo) or
on two pitches (as a gruppo].
L'INTONAZIONZOR INTONATION
Seventeenth-century singers were always expected to sing with true and clear
intonation. Tosí recommends adjusting ones intonation with such subtlety
when singing with stringed instruments that a D-sharp would not be confused
with Ε-flat and sound out of tune.35 "Tuning the voice" had another meaning
as well: Caccini describes starting a phrase on the note a third or a fourth below
the written note, depending on the harmony, touching it very lightly and then
sliding gently to the intended note. He does not recommend this grace for be-
ginners, nor does he recommend lingering on the approach note.36 Giovanni
Bovicelli called this ornament cercar della nota (searching for the note), and he
instructed singers that "to lend grace to the voice. . . you begin a third or fourth
below [the given note] depending on the harmony of the other parts . . . [and]
the longer you hold the first note, the shorter the second, the more grace will
the voice gain."37
ESCLAMAZIONEAND MESSA DIVOCE
An alternative to cercar della nota for starting a phrase on a long note is to begin
softly and then increase the volume in an esclamazione, which Caccini thought
was "the proper way to put forth the voice with grace."38 Another possibility is
to "begin singing with a decrescendo, then on to an esclamazione, which is the
most basic means of moving the affect [and] is really nothing but a certain
strengthening of the relaxed voice."39 A messa di voce, or "increasing and abat-
ing of the voice" as Caccini calls it (crescendo and decrescendo as we know it),
could also be used as an expressive device on a long note.
SPREZZATURA OR RUBATO
Caccini explains that, in order to successfully convey the passions of the text in
his "new music," a singer must have a certain "noble neglect of the song" or
sprezzatura, which we would understand as rubato. To achieve the "speaking in
singing" effect of the stile recitativo, the singer is allowed both to bend the no-
tated rhythms to fit the natural rhythm of the words and to shape the tempo to
fit the texts mood. Tosi describes this practice as "stealing the time" and rec-
ommends it for use with a single instrument in music of a pathetic or tender na-
ture. He instructs that the bass must continue in an exact time, while the voice
can delay or anticipate for the sake of expression. In lively or joyful works, how-
ever, the tempo must be kept even and steady, particularly to show offpassaggi
and dispositione.
The stile recitativo of Monteverdi s Due lettere amorose, for example, demands
a flexibility in rhythm and tempo in order to communicate the variety of
moods and emotions in the text while avoiding monotony. In example 1.6, I
might decide to sing the first phrase in m. i a little more slowly and thought-
fully, then increase speed during m. 2 and aim for a strong arrival on the first
half note of "viva" in the middle of m. 3.1 could also choose to start a little faster
and then linger and elongate "tenebre mi viva" to highlight the shadowy dark-
ness suggested by the text. I would treat all the repeated notes in m. 6 with a lot
of flexibility, giving more emphasis and length to the strong syllables on
"cofcmto," "таш'ге," and "sconso/^to" while moving more quickly and lightly
Example 1.6. Monteverdi: Έ pur destina," mm. 1-6, from Due lettere amorose (ybnice,
1623), facsimile edition, Fondazione Centro Studi Rinascimento Musicale, 1973i
over the less important syllables. A wide range of deliveries are possible, de-
pending on how you choose to interpret the text and convey the drama.
Monteverdi also composed highly ornamented vocal music. In the 1609
score for the opera L'Orfeo, the aria "Possente spirito" shows two vocal lines, one
plain and one highly ornamented. Perhaps the singer was allowed to choose
whether he wanted to leave the aria as written, improvise his own ornaments, or
use Monteverdi's. In the motet "Laúdate Dominum," Monteverdi writes out
trillos &nd diminutions for some of the cadential phrases (ex. 1.7). Exactly how
much additional ornamentation singers used is something we can only guess at,
but we can explore composers' preferences by examining scores with their own
added ornamentation.
In both Le nuove musiche of 1602 and Nuove musiche e nuova maniera di
scriverle of 1614, Caccini indicates that in order to avoid the tasteless excesses
practiced by some singers he has written out most of what should be sung in the
score. He includes trillos, gruppos, some cercar delle note, and diminutions. He
instructs singers to add esdamazioni, messe di voce, sprezzature, and some deco-
ration of final cadences, and provides examples of each in his preface (ex. 1.8).
One of the most famous songs to come from Caccini's collections is "Amar-
illi, mía bella" which was known all over Europe and continued to be popular
Example 1.7. Monteverdi: "Laúdate Dominum," mm. 68-75, from Tutte le opere, vol. 15,
1940.
well into the nineteenth century, when it was included in a collection of Arie
antichepublished by Ricordi in 1885. That version, which is significantly altered
in the Romantic style, is the one printed in G. Schirmer s Twenty-four Italian
Songs and Arias. If you examine the original score printed in Hitchcock's A-R
edition (see note 3) and consider using a gentle, soft tone, adding rubato, in-
corporating some esclamazioni or messe di voce on the long notes, perhaps a
trillo or gruppo at a cadence, and understand the diminutions in the final mea-
Example 1.8. Caccini: from Le nuove musiche (1602), reproduced in Source Readings in
Music History: The Baroque Era, ed. Oliver Strunk, Norton, 1950, pp. 26—27.
Example 1.8. Continued

Example 1.8. Continued

sures as a notated cadential flourish, you may gain a new appreciation of the
song in the context of seventeenth-century Italian style. To learn how to pre-
pare your own cadential ornaments, start by becoming familiar with the exam-
ples given in this section and then experiment with your own variations. As you
find more examples in other repertoire from the period, you will begin to un-
derstand what kinds of contours and rhythms are typical for early seventeenth-
century Italian music.
RECITATIVE AND ARIA
By the middle of the century, the ornate stile recitativo had become less fashion-
able. In operas, oratorios, and cantatas, Italian composers began to make a clear
distinction between two basic modes of expression: recitative and aria. Recita-
tive involved setting the text in a way that imitates the rhythm, inflection, and
syntax of the words. Recitatives may be secco or dry with only continuo accom-
paniment and should be speechlike. Accompanied or accompagnato recitatives,
in which the orchestra participates, were often used for particularly dramatic
moments. An arioso is a lyrical version of a recitative; the words are still a deci-
sive factor, but there will often be a more ornate and expressive melody. In an
aria musical considerations take precedence. Italian arias in the middle of the
seventeenth century were usually quite brief, but as the century went on they
became longer and more elaborate, with more extensive instrumental accom-
paniments. By the third quarter of the century, the majority of anas were writ-
ten in da capo form (literally "from the head"): an ABA form in which the
repeat of the A section was usually ornamented. G. Schirmer's Twenty-four Ital-
ian Songs and Arias includes a number of examples of seventeenth-century arias
by Alessandro Scarlatti, Giacomo Carissimi, Alessandro Stradella, and others,
but there are hundreds of others from operas and cantatas that are just as re-
warding to sing and fascinating to study.
ENGLAND
If the music of Italy was influenced by the singable language and passionate
emotions of its people, the music of England was in turn shaped by its strong
theatrical tradition, magnificent poetry' and rhetorical approach to language.
The early seventeenth century was still the time of Shakespeare, when the text
of a lute song was often of higher quality than the music. Decisions about in-
flecting the grammar of a text were just as important if not more so than vocal
display or ornamentation. Few treatises about singing in England survive, but
much historical information exists about English oration, gesture, and acting;
Robert Toft s Tune ThyMusicke to Thy Hart, in particular, offers much in-depth
40
discussion and extensive references. Musical performances in England saw
many changes throughout the seventeenth century: from the country houses of
great families to the court entertainments of James I, to the period of civil war
when all theaters were closed and church music was discontinued, to the
Restoration of the i66os, which brought an even greater flourishing pf theatri-
cal diversions. The end of the century was dominated by Henry Purcell, who
was responsible for many significant contributions and innovations in vocal
music—both sacred and secular, solo and choral, orchestral and theatrical.
English music was influenced by both Italian and French traditions. Italian
monody was available in England as early as 1610. Caccinis "Amarilli" of 1601
was included in Robert Dowland's A Musical Banquet of 1610 as well as in six
other sources, some in highly ornamented versions. By the i6ios the polyph-
ony of John Dowland's lute songs seemed old-fashioned in comparison with
the newer declamatory songs based on the Italian stile recitativo. Figured bass
came into use in the 16305, and Playford included a translation of Gaccini's Le
nuove musiche preface (entitled "A Brief Discourse of the Italian Manner of
Singing") in his 1653 edition of An Introduction to the Skill of Musm Playfords
Introductionvra$ reprinted many times through to the end of the century, indi-
cating the prominence it held in vocal approach and instruction. The influence
of French dance was also felt during the Restoration period as the popularity of
triple-time dances found their way into vocal music. The rhythmic and har-
monic organization of these dance forms provided a musical rather than textual
foundation for English song settings. Purcell defined the height ι of English
rhetorical style by combining both Italian and French influences in his vocal
music. He set longer, more complex texts by juxtaposing contrasting sections of
ornamented recitative with lively duple- and triple-meter dances.
Voice Types and Ranges
Before the Restoration, women were not a part of professional performances in

England. They certainly could have sung for friends or family as part of infor-
mal or amateur gatherings, but the treble parts sung in churches and theaters
were performed by boys, as were the female spoken roles in plays. Women were
allowed to perform professionally after the theaters reopened in 1660, and Pur-
cell had a number of extremely capable women (and girls as young as twelve or
thirteen) singing for him. Earlier in the century, however, there would not have
been the kind of trained virtuoso female performers known in Italy. The men
and boys, on the other hand, would have been exceptionally well trained, as
Purcell was himself from the age of seven in the choir of the Chapel Royal.
Choral parts were typically indicated as treble, countertenor, tenor, and bass,
and it is possible that in some situations the soloists' parts would have been des-
ignated similarly. Yet this was not always the case, and determining the ranges
and possible voice types for specific English choral and solo vocal music of this
period is a complicated business. For a thorough discussion of this issue, see the
chapters by Timothy Morris and by Olive Baldwin and Thelma Wilson in Per-
forming the Music of Henry Purcell.^ *
The particular question of what kind of male voice sang the countertenor
parts has caused much speculation and controversy among musicologists and
performers. Taking the lower pitch of the day into account, many of the "alto"
or countertenor parts would have been too low for what we call a countertenor
today. Extending down to G or even E below middle С would require some use
of chest voice for the lowest notes.42 Some scholars suggest that Purcell wrote
for two different types of countertenors with lower and higher ranges, while
other authorities have settled on the idea that the seventeenth-century English
countertenors were high light tenors singing with their natural voices rather
than falsettists. Yet Peter Giles, in his detailed and informative The History and
Technique of the Counter-Tenon argues that this voice type did use head voice or
falsetto, either consistently throughout the entire range or blended into the
high notes, depending on the music and the particular voice.
Lower pitch also influenced other voice parts. Tenor parts could have been
sung by what we today would call baritones, and the bass voice, which enjoyed
much more popularity than in Italy or France (particularly in the music of Pur-
cell), would have sung the bass parts. In an article about Purcell s sopranos,
Olive Baldwin and Thelma Wilson recommended that sopranos today sing the
songs of Purcell at a lower pitch level in order to achieve "clarity and ease."43
Other circumstances may have shaped vocal assignments. Songs for the the-
ater were often written for a particular performer, but songs were frequently
lifted from one context and freely adapted for another play or performer, a typ-
ical and important Baroque tradition. Sometimes the singers in a play also
acted and the actors sang, and it is worth noting that while some theater singers
were also trained members of the court or chapel choir, many were not and
would have had more modest vocal ability. It is interesting to imagine how our
theater singers of today—Broadway performers or actors who sing, as opposed
to singers who act—might approach Purcell's theater music.
The Earjly Baroque 35
Ornamentation
English ornaments, like the Italian ones, are divided into two basic categories:
small graces and more elaborate divisions. A number of English sources show
ornamentation tables, including Playford s Introduction, Thomas Mace s Mu-
sicks Monument (1676), and Christopher Simpsons The Division-Viol (1665).
In The Performance of English Song, Edward Huws Jones suggests that because
there was such close interaction between singer and lutenist or viol player, and
therefore some imitation of ornamentation was bound to have taken ¡place, it is
helpful to consult these instrumental ornamentation tables when considering
vocal graces.44 (Jones also includes a detailed discussion of accompaniment in-
struments and thoroughbass.) We are also fortunate to be able to study a num-
ber of manuscripts of highly embellished songs from the first decades of the
seventeenth century. These provide a glimpse into the use of graces and divi-
sions in early English song. (These manuscripts are discussed in detail in Jones,
and some are printed clearly in Toft, pp. 94-100.)
GRACES
Sifting through the conflicting names of graces can be confusing and frustrat-
ing, yet it is more important to be familiar with the musical gestures and how
to use them than with what they are called. Many important graces taken from
Caccini are reprinted in Playford (ex. 1.9). English sources rename the Italian
graces with such colorful names as "beat," "backfall," "elevation," "cadent,"
and "springer." These can also be turned into varieties of "shaked graces." Cac-
cini s trillo is called a "trill or plain shake," while the gruff о is referred to as a
"gruppo or double relish." But remember, don't be intimidated by the names or
the ornament tables; get to know the gestures so you can apply them.
Sometimes graces were written out in the music, and sometimes they were
indicated with signs. The signs, which appeared in manuscripts only occasion-
ally, were inconsistently used and could be misleading as to the relative impor-
tance of the figure. The graces must be lighter and more delicate than the prin-
cipal notes of the melody, yet inflected to highlight the harmony and the text.
Roger North wrote about the difficulty of notating graces: "It is the hardest task
that can be to pen the manner of artificiall Gracing an upper part. It hath bin
attempted, and in print, but with woeful effect. . . . The spirit ofthat art is in-
communicable by writing, therefore it is almost Inexcusable to attempt it."45
An anecdote involving Purcell, who was known for writing out much of his
Sources: Manchester. Public Library, Watson Collection, ms 832 Vu 51 [ca 1660]

Simpson, Christopher. The Division-Viol. London 1665
Playford, John. An Introduction to the Skill of Musick. London 1674
Mace, Thomas. Musick's Monument. London 1676
Example 1.9. Graces from later seventeenth-century English sources, reproduced by

permission of the author from Robert Toft, Tune ThyMusicke to Thy Hart: The Art of
Eloquent Singing in England, 1597-1622, University of Toronto Press, 1993, p. 102.
desired ornamentation, is described in the memoirs of an actor, Colly Gibber:

in discussing the performance of Jemmy Bowen, a young singer vyho often
worked with him, Purcell is quoted as having said, Ό let him alorie, he will
grace it more naturally than you or I can teach him."46 This remark is difficult
to interpret. Drew Minter suggests that Jemmy Bowen's fellow singers may
have complained that he wasn't adding enough extra ornamentationl and Pur-
cell defended the boys simpler approach. Baldwin and Wilson caution that
"more naturally" does not necessarily mean more simply. Michael Burden be-
lieves that Bowen most likely did add extra graces, and that Purcell's comment
"meant that he would naturally (i.e., with more innate musicality) ornament
the piece. I don't think it implies anything about the ornamentation one way or
the other."47
For modern singers who want to grapple with this difficult situation, it is
best to recognize what type of a figure can be considered a grace, so that even if
it is written into the music it can be performed as an ornament. The same kinds
of figures can then be added to other passages that seem in need of further dec-
oration. This is particularly helpful in Purcell, who, as Roger North ¿aid, "hath
given us patternes of all the graces musick can have."48
In the opening phrase of "The Fatal Hour" (ex. i.ioa) Purcell has incorpo-
rated variations on the elevation, backfall, double backfall, and shaked backfall
into his composed melody. Example i.iob shows a skeleton melody onto which
the ornaments have been layered. If you compare Purcell's melody with Toft's
list of graces, you can see which gestures and pitches are ornamental and which
are part of the more fundamental structure. With these distinctions in mind, I
would probably give more importance to the first G in m. i, move lightly
through the sixteenth notes that follow, and make a strong arrival on the first F
in m. 2. Similarly, the first F in m. 2, which is really an appoggiatura to the E,
would be stronger than the sixteenth notes and resolution that follow. Consid-
ering the melody in terms of the graces it contains will certainly influence how
much weight or importance you choose to give to each sixteenth note and how
you shape the phrase overall.
DIVISIONS
Divisions were not often written into published song books, except in a few
songs of the Ayres (1609) of Alfonse Ferrabosco. These brief examples and the
divisions seen in manuscript scores indicate that a florid Italian style of embel-
lishment was occasionally used in early English song. The florid passages in
mm. 5, ίο, and 15, of example i.n, for instance, are not notated in a precise
Example i.io. Purcell: a) "The fatal hour," mm. 1-7, from The Works of Henry Purcell, vol.
25, Novello; b) "The fatal hour," mm. 1-5, skeleton melody.
Example ι.π. Ferrabosco: "Why stayes the bridegroome," manuscript source with ornaments, reproduced by permission of the author from Robert
Toft, Tune ThyMusicketo Thy Hart: The Art of* Eloquent Singing in England, 1597-1622, University of Toronto Press, 1993, p. 97.
Example 1.12. Purcell: "Fly swift, ye hours," mm. i—6, from The Works of Henry Purcell, vol.
25, Novelle.
mathematical relationship to the original rhythm. This seems to indicate a flex-

ible tempo and the use of rubato in order to accommodate all the notes, prac-
tices in keeping with the declamatory style popular in the early decades of the
century. This manuscript shows examples of longer divisions as well as smaller
graces in the first measure alone. In Purcell s music, divisions are precisely writ-
ten out and governed by the flow of the musical form, as illustrated in "Fly
Swift, Ye Hours" (ex. 1.12).
Both Toft and Jones have concluded that the historical evidence indicates a
wide variety of valid approaches to performing seventeenth-century English
song. Some sources show elaborate ornamentation, while others suggest more
modest added graces or even none at all. A book of music owned by one Eliza-
beth Seger in 1692 contains a number of Purcell songs with extensive added em-
bellishments. Baldwin and Wilson suggest that these could either be examples
of the bad taste of an amateur singer or her teacher, or inducements to singers

to start out with too much ornamentation, which is easier to correct than too
49
little. Decisions regarding pitch and instrumentation in the earlier repertoire
can also be made based on the performance circumstances and personal prefer-
ences and abilities of the performers. Jones offers a fascinating comparison of
several manuscript versions of John Wilsons song "Take, О take those lips
away."50 The contrast between the ways in which professional and amateur
singers of the seventeenth century might have performed this song can show
modern singers some of the many options open to them.
FRANCE
Grace and beauty are two of the most important defining characteristics of
French Baroque music, which strove to bring to life the French language and the
dance. Louis XIV, a dance enthusiast, set a tone for the entertainments at his
court, which provided ample opportunities for dancing by the courtiers and the
king himself. The vocal music, performed as part of the ballet de court (conn bal-
let), was often meant to accompany dance as well. In his French Baroque Music,
James Anthony notes that there was no Monteverdi in early seventeenth-century
France to develop the dramatic possibilities of these lavish entertainments, bal-
lets mélodramatiques, into operatic forms. It was not until Jean-Bap tiste Lully
served the king as court composer and master of the musique de la chambre m the
16505 and '6os that vocal music in a theatrical context began to reach its full po-
tential. The French theater also had a strong influence on the developing vocal
music, and composers took special pains to preserve the intrinsic rhythms and
stresses of the French language. Theatrical declamation at that time was highly
stylized and quite different from modern spoken French. Bénigne de Bacilly de-
votes the entire third section of his Remarques to a discussion of word stress,
vowel length, and the distinction between long and short syllables. He also com-
pares the French and Italian languages and concludes that "the Italian language
permits more freedom than the French, whose strictness (which is perhaps ex-
cessive) tends to hold composers in check." The flexibility of musical treatment
in Italian vocal music is particularly due to the elision of syllables and to word
repetition, which were not allowed in French.51
As Italian singers came to perform in France, the distinct separation and
even rivalry of the two styles were noticed and discussed by numerous writers:
As to the Italians, in their recitatives they observe many things of which ours are de-
prived . . . the passions and affections of the soul and spirit, as for example anger,
furor, disdain, rage, the frailties of the heart, and many other passions with a violence
so strange that one would almost say that they are touched by the same emotions
they are representing in the song; whereas our French are content to tickle the ear,
and have a perpetual sweetness in their songs. (Mersenne, 1636)52
Moreover, in vocal settings of Latin and Italian, all kinds of words are utilized with-
out resulting in outcry from the ranks of the critics. Lightning, Thunder, Stars, Pur-
gatory, Hell. . . and also a large number of expressions which might seem quite odd
in French. . . . The very use of such expressions would be considered barbarous in
French airs, which can accept only sweet, flowing terms and familiar expressions; for
an air to be of good quality it is not enough that the music be beautiful: it is also nec-
essary that the words be beautiful,. . . and above all that they are not shocking or re-
pulsive. (Bacilly, 1668)53
Anthony has compiled a list of nouns and adjectives from numerous French
sources describing the differences between French music, which was consid-
ered the embodiment of le bongoüt (good taste), and Italian, which was viewed
as la corruption dugout. The Italian list includes bizarre, brilliant, colere, extrav-
agance, rage, violence, and vivacité. The French list includes beauté, calme,
charme, delicate, douceur, elegante, grace, noble, tendresse, touchant. Modern per-
formers approaching this repertoire should keep in mind how the French saw
themselves and their music during this period.
Just as the French were not comfortable with the excessive emotional expressiv-
ity of Italian music, they were also uneasy with the castrato voice and the sexual
ambiguity, so popular in Italian opera and culture of the time, that it personi-
fied. With the exception of imported Italian music performed in Paris, castrad
were rarely used in France. Some women sang at court, and groups of "pages,"
or boys, were also kept to sing the dessu parts for the musique de la chapelle and
the musique de la chambre. But the favored voice for the leading roles was the
haute-contre, the specific nature of which has provoked considerable debate.
Most authorities understand this voice, thought to be similar to the English
countertenor, to be a high natural tenor with a range from d to b'. Given the
probable low pitch standard in France, it is possible that falsetto was not used,
or perhaps it was only used for the very highest notes or for a special sweet and
touching effect. Bacilly defends tenors against the abuse and inappropriate re-
marks commonly piled on them for using falsetto, because "it is soon realized
that the vocal art owes everything to this high falsetto voice, because of the fact
that it can render certain ports de voixy intervals, and other vocal decorations in
a fashion entirely different from that of the normal tenor voice."54
In choral music the haute-contre sang the alto part in an arrangement that
was similar to that of the English choir. Voice parts were designated as dessus
(sopranos: women, boys, falsettists, or castrad), haute-contres (altos: men),
failles (tenors), basses-tailies (baritones), and basses-contres (basses).55 The lower
tenor voice was neglected in this period, and the solo bass voice was reserved for
special circumstances according to Bacilly: "The Bass voice is suitable for al-
most nothing but the emotion of anger, which appears rarely in French airs. As
a result, this voice range must be content with partsinging and . . . singing the
bass part rather than the melody."56 Low male voices were also conveniently
used to sing operatic demons or villains.
Since most of the significant music making was at the court of Versailles, it
followed that most of the performers, both musicians and dancers, were mem-
bers of the court. The musicians were most likely professional, but professional
dancers, particularly women, were not used until well into the middle of the
century. Solo singers were known as acteurs and actrices pour les rolles. Though
the system of training could not compare to that in Italy, and some singers went
to Italy to study, Bacilly, who was a respected voice teacher himself, devotes
considerable space in his Remarques to a discussion of choosing a suitable in-
structor. From his comments, it is apparent that many masters professed to
teach singing, but not all possessed the qualifications he thought they should.
Whereas some of the writing about singers and singing from the seventeenth
century is difficult for us to relate to, much of Bacilly s discussion—particularly
of types of voices—is extremely clear and remarkably valid for singers today.
He cautions big voices not to try to reduce their total volume and thereby com-
promise vocal quality; rather they must "relax and never force the voice so that
it will have the vocal quality it ought to have."57 From his discussion of voices
it is obvious that though smaller, higher, sweeter voices were preferred, there
were as many different-sounding singers then as there are today. i
Dance Forms
One of the important keys to tempo and phrasing in French Baroque music is
a familiarity with the particular dances which served as the basis of the musical
forms. Dances from the period include familiar names such as gavotte, minuet,
sarabande, and gigue, as well as lesser-known forms including the loure, bour-
reé, canarie, passacaille, chaconne, and passepied. Each has its own distinct
character, meter, and tempo, which are described in depth by Anthony.58
These dances were central to the large dramatic and operatic productions. At
the same time, many of the smaller vocal forms such as фе air de cour and
motet were also based on dances, which were an inseparable part of French mu-
sic. If you have a chance to take a Baroque dance class you can learn firsthand
how these dances feel in the body. If you have a sense of how to dance to a cer-
tain piece of music, it will tell you a lot about how you ought to sing it.
Notes inegales
Another particular characteristic of French Baroque music is inégalité, a tradi-

tion of performing music in a rhythmically uneven fashion even though it may
have been written evenly. As our modern jazz players know, in certain situa-
tions a series of eighth notes might need to "swing" instead of being played with
metronomic evenness. The rules for inégalitéztz complicated, but basically the
performer alters the rhythm of the written notes to make them anywhere from
slightly to very unequal. This can range from turning a pair of eighth notes into
a softly lilting triplet all the way to double dotting. It is usuajly applied to notes
in stepwise motion rather than skips; it is used on the fastest-moving predomi-
nant note value in the work (often eighth notes, but sometimes sixteenths or
quarter notes); and it should add an appropriate gracefulness to the music and
avoid choppy or jerky motion. In his article on notes inegales in the New Grove
dictionary, David Fuller suggests that the degree of inequality should be deter-
mined by the character of the situation, the expressive needs of the work, and
the taste of the performer.59 For vocal music, Bacilly cautions that "given two
notes of equal length, one of them is interpreted as being dotted while the other
is not. However, the student will also notice that this indication is never indi-
cated in the printed notation and for good reason. If this interpretation were
written out in dotted rhythms, the probable result would be that the singer
would perform them in the jerky or jumping style typical of the old 'Gigue/
This style of vocal interpretation is no longer acceptable; therefore, it is neces-
sary to interpret this dotted rhythm as delicately and subtly as possible so that
it doesn't seem overdone."60
Sometimes composers indicate that they do not want inequality by writing
"notes equal" or "tempo equal" in the score. Staccato markings, or instructions
for marqué, détaché, or mouvement decide, also indicate that inequality is not de-
sired. It is required, however, when the score indicates note púntate, notes ine-
gales, pointé, or couler. Piqué, which is also used to describe a ballet step, can
mean either sharply dotted, or staccato and not dotted. Many questions and
choices thus face the performer, and, not surprisingly, the issue is filled with
controversy. Much has been written about the general question of rhythmic al-
teration in Baroque music if you wish to investigate it further. (See "Chapter i:
General, and French" in "For Further Reading.")
Ornamentation
Most of the following discussions of ornamentation and diction are taken from
Bacilly. His lengthy and detailed presentation in the Remarques has been trans-
lated into accessible English by Austin Caswell and is highly worth reading.
DISPOSITION
Like the Italians, the French described the ability of a voice to move with speed
and agility, and to execute subtle trills and appoggiaturas with delicate finesse,
as disposition. The articulation is described as coming from the throat, and a
lighter tone and steady breath are recommended for ease of movement. Yet
Bacilly recognizes that some voices will have a natural ability to execute one
kind of agility while other voices will be naturally good at another kind of
movement. Patient practice, in the morning, was considered the best way to
improve disposition.61
AGRÉMENTS
Agréments, the small delicate ornaments that the English called graces, were
used in the French air de cour. These solo songs, usually accompanied by a sin-
gle stringed instrument, were composed of two verses. The first was a simple
melody, and the second was more ornately composed. While certain agréments
were needed in the first verse, it was the second verse, commonly called the
double, which the singer was expected to embellish with more elaborate orna-
ments. Bacilly says it is permissible to vary the tempo of a song in order to add
ornaments that bring greater refinement and charm to the performance, even if
it is a dance piece: "It is completely unfair to criticize this style of performing by
saying that the airs aren't danceable, as thousands of ignoramuses have done. If
this were to be the intention of the performing singer, then his function would
be no more than that of a viol."62
On the nature of notating ornaments, Bacilly expresses an opinion similar to
that of Roger North's: "The majority of these ornaments are never printed in
the music, either because they cannot accurately be reduced to print because of
a lack of appropriate musical symbols, or because it may be thought that a su-
perabundance of markings might hinder and obscure the clarity of an air and
thus result in musical confusion."63 He also notes that the following agréments
are almost always used to ornament long syllables, not short.
PORTDEVOIX
This is not to be confused with portare la voce or portamento, whose seven-

teenth-century Italian meaning, "to carry the voice," is slightly different from
what we understand it to mean today. The port de voix (ex. 1.13) is similar to an
appoggiatura from a half or whole step below, comparable to Playford s beat,
and can be used at cadences and half cadences. In the French context the
Example 1.13. Bacilly: port de voix, from Remarques curieuses sur Гап de bien chanter (1668),
ed. and trans. Austin В. Caswell, Institute of Mediaeval Music, 1968, pp. 66-67.
rhythm is altered slightly so as to delay the arrival on the upper note¿ which is
then repeated gently in a coup de gosier, or slight glottal rearticulation. In a
demi-port de voix the long upper note is not rearticulated. Port de voíx can also
be applied to rising thirds and fourths at cadences, but it is slightly more com-
plicated to determine when those are appropriate.
TREMBLEMENT
This term can be understood to be comparable to the Italian tremolo, which, as

previously mentioned, had several different meanings. Caswell has chosen to
translate it as "vibrato" in some situations. The trembling quality is also some-
times referred to as flexion de voix, which adds brilliance to singing and could
also be understood as vibrato. The other meaning of the term tremblement,
sometimes called cadence, is a cadential ornament like a shake or trill ι that starts
on the note above the written pitch and can be used in a descending melody
(ex. 1.14). It is like a mirror image of the port de voixzxia can be used as an alter-
native to that ornament. Bacilly states that the tremblement is "one o|f the vocal
arts most important decorations," which must be performed quickly, deli-
cately, and in the throat. The preparatory note before it is the аррг^у, and the
note leading gently from the tremblement to the final note is called the liaison.
Two other terms closely associated with tremblement, which may also indi-
cate some kind of ornamental vibrato, are étouffézna doublement du gosier. The
doublement du gosier Is described as a repetition of one note in the throat, simi-
lar to the trillo. It is called animerwhen used on the viol. The étouffé is described
as follows: "after having sung the appuy,. . . the voice starts to trembler but ac-
tually only seems to as if only wanting to repeat the note on which it would or-
dinarily make a complete cadence."64
ACCENTOR ASPIRATION
The accent is a tone touched very lightly with the throat, as if an appoggiatura,
leading either up or down, were phrased off to a gentle staccato. Sometimes
called bplainte, it is used where the meaning of the text warrants it. Depending
on the situation, a long note can be shortened using an accent, but Bacilly warns
not to use it too often on an air s final notes, where a gentle surge and ebb, com-
parable to a messa di voce, would be more graceful.
DIVISIONS
While all the above agréments could be added to the first verse of an air de cour
or similar vocal piece, more extended and complex diminutions (which were
Example 1.14. Bacilly: tremblements, from Remarques curieuses sur Гап de bien chanter
(1668), ed. and trans. Austin B. Caswell, Institute of Mediaeval Music, 1968, pp. 69, 84.
roundly criticized and derided by some) were allowed in the double, or second,
more ornate verse. The challenge was to add notes and readjust the new text of
the second verse to the melody of the first verse without disturbing any of the
requirements for long and short syllables. As was the case throughout the
Baroque, some singers wanted to show off vocally while their critics wanted to
preserve the intelligibility of the text. Bacilly generously encourages, diminu-

tions, but within the bounds of good taste and respect for the language. (See
H. C. Wolf's edition of Original Vocal Improvisations for ornamented versions
of airs by Bacilly.)65 Lully is said to have disapproved of improvised diminu-
tions in general, and he severely limited ornamentation, particularly! in recita-
tives.
Diction
Bacilly stresses the importance of clear, precise diction, but he warns that it is
not enough for the words to be easily understood as in familiar speech. In
singing, as in public speaking or the theater, proper declamation gives more
weight and gravity to the words by infusing them with appropriate force and
energy. Enunciating all the necessary syllables, plural endings, and articles, em-
phasizing expressive consonants, and highlighting the difference between long
and short syllables are all elements of declamation for singing. Yet these tech-
niques must not be confused with expressive singing in general and must be
used in combination with overall dramatic expression.
VOWELS
Bacilly cautions seventeenth-century French singers against some of the same
mistakes that modern singers make in singing in French. The mute "e" must be
pronounced, in many situations where it would not be in spoken language in
order to contribute to the rhythm and rhyme scheme of the poetry. He suggests
a closed, dark vowel, such as "eu" and warns against an open sound of "a" or
"eh." He advises singers to be careful when singing the nasal vowels "an," "en,"
and "on" not to let the "n" sound come in too soon, and he remarks ¡that this is
the only situation when singing through the nose is acceptable. He also cau-
tions singers when considering the two possible pronunciations of "oi," which
can sound like "ai" in some situations. Today we would understand reconnoisto
be an old-fashioned spelling of reconnait. In general Bacilly recommends using
the written pronunciation to avoid fooling the ear with words that rhyme in
spelling but not in sound, or words whose meaning is altered whenj adapted to
the rhyme. Yet if words like reconnois and fois do rhyme, then you ¡pronounce
them as written.
CONSONANTS
Certain consonants deserve special consideration and attention, particularly

"r," "1," "n," "m," and "s." In most situations when these consonants are found
between two vowels, they should be pronounced simply and lightly, even if
they are doubled as in belle. When "r" and "1" are used preceding or following
another consonant, they need more force and intensity, particularly when they
precede, almost as if they were doubled. Pardon and charmant need more em-
phasis than preñare or agréable, but cruelle or ingrate can take more energy if
the context and the meaning indicate it. The "1" in malgreczn take more energy
than the one mpL·isir, but you must be careful not to insert a shadow vowel be-
tween the consonants, resulting in malegré. This caution holds for other conso-
nant combinations as well. Bacilly advises that "n" must be more gentle than
the vigorous "r," and singers must be careful not to add an extra "n" before the
initial consonant of a word, a mistake modern students make as well.
Consonants before a vowel can be elongated or suspended, a technique
Bacilly calls gronder (growling). The most important consonant to suspend is
"m" as in mouriror moment, in which the first "m" is elongated but the second
one is not because it is between two vowels. Other consonants can be treated
this way as well, including "n," "f," "j," "v," and "s," as in non, enßn, jamáis,
volage, and severe. Final consonants follow complex rules regarding plurals,
persons, and verb forms, as well as liaison and separation made for phrasing and
breathing. All these devices are used according to the context and sense of the
poetry and are governed overall by le bon gout.
SYLLABLE LENGTH
Bacilly s discussion of long and short syllables makes up the entire third part of
his Remarques and was one of his primary reasons for writing the treatise. He
emphasizes the need to identify the long syllables so the singer will know where
to add ornaments and how to adjust the melody of an air de courfor the second
verse of text. Within a series of long syllables, it is further necessary to establish
a hierarchy of slightly longer and shorter ones. For the modern student of
French language and music, some of these distinctions are obvious, while other
circumstances present difficulties and questions. Bacilly considers many prob-
lematic situations in great detail. For the present discussion I will try to sum-
marize some of his rules without venturing into the vast areas of exceptions.
Syllables that are usually long:
penultimate syllables of feminine words: imaginable, cruelle

monosyllables ending in "s," "z," and "x": aux, bas, prez
monosyllables with "n' following a vowel: топ, rend
syllables within words with "n" following a vowel: danger, défendre
Syllables that are usually short:
penultimate syllables of masculine words: secret, espoir

monosyllables: de, te, me, que
Elisions and other surrounding syllables can change the relative length of an
otherwise long syllable. The final syllable of a song or air is usually considered
long regardless of its length in another context, particularly the mute "e." The
context and dramatic meaning of the poem also contributes to the relative im-
portance of long and short syllables. Phrasing, inequality, and ornamentation
are all affected by the rhythmic contours of the text, resulting in small, delicate
phrases, ¿/^^/articulation, and graceful embellishments. '
Here are some of the main issues introduced in this chapter: A score of early
Baroque music is not a complete representation of the composer s wishes or a
record of the first performance. When you approach a seventeenth-century
work, you will need to make choices about a variety of issues that may include
instrumentation, realization of a figured-bass accompaniment, tempo, charac-
ter, articulation, and ornamentation. We have seen that the text is always an im-
portant guide in these matters. You will also need to make informed choices
about the kind of edition you use. Look for modern scholarly editions and
check the accompanying commentary. Also become familiar with reading the
increasing number of reproductions or facsimile editions of composers' manu-
scripts and the early prints that are available from publishers and in libraries.
They can give you access to a wealth of wonderful seventeenth-century reper-
toire otherwise unavailable.
By considering the period s writings about vocal production and technique
as well as the instruments and performing spaces, you can develop a sense of the
sound ideal for voices in the seventeenth century. It appears that in some re-
spects it was quite different from what we expect to hear from classically trained
singers today. The language used then to describe agility and vibrato is ex-
tremely confusing and foreign to us. Yet the writings give the general impres-
sion that singers were instructed to sing with less breath pressure and muscular
force than we use today, resulting in a gentle and flexible tone that allowed the
voice to move with increased speed and a small, hardly noticeable vibrato.
Singers today should not try to remove the vibrato from their voices or sing
with a straight tone or tight throats for early Baroque repertoire. Rather they
should experiment with their flow of breath to achieve a softer, gentler tone.
Trends in ornamentation varied dramatically both from nationality to na-

tionality and over the course of the century. We have seen, and will continue to
see throughout this book, how fashions in ornamentation have ranged from
modest tastefulness to unimaginable virtuosity. From the seventeenth century
to the present we will see an ongoing negotiation between singers and com-
posers about who is responsible for what kinds of ornamental additions.
Singers today must first become familiar with the specific kinds of ornamental
patterns used in music of different times and places in the early Baroque. The
only way to do this is to study and practice many examples from specific works
and primary sources. Then you can try to create your own variations on those
patterns within the bounds of the specific style you are performing.
Chapter 2 The Late Baroque
This chapter examines music from the first half of the eighteenth cen-
tury, including works by]. S. Bach, Handel, and Rameau. It discusses
the development of vocal music and singing styles in Germany,
France, and England and considers how they were affected by the
overwhelming popularity of Italian opera and the fame of Italian
singers. It reviews how the relationship between words and music con-
tinued to develop, and it discusses different national approaches to-
ward ornamentation. Issues presented in chapter i, such as scores and
notation, figured bass, early instruments, vibrato, pitch, rhythmic al-
teration, dance forms, and the relationship between composer and
performer, are reconsidered from an eighteenth-century perspective.
This chapter also addresses the larger issue of articulation in Baroque
music and its relationship to harmony, dissonance, rhythm, tempo,
and text.
Many of the best-known eighteenth-century treatises providing
useful information about style and performance practice are German:
C. P. E. Bachs on keyboard playing, Leopold Mozart s on violin play-
ing, and Johann Joachim Quantz s on flute playing. Though written
53
54 The Late Baroque
during the transition between the late Baroque and early Classical periods, they
reflect the practices of the previous generation. Most helpful for singers is the
1757 Anleitungzur Singkunst: a translation with commentary by J. S. Bach's stu-
dent Johann Friedrich Agrícola of Tosí s 1723 treatise. Julianne Baird, the well-
known soprano and translator of Agrícola, maintains that vocal music was con-
sidered more important than instrumental music during the early to middle
eighteenth century, and thus Agrícolas Anleitung was possibly more popular in
its time than the manuals by Quantz and C. P. E. Bach.λ The Italian approach
of Tosí, upon which Agrícola is based, exerted a major influence on singing
style. Also important were French treatises by Jean-Baptiste Bérard and Michel
Pignolet de Montéclair, which helped to define and set apart the French style of
declamation and ornamentation. (See "Chapter 2: Primary Sources" in "For
Further Reading" for all these treatises.)
As the antagonism between the Italian and French styles raged on among
musicians and critics, Italian singing and French dance continued to have
widespread influence on music all over Europe. By the second half of the cen-
tury, German composers such as Telemann had combined the best features of
the rival camps with their own typically German style. By knowing something
about each of the three major national schools of composition and perfor-
mance practice—Italian, German, and French—a singer can evaluate which
elements of each are present in a particular work and choose appropriate stylis-
tic approaches.
Scores and Notation
As discussed in chapter i, Baroque composers expected that their performers

would be familiar with basic principles of style and performance practice. In-
formation that we now expect to find in a score, such as instrumentation,
tempo, or exact rhythms, was often omitted or indicated in a way that we find
problematic. As in the seventeenth century, singers and instrumentalists were
expected to add appropriate ornamentation, both small and quite elaborate,
which was not written in the score. Nikolaus Harnoncourt suggests that a
Baroque score is actually more a record of the composition itself than a set of in-
structions for performance.2 If the composer was part of the performance, as
was often the case, questions of notation or interpretation could easily be an-
swered in rehearsal. Different courts or churches throughout Europe had vary-
ing numbers of musicians in their service, depending on their budget or the
taste of their patron. Instrumentation, transposition, ornamentation, and even
tempo could depend on the available performers or on the acoustics of the
The Late Baroque 55
room. Music was often recomposed or adapted for different performers or per-
formance situations. Revising a work could also generate renewed enthusiasm
and prevent pirated performances. The score was just a starting place; com-
posers and performers made generous use of the variable resources available to
them. ι
Modern Editions
In the second half of the nineteenth century, historians and editors started pro-
ducing historical editions of Baroque music. While this effort resulted in much
fine work, these editors tried to fix the "mistakes" and correct the "inconsisten-
cies" found in eighteenth-century scores. Their editorial adjustments often led
to interpretations from a nineteenth-century perspective. More recent editions
of Baroque music try to present scores closer to their original "bare-bones"
form, while still providing suggestions for solving some of the more problem-
atic issues. These critical or Urtext editions often include extensive commen-
tary about sources and editorial procedures. Paul Steinitz, in his Performing
Bach's Vocal Music, warns that "dynamics, phrasing and tempo marks given in
nearly all vocal scores, other than those from Bärenreiter and Curwen, should
be viewed with the utmost suspicion and checked against Urtext full scores and
musical common sense."3 While this opinion is a bit extreme, it is advisable to
be cautious in evaluating Baroque scores, particularly vocal scores.
On the other hand, using collections of Handel arias transposed ι into keys
other than the original is not necessarily inconsistent with eighteenth-century
practice. Arias were frequently transplanted from one situation to another,
from opera to opera or theater to chamber, as well as being transposed or re-
composed to suit a particular singer. French cantatas were often designated for
soprano or tenor, depending on who was available. It is always interesting to
learn of a works originally intended voice type, but it is equally appropriate to
sing Baroque arias out of context or in transposition.
Another important issue regarding modern piano-vocal scores of Baroque
works is the piano reduction of the continuo and obbligato lines. The majority
of vocal works from the early eighteenth century are for voice andi continuo
alone or for continuo with one or more obbligato instruments. Because musi-
cal textures became more complex in the eighteenth century, it is more impor-
tant for performers to be able to see the different lines of counterpoint and how
they interact with one another. It is also necessary to distinguish the composed
instrumental lines from improvised continuo filler. A piano reduction that
combines all the different components into one homogenized texture may be
56 The Late Baroque
easier to play at a choral rehearsal, but it is not helpful for studying the work and
making decisions about rhythmic alteration, ornamentation, and articulation.
When evaluating the score of a Baroque work, check the date of publication and
read the preface (if there is one). Consider the editorial markings, the realization
of the continuo part, the arrangement of the instrumental lines, and any accom-
panying scholarly commentary. Decide whether the edition has made too many
restrictive decisions for you or has left some room for flexibility.
Tempo
Many treatises advise considering the music itself and its mood or "affect" in or-
der to determine a proper tempo:
The pace of a composition, which is usually indicated by several well-known Italian

expressions, is based on its general content as well as on the fastest notes and passages
contained in it. Due consideration to these factors will prevent an allegro from being
rushed and an adagio from being dragged. (C. P. E. Bach, 1753)^
One must deduce [the tempo] from the piece itself. . . . Every melodious piece has at
least one phrase from which one can recognize quite surely what sort of speed the
piece demands. (Leopold Mozart, 1756)5
For singers, the text is the best place to look to determine a works dramatic
"affect" and mood. The rhythm of the diction plus the emotional quality of the
words and music provide invaluable clues to an appropriate tempo. These ele-
ments, combined with a review of the faster note values or more intricate pas-
sages, should yield a good notion of how the piece needs to move. In the eigh-
teenth century, words given as tempo indications were often more descriptive
of a mood than a specific speed: allegro—cheerful; allegretto—rather gaily but
gracious; andante—walking comfortably; adagio—at ease; grave—serious.
Meter indications also suggested a certain kind of movement based on the
number of strong beats per measure: 3/4, which was felt as three beats, had a
very different character than 3/8, which was felt in one. Other factors con-
tributing to tempo choice included a singer's personal temperament, the way
his or her particular voice moved, the acoustics of the room, and the needs of
the instrumentalists. In general, Baroque tempos can be faster than what we
may be used to. (For more about choosing Baroque tempos, see Donington,
Style and Performance, 11-19, an(i Cyr, Performing Baroque Music, 29-47; both
sources are listed in "Chapter i: General" in "For Further Reading.")
If a vocal work is based on a dance form, the traditional movement ofthat
dance also implies an appropriate tempo. Meredith Little and Natalie Jenne, in
The Late Baroque 57
their book Dance and the Music ofj. S. Bach, suggest that the "Et exsultavit"
movement of the Magnificat is a minuet (ex. 2.ia). The character of this classic
French dance in 3/8 has been described as noble, with an elegant simplicity.
The recommended speed of the dance ranges from very lively to moderate,
with one strong beat per measure and the phrase starting on the downbeat.
This seems appropriate for the "Et exsultavit," which has no tempo indication
Example 2.1. J. S. Bach: a) "Et exsultavit," mm. 1-14, from Magnificat, NBA Il/ш/бу, b)
"Quia fecit," mm. 1-6, from Magnificat, NBA II/iii/96.
58 The Late Baroque
but does have strong syllables on each downbeat in a 3/8 meter. The thirty-sec-
ond-note figures scattered throughout the bass line, first violin, and vocal lines
function as written-out ornaments and can help determine the upper limits of
a brisk tempo. Little and Jenne also suggest that the "Quia fecit" aria for bass is
a gavotte (ex. i.ib). This popular dance was characterized by two large beats per
measure, with the phrase starting on the upbeat. The character was usually de-
scribed as graceful, and it could be either joyful or serious in a variety of tempos
within the moderate range. Bach's aria has no tempo indication, and it is im-
portant to bring out the gavotte phrasing and rhythm—-"quia" leading to
"fecit" "mini" leading to "/^^gna," and so on—in a moderate tempo.
Articulation
In Music as Speech, Harnoncourt explains that Baroque music has a discernible

grammar. It is made up of small pieces, or figures, that fit together like words to
form phrases, sentences, paragraphs, and larger forms. The seventeenth-cen-
tury idea of constructing music based on the rules of rhetorical speech and ora-
tion reached new levels of complexity in eighteenth-century music. Many writ-
ers of the time distinguish between musical figures that should be grouped
together and those that should be separated. Most instrumental treatises dis-
cuss specific bowing or tonguing techniques to produce a wide variety of de-
tached and slurred articulations. Often the goal was to create the impression
that the instrument was playing syllables and words of different relative lengths
and importance. Emphasizing the contrast between stressed and unstressed
notes, which were sometimes referred to as "good notes" and "bad notes," is
one of the fundamental principles of Baroque articulation. Highlighting the
difference between heavy and light also applies to measures and whole phrases,
building a complex hierarchy of tension and release, dissonance and resolu-
tion.6 As Quantz wrote, "Musical ideas that belong together must not be sepa-
rated; on the other hand, you must separate those ideas in which one musical
thought ends and a new idea begins. . . . You must know how to make a dis-
tinction in execution between the principal notes, ordinarily called accented от
in the Italian manner, good notes, and those that pass, which some foreigners
call ¿¿¿/notes. Where it is possible, the principal notes always must be empha-
sized more than the passing."7
Singers have an easier time making decisions about articulation because we
have a text with built-in syllabic stresses and grammar. We also have conso-
nants, which serve to separate the vowels in each syllable. Baroque composers
used the parameters of syllable stress and length, metric placement, and har-
The Late Baroque 59
mony and dissonance to create different levels of emphasis and relaxation

throughout a work. As Bacilly discussed at length in 1668, using the varying
lengths of vowels and consonants can help to clarify the difference between
long and short syllables in French. Fifty to one hundred years later in the late
Baroque period, this technique still helped to highlight the stressed and un-
stressed notes and to bring out interesting contours of the melody line. The
idiosyncrasies of the German and Italian languages lend themselves to this ap-
proach as well. German has many percussive consonant clusters and guttural
sounds that produce a different kind of articulation from the fluid, legato flow
of Italian. Each language possesses its own particular rhythms and contours
that contribute to the distinct national flavor and character of the music. The
metric placement of syllables and words also contributes to the contrast be-
tween stressed and unstressed notes. Most of the time, stressed syllables and im-
portant words fall on strong beats of the measure, reinforcing the expected met-
ric hierarchy of "good" and "bad" notes. When a strong syllable falls on a weak
beat, the resulting rhythmic clash can add more interest and texture to the mu-
sical fabric.
The harmonic movement of the music also affects the heightened impor-
tance of some notes over others. Dissonance was considered more expressive
than consonance. Performers were expected to add dissonant appoggiaturas to
increase expressive tension and to highlight the contrast between heavy and
light. "Above all," Agrícola wrote, "it is the dissonances which make the har-
mony more interesting. According to the rules of good taste, dissonant notes,
in general, are produced more loudly than consonant notes."8
In example 2.2, the first few measures of J. S. Bach's "Wenn kömmst du,
mein Heil?" duet provide an opportunity to see how these tools can be used. In
the opening phrase the soprano starts on a weak beat of m. 8 with the relatively
short and unimportant syllable "wenn." "Kömmst," more important because it
is the verb, falls on a strong downbeat and has an active and energetic ornament
attached to it. The "k" has a percussive attack that, when combined with the
more emphasized and ornamented downbeat and the weaker "wenn" syllable,
should produce the desired contrast and slightly separated articulation that are
characteristic of early eighteenth-century style. "Kömmst" itself is not a long
syllable. In getting from "kömmst" to "du," you must sing through a slightly
noisy consonant cluster of "-mmst d-". If you take a little time off the "o" vowel
and linger briefly on the "mm" before moving on to the shorter and weaker
"du," you will create a graceful valley after the peak of the downbeat. This ap-
proach will also highlight the weaker rhythmic placement of "du mein" as you
Example 2.2. J. S. Bach: "Wenn kömmst du, mein Heil?" a) mm. 1-3; b) mm. 8-10, from BWV140, NBA I/xxvii/i5i.
The Late Baroque 61
prepare for the next strong arrival on "Heil." "Heil" is an important word in the
text; it is a long, smooth syllable and is further emphasized by a dissonant ap-
poggiatura on a strong beat.
Late Baroque musicians preferred a slightly separated articulation even in
long melismas on one syllable. Agrícola instructs students to "imagine that the
vowel sound of the division is gently repeated with each note . . . just as with a
stringed instrument, [where] a short bow stroke belongs to each note of the di-
vision." If divisions are slurred, "the singer must guard against allowing the
notes to become unclear." He also suggests that the first note in a group of three
or four fast notes should receive a slight emphasis to maintain clarity and a
steady tempo.9 Deciding which notes to stress in melismas depends on the pat-
tern and sequence of the melody and its relationship to the harmony. Dividing
a long passage of fast notes into smaller pieces makes the melisma easier to sing
and highlights the music's stressed and unstressed contours. Rather than apply-
ing a modern conception of a smooth, consistent flow of tone throughout, we
can bring clarity to the texture by using expressive diction and varied! articula-
tion. By looking for the natural rhythmic variety of the language, as well as the
intricate contours of melody and harmony, we can find the tools of stylistic ar-
ticulation in the music itself
Rhythmic Alteration
By the early eighteenth century, rhythmic alteration was not only limited to
employing inégalité in French music. Dotted figures, which we today under-
stand to have specific proportion and duration, were interpreted with much
more flexibility. A dotted eighth and sixteenth note, for example, could be per-
formed with different degrees of length and crispness, ranging from a gentle
triplet to a "double dot" (this is not necessarily a precise mathematical relation-
ship) depending on the character of the situation. Standard and idiosyncratic
conventions in notation led to frequent inconsistencies in rhythms and articu-
lation marks. Some of these differences could have been intentional oiji the part
of composers, but some could also have been due to sloppy penmanship and
scribal errors. While triplets, dotted figures, and even eighth-note figures often
appear to be interspersed interchangeably throughout a work, it may ¡be neces-
sary in one situation to change an even figure to agree with a dotted ¡one, or a
dotted figure to agree with a triplet. When different rhythms appear simultane-
ously, they pose particular problems that need to be solved on a case-by-case ba-
sis. Again, depending on the character of the music, dotted figures would most
62 The Late Baroque
likely be softened to accompany simultaneous triplets, since it was unusual to

have a two-against-three rhythm in Baroque music.
In the air from Montéclair's cantata Arlane et Backus shown in example 2.за,
the flute plays a melody notated with dots. When the voice enters in m. 15 with
the same melody, it is notated with equal eighth notes. Later^ in m. 27 (ex. 2.30),
when the flute and voice play the melody together in thirds, it is notated with
dots in both parts. Still later the voice sings an even eighth-note figure over a
dotted countermelody in the bass. The performers must decide how to recon-
cile these rhythmic inconsistencies. Given that the tempo marking is Lent and
the text is sad and mournful, you would probably not want to emphasize the
crisp, bright aspect of the dotted figures. Considering that the music is French,
you could also decide to use inégalité to soften both the dotted and nondotted
figures into some degree of gentle, lilting triplets. In another work you may de-
cide that the different rhythms are deliberately composed into the motivic
landscape and therefore need to be highlighted to provide variety and excite-
ment. It should not be taken for granted, as we do today, however, that the lit-
eral rhythms shown in the score were exactly what the composer wanted.
The inconsistent use of slur marks in Bach scores have left performers and
scholars perplexed for generations. John Butt, in his detailed study of Bach's ar-
ticulation marks, refuses to believe that the incongruities are due to careless and
hasty penmanship in the heat of composition or reflect indifference to a hap-
hazard performance.10 Harnoncourt feels that the differences in articulation
markings found in many Baroque scores add to the complex texture of distinct
layers typical of Baroque music.11 Still, some of these discrepancies could
merely be mistakes or oversights. In vocal repertoire, slur markings usually in-
dicate syllabification and word underlay, which may in turns inform the articu-
lation of an instrumental line. For singers, the text offers the best guide for
making decisions about phrasing and rhythmic alteration. The dramatic and
emotional quality of a work will tell you how gentle or energetic to make its
dotted figures. The internal rhythm of long and short syllables of words can
also help suggest how to phrase notes together or ways to reconcile inconsistent
rhythms.
Legato
While the primary approach to early Baroque articulation was separated and
detached, the use of legato gained favor throughout the eighteenth century.
Singers were the model for both. Just as instrumentalists were encouraged to
The Late Baroque 63
Example 2.3. Montéclair: Arlane et Bachus, a) mm. 11-19; b) mm. 27-31, from Cantatas:
Book $ (1728), modern edition Cantatas for One and Two Voices, A-R Editions, 1978, pp. 33-
34·
64 The Late Baroque
create diction and grammar with their articulation, they were also advised to
emulate the cantabile style of singing:
Each instrumentalist must strive to execute that which is cantabile as a good singer
executes it. The singer, on the other hand, must try in lively pieces to achieve the fire
of good instrumentalists. (Quantz, 1752)12
The human voice glides quite easily from one note to another; and a sensible singer
will never make a break unless some special kind of expression, or the divisions or
rests of the phrase demand one. . . . Indeed it is good practice to sing instrumental
melodies in order to reach an understanding of their correct performance. (C. P. E.
Bach, 1753)13
Agrícola tells the voice teacher "he should take care that the notes emitted by
the student are well connected to one another and thus legato."14 Bérard men-
tions a kind of legato singing in which syllables are connected by a linking
sound that is sung softer than the words themselves.15 Butt proposes that the
emphasis on legato singing grew as the eighteenth century ¡progressed, but he
cautions us not to apply a modern understanding of continuous legato to an
eighteenth-century context.16 Leopold Mozart compares an ideal adagio bow
stroke to the gentk crescendo and decrescendo that singers use on long, sus-
tained no tes.17 The soft swelling and diminishing of the messa di voce was still a
commonly used gesture for both instrumentalists and singers. The stmscino
(drag) and the portamento or portare la voce (carrying the voice) are explained
by Agrícola as gestures in which the notes are bound together into a smooth
flow. But the very fact that writers explained how to execute these special ges-
tures suggests that they were not used as a matter of course. They were to be dis-
tinct events, interspersed with the more detached kinds of articulations as the
music and the text demanded.
Vibrato and Breath
Vibrato in the early eighteenth century was still considered an ornament, to be

used selectively in appropriate situations. However, it was written about in a
significantly more positive light than in the previous century:
The flate is a type of vibrato which the voice makes by means of several small, gentle
exhalations without raising or lowering the pitch. [This is done] on a note of long
duration or on a note of repose. . . . If the y&zré were used on all important notes, it
would become unbearable in that it would render the melody tremulous and too
monotonous." (Montéclair, 1736)18
The Late Baroque 65
The Tremolo [vibrato] is an ornamentation which arises from Nature herself and
which can be used charmingly on a long note, not only by good instrumentalists but
also by clever singers. . . . it would be an error if every note were played with the
tremolo. (Leopold Mozart, 1756)19
The vibrato on one note—is also an ornament that in singing is especially effective
on long sustained notes, particularly when applied toward the end of such notes . . .
but not all throats are capable of this type of execution. (Agrícola, 1757)20
The use of vibrato can be thought of as part of a varied palette of articula-

tions. Stressed dissonances surely need little or no vibrato, but long notes can
benefit from a tasteful amount, depending on the harmony. Modern singers,
who are accustomed to using a steady flow of breath and tone, should try to
find flexibility in their use of breath, rather than tightness in the throat, in or-
der to negotiate different types of vibrato, ornaments, and articulation. Bérard
includes a lengthy discussion of breath and different kinds of exhalations to go
with different "affects" or moods: "For good expirations, the interior air must
be made to leave with more or less force, with more or less volume, according to
the character of the song. There is no doubt that the art of good inspiration and
expiration will multiply the powers of a singer, and will give him great facility
in executing the ornaments well."21 Specific ornaments will be discussed below
in the sections on national styles.
Pitch
The pitch situation was still quite complicated in the eighteenth century. The
works of Bach, written for various local organs and ensembles, are a maze of
complex transpositions for organ, trumpets, woodwinds, and strings, depend-
ing on where each work was composed and performed. Since the organ in
Leipzig was tuned particularly high, the continuo parts composed for those
performances were usually written a major second lower than the rest of the in-
strumental parts. In order to be performed in Leipzig, works written in Weimar
had to go through another layer of transposition and adaptation in order to be
suitable for the instruments. The largest concern, however, was for the range
and comfort of Bach's singers and particularly the male sopranos and altos, who
could not sing comfortably if the ranges were too high.22
Agrícola includes a lengthy discussion of such problems. He mentions the
differences between the high tuning in Lombardy and Venice and the much
lower pitch in Rome. He describes the difficulties of learning an aria in one key
and then having to sing it higher or lower in another city. In addition to the
66 The Late Baroque
stresses they put on the extremes of a singers range, such adjustments caused
great confusion in negotiating the transitions from the natural voice to the fal-
setto or the middle range to the head voice: "under such circumstances, the
happiest singers are those in whom the unification of the natural voice and the
falsetto does not cause much difficulty and who, in case of necessity, have a few
extra notes in their high and low registers."23 The same would be true today for
a singer who had to sing performances of a Bach Passion or Handels Messiah
with a modern orchestra one day and a period instrument ensemble the next.
Period Instruments
Singers can learn a great deal about articulation, sound, and nuance in Baroque
music by working with period instruments (restored original instruments or
new reproductions of old instruments). Players who are skilled and knowledge-
able about Baroque style can demonstrate characteristic instrumental gestures
such as crisp, detached articulation, inégalité> or the gentle rising and falling of
bow pressure that Leopold Mozart likened to the messa di voce. The particular
construction and playing techniques of early instruments produce an articula-
tion model that we can imitate.
The sound of many of these instruments is softer, gentler, and more mellow
than their modern counterparts. Due to the complexities of early tuning sys-
tems, many early instruments can create interesting variety in tone color and
intonation, depending on what key they are playing in. Singers can emulate
this variety of sound, which differs so markedly from the consistency of
strength and range of modern instruments. Reproductions of early instruments
are generally built to conform with the now standardized Baroque pitch of a' =
415 cycles per second. Singing a half step lower than the modern a' = 440 al-
lows for more relaxed sound production and makes some of the subtle gestures
of articulation easier to employ, particularly in the high range. Since the total
volume of sound produced by a period orchestra or chamber group is much less
than that of the comparable modern group, the total amount of effort for
singing and projecting can be similarly less. It is possible, of course, to apply el-
ements of Baroque style to performances with modern instruments, but certain
compromises undoubtedly have to be made because of their higher pitch,
greater volume, and more consistently brilliant tone.
Recitative
Recitative in the early eighteenth century was written for three different types
of performance venues: church, chamber, or theater. Each demanded its own
The Late Baroque 67
stylistic approach. Tosi instructs that the singer should be allowed the most free-
dom in Italian church recitatives to shape the rhythm to the text and to add ex-
pressive ornaments. In theater recitatives, by contrast, the singer has little free-
dom to play with the rhythm and should add no ornaments. Chamber recitatives
fall somewhere in between, being less serious than those for the church but
somewhat closer to those for the theater. Agrícola comments: "All three types of
recitative share in common that which the author has specified above for the
church recitative: that it is not sung in strict time. One must be guided more by
the length and shortness of syllables in common speech than by the written value
of the notes in the recitative."24 He shows other additions to the written notation
that the singer must make, such as passing appoggíaturas and trills, both within
the body of the recitative and particularly at cadences. In example 2.4, for in-
stance, the singer should fill in descending thirds with passing notes but keep the
penultimate note on the upper portion when a cadence is approached by a fourth.
Much has been written, both in the eighteenth century and more recently,
on the accompaniment of recitative. Particularly problematic issues include the
timing of cadences and the length of bass notes. In the theater, cadences were
often telescoped or truncated in order to keep the dramatic action moving.
Tosi s term for these was cadenze tronche. If a fast dramatic pace was desired, the
penultimate and final chords of the continuo were often played under the
singer s penultimate or final note, as in the above examples. This was not neces-
sarily the only choice in period practice. If the dramatic situation were more se-
rious or contemplative, the cadence might come after the singers last note.
Again, the words provide the most important guide to making decisions about
pacing and character of both the vocal line and the accompaniment.
Bass notes in recitatives were often written as long notes, either whole or half
notes. Yet it was understood that they might not be held for their full notated
value. Scholars today speculate that writing long note values may have been
faster and easier than writing quarter notes and rests and would have shown the
harmonic progression more clearly.25 Figures written to go with the bass notes
varied widely in type and frequency from composer to composer and even
within a particular composers work. In the cases of J. S. Bach or Handel, who
often played their own continuo parts, the notation did not need to be precise
or complete because the composer, who was also the performer, knew how long
or short to play the bass notes and what figures to add.
In the Italian opera tradition secco (dry) recitatives moved along briskly and
were usually accompanied by a harpsichord whose sound decayed quickly. In
contrast, recitatives in the church music of Germany would move more slowly
68 The Late Baroque
Example 2.4. Agrícola: appoggiaturas in recitatives from Anleitung zur Singkunst ( i j ^ j ) ,

reproduced in Julianne Baird, Introduction to the Art of Singing, Cambridge University
Press, 1995, pp. 174-75-
The Late Baroque 69
and be accompanied by the sustaining sound of the organ. Accompanied

recitatives, in which strings or winds are composed into the texture, were
treated more like arias or ariosos. They were performed with less rhythmic flex-
ibility and more sustained sound. For the secco recitatives, which emulated the
Italian style, German sources cautioned the continuo players against prolong-
ing the annoying organ drone, which might drown out the vocalist s words. In-
stead, they encouraged both keyboard and cello/viol players to play shorter
notes. In his St. Matthew Passion Bach wrote long note values in the score for
the secco recitatives but wrote quarter notes and rests in the continuo parts.
Many scholars and performers have concluded that it was conventional prac-
tice to play all secco recitatives with short bass notes. Yet Patrick Rogers, in his
Continuo Realizations in Handels Vocal Music, presents evidence to suggest that
some recitative situations, particularly those involving harpsichord, demand
something more sustained than the highly detached approach recommended
for the organ. The existence of changing figures during rests or long sustained
bass notes requires a flexible interpretation of the length of bass notes and the
placement of chords. This illustrates that one bit of advice from a primary
source cannot necessarily be applied to all situations. Singers should be aware of
the complex issues involved and the kinds of choices that need to be made re-
garding Italian and German recitatives.
French composers had a completely different approach to setting text in
recitatives. From Lully to Rameau, recitative was written out in a complicated
sequence of changing meters intended to capture the poetic construction and
heightened rhythmic execution of theatrical declamation. Singers and instru-
mentalists were expected to follow the flow of the written rhythm dosely. Ex-
pressive diction and artfully placed ornaments, rather than rhythmic freedom,
helped to convey the emotion of the text. In the example from Rameau's can-
tata Le berger fidele (ex. 2.5), the singer and the accompanying instrumentalists
must observe the notated rhythm more strictly than in German or Italian styles
in order to stay together through the changing meters of fours, twos, and
threes. Italians who criticized French music often complained that they could-
n't tell where the recitatives ended and the arias began, because the contrast in
pacing and flow between recitative and air was not as great as in Italian music.
ITALIAN INFLUENCE
The most characteristic feature of the Italian style of singing in the late Baroque
period was the virtuosic display of dazzling vocal showmanship. The famous
70 The Late Baroque
Example 2.5, Rameau: Le bergerfidele, recitative no, 5, from Cantatas: Booki(ijx^),

modern edition Deutscher Verlag, 1969.
The Late Baroque 71
Italian singers were sought all over Europe: the great castrati Farinelli, Pistoc-
chi, Bernacchi, Senesino, and Guadagni were known for their phenomenal
breath capacity and control. The renowned Italian women singers Faustina and
Cuzzoni created a sensation in London by competing for attention in Handels
operas.26 Singers tried to best each other and the accompanying instruments in
an attempt to gain the audience s favor and to steal their attention away from
eating, talking, and general distractions. By the time Tosi wrote his Observa-
tions on the Florid Song in 1723, he condemned what he considered to be the
horrid excesses in ornamentation that singers were inflicting on Italian music.
Yet contemporary reports indicate that ornamentation became even more elab-
orate as the century progressed. Charles Burney, a noted historian and critic of
the time, wrote a great deal about the performances of singers in the early eigh-
teenth century. In his General History of Music, written in the latter half of the
century, he comments that Farinelli's ornamental display "which excited such
astonishment in 1734, would be hardly thought sufficiently brilliant in 1788 for
a third rate singer at the opera."27
It is difficult to know exactly what kind of ornamentation singers added or
what was fashionable at a given point in the century. From accounts of perfor-
mances, instructions in treatises, and some surviving manuscripts with written-
out ornamentation and cadenzas, we can glean some ideas of how to re-create
the Italian virtuoso style.
Ornamenting Arias
The da capo aria, whether in opera or cantata, became the place to show off
one s vocal and compositional skills. There were four principal ornamentation
types: appoggiaturas and trills; first-section aria variations and interpolated di-
visions; slow-section melody recomposition; and thrilling cadenzas.
APPOGGIATURAS
As already mentioned, the use of certain musical gestures was common knowl-
edge to musicians of the early eighteenth century. This was certainly the case
with appoggiaturas and trills, which were often not notated but rather expected
to be added by singers in appropriate places throughout arias and recitatives.
Tosi even mentions that Italian singers would laugh if a composer indicated an
appoggiatura that they would have naturally added on their own.28 Appoggiare
in Italian means "to lean," and an appoggiatura is made by leaning on an auxil-
iary note. This usually adds dissonance to a strong beat, creating more contrast
between strong and weak or tension and release. (This type of ornament in the
72 The Late Baroque
seventeenth century was called a beat or backfall in England, and an accent от

port de voix in France.) I have already mentioned cadential appoggiaturas in
recitatives. Most of the writers of the major eighteenth-century treatises devote
entire chapters to the proper use of appoggiaturas.29 Their commentary is ex-
tensive and can be confusing and contradictory. Sometimes appoggiaturas are
indicated by small notes written before the main note to be ornamented. Many
times composers notated these ornaments in an instrumental part but not in
the vocal line. In other scores appoggiaturas are composed directly within the
melody. By studying the appoggiaturas that are indicated or written into the
music, you can learn what kinds of situations demand or can support added ap-
poggiaturas. Tosi himself was confident that the ear could tell you when an ap-
poggiatura is needed and when it is not.
The well-known aria "Sebben, crudele" by Antonio Caldara provides many
instances of both written and unwritten appoggiaturas (see ex. 2,6). As the aria
appears in the Schirmer Twenty-four Italian Songs and Arias edition, it features
an eighth-note accompaniment that was added for the 1885 Ricordi edition of
Arie antiche. The original 1710 manuscript shows an accompaniment of half
and quarter notes, which gives the Ε-minor aria more of a minuet-like feel and
makes the appoggiaturas easier to see. In example 2.6a Caldara includes a writ-
ten-out appoggiatura on the first beat of m. 8, the second syllable of "langer."
In the second rising chromatic theme (ex. z.6b), he writes 4-3 suspensions in
mm. 28 and 30 that feel like appoggiaturas on the second syllables of both
"crupié" and "1ап£ш>." In the answering descending phrase he writes disso-
nant suspensions in each measure on "stmpre fe-de\¿' and "ri wglio." All these
effects (marked with arrows in the example) act like appoggiaturas, adding ten-
sion to a strong beat by creating a stressed dissonance that is then resolved into
a relaxed consonance.
If you wanted to add more appoggiaturas as additional ornaments, in the
opening phrase you could repeat the E of "cm-" as a dissonant upper neighbor
note on the "de-" of the downbeat of the next measure. The "-le" on C, which
passes from the D of the first beat to the В of the third beat, can also be thought
of as a passing appoggiatura, though it doesn't receive as much emphasis be-
cause it falls on a weaker beat. (It is interesting to note that in the 1885 version
of the aria included in the Schirmer anthology this "-le" syllable has been
moved to the third beat, smoothing out the phrase and destroying the feeling of
a passing appoggiatura.) Dissonant lower neighbor notes can also act as appog-
giaturas. In the first statement of "Sempre fedele" you could add a dissonant
lower neighbor note to the "fe^le" syllable by repeating the В of "yódele" on the
Example 2.6. Caldara: "Sebben, crudele," a) mm. 5-8; b) mm. 27-34, modern
transcription from 1710 manuscript in 26 Italian Songs and Arias, ed. John Glenn Patón,
1991. Used with permission of the publisher. © Alfred Publishing Co. All rights reserved.
74 The Late Baroque
downbeat of the next measure. In this situation it would also be permissible to

add an upper neighbor appoggiatura by jumping up a minor third to the half
step above the "fe¿j5?le" syllable.
Usually appoggiaturas are stressed and sung or played on the beat. They can
be performed louder than the neighboring consonant notes, except for passing
appoggiaturas, which fill in thirds. These can be lighter in emphasis and come
before the beat, as in "fede/^" in the above example. Stressed appoggiaturas usu-
ally take half the value of the note they are decorating (or two-thirds the value
in triple time), regardless of the value of the written small note if there is one.
They can be shorter or longer than this, depending on the harmonic and rhyth-
mic context, but the point is to prolong the expressive dissonance so the release
into consonance is all the more welcome.
TRILLS
While the trill in the early eighteenth century was similar to its seventeenth-cen-
tury ancestor, it was also used decoratively and structurally. It was expected at
major cadence points and could also be used to decorate internal melodies and
less important cadences. It could be combined with appoggiaturas and other
types of preceding ornaments, as well as with various kinds of concluding fig-
ures. In its simplest and most common form it starts on the upper note, which is
usually stressed, with a trill between that and the main note, resolving on the
note below. A great deal of controversy concerns the upper-note start of trills.
Some scholars and performers believe that all Baroque trills should start from
above. Frederick Neumann, in his Ornamentation in Baroque and Post-Baroque
Music, provides evidence that some situations demand a main-note start in-
stead.30 In any case, it is important to evaluate the surrounding melodic and
harmonic circumstances, as well as other motivic details in the music, when
making decisions about trills. Using the Caldara aria as an example again, you
could add a trill to "lang-ш'г" in mm. 7-8 by starting on the D above the written
С on the downbeat, trilling on the С and D, and finally resolving to the B. You
must decide, however, whether the trill is appropriate for the dramatic effect you
want to create here. In the final phrase before the middle section (m. 33), an ex-
pected cadential trill should be added to the second beat of "wglio" by starting
with a reemphasized G from the downbeat and then trilling on G and F-sharp.
DIVISIONS AND IMPROVISED VARIATIONS
Tosi recommends that the first part of a da capo aria needs only a few tasteful
added decorations, perhaps appoggiaturas and trills, while the middle section
The Late Baroque 75
can withstand slightly more. However, the restatement of the first section of the
aria, or the da capo (from the head), requires the most creative invention on the
part of the singer/composer. If the rhythms are slow, you can use the seven-
teenth-century technique of diminution, filling in parts of the melody with
passing notes and smaller note values. If it is a fast aria, you can try varying the
patterns and contours of the coloratura. Bear in mind that any alterations you
make must be in keeping with the rules of voice leading and counterpoint. Just
as important, the style of the ornaments must be from an eighteenth-century
vocabulary. You must be careful not to use familiar coloratura patterns from vo-
cal music by Mozart, Rossini, Donizetti, or other later composers w¡hen orna-
menting Baroque arias. The more you are familiar with early eighteenth-cen-
tury repertoire (both vocal and instrumental), the more appropriate patterns
and figures you can choose from and copy when planning ornaments and ca-
denzas for a da capo aria. It is also invaluable to listen to the ornaments of expe-
rienced singers and to review the examples of ornamented manuscripts that
survive from the period. Ultimately you want to highlight your own technical
strengths and musical good taste.
Handel was a German-born composer whose formative training had been in
Italy. Since he wrote Italian opera in England primarily for Italian singers, the
Italian style influenced most of his vocal music. He worked closely with many
famous singers and wrote music specifically designed to suit their particular
strengths. He also often reworked music originally written for one singer to be
more suitable for another singer. Example 2.7 shows an aria from his opera Ot-
tone (1723) transcribed from a manuscript with added graces. Hellmuth Chris-
tian Wolff suggests that this manuscript was in a collection of transposed arias
that were ornamented and performed in concerts by the castrato Guadagni. In
contrast, Winton Dean concludes from the scholarly evidence that the orna-
ments were added to the transposed music by Handel himself for a perfor-
mance of the opera in which a contralto had to substitute for the famous so-
prano Cuzzoni, who was ill. Whatever the circumstances, the added graces
provide a fascinating example of ornamentation in a period style.31
CADENZAS
Agrícolas German translation and extensive commentary on the treatise of Tosi

offers us the most detailed instructions for improvising cadenzas in the Italian
style. He describes three main cadence points in an aria that, in the seventeenth
century, would have been decorated with a small trill. As time went on, further
embellishment was added before the trill, which gradually began to cause a de-
76 The Late Baroque
Example 2.7. Handel: "AfFanni del pensier" from Ottone(1723), mm. 7-17, transcription of
ornaments in G. E Handel: Three Ornamented Arias, ed. Winton Dean, © Oxford
University Press, 1976. Used by permission. All rights reserved.
The Late Baroque 77
lay in the arrival of the final note. "Finally there was an attempt to adorn this
delay with all kinds of improvised divisions, runs, drags, leaps—in short, all the
possible figures that the voice is capable of executing."32 These cadenzas, which
Agrícola supposes originated between 1710 and 1716, can be added at the final
cadence at the end of an arias first section and possibly also at a cadence within
the aria if it seems appropriate. A cadenza can also be added at the final cadence
of the middle section and certainly at the final cadence of the concluding sec-
tion or at the repeat of the opening section. Tosi disapproved of an accompani-
ment's coming to a complete halt for a cadenza.33 Julianne Baird supposes that
cadenzas in the early part of the century were performed over a continuously
moving bass line, at a slow tempo marked adagio or andante. Later, as cadenzas
became more elaborate, the continuo players would have to wait longer for the
singer to reach the final note.
Agrícola instructs that improvised embellishments must maintain the affect
or character of the work. If possible, thematic material or motives from the aria
should be used, while also including some sort of surprise—perhaps notes
from outside the range of the aria. The singer should show off his or her
strengths, while also following the rules of good voice leading and composition.
Usually the cadenza is sung on the syllable preceding the trill, though the word
underlay can be adjusted somewhat to provide a comfortable vowel. Most im-
portant, it should be performed all in one breath, though singers in the eigh-
teenth century devised ways to sneak breaths in order to prolong their moment
of glory. Some singers performed truly improvised cadenzas, but there is evi-
dence to suggest that they also wrote them out and practiced them ahead of
time. This was probably more necessary for double cadenzas involving two
singers or a singer and an instrumentalist. Both Agrícola and Quantz recom-
mend planning ahead for double cadenzas to avoid mistakes in counterpoint or
a collision in performance.34
Handel's Messiah
One of the best-known works of Baroque vocal music, Handel's Messiah has
been performed virtually every year since its first performance in Dublin in
April 1742.35 In Handel's lifetime, each performance used somewhat different
performing forces according to the location and the personnel available. Con-
sequently, Handel made abundant changes to the score, rearranging and
rewriting many passages and redistributing the arias to different singers. Each
singer, of course, brought personal strengths and idiosyncrasies to ¡the music's
interpretation and ornamentation. It is difficult to determine what changes
78 The Late Baroque
were made for purely practical reasons and what revisions were for composi-
tional considerations or Handel's own changing preferences. It is certain that
there is no definitive version that is absolutely "authentic" юг "correct." By the
end of Handels lifetime, a slightly more standard version had evolved, incor-
porating changes the composer made in 1750 for the famous Italian castrato
Guadagni, who performed some of the arias formerly sung by English female
altos. In 1767 a full score was published, further standardizing the work and
making it available to the general public. By the end of eighteenth century, in-
creasingly larger performing forces were used, taking the work farther away
from its modest beginnings and on toward the nineteenth-century tradition of
grandiosity.
Today a wide variety of styles in Messiah performances cover a broad spec-
trum of possibilities from small period-instrument orchestra with early music
singers to full modern symphony orchestra, large chorus,; and opera-singing
soloists. As in Handel's day, different approaches to interpretation and orna-
mentation abound. A recent New York Times review of a Christmas season of
Messiah presentations bemoaned the misunderstood and misused excesses in
ornamentation, both instrumental and vocal, in all but a few of the perfor-
mances.36 The Times critic's complaints of lack of good taste sound curiously
reminiscent of some seventeenth and early eighteenth-century writers, includ-
ing Tosi. Whatever performance situation you may fina yourself in, par-
ticularly for Messiah or other Handel oratorios, you will undoubtedly have to
compromise to get along with your musical colleagues, including singers, in-
strumentalists, and conductor. The more information you have about different
stylistic approaches, the more choices you can make in trying to achieve the
goal of singing with good taste.
GERMANY
Music in Germany had been under the influence of the Italian style even in the
seventeenth century, when Christoph Bernhard followed in the footsteps of his
teacher Heinrich Schütz and went to Italy to bring back singers for the Dresden
chapel. He also brought back the principles of secondapratticam his treatise of
1650, On the Art of Singing; or Manier?7 In the eighteenth century Italian
singers were still sought to adorn the operas and chapels of courts and noble
households in Germany, just as French dancing masters were employed to
teach the German aristocracy the most fashionable French court dances. While
Italian singing and French dancing exerted strong influences on music in Ger-
The Late Baroque 79
many, two other elements helped to give German Baroque music its own dis-
tinct style. One was the importance of Lutheran doctrine in church music and
education, as exemplified in the music of J. S. Bach. The other was the involve-
ment of Frederick the Great, king of Prussia, in the musical life of the court of
Berlin.
Frederick, who ascended to the throne in 1740, was an enthusiastic amateur
performer and composer. He studied flute with Quantz and performed with
C. P. E. Bach at the keyboard. Both musicians, along with Agrícola, were also in
his employ. Frederick enjoyed involving himself with every aspect of musical
life and production at his court, yet his energetic participation in musical mat-
ters led to strict rules and limitations governing composition and improvised
ornamentation. Some of these restrictions were specifically designed to prevent
tasteless excesses by Italian singers. Others may have been a result of Fredericks
generally conservative musical taste. The king ordered all ornaments in da capo
arias to be written into the music by the composer, including all appoggiaturas
and trills. He is said to have forbidden any improvised ornamentation, yet he
probably studied this very technique with Quantz. (One of Fredericks own
composed cadenzas and ornaments for an aria by Johann Hasse can be found in
the Wolff collection of vocal improvisations.38) Agrícolas interpretation of
Tosi s Italian style, particularly his discussion of ornamentation, reflects the
more restrictive approach characteristic of the Berlin School.
Ornamentation
APPOGGIATURAS OR VORSCHLAGS
The critical question regarding appoggiaturas has to do with the placement of

the ornamental note in relation to both the main note and the beat. Varying so-
lutions can be seen in different examples of German music from throughout
the late Baroque period. Agrícola, as well as Quantz, C. P. E. Bach, and Leopold
Mozart, translated appoggiatura as Vorschlag. This was not to indicate that the
ornament should come before the beat, but rather that the decoration should
come before the main note. Agrícola lists four reasons to add appoggiaturas: "i)
better to connect the melody, 2) to fill in the movement of the melody when it
seems somewhat empty, 3) to enrich the harmony, 4) to impart to the melody
more liveliness and brilliance."39 He also instructs that appoggiaturas should
be slurred to their main notes, and that when the main note precedes a rest, the
appoggiatura may take the entire value of the note, which then resolves on the
rest. An appoggiatura should not be added after a long pause, on the first note
80 The Late Baroque
of a work, or in a situation where parallel fifths or octaves would result. A short

appoggiatura that comes before the beat could be used in a passage where pass-
ing notes fill in descending thirds. Agrícola cautions singers and composers not
to use appoggiaturas to change the mood of a work: serious into brilliant or
lively into melancholy. The appoggiaturas must also not be used so consistently
as to become monotonous or too busy.40
Other ornaments which come before the main note include the Schleifer
(slide) and the Anschlag (compound or double appoggiatura). In both cases the
main note is preceded by two or sometimes more ornamental notes. In the slide
the auxiliary notes come from below or sometimes above. In the compound ap-
poggiatura the main note is surrounded by decorative notes from above and be-
low. Agrícola, as well as Quantz, C. P. E. Bach, and others, indicate that these
ornaments should be performed on the beat.
The alto aria "Erbarme dich" from Bach's St. Matthew Passion (ex. 2.8) illus-
trates a situation where this is not always the case. In this aria the slide orna-
ment is notated in a variety of different ways and poses some interesting prob-
lems of execution. It first appears as an ornament sign before the first downbeat
of the solo violin melody (see ex. 2.8a). (Neither the slide sign nor the appog-
giatura before the second beat appear in Bach's autograph score; they were
added to the original violin part by Bach's chief copyist, presumably under his
supervision.)41 In the first beat of the second full measure (not shown), a slide-
type figure is written out in the violin line leading the melody up to the second
beat, clearly coming before the beat, not on it. When the voice enters with the
same violin melody, no ornaments are indicated in the first measure, yet the
written-out slide before the beat occurs in mm. 16,17, and other places where
the voice sings the comparable thematic figure from m. 2 of the violin melody
(see ex. 2.8b). In m. 20, however, a sign for a slide appears in the voice part be-
fore the second beat, while in the instrumental line the violin plays a downward
written-out slide rhythm that starts on the second beat (see ex. 2.8c). This jux-
taposition may suggest that, at this particular moment, the slide in the voice
part should come on the beat as well in order to make a parallel rhythm with
the violin.
In the soprano aria "Zerfliesse, mein Herze" from Bach's St. John Passion, the
slide is used as a thematic figure throughout the aria, but it is always written out
in both the voice and instrumental parts and always comes on the beat (see ex.
2.9). In the duet "Wenn kömmst du, mein Heil?" from Bach cantata 140 (see
ex. 2.2), the slide figure is notated as an ornament sign in both the violin part
The Late Baroque 81
Example 2.8. J. S. Bach: "Erbarme dich," a) m. ι; b) mm. 16-17; с) mm. 19-20, from St.
Matthew Passion, NBA 11/5/179,180,181.
and the two voice parts whenever it appears. It could be performed on the beat
throughout, yet Neumann points out that in m. 33 of the duet (as well as in sev-
eral places in "Erbarme dich") an on-the-beat rendition of the ornament would
result in parallel motion to an octave with the bass line, voice-leading mistakes
"which were not likely to have been intended."42 John Butt counters this opin-
82 The Late Baroque
Example 2.9. J. S. Bach: "Zerfliesse, mein Herze," mm. 16-18, from St. John Passion, NBA
11/4/143.
ion by suggesting that the ear would not catch such a voice-leading mistake as
part of an ornament. He and other scholars therefore argue that ornaments
should not be subject to such strictness in the laws of harmony.43 Once again,
no single rule from a treatise can be applied to all situations. Each question
must be answered using the surrounding material as a guide.
TRILLS
Many writers describe what they consider to be a well-executed trill, but Agri-
cola is one of the few to give practical advice on how to achieve it. He recom-
mends practicing both a whole-step and a half-step trill (because some singers
find one easier than the other) and alternating the notes slowly at first as if they
were slurred and dotted, the lower note longer, the upper note slightly shorter
(ex. i.ioa). This approach helps ensure that the main note is sung louder than
the auxiliary note. The singer can then gradually increase the speed and dimin-
ish the dotted quality, all the while making sure to keep the intonation of the
two pitches clear and true. Eventually the singer should begin the trill on the
upper auxiliary note and possibly attach an ending or termination (ex. 2.iob).
The termination is usually performed at the same speed as the rest of the trill
without being newly articulated or detached. If the music is slow or serious,
however, the termination may be held slightly longer—particularly when fol-
lowing a cadenza (ex. z.ioc). Depending on the syllabification and the charac-
The Late Baroque 83
ter of the orchestral entrance after a cadenza, no termination may be neces-

sary.44
The speed and duration of a trill also depend on the situation. A slower, sad
aria could withstand a slower, longer trill. Slow trills were generally frowned
upon by Italians and Germans, who thought them to be typically French (even
though the French thought their own trills should be fast and light). Quantz
complained that an excessively rapid trill sounded like a flutter, and he pre-
ferred a trill of moderate speed.45 Agrícola describes a short or half trill for fast,
Example 2.10. Agrícola: trills horn Anleitung zur Singkunst, reproduced in Julianne Baird,
Introduction to the Art of Singing, Cambridge University Press, 1995, a) p. 129; b) p. 129; c)
p. 142; d) p. 134; e) p. 134; f) p. 132.
84 The Late Baroque
lively pieces (ex. i.iod). He also describes a mordent, which is more typically a
keyboard ornament. Similar to a half trill, it starts on the main note and alter-
nates with the note below (ex. i-.ioe). Other vocal trills that were primarily key-
board ornaments included prefixes either from above or below (ex. i.iof). Ide-
ally, a singer of late Baroque music should be able to master many different
kinds of trills—major, minor, short, long, rising, and descending, with prefixes
and suffixes—and know how to use them tastefully and effectively in the ap-
propriate situations.
). 5. Bach and His Singers
Johann Sebastian Bach wrote more ornaments into the fabric of his music than
many of his contemporaries. He was even criticized for writing notes normally
left to the performers discretion. As previously mentioned, his ornaments
could be notated in different ways according to the situation and were used
both as decorative and structural musical elements. In general, the complexity
of Bach's music leaves little room for improvised divisions or cadenzas. Simple
appoggiaturas or trills can be added to some passages. In arias instrumental
parts can guide decisions on small ornaments to complement thematic gestures
in obbligato lines. However, the text is most important in determining orna-
ments and articulation.
The main purpose of Bach's vocal music was to serve the Lutheran Church
and its doctrine. In eighteenth-century Germany music held a much more
prominent role in educational and religious life than it does today. It was a
powerful tool for transmitting the word of God to the community. Communi-
cating the text and conveying the proper "affect" had been important goals in
vocal music in the seventeenth century. In the context of Lutheranism, what
mattered more than tickling the ear and moving the passions was enhancing
the message of the sermon and saving souls. Thus it is crucial to consider the
text in shaping the tempo, character, dynamics, articulation, and ornamenta-
tion.
When making decisions about performing Bach's music, we may not re-cre-
ate the exact conditions under which his music was originally performed.
Knowing something about those conditions, however, can further our appreci-
ation and understanding of the music. While Bach was cantor in Leipzig, from
1723 until his death, he provided music for the Thomaskirche and taught the
boys at the adjoining Thomasschule. Responsible for the music at several other
churches in Leipzig as well, Bach used the schoolboys as singers and instru-
mentalists for various services throughout the week. It is thought that at that
The Late Baroque 85
time boys' voices changed much later than they do today, perhaps as late as six-
teen or seventeen, resulting in stronger, more experienced trebles and altos. It is
also possible that some of Bach's altos continued to sing in falsetto after their
voices had changed. Tenors and basses may have been older students at the
school or recent graduates, though little information on the identitiek of indi-
vidual singers exists. j
The number of Bach's singers has proved to be a fascinating and controver-
sial question. Probably only his best pupils sang his music at the Thomaskirche.
The more inexperienced ones performed less demanding music in the other
church services. A famous and much-discussed document known as the Ent-
\
wurff reveals that Bach barely had enough musicians to cover all the services,
especially taking into account absences due to illness or singers doubling on in-
struments. On August 23, 1730, he requested that the town council pay for
twelve or preferably sixteen singers in each choir instead of the usîal eight.
Joshua Rifkin has suggested that since Leipzig was not a wealthy towfi it prob-
ably could not afford as many musicians as Bach wanted. Rifkin hypothesizes
that most of Bach's choral works were performed with only one solo singer per
part.46 Scholars before Rifkin knew that Bach's cantatas were usually written
for both "concertists," who sang all the choral and solo music, and "ripienists,"
supplemental singers who filled out the larger choral sections. Since Rifkin s
revelatory articles and recording of the B-Minor Mass in the early 19868, no one
has unearthed sufficient evidence to prove him wrong, yet scholars today be-
lieve the question is still open for debate. Todays concert promoters and audi-
ences have not embraced the "one voice on a part" performance tradition, and
the convention of using chorus (small or large) and soloists (from within the
chorus or not) remains preferable. As with Handel's Messiah, when performing
Bach's choral works you may find yourself in a variety of performance situa-
tions. The most comprehensive discussion of all the relevant issues is Andrew
Parrott's The Essential Bach Choir.47
FRANCE
It is from Quantz, through his opinionated criticisms of both Italian and

French music and singing, that we get an interesting glimpse into s^me of the
elements that make up the early eighteenth-century French style. In the final
chapter of his Versuch, Quantz compares the three prominent national styles.
He proclaims that German mid-eighteenth-century music combines the best
elements of all the nationalities in a forward-looking, mixed or universal style.
86 The Late Baroque
On his way to that conclusion, he reveals several useful details about the rival
camps:
The French manner of singing is not designed, like the Italian, to train great virtu-
osos. It does not at all exhaust the capacities of the human voice. French arias have a
spoken rather than a singing quality. They require facility of the tongue, for pro-
nouncing the words, more than dexterity of the throat. That which should be added
in the way of graces is prescribed by the composer, hence the performers do not have
to understand harmony. They make hardly any use of passage-work, since they
maintain that their language does not allow it. As a result of the lack of good singers,
their arias are mostly written so that anyone who wants to may sing them; this
affords satisfaction to the amateurs of music who do not know much, but offers good
singers no particular advantage. The only distinctive quality of their singers is their
acting ability, in which they are superior to other people.48
Quantz explains that the French thought Italians placed too much importance
on passage work and vocal display at the expense of dramatic expression and
text. Quantz agrees with this opinion in some cases. The French also found Ital-
ian arias endlessly repetitive and too full of vulgar bravura. Yet within the con-
text of the secular cantata, the Italian style had an important influence on
French music and singing.
Secular Cantatas
After the death of Louis XIV the court was no longer the center of musical life
in France. In the salons of Paris, the poetic form of the cantata became popular
and found musical expression in the Italian-inspired cántate frangaise. These
secular cantatas were composed both for small informal gatherings and for
larger events in fashionable society. They were performed by both amateurs and
professionals. Like their Italian counterparts, French cantatas were written for
one or more solo singers and continuo, with or without obbligato instruments.
They usually consisted of three recitatives and three airs, portrayed some
mythological or allegorical story, and included flowery poetry, amorous entant_
glements, and a concluding moral. They became wildly popular in Paris in the
period between Lully and Rameau. Most composers of the time, including An-
dre Campra, Nicolas Bernier, Louis-Nicolas Clérambault, Montéclair, and
Rameau were drawn to the form. In his book The Eighteenth-Century French
Cantata, David Tunley presents a thorough introduction to this wealth of
repertoire, which he says "provides a fascinating study in musical style."49 He
also includes a list of facsimile scores and performing editions in appendix Aii
of the book.
The Late Baroque 87
Most cantatas were scored for a single voice, usually soprano or tenor, using
а С clef on the first line (soprano clef) or sometimes our more familiar G clef.
Some cantatas were scored for lower voices in alto and bass clefs, but transposi-
tion was commonplace. Instrumentation could be dictated but could also re-
main flexible, depending on the musicians available. The most common in-
struments used were violin and/or flute, harpsichord, bass viol, and/br cello.
Ornamentation
As Quantz mentioned, improvised diminution was less common in the eigh-

teenth century. Some appoggiaturas were written into scores, but most vocal
ornaments were indicated with a simple " + " or "t." The performer could de-
termine the specific ornament based on the dramatic context and the nature of
the syllable. In his 1755 Uart du chant, Jean-Baptiste Bérard, a leading Parisian
opera tenor and respected voice teacher, presents common diction mistakes
and shows the proper pronunciation of both vowels and consonants.?0 His rec-
ommendations are remarkably similar to those of Bacilly s Remarques a century
earlier. Bérard complains of the lack of ornamental notation and asks singers to
look within the character of the word to avoid choosing the wrong ¡ornament
for a particular passage. His discussion of specific ornaments is worth reading
but somewhat frustrating, owing particularly to the lack of notated musical ex-
amples.
Other useful sources for information on French ornaments include Michel
Correttes Le parfait maitre a chanter (1758) and Montéclair s Principes de
musique (1736). Tunley has compiled a table of vocal ornaments with musical
examples from Corrette on pages 205-8 of his The Eighteenth-Centμry French
Cantata. A-R Editions' volume of Montéclair's Cantatasfor One and Two Voices
includes a translation of the vocal ornamentation section from his Principes m
an appendix.
There are some discrepancies in the use and translation of certain terms
when describing French vocal ornaments. As mentioned in chapter i, Austin
Caswell translated Bacilly s use of cadence to mean "vibrato." Sidney Murray
translates Bérard s use of cadence to mean "trill," while Tunley specifies cadence
as "final trill." In James Anthony and Dirán Akmajian's translation, Montéclair
clarifies the term by distinguishing between cadence, which is "an ending or
conclusion of the melody which is to music what a period is to prose," and
tremblement, which is an instrumental term for "trill." "There are ^cadences or
conclusions of melodies without tremblements? Montéclair writes, "just as
there are tremblementswithout cadences"*^1 Trills can be long or short, with or
88 The Late Baroque
without a preparation or prefix. They can flow right into the continuing
melody or prepare the final note of a phrase with an anticipation. They can be
"well beaten" and clearly articulated or give a subtler shimmer to the line, de-
pending on the sense of the text. Montéclair also distinguished between bal-
ancement (tremolo) and flute (vibrato). Other subtler inflections and shadings
of a note included the son filé (spun sound), perhaps without any vibrato, and
the son enfle et diminué> comparable to a messa di voce. Once again, as with sev-
enteenth-century ornaments, it is more important for singers to know how and
where to use a particular ornamental gesture to shape a phrase or add expres-
sion to a word than it is to know what it is called by whom. Example 2.11 illus-
trates some of the most basic ornaments from Montéclair, Bérard, and Cor-
rette. Examples a-e are sometimes notated with small notes; these ornaments
can also be added in appropriate places where not indicated. Examples f-j are
usually written in scores as a "t" or " +."
In Monteclair's cantata Arlane et Backus (see ex. 2.3), a variety of ornaments
are indicated by small notes and " +." The appoggiatura on the downbeat of m.
16 should be stressed and on the beat to add tension and anguish to the word
"cru^/." The small note on the downbeat of m. 17 functions as the appoggiatura
to a prepared trill on the syllable "Minotawre." It should also come on the beat.
The small note attached to the second beat of m. 17 should come slightly before
the beat, forming a conclusion to the trill that leads into the unaccented silent
syllable "Minotaure." This is a common rhythmic gesture, particularly for set-
ting a final mute "e," similar to the passing appoggiatura in Italian and German
music. The trill on the downbeat of m. 18 is also prepared by a written appog-
giatura. In this case both the appoggiatura and the trill may be shorter and
faster than those of the previous downbeat, owing to the shorter syllable and
the need for dramatic emphasis on the word "mgrat." In contrast, the trill on
the downbeat of the following measure could be longer and slower, prepared by
a dissonant B-natural appoggiatura, to create the appropriate "affect" for Ari-
anes sadness ("douleurs"). One might even add a passing note, by itself or with
a trill, between the С and Α-sharp in m. 18 before the half cadence at "dou-
leurs." In fact, when the opening phrase is restated with the flute in m. 30, the
arrangement of ornaments is altered to include this figure.
Example 2.5, from Rameaus Le berger fidele, shows how various ornaments
can function in a recitative. In the first measure, the trill on "berg??"" can start
on the С of the preceding syllable and be quite short so as not to disturb the
flow of the line to the half cadence on the second measure's downbeat. Here,
though no ornament is indicated, you could approach the D on "présente" with
Example 2.11. French ornaments from Montéclair (in Cantatasfor One and Two Voices, A-R
Editions, 1978); Corrette (in David Tunley, The Eighteenth-Century French Cantata,
Clarendon Press, 1997); and Bérard (in L'artdu chant, Pro Música Press, 1969).
90 The Late Baroque
^ port de voix> either by itself or with a mordent. (In his annotated examples of
repertoire at the end of Uart du chant, Bérard adds many such small graces to
both airs and recitatives.) In m. 3, if the trill on "funeste" starts on the G of the
preceding note, it would give dramatic color to the word. At the full cadence in
m. 4, a coulé can be added on the beat. The expressive dissonance need not be
hurried here, because of the fermata. Likewise, in mm. 7 and 8 on "rare" and
"beau," prepared trills can highlight the words and fill the elongated note val-
ues. Or you might prefer to save the trill for "beau" and instead color "rare"
with a subtler inflection. As Quantz pointed out, the French style has a more
spoken than sung quality. Each of these ornamental gestures is a distinct event
that, when combined, will result in a collection of small, detached phrases. As
you become more familiar with French music from this period, you will recog-
nize similar situations in which to use standard ornamental gestures.
One of the important points presented in this chapter is the issue of tempo. In
Baroque music the tempo is often based on a combination of the meter, the
note values, and the tempo term. Popular court dances also influenced the
tempo and character of certain works based on those dances. Remember to take
all these elements into account when choosing tempos for Baroque music.
This chapter continued to reflect issues introduced in chapter i, including
choosing scores wisely and working with figured bass. Whenever possible, try
to find an Urtext or critical edition score that shows all the lines of counter-
point, rather than a piano reduction. Work with early instruments if you can,
and take advantage of their distinctive qualities of sound and articulation.
Singing at the lower pitch used in the eighteenth century will also allow you to
use subtler shadings of tone color, vibrato, and articulation.
Articulation is another significant element in late Baroque music. Remem-
ber that the music is made up of small pieces joined together like words in a
sentence. These small musical pieces can be combined to form phrases, clauses,
sentences, and paragraphs. Make sure to highlight the differences between the
stressed and unstressed, the heavy and light, and the dissonant and consonant
notes and phrases. Use the words to emphasize these contrasts. Remember that
a legato or cantabile approach was used at times but that the favored articula-
tion was separate and detached. Take advantage of the rhythmic differences in
the Italian, German, and French languages to give the music from each country
its own particular style of articulation.
Each of those countries took a different approach to ornamentation. You
must know both how to execute particular ornaments written into the music as
The Late Baroque 91
well as how and where to compose or improvise additional ornaments and ca-
denzas, if appropriate. As with seventeenth-century ornaments, the best way to
become familiar with the eighteenth-century variety is to research and study as
many examples of ornamentation from this period as possible, and from each
nationality. Once you are comfortable using what you have found, you can ex-
periment with creating your own ornaments in the same style. Remember that
arias from operas and cantatas were often transplanted, transposed, and recom-
posed to suit the needs of the particular singer who was singing them at a given
time. Feel free to take advantage of this practice.
Chapter 3 The Classical Era
This chapter considers the vocal style of the Classical era, including
the music of Haydn, Mozart, and Beethoven. The term "Classical"
was not coined until after 1830 and was not used b|y the composers of
the late eighteenth and early nineteenth centuries. Both performers
and composers of the time continued many of the traditions of the
late Baroque, while adapting their musical style to fit new social and
artistic circumstances. Charles Rosens landmark book The Classical
Style indicates 1750 as a boundary marking the end of the high
Baroque and 1775 as an important transition point for Classical mu-
sic.1 Sandra Rosenblum's Performance Practices in Classic Piano Music
documents late Baroque practices in use from 1690 to the 17405, early
Classical practices starting from the 17305 and reaching into the 17705,
and "mature" Classical style in place from the 17705 to the early
18205.2 This chapter examines how many important ideas from the
Baroque period continued to exert strong influence on vocal and in-
strumental music during this period of transition.
Performers and composers used and respected the mid-eighteenth-
century treatises by Quantz, C. P. E. Bach, and Leopold Mozart, until
92
The Classical Era 93
well into the nineteenth century. Treatises from later in the eighteenth century
by Daniel Gottlob Turk, Domenico Corri, Johann Adam Hiller, and Giambat-
tista Mancini contributed new thoughts on familiar topics (see "Chapter 3: Pri-
mary Sources" in "For Further Reading"). The rhetorical principles of music as
speech, inflecting the difference between good notes and bad notes or strong
and weak syllables, were still important ideas, as were using rubato and com-
municating the "affect" or emotional tone of a work. Dance forms, with their
characteristic rhythms and tempos, continued to play a significant part in mu-
sic of all kinds. Performers also continued to participate in the compositional
process by adding ornamentation. As in the past, the ornaments, whether small
graces or more elaborate improvisations, still depended on the situation and
the individual tastes of composer and performer.
Yet many things were changing, including the kinds of performance venues
and the people who were performing. During the period leading up to the
French and American Revolutions, the court patronage system was j gradually
replaced by the rise of a new middle class. Professional performances moved
out of the private domain of the aristocracy and into a more public ¿ealm that
sometimes demanded larger performance spaces. This in turn led to the need
for louder instruments and a more projected performing style for both instru-
mentalists and singers. At the same time more amateurs wanted to make music
in their homes, and the growth of the music publishing industry made it possi-
ble for more people to acquire performance materials. All these elements con-
tributed to the development of a new compositional style and a new interaction
between composers and performers to suit the changing circumstances. This
chapter will revisit issues of scores and notation, instruments, tempo, articula-
tion, legato, vibrato, recitative, and ornamentation, as well as examine how the
developments of the Classical era contributed to changes in performance prac-
tice.
Scores and Editions
In the Classical era, composers started to nótate their scores more carefully and
completely. As in the Baroque period, if the composer were performing himself
or leading a performance, usually from the keyboard or violin, it was possible to
omit certain details from the score. Issues of tempo, articulation, and orna-
mentation could be worked out in rehearsal or determined spontaneously in
performance. Today it is impossible to pinpoint definitive versions of most
eighteenth-century works owing to the flexible nature of performance situa-
tions, and the changes made by and for particular musicians. Yet as hotational
94 The Classical Era
practices became more sophisticated, composers tried to include more specific

performance instructions in their scores to assure a proper rendition. This
effort was not always successful, however, as we learn from Wolfgang Amadeus
Mozart's letter of January 14,1775, regarding La finta giardiniera: "Next Friday
my opera is being performed again and it is most essential that I should be pre-
sent. Otherwise my work would be quite unrecognizable."3
As the music printing industry grew, published scores became more com-
mon than manuscripts and were used by both professionals and amateurs. For
the growing new market of amateur musicians, composers prepared easy-to-
play and easy-to-acquire scores. The numerous folk-song arrangements for
voice and piano trio composed by Beethoven were conceived for performance
by amateurs. George Thomson, the Scottish publisher who; organized the pro-
ject, implored Beethoven to write the parts in an easy and familiar style so that
society ladies could play them. He even requested that some of the accompani-
ments be rewritten because they were too difficult.4 According to A. Peter
Brown, the first edition of Haydn's already popular The Creation was prepared
as a souvenir or collector's item and was never used by Haydn in performance.
Brown suggests that this published score, printed on special paper and super-
vised and signed by Haydn, was not necessarily an accurate record of the per-
formance tradition but was designed instead to appeal to the concertgoer or
amateur musician for home use or as a keepsake.5
Sometimes first editions followed the composer's manuscript closely. Often,
though, there were discrepancies such as last-minute alterations by the com-
poser or copying and engraving mistakes. Some publishers in London began is-
suing collections of popular arias with "realized" accompaniments and the or-
namentation, printed in small noteheads, used by famous artists. For some
works of Beethoven, no autograph manuscript or corrected copy exists. The
first edition, published either in his lifetime or shortly after his death, is the
closest to the original source available.
By the second half of the nineteenth century, Breitkopf & Härtel had pub-
lished the complete works of Mozart and Beethoven. A complete set of Haydn's
works followed in the first decades of the twentieth century. As mentioned in
chapter 2, the editors of these scores "fixed" inconsistent articulation and orna-
mentation markings, and they added phrasing marks to create the sense of long
line and style consistent with nineteenth-century performance traditions.
Many of today's modern editions of Classical works are still based on these
nineteenth-century editions. As with similar scores of Baroque music, these
nineteenth-century texts should be viewed with a degree of suspicion and com-
pared, when possible, with the more recent critical Urtext editions of Haydn,
Mozart, and Beethoven.
The following examples illustrate some of these discrepancies. In the old
Mozart complete works (WA. Mozarts Werke, edited by Ludwig von Köchel
and others), the song "Dans un bois solitaire," K. 308, has no tempo indication
at the beginning but does have an Adagio marked at m. 54 for the words "il me
blesse au coeur." This is reproduced in the Peters and Dover editions, with an
editorial Allegro added at the beginning of the song. In the new Mozart edition
published by Bärenreiter, Neue Mozart Ausgabe (NMA), the opening of the
song is marked Adagio, with no Adagio at m. 54. Changing the opening tempo
makes a big difference in the pacing of this song. Similarly, the Peters and
Schirmer piano-vocal scores of Haydn s "Nelson" Mass, based on the old Breit-
kopf & Härtel Haydn complete works (/. Haydns Werke, edited by Eusebius
Mandyczewski and others), show various articulation marks that were added to
the vocal line. These include long and short slurs and staccatos that do not ap-
pear in the new Haydn critical edition published by Henle. The Peters and
Schirmer scores have also realized some of the ornaments in ways that conflict
with Henle's Urtext edition. Finally, a Peters piano-vocal reduction of Mozart's
Sechs Notturni for two sopranos, bass, and three clarinets is full of crescendos,
diminuendos, and mp and ra/*markings, which Mozart rarely used. These dy-
namics are nowhere to be seen in the original sources, according to the NMA.
This does not mean that we should not add dynamic shading, articulation,
and phrasing to Mozart and Haydn. We should, but we must be aware of what
was originally indicated by the composer and what is an editor's suggestion.
The autograph manuscript is the best source for determining the composer's
intentions. Early editions can also offer valuable information, but a reliable Ur-
text edition with the latest critical scholarship will show what today's editors
believe to be the most accurate presentation of the score.
Classical Instruments
KEYBOARDS
The development of the fortepiano had a profound effect on the kinds of artic-
ulation that were central to Classical style. According to Sandra Rosenblum,
the fortepiano began to appear in public venues by the 17605; previously it had
been used only for private gatherings. During the 17705 the harpsichord re-
mained the instrument of choice in most situations, but the piano began to be
heard in some orchestral and opera performances. By the 17805, even though
the harpsichord was still used, particularly for continuo parts, the piano had
become the dominant keyboard instrument. Its more substantial sound and its
ability to vary dynamics and articulation inspired composers to write music
that took advantage of these innovations. Of course the fortepianos of the late
eighteenth century probably more closely resembled the harpsichords of that
time than todays concert grand pianos.
While undergoing changes and improvements throughout the period, the
early instruments varied from country to country. The technical differences be-
tween the early pianos of the English and Viennese schools influenced both the
style in which they were played and the kind of music composed for them. En-
glish pianos were generally larger and heavier than their Viennese counterparts.
Their hammer action produced a more powerful blow to the strings, and the
damping mechanism enabled the sound to ring a bit. English instruments also
sported larger hammers, wider keys, thicker strings, heavier cases, and more
notes on the keyboard. These features produced a louder, fuller, more sustained
sound compared with the lighter, clearer, quickly decaying sound of the Vien-
nese pianos, known for their sprechend (speechlike) style. Mozart was arguably
one of the greatest keyboard virtuosos of his day. His playing of the Viennese
fortepiano is described by his contemporaries as brilliant, full of character and
grace. Yet Beethoven, a renowned pianist himself, played both Viennese pianos
and a Broadwood piano from London and reportedly described Mozart's play-
ing as choppy, with no legato.6 This anecdote reveals that the variety of the pi-
anos available in the Classical era had an important impact on the taste and
style of the time, particularly regarding articulation and legato.
STRINGS
The development of a new violin bow by Francois Tourte in the 17805 had a
comparable impact on string playing. The new bow was heavier, with more hair
in a flatter, wider configuration. It was more balanced and could withstand
greater tension, allowing the player increased volume and brilliance. It was now
possible to sustain a longer line and approach a more legato style of playing. Yet
Classical string playing was still closer to the light, articulated style of the
Baroque than to our modern conception of a projected, sustained sound.
Baroque, transitional, and Tourte bows were all used in the late eighteenth cen-
tury, and it wasn't until the second decade of the nineteenth century that the
full potential of the Tourte bow began to be realized.
Both Jaap Schröder and Robin Stowell believe that some of the relaxed clar-
ity and elasticity of the Baroque string sound was sacrificed with the use of the
Tourte bow. The key to achieving a Classical style is to combine well-articu-

lated gestures, dynamic variety, and a clear, relaxed sound that is "intense but
not tense."7 This seems to be good advice for singers as well: When performing
Classical works with modern instruments, you may need to make compromises
regarding tone color, volume, and intensity to match the more powerful and
sustained instrumental sound. When singing with a fortepiano or with Classi-
cal strings, you can use a lighter, more relaxed tone, as well as a less sustained,
more articulated sense of phrasing.
Articulation and Legato
The eighteenth century witnessed a gradual progression along a continuum

from detached articulation toward a more legato style. In the Classical era, the
Baroque ideal of crisp, detached articulation, which highlighted the differences
between heavy and light and made the music speak as if it had text, was still im-
portant. In elaborating on Alfred Brendel's remark that a phrase in a Mozart pi-
ano sonata was like a vocal line, fortepianist Malcolm Bilson has commented
that "in Mozarts time a vocal line meant clear inflection between strong and
weak syllables, strong and weak beats, stressed versus unstressed. This is spelled
out clearly by Mozart's careful slurring and articulation marks."8 Dominico
Corn's vocal treatise of 1810 reveals an approach to Classical singing based on
his studies with Nicola Porpora, the famous teacher of Farinelli and Bernacchí.
While Corri acknowledges that the character and inflection of the words is of
utmost importance, the advice he gives for singing solfeggio without words
offers valuable clues for vocal articulation:
To give them meaning requires the rise and fall of the voice, which as I have before
explained, is, in Songs regulated by the sense of the words; Solfeggio not having the
assistance of the words, the following rules may be of use . . .
ist. On every note of any duration, use the Messa di Voce.

id. Every note ought to have, as it were a different degree of light and shade accord-
ing to its position.
3d. When the passages are, of gradual notes, Slur and join them with nicety; When
leaping passages, give a well articulated accent.
4th. On the last note of a passage, always die the Voice, and at each note of the final
phrase, end thus < > this swell must be done as gently as possible, only as much
to accent the sound, and immediately die it away.9
As the eighteenth century drew to a close, preference shifted away from de-
tached articulation toward a smoother, more legato flow. This new ideal was in-
spired by cantabile singing and manifested in the English pianos and Tourte
bow. In his Klavierschuleof1789, Daniel Gottlob Turk quotes С. P. E. Bach and
disagrees with his instruction from 1753 that notes that are rieither slurred nor
detached should be held for half their notated value. Turk recommends that for
notes played in the "usual manner," the finger should be lifted from the key
only a little earlier than required by the value of the note. He also suggests us-
ing a variety of articulations according to the tempo and meter. Faster, brighter
compositions demand a lighter, shorter touch, whereas serious, solemn, or pa-
thetic music demands a heavier, more sustained execution.19 By the end of the
century a number of new keyboard treatises, including one by Muzio Clementi
in 1801, recommended legato articulation: "When the composer leaves the
Legato and Staccato to the performer s taste; the best rule is, to adhere chiefly to
the Legato; reserving the Staccato to give spirit occasionally to certain pas-
sages."11 In the 1804 fourth edition of Leopold Mozart's Violinschule, the earlier
assumption that non legato was the "usual" manner of playing was revised in fa-
vor of a slurred, sustained sound for cantabile passages, "even more imperative
in an Adagio than in an Allegro."12 Neal Zaslaw observes that "this evolution
from a more detached style of playing to a more legato one means that in mu-
sic from the beginning of the period the slurs are the most important marks of
articulation whereas in music from the end of the period the dots and strokes
indicating detached playing have acquired that role."13
We must remember, however, that the "cantabile" legato ideal for music of
the late eighteenth century is nowhere near the long, continuous lines associ-
ated with music from the late nineteenth or early twentieth centuries. Corri re-
minds singers to shape phrases by using crescendo and decrescendo (ex. 3.1),
similar to the messa di voce used on one note. In vocal music, the text and dra-
matic context set the tone for articulation, and so it usually has fewer articu-
lation markings than instrumental music. But just as eighteenth-century in-
strumentalists used good singing as a guide for their style, modern singers can
also use Classical instrumental style as a guide for their approach to Classical
singing.
Tempo and Rhythm
DANCE FORMS
The traditional court dances of the seventeenth and early eighteenth centuries,
though no longer fashionable for dancing, remained well known to composers
and musicians in the Classical era. Each dance form had a physical association
Example 3.1. Corrí: from The Singer's Preceptor (1810), reproduced in ThePorpora Tradition, ed. Edward Foreman, Pro Música Press, 1968,
p. 65.
with a character and tempo that carried over into new musical contexts. In the
Baroque period each dance had been linked to a particular "affect" that colored
the entire piece. In Classical music the expression of dramatic gesture became
more flexible within a single composition. Now a composer could express more
than one mood or character by juxtaposing contrasting "affects" within one
piece, and a particular dance type could be used to express a variety of emo-
tions.14 For example, a minuet could be slow, creating a noble and stately at-
mosphere, or faster, with a lighter, more graceful feeling. The sarabande, which
had been slow and haughty in the Baroque, was now somewhat faster, almost
closer to a slow minuet, yet still serious and grand. The gigue was a fast country
dance in 6/8 with a rustic, peasant feeling. (Mozart's Don Giovanni offers many
examples of dances that form the basis of much of the vocal music.)
METER AND TEMPO TERMS
The central place of dance in musical life made eighteenth-century musicians

familiar with the connection between meter and tempo. Specific meters had
particular associations with a kind of movement related to a dance. A tempo
term in combination with a time signature could also create many different
possible degrees of relative speed. Allegretto 4/4 is a very different situation from
allegretto 2/4, which is different from allegretto alia breve or 2/2. Similarly, an
andante m 3/4 would move quite differently from an andante т з/8. 15 Accord-
ing to theorists of the time, the larger the note value described by the denomi-
nator of the time signature, the heavier and slower the movement should be;
conversely, the lightness and liveliness of the tempo would increase as the bot-
tom number increased. Thus 3/8 would be more lively than 3/4, and 2/2 would
1б
move with more weight than 4/4. Beethoven wrestled with the relationship
between note values, meter, and tempo as he recomposed his song "Klage,"
WoO 113. In the second version of the song (ex. з.2с and d) he added descrip-
tive tempo terms, changed the meter of the minore section, and offered these
observations:
That which now follows will be sung still more slowly, adagio or, at the most, an-
dante quasi adagio. Andante in 2/4 time must be taken faster than the tempo of the
song here. As it appears, the latter cannot remain in 2/4 time for the music is too slow
for it. It appears best to set them both in С time.
The first [part], m E major, must remain in 2/4 time, otherwise it would be sung
too slowly.
In the past, longer note values were always taken more slowly than shorter ones;
for example, quarters slower than eighths.
Example 3.2. Beethoven: "Klage," first version WoO 113, a) mm. 1-4; b) mm. 15-17; second
version WoO 113, c) mm. 1-4; d) mm. 14-17, Beethoven Werke XII/i/i8o, 182.
The smaller note values determine the tempo; for example sixteenths and thirty-
seconds in 2/4 time make the tempo very slow.
Perhaps the contrary is also true.17
Tempo terms discussed in treatises by Quantz, C. P. E. Bach, Leopold

Mozart, and Turk and in Jean-Jacques Rousseau's 1768 Dictionnaire de musique
fall into the following basic order from slow to fast:
largo—slow
adagio—moderately slow
andante—going; an easy walking pace between slow and fast
allegro—quick, or moderately fast
presto—fast18
Qualifying adjectives or diminutives can further refine the shade of the tempo.
For example, larghetto (a little slow) is less slow than largo; piu allegro or molto
allegro is faster than allegro-, and allegretto (a little fast) falls between andanteand
allegro. The qualification of andante created much confusion in the late eigh-
teenth century. Some thought piu or molto andante meant ¡"more going" and
therefore faster, while others understood it to mean slower. Andantino was the
most confusing of all, sometimes falling between larghetto and andante and
other times indicating a tempo faster than andante. In a letter to Thomson re-
garding the folk-song arrangements, Beethoven asked for clarification regard-
ing the uncertain nature of andantino-. "If among the airs that you may send me
to be arranged in the future there are Andantinos, please tell me whether An-
dantino is to be understood as meaning faster or slower than Andante, for this
term, like so many in music, is of so indefinite a significance that Andantino
sometimes approaches an Allegro and sometimes, on the other hand, is played
like Adagio."19
Classical tempos must thus be chosen with great subtlety, yet they have a
crucial impact on the flow and character of the music. To find the right tempo
for a Classical song or aria you must evaluate the general mood of the text and
music as well as the intricacy of the ornamentation. You should also take into
consideration the rate of harmonic movement, which proceeded more slowly
than in Baroque music owing to the increased importance of the four-measure
phrase.20 Keep in mind that early instruments, with their lighter, clearer
sound, can sometimes navigate faster tempos than modern instruments.
TEMPO FLEXIBILITY AND RUBATO
"Playing in time" was extremely important to Classical composers. Theorists

demanded it in treatises, and Mozart complained bitterly when his students or
colleagues did not keep strict time. Yet Mozart himself played with rubato in
slower, more lyrical movements, and he described his method in a famous let-
ter to his father on October 24, 1777: "Everyone is amazed that I can always
keep strict time. What these people cannot grasp is that in tempo rubato, in an
Adagio, the left hand should go on playing in strict time. With them the left
hand always follows suit."21
Both Turk and Corri explain similar approaches to tempo rubato:
There is something taken away (stolen) from the duration of a note and for this, an-
other note is given that much more. . . . from this it can be seen that. . . the tempo,
or even more, the meter as a whole is not displaced... for the bass voice goes its way
according to the meter (without displacement), and only the notes of the melody are
moved out of place. (Turk)22
It is a detraction of part of the time from one note, and restoring it by increasing the
length of another, so that, whilst a singer is, in some measure, singing ad libitum, the
orchestra which accompanies him keeps the time firmly and regularly. (Corri)23
The notion of "stolen time" goes back to Tosi, who described vocal rubato
for pathetic airs in just this way (see chapter i). Classical instrumentalists con-
tinued to be encouraged to listen to distinguished singers as a model for the use
of flexible rubato in slow or "cantabile" music. The adagio style allowed for the
inclusion of more extemporaneous ornamentation and a flexible tempo for ex-
pressive purposes. Allegro arias left less room for added ornaments and hence
maintained a stricter tempo. Contemporary accounts describe Beethoven's
playing as both rhythmically flexible and strictly in time. He chose to vary his
handling of the rhythm and tempo depending on the character of the music.
Certainly the notion of flexible tempo for Classical music is not as exaggerated
as what we would apply to late Romantic music, but it should be employed to
some extent, perhaps more than we might think.
THE METRONOME
Before the metronome came into use around 1812, eighteenth-century musi-
cians may have based certain tempos on the speed of walking or on the ticking
of a grandfather clock. Corri recommends that his vocalises and exercises be
practiced at an andante pace that can be determined by the pulse of a watch.24
In their book Interpreting Mozart on the Keyboard, Paul and Eva Badura-Skoda
mention another pendulum-based measuring device called the Rhein Zoll,
which was used in 1793 by an acquaintance of Mozart s to gauge the tempo for
Paminas aria "Ach, ich fühls." Converted into a metronome marking, the
tempo measured by this device would have been eighthi note = 138~48.25
Given that the aria is indicated as Andante 61% the combination of tempo term
(Andante meaning between slow and fast) and meter (6/8 suggesting a more
lively movement than 6/4) suggests that it should not move as slowly as it is of-
ten sung today. If the tempo measurement from 1793 is at all accurate, it sug-
gests an even faster tempo than might be imagined.
It is well known that Beethoven loved the metronome: "So far as I am con-
cerned I have long thought of giving up the nonsensical designations of Alle-
gro, Andante, Adagio, Presto; Mälzel's metronome gives us the best opportu-
nity to do this."26 While the metronome markings he assigned to his works
leave clear evidence as to his desired tempos, many of which|are on the fast side,
some can be puzzling and difficult to make work in performance.27 Johann
Nepomuk Hummel, Mozart s student, and Carl Czerny, Bbethovens student,
added metronome markings to various works of Haydn, ¡ Mozart, and Bee-
thoven, based on their supposed knowledge of the composers' preferred tem-
pos.28 Both Antonio Salieri, who played continuo for Haydn, and Sigismund
Neukomm, one of Haydn s pupils, wrote down metronome markings for the
tempos they believed Haydn used when he conducted The Creation2^ It is ex-
tremely difficult to evaluate how accurate these early markings are. Even today,
composers often misjudge their own music when assigning metronome marks.
While most vocal music from the Classical period did not receive metronome
markings, it is still worth considering the implications of early metronome
markings. The generally fast tempos assigned to instrumental music may well
reflect the prevailing eighteenth-century tempo preferences.!
RHYTHMIC ALTERATION
As composers included more carefully notated performance instructions in

their scores, they became more precise in notating rhythms that had been prob-
lematic in Baroque scores, particularly "mixed meters" such as two eighth notes
against three or dotted notes against triplets. As the eighteenth century drew to
a close, these rhythmic situations were understood to be more literal and less in
need of fixing or "alteration." In example 3.3 from Beethovens "Adelaide," the
eighth notes in mm. 7-8 should contrast with the triplets in the accompani-
Example 3.3. Beethoven: "Adelaide," op. 46, mm. 5-10, Beethoven Werke XII/i/25.
ment. In m. 9 Beethoven indicates the triplet rhythm in the vocal part that
should coordinate with the triplets in the piano part.
At the same time there are instances in the music of both Mozart and Haydn
where it is desirable to make rhythmic alterations, particularly concerning
overdotting or double dotting (described in chapter 2). Corri mentions that
different styles of music, from cantabile to bravura, take slightly different inter-
pretations of dotted figures. In music in the French style, which can be seen in
both Haydn and Mozart, overdotting is particularly welcome. For "national

airs" of Scottish, Irish, and Welsh character, Corrí recommends overdotting
and the use of the "Scotch snap" rhythmic figure, in which a dotted note is pre-
ceded rather than followed by its complementary shorter note.30
The Classical period saw the gradual decline of the castrato voice and the in-
creased use of female sopranos and mezzo-sopranos in opera and concert mu-
sic. Both Haydn and Mozart wrote music for castrati, but by Beethoven's time
i
castrati were much less popular. Women still did not sing in church (with the
exception of nuns in convents), and therefore castrati, adult male falsettists,
and boys continued to supply churches and concert choruses with treble voices.
Sometimes, as for performances of Haydn s The Creation,] which was usually
produced in a theater and not in a church, female soloists joined the male cho-
rus for the tutti sections.31 The tenor voice became more popular and was fea-
tured in more substantial opera roles. Tenors still blended their chest voice into
a falsetto for their high range. Some tenor parts from the pleriod extend lower
than would be comfortable for most tenors today and could be considered to
fall in the baritone range. The baritone voice was not comnionly distinguished
as its own category yet, however, and so low tenors navigated these roles using
their chest voice for the low range and head voice for the higher notes. The bass
voice was featured in comic and buffo opera roles, as well as in concert and or-
atorio, although not as prominently as the higher voices.
Sopranos, on the other hand, were singing higher and higher, as Mozart de-
scribed in a letter of March 24,1770. He was visiting the house of a famous so-
prano in Parma, and he jotted down her after-dinner vocal feats, which soared
to well above high C.32 Yet Corrí warns singers not to push the voice beyond its
normal compass. It is possible to extend the range, he says, but it is like the
stretching of a piece of leather: it must be done gently and carefully, or the ma-
terial will break. He recommends transposing music to suit the natural range of
the voice, sacrificing the low notes in favor of relaxed high notes.33 Transposi-
tion was still a common practice, so voices with in-between ranges could sing
comfortably.
Mozart on Singers
Mozart complained bitterly when he heard arias that didn't suit their singers.
He preferred to become familiar with a particular singer's voice and abilities be-
fore he composed, and to "fit the costume to his figure," as his father de-
scribed.34 The favorite singer of his early years, with whom he was also in love,
was Aloysia Weber. He loved her cantabile singing, which, he wrote, "goes to
the heart."35 This expressive quality impressed him much more than all the
runs, roulades, and general fireworks of famous divas singing bravura arias. In a
letter of July 30,1778, Mozart described to Aloysia how he wanted her to work
on the concert aria "Ah, lo previdi" that he had sent her: "I advise you to watch
the expression marks—to think carefully of the meaning and force of the
words—to put yourself in all seriousness into Andromeda's situation and posi-
tion!—and to imagine that you really are that very person."36 Of course Mo-
zart did compose his own fireworks for singers, particularly in The Abduction
from the Seraglio: "I have sacrificed Constanze's aria a little to the flexible throat
of Mile. Cavalieri. . . . I have tried to express her feelings, as far as an Italian
bravura aria will allow it."37 From this and other letters it is clear that he pre-
ferred a genuine dramatic expression of simple and honest feelings to empty
technical display or to overly sentimental, or false, cloying sweetness.38
Vibrato and Portamento
Mozart describes his taste for vibrato in relation to a bass singer in a letter of
June 12,1778:
Meisner, as you know, has the bad habit of making his voice tremble at times, turn-
ing a note that should be sustained into distinct crotchets, or even quavers—and this
I never could endure in him. And really it is a detestable habit and one which is quite
the contrary to nature. The human voice trembles naturally—but in its own way—
and only to such a degree that the effect is beautiful. Such is the nature of the voice;
and people imitate it not only on wind instruments, but on stringed instruments
too and even on the klavier. But the moment the proper limit is overstepped, it is no
longer beautiful—because it is contrary to nature.39
In writings from the Classical period we see more positive remarks concern-
ing vocal vibrato. It was still, however, a much different phenomenon from to-
day's wider, more continuous vibrato. It is difficult to determine from Mozart's
remarks above whether he meant that natural "trembling" occurred all the time
or only in special circumstances. String treatises recommend a finger vibrato
combined with a swell to imitate a singer's messa di voce on a sustained sound.
Flute players also used both finger vibrato and breath vibrato, mostly on long
notes and in slow, tender music, modeled after good singers. So we know that
some kind of small vibrato was occasionally used by singers and instrumental-
ists.
The term portamento received a number of different definitions and usages

in Classical period treatises. Primarily a vocal ornament, portamento was also
used sparingly by instrumentalists. While it is not what we associate with the
late nineteenth-century gesture, Corrí calls portamento di voce the "perfection
of vocal music": "It consists in the swell and dying of the voice, the sliding and
blending one note into another with delicacy and expression—and expression
comprehends every charm which music can produce; the Portamento di voce
may justly be compared to the highest degree of refinement in elegant pronun-
ciation in speaking. Endeavor to attain this high qualification of the Porta-
mento, and I must again repeat, deliver your words with energy and emphasis,
articulate them distinctly."40
The fact that Corri mentions portamento in association with clear articula-
tion of the text implies that it was used in the service of the words, rather than
for vocal show and heightened general expressivity as we would understand it
today. Charles Burney, in The Present State of Music in France and Italy, de-
scribed the term as "conduct of the voice: the portamento is said to be good,
when the voice is neither nasal nor guttural." In describing French singers
whom he disliked, he said "the French voice never comes further than from the
throat; there is no voce di petto, no true portamento or direction of the voice."41
From these remarks it seems that the term in the eighteenth century had more
to do with a general kind of placement of the voice, a connection of breath and
tone that produced a pleasing flow and helped to make the text clear and in-
flected, rather than an audible slide from one note to another.
Recitative
The rules for Classical recitative remain similar to those for recitative in
Baroque music. The words are the most important element for determining the
rhythm, tempo, and character of the music. According to Corri, recitative "re-
sembles speaking in Musical notes—but mere description cannot fix the exact
measure of that inflection of tone necessary to be used, which should be a
medium of sound between speaking and singing. . . . No particular degree of
Time is marked to Recitative, but is left to the Singer to prolong or shorten
notes which he ought to do agreeable to the passion and accent of the words."42
Mozart heard several melodramas performed in 1778 that delighted him: "there
is no singing in it, only recitation, to which the music is like a sort of obbligato
accompaniment to a recitative. . . . I think most operatic recitatives should be
treated this way—and only sung occasionally, when the words can be perfectly
expressed by the music'."43
Most theorists continued to instruct that recitatives for church, chamber,

and theater be delivered differently. According to Corrí, church recitatives re-
quire a slower, more serious delivery with few if any ornaments. Theater recita-
tives require a pace appropriate to the dramatic action and need no ornaments
except an occasional mordent or short trill. Chamber or concert recitatives
could include more ornaments, particularly at fermatas or on final badences.
Hiller (1780) distinguishes between recitatives that are secco (dry) and those that
are accompagnato, or accompanied by instruments or composed in an arioso
style. Accompagnato recitatives can be more sung than declaimed arid can in-
clude more elaborate and expressive ornamentation. As for the acbompani-
ment, according to an item in the Leipzig Allgemeine Musikalische Zeitung of
1810, the Baroque practice of shortening the bass notes was considered old-
fashioned. From the newspaper's imaginary dialogue between representatives
of the "old way" and the "new way," it is clear that "modern" Classical com-
posers notated their works as they intended them to sound, and therefore the
bass notes would be sustained for their full value.44
APPOGGIATURAS IN RECITATIVE
Singers were still expected to add appoggiaturas within the body of a recitative
and at cadences even if they were not written in the score. Standard practice in-
cluded filling in descending passing notes between thirds and repeating the
penultimate note at cadences approached by a falling fourth (see ex. 2.4). In
this situation Hiller also allowed jumping down a third to the uppei neighbor
note.
Much disagreement, however, exists among modern scholars regarding the
general treatment of appoggiaturas in words with feminine endings (unstressed
final syllables). Will Crutchfield offers evidence from the period to support the
ideas that a "blunt ending" (two of the same pitches in a row) would always be
undesirable and that any feminine ending, whether in a recitative or an aria, de-
mands an appoggiatura.45 Frederick Neumann, on the other hand, contends in
his Ornamentation and Improvisation in Mozart that contemporary treatises in-
dicate that additional appoggiaturas were optional, depending on the situation
and the meaning of the words. He suggests that note repetition could convey
insistence, determination, anger, or defiance, while an added appoggiatura in
such a situation would soften or weaken the dramatic effect.46 Appbggiaturas
from below were much less common than those from above, though one could
be used to create a yearning quality or a question. It is also possible to jump to
an upper neighbor note to create a falling appoggiatura. For words with mas-
culine endings (stressed final syllables), Neumann recommends an unorna-

mented approach in most cases, and Hiller (1774) advises avoiding monotony
by not adding too many appoggiaturas.
Overall, added appoggiaturas in recitatives can help shape the inflection of
the text. In deciding where to use them, let the dramatic intent of the words be
a guide. In the following lines from the recitative preceding Mozart s concert
aria "Bella mia fiamma," K. 528 (ex. 3.4), I would choose not to fill in the third
between "Ah" and "tutto" with a C-natural, even though it is possible. I would
rather not use an appoggiatura to weaken the strong impact of "tutto," and I
would save the passing note С for the word "mio" leading to "furor" because it
adds more tension to the sense of "my fury." I would also not add an appog-
giatura to the E-natural of "furar," because of the angry feeling of "fury"; in a
different textual and dramatic context, I might want to soften the masculine
ending with an F appoggiatura. In contrast, I would add an appoggiatura to the
first syllable of "mio" in the next measure, either a G lower neighbor or a B-flat
upper neighbor, to add a more anguished, sighing feeling to "my death." (The
NMA edition suggests the B-flat.) In the next phrase, "Cerere" could take a C-
natural upper neighbor on the first syllable, "Alfeo" could take an F in place of
the first Ε-flat, the first syllable of "sposa" should take a D-natural to fill in the
third from the E of "diletta" (the NMA suggests that one as well), and the raised
fourth for the cadence is already written into the "addio." Singing all of these
appoggiaturas consecutively might be too monotonous, as Hiller warns, so I
would probably leave out either the one on "Cerere" or on "Alfeo."
Ornamentation
As in the Baroque period, ornaments in Classical music fall into two basic cat-
egories: essential graces and added improvisation. Throughout the eighteenth
century, singers wanted to show off their vocal talents by adding embellish-
ments and cadenzas to the score. In many instances, composers expected and
appreciated these personal contributions. Yet singers could just as easily go far
beyond the bounds of what composers wanted. Contemporary evidence reveals
accounts of both dazzling ornamental display and horrid excess. Treatises of the
period advise using taste and good judgment and warn against abuse, yet sur-
viving manuscripts with added ornamentation show virtuosity ranging from
the modest to the unimaginable. Styles varied from region to region, with Ital-
ian music and singers using the most elaborate and rhythmically flexible orna-
ments. German works were more restrained and specific, and tended to include
more ornamentation composed directly into the music. Christoph Willibald
The Classical Era Ш
Example 3.4. Mozart: "Bella mia fiamma," K. 528, mm. 33-37, NMA II/7/iv/4O.
Gluck stirred up controversy in France with his harsh critique of operatic prac-
tices that put the vocal display of singers above the simple and honest flow of
the drama. Singers today must find a way to reconcile these conflicting ele-
ments and decide how to incorporate the vocal traditions of the time with the
composers' desires.
As in recitative, the choice or execution of a particular ornament should be
innately connected to the dramatic sense of the text. In Classical vocal music,
embellishments such as appoggiaturas, trills, and turns can be variously indi-
cated with ornament signs, small notes, or regular-sized notes or added to an
unadorned line at the discretion of the singer. Often an accompanying instru-
mental line can suggest either how to execute an indicated vocal ornament or
how to add a complementary one. In certain situations it may be desirable to
coordinate the two lines exactly. That is not always necessary, however, and it
may even suit the dramatic context to have contrast. Piano reductions of or-
chestral scores often simplify instrumental rhythmic and ornamental notation,
so when in doubt, consult a reliable full score.
APPOGGIATURAS IN CONCERTED PIECES
The appoggiatura, the most popular and important of the essential graces, can
be indicated either with a small note of varying value or written into the music
using a regular-sized note. This ornament was notated inconsistently even in
the late eighteenth century. The execution of the appoggiatura can be handled
in a number of ways as well. Unlike Baroque appoggiaturas, which are most of-
ten stressed and on the beat, Classical appoggiaturas can be variously short and
on the beat, short but before the beat, or long and on the beat.
In the accompanying examples from "Adelaide," op. 46, appoggiaturas are
notated in several different ways. In m. 13 (ex. 3-5a), Beethoven writes out a
short appoggiatura using two regular eighth notes. In mm. 15 and 17, he writes
out long appoggiaturas using two regular quarter notes. In m. 19 he writes a
small eighth note next to a large quarter note; this could be performed either as
two eighth notes to match the articulation of "zittert" in m. 13 or as two quarter
notes to match the gestures in mm. 15,17, and 21. The mid-eighteenth-century
treatises comment that an appoggiatura before a rest can take the full value of
the note and resolve on the rest, although the Badura-Skodas maintain that
Mozart never applied this rule.47 "Adelaide" also has other interesting elements
to consider. The appoggiatura in m. 19 is the first one on a single-syllable word;
it is also the first to be approached from below. This may be the reason
Beethoven wrote it as a small note instead of with a regular note value. Mm. 105
Example 3.5. Beethoven: "Adelaide," op. 46, a) mm. 12-21; b) mm. 104-105,145-146; c)
mm. 59-61, Beethoven Werke XII/i/25-зо.
and 146 (ex. 3.5!)) show another inconsistency. In the earlier measure Beethoven
writes out an appoggiatura as quarter notes. In the later measure the same mu-
sic is written as a small quarter note before a half note. Here Beethoven indi-
cates the desired execution of the ornament seen elsewhere in the piece. Clues
Example 3.6. Mozart: "Das Veilchen," K. 476, mm. 1-39, manuscript Zweig 56, used by
permission of the British Library.
to appoggiatura lengths can also be found in the text. The small sixteenth notes
in mm. 59 and 61 (ex. 3.50) should reflect the "rustling evening breezes" of the
words: thus they must be sung fast and light and before the beat as grace notes.
According to Neumann, Mozart varied his notation to indicate different
lengths of appoggiaturas: a quarter-note appoggiatura should be a long note; an
eighth note could be long or short; and a sixteenth or thirty-second note usu-
ally indicates a short note.48 The Badura-Skodas prefer to distinguish between
"stressed" and "unstressed" appoggiaturas in Mozart, rather than long and
short, and they counsel that the accentuation of the appoggiatura is more im-
portant than its actual length.49 The harmonic context can also have an impor-
tant influence on whether the appoggiatura should be stressed or unstressed. If
the appoggiatura adds expressive dissonance to a situation, it can be stressed. If
the main note is itself a dissonant passing note or upper neighbor (as in ex. 3.5C
from "Adelaide"), then the appoggiatura should not distract from the harmonic
tension already present.
In the opening of Mozart's "Das Veilchen" (ex. 3.6), the autograph clearly
shows eighth-note appoggiaturas. Since they each add a dissonant upper neigh-
bor to their main note, I would sing all of them on the beat and gently ¡stressed,
but not too long. Here is a situation where the vocal part and the accompany-
ing line do not have to coordinate exactly. Some modern scholars and editors
have latched onto the rule that an appoggiatura before a dotted note gets two-
thirds the value of the main note. They would suggest therefore that the ap-
poggiaturas on " Wiese' and "««bekannt" should last longer than the Others: an
eighth note, with the resolution only taking a sixteenth. A rule is less impor-
tant, however, than the specific musical context, and in this situation I would
rather inflect the appoggiaturas to convey the dramatic sense of the text: a sim-
ple, innocent flower standing unobtrusively in a meadow.50
Appoggiaturas can also be added where they are not indicated. As previously
mentioned, many blunt endings and masculine endings in concerted pieces
can take added appoggiaturas. Appoggiaturas indicated in an accompanying
instrumental line may also be added to an unadorned vocal line.
TRILLS AND TURNS
The notation of trills in Classical vocal music is more consistent than other or-
naments; they are usually indicated simply as tr. That one sign, however, can
indicate a number of different trills, all depending upon the situation: prepared
or unprepared; long or short; slow or fast; and with a variety of concluding
suffixes. The upper-note start, or the prepared trill, was accepted for most
Baroque trills and continued to be popular well into the nineteenth century.
However, many passages in Classical music demand an unprepared trill with a
main-note start. As with appoggiaturas, the variables in the performance of a
trill should be determined by the musical and dramatic context. In general, ca-
dential trills, like those in example 3.7, can take an upper-note start.
A main-note start is preferable when the addition of an upper-note appog-
giatura would confuse the harmony, melody, or rhythm. In example з.8а from
The Creation, the melody includes a passing-note figure, even without the trills.
This figure stresses the upper note, creating a sighing quality. The addition of a
gentle trill is meant to imitate the soft cooing of the doves. Starting the trill on
the note above the upper note would weaken this effect by making the whole
figure too busy.
In example 3.8b a main-note start would simplify a complex situation.
Melodically, the trill note is already decorated with a lower neighbor. It is also
approached from the opposite direction by an instrumental line, which then
echoes the vocal line. A main-note start would thus assure the clarity of the
melody and the counterpoint within the relatively fast-moving tempo of Alle-
Example 3.7. Mozart: a) Exsultate, jubilate, K. 165, mm. 62-65, NMA1/3/157;

b) "Domine" from Mass in С Minor, K. 427, mm. 26, 41, NMA I/i/v/46.
The Classical Era ΊΊ7
Example 3.8. a) Haydn: "On Mighty Pens" from The Creation, mm. 68-69, Novello, 1904;
b) Mozart: "Domine" from Mass in С Minor, K. 427, mm. 14-15, NMA I/i/v/45.
gro moderate. Most modern scholars agree that trills approached melodically by
51
stepwise motion can take a main-note start. Frederick Neumann offers sensi-
ble advice: If the trill is taken away and the plain melody can use the addition of
some kind of appoggiatura, then the trill can be started with the corresponding
type of preparation. If not, then a main-note start is probably advisable.52 This
Example 3.9. a) Mozart: "Laudamus te" from Mass in С Minor, K. 427, mm. 31-34, NMA
I/i/v/35; b) Haydn: "With Verdure Clad" from The Creation, mm, 7-8, with possible turn
executions; c) Haydn: "With Verdure Clad" from The Creation, m. 2,0, Novello, 1904.
test can also be used in deciding whether to use a suffix. Most long trills de-
mand a concluding turn, but it can be omitted from a short trill if it would dis-
turb the melodic flow.
Turns can be indicated with an ornament sign or written out with small
notes, which is more common in vocal music. In example з.9а Mozart adds six-
teenth-note appoggíaturas in mm. 32 and 34 of "Laudamus te" from his Mass in
С Minor. In the first measure the small notes transform the figure into a writ-
ten-out turn. In the second measure they indicate a pattern of four equal six-
teenth notes. Singers were allowed a certain amount of freedom in the rhyth-
mic execution and placement of a turn to accommodate the needs of drama
and diction (see ex. 3-9b). Singers were also allowed more expressive freedom
with the placement of a Schleifer (slide), which could come on the beat, before
the beat, or somewhere in between, depending on the needs of the text (see ex.
3.90).
IMPROVISATION
Most performers of the late eighteenth century had a much more complete mu-
sical education than performers do today. Training in harmony, counterpoint,
and composition gave performers the needed skills to prepare or improvise em-
bellishments in a style complementary to the music. The adagio style left room
for singers to add trills, turns, appoggiaturas, and diminutions, the smaller note
values that fill in a simple melody line without necessarily changing its contour.
The allegro or bravura style could also use added appoggiaturas, trills, and
diminutions, but within a more active and exciting rhythmic context. Hiller
(1780) distinguishes between pulling, dragging ornaments (gezogene) for the
pathetic style and pushing, thrusting ornaments (gestosseneri) for the allegro
style. He and Mancini (1777) both recommend recomposing passagework in a
bravura aria to suite the taste and ability of the singer. They describe what is
written in the score as a mere sketch or outline, a point of departure for the
imagination of the singer.53 As the rondo form eclipsed the popularity of the
da capo aria, the return of the principal melody became an important opportu-
nity for singers to display their ornamental and compositional skills.
The following examples from an aria by Luigi Cherubini with prepared vari-
ations by the famous castrato Luigi Marchesi show the difference between ada-
gio and allegro ornamentation, as well as the dazzling array of virtuosic pos-
sibilities within each style. Example 3.10 shows a brief excerpt of possible
ornamentation for the opening adagio section; example 3.11 shows a number of
variations on the first few measures of the rondo theme.54 Marchesi was greatly
Ί20 The Classical Era
admired in his day, even by Burney, who found him "elegant and refined," al-
though he was occasionally criticized for excessive ornamentation in recitatives.
Fermatas also provided a showplace for cadenzas of varying lengths. Corri
offers suggested generic cadenzas in every possible key, some with as many as
fifty notes of scales, turns, and trills. These would be used at a final cadence or
at a pause within an aria, perhaps before the reprise of the rondo theme (see ex.
3.12). Burney harshly condemned the excessive cadenzas of some Italian singers
Example 3.10. Marchesi: Cherubini variations on opening adagio measures, reproduced in

Robert Haas, Aufführungspraxis der Musik: Handbuch der Musikwissenschaft, Akademische
Verlagsgesellschaft Athenaion, 1931, p. 225.
Example 3.11. Marches!: Cherubini variations on rondo theme, reproduced in Robert

Haas, Auffuhrungspraxis der Musik: Handbuch der Musikwissenschaft, Akademische
Verlagsgesellschaft Athenaion, 1931, p. 230.
he heard: "This fault is general throughout Rome and Naples, where such a
long-winded licentiousness prevails in the cadences of every singer, as it is al-
ways tiresome, and often disgusting; even those of great performers need com-
pression . . . A few select notes with a great deal of meaning and expression
given to them, is the only expedient that can render a cadence desirable."55
NB. The following Cadenzes may be Shorted by omitting the Notes under this
Signe If they are found toolong or too difficult
REPRISE
A Pause of Suspenstion
Example 3.12. Corrí: cadenza suggestions from The Singer's Preceptor(1810), reproduced in ThePorpora Tradition, ed. Edward Foreman, Pro
Música Press, 1968, pp. 75-76.
Other fascinating examples of period ornamentation dan be seen in Will

Crutchfield s article "Voices" in the Classical section of Performance Practice:
Music after 1600. It includes ornamented music by Giuseppe Sarti, Nicola An-
tonio Zingarelli, Domenico Cimarosa, and Corrí > as well as an extensive exam-
ple of Mozart's suggested ornaments for J. C. Bachs aria "Cara la dolce fi-
amma." Mozart's ornamented versions of his own arias ¡Ah se a morir mi
chiama," K. 293, from Lucio Silla and "Non so d'onde viene," К. 294, written
for Aloysia Weber, can be found in Neumann's Ornamentation and Improvisa-
tion in Mozart. While the J. C. Bach aria offers an invaluable example of the
best kind of adagio ornamentation, it is interesting that Mozart's ornaments for
his own music are much more simple and modest than what he added to the
J. C. Bach aria, Haydn's ornamentation of his own music! can be seen in the
tenor aria "Quando mi dona un cenno" from the oratorio // ritorno di Tobia in
the Henle complete works critical edition.56 Ornamentation in surviving man-
uscript sources for The Creation can be found in A. Peter Brown's Performing
Haydn's "The Creation." Here it is again illuminating to note the differences be-
tween the extensive and elaborate Italian-style additions to the Tobia aria com-
pared with the conservative ornamentation for The Creation.
Many modern scholars concur with some early nineteenth-century critics:
examples of early Classical vocal music by second-echelon composers can be
considered skeletons to be filled in by the imagination of thej performers. How-
ever, the mature works of the greatest composers of the Classical era represent a
more complete working out of the musical material and hence need much less
in the way of ornamental additions. For the music of Haydn, Mozart, and
Beethoven, significant disagreement exists among modern performers and
scholars over how much and what kind of ornamentation arid improvisation to
employ.57 Singers today must study the surviving examples of Classical orna-
mentation and also become familiar with the general patterns and gestures used
throughout the Classical repertoire in order to choose those "few select notes"
that Burney preferred.
Important issues revisited in this chapter include articulation, tempo, orna-

mentation, and the need to choose scores wisely. Many editions of Classical
music based on nineteenth-century sources contain mistakes and discrepan-
cies; they also may include too many editorial additions regarding dynamics,
tempo, and articulation. Whenever possible, consult an Urtext edition to ob-
tain the most recent scholarly commentary.
Articulation in Classical music was still under the influence of the early eigh-
teenth century's crisp, detached approach. Legato was becoming more popular,
but the prime performance objective should be to reveal the differences be-
tween light and shade, or stressed and unstressed syllables and musical gestures.
If you can work with a Viennese fortepiano or Classical strings, their lighter,
clearer texture will help you achieve the delicate contrasts so necessary for Clas-
sical music. When choosing tempos for Classical repertoire, keep in mind that
the meter and the note values have an important influence on the flow of the
music. Dance forms, the text and general character of the work, and various
tempo terms and metronome markings should also factor in the choice of
tempo. Using subtly flexible rubato is appropriate in slower music, but keep the
bass line steady as you gently shape the melody line.
Classical music demands significant but well-chosen ornamentation. You
must know how to execute small graces written into the music as well äs how to
add appoggiaturas, trills, and turns in recitatives and other appropriate places.
You can also improvise diminutions and cadenzas. As with Baroque ornamen-
tation, you should study as many examples from period sources as you can and
then aim to prepare your own variations in the same style.
Chapter 4 Italian Bel Canto
This chapter discusses the Italian vocal style of the early nineteenth
century, including music by Rossini, Bellini, and Donizetti. This
repertoire is commonly called bel canto, a term with a tangled web of
meanings and associations. Literally, it means "beautiful singing,"
which is probably how it was used in seventeenth- and eighteenth-
century vocal treatises. In the mid-nineteenth century the term took
on a larger significance. The 1838 publication of a collection of songs
by Nicola Vaccai entitled Dodici ariette per camera per Vinsegnamento
del bel canto italiano may have been the first time the term was seen in
print in this particular usage.1 Later in the nineteenth century, as part
of the wave of nationalism that swept through Europe, "bel canto"
was used to contrast the Italian vocal approach with the German
declamatory style of the Wagnerians. Ferdinand Sieber's collection of
eighteenth- and nineteenth-century songs entitled II bel canto (Berlin,
1887) was offered as an antidote to "offensive shrieking of dramatic
singing" in the German style.2 In dictionaries after 1900 the term of-
ten describes the florid vocal style of the eighteenth-century Italian
Baroque, epitomized in the art of the castrati. German musicologists
126
Italian Bel Canto 127
of the early twentieth century extended the definition to include seventeenth-

century Italian vocal repertoire and style as well.
Some writers and voice teachers insist that bel canto is a vocal technique;
others believe it to be a conceptual approach characterized by smooth, flowing
melodies. Still others say it describes a general approach to singing that culmi-
nated in the florid style of Rossini. In a conversation from 1858, Rossini himself
expressed his requirements for a bel canto singer as having a naturally beautiful
voice, blended registers and an even tone from the low to the high range, dili-
gent training resulting in effortless delivery of highly florid music, and a com-
mand of style not taught but rather learned from listening to the best Italian
singers. It is not clear from the context whether he was using the phrase in its
literal or more general meaning, but he did regret that the tradition Was already
dying out by the middle of the nineteenth century.3
The early nineteenth century saw the continued growth of a middle-class au-
dience hungry for diversions and entertainments of all kinds, including music.
Public concerts increasingly required larger halls, which in turn required louder
orchestras and soloists. The greater demand for performers resulted in the ex-
pansion of the conservatory system, which now provided more thorough and
systematic instruction to both instrumentalists and singers. Choral societies
flourished as more people, including women, were invited to sing in choruses
not affiliated with churches or choir schools. The public eagerly followed the
latest new symphonies, concertos, operas, and star performers, and wanted to
be able to play their favorite melodies at home. Private concerts took place in
fashionable households, including music making by amateurs and appearances
by celebrities. Virtuoso musicians such as violinist Niccolo Paganini and pi-
anist Franz Liszt also took to the road, performing across Europe.
As the opera industry grew, practices that had begun in the seventeenth cen-
tury gained momentum and popularity. New Italian operas were Repeated in
different cities as word of their success spread and audiences demanded their
own productions of the latest hits. More operas were thus being regularly per-
formed by singers other than those for whom they were originally written. It
had been common in the eighteenth century for singers and cojnposers to
make changes and substitutions in an opera production, and in the early nine-
teenth century composers continued to help fashion changes and adaptations
for different singers and new productions. Increasingly, however, the composer,
who typically was paid far less than the prima donna, was not involved in pro-
ductions following the premiere or had little control over the whims of star
singers.
128 Italian Bel Canto
This book focuses primarily on vocal repertoire outside the realm of opera.
In the case of the early nineteenth-century bel canto period, however, the de-
finitive technical and stylistic decisions regarding new works were made by
opera singers, collaborating with composers. While the musical examples for
this chapter are chosen mostly from the nonoperatlc repertoire, we must con-
sider the early nineteenth-century opera tradition to understand the singing
style of the period.
Scores and Editions
Because composers now had less to do with the performances of their works,
they began to mark their scores more carefully, aiming to ensure a reasonable
performance without their presence. Italian composers were a bit more casual
about this need than their more exacting German neighbors. As opera produc-
tions took to the road, publishers became increasingly involved in the preserva-
tion of the composer s wishes by producing carefully printed sets of orchestra
parts that were rented to local opera companies. The wide availability of piano-
vocal scores, however, enabled the more unscrupulous to reorchestrate the mu-
sic and produce operas while avoiding the rental fees for the published parts.
Donizetti complained bitterly to his publisher Giulio Ricordi (founder of the
foremost publishing house of Italian music) about these kinds of abuses. By
1840, copyright laws prevented such piracy, and the publishers kept track of
which operas were being performed where and by what singers.
Publishers also issued much new music for home use by amateurs. Since the
only way for most concertgoers and opera lovers to hear their favorite music
again was to play it themselves on their pianos at home, publishers, with the
help of composers, prepared editions of popular arias, sometimes with the or-
naments of famous singers included. Arrangements of well-known orchestral
pieces for piano, piano duet, or chamber ensemble were also popular. The bel
canto composers also wrote songs and miniature versions of arias for perfor-
mance in the home.
In general, first editions of songs or piano reductions of opera scores from
this period are reasonably reliable expressions of the composers' wishes. Many
modern piano-vocal scores of Bellini and Donizetti operas are essentially
reprints of early- or mid-nineteenth-century publications. Facsimiles of various
opera score manuscripts, with fascinating editorial commentary by Philip Gos-
sett and others, are available in some libraries. Some scholarly work has been
done on Donizetti, and a critical edition of Bellini s works is in preparation but
not yet available. Ricordi s modern editions of Bellini s Composizione da camera
(chamber songs) are usually based on their own previous or original editions,
either as exact reprints or lightly edited. A semicritical edition of Bellini songs,
edited by Francesco Cesari, was published in Milan in 2000. Ricordi is in the
process of publishing new critical editions of Rossini's works, edited by Philip
Gossett, Bruno Cagli, and Alberto Zedda, Many of the Rossini operas have
been issued, and some piano-vocal reductions are now available; many include
appendixes with extra variants or cadenzas; the volumes of songs have not yet
been published.
Thus, for early nineteenth-century Italian bel canto repertoire you don't
have to be as careful in choosing a score as for earlier repertoire. However, the
more specific information you can find at the beginning of a score about its
sources, the better.
Articulation and Phrasing
In the early nineteenth century, legato style gained favor over the usual de-
tached articulation of the eighteenth century, particularly in Italian vocal mu-
sic. If composers wanted a detached articulation they now asked for it by mark-
ing the score with various accentuation marks. These could include dots,
strokes, dashes, and accents for differing degrees of stress, separation, or stac-
cato. Composers also used a wide variety of dynamic accents including $fz, sfp,
rfi sfyfy or ffto highlight variations in articulation.4 In vocal music the text
continued to be the main inspiration for articulation, and by including various
accentuation marks in their instrumental works, composers may have been en-
couraging instrumentalists to shape their lines as if they had words. As men-
tioned in previous chapters, many instrumental treatises advised players to
listen to good singing for examples of expressive articulation. The violinist
Charles de Bériot, husband of the famous singer Maria Malibran, discusses the
importance of emulating good singing and expressive pronunciation in his 1858
Methode de violón: \
We cannot repeat too often that the performer will not be perfect until he can repro-
duce the accents of song in their most delicate forms. . . . It is then of the highest im-
portance for the singer to articulate clearly the words which he undertakes to inter-
pret. . . . It is well understood that the degree of intensity of this pronunciation
should be in harmony with the spirit of the piece. . . . These are the varied and di-
verse shades of expression which the violinist should render, giving to his bow a soft
pronunciation for calm and serene music, and employing it with graduated force in
passionate music. This accentuation gives to the instrument the prestige of words:
we say that the violin speaks in the hands of the master.5
Vocal method books of the time, including, among many others, those of
Gesualdo Lanza (1809), Domenico Corri (1810), Giacomo Gotifredo Ferrari
(1818), and Isaac Nathan (1823) stress the importance of communicating the
text by using appropriately placed breaths, phrasing, and expression to comple-
ment the dramatic meaning of the words.6
The most famous discussion of vocal technique and style from the period
(and probably the most widely available in libraries today) is the younger
Manuel García's Traite complet de Van du chant. Written in 1841 and revised in
1847,1854, and 1872, it was translated into English as Hints on Singing in 1894/
The Garcia family had an extraordinary influence on opera and singing in the
nineteenth century. Garcías father was the famous tenor Manuel del Popólo
Vicente Rodríguez García, the first Almaviva in Rossini s The Barber of Seville.
In addition to singing florid tenor roles, Garcia the elder was also famous for his
portrayal of Mozart s Don Giovanni, which he sang in transposition. As a com-
poser, performer, and impresario, he brought his family of singers to New York
in 1825, introducing Italian opera to American audiences by producing operas
of his own as well as works by Rossini and others. The company included his
wife as well as his son Manuel and his daughter Maria. Both of his daughters,
Maria Malibran and Pauline Víardot, later became famous divas.
While Manuel the younger was not as successful onstage as his father or his
sisters, he went on to become one of the most important voice teachers of his
day. His treatise, though written years later, captures the approach and style he
learned from his father in the first decades of the century. It is divided into two
parts: the first deals with proper production of tone and vocalization of vowels,
while the second begins with a detailed discussion of using phrasing and artic-
ulation to communicate the meaning of the text, which Garcia says is of the
greatest importance. He stresses the correct pronunciation of consonants and
the appropriate stress of syllables with regard to grammatical position and spo-
ken inflection. One of Garcías students, the noted English baritone Charles
Santly, remarked that while it may seem easier to sing in Italian than in English,
the proper pronunciation of the Italian language takes much care and atten-
tion. Santly stressed to his own students "the necessity for refinement, delicacy,
and finish in the execution of all detail, whether of music or language; attention
to these distinguishes the artist from the artisan."8 Garcia includes many exam-
ples of passages requiring noble, pathetic, impassioned, and buffo effects from
operas by Rossini, Bellini, Meyerbeer, Mozart, and others, with detailed in-
structions on the rhythmic placement of consonants in order to achieve the de-
sired dramatic results.9
Even more interesting is the discussion of phrasing and breathing that fol-
lows. Garcia presents two kinds of breaths: full and half. A full breath would
take enough time for a deep intake of air, while a half breath would be more like
a short catch breath. He recommends changing the written rhythm of |a phrase,
if necessary, to fit in an appropriate place to breathe. A particularly fascinating
example is his recommendation for Zerlinas aria "Batti, batti" from Don Gio-
vanni (ex. 4.1). In the phrase "staro qui come agnellina," he suggests changing
the rhythm of the dotted eighth and sixteenth notes in the measure of "qui
come angnel-" to an eighth note followed by a sixteenth note and a sixteenth
rest for "-lina" in the next measure, in order to catch a quick half breath before
the next phrase. In some scores of this aria, particularly in aria collections, the
rhythm is printed as dotted in the first measure followed by two even eighth
notes in the second measure. In 1789 Mozart wrote two dotted rhythms, but
what we see in some modern editions is Garcías suggested alteration from 1841.
Garcia's influence was so strong that his recommendations became part of the
arias performance tradition and made their way into published scores.
In the early nineteenth century, singers considered the score a flexible start-
ing place, tailoring the music to fit their particular needs and abilities. For ex-
ample, Garcia suggests changing the underlay of syllables, if necessary, to facil-
itate breathing and phrasing or to correct a composer s poor text setting. He
advises against breathing in the middle of a word unless absolutely jnecessary,
and he recommends that any changes made to accommodate breathing be as
unobtrusive as possible. As well as adding rests when the composer did not in-
dicate them, Garcia suggests places to sing through written rests in order to
connect phrases for dramatic effect. He also discusses altering triplets or dotted
figures to fit the character of an aria. For majestic or martial music, oyerdotting
or double dotting may be called for, while for more gentle sentiments, soften-
ing a dotted figure into a triplet can add expression. For passages of coloratura
Garcia suggests emphasizing or accenting the first note in a group in order to
shape the phrase and avoid monotony. He also recommends giving dissonant
notes slightly more emphasis, and he shows how and where to use the breath to
add expressive sighs and sobs (a device most modern singers avoid). Many an-
notated examples from Handel, Mozart, and early nineteenth-century opera
punctuate his discussion. Garcías eighteenth-century musical examples offer a
fascinating glimpse at a nineteenth-century perspective on Classical style, while
his early nineteenth-century examples give a definitive view of singing style at
the time. For singers working on early nineteenth-century Italian opera reper-
toire, I strongly recommend reading and studying the entire treatise.
Example 4.1. García: rhythmic alterations for Mozart's "Batti, batti" from Don Giovanni m
Λ Complete Treatise on the Art of Singing: Part Two, Da Capo Press, 1975, p. 64.
Tempo, Rhythmic Flexibility, and Rubato
Just as the flow of rhythm became more flexible in the early nineteenth century,
so did the flow of tempo and the use of rubato. Since traditional court dances
now held a much less prominent place in social life, they had less influence on
the shape of music than in the Classical era. The combination of meter and pre-
vailing rhythm, which used to give a strong indication of the appropriate
tempo and character of a work, no longer had as important an impact. In addi-
tion, by the end of the eighteenth century a single work often contained a num-
ber of different thematic elements, each with its own character or mood and
each demanding a slightly different tempo. Nineteenth-century composers
started using more qualified and descriptive tempo terms such as Allegro agi-
tato, Larghetto espressivo, and Andante sostenuto, along with terms such as an-
dantino and cantabile to describe the tempo they wanted.
Unfortunately, these vague terms did not necessarily clarify the distinctions
among subtly different tempos. Clive Brown notes that Rossini, in his Paris op-
eras, used a variety of slow tempo terms without any specific hierarchy of slow-
ness. Brown also notes that in many situations Rossini seems to have indicated
a 4/4 or alia breve meter without any obvious reason. Sometimes a piece
marked 4/4 feels decidedly "in 2," while another piece marked alia breve feels
"in 4."10 Without an accompanying metronome marking, it is sometimes

difficult to determine precisely what a composer meant by a certain tempo
marking. Even a surviving metronome marking, while supplying a reference
point or starting place, can be misleading. The character and flow of the music
must provide the clues, unless one is thoroughly familiar with a particular
composer's tempo preferences.
In vocal music of this period, the text often provides answers to questions of
tempo. In the early nineteenth century, the singer had more license to surge
ahead if the text became excited or impassioned, or to hold the tempo back a bit
if the dramatic situation became sad or tender. It was common for a crescendo
to be accompanied by an accelerando, just as a diminuendo would accompany
a rallentando. García remarked that "the works of Donizetti and Bellini contain
a great number of passages which, without bearing the sign of the rallentando
or accelerando, yet require their use."11 Often singers might do exactly the op-
posite of what the notation suggests. Charles Rosen states in The Romantic
Generation that "the sudden increase of speed at the end of the phrase would
have been executed by Rossini singers with a slight expressive slowing down."12
This kind of tempo flexibility required the accompanist or orchestra to adjust
tempo along with the soloist, whether it be for an entire section or just a mea-
sure or two. Such coordination would be even easier to accomplish with a piano
accompaniment in a chamber setting.
Of course the amount of rhythmic flexibility employed depended upon the
context of the music and the choices made by a particular singer. Certainly the
famous singers of the time had different strengths and abilities when it came to
flexibility and agility. The renowned Giuditta Pasta, for whom Bellini and
Donizetti both wrote important roles (including Amina, Norma, and Anna
Bolena), was said to have excelled in unadorned cantabile singing. Contempo-
rary accounts report that her voice was not as flexible or fast as those of some of
her rivals. In contrast, Isabella Colbran, who eventually married Rossini and
most probably influenced his vocal writing, is said to have had more difficulty
sustaining a steady tone or singing in tune in slow, simple passages, yet she ex-
celled at fast, brilliant fioritura.
Just as the amount of flexibility varied from singer to singer, it also varied
from composer to composer. Garcia states that the music of Haydn, Mozart,
Cimarosa, and Rossini demands a complete exactness in the rhythmic move-
ment; any changes one might make in the values of the notes should not inter-
fere with the steady movement of the measure.13 While Rossinis music may
demand more precision and exactitude than the music of Bellini and Donizetti,
there is no question that the flow of rhythm within measures and phrases was
becoming more flexible for both solo instrumentalists and singers.
The other kind of rhythmic flexibility was tempo rubato. Here the voice or
solo instrument goes out of phase with the accompaniment, which continues
in a steady flow as the soloist steals or borrows time, only to return it later in the
measure or phrase. This practice, introduced as sprezzatura in the seventeenth
century, had been used throughout the eighteenth century.;The agogic accent,
which falls into the realm of rubato, was a particularly important expressive de-
vice throughout the nineteenth century. If a certain note merits special accen-
tuation for textual or general expressive reasons, the soloist may stress and hold
it slightly longer than notated. The stolen time for the agogic accent must be re-
turned, however, and the general flow of the music left undisturbed. Garcia
offers many operatic examples in which the rhythm can be bxpressively altered
using rubato and agogic accents. He encourages the use of rhythmic alteration
and flexibility to suit a wide variety of situations, and he sternly admonishes
against performances that are metronomic and stiff. Many nineteenth-century
tutors encouraged singers to be expressive by varying the rhythm within the
measure, the flow of phrases across several measures, and the tempos of larger
sections within an entire song or aria.
A singer s taste and abilities determined the degree to which he or she em-
ployed these devices. As we will see with portamento, vibrato, and ornamenta-
tion, tutors recommended using caution and moderation whh regard to tempo
modification. The use of rubato by a sensitive musician should result in an al-
most imperceptible flexibility throughout a performance. Contemporary ac-
counts, however, indicate that many singers and instrumentalists used flexible
rhythm and tempo to a great degree, often beyond what sorne now would con-
sider good taste.
Voice Types
A singer in his youth, Rossini considered the castrato voice his ideal, both for its
virtuosic and expressive capabilities. He wrote one role in an early opera, Aure-
liano in Palmira (1813), for the last castrato who appeared on the operatic stage,
Giovanni Battista Velluti. The final operatic role written jfor Velluti was in
Meyerbeer s II ero ciato in Egitto (1824). By then the public had lost interest in
the castrato voice, particularly in London, where Velluti was seen as a curiosity
at best and a disgusting freak at worst.14 Opera composers turned to the heroic
contralto voice as the closest replacement for the primo uomo castrato. Some
roles, including several for which Giuditta Pasta was famous, required the fe-
male lead to be in male dress, thus putting a woman in what would have been
the castrato's part. Both Pasta and Malibran were said to sound male in their
lower registers and female in their higher registers, thus appealing to Italian
opera audiences' love of sexual ambiguity. Rossini used this low female voice
type for comic heroines such as Rosina in The Barber of Seville and Angelina m
La Cenerentola. Both Pasta and Malibran sang soprano roles as well.
The tenor voice gained full acceptance in the roles of the young hero or lover
and was also used in character parts. Rossini s Otello includes three demanding
tenor parts. Most tenors still blended into their head register above g' l Adolphe
Nourrit, who premiered numerous roles written for him by Rossini, Meyer-
beer, and others, sang the many high Cs in Rossini's Guillaume Tell in his head
voice. He reigned at the Paris Opera from 1826 to 1837, when his rival, Gilbert-
Louis Duprez, made his debut as Arnold in Guillaume Tell. Duprez caused a
sensation, assuring his success in Paris and his place in tenor history by singing
the famous high Cs in full chest voice. This was too much for pooi: Nourrit,
who withdrew from the stage and later committed suicide. Rossinij however,
much preferred his singing to Duprez's. "That tone," Rossini said, referring to
the chest-voice high C, "rarely falls agreeably upon the ear. Nourrit sang it in
head voice, and that is how it should be sung. [Duprez's note sounds] like the
squawk of a capon whose throat is being cut."* ^
The lowest male voices still largely played buffo characters or wise old teach-
ers and fathers. Their parts tended to have more robust singing and ¡fewer pas-
sages of ornate coloratura. Antonio Tamburini's vocal gifts, however, led Bellini
and Donizetti to compose a new kind of baritone role, making use of a higher
tessitura and longer, more legato lines. The baritone Giorgio Ronconi also in-
spired Donizetti to write roles that took advantage of his superior acting skills.
The collaborative efforts of these singers and composers paved the way for the
dramatic baritone roles later seen in Verdi's operas.
When writing a new opera, Bellini is said to have worried more about the
contracted singers than the libretto. Despite the celebrity of Rossini, Bellini,
Donizetti, and other successful composers, singers continued to transpose and
substitute numbers to suite their taste and comfort. They also used an approach
called puntatura, whereby they would keep the music in its original key but
readjust certain pitches to fit their ranges. This could involve lowering ex-
tremely high notes or, more commonly, moving particularly low notes higher.
Yet starting in the 18408, the young Verdi began to place ever more importance
on the dramatic and musical flow of the entire opera, insisting on making all
such changes himself rather than leaving them to the discretion of the singers.
Technique
In Bel Canto in Its Golden Age, Philip Duey observes that the principles of good
singing remained remarkably consistent throughout the many treatises from
Caccini to Garcia. Writers and teachers since the seventeenth century have
stressed natural methods and the cultivation of natural abilities, and they have
even reproduced some of the same phrases to describe desirable qualities in a
singer.16 From the early seventeenth century to the time of Rossini, the musical
goals for singers were a sweet, pure tone, blended registers, command of the
messa di voce> a facility for executing florid ornaments, and an ability to convey
the emotions of a text.
Writers in the early nineteenth century tried to explain how to achieve these
goals in more concrete terms. Isaac Nathan discusses the transition from the
middle register to the high voice and distinguishes between the falsetto and
"feigned" voice in the upper range. The falsetto, he reports, is produced by the
aperture of the mouth "in the small cell or cavity above the arch of the mouth,
called the internal nose." The feigned voice is formed "at the back part of
the head and throat just above the glottis where the uvula is situated."17 He
warns against carrying the chest voice up too high. Instead the singer should
strengthen the higher tones using a messa di voce in the feigned voice before
joining it to the chest voice.
By contrast, writers described how Pasta could use different registers of her
voice in different ranges. For example, she could carry her chest voice up to g"
above the staff but also use her head tones for pitches in her middle range.18
Garcia the elder taught Malibran how to develop this abilit^ so that she could
reportedly sing any note except those at the extremes of her vocal range in ei-
ther chest, middle, or head register. She used this capacity to great dramatic
effect.19 Most writers from this period and earlier urged singers not to force
their voices in either volume or compass. While Malibran was famous for her
variety of tone colors and dramatic flair, she rarely chose to make a sound that
was not beautiful.
After experiencing severe vocal trouble and leaving the stage, Garcia the
younger studied the anatomy of the vocal organ while working in an army hos-
pital. He presented his findings to the French Academy of Sciences in 1840 as
Mémoire sur la voix humaine. His continued interest in the physiology of the
larynx led to his development of the laryngoscope in 1854. He incorporated his

findings into the various versions of his Traite complet de Vart du chant and. dis-
cussed the position of the larynx in relation to the timbres clairs et sombres (clear
and dark tones), but he continued to place great importance on blending regis-
ters and the development of beautiful tone. He also presented over ioo anno-
tated exercises on various scale patterns for the development of agility. Garcia
instructed that these scales and roulades should be performed legato in most
cases, with each pitch "perfectly connected and distinct/'20 He expected his
students to be able to execute each exercise on all five Italian vowels at the
tempo of quarter note — 120: a daunting challenge indeed.
Another important vocal pedagogue of this time was Nicola Vaccai. Frus-
trated by his lack of fame as an opera composer, Vaccai enjoyed great ¡success as
a voice teacher to members of society. As an alternative to the endless hours of
tedious scales required of professional singers and students, he composed short
exercises with texts by Pietro Antonio Metastasio, the most important Italian li-
brettist of the eighteenth century. Through this Métodopratico di canto italiano
per camera of 1833 (the camera indicating salon singing as opposed to opera
singing), his amateur students learned all the fundamentals of technique and
ornamentation while singing charming little songs.
Rossini, like many composers before him, wrote a series of vocalizzi and
solfeggi for the more professionally minded student. His vocalises are exercises
in agility to be sung on one vowel, usually "ah." The solfeggi are wordless
melodies to be sung on the sol-fa or solfege syllables. Rossini believed that these
exercises were critical for developing and maintaining flexibility and should be
practiced every morning: first softly and slowly, the second time softly and
faster, and finally full voice and very fast.21 He composed a mere eighteen vo-
calises, which would probably take about thirty minutes to complete as he in-
structed. According to Garcia, this kind of exercise should be followed by a pe-
riod of rest; he cautioned his students not to practice for more than a half hour
at a time. If singers today want to develop the vocal suppleness and flexibility
necessary for this repertoire, thirty minutes each morning is a worthwhile in-
vestment.
Vibrato
Vibrato in this period was still much more restrained than what we are used to
today. The terminology used to describe it also remained confusing and diffi-
cult for us to interpret. A pure, steady tone was the ideal, and Nathan suggests
that "any unsteadiness or tremor of voice is to be remedied by taking the note
softer; a contrary course only serving to increase and confirm the defect."22
This comment probably doesn't mean to sing with a tight, straight tone but
rather with less of the muscular pressure that creates the kind of vibrato com-
monly heard today. In A Complete Dictionary of Music (1806) and the subse-
quent A Musical Manual or Technical Directory (1828) by Thomas Busby, vi-
brato is defined as "A term used in the Italian opera, to signify that at the note,
or passage, to which it refers, the voice is to be thrown out, in a bold, heroic
style."23 This definition suggests that vibrato was used as a special effect only in
certain dramatic circumstances. Garcia calls this expressive device the tremolo
and discusses its use under the large heading of "Passions and Sentiments," in
which he also discusses the use of sobs, sighs, and laughs:
The tremolo should be used only to portray the feelings which, in real life, move us
profoundly; the anguish of seeing someone who is dear to us in imminent danger,
the tears which certain movements of anger or of vengeance draw from us etc. Even
in those circumstances, the use of it should be regulated with taste and moderation
[mesure]; as soon as one exaggerates the expression or the length of it, it becomes tire-
some and awkward. Outside of the special cases which we haie just indicated, it is
necessary to guard against altering in any way the security of the sound, for the re-
peated use of the tremolo makes the voice tremulous [chevrotante]. The artist who
has contracted this intolerable fault becomes incapable of phrasing any kind of sus-
tained song. It is thus that some beautiful voices have been löst to the art.24
This discussion is accompanied by many operatic examples in both declam-

atory passages and in combination with portamento in both recitatives and
concerted numbers. Garcia indicates use of the tremolo by a wavy line; this
marking can occasionally be found in scores. At times Donizetti and others
wrote vibrato as a special instruction. An accent mark could also indicate a
slight vibrato more in keeping with Busby s definition. At an earlier point in his
treatise, Garcia discusses other expressive effects that he calls "echo notes" and
"repeated notes." These involve a controlled regular pulsation of the voice on
one pitch: "One will carefully avoid aspirating the different articulations of the
same tone, or performing them by a trembling or quivering of the voice."25 It
is a delicate, hardly noticeable effect that can be used on long notes and is suit-
able only for high, light female voices. Whether the vibrato ¡used is the natural
trembling of the voice that Mozart described or a special gesture used for emo-
tional and dramatic effect, it should be hardly noticeable and should not alter
the pitch.
To gain an idea of the small, controlled vibrato fashionable in the late nine-
teenth century, we can listen to early twentieth-century recordings. Soprano
Adelina Patti, who had an unusually long performing career from the i86os un-
til 1914, must have carried with her some of the style she heard growing up in
the 18505. Her recordings from 1905, and many by other turn-of-the-century
singers, offer invaluable examples of the use of vibrato, tempo flexibility, ru-
bato, portamento, and ornamentation.26
Portamento
Many early nineteenth-century writers distinguish between legato and porta-

mento. Legato is the smooth connection of one note to another without inter-
rupting the sound, while portamento connects two distant notes by gliding
lightly and smoothly from one to the other. Different writers give varying in-
structions on whether the intermediary notes should be clearly heard or hardly
noticeable.27 In his Practical Method of Italian Singing, Vaccai offers clear ad-
vice on two kinds of portamenti:
By carrying the voice from one note to another, it is not meant that you should drag
or drawl the voice through all the intermediate intervals, an abuse that is frequently
committed—but it means, to unite perfectly, the one note with the other. When
once the Pupil understands thoroughly how to unite the Syllables, as pointed out in
the first Lesson, he will more easily learn the manner of carrying the voice as here in-
tended: of this however, as before observed, nothing but the voice of an able Master
can give a perfectly clear notion. There are two ways of carrying the voice. The first
is, by anticipating^ it were almost insensibly, with the vowel of the preceding Sylla-
ble, the note you are about to take as shown in the first example [ex. 4.2a]l In Phrases
requiring much grace and expression, it produces a very good effect; the abuse of it
however, is to be carefully avoided, as it leads to Mannerism and Monotony.
The other method, which is less in use, is by deferring, or postponingas it were al-
most insensibly the note you are going to take, and pronouncing the Syllable that be-
longs to it, with the note you are leaving [ex. 4.zb] .28
Garcia confirms that the first method, the "anticipation grace," had gained
favor in the nineteenth century and that the second method, the "leaping
grace," seen frequently in Corris A Select Collection of the Most Admired Songs,
Duetts, &c. (1782), was more fashionable in the eighteenth century. Garcia
writes about portamento using the French term port de voix, whicn no longer
retained its Baroque meaning of appoggiatura from below. In contrast to Vac-
cai, he recommends a more audible slide between pitches, both distant and
close together. He advises that the character of the words and music should de-
termine the quality of the portamento: full and rapid for vigorous sentiments,
slower and more gentle for tender and gracious movements.29 He also gives ex-
Example 4.2. Vaccai: Practical Method of Italian Singing, lesson XIII, "Portamento,"
a) anticipating, p. 30, b) postponing, p. 33. The modern Schirmer edition, based on the
London 1834 edition, has remained in print since 1894.
ampies where a singer can breathe between the note of the portamento that an-
ticipates the main note, and the main note itself (exs. 4»за and b).
Slurs marked in vocal music could indicate either that a vowel should con-
tinue for more than one pitch or that a legato articulation is required. Occa-
sionally portamento was marked specifically in scores, but a slur mark could
also indicate portamento, especially if the slur connects notes with two differ-
ent syllables. In the Donizetti song excerpted in example фзс, the portando is
indicated in the score (either by Donizetti or an editor). (In a similar passage
from La regata veneziana, shown in example 4.3d, Rossini also marks portando
in the score but has written the gesture out in regular-sized notes.) If you want
to breathe before "Ti giuro amore," carry the С up to the F with a crescendo,
sing the F on the "o" syllable of "desio," then breathe and reattack the F on "Ti"
as a subito piano. The portamento here can also be performed the same way
without a breath between "desio" and "Ti." The change in dynamic level and
accompaniment texture at "Ti giuro amore" implies a new dramatic feeling for
that section of text; perhaps some tempo flexibility or rubato could also be used
here to good effect.
Contemporary accounts reveal terrible abuses of portamento, sometimes de-
scribing Italian singers sounding like meowing cats. These sorts of complaints
only got worse as the nineteenth century progressed. Singing treatises had of
course been discussing portamento in its various guises since the seventeenth
century: Caccini s strascino and cercar della nota (see chapter i) developed into
Corn's portamento and Garcías port de voix. Early twentieth-century record-
ings clearly show the use of this technique, which took on the derogatory con-
notations of crooning and swooping later in the twentieth century. Both Ellen
Harris and Will Crutchfield, in their "Portamento" articles in The New Grove
Dictionary of Music and Musicians (1980) and The New Grove Dictionary of
Opera, point out that singers specializing in early music have not incorporated
this element of voice production into their style. Just as we may be reluctant to
use Garcías sobs and sighs, singers today may feel uncomfortable adding porta-
mento to the "pure" style of singing that became fashionable by the late twenti-
eth century. Certainly combining portamento with todays fuller and more
muscular vocal sound and continuous vibrato would produce a very different
effect from that of most of the singers we hear in early twentieth-century
recordings. Clive Brown has tried to nótate the use of portamento heard in
Pattis recordings.30 Once the gesture is notated, however, it seems to indicate
something less subtle than what we hear on the recordings. Vaccai s advice re-
Example 4.3. Examples of portamento, a) and b) Garcia: A Complete Treatise on the Art of
Singing: Part Two, Da Capo Press, 1975, p. 20, passages from Rossini's Lagazza ladra and
Sigismondo'y c) Donizetti: "Eterno amore e fe" from Collezionedi canzonette (undated),
mm. 37—41, reprinted in Composiziont da camera, Ricordi, 1961; d) Rossini: "Anzoleta
avanti la regata," mm. 16—18, from La regata veneziana (1835), Ricordi, 1967.
mains true: this is an art "which nothing but the voice of a skillful Master can
communicate perfectly to the learner."31
Ornamentation
Continuing the trend of the late eighteenth century, composers in the early
nineteenth century wrote more ornamentation into their scores. Treatises of
the time continued to advise the use of good taste and moderation, yet con-
temporary accounts describe performances so highly ornamented that they
rendered the music unrecognizable. According to a famous anecdote, the great
castrato Velluti was said to have added so much embellishment to Aureliano in
Palmira that Rossini couldn't recognize his own melodies. Rossini supposedly
vowed from that moment on to write all ornamentation into his scores to avoid
having his music disfigured and upstaged by the vain display of singers.32 In
another tale involving a much older Rossini, the young Patti sang a particularly
florid rendition of "Una voce poco fa" at one of his Saturday soirees. The com-
poser then remarked, "Very nice, my dear, and who wrote the piecei you have
just performed?"33 Despite these stories, scholars believe that Rossini and other
early nineteenth-century Italian composers expected and encouraged singers to
add their own ornamentation. As in earlier repertoire, singers were obliged to
choose ornaments suited to an aria's text and dramatic context. The more ex-
ceptional artists were masterful at this, while less experienced or talented
singers were criticized for using generic formulas that showcased theif technical
strengths at the expense of both words and music.
Composers were generally becoming more careful about notating the
rhythms of small graces, yet these could also be added where not indicated.
More elaborate embellishment could be used in cadenzas, or substituted for
coloratura supplied by the composer. Some singers truly improvised on the
spot. In 1829 Rossini contrived for the competing divas Henriette Sontag and
Maria Malibran to appear at a private salon concert in Paris. After each sang
some solo songs and arias, the composer and the guests persuaded the rivals, af-
ter much protesting, to sing a duet from his Semiramide. What followed was an
impromptu battle of ornaments, but one that began a new friendship and a leg-
endary collaboration.34 Some singers kept notebooks in which they worked
out possible variations and cadenzas in advance. Rossini composed several lead-
ing roles for the French singer Laure Cinti-Damoreau, who excelled in the Ital-
ian style. Her detailed notebooks, housed in the Lilly Library at Indiana Uni-
versity, are now available from Dover Publications.35 During the period of
Italian bel canto opera, singers and composers worked closely together on mat-
ters of ornamentation. In the end, however, whether the composers approved

or not, it was the singers—particularly the famous stars—who controlled what
happened in performance.
APPOGGIATURAS AND GRACE NOTES
Composers in the early nineteenth century also wrote more appoggiaturas than
their predecessors, with the precise values they wanted shown by regular-sized
notes. This practice helped to clarify the often difficult decision about how long
or short to make an ornamental note. Sometimes composers still wrote appog-
giaturas as small notes, perhaps to communicate a desire for a slightly different
type of stress or emphasis, or perhaps merely as an inconsistency. Many pas-
sages in songs and arias, and particularly in recitatives, of this period still need
additional appoggiaturas for words with unstressed final syllables or to add
heightened expression to a dissonant note. Some composers, notably Rossini
and Donizetti, rarely indicated appoggiaturas, assuming that the singers would
know where to put them. Bellini and Verdi were more conscientious about no-
tating appoggiaturas where they wanted them.
Eighteenth-century rules governing the length of appoggiaturas still applied
to music of this period. An appoggiatura (either printed or added) preceding
same pitched notes would often take the entire value of its main note. For ex-
ample, in the recitative shown in example 4-4a, the first A of "valle" should be
sung as an added B-flat appoggiatura. The small notes in the following mea-
sures should take the entire value of the note they precede. In example 44b, the
small note can be either a quarter to make the rhythm match the quarter/
eighth rhythm of the previous half cadence, or it can be an eighth if you don't
like the clash that occurs between the G in the vocal line and the A in the right
hand of the piano part. A little rubato would easily solve that problem, how-
ever.
If a small ornamental note adds expressive dissonance to its main note it is
probably intended as an appoggiatura. In example 4-4C, for instance, the small
note should receive more length and emphasis—an eighth note in this case—
than its shorter cousin, the grace note.
By 1830 most small ornamental notes written with a slash across their tails
were understood to be grace notes. (As we have seen in Bellini s "Quando in-
cise" from 1824 [ex. 4-4a], in earlier music that was not always the case.) In con-
trast to the appoggiatura, the grace note (acciaccatura) is performed so lightly
and quickly that it is almost impossible to tell whether it happens on or before
the beat. This grace was introduced in the Classical era as a short appoggiatura
Example 4.4. Bellini: a) "Quando incise su quel marmo," mm. 8-12; b) "Vanne, o rosa
Fortunata," mm. 11-12; с) "Quando incise," mm. 37-39, from Composizioni da camera
(1829,1835), Ricordi, 1948.
Example 4.5. a) García: Traite complet de l'art du chant: Part /(1847, facs. Geneva, 1985),
p. 67; b) Rossini: "L'orgia" (1835), mm· I 7~ 2I > reprinted in Serate musicali, Ricordi, 1997;
c) Bellini: "La farfalletta," mm. 8-12, from Composizioni da camera (1829,1835), Ricordi,
1948.
(see chapter 3) and became more popular in the nineteenth century. In lesson
VIII of his Practical Method, Vaccai notates grace notes as small sixteenth notes
and instructs that "the Acciaccatura differs from the Appoggiatura in as much
as it does not interfere with the value or the accent of the note to which it is pre-
fixed."36 Garcia notates the acciaccatura as a small eighth note and warns that
it must not relax into triplets lest it lose its lively and resolute character (see ex.
4-5a). In the passage from a Rossini song in example 4.50, the principal note G
on the downbeat of m. 19 is already a dissonant passing tone between the A and
the F, and so the small ornamental A must be performed as quickly and lightly
as possible. In example 4.5c the ornamental note should also be performed
rapidly on the syllables "ci" in m. 9 and "pre" in m. n.
TRILLS AND TURNS
Just as the shorter, faster version of the appoggiatura gained favor in the early
nineteenth century, so did the main-note start for trills. Garcia repeats familiar
advice: the trill should be clear and distinct, and students should practice it
slowly at first, using a dotted rhythm and gradually increasing speed and even-
ness as the voice becomes more flexible. For an isolated trill he shows a special
preparation from below, but for most situations he says no preparation or ter-
mination is needed.
Turns and mordents were becoming more popular, written into music by
composers and added freely by singers. Composers continued to use different
kinds of signs for accented and unaccented four- and five-note turns, which
were easily confused. To help clarify which pattern they wanted, composers in-
creasingly wrote out more turns using both small and regular-sized notes.
Singers probably substituted the patterns they preferred, adding and inter-
changing the ornaments at their pleasure.
As with the trill, a main-note start for turns was becoming more popular, but
other turn patterns were acceptable as well. In example 4.6a one can see that
Bellini has written a turn sign in m. 26 and then a slightly different turn pattern
in regular-sized notes in the following measure. Donizetti has written two
different patterns of turns for the cadenza in example 4.6b. The speed of a turn
is determined by the musical context: If the accompaniment stops, as in exam-
ples 4.6b and d, then the singer can take time to shape the turn. If the accom-
paniment continues as in example 4.6c, then the turn should be fast to keep
pace with the flow of the music. In general, turns were performed more slowly
as the nineteenth century progressed.
Other small graces called mordents, gruppetti, or compound appoggiaturas
Example 4.6. a) Bellini: "Quando incise su quel marmo," mm. 26-27, from Composizioni
da camera (1829,1835), Ricordi, 1948; b) Donizetti: "La ninna nanna" (1839), mm· 12-1-22,
reprinted in Composizioni da cameray Ricordi, 1961; c) and d) Rossini: "Anzoleta avanti la
regata," mm. 40-41, 44-45, from La regata veneziana (1835), Ricordi, 1967.
Italian Bel Canto Ί49
added several delicate ornamental notes to a main note. Vaccai explains that
"the mordent is the ornament which offers the greatest variety, as well as the
greatest difficulty in its execution, on account of the lightness and neatness
which it requires. It consists of two, or three notes, and adds much to the grace
of the Phrase, without taking anything away, or interfering at all with the in-
tentions of the Composer."37 In his examples (see ex. 4.ya), the ornamental
notes all come before the beat and thus steal time from the preceding note.
These kinds of ornaments could be written into the music with small or reg-
ular-sized notes (see exs. 4./b and c) or added at the discretion of the singer.
The use of rubato and rhythmic flexibility in combination with these small
graces gave singers some freedom in their execution. The exact pattern of a trill
or turn or the rhythmic placement of a grace note or mordent was less impor-
tant than the charm and expression the figure added to the melody.
Cadenzas and Improvised

Florid Embellishment
Manuel Garcia recounts an incident from 1815 in which his father was asked to
sight-read his part at the first rehearsal for a new opera: "When his first aria had
been reached he sang it off with perfect phrasing and feeling, but exactly note
for note as written. After he had finished, the composer said 'Thank you signor,
very nice, but not at all what I wanted/ . . . The Elder Garcia was skillful at im-
provising . . . he made a number of alterations and additions, introducing runs,
trills, roulades and cadenzas. . . . The old composer shook him by the hand.
38
'Bravo! magnificent! That was my music as I wished it to be given.'"
The art of improvising florid embellishments reached its peak with the Ital-
ian singers of the early nineteenth century. The best way to learn stylistically ap-
propriate ornamental additions is to study surviving examples of ornamenta-
tion from the period. According to Will Crutchfield s article "Voices" in Brown
and Sadies Performance Practice: Music after 1600, hundreds of examples of
added embellishment exist in manuscript form and in published scores from
the period. These contain ornaments by famous singers, composers including
Rossini and Donizetti, and various unattributable sources. Some of these
sources are in European libraries or private collections. The easiest place to see
some of these examples is in the New Grove (2001) articles on "Rossini" and
"Improvisation: Nineteenth Century," and the New Grove Dictionary of Opera
"Ornamentation" article. Crutchfield s article includes many examples of
added cadenzas and flourishes, mostly from well-known opera arias. Robert
Toft's Heart to Heart: Expressive Singing in England, 1780—1830 includes orna-
Example 4.7. a) Vaccai: Practical Method of Italian Singing, lesson IX, pp. 20-21, mm. 21-
27; b) Rossini: "La promessa" (1835), mm· 18-19, reprinted in Serate musicali, Ricordi, 1997;
c) Bellini: "Sogno d'infanzia," mm. 46—47, from Composizioni da camera (1829,1835),
Ricordi, 1948.
mented examples of arias by lesser-known Italian composers, songs by English

composers, and some examples from Handel and Mozart. These sources can
provide a starting place for the more adventurous reader. Another invaluable
source for ornamentation of this period is Austin Caswell s collection "Em-
bellished Opera Arias," published by A-R Editions. It presents over twenty
complete arias by Bellini, Rossini, and others, often including several alterna-
tively ornamented versions of particular sections of the aria. Caswell s initial
commentary presents detailed information on sources and specific ornaments.
He also includes general information on the period, the music, and ¡the sing-
ers. (See "Chapter 4: Performance Practice-Ornamentation" in "For Further
Reading.")
Several basic trends can be discerned in the embellishments of this period.
The first was embellishments which did not disturb the basic flow of the
melody. Garcia recommends varying a simple melody when it is repeated. In a
letter to Rossini, the English soprano Clara Novello asked whether it was nec-
essary or permissible to make variations in the da capo section of an aria. Rossini
replied that "the repeat is made expressly that each singer may vary it^ so as best
to display his or her peculiar capacities."39 This kind of embellishment could
also be used in strophic songs. The variations could be slight, such a¿ changing
the rhythm or adding some of the small graces discussed above. Or they could
include a bolder alteration of the melody, perhaps introducing a higher tessi-
tura, a more florid line, or a more technically difficult pattern, all the while
keeping the larger rhythm of the section steady. Caswell comments that in
arias, particularly those by Bellini, singers would often leave a cantabile melody
in the slow cavatina section plain and unadorned, varying only the fioritura to
suit their own abilities. Many times the composer wrote customized coloratura
for a particular singer performing a role or aria. The alternative embellishments
Rossini wrote for Pasta in the aria "Non piu mesta" from La Genérentela take
the vocal line higher than the original and also simplify some of the long
runs.40
The final coda section of a fast cabaletta was a prime place to show off. Gar-
cia includes four pages of suggested variations on standard coda formulas at the
ends of arias. Example 4.8 shows a brief glimpse of the vast possibilities he offers
in his treatise; the lines of text are from the various arias used to make the vari-
ations. According to Crutchfield, the now common final cadential fermata is a
late nineteenth-century invention. In Rossini s time the coda would continue
into the final cadence with a burst of florid activity and no pause or ritar-
dando.41
Example 4.8.Garcfa: coda variants fro Traite complete de lart du chant: Part II(1847, Facs, Geneva, 1985), p.72
Example 4.9. García: cadenzas from Traite complet de Van du chant: P¿zrí//(1847, facs. Geneva, 1985), p. 76.
The other basic kind of embellishment involved adding cadenzas. Garcia in-
cludes six pages of suggested cadenza formulas to suit a variety of harmonic sit-
uations. Again, example 4.9 shows only a fragment of his offerings. Cadenzas
from this period served the function they did in earlier music, namely to high-
light the end of an aria or section (ex. 4.ioa). They could also now appear at sev-
eral interior moments of pause within an aria, perhaps denoting a change of
thought or musical idea (ex.4.iob). Sometimes they were indicated with a fer-
mata; sometimes the composer wrote out a specific cadenza, which the singer
could change to suit his or her taste and abilities (see also ex. 4.6b). Other times
the score would be marked merely a piacere, col cantoy or colla voce, which al-
lowed the singer the freedom to proceed "at their pleasure" by adding some-
thing simple or more elaborate depending on the dramatic context (ex. 4.na).
In a strophic song such as Rossini's "La partenza," perhaps the instruction con
molta grazia inspired the singer to vary this cadential figure in every verse. In
other situations, when the accompaniment becomes sparse or drops out alto-
gether, the singer does not have to keep pace with a steady rhythmic flow and
has more freedom to adorn or vary the melody (ex. 4.nb; see also ex. 4.6d). A
cadenza need not be as long or elaborate as Garcías suggestions, especially in
the more intimate songs.
Cinti-Damoreau, who taught at the Paris Conservatoire after retiring from
the stage, encouraged her students not to perform her ornaments verbatim, but
rather to use the varied formulas as the basis for their own inventions. This is
still good advice today. As with the improvised embellishments of earlier peri-
ods, you must first become familiar with the particular formulas and patterns
of this period and then use that vocabulary to create your own ornamentation.
The early nineteenth century was the culmination of the Italian bel canto ap-
proach to technique and ornamentation. This period's stylistic principles came
from the opera house but were equally at home in the realm of chamber song.
The gentle, relaxed production favored since the seventeenth century was still
central to this style. Legato was more popular than it was in the eighteenth cen-
tury, but the expressive and dramatic articulation of the text remained crucial.
Vibrato was still small and shimmering, but now it was sometimes specifically
notated as a special effect. Portamento was much more common, but it was not
meant to be overused.
Florid ornamentation had reached its most elaborate state and could be used
to vary the return of a da capo aria or rondo form, as well as to differentiate the
verses of strophic songs. More modest variation could be added to slower sec-
Eample 4.10. a) Rossine: "La partenza" (1835), mm. 55-57 reprinted in Serate musuicali Ricordi, 1997; b) Donizetti: "Una lacrima" ("preghiera")
(1841), mm.31-32, reprinted in composizioni da camera, Ricordi, 1961.
Example 4.11. a) Rossini: "La partenza" (1835), mm. 20-22, reprinted in Serate musicali, Ricordi, 1997; b) Bellini: "Per pieta, bell'idol mio," mm. 32-35,
from Composizioni da camera (1829,1835), Ricordi, 1948.
tions of arias or songs, and small graces, including appoggiaturas, grace notes,
trills, and turns, were added at the singer's discretion. Cadenzas small or large
were often included at a variety of places throughout arias or songs. Todays
singers wishing to learn how to compose or improvise their own ornamenta-
tion for Rossini, Bellini, Donizetti, and other bel canto composers should
study as many examples from this repertoire as they can and be careful not to
confuse these embellishments with the patterns from earlier periods.
New issues introduced in this chapter have to do with rhythm and tempo.
The scores of the early nineteenth century started to include more descriptive
tempo terms that don't necessarily help in selecting an appropriate speed.
Tempo is also much more flexible in general for this repertoire, often ¡changing
frequently to accompany shifts in text and dynamics. Rubato should be used as
well to shape the rhythm and enhance expression. Garcia instructs singers to
adjust the rhythm and tempo to suit breathing, phrasing, expression, and orna-
mentation. You can enjoy considerable freedom in your approach to rhythm
and tempo for this repertoire, all in the service of dazzling ornamentation,
beautiful singing, and heightened expressivity.
Chapter 5 German Lieder
This chapter discusses the German lied and its development through
the nineteenth century. It will consider songs by Schubert, Schu-
mann, Mendelssohn, Brahms, Hugo Wolf, and Richard Strauss, as
well as by Louis Spohr, Liszt, Wagner, and others. The lied was first
viewed as a marginal form in such compositions as Mozarts "Das
Veilchen" and "Abendempfindung" and Beethoven's "Adelaide" and
his song cycle An die ferne Geliebte. In the hands of Schubert, the lied
and the song cycle began to realize their greater potential. Many com-
posers considered their Heder insignificant in comparison with their
larger symphonic compositions. This smaller form, however, was able
to convey many important themes of nineteenth-century Romanti-
cism while appealing to a large audience, both in private and public.
The richness of the German poetic tradition in the hands of great
masters such as Goethe, Heine, Riickert, and others helped shape the
course of the Romantic movement. Poems and vast landscape paint-
ings depicted the individual's place in the natural world and helped
define the German national identity. The growing educated middle
class in German-speaking countries showed a heightened interest in
160
German Lieder Ί6Ί
poetry and art, as well as an increased appetite for literature, science, and mu-
sic. Journals of all sorts emerged to satisfy the publics need to discuss intellec-
tual and philosophical issues. Many Romantics sought to synthesize numerous
artistic elements into a unified whole, eventually epitomized by Wagrier in his
music dramas and his concept of the Gesamtkunstwerk. The lied was a perfect
miniature form combining music and poetry, art and literature, natural land-
scapes and the interior life of the individual seeking a place in the world.
Composers wrote Heder for a variety of reasons. Most nineteenth-century
Heder were inspired by particular people, circumstances, or emotions and of
course by the wealth of magnificent poetry that begged to be set to music.
Schubert set poems chosen or written by his close friends, and he intended his
songs for performance in his inner social circle. He may also have foreshadowed
his own death with his final great song cycle Winterreise. Schumann responded
to the tempestuous circumstances leading to his joyous marriage to Clara
Wieck with an outpouring of 138 songs within the single year of 1840. Brahms
fancied a number of female singers with rich, low voices and wrote songs with
their instruments in mind. He also had a close working relationship with bari-
tone Julius Stockhausen, who was a performing colleague as well as the recipi-
ent of music composed for his voice. Composers also set poems to music as an
expression of the ideals of the Romantic movement, and they discussed these
philosophical concepts in journals. Schumann founded one of the first music
journals, Neue Zeitschrift für Musiky which included theoretical writings as well
as reviews and critiques of new songs. Wolf, who also spent time as a music
critic, declared that his songs were written "for epicures, not amateurs."1
Yet it was the general amateur public that demanded a constant supply of
newly published songs to be used for home entertainment. Many composers
took advantage of the fact that a good deal of money could be made from the
sale of songs. Carl Loewe, Robert Franz, Mendelssohn, Schumann, and
Brahms all wrote songs, duets, and quartets geared to the abilities of amateur
musicians. Private music making took place in virtually every educated home.
Piano and voice teachers in the cities had no want for students from wealthy
and middle-class families. The lied was a welcome guest at private gatherings
both small and large, and many composers performed their songs in these less
formal situations.
The lied figured in the development of public concerts as well. As the nine-
teenth century progressed, professional public concerts became more numer-
ous and more varied in programming. Songs, solo instrumental selections,
melodramas, and poetry readings gradually began to appear on the same con-
162 German Lieder
cert programs with symphonies, concertos, and choral works. Even touring vir-
tuosos such as Liszt and Clara Schumann appeared with assisting artists, both
vocal and instrumental. Influential or charismatic performers such as Men-
delssohn, Liszt, and Strauss had the power and freedom to promote their own
lieder and those of their colleagues. Liszt, a particular favorite with the public,
was an enthusiastic champion of many of his fellow composers. He introduced
audiences to many Schubert songs through his dazzling piano fantasies on their
themes. Brahms was not a virtuoso like Liszt or Clara Schumann, but as a pi-
anist he often performed chamber music and lieder. In the 18505 and '6os his
collaboration with Stockhausen in public performances of Die schöne Müllerin,
Winterreise, and the cycles of Schumann brought this repertoire to a wider pub-
lic. It also introduced the idea of the solo vocal recital. At first contemporary
critics had "an apprehension of great monotony" comparecí with the familiar
marathon smorgasbord concerts of the day.2 The public enjoyed the more fo-
cused experience, however, and the concept took hold.
By the end of the century, homogenous concert programs featuring only or-
chestral selections, chamber music, or lieder had become common. Orchestra-
tions of songs, or songs conceived for voice and orchestra, like those of Mahler,
were in demand for orchestral concerts. Many lieder composers, in addition to
performing as pianists, were also well-known conductors who could program
their songs on their orchestral concerts. Mendelssohn, Spohr, Brahms, Liszt,
Mahler, Strauss, and Wagner all conducted their own works. Over the course of
the nineteenth century, the lied was thus transformed from a miniature form
for prívate consumption to a large, expansive form for the jmblic concert hall.
Yet, whether performed by Schubert and his amateur musician friends at a pri-
vate gathering in 1820 or by Strauss and his wife Pauline at an orchestral sub-
scription concert in 1903, the lied demanded close communication between
singer and accompanist and provided an intimate experience between per-
formers and audience.
Editions
A wide variety of German lieder editions are available today. ¡Some are based on
nineteenth-century publications, some on recent scholarship. Many com-
posers meticulously prepared their lieder for public use, and the most reliable
presentations of their original markings and performance instructions are edi-
tions published in their lifetimes. Songs also appeared in journals or monthly
music magazines, carefully prepared for public appreciation. By the end of the
nineteenth century, Breitkopf & Härtel had published complete editions of the
German Lieder 163
works of Schubert, Schumann, Mendelssohn, and Brahms. These were most

often based on scores published while the composers still lived. Most modern
editions of these composers' works are either reprints of these sources or based
on them. Bärenreiter has published a new scholarly Urtext edition of Schubert,
the Neue Schubert Ausgabe (NSA), and new editions of Schumann, Mendels-
sohn, and Brahms are under way, although they do not yet include all the vocal
music. For the most part, modern editions of German Heder that are based on
nineteenth-century sources are reliable representations of the composers' origi-
nal intentions.
Schubert was not aggressive about publishing his music. He often wrote out
extra copies of songs to give to friends, making slight changes in each copy—
hence the five versions of "Die Forelle."3 He also revised songs for various rea-
sons, including compositional choices and performance adjustments (the
eighth-note accompaniment of "Erlkönig" was easier for him to play than the
triplets), resulting in multiple versions of many songs. In 1821 several of his
friends had "Erlkönig" published at their own expense, and it became hugely
popular. This opened the door for more songs to be published and performed
at fashionable society soirees. Yet while 187 of his 600 songs were published in
his lifetime, most of his music remained unknown until other leading com-
posers championed it after his death. Brahms was involved in editing Schu-
berts instrumental music for the Breitkopf & Härtel E Schuberts Werke edition,
and Eusebius Mandyczewski oversaw the Heder. (The modern Dover volumes
of songs reproduce this work, and most other readily available editions are
based on it.)
Max Friedlaender, an oratorio and Heder baritone who had studied voice
with Garcia and Stockhausen, undertook an important second career as a mu-
sicologist. In 1887 he wrote his dissertation on Schubert and thereafter devoted
all his attention to scholarly work on German folk songs and Heder. His edi-
tions of Schubert songs, published by Peters, are based on nineteenth-century
sources.
Bärenreiters NSA versions of Schubert's songs, edited by Walter Dürr in the
1970s, include many variants based on manuscript sources not included in the
older complete-works edition. Dürr s scholarly commentary, in German, offers
fascinating details about each song.
Clara Schumann edited the first complete compilation of Robert s works af-
ter his death, and most editions of the songs are based on it. Although
Mendelssohn did not consider his songs a high creative priority and composed
them mostly for amateurs, he carefully prepared his song manuscripts for pub-
164 German üeder
lication. Most modern editions of Mendelssohn's Heder are reprints. While

Brahms s inspiration to write songs was more personal, he, too, carefully super-
vised their publication. Wolf took particular pains to ensure that every detail
was just right in the publication of his songs; once he was well known, he spent
extra money on expensive paper stock and decorative borders and often placed
the poet's name in larger type, above his own. The first complete edition of his
works was published in the 19605. !
Strauss also supervised the publication of all his songs for the general public,
except for the Four Last Songs, which were discovered in manuscript after his
death. Most current editions of his songs contain very few differences from the
first editions and are in fact reprints. In 1964 a deluxe commemorative collec-
tion of his songs was published in honor of the looth anniversary of his birth.
Unfortunately, this Gesamtausgabe der Lieder, in four volumes, was prepared
hurriedly for the centenary and contains various mistakes. Barbara Petersen, in
Ton und Wort: The Lieder of Richard Strauss, cautions against relying on the edi-
tions accompanying critical commentary and details its other shortcomings.4
She also offers critical evaluation of other editions, manuscripts, and sources
and considerable in-depth information about the songs. Most first editions of
Strauss songs included singable English translations. Some later printings even
included translations m French, Italian, and Russian.
Transpositions
In the nineteenth century, attitudes toward transplanting and transposing indi-

vidual songs evolved from the flexible approach of earlier centuries. Different
composers had definite opinions about changing the order or the key of their
songs. While Strauss wrote most of his songs for high voice, he was happy to
have them transposed for the comfort of the general public and presented pub-
lished groups for high, medium, or low voice. Rather than transposing his en-
tire opus or a group as a block, however, Strauss recommended specific trans-
positions for individual songs, taking into consideration the sound of a
particular key for the piano accompaniment and the comfort of the vocal tessi-
tura.
Wolf was notoriously particular about the performance off his songs; once he
even berated a singer in the middle of a performance for taking liberties he
found offensive. He tried to interest several famous singers, including Lilli
Lehmann, to perform his songs, but because of his reputation for being diffi-
cult, many did not want to work with him. Given his remark about his songs
not being for amateurs, he probably would not have approved of his songs be-
German Lieder 165
ing transposed. The Wolf complete-works edition presents the songs in their
original keys, as do the Dover scores, which are based on editions from Wolf's
lifetime.
Brahms, on the other hand, was happy to change the key of a composition to
suit the comfort of the singer, and while he conceived certain songs with a low
voice in mind, he published many of them in a high key. He was more particu-
lar about preserving the keys of songs based on folk music sources, the Volks-
Kinderliederand Deutsche Volkslieder, even though many lie too high for the av-
erage child or folk singer.
Schumann had definite opinions about transposition and the naáire of cer-
tain keys. He published his thoughts in an article entitled "Characterization of
the Keys":
It is as inadmissible to say that this or that feeling, in order to be correctly expressed

in music must be translated in this or that key (anger, for example in C-snarp minor)
as to agree with Zelter who declares that any feeling may be expressed in any key. . . .
The process by means of which the composer selects this or that principal key for the
expression of his feelings is as little explicable as the creative process of genius it-
self. . . . No one will deny that a composition, transposed from its original key into
another produces a different effect.5
Schumann did not seem to be against transposing in principle, but he did indi-
cate his strong feelings on the important contribution the choice of key makes
to the composition as a whole. Since Schumann wrote many of his songs for an
idealized voice rather than a particular singer, his choice of key was important
in establishing the general mood and feeling of a piece. He also introduced a
more important role for the piano in his songs. Transposing would change not
only the vocal tessitura but also the texture and color of the piano writing. The
key relationships among songs within his cycles are particularly important as
well. In Dichterliebe and Liederkreis, the individual songs are fragments that
cannot stand on their own but are strung together like beads on a necklace. To
preserve the integrity of the harmonic construction and flow of the whole com-
position, the entire cycle must be transposed by the same interval.
Some of Schumann's later songs are in keys that take standard voice types
into an uncomfortable range. Boosey & Hawkes published volumes of his
songs edited by Clara and based on her complete works but transposed for high
or low voice. Excluding the cycles, some individual songs have been transposed
into more suitable keys.
Schubert composed many of his songs to suit his own voice, which was a
166 German Lieder
light high baritone. He also wrote songs for his friend Karl Freiherr von Schön-
stein, a gifted amateur singer, who was also a high baritone. Schönstein reports
that "Schubert had grown fond of me and enjoyed making music with me,
which he did often; he admitted to me repeatedly that from this time on [1818],
in his songs, he generally had in mind a voice of my range .i"6 Schönstein also
mentions that another close friend of Schubert's, the well-lmown court opera
singer Johann Michael Vogl, had one of Schubert s songs transposed into a
lower, more comfortable key. Two weeks after Schubert had given him the
song, Vogl sang it for the composer. "After Vogl had sung it, without saying a
word beforehand about the song and without having made the slightest change
in it—sometimes he liked to take liberties with Schubert's songs—Schubert
cried out in his simple way, cYou know that song isn't bad! Who's it by?' After a
fortnight he no longer recognized his own creation."7
Schubert and Vogl had an interesting working relationship. The young com-
poser idealized the older, famous singer and was indebted to him for promoting
his songs and offering moral and financial assistance. According to some re-
ports, Schubert was thrilled with the way Vogl sang his songs and loved playing
for him: "the manner in which Vogl sings and the way I accompany, as though
we were one at such a moment."8 In other accounts, Schubert was sometimes
less than happy with Vogl's operatic approach. The composer was willing, on
occasion, to compromise with Vogl regarding transposition, but he probably
would have wanted to maintain the integrity of the cycles' tonal construction.
A word about gender-specific songs: plenty of accounts exist from the nine-
teenth century of famous sopranos singing An dieferne Geliebte, Winterreise, or
the Müllerlieder, in which the protagonist or speaker in the poems is obviously
male. By the same token, Stockhausen sang Frauenliebe und Leben, in which
the voice in the poetry is decidedly female. Thus the boundaries of gender were
not considered to limit a singer's choice of songs in the days of Schubert, Schu-
mann, or Brahms.
German Diction and Technique
The percussive characteristics of the German language exerted a strong influ-

ence on German singers of the nineteenth century, who aimed to combine Ital-
ian vocal production with their own approach to articulation and expressivity.
Many German singing manuals throughout the century advocated the schönen
Tone of the Italian school, and some of the most famous German singers, in-
cluding Stockhausen and Lilli Lehmann, were students of Garcia or his disci-
German Lieder 167
pies. Yet many German singers were criticized for shrieking, yelling, shouting,
whispering, and other affronts to bel canto.
Clara Schumann's father, Friedrich Wieck, advocates beautiful singing in the
Italian style in his 1853 manual Klavier und Gesang. In the 1878 edition of his
treatise, he chastises voice teachers and singers for producing tones that were
"cold, ugly, tight, forced, and throaty."9 An acquaintance of Schubert's and
Vogl's recalled the singer's remarks on interpreting German lied: "in particular
he emphasized the necessity for a clear enunciation of the words. If you have
nothing to say to me, ran his motto, you have nothing to sing to me either."10
Vogl was also credited with introducing Schubert's circle to the "declamatory
style of singing."11 Schubert's good friend Leopold von Sonnleithner, writing
in 1857 and 1860, disapproved of the current "dramatic" style of singing Schu-
bert: "According to this, there is as much declamation as possible, sometimes
whispered, sometimes with passionate outbursts." He commented that, after
having heard Schubert rehearse and perform his songs "more than a hundred
times," the "singer must conceive the song lyricallynot dramatically.. . . Every-
thing that hinders the flow of the melody and disturbs the evenly flowing ac-
companiment is, therefore, exactly contrary to the composer's intentions and
destroys the musical effect.—Consequently singers with good voices and just
a natural way of singing have frequently achieved great effect with these
songs. . . . Schubert, therefore, demanded above all that his songs should not
be so much declaimed, as sung flowingly."12
Jenny Lind was a particularly notable exponent of this kind of lyrical ap-
proach. A student of Garcia, Lind was famous for her Italian, French, and Ger-
man operatic roles, as well as for her charming renditions of Swedish songs.
Mendelssohn, who wrote songs with flowing melodies and simple accompani-
ments, loved her singing and often accompanied the "Swedish Nightingale" on
his songs. The Schumanns greatly admired her as well, and Robert dedicated
his op. 89 songs to her. Clara wrote in her diary: "Her singing is ever so sincere.
There is no showmanship, no great display of emotion, yet she touches your
heart. . . . No weeping, sobbing, or quavering sounds, no bad habits of any
kind. . . . If we could only persuade her to sing nothing but good music and to
get rid of all that rubbish by ... Meyerbeer, Bellini, Donizetti, etc.; she is too
good for that."13
A very different kind of singer who epitomized the dramatic style of singing
was the German opera star Wilhelmine Schröder-Devrient. She too was ad-
mired by the Schumanns, and she received the dedication of Robert's op. 48
168 German Lieder
Dichterliebe. Mendelssohn enjoyed performing with her and persuaded her to

join him, substituting at the last minute for a singer who was ill, in an 1840 per-
formance of some Beethoven songs.14 It was in another Beethoven work, Fide-
lio, that she first made her mark on the German opera world. Schröder-Devri-
ent trained initially as an actress, and she never quite mastered the art of
beautiful singing. Henry Chorley, an English music critic partial to bel canto,
did not care for her performances of Italian opera: "Her tones were delivered
without any care, save to give them due force. Her execution was bad and
heavy. There was an air of strain and spasm throughout her performance."15
Yet when she brought her mesmerizing acting ability to Fidelio, she used ex-
pressive declamation to heighten the intensity of her singing, deliberately sacri-
ficing beautiful tone for dramatic effect. Her performances were a revelation,
particularly to Wagner. When he heard her sing Leonore in 1829, it changed his
life. She inspired him to fashion a new kind of German singing, and he went on
to create several roles for her, including Senta in Der fliegende Holländer and
Venus in Tannhäuser.
Writing in 1865, Wagner complained that German singers did not receive
proper training, and he advocated the establishment of a national singing
school. He believed that German singers were ill equipped for dramatic singing
because they were trying to apply the Italian method to the German language
without success: "The Italian style of vocalism, our only present model for the
classic, is inappropriate to the German language. . . . A right development of
Song, upon a groundwork of German Speech, is therefore the first task for us
. . . that any such method should never permit euphony to be tampered with,
goes without saying."16 Ideally, Wagner wanted singers to preserve the beauti-
ful tone of the Italian approach and infuse it with clear German diction and
pronunciation so that the words and the music would have equal importance.
He wanted German singers to speak in singing and above all to sing and act
naturally.17
Clear diction and natural declamation were also extremely important to the
enthusiastic Wagner disciple Hugo Wolf, as demonstrated by his reviews of
singers from 1884 to 1889. Writing in the Viennese Sunday journal the Wiener
Salonblatt, he reviewed Lilli Lehmanns first performance as Isolde and criti-
cized her poor diction: "A shame that she enunciates so badly. Not a single word
was comprehensible. A serious deficiency!"18 Several weeksllater he again men-
tioned his frustration with her enunciation as Donna Anna and Norma, most
likely sung in German. By contrast, Lehmann, who went oh to become a cele-
brated Wagner singer, writes in her book How to Sing that she sang Isolde "in a
German Lieder 169
beautiful legato in the fine Italian style." She also criticizes Wagner adherents
for having misguided ideas about technique and advocates a way of connecting
vowels to consonants to facilitate a distinct articulation without being destruc-
tive to the voice,19 On the other hand, Wolf did not like exaggerated diction ei-
ther, and he criticized a singer portraying Alberich in Wagner's Das Rheingold
for going too far: "Herr Horowitz seemed obviously concerned to unite the
singer with the declaimer. The intention is admirable. It is only that in his over
enthusiasm he went so far as to not sing at all, but only to declaim. He should
try to achieve a blend of both and he will succeed."20 Nor did he approve
of a melodramatic performance of Schubert's "Erlkönig" and Mozart's "Das
Veilchen" by another celebrated Wagner singer, Pauline Lucca: "This rough
theatricality does not work to best advantage in the concert hall."21
All these accounts demonstrate that German singers in the nineteenth cen-
tury were trying to define their own style and to sing expressively in their own
language while still achieving a beautiful, melodic sound. Some were more suc-
cessful than others at finding a workable compromise.
Articulation and Expression Marks
Most Heder composers included relatively few articulation marks in their vocal
music. They expected that singers would use the text to shape the melody ap-
propriately. Articulation marks and performance instructions added in purely
instrumental music or vocal accompaniments were meant to help the instru-
mentalists play expressively, as if they had diction and text. In Schubert, for in-
stance, an accent mark can mean either a slight elongation of a stressed syllable
or an added emphasis for an important word or dissonant neighbor note. For a
more detailed discussion, see Clive Brown's Classical and Romantic Performing
Practice.22 Schubert also used varying lengths of crescendo and decrescendo
marks that can sometimes be confused with accentuation marks. In the Neue
Schubert Ausgabe volumes, the editors have tried to standardize the length of
the shorter decrescendo marks into > accent marks. This approach doesn't
cause a problem if you understand a > mark to mean the elongation of an ex-
pressive syllable. It can be quite confusing in the piano part, however, when try-
ing to choose between an accent that adds more energy to a note and a slight de-
crescendo that would soften the energy of the gesture.23
For example, compare the two editions of "Suleika Π" seen in example 5.1. In
this song Schubert uses swelling crescendo and decrescendo marks throughout
to shape the phrases. In m. 47 the NSA (ex. 5.ib) makes an accent mark out of
what was a decrescendo in the old Schuberts Werke edition (ex. 5.ia). In this par-
170 Germán Lieder
Example 5.1. Schubert: "Suleika II," mm. 45-48, in a) Schuberts Werke XX/vi/ioi; b) Neue
Schubert Ausgabe IV/11/99.
ticular spot, under the top of an arching melody line, an accent in the piano
part on the word "stilles" would disturb the sense of awakening silent longing
in the breast. The same marking in m. 57 seems equally inappropriate under the
word "Hauch," since the phrase is describing a soft breath. Singers and their ac-
companists must be alert for these kinds of situations when using the NSA. As
always, the dramatic context created by the words will suggest answers to such
questions.
Brahms, too, did not include many expression marks in his vocal music. His
decision to give minimal instructions to singer and pianist was apparently
modeled on Schubert's approach.24 He also had little interest in bowings or fin-
German Lieder 171
gerings for his string writing, usually letting his good friend violinist Joseph
Joachim work out these details.25 Similarly, Brahms left most decisions about
articulation and dramatic interpretation to the singer—especially a singer he
admired. If he was accompanying, he would make suggestions or show his in-
tentions by the way he played.
Strauss had much the same approach to his songs. Performance instructions
such as accents, tenutos, accelerandos, ritardandos, or indications of breathing
places are sparse. Most of his early songs were for his wife Pauline de Ahna, who
sang while Strauss accompanied at the piano or conducted the orchestra.
Strauss clearly felt they could work out such details together. In working with
other singers he admired, he would often defer to their judgment and adapt his
interpretations accordingly. He wrote the Brentano Lieder, op. 68, for Elisabeth
Schumann and frequently accompanied her. Recordings of her interpretations
of Strauss songs offer a glimpse into the style they created together. According
to the page turner at one of their many Liederabenden, Strauss had a very flexi-
ble approach toward playing his own piano accompaniments. He often impro-
vised a more orchestral rendering of the piano part, arpeggiating chords,
adding octaves, and enriching the harmonies. He was even known to improvise
interludes between songs, sometimes including tunes from his operas.26
Embellishment
The attitude toward ornamentation and improvisation in nineteenth-century

German vocal music was vastly different from the practices associated with Ital-
ian bel canto. In the early decades of the century, musicians reacted against the
tasteless excesses of some singers, preferring honest, genuine expression to vir-
tuosity and showmanship. Yet it was understood that certain small ornaments
such as appoggiaturas, trills, turns, and mordents could be added at the discre-
tion of the singer. In his 1821 article on dramatic singing, theorist Ignaz Franz
von Mosel instructs the performer:
Who on the one hand embellishes a simple melody with taste, adding a turn or two
to the original form to produce a pleasing variety but who, on the other hand, does
not succumb to licentious fantasy or to a desire to excel by sheer mechanical ability;
instead he relies on reason, sensibility and taste, which are the prerequisites for every
dramatic singer. Frequently a single suspension, a well expressed mordent, an orna-
ment of three, four, or at most six notes, overtime new and gracious in its form, over-
time pure and clear in its performance, overtime appropriate to the sense pf the text,
to the character of the song, the singer and the accompaniment—these may gen-
uinely embellish the melody, intensify its expression and increase its overall effect.27
172 German Lieder
Vogl, who started his career in the late eighteenth-century opera houses and
was the product of an earlier stylistic tradition, influenced Schubert when it
came to ornamentation. Some contemporary writers report that Vogl freely al-
tered songs and persuaded the composer to change details inj published versions
to reflect his interpretation.28 Other accounts relate good-natured disagree-
ments between the two in which Schubert s compositional decisions prevailed.
According to Eduard von Bauernfeld, another of Schubert's inner circle of
friends, "Small alterations and embellishments, which the skillful singer, a past
master of effect, allowed himself, received the composer's consent to some ex-
tent, but not infrequently they also gave rise to friendly controversy."29 Vogl
kept notebooks of many versions of the songs with his own added ornaments.
He may have had a hand in an 1829 Diabelli edition oí Die schöne Müllerin that
includes substantial alterations. Examples of Vogl's ornamented alterations can
be seen in the NSA, as well as in articles by Walter Dürr and David Mont-
gomery.30 Vogl continued to sing Schubert's songs until shortly before his own
death in 1840 at the age of seventy-two. As his vocal powers diminished, he is
said to have relied increasingly on exaggerated dramatic declamation and spe-
cial effects such as whispering, falsetto, and sudden outcries. Sonnleithner re-
lates that "Michael Vogl, it is true, overstepped the permissible limits more and
more as he lost his voice . . . and he merely helped himself out as well as he
could, in the manner of the experienced opera singer, where his voice and
strength did not suffice. And Schubert would certainly not have approved his
manner of performance as it developed in his last years."31
As the century progressed, German composers took a firmer stand against
added embellishments. Carl Maria von Weber is said to have forbade his singers
any ornamentation whatsoever of his music. Mendelssohn's attitude was also
conservative, but he may have allowed small additions and alterations in some
cases. Chorley reports that "Mendelssohn wrote so as to allow no space or exer-
cise of fancy for the vocal embroiderer; and thus to alter or add to his music,
would be to injure it, by showing arrogant disloyalty to the!master's wishes and
meanings. Nevertheless, I well recollect the quiet smile of pleasure with which
even Mendelssohn used to receive a shake exquisitely placed in the second verse
of his delicious 'Frühlingslied' (op. 47)."32 Chorley also relates that Mendels-
sohn allowed a small change of pitch in a recitative in Elijah\that heightened the
dramatic effect. Yet in some of his songs, including the op. 8 "Frühlingslied,"
rather than allowing the singer freedom to improvise, he wrote out his own ver-
sion of an Italianate cadenza (see ex. 5.2).
Clara Schumann made her disapproval of the Italian vocal style clear in her
German Lieder 173
Example 5.2. Mendelssohn: "Frühlingslied," op. 8. no. 6, mm. 65-68, Mendehsohn Werke
XIX/I3.
remarks about Jenny Lind's choice of repertoire. She also shunned empty virtu-
osic display in her own restrained and pure playing. She and violinist Joseph
Joachim strove to express the music simply and directly rather than to indulge
in extravagant emotional outpourings. Brahms, who shared a close personal
and artistic relationship with Clara, was of the same mind. Wagner, though on
the opposite side of a protracted and bitter battle against Brahms and his con-
servative musical ideals, allowed his singers absolutely no alterations to the
score. Wolf, his devoted acolyte, hated the vain displays of Italian opera and was
similarly opposed to singers' making any changes to the score.
Notated Ornaments
APPOGGIATURAS AND GRACE NOTES
As notational practices developed over the course of the nineteenth century,

composers wrote more ornaments into their scores. Increasingly they used reg-
ular-sized notation for writing out appoggiaturas and turns. By the midcentury
174 German Lieder
the small ornamental note, particularly with a slash through it, came to be un-
derstood almost universally as a fast, light grace note to be performed either just
before the beat or right on it (see the discussion of acciaccatura in chapter 4.) In
Schubert's songs, however, the notation of appoggiaturas and grace notes is
very inconsistent. The NSA offers suggested solutions to Questionable situa-
tions, but they are not always the only answer. Since Schubert composed in a
transitional period between late Classical practices and early Romantic ones,
his music can follow the rules for graces outlined in the eighteenth-century
treatises of C. P. E. Bach and others. Often a small ornamental note will add ex-
pressive dissonance in a Schubert song and should be stressed and performed
on the beat. Sometimes an appoggiatura should take half the value of the note
it decorates, while other times it will demand the entire value of the main note.
In some situations the textual content or the notation in another part of the
song may reveal the most appropriate execution. Still other situations have no
clear solution.
One of the most common figures in Schubert's songs is an appoggiatura
above or below two repeated notes, the second of which is unaccented. In ex-
amples 5.3a-d, the appoggiatura should take the entire value of the main note.
In example 5«зЬ, in fact, many editions show the appoggiatura written out in
large notes.
In examples 5.4a-d, the notation in another part of the song may offer clues
as to how to perform a particular appoggiatura. In "Rastlose Liebe" (ex. 5.4a),
Schubert writes the long kind of appoggiatura at the opening of the song, using
two regular-sized quarter notes. In the Ε-major section at "Lieber durch Lei-
den" he substitutes the shorter version, with an eighth-note rhythm in large
notes on "Leiden" m. 26 and on "schlagen" in m. 28. At the word "tragen" in
m. 33 he writes a small ornamental note before two quarter notes. Here the per-
former must decide whether this ornamental figure should match the shorter
execution on "schlagen" in m. 28 or the longer version on "Herzen" in m. 38.
The NSA suggests the longer quarter-note execution for "tragen" as well as for
"vergebens" later in the song at the return to E minor. That is what I prefer as
well. |
In " Frühlings träum" (ex. 5-4b), Schubert seems to indicate that he wants the
shorter sixteenth-note execution for "malte" in mm. 31 and ¡35 by the way he sets
"an den" in m. 33 and "lacht wohl" in m. 37, and this is the solution suggested
by the NSA. However, Ernest Walker, in his article "The Appoggiatura," sug-
gests that the treatment of "Blumen" in m. 39 might instead be Schubert's in-
tended execution for "malte."33 It is also worth noting that the eighth-note ap-
German Lieder 175
Example 5.3. Schubert: a) "Suleika II," mm. 98-100; b) "Halt," mm. 11-13; c) "Ave Maria/
m. π; d) "Frühlingsglaube," m. 18, Peters, ed. Friedlaender.
176 Germán Lieder
poggiaturas on "malte" and "Blumen" are followed by sixteenth notes, whereas

the written sixteenth notes on "an den" and "lacht wohl" are followed by
eighths. There doesn't seem to be a conclusive solution either way.
Example 5.40 from "Lachen und Weinen" shows what could be a simple long
quarter-note appoggiatura, yet on many early recordings you hear two eighth
notes or a rhythmically indeterminate slide from the Ε-flat to the D in m. 65.1
would say that this moment depends entirely on what kind of dramatic gesture
you want to give the word "Lachen," be it coy, sassy, wistful, or some other
choice. Example $.^а also shows what could be a simple long appoggiatura;
Vogl preferred it short, while Schubert thought about making it even longer in
an earlier version by changing the meter to 6/8.
In situations involving triple time or a dotted note, eighteenth-century rules
would require an appoggiatura to last two-thirds of the value of the main note.
In examples 5«5a and b, both Walker and the NSA recommend that the appog-
giatura be executed this way. In "Ave Maria" (ex. 5-5а), however, you rarely hear
it performed that way. The length usually varies from verse to verse depending
on the specific syllables. In early twentieth-century recordings you often hear a
note of indeterminate length, as in "Lachen und Weinen." When combined
with rubato and portamento, this treatment creates a very expressive yet impre-
cise rhythm. In "Frühlingsglaube" (ex. 5-5b), you could make "Herz" in m. 41
match the rhythm of "Herze" earlier in the song (m. 18: see ex. 5.3d) and use an
eighth-note appoggiatura, or you could match the rhythm of "tiefste" in m. 40
and use a sixteenth-note appoggiatura. For "An die Nachtigall" (ex. 5.5c),
Walker and NSA recommend a quarter-note appoggiatura for the D in m. 13.
In early recordings you hear both an eighth-note appoggiatura and a gentle
portamento from the D to the С that smudges the exact rhythm. The orna-
mental note on "guter" in m. 14 should be light, fast, and before the beat.
In Der Hirt auf dem Felsen (ex. 5.6), the NSA recommends a dotted half-note
appoggiatura in m. 147 on the word "wich," resolving on the downbeat of the
next measure. I prefer using a half-note appoggiatura and making the rhythm
match the earlier phrases in this slow section: "Gram" in m. 131 and "mir" in m.
143. Again, you could use portamento and rubato to create an expressive gesture
of unspecific length.
Example 5.7 shows situations in which you could treat the appoggiatura ei-
ther as a slightly stressed dissonant upper neighbor note or as a fast grace note.
The NSA suggests a sixteenth note on the beat for example 5-7a but a fast, be-
fore-the-beat grace note for examples 5.7b-d. In "Tränenregen" (5.7b), the
tempo you choose and the way you want to inflect the different syllables in each
Germán Lieder 177
Example 5.4. Schubert: a) "Rastlose Liebe," mm. 25-38; b) "Frühlingstraum," mm. 29-40;
c) "Lachen und Weinen," mm. 62-65, Peters, ed. Friedlaender; d) "An Emma," NSAIV/
ia/Vorwort, p. xiii.
178 Germán Lieder
Example 5.5. Schubert: a) "Ave Maria," mm. 11-12; b) "Frühlingsglaube," mm. 40-42;
c) "An die Nachtigall," mm. 10-17, Peters, ed. Friedlaender.
verse may affect how you choose to perform the ornament. A fast, springy ges-
ture might not feel appropriate in a slower tempo. The more lively movement
of "Halt" (5.yc), however, does invite that gesture. In "Der Einsame" (5·7<1), Ι
personally like the melodic shape that results from making two even sixteenth
German Lieder 179
Example 5.6. Schubert: Der Hirt auf dem Felsen, mm. 141-148, International, 1972.
notes on "stilles," and so I save the grace-note gesture for "daÁér gebracht" later
on in the song.
Because Schubert composed at a time of transition, he employed both old
and new ideas. Thus the suggestions for performing appoggiaturas in the NSA
are a good place to start, but they are not necessarily the only solutions.
TRILLS AND TURNS
The preference for starting trills on the main note continued to grow during
the nineteenth century. Noted musicians such as Johann Nepomuk Hummel
and Louis Spohr recommended that if composers wanted an alternative to the
main-note start, they should indícate it. Yet scholars believe that musicians
most likely felt free to vary the beginnings and endings of trills to suit their taste
and the particular musical situation. Composers used both ornament signs and
small and large notes to indicate turns with a wide variety of patterns and
speeds. Example 5.8 shows two different autographs by Spohr of his song
"Nachgefühl." In the first copy from 1834, he notates the turn figure with a reg-
180 German Lieder
Example 5.7. Schubert: a) "Danksagung an den Bach," mm. 5-8; b) "Tränenregen," mm.
5-9; c) "Halt," mm. 27-30; d) "Der Einsame," mm. 17, 29, Peters, ed. Friedlaender.
Germán Lieder 181
ular turn sign and large notes for the D and C; this would indicate that the turn
should start in the upward direction. In another handwritten copy from 1839,
he uses an inverted turn sign and an appoggiatura for the D leading to the C;
this might indicate starting in the downward direction, as Spohr instructed in
his Violinschule. Yet his casual attitude toward the specific notation of the figure
probably meant that he trusted the singer to execute something graceful and
appropriate. Many examples of inconsistent notation of turns and controversy
over their execution exist in nineteenth-century music, even in the early operas
of Wagner. The same kind of controversy—one could call it freedom—exists
regarding mordents. It seems there are no definitive answers.34
Vibrato and Portamento
Since mid-nineteenth-century German vocal technique remained based mostly

on the Italian models (except, of course, with regard to language and diction),
the use of vibrato and portamento still echoed earlier Italian recommendations
(see the discussion in chapter 4). Some German sources discuss vibrato in con-
junction with a gesture of accentuation. In his Violinschule, Spohr described vi-
olin vibrato as an imitation of "the singer s voice in passionate passages, or when
he forces it to its most powerful pitch. . . . Avoid however its frequent use, or in
improper places. In places where the tremolo is used by the singer, it may also
advantageously be applied to the Violin. This tremolo is therefore properly
used in passionate passages, and in strongly marking all the Jz or > tones."35
The Wiener allgemeine musikalische Zeitung of 1813 described vibrato as "here,
the notes will not be pulled out at the roots but only tickled at their tips,"36 also
indicating some kind of gentle accent. In examples 5-9a-d the > or sf 'could
indicate a stress with vibrato.
Wagner occasionally went so far as to indicate vibrato with a wavy line or to
include instructions for a trembling voice, steering singers to produce a special
dramatic effect. While the use of instrumental vibrato gradually increased over
the course of the nineteenth century, significant and continuous vocal vibrato
was still considered a problem that needed to be controlled. In 1884 Wolf criti-
cized a singer in one of his reviews: "she seldom manages to attack the upper
notes securely, and this tremulous vibrato is distressing to the listener."37 Lilli
Lehmann, born in 1848, grew up listening to singers schooled in bel canto, and
she developed her technical approach based on their models. In her book How
to Sing, she states emphatically, "Even the vibrato, to which full voices are
prone, should be nipped in the bud, for gradually the tremolo, and later even
182 German Lieder
Example 5.8. Spohr: "Nachgefühl," two manuscript reproductions from Clive Brown,
Classical and Romantic Performing Practice, Oxford University Press, 1999, p. 508, used by
permission of the publisher.
Germán Lieder 183
Example 5.9. a) Schumann: "Schneeglöckchen," mm, 7-9, Peters; b) Mendelssohn: "Bei

der Wiege," mm. 17-19, Mendelssohn Werke XIX/87; с) Wolf: "Du denkst mit einem
Fädchen," mm. 12-15, Peters, 1904, reprint Dover, 1989; d) Strauss: "Du meines Herzens
Krönlein," mm. 5-12, Joseph Aibl, 1890, International, 1955.
worse, is developed from it."38 In her late-career recordings from the early
twentieth century she sings with a hardly noticeable shimmer to her voice.
Portamento was used much more freely, sometimes to great excess. Leh-
mann warns against its overuse, resulting in "a bad habit lacking in good
taste."39 Yet in early twentieth-century recordings, singers who had begun their
careers in the nineteenth century frequently used portamento for expressive
purposes. Portamento could enhance legato lines and connect small and large
184 German Lieder
Example 5.10. a) Schubert: "An die Musik," mm. 4-6, Peters, ed. Friedlaender; b)
Schumann: "Ich kann's nicht fassen, nicht glauben," mm. 8-n, Peters; c) Schumann: "Ich
kann's nicht fassen, nicht glauben," mm. 20—23, Peters.
leaps in either direction. It could color a specific word for dramatic reasons.40
As mentioned in the appoggiatura section, it could help in performing small
notated ornaments as in "An die Musik" (ex. 5.ioa). The figures in examples
5.iob and c, common in Schumann, could be executed as springy, heroic ges-
tures, depending upon the text, musical context, and intended dramatic inter-
German Lieder 185
Example 5.11. Strauss: "Wie sollten wir geheim sie halten," mm i—6, Joseph Aibl, 1888,
International, 1955.
pretation. In Strauss s "Wie sollten wir geheim sie halten" (ex. 5.11), the wavy
lines seem to request both portamento and vibrato between the E and the low
A and between the F-sharp and the high A.
Phrasing marks are used differently by individual composers, but most often
in vocal music they indicate a single syllable on more than one note. One could
186 German Lieder
use portamento here, but it is not necessary for simple legato singing. Some-
times, particularly in songs by Strauss, a long phrase mark may indicate the
larger shape of a phrase. This could show where to breathe or merely illustrate
the long-term destination of the phrase.
Portamento thus can be used in Heder for expressive, dramatic purposes both
in large and small gestures, but it should be motivated by the words and the
musical context. It should not be used so much that it becomes a tasteless
mannerism, though what many listeners might consider tasteless today is quite
different from what was accepted as the norm a century ago;
The Piano
Between 1800 and 1880 the piano underwent a phenomenal transformation

unlike that of any previous instrument. Its modifications and technical devel-
opments most certainly contributed to the blossoming of liieder. The English
and Viennese approach to piano building continued to produce instruments
with different sounds and technical capacities. Most German composers had
access to and preferred the Viennese fortepianos. The instruments made by the
Viennese piano makers Nannette and Johann Andreas Streicher and by Conrad
Graf had a light touch, even tone quality from bottom to top, and a beautiful,
delicate sound. Their crystalline quality inspired many of the shimmering ac-
companiments in Schubert's songs. The clear sound in the bass range also high-
lighted many of the melodic figures in the left hand of Schubert's accompani-
ments.
Schumann had already written a great deal of piano music before composing
most of his songs, and he had worked out a pianistic language to take greater
advantage of the instrument's tonal palette and expressive potential. His songs
used the piano in a newly integrated way, building layers of melody shared by
voice and piano and colored by rich harmonic shadings. By the middle of the
century, piano makers continued to experiment with every ¡element of the in-
strument. They investigated its basic shape, the thickness and number of
strings, the striking action, the hammer action, the soundboard material, and
the ability of the frame to withstand increased string pressure. Innovations in
pedal technology contributed new kinds of tonal and dampening effects as
well. Piano builders competed to satisfy the technical desires of famous virtu-
oso players, hoping the celebrities would play and endorse their instruments.
By 1880 most of the major technical innovations were complete, and the mod-
ern concert piano, with its full, brilliant sound that could carry over an orches-
tra, became available. Yet even Brahms preferred his more delicate Viennese
Germán Lieder 187
Streicher and wrote much of his piano music with its sound and action in
mind.
The spaces in which songs were performed in the nineteenth century were
likely to have been more resonant than most concert halls today. The quick at-
tack and rapid decay of the pre-i85O Viennese fortepiano provide a clear texture
for such an acoustic. The total volume of this instrument is considerably less
than that of a modern grand piano, which provides greater sustaining power
and accumulating overtones. Singing Heder with a Viennese fortepiano thus
immediately solves many balance issues between voice and keyboard. It allows
the pianist to choose from a wider variety of dynamic levels without having to
worry about overpowering the voice. Singer and pianist can employ more vari-
ety in accentuation and articulation as well. The fortepiano also enables a vo-
calist to sing more gently, with less pressure and effort. This invariably relaxes
the vibrato, perhaps revealing that delicate shimmer so characteristic of early
recordings.
Tempo and Rhythm
As discussed in chapter 4, the nineteenth century brought a more flexible ap-

proach to tempo and rhythm. This tendency became increasingly exaggerated
as the century progressed. In the early iSoos, most Austrian and German com-
posers, including Schubert and Mendelssohn, insisted, as Mozart had, that
their music be played "in time." Yet they also showed a certain degree of flexi-
bility in choosing tempos and often varied them within a work for expressive
purposes. By 1833, a writer in a German musical journal complained that "ritar-
dando and accelerando alternate all the time. This manner has already become
so fixed in the minds of the musical public that they firmly believe a diminu-
endo must be slowed down and a crescendo speeded up; a tender phrase (e.g.,
in an allegro) will be performed more slowly, a powerful one faster."41 Sonn-
leithner, writing in 1860, reflected midcentury practices as he described Schu-
bert s expectations with regard to flexible tempo:
The musical idea should be displayed in its purity. A necessary corollary to this is the
strictest observation of tempo. Schubert always indicated exactly where he wanted or
permitted a ritardando, an accelerando or any kind of freer delivery. But where he did
not indicate this, he would not tolerate the slightest arbitrariness or the least devia-
tion in tempo. . . . A trotting or galloping horse permits of no deviation of strict
time; . . . —a quickly beating heart cannot suddenly stop (except from a stroke) in
order to let the singer dwell on his high A at the words "Dein ist mein Herz und wird
es ewig bleiben" and give rein to his excess sentimentality;. . . these wretched exam-
188 Germán Lieder
pies, are merely intended to serve as isolated instances, for this senseless manner of
interpretation has unfortunately already become the rule. . . . This does not mean in
the very least that Schubert wanted to hear his songs ground out merely mechani-
cally. An accurate, purely musical performance in no way excludes feeling and sensi-
tivity; but the singer should on no account give himself airs.42
Schubert assigned metronome markings to his first few sets of published

songs but quickly gave them up in favor of descriptive tempo terms. Singers of
lieder are probably all too familiar with such vague tempo qualifiers as mäßig,
ziemlich, or etwas and emotional descriptions such as zart, leise, innig, or aus-
drucksvoll. Sometimes these German terms impart a sense of actual speed, but
more often they simply suggest the general mood or character of a song.
Mendelssohn, who used Italian tempo terms in his songs, was also reluctant to
assign metronome markings. It was reported of him that "though in playing he
never varied the tempo once taken, he did not always take a movement at the
same pace, but changed it as his mood was at the time."43 Weber, writing in
1824, had a slightly different approach toward tempo, but it vvas one that repre-
sented a prevailing attitude: "The beat should not be a tyrannical restriction or
the driving of a mill hammer. On the contrary, it should be to the music what
the pulse beat is to the life of man. There is no slow tempo in which passages
which demand a faster movement do not occur, and thereby prevent the feeling
of dragging. Conversely, there is no presto, which does not call for the slower ex-
ecution of certain passages, so that expression will not be marred by overzeal-
ousness."44
Brahms, too, believed that the tempos in his music should not be constant,
nor was it possible to determine the "correct" tempo with a metronome mark.
As he commented in a letter to the conductor George Henschel, "I am of the
opinion that metronome marks go for nothing. As far as I know, all composers
have as yet retracted their metronome marks in later years."45 Most reports de-
scribe Brahms's piano playing as extremely variable with regard to tempo. Clara
Schumann wrote in her journal, "It is not easy to play with Brahms; he plays
too arbitrarily, and cares nothing for a beat more or less. . . . I cannot quite get
used to the constant change of tempo in his works, and he; plays them so ac-
cording to his own fancy."46 It is tantalizing to imagine these two close friends
playing the four-hands piano accompaniment at the premiere of Brahms's
Liebeslieder waltzes.
We are fortunate to have recordings of Strauss accompanying his own songs,
none of which have assigned metronome markings. Strauss was perfectly will-
ing to adapt the flow of a song to fit the tastes and needs of different singers, and
German Lieder 189
his recordings reveal considerable flexibility and variety in tempo.47 By the end
of the century, charismatic conductors (including Strauss) demonstrated their
virtuoso control of orchestras with wildly fluctuating tempos. Achieving a
heightened emotional state with extremely slow tempos was also popular with
certain conductors and singers, who were happy to show off their breath con-
trol.48
As nineteenth-century music became more complex and performers took
increasing liberties with tempo, many composers decided to include more de-
tailed performance instructions, aiming to exert more control over the perfor-
mances of their works. Wolf wrote frequent tempo indications in his songs,
sometimes every few measures. He and Mahler both used elaborate German
descriptions as well as Italian terms to specify subtle shifts in tempo. Com-
posers also started using time signature changes and more precise note values to
control the flow of their music.
DOTTED FIGURES
The notation of dotted figures, duples, and triplets became more meticulous as
the nineteenth century progressed. The songs of Schubert and Schumann,
however, were still somewhat under the influence of ambiguous eighteenth-
century notational practices. (See the discussion of rhythmic alteration in
chapters 2 and 3.) Often a singer and accompanist must decide whether a triplet
figure should agree with a dotted figure or be performed as a distinctly con-
trasting rhythm.49
There has been much debate, for instance, about the preferred approach to
this issue in "Wasserflut" (ex. 5.i2a) from Schubert s Winterreise. A1963 perfor-
mance by Peter Pears with Benjamin Britten at the keyboard provoked strong
reactions when Britten made the dotted figure in the left hand agree with the
triplet in the right hand. Some scholars argued that there was enough evidence
to support the aligning of the rhythms in the eighteenth-century manner. Oth-
ers disagreed for a variety of reasons, including their personal preference.50 In
"Des Mädchens Klage" (ex. 5. i ib), the singer must decide whether to fit the du-
ple rhythms into the steady triplets in the piano, to pull against them within the
triplet, or perhaps to place the syllable after the triplet using a double dot. The
text should help you weigh such choices in any song, but especially when the song
is strophic. The best approach is to allow each verse to differ, depending upon
the length and weight of certain syllables as well as the dramatic inflection de-
sired. In mm. 9 and 10, it would be too complicated to simultaneously perform
a duple in the voice part, a triplet in the treble of the piano, and a double dot in
190 German Lieder
Example 5.12. Schubert: a) "Wasserflut," mm. 5-8; b) "Des Mädchens Klage," mm. 9-10;
с) "Frühlingsglaube," mm. 8-ίο, Peters, ed. Friedlaender.
German Lieder Ί9Ί
the bass. Even if Schubert did want a contrasting rhythm in "Wasserflut," it is

unlikely that this was his intention here. The placement of the bass note should
help convey the diction, not work against it. So, for example, the second verse
might lend itself to observing triplet rhythms in all parts, while the third verse
might benefit from more marked dots from the voice and the bass line. Both
"Frühlingsglaube" (ex. 5.i2c) and "Ave María" (ex. 5»зс) include the same kinds
of situations. Again, the text should be the deciding factor, with the under-
standing that a flexible approach will allow the greatest freedom for musical
and dramatic expression.
A good example of this freedom and variety can be found in early recordings.
Here singers frequently make duples into triplets and triplets into double dots,
all to suit the text and their expressive impulses. We must remember that ex-
pectations of rhythmic exactitude were an invention of the twentieth century.
In 1886 Wolf applauded a singer for his particularly good rhythm: "Especially
admirable in Blauwaert s singing is its accuracy, its rhythmic precision and its
lively expression."51 He mentions this as if it were something unusual^ which it
probably was.
RUBATO
Another element contributing to the larger picture of rhythmic freedom was

rubato. The eighteenth-century tradition of rubato, advocated by both Tosi
and Mozart, allowed the melody line to become independent from the accom-
paniment for heightened expression. Tosi cautioned that if you stole some time
you must replace it, and Mozart taught that the left hand must continue play-
ing in time while the right hand wove around it. In the nineteenth century this
concept was stretched further: the basic tempo was constantly shifting, and in-
ternal rhythms were being pushed and pulled as well. The use of an agogic ac-
cent to linger on a particular note, letting the melody go out of sync here and
there, only added to the music's expressivity and fluidity. In addition to speed-
ing up for a crescendo and slowing down for a decrescendo, it was common to
rush through fast notes and stretch out longer ones. When Brahms played his
own Hungarian Dances, the rhythm was described as "darting and Halting."52
Liszt, writing about Chopin's playing, described his use of tempo rubato as "a
stolen, irregularly interrupted time-measure, desultory and languishing, then
flickering like a flame in a d r a f t . . . or like treetops leaning this way and that ac-
cording to the capricious whims of the wind."53
Of course, different artists used all these devices to varying degrees, and com-
posers sometimes expressed opinions about rhythm and tempo at odds with
Ί92 German Lieder
how they themselves played. Joseph Joachim's violin playing was described as
conservative and noble, yet a 1903 recording reveals tempos that Robert Philip
describes as quirky and volatile, with rhythms that sound almost out of control.
Philip suggests that when we listen to early twentieth-century recordings we
sometimes hear playing that underemphasized rhythmic detail, almost to the
point of carelessness. Certainly over the course of the twentieth century we
have become much more disciplined about rhythmic accuracy and precision.
But Philip cautions that this lack of precision heard in the playing of the great-
est soloists and virtuosos of the late nineteenth and early twentieth centuries
was not a question of sloppiness but rather of style.54
This chapter has continued to track the changing attitudes toward ornamenta-
tion, vibrato, portamento, rubato, and tempo flexibility. In German Heder
singing, elaborate added ornamentation virtually disappeared. You might feel
free to add some small graces, including appoggiaturas, trills, or turns, but only
to songs from early in the period. By the end of the nineteenth century, com-
posers were notating almost all ornamentation, sometimes with small notes or
signs but increasingly with regular-sized notes. The execution of notated orna-
ments, including appoggiaturas and turns, is still open to some interpretation,
particularly in songs by Schubert, which can be influenced by eighteenth-cen-
tury practices. Songs by Schubert and Schumann can also follow earlier rules
for rhythmic alteration, but this remains a controversial subject.
Vibrato can be used as an expressive device in Heder and is sometimes indi-
cated with a wavy line or a variety of accent markings. A wide, continuous vi-
brato, however, is still inappropriate. Portamento, on the other hand, can be
used liberally in many situations for both small and large gestures. It should be
employed for expressive and dramatic purposes. I
Rubato and tempo flexibility can also be used to a great degree to enhance
expression. By the end of the century, variation of tempo and rhythm had be-
come so extreme that composers started trying to regulate the flow of tempo
with meter and note values as well as an increased use of descriptive tempo in-
structions. Even though wildly fluctuating tempos, exaggerated rubato, and
pervasive use of portamento were fashionable by the end of the nineteenth cen-
tury for this repertoire, singers today may want to avoid going too far.
New issues presented in this chapter include approaches to editions and
transposition and German diction. Scholarly Urtext editions of Heder are either
available or under way, but the readily available editions based on nineteenth-
century publications are generally reliable. You should be careful about trans-
Germán Lieder Ί93
positions, especially for song cycles and repertoire from the end of the nine-
teenth century. You can bring a flexible attitude to gender-specific poems,
which are appropriate when sung by either male or female voices. The specific
idiosyncrasies of the German language have a major impact on articulation and
expression for lieder. Many prominent nineteenth-century singers strove to
combine Italian vocal production with German diction, yet many German
singers employed highly exaggerated dramatic declamation. You certainly need
to highlight the special qualities of German without, perhaps, going to the ex-
tremes seen in the nineteenth century.
Chapter 6 French Melodies
This chapter surveys the French art song of the nineteenth and twen-
tieth centuries, beginning with the romance of the mid-nineteenth
century and focusing on the melodies of Fauré, Qebussy, Ravel, and
Poulenc. It also includes the contributions of such prominent singers
as Reynaldo Hahn, Mary Garden, Jane Bathori, Claire Croiza, Charles
Panzéra, Maggie Teyte, and Pierre Bernac. Chapters ι and 2 noted the
importance of le bon gout (good taste) and the sanbtity of the French
language in music of the Baroque. Matters of taste and language con-
tinued to play a significant role in the nineteenth-century develop-
ment of the melodic, influencing the relationships between both po-
etry and music and composer and performer. We are fortunate to have
a close link with the creators of this repertoire through recordings by
singers who worked with Fauré, Debussy, and others, as well as re-
cordings performed and supervised by Ravel and Poulenc. Some stu-
dents today also still have access to teachers who studied with the ma-
jor figures of this period. Many books on the French song repertoire
offer advice on diction, translations of poetic textis, and interpretive
suggestions (see "Chapter 6: Texts and Diction" in "For Further Read-
194
French Melodies 195
ing"). This chapter does not attempt to repeat such material but rather to sup-
plement it by considering the roots of French style as developed by the com-
posers and singers who worked together.
The French melodic took longer than the German lied to come into its own.
This delay was due in part to the devastating effects of the French Revolution.
It did not take long, however, for a new aristocracy to emerge, hungry for artis-
tic stimulation. By the iSzos Italian opera was all the rage in Paris, where Ros-
sini was the director of the Theatre-Italien. Touring celebrities dazzled Parisian
audiences, and the wealthy held private salons to discuss political, philosophi-
cal, and artistic ideas. Often composers, or perhaps their hostesses, sang ro-
mances, simple s trophic songs with modest piano accompaniments. However,
music in the salons was often merely incidental to the conversation.
Schubert s Heder were first published in French translation in the 18305 and
championed in performances by the well-known tenor Adolphe Nourrit, ac-
companied by Liszt. Translated as "melodies," they became wildly popular and
inspired a new level of creativity in the composition of romances. Berlioz com-
posed a set of songs to texts by Irish poet and musician Thomas Moore known
as "A Selection of Irish Melodies." Published in 1830 as Neuf melodies imitées de
VanglaiSy they helped introduce the new term mélodie. The French poetry avail-
able to composers at this time was also improving. Gounod, Massenet, Saint-
Saens, Delibes, and Bizet all composed songs to French texts for the delight of
accomplished amateur and semiprofessional salon performers. Some songs by
Berlioz, Liszt, Edouard Lalo, and Henri Duparc may have been written with
professional performers in mind. Much of this Romantic song repertoire was
written under the predominant influence of Italian opera stars and traveling
virtuosi. Within the context of French taste, it can be approached by following
the stylistic advice given in chapters 4 and 5.
The later music of Fauré, Debussy, and Ravel moved French music in a new
direction and style. The Third Republic (1870-1940) witnessed a flowering of
all the arts, including painting, literature, and music, transforming Paris into
the artistic capital of Europe. Salons now provided a haven for a sophisticated
elite and by the end of the century had become the refuge of the avant-garde.
The music of Wagner had an important influence on some French composers.
The International Exposition of 1889 had a profound impact on Debussy,
Ravel, and many others by introducing exotic elements such as Javanese game-
lans and Russian orchestral music conducted by Rimsky-Korsakov. Not long
after, two other Russians, Sergei Diaghilev and Igor Stravinsky, took Paris by
storm, revolutionizing the theater and involving the entire artistic community.
196 French Melodies
Composers began creating a new harmonic landscape, highlighting the sound

of the music rather than the virtuosity of the performer. They sought to pre-
serve the integrity of their music with precise notation. They also insisted on
accurate performances and found performers who were willing to cooperate.
The role of the performer increasingly became less about displaying skill and
more about serving the needs of the composer.
Composers gravitated to performers who could fill this role, often develop-
ing close relationships such as those of Fauré and Claire Croiza or of Debussy
and Jane Bathori. Some of these have been documented in fascinating memoirs
that reveal the inner workings of rehearsals and performances.l Perhaps the
most fruitful of these relationships was between Francis Poulenc and Pierre
Bernac. They first met in 1926 and began a performing relationship m 1934 that
lasted for twenty-five years. Poulenc composed two-thirds of his songs for
Bernac, tailoring his music to the baritone's particular strengths, including his
ease in singing softly in his high range and loudly in his low range. It was a truly
collaborative relationship, and the singer strongly influenced the composer re-
garding issues of programming, tempo, dramatic presentation, and general
style. As Bernac writes in his book, Francis Poulenc: The Man and His Songs, "we
created them together."2 We are left with many invaluable recordings of this
magnificent pair, as well as with Bernac s indispensable book The Interpretation
of French Song, which every singer of melodies and chansons should keep close
to his or her piano.3 Yet this book, written in 1970, reflects the culmination of
an evolutionary process; the advice Bernac gives is not necessarily how he sang
in his early recordings. His approach also reflects the particular style he created
with Poulenc, incorporating all their idiosyncrasies. It is also worth exploring
the collaborative relationships of some other singers and composers to get a
broader sense of French singing styles.
Editions, Transpositions, and Voice Types
Most nineteenth-century Romantic melodies were published for the general

public, and many composers earned considerable profits frqm the sale of their
songs, which were printed both as sheet music and in journals, newsletters, and
songbooks. Often songs were transposed into a variety of keys and arranged in
different combinations and orders, sometimes with new titles or even new
texts. In his French Songfrom Berlioz to Duparc, Frits Noske pbserves that both
publishers and composers treated songs with a certain casuálness, considering
them to be "mere trifles."4 The variety of nineteenth-century French song edi-
tions can create confusion in determining whether composers intended songs
French Melodies 197
to be presented as a distinct cycle in a certain order or whether the songs could

stand alone or be recombined.
A good example of this ambiguity is Berlioz's Les nuits d'été, first published in
1841 as a set of six songs with piano accompaniment. Despite the awkward writ-
ing for the piano, the songs set a new standard of quality for the melodic. In
1843 Berlioz orchestrated "Absence" for the mezzo Marie Recio, who was tour-
ing with him in Germany and whom he later married. In 1854 he orchestrated
the rest of the songs, transposing them and dedicating each to a different singer.
While he may have had a particular order in mind for the publication of the
piano-vocal score, he never performed all the orchestrated songs together as a
strict cycle. Singers today therefore can decide whether they want to perform
Les nuits d'été with piano or with orchestra, in what order, and in what keys.5
As the century progressed, composers grew more careful about supervising
the publication of their songs. They also became more particular about trans-
position. Fauré s songs appeared in three collections of twenty songs each, pub-
lished by Hamelle in 1879,1897, and 1908. The first book was clearly intended
for the accomplished amateur, while the later songs demanded higher levels of
technical skill and musicianship from both singer and pianist. Fauré preferred
a voice of medium range and wrote most of his songs in keys that avoided op-
eratic high notes. The "Pie Jesu" solo in his Requiem was written for a boy so-
prano only because the clergy of the Madeleine, where it was first performed,
forbade women in the choir. In subsequent performances Fauré entrusted the
solo to a female soprano.
Debussy wrote many of his early songs for the high, light voice of his first
muse, Marie Vasnier. He met her while accompanying a singing class for soci-
ety women, and she and her husband gave moral and financial support to the
young composer. The Quatre chansons dejeunesse, written for and dedicated to
her, were published posthumously in 1926. Jane Bathori, who worked with De-
bussy on all his later songs, believed they were issued as curiosities and that the
composer had not wanted them published.6
Later Debussy, as well as Ravel and others, developed a preference for a lower
voice type and fashioned compositions primarily for professional musicians. As
their music became more subtle and harmonically complex, the choice of key
was directly linked with a certain vocal tessitura and a distinct color palette for
the piano or instruments. Transposing the music would disturb this delicate
arrangement of sounds. Poulenc stated that he never transposed his music once
he conceived it for a specific key.7 (Pianist Dal ton Baldwin, however, has said
that Poulenc did transpose certain songs for Gerard Souzay.)8 Bernacs book on
198 French Melodies
Poulenc lists all the songs with date of composition, publisher, range, and rec-
ommendations for male or female singer (some songs are only appropriate for a
certain voice type). According to the baritone Martial Singher, Ravel admitted
that he had originally conceived "Asie" from Shéhérazade and the Chansons
madécassesíot a male voice, though only female singers who were excellent mu-
sicians had been interested in performing them. Singher, who premiered Ra-
vels Don Quichotte ä Dulcinéesongs, did eventually perform Chansons made-
casses and "Asie," but not until after Ravel's death.9
Most modern editions of French songs are reprints of nineteenth-century
editions or publications supervised by the composer. The original French edi-
tions, including those by Durand, are the best places to start for this repertoire.
Information on publications is available in the New Grove articles on individual
composers as well as in an extensive appendix in Frits Noske's French Song from
Berlioz to Duparc. The Dover editions of French Art Songs of the Nineteenth Cen-
tury (and of the songs of Fauré, Duparc, Ernest Chausson, and Debussy) all list
the original sources for the reprinted scores. The Dover edition of Ravel songs
was edited by Ravel scholar Arbie Orenstein and includes introductory infor-
mation, notes on each song, and further references. Much of the earlier nine-
teenth-century repertoire is suitable to sing in a comfortable transposition,
since that was the practice of the amateur singers for whom; it was intended. It
is advisable, however, to sing the later repertoire in the original keys, especially
works by Debussy, Ravel, and Poulenc.
The International editions of Fauré and Chausson list the original keys of
transposed songs, but these don't always agree with the original keys listed in
the New Grove or reprinted in the Dover editions. For example, both the New
Grove and the Dover reprint of 1908 list the original key of Fauré's "Revé
d'amour" as Ε-flat major. The International score, however, Ishows the original
key as F. It also prints some alternative rhythms in the vocal part. The New
Grove Chausson article unfortunately does not list original keys for the songs.
While the International edition of Debussy does not indicate transposed songs
or original keys, the 1993 Hal Leonard Vocal Library edition of Debussy songs
did publish all the songs in original keys, arranged in two volumes for high and
medium voice. Edited by James Briscoe, it is a critical edition that includes
translations, notes, and historical sources. A new scholarly edition of Debussy s
complete works was begun by Durand-Cdstallat in 1985, but the songs have not
yet been published. A new edition of the complete works of Berlioz was begun
in 1969 by Bärenreiter and includes a volume with orchestral songs, but the vol-
ume of songs with piano has yet to be published. Thomas Grubb, in his diction
French Melodies 199
manual Singing in French, warns students to be aware of the unavoidable mis-

takes found in most editions, particularly regarding texts and translations. He
is most suspect of anthologies with unacknowledged transpositions and ossia
interpolations.10
French Taste and Aesthetics
French music of the seventeenth and eighteenth centuries had been character-
ized by restrained grace and beauty, especially compared with the obviously
emotional and heavily ornamented style of the Italians. In his 176^ Diction-
naire de musique, Rousseau emphasized the need for simplicity in the composi-
tion and performance of the romance. He believed that songs should be written
in a simple, moving style, according to the character of the words: "an accurate,
clear voice that articulates well and sings without affectation is all that is re-
quired for singing a romance."11 In the nineteenth century, this desire for pu-
rity and simplicity of style continued, along with a growing aversion to virtuos-
ity. An overly demonstrative outpouring of emotion or technical skill was not
considered appropriate to the intimate salon setting where romances were
sung. Martin Cooper's French Music from the Death of Berlioz to the Death of
Faurésuggests that the French composer was interested primarily in "arranging
sounds in agreeable and intellectually satisfying patterns. . . . He would prefer
the humbler search for perfection in all forms, however small. . . than in a work
planned on a larger scale, with more obvious pretensions to greatness but less
λ2
aesthetically satisfying."
In 1846 Antoine Romagnesi, a well-known composer and singer of ro-
mances, published L'art de chanter les romances, les chansonnettes et les nocturnes
etgénéralement toute la musique de salon. In this short treatise he presented sim-
ple suggestions that teachers and singers of French repertoire have practiced
throughout the twentieth century. In distinguishing the difference between the
opera singer and more delicate salon singer, "between a large and intractable
voice and the more modest and manageable kind found in the larger number of
non-professional musicians," he advised the salon singer to treat his or her
voice with care and never to force it beyond its natural abilities; he should sing
fort bien rather than bien fort (very well rather than very loudly). The singer
should study the text before he sings it and convey the sense of the poem with-
out vocal display. Breathing and facial expressions should be natural and un-
forced. "Good taste will warn him to search only for those means that are sim-
ple and natural; to avoid mannerism and exaggeration; finally, to model his
vocal inflections on the sentiments that he is called upon to express."13
200 French Melodies
Fauré started performing in aristocratic salons as a young student. He was

not fond of Italian opera and opera singers or "big theater bellowers" as he
called them, and while he admired Liszt, he was not interested in the flashy
world of the virtuoso pianist. He was an accomplished pianist himself but pre-
ferred a collaborative role to the soloists spotlight. One of his favorite singers
was Emma Bardac, the wife of a wealthy banker. She sang in an easy soprano
voice and was an excellent sight-reader who preferred to perform in private. Af-
ter a performance of "Soir," which Fauré had written for her, he told the poet
Albert Samain, "You'll never hear it sung better." Samain recounted that "she
has a feeling for nuances and especially a purity of expression which are ex-
tremely rare."14 In a 1902 letter to a patron, Fauré wrote about his songs: "I
dream of hearing them performed by perfect singers, but I don't know of any
among the professionals. It's the amateurs who understand and interpret me
best."15 Since many of the singers he preferred did not possess great instru-
ments—including the natural, untrained baritone of his friend, composer/
singer Reynaldo Hahn—some assumed (much to Fauré's annoyance) that one
didn't need a voice to sing his songs. He entrusted his later cycles to two young
newcomers, Madeleine Grey and Charles Panzéra, whose singing was charac-
terized by extreme refinement and purity.
Debussy's early experience also included accompanying amateurs in the sa-
lon, and in 1904 he married Emma Bardac, whose previous husband had finally
lost out to her fascination and involvement with young composers. Debussy
strove for a purity and simplicity of style that was governed by nuance. He in-
cluded warmth and passion in his music but knew when to stop before the
emotion became vulgar: "One should sing when there is something to sing
about, and keep the emotional power of the voice in reserve."16 Debussy par-
ticularly admired and worked with Mary Garden, whose intense yet simple and
restrained acting made her his ideal Mélisande, and Jane Bathori, whose
singing was described by Darius Milhaud as having "a pure style, clear diction
and discrete interior charm."17 Bathori, who championed the music of many
early twentieth-century French composers, was also an accomplished pianist
and often accompanied herself when the composers were not at the piano. The
critic Georges Jean-Aubry, in an article from 1913, hailed her as "modern French
song incarnate . . . her art is all of nuances . . . nevertheless it is impossible to
surpass it in naturalness and in that important spontaneity which alone has im-
portance for the listener. . . . [she possesses] vocal simplicity the same scrupu-
lous care of refinement, and the same absence of display even in the most ardent
passion."18
French Melodies 201
The 1907 premiere of Ravel's Histoires naturelles, with the composer accom-
panying Bathori, incited a riotous demonstration in the concert hall and lively
discussion following in the press. Even though Emmanuel Chabrier nad previ-
ously set poems about animals, the public was astonished at Ravel's unusual
choice of Jules Renard's text and his revolutionary treatment of it, particularly
his elision of many mute Vs." In his review, Louis Laloy described Bathori as
having a childlike grace: "a little girl looking at the beasts and laughing at their
comical appearance or the antics they get up to, but feeling very fond of them
just the same. As for the author, seated at the piano, stiff and impassive, he was
the only person in the entire hall who betrayed no emotion."19 Ravel also ad-
mired the clear voice and impeccable diction of Madeleine Grey and frequently
requested that she perform his music. Their recording of Chansons madécassesls
decidedly restrained emotionally. Poulenc, in his Diary of My Songs, cautions
singers not to overdo the dramatic emotions of his songs. He recommends a
simple seriousness, especially in some of the humorous songs where a knowing
look would spoil the irony.20 Bernac sums up the French aesthetic in writing
that lyricism and passion are not absent in French music, rather that "emotions
and feelings are refined, purified and controlled by reason."21
Ornamentation
For the most part, added ornaments were not needed or desired in French vo-
cal repertoire, especially by the final decades of the nineteenth century. Earlier
in the century, opinion had been mixed. The critic Fraôis-Joseph Ijetis, writ-
ing in 1827, disagreed with the common view that ornaments were not suited to
the French language; he believed that the Italian style had a positive impact on
French singing in general and particularly in ornamentation.22 Most other
French writers, however, warned singers not to go too far. Berlioz made no se-
cret of his dislike of ornamented singing in the Italian style. Romagnesi men-
tioned that some types of songs—gracious, or light and courtly—may tolerate
some tiny embellishments, "but they must be used sparingly so as not to stifle
the basic idea under a web of parasite-like details."23 For tender and sad ro-
mances, he cautioned to "guard against pretentious flourishes, or against
clumsy appoggiaturas [ports de voix\, which some singers take to be expression
but which is only parody. He should avoid those long fermatas on the penulti-
mate note of the phrase that poor taste on the part of several singers has made
fashionable."24 Writing about the romances of Henri Reber, Saint-Saens men-
tioned that Reber "forbade in his songs neither trills nor scales nor arpeggios;
gracious arabesques frequently accompany the lines of his architecture, ara-
202 French Melodies
besques of a very pure style, it goes without saying, and having nothing to do
with those gargoyles of the old Italian school."25
The Performer's New Role
As the nineteenth century progressed, composers included considerable perfor-

mance instructions in their scores. Singers needed only to execute what was
written accurately, without adding any embellishment of their own. Bathori
advised young singers not to add any personal interpretation to Debussy's
songs: only to deliver what is written in the music exactly and to have "com-
plete musical respect."26 When Debussy first heard Claire Croiza, another
champion of contemporary French music, he said, "What a joy to hear my
songs sung as I have written them!"27 Croiza was also a favorite of Faure's. In
her master classes she said of his songs: "we must try to enter into the atmo-
sphere he has created, and not allow our own personalities to disturb it."28 She
also insisted on scrupulous accuracy in following every detail of a composer s
indications: "A singer's first duty is to sing the music as it is written and to give
every note the value the composer has given it. ... Perhaps it is because the pre-
cise composers of today mistrust the possible bad taste of ¡some interpreters,
that they note everything to the least detail. They indicate everything that
should be done and the interpreter has nothing to add, nothing to modify. His
magnificent role is to serve and not to collaborate."29
As early as 1846 Romagnesi had advised that romances should be "sung sim-
ply and following the composers indications."30 At the turn of the century
Jean-Aubry railed against performers who used their technical gifts merely to
dazzle the audience and gain applause for themselves, "whereas their sole duty
is to be the humble and respectful servants of the works."31 Ravel said, "I do
not ask for my music to be interpreted, but only for it to be played."32 Bernac,
too, stressed the importance of being faithful to the work by achieving a preci-
sion of performance; he quoted Stravinsky, who required an aspiring inter-
preter to first be a "flawless executant." This credo, which some today mistak-
enly take as a given for all music, has its roots in the late nineteenth-century
French reaction against the excesses of Romanticism.
Words and Music
The French have always held the beauty and refinement of their language in
high esteem. The wealth of poetry by Paul Verlaine, Charles Baudelaire,
Stephane Mallarme, and others provided composers with rich opportunities
for song settings. Curiously, the relationship between French words and meló-
French Melodies 203
dies was not as mutually supportive as that of German poetry and Heder. The
rigorous rules required for French poetry made setting it to music problematic.
This resulted in frequent prosody mistakes, even in the songs of Berlioz, Bizet,
and Fauré. In an essay on the relationship between words and music in his 1885
Harmonie et mélodie, Saint-Saens complained about the woeful state of text set-
ting, which he found worse at the opera than in the salon. Since French does
not have the strong rhythmic inflection of Italian or the percussive consonant
clusters of English or German, syllable stress is made by duration rather than
force. Because French musicians and poets believed that no syllable had an ac-
cent except either the final masculine or penultimate feminine syllable of a line,
they thought that musical accents could be placed at will, often resulting in a
confusing gibberish.33
Debussy chose to attack this problem by setting poetry in a recitative style,
preserving the subtle stresses and proper syllable lengths. He was compared to
Monteverdi for his natural, speechlike text setting. By contrast, Fauré often
chose to stress the general mood or subtext of a poem. Claire Croiza pointed
out the rhythmic differences between the settings of Verlaine's "Clair de lune"
by Fauré and Debussy: In example 6.ia, Fauré sets the sung silent syllables of
"paysag^" in m. 14 and "masques" in mm. 16 and 17 with rhythmic placement
and melodic contours that unnaturally accent those syllables. In example 6.ib,
Debussy follows the traditional practice of sounding the silent syllables while
deemphasizing their importance, aiming for a more natural, speechlike rhythm
and contour. In learning Fauré s songs, Croiza advised singers to study the po-
ems away from the music so as not to be led astray by the false musical rhythms.
In contrast, she observed that one could declaim the poems naturally using De-
bussy's settings.34 Ravel preferred to set free verse and prose so that he would
not be bound by the rhythmic constraints of the text.
Diction and Articulation
Most French singing manuals from the late eighteenth and early nineteenth
centuries stressed dramatic expression and diction without placing much em-
phasis on vocal technique. French singers of that period prided themselves on
their precise, clear diction and dramatic presentation, but their singing often
lacked line and cantabile. The proximity and popularity of Italian singers at the
beginning of the nineteenth century helped to solve this problem, considerably
improving the technical level of French singing. Diction never lost its primary
place, however, and Romagnesi urged singers to practice speaking the text of a
song before singing it, giving the words serious attention: "In singing, not only
204 French Melodies
Example 6.1. a) Fauré: "Clair de lune," mm. 12-17, Hamelle, 1897, rey. 1908, reprint Dover,
1990; b) Debussy: "Clair de lune" from Fetes galantes I, mm. 5-8, E. Fromont, 1903, reprint
Dover, 1981.
must one stop appropriately at points of repose, like the comma and full stop—
as will have been done when declaiming the words—and not only must one
lean on the long syllables and pass lightly over the short ones, but also it will be
necessary to emphasize all the words much more vigorously than in speaking;
one has to articulate them as if being heard by a slightly deaf person, but how-
ever, avoiding all affectation and grimacing."35
Reynaldo Hahn opposed singing in which virtuosity was the sole interest.
For him, good singing with bad enunciation was as offensive as excellent dic-
tion with poor vocal quality. In his own singing he tried to jfind a balance be-
tween pure vocalism and clear, accurate, and expressive diction. His recom-
mendations in On Singers and Singing include suggestions that we might find
exaggerated today, including expressive breathing and sighing with the mood
of the poem and enunciating double consonants for special emphasis.36 Croiza
also suggested using consonants to help convey the expression of the text: "The
accent in singing is given by the attack of the consonants, but an energetic ar-
ticulation of the consonants does not mean a brusque articulation. Good artic-
ulation must be searched for with suppleness. . . . In Fauré's cSoir,' all the ex-
French Melodies 205
Example 6.2. Fauré: "Soir," mm. 34-37, Hamelle, 1908, reprint Dover, 1990.
pression of the end, 'tes yeux leves au del, si tristes et si doux/ depends on the
way we pronounce the consonant of this word, cdoux.' French has some conso-
nants more 'explosive' than others, and use must be made of them in interpre-
tations. The absence of the tonic accent in French must be supplemented by
the accentuation of the consonants."37 Perhaps she meant elongating the at-
tack of the d in "doих" by making it softer and more aspirated (instead of per-
cussive and dentalized) to complement the rolled chord in the piano accompa-
niment (see ex. 6.2).
Croiza also stressed knowing how "to give more value to certain syllables that
give the words their significance." While a student at Juilliard, I sang in a mas-
ter class for the Swiss tenor Hugues Cuénod, who had made his debut in Paris
in 1928. He encouraged me to vary the length and weight of all the sixteenth
notes in Fauré s "Notre amour" to try to approximate a more natural spoken
rhythm (see ex. 6.3). The underlined syllables in the first line should be slightly
longer: "Afotre amour est chost legere"; and the mute "e" syllables should receive
206 French Melodies
Example 6.3. Fauré: "Notre amour," mm. 3-5, Hamelle, 1897, rev. 1908, International,
1953-
less emphasis than the surrounding vowels, especially if they fall on a strong
beat like "cho^" in m. 3. This is what Romagnesi meant by leaning on the long
syllables and passing lightly over the short ones. This approach, of course, con-
tinues a long tradition: as mentioned in chapter i, in 1668 Bénigne de Bacilly
French Melodies 207
stressed the importance of emphasizing the differences between long and short
syllables.
Ravel incorporated this approach to diction in the very fabric of his settings
of Renard's Histoires naturelles: "The direct clear language and the profound
hidden poetry of the poems in prose tempted me for a long time. My authors
text demanded a particular kind of musical declamation from me, closely re-
lated to the inflections of the French language."38 He believed that diction
must lead the music, and he therefore tried to approximate the rhythms of
everyday speech. Rather than set the mute "e" syllables with their own sung
pitch and rhythm, he chose to elide them with the surrounding syllables and
leave them silent, as in spoken French. For example, the final syllables of "ar-
nve" in example 6.4a, "appeW in example 6.4b, and "regard" and "nua^" in
example 6.^c do not get their own articulated pitches and rhythms. Rather they
are silent and connected to the next syllable, as they would be in speech. In 1907
this radical approach to text setting was much more common in the music hall
than on the concert platform.
These examples also show how Ravel captured the contours of the French
language. A friend of his, the music critic Emile Vuillermoz, observed: "When
Ravel made one of those razor-edged remarks of which he alone possessed the
secret, he used to make a characteristic gesture: he put his right hand quickly
behind his back, described a sort of ironical pirouette, cast down his mischie-
vously sparkling eyes and let his voice suddenly drop a fourth or fifth. In the
Histoires naturelles and L'heure espagnoleone finds this characteristic intonation
in all sorts of places. It is Ravels own voice, his pronunciation, his well-known
mannerisms, that have produced this quasi parlando melody."39
In contrast, Debussy cautioned against a parlando approach. His friend and
biographer Louis Laloy included the following instructions to singers with the
composers approval: "A very widespread prejudice would have it that, in all
modern music, it is only necessary to 'speak,' with as little voice as possible,
without even observing the correctness of the intervals. This is an absurdity. . . .
Even when the melody remains motionless on one note, the purpose is not to
imítate the intonation of speech, which is endlessly variable. It is to paint the
half-light of contemplation and silence by means of a thoroughly musical
metaphor."40 Poulenc was also opposed to a parlando approach. He loved Ital-
ian opera and beautiful singing, and admitted that he wanted some of his larger
vocal works such as Le bal masqué and La dame de Monte Carlo to be sung al-
most in the style of Tosca or Otello. He hated what he called the "pseudo-intel-
ligent singer, usually without voice. . . . I like to hear some singing with a good
208 French Melodies
Example 6.4. Ravel: Histoires naturelles, a) "Le paon," m. 26; b) "Le paon," m. 33; c) "Le
cygne," mm. 15-18, Durand, 1907, reprint Dover, 1990.
French Melodies 209
sauce of pedal (the butter!), without which my music is destroyed."41 In addi-

tion to insisting on impeccable precision and accuracy in diction, Bernac also
promoted the legato, non parlando style favored by Poulenc in most situations.
Most French diction manuals today stress an even, legato approach with fast,
clear consonants and consistent long vowels. For American singers, the empha-
sis on the International Phonetic Alphabet and the lack of tonic accent in
French often omits the issue of long and short syllables altogether. In The Sing-
er's Debussy Marie-Claire Rohinsky comments that classical French singers to-
day favor a diction close to normal speech, not as informal and casual as that
used for cabaret or musical theater, but not as affected with mannerism as that
of nineteenth-century poetic readings. Her observation implies that French
diction at the turn of the twentieth century was more exaggerated than what we
commonly hear today, including more variety in the articulation of consonants
and in the length and weight of vowels. Many early recordings of French singers
such as Croiza, Garden, Bathori, Panzéra, and others are available on CD and
provide fascinating demonstrations of this approach to diction.
Portamento and Vibrato
Early twentieth-century recordings also show French singers using portamento

as a gentle slide or stretch to enhance legato and connect small and large leaps.
As with everything else in the French style, portamento was used with refine-
ment and subtle expression. Hahn distinguishes between the technical and ex-
pressive use of portamento: "Singers cannot totally avoid portamento, for with-
out it, their singing would be too dry. Unconsciously, most singers make
frequent use of this sliding from note to note, barely audible sliding which in
fact contributes smoothness and connection to singing, which makes legato
passages possible. But this subtle sliding should not really be called portamento.
The real thing is much more obvious, more conscious. It is a means very often
abused in singing to add expression; but the quality of expression thus pro-
duced very easily becomes whimpering, silly and odiously vulgar."42 In some
situations he even advises attacking a note gently from below to achieve a cer-
tain expressive effect. In a discussion of Fauré's "Le parfum impérissable" (ex.
6.5), for instance, he suggests inflecting the last note of m. 16, "par/í/mée," from
below.43 In this particular harmonic context, the F-sharp at the end of m. 16
helps to turn the harmony back toward the tonic E major after a brief excursion
into more distant flat keys; Hahn probably believed that shading the pitch from
below would clarify the harmony in addition to being expressive. I might give a
special inflection to the F-sharp at the beginning of m. 17 on "parfuraê," but
210 French Melodies
Example 6.5. Fauré: "Le parfum impérissable," mm. 15-17, Hamelle, 1908, reprint Dover,
1990.
probably not on the previous note and syllable. Hahn does caution that "of
course these little tricks must be used with extreme discretion and tact."44 This
particular trick of gently sliding into a pitch from below, which singers today
rarely employ, is heard frequently in early recordings.
Jane Bathori gives specific recommendations for discreet, expressive porta-
menti in Debussy songs. In "Fantoches" she encourages a slight portamento
from the low G to high A on "sous la charmille" (ex. 6.6a). She also allows some
very gentle portamenti in "Placet futile" (Troispoemes deMallarmé) on "Prin-
cesse!" (ex. 6.6b) and the final word "sourires" (ex. 6.6c). For "La mort des
amants" she cautions, "beware of effects which the tessitura of certain descend-
ing phrases could permit: I speak of too-obvious portamenti which would be in
bad taste."45
Example 6.6. a) Debussy: "Fantoches" from Fetes galantes I, mm. 37-39, E. Fromont, 1903,
reprint Dover, 1981; b) Debussy: "Placet futile" from Troispoemes deMallarme, m. 3,
Durand, 1913, reprint Rolland Pere et Fils, 1971; с) Debussy: "Placet futile" from Trois
poemes deMallarmé, mm. 32-33, Durand, 1913, reprint Rolland Pere et Fils, 1971; d) Ravel:
"Placet futile" from Trois poemes de Stephane Mallarme, m. 17, Durand, 1914, reprint Dover,
1990.
212 French Melodies
Ravel occasionally used portamento as a special effect, marking it into his

setting of Mallarmés "Placet futile" on "coiffeurs," (ex. 6.6d) and "Princesse!"
(see ex. 6.10, m. 6). Poulenc, in keeping with his fondness for Italian operatic
style, allowed portamento as long as it was used with discernment and taste. He
always cautioned against exaggerating any effect that would become vulgar.46
Thomas Grubb explains the difference between the types of portamento ap-
propriate for French versus Italian singing: [in French] "the syllable upon
which the portamento takes place must give way to the following one as soon as
this syllable's note is reached at the end of the portamento, unlike the porta-
mento in Italian where the old syllable is extended indefinitely into the note of
the new syllable."47
Vibrato in the French style is not discussed by most writers and teachers. Ro-
magnesi mentions that some singers employed a light trembling of the voice to
heighten the expression of grief. But he warns that, if used too much, "their in-
cessant little tremolos become the most unbearable thing that one can hear.
Therefore this effect must only be employed relatively rarely where it can be
used in a rational way."48 Early recordings of French singers reveal the typical
small fast vibrato mentioned in previous chapters. Though still mostly a deli-
cate shimmer, it is slightly more pronounced than in recordings of Italian and
German singers.
Rhythm and Tempo
Chapter 5 discussed how a free approach to tempo and rhythm gained force
throughout the nineteenth century. This resulted in performances that were
sometimes described as sloppy and careless, in which the singer may have
sounded unconcerned about coordinating with the accompaniment. Perfor-
mances of the early romances and melodies most likely enjoyed this same kind
of freedom within the tighter bounds of conservative French taste. Songs by
Bizet, Massenet, César Franck, Delibes, Chausson, Chabrier, and Duparc can
be approached within the context of this Romantic style. Bernac, writing in
1970, even allows certain purely vocal effects and rubato for the songs of
Gounod, though he cautions that "we should not go so far as the interpreters of
the period, whose style would undoubtedly be unacceptable to us now."49
Fauré, with his dislike of flamboyance, wished for a simpler, more accurate ap-
proach to rhythm and tempo in his songs. As he became more famous and re-
spected, his attitude influenced the younger generation of composers and per-
formers. They in turn began to insist on a precise and accurate execution of the
rhythms and tempos specified in the music. Yet we must remember that what
French Melodies 213
they considered to be precise and accurate is quite different from what we are
used to hearing today. They were reacting against the excesses of the Romantic
style while still somewhat under its influence.
RHYTHM AND RUBATO
Fauré's son described his father's musical values as follows: "He had a horror of
virtuosity, of rubato and effects aimed at making the audience swoon. He fol-
lowed the printed notes meticulously, keeping strict time."50 Croiza called him
a "metronome incarnate" and advised students working on his "Clair de lune"
to practice with a metronome.51 Marguerite Long, a noted concert pianist who
worked closely with Fauré from 1902 to 1912, believed that she played his solo
piano music as it should be played and—much to his chagrin—designated
herself his disciple and champion. Fauré did not necessarily approve of her vir-
tuoso approach, but she left many fascinating observations in her book Аи pi-
ano avec Gabriel Fauré. "This music requires that one 'play it straight' without
camouflage or trickery," she writes. She also mentions that rubato in Fauré is
for pacing the long line and rounding the phrase, cautioning that "Fauré
thought that the search for effect was the worst sin of all. . . with Fauré it is the
line that counts."52
Long also worked with Debussy and describes his playing: "Debussy was an
incomparable pianist. How could one forget his suppleness, the caress of his
touch? While floating over the keys with a curious penetrating gentleness, he
could achieve an extraordinary power of expression." She relates how he played
his preludes with an almost metronomic precision and became annoyed at pi-
anists who took any liberties with his music. She tells of a pianist who came to
play some of the composer's music for him. At a certain passage the pianist
stopped and said, "Master, according to me this should be 'free.'" Debussy re-
sponded by looking at the carpet and vowing that the offending pianist would
never tread on it again. Rubato in Debussy, Long explains, "adds up to a series
of nuances that are not to be defined unless they are felt. . . . it is confined by a
rigorous precision, in almost the same way as a stream is the captive of its banks.
53
Rubato does not mean alteration of line or measure, but of nuance or ¿¿ш."
Bathori, Croiza and Hahn all complain in their writings about singers'
rhythmic imprecision. They frequently admonish their students to study the
rhythm and to respect the note values. Croiza even begs singers to listen to the
accompaniment and try to sing with it. Hahn argues that "good singing is sus-
tained by steady, firm rhythm, allowing the diction to remain flexible . . .
within the limits set by the rhythm."54 In recordings in which Bathori accom-
214 French Melodies
Example 6.7. Debussy: a) "Les ingénus" from Fetes galantes II, mm. 1-8, Durand, 1904,
reprint Dover, 1981; b) "La flute de Pan" from Trois chansons de Bilitis, mm. 3-5, E.
Fromont, 1899, reprint Dover, 1981.
French Melodies 215
pañíes herself at the piano, her rhythm ranges from absolutely metronomic in
Debussy's "Les ingénus" (see ex. б.уа) to subtly flexible in "La flute de Pan" (see
55
ex. б./Ь). She advises studying the latter song "WiTH RIGOROUS ATTENTION
то RHYTHMIC VALUES," making sure to give the duplets and triplets their
proper length. She cautions that Debussy's marking, Lent et sans rigueurde ry-
thme, "does not mean that one's fancy can be the basis of interpretation, but
that the sounds must succeed one another without stiffness, with a flow so nat-
ural that the phrase seems to be improvised." In a later recording of this song by
Maggie Teyte, the rhythm is precise throughout most of the song, yet she takes
greater liberties with exaggerated ritardandos.
Teyte succeeded Mary Garden in the role of Mélisande, but she caused De-
bussy to complain bitterly about her. In a recording of "La chevelure" from
Chansons de Bilitis, she holds the high note on "bouche," forcing the pianist to
stretch the last four beats of m. 12 considerably (see ex. 6.8a). She also elongates
"tendré" in m. 24, almost putting a fermata on the C-flat. She follows this with
an indulgent portamento down to the Α-flat and an expressive double "s" on
"frisson" in m. 25 (see ex. 6.8b). Croiza cites this song when admonishing
singers to follow the rhythms indicated in the score: "if the composer had
wanted a pause on the word 'tendré,' he would certainly have indicated it; most
singers, however, make one."56 Bathori discusses the same moment in the song:
"The last phrase is full of gentle and subtle emotion: et Urne regarda d'un regard
si tendré·, queje baissai les yeux, separating this with a tiny break from the fol-
lowing words avec un frisson, as simply as possible. And above all, no shiver,'
Claude Debussy said before the last phrase, so painful was it to him to hear the
word frisson sung in this way."57
Bathori describes rubato in Debussy as we have encountered it before: "in an
expressive passage one steals a fragment of time from one measure by hurrying
a bit and, in the following measure, one gives it back by slowing down . . . one
must have good taste and sensitivity." Interestingly, the tempo marking in De-
bussy's "Le faune" is Andantino (Tempo rubato), yet Bathori plays only the
opening three measures with rubato and adopts a metronomically even tempo
from m. 4 on. In contrast, in her recording of Chabrier's "L'ile heufeuse" (see
ex. 6.9) she uses a great deal of rhythmic flexibility and rubato, as the tempo
marking Animato, molto rubato edappassionato indicates. Croiza, too, mentions
"L'ile heureuse," observing that Chabrier demands special treatment: "there is a
rubato which is not in the French character. There is a kind of supple right
hand, whilst the left hand Va son train [goes on its way], unlike the exactness
of a Ravel or a Debussy. . . . For the liberty allowed to the interpreter is not a
Example 6.8. Debussy: "La chevelure" from Trois chansons de Bilitis, a) m. 12; b) mm. 24-25, E. Fromont, 1899, reprint Dover, 1981.
French Melodies 217
Example 6.9. Chabrier: "L'ile heureuse," mm. 1-9, Enoch & Costallat, 1890, reprint
Belwin-Mills, [197?].
218 French Melodies
modern characteristic."58 Bernac notes that the lyricism in this song ap-
proaches caricature, and he writes that the dynamics, nuances, and rubatos
"should be made without fear of exaggeration."59
Poulenc demanded strict observance of note values, and his playing on
recordings reveals steady and accurate rhythm. He said he was against rubato of
any kind, but Bernac acknowledged that "because of his love for the voice, he
was a little less strict at times (which was in fact true) in order to allow the singer
to 'sing,' to prepare certain effects, and to breathe."60
TEMPO AND METRONOME MARKS
Berlioz—known for his rhythmic precision as a conductor—used metronome

markings on his orchestral songs. Given the importance he assigned to rhyth-
mic accuracy, his markings probably reflect the tempos he wanted. Fauré com-
plained that his songs were usually sung too slowly. Croiza acknowledges that
"I am certainly among the interpreters of Fauré who adopt the fastest tempi,
and yet the composer himself often found I did not go fast enough." She asked
him what tempo he wanted for "Apres un revé," and he replied "sans ralentir!"
Fauré deplored the common practice of slowing down at a diminuendo or at
the end of a song, and he rarely specified a rallentando unless he really wanted
it. The Dover score of all sixty Fauré songs includes many metronome mark-
ings, mostly in the second and third collections. These may have been added
with Fauré's consultation or by the original editors. Duparc also lamented that
singers generally took his songs too slowly. He confided to Croiza that "If I had
known what some singers do with them I would never have put any rallentandi
in my songs."61 The tradition of greatly speeding up for crescendos or for high,
loud passages and drastically slowing down for diminuendos^ rallentandos, and
softer moments was so ingrained in most performers that coinposers had to in-
sist, demand, plead, and beg for their music to be performed at a consistent
tempo. Even in a recording by Charles Panzéra, the baritone who sang the Re-
quiem solo at Fauré's memorial service and for whom Fauré wrote L'horizon
chimérique, singer and pianist make a significant ritardando during the last five
measures of "Apres un revé." l
Debussy wrote many tempo indications into his scores to try to control the
flow of the rhythm and the tempo. Bathori advises following all these tempo
markings precisely. She relates an interesting story about the tempo and
metronome marking of "Mandoline": "I remember having sung 'Mandoline'
for Debussy in the tempo indicated in the first edition: allegretto, without a
metronome marking. After the first measures Debussy exclaimed, 'It's twice as
French Melodies 219
fast as that.' When I told him no one would understand the tempo with alle-
gretto as only the indication, he had later editions emended to include the
62
metronome marking." In 1904 recordings of Debussy accompanying Mary
Garden, the tempos for "Green" and "П pleure dans mon coeur" are surpris-
ingly fast yet not hectic.63 They both achieve a simplicity and calmness domi-
nated by a long flowing line.
Ravel wanted exactitude of rhythm, but as pianist Vlado Perlemuter ex-
plains, "Ravel's strict approach doesn't scorn subtlety."64 Perlemuter, who stud-
ied all Ravel's solo piano music with the composer, cautions pianists not to rush
in fast passages or to slow down where it is not marked. Ravel does ask for ru-
bato and a flexible flow of tempo, but only where he specifies it in the score. His
tempo indications often require merely a subtle nuance without a significant
change of speed. In "Placet futile" (see ex. 6.10), Ravel gives a metronome
marking but also controls the flow of rhythm and tempo with meter changes
and several different tempo indications, including rubato, in each of the first
eight measures. The rhythm thus must be precise yet infinitely flexible within a
subtle range of tempos.
Poulenc demanded a steady tempo and disliked slowing down at the end of
a song. In 1960 he declared that "all my metronomic speeds, worked out with
Bernac, are exact." Yet in writing about Tel jour, teile nuit he cautioned that "the
tempo, which no metronome can indicate exactly, must be felt instinctively."65
Bernac observed from countless performances and recordings made at different
times that Poulenc s tempos varied over the course of their long association:
"He played more quickly (too quickly, in my opinion) in his youth than in his
maturity." Bernac also allowed that "Poulenc knew better than anyone how to
take a breath to prepare an unexpected modulation, or to hesitate for a sudden
change of dynamics, but this without ever altering the basic tempo."66 Poulenc
and Bernac provided an important link between the new attitudes of the turn
of the twentieth century and our present modern style. They helped transmit to
the next generation of singers the French ideals of tasteful reserve and scrupu-
lous attention to details in the score.
The most important new ideas presented in this chapter have had to do with
the turn away from virtuosity toward a strict observance of the notated score.
Performers were expected to serve the composer, following the details of the
music with scrupulous accuracy. This was the genesis of an approach that is
sometimes mistakenly applied to music of all earlier periods. French composers
of the late nineteenth century reacted against the earlier excesses of Romanti-
220 French Melodies
Example 6.10. Ravel: "Placet futile" from Troispoemes de Stephane Mallarme, mm. ι-8,
Durand, 1914, reprint Dover, 1990.
cism and demanded accurate rhythms and steady tempos. Portamento and ru-
bato were governed by French "good taste" and used only with subtlety and re-
straint. As they had since the seventeenth century, French composers and per-
formers shunned empty display in favor of a refined purity of expression and a
French Melodies 221
simple, natural approach. The performers' job by the beginning of the twenti-
eth century was to enter into the world of the composer and to leave their own
showmanship—including ornamentation and even expressive interpreta-
tion—behind. For this repertoire, subtlety and simplicity are crucial. Follow-
ing exactly what is written in the score will probably feel familiar and comfort-
able.
Attitudes toward French diction were quite different through the nineteenth
and early twentieth centuries than they are today. In contrast with the even
vowels and fast, light consonants taught today, singers at the turn of the century
used vowels with a variety of weights and lengths, as well as expressive double
consonants. You might consider observing some of Bénigne de Bacillys seven-
teenth-century rules for long and short syllables when you approach French
melodies of the late nineteenth and early twentieth centuries.
Chapter? Second
Viennese School
This chapter discusses works by Arnold Schoenberg and his students

Alban Berg and Anton Webern. These three composers, commonly
known as the Second Viennese School, ventured beyond the increas-
ing chromaticism of late nineteenth-century musk to explore a new
realm of atonal and twelve-tone composition. In addition to demand-
ing new techniques for singing and for solo instrumental and orches-
tral playing, Schoenberg and Berg refined the use of melodrama, or
speaking over music, into the more exacting art of \Sprechstimme, also
known as Sprechgesang, fully exploited in one of the most important
early twentieth-century works, Schoenberg's Pierrot lunaire.
All three composers wrote much vocal music, from songs and
chamber music to choral and stage works. Yet their relationships with
performers, and particularly with singers, were quite different from
those of their predecessors. Their motivation for writing vocal music
was primarily either the text or the composition itself, rather than a
potential performance or a relationship with a singer. Like their
French predecessors and contemporaries, all three reacted against the
excesses of nineteenth-century Romanticism, developed an aversion
222
Second Viennese School 223
to virtuosity for its own sake, and stressed the need to respect the score. Because
of World War I and the resulting political antagonism between Austria and
France, however, these ideas emerged in Vienna largely independent of direct
French influence. The main goal of the Second Viennese School composers,
through performance or intellectual contemplation, was to communicate the
meaning of the music with a clarity and precision that would reveal its inner
workings and construction.
At the turn of the twentieth century, the general Viennese public was particu-
larly conservative in its musical tastes, and waltzes and operetta reigned supreme.
Simultaneously, however, the city was a magnet for avant-garde experimenta-
tion in all the arts, including art, architecture, poetry, literature, philosophy,
and music. Schoenberg's circle of friends included painters Oskar Kokoschka,
architect Adolf Loos, and writer Karl Kraus. Kraus, in his magazine Die Fackel
(The Torch), sought to preserve his beloved German language from deteriora-
tion, choosing his words precisely in order to achieve exact meanings. Architect
Loos eliminated ornament from his designs in order to reveal and clarify the
formal structure of a building. Kokoschka chose to portray the inner reality of
his subjects rather than a literal exterior view. All these ideas had an important
impact on Schoenberg as he embarked on the intellectual journey that Webern
would later describe in his book The Path to the New Music.l
Schoenberg had an overwhelming effect on the lives and work of Berg and
Webern. He was a demanding teacher and imprinted both his ideals and his
difficult personality on his friends and students. They called him "the master"
and said he was as possessive as an octopus.2 He and his inner circle often felt
unappreciated and misunderstood. On the one hand, artists in Vienna in the
first decades of the twentieth century deliberately created work that would
shock and provoke the city's complacent, philistine audiences. On the other
hand, the hostility and isolation they felt drove them into their own creative
worlds. The composers, particularly Schoenberg and Webern, had an ambiva-
lent attitude toward performers and audiences: they wanted their works pre-
sented in public, but, anticipating a lack of understanding on everyone's part,
felt contempt for the very people who could bring their music to life. They
therefore wrote music primarily for themselves and for their art, and in some
cases—such as Berg's Wozzeck and Schoenberg's Moses undAron—didn't know
if their work was even performable.
Yet these composers also believed they were continuing the long tradition of
German /Aus trian masters such as Bach, Beethoven, and Brahms. The young
Schoenberg, Berg, and Webern, along with everyone else at the time, were
224 Second Viennese School
greatly affected by Wagner's music and philosophy. Mahler's towering presence

in Vienna at the turn of the century also had a powerful impact on them. After
taking tonality and free chromaticism as far as it could go, all three, following
Schoenberg's lead, turned to the twelve-tone method of composition, which
they believed was the inevitable destination of German Romanticism. It is
ironic that later their own country, under the Nazi regime, rejected their work
as "degenerate."3
Publications and Editions
Most editions of vocal music by Berg, Webern, and Schoenberg are extremely
reliable because they were scrupulously prepared by the composers. Universal
Edition in Vienna published most of their music, and now Belmont Music in
Los Angeles handles all of Schoenberg's works. Some scores; were published in
the composers' lifetimes, and some were not commercially available until after
their deaths. The New Grove articles on Schoenberg, Webern, and Berg list
publication dates in the works lists. A critical edition of Schoenberg's complete
works, published by Universal, was begun in 1966 and is still under way: the
first three volumes include the songs, and series 6, vol. 24, issued in 1996, in-
cludes Pierrot lunaire and other vocal chamber music. The Dover collection of
Schoenberg songs, based on publications in his lifetime, includes sources and
translations. Pierrotwus first published by Universal in 1914 and reprinted sev-
eral times in Schoenberg's lifetime. In 1990 Belmont issued a new edition in a
larger, easier-to-read format with some earlier mistakes corrected. Universal has
published several critical editions of vocal music by Berg, including Der Wein,
the Alten berg Lieder, and Sieben früher Lieder.
Unlike nineteenth-century German Heder and French melodies, vocal music
by the Second Viennese School composers was not published for the general
public. It was composed out of necessity and inspiration, and published when
possible. Sometimes works would appear in journals or magazines to be con-
sidered and appreciated intellectually, in the same spirit as'Schumann s early
nineteenth-century journal. Sometimes the composer would publish a score
himself. After quarreling with Berlin publishers over their version of his opera
Von heute auf morgen, Schoenberg published the piano-vocal score at his own
expense. Berg paid to publish a piano-vocal score of Wozzeck to show to pro-
spective producers and conductors. Georg Büchner's play] upon which the
opera was based, was well known in literary and theatergoing circles; as the
score circulated and was evaluated, Berg was transformed from an unknown
young composer into the notorious creator of an unperformable atonal specta-
cle. The conductor Hermann Scherchen was intrigued, however, and per-
formed three excerpts with a singer at a well-received concert in 1924. The 1925
premiere of the full opera changed Bergs life overnight: after the public and
critics enthusiastically hailed the performance, Berg was catapulted into inter-
national recognition and success.
This was not the case for Schoenberg, who was always searching for supple-
mental income that would enable him to compose. In the first years of the cen-
tury he worked as a conductor at the Überbrettl, a literary cabaret in Berlin. Al-
though he appreciated good performances of Viennese operetta, he believed
the gulf between serious and popular entertainment was too wide, and he tried
to fill the gap with his early Brettl-Lieder (cabaret songs). Only "Nachtwandler"
was ever performed, and it was poorly received because of an unsuitable trum-
pet player. The rest of the Brettl-Liederwere not published until 1970. He also
spent some of his early years working for the new Vienna-based publisher Uni-
versal Edition, preparing orchestrations, arrangements, and piano reductions
of light music and operetta. Later Universal agreed to publish Schoenberg s
own scores under his supervision.
Webern was also constantly seeking sources of income to support not only
himself but also his mentor, Schoenberg. He found his way into and out of nu-
merous conducting positions: semiprofessional orchestras and choruses seemed
to suit him better than jobs in the theater, which he disliked and never stayed with
for long. He programmed mostly standard German repertoire, as well as new
music of his own and by Schoenberg, Berg, and others. In his diary entries about
his various performances, even in works with voice, he restricts his comments
to how the orchestra played or how he felt about the overall concert; he makes
very few remarks about how the chorus or soloists sang.4 Despite this distant at-
titude, Webern composed many groups of songs, some with piano accompani-
ment, others with instrumental ensembles. Some of these collections were pub-
lished by Universal in the 19205 under his supervision, others posthumously in
the 19505 and '6os. But since Webern prepared all his manuscripts with metic-
ulous care, it was easy for editors to produce final scores that reflected his wishes.
Berg had composed songs long before he started studying with Schoenberg.
His self-taught style of song writing was modeled after the works of Schubert,
Schumann, Wolf, Wagner, and Mahler. He was more interested in French com-
posers than his colleagues, and he was influenced by the music of both Debussy
and Milhaud. According to Berg s wife, Helene, Berg believed that his youthful
compositions, aside from the Sieben früher Lieder, were not of a high enough
quality to be published.5 Nevertheless, after her death in 1976, much previously
guarded material, including the early songs, became available. This shed an en-
tirely new light on Berg scholarship. In 1985 Universal published the unknown
songs in two volumes, and Dietrich Fischer-Dieskau recorded them.6 In 1904,
when the nineteen-year-old Berg first came to study with Schoenberg, the mas-
ter remarked that his new protege was "an extraordinarily gifted composer. But
the state he was in when he came to me was such that his imagination appar-
ently could not work on anything but Lieder. . . . He was absolutely incapable
of writing an instrumental movement or inventing an instrumental theme. You
can hardly imagine the lengths I went to in order to remove this defect in his
talent. . . . [I] am convinced that m time Berg will actually become very good
at instrumentation."7 The Sieben früher Lieder, composed from 1905 to 1908,
were orchestrated and performed in Vienna in 1928. A revised piano-vocal
score, which some think is a reduction of the orchestration, was published at
that time by Universal. The orchestrated version was not issued until 1969.8
A powerful and unfortunate incident had a decidedly negative impact on
Berg the song composer. In 1911, while he was preparing a piano reduction of
Schoenberg s cantata Gurrelieder, he was inspired by the premiere of Mahler s
orchestral song cycle Das Lied von der Erde and set out to compose his own or-
chestral songs. In 1913 he sent his Five Orchestral Songs on Picture-Postcard Texts
of Peter Altenberg (Fünf Orchesterlieder nach Ansichtskartentexten von Peter Al-
tenberg) to Schoenberg, who performed two of the songs on a famously scan-
dalous concert in Vienna.9 The audience's hostile and violent reaction, along
with some critical remarks from the father figure Schoenberg, devastated the
young Berg, who subsequently turned away from song writing. These Al-
tenberg Lieder, op. 4, were not performed in their entirety or published until
1953. In 1929 Berg was commissioned by the singer Ruzena Herlinger to com-
pose a concert aria with orchestra. Herlinger performed Der Wein m June 1930,
at which point Universal published a piano-vocal score. The orchestral score
did not appear until 1966.
Given the importance of the inner workings of these compositions, as well as
the delicate relationships among pitches and lines of counterpoint, all of
Schoenbergs, Berg's, and Webern's vocal works should be performed only in
their original forms and keys. Transpositions are unacceptable.
Performance and Analysis
All three composers were performers of a sort and enjoyed getting together to
play chamber music with a close circle of friends. Schoenberg, self-taught on
the cello, learned basic skills for the violin and viola as well. ¡Webern played the
cello and (less frequently) the viola, and Berg could play the piano and harmo-
nium. Schoenberg presided over famous Sunday chamber music marathons at
which both professionals and amateurs played and discussed the classics. The
object was not to deliver a polished performance, but rather to discover the in-
ner workings and meaning of the music and to experiment with interpretive
choices that might convey and clarify these ideas. Schoenberg was more of a
coach than a collaborator, discussing his philosophy of music during the breaks
in the playing and helping his colleagues to penetrate the harmonic and rhyth-
mic mysteries of musical form and structure. As a conductor he also had lim-
ited technical facility but great insight and understanding, and he applied a
similar coaching style to his rehearsals. Webern adopted this detailed, inves-
tigative coaching style in his teaching and conducting as well. The violinist
Louis Krasner remembers that when Webern gave some lectures on the Bee-
thoven piano sonatas, his passionate analysis of the music was a performance in
itself: "His voice was quiet and sometimes hesitant, but always enormously ex-
pressive and emotionally captivating."l °
As a teacher, Schoenberg was famous for his fascinating analyses of the stan-
dard German repertoire, and he believed that a thorough understanding and
appreciation of a work could result from merely studying the score. As his idol
Brahms had declared when invited to see Don Giovanni at the Vienna Opera,
he heard the best performance in his head when he read the score.11 Unfortu-
nately, after years of bitter disappointment and rejection, Schoenberg's attitude
toward performers became even more sour. Dika Newlin, one of his students at
the University of California at Los Angeles in 1940, recalls him railing against
conductors: "The interpretation, good or bad, is much less important than the
music! Music need not be performed any more than books need be read aloud,
for its logic is perfectly represented on the printed page; and the performer, for
all his intolerable arrogance, is totally unnecessary except as his interpretations
make the music understandable to an audience unfortunate enough not to be
able to read it in print. 'Now do not tell anyone I said this!' he added; 'for there
are those who would stone me for it!'"12 Sadly, he passed this attitude onto
many of his disciples, and it still pervaded the atmosphere in the music depart-
ment at Princeton when I was an undergraduate in the late 19705. (Happily,
this is no longer the case today.) Yet Schoenberg's brilliance in applying analysis
to performance also inspired many performers, including Rudolf Kolisch and
the Kolisch Quartet, Felix Galimir and the Galimir Quartet, pianists Edward
Steuermann and Rudolf Serkin, and conductor Otto Klemperer, who would
carry on his ideals in their performing and teaching.
The Society for Private Musical Performance
A professional organization that did fulfill Schoenberg s performance ideals was

his Society for Private Musical Performance (Verein für Musikalische Privatau-
fführungen). It was founded in Vienna in the summer of 1918 specifically for
the performance of new music. Schoenberg was the president, with Berg and
Webern handling much of the logistics and organization and other close
friends filling out the board. The society presented weekly Sunday concerts of
chamber music, including songs and vocal chamber music as well as reductions
of orchestral works for piano or chamber ensemble. Critics were banned, and
only an invited audience of members could attend. Virtuosity was shunned,
and rather than engaging established artists, Schoenberg favored excellent
young professional musicians who were less well known. Each work had one re-
hearsal leader or coach, usually Schoenberg, Berg, or Webern; pianist Edward
Steuermann, who studied composition with Schoenberg and performed most
of the piano works of the Second Viennese School, and Erwin Stein, another
Schoenberg student, also served as Vortragsmeister. They scheduled plenty of re-
hearsals, and only performed a work when it was thoroughly ready. New works
were repeated on subsequent concerts so that the audience could become fa-
miliar with them, and the programs were not announced in advance (so that at-
tendees would arrive without preconceived expectations). Audience response,
either positive or negativeîncluding the clapping, hissing, whistling, or shout-
ing then typical at public concerts of new music—was strictly forbidden. In the
three years (1918-21) during which the society operated, it gained 320 members
and presented over 100 concerts.13 Rudolf Serkin, who participated m many
society performances, later transferred most of these practices, not including
the ban on applause, to his Marlboro Music Festival in Vermont.
In a statement of the aims of the society, Berg described its performance
ideals as "the attainment of the greatest possible clarity and the fulfillment of all
the composer's intentions as revealed in his work."14 Schoenberg s basic idea,
taken up by the Kolisch Quartet and other performers in his circle, was vividly
described by another society pianist, Lona Truding: "[let] the music speak
without any interference of personal emotions."15 The fundamental goal was
to analyze the music, understand what the composer had in mind, and transmit
this without letting personal interpretations, idiosyncratic mannerisms, or
technical display get in the way. While this approach resembled what French
composers and performers were also striving for at the time (see chapter 6),
Schoenberg claimed to be the source of these new ideas. (The society did, how-
ever, program French and Russian works, and in 1920 Ravel visited Vienna and
performed some of his own music at a society concert.)
Singers who participated in the society's concerts included Marie Gutheil-
Schoder, a well-known soprano at the Vienna Opera, who sang the premieres
of Schoenbergs Second String Quartet, op. 10, with the Rosé Quartet in 1908,
and of Erwartung in 1924. Another favorite was Felicie Huni-Mihacsek, a
Mozart soprano at the Vienna Opera, who sang music of Debussy, Berg, and
Webern. Other singers included Marya Freund and Erika Stiedry-Wagner, ma-
jor players in the story of Pierrot, as well as Emmy Heim, Stella Eisner, Stefanie
Bruck-Zimmer, and Arthur Fleischer.
Music and Text
Of the three composers, Berg had the most Romantic approach to text and mu-
sic. As mentioned above, his earliest compositions were songs that were
strongly influenced by the nineteenth-century Heder tradition. Even in Woz-
zeckhe preserved the Wagnerian principle of music serving drama.16 He was
also somewhat willing to consider the needs of a singer. In writing Der Wein,
because he was uncertain about the capabilities of the soprano who commis-
sioned the work, he composed a vocal line with a more moderate range than in
his other works. He also included some optional lower notes in the vocal part
and instrumental lines to double and support the voice (see ex. 7.1). He hoped
that the work would also be performed by tenors, but it never was in his life-
time.
Webern came to songwriting from an early interest in Heder concerts and
opera. He developed a strong interest in poetry, and it is hard to say whether his
love of poems led him to compose songs, or that working on songs inspired his
exploration of poetry. He was also passionate about nature, wildflowers, and
mountain hikes, and therefore chose poems that expressed a devout and sincere
love of nature. In 1926 Webern met poet Hildegard Jone, and they developed a
close relationship; from then on, he set only her poems. He often turned to
songwriting in times of compositional uncertainty, using the text to give the
new atonal and twelve-tone music a form.
Weberns songs are subtle and intimate, never grandiose or vulgar. The
rhythms and contours of the music fit the text well, achieving expressive set-
tings of the poems. Yet the vocal lines often contain large leaps and challenging
rhythms (see exs. 7.9 and 7.10). Steuermann relates an anecdote revealing We-
berns attitude toward singers: "I expressed a doubt to Webern once about the
possibility of singing some melodic patterns. He said 'Don't worry; we feel and
Example 7.1. Berg: Der Wein, mm. 32-34, Copyright 1931 by Universal Edition, A.G.,
Wien. Copyright renewed. All Rights Reserved. Used by permission of European
American Music Distributors LLC, US and Canadian agents for Universal Edition, A.G.,
Wien.
write, they will find a way.'"17 In 1938, when a performance of his choral work
Das Augenlicht VJ-AS being prepared in London, only the BBC Singers, a select
group of sixteen exceptional musicians, could handle the challenge, because
"no normal choir can do it."18 Webern was more concerned with the poetry,
the expression, and the form than with the singability of his music.
Schoenberg s attitude toward the relationship between text and music most
decidedly favored the music. In a 1912 article, "The Relationship to the Text,"
he explains:
A few years ago I was deeply ashamed when I discovered in several Schubert songs,
well known to me, that I had absolutely no idea what was going on in the poems on
which they were based . . . since the poems did not make it necessary for me to
change my conception of the musical interpretation in the slightest degree. . . . it ap-
peared that, without knowing the poem, I had grasped the content, the real content,
perhaps even more profoundly than if I had clung to the surface of the mere thoughts
expressed in words. For me, even more decisive than this experience was the fact that,
inspired by the sound of the first words of the text, I had composed many of my
songs straight through to the end without troubling myself in the slightest about the
continuation of the poetic events . . . I divined everything that obviously had to fol-
low this first sound with inevitability.19
In fact, in his Das Buch der hängenden Gärten, op. 15, settings of fifteen po-
ems by Stefan George, he achieved exactly this effect: "for the first time [I was]
successful in coming near an ideal of expression and form which I had had in
mind for years. But up to that point I had lacked the strength and certainty to
realize it. Now that I have finally embarked upon this path I am conscious that
I have broken all barriers to past aesthetic."20 The George songs are completely
free from any sense of key or tonality, creating what Schoenberg described as a
world of "sound and mood."21 Martha Winternitz-Dorda, who had absolute
pitch, sang the premiere at a Berlin concert m February 1912. Schoenberg
thought she was very musical and that her voice was beautiful, but he found her
interpretation of the songs "much too dramatic, taking everything from the
words instead of from the music." He admonished another singer in 1913: "not
too accented pronunciation of the text but a musical working out of the
melodic lines!—So don't emphasize a word which is not emphasized in my
melody, and no 'intelligent' caesuras which arise from the text. Where a
'comma' is necessary, I have already composed it."22
In the instructions at the beginning of the Pierrot lunaire score, Schoenberg
cautions the singer/speaker: "It is never the task of performers to recreate the
mood and character of the individual pieces on the basis of the meaning of the
words, but rather solely on the basis of the music. To the extent that the tone-
painting-like rendering of the events and feelings given in the text was impor-
tant to the author, it is found in the music anyway. Where the performer finds
it is lacking, he should abstain from presenting something that was not in-
tended by the author. Otherwise he would be detracting rather than adding."23
In the 1926 foreword to his texts for Die glückliche Handaxia other works, he
continues to explain his approach to text setting, as well as his deep mistrust of
singers: "These are texts: that means they only yield something complete to-
gether with the music. . . . The musician is in the position of being relatively
untouched by his text. He needs it mainly in order to arrange the succession of
vowels and consonants according to principles which would also be decisive
without the text. . . . singers are most satisfactory when they perform it as eu-
phoniously as a well understood do-re-mi."24
In a 1932 radio lecture about his Orchestral Songs (Vier Orchesterlieder, op.
22), we see him softening somewhat in his campaign against dramatic presen-
tations of his texts. He still does not set evocative words such as "Wilder See"
(wild sea) or "Finster Sturm" (dark storm) literally, "reflected by some musical
symbol," but rather treats them on the same level as abstract words in the way
they relate to the meaning of the whole text (see ex. 7.2, mm. 22-23). Yet ne al-
lows that "if a performer speaks of a passionate sea in a different tone of voice
than he might use for a calm sea, my music does nothing else than to provide
him with an opportunity to do so, and to support him."25
Example 7.2. Schoenberg: "Seraphita" from Four Orchestral Songs, op. 22, mm. 20-23,
piano reduction, Belmont, 1968. Used by permission of Belmont Music Publishers.
By 1949 Schoenberg felt the need to address a situation that had gotten out
of hand. In an article entitled "This Is My Fault," he amended the stern warn-
ings in the preface to Pierrot, acknowledging that music must both heighten
the expression of the text and express something provoked by that text.26 Per-
formers and composers had taken his ideas too literally. After World War II,
many composers were writing songs that had nothing whatsoever to do with
their chosen texts. The pendulum had swung too far in the direction of execut-
ing music without any expression of the text or any glimmer of personal inter-
pretation. Performances of serial and twelve-tone music had become cold, stiff,
and detached, which was not what the Second Viennese School composers had
originally intended.
Pierrot lunaire
Stravinsky called Pierrot "the solar plexus of Twentieth Century music."27 It

was revolutionary in many ways, including its treatment of the voice, its con-
struction as a song cycle, and its use of an instrumental grouping that became
the twentieth-century chamber ensemble of choice. The last work in Schoen-
berg's atonal period (1908-21) before he unveiled the twelve-tone method in
1923, it has probably been performed and recorded more than any other vocal
work by Second Viennese School composers. Schoenberg included detailed in-
structions for the vocal performer, yet the exact meaning of these remarks has
proved elusive and controversial. Learning how to deliver the vocal style and
Sprechstimme needed for this work is almost as difficult as learning the music it-
self. Both require a familiarity with its performance history.
BACKGROUND
In early 1912 Schoenberg received a commission from a well-known actress, Al-

bertine Zehme, to compose some melodramas for the poems from Pierrot lu-
naire by Albert Giraud, in the German translation by Otto Erich Hartleben.
Zehme had already been touring Germany reciting these poems to other music
that she felt was unsuitable. She originally requested piano accompaniment,
but as Schoenberg worked on the project he asked whether he could add a clar-
inet, then a violin, a flute, and finally a cello, reaching the full instrumental
complement. (There are also passages for piccolo, bass clarinet, and viola that
are played by the flutist, clarinetist, and violinist.)
Melodrama, or speaking text over music, was not an uncommon dramatic
form in the late nineteenth century. As mentioned in chapter 5, some German
singers were even criticized for overly declaiming music that should be sung.
Schoenberg's friend Karl Kraus was famous for recitation performances of

everything from his own work to plays by Goethe and Shakespeare and to the
operettas of Offenbach, in which he sang/declaimed all the different parts. His
singing style reportedly was similar to Schoenberg's Sprechstimme.,28
Pierrot, however, made musical demands of the speaker far beyond any
melodrama before it. Steuermann was assigned the daunting task of teaching
Albertine Zehme her part. He remembers that "she was an intelligent and artis-
tic woman, but by profession an actress and only as musical as the well-bred
German ladies ofthat time. I still remember that sometimes, in despair of ever
making her feel the exact difference between three-part and two-part rhythm, I
would ask her to dance a few bars of a waltz, then a polka^ alternating in ever
shorter intervals to try finally the first bars of the 'Dandy.'" (see ex. 7.3).29
Steuermann also recalls that though the instrumentalists were all excellent
players from Berlin, they needed twenty-five composer-conducted rehearsals
before the first performance: "Schoenberg's magnetic personality always pro-
vided enough thrill and inspiration. . . . I well remember his trying to get the
speaker away from the tragic-heroine expression she was inclined to assume."30
EARLY PERFORMANCES
At the Berlin premiere on October 16,1912, Zehme appeared in the costume of

Pierrot, the commedia dell'arte clown. She insisted that she be alone onstage,
with the instrumentalists and conductor behind a screen in such an arrange-
ment that they were visible to her but hidden from the audience. The perfor-
mance caused a scandal of sorts, but it also received a huge ¡ovation. The audi-
ence hissed and applauded after the individual pieces, and some poems were
repeated. The reviews of Zehme's performance were mixed, but on the whole
the event was considered a tremendous success. A tour of eleven German. and
Austrian cities followed, Hermann Scherchen conducting^ with enthusiastic
receptions everywhere.
Albertine Zehme felt she had exclusive performing rights to PierrotTor a
number of years after the premiere, but in 1920 Schoenberg (started looking for
other reciters. Some reports suggest that he maintained a friendly relationship
with Zehme, while others claim that he was frustrated and unsatisfied with her.
Steuermann tried to teach the part to his sister Rosa, an actress: "he laid special
emphasis on rhythmic precision and avoidance of singing tone."31 Schoenberg
wanted Marie Gutheil-Schoder to learn the part for a performance with the So-
ciety for Private Musical Performance. She started coaching it with Erwin
Stein, but she said it hurt her voice. Stein replaced her with actress and singer
Example 7.3. Schoenberg: Pierrot-hmaire, "Der Dandy," mm. 3-5. Used by permission of Belmont Music Publishers.
Erika Stiedry-Wagner, who ultimately performed Pierrot in Vienna in 1921, on

tour, and in a 1940 recording with Schoenberg conducting. Stiedry-Wagner
later related that, since she was not a trained musician, she had to work ex-
tremely hard to learn the part. She had to feel it in her body. After she worked
on it alone with Stein, she had thirty rehearsals with Schoenberg and the musi-
cians, who already knew it well. Schoenberg told her "it is absolutely wrong if
you sing the notes. It is not meant this way. He always said that to me. And he
was satisfied because I was—I mean I am not a musician, but I was quite musi-
cal and I could speak and I could give expression. . . . Nobody laughed at me—
nobody. . . . You have to make it very strong and very suggestive and then no-
body laughs. You have to have the expression for it."32
The 1921 performances in Vienna were a great success arid inspired a French
performance in Paris, conducted by Darius MUhaud. The Paris speaker was
Marya Freund, who had sung the Wood Dove in performances of Schoenberg s
Gurrelieder m 1913 and 1914. She prepared a French translation of the German
text after finding the original French poems by Giraud unsuitable. She and
Milhaud had twenty-five rehearsals with the Parisian ensemble. When Mil-
haud, Poulenc, and Ravel visited Vienna later that year, a special evening was
arranged at Alma Mahler's house, where Pierrot was performed twice: once by
Erika Stiedry-Wagner in German, with Erwin Stein (some reports indicate it
was Schoenberg) conducting the instrumentalists from the society perfor-
mance, then by Marya Freund in French with Milhaud conducting the same
ensemble. Milhaud left the following impressions of the evening:
It was a most exciting experience; Schoenberg's conducting brought out the dra-
matic qualities of his work, making it harsher, wilder, more intense; my reading on
the other hand emphasized the music s sensuous qualities, all the sweetness, subtlety
and translucency of it. Erika Wagner spoke the German words in a strident tone and
with less respect for the notes written than Marya Freund, who if anything erred on
the side of observing them too carefully. I realized on that occasion that the problem
of recitation was probably insoluable.33 . . . It was interesting to note the differences
in atmosphere that were produced. Perhaps the French and German languages used
by the singers determined the "colors" that were imposed upon the instruments. . . .
The French language being the softer, made all the delicate passages appear the more
subtle; but in the German interpretation the dramatic passages seemed more power-
ful, while the delicate ones assumed more weight.34
Alma Mahler left this account:

It was first spoken [gesprochen] by Erika Wagner, coached by Schoenberg himself—
then sung [gesungen] by Marya Freund coached by Milhaud. Schoenberg scarcely
recognized his work—but the majority of those present were for Milhaud s interpre-
tation. Doubtless it was more original in Schoenbergs more rhythmical style of ac-
cented speaking than in the song, where one noticed rather the similarity with De-
bussy. The authentic interpretation was naturally the one by Schoenberg-Stein.35
It is interesting to note, in light of the performance ideals discussed in chapter

6, that the French performance was more accurate than the German, even to a
fault. Yet Schoenberg criticized Marya Freund's performance for being too
sung throughout.
In 1940 Schoenberg performed and then recorded Pierrot m New York with
Erika Stiedry-Wagner. In arranging the rehearsals he wrote to her: "We must
freshen up the speaking part fundamentally—at least I am going to try this
time to see if I can bring out perfectly the light, ironic and satirical tone in
which the piece was really conceived. . . . It is difficult to immortalize the au-
thentic performance on records in two weeks."36 The recording, aside from a
dry acoustic and muffled balance, is fascinating as a document of Schoenbergs
intentions: Stiedry-Wagner s recitation is quite reserved dramatically overall,
compared with more recent recordings and performances. Her basic expres-
sion, while not exactly light and satirical, is not overly tragic either. The con-
tours and pitches of her vocal line are only approximations of those notated,
while the few sung pitches are quite accurate. She uses a large palette of vocal
colors, including glissandi with both wide vibrato and straight tone. She makes
interesting transitions in and out of chest voice and head voice, sometimes even
going against the written contours of the vocal line. Her diction is virtuosic,
with a wide variety of consonants, both fast and percussive and slow and sus-
tained. Milton Babbitt, who attended one of the recording sessions, told me
that while Schoenberg was extremely demanding about details in the instru-
mental parts, he didn't have many comments for Stiedry-Wagner.37
At the time of the New York performance and recording, Schoenberg wished
that Pierrot could be performed in English for an English-speaking audience. It
eventually was, to Schoenbergs delight, and a number of translations, includ-
ing one by his student Dika Newlin, have been made. On a 1974 RCA record-
ing, jazz singer Cleo Laine recites the piece in English.
MORE RECENT PERFORMANCES
As singers and instrumentalists became more familiar with Pierrot particularly,

and atonal music in general, it became much easier to learn the work. Steuer-
mann, who taught at Juilliard from 1948 to 1964, remarked that "young people
today [1963] perform the work sometimes after five rehearsals, but one should
not forget what an unknown language this was at the time' and also that works
become easier when already performed."38 Problems of rhythm and ensemble
became less daunting. Singers also came closer to Schoenberg's instructions for
the Sprechstimme: "to transform it into a speech-melody, taking into account
the given pitch. . . . speaking tone gives the pitch but immediately leaves it
again by falling or rising."39
In her 1971 recording for Nonesuch, Jan DeGaetani achieved a new kind of
vocalism, coming much closer than previous reciters to many of Schoenberg's
notated pitches while avoiding a sustained singing tone. In 1992, when I was
performing at the Marlboro Music Festival, one of Jans former students was
rehearsing Pierrot with violinist Felix Galimir, who had worked with Schoen-
berg in Vienna. Mezzo Mary Nessinger, an extraordinary musician and com-
pelling performer, had performed Pierrot before, sounding many of the notated
pitches in a rising or falling spoken tone that moved immediately from one
note to the next. Galimir insisted that this was not the correct Viennese style,
encouraging her not to speak on the pitches because no one in Schoenberg s cir-
cle ever did.
Yet Schoenberg did nótate specific pitches that, if vocalized, do contribute to
the overall sound and architecture of the composition. And he did specify that
the speech-melody should take into account the given pit:ch. Should todays
singers, who have the musicianship to achieve this accuracy, choose a less pre-
cise approach because that is what Stiedry-Wagner did? If Schoenberg had
heard Jan DeGaetani's performance, would he have liked it, or would he have
criticized her, as he did Marya Freund, for singing too much? Since what
Schoenberg said he wanted regarding the Sprechstimme in Pierrot and what he
got in his lifetime were quite different, we can never determine what the "cor-
rect" style really is.
I personally favor using both approaches depending on the situation, which
can change from measure to measure. In busy, noisy passages, where there are
balance issues and thick counterpoint, I often choose a pitch level higher or
lower than notated to produce the most volume and projection without strain.
I also give more energy to the consonants and less definable pitch to the vowels,
while still trying to maintain the relative contours and distance between notes.
Such passages can be found in "Rote Messe," "Galgenlied," and "Enthaup-
tung" (see ex. 7.4). Soprano Bethany Beardslee, celebrated for her many perfor-
mances and recordings of Second Viennese School works, told me that she fa-
vors a lower placement to create atmosphere in "Eine blasse Wäscherin" and
"Nacht," and finds that certain pieces, such as "Der kranke Mond" and "Paro-
Example 7.4. Schoenberg: Pierrot lunaire, "Enthauptung," mm. 11-15. Used by permission of Belmont Music Publishers.
Example 7.5. Schoenberg: Pierrot lunaire, a) "Der kranke Mond," mm. 1-5; b)
"Heimweh," mm. 25-27. Used by permission of Belmont Music Publishers.
die," have more pitch content than others. Where the texture is more transpar-
ent, as in "Madonna," "Der kranke Mond," "Heimweh," or "Heimfahrt," I
find speaking on the notated pitches enhances the harmony (see ex. 7.5). In
"Parodie" (ex. 7.6a) the viola is in canon with the vocal line through much of
the piece, and in Ό alter Duft" (ex. y.6b) the piano doubles the vocal line.
Reciting (without singing!) close to the written pitches helps clarify the instru-
mental lines.
If you want your Sprechstimme to be close to the notated pitches, there are
two approaches. You can learn to sing the pitches first and then move toward a
speaking tone. Alternatively, you can speak the text in rhythm in an approxi-
mate contour and then get closer to the pitches as you learn the music better. I
prefer the first method. For a more detailed discussion of Sprechstimme tech-
nique, see Sharon Mabry s Exploring Twentieth-Century VocalMusic.40
Sprechstimme ¡n Other Works
Schoenberg used Sprechstimme in other works, including Gurrelieder, Λ Sur-

vivor from Warsaw, Moses und Awn, Ode to Napoleon, and Kol nidre. His nota-
tion and instructions for the speaking voice vary. In Gurrelieder, the Sprech-
stimme is notated with empty square noteheads on a five-line staff (see ex. y.ya).
Example 7.6. Schoenberg: Pierrot lunaire, a) "Parodie," mm, 10-15; b) Ό alter Duft," mm.
1-5. Used by permission of Belmont Music Publishers.
Yet in a letter to Erika Stiedry-Wagner, who was preparing for a performance of

the speaking part, Schoenberg explained what he desired: "As against Pierrot,
there are no pitches here. The only reason why I wrote notes was because I
thought that I could make my phrasing, accentuation and declamation clearer.
So no speech melodies. The only likeness with Pierrot is in the necessity of re-
maining in time with the orchestra."41 In complaining about the Stokowski

recording of Gurrelieder, Schoenberg chastised both the speaker for singing his
part too much and one of the singers for speaking too much: "Finally I found
out the best thing is to give [the speaker role] to a singer who no longer has the
necessary beauty of voice to sing great parts. It should if possible, be a higher
voice, about Tenor range or high Baritone. For the first performance, which I
conducted myself in 1914,1 had the part of the speaker carried out by Mrs. Al-
bertine Zehme. She was very, very good and it was quite possible to make this
piece also with a woman's voice, because it is orchestrated in this manner."42
In Moses undAron (ex. y.yb), the Sprechsumme'is notated with X'ed note-
heads on a five-line staff, while in Die Jakobsleiter, Die glückliche Hand, and
Pierrot it is notated with regular noteheads with X 's on their stems. In A Sur-
vivor from Warsaw and the Ode to Napoleon (ex. y.yc), Schoenberg used a one-
line staff with regular notes distributed above and below it.
Berg used Sprechstimme for a few isolated moments in songs and more ex-
tensively in his operas Wozzeck and Lulu. In his 1929 article "The Preparation
and Staging of Wozzeck" which was written to accompany the performance
materials, Berg discussed the different kinds of vocal writing in the score indi-
cated by "recitative," "parlando," and "cantabile":
It is important to differentiate between these directions, which indicate where a

cantabile, that is a bel canto treatment of the voice is clearly required and feasible, and
where it is not. . . . The following remarks amplify the instructions for the execution
of the Sprechstimme as given in both the full score and the vocal score. No singing,
under any circumstances! Still, the pitches are to be stated and held exactly as indi-
cated by the notes, but held with the tone quality of the speaking voice. To be sure,
such a speaking voice need not be restricted to chest tones throughout. Head tones
are also possible, even necessary, since the normal speaking voice is often too low and
of too limited a compass. For this reason these spoken pitches are placed where they
will cause the singer no more difficulty or harm than the sung tones suitable to the
different registers and to various changes in expression.
Only in those cases where, in spite of these directions, an unnatural and mannered
timbre is unavoidable, is it permitted to execute the spoken melody in a narrower
compass, within which, however, "the relationship of the individual pitches to one
another" must be absolutely maintained.43
Berg's instructions thus differ from Schoenberg's in that he wants the singer to
sustain the notated pitch with a speaking tone, rather than move continuously
toward and away from it. Berg was just as demanding and exacting in rehearsals
as Schoenberg and Webern, but he was warmer and more appreciative of per-
Example 7.7. Schoenberg Sprechstimme notation: a) Gurrelieder, b) Moses undAron\ c) Ode

to Napoleon. Used by permission of Belmont Music Publishers.
formers' efforts. Webern did not include Sprechstimme in his vocal composi-
tions.
Dynamics and Articulation
In a 1929 letter about the preparations for his opera Von heute auf morgen,
Schoenberg revealed his thoughts on dynamics and articulation:
Above all: my music calls for singers who are able to sing 9/10 of their parts between
pp and mfy in order to achieve appropriate climaxes with an occasional forte and with
very few fortissimos. In other words, they should not just scream the text all evening
(as is too customary in Berlin and elsewhere in Germany). . . . The singers must be
able to sing piano and mezza voce, and legato when that is called for; even in détaché
passages, their style must remain cantabile. They must cut their tones short only
where staccato is indicated; but this is for purposes of special musical characteriza-
tion, not because of the text. . . . The singing must always remain dignified. The
singers must never characterize-at the expense of vocal beauty, must never exaggerate.
Better colorless than crude—it is not necessary to "help out" my music; in itself it is
so characteristic that, when it is performed correctly, all the characterization is auto-
matically present. This has proven to be the case with all my works up till now.44
Schoenberg also disapproved of too much portamento and vibrato, exagger-

ated sforzandi, and the familiar Romantic tradition of speeding up with a
crescendo and slowing down with a decrescendo.45 He did indicate where he
wanted portamenti with a wavy line connecting two notes (see ex. 7.8).
Webern, another perfectionist about detail, wanted every note to be clear
and just right. His most important musical goals were clarity, comprehensibil-
ity, emotion, and expression; unfortunately, most audiences in his lifetime
found his works incomprehensible. He often preferred soft dynamics in his
music (see ex. 7.9), and many of his songs are sparse and transparent minia-
tures. As a conductor, Webern sometimes spent so much time perfecting the
opening measures of a work that he didn't have enough time to rehearse the rest
of the program. Yet his conducting was described as emotional and sensitive,
achieving an intensely intimate expressiveness. He believed that "every note has
its own life"46 and deserved to be heard completely. In his recording of the Berg
Violin Concerto with Louis Krasner and the BBC Orchestra, he never hurries,
Example 7.8. Schoenberg: Das Buch der hängenden Garten, op. 15, "\ζ" mm. 13-15,
Universal, 1914, reprint Dover, 1995.
Example 7.9. Webern: "Die Sonne" from Sechs Lieder, op. 14, mm. 4-6, Universal, 1924.
lingering lovingly over every harmony and caressing every turn of phrase. His
friends thought he was the greatest conductor since Mahler,
Tempo and Rubato
To counteract the excessive abuses of rhythm and tempo common in the late
nineteenth century, the Second Viennese School composers wrote many more
tempo indications into their scores. They also controlled ¡the flow of tempo
with notated rhythms, which became dauntingly intricate and complex. In
"Des Herzens Purpurvogel" from Drei Liedery op. 25 (ex. 7.10), for instance,
Webern constantly adjusts the pacing with ritardandos and A tempos while bar-
ring groups of notes across the bar lines. ;
Yet all three composers remained still under the influence of the Romantic
tradition, and a subtle flexibility was part of all their music making. In an in-
terview with Joan Allen Smith, Felix Galimir discussed rhythm and flexibility
in rehearsing with Schoenberg and Webern:
They were terribly meticulous about rhythms, and you know, that these sixteenths
or the triplet comes after the second sixteenth, and you played and you finally could
Example 7.10. Webern: "Des Herzens Purpurvogel" from Drei Lieder, op. 25, mm. 25-32.
Copyright 1956 by Universal Edition A.G., Wien. Copyright renewed. All Rights Reserved.
Used by permission of European American Music Distributors LLC, US and Canadian
agents for Universal Edition, A.G., Wien.
make it just right, and that was it. And when you finally got it and he says, "Yes, but
it sounds stiff and . . . it has to be free," and that was really very important that one
does play the music although very correct but with a certain freedom and not in a
straitjacket because of the complications or the expression. I think the paramount
thing was that one should not, especially because of the row and the intonation and
dissonances, overlook the expressional aspect of the piece. And you know that
Schoenbergs or Weberns music is so terribly expressive.47
Edward Steuermann remembers that when Webern played a new score for
him at the piano, "Webern himself was the freest interpreter of his own music
that could be imagined . . . he played so freely that I hardly could follow the
music, but it was extraordinary."48 Otto Klemperer was similarly struck when
Webern played some of his symphony for him on the piano: "He came and
played every note with enormous intensity and fanaticism, not coolly, passion-
ately!"49
Finding the correct tempo was essential to Schoenberg, but much of what he
wrote over the years about it is extremely contradictory. He complained about
conductors who were too free with violent changes of tempo in his music, but
then he confessed that some of his tempo markings were too exaggerated. In
1926 he wrote that performers should follow composers' metronome markings,
but then in 1950 he cautioned that we shouldn't take them too literally.50 Some
of these discrepancies resulted from the large body of writing that Schoenberg
produced over an extended period of changing performance styles. Many re-
ports say that Schoenberg took very slow tempos when he conducted his mu-
sic. This could be simply because orchestras in the early part of his career found
his music hard to understand and difficult to play. In his 1940 Pierrot recording,
Schoenberg's tempos are generally close to or a little faster than his metronome
markings. Erika Stiedry-Wagner tends to be a bit slower when the vocal line
and diction dominate the instrumental lines. Rubato and changes of tempo
within a given poem are flexible, but not overdone.
Edward Steuermann, in discussing Schoenberg's Six Little Piano Pieces, op.
19, gives the following advice: "In the case of Schoenberg you must try to keep
the balance between unity of tempo and the expressive demands. . . . My ad-
vice is: play always so that the expression and the characters are established ab-
solutely and in the way you feel it. Then try to give the piece the unity of move-
ment which comes out of your feeling. . . . The deviations from the general
tempo should be as slight as possible. The more you are able to give the impres-
sion of a constant tempo, the better."51 Although Schoenberg mistrusted most
performers and counseled them not to add emotion or characterization to his
music, Steuermann was one of his most faithful interpreters. We can rely on his
advice when it comes to style.
Learning This Music
In his 1929 letter about Von heute auf morgen, Schoenberg states, "I think that
only singers who can sight-read these parts and who have absolute pitch can re-
ally learn them reliably."52 Certainly this does not have to be the case today. Af-
ter having had atonal music in our ears for almost a century, we are much more
accustomed to learning works with difficult pitches and rhythms. Singers with
absolute pitch do have an advantage, but even those without perfect pitch can
master this language.
Here are some tips to make the learning process a little easier: learn a work in
layers; deal separately with the rhythms, the pitches, the language, the expres-
sion, and the coordination with the accompaniment. Don't attempt to put it all
together until you feel comfortable with each individual task. In a work with
large intervals, try thinking of them as their smaller inversions; for example,
jumping up a minor ninth is the same as going down a half step and then up an
octave. Calculating intervals like this resembles solving a puzzle, and the more
you do it the better you get at it. Work on tough rhythms and pitches only for
short periods of time. When you reach the point of mental fatigue, stop and
then do more the next day.
I find that if I chip away at difficult pitches over a number of days, eventually
they begin to make sense and to sound5 familiar. At this point, muscle memory
starts to work as well, and, as a friend of mine who sang Moses undAron at the
New York City Opera remarked, "once it's in there, it's in there." He also cau-
tioned that it can be impractical to "mark" this music, either by singing down
an octave or singing half voice, because of the large intervals and wide range
combined with subtle gradations of dynamics. When you need to find your
note from the accompaniment and remember it over other distracting pitches,
a very soft, breathy whistle on the necessary pitch will help keep it in your ear
without anyone else hearing.
But merely executing the correct pitches is not the end of our job. We must
sing our notes in relation to each other and to the accompaniment, just as we
did within the bounds of tonality. In the realm of atonal or twelve-tone music,
it feels a bit different. In twelve-tone music it is possible to learn the row and to
recognize repeated contours and groupings either by hearing them or feeling
them in your voice. In atonal music that is not twelve-tone, most composers es-
tablish some kind of context that determines a hierarchy among the notes.
Pitches are no longer merely consonant or dissonant, but they do seem to have
more or less tension or to belong with what's going on to a greater or lesser de-
gree. In example 7.8., for instance, on the first beat of m. 13, the F-sharp in the
vocal line feels almost like a G-flat in a B-flat-minor chord against the B-flat/E-
flat in the piano. The top A seems like the dissonant note. On the second beat,
however, when the chord in the piano changes, the F-sharp in the vocal line
feels like part of an Ε-major seventh chord. Then the С on "ich" feels like a
more dissonant passing note. The D and E on "meine" have slightly less tension
because of the unisons with notes in the accompanying piano chords. The F-
natural on the downbeat of the next measure, however, seems to have slightly
more tension, but only until the piano chord changes on the second beat, cre-
ating even more dissonance as the F squeezes between F-sharp in the bass and E
and G in the treble. When the voice slips down to E on "Wang?," some ofthat
tension seems relieved, but then a new kind of tension builds toward the sus-
pension on the downbeat of m. 15. Once you get used to hearing and feeling
this music in a new kind of harmonic context, you can shape and inflect it just
as you would Brahms, Wolf, or Strauss.
Schoenberg, Berg, and Webern had an unusually ambivalent attitude toward

both performers and audiences. They felt unappreciated and frequently dis-
trustful of singers in particular. Their vocal music was often written without
specific singers or performance situations in mind.
Like their French colleagues, the Second Viennese School composers wanted
their works presented with clarity and precision. They wanted performers to
transmit the meaning of their music without added personal interpretation.
They used increasingly detailed notation to control every aspect of the music,
including dynamics, articulation, rhythm, and tempo. They expected singers
to execute all these elements accurately while negotiating increasingly difficult
atonal pitches. Yet these composers did not want dry, unemotional perfor-
mances of their vocal music. This repertoire still reflects the influence of nine-
teenth-century Romantic traditions, even though tempo fluctuations, rubato,
and expression were now specifically written into the music. Singers today
should not be intimidated by the challenging pitches and rhythms, but instead
approach this rich music as an extension of nineteenth-century German lieder.
The new issue introduced in this chapter is Sprechstimme. This technique in-
volves a variety of notations and a wealth of confusing, contradictory advice
about exactly how to do it. The crucial points concern how accurately the
pitches, if indicated, should be sounded, and with what kind of tone. Different
composers had different kinds of sounds in mind for specific pieces. Despite all
his demands for precision and accuracy, it is particularly interesting that
Schoenberg found a French performance of Pierrot lunaire too precise and ac-
curate, and too tastefully sung. It is possible that singers today can come closer
than many of the original performers to what these composers actually asked
for. But whether the composers would ultimately approve of this approach re-
mains unanswerable.
Chapters Early Twentieth-
Century Nationalism
This chapter considers nationalistic trends in the first half of the

twentieth century in Russia, Spain, England, and the United States,
focusing on vocal music by Musorgsky, Rachmaninov, Prokofiev,
Shostakovich, Falla, Rodrigo, Britten, Ivés, Copland, and Barber. It
investigates cultural and political issues as well as the folk heritage that
contributed to the distinctive musical flavor of each country. It also
examines what motivated these composers to write vocal music as well
as some of the singers who inspired and championed them; Fyodor
Chaliapin, Maria Olenina-d'Alheim, Galina Vishnevskaya, Victoria
de los Angeles, Peter Pears, Eleanor Steber, and Leontyne Price. It is
much easier to get a sense of these singers' individual styles because of
their many recordings, which include premieres of works written for
them. We can also hear composers playing and conducting their own
works. We can use the recordings to learn these composers' intentions
and to glimpse their working relationships with singers.
In the early twentieth century, greater ease of travel allowed for
more extensive exchange of ideas between different national cultures.
Performances of Russian music in Paris had a tremendous effect on
251
252 Early Twentieth-Century Nationalism
French composers. Spaniards and Americans studying in Paris absorbed the in-
fluence of French composers and included French ideals in their own music.
Spanish-flavored music by Russian composers also influenced the development
of Spanish nationalism. Russian composers performing in the United States
adapted to the demands of American audiences. Friendships between Copland
and Britten, Britten and Shostakovich, and Barber and Poulenc yielded inspi-
ration for composition and performance. Despite this cross-pollination of ide-
ology and tradition, composers still established distinctive national styles, in-
corporating their folk heritage and adapting to the circumstances of their
homelands.
The format of this chapter differs slightly from previous chapters. It consid-
ers questions of text, tempo, rhythm, rubato, metronome markings, and edi-
tions within the context of particular composers and singers.
RUSSIA
The discussion here of Russian national style focuses primarily on vocal cham-
ber music from the "realist" period. I will not address songs by Tchaikovsky,
Glinka, or Rimsky-Korsakov from the salon romance tradition because their
style follows closely that of nineteenth-century European ¡ Romanticism (see
chapter 5). As in chapter 4, opera will be considered only as an aid to discover-
ing the vocal styles created by singers and composers.
Industrialism did not reach Russia until near the end of the nineteenth cen-
tury. Before the 1917 revolution, mucli of the country remained agrarian, with
peasants working on country estates owned by landed gentry. The peasants' life
and culture had changed little over the nineteenth century. Singing accompa-
nied many aspects of peasant life, including work in the fields and religious and
social rituals. Peasant songs that were often not much more than sung speech
captured a wide range of emotions, including lamentation and celebration.
The vocalism was throaty and guttural, with glottal ornamentation.1
The aristocracy, particularly in the cities, was more interested in the latest
operas by Verdi and touring European virtuosos than in new Russian music. In
the i86os a group of composers began gathering to perform and discuss their
own new compositions as well as to study works by European masters. These
five—Mily Alexeyevich Balakirev, Alexander Borodin, Modest Musorgsky,
César Cui, and Nikolay Rimsky-Korsakov—became knovyti as the Mighty
Handful (moguchaya kuchkd) or the New Russian Musical jSchool. They be-
lieved in promoting "musical truth," which, in terms of vocal music, required
Early Twentieth-Century Nationalism 253
words and music to work together to present a realistic flow of action and
drama. Their text settings highlighted natural inflection and phrasing, favoring
the rough, spoken quality of peasant songs. Songs, particularly those by Mu-
sorgsky, presented everyday, commonplace themes with comic or vulgar texts
about children, fishwives, and beggars, in sharp contrast with the lofty love po-
etry found in most European chansons or Heder.
By the start of the twentieth century, Russian artists of all types were collab-
orating, both embracing and rejecting realism. This richly creative environ-
ment spawned such personalities as the impresario Sergei Diaghilev and the di-
rector Konstantin Stanislavsky, and endeavors such as the Moscow Art Theater
and the Evenings of Contemporary Music in St. Petersburg.
The 1917 revolution and succeeding civil war of 1918-20 brought terrible
suffering to Russia. Some artists who came from bourgeois backgrounds fled
the country, and their property was confiscated. Conservatories and theaters
stayed open, but only the most dedicated music lovers attended concerts, be-
cause finding heat and food became all-consuming needs. Conservatory stu-
dents were ordered to perform at Red Army headquarters, factories, and worker
communes in an attempt to lift people's spirits.2 Once the Communist Party
was in power, it encouraged music with a social message and wide public ap-
peal. Gone were the elegant salons of the wealthy and educated, which had fea-
tured solo songs expressing personal emotions. The cultural revolution of 1928
and the state-sponsored Russian Association of Proletarian Musicians increas-
ingly required that music glorify Soviet ideals. Artists who believed otherwise
had to change their views or suffer official criticism and banishment. Some, de-
spite great personal hardship, still managed to find a way to get their musical
messages across.
M u sorg sky and Olenina
Although Modest Musorgsky died well before the revolution, he had been a
radical in his own time and was both misunderstood and underappreciated.
His music challenged the prevailing boundaries of rhythm and harmony and
was considered by many to be coarse, crude, and ugly. Tchaikovsky grudgingly
acknowledged Musorgsky's remarkable originality and believed that he had in-
troduced a new musical language despite his lack of formal training. This fresh
language came from a desire to capture the natural rhythms and dramatic in-
flections of Russian speech. In Musorgsky's words, "my music must be an artis-
tic reproduction of human speech in all its finest shades, that is the sounds of hu-
man speech. . . must become music that is truthful, accurate, but artistic, highly
Example 8.1. Musorgsky: "With Nanny" from The Nursery, mm. i—10, ed. Rimsky-
Korsakov, 1882, International, 1955.
artistic."3 Musorgsky based his song cycle The Nursery on his own transcrip-
tions of typical children's conversations. The changing meters and declamatory
rhythms of the first song, "With Nanny" (see ex. 8.1), were unlike anything that
had come before it in Russian song.
A gifted mimic and pianist, Musorgsky delighted his young cousins as he

sang and accompanied himself in his "enactment" of songs. In his operatic set-
ting of Gogols play The Marriage, he tried to reproduce the inflection and
phrasing of the actors onstage. He sang through this opera for his friends, who
were perplexed by both the harmonic progressions and his antilyrical approach
to ordinary conversations of the play. Musorgsky abandoned The Marriage af-
ter completing the first act, but he triumphed in 1874 with his next operatic at-
tempt, Boris Godunov.
Despite the successes of Boris, The Nursery, and other works, Musorgsky suc-
cumbed to alcoholism and poverty in the last years of his life. He often accom-
panied singers as a means of income, only rarely performing his own songs. He
was grateful for any bit of work or public recognition and became dependent
on his friends for moral and financial support. Penniless and miserable, he died
in 1881 at the age of forty-two. Many of his compositions were left unfinished
and were completed and published posthumously by his friends. Rimsky-Kor-
sakov, who first edited most of his works, smoothed over what he considered to
be coarse harmonies and offensive part-writing. As avant-garde composers in
the early twentieth century came to know Musorgsky's works, they objected
to Rimsky-Korsakov's corrections and tried to uncover the original rough and
innovative music. Musorgsky's music profoundly influenced many later com-
posers.
The singer Maria Olenina-d'Alheim was one of Musorgsky's most impor-
tant posthumous champions, and she helped introduce his vocal music to the
rest of the world. Olenina became a link between centuries: the nineteenth cen-
tury of her childhood and the twentieth century in which she performed and
taught. In 1963, at the age of ninety-five, she coached two young professional
Russian vocalists in songs by Musorgsky. In these master classes she transmitted
her firsthand experience with Musorgsky's world and style.4
Olenina, born to a noble family, had grown up on a country estate. In 1887
her family moved to St. Petersburg, where she sang in prominent salons and
came to the attention of the Mighty Handful. At first she studied voice with
Yulia Platonova, a famous opera singer who had championed music by the
New Russian Music School and worked with Musorgsky in the first staging
of Boris. Through Platonova, Olenina heard stories of Musorgsky and worked
on his music. Platonova encouraged Maria to pursue an opera career, but Ole-
nina had other plans. At the suggestion of the Mighty Handful, she began study-
ing with Rimsky-Korsakov's sister-in-law, Alexandra Purgold-Molas. Purgold-
Molas, an exquisite lieder singer, was not a professional, but she had performed
at many of the Mighty Handful s musical evenings. Attracted to Musorgsky,

she had hoped, in vain, that the great composer would return her affections.
Though disappointed in love, she had remained one of his most gifted and
faithful interpreters. Musorgsky s friends were delighted that she would pass
her knowledge and dedication on to the talented young Olenina.
In 1893 Olenina moved to Paris and became involved with her cousin, writer
and historian Pyotr d'Alheim, whom she eventually married. Together they
created unusual performance experiences, introducing Paris's musical elite to
the works of Musorgsky. Their lecture recitals, known as conferences, featured
commentaries by Pyotr followed by Maria's recitals. Begun in 1896, these events
were a tremendous success, attracting the attention of both the French and
Russian press. In 1901 Debussy reviewed a performance of The Nursery, hailing
it as a masterpiece: "Mme Marie Olénine sang these songs at a recent concert of
the Société Nationale, so well that Mussorgsky himself would have been
pleased—if I may dare speak on his behalf."5 Ravel heard these performances
as well, and his 1907 work Histoires naturelles was clearly inspired by The Nurs-
ery in both text choice and declamation style.6
In the midst of this rich Parisian cultural life, Olenina developed a personal
artistic philosophy based on the idea that a performer should renounce one's
self in the service of the composer and artistic truth. Some of this credo may
have been inspired by Musorgsky s ideals, but many performers in Paris at the
turn of the century were pursuing a similar path. Perhaps she absorbed the pre-
vailing wisdom; perhaps she contributed to it as well.
In the first decades of the new century, Maria concertized in Paris, London,
Moscow, and St. Petersburg. She established a new standard for recital pro-
gramming, constructing groups of songs by a single poet or composer, orga-
nized to highlight the music, rather than presenting audience favorites. Her
concerts, without the use of assisting artists or guest virtuosos, were seen as a
new genre of musical performance, and she was hailed as the first Russian
chamber singer. A review by César Cui declared her a revelation: "Mme. Ole-
nina is an artist of conviction. She does not follow the public, she does not
chase after success and does not toady to public taste. . . . [She] has a medium-
strength mezzo-soprano voice, beautiful, high and even in all its registers. Her
technique is superb and consists in splendid diction, phrasing, rich, varied and
subtle nuances, and a rare ability to impart various timbre to her voice—now
metallic, now gentle and caressing. Every nuance of human feeling is equally
accessible to her—both a childlike, touchingly naive lisp that evokes smiles of
sympathy (Musorgsky's Nursery), and profound, fateful and soul-shaking trag-
edy (Schubert's 'Der Doppelgänger')."7 Other reviewers, while sometimes find-

ing fault with her voice, praised her declamatory expression and captivating
dramatic presentation.
From 1908 to 1918 she and her husband directed the Dom pesni (House of
Song), based in Moscow, which sought to promote the music of Musorgsky
and other nineteenth-century Russian composers. The organization produced
concerts, lectures, and publications and by 1913 had 900 members. It also spon-
sored a competition for translating song texts and harmonizing folk songs;
Ravel won in 1911. After the revolution, Maria and Pyotr fled to Paris, where
they continued their activities in the transplanted Maison du Lied, involving
singers such as Marya Freund, Madeleine Grey, Claire Croiza, and Jane Ba-
thori. After her husband died in 1922, Olenina occasionally participated in con-
certs of new music by French composers, but mostly she continued to promote
and teach the Russian style of Musorgsky. Tragically, she descended into pov-
erty and eventually returned to Russia, alone and forgotten.
Olenina's artistic goals set an example for all Russian chamber singers who
followed. We can follow her advice today when approaching this repertoire: she
believed that a singer must give special attention to the text, dedicate all vocal
gestures to the needs of the composer, and lose oneself completely in the dra-
matic world of the song.8 In her master classes in the 19605, she told singers that
they would find everything they needed in the text and the music.9 They must
perform simply and naturally and not invent or exaggerate. The presentation of
the characters in the songs should, however, be genuine and realistic.10 As she
discussed "The Hobby Horse" from The Nursery, she warned soprano Galina
Pisarenko not to force the melody and sing too sentimentally when the mother
comes to comfort the child: "You made the mother too fancy sounding. . . the
mother speaks in a much more simple fashion. You don't have to make her
sing." Pisarenko then explained that she imagined the mother as an aristocratic
lady, all in lace. Olenina responded with a laugh, "And my mother was also an
aristocratic lady, but she never spoke with me that way."11
In discussing "Evening Prayer," Olenina reminisced about how her own
nanny had taught her to recite her prayers before bed, revealing that she not
only absorbed Musorgsky's desired style of performance but also that she had
grown up in the very world of his Nursery. In the first song from the cycle, she
demonstrated that the dramatic expression and motivation of the text must de-
termine the changing tempo: the little girl is fascinated by various frightening
and funny fairy tales, and the music's rhythm and pacing must respond to her
changing thoughts as she describes different scenes and characters. On the tape
of these master classes, Oleninas voice is somewhat ragged from her advanced
age, yet it still conveys a childlike wonder and excitement and a vibrant flexi-
bility of rhythm and tempo as she half speaks, half sings through the opening of
"With Nanny." Musorgsky writes a series of quarter-note rhythms in changing
meters (see ex. 8.1). It is crucial for the singer to be familiar with the spoken in-
flection of the Russian text, so each quarter note, rather than having equal
length and weight, can be shaded and shaped like natural speech. The singer
must also fully envision each changing character, from the bogeyman to the
queen who sneezes, so that the characterizations can be genuine.
Rimsky-Korsakov, Rachmaninov,
and Chaiiapm
Rimsky-Korsakov is commonly considered to have been both the first architect

of the Russian national style and its first ambassador to western Europe. In the
18705 he collected and arranged a groundbreaking volume of Russian folk
songs. As he did in editing Musorgsky s music, however, he corrected and
smoothed over rhythms and harmonies that he found questionable or prob-
lematic, compromising the musics authenticity. His Capriccio espagnoleofiSSS
captured a Spanish flavor that would later inspire Spanish composers. His or-
chestral concerts at the Paris Exposition of 1889 sparked a western European
fascination with Russian music that would continue into the new century.
Most of his songs were in the traditional Romantic style, melodious and
singable, and by his own admission "conforming to the tastes of singers and of
the public at large."12 In his later operas he perfected the kind of declamatory
recitative heard in Musorgsky s vocal music, composing several leading roles for
Fyodor Chaliapin, a master of this style of singing. According to Rimsky-Kor-
sakov's son, Andrey, the composer "demanded from his interpreters accurate
intonation, tempo and rhythm; he did not tolerate arbitrary cuts, fermatas,
hums or shouts, and so on. But my father never accused Chaliapin of any of
these things."13
Chaliapin was probably the most celebrated singer in Russian history. He
created a sensation with his 1895 debut at age twenty-one with the Imperial
Mariinsky Theater as Mephistopheles in Gounod's Faust, and he soon became
famous throughout Russia and the world. In addition to his magnificent voice,
he was renowned for his intense acting and overwhelming stage presence. Al-
though he left many recordings, a British critic wrote in 1926 that "he is among
all singers the one most in need of being seen in order to be properly heard.
Those who could only know his art from gramophone records can have but a
faint idea of his power of swaying and subduing the audience."14 Stanislavsky
often credited the bass for inspiring his system of method acting and said that
his opera workshop was really a Chaliapin school.15 Chaliapin favored Mu-
sorgsky's music, performing some of the songs and offering the first of his many
unforgettable renditions oí Boris Godunovm 1898. The Russian public, how-
ever, continued to prefer Italian opera, much to his frustration.
Around the turn of the century Chaliapin became friendly with Sergei Rach-
maninov, then a conductor at Sawa Mamontov's Private Russian Opera in
Moscow. The two enjoyed giving concerts together, both in public and at pri-
vate gatherings of friends. These occasions have been described in superlatives:
"two giants stimulated each other and literally produced miracles. This was no
longer singing or music in the generally accepted sense of the words. This was
inspiration breaking out in two great artists."16 Rachmaninov said of Chali-
apin, "I cannot name a single artist who has given me such deep and pure artis-
tic pleasure. To accompany him was the greatest joy of my life."17 In turn,
Chaliapin said of Rachmaninov, "When he is at the piano, I am not singing
alone—we are both singing. He interprets the very soul of the composition
with the utmost delicacy, and if a pause or a suspended note is required, the
singer may be sure that he will indicate them perfectly."18
The relationship proved mutually beneficial. Rachmaninov composed songs
for Chaliapin, one of which they performed for Leo Tolstoy on a particularly
memorable occasion. Rachmaninov also helped the bass learn the parjt of Salieri
in Rimsky-Korsakov s new chamber opera Mozart and Salieri, demanding the
utmost attention in accurately following the score. Rimsky-Korsakoy was sen-
sitive to criticism and reluctant to make changes to suit the singer. Rimsky-
Korsakov may have worried that Chaliapin's charisma would overshadow his
music, yet he and the singer maintained a close relationship and traveled to
Paris in 1907 for concerts with Diaghilev.
Because of his peasant background, Chaliapin had no need to leave Russia
after the revolution. Ironically, unfounded accusations of involvement with the
Bolsheviks tarnished his reputation in the West. Rachmaninov, in contrast, was
from an aristocratic family and left Russia forever in 1917. He became famous as
a concert pianist, performing his own music and other works in Europe and the
United States. He was one of the first virtuoso pianists of the modern era who
took surprisingly few liberties and accurately followed the text of the score. He
left many recordings that display both dazzling technique and rich, warm mu-
sicality. His thrilling virtuosity was not just for show. While he did employ
rhythmic flexibility and gentle rubato, it was always used to shape the musical
phrase and highlight the structure of the piece. Rachmaninov believed that
every composition had a culminating moment, "the Point" as he called it. He
and Chaliapin always tried to construct their performances to reveal that mo-
ment.19
Chaliapins principal voice teacher, Dmitri Usatov, had been trained accord-
ing to the Italian school of Garcia and Francesco Lampera and had been the
first Lensky in Tchaikovsky's Eugene Onegin. He passed on a firm grounding in
bel canto technique as well as a knowledge and love for Russian music. He also
introduced his student to the idea of dramatic intonation or coloration of the
voice. In his book Man and Mask, Chaliapin asks "how must breathing be con-
trolled so as to express a musical situation, the mental state! of a character, with
the appropriate intonation? I am speaking here not of musical intonation, of
the production of such and such a note, but of the colour of the voice which,
even in ordinary conversation, assumes various shades."20 In fact, this quality
of his singing, along with its precise diction, gave the impression of human
speech: "The realism of his declamation was amazing," wrote critic Nikolay Fy-
odorovich Findeyzen. "I sometimes found it positively unbelievable that I was
listening to a phrase which had been learnt from written-down music. It was
live, free speech, adapting itself to convey all the shades of emotion."21
In his recordings Chaliapin used a wide variety of sounds—light and broad,
round and dark—but always easy and never forced, even when the full richness
of his voice poured forth. He was not afraid to make an ugly sound to enhance
the drama, and his articulation could be quite conversational or extremely
legato. He allowed for flexible tempos, which complemented his dramatic in-
flection, and his suspenseful pauses were legendary.
Chaliapirís biographer brings up an interesting point about the title of his
book, Maska i dusha, translated into English as Man and Mask. The Russian
word dusha can be translated as "soul," but it really has a much more subtle as-
sociation that includes a sense of "innermost feelings."22 Chaliapins commit-
ment to revealing his innermost feelings with colorful declaration and honest
dramatic interpretation epitomized the soul of the Russian style.
Prokofiev
Just as Rachmaninov had studied piano, conducting, and composition at the

Moscow Conservatory, Sergei Prokofiev studied all three subjects at the St. Pe-
tersburg Conservatory. In 1907 he began to participate in the Evenings of Con-
temporary Music concerts as a pianist, revealinghimself as a controversial and
innovative young composer. In 1913 he visited Paris, where he heard Ravel and
Stravinsky. He met Diaghilev, which paved the way for future collaborative
theater projects. He also met Anna Zherebtsova-Andreyeva, a voice professor
from the St. Petersburg Conservatory. Her husband, a tenor, was performing
opposite Chaliapin in a sensational Diaghilev production of Boris Godunov in
London. Zherebtsova-Andreyeva appeared often at the Evenings of Contem-
porary Music and performed several of Prokofiev's songs, including the first
version of The Ugly Duckling, op. 18. This extended through-composed song
demonstrates Prokofiev's reverence for Musorgsky and the particular influence
of The Nursery. The text is Prokofiev's own version of the Hans Christian An-
dersen tale. The music highlights the changing narrative and different charac-
ters of the story with sections of declamatory recitative interspersed with lyrical
passages (see ex 8.2). As with The Nursery, the singer must "enact" the story us-
ing expressive, dramatic diction and a large palette of vocal colors.
Prokofiev left Russia in 1918, committed to return someday despite his mixed
to negative feelings about the Bolshevik regime. He spent several years in the
United States but became frustrated with the conservative taste of American
audiences. He felt overshadowed by Rachmaninov, who had wider public ap-
peal as a pianist and composer. While he tried to simplify his compositional
language, he didn't want to compromise his artistic standards. He returned in
1923 to Paris, where he married the Spanish-born singer Lina Llubera, who of-
ten performed his songs with him on concert tours. In 1932 Prokofiev revised
and orchestrated The Ugly Duckling so that they could perform it together when
he conducted orchestral concerts.
Llubera also helped prepare French translations of his vocal works for publi-
cation. Prokofiev wanted his vocal works to be accessible to as wide a public as
possible and often participated in or sanctioned performances of his songs and
operas in French or English. Yet because they were most often conceived with a
Russian text for a Russian-speaking audience, he preferred them in the original
language.
European audiences appreciated his music and he had many successes, but
the dominating presence of Stravinsky kept him in second place. In the spring
of 1936 he returned to Soviet Russia, where he adapted his musical language to
suit the demands of the Communist Party. Russian singers Nina Dorliak and
Zara Dolukhanova performed some of his songs. As the years went on, he
wrote fewer songs and more large-scale works for public occasions and for the
theater. Ironically, while his music enjoyed increasingly positive receptions in
Europe and the United States, it met with more criticism and restriction in the
Soviet Union. By 1948 many of his works were withdrawn from public perfor-
Example 8.2. Prokofiev: The Ugly Duckling, op. 18 (first version), mm. 116-25, Breitkopf &
Härtel, 1922, reprint Belwin-Mills, 1970.
manees, his concert tours abroad were canceled, and his wife was arrested and
later sent to a labor camp.
Prokofiev left a legacy of solo piano recordings that reveal his personal style.
Soviet recordings from midcentury also document state-authorized perfor-
manees of his larger works from that period. Poulenc observed that Prokofiev
played "quite straightforwardly. . . without any sham 'effects' and with impec-
cable technique." Another reviewer evaluated his playing as "brilliant, rather
dry, but extremely polished, pure and finished . . . perfectly balancing spon-
taneity with rigorous precision."23 In his book The Great Pianists, Harold
Schonberg writes that "as a pianist Prokofiev was the New Man of the century.
He had little in common with the past, and his playing was completely origi-
nal."24 In a recording of some small piano studies entitled Visions fugitives, Op.
22, the composer uses a subtle and flexible rubato to shape the line, often
adding slight ritardandos at the ends of phrases. Fast, brilliant passages almost
sound rushed, yet everything he does enhances the musical gesture. Nothing is
heavy or overdone; slow, introspective passages are touching and a bit sad with-
out being overly serious. His playing of these miniatures might suggest how he
accompanied his songs.
Shostakovich
Dmitri Shostakovich, fifteen years younger than Prokofiev, entered the St. Pe-
tersburg Conservatory just after the revolution. He was a brilliant pianist and
earned money in the difficult war years improvising accompaniments for the
cinema. As a composer, he achieved early fame with his First Symphony. In a
sense he led a double life, keeping his public persona and ideology separate
from his private thoughts and feelings. He produced some symphonic works
that appealed to the masses and supported Soviet doctrine, while writing "seri-
ous music for 'the desk drawer"'25 that had no chance of public performance
but expressed his true creative ideals. During his career he both enjoyed
tremendous success and suffered terrible criticism and humiliation, depending
on the whims of the Communist Party. His music has been criticized as Soviet
propaganda, yet much of it contains a bitter, ironic condemnation of the very
institutions it was meant to celebrate.
As a pianist, he was known for incredibly fast tempos. He had natural facil-
ity and technique and rarely practiced. Nathan Perlman, a fellow pianist in
Leningrad (St. Petersburg), describes Shostakovich's playing: "It was an alto-
gether idiosyncratic way of playing. . . . Shostakovich emphasized the linear
aspect of music and was precise in all the details of performance. He used little
rubato in his playing, and it lacked extreme dynamic contrasts. It was an 'anti-
sentimental' approach to playing which showed incredible clarity of thought.
. . . His Chopin playing didn't resemble anything I have heard before or since.
It reminded me of his performances of his own music, very direct and without
much plasticity, and very laconic in expression."26 By the 19308 he performed

only his own music. Cellist Arnold Ferkelman enjoyed collaborating with him
on his cello sonata: "In general he was very modest, and it was very easy to play
with him. He had no sense of ambition or pride, and wasn't offended if one
made a suggestion. Most of his comments in regard to his sonata related to the
tempi. Dmitri Dmitriyevich was a brilliant pianist and had incredible tech-
nique . . . he liked playing quickly and loudly, and he took incredibly fast
tempi. His playing was on the dry side, but on the other hand he played very
loudly, no doubt because of his great force of temperament'"27
As a composer, Shostakovich heard all the details of a work in his head and
notated them with clarity and precision. His scores are flawless and include
both exactly what he wanted and everything the performer needs. At rehearsals
he would listen quietly and answer questions politely but rarely allow any
changes to be made in the score. Occasionally he would make comments about
the tempo and acknowledge that the indicated metronome| markings were not
quite right. In general, he trusted and respected the performers he asked to play
his music and allowed them their own interpretations. Yet soprano Galina
Vishnevskaya admits that Shostakovich's musical authority was unshakable,
and she always followed his scores to the letter: "Before Shostakovich everyone
was intimidated. Even Slava [her husband, cellist and conductor Mstislav Ros-
tropovich] became nervous in front of him as with no one else. When we per-
formed Shostakovich's Blok cycle for him, the hands of that great artist David
Oistrakh shook from nerves."28
In the 19405 Shostakovich, like Prokofiev, found performances of his works
banned. In 1948 he showed a newly composed cycle of songs, From Jewish Folk
Poetry, op. 79, to the soprano Nina Dorliak and asked her to find a mezzo and
tenor to fill out the ensemble. Small, private performances of the cycle for
groups of friends were well received, and the work gained a large underground
following despite a public campaign against it. Even though Dorliak was mar-
ried to the eminent pianist Sviatoslav Richter, Shostakovich always insisted on
accompanying these performances. He believed that the piano lid should al-
ways be fully open, even when accompanying, because the'piano sounded its
best that way. In 1955 the songs were finally performed publicly and subse-
quently recorded with soprano Dorliak, mezzo Zara Dolukhanova, tenor
Alexei Maslenikov, and the composer at the piano. The singing is expressive,
with wonderfully inflected Russian text. Shostakovich's placing is simple and
straightforward, with tempos slightly different than indicated by the score's
metronome markings.29
In the last years of his life, Shostakovich wrote works primarily for a small
circle of trusted friends and colleagues. These included song cycles for mezzo
and bass as well as a number of works for Vishnevskaya and Rostropovich. The
original artists have recorded some of these.30 In her 1984 autobiography, Vish-
nevskaya relates horrifying details of the political entanglements surrounding
many performances and recordings of Shostakovich s music. She also shares her
views on the "Soviet style" of singing: The average audience that came to the
opera was made up of uncultured workers. While they were thrilled to sit in the
great Bolshoi Theater, they were also frequently exhausted from a day battling
the frustrations of Soviet life:
Something had to be done to rouse that impassive crowd and involve them in the
performance. Unless everything was greatly overemphasized, the audience wouldn't
understand. And to communicate the meaning of the operas, the words had to be
given precedence over the music. Hence the excessive emotions, the exaggeration of
words and gestures, the forced voices that have come to characterize Soviet opera.
When Soviet singers perform abroad, they are often criticized for their overacting,
the shrillness of their voices, and a lack of vocal blending and musical phrasing in
their singing. But that is our style, the style of Soviet theater.31
It is a fascinating journey from Chaliapin's first experiments with method

acting and declamatory recitative to the overexaggeration of these elements de-
scribed by Vishnevskaya. Perhaps few other examples could illustrate so starkly
how the cultural and political environment can influence the development of
singing and acting styles.
Editions and Language
Both Rimsky-Korsakov and Rachmaninov prepared editions of their songs for

the general public, and both Russian and European editions are available. The
Dover collection of Musorgsky songs is a reprint of the 1960 edition prepared
by the Soviet State Music Publishing House. It includes new translations and
transliterations. Other Russian and European editions of Musorgsky songs
contain significant discrepancies. A critical edition of Musorgsky is under way
as a collaborative effort of Muzyka in Moscow and Schott Musik International
in Mainz, Germany. It includes all the various versions of his works edited by
Rimsky-Korsakov, as well as orchestrations by Ravel and Shostakovich and
original reconstructions; unfortunately, the volumes containing the songs have
not yet been completed. Prokofiev prepared scores of his vocal music for
Boosey & Hawkes before he returned to the Soviet Union; these are quite reli-
able. The complete works of Shostakovich, published by the Moscow State

Publishing House, includes all the vocal music and is the most dependable,
though a new complete edition is under way. Many editions of Russian vocal
music print only the Russian text in Cyrillic. Some editions include transla-
tions in English, German, or French. However, it is best to preserve the in-
tegrity of the Russian language with its idiomatic stresses and accents.
The Russian vocal music discussed in this chapter is highly emotional and
dramatic, yet genuine and realistic, capturing the natural rhythms of the lan-
guage in a declamatory style. Acting is an integral part of the performance tra-
dition. For these reasons it is especially important for singers approaching this
repertoire to learn the Russian language and to understand both the correct
pronunciation and the meaning of each word. Russian language and diction
classes are not always offered in conservatory vocal programs, and good diction
books are few. Private coaching with a native speaker is the best way to get the
sounds and rhythms of this language in your ear and mouth.
One other cautionary note: the names of Russian musicians have received a
variety of transliterated spellings over the years. For example, the current New
Grove Dictionary of Music and Musicians spells Musorgsky with one "s," while
his name may appear as Mussorgsky on many scores, books, or recordings.
Most recordings or books about Chalíapin spell his name with a "ch" while the
New Grove offers both that spelling and Shalyapin, which ¡is a more accurate
transliteration. Titles of individual songs and larger vocal works are known by a
variety of English translations as well. When searching for Russian composers
and repertoire in library catalogues or other databases, be prepared to try sev-
eral different spellings or translations.
SPAIN
For most of the nineteenth century, Spain remained culturally and politically
isolated from the rest of Europe. Its musical traditions included zarzuela (a
popular operetta-style of entertainment) and flamenco or cante jondo. The folk
songs of flamenco, originally associated with a style of singing as opposed to
dancing, expressed the desperate existence of gypsy life, with such turbulent
emotions as jealousy, rage, anguish, recklessness, boisterousnéss, and generosity
revealing an intensity of spirit similar to that of Russian peasant music. They
often accompanied social rituals such as weddings and funerals; at the latter,
professional mourners were hired to sing songs of grief. The micro tonal orna-
mentation and modal cadences in cante jondo have roots in Moorish, Arabian,
Indian, and Oriental chants. Some songs feature a flexible recitative style with
much rubato and an improvised guitar accompaniment between the verses.
Often in these types of songs the singer will "tune" his or her voice with an ex-
clamation on the syllable "ay" before proceeding with the first line of the sung
text. Other songs are based on traditional dances such as the caña, tona, polo,
seguidilla, and fandango—each with its own rhythmic patterns. Complex
polymeters between vocal line and accompaniment allow for the characteristic
hand-clapping/foot-stomping cross rhythms of flamenco dance.
By the beginning of the twentieth century, many Spanish composers and
performers were studying in Paris. Ironically, they often found inspiration in
the "Spanish" music of Debussy, Ravel, Glinka, and Rimsky-Korsakov. The
Russian composers took the rough, raw flamenco material and gave it an exotic
eastern sheen, while the French composers objectified it with their cool, de-
tached sensibility. Spanish composers absorbed the stimulating Paris atmo-
sphere and included all these elements in their own national music. Oddly, the
prevailing musical taste in Spain at the turn of the twentieth century rejected
Debussy and Ravel in favor of a fascination with Wagner and Strauss. Spanish
music by Spanish composers remained largely unappreciated until much later
in the century.
Falla
Manuel de Falla was a gifted pianist, although he never became as famous as his
compatriots Isaac Albéniz, Enrique Granados, or Ricardo Vines. He had a pas-
sion for composing. In order to support himself and raise money to study in
Paris, he wrote several zarzuelas. The genre frustrated him because he felt it
lacked national flavor and depended on Italian opera style. Nevertheless, the
zarzuelas gave him valuable experience in orchestral writing and managed to fi-
nance his travels.
In Paris he studied with Debussy and Paul Dukas, who taught him the im-
portance of precisely notated scores and the expectation that performers should
play exactly what was written. Ricardo Vines, well established in Paris as a vir-
tuoso, played some of Fallas piano works. In a 1910 performance at the Société
Musicale Indépendante, Falla accompanied a French singer in his Three Songs
on Poems of Théophile Gautier. In 1914 he composed the Seven Popular Spanish
Songs, hoping to perform them while he was still in Paris. However, the pre-
miere actually took place in Madrid the following year at a concert honoring
his return from France. Falla accompanied soprano Luisa Vela and was very
pleased with her voice and her sense of style. These songs, now the most widely
performed in the Spanish repertoire, were greeted at that occasion by complete

indifference.32
THE CANTE JONDO COMPETITION
In 1922 Falla organized a competition for singers of cantejondo (literally, "deep

song"). His goal was to restore the original purity of traditional flamenco sing-
ing, which he believed had become overly commercial in the twentieth century.
In the eighteenth century cante jondo existed in peasant villages as a tough,
gritty, macho style of singing in marked contrast with that used for the languid
and ornate vocal music of France and Italy. In the nineteenth century it became
fashionable in city cafés and cabarets, developing a broad popular appeal, par-
ticularly in Andalusia. Here it embraced a brilliant virtuosity, not at all in the
Italian opera style, delivered in a distinctively hoarse, nasal vocal tone. It also
came to be associated with the alcohol and prostitution found in café life.33 In
the twentieth century, charismatic stars including Pepe Manchena indulged in
elaborate, operatic ornamentation and developed their own schools of fla-
menco singing. Flamenco culture was increasingly lifted from its poor gypsy
roots toward professionalism and commercialization.
Falla excluded all professional singers from his competition and forbade
showy, theatrical ornamentation. He wanted to purge cante jondo of its more
seedy, grimy, and depraved elements. Ironically, these very elements gave the
tradition its distinctive flavor.
EL AMOR BRUJO
Falla's most valuable working relationship was not with a particular singer, but
with Gregorio Martínez Sierra, the librettist for many of his stage works. Not
long after Falla's return to Madrid, the two were asked to create a song and
dance for the famous Andalusian gypsy dancer, Pastora Imperio. This commis-
sion resulted in El amor brujo (Love the Magician). At the time, Imperio was the
most important performer of authentic Spanish dance. Sierra wrote that she
"was truly the empress of all the Spanish dancers who have shaken the floor
boards with the drumming of their heels during the twentieth century." Falla
reported that "I have never worked with more enthusiasm or satisfaction than
during the months it took me to write Love the Magician. The work is an emi-
nently gypsy piece. To write it I employed only ideas of a popular character,
some of them borrowed from Pastora Imperio herself, who has the custom of
singing them, and whose 'authenticity' cannot be denied. . . . Her ear for mu-
sic is such that at first one takes her to be a trained musician, when she knows
not a single note."34
The three songs in the work feature typical elements of cante jondo, includ-
ing an introductory and decorated "Ay!" in the first song (see ex. 8.3). This ges-
ture can also be seen in "Polo" from Seven Popular Spanish Songs (see ex. 8.4).
These melismas should not be executed like Italian coloratura. Rather they
are an attempt to nótate the microtonal kind of embellishment found in fla-
menco song. Falla explains: "They have to be considered, therefore, as extensive
vocal inflections rather than as ornamental turns, although they sound like the
latter when they are translated' into the geometric intervals of the temperate
scale."35
El amor brujo premiered in Madrid in 1915 and was repeated some thirty
times in twenty days. The pianist Arthur Rubinstein left a wonderful descrip-
tion of one of these first performances:
Example 8.3. Falla: "Canción del amor dolido" from El amor brujo, mm. 1-6, Chester,
I9M.
Example 8.4. Falla: "Polo" from Siete canciones populares españolas, mm. 1-9, Max Eschig
Editions, 1922.
The performance took place very late at a theater after the work they were showing
finished. The music was played by six or seven musicians—the type of group that
tends to play in night clubs—and the pianist had an upright piano. But that music
fascinated me—m particular the dance called "The Fire Dance" which Pastora exe-
cuted so marvelously. "Could you lend me the score of the music ofthat dance?" I
asked Falla, "I'd like to do an arrangement for piano to play at a concert." He replied
smiling: "Of course I'll give it to you. Though I strongly doubt that the music will
have the least effect." I did the arrangement basing myself solely on the original. And
when I played it as an encore, the audience literally went mad. I had to repeat it three
times.36
El amor brujo thus started out much like Pierrot lunaire: a chamber work of
music theater written as a showcase for a nonprofessional musician. Falla re-
vised and expanded it into both an orchestral version and a fully staged ballet.
The orchestral version premiered in Paris in 1924 on a concert conducted by
Serge Koussevitzky, during which Prokofiev played the premiere of his Second
Piano Concerto as well as the extensive piano part in the Falla work. In this
arrangement, a concert singer performs the songs from the original cabaret en-
tertainment. The ballet version, without the songs, was produced in many the-
aters in Paris, Buenos Aires, and Madrid, where the 1934 production featured
Pastora Imperio.
Falla clearly had in mind a "folk" singer or traditional cantaora for these
songs. The intimacy of the original chamber version makes it easier to achieve
the intended flamenco flavor of the work. Projecting over the full orchestration
demands the abilities of a trained concert singer, even though the range of the
songs is low and often requires the use of chest voice. The classically trained
singer must find a way to include a taste of that raw, hoarse quality of cante
jondowhile upholding Falla's prescription to follow the score with accuracy and
precision.
Rodrigo
Like Falla, Joaquín Turina, and Granados, Joaquín Rodrigo was another Span-
ish composer whose studied in Paris. He worked with Dukas from 1927 until
1935, and while in Paris he married the Turkish pianist Victoria Kamhi. Al-
though he had become blind from diphtheria at the age of three, he composed
and performed as a pianist well into his old age. His wife was an invaluable help
to him, transcribing the music that he composed in Braille. Rodrigo loved
composing songs for the soprano voice and accompanying such singers as Vic-
toria de los Angeles, Montserrat Caballé, Teresa Berganza, and Pilar Lorengar.
While many of his songs were written for particular well-known artists, he also
wanted his songs to be accessible to all: "I do not object to transposition if this
brings them into the reach of a greater number of singers. However, most of my
songs have been composed for a specific soprano voice, keeping in mind the
particular range and qualities or characteristics ofthat voice."37 He and his wife
heard the debut of the twenty-year-old de los Angeles and were captivated:
"She sang two of Joaquín s songs with a purity of style and diction, and a sensi-
tivity that left us enthralled. And what a voice she had! That twenty-year-old
girl was already a diva!"38 This was the beginning of many years of collabora-
tion.
His Cuatro madrigales amatorios are the most well known and frequently per-
formed of his songs. However, he composed nearly one hundred songs, many
of which are not known or not available in the United States. It has been said
that Rodrigo is not "merely another musician dealing in Andalusian scales and
dance rhythms inspired by flamenco singers, dancers and guitarists."39 His

songs employ a variety of texts and styles. Suzanne Draayer's A Singer's Guide to
the Songs of Joaquín Rodrigo lists invaluable information on how to obtain his
music. She also includes a selected discography with recordings by de los Ange-
les, Caballé, Plácido Domingo, and others. These Spanish performers certainly
sing with a modern technique and sound, but they also provide an inside look
at their own national style.
Editions
Manuel de Falla had a publishing contract with Max Eschig in Paris, and he su-
pervised the publication of his music. As yet there is no critical edition of his
works. Nico Gastéis A Singer's Manual of Spanish Lyric Diction includes a refer-
ence list of Spanish repertoire and publications.40 Draayer also includes publi-
cation and ordering information for music of Rodrigo, and she recommends
the Schott edition of Joaquín Rodrigo: Thirty-five Songs for Voice and Piano>.41
Cockburn and Stokes's The Spanish Song Companion is mostly a collection of
song texts, but it includes a useful discography in an appendix.42
ENGLAND
In the first decades of the twentieth century, London was a magnet for interna-
tional performances, attracting favorite Italian and German opera stars, Rus-
sian artists such as Chaliapin, Olenina, Prokofiev, and Rachmaninov, and Di-
aghilev's touring productions of Russian, Spanish, and French ballets. There
was also a marked countervailing taste for local singers and homegrown enter-
tainments, including music hall shows and drawing-room ballads performed
by amateurs in private settings. Songs by Vaughan Williams, Hoist, Delius, Pe-
ter Warlock, John Ireland, and Gerald Finzi were written in a Romantic style
for nonprofessionals to sing and enjoy. Even conservatory training stressed the
amateur ideal, catering to the "English gentleman" who had no interest in "su-
perficial and insincere technical brilliance."43 Opera had never been as popular
or influential in England as church, oratorio, or concert music. "As a result,"
writes Sydney Northcote, "the English style of singing is at its best in religious
music, and reveals a sympathetic aptitude for the more intimate kind of vocal
music which is exemplified in solo song."44 British audiences did not favor
much contemporary music either, and most English composers were not inter-
ested in pursuing avant-garde ideas. As late as the 19305 the Royal College of
Music had no score of Schoenbergs Pierrot lunaire in its library.45
An English singing manual from 1912 by Harry Plunkett Greene complains

about singers' poor rhythm and diction and reveals a performance environ-
ment more reminiscent of nineteenth-century Germany than of contemporary
France or Russia:
During no period of our musical history have technique and inventiori made such
strides as in the last generation. The Wagner score of thirty years ago, the f error of the
orchestral player and the wonder of the composer, have become commonplace of the
one and textbook of the other. Each has risen to the level of his responsibilities and
played a man's part. The singer alone has stood still. The reasons need not be dis-
cussed here. The fact remains that our platforms are overrun with voices half-devel-
oped and quarter-trained, singers without technique, without charm and without
style, to whom rhythm is of no account and language but the dead vehicle of sound,
whose ambitions soar no higher than the three-verse song with organ obbligato, and
to whom the high-note at the end and the clapping of hands spell the sublimity of
achievement.46
In his Interpretation in Song, Greene also instructed English singers in the

most basic issues of musicianship: don't break the flow of rhythm qr add extra
beats when taking a breath; don't chop up the phrase by taking too many
breaths; don't distort the rhythm by holding a note (complete with arms spread
wide) just for effect. He warned accompanists not to wait for singers or to fol-
low them should they tastelessly distort the rhythm. He did allow singers to add
an indulgent ritardando if they had a particularly long phrase. He bemoaned
Italianate diction that made the text from The Messiah all but unintelligible:
"arnd crorse tar-r-rknaiss-a the peopurlla." He urged singers to create atmo-
sphere and mood in a song, without artificiality. He also suggested that they
disregard expression markings in scores (though perhaps he is referring to the
editorial additions of nineteenth-century publications) because they might not
suit the individual singer and his or her interpretation. Altogether, Greene
painted a fairly dismal picture of British singing at the time.
Britten and Pears
Composer Benjamin Britten and tenor Peter Pears were key figures in revitaliz-
ing this stale and old-fashioned English vocal scene. As in several other famous
composer-singer working relationships (Schubert and Vogl, Poulenc and
Bernac), each had an immeasurable impact on the other in creating a body of
work and a performance style. In addition to their musical collaboration, their
devoted life partnership colored everything they did. Britten told Imogen
Hoist that "if it hadn't been for me, he'd never have been a singer, and although
he might have been happier, lots of people wouldn't have been."47 Pears in-
spired Britten to write opera roles, concert works, and so|ng cycles that they
could perform together. Pears also contributed advice about vocal, textural,
and theatrical matters. Britten wrote music that suited Pears's particular gifts
but also challenged him to greater vocal accomplishments, as well as providing
performance exposure that enhanced his career. As a friend of theirs put it,
"Ben wanted, and found, imagination and sensitivity in Peter's singing and ap-
proach, and from this he conjured what he wanted for his music."48 Pears, in
return, provided the motivation for Britten to become one of the most im-
portant twentieth-century composers of opera and vocal music. Aside from
Gilbert and Sullivan's operettas, the 1945 premiere of Britten's Peter Grimes
marked the first enduringly successful English opera since'Purcell's Dido and
Aeneas. \
Britten was fortunate to enjoy both recognition and financial success. Unlike
his Russian colleagues, he had the freedom to produce his own projects and to
choose his texts and forms. From the English Opera Group to the Aldeburgh
Festival, he was involved in all aspects of planning and performance. He wrote
for churches and theaters, schoolchildren and adults, amateurs and profession-
als, and always had specific performers and situations in mind for the works he
created. As he explained, "almost every piece I have ever written has been com-
posed with a certain occasion in mind, and usually for definite performers. . . .
When I am asked to compose a work for an occasion, great or small, I want to
know in some detail the conditions of the place where it will be performed, the
size and acoustics, what instruments or singers will be available and suit-
able. . . . I prefer to study the conditions of performance and shape my music
to them."49 Although he was internationally known, Britten preferred writing
for people and places that were close to home. He lived and worked in Aide-
burgh, which provided occasions for specific compositions and models for
characters and situations in his operas, coloring the moods of his works. People
who visit Aldeburgh receive a new appreciation for Britten's music when they
see the gray sky, feel the penetrating cold, and hear the sea pounding into the
East Anglican shore.
Britten was not fond of the "pastoral school" of British ¡ composers repre-
sented by Vaughan Williams and Elgar. He didn't like overexaggerated senti-
mentality in compositions or performances, nor did he approve of deliberately
obscure modern music. He was interested in Berg, not sure about Schoenberg,
and didn't feel the need to be original just for the sake of originality. He wanted
to find his own way, but he also freely borrowed from the great masters and felt
a particular connection to the music of John Dowland and Purcell. As he put it,
"it doesn't matter what style a composer chooses to write in, as lon£ as he has
something definite to say and says it clearly."50
Britten s scores are clearly and precisely marked. "To achieve perfect clarity of
expression, that is my aim," he once remarked.51 Steuart Bedford, a conductor
Britten favored because he tried to understand composers' intentions, points
out that there "is hardly a vocal line that does not have some meticulous in-
struction as to how it should be performed, or how it should sound. Even if he
was doing a quick rewrite, Britten would automatically add those kinds of
marks. Such precision can be a terrifying prospect for a singer coming to the
music afresh."52
Britten played the piano with a natural and precise technical ability, though
he was a nervous performer and disliked practicing. Fellow pianists speak with
awe and reverence about his playing and say "it went straight from his musical
brain to his fingers without hindrance."53 Rather than compose solo piano mu-
sic, he preferred to write challenging and virtuosic song accompaniments for
himself to play with Pears. He was not comfortable in the role of conductor, but
musicians loved playing and singing for him.
Britten's complex and contradictory personality made him extremely de-
manding of musicians, but he was also supportive. According to his longtime
assistant Imogen Hoist, "The professional singers and players who work with
him are not just 'good' musicians: they have to be superlatively good, and they
have to be prepared to go on getting better and better all the time! There can
never be any possibility of a compromise, because the music always comes
first."54 Director Basil Coleman observed of Britten's rapport with singers that
"if he saw that they were committed wholeheartedly to him, there was nothing
he would not do to encourage them, and they felt they could give their best
with the composer there to support them."55 The composer wanted his singers
to be completely meticulous yet flexible. According to tenor Robert Tear, "he
managed to give an artist great space in which to work, and yet maintain total
control over the piece as a whole."56 Dame Janet Baker recounts that "he gave
artists the framework in which to work out their vocal problems. Whatever the
music, one knew with him that the tempo would be ideal."57
Though sensitive and patient, Britten could also be relentless in his pursuit
of what he wanted, offering singers breathing and phrasing advice to get the
performances he desired. He was frustrated working with musicians he felt
were not up to his standards, and he loathed bad performances. As he wrote to
Imogen Hoist, "How essential good performances are! I have recently heard
several performances of my own pieces and I felt so depressed that I considered

chucking it all up! Wrong tempi, stupid phrasing and poor technique—in fact
nonsense."58
In some situations he preferred working with amateurs and children who
may not have been technically polished but had the right spirit: "There is some-
thing very fresh and unrestrained in the quality of the music produced by ama-
teurs. What annoys me more is the ineptitude of some professionals who don't
know their stuff. I have no patience with that."59 He composed a good deal of
music for young boys and had definite opinions about their singing style. Early
sketches for A Ceremony of Carols indicate sopranos and altos, but he changed
his mind and published the score for treble (boys') voices. According to his as-
sistant Rosamund Strode, he found the traditional "pure" English cathedral
boys' choir sound "insipid and inadequate."60 He preferred a rougher and
more raucous sound that he felt represented boys' natural robust energy.
A big, beautiful, creamy voice was not as important to Britten as the charac-
ter in the voice. He once told friend and recording producer John Culshaw,
"Frankly, I'm not interested in beautiful voices as such. I'm interested in the
person behind the voice."61 Most important was the singer's ability to commu-
nicate character and drama: "The singers must, of course, have good voices, but
these should be used to interpret the music, not for self-glorification."62 The
singers he wrote for, including Baker, Dietrich Fischer-Dieskau, Kathleen Fer-
rier, and Galina Vishnevskaya, all possessed a depth of soul in their singing as
well as the impeccable musicianship that Britten required. They were also peo-
ple he admired. Britten and Pears traveled to Russia and were warmly hosted by
Rostropovich and Vishnevskaya, who in turn visited Britten and performed at
the Aldeburgh Festival. Britten wrote cello works for Rostropovich and a song
cycle in Russian for Vishnevskaya. His War Requiem was conceived for Pears,
Fischer-Dieskau, and Vishnevskaya; in a gesture of peace and reconciliation, he
wanted to have an Englishman, a German, and a Russian all performing on the
same stage together. Though the Soviet government contrived to make Vish-
nevskaya unavailable for the premiere, she ultimately performed and recorded
it with Britten conducting.63
In order to facilitate characterization, diction was of supreme importance to
Britten: "I have always been interested in the setting of words. . . . My interest
in the human voice has grown, especially in the relation of sound to sense and
colour . . . to the English voice in particular, singing our subtle and beautifully
inflected language. Also I believe passionately in the intelligibility of the
words."64 He particularly admired Pears's ability to color words dramatically
and to infuse them with character. This set a high standard for others (and par-
ticularly tenors) who sing Britten's works, and in fact the composer often didn't
like other tenors singing music he'd written for Pears. In general, he felt that the
first production of a work was definitive; as a result, subsequent performances
with other singers would be considered inferior.
Robert Tear, who worked with Britten and developed professionally in
Pears's shadow, partly modeled his style and technique on Pears. He believed
that if the music written for Pears were "not sung in the Pears fashion, it will
sound wrong-headed." He also believed, however, that Pears's style and tech-
nique were difficult to use in other repertoire. He noted that since the Les illu-
minations songs were not originally conceived for Pears's voice (this cycle was
written for Sophie Wyss, who was Britten's favorite singer until he met Pears)
they are easier to sing.65
Tenors today can listen to the many recordings of Peter Pears singing the mu-
sic written for him and appreciate his unique vocal and dramatic gifts. Trying to
copy his highly individual sound, technique, and style, however, would be a mis-
take. Instead, singers should strive to emulate the underlying qualities of musi-
cianship and dramatic coloration that made his performances so compelling.
THE PURCELL REALIZATIONS
Britten and Pears gave countless recitals together, performing a wide variety of
repertoire in addition to Britten's works. They were particularly famous for their
Schubert cycles. After hearing one of their performances, Poulenc declared to
Bernac that he would never play Schubert again.66 Britten and Pears always in-
cluded a group of Purcell songs. In the late 19305 and early 19405 the music of
Purcell was surprisingly unknown and unavailable in England. Few songs were
obtainable in print, and those that were had old-fashioned realizations of the
bass line. Britten wrote his own realizations, some of which were published. He
believed that "since the accompaniments were originally intended to be impro-
vised, they must be personal and immediate."67 He also thought that since tastes
change over time, the accompaniments should be modified to suit the times.
Thus he rewrote his realizations every few years. He hoped to reintroduce Pur-
cell's songs to a wider public, and he encouraged more people to prepare their
own arrangements "with plenty of boldness and imagination."68
Singers who encounter Britten's realizations of Purcell must understand that
they were intended for piano and a modern style of performance. Purcell per-
formance-practice authority and opera director Michael Burden observes that
"the Britten realizations actually are rather good; but only when played by Brit-
ten! I think the point to emphasize is that they are designed for a very particu-
lar performance approach, and one that is piano based. It also, I think, relies on
a style of singing which we would not today accept. Pears simply sang the songs
the way he wanted to sing them. (Not actually a bad policy, of course!)"69
Pears's particular sound and technique featured a use of legato, phrasing, and
vibrato representative of a mid-twentieth-century style of singing; it reflected
an approach toward early music that was prevalent more than fifty years ago.
Burden points out that the quest for an appropriate or "authentic" Purcell per-
formance style is different today than it was twenty or even ten years ago, much
less fifty. He recommends, like Britten, that performers tailor their own realiza-
tions to be flexible, personal, and not too elaborate. The Britten realizations,
while preserving this spirit of improvisation and flexibility, should definitely be
considered and credited as music by Britten. Singers may even choose to pro-
gram them as twentieth-century music rather than music of the Baroque.
Editions
Britten was extremely meticulous about preparing his scores for publication,
and strove to complete any revisions or corrections before sending a work out
into the world. Usually a score was not published until after the first perfor-
mance. In order to make his wishes absolutely clear to the performer, he was
particular about the visual layout of a score. His assistant Rosamund Strode had
the daunting task of checking for discrepancies between various prepublication
sources, including manuscript, conducting score, and piano-vocal score, in or-
der to ensure the most accurate presentation of the performance instructions.
In 1964 Britten founded Faber Music Ltd. in order to publish his music; most
of his earlier works were published by Boosey & Hawkes.
THE UNITED STATES
In the early twentieth century, the United States attracted touring virtuosos
from Europe and Russia. American audiences flocked to hear famous artists
play and sing well-known Romantic repertoire. With their conservative tastes,
these listeners tended not to be interested in new or avant-garde music. Rach-
maninov was a sensation playing his own music because of its traditional Ro-
mantic language. Prokofiev and other more innovative composers did not
receive the same warm welcome. Yet as the century progressed, American com-
posers found encouragement and opportunity in the form of fellowships,
grants, competitions, and prizes. University music departments and conserva-
tories flourished and produced large numbers of hopeful young composers

who had growing opportunities to have their music performed. Some com-
posers included elements of American folk and popular music in their works,
giving them a nationalistic flavor. Other young composers followed in the foot-
steps of famous Europeans such as Stravinsky, Hindemith, Schoenberg, and
Milhaud, who were all living and teaching in the United States by the late
19405. Yet mainstream audiences remained wary of music that was strange or
unusual, preferring the familiar and accessible. Most well-known performers
and concert series were reluctant to embrace new American music, which
thrived primarily in small, specialized communities of performers and appre-
ciative audiences. Samuel Barber was one of the few composers of vocal music
who managed to bridge this gap, writing music that large audiences enjoyed
and famous artists wanted to perform.
Ivés
Charles Ivés, who made his living as an insurance executive, stopped compos-
ing in 1927. It wasn't until years later that his music became well known and
widely performed. He grew up in Danbury, Connecticut, singing hymns at
church camp meetings and listening to town bands. His father encouraged in-
dependent thinking and musical experimentation, helping him develop his ear
by having him sing a song in one key while playing the accompaniment in an-
other. "This was to stretch our ears and strengthen our musical minds," ac-
cording to Ivés.70 Trained as an organist, he began playing in churches while
still in his teens. By the turn of the century he was attending Yale and compos-
ing marches and other music for the glee club and popular shows. He was also
writing organ and other church works based on Romantic models.
Ivés s early vocal music was in the style of French and German parlor songs.
However, he also composed music for himself that was more adventurous.
When he showed the song "At Parting" to his composition teacher at Yale, he
was chastised for leaving unresolved dissonances. His supportive father replied
that "every dissonance doesn't have to resolve, if it doesn't happen to feel like it,
any more than every horse should have to have its tail bobbed just because it is
the prevailing fashion."71
Although Ivés tempered his imagination for his church singers, they often
found his songs extremely difficult to learn. After his father died, Ivés had only
a few friends who encouraged his compositions. One was Dr. John C. Griggs,
the choirmaster and baritone soloist at the Center Church in New Haven. In
1914, when Ivés showed him the song "General William Booth Enters into
Heaven," Griggs commented, "it's not difficult vocally, its difficult mentally.
Unless the mind grasps and senses a tone from the ear, the'voice can't sing it."
Ivés replied, "If you can't sing it, nobody can."72 In fact Ivés was quite frus-
trated by the limited ability and interest he found in most singers, and some-
times he composed songs with instrumental parts doubling or substituting for
the vocal lines: "The principal reason for this was because singers made such a
fuss about the intervals, time etc. . . . My experience has been, not always but
most always:—the more voice a man has, the less music he has. Apparently in
a singer's education the muscles of the throat get the training, and not the mus-
cles of the ear and brain."73 Most Ives-supervised performances took place be-
tween 1902 and 1924. The majority of these were in the Connecticut churches
where he played the organ. It is impossible to know the level of musicianship
among the church performers, or how his music might haye been received by
more skilled musicians. It may well be that he took a certain pride in his music
being difficult to play, sing, and understand.
In the early 19208 he published some works himself, including 114 Songs. He
sent them to friends and prominent musicians, hoping to stimulate wider in-
terest in his music. The publications met with little initial success, and only a
few isolated professional singers included the songs in their (recitals. It was not
until 1932, when Aaron Copland and Hubert Linscott presented seven of his
songs at the Yaddo new-music festival in Saratoga Springs^ New York, that a
wider public became aware of Ives's songs. Copland admitted that he had "only
an inkling of the existence of the music of Charles Ivés in the twenties."74
More performances followed throughout the 19308 and 405, and many of
his works were revised, published, and finally performed. By the midcentury,
professional singers interested in new music had more experience with chal-
lenging pitches and rhythms, and they had a much easier time with Ives's songs
and could appreciate his quirky combination of band music, hymn tunes, and
early Americana. These were the performances that established a successful
stylistic approach to Ivés and that singers today can use as models. Unlike other
composers discussed in this chapter, the early performances involving Ivés him-
self didn't really launch a definitive tradition; rather, they reflected the conser-
vative attitude of American performers in the early decades of the century.
Copland
Aaron Copland was more than a composer who captured a uniquely American
sound; he was also a writer and critic, a concert organizer, a faithful supporter
of other composers, and an invaluable promoter of American new music in
general. In the 19208 in Paris he studied piano with Ricardo Vines and compo-
sition with Nadia Boulanger. Stravinsky was his hero, and he also admired
Fauré, Milhaud, and Mahler. In 1923 he met conductor Serge Koussevitzky,
with whom he began a long and rewarding collaborative relationship). Kousse-
vitzky, who had shown a devotion to new works and composers throughout his
career (Scriabin, Stravinsky, and Prokofiev in Russia; Ravel, Arthur rflonegger,
and others in Paris), brought this enthusiasm with him when he moved to the
United States in 1924. As the new conductor of the Boston Symphony Orches-
tra, he frequently programmed the music of many American composers, in-
cluding Copland, Barber, Howard Hanson, Roy Harris, Walter Piston, and
William Schuman.
In 1928 Copland joined the New York-based League of Composers. With
Roger Sessions, he produced a series of new music concerts in New York that
presented works by a range of young American composers. He also helped
found the Cos Cob Press to publish new music. In 1932 he visited Mexico and
in the 19405 he traveled to South America, absorbing sounds and colors that
would find their way into his own music. In 1938 he visited England, show-
ing Britten his opera for schoolchildren, The Second Hurricane. This example
may well have inspired Britten to write dramatic works for young performers.
Copland was also especially interested in composing for choreographers and
dancers.
Copland wrote only a few songs, including two sets of American folk songs,
but these are beloved and frequently performed. The 1964 recording of the
Twelve Poems of Emily Dickinson, with the composer accompanying Adele Ad-
dison, offers an exquisite presentation of the work.75 The clear singing and sim-
ple playing are models that still inform the way the songs are usually performed
today. Copland rejected the idea of a definitive performance of his works, but
he did demonstrate "how my music should go"76 through conducting and
playing his own works on recordings. He criticized the emotional excesses of
some Romantic composers but didn't approve of the overly intellectual ap-
proach of many modern composers. Copland wanted performances of his
works to be expressive yet not overly sentimental, and he preferred simple,
straightforward playing without excessive vibrato.
Barber
Samuel Barber had a charmed career: virtually every door he came to opened
for him. He started to compose vocal works at the age of seven, and entered the
Curtis Institute of Music in Philadelphia at fourteen, studying piano, voice,
and composition. His aunt and uncle Louise and Sidney Homer, a well-known
contralto and composer respectively, had a great influence on him. In 1934 he
went to Vienna to study voice and became interested in early music. He
searched the archival manuscript collections in the Vienna libraries and learned
works by Monteverdi, Purcell, Francesco Cavalli, Emilio de Cavalieri, Gio-
vanni Gabrieli, and other seventeenth-century composers. When he returned
to the United States the following year, he performed some of this repertoire,
accompanying himself on a little spinet he had brought back from Munich. A
review of one of these concerts praised Barber as "an artist in every sense, for he
combines a beautiful voice with remarkable ease of style1, and a musicality
which never fails to 'cross the t's and dot the is/ Mr. Barber ¡has learned the in-
valuable art of voice modulation, and wisely believes that an ounce of singing is
worth a ton of shouting."77 Certainly his understanding of early music style
was different from ours today, but he apparently found a gentle, easy musical-
ity and vocalism that made this unfamiliar music appealing.
In 1935 Barber got a job singing on the radio for the NBC Music Guild series.
On the February 4 broadcast he played some of his solo piano works, accompa-
nied himself singing some of his own songs, and sang his\Dover Beach cycle
with the Curtis String Quartet. Barber had written Dover Beach in 1931 for his
own baritone voice range, but he occasionally asked a Curtis friend, mezzo-
soprano Rose Bampton, to perform it. Critics declared of Barber's radio debut
that his music had "singular charm and beauty . . . intelligent music intelli-
gently sung—and with a naturally beautiful voice." They also agreed that he
had won for himself "instant recognition as a triple-fold musician—composer,
singer and pianist."78 Poulenc was so impressed with Barber's singing on a 1936
recording oí Dover Beach that he offered to compose some songs for the young
American. In 1935 Barber also won the prestigious Rome Prize, which secured
his rising reputation as a composer. From then on, all his works were commis-
sioned by well-known artists or ensembles and were premiered in prominent
venues.
In many instances, a text inspired Barber to compose a song setting. In 1947
he came across James Agee's prose poem "Knoxville: Summer of 1915" and was
powerfully moved. It was around this time that Koussevitzk^ requested a piece
for voice and orchestra and that soprano Eleanor Steber also approached Barber
about writing a vocal work for her. No American singer had previously com-
missioned a work for voice and orchestra, and Steber's manager believed that
using new American music would be an innovative way to help promote her ca-
reer. As he composed Knoxville, Barber had some reservations about both Ste-
ber and the orchestral setting, since his original conception had been for an in-
timate piece. As negotiations developed and he realized that she would proba-
bly sing the premiere he modified his plans: "I suspect that what she really
wanted was a big, whooping thing to do with Koussevitzky and the Boston
Symphony, but of course, Knoxvilleis not that kind of piece. I also knew that
Koussevitzky preferred the full orchestra, so I continued with the original scor-
ing, and it was premiered that way."79 During the rehearsals he raised parts of
the vocal line so that Steber s voice could be heard over the orchestra. Later he
considered the possibility of a chamber orchestra version, and he ultimately re-
duced the full orchestration for the 1949 Schirmer score.
In 1950 Eileen Farrell and twenty players premiered a chamber version in
Washington, conducted by William Strickland. The score for this version was
published in 1952. For the same concert, Strickland asked Barber to compose a
song cycle for Farrell. Since the majority of his music was vocal, Barber spent a
great deal of time reading poetry, looking for suitable texts. He had come across
several unusual French poems by Rilke and was happy for an opportunity to set
them. When he asked his friend Poulenc to look at the songs and check his
French prosody, Poulenc immediately wanted to give the premiere with Ber-
nac. They did eventually perform Melodiespassageres, op. 27, but not ¡until after
Barber accompanied Farrell at the Washington concert.
Many of Barbers stage works were performed and recorded in ¡the 19505,
which kept the composer busy. Between large projects he often worked on
songs. Having come across some medieval Gaelic poetry from various sources,
translated into English by several different people, he began composing the
Hermit Songs. A year later he received a commission from the Elizabeth Sprague
Coolidge Foundation for a song cycle for a specific event. It was not until then
that he started searching for the ideal singer to premiere the work. Leontyne
Price was already known for her performances in Porgy and Bess, but she was
still a relative newcomer and had not yet made a recital debut. After her voice
teacher introduced her to Barber, she learned the songs and sang them for him,
thus beginning a long collaborative relationship. She premiered the songs at the
Library of Congress in 1953 with Barber accompanying, and they recorded
them the next year.
This recording, like Copland's and Addison s Twelve Poems of Emily Dickin-
son, shows a wonderful presentation of the composer s wishes. Most perfor-
mances today of the Hermit Songs are modeled after it.80 Barber's playing is
warm and richly expressive. Price's singing is young and fresh, not quite the
opera diva we associate with her later recordings. After Barber got to know and
love her singing, he wrote other works specifically tailored to her abilities. He
also composed Andromache s Farewell, op. 39, expressly for Martina Arroyo and
her particular strengths. For the most part, however, Barber conceived of his
music first and then searched for appropriate performers after: "It seems to me
that the most practical thing is to simply write your music the way you want to
write it. Then you go out and find the interpreters who will' give it voice."81
Barber's music continued the traditions of nineteen th-century Romanticism
and appealed to a broad audience. We don't really need to find a new style in or-
der to sing it. In fact the approach we use could be called the prevailing "mod-
ern" style of singing: it is based on nineteenth-century traditions, but with
more faithful adherence to the details in the score and with fewer wild tempo
fluctuations, rubato, and portamento. This approach is, however, still rich,
flexible, and expressive.
Editions
In the majority of cases, the available editions of twentieth-century American

music offer reliable presentations of the composers' wishes! Ivés prepared the
early editions of his songs and supervised the later corrected and revised edi-
tions of various works. Copland facilitated the publication of many new works
and helped set a high standard of care and precision that composers and pub-
lishers subsequently followed as a matter of course.
This chapter has surveyed how a complex sharing of different national tradi-
tions resulted in the flourishing of distinctive national styles of vocal perfor-
mance in the early twentieth century. Two central artistic goals were common
to all: the importance of communicating the text with a more realistic acting
style, and a growing demand for accuracy and precision in!performance. Be-
cause Paris was the most vibrant center of cultural life at the time, many musi-
cians from Russia, Spain, and the United States performed and studied there,
absorbing the French ideal of serving the score with precision and clarity. Per-
formances that offered a more accurate presentation of the composers' wishes
replaced Romantic offerings of virtuosic showmanship and heightened expres-
sivity.
A straightforward projection of texts and a more restrained, realistic style of
acting became preferable to the overly showy and sentimental displays often as-
sociated with nineteenth-century Romanticism. Chaliapin and Olenina estab-
lished new standards of vocal acting that would have far-reaching influence on
singers in Russia and elsewhere. Britten chose his singers not for the beauty of
their voices, but for their ability to color their characterizations and to convey
the drama.
The particular qualities of the Russian, Spanish, and English languages con-
tributed the most important element to each national vocal style. The folk song
tradition in each culture also conveyed a distinctive emotional and rhythmic
flavor to its concert vocal music. When learning Russian or Spanish repertoire,
the more comfortable you are with the language, the more subtle understand-
ings you will be able to bring to your pronunciation and acting. If you have an
understanding of the folk music heritage of these cultures, it will help you in-
fuse the concert music of Musorgsky, Rachmaninov, Prokofiev, and Shostako-
vich or of Falla and Rodrigo with an appropriate sense of style. For repertoire in
English, we naturally have an easier time dealing with our native language. Re-
member, however, that the inflections and idiomatic expressions of British En-
glish are a bit different from those of American English; the folk music roots are
different as well.
This chapter has also described the gradual emergence of what could be
called the "modern" vocal style. Singing with modern piano and orchestral in-
struments in larger concert halls encouraged a fuller, more projecting style of
voice production. We can hear on recordings from the middle of the twentieth
century onward a use of vibrato, portamento, legato, and articulation that
sound familiar and resemble the ways in which we are trained to sing today.
Chapter 9 Working with
Living Composers
This chapter explores the working relationships between American

singers and composers over the last thirty years. It draws from conver-
sations with several well-known singers, including sopranos Phyllis
Bryn-Julson and Benita Valente and tenor Paul Sperry, who have
worked with composers such as Pierre Boulez, György Ligeti, David
DelTredici, George Crumb, Daron Hagen, and William Bolcom.1 It
also includes anecdotes from Cheryl Bensmans years as a member of
the Steve Reich ensemble,2 as well as my own experiences from twenty
years of working with composers at Princeton University and various
new-music groups.
Most of these stories have parallels in earlier chapters. Composers
write for specific singers, but they frequently make changes for subse-
quent performances, recordings, publication, and other general pur-
poses. Singers collaborate in the compositional process, grapple with
the meaning of notation, and add their own idiosyncratic styles to a
work. Recordings are now a much more important part of the picture;
when performed or supervised by the composer, they can define a
work. Recordings can also differ from both the published score and
286
Working with Living Composers 287
live performances. They often present several interpretations of a particular

piece or show the development of a composer s thought on a work years after it
was composed.
Over the course of the twentieth century, especially from the second half of
the century to the present, the scope of serious vocal concert music has ex-
panded to include a wider variety of compositional and singing styles than ever
before. "New music" (and some of what is still considered contemporary is
hardly new anymore) can range from one end of the vocal spectrum to the
other. Some works may feature an untrained or pop singing style, while tradi-
tional, tonal art songs require classically trained vocalism and other songs call
for a jazzy or cabaret approach. More experimental works often include chal-
lenging rhythms, difficult pitches, microtones, nontraditional use of text or
phonemes, electronics or amplification, as well as extended vocal techniques
including a variety of spoken vocalization, whispering, shouting, laughing,
humming, trilling, singing with a straight tone or a breathy tone, and so on.
Singers are also sometimes asked to improvise, to play percussion instruments
or to use props and movement.
This chapter does not delve into the vast realm of innovative vocal tech-
niques. For that, I recommend Sharon Mabrys Exploring Twentieth-Century
Vocal Music? It also will not attempt to cite the ever-growing list of new reper-
toire for the voice. (For that, see the books by Jane Manning and Patricia Lust
under "Chapter 9: Repertoire" in "For Further Reading.") Instead, the chapter
examines how composers and singers jointly develop new ideas and styles, how
performers work to realize composers' intentions, and how singers can con-
tribute their gifts and experience to the final outcome. The stories are the same
as those discussed throughout this book, but now we are the singers in the sto-
ries. When you work with a composer on a new work, you are participating in
the creation of a new performance-practice tradition that may in turn give you
invaluable insights into performance-practice traditions of the past.
Voice Types
Contemporary composers write vocal music both for general voice types and
for specific singers with unusual abilities. In some situations a particular voice
inspires a composition. In others a number of factors may come into play, in-
cluding the politics of commissions, performers' availability, and venue possi-
bilities. Luciano Berio wrote much of his vocal music for his wife, Cathy
Berberian, who was well known for the wide variety of dramatic effects and spe-
cial sounds in her vocal palette. She inspired several other composers, including
288 Working with Living Composers
John Cage, to write music for her, and she also composed for herself. Steven
Mackey wrote his chamber opera Ravenshead for Rinde Eckert, who possesses a
huge range and unique sound, as well as a large assortment of uninhibited spe-
cial effects. In some cases it may be difficult or unlikely to find another singer
who could perform such works in the future. In other cases, new singers will
rise to the challenge.
Many composers have written music for soprano Phyllis Bryn-Julson. She
speaks of her special connection to David Del Tredici, whom she met when she
was a student atTanglewood in the mid-1960s. His numerous Alice works were
conceived with her voice and personality in mind. "He blames me and I blame
him," she says of the wild and very high music of the complete An Alice Sym-
phony, which she performed a number of times and recorded in 1991. But it was
Barbara Hendricks who premiered and recorded his Final Alice with the Chi-
cago Symphony in 1975. Benita Valente also had a working relationship with
Del Tredici, whom she met at Marlboro in the mid-'6os. She recorded a num-
ber of his works and performed the premiere of "All in the Golden Afternoon"
(ChildAlice, part 2, movement 3) with the Philadelphia Orchestra in 1981.
Sometimes composers write with opera singers in mind, but other times they
want another kind of singer altogether. In his La Pasión según San Marcos, Os-
valdo Golijov requires classically trained concert singers for some of the solo
parts and Latin American pop or jazz singers for others. He composed the work
for a nonprofessional choir, the Schola Cantorum de Caracas, and requests that
they take part in any performances. The choir understands and vocalizes the
Latin American sound and style that is central to the work. "Hopefully," Goli-
jov told critic Paul Griffiths, "in ю or 20 years, singers will think it s as essential
to sing like this as it is to sing in the Italian or German style, because this is a
world of emotion that is as big as those other European traditions."4
Some composers have a folksinger sound in mind. When Cheryl Bensman
auditioned for Steve Reich after training as a classical lieder singer, he asked
whether she could sound like Joni Mitchell. "I can do that,]' she said, and she
was hired. Reich wrote parts with scat syllables for a jazzy kind of sound. He
also wrote very high parts for a particular singer he knew could easily sing soft
and straight-toned high Cs. In his notes for the recording of Tehillim, he writes
that "the non-vibrato, non-operatic vocal production will also remind listeners
of Western music prior to 1750."5 Other composers such as Arvo Part write vo-
cal music with a Renaissance choral sound in mind. I had á fascinating time
working on Princeton faculty member J. K. Randall's "Clebphila," an a cap-
pella setting of a Christina Rossetti poem. On the page the music looked
straightforward and modest in range. But by working closely with me, Jim ex-
plained the sound and approach to articulation he wanted, which was not really
something he could nótate. This newly created style felt like a unique combi-
nation of singing an American folk song and an early seventeenth-century lute
song.
Some composers write for a cabaret or musical theater singer. William Bol-
com has written many volumes of cabaret songs, both to perform with his wife,
Joan Morris, and for use by the general singer population. I recently .talked with
composer Jon Magnussen about his latest vocal project, which was inspired by
a play and his collaboration with a librettist. When asked what kind of singers
he had in mind, he said he wanted natural voices that could convey the words
easily: "probably a musical theater sound, not beefed-up or overwrought like
opera singers."6 George Crumb, now in his seventies, has started writing music
for his daughter, Ann, who is an experienced Broadway singer and i television
actress. Yet these same composers have also written music for classically trained
voices: Golijov for Dawn Upshaw, Bolcom for Benita Valen te, Crumb for Jan
DeGaetani, and many others.
The most satisfying new-music experience is when a composer writes a piece
that fits your voice, as Mozart put it, like a custom-tailored suit. A few years ago
Philadelphia's Network for New Music commissioned David Rakowski to
write a piece for small chamber group and soprano. Not knowing that David
and I had been students together at Princeton, they had asked me to sing the
premiere. I hadn't seen David for many years, and we had a happy reunion at
the first rehearsal. He told me that he had known I was to sing The Gardener
when he wrote it. Even though David hadn't heard me sing in a long time, and
I now had many more technical tricks up my sleeve, his work sat in the most
comfortable part of my voice and fit my musical personality perfectly. It was a
complete delight to sing.
Sometimes it takes a little work to arrive at a perfect fit. When Bolcom wrote
Let Evening Come for Valente, she asked him to make some changes iso that cli-
mactic phrases would lie in the most beautiful parts of her voice. "Dead Calm"
by Earl Kim had already been conceived and composed when Valente first sang
it in 1966. She liked it so much, however, that she asked the composer to write
more for her voice. He complied by expanding the work into Exercises en Route,
which was fitted to her strengths. Kim and Valente worked closely together to-
ward the 1971 premiere, and she also asked him to be involved, either through
coaching or conducting, whenever she performed the work in the following

years.
When another person sings a work written for a specific Isinger, it can bring
out different qualities in the composition. Aaron Jay Kernis heard me sing
songs he had composed for one particular soprano, which;had also been per-
formed by a number of other singers. He commented on !what our different
voices had brought to his music: one singer had more powerful high notes but
less voice in the lower middle, another had a stronger middle but less comfort-
able top, I had a good middle and easy high notes but a lighter quality overall.
A new singer can also influence a composer, making a work more accessible.
Paul Sperry heard a tape of The Seven Deadly Sins, a song cycle by Robert
Beaser, and wanted to sing it. The cycle had originally been written for and pre-
miered by a baritone; Paul believed that the cycle as a whole sat too high in
places for most baritones yet was also too low for most tenors. He approached
the composer and suggested that more singers would be able to sing his work if
he offered some alternatives to the extremes of range. In response to Paul's sug-
gestions, Beaser prepared a score for publication with a piano part and two
slightly different vocal lines, one for tenor and one for baritone. Paul went on
to record the cycle and perform it often.
Manuscripts, Revisions, and Published Scores
When you work on new music, you can come in contact with a variety of
scores: handwritten manuscripts, computer-generated manuscripts, and pub-
lished, printed scores. These come quite literally in all shapes and sizes. Hand-
written manuscripts can be difficult to read. They may include mistakes and
corrections, irregularly spaced rhythms, or imprecise vertical alignment, not to
mention messy or illegible penmanship. Yet they can also transmit an indefin-
able aspect of the composers personality as well as his or her feeling for the mu-
sic. Phyllis Bryn-Julson much prefers handwritten scores, especially when the
composer's penmanship is neat and clear like Boulez's. The autograph score for
his Pli selon pli, she says, looks like lace, and the texture of the music also re-
sembles lace. The work must be performed with the delicacy and transparency
that is reflected in the appearance of the handwritten score. Some autograph
scores can be visual works of art. I recently performed some settings for voice
and vibraphone of ninth- and tenth-century Japanese poetry about the moon.
The composer, Brooke Joyce, carefully hand-copied the music onto special
translucent vellum, set over small pieces of dark blue paper with a hand-tied
binding. The title and notes were written in silver ink, and the singer was in-
structed not to use a music stand in performance, but rather to hold the small
pamphlet "as if reading from a book of poetry." The score was as beautiful as the
songs.
Computer-generated scores are certainly the most practical to create and
edit, though they lack somewhat in charm or personality. Composers can make
parts from a score, add changes or corrections, reformat the layout, and print as
many versions as necessary. One wonders what musicologists of the future
might make of the remains of different computer-based versions of a work in
progress.
Published scores may present music in a typeset format or simply reproduce
a computer-generated score or a handwritten manuscript. Some published
scores of Crumb or Cage are as interesting visually as they are aurally. Published
scores are not necessarily the last word on a work, however; they might contain
mistakes. A composer may also change various details post-publication. When
rehearsing Aaron Jay Kerniss Songs of Innocents, the pianist and I felt that one of
the songs should move more quickly than the indicated metronome marking.
Perhaps it was for dramatic reasons or for vocal comfort, but the song just
seemed to flow better faster. In the end, not having a chance to coach the songs
with the composer, we decided to try to make the song work at his indicated
slower tempo. After the performance Kernis told us that we could move that
song a bit more than the written metronome marking if we wanted to, con-
firming our gut reaction.
Sometimes the published score is very different from what was performed at
the premiere. A few years ago I was asked to sing John Corigliano's song cycle
Mr. Tambourine Man: Seven Poems of Bob Dylan at a concert in his honor in
Philadelphia. I found the thirty-five-minute work challenging, not because of
difficult pitches or rhythms but for reasons of dramatic and vocal pacing. The
three songs before the postlude were angry and emotionally intense, with lots
of loud, sustained high notes. The final song began as a floating unaccompa-
nied hymn tune with large leaps from very low notes to luminously high
ones—sheer terror for me. I wasn't able to coach the songs with the composer
before the concert, but he sent me a tape of Hila Plitman, a colleague of mine,
with whom he had worked extensively. He said she sang the songs exactly as he
wanted them. At the performance, after the sixth song, my heart pounding in
my chest and my head reeling from sustained triple fortes, I did the best I could
to calm down for the last song. Mr. Corigliano, whom I had not yet met, sat in
the front row, listening with his eyes closed. Composers are usually warm and
complimentary after you perform their music, and Mr. Corigliano seemed
292 Working with Living Composers ί
pleased with the performance. He told me that he had not really enjoyed a live
performance of the cycle before. Surprised, I asked him about Sylvia McNair's
premiere in Carnegie Hall. He said that she had asked him to significantly
shorten the fifth and sixth songs and to recompose some of the large leaps in the
final postlude. Astonished, I asked him about Hila Plitmans tape. He re-
sponded that she had not yet performed the songs live and had made the
recording in a studio!
When a composer is involved in the performance of his or her own work, the
score may be intelligible only to the performers. Cheryl Bensman said she sang
Reich's hour-long Music for 18 Musicians from a one-page cue sheet. All the per-
formers knew what the repetitions were and how to give visual cues to move on
to the next section, but the work could only be performed by that ensemble. As
Reich explains, "I wrote the piece in a kind of musical shorthand directly into
my music notebook. From there I wrote out individual partp for the musicians
in my ensemble without making a full score. The parts themselves were also
in shorthand that was only fully intelligible with considerable oral explana-
tion—spoken, then finally jotted down, i.e. 'look at Russ,' ptc. The result was
that from 1976 until 1997 Music for 18 Musicians was, with a few unusual excep-
tions, only performed by my own ensemble."7 In 1995 a graduate student at
Cornell wanted to write a doctoral dissertation on the work and ended up cre-
ating a full score and new set of parts. These were ultimately published, with
Steve Reich's supervision, by Boosey & Hawkes. The first score he created for
general use was in 1981 for Tehillim. This work was intended for his group of
four singers and chamber ensemble, but he also created an ¡orchestral version,
and therefore the performing materials had to be made accessible to a larger
group of musicians.
Meredith Monk is another composer (as well as singer and choreographer)
who creates pieces directly on performers. She rarely writes any of her work
down, instead developing and teaching the music and the movement as part of
the rehearsal process. If a group other than her own wants to perform one of her
existing works, she collaborates with them to teach them ι the piece and the
style. When she was commissioned by Michael Tilson Thomas to create Possi-
ble Sky for the New World Symphony in 2003, she grappled for the first time
with the task of notating her work. She is currently working with Boosey &
Hawkes to prepare her other works in written form so that future generations
of performers will be able to re-create them.8
Notation
In the past half century musical notation has undergone a revolution. Experi-
mentation in the 19505 and '6os resulted in two innovative and divergent paths:
one attempted to nótate every element of performance with more precision
and control, and the other rejected precision by seeking more freedom in im-
provisation and indeterminacy. Composers developed new systems and sym-
bols for notating tempo, pitch, duration, intensity, articulation, and color, us-
ing them alone or in combination with traditional notation.
Unfortunately, these new systems and symbols are not used consistently
from composer to composer. Sharon Mabry devotes considerable space in her
book to a discussion of new kinds of notation. She also recommends a number
of specialized books on notation, including Howard Risatti s New Music Vocab-
ulary, Kurt Stone s Music Notation in the Twentieth Century, David Cope s New
Music Notation, and Erhard Karkoschkas Notation in New Music (see "Chapter
9: Notation" in "For Further Reading"). The lists of symbols and explanations
in these books are extensive and complex. In some ways they are reminiscent of
the seventeenth- and eighteenth-century treatises on ornamentation discussed
in chapters ι and 2.
Many composers, including Boulez, György Kurtag, Ligeti, and Crumb, try
to incorporate as much information as possible in their scores. Some even in-
clude pages of instructions. In his scores to Aventures and Nouvelles aventures,
Ligeti writes descriptive words over almost every vocal or instrumental gesture,
as well as paragraphs of explanation in the margins. The Peters study score
comes with two supplements: one a translation of all the handwritten German
annotations in the score, and the other an explanation of all the special nota-
tional symbols. Deciphering such instructions can be as challenging as singing
the works. But, as Bryn-Julson cautions, even if you know what the instruc-
tions mean, can you do them? Or if you can do them all, is that really what the
composer wants?
In preparing Milton Babbitt s Phonemena, which was written for Bethany
Beardslee, I worked diligently to execute the serial dynamics indicated over al-
most every note. When I coached the work with the composer, however, he
told me to use the dynamics to make shapes with the lines instead of isolated
events. He may not have given this comment to another singer, but it was obvi-
ously a suggestion I needed at that point in my work. In any case, I was pleas-
antly surprised that I could use a little more flexibility in my approach.
Conversely, some composers don't put enough information into their scores.
Then you have to discover if they really want the music plain and uninflected,
or if they expect you to add your own coloring, articulation, and dynamics. If
you work with the composers, you can ask them what they want. If they are not
available, then you must decide on your own, in collaboration with your fellow
musicians.
Even a familiar notation like a grace note can take on a completely unique
meaning and require a special type of execution, depending upon the composer
and context. For example, the atmosphere and text by Garcia Lorca in Crumb's
Ancient Voices of Children invite a certain execution of the notated grace notes.
The Chinese poetry of the Song Dynasty and the music of Chen Yi s As in a
Dream inspire a completely different approach. Barry Shiflfman, the second vi-
olinist in the St. Lawrence String Quartet, describes how Osvaldo Golijov
demonstrated the grace notes in his Yiddishbbuk:
If you played literally what Ozzie had on the page, it sounded pretty empty. How
were we to know that his strange little crushed grace notes that preceded a difficult,
dissonant and ugly chord were echoes of the vocal inflections ofjscreams? We tried to
make his music sound "good" before meeting him, and we got more and more frus-
trated. When we met Ozzie, he listened and heard our frustration. He began to de-
scribe what he was after. When we asked him to sing it, well, everything changed. He
was not "singing." It was more like crying, screaming, praying, all mixed up together.
We got it.9
Tañía León wrote many grace notes into the vocal part of Singin Sepia, com-
posed for the New York new-music group Continuum. Some were within a
half or whole step of the following main note, but some were over an octave
away. Some were to be sung on vowels, some on consonants, and others
hummed. She indicated in the score that the vocal style she wanted should have
elements of jazz and gospel, while imitating the inflections ¡of southern black
speech. Yet the tessitura was consistently above the staff. When I first started
learning the work I was unsure how to combine all those elements. When I fi-
nally met Tania and she demonstrated what she wanted, it sounded natural and
colorful, evoking exactly the elements she described. I still had a difficult time
transferring those gestures to my voice, but hearing her do it made all the diff-
erence.
Special Vocal Effects

Special vocal effects, also known as extended techniques, are often seen in new
vocal music. Jane Manning regards them as "a variety of everyday sounds,
which would all be familiar in a different context."10 Yet they are even more
problematic to nótate than traditional singing gestures. Speaking or recitation
can be unpitched or intoned, in rhythm or freely declaimed. Composers can
use a number of now-standard notations to describe spoken rhythms and con-
tours, but it is still impossible to nótate all the subtleties of speech. When
speaking or whispering, singers become actors with an infinite variety of nu-
ances at their disposal. We are also asked to shout, whoop, whistle, and laugh—
sometimes on pitch and in rhythm. We are asked to use our lips and tongues to
make trilling or percussive sounds, or to distort a sung sound. We can also be
asked to modify sung sounds by placing a hand over the mouth, or opening and
closing our mouths to different degrees. Growling, grunting, and heavy breath-
ing can be required in a variety of situations as well.
It is best when the composer can be there to demonstrate these effects or to
help you experiment until you arrive at the sound he or she wants. Bryn-Julson
says that there was nothing she had to do in Aventures that Ligeti couldn't
demonstrate himself. Continuum performs a number of fanciful theater pieces
by Francis Schwartz that require unusual and uninhibited effects, including
talking, improvisation, and expressive breathing. The directors of the group
have worked with Schwartz over many years and can explain the techniques to
new players and singers. It is a special treat, however, when Schwartz himself is
there to coach an existing piece or to unveil a new creation along the same lines.
If the composer is not available for the rehearsals of a new work, you have to
be creative and make educated guesses. If you are working on an existing piece,
you may be able to get advice from someone who has sung or played it with the
composer's supervision. If a recording exists by the players or singers for whom
it was written, it may well offer answers to questions about special vocal effects.
It can be liberating to be adventurous and daring with vocalization. How-
ever, if a composer asks you to do something that feels uncomfortable or that
you think might be harmful vocally, you must be polite but honest about your
reservations. In choosing repertoire with extended techniques, you must make
sensible decisions in the interest of vocal health. Bryn-Julson was once asked to
crack her voice in the middle of a swoop from high to low range; of the wide va-
riety of vocal effects she has performed and recorded, that was one thing she
was not comfortable doing. Sometimes composers ask us to re-create vocaliza-
tions heard in native or indigenous music from exotic cultures. Some singers
will have less trouble using their instruments and imaginations to achieve these
effects. Remember that many seemingly outrageous and bizarre vocal gestures
were written for singers who could do them easily.
Other Special Instructions or Requirements
Some new works include instructions for special stage setups or movement and
choreography. Francis Schwartz's scores often indicate hand gestures, foot
stomps, and facial expressions to be performed by both singers and instrumen-
talists. Sometimes the instrumentalists are required to make vocal sounds. (Per-
cussionists are usually game, but other instrumentalists often feel uncomfort-
able doing this.) Sometimes singers must use props, as in Crumbs Ancient
Voices of Children, where the singer is instructed to speak through a cardboard
tube. Performers may be placed offstage or at different locations in a hall. Many
new works call for special instruments such as a prepared piano or exotic per-
cussion equipment, or they require unusual uses of standard instruments. You
may be asked to sing into the piano to create sympathetic vibrations or to play
a small percussion instrument. It is advisable to learn whatever techniques are
required for playing that instrument and to practice it as you would the
singing. I have always wanted to play the triangle, but when Cynthia Folio
wrote a work for me in which I had to sing complex rhythms and play the tri-
angle at the same time, I had to practice more than doubly hard to master the
coordination. It was a challenge, but ultimately the experience was rewarding.
Some new vocal works use amplification or electronics. Others use prepared
tape with synthesized or computer-generated sounds. Babbitts Philomel, for
soprano and synthesized tape, was written for Bethany Beardslee and includes
her voice on the tape. Anyone who performs the work now and rents the tape
from the publisher sings with Bethany's voice. When I was still a student, I was
asked to record some bits of singing and talking that were then processed and
combined with special effects to produce a seven-part chorus for Martin But-
ler s opera The Sirens' Song. The premiere took place in an outdoor amphithe-
ater on the island of Crete. Unfortunately I was not there, but my voice was.
In some works with electronics, the special effects can be created during the
performance. My cousin, a cellist, once played a new piece originally written as
a cello solo. When the composer asked her to perform it again ten years later, he
decided to add a recording engineer who manipulated her sound electronically
as she played. This process created an entirely new work that could be different
each time it was performed. In some pieces with prepared tape, the electronics
are not meant to be synchronized with the live performer. In those works where
the voice and tape are meant to be coordinated, the score indicates cues that are
more or less accurate, depending on the nature of the taped sounds.
Working with a tape can be comforting because it is always exactly the same,
whether in practice or performance. Babbitt uses the synthesizer to create elec-

tronic works because it enables him to retain precise control over the flow of
events, "particularly temporal events. As I don't have to tell you, much of the
problem of having unusual, intricate rhythms performed has nothing to do
with the incapacity of the performer or the incapacity of the ear. Its the damn
notation."11 His Phonemena, which exists with both a synthesized tape and a
notated piano accompaniment, was fascinating for me to perform in both ver-
sions. They felt in many ways like completely different works, both from the
ways they sounded and the ways the rhythms worked when collaborating with
the pianist or the tape.
It is undeniably exciting to participate in creating new and unusual effects or
a tape or video portion of a work. As with special vocal effects, however, you
must be honest with the composer about what you are able and willing to do.
When I was a still relatively inexperienced performer, I was rehearsing a new
work in which the composer asked me to walk all over the stage, face different
directions, and even lie on the floor face down. I was still struggling with the
rhythms and pitches and really needed to keep my eyes glued on the conductor.
I told the composer I was sorry, but I wouldn't be able to manage the choreog-
raphy. (Hopefully he was able to have the piece performed later with a singer
who could include all the movement.)
Learning Difficult Rhythms and Pitches
The first task in approaching any new piece should be to learn the correct
rhythms and pitches. Singers can have a bad reputation for not taking notation
seriously, especially when it comes to rhythm. In music from past periods and
styles, the rhythm can be subtly flexible to support and highlight the text. This
can also be the case in new music written in a traditional, tonal style. But the
more complex and abstract the music, the more exacting and precise the
rhythm must be. Bryn-Julson has never heard a composer tell her not to worry
about the rhythm; she remembers Boulez saying, "I took the time to write it,
you should sing it that way." If Crumb writes a series of thirteen repeated six-
teenth notes accelerating over three seconds, or a two-note trill repeated three
times (for a total of six notes), she believes one must do exactly that number of
repetitions; otherwise it doesn't sound right. When the vocal line forms a deli-
cate counterpoint to a piano accompaniment or instrumental lines, I have been
grateful to my instrumental colleagues for insisting that we figure out the align-
ments exactly. Often this extra effort makes the difference between a piece's
sounding muddy and vague or magically transparent and clear.
Mastering complex rhythms is like doing a crossword puzzle or a math prob-

lem. Often if you find the smallest common note value (the lowest common
denominator), you can line up each tiny pulse and put your notes exactly where
they belong. Then you can gradually increase the speed until you reach the cor-
rect tempo. Arthur Weisberg's Performing Twentieth-Century Music includes a
detailed discussion about figuring out complex rhythms. My ear-training
teacher at Juilliard made us perform difficult rhythms by conducting and
speaking on "ta." This practice has served me well over the years. I remember
learning Phonemena during a Christmas season when my daughter was in-
volved in a production of The Nutcracker. During the long hours of rehearsal
that parents had to stay in the theater, I sat in the lobby using the swing of
whichever Tchaikovsky dance happened to be playing in the background as my
sixteenth-note or eighth-note pulse. I then conducted the measures and spoke
the complex rhythms until I knew where all the groups of five or seven went.
(The other parents couldn't imagine what I was doing.) If you practice enough,
tough rhythms will start to make sense. Cheryl Bensman says that the Reich en-
semble was lucky to be able to rehearse many hours until the group rhythm
"locked in." She also reports that sometimes she would rebar the rhythms for
herself so the stresses made sense to her. This can be helpful in a lot of music, as
long as you don t rewrite the rhythm itself.
Learning difficult pitches in nontonal music can be a real challenge for
singers without perfect pitch. But plenty of well-known new-music singers
(Jan DeGaetani, Bethany Beardslee, Lucy Shelton, Benita Váleme) have
achieved great things with hard work in place of perfect pitch. I spend a lot of
time working with intervals. Sometimes it helps to transpose large leaps to
smaller intervals within an octave, enabling you to really get the pitches in your
ear. For example, a major seventh would become a half step down, or a minor
tenth would be a minor third. If the spelling of the pitches is confusing, I often
write the generic interval above the notes. I also check my pitch against pitches
in the accompaniment to make sure I'm in the right place. I always try to un-
derstand my note in some harmonic or dissonant context (see the discussion re-
garding learning twelve-tone music at the end of chapter 7). One of my voice
teachers used to say that singers need "infinite patience, infinite faith," and this
is certainly true when learning tough pitches. With enough gentle and patient
practice, you will find that eventually the pitches start to make sense in your
head and feel comfortable in your throat.
Singing microtones is a special challenge for the brain and ear, involving an
additional level of calculation to the procedure: sing a major third up plus a
quarter tone down, or sing a perfect fourth a little bit sharp. Even Bryn-Julson,
who with her perfect pitch impressed Del Tredici by sight-reading his I Hear an
Army, has been stumped by some difficult micro tonal music with large leaps.
She asked Boulez if she could trade parts with the flute player in one particu-
larly challenging piece; he agreed, but the flute part proved to be even harder. If
Bryn-Julson couldn't sing the part, it would be difficult to find another singer
who could. Yet composers continue to write very difficult pitches and rhythms,
expecting or hoping to find singers who can manage. Throughout the centuries
we have seen composers challenging singers to extend their boundaries of skill
and musicianship. Singers today must continue to rise to the challenge and
learn to execute difficult pitches and rhythms accurately. We have an ¡advantage
with todays technology—it is possible to generate a computer realization of a
new work as an aid in learning a difficult part.
Vibrato
This book has traced the changing attitudes toward the use of vibrato in vocal
music. By the middle of the twentieth century, most singers' vibratos were big-
ger and more consistently employed than ever before. Today many composers
ask singers to remove their vibrato as a color contrast in an isolated moment, or
even for an entire work. Bensman says she often thought of a child's voice in-
stead of a woman's to achieve the straight-toned sound required in Reich's mu-
sic. She always sang softly to keep her throat from getting tight. This was easy
for her to do, since Reich's singers were amplified. Bryn-Julson recommends
not singing loud or high for straight tones and keeping the breath and throat re-
laxed. She says she always adds some vibrato to notes above the staff. Some-
times, however, composers specifically ask singers to sing straight tones high
and/or loud; whether this can work depends on the ability and corrifort of the
singer. If you are rehearsing with a composer, you may be able to negotiate such
issues. A wonderful interaction for me came when I was working with Brooke
Joyce on a new work and asked him whether he wanted a particular note or sec-
tion sung straight-toned or not. He said, "What do you think?"
Vibrato can be varied and straight tones added in places not specifically
marked in a score. You might use a straight tone to create an exotic or native
style of singing, or to make the intonation of micro tones clear. You can also ad-
just vibrato to match or blend with different accompanying instruments. I use
different kinds of vibrato to color harmonies and inflect dissonances. Bryn-Jul-
son says she varies her vibrato to inflect the text and suit the dramatic situation.
Paul Sperry lets the text help him make vibrato choices to create anything from
a cold, stark sound to a jazzy or cabaret style. Your choice of vibrato should
complement the harmony, instrumental textures, and the text in music of all
periods and styles.
Text and Musk
Composers today treat the relationship between text and music in a wide vari-
ety of ways. In some traditional approaches, the meaning ¡of the text can be
clearly and directly related to the musical setting. In others it can seem com-
pletely independent or irrelevant. A composer can set an expressive text in an
abstract way and instruct the singer not to add emotion or inflection to the po-
etry. Composers can use the voice instrumentally, with vowels and consonants
merely providing texture. I find that when I am asked to sing without expres-
sion, I have to be careful not to sing without energy and coKimitment as well.
The composer may want the text to be neutral, but I don't want my perfor-
mance to be lifeless. This can be a tricky balance to achieve.
Composers can also use a nonsense text, phonetic symbols, or phonemes to
create singable sounds. These can be used in an abstract and unexpressive con-
text, or they can take on an inner meaning all their own. Bryn-Julson says that
Ligeti's Aventures contains the largest number of phonetic symbols she has ever
seen. Its list of over 100 requested sounds creates a nonsense language that,
when combined with the music, communicates a compelling drama. The com-
bination of phonemes and music in Babbitts Phonemena suggests a vivid sub-
text to me. This was most likely not part of Babbitt s conception of the work,
but it helps me shape the phrases, as he suggested, and makes the performance
livelier and more interesting.
Many contemporary composers set text in a traditional way with appropri-
ate prosody and rhythm. For Tehillim> Reich modeled the vocal lines on the
natural spoken inflections of the Hebrew text. According to Bensman, Reich
preferred a non legato, speechlike approach to the singing, reminiscent of
Monteverdi. Combined with the music's intricate counterpoint and the re-
quired nonvibrato production, this creates a unique sound world that is at once
both abstract and expressive. Benita Valente reports that all the vocal rhythms
in Exercises en Route жъ derived from the ways Earl Kim spoke the text. His set-
ting combines spoken and sung declamation, with the instrumental parts fol-
lowing and complementing the singers flexible approach to declaiming the
text.
Some composers expect singers to communicate the emotional and dra-
matic meaning of a text by using the full range of colors available to us. The
subject matter of texts chosen by contemporary composers ranges much farther

afield than nineteenth-century poems about love and nature. Singers today
thus must be able to express a wide variety of poetic and dramatic situations.
Sperry says that he doesn't think about colors specifically, instead concentrating
on the poem and the music. He first tries to find a speaking voice that would
work for the text, whether it is serious, humorous, or theatrical: "Out of my
speaking voice and what I hear in the music comes what the singing voice
ought to be for the song." From a wealth of possibilities he picks what seems ap-
propriate and hopes that it works. Daron Hagen heard Sperry sing a song in
which he used a white, hollow tone to create a particular dramatic atmosphere.
The composer then wrote a work for him, choosing a text that would require
that very color. Some composers try to control and nótate every aspect of the
expression of the text, using articulation marks, dynamics, and color sugges-
tions. This can feel restrictive, but it can also assure us that we really know what
the composer wants.
If you are involved in the compositional process, you may be able to have in-
put regarding the text setting's prosody, vocal comfort, and expression. Some
composers are happy to make changes to suit the singer. Others are less ap-
proachable or less flexible. Peter Westergaard, a Princeton professor with whom
I have often worked, offers a glimpse of a composer's approach to text and mu-
sic. As he finished work on a concert setting of scenes from Melville's Moby
Dick, he sent the score to a singer he thought might be good for the role of
Ahab. Baritone William Parcher had sung the role of Prospero in Westergaard's
operatic setting of Shakespeare's The Tempest. Peter liked working with him and
admired his singing. Yet the role of Ahab was not conceived for Bill or for any
particular singer, but rather for a standard high dramatic baritone. When
Parcher looked at the score, he considered the role both vocally and dramati-
cally, and then phoned Peter with some concerns about how the high tessitura
might affect his characterization of Ahab. Peter responded, trying to combine
the needs of the singer with the requirements of his compositional structure:
31 October 2003
Dear Bill:
I can't tell you how exciting it was for me to get your read on Ahab. To say I learned
a lot is the understatement of the week. It was particularly important for me to un-
derstand that it was not a question of whether you could or could not sing the high-
lying lines, but what staying that high that much and that early on would mean for
our perception of Ahab's mental state. I wholeheartedly agree that the last thing we'd
want would be an Ahab verging on hysteria from the moment we meet him. Flashes
of something the crew might find spellbinding and we in retrospect might under-
stand as madness, yes, but he must be resolute, articulate, not frantic.
Unfortunately, the way the piece is put together does not allow easy fixes at the lo-
cal level. One reason for this is the way the vocal lines, at least Ahabs and Ishmaels,
are conceived of as central to the long-range pitch trajectories. (I really hate the kind
of opera where the orchestra plays a symphony and the singers do a kind of recitative,
using whatever pitches are flowing by. I want the vocal line to be at the core of what
we hear.) What this usually comes down to in MobyDickis that Ahab's and Ishmaels
lines create their own four-part harmony which is supported by the instruments.
The top line ofthat four-part texture is of course audibly the most obvious, but its
motion is supported by the others. Furthermore, the others are often supported by
the instruments. In short, this makes it difficult to make changes to a couple of mea-
sures of the vocal line without changing the surrounding measures as well, and it is
usually impossible to change the vocal line without also changing the instruments.
So what I am sending you here represents a set of minimal fixes, changes that I can
make without upsetting the whole apple cart. The question is, are they sufficient—
I'm sure from what you've said they'll make things better, but ¡are they sufficient to
make Ahab the character I think we both envision?12
Westergaard's music includes challenging pitches and rhythms, but it always

reflects the way he wants the line of text read. Large leaps magnify the contours
of spoken inflection, and complex rhythms attempt to nótate the natural sub-
tle patterns of speech. The changes he offered, which Bill Parcher gladly ac-
cepted, included moving pitches and resetting rhythms to emphasize different
words and syllables. This kind of close collaboration is truly rewarding.
Composers as Performers
Composers have been performing their own vocal works throughout music
history. They have sung, accompanied from the keyboard, or directed from the
podium. This still happens today. Steve Reich leads rehearsals of his group from
a percussion instrument or the piano in works that don't require a conductor
and usually participates in performances, either playing percussion or adjusting
the electronics. Pierre Boulez remains one of the great conductors of our time
in his own music or any repertoire. Leonard Bernstein was the twentieth cen-
tury's answer to Brahms or Liszt, achieving multiple renown as a composer,
conductor, and pianist; Paul Sperry remembers Bernstein's overwhelming gen-
erosity of spirit as he conducted his Dybbuk Suite No. ι or coached singers for
his Songfest. Many other contemporary composers who do not have full-time
careers as professional conductors conduct their own works.
Composers who are pianists often accompany their songs. Louise Talma, a
prodigious pianist, asked Sperry to perform many of her songs (though they
were not written specifically for him). Sperry recalls that she played her music
exactly the way she wanted it to go, performing her prescribed metronome
markings perfectly. George Walker, an accomplished pianist, gave many more
comments to the pianist than to me when he coached us for a performance of
his songs.
When a composer rehearses and performs music with you, the communica-
tion becomes immediate and intimate. You can achieve exactly the right kind
of articulation or phrasing, precisely the right tempo. Of course the experience
can also be problematic if what the composer wants in performance is contrary
to what you want or need to do in the work, or what may seem to be indicated
in the score. Delicate negotiations must be worked out in any kind of chamber
music or collaborative situation, whether between singer and accompanist,
singer and conductor, among instrumental colleagues, or between the com-
poser and the other players.
Composers who are also singers can write for their own instruments and sing
the music exactly the way they want. Joan La Barbara has composed works for
her own unique vocabulary of extended techniques. She has also premiered
works written for her by many experimental composers, including Reich,
Cage, and Morton Feldman. Some of her own works are meant to be per-
formed live, usually with the aid of amplification or electronics. Others were
created in the studio, with multiple vocal tracks and various kinds of electronic
and digital manipulation, for radio or recordings. Meredith Monk started by
experimenting with her own solo voice and then added other voices to her
compositions:
What happened to me was that I started working alone, and I didn't really have any
references except that I wanted to make my voice as flexible as my hand could be.
And I wanted to find a vocabulary that was built on my own instrument. . . . There
are universal vocal sounds that you just come upon as you're working on expanding
your own vocal palette. At the same time, each person's vocal apparatus is totally
unique, so there are certain things, for example, that members of the ensemble do
that I can't quite figure out, and there are certain things that I can do that they can't
quite get exactly. . . . We all share a common, cross-cultural huge palette of arche-
typal sounds that people in the world that sing have in common. And that's a won-
derful combination of things, and it's kept me captivated all these years.13
Susan Botti also writes and performs music conceived for her own soprano
instrument: "I try to create a satisfying singing experience so that it's not simply
sounds but also lyricism and enjoyment of what the voice can do."14 Unlike
Monk, her music is notated and intended for other singers as well. Her style
combines elements of classical, jazz, and world-music vocalism, reflecting her
eclectic background.
Using Recordings
Recordings of new music can be an invaluable help in discovering composers'

intentions. They reveal the choices made by particular performers and can in-
spire us to imagine our voices making new sounds. They can also be mislead-
ing. A recording made by the performers for whom a work was written, with
the composer's supervision or participation, can be the definitive version ofthat
piece. If the work was written for special forces or particularly distinctive per-
formers, the recording may be the last word on the work and even its last per-
formance for many years. But a performance captured in a recording studio can
be very different from what the performers could achieve in a live concert.
Composers also often make changes for a recording. Discrepancies between the
score and a recording can result from a number of factors, including mistakes in
the score, mistakes made by performers, or changes agreed on by all involved. A
recording of a new work may also be made by performers other than those who
performed the premiere, in a different location or years later. Popular works
may be recorded numerous times by different performers, resulting in a range
of different interpretations, even with the composer's involvement. Later gen-
erations of players may be able to approach a difficult work with more techni-
cal facility than the original players on the recording. All these variables suggest
that, as with recordings of earlier music, we should use a recording only as a
starting place from which to craft our own rendition of a work.
When I was preparing for a performance of Lucas Foss s Thirteen Ways of
Looking at a Blackbird, I listened to a number of different recordings of this
now well-known cycle. Some things about the various recordings were similar,
while others were strikingly different. Some recordings were made with Foss's
involvement, others not. The work contains many innovative notational de-
vices, instructions for using unusual percussion instruments inside the piano,
and a variety of special vocal effects. Unfortunately, Foss was unable to come to
our rehearsals or performance, but when one member of our group called him
to inquire about several issues, he actually couldn't remember definitive deci-
sions he had made about those items. We seemed to have his permission to do,
within a certain range of possibilities, what would work well for us in that per-
formance situation. Ultimately, that is what musicians must do in every new re-
hearsal or performance situation. A recording can be a helpful place to start,

but we must make the piece work for the given players and circumstances.
When I was working on As in a Dream by Chen Yi, I wasn't able to [work with
the composer but I was able to hear a recording by the Chinese soprano Rao
Lan, who had inspired the composition. Listening to the recording was helpful,
but it didn't answer all my questions. The work is sung in Chinese and includes
elements of traditional Chínese opera. Two versions exist: one for violin and
cello, and the other (used for the recording) for traditional Chinese instru-
ments. I was working with a cellist and a Chinese violinist, who luckily could
help me with pronunciation and word-for-word translation of the text. But
there were many differences between what we saw in our scores for modern in-
struments and what we heard on the recording. Rao Lan s instrument and vocal
production were also quite different from mine. I had difficulties deciphering
which gestures were indicative of Chínese opera and which were her personal
idiosyncrasies, either technical or interpretive. I knew I could not just copy
what I heard but instead would have to discover how to make the piece work
for me both vocally and dramatically. Inspiration eventually came when I was
singing the work on tour with Continuum in Mongolia. Just before we were to
perform, a group of Mongolian folk musicians presented some songs with tra-
ditional instruments. The micro tonal sliding grace notes I heard from the
Mongolian woman created a sound model that suddenly made sense for my
voice, and I decided to use some of those stylistic elements in As in a, Dream. I
don't know whether the composer would have approved, but it helped me
make the piece work in that situation.
Conclusions: Finding the Composers'

Intentions
If a composer is present when we are working on his or her piece, we will make
our best effort to do what the composer wants. Yet the nature of composer-per-
former interactions can vary widely. Some composers defer easily to the needs
of performers. Others are unapproachable and unyielding in their conception
of how their music should go. Younger composers can be quite picky about
what they want from performers, yet they often gain flexibility with experience.
Sometimes—because the experience of performing and listening to a work can
be completely different—a piece may make absolutely no sense to the per-
formers who are playing it. With some new pieces I have been so busy counting
or trying to find my notes that I have initially had no real idea what the work
sounds like as a whole. Only much later, after I heard a recording of the perfor-
manee, would I realize what an interesting composition it was. This is not an

uncommon experience. In these kinds of situations, you must trust the com-
poser. You could be working with a genius and not realize itl At least if you do
what they ask and it doesn't work, they will learn.
In other situations, however, it is important to make changes in order to
achieve the composers' goals and assure that the work is successful. Some of the
most accomplished new-music players and singers I have asked agree that they
certainly adjust details in a score in order to make a piece work well. Serving the
new-music composer does not necessarily mean a slavish devotion to the details
of notation. Most composers want performers to feel comfortable performing
their music, so that it will sound good. If they are working with excellent play-
ers and singers whom they admire and trust, they will welcome suggestions that
would help transmit their ideas more effectively. If you try an adjustment that
works well and you ask the composer for a reaction, he or she will most likely be
pleased. Of course it all depends on the particular chemistry between composer
and performer. There can be a large gray area between serving the score and
serving the composer, between doing what the composer asks and making a
work effective. Performers of new music face these kinds of choices all the time
and must find workable solutions for each situation.
Singing new music is both challenging and liberating. In creating a new vo-
cal work you get to explore your own instrument and discover uncharted
realms of vocalism and expression. Working with a composer enables you to ex-
perience a fluid and dynamic relationship, out of which new vocal styles may
emerge. You also get to contribute to the continuation of vocal art by adding to
the body of literature for the voice. Not every work will be a masterpiece, but
you will help create an environment in which masterpieces can come into be-
ing. In the process you will gain a much more personal understanding of the
working relationships between singers and composers in history. You will also
gain an appreciation for all the elements that go into creating a style of singing
from any period. In the interest of serving the composers of the present, we can
bring our gifts, our experience, and our hearts to the collaborative process. The
insights we gain from these interactions can also help us as we try to serve the
composers of the past.
Notes
CHAPTER 1.THE EARLY BAROQUE
1. Bénigne de Bacilly, Remarques curieuses sur Van de bien chanter (Paris, 1668),
trans. Austin B. Caswell as A Commentary upon the Art of Proper Singing (New
York: Institute of Mediaeval Music, 1968), 16.
2. Robert Donington, Baroque Music: Style and Performance (London: Faber
Music, 1982), 12. ι
3. Giulio Caccini, Le nuove musiche (Florence, 1602), trans, and ed. H. Wiley
Hitchcock, Recent Researches in the Music of the Baroque Era 9 (Madison, Wis. :
A-R Editions, 1970), 56. ;
4. Giovanni Piccioni, in the preface to his Concern ecclesiastici ai-+8 voci con il
suo Basso seguito (Venice: Vincenti, 1610), quoted in Franck Thomas Arnold,
The Art of Accompaniment from a Thorough-Bass as Practiced in the Seventeenth
and Eighteenth Centuries (New York: Dover, 1965), 66, n. 10.)
5. Howard Mayer Brown, "Editing," The New Grove Dictionary of Music and
Musicians, ed. Stanley Sadie (London: Macmillan, 1980), vol. 5, 846-48.
6. John Wilson, ed., Roger North on Music (London: Novelle, 1959)i 26.
7. Ibid., 238.
8. Lodovico Zacconi, Prattica di música utile et necessaria si al compositore. . . si
anco alcantore (Venice, 1596), quoted in Carol MacClintock, Readings in the
History of Music in Performance (Bloomington: Indiana University Press,
i979)> 73·
307
308 Notes to Pages 14-22
9. Michael Praetorius, Syntagma musicum, quoted in MacClintock, Readings, 164.

ίο. Christoph Bernhard, Von der Singe-Kunst, oder Manier (ca.. 1649), trans, in Walter Hilse,
"The Treatises of Christoph Bernhard," Music Forum 3 (1973): 14.
11. Bacilly, Remarques, 83.
12. Wilson, Roger North on Music, 18.
13. Pier Francesco Tosi, Opinioni de canton antichi e moderni (Bologna, 1723), trans, and ed.
John Ernst Galliard as Observations on the Florid Song (London, 1742), ed. with addi-
tional notes by Michael Pilkington (London: Stainer and Bell, 1987), 8.
14. Hermann Finck, Practica música (Wittenberg, 1556), quoted in MacClintock, Readings,
6г.
15. Gioseffo Zarlino, Le istitutioni harmoniche (Venice, 1558), iii, chap. 46, 253; quoted in
Richard Wistreich, "'La voce e grata assai, ma . . .' Monteverdi oh Singing/' Early Mu-
sic 22 (1994): 9.
16. The Letters of Claudio Monteverdi, ed. Denis Stevens (Oxford: Clarendon Press, 1995),
66-67.
17. Tosi, Opinioni, 4.
18. Caccini, Le nuove musiche, 44.
19. John Dowland, Andreas Ornithoparcus His Micrologus, or Introduction: Containing the
Art of Singing (London, 1609), quoted in MacClintock, Readings, 162.
20. Claudio Monteverdi, letter of December 9,1616, quoted in MacClintock, Readings, 179;
see also Monteverdi, Letters, in.
21. Marin Mersenne, Harmonie universelle, contenant la théorie et la\pratique de la musique
(Paris, 1636), quoted in MacClintock, Readings, 173.
22. See also Bacilly, Remarques, parts 2 and 3; Sally Sanford, "Solo Singing i," in A Performer's
Guide to Seventeenth-Century Music, ed. Stewart Carter (New York: Schirmer, 1997), 5;
and Sanford, "A Comparison of French and Italian Singing in the Seventeenth Century,"
Journal of the Society for Seventeenth-Century Music ι (1995), http://sscm-jscm.press.
uiuc.edu/jscm/vi/noi/sanford.html.
24. Tosi, Opinioni, 6; see also Julianne Baird, "Solo Singing 2" in A Performer's Guide to Sev-
enteenth Century Music, ed. Stewart Carter (New York: Schirmer, 1997).
25. Rodolfo Celletti, A History of Bel Canto, trans. Frederick Fuller (Oxford: Clarendon
Press, 1991), 15, n. 2.
27. Wistreich, " 'La voce e grata,'" 7-15.
28. Giovanni Gamillo Maífei, Libri due: Discorso . . . /' cantar di \garganta senza maestro
(Naples, 1562), quoted in MacClintock, Readings, 53.
29. Ibid., 52.
30. Zacconi, Prattica di música, 70.
31. Adrian Petit Coclico, from Compendium musices, 1552, quoted in MacClintock, Read-
ings, 33.
32. Minter also recommends the following sources for examples of diminutions: Diego Or-
tiz, Tratado de glosas sobre clausulas y otros géneros de puntos en la música de violones (Rome,
1553), ed. Max Schneider (New York: Bärenreiter, 1967); Italienische Diminutionen—
Notes to Pages 22-45 309
Italian Diminutions: The Pieces with More than One Diminution from 1553 to 1638, ed.
Richard Erig (Zürich: Amadeus, 1979).
34. See H. Wiley Hitchcock, "Vocal Ornamentation in Caccini's Nuove musiche" Musical
Quarterly 56 (1970): 391, and Edward Huws Jones, The Performance of English Song,
1610-1670 (New York: Garland Press, 1989), 70.
36. Caccini, Le nuove musiche, 48-49.
37. Giovanni Battista Bovicelli, Rególe, passaggi di música (Venice, 1954), π, as quoted in
Hitchcock, "Vocal Ornamentation," 392, n. 10.
39. Ibid., 49.
40. Robert Toft, Tune ThyMusicketo Thy Hart: The Art of Eloquent Singing in England, 1597-
1622 (Toronto: University of Toronto Press, 1993), 15-56,108-26.
41. See Olive Baldwin and Thelma Wilson, "Purcells Stage Singers," and Timothy Morris,
"Voice Ranges, Voice Types, and Pitch in Concerted Works," in Performing the Music of
Henry Purcell, ed. Michael Burden (Oxford: Clarendon Press, 1966), 105-29,130-42.
42. See Andrew Parrott, "Performing Purcell," in The Purcell Companion, ed. Michael Bur-
den (London: Faber and Faber, 1995), 387-444.
43. Olive Baldwin and Thelma Wilson, "Purcells Sopranos," Musical Times 123 (1982): 602.
44. Jones, Performance of English Song, 56.
46. Baldwin and Wilson, "Purcells Stage Singers," in Burden, Performing the Music of Henry
Purcell, 124.
47. Michael Burden, e-mail message to author, January 14, 2004.
49. Baldwin and Wilson, "Purcells Stage Singers," 119.
50. Jones, Performance of English Song, 233-43.
52. Mersenne, Harmonie universelle, 173.
53. Bacilly, Remarques, 42-43, 32-33.
54. Bacilly, Remarques, 23; see also Peter Giles, The History and Technique of the Counter-
Tenor (Aldershot, U.K.: Scolar Press, 1994), 37-38.
55. Neal Zaslaw, "The Enigma of the Haute-Contre," MusicalTimes115, no. n (1974): 939-
41·
57. Ibid., 22.
58. James Anthony, French Baroque Music: From Beaujoyeubc to Rameau, rev. ed. (Portland,
Ore.: Amadeus Press, 1997), 134-39.
59. David Fuller, "Notes inegales," The New Grove Dictionary of Music and Musicians, ed.
Stanley Sadie (London: Macmillan, 1980), vol. 13, 420-27.
61. Ibid., 24.
62. Ibid., 49.
63. Ibid., 64.

64. Ibid., 95.
65. Hellmuth Christian Wolff, ed. Original Vocal Improvisations from the i6th—i8th Cen-
turies, Anthology of Music 41 (Cologne: A. Volk, 1972): 95=100.
CHAPTER 2. THE LATE BAROQUE
1. Johann Friedrich Agrícola, Anleitung zur Singkunst (Berlin, 1757), trans. Julianne Baird
as Introduction to the Art of Singing (Cambridge: Cambridge University Press, 1995), 34.
2. Nikolaus Harnoncourt, Baroque Music Today: Músicas Speech (Portland, Ore.: Amadeus
Press, 1982), 29.
3. Paul Steinitz, Performing Bach's Vocal Music (Croydon, U.K.: Addington Press, 1980), 26.
4. Carl Philipp Emanuel Bach, Versuch über die wahre Art das Clavier zu spielen (Berlin,
1753; part 2,1762), trans. W. J. Mitchell as Essay on the True Art of Playing Key board In-
struments (New York: Norton, 1949), sec. Шло, 151.
5. Leopold Mozart, Versuch einer gründlichen Violinschule (Augsburg, 1756), trans. Editha
Knocker as A Treatise on the Fundamental Principles of Violin Playing, 2nd ed. (Oxford:
Oxford University Press, 1985), sec. I.ii.y, 33. I
6. Harnoncourt, Músicas Speech, 39-40.
7. Johann Quantz, Versuch einer Anweisung die Flöte traversiere zu spielen (Berlin, 1752),
trans. Edward Reilly as On Playing the Flute, 2nd ed. (New York: Schirmer, 1985), sec.
XI.io, 122-23.
8. Agrícola, Anleitung, 102.
9. Ibid., 152-53.
ίο. John Butt, Bach Interpretation: Articulation Marks in Primary Sources of], S. Bach (Cam-
bridge: Cambridge University Press, 1990), 2-3.
11. Harnoncourt, Music as Speech, 44.
12. Quantz, Versuch, sec. XI.19,127.
13. С. P. E. Bach, Versuch, sec. V.I4, ιοί; sec. III.12,151-52.
15. Jean-Baptiste Bérard, L'art du chant (1755), trans, and commentary by Sydney Murray
(New York: Pro Música Press, 1969), 27,103-4.
16. Butt, Bach Interpretation, 15.
17. L. Mozart, Versuch, sec. V.4, 97.
18. Michel Pignolet de Montéclair, Principes de musique (Paris, 1736), trans, in Cantatas for
One and Two Voices, ed. James R. Anthony and Dirán Akmajian (Madison, Wis.: A-R
Editions, 1978), xiii.
19. L. Mozart, Versuch, sec. XI.i.3, 203.
21. Bérard, L'art du chant, 71-73.
22. Arthur Mendel, "On the Pitches in Use in Bach's Time," Musical Quarterly 41 (1955):
332-54,466-80.
24. Ibid., 173.
25. See Laurence Dreyfus, Bach's Continuo Group: Players and Practices in His Vocal Works
(Cambridge, Mass.: Harvard University Press, 1987), 106.
26. See Henry Pleasants, The Great Singers (New York: Simon and Schuster, 1966); John
Rosselli, Singers of Italian Opera (Cambridge: Cambridge University Press, 1992); C.
Steven Larue, Handel and His Singers: The Creation of the Royal Academy Operas, 1720-
1/28 (Oxford: Clarendon Press, 1995).
27. Charles Burney, A General History of Music (London, 1789), 814.
29. See Agrícola, chap. 2; Quantz, chap. VIII; C. P. E. Bach, chaps, ι and 2; and L. Mozart,
chap. IX.
30. Frederick Neumann, Ornamentation in Baroque and Post-Baroque Music, with Special
Emphasis on]. S. Bach (Princeton, N.J.: Princeton University Press, 1978), part V, 29.
31. Winton Dean, G. E Handel: Three Ornamented Arias (Oxford: Oxford University Press,
1976), 3-4. See also Hellmuth Christian Wolff, Original Vocal Improvisations from the
i6th-i8th Centuries, Anthology of Music 41 (Cologne: A. Volk, 1972), 109-17.
33. Ibid., 192-93-
34. See also Howard M. Brown, "Embellishing Eighteenth-Century Arias: On Cadenzas,"
in Opera and Vivaldi, ed. Michael Collins and Elise K. Kirk (Austin: University of Texas
Press, 1984), 268-76.
35. See Jens Peter Larsen, Handel's Messiah: Origins, Composition, Sources (Westport, Conn.:
Greenwood Press, 1990); Donald Burrows, Handel: Messiah (Cambridge: Cambridge
University Press, 1991); and Watkins Shaw, A Textual and Historical Companion to Han-
del's 'Messiah'(London: Novello, 1965).
36. James R. Oestreich, "Rejoicing in a 'Messiah' of Good Taste (No Mad Scenes, Please),"
New York Times, sec. E, December 23,1999.
37. Christoph Bernhard, Von der Singe-Kunst, oder Manier (ca.. 1649), trans, in Walter Hilse,
"The Treatises of Christoph Bernhard," Music Forum 3 (1973): 13—29.
38. See Wolff, Original Vocal Improvisations, 143-68.
40. Ibid., 97, ιοί, ΙΟ7,109, no.
41. Neumann, Ornamentation in Baroque, 145-46.
42. Ibid., 219.
43. John Butt, "Ornamentation," in The Sacred Choral Music of]. S. Bach: A Handbook
(Brewster, Mass.: Paraclete Press, 1997), 52.
44. Agrícola, Anleitung, 129,139.
45. Quantz, Versuch, ιοί.
46. See Joshua Rifkin, "Bachs Chorus: A Preliminary Report," Musical Times 123 (1982):
747-51; Robert Marshall, "Bachs Chorus: A Reply to Rifkin," Musical Times124 (1983):
19-22; Rifkin, "Bachs Chorus: A Response to Marshall," Musical Times124 (1983): 161-
62; Rifkin, "Bach's Choruses: The Record Cleared," High Fidelity (October 1982): 58-59;
Marshall, "Bach's Choruses Reconstituted," High Fidelity (October 1982): 64-66; and
Andrew Parrott, "Bach's Chorus: A 'Brief yet Highly Necessary' Reappraisal," Early Mu-
sic24 (1996): 551-80.
312 Notes to Pages 85-Ί 02
47. Andrew Parrott, The Essential Bach Choir (Woodbridge, Suffolk: Boydell Press, 2000).
48. Quantz, Versuch, 328-29.
49. David Tunley, The Eighteenth-Century French Cantata (Oxford: Clarendon Press, 1997),
viii.
50. Bérard, L'artdu chant, part II, chap. 3.
51. Montéclair, Principes, xiii.
CHAPTER 3. THE CLASSICAL ERA
1. Charles Rosen, The Classical Style: Haydn, Mozart, Beethoven (New York: Norton, 1972),
20-23.
2. Sandra P. Rosenblum, Performance Practices in Classic Piano Music: Their Principles and
Applications (Bloomington: Indiana University Press, 1988), 2.
3. Emily Anderson, trans, and ed., The Letters of Mozart and His Family, 3rd. ed. (London:
Macmillan, 1985), 259.
4. Elliot Forbes, ed., Thayer's Life of Beethoven (Princeton, N.J.: Princeton University Press,
1967), 405, 554, and 716.
5. A. Peter Brown, Performing Haydns "The Creation" (Bloomington: Indiana University
Press, 1986), 74.
6. Rosenblum, Classic Piano Music, 24.
7. Jaap Schröder, "A Performers Thoughts on Mozart's Violin Style," in Perspectives on
Mozart Performance, ed. R. Larry Todd and Peter Williams, 121 (Cambridge: Cambridge
University Press, 1991). See also Robin Stowell, Violin Technique and Performance Prac-
tice in the Late Eighteenth and Early Nineteenth Centuries (Cambridge: Cambridge Uni-
versity Press, 1985).
8. Malcolm Bilson, "Restoring Ingredients: Malcolm Bilson on the Fortepiano," in Inside
Early Music: Conversations with Performers, ed. Bernard D. Sherman (New York: Oxford
University Press, 1997), 299. I
9. Domenico Corrí, The Singer s Preceptor (London, 1810), reprinted in Edward Foreman,
ed., ThePorpora Tradition (New York: Pro Música Press, 1968), 52.
ίο. Daniel Gottlob Turk, Klavierschule, trans, and ed. Raymond H. Haggh as School of
Clavier Playing (Lincoln: University of Nebraska Press, 1982), 345-48.
11. Clementi quoted in Rosenblum, Classic Piano Music, 154.
12. Leopold Mozart quoted in Rosenblum, Classic Piano Music, 154. |
13. Neal Zaslaw, "Introduction to Part II, The Classical Era," in Performance Practice: Music
after 1600, ed. Howard Mayer Brown and Stanley Sadie (New York: Norton, 1990), 213.
14. Wye Jamison Allanbrook, Rhythmic Gesture in Mozart (Chicago: University of Chicago
Press, 1983), 27-28.
15. See Jean-Pierre Marty, "Mozart's Tempo Indications and the Problems of Interpreta-
tion," in Perspectives on Mozart Performance, ed. R. Larry Todd and Peter Williams, 55—
73 (Cambridge: Cambridge University Press, 1991).
16. See Rosenblum, Classic Piano Music, 306-11.
17. Richard Kramer, "Notes to Beethoven Education," Journal of the American Musicological
Society28, no. ι (Spring 1975): 75.
Notes to Pages Ί 02-109 313
18. Rosenblum, Classic Piano Music, 312.

19. Forbes, Thayer's Life of Beethoven, 555.
20. For a more complete discussion of the periodic phrase, see Rosen, Classical Style, chap. 3;
Roger Norrington, "Taking Music Off the Pedestal: Roger Norrington on Beethoven,"
in Inside Early Music: Conversations with Performers, ed. Bernard D. Sherman, 341 (New
York: Oxford University Press, 1997); and Charles Rosen, The Romantic Generation
(Cambridge, Mass.: Harvard University Press, 1995), 261.
21. Anderson, Letters of Mozart, 340.
22. Turk, Klavierschule, 363-64.
23. Corrí, Singer's Preceptor, 6.
24. Ibid., 35.
25. Paul and Eva Badura-Skoda, Interpreting Mozart on the Keyboard, trans. Leo Black (New
York: St. Martin's Press, 1962), 31.
26. Forbes, Thayer's Life of Beethoven, 688.
27. See Clive Brown, "Historical Performance, Metronome Marks, and Tempo in Beetho-
ven's Symphonies," Early Music 19 (May 1991): 247-58; Willy Hess, "The Right Tempo:
Beethoven and the Metronome," The Beethoven Newsletter 5 (1988): 16—17; Peter Stad-
ien, "Beethoven and the Metronome," Soundings 9 (1982): 38-73; and Rosenblum, Clas-
sic Piano Music, 3 23 - 48.
28. See William Malloch, "Carl Czerny's Metronome Marks for Haydn and Mozart Sym-
phonies," Early Music 16 (February 1988): 72- 82.
29. Nicholas Temperly, "Haydn's Tempos in The Creation," Early Music 19 (1991): 235-45.
30. Corri, Singer's Preceptor, 69.
31. Brown, Performing Haydn's "The Creation, "23.
32. Anderson, Letters of Mozart, 121-22.
35. Ibid., 486.
36. Ibid., 581.
37. Ibid., 769.
38. See also Thomas Bauman, "Mozart's Belmonte," Early Music 19 (1991): 557—64; Patricia
Lewy Gidwitz, "'Ich bin die erste Sängerin': Vocal Profiles of Two Mozart Sopranos,"
Early Music 19 (1991): 565—79; and Alessandra Campana, "Mozart's Italian Buffo Sing-
ers," Early Music 19 (1991): 580-84.
40. Corri, Singer's Preceptor, 3-4.
41. Charles Burney, The Present State of Music in France and Italy (London, 1773; facs. New
York: Broude Brothers, 1969), vii, 19.
44. Laurence Dreyfus, Bach's Continuo Group: Players and Practices in His Vocal Works (Cam-
bridge, Mass.: Harvard University Press, 1987), 87-88.
45. See Will Crutchfield, "The Prosodic Appoggiatura in the Music of Mozart and His Con-
temporaries," Journal of the American Musicological Society 42 (1989): 229-74, and
314 Notes to Pages Ί 09-129
Crutchfield, "Voices," in Performance Practice: Music after 1600, ed. Howard Mayer
Brown and Stanley Sadie, 298-99 (New York: Norton, 1990).
46. Frederick Neumann, Ornamentation and Improvisation in Mozart (Princeton, N.J.:
Princeton University Press, 1986), 184-215. See also Neumann, "A New Look at Mozarts
Prosodic Appoggiatura," in Perspectives on Mozart Performance, ed. R. Larry Todd and
Peter Williams, 92-116 (Cambridge: Cambridge University Press,ji99i).
47. Badura-Skoda, Interpreting Mozart, 76.
48. Neumann, Ornamentation in Mozart, 16.
49. Badura-Skoda, Interpreting Mozart, 70.
50. See the discussions of this particular problem in Neumann, Ornamentation in Mozart,
an
22-23, d Badura-Skoda, Interpreting Mozart, 87.
51. Paul Badura-Skoda, "Mozart's Trills," in Perspectives on Mozart Performance, ed. R. Larry
Todd and Peter Williams, 8-12 (Cambridge: Cambridge University Press, 1991). See also
Rosenblum, Classic Piano Music, 246—47. j
52. Neumann, Ornamentation in Mozart, 114.
53. See also Crutchfield, "Voices," 301, and John Spitzer, "Improvised Ornamentation in a
Handel Aria with Obbligato Wind Accompaniment," Early Musici6 (1988): 515=16.
54. For the complete aria, see "Luigi Marchesi Embellishing a Cherubini Rondo" as repro-
duced from a Vienna manuscript in Robert Haas, Auffuhrungspraxis der Musik: Hand-
buch der Musikwissenschaft (Wildpark-Potsdam: Akademische Verlagsgesellschaft Athe-
naion, 1931), 225-30.
55. Burney, Present State of Music, 377.
56. For the complete aria, sec Joseph Haydn Werke (Munich: G. Henle), XXVIII/i:i8o-2Oi.
For an excerpt from the aria, see Eva Badura-Skoda, "Improvisation: I. 3. The Classical
Period," The New Grove Dictionary of Music and Musicians, ed. Stanley Sadie (London:
Macmillan, 1980), vol. 9, 43-48.
57. Neumanns general attitude is one of cautious conservatism, while pianist Robert Levin
advocates a more adventurous approach: see his review of Neumanns book m Journal of
the American Musicological Society 41 (1988): 355-68. See also Levin, "Improvised Em-
bellishments in Mozart's Keyboard Music," Early Music 20 (1^92): 221-33; "Robert
Levin on Mozart and Improvisation," in Inside Early Music: Conversations with Perform-
ers, ed. Bernard D. Sherman, 341 (New York: Oxford University Press, 1997), 315-38; and
Christoph Wolff, "Cadenzas and Styles of Improvisation in Mozart's Piano Concertos,"
in Perspectives on Mozart Performance, ed. R. Larry Todd and Peter Williams, 228-38
(Cambridge: Cambridge University Press, 1991).
CHAPTER 4. ITALIAN BEL CANTO
1. Philip Duey, Bel Canto in Its Golden Age (New York: King's Crown Press, 1951), 5.
2. Owen Jander and Ellen T. Harris, "Bel canto," The New Grove Dictionary of Music and
Musicians, ed. Stanley Sadie (New York: Grove's Dictionaries, 2001), vol. 3,161.
3. Ibid.
4. For a more detailed discussion, see Clive Brown, Classical and Romantic Performing Prac-
tice, 1750-1000 (New York: Oxford University Press, 1999), chapters 2-6.
5. Charles de Bériot, Methode de violón (Mainz, 1858), iii, 219-20, quoted in Brown, Cías-
steal and Romantic, 55-57.
6. Gesualdo Lanza, The Elements of Singing (London, 1809); Domenico Corrí, The Singer's
Preceptor (London, 1810); Giacomo Gotifredo Ferrari, A Concise Treatise on Italian Sing-
ing (London, 1818); and Isaac Nathan, An Essay on the History and Theory of Music, and
on the Qualities, Capabilities, and Management of the Human Voice (London, 1823; rev. as
Musurgia Vocalis, 1836). For a more detailed discussion of all these sources, see Robert
Toft, Heart to Heart: Expressive Singing in England, 1/80-1830 (New York: Oxford Uni-
7. Manuel García, Traite complet de l'art du chant (fans, 1874; facs. Geneva, 1985), trans.
Donald V. Paschke as A Complete Treatise on the Art of Singing: Part Two (New York: Da
Capo Press, 1975) and A Complete Treatise on the Art of Singing: Part One (New York: Da
Capo Press, 1984); Garcia, Hints on Singing (New York: Edward Schuberth, 1894).
8. Charles Santly, Student and Singer: Reminiscences of Charles Santly (London, 1892), 79—
80.
9. Garcia, Complete Treatise: Part Two, 26—31.
ίο. Brown, Classical and Romantic, 344, 321; for a more detailed discussion of tempo terms,
see chaps, ю and n.
11. Garcia, Hints on Singing, 6г.
12. Charles Rosen, The Romantic Generation (Cambridge, Mass.: Harvard University Press,
1995), 611.
13. García, Traite complet, гг.
14. Howard Bushnell, Maria Malibran: A Biography of the Singer (University Park: Pennsyl-
vania State University Press, 1979), п.
15. Henry Pleasants, The Great Singers (New York: Simon and Schuster, 1966),1167; see his
entire entry on Nourrit and Duprez, 161-70.
16. Duey, Bel Canto, 153.
17. Isaac Nathan, Musurgia Vocalis: An Essay on the History and Theory of Music, and on the
Qualities, Capabilities and Management of the Human Voice (London, 1836), reprinted in
ThePorpora Tradition, ed. E. Foreman (New York: Pro Música Press, 1968), 145.
18. Bushnell, Maria Malibran, 9, quoted from an 1828 musical magazine.
19. Ibid., 4.
20. Garcia, Complete Treatise, Part One, 66.
21. Gioacchino Rossini, VocalisesandSolfeges(Bloomington, Ind.: Frangipani Press, 1986), 3.
22. Nathan, Musurgia Vocalis, 145. ¡
23. Thomas Busby, A Musical Manual or Technical Directory (London, 1828; facs. New York:
Da Capo Press, 1976), 182.
24. Garcia, Complete Treatise: Part Two, 150.
25. Garcia, Complete Treatise: Part One, 137.
26. See The Era of Adelina Patti, Nimbus Records (N1 7840/41). See also Robert Philip,
Early Recordings and Musical Style: Changing Tastes in Instrumental Performance, 1000—
ip$o (Cambridge: Cambridge University Press, 1992), and Will Crutchfield, "Vocal Or-
namentation in Verdi: The Phonographic Evidence," lyth-Century Music j (1983—4): 3—
54-
27. See Toft, Heart to Heart, 58-64, and Brown, Classical and Romantic, chap. 15.
28. Nicola Vaccai, Practical Method of Italian Singing (New York: Schirmer, 1975), 18.
29. Garcia, Complete Treatise, Part Two, 82-83.
30. See Brown, Classical and Romantic, 429-38 and chap. 15.
31. Vaccai, Practical Method, 3.
32. Pleasants, Great Singers, 91-93.
33. Richard Osborne, "// Barbiere di Siviglia (ii)," Gr0i# Music Online, ed. Laura Macy,
http://wwwgrovemusic.com (accessed June 2, 2004).
34. Bushnell, Maria Malibran, 88.
35. Laure-Cinthie Damoreau, Classic Bel Canto Technique, trans, and introduction by Victor
Rangel-Ribeiro (New York: Dover, 1997).
36. Vaccai, Practical Method, 16.
37. Ibid., 18.
38. Malcolm Sterling Mackinlay, Garcia the Centenarian and His Times (Edinburgh, 1908),
34, quoted in Brown, Classical and Romantic, 419.
39. Will Crutchfield, "Improvisation—II, 5, ii. Nineteenth-Century Vocal Music," The
New Grove Dictionary of Music and Musicians, ed. Stanley Sadie (New York: Groves Dic-
tionaries, 2OOl), Vol. 12, 121. i
40. See Austin Caswell, ed., "Embellished Opera Arias," Recent Researches in the Music of the
ipth and Early 20th Centuries, vols. 7 and 8 (Madison, Wis.: A-R Editions, 1989), no. 13.
41. Crutchfield, "Improvisation," 122.
CHAPTER 5. GERMAN LIEDER
ι. Eric Sams and Susan Youens, "Hugo Wolf," Grove Music Online, ed. Laura Macy, http://
www.grovemusic.com (accessed January 8, 2002).
2. Edward E Kravitt, "The Lied in 19th-century Concert Life," Journal of the American
Musicological Society 18 (1965): 210. See also Lorraine Gorrell, The Nineteenth-Century
German Lied (Portland, Ore.: Amadeus Press, 1993), 89.
3. Otto Erich Deutsch, Schubert: Memoirs by His Friends (New York: Macmillan, 1958), 49.
4. Barbara A. Petersen, Ton und Wort: The Lieder of Richard Strauss (Ann Arbor, Mich.:
UMI Research Press, 1980), 104.
5. Robert Schumann, On Music and Musicians (Berkeley: University of California Press,
1983)» 59-6o.
6. Deutsch, Memoirs, ιοί.
7. Ibid. i
8. Robert Winter, "Franz Schubert," Grove Music Online, ed. Laura Macy, http://www.
grovemusic.com (accessed August 30, 2001). See also Walter Dürr, "Schubert and Jo-
hann Michael Vogl: A Reappraisal," ipth-Century Music3 (1979/80): 128.
9. Philip Duey, Bel Canto in Its Golden Age (New York: King's Crown Press, 1951), 8-9.
ίο. Deutsch, Memoirs, 216.
π. Ibid., 273.
12. Ibid., 116, 336.
13. Quoted in Dietrich Fischer-Dieskau, Robert Schumann: Words and Music, the Vocal Com-
positions, trans. Reinhard G. Pauly (Portland, Ore.: Amadeus Press, 1981), 135,176.
14. Gorrell, German Lied, 83.
15. Henry Pleasants, The Great Singers (New York: Simon and Schuster, 1966), 155-56.
16. William Ashton Ellis, ed. and trans., RichardWagner's Prose Works, vol. 4, Art and Politics
(1895; repr., New York: Broude Brothers, 1966), 182.
17. Ellis, Wagners Prose Works, vol. 5, Actors and Singers, 204.
18. Henry Pleasants, trans, and ed., The Music Criticism of Hugo Wolf (New York: Holmes
and Meier, 1979), 108.
19. Lilli Lehmann, How to Sing, trans. Richard Aldrich (New York: Macmillan, 1955), 281,
239.
20. Pleasants, Music Criticism of Wolf, 95.
21. Ibid., 259.
22. Clive Brown, Classical and Romantic Performing Practice, 1/50—1000 (Oxford: Oxford
University Press, 1999), 114-27.
23. See David Schroeder, "Schubert the Singer," Music Review 49 (1988), 259.
24. Gorrell, German Lied, no.
25. Jan Swafford, Johannes Brahms: A Biography (New York: Knopf, 1997), 448.
26. Petersen, Ton und Wort, 156.
27. Quoted in Dürr, "Schubert and Vogl," 136.
28. Ibid., 126.
29. Deutsch, Memoirs, 226.
30. David Montgomery, "Modern Schubert Interpretation in the Light of Pedagogical
Sources of His Day," Early Music г1) (1997): ioi-i8; Montgomery, "Franz Schubert's Mu-
sic in Performance: A Brief History of People, Events, and Issues," in The Cambridge
Companion to Schubert, ed. Christopher Gibbs, 270-83 (Cambridge: Cambridge Uni-
31. Deutsch, Memoirs, 116-17.
32. Quoted in Brown, Classical and Romantic, 426.
33. Ernest Walker, "The Appoggiatura," Music and Letters 5 (1924): 140.
34. For a more complete discussion of turns and mordents, see Brown, Classical and Roman-
tic, 493-516·
35. Quoted in Robert Philip, Early Recordings and Musical Style (Cambridge: Cambridge
University Press, 1992), 208.
37. Pleasants, Music Criticism of Wolf, 22.
38. Lehmann, How to Sing, 143.
39. Ibid., 254.
40. See Philip, Early Recordings, 172-76, 216.
42. Deutsch, Memoirs, 337-38.
43. Quoted in R. Larry Todd, "Felix Mendelssohn-Bartholdy," Grove Music Online, ed.
Laura Macy, http://www.grovemusic.com (accessed October 13, 2001).
44. Quoted in Philip, Early Recordings, 219.

45. Ibid., 218.
46. Quoted in Swafford, Brahms, no.
47. Petersen, TOW und Wort, 156.
48. Edward F. Kravitt, "Tempo äs an Expressive Element in the Late Romantic Lied," Musi-
cal Quarterly 59 (1973): 516.
49. See Gwilym Beechey, "Rhythmic Interpretation: Mozart, Beethoven, Schubert, and
Schumann," Music Review 33 (1972): 233-48.
50. Desmond Shawe-Taylor, "Schubert as Written and Performed^" Musical Times 104
(1963): 626-28.
51. Pleasants, Music Criticism of Wolf 203.
52. Swafford, Brahms, 343.
53. Quoted in Pleasants, Music Criticism of Wolf, 174.
54. Philip, Early Recordings, 6.
CHAPTER 6. FRENCH MELODIES
1. See especially Betty Bannerman, ed. and trans., The Singer as Interpreter: Claire Croiza's
Master Classes (London: Victor Gollancz, 1989); Jane Bathori, On the Interpretation of the
Melodies of Claude Debussy, trans. Linda Laurent (Stuyvesant, N.Y.: Pendragon Press,
1998); and Pierre Bernac, Francis Poulenc: The Man and His Songs (London: Victor Gol-
lancz, 1977).
2. Bernac, Poulenc, 13.
3. Pierre Bernac, The Interpretation of French Song, 2nd ed. (New York: Norton, 1976).
4. Frits Noske, French Song from Berlioz to Duparc, trans. Rita Benton (New York: Dover,
1970), 328.
5. See Graham Johnson, French Song Companion (Oxford: Oxford University Press, 2000),
17. See also Songsfor Solo Voice and Orchestra, ed. Ian Kemp, vol. 13 of Hector Berlioz: New
Edition of the Complete Works (Kassel: Bärenreiter, 1967).
6. Bathori, Debussy, 69.
7. Francis Poulenc, Diary of My Songs, trans. Winifred Radford (London: Victor Gollancz,
1985). 77·
8. Paul Sperry, telephone conversation with author, December 8, 2003.
9. Arbie Orenstein, ed., A RaveÍReader (New York: Columbia University Press, 1990), 507.
ίο. Thomas Grubb, Singing in French: A Manual of French Diction and French Vocal Reper-
toire (NewYork: Schirmer, 1979), 18. , \
11. Quoted in Noske, French Song, 2.
12. Martin Cooper, French Music from the Death of Berlioz to the Death ofFauré (New York:
Oxford, 1951), 1-2.
13. Romagnesi s treatise is excerpted and translated in David Tunley, Salons, Singers, and
Songs: A Background to Romantic French Song, 1830-18/0 (Burlington, Vt.: Ashgate,
2002); for these passages, see pp. 258, 265, and 262. ¡
14. Jean-Michel Nectoux, Gabriel Fauré: A Musical Life, trans. Roger Nichols (Cambridge:
Cambridge University Press, 1991), 27,51,189.
Notes to Pages 200-213 3Ί9
15. Quoted in Nectoux, Musical Life, 471-72. See also Gabriel Fauré: His Life through His
Letters, ed. Jean-Michel Nectoux, trans. J. A. Underwood (New York: Marion Boyars,
1984), 252.
16. Quoted in Cooper, French Music, 114.
18. Georges Jean-Aubry, French Music of Today, trans. Edwin Evans (Plainview, N.Y.: Books
for Libraries Press, 1976), 222-23.
19. Louis Laloy, Louis Laloy (1874-1944) on Debussy, Ravel, and Stravinsky, trans. Deborah
Priest (Burlington, Vt.: Ashgate, 1999), 248-49.
20. Poulenc, Diary, 21,51, 63, 71, 79.
21. Bernac, Interpretation of French Song, 33.
22. Tunley, Sabns, 43.
23. Ibid., 259.
24. Ibid., 263.
25. Ibid., 103.
27. Bannerman, Singer as Interpreter, 13.
28. Ibid., 81.
29. Ibid., 36-37.
30. Tunley, Salons, 259.
31. Jean-Aubry, French Music, 212.
32. Marguerite Long, At the Piano with Ravel, trans. Olive Senior-Ellis (London: J. M. Dent
and Sons, 1973), 16.
33. See Noske, French Song, 163.
35. Tunley, Salons, 260.
36. Reynaldo Hahn, On Singers and Singing, trans. Leopold Simoneau (Portland, Ore.:
Amadeus Press, 1990), 29, 60-65.
38. Orenstein, Ravel Reader, 30-31.
39. Arbie Orenstein, Ravel: Man and Musician (New York: Columbia University Press,
1975), 163.
40. Laloy, Louis Laloy, 108.
41. Poulenc, Diary, 79,113,41, 69.
42. Hahn, Singers, 114—15.
43. Ibid., 63.
44. Ibid., 62.
45. Bathori, Debussy, 41, 85, 81.
46. Bernac, Poulenc, 45, 48.
47. Grubb, Singing in French, 104.
48. Tunley, Sabns, 265.
50. Nectoux, Musical Life, 43, 294.
52. Marguerite Long, At the Piano with Fame, trans. Olive Senior-Ellis (New York: Tap-
linger, 1981), 68.
53. Marguerite Long, At the Piano with Debussy, trans. Olive Senior-Ellis (London: J. M.
Dent and Sons, 1972), 19, 63, 71,13, 25.
54. Hahn, Singers, 115.
55. Bathori's singing is now available on a CD, Jane Bathori: Complete Solo Recordings
(Marston 51009).
57. Bathori, Debussy, 33,
60. Bernac, Poulenc, 44.
61. Bannerman, Singer as Interpreter, 92, 82, 95.
63. These recordings are available on a CD, Claude Debussy: The Composer as Pianist (Pier-
ian Recording Society, Pierian oooi).
64. Vlado Perlemuter and Hélene Jourdan-Morhange, Ravel according to Ravel, trans. Frances
Tanner (London: Kahn and Averill, 1988), 13.
65. Poulenc, Diary, in, 33.
66. Bernac, Poulenc, 45. :
CHAPTER 7. SECOND VIENNESE SCHOOL
1. Anton Webern, The Path to New Music, ed. Willi Reich, trans. Leo Black (Bryn Mawr,
Pa.: Theodore Presser, 1963). The essays in this publication were taken from lectures
given by Webern in a private home in Vienna in 1932 and 1933.
2. Quoted by Steuermanns sister, Salka Viertel, in Joan Allan Smith, Schoenberg and His
Circle: A Viennese Portrait (New York: Schirmer, 1986), 97.
3. Kathryn Bailey, The Life of Webern (Cambridge: Cambridge University Press, 1998), 50.
4. See Bailey, Webern, 54-58,116,140-47.
5. Nicholas Chadwick, A Survey of the Early Songs ofAlban Berg (Oxford: Bodleian Library,
i97*)> 81.
6. Alban Berg, Jugenlieder, ed. Christopher Hailey, 2 vols. (Vienna: Universal Edition,
1985); Alban Berg: Jugenlieder, Dietrich Fischer-Dieskau, baritone; Aribert Reimann, pi-
ano (Angel / EMI 27 019 51).
7. Quoted in Mark DeVoto, "Berg the Composer of Songs," in The\Berg Companion, ed.
Douglas Jarman (London: Macmillan, 1989), 37.
8. See George Perle, The Operas ofAlban Berg (Berkeley: University of California Press,
1980), 1:3. See also Chadwick, Survey, 49.
9. See DeVoto, "Composer of Songs," 47-52.
ίο. Louis Krasner and D. C. Seibert, "Some Memories of Anton Webern, the Berg Violin
Concerto, and Vienna in the 19305," Fanfare n, no. 2 (1987): 335.
ii. H. H. Stuckenschmidt, Arnold Schoenberg, trans. Edith Temple Roberts and Humphrey
Searle (New York: Grove Press, 1959), 147.
Notes to Pages 227-242 32Ί
12. Dika Newlin, Schoenberg Remembered: Diaries and Recollections (1035-76) (New York:
Pendragon Press, 1980), 164.
13. See Smith, Schoenberg and His Circle, app. 3, 255-68, for all of the Society for Private
Musical Performance programs and performers.
14. Ibid., 246.
15. Ibid., 96.
16. See Berg's article "The 'Problem of Opera " in Willi Reich, Alban Berg, trans. Cornelius
Cardew (New York: Harcourt, Brace and World, 1963), 63—64.
17. Edward Steuermann, The Not Quite Innocent Bystander: Writings of Edward Steuermann,
ed. Clara Steuermann, David Porter, and Günther Schuller (Lincoln: University of Ne-
braska Press, 1989), 83.
18. Lewis Foreman, "Webern and the BBC and the Berg Violin Concerto," Tempo, no. 178
(September 1991): ίο.
19. Arnold Schoenberg, Style and Idea: Selected Writings of Arnold Schoenberg, ed. Leonard
Stein, trans. Leo Black (Berkeley: University of California Press, 1984), 144.
20. Stuckenschmidt, Arnold Schoenberg, 45.
21. H. H. Stuckenschmidt, Schoenberg: His Life, World, and Work, trans. Humphrey Searle
(London: John Calder, 1977), 117.
22. Ibid., 158,188. \
23. Arnold Schoenberg, Arnold Schoenberg Self-Portrait: A Collection of Articles, Program
Notes, and Letters by the Composer about His Own Works, ed. Nuria Schoeriberg Nono
(Pacific Palisades, Calif.: Belmont Music, 1988), 14.
24. Schoenberg, Self-Portrait, 27.
25. Ibid., 62.
26. Schoenberg, Style and Idea, 146.
27. From Pierrot to Marteau: An International Conference and Concert Celebrating the
Tenth Anniversary of the Arnold Schoenberg Institute (University of Southern Califor-
nia School of Music, 1987), 4.
28. Smith, Schoenberg a nd His Circle, 52-53.
29. Steuermann, Innocent Bystander, 36.
30. Ibid., 37.
31. Stuckenschmidt, Life, World, andWork, 214.
32. Smith, Schoenberg and His Circle, 100.
33. Stuckenschmidt, Life, World, and Work, 279.
35. Ibid.
37. Milton Babbitt, conversation with author, October 13, 2002.
38. Steuermann, Innocent Bystander, 36—37.
39. From the preface to the Pierrot Lunaire score (Pacific Palisades, Calif.: Belmont Music,
1990).
40. Sharon Mabry, Exploring Twentieth-Century Vocal Music: A Practical Guide to Innovations
in Performance and Repertoire (New York: Oxford University Press, 2002).

43. Perle, Operas of Berg, vol. i, 2,03-5.
45. See Schoenberg, Style and Idea, 346, 300.
46. Krasner and Seibert, "Memories of Webern," 340.
49. Peter Heyworth, ed., Conversations with Klemperer (London: Victor Gollancz, 1973), 76.
50. Schoenberg, Self-Portrait, 117-18; see also Schoenberg, Style and Idea, 342.
CHAPTER 8. EARLY TWENTIETH-CENTURY NATIONALISM
1. See Caryl Emerson, The Life ofMusorgsky (Cambridge: Cambridge University Press,
i999)> 15-
2. See Elizabeth Wilson, Shostakovich: Λ Life Remembered (London: Faber and Faber, 1994),
20-23.
3. Richard Taruskin, "Musorgsky" (opera entry), Grove Music Online, ed. Laura Macy,
http://www.grovemusic.com (accessed June 20, 2003).
4. See Alexander Tumanov, The Life and Artistry of Maria Olenina-d'Alheim, trans. Christo-
pher Barnes (Edmonton: University of Alberta Press, 2000).
5. Claude Debussy, Debussy on Music, ed. Richard Langham Smith, trans. Francois Lesure
(Ithaca, N.Y.: Cornell University Press, 1977), 21.
6. Arbie Orenstein, Ravel: Man and Musician (New York: Columbia University Press,
i975)> 53·
7. Tumanov, Olenina-d'Alheim, 94.
8. Ibid., 142.
9. I am especially grateful to Caryl Emerson for providing a recording of Oleninas 1963
master classes and to Ivan Eubanks for translating the transcript of the recording.
ίο. Tumanov, Olenina-d'Alheim, 280.
11. Eubanks, translation of master class transcript, 15.
12. Robert Kenneth Evans, "The Early Songs of Sergei Prokofiev" (Ph.D. diss., Ohio State
University, 1971), 74.
13. Victor Borovsky, Chaliapin: A Critical Biography (London: Hamish Hamilton, 1988),
221.
14. Ibid., ίο.

15. Ibid., 4.
16. Ibid., 206, n. 140.
17. Ibid., 206.
18. Feodor Chaliapin, Man and Mask: Forty Years in the Life of a Singer, trans. Phyllis Mégroz
(London: Victor Gollancz, 1932), 184.
19. Harold C. Schonberg, The Great Pianists (New York: Simon and Schuster, 1987), 391.
20. Chaliapin, Man and Mask, 60.
21. Borovsky, Chaliapin, 196.

22. Ibid., 13.
23. Prokofiev, Piano Concerto No. 3 and Other Works, Great Recordings of the Century, An-
gel Records COLH 34 (33 rpm), liner notes, pp. 2, n.
24. Schonberg, Great Pianists, 417r.
25. Wilson, Shostakovich, 200.
26. Ibid., 58-59.
27. Ibid., 104-5.
28. Galina Vishnevskaya, Galina: A Russian Story, trans. Guy Daniels (New York: Harcourt
Brace, 1984), 356.
29. This recording was first issued in the West as an LP on Monitor MC 2020.
30. The Suite on Poems by Michelangelo, op. 145, was written for bass Evgeny Nesterenko and
issued on LP as Melodiya Cio-o6i6i-62. The Blok cycle, Seven Songs, op. 127, was writ-
ten for Vishnevskaya, Rostropovich, Oistrakh, and pianist Moisei Vainberg; Vish-
nevskaya and Rostropovich later recorded it with Ulf Hoelscher and Vasso Devetzi for
EMI, issued on LP as SLS 5055. Symphony No. 14, also written for Vishnevskaya and
Rostropovich, was issued on LP as Melodiya Мз45<Э7.
31. Vishnevskaya, Galina, 179.
32. Gonzalo Armero and Jorge de Persia, eds., Manuel de Falla: His Life and Works, trans.
Tom Skipp (Madrid: Ministerio de Cultura, Ediciones Opponax, 1996), 69.
33. See Timothy Mitchell, Flamenco Deep Song (New Haven, Conn.: Yale University Press,
1994).
34. Armero and de Persia, Falla, 108.
35. Manuel de Falla, On Music and Musicians, trans. David Urman and J. M. Thomson
(London: Marion Boyars, 1979), 105.
36. Armero and de Persia, Falla, 109.
37. Suzanne Rhodes Draayer, A Singer's Guide to the Songs of Joaquín Rodrigo (London:
Scarecrow Press, 1999), 8.
38. Ibid., 23.
39. Ibid., ix.
40. Nico Castel, A Singer's Manual of Spanish Lyric Diction (New York: Excalibur Publishing,
1994), 137-41-
41. Draayer, Songs of Rodrigo, 8-и.
42. Jacqueline Cockburn and Richard Stokes, trans, and eds., The Spanish Song Companion
(London: Victor Gollancz, 1992), 249—52.
43. See Humphrey Carpenter, Benjamin Britten: A Biography (London: Faber and Faber,
1992), 35-
44. Sydney Northcote, Byrdto Britten: A Survey of English Song (London: John Baker, 1966), 22.
45. Imogen Hoist, The Great Composers: Britten (London: Faber and Faber, 1966), 26.
46. Harry Plunket Greene, Interpretation in Song (New York: Macmillan, 1912), x.
47. Carpenter, Britten, ιοί.
48. Ibid., II2.
49. Benjamin Britten, On Receiving the First Aspen Award (London: Faber and Faber, 1964),
11-13.
50. Holst, Britten, 35.

51. Alan Blyth, Remembering Britten (London: Hutchinson, 1981), 66.
52. Ibid., 94.
53. Ibid., 40.
54. Hoist, Britten, 64.
55. Blyth, Remembering Britten, 116.
56. Ibid., 154.
57. Ibid., 138.
58. Ibid., 56.
59. Carpenter, Britten, 265.
60. Rosamond Strode, "Working for Benjamin Britten (II)," in The Britten Companion, ed.
Christopher Palmer (Cambridge: Cambridge University Press, 1984), 53.
61. John Culshaw, "'Ben-A Tribute to Benjamin Britten," in The Britten Companion, ed.
Christopher Palmer (Cambridge: Cambridge University Press, 1984), 65.
63. Britten, War Requiem, Decca OSA 1255, 33 rpm. Reissued on CD as Decca 414383.
64. Benjamin Britten, "On Writing English Opera," Орегагг, no. ι (1961): 7.
66. Anthony Gishford, ed., Tribute to Benjamin Britten on His Fiftieth Birthday (London:
Faber and Faber, 1963), 13.
67. Benjamin Britten, "On Realizing the Continuo in Purcell s Songs," in Henry Purcell,
1659-1695: Essays on His Music, ed. Imogen Hoist (London: Oxford University Press,
1959), 8.
68. Ibid., 13.
69. Michael Burden, e-mail message to author, September i, 2003.
70. Charles E. Ivés, Memos, ed. John Kirkpatrick (New York: Norton, 1972), 115.
71. Ibid., 116.
72. Ibid., 116-17.
73. Ibid., 127,117.
74. Aaron Copland, Copland on Music (New York: Doubleday, 1960), 166.
75. This recording was first issued as an LP on CBS Masterworks 3211 0017. It has been reis-
sued on CD as part of vol. 3 of A Copland Celebration (Sony 89329).
76. Howard Pollack, "Aaron Copland," section I, Grove Music Online, ed. Laura Macy,
http://www.grovemusic.com (accessed July 23, 2003).
77. Barbara B. Heyman, Samuel Barber: The Composer and His Music (New York: Oxford
University Press, 1992), 104.
78. Heyman, Barber, 107,122.
79. Ibid., 287.
80. The Price/Barber studio recording of the Hermit Songs was reissued on LP as Odyssey 32
16 0230, also including Steber singing Knoxville; both of these performances have been
reissued on CD as part of Sony Masterworks Heritage MHK 60899.
81. Heyman, Barber, 326.
CHAPTER 9. WORKING WITH LIVING COMPOSERS
1. Phyllis Bryn-Julson, conversation with author, Ocean Grove, N.J., November 15, 2003;
Benita Valente, telephone conversations with author, January 13, 28, 2004; and Paul
Sperry, conversation with author, Princeton, N.J., December 8, 2003, and by telephone,
January 17, 2004.
2. Cheryl Bensman, telephone conversation with author, January 3, 2004.
3. Sharon Mabry, Exploring Twentieth-Century Vocal Music: A Practical Guide to Innovations
in Performance and Repertoire (New York: Oxford University Press, 2002).
4. Quoted in Paul Griffiths, "Writing Music That Sings, Cries, Screams and Prays," New
York Times, October 27, 2002.
5. Steve Reich, Tehillim LP recording (ECM Records 1-1215), liner notes.
6. Jon Magnussen, conversation with author, Princeton, N.J., January 6, 2004.
7. Steve Reich, "About This Edition," in Music for 18 Musicians (London: Boosey &
Hawkes, 2000).
8. Meredith Monk, remarks made at a Composers Colloquium at Princeton University,
March 13, 2003.
9. Griffiths, "Writing Music."
ίο. Jane Manning, New Vocal Repertory, vol. 2 (London: Oxford University Press, 1998), i.
π. William Duckworth, Talking Music: Conversations with John Cage, Philip Glass, Laurie
Anderson, and Five Generations of American Experimental Composers (New York: Schir-
mer, 1995), 91.
12. Peter Westergaard to William Parcher, October 31, 2003. Used by permission.
13. Duckworth, Talking Music, 357.
14. Cori Ellison, "Downtown Divas Expand Their Horizons," New York Times, sec. 2, Octo-
ber 28, 2OOI.
For Further Reading
CHAPTER 1. THE EARLY BAROQUE
Primary Sources
Bacilly, Bénigne de. Remarques curieuses sur I art de bien chanter. Paris, 1668. Trans-
lated by Austin B. Caswell as A Commentary upon the Art of Proper Singing.
New York: Institute of Mediaeval Music, 1968.
Bardi, Pietro de'. Letter to G. B. Doni. Florence, 1634. In Source Readings in Mu-
sic History: The Baroque Era, edited by Oliver Strunk. New York: Norton, 1965.
Bernhard, Christoph. Von der Singe-Kunst, oder Manier. Ms., ca. 1649. ^n Die
Kompositionslehre Heinrich Schützens in der Fassung seines Schulers Christoph
Bernhard, edited by Joseph Maria Müller-Blattau. Kassel: Bärenreiter, 1963.
Translated by Walter Hilse as 'On the Art of Singing: or, Manier." Music Fo-
rum 3(1973): 13-29.
Caccini, Giulio. "Euridice—Dedication." Florence, 1600. In Source Readings in
Music History: The Baroque Era, edited by Oliver Strunk. New York: Norton,
1965.
. Le nuove musiche. Florence, 1601. Reprint, New York: Broude Brothers,
1987. Modern edition edited by H. Wiley Hitchcock in Recent Researches in the
Music of the Baroque Era, vol. 9. Madison, Wis.: A-R Editions, 1970. Preface
translated into English in John Playford, An Introduction to the Skill of Mustek.
London, 1654. (Preface also in Source Readings in Music History: The Baroque
Era, edited by Oliver Strunk. New York: Norton, 1965.)
327
328 For Further Reading: Chapter Ί
Cofone, Charles J. F., ed. Elizabeth Rogers, Hir Virginall Eooke. 1656. Reprint, New York:
Dover, 1975.
Dowland, John. Andreas Ornithoparcus His Micrologus, or Introduction: Containing the Art of
Singing. London, 1609. Excerpted in Readings in the History of Music in Performance,
edited by Carol MacClintock. Bloomington: Indiana University Press, 1979.
Finck, Hermann. Practica música, book V. Wittenberg, 1556. In Readings in the History of
Music in Performance, edited by Carol MacClintock. Bloomington: Indiana University
Press, 1979.
MafFei, Giovanni Gamillo. Letter on singing. Naples, 1562. In Readings in the History of Mu-
sic in Performance, edited by Carol MacClintock. Bloomington: Indiana University Press,
1979.
Mersenne, Marin. Harmonie universelle, contenant la théorie et la pratique de la musique.
Paris, 1636. Excerpted in Readings in the History of Music in Performance, edited by Carol
MacClintock. Bloomington: Indiana University Press, 1979.
Monteverdi, Claudio. The Letters of Claudio Monteverdi. Edited by Denis Stevens. Oxford:
Clarendon Press, 1995.
Ortiz, Diego. Tratado de glosas sobre clausulas y otros géneros de puntos en la música de violones.
Rome, 1553. Edited by Max Schneider. New York: Bärenreiter, 1967.
Peri, Jacopo. "Euriaice—Foreword " Florence, 1601. In Source Readings in Music History: The
Baroque Era, edited by Oliver Strunk. New York: Norton, 1965.
Playford, John. An Introduction to the Skill of Mustek. London, 1694. nth ed. corrected and
amended by Henry Purcell. Edited by Franklin B. Zimmerman. Reprint, New York: Da
Capo Press, 1972.
Praetorius, Michael. Syntagma musicum. Wolfenbiittel, 1619. Excerpted in Readings in the
History of Music in Performance, edited by Carol MacClintock. Bloomington: Indiana
University Press, 1979.
Simpson, Christopher. The Division Violist. London, 1659. Reprint, New York: Schirmer, 1955.
Tosí, Pier Francesco. Opinioni de'cantonantichi e moderni. Bologna, 1723. Reprint, Bologna:
Forni, 1968. Translated by [John Ernst] Galliard as Observations on the Florid Song. 2nd
ed., London, 1743. Reprint edited by Michael Pilkington. London: Stainer and Bell, 1987.
Vicentino, Nicola. L'antica música. Rome, 1555. Excerpted in Readings in the History of Mu-
sic in Performance, edited by Carol MacClintock. Bloomington: Indiana University Press,
1979.
Wilson, John, ed. Roger North on Music. London: Novelle, 1959.
Zacconi, Lodovico. Prattica di música utile et necessaria si al compositore. . . si anco al cantore.
Venice, 1596. Excerpted in Readings in the History of Music in Performance, edited by Carol
MacClintock. Bloomington: Indiana University Press, 1979.
General
Brown, Howard Mayer. "Editing." The New Grove Dictionary of Music and Musicians, edited
by Stanley Sadie, vol. 5, 846—48. London: Macmillan, 1980.
. "Performing Practice." The New Grove Dictionary of Music and Musicians, edited by
Stanley Sadie, vol. 14, 370-93. London: Macmillan, 1980.
For Further Reading: Chapter Ί 329
Brown, Howard Mayer, and Stanley Sadie, eds. Performance Practice: Music after 1600. New
York: Norton, 1989.
Carter, Stewart, ed. A Performer's Guide to Seventeenth-Century Music. New York: Schirmer,
1997.
Charles, Sydney Robinson. "Editions, Historical." The New Grove Dictionary of Music and
Musicians, edited by Stanley Sadie, vol. 5, 848-69. London: Macmillan, 1980.
Cyr, Mary. Performing Baroque Music. Portland, Ore.: Amadeus Press, 1992.
Donington, Robert. Baroque Music: Style and Performance. London: Faber Music, 1982.
. The Interpretation of Early Music. New version. London: Faber, 1974.
. "Ornaments." The New Grove Dictionary of Music and Musicians, edited by Stanley
Sadie, vol. 13, 827-67. London: Macmillan, 1980.
. A Performer's Guide to Baroque Music. New York: Scribner, 1973.
Fuller, David. "Performer as Composer." In Performance Practice: Music after 1600, edited by
Howard Mayer Brown and Stanley Sadie. New York: Norton, 1989.
Harnoncourt, Nikolaus. Der musikalische Dialog: Gedanken zu Monteverdi, Bach und
Mozart. Salzburg: Residenz, 1984. Translated by Mary O'Neill as The Musical Dialogue:
Thoughts on Monteverdi, Bach and Mozart. Portland, Ore.: Amadeus Press, 1989.
Harris, Ellen T. "Voices." In Performance Practice: Music after 1600, edited by Howard Mayer
Brown and Stanley Sadie. New York: Norton, 1989.
MacClintock, Carol, ed. Readings in the History of Music in Performance. Bloomington: In-
diana University Press, 1979.
Neumann, Frederick. Essays on Performance Practice. Ann Arbor, Mich.: UMI Press, 1982.
Rangel-Ribeiro, Victor. Baroque Music: A Practical Guide for the Performer. New York: Schir-
mer, 1981.
Rogers, Nigel. "Voices." In Companion to Baroque Music, edited by Julie Anne Sadie. Lon-
don: J. M. Dent and Sons, 1990.
Sadie, Julie Anne, ed. Companion to Baroque Music. London: J. M. Dent and Sons, 1990.
Sanford, Sally. "Solo Singing i." In A Performer's Guide to Seventeenth-Century Music, edited
by Stewart Carter. New York: Schirmer, 1997.
Sherman, Bernard D., ed. Inside Early Music: Conversations with Performers. New York: Ox-
ford University Press, 1997.
Stevens, Denis. "Performance Practice in Baroque Vocal Music." College Music Symposium 18
(1978): 9-19. i
Strunk, Oliver, ed. Source Readings in Music History. Vol. 3, The Baroque Era. New York: Nor-
ton, 1965.
Wolff, Hellmuth Christian, ed. Original Vocal Improvisations from the i6th—i8th Centuries.
Anthology of Music 41 (Cologne: A. Volk, 1972), 95-100.
Accompaniment Instruments
Buetens, Stanley. "Theorbo Accompaniments of Early Seventeenth-Century Italian Mon-
ody." Journal of the Lute Society of America 6 (1973): 37-45.
Dart, Thurston, and John Morehen. "Tablature." The New Grove Dictionary of Music and
Musicians, edited by Stanley Sadie, vol. 18, 506-15. London: Macmillan, 1980.
330 For Further Reading: Chapter Ί
Lindley, Mark. Lutes, Viols and Temperaments. Cambridge: Cambridge University Press,
1984. ,
North, Nigel. Continuo Playing on the Lute, Archlute and Theorbo. Bloomington: Indiana
Poulton, Diana. "Lute." The New Grove Dictionary of Music and Musicians, edited by Stan-
ley Sadie, vol. π, 342—65. London: Macmillan, 1980.
Spencer, Robert. "Chitarrone, Theorbo and Archlute." Early Music4 (1976): 409—22.
Figured Bass
Arnold, Franck Thomas. The Art of Accompaniment from a Thorough-fiass as Practiced in the
Seventeenth and Eighteenth Centuries. London: Oxford University Press, 1931. Reprint,
New York: Dover, 1965.
Borgir, Tharald. The Performance of the Basso Continuo in Italian Baroque Music. Studies in
Musicology 90. Ann Arbor, Mich.: UMI Research Press, 1987.
Buelow, George J. Thorough-Bass Accompaniment According to Johann David Heinichen. Rev.
ed. Studies in Musicology 86, 219-36. Ann Arbor, Mich.: UMI Research Press, ca. 1986.
Hill, John Walter. "Realized Continuo Accompaniments from Florence c. 1600." Early Mu-
sic u, no. 2 (April 1983): 194-208.
Holman, Peter. "Continuo Realization in a Playford Songbook." EarlyMusic6 (1978): 268-69.
Mangsen, Sandra. "The Unfigured Bass and the Continuo Player: More Evidence from
France." Early Key board Journal ¿ (1984-85): 5-12.
O'Dette, Paul, and Jack Ashworth. "Basso Continuo." In A Performer s Guide to Seventeenth-
Century Music, edited by Stewart Carter. New York: Schirmer, 1997.
Williams, Peter. Figured Bass Accompaniment. Edinburgh: Edinburgh University Press, 1970.
Pitch
Ellis, Alexander John. "On the History of Musical Pitch." Journal of the [Royal] Society of the
Artsiu (1880): 293-336,400-403; 29 (1881): 109-12.
Haynes, Bruce. A History of Performing Pitch: The Story of "A." Oxford: Scarecrow Press,
2OO2.
Karp, Cary. "Pitch." In Performance Practice: Music after 1600, edited by Howard Mayer
Brown and Stanley Sadie. New York: Norton, 1989.
Lindley, Mark. "Tuning and Intonation." In Performance Practice: Music after1600, edited by
Howard Mayer Brown and Stanley Sadie. New York: Norton, 1989.
Mendel, Arthur. "Pitch in the Sixteenth and Early Seventeenth Centuries." Musical Quar-
terly^ (1948): 28-45,199-221, 336-57> 575-93-
. "Pitch in Western Music since 1500: A Re-Examination." Acta Musicologica 50
(1978): 1-93, 328.
Meyers, Herbert W. "Praetoriuss Pitch." Early Music-12, no. 3 (August 1984): 369-71.
. "Pitch and Transposition." In A Performer's Guide to Seventeenth-Century Music,
edited by Stewart Carter. New York: Schirmer, 1997.
Pirrotta, Antonino. "Temperaments and Tendencies in the Florentine Camerata." Trans-
lated by Nigel Fortune. Musical Quarterly 40, no. 2 (1954): 169-89.
For Further Reading: Chapter 1 331
Rhodes, J. J. K., and W. R. Thomas. "Pitch." The New Grove Dictionary of Music and Musi-
cians, edited by Stanley Sadie, vol. 14, 779-86. London: Macmillan, 1980.
Vibrato
Donington, Robert. "Vibrato." The New Grove Dictionary of Music and Musicians, edited by
Gable, Frederick K. "Some Observations concerning Baroque and Modern Vibrato." Per-
formance Practice Review 5 (1992): 90-102.
Moens-Haenen, Greta. Das Vibrato in der Musik des Barock: Ein Handbuch zur Aufführung-
spraxis für Vokalisten und Instrumentalisten. Graz: Akademische Druck- und Verlagsan-
stalt, 1988.
. "Vibrato." The New Grove Dictionary of Opera, edited by Stanley Sadie, 982-83.
London: Macmillan, 1992.
Neumann, Frederick. "Authenticity and the Vocal Vibrato." In New Essays on Performance
Practice. London: Research Press, 1989.
Historical Pronunciation
Copeman, Harold. Singing in Latin. Oxford: by the author, 1990.

Dobson, EricJ. English Pronunciation, 1500-1700. 2 vols. Oxford: Clarendon Press, 1968.
McGee, Timo thy J., ed. Singing Early Music: The Pronunciation of European Languages in the
Late Middle Ages and Renaissance. Bloomington: Indiana University Press, 1996.
Termini, Olga. "The Role of Diction and Gesture in Italian Baroque Opera." Performance
Practice Review 6 (1993): 146- 57.
Italian: D/spos/t/one and Ornamentation
Carter, Stewart. "Francesco Rognoni's Selva de varii passaggi (1620): Fresh Details concern-
ing Early Baroque Ornamentation." Performance Practice Review г (1989): 5-33.
Dickey, Bruce. "Ornamentation in Early-Seventeenth-Century Italian Music." In A Per-
former's Guide to Seventeenth-Century Music, edited by Stewart Carter. New York: Schir-
mer, 1997.
Erig, Richard, ed. Italienische Diminutionen—Italian Diminutions: The Pieces with More than
One Diminution from 1553 to 1638. Zürich: Amadeus, 1979.
Galliver, David. "'Cantare con la gorga: Coloratura Technique of the Renaissance Singer."
Studies in Music j (1973): 10-18.
Greenlee, Robert. "Dispositionedi voce: Passage to Florid Singing." Early Music ц (1987): 47-
55-
Hitchcock, H. Wiley. "Caccinis Other' Nuove musiche." Journal of the American Musicolog-
ical Society гJ (1974): 438-60.
. "Vocal Ornamentation in Caccinis Nuove musiche." Musical Quarterly 56 (1970):
389-404.
MacClintock, Carol. "Caccinis Trillo: A Re-examination." National Association of the Teach-
ers of Singing Bulletin, October 1976, 38—41.
332 For Further Reading: Chapter 1
Sanders, Donald C. "Vocal Ornaments in Durantes Arie devote (1609)." Performance Prac-
ticeReview 6 (1993): 60-76.
Sherman, Joy, and Lawrence R. Brown. "Singing Passaggi: Modern Application of a Cen-
turies-Old Technique." ChoralJournal (1995): 27-36.
Uberti, Mauro. "Vocal Technique in Italy in the Second Half of the Sixteenth Century."
Translated by Mark Lindley. Early Music 9 (1981): 486-95.
Italian: General
Aldrich, Putnam. Rhythm in Seventeenth-Century Italian Monody, with an Anthology of Songs

and Dances. London: J. M. Dent and Sons, 1966.
Baird, Julianne. "Solo Singing 2: The Bel Canto Style." In A Performer's Guide to Seven-
teenth-Century Music, edited by Stewart Carter. New York: Schirmer, 1997.
Celletti, Rodolfo. A History of Bel Canto. Translated by Frederick Fuller. Oxford: Clarendon
Press, 1991.
Palisca, Claude V. The Florentine Camerata. New Haven, Conn.: Yale University Press, 1989.
Rosselli, John. Singers of Italian Opera. Cambridge: Cambridge University Press, 1992.
Sawkins, Lionel. "For and Against the Order of Nature: Who Sang the Soprano?" Early Mu-
¿¿15(1987) 1315-24.
Wistreich, Richard. "'La voce e grata assai, ma. . . ' Monteverdi on Singing." Early Music 22
(1994): 9-15.
English
Baldwin, Olive, andThelma Wilson. "Purcell's Sopranos." Musical Times 123 (1982): 602-9.
Burden, Michael, ed. Performing the Music of Henry Purcell. Oxford: Clarendon Press, 1996.
, ed. The Purcell Companion. London: Faber and Faber, 1995.
Butler, Gregory. "Music and Rhetoric in Early Seventeenth-Century English Sources." Mu-
sical Quarterly 66 y no. ι (1980): 53-64.
Duckies, Vincent. "English Song and the Challenge of Italian Monody." In Words to Music:
Papers on English Seventeenth-Century Song. Los Angeles: William Andrews Clark Memo-
rial Library, University of California, 1967.
. "Florid Embellishment in English Song of the Late Sixteenth and Early Seven-
teenth Centuries." AnnalesMusicologiques 5 (1957): 332—45.
Giles, Peter. The Counter Tenor. London: Frederick Muller, 1982.
. The History and Technique of the Counter-Tenor. Aldershot, U.K.: Scolar Press, 1994.
Jones, Edward Huws. The Performance of English Song, 1610-1670. New York: Garland Press,
1989.
Parrott, Andrew. "Performing Purcell." In The Purcell Companion, edited by Michael Bur-
den. London: Faber and Faber, 1995.
Rose, Gloria. "A New Purcell Source." Journal of the American Musicological Society 25 (1972):
230-34.
Spink, Ian. English Song: Dowland to Purcell. New York: Scribner, 1974.
. "Playfords Directions for Singing after the Italian Manner." Monthly Musical
Record^ (i959):i3<D-35.
Toft, Robert. Tune Thy Musicke to Thy Hart: The Art of Eloquent Singing in England, 1597-
1622. Toronto: University of Toronto Press, 1993.
Wells, Robin H. "The Ladder of Love: Verbal and Musical Rhetoric in the Elizabethan Lute-
Song." Early Music 12 (1984): 173 - 89.
French
Anthony, James R. French Baroque Music: From Beaujoyeubc to Rameau. Rev. and expanded
ed. Portland, Ore.: Amadeus Press, 1997.
Fuller, David. "Notes inegales." The New Grove Dictionary of Music and Musicians, edited by
Gérold, Theodore. L'art du chant en France au XVIIe siecle. Strasbourg: Commission des
publications de la Faculté des lettres, 1921.
Hefling, Stephen E. Rhythmic Alteration in Seventeenth- and Eighteenth-Century Music: Notes
Inegales and Overdotting. New York: Schirmer, 1993.
Mather, Betty Bang, with Dean M. Karns. Dance Rhythms of the French Baroque: A Hand-
book for Performance. Bloomington: Indiana University Press, 1988.
Rosow, Lois. "French Baroque Recitative as an Expression of Tragic Declamation." Early
Music ii (1983): 468 -79.
Sanford, Sally. "A Comparison of French and Italian Singing in the Seventeenth Century."
Journal of the Society for Seventeenth-Century Music ι (1995), http://sscm-jscm.press.uiuc.
edu/jscm/vi/noi/sanford.html.
Scares, Margaret. "Mersenne on Vocal Diminutions." Performance Practice Review 6 (1993):
141-45·
Zaslaw, Neal. "The Enigma of the Haute-Contre." MusicalTimes115, no. n (1974): 939-41.
CHAPTER 2. THE LATE BAROQUE
Primary Sources
Agrícola, Johann Friedrich. Anleitung zur Singkunst. A translation [with additions] of Pier
Francesco Tosi's Opinioni de' cantón antichi e moderni. Berlin, 1757. Translated by Ju-
lianne Baird as Introduction to the Art of Singing. Cambridge: Cambridge University
Press, 1995.
Bach, Carl Philipp Emanuel. Versuch über die wahre Art das Clavier zu spielen. Berlin, 1753;
part 2,1762. Translated by W. J. Mitchell as Essay on the True Art of 'Playing Keyboard In-
struments. NewYork: Norton, 1949.
Bérard, Jean-Baptiste. L'art du chant. Paris, 1755. 2nd ed., 1756. Translated with commentary
by Sydney Murray. Milwaukee: Pro Música Press, 1969.
Montéclair, Michel Pignolet de. Principes de musique. Translated in Cantatas for One and
Two Voices, edited by James R Anthony and Dirán Akmajian. Madison, Wis.: A-R Edi-
tions, 1978.
Mozart, Leopold. Versuch einer gründlichen Violinschule. Augsburg, 1756. Translated by
Editha Knocker as A Treatise on the Fundamental Principles of Violin Playing. 2nd ed. Ox-
ford: Oxford University Press, 1985.
Quantz, Johann. Versuch einer Anweisung die Flöte traversiere zu spielen. Berlin, 1752. Trans-
lated by Edward R. Reilly as On Playing the Flute. 2nd ed. New York: Schirmer, 1985.
General
Glover, Cedric Howard. Dr. Charles Burney's Continental Travels. London: Blackie and Sons,
192-7-
Harnoncourt, Nikolaus. Baroque Music Today: Music as Speech. Translated by Mary O'Neill.
Portland, Ore.: Amadeus Press, 1988.
Hefling, Stephen E. Rhythmic Alteration in Seventeenth- and Eighteenth-Century Music: Notes
Inegales and Overdotting. New York: Schirmer, 1993.
Pleasants, Henry. The Great Singers. New York: Simon and Schuster, 1966.
Yorke-Long, Alan. Music at Court: Four Eighteenth-Century Studies. London: Weidenfeld
and Nicolson, 1954.
Ornamentation
Baird, Julianne. "An i8th-Century Controversy about the Trill." Early Music15 (1987): 36-
44.
Brown, Howard Mayer. "Embellishing Eighteenth-Century Arias: On Cadenzas." In Opera
and Vivaldi, edited by Michael Collins and Elise K. Kirk. Austin: University of Texas
Press, 1984.
Buelow, George J. "A Lesson in Operatic Performance Practice by Madame Faustina Bor-
doni." In A Musical Offering: Essays in Honor of Martin Bernstein, edited by Edward H.
Clinkscale and Claire Brook. New York: Pendragon Press, 1977.
Dean, Winton. "Vocal Embellishment in a Handel Aria." In Essays on Opera. Oxford:
Clarendon Press, 1990.
, ed. G. E Handel: Three Ornamented Arias. Oxford: Oxford University Press, 1976.
Haböck, Franz. Die Gesangskunst der Kastraten. Vienna: Universal, 1923.
. Die Kastraten und ihre Gesangskunst. Stuttgart: Deutsche Verlags-Anstalt, 1927.
Hall, James S., and Martin V. Hall. "Handels Graces." HandelJahrbuch ?> (1957): 25-43.
Neumann, Frederick. Ornamentation in Baroque and Post-Baroque Music, with Special Em-
phasis on]. S. Bach. Princeton, N.J.: Princeton University Press, 1978.
Handel
Burrows, Donald. Handel: Messiah. Cambridge Music Handbooks. Cambridge: Cambridge

Larsen, Jens Peter. Handel's Messiah: Origins, Composition, Sources. Westport, Conn.: Green-
wood Press, 1990.
Larue, C. Steven. Handel and His Singers: The Creation of the Royal Academy Operas, 1720—
1728. Oxford: Clarendon Press, 1995.
Rogers, Patrick. Continuo Realization in Handel's Vocal Music. Ann Arbor, Mich.: UMI Re-
search Press, 1989.
Bach
Butt, John. Bach: Mass in В Minor. Cambridge Music Handbooks. Cambridge: Cambridge
. Bach Interpretation: Articulation Marks in Primary Sources of'/. 5. Bach, Cambridge:
Cambridge University Press, 1990.
. Music Education and the Art of Performance in the German Baroque. Cambridge:
, ed. The Sacred ChoralMusic of J. S. Bach: Л Handbook. Brewster, Mass.: Paraclete
Press, 1997.
Dreyfus, Laurence. Bach's Continuo Group: Players and Practices in His Vocal Works. Cam-
bridge, Mass.: Harvard University Press, 1987.
Little, Meredith, and Natalie Jenne. Dance and the Music of ]. S. Bach. Bloomington: Indi-
ana University Press, 1991.
Marshall, Robert L. The Music of Johann Sebastian Bach. New York: Schirmer, 1989.
Mendel, Arthur. "On the Pitches in Use in Bach's Time." Musical Quarterly 41 (1953): 332-
545466-80.
Parrott, Andrew. The Essential Bach Choir. Woodbridge, Suffolk: Boydell Press, 2000.
Sherman, Bernard D., ed. "Triple Counterpoint: Jeffrey Thomas, Philippe Herreweghe, and
John Butt on Singing Bach." In Inside Early Music. New York: Oxford University Press,
1997-
Steinitz, Paul. Performing Bach's Vocal Music. Croydon, U.K.: Addington Press, 1980.
French
Cyr, Mary. "Declamation and Expressive Singing in Recitative." In Opera and Vivaldi,
edited by Michael Collins and Elise K. Kirk. Austin: University of Texas Press, 1984.
. "On Performing Eighteenth-Century Haute-Contre Roles." Musical Times 118
(1977): 291-95.
. "Performing Rameau's Cantatas." Early Musicii (1983): 480-88.
Rosow, Lois. "French Baroque Recitative as an Expression of Tragic Declamation." Early
Music и (1983): 468-79.
Sawkins, Lionel. "Doucement and légerement: Tempo in French Baroque Music." Early Mu-
sic2i (ΐ993): 365-74.
Tunley, David. The Eighteenth-Century French Cantata. Oxford: Clarendon Press, 1997.
CHAPTER 3. THE CLASSICAL ERA
Primary Sources
Anderson, Emily, ed. The Letters of Mozart and His Family. London: Macmillan, 1985.
Burney, Charles. The Present State of Music in France and Italy. Facsimile of the 1773 London
ed. New York: Broude Brothers, 1969.
Corrí, Domenico. The Singer's Preceptor. London, 1810. In The Porpora Tradition, edited by
Edward Foreman. New York: Pro Música Press, 1968.
Hiller, Johann Adam. Anweisung zum musikalisch-zierlichen Gesänge. Leipzig, 1780. Trans-
lated and edited by Suzanne J. Beicken as Treatise on Vocal Performance and Ornamenta-
tion. Cambridge: Cambridge University Press, 2001.
Mancini, Giambattista. Riflessionipratiche sul canto figurato. Vienna, 1774,1777. Translated
by Edward Foreman as Practical Reflections of Figured Singing. Champaign, 111.: Pro Mú-
sica Press, 1967.
Turk, Daniel Gottlob. Klavierschule. Leipzig, 1798. Translated and edited by Raymond H.
Haggh as School of Clavier Playing. Lincoln: University of Nebraska Press, 1982.
General
Bilson, Malcolm. "The Viennese Fortepiano of the Late Eighteenth Century." Early Music%
(1980): 158-62.
Brown, Howard Mayer, and Stanley Sadie, eds. Performance Practice: Music after 1600. [See
Part Two: The Classical Era.] New York: Norton, 1989.
Heyer, Anna Harriet. Historical Sets, Collected Editions, and Monuments of Music: Λ Guide to
Their Contents. 3rd ed. Chicago: American Library Association, 1980.
Hill, George R., and Norris L. Stephens. Collected Editions, Historical Series and Sets, and
Monuments of Music: A Bibliography. Berkeley, Calif: Fallen Leaf Press, 1997.
Krummel, D. W, and Stanley Sadie, eds. Music Printing and Publishing. New York: Norton,
1990. |
Rosen, Charles. The Classical Style: Haydn, Mozart, Beethoven. New York: Norton, 1972.
Rosenblum, Sandra P. Performance Practices in Classic Piano Music: Their Principles and Ap-
plications. Bloomington: Indiana University Press, 1988.
Sherman, Bernard D., ed. Inside Early Music. New York: Oxford University Press, 1997.
Stowell, Robin. Violin Technique and Performance Practice in the Late\ Eighteenth and Early
Nineteenth Centuries. Cambridge: Cambridge University Press, 1985.
Tempo and Rhythm
Allanbrook, Wye Jamison. Rhythmic Gesture in Mozart. Chicago: University of Chicago

Press, 1983.
Brown, Clive. "Historical Performance, Metronome Marks and Tempo in Beethovens Sym-
phonies." Early Music 19 (1991): 247-58.
Malloch, William. "Carl Czerny's Metronome Marks for Haydn and Mozart Symphonies."
Early Music 16 (1988): 72- 82.
Ornamentation
Badura-Skoda, Eva. "Improvisation: Western Classical." The New Grove Dictionary of Music
and Musicians, edited by Stanley Sadie, vol. 9,43-48. London: Macmillan, 1980.
Crutchfield, Will. "The Prosodic Appoggiatura in the Music of Mozart and His Contempo-
raries." Journal of the American Musicological Society 42 (1989): 229 -74.
Haas, Robert. Aufführungspraxis der Musik: Handbuch der Musikwissenschaft. Wildpark-
Potsdam: Akademische Verlagsgesellschaft Athenaion, 1931.
Levin, Robert. "Improvised Embellishments in Mozart's Keyboard Music." Early Music 20

(1992): 221-33.
. Review of Frederick Neumanns Ornamentation and Improvisation in Mozart in
Journal of the American Musicobgical Society 41 (1988): 355-68.
Neumann, Frederick. Ornamentation and Improvisation in Mozart. Princeton, N.J.: Prince-
ton University Press, 1986.
Smiles, Joan E. "Directions for Improvised Ornamentation in Italian Method Books of the
Late Eighteenth Century." Journalof *theAmericanMusicologicalSociety31 (1978): 495-509.
. "Improvised Ornamentation in Late Eighteenth-Century Music: An Examination
of Contemporary Evidence." Ph.D. diss., Stanford University, 1975.
Spitzer, John. "Improvised Ornamentation in a Handel Aria with Obbligato Wind Accom-
paniment." Early Music16 (1988): 514-22.
Haydn
Brown, A. Peter. Performing Haydn's "The Creation!' Bloomington: Indiana University Press,
1986.
Landon, H. C. Robbins. Haydn: Chronicle andWorks. 5 vols. Bloomington: Indiana Univer-
sity Press, 1976-80.
Temperley, Nicholas. Haydn: The Creation. Cambridge Music Handbooks. Cambridge:
Mozart
Badura-Skoda, Paul and Eva. Interpreting Mozart on the Keyboard. Translated by Leo Black.
New York: St. Martins Press, 1962.
Bauman, Thomas. "Mozart's Belmonte." Early Music'19 (1991): 557-64.
Bilson, Malcolm. "Interpreting Mozart." Early Music12 (1984): 519-22.
Campana, Alessandra. "Mozart's Italian Buffo Singers." Early Music 19 (1991): 580-84.
Eisen, Cliff. "The Old and New Mozart Editions." Early Musiciy (1991): 513-32..
Gidwitz, Patricia Lewy. "'Ich bin die erste Sängerin': Vocal Profiles of Two Mozart Sopra-
nos." Early Music 19 (1991): 565-79.
Söderström, Elisabeth. "A Life of Mozart Singing." Early Music 20 (1992): 139-41.
Todd, R. Larry, and Peter Williams, eds. Perspectives on Mozart Performance. Cambridge:
Beethoven
Cooper, Barry. Beethoven's Folksong Settings: Chronology, Sources, Style. Oxford: Clarendon
Press, 1994.
, ed. The Beethoven Compendium: A Guide to Beethoven's Life and Music. London:
Thames and Hudson, 1991.
Forbes, Elliot, ed. Thayer's Life of Beethoven. Princeton, N.J.: Princeton University Press,
1967.
Stowell, Robin, ed. Performing Beethoven. Cambridge: Cambridge University Press, 1994.
CHAPTER 4. ITALIAN BEL CANTO
Primary Sources
Damoreau, Laure-Cinthie. Classic Bel Canto Technique. Translated and introduced by Victor
Rangel-Ribeiro. New York: Dover, 1997.
Garcia, Manuel. Hints on Singing. New York: Edward Schuberth, 1894.
. Traite complet de l'art du chant. Paris, 1847, âcs· Geneva, 1985. Translated by Don-
ald V. Paschke as A Complete Treatise on the Art of Singing: Part Two, New York: Da Capo
Press, 1975, and A Complete Treatise on the Art of Singing: Part One, New York: Da Capo
Press, 1984.
Nathan, Isaac. Musurgia Vocalis: An Essay on the History and Theory of Music, and on the
Qualities, Capabilities, and Management of the Human Voice. London, 1836. In The Por-
pora Tradition, edited by Edward Foreman. New York: Pro Música Press, 1968.
Vaccai, Nicola. Practical Method of Italian Singing. London, 1834. Reprint, New York: Schir-
mer, 1975.
Performance Practice
Brown, Clive. Classical and Romantic Performing Practice, 1750-1000. Oxford: Oxford Uni-
versity Press, 1999.
Crutchfield, Will. "Voices." In Performance Practice: Music after 1600, edited by Howard
Mayer Brown and Stanley Sadie, 424-58. New York: Norton, 1989.
Toft, Robert. "The Expressive Pause: Punctuation, Rests and Breathing in England, 1770-
1850." Performance Practice Review 7 (1994): 199-232.
. Heart to Heart: Expressive Singing in England, 1/80-1830. Oxford: Oxford Univer-
sity Press, 2OOO.
General
Bushnell, Howard. Maria Malibran: A Biography of the Singer. University Park: Pennsylvania
State University Press, 1979.
Celletti, Rodolfo. A History of Bel Canto. Translated by Frederick Fuller. Oxford: Clarendon
Press, 1991.
Duey, Philip A. Bel Canto in Its Golden Age: A Study of Its Teaching Concepts. New York: Kings
Crown Press, 1951.
Franca, Ida. Manual of Bel Canto. New York: Coward-McCann, 1959.
Manen, Lucie. Bel Canto: The Teaching of the Classical Italian Song-Schools, Its Decline and
Restoration. Oxford: Oxford University Press, 1987.
Pleasants, Henry. The Great Singers. New York: Simon and Schuster, 1966.
Rosen, Charles. The Romantic Generation. Cambridge, Mass.: Harvard University Press,
1995.
Rosselli, John. The Life of Bellini. Cambridge: Cambridge University Press, 1996.
. The Opera Industry in Italy from Cimarosa to Verdi. Cambridge: Cambridge Uni-
. Singers of Italian Opera. Cambridge: Cambridge University Press, 1992.
Ornamentation
Caswell, Austin. "Mme Cinti-Damoreau and the Embellishment of Italian Opera in Paris,
1820-1845." Journal of the American MúsicologicalSociety 28 (1975): 459-92.
. "Vocal Embellishment in Rossini's Paris Operas: French Style or Italian?" Bollettino
del Centro Rossiniano di Studi 1-2 (1975): 5-21.
, ed., "Embellished Opera Arias." Recent Researches in the Music of the Nineteenth and
Early Twentieth Centuries, Vols. 7 and 8. Madison, Wis.: A-R Editions, 1989.
Crutchfield, Will. "Vocal Ornamentation in Verdi: The Phonographic Evidence." ipth-Cen-
tury Music 7 (1983 - 84): 3 - 54.
CHAPTER 5. GERMAN LIEDER
Primary Sources
Deutsch, Otto Erich, ed. Schubert: Memoirs by His Friends. London: Adam and Charles
Black, 1958.
Ellis, William Ashton, ed. and trans. Richard Wagner's Prose Works. London, 1895. Reprint,
New York: Broude Brothers, 1966.
Holland, Henry Scott, and W. S. Rockstro. Jenny Lind the Artist, 1820-1851: A Memoir of
Madame Jenny Lind Goldschmidt, Her Art-Life and Dramatic Career. New York: Charles
Scribner, 1893.
Lehmann, Lilli. How to Sing. Translated by Richard Aldrich. New York: Macmillan, 1902.
Nauhaus, Gerd, ed. The Marriage Diaries of Robert and Clam Schumann. Translated by Peter
Ostwald. Boston: Northeastern University Press, 1993.
Pleasants, Henry, ed. The Music Criticism of Hugo Wolf. New York: Holmes and Meyer, 1979.
Schumann, Robert. On Music and Musicians. Edited by Konrad Wolf. Translated by Paul
Rosenfeld. New York: Pantheon, 1946.
Performance Practice
Beechey, Gwilym. "Rhythmic Interpretation: Mozart, Beethoven, Schubert and Schu-

mann." Music Review 33 (1972): 233 -48.
Brown, Clive. Classical and Romantic Performing Practice: 1750-1000. Oxford: Oxford Uni-
Dürr, Walter. "Schubert and Johann Michael Vogl: A Reappraisal." ipth-Century Music 3
(1979-80): 126-40.
Kravitt, Edward F. "Tempo as an Expressive Element in the Late Romantic Lied." Musical
Quarterly (1973): 497-5*8.
Montgomery, David. "Franz Schubert's Music in Performance: A Brief History of People,
Events and Issues." In The Cambridge Companion to Schubert, edited by Christopher
Gibbs. Cambridge: Cambridge University Press, 1997.
. Franz Schubert's Music in Performance: Compositional Ideals, Notational Intent, His-
torical Realities, Pedagogical Foundations. Hillsdale, N.Y.: Pendragon Press, 2003.
. Historical Information for Musicians: Sourcebooks for the Style of Performance Prac-
tices in European Classical Music. Vol. i, Musical Tutors, Methods and Rebted Sources: c.
1650-1005. Huntingdon, U.K.: Kings Music, 1997.
. "Modern Schubert Interpretation in the Light of the Pedagogical Sources of His
Day." Early Music 25 (1997): 101-18.
Philip, Robert. Early Recordings and Musical Style: Changing Taste in Instrumental Perfor-
mance, 1000—1050, Cambridge: Cambridge University Press, 1992,.
Schenkman, Walter. "Beyond the Limits of Urtext Authority: A Contemporary Record of Early
Nineteenth-Century Performance Practice." College Music Symposium 23 (1983): 145-63.
Schroeder, David. "Schubert the Singer." Music Review 49 (1988): 254-66.
Shawe-Taylor, Desmond. "Schubert as Written and Performed." Musical Times 104 (1963):
626-28.
Van Tassel, Eric. "'Something Utterly New': Listening to Schubert Lieder—ι. Vogl and the
Declamatory Style." Early Music 25 (1997): 703-14.
Walker, Ernest. "The Appoggiatura." Music and Letters 5 (1924): 121-44.
Winter, Robert. "Keyboards." In Performance Practice: Music after 1600, edited by Howard
Mayer Brown and Stanley Sadie. New York: Norton, 1989.
General
Bell, Craig. The Songs of Schubert. Lowestoft, U.K.: Alston Books, 1964.
Daverio, John. Crossing Paths: Schubert, Schumann, and Brahms. New York: Oxford Univer-
sity Press, 2002.
. Robert Schumann: Herald of a "New Poetic Age." New York: Oxford University
Press, 1997.
Ferris, David. Schumanns Eichendorff "Liederkreis" and the Genre of the Romantic Cycle. New
York: Oxford University Press, 2000.
Fischer-Dieskau, Dietrich. Robert Schumann: Words and Music- The Vocal Compositions.
Translated by Reinhard G. Pauly. Portland, Ore.: Amadeus Press, 1981.
. Schubert's Songs: Λ Biographical Study. New York: Limelight Editions, 1984.
Gorrell, Lorraine. The Nineteenth-Century Lied. Portland, Ore.: Amadeus Press, 1993.
Kravitt, Edward F. The Lied: Mirror of Late Romanticism. New Haven, Conn.: Yale Univer-
sity Press, 1996.
. "The Lied in 19th-century Concert Life." Journal of ^theAmerican MusicologicalSo-
ciety18(1965): 207-18.
Miller, Richard. Singing Schumann: An Interpretive Guide for Performers. New York: Oxford
Petersen, Barbara. Ton und Wort: The Lieder of Richard Strauss. Ann Arbor, Mich.: UMI Re-
search Press, 1980.
Porter, Ernest Graham. Schubert's Song Technique. London: D. Dobson, 1961.
Rosen, Charles. The Romantic Generation. Cambridge, Mass.: Harvard University Press,
1995-
Sams, Eric. The Songs of Hugo Wolf. Oxford: Oxford University Press, 1962.
. The Songs of Johannes Brahms. New Haven, Conn.: Yale University Press, 2000.
Stein, Deborah, and Robert Spillmann. Poetry into Song: Performance and Analysis of Lieder.
Oxford: Oxford University Press, 1996.
Swafford, Jan. Johannes Brahms: A Biography. New York: Knopf, 1997.
Youens, Susan. Hugo Wolf: The Vocal Music. Princeton, N.J.: Princeton University Press,
1992.
. Hugo Wolf and His Mörike Songs. Cambridge: Cambridge University Press, 2000.
. Retracing a Winter's Journey: Schubern "Winterreise. "Ithaca, N.Y.: Cornell Univer-
sity Press, 1991.
-. Schubert: Die schöne Müllerin. Cambridge Music Handbooks. Cambridge: Cam-
bridge University Press, 1992.
. Schubert's Late Lieder: Beyond the Song-Cycles. Cambridge: Cambridge University
Press, 2OO2.
CHAPTER 6. FRENCH MELODIES
Primary Sources
Bannerman, Betty, ed. and trans. The Singer as Interpreter: Claire Croizas Master Classes.
London: Victor Gollancz, 1989.
Bathori, Jane. On the Interpretation of the Melodies of Claude Debussy. Translated by Linda
Laurent. Stuyvesant, N.Y.: Pendragon Press, 1998.
Bernac, Pierre. Francis Poulenc: The Man and His Songs. Translated by Winifred Radford.
. The Interpretation of French Song. New York: Norton, 1970.
Garden, Mary, and Louis Biancolli. Mary Gardens Story. New York: Simon and Schuster,
1951.
Hahn, Reynaldo. On Singers and Singing. Translated by Leopold Simoneau. Portland, Ore.:
Amadeus Press, 1990.
Jean-Aubry, Georges. French Music of Today. Translated by Edwin Evans. London, 1919.
Reprint, Plainview, N.Y.: Books for Libraries Press, 1976.
Laloy, Louis. Louis Laloy (1874-1044) on Debussy, Ravel, and Stravinsky. Translated by Debo-
rah Priest. Burlington, Vt.: Ashgate, 1999.
Long, Marguerite. At the Piano with Debussy. Translated by Olive Senior-Ellis. London: J. M.
Dent and Sons, 1972.
. At the Piano with Fame. Translated by Olive Senior-Ellis. New York: Taplinger,
1963.
. At the Piano with Ravel. Translated by Olive Senior-Ellis. London: J. M. Dent and
Sons, 1973.
Perlemuter, Vlado, and Hélene Jourdan-Morhange. Ravel according to Ravel. Translated by
Frances Tanner. London: Kahn and Averill, 1988.
Poulenc, Francis. Diary of My Songs. Translated by Winifred Radford. London: Victor Gol-
lancz, 1985.
Stravinsky, Igor. An Autobiography. New York, 1936. Reprint, New York: Norton, 1962.
Texts and Diction
Grubb, Thomas. Singing in French: Λ Manual of French Diction and French Vocal Repertoire.
New York: Schirmer, 1979.
Johnson, Graham, and Richard Stokes. A French Song Companion. New York: Oxford Uni-
versity Press, 2OOO.
Meister, Barbara. Nineteenth Century French Song: Fauré, Chausson, Duparc, and Debussy.
Bloomington: Indiana University Press, 1980.
Rohinsky, Marie-Claire. The Singer's Debussy. New York: Pelion Press, 1987.
General
Briscoe, James R., ed. Debussy in Performance. New Haven, Conn.: Yale University Press,
1999.
Cooper, Martin. French Music from the Death of Berlioz to the Death of Fauré. London: Ox-
ford University Press, 1951.
Nectoux, Jean-Michel. Gabriel Fauré: A Musical Life. Translated by Roger Nichols. Cam-
bridge: Cambridge University Press, 1991.
, ed. Gabriel Fauré: His Life through His Letters. Translated by J. A. Underwood. Lon-
don: Marion Boyars, 1984.
Noske, Frits. French Song from Berlioz to Duparc. Translated by Rita Benton. 2nd ed. New
York: Dover, 1988.
Orenstein, Arbie. Ravel: Man and Musician. New York: Columbia University Press, 1975.
, ed. A Ravel Reader: Correspondence, Articles, Interviews. New York: Columbia Uni-
CHAPTER 7. SECOND VIENNESE SCHOOL
Primary Sources
Foreman, Lewis. "Webern, the BBC and the Berg Violin Concerto." Tempo no. 178 (Sep-
tember 1991): 2-IO.
Heyworth, Peter, ed. Conversations with Klemperer. London: Victor Gollancz, 1973.
Krasner, Louis, and D. C. Siebert. "Some Memories of Anton Webern, the Berg Violin Con-
certo, and Vienna in the 19305." Fanfaren, no. 2 (1987): 335-47.
Milhaud, Darius. "To Arnold Schoenberg on His Seventieth Birthday: Personal Recollec-
tions." Musical Quarterly30 (October 1944): 379-84.
Newlin, Dika. Schoenberg Remembered: Diaries and Recollections (1958—76). New York: Pen-
dragon Press, 1980.
Schoenberg, Arnold. Arnold Schoenberg Self-Portrait: A Collection of Articles, Program Notes,
and Letters by the Composer about His Own Works. Edited by Nuria Schoenberg Nono. Pa-
cific Palisades, Calif: Belmont Music, 1988.
. Style and Idea: Selected Writings of Arnold Schoenberg. Edited by Leonard Stein.
Translated by Leo Black. Berkeley: University of California Press, 1975.
Searle, Humphrey. "Conversations with Webern." Musical Times 81 (1940): 405-6.
Smith, Joan Allen. Schoenberg and His Circle: A Viennese Portrait. New York: Schirmer,
1986.
Steuermann, Edward. The Not Quite Innocent Bystander: Writings of Edward Steuermann.
Edited by Clara Steuermann, David Porter, and Günther Schuller. Translated by Richard
Cantwell and Charles Messner. Lincoln: University of Nebraska Press, 1989.
. "'Pierrot lunaire in Retrospect." Journal of the Arnold Schoenberg Institute 2, no. ι
(1977): 49-51.
Webern, Anton. The Path to the New Music. Edited by Willi Reich. Translated by Leo Black.
London: Universal, 1963.
Schoenberg
Brinkmann, Reinhold. "On ΥΙΖΪΪΟΪ$ΎΪΜ\" Journal of the Arnold Schoenberg Institute г, no. 2
(1978): 42-48.
Dunsby, Jonathan. Schoenberg: Pierrot lunaire. Cambridge Music Handbooks. Cambridge:
Frisch, Walter, ed. Schoenberg and His World. Princeton, N.J.: Princeton University Press,
1999.
Stein, Leonard. "A Note on the Genesis of the Ode to Napoleon." Journal of the Arnold
Schoenberg Institute г, no. 1^(1977): 52-54.
Stuckenschmidt, H. H. Arnold Schoenberg. Translated by Edith Temple Roberts and Hum-
phrey Searle. New York: Grove Press, 1959.
. Schoenberg: His Life, World, and Work. Translated by Humphrey Searle. London:
John Calder, 1977.
Berg
Chadwick, Nicholas. "Bergs Unpublished Songs in the Österreichische Nationalbibliotek."

Music and'Letters 52, no. 2 (1971): 123-40.
. "A Survey of the Early Songs of Alban Berg." Bodleian Library thesis, 1972.
Defotis, William. "Berg's Op. 5: Rehearsal Instructions." Perspectives ofNewMmicij, no. ι
(1986): 131-37.
DeVoto, Mark. "Some Notes on the Unknown Altenberg Lieder." Perspectives of New Music
5, no. ι (1966): 37-74·
Jarman, Douglas, ed. The Berg Companion. London: Macmillan Press, 1989.
Perle, George. The Operas of Alban Berg, г vols. Berkeley: University of California Press,
1980,1985.
Reich, Willi. Alban Berg. Translated by Cornelius Cardew. New York: Harcourt, Brace and
World, 1963.
Webern
Bailey, Kathryn. The Life of Webern. Cambridge: Cambridge University Press, 1998.
Moldenhauer, Hans and Rosaleen. Anton von Webern: A Chronicle of His Life ana Work. Lon-
don: Victor Gollancz, 1978.
General
Austin, William. Music in the Twentieth Century: From Debussy through Stravinsky. New
York: Norton, 1966.
Heyworth, Peter. Otto Klemperer: His Life and Times. Vol. i. Cambridge: Cambridge Uni-
Mabry, Sharon. Exploring Twentieth-Century Vocal Music: A Practical Guide to Innovations in
Performance and Repertoire. New York: Oxford University Press, 2002.
Simms, Bryan R. "The Society for Private Musical Performance: Resources and Docu-
ments. " Journal of the Arnold Schoenberg Institute?» no. 2 (1979): 127=49.
Slonimsky, Nicolas. Music since ipoo. New York: Norton, 1938.
Weisberg, Arthur. Performing Twentieth-Century Music: A Handbook for Conductors and In-
strumentalists. New Haven, Conn.: Yale University Press, 1993.
CHAPTER 8. EARLY TWENTIETH-CENTURY NATIONALISM
Russia
Borovsky, Victor. Chaliapin: A Critical Biography. London: Hamish Hamilton, 1988.

Chaliapin, Feodor. Man and Mask: Forty Years in the Life of a Singer. Translated by Phyllis
Mégroz. London, 1932. Reprint, St. Clair Shores, Mich.: Scholarly Press, 1973.
Emerson, Caryl. TheLifeofMusorgsky. Cambridge: Cambridge University Press, 1999.
Evans, Robert Kenneth. "The Early Songs of Sergei Prokofiev and Their Relation to the Syn-
thesis of the Arts in Russia, 1890-1922." Ph.D. diss., Ohio State University, 1971.
Nice, David. Prokofiev: From Russia to the West, 1801-1035. New Haven, Conn.: Yale Univer-
sity Press, 2003.
Redepenning, Dorothea. "Sergei Prokofiev." Grove Music Online, edited by Laura Macy,
http://www.grovemusic.com, 2003.
Schonberg, Harold C. The Great Pianists. New York: Simon and Schuster, 1987.
SemeonofF, Boris. "Chaliapins Repertoire and Recordings." Record Collector го (1971-72):
171-230.
Taruskin, Richard. "Musorgsky" (opera entry). Grove Music Online, edited by Laura Macy,
http://www.grovemusic.com, 2003. ι
. "Prokofiev" (opera entry). Grove Music Online, edited by Laura Macy, http://www.
grovemusic.com, 2003.
Tumanov, Alexander. The Life and Artistry of Maria Olenina-d'Alheim. Translated by
Christopher Barnes. Edmonton: University of Alberta Press, 2000.
Vishnevskaya, Galina. Galina: A Russian Story. Translated by Guy Daniels. New York: Har-
court Brace Jovanovich, 1984.
Wilson, Elizabeth. Shostakovich: A Life Remembered. London: Faber and Faber, 1994.
Spain
Armero, Gonzalo, and Jorge de Persia, eds. Manuel de Falla: His Life and Works. Translated
by Tom Skipp. Madrid: Ediciones Opponax, 1996.
Cockburn, Jacqueline, and Richard Stokes, eds. and trans. The Spanish Song Companion.
Draayer, Suzanne Rhodes. A Singer's Guide to the Songs of Joaquín Rodrigo. London: Scare-
crow Press, 1999.
Falla, Manuel de. On Music and Musicians. Translated by David Urman and J. M. Thomp-
son. London: Marion Boyars, 1979.
Hess, Carol A. "Manuel de Falla." Grove Music Online, edited by Laura Macy, http://www.
Katz, Israel J. "Flamenco." Grove Music Online, edited by Laura Macy, http://www.
Mitchell, Timothy. Flamenco Deep Song. New Haven, Conn.: Yale University Press, 1994.
England
Blyth, Alan, ed. Remembering Britten. London: Hutchinson, 1981.

Brett, Philip. "Benjamin Britten." Grove Music Online, edited by Laura Macy, http://www.
Britten, Benjamin. "On Realizing the Continuo in PurcelPs Songs." In Henry Purcell, 1659-
1695: Essays on His Music, edited by Imogen Hoist. London: Oxford University Press, 1959.
. On Receiving the First Aspen Award. London: Faber and Faber, 1964.
. "On Writing English Opera." Operaiz, no. ι (1961): 7-8.
Carpenter, Humphrey. Benjamin Britten: A Biography. London: Faber and Faber, 1992.
Duncan, Ronald. Working with Britten: A Personal Memoir. Bideford, U.K.: Rebel Press,
1981.
Gishford, Anthony, ed. Tribute to Benjamin Britten on His Fiftieth Birthday. London: Faber
and Faber, 1963.
Greene, Harry Plunket. Interpretation in Song. New York: Macmillan, 1912.
Hoist, Imogen. Britten. London: Faber and Faber, 1966.
Johnson, Graham. Britten, Voice and Piano: Lectures on the Vocal Music of Benjamin Britten.
London: Guildhall School of Music and Drama, 2003.
Northcote, Sydney. Byrdto Britten: A Survey of English Song. London: John Baker, 1966.
Palmer, Christopher, ed. The Britten Companion. Cambridge: Cambridge University Press,
1984.
Reed, Philip, ed. The Travel Diaries of Peter Pears, 1956-1978. Woodbridge, U.K.: Boydell
Press, 1995.
United States
Burkholder, Peter. "Charles Ivés." Grove Music Online, edited by Laura Macy, http://www.
Copland, Aaron. Copland on Music. New York: Doubleday, 1960.
Heyman, Barbara B. Samuel Barber: The Composer and His Music. New York: Oxford Uni-
Ivés, Charles E. Memos. Edited by John Kirkpatrick. New York: Norton, 1972.
Pollack, Howard. "Aaron Copland." Grove Music Online, edited by Laura Macy, http:
//www.grovemusic.com, 2003.
Diction
Castel, Nico. A Singer's Manual of Spanish Lyric Diction. New York: Excalibur Publishing,
1994·
Cheek, Timothy. Singing in Czech: A Guide to Czech Lyric Diction andôcal Repertoire. Lon-
don: Scarecrow Press, 2001.
Piatak, Jean, and Regina Avrashov. Russian Songs and Arias: Phonetic Readings, Word-by-
Word Translations, anda Concise Guide to Russian Diction. Dallas: Pst. Inc., 1991.
CHAPTER 9. WORKING WITH LIVING COMPOSERS
General
Duckworth, William. Talking Music: Conversations with John Cage, Philip Glass, Laurie An-
derson, and Five Generations of American Experimental Composers. New York: Schirmer,
1995.
Mabry, Sharon. Exploring Twentieth-Century Vocal Music: A Practical Guide to Innovations in
Performance and Repertoire. New York: Oxford University Press, 2002.
Rockwell, John. All American Music: Composition in the Late Twentieth Century. New York:
Da Capo Press, 1997.
Weisberg, Arthur. Performing Twentieth-Century Music: A Handbook for Conductors and In-
strumentalists. New Haven, Conn.: Yale University Press, 1993.
Repertoire
Lust, Patricia. American Vocal Chamber Music, 1045-1080: An Annotated Bibliography. Lon-
don: Greenwood Press, 1985.
Manning, Jane. New Vocal Repertory: An Introduction. London: Macmillan Press, 1986.
. New Vocal Repertory 2. Oxford: Clarendon Press, 1998.
Notation
Cope, David. New Music Notation. Dubuque, Iowa: Kendall/Hunt, 1976.

Karkoschka, Erhard. Notation in New Music. New York: Praeger, 1972.
Read, Gardner. Music Notation: A Manual of Modern Practice. New York: Taplinger, 1979.
Risatti, Howard. New Music Vocabulary: A Guide to Notational Signs for Contemporary Mu-
sic. Urbana: University of Illinois Press, 1975.
Stone, Kurt. Music Notation in the Twentieth Century: A Practical Guidebook. New York:
Norton, 1980.
Index
absolute pitch, 213, 248, 298, 299 Addison, Adele, 281, 283
accentuation marks, 129,138,169,181,192 affect, 18,56, 65, 77, 84, 88, 93,100
acciaccatura, 147,174 Agrícola, Johann Friedrich, 54, 59, 6ι, 64,
accompaniment: realized, 56, 94, 277-78; 65,67-68,75,79-84
bel canto, 133,134,141,147,156; lieder, appoggiatura: early Baroque, 46—47; late
164,165,167,169,171,186,188; melodies, Baroque, 59, 67, 71-74, 79-82, 87-89,
:
95' X97> 2OO> 2O5> 2I2; Second Viennese 112; in recitatives, 67, 71-72,109-10,125,
School, 225, 233, 248, 249; new music, 144; Classical, 112—15,I25> bel canto,
298 144—47, *595 lieder, 171,173—79,184,192;
accuracy in performance, 196, 202, 209, melodies, 201
212-13, 2,18-20, 237, 238, 247, 249, 250, aria: early Baroque, 32; late Baroque, 55, 69,
258, 259, 267, 275, 284, 299 71, 75, 77, 79, 86, 91; Classical, 94,103,
acoustics of performing spaces, 54, 56, 93, 104,107,119-20; bel canto, 128,131,143,
127,187, 274, 285 149,151,156,159
acting: early Baroque, 33; lieder, 167-69; Arroyo, Martina, 284
melodies, 199, 200, 201; Second Vien- articulation: early Baroque, 5,18, 21—22, 51;
nese School, 231—32, 234, 244; 20th- late Baroque, 58—64, 65, 66, 84, 90, 96—
century Russian, 257, 258-59, 260, 261, 97; Classical, 93, 94, 95, 96, 97-98,124-
265, 266, 284; 20th-century English, 25; bel canto, 129-32,141,156; lieder,
276—84; new music, 295, 301—2 166,169-71,193; melodies, 199, 203-9;
347
348 Index
articulation (continued} Beaser, Robert, 290

Second Viennese School, 243—44, 2495 Beethoven, Ludwig van, 92, 94, 95, 96,
20th-century, 260, 285; new music, 289, 100—106,124,168, 223, 227
294, 301, 303 "Adelaide," 104-5, H2--I4,160
atonal music, 222, 224, 229, 233, 237, 248, An dieferne Geliebte, 160,166
249, 298 Fidelio, 168
avant-garde: in Paris, 195; in Vienna, 223; in Folk Song Arrangements, 94,102
London, 272; in US, 278 "Klage" WoO 113,100-102
Bellini, Vincenzo, 126,128,129,130,133,135,
Babbitt, Milton, 237, 293, 297 144,147,151,159,167
Philomel, 296 "La farfalletta," 146
Phonemena, 293, 297, 298, 300 "Per pieta, bell' idol mió," 158
Bach, Carl Philipp Emanuel, 53, 54, 56, 64, "Quando incise," 144-45,148
79, 80, 92, 98,102,174 "Sogno d'infanzia," 150
Bach, Johann Christoph "Vanne, o rosa Fortunata," 145
"Cara la dolce fiamma," 124 Bensman, Cheryl, 286, 288, 292, 298, 299,
Bach, Johann Sebastian, 54, 55,59, 62, 65, 300
66, 67, 79, 84-85, 223 Bérard, Jean-Baptiste, 54, 64, 65, 87, 90
Cantata 140 "Wenn kömmst du, mein Berberían, Cathy, 287—88
Heil?," 59-61, 80 Berg, Alban, 222, 223, 224—26, 227, 228,
Magnificat, "Et exultavit," 57 229, 242-43, 249, 274
Magnificat, "Quia fecit," 57—58 Altenberg Lieder, 224, 226
Mass in B minor, 85 Sieben frühe Lieder, 224, 225—26
St. John Passion, "Zerfliesse mein Herze," Der Wein, 224, 226, 229, 230
80,82 Wozzeck, 223, 224, 229, 242
St. Matthew Passion, 69 Berio, Luciano, 287
St. Matthew Passion, "Erbarme dich," Berlioz, Hector, 195,197,198, 201, 203, 218
80-81 Les nuits d'été, 197
Bacilly, Bénigne de, 6,14,41-51, 59, 87, Bernac, Pierre, 194,196,197, 201, 202, 209,
206-7, 221 212, 218, 219, 273, 277, 283
Baird, Julianne, 25, 54, 68, 77 Bernacchi, Antonio, 71, 97
Baker, Janet, 275, 276 Bernstein, Leonard, 302
Bampton, Rose, 282 Bizet, Georges, 195, 203, 212
Barber, Samuel, 251, 252, 279, 281—84 Bolcom, William, 286, 289
Andromache's Farewell, 284 Let Evening Come, 289
Dover Beach, 282 Botti, Susan, 303—4
Hermit Songs, 283 Boulez, Pierre, 286, 293, 297, 299, 302
Knoxville: Summer of 1915, 282-83 Pliselonpli, 290
Melodiespassageres, 283 boys' voices, 33,42-43, 84-85,106,197, 276
Bardac, Emma, 200 Brahms, Johannes, 160,161,162,163,164,
Bathori, Jane, 194,196,197, 200, 201, 202, 165,166,170-71,173,186,188,191, 223,
209, 210, 213-15, 218-19, 257 227, 249, 302
Bauernfeld, Eduard von, 172 Liebeslieder Walzes, 188
Beardslee, Bethany, 238, 293, 296, 298 breathing: and vibrato, 65; and cadenzas,
Index 349
77; bel canto, 130,131,141,159; melodies, clefs: early Baroque, 7, 9; late Baroque, 87
199, 204, 218, 219; 20th-century (Russia), Coclico, Adrian Petit, 23
260; 20th-century (England), 275; new Colbran, Isabella, 133
music, 295 coloratura, 21, 75,131,133,143,151, 269
Britten, Benjamin, 189, 251, 252, 273-78, computer generated scores, 290, 291
281, 284 continuo: early Baroque, 32; late Baroque,
A Ceremony of Carols, 276 55-56, 65, 67, 69, 77, 86; Classical, 96.
Les illumination, 277 See also figured bass; thoroughbass
Peter Grimes, 274 Copland, Aaron, 251, 252, 280-81, 284
War Requiem, 276 Emily Dickinson Songs, 281, 283
Broadway, 10, 34, 289 The Second Hurricane, 281
Bryn-Julson, Phyllis, 286, 287, 290, 293, Corigliano, John, 291—92
295, 297, 299, 300 Corrí, Domenico, 93-97, 98,103,105-6,
Burney, Charles, 71,108,120,124 108,109,120,124,130,139,141
Butler, Martin, 296 countertenor, 14, 34, 42
Croiza, Claire, 194,196, 202, 203, 204-5,
Caballé, Montserrat, 271, 272 209, 213, 215, 218, 257
cabaret, 287, 289, 300 Crumb, George, 286, 289, 291, 297
Caccini, Giulio, 6,10,12, 20, 22, 25-31, 33, Ancient Voices of Children, 294^ 296
35,136,141 Cuénod, Hugues, 205
"Amarilli mía bella," 27-28, 33 Cui, César, 252, 256
cadence (vibrato), 14, 87 Cuzzoni, Francesca, 71, 75
cadential trill, 47
cadenza: late Baroque, 71, 75, 77, 82, 84, 91; d'Alheim, Maria Olenina. See Olenina
Classical, no, 120,125; bel canto, 129,143, d'Alheim, Maria
147,149,154-56,159; Heder, 172 d'Alheim, Pyotr, 256—57
Cage, John, 288, 291, 303 dance forms: early Baroque, 33, 41, 43-44,
Caldara, Antonio 45; late Baroque, 54, 56-58, 78, 90; Clas-
"Sebben, crudele," 72-73 sical, 93, 98, loo, 125,132; Spanish, 266-
cantabile style, 64, 90, 98,103,105,107,132, 67,268
151, 203, 242, 244 de Los Angeles, Victoria, 251, 271, 272
cantata: early Baroque, 32; late Baroque, 55, Debussy, Claude, 194,195,196,197,198,
85, 86-87, 91 200, 202, 203, 213-15, 218-19, 225> 237>
cantejondo, 266, 268, 269 256, 267
castrad, 20,42, 43,71, юб, 126,134-35,143 Chanson deBilitis, "La Chevelure," 215-
cercar della nota, 26,141 16
Chabrier, Emmanuel, 201, 212 "Claire de lune," 203-4
"L'íle heureuse," 215, 217 "Fantoches," 210-11
Chaliapin, Fyodor, 251, 258-60, 272, 284 "Le faune," 215
Chausson, Ernest, 198, 212 "La Flute de Pan," 214—15
ChenYi "Green," 219
As in a Dream, 294, 305 "II pleure dans mon coeur," 219
Chorley, Henry, 168,172 "Les ingénus," 214-15
Cinti-Damoreau, Laure, 143,156 "Mandoline," 218
350 Index
2 2
Debussy, Claude (continued] ond Viennese School, 243—45, 48> 49;
"La mort des amants," 210 new music, 293, 294, 301
"Placet futile," 210-11
Quatre chansons dejeunesse, 197 early instruments. See instruments,
declamatory style, 19,40,167,172,193, 233- period
34; 20th-century Russian, 254, 256, 258, editions: early Baroque, 9, n, 51; late
261, 265, 266 Baroque, 55, 86, 90, 94; Classical, 93-95,
DeGaetani, Jan, 238, 289, 298 124; bel canto, 128-29; lieder, 162-64,
DelTredici, David, 286, 288, 299 192; melodies, 19^—99; Second Viennese
Alice Symphony, Final Alice, Child Alice, School, 224—26; 20th-century Russian
288 255, 265—66; 20th-century Spanish, 272;
I Hear an Army, 299 20th-century English, 278; 20th-century
Delibes, Leo, 195, 212 US, 280, 281, 284;!new music, 290, 293
Diaghilev, Sergei, 195, 253, 259, 261, 272 electronics in new music, 287, 296—97, 302,
diction: French, 41, 49-51, 59, 86, 87,194, 303
2I 22J
198-99, 200, 203-9, 3> ; Italian, 41, esclamazione, 21, 26
129,130; and late Baroque rhythm, 56,
59, 64, 90; German, 59—61,166—69,191, facsimile, 7, 51, 86, i¿8
192-93, 231, 237, 247; in recitatives, 67, Falla, Manuel de, 251, 267-71, 272, 285
69; Russian, 253-54, 258, 259, 261, 266, El amor brujo, 268^-71
285; Spanish, 271, 285; English, 273, 276, Seven Spanish Popular Songs, 267-68,
285 269-70
disposition, dispositione, 22, 26, 45 falsetto, 20, 34,42, 66, 85, ю6,136,172
divisions, diminutions, 21-22, 28, 37-41, Farinelli (Carlo Broschi), 71, 97
47-49, 61, 71, 74-75» 77> 84, 87,119,125. Farrell, Eileen, 283
See also coloratura Fauré, Gabriel, 194,195,196,197,198, 200,
Dolukhanova, Zara, 261, 264 202, 203, 212-13, 218, 281
Domingo, Placido, 272 "Apres un revé," 218
Donizetti, Gaetano, 75,126,128,133,135, "Claire de lune," 203—4, 2I3
138,141,144,147,149,159,167 L'horizon chimérique, 218
"Eterno amore e fe," 142 "Notre amour," 205—6
"La ninna nanna," 148 "Le parfum impérissable," 209-10
"Una lacrima," 157 Requiem, 218
Dorliak, Nina, 261, 264 Requiem, "Pie Jesu," 197
dotted figures, 44, 61-62,104-5, Π5> I3I» "Revé d'amour," 198
176,189-91 "Soir," 200, 204—55
Dowland, John, 9, n, 15-17,18, 33, 275 Faustina, Bordoni, 71
"I saw my lady weepe," 15-16 feigned voice, 20,136
Dukas, Paul, 267, 271 Ferrabosco, Alfonso, 37, 39
Duparc, Henri, 195,198, 212, 218 Ferrari, Giacomo Gotifredo, 130
Duprez, Gilbert-Louis, 135 Ferner, Kathleen, 283
dynamics: early Baroque, 6; late Baroque, figured bass, ιο-ιι, 33^ 51, 69, 90, 277. See
55, 84; Classical, 95, 96,124; bel canto, also continuo; thoroughbass
141,159; lieder, 187; melodies, 218; Sec- Fischer-Dieskau, Dietrich, 226, 276
Index 351
Folio, Cynthia, 296 Haydn, Franz Joseph, 92,94, 95,105,106,

folk music, 251, 252, 285; Russian, 252, 253, 133
258, 266; Spanish, 266, 268; American, The Creation, 94,104,106,115,117,118,
279, 281, 289; Mongolian, 305 124
folksingers, π, 288 IlRitorno di Tobia, 124
fortepiano, 95-96, 98,125,187 Hendricks, Barbara, 288
Foss, Lukas, 304 Hiller, Johann Adam, 93,109-10,119
Freund, Marya, 229, 236-37, 238, 257
Friedländer, Max, 163 Imperio, Pastora, 268
improvisation: early Baroque, 10, 27, 277-
Galimir, Felix, 227, 238, 246-47 78; late Baroque, 55, 74, 75, 77, 79, 91;
Garcia, Manuel Patricio Rodriguez, 130— Classical, 93, no, 119-25; bel canto, 143,
40,141,146,147,149,151-56,159,163, 149-59; Heder, 171,172; Spanish, 267;
166,167, 260 new music, 287, 293, 295
Garcia, Manuel (senior), 130,136,149 inégalité, 44-45, 61-62, 66. See also rhyth-
Garden, Mary, 194, 200, 209, 215, 219 mic alteration
Glinka, Mikhail, 252, 267 instrumentation: early Baroque, 6, 41, 51;
Gluck, Christoph Willibald, 112 late Baroque, 54, 87
Golijov, Osvaldo, 288, 294 instruments: Baroque accompaniment, 10—
gorgia, 14, 2i 13, 87; modern, 66, 96, 97,102; period,
Gounod, Charles-Francois, 195, 212 66, 78, 90, 95-97,102; new music, 287,
grace note, 114,144-47,149,159,174,176, 296, 299, 305. See also fortepiano; harpsi-
294, 305 chord; lute; organ; piano; theorbo
graces: early Baroque, 21, 32, 33, 34, 35, 43; Ivés, Charles, 251, 279-80, 284
early Baroque (Italian), 21—32; early "At Parting," 279
Baroque (English), 35-41; early Baroque "General William Booth Enters into
(French), 45-49; late Baroque, 75, 86; Heaven," 279-80
Classical, 93, no, 112-19,125; bel canto, 114 Songs, 280
143, i44-49> 151» 159
Granados, Enrique 267, 271 jazz, 7, 44, 287, 288, 294, 300, 304
Grey, Madeleine, 200, 201, 257 Jean-Aubry, Georges, 200, 202
gruppo, 25, 28, 35 Joachim, Joseph, 171,173,192
Guadagni, Gaetano, 71, 75, 78 Joyce, Brooke, 290, 299
guitar, n, 12, 267
Gutheil-Schoder, Marie, 229, 234 Kernis, Aaron Jay, 290, 291
Kim, Earl, 289, 300
Hagen, Daron, 286, 301 Klemperer, Otto, 227, 247
Hahn, Reynaldo, 194, 200, 204, 209, 213 Koussevitsky, Serge, 271, 281, 282-83
Handel, George Frideric, 55, 67, 69, 71, 75,
131,151 La Barbara, Joan, 303
Messiah, 66, 77-78, 85, 273 Laine, Cleo, 237
Ottone, "Affanni del pensier," 75-76 Lanza, Gesualdo, 130
harpsichord, 12, 67, 69, 87, 95-96 legato: early Baroque, 22; late Baroque, 59,
haute-contre, 14,42—43 62-64, 90; Classical, 96, 97-98,125;
352 Index
legato (continued) century, 252; 20th-century Russian 264;

bel canto, 129,137,139,141,156; Heder, new music, 291
169,183,186; melodies, 209; Second Meyerbeer, Giacomo, 130,167
Viennese School, 244; 20th-century // Crociato, 134
Russian, 260; 20th-century English, microtones, 287, 298-99, 305
278; 20th-century, 285; new music, 300 Milhaud, Darius, 200, 225, 236, 279, 281
Lehmann, Lilli, 164,166,168-69, iSi» 183 Minter, Drew, 22, 24, 37
Leon, Tania, 294 Monk, Meredith, 292, 303-4
Ligeti, György, 286, 293 Montéclair, Michel Pignolet de, 54, 62, 64,
Aventures, 293, 295, 300 86,87
Nouvelles aventures, 293 Ariane et Bachus, 62-63, 88
Lind, Jenny, 167,173 Monteverdi, Claudio, 7,12,17-18, 20, 21,
Liszt, Franz, 127,160,162,191,195, 200, 302 22, 24, 27-28,41, 203, 282, 300
Llubera, Lina, 261 "Lasciatemi moriré," 7—8
Lully, Jean-Baptiste, 12,41, 49, 69, 86 Laúdate Dominum, 27-28
lute, ίο, и, 12,13, 32, 33, 35, 289 Lettere amorose, "E pur destina," 26—27
Orfeo, 12, 27
Mackey, Steven, 288 mordent, 84,147,149,171,181
Magnussen, Jon, 289 Morris, Joan, 289
Mahler, Alma, 236, 246 Mozart, Leopold, 53, 56, 64, 65, 79, 92, 98,
Mahler, Gustav, 162,189, 224, 225, 281 102
Das Lied von der Erde, 226 Mozart, Wolfgang Arnadeus, 75, 92, 94, 95,
Malibran, Maria, 129,130,135,136,143 96, 97,103,105,106,108,109,112,114,
manuscript, 7-9, 35, 37, 39-41,51, 72, 75, 124,130,131,133,138,151,187,191, 289
94, 95, no, 114,124,128,149,163,164, Abduction from the Seraglio, 107
182, 282, 290 "Abendemfindung," 160
Marchesi, Luigi, 119—21 "Ach ich fühl's," 104
Massenet, Jules, 195, 212 "Ah, lo previdi," 107
McNair, Sylvia, 290 "Batti, batti," 131-32
Mendelssohn, Felix, 160,161,162,163,164, "Bella mía fiamma," no—n
167,172,187,188 "Dans un bois solitaire," 95
Elijah, 172 Don Giovanni, ιοο, 130, 227
"Frühlingslied" Op. 8,172—73 Exultate jubilate, n6
"Frühlingslied" Op. 47,172 La finta giardiniera, 94
messa di voce, 21, 26, 47, 64, 66, 88, 97, 98, Lucio Silla, "Ah se a morir mi chiama,"
107,136 124
meter: late Baroque, 56-58, 69, 90; Classi- Mass in C Minor, "Domine," n6,117
cal, 98,100-104,125; bel canto, 132-33; Mass in С Minor, "Laudamus te," 118,
Heder, 176,192; melodies, 219; 20th- 119
century Russian 254, 258 "Non so d'onde viene," 124
metronome, 103—4, 2I3> 2I5 Sech Notturni, 95
metronome marks: Classical, 104,125; bel "Das Veilchen," 114—15,160,169
canto, 133; Heder, 188; melodies, 218—19; Musorgsky, Modest, 251, 252, 253-58, 259,
Second Viennese School, 247; 20th- 261, 265, 285
Index 353
Boris Godunov, 255, 259, 261 phrasing: Classical, 94, 95, 97; bel canto,
The Nursery, 254-55, 2,56-58, 261 130-31,138,159; Heder 185-86
piano, 13, 96,186—87, 277> 285; prepared,
Nathan, Isaac, 130,136,137-38 296
Nessinger, Mary, 238 piano reduction, piano/vocal score, 55-56,
Newlin, Dika, 227, 237 90, 95,112,128,197, 224, 225, 226, 278
North, Roger, 12,15, 35, 37,45 pitch: early Baroque, 5,13—14, 34, 41, 42;
notation: early Baroque, 6—7, 9, 44; late late Baroque, 65—66, 90
Baroque, 54, 61, 67; Classical, 93—94, Platonova, Yulia, 255
104,109; Heder 173,189,192; melodies, Playford, John, 33, 35, 46
196, 219; Sprechstimme, 242-43, 249-50; Porpora, Nicola, 97
new musk, 286, 289, 292, 293-94, 295, port de voix, 43, 46—47, 72, 90,139,141, 201
297, 301, 304, 306 portamento: early Baroque, 46; late Ba-
Nourrit, Adolphe, 195 roque, 64; Classical, 108; bel canto, 134,
138,139-43,156; Heder, 176,181,183-86,
Olenina d'Alheim, Maria, 251, 255-58, 272, 192; melodies, 209-12, 215, 220; Second
284 Viennese School, 244; 20th-century, 284,
opera: early Baroque, 42; late Baroque, 67, 285
71, 75, 91; Classical, 95,106,112; bel portare la voce, 46
canto, 127-28,130,134-36,138,143,149, Poulenc, Francis, 194,196,197, 201, 207,
156; German 168, 225, 229; Italian, 173, 209, 218, 219, 236,252, 273, 277, 282, 283
195, 200, 212, 259, 267, 268, 272; French, Le bal masqué, 207
203; Russian, 252, 265; Spanish, 266, 267; La dame de Monte Carlo, 207
English, 272, 274 "Tel jour, teile nuit," 219
operetta, 223, 234, 274 Price, Leontyne, 251, 283
organ, 12,13,15, 65, 69 Prokofiev, Sergei, 251, 260-63, 2.64, 265,
ornamentation: early Baroque, 5—7,10,12, 271, 272, 278, 281, 285
15,18,19, 21—32; early Baroque (Italian) The Ugly Duckling, 261
35-41; early Baroque (English) 45-49; Puccini, Giacomo, 18, 207
early Baroque (French) 51-52; late Ba- Purcell, Henry, 12, 33, 34, 35-38, 40, 275,
roque, 54, 58, 64-65, 67, 71-78; late 277-78, 282
Baroque (Italian), 79—84; late Baroque Dido and Aeneas, 274
(German), 87—90; late Baroque "The Fatal Hour," 37-38
(French), 91; Classical, 93, 94,102,103, "Fly Swift, Ye Hours," 40
108,109,110—25; bel canto, 128,134,136, Purgold-Molas, Alexandra, 255
137,139,143—59,199, 201—2; Heder, 171—
81,192; melodies, 201—2, 221 Quantz, Johann Joachim, 53, 54, 58, 64, 77,
79, 80, 83, 85-86, 87, 90, 92,102
Panzéra, Charles, 194, 200, 209, 218
Pasta, Giuditta, 133,135,136,151 Rachmaninov, Sergei, 251, 259-60, 261,
Patti, Adelina, 139,141,143 265, 272, 278, 285
Pears, Peter, 189, 251, 273-78 Rakowski, David, 289
perfect pitch. See absolute pitch Rameau, Jean-Philippe, 69, 86
period instruments. See instruments, period Le bergerfidel, 69—70, 88
354 Index
Randall, J. К., 288-89 romance, 194-95,199, 202, 212, 252

Ravel, Maurice, 194,195,197,198, 201, 202, Rossini, Gioacchino Antonio, 75,126,127,
203, 207, 211, 212, 215, 219, 229, 236, 256, 129,130,132,133,134,135,136,137,143,
257, 260, 265, 267, 281 144,147,149,151,159,195
Chansons madécasses, 198, 201 Aureliano in Palmira, 143
Don Quichotte, 198 Barber of Seville, 130,135
L'heure espagnole, 207 Cenerentola, 135,151
Histoires naturelles, 201, 207-8, 256 "L'orgia," 146
"Placet futile," 211-12, 219-20 Otello, 135
Shéhérazade, "Asie," 198 "La partenza," 156-57,158
recitative: early Baroque, 26, 32, 33,49; late "La promessa," 150
Baroque, 66-69, 86; Classical, 108, no, La regata veneziana, 141-42,148
112,125; bel canto, 138,144; melodies, 203; Semiramide, 143
in Wozzeck, 242; 20th-century, 258, 267 "Una voce poco fa," 143
recordings: early 20th-century, 138-39,141, William Tell, 135
187,191,192, 212; Heder, 176,183,189; Rostropovich, Mstislav, 264, 265, 276
melodies, 194,196, 201, 209, 210, 212, rubato: early Baroque, 26-28,40; Classical,
213, 215, 218, 219; Second Viennese 93,103,125; bel canto, 132,134,139,141,
School, 236-37, 238, 242, 247; 20th- 144,149,159; Heder, 176,191—92; melodies,
century, 251, 258, 285; 20th-century Rus- 212-20; Second Viennese School, 246-
sian, 258, 259, 260, 262-63, 264-65; 48, 249; 20th-century, 252, 284; 20th-
20th-century Spanish, 272; 20th-century century Russian, 259, 263; 20th-century
US, 281, 282, 283; new music, 286—87, Spanish, 267
288, 290, 295,304-5
Reich, Steve, 286, 288, 292, 299, 302, 303 Saint-Saens, Camille, 195, 201, 203
Music for 18 Musicians, 292 Santley, Charles, 130
Tehillim, 288, 292, 300 Scherchen, Hermann, 225, 234
rhythmic alteration: early Baroque, 6, 45; Schleifer, 80-82,119
late Baroque, 61-62; Classical, 104-6; Schoenberg, Arnold, 222, 223, 224-27, 228-
bel canto, 134; Heder, 189,192 29,230-44, 246-48, 249,274,279
rhythmic flexibility: bel canto, 132—34,149, Book of the Hanging Garden, 231, 244
159; Heder, 176,187-193; melodies, 215, Brettl-Lieder (Cabaret Songs), 225
219; Second Viennese School, 246, 247; Erwartung, 229
20th-century, 258, 259, 260, 273; new Die glückliche Hand, 231, 242
music, 297 Gurrelieder, 226, 236, 240, 242-43
rhythmic precision, 133,191,192, 202, 209, Jakobsleiter, 242
234, 247, 250, 297 Kol nidre, 240
Rimsky-Korsakov, Nicolay, 195, 252, 255, Moses undAron, 223, 240, 242—43, 248
258, 259, 265, 267 Ode to Napoleon, 240, 242-43
Mozart and Salieri, 259 Orchestral Songs, Op. 22, "Seraphita,"
Rodrigo, Joaquín, 251, 271-72, 285 232
Cuarto madrigales amatorios, 271 Pierrot lunaire, 222, 224, 229, 231, 233-
Romagnési, Antoine, 199, 201, 202, 203, 40, 241, 242, 247, 250, 270, 272
2O6, 212 Six Little Piano Pieces Op. 10, 247
Index 355
Survivor from Warsaw, 240, 242 Society for Private Musical Performance,
Von heute auf morgen, 224, 248 228-29
Schröder-Devrient, Wilhelmine, 167-68 Sonnleithner, Leopold von, 167,172,187
Schubert, Franz, 160,161,162,163,165-66, Sontag, Henriette, 143
167,172,174-79,186,187-88,189-91, Souzay, Gerard, 197
192,195, 225, 231, 257, 273, 277 Sperry, Paul, 286, 290, 299, 301, 302—3
"An die Musik," 184 Spohr, Louis, 160,162,179,181
"An die Nachtigall," 176,178 "Nachgefuhl," 179,182
"An Emma," 177 Sprechstimme, 222, 233-43, 249-50, 295
"Ave Maria," 175—76,178,191 sprezzatura, 26—27,134
"Danksagung an den Bach," 180 Stanislavsky, Konstantin, 253, 259
"Der Einsame," 178,180 Steber, Eleanor, 251, 282-83
"Erlkönig," 163,169 Steuermann, Edward, 227, 228, 229, 234,
"Die Forelle," 163 237-38, 247-48
"Frühlingsglaube," 175-76,178,190-91 Stiedry-Wagner, Erika, 229, 236-37, 238,
"Frühlingstraum," 174,177 241, 247
"Halt," 175,178,180 Stockhausen, Julius, 161,162,163,166
Der Hirt auf dem Felsen, 176,179 Strauss, Richard, 160,162,164,171,185-86,
"Lachen und Weinen," 176-77 188-89, 249, 267
"Des Mädschens Klage," 189-91 Brentano Lieder, 171
"Rastlose Liebe," 174,177 Four Last Songs, 164
Die Schöne Müllerin, 162,166,172 "Wie selten wir geheim sie halten," 185
"Suleika II," 169-70,175 Stravinsky, Igor, 195, 202, 233, 261, 279, 281
"Tränenregen," 176,180
"Wasserflut," 189-91 Talma, Louise, 303
Winterreise, 161,162,166 Tchaikovsky, Pyotr, 252, 253, 260
Schumann, Clara, 161,162,163,165,167, The Nutcracker, 298
172-73,188 Tear, Robert, 275, 277
Schumann, Elizabeth, 171 tempo: early Baroque, 5, 6, 26, 40, 43, 45,
Schumann, Robert, 160,161,162,163,165, 51; late Baroque, 54, 55, 56-58, 61-62,
166,167,186,189,192, 224, 225 77, 84, 90; Classical, 93, 95, 98,100-104,
Dichterliebe, 165,168 108,124-25; bel canto, 132-34,139,141,
Frauenliebe und Leben, 166 159; Heder, 178,187-89,191-92; melo-
"Ich kann's nicht fassen, nicht glauben," dies, 196, 212—20; Second Viennese
184 School, 246—48, 249; 20th-century, 252;
Liederkreis, 165 20th-century Russian, 257-58, 260,
Schwartz, Francis, 295 263-64; 20th-century English, 275, 276;
Senesino, Francesco Bernardi, 71 new music, 291, 293, 303
Serkín, Rudolf, 227, 228 text: early Baroque, 9,16,18-19, 26-27,
Shelton, Lucy, 298 32-33, 35, 48, 50-51; late Baroque, 56, 58,
Shostakovich, Dmitri, 251, 252, 263—65, 62, 67, 69, 84, 90; Classical, 97, 98,102,
266, 285 107,108, no, 112,115,125; bel canto, 129,
From Jewish Folk Poetry, 264 130-131,136,143,159; Heder, 169-70,171,
Singher, Martial, 198 174,184,186,189,191; melodies, 196,199,
356 Index
text (continued] Vaughan Williams, Ralph, 272, 274

201, 202-3, 2I5; Second Viennese Velluti, Giovanni Battista, 134,143
School, 222, 229-33, 2445 20th-century, Verdi, Giuseppe, 18,136,144, 252
252; 20th-century Russian, 253, 257, 261, vibrato: early Baroque, 14—18, 21, 25, 47, 51;
264; 20th-century English, 274; new late Baroque, 64-65, 87-88, 90; Classi-
music, 287, 294, 297, 299-302, 305 cal, 107; bel canto, 134,137-39, H1» T5^5
Teyte, Maggie, 194, 215 Heder, 181,183-186,187,192; melodies,
theorbo, 12 212; Second Viennese School, 237, 244;
Thompson, George, 94,102 20th-century English, 278; 20th-century
thoroughbass, 9-11,12. See also continuo; US, 281; 20th-century, 285; new music,
figured bass 288, 299-300
time signatures: early Baroque, 9; Classical, viol, n, 12, 35,45, 69, 87
100; Heder, 189 Vishnevskaya, Galina, 251, 264, 265, 276
Tosi, Pier Francesco, 15,18, 20, 22, 26, 54, vocal flexibility, 22,133,137,147
67, 71, 72, 74, 75, 77, 78, 79,103,191 vocal ranges, 33—34,42—43, 65—66,106,
Tourte bow, 96—97, 98 288, 290
transposition: early Baroque, 13; late vocal registers, 20-21, 66,106,108,126,135,
Baroque, 54, 55, 65, 75, 87, 91; Classical, 136, 237, 242
106; bel canto, 130,135; Heder, 164-66, vocal technique: and low pitch, 14,17—18;
192; melodies, 196—99; Second Viennese bel canto, 136—37,156,166—67, J68,169;
School, 226; 20th-century Spanish, 271 German, 166-69, 222> 24°> 242; French,
tremblement (vibrato), 14,15, 47, 87 203, 209; extended, 287, 294-95, 303
tremolo (vibrato), 14,15,47, 65,138,181, 212 Vogl, Johann Michael, 166,167,172,176,
trill: early Baroque, 15, 21, 22—25, 35; late 273
Baroque, 67, 74, 75, 77, 82-84, 87-90, voice types, 55,135,197,198
115; Classical, 112,115-120,125; bel canto,
147-149,159; Heder, 171,179,192; Wagner, Richard, 18,126,160,161,162,168,
melodies, 201. See also trillo 173,181,195, 224, 225, 229, 267, 273
trillo, 14,15, 2i, 25, 35, 47. See also tr'ál·, Der fliegender Holländer, 168
vibrato Das Reingold, 169
Turina, Joaquín, 271 Tannhäuser, 168
Turk, Daniel Gottlob, 93, 98,102,103 Walker, George, 303
turn: classical, 112,119,120,125; bel canto, Weber, Aloysia, 107,124
147-149,159; Heder, 171,173,179,182,192 Weber, Carl Maria von, 172,188
twelve-tone music, 222, 224, 229, 233, 248 Webern, Anton, 222, 223, 224, 225, 226-27,
Twenty-four Italian Songs and Arias, 7, 28, 228, 229-30, 242-43, 244-47, 249
32,72 Das Augenlicht, 230
Drei LiederOp. 25, 246
Upshaw, Dawn, 289 Sechs Lieder, Op. 14, "Die Sonne," 245
Westergaard, Peter, 301—02
Vaccai, Nicola, 126,137,139—40,141—43, Wolf, Hugo, 160,164,168-69, i73> 181,189,
147,149,150 191, 225, 249
Valente, Benita, 286, 288, 289, 298, 300
Vasnier, Marie, 197 Zehme, Albertine, 233-34, 242

Singing in Style - Elliott, Martha PDF

Uploaded by

Copyright:

Available Formats

Singing in Style - Elliott, Martha PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Singing in Style - Elliott, Martha PDF

Uploaded by

Copyright:

Available Formats

This page intentionally left blank

Yale University Press

New Haven and London

The Library of Congress has cataloged the hardcover edition as follows:

ISBN 978-0-300-13632-6 (pbk.: alk. paper)

1 The Early Baroque, 5

2 The Late Baroque, 53

3 The Classical Era, 92

4 Italian Bel Canto, 126

5 German Lieder, 160

6 French Melodies, 194

7 Second Viennese School, 222

8 Early Twentieth-Century Nationalism, 251

9 Working with Living Composers, 286

This book introduces the issues of historical performance practice as

This chapter presents an overview of seventeenth-century vocal music

Scores and Editions

in the medium-high edition or С minor in the medium-low edition, bar lines

an organ is certainly appropriate, though you must decide whether to work

m. 6. Here is a situation where a tone with little or no vibrato can be used as an

Monteverdi's sense of "powerful and sustained" was quite different from

Words and Music

In re-creating historically informed performances, historical pronunciation is

Voice Types and Registers

Ornamentation in the seventeenth century falls into two general categories:

DISPOSITIONE AND DIVISIONS

ESCLAMAZIONEAND MESSA DIVOCE

Example 1.8. Continued

Example 1.8. Continued

RECITATIVE AND ARIA

Voice Types and Ranges

Before the Restoration, women were not a part of professional performances in

Sources: Manchester. Public Library, Watson Collection, ms 832 Vu 51 [ca 1660]

Example 1.9. Graces from later seventeenth-century English sources, reproduced by

desired ornamentation, is described in the memoirs of an actor, Colly Gibber:

mathematical relationship to the original rhythm. This seems to indicate a flex-

of the bad taste of an amateur singer or her teacher, or inducements to singers

Voice Types and Ranges

Another particular characteristic of French Baroque music is inégalité, a tradi-

This is not to be confused with portare la voce or portamento, whose seven-

This term can be understood to be comparable to the Italian tremolo, which, as

preserve the intelligibility of the text. Bacilly generously encourages, diminu-

Certain consonants deserve special consideration and attention, particularly

penultimate syllables of feminine words: imaginable, cruelle

Syllables that are usually short:

penultimate syllables of masculine words: secret, espoir

Trends in ornamentation varied dramatically both from nationality to na-

Scores and Notation

As discussed in chapter i, Baroque composers expected that their performers

The pace of a composition, which is usually indicated by several well-known Italian

In Music as Speech, Harnoncourt explains that Baroque music has a discernible

mony and dissonance to create different levels of emphasis and relaxation

likely be softened to accompany simultaneous triplets, since it was unusual to

Vibrato and Breath

Vibrato in the early eighteenth century was still considered an ornament, to be

The use of vibrato can be thought of as part of a varied palette of articula-

Example 2.4. Agrícola: appoggiaturas in recitatives from Anleitung zur Singkunst ( i j ^ j ) ,

and be accompanied by the sustaining sound of the organ. Accompanied