nside

eauii

co/ioe ua/io7i s uu/n pezmzmezd

,-e attempt to play music wnn ine
es and instruments of its era—
.mmonly referred to as the early
music movement— has become im-
mensely popular in recent years. But
the movement has generated as much
controversy as it has best-selling

records, not only about the merits of

its results, but also about the validity

of its approach. To what degree can we

recreate long-lost performing styles?

How important are historical period

instruments for the performance of a

piece? Why should musicians bother
with historical information? Are they
sacrificing art to scholarship?

Now, in Inside Early Music, Bernard

D. Sherman has invited twenty-three of

the leading practitioners to speak out

for early music- T> k.
Ci.
o
about their passion OQ cd
why they are attracted to a historically 1
=3
oriented approach and how it shapes C3 cr
CO
their work. Readers listen in on con- CQ >
versations with conductors John Eliot CL. (D
Roger Norrington, William ^S^ a.
Gardiner,
5=> o
and Philippe Herreweghe; choral O
Christie,

director Peter Phillips of the Tallis g

Scholars; vocalists Susan Hellauer of
Anonymous 4 and Barbara Thornton of

Sequentia; fortepianists Robert Levin

and Malcolm Bilson; harpsichordist

Gustav Leonhardt; cellist Anner Bylsma;

and many others. The book is divided

into musical eras— Medieval, Renais-

sance, Baroque, and Classical and

Romantic — with each interview focus-

ing on particular composers or styles,

touching on heated topics such as how
histohcal evidence should be used,
why period instruments might matter,
and what "authenticity" is. Whether
debating how to perform Monteverdi's
madrigals or comparing Andrew
Lawrence-King's Renaissance harp
improvisations to jazz, the performers
 Digitized by the Internet Archive
in 2010

http://www.archive.org/details/insideearlymusicOOsher
INSIDE EARLY MUSIC
NO LOGGER THE PROPERTY
OFBC JCUBRAR^-

INSIDE
EARLY MUSIC
-sn oQO^
Conversations with Performers

BERNARD D. SHERMAN

New York Oxford

Oxford University Press
1997
 Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota Bombay
Buenos Aires Calcutta Cape Town Dar es Salaam
Delhi Florence Hong Kong Istanbul Karachi
Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore
Taipei Tokyo Toronto

and associated companies in

Berlin Ibadan

Copyright © 1997 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.,

198 Madison Avenue, New York, New York 10016

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Sherman, Bernard D.
Inside early music : conversations with performers /

Bernard D. Sherman.
p. cm. Includes bibliographical references (p. ) and index.
ISBN 0-19-509708-4
1. Performance practice (Music). 2. Style, Musical.
3. Musicians —Interviews. I. Title.

ML457.S52 1997 781.4'3— DC20 96-6341

135798642
Printed in the United States of America
on acid-free paper
To my family
 CONTENTS

Preface and Acknowledgments, vii

Introduction: An Atmosphere of Controversy, 3

I. MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS," 23

1. A Different Sense of Time: Marcel Peres on Plainchant, 25

Selected Discography, 41 For Further Reading, 42
2. You Can't Sing a Footnote: Susan Hellauer on Performing
Medieval Music, 43
Selected Discography, 52 For Further Reading, 53
3. Vox Feminae: Barbara Thornton on Hildegard of Bingen, 54
Selected Discography, 69 For Further Reading, 70
4. The Colonizing Ear: Christopher Page on Medieval Music, 71
Selected Discography, 86 For Further Reading, 87
Medieval Music, Plainchant,
Postscript:
and "Otherness," 88

II. THE RENAISSANCE, OXBRIDGE, AND ITALY, 97

5. There Is No Such Thing as a Norm: Paul Hillier on Renaissance

Sacred Music, 100
Selected Discography, 114 For Further Reading, 115
6. Other Kinds of Beauty: Peter Phillips on the TalHs Scholars and
117
Palestrina,
Selected Discography, 130 For Further Reading, 132
7. Singing Like a Native: Alan Curtis, Rinaldo Alessandrini, and
Anthony Rooley on Monteverdi, 133
Selected Discography, 153 For Further Reading, 155
Postscript: Nationalism and Early Music, 156
8. Emotional Logic: Andrew Lawrence-King on Renaissance
Instrumental Music and Improvisation, 157
Selected Discography, 169 For Further Reading, 170
 III. THE BAROQUE, 171

9. Consistent Inconsistencies: John Butt on Bach, 173

Selected Discography, 190 For Further Reading, 191

10. "One Should Not Make a Rule": Gustav Leonhardt on Baroque

Keyboard Playing, 193
Selected Discography, 204 For Further Reading, 206

11. Aladdin's Lamp: Anner Bylsma on the Cello (and Vivaldi, and
Brahms, 207
Selected Discography, 222 For Further Reading, 223

12. Beyond the Beautiful Pearl: Julianne Baird on Baroque Singing, 225
Selected Discography, 241 For Further Reading, 242

13. You Can Never Be Right for All Time: Nicholas McGegan on
Handel, 243
Selected Discography, 255 For Further Reading, 256

14. At Home with the Idiom: William Christie on the French

Baroque, 257
Selected Discography, 272 For Further Reading, 274

15. Triple Counterpoint: Jeffrey Thomas, Philippe Herreweghe, and John

Butt on Singing Bach, 275
Selected Discography, 291 For Further Reading, 293

IV. CLASSICAL AND ROMANTIC, 295

16. Restoring Ingredients: Malcolm Bilson on the Fortepiano, 297

Selected Discography, 312 For Further Reading, 313

17. Speaking Mozart's Lingo: Robert Levin on Mozart and

Improvisation, 315
Selected Discography, 334 For Further Reading, 336

18. Taking Music Off the Pedestal: Roger Norrington on Beethoven, 339
Selected Discography, 358 For Further Reading, 359
Postscript: "Classical" and "Romantic"
Performance Practice in Beethoven, 360

19. Reviving Idiosyncrasies: John Eliot Gardiner on Berlioz and

Brahms, 364
Selected Discography, 374 For Further Reading, 376

20. Reinventing Wheels: Joshua Rifkin on Interpretation and

Rhetoric, 378
Selected Discography, 389 For Further Reading, 390

Epilogue, 391

Index, 403
 PREFACE AND ACKNOWLEDGMENTS

Many readers will wonder why I chose these specific artists to interview —and,
more to the point, why I didn't choose others. Some decisions were obvious:
anyone would include pioneers like Anner Bylsma, William Christie, or Gustav
Leonhardt. But where, you ask, are . . . ?

One was space: to do justice to all the branches of the

serious constraint
historical-performance scene would take a book at least twice this size. I would
have been happy to write that book (my editor will attest to that), but you
would have had to pay twice as much for it. Some interesting interviews were
sacrificed to reduce the book's length. I was distressed that we couldn't include
the fascinating interview that Monica Huggett granted me, because this out-
standing artist has not received all the recognition she deserves. But her inter-
view had too much overlap with Anner Bylsma's; since hers was published else-

where,' his was kept. The gifted soprano Judith Nelson and the distinguished
harpsichordist Christoph Rousset had to be left out, with just as much regret,

for similar reasons.

Other factors entered into my choice as well. One —not the least of them
was sheer accident. Had I searched internationally for a Bach expert, I would
probably not have chosen the as yet little-known scholar and keyboardist John
Butt. But after spending three fascinating hours with him, I believed that a more

insightful interviewee would have been impossible to find. The "sheer accident"
in this case was geographic: Butt was a neighbor of mine. And geography partly
explains why there are too few Continental musicians in the book.
The choices also reflect, inevitably, the accident of my tastes and interests,

though I did try to go beyond that. And some interviews weren't possible. Mary
Springfels was too busy; Jordi Savall was disinclined; Nikolaus Harnoncourt
won't do telephone interviews, and I couldn't afford air fare to Austria. The
exceedingly busy Christopher Hogwood had just invested a great deal of time
in interviews that were never published, and I didn't yet have a publisher at the
time I approached him. I would have liked to include more women. I could
argue, accurately, that even in the 1990s, men dominate the early-music field;
but several serious attempts I made to do more interviews with women (in ad-

1. Bernard D. Sherman, "Monica Huggett," Strings 10 (March/April 1996), pp. 54-61.

X PREFACE AND ACKNOWLEDGMENTS

dition to the ones I've mentioned) fell through. I also wanted more Continen-
tal musicians, such as Reinhard Goebel and Eduardo Lopez Banzo; that, too,

just didn't turn out to be feasible.

For all these reasons, a number of artists I admire enormously were not even
approached. And many important areas are omitted. There are no instrument
makers, for example. Perhaps a future volume will fill some of the gaps, but
for now I hope this one gives a reasonable sense of the current scene and its

issues. (This book, by the way, does not attempt to give a history of the his-

torical-performance movement; a good one is Harry Haskell's The Early Music

Revival ILondon: Thames and Hudson, 1988].)
Instead of just giving my own opinions, my discographic notes often quote
other critics. If I were reading this book, I would prefer to get an idea of the
critical consensus, instead of just one listener's opinions (although some of
those are present as well). Also, I rarely attempt to provide a full critique of an
artist's recordings, but instead suggest what I think are high points and good
places to begin investigating.
Performers are now playing Mahler with historical styles and instruments,
so the term "early music" is no longer quite apt. For convenience, though, I

will use it as well as "historical performance."

In preparing this book, I have been a great borrower; my voracious need for
books and scores was fed by such uncomplaining borrowees as Belle Bulwin-
kle, Jonathan Harris, Julie Jeffrey, James Meredith, Zoe Vandermeer, and Rick

Weller. On occasion, Joseph Spencer and Anna Shtutina lent me CDs. My

thanks to all of them for their generous loans, which, I am relieved to say, have

all been returned.

I can't see how the music library at the University of California could be
surpassed. If something I needed was unavailable there, I could usually find it

at the excellent public libraries in Berkeley and Oakland and the private one at
Mills College (late in the book's preparation, the University of Iowa library was
very helpful). Julie Jeffrey and Jim Bates provided expert last-minute reference
aid.Many record companies helpfully sent me review copies of CDs.
Many experts were kind enough to answer my queries about specific issues.
They included Wye Jamison Allanbrook, George Barth, Katherine Bergeron,
Jane Boothroyd, Alfred Brendel, John Butt, Mary Cyr, Laurence Dreyfus, David
Fallows, Donald Greig, Ralph HoUoway, D. Kern Holoman, Chris Hunter,
David Lasocki, Carol Lems-Dworkin, Hugh Macdonald, Daniel Melamed,
William Meredith, Donald Mintz, Herbert Myers, Marc Perlman, Lawrence
Rosenwald, the late Max Rudolf, Sally Sanford, Howard Schott, David Schu-
lenberg, and Richard Taruskin.
Finally, let me thank all those who took time to comment on the book's
contents. To the interviewees themselves, of course, my thanks for putting what
was in some cases considerable time and energy into their interviews. Thanks
 PREFACE AND ACKNOWLEDGMENTS XI

also to their agents or concert contacts for helping me gain access. Reviewers
at Oxford and one other press gave me valuable and much-appreciated feed-
back; I can thank Oxford's reviewers by name — Fallows, Taruskin, and Tess
Knighton — as they all revealed their identities to me. I didn't have the chutz-
pah to ask anyone else to read the whole manuscript, but many people took
valuable time (something they don't have a lot of) to look at and comment on
chapters or subsections of them. I'm very grateful for their feedback. These
readers included, among musicians and musicologists, Allanbrook, Barth,
Butt, Bulwinkle, Michelle Dulak, Virginia Hancock, Tamara Loring, William
S. Newman, Perlman, Vandermeer, and Robert Winter; among non-musico-
logists, Bryan Aubrey, Steven Bensinger, Gerald Geer, Beverly Hill, Leon
Honore, Christopher Minkowski, James Obertino, Amy Sherman, Brian Stains,

and Elizabeth Van Schoick. Stains and Van Schoick also served as sounding
boards whenever asked, and were so perceptive that I asked more than once.
Several non-musicologists read the Postscript to the Christopher Page interview
and gave valuable input: Silvine Marbury Farnell, William Jankowiak, James
Karpen, Lee Kirkpatrick, Steven Pinker, Dane Waterman, and Robert Wright.
Harmon R. Holcomb III wrote a detailed critique of an early version of that
Postscript; I am indebted to him for that, and for having (in the process) sug-
gested the angle taken in the final version. Jonathan Harris not only gave me
helpful feedback but, along with Myrna Melgar, devoted an extraordinary
amount of time to helping prepare and then translate questions to be sent to
the elusive Jordi Savall (who, through no fault of theirs, never responded).

Elliot Hurwitt and Jerome Weber generously gave discographic input. Joel
F.

Flegler kindly granted permission to use material in the Jeffrey Thomas inter-
view that had previously appeared in Fanfare. I'm grateful to Oxford Univer-
sity Press for honoring my request that Eric Van Tassel be asked to copyedit

the book, and even more grateful to him for agreeing to do so, and for doing
so expert, thorough, and helpful a job of it. I'm also grateful to Kimberly
Torre-Tasso of Oxford University Press for being so helpful throughout the pro-
duction process. Finally, special thanks to my editor, Soo Mee Kwon, for her

enthusiasm for this project and for her unfailing patience and skill.

Berkeley, California/Fairfield, Iowa

1995/1996
INSIDE EARLY MUSIC
 INTRODUCTION:
AN ATMOSPHERE OF CONTROVERSY
Torniamo all'antico: sara un progresso [Let us return to
old times: that will be progress] — Giuseppe Verdi

The "early-music revival" has been around for a whole century, but it was only
in the 1980s that its recordings suddenly began to top the Billboard charts. A
decade later those who called "historical performance" a fad have been proved
just as wrong as those who called it a revolution. It shows no signs of disap-
pearing. But even now, it can still make us wonder, "Why do they play like
that?"
That question inspired this book. Musicologists have published whole li-

braries of historical evidence, but few performers have put their experiences

into print. Yet they have special insights to share. Charles Rosen says that "mu-
sicology is for musicians, what ornithology is to the birds."' While some of my
interviewees do say a good deal about musicology, the musicologists can't tell

us how it feels to fly.

The relationship between musicologists and early-music performers might

seem to be a simple matter: musicologists do research, performers put the re-

search into practice. In fact the relationship is complex,^ because the two dis-

ciplines make an uneasy match. Music history tries to restrict itself to what is
supported by data, but performance suffocates under that restriction. Music
historians try to find out what happened in the past, performers try to make
something happen now. In some ways, the purposes conflict: as Rosen says,

"Paradoxically, in so far as the purpose of a performance of a Mozart concerto

is reconstruction of eighteenth-century practice rather than pleasure or dra-
matic effect, just so far does it differ from an actual performance by Mozart."^

1. Rosen, The Frontiers of Meaning (New York: Hill and Wang, 1995), p. 72.
2. Discussed by Joseph Kerman, Contemplating Music (Cambridge, Mass.: Harvard Uni-
versity Press, 1985), chap. 6;Laurence Dreyfus, "Early Music Defended Against Its Devotees,"
The Musical Quarterly 69 (1983), pp. 297-322; and Richard Taruskin, "On Letting the Music
Speak for Itself," reprinted in his Text and Act (New York: Oxford University Press, 1995),
first pubhshed in the Journal of Musicology 1/3 (1982), pp. 338-49.

3. Rosen, The Classical Style (New York: Norton, 1971), p. 107.

4 INTRODUCTION

But a performance can, of course, try both to reconstruct early practice and
to give pleasure or dramatic effect. The results can be surprisingly vital. Still,

the tension between the goals may explain part of why the early-music move-
ment has, as Joseph Kerman says, "always flourished in an atmosphere of mul-
tiple controversy.'"* This book presents insiders' views of many of the contro-
versies —what they are about, and why they might matter. Without such views,
we can't really understand why these artists play as they do.
Some of the controversies arise within the realm of musicology, but others
reflect the tension between scholarship and art. The crux of this book can be
expressed in a question: How can you use historical information to enliven
modern performance? Answers to that question fall on a spectrum from ig-

noring the evidence to following it to the letter. Those two responses, and oth-
ers less extreme, underlie the most obvious of the controversies —what Kerman
called "disputes over turf."

"Manichean Struggles ": The Turf Wars

The turf wars have usually pitted mainstream musicians against the history-
minded upstarts who encroach on their territory, as in, "Now they're playing
Brahms?" or, from the other side, "How can they still play Bach on the piano?"
If the "war" image is extravagant, it does at least suggest how strong the emo-

tions could become on both sides. When historical performers of the 1970s and
'80s compared using old instruments to cleaning a dirt-encrusted Rembrandt,
it was more than an analogy; it implied that the ignorant mainstreamers were

trashing the classics.^ The historicist pronouncements often involved not just
art, but also morality. Bernard Holland recalls "fierce Manichean struggles of

good versus evii."^ I remember an early-music advocate describing her col-

leagues' work as "the responsible performance of Baroque music." Another, ex-
pressing the zeal many of his colleagues felt in the 1970s, argued that musi-

4. In "The Early Music Debate" (an edited transcript of a symposium featuring Kerman,
Laurence Dreyfus, Joshua Kosman, John Rockwell, Ellen Rosand, Richard Taruskin, and
Nicholas McGegan), Journal of Musicology 10 (Winter 1992), pp. 113-30.
5. It was, moreover, a problematic analogy. It assumes that underneath all the accumulated
grime there is an authentic musical "original" waiting to be restored to pristine condition.
This conceives of a piece of music as a timeless thing —a concept that raises enormous prob-
lems when applied to an active, temporal process like music, as several writers have discussed,
such as Lydia Goehr in The Imaginary Museum of Musical Works (Oxford University Press,

1992). Another problem is that the analogy, in comparing tone color to the pigment in paint-
ings, exaggerates the importance of tone color in pre-nineteenth-century music; this is dis-

cussed in the Gardiner interview, Chapter 19. However, the charge that historical performers
put an anachronistic emphasis on tone color ignores the fact that many of them justify the

use of period instruments not in terms of sound, but in terms of the clues that the instruments
give performers about style and articulation. But the analogy itself misses this point.
6. Bernard Holland, "A Streak in the Heavens Has Become a Straggler," The New York
Times, Sunday Arts and Leisure section, 31 October 1993, p. 31.
 INTRODUCTION 5

cians are [my italics] "under an absolute injunction to try to find out all that

can be known about the performance traditions and the sound-world of any
piece that is to be performed, and to try to duplicate these as faithfully as pos-

sible." So much for Horowitz, Gould, and Rachmaninoff, who violated that in-
junction without apology, and for other mainstream performers, who also vio-
late it (if less audaciously). It's not surprising that mainstreamers often accused
the historical performers of pedantry —of "restraining any and all of the inter-
"^
preter's natural urges.

As that shows, mainstreamer charges could be belligerent and moralistic in

their own way. The historicists were accused of amateurism, and — to turn the
tables —of trashing the classics. Although historical performers did sometimes
take speed, lightness, or inflection to the point of mannerism, the critiques went
beyond that. One critic wrote of the unfamiliar instruments, "it is impossible
to listen without discomfort, nausea, without clenching one's teeth," and urged
that performers "guilty of musical outrages" be given prison sentences.
**

The turf wars sometimes seem a textbook case of "ingroup and outgroup"
psychology.' Psychologists have found that people in an ingroup tend to see the
outgroup in terms of simplistic stereotypes —and while historical performers are

a varied group, you wouldn't know it from many of their critics. Similarly, the

"mainstream" has eddies that historicists sometimes ignore.'" Also, ingroups

tend to see themselves as virtuous but beleaguered, and the outgroup as malev-
olent and powerful. Both camps have been known to describe each other in
such terms: I recall a review of the Cecilia Bartoli/June Anderson recording of
Pergolesi's Stabat Mater, which praised the two popular divas for bravely re-
claiming the piece from the early-music mafia.
Of course, both sides also make substantial points. Regarding the main-
stream complaints, even the most sympathetic observer must admit that his-

torical performance did have its phases of both amateurism and pedantry. Both
phases, however, were necessary. Regarding amateurism, it has often been hard
to make a living at historical performance, so those who tried to master it had
to earn their keep doing something else. Besides, even in supportive circum-
stances it takes a while to master an instrument. In 1963, Nikolaus Harnon-

7.Donald Vroon, American Record Guide 56 (MarchyApril 1993), p. 220.

8. Gerard Zwang,A contre-bruit (Paris, 1977). The translation comes from Laurence Drey-
fus's "Early Music Defended Against Its Devotees."

9. Marilynn B. Brewer and Roderick M. Kramer, "The Psychology of Intergroup Attitudes

and Behavior," in Annual Review of Psychology, vol. 36, ed. Mark R. Rosenzweig and Lyman
W. Porter (Palo Alto: Annual Reviews, 1985), pp. 219-43.
10. Hermann Danuser argues that there are actually three modes of performance in art
music today: the "traditional" mode, which includes what I call "mainstream" musicians; the
"actualizing" mode, which interprets old music in light of modern compositional styles, and
includes such artists as Glenn Gould and Pierre Boulez; and the "historical-reconstructive,"
which includes the artists interviewed in this book. See Danuser's Mustkaltsche Interpretation
(Laaber: Laaber Verlag, 1992).
6 INTRODUCTION

court's fledgling Concentus Musicus recorded the Brandenburg Concert! in ten-

second takes, because the period winds could stay in tune only that long;'^ but
in 1993, John Eliot Gardiner's Orchestre Revolutionnaire et Romantique
recorded three Beethoven symphonies in live concerts, and the winds stayed in

tune throughout. As for pedantry, it too was an inevitable phase. It takes time
to unlearn old habits, and a longer time before new practices become ingrained.
In the gap between the old habits and the new ones, musicians play pedanti-
cally. In addition, sometimes the pedantry was deliberate; a program note circa
1950 said that "Early music was a highly aristocratic art and restraint governed
even the display of emotion."'^ In 1994, however, Clifford Bartlett writes of
British Baroque playing that "what was at first a fairly stiff, somewhat puri-
tanical approach has become much more free";'^ and that applies elsewhere
(though not universally —medieval performers, who often have links with folk
music, sounded anything but puritanical in the 1950s, and the old Harnoncourt
Baroque recordings were not exactly stiff either). In any case, I don't find much
evidence for pedantry in my interviewees' playing, or in their words. Several of
them praise historically uninformed moderns from earlier in our century. Anner
Bylsma admires Fritz Kreisler; William Christie loves Sir Thomas Beecham's
recordings, and doesn't mind at all that Beecham detested musicology and in-
sisted on using corrupt editions. These interviewees have no fetish about his-

toricism, but they do object to an undiscriminating obsession with smooth,

powerful surfaces, a fault many music critics observe in "competition-winner"
mainstream playing today.
In general, the turf wars appear to be subsiding in favor of what Alfred
Brendel calls "true cross-fertilization.'"'* Period string sections, once famous for
astringency, are now sometimes praised and even criticized for sweetness of
tone'^ (which, according to Michelle Dulak, proves that the astringency was
not determined by the instruments but instead reflected the performers'
choices). ^^ Wendy Carlos uses historical tunings in her electronically synthe-
sized Bach, some mainstream conductors explore period styles, ^^ and many

According to the producer, Wolf Erichson, in James Keller, "Wolf Erichson," Histori-
11.
calPerformance 6 (Spring 1993), p. 32.
12. Erwin Bodky, program notes for the Cambridge Society for Early Music, quoted in
Harry Haskell's The Early Music Revival (London: Thames and Hudson, 1988), p. 178.
and Others," Early Music 22 (August 1994), p. 521.
13. Clifford Bartlett, "Pandolfi Mealli
Music Sounded Out (London: Robson, 1990, and New York: Farrar,
14. Alfred Brendel,
Straus and Giroux, 1991), p. 224. See also Michelle Dulak, "The Quiet Metamorphosis of
'Early Music,'" Repercussions 2 (Fall 1993), pp. 31-61.
15. Stanley Sadie praises Trevor Pinnock's strings for their sweetness in Gramophone 72
(January 1995), p. 50; Raymond Knapp, in an informative review of Norrington's Brahms
First Symphony, praises the strings' sweet tone but faults their "reluctance to abandon" that
tone in some passages {American Brahms Society Newsletter 11 [Spring 1992], pp. 4-7).
16. Dulak, "The Quiet Metamorphosis," p. 39.
17. For example, Yehudi Menuhin, David Zinman, and Michael Morgan; Sir Charles
Mackerras and Sir Simon Rattle have even conducted period-instrument groups.
 INTRODUCTION 7

players in both camps cross over regularly. Gardiner conducts the Vienna Phil-
harmonic, and Yo-Yo Ma, who has performed the Beethoven Triple Concerto
with Itzhak Perlman, Daniel Barenboim, and the Berlin Philharmonic, has also
played it with Ernst Kovacic, Robert Levin, and Roger Norrington on historic
instruments (Ma using a gut-strung cello). Even eminent singers cross over: Bar-
toli has recorded Mozart with Christopher Hogwood (using period instru-
ments), and Dame Joan Sutherland is featured in Hogwood's period recording
of Handel's Athalia. When told that the recording would involve old instru-
ments, she is said to have replied, "That's all right, I'm a bit of an old instru-
ment myself."
Listening tastes, too, seem less segregated. Today, writes James Jolly, the ed-
itor ofGramophone, "a Mozart opera almost has to be performed on period
instruments to make reasonable headway in Ithe CD] catalogue.""* He notes
that recent exceptions, by Harnoncourt and Sir Charles Mackerras, have ex-
plored period styles, if not period instruments. At the same time, a few per-
formers of pre-mainstream repertoire —the monks of Silos, Anonymous 4, the
Tallis Scholars, Jordi Savall, and some others —have reached audiences far
larger than the usual medieval/Renaissance music subculture. These performers
may even cross over to contemporary classical music (as in Paul Hillier's work
with Arvo Part, or Fretwork's commissioning of new music by Gavin Bryars)
or, to everyone's surprise, to popular musics (as in the Hilliard Ensemble's disc
with the saxophonist Jan Garbarek).''*
Nonetheless, it would be mistaken to declare either a victory for early music
or a cease-fire in the turf wars. The later interviews in this book contain plenty
of salvos against the mainstream. On the other side, some critics still speak of
"the authenticity craze['s] . . . arid Imusic making,] the often out-of-tune in-
struments Isounding] ghastly."^" As for performers, in a recent interview the vi-

olinist Pinchas Zukerman said that historical performance is "asinine

STUFF ... a complete and absolute farce * * * AWFUL," and adds, "Nobody
wants to hear that stuff. I don't."-'

The Authenticity Debates: What Is Possible, What Desirablef

As large audiences began to want to hear that stuff and historical performers
began to make money, new controversies arose, often within the historical
ranks. Beneath the disputed turf lay certain assumptions; as the historicists
gained ground, they began to question those assumptions. For example, many

18. Editorial, Gramophone 71 (April 1994), p. 1.

19. Officium (ECM New Series445 369).

20. Sedgwick Clark, in "North American Retrospect," Gramophone 70 (May 1993),
North American edition, p. A3.
21. David K. Nelson, "An Interview with Pinchas Zukerman," Fanfare 13 (March/April
1990), p. 38.
8 INTRODUCTION

flinched from their critics' mockery of the use of the term "authenticity" to
mean historical accuracy, so that the term has become virtually taboo among
historicists (unless it's enclosed in ironic quotation marks; though some artists

who sneer at the term don't seem to be able to stop their record companies
from using it). Other disagreements (often still internecine) continue, reflecting
the tensions between art and scholarship — and, some would say, between artis-

tic idealism and marketplace pragmatism. And these disagreements, like the his-
toricists' disputes with the mainstream over turf, reflect a further tension: be-

tween the past, when the music was written, and the present, when we're
playing and hearing it.

For example, many doubt that reviving old playing styles is entirely possi-

ble today. The goal of historical performers is something like playing Strauss
waltzes with the special rhythmic lilt that makes the oom-pah-pah uniquely Vi-
ennese; but the equivalent of echt Viennese style may be forever lost to us in
Dufay or Monteverdi. Admittedly, our growing knowledge of performance
practice lets us "resolve certain problems about how various musical notations
were meant to be rendered. "^^ But evidence about early playing styles is almost
never complete. We must fill in the gaps with our imaginations, and we have
twentieth-century imaginations.
To honor St. Patrick's Day, a small Iowa town once hosted an Irish-accent
contest. A visiting Dubliner —the only authentic Irishman in town —signed up
and seemed certain to win. He lost, to an lowan. To lowan judges, the real

thing didn't have enough of the ould sod to it. Perhaps when we fill in the gaps
of authentic Bach or Beethoven style, our sense of what's authentic is just as

biased.
What might also make historical re-creation impossible is that performing
contexts influence music-making, and old music is almost never performed now
in the contexts it was written for —chapel, feasting hall, music room, salon. ^^
Even music written for the concert hall has to contend with radically different
concert-hall sizes, acoustics, and audience behavior.^'* In a subtle example of a
context shift, Robert Philip observes that a modern audience's main listening
context is recordings, which are "perfect. "^^ If a concert performer, in the heat
of inspiration, flubs a few notes, many of his listeners, recalling their CDs, will
say not "How inspired!" but "Why should I pay good money to hear someone
who hasn't practiced?" Cowed by such responses, most modern performers

22. Neal Zaslaw, Notes 8 (March 1994), p. 948.

23. See Charles Rosen, "The Shock of the Old," New York Review of Books, 19 July
1990, pp. 46-52, and The Romantic Generation (Cambridge, Mass.: Harvard University
Press, 1995), pp. 383-85.
24. Nicholas McGegan and Robert Levin discuss the changes in audience behavior in their

interviews. Julianne Baird, in her interview, discusses how hall sizes have affected singing tech-
nique.
25. Early Recordings and Musical Style (Cambridge University Press, 1992), pp. 230-31.
 INTRODUCTION 9

spend hours practicing a concerto so that they get the notes exactly right, and
make that a higher priority than the flight of inspiration, which the orchestra
might not be able to follow neatly. This attitude is the opposite of those held
by many performers in even the relatively recent past (as early recordings
show); to the extent that recordings have changed our outlook, some critics say,

we can never play as artists played in the past.

Some historical performers try to "recontextualize" old music — for instance,

by embedding a Renaissance mass in re-enacted liturgy, or preceding a Mozart

concert by giving the audience minuet lessons. Such approaches, however, may
not quite bridge the chasm that divides us from the past. They do not solve a
deeper problem that is often raised: that our musical aesthetics reflect our emo-
tional, intellectual, and spiritual lives, which differ from those of past eras.

Even if we embed a Dufay mass in its liturgy, we probably won't feel, as

Dufay's listeners did, emotional associations between the chants and specific re-
ligious holidays (as Susan Hellauer explains in her interview). Even if we learn
to dance a few Mozartian dances, they probably won't signify class distinctions
for us as they did for Mozart and his listeners; unless we're told, we won't un-
derstand that the Count would dance a minuet but not a "relatively rowdy"
contredanse.-^*" And the problem might go deeper: If a passage in Mozart ex-
presses yearning, says the fortepianist Steven Lubin, we should know that eigh-
teenth-century yearning was different from twentieth-century yearning: it was
"more innocent and trusting" than the sort habitual to us, who have fewer
"hospitable realms" to yearn for.^^ If such "otherness" applies to Mozart, a
contemporary of Thomas Jefferson, how much more to Hildegard, who pre-
dates Thomas Aquinas?
Lubin believes, however, that we can re-create eighteenth-century yearning,
through historical immersion and through searching within ourselves. I myself
wonder whether yearning in Mozart's day was all that "innocent and trusting."
Mozart's father constantly warned him that "all men are villains," that "all
friendships have their motives," and that he should "trust no one"; Mozart
ended up "[sjkeptical, wary of easy solutions, doubtful of men's motives,
disdainful of panaceas."-** Wye J. Allanbrook is right, I think, to say that we
value the dark side of Mozart more than did he or his contemporaries, who
didn't share our post-Romantic "ingrained assumption that profundity and
melancholia go hand in hand":'" we do seem more prone to melancholy than

26. See Wye J. Allanbrook, Rhythmic Gesture in Mozart (University of Chicago Press,
1983), chap. 2, and p. 81.
27. Lubin, "Authenticity Briefly Revisited," Historical Performance 4 (Spring 1991),
p. 46.
28. See Maynard Solomon's Mozart (New York: Harper Collins, 1995), pp. 8 and 90 on
Leopold Mozart and and
355 on Mozart's skepticism.
trust, p.
29. Allanbrook, "Mozart's Tunes and the Comedy of Closure," in On Mozart, ed. James
Morris (Cambridge University Press, 1994), pp. 169-86; quote, p. 176.
10 INTRODUCTION

Mozart's contemporaries,^" and our era is clearly less optimistic than theirs
was, with its faith in the triumph of reason and the perfectibility of humanity.
But a close look at the joys and sorrows of both eras suggests that underlying
human emotional equipment has changed little since then.^' So the claim that
"otherness" makes historical re-creation impossible may be exaggerated (we'll
return to this in Christopher Page's interview). Still, there's no denying that we
differ in some ways from our predecessors, and that it affects how we hear or
play their music.
For all these reasons, almost no one today makes bold claims about play-
ing music exactly as it was played in the past. Attaining that goal seems too
difficult.

Of course, a goal might still be worth seeking even if it's impossible to attain.
If 100 percent historical accuracy were really ideal, 50 percent accuracy would
be better than none at all; and while a minuet lesson might not give us eigh-
teenth-century ears, it could still enrich our understanding. And the argument
that our imaginations must fill in some factual gaps can apply to any histori-

cal enterprise (and doesn't invalidate it). But some critics argue that even if per-
fect historical reconstruction were possible, it would not be worth the effort.
We can hear Elgar conducting his music on early recordings and could re-cre-

ate his performances accurately. But why should we play Elgar as Elgar did,
these critics ask, if we prefer it played differently? Why shouldn't we use
Mozart to express twentieth-century yearning? That may be what we need
Mozart for. And some listeners might prefer a Dufay mass with as little liturgy
as possible.
The original early-music view was, of course, that music sounds best when
played as the composer expected it would be played. In this spirit, the inter-
views in this book include many specific claims that historical practice improves
performance. If, for example, the composer wrote down only a bare skeleton
and expected the performer to flesh it out with ornamentation, then learning
how to ornament in the composer's style will generally give better results than
playing the bare bones. And (though this doesn't arise in the interviews) sev-

enteenth-century keyboard music has proven to be vastly more interesting when

played with historical instruments and tunings. ^^ These and other arguments
can be convincing; but in the end the principle is not as universal or as self-ev-
ident as it may seem.^' No one, for example, has tried to revive the French

30. Or Ravel's: in recent decades, rates of serious depression appear to have doubled every
ten years in many countries. See the Cross-National Collaborative Group, "The Changing
Rate of Major Depression. Cross-National Comparisons," Journal of the American Medical
Association 268 (1992), pp. 3098-105.
31. This is discussed in the Postscript to the Christopher Page interview.
32. John Butt pointed this out; personal communication, 1996.
33. Sometimes an informed critic even applauds the bare-bones avoidance of ornamenta-
 INTRODUCTION 11

Baroque practice of conducting by beating time on the podium with a large

wooden staff. And when we imagine shivering Thomasschule students, at
seven-thirty on a winter morning, performing a virtuoso chorus written three
days earher, we might ask whether we could tolerate truly historical Bach.
Whether or not the original way is best may also depend on who's listen-
ing. I once saw a Chicago production of Under Milk Wood with an English

friend. She informed me afterwards that most of the actors had vaguely Gaelic

accents, and that only Captain Cat's was truly Welsh. His was the only accent
I'd found difficult to understand. Perhaps Chicagoans are better served by er-

satz Welsh. Such thinking, some fear, might lead a musician to pander to an
audience rather than challenge its preconceptions; but principle may have
deeper musical applications. Joshua Rifkin is far from alone in noting that a
typical performance in the eighteenth century or earlier was what we would
call a barely rehearsed run-through. We could re-create that, he says, but maybe
an eighteenth-century audience, hearing a piece for the first time, needed only
a run-through, while we, who have heard Mozart and Bach so often, need an
interpretation. And the fact that performance context has changed also raises
questions about why we should want historical accuracy. Charles Rosen points
out that much of Bach's keyboard music was written for private use; if we tried
to play it in public in the same way that an eighteenth-century musician would
have played for himself —that without trying to project the musical events
is,

to an audience —we would defeat the purpose of concert-giving.'"

Such examples suggest that even if you believe that historical evidence mat-
ters, you still have to decide whether each particular historical practice does. Is

it important to making the music work today, or is it a meaningless accident

of history — irrelevant or even harmful from a musical standpoint? The "mean-
ingless accident" is what Donald Tovey had in mind when he said that if we
want to be truly authentic in performing Bach cantatas, we would have to "flog
the ringleaders of the choir after an atrocious performance.'"'
The distinction is by no means lost on most historical performers. Of course,
making the distinction is not an objective science; sometimes a conclusion (one
way or the other) is all but inescapable, but more often it depends on the per-
former's assumptions and priorities —and those of the performer's era. Our era
is more likely to consider historical practice important than the nineteenth cen-
tury was. And as we'll see, historical performers themselves draw the line in a
variety of places, from near-purism to near-rejection of the historical ideal.

The question of the accidental versus the relevant distills some of the "au-
thenticity" issues into the kind of practical problems a musician faces every day.

tion; see Richard Taruskin's praise of Artur Schnabel's unhistorical Mozart in Text and Act,
pp. 290-91.
34. Rosen, "The Shock of the Old," p. 50.
35. Tovey, A Musician Talks (Oxford University Press, 1941), p. 66.
12 INTRODUCTION

It applies, for example, to the original calling-card of historical performance

period instruments. Harnoncourt now focuses on modern-instrument groups; an
instrument, he says, is "a tool, not a religion." Nonetheless, when he conducts
modern orchestras he often uses historical brass and percussion instruments,
which he thinks have an inimitable effect. Anner Bylsma sometimes plays Bach
with instruments other than those specified, or with a modern bow; but he con-
siders using gut strings crucial to Boccherini's cello music.

As that suggests, how much of period practice is important might differ in

different sorts of music,

"lit] more acceptable," wrote Howard Mayer
is

Brown, "to play Bach's music on modern instruments than Rameau's,"^^ and
Bach seems robust in many other ways. But French Baroque music often re-
quires period instruments and a firm grounding in historical style even to be
interesting, much less effective.

The issue of accidence versus importance relates also to a major trend

among musicologists today, which seeks to understand music by reference to
its larger contexts — social, political, economic, religious, and so on.^^ Such mu-
sicologists often seek to understand what music meant in its own time, which
makes their project seem, at least at first glance, like a natural ally to histori-

cal performance. In fact, historical context comes up often in the interviews. I

suggested earlier that changes in performance context may make it impossible

(or undesirable) for us to play as people did in an earlier century; here we may
ask which aspects of larger historical contexts must be considered when trying
to make the music live in modern performance.
The discussions in this book often suggest an ecological web: change one
part of the system, and you change what is incidental to the music and what
is necessary to it. An authentic historical performance practice may no longer
fit, because we or our contexts have changed. If you re-create the exact size

and layout of the orchestra that premiered a Mozart symphony, but put it in

Carnegie Hall, it will sound puny. The "necessary" element is an adequately

powerful sound; the "historical accident" is the size of the premiere's orches-
tra, which made a big sound in the small, resonant halls of Mozart's day.^^
Some historical performers, perhaps, have focused on accidental features —such
as the exact size of an ensemble —rather than the important ones. My inter-

viewees are often more discriminating.

Should We Care about the Composer's Intentions?

Some doubt that even selective historical accuracy is a worthy goal. In par-

ticular, they question whether performers should be concerned with honoring

36. Brown, "Pedantry or Liberation?" in Authenticity and Early Music, ed. Nicholas
Kenyon (Oxford University Press, 1988), p. 30.
37. Discussed briefly in the Postscript to the Page interview.
38. See Neal Zaslaw, "Mozart's Orchestra," Early Music 20 (May 1992), pp. 204-5.
 INTRODUCTION 13

composers' intentions on how to perform their own music. During the

nineteenth century this ideal became widespread,''' and today, says Robert
Martin, "In general, the best performers have a strong sense of their roles
as servants of the composer.""" Admittedly, some critics believe performers'
subservience to the composer is lip service —^John Butt has called it little more
than "crocodile humility" — but few would agree. In any case, many in the
early-music movement (for example, Malcolm Bilson and Robert Levin) share
Martin's "strong sense."
Others in the movement appear to serve a slightly different purpose: they
try to play as the composer's contemporaries did. This may seem to make ar-
guments about composer's intentions somewhat less relevant to the authentic-

ity debates, but critiques of the composer's intentions raise arguments that, if

accepted, could undermine this approach as well. For example, Richard

Taruskin argues, to paraphrase him, that what really counts in the arts is the

experience delivered to the audience; what doesn't count is who comes up with
the means of delivery."" A distinguished critic has denounced conductors who,
like Barenboim, at the chorus's first entry in Brahms's German Requiem ignore
the p marking and instead have the chorus enter pp. Barenboim might respond
by pointing out that in his later years Brahms told a choral conductor that the
chorus should enter with "the softest pp.'"*^ But this argument against inten-
tions is different: it would say that if the effect is beautiful, it doesn't matter
who thought of it — Brahms, Barenboim, or Brahms's contemporaries. Another
argument against privileging the composer's intentions distinguishes between a
performance art and a textual one, and thus between the composer's intentions
regarding performance and those regarding notes. As Peter Kivy says, it's easy
to disprove the simplistic belief that the composer always knows best how to
play his or her music;"" and the composer's contemporaries could be equally
fallible.

Of course, as Kivy notes, composer's views on how to perform their own

works deserve special consideration —and sometimes composers do know best.
In previous centuries, composers were often great performers, so their perfor-
mance instructions reflect their expertise in that area as well as in composition;
and their compositions may have been partly determined by performance con-

Bowen, "Mendelssohn, Berlioz, and Wagner as Conductors: The Origins of the

39. Jose A.
Composer,'" Performance Practice Review 6 (Spring 1993), pp. 77-88.
Ideal of 'Fidelity to the
40. Martin, "The Quartets in Performance: A Player's Perspective," in The Beethoven
Quartet Companion, ed. Robert Winter and Robert Martin (Berkeley: University of Califor-
nia Press, 1994), p. 140.
41. See Taruskin, "Tradition and Authority," reprinted in Text and Act, p. 190, originally
published in Early Mustc 20 (May 1992), pp. 311-25.
42. The conductor was Siegfried Ochs; see Max Rudolf, "A Recently Discovered Com-
poser-Annotated Score of the Brahms Requiem," Quarterly Journal of the Riemenschneider
Bach Institute IIA (October 1976), p. 13.
43. Kivy, Authenticities (Ithaca, N.Y.: Cornell University Press, 1995), pp. 162-87.
14 INTRODUCTION

siderations. One must then distinguish between a performance instruction that

is part of the piece's identity (just like the notes themselves), and an instruction
that is, to quote Virgil Thomson, "not part of [the composer's] original cre-
ation, but rather one musician's message to another about it, a hint" on how
to put it over.""* If you play the slow movement of the "Hammerklavier" Sonata
at a presto tempo, you are not only contradicting Beethoven's performance in-

structions, you are, in a sense, creating a new piece. On the other hand, if your
slow tempo is slower than Beethoven's metronome marking, you may be choos-
ing what is (in your particular situation) a better tempo for the movement. You
may be serving one of Beethoven's intentions —winning over the audience with
a great performance of the movemenf*^ — by overriding another one'** (playing

at eighth note = 92).

Arguments about the composer's performance intentions can get more com-
plex than this, and the above is far from a complete survey. In practice, the res-

olution may be the same as with historical practices: it may boil down to case-

by-case decisions by the performer about whether each performance instruction

is essential, beneficial, insignificant, or inferior to some other alternative. Which
category each intention ends up in will, again, depend on the artist's (and the
era's) priorities. Furtwangler was at least as devoted to serving Beethoven's in-

tentions as Norrington is; but Norrington, unlike Furtwangler, defines such ser-
vice so that it includes the metronome marks.

Why Did the Early-Music Movement Happen f

This brings us to the most celebrated of the authenticity debates, those in-

volving motivation. These take my originating question

— "Why do they play
like that?" —to a deeper level. They ask not "On what grounds does Nor-
rington justify following Beethoven's metronome markings?" but "Why does
Norrington think it important to do so?" If the goal of historical performance
might not be completely attainable or desirable, why do so many people seem
to share it? The simplest explanation, as we've seen, is that they just think
the music sounds better played historically. That may seem an adequate rea-

son, and in some cases it's undeniably convincing. But, in general, one can't
get away with so simple an explanation of motives. A critic could, for ex-
ample, respond, "You can get used to many things once you decide that you

44. Thomson, The Art of Judging Music (New York: Knopf, 1948), p. 296. See also Kivy,
Authenticities, pp. 28-32.
45. Beethoven could show more concern for such things than we might expect. In an 1819
letter to his piano student Ferdinand Ries, he said that if the "Hammerklavier" Sonata "should
not be the right thing for London," Ries could leave out the slow movement or reorder the
internal movements. In Beethoven's Letters, with notes by A. C. Kalischer, trans J. S. Shed-
lock, ed. A. Eaglefield-Hull (London: Dent, 1926), p. 268.
46. Kivy, Authenticities, pp. 24-44.
 INTRODUCTION 15

should — but if performers ever run into a case where it sounds worse to them
played historically even after they get used to it, would they ignore the his-
tory, or would they play in a style they didn't like?" In short, for historical
performers is history a means or an end?
Some say these performers (many of them, anyway) do make it an end in

itself. The motivation, they say, is antiquarian, like Civil War battle re-en-
actments and other "living museums." But that kind of historical accuracy,
some add, should be an end only for scholars, not artists (the "uneasy match"
idea). Of course, when a musical culture is as fixated on the past as ours has
become, it might seem inevitable that some musicians would want to re-cre-
ate history for its own sake. But is it? After all, the emphasis on past mas-
terpieces needn't be antiquarian, since those masterpieces are believed to speak
to all times. In fact, many musicians understand the composer's intentions as
involving timeless elements of the work, which transcend historical circum-
stance.
Robert Morgan finds another explanation for our concern for historical
accuracy. Most musicians before our era, he says, believed themselves part of
a living tradition with a direct connection to the musical past, so they usu-
allysaw nothing wrong with playing Bach or Handel in the performer's own
modern style. That attitude still holds today among, say, rock musicians cov-
ering a Beatles tune: there's no expectation that they will simply reproduce
the original, because the tradition is still alive. In "classical" music today,
though, when modern musical styles seem unconnected with those of the past,
the musical past has become a museum. Its artworks are "no longer ours to
interpret as we wish" —that would seem like painting airplanes over a Con-
stable landscape — but "ours only to reconstruct as faithfully as possible." Mor-
gan contends that "concern for historical authenticity represents ... a situa-
tion characterized by an extraordinary degree of insecurity, uncertainty, and
self-doubt" —that is, a fragmented musical culture, which lacks a strong iden-
tity of its own."*^
Some go further and speculate that the concern with historical accuracy
"may be a symptom of a disintegrating civilization.""*** Perhaps; regardless of
whether eighteenth-century yearning was more naive and trusting than
really
ours, many of us yearn for a more naive and trusting world. Some listeners
and players want music to take them out of our world and into Bach's or
Hildegard's, whether that is impossible or not. A sense that our culture has
taken a wrong turn (or even that it is "disintegrating") is no longer the pre-
serve of fundamentalists, fascists, and reactionaries; as a college professor

47. Robert Morgan, "Tradition, Anxiety, and the Current Musical Scene," in Authenticity
and Early Music, ed. Kenyon, pp. 57-82.
48. Donald J. Grout, "On Historical Authenticity in the Performance of Old Music," in
Essays on Music in Honor of Archibald Thompson Davison (Cambridge, Mass: Harvard Uni-
versity Press, 1957), pp. 341^7.
16 INTRODUCTION

wrote in 1994, "Nearly everyone I know lives with the sense of serious de-
cline if not impending fall.'"*^

Since Berlioz,'" various people have questioned the assumption that modern
instruments are necessarily better than their predecessors; but, with a few ex-
ceptions, such questioning has become a force in the musical marketplace only

since the 1960s. That this date seems a watershed may reflect nothing more
than the growth of the recording industry; but as we've just seen, it is tempt-
ing to speculate about deeper social causes. Such speculations are, of course,
slippery. On the one hand, one could note that this rise in the market popu-
larity of historical performance coincides with when Robert Heilbroner sees so-

ciety on a large scale losing faith in human progress and perfectibility — and in

particular, in technological progress.^' (Obviously, he is aware that such doubts

had arisen in the nineteenth century and in the early twentieth, but says that
these doubts were localized to intellectual circles, and not nearly so widespread
as those of our era.) This fits in with the idea that yearning for an Arcadian
past motivates some historical performers. On the other hand, John Butt spec-
toward historical performance could equally well reflect the
ulates that the shift
"bombshell in late nineteenth- and twentieth-century historiography and
hermeneutics: that one's viewpoint is not neutral and absolute" but is instead
contingent on one's place in history.^^ Such an attitude could have made musi-
cians less likely to assume that their "natural" way of playing music was the
best way.
Indeed, some other suspected motives are not especially oriented to the past.
Some see the motivation behind the early-music movement as simple competi-
tiveness: if Karajan and Co. have already perfected mainstream style, the only
way to make your mark is to stake out radically new territory. Other observers
see this in a less cynical light. They think that historical performers are trying

to inject a dose of novelty into a flagging concert life, in which a limited reper-
tory and an increasingly uniform style have led to shrinking audiences.^^

49. Joseph Epstein, "Decline and Blumenthal," The American Scholar, Winter 1994 (Ep-
stein is a neo-conservative, but many liberals feel the same way). There are many optimists

too, of course —
for example, those who believe that computer technology will create a new
golden age. But even some techno-optimists exhibit nostalgia for a golden past.
50. Berlioz, "Instruments Added by Modern Composers to Scores of Old Masters," from
hisA travers chant (Paris, 1862). In The Art of Music and Other Essays, trans, and ed. Eliz-
abeth Csicsery-Ronay (Bloomington: Indiana University Press, 1994), pp. 148^9.
51. Heilbroner, Visions of the Future (New York: Oxford University Press, 1995). He sees

a somber, "apprehensive" view of the future becoming common only during the last thirty

years; the fin-de-sieclism of the 1890s was, he argues, localized, and should not be given undue
prominence. (Whether or not the world is in fact in a state of decline today is a separate ques-

tion, as he notes, and for purposes of the present discussion is irrelevant.)

52. Butt, personal communication, 1996.

53. See Nicholas Temperley, "The Movement Puts a Stronger Premium on Novelty than on
Accuracy . . .
," Early Music 12 (February 1984), pp. 16-20; Will Crutchfield, "Fashion, Con-
 INTRODUCTION 17

With few newly composed works reaching a significant pubhc, performers

sought a more appeahng brand of novelty and the past provided plenty of —
neglected works and new ways of playing familiar pieces. But if novelty is

all that historical performance offers, what will happen when the novelty
wears out?'"
Richard Taruskin argues that this novelty-making reveals some life in our
musical culture. Historical performance, he says, is not about putting musicol-
ogy into action — it's about trying to make old music suit our modern (and, he
argues, modernist) tastes. Regarding the acid-test question
— "if the music
sounded worse to them played historically, what would they do?" —Taruskin
documents cases of supposed historical purists ignoring inconvenient historical
evidence. These performers, he says, privilege the evidence they liked and ig-
nored or devalued the evidence they didn't. He concludes that by using evidence
selectively, even the most uncompromising historicist performers unconsciously
try to create the sound not of "then," but of now.^'' They are doing, in other
words, exactly what Morgan says musicians at most times have done —playing
earlier music in the style of their own day. The only difference is that because
of the museum-curator ethos that Morgan talks about, historicists have to pre-
tend (even to themselves) that they're being historically accurate. This makes it

irrelevant to argue about the "impossibility" of re-creating the past: the past is

something we construct to suit our needs. In many of his writings, Taruskin

tries not to deplore this or to dismiss it as the equivalent of the lowan Irish ac-

cent; instead, he calls it far more reassuring (and, in the deepest sense, more

viction,and Performance Style in an Age of Revivals," in Authenticity and Early Music, ed.
Kenyon, pp. 19-26; and Joshua Kosman's section of "The Early Music Debate," pp. 117-19.
54. Peter PhilHps, "Beyond Authenticity," in Companion to Medieval and Renaissance
Music, ed. Tess Knighton and David Fallows (London: Orion, and New York: Schirmer, 1992),
pp. 44-47. Phillips believes that this has in fact happened, and that historical performers must
now win attention on "the strength of their musical vision."
55. In many of the essays reprinted in Taruskin's Text and Act. His claim that early-music
style belongs to modernism — by
which he means a Satie/Stravinsky-like concern with light-
ness, —
formalism, and impersonality applied most effectively to some of the dominant British
artists of the 1980s; it was, perhaps, a specific example of his general point, an example
Taruskin was able to document at length. But even at the time, Taruskin acknowledged that
some of the Continental early-music leaders, like Harnoncourt and Leonhardt, were not
Stravinskyan objective modernists.
This does not necessarily disprove Taruskin's basic idea that historicists are creating a mod-
ern, not a historical, sound. Jordi Savall's gamba-playing does not reflect Stravinskyan mod-
ernism, but it does use more legato than French Baroque playing probably did (see Taruskin's
"Of Kings and Divas," The New
December 1993, pp. 31-44; on p. 43 he dis-
Republic, 13
cusses Savall with enthusiasm). This too may
modern tastes rather than history.
reflect Savall's
Romantic though it sounds, it may not entirely escape the label of "modernism." Rosen writes,
in "The Shock of the Old," p. 46, that modernism "has its neo-Romantic side"; Taruskin, in

the introduction to Text and Act (p. 10-11), describes modernism as a late manifestation of
Romanticism, not its antithesis.
18 INTRODUCTION

authentic) than true antiquarianism would be. "Being the true voice of one's
time," he writes, "is . . . roughly forty thousand times as vital and important
"^^
as being the assumed voice of history.
Despite the compliment, this argument has not endeared Taruskin to the
early-music world. And, of course, Taruskin's argument seems overstated if

it's put simplistically. Early-music performers have been known to ask in-
strument makers to use wood only from the very region used by the seven-
teenth-century instrument maker they want copied. Marcel Peres tells me that
at home he never uses electric lights, but only candles, in order to better un-
derstand the mentality of the Middle Ages. Telling such performers that they
aren't really trying to re-create historical practice does not win their grati-
tude, even if you then tell them that what they're actually doing is forty thou-
sand times more important.
This is part of why Taruskin is generally seen as early music's most fero-
cious detractor. That perception needs more discussion, because the published
record suggests something more complicated. Taruskin agrees, after all, that
"the fruits of scholarship can mightily assist the performer's purposes"; pur-
suing historical practice, he thinks, can free one from deadening habits. He
even says that "the best specialist performers get much closer to their chosen
repertory than their mainstream counterparts manage to do."^^ And as a re-
viewer he has warmly praised, in detail, the musicianship of many early-music
performers — I count twenty in the reviews I've surveyed —and has attacked,
according to my survey, only seven. But the impression that remains from his
work is of a scathing denunciation. The reason may lie partly in Taruskin's
rhetorical emphases, partly in his internal contradictions (his critics have noted
some),^^ partly in what got published where (the praise has been buried in
smaller-circulation magazines), and partly in how entertaining the attacks can
be. But another part of it may lie in the fact that his praise for the playing

is often mixed with complaints about claims of historical fidelity. Consider

Taruskin's response to one of many angry letters." Here he takes an approach
different from the one I summarized: he says he admires early-music per-
formers' idealism and musicianship, but continues, "What I am waiting for
is an end to the pretense that what Early Music performers are doing is being
[merely] historically correct. They are not ransacking history in pursuit of
truth. What they are seeking is permission. . . . Being human, when they find

"The Modern Sound of Early Music," originally published as "The Spin Doctors of
56.
Early Music" in The New York Times, 29 July 1990, Sunday Arts and Leisure section, p. 1.
Reprinted in Text and Act; the quote is on p. 166.
57. Text and Act, p. 306.
58. John Butt, "Acting up a Text," Early Music 24 (May 1996), pp. 323-32.
59. The letter is from James Richman, and was printed in the New York Times Sunday
Arts and Leisure section letter column on 26 August 1990; parts of it are reprinted in Text
and Act, p. 171, along with the commentary from Taruskin that I quote.
 INTRODUCTION 19

permission, they are apt to believe that they have found the truth and be-
come 'certain.'"*'"

Just how apt they are to beheve this you can judge for yourself as you read.
Whatever you decide, this idea of "seeking permission" brings us to another
proposed motive in historical performance. Needing permission reflects, of
course, submission to authority. Perhaps thinking of the moral tone of the turf
wars, Nikolaus Harnoncourt calls the authenticity ideal "very close to the kind
of political dogmatism and religious fundamentalism that are so much part of
our times.""' Taruskin and the gambist/musicologist Laurence Dreyfus,*^^
among others, argue more generally that the need to get "permission" is all but
universal in classical performance today. This authoritarian need, Taruskin

says, reflects the exaltation of the composer and the musical work over the per-
former and the performance. Permission usually comes from the composer's
score, he and Dreyfus say, but among early-music purists permission must also
be granted by another authority: scholars of performance practice. Not sur-

prisingly, Taruskin considers this "tyrannically limiting."

All the same, another motivation for historical performers might be the
opposite of adding more constraints of authority: it might be to sidestep the
conflicting demands makes of performers today. Over
that musical authority
the past few centuries, our art-music culture has given performers less and
less latitude in determining what notes to play or how. But since about 1800

our culture has also put far more emphasis on individual expression and cre-

ativity than in pre-Romantic times. Performers are supposed to express the

composer's emotions and intentions, not their own; but they are also sup-
posed to be original, insightful, and creative.*'^ Some believe they can't win:

whatever they do, someone will attack them, either for serving the composer
too slavishly or for expressing themselves too willfully.*"* This conflict is at

the root of the critics' arguments about inspirational artists like Bernstein and
Furtwangler. Perhaps historical performance offers a way out of this dilemma.
Historical performance obliges you to serve the composer's intention by learn-

ing all you can about his era's playing style, including idiomatic features (like

the Viennese waltz rhythm) that were never explicitly notated. (Let's overlook

60. Text and Act, p. 171.

61. Stephen Johnson, "Making It New," Gramophone 69 (May 1992), p. 26.
62. Personal communication, 1995.
63. The ethnomusicologist Marc Perlman, who is doing a study of the early-music scene,
points out that there are other music cultures which give their performers little latitude in de-

termining the notes but do not lavish attention on the figure of the composer. What makes us
distinctive is that we lionize the composer and freedom but also
limit the performer's creative

value originality highly. We thereby sever the functions of composer and performer, and place
them in a relationship of mutual dependence in which a certain tension inheres. Personal com-
munication, 31 December, 1994.
64. For an example of the latter, see Bernard Holland, "When the Musician Upstages the
Music," The New York Times, 24 May 1995, p. B2.
20 INTRODUCTION

the inconvenient fact that composers sometimes ignore their own idiom, as
in Bernstein's recording of West Side Story.) To regain the style of a past era

means that you have to improvise, ornament, add notes skills that are hardly —
recognized in mainstream conservatory training —
and phrase and articulate
quite differently than you learn to do in such a conservatory. As a result, you

can creatively rethink how Even better, especially if you play music
to play.
about which there's little or no evidence regarding performance style, you can
construct your own style more or less from scratch. An advantage of such
stylistic creativity is that it is less likely than mere originality to get you ac-
cused of willful self-expression — after all, you are just doing what was done
by the composer's contemporaries, and musicology says so. Historical per-
formance, then, may
some performers have it both ways, by combining
let

creativity and fidelity.^^ Michelle Dulak argues that the true defining charac-
teristic of the historical-performance movement today is that it offers "radi-

cal freedom from mainstream convention"; its players "are expected merely

to sound different and are given such wide latitude that they can be differ-
ent in nearly any way that pleases them."**
It could be, of course, that all of the motivations I've mentioned — and oth-
ers too —act on different historical performers, or perhaps on the same per-
former in varying degrees. It's not difficult to find evidence both for and against
every one of these motivations somewhere in this book. And they hardly ex-
haust the possible motivations. For example, John Butt, at the outset of his in-

terview in Chapter 9, gives some reasons for his interest in historical perfor-
mance that are quite different from those suggested above.

Why Does It Matterf

Das Beste, was wir von der Geschichte haben, ist der Enthusiasmus, den sie

erregt [The best that history has to give is the enthusiasm it arouses]. (Jo-

hann Wolfgang von Goethe)

The musicologist Leo Treitler recently began an article about a Marcel Peres
concert by remarking on the "decline of Early Music Talk," by which he meant
the controversies discussed above. He then proceeded to give a fascinating ex-
ample of Early Music Talk.*^ It seems to me that the Talk is, if anything, get-

65. An argument against the proposals I've made in the preceding two paragraphs is that
they may overestimate the prestige of the composer relative to the performer. After all, some
say, it is the performers who are idolized and enriched, not the critics or composers.

66. Michelle Dulak, "Early Music Circles Its Wagons Again," The New York Times, Sun-
day Arts and Leisure section, 11 June 1995, p. 40.
67. Leo Treitler, "Remembering 'Early Music,'" in Thesis 8 (Fall 1994), pp. 32-33. "The
decline of Early Music Talk" is a phrase quoted from Bernard Holland's "A Streak in the
Heavens."
 INTRODUCTION 21

ting more interesting these days, interesting for what it says about both the past

and the present. And I consider the extent of controversy surrounding histori-
cal performance to be a sign of artistic vitaUty.

All the same, the controversies may not convey why the fuss is worth it.

Analyzing premises and hidden motives can obscure creative achievements. Two
of the most vocal critics of the premises of historical performance, Taruskin
and Rosen, concur about these achievements. Rosen says that through taking
"the indefensible ideal of authenticity" seriously, historical performers have in-
creased "our knowledge . . . and our musical life [has beenl enriched."*^

Taruskin calls historical performance "the moribund aspect of our classi-

least

cal music life";"'' it allows a musician to "remake oneself," to challenge all of

his or her "knee-jerk habits."™ (Laurence Dreyfus argues that it can impose
uniformity and knee-jerk habits of its own;"' but I don't find these faults af-
flicting the artists in this book.) I will be less circumspect than Taruskin and
Rosen. As someone who has reviewed and played mainly modern instruments,
I've found that the best historical performers provide some of today's most in-
sightful, original music-making. Historical reconstruction has its own fascina-

tions, but its ultimate justification has been in the moving performances of
many artists like those interviewed in this book. Moreover, no one can deny
that only historical performers have made it possible for us to live with great
music — from such giants as Hildegard, Dufay, or Josquin —that had been
buried for centuries.
I've spoken of the incompatibilities of musicology and performance, but the
two share a common interest: the same works of music. The interactions of

musicology and performance can reveal important things about those works,
things that would stay hidden without the joint effort. Thus my interview sub-
jects offer us entryways into composers many of us had never heard of, like the

sixteenth-century Spaniard Alonso Mudarra; they share insights into some

whose music may be less familiar than we think, like Palestrina; and they make
us reconsider some whose music we thought familiar, like Mozart. And my in-
terviewees often give revealing answers to my original question, "Why do they
play like that?" In doing so, they often challenge assumptions that many of us

make about music.

68. "The Shock of the Old," p. 52.

69. "The Modern Sound of Early Music," p. 170.
70. "The New Antiquity," p. 231.
71. "The Early Music Debate," pp. 114-17.
 MEDIEVAL MUSIC, PLAINCHANT,
AND "OTHERNESS"

We have so little evidence about how medieval music was played that any per-
formance of it becomes, as Jonathan Harris puts it, "something of a personal
vision quest."' We are often unsure about things as basic as a piece's rhythm,
and will probably never have anything but educated guesses about less intrin-

sic matters such as bowing, tonguing, and voice-production techniques. As my

interviewees show, however, the dearth of evidence brings forth some ingenious
sleuthing.
Perhaps because the era is so distant, medievalists tend to raise the question
ofwhy we bother playing this music at all. Sometimes their answers involve
what we might call nostalgia a sense — that medieval music conveys a vision of
the divine, or a connectedness to the rhythms of life, that the modern world
has lost. But in one way or another, not necessarily nostalgic, all the medieval

performers I've interviewed express the view that encountering music from this
remote time and culture can enrich modern life; the later interviews raise this
topic less often.
That medieval music might bear on modern life is suggested by recent
record sales. In 1994, a reissued CD of Gregorian chant by the monks of Santo
Domingo de Silos reached number one on Billboard's classical charts (as of

mid-1996 it had never fallen below the number three rank) and number three
on the magazine's overall charts. At last count the CD had sold over six mil-

lion copies worldwide.' Dozens of other plainchant-related CDs — particularly

1. Sarah Cahill, Jonathan Harris, and Bernard D. Sherman, "Berkeley Festival Stretches
the Boundaries," Historical Performance 7 (Fall 1994), p. 131.
2. David Littlejohn, "Chant Meets Culture," Early Music America 2 (Fall 1996), pp.
24-32.

23
24 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

those of Anonymous 4 — are selling briskly, at least by the modest standards of

classical records. This wave clearly reflects our era more than it does medieval
music. As we'll see, plainchant served its first audiences, who were also its

singers, less as music than as prayer; for modern CD buyers, it usually provides
ambience. Medieval monks did a lot of praying; we, it seems, need a lot of
background music.
But are prayer and ambience as unrelated as they seem? James McKinnon
informs us that early Christian chant was "a form of meditation," sung by
fourth-century monks "for extended periods in an effort to maintain a medi-
tative state. "^ Later on, monastic life became more communal and regulated,
but, says Christopher Page, chanting was still meant to calm the mind, body,
and senses in order to create a "meditative quiet" in which "a monk could hear
the voice of his Creator.'"' Such total involvement is obviously very different
from background music; but there may be a connection. When we use chant
for ambience, the only "meaning" involved seems to be a vague feeling of time-
—
otherworldly purity and calm which we hope will give temporary relief
less,

from the pressures of time and the world. Although chant seems to have con-
veyed more specific feelings and meanings to its monastic singers, didn't they
also use it — as we do —to screen out such pressures?
The four interviews that follow all include attempts to understand what
chant meant to its originators (the recurring emphasis on chant is not by my
design). We'll see that theorists of the later Middle Ages discussed conventional
associations between the elements of plainchant and specific emotions (though
we could question whether such theory governed practice), and that many
chants were associated in people's minds with specific liturgical events —asso-
ciations that we lack. Still, as I've suggested, it might be worth asking whether
what chant and other music meant to medieval people has any continuities with
what it means to us, and what such continuities might be based on. As v/e've
seen, "otherness" —the idea that the experiences of people from remote times
and places have little common with our own — has been claimed
significant in
for a composer as recent as Mozart; how much more might it apply to a me-
dieval hermit or minstrel? On the other hand, nostalgia may tempt us to ex-
oticize those who were not really so different from us. Just how "other" were
our medieval forebears? And would serious musicians bother with their "vision
quests" if they believed that continuities with medieval experience were im-
possible?

3. James McKinnon, "Desert Monasticism and the Later Fourth Century Psalmodic Move-
ment," Music and Letters 75 (November 1994), p. 507.
4. Page, "Musicus and Cantor," in Companion to Medieval and Renaissance Music, ed.
Tess Knighton and David Fallows (London: Orion, and New York: Schirmer, 1992), p. 76.
 1
A. Different Sense of Time
Oo^
Marcel Peres on Plainchant

Newspaper reports on the monks of Santo Domingo de Silos have called their
best-selling CD "an album of 1,000-year-old Gregorian chant" or even "1,500-
year-old chant." But both estimates make the same mistake: they assume that
the chants on the CD are unaltered relics from the Middle Ages. If anything in

music can be shown clearly, it's that the chanting of modern monks bears only
a general resemblance to what was sung a thousand years ago. The very con-
cept of an "original form" of the chants is problematic.
Not that people haven't tried to find original forms. It can be argued that
today's historical-performance movement began (like written-down Western art

music itself) with plainchant. Like the CD's target audience, the nineteenth-cen-
tury religious "Isought refugel from the unwonted strangeness of the present"
in ancient church music' Among them were a group of Benedictines at

Solesmes near Le Mans, whose attempt to resurrect ancient plainchant proved

momentous. Like today's early-music performers, they wanted to get back to
the way it was — body of chant that, according to tradition,
in this case to a

had been whispered by the Holy Spirit, in the form of a dove, into the ear of
Gregory I, the pope who reigned from 590 to 614. Today it appears that Gre-
gory had "virtually nothing to do with either liturgy or chant. "-^ The misattri-

1. Carl Dahlhaus, Nineteenth-Century Music, trans J. Bradford Robinson (Berkeley: Uni-

versity of California Press, 1989), p. 181.
2. See James McKinnon's "The Emergence of Gregorian Chant," in his Antiquity and the
Middle Ages (London: Macmillan, 1990, and Englewood Cliffs, N. J.: Prentice-Hall, 1991).
This quote comes from earUer in his book, p. 19; the following discussion of the misattribu-
tion simplifies his explanation, which is given on pp. 115-17.

25
26 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

bution had several causes; for one thing, Gregory's sainted name lent author-
ity to the Carolingian emperors as they replaced their subjects' many chant tra-
ditions with their single official one. They did this not only from political mo-
tives, however, but also because they believed their "Gregorian" chant to be
the authentic early Roman one. They too wanted to get back to the way it was.
The Solesmes research into the so-called Gregorian repertory began in mid-
century, and the monks eventually collected an enormous amount of original
material. In 1903, after decades of internecine struggle. Pope Pius X threw his
authority behind their project. Since then, the chant repertoire and style of
singing developed by Solesmes has been canonized and used in most Catholic
chant, including that of the monks of Silos.
But despite their stated goal, what the monks of Solesmes actually produced
was very different from medieval Gregorian chant. Regarding singing style, for
which evidence is scant, it seems that what they created reflects, as Joseph Ker-
man said, "the ideals of the Cecilian or Pre-Raphaelite movements more closely
than anything that can conceivably be imagined from the ninth century."^ It
seems a textbook example of Taruskin's idea, discussed in the introduction, that

historical re-creations unconsciously reflect the re-creator's taste. In recent

decades, many elements of the Solesmes method, having to do with details such
as ornamentation but above all with rhythm (which Susan Hellauer will dis-

cuss), have been vigorously debated and revised. Some of the best work has
come from within the walls of Solesmes itself. An obvious question is whether
a century from now it will seem to reflect our era or an advance in historical

accuracy (or, perhaps, both).

Nor were the actual chants canonized by Solesmes historically accurate.
Solesmes, as we've said, made the crucial assumption that there had been a pris-
tine repertory of chant centered in Rome at the time of Gregory I, and that the
rest of Europe sang distorted variants of it. But David Hiley speaks for most
scholars today when he comments, "It is not at all certain that an 'original'
form of this type ever existed. The manuscript tradition
. . . is too variable for
a single 'authentic' reading to be deduced even from a small group of the ear-
liest sources."'*

Since the Second Vatican Council in 1963, Gregorian chant has become a
rarity in Catholic churches. As Mary Berry points out, this has proved a bless-
ing in disguise for modern performers interested in re-investigating chant. ^ It

3. Kerman, "A Few Canonic Variations," reprinted in his Write All This Down (Berkeley:
University of California Press, 1994), p. 47.
4. David Hiley, Western Plainchant: A Handbook (Oxford University Press, 1993), p. 628.
McKinnon seems "The Emergence of Gregorian Chant," pp. 111-17,
to disagree; see his
where he speculates about the possibility of a stable core repertory dating from the time of
Gregory the Second (pope from 715 to 731).
5. Mary Berry, "The Restoration of the Chant and Seventy-five Years of Recording," Early

Music 7 (1979), pp. 197-217.

 A DIFFERENT SENSE OF TIME 27

has encouraged them to pursue other approaches to Gregorian chant perfor-

mance, and to explore the chant repertoires that were suppressed centuries ago
in favor of the Gregorian. In these explorations, no one has been more adven-
turous than Marcel Peres. Peres has devoted his career to exploring such chant
traditions as the Old Roman (which was sung in Rome, except perhaps for the

Vatican, until the thirteenth century), the Beneventan (sung in southern Italy
until the eleventh century), and the Mozarabic (forbidden in Spain at the end
of the eleventh century, but still sung in some places until the fifteenth). Peres
has also explored Ambrosian chant, which escaped suppression and was sung
in Milan — though not necessarily in its original form — until our own time, and
various Gregorian "dialects," such as the seventeenth- and eighteenth-century
plainsong repertoires of Auxerre, Paris, and the Italian Franciscans.
In approaching lost repertoires, Peres has collaborated with distinguished
modern representatives of non-Western chant traditions, especially the Greek
Byzantine and the Syriac (Syria was the first Christian center outside Palestine).
He argues that these repertories have important links with Old Roman, Am-
brosian, and Beneventan chant, and that his collaborations have solved other-
wise impossible performance problems. This approach, not surprisingly, has
been controversial and has often been criticized by scholars. But the musical re-

sults have been, it is generally agreed, mesmerizing.

Part of the mesmerism comes, I think, from Peres's choirboy background. He
never forgets that chant was not music in the modern sense, but (since about the
fourth century) the prayers of monks whose lives revolved, all day, every day,
around the church liturgy. In this sense, the monks of Santo Domingo de Silos
are authentic in a way that few non-monastic chant performers can be; as the
interview makes clear, Peres takes this very seriously. He emphasized this music's

unmechanized, unhurried sense of the unfolding of time, and the idea that the
West's experience of time has changed over the millennium. Perhaps the older
sense of time, he implies, is part of what appeals to us in this music. Just how
different, we might ask, was that sense of time from ours?

Gregorian Chant

For modern listeners, plainchant is usually taken to mean what we now call

Gregorian chant. You've been exploring other aspects that chant has taken.
In a general sense, Gregorian chant means the chant of the Church of Rome.
But in different times, places, and ideological centers, the content of this reper-
toire changed. Today it means essentially the repertoire that was printed in
1908 in what we call the Vatican edition, the official publication of Gregorian
chant by the monks of Solesmes. In this edition, they collated most of the sur-
viving manuscript versions of a specific chant and, using statistical methods,
abstracted something they called the "authentic" version of the Gregorian
melody. But they were deriving a specific chant from chants composed in dif-
28 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

ferent parts of Europe and different eras, ranging from early Christianity
through the nineteenth century.

So a specific chant in the Vatican edition might never have existed before
Solesmes.
Right. If you want to reconstruct how it was in the Middle Ages, you have
to consider many other sources of information. Above all, you have to go back
to the manuscripts themselves, because even the same melodies often vary
greatly from one place to another. Also, in the Middle Ages each place had its

own repertoire of polyphonic settings and tropes (poetic and musical comments
on the canonical texts); in the Gregorian revival of the nineteenth century, they
didn't want to deal with these important aspects of medieval music. Only
polyphony in the style of Palestrina was held to express the "Catholic spirit in
music."

Solesmes also canonized a specific style of singing, which we still hear today
from most groups of monks.
Yes. Regarding Solesmes, we have to be precise about which aspect of this

community we are talking about: the scholarly one, the liturgical one, or the
aesthetic one.
First is the scholarly work they've done on collating manuscripts, which is

very important; they were the first to publish a collection of manuscripts in fac-
similes. Second is the liturgical aspect of their work, which has been focused
on the idea that the Catholic Church must live in unity, having throughout the
world the same liturgical practices, the restored Roman rite being the norm. So
they wanted to get rid of the local traditions that still existed in the nineteenth
century.
And third is the aesthetic side, whereby they developed a style that was just

the opposite of the singing style of traditional church singers of the nineteenth
century. These singers used to have a very strong and deep bass voice, and their

chant was highly ornamented, just the contrary of what we're used to now. To
Solesmes, that was an eighteenth-century tradition, and they could not imagine
that that way of singing might have any links with medieval singing. We must
be precise in noting that they were not at all interested in re-creating the me-
dieval aesthetic; they wanted only to reconstruct a tradition they believed to be
of the time of St. Gregory in the sixth century. The way of singing that the
monks of Solesmes developed chiefly in the beginning of this century, then, was
with a very high voice with an almost uncolored timbre and no ornamentation.
Their publication was not meant as a critical edition; it was a useful, prac-
tical book. It was to be used by an amateur parish choir, so they had to imag-
ine a very simple method that would not require that the singer be able to read
complex music. Dealing with theoreticians of the Middle Ages would have been
too complicated. So they developed a method that most people could sing.
 A DIFFERENT SENSE OF TIME 29

The Solesmes method of singing has been called "This very beautiful, very
Romantic, and somehow very French tradition of singing (that] has never
ceased to dominate our notion of Gregorian chant. "**
Could you explain why
it might be considered Romantic and French?
Romantic, because the aesthetic beginning of this restoration was Hnked
with the Romantic idea of a mythic Middle Ages, the "age of faith" as they
used to say. Musically speaking, most of the elements of nineteenth-century mu-
sical performance are found in the Solesmes performances: the legato phrasing,
the lack of ornamentation.
As for it being French, that is simply because Solesmes is in France. But we
must keep in mind that all regions and cathedrals in France used to have their
own styles, which disappeared after the normalization that the Solesmes style
created at the beginning of this century.
Indeed, I don't try to find the authentic way of performing Gregorian chant.
I am much too aware of all the different styles that coexisted throughout the
centuries. For each manuscript, period, or repertoire, I try to create a specific

performance. But above all, I try to remain open-minded and to change my in-

terpretation if a new aspect I was not aware of comes into consideration. Each
of my records shows a different approach to chant.

Reconstructing Ancient Chant

There was a range of different chant repertories and styles in Europe before
Gregorian chant. Traces of some of them survive; and you've been the most ac-
tive of anyone in resurrecting them for performance. How do you go about

that?
I think the revival of ancient music is a sort of equation. On one side are
the documents. On the other side is the performer, with his personality, voice,
education, and skill in doing music and living with it. And then you have the
understanding of the source. By that I mean all that the performer, and the
scholars he refers to, have understood — not only of the music and its function,
but also of the nature of the tools they are using today to re-create the past.
That last aspect is why you must always work on the original notation, with
good musicians who come from different worlds, and why you must have re-

lations with researchers not only in your field but also in other subjects con-
nected with yours.
In Ensemble Organum, we use singers who come from all parts of the mu-
sical landscape: from folk music, liturgical music, early music, opera, and so

6. Katherine Bergeron, "Chant, or the Politics of Inscription," in Companion to Medieval

and Renaissance Music, ed. Tess Knighton and David Fallows (London: Orion, and New York:
Schirmer, 1992), pp. 101-3. Mary
"The Restoration of the Chant," describes Solesmes
Berry,
chant's "smooth expressive legato with its undoubted 'spiritual' quality lilted accents, and . . .

the softening of the melodic peaks which gives the style its extraordinary elasticity."
30 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

on. Others are instrumentalists who come to me to learn to sing. For me the
important thing is to work with musicians who can add something to what I
think I've figured out. I try to be aware of my limits.

How about musicology: how do you use that?

I work from time to time with musicologists on certain specific subjects, but
also with historians, philologists, liturgists, and ethnomusicologists. Since 1984,
I have managed a center for the research and interpretation of medieval music
at the Fondation Royaumont, near Paris. We work on research programs that
may last one year or many. The role of Ensemble Organum is to make known
the musical result of this research. We pubUsh books, organize symposiums,
and make instruments. We also invite a group for residence at Royaumont each
year; they may also be musicologists or historians or instrument builders. We
offer them the opportunity and the tools to study a specific problem. In this
way, we try to be in touch with most of today's leading personalities in the

study of medieval civilization.

For instance, we are engaged now in a three-year study of how aesthetics
changed in relation to changes in political power, in different cathedrals in Eu-
rope. We aim to figure out how an aesthetic gained coherence in coordination
with all the aspects that made up the life of the cathedral —the economy, pa-
tronage, architecture, painting, sculpture, music —that is, all the fields that
work towards the celebration of the liturgy.
In the project, we are studying four cathedrals. In three of them they had a
complete change of repertory at specific dates, while in the fourth. Sens, they

were still singing from thirteenth-century books as late as the eighteenth cen-
tury. Throughout the centuries, they wanted to keep the Carolingian traditions,
because the Archbishop of Sens received his of "Primat de Gaule et de Ger-
title

manie" from Charlemagne. This shows us one of the problems often met in the
history of music: how to appreciate the continuity of a tradition in one place
while other musical events, sometimes very different, occur elsewhere. We tend
to think, for instance, that at the time when Machaut composed his Mass
everybody knew the work, and that everybody was doing music that way. In
history it's been realized for several decades that this is nonsense, but in musi-
cology you still find this way
The popularizing history of music
of thinking.
tends to be much too factual.^ And we lose what is, to my mind, one of the
most important things we must be aware of, which is the persistence in some
places of some practices in music. When you realize that in Sens Baroque music

7. On Reinhard Strohm, "Centre and Periphery," in Companion to Medieval

this point, see
and Renaissance Music, pp. 55-59, or in his The Rise of European Music (Cambridge Univer-
sity Press, 1993), pp. 62-105. Machaut's Messe de Nostre Dame, from c. 1360, is the earUest

surviving cycUc setting of the Ordinary of the Mass to show "conception as a unit." As Phihp
T. Jackson writes, "In one of the ironies of music history, there is no evidence that Machaut's

unparalleled work had any direct influence on future developments" {Companion, 120).
 A DIFFERENT SENSE OF TIME 31

was coexisting with some forms of medieval music, it changes your conception
of periods in music history.
I have the same kind of question regarding the changing of notation in the
thirteenth century from neumes to square notation.** When they made this

change, did they also change, in every place, the way they sang? A lot of schol-
ars think so, but I am not so sure.
Rome is another good example of an aesthetic shift, when the Old Roman
chant was forsaken and supplanted by the Gregorian at the end of the thir-

teenth century. One thing have been wondering about for years, but have not
I

been able to come up with an answer to, is this: When Rome changed from
the Old Roman to the Gregorian chant, was there a change in their voice pro-
duction and in the style of the music itself? The liturgy changed, but I'm not

Old Roman chant, to define the difference

able, even after ten years of singing

in aesthetic between Old Roman and Gregorian chant, because some Grego-

rian chants can be understood in a certain way, in which the notation refers to
certain ornamentation formulas very similar to Old Roman chant.

David Hiley notes that "the main difference between the Gregorian and the
Old Roman chant concerns surface detail: Old Roman is more ornate, '"* much
more ornamented. He adds, "In many places the two versions [of a chant] are
almost identical [in Gregorian and Old Roman chant] and there is evidently a
close relationship between them." Scholars have argued over what this might
mean.^^' What is your opinion on how the two repertories may have interacted?
At first glance, the Old Roman chant seems to be more ornamented. But in

8. From the ninth century to the twelfth, the term "neume" referred not to a form of no-
tation, but to "a sounding melody, or phrase, in particular one which has no words" (Hiley,
\(/estern Plainchant, p. 345). Nevertheless, today the term has gained currency as a label for
the kind of notation used from the ninth century to the thirteenth. Neumes did not record ei-

ther pitch or rhythm precisely; their function was to indicate a melody's direction and con-
tour, as well as certain nuances of performance. This partly reflected the orality of the tradi-
tion —the notation aided people who had already learned the repertory—and also the concept
of the music, where "melodic identity
meant identity of contour, not a literal identity of notes"
(D. Fenwick Wilson, Music of the Middle Ages [New York: Schirmer, 1990], p. 25). This is
true of many modal traditions elsewhere in the world.
Square notation, developed in the second half of the thirteenth century, used a four-line
staff indicating the height of each pitch with square note heads (originally developed by Guido
of Arezzo in the eleventh century); the notation gave some rhythmic indications as well.
9. Hiley, Western Plainchant, p. 532.
10. Scholars have contended that this could mean that the Roman style, in the two cen-
turies before it was finally written down, grew more elaborate compared to the forms in use
when Pepin, the father of Charlemagne, had the pope send Romans to teach their form of
chant to the Franks (thus creating Gregorian chant); or that the Franks didn't understand elab-
orate music, so they simplified what they learned from Rome. There is also evidence that the
Gregorian chant had influenced the Old Roman by the time the latter was written down. See
Hiley, pp. 561-62, for a summary of scholarly arguments. Peres argues for another interpre-
tation.
32 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

the Gregorian neumatic notations, a lot of the signs can be performed as or-
naments or even formulas.

So the traditional Solesmes concept of Gregorian chant underestimated how

ornamented its early form was?

Yes. The paradox is that it's easier to know how to perform the Old Roman
than the Gregorian chant. I have the impression that in most of the pieces
(some are very different), we have in the manuscript of Old Roman chant a
sort of recording of what and creative singer could do when per-
a skillful
forming what we call Gregorian chant. That is to say that in the Old Roman
chant the ornaments and cadential formulas had been written down, but we
have very few examples of notated ornaments in Gregorian chant.
The only thing we are sure of is that every important place had its own tra-

dition. An example can help us to understand this. At the beginning of the

twelfth century, the first Cistercians from Burgundy were sent to learn the
Roman tradition from the singers of Metz. In music, as in all aspects of monas-
tic life, the Cistercians wanted to go back to the original traditions, and for
music this meant Rome. But the two Cistercians sent to Metz were shocked by
what they heard there. They could not believe that it was the true tradition
which shows that chant singing in Burgundy was very different.
Now, their mission to Metz had been instigated by their abbot, Etienne
Harding. Harding knew the Roman tradition because he had made a pilgrim-
age to Rome in his youth. Even if on paper the Gregorian version from Metz
looks different from the Old Roman chant, it is interesting to notice that they
sounded much the same, at least in the mind of Harding. So he decided, against
most of his monks, to follow the Metz model. It was only after his death that

the Cistercians made their reform of chant. The story reminds us that notation,
even when we believe it's precise, conveys only part of the musical event. It

never tells you the sound of the voices. By this I refer not only to the voice pro-

duction, but also to the value of the intervals, and especially to musical prac-
tices that include the way of doing ornaments. Two melodies with different
notes can be perceived as the same if they are sung with the same vocal style,
and two melodies written with exactly the same notes can, if they are sung in
different styles, be perceived as different.

Old Italian Chant and Non-Western Traditions

I'd like to ask you about your work in reconstructing Old Italian chant.

I started to study the Old Roman chant in 1984. When reading this music,
I was missing something. I couldn't understand the aesthetic of this
realized I

You can't catch it with a standard modern Western approach; you

repertoire.
need something else. Most of the scholars who described the Old Roman chant
talked about tedious, boring, unimaginative music, chiefly because it contains
 A DIFFERENT SENSE OF TIME 33

a lot of apparently repetitive formulas. It uses a different logic from the Gre-
gorian one, and some musicologists, who didn't understand the way it works,
concluded it was a decadent system.
I noticed that this repertoire preserved until the thirteenth century some
pieces in Greek. To try to understand what was going on in this music, I

thought it would be interesting to work with a Greek singer and a Greek mu-
sicologist. So I contacted Lycourgos Angelopoulos; it was really intuition, be-

cause I had heard him in concert in Barcelona a few years before, and I had a
sense that he was living a lot of the things I was trying to understand — it was
everyday to him. I asked him if he could be interested in working with me on
the Old Roman repertoire. He told me, "I know nothing about Gregorian
chant; I cannot be useful to you." I said, "That's exactly why I wanted to get
in touch with you, because you don't have preconceptions. You'll come to Old
Roman chant like a virgin but with all your own background."
And it was a revelation, maybe the biggest of my life. After three or four
difficult days, he was able to get into the music; he brought a different men-
tality to dealing with the modes, the rhythms, the intervals, and so on.

You argue for a strong Byzantine influence in Old Italian chant. But Darid
Hiley, after reviewing the evidence, argues that the ^'overwhelming impression

is that Roman chant developed largely independently from Greek models, " and
that "Byzantine musical influence can be seen to reduce itself largely to a num-
"
ber of individual instances. (He saw more examples of Byzantine borrowings
in Milanese and Beneventan chant. }^^ How would you respond?
First of all, we must consider the words we are using to talk of the past. In
your question you use the words "Byzantium," "influence," and "Italian."
From the beginning you assume that Byzantium and Italy had two distinct cul-
tures and that the first influenced the second. But let us consider the facts from
the beginning. From the second century B.C., the Roman and Greek cultures
not only had relations, but very quickly the Greek model and its opening on
Eastern cultures became the reference for Rome. And when we go to the first

centuries of the Christian era, it's impossible to locate the boundary between
the Roman and Greek liturgical cultures. The Roman liturgy was exclusively in
the Greek language until the fourth century, and it retained a lot of Greek until
the ninth century. Even in a twelfth-century manuscript of Old Roman chant
you find seven Alleluias with Greek verses.'-^

11. Hiley, Western Pbinchjnt, p. 52''. Leo Treirler rejects the connection of Old Roman
and Byzantine chant too, in that the oldest records ot Byzantine chant were written down in
the twelfth century and bear no resemblance to Old Roman chant. As for modem Byzantine
it originated in 1300, when the Old Roman chant repertory was no longer sung,
chant, he says
and he adds that it has since been influenced by centuries of Turkish occupation. See "Re-
membering 'Early Music,'" in Thesis 8 (Fall 1994), pp. 32-33.
12. See Hiley! p. 538.
34 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

In the mind of Boethius, the sixth-century Roman philosopher and theo-

retician of music, there is no distinction between Roman and Greek music. For

him, it's very clear that the theory of music is Greek. This conception will stay
the norm among most of the medieval theorists; even if some authors are very
far from the Greek original, they will go on using Greek terms in order to look
educated. This community of culture is obvious not only in music but in many
other fields.

After the Gothic domination, Byzantium reconquered Italy in 533, and its

domination, although merely symbolic from the eighth century on, lasted until

Charlemagne's coronation as Western Emperor. During the seventh and eighth

centuries, fourteen popes were Greek, most of them from Sicily, where there
was a very strong Greek community from antiquity until the thirteenth century.

Also, the iconoclastic persecution in Byzantium in 726 sent many of the Greek
religious to Italy. Let us keep in mind also that the Emperor of Byzantium had
the title of Roman Emperor.
Once this has been settled, it's possible to observe how with time the Ital-

ian and Greek churches and chants evolved in different ways. Western, East-
ern, and Greek aesthetics must not be considered homogenous blocks. The dif-

ferent styles found between and within the three Italian repertoires show us that
important diversities existed.
To come back to our subject, which is musical performance, we are in the
same situation as a linguist who tries to find the sound of medieval Latin by
studying today's Romance languages. Of course, for centuries there have been
differences between all these languages and the original Latin. But some words
have not changed, such as sol for "sun" in Spanish, Stella for "star" in Italian.

We could find thousand of examples like them. In music it's the same process.
You find in the Italian repertoires —the Roman, the Milanese, and the Ben-
eventan —some formulas or ornaments that are still living in Byzantine, Syriac,

or Coptic pieces. The process in experimentation is not to imitate the models

slavishly but to use their information to figure out the dynamism of these dead
musics. The common roots of Eastern and Western chant should be studied,
but not in order to prove anything.

One repertoire you've recorded did survive in a living form until our time:
the Milanese " Ambrosian" chant. But in your CD notes you point out that it
was much influenced over the centuries by Gregorian chant, and that originally
it had been sung, according to Ambrose, "in the manner of the East. " You

worked on that, too, with Eastern collaborators.

Yes, there we went further in our experiments. The Milanese liturgy had
roots in the Antiochan ISyrian Christian] liturgy, and at different times in its
early history Milan had been in relation with Syria and even had Syriac bish-
ops. We tried, as a working hypothesis, to distinguish traces of Syriac chant in

the Milanese repertoire. I worked not only with the same Greek singer, but also
 A DIFFERENT SENSE OF TIME 35

with experts in Syriac chant, with the Lebanese singer/musicologist, Sister

Marie Keyrouz, and with the Lebanese musicologist Ehe Kesruani. They opened
another field that I had not imagined at all, because Marie Keyrouz had an-
other approach to music, to modality, and to the value of intervals. From the

beginning, she told me something very important: "This music IMilanese chantl
is a music of intervals." That means you really have to be aware of the value
of each interval, because it's what creates the mode; the ornamentation is there

to throw the intervals into relief. This was something quite new to me. West-

ern musicians, when singing monody, are too little aware of the quality of in-

tervals — but that is what produces the real character or mood of the mode.

Could you give an example?

In the offertory of the Milanese Christmas mass, Ecce apertum est, there's

a mode that alternates Bb, low B^l, and high Bll. When this formula reaches its

highest point on a low Bt?, the A is sharpened. When the formula reaches its

top on the C, the Bti is high. So you can imagine the complexity of the music.
In each formula, you must always discern which note exercises a power of at-

traction that redefines the value of the intervals of the scale.

And regarding the role of ornamentation she mentioned?

Marie Keyrouz is very sophisticated in the art of ornamentation. This is

something we have lost in the West. Even in Baroque music, most singers don't
have enough imagination to go very far in ornamentation. Some jazz players
do it, and some Baroque players, like Jordi Savall, have a freedom in orna-
mentation and a quality of nuance that you don't find in many singers. Singers

like Marie Keyrouz have this knowledge. I think what will be really important
in early music in the next few years will be the progress of singers in the art

of ornamentation. In old traditions that's what made the quality of a musician:

someone was a distinguished musician because he had his own way of orna-
menting.

Ornamentation has been one of the hallmarks of your work, but in some
repertory it has been controversial; for example, some argued that there was

no evidence for what you did in your reconstruction of the Gradual of Eleanor
of Brittany. How would you respond to such critics?
That they should improve their knowledge of the thirteenth century. In fact,

it's from that century that we have the first precise description of ornamenta-
tion, with the treatise of Jerome of Moravia, a Dominican friar. I have man-
aged to do two books on him." As I said, ornamentation is the big lacuna in

13. Jerome de Moravie, un de la musique dans la milieu intellectuel parisien du

theoriticien
Xlle siecle (Paris: Creaphis, 1992); and Jerome de Moravie, Traite sur la musique, ed. Christ-
ian Meyer, trans. Esther Lachapelle, Guy Lobrichon and Marcel Peres (Paris: Creaphis, 1996).
36 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

the early-music revival. Now it's accepted for seventeenth- and eighteenth-cen-
tury music, but not yet for Renaissance and medieval music. I feel like saying
to those who believe people started to add ornaments on New Year's Eve of
1600: Wake up!

One other interesting area you've explored is the use of microtones —tones
that fall between the usual twelve pitches used in the modern Western scale.

You say microtones were used in early chant traditions?

There are two approaches to this problem: the manuscripts, and the oral
traditions that still exist. From the written documents, we know that micro-
tones were known, because a lot of theoreticians talked about them, and we
have at least three manuscripts that refer more or less explicitly to some mi-
crotone practices. To my knowledge, the first mention of microtones in West-
ern writings, after Boethius, are in Remigius of Auxerre (d. c. 900), a Prank-
ish theorist. In one of his texts he uses a Greek musical vocabulary, meaning
that at this time the Greek vocabulary was in use among educated musicians.
He uses this vocabulary to talk about quarter tones and thirds of tones: so
such intervals were known, though we don't know how they occurred in the
music. After that, the next book to give us this information more precisely is

the Montpellier Treatise [copied c. 1100]. In this manuscript you have a dual
notation: one form uses neumes, and one uses letters of the alphabet from A
to P to cover a two-octave range. In this notation you have two ways of sig-

nifying the quarter tone. The two other manuscripts I referred to —a twelfth-
century antiphonary from Utrecht and another one from Cluny —have the
same kind of chromaticism, often at the same places. But as they are just
neumatic manuscripts, they use different neumes' shapes to express these vari-
ations.
Now, all this is useful information, but the problem is how to deal with it.

For some examples we were some correspondence with Byzantine

able to find
or Arabic theory, but for some examples we were not. The latter cases may
have involved things that had disappeared in Byzantine tradition, or that oc-
curred in Latin music only.

Another practice you've taken from Greek tradition in singing Old Roman
chant is the use of ison singing —the use of a vocal drone pedal point—which
is first documented in Byzantine chant in the fifteenth century, but of course
may be older. Many critics find it hypnotic, but nonetheless it is controversial;
how would you respond to critics?
There is some evidence that this practice might come not from the Greek
but from the Latin. As you said, the use of the ison seems to be known in the
Byzantine tradition around the fifteenth century, but not in other Eastern
churches. The first clear description I know of this technique, though, comes
 A DIFFERENT SENSE OF TIME 37

from a Western source, the Micrologus by Guido d'Arezzo in the eleventh cen-

tury. For him it was a sort of organum.'"* He teaches us that this practice was
common in Rome. We know from the Ordines Romani that by the seventh and
eighth centuries there were traditions of organum singing in the pontifical
chapel.'"^ Later the anonymous author of the Summa Musice,*^ a treatise writ-
ten around 1200, describes the sort of organum that consists of a drone. He
calls this manner diaphona basilica: that's very interesting, because the term
basilica in liturgical matters often refers to the Roman tradition. So in the thir-

teenth century there was still in the vocabulary of singers a word that seems to
referred to the Roman Basilican tradition and that means a vocal drone. It is

very possible that the Greeks borrowed this practice from the Italian singers.

We find in some fifteenth- and sixteenth-century Greek sources, written in

Byzantine notations, some instances of polyphony in this style with parallel

fifths and contrary motion. In one manuscript a rubric says, "This is done in

the Italian way." We know that from the thirteenth century the Italians, chiefly

the Venetians, had a very strong influence in some regions like Crete and in
Byzantium itself, where there existed a strong Latin government for almost sev-

enty years. So there is a strong basis for this scenario.'^

But, you know, above all it is important when you make a theory to ex-

periment and see how it works. In this matter, the big question is, Why do we
have so few recorded instances of drone singing? Was it so common that it was
not necessary to talk about Or maybe some people did not consider it a
it?

form of polyphony at all, as is the case today in Greece, so that maybe it was
assimilated into monody. Or maybe it existed in only a few places. But musi-
cally speaking it works, and that helps us to better hear the modal structure of
a piece.

14. "Organum" meant several things, but in this context it meant, in general, singing two
or more related lines, as opposed to just the single line of plainchant. See Sarah Fuller, "Early
Polyphony," in New
Oxford History of Music, II, rev. ed., ed. R. Crocker and D. Hiley (Ox-
ford University Press, 1990), pp. 484 et passim.
Guido describes several organum practices, and he notes that practices varied from one lo-
cale to another. His text has been published in English translation in Warren Babb's Hucbald,
Guido, and John on Music (New Haven: Yale University Press, 1978). The section suggesting
what we call ison singing is on p. 80 (section 211).
15. The Ordines Romani were "Prankish reports of Roman practice." See Richard Crocker,
"Liturgical Materials of Roman Chant," in New Oxford History of Music, U, rev. ed., p. 139.
16. Summa Musice, ed. and trans. Christopher Page (Cambridge University Press, 1991).
The description of diaphona basilica is on p. 124.
17. Peres recommends Michael Adamis, "Some Instances in the Byzantine Manuscripts In-
dicating a Relation to the Music of the West," in Polyphonies de tradition orale: Histoire et
traditions vivantes, ed. Michel Huglo and Marcel Peres (Paris: Creaphis, 1993). On the other
hand, Dimitri Conomos's "Experimental Polyphony in Late Byzantine Psalmody," in Early
Music History 1 (Cambridge University Press, 1982), pp. 1-16, suggests that ison was intro-
duced into Greek singing several decades after any Italian influence which was in any case —
quite "isolated" — had run its course.
38 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

Time and the Nature of Plainchant

Katherine Bergeron writes: "It may be well to ask from the start whether chant
can properly be considered 'music' at all. [It had] an eminently
. . . practical
purpose: to make ritual words audible, memorable, and powerful . . . in this
sense a particular chant hardly different from a spell or incantation, a set of
is

specially pronounced words designed to bring about a certain magical result. "^^
Could you comment on this, and on how that affected both the way chant was
composed and sung in earlier centuries and the way you perform its'
The answer is very simple. To really restore these musics, and clarify their
vocal aesthetics, it's better to reconstruct the liturgies they belonged to and to
believe in what you are doing.
This is what I found, for example, when I started working with Lycourgos
Angelopoulos. He has another way of approaching time. This is because he is

a real church singer. He is used, for instance, to singing for liturgies that go all

through the night. So he really has experience of the pace of the liturgy, and
that's very important. The biggest criticism I would make of many reconstruc-
tions of medieval music is that, listening to them, I don't feel the atmosphere
of the ancient liturgies. I think you must be able to visualize all the stages and
the movements of the liturgy whose music you are singing. To know the East-
ern liturgies can help us, since today the Catholic traditions are almost dead. I

had a traditional Catholic education; and fortunately I work with Syriac and
Byzantine church musicians.

For your concerts, you dress liturgically.

Only for our liturgical dramas. It is really necessary to play dramas because
it's the best way to get into the spatial and temporal dimensions of these mu-
sics, as it's very rare to have the opportunity to perform them in true liturgies.
It is also fundamental to get used to working with candlelight. That was a con-
stituent part of the liturgy. With candles, you have lights around you, but large
parts of the room are darkened. Today, due to electric light, people have lost
the habit of living with the night. When people have lost the habit of living
with the night, the night doesn't exist. To live surrounded by the night gives
you another way of understanding the mentalities of the past. The candle is a
living light as fragile and powerful as human life. It always reminds you that
light is not to be taken for granted; you have to be conscious of it.

Contemporary Catholic clergy have lost the sense of light and sound. Even
in Europe, where we have old churches planned according to the position of
the sun, they use electric lights during the day. For today's Christians that's a
secondary, even nonexistent matter, but in the tradition of the Church of all

centuries it was a crucial point. It's a disaster that the Roman Church aban-

18. "Chant, or the Politics of Inscription," in Knighton and Fallows, Companion, p. 101.
 A DIFFERENT SENSE OF TIME 39

doned the liturgy in Latin, because all the people who were able to transmit
the tradition are now very old, so we'll have a break in the transmission. But
I think Latin could come back to the liturgy because young people have an at-

traction to plainchant . . .

As the monks of Silos found out . . .

Yes, and it's significant that the age range of the customers in Europe was
18-25, because these people did not grow up with the Latin liturgy and plain-
chant. That suggests that the Church made a mistake.

The earliest notation of plainchant had to do with the melodic gestures' mo-
tion, rather than with the exact pitches or rhythms:^"^ the time wasn't notated
exactly.That was because of the oral nature of the tradition; but does it fit in

with your view of liturgical time?

Yes, and that's why we're a little bit lost with these notations. Now we are
used to a mathematical division of time, but we must remember that this has
been true only since the end of the thirteenth century. Before that, people had
no way to write these things down.

Is this why you've been opposed to singing Notre-Dame organum with

strict proportional rhythm?^°
Yes! In the polyphony of the twelfth century, we know from the notation
that this note is longer and that shorter, but not exactly how much longer one
note is than another. We can get a sense of what this might mean from music
we still have that reflects a mentality that treats time differently than we do in

the West today. For example, in Corsican polyphonic singing they don't have
a tempo with a beat, they just have the time of the chords, and when the en-
ergy of the chord starts to diffuse it changes. Time becomes a succession of fo-
cuses of energy each with a period in itself, and almost every chord has its own
period. When you feel the end of the period of this chord you move; it's not
something you can divide arithmetically, saying this chord is two times longer
or three times longer than the last one.
So it seems that there are two ways of perceiving time —qualitative and

19. See above, note 8.

20. Organum at Notre Dame in the thirteenth century seems to have been sung using what
are called "rhythmic modes," codified around 1240, whereby notes had specific durations rel-
ative to each other. There were six rhythmic modes, all in triple meter; they were developed
for singing polyphony, but their interpretation is not entirely clear. (Their application to
monophony is much more troublesome
Hendrik van der Werf writes that a belief in the
still.)

omnipresence of modal rhythm in Notre Dame polyphony "is now waning." He specifically
notes that in polyphonic passages in "sustained pitch" style as opposed to "pitch against —
pitch" style —
the note lengths were probably not modal; his interpretation is congruent with
Peres's (van der Werf, "Early Western Polyphony," in Companion to Medieval and Renais-
sance Music, p. 112).
40 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

quantitative. The quantitative manner began to be created during the end of

the thirteenth century and developed in the later centuries. But the older way
of thinking about music co-existed as well. It's funny, because the trend in all

the spheres of social life and science of this time is to rationalize things. For
example, the first mechanical clock was invented at the end of the thirteenth
century.^'
We did a symposium on this four years ago,^-^ in which we tried to figure

out how in a place like Paris many thinkers in different fields intended to give
a description of time and to find a tool to describe it. In that century many dif-

ferent authors proposed a system, but the systems don't The all fit together.

most exciting of the treatises was that of Jerome of Moravia, who wanted to
put in a book all the musical knowledge of his time. Writing of polyphony, he
says, "Many different authors have their own way of describing the rhythm,
and I think the best thing to do is to present all these treatises and let the reader
make up his own mind." At this time, around 1265, they knew they were on
the verge of reaching something, but it was only in the fourteenth century that

a notational system would be standardized.^^

Even when this notational system was standardized after the fourteenth or
fifteenth century, the old mentality continued to exist. Although the mechani-
cal clock was invented at the end of the thirteenth century, it doesn't mean that
a few years later there was a clock in every home. In the country today, farm-
ers still have to live with the seasons; for them, in the winter and summer 8 o'-
^'^
clock doesn't mean the same thing.

Music is a tool that can help us to better understand history, how human
beings used to be, used to live. It also can help us to increase our sensibility,
our aesthetic sense. When we learn the ancient arts, we start to develop our
sensibilities to be able to perceive more things in the reality of our human re-

lationships and ways of living. Quality of life is one of the most important
things we can learn from people of the past, because one thing we have to learn
from the past until the nineteenth century is that people had a different qual-

21. See David Landes, Revolution in Time (Cambridge, Mass.: Belknap, 1983), p. 53,
though he considers references to clocks from before the fourteenth century a little uncertain.
22. Proceedings published in La rationalisation du temps au Xlle siecle (Paris: Creaphis,
1995).
23. Christopher Page's essay "Ars Nova and Algorism" in his Discarding Images (Oxford
University Press, 1993), pp. 112-39, relates the rise of measured music to scientific trends of
the era, especially the wider adoption of Arabic numerals — but not to the clock. Page also dis-

cusses the issue of time and notation in his interview, and arrives at conclusions opposite to
those of Peres regarding medieval polyphony.
24. I am not convinced that clocks had much impact on the development of quantitative
rhythm; see the Postscript to the Christopher Page interview, where I mention my reasons
briefly. This is not to say that I think Peres is necessarily wrong about rhythm in Perotin or
in Corsican chant; but that the explanation for the change to more quantitative rhythms in
music might lie elsewhere.
 A DIFFERENT SENSE OF TIME 41

ity of life —one that, at most social strata, had certain cultural advantages that
we have lost, in spite of all the technical progress . . .

. . . sometimes because of it . . .

. . . yes, but we make poorer the quality of everyday life. For instance, in
churches, even in Europe, they use microphones for the liturgy. That means we
have lost a quality of hearing and of voice production. The same thing for light-
ing; I talked before about candles. If you light the church with candles, the
mood you create —the quality of the space and time— is really something dif-

ferent and is worth the experiment. There is no reason to lose these things.

SELECTED DISCOGRAPHY
Marcel Peres has made a number of recordings for Harmonia Mundi, and even
readers with little interest in plainchant might find them fascinating. The best
starting points may be the discs that include non-Western singing, especially the
one featuring Beneventan chant (HMC 901476) and the second of those fea-

turing Old Roman chant (HMC 901382). Both feature Lycourgos Angelopou-
los, with his extraordinary microtonal ornamentation and un-Western voice
production; the results in both cases are hypnotic —a word many enthusiastic
critics have applied to it. The same adjective is often applied to his recordings
of other repertory discussed above — e.g., the Milanese chant CD (HMC
901295), which includes Marie Keyrouz.
It would be misleading to imply that the reviews have been unanimous in

praising Peres's chant CDs. One persistent critic has been Jerome F. Weber, who
thinks that many of the performance practices used have no scholarly basis; he
often objects, for example, to the use of drones and of "Eastern" ornamenta-
tion. Reviewing a recent Peres CD of Mozarabic chant, Weber says that it

sounds "less like any other recording of Mozarabic chant (few as they may be)
and more l-ke [Peres's] own recordings of Gregorian, Cistercian, Old Roman,
Ambrosian, and Beneventan, and Neo-Gallican chants." (He says same
this

criticism applies to another wide-ranging director of plainchant, Laszlo Dob-

szay.)^'' Weber is, however, enthusiastic about Peres's recording of the Mass of
Tournai (HMC 901 353), which he thinks is clearly the best realization of that

manuscript.-*'
From the chant recordings, one might turn to Peres's recordings of later
repertoire. These bring up a theme that recurs often in the other medieval and
Renaissance interviews in this book, and even in some of the Baroque ones
English versus Continental singing styles. Fabrice Fitch describes the difference
well in his review of Peres's Ockeghem Requiem (HMC 901441): "The tenors'

25. Weber, Fanfare 19 (November/December 1995), p. 453.

26. Personal communication, 1996.
42 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

emphasis on chest-tone clearly differentiates them from their English counter-

parts. It is as though English ensembles match their lower voices to the high

partials of the choirboy and the countertenor, whereas ensembles like Organum
start from the basses' rich, deeplow Cs and build upwards. "^^ Despite reser-
vations regarding a few pitch standards and tempos, Fitch is enthusiastic about
this recording. Peres interpolates plainchant, treating the Requiem as the Mass
for the dead it was meant
and while Fitch usually experiences such in-
to be,
terpolations "as so many distractions from the polyphony," here he finds them
"literally awe-inspiring." He also calls Peres's recording of Josquin's Missa
Pange Lingua (HMC 910239) "superb."

FOR FURTHER READING

Plainchant has inspired an extremely active and wide-ranging body of research
over the last To do justice to it all in one book would clearly be
generation.
impossible; David Hiley has done the impossible in his Western Plainchant: A
Handbook (Oxford University Press, 1993). This magnum opus is indispens-
able for anyone interested in chant.
I don't know of a better introduction to chant than James McKinnon's chap-
ter, "The Emergence of Gregorian Chant in the Carolingian Era," and Hiley's
"Plainchant Transfigured," both in McKinnon's Antiquity and the Middle Ages
(London: Macmillan, and Englewood Chffs, N.J.: Prentice-Hall, 1991), the sin-
medieval music. For those wanting more musical de-
gle best introduction to

tail McKinnon aims to provide, David Fenwick Wilson's Music of the

than
Middle Ages (New York: Schirmer, 1990) is excellent. All these books also dis-
cuss non-Gregorian chant; those seeking a more advanced discussion might try
the New Oxford History of Music, II, rev. ed., ed. Richard Crocker and David
Hiley (Oxford University Press, 1990), Part II.

Katherine Bergeron's essay on the nature of plainchant in Companion to

Medieval and Renaissance Music, ed. Tess Knighton and David Fallows (Lon-
don, Orion, and New York, Schirmer, 1992), quoted above, is a high point in
a very stimulating book. Her essay on Peres and the Silos monks, "The Virtual
Sacred" (The New Republic, 27 February 1995, pp. 29-34), has been contro-
versial partly, I think, because it has been misread: her real subject is not early
music, but what the chant phenomenon says about spiritual life in the 1990s.
David Landes's Revolution in Time (Cambridge, Mass.: Belknap, 1983) is a
first-rate history of the clock and its impact on Western civilization. Paul
Fraisse's "Rhythm and Tempo" Diana Deutsch's The Psychology of Music
in

(New York: Academic Press, 1982) summarizes a wealth of research on time

perception in music.

27. Early Music 11 (February 1994), p. 155.

 2
You Can t Sing a Footnote
— NoOo^
Susan Hellauer on Performing
Medieval Music

Medieval composers rarely expected their sacred music to be listened to for

its own sake. They designed accompany church services events of solemn
it to —
meaning for medieval worshippers, but not for modern concert audiences. On
top of that, they set texts with little appeal or resonance for most modern
listeners. For these reasons, their music translates to the modern concert hall

with difficulty. Anonymous 4 have been unusually successful in this act of

translation; my discussion with Susan Hellauer focused on how the group ap-
proaches it.

We also discussed their extraordinary popularity, their appeal to a "cross-

over" audience whose usual interests do not include medieval motets and se-

quences. Popularity, the group told me, was something they had neither sought
nor expected. At the time of the interview, February 1994, their first two CDs
were bestsellers, and they had appeared on Garrison Keillor's radio show and
in an interview in USA Today, a publication not known for the height of its

brow. But it wasn't until a few months later that the four singers, all veteran
performers of medieval music, felt secure enough to at last become full-time

musicians. Their musicianship deserves no less; but what exactly was its large-

scale appeal? Their pure, celestial sound might be at least a factor. I discussed
that sound, too, with Hellauer.
Historians writing about American beliefs at the end of the second millen-
nium may note that in 1994 at least a score of angel books were published,
that two of them made the New York Times bestseller list, and that angelic im-

43
44 MEDIEVAL MUSIC, PLAINCHANT, AND " OTHERNESS "

ages adorned T-shirts and college dorm rooms, side by side with posters of
death-rockers/ All of this, too,may be relevant to understanding the wave that
Anonymous 4 has caught; as a German critic wrote of them, "They sing like

angels."
Offstage, these singers are wonderfully down-to-earth —and humorous, as

suggested by their group name (though the name is, as our interview shows, a
meaningful one, it is on
actually a play the designation given by modern schol-
ars to a medieval theorist whose name is not recorded). I met with Susan Hel-
lauer on the morning before an Anonymous 4 concert, and began by asking
her about their new-found (and still baffling) celebrity.

An English Ladymass reached number three and On Yoolis Night reached num-
ber one in the Billboard charts. [So did, since then, their third release. Love's
Illusion; it was Billboard's "Classical Disc of the Year" for 1994, as Ladymass
had been for 1993.] You've said that this took you by surprise; why do you
think it's happening?
We may not be the right people to ask! Perhaps the answers lie in the hearts
of our listeners, and in the minds of the record company who keepexecutives,
their fingers on the popular pulse far more than we do. An audience member
at a recent concert suggested that our performances create a space for con-
templation, an opportunity often lacking with the frantic nature of most daily
routines. Much of our repertoire is contemplative in nature. Perhaps there's
something in that explanation ... In any case, we are mystified and delighted
at the response to our recordings and performances.

Many people think that the repertoire you sing, and also the music of Arvo
Part and John Tavener, taps into New Age spirituality just like angel books, —
for example. Have you or Harmonia Mundi USA done anything to reinforce a
connection to these things?
We have never tried to tailor either the repertoire we sing or the way we
sing it to any particular audience. As an ensemble, we have not attempted to

connect ourselves to New Age spirituality. Whatever our listeners' religious or

spiritual leanings may be, there would probably be a common belief among us
that creating and appreciating beauty are food for the spirit. As for Harmonia
Mundi, they give us free rein in terms of choice of repertoire and its perfor-
mance, and they fully support our musical decisions. We leave the marketing
to them.

Much of your music was written for a liturgical context, but now it's lis-

tened to in a concert or on a recording. Those are very different listening con-

texts —we go to concerts not to worship in a service that's supported by music,

1. "Angels Everywhere," The New York Times, editorial, 4 September 1994.

 YOU can't sing a footnote 45

hut primarily to listen to the music. How do you translate music written for
one context to others that are very different^
It's a very complex issue, one that has been on our minds from the very
start in 1986. There are several possible solutions. To illustrate one extreme,
some ensembles present medieval music in the tradition of the nineteenth-cen-
tury song recital. They wear formal dress. They begin their concerts by bow-
ing to the applause of their audiences. They sing a number of pieces, receiving
applause after each work or "set." Their concerts include an intermission, and
may be followed by one or more encores. Some groups use this format very
convincingly. The other extreme is the complete reconstruction of, say, a litur-
gical service with everything but the priest — as, for instance. Ensemble Or-
ganum does so superbly. (Of course, the liturgical drama is a different category:

the music, its order, and the dramatic continuity at least are provided within
the structure of the work.)
One of our first ideas about how we were going to structure our programs
came from an uneasy feeling we had all occasionally had when hearing or per-
forming in concerts of this music presented recital-style. These might be pro-
grams of wonderful music, well performed, but they consist of a succession of

beautiful miniatures, often with little connection beyond their all being from
more or less one time or possibly from one place. After about the tenth or fif-

teenth similar piece, it can be almost impossible to absorb any more. And when
the pieces share little similarity, it can be hard to concentrate, or to come away
with a feeling of completeness. From the very start, our most important objec-
tive has been to make stylistically cohesive programs, each built around a the-
matic concept, but including enough internal contrast to show off each work
to its best advantage. This might occur naturally in a liturgical service, but we
didn't feel that the answer for us was in going all the way to complete recon-
struction.
The development of each of our new programs is a major effort much —
more work, we believe, than making a recital program. Each one needs quite
a long gestation. We begin with the concept: a particular manuscript (the music
for Love's Illusion comes from the thirteenth-century French Montpellier
Codex); a historical liturgical practice (An English Ladymass is an evocation of
the thirteenth- or fourteenth-century votive mass to Mary as it might have been
sung at the Cathedral of Salisbury); or a portrait of a historical or legendary
figure (The Lily and the Lamb is a depiction of the intensely personal suffer-
ing of Mary at the foot of the cross). Once we have decided on the theme for

a particular program, we have to find and select the music. We look for both
continuity and variety in the music, and for musical texts which help to illu-

minate the theme in question.

Some of those texts involve concepts modern people can relate to devo- —
tion, romance, suffering— but others are more difficult. An extreme case is
when a medieval text is objectionable to us in some way. I'm thinking of the
46 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

anti-Semitic passages in medieval and Renaissance sacred music and in chant;

and secular songs that treat wife-beating or rape as a huge joke. How do you
deal with such texts?-
Depending on the situation, we've dealt with it in several ways. For exam-
ple, texts about rape (or attempted rape) do occur in the motets from the Mont-
pellier Codex, the source for our program Love's we had
Illusion. Luckily, so
many other pieces illustrating aspects of courtly love that we could simply ig-

nore the most repulsive ones. However, the four of us didn't always respond
identically to what I would call "borderline" texts, which denigrate women in

various ways but don't go all the way to force or violence. It was a matter of

negotiation, based on the shape and requirements of the program, and on our
own feelings. I don't think any of us would sing a song trivializing rape or wife-
beating.
We've also come up against anti-Semitic texts. If a work is strophic and the
anti-Semitic text is in isolated verses, as in Perotin's monophonic conductus
Beata viscera, we omit the offending verses in performance (although we
haven't yet found a solution for recording such a work). Where the anti-Semitic
text consists of only a few words, we actually change the Latin and explain in

our program notes what we have done.

The way we choose to handle these problems is very personal; we wouldn't
presume to mandate it for anyone else. Of course, we recognize that these pieces,
with texts that are in some way objectionable to us today, are important his-
torical documents, reflections of their times. But we don't want to contribute in

any way to a pervasive sense that women are things or that hate is an accept-
able way of life.

Getting back to the program-development process: once you've decided on

the theme and chosen what next?
the music,
We use either our own good modern editions, and prepare
transcriptions or
the music to our satisfaction. The order of the pieces and their key relation-
ships are crucial, as are variations in texture and voicing. We complete the de-
sign with poetic or prose readings, which add a sense of the unfolding of a nar-

rative, bringing the program to about 75 minutes in length. Our programs have
no intermission, and we request that our audiences not applaud between pieces.

The more like a story than a concert. This story-like

result feels to us function,
with continuous communication and dramatic flow, is really at the center of
everything we do.

2. This topic was discussed by Lawrence Rosenwald, "On Prejudice and Early Music,"
Historical Performance 5 (Fall 1992), pp. 69-71, and by a number of respondents to his essay,
including Barbara Thornton and Richard Taruskin, in the pages that followed and in the fol-
lowing issue of Historical Performance (Spring 1993). Taruskin has since written about it else-
where, notably in "The Trouble with Classics: They're Only Human," New York Times, Sun-

day Arts and Leisure section, 14 August 1994, pp. 25 and 31, reprinted as the title essay in
his Text and Act (New York: Oxford University Press, 1995), pp. 353-58.
 YOU can't sing a footnote 47

Of course, your earlier observation is correct about liturgical polyphony not

being meant for concert performance as we know it. There are some similari-

ties, in that in the medieval church service there was an audience listening (at-
tentively, one presumes), undistracted by food, conversation, or other pursuits.
But there are three principal differences between medieval sacred music and
music intended for concert performance: context, anonymity, and textural
rhythm.
First of all, the medieval audience for liturgical polyphony was listening in
a very different context than most modern audiences. As you said, these lis-

teners were there primarily to worship and only secondarily to hear music (al-
though there undoubtedly were exceptions!), and the meaning of the particu-
lar feast being celebrated created a context and emotional framework for the
music. In the U.S., at any rate, only one Christian feast still has this kind of
power, and that is Christmas. If you grew up in —or in close contact with
Christianity, you know what strong emotions a hymn like Adeste fideles can
create on Christmas Eve. But hearing that hymn on a hot summer day is like

finding shoes in your refrigerator —completely out of place. In the Middle Ages
there were many feasts besides Christmas: each one had its identifying music
that was heard only once a year and that had tremendous emotional power.
We feel that it is our job, through the selection and arrangement of musical
texts and poetry or narrative, to create something approaching that emotion-
ally powerful context for the music we sing.

The second point of difference acknowledgment for the me-

is the lack of
dieval composers/performers themselves: what we would call anonymity. Me-
dieval church musicians composed and sang as do their modern counterparts;
but it was soli deo gloria [to God alone the glory]. This is why so few of their
names have come down to us. And the singers were not supposed to be the
center of attention in a liturgical service (although there were certainly instances
of church singers in the Middle Ages receiving reprimands for their tendency
to show off).^

We try to evoke this sense of anonymity by structuring our programs in a

continuous flow, and by avoiding extensive solo singing, suppressing applause

between pieces, and refraining from encores, all elements which tend to draw
the audience toward the performers and away from the narrative. The only star
in each of our programs is the program itself.

The third element of difference is the naturally occurring textural contrast

or rhythm in a festive medieval liturgy. We frequently use liturgy — both Mass
and Office — as a structural framework; we find that it helps to provide us with
the right balance of continuity and contrast. Along with varied textures and
types of polyphony during the celebration of the feast, there was also plenty of

3. References to vocal "excess" by clerical singers in the thirteenth century can be found
in chap. 6 of Christopher Page's The Owl and the Nightingale (Berkeley: University of Cali-
fornia Press, 1989) and in his interview, below, p. 74.
48 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

chant — in fact, the service was mostly chant —and it varied greatly, from the
simplest psalm tones, prayers and readings to long, elaborate, and virtuosic
graduals and responsories. And so from the beginning we were determined not
to neglect chant, either in the amount we include in our programs or in the at-
tention we give to its performance.
Even an all-chant program needs textural contrast. While developing our
program of music of Hildegard von Bingen, we found that alternating her
strongly expressive, pungent works with plainer chant in something like their

original liturgical order greatly heightens their effect and, we think, comes
closer to reproducing the startling first impression they must have made. We
think of Hildegard's virtuosic chant, or virtuoso polyphony, as exquisite
gourmet dishes. As wonderful as they are, we wouldn't think of serving them
alone; we prefer to serve them with good plain rice to clear the palate and act
as a foil to their complexity.

/ wanted to ask you about your performance of chant. For one thing, if one
doesn't know anything about chant, one assumes that it's easy to sing, whereas,
oh, Wagner and Berg are hard to sing. But people who sing chant assure me
that it's difficult both to understand and to sing.

The assertion that singing chant is easy is based on the assumption that
increasing vertical and harmonic complexity in music represent definite

progress along a continuum from primitive to civilized, from immature to ma-

ture. The truth is that the medieval chant from western Europe that has sur-

vived, and that we can decipher, is the flowering of a long and marvelous de-
velopment, and its masterworks are as subtle and artfully crafted as any
polyphonic work could The means of communication are different, but
be.

the performance of these works deserves and requires no less skill and devo-
tion. On average, it takes us longer to prepare a substantial work of chant

than it does a polyphonic work. We must first pay close attention to nuance
of verbal accentuation, musical line, and phrase relationships; only then can
we begin to make music — but it is music of the highest level, very exposed,
very intimate, very rewarding. There is a kind of ego submersion that is nec-
essary as well —to create something that becomes greater than the four of us,

yet remains part of each of us. It's a mystical thing in its own right, bring-

ing together four very different voices and four very different personalities,

so that in any particular chant, the listener hears a sound that is all of us
and none of us. Singing chant is great training for the kind of ensemble con-
sciousness that we strive for when we prepare and perform all of our reper-
toire, both monophonic and polyphonic.

That could be part of the popularity of chant maybe it appeals to those —

who want to experience something communal, as opposed to isolated individ-
ualism. Anyway, there's another issue: expressiveness or not in singing chant?
 YOU can't sing a footnote 49

Aha! There's a word

— "expressiveness." We throw it around easily, Hke the
word "progress," but what does it mean?
really Singing expressively can mean
different things in different vocal repertoires. The uninflected delivery of a tra-
ditional Scottish ballad singer telling of blood-curdling deeds is just as riveting

as the vocal intensity of Tosca pouring out her anguish.

There are some who believe that it is inappropriate to respond to word
stress, accentuation, and meaning when singing chant. However, we feel that

our individual emotional responses to the music and texts of chant contribute
to the ensemble's artistic interpretations. For us, the key to expressiveness lies

in definite nuances in rhythm and dynamics in response to the text and its

meaning, and to word stress and musical line. In a leaderless group like ours,

it's the response that's tricky, because no two people (let alone four) respond
to a musical line or a text in the same way. Working out the nuances that re-

flect our emotional and artistic responses to a work is just as important as

working on pitch and rhythm. We hope that comes across."

Some people have remarked to us that our singing seems much more
and emotionally involved than do recordings of chant performances
flexible

by monks or nuns, and they ask what theory of chant performance we fol-
low. We don't subscribe to any particular modern theory, but we are proba-
bly closest in spirit to one of the earliest leaders of the Solesmes chant re-
vival, Dom Joseph Pothier (1835-1923), who favored a free, oratorical rhythm,
based on speech. It was the later Solesmes-movement leaders who devised the
basically equalist, "ictus"-ized rhythmic guidelines, in an effort to create uni-
formity of performance.' In fact, the earliest Solesmes editions from the 1880s
don't contain the ictus and other phrase marks so characteristic of the mod-

4. For a somewhat different view of expressiveness in chant, see Christopher Page's inter-

view, pp. 79-81.

5. Solesmes was discussed in detail in the Marcel Peres chapter. One enormous challenge
Solesmes faced was reconstructing the rhythm of early Gregorian chant. The Solesmes edition
of 1905, which set the monastic chant-singing style for much of this century (as heard in the
recordings of monks and nuns that Hellauer mentions), included a number of rhythmic indi-
cations; these included accents to mark the "ictus" —that is, in this usage, the first pulse in

what Solesmes understood as groups of two of three pulses. As Hellauer notes, these indica-
tions are now considered misleading and without historical justification. In the last forty years,

new scholarship — particularly that of a Solesmes monk, Dom Eugene Cardine — has led to

some advances. (See David Hiley's Western Plamchant: A Handbook lOxford University Press,
1993], pp. 373-85, for a detailed discussion of rhythmic notation of chant.) Among the
somewhat discredited are the mensuralists, who argue that "the notes
schools that have been
of chant have various fixed time-values, or that some system of strictly proportional mea-
surement is and (as Hellauer notes) the equalists,
to be applied to the performance of chant,"
"who hold that all notes of Gregorian chants are more or less of equal duration so that there
is a single basic time-value" (the definitions quoted are by Mary Berry, in her "The Restora-

tion of the Chant and Seventy-five Years of Recording," Early Music 7 [1979], p. 217). Car-
dine's views are similar but not identical to those of the equalists: "a theory of syllabic equiv-
alence, relating the length of notes to the normal delivery of a syllable with a single note"

(Hiley, p. 382).
50 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

ern Solesmes books —with a few exceptions, these markings are not at all me-
dieval.

This brings up a broader authenticity issue. Scholars need to doubt their

convictions or at least test them against the evidence; but performers, to really
perform well, need complete conviction, beyond all doubt. To reach such con-
viction, they have to go beyond the evidence. How do you deal with these is-

sues as artists performing music of the past, music that today is so involved
with scholarship?
Since it's impossible to know what medieval music originally sounded like,
no one can ever claim that they know exactly how to perform it. Scholars can
argue and argue; they may even throw knives at each other early in this cen- —
tury,one French musicologist challenged another to a duel over whether or not
troubadour music should be transcribed in rhythmic modes.*" But since our job
is to present this music in performance, we are forced to make musically viable
choices. As I often say, You can't sing a footnote.

That's a great motto —but could you say a

more about it? bit

Well, I don't mean to imply that the performers can do without the schol-
ars; they need to work together in order to make informed decisions about per-

formance. We've had important help from several outstanding scholars, and
they've been as generous and understanding as they have been meticulous in
their work.
The facts that scholars have uncovered are the firm foundation on which
our imaginations can build. But sometimes the facts are insufficient to complete
the picture, or scholars disagree about what they mean. Then we have to make

choices that convince us, and that will convince our audiences. Our working
goal is very simple: to gather the sources and the available interpretations of
the facts and, with those in mind, to come to an agreement about how to pre-

sent each piece. Then each of us can express whatever it is she wants to ex-
press within the boundaries we've set. Each piece is very carefully worked out,
but it has to sound effortless in performance. That's where the hours and hours
of rehearsal come in.

It shows. For one thing, your intonation is amazing.

We'll never stop working on it. We don't follow a particular theory of tun-
ing, but we do have to take care in tuning certain intervals.

6. Actually, the two musicologists, Jean Beck and Pierre Aubry, both agreed that the trou-

badours had used the modes (six specific rhythmic patterns used, at least sometimes, in thir-
teenth-century Parisian polyphony); the argument was over which of the two had thought of
it first. Ironically, it is now felt that the troubadours did not use the rhythmic modes. The duel

never came to pass: while practicing his fencing, Aubry was fatally wounded. It should be
noted that in France in 1910 literal dueling was still possible; today's musicologists stick to

verbal swordplay.
 YOU can't sing a footnote 51

In the full triads found in later music, there is a wider acceptable range for
the placement of the harmonic third. But for medieval repertoires, we need to

adjust the size of our thirds according to the melodic mode. And in twelfth-

through fourteenth-century polyphony, in which many of the nodal points (such

as cadences) consist of an open fifth (with no third), the intonation bull's-eye

is very small.
Accuracy of tuning is especially crucial in pieces that are very dissonant,

such as thirteenth-century French motets. We sometimes sing ten beats in

a row with seconds or sevenths occurring vertically, and if we're not perfectly
in tune they all sound like mistakes; but when they're right, they sound
really spicy. It's similar to some twentieth-century music, where the intona-
tion has to be very crisp and clean so that sense can be made out of disso-
nance.
Singing in tune is both an individual and an ensemble skill. In order to bring
our tuning to its finest point, we have to examine the way in which our indi-

vidual voices work together. For example, we often have to modify one per-
son's pronunciation or shaping of a vowel sound so that even our overtones
will match. ^ That's where more hours of rehearsal come in!

Could you discuss one issue that musicologists disagree about: the role of
women in the music of this era? Scholars have argued, for example, over
whether or not women sang fourteenth-century English sacred polyphony.^
Women in convents probably sang chant just as much as men in monaster-
ies did. Hildegard of Bingen and her community certainly provide evidence of
that. We don't know whether or not women sang sacred polyphony; but since
much of this music is for equal voices (and since we can sing it), we think it's

possible, and even probable, that they did. Of course, women were not wel-
come to sing in cathedrals (unless they disguised themselves as men, as they did
for other reasons). With regard to secular music, there are many paintings and
illuminations, as well as literary references (Boccaccio's Decameron, for one),

that depict women participating both as singers and as instrumentalists in the

high Middle Ages.

7. Christopher Page writes about a related issue: "In every language there are dark vow-

els and bright vowels, and the dark ones must often be sung slightly sharp in order to avoid
giving an impression of being more than slightly flat. (The radical vowel in British English 'fa-
ther' is an example of a dark vowel; it accounts for the flatness which is often heard when
performers of plainchant sing the final note of a melisma on the word 'AUeluya')": in Per-
formance Practice: Music Before 1600, ed. H. M. Brown and S. Sadie (London: Macmillan,
1989, and New York: Norton, 1990), p. 84.

8. For example, Frank LI. Harrison argued that a certain fourteenth-century English motet
was sung by nuns, but Roger Bowers has strongly disputed this point. See Bowers's "English
Church Polyphony, c. 1320-c. 1390," in Studies in the Performance of Late Medieval Music,
ed. S. Boorman (Cambridge University Press, 1983), p. 188. In any case, the narrow pitch-
range of medieval music, discussed in Christopher Page's interview, allows a quartet of women
to sing it without revision.
52 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

Apart from historical do you think there are any purely musical ad-
issues,

vantages to having polyphony sung by women's voices rather than by men's?

One of our reasons for starting Anonymous 4 was to experiment with the
sound of this music as sung by higher voices. And some of our hsteners have
commented that the individual lines of polyphony sound more clearly etched

in our voices than in lower voices. We've been told that this might be explained
by the fact that low voices singing polyphonically produce more acoustical
combination tones in the audible range than do our higher voices. But, of
course, the study of acoustics is not our area of expertise!

Clearly the women's voices appeal to audiences. So many of your reviewers

say that you sound like angels. Care to comment on that?
The acoustical phenomenon may help to explain it. Also, since its revival in

this century (until quite recently), it's been outside the norm for women to sing
medieval music. Of course, in real life we're not exactly angels.

SELECTED DISCOGRAPHY
Since their first release, Anonymous 4 has been recording new CDs for Har-
monia Mundi The releases so far have been consistently fine.
at least annually.

Of their first. An English Ladymass (HMU 90 7080), David Fallows writes,

"There may be some question of whether it is historically appropriate to sing
the English polyphony of the years around 1300 with women's voices only; but
that is not the point. What is clear is that some of it sounds magical that way
on their record. . . . This is because the higher voices help to clarify the tex-
ture in a way that pays enormous dividends for complicated music. "^ As an ex-
ample he mentions the four-voice motet Salve mater redemptoris, "where the
details and textures are far more transparent than in any all-male performance

and the music comes across with an irresistible swing." Most critics have
greeted the group's other CDs with similarly warm praise.

The only criticism of the group I've encountered, in fact, is Tess Knighton's
concern that in The Lily and the Lamb (HMU 90 7125) "the harsh and dra-
matic nature of the texts is here too often belied by the beguiling perfor-
mances. "^° She says the group provides "the ethereal qualities of The Tallis

Scholars . . . further disembodied." Knighton singles out ]esu Christes milde

moder as the only piece that conveys "the emotional and physical asperity of
the Crucifixion." Nicky Loseff says of the same performance that it "evokes
the pain of the crucifixion . . . intensely." But she considers this typical rather

than exceptional, saying that in general "it is this direct connection between
poetic meaning and musical interpretation in polyphonic items which most sep-

9. Fallows, "Quarterly Retrospect," Gramophone 71 (December 1993), p. 33.

10. Knighton, Gramophone 73 (September 1995), p. 98.
 YOU can't sing a footnote 53

arates Anonymous 4's approach from that of other high quaUty ensembles.""
The two critics' disagreement may reflect nothing more than differences in sub-
jective response; but it may also reflect different ideas about the nature of ex-
pressiveness in medieval singing. Hellauer discusses this in her interview, and
the issue returns in the Christopher Page interview.
One item in the group's discography is an example of the crossover, men-
tioned in the introduction, between medieval/Renaissance performers and con-
temporary composers. Richard Einhorn's oratorio Voices of Light (Sony SK
62006) uses Anonymous 4 to collectively represent the voice of Joan of Arc.
The piece refers to early music in terms both of idioms (e.g., parallel organum
and plainchant) and of instruments (viola da gamba).

FOR FURTHER READING

A good introduction to the English medieval repertory —which Anonymous 4
has recorded on three of its discs — is found
David in the relevant sections of

Fenwick Wilson's Music of the Middle Ages (New York: Schirmer, 1990). More
detail can be found in chaps. 9 and 14 of The New Oxford History of Music,

II, rev. ed., ed. Richard Crocker and David Hiley (Oxford University Press,
1990). More detail still can be found in John Caldwell's The Oxford History
of English Music, I (Oxford University Press, 1992).

11. Losseff, "Anonymous 4, an Ensemble Apart," Early Music 24 (February 1996), p. 176.
 3
Vox Feminae
— oOo^g^

Barbara Thornton on Hildegard

of Bingen

Hildegard (1098-1179), the founder and abbess of a Benedictine nunnery in

Bingen, Germany, first came to the attention of modern America in a book

about headaches. Oliver Sacks's 1970 book Migraines included an essay' ar-
guing that Hildegard's mystical visions were "indisputably migrainous," al-

though with characteristic open-mindedness he has also written that this "does
not detract in the least from their psychological or spiritual significance."^
What is significant for my discussion, though, is that in describing Hildegard's
"exceptional intellectual and literary powers" —shown in her work as, among
other things, a mystic, poet, naturalist, and playwright — Sacks made no men-
tion of her music. This oversight was by no means unusual; even many music
historians in 1970 knew little about Hildegard.
Today Hildegard is known chiefly for her music. What brought about the
change was a pair of 1982 CDs, Gothic Voices' A Feather on the Breath of
God, and Sequentia's Ordo Virtutum. Both still stand among the best-selling
recordings yet made of early music. Several factors have been suggested for
the discs' popularity, among them Hildegard's appeal to the rising interest in
women's spirituality, the feminist search for great women composers, and —
factor we have already encountered — the quest for transcendence, by the same
audience that has since made best-sellers of the monks of Silos and Anony-

1. The essay, "The Visions of Hildegard," is reprinted in Sacks, The Man Who Mistook
His Wife for a Hat (New York: Summit Books, 1986), pp. 166-70.
2. Ibid., p. 130; the comment comes from Sacks's introduction to the section of the book

that contains the Hildegard essay, and appHes not only to Hildegard but to others discussed
in the section.

54
 vox FEMINAE 55

mous 4. All these factors intertwine, no doubt, with an essential one: Hilde-

gard's musical and poetic genius. No one can reasonably claim that her works
need any special pleading based on her gender.
Hildegard's "everflowing, rapturous outpouring of melody"' seems unusual
even to those who know little of her contemporaries' music; according to Bar-
bara Thornton, who knows a great deal of such music, Hildegard's musical lan-
guage surpasses that of her contemporaries in "intensity and breadth."** Thorn-
ton's own contributions to the Hildegard revival include scholarly articles and
the reconstruction of the Ordo Virtutum (The Play of the Virtues), an allegor-
ical music-drama, which she has not only recorded but has also published,
staged, and had filmed; above all, they include her performances, which con-
vey Hildegard's vision with special fervor. Thornton is now recording Hilde-
gard's complete works, a process to be completed in time for Hildegard's 900th
birthday in 1998.
Thornton, an American-born soprano, began working on medieval music
with her partner, Benjamin Bagby, in 1974, when the two were graduate stu-

dents at the Schola Cantorum Basiliensis in Switzerland. In 1977 the pair

founded Sequentia, which performs a wide range of medieval music, from Oc-
citan troubadour songs to Old English sagas to early polyphonic motets. The
group is based in Cologne, about 75 miles (as it happens) from Bingen. Our
discussion of Hildegard reveals Sequentia's general approach to preparation: an
uncompromising immersion in the music, language, texts, milieu, and cultural
as well as musical sources. It includes using only the original medieval nota-
tion, and memorizing everything performed, for reasons that become clear

when Thornton speaks about orality in medieval culture.

I interviewed Thornton when she was touring with Sequentia's ensemble of

five women. Vox Feminae, in a program of music by Hildegard. Their concert

the next day was in San Francisco's Grace Cathedral, a Gothic structure with
a reverberation time of about six seconds. The architecture was appropriate
the Gothic style emerged during Hildegard's lifetime —and with its stained-glass
light reflected Christendom's turn toward subjective, personal religious feeling,
which finds such strong expression in Hildegard's music. The acoustics proved
appropriate, too: while a twelve-part Renaissance Mass would have been lost
in the echo, Hildegard's "ecstatic devotions" (as the program notes called them)
seemed all the more celestial.

was innovative and transitional the era

Hildegard's era, the twelfth century, —
when Europe began to rediscover Aristotle, when modern scientific thought
began to emerge, when Christian piety became more personal and emotional

3. Mary Berry, Gramophone (May 1995), p. 94

4. Thornton, ''Vox Feminae: The Ecstatic Devotions of Hildegard von Bingen," (program
notes, 1993), p. 3.
56 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

as opposed when the Virgin Mother really took a central place

to sacramental,
when the idea of "courtly love " emerged, the seeds of
as a figure of devotion,
humanism were sown, and so on.^
In many areas of thought, certainly in those of Hterature and music, the
eleventh and twelfth centuries demonstrated an enormous burst of energy
which modern writers have often called a renaissance of sorts. ^ However, I
think it is now accepted that "renaissances" are actually very carefully prepared
in previous times: for example, the great surge of so-called Marian worship,
which came to pervade every aspect of twelfth-century art, had been gaining
momentum since the ninth century.
Musically, the twelfth century seems to me in a way the apogee of the ar-
chaic mind. In mind has not yet become technological in our modern
it the
sense, but has reached an enormous imaginative flowering according to pre-
existing rules. In this sense, twelfth-century music fits a certain modern ap-
proach to the idea of what an avant-garde is: in its search for innovation and
new expression, it takes even the most archaic forms and uses them to make
new music. Our era has become interested in so-called primitive ways of life

and of thinking —African art, modal music of all varieties, shamanistic tradi-
tions, etc. — as it has, in a sense, exhausted the rationalistic way of doing
things; our era has also discovered that there is an attractive radicalism in
so-called archaic world views.
The musical/poetic spirit of the twelfth century entails both tradition and
innovation. It is very forward-looking and avant-garde, because creators con-
sidered their own creations as utterly new; but the older rules of operation were
never abandoned. There's an insistence in certain musical and poetic reperto-
ries on the idea: "This is nova cantica," "this is new song." Of course nova
cantica is a term that comes from the Bible, in several different contexts, but
it also reflects the sense in which we today say "this is new music" — this is the
latest, this is the hottest, this is the newest, this is refreshing, this is going to
wake up parts of your spirit that have been asleep. And yet there is no sense
at all of "to do this we have to cut off from what came before." It's rather very

5. For more thorough discussion, see the relevant chapters in Norman Cantor's introduc-
tory history.The Civilization of the Middle Ages (New York: Harper CoHins, 1993), notably
pp. 306-56. To quote another writer, A. C. Crombie, it was in the twelfth century that "men
of philosophic temperament began to turn away from the vision, given them by St. Augustine,
of the natural world as a symbol of another, spiritual world, and to see it as a world of nat-

ural causes open to investigation by observation and hypothesis." In Crombie, Science, Op-
tics, and Music in Medieval and Early Modern Thought (London: Hambledon Press, 1990),

quoted in Christopher Page's Discarding Images (Oxford University Press, 1993), p. 11.
6. See Christopher Brooke's The Twelfth-Century Renaissance (New York: Harcourt, Brace

and World, 1970); Renaissance and Renewal in the Twelfth Century, ed. R. L. Benson and G.
Constable (Oxford University Press, 1982); and Dom Jean LeClerq's The Love of Learning
and the Desire for God, 3rd ed., trans. C. Misrahi (New York: Fordham University Press,
1982) on monastic life in the period. Finally, the "dark underside" of the twelfth-century re-
naissance, the increased persecution of groups such as the Jews, is discussed in R. I. Moore's
The Formation of a Persecuting Society (Oxford University Press, 1987).
 vox FEMINAE 57

much in the spirit of this era to imagine a tree of Jesse, coming from deep roots
and spreading infinitely from its fertile origins.

How did Hildegard, with her huge range of accomplishments, fit into
this era?

There were not many people like Hildegard in her era. There were some
other "universal geniuses," however. One, for example, was the theologian and
poet-composer Peter Abelard, who added a great deal of original work to the
tradition and was an accepted though controversial master. Another was the
poet-theologian Alan of Lille. Hildegard can easily take her place among those
two. She's also been compared to the Arab philosopher-encyclopedist-healer
Avicenna 1980-1037]. ~ She was well known in her period, though not as well
known as Abelard or as widely read as Alan of Lille. But unlike Abelard, she
took the necessary steps to ensure that she not fall into disfavor with the higher
political powers of her day and that her position stay secure within the Church.
Over time she also gained official recognition of her gifts as a prophetess and
visionary. Without that official recognition, she probably would have had more
clashes with authority than she did have. In fact, toward the end of her life the
prelates of Mainz decreed a temporary ban on music in her cloister, in response
to a perceived disobedience on her part.** But in general, she knew which im-
portant steps to take to ensure her creative freedom. If she had come under the
shadow of serious suspicion, the authorities would presumably have cut her off
very quickly.
In Hildegard we also see someone who, within that secure position, saw to
it that her works were documented. She seems to have had scribes of the high-
est caliber, copyists with great expertise in music notation, even very gifted il-

lustrators; she was obviously in a position to call on the best people. In this
she resembles composers of later centuries such as Guillaume de Machaut and
Oswald von Wolkenstein, poet-musicians who also supervised the creation of
manuscripts of their works.

Could you say something about her background?

At the age of eight, she entered a Benedictine double cloister —a monastic
community consisting of men and women, where all labor and forms of wor-

7. By Perer Dronke, Women Writers of the Middle Ages (Cambridge University Press,
1984), p. 144. He also compares her to Goethe.
8. "Many wealthy people" buried family members at her convent. One such burial, in

1178, was of a nobleman who had been excommunicated, though end he was recon- in the
ciled with the Church. The prelates of Mainz alleged, however, that he had died excommuni-
cate, and they ordered Hildegard to remove his remains or be excommunicated herself. She
refused, and also defied (and challenged in writing) their ban on music in her convent. Her
nuns chose to remain with her and thus be excommunicated as well. During this episode, the
Archbishop of Mainz was in Italy mediating between the Pope and the Emperor Frederick
Barbarossa; in 11 79 the Archbishop made it possible for the ban and excommunication to be
lifted. See Dronke, pp. 196-99.
58 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

ship were carried out equally but separately —called Disibodenberg, near to her
birthplace. This cloister must have been very aristocratic, and intellectually of

very high standards. For most of her early life, Hildegard was cloistered alone
with an older woman, Jutta of Spondheim, who taught her in isolation for
many Her education must have included a formidable body of religious
years.
knowledge. Apparently, this woman was also her teacher in areas such as heal-
—
ing what we would call medicine —
and in what we would call natural sci-
ences. A recent article by the medieval scholar Peter Dronke** identifies some of
the more obscure writings that Hildegard must have known when she was
putting her own writings together. He concludes that she had one of the better
educations available at the time, ifwe can speak in such terms. Perhaps the
terms "privately tutored" and "self-educated" might be appropriate for de-
scribing how she came to command such a breadth of knowledge and origi-
nality of thought.
Dronke in this article refers to some very esoteric works known to Hilde-
gard. She seems to have had access to some quite progressive and up-to-date
material whose Latin was rich and full of images, as was fashionable then.
More came from her lifelong familiarity with the Bible and
general influences
its commentaries, and the writings of the Church fathers. In addition, twelfth-
century thinkers were drawing on much Platonic literature, and on many
sources of music anddrama from Antiquity. Perhaps the most pervasive influ-
ence of came from religious rites at the convent, and all the types of con-
all

temporary music and poetry that would have been heard in a well-to-do, main-
stream monastic institution of the time.

So this cloister presumably had a large library, and a source of new books?
Yes; but it seems, according to newer assessments of medieval intellectual
life, such as those of Mary Carruthers and her school, ^° that in this era books
were crucial to the tradition but were not used manner we are accustomed in a

to. A book, as a repository of thought and wisdom, was almost a symbol of

the intellect's ability to store words in memory so that they could be used at

will for reasoning, speaking, and composing. It has been crucial for us, dealing

with all periods of the Middle Ages no matter how advanced, to be sure that
we understand that basic premise. It was common that in one's education the
available works of Plato and other philosophers, the Bible, commentaries, and
other favored works, were interiorized through active memorization more than
just being read off the page. An
would have memorized using cer-
intellectual —
tain systems for facilitating the process —
vast amounts of knowledge in his life-
time. Therefore learning, certainly of music and literature, took place mostly
on an oral basis.

9. Dronke, "Platonic-Christian Allegories in the Homilies of Hildegard von Bingen," in

From Athens to Chartres: Neo-Platonism and Medieval Thought, Studies in Honor of
Edouard Jeaneau, ed. J. H. Westra (Leiden: E. J. Brill, 1992).
10. Carruthers, The Book of Memory (Cambridge University Press, 1991).
 vox FEMINAE 59

How does this orality apply to Hildegard?

Hildegard wrote some very intriguing things about her own processes of
musical creation. As you know, she claims to have heard her songs directly
from the Spirit with her "inner ears," although (she explains) she did not
"know neumes" — meaning she did not know how to notate music. Some mod-
ern writers have taken this statement to be an expression of false modesty on
her part, or a way of strengthening her claims to having "mystic" or direct con-
nection to divine inspiration, which would seem to use her ignorance as a
medium for reaching her senses. I have no way of knowing in exactly what
spirit she makes the claim of "seeing" and "hearing" Divine Light. I do not

think it beyond imagining that she would assume a humble personality as a lit-
erary device, or attempt to create a mythology out of her gifts and works in
order to give them authenticity. She may, indeed, have created an acceptable
persona for her peers, in order to remain in a position to continue her various
enterprises. In any case, what is significant from the practicing musician's point
of view is that her claim is, in fact, a plausible statement from any accom-
plished musician of the period. The realms of theorizing about or notating
music were separate from that of actually producing sounding music. In both
cases we're dealing with skills of a highly specialized nature reserved for very
few people. In general, the class of the musically active was not musically lit-

erate (this situation changed over the next few centuries).

Once you take that premise of orality seriously, it changes very funda-
mentally all your opinions and feelings about the evidence that comes from
the period. In particular, people had interiorized their knowledge to such an
commentary or contribution of one's own to the body
extent that adding a
of knowledge was already an enormous step. For us today it's easy to say
that an intellectual contribution should be above all "original," but for twelfth-
century thinkers originality would be expressed differently. Probably the ma-
jority of one's intellectual life consisted of exchanging and transmitting knowl-
edge with other individuals who had the same sort of deeply stored textual

information; exchanging ideas on this material was already one vital level of
intellectual creativity in the period. To take the next step, then, and docu-
ment in written form one's original opinion or commentary upon the basic
body of knowledge would imply that a very creative contribution was being
made. While to us it may seem that commentators of this type were adding

minimal amounts of information and new thought, it is actually because we

approach "knowledge" with a fundamentally different philosophy from intel-

lectuals of that period. We tend to "project" knowledge into books; they

stored it, had it readily available, used it to build their souls. When we see

representations of monks laboriously copying books and adding their com-

mentaries, we are actually in the presence of those people who were recog-
nized as masters in their tradition.
In musical situations as well, the risk, the audacious step, of adding
melodies to the existing canon was one taken only by a master.
60 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

Regarding Hildegard, what was the body of existing music that she was
working from?
Her compositions are strictly religious, and seem to be very much based on
the ideasand traditions of Gregorian chant; they take the elemental ideas, such
as modes, formulas, and subject matter, and widen the scale of the whole ex-
perience both technically and emotionally.

So Gregorian chant is the tree of Jesse you spoke about at the outset?

That would be a way of putting it, yes. To people who are involved in the
notation and realization of Gregorian chant, tenth- through twelfth-century
new music must seem very simple, actually. The complexities in notation and
in theology and literary exegesis are heavily concentrated in Gregorian chant;
twelfth-century pieces add a certain transparency, certain "performance quali-
ties" if you will. Such pieces are usually intended to embroider central events
in religious services, be it a Mass or an Office. This embroidery creates new
forms and attitudes, and gives an outlet for contemporary feeling and thinking.
On the level of real composition and experience, our pieces are elementary
compared to Gregorian chant —they were intended to be that way. Anyone
who's interested in the molten core of this whole process needs to look at Gre-

gorian chant, its notation, its history, its performance, its repertoire. Then the
newness of our music becomes apparent.
From modern writers we might get the impression that there is something
simple and cute about medieval melody; this is a fundamental misunderstand-
ing. There enormous power implied in these kinds of modal melodies, but
is

only with a great deal of study can a modern person add the imagination nec-
essary to understand what is implied and to realize the intended effects. Al-
though we do have direct evidence about instrumental music —we know there
was an art and tradition of arranging notes without texts —the heart of me-
dieval tradition is liturgical, which is preserved as an untouchable and unalter-
able repertory of chant. As a composer, you didn't try to change that repertory;
you spent your Hfe trying to realize it. Then, if you wanted to entertain your-
self and your friends and have literary/musical experiences of a new sort, you
might, based on this intense relationship to certain themes, add individual
arrangements and pitches and so on, but drawing from an enormous reservoir
of communally interiorized associations. Modal music lives from these kinds of
associations, just as language does.

So you're talking about associations with specific Gregorian formulas and

melodies from a standard repertoire?
Yes. All modal musics seem to enjoy this fluidity, this protean quality which
allows a gesture to take on one meaning within one context and a different one
in the next context. People are writing very interesting things in ethnomusico-
 vox FEMINAE 61

logical circles" about how modal or proto-modal music systems work which
shed light on these processes. In our experiences in Sequentia, we have found
that a modal gesture does not unfold uniquely on the "horizontal" level as a
melody might. A modal system is much more a matter of defining some very

simple constructive principles, attaching meaning to those simple elements, and

'~
then learning to combine them significantly.
Now, the modal system is always significant, but it's relatively undog-
matic despite its elemental laws. It is also open to exceptions (here too it is

a lot like language and grammar). But one must build up associations to the
elements and combinations in order to understand the next level of signifi-

cance.
Now, I don't suppose there is an essential difference in this respect between
modal and tonal music. I suppose they could be considered ancient and mod-
ern versions of the same language, the only difference being that the average
listener has built up these associations only for tonalities and not for modali-

ties. As a result of our lack of associations, though, when we hear a modal fig-

uration we make an inner judgment as if it were a melody. Because we're so

busy with the melody on the horizontal level,we can't appreciate the very spe-
cific messages of modes. Each gesture shows a new understanding of how to

combine elements and colors. If we don't relate to the fundamental colors, we

won't hear the individual shadings.

What do you mean by color f You don't mean timbre, obviously.

I don't, you're right. The "color" is the emotional effect of the relative
placement of pitches in a modal gesture. It means, for example, that intervals,
especially imperfect ones,'^ are never "absolute" but always derive their sig-
nificance from their modal context. On the other hand, the actual pitches of
the gamut (the known spectrum of pitches) do have some "absolute" associa-

1 1. Leo Treitler cites the ethnomusicologist Robert Ridgely Labaree's studies of troubadour
songs, which suggest applying to plainchant ideas developed in the study of Irish music. See
section VII of Treitler's "Sinners and Singers: A Morality Tale," Journal of the American Mu-
sicological Society 47 (Fall 1994), pp. 137-71.
12. See Leo work, for example "Centonate Chant," Journal of the American Mu-
Treitler's

sicological Society 28 (1975), pp. 1-23. Treitler differs with some predecessors who argue that
chant composers put together pre-existing formulas into an artistic whole; he says instead that
as all these chants were memorized, it was natural that certain "well-defined moments in the
progress of chants (openings, cadences, continuations after cadences, settings of words with
particular syntactical functions such as conjunctions)" would tend to become standardized
within a genre of chant, and that what was in between would be less fixed (I quote from "Sin-
ners and Singers," p. 149).
Regarding oral cultures, see Walter Ong, Orality and Literacy (London and New York:
Methuen, 1982), chaps. 2 and 3.
13. The "perfect" intervals are the octave, the fourth, and the fifth; all the rest are "im-
perfect."
62 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

tions. The composer plays with both levels when he sets up an arrangement of
modal gestures.
Not surprisingly, I think one's real associations to modes develop through
practice, not through descriptions such as I've tried to give. Medieval modal
music, like other modal musical cultures, is a very practice-oriented tradition.

You can't learn to associate from music theory books, or from schemes that
show modes in a scale, because a mode doesn't exist in a scale. Presenting a
mode in a scale is the worst thing you can do to it. It would be like writing a

treatise on painting by saying. Here is the row of pigments — red, orange, yel-
low, green, blue —now you understand how to combine them. It doesn't tell

you anything about, for example, how modal colors relate to emotions.
This is part of why text is so important to our music, for the emotional
world of the text constitutes the fundamental color of a musical composition.
We have also found some old treatises, such as ninth- through eleventh-century
handbooks contemporary ways of look-
for cantors, to be helpful in revealing
ing at "Gregorian" modes.'"* The authors of such books developed a tradition
of terms and affects associated with the several modes. '^ For example, what we
call Dorian mode —
the D-oriented or "protus" mode for them had the quality —
of gravitas (it was grave in the sense of solemn) and also had the association of
nobilitas, nobility. It was called fons et origo, the source and origin of all other

modes. If you superimpose these ideas upon a string of pitches in this mode, you
also get a concrete result. Superimposing upon the music the emotional idea of
a text then represents another level which must be interiorized in a given piece.

You mentioned text as a color. It's sometimes asserted that in medieval

music the text was primary, and the music decorated it; and sometimes the op-
posite is claimed.^^ How do you apply this question to Hildegard?
I think that in making music of any sort, but especially in singing religious
music, the process which results when text and music combine cannot be ana-
lyzed in this way at all. I can't see that one element in this process could possi-
bly dominate the other. Leo Treitler, whose writings have influenced our work,
uses the term "text carrier" or "text vehicle" for medieval music. That usage
has been interpreted as relegating music to a subservient role; but I think that
perhaps demonstrates a certain lack of understanding of how medieval people
thought of "word." Word (logos) is practically the Tao of Western civilization.

That fits in with what you've been saying about its being an oral tradition:
Walter Ong writes that oral cultures commonly, and probably universally, con-

14. See Harold Powers, "Mode," III, 1, in the New Grove Dictionary of Music, vol. 12
(London: Macmillan, 1980), pp. 397^01.
15. See Christopher Page's interview, p. 80-81, for a more detailed discussion of this.
16. See Page's interview for his views —very different from Thornton's —on text expression
in medieval music.
 vox FEMINAE 63

sider words to have enormous power,^" though that implies the primacy of the
word—as in "in principio eratverbum," in the beginning was the Word.
But ascribing great power to words doesn't necessarily grant them primacy.
It is hard to convince modern people of one basic truth: that the medieval

poems are music in and of themselves. This idea is fundamental to the classic

definition of poetry
— "words set to music" (with or without pitches). It de-

pends on how intensely you relate to the poetry to know just how the music

of a text operates. So to say, as some people do, that the word comes first and
the music comes second is an absurdity, because to the medieval person the
concept of music is an all-encompassing idea; it's the most divine thing a human
being has at his or her disposal. It's even more divine than Word, because it

reaches above human words to superhuman word, and it's thoroughly inex-
plicable in its origins and effects (although it obeys certain laws). In the mind
of a medieval person, it comes as close as anything can to giving an image of
what cosmos and soul are like. Augustine's treatise De Musica deals with

music, as such, in terms of texts because, he says, one experiences the propor-
tions of all types of harmony more clearly through texts. This was the accepted
way of looking at music in the Middle Ages. So to combine the music of word
and the sounding music of pitch represents an extreme potency in and of itself;

we ought not try to pull it apart and say which should come first, word or
music, because it puts us in the wrong frame of mind.
Moreover, a medieval cleric heard and sang vast amounts of music in the

Latin language in his lifetime. Clerical musicians experienced music and text in
the highly integrated form of Gregorian chant, and it formed the psychic back-
ground of their own music-making. So the question of which comes first would
be meaningless to them. They'd have lived with the power of chant all their

lives.

Regarding Hildegard's own use of the word, Peter Dronke writes, in the

liner notes to your Symphoniae CD, that her "poetic effects are often strange

and violent, " not smooth of most of her contemporaries, and that
like those

her anaphoras, superlatives, exclamations, "daring mixed metaphors, " and in-
tricate grammatical constructions make her Latin stand out in her era.^^

All I could add from the performer's point of view is just how effective these

metaphors and images (and even the strange things she attempts) really are. In

his article about her influences, Dronke traces the origins of some of these ob-

scure images. He quotes a passage where Aristotle says that the merit of a poet
lies in his metaphors. A more pragmatic statement might be that the merit of
a poet is tested by how much life is in the images he or she creates. One cri-

17. Ong, Orality and Literacy, p. 32. On p. 93 he discusses this with reference to soci-

eties where writing is restricted to certain sectors, as in Hildegard's time.

18. Symphoniae, Deutsche Harmonia Mundi (DHM) 77020-2-RG; booklet, p. 9.
64 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

terion of poetic merit from the memorization and thetreatises of Antiquity

Middle Ages is making it suitable

the vividness of a given allegory or image,
for memory. Their theory of memorization had to do with making one's soul
receptive to the energy of imaginative material. A creative person was an image
maker, and the effectiveness of a creation could be judged by the relative ease
or difficulty with which one could interiorize the images. If they were easily in-
teriorized and seemed to live within one, both immediately and over the long
term, then perhaps one could say that the poet was very good. Without ques-
tion, Hildegard is such a poet-creator-musician. You never tire of her images.
They're very specific, very intricate, very alive!

Could you give an example?

One might be her imagery for Wisdom from her antiphon O virtus sapen-

tiae: "You power of Wisdom/ that circled circling/ and embracing all/ in a

course that is filled with life I You have three wings:/ one soars into the
heights,/ another has moisture from the earth,/ the third flies all around."''

As that example may show, the ways her images are chained together al-

ways has an extremely organic logic that the body and the mind are more than
willing to follow, and there's something so musical and pleasing about the
process of interiorizing these images. Of course, the music facilitates that. It

bring these image-experiences into the self in a very specific —modally spe-
cific —way. Everything is crafted to the utmost, so that one is very quickly in a
position to live what's being composed. I think that's a good criterion for judg-
ing a poet-composer.

You said that in music she derives her procedure from the tradition of Gre-
gorian chant, but expanded it. Could you give an example?
Remember, Hildegard was by no means unique in having based her flights
of imagination upon the stable repertoires of her experience. It is the power
and scope and degree of surety in her flights that contributed to her individ-
ual, recognizable musical style, and this is extremely rare in the Middle Ages.
As an example, one might look at the (presumably) tenth-century antiphon
Alma Redemptoris Mater and a composition of Hildegard's called Ave Maria,
O auctrix vite. Her piece could be called a composed improvisation upon all
the elements found in the older liturgical piece —
its mode, its message, its in-

dividual gestures, its overall curve and construction. One can truly appreciate
the inexhaustible source of invention which was Hildegard's when one sees how
^°
skillfully she has embroidered upon the original material.

19. The piece is recorded on Sequentia's Symphoniae CD.

20. These two pieces are presented in succession on Sequentia's CD Canticles of Ecstasy,
DHM 05472-77320-2.
 vox FEMINAE 65

Regarding the element of musical craft, in describing how you developed

the Ordo Virtutum, one thing you mentioned was that to Hildegard different
—
instruments represented different things the strings earthly striving, the harp
heavenly blessedness, the flutes the presence of God. You used this information
in scoring the piece.

That was based on her other theological works, which were very helpful to
our preparation because she expresses herself so vividly and repetitively, so —
that certain ideas gain stable associations in the mind. And her feeling for music
is prevalent no matter what she's writing about.

We still take the poetic license offered to us by her poetic vision of seeing
everything, including instruments, in terms of the divine scheme. She gives very
specific information about how she thought various aspects of music affect the
soul. This is completely in keeping with other things you can read from the pe-
riod, that all disciplines — rhetoric and grammar, arithmetic and so on — are in-

tended for the enrichment of the soul. Therefore, instruments, by their very na-
tures, communicate diverse modes, emotions, and symbolic associations, in

addition to being embodiments of musica instrumentalis.

We don't use instruments nearly as much now as we did when we recorded
the Ordo, which was over ten years ago. Instruments were very helpful for
singers not initiated into modal thinking, and we're a little more secure in our
feeling for modes. Instruments —or should I say instrumentalists — realize these

things so quickly and clearly. Instruments stimulate imagination: they are very
potent, and they should be used as judiciously as possible, so that the word
and the modal gesture, whether realized by voice or instrument, remain the
main sources and means of impregnating the imagination.

That brings to mind Hildegard's mysticism; does it influence your approach

to her music?
One must probably take seriously her accounts of how she had visions and
heard music; she says that her compositions are not, strictly speaking, her

works, but that she functioned as a medium for them. I do take that seriously,

even if I don't always understand what it might actually imply. Today the cre-

ative personality is made the object of a cult and is exalted as such; in her era
that concept of the solitary creative genius hadn't developed to such an extent.
Her claims of being a medium are not really that different from what many
people today are saying about the creative process generally, but hers are ulti-

mately bigger claims.

Sometimes one is tempted to feel she is justified in claiming "mystic" in-

spiration — her music is so perfect in certain ways. She's a first-rate composer

by any criterion of composing. This is not an instance of having to bend around
the issues because she's a woman or because we are dealing with music from
the remote twelfth century — hers is first-rate music by any definition or
 "
66 MEDIEVAL MUSIC, PLAINCHANT, AND " OTHERNESS

method. Many things in her upbringing explain the coherence of her concepts,
but if mystical inspiration is how she explains or defines her personal genius,
then by taking it seriously we are also taking the religious content of her po-
etry seriously when it isn't second nature for us to do that at all. I don't know
anyone today whose own history prepares them for the kind of detail and in-

tensity of her religious experience. In that respect, her claim has helped us put
ourselves in a position to receive this music as we might not have done other-
wise —as we do have a rather secular way of looking and inspira- at creation
tion nowadays. She made sure with her explanations of her creative process
that we not secularize her works.

Dronke writes in those notes about her world view, which was definitely
not secularized.
In all eras, there are probably prevalent, archetypal views of the world that
people generally share with varying degrees of consciousness, and then there
are learned world views which might be built up on a system of central and
secondary texts or belief systems. The learned world view in the Middle Ages
might take various forms but seems to have had as its basis a synthesis of sa-
cred disciplines and systematically rational ones (which they called "scientia").
One author who influenced this synthesis greatly was Plato —and he came into
the medieval spheres of knowledge through various channels. For example, a
Platonic-Boethian theory of music is based on a vision of the cosmos in which
three "musics" are recognized: a music of the spheres, or harmony in the heav-
ens {musica mundana); an unexpressed harmony above that, the harmony of
God; and below it the harmony of the soul (musica humana). Sounding music
(which medievals called musica instrumentalis) brings these together in a man-
ifestation, so that the soul which "hears" it becomes "symphonic," expressing
itself in the inner accord of the soul and the body; thus music-making is both
earthly and heavenly.
In our day, this cosmic scheme seems like an archaic myth. Some call it the
myth of Timaeus, from Plato's dialogue of that title.-' But that's where the term
"myth" falls short. In the twelfth century, people didn't think of this scheme of
things as myth; they thought of it as reality, just as we consider it reality that
we turn a light switch and have the lights go on, or that the earth goes around
the sun. For them, the harmony of the celestial spheres and its mirror on earth
in the soul was just as real as what we call scientific fact.

Some people say that medieval people had a degree of faith in their religious
beliefs that went beyond most modern believers' experiences.

21. According to David Hiley, medievals knew the Timaeus in a reworking by the fourth-
century writer Calcidius; see Hiley 's Western Plainchant, pp. 443^4. The formulation of the
three types of music came into medieval thought from Boethius, the Roman statesman and
philosopher (c. 475-524), in his book De institutione musica.
 vox FEMINAE 67

Well, textbooks call the Middle Ages the Age of Faith, but even that term
doesn't explain the phenomenon well. Even the word "faith" has something
Protestant about it. We are all post-Reformation people, and also post-Inquisi-
tion people. The Church as an organization discredited itself over the centuries.
It hadn't yet reached that turning point in the twelfth century, so there was lit-

tle necessity to isolate "faith" from any other phenomena in the intact world
view of that period.
We live in an era where things have become much more relative. We em-
brace thousands of different realities in a given day. Consider all the styles, pe-
riods, and types of music we listen to in a given day, all the different traditions,
epochs, and experiences we confront through TV or film or books. In a sense,
there are elements to living within an intact world view that we will never un-

derstand.

How does her world view influence your work on Hildegard?

For me, she was a very important gateway to all medieval thought. Her in-

tensity and immediacy, and that quality to her work that one could identify as
feminine, made it a lot easier for me to come into a profound medieval expe-
rience. Working on Hildegard's music is demanding, and that is part of the in-
tensity of the experience, part of the reason one says, in retrospect, "I think I

really lived through something doing this." Yet we don't necessarily have time
to think about these things when we're performing it. It is extremely hard work
to sing her music: not in the sense that it is drudgery, but because we must
know where to put our concentration at all times. Her music provides a strong
incentive to learn the craft of medieval song. And that's also a reason why she
was a pivotal experience for me, because she helped me identify my craft and
make such a full identification with it that a lot of other issues fell away af-

terwards.

You spoke about the feminine elements of Hildegard; her spending her life
in an abbey was presumably a basic part of this? I think of texts where she ex-
alts virgins, an important theme in her work.

You know, I spent a lot of my early life among women, in school and at
home, and I think I do have an inkling now of how female religious commu-
nities might function. I've been doing professional work with women for quite
a while, and I really enjoy it. I can easily identify with what Hildegard is aim-
ing for in writing music to be sung by women. There is something normal and
everyday about valuing women — if you spend your time with women, you
come to understand their special natures. I suppose it's the same for any man
who spends his life married to a woman; at a certain point womanly things be-
come treasured and yet "everyday." Hildegard von Bingen has been able to ex-
press certain of these treasured feminine things through the themes she chooses:
Mary, praise of the virgins, stories and meditations based on the Saint Ursula
68 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

legend, Wisdom literature, and so forth. From Hildegard, I've learned a lot
about women's musicality, women's relationship to their own spirituality. It
sounds trite, but it is worthwhile to seek out "feminine Divinity" in addition
to the all-pervasive Father principle; feminine spirituality is something very nat-
ural, uncomplicated, and yet intense, in my experience.

Is it just her themes, such as the ones you mentioned, or is it something be-
yond that?
Certain poetic "paintings" of the idea of womankind, even in the way she
composes around it, are very lovely and revealing of her profound feelings
about womanhood. Look at pieces like Quia ergo femina or O tu suavissima
virga. She uses standard medieval images like comparing women to flowers, but
the distinction of her style is in the exalted way she does it. She makes it easy
to feel high and sensuous things, and, based on these good feelings, she will try
to insert grander ideas about archetypal womanhood as well (drawing on es-
tablished Judeo-Christian tradition), so that suddenly imagining something like
a beautiful Mother principle governing the world becomes conceivable to the
singer, where it was not conceivable before. Such ideas had previously been pre-

sented in contexts I absolutely could not accept.

This is most of history, women were

often spoken of as an era when, as in
subjugated; and people like Norman Cantor have applied that idea to Hilde-
gardr^ But some argue that this was also the era in which the seeds of the even-
tual recognition and emancipation of women first emerge the adventure-ro- —
mances like the Arthurian cycles that were written in that century put women
and feminine qualities in a more positive light than before.-^
I would certainly want to distance myself from Cantor's statement about

Hildegard's creative motivations being frustrated and rebellious. There's some-

thing glorious about the twelfth century that you can feel in many male writ-
ers as well as female writers (of which there are surprisingly many):^"* an ad-
miration for women's spontaneous musicality and spirituality. Many spiritual
men were able to recognize, pay homage to, and aspire to that quality without
feeling any conflict. Their homage is couched in specific terms. For example, in
our program tomorrow, one section is formulated with the feeling and vocab-
ulary of the biblical Song of Songs, which has been a pivotal text for musical
composition in all eras (after all it's the song of songs). The work can be seen
as a dramatization of what is yin and yang in the universe, of how beautifully

22. Cantor, The Civilization of the Middle Ages, p. 356. He argues that Hildegard's cre-
ations were "a form of women's revolt against the male-dominated society. Confined and frus-
trated . . . she single-handedly created an alternative culture in her imagination ... [in which]
religious women
have a special claim to articulate God's word in the world."
23. Cantor, pp. 354-56.
24. See Dronke, Women Writers of the Middle Ages.
 vox FEMINAE 69

they move together in nature and in reaHty, and how desperately it hurts the
soul when they're forced apart. These were conscious issues in the twelfth cen-
tury, and the fact that this text served as such an important basis for theolog-
ical discourse as well as for lyric art says a lot about the imaginal quality of
the period.

SELECTED DISCOGRAPHY
Andrew Porter, in a review of a Sequentia concert, asserted that "singing more
beautiful than theirs is not to be heard today" and followed this with a cata-
logue of virtues that is worth quoting at length: "pure, steady tone from open
throats; supple movement through the phrases; pure intonation, with precise
intervals; verbal force and clarity, with vowels distinctly colored; rhythmic live-

liness serving at once the sense of the word and the musical structures." He
concluded, "lT]his was a demonstration that bel canto — beautiful, eloquent
singing that by its sound can move listeners to rapture — is an art not lost."^^

One might compare Porter's list with one that Christopher Page gives, near the
end of the next chapter, of his own desiderata for medieval singing; both the
overlaps and the differences are instructive. Sequentia attempts to express the
text more in its singing style; that, plus the fact that it uses instruments more
(although, as Thornton says, they use them less now than they once did), has
sometimes caused controversy in the British musical press. ^^
Of the group's many recordings for Deutsche Harmonia Mundi, the most
relevant to this chapter are, of course, those of Hildegard. Their 1994 Canti-
cles of Ecstasy (DHM 77320) gives a good idea of what I heard at Grace
Cathedral. It is described as the first in the projected complete Hildegard se-
ries, which suggests, Jerome F. Weber writes, that "the ensemble recognizes how
far it has come since their" early 1980s Hildegard recordings —though some re-

garded their Symphoniae (DHM 77020-2-RG, rec. 1982-83) as having sur-

passed other available Hildegard recordings at the time of its release. Weber
praises the "ecstatic interpretations" in Canticles and in the second CD of the
series, Voice of the Blood (DHM 77346-2).-^ In her Gramophone review of
Canticles (May 1995, p. 94), Mary Berry praises the singers' ability to "give a
real shape and meaning" to Hildegard's soaring phrases, and adds, "Their vocal
quality is very much what I would like to think Hildegard herself would have
expected from her —
own company of nuns firm, unwavering, exultant."
Sequentia's many other CDs display their wide range of sympathies. A sense
of this range can be gained from Vox Iberica, their three-CD set (DHM 77333)

The New Yorker, 11 October 1990, p. 100.

25. Porter, "Reanimations,"
26. See, David Hiley's review in Early Music 13 (November 1985), p. 597, and Ben-
e.g.,

jamin Bagby's essay, written in response, "Musicology and Make-Believe?" Early Music 14
(November 1986), p. 557.
27. Weber, Fanfare 19 (May/June 1996), pp. 167-68.
70 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

of Hispanic repertoire, since it covers many medieval styles: polyphony, im-

provised polyphony, monophony, popular music, troubadour songs, and so
forth. The discs are available singly as Sons of Thunder (DHM 77199) for

men's ensemble. Codex Las Huelgas (77238) for women's ensemble, and El
Sabio (77173) for mixed ensembles. El Sabio is probably the disc to start with
if you buy them separately; Ivan Moody especially praises the "marvellous 'Bul-
garian' folk sound (and, indeed, harmony) of the women's choir in the incan-

tatory Sobelos fondos do mar."^^

FOR FURTHER READING

A good biography of Hildegard is Sabina Flanagan's Hildegard of Bingen: A
Visionary Life (London: Routledge, 1989). Hildegard's own words are avail-
able in Barbara Newman's Symphoniae (Ithaca, N.Y.: Cornell University Press,
1988) and Joseph Baird and Radd K. Ehrman's translations of The Letters of
Hildegard of Bingen (New York: Oxford University Press, 1994). Newman's
Sisters of Wisdom: Hildegard's Theology of the Feminine (Berkeley: University
of California Press, 1987) provides a different approach. Norman Cantor's Me-
dieval Lives (New York: Harper, 1994) contains an enjoyable short story fea-
turing Hildegard, though many people, including Thornton, might take issue
with it.

Peter Dronke's work on medieval poetry includes some important discus-

sions of Hildegard. See especially chap. 6 of his Women Writers of the Middle
Ages (Cambridge University Press, 1984), and also his Poetic Individuality in
the Middle Ages (Oxford University Press, 1970).
Regarding the effects of oral tradition on plainchant — its structure, creation,
and transmission —the pioneering work in English Leo Treitler. As I
is that of
write, this has not yet appeared in book form but is spread over many journal
articles; the best summary is "'Unwritten' and 'Written' Transmission of Me-
dieval Chant and the Start-up of Musical Notation," the Journal of Musicol-
ogy 10 (1992), pp. 131-91. Peter Jeffery's Re-Envisioning Past Musical Cul-
tures: Ethnomusicology in the Study of Gregorian Chant (University of Chicago

Press, 1993) is an important book, especially for its voluminous bibliography.

Two critical reviews of the book have been Treitler's "Sinners and Singers: A
Morality Tale," Journal of the American Musicological Society 47 (Fall 1994),
pp. 137-71, in essence a defense of his work against Jeffery's, and Edward
Nowacki's review in Notes 8 (March 1994), pp. 913-17.

28. Moody, Early Music 21 (August 1993), p. 491.

 4
Tne Colonizing Ear
— oOo^
g^

Christopher Page on
Medieval Music

Christopher Page first ruffled the early-music world in the late 1970s, when he
(and, independently, the American scholar Craig Wright) put forward a radical

hypothesis. It held that the instruments popularized by Noah Greenberg,

Thomas Binkley, and David Munrow — their shawms, rebecs, nakers, and the
rest —were modern impositions; that medieval polyphony was usually per-
formed with voices taking every line and with little or, more often, no instru-

mental accompaniment; and that even monophonic music may often have been
sung unaccompanied.' Audiences and performers had come to love those
shawms and rebecs, and scholars had supported their use, so it's not surpris-
ing that the idea was hardly welcomed except,— significantly, in Britain, with
its wealth of cathedral-trained singers.

1. "Going Beyond the Limits: Experiments with VocaHzation in the French Chan-
Page's
son, 1340-1440," Early Music 20 (August 1992), pp. 447-59, begins with a review of the ev-
idence for this hypothesis. See also David Fallows's crucial paper "Specific Information on the
Ensembles for Composed Polyphony, 1400-1474," in Studies in the Performance of Late Me-
diaeval Music, ed. S. Boorman (Cambridge University Press, 1983), pp. 109-59; Page's "The
Performance of Songs in Late Medieval France," Early Music 10 (October 1982), pp. 441-50;
and Craig Wright's "Voices and Instruments in the Art Music of Northern France during the
15th Century," in Report of the Twelfth Congress {of the International Musicological Society]
(Berkeley, 1977), ed. D. Heartz and B. Wade (Kassel: Barenreiter, 1981), pp. 643-49. Espe-
"Secular Polyphony in the Fifteenth Century," in Performance Prac-
cially useful are Fallows's

tice:Music Before 1600, ed. H. M. Brown and S. Sadie (London: Macmillan, 1989, and New
York: Norton, 1990), esp. pp. 203-12, and in the same volume. Page's "Polyphony before
1400," esp. pp. 92-99. See also Reinhard Strohm, The Rise of European Music (Cambridge
University Press, 1993), pp. 357-58.

71
72 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

The musicologist Howard Mayer Brown, reviewing Page's early recordings,

dubbed the new hypothesis "the English a cappella heresy." In the last fifteen
years it has overcome its heretical status and become the majority position
among scholars. Not that all performers have given in: they tend to fall,

roughly speaking, into two camps, instrument-abstainers (mainly British) and

instrument-users (often American or Continental, e.g., the Newberry Consort
and the Ensemble PAN). The latter camp seems, if anything, to be gaining
strength —although it also seems to be using instruments more selectively than
before the a cappella ideal was mooted.
Page is far less absolute about that ideal than are some of his followers; he
told me that he is sure that all options were used in the Middle Ages, but that
a cappella performance was a common and sometimes standard one. And he
doesn't consider instruments a sort of enemy —he
is also a lutenist and an im-

portant scholar of medieval instruments. was Page who published the a

Still, it

cappella hypothesis first, and Page who has been the most influential in estab-
lishing it. This was in part because he is also a distinguished performer. As the
leader of Gothic Voices, he has recorded a great deal of medieval repertoire in
the a cappella style, and those recordings, more than any documentation, have
proven how well the approach can work.
Page is one of the most influential of musical scholars, but he is not a mu-
sicologist. He is a philologist who teaches Anglo-Saxon, Middle English, Old
French, and Latin in the English department at Cambridge. The combination
of philological expertise with a direct, deep involvement in music has been cen-
tral to his work. It has allowed him to look at musicological issues in a fresh

manner. It has focused him on literary evidence that had been ignored, and has
allowed him to read that evidence skillfully. It has prompted him to look at the
field of musicology itself from an outsider's perspective.
It has also prodded him to consider, perhaps more than full-time perform-
ers would, how the agendas of scholarship conflict with (or support) those of
performance. His published thoughts about the relations between performance
and scholarship —particularly his views on the English choral tradition and its

relevance to early vocal music —have evoked new controversy, making some
performers hopping mad, as we discuss late in the interview. Another of his
suggestions has galled some scholarly critics: that performing medieval music
today can tell us something about musical experience in the Middle Ages. This
idea was almost certain to be controversial, as Page explains, given the acade-
mic trends of our day. The fact that my other interviewees said almost nothing
about these trends is symptomatic of how far the concerns of academics and
non-academics have diverged. But in this case the academic concerns brought
out some core issues of historical performance.

In your writings I find clues to what the medieval experience of music might
have involved. Could you discuss this?
 THE COLONIZING EAR 73

It's a treacherous issue, but I'll try. For one thing, the music of Latin Chris-
tendom probably gave its listeners a clear awareness that a piece, whether
monophonic or polyphonic, was not occupying the full range of frequencies
that it could occupy. The compass of most medieval pieces lies within two oc-
taves. Listeners probably sensed that a piece they heard had the hue of a cer-

tain area in the larger spectrum of usable pitches. In other words, when they
heard a piece they were aware that it could be sung significantly lower or
higher, because the compass of human voices from the lowest men to the high-
est women or boys is much greater than two octaves. By the later fifteenth cen-

tury the range employed had expanded significantly; when we listen to Josquin
we begin to feel that the composition has colonized much more of the avail-
able space. With Tallis, and other members of the English School, we marvel
at the sheer expanse of the musical territory.

Of course, when Josquin and his generation were colonizing the whole mu-
sical range their patrons were colonizing the world,^ but before that "Latin
Christendom " colonized a lot of territory in Europe. Why did the medieval mu-
sicians stay within a narrow spectrum?
Because the medieval music which survives to us suggests a dependence
upon the human voice as the means of colonizing territory for the ear to rule.
Theorists of the Middle Ages state that the absolute limit of one human voice
is two octaves or less. Medieval singers were clearly not required to exercise in

anything like the gymnasium of the modern singing lesson. The theorists fre-
quently say that musical instruments can ascend considerably further upwards
(I cannot think of one who says they may descend considerably further down-
wards), but this clearly did not affect their conception of the voice as the cen-
tral resource of artistic music.

Which has of course been reflected in your work on secular song perfor-
mance. To consider another element of this, though, what do we know about
the specifics of how they listened to music?
The evidence is even sparser than the evidence about performing practice,
but it would be worth putting together. I think we need not only a discipline

of performing practice, but also a discipline of listening practice. It would be

good to know more, for example, about the patterns of attention that a per-
former could expect. Did people gather together specially to hear music, or was

2. Edward Lowinsky, over fifty years ago, found connections between the Renaissance ex-
pansion of musical space and the expansions of the Renaissance concept of physical space pro-
duced by Copernicus and Columbus; see his "The Concept of Physical and Musical Space in
the Renaissance," reprinted in his Music in the Culture of the Renaissance (University of
Chicago Press, 1986), pp. 6-18. In one of a number of parallels, Lowinsky points out, for ex-
ample, that the medieval scheme of the solar system not only was geocentric, but also placed
the planets much closer to the earth than in the Copernican scheme, because empty space was
considered useless, and nature did not include anything useless.
74 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

music essentially inserted into what was another kind of occasion? And how
consistently did people pay attention? There is more evidence about that than
one would expect. As early as the thirteenth century there is a reference, from
troubadour sources, to listeners who have started to chatter by the time a singer
reaches the third word of his song.

What would be the listening practice for those complex three-texted motets?
The one person who says anything about that is Johannes de Grocheio, and
he is quite remarkable for c. 1300 in that he specifies something of the audi-
ence for each kind of practice. For the motet, which in his day is on the way
to transmuting itself to a precious ars nova creation, his comment is that it

should be performed before the literati, which I think means the clergy, or a
little more than that —the more conscientious clergy, those who take seriously
their duty to know Latin and to study books of theological and pastoral im-
portance. That suggests that the listening practice for something like a motet
of c. 1300 could well be the following: the bishop in his palace —every bishop
has one, although the word "palace" may be rather grand for what it actually
is —has dinners to which he invites clergy from his diocese, possibly another
bishop, and quite possibly some leading guildsmen or the mayor. One can imag-
ine that an occasion like that might have a motet. The context for so much
music-making in the Middle Ages is festive in the strict sense of the word: a
feast, actual dining. Meal times are probably the only times when people in a
medieval household converged, and therefore there was a tendency for music
to gravitate towards them.

As for the listening practice of sacred music, you once described chant as
being not so much among monks.^
a performance as an incantation or prayer
At what point do sacred musicians become performers, as of course they were
by the time of, say, Monteverdi?
By the fourteenth century, there are clear references, most of them dis-

paraging, to singers in England who sing elaborate polyphony so that they will
be congratulated afterwards by laypersons. One thing involved here is the nat-
ural vanity of people who have good voices, which I imagine is simply irre-

pressible and always has been. But there is another factor. We see it in the

twelfth century, with the Parisian organum of Leonin and Perotin. One of the
striking things to emerge from Craig Wright's Music and Ceremony at Notre
Dame'* is that much of theorganum of Notre Dame is processional: it was per-
formed with the full theatricality of a procession. Now, if lay men or women
were impressed, the more wealthy amongst them would endow masses or give

3. In "Musicus and Cantor," in Companion to Medieval and Renaissance Music, ed. Tess

Knighton and David Fallows (London: Orion, and New York: Schirmer, 1992), pp. 74-78.
4. Cambridge University Press, 1989.
 THE COLONIZING EAR 75

money to raise the rank of the hturgical feast so the organum could be sung.
Most of the polyphonic singing was done by deputies, some of whom were free-
lance performers, the so-called "masters of organum."^ One could say to them:
"I am the bishop of Amiens, and we do not have anything like this at the mo-

ment. So I will buy your services; will give you a benefice of so many florins
I

per year if you will come and sing it in my church." This was a living for some
—
young singers even if probably not a very good living, and one they rejected
once they found some better job in the Church. So here there is a link between
organum and money. Sometimes when organum is performed very vigorously,
as it was by David Munrow, it seems to me that can almost hear turned into
I

sound one of the great themes of twelfth- and thirteenth-century history: am-
bition.

This suggests an approach to understanding their experience: how music re-

lated to other social developments. We discussed the colonizing spirit, but

you've also written about connections between music and mathematics and sci-

ence —for example, how the introduction of Arabic numerals led to notational
advances that made ars nova possible.^
It's clear that the idea of measurement does intrigue medieval musicians, and
of course there are two things you can measure —how far apart notes stand

(their intervals) and how long they last. The idea that the materials of music
might be objectively and scrupulously calibrated, with regard to both the height
and the duration of the notes, bit deeply into Western musical consciousness in
the thirteenth century, and retained its intensity for a very long time indeed.
One obstacle to our understanding their experience, though —and this ap-

plies, I suppose, to any period — is that you have to compare what contempo-
raries say about their own art with what they appear to do within their art.

The example I use in Discarding Images is that if we were to form our assess-
ment of Geoffrey Chaucer's poetry on the basis of what fifteenth-century read-
ers praise in it, we would end up with a very different Chaucer from the one

—
we have now and probably a poet that Chaucer himself would find surpris-
ing. Fifteenth-century readers praise him for all the virtues that they have terms

for— his "high sentence" and "moral learning" and so on. They don't praise

—
him for the things we praise for irony, for humor, for realism, and (to come
more up to date) for a sympathetic and complex attitude to women. But when
fifteenth-century poets try to imitate Chaucer, often to complete the incomplete
Canterbury Tales, they show by their pastiches that they understand things
which they did not have terms to explain. To attach great importance today to
what medieval theorists say about measurement is to lean heavily upon what

5. Page, The Owl and the Nightingale (London: Dent, 1989, and Berkeley: University of
California Press, 1990), chap. 6.
6. See Page's Discarding Images (Oxford University Press, 1993), chap. 4.
76 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

medieval people were able to say about what they found in music. I suspect
that what they actually found was broader than what they could name. It is

surely the same with us.

Your mention of ambition and of marketing oneself at Notre Dame brings

Hollywood deal-making to mind, and you say now that appreciation of

Chaucer by the generation that followed him was much more like ours than
their critical writing suggests. This brings up a key issue in this whole exercise

I've tried to involve you in: how much can we understand and relate to them —
are we just too different, as some today believe?

If there were no such thing as "transhistorical humanness," then trying to

write a history of fifteenth-century music by composers like Obrecht would be
like trying to write a history of the parrot community in Australia on the basis

of interviews with living parrots. We can't understand what the parrots are say-

ing, and we don't know how they structure their experience or what memory
they have of the past. Obrecht does not sound to me like a parrot.

Regarding the question of transhistorical humanness and performance, a

particularly hot debate has arisen over your suggestion that "sustained expo-
sure to the sound of medieval music [today] contributes to a vital sense of pro-
portion (in the colloquial sense of those words), not only in the analysis of spe-
cific musical details but also in conceiving what the music may have meant"
(Discarding Images, p. xxiv). That is, you seem to say that our playing the
music now can give us insight into what it meant to them then. What would
you say to critics of this view?
Anybody who denies that listening to the music of the past is a vital part

of studying the music of the past will never write a book or article that I want
to read. Why, though, has this issue become so controversial? It may be partly
because we live in an academic milieu where anyone who follows a critical
practice of any kind is required to clarify the "theory" of what they do —which
usually entails the presentation of an argument that can be recognized as
methodologically sophisticated and self-aware by those who, wittingly or un-
wittingly, espouse some form of post-structuralism, deconstruction, discourse
theory or whatever it may be. Many influential literary critics argue that in

speaking of texts —
some would say, of anything it is not legitimate to take
or, —
a particular standpoint and try to discredit other standpoints. It is not legiti-
mate to insist upon a particular value judgment; when we admire texts we are
really only reacting to the fact that they seem flattering to us, that they tell us

what we already know. If you are to substitute "listener" or "performer" for

"reader" or "critic" in statements of that kind, you will end up with something
like the position which some modern scholars have taken on the question of
historical performance.

It would be one thing to state that certain performances now are actually
 THE COLONIZING EAR 77

recovering something that was done then; I've never pubhshed such a view. But
it's another thing to say that some performances are more historical than oth-
ers. If they're all equal in their historicity, then the historicism disappears. It's

only when you can have some kind of a scale of judgment that you can invoke
a notion of historical performance at all.

That "scale of judgment " is exactly what critical theories would try to dis-

miss.
My conception of what we do in Gothic Voices implicitly rejects some major
developments in critical theory during the last thirty years. Up to a point, per-

haps, this is true of any committed early-music ensemble, and is not just a mat-
ter of the inevitable slippage between theory and practice. From the vantage
point of the later 1990s, it is obvious that thinking of the kind found in vari-

ous kinds of post-structuralism was bound to corrode the notion of "authen-

ticity," so often invoked in the 1970s as the pursuit of "what the composer in-
tended." During the last three decades, various forms of deconstruction have
attacked the belief that a coherent, individual subject —an author—endows an
artistic creation with meaning. Further, many critical theorists have tended to
silence discussion of ethical issues. One may now read sophisticated presenta-
tions of the view that the study of ethical issues is the study of what an indi-

vidual society finds acceptable or expedient, or that it is the study of a certain

way in which language operates. A short step takes us from here to the asser-
tion that there is no such thing as "human nature," and that we have no firm
ground upon which to stand in judging the promptings of conscience. To bring
the matter down to the level of historical performance, it means that, as you
said, there would be no way of talking about continuities of musical taste or
understanding from the Middle Ages to the present.
I am far from clear where I stand on some of these issues at this point. But
I do wish to state boldly here that I believe the modern performance of me-
dieval music to be an ethical concern, and that any "historical performance the-
ory" which tends to silence ethical issues can have no interest for me. By in-

voking "an ethical concern" in this context, I mean that my answer to the
question of why we should perform medieval music is a part of my answer to
the question of how we should live.
This may seem too portentous an edifice to build on waste ground that is
strewn with old reviews, performer profiles in newspapers, CD booklets, record
company publicity, and all the other ephemera which sometimes cause profes-
sional scholars such alarm. I hope it is not. An alert and compassionate ap-
proach to the arts of other civilizations, including the arts of one's own civi-
lization in the past, is a great human good. The aim of people in many areas
of cultural life, as we approach the year 2000, is to inspire what might be called

"compunction" in this regard: a sudden turning of the mind's attention to the

existence of something that has beauty where one might not have expected to
78 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

find it, or that offers a challenge where one might not have expected the re-
sources for mounting a challenge to be present. I hope that recordings and con-
certs by medieval ensembles, Gothic Voices included, may do something to pro-
mote such awareness. That is an enterprise which is independent of scholarly
progress or controversy, because it does not depend upon a belief that one's
own artistic solutions have scholarly support.

True, but when you talk about a compassionate approach to the arts of
other civilizations and the value of encountering them, it returns us to the issue

of what these performances can tell us about the experience of people in these
civilizations —
presumably, that's part of their value to us. Can our perfor-
mances tell us about such things, specifically about our own past eras?
Yes, why not.'' Music is part of our evidence about the past. It is certainly
possible for performers to influence in subtle and powerful ways what the
public (and that includes scholars) hear in the music, and to define what
scholars think could have happened in the past. If modern performers are un-
able to produce a certain kind of result, it will often produce a consensus
among scholars that the result could not have been achieved in the past. A
specific instance of this is the performance of French secular music with voices
only. One of the reasons why the idea was rejected in the late 70s, when it

was first proposed, was that at the time no performer had done it success-
fully. When performers learned how to do it, the idea became acceptable. So
in that sense, if there is a change in performing style, it can affect the way
we hear the music.
However, that's not necessarily the same as saying that one is discovering
something about the music. One may be discovering something for oneself, a
possibility one didn't think was there, but the question is whether performance
can imply something about the musical past. would agree that it can't verify I

anything. I would still suggest, however, that some performances are more his-
torical than others, and that those that are more historical should be able to

imply something that the less historical performances don't.

That seems reasonable, but I suppose a crucial question then is how we de-
cidewhich performances are the more historical ones —who decides, and on
what basis?
Let me give an example to illustrate how one might approach it. Accuracy
of tuning is something that performers of the 1940s, as one can tell from their

recordings, simply couldn't accomplish in their performance of medieval music.

It's just a plain fact; they couldn't do it, partly because they were using vari-
ous vocal techniques that prevent the listener's ear from seizing and savoring
true, clear pitch. Nor could they sing some of these very syncopated rhythms
of Machaut, for example, in correct time. Now it seems to me that everything
one can deduce about medieval performance style from what theorists say and
 THE COLONIZING EAR 79

from what composers write suggests that performers who can do those things
are being more historical in their approach than performers who can't.

I suppose the safest position to take is that performance can be educative.

I think we're all agreed on that. Like most performers, I have certain intuitions
that some things are correct and others are not. Now, that's a very shaky basis
on which to go into print, and it may be that as a scholar one should keep out
all such intuitions and deal only with what can be deduced from the evidence
in a positivistic way, but I still do believe that progress has been made in the

performance of medieval music in the last thirty years.

One thing that dramatizes the uncertainties, and that ismore likely to arise
in medieval than in, say, nineteenth-century repertoire, is when two performers
come to opposite conclusions. An example has to do with meaning in medieval
music: whether texts should be sung expressively. Here your views as you've
written about them seem to conflict with those of some others, including some
interviewed in this book.^ Could you address this?
To take an extreme view for the moment, one could say that medieval mu-
sicians set a text as neutral syllabic material, while Renaissance musicians, im-
itating what they imagined to be classical example, discovered the power of
music to express the meaning of text, not only by adopting more "rational"
methods of word-setting but also by attending to particular moments of verbal
meaning. Medieval writers on chant — with whom any consideration of this

matter begins —sometimes say that composers of new chants should ensure
"that the music seems to express what the words say."^ Notice that this is a

reference to how a composer should compose, not to how a performer should

perform; expressiveness in that regard is something which medieval theorists of
music do not so readily discuss. As I read the several hundred words which our
twelfth-century author has devoted to the composition of new chants, I cannot
find anything that is easily recognizable as a summons to what a modern singer
may intuitively regard as "expressive" means. Indeed, this author thinks that
the best way to make music project the sense of the words is to lead every mu-

7. Page, Discarding Images, pp. 85-86. See the interviews with Barbara Thornton and
Susan Hellauer.
8. Page informs me that he was referring to the treatise of Johannes (c. 1100); see J. Smits
van Waesberghe, ed., Johannis Affligemensis De Musica cum Tonario, Corpus Scriptorum de
Musica 1 (Rome: American Musicology, 1950), p. 117. As early as the ninth cen-
Institute of
Musica Enchtriadis, we read, "In peaceful subjects let the
tury, in the Uturgical-music treatise
notes be peaceful, happy in joyous matters, grieving in sad ones; let cruel words or deeds be
—
expressed with harsh sounds sudden, loud, and swift shaped according to the nature of —
events and the emotions." In line with Page's view that this has to do with composition only,
this passage has been called "rather irrelevant to actual singing technique" (David Hiley, in
Performance Practice: Music before 1600, p. 44) and "an elegant formulation of the medieval
aesthetic of the affections" (J. Dyer, "Singing with Proper Refinement," Early Music 6 [April
1978], p. 211).
80 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

sical phrase towards the final of the mode at a point where the text completes
a sentence or other major unit of sense. This is an analytical —even a gram-
matical —approach to the idea of projecting meaning, not a synthetic or dra-
matic one. It is also striking, I think, that one of the most influential of all

plainsong theorists, Guido of Arezzo Iwhose treatise Micrologus was finished

c. 1030], describes a way of composing new chants in which the composer is

to write the five vowels of the Latin alphabet in one, two, or three sequences
beneath the gamut, and to produce a new composition by matching the vow-
els of the syllables to their corresponding notes. This is a system which sug-
gests a remarkably undramatic conception of how words and music might be
married.
The plainsong modes bring us a little closer to the kind of expressiveness
that a modern singer is perhaps most likely to wish to discuss. Although there
can have been no standardized pitch in the Middle Ages, the system of eight
plainsong modes was not regarded as a way of defining eight different se-
quences of tone and semitone steps within the same notational octave but was
conceived as a movement in which those sequences were assigned to various
areas in the voice from the lowest usable region to the highest. To say, for ex-
ample, that a chant is "Authentic Phrygian" (and the plainsong theorists are
overwhelmingly concerned with such classifications) is to say roughly where it

should lie in the voice. One sang the Plagal Dorian (mode 1) lowest of all, for
example, and the Authentic Mixolydian (mode 2) the highest of all. Let me re-

peat, this has nothing to do with actual sounding pitch: men and women could
establish this system of differentiation within their voices at quite different
pitches. The eight modes were therefore associated with a move upwards from
a thicker, darker region of the voice to a thinner, brighter tone, and this is the
language which medieval writers repeatedly use to describe the sonorities of the
human voice as pitch rises. This movement from darkness to brightness was
naturally associated with the most fundamental association that the human
mind makes between emotion and musical sound, namely that a high, bright
sonority befits a mood of exaltation. (The words "gravity" and "exaltation"
both have etymological roots which display the potency of the simple concepts
I am describing.) Medieval writers on plainsong say exactly this; and while it

is clear that there are lamenting texts set to higher modes and joyous texts set
to lower ones, the association between vocal pitch and meaning that I have just
described exists in their minds and is described in several treatises. Once again,

however, this is an expressiveness which is created in a general fashion by a

major resource of the entire piece — its place in the voice —and not by a re-

sponse to certain individual moments in the text.

The association of modes with certain states of feeling an association that —

a number of plainsong theorists make with exuberant metaphor is more than —
a matter of vocal color, however. The modes were not just scalar patterns of
an octave associated with certain regions of the voice; a mode was a complete
 THE COLONIZING EAR 81

Stylistic grammar for all the melodies cast in it, involving certain characteristic
melodic movements, especially at the beginnings and ends of phrases. These as-

pects of mode also seem to have been associated, at least in the minds of the-

orists, with certain emotional states. "^

I have tried to illustrate in my edition of

the thirteenth-century Summa Musice how references to "the haughty prancing

of the third mode" and so on do make sense in terms of brief, recurrent melodic
configurations characteristic of the modes concerned."' I emphasize once again,
however, that this is a system of expressiveness which relies upon the singer's
(and the hstener's) internalization of the whole modal system of plainsong; it

is not something that can be approximated by the modern singer who reads the
text of a piece, forms an idea of its "meaning," and then proceeds to project
it rhetorically moment by moment. There is certainly evidence in medieval

monophonic settings that composers might respond to verbal moments in a

text, but the crucial difference between their practice in this regard and what
a modern listener might intuitively expect is that the medieval composer does
not actually "mime" the affective meaning of the word with some musical fig-

ure which he considers a correlate for it; when he draws attention to a word
or words, it is by exploiting something —a melodic formula, a particular inter-

vallic step —that is part of the general resources of the piece.

How does this relate to expressiveness in later medieval repertoire, espe-

cially polyphony?
The question of whether such "expressive" resources were ever transferred
to measured polyphony, whose melodic modal character (when it had any) was
clearly of a very different sort, is easy to ask but probably impossible to an-

swer. My own suspicion is that when the interests of composer and theorists

shifted to mensural music and to the endlessly intriguing (and exasperating)

problems of its notation, the older language for talking about expressiveness in
general terms of the plainsong modes was simply dropped from the agenda of
trained musicians, together with the issues broached in that language. One
achievement of the fifteenth century will be to reinvigorate that language and
to apply it to polyphony. (In this regard, as in some others, the Renaissance in

music is not so much a rediscovery of Antiquity as a rediscovery of twelfth-

century teaching about plainsong, now applied to polyphony.") I have never

read a medieval treatise that cites a set of polyphonic pieces for their contrast-
ing "affects" in the way plainchant theorists cite items of plainsong from at
least c. 1100 on.

9. Harold Powers, "Mode," III, 1, in The New Grove Dictionary of Music, vol. 12 (Lon-

don: Macmillan, 1980), pp. 397-401.

10. The Summa Musice: A Thirteenth-Century Manual for Singers (Cambridge University
Press, 1991). The passage Page refers to here is from chap. 22, on p. 118; his discussion and
illustrations are on pp. 18ff.

11. Powers, "Mode," III, 1, b.

82 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

Of course, you've already discussed the discrepancy between what theorists

talked about in Chaucer and their pastiches of Chaucer.
Absolutely, and there's another factor to consider as well. All of what I've

said about chant and polyphony is true, I hope, but it is also a delaying tactic;
for I am at last compelled to admit that I often find performances of medieval
monophony or polyphony unsatisfactory if singers are zealous "to make some-
thing of the words." I cannot endure performances of any medieval music in
which singers act, and singers who do perform in that way are, in my view,
making a grievous mistake. "Expressiveness," as most modern singers are in-
clined to interpret it, often includes a pervasive and subtle (or not so subtle)
rubato which is in my view fundamentally opposed to the ethos of musica men-
surabilis, an art which thrived upon the scrupulous calibration of durations and
intervallic steps, and which is constantly praised by its devotees for its innate
resistance to waywardness and caprice.

This brings up another area of divergent opinion: accuracy of duration.

Marcel Peres has a different attitude: his view is that because they weren't used
to the clock in the twelfth century, they may have notated duration precisely
but still didn't think of time arithmetically — instead, they thought of it quali-
tatively.

It's definitely a different attitude. Certainly the rise of measured notation in

the late twelfth century and its increasing elaboration in the thirteenth and
fourteenth centuries has been linked to the rise of mechanical clocks, and in-
deed to the establishment of civic, mercantile time as opposed to contemplative
monastic time.'^ But I don't think that this has anything to do with issues of
performance or notation at any time. I can quite well believe, however, that
performances of some kinds of polyphonic conducti in the twelfth century had
a wondrous flexibility, and that by the fourteenth this had all changed. When
one is investigating the music of, say, Machaut, one surely discovers that the
sonorous events are calibrated against a pulse with great scrupulosity, and that
the placement in time of each note, not before and not after, so that the dis-
sonance is precisely disciplined, is an essential part of the medium.
Here I must declare my affiliation. My genealogy as an interpreter of me-
dieval music goes back through singers such as Margaret Philpot to Michael
Morrow and John Beckett of Musica Reservata, the ensemble whose record-
ings I first heard as a schoolboy. How does this affect me? For one example
among many, I take it to be absolutely fundamental that unless a crescendo is

planned, a note should begin with as much commitment and tone as it is ever
going to possess; but that's something that Michael Morrow taught John Beck-
ett who taught it to Margaret Philpot who taught it to me. So perhaps it be-

12. Jacques Le Goff, Time, Work, and Culture in the Middle Ages (University of Chicago
Press, 1980).
 THE COLONIZING EAR 83

hooves me to stand back a moment before I reach for medieval theorists whom
I might cite to support my views in this regard — theorists who insist that tenor
parts, for example, be sung "in a firm and rigorous fashion." Influences reach-

ing back long before I had heard of the medieval theorists must have influenced
the uses I find for their statements.

After twelve years, the ensemble Gothic Voices has evolved a style. Our styl-

istic ideal —one which it is quite beyond us to achieve with any consistency
is this:

All the polyphonic music of the Middle Ages must be sung with scrupu-
lously accurate tuning (would that it were always possible!). Standards of

ensemble must be of the highest order. In most cases polyphony should be

sung one voice to a part, by singers with a natural ability to phrase and an
ability to move very cleanly from one note to the next. Blend is vital. Close
attention to tactical tuning —the widening and narrowing of intervals for
artistic effect — is of great importance, and for most French, Italian, and
Franco-Flemish music before the mid fifteenth century tuning should nor-
mally follow Pythagorean principles. Next, the music must be sung with a
meticulous sense of pulse, with a light, transparent tone, and (for the most

part) with bright, clean vowel sounds produced quite far forward in the

mouth.

Here are some of the difficulties I believe I encounter in performing me-

dieval music. Some of what I said to describe the Gothic Voices' stylistic ideal

can be classified at once as personal artistic preference (the bright vowels, the
clean movement from note to note), vital to any sense of style in the perfor-

mance of music but not demonstrable by reference to the surviving evidence.

However, it is only by the most laborious self-scrutiny that I can begin to
glimpse how many of the remaining matters are what I know and how many
are what I believe. Up to a point, questions of scoring (one voice to a part) rest
upon literary and documentary evidence, although the testimony is far from

unanimous. The claim about scrupulosity of tuning can also be urged, if not
actually proved, on the basis of the obsessive interest which the musical theo-
rists of the Middle Ages show in the calibration of musical intervals, not to
mention the strictures of someone like Jacques of Liege [c. 1325), who cannot
abide singers who, when they are required to sing an octave, "ascend more, or
ascend less, than the full five tones with two minor semitones." As for blend,

we have Jacques again, who says that in polyphony the various parts should
resolve as if into one voice (this was an ideal carried over from plainchant).
But I can't claim that any of this is conclusive.

Your distinction between house style, knowledge, and belief indicates cau-
tion about claims that your approach is more historical than others', even if
84 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

you posit that such a thing is possible in principle. However, you've been crit-

icized for (your critics say) suggesting not only that English musicians are dis-
tinctive in some respects, but that their priorities are like those of medieval and
Renaissance singers, which, you said, might "complicate (but not confute) " the
argument that some ideals of historical performance are "a projection of cer-

tain 20th-century musical ideas and tastes.'" ^^ Some non-British musicians take
this to imply a Kiplingesque assertion of British privilege and, also, of histori-

domain of music. How would you answer such charges?

cal accuracy in this
When the article to which you refer was still in typescript, it was read by
more than a dozen English musicians, scholars, and critics. As I remember, none
of them quite foresaw the reaction of the "non-British musicians" whom you
mention. Indeed, the general view of the English readers whom I consulted was
that the article offered a timely deconstruction of the English a cappella scene
and one liable to offend many working within it, not a complacent or some-
how what is being done there. Perhaps the non-British
political vindication of

readers whom you mention do not realize what is implied by saying (as I do
in the article) that the English a cappella scene might be described as "the mu-

sical equivalent of the Whitehall civil service." That is not a compliment.

I think I now see it this way. The medieval writers who discuss measured
music sometimes say that expertise in plainsong is the basis of an expertise in
polyphony. At least two things are involved here. A training in chant was usu-
ally the way in which a singer learned to read from staff notation and came to
recognize note shapes. On the other hand, however, those writers also mean
that no singer can be a skilled performer of measured music until he has ac-

quired long experience of the choral discipline of performing chant. We possess

many attempts by medieval writers to codify good practice in the performance
of plainsong. They emphasize that each singer should listen carefully; they re-
quire that all pauses should be executed in a neat and unanimous way; some

13. Page's "The English a cappella Renaissance," Early Music 21(August 1993), pp. 453-71
(quoting p. 470). This article should not be missed by anyone interested in this issue. It argues
that the English preference for a cappella singing of medieval and Renaissance music and the —
proliferation of groups such as his, the Tallis Scholars, the HiUiard Ensemble, and many more
reflect partly a scholarship that has mostly originated in England and, more than that, reflect
an English tradition of a cappella choral singing cultivated network with hubs at Ox-
in a tight
ford and Cambridge. He summarizes from the premiss that Eng-
his theory, early on: "It begins
lish singers performing a cappella are currently able to give exceptional performances of me-

dieval and Renaissance polyphony from England and the Franco-Flemish area because the
ability of the best English singers to achieve a purity and precision instilled by the discipline of
repeated a cappella singing in the choral institutions is singularly appropriate to the trans-
parency and intricate counterpoint of the music. From that premiss we proceed to the theory
that, in certain respects, and especially in matters relating to accuracy of tuning and ensemble,
these performances represent a particularly convincing postulate about the performing priori-
ties of the original singers" (p. 454). It was the last proposition especially that touched off con-

troversy; for one argument, see Donald Greig's "Sight- Readings: Notes on a cappella Perfor-
mance Practice," Early Music 23 (February 1995), pp. 125-48.
 THE COLONIZING EAR 85

authors, such as Jerome of Moravia, insist upon a blended sound, and sets of

regulations for the conduct of religious life sometimes emphasize the essentially

self-effacing nature of the ideal chant performance: "Let one voice be scarcely
distinguishable from another." All of these stipulations must be read with care,
needless to say, for their ultimate goals are spiritual and not aesthetic; more-
over, many of them are specifically concerned with psalmody. None the less,

their regulations sound like a call for something which can be described, I be-

lieve without grievous anachronism, as "good consort singing." By this I mean

an essentially non-soloistic manner of singing that values blend and good en-
semble, honed during the daily discipline of preparing music for the day's ser-
vices. England is a country where singers are still trained in a discipline of this
kind — so is America, if you relax the point about "preparation for the day's

services" and then count some of the vocal ensembles in the soul and close-har-
mony The performers of Take Six are by far the best consort singers
traditions.

I have ever heard, and it was Rogers Covey-Crump, a doyen among English
early-music singers, who first introduced me to them.

As I've said, for me, working with an a cappella group in England, there is

already a danger that hidden preferences are making themselves felt even as I

speak about the meaning of medieval evidence. For various reasons, the tradi-
tion of consort singing in England has produced (by general agreement) stan-
dards of excellence in intonation, blend, and ensemble, and this provides a very
suggestive —perhaps a very seductive —context for understanding what Jacques,
Jerome, and others are referring to. I attempted to scrutinize that context in the

article you mentioned, even going so far as to say that, because most of the
English ensembles have a very strong (indeed overwhelming) Oxbridge contin-
gent, their singing "turns the memories and dreams of a social class into

sound."'^
And here is the heart of the problem. All serious musicians, I take it, have
something we may call their artistic integrity; it is a specialized working of con-
science that they bring to what they do. When musicians are involved in his-

torical performance, however, they must scrutinize the relationship between

their artistic integrity and their intellectual integrity —the relationship between
what they believe about the music, and can satisfactorily express in no other
way than by performance, and what they know about the music. The problem
for many performers is that it may be difficult for them to distinguish what

they believe about the music from what they know about it. I do not believe

that the modern performer has privileged access to any special way of know-
ing; but, as I said, I believe that our understanding of the technical resources
required to make early music musical can improve with time, and has done so
in medieval music. As a scholar, my greatest difficulty has been to find a
rhetoric with which to express that idea while acknowledging the need to scru-

14. "The English a cappella Renaissance," p. 458.

86 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

tinize themeans that have enabled me to conceive it, namely the English choral
tradition.As a performer, my greatest difficulty has been, in a Yeatsian phrase,
to acknowledge that I "lack all conviction," for most of what I do to make the
music musical rests upon nothing that can be called evidence, and yet I am "full
of passionate intensity" about what we do.
You mentioned Kipling. Well, the a cappella renaissance in England will not
last forever. During the next twenty years the universities of Oxford and Cam-
bridge, with their choral foundations and chapels, are set to change a great
deal. Nobody knows what effect that will have on the choral scene. Come back
in 2015, and you might well find me quoting Kipling to myself:

"Over then, come over, for the bee has quit the clover.

And your English summer's done."

SELECTED DISCOGRAPHY
Page and Gothic Voices have made over a dozen recordings for Hyperion. In a
review that includes an index of almost all of them through 1993, the musi-

cologist Lawrence Earp says that "no recordings are more eagerly awaited by
enthusiasts of medieval music. "'^ Page feels that the group hit full stride in
1991 with their ninth CD, The Medieval Romantics (CDA66463); it was the
first in which, according to Page, the group had solved the problems of how
to vocalize non-texted parts and had found two bass-baritone singers appro-
priate to the music. One of its songs, Solage's ]oiex de cuer, inspired Pierre
Boulez to approach Page after a concert and say, "This Solage — it is extraor-
dinary!" The and the two that followed, Lancaster and Valois (66588) and
disc
The Study of Love (66619), explored the French repertoire of the fourteenth
and early fifteenth centuries; any one of them would be a good place to start
exploring Page's work. Even better might be his most recent (as I write) series,
The Spirits of England and France, since its Volume One (66739) is an an-
thology that surveys several areas of repertory. Volume Two (66773) is one of
the finest extant recordings of trouvere songs; Volume Three (66783) is an ac-
claimed sampling of the fifteenth-century composer Binchois and his contem-
poraries.
Howard Mayer Brown criticizes the performances of chansons on the CD
Castle of Fair Welcome (66194) as "a bit too perfect, slightly without individ-
ual personality or nuance. One cannot tell, for example, just from listening
whether the performers are singing a happy or a sad song." Brown notes, how-
ever, that this is "clearly intentional."'* Page's view is that the music itself
would tell one nothing about whether a song is happy or sad (only the lyrics
would) and that the performance needn't either; what the original performances

15. Earp, Early Music 21 (May 1993), pp. 289-95; quote, p. 289.
16. Brown, Early Music 15 (May 1987), p. 277-79; quote, p. 277.
 THE COLONIZING EAR 87

expressed was the "social register" of a piece —whether it is in high style or a

lower one. Richard Taruskin, heartily praising the same CD, paraphrases Page's
views on expressiveness in late-medieval songs, saying that "what is 'expressed'
in these chansons, in short, is the quality of hauteur, that is 'elevation' ... in

tone, in diction, in delivery, all reflecting the elevated social setting in which
the performance took place." He says that Gothic Voices' application of this
approach, in one piece, gave him "goose bumps," adding that such "historical
gooseflesh is 'authenticity' at its best and lin existential termsl its most au-
thentic."'' That two prominent musicologists could differ on this issue suggests

how challenging the historical evidence can be.

I am especially fond of The Voice in the Garden (66653), which takes a very

different approach to the Spanish cancioneros repertoire from the more popu-
lar one of Jordi Savall. Savall's colorfully orchestrated versions are infectious,

but I was surprised to find that the same is true of Page's mostly vocal ap-
proach. Page's approach appears to be more musicologically sound."*

FOR FURTHER READING

Page's own writing not only is among the most important in this field but is

some of the most readable. His own favorite among his books is The Owl and
the Nightingale (London: Dent, 1989, and Berkeley: University of California
Press, 1990). The book's importance goes beyond details of music: it is really

a social history that considers such issues as the role of music in the life of a
society, and how that role changed during a time and place crucial in Western
history. Page also discusses the period in an excellent brief essay, "Court and
City in France, 1100-1300," in Antiquity and the Middle Ages, ed. James
McKinnon (London: Macmillan, 1990, and Englewood Cliffs, N.J.: Prentice
Hall, 1991), pp. 197-217.
Page's Voices and Instruments of the Middle Ages (Berkeley: University of
California Press, 1987) includes a fascinating discussion of the troubadour
repertoire and its use of instruments. Some of his other writings on the use of
instruments and voices are listed in note 1, above.
Page's most controversial book. Discarding Images (Oxford University
Press, 1993), is about musicology's construing of the Middle Ages. One im-
portant record of the controversies it evoked is an exchange between Page and
two musicologists, Margaret Bent and Reinhard Strohm, which appeared in
Early Music 21 (November 1993), pp. 625-33 (Bent); 22 (February 1994), pp.

17. Taruskin, "High, Sweet, and Loud," originally published in Opus (June 1987), pp.
36-39, reprinted in Text and Act (New York: Oxford University Press, 1995), pp. 347-52;
quotes, pp. 350 and 352.
18. See Tess Knighton, "The a cappella Heresy in Spain: An Inquisition into the Perfor-
mance of the Cancionero Repertory," Early Music 20 (November 1992), pp. 562-81), and
Kenneth Kreitner's "Minstrels in Spanish Churches, 1400-1600," in the same issue, pp.
533^4.
88 MEDIEVAL MUSIC, PLAINCHANT, AND OTHERNESS

127-32 (Page's reply); 22 (November 1994), pp. 715-19 (Strohm's comments).

Incidentally, Strohm's The Rise of European Music: 1380-1500 (Cambridge
University Press, 1993) may be challenging (Alejandro Enrique Planchart's re-

view^ began with the memorable line, "This book is not for wimps"), but it is

fascinating. Far more accessible is Strohm's article "The Close of the Middle
Ages" in Antiquity and the Middle Ages, ed. McKinnon, pp. 269-312; Page
discusses earlier eras in light of (and in contrast to) Strohm's book in his arti-

cle "Towards: Music in the Rise of Europe," in Musical Times 136 (March
1995), pp. 127-34.
Page's essays in other books give accessible entries into his thought: see for
example "Instruments and Instrumental Music before 1300," chap. 10 in New
Oxford History of Music, II, rev. ed., ed. Richard Crocker and David Hiley
(Oxford University Press, 1990); and "Polyphony before 1400," in Perfor-
mance Practice: Music before 1600, ed. Howard Mayer Brown and Stanley
Sadie (New York: Norton, 1989), pp. 79-104.

Postscript: Medieval Music, Plainchant, and "Otherness"

In the preceding chapters, we've heard Marcel Peres speak of an era before
clocks and electric lights, Susan Hellauer discuss medieval associations between
plainchants and holy days, and Barbara Thornton emphasize the oral nature of
Hildegard's tradition.'*' Page, too, argues that the era's idea of musical expres-
siveness differed from ours.
Yet he also disputes those who proclaim the "otherness" of medieval peo-
ple. Unabashed humanist that he is. Page asserts that beneath the obvious
differences lies a "transhistorical humanness," which he has described else-

where as "an appreciable continuity of human thought and feeling from age
to age."2°
This assertion has caused a flap. In a critical essay, the musicologist Rob
Wegman dismisses Page's "transhistorical humanness" as an expression of the
"Enlightenment ideal of universalism." Like (among others) many postmod-
ernists, Wegman has no sympathy for that ideal. He mentions its "deeply prob-
lematic legacy"-^ and calls it "a typical product of a society that appropriates
the thoughts and artifacts of other societies while it tries to understand them.
Not surprisingly, it was closely allied to imperialism and nationalism in the

19th century. "^^

19. A longer version of this essay will appear in Early Music in February 1998.
20. Page, Discarding Images, p. 190.
21. Rob C. Wegman, "Reviewing Images," Music and Letters 76 (1995), pp. 265-73;
quote, p. 270.
22. Rob C. Wegman, "Sense and Sensibility in Late-Medieval Music," Early Music 23
(1995), p. 312.
 THE COLONIZING EAR 89

We could question the tone of moral condemnation; almost any idea, in-

cluding the ideologies Wegman seems to prefer, can be used for harmful pur-
poses. We could also question whether ideas of transhistorical humanness could
express only "Enlightenment" concepts. But I will ignore all that, and note just

that if we were to grant Wegman the point about hidden agendas it would set-

tle nothing. It would tell us about the biases that predispose people to see the
world a certain way, but it would not address the real question here: whether
Page's way of seeing things is, in the instance of transhistorical humanness,
more plausible than its rivals.
To help determine that, we must consider the relevant empirical evidence.
Theory-laden though it may be, it can at least disprove faulty ideas. Wegman,
too, is interested in empirical evidence. He says it "persistently denies" the
idea that medieval people "heard and felt just like us." The phrase "just like
us" makes this hard to challenge; it may also be unfair to Page, who speaks
only of "appreciable continuities." "Continuities" demand not exactness (hear-
ing just like us) but similarity. Does the empirical evidence really show that
we have no appreciable continuities of thought and feeling with people from
other times and places?
Wegman does not say which empirical evidence he means, but his references
suggest he's thinking of anthropology.^^ In mid-century, almost any generaliza-
tion about humanity could be countered by a report of a people so different
from us as to seem almost another species. There were Polynesians with no sex-
ual attachment or frustration, Chinese without romantic passion, Chambris
with reversed sex-temperaments, "gentle" Arapeshes without aggression, Hopis
with no sense of past, future, or temporal flow (they sounded a bit like

Alzheimer's patients), tribes with few color terms and, was assumed, just as
it

few colors experienced, and so on. A common explanation was that human be-
ings were, in Margaret Mead's words, almost unbelievably malleable, with the
molding process done entirely by culture and language.
But in recent decades every one of the above reports, and some of anthro-
pology's other exotica as well, have turned out to be myths.-"* A few anthro-
pologists have begun to complain about their field's tendency to overly exoticize

23. For example, in discussing transhistorical humanness in the Music and Letters review,
p. 270, Wegman quotes the anthropologist Clifford Geertz.
24. Donald Brown, Human Vniversals (New York: McGraw-Hill, 1991) chaps. 1 and 3;
Helen Harris, "Rethinking Heterosexual Love in Polynesia: A Case Study of Mangaia, Cook
Island," in Romantic Passion: A Universal Experience?, ed. William Jankowiak (New York:
Columbia University Press, 1995), pp. 95-127; Jankowiak, "Romantic Passion in the People's
Republic of China," in ibid.; Derek Freeman, Margaret Mead and Samoa: The Making and
Unmaking of an Anthropological Myth (Cambridge, Mass.: Harvard University Press, 1983);
Ekkehart Malotki, Hopi Time: A Linguistic Analysis of the Temporal Concepts in the Hopi
Language (Berlin: Mouton, 1983); Brent Berlin and Paul Kay, Basic Color Terms: Their Uni-
versality and Evolution {Berkeley: University of California Press, 1969). The "Alzheimer's"
comparison is from Jankowiak (personal communication, 1996).
90 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

"others" and to overstate cultural determinism;^^ and over a hundred behav-

ioral or psychological traits many once thought to result from "Western accul-
turation," have been shown to be universal.^'' For example, human beings in all

cultures have words for logical relations, including "not" and "same," and
"equivalent" and "opposite"; and all cultures make binary distinctions, includ-
ing "male and female," "black and white," "nature and culture," "self and oth-
ers," and "good and bad."^^ None of these cultures, obviously, subscribe to de-
constructionism, which teaches that "oppositional thinking" is a Western
hang-up.
It's unlikely that any of these human universals have changed since the Mid-
dle Ages. That alone gives us reason to take the idea of transhistorical human-
ness seriously.
Wegman's complaint may reflect an influential mode of thought, which holds
that there is "not such a thing as human nature" and that "socialization, and
thus historical circumstance, goes all the way down —that there is nothing 'be-
neath' socialization or prior to history" in any of us.^** We are entirely "socially
constructed." Some even imply that we are born blank slates. In view of these
ideas, transhistorical continuity has to seem an outmoded concept.
This viewpoint has little place for universals; and it is true that a given cross-
cultural universal doesn't necessarily imply an inborn detail of human nature;
but consider the other evidence this mode of thought must answer to in the mid-
1990s. Some examples are:

• Universals discovered by psychology. An extensive research program has

shown that certain basic emotions (happiness, sadness, anger, fear, disgust, con-

tempt, and surprise) are expressed by the same facial expressions in cultures as
disparate as Japan, the United States, Europe, South America, and preliterate
New Guinea. ^^ This basic repertoire of expressions is widely agreed to be an in-

born universal. Our culture may affect our sense of when it's appropriate to smile,
but it does not create our association of smiling with happiness. This is evidence
of our coming into the world with at least some unlearned content and detail

built into our psychologies. A more complex example of universals in psychol-

ogy involves certain patterns of mate preference found in every culture so far.^°

25. Brown, Human Universals, p. 155, and chap. 1. See also Freeman, Margaret Mead and
the Samoans, and Maurice Bloch, "The Past and the Present in the Present," Man 12 (1977),
esp. pp. 283-85.
26. Brown, Human Universals, chap. 6.
27. Brown, ibid., p. 134 and elsewhere.
28. Richard Rorty, Contingency, Irony, and Solidarity (Cambridge University Press, 1989),
p. xiii.

Human Face, ed. Paul Ekman (Cambridge University Press, 1982).

29. E.g., Emotions in the
30. David M. Buss, "Sex Differences in Human Mate Preferences: Evolutionary Hypothe-
ses Tested in 37 Cultures," Behavioral and Brain Sciences 12 (1989), pp. 1-14; reprinted in
Human Nature: A Critical Reader, ed. Laura Betzig (New York: Oxford University Press,
1997), pp. 175-90.
 THE COLONIZING EAR 91

• Some patterns of behavior are found not only in all humans but m many
other species —which raises the possibility that they are not entirely the result

of human acculturation." An example involves social status: both the body

language used to signal status and the biochemistry that accompanies changes
in status are found not only among humans but also among other pri-

mates.'^
• Some preferences may be present from birth, before one can be socialized^^
including, possibly, a preference for consonant over dissonant intervals.^"
• Instances exist of "prepared learning," where animals (including us) are in-
nately "primed" to learn certain things and not other things. An example is cer-

tain phobias: it has been shown that monkeys are pre-wired to develop fear of

snakes;" and human phobia patterns suggest that we, too, are primed to fear,

among other things, snakes, heights, and spiders — but not more significant mod-
ern hazards like, say, electrical outlets.'* These instances may provide evidence
for innate mental "modules" designed to process specific kinds of information

in specific kinds of ways.

• There is evidence of genetic bases for highly specific cognitive or mental prob-
lems, e.g., impairments in language and in visual-spatial cognition.'^ Again,
these may suggest inborn mental "modules."
• Evidence of brain-cell specialization. Although the development of the brain

31. Some examples are research on incest-avoidance mechanisms in humans and other
sexually reproducing species, discussed in chap. 5 of Brown's Human Universals; on family
dynamics in vertebrate species, in Stephen Emlen, "An Evolutionary Theory of the Family,"
Proceedings of the National Academy of Science, USA, 92 (August 1995), pp. 8092-99; on
sex differences in violence, e.g., Martin Daly and Margo
intra-sex homicide rates, discussed in
Wilson, Homicide (New York: Aldine de Gruyter, 1988), chaps. 6, 7, and 8; and on sex-
ual strategies, reviewed in Robert Wright's The Moral Animal (New York: Pantheon, 1994),
chap. 2.
32. This evidence is reviewed in Wright, The Moral Animal, pp. 236-62.
33. E.g., Judith H. Langlois, Lori A. Roggman, Rita J. Casey, Jean M. Ritter, Loretta
Rieser-Danner, and Vivian Jenkins, "Infant Preferences for Attractive Faces: Rudiments of a
Stereotype?" Developmental Psychology 23 (1987), pp. 363-69; Curtis A. Samuels, George
Butterworth, Tony Roberts, Lida Graupner, and Graham Hole, "Facial Aesthetics: Babies Pre-
fer Attractiveness to Symmetry," Perception 23 (1994), pp. 823-31.
34. Marcel Zentner and Jerome Kagan, "Perceptions of Music by Infants," Nature 383 (5

September 1996), p. 29.

35. Susan Mineka, "A Primate Model of Phobic Fears," in Theoretical Foundations of Be-
havior Therapy, ed. Hans J. Eysenck and Irene Martin (New York: Plenum, 1987), pp.
81-111.
36. Martin E. P. Seligman, "Phobias and Preparedness," Behavior Therapy 1 (1971), pp.
307-20; D. R. Kirkpatrick, "Age, Gender and Patterns of Common Intense Fears among
Adults," Behaviour Research &
Therapy 22 (1984), pp. 141-50.
37. Studies indicate genetic linkages for Specific Language Impairment; see, for example,
Dorothy Bishop, et al., "Genetic Basis of Specific Language Impairment: Evidence from a Twin
Study," Developmental Medicine and Child Neurology 37 (1995), pp. 41-55. Genetic link-
ages have been found for developmental dyslexia, and for visuospatial skill impairments in
Williams Syndrome. Regarding the latter, see "Gene Connected to Human Cognitive Trait,"
Science News (20 July 1996), p. 39.
92 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

allows for a great deal of plasticity, damaging a specific section of a mature

brain often results in the loss of a very specific function. This might be said to
provide evidence that the mind includes not just general mechanisms, but also
specialized ones. That some of these brain-cell localizations (as well as some el-

ements of neurochemistry) are found across species suggests that these may be
built-in, not the products of learning.

This is far from a complete list. But it's hard to see how the belief that
"there is nothing beneath socialization" could possibly account even for this

subset of the evidence. Rather, the full range of the evidence seems better ex-
plained by the viewpoint that we are born with at least some shared content
built into all of our brains. How much content is built in, what its exact na-
ture is, how it operates, and how it got there in the first place are matters
of debate and of active research programs.^** And whatever the answers are,
they clearly will not equate to "genetic —
determinism"^'' when genes influ-
ence behavior it's complex interactions with culture and en-
usually through
vironment. We are astonishingly malleable, and culture has an enormous in-
fluence on how we think and act. But enough evidence exists to cast grave
doubt on the idea that we have no inborn natures at all. Even the more mod-
erate idea that we have only a minimal human nature, consisting of a few
general mechanisms and basic drives — a view common among social scien-
tists —would find it hard to account for all of the above evidence (it's not ob-
vious that it could). At the very least, then, "transhistorical humanness" has
earned something more than a curt dismissal.
(By the way, the political and ethical implications of the findings I've listed
are by no means clear. Robert Wright says that to the degree that the emerg-
ing picture of innate human nature has "reasonably distinct political implica-
tions —and as a general rule it just doesn't —they are about as often to the left

as to the right. In some ways they are radically to the left."'"' None of what
I've covered implies Social Darwinism. Also feminists have tended to object to
arguments for innateness, but today many leading researchers in this field are

avowed feminists. And many thinkers in this field dismiss racial differences as

insignificant. I think we should try, therefore, to base our verdict not on poli-
tics but on the weight of the evidence.)
It's harder to say how all this applies to music. The musical universals we
know about (involving such things as phrase grouping) are so basic and gen-

38. See Human Nature: A Critical Reader, ed. Laura Betzig; also, Harmon R. Hoicomb
III, Sociobiology, Sex, and Science (Albany: State University of New York Press, 1993). See
alsoThe Adapted Mind: Evolutionary Psychology and the Generation of Culture, ed. Jerome
Barkow, Leda Cosmides, and John Tooby (New York: Oxford University Press, 1992).
39. For a sophisticated discussion see Hoicomb, Sociobiology, Sex, and Science, pp.
132-48.
40. Wright, The Moral Animal, p. 13.
 THE COLONIZING EAR 93

eral that they don't usually bear on issues in early music. An example: it seems
likely that the "natural intervals" — such as the octave, fourth, and fifth — have
affected the development of scales in many cultures;"*' and it has been suggested
that we tend to prefer these intervals because they play a role in how the brain
analyzes sound in natural environments."*- But I can't think of any early-music
performance issues that this could illuminate — especially since research suggests

that tuning and intonation, as opposed to scale structure, are determined by

culture, and that even scales can deviate from the natural intervals."" On the
other hand, it's possible that evidence for inborn elements of rhythm disproves
Peres's idea that strictly measured quantitative rhythm developed as a result of

our ancestors becoming used to the clock. Instead, such rhythm seems to re-

flect innate timing and motor mechanisms. '*'*

(This also may account for why
quantitative rhythms can be found in some cultures that have no clocks.) But

such reasoning doesn't take us very far.

The evidence I reviewed for a trans-cultural, transhistorical human nature

may have another kind of relevance to the quest to understand early music. In
recent years many scholars have emphasized how social elements bear on and
shape musical meaning. *'^ Some of their work has been unconvincing, and some
can even seem laughable; but some has demonstrated that at times we misun-
derstand early music because we fail to recognize how differently people saw
things in previous eras. This book includes many examples of how culture has
influenced the meaning of music — say, the meaning of different dances in

Beethoven's day (in addition to whatever meaning those dances may convey in-

trinsically). But we should take care not to exaggerate the "otherness" of those
who did the dancing. Several myths of medieval "otherness" have been de-
bunked in recent years: it is no longer reasonable to say that medieval people

41. Edward M. Burns and W. Ward Dixon, "Intervals, Scales, and Tuning," in The Psy-
chology of Music, ed. Diana Deutsch (New York: Academic Press, 1982), pp. 241-69.
42. Do the component frequencies that strike our ears at a given moment come from sep-
arate sources in our environment, or are they parts of a single sound from one source? To fig-

ure this out, one of the tricks the brain uses is to analyze whether any of the frequencies be-
long in the same harmonic series. If some do, the brain guesses that these ones are parts of a
single tone —and in the natural world, this is usually accurate. Steven Pinker relates this to
our response to musical intervals, in a brief section of his forthcoming How the Mind Works
(New York: Norton, 1997; page numbers unavailable as go to press). I

43. Burns and Dixon, "Intervals, Scales, and Tuning," pp. 258-59.
44. See Paul Fraisse, "Rhythm and Tempo," especially pp. 151-55, in The Psychology of
Music, ed. Deutsch.
45. Richard Leppert writes, "(Meaning) develops not only, and maybe not even principally,

from what's 'in' the music . . . but more from the purposes or functions to which music is put
(inevitably different for different people at any given moment, and inevitably changing in his-

torical time)." "The Postmodern Condition and Musicology's Place in Humanistic Studies,"
journal of Musicological Research 12 (1995), pp. 235-50.
A caveat regarding these attempts is Charles Rosen's "Music a la Mode," New York Re-
view of Books, 23 June 1994, pp. 55-62.
94 MEDIEVAL MUSIC, PLAINCHANT, AND "OTHERNESS"

had no idea of childhood as a separate stage"**^ or, as I'll discuss below, no ex-
perience of romantic love before the twelfth century. (And in music it is no
longer reasonable to say, as once was said, that medieval people had only a
weak sense of harmony.'*^) If music's human context influences its meaning,
then we had better understand humans as accurately as we can. We need to
recognize not only the historical differences, but also the transhistorical conti-
nuities.

To illustrate why this might be worthwhile, consider one of the most cen-
tral aspects of musical meaning: emotion. Wegman may be right that we do not
"hear just like" medieval people; and there's no denying that our world views,
lifestyles, technologies, economies, political structures, class structures, and —to
return to music — listening contexts differ markedly from theirs. So do our aes-
thetics and our pool of artistic experiences. Some of these things affect musi-
cal experience. But given the research I've summarized, it is no longer naive to
suspect pace Wegman, that we feel more or less as medieval people did. Many
psychologists now accept that emotions often have functional origins in our
species' evolution, and are not purely a matter of acculturation. ''^ Some of our
emotional "calls" (such as crying and laughing) are like the facial expressions
I discussed: they convey the same emotions in all cultures. Romantic love was
once written off as an invention of the West in the twelfth century,'*^ and an-
thropologists portrayed certain cultures as being free of it;^° but it is now re-

garded as a human universal.^' Of course, different cultures place different val-

uations on romantic love. In our culture, it's considered one of life's highest
achievements, but some cultures regard it as something to avoid or at least hide.

But even in those cultures romantic love erupts regularly, with all the intensity
(and all the components) that it has for us.^^
Thus, the longing of a medieval song may not be so "other" from modern

462-63 of Michel Rouche, "The Early Middle Ages in the West,"

46. See, for example, pp.
in A I: From Pagan Rome to Byzantium, ed. Paul Veyne, trans. Arthur
History of Private Life,
Goldhammer (Cambridge, Mass.: Harvard University Press, 1987).
47. As was shown by R. Crocker, "Discant, Counterpoint, and Harmony," Journal of the
American Musicological Society 15 (1962), pp. 1-21.
48. The Nature of Emotion: Fundamental Questions, ed. Paul Ekman and Richard David-
son (New York: Oxford University Press, 1994), pp. 15-25, 146-77; and Richard Lazarus,
Emotion and Adaptation (New York: Oxford University Press, 1991).
49. Denis de Rougemont, Love in the Western World, trans. Montgomery Belgion, rev. ed.
(New York: Pantheon, 1974).
50. Re the Chinese, ibid., p. i; Margaret Mead, Coming of Age in Samoa
re Polynesians,
(New York: Morrow, 1961 11928]);Donald Marshall, "Sexual Behavior in Mangaia," in
Human Sexual Behavior, ed. Marshall and R. Suggs (New York: Basic Books, 1971), pp.
103-62.
51. Romantic Passion: A Universal Experience?, ed. Jankowiak; Jankowiak and Ted Fisher,
"A Cross-Cultural Perspective on Romantic Love," Ethnology 31 (1992), pp. 149-55.
52. Helen A. Regis, "The Madness of Excess: Love among the Fulbe of North Cameroun,"
in Romantic Passion, ed. Jankowiak, pp. 141-49.
 THE COLONIZING EAR 95

longing. Of course, that and other basic emotions may be embedded in beHef
systems we find foreign and in musical idioms we don't speak fluently. Still, it
seems reasonable to suppose that transhistorical emotional resonances have
something to do with why old music means more to us than wild parrot
squawks. They may also give the historical performer a bit of hope that what
he or she is expressing can be, if not identical, at least closely related to what
the music was expressing in its own day. At the very least, they provide one
more reason for giving the idea of transhistorical humanness a careful hearing.
 THE RENAISSANCE, OXBRIDGE,
AND ITALY

In 1994 a lecturer in the music department at Cambridge University asserted,

in a memo to his students, that "Recent recordings [of Renaissance music by
British artists] are in many cases unspeakably dull and performed with a des-

perate absence of historical awareness." He went on to damn most of the

prominent English ensembles. The leader of an ensemble that was treated fairly
leniently —
Christopher Page, who teaches in another Cambridge department
relayed thedocument "in amazement" to the readers of Gramophone.^
A couple of months later, the author of the document, the music historian
Roger Bowers, responded with a letter backing up his complaints.^ He com-
pared using women's instead of boys' voices to using "members of the violin
family for music conceived for viols." He called upward transposition by a tone
or a third "a pretty miserable experience" (such transposition is now discred-

ited, largely by Bowers's own work, but is still employed by some British per-

formers). Beyond that, he argued that "the performances are far removed from
the spiritual and aesthetic sensibilities of the period. . . . Rather, they tend to
respond to the wholly alien concepts of the modern Anglican tradition. . .
.^

1. Christopher Page, "The Listening List," Gramophone 72 (December 1994), p. 8.

2. Roger Bowers, "The Listening List," Gramophone 72 (February 1995), p. 6.

3. American readers may wonder what Bowers means here. Chris Hunter (personal com-
munication, 1995) answered my query by saying that it probably refers to a performing style

in —
which the chorus dramatizes the text e.g., slowing down for the Crucifixus or speeding
up for the Et resurrexit. This approach grew out of the Victorian revival of the Anglican cho-
rus, and appears not to have been part of church music performance in the Renaissance. The

issue is discussed briefly in the Peter Phillips interview.

97
98 THE RENAISSANCE, OXBRIDGE, AND ITALY

The large and resonant tone production of the singers and its expansive pro-
jection conflicts with the conditions under which the music was originally con-
ceived and performed —namely, the sense of intimacy and introspection gener-
ated within a late medieval chancel or chapel by singers performing for
themselves and for their deity alone, there being no congregation in attendance
to edify or entertain."
All of which may be valid —though the success of the Tallis Scholars sug-
gests that many people find a little upward transposition far from "miserable,"
and reminds us that when Bowers sets out aesthetic judgments he can no longer
claim to be speaking objectively. But what strikes me most is that Bowers ob-
jects mainly to practices that present the music in modern conditions. If you
use boy choristers, forget about grueling international tours. And if you are per-
forming for an audience, you may well find it impossible to sing like people
who didn't have one. If you do manage not to project to your audience, you
may not have one for long. If it comes down to it, which would you sacrifice:

the history or the audience?

This is an issue that recurs throughout early music: how far should (and
can) we go toward the past, as opposed to trying to bring it toward us? Peter
Phillips argues cogently for the latter orientation; Bowers and others, also with
good reasons, argue for the former. And Paul Hillier argues that the issue itself

may be anachronistic.
If we do want to go toward the past, one obstacle we face is that much of
what musicians did in the past they did not write down. Some of the notes they
performed were improvised; others were written down without all the infor-
mation we'd need about inflection and even pitch. Modern classical training

enjoins us to honor the written notes, so (as Andrew Lawrence-King observes)

learning to go beyond the notes is in some ways subversive, or at least terra
incognita. Paul Hillier mentions another problem: even when we feel that all

the notes we need were written down, modern forms of notation differ from
older forms so profoundly that it can change how we sing the music.
Finally, "Renaissance" is a more likable term than "Middle Ages": most me-
dievalists object to their period being labeled as a "middle" ground lying be-
tween other, implicitly more interesting times; but who could dislike a rebirth?
And unlike "medieval," "Renaissance" does seem to apply to a real historical
development. But for musicians, the latter term is problematic. For one thing,
compared to art or literature, it's harder to say just what is so "Renaissance"
about the era's most prominent music, the polyphonic vocal works. Christo-
pher Page speculates about this in his interview, and Peter Phillips adds further
observations on the issue. In Andrew Lawrence-King's chapter, definition again
proves troublesome, in that we find the dividing line between "Renaissance"
and "Baroque" to be a little arbitrary. Not long ago, Monteverdi was usually

described as the father of the Baroque, but an influential recent book called
 THE RENAISSANCE, OXBRIDGE, AND ITALY 99

him "the foremost ItaHan composer at the end of the Renaissance.'"* I include

his chapter here, rather than in the section on Baroque singing, because it re-

lates to the issue of "Oxbridge" style, discussed by Hillier and Phillips, and to
the views of Italophiles who would prefer that Britons find other work for their

vocal cords.

4. Gary Tomlinson, Monteverdi and the End of the Renaissance (Berkeley: University of

California Press, 1987), p. ix.

 5
There Is No Such Thing as a Norm
— oOo^ g^

Paul Hillier on Renaissance

Sacred Music

The mainstream repertory is so heavily Germanic that a novice of an earher

generation might easily have imagined that classical music had always been
Teutonic territory. But in the art music of the Renaissance, the crucial region
was what is now Belgium and northern France. Franco-Flemings like Dufay
and Ockeghem dominated the fifteenth century, and their follower Josquin
Desprez established the polyphonic style that dominated the sixteenth century.
It's true that the English were an early stimulus to these Franco-Flemish mas-
ters —the was a
sweet, thirds-dominated "English countenance" of Dunstaple
—
on Dufay and Binchois but England usually remained some-
crucial influence
what separate from developments on the Continent. As far as the sixteenth-
century Franco-Flemings and Italians were concerned, England was a musical
backwater.
Thus there is a certain irony in today's early-music revival: the great Franco-
Flemish polyphonists have been popularized mainly by English groups, such as
Pro Cantione Antiqua, the Clerkes of Oxenford, the Sixteen, the Taverner
Choir, the Oxford Camerata, and various cathedral and university choirs. An
irony; but not an accident. The choral tradition in England's colleges and
churches, has, as Christopher Page points out, produced an astonishingly large
pool of skilled choral singers, who can read difficult music at sight, stay per-

fectly in tune even when singing challenging harmonies, and blend skillfully
with each other. No other country has such a resource concentrated in so small
a geographic area. Moreover, much of the classical recording industry is in Lon-
don, around which this brigade of singers stands on call, and recordings have

100
 THERE IS NO SUCH THING AS A NORM 101

long been the main means of propagating early-music performance. Thus, most
listeners have heard their Dufay, Ockeghem, and Josquin with a new English
countenance, called by many the "Oxbridge style" in honor of its main prop-
agators' alma maters.
As Page noted in his chapter, on the Continent this style has won approval
but far from universal acclaim. Even within Oxbridge it has detractors, as the
Roger Bowers controversy shows. All the dissent, though, is at least in part a

reaction against enormous success. David Fallows writes that, although the sit-

uation is changing, since the early 1970s "it has looked very much as though
British musicians were the almost unchallenged leaders in the performance of
music before 1550."'
Our next chapters feature two leading representatives of the Oxbridge style
who have taken opposite paths. Peter Phillips, an Oxford graduate, founded
the Tallis Scholars in 1978 (though the group had been singing without the
name before then); Patrick Russill writes that the group has always seemed to
him "not so much a product of the 'early music and authenticity' movement,
but rather a perfectly logical development of the English choir-stall tradition."^
In 1974, the baritone Paul Hillier, a graduate of the Guildhall School of Music,
co-founded something rather different to perform much the same repertory: a
quartet with three other singers, two of them Oxonians. (The group was named
after the sixteenth-century miniaturist painter Nicholas Hilliard —not, as many
assume, after Hillier.)

Phillips shows no inclination to do anything but continue exploring the

Renaissance repertory with the Tallis Scholars, using London as his home
base. Hillier, though, left the Hilliards in 1990 and moved across the Atlantic.

He first joined the music faculty of the University of California at Davis,

where he formed a new ensemble, the Theatre of Voices. Then, in 1996, he
took the directorship of the Early Music Institute at the University of In-
diana (a post previously held by the late Thomas Binkley, a pioneer of the
early-music movement). Hillier has also worked closely with the Estonian com-
poser of mystical minimalism, Arvo Part — an interest that Phillips, as we'll

see, doesn't share.

Phillips and Hillier have differed on many musical issues: choral size (the
Tallis Scholars use ten or so singers; the Hilliards sing one-per-part, as the The-
atre of Voices occasionally does); pitch (the Tallis Scholars transpose upward);
pronunciation (Phillips has little interest in regional differences in Latin pro-

nunciation, Hillier finds them crucial); use of musica ficta (Hillier applies these
unnotated sharps and flats more freely); the use of a conductor (the Hilliards
don't have one, the Tallis Scholars —and, usually, the Theatre of Voices —do;
1. "Quarterly Retrospect," Gramophone 71 (December 1993), pp. 33-34. The article is a
stimulating survey of the rise of non-British groups who are performing this music.
2. Gramophone 73 (June 1995), p. 104.
102 THE RENAISSANCE, OXBRIDGE, AND ITALY

Renaissance choirs generally had no conductor, relying on rehearsal, stylistic

empathy, and, probably, eye contact to stay together). Above all, the two men
differ more than one might expect in their basic aesthetic orientation — compare
Hillier's doubts about "seraphic refinement" to Phillips's admiration for the
Berlin Philharmonic, or the two men's views on "Continental" versus
"Oxbridge" sound.

As these differences demonstrate, "English" ideas of style and taste can vary
considerably —though that's influenced by (among other things) the inconclu-
siveness of much of the evidence, and by questions about how carefully we
should attend even to evidence that is firm. These were the first issues I raised
when I interviewed Hillier in his office in the music building at Davis.

When singing a Renaissance piece, we usually have less certain information

about performance practice than when we perform a Classical or Romantic
piece.

Yes, but I think the music in itself contains virtually all the information you
need to unlock its beauty; and while the particular way in which this beauty
manifests itself is obviously not unimportant, it is liable to adaptation accord-
ing to the means used to perform it.

Definitive realization —the idea that there's one right way to orchestrate and
perform a piece— is a much later concept.

I think the most important thing to remember regarding this music is that
there is no such thing as a norm to which we must try to adhere, which means
that the question is very open. It is important to avoid assuming that this music
should sound the way it has always sounded to us — in other words just adopt-
ing a standard choral format and singing style, and assuming that the sound
and balance they're producing is the right one for the music.

I'd like to deal with various areas of uncertainty one by one. One of them
is the size of the ensemble. There's evidence of group sizes ranging from a total

of four through fifty or more.

It's pretty obvious, as you say, that choral sizes varied tremendously, and
not only to fit the size of the given building they were singing in, but also to
reflect different tastes about what the music should sound like. The same piece
would surely have been performed differently in different places.
My approach to this may be excessively pragmatic, but I believe that the
music is there for us to use and to suit to the forces we have, rather than al-

ways requiring that we play around with the forces to try to suit the music. We
may feel there is an ideal grouping for a given piece or genre, but that doesn't
mean to say one should never perform it with groups of different sizes — large
 THERE IS NO SUCH THING AS A NORM 103

choirs, small choirs, one to a part. Of course, it also depends on the piece
you're talking about. But I think that one can be very open-minded about it,

allowing for the fact that the more you do a certain repertoire the more you're
likely to develop a relatively fixed sense of your ideal.

Is there music you prefer with small or, for that matter, large ensembles?
Well, for example, when I work on Dufay and his period, I've been inter-

ested in the research of scholars who are trying to work out the size and bal-

ance of his ensembles. The evidence seems to point towards a larger number
of singers — falsetto male sopranos —on the top lines, and fewer on the lower
lines.' That music tends to have a strongly melodic upper line that clearly can
take a certain amount of emphasis. This seems to tell us what to do; but then
we also have to remember that our chorus members surely sing in a different
way than people did four hundred years ago. Today, the singing voice is

"placed" in a way that differs radically from the placement we use when we
speak, but I (and others) suspect that back then it wasn't" —as it still isn't, for

example, in folk singing. Today's singing technique creates a legato stream of

sound and breath support that tends to increase the size of the voice. So maybe
our falsettists, for example, produce an individually stronger sound than their
fifteenth-century counterparts did; if so, maybe the need for greater numbers is

obviated. Even we can establish exactly how many singers there were then,
if

we still don't know if we're reproducing the sound of the original, because our
ways of singing have changed.
So you're left, really, relying on your own musical intuition as to what the
ideal balance is, how the music works best. You can't really get away from that.
To give another example, I'm very interested at the moment in the music of
Byrd, and there have been a lot of very fine choral recordings of his music. But
I'm convinced that the Gradualia, which is a very significant part of his out-
put, was performed by solo voices, one per part. The music was probably per-
formed at secret Catholic services,^ and though they could perhaps have had

more than one singer per part, it's unlikely. And if you look at the music, the
nature of the individual part-writing is often much more florid, sometimes only
in one part at any given time, and I find it sounds much more effective with

one singer to a part than with a whole choir. Again, I want to avoid a blanket
assertion that I don't think it should be sung by a choir; but I just think it's

3. See especially David Fallows's essay, "Specific Information on the Ensembles for Com-
posed Polyphony, 1400-1474," in Studies in the Performance of Late Medieval Mustc, ed. S.

Boorman (Cambridge University Press, 1983), esp. pp. 120-33.

4. See John Potter, "Reconstructing Lost Voices," in Companion to Medieval and Renais-
sance Music, ed. T. Knighton and D. Fallows (London: Orion, and New York: Schirmer,
1992), pp. 311-16.
5. See Joseph Kerman's essay (aptly described on its book jacket as "moving") entitled

"William Byrd and Elizabethan Catholicism," in his Write All These Down (Berkeley: Uni-
versity of California Press, 1994), pp. 77-89.
104 THE RENAISSANCE, OXBRIDGE, AND ITALY

time we had a look from the other end of the telescope. There, you see, you
have a special case, Byrd and the position of the Catholic Church in England
at that time. It's not a rule you can apply indiscriminately.

How about Byrd's Anglican works, such as the Great Service.^

That's different again. We know the standard Anglican choir of today is

quite appropriate for that — at least in terms of overall size.*^ It's a different kind
of music, too; it doesn't have those little virtuosic moments in quite the same
way that you find in the Gradualia.

Another area of uncertainty is instrumental accompaniment. For example,

written-out organ parts exist for Victoria, among others.
Again, the probable answer is that there were varying practices. There is

strong evidence for instrumental participation in Isaac, for example, in the

court of Maximilian, where he worked, and in Lassus too. It seems some-
what more common later in the Renaissance than earlier. But I happen to be
biased towards the sound of a cappella singing: it's as simple as that. So even
where there's strong evidence for using instruments, both for practical rea-
sons and for reasons of personal taste I usually opt for a cappella. I suppose
it's a case of liking what I know. But there is some music —for example the
Lassus Penitential Psalms —which I've recorded using voices and instruments.
And I can still imagine doing that music in both ways and getting a great
deal of pleasure out of it. They're different, but both options seem to me
equally viable, and in either case the character of the music is altered —though
not fundamentally.

You mentioned voice production in Dufay's time; what do we know about

Renaissance voice production? I'd imagine this would be especially hard to re-
construct. And do you have any comments about using vibrato in this music?
What we "know" is actually very little. What we can imaginatively recon-
struct depends on cross-referencing all sorts of information (from linguistics,
the nature of instruments, the history of liturgy, etc.) —none of which is re-

motely conclusive from the singer's point of view. The reason I would give for
not using excessive vibrato is that it obscures the counterpoint and muddies the
tuning, making the music of the era quite simply less effective —disastrously so
in some cases.

Another area where information is incomplete is the use of unwritten acci-

dentals —the convention of musica ficta, as it was called. It suggests that some

6. Eric Van Tassel, however, observes that many AngHcan choirs today have a higher pro-
portion of trebles than in the sixteenth and seventeenth centuries. See, e.g., his "Purcell's Sa-
cred Music on Record," Part 2, Early Music 24 (February 1996), p. 92 n. 11.
 THERE IS NO SUCH THING AS A NORM 105

of the accidentals that the composer assumed would be sung in a piece weren't
written down, but were understood by convention.
I'm going to have to generalize, but providing you're using a good edition
(and there are a lot of them nowadays) that issue will have been taken care of.

The best advice, then, is to follow the editorial suggestions (usually written
above the notes) unless you really know what you're doing. It's a very involved
issue, since the more you deal with that kmd of music, the more you will want
to introduce changes, and the more you'll be in a position to do so. But it's im-
portant to know clearly what is original and what the editor has added. Dif-
ferent editors will solve the same ficta issues in slightly different ways.
There's been a tendency in the past years towards adding more ficta so that,
"^
particularly in the middle Renaissance period, the music is not so "modal.
Also, the guiding principle gives weight more to the logic of the individual line

as against the resulting harmony. We have to remember that in earlier centuries

each singer had only his own part in front of him —they didn't sing from a

score. So their initial response would have been based very much on their own
part first, and then the other parts and the total harmonic picture afterwards.
That brings up something else that maybe you're going to ask me about
the appearance of that part, as opposed to the way it appears in a modern
score. Modern notation puts a bias on the information you're receiving that is

very different from what the original notation tells us. It's not only that early
notation usually presents a single part by itself rather than a score. It's also,

first of all, that you do not have barlines, which divide the music up far too
frequently: you get a long line of sound, which in a modern edition is broken
up into little segments of I or i or something. The barline not only divides it

up, it also suggests a regular emphasis or a beat which is inappropriate to the

music most of the time. Visually, the flow of the line and phrase is impeded by
these barlines. And the second thing that's different is the nature of the note
values that we use, as opposed to the originals — all this does affect the way we
sing the mus'c. As just one example we might consider a well-known motet by
Byrd, such as Ave verum corpus. Looking at the familiar editions used by most
choirs, we can see straight away that the editor has responded as well as he
can to the music's demand for a flexible sense of meter, changmg frequently
throughout the piece between duple and triple time signatures. The problem is

that there are still many, many passages where the voices either are in a mix-

7. One group that has tended to add ficta very sparingly in earlier repertoire is the Tallis
Scholars (I failed to ask Peter Phillips about this). While their practice may represent an
anachronistic preoccupation with the written text and may not re-create period practice, it

may not be entirely unrelated to Renaissance musical philosophy. Charles Rosen says that the
very term musica ficta ("fictive music") suggests that a version that was never actually
sounded —the purely —
modal one on the page was considered "real" (Rosen, The Romantic
Generation (Cambridge, Mass.: Harvard University Press, 1995], pp. 28-29). That "real ver-
sion" is what the Tallis Scholars are sounding in some of their recordings.
106 THE RENAISSANCE, OXBRIDGE, AND ITALY

ture of duple and triple rhythms, or else simply have their "downbeats" at dif-

ferent times. No system of barring can do justice to this, and the result is a
mixture of seemingly displaced accents and superfluous syncopations. Even tak-
ing a line at a time, the use of barlines, however flexible, imposes a set response
in place of the wonderfully supple and expressive ambiguity of the original.
Skeptics will argue that the sounding result is pretty much the same in either

case, but I beg to differ. We hear the same sequence of abstract notes, certainly,
and the difference may be rather subtle at any given moment, but the net re-

sult is like viewing the music through tinted glasses. Ultimately, we lose a sense
of rhythmic freedom and "natural" phrasing, replaced by a kind of foursquare
security borrowed from the conventions of a different age.
It would be wonderful if people wanting to perform early music could use
either a facsimile of the original or a transcription (known as "diplomatic")
using a more readable adaptation of the original style of notation. It might be
impractical to do that all the time; but it would be a very informative experi-
ence if singers would once or twice work their way through a piece in the orig-
inal notation, and just learn the kinds of information that are conveyed by that

notation. Sometimes the differences are very subtle; one wouldn't want to in-
sistupon them too heavily. But then it would, I think, be possible to go back
to using modern editions armed with this information and not simply having
been told by someone like me that the trouble with these editions is that there
are too many barlines. It's one thing to say that, and another to have the ex-
perience of singing that music without the barlines.

When do you feel that the barline becomes a musically significant element,
as opposed to a modern intrusion?
I suppose, loosely speaking, the middle to late sixteenth century. It seems to
develop first in lute songs; soon after, in polyphony and larger court music, the
barline is more clearly related to an audible metrical emphasis. There is a fairly
clear division in the polyphonic field, at least in my mind, and I suppose it falls

right around the end of the sixteenth century, where suddenly the use of the
barline makes sense and doesn't get in the way.** But with Palestrina and all

those people it doesn't really help. It helps us because we're used to it, but it

doesn't help the music.

Returning to the theme of incomplete information, I wanted to ask also

about transposition —the possibility that a piece was meant to be heard at a

higher or lower pitch than the notation appears (to us) to imply.
I think the important thing is to be clear that there was no such thing as a
fixed pitch standard, that it varied from place to place —maybe not wildly,

8. For an explanation of why the barline became musically important after this point, see
the interview with John Butt, Chapter 9, in the section on "grammar."
 THERE IS NO SUCH THING AS A NORM 107

though maybe more than we might expect. And because they didn't use ex-
treme key signatures — in the Renaissance, usually at the most two flats or
sharps —the concept of color that later centuries attach to keys just did not
exist. So it's perfectly appropriate to perform a piece at more or less any pitch.
In practice, one's talking about transposing a half-step or so, depending upon
what your choir needs. Again, I would say, let practical considerations rule:

make sure it's pretty much in the range of your choir or your singers, and then
you can't go far wrong. Obviously, there are pieces where low registers are
being exploited, so you wouldn't want to transpose them from low to high
though you could argue that as long as it sounds low on the spectrum of what-
ever voices you have, then the relative sense of depth has been retained.
There is a separate issue with regard to certain combinations of clefs. It

seems that in sixteenth-century polyphony, some pieces were actually sung a

third or a fourth lower, and certain clefs were used to avoid the necessity of
unusual key signatures. You may have a collection of motets in which suddenly
there's this motet where every voice is at the top of its range, and it all seems
very impractical; you look at the clefs, and sure enough, the bass part is in the

baritone clef, and so on. If you transpose it down, it's in line with the range of
the other motets. In modern editions, usually the original clef is printed before
the new one, so that even if everything is transposed into "normal" clefs you
can see at the beginning what the original was.
Then there's another side issue, the question of English polyphony of the
early sixteenth century, which certain scholars'' are sure sounded a minor or
even major third higher. I'm not totally convinced by this, just because I feel

that the result is wonderful for five minutes but then is quite tiring. I'm not
saying it shouldn't be done, but I don't think we need all follow in that direc-
tion. Instead —and this would again be a pragmatic approach —we might keep
in mind the possibility that generally the music needs to be shifted up, say, a
half-step or so.

Another topic of uncertainty: intonation systems. Modern equal tempera-

ment, of course, hadn't developed yet, so one reads about just and mean-tone

and Pythagorean tunings.

Well, again, unfortunately, it's a very large topic, and the application of it

changes as you move through history. Ultimately I would argue that it's im-

9. Notably David Wulstan: see his "The Problem of Pitch in Sixteenth-Century English

Vocal Music," Proceedings of the Royal Musical Association 93 (1966-67), pp. 97-112. Wul-
stan's view has been countered by Roger Bowers, who demonstrates that {as Hillier says) there
were no established pitch standards at this time; see his "The Performing Pitch of EngHsh
15th-century Church Polyphony," Early Music 8 (January 1980), pp. 21-28; "Further
Thoughts on Early Tudor Pitch," Early Music 8 (August 1980), pp. 368-75; and "The Vocal
Scoring, Choral Balance, and Performing Pitch of Latin Church Polyphony in England, c.
1500-58," journal of the Royal Musical Association 112 (1987), pp. 38-76.
108 THE RENAISSANCE, OXBRIDGE, AND ITALY

portant not to rehearse with a piano. The choir will probably settle on their
own naturally into a good tuning system, all other things being equal — if they
aren't tired, if the music isn't too high for them, if they're reasonably good
singers, and so on.
Talking about unaccompanied polyphony, one thing I've found very helpful
when there's an intonation problem is to transpose the music up or down a

half-step (so we return to the topic of transposition). In other words, you're

moving the music from the white keys of the piano to the black keys. And for

—
some reason and I have no idea what it is that very often solves all the — in-

tonation problems. Now this shouldn't be the case, unless we have the piano's
tempered tuning so ingrained that we can't get away from it. For example, F
major is a very problematic key when it comes to tuning, so I move F major a

half-step up and usually it falls into place. That's totally unscientific, but it just

seems to be something that works.

I think that, beyond that, the most important thing in the question of into-
nation, rather than worrying about given systems, is to give very close aural at-
tention to the principal intervals, the octave, the fifth, and the fourth; also, the

note just above the tonic is worth watching. If those are in tune, everything else
will probably follow. Just to get a choir to practice tuning a perfect fifth is ac-

tually quite an interesting exercise. There are lots of other little practical tips

that one can follow up with. For example, basses generally need to think a lit-

tle bit higher, particularly in a descending phrase. So if the basses think more
like high baritones, even when they're singing at the bottom of their range, and
the higher voices are a little bit more deeply rooted, then they can begin to
come together. This is verging more on choral music generally —these are im-
portant considerations for any period of music, not just the Renaissance — but
given the relative lack of chromaticism in Renaissance music, those principal
intervals become that much more crucial.

Another area of uncertainty is the question of improvisation. It's been sug-

gested that there were two kinds of music in the High Renaissance: what the
composer had put on paper, and then what was performed, which was heavily
°
ornamented. '

The fact that polyphony was probably ornamented some of the time does
not make me want to ornament it, unless I am performing one to a part and —
even then, quite frankly, not very much. I also think that if anyone is going to
concern themselves with ornamentation they need to spend a considerable
amount of time not only practicing it but exploring the treatises themselves and
thoroughly absorbing the kinds of configurations that were used.^^ I have to be

10. John Butt, personal communication, 27 April 1993. See Howard Mayer Brown's Em-
bellishing Sixteenth-century Music (Oxford University Press, 1976).
11. In his interview, Robert Levin makes a similar point with respect to Mozart.
 THERE IS NO SUCH THING AS A NORM 109

honest: it's something that holds very Httle interest for me. With a lot of peo-
ple — especially singers who come to Renaissance music from Baroque music
almost the first thing they want to know is how to ornament, but I would say
it's the last thing they should worry about. Yes, it undoubtedly played a role
in performance, but there's so much else to learn that it seems to me it's putting
the cart before the horse, or whatever the appropriate saying is. Especially in
polyphony, apart from an occasional cadential thing, I have to be honest: it just

doesn't interest me. It's a purely personal reaction. When you go into the

Baroque, that's different, of course. But I've often heard it done so badly.

Bovicelli, Rogniono, and others wrote down examples of how singers or-
namented Palestrina motets, but there were complaints from contemporaries
about this practice. More recently, Alfred Einstein said the Bovicelli versions
were "monstrosities," and Howard Brown said they show that "bad taste is

not the exclusive property of the present century."^ This relates to the debates ^

about authenticity: one ideal of authenticity is to play the music the way it was
played in the composer's era, but could this be a case in which the modern style

of playing it, unornamented, would — at least to our ears —serve the music bet-

ter than the historical practice?

Maybe. Music that wasn't virtually contemporary in those days didn't get
performed, or, if it did, we can be sure that was reworked according to the
it

tastes of the day; so this whole notion of authenticity is thrown out the win-

dow by the very people we're trying to emulate. They lived and breathed one
style of music essentially, which was the style of their contemporaries. This is

something we have no idea about; I think we have no conception of what that

could feel like. If we grew up hearing one kind of music then we might be in
a position to start embellishing it authoritatively. But I still say, though I've

heard some very good performances of sixteenth- and seventeenth-century

music that used embellishment, many have been awful. Another important
thing to remember is that embellishment exists for only one performance. It

shouldn't be reproduced exactly from performance to performance, otherwise

it's going to become rigid. So, again, if it's done very well, I'm interested, cer-

tainly. In fact, in the Theatre of Voices we've had Drew Minter singing, and
he's someone who embellishes as naturally as he breathes; you don't hear the
extra notes as "ornaments," but as part of a natural response to the music.
Done like this, it works — but now we're talking about one-to-a-part singing.
All the same, are there any places where you find ornamentation or em-
bellishment necessary?
Before about 1600, I wouldn't say so. Again, I would take any given ver-

sion on its own merits. There's a school of thought that says the use of fer-

12. Brown, Embellishing Sixteenth-century Music, discusses this on p. 73.

110 THE RENAISSANCE, OXBRIDGE, AND ITALY

matas in some passages in Dufay indicates a place where someone could em-
bellish the notes. Fine: let's hear someone do it. But it takes a lot of know-
how and just sheer application over a period of time to get inside the music
and the history of it —and not just the history, but also the kinds of things
that might have been done to make the embellishment work.
The theme we've been returning to here, with embellishment being one
example [and ficta another], is that the music on the page is not necessarily
quite representative of what was actually heard. '^ I totally agree with that;

but I would also say that it allows one to take other kinds of liberties as
well, and not just the addition of extra notes. And these concern articulation,
expression, tempo, all the things that we don't really hear about in Renais-
sance music.

Let me ask you about them, then. Can any general principles be stated

about articulation in Renaissance music?

No, I don't think so. I don't think that one can lay down any specific

guidelines as to how to perform a piece of Renaissance polyphony in quite

the way you can for some later music. And even for later music there are
lots of ifs and buts. There are a number of things that can be said on this

subject, and the more you perform this music the more possibilities you see

in it. The only general statement I can make is that, again, you shouldn't as-

sume that what you've been hearing is the way it should be. I don't think
one should automatically turn on the all-purpose church legato when there
are so many other possibilities.
Again, it has to be ultimately a matter of taste, as to how you treat the

fast figuration versus the slow notes, whether it's clearly articulated, whether
you should ever use what we would call a staccato, and so on. Unfortunately,
without specific examples it's very hard to talk about this issue, which I think
is one of the most important.

Can you give an example?

In the music of one of my favorite composers, Ockeghem, there are often
many different things happening at once. And I don't think there should ever
be any dead ground. So even the slower-moving phrases should be phrased
in an expressive way, whatever that may mean. I'm thinking also of the music
of Dufay, where the slow-moving parts should be sung with just as much care
for their shaping as the more obviously expressive fast-moving parts. This

13. James Haar suggests another angle to this: that notated music, e.g. some florid lines

in Josquin Masses, attempted to capture something of how

might have impro-
skillful singers

vised. This, he argues, suggests a different approach to performing such pieces than that which
is common today —
a more improvisatory, less solemn manner. See Haar, "Monophony and
the Unwritten Tradition," in Performance Practice: Music before 1600, ed. H. M. Brown and
S. Sadie (London: Macmillan, 1989; New York: Norton, 1990), pp. 240-66; reference, p. 260.
 THERE IS NO SUCH THING AS A NORM 111

brings the music into sharper rehef, and makes it more interesting and more
expressive. I have no idea whether it was performed that way, but that's the
way it interests me.

How about specific issues of articulation per se, though?

Your first clue, I think, has to come from the The nature of the lan-
text.

guage, in so far as you can understand how was pronounced and used, has
it

to have a strong influence on the way in which you sing and enunciate the
text through the music.'" If you go back to styles of polyphony where one
syllable stretches over a long period of time, you lose that particular crutch,

obviously. But in less florid music, clearly it's important to understand how
the language that you're singing was pronounced. This refers not only to the

vernacular languages — French, English, Italian, and so on but also to — Latin,

which we know varied quite considerably from country to country. It still

doesn't give us nice easy solutions, hard and fast rules, but it certainly is a

way of opening up our ears and minds to what kinds of possibilities are avail-

able. It is generally accepted that Latin pronunciation in sixteenth-century

England lacked the smooth Italianate quality that we give Latin today. Its

pronunciation and vocal placement were probably much closer to that of the
spoken English of the time. As a result, the Latin pronunciation would have
had a "lived-in" quality —and of course for some people Latin was still po-
tentially a second language of communication.

Can you give an example of how it differed from modern Latin pronun-
ciation?
Certain consonants and vowels were different. The diphthongs, for exam-
ple, were emphasized differently. We tend in Italianate Latin to iron them out,
making the vowels as "pure" as possible; singers are trained to extend the

first half of the diphthong and then slip in the second half as briefly as pos-
sible. By contrast, my philology teachers tell me that the first element of a

14. Hillier discusses this issue in more detail in his essay "Framing the Life of the Words,"

in Companion to Medieval and Renaissance Music, ed. Knighton and Fallows, pp. 307-10.
He says that modern voice production "is significantly different for singing than for speech"
modern vocal technique being designed to sustain sounds for "quite long periods" at "con-
siderable volume." As a result, when we sing "the physical properties of the language do not
seem to be felt ... as they are when the language is spoken. Instead there is a barrier of pro-
duced sound, between the singer and the words."
a continual deposit of 'expressive' sonority,
Most early-music modern, but "the singer is told not to use vibrato, not
singing technique is

to do this or that with regard to dynamics, tempos, tone-colour, contrast, and other expres-
sive devices. Thus, the natural connection between emotion and voice, thought and its ex-
pression, words and music is inhibited and enveloped in a set of cautionary restrictions." He
argues that period pronunciation can provide a positive framework for enlivening early-music
singing.
Regarding the "authenticity" of period pronunciation, Hillier notes that Josquin's music
was sung in many different parts of Europe, where Latin pronunciation was far from uniform.
112 THE RENAISSANCE, OXBRIDGE, AND ITALY

diphthong would have been sounded quite briefly, giving longer emphasis to
the second element. In thickly texted polyphony this is going to make a big
difference.

Can you give an example of a work where this makes a differenced

Well, we've already mentioned Byrd. So you could experiment, in his Latin
works, by pronouncing the Latin with a heavy regional English accent (as op-
posed to the standard BBC accent that's preferred today). First just notice the
difference that produces, and then try to sing it in the same way. This is not
producing a "correct" result, but it's producing a different result (similar in

kind to the one I'm talking about) and, hopefully, opening one's ears to the idea
that there are more than one or two ways of doing it.''^
It's making something strange
a difficult issue, because merely the fact of
in this case the pronunciation —
might cause you to overreact and say, "Oh yes,
the music has to be very different here like this." And we haven't proven any-
thing. All we've done is to create a different set of possibilities. That's all, but
I think that's important.
Also, you have to accept that the result won't seem so comfortable. The
kind of thing that's regarded as proper blend, the smooth homogeneity we are

so used to, will disappear out the window. In its place will be something un-
compromisingly active and grainy, a sense of real people singing about some-
thing that is real to them. How does this square with the beatific view we all
have of Renaissance church music? Not very easily, at first anyway. And you
can be sure that the critics, even those who are supposedly "informed," will be
alarmed at the sound of something actually new.

After articulation, you mentioned tempo as another thing that's not notated
but that is important.
Obviously, the key —perhaps the most important key—to a piece of music
is finding the right tempo. applies to
It music, and
all applies no actually, it less

to the fifteenth or sixteenth century. My — a very general observa-

feeling it's

tion — is that there's a tendency to adopt tempos that are too fast in this music.
I used to take Renaissance sacred music rather slowly, partly because of the
singers I was working with. Now I'm enjoying a somewhat different approach.
But I never like doing a piece exactly the same way twice —sometimes I'll adopt
a different approach just to avoid solidifying my approach and to see if some-
thing fresh and interesting will emerge. As a general observation, I would say
that when choirs take this music "too fast," even just a little too fast, they tend
to skate across the surface of it, rather than go right through the middle. Par-

is. An example of how Byrd's Latin may have been pronounced is given on pp. 60-61 of
Singing Early Music, ed. Timothy J. McGee (Indiana University Press, 1996). The text given
is also recited on the book's accompanying CD; it does indeed sound Hke Latin with a heavy
regional English accent.
 THERE IS NO SUCH THING AS A NORM 113

ticularly in America, I've noticed a tendency towards that in college choirs,

making the music sound a little bit flippant. It's very beautiful in an airy-fairy
way, but it doesn't seem to have any guts. That's what I miss too often. Of
course, the other extreme is to become too self-indulgent, and the music dies a

different kind of death. But I think that in Renaissance music one can experi-
ment by taking it too slowly to start with, so that you're really brought face
to face with what's going on inside the music. And then you can somehow re-

tain that intensity but take it at a more flowing tempo.

Your mention of American choirs and guts, and before that of "blend" an —
—
Anglican choral virtue versus graininess, brings up another often-discussed
issue, that of the Oxbridge/Anglican dominance of Renaissance and medieval

performance in recent decades, and how that has affected our perception of this
^^
music.
It certainly has had a clear impact on the accepted sound of this music
today, because these are the people who have mostly been recording it. And
from a purely musical point of view it is perhaps a shame that there is so much
duplication of personnel from one well-recorded group to another. But it's in-

evitable. You have a resource of very skilled singers who can virtually perform
this music at sight. There is a danger there, of course, that the results will be

a little superficial unless they do actually spend time on a given piece; but the
disadvantages are outweighed by the advantages in purely practical terms. It's

true, there are groups in France and Belgium that are also performing this
music, like the Ensemble Clement Janequin, the Huelgas Ensemble, and the
choirs conducted by Philippe Herreweghe, and although they use a few English
singers one can clearly hear a different kind of timbre, one that's more grainy
or personalized than the more ethereal Oxbridge sound. I'm very interested in

that: in fact, I prefer it! Perhaps it's just that I'm so used to Oxbridge that it's

nice to go and hear something different, something fresh. I'm interested in hear-

ing the music sung by different kinds of choirs. It's the same with plainchant:
you hear different countries bring a totally different sound to music which is

on paper exactly the same. It's like hearing the music of Victoria sung by a
Spanish choir; although there the point is a little bit disproved, because West-
minster Cathedral has recorded some Victoria very beautifully, very powerfully,
and doesn't sound at all like Oxbridge.

They have a more Continental sound . . .

Well, what we English call a Continental sound, though it's hard to find it

on the Continent. It's something different again. But, yes, relative to the Eng-
lish tradition, it /5 more Continental.

16. See Christopher Page's interview, above, and Howard Mayer Brown's "Pedantry or
Liberation?" in Authenticity and Early Music, ed. N. Kenyon (Oxford University Press,
1988), p. 47.
114 THE RENAISSANCE, OXBRIDGE, AND ITALY

Your Theatre of Voices is American.

Mostly American, though even there I have a couple of English people! But
the overall sound is definitely different; and anyway, the sunny California cli-

mate gives one a different feel for things than all those gray English skies!

SELECTED DISCOGRAPHY
Paul Hillier's talents are represented on disc in music from the eleventh century
through our own day (the latter notably in the music of Arvo Part, on a num-
ber of ECM releases). The earliest repertoire is featured in two of Hillier's best

recordings — and in one of his most controversial ones. The two acclaimed ones
preserve gripping performances of eleventh- to thirteenth-century troubadour
songs (Hyperion CDA66094, and Proensa, ECM 21368 —on the latter disc he
deploys a less smooth singing style, perhaps in line with the ideas he discusses
toward the end of his interview; of the two discs, I prefer it). But when he and
the Hilliard Ensemble recorded music of the twelfth-century Parisian master
Perotin (ECM 21385), Mark Everist found it "a misguided attempt to perform
this music as if it has no surface interest," due to Hillier's "attempts to inter-
pret Perotin's music in terms of a post-60s minimalist aesthetic." ^^ That he and
Hillier have such different views on the music may —as we've seen in earlier
chapters — be symptomatic of (or is at least allowed by) the incompleteness of
the evidence.
One's response to the Hilliard Ensemble depends in part on one's response
to its individual voices; but describing the group collectively, Richard Taruskin
says its strengths in Renaissance music include "a wonderful sense of line," a
"phenomenal" level of intonation, and a knowledge of "what relative harmonic
tension is, and what it can contribute to keeping music of slow tempo afloat."'**

The ensemble made numerous recordings during Hillier's tenure with them.
Naturally, not all the recordings are successful, but many are first-rate, includ-
ing CDs of Dufay (to Gareth Curtis, "by far the most satisfactory recording of
any of Dufay's Masses at present available"''' —EMI 47628); Josquin's Missa
Hercules Dux Ferrarie (according to Todd McComb, it is "uncommonly rich

in sonority and color" for the Hilliard, and "may be their best recording,

among many excellent ones"; EMI 49960);^° music from the Old Hall Manu-
script (to Mary Berry, the CD is "a major service to early music"-' EMI —
54111); and what David Fallows calls the "blandly titled but stunningly per-
formed"^^ Sacred and Secular Music from Six Centuries (Hyperion 66370).

17. Everist, Early Music 18 (August 1990), p. 486.

18. Taruskin, Opus (October 1985), p. 42.
Music 16 (February 1988), p. 127.
19. Curtis, Early
20. From whose URL is http://www.medieval.org.
his online database,

21. Berry, Gramophone 69 (January 1992), p. 90.

22. Fallows, Gramophone 69 (January 1992), p. 39.
 THERE IS NO SUCH THING AS A NORM 115

Hillier has made two recordings of Byrd's great Mass for Four Voices. The
earher one, with the HilHard Ensemble (EMI 63441), has detractors (Eric Van
Tassel calls it "low-temperature to a fault, with excessively slow tempos and
lacklustre phrasing," though he liked the Mass for Five Voices, on the same
CD),-' but it also has strong advocates (Jerome Weber not only likes the tem-
pos but thinks the performance has a subtlety and depth that no other record-
ing has matched).-'* Hiilier's later recording of the Mass for Four Voices (ECM
21512) was among his first with the Theatre of Voices — which, in this record-

ing, consists of a solo quartet. Tess Knighton writes that its American soprano
and countertenor have noticeable vibratos that obscure the counterpoint, and
^^
voices that don't blend with those of the Englishmen who sing the lower lines.

By contrast, the Theatre of Voices' more recent larger-ensemble recordings,

on Harmonia Mundi, have garnered quite a bit of praise. Regarding their
Josquin Missa de Beata Virgine (interspersed with beautiful Marian motets by
Jean Mouton; HMD 907136), Fallows calls the singers "effortlessly clear, won-
derfully in tune, beautifully balanced," and comments on their control of "a

range of vocal timbre, from the sweetest to something really quite direct." Be-
yond that, he adds, "there is an energy in the performances that keeps every-
thing marvellously alive."-" (By the way, I bring up the idea of interspersing
Renaissance masses with motets in the next chapter to Peter Phillips, who
prefers a different approach.) Patrick Russill praised the Theatre's Tallis col-
lection (HMU 907154) for the singers' "dark-browed gravitas and warmth of
feeling" and their realization of "the sombre harmonic undertow so character-
istic of Tallis."-^
Clearly, Hiilier's move toward new sounds proceeded step by step. His (I

believe) first American recording, from 1991, was of the Cornago Missa de la

mapa mundi (HMU 907083) with an ad hoc group of American singers who, —
according to Fallows, sound like Englishmen. And not just any Englishmen:
"Anyone on a blind tasting," he says, "is likely to conclude that this is indeed
The Hilliard."^"

FOR FURTHER READING

Howard Mayer Brown's Music in the Renaissance (Englewood Cliffs, N. J.:

Prentice Hall, 1976) is an excellent introductory text on Renaissance music. For

a discussion of what is known about Renaissance performance practice, the in-

dispensable starting point as I write this is Performance Practice: Music before

23. Van Tassel, Early Music 13 (August 1985), p. 463.

24. Weber, personal communication, 1996.
25. Knighton, Gramophone 72 (September 1994), pp. 86-87.
26. Fallows, Gramophone 73 (November 1995), p. 137.
27. Russill, Gramophone 73 (July 1996), p. 91.
28. Fallows, "Quarterly Retrospect," Gramophone 71 (December 1993), p. 34.
116 THE RENAISSANCE, OXBRIDGE, AND ITALY

1600, ed. H. M. Brown and S. Sadie (London: Macmillan, 1989 and New
York: Norton, 1990); the chapter on sacred polyphony by Christopher
Reynolds is especially relevant to this interview. Alexander Blachly and Ale-
jandro Planchart provide more practically oriented guides for Renaissance vocal
ensembles in chaps. 2 and 3 of A Performer's Guide to Renaissance Music, ed.
Jeffrey Kite-Powell (New York: Schirmer, 1994); they, too, are first rate.

Regarding historical pronunciation, the essays by Hillier and by Alison

Wray in Companion to Medieval and Renaissance Music, ed. T. Knighton and
D. Fallows (London: Orion, and New York: Schirmer, 1992), are stimulating
introductions. The best complete book on the subject is Singing Early Music,
ed. Timothy J. McGee (Bloomington: Indiana University Press, 1996), which
also includes a helpful CD with texts recited using historical pronunciations.
For a discussion of William Byrd's sacred music, the most important book
is undoubtedly Joseph Kerman's The Masses and Motets of William Byrd
(Berkeley: University of California, 1981); and there are four fascinating essays
on Byrd and Tallis in Kerman's Write All These Down (Berkeley: University of
California, 1994). An excellent book about another composer Hillier discusses

is David Fallows's Dufay (London: Dent, 1982).

 6
Other Kinds of Beauty
— oOo^ 6^

Peter Phillips on the Tallis Scholars

and Palestrina

One could locate an endless number of beginning points for the early-music re-

vival, from late-eighteenth-century Handel-reverence' to early-twentieth-cen-

tury clavichord making. Wherever the revival first sprouted, though, one could
argue that was foreshadowed by the sixteenth-century composer Giovanni
it

Pierluigi da Palestrina. Palestrina was the first composer to remain influential

not just for a few decades after his death, but for at least 300 years.
His influence lay not in his works, though they went on being performed
sometimes, nor even in his style, which was drawn upon mainly in some reli-
gious music or in the odd work of, say, late Bach or Beethoven.' Instead, what
kept Palestrina influential was something that theorists derived (more or less)

from his music —a set of rules of counterpoint. These became central to musi-
cal education and remain so today. In our interview, however, Peter Phillips ob-

serves that these rules of counterpoint have distorted our concept of Palestrina:
they represent only one aspect of his style.
Palestrina foreshadowed today's early-music revival in another way: in the

nineteenth century, both Catholic and Protestant musicians restored his music
to more frequent performance. That they sang it at what we would now regard
as half speed was due partly to their misreading of his notation and partly to

1. Ellen T The Composer's Posthumous Reputation in the Eigh-

Harris, "Handel's Ghost:
teenth Century," in CompanionContemporary Musical Thought, vol. I, ed. J. Paynter et
to
al. (London: Routledge, 1993), pp. 208-25.

2. See Christoph Wolff's "Bach and the Tradition of the Palestrina Style," in his Bach: Es-

says on His Life and Work (Cambridge, Mass.: Harvard University Press, 1991), pp. 84-104.

117
118 THE RENAISSANCE, OXBRIDGE, AND ITALY

their longing for the seraphic purity of the lost age of "true church music. "^
We sing the music closer to an appropriate tempo now; but Peter Phillips be-
lieves that we may be even further from its aesthetic. Perhaps this legendary
composer needs a re-evaluation.
Phillips thinks so and has done his part to contribute to one — above all

through his group, the Tallis Scholars, who, he says, perform Palestrina more
than any other composer in their vast repertory. Moreover, he says that the
Tallis Scholars' style "is, to a considerable extent, formed on Palestrina."" If

you wonder how a group of twentieth-century British men and women can
claim to sound like a group of sixteenth-century Italian males, read on. As
you'll see, Phillips has never aimed to re-create original performing styles, and
was among the first in the early-music world to publicly question the goal of
doing so.

Phillips also spoke about the Tallis Scholars' popularity. Musicians often
consider such questions crass, but few issues could be more urgent today. One
possible factor in their success is something they share with Anonymous 4, an-
other group with rare market appeal: a celestial sound. We discussed the Tallis
Scholars' sound as well.
I interviewed Phillips a few hours before he led a sold-out Palestrina con-
cert. The program was similar to one the Tallis Scholars had given a few
weeks earlier in Rome, in the church where Palestrina was trained. It's in-

teresting that when the Accademia Nazionale di Santa Cecilia scheduled this
commemoration of the 400th anniversary of the composer's death, they in-
vited an English group to sing. Even Italians, it seems, loved the Oxbridge
achievement in this music and were, like Phillips, not convinced that natives
had come up to snuff in it yet. We'll see something quite different in the next
chapter, on Monteverdi.

On Performing Renaissance Music

You once wrote that the issue of authenticity didn't really apply to the choral
music of the Renaissance, partly because there's no way of ever knowing what
it sounded like in the first place.^

Well, I haven't changed my view. I'm uncomfortable, though, with the im-
plication that because we can't know exactly how they sounded we are there-
fore absolved of all responsibility to try to find out.

3. Carl Dahlhaus, Nineteenth -Century Music, trans. J. B. Robinson (Berkeley: University

of California Press, 1989), pp. 181-82.
4. Hilary Finch, "The Road to Rome from Oxford," Gramophone 72 (September 1994),
p. 17.
5. Peter Phillips, "Performance Practice in 16th-century English Choral Music," Early
Music 6 (April 1978), p. 195.
 OTHER KINDS OF BEAUTY 119

But you also made the point that you think the evidence suggests that we
—
may sing it better than they did their archives show some very old singers,
who probably would have become woolly in voice production.''
—
That does follow we may be doing it better in our own terms. Again, I

don't think that's necessarily a reason for not trying to find out as much as we
can about how they would have sounded. But it stands to reason that a choir
some of whose members are quite old and have been in it almost the whole of
their lives probably didn't sound fresh, agile, and youthful, as we like the music
to sound now. Moreover, the older singers probably were not careful about
their tuning and in control of their voices, as we think essential.
Of course, there's always the converse question of what Palestrina would
think if he heard us. But what I'm really saying is that we have to carry on in

making the sound good in our own terms. I don't think there's much future in

making it disagreeable to our ears to satisfy some theory, even on the rare oc-
casion when we can substantiate the theory, because then modern audiences
won't go for it and we as performers will cut the ground from underneath our

feet. I do think that occasionally groups make an error of judgment, in a way

that I would hope to avoid, of not trying to appeal to modern people. I won't
go into huge detail about this, because it sounds as though I'm swiping at my
colleagues; but just to give one example, I am struck by a recent recording of
the Allegri Miserere that in all five verses leaves out the top C that everyone
wants to hear. Predictably, it wasn't very successful in commercial terms. I say
this only to illustrate that we try hard to interest people in what interests us. I

hope we communicate with people; and I think that the numbers of records
sold and of people who come to our concerts all over the world mean that
we're getting this intellectually difficult, taxing music across to many who
would never have believed that they would like it.

It's not only difficult; it's also written for liturgical purposes quite different
from the concert experience.
At least half the time we're performing in churches; but even then it's not
a service, it's a concert. We're singing to people who've come and paid to hear
it. To bridge the gap between our performing contexts and the original ones, I

think we —deliberately— and the audience —also deliberately —turn this music
into something they're more familiar with. So a Mass setting, which lasts half

an hour and has five movements, begins to turn into something like a sym-
phony. I'm normally quite careful to program one big work, which will prob-

6. This problem clearly existed and caused consternation in the papal choir in Palestrina's

time; see Richard Sherr's "Competence and Incompetence in the Papal Choir in the Age of
Palestrina," Early Music 11 (November 1994), pp. 607-29. Sherr concludes, "we may not re-
ally want to hear the music the Sistine choir sang in the Age of Palestrina in the way that they

sang it. (So much for 'authenticity.')"

120 THE RENAISSANCE, OXBRIDGE, AND ITALY

ably take up the first part of the concert, so that people can come away with
the feeling that they've had to grapple with something substantial. A Requiem
Mass is ideal —even longer than a Mass, even more of a grapple in many cases,
and potentially highly emotive. I don't like to program too many short items.
The same principle applies to records. So new guidelines come into play in
modern circumstances.

An issue that has been raised in this respect is, to quote Richard Taruskin,
"A Palestrina mass done as a five-movement choral symphony defeats the com-
poser's purpose, which is to unify the service of an hour's duration or more by
periodic inspiring returns to familiar and symbolic sounds. " The sections were
made to sound similar, with a single cantus firmus or head motif, so people
would recognize each return as part of a unity. "In hearing them in the artifi-

cial context of a recording, it's like hearing a gigantic rondo with the episodes
removed. Why not record cyclic masses with motets interspersed?"^ How
would you answer that?
I would answer by saying that I think we've changed the terms of reference
under which these pieces are being sung. We're singing either in concert or on
disc, but not in a service. Now, Richard Taruskin suggests a sort of halfway
house there, which would be very interesting.

His concern is partly musical —he thinks it's too repetitive to hear the same
ideas again and again for five movements in a row, when they're meant to be
interrupted with other ideas.
Well, I think he has a point. But I like listening to the Masses straight
through. There's a musical argument in so many of them that doesn't neces-
sarily benefit from having long interruptions between movements for liturgy. It

would perhaps be even worse, in a sense, to interrupt it with other pieces of

music that distract you from the musical argument that is unfolding from
movement to movement. You can argue this either way; I think he has a point,
but it doesn't worry me, nor does it seem to worry the people who buy our
discs. I invite him to look at the sales figures.
Regarding that last point, I'm sorry to argue from the point of view of "it's

a great success and that's all there is to it." It may cease to be a great success
in ten years' time, in which case the argument will be turned around on me. In
any case, even if it ceased to be a great success I would not change the way I
do it, because I actually believe it presents the music to best advantage.

One question regarding performance of masses has been that of dramatiz-

ing certain sections by such things as tempo variations, slowing down for the

7. Richard Taruskin, "A Glimpse of the New British Choral Sound at Its Best," Opus (Oc-
tober 1986), pp. 1\-1S —an interesting (and generally very enthusiastic) discussion of the
Tallis Scholars' early recordings.
 OTHER KINDS OF BEAUTY 121

Passus et sepultus est and speeding up for the Et resurrexit. This is something
that people often disapprove of. What are your views on that?
My views on that have in fact changed —among the few that have. (I don't
mean to say that my views on this music haven't matured, but we have been
quite consistent in what we've done as far as sound goes, as you can hear from
the old recordings.) This business about changing speeds is difficult, because
we're so used to pulling out the phrases for the Miserere and speeding up for
the Quoniam; we do it because the words suggest to a modern mind that some-
thing should be slower or faster. But I'm no longer prepared to do such things,
and I think the change came with more understanding of what polyphony is
really about. If the polyphonic lines are going without a break through one of
what we used to call "slow" sections into one of the "fast" sections, I'm not
prepared now to speed up in the middle of a phrase. That's what used to hap-
pen, but I now think it spoils, and must spoil, the nature of the counterpoint.
After a double bar, when it is possible to start again, may take on a new I

tempo, provided that at the end of the new passage the tempo fits satisfacto-
rily with what follows afterwards. But wrapped up in a very
this question is

complicated issue about whether you maintain strict tempo relationships be-
tween all the time signatures from the start to the finish of a whole Mass, not
just from the start to the finish of a movement.

So you wouldn't go to a strict proportional tempo ideal, with all the tem-
pos based on an unchanging tactus.''*

I am prepared to do that within a movement. By and large, we try to keep

to the proportions implied within a movement, so that the triple times relate
to the duple times and the duple times relate to each other. What I'm not pre-
pared to do is then start the next movement at exactly the same speed, espe-
cially not in the Credo and Sanctus.
You could argue that this has come about through concert performance. If

you're doing five movements on the trot without any interruption, you're more
you would be
inclined to vary the speeds, to give variety to the audience, than
if movements were broken up. In fact, under the "gigantic rondo" liturgical
the
situation that you spoke about, it would actually help the audience to under-
stand what's going on in the music if you maintained the same speeds, because
then they would hear all the connections undisturbed. Whereas in a concert per-
formance, running things against each other, you can get new perspectives on
the head motif and so on by taking them at slightly different speeds.
We try to throw off what might be called a nineteenth-century view of in-

8. For a discussion of this, see Alejandro Enrique Planchart's "Tempo and Proportions,"

in Performance Practice: Music before 1600, ed. H. M. Brown and S. Sadie (London: Macmil-
lan, 1989, and New York: Norton, 1990), esp. p. 134, where he writes, "It appears that the

tactus was meant to stay constant throughout a composition, though there is a small amount
of evidence that it could vary slightly."
122 THE RENAISSANCE, OXBRIDGE, AND ITALY

terpreting certain passages when it clashes with what I think the polyphony re-
quires. And it's very difficult to do so, however hard you try, especially in pieces
that you're used to doing in the old way and, indeed, using editions that insist
that you do it in the old way. In Palestrina's Missa Papae Marcelli I've discov-

ered that there should be no double bar before the Et incarnatus est —you
should go straight on without any break at all. I can't do it; I've done it for so
many years from bowdlerized old editions that have always had double bars
there. And even if I were able to make myself do it, the singers would have to
as well; I would just be asking for instant trouble. You ask yourself what's
gained.
If we had come to it fresh, we would have gone straight on, but I also bet
that however much we tried to keep the speed up for the Et incarnatus est, we
would actually have gently slowed down even if we'd never sung that piece be-
fore, since this is what people expect at those words. This is the astonishingly
difficult process of undoing received opinion in all our training. It's difficult to

do this and make the music live.

That of course what historical performers of Beethoven or Bach

is exactly
must do every time they play; as you say, in some repertoire. Renaissance per-
formers can run into it too. But to return to the tempo issue: how do you
choose your speeds? Has this anything to do at all with such elements as dis-
sonance, and suspensions'^ being prepared and resolved?
We adopt speeds that are convenient to ourselves in the singing, so that we
can make the phrases breathe, make them expressive to people, and let people
hear how the music is constructed.
Speed is largely in the hands of the singers. You have to find a compro-
mise — you can't go so slow that the singers can't breathe the phrases so that
they fall naturally, but you can't go so fast that the suspensions sound rushed
and lack their full weight. If you want to take a passage fast, the singers have
to know the music well, so that they stand a chance of being able to feel it,

phrase it, and make it sound convincing. Otherwise you need to slow down, it

and then they'll be in trouble because they can't breathe. And then you have
to take into consideration the building that you're singing in. If it's very dry
you don't want to take things too slowly because then the voices get dried out
and they can't sing the phrases. Performance is a living art, and you have to
take all these little factors into consideration.

There has been a lot of discussion lately of the Oxbridge sound, which has
dominated modern performance of Renaissance music, versus the so-called

9. A suspension is a note that begins as a consonance but is held over (suspended) when
the other voices change, creating a new harmony
which the suspended note is dissonant.
in
At this point the suspension must be resolved onto a consonant note, which it almost always
does by moving up or down a step. The suspension is a basic source of expressiveness in Re-
naissance polyphony, and indeed in most later music as well.
 OTHER KINDS OF BEAUTY 123

"Continental" sound. ^^^

You've described the "traditional" English sound: "var-
iously described as hooting or floating, depending on the commentator's point
"^^
of view, with indistinct words and good, if rather hazy blend and balance.
And you've noted that from what you do using women in-
this is different —
stead of boys, for example. Richard Dyer called your sound "Phillips's great
imaginative creation. It is a sound that requires [acoustic] coddling because it
is so bright and piercing in Tudor tessitura."*^
That's amusing, because the criticism that's made of the TaUis Scholars in
England is that we're narcissistic in the sound we make: that we're just wal-
lowing in beautiful noises. And I think that ties up with what you just quoted
about the English choral tradition. The King's sound in its heyday was a very
beautiful basic sound. Itwas much helped by the acoustics of King's Chapel,
but in addition, certain things were done to cut off the sharp edges of the sound
and round it off into a mellifluous ensemble. The Continental sound is much
edgier, less purely beautiful,more dramatic and arresting.
I think it's true that we try to make beautiful sounds. Having said that, I

can't see the problem with it. What is the matter with making a beautiful
sound? It's extremely difficult to do. It's like complaining that the Berlin Phil-

harmonic sounds too beautiful.

Which people do.

Yes, well, they're nuts. It's significant probably that I have a vast collec-

tion of Berlin Philharmonic recordings, because I think that they are the best
orchestra. They have exactly the same aim that we do, a blended overall
sound, in which every different timbre has its place, but not so that it

dominates or distorts. In purely musical terms, you can get wonderful effects
with this. The crescendo, for example, is a remarkable thing, a thrilling

opening-out of the sound in completely seamless stages — that's something,

for a start.

I think this blended sound suits the nature of the polyphonic writing, which
is always my concern. The lines in polyphony need to be equal, but distin-

guishable from each other. You could make a nasty noise on each line and
they would then be equal, but I think to be able to distinguish them all in a
seven- or eight-part texture there has to be a background out of which the
lines emerge, rather than a nasty up-front jangle, which I think is perhaps the
danger with Continental-sounding choirs. The lines are highly distinguished

in timbre from one another, but the ear can't always get to them because of

10. A thorough discussion of this can be found in Christopher Page's essay, "The English
a cappella Renaissance," Early Music 21 (August 1993), pp. 453-71, and, of course, it is dis-

cussed in his chapter above.

11. Peter Phillips, "The Golden Age Regained," Part II, Early Musk 8 (April 1980), pp.
180-98; quote, p. 180.
12. Dyer, "The Boston Early Music Festival and Exhibition," Historical Performance 4
(Fall 1991), p. 126.
124 THE RENAISSANCE, OXBRIDGE, AND ITALY

what's up front. And the tuning may be bad, which is always a disaster for
picking out individual lines in the polyphony.

The beautiful sound is also consistefit with your concern for reaching large
audiences.
Yes, many people have told us that what recommended polyphony to them
has been the basic sound: before they've actually listened to any piece of music
very carefully, they've been seduced by the sound. anybody I don't object if

comes to polyphony that way, but I would go onmore to it

to say that there's
than that. If they want to make an effort, if they can be bothered to buy a sec-
ond or third record, they won't get bored by it, as they may get bored by mod-
ern religious minimalism. In twenty or thirty years we'll see where we stand
vis-a-vis Gorecki, Tavener, and Part. I suspect that Palestrina et al. will be lis-

tened to just as much, if not more, while the minimalists will lose appeal.

In her review of your Isaac record, Tess Knighton said that your sound
had "developed over the years into something much more direct and full-
blooded."^^ Have you noticed such a change?
Yes, certainly since the amateur days. Nowadays we use fewer singers who
make more noise. Because they have trained voices, our ten singers today sing
more powerfully, more excitingly, and with more varied dynamic control than
the twenty amateur singers we used to have. I've always encouraged people to
sing; however it blends and tunes, new singers must sing out, and then I can
tell them what to do. There's no good holding back or singing half-voice. In a
concert, if you stand next to these singers it's deafeningly loud. It's just that it's

a certain sort of voice, one that is not operatic.

On another issue, using women instead of boys, you said, "one of the most
blatant contraventions of the ground rules of historical accuracy is perpetrated
by chamber choirs who claim, while using female sopranos and altos, to come
nearer to what the composer had in mind than our cathedral choirs that con-
sist entirely of male singers."^'*

Exactly, this is one of the hypocrisies of the whole thing. Boys in the six-

teenth century sang differently from the way women sing now — I'm sure of that.

But they also may well have sung differently from boys now; boys' voices crack
much sooner today, so the old treble sound may be biologically unavailable to
us. It would have been very interesting to hear the treble voices of the sixteenth
century, but we've got to get on with it now. And using female singers is much
better than using boys in terms of, among other things, concert touring.
I feel that our job in the Tallis Scholars —what it comes down to — is to in-
troduce the public to as much of this unknown repertoire as we can perform

13. Knighton, Gramophone 69 (October 1991), p. 165.

14. Phillips, "The Golden Age Regained," Part II, p. 180.
 OTHER KINDS OF BEAUTY 125

well. That is a pioneering stage, and I'm aware of it. The next stages should
be more varied. For example, if enough people get interested in Renaissance
polyphony, then the market will support many more groups doing what we're
doing. When that happens, those groups will be able to specialize even more
than we can. We have a wonderful role at the moment, which is to roam
around doing whatever attracts our attention. But the next stage is obviously
going to be more specialized than that.
I think many other things should be tried out that never are. We can't do
them because we can't do everything; we already have a lot to do. One exam-
ple, briefly, is that I wish a professional Italian choir would sing Palestrina ac-
cording to reasonable rules of Renaissance performance practice. That would
be fascinating, to hear the Italian spirit behind fully committed performances
of Palestrina. Maybe one day this will happen, but it certainly hasn't yet.'^

On Palestrina

Please comment on Palestrina, since we're observing the 400th anniversary of

his death.

Palestrina is unlike every other Renaissance composer: it's not that he was
unfairly neglected but that he was unfairly pushed forward. Being pushed for-

ward, in his case, meant that certain aspects of his style were studied very care-
fully and others were ignored. What was studied especially was his writing in

the quasi-Franco-Flemish style, because that was a style that could be imitated.
There's a mathematical basis to good counterpoint that students can study and
regurgitate: examiners would presuppose a missing line, or give students one
line and leave them to compose two or three more, largely according to math-

ematical principles. It would have been very much more difficult to teach a stu-
dent to write a pastiche of homophonic music.
But this obscured the fact that Palestrina changed his style at least once and
possibly, I think you could argue, twice, in the direction of homophony. He did
vary, like Tallis. Basically, both were fine craftsmen who did what they were
told. And what they had been told, whichever side of the argument you were

on, was that music had been too elaborate, and that it should come down to
earth, with the words to the fore. Palestrina managed to do that: famously,
that's the Missa Papae Marcelli story.'" But what is ignored is the music that

came out of it.

The Missa Papae Marcelli is a sort of halfway point. Some of the move-

15. Since this interview took place, some Italian recordings of Palestrina, on the Bongio-
vanni label, have received enthusiastic praise (see Jerome F. Weber, Fanfare 20 [September/Oc-
tober 1996], p. 283). But the ensemble, led by Sergio Vartolo, doesn't actually address
Phillips's "wish": it sings one-voice-per-part with discreet organ accompaniment (which may
well, however, be historically correct).
16. An account of the Counter-Reformation legend that with this Mass Palestrina "saved
polyphonic music" for the Church — alas, not true —can be found in theNorton Critical Score
of the Missa Papae Marcelli, ed. Lewis Lockwood (New York: Norton, 1975).
126 THE RENAISSANCE, OXBRIDGE, AND ITALY

ments are in the old Franco-Flemish style, but some are looking towards the
Baroque and are much more homophonic. After that come all the double-choir
motets and other Mass movements which are so chordal, so homophonic, that
it's clear he's thinking entirely from a harmonic starting point —which shows
that he's looking forward to the next musical period, the Baroque.
Now, that's almost never said about Palestrina. People go on about how
progressive Gesualdo was, but Gesualdo was a freak. No one else was like
Gesualdo: of course he looked forward. And some of Lassus's music looks for-
ward. But Palestrina was not so conservative as is made out. That's one thing
that I think needs serious correction.
From this perspective, you can make up interesting programs with early,
mid-period, and late Palestrina. The Missa Papae Marcelli is the quintessential
mid-period work, because you can see the way the style is changing. You've got
the two beautifully elaborate Kyries, and especially the second Agnus, which is

canonic —an old-fashioned technique. It's not a difficult canon by some stan-
dards, but it's very effectively worked out, and it's a substantial movement. And
then you've got some other sections where he's obviously come right down on
the style and cleaned it out. Then you can go on from there to the late-period
works, especially the double-choir motets. A particular instance is the double-
choir Magnificat primi toni. There's not a note of counterpoint in it — well,
that's an exaggeration, but basically it's so. And it's such a clever piece. I have
looked at it on the page and thought. We can't do this, this is boring compared
with all those wonderfully elaborate florid lines that overlap each other so
beautifully in his earlier works. And yet when you come to sing it you realize
that every element of the composition is under perfect control. The progression
of the chords is so finely calculated that it thrills audiences; I'm so impressed
because you really can't see why. It's almost inexplicable when you just look at
it on the page.'^
The other thing I'd like to say about Palestrina, which is more offbeat, I sup-
pose, is that the word "Renaissance" is a problematic one to describe music.
After all, there was no rebirth of interest in Greek and Roman music to inform
anything that happened in Renaissance polyphony. It's a convenient term to join
music with the other arts. All I can say about this is that if ever there was a tie
to be made between Renaissance choral music and classical thought, Palestrina
would exemplify it, because he always does remind me of magnificent classical
edifices. It's with architecture that the comparisons can be made. Palestrina has
Roman grandeur in his style. I don't think this is just fanciful. His music is beau-
tifully sonorous, like a large classical arch, if you like, or a large classical struc-

17. In writing to a ducal patron who had asked the composer to critique some of his own
compositions, Palestrina emphasized, in James Haar's words, "that the sound of the music is

much more important than scholastic rigidity of technique." From "Value Judgments in Music
of the Renaissance," in Companion to Medieval and Renaissance Music, ed. Tess Knighton
and David Fallows (London: Orion, and New York: Schirmer, 1992), p. 20.
 OTHER KINDS OF BEAUTY 127

ture. It's cool, as these facades were, really big and massive, and that's caused
by his clever use of sonorities perfectly contrived to give a solid basic sound. My
experience of Palestrina is that if you take him too fast it seems as though you're
making the edifice squat; it's reduced, spoiled, in a way that other composers of
this period are not necessarily spoiled. This has a lot to do with the sonority,

the grandeur in this music. If you take it too fast you spoil it — it's as if a Roman
architect had got the proportions wrong in a building.

To continue with this comparison, people speak of Palestrina using words

like "serenity," "moderation," and "balance." Musically, that means, for one
thing, dissonancesand suspensions being carefully prepared.
He's more careful than most. The telling comparisons here are not really
with madmen like Gesualdo, but with the style that Palestrina grew up with, the
Franco-Flemish style. I wouldn't say his suspensions are more bizarre than
Josquin's; he is a true successor to Josquin in some important respects. And
you can compare him with the Franco-Flemish composers who lived at the
—
same time as he the really great men, Clemens and Gombert. The Palestrina/
Gombert comparison is a beautiful one to make, though it would be more illu-
minating more people knew any of Gombert's music. But there's no doubt
if

now that Gombert was one of the great polyphonic thinkers of those decades,
for all that his lines work completely differently from Palestrina's. He prepares
his suspensions, but he's much more cavalier with the way they resolve; yet it's

a cavalier attitude that's completely the opposite of Gesualdo's. The music seems
to grow organically out of itself. There's something in Gombert's style which
marks him out as an original thinker, and not just a quirky one. He was a very

great composer, consistently so over his entire output, which can't be said of
some of the more way-out names. I think Lassus'** was as good as Palestrina
when he was writing sonorous music; but he was prepared to experiment (which
one admires him for), and some of the experiments didn't work. Palestrina
seems not to have done that; there's a much higher average in Palestrina, a
craftsman-like technique that maintained a minimum level. Like Bach, he had
an absolutely rock-solid technique that rarely let him down.

To return to the descriptions of serenity and balance, they may be rooted

partly in melodic aspects of the "Palestrina style" —
avoiding leaps beyond a
fifth except in rare instances, and preferring stepwise motion, and immediately
reversing the direction of the line after intervals of more than a third^"^ —almost
plainch ant-like.

18. Phillips discusses Lassus in detail in the essay "Great Men Think Alike," Musical Times
135 (June 1994), pp. 357-63; the article includes comparisons with Palestrina.
19. These characteristics, as discerned by Knud Jeppesen, are discussed clearly by Howard
Mayer Brown in Music in the Renaissance (Englewood Cliffs, N. J.: Prentice-Hall, 1976), p.
286.
128 THE RENAISSANCE, OXBRIDGE, AND ITALY

This is partly because he's a very vocal composer. Palestrina is grateful to

sing; he's an ideally vocal writer. Lassus is not. Lassus thought of things from
an instrumentalist's standpoint very often, which is why some of his effects
—
don't work or at least why you have to impose yourself on the singers to
make them work, which in the long term, at the hundredth performance, is not
ideal. We give up Lassus pieces sooner than we give up Palestrina or Tallis, both

of whom were singers and thought like vocalists, so that their lines are conve-
nient lengths and tend not to do anything bizarre or pose technical problems
that have no expressive meaning. So to return to your question, yes, it doesn't
help to have to sing a major sixth or major seventh in a melody line. It's diffi-

cult to do; you can practice until you get it right, but in the end —as I say at
the hundredth time —you'd rather have a composer who was sympathetic to
the basic needs of the performers, which Palestrina certainly was.

Another comparison might be with Victoria, and this might connect to the
question of Palestrina's relationship to words. Howard Mayer Brown remarks
that Palestrina would seldom go to great lengths to imitate or paint the mean-
ing of an individual word; he didn't want to disturb the continuous stream of
music in a dramatic way by abrupt changes of pace or texture, for example, in
order to insist on the priority of text over music. Victoria, on the other hand,
"was willing to disturb the even flow of counterpoint; to emphasize a word or
a phrase he would tolerate an 'ungraceful' leap of a major sixth, for example,
which Palestrina would have avoided, or he would allow a strong melodic line
"^°
to proceed without immediately reversing its direction.
Victoria wasn't a contrapuntist like Palestrina, so what you say about his
interrupting the counterpoint doesn't necessarily apply. Maybe; but you always
feel with Victoria that he didn't have the same ability to hold in his head six

lines in genuinely independent counterpoint. His effects are consistently more

Baroque than Palestrina's. He groups voices against each other, and his phrases

are generally shorter, more harmonically conceived. Wonderful music; I actu-

ally think Victoria produced the greatest single work of the whole period, his

six-voice Requiem. And his Responsories for Tenebrae — I'd take that set over
the Gesualdo any day. I think Victoria was a very great musical thinker, who
also is appreciated only in part.
And yet Palestrina doesn't bang on in a six-part texture page after page after
page — Clemens does, as a real Franco-Flemish composer would have done,
where they get going and stay going, point after point unworking itself, never
stopping, never cadencing fully. Palestrina is almost never like that. He cadences
fairly frequently; he's prepared to group. In his motet Tu es Petrus, he's con-
stantly grouping and regrouping, not using imitation as the only means of start-

ing a phrase. There is a seamlessness to Palestrina's writing which distinguishes

20. Ibid., p. 315.

 OTHER KINDS OF BEAUTY 129

him from many late Renaissance composers, but he's Hke a halfway house be-
tween the Franco-Flemish and the late Renaissance/Baroque composers,
amongst whom I would put Victoria, Andrea and Giovanni Gabrieli — especially

Andrea — and the early Monteverdi of the In illo tempore Mass. Palestrina was
a crucial halfway house, a man of utmost importance even in his own lifetime.

Victoria was completely in awe of him.

What did Palestrina transmit to the later generation?

I think a lot of scholars would say he was overtaken by events; this is the

argument that Palestrina was essentially a conservative composer. I would say,

instead, that he was required by his employers and by the Council of Trent to
update or change his style, and this encouraged him into a harmonic way of
thinking.'' This harmonic approach is so gently done, so unobtrusively done,
that most people now don't notice it, and I wonder whether it was noticed then.

But there are books' worth of great works by Palestrina which are essentially
homophonic pieces; and I really think it must have made a difference to peo-
ple at the time. It showed them how to organize their thought.

A difference between him and early Baroque composers seems to be this

issue of imitating and painting the text.

The words do come out more in later Palestrina than in early Palestrina. He
was always reluctant to be obvious about word painting, though there is word
painting in early works as well. But I don't think his musical thought depended
on the words, not in the way that certain composers throughout history needed
words to get their musical thought processes going one thinks of Cluck, Wag- —
ner, and Verdi, for example. There's another type, a pure contrapuntist like

Josquin, Bach, or Gibbons, who could turn anything into a beautiful succes-
sion of interrelated melodies, which is what polyphony is. Palestrina was one
of those.
A and Victoria is that Victoria seems
further difference between Palestrina
to have taken a much more impassioned view of the text; Palestrina was a
cooler composer, a more understated composer, as understated as Clemens and
Gombert in the Franco-Flemish tradition. It makes one wonder if Palestrina
was fully Italian in the way we now view the Italians.^^ He maintained a cer-
tain distance; I think his intention was to give a sort of overall picture of the

21. See Phillips's article "Reconsidering Palestrina," Early Music 11 (November 1994), pp.
574-86, for another discussion of this point.

22. On this point, Phillips has written: "Yet although Palestrina's Italianness is now less

familiar to us than the other, more vaunted kind, what he represented is just as typical, and,
of course, can be found throughout the mainstream of Italian intellectual life. Perhaps the best
visual proofs of it are contained in the paintings of Botticelli: a directness of expression, art-
iessness, even naivety of effect, a certain ineffable sweetness." Phillips, "Fiirst der Musik aller

Zeiten," Musical Times 135 (February 1994), pp. 74-79.

130 THE RENAISSANCE, OXBRIDGE, AND ITALY

words, a dignified one. And if that puts modern people off, then I'm sorry,
there's nothing to be done about it. I think his is a position which is full of in-
terest; it's just rather untrendy. I was told by a BBC interviewer that Palestrina
was "cold and sexless," and that she preferred Victoria because it made her
sob. Now, I'm sure there's a lot of truth in what she says. But why is it that
music must be sexy, tear-inspiring, and warm (a particularly nasty concept) in
order for us to be moved by it? I don't say that all music should be cold, re-

mote, and understated —rather elitist, one might almost say aristocratic. But
some elements of that are in Palestrina and I think have their expressive force. ^^
If you want to be an instant success in the modern world, it's best if you're

experimental, obviously human, and obviously suffering, which Lassus was.

The very fact that Lassus produced some bad pieces, as I think he did, only
adds to his image now. The idea of the perfect Palestrina is not attractive to
people; that's a modern problem. Yet I'm sure Palestrina is great enough that
in another fifty years he'll be a trend again. In this we are saying that we hope

his moment will come; but his moment, compared to any other composer of

his period and later, has come and come and come. It's never stopped coming.

When people say they can't love Palestrina's music, I say, "Just listen to it

without any preconceptions about whether it's warm and sexy; imagine other
kinds of beauty and other forms of enjoyment." Then he comes into his own.
On this tour we are doing a whole program of Palestrina, and people are com-
ing up to me and saying, "This is it; this is the greatest music of the period
that I can imagine." So it is for me.

SELECTED DISCOGRAPHY
In the Palestrina issue of Early Music (November 1994), Graham Dixon con-
cludes that one-per-part may have been the most common configuration used
in contemporary performance of Palestrina; ornamentation of his lines may
have been common; and the organ and other instruments may have played
more of a role than we usually recognize (though not within the a cappella Sis-
tine choir). Yet in the same issue Richard Sherr, who has done much of the re-
search on these issues, notes that the original performances may not have
sounded very good even to contemporaries' ears. It's hard to imagine Palestrina
not enjoying what the Tallis Scholars do with his music, even if (as Philip might
acknowledge readily) it sounds like nothing he ever heard in his own day.

23. About which Phillips writes: "[Palestrina's] contribution to the palette of expression
through music was to broaden it in ways which most composers have found unsympathetic.
Few writers have gone unreservedly for happiness, since it is much easier to make an effect

through its opposite. . . . movement after movement of irrepressibly positive, hopeful music
is an experience to be valued, all the more for being rare" (ibid., pp. 78-79). Phillips also
writes about the power of understatement in Palestrina's setting of penitential or despairing
words.
 OTHER KINDS OF BEAUTY 131

Indeed, for investigating Palestrina, there may be no better starting point

than the recording that was made of the 400th anniversary concert mentioned
in this chapter's introduction (Phihps 454 994; the concert is also available in

video and laser disc). In it the Tallis Scholars sing the Missa Papae Marcelli and
other works; it is essentially the same program that Phillips discusses in our in-
terview. The Tallis Scholars have recorded six other Palestrina Masses, which,
says Michael Oliver, are "most beautifully, but not too beautifully, sung."'"* Tess
Knighton praises their "flexible, expressive, and at times overwhelmingly beau-
tiful" performances of the Masses Sicut lilium and Assumpta es Maria (Philips

454 920).-' Noel O'Regan, however, complains about the same expressive qual-
ities, calling these performances "somewhat overshaped," with more a "con-

cert" than a liturgical feeling —though that is, as we've seen, Phillips's intention.

O'Regan also says that the singers' perfect blend "entails a loss of contrasts be-
tween registers," and says that their underuse of musica ficta in Sicut lilium gives

it an "unjustified antiquated feel."-" But he finds that their attention to detail

"gives the listener a constant insight into Palestrina's compositional process."
In apparent contrast to O'Regan, Patrick Russill finds the Tallis Scholars

"on occasion blandly under-inflected,"'^ and Fabrice Fitch speaks of their

. . .

"cool and rather detached interpretive stance."'^ As this shows, for all their
success, the Tallis Scholars evoke varying responses from early-music aficiona-
dos, ranging from devotion to dismissal. But critical consensus seems to be that
the Tallis Scholars' recordings are of reliably high quality and are sometimes
inspired (Russill's and Fitch's remarks come from their enthusiastic reviews of

the Tallis Scholar's CD of White, 454 930). Three outstanding ones are their

beautiful CDs of Isaac (454 923) —which, by the way, includes chant interpo-
lations—of Cardoso (454 921), and of Shepherd (454 916). Their CD of
Josquin's Masses Pange lingua and La sol fa re mi (454 909) was the Gramo-
phone "Record of the Year" in —
1987 the first time in the history of the
awards, we are told, that all the critics "who voted in a category placed the
—
same record first" and their Rore CD (454 929) won the Gramophone 1994
Early Music Award (according to Iain Fenlon, it is "moving and beautiful . . .

the best record [Phillips has] ever made").-** On the other hand, in the Victo-
ria Requiem (454 912), Jerome Roche says that the Tallis Scholars, though
beautifuland not cold, are "less overtly expressive" than the Westminster
Cathedral Choir (Hyperion 66250), which is "well imbued" with the "plan-
gent, intense" spirit of Victoria'" — more likely to make one weep, perhaps.

24. Oliver, Gramophone 71 (January 1994), p. 86.

25. Knighton, Gramophone 69 (December 1991), p. 49.
26. O'Regan, Early Music 19 (February 1991), pp. 136-38.
27. Russill, Gramophone (June 1995), p. 104.
28. Fitch, Mustcal Times 136 (September 1995), p. 495.
29. Fenlon, Gramophone 72 (November 1994), p. 40.
30. Roche, Early Music 16 (February 1988), p. 137.
132 THE RENAISSANCE, OXBRIDGE, AND ITALY

FOR FURTHER READING

The November 1994 issue of Early Music is a very good introduction to cur-

rent issues in Palestrina performance. Another good starting point is Phillips's

own articles on Palestrina — in the same issue of Early Music and in the Musi-
cal Times for February and June 1994. Howard Mayer Brown's Music in the

Renaissance (Englewood Cliffs, N. J.: Prentice-Hall, 1976) contains a very use-

ful chapter on Palestrina. The idea of three creative periods in Palestrina's out-
put was, as far as I know, first broached in a 1971 monograph by Jerome
Roche (Oxford University Press), Palestrina, which is an excellent introduction
to the composer.
 7
Singing Like a Native
— oOo^ ^^

Alan Curtis, Rinaldo Alessandrini, and

Anthony Rooley on Monteverdi

I read once about a blues musician who claimed he could immediately tell a
recorded blues singer's race. On a blind listening test he scored perfectly, until
he mistook the white Englishman Eric Burdon for a black American.
I wish I had thought of a similar test for Alan Curtis and Rinaldo Alessan-
drini. I wouldn't have bet on the outcome. Both are gifted keyboardists who
have founded vocal groups to explore the madrigals of Monteverdi and other
Italian composers. The gist of their efforts is to have the music sung not by the
early-music specialists, often British, who have dominated the field, but by na-
tive Italian speakers. No one else, the two believe, can equal Italian singers in

this music.

That may suggest nationalism, and in the 1990s many of us find ourselves
recoiling from nationalism, or to be more precise, from the ideal of ethnic pu-

rity. Yet most of us do like local color. When I challenge the ideal of keeping
a national culture "pure," I like to mention that Italy got pasta from the Chi-
nese, tomatoes from the New World, and pizza from Sephardic refugees. But if

I ever get to Italy I will be distressed if I have to eat at a McDonald's or, more
to the point, a Pizza Hut. I fantasize about "pure" cuccina Italiana, not the
American mass-market culture that so much of the world is adopting.
Which, then, of the two faces of nationalism show in the Monteverdi de-
bates: the bad essentialism, or the good —what shall I call it — cultural integrity?
(I mean something like a culture that assimilates valuable input from other cul-
tures, but without losing its own best features.) Certainly, getting the local color
right pays dividends in some music. Chopin played his mazurkas in Polish

133
134 THE RENAISSANCE, OXBRIDGE, AND ITALY

dance rhythms that were so different from the notated I that sometimes they
could be counted in \.''
Played as written, many mazurkas lose something.
But just being Polish wouldn't equip you to re-create Chopin's rhythms;
some Polish pianists (clearly not dancers) have played the mazurkas without
ever straying from ?.The music of Monteverdi, however, involves a cultural
legacy more basic than dance steps. Monteverdi wrote some of the most text-
centered music in Western history. In sections of a few madrigals, he notated
no rhythms at all: clearly, the rhythms are to come from the words. If this de-
mands native fluency, as some argue, that would rule out a lot of us. Gaining
native fluency in a language usually requires mastering it before puberty,^ which
usually means being a native. Perhaps that's why connoisseurs of art song,
which is also text-centered, have argued that such songs respond best to native
speakers.^ And even in the eighteenth century it was sometimes said that only
Italians could give Italian vocal music its "true accents and expressions." Thus
I can't dismiss Curtis's and Alessandrini's preferences for Italian singers in this
music.
And yet Ivan Moody writes that his earlier conviction that "it was almost
impossible for the secular song repertory of Europe to be sung by ensembles
from countries other than those in which they were created" was shaken by
"highly idiomatic performances of Italian madrigals" by some non-Italians." To
my ears British Monteverdi of the 1990s can be far from the musical equiva-
lent of Pizza Hut, and needn't sound "Anglican," "restrained," or — well, any
of the things that Alessandrini accuses it of. It is fair, in an ironic way, that I

give the last word to a British Monteverdi pioneer, Anthony Rooley. Just as the
main audience for the blues in the late 1970s was college-age white kids, the
main impulse for today's Monteverdi revival originated not in the Italian sun-
shine but in the English fog.

"Every Detail of This Music Has

Something to Do with Language"
Alan Curtis

After graduate study at the University of Illinois in the late 1950s, the harpsi-
chordist/conductor Alan Curtis went to Amsterdam to work with Gustav Leon-
hardt, whom he has called his "chief mentor." In 1960 he joined the faculty of
the University of California at Berkeley. There, in 1966, he conducted his first

1. Jean-Jacques Eigeldinger, Chopin: Pianist and Teacher as Seen by His Pupils, trans.
Naomi Shohet with Krysia Osostowicz and Roy Howat (Cambridge University Press, 1986),
pp. 110-12.
2. (New York: William Morrow, 1994), pp. 290-91.
Steven Pinker, The Language Instinct
3. Alan Blyth, Gramophone 68 (February 1991), p. 1546.
4. Ivan Moody, Early Music 20 (November 1992), p. 685.
 SINGING LIKE A NATIVE 135

opera (and Monteverdi's last), L'incoronazione di Poppea. Since then he has

been some ways the work's most important advocate. He has pubHshed the
in

pre-eminent modern edition of it, recorded it twice, and, in an influential study,

shown that some of its music, including the celebrated final duet, was not writ-
ten by Monteverdi.''
Over the years, Curtis's work has centered more and more in Italy; he now
lives in Venice, and conducts opera regularly in Italy and elsewhere in Europe.
He has conducted at, among other places, Spoleto, La Scala, the Rome Opera,
San Carlo, Bologna, the Netherlands Opera, the Lisbon Opera, and the Inns-
bruck Festival.

Curtis has pioneered several elements of Baroque performance practice. He

was (as far as he can tell) the first person since the seventeenth century to have
a chitarrone built, for use in a 1962 Monteverdi recording. He was the first
since the eighteenth century to use lutes in the continue^ group in Handel op-
eras, a practice that is now common. He was also the first to revive Rameau
operas and ballets with period instruments and choreography. More relevant to

this chapter, he has been the leading advocate of what was at first a radical ap-
proach to performing Monteverdi's late operas: without any added orchestra-
tion, but with just the documented band of a few instruments^ playing mainly
just the notes indicated. This challenged what had long been the dominant view,
espoused by (among many others) Nikolaus Harnoncourt, who wrote.

In contrast to the all-too-liberal arrangers, there are those representing the

opposite extreme: super purists who only want to realize the handed-down,
skeletal score and reject any additions. This sort of loyalty to the work does
not serve the intention of the composer, since it negates the presuppositions

on which he has based his work. It is just as incorrect to reveal only the
"skeleton" which was written down by Monteverdi as to cover it with in-

appropriate "flesh" of a much later age —as frequently happens.*

5. The edition was published by Novello, 1989. The recordings were made in 1962 (Vox)

and 1980 (Nuova Era). The article is "La Poppea impasticciata, or Who Wrote the Music to
L'incoronazione}" Journal of the American Musicological Society 42 (1989), pp. 23-54.
6. In the "continuo" —
essential to the Baroque style from mature Monteverdi through Bach
—
and Handel a keyboard or plucked instrument both played the bass line and filled out the
harmonies above it throughout. Often a melody instrument (gamba, cello, bassoon, etc., de-
pending on the era and location) doubled the bass line.
7. In his preface to his edition of Poppea, p. xii, Curtis explains that the typical seven-

teenth-century Venetian opera orchestra consisted of three to five string parts, often with one
on and usually with a violone or other 16-foot instrument doubling the bass. Occa-
a part
were added. The continuo used, Curtis says "two
sionally trumpets, cornetts, or recorders
harpsichords and one or two instruments of the lute family, such as an archlute and theorbo
(or chitarrone)." Poppea and Ulisse were scored for such small groups; Poppea, Curtis says,

should not have trumpets added, though he seems to approve of a recorder or two.
8. Nikolaus Harnoncourt, article on Monteverdi's Ulisse in The Musical Dialogue, trans.

Mary O'Neill (Portland, Oregon: Amadeus, 1988).

136 THE RENAISSANCE, OXBRIDGE, AND ITALY

How do you respond to Harnoncourt's critique of "super purists" in Mon-

teverdi?
The notion that early opera must be orchestrated comes from the period
when it was rediscovered. When Vincent D'Indy conducted the Monteverdi op-

eras early in our century, he found it necessary to orchestrate them, because no

one then could conceive of opera without a modern orchestra. When I came
along, around 1 960, there were people who believed that in Venice they didn't
use an orchestra but who would
some excuse or historical fantasy to the
invent
effect that"Of course, they must have improvised, and probably improvised
accompaniments to the singing." I took the stance that we don't have to do
anything beyond simply reconstructing some missing parts in the ritornellos [re-
peated instrumental sections that come between the sung sections].

Why do you think it's wrong to do more?

I stick to what I believe was Monteverdi's notion, which we see lasting on
into Alessandro Scarlatti in the early eighteenth century. When Scarlatti writes
an aria that is delicate, he might specify one violin to a part and then write
those parts only when the voice is resting. He's writing an accompanied aria,

which Monteverdi almost never did —Monteverdi alternates the ritornellos with
the voice — but even though Scarlatti could have written obbligato counterpoint
to the voice, he doesn't, because he doesn't want to cover the voice. He still

wants the voice to be able to have its subtle expression.

Subtleties, as a form of expression, are later replaced by virtuoso display,
which is much less related to the text. With that, of course you can have in-

struments. We hear of Ithe eighteenth-century castrato] Farinelli having a con-

test with a trumpet to see who could do the most coloratura. But that's not
Monteverdi. It's just as wrong musically to put that kind of eighteenth-century
concept back a hundred years as it is to put a Respighi concept back three hun-
dred years. That's what would like to convince Harnoncourt of. Not only is
I

it hard to imagine what would have been improvised, but I'm simply not con-

vinced that it would have been tolerated for a moment for a trumpet to be im-
provising while the singers were trying to sing expressively about fortune or
fate or the gods. Harnoncourt has this idea that whenever the gods speak or
whenever there's something godlike in the text, it's Baroque to have a trumpet
in the background. So when Jove makes a pronouncement, in the background
you hear military trumpet arpeggios. To me, that's distracting. What I object
to in the versions by Rene Jacobs, Harnoncourt, Leppard, and others is that
their composed additions (let us all confess that we are not really talking about
true improvisations!) do not indeed, cannot — —
follow the voice but rather tend
to cover it, distract from it, impede its rhythmic freedom in short, lessen the —
potential dramatic impact of Monteverdi's musical line.
What I think we really have to do is not orchestrate, but find out why this

music was so interesting without an orchestra. What would make it interesting

 SINGING LIKE A NATIVE 137

to a modern audience with only a continuo accompanying it? That is still what
we are struggling to find out. We must continue to seek expression that is so
complex and gripping, immediately gripping, that it will fascinate a wide au-
dience — not just scholars who get all the complexity on the first hearing (if

there are any such people), but also people who've never been to an early opera
or heard any early music.
The solution to this, in my view, is that if we're going to revive early
Baroque Italian music we have to revive the notion that language is the most
important element. The libretto comes first, and the libretto is then set to
music. This is opposed to the notion we've inherited from the nineteenth cen-
tury, when people came to believe that the libretto wasn't very important, and
that music was all that really counted.
Now, if you don't have singers capable of bringing off a Monteverdi opera
with the subtle response to the words that it needs, then you're probably bet-
ter off orchestrating it or doing something like what Harnoncourt does: using
instruments that don't cover the voices completely, or that give color to voices
that don't have color, or that interest the audience in singers who wouldn't be
interesting otherwise. What I prefer, though, is to find singers and teach them
to be expressive in terms of the vocal writing itself —that is, not as they would
be in Wagner or Puccini, but as they are relearning to be specifically for Mon-
teverdi — so expressive that they knock people off their feet.

What's your view of how one learns to sing specifically for Monteverdi?
I try not to be dogmatic, which doesn't mean that there aren't some things
I like and some I don't like. Among the things I don't like are the extremes.
One extreme, of course, is obvious: I wouldn't want to use Birgit Nilsson or
Montserrat Caballe. In my opinion, as in that of the majority of early-music
people, the standard opera singer of today —with the Wagnerian unending line

and lack of articulation, plus the volume (at times bordering on shouting) and
the vibrato rhat seem to be felt necessary for large halls — is not suitable for
music that requires such subtle inflections. But where I differ from the vast ma-
jority of early-music buffs is that I also think the light-voiced, pure, "non-vi-
brato" (often English) "early-music voice" is not on the whole appropriate for
the dramatic music of the Baroque, although it may be ideal for other Baroque
music. For opera, we should try to explore more things in between.
What I try to do is to get potential future Fischer-Dieskaus before they've
become too expensive and, more important, too inflexible to be persuaded to
do something other than the modern style. If you find young people who have
enormous talent, and I'm finding quite a few of them coming up now in Italy,
you can get them to do practically anything. I like to work with young opera
singers: the result can be voices that are good enough, and dramatic enough,
to be on the stage, but that also have control of vibrato and intonation, and
of the stylistic patterns and the flexibility needed for the style.
138 THE RENAISSANCE, OXBRIDGE, AND ITALY

You mentioned that the late-Romantic "unending line" and so on would be

out of place in Monteverdi. Can you give an example of why?
Again I turn to what we all say, the text. Let's take Monteverdi's setting of
the words "Lasciatemi morire" ("Let me die"), which opens the Lament of Ar-
ianna. If you say the words simply and nobly, it can be very touching, but it

may not really grab you. On the other hand, if you say more vehemently, as I

think you should, [emphasizing the consonants] "La-SHA-te-mi mo-RIR-e,"

with enormous intensity, it can be overwhelming. But suppose you say it with
intensity and great beauty of sound, but with an absolutely smooth legato, huge
volume, and wide vibrato— "LAASHAAAATEMI-M MORI-I-LLRE-E"— with
that much unvarying volume, legato, and vibrato, it can't have the same kind
of effect. It may overwhelm you with Caballe's vocal strengths, but not with
the meaning of the words, so the power and expression have to come from
somewhere else.

This relates to another issue — and here too I differ from many of my early-
music colleagues: I strongly favor the idea that the singer's native language is

very important.

What advantage would Italians have in singing Monteverdi —couldn't pro-

nunciation, for example, be developed by a good Italian coach?
Pronunciation, as you say, is not so crucial, and perhaps its importance is

more for Italian audiences. But what comes across to everybody is the partic-
ular relationship between the text and the music, and the emotions that arise
from that relationship. This includes such things as extremes of diction and col-
oring and accent, and freedom of rhythm, and knowing where the rhythm
should be free and where it can be more metrical. Also, knowing which words
need to be strongly accented and which are ambiguously accented (not that
there are many in Italian, but there are some); the color of the vowels; where
you would make a crescendo and decrescendo —every detail of this music has
something to do with language.
Now, all of that can be taught to a foreigner, but it just doesn't come to-
gether in the same way. Listen to certain foreign-born singers in your own na-
tive language; you hear a very clever mimicry, but not the substance. You can't
even say specifically what they're doing wrong. But there's something essential
that's not there.

How about when the singer is fluent, flawlessly fluent, in the foreign lan-
guage? Can't they get the same result as a native?
No, although they can get very close. Now, anything that I say could be
contradicted demonstrably, because you can find Italians brutalizing their own
language. In fact, many Italians don't care about or even know the texts they're
singing. And they don't necessarily declaim the words instinctively; but even
when they do, it would be better if they would pay more attention to these
things. That's what I focus on, as a conductor working with singers.
 SINGING LIKE A NATIVE 139

Of course, your own example could be said to prove that a foreigner can
master the subtleties of Italian.

It would seem that way; but a famous Italian critic once noted that 1 de-
mand from my singers much better diction than I could produce myself.
What is also important for the conductor is to hear subtle relationships (that
are present or are added appropriately) even in the accompaniment, because
everything should revolve around the text. By "text" I mean not just the words,
but also the meanings and above all the emotions that go with those words.
And there, too, I think Italians have an advantage. What does "ahi, lasso"
mean? Well, you can explain to a non-Italian that it not only means languish-
ing and exhaustion but also can have certain sexual connotations. But unless
that understanding has been there from the beginning, it won't be there in the
same way. There are many such instances where nuances persist in the lan-

guage, unstated but implicit. It's not just a matter of reproducing the sound,
the pronunciation and phrasing of a sentence; it's knowing the subtle connota-
tions of those words.

Vve read that what has become modern Italian was, in Monteverdi's day,

justone of many regional dialects, some of them mutually unintelligible (some

of these persist today). Elsewhere in Italy this dialect, Florentine Tuscan, had
great prestige as a literaryand courtly language, but "only the highly educated
could master and they used it almost exclusively in writing or on very for-
[it],

mal, solemn occasions. '"* Do historical and regional dialects such as those no —
doubt spoken by Monteverdi's singers, for whom Italian was probably not the
first language^^- —
have any bearing on your argument that native Italians speak-
ers have an advantage in Monteverdi?
No. For one thing, Monteverdi's singers would have understood "Floren-
tine Tuscan" perfectly —as any educated audience would have —even if edu-
cated ears might have been able to detect a slight accent (and remember that
regional accents are much less noticeable in singing than in speaking). And
I've said that pronunciation is not the main advantage of using native speak-
ers. Still, sometimes pronunciation does matter: when "historical" pronunci-
ation is significant to Monteverdi's music, as occurs sometimes, then I retain
it. But one can often modernize the pronunciation without doing any harm
to the music or the verse, and in those cases, I prefer to modernize for the
sake of native listeners.

9. Gianfrenzo P. Clivio, "Italian," in Singing Early Music, ed. Timothy McGee (Bloom-
ington: Indiana University Press, 1996), pp. 187-88; he also points out that because modern
Italian was primarily a written language until the nineteenth century, it has changed very lit-
tle since the fourteenth century, unlike English or French.
10. Tuscan "would not have held any real sway amongst the singers in Mantua and
Venice." Alison Wray, "Restored Pronunciation for the Performance of Vocal Music," in Com-
panion to Medieval and Renaissance Music, ed. Tess Knighton and David Fallows (London:
Orion and New York: Schirmer, 1992), p. 296.
140 THE RENAISSANCE, OXBRIDGE, AND ITALY

Your arguments about the text-centered nature of Monteverdi may be qual-

ified by something you said about the rise of virtuosity, when you mentioned
Farinelli and the trumpet. People argue that even in late Monteverdi there's
some move away from, the primacy of the text; and when you get to Han-
del you get coloratura that's not text-based. Wouldn't that reduce the ad-
vantage of native speakers?
I don't see that happening significantly in Monteverdi, ahhough Monteverdi
does something different, even more dramatic: he becomes less word-oriented
and more emotionally oriented. Here I disagree with Gary Tomlinson." I

would go along with Gary and others in saying that the text-centered style
starts to die out with later Monteverdi and with Cavalli; but it's only the very
slightest beginning of its demise. ^-
Now, text-centeredness dies very gradually, and it gets revived regularly

such a revival in Hugo Wolf is one reason he was a great song writer. Still,

you can't deny that the overall trend is away from text-centeredness. To me,
the real change comes after Mozart, in the 1790s.'^ Cimarosa starts doing
things that actually cover up the voice, like having an offstage band, or cho-

ruses that try to outshout the lead singer. All these things become part and
parcel of nineteenth-century opera, carrying on from Gluck. For me, the big
change is then, just as there's a big change in history with the French (and
other) revolutions followed by the industrial revolution.

In terms of patronage, what happens with both revolutions is that the aris-
tocracy is no longer reliably the main employer and supporter of musicians,
and the audience is more and more middle class. But the beginnings of that
change are at least slightly hinted at in the Venetian opera of the 1630s.
Oh, yes, everything is gradual; you can trace everything backward and for-

11. Gary Tomlinson, Monteverdi and the End of the Renaissance (Berkeley: University of
California Press, 1987) argues that Monteverdi's early works reflect Renaissance humanism as
it flovi^ered in a moment of Italian prosperity and liberalism c. 1600, but that his later works
an economic collapse, a more repressive Church, and a withering of culture toward a
reflect

preoccupation more with surface and less with ideas. Thus, Orfeo's recitatives "aspire [to] . . .

something like the sustained persuasive force of Ciceronian oratory" (p. 237), whereas many
of Monteverdi's later madrigals exhibit "a loosening of the tightly woven fabric of musical
and poetic rhetoric he had spun in earlier works" (p. 172). Tim Carter reviews Tomlinson's
book scathingly in Early Music History 8 (Cambridge University Press, 1988), pp. 245-60,
but in general it has been applauded.
12. Several authors, including Tim Carter and Gary Tomlinson, have discerned in Mon-
which musical concerns dominate textual ones to a
teverdi last years a "Third Practice," in
greater degree. In "From Madrigal to Cantata," in Music in Late Renaissance and Early
Baroque Italy (London: Batsford, and Portland, Oregon: Amadeus, 1992), p. 253, Carter
writes, "The new role given to the aria prompts the notion of a 'terza prattica'"; Julianne
Baird sees this as a broader Italian trend (in the second part of her interview, below).
13. For an interesting discussion of things) the changing relationship between
(among other
words and music and nineteenth centuries, see Charles Rosen's The Romantic
in the eighteenth

Generation (Cambridge, Mass.: Harvard University Press, 1995), pp. 58-78, esp. p. 66.
 SINGING LIKE A NATIVE 141

ward. But any historian would have a hard time denying that a lot changes
in the 1790s.

Some people say that the opening of the opera to paying customers in
Venice was part of the reason Monteverdi changed his style^"* (as opposed to

what he wrote for a noble audience in Mantua). Do you think there's any
fruit to be plucked off that tree?
Not a lot. I think it's been overplucked. It's an idea that we like because
we like any chance to exalt democracy and put down the "decadent aristoc-
racy" —
which has become almost one hyphenated word. More recently, it's
become fashionable to paint a rosy picture of the aristocracy, which of course
is equally false. They were scoundrels. But you can't say they were uncul-

tured scoundrels.
I think a historian has to realize that many of the horrors of the modern op-
eratic environment — the things that singers do to composers and their music
were there almost from the beginning. Nevertheless, there are big changes from
the early Baroque to the nineteenth and, especially, the twentieth century.

"People Say, 'It's Incredible to Hear

Monteverdi Sung by Italians'"

Rinaldo Alessandrini

I first heard Rinaldo Alessandrini's group, Concerto Italiano, after their CD

of Monteverdi's Fourth Book of Madrigals won the 1994 Gramophone Award
for best Baroque vocal recording. In a brief review accompanying the an-
nouncement, the Monteverdi scholar Iain Fenlon said, "These are performances
infused with such a strong sense of the drama of the text that it is almost
overwhelming," and concluded that "this is the finest recording of Monteverdi
madrigals ever made." That last sentence is what made me take notice; ex-

perienced reviewers like Fenlon rarely allow themselves absolute superlatives.

I arranged an all-too-brief interview, which took place by transatlantic tele-

phone and required that I rouse Alessandrini at daybreak in his hotel room
in Nice. I didn't have time to ask about his background, but according to
Fenlon, it involved some study with Ton Koopman, and beyond that a great
deal of self-education." He has directed Concerto Italiano, whose changing
personnel is always all-Italian, since the late 1980s.

14. However, Iain Fenlon argues that the Venetian audience was, to judge from the high
price of tickets, probably also aristocratic; what made Venice different from Mantua, where
Orfeo was written, were other factors, such as Venice's long-standing interest in spectacle. Fen-
lon, "Monteverdi, Opera and History," in The Operas of Monteverdi, ed. N. John (London:
Calder, and New York: Riverrun, 1992), p. 11.
15. Iain Fenlon, "Their Way," Gramophone 72 (October 1994), p. 27.
142 THE RENAISSANCE, OXBRIDGE, AND ITALY

People often argue about Monteverdi singing style. The latest example I've seen

is Jeffrey Kurtzman's in the second Monteverdi issue o/^ Early Music (February
1994). He says that in Paul McCreesh's award-winning recording of Venetian
sacred music, the singers "substitute loveliness and elegance for affect. I find
the expressive devices in this performance too limited — the vocal sonority is the
typically English 'white' sound, and tempo and dynamics are the principal
sources of 'affect.' I would like to hear the singers utilize consonants more as
vehicles for vocal expression . . . [the performers are] too concerned with re-

finement and too little with the manifold means of emotional expression avail-
able to them. " What would you say about that?
I think it's typical of the culture of English musicians that, having been
trained in choirs from childhood, they have a very collective approach to
singing. Often, they sing the solo repertoire in the same style that they'd use in

choral music. For me this is not good, of course: the voice is not flexible or
elastic enough for expressing the mood of the madrigal. In the madrigal we
have a lot of musical changes as the text calls for different emotions. I have the
feeling that the color and dynamics of many English singers is too often un-
varied.
I also have the impression that English performances tend to be anti-Ro-
mantic —that they are performing in the opposite style from that used in
Brahms or Verdi. If vibrato is right for Verdi, than it must not be for early
music. To me, this shows a lack of historical awareness about early Baroque
singing. We have a lot of documents about singing in the seventeenth and eigh-
teenth centuries from which we can realize that the style of singing was very
rich; and I feel that the English tone of voice is not a rich style. We have no

interdiction against vibrato, or against using all the possibilities of the human
voice in order to express all the different things we find in the text.

One fine English singer, Richard Wistreich, has examined a Monteverdi let-

ter and some other documents and finds that singers were praised for their abil-
ity to articulate in the throat, to support gorgie (rapid virtuoso passages) from
the chest, to do a trill, to sing loud enough to fill a church, to declaim clearly.

And they were praised for subtle shadings and for their ability to make audi-
ences weep.^^ What would you say about the Monteverdi singing style?
It seems that the most important ability was to express the mood of the
text, in the largest way possible. About the technique, the gorgie in fast pas-
sages 1 where the voice can't be as loud], we have to consider that the theatres
were not as big as those of today. We have to use a very elastic voice; it does-
n't need a powerful voice so much. Perhaps it's better if a voice is not so pow-
erful, because it lets one pass from a declamatory style, in which the voice is

16. Richard Wistreich, "'La voce e grata assai, ma . .

.
': Monteverdi on Singing," Early
Music 11 (February 1994), p. 7-19.
 SINGING LIKE A NATIVE 143

used ar the maximum of its power, to a style in which we need a more elastic

voice —with very little breath, because in the gorgie the breath is beating against
the glottis.'^ We have to consider a lot of possibilities for the voice, not just
power —though all of it is supported by the breath from the diaphragm. Of
course we don't need the same volume as for Verdi; that's historical, absolutely.
But it's without sacrificing from the voice any possibilities, such as vibrato and
portamenti, because they are so characteristic of the voice. We cannot find any
vibrato indications in the music, but that doesn't mean we can't use vibrato.
We do find a lot of portamento indications in the music. When we find slurs
in the music, they indicate that we have to sing portato. But do you hear Eng-
lish singers sing portato} Never.

[See Anthony Rooley's interview, below, for an opposite view.] Nowadays

opera singers have a wide vibrato; is it certain that the seventeenth century fa-
vored a narrower, faster vibrato?
I think that vibrato is different in each voice; every voice has its own char-
acteristic vibrato. Of course the vibrato that's good for Wagner is not for Mon-
teverdi. So we have to find a vibrato agreeable for the music — not so large. And

it's also true that vibrato was a sort of ornamentation.

In reviewing your recording of the Fourth Book of madrigals, lain Fenlon

found that your being native Italian speakers was a crucial advantage in terms
of your "strong projection of text geared to a determination to allow each de-
"^^
tail of the words to speak with due force.
We've had the same reaction from many listeners; they say, "It is incredible
to hear Monteverdi sung by Italians."
For us, it was a sort of rediscovery of the possibilities of the Italian lan-
guage. We take a lot of care in that, especially in the pronunciation and the
artistic declamation of the text. Of course, being Italian, we don't need a lan-
guage coach! Also, we know that before rehearsing the music we have to do
enormous work just to conceive a sort of theatricalization of the text. Nor-
mally, we consider each madrigal as a sort of opera scene. Before we rehearse
the music, we try to identify a certain theatrical rhythm in the text with pro-
nunciation, declamation and so on. After that we add the music.

So the text solves musical problems. Does the music ever solve textual prob-
lems?
One of the most important sources regarding declamation in music is the

17. See Julianne Baird's chapter for a further discussion of early Italian singing style. In
Iain Fenlon's "Their Way," Alessandrini is quoted as seeing aspects of the (modern) Italian
vocal sound and technique as "distinguishing features of a tradition which has not changed
since the sixteenth century"; Baird discusses this issue in more detail.

18. Fenlon, Gramophone 71 (December 1993), p. 106.

144 THE RENAISSANCE, OXBRIDGE, AND ITALY

score of Orfeo. I think it was a personal study by Monteverdi of the possibil-

ities of declamation in music. Monteverdi is very precise in notating the dif-

ferent rhythmic values in the score. So it is possible from the score of Orfeo to

create some rules for realizing declamation. For example, some words are al-

ways very important. If you find in a phrase certain words, especially adjec-

tives, that are very powerful in expression, they must be pronounced slowly. So
for me that was the first source about old declamation. It is so clear, especially

in the solo music. Normally I try to respect the relationship between the dif-

ferent values of the notes. Even if the general rhythm is a little bit more elas-

tic, and even though the bar must be really elastic, still if we read a half note
and an eighth note and a sixteenth note, that relationship must be more or less

conserved. This is very good for getting an idea of the declamation.

When we understand the structure and power of the text, we find regularly

that the music is clear. But if we start from the music, the music alone is not
always clear. When we consider the music without the text, we can come up
with multiple solutions; if we start with the text, we have only one solution.

The new seconda prattica, Monteverdi said, made music the servant of the
words, whereas the older practice had it the other way around. Yet in Denis
Arnold's view, many of Monteverdi's finest works were backward-looking ones.^'^

This is true. Especially in the sacred music, you find a lot of pieces in the

new concertato style and a lot in the older contrapuntal prima prattica style,

and it's true that much of the latter is very special. Still, I think in either style
Monteverdi realized the enormous power of the word. We have to consider that
before the music; and we have to realize that the music must be servant to the
words. This is the story of history. There's no sense in defying it in any way.

In fact, Tim Carter points out that the emphasis on text that defined the

seconda prattica was not necessarily something that emerged around 1600; it

had been part of Renaissance humanist thought for a while, and the dividing

line may not be so clear.^°

The relationship with the Renaissance is very difficult to define. We cannot

19. Arnold, Monteverdi Madrigals (London: BBC, 1967), p. 44.

20. In "Renaissance, Mannerism, Baroque?" in his Music in Late Renaissance and Early
Baroque Italy, Tim Carter suggests that "many of the aesthetic ideals of the seconda prattica —
for example, its emphasis on the close relationship between music and word —had their roots

in the Humanist movement so characteristic of the Renaissance. Similarly, the rise of opera in
Florence was claimed to be a product of that typical Renaissance activity', looking back to the
Greeks and Romans. Even the supposed shifts in compositional style and performing practice
at the end of the sixteenth century may be more apparent than real, being rooted in impro-
visatory procedures that had developed over the preceding century. It would be ironic if those
new musical styles were in fact 'Renaissance' styles in their most representative form" (p. 20).
When I read this to Andrew Lawrence-King, he strongly endorsed it, but he noted that what
was new about the seconda prattica may have been, first, that it allowed written music to take
unwritten music had been taking for years and, second, that composers like the
liberties that

later Monteverdi and D'India deliberately set texts that involved more extremes of emotion.
 SINGING LIKE A NATIVE 145

say where exactly is the dividing line between Renaissance and seconda prattica.
But at the beginning of the seventeenth century, especially with the first operas,
by Caccini and we do have a sort of reaction against the prima prattica. I
Peri,

think it was a real shock when the seconda prattica came along, to get people
to sing in this elastic way, and especially to move between singing style and recit-
ing style in a single passage. We have to wait for a musician like Monteverdi to
create really artistic results from this revolution; I think the first opera we can
listen to is Orfeo. Personally, I find the music of Caccini or Peri a little bit bor-
ing; at the time it was quite important, but at the beginning of the seventeenth
century polyphonic madrigals are more interesting than solo music.

Why do people regard Monteverdi as greater than, say, Grandi?

For me, he's not greater; for me, he's one of many great Italian musicians.
I love Marenzio, and Marenzio was composing in exactly the same style as

Monteverdi. We know that he created seconda prattica music before Mon-

teverdi. I recorded Marenzio some months ago, and it was a surprise to see that

there are elements in the style of Marenzio that are sometimes more amazing
than in Monteverdi.
Generally speaking, the standards of Italian musicians were very high. So
perhaps the problem is that we know only Monteverdi and know so little by
Grandi, Marenzio, and others. For me, the seventeenth century had so well-
formed and complete a musical language that in a certain way it was very easy
for composers to speak in that language. Monteverdi's role was very important,
but every time we do music by other composers we are astonished by its qual-
ity. Monteverdiis very well known —
okay, I'll grant that he is the greatest Ital-
ian composer of the time —
but as we have the possibility of discovering other
music, perhaps we will be obliged to place Monteverdi in a larger cultural con-
text.

Kurtzman [in the review I quoted earlier j says that in these other composers'
sacred works "there is invention of melody and musical figures fully compara-
ble to those of Monteverdi. What is lacking, however, is Monteverdi's masterly
structural sensibility; his melodic shapes, as they unfold through a motet, have
a more pronounced sense of direction and purpose, and the succession of fig-

ures and psalms have a similar sense of solidity and di-

styles in his large-scale

rection, even with his sometimes violent juxtapositions. Monteverdi gen- . . .

erates a greater perception of inevitability than any of the others. " Do you
agree with regard to the secular music?
Not totally. I repeat that Marenzio is for me one of the greatest composers
of madrigals. I cannot imagine how we can listen to the early madrigals of
Monteverdi and not see that Marenzio is at exactly the same level. Especially

in the early madrigals of Marenzio, there are very individual, very fine ideas
about structure, and about the relationship between words and music. It's like

Monteverdi, and sometimes more complex than Monteverdi.

146 THE RENAISSANCE, OXBRIDGE, AND ITALY

"It's Like the Difference Between Oils and Acrylics"

Anthony Rooley

I interviewed Anthony Rooley when he and the Consort of Musicke were on a

1995 tour of the United States, giving Purcell anniversary concerts. I had hoped
the justly renowned soprano Emma
member of the Consort, would
Kirkby, a
join us, but she declined to what we discussed as "the debate" over
engage in

English versus Italian Monteverdi singing (besides, she had errands to run on
the only free day of the tour). Rooley, indeed, had agreed only reluctantly to
address the topic. In a letter to me (20 December 1994) he said that he re-
spected Alessandrini's work and wanted to avoid polemic. Besides, Alessan-
drini's "stated aims (fidelity to the text, return to source materials, freedom
from tactus, etc.) are all entirely in accord with my aims and intentions and are
not new at all."

In the end, fortunately from my standpoint, Rooley gave in. His views do
have quite a few points of overlap with those of Curtis and Alessandrini: Roo-
ley has, for example, worked for years on the music of Marenzio, Arcadelt,
Rore, Wert, and other composers who help us see Monteverdi in context; and
he strongly emphasizes the primacy of the text. But in other respects, his per-
spective differs — he mentions, for example, his long-standing interest in the es-

oteric neo-Platonic philosophy of the Italian Renaissance.

Rooley, a lutenist, founded the Consort of Musicke in 1969 to explore the

secular repertoire of the late Renaissance and early Baroque. The group was
originally oriented toward instrumental music, but began to focus on madrigals
in 1978, when a British Arts Council bursary and a British Decca recording
contract made possible several months of rehearsal. Since then the group has
brought a wide range of long-hidden vocal repertory to the record-buying pub-
lic. As Rooley points out, the group's large (and often pioneering) discography
can obscure how it has evolved over the years.

Some critics and artists are claiming that Monteverdi is best served by Italian
singers. How would you respond?
If Monteverdi is as great as we all say he is, there is room for at least a
dozen ways of approaching him. I don't want to add one word to a debate
based on vehement polemic; I don't believe that that's what music is about.
Music is fundamentally about harmony and bringing health and well-being to
the soul. That's how it started out, and that's why the Gonzagas were paying
Monteverdi for it in Mantua. Itwas entertainment, but underneath it's about
well-being and harmony. So I would like any contribution I make to carry
something of that feeling. I'd like to embrace any "competitors" who appear
to be there and congratulate them for what they're doing and encourage them.
 SINGING LIKE A NATIVE 147

With that as a primer, then it's possible to say some things that I beheve, be-
cause I don't mean to say that I'm losing my critical faculties.

Let's talk about it historically for a moment. It is great for Italians now to
be desiring to explore Italian repertoire. To this end I and my colleagues in the
Consort of Musicke have been encouraging Italian singers, including many of
Alessandrini's singers, for about fifteen years, by teaching courses, coaching,
and so on. They needed encouragement. There was sense of failure on their

part, partly because of the strength of the living tradition of Verdi and
Rossini —they felt unconfident about anything before then —and this also af-

fected economic resources. When you have so much of a nation's money poured
into the culture of the nineteenth century, "real" singing becomes Verdi singing,

not Monteverdi singing. From officialdom downwards, pre-Rossini culture was

not really praised. You had to be somewhat unusual, even eccentric, to take an
interest in that repertoire.

We were welcomed in Italy with open arms, because we gave a fresh ac-
count of this repertoire and, to some extent, brought back with it a sense of

Italian self-respect and self-esteem. They liked us for it, and they thanked us
for it and for the care we had taken with the Italian pronunciation. For the
first time the Italians were able to hear the words to these things, because we
put the language first —always did and continually have done.
Now, another historical perspective. The musical lingua franca throughout
Europe around 1600, outside of was Italian repertoire. The
liturgical music,

whole of Europe, if it was sophisticated at all, had an awareness of what was

happening in Italian courts. Anyone who claims that this is a repertoire for the
Italians alone is clearly drawing a false conclusion from the myriad of evidence
that shows the spread of Italian culture across Europe.

The argument that's given, though, is not historical but linguistic. People
say that Monteverdi is so incredibly word-oriented, so responsive to every syl-
lable, that you can't do him justice if you're not an Italian native. Otherwise,
they say, you won't know that this word has hidden connotations, that that
syllable needs a special weighting, and so on.
It is incredibly word-oriented, but of course people all over Europe were
struggling with the very same problem then. But they were determined to work
with it. In most places they had a few Italians on hand. And the English, in a

way almost more than the Italians themselves, nursed an interest in some of the
most esoteric things that were happening. A lot of the Italian repertoire would
not have been transmitted to us were it not for these English students, as it were,
of Italian music. The English had a sense of the special quality of the Italian
repertory being created at that time, and they took care of it, nurtured it, stud-
ied and learned from it. Now that, I would suggest, is a relationship that I have
attempted to continue. We have nurtured an awareness of Italian culture, albeit

transmitted through the eyes, ears, hearts, and minds of English people, but with
148 THE RENAISSANCE, OXBRIDGE, AND ITALY

real love and care. We have never for one moment pretended that we are Ital-
ian or could do what an Italian group could do. We've listened to what Italians

have to say and learned from it, I hope. Yes, there will always be passages for
which we can learn about the Italian language, and how it should be accented
and nursed. But I would suggest that some of what's being done now by Ital-

ians has learned a great deal from what we've done in the past.^'
Regarding which, another historical perspective to recall in this "debate" is

that our performing is a continuing journey. Because we have recordings going

back almost two decades, I can now put on my CD player something we
recorded in 1980 and compare it to what we're doing presently. The changes
are absolutely phenomenal. Our first recording project was, believe it or not,
Gesualdo's Fifth Book of Madrigals. What a place to step in! When I listen to

that now, I hear that it's a study project. It's full of effort. The fact that it sits

in the bins and looks as current as the thing we did last year is just in the na-

ture of the CD industry.

What makes a very interesting little experiment is to compare our 1984
recording of Monteverdi's Fourth Book, Concerto Italiano's [1993] recording
of the Fourth Book, and our 1992 version of some of the madrigals from the
television film we did in the Gonzagas' Palazzo Te in Mantua. That is a fair

comparison, and brings this dialogue rather more up to date. Anybody who lis-

tens to those recordings will hear them as if there were three different ensem-
bles approaching this repertoire, not just two.
Because we've recorded so frequently, others can listen to what we've done
and react to it and in dialogue with it go a step further. You can hear Alessan-
drini reacting to something we've recorded before, and because we did it this

way, he's going to do it that way —which is perfectly natural. You can also take

a step back from where we are now. Indeed, we've gone so far since then that

there are others now who are saying "Oh, the Consort have gone too far,

they're too theatrical, they're too dramatic." You can't get it right, of course.

People either like where you're coming from or they don't, and can find all

kinds of reasons to criticize it.

In the beginning, too, you were exploring new ground for the early-music
movement. You were, for example, the first group to record integral books of

madrigals.
The reason I was doing that was that if a composer had felt there was a

certain integrity to bringing together a collection of separate works and pre-

21. In Fenlon's "Their Way," Alessandrini isquoted as saying that twenty years ago, "all
Italian singers of early repertoires wanted to sound like Emma Kirkby; their efforts were
doomed to failure from the start, he believed, since they inevitably involved the destruction
of cultural backgrounds and traditions." On the other hand, in some Monteverdi recordings
by Italians (not Alessandrini's, obviously, or those of some others) the singing has been too
Romantic in style. These instances support the view that non-Italians like Curtis and Rooley
have played a role in fostering the current risorgimento of Italian early-music singing.
 SINGING LIKE A NATIVE 149

senting it as a monument to his patron, or whatever other impulse it was, that

was an integrity worth honoring.
We were also pacemaking in our choice of singers. They were all singers ca-
pable of singing with great clarity, with a voice that's centered and relatively
vibrato-free, so we could start from that and then use colors as they came from
the various consonant and vowel sounds of the language and from the mood
and sense of the poetry. You could build out, in almost a sculptured way, a per-
formance that was molded and bent in terms of vocal color and vibrato. If you
start with a voice that has a directness and simplicity about it, it allows ex-
pressive additions to have meaning, rather than starting with a voice that is so
fruity in itself that what you'd have to do is bring it down rather than allow
it to expand.

Some specific charges that Alessandrini made about English singing were,
indeed, that "they don't use vibrato" and that they don't use portamento. Yet
I heard both yesterday in your concert and on, for instance, your Musica Os-
cura CD of Monteverdi.
good measure, and used with some considerable
In skill, so it's clearly not
something that we've just taken on recently.

Another complaint is that "the English" don't make enough of the vowels
and, especially, the consonants of Italian.
It has nothing to do with being English or Italian, I believe. We've got some
pretty exciting consonants in English too, and when you get to a writer like
Dowland or Purcell, they're using the English consonants as well as the Eng-
lish vowels. How you use consonants depends upon your awareness as a per-
former of the space you're in, and of the mood and mode that you're present-
ing in that space. Then it devolves on the director, if there is one, as to how
far he or she wants to take explosive consonants and their use. I like them, and
I use them quite a lot, and we are pretty theatrical in our approach. You won't
hear that in our 1984 recording of Monteverdi's Fourth Book, but you will get
it if you look at our film from 1992. If you listen to our performances in con-
cert today, you will find that the consonants are as alive as those coming from
anybody else. We've stepped out; we've learned how to handle it. And we're
learning all the time.

Another thing people criticize "the English" for in this music is suppressing
solo expression in favor of ensemble.
Well, we don't search for "blend." I hear that word used a lot with regard

to English vocal ensembles in general. The King's Singers, for example, is an

ensemble famous for blending. We couldn't be further apart on the spectrum,
because what we want is to have four, five, or six voices as individuals, more
like flutes, violins, oboes, and violas interweaving together. Certainly it has to
be an ensemble. My model was to bring my group's work to the level of con-
150 THE RENAISSANCE, OXBRIDGE, AND ITALY

sistency you expect of, once again, a string quartet, but using the voices and
their special quahties individually. Emma Kirkby's soprano voice against Eve-
lyn Tubb's — there's a wonderful difference, and I've wanted to exploit it.

Now, to compare this approach not to the King's Singers but in the other

direction, I'd suggest listening to Rimanti in pace, from Book Three —one of
Monteverdi's best compositions, hardly known, but a masterpiece, absolutely
sublime. Listen to our recording, and then to what the young Italian groups
have been turning out recently; it clarifies the contrast. I find that hearing the
Italian recordings is like enjoying the powerful effect when you first see acrylic
colors. You think Wow! I've never seen colors like that! They're so bright,
they're so vibrant, they're so alert, they're so up-front! But contrast that to
painting in oils. The oils take you deeper and have a sense of chiaroscuro, sub-
tleties of shading. Now I would say our Rimanti in pace is Titian using oils,
and some of the Italian recordings have a more slightly edgy quality, belonging
more to acrylics.
The reason is partly to do with recording technology. Almost all my record-
ings since 1982 have been done in Forde Abbey in the West Country in Eng-
land. It's a medieval space, with a stone flag floor well-worn by centuries of
feet going across it, so that it's an undulating surface. There are no flat sur-

faces in the room, really, except for some glass. It gives a most magical expe-
rience of sound. it's warm, it's sweet, and the sound seems
It's vibrant, it's alive,

room for you to view and enjoy, and then

to present itself in the middle of the
gently evaporate to make room for the next sound. When we record, I try to
use the natural dynamics and ambience of the room and capture as much of
that as possible on the final tape. These things form part of the alchemical mix
that performance is. If you cut out an awareness of the acoustics of the room,
you've cut out an essential ingredient.

But you believe "the Italians" don't record that way.

What we're hearing nowadays increasingly goes right into the midst of the
voices. Because the recording engineers are having a bigger and bigger say in

how the end sound comes around, we're getting a closeness of recording where
you hear the warts and all of each individual voice. At first this is incredibly
exciting to the ear. It can be mind-blowing, because the ears have never heard
it like this before. But it's the aural equivalent of using a microscope, or a tele-
vision lens that can go so close to the mouth you can see almost the larynx of
the singer. Nobody in Monteverdi's time would have heard it sung this way,
because if you're in an ensemble you've got your own space in which you're
singing, and you're hearing your colleagues in their space — but that is not like
an independent microphone sitting in the middle.

It may have not been a historical experience, but what about its musical
costs or benefits?
I feel it produces anti-polyphony. Polyphony is about the melding together
 SINGING LIKE A NATIVE 151

(and, again, I'm not talking about "blend") of the various elements to make a
dialogue of equals. What we microphone technique is anything
get with close
but a dialogue of equals. It causes the breakdown of polyphony, because it puts
the emphasis on magnified detail rather than on the whole. Once you've got
accustomed to the and sharpness of sound and the bright colors
brilliance as-
saulting your ears, you begin to hear the strengths and weaknesses of each in-

dividual voice as it comes and goes into the line. This is highly distorting to
your ability to keep an overall picture of where the composition is going. And
there are very few singers, I would suggest, whose technique is sufficiently, con-
sistently under control to be able to give us a line that can stand that degree
of close inspection all the way through. It's certainly not true of some of these
rather inexperienced young singers who are being used to present these over-
vivid realizations. You hear the flaws, they're disturbing, and it's not fair to the

singers. It is vivid, but it's a very modern experience. It's almost a surreal ex-
perience.
Now, it's a style of its own. And it may be that some of us want to go in
that direction in the future. But it's important to state that this is not how any-
body could have heard it previously. The squeaky-clean CD technology has
now taken us so far that we can in fact begin to distort our perspective of how
music should sound, because the technology begins to interfere. We've stepped
into a new technology and new use of polyphony (which, again, I happen to
a
think is anti-polyphony). Then when you step back from it again you're in great
danger, because you lose the brightness and clarity which was so exciting to
the ear at first. It's like being used to music being turned up several degrees too
loud: when you turn it back down, it sounds tame.

As you say, though, some might prefer it even if they acknowledge that it

is unhistorical.
Of course. And even stating the contrast as I have is making too big a case,
because when you come to compare the recordings they're often not that far
apart. Very often the difference is, "What's the big difference?" So let us get all

these elements out of the way; then we can begin to have a proper dialogue
about niceties of emphasis here or there, taking more the overall architecture
of a work and seeing how the parts of it can be expanded and compressed. A
shift of perspective would take the debate further in.

One thing that CDs don convey 't is something we see in your concerts, and
also in the film you made in the Gonzagas' pleasure dome: acting and chore-
ography to accompany the madrigals. Could you talk about this?
Contemporary sources tell us that madrigals were performed with suitable
facial and hand gestures and so on. They became theatrical. The works them-
selves have theatrical implications and possibilities. So the film I made was a

serious attempt to marry all of those things together, with the rider that in the
end there's a degree of compromise because it is for television — just as you have
152 THE RENAISSANCE, OXBRIDGE, AND ITALY

to decide what kind of sound you want for polyphony when you're making a
record. This is not for the Gonzagas' ears.
Still, in making the film, and when recording a CD, I have in mind the re-
sponse of a group of the kind of people the Gonzagas, or any other patron,
would have gathered together. People who were cognoscenti; people who knew
not only about the music and the poetry, but also about the philosophies of
performance lying behind them. People who knew why the arts were there: to
raise one's mind upwards. It wasn't simply for entertainment; it wasn't just
dainty music to adorn. was music which was there to some extent to im-
It

prove and educate and lift up. It was a music which was addressing not only
the mind, not only the literate intellect, but the soul as well. It was a music
which was created — as was all music, in Renaissance thought — in order to cre-
ate a sense of harmony and well-being, and also to carry, to some extent, a di-
vine furore, such as Orpheus's divine frenzy represented. All music-making had
that behind it. It would be so taken for granted that it would not always nec-

essarily be forward consciously in the mind, but it would be there in the back-
ground of thought: that it's impossible to divorce musical performance from
the quality of Orphic frenzy that we find expressed by Ficino^^ and his like. So
itwas just there, a given, which it isn't for us. We have to work hard to re-
mind ourselves of it, and that sophistication of awareness is something we find
very hard to reproduce today.

How do you relate that, then, to modern audiences, ivho don't have the
background you describe, and usually will not catch all the classical references

that might have been meaningful to contemporary Academy members?

I think that an attuned performer is going to adapt his or her performance
differently to each audience and space. When you gather people in a space for
a performance, the moment of creativity is very exciting. It's like watching an
artist about to put the first brush stroke onto a plain bare canvas. The audi-
ence witnesses that taking place; it's a special moment.
Now, a seasoned, subtle performer will adapt to that, and in that is the

translation for modern audiences of the Renaissance ideal I described. For that's
really where art becomes life. It's not a seeming of life, it's not a play, it's the
being of life. I think therein is the great shift. How far is this art.-* How far is

it just a play, and how far is it truthful? How far is it a reflection of the human
condition in one aspect or another, be it meditative or theatrical or whatever
else.' And I feel that this is why the seemingly over-esoteric debates of the acad-
emies of Monteverdi's time are very important for us to tune in to. Because life

is too short just to play. Life is too short to say, "My approach to Monteverdi
is better than yours." In approaching the music, you raise curiosity and you

22. Marsilio Ficino, a Florentine Platonist (1433-99) whose translations and commentaries
were extremely influential. He is discussed in depth in Gary Tomlinson's Music in Renaissance
Magic (University of Chicago Press, 1993), esp. chap. 4,
and from a performer's standpoint
in Rooley's Performance (Longmead, Dorset: Element Books, 1990).
 SINGING LIKE A NATIVE 153

entertain, but it's saying much, much more than that. If it doesn't have that di-
mension, then I think I've been in the wrong business!

SELECTED DISCOGRAPHY
Alan Curtis —whom Andrew Porter calls "an interpreter with a rare instinct for
living rhythms, pulses, and inflections"^^ — has been ill served by the record
companies, and many of the recordings he has made are out of print. His finest
opera recording is probably Handel's Floridante (CBC SMCD5110); but the
disc preserves only excerpts. Still, many listeners will not really mind that in

opera seria; and Stanley Sadie, a leading critic of Handel opera performance,
says that the recording "sets new standards in the presentation of Handel as a
dramatic composer." Sadie praises Curtis's "full-blooded . . . alert, knowledge-
able direction," and "a cast who show a real grasp of how to convey power-
ful emotion through this idiom." He also mentions "a good deal of music of
supremely high quality."^"'

As for Monteverdi opera, Curtis's // ritorno d'Ulisse in patria (Nuova Era

7103, 3 CDs) gives a sense of the kind of singing he discusses in his interview.

It is, moreover, a powerful performance. It is indeed a performance; it was

recorded live in (I am told) essentially one night at the Siena Opera. As one
would expect of live opera, it has moments of out-of-tune singing and must not
come as close to Curtis's ideal as a studio recording might. For listeners who
share my disdain for the modern preoccupation with polish, though, it is

strongly recommended. I have not been able to procure Curtis's 1980 Poppea
(Fonit Cetra LMA 3008/A-D); Thomas Walker admires it but complains about
its recorded sound and its cuts, while Ellen Rosand praises its conviction and
Curtis's deep understanding of the work.-' Among recent releases are an inter-
esting exploration of Neapolitan chromatic seventeenth-century harpsichord
music (Nuova Era 7177), and his first recording with his new group, I Febi Ar-
monici (Symphonia SY 93S25), featuring pieces by d'India, Marenzio, and
Monteverdi. Margaret Mabbett finds the Monteverdi the "least secure perfor-
mance," but says that "this fine group . . . otherwise gives Concerto Italiano
some real competition."^*
Having expressed reservations about the priorities of the recording studio,
I should note that in a review of Monteverdi madrigal recordings Eric Van Tas-
sel argues that recordings may allow us closer access to madrigals than mod-
ern concerts do: madrigals, he declares, depend on intimate subtleties that are
lost in today's concert halls. -^ Van Tassel's review is the most balanced com-
parison I've read of the two "rival" groups whose leaders I interviewed. An-

23. In Music of Three Seasons (New York: Farrar Strauss Giroux 1978), p. 373.
24. Sadie, Gramophone 70
(January 1993), pp. 58, 60.
25. Both quotes are from reviews in Historical Performance 4 (Spring 1991), p. 53
(Walker), p. 73 (Rosand).
26. Mabbett, "Monteverdi and Other ItaHans," Early Music 24 (May 1996), p. 359.
27. The New York Times, Sunday Arts and Leisure section, 9 April 1995, pp. 31-32.
154 THE RENAISSANCE, OXBRIDGE, AND ITALY

thony Rooky's Consort of Musicke, Van Tassel says, sings with the "bold elas-

ticity of dynamics and phrasing" that the historical evidence calls for, and its

"singers spur one another on more daring than

to interpretive flights of fancy
any of them might have ventured on their own." Rinaldo Alessandrini's Con-
certo Italiano "[declaim] and [act] out the words without inhibition, with
swooping glissandos on words like 'alas' and unabashed accelerandos as cli-
maxes .approach; some pauses between phrases are so generous that we
. .

lose track of the meter altogether." In the Second Book (Opus 111 30-111) and
Fourth Book (Opus 111 30-81) of Madrigals, "there are abrupt staccatos
where Monteverdi writes whole-notes; spectacular changes of tempo; an almost
lubricious rallentando on 'squeeze me till I faint.'" He believes that Concerto
Italiano is "imperfectly housebroken in matters of 'authenticity,'" but holds
that "even the most extravagantly overdone moments spring from an authen-
tically madrigalian concern for the words." Their great strength, in fact, is their

fluency in the Italian language; they can sing "just the way they speak across
a dinner table." But the group's youthful voices "do not yet interweave as
gracefully as the Consort's"; the "interplay among the voices still needs work."
Van Tassel concludes by dismissing the rivalry: "these groups are complemen-
tary, not competitive."
Concerto Italiano has also recorded some Monteverdi sacred music (Opus
111 30-150) —an interesting case, since the music is in Latin, not Italian, so
their linguistic advantage is irrelevant. Iain Fenlon, who has called their Fourth
Book "miraculous," and generally much prefers the group to English rivals,

finds the sacred music disc brimming with "revelations and surprises —no seri-

ous Monteverdian can afford to be without it."^^ The group's many other
recordings have generally been quite warmly received. This is also true of
Alessandrini's solo keyboard recordings; a good sampler of his work in this
field is his "150 Years of Italian Music" (Opus 111 30-118). His harpsichord

recordings of Bohni (Astree E 8526) and Buxtehude (Astree E 8534 show how )

superbly an Italian can play German music, should anyone doubt that.
When we talked, Rooley gave frank assessments of his own earlier record-

ings of Italian music (those of English composers have been almost universally
praised). The earliest he considered "study projects." He hked the group's Vir-
gin recordings of Monteverdi — particularly, the first three books of madrigals
(Book One, Virgin 45143; Book Two, 59282; Book Three, 59238)— but he was
less pleased with the Balli of Book Eight, in which the microphones caught the
Consort on an off day. He has recently reissued some recordings (many made
in the 1980s by West German Radio) on his own label, Musica Oscura. Among
those relevant to this chapter are the CDs of Marenzio (many consider this one
of the group's best recordings; Musica Oscura 070992), Monteverdi (Jonathan
Freeman-Attwood notes that it is "more overtly impassioned"^^ than their mid-

28. Fenlon, Gramophone 74 (July 1996), p. 90.

29. Freeman-Attwood, Gramophone 71 (March 1994), p. 94.
 SINGING LIKE A NATIVE 155

1980s recording; 070995), and Notari (Fenlon describes some of the perfor-
mances on this disc as "quite breathtaking both Hterally and metaphorically";^"

070983). On the other hand, Mabbett says that their CD of Pallavicino (07096)
shows the group "tiring of late 16th-century repertoire" and reluctant to "at-

tack the harsher sounds of the text." She complains that the radio recording
constricts their true dynamic range, and in general prefers the group's "excit-
ing live performances.""" Yet this same CD stirred Anthony Pryer to praise the

Consort for providing an object lesson in how to avoid "the dangers of habit

in a life of music-making, in how to expand the personality of a group to fit

"^-^
the music.

FOR FURTHER READING

Excellent discussions of Monteverdi's operas and of their performance can be
found The Operas of Monteverdi, English National Opera Guide no. 45, ed.
in

Nicholas John (London: Calder, and New York: Riverrun, 1992). A good gen-
eral introduction to Monteverdi can be found in The New Monteverdi Com-

panion, ed. Denis Arnold and Nigel Fortune (London: Faber and Faber, 1985).
A good introduction to the madrigals is the BBC Music Guide by Denis Arnold
(London, 1967).
The foremost discussion of early Italian opera today is Ellen Rosand's
Venetian Opera in the Seventeenth Century (Berkeley: University of Califor-
nia Press, 1991). Some provocative recent work on Monteverdi includes Gary
Tomlinson's Monteverdi and the End of the Renaissance (Berkeley: Univer-
sity of California Press, 1987), although I recommend reading the critical re-

view by Tim Carter in Early Music History 8 as well. Carter's own book of
essays, Music in Late Renaissance and Early Baroque Italy (London: Bats-
ford, 1992; Portland, Oregon: Amadeus, 1992) contains fascinating material
as well.
A scholarly discussion of Renaissance Neo-Platonism and music is Gary
Tomlinson's Music in Renaissance Magic (University of Chicago Press, 1993),

His discussions, in chaps. 3-5, of Ficino et al. are fascinating, whatever one
makes of the Foucaultian philosophizing that occupies the outer chapters.
Tomlinson has disavowed any interest in performance; Rooley, of course, is

preoccupied with it book on Renaissance musical mysti-

and even called his

cism Performance (Longmead, Dorset: Element Books, 1990). The book re-
lates Neo-Platonic Renaissance ideas to the modern performer's problems; its

New Age slant puts off some readers, but others find it engagingly personal.
Rooley has also published a fascinating article on passion in Monteverdi
singing, called "L'humore universale," in Musical Times 134 (September 1993):
490-95.

30. Fenlon, Gramophone 73 (August 1995), p. 117.

31. Mabbett, "Monteverdi and Other Italians," p. 357.
32. Pryer, "Assuming Personalities," Musical Times (September 1995), p. 493.
156 THE RENAISSANCE, OXBRIDGE, AND ITALY

Postscript: Nationalism and Early Music

The debate over who can best sing Monteverdi gives an example of the influ-

ence of nationalism in the early-music revival. If Bach scholarship has flour-

ished, it was at least originally through the work of German scholars; if Han-
del scholarship lagged behind for decades, it was, as Winton Dean noted,
because no one country claimed Handel —a Saxon who immigrated to Lon-
don —as a national asset. ^^ As for performers, England would probably not have
emerged as such an important center for early music if its own music hadn't
reached an apex before 1700, and if modern Britons had been less fascinated by
their own earlier history. Conversely, as Rooley points out, Italy might have
emerged sooner if it hadn't been intoxicated with Verdi and Puccini.
But if nationalism has been beneficial for early music, national borrowings
and interactions have been even more salutary.^"* Think of Bach, whose mature
style depended crucially on Italian and French influences. Such interaction has
raised the current level of Monteverdi madrigal performance in both England
and Italy to a new high. The Italian/English controversy, however absurd it may
seem, will probably continue to bear artistic fruit in both countries.

33. Dean, "Scholarship and the Handel Revival," in Handel Tercentenary Collection, ed.
S. Sadie and A. Hicks (Ann Arbor: UMI Press, 1987), p. 3.
34. A classic discussion of this with respect to folk music is Bela Bartok's "Race Purity in
Music" (1942; reprinted in Music in the Western World, ed. Piero Weiss and Richard Taruskin
[New York: Schirmer, 1984]). Bartok shows that borrowings back and forth between Slavic
and Hungarian folk music repertoires have almost always improved upon the "pure" origi-

nals.
 8
Emotional Logic
—^eOe^
Andrew Lawrence-King on
Renaissance Instrumental Music
and Improvisation

Conventional wisdom holds that during the Renaissance, instrumental music

was a sideshow. Vocal music, such as that discussed by Peter Phillips and Paul
Hillier, had suchprestige that only after 1600 did instrumental repertory come

into its own. But according to the harpist-keyboardist Andrew Lawrence-King,

conventional wisdom is wrong.
It became the conventional view anyway, he argues, partly because of the

modern preoccupation with written scores. In Renaissance instrumental music,

much of what was important was not written down but was improvised. This
raises, he points out, a paradox at the heart of the early-music movement: to

be faithful to the spirit of the past often means being unfaithful to the written
notes that survive from the past.
It also points to another issue. Structure has been a critical element of West-
ern art music; composers like Bach, Haydn, and (most influentially) Beethoven
used large-scale structure as a powerful expressive device. Music theorists in

the past century or two have focused much of their attention on understand-
ing large-scale integration in music. But it has been argued that this focus has
led us to undervalue local, non-structural musical elements — most ornamenta-
tion, for example —and to overestimate the emotional significance of structure.

Meaning and emotional power, some argue, reside on the music's surface at
least as much as in its undergirding. The way a jazz singer bends a note may

have as much expressive significance as the way a composer provides catharsis

157
158 THE RENAISSANCE, OXBRIDGE, AND ITALY

by bringing back an unresolved musical progression and at last resolving it.

Lawrence-King's discussion of how he improvises gives a sense of why we

shouldn't ignore local elements: compared to written music, he says, improvi-
sation is structurally much looser, yet in concert it compels an audience just as

much as pre-composed music does.

Preoccupation with perfect musical integration may be a later development,
but composers from Vitry' to Mozart gave structures —often remarkably tight
ones —to their music. The tendency to seek order in experience is one of the
deepest of the human mind. But Lawrence-King reminds us that in music, order
can take different forms, some of them quite fluid.

Lawrence-King acquired his first harp, a medieval Irish model, during a

party at a harpmaker's house. Because no one knew how to play the instru-
ment well enough to teach it, he taught himself. He then mastered the Italian
Renaissance double harp and other historic harps as well. He has been in de-
mand ever since as an accompanist or continuo player for many leading artists,
including several in this book, such as Christopher Page, Paul Hillier and Gus-
tav Leonhardt; and as a soloist, he has won high praise from the critics. His
solo playing shows us why the harp was in the Middle Ages no variety-show
act, but the most prestigious and admired of instruments, and was in the Re-
naissance an instrument for master virtuosi only.

Perfect Instruments

It's widely held that the prestige of vocal music militated against the develop-
ment of purely instrumental music in the sixteenth century, except for dances,
and that it was in the seventeenth century that true instrumental genres devel-
oped. What do you think?
I'd like to answer that by sidestepping the question. The important ques-

tion to ask is subtly different: not Where is Renaissance instrumental music?

but, rather, What did Renaissance instrumentalists play? The answer has to
be in the first place that they mostly played vocal music. That doesn't down-
grade what they did; it's simply that they took vocal music as their reper-
toire, which is very similar to what happens today in popular music. If you

go into a restaurant, you'll hear a lot of instrumentals being piped in, which
are really vocal pieces in instrumental arrangements. Of course, the ideal that
instrumentalists strove for was to have the same flexibility and expressivity
1. The fourteenth-century French composer Philippe de Vitry, who coined the term Ars
Nova new style, devised one of its main forms, the isorhythmic motet. About this
for his era's
Reinhard Strohm says, "In order to absorb the world into its own structure, music had first

to acquire such a structure, at least in principle. This happened —only on the level of written
art music, to be sure —
in the theory and practice of the French Ars Nova" specifically, —
Strohm then explains, in the isorhythmic motet. Strohm, "The Close of the Middle Ages," in
Antiquity and the Middle Ages, ed. James McKinnon (London: Macmillan, 1990, and Engle-
wood Cliffs, N. J.: Prentice Hall, 1991), p. 270.
 EMOTIONAL LOGIC 159

in an instrumental version of a piece that one could hope for in a vocal ver-
sion. That was, in fact, one of the roles of ornamentation in some of the in-
strumental repertoire.-^

But did this general picture change in the seventeenth century?

It is the conventional view that the early seventeenth century brought a
change: instrumental pieces are independently published with titles, the two sig-

nificant ones being canzona and especially sonata —which means "what is

played" as opposed to cantata, "what is sung." In fact, though, these early

sonatas are more parallel to vocal music than might be assumed. I don't think
there's a great splitting away of explicitly instrumental music from vocal music;
I think, rather, that the early sonata repertoire is a way of translating the new
gestures of vocal music into instrumental terms. So all the new development
that was taking place in vocal music around 1600 monody, the new reciting —
style, and the episodic style of writing vocal pieces according to the text, which

meant that from phrase to phrase the music changes radically from recitative
style to arioso style, or changes speed or mood —
all of this episodic style was

taken over by the sonata.

There's a second reason why I don't think 1600 is such a sharp dividing
point. True instrumental styles actually developed earlier, in a different genre
namely, the various styles of writing for what they called the "perfect instru-
ments," that is, instruments that could play polyphony as well as a single
melodic line.

Such as the lute, harp, and keyboard instruments?

Also the viola da gamba, which I think had a special place because its early
history was so associated with a lute-like Spanish instrument, the vihuela. The
vihuela could be played either with the hand, like the lute, or with the bow;
and the later viola da gamba was a descendant of the bowed vihuela. From
that point of view it had an especial association with polyphonic playing and
with the style of the perfect instruments.

So the perfect instruments are really where the interest begins with instru-

mental music.
Yes, because they're instruments that have a lot of functions. They can of
course play single lines, and we know that ensembles existed where these in-

struments would play single lines (or, at most, thickened-up versions of single
lines). But because they could play polyphony, perfect instruments could ac-
company a solo voice or instrument by taking the remaining voices of a com-
position and combining them into one accompanying part. Many solo-voice-

2. See Lawrence-King's essay "Perfect Instruments," in Companion to Medieval and Re-

naissance Music, ed. T Knighton and D. Fallows (London: Orion, and New York: Schirmer.
1992), esp. pp. 356-58.
160 THE RENAISSANCE, OXBRIDGE, AND ITALY

with-accompaniment publications were in fact arrangements of four-part vocal

pieces, based on exactly that approach.

And this pre-dates the seventeenth century?

The first printed works of this kind are right at the beginning of the six-

teenth century, in Petrucci's publications.

Of course, the other function of the "perfect" instruments, besides accom-
panying, is that once you can play polyphonic lines you can play an entire poly-

phonic solo on one instrument: you can play vocal works as solo instrumental
pieces.

So I'm making a different distinction here than vocal/instrumental. As I sug-

gested at the outset, one can't simply make a distinction between vocal and in-
strumental music, because vocal music was routinely played by instruments as
part of their normal repertoire. And, equally, instrumental music could be
sung.^ But as we begin to get published sources of instrumental music in the
sixteenth century, we find polyphonic pieces that are certainly related to the
normal styles of vocal polyphony, but that have a number of distinctive fea-
tures that are not only non-vocal but are also non-consort. These features come
about because the music is to be played on one instrument. The usual practice
may be for vocal consort music to be played by an instrumental consort; but
there's something else involved when one person, on a perfect instrument, can
play music that normally involves several people. It's that special capacity
which led to the development of particular instrumental styles for these instru-

ments.

What are some of the distinctive features of these instrumental styles, and
what capacities of perfect instruments led to their emergence?
Some key features are liberties that a soloist can take with the polyphony
that a consort can't take. An important such liberty is playing around with the
number of voices in a free way, so that on a particular chord, or for a few
notes, you add extra voices simply more resonance or sonority. You haven't
for
created that voice strictly polyphonically, nor do you lose it polyphonically: it
just appears as needed. You do occasionally see this in vocal writing, especially

in English sources where the top part splits into what is called a "gimell," but

you see it very often in the perfect-instrument pieces, in ways that would not
be practicable even if you had a couple of singers on one of the parts who could
separate to sing chords.
A second uniquely instrumental feature is a way of creating polyphony that
sounds stricter than it really is, where the number of parts stays constant but

3. Stewart McCoy, "Edward Paston and the Textless Lute Song," Early Music 15 (May
1987), pp. llX-li, describes an English practice whereby a piece that appears to be for lute
and solo instrument could also be performed by lute and solo singer, with the singer either
vocalizing or using solmization syllables.
 EMOTIONAL LOGIC 161

a part may exit as a bass part and reappear as a high soprano part. That's very

convenient on one of these perfect instruments; you're only holding down three

or four notes at a time, so your three- or four-part polyphony can stay strict.

You see that a lot in early organ music, for instance that of Tallis. But if you
attempted to realize this in ensemble music you'd be faced with a bass singer
having to make a high soprano entrance; even in an instrumental consorts, the
bass instrument would have to make an entry that is impossibly high in its

range.
I think even more significant is something you see in the first published
source of vihuela music, a book of 1536 called El Maestro by the Valencian
composer Luis Milan, in the many pieces that he calls either fantasia or tiento.

This is the style that Milan calls consonancias y redobles, consonances and re-

doubled lines.The redobles are fast passages and scales, which run up and
down the whole compass of the instrument. This is virtually impossible to do
by combining several instruments or several singers, each of which works in a

limited range; but it's simply an idiomatic style for a single instrument with an
extended range.
The consonancias y redobles style leads to another point that I think de-
serves emphasis. Because the cutting edge in instrumental composition was with
these perfect instruments, and because these kinds of instruments did a lot of
accompanying, I see numerous links between accompanying techniques and
solo instrumental techniques. To connect this to the specific example of conso-
nancias y redobles, the correspondence is with the styles for the so-called ro-

mances, where the instrument plays simple chords while the singer is singing.

Every time the singer pauses at the end of a phrase, the instrument takes over,
and instead of playing back something similar it plays something of great con-
trast —
you have simple chords accompanying the singer, and then you have fast
scale passages in between the phrases. The fast scale passages employ the whole
compass of the instrument, often including chords in unusually high positions,
so it's something that displays the abilities of these perfect instruments, in par-
ticular their ability to cover the whole compass. The style of these romances,
with their alternations between the vocal passages accompanied by simple
chords and the instrumental passages of fast scales, seems to be what is mim-
icked in the consonancias y redobles style of fantasia.
Another thing you see in the fantasia are tantalizing hints of the improvis-

ing style, a style sometimes consciously imitated in these works. The most fa-

mous example of this is by the Spanish composer Alonso Mudarra: a piece for
vihuela, dated 1546, written in imitation of the improvising style of the Span-
ish harp player Luduvico. It has features which are clearly imitations of the
sound of the harp, features which seem to be those of the improvising style,
and features that break the normal rules of polyphony by the inclusion of
strange notes. It also exhibits a much freer way of writing, neither strictly poly-
phonic nor strictly melodic, but a combination of melodic fragments and
162 THE RENAISSANCE, OXBRIDGE, AND ITALY

chords. These features are very characteristic for any one of these perfect in-
struments, but very uncharacteristic of polyphonic music whether vocal or in-
strumental, where you're writing for a certain number of singers or players with
limited ranges.
Finally, what we also see in the Spanish sources, right from the beginning
with Mudarra, is an interest in falsas, which usually means illegal harmonies
that disobey the normal rules of polyphony. These give you particularly inter-

esting chords and sounds. We then see falsas —called stravaganzas in Italy — in

the Giovanni Trabaci and Giovanni de Macque generation of keyboard and

harp compositions in Naples around 1600, and we see similar dissonances later

on in Spain in sources for the guitar. That is quite extraordinary, because the
guitar —with, at that time, five courses and a rather narrow compass — is in

some ways quite a limited instrument. It is particularly important in this ex-

amination of instrumental music, because it uses styles which are purely in-

strumental and solo, with no vocal imitation. Originally, the guitar didn't play
in the polyphonic style, it only played chords which could be made rhythmic
by strumming. As the guitar repertoire developed, this interest in strange har-

monies was always well to the fore.

The guitar was at the time a purely harmonic, non-polyphonic instrument,

—
and it was not the only one another one to emphasize was the lyra. Such in-
struments indicate an awareness of harmony in the mid-sixteenth century which
we tend to overlook. The conventional wisdom used to be that music was con-
ceived entirely polyphonically until the invention of the continuo, at which
point composers and players for the first time begin to think harmonically,
whereas before the harmonies were something that resulted from the accumu-
lation of polyphonic lines. I think the use of instruments like the guitar and
lyra show that this wasn't always true. The players of those instruments were

entirely harmonically aware, and both of those instruments abandoned

polyphony in the sense that they abandoned the regular movement of the bass
line. It's normal on the guitar —
since it doesn't have a very large bass com-
pass —to play chords as they arrive, without requiring that the bass note be the
lowest sounding note of the chord. A G chord with D as its lowest note did-
n't bother the guitarists at all and was not perceived as inverting the chord po-
sition.

It's interesting that among the early sources that talk about continuo play-
ing is one that describes the sound of an entire piece being re-created from the
realization of the continuo line. That suggests to me that even in the "Renais-
sance," and right at the beginning of the Baroque, musicians were able to rec-
ognize a piece by harmonic content as much as by the polyphonic lines that
supposedly made it up.

Did these elements from the stylesof perfect instruments influence other
genres of music writing? An essay of yours refers to Haar and Pirrotta specu-
 EMOTIONAL LOGIC 163

lating that the sound of the Italian fourteenth-century madrigal derived from
improvisatory organ playing, and you suggested that the Burgundian chanson
might have been influenced by the kind of accompanying you spoke about.^
With the idea that the three-part chanson was influenced by instrumental
style, I'm referring particularly to the level of quick activity that you often find
in the tenor and, especially, the contratenor parts of these chansons, where
fairly static harmonies are kept on the move by the voices repeating or ex-
changing the notes: the harmony doesn't change, but the two voices keep ac-
tive. That seems very reminiscent of the way a plucked-string player keeps the

harmony sounding, just by repeating notes, exchanging parts if necessary, and

using rhythmic repetition to keep harmonies in the air. That has the effect of

giving music rhythmic impulse and rhythmic life, which is why it would have
been interesting to take it into vocal music.

You're talking about fifteenth-century chansons —Dufay, Binchois et alJ

Yes, the standard Burgundian chanson, which typically has a vocal style in
a texted upper part, and two lower parts which appear to be less what one
might call vocal in style. Of course, the trend nowadays is to perform these
works entirely vocally, with those lower parts either texted or sung to solmiza-
tion syllables. Yet we know that those parts were also played instrumentally, in
such a combination as lute and harp accompanying a singer, or with the two
lower parts combined for one lute or one harp.
Perfect instruments also influenced vocal styles in the sixteenth century. The
four-part frottola style of the early Petrucci prints was derived from lute im-

provisation in essentially two-voice polyphony.^ These frottolas were then re-

arranged as lute songs, bringing the vocal style back to the perfect instruments.

How about influences on music for purely melodic, non-perfect instru-

ments f
Although at the outset I asserted that early instrumental sonatas are less a
novelty than an instrumental equivalent of what was happening vocally, it's true

that they begin to explore elements that are non-vocal. These include the ex-
tended use of the kind of instrumental flgures that suit a particular instru-

ment — particular combinations of notes, ornamentation, and tricks and turns

of phrase. That's one aspect, the use of special effects for instruments. But, yes,
the other aspect is the copying by melodic instruments of the kinds of non-con-
sort effects that were being used a long time before by the perfect instruments.
An example is the use of strange tunings for the melodic instruments, tunings
that give you more harmonic and solo possibilities, such as Biber's for the vi-

olin. Violins experimented more and more with double stops, allowing the re-

4.Lawrence-King, "Perfect Instruments," p. 355.

5. F. Prizer, "The Frottola and the Unwritten Tradition," Studi Musicali 15, ed. Leo
W. S.

Olschki (Rome: Accademia Nazionale di Santa CeciUa, 1986), pp. 8-12.

164 THE RENAISSANCE, OXBRIDGE, AND ITALY

alization of polyphony in forms specially adapted to the violin. That I think is

a very clear example of something that was happening with many of the
melodic instruments. Seeing the freedom and the possibilities open to the per-
fect instruments, they explored the possibilities of working in that direction.
One thing that I think is very important in this regard is the so-called style
brise which evolved from the lute, where you play a piece which is progressing
in a completely normal polyphonic way with a melody and a bass, and poly-
phonic parts sitting in the middle, but instead of playing the chords vertically
and simultaneously, you play the notes one after another, sometimes in a very

subtly varied order. You don't just play the arpeggio up from bass to treble or
down from treble to bass, but you pick the notes out one by one in a subtle

and subtly changing way. It produces a very rich texture and all kinds of sub-
tle and jazzy syncopated rhythmic effects, although the underlying music is

quite simple. It's obviously a very idiomatic technique for any instrument that
arpeggiates, like the lute or the harp. But it was taken over by the harpsichord

and even by the organ in, for example, some Buxtehude and Bach organ pieces.
Moreover, this technique of playing a polyphonic texture by playing the notes
one after another allowed instruments that weren't polyphonic at all to create
the illusion of polyphony. Thus we even have quasi-polyphonic works by Tele-
mann and Bach for —
unaccompanied flute the instrument is purely melodic, but
you can make a solo suite for a flute by imitating the solo perfect instrument's
style brise.

Another issue: later instrumental music is usually for specific instruments —

sonatas for piano, or for flute, harp, and violin. When did this specificity
come in?
There's a growing body of evidence to indicate that instrumental music went
on being unspecific much later than we used to assume. The harp, for exam-
ple, had a small repertoire of music specifically "per I'arpa, " and an enormous
repertoire of pieces shared with singers and other instruments. We certainly
know for the Baroque period that the repertoire of harp players was identical
to that of keyboard players. There recently was discovered a mid-nineteenth-
century library belonging to a harpist in Wales, which contains mostly vocal
and keyboard music. And that points out a consistent factor from at least the
Renaissance through the Baroque, which is that instrumentalists of all kinds
mostly played vocal music, and otherwise shared a common repertoire of key-
board music. This instrumental repertoire, especially in the earlier eras, was
playable either by consort or by any of the perfect instruments. Specificity was
the exception rather than the rule.
The lists of instruments we see on title pages of printed works in the sev-

enteenth century are determined more by market forces than by anything else.

They are an indication of performances by the — mostly — amateur players or

the lower-ranking professionals who would be the target of that kind of mar-
 EMOTIONAL LOGIC 165

keting, rather than very precise information about performances by the top
ranks of performing virtuosos. Record companies searching for repertoire for
specific instruments lead to an inauthentic concentration on music performed
by "original" forces as described on title pages. Literary descriptions of per-
formances suggest that instrumentalists spent more time playing "cover ver-
sions" than "original hits."

To digress: you've mentioned Spain several times —a country widely ignored

in popular histories of music. You've written elsewhere that there was a well-
trodden path from Spain to Naples to Rome to Paris, taken by many of these
aspects of instrumental composition, and taken by the harp.^
And by the guitar. Those clearly are the routes taken by the improvisatory
styles of composition — fantasia, toccata, prelude — and by the large double harp

and guitar. It may be no more than coincidence, but I find it a strong one, es-
pecially given the link between the harp and improvisatory music, and given
that the harp and the guitar as perfect instruments make very particular in-

strumental colors rather than only trying to imitate the voice. This route is also
that taken by the chaconne and the passacaglia. Their descending four-note
bass theme is definitely Spanish —
it's what the whole of flamenco is based

upon! It takes the same route through Naples up through Rome, and there is
Roman upon it, in Luigi Rossi for instance. And then
repertoire entirely based
the chaconnes and passacailles become associated with the high Baroque French
keyboard composers, moving from there even to Bach.

On Improvisation: "Emotional Logic"

I want to ask you about improvisation, because it has to do with the nature of
music as we think of it. It's not done in the mainstream classical world.
The theme in our time has been a separation between improvisation and the
performance of great masterworks of the past. It's a separation particularly

painful for early musicians who are trying to deal with a paradox in being
faithful to the great masterworks of the past, because half of their fidelity must
include what was called sprezzatura, the willingness to disdain, not to take too

seriously the object that you're dealing with.^ In other words, to be faithful to
the spirit of the music one must be prepared to alter the written notes. In a

way, it's a paradox that impinges on all period performances, since they require
ornaments, but it's particularly important when thinking about instrumental
music and the perfect instruments, because these, being able to play completely
solo, are the instruments that are most convenient to improvise on. Obviously,
it's easier for one player to improvise as a soloist than it is for pairs or larger

the notes to the CD The Harp of Luduvico (Hyperion 66518).

6. In

The idea was expounded in Castiglione's 1528 // libro del cortegiano (The
7. Courtier)
and apphed to music by, among others, Caccini in Le nuove musiche (1602).
166 THE RENAISSANCE, OXBRIDGE, AND ITALY

groups to agree on ground rules for improvising together, although they did
that too.
We know that Renaissance performers added improvisation to their com-
plete performances. For example, if a performer played a suite, he'd probably

improvise a prelude to begin that suite. In fact, "suite" means "that which fol-

lows on." It's the continuation, and the thing it continues is the prelude. Of
course, there are written preludes, but the best performers would usually im-
provise their own, which could be as long as the whole rest of the suite. (That
gives you a nice balance between this first piece, which is in the improvised and
solo instrumental style, and what followed —a group of dance pieces which
tend to symmetry and fixed forms, and perhaps more strict polyphony.)
Today we tend to venerate the great written masterpieces and see music
made up by performers as irreverent, but in the past that certainly was not the
case. Improvisation had a parallel status to written music, even in the fairly re-

cent past: some of the great nineteenth-century pianists were admired as much
for their improvisations as for theircomposed works. Many of the great mas-
ter composers were actually the main performers of their music, sometimes

being the only ones who could play it at the time: the separation between com-
poser and performer was a lot less strong. And this was much more so in the
Baroque era. We know of J. S. Bach improvising at considerable length and in
fact taking part in a competition to improvise; and that skill was valued for its

own sake. We know also of the French art of preluding, from surviving writ-
ten preludes that attempt to give a sense of the freedom, and the different way
of structuring, found in improvised preludes. As I mentioned, regarding those
preludes and the toccatas/fantasias and so on, I think improvisation had a very
important effect on written music.
I think that in general music history tends to place too much emphasis on
the first notated evidence of each new departure in music, simply because it's

more easily traceable and so more convenient to write about. But many of the
important developments were already happening way before the notation was
established. Consider, for example, the notation of basso continuo as a way of
accompanying music: I think the sound of basso continuo was happening long
before the notation was developed. And I think that's especially true for the
question of instrumental performance and the kinds of freedom that were no-
tated just by chance in the Mudarra piece imitating the improvising style of
Luduvico. This was clearly a recognizable improvising style, because that was
the point of the piece: people could say "Aha! Yes, that's how Luduvico used
to sound." In the written repertoire, apart from this one piece, we don't see
these harmonic and structural freedoms for decades afterward; but they clearly
were there among improvising instrumentalists.
I think that's a particularly important point: the structural freedom. If

you're improvising, it's very hard to keep hold of a formal design, where you
remember a phrase well enough to repeat it, for instance, or where you work
 EMOTIONAL LOGIC 167

out a long, elaborate formal pattern. The planning of the piece tends to be in-

spirational: what you play at this moment gives you an idea of what to play
in the next moment. This "organic" development is very satisfying for an au-
dience; they'll follow the piece along with you. It's very different from the kinds
of formal organization we see in written pieces. We see this kind of inspira-
tional planning in some of the fantasias, in particular the Mudarra piece. We
don't see it again in written music until we get to the very late ornamented ver-
sions of madrigals, where the ornaments themselves develop in this inspira-
tional rather than formal way. These pieces have the formal underpinning of
the original piece on which the ornaments are being written, but the ornaments
themselves develop in this more organic, less formal way. Otherwise we see it

best in monody, where the structure of the piece follows the poetry and its emo-
tional development, rather than formal elements. And of course the monodists
tended to choose particularly emotional and dramatic texts, which justify an
emotional logic for a piece rather than a formal logic. When they were setting
arioso or dance movements, they'd choose texts with a more coherent struc-

tural form. As I said, this is imitated in the solo sonatas at the beginning of the
seventeenth century.
That point is made concrete by Frescobaldi, who says his toccatas should
be played like the modern madrigal, by which he means that you should vary
the Affekt [predominant emotion], vary the speed, and feel free to omit certain
passages if you want to— again, a lot of freedom for the performer. The point
he's making there is that the formal organization of the piece is not based on
strictly instrumental forms but, rather, is copied from the vocal recitatives,
which themselves take their organization from the emotion of the text. So you
have this emotional logic behind the piece rather than a formal logic. You're
able to hold it together by performance rather than by composition. If you have
a piece which is unified by emotional development, rather than by fixed for-
mal elements, then, assuming the performer can supply this emotional content,
it isn't damaging to the piece to leave out one of the sections as long as the
emotional development is still coherent. Whereas in a piece designed in a fixed
form such as a dance — let's say a galliard, with three strains, each of which is

repeated — it would be quite strange to leave out one of the strains. It would
have an odd feeling, because it's upsetting a fixed formal progression.

You improvise yourself in concert and even on your CDs.

Yes, I think that this is very important. In the same way, over the last twenty
years we've come to realize that adding ornamentation to Baroque music isn't

being sacrilegious to the great masterworks, but what was intended is rather
and must be done. Many performers now have learned the delight of making
that added ornamentation spontaneous.
I see the revival of the Renaissance art of fantasia not as a radical new de-
parture but simply as the next stage on —from improvising ornaments for the
168 THE RENAISSANCE, OXBRIDGE, AND ITALY

dances, to improvising the prelude which would normally begin a suite of

dances. Of course, it's quite a challenge to be free enough to improvise but re-
main within the style boundaries, especially when one knows that improvisa-
tion is where the style boundaries were stretched the most anyway.

You mentioned that they also improvised in groups.

I think that one of the important things to consider in the earlier period is

that opinion is beginning to shift now with regard to large-scale works, to see
(for example) the Monteverdi Vespers not as a choral and orchestral piece, but
rather as concerted music sung by a team of soloists. And seeing these large
works in this way means that the possibilities for spontaneous ornamentation
and other kinds of improvisation are greater than they appeared to us before.
To give another example of how supposedly large-scale pieces could still in-

corporate improvisation, there seems to have been a practice in Rome, which

is preserved in a few pieces now in the Uppsala library, involving the continuo
players who accompanied a four-, five-, or six-voice canzona —which likely was
played by a violin band —with a continuo of perhaps a couple of harpsichords,
a few harps, and a couple of lutes. There are certain sources which between
the sections of the canzona give passages that have simply a very slow-moving
bass line as a basis for improvisation for each of the accompanying instruments.
In fact, in some of the sources, even the violins get a passage like this to im-
provise over. And it seems probable that the few sources that notate this prac-

tice are the tip of a performance-practice iceberg, in that this would normally
happen even when it wasn't notated. So in the middle of, say, a Frescobaldi
canzona, the violin band or the wind band would stop and take a breather
while their continuo improvised for a few bars.

Have you experimented with group improvisation?

Actually, with my new group, the Harp Consort, the idea is to take varied
ensembles of perfect instruments — harp, keyboards, viola da gamba, lutes, and
guitar —and explore the links between improvised and written music. We are
all continuo players, and obviously continuo is one such case, where the player
has a written bass line but the reaUzation is improvised. We're attempting to
take the continuo player's mind set —where one is trying to produce a realiza-
tion that is free and spontaneous, but also very historically informed and ap-
propriate to the period, and apply it to other areas of performance besides con-
tinuo — if you like, taking continuo thinking as an inspiration for carefully
stylized improvisations. There are enough indications in period sources that

show that they too saw this link existing between continuo improvisation and
solo improvisation.
It's fun. It's quite a challenge, because of this paradox of trying to be as free
as possible, as spontaneous as possible, while also trying to remain carefully
within the style boundaries. It's not enough to play anything and say, "Well,
 EMOTIONAL LOGIC 169

that was free." But the advantage that continuo players have, aside from the
fact that they're playing the very instruments that used to improvise, is that
they're used to dealing with this paradox in the job of continuo playing. From
there it's a relatively short step to doing it in the context of solo playing or
group improvisation away from accompanying, as well as improvising in ricer-
cars and canzonas.

To what degree does improvising well depend on education in composition

and theory?
I think this again brings us to the paradox we spoke about earlier, that to

improvise you need as much technical information as possible about the com-
positional style of the period, the rules of composing. But you also need to com-
bine that with the freedom to just play. I do quite a lot of teaching of impro-
visation, and I usually start with the second of those, the freeing up, as the first

phase. Then, as a second phase, I go almost in contradiction and try to refine

it somewhat. Usually I've got ten people who are terrified of improvising; but
once they have the feeling for it, I change tack and, for example, teach about
how to cadence within a particular style, since once you get going on an im-
provisation one of the most important things is to know how to stop! It really

does need an awareness of styles, whether that comes from training in compo-
sition or training in analysis, or simply from playing a lot. I certainly would-
n't want to underemphasize that side of it, the awareness of style that one
couldn't describe formally but nevertheless is very strong. That's how all jazz
players improvise. They know what sounds right and what sounds wrong,
based on a very subtle and complicated set of style rules that few of them
would want to articulate but that they're all extremely aware of. The challenge
is trying to get that same sort of deep-down awareness of a Renaissance or

Baroque style, so that you can bend the rules but not break them.

SELECTED DISCOGRAPHY
Lawrence-King and Harp Consort have both signed long-term recording
his

contracts, so it many more CDs, both solo and group, will be

is likely that

available by the time you read this. As I write, Lawrence-King's solo discogra-
phy consists of two CDs on Hyperion. The second of them, The Harp of Ludu-
vico (Hyperion 66518), has better sound and freer performances; it includes the
Mudarra fantasy spoken of above. It's an extraordinary CD that I recommend
strongly (though the earlier CD is certainly recommendable too).

Lawrence-King has appeared on over eighty recordings by various ensem-

bles. He plays on several of Christopher Page's CDs, which include some of his

harp solos, and on some of Jordi Savall's. In addition there are his Teldec
recordings with the group Tragicomedia, which he formed with two other con-
tinuo players, Stephen Stubbs and Erin Headley. The high point of their discog-
170 THE RENAISSANCE, OXBRIDGE, AND ITALY

raphy may be a selection from the Notebook of Anna Magdalena Bach (Teldec
91183). EUiot Hurwitt, reviewing the disc in the November 1994 Fanfare,
called it his "CD of the year . . . and I imagine of the decade as well." Hur-
witt called one of the disc's numbers, an arrangement of Couperin's Les berg-
eries, "quite simply the most delicious single performance of a piece of music
that I can remember ever hearing."
Lawrence-King's new group, the Harp Consort, has released two CDs as I

write: Mustek's Hand-maid (Astree Auvidis E 8564), a "colourful and inven-

tive"^ Purcell program; and Spanish Dances — "a joyous experience, not to be
missed"^ —which offers selections from Ruiz de Ribayaz's 1677 collection Luz
y norte musical (Deutsche Harmonia Mundi 77340).

FOR FURTHER READING

Companion to Medieval and Renaissance Music, ed. Tess Knighton and David
Fallows (London: Orion, and New York: Schirmer, 1992), has several short es-

says on Renaissance instrumental music (chaps. 22-26, 32, and —most stimu-
lating, to my mind — Lawrence-King's own chapter, "Perfect Instruments,"
chap. 49). Performance Practice: Music before 1600, ed. Howard Mayer Brown
and Stanley Sadie (London: Macmillan, 1989 and New York: Norton, 1990),
contains Brown's rigorous surveys of known and not known about in-
what is

strumental performance in the Middle Ages and Renaissance. Brown also wrote
a very good introduction to the subject of Embellishing Sixteenth-Century
Music (Oxford University Press, 1972). Finally, A Performer's Guide to Re-
naissance Music, ed. Jeffrey Kite-Powell (New York: Schirmer, 1994), contains
valuable background and advice on a number of instrumental genres, includ-
ing the plucked strings. Herbert Myers's chapter on harps (pp. 154-60) is con-
ciseand useful, as is Paul O'Dette and Jack Ashworth's section (pp. 201-14)
on "proto-continuo."

8. Jonathan Freeman-Attwood, Gramophone 73 (January 1996), p. 76.

9. John Duarte, Gramophone 73 (January 1996), p. 76.

 C?Qj J J J ^.^o

THE BAROQUE

Robert Benchley once remarked that there are two kinds of people: those who
divide the world into two kinds of people, and those who don't. In recent
decades, two national schools have dominated Baroque performance: the
Dutch, whose style has been described as more inflected (or, if you don't like

it, mannered), and, again, the English, whose style has been called more direct

and energetic (or, if you don't like it, boring and modernist). Yet talking with
musicians suggest that this oversimplifies. The Londoner Monica Huggett, for
example, developed her unique, imaginative style in the Netherlands, and the
Dutchman Anner Bylsma considers swelling on most long notes (which some
consider a "Dutch" trademark) tasteless.
A second oversimplification is that this ignores other nations. Austria played
an influential role in fostering early-music Baroque style, thanks to Nikolaus
Harnoncourt (Julianne Baird is one of many artists who studied with him). In
more recent years, some critics believe, Cologne has displaced Amsterdam and
London as the international capital of early music' Yet I represent Cologne
only by the American expatriate Barbara Thornton, and — regretfully — I neglect
its father figure, Reinhard Goebel. Also regretfully, I haven't covered the excit-
ing scene that is emerging in Spain, though the one in Italy has already been
represented by Rinaldo Alessandrini, who of course argues that Italians sing
Monteverdi best. By contrast, the French scene is represented by a non-native,

1. James Oestreich, "The New Sound of Early Music," The New York Times, Sunday Arts
and Leisure section, 21 July 1996, pp. 1 and 32.

171
172 THE BAROQUE

William Christie —
the American who, more than anyone, catalyzed its current
excitement. Home-based Americans are represented as well, but as to whether
there's an "American" Baroque style, as the record producer Wolf Erichson ar-

gues,^ I would suggest that if maestros Gardiner and Norrington were Ameri-
cans, people would describe their styles as "very American."
Along with the by-now-familiar issue of nationality. Baroque playing brings
up some new issues. Renaissance players don't need to defend the use of his-
toric instruments, for example, but in repertoire that people had been playing
for decades or generations on modern instruments —Bach, Vivaldi, and Han-
del —the issue does arise. (The first three interviews delve into various aspects
of using early instruments.) For similar reasons, vibrato (or its absence) be-
comes more controversial. Improvisation and ornamentation, which have been
broached in some of our discussions of Renaissance and medieval music, be-
come central concerns; a number of references are made to jazz. And a new
analogy surfaces repeatedly in this section (and later, too, in discussions of the
Classical era). Musicians in the seventeenth and eighteenth centuries often com-
pared music-making to speech. Joshua Rifkin, however, sounds a cautionary
note later in the book: was the Baroque concern with rhetoric just a way of
describing what good musicians of any era have always done?

2. In an interview by James Keller Performance 6 (Spring 1993), p. 34. After

in Historical

saying that many want to sound critical,

of the best players are American, he adds, "I don't
but it is a typically American characteristic to be rather extroverted. A bit of what you might
call a 'Juilliard Style.'" Queried on that —
since Juilliard pretty much neglects early music
Erichson answered, "But the musicians are coming from there, or at some point they're trained
by people who did. . . . It's a different aesthetic."
 9
Consistent Inconsistencies
— oOo^
g^

John Butt on Bach

In 1827, Goethe wrote that the Well-Tempered Clavier sounded "as if the eter-
nalharmony were communing with itself, as might have happened in God's
bosom shortly before the creation of the world." In 1950, Pablo Casals spoke
of Bach's ability to "strip human nature until its divine attributes are made
clear, to inform ordinary activities with spiritual fervor, to give wings of eter-

nity to that which is most ephemeral." It was such expressions, perhaps, that
led the composer Lou Harrison ("a Handel man myself") to grumble, "Why
don't they simply canonizehim and be done?"'
To which many would respond. Good idea. To his devotees, Bach's mys-
tique originates in his music: one feels that there's much more to it than meets
the ear, and that the extra something is profound. His mystique may be height-
ened, though, by our tantalizingly limited knowledge of his life and his per-

formance style. These limitations have given rise to a field of research distin-
guished by cunning detective work and ongoing ferment. As one leading Bach
scholar, Christoph Wolff, said, "There hardly exists a more fascinating and re-

warding subject in the history of art than the music of Bach."'

For an example of the challenges of Bach scholarship, consider Bach's ar-

1. Lou Harrison, "Cloverleaf," in Companion to Contemporary Musical Thought, vol. 1,

ed. John Paynter et al.(London: Routledge, 1993), p. 254. The Goethe quote is from the post-
script of a letter to Karl Friedrich Zelter, 17 July 1827; the translation is from Robert Mar-
shall, The Music of Johann Sebastian Bach (New York: Schirmer, 1989), p. 71. The Casals
quote is from an essay written for the 1950 Prades Festival.

2. In his Bach: Essays on His Life and Work (Cambridge, Mass: Harvard University Press,
1991), p. ix.

173
174 THE BAROQUE

ticulation marks. These slurs and dots may have aided Bach's contemporaries,
but they flummoxed modern scholars for decades. The confusion lay in what
Erwin Bodky called the markings' "incredible discrepancies";^ for example,
three simultaneous parts playing the same notes may have three different slur-
rings. Attempts to find a consistent interpretation of the markings kept failing,

a fact that Bodky found "bitterly disappointing."

A 1987 Cambridge University doctoral dissertation yielded more palatable
results. Through an exhaustive survey of Bach sources and Baroque perfor-

mance treatises, the author, John Butt, discerned some of the motivation be-
hind Bach's markings. He showed, moreover, that the markings give special in-

sight into how Bach understood his music. A book based on the dissertation
won the "William Scheide Award of the American Bach Society in 1992."* Those
findings formed the center of our conversation. Eventually the discussion
moved on to Bach's metaphysics —which. Butt showed, connects meaningfully
with Bach's perfect technique.
The conversation took place in Berkeley, where Butt was serving the Uni-
versity of California as associate professor and University Organist. (He has
since then accepted a position at his alma mater, Cambridge.) We spoke in

Butt's basement, which he had converted into an office with the usual equip-
ment —computer system, fax machine, etc. — as well as the specific tools of his

trades of scholarship and performance: at the periphery, taking up every square

foot of wall space, shelves crammed with books and scores; near the center of

the room, a harpsichord and a piano.

On Using Original Instruments

You've written that even when Bach's music is played on synthesizers, "if the
notes are correct and played in the right place, Bach's genius is still somehow
"^
there.

I was partly paraphrasing Bach's own words. ^ One conception of Bach,

which I think is valid, is that his music is there whatever you do to it inter-

pretively. It's very hard to play Bach well, but if you get the notes right it's very
hard to ruin. With Handel, on the other hand, you can get the notes right and
nothing else and it will sound absolutely terrible — there's nothing there at all.

But with Bach, such is the tautness and tightness of the writing that if you are

3. Erwin Bodky, The Interpretation of Bach's Keyboard Works (Cambridge, Mass.: Har-
vard University Press, 1960), p. 214.
4. John Butt, Bach Interpretation: Articulation Marks in Primary Sources of J. S. Bach
(Cambridge University Press, 1990).
5. John Butt, recording review. Early Music 17 (February 1989), p. 116.

6. Bach, when comphmented on his organ playing, is reported to have said, "There is noth-

ing remarkable about it. All one has to do is hit the right notes at the right time, and the in-
strument plays itself." From The Bach Reader, ed. Hans David and Arthur Mendel (New
York: Norton, 1966), p. 291.
 CONSISTENT INCONSISTENCIES 175

inclined to appreciate it, you'll get something out of it. More than with virtu-

ally any composer up to our time, you can appreciate it however bad the per-
formance is. From that point of view it doesn't matter in the slightest how it's

played, or on what instrument.

So why bother with historical information?

What interests me in terms of historical performance, if we're talking about
Bach, is that the performer can perhaps gain some insight into the way this

music was created in the first place —the tradition of improvisation, the tradi-

tion of performance, the traditions relating performer and composer in terms

of their professions and practice, and so on. As a historian and a performer, I

like to put my two disciplines together and try to recover and experience some-
thing of the creative process behind the music.
You see, one error I think there is in historical performance is the empha-
sis on what the first performers did, when they encountered this music. I think
historical performance has been rather superficially concerned with what's the
right way to play and the wrong way to play, with what should be "done to"
the work, which itself is taken as a given. As far as I'm concerned, I never know
what's right or what's wrong, but I'm always keen to look for new ways of dis-
covering the music. I'd rather go beneath the surface and ask, "How did this
music come to be written like that in the first place?" What are the origins of
this particular kind of form, this particular kind of figure, this ornamentation,
this note, and so on? I think questions of historical performance should be
placed in the realm of the original creation of the music, rather than merely the
original reception of the music.

Where, then, might you start?

Well, for one thing, it's nice in some ways to be able to see Bach's music al-
most from the perspective of the seventeenth century. As a performer, I like to

see this music emerging out of the raw material of seventeenth-century musical
practice —the musical figures, performance conventions, and so on. Almost all

of the ingredients are there — not quite the formal structures, such as the con-
certo ritornello, but most of it. By limiting yourself like that historically, and
putting yourself notionally in the positions of the creator of that music and the
creator of the original performances —the two roles I think were very close,
closer than we often imagine for Bach, —
anyway you can get a particular in-
sight, as a performer, into music that has a lot of colors, levels, and implica-
tions in it. Now what I do as a performer might not come across to any lis-

tener (indeed, it doesn't to some!); and even if it does, I can't prescribe how
one listens to Bach. But this is how I like to approach it.

So to understand Bach you immerse yourself music of Buxtehude, in the

Reinken, and other seventeenth-century composers he would have known?

176 THE BAROQUE

Definitely, and in the general things that Bach would have been thinking
about. Above all, the function —what is this music for? What is he trying to do
in this music? How is he trying to use conventions in new ways? That is cen-
tral to a lot of Bach's compositional thought. It's inventing nothing new, virtu-
ally, but constantly using conventional musical ideas, conventional forms, mix-
ing them up, and combining them in new ways.

Robert MarshalF gave the example of Cantata 78's first movement which —
Baroque conventions, the chaconne en
unites different, apparently incompatible
rondeau, the Lutheran chorale, and the cantus firmus. Which means you need
to know all those conventions, and all the other ones too.
Yes, and what was on his —
mind was he concerned with me, the performer,
when he wrote this? Sometimes he might not have been.

As in some late contrapuntal works.

Right. In other words, what were the issues involved in creating a piece of
music? What was it meant to be doing, what was it doing, and how does it re-

late to other pieces of music?

The playing techniques have a place, but I think that if there's any mileage
in historical performance, it's in other issues, such as the issue of what counts

as a "piece of music." Is it the notation as we see it today? Is it the original

manuscript? Or is it something more subtle than that?

Such asf
A combination of influences: the historical prejudices on the part of a par-
ticular composer, and the resources available to him, the limitations of any par-
ticular performance medium. I think historical performance can help us under-
stand the intricacies of what actually counts as being a musical work.

Which is more than the notes, or than the notes played well.
Yes. To put it differently, where does the work stop? That's like asking
where the human being stops: is it a single person, or is it a whole range of
cultural and genetic influences?

The Goldberg Variations might be a good example. We can look at them

we can look at them as a range of culturally
as a self-sufficient structure, or
—
derived genres sarabande and Scotch snap, fughetta, French overture, quodli-
bet, and, of course, canon.
Yes, the later works of Bach are often trying to encompass everything. You
might look at the Goldbergs as a compendium of all the different styles of the
age, or you might look at it as a compendium of all the harmonic implications

7. Marshall, The Music of Johann Sebastian Bach, pp. 76-79.

 CONSISTENT INCONSISTENCIES 177

of one particular bass line —you can appreciate it very deeply on one of these
counts while ignoring the others, and still get an incredible amount out of it.
And there's also some sense of thoroughness, particularly in the way the canons
work, of going full circle, of almost creating a curved universe.

So historical instruments (getting back to that issue) would relate to one of

those ways of looking at Bach — the cultural and creative origins — but not to
others.
Well, historical instruments might correspond to the conditions under which
he wrote the music. So in some ways historical instruments can help show you
how he came to conceive of writing music in one particular way rather than
another completely different way. It gives you something of the context, just as

in order to understand a human being you might look at their parents and their
family.
On the other hand, you can appreciate human beings apart from their par-

ents; whether they have problems or not, you can appreciate them on their own
merits in relation to where they happen to be now. And the same might be true
of the music. Where does the music happen to be now.-* How does it relate to

other music that we're using today? How does it fit into that equation? From
that point of view, original instruments, logically speaking, are irrelevant.
As far as I'm concerned there are strong intellectual problems with a strict

insistence on using original instruments. I have a subjective affinity to using his-

torical instruments, because I find that their limitations are analogous to the
limitations the composer knew in his own age, which may have influenced him
to write in a certain way, to create works in one way rather than some other
way. So again it's bringing us back to the creative process.

On Bach's "Inconsistent" Articulation Marks

One avenue we might use to enter Bach's creative process is explored in your
book on Bach's articulation marks. In the past, they hadn't seemed to admit of
any consistent interpretation. You found that they do, if one understands the
historical context.
There are consistencies within the inconsistencies, in various ways. For ex-
ample, inconsistent articulation marks within the same piece sometimes reflect

differences in the instruments used. That's the case in the gamba sonatas, where
a harpsichord using a specific slurring would produce a result different from a
gamba slurring the same way. In those circumstances, what might superficially

seem a different slurring could produce the same musical result.

On the other hand, sometimes I feel that Bach just wanted a basic slurred
idea and didn't really care how it was realized. So that's another issue, that quite

often he's not consistent because he doesn't consider it essential —or even, per-
haps, thought it counterproductive. An example are those movements where the
178 THE BAROQUE

basic 4 meter is broken into triplets. In the last movement of the A Minor V'io-

iin Concerto, for example, it's not clear whether these are to be slurred 2+1 or
3. Quite often the player's technique or choice will determme it. Of course, some
may argue that the markings are ambiguous only to us, and that the origmal
would have known from local convention what Bach desired. But if they
players
knew what he wanted all along, why did he bother to mark virtually everv' slur-
ring m the autograph solo violin part? Just one or two slurs at the outset would
have sufficed. Li other words, it was crucial to him that a slurred st>'le be
adopted —three slurs per bar — but the details were not to be notationally carved
in stone.

So the exact details of the slurring weren't critical to him there? You've given
the last chorus of the St. Matthew Passion as a related example —as in, say, the
last bar, where the voices have no slurs, the first violin a four-note slur, and the
second violins two-note slurs.

Yes, that's an example of where he wants a general slurred effect to make

the strmgs more of a background to the voices. So it's just a device to make the
strings quieter or less forward at that point, and he's not concerned about how
you do it (there are similar B Minor Mass, among other works).
examples in the

So part of my view on the inconsistencies is that you have to be ver\' care-

ful about whether the slurring is actually important at a particular point. Some-

times there's even.' e\idence that to Bach it didn't matter. Of course, it's hard to
convince people today that different players in an ensemble should play some-
thing differently or completely randomly!

We're more concerned than most past ages were with ensemble unanimity.
But how do you tell whether the exact slurring does matter in Bach:
It might depend on the music. For one thing, if it's an inner part or an ac-

companying part, I would say he was not concerned as much with motivic co-
hesion as \A,ith foregrounding certain elements in the texture and not others. In
that context, slurring acts as a way of shading out certain areas of the texture.
So one has to decide, within the overall texture, how important the music is

that's being slurred.

one of the trends of twentieth-century performance has been

We//, to remove
the differentiation between foreground and background material. Yet it appears
to be an important element of Bach interpretation.
Oh, certainly, yes, very important. I think in Bach's music, and that of many
other composers of the time, you need to see the music as a simple structure
which has been ornamented at several overlapping levels, each of which might
bring out other relationships within the music. You shouldn't see the music as
cast iron, note after note after note, as you would in Brahms, where every note
is of the same importance — to a certam extent —
withm the larger Imes.
In other words, within the music you need to see hierarchies of pulse, disso-
 CONSISTENT INCONSISTENCIES 179

nance, and so on, all the way through to the overlaying of different types of fig-
ures and ornamentation. Now, this is a theory I've drawn from a variety of his-
torical data, and there's no book I can point you to from, say, 1715 that puts it

exactly as I do.

It seems that performance-practice studies of a generation ago were often al-

gorithmic — —
when you see X, do Y about such thmgs as ornamentation and
rhythmic alteration. You're saying that you can't read Bach's articulation marks
simply, as if they were algorithms.
Right: you first must understand something of the music, its structure and
its various historical implications.

Can we discuss that, beginning with the nature of its structure:

The simplest hierarchical level, you could say, is perhaps the grammar, and
then there are other aspects of syntax, rhetoric, emotion, and decoration which
go above that.

Like verbal language? Let's begin with grammar.

The "grammar," the first of several parameters that might define articula-
tion —the first level in the hierarchy — is the way the music is put together in
terms of meter, accent, and bar. These provide the metrical hierarchy of the
music. From that point of view you come up with a fundamental st>le of artic-

ulation in which what Baroque theorists called "good notes" — notes on strong
beats, like the first beat in a bar —are stronger than what they called "bad
notes." And that concept always seems to be in the background of Bach's music.

Why was the downbeat stronger in this era?

You only have to know the rules of harmony as they were codified from the

early seventeenth centun.-, up until Mozart and beyond, to understand that the

downbeat is assumed to have a function that the upbeat doesn't have. This is in

the very structure of music, in the way dissonance is used. You prepare a struc-

turally significant dissonance, like a suspension, on the weak beat, sound it on

the strong beat, and resolve it on the weak beat. That's there all the way
through, and if the suspension is done in the wrong place, unless it's skillfully

handled, it's a compositional anomaly.

So the strong beat gets the dissonance.

Right. But if you're learning to write a passing dissonance, one that carries
less structural weight, the first dissonance you learn is the unaccented passing
tone, which comes between two harmony notes and mustn't be on the strong
part of the beat.
In some sense, then, melody and harmony have rhnhmic dimensions. You
cannot understand meter unless you understand dissonance, and vice versa.
And dissonance is perhaps the ver>' substance of tonal harmony. Particularly in
180 THE BAROQUE

late Baroque harmony, the control of dissonance what gives the music
is really

its power. Directly connected to that, of course, is where you hear the disso-

nances and where you learn to put the dissonances; and that brings us right
back to the hierarchy of beats.

I'd like to return to the dissonance issue later, but first get back to the ar-
ticulation marks. One thing you spoke about in your book is how Bach's marks
relate to metrical stresses.

Or bring out contrasts in metrical stresses, or deviations from metrical

stress, as in the slurs in the Kyrie theme of the B Minor Mass.^ The articula-
tion of that piece is tied in with the appoggiatura^ on the offbeat. It rubs
against the meter — it's a metrical and melodic dissonance although not strictly

a harmonic dissonance.

Largo

Your book mentions some other slurs and dots that indicate deviations from
the basic grammar of "good" and "bad" beats.^°

It when the articulation marks themselves are inconsis-

also suggests that
tent, that may be when the hierarchy of good/bad beats is in routine operation.

Not all Bach performances, especially mainstream ones, observe that hier-
archy of beats —a lot still play Bach as Proust's "divine sewing machine, " with
all the beats more or less equal in importance. Sometimes even early-music
groups do that. Other early-music groups have gone to the opposite extreme,

and have been attacked for bashing the downbeats.

The divine sewing machine was characteristic of the 1950s; it was a Stravin-
sky- and Hindemith-influenced approach. As for downbeat-bashing, you find
it for example in some of the Teldec IHarnoncourt/Leonhardt] cantatas, in the

chorales, in the manner of attacking the words. So even within the early-music
movement, you can get the whole range, from the totally undifferentiated
ticker-tape approach to the downbeat for children's dance class.
The big problem of downbeat-bashing is the view that the barline is sacro-
sanct, when quite often it's just a notational aid, and in fact the grouping might

8. Bach, Mass in B Minor, BWV 232: subject of Kyrie I.

9. An an ornamental note that leads melodically into the main note that
appoggiatura is

follows, usually from a half-step or whole step away. In this example, the first note under
each two-note slur is an appoggiatura.
10. E.g., Bach Interpretation, pp. 179 and 176; the latter concludes: "players unless oth- —
—
erwise informed geared their articulation to the natural hierarchy of pulse, and to their
knowledge of harmony and figuration."
 CONSISTENT INCONSISTENCIES 181

well go beyond that bar. In this regard, dance patterns can be of crucial impor-
tance, even in pieces which are not specifically labeled as dances." The other
point regarding the grammar, this metrical hierarchy, is that it's much more im-
portant that you have the idea of it than that you play it out rigorously. By anal-
ogy, we manipulate grammar rhetorically when we're speaking. Sometimes we
use standard grammar more or less unconsciously, but sometimes we follow the
strictest grammatical rules conspicuously for a specific effect, and at other

times — for example, when we're being poetic—we might flout the rules.

You mentioned Bach's syntax as the next level in the hierarchy; could you
define it?

You might think of the way a musical phrase is structured as being the syn-
tax of the music. "Syntax" is a question of how Bach ties all the various de-
vices — the figures and so on —together. It's concerned with what's appropriate
at any particular point of the musical sentence, as it were. By contrast, "gram-
mar" helps us to know how various musical "words" fit into a background
metrical scheme —whether you stress this note or that note.

While syntax deals with putting those words together into sentences.
Right. And with syntax you can blend together several issues at once: the
question of metrical and harmonic underpinnings (the grammar), the question
of phrasing, the "rhyming" of various motivic figures with one another, and so
on. So it's a complicated issue, but it's crucial to how the music works.
It relates to the inconsistencies of articulation marks as well. For example,

Bach's syntax often involves the interaction of several lines; their slurs might
well agree on the strong part of the beat and differ on the weak parts. '^ The
articulation marks of the different lines in these cases may just be meant to cre-

ate resonance and underline metrical accents —to remind the players that there
should be three accents per bar, let's say — rather than to enforce a particular

articulation. (If these lines are involved in a subtle interplay of contrasting fig-

uration, though, the slurs might be meant more exactly. And we could also dis-
cuss the "melodies" that emerge from the interaction of multiple lines.)
Also, the role of a motivic figure is often ambiguous —does it relate to what
preceded it or to what follows, or both? That influences the player's articula-
tion. Now, these ambiguous notes often come at the weak part of the bar. In
other words, you could look at any weak note in Bach, whether on a weak
beat or a weak subdivision of a beat, and ask yourself whether it prepares the

11. Butt discusses this in chap. 6 of his Cambridge Handbook Bach: Mass in B Minor
(Cambridge University Press, 1991). Some examples are the Gloria in excelsis deo and the Os-
anna, which use a two-bar rhythmic grouping derived from a pair of similar dance types, the
Gigue and Passepied. For more extensive discussion see Meredith Little and Natalie Jenne,
Dance and the Music of J. S. Bach (Bloomington: Indiana University Press, 1991).
12. Bach Interpretation, p. 132.
182 THE BAROQUE

way for the next strong beat or is the dying breath of the strong beat before
it. Or you might play it both ways. In the Second Organ Sonata, the figure (a)

could relate to the previous downbeat, or to the following one:^^

And, as I said, I like to have it both ways, if possible. To be able to point to

that ambiguity in the way the music emerges within notation and bring it out
in performance is a very great challenge. Here I'm talking as a performer rather

than as a listener. I try to preserve the ambiguity when I play such figures.

The next you mentioned are those of rhetoric and emotion. You've
levels

been talking about music by analogy with language in terms of grammar and
syntax, but Baroque theorists often talked about music-as-language in terms of
Latin rhetoric and oration. How does that relate to understanding Bach's cre-
ative process and his articulation marks?
First, I should say that I think the connection between rhetoric and music
is often overstated. Some modern writers believe that a study of rhetoric en-
ables them to discern specific meaning in the music — holding, for example, that
the various musical figures each had a specific meaning'"* —but rhetoric a very is

different field from semantics. It power of persuasion, and the

deals with the
nuances of figurative speech. Most people, including Bach, had been taught
rhetoric at school'^ (though not necessarily to a very sophisticated level), so the
idea of how to construct a persuasive speech would have been clearly in their
consciousness —the basic building-blocks of a speech, the order of the argu-
ments, and their elaboration. Moreover, they clearly would have understood the
different levels of rhetorical delivery: inventio, the invention of the basic idea;
dispositio, laying out the idea; elaboratio, filling in or "elaborating" the laid-
out idea; pronuntiatio and enunciatio, the actual performing of that idea.'^

Its "pronunciation" and "enunciation"? Where do the articulation marks

fit into this sequence of stages in making a persuasive speech?

13. Bach, Trio Sonata No. 2 in C Minor, BWV 526: opening.

14. Butt writes, in Bach: Mass in B Minor, "Most figures, like words, create different af-
fekts in different contexts, so it is certainly a mistake to interpret them as fixed tokens of
meaning"(p. 85). This flexibility, he notes, accounts for the ease with which Bach could reuse
his own music with different texts.
15.One of Butt's books studies musical education in German Baroque schools: Musical
Education and the Art of Performance in the German Baroque (Cambridge University Press,
1994).
16. See Bach Interpretation, pp. 15-19, for detailed source references on how these terms,
derived from Cicero, were used in the German Baroque. On the other hand, as Butt points
out, this can be overstated: see Peter Williams, The Organ Music of J. S. Bach, vol. Ill (Cam-
bridge University Press, 1984), pp. 69-72, and Joshua Rifkin's interview later in this book.
 CONSISTENT INCONSISTENCIES 183

Articulation marks deal primarily with the levels having to do with perfor-
mance, how you "enunciate" and "pronounce." But they also relate to the pre-
vious level, called decoratio^ decoration. Decoration is part of the way you
"elaborate" the basic "inventions," which in music are more like musical
themes or ideas for the potential development of the entire piece. Decoration
is taking that elaborated form of the invention and, in music, adding little mo-
tivic figures to it, the little ornamental figures that help convey the work's spe-
cific mood or character. These decorations were, by the way, one of the things
that distinguished Baroque music from Renaissance music, which at least on
paper allowed only a few "primary" figures
— "exceptions" that eventually, in
the Baroque era, became a new "rule."

So the decoration is the detailing of the elaborated idea, and it's with this
level that the articulation marks in Bach primarily deal.

Right. Now, in music decoration is the level (in terms of ornaments and
diminution) which is often added by the composer but can also be added by
the performer. And that's why it interests me — you can think of decoration as
being the hinge between what's notated and what's performed.

You note in your book that Bach's articulation markings are concerned with
bringing out the roles of the decorations, the motivic figures, in the musical ar-
gument, and that this differs from later eras' markings, which have to do more
with indicating longer phrases. You also said that if one articulates the deco-
rations in the detailed way Bach's markings suggest, it clarifies his contrapun-
tal textures.

That's right. I would emphasize that while articulation marks belong to the

same mode of thought as the decorative level of the music, they don't always
simply deliver these motivic figures — indeed they might sometimes contradict a

simple motivic analysis.

In fact, the role of figures in the musical argument is changeable, and this
bears on the question of inconsistencies in articulation marks. For example,
sometimes changes in slur markings are meant to highlight a moment when a

figure begins to play a different role in the musical design. At one point in the

first movement of the A Minor Violin Concerto,'^ a three-note figure (a) is

slurred four times, but not the fifth time (a'): The dropping of the slur at this
fifth occurrence disguises the way that the three-note figure is suddenly incor-
porated into a long sequence."* In that sequence the three-note slur then be-

17. Bach, Violin Concerto in A Minor, BWV 1041: first movement, bars 24-30.
18. Butt, Bach Interpretation, p. 197-99.
184 THE BAROQUE

comes a musical element in its own right — it is transferred to the first beat of
the bar over a different figure (b).

This reminds me of a review you wrote, where you contrasted Leonhardt's

Bach playing— which a given motif may be played
in slightly differently in dif-

ferent contexts — with that of a harpsichordist for whom each note in a Bach
theme or motif "is definitively legato, staccato or midway. "^^ So it seems that
with Bach the decoration interacts dynamically with the structure.
To understand this, I would suggest that in Bach —to an extent that's un-

usual for his era —the basic idea or inventio of any particular piece is played
out in all the levels right up to the figures, the decorations. For example, quite
often the little figure might be the subject of the piece of music, as in the Two-
Part Inventions. In No. 1, for instance, it becomes a subject that is discussed
and developed in the course of twenty-two measures.^" In fact, Robert Mar-
shall finds that Bach usually began a piece by composing the essential figures

in the primary melodic voices.

So one has to think on several levels —not just in a sequential line from in-

vention to performance, but backwards and forwards and sideways. The small-
est element of decoration might in fact be the seed to the invention, and so on.

This connecting of the different levels in Bach —where the seed idea is also
the surface ornament —does this account for your statement that although it's

very hard to play Bach well, something of Bach always comes through if the
notes are all there?

Yes. In some ways I like to think of Bach's music as being a notated form
of performance —that he's actually recording in notation almost what he did as
a performer. And, yes, that's why even if you play the notes of Bach's music
without any interpretative intent at all, you're bringing out a performance of
the music, because the performance is already there in the notation itself. It's

already ornamenting itself; there's already an aspect of what the performer

might have done.
Indeed, it's always assumed nowadays that in the Baroque era the performer
played the ornaments, and the composer wrote the "real" music. But with
Bach, and with many Baroque composers for that matter, the boundary be-
tween notated ornamentation and improvised ornamentation is hard to define.
From that point of view, the articulation markings extend the idea that the
performance is in the notation. The articulation markings bring out more and

19. Butt, Early Music 17 (February 1989), p. 117. In this review of Davitt Moroney's Mu-
sical Offering recording, Butt does note that the approach taken by Moroney may be appro-
priate for Bach's late works, which were written when the general style of keyboard articula-
tion was moving in this direction. This relates to Butt's comment that in Leonhardt's Bach
playing the hierarchy (discussed in this chapter) is always clear.

20. "[Baroque] music |was] understood as a complex discussion of motives and figures
very much like classical rhetoric." Butt, Bach: Mass in B Minor, p. 84.
 CONSISTENT INCONSISTENCIES 185

more of the implications of figures and ornamental patterns of the music. The
articulation is actually a part of the interpretation, which is why I called my
book on articulation marks Bach Interpretatiori.

Earlier you related dissonance to grammar which notes one stresses in a

bar. But now that you're discussing the rhetorical element, where does disso-
nance fit inf
Many dissonances are just part of the substance of the music —what moti-
vates it — and you don't notice them particularly. But others give you a specific
emotional effect. I think the Kyrie I theme in the B Minor Mass is a good ex-
ample — its character comes partly from its dotted rhythm, but more impor-
tantly from the melodic dissonance. It's the progression of various kinds of dis-
sonance and the implication of dissonances that give it its character, its

emotion. If you stripped it of that and made it a continual line, it wouldn't

have any character at all. In general, you might think of particular notes that
stick out —stand out as dissonances or for other reasons —as being rhetorical.
They're "decoration" as well as " invention" f The Kyrie theme relates to a
point in your book, where you say that some articulation marks have to do
with bringing out the emotion, the affect.

In these marks, a particular slurring may be associated with a particular af-

fective figure and therefore brings it out (as in the Kyrie theme), or there is the
idea of a slurred affect, where everything is somewhat slurred to give a partic-
ular mood (as in the St. Matthew example).

That reminds me of your statement that the dissonances are what give this

music its power.

Yes. To bring in metaphysical issues, which are common in the writings of
the time, there was the idea of introducing dissonance as a way of appreciat-
ing consonance. You must have the opposites, you must have the yin to ap-
preciate the yang, you can't have one without the other. If it's just a continu-

ous consonance, you don't know how "good" consonance is.

You can relate this to religion. You have to appreciate evil in order to un-
derstand good. I don't think the Baroque use of the terms "good" notes and
"bad" notes is casual thinking: there's an analogy here that they were conscious
of at the time. Bach's cousin J. G. Walther, who was a theorist, said, "You must
think of dissonance as night and of consonance. as day." And so on: the inter-
play of the two concepts.

On How Bach's Metaphysics Influenced His Music

The understanding of Bach's religiousness has gone through a big shift since
1 950, from the idea of a "devout Lutheran, his art and life wholly directed to-
186 THE BAROQUE

wards the improvement of church music," to a more confusing picture.^^ Still,

his religious faith would seem to be an entryway into his creative process. What

would you make of the connection between the two?

The traditional way of looking at this, particularly by those who have a re-
ligious ax to grind, is to find religious symbolism and all sorts of levels of piety
in what Bach does. Some of these findings seem too convincing to be fortu-

itous —
the emphasis on "threeness" in Clavieriibung III, for instance, or the ten
entries for the Ten-Commandments fugue^^ —
but of course it's in the nature of
Baroque music to paint or gloss on a text, so from that point of view there's
not a tremendous difference between Bach and many of his contemporaries
(and not just those in the sacred field). I think it's more or less a dead end to
try to make Bach's music a medium for hidden esoteric messages, merely on ac-

count of his superlative quality as a composer and his evident piety.

In your book on the B Minor Mass you dismiss attempts to apply nu-
merology, Old Testament prophecies, and the like to the composition of the
Mass; you say that "there is at once no supporting evidence for verification [of

this approach] and, on the other hand, no possibility of refutation. " But is there

a fruitful way to use Bach's religion to understand his creative process

One approach towards answering this question is to get to the main reli-

gious dispute of his age, which was between the Pietists and the Orthodox
Lutherans. It has become quite clear, perhaps only in the last ten years or so,

that Bach was very definitely an Orthodox Lutheran, not a Pietistic one. That
doesn't mean that he lacked piety, but he wasn't part of the movement of
Pietism with a capital "P," which looked towards a very direct relationship with
God, one which didn't require the Church —congregations, buildings, altars,

and such things —as intermediary. They wanted a personal relationship that
could be expressed in the language and discourse of the early Christians. In
many ways, it's like a modern, homespun Evangelical movement, where the
great emphasis is on worship at home, with the singing of sacred songs. From
that point of view, the Pietists shunned the whole apparatus of the church year
and church music.
It's quite clear that throughout Bach's career he was escaping the Pietists.
His second post, at Muhlhausen, fell through in a couple of years because of
what was very clearly the Pietistic attitude of certain authorities, which didn't
allow complex church music. Now, you can articulate this issue from two an-
gles. First of all, you could say, "Bach liked to write great music, and therefore

21. Malcolm Boyd, Bach, The Master Musicians (London: Dent, 1983, and New York:
Viking, 1987), p. xiii, in explaining the ongoing Bach revolution.
22. Fughetta on "Diess sind die heil'gen zehn Gebot'" ("These Are the Holy Ten Com-
mandments"), BWV 679. The "threeness" of Clavieriibung III (BWV 552, 669-89, and
802-5), which represents the Holy Trinity, is evident in (among other things) the collection's
having three settings each of the Kyrie, Gloria, catechism, and sacrament movements.
 CONSISTENT INCONSISTENCIES 187

sided with the —

Orthodox" that Bach was really only interested in music, and
therefore chose the wing of Lutheranism that favored music. But I suspect it
was the other way around: that Bach was an Orthodox Lutheran and therefore
believed in the order of the church year and the order of the church ceremony,
the history of the church ceremony and liturgy, and the indisputable history
and place of music within that liturgy as part of the mechanics of religious
faith.

This brings us beyond his theology to his metaphysics. Bach sees music as
being part of a mechanical process by which humankind comes to terms with
the divine. And that mechanical process involves aspects of liturgy, going
through all four Gospels day after day, week after week in a specific order, cov-
ering the Bible and the Psalms in an ordered way throughout the course of the
liturgical year. It's organization, what you might think of as cultured religion,

as opposed to personal and immediate religion, religion that's based on one's

immediate reactions, feelings, and notions of faith. So from this point of view
there's an aspect of his music that falls into a larger picture of what a religion

is concerned with.
So Bach's beliefs might have influenced the way he approached the task of
composition more abstract sense than that of theology. What he might
in a

communicate more in his music is his sense of order, which is in fact more a
metaphysical concept than a religious one. His type of faith is one that looks
for Godly order on earth. It's not a million miles from Pantheism in some ways,
but it would never have been articulated thus in his time.

Could you say more about this metaphysics?

Well, one thing that interests me is the natural philosophy of the time.
Whether or not Bach was familiar with the work of, say, Leibniz, I think there's
a strong connection between that naturalistic philosophy of Bach's age and the
way Bach's music works. Leibniz sees the smallest substance in the universe as
being a microcosm of the greatest —the concept of the windowless monad. You
could infer the whole of creation from a single monad and versa — he
vice in-

fers monads from looking at the whole of creation as he knows it.^^ I myself
am particularly fond of linking Bach with the thought of Spinoza — not that this

is plausible historically. But Spinoza seems to come closest in describing the

"one substance" of Bach's music, and the "immanentist sacrality" of music

23. Bach's analogies with Leibniz were discussed in a 1963 paper by Edward Lewinsky,
"Music and the History of Ideas" (reprinted in his Music in the Culture of the Renaissance
[University of Chicago Press, 1989], pp. 67-86). Lowinsky footnotes earlier discussions of the
relationship, and points out that Bach's student and advocate Lorenz Mizler studied with
Christian Wolf, a translator of and authority on Leibniz. Regarding Butt's idea that there are
even deeper analogies between Bach's metaphysics and Spinoza, see his "'A Mind Unconscious
that It Is Calculating'? Bach and the Rationalist Philosophy of Wolff, Leibniz and Spinoza,"
in The Cambridge Companion to Bach, ed. Butt (Cambridge University Press, 1997), pp.
60-71.
188 THE BAROQUE

the idea that Bach expressed in his jotting in the margin of his Bible, "Where
there is devotional music, God is always present with His grace." Spinoza's
term "the intellectual love of God" seems remarkably appropriate for Bach.
There are other historical streams feeding into Bach's metaphysics; you
might even think of the Lutheran sense of the mystic function of music as being
a late manifestation of medieval thought. The conception of music as mirror-

ing God's universe, and having mathematical proportions and mysterious as-

pects that mirror the human soul, is in some ways a medieval conception all

along.^"* On humanism brings in classical texts in which the

the other hand,
power of more human-based than God-based. So in some ways the phe-
art is

nomenon in Bach and the late Baroque is a sort of combination of these two.
The God-based medieval conception and the more human-based Renaissance
conception melded together in what we think of as Orthodox Lutheranism.
Pietism is a more progressive movement; it's actually pointing more towards
Enlightenment types of thought, where the individual is responsible for his own
salvation, his own faith. It's more immediately put down to the personal, the

individual.

Regarding the Pietist movement, wasn't Bach criticized in his own lifetime

for the over-complexity of his writing?

Oh, yes, by Scheibe in some ways speaking for a later
1737. Scheibe is in

age, the early Enlightenment age, in melody and of affect,

which simplicity, of

is thought to be more natural than complexity in music. And, as you said, here

is another relation to the Pietistic movement. Scheibe was not a Pietist as such,

but the artistic side of the Pietistic movement (if you can think of it as having
an artistic side) was very much that of mid-eighteenth-century mainstream
music, which favored lightness, simplicity, and directness.

Like American classical radio today. Or Rousseau, a little later . . .

... the same sort of thought. Anything that apparently confuses and makes
complex was thought of as being bombastic. Indeed, one of the things which
Scheibe criticizes is the way Bach notates every little ornament. He puts down
every little figure that the singer would normally sing completely naturally him-
self, and much more gracefully than this ghastly fixed notation of Bach allows.
That points to my view that Bach wanted a lot of control over the perfor-
mance, and that part of that control was gained by writing down ornamenta-
tion. In his mind, this made the music that much more complex and cohesive,
because it always relates to other aspects of the structure which an improvising
performer would surely miss. It's a particular conception, not unlike that of

24. Lewinsky points out that Bach's student J. P. Kirnberger was still writing about the
medieval classification of musica mundana, humana, and instrumentalis (see Barbara Thorn-
ton's interview for a discussion of these ideas in Hildegard of Bingen). Lowinsky also men-
tions Mizler's advocacy of Pythagorean principles.
 CONSISTENT INCONSISTENCIES 189

twentieth-century modernists such as Stravinsky or Schoenberg. Both of those

composers have a very similar view of control over the performer, and control
of the insignificant; the insignificant is to be as significant as the significant.

That seems to indicate a limit to improvisation, namely that you can't get

perfect integration of the parts and the whole (though, of course, not all music
aims for that).

That's right: there's going to be a sense of chance, or a sense of diffusion.

Bach in some ways has more nineteenth-century aesthetic, that the music
a

must form its own coherent whole. That means, on the other hand, that as a
performer you must play the piece as if for the first time, as if you were just
discovering it. It doesn't mean you take away the spontaneity. You can keep
the spontaneity and have the sense of the coherent whole, which is the best of
both worlds.

And Bach's concern for coherent wholeness derived from the metaphysics
you described.
Very certainly, yes. He had an attitude that everything human, natural, and
musical existed in a neatly ordered hierarchy. This included even his patrons
and the aristocracy — he felt that the domination of the upper classes was
strictly analogous to the domination of God. It's hard, particularly for religious
people, to accept today that Bach must have believed in the divine right of the
aristocracy as much as he believed in the greater divine right of God. But the
rv\'0 do belong together.
That concern for coherent wholeness also may account for why Bach has
been so fully accepted into the canonical mainstream of music. He had no in-
tention of being up there with Beethoven, Mozart, and Co.; he couldn't con-
ceive of that sort of thing in his day. But it's no surprise that he's the one com-
poser from that era who's been taken into common practice. ^^

Because his contemporaries' music is less coherent, and coherence is what

we value now.
In other words, in Bach's time it was not so important to write music that
was really tightly organized.

So his pre-Enlightenment metaphysics helped bring him into the post-En-

25. This is a nice rwist on Lydia Goehr's argument that Bach did not conceive of his pieces
as "works of music," which she considers a nineteenth-century concept (though she does not
deny that his pieces can legitimately be considered "works"). See her essay "Being True to the
Work," Journal of Aesthetics and Art Criticism 47:1 (Winter 1989), esp. pp. 56, 57, and 61.
Butt's next remark does seem to concur with her belief that in Bach's era, musical practice
"was not regulated by the idea of the work." Recall also Butt's thoughts on what constitutes
a "piece of music," in the early pages of this interview.
190 THE BAROQUE

lightenment canon. Richard Taruskin^'' points out the irony that twentieth-cen-
tury composers like Stravinsky have taken Bach as the standard for treating
music as a purely formal arrangement of sound, of pure form, abstracted away
—
from the rest of life his music is so perfectly wrought that he can be taken as
"the formalist supreme" —when Bach music was not that at all.
to
In some sense, Bach was something even more extreme than a formahst;
Bach more or less saw music as reflecting the whole substance of everything
around him: substance and form, actually. And he very much believed that
music has value in its own way. His underlinings in his copy of the Bible are
all under texts that mention music as being crucial to God's creation, to show-
ing God's work.

In your notes to your CD of the Trio Sonatas, you say that Bach's "reli-

gious faith probably led him to believe that the musical language of his time
was divinely developed and fixed; all he needed to do was to understand its

deepest implications to discover the unity behind the diversity, but also to cre-
ate something fresh and unique each time he composed." So that mixing of
Baroque genres we talked about —the way he melds so many conven- different
tional genres into unified forms, as Cantata 78 — was motivated by
in his reli-

gion.
I suspect it was the religion, not the metaphysics, that he thought about, be-
cause that's what you were supposed to think about; the metaphysics was a
given. In a thoroughbass method he wrote down for his students in 1738, Bach
said, "The end or final cause of all music, and also of thoroughbass, is the glory

of God and the permissible enjoyment of the spirit. Wherever this is disre-
garded, there is no longer actual music but a devilish bawling and singsong."
In other words, following the rules of music as he knew it was almost like fol-
lowing an ethical rule, a cosmological rule.

I would repeat, though, that it's a mistake to reduce the music's quality and
effect to the religious values he held, as some people have tried to do. After all,

there are religious composers who don't have that kind of effect, and some non-
religious composers, like Debussy, whose music is highly ordered. I think a lot
of writings on Bach miss the fact that his religious faith rests upon certain meta-
physical premises, of order, of connection, of thoroughness —that his faith rests

upon those premises, rather than supporting them.

SELECTED DISCOGRAPHY
John Butt's research into articulation tells throughout his recording of the Bach
Trio Sonatas for organ (Harmonia Mundi HMU 907055), whose articulation

26. Taruskin, "Facing Up, Finally, to Bach's Dark Vision," in The New York Times, Sun-
day Arts and Leisure section, 27 January 1991, pp. 25, 28, reprinted in his Text and Act (New
York: Oxford University Press, 1995), pp. 307-15.
 CONSISTENT INCONSISTENCIES 191

marks are especially detailed. Reviewing the CD, David Mulbury praises Butt's

"splendid instinct" for the "most problematic of instruments," the organ, and
says that "a strong musical personality permeates his music making."'^ Mul-
bury thinks his articulation works well in the "buoyant" fast movements, but
is disturbed by the "rhythmic Quixotism" in some slow movements; this refers

to Butt's use of rubato, which I think is eloquent.

Butt has also recorded music that may have influenced the young Bach. This
includes Hexachordum
Pachelbel's Apollinis (on organ; Harmonia Mundi
HMU 907029), and two recordings of music by Kuhnau, Bach's predecessor at
the Thomaskirche: the Frische Clavier Friichte (on harpsichord; HMU 907097)
and the Biblical Sonatas (on harpsichord, organ, and clavichord; HMU
907133). Nicholas Anderson says, "Butt's Kuhnau playing is fluent, rhetorical
and virtuosic and he makes more sense of these extraordinary, often theatrical

pieces [the Biblical Sonatas] than I have previously experienced."'** Butt's

recording of the organ composer Cabanilles (HMU 907047) appears regularly
on lists of the best CDs of Spanish Baroque music. As for his recording of the
complete organ music of Purcell and Blow (HMU 907103), Marc Rochester
likes it but finds it "somewhat remote";^** by contrast, the organist Haig
Mardirosian praises Butt's ornamentation, rhythm, and "verve and grace" in

these works. He concludes, "Butt plays Bach and earlier organ music with un-
"^°
challenged expertise.
With the violinist Elizabeth Blumenstock, Butt has also recorded Bach's
complete violin and harpsichord sonatas (HMU 907084). Reviewing the set,

Anderson calls Butt's playing "impressive" and praises Blumenstock's "invigo-

rating, perceptive, and often very sensitive" playing, which "reaches the heart
"^'
of the music.

FOR FURTHER READING

Butt's Bach: Mass in B Minor, a Cambridge Handbook (Cambridge University
Press, 1991) is an excellent general introduction to Bach's formal techniques
and procedures, especially in the vocal works. Donald Francis Tovey's 1937
essay on the Mass in B Minor is also a superb introduction to the ritornello in

Bach; it is reprinted in his Concertos and Choral Works (Oxford University

Press, 1989). Butt's Bach Interpretation (Cambridge University Press, 1990) is

important, even essential, for performers and scholars. His Musical Education

and the Art of Performance in the German Baroque (Cambridge University

Press, 1994), an important work for specialists, sheds light on the one-per-part

27. Mulbury, American Record Guide 55 (May/June 1992), p. 22-23.

28. Anderson, Gramophone 73 (March 1996), p. 71.
29. Rochester, Gramophone 72 (June 1994), p. 87.
30. Mardirosian, Fanfare 17 (March/April 1994), p. 282-83.
31. Anderson, Gramophone 71 (October 1993), p. 60.
192 THE BAROQUE

debate (see Chapter 15 below) among other performance-practice issues. Mal-

colm Boyd's Bach in the Master Musicians series (London: Dent, 1983, and
New York: Viking, 1987) is the best introductory biography at present. Robert
Marshall's The Music of Johann Sebastian Bach (New York: Schirmer, 1989)
is illuminating. Best of all, perhaps, is Christoph Wolff's Bach: Essays on His
Life and Music (Cambridge, Mass.: Harvard University Press, 1991). Regard-
ing Bach's metaphysics, see Robert Marshall's essay "On Bach's Universality"
— —
book, and Butt's two essays chaps. 4 and 5 in a book he edited, The
in his

Cambridge Companion to Bach (Cambridge University Press, 1997).

 10
"
''One Snoula Not Make a Rule
— oOo^ g^

Gustav Leonhardt on Baroque

Keyboard Playing

In the 1950s and 1960s, a group of Dutch musicians created one of the world
centers of what we now call the early-music movement. Through the work of
among many —the Kuijken brothers, the recorder virtuoso Frans Brueggen, and
my interview subjects Gustav Leonhardt and Anner Bylsma, the Netherlands
became to Baroque performance what Switzerland is to chocolate, watches, and
banks. Just as young musicians had long flocked to the Schola Cantorum in
Basel for instruction, they now came from all over the world to Amsterdam.
Leonhardt and his associates raised their instrumental technique to new
heights. More significantly, they developed a new approach to playing Baroque
music. In contrast to the motor-like "sewing machine" style prevalent in pre-

ceding decades, these players emphasized the metrical hierarchy that John Butt
explains in the last chapter; and their playing, says Laurence Dreyfus, sounded
"strikingly speech-like by mimicking ever-shifting patterns of thought."' Their
approach has had enormous influence, either by being adopted elsewhere or by
being reacted against.
In the next two chapters, Leonhardt and Bylsma speak about their approach
to performance and to the exploration of early instruments.

"The harpsichord is perfect as to its compass and is brilliant in itself," wrote

the great French composer Francois Couperin in 1713, "but as it is impossible
to swell or diminish its tones, I shall always be thankful to those who, by means

1. Dreyfus, in "The Early Music Debate," Journal of Mustcology 10 (Winter 1992), p. 115.

193
194 THE BAROQUE

of infinite artistry borne up by good taste, shall succeed in making the instru-
ment capable of expression."^ He might have been especially grateful to Gus-

tav Leonhardt.
Harpsichord playing in the middle of our century (in reaction against the
pioneering romantic, Wanda Landowska) was often rigid and prickly. My pi-

anist friends still occasionally say things like, "But the harpsichord can only
play staccato, right?" More than anyone else, Leonhardt has developed a bat-
tery of techniques that allow the harpsichord to speak and sing, to create an
illusion of rich variety in both sound and touch. How he did this was one of
the topics we discussed when I telephoned him at his seventeenth-century Am-
sterdam
—
house "a dwelling," says Howard Schott, "filled with beautiful old

furnishings and fine instruments."^

Leonhardt, born in Amsterdam in 1928, grew up playing piano and cello,

and war he studied harpsichord and organ at the Schola Cantorum in

after the

Basel. He returned to his native city in 1954 to take up a professorship at the

Conservatory. Since then he has played a key part in the extraordinary devel-
opment of Dutch Baroque performance. He has been, beyond question, the

most influential harpsichordist of our time. Name a leading harpsichordist, and

the chances are good that they studied with Leonhardt. Among his pupils are

Bob van Asperen, Goode Crawford, Alan Curtis, John Gibbons, Pierre
Lisa
Hantai, Ketil Haugsand, Christopher Hogwood, Ton Koopman, Edward Par-
mentier. Skip Sempe, Colin Tilney, Anneke Uittenbosch, Glen Wilson, and many
others. The French harpsichordist Christophe Rousset says that it was a mas-
ter class with Leonhardt that liberated him at the keyboard:

I learned all I know about harpsichord technique in Holland. They know

everything about the harpsichord there. It's amazing how Gustav Leonhardt
has thought about everything, every little reaction of the instrument, the ac-

tion of the keyboard and the plectrum and the string. . . . after I took a
master class with Leonhardt, I finally understood . . . what is possible on
a harpsichord and how to make it sound. ''

Some harpsichordists become a bit lost in the "infinite artistry" —the subtle
techniques that Leonhardt pioneered. Alfred Brendel writes that "nowadays we
hear Couperin on the harpsichord played in a way that amazingly resembles
the 'romanticism' of Paderewski's records: no chord without an arpeggio and

2. "Le Clavecin est parfait quant a son etendiie, et brilliant par luy meme; mais, comma
on ne peut enfler ny diminuer ses sons, je s^auray toujours gre a ceux qui, par un art infini

soutenu par le goit, pouront ariver a rendre cet instrument susceptible d'expression."
Couperin, Preface to the Premier Livre of the Pieces de clavecin.
3. In an interesting tribute to Leonhardt, "Ein vollkommener Musik-meister," Musical

Times 133 (October 1992), pp. 514-16.

4. Bernard D. Sherman, "Finding One's Own Recipes: Christophe Rousset Ponders the In-

gredients," Piano and Keyboard, May/June 1994, p. 29.

 ONE SHOULD NOT MAKE A RULE 195

the left hand constantly anticipating the right."' In fact, a "straighter" school

of period-instruments Baroque playing, which emerged in London in the 1970s,

developed in conscious reaction against what its players considered the "highly
mannered style" of Dutch (and Austrian) Baroque playing. In our interview,
Leonhardt discusses how he integrates his subtleties into a natural whole.
Leonhardt clearly feels uncomfortable with the roles of eminence grise,

doyen, or guru; "I don't regard myself as a pioneer," he once told a German in-
terviewer.*" It is equally characteristic, and a tribute to his teaching, that many
of his students have distinctive styles of their own. Rousset, for example, does-
n't play in what he would call a "Dutch" interpretative style, as described above
by Dreyfus; as he told me, "If Leonhardt does something, it's not a recipe. It's

good for him; it's convincing, but as a system of interpretation which works
every time, I don't believe in it." Neither, as you will read, does Leonhardt.

In the Baroque, three keyboard instruments —the harpsichord, the clavichord,

and the portative organ — were in common domestic use. To what degree were
they interchangeable, or to what degree specific in a composer's mind?
In general, it's difficult to say; but I think they were largely overlapping, and
the composers couldn't care less. Someone at home would use indifferently
whatever was practical (of course, the large organ is a different matter). It's rare

to find specifications that make it dead certain that the composer wanted a
piece played on this or that instrument. Sometimes we think that we know that
this must be an organ piece, and that a harpsichord piece, and it cannot be
anything else; but usually we cannot prove it. And I think it's rather good that
we cannot prove it, because it should remain rather flexible.

Regarding Charles Rosen argues that few of Bach's

this inter changeability,

fugues "exploit the of any particular instrument" harpsichord,

resources —
clavichord, organ, or piano; most were meant to be playable on whatever in-
strument one had at homeJ If, as he says, "the type of sonority . . . is rarely

a matter of interest," why not use the piano

I don't agree at all with the conclusion. I think the historical facts sur-
rounding the composer are much more important than that argument suggests.
It's a unity; you can't break it up, although there's a lot of liberty within it.

Of course, one should not say the harpsichord is always the thing to use,
because so often it might be a small organ or a clavichord. But they, unlike the
piano, all belong to a certain way of thinking.

5. Alfred Brendel, Music Sounded Out (London: Robson, 1990, and New York: Farrar,
Straus and Giroux, 1991), p. 221.
6. Arnd Richter, "Ich fordere nichts vom Publikum," Neue Zeitschrift fiir Musik 147/9
(1986), pp. 34-38; quote, p. 34.
7. Bach: The Fugue, ed. Rosen (Oxford University Press, 1975), Introduction, p. 3.
196 THE BAROQUE

What defines that way of thinking?

Well, it also depends on the particular pieces, but most pieces of the seven-
teenth and eighteenth centuries seem to want to speak instead of sing —though
this is too simplistic a statement —whereas in the nineteenth century one thinks
primarily about singing in long, sustained phrases.
I thmk the nineteenth century, to put it roughly, is for sustained sounds,

which are always under tension and always nourished; but I think before that
it was exactly the contrary, it was more like speaking, which means wave-like,

constantly rising then loosening up even within a single sentence. In this music,
you push it from a "good" beat,** and then the following group of notes goes
by itself. Unlike later music, you don't have to push a phrase all the time; it

often rolls by itself.

Put differently, there is more attention to the details of the phrase, as op-
posed to projecting a long, sustained line, which is what the modern piano is

designed for.

Pianists will always play Bach; so what would you say to a pianist regard-

ing what, if anything, they can learn from harpsichordists? Some pianists have
tried to use harpsichordistic techniques.
I think such imitations are useless, because the piano only sounds worse if

you try to imitate the harpsichord. The piano has its own ideals and capaci-
ties; you can't mix the two instruments. I don't want to be a policeman, but I

think that's not the right approach.

Harpsichord technique has changed dramatically since World War II, and

you've been very much involved in that. What some of the ways developed
are
since then through which one creates a sense of speaking and dynamics on the
harpsichord?
I cannot say that it's a secret, but it's almost impossible to describe with
words; it's even difficult to show at the keyboard. Essentially, it must be based
on a dynamic wish. The imagination of the player, fed by analysis and by study
of the whole period in which the music was written, must have become very
dynamic. Again, I cannot describe how one puts that on the keyboard, but it

must be based on the musical imagination.

8. A "good beat" was defined in John Butt's discussion of Bach in Chapter 9. What Leon-
hardt is discussing here exemplifies Butt's point that the music of Bach and his contemporaries
has a hierarchy of different levels, so that all notes are not equally important. Here Leonhardt
discusses what Butt calls the "grammar," the hierarchy of strong and weak beats. Regarding
the idea that the music "rolls by itself," Butt again offers some support: in reviewing Leon-
hardt's second recording of the Bach Partitas (reissued on Virgin VER5 61292), Butt mentions
Leonhardt's attention to the "grammar," coupled with his insight into longer-term matters
motivic development, voice-leading, and harmonic tension —and says that together these lead
to "a performance where the music provides its own momentum" (Early Music 17 [February
1989], p. 117).
 ONE SHOULD NOT MAKE A RULE 197

Indeed, I think that the changes made in the last fifty years are based on
the fact that the imagination has changed. Ignoring for a moment the kinds of
instruments the earher harpsichordists were playing, which were not historical
at all, the crucial element behind the wish of some players not to play Bach on
the piano was
wanted to get rid of that dripping Romanticism they
that they
did not like (forgood reasons, we now think). But they then threw out the
baby with the bath water; their approach was only negative, with no expres-
sion, no dynamic levels. They had the idea that the music should "speak for it-
self." So the Helmut Walcha school, the neue Sachlichkeit,"^ put everything at

zero.
In the last fifty years, we have gradually begun to see that Baroque music
is, if more expressive than Romantic music, but in detail rather than
anything,
in large lines. With that, a technique developed, but not by itself; it's only that

the wish has changed, that our imagination of the music has changed com-
pletely.

That's fascinating. Nonetheless I would like to ask a little about the ways
technique has developed: for example, subtleties in how the plectrum strokes
the string.
Yes, through the harpsichord's key we can feel the plectrum touching the
string, so we can rest the plectrum on the string before we pluck, and then
pluck quickly, slowly, overlapping with other notes, or with a range of other
subtleties. The pianist never has this really close contact with the string, be-
cause with the piano it's all indirect. He can of course compensate with dy-
namics and little rhythmic subtleties, but he can never caress the string.

Let me take that technique as an example: how did you learn to do it?

I can't remember that I learned it. Once again, by studying so many fields

of this period, not just music but all of the arts, you see that the Baroque is

the most expressive period we have had in the whole history of Western civi-

lization.^" So I thought the music cannot be dull like a block of concrete, with-
out any life in it, without any undulations. And then, one reads in historical
sources about such techniques as overlegato — in which you release a note later

than it is marked —and other techniques, all used in order to achieve dynamic
subtleties. But again, it's based on the wish. I have never worked on technique
as such. It came by expanding one's understanding of the music

9. The "new objectivity" or "new actuality" —a post-World War I aesthetic movement

that radically rejected nineteenth-century Romanticism.
10. Compare, e.g., Joseph Kerman on the Baroque's "exhaustiveness of emotional effect":
"[high Baroque] music hammers away at a single feeling, intensifying it and magnifying it to
a remarkable extent. . . . The Baroque theater concentrated on grandiose gesture and high
passion. . . . Theatricality is the key to the emotional world of Baroque art, whether in music,
the visual arts, or poetry" {Listen [New York: Worth, 1980], p. 197).
198 THE BAROQUE

/ was going to ask about overlegato: holding some notes longer than writ-
ten. To anyone familiar with older harpsichord playing, the effect was of an al-
most Impressionist wash of sound. It's another tool for making the harpsichord
expressive.
If not overdone.

What determines whether it's overdone — ^5 it a matter of harmonic tension,

or what?
It's very hard to say, and one should not make a rule. It is one of the means.
But again, you use that means in order to achieve a dynamic effect. Now, on
one instrument in one hall you do it a little, and in another you do it a lot, in

order to achieve the same effect. So the circumstances play an enormous role.

Pianists usually find that with the sustaining pedal they can make the sur-
roundings more or less the same; we cannot, so we depend much more on the
hall's having good acoustics, and we must adapt our way of playing very much

to the circumstances of the acoustics. So that changes the technique. You use
different tricks hoping to achieve the same results.

Another crucial element of your style has been the use of timing —delaying
an important note a little, or holding it a little longer than written — to clarify
which beats are strong beats and which are weak. Can you discuss that?
Well, it's not that I don't want to speak, but it's too subtle to explain. Again,
one doesn't make up one's mind, for example, to delay a certain note. Rather,
the wish is to stress that note, and delaying it is one of the means which some-
times one hopes will work. And you can delay to make a note weaker, or to
make it stronger, anything you like; but it's not with the thought of delaying.
When one is a student one does things consciously, but when one is more
experienced one does not play intellectually any more.One doesn't think; one
has thought. You must have done so before, but when you perform it is too
late to think; you are only making music, without any thought of "now delay

here" and "now articulate there." The only thing is music. It is like when we
speak; we don't think, "Now this 'S' must be strong," or "Here let's pause."
Those things are done automatically, depending on what you intend to say.

How does one learn to integrate these things? And can the ability to inte-

grate the parts into a whole be taught?^^

No; I think it comes from two things. The first is a gift. You are born with
and then you must develop it; but either you can develop it or you can't.
it,

The second is probably experience or age, if you like. It takes time, I think. A
young person usually has not that feeling for integration. You see that in all

11. On this topic, I recall something Leonhardt said at a lecture in Berkeley about Fres-
cobaldi; he cautioned against over-emphasizing repeated motifs, quoting Oscar Wilde's last
words (delivered in a characteristic Parisian hotel room): "The wallpaper is killing me."
 ONE SHOULD NOT MAKE A RULE 199

the creative arts, especially the visual arts. I don't know of any great artist who
in his youth was sloppier than in his old age — sloppy not in the negative sense,
but in being more concerned with the whole than the details. Think of Titian
or Rembrandt; they were meticulous when they were young, but rough —con-
cerned only with the major things and forgetting about small details —when
they were old. So I think it's a general tendency in a human being: when you
get older, the whole becomes more important than the details.

I'd like to ask you about some specific performance issues. One is so-called

early fingerings.
The question is, what do we know of good fingerings of important pieces

by important players or composers? Hardly anything remains except by

Francois Couperin. The other fingerings are not necessarily written by the com-
poser, but either by a slow pupil or by a good teacher for a silly little student
learning the piece and doing the thing wrong. So they wrote out, "Here use the
second finger." Now we write a whole book on the second finger being on that
note. So it's dangerous; one must consider it, but it doesn't explain things.

In fact, we see the whole thing in an inverted way. They used a certain fin-

gering in order to achieve a certain effect in the easiest way. Now we go back-
wards; we see a fingering and try to see what effect they meant by it. That is

why it can be dangerous —worth doing, but not to be followed absolutely.

Of course, I've studied early fingerings as much as I could, but the incon-
sistency of early written fingerings is incredible. Even in one piece, with all the

fingerings written by one person, the same motif recurring gets totally differ-

ent fingerings, suggesting totally different articulations. It's a very common

thing. Sometimes, a certain articulation is determined if you use a certain fin-

gering, and It seems to make sense. However, other early fingerings, which a
modern pianist would never use, can indicate articulation, but not necessarily
so. These things are easily exaggerated; one can get pedantic.
Generally, then, I thmk the early fingering is hardly ever known. Even in
Couperin, I'm not sure in my heart that he himself used his own fingerings all

the time. Of course, his fingerings are cleverly thought out and are marvelous
to know; and I probably use them most of the time, because they make sense,
and fall easily for that clear, gentle, clean speaking style of his.

Temperament is another issue. Howard Schott writes that you tune the
harpsichord "as the tonalities and enharmonic notes in [your] programme re-

quire, seeking to preserve as many pure, or at least less than very wide, thirds
as possible. "^^ I take it that you don't feel the need to adhere rigidly to a spe-

cific historical tuning system?

12. Schon, "Ein vollkommener Musik-meister," p. 516. Schott tells me that Leonhardt pi-
oneered the use of historical tunings, and was recording with them in the early 1960s (per-
sonal communication, 1996).
200 THE BAROQUE

Yes, on the harpsichord I my tuning, depending on

adapt the program. The
temperament is unequal, but how unequal and where it is unequal I cannot
standardize. Indeed, as Howard Schott said, the main issue is the major third
the purer it is, the better it is for that music — but inevitably there are conflicts

and practical drawbacks. All the various systems were just attempts to answer
the question, "How do I get around the conflicts? How much can I suffer on
one side in order to enjoy the other side the more?" Anyway, as soon as the
audience comes in, the room heats up and the tuning is already gone a little,

at least the very fine things.

Another issue is registration. A person who hadn't heard an organist play

Bach since 1 950 would notice that some of the current players change regis-
tration much less.

I think the very active registration shifts came into being in the late nine-
teenth century because the organs were so awful, so totally unhistoric, that it

was unbearable to hear most sounds for any length of time. So in a way, it was
a musical thing. But since one has discovered better organs, and I think also
looked at the music very precisely, one sees that many of the registration
changes suggested by Widor or Dupre or whomever are not clean; they break
up one voice in order to make another voice's entry clear, so they butcher the

piece. I think the most important thing is that if you make a manual change it

must be clear in all parts. And then, yes, of course, why not? They had four-
manual organs in the Baroque era.

Can you and where not to change manuals?

give an example of where
Well, it depends on the B Minor Prelude of Bach IBWV
piece. In the big

544] you cannot change manuals properly; more specifically, while you can
get out of the opening registration properly when the opening ritornello ends
and the first episode begins, you never can come back properly from the
episode to the returning ritornello. You have to break a thematic element in
On the other hand, today one often hears the Passacaglia in C
the middle.
Minor [BWV 582] played in a roaring fortissimo from beginning to end; I

think that, first of all, it is unbearable that way and, second, it is contrary
to the character of several of the variations. So one can make a very clean
change of manuals between variations in that piece, which is typical for vari-
ations anyway.'^

Harpsichord registration raises similar issues. For example, what do you

think about varying registration in repeats?

13. See Christoph Wolff's "The Architecture of the Passacaglia," in his Bach: Essays on
His Life and Music (Cambridge, Mass.: Harvard University Press, 1991), pp. 306-16. A de-
tailed discussion of manual changes is in Peter Williams, The Organ Music of J. S. Bach, vol.
3 (Cambridge University Press, 1984), pp. 171-82.
 ONE SHOULD NOT MAKE A RULE 201

Well, I may think something about it, but I don't knoiv anything about it.

To my knowledge, it is never written about in early sources. I think it is not so

important.

You are much more sparing than many harpsichordists in taking repeats —
in many recordings people take them all, while you often take few or some-
times none.
think there are no rules. One does what is required, and a repeat
Again, 1

that may make the piece too long in one setting is just right in another. One
doesn't know the historical practice, really, but think it has always been like I

that. There was no "You must," ever. 1 think it's a bit modern and pedantic
when people think that way. And recording is a rather unhuman thing. I find

that repeats can be pedantic on a recording.

This brings up the issue of recordings. They have been central to the early-
music movement, but of course nothing could be less historical. And yet they
do allow the harpsichord more intimacy than a large hall does. In your expe-
rience of recording, how does it influence the music-making?
Music-making? That I don't want to say. It has influenced enormously the
spread of what I consider good taste in music. This is apart from the fact that
it can offer some bliss to who want to hear
people in the middle of the Sahara
some music; for people in the middle of a big town it's not so necessary. But

the influence of recording has been enormous, and I think beneficial on the
whole. Though that's not what you wanted to know.

/ suppose the question is, in making a recording is there a conflict between

mspiration and technical perfection? Would you let a take go out with wrong
notes because it was inspired or had the right feeling?

For me, a recording is quite a different thing from playing a concert. That's
the reason why I refuse to combine them. Often people, either the radio or

someone else, want to record a concert: I always refuse, because my playing is

totally different from concert to concert. In a concert I adapt my wavelengths
to how is, how far away the audience is, the relationship with
large the hall
that audience at that moment, and, of course, what the acoustics are. And also
it's only for once, so I take risks. Whereas on record, a risk is a silly thing. You

have to do it over if it doesn't go. So I play neatly, and as well as I can on the
record, and in the concert 1 try to play beautifully.

Is something lost from the music in playing neatly?

If you do outrageous things which may perhaps be fine in a concert, they

are ridiculous if you hear them several times in a recording. So I think it should
be a sort of quasi-ivory-tower perfection on record — not too much, not too lit-

tle; exemplary in your own mind. It's a document, really.

202 THE BAROQUE

Regarding the concert experience: Charles Rosen differentiates between

playing an organ fugue written for public performance — in which "the en-
trances of the theme . . . are easily heard and appear with dramatic effect" —
and "private" keyboard fugues written to be played mainly just for oneself. In
these, the entrance of the theme is often hidden, "its opening note tied to the

last note of the previous phrase. " Because it was being played in private, how-

ever, these entries of the theme "needed no illustration or emphasis from the

performer," who could "hear [the theme] himself as he knew where it was, and,
even more, he could feel its presence in his fingers. " But now, when one plays
in a concert hall, it becomes "imperative to allow the listeners to perceive what

[goes] on in the fugue, to give them an idea how the individual voices move.
At its worst, this leads to the pianist who will always bring out the subject to
the detriment of everything else; but for any performer, Rosen says, it creates
a unavoidable tension between modern concert life and "what the composer
wanted or what he expected to get."^^
Well, I think it's a nice idea. It may be a bit of a modern idea to make a
distinction between private and public works; I'm not aware that Bach or any-
body else thought in these terms, though I must say I cannot deny it either. But
I suspect it's a new thought, which may be a bit overdone. I quite agree that
there are some hidden entrances, but I don't know if the reason is that it was
only for oneself; the thought is nice, but I would not like to say that that's the
reason. That's definitely one of the pleasures of polyphony, that there are some
surprises, subtleties, that make you think "Ah!" because they are hidden at
first. You may say these are literary private jokes, for the connoisseur, but such
people may also sit in a large hall. And I think the general music lover, who is

not really listening closely to what happens, may enjoy a piece that is full of
hidden entries without ever noticing them. It may be a bit academic to make
the distinction, but the fact happens that there are clear-cut and hidden pieces.

In discussing the change in harpsichord playing, we set aside the unhistor-

ical instruments of fifty years ago, but one of the developments of the last thirty
years has been the historical re-creation of specifically Baroque harpsichords,
and of nationalstyles of harpsichord —
the French, Italian, and German instru-
ments for example. You've been very much involved with the revival of these.
How much does it matter to have say, a French harpsichord for French music?
What's your experience?
I thmk the whole development, which I consider a sound one, leads to more
refinement. One discovers more important than we
little things that are
thought, so they don't remain whole view perhaps
little things — well, in the
they are still little things, but the refinement means that more and more one
hears and experiences how certain instrument types may have inspired com-

14. Rosen, "The Shock of the Old," New York Review of Books, 19 July 1990, p. 50.
 ONE SHOULD NOT MAKE A RULE 203

posers of that period. Things that did not sound well, or even that were dull
on one instrument, would, with the proper instrument, all of a sudden start to
live. So that, indeed, an instrument of the composer's period and country is cer-

tainly the best. There's no doubt about that; it's been proved for people with a
refined ear. On the other hand, with modern concert life, which is different
from anything before the 1800s, we should not be too stubborn, so I person-
ally think that although it is perhaps not perfect to play Bach on a French eigh-
teenth-century harpsichord, it doesn't matter so much. On the other hand, to
play Frescobaldi [an early-seventeenth-century Italian] on a Kirckmann [a late-

eighteenth-century English builder] is so unsuccessful that I don't do it. But,

then, I'm not angry if somebody else does.

Organs also varied greatly from place to place. You've recorded on quite a
variety of organs; has that had similar benefits?
Organs are of such individuality that I find it impossible sometimes, tech-
nically, to play certain music on certain organs. For example, French music is

thought out for French instruments with certain standard registrations; you
hardly ever play any of them on organs of another country. And even Italian
organs are, again, so specific and standardized in layout that most of the pieces
only come off on that kind of organ. So there we are much more restricted. I

think the idea of taking one specific type of instrument is the only good solu-
tion to arriving at a better understanding of the music.

Finally, I want to ask about national styles in the current music scene. The
Netherlands has produced an unusually large number of prominent early-music
players, like Italy producing opera singers; do you have any theories to explain
why that might be the case?
No, just chance. Although a thing that happened by chance may perhaps
spread because there's an example.

/ think of the main examples as coming from yourself, and Brueggen,

Bylsma, and the Kuijkens. Are the kinds of performance practices one associ-
ates with the Netherlands — the metrical hierarchy, rhetoric, and so on — the
sorts of things you were thinking about?
Oh, we never thought about them. We never thought about developing any-
thing much. We never talked about any issues. We didn't make a point of any-
thing, ever. We played, and each one studied the pieces. We played — we had no
theories. Perhaps in secret; but no, I never had theories. I was investigating all

the time, but from a tradition to a wealth of general concepts. And maybe it

[our style] is all wrong; I don't know, it could be.

Is there anything else you would like to add to what we've discussed?
No, I have nothing to say, I am only a player.
204 THE BAROQUE

As opposed to?
To a real musician, which is a composer.

SELECTED DISCOGRAPHY
What Leonhardt says about playing in an exemplary way on recordings may
sound off-putting, and some of his discs do seem to lack the spark of live

performances; critics have been known to comment on this (e.g., Teri Noel
Towe considers his B Minor Mass "almost completely devoid of excite-
—
ment"'^ though some critics appear to have found more of it; Deutsche Har-
monia Mundi 77040-2-RG, 2 CDs). But listeners with a taste for Leonhardt's
intensive approach may find many of his 200-plus recordings gripping, and
sometimes overwhelming. For example, Nicholas Anderson writes that the
Dies irae of the Biber Requiem a 15 (DHM 77344) "inspires Leonhardt and
his musicians to deliver it with fearful fervour," and says that Leonhardt

"makes a good deal more of the drama" of the work as a whole than Ton
Koopman does in his recording.'* Below are listed a few of Leonhardt's best
recordings, with no further attempt to give a balance of pro and con.
For a sample of Leonhardt's keyboard work, you might start with his

second recording of the Prelude, Fugue, and Allegro in Eb (BWV 998) on

an excellent 1986 Philips Bach recital (416 141). This is one of my favorite
Bach performances —Leonhardt has a unique feeling for this work's myster-
ies. It's interesting to with his 1965 recording of the same work
compare it

(DHM 77013-2-RG, 2 CDs) to see how his conception of the Prelude

changed over two decades. Bach lovers should definitely hear the other works
on the earlier CD, especially the Art of Fugue (his second recording of it,
from 1969). Leonhardt has since the 1950s argued that this work was in-
tended for the keyboard —an argument that today is widely accepted —and his

performance here (as well as his booklet essay) makes a strong case for that

position.
Leonhardt has recorded almost all of Bach's keyboard works and most of
the concertos and chamber works, plus many of the choral pieces (discussed
in the discography of Chapter 15) and organ works. A good example of his

Bach keyboard recordings is the late-1970s French Suites (RCA Seon GD

71963, 2 CDs): the Allemande of the Fourth Suite, for example, has a mys-
tical gravitas, Scott Cantrell describes the Gigue of the Fifth as "limpid," and

about the Courante of the Second he says he has never "so viscerally felt the
momentum. "^^ Leonhardt has recorded the English Suites and Partitas twice.
Howard Schott prefers the earlier version of the English Suites (no longer

15. Towe, "J. S. Bach: Mass in B Minor," in Choral Music on Record, ed. Alan Blyth
(Cambridge University Press, 1991), p. 58.
16. Anderson, Gramophone 73 (August 1996), p. 81.
17. Cantrell, "The Multifaceted Mr. Leonhardt," High Fidelity, March 1979, p. 85.
 ONE SHOULD NOT MAKE A RULE 205

available, on Philips-Seon);"* I have heard only the second version (Virgin

61157, 2 CDs), recorded in 1984, and I find it excellent. The Prelude of the
Fourth Suite, for example, is an exhilarating rebuttal to critics who think
Leonhardt is at home only in sober, grave music; so, for that matter, is his

lyrical Allemande and his smiling Menuet I. The First and Third Suites also

stand out particularly. Leonhardt doesn't take many repeats (the earlier record-
ing takes many more, I am told), but those he observes show his skill at or-

namentation.
Alan Curtis calls Leonhardt "certainly the greatest living improviser in
Baroque styles of accompaniment as well as in solo organ and harpsichord
playing, [whol far surpasses in both authenticity and imagination anything I

can remember having heard before."'^ We in the public are unlikely ever to

hear Leonhardt's solo improvisations, but we can at least hear his outstand-

ing work as an accompanist, sometimes improvising from a bass, on a num-

ber of recordings. Listen especially to his recordings of Bach chamber works
with the Kuijken brothers, on Deutsche Harmonia Mundi. Of the Flute Sonatas
with Barthold Kuijken (DHM 77026-2-RC, 2 CDs), Richard Taruskin writes,
"Those attuned to subtlety will find more of it in these poised, deeply con-
sidered renditions than in the work of any other flutist," and praises the
"warmth and scope" of Leonhardt's playing.'"
Leonhardt was a pioneer in recording the music of keyboard composers
before Bach. One of his best is of Georg Bohm, Bach's friend and possibly
his teacher (Sony SK 53114); Kevin Bazzana writes, "Order and improvisa-

tion are expertly balanced here: Leonhardt preserves distinctive rhythmic pro-
files in the dance movements while injecting a considerable amount of per-
fectly judged rubato.'"^^ Another highlight is a profound 1989 recording of
the seventeenth-century giant Johann Jacob Froberger (DHM 7923-2-RC). A
recital of French Baroque music (DHM 77924-2-RC) conveys, according to

Julie Ann Sadie, a range of emotions and displays a "superb sense of timing
[that is] unrivalled today."'^ Another notable harpsichord CD is a 1988 recital

featuring various composers (Philips 426 352), which gives an idea of Leon-
hardt's range, both stylistically and expressively.
"Leonhardt's recordings as an organist," writes Patrick Russill, "are all

too rare. No other player can draw such poetic intensity . . . from just a sin-

gle stop."" Only one Bach organ CD is readily available (DHM 7868-2-RC),
but it includes some of the most masterful organ playing in my collection.

18. Schott, "Ein vollkommener Musik-meister," p. 516.

19. In Curtis's edition of Monteverdi's L incoronazione di Poppea (London: Novello,
1989), p. XV.

20. Taruskin, Musical America, May 1990, pp. 69-70.

21. Bazzana, Fanfare 17 (November/December 1993), p. 193.
22. Sadie, Gramophone 69 (May 1992), p. 79.
23. Russill, Gramophone 72 (October 1994), p. 164.
206 THE BAROQUE

Another fine recital (Sony SK 66262) is titled North German Organ Masters;
Haig Mardirosian praises its "remarkable freshness and admirable insights."^'*

Leonhardt has also recorded an entire CD on the clavichord. The 1988 recital

on Philips (422 349) is a classic, with inspired performances of C. P. E. and

W. F. Bach —whose polonaises, in Leonhardt's hands, are heartbreaking —and
of their father's Second French Suite, played with a piquant touch and elo-
quent rubato.

FOR FURTHER READING

Richard Troeger's Technique and Interpretation on the Harpsichord and Clavi-
chord (Bloomington: Indiana University Press, 1987) is excellent, as is Peter
Williams's article "Keyboards," the first entry in Performance Practice: Music
after 1600, ed. Howard Mayer Brown and Stanley Sadie (London: Macmillan,
1989, and New York: Norton, 1990). An approachable discussion of perfor-
mance-practice issues, including many of those Leonhardt speaks about, is Peter

le Huray's Authenticity in Performance: Eighteenth-Century Case Studies

(Cambridge University Press, 1990).

In recent years, several guides to Bach's keyboard works have appeared.

David Schulenberg's superb The Keyboard Music of J. S. Bach (New York:

Schirmer, 1992) is at the moment the indispensable handbook for anyone seri-
ously interested in the repertoire. Peter Williams's three-volume The Organ
Music of J. S. Bach (Cambridge University Press, 1980-84) is masterful and,
once again, indispensable; the first two volumes analyze the works one by one,

and the third is a brilliant discussion of performance-practice issues. Paul

Badura-Skoda's Interpreting Bach at the Keyboard (Oxford University Press,
1994) contains stimulating ideas about interpretation from a leading performer,
but must be used with caution: it does not clearly distinguish between current
scholarly findings and the author's personal solutions, and its scholarship is not
always reliable.

24. Mardirosian, Fanfare 19 (July/August 1996), pp. 376-77.

 11
Alaaain 's Lamp
— oOo^ g^

Anner Bylsma on the Cello

(and Vivaldi, and Brahms)

William H. Youngren, who writes frequently about Wagner recordings, com-

plains that most modern string performances "have a certain oppressive mo-
notony. . . . Almost every phrase, regardless of content or context, is delivered
in the same big, luscious, vibrato-ridden tone. ... a thick wall of throbbing
sound comels] between us and the music."' A different complaint was lodged
by no less mainstream a musician than George Szell: "Any subtle function of
the wrist and fingers of the [bowing] hand is practically unknown to [most
modern violinists]. They have never been told that the bow has to articu-
. . .

late the music. "^ For these and other reasons, Hans Keller, the music analyst,

string-quartet coach, and devotee of Furtwangler and Casals, concludes that

"contrary to official, professional views, modern [string] technique has actually
narrowed down our expressive range. "^
Obviously, these critics (who are not in the early-music camp) don't mean
to damn all mainstream players. Only extremists would dismiss such talents
as— —
to mention only cellists ^Jacqueline Du Pre or Yo-Yo Ma. And I've read
about the violinist Anne-Sophie Mutter making extensive use of non-vibrato in
the Sibelius Concerto. Still, the group to whom the complaints about vibrato
and bowing most clearly don't apply are the outsiders —the period-instrument
players, about whom mainstreamers are quick to lodge their own complaints.

1. Youngren, "Vocal Violin," The Atlantic, November 1992, pp. 144—48. The article is an
appreciation of Josef Szigeti.
2. Ibid, p. 148.
3. Keller, "Whose Authenticity?" Early Music 12 (November 1984), p. 517.

207
208 THE BAROQUE

Whichever side you take, when you discuss the Baroque cello you have to put

Anner Bylsma on the shortest of lists. Bylsma took up the Baroque cello in the
1960s, by which time he already had a distinguished career as a modern cellist
(which we discuss below). In the ferment of the early-music scene of Amster-
dam, he says, it was inevitable that he would be drawn to the early cello. That
he still plays both instruments makes him a sharp observer of their differing
challenges, and of current issues in their use.

He also spoke about a favorite composer of his, Vivaldi. Vivaldi was the
most influential Italian composer of his day, and the only one of his time and
become truly popular in ours. But serious musical thinkers often dis-
place to
miss him.The modern Italian composer Luigi Dallapicola uttered the most fa-
mous put-down: "Vivaldi wrote a great concerto, 500 times." Charles Rosen
thinks it wasn't even a great concerto. Bylsma, however, thinks that musicians
play and hear Vivaldi wrong; he believes that the key to Vivaldi is his interest

in depicting character.

This point may reflect The nineteenth cen-

changes in musical aesthetics.'*

tury, as John Butt said in Chapter 9, placed new emphasis on the tight inte-

gration of musical works; and in our century some have regarded music as a
set of formal structures and nothing else. People may not have really believed

that even in the high modernist era, but it may lie behind some of the Vivaldi-
bashing. Looking at Vivaldi in purely formal terms may impoverish him, which
it does not do to Bach, who. Butt said, wrote music that was unusually inte-
grated for its time —and who, ironically, discovered how to handle large-scale

form by studying Vivaldi.

Another obstacle for Vivaldi is that later in the eighteenth century thinkers
began to emphasize musical genius and innovation, while in Vivaldi's time a
musician was, above all, a craftsman^ (he prided himself on being able to turn

out a concerto in minutes). Borrowing from one's own work was normal. Ad-
mittedly, Vivaldi reused the same patterns to a degree unusual even for his own
day. But he was often brilliantly inventive within his form's parameters —para-
meters that to a significant extent he created himself.
Bylsma and I met briefly in an airport coffee shop while he was waiting for
a flight; we spoke again later by telephone. Both conversations exhibited

4. See Carl Dahlhaus, "The Metaphysics of Instrumental Music," in his Nineteenth-Cen-

tury Music, trans. Bradford Robinson (Berkeley: University of California Press, 1989), pp.
J.

88-96; and "Theme and Character," chap. 6 of his Ludwig van Beethoven, trans. Mary Whit-
tall (Oxford University Press, 1991).
5. Edward Lowinsky says that Rousseau was the first to write about the concept of genius
in music, as opposed to the older ideal of craftsmanship; Lowinsky, "Musical Genius: The Evo-
lution and Origins of a Concept," in his Music in the Culture of the Renaissance (University
of Chicago Press, 1986), originally published in Musical Quarterly 50 (1964), pp. 321-40,
476-95. Charles Rosen observes, though, that "the late eighteenth century does not mark the
first appearance of the concept of the temperamental genius in music"; he points to a fifteenth-
century Netherlandish composer Hke Josquin, and says that in the arts in general Michelangelo
provided the "basic model of the temperamental genius." In "Did Beethoven Have All the
Luck?" New York Review of Books, 14 November 1996, pp. 57-63, quote p. 58.
 Aladdin's lamp 209

Bylsma's gifts for figurative language and for aphorism, and his originality of

thought on many aspects of music —among them the issues raised by Keller,

Youngren, and Szell.

Performance Practice

You started out on the modern cello, winning the Casals Competition in 1959
and playing first cello in the Concertgebouw Orchestra.
I still do play the modern cello. And of course, everybody played the mod-

ern cello at that time. I played in the Concertgebouw in the 1960s, and left
about twenty-five years ago. I never regretted leaving, although I like the or-

chestra very much.

I began working on the Baroque cello during my days at the Concertge-
bouw, because playing early music with musicians who were playing Baroque
instruments made it a natural thing to do. Playing with instruments like the
harpsichord and recorder, which were so clear and at the same time so fast and
soft, made it hard to deliver on the modern cello. 1 was always having to play

between pianissimo and pianississimo. It sounded so horribly unnatural that ex-

perimenting with Baroque cellos came easily.

You, more than anyone, developed the prevailing style of Baroque cello
playing. How did you go about doing this —using Baroque treatises, or what?
Well, I think the mastering of an instrument never goes through reading
first, and then playing. It goes through playing first, and then reading —and
having good colleagues, especially people who play other instruments.

One aspect of the style heard in your playing and that of your colleagues,
such as Leonhardt and Brueggen, is an eloquent use of rubato. Could you
speak about that?
The word "rubato" means "robbed" — you rob from the bar. Rubato is pos-
sible only if there is a very keen sense of the pace of the music. If you see a

film, and somebody is in a cell, and he grabs the bars and wants to get out be-
cause the government wants to shoot him, the whole audience wipes its brow
in sympathetic anguish. Now, if this fellow were standing in an open meadow
there would be nothing to grab —there'd be no rubato. There's no feeling of
having to get out of anything.

It's the tension of freedom versus order.

It is the individual against law. One person's rebellion against restrictions.

There seem to be different approaches to rubato in different eras. Is there such

a thing as a Baroque approach to rubato, as opposed to a Romantic rubato?
No, the thing is that there's no "Baroque" in the first place. There are so
many variations. What is a twentieth-century man.-* Just look around and you
210 THE BAROQUE

see so many types; some would fit very well in the eighteenth or nineteenth cen-
tury. Or the thirty-first, but we don't know that.

The Gramophone reviewer of your Sony recording of the Bach Cello Suites
was enthusiastic, but also remarked that "if Pierre Fournier had allowed him-
self the expressive licence demonstrated by Bylsma he would most probably
"^
have been roundly condemned for excessive romanticism.
All these things go in waves. First, people do too much rubato, and then
somebody comes up with the new idea —don't do any at all. And it's the same
with vibrato and portamento,^ and all these things. It's like fashion — if every-
one is wearing long skirts, you can be sure that the next thing will be short

skirts. And to be honest, after so many long skirts you'd like to see some leg.

How about vibrato and portamento? Those are two other aspects of string
playing that have gone in and out of fashion. There's confusion over how much
they used vibrato in the past.
I have at home
method by Kummer, a cellist, written about 1840, saying,
a
"Formerly, people vibrated much too much; the modern style of playing de-
mands that we also be able to draw with clear lines." And you can find this
kind of thing time and again; we think that something is quite new at a cer-
tain point, when generations before people were saying it was old-fashioned.
Praetorius in 1619 said the instruments of the violin family sound best with
iron strings. Cambists vibrated on the frets, flutists in eighteenth-century France
used a vibrato that sometimes covered almost a whole tone. And you can tell
that some string players vibrated a lot and some not at all. Geminiani in his
violin school of 1751 said you should vibrate as often as possible.^

Though since he didn't hold his violin under the chin, he couldn't have meant
a constant vibrato, like that of Fritz Kreisler — it would have been too difficult.

True; but violinists were very good at using their chins when
needed, and
of course it depends on how long your neck is. And Kreisler was not such a
vibrator; it's more the American and Russian violinists of today. Have you
heard Kreisler's recordings? What a beautiful player, and his vibrato is not so
like an electric bell, the way many people sound now. And great rhythm! tSings
the opening of the finale of the Brahms Violin Concerto as Kreisler played it.]

Kreisler and his contemporaries grew up playing gut strings, and on gut strings
vibrato works quite differently. One does not have to hide the ugliness of steel

strings.^

So it's difficult to give a straightforward answer, like "they vibrated not at

all in the seventeenth century, a little in the eighteenth, and more in the nine-

6. Nicholas Anderson, Gramophone 70 (January 1993), p. 49

7. "Portamento" is from one note to another.
the audible sliding
8. Francesco Geminiani, The Art of Playing on the Violin (London, 1751), p. 8.

9. The top (E) string on the violin remained pure gut until about 1920, when steel strings

—
generally took over though gut E strings did not entirely disappear from mainstream play-
 Aladdin's lamp 211

teenth, and most today." That would not be true at all. Joseph Joachim
[1831-1907] still vibrated very little, while the cellist David Popper
[1843-1913] vibrated much more than his contemporaries. The cellists around
Brahms, Robert Hausmann and Alfredo Piatti [1822-1901], vibrated very lit-

tle. It would be a pleasure if people were to start using less vibrato in Brahms,
because you get so tired —you hear string players in Brahms and in five min-
utes you are already at the end of what you can digest.

How about in orchestral playing^

Weil, my 91 -year-old father-in-law, who is a very fine violinist, had his first

orchestral job in 1916, in the second violin section. He was a poor boy and
was so happy to have a job that he tried to play as well as he could. They had
just started playing when the conductor stopped and said, "Hey, young fellow,
no solo tone here." Because he was vibrating. So all the comments by people
in the early-music world about how the mainstream players are applying nine-
teenth-century style to old music are ill-informed —they use a twentieth-century

style, one that developed between the world wars.

The older recordings prove it.^" On them, even the soloists use vibrato only
on expressive notes, not in passages or background material.
That we now vibrate all the time is a pity. It feels as if people are afraid
that their neighbor in the orchestra will say behind their backs, "He cannot vi-

brate correctly." It seems as if people don't use their judgment and don't listen

to whether a particular passage or note is dissonant or consonant, but just keep

vibrating.

To be specific, would you stop vibrating on dissonant notes, to increase the

discomfort^ Or do I have it backwards^
I think one should not vibrate on the consonances. A consonance is unin-
teresting . . . you never read in the paper about father putting on his slippers

and lighting a cigar, except when the house, being full of petrol fumes, blows
up. A good time for vibrato.*'

ing until later. The two middle strings remained (for the most part) pure gut until about 1950.
Sir Adrian Boult, who was conducting during the transition, lamented that the adoption of
steel E-strings had ruined orchestral string tone.
The lowest string on the"overspun" since the eighteenth century, though
violin has been
authorities differon when this started being done with metal winding; Robin Stowell says
metal winding was "increasingly common" in the late eighteenth century ("Strings," in Per-
formance Practice: Music after 1600, ed. Howard Mayer Brown and Stanley Sadie [London:
Macmillan, 1989, and New York: Norton, 1990], p. 239).
10. See Robert Philip's Early Recordings and Musical Style (Cambridge University Press,
1992). These matters are discussed in the Postscript to the Norrington interview.
11. In Chapter 12, below, Julianne Baird suggests the opposite; this probably reflects the
differences between the way most Baroque singers use vibrato (as a continuous style of tone
production, which can be varied) and the way most Baroque instrumentalists use it (as an oc-
casional ornament).
212 THE BAROQUE

How about vibrato in Bach's or Beethoven's time?

Again, it is difficult to make generalizations. For instance, in Berlin in

Beethoven's time there were three cellists with quite different styles of playing
Jean-Louis and Jean-Pierre Duport on the one side, and Bernhard Romberg on
the other. Romberg, who was a friend of Beethoven (and you may take it that
his playing was an example for most of what Beethoven wrote for the cello so —
we'd better mind that), hardly vibrated, and you can see that from the way he
held his fingers at a slant, like a violinist.'^ Jean-Louis Duport, who wrote a very
important essay that has since been the basic text for all cellists on fingering,

held his hand in a more perpendicular way, where one could vibrate easily. I'm
sure there must have been quite a difference in the way these two people played.

So when you play Beethoven, you follow the non-vibrato approach of

Romberg?
I try, because I like it very much. If you don't vibrate, you're much more
alert about your bowing;'^ also, instead of vibrating you can use portamento
slides —as another way of enlivening a note. And it leaves you with vibrato as

a trick up your sleeve, so that if you really feel you have to do something, you
can vibrate all of a sudden. That is often in the places where a modern cellist

would stop vibrating all of a sudden. In the first movement of the Second
Sonata of Beethoven, in the coda, there's a place where people tend to stop vi-

brating, eight long notes at the end; I like to suddenly use vibrato there, and
it's just as special.

And how about portamento?

That was always there, of course. In the naughty or sad times of one's life,

one's shifts lof a finger from one note to another] tend to become a little more
pronounced. I personally like fingerings which use the same finger more times
in a row, which naturally brings forth more portamento. You often find these
fingerings in nineteenth-century texts.
But the comparison of Romberg and Duport shows that, in general, because
string instruments have so many more possibilities than a recorder or a harp-
sichord, the styles of playing must have been vastly different not only over the
years, but also within the same era. We know that Corelli, in Naples, told Han-

12. David Watkin details Romberg's "preference for expression through the varied use of
the bow rather than vibrato." Romberg appears to have used vibrato only as an ornament,
and then only in the first one-third of a note. Watkin, "Beethoven and the Cello," Perform-
ing Beethoven, ed. Robin Stowell (Cambridge University Press, 1994), pp. 110-11.
13. See previous footnote. Hans Keller observes that in coaching young string quartets his
instructions to play passages non-vibrato often produced regrettable results: "Owing to the
narrowing down of our expressive range, the player's right arm is, in most cases, no longer
capable of producing the tone modulation required for his vibrato-less execution ... his right
arm has grown up behind the screen of a vibrato" ("Whose Authenticity?" p. 518). Bylsma's
remarks suggest that the converse also applies: playing without vibrato forces one to bow
more subtly.
 ALADDIN'S LAMP 213

del he couldn't play French music. Now, that is only the difference between two
major styles, the French and Italian; but the artists within each style must also
have been very individual.

One Italian played very differently from another Italian.

Oh, I am sure of that. And how can we tell? Well, we have a sentence here

or there saying how this or that one played, but also we have their composi-
tions and the way they use slurs in notation. And sometimes they wrote
books — like Geminiani or Leopold Mozart. The most outspoken were the
French composers, who were more schoolmasters than the Italians. The Ital-

ians would someone else would

generally rather not write a treatise, for fear
steal their secrets, but the French were obsessed with what they called "good

taste," to the point that you think good taste is another word for "jealousy,"

because somebody else has more invention than you have the Italians cer- —
tainly had more imagination than the French. Even the native French suites are

often less interesting than French suites written by Italians, let alone those writ-
ten by the Germans, who mixed the two styles together and made out of them
the incredible thing we find in Bach.

The French and Italian styles were dominant at the time of Bach. Can you
discuss how the French and Italian string-playing styles differed?

One example is the downbow rule, an especially French rule. If you look at
Muffat's introduction to his Florilegium Secundum, you'll see how he says that
the French do a downbow on the first beat of every bar, and that the Italians
go back and forth. It's a very interesting statement. You should keep in mind
that in France, the famous orchestra of the twenty-four violins of the king
played a great deal of ballet music, for which it's wonderful if you have a strong
accent on the first beat. The French possibly also had shorter bows.
I think the famous story of Lully dying of gangrene after hitting his foot

with his big staff, with which he beat time loudly, shows that it must have been
hard for all these guys to have their bow on the right spot, barline after bar-
line. It must have had to do with this French technique where you have to go

back to the frog of the bow on every downbeat.

In the notation of rhythm, the eighteenth century saw imprecise conven-

tions, like the French notes inegales where the first note in each pair is held
longer and the second note shortened —and also overdotting.
No, I think the French were very precise about where their inegalite was ap-
and they gave you all the exceptions to it also. Inegalite is always pairs
plied,

of notes slurred together, in runs like scales; there are many, many exceptions.'"*

14. See Stephen Hefling's Rhythmic Alteration in Seventeenth- and Eighteenth-Century

Music (New York: Schirmer, 1993). Chaps. 1 and 2 discuss French practice; later chapters dis-
cuss controversies concerning the application of this practice outside of France, and also the
issue of overdotting. See also note 13 in the William Christie interview, below.
214 THE BAROQUE

On the other hand, the Itahan style of "Lombardic" notation/^ which is the
opposite of notes inegales, is mostly not very precise as to where it should be
applied —and I'm sure Italians like Geminiani did everything they liked in the

French style as well. Inegalite is like shaking someone's hand with a very soft
handshake, while the Italian Lombardic style —where the first note is short and
the second is long — is vigorous and proud.

What about the issue of assimilating dotted notes to triplets?

One must decide case by case, but mostly, to make these things equal seems
dull. The clashing of triplets and duplets can also be found in later music; for
example, in Beethoven's "Moonlight" Sonata, where the right hand is playing
triplets and the left hand is playing dotted rhythms, you wouldn't reconcile
them.

The French and the Italians ornamented differently?

The French would have notated all the ornaments, while the Italians would
mainly have just ad-libbed it; that's the general modern opinion.

75 that why Quantz said that the Italian style required good knowledge of

harmony to extemporize ornaments, while the French didn't?

All good players must understand harmony in adding ornaments, because
an ornament must express an excess of emotion. Too often now people add or-
naments like babbling at the mouth; they add runs and chords that don't mean
much, and take away from the real character of the piece. Indeed, one very nice
way of ornamenting is to leave out notes. Often a single note, a dissonance re-
solved much better ornament than a whole exercise
late, is a of scales. Orna-
ments must come from the heart, not from the fingers.
To get back to Quantz's point, the reason the French notated the ornaments
and the inegalite so precisely, I think, is that the French were so concerned with
comme il faut; Marin Marais played all of his pieces through every two weeks,
and I think it was just to make each one a littlemore meticulous and precise,
not different. Giveme Vivaldi any time, his incredible amount of fantasy, imag-
ination, and daring. And give me Vivaldi a thousand times over Corelli, who I
think tried to make music palatable for the upper classes, the bored rich, of his
day. I think Vivaldi must have hated him; the violin sonatas, Op. 2, of Vivaldi
are like Corelli, but there's a little kick in the pants here or there —not that it's

wrong, actually, but it's like a finger in the eye. It's very interesting — Corelli
seen in the light of Vivaldi.

15. The "Lombardic" rhythm is described by Johann Joachim Quantz, Frederick the
Great's flutist, in his treatise on flute playing, Versuch einer Anweisung die Flote traversiere
zu spielen (BerHn, 1752): On Playing the Flute, trans. Edward R. Reilly,2nd ed. (New York:
Schirmer, 1966), p. 323. Quantz and other Berliners thought that the rhythm derived from
Corelli, Torelli, and Vivaldi.
 ALADDIN S LAMP 215

Vivaldi

Vivaldi is the most often dismissed among the major composers. Stravinsky, for
example, called him "greatly overrated —
a dull fellow who could compose the
same form over and so many times over."^*'

Stravinsky's remark —or is it Craft's? — is very superficial, and a pity for

somebody buried in Venice. I feel shy about daring to criticize a genius like
Stravinsky; but could it be that (if it was not unfamiliarity on his part) he felt

irritation in meeting a spirit in many ways like himself: wit, irony, a certain

"coldness" or, at least, professed coldness?

Of course, in Stravinsky's time Vivaldi was played like a tenth-rate
Brahms,'" without any idea of how to characterize all the commedia dell'arte

figures, plus those one could encounter on the streets of Venice. This is the most
surprising thing about him —the endless variety of characters depicted and car-
icatured. All of his music is always depicting a character.

Can you give an example?

The first movement of the G Major Cello Concerto seems like a fat fellow

walking down the street, enjoying himself immensely:'^

Allegro

I
i^^
^ ^^ ^
16. Igor Stravinsky and Robert Craft, Conversations with Igor Stravinsky (Garden City,

N.Y.: Doubleday, 1959), p. 76.

17. Stravinsky did complain about "sewing-machine performances of Vivaldi." Robert
Craft, Stravinsky: Chronicle of a Friendship, 1948-1971 (New York: Knopf, 1972), p. 178.
18. Vivaldi, G Major Concerto, RV 413 (recorded by Bylsma on Sony SK 48044), opening.
216 THE BAROQUE

The famous E Minor Cello Sonata is somebody wallowing in an opera on

stage/' But Vivaldi is always theatre; it's always characterizing, as you would
do in a play. Also, his love of color and sonorities reflects the splendor of
Italy.

Nowadays, he is not just misinterpreted; too often he is uninterpreted.

Many modern soloists when playing Vivaldi do not seem to have ever looked
at any caricatures by, say, Ghezzi, Tiepolo, or Guardi. People often just play
scales and chords, with no feeling for that period or nationality. There is so
much more there than you think. I once got a compliment after I played a con-
cert of Vivaldi, Frescobaldi, Gabrielli, and Boccherini: an Italian colleague said,
"I loved your Boccherini, because it was exactly the dialect of Lucca." This did-
n't —
make me feel so safe I thought, "What about the dialect of Venice, the di-
alect of Rome, the dialect of Bologna?"

For all the bad-mouthing he was the most influential composer

gets, Vivaldi

Even older composers like Albinoni

of the first half of the eighteenth century.
modified their styles in mid-career to be more like his. Vivaldi had a decisive
influence on Bach, who adopted the ritornello principle from him. Bach told
his sons, apparently, that studying Vivaldi "taught him how to think musi-
"^° Could you comment?
cally.

He was an incredible form-maker: witness what Bach said about him. And
Vivaldi, apart from being a great contrapuntist, and apart from the many ways
of bowing and left-hand virtuosity, is an unfailing stage-setter. And what in-
vention!

Eleanor Selfridge-Field writes that "Vivaldi's talent for extracting a succes-

sion of motivic variations, or the process of Fortspinnung, from sedate open-
ing phrases distinguishes him and enables soloists to enter with an air of drama
rather than of mere duty." Any comments?
One main kind of Fortspinnung — literally, "spinning out" — is of course the

19. RV Harmonia Mundi 7909.

40; recorded by Bylsma on Deutsche
20. Bachwas introduced to Vivaldi's music when he was serving the young music-loving
Duke Johann Ernst at Weimar in 1713. The ritornello principle derived from Vivaldi's con-
certos became the basis for the large-scale organization of most of Bach's music thereafter. In
this design, an opening section for the whole orchestra (tutti) returns in full at the end of the

movement, and in part, in different keys, at intervals in between. These ritornellos ("little re-
turns") tie together the "episodes," which are played by solo instruments and often involve
changes of key. A clear discussion of the Vivaldi ritornello, by Walter Kolneder, appears in

The New Oxford History of Music, vol. VI, pp. 302-39. For discussions of the influence of
the ritornello on Bach's concertos, see Malcolm Boyd's Bach: The Brandenburg Concertos
(Cambridge University Press, 1993), chap. 5; on Bach's choral works, see John Butt's Bach:
Mass in B Minor (Cambridge University Press, 1991), chap. 5. For a discussion of the more
subtle influences that Vivaldi had on Bach's style, see Christoph Wolff's "Vivaldi's Composi-
tional Art, Bach, and the Process of 'Musical Thinking,'" in his Bach: Essays on His Life and
Work (Cambridge, Mass.: Harvard University Press, 1991), pp. 72-83.
 Aladdin's lamp 217

sequence.'' In itself it's a boring device used to modulate, but in Vivaldi's hands
it is full of wit. For instance, sometimes he'll put in one sequence too many, as
a caricature,

Michael Talbot wrote that Vivaldi can be seen as a "harbinger of musical

Romanticism'' partly because of "the higher value he placed on expression than
on perfection of detail. "~~ Could you comment?
I think Vivaldi certainly had a better sense of what the important and unim-
portant notes were in a bar, and definitely played the important notes in the
melody more in tune than any of us do now. To hell with the perfection of the
highway tarmac!
It's the same issue in playing any of the great composers. Imagine what goes
through your body when you have composed a piece like the St. Matthew Pas-
sion, and you lead it. Or imagine that you're Maurice Ravel, and you're sitting

in the hall and hearing Daphnis et Chloe. It must be so much made out of your
own soul that it must be hard to take. All these great people, our demi-gods,
are so much greater than we realize. I've worked on the Bach cello suites for

over forty years and keep coming back to them, and still they don't get bor-
ing — still there is more depth.

The Baroque Cello

The obvious differences between the modern and the Baroque cello are the

gut strings, the bow, and the endpin. I'd like to ask about each of them in
turn.
Technical changes are never only technical. They always are more than just
that and give new possibilities where other ones mostly are not perceived to
disappear. The steel string of course is not just technical. The heyday of the
steel string is the time of the big swing orchestras, especially the Glenn Miller
band —everything smooth and round.
People argued that steel strings were more powerful and more reliable.

The arguments for steel are mostly cheap —cheapness, for instance, ease of

use, loudness, and, worst of all, equality across all the strings.

21. A sequence, in Baroque and later music, means the repeating of a musical phrase at a
it was the primary means that high Baroque composers used to keep their
different pitch;
music moving forward harmonically. The term fortsptnnung is a modern one, first coined to
describe the second (of three) sections of a typical ritornello tutti. The ritornello's introduc-

tory section, usually a few bars long, establishes the home key but cadences on the dominant;
the Fortspinnung section then modulates through various keys (usually through sequences);
and the third, final, section is an "epilogue" in the tonic.

22. Talbot, "Vivaldi", §7, in The New Grove Dictionary of Music and Musicians (Lon-
don: Macmillan, 1980), vol. 20, p. 38.
218 THE BAROQUE

Can you give an example of why equality of strings is so bad?

One example is that they make it very hard to play Boccherini on modern
instruments. I hardly can do it. And it's so easy on an old instrument. The main
reason is that Boccherini makes use of the different unequal sounds from string
to string. When you play high notes on low strings with gut, the sound is a bit
muffled —and Boccherini uses this effect like a stopped horn. Boccherini is a
sound-oriented composer anyway. Almost all of the main notes in his melodies
are bolstered in sound by the harmonics of the other open strings. It's all made
on sound.

How about the endpin, which Baroque cellists today often don't use and
mainstream cellists do use —although nineteenth-century cellists generally did
not use itf"
I don't know why up to our century most cellists played without endpins.
Imagine — great nineteenth-century cellists like Popper, Hausmann, and Piatti
playing Brahms without an endpin, when an endpin definitely makes the in-
strument louder and also easier and more stable.

What do you do about endpins?

I have two cellos, one with, one without.

It's not like gut strings, which really do make a difference?

It may perhaps make a difference, but I have never been able to form an
opinion about it.

How about bowing? It's been said that in the Baroque "the bow, to a much
greater extent than in modern playing, became the primary source of expres-
"^'*
sive inflection. What would you say about that?
All good artists with any kind of bow at any time have their primary source
of expressive inflection in the bow. I did say, though, that when you vibrate
less you have to be more alert about bowing, and also that you don't feel the
need to vibrate as much on gut strings. So perhaps those are factors.

How about the different bows themselves —the modern "Tourte" -style bows
versus the earlier pre-Tourte bows?^^
Of course there were many different bows —the great violin makers also

23. Tilden A. Russell,"New Light on the Historical Manner of Holding the Cello," His-
toricalPerformance 6 (Fall 1993), pp. 73-78. This also bears on the question of vibrato, by
the way: "Without the security of an end-pin, cellists had to rely on the contact of the left
hand with the neck and strings for certainty of intonation, and a continuous, fierce vibrato
was less practicable." Watkin, "Beethoven and the Cello," pp. 92-93.

24. Peter Walls, "Strings," in Performance Practice: Music after 1600, ed. Brown and
Sadie, p. 52.
25. Named after the French bow maker Francois Tourte (1747-1835). According to David
 Aladdin's lamp 219

made their own bows. But I think one should not play with too many differ-

ent bows. Your bow is your magic wand, and you need to know it well. So if

you have a good old bow and a modern Tourte-type bow, that's enough. For
that reason, I don't know whether you should also have a so-called transition
bow, a Classical bow. It's interesting to note that Duport in his Essay wrote,
"The best bows made now are those by Tourte. That's not a compliment, it's

true." It's the last line in his book. was the man who played
One of the Duports
the Op. 5 Beethoven sonatas for the first time; that was in 1795, so from that
date onward there's no doubt that you can play Tourte bows, modern bows,
but possibly in the old-fashioned way.^* For some time, the cellists in France
went on holding the modern bow in the old-fashioned way, as you often see in
caricatures of French cellists.

The earlier bows had a much thinner band of hair, had no ferrule to keep
the band of hair flat, and lacked an even balance. How do such things affect
one's approach? Do they relate to the style of articulation?
Yes, the made much more for speaking, the modern bow much
old bows are
more for singing. But when you use the modern bow in the old-fashioned way,
the difference is not so great as when you hold it the modern way, at the frog.
And, of course, you should not think too much of instruments, because for a
good player, what his inner ear hears his if you know what
hands can do. So
you want, then you can usually do it modern bow and in-
quite nicely with a
strument. But first you have to know what you want. And therefore I would
say, take an old instrument and keep using it — but this is just part of one quest;
in fact, an independent thinker is always developing his ideas.

This question of speaking versus singing, of a rhetorical understanding of

music, seems crucial to Baroque and Classical approaches to music.
Yes; and the dynamics when you speak are much more detailed than when
you sing. When you say four words you have four mezzofortes, one or two
fortes, and several pianos, and then it diminuendos into nothing. So if you tried

to notate your own speech, you'd need all the dynamic signs in a matter of two
bars. That's also the pleasure of it. Sometimes somebody has engaged the
wrong singer for a Baroque piece — if you're used to a more speaking style, it's

horrible when somebody just bellows. With these opera bellowers, you cannot
hear the words anymore.

Boyden, "the type of bow he established about 1785 has continued (a few details apart) to be
the standard 'modern' type." It was not a radical innovation; rather, Tourte's design "succeeded
in .combining the best features of his predecessors in so satisfactory a manner that it set a
. .

standard for bowmaking." Boyden, "Tourte," The New Grove Dictionary, vol. 19, p. 100.
26. This includes, among other things, not holding the bow at the end (the "frog"), as
modern string players do, but a little way up from it. The universal supremacy of the Tourte
bow, by the way, was attained not overnight but over the course of the nineteenth century.
220 THE BAROQUE

One reflection of that more inflected, speaking style is the ornament called
messa di voce the gradual crescendo and decrescendo on a single long note.
Some say that was overdone in some of our era's earlier experiments in
Baroque string playing.
Yes, that was notorious. To be sure, there's also the mezza di voce, where
each note starts very soft; most modern string players use that too little. Both
groups are too extreme: those early-instrument players who swell every note

in the modern players whose notes have no real beginning.

middle, and those
To start a note is woodwinds always practice and strings never
a thing that
practice. And they should. If you listen to an orchestra warming up, and stand

over near the horns, you'll hear them do all these things a note without a —
beginning, a note with a small beginning, a note that is a hundred percent
from beginning to end, a note with diminuendo, a note with a swell in the

middle. If you go to the oboe section, you'll hear the same things. But when
you stand near the violin section, you'll hear all the violin concertos at the

same time, but you won't hear anybody try to start a note well. I am always
surprised when I hear famous cellists and violinists who, it seems, do not
—
know how to start a note they seem to speak a language without conso-
nants.

So starting the note well relates to the aesthetic of speech — like enunciat-

ing a consonant.
I think this idea of speech is so fruitful that you can apply it to almost

all music. The better the music, the more it speaks.

Historically, there was a change from thinking of music in terms of speech

to thinking of a long, unbroken singing line. How did that change in aes-
thetic come about?
A lot more speaking and less singing, such as
of later music does need
Brahms. The influence of Wagner is rather strong. The interesting thing is
that, historically, players have almost always slurred a chromatic figure. Think

of the famous Musical Offering theme of Bach —actually Frederick the Great.
It starts with "words" and "syllables," but when it descends chromatically,
you would not think of syllables: you slur it together. And because Wagner
is so chromatic, that might have contributed to the triumph of the long line.

Another influence modern conductor, with this totally inadequate

is the
musical instrument, the baton, which cannot show motifs in any detail maybe —
once every ten seconds at most. Just try to show with a stick all the inflec-
tions you can make with your bow or your voice or your flute. The prove-
nance of the long line in music is the conductor's, because it's easier to con-

duct that way. Conductors always want to beat slow pieces in four that are
written in two, and fast pieces written in four in two; it damages music very
 Aladdin's lamp 221

much, but then they can conduct it. The answer, of course, is to never look
at these guys.'^^

How would you relate this to early-music conducting?

I sometimes doubt whether there should be a conductor. Of course, what
you can't show with the baton you have to talk about in rehearsal, so in that
case it's not the baton that speaks but the voice.

Could you comment more about the state of modern string playing, partic-
ularly soloists?
When I heard a modern violinist in Khatchaturian the other day, I thought
the playing was fantastic. But often you hear Brahms played too hot, too
slurred together, too egocentric. When you hear Beethoven, it's worse, because
it's all in the same vein; and when you hear Mozart and Bach, it gets worse
and worse. And when you hear a famous violinist play Debussy, it's mostly hor-
rible too.

The thing is that so many of the famous players of today are too dependent
on their teacher; they are not their own man. They are great pupils, fantastic
pupils, without a voice of their own. Give me an amateur anytime a good am- —
ateur, hearing what he thinks he does with tears in his eyes — super! — rather
than a famous pupil. All these pupils' fingerings come from their teachers
Galamian, or somebody in France or in Russia. And of course you'll never find
any of the teacher's fingerings in the playing of Kreisler, or Josef Szigeti, or Bro-
nislaw Huberman, or Adolf Busch. It's personal playing. Those were masters.

Some people might be surprised to hear you praise early-twentieth-century

violinists like Kreisler, because they would expect you to be terribly concerned
about "historical authenticity.

I'd like to say something about being authentic. "Authentic" means "just as

alive as it ever was." Being authentic is, most of all, Aladdin rubbing his lamp:
we rehearse some music, and all of a sudden we have the feeling, "Hey! This
is right. This is the way it must go." And I guarantee you that in a year's time,

when we hear the tape of that, we will agree that it's not at all how it should
go. But it's a very wonderful feeling ''This is how it should go!" —and that's

authenticity. And I think it's worthwhile. But it has nothing to do with being
historically correct. Maybe the motive behind what we do today has something

27. It may be worth noting, in light of Bylsma's remarks, that two of the conductors most
often lionized in our century had an unclear beat, namely Toscanini (according to an admir-
ing Sir Adrian Boult) and, especially, Furtwangler.
In their interviews below, Julianne Baird, William Christie, and Philippe Herreweghe all
comment on the conducting of Baroque music, and Robert Levin talks about playing Mozart
concertos without a conductor. A fairly consistent picture emerges.
222 THE BAROQUE

to do with history
— "This is how it must have been." But one's view of history

changes with the times. When you see some buildings from the 1880s, such as
our Central Station in Amsterdam, where the architect fully believed that he
had conceived a medieval castle, what we see is definitely 1880; it would not
fool anyone now.

SELECTED DISCOGRAPHY
Anner Bylsma's feeling for the Italian Baroque has been captured deliciously on
record. You might begin with his Deutsche Harmonia Mundi selection of rarely
played seventeenth-century Italian cello music, which in his hands is captivat-
ing (DHM 7978). His Vivaldi concertos on Sony with the excellent Canadian
group Tafelmusik (SK 48044) are an antidote to the kind of Vivaldi playing he
complains about, and his recording of the composer's cello sonatas (DHM
7909) gives, among other things, a vivid demonstration of the element of "car-
icature" he speaks of.
Reviewing Bylsma's recording of Boccherini Sonatas (Sony SK 53362), Stan-
ley Sadie praises his "high spirits and nervous energy," "quite extraordinary
rhythmic vitality," "sharp and precise" articulation, and "close to reckless
abandon." In the slow movements, he says, Bylsma plays with much intensity,

eloquence and a "natural feeling for . . . expressive tension"; but these move-
ments, he adds, could occasionally benefit from "a little more relaxation (not
his strong point), warmth, and grace. "^^ Bylsma's recordings of Boccherini con-
certos with Tafelmusik (on DHM 7867 and Sony SK 53121) have met with en-
thusiasm; several critics speculate that Boccherini himself might have sounded
like Bylsma.^'
Bylsma has recorded a great deal of chamber music, much of it with
L'Archibudelli, a group he formed with his violinist wife Vera Beths (admired
for her work in contemporary as well as early music), Lucy Van Dael (the most
unjustly neglected of today's Baroque violinists), and the unsurpassed violist

Jiirgen Kussmaul. Among many fine recordings, some highlights are three sub-
lime quintets —the Schubert (Sony SK 46669), the Bruckner (SK 66251), and
the Mozart Clarinet Quintet (and Trio, with Charles Neidlich and Robert
Levin, SK 53366) —
and Mozart's great Divertimento K. 563 (SK 46497).
Bylsma's comments about how Bach's solo cello suites keep changing for
him are verified by his two recordings of them, made only a decade apart yet
differing significantly. Each set has its supporters, and I suggest that you com-
pare for yourself (unfortunately, as I write the earlier set [RCA RD 70950] is

not distributed in North America). Some critics describe the later set (Sony S2K
48047) as mannered, but others find it especially imaginative and individual. I

28. Sadie, Gramophone 71 (March 1994), p. 69.

29. E.g., Alan George, "Classical and Romantic Chamber Music for Strings," Early Music
23 (November 1995), p. 342.
 Aladdin's lamp 223

side with the latter group. Bylsma, by the way, uses a modern bow in Sony's
first five suites: no listener, no matter how expert, could have guessed that it's

not a Baroque bow, which suggests that the instrument matters less here than
the player.
His Beethoven Cello Sonatas with Malcolm Bilson (Elektra-Nonesuch
79152 and 79236) have been said to "stand beside the finest on modern in-

struments."'" In Brahms's Second Cello Sonata Op. 99 (Sony SK 68249),

Bylsma's tempos are the fastest I've heard. In the first movement this is brac-
ing, but Joan Chissell finds him "marginally too fast for [the slow movement's]
affetuoso to speak as it can."" Chissell praises the disc's Schumann perfor-
mances for allowing "fantasy its full, free rein" and the Brahms First Cello
Sonata for "its unflaggingly strong and purposeful sense of direction"; still, she
suspects that "there has never been a real love-affair between Bylsma and
Brahms."
She might reconsider —or suspect that one has been kindled — after hearing

L'Archibudelli's Brahms string sextets (Sony SK 68252). The first movements

of both sextets have prominent solos for the first cello, and Bylsma plays them
with passion.An example is the second subject in Op. 18's first movement,
marked espressivo and animator Bylsma picks up the tempo a little, with a
sense of surging emotion (tr. 1, 2:11-2:21). In an enthusiastic review of the disc,
James Oestreich says, "L'Archibudelli responds brightly and with seeming spon-
taneity to each peculiar turn in the scherzos but gives full scope to more med-
itative or passionate moments."'^ In these CDs Bylsma doesn't really sound like
the string players Brahms knew, some of whose recordings survive^^ (he uses
less portamento, for one thing); but sounding like them is not what he seems

to be after. Instead, this performance gives an idea of what he means by "rub-

bing Aladdin's lamp."

FOR FURTHER READING

Michael Talbot's Vivaldi in the Master Musicians series is a first-rate biogra-
phy (London: Dent, 1992, and New York: Schirmer, 1993). Vivaldi's critics de-

serve a hearing as well; a representative might be Robert Craft, in his review

"Women Musicians of Venice and the Red Priest," New York Review of Books
27 (2 November 1995), pp. 58-59. Regarding Brahms, see the reading list after

the interview with John Eliot Gardiner. As for historical cello practice, Klaus
Marx's article "Violoncello" in the New Grove Dictionary (vol. 19) is very

30. David J. Fanning, Gramophone 69 (April 1992), p. 93.

31. Chissell, Gramophone 73 (January 1996), p. 69. "Affetuoso" means "affectionate" or
"tender."
32. James R. Oestreich, "Critic's Choice Classical CDs: Brahms' New Day in the Sun,"
New York Times, 8 August 1996, p. C2.
33. See Will Crutchfield, "Brahms by Those Who Knew Him," Opus, August 1986, pp.
13-21, 60, esp. pp. 15-16.
224 THE BAROQUE

good, as are the discussions on "Strings" by Peter Walls and Robin

in the essays

M. Brown and S.
Stowell in Performance Practice: Music after 1600, ed. H.
Sadie (London: Macmillan, 1989, and New York: Norton, 1990). David
Watkin's "Beethoven and the Cello," in Performing Beethoven, ed. Robin
Stowell (Cambridge University Press, 1994), is thorough and informative.
Finally, David Blum's Casals and the Art of Interpretation (Berkeley: Uni-
versity of California Press, 1977) is one of the great books on musical perfor-
mance. Its significance extends well —
beyond string playing though Casals's
views on that relate ways to those expressed by Bylsma. See, for ex-
in certain

ample, the discussion of vibrato (pp. 133-37), which includes a detailed analy-
sis of how Casals would vary his. Says Casals, "When you hear all the time a
beautiful vibrato — well, you've had enough!"
 12
Beyond the Beautiful Pearl
— oOo^ g^

Julianne Baird on
Baroque Singing

More than anything else in music, the art of singing can turn reasonable peo-
ple into red-faced absolutists. That by itself guarantees that the most contro-
versial area of Baroque performance today is singing. It probably explains why
outsiders often have problems with early-music singing: after years of bathing
in rich, warm voices like that of Jessye Norman, critics often can't adjust to the

small, light, clear voices of many early-music specialists. And after the opulent

beauty of the singing of Renata Tebaldi, they often object to such elements as
limited vibrato —though it's a mistake to assume that Baroque singers never use
any vibrato. Such objections sometimes remind me of Paul Hindemith's remark,
"I don't know how, with no vibrato. Bach could have had so many sons."
But as the next half-dozen interviews (and, for that matter, those of Alan
Curtis, Rinaldo Alessandrini, and Anthony Rooley) all demonstrate, the early-
music community itself argues over Baroque singing. Reconstructing historical
singing styles has proved more challenging than reconstructing instrumental
styles. We have more or the same kinds of vocal cords and larynxes that
less

Baroque singers had, but our singers don't necessarily use them in the same
ways; and our vocal priorities can be quite different from those of past eras.

Unfortunately, it didn't occur to singers or critics in earlier centuries to make

their priorities clear to us. And in the field of singing, far more than in the

world of the harpsichord or the gamba, modern mainstream style weighs heav-
ily upon us. It's not necessarily easy for a cellist to adjust to limiting their vi-
brato, but for a singer the effort can be far more difficult. Will Crutchfield
writes, "Most pedagogues agree that a fully developed singer cannot continu-

225
226 THE BAROQUE

ously suppress his customary vibrato without some kind of unhealthy tension
of the vocal mechanism."'
The debates about Baroque singing also reflect an aspect of the Baroque it-

self: there was more than one style of singing. In the following four chapters,

we look at the various schools from different angles. William Christie explains

the French style; John Butt, Philippe Herreweghe and Jeffrey Thomas look at

the German, as represented in Bach. But all agree that the style that dominated
the era was the Italian. In the next chapter, Julianne Baird proves a knowl-
edgeable guide to Italian style and to its impact outside Italy. In the chapter fol-
lowing hers, Nicholas McGegan discusses Handel, a Saxon who spent his ma-
ture years in London, but whose vocal writing was essentially Italian.

It might surprise some to learn that Julianne Baird, the American Baroque spe-
cialist, was attracted to singing because she loved Maria Callas and Mirella
Freni in verismo opera. It might not surprise Baird's admirers. Tom Moore, for

example, praises her in the same terms used for Callas: he writes of Baird's "un-

canny ability to connect with the emotional core of the music she sings, so that
every thought or mood of the poetry finds a compelling musical response, the
phrase hastening or alighting on a word made special, the voice sighing,
"^
dying . . .

Baird's soprano voice is far from the verismo type, and its vernal beauty
blends perfectly with early instruments. As with Callas, though, it's what she
does with the voice that makes her work matter. In Will Crutchfield's judgment,
"hers is the most perfect integration of style and sense, of period practices and
dramatic persuasion, that I have yet encountered in Baroque opera." Crutch-
field, an authority on vocal ornamentation, praises her as one of the few singers
to have mastered the art of decorating extempore, again emphasizing the emo-
tional basis of Baird's art: "It is not just that the ornaments are wonderfully
apt, stylish and fluent, but that they are dramatic. Not, as one sees all too often,
that they have been tagged to some obvious dramatic concept . . . but rather
that Baird herself is charged with an emotional certainty that floods through
the theatre, and the ornamentation flows out as an unselfconscious means of
expressing the feeling."'
Baird herself has said that she prefers "a more personal, more dramatic ap-
proach" that "tries to express Ithe character she is portraying] in the music."'*
Not that she neglects scholarship. Baird has a doctorate in music history from
Stanford and holds a tenured faculty position at Rutgers. When we spoke on

1. Crutchfield, "Voices," in Performance Practice: Music after 1600, ed. H. M. Brown and
S. Sadie (London: MacmiUan, 1989, and New York: Norton, 1990), p. 295.
2. Moore, Fanfare 16 (November/December 1992), p. 410.
3. "Handel Opera: A Tercentenary Report from New York," Early Music 14 (February
1986), p. 149.
4. In Octavio Roca, "A New Authenticity," Gramophone 68 (January 1991), p. 1351.
 BEYOND THE BEAUTIFUL PEARL 227

a sunny July day in a church in CaHfornia's Marin County she was singing —
Bach that night —she was engaged in preparing her doctoral thesis for book
publication/ It deals with two major eighteenth-century singing masters, the
Italian Tosi and his German translator and commentator, Agricola, the latter of
whom she made available in English for the first time. Tosi, in an attitude typ-

ical of his era, defended the singer's right and even obligation to embellish
that is, to create part of the music. Agricola, however, takes an attitude that,
as we've seen, has become the modern orthodoxy: he insists on the primacy of
the composer and the score. (As this shows, the "composer's intention" was by
no means a brand-new idea in the nineteenth century.) Though Baird has ob-
viously learned a great deal from Agricola, it is to Tosi's philosophy that she

subscribes. In doing so, she challenges the essence of today's text-based per-
formance theory, and I took up that point early in our conversation.

You've told me that Callas and Freni inspired you to become a singer. What at-

tracted you to them?

I admired the assurance of Freni's impeccable technique and the sheer
beauty of her voice. Of course, Callas's voice and technique could hardly have
been more different. To me, what was especially impressive in Callas was an
absolute fearlessness that let her overcome vocal faults by her sheer drama and
passion.

But wouldn't the modern approach to their repertoire the idea that we —
—
sing just what the composer wrote conflict with some of what you now do:
specifically, improvising ornaments?
But that is exactly what singers were still doing in the early Romantic era
especially among the Italians, where singer-dominated traditions held on from
the high Baroque. In Verdi's day, as in earlier times, performances were often

a rather "seat of the pants" affair, with the music geared to whoever was avail-

able to sing. When Verdi, for example, was involved in a production he, like
his predecessors, made alterations to accommodate the voices available. From
Handel to Verdi, composers didn't dare fashion arias without intimate knowl-
edge of the abilities, both technical and dramatic, of the singers waiting to be
fitted. Almost like tailors —
to whom they sometimes compared themselves
composers made alterations to reveal the strengths and hide the faults of the
singer the aria was being written for. At least in the early part of Verdi's ca-

reer, the primacy of the singer was still respected.

5. Introduction to the Art of Singing, by Johann Friedrich Agricola, trans, and ed. Julianne
C. Baird (Cambridge University Press, 1995). Agricola's Anleitung zur Singekunst (1757) is a

translation of and commentary on Pier-Francesco Tosi's Opinioni de' cantori antichi e mod-
erni (1723), the most influential treatise on singing in the first half of the eighteenth century.
Baird's version is cited in the following notes as "Agricola/Baird."
228 THE BAROQUE

When did this start to change?

During the middle of Verdi's career —the middle of the nineteenth century.
The usual contractual arrangement with the opera house at the beginning of
Verdi's career required the composer, like his predecessors in the Baroque era,

to be available for the first few performances. He was to be in the pit, to be

available at the keyboard, to help direct, and to give various kinds of assis-

tance, such as page turning. Only after the first few performances did he get
paid for his work. Following the publication of his text, which occurred usu-
ally around the time of the first performance, Verdi frequently lost control

over what the singers did with his music. Without his presence in the pit and
his availability to tailor the arias to individual requirements, singers refash-

ioned them to their own needs and whims, sometimes (to Verdi's dismay) even
inserting passages with no basis whatsoever in the published texts —such as
arias created by other composers!
To see the situation from the singers' point of view, they were merely ex-
ercising their prerogative, their right, their obligation to embellish so as to
emphasize their own virtuosity —to produce the most stunning performance
possible and to ensure the success of the opera.
The modern idea of the sanctity of the composer's score simply didn't exist
in the Baroque tradition, especially among
Music was a living
the Italians.
art, created anew by was not a dead arti-
the singer in each performance. It

fact. When the composer in the Italian tradition set down his directions on

paper, he was not killing possibilities: he was creating possibilities. He was

providing a sort of blueprint for an artist who would then realize the music
in his or her own creative way.
That what the Baroque era was all about: uplifting not just the cre-
is

ativity of the composer but also the creativity of the singer, honoring what

an individual might bring to something that was already partly fashioned. If

one thinks of written-down music as a blueprint, the singer or instrumental-
ist was a co-designer; this is a very sensible and human approach. These days,
the conductor is entrusted with the "re-creation" of the music, and the singer
is relegated to one category or, in the widely used German term, one Fach.
Even though forensic medicine and criminology recognize the individuality of
voices in voice prints, the prevailing attitude in opera is that the singer must
fit into a Fach. If the coloratura Fach includes the roles of Zerbinetta, Con-
stanze, the Queen of the Night, and several others, God forbid that the singer
be anything other than just the perfect tomato for each of those boxes.

One result today is that many opera singers feel compelled to take on
"big" roles that do not suit their voices, and this contributes to the voice's
early demise.
Exactly. This development, of making the score primary and suiting the
singer to it, is rather recent. For example, in remounting an opera in Paris,
 BEYOND THE BEAUTIFUL PEARL 229

Verdi had to adapt the tenor role for the star tenor of the Paris Opera, who
had damaged his voice by trying to sing full-voice high Cs. Because the tenor
had lost all of his range above the staff, Verdi transposed or rewrote the part
to eliminate every note above the staff.

How would you explain the rising power of the composer and reverence
for the score?
It comes largely from the German philosophy of singing that had its roots
among such personages as Frederick the Great, who forbade added orna-
mentation among his singers. Gluck and other musicians and theorists of this

persuasion (Agricoia, for example) were offended by the extravagances of the

all-powerful virtuoso singers, and sought to rein them in and effectually de-

prive them of their creative spontaneity. The insistence on Urtext tthe the

score as the composer wrote it, free of any editor's additions] reminds me of
what we learn from nineteenth-century German philology, where the term was
coined. Scholars were concerned with establishing valid original texts for the
written-down epics, sagas, and chansons that are believed to have originated
in the oral tradition. Once committed to paper and then rigidified as "the

text," an oral tradition is dead in its tracks. It has become a written tradi-
tion. A vital, creative singing tradition is somewhat similar to the oral tradi-

tion of literature in that each performance is something of an original cre-

ation.
In the course of his career, Verdi became more and more powerful, more
able to control what singers did with the score.* By the end of the Roman-
tic power had shifted almost entirely to the composer's
period, the balance of
corner, but this arrangement still worked so long as the composer fitted his
music to the individual singer. In the twentieth century with its atonal com- —
positions, which discourage both singers and audiences — has arisen a predilec-
tion for the performance of music of earlier eras and the unhappy situation

in which a singer is forced to fit his or her body into the body (the larynx)

of another person in a role that was tailored specifically for that other per-
son. The singer can be judged harshly for not being another person. That is

the problem.
This textual rigidity diminishes music as a living art; it denies singers the

full power of their creative potential; it pigeonholes them; it creates vocal ath-
letes; and sometimes it even ruins singers' voices. I am a staunch advocate of

singers' rights and of a more sensitive and, I think, effective relationship be-

tween composer and performer.

This is why I subscribe to the Italian philosophy. This is why I chose the
Baroque era.

6. As Will Crutchfield has shown, though, until the end of his life Verdi expected the
singers to add ornamentation of a characteristic sort. For details, see "Vocal Ornamentation
in Verdi: the Phonographic Evidence," in 19th Century Music 7 (1983-4), pp. 3-54.
230 THE BAROQUE

Let me ask you for some details about Italian Baroque singing. Is it true
that Italian vocal writing changedfrom the early Baroque the Monteverdi pe- —
—
riod with its ideal of the music following the words very closely, to the mid-
dle Baroque, which put more emphasis on melody, and to the late Baroque,
when ornamental display becomes most important?
I once had the opportunity to look at five Itahan Baroque recitatives setting

the same text. The later the setting, the more formulaic it was, and the less in-
teresting harmonically and melodically. The text of the recitative was very pow-
erful: "Oh Heaven! In one instant, I have lost my country, my family, and my
father." In the earliest setting, pre- 1700, the anguish of the text is represented
both in the harmony and in the leaps over dissonant intervals and appoggiat-
uras in the melody. The later settings have much less poignancy. There are
fewer large melodic leaps, fewer dissonant intervals: it is a flatter, more for-

mulaic workaday style. The vocal part contains mostly consonant leaps, and
the harmony is nondescript; it almost sounds as if the text could be, "Oh Heav-
ens, I've lost my hairbrush."
In later opera composition (about were many conven-
1720 to 1750) there
tions —opera-goers presented with the would recognize the
first line of text
chief sentiment or affect of the aria. Examples might be the aria di bravura and
the aria di pathetica. Some arias were structured around specific metaphors,
such as nightingale arias or shipwreck arias
— "Poor me, I'm like a little vessel

that gets tossed from one wave to another; I'm in love with two people, and I

can't decide between this port and that port." The audience really wouldn't pay
attention to the text any more; they were satisfied with knowing the general
affect.

Did this loss of interest in the words reflect in part an increasing priority
placed on showing off the singers' virtuosity?
Yes. In fact, as singers gained in prominence, librettists began to be severely
limited as to the words they could use. By the late eighteenth century the promi-
nent librettist Pietro Metastasio complained^ that had
in his time, the librettist

become two vowels, a I"ah"] and o l"oh"], for the final vowels of
limited to
lines set to coloratura passages —
and that, eventually, even the latter vowel had
been eliminated. Thus, all the final syllables had to be a, which limited the
rhyme and thus the choice of words.

Early Baroque Italian singing was genuinely text-centered, but it also in-
volved very florid embellishment. Robert Greenlee wrote that "when performed
with the vocal techniques commonly taught today, these embellishments are un-
wieldy, often resulting in phlegmatic tempos and awkward phrasing."^ Could

7. Charles Burney, Memoirs of the Life and Writings of Abate Metastasio, 3 vols. (Lon-
don, 1796; reprint, New York: Da Capo, 1971), vol. 2, pp. 135-36.
8. Greenlee, "Dispositione di voce: Passage to Florid Singing," Early Music 15 (February
1987), p. 47.
 BEYOND THE BEAUTIFUL PEARL 231

you talk about the technique that allowed Italian singers of the early Baroque
to sing these tricky passages?
In the period from 1580 to 1640 or so, dominated by such composers as

Monteverdi and Caccini, Italian singers negotiated the fast passages with a
Hghtning-quick kind of glottal articulation, which was performed on the soft
palate. The air percussed against the soft palate the way it does in a giggle. Sev-

eral treatise writers called this technique the dispositione di voce or "disposi-

tion of the voice," and said it gave the voice the ability to move quickly or with
agility."

In vocal writing, these fast notes were alternated with long-held notes. Flex-
ibility in the alternation of the fast notes with the sustained notes, in passages,*"
was considered the essence of good singing throughout the seventeenth and
eighteenth centuries. A singer was considered accomplished if he could exhibit
both types on every note in his range: that is, to sustain a note while making
a messa di voce (crescendo-decrescendo on a long-held note) and to trill (to ef-

fect a rapid alternation between two notes that are one full tone apart)." When
Roger North, a contemporary of Pepys, suggests that one begin instruction on
the viol by playing long-held notes with a messa di voce, I believe he was im-
itating singing instruction. This way the voice is kept in balance, and main-
tained in health.

In contrast to that glottal articulation, singers nowadays are often taught to

treat the throat as if it were a hollow tube through which air passes. When did
the change in attitude occur?
The old style of glottal articulation became unpopular when, in the late sev-

enteenth century and the eighteenth century, performance moved out of

princely chambers and private rooms and into larger spaces such as the opera

houses and concert halls. Glottal articulation became unpopular because it is

difficult to hear at the back of a larger space, even one that holds just 500 peo-
ple. During the high Baroque (1700-1750), in every nation, singers had largely
discarded this technique. By 1723, it was derisively called the sgagateata (cack-

ling like a chicken). Agricola, a student of Bach's on whom I did my doctoral

thesis, explains the acoustical factor behind the demise of glottal articulation.
Agricola recommended instead a detached articulation of passages, or battuta,
as Tosi names it. (The term is not commonly used by modern musicians.) Ac-

cording to the 1757 translation of Tosi by Agricola (who rendered this term

9. Ibid, pp. 47-55.

10. "Passages" here refers to "fast-moving, stereotyped melodic formulas" substituted for
longer notes or groups of notes (as opposed to "specific ornaments applied to single notes"):
also called diminutions. The definitions come from Howard Mayer Brown's Embellishing
16th-century Music (Oxford University Press, 1976), p. 1.

11. The trill is ornament to master; it was highly prized in the Italian Baroque
a difficult
but is rarely emphasized in modern voice instruction. See Julianne Baird, "An 18th-century
Controversy about the Trill: Mancini v. Manfredini," Early Music 15 (February 1987), pp.
36-45.
232 THE BAROQUE

gestossen), it involves re-articulating the vowel on each note. To avoid the ef-
fect of giggling, detached articulation required a diaphragmatic impetus for
each note, together with a light throat articulation for the sake of clarity. This
style of articulation, Tosi's battuta, was used for most passages of the Italian

and German high Baroque.

What about what we think of as Italian style now, which joins the notes in
a legato?
The smoother articulation of the legato, which was discussed by both Tosi
and Agricola, not to mention other Baroque writers, was far less common in
the seventeenth and eighteenth centuries than it is now. In this type of articu-
lation (according to Agricola), the singer articulated the vowel only once, at the

very beginning of the passage or division.'^ The legato was typically employed
in pieces in the "pathetic" style, pieces with an affect that was sad, tender, ex-
pressive or nobly serious.'^

So was articulation more marked in Baroque singing, generally, than today?

While that is true, it should also be noted that Tosi enthusiastically praises
slurred articulation or the "drag," for which he uses the term strascino (the

Germans Das Ziehen). The technique is like a glissando.^"* It did not

called it

appear suddenly in the high Baroque. It was already being employed in the
works of Monteverdi and D'India, particularly upon affective words like pi-
ango, "I cry," or lagrime, "tears." It is often notated on a rising note for ex- —
ample, on a B\) going to a B with a slur mark connecting the pitches. In the
drag, the singer was expected to slide by almost imperceptible increments from
one note to the next, touching all the pitches in between.

How does this relate to what we call portamento? Wasn't portamento a part
of the Italian cantabile style of singing in Mozart's era, too? Didn't Corri dis-

cuss it as well?
Portamento is a more difficult concept to explain. Tosi's phrase "porta-
mento di voce" literally means "carriage of the voice," and Hiller uses the term
to mean "good use of the voice." Corri, a student of Porpora, for example,
calls it "perfection of vocal music" and "the refinement in elegant pronuncia-
tion in speaking." In general usage, portamento refers simply to the process of
"singing well.""

But wasn't portamento more specific than that?

12. "Divisions" decorate a melody by replacing each long note with livelier melodic move-
ment in shorter note values.
13. See Agricola/Baird, p. 196.
14. Ibid, p. 279. Baird says that there is no good English equivalent of strascino, and that
she adopted "the drag" from John Galliard's 1742 English translation of Tosi, Observations
on the Florid Song.
15. Agricola/Baird, p. 271.
 BEYOND THE BEAUTIFUL PEARL 233

Yes, in fact it was. Corri is getting to the essence of the concept when he
says that the "good use of the voice" involves the "sHding and blending [of]
one note into another with delicacy and expression." To improve portamento,
Mancini provides exercises designed to "connect the voice from one note to the
next with perfect proportion and union in ascending and descending motion
alike." He warns that no "unpleasant slide or dragging through smaller inter-

vals" must be heard, so that the singer brays or howls. Mancini says that when
the singer attacks the tones too strongly and pushes them forward, because his

chest is too weak to sustain the tones evenly, he brays: when he produces an
intermediate note that does not have a harmonic relationship to the notes on
either side of it, he howls.
Students were encouraged to work on the portamento only after they had
mastered music reading and eliminated the defect of "singing through the nose."

I've heard it said that there was a continuity of the Italian singing tradition
from Caccini to Rossini.^*" Do you think there was a continuity of the bel canto
style over the centuries?
A few years ago I saw a televised La Scala production of La Cenerentola
with Frederica von Stade.'^ In that production, two native Italian male opera
singers used glottal articulation when performing the fast passages —the battuta
of Tosi or gestossen of Agricola. I remember wondering at the time about an
unbroken tradition of bel canto teaching in Italy, particularly because glottal
singing is usually discouraged today in the USA. The music of BeUini, Donizetti,
and early Verdi certainly required agility and a good trill (as did eighteenth-
century music —which might indicate continuity of technique). In Italian

Baroque singing, the trill was stressed as not only as the hallmark of agility,

but also as the source of the agility — one practices the fast notes in order to

get the trill to happen more evenly, and vice versa. '^ And usually a light voice
production was also a component.

16. The from Caccini to Rossini is discussed in an evenhanded way in

issue of continuity

John of Italian Opera (Cambridge University Press, 1984) pp. 103-4. He

Rosselli's Singers

cites a 1968 doctoral dissertation by E. V. Foreman at the University of Illinois, "A Compar-

ison of Selected Italian Vocal Tutors of the Period from 1550 to 1800," as arguing for such
a continuity, but he neither endorses or dismisses Foreman's conclusion.
According to Owen Jander, the term bel canto was not used until the mid-nineteenth cen-
tury, and referred not only to a beautiful voice but to "effortless delivery of highly florid
music," as opposed to the more stentorian verismo style that followed. (See his article "Bel
Canto," in The New Grove Dictionary of Opera, vol. 1, ed. Stanley Sadie (London: Macmil-
lan, 1992].) It is anachronistic, Jander says, to apply the term to mid-seventeenth-century lyri-
cal singing (in, say, Carissimi) that reacted against the "representative" style of, say, Mon-
teverdi; I do so with that disclaimer. Also, the early-nineteenth-century Italian style was
distinctfrom earlier singing in at least some known respects. For instance, Manuel Garcia II

observed in 1840 that a lowered larynx position was a new development in his generation; it

has since become standard operatic technique.

17. The laser disc is available on Deutsche Grammophon 072502.
18. Agricola/Baird, chap. 3.
234 THE BAROQUE

Unlike today, where we value warm, rich timbre and sheer size of voice
more than anything else.

Listening to the recordings of Galli-Curci, Jenny Lind (a Manuel Garcia stu-

dent), and other famous singers from the early twentieth century, one is struck
by their light production, much like that of today's early-music singers.

Was it the Wagnerian influence that changed things?

Perhaps, but I think it was more the size and acoustics of larger performance
areas such as today's opera houses, and the decibel levels of today's orchestras.
It's much like the conditions that caused the changes in use of glottal articula-
tion in the Baroque. The old Met was a horseshoe-shaped theater with excel-
lent acoustics; it was like an Italian opera house, easy to sing in and project in,

much better than huge rectangular halls. Now we have halls like the 3,300-seat
San Francisco Opera [built in 1934] and the 3,800-seat Met [built in 1966].

And we have louder orchestras, with instruments (like steel-string violins and
larger-bore brass) that can produce greater volumes. Many voices break down
early because of the stress caused by singing at the volume necessary to be
heard in such conditions."
Today, in some operatic circles, few words are so damning as "small voice";
but the "large voice" has not always been needed or even relevant, and attempts
to develop or keep on using a large voice can sometimes ruin it. One might com-
pare the voice to an earthquake-proof building, with a lot of architectural give
and take so that it adjusts to seismic shocks. The vibrato enables the voice to
sustain a volume that's audible in a huge hall over a loud orchestra. In these cir-

cumstances, singers need to rely on a wide vibrato to avoid hurting the voice.

This brings up a much-debated question: how much was vibrato used in

Baroque singing?
Vibrato was and is used as part of the singer's sound vocabulary. It is dif-

ficult for a singer to execute a messa di voce or crescendo without some vi-

brato at the top. One indication that vibrato was used are the many organ stops
called the vox humana, which always had vibrato (the earliest are in fifteenth-

century Spain). ^°
Whether or not vibrato was employed constantly by seventeenth- and eigh-

19.Another factor demanding more vocal volume is that conductors have moved up in
the operatic pecking order, and often want more of the limelight in performance. Many have
their orchestras play much louder than previous generations of conductors did, making it even
harder for singers to be heard.
20. The leading study of Baroque vibrato is Greta Moens-Haenen's Das Vibrato in der
Musik des Barocks: Ein Handbuch zur AuffUhrungspraxis fiir Vokalisten und Instrumentalis-
ten (Graz: Akademische Druck- u. Verlagsanstalt, 1988). It is not available in English trans-
lation; but Moens-Haenen's entry on "Vibrato" in the New Grove Dictionary of Opera, vol.

4, pp. 982-84, makes points similar to Baird's. Eighteenth-century opera singers seem to have
had a basic vibrato, but it appears to have been small and almost inaudible, allowing for the
 BEYOND THE BEAUTIFUL PEARL 235

teenth-century singers is a different question. The writings of the eighteenth

century indicate an aesthetic in which a singer was most prized for the abiHty
to change his or her sound to express the words. By contrast, many teachers
today promote an even vibrato with an equal width and speed on each note,
to create a series of "beautiful pearls." You make your beautiful pearl, and it

doesn't matter what the words are.

Many opera singers today don't do much with varying their sound and color
to suit the words — though Callas did, of course, as well as Fischer-Dieskau and
Schwarzkopf. You've praised Fischer-Dieskau and Victoria de los Angeles for
their use of non-vibrato.^^
Yes. One of the things that made Callas's singing so exciting was that she
wasn't afraid to make an ugly sound. While that may sound like a back-handed
compliment, I applaud her willingness to go to great lengths to portray her
characters dramatically and to move the audience. For her, it was most impor-
tant to play the part, rather than to produce a "pearl." Vocal beauty as such
should not always be prized; the "pearl" is not always appropriate.

How does this "beautiful pearl" issue apply to the use of vibrato
There are places —on a dissonant note, a leading tone, chromatic tone, or a

a tonus diabolus (an augmented fourth) — which are better sung without vi-

brato, because the vibrato softens the effect, makes it a beautiful pearl, as if

application of more vibrato (though still narrow) as an ornament, and the suppression of vi-

brato as an effect.

Interestingly, Moens-Haenen concludes that "vibrato was used less often in early Roman-
tic opera than in the opera of the late 18th century"; singers still used it as an ornament.The
increased size of orchestras and opera houses and the development of the verismo and Wag-
nerian styles of singing led to increased use of vibrato over the century. Continuous vibrato,
however, is "a 20th-century phenomenon," she says; the same can be said of the wide, slow
continuous vibrato that has come to signify "operatic singing" for many listeners.

Robert Philip points to one bit of evidence for this last assertion — early recordings, in
which "distinguished singers" use a vibrato that is usually "too fast and too shallow to be
perceived as a fluctuation in pitch. What is clear from recordings is that many singers of the
early years of the century used a shallower vibrato than singers later in the century" (Philip,
"1900-1940," in Performance Practice: Music after 1600, ed. Brown and Sadie, p. 477). In
the same book, p. 453, Will Crutchfield has a good deal to say about these recordings; he
mentions Adelina Patti, an adherent of the older bel cantu style, who was praised in 1886 for
"her judicious 'refusal to sing tremolos' in spite of the growing [verismo] vogue for them."
Recordings show that her vibrato was indeed minimal.
In the opposite historical direction and John Butt finds that the sur-
in a different country,

viving German Baroque documents was used

as an ornament rather than
suggest that vibrato
something continuous: see his Music Education and the Art of Performance in the German
Baroque (Cambridge University Press, 1994), pp. 70, 138, and 144.
21. "Fischer-Dieskau commonly uses non-vibrato, and we just think of him as an extremely
expressive singer without examining why he does it. And Victoria de los Angeles is the same."
Baird, quoted in Roca, "A New Authenticity."
236 THE BAROQUE

the singer is oblivious to the affect of the words. The modern aesthetic is very
different from that of the Baroque. Another example is the appoggiatura. Three
rules apply to the appoggiatura: it is longer than the main note, louder than
^^
the main note, and performed with less vibrato.

Does the appoggiatura get less vibrato because it's dissonant? Wasn't a mo-
mentary increase of vibrato itself used as an ornament? If so, how and where?
Some good examples in the music of Bach are the soprano aria "Zerfliesse,
mein Herze" in the St. John Passion —which involves Bebung a (vibrato), indi-

cated with a long chevron sign on the word "Tod" —and similar passage on a

"timentibus ejus" from the alto-tenor duet "Esurientes" in the Magnificat.

Also, many Italian composers from around 1580 to 1630 employed an or-

nament called the trillo, a type of intensity vibrato. ^^ Itwas performed both
quickly in glottal fashion and slowly von der Brust (in the chest), or from the
diaphragm.

How about the width and speed of the vibrato?

Early-twentieth-century studies showed that the average speed of vibrato in
modern singers is around six or seven cycles per second. ""* Vibrato speeds were
probably not very different in the seventeenth and eighteenth centuries — al-

though I do think there is a correlation between speed in smging divisions in

rapid passages and speed of vibrato, and rapid passagework was far more im-
portant in Baroque music than in Wagner or Puccini. But as I suggested, the
width of today's vibrato, sometimes approaching a minor third, is quite another
matter. It is a product of volume and stress or pressure on the voice.

Another technical question is how much singers blended the head and chest
registers in the Italian Baroque style. Is it true that they distinguished the

11. Agricola/Baird, pp. 92-93.

23. "The trillo is a tone repetition which can be the distinct (almost staccato) articulation
more legato pulsing on the same note" (Agricola/Baird, p. 246). There
of one note or can be a
issome controversy about this: see David Fuller, "The Performer as Composer," in Perfor-
mance Practice: Music after 1600, ed. Brown and Sadie, p. 124, and Brown, Embellishing
ISth-Century Music, p. 10.
24. A number of these studies are collected in Carl Seashore's The Vibrato (Iowa City: Uni-
versity ofIowa Press, 1932). Studies of famous singers of the first part of the twentieth cen-
tury found that the average speed of vibrato was about 6.5 cycles per second. The speed could
vary, however, both for an individual singer and between singers. The most extreme case was
that of Giovanni Martinelli, whose vibrato ranged from 5.5 to 12.5 cycles per second; but he
too averaged around 7 cps. Vibrato width averaged about a half-step, though it could vary
from a quarter-step to (rarely) more than a whole step. The minor third that Baird mentions
was very rarely approached by the earlier generation. Among these singers, 95 percent of their
tones had some vibrato; these studies found that among untrained adults only about 20 per-
cent had a vibrato. This last finding shows that vibrato is not as "natural" as some claim.
(The further assertion that it is impossible to sing without vibrato is misleading: it is based
on a redefinition of "vibrato" to refer to the vibrating of the vocal cords, but in common
usage "vibrato" refers to the sound as heard, not to its source.)
 BEYOND THE BEAUTIFUL PEARL 237

sounds of the different registers, instead of trying to minimize the differences

as modern singers do?
The Italians recognized two
head and chest, with the falsetto
registers, the

evidently included in the head voice. The Germans, such as Agricola, distin-
guished the falsetto from the head and chest voices. Blending the registers
around the break between them is mentioned in singing treatises of every era.
The singer is encouraged to practice certain notes around the break both in
chest voice and in head voice, in order to make the transition between them
gracefully. When the chest voice is extended into the domain of the head voice

a sound occurs that is the same as belting in Broadway singing. Eighteenth-cen-

tury writers discouraged the singer from extending the chest voice too high be-
cause of possible damage to the voice.

It's said that eighteenth-century singers were encouraged to not belt out the
high notes, but to hit them lightly, which is the opposite of what we admire
now.
That's absolutely true. Tosi said, "Let the singing master be careful not to
let the student who's seeking to attain the high notes shout them or scream
them." He adds, "... lest he not only lose his soprano voice ..."

. . . as happened to Verdi's tenor . . .

Right, "... but also lose his health."''' One may ask, what does singing
loud high notes have to do with general health.^ There is, however, an account
of a castrate singer who died of a hemorrhage from singing too loud in the
high register.^*"

Which, however, indicates that it was done.

was done. Not every singer is aware of his or her limits. There are other
It

examples. The treatises admonished singers not to memorize cadenzas they —

were supposed to listen to those of the best singers, study composition, and im-
provise in the style of the aria. But some of the most famous male and female
singers of the day had collections of cadenzas, and probably drew from them,
mixing and matching, so to speak. Perhaps they operated somewhat like a good
jazz musician, who can begin with various standard riffs.

Similarly, the books warned against memorizing and simply reusing the
same ornaments in a da capo.-^ But in the memoirs of Madame Mara, Eliza-

beth-Gertrud Schmelling, we learn that she prepared by writing many different

25. Tosi, Opinioni de' cantori antichi e moderni, chap. 1, section 9. The translation is an
informal one by Baird.
26. Agricola/Baird, p. 263, n. 37, tells of the strange end of Luca Fabbris in 1765.
27. The da capo form, a common high Baroque form (it is used often in Handel and Bach
arias), has an elaborate first ("A") section, followed by a contrasting second ("B") section in

and then repeats the A section. The second appearance of the A section
a related key, is called
the da capo (lit., "from the beginning"); it was used to demonstrate a singer's skill at orna-
mentation.
238 THE BAROQUE

ornamentations, from which she would pick and choose. This gave the im-
pression of improvisation even if the audience heard repeat performances. Mara
^**
attributes her triumph over her rival, Brigida Banti, to this trick.

Your mention of the sopranos brings up another topic I wanted to mention.

Wasn't it in the Italian Baroque that women singers first emerged as star

soloists?
Yes. Not that their paths were always lined with roses. Many married their
impresarios —managers— in order to avoid gossip about their personal lives, or
traveled with their mothers. And a great number died in poverty.^**

To return to singing style —and interpretative freedom —how was rubato

used in Italian Baroque singing?
One of the first mentions of the rubato is in Tosi's treatise. He specifies two
kinds of rubato, one in which time is lost in order to be gained, and the other
in which it is gained in order to be lost. Our first response is to wonder what
he could mean by that; I think it means one type was rushed and then dragged,
the other dragged and then rushed. The point is that the rubato is not a slow-
ing down of a whole section, only to have the tempo resume later; the bass
stays pretty much in the original tempo, and the bass and melody come to-
gether eventually on a downbeat.

It sounds like what is said of the piano playing of Mozart and Chopin —
both of whose approaches to melody were clearly derived from Italian operatic
singing.
Yes, the rubato clearly did find its way into keyboard playing. For an ear-
lier example, Carl Philipp Emanuel Bach in his Sechs Sonaten mit verdnderten
Reprisen (Six Sonatas with Varied Repeats) writes out this kind of rubato, giv-
ing sometimes seventeen in the right hand against four in the left, or twenty-
three against six. Similarly, Mozart writes it out in the Rondo in A Minor, K.
511. In both cases, it's clear that the right hand plays more freely while the left
is more steady.
This is one of the reasons why a singer was taught to accompany himself:^"
so that the right hand can be free while the left hand stays steady. Rubato was
one of the things a singer was expected to learn. Skill in using rubato was an-
other little tick on the chart as to whether the singer was triumphing that night
or failing.

I'd like to ask you about Italian influences in England and Germany.
You've written that English composers kept to a simpler, less florid style,

28. Agricola/Baird, p. 28.

29. See Rosselli, Singers of Italian Opera, chap. 3.
30. Agricola/Baird, p. 252.
 BEYOND THE BEAUTIFUL PEARL 239

where the words were paramount, until the Italian style took over in the sev-
'
enteenth century.
Yes. For example, in many pieces by Campion or Dowland, the word set-

tings, involving one syllable per note, makes the music subservient to the text
and makes it much more difficult to sing. Campion was of course a poet, and
he wrote even his second and later stanzas so that they would fit the music.
But once the English fell under the Italian influence, especially with Nicholas
Lanier [1588-1666] and Robert Johnson [1583-1633], we start to see a slow-

ing down of the harmonic movement. Instead of a chord every beat in a J piece,

they use a chord every bar, or every two bars. So the harmonic movement gets

much slower, and this means that the singers get much more active with a great

deal of florid expression of the text —sometimes, curiously, on those words

where you would least expect it.

Suppose, for example, you have the phrase "loss of breath." You might
think that the way to make a joke would be a
on it florid run on the word
"loss," so that the singer almost runs out of breath on it. But, in fact, what we
typically find in English music of this period is that the coloratura passage is

on the word "breath."

Because a florid run on an insignificant word turns out to help make the
text more understandable, it helps listeners get a double entendre. It's a little

hat-tipping to the ideal of understanding the words. It doesn't fulfill that ideal
completely, but it does help. You'll find it also in Johnson's Care-charming
Sleep., where long elaborate ornaments occur on very insignificant words, like

"in," "of," and "to."

By contrast, among Italian composers the ornamentation is on poignant
words. But the result is that the minute the singer starts the wild ornamenta-
tion, it becomes hard to hear the text. Although in theory the ornamentation
is meant to develop the meaning of the word it occurs on, in practice the lis-

tener misses both that word and the point of the ornamentation. You hear the

notes but you don't understand their function. So the English didn't simply ape
the Italian manner.

You mentioned that as the line got more florid, the harmonic rhythm slowed
down. Could you say more about the connection between harmonic rhythm
and floridity?
Regarding that, I think of the most influential composer in England, the Ital-

ianized German Handel, and of the Italian singers who came to London to sing
his music. Some of them were annoyed because his instrumental parts had too

much busy-ness, whereas in Italy it was often the singers who provided all the
melodic interest, perhaps with instrumental doubling. Handel's writing for the
accompanying instruments was far more complex than what Italian singers

would have received from Galuppi or Torelli. And the more complex and
31. Booklet note, Dorian CD 90109, The English Lute Song.
240 THE BAROQUE

rapidly changing harmony makes it harder for the singer to ornament —you
have to really know what the bass is doing. You can't assume anything.
When Handel wrote a strictly continuo aria, that was one of his highest

compliments to a singer, because then the singer was completely on her own.
By the way, I'm always a little piqued when I get a Handel aria that has the
violins in unison with me. My first thought is, "The soprano must not have
been very good." But in fact, it wasn't necessarily a slur on the singer. Quite
often the Italian opera orchestras were brilliant in their ability to do
chiaroscuro^^ — a hundred violins may have been accompanying a singer, but
when the singer started soloing the violinists knew how to cut back so far as
to be almost an ambience around the singer, to support without competing. But
that's a concept that most modern orchestras don't have, and they often have
to be told over and over again to cut back.

In eighteenth-century Germany and England, Italian singers were hot com-

modities.
This is still true. Agricola writes that there were probably ten or fifteen good
German singers who left Germany to make their careers in other countries,

sometimes even changing their names —so great was the prejudice against them
in their own country and the preference for Italians. The Irish singer Michael
Kelly —Mozart's first Don Basilio —changed his name to Michele Ochelli!

It's like our own country now!

That's very true!
When Frederick the Great founded the Berlin Opera, he insisted on having
only Italian singers —which resulted in angry editorials complaining that
Graun^^ had been sent by Frederick to Italy to once again audition Italian
singers for the Berlin opera, and "because of such circumstances, our best per-
formers leave the country in shocking numbers, while wretched Italians are ac-
corded great honor. "^'' And the castrati made so much more money than the
average singer that it's like comparing Pavarotti to a church singer. The Eng-
lish, Spanish, and German courts employed so many castrati and other Italian

singers that when Frederick was getting his opera going in the middle of the
eighteenth century Italy was practically depleted of her best singers. And al-

though Graun returned with second-rate talent, they nonetheless earned far
more than the native German singers.

Dresden had its stable of Italian singers, including the famous Faustina Bor-
doni, Johann Adolph Hasse's wife.

32. Varying the loudness. Agricola/Baird, p. 279, n. 4.

33. Karl HeinrichGraun (1703--1759), Royal Kapellmeister to Frederick the Great, as well
as chiefcomposer to the Berlin Opera.
34. Quoted in Agricola/Baird, p. 254, n. 3.
 BEYOND THE BEAUTIFUL PEARL 241

Yes. By. the way, Agricola and Quantz both have a lot to say about
Faustina's technique, particularly her ability to sing divisions loudly but with
glottal articulation. Agricola notes this as an exception to his ban on glottal ar-

ticulation.

Your main interest regarding singers seems to he to promote the exception —

the not fitting into boxes.
My general thought is that modern music teachers, and I am one, should
respect the particular strengths and weaknesses of the singer standing before us
to be taught. We must resist the pressure to fit the singer into some procrustean
bed inimical to his talents. We must teach him the first principle: Know thy-

self. We must realize that there are important values —wit, ingenuity, feeling,

for example —to strive for, other than the large voice and the plastic sameness
in quality of sound for every note —other than the "beautiful pearl."

SELECTED DISCOGRAPHY
Julianne Baird's relative absence from British Baroque recordings can be puz-
zling. One explanation might be that Newcastle already has so much coal; but

Baird suggests, in an Opera News article by David Patrick Stearns, that another
reason is "the individualistic approach that I take —that rubs them [conductors,
if I read correctly] the wrong way. It's hard to be a singer when you're sup-
posed to stumble from one Svengali figure to another, being molded in their

own image.'"'' Nicholas McGegan, in the next interview, seems to oppose the
"Svengali" approach.
Baird's way with much of the repertory she discusses is preserved on CD
thanks mainly to American independent labels. Dorian, an audiophile company,
has released a number of her CDs. Among the best are two of early Italian

Baroque music: Musica Dolce (Dorian 90123) —the one praised by Tom Moore
in the introduction to this chapter —and
Songs of Love and War (Dorian
90104), in which, says Stearns, her handling of the ornamentation "gives a re-
peating sequence of notes an element of surprise . . . that drives home the ar-
chitectural variety and unity." Regarding her The English Lute Song (Dorian
90109), David Fallows remarks that Baird's "dramatic sense pays rich divi-

dends."'" Nicholas Anderson find her "particularly beguiling" in Handel solo

cantatas (Dorian 90147).'^
She has also recorded for Newport. Her Bach Arias with Flute (Newport
NDP 85530) has, according to Stearns, "great rhetorical variety, color and im-
mediacy." Her Handel Arias (Newport 85530) demonstrates her mastery of the
art of ornamentation; says Stearns, "one doesn't realize how effortful much

35. Stearns, "Baroque Rebel,"Opera News (October, 1995), p. 33.

36. Fallows, Gramophone 67 (April 1990), p. 1858.
37. Anderson, Gramophone 69 (April 1992), p. 126.
242 THE BAROQUE

modern Handel singing is until one hears an almost complete absence of this

quality" in her singing on this CD.

Baird's recorded Handel opera and oratorio roles are mostly on the New-
port and Vox labels. Stanley Sadie objects to the conductor's cuts in Imeneo
(Vox 115451), but praises the "really lovely singing of Julianne Baird, an ex-
ceptionally tasteful artist and in excellent voice. "^^ In Muzio Scevola (Act 3;
Newport 85540) Nicholas Anderson dislikes Baird's vibrato, but calls her char-

acterization "imaginative" and says that she "brings a coquettish sparkle" to

her role.^''

In Alessandro Scarlatti's oratorio Ishmael (Newport 85558/2), her singing

of the final scene, says Stearns, is "one of the most convincing portrayals of
physical exhaustion since Maria Callas's recording of Mimi's farewell in La Bo-
heme." Her "most fully realized opera portrayal," says Stearns, is on Omega
OCD 1016: Pergolesi's La serva padrona, in its century the most celebrated and
influential of opera buffas. Finally, her Lullabies and Dances, with Bill Crofut's
ensemble of folk instruments (Albany 048) is an ideal starter album for young
children, though I'm hardly the only grownup to enjoy it.

FOR FURTHER READING

As far as I know, no good one-volume introduction to Baroque singing is yet

available. I look forward to Sally Sanford's and Baird's chapters in Schirmer's

forthcoming Performer's Guide to Seventeenth-Century Music, edited by Stew-
art Carter. Sanford's article, "A Comparison of French and Italian Singing in
the Seventeenth Century," in the on-line Journal of Seventeenth-Century Music
1 (1995) — its URL is www.sscm.harvard.edu/jscm/vl/nol/sanford.html —con-
veys a great deal of information in an accessible style. Baird's edition of Intro-
duction to the Art of Singing by Johann Friedrich Agricola (Cambridge Uni-
versity Press, 1995), especially her commentary (in footnotes), is a good
entryway into the primary sources. Will Crutchfield's two essays on "Voices"
and David Fuller's essay "The Performer as Composer," in Performance Prac-
tice: Music after 1600, ed. H. M. Brown and S. Sadie (London, Macmillan,
1989, and New York: Norton, 1990) are indispensable. John Rosselli's Singers
of Italian Opera (Cambridge University Press, 1984) is an excellent historical
study of that nation's singing traditions, and John Steane's The Grand Tradi-
tion, 2nd ed. (Portland, Oregon: Amadeus, 1994) is an excellent study of
singers on record, including early recordings.

38. Sadie, Gramophone 71 (August 1993), p. 79.

39. Anderson, Gramophone 70 (March 1993), p. 96.
 13
You Can Never

Be Rignt for All Time

— oOo^ gN

Nicholas McGegan on Handel

Christoph Willibald Gluck, the reformist opera composer, kept a full-length

portrait of Handel by his bed so that it would be saw when
the first thing he
he woke up. Mozart, "Handel
three decades after Handel's death, said that
knows better than any of us what will make an effect." Beethoven called Han-
del "the greatest composer that ever lived," and added, "to him I bow the
knee."'
Handel was the first composer in history whose works never fell out of the
concert repertory. A few of his works, at any rate: in spite of the reverence for
him, only a little of his music was actually performed in the nineteenth century.
Messiah was one of them, of course, sung with increasingly gargantuan cho-
ruses and orchestras, and so were Samson, Israel in Egypt, and Judas Mac-
cabeus. But such masterpieces as Theodora, Giulio Cesare, Jephtha, Orlando,
and L' Allegro, il Penseroso, ed il Moderato occupied few beyond the occasional
scholar.
It was in our century of revivals that these works returned to the stage and
concert hall. The Gottingen Festival in Germany began staging Handel opera in
1920, and over time these pieces (and the less well-known oratorios) have re-

ceived more frequent performances. Since the Second World War, the growth of
the Handel discography has been vastly accelerated by the early-music revival.

1. For a discussion of Handel's posthumous influence, see Ellen T. Harris, "Handel's Ghost:

The Composer's Posthumous Reputation in the Eighteenth Century," in Companion to Con-

temporary Musical Thought, vol. 1, ed. John Paynter et al. (London: Routledge, 1993), pp.
208-25.

243
244 THE BAROQUE

The British conductor Nicholas McGegan has recorded more of Handel's

operas than anyone else, as far as I know —
as I write, at least nine of them,

out of a total of thirty-nine. As artistic director of the Gottingen Festival, he

has helped bring that Handelian mecca into line with current ideas about Han-
del playing. As Music Director of the Philharmonia Baroque Orchestra, he has
fostered its emergence as one of North America's few world-class period-in-
strument orchestras, and the majority of his recordings with them are of Han-
del. He has also recorded Handel with a Hungarian group, the Capella Savaria.

In addition, he leads Sweden's Drottningholm Festival and is principal guest

conductor of the Scottish Opera.

The coincidence of the Handel revival with the historical-performance
movement has raised a number of issues. Consider, for example, the Baroque
convention of the da capo aria, in which a long first ("A") section is sung again
after the shorter second section —
the repeat lets the singers show off their abil-
ity to ornament. Handel singing in recent decades has (in the words of Winton

Dean) "moved from a period when all da capos were literal repeats ... to a

fashion where decoration is allowed to sprout anywhere, even in A sections,

and da capos release salvos of rockets in the style of Rossini or Bellini."^

Doubts also arise about the Anglican purity of much early-music singing: does
it really fit the music that Handel wrote for Italian opera singers? Then there
are the roles Handel wrote for castrati, a voice type whose cultivation was il-
legal in Italy even in its heyday. The castrato voice is, we can rest assured, a

historical instrument that won't be revived; but who, then, should sing the parts
written for it?

A more basic issue is the works themselves. When certain works by a widely

revered composer are almost never played, one might be forgiven for suspect-
ing that these works are of lesser quality. Such suspicions have faded in recent
decades, but doubts about the stageworthiness of the operas persist.

I discussed these and other issues with McGegan in his home office, which
overlooks a scenic canyon from atop the Berkeley hills. McGegan's gift for wit
and vivid language and his congenial manner all somehow fit the subject mat-

ter, for Handel's personality inspired an exceptionally large fund of anecdotes

some of which came up in our conversation.

People sometimes complain that using countertenors for the operatic roles Han-
del wrote for castrati is a sort of spurious authenticity. Handel, they say, al-
ways used a woman for such roles when a castrato wasn't available, because
the voices were more similar: like a castrato, a woman uses both chest and head
registers, whereas a countertenor typically uses only head voice.

2. Dean, "Scholarship and the Handel Revival, 1935-85," in Handel: Tercentenary Col-
lection, ed. Stanley Sadie and Anthony Hicks (London: Macmillan, and Ann Arbor, Michigan:
UMI Press, 1987), pp. 1-18; quote p. 17.
 YOU CAN NEVER BE RIGHT FOR ALL TIME 245

I accompanied the countertenor Paul Esswood in a recording of Schumann's

Dichterliebe, which no one would have thought of singing in countertenor
voice in Schumann's day. What mattered, I think, was that Esswood sings it
very well. In Handel opera, I'm so grateful to find singers (like Drew Minter)
who can hold the stage as well as sing the notes that I don't hesitate to use
them if they're countertenors. To me it doesn't really matter. There are plenty
of people who can sing but can't act, and plenty who can act but can't sing,

so if you find any who can do both you use them. It's a question of giving a

really good and exciting performance. Sarah Bernhardt could play Hamlet.

Someone called her Hamlet the "Princess of Denmark " —that suggests a
theatrical advantage of using countertenors, men playing male roles.

On the other hand, it's a little odd to hear Alexander the Great sing in the
range of Dame Janet Baker. And women can be great in trouser roles.

People have raised objections to "early-music singing" in Handel. Do you

have any comments on that?
I don't think that you should make singers too uniform. They certainly

weren't in the eighteenth century. Ultimately, a really successful singer, apart

from having good technique and so on, is a total package. The personality, the

technique, the diction, the way they sing, the way they put it across — it's all

part of them. I don't think you should say that Madame Cuzzoni' was neces-
sarily the most perfect singer in the whole universe, but I think all of Handel's
singers were great personalities and sounded like themselves. There were criti-

cisms of them —one had too much vibrato, another didn't have a very good
trill — but it seems that they had a lot of color in their voices, and sang to the
maximum range available, and weren't too Anglican.
You can don a cloak of authenticity in some spurious way, but if that cloak
doesn't fit, it's a disaster. Still, there are quite clearly certain things that you can
learn from early singing treatises, which singers have to pay attention to. A
singing teacher of the late eighteenth century named Domenico Corri,'* for ex-

ample, provides breath marks, which are a lot more expressive than most
singers' breathing now. Modern singers like to sing very long phrases, but a lot
of Corri's breath marks are for much shorter phrases — it's breathing for ex-
pression. On the other hand, his written-out ornamentation gives you an idea

3. Francesca Cuzzoni (c. 1698-1770), a leading Italian soprano of Handel's day, "an ec-

centric and temperamental artist, neither beautiful nor a great actress. Her voice was her
gift . . . ornamentation and breath control were extraordinary; the sheer ex-
[her] intonation,
pressive power of her voice was praised by Tosi [et al.]" (Julie Ann Sadie, Companion to
Baroque Music [London: Dent, 1990, and New York: Schirmer, 1991], p. 83).
4. Domenico Corri (1746-1825), whom Will Crutchfield calls "the most valuable single

theorist" of his era for modern scholars to study. From Crutchfield's essay "Voices" in Per-
formance Practice: Music after 1600, ed. H. M. Brown and S. Sadie (London: Macmillan,
1989, and New York: Norton, 1990), p. 293 et passim.
246 THE BAROQUE

of how ornamentation was done fifty years after Handel, so while some of it

is great, most of it is not particularly useful.

One thing that what Corri said were the three things a singer
is useful is

needed: the messa di voce (the swellingand diminishing on a note), the trill,
and a thorough understanding of harmony and counterpoint so as to be able
to improvise cadenzas and ornaments. And you can tell that people like Lor-
raine Hunt can improvise ornaments, or can make them sound that way even
when they're not improvised.

How?
Ornaments that are written out often sound as if they're part of the text,

instead of something added spontaneously; for instance, they may be too rhyth-
mically correct, without enough rubato. Of course, somebody who's a very

good performer can make you think they're improvising even when they're not,
just as a really good actor can make you think he's scratching his ear because
it itches, even though the scratch was rehearsed to the last detail.

I've always thought that every conservatory should have a compulsory jazz

course for all keyboard players so they can learn to play continuo, and for all

singers so they can free up their singing. What you learn at most conservato-
ries is how to play music that's put in front of you —preferably sight-reading
it, you turn to an orchestral player and say,
but not necessarily listening to it. If

"Now, why don't you play Mozart's Flute Concerto in G, and can you just im-
provise the cadenza?" they've had no training in that, which is absolutely one
of the most important things in all music up until Verdi. This seems to me a

whole dimension of music teaching which is a desperate failure.^ It's a great

shame that it's a failure in the States, because this is, after all, the land of jazz.
Whether you like jazz or not, it still has that free spirit that was expected of
musicians in the eighteenth century, the ability to improvise.

This reminds me that you once said a Baroque orchestra is in some ways
jazz-like.

It's very much continuo- and bass-section-led, in the same way that the

rhythm section and bass form the foundation of a jazz group. In both, the tre-
ble parts are the free parts which sit on top of that. A Baroque piece is very
much like a classical building, where you have structure, which the bass gives
you, and ornament, which is provided by the melodic instruments on top. And
if the structure is strong, then the ornaments can float freely. It's the same in a

good jazz piece; it's not the same in a Tchaikovsky symphony, which is very
often driven by the tune. Even the cellos are often playing the tune. It's just the

double basses that are providing the harmony, limping along underneath.
When I conduct Baroque music with modern orchestras, one thing I do is

to ask the cellos and basses to drive it a bit more, and I ask the violins simply

5. See the interview with Robert Levin for a discussion of this.

 YOU CAN NEVER BE RIGHT FOR ALL TIME 247

to relax. You can often get very good results that way. It's the opposite of what
they're used to. The cellos and basses love it.

How do you get such orchestras used to the element of improvisation?

I was doing Handel's Ariodante at the English National Opera, and asked
the oboe player and the bassoonist to make up some twiddles if they felt like

it. At the first couple of performances nothing happened, but by the end it was
encrusted, and I actually had to ask them to put less in. In the other Handel
operas they'd done, the conductor or editor wrote everything out that he
wanted added, which I think is not the spirit of the thing. Somehow it betrays
a great lack of trust in the performers.

Is your trust ever ill-placed —do you ever find that singers or players im-
provise ornaments badly, and if so what do you do? Or do they overdo it, as
Winton Dean complains happens nowadays?
Well, occasionally you get that. Sometimes people put more than enough
ornamentation in and expect you to edit it out. And there's also the question

of whether what works in the theatre, where it's tied to something physical,
will work in a recording. And those of my singers who improvise, like Drew
or Lorraine, will sometimes just try things out and say, "Well, that one didn't
work" —though as far as I'm concerned, the attempt is itself laudable. Also, I

think in general you'd use more ornamentation in opera than in oratorio. This
was partly because of the needs of the operatic singers: opera was more about
brilliant effects by prima donnas, oratorio more about making sense of the dra-
matic situation.

Do you ever use the ornamentation that Handel wrote out for his singers?*'

I always ask my singers first. In one case the singer simply didn't want to
do them. This was in a Handel opera, Ottone, where he once had to use a

mezzo for a soprano part, and this mezzo obviously was hopeless so Handel
had had to write out the ornaments for her. But her arias were transposed down
a fourth or a fifth, and when you put it back up to the soprano range, the or-
naments sound like the Chipmunks. It was simply too high. So we did a cou-
ple of them, but basically not.
In general, such ornaments were tailor-made for a particular singer. So I tend
to show those ornaments to a singer, and then say. Go and do thou likewise, but
not necessarily copy. Also, I think some of those ornaments are too much.

It's said that rising to high notes in ornaments and improvisations is a mod-
ern idea, that eighteenth -century singers weren't so enamored of heights.

6. See Winton Dean, G. E Handel: Three Ornamented Arias (Oxford University Press,

1976), and "Vocal Embellishment in a Handel Aria," a 1970 essay reprinted in his Essays on
Opera (Oxford University Press, 1990), pp. 22-29.
248 THE BAROQUE

It's very hard to say, because I'm not sure that we have enough evidence to
know what every eighteenth-century singer did. We often read that a singer had

a particular range, but when you actually see the music written for them it

doesn't go nearly that high; so if they had a range that went up to there, when
did they use it? —perhaps in the cadenzas. You can generally assume that if a

singer had a good top C, Handel would use it. Maybe a lesser composer would-
n't. But, yes, Handelian tenor parts don't have as many high notes as a lot of

modern performers would like, so they do tend to throw in extra high notes.
There's some evidence that sopranos used rather more head voice at the top of
their range,and tenors too, rather than belting things out at full voice as nowa-
days they're trained to do. Sometimes the tendency to do that always, and end
everything like the Toreador Song, is a little unsubtle. But I don't think we can
ever say that nobody ever did a particular thing, because we just don't know.

You quoted Corri as saying that knowing harmony was crucial to orna-

menting well.

"Well, this business of knowing harmony is extremely important. When I was

teaching at the Royal College of Music, I was amazed that the singers were often

excused from harmony classes on the grounds that they didn't need it, when in

my view they're precisely the people who do need it. Most string instruments

can provide harmony of their own, and wind instruments can at least get a sense
of what harmony is about, but singers generally sing only one line, and if you're
a tenor or soprano singing the top line you don't really get any idea of structure
and harmony. But you really need that fundamental knowledge.
The other thing you need, which is also very poorly taught to singers, is

rhythm. I find when I'm working with singers who were or still are good in-

strumentalists —Lorraine Hunt, for example, used to play the viola profession-
ally —that their sense of rhythm is so much stronger, which means that when
they want to depart from the beat they know what they're doing. Some singers
merely sing out of time because they've never been disciplined to sing in time.
Somehow the normal rules don't apply to voices. But the great thing about the
eighteenth century is that they did.

Cuzzoni was known for her wonderful rubato.

The biggest problem with all those things is not in performing them but in
puttingthem on record, because they're very fragile things. If you start to put
too much on record, what tends to happen is that it starts to sound fixed and
structured, especially when you've listened to it three or four times. What you
really some wonderful machine that you attach to your CD player that
need is

can change all these things; you would record a basic performance and a bunch
of ornaments and program them randomly.
With LPs, by the time you listened to them three or four times they were
so scratched that you'd never listen anyway. I wish that CDs would self-de-
 YOU CAN NEVER BE RIGHT FOR ALL TIME 249

Struct in the same way after five years, so you'd have to make them all over
again. I was horrified that a record I'd made in 1974 was reissued on CD re-

cently. I think of those pieces totally differently now than I did then. All I could
say was, Thank God there wasn't a photograph inside, because feel about I

those CDs
the way I feel about photographs of myself from fifteen years ago:
"Was there really so much hair? God, those sideburns are awful." Maybe some
of it is okay, but a lot of it isn't and you'd like to redo it.

The recording issue is also related to responding to the audience. If an au-

dience is dreary and sleepy, you have to pull all the stops out not to be dreary
and sleepy yourself, but if there's a lot of energy coming from the audience it

can inspire you to great things. And everything we hear about eighteenth-cen-
tury audiences says that they were very participatory. If they didn't like some-
thing, they threw things. If they did like something, they followed you all over
town, giving you diamonds. was much more Italian in that sense than, say,
It

going to an opera in Washington, D.C., where they tend to just applaud po-
litely. (In Italy, the difference between a crowd at a football game and at the

opera is that the football game crowd is maybe a little louder.) The audience
can certainly inspire you to do your best. So if you extend that principle into
the recording studio where there's no audience, you're having to produce the
music in a way that it was never intended to be produced. You have to fake it.
It's a very different art, especially if you have to do the same thing five times

so they can edit it.

Your recordings often have a sense of live-ness.

Well, recording in the United States is much more expensive than in a lot

of places in Europe, so we make our American recordings in one-third to half

the time of many recordings in Europe. One European recording of a Handel
oratorio took eighteen sessions; we did the same oratorio in six and a half.
Therefore what you're getting very often is one take. Often I'm happy to let

the odd mistake stay in, because it seems to me more important that the spirit

is there, even if there might be one little plonk from the oboe; otherwise, every-
thing might be perfectly manicured but perfectly dead.
In a metaphor I've used before, there's a great deal of difference between a
butterfly flying about and a butterfly in a collection, which is beautifully col-
ored but has this bloody great pin through it. It's as dead as it could possibly
be. What you're trying to do at the recording is to fake the live butterfly, not
the dead one. And you maybe can't see the details of the live butterfly so col-
orfully as you can the one in the collection; on the other hand, it has all the
beauty of the live creature. We have a rule that we only record pieces we've
performed. In the Handel operas, we've performed them on stage three or four
times before we even take them to the studio, and the singers usually sing from
memory. We recorded Handel's Susanna live, and Messiah in whole acts, where
we started with the beginning of the act and went to the end of it as one enor-
250 THE BAROQUE

mous take, and then went back and patched that. The base take is actually sort
of a performance without the audience, so you get some of that tension and
drama, and some of the feeling of taking the energy from the previous piece
the sort of bleed-through from one movement to the next that inevitably hap-
pens in a concert.

So you've tried to turn the limitation of studio time in America into a virtue.

Yes. Correct the odd mistake here or there, but generally get that sense of

making people tap their feet, at least in a happy piece.

It relates to that spirit you've talked about: technical perfection is a twen-

tieth-century concern.
We assume that they were just as persnickety about ensemble as if the CD
microphones were on in their day; yet there's strong evidence that not every
string player even bowed in the same way. I think you can say that on certain
matters in certain acoustics in certain times and places they were being maybe
a little more careful, but in general I think we can be a little holier-than-thou
about it.

One rehearsal was the norm for a new instrumental piece. Early recordings

support your point; even well-rehearsed chamber groups play in what we

would consider a sloppy way.^
You can hear it all the time on those recordings. If you want perfection, buy
a synthesizer. Get the human element out of it altogether. And there are some
people who think that's exactly what they should do. That's absolutely fine; but
it's completely counter to the eighteenth century. The fun thing about doing so
much eighteenth-century music is that it has this free spirit about it.

This also relates, perhaps, to the controversies about Handel's rhythmic no-
tation —whether dotted rhythms and so on should be read literally?

Yes, one has to be very careful of the assumption that we read notes in the
same way they did in the eighteenth century. I think that the danger of asking,
"Do you dot this precisely, and do you make this a sixteenth as opposed to an
eighth?"^ is that it implies that they tried to be as mathematically precise as we
try to be playing Boulez. Perhaps it was more like taking a Charleston, which

7. Robert Philip, Early Recordings and Musical Style (Cambridge University Press, 1992),
chap. 9 et passim. For example, even the four players in a quartet didn't all apply portamento

in the same places and didn't bow or use vibrato uniformly. The rhythms, in particular, are

much less literal than modern performers would allow. All of this occurred in orchestral

recordings as well.
McGegan is referring to a controversy (driven in recent decades by the musicologist
8.

Frederick —
Neumann) about "overdotting" playing certain notes longer or shorter than the
notation indicates. Neumann tended to favor playing the notes as written, a view that put him
at odds with most musicologists. I discuss this at length in note 13 of the William Christie in-
terview.
 YOU CAN NEVER BE RIGHT FOR ALL TIME 251

looks dead written down, but is lively and fluid when played in the authentic
Charlestonized idiom. (I think a close study of Baroque dance would teach us

a lot more about how to do a lot of these little rhythmic things.) In general,
it's dangerous to assume that everybody had that very mathematical approach
to writing down music; some people obviously did, and many people obviously
didn't. Indeed, you can assume that people knew how to ornament, you can
if

assume that they weren't tied to the notes or the rhythms as they were written.
We talked about Cuzzoni doing rubato; she obviously could sing in time but
sometimes chose not to.

How do you handle overdotting?

What we do is based on the fact that if you do a crescendo, the dotting nat-

urally gets a little sharper.** I'd almost rather that the notes were written out
equally and you just played them a lot: I think if you played Lully overtures all

your life, you'd find a way of playing them. It's fine to read a treatise saying the

second note should be a sixteenth note, but ultimately, you just have to play a lot.

I have a computer next door, which lets me play into it and notates exactly
what I play. I think I'm being incredibly accurate, but very often I'm a sixteenth
note early or a thirty-second note late —and I'm just trying to play Frere Jacques
or something. Because we don't actually play in time.'" I think to reduce it to this

sort of organ-loft mentality is uhimately pedantry. A Hungarian musicologist,

Laszlo Somfai, did some work on Bartok's recordings of his own piano works,
works which he notated every timing, every pedal mark, every slowing, and
for

every metronome mark. He found that not a single performance by Bartok con-
forms to what he wrote." If the composer doesn't do it, why should we?

To return to opera performance, you once said that you could conduct
Verdi's comic opera Falstaff but not his Otello;'^ but one of the big areas of

9. It's McGegan's point that performers in the first third of our century over-
relevant to
dotted routinely in —
most music not just Baroque music, and not in obedience to theoretical
—
exhortations and that the overdotting was more pronounced in loud than in quiet passages
(see Philip, Early Recordings, chap. 3, esp. p. 84, last paragraph, and p. 90, second para-

graph).
10. psychologists Ingmar Bengstsson and Alf Gabrielsson have done exten-
The Swedish
sive studies that found that when classical musicians play rhythms in a 2:1 ratio (say, a quar-
ter note and an eighth note), even though it sounds like two to one, it is never mathemati-

cally precise (it's usually less than 2:1). Their work is published in Studies of Musical
Performance, ed. Johan Sundberg (Stockholm: Royal Swedish Academy, 1983), p. 58.
11. Though it's not the study McGegan refers to, Somfai's Bela Bartok: Composition, Con-
cepts, and Autograph Sources (Berkeley, University of California Press, 1996), pp. 279-95,
discusses Bartok's recordings and what they tell us about his performance style —and, for that
matter, about the inadequacy of musical notation systems. Bartok's playing goes well beyond
what he notated, and that includes the rhythms and even sometimes the notes; one could not
determine from the notation what Bartok actually plays.
12. Eric Van Tassel, "An Interview with Nicholas McGegan," Fanfare 13 (January/Febru-
ary 1990), pp. 76-84; quote p. 83.
252 THE BAROQUE

your Handelian repertoire has been his operas, which are in the vein of opera
seria does that contradict your first statement?
Not really. I find Handel's operatic characters —not as titled figures, the king

of this and that, but as actual people —a lot more interesting than those in the
really bourgeois operas of the nineteenth century. Wagner is very bourgeois in
terms of his plots and characters. In Tannhduser and Dutchman the women are
really suburban, these dreadful singing Hausfraus. They always have to have
the House and Garden virtues: they all cook and spin (and sing top C). And
the men, it seems to me, are perfectly fine to begin with, but they all have to
be redeemed in some ghastly way. It's a problem in Wagner that I don't find
interesting. There was a wonderful cartoon in the New Yorker, showing the
end of one of these Verdi or Wagner operas, with everyone lying dead on the
stage;two blase people are sitting in a box, and one of them says to the other,
"You know, with a little early counseling, all of this could have been avoided."
You don't get that so much in earlier opera.

A question often raised about these operas (though, admittedly, it has sub-
sided) was whether they held the stage. You've conducted them in the opera
house; what do you think?
would say that you can say the same about Bellini, or about any Rossini
I

opera seria. The problem is not whether it's Baroque music as opposed to bel

canto. I think there are a number of ways through which you can make them
work on the stage. '^
First, as far as cuts in the text go, sometimes you can trim a little bit, but

—
where it becomes a mistake which has often been made is when you simply —
trim the arias of the lesser characters. Usually, these characters are the ones
who make the plot flow. Handel's own cuts, too, can riddle the plot with non
sequiturs, but they were often made in desperation, sometimes because he did-
n't have a good singer for a role.

One thing I think is extremely important in these operas is to cast grandly

enough. When Handel was casting for the best, the most fa-
did an opera he
—
mous almost as you would cast a Broadway musical now. These people were
known offstage as well as on, so when that particular famous person walked
onstage all the audience would go "Oooh!" just as you would if Barbra
Streisand walked onstage.
Another thing is that the stagings were often spectacular. A lot of opera
houses now decide that if they do a Handel opera it's going to be their cheap
show of the season. It's only got a cast of five, no chorus, so let's just save all

around, shall we? But actually, those shows should be gaudy and expensive and
glorious.
Another thing that I think is the death of Handelian opera is the orchestra

13. McGegan goes into this topic in more detail in his article "Movements by Candle-
light," Musical Times 135 (April 1994), pp. 210-15.
 YOU CAN NEVER BE RIGHT FOR ALL TIME 253

pit, which is a nineteenth-century invention. In the eighteenth century the or-

chestra was same level as the stalls, '* or maybe a foot lower, and the
at the

most important players faced the stage, so that they could accompany the
singers directly just as they would in a concert. That's why you have these fan-
tastic oboe and violin obbligatos: they are actually making music with the

singers, without the silly medium of the conductor doing semaphore to relay
between two people who can't hear or see each other.
Another thing is that you have to be very careful about doing these op-
eras in large opera houses — theatres where you can't see the whites of the

singer's eyes. Handel's opera house held about 1200 people, and we're very
short of such houses. To get a feeling of intimacy, a sense of being very close
to the scene, is very important for most of these operas. Of course, doing
these operas in a small house isn't a terribly good idea financially, because
the tickets have to cost too much — though in Europe there's generally gov-
ernment subsidy.

Another issue regarding Baroque opera performance: Paul Griffiths recently

commented ^^ that Baroque staging and gestures seem stilted to modern audi-
ences, even though Baroque performance practice in music has proven quite
appealing.
I don't know Baroque gestures really done. We've
that any of us have seen
seen some attempts, but if them under candlelight there's no point
you don't do
in doing them. Once you start raising all the light, the gestures remind you of

that Monty Python skit of doing Wuthering Heights by semaphore. The only
reason Baroque actors used those gestures was so they could get across what
they wanted to get across in the dark. A gesture is of a certain size in order to
be seen, and if you can see everything so clearly that all you need to do is raise

your finger a little, then everything else is overacting. I think the only way in

which Baroque gesture works is if it's part of a total package. The proportions
of the stage, the sets, the costumes and the lighting all have to be right.

Another issue: Handel would revise a part, you've observed, to suit what-
ever particular singer he had on hand. Would you?
No. But you can put a part up or down with ornamentation you can fudge —
it a little in that way. I think it's better to try and find somebody whom the

part does fit, since so often the roles are dramatically somewhat generic."" Fun-
nily enough, the heroes in these operas are usually the most boring people; it's

14. The American term is "orchestra seats."

15. New Yorker, 5 July 1993, p. 98.
16. Charles Rosen, in The Romantic Generation (Cambridge, Mass.: Harvard University
Press), p. 605, writes that in eighteenth-century opera serta "the psychology, if that is the word
for it, is . . . simplistic, even primitive: there were rarely any characters at all in opera seria,

only a succession of dramatic situations which allowed the singers to express a series of emo-
tional states."
254 THE BAROQUE

the villains who have all the fun, and they're usually basses or baritones. The
heroes are often not heroic. The lead castrato role, let's say, is a Roman em-
peror; but he doesn't do anything except moon about the stage in love. It's the
women and, as we've said, the lesser characters who do everything in the
drama.

Regarding another genre you've worked in, the English -language oratorios,
one issue might be period pronunciation. Was Handel writing for a specific
English pronunciation we no longer have?
He was: there are certain words that have changed. If I were to speak eigh-
teenth-century English I could say that I am part of the audience at a dray-ma,
and listen to air-ias (instead of arias) and in the bal-CO-ney. One of the
sit

American critics made a big stink because when Philharmonia Baroque did
Judas Maccabeus one of the singers didn't make "hands" and "commands" a
full rhyme, but in eighteenth-century England they weren't.
On the other hand, there hasn't been a standard English pronunciation until
this century. If you had a regional accent you kept it. And beyond that, from
what one can tell half of the oratorios, although they're written in English, were
not sung entirely in it. At least six arias in UAllegro were sung in Italian, be-

cause the singers were Italian. And even when the Italians sang in English, one
critic said, "I thought they were singing in Hebrew." The famous story is about
the revival of Esther, where an Italian singer made "I come, my queen, to
chaste delights" sound like "I comb my queen to chase the lice." And there
were Italian singers in almost all the oratorios right up to Handel's death.

It's a nice argument against the idea of authenticity being "re-creating just
what they did at a performance in the composer's time.
Yes. When we did Judas Maccabeus we got roundly criticized because Guy
de Mey's very good English was not absolutely perfect. I wrote to the critic and
pointed out that most of Handel's singers were foreigners.
This question of authentic re-creation becomes, in the end, meaningless. In
Ariodante, which I've just edited, there's an aria for the bass where Handel re-

moved quite a lot of the coloratura for the first performance, and it's in this

simplified version that the piece is now printed. If you do the simple version
you're doing what Handel did in his lifetime, so it has a certain cachet of au-
thenticity. But the only reason it exists is because that particular bass couldn't
sing the more difficult original version. At Gottingen we have a singer who can
sing the more difficult version, and we prefer that version, so it's what we're
going to use. The great thing about performance is that you can never be right
for all time, the way a scientist can be right about the earth going around the
sun. These things are much more fluid. In the arts, one thing you can never be
is absolutely right.
 YOU CAN NEVER BE RIGHT FOR ALL TIME 255

SELECTED DISCOGRAPHY
Nicholas McGegan's catalogue of Handel operas on Harmonia Mundi has
grown at the rate of one per year. As I write, the best of the 3-CD sets are the

most recent: the Gramophone Award-winning Ariodante (HMU 907146.48),

which "may be the crown jewel of the series" according to David Johnson, who
praises the "generosity, verve, and elegance" of McGegan's conducting;'^
Giustino (HMU 907130.2), in which, Johnson wrote earlier, "McGegan sur-

passes himself";"* and the previous entry, Radamisto, which Stanley Sadie
called "(McGegan'sl best by far. ... as compelling as any Handel opera per-
formance I have heard"'** (HMU 907111.3). Sadie, an expert on Handel opera
performance, has not always admired McGegan's work in this repertory: writ-

ing of the 1990 Floridante (Hungaraton HCD 31304.6) Sadie calls it lively but
emotionally detached and musically mannered, with "persistent and ultimately
irritating little swells and squeezes and . . . coldly abrupt phrase endings."^"
But Sadie has recently called McGegan's ariodante "the best Handel opera
recording we have yet had."''

Like the middle-class London audiences of the 1740s, I prefer the oratorios
to even the best of the operas. Thus my favorite McGegan Handel recording is
of the late masterpiece Theodora (HM 970060.62, 3 CDs); Colin Tilney calls

the disc "impossible to recommend too highly."^^ McGegan's performance of

the early cantata Clori, Tirsi e Fileno (HM 907045) is, in Tilney's words, a "les-

son in style and joie de vivre." As for Messiah (HMU 40 7050.52, 3 CDs)
McGegan's recording has the distinction of including almost all the variants
from Handel's many versions, so that you can program any (or your own) ver-

sion. The performance itself has proved controversial: Classic CD named it the
best Messiah on CD, but Nicholas Anderson (among others) is disappointed by
its "undercharacterization of Handel's music"; he misses "Handelian grandeur
"^^
and nobility.

McGegan has recorded a good deal of other Baroque music, including

Corelli and Vivaldi. Among the best-received have been his CDs with a cham-
ber ensemble, the Arcadian Academy, consisting of McGegan and three mem-
bers of Philharmonia Baroque — Elizabeth Blumenstock, David Tayler, and Lisa
Weiss. Their recordings of Matteis and especially of Uccellini (La Bergamasca,
HM 907094) are exquisite examples of Italian Baroque chamber music. Of
Philharmonia Baroque's recording of instrumental suites from Rameau's Na'is
and Le temple de la gloire (HM 901418), Jan Smaczny writes that "The first

17. Johnson, Fanfare 19 (May/June 1996), pp. 163-64.

18. Johnson, Fanfare 19 (January/February 1996), p. 217.
19. Sadie, Gramophone 72 (June 1994), p. 109.
20. Sadie, Gramophone 70 (January 1993), p. 61.
21. Sadie, Gramophone 74 (November 1996), p. 54
22. Tilney, "Theodora: Two Views," Historical Performance 5 (Fall 1992), p. 91.

23. Anderson, Gramophone 69 (October 1991), p. 162.

256 THE BAROQUE

forty seconds of the overture to Nat's should be enough to persuade anyone that
they are Hstening to perhaps the most thrilling sounds of the late Baroque."^''

FOR FURTHER READING

Handel had a strong personality but, as Donald Burrows points out, we have
"surprisingly little firm evidence about his private life and many aspects of his

personality."^^ Nonetheless, many fine biographies exist; the best so far is Bur-
rows's Handel (London: Macmillan, and New York: Schirmer, 1994). Christo-
pher Hogwood's Handel (London: Thames and Hudson, 1984) is also good,

and Otto Erich Deutsch's Handel: A Documentary Biography (London: A. and

C. Black, 1955) is a still a valuable resource (as Burrows notes, we do have de-
tailed records of Handel's public and professional Hfe). For a shorter biogra-
phy, the NewGrove Handel by Winton Dean (New York: Norton, 1983) is
one of the most appealing in that series.
As for discussions of the music, Winton Dean's Handel's Dramatic Orato-
rios and Masques (Oxford University Press, 1959) and, with John Merrill
Knapp, Handel's Operas, 1704-1726 (Oxford University Press, 1987) are clas-
sics of scholarship and style. Those who want to read more about Handel's

own surviving ornamentation might consult Dean's edition, G. F. Handel: Three

Ornamented Arias (Oxford University Press, 1976), and his 1970 essay "Vocal
Embellishment in a Handel Aria," reprinted in his Essays on Opera (Oxford
University Press, 1990).

24. Smaczny, BBC Music Magazine (December 1995), p. 69.

25. Burrows, Handel (London: Macmillan, and New York: Schirmer, 1994), p. ix.
 14
At Home with the lai wm
— oOo^
g-

William Christie on the

French Baroque

A Gramophone critic, praising a French Baroque opera recording, noted that

it was "a distinct improvement on what has been [the French conductor's]
rather dodgy past." But how much of the improvement, he asked, was due to
the "standards of technical and dramatic excellence which William Christie has
helped to establish?" He pointed out that the singers, and for that matter those
in the other French Baroque recordings under review, had all been trained by

Christie, an American harpsichordist/conductor. "Do we have then," he asked,

"a French school of performance or a Christie one?"'

The answer could be, both. The New York Times says that Christie is

"chiefly responsible for reviving interest in French Baroque music,"' and few
would disagree. After studying the harpsichord with Ralph Kirkpatrick at Yale,

Christie moved to Paris in 1971; there he immersed himself in the available

documentation on French Baroque music, culture, and performance practices.

From this he extracted not only a wealth of details but also the essence of a
living style. Fie has managed to share that style with a pool of young musi-
cians, many of them his students at the Paris Conservatoire, where he was the
first American ever to be given a professorship. His group, Les Arts Florissants
(named after a Charpentier opera), and his proteges, such as Christophe Rous-
set, have been at the core of the recent explosion of activity in the French early-

1. Lindsay Kemp, "Quarterly Retrospect," Gramophone 71 (September 1993), p. 33.

2. Alan Riding, "Where Is the Glory That Was France?" The New York Times, Sunday
Arts and Leisure section, 14 January 1996, p. 1.

257
258 THE BAROQUE

music world —an explosion that led one critic to call Paris the current "hotbed"
of early music.
It may seem puzzling that France, the country that gave us the word "chau-
vinism," would adopt an American to teach it how to sing in the true French

Baroque style. It's not as uncharacteristic as seems, though. The French arts
it

scene, even at its most florissant, has usually welcomed foreign talent (think of
Chopin or Picasso). It may be especially welcoming in its less-than-flourishing
present state: many people believe that what's exciting in the arts in France
right now is, for the most part, the work of immigrants like Christie."* Whether
that's true or not, France's reception of Christie is poetically just: Jean-Baptiste

LuUy, the father of the French Baroque style —a style that self-consciously dis-

tanced itself from the Italian style —was himself Italian.

Christie has become welcome enough to have been awarded, in 1993, the
rosette of a Chevalier de la Legion d'Honneur (the award led the immigration
office to at last grant him a permanent visa).The award honored his resurrec-
tion of French Baroque opera, an achievement that Richard Taruskin calls "per-

haps the finest" of the early-music movement, the only one to which even
Taruskin "would willingly grant the freedom of the term" restoration.^ To ac-
complish that restoration, Christie's musical and theatrical gifts, though excep-
tional, weren't enough; they had to be allied to his unwavering quest for an en-
semble that all but breathes in a unified, idiomatic style —precisely what had
been hardest for performers to attain in French Baroque music.
Some musicians, such as the pianist Andras Schiff, complain about "so-
called stars of early music . . . Iwho go] through the literature of music at the

speed of a Concorde, maybe performing it once: this week it's Mozart, next
week. Bach, then it's all of Monteverdi. In a couple of years they cover the field
from William Byrd to Stockhausen. Nobody can digest that much music."" The
critique may be valid in certain cases —you can judge that for yourself— but the
majority of the artists interviewed book are specialists. Christie, for one,
in this

rarely ventures beyond the seventeenth and eighteenth centuries, and his focus
has most often been French. In his interview (conducted by trans-Atlantic tele-

phone) he argues that specialization is a path to musical freedom.

Many is the most

people, including yourself, have said that the French Baroque
difficult style formodern performers to master. Can you discuss why?
I think it's partly because you must have a linguistic approach to French

Baroque music. The basis of all French musical art is declamation; you have to

3. Tim Pfaff, "Les Talens Lyriques: The Next Generation," Strings 9 (March/April 1995),
p. 74.

4. Riding, "Where Is the Glory?"

5. Richard Taruskin, "Of Kings and Divas," The New Republic, 13 December 1993, p. 40.
6. Quoted by Harriet Smith, "Far from the Madding Crowd," Gramophone 72 (October
1994), pp. 22-23.
 AT HOME WITH THE IDIOM 259

(b)

'S''
;i I F 1^
(ob)-jct j'a - do - re dc bril - lent de - ja

EXAMPLE 1 (a) A port de voix is a short added note that ascends to the main note;
it usually repeats the preceding note. This port de voix (circled) moves from

unstressed to stressed syllable, (b) A coule is a short added note that descends to the
main moving from unstressed to stressed syllable, with coule to the
note, here seen
latter. From Rameau's L'impatience, realized by Mary Cyr, "Performing Rameau's

Cantatas," Early Music 11 (October 1983), p. 485.

understand French declamation and declamatory patterns to understand, for

example, all the baggage of French Baroque ornamentation, which is gram-
matical to a large degree.
This has to do with what the French call syllabic quantity, which essentially

is a way of giving stress to certain syllables and not to others. Now, linguistic

patterns that stress specific syllables can conflict with musical patterns that
stress specific notes. You can have a conflict sometimes between the text and
a musical line, where the musical value of the note is not the real value of the
syllable upon which it's placed, especially in strophic music. The trill, port de
voix, coule, and other ornaments, which one finds in all Baroque music, in

France become essentially ways of lengthening syllables to resolve such con-

flicts. Used this way, the ornaments and so on give the listener a better com-
prehension of the text. One might sum up their use in the following recipes.

When leaving an unstressed syllable to go to a stressed syllable in a rising

melodic line, one uses the port de voix in one of its various manifestations (Ex.
1 [a]). When leaving an unstressed syllable to go to a stressed syllable in a de-

scending melodic line, the musician can use the coule (Ex. 1 [b]) or the port de
voix on the stressed syllable.

Why ivould stressing certain syllables lead to better comprehension of the

textf I had understood that in French, stress and vowel length are not usu-
ally fixed factors in a word, and don't help one pick out the word in a sen-
tence.
That's true in modern French, but in seventeenth- or eighteenth-century
French, stress was an extraordinarily important factor. (These patterns gener-
ally changed about the end of the nineteenth century.) You can't have syllable

emphasis without syllable quantity; and for that, vowel length is critical.

Syllabic quantity is important not only in the matter of ornamentation.

Phrasing, too, becomes a question of long or short syllables, and non-legato
treatment of music also arises out of this. These are difficult questions for peo-
ple to solve, because essentially what you're doing is putting something into the
music that isn't in the score. There's a lot you must put into French music,
260 THE BAROQUE

which is not the case in Bach or some other composers, where the music seems
to be more indestructible —which of course is why their music is played more.

You've praised Beecham's Messiah, Horowitz's Scarlatti, and Casals's and

Landowska's Bach, saying that while they weren't historically informed, they

had a grandeur and eloquence missing from much modern historical perfor-
mance.^
Yes. Incidentally, Beecham's 1938 recording of Die Zauberflote is one of
the best things in the world.

But what you were saying earlier brings up a question: is historical in-
formation more important for French music than for Bach and Scarlatti?
Would Beecham have a harder time in Rameauf
One doesn't know, of course. But as for the first question, yes, I'd have
to agree. French music dies when one doesn't have the tools to bring it back
to life. As I said, to bring it back to life is largely a literary consideration,
in many ways, having to do with declamatory and linguistic questions.

That raises another question: how do these linguistic concerns apply to

French instrumental music?
To such an extent that A. Pherotee de la Croix said that all dances have
texts —imaginary texts in some cases. Any allemande, any courante, any dance
form (with a very few exceptions) is essentially a dance with specific num-
bers of syllables per line; one could say that instrumental dances are always
accompanied by imaginary texts. ^ Indeed, all good instrumental styles have
the voice as their model.
Almost everything has a verbal basis. We also have a kind of paradox,
though, which singers don't like to hear: while instruments are supposed to
imitate voices, we've got to recognize the fact that in France, Italy, and Ger-
many and elsewhere, voices were sometimes asked to behave like instru-
ments. A Mondonville psalm for harpsichord obbligato, ad libitum violin, and
voice requires the voice to behave in a completely instrumental fashion.
This is true of some Rameau and Handel arias as well, and Bach cantatas,
of course. That's just something that one has to do in Baroque vocal music.
There is a kind of extraordinary mimicry between voices and instruments.
Voices have to identify themselves with the instrumental context to make co-
herent music.
Why do I like certain voices? Because they essentially know how to sing

7. Jean-Francois Labie, William Christie: Sonate-Baroque (Aix-eii-Provence: Alinea, 1989),

pp. 86, 91. This work features extensive interviews.
8. A valuable discussion of this point can be found in Patricia Ranum's "Audible Rhetoric
and Mute Rhetoric: The 17th-century French Sarabande," Early Music 14 (February 1986),
pp. 22-39.
 AT HOME WITH THE IDIOM 261

with a harpsichord or how to sing with a Baroque oboe or vioMn. It's a ques-

tion of technique, it's a question of writing, and it's a question of sound —of
mimicry. That's very important. It's true, also, that instrumental schools were
bound and beholden to obey the same principles that singers were. An in-

teresting thing is that the prototypes of instrumental ornamentation, as cod-

ified by people like d'Anglebert in the 1680s and 1690s, are vocal prototypes.
We're aware that smgers were already using these. There's an extraordinary
sharing of ornamentation, vocal and instrumental.

The declamation, and emphasis on words, I understand reflected the pres-

tige —
of French theater Italy had nothing to compare with it; and to the sev-
enteenth-century French, the Italian opera seemed weak from a dramatic stand-
point. Is that correct?
There are all kinds of styles. You're talking about the tragedie lyrique, of
course, and there that's true. But not everything in France comes out of
Corneille and Racine. The big stuff does, of course, the big vocal forms and,
by a kind of inevitable extension, the recitative —which was already being de-
claimed in the theatre in a very sung style. LuUy essentially developed some-
thing that was already happening amongst the actors, the desire to burst forth

into song.

He is said to have sat with the great actress Champmesle and studied her
declamation of Racine; Racine apparently wrote down pitches for her to use

in declaiming.''

That was a constant of French declamation up until the end of the nine-
teenth century —there was something very sing-song in the way they declaimed.
One hears it in the Sarah Bernhardt wax cylinders from the beginning of our
century. The declamation patternswould be immensely complicated to graph,
because they're so sing-song. And the French have had a kind of love-hate
relationship with this. It was already criticized and defended back in the eigh-
teenth century. This kind of grandiloquent way of declaiming is something I

can sometimes hear nowadays in a bastardized form; one can still hear some
echoes of it in modern-day public speaking. It's dying out now, but the older
clergymen that I knew twenty-five years ago still had it.

An obvious corollary: the style of theatrical delivery was not naturalistic.

Well, it imitates nature: "nature" at the end of the seventeenth and early

eighteenth century was very different from that word today. It meant, essen-
tially, truth in terms of emotions and feeling. It's a way of heightening the
pleasure by exaggerating speech and gesture.

9. See Lois Rosow, "French Baroque Recitative as an Expression of Tragic Declamation,"

Early Music 11 (October 1983), pp. 468-79.
262 THE BAROQUE

When you prepared Medee, you began rehearsals by having the singers

first recite the text according to the rules of seventeenth-century French

tragedy.
Yes, though I can't claim that we're doing exactly what was done then.
There are immense holes what I do. I'm not seeking authenticity per se. For
in

example, I don't really insist on historical pronunciation, though I recognize

that eventually I have to get around to doing so. There are several reasons I
don't insist, but one is that I've burned my fingers with it with French audi-
ences many times; they simply balk at it.

It's like doing Shakespeare with Elizabethan pronunciation for English au-
diences.
Exactly, and people have tried that, you know.

With unfortunate results.

It goes back to what began all this: declamation and comprehensibility. I

want to go further in terms of pronunciation, and I will. It is only a question

of time.

You have used historical pronunciation, though, with Latin works by Char-
pentier and other French composers, using the seventeenth-century French style

of Latin rather than the Italian style prevalent today. Can you say anything
about how that benefits these works?
Pronouncing Latin texts as a Frenchman would have causes the singers to
place their voices differently. This means you have a large number of half-closed
vowels, which of course is not the case in Italian Latin, and they give a very
particular kind of color to the piece, which you can use.

Let me ask you about the controversy over one very characteristic aspect of
French Baroque music, notes inegales; do you have a general statement about
them in French music f^°
I'd say that if you don't subscribe to notes inegales you might as well stay
away from French music. Anybody who has any notion of spoken French, even
uncultivated twentieth-century spoken French, understands that one never pro-
nounces a chain of monosyllables with equal insistence. Just listening in a stu-

dent cafeteria, you understand very clearly why inegales exist. It's an extension
of speech.

David Fuller questions the idea that notes inegales are based on French

10. As was observed in the Anner Bylsma chapter, in the French convention of notes ine-
gales certain pairs of notes that were written at equal lengths were played unequally —more
often long-short, long-short, but sometimes short-long, short-long.
 AT HOME WITH THE IDIOM 263

declamation; he argues that they were mainly sung to diminutions on long syl-

lables.''

I'd say that I hear inequality all around me, in modern-day French speech.
I have a feeling it did come out of French declamation patterns, very much so.

Christophe Rousset says that applying notes inegales to a sung text in

French can have the same effect that you described with ornaments, that of
making the text easy for listeners to comprehend.'^ And so it would be in-

evitable in instrumental music with imagined text. What are your views on the
awareness of notes inegales in Germany and England^'"'
I can't imagine someone like Henry Purcell shying away from the practice.
I mean, he's the most consummate Frenchman I know the parallels, the bor- —
rowings, it's amazing. And all you have to do is listen to some of the eigh-

11. Fuller, "The Performer as Composer," in Performance Practice: Mustc after 1600, ed.
H. M. Brown and S. Sadie (London: Macmillan, 1989, and New York: Norton, 1990), pp.
145-46.
12. In Bernard D. Sherman, "Finding One's Own Recipes," Piano and Keyboard,
May/June 1994, p. 26.
13. The idea that Baroque musicians outside France generally did not know about such
French rhythmic alterations as notes inegales and overdotting (sharpening dotted rhythms) was
argued by the musicologist Frederick Neumann (see his Essays on Performance Practice (Ann
Arbor, Mich.: UMI Press, 1982). Neumann's view that the notes should be played as written
put him at odds with most early-music performers and scholars. A book by his student

Stephen Hefling, Rhythmic Alteration in Seventeenth- and Eighteenth-Century Music: Notes

Inegales and Overdottmg (New York: Schirmer, 1993), demonstrates beyond a reasonable

doubt that Neumann was largely mistaken about this issue. (See also the review of Hefling's
book in the Fall 1994 issue of Performance Practice Review 7, pp. 120-32, by Neumann's
principal critic, David Fuller.) Hefling summarizes his findings on notes inegales thus: "The
was certainly known to musicians in several
available evidence suggests that French inegalite
areas ofGermany as well as in the Netherlands and England, and also in Modena. The . . .

custom was transmitted by Frenchmen who went abroad, as well as by visitors to Paris who
subsequently imitated French style, and also by French musical writings. . . . Very likely in-

equality was applied to music of French origin and to works that obviously emulated French
style; to what degree [it was] extended to pieces less closely related to French models remains
uncertain. . . . [However,] there seems to be no evidence that anyone outside France assumed
inequality as a matter of course, with the exception of Quantz and possibly some of his col-

leagues in Dresden and Berlin" (pp. 60-61).

In music by Bach in the French style, Hefling shows, overdotting and notes inegales were
probably applied (see pp. 41-50, and chaps. 5 and 6, esp. pp. 98-100). In Purcell (in whose
music Christie has been criticized for using inegalite), Hefling seems to find the evidence less

clear; he argues that Purcell was probably familiar with inequality in French music, but that
when Purcell wanted it used he may have indicated it with dotting (pp. 51-55).
While Neumann's positions on these issues have not held up under careful scrutiny, they
did a service by forcing scholars to undertake that scrutiny. Moreover, his thorough critiques
elsewhere have led to the relaxing of some early-music orthodoxies about ornamentation, such
as beginning every —
ornament on the beat as he showed, original practice was not so rigid. I
would also mention that Neumann was not desk-bound; he was a fine violinist, and was mo-
tivated by practical experience of playing early music, as well as by his distaste for much of
the "early-music" style.
264 THE BAROQUE

teenth-century English barrel organs that are still around to understand that the

quirky rhythms inegales, jerky lombards, scotch snaps —were very much pre-
sent there.
As for Germany, again you have a predominant aesthetic. Good taste, la

mode, was of course a kind of European disease in the eighteenth century, and
the most important capital of la mode was France. If in fact this taste traveled,

because French people did travel and were asked to travel to show people how
to do these things, then everywhere — in St. Petersburg, in remote parts of Ger-
many, in Edinburgh —you're going to find people doing things that were done
in France in 1710 or 1730. It seems to me blatant: we have French musicians
in Germany; we have German musicians who have spent time in France. We '"^

have immense amounts of music being sent from France to Germany or being
copied in Germany —Bach copying de Grigny, for example, or Couperin. It

seems to me would be very difficult to avoid the essential features of

that it

French style. Ornaments —

we know they did no one quibbles about that. The
French ornament tables we know were in Germany from very early on. If that's
the case, it seems to me that one would also have performance-practice ingre-
dients like notes inegales, especially coming into forms that were essentially
French to begin with. It's perverse to buy a bit of a product and not the whole
product. But of course this is grist for the musicologists, some of whom have
no more important things to talk about, and some of whom have no practical
musical ideas.

/ also want to bring up the distinction between the dominant style of

Baroque singing, which was the Italian, and the French style. Although the Ital-
ians at the time of early Monteverdi spoke a lot about text, by the 1670s, when
French opera gets going, the Italians are no longer paying that much attention
to it.

Well, that happens to French opera in the 1750 and 60s: you lose text to
melody and technique. I like the French Baroque immensely in the earlier pe-

riods, the 1670s and 80s, because there is that kind of equilibrium between text
and music, and the fluidity that you find amongst the Italians fifty years be-

fore.

14. According to his obituary, the young Bach's excursion to Hamburg gave him the op-
portunity to "acquire a thorough grounding in the French taste," by allowing "frequent hear-
ings of a then famous band maintained by the Duke of Celle, consisting mainly of French-
men" [The Bach Reader, ed. Hans T. David and Arthur Mendel [New York: Norton, 1966],

p. 217). Peter Williams points out the significance of this by contrasting it to Bach's experi-
ence of Italian music. Exposure to Vivaldi scores influenced Bach's compositional style pro-
foundly, but his music in this style often cannot be played at the rapid speeds for which Ital-

ian virtuosi were known, presumably because Bach didn't know about those speeds. See
Williams's The Organ Music of J. S. Bach, vol. 3 (Cambridge University Press, 1984), pp.
91-102, for a subtle discussion of what Bach learned from France and Italy.
 AT HOME WITH THE IDIOM 265

But would differences between the of singing, though — such

styles as the

use of legato in Italy hut not in —

France create a problem for modern singers,

since most Baroque singers today are essentially singing in an Italianate style?

No, I think the basis for any kind of singing really is Italian. That was rec-

ognized even by the French. Rameau said if you want to have agood vocal ed-
ucation, go off to Italy. He was very much in favor of that. And you find the
French nodding their heads towards the Italians, in terms of technique at any
rate, from the beginning. It's funny that the French, whose singing style essen-

tially goes counter to all that the bel canto aesthetic wants — legato, long-line

singing — insist that you have to learn Italian style before you learn their style.

To make coherent sense of the French music, of course, you have to do some-
thing different from the Italian; but just as you have to learn how to walk be-
fore you can run, you have to learn essentially how to sing legato before you
can sing non-legato phrases.
The is that you get enthusiastic people who want to sing in the
great error
French and have no vocal technique whatsoever, and that's been a prob-
style

lem in the States and in other places as well. I've had far more success with
people in French music when they have a first-rate base in classical Italianate
—
techniques the support, the complete muscular freedom, everything that bel
canto wants, essentially.

In the Labie book you expressed the thought that singing and dance were
two areas of early-music style that had trailed behind.^^
I think there's been improvement since then. Ten years ago, when that book

was written, it was painfully obvious that the Baroque violin school, or flute
playing or harpsichord playing or theorbo playing, had come a longer way than
singing or dance. Now I have thirty young student singers working with me all
over Europe, and they do things instinctively that professionals ten years ago
wouldn't do. For example, in ornamentation they know exactly what I'm talk-

ing about and how to do it, in a way that already suggests a specialty approach.

So that's enjoyable.

You mentioned back then that you felt the English singing style was vigor-
ous and accurate but sometimes lacked the necessary passion for Baroque
music.
I'm not wildly fond of a lot that I think is going on in England. I remem-
ber talking to one of the foremost early-music leaders from London, and we
talked about apathy and routine, and these seem to be a kind of disease that's
happening right now in London. Along with it comes a kind of lackluster way
of dealing with style.

15. Labie, William Christie, p. 72.

266 THE BAROQUE

Is it the training or the plethora of recording gigs? Is it because the same

people play all the recording sessions?
Perhaps. Many no longer think about what they're doing, essentially. If it's

in tune and it's together, that essentially satisfies what they're after. They don't
take risks.
As for the training, though, I think the vocal education in England is fabu-
lous,which is why oftentimes I'd rather use English singers especially those —
coming out of the non-specialty schools, that is, coming from the mainstream
repertories at, say, Guildhall or Northern College, where voices are given a
good technique. There are some very fine countertenor voices from England;
someone like Michael Chance is exquisite.
What I was harping on in the Labie book was this specialty "early-music"
voice, which still sometimes gets my dander up. As it did back in the States
too —people who somehow insist on squeaky, small, non-vibrato voices as the

key to correct early-music style.

What is your view on vibrato in eighteenth-century singing?

My view is simply physiological. If you want to have a voice and a decent

technique, and you want to keep your voice, you have to recognize that vibrato

is an essential part of vocal production. You can't subscribe to good bel canto

technique, you can't be a bel canto singer without vibrato. With a column of
air with good breath support and no muscular tension if it's free and easy —
you are going to vibrate the two vocal cords. Aside from stentorian high-deci-
bel verismo stuff, which is part of the twentieth century the overblown Amer- —
ican-monster style of singing where you blow your guts out —with the
exception of that, sensitive singers of any kind of music have to know how to

use the vibrato. Vibrato is more important in singing a Handel portamento aria,

or Mozart, than perhaps in French music. But to say that one doesn't vibrate
in French music is perfect nonsense.
One hardly talked about vibrato in the eighteenth century. You get occa-

sional references: Rousseau '*"

says that voices without oscillations can be just
as pretty as ones that had them, which means of course that people were using
vibrato. But it's very simple to fulfill all the requirements —conjunct-note
singing, close-interval singing, obeying certain instrumental principles of vi-
brato and non-vibrato — if you are indeed a singer with vibrato. You just have
to know when to turn it off or when to use less of it. If one is beginning a trill,
since the trill itself is the beating of two notes, you want to make sure that the
voice, if it is using vibrato, is using less of it. The key concepts are dissonance
and good tuning, which are crucial parts of seventeenth- and eighteenth-cen-
tury singing; these are things that require attention to vibrato. But to say that
vibrato can't be used at all is nonsense.

16. Jean-Jacques Rousseau, Dictionnaire de musique (Paris, 1768), "Voix," pp. 541-42.
 AT HOME WITH THE IDIOM 267

Was it used as something ornamental, to heighten or lessen, or as something

constant?
Vibrato can be used as something constant, with non-vibrato being used as
an ornamental feature. Or we can say — as instrumental schools did — that vi-

brato itself is an ornament, to be used, more of it or less of it, as an ornament.

Your work with Les Arts Florissants brings up the issue of specialization.
David Fuller writes: "One does occasionally hear today what seem to informed
ears to be fine, stylish idiomatic performances of this music, and the secret
seems to be specialization . . . by steeping oneself in one repertory and all that
surrounds it —cultural background, organology, the dance, matters of diction,
prosody and gesture, physical surroundings, and above all large amounts of

music in a narrow range of styles — . . . one discovers that features which could
not be reconstructed on the basis of any documents are somehow shaped by
the pressure of everything else that is right about the performance."^^ Do you
have any view on this?

Yes, I talk about specialization a great deal. I use the word "specialization"
far more than the word "authenticity," which I hate. I'm very specialized in my
own way. I may conduct a Missa Solemnis or a Zauberflote, but even then I

feel very secure about all the components. I feel most at ease when I'm in a

context, and that context has to do with specialization. For people at the end
of the twentieth century who are dealing with music that's three centuries old,
total immersion is very important. I've been with French style for a long time,
and most of the pupils I work with are in this field because they love French
music. They're not doing it because it's fashionable now, or because it's a good
idea to do it; it's something they actively do themselves, and this you have to
do. They probably can talk a great deal about what was going on elsewhere in

the culture of France: they have notions of painting, sculpture, literature, his-
tory — all of these things are very important.
A case in point: I heard a production not too long ago of seventeenth-cen-
tury French music. The cast was a motley crew that didn't have much style, but
the biggest problem was a German orchestra that essentially had no notion
whatsoever of playing in the French style. In many places, because the orches-

tra seemingly had no regard for the fact that they were playing texted music,
it became unbearable to listen to.

Another thing you've done is emphasize ensemble, not famous divas or

divos; you 've had great interest in stylistic homogeneity.
Well, the greatest sin we're experiencing now, in a lot of new groups, is

that they seem to think that with a big orchestra and fancy soloists you can

17. David Fuller, "Ornamentation," in Companion to Baroque Music, ed. Julie Ann Sadie
(London: Dent, 1990, and New York, Schirmer: 1991), p. 433.
268 THE BAROQUE

whip into shape Rameau's Hippolyte et Aricie, or give a convincing perfor-

mance of a Charpentier/Moliere, or do a marvelous Monteverdi Poppea.
What's sadly lacking is the idea that you need homogeneity of style. It's so
obvious when you hear any kind of group where people were doing just
that. Just take a look at jazz of the 1930s or 40s or 50s: that's a real spe-
cialist medium. The people who created it knew only how to do that one
style, but they did it instinctively, reflexively, with an extraordinary sense of
ensemble. That always amazes me when I listen to old jazz recordings or to
old tango people in Buenos Aires — this fabulous, extraordinary sense of fit-

ting with each other, making a style which is so coherent, so easy, so effort-
less. The difference is enormous between that and what I heard a couple of
months ago in a performance of Hippolyte et Aricie presented by a young
colleague of mine, where there was no attempt to coordinate the main soloists

in any stylistic sense, and there were wild differences of vocal style and tech-
nique. It doesn't work, because essentially it's as if the opera had become
polyglot, as if people were singing in different languages. It's that blatant.
And I said years and years ago that probably the most wonderful element of
French art is its extraordinary unifying rhetoric and style. Given wildly dif-

fering interpretations, people were unified and linked by so many common

things — rhetorical things, essentially. That's something I try to give off to the

ensemble. And of course it's easier to do with younger people than it is with
older people. If you have people who have wonderfully individual personali-
ties, but who are united by a common aesthetic, it's far more interesting than
the contrary.

/ suppose one of the results of specialization may be the element of the im-
provisatory, which mastery would allow.
Of course, things become spontaneous and reflexive —two very important
words, I think. And this imparts a sort of easy sense that one has with the
score, especially scores that are incomplete —with these it becomes very appar-
ent. Being at home with the idiom allows you to be fluid with it.

That's really the answer to the complaint that they couldn't have overdot-
ted and stayed together.

Musicology of that type is living in the office, in the sense of looking at the

score and not living with the reality of the score played. And one can open
one's eyes so easily just by listening to ensembles that have been playing to-
gether for a long time. I can ask a choir with fifteen sopranos to sing very com-
plex ornaments and they do it perfectly well. To get an entire orchestra play-
ing notes inegales in a Rameau dance is not difficult.

The degree of documentation for French Baroque music is much greater

than for Bach or Monteverdi. Could you summarize some of the areas in which
this applies?
 AT HOME WITH THE IDIOM 269

That would take hours to describe. But it's true that we have immense quan-
tities of treatises from the 1680s through the entire eighteenth century about
how to play and sing, and how not to play and sing. Some of these are quar-
rels, silly texts in which someone takes somebody else to task. Some are dread-
fully simple, because they're written for provincial amateurs, who want to
know how to do it as they do it in Paris; you get these sort of plain lessons.

But some are much better. have certain bibles I rely on constantly, such as the
I

Remarques curieuses sur I'art de bien chanter by Baciily (in 1681): a kind of
primer for voices, and use other texts to round it out. The resources are more
I

complete, for example, than in Italy —absolutely. There are German treatises on
the Italian style in the mid-seventeenth century, which are good but much less

complete, of course, than one finds among the French.

There are German treatises about French style, too, of course, such as Muf-
fat's. I want to ask you about the correspondence between Graun and Tele-
mann (a lifelong student of French music) about French style.
It's a good text, because it's one of the rare documents where you really get

a few glimpses of what staging was probably all about. Also there's the busi-
ness about the quarter-tone and the portamento slides.

/ was interested in its hit about how the frequent changes of time signature
in the recitatives were necessary for putting the stresses on the strong sylla-

bles.^^ The issue I want to raise has to do with how much additional tempo
flexibility is called for in recitatives, over and above notated signature changes;
at least one musicologist has called for limiting that flexibility.

You can't sing recitative without an extraordinarily free sense of tempo.

There is no fixed movement. The first thing I tell singers is that they've got
to read what's on the page without confusing rhythmic precision as written
by the composer with rhythmic freedom in terms of the overall line. You can't
sing drama if everything is being beaten in a metronomic way. And that's a

problem with a number of conductors today —who after we've spent years
trying to get away from the conductor's tyranny — are being tyrannical with
the Baroque groups. People are conducting three hours of recitative. Yet there

are no indications whatsoever that the batteur de mesure beat time in the
recitative; he was there to maintain order in the ballet and the large choruses.
Remember also the division they had, the breakdown of the orchestra into a
small improvising orchestra, the petit choeur, within the very large one, the
grand choeur. That's essentially to allow the petit choeur to be supple and to
hug on to the continuo during the recitative —without the intervention of a
third party. Of course there would be immense tempo fluctuation.

Goldoni, on his first visit to the Academic Royale de Musique in Paris, was

18. Lois Rosow, "French Baroque Recitative," p. 468.

270 THE BAROQUE

unable to tell when the arias were, and thought it was all recitatives —I'm won-
dering about tempo flexibility in arias.^'^

I more organized, and

think that because in the arias, the bass hnes are far
there some element of melodic and rhythmic regularity, you are probably
is

going to have more regular rhythm. Goldoni had a problem because it is very
different from Italian music. Though arias could be long, an aria could often
—
be just five or six measures of more organized music and text refrain texts or
moralizing texts — and might be over before Goldoni could notice it.

Along with material about orchestra size, bowing, tempo, and layout, don't

the treatises also discuss choreography,hand gestures, and so on?

I don't think we know much about hand gestures in France in 1670; we

know a lot more about hand gestures in Germany in 1780. That's a field that
is painful for me right now, I must say, the whole idea of gestural art, espe-

cially when it comes to the late-seventeenth-century style. We have a few good

iconographical references, and certainly the old adage about a good statue or
painting can be taken to heart. But as for how people used hands and bodies,
how they moved onstage, and the bigger issues of what do you do in terms of
staging — that's a terrible gray area. There are a very few things —a few French
burlesques, a few stage directions —and these essentially just tell you who came
in from the left and who came in from the right, and how long they stood on-
stage. Most of the information we deal with in terms of gesture and steps is

not French and is of a later epoch.

Do you think that much of what is done in the name of Baroque gesture is

conjectural?
I think that it has to do with conjecture, and I think that sometimes it ap-
plies, as I said, to a later date or to some other place. I think it's very neces-
sary, but I also think that you simply can't come in a month before a perfor-
mance and expect someone to learn the essentials of rhetorical gesture and then
make it convincing for the audience. In a lot of these productions I've seen peo-
ple look like berserk windmills.

I think there's another side of the coin, too; there's a great need for many
of the stage directors to do some homework, to find out what remains of con-
temporary information about Baroque theatrical comportment, rather than
doing wildly excessive and exaggerated staging because that's what the twenti-
eth century requires of stage directors. When the stage director becomes more
important than the music and more important than the composer, I have to
balk a great deal.

19. Carlo Goldoni (1706-85), playwright, librettist (to Galuppi, Salieri, and many others),
and "father of modern Italian comedy." He moved to Paris in 1762, and he recounted this
tale in his 1787 memoirs.
 AT HOME WITH THE IDIOM 271

Could you discuss French Baroque opera's socio-political origins — the pa-
tronage of the aristocracy in the seventeenth century versus the growing influ-
ence of bourgeois audiences in the eighteenth?
I don't really want to comment on that — it's a very difficult issue to talk

about, especially on the telephone — but I don't really believe simply that the
tragedie lyrique and tragedie were exclusively aristocratic forms. The court
commissioned the big Lully works — that's a historical fact — and Lully was the
superintendent to the king, and he wrote a tragedie lyrique a year for the court.
But the fact is that while you can call this music aristocratic, you can also call

it in some ways popular. And certainly he wasn't aiming only at a very narrow
part of society, the French court. The fact is that the music in its own time had
this extraordinary popular allure. People said that every bootblack and chim-
ney sweep was whistling a Lully tune on the Pont-Neuf a few days after the
opera premiered. Don't forget, these things were also performed for a bour-
geois audience in Paris —almost simultaneously in many cases. The parodies,
the pastiches, all this gives you the idea that there's more to it than just aris-

tocratic entertainment. Lully, after all, becomes a national figure very early;
even forty or fifty years after his death he continued to be canonized, not only
by the aristocracy, the literati, and the musically educated, but by everybody in

France.

How about the idea that the opera was a way of propagandizing for the
ancien regime.^
Well, I think it obviously had that value. I think every society has a cultural
appendage that helps it out and is its own mirror. But to think that Lully be-
came famous and stayed famous for the better part of the eighteenth century

just because he mirrored the aristocratic, Versailles court ideal is stupid. This
music does have a very strong popular history as well: Lully as the man on the
street. His was the song that everyone was humming. That's something one
tends to forget.

As revered as Lully was in his day, today people often express a preference
for Charpentier, and in the process patronize Lully. Are we missing something
in Lully or was he overrated in his time? Is it our distaste for his personality,

his political machinations, or for his music?

I haven't figured that one out yet, quite frankly. I find that there are times
when Lully simply knocks me over, by his extraordinary sense of equilibrium

between simplicity and doing essentially what the form wants him to do, in the

tragedies lyriques, to give a wonderful meeting of words and music. There are
other times when Lully bores me. There are times when I think, yes, Quinault
has wonderfully well-wrought libretti in the tragedies lyriques, but there are
others with very weak endings. I wouldn't want to do a Phaeton or Bel-
lerophon, as I find that in a curious way they just collapse at the end. So there's
272 THE BAROQUE

verygood and very bad. Sometimes the four-squareness of the music gets me
down; I don't Hke it. But Atys and Armide I think are two of the finest exam-
ples of musical theatre in the world.
If I had to take ten scores of Charpentier or Lully to a desert island, I would
probably take Charpentier. I think it's musically richer, more satisfying, and
more individual. There's a musical personality that's more complex, more a mu-
sician. But then we're looking at these things very differently than they were
looked at then. And Lully's extraordinary simplicity is something that the
French revered. "Simplicity" is not a word that we like today.

This brings up the question of relating French opera to modern audiences.

One issue is French audiences, who understand the words, versus other audi-
ences; didn't you use supertitles in Medee in New York?
Yes, but I don't believe we used supertitles in Atys in the beginning. My im-
pression was that Atys had an extraordinary success; you had people in England
and France saying it was a milestone, a watershed —
there's prt-Atys and there's

^osi-Atys. And was an enormous success not only in France but abroad, in-
it

cluding the States. I think that essentially it has to do with its being a good show,
well wrought, with fine singers and fine costumes and very good staging, and,
again, this broad aesthetic coherence, a very tight structure, and a total bond-
ing of the balance of dance and theater and music. People had given some
thought to what they were doing and what their partners were doing. And that
was with very good raw material. Atys and Medee are fabulous theatre pieces.
They're very convincing. They deal with issues that are as relevant at the end of
the twentieth century as they were at the end of the seventeenth. We adapted
them, obviously; I don't claim for a minute that we were interested in doing a

historical re-creation — by no means. But I think the essential is that there's a

strong text to be communicated which includes very strong emotions which one
tries to play upon and provoke among the audience.

SELECTED DISCOGRAPHY
The recent French Baroque revival rescued Charpentier from the library shelves;
to many modern ears he seems its greatest discovery. To sample both him and
Lully, ambitious listeners might invest in the 3-CD recordings of the tragedies-
lyriques discussed above — Lully's Atys (Harmonia Mundi 901257-59; a single
disc of excerpts, HM 901249, lets one hear the great "Sleep" scene) or Char-
pentier's Medee. In the Erato CD booklet for Medee Christie calls it the "most
important of all [Charpentier's] works," while H. Wiley Hitchcock says it is

generally agreed to be his"consummate masterpiece." Christie's 1984 Harmo-

nia Mundi Medee (Harmonia Mundi 901139, 3 CDs) won a Gramophone
Award; but he re-recorded the work ten years later in a performance that is
widely preferred (Erato 96558, 3 CDs; excerpts, Erato 99486). The composer
 AT HOME WITH THE IDIOM 273

Eric Salzman writes of Christie's second Medee, "You will rarely hear a dra-
matic work treated with as much depth and passion as this."'"
Those wanting a gentler (or less expensive) introduction to Charpentier
might sample some of his sacred works, such as the hauntingly beautiful Christ-
mas Pastorale (HM 901082) or some lighter Charpentier, such as one of his

iMoliere collaborations, the intermezzos to Le malade imaginaire. This record-

ing (HM 901336) lets us hear Christie's own well-timed comic acting; he plays
a part in Moliere's still-funny satire of a medical-degree examination. The scene
contains the famous exchange about why opium puts people to sleep: the
"Bachelerius" impresses his examiners by answering that it's because "it has
sleep-inducing powers."
Another beneficiary of Christie's work has been the greatest French opera
composer of the eighteenth century, Rameau. You might begin with Castor et

Pollux (HM 901435-37, 3 CDs; excerpts, HM 901501)— according to Barry-

more Laurence Scherer, "a gem of its kind . . . melodically ingratiating, rhyth-
mically vivacious"^' —or with the Gramophone Award-winning Grands Motets
(Erato 96967). The 3-disc Les Indes galantes (HM 901367) received an inter-

estingly mixed review from Graham Sadler, who is enthralled by the work (it

"raised the traditionally lightweight genre of opera-ballet to a new level") and

pleased with the performance — but with reservations. Sometimes, he says, the
"polished and stylish" orchestral players "do not characterize the music as
vividly as they might"; and he says that as with all modern Rameau perfor-
mances, in this one "we have still to hear anything resembling a 'period voice'
or the full range of vocal ornamentation."'^
Christie has also recorded, on Harmonia Mundi, some neglected lesser mas-
terpieces of the French Baroque, notably Campra's Idomenee (HM 901396-98,
3 CDs; excerpts, 901506), Monteclair's ]epthe (HM 901424-25, 2 CDs),
Bouzignac's strange, fierce motets (HM 901471), and much else.

Also significant are Christie's forays into non-French repertoire, such as his
Handel Messiah (HM 901498-99), with its unforced alertness to the implica-
tions of the words. Christie's Mozart is unlike that of anyone else either in the
mainstream or the early-music movement. Stanley Sadie praises his Requiem
(Erato 106972) for being "ready to make the most of changes in orchestral
colour or choral texture, to mould the dynamics more than the (very sparse)
original indications, and indeed to dramatize the music to the utmost," and for
treating the music as "operatic, almost romantic."-' Other critics have com-
plained about this; Elliot Hurwitt found the performance "puzzling," with
some "incredibly slow tempos."'" Regarding Christie's well-cast Magic Flute

20. Salzman, "A Medea for Our Time," Stereo Review 60 (September 1995), p. 90.
21. Scherer, Gramophone 71 (North American edition), September 1993, p. A2.
22. Sadler, Early Music 20 (May 1992), pp. 353-54.
23. Sadie, Gramophone 73 (November 1995), p. 142.
24. Hurwitt, Fanfare 19 (January/February 1996), p. 265.
274 THE BAROQUE

(Erato 12705, 2 CDs), Sadie writes of "light and soft textures and graceful
phrasings," and says, "Some may find Christie less readily responsive [than
some other conductors] to the music's quicksilver changes in mood, but this is

part of his broad and essentially gentle view of Die Zauberflote," which falls
"^^
"more sweetly and lovingly on the ear than any I can recall.

Christie's Purcell Fairy Queen (HM 901308-09, 2 CDs) was decried for its

liberal use of French rhythmic conventions and both praised and blamed for its

unabashed theatricality. Some Purcell lovers regard King Arthur as a greater

work than The Fairy Queen, and Jonathan Freeman-Attwood regards Christie's
Gramophone Award-winning recording (Erato 98535, 2 CDs) as making "the
strongest case for this music to date."^*^ Eric Van Tassel prefers Gardiner and
especially Trevor Pinnock in this work, but praises Christie as offering "what
I missed in [his] 1980s readings of Dido and Aeneas and The Fairy Queen: a
true marriage of French and English seventeenth-century idioms, evidence that
this group's long and passionate engagement with the music of Lully and Char-
pentier can throw new light on their greatest English contemporary."^^ This

light is also present in Christie's vivid 1994 recording of Dido and Aeneas
(Erato 98477).

FOR FURTHER READING

The basic English text on French Baroque music remains James Anthony's pi-
oneering French Baroque Music (revised edition, New York: Norton, 1975); it
may be time again for the author to update or expand it. French Baroque
singing has not yet received a clear modern exposition in English, although
rumor has it that Thomas Grubb includes a section on it in the as-yet unpub-
lished third edition of his well-known textbook Singing in French. Albert
Cohen's bibliography of recent writing in English (and other languages) about
French Baroque music appeared in Performance Practice Review 1 (Spring/Fall

1988), pp. 10-24. This includes, among many other sources, some ambitious
French-language books on French Baroque performance. The foremost study of
French rhythmic conventions is Stephen Hefling's Rhythmic Alteration in Sev-
enteenth- and Eighteenth-Century Music: Notes Inegales and Overdotting
(New York: Schirmer, 1993).

25. Sadie, Gramophone 73 (May 1996), p. 119.

26. Freeman-Attwood,Grawo/7fco«e 72 (May 1995), p. 110.
27. Van Tassel, Fanfare 19 (September/October 1995), pp. 282-84.
 15
Triple Counterpoint
— NoOo^
Jeffrey Thomas, Philippe Herreweghe,
and John Butt on Singing Bach

"It is right, however tedious," writes Joseph Kerman, "that discussion of his-

torical performance should always keep circling back to Bach. Musicians in

general care as deeply about Bach's music as any other; they know they are at-
tuned to its 'spirit' and consequently have strong feelings about its interpreta-

tion.'"
Consider Bach's choral works. It's easy to forget the strong feelings aroused
by Nikolaus Harnoncourt's pioneering mid-1960s recordings, using small cho-
ruses and period instruments;- among those who reviled them, quite a few were
missing the grandeur and massiveness of large choruses, and the slow tempi
they required. Today, though, it is large choruses that are often dismissed, with

such terms as "elephantine," "gargantuan," and "bloated." Perhaps we might

also call them "brontosauran," since their Bach performances have all but dis-
appeared from the recording studios. New recordings of the B Minor Mass usu-
ally have choruses of about twenty-five, which a majority of historical per-
formers have regarded as close to what Bach wanted. Many listeners (to judge

from what the record companies think will sell) seem now to prefer the trans-

parency and litheness of small groups, which serve Bach's polyphony and
dance-based rhythms well — although early-music devotees sometimes forget
that the older approach did yield some great performances.

1. Contemplating Music (Cambridge: Harvard University Press, 1985), p. 203.

2. For example, the musicologist Paul Henry Lang (High Fidelity, July 1969, p. 77) called
one such recording "pitiful."

275
276 THE BAROQUE

Small chorusesmay command the scene today, but Bach's choral works are
still some respects embattled terrain: the arguments keep circling back to
in

Bach. For some listeners, historically informed performances "are so supple and
elegant, so refined and light, [that] they miss the gravity" implicit in some of
Bach's scores.^ Among the historicists themselves, the performance-practice is-

sues are by no means resolved. Chief among the problem areas (and I am ig-

noring minor ones, like those involving Bach's continue group) are the very
concepts of "singing" and, even more, of a "chorus." This chapter examines
this pair of issues, and others, including the question of historical authenticity
itself. The discussions bear out Kerman once again: the feelings stirred up can
be, as you will read, quite strong.

"A Little More Direct"

Jeffrey Thomas and the American Bach Soloists

The tenor Jeffrey Thomas, a Juilliard graduate, did not begin his career in early
music. He spent three years in the early 1980s with the San Francisco Opera,
where he won a prestigious Adler Fellowship. Since then, he has performed
with most of the major US orchestras, under such conductors as Zinman,
Ozawa, Blomstedt, and Shaw; he has also been the dedicatee of new vocal
works by such composers as Ned Rorem. But Thomas is now best known for
his early-music work with Hogwood, Koopman, Leonhardt, McGegan, Nor-

rington, Parrott, and many others. His central focus is clearly the American
Bach Soloists (ABS), which he founded in 1988 with the organist Jonathan
Dimmock, to give American early-music performers a domestic arena for ex-
ploring Bach cantatas. As a Bach conductor (and singer, for that matter),
Thomas has demonstrated a special concern with conveying character.
In our interview, he discussed not only his approach to Bach performance,
but the larger issue of authenticity. Many in the early-music community will
bristle at his rejection of the ideal of historical re-creation — his attitude can
lead, many would argue, to complacency —while others in the community will
applaud it as a sign of the movement's maturity.
I began the interview by asking about Bach's Cantata No. 198, the Trauer
Ode, the high point of an ABS cantata disc I had just received.

In your recording of Cantata 198, in the opening chorus, your singers sound
as if they're really in grief —which is not the case with two other recordings I
compared it to.

3. Edward Rothstein, "CDs in the Spirit of the Easter and Passover Season," The New
York Times, 7 April 1995, p. B16.
 TRIPLE COUNTERPOINT 277

I hope this is one of the things that come across about any of our perfor-
mances. If critics want to blame us for something, let it be that we try to be

much more emotionally direct, as in the first chorus of Cantata 198 when they
scream [sings] "Lass, Fiirstin" with all they've got. The same goal applies to
the instrumentalists. We get them to really make some sound. It just kills me
to hear what often happens with these beautiful old instruments — if people
don't play all the way to the bottom of the sound the instrument can make, it

seems like a waste.

In that same cantata, there's a recitative, "Der Glocken bebendes Geton,"

about death-knells and the terror of the soul. Unlike the other recordings I've
heard, the feeling of terror comes through in your instrumental accompani-
ment.
I tend to ask them to play a lot more soloistically. It's a smaller group, but

I don't necessarily think it should sound small. So I ask them to play a lot more
deeply than they might in a different setting with a different conductor. Also,
it's hard for the players sometimes, right off the bat, to play a phrase as long
as I'd like. The opening of the Agnus Dei in the Mass in B Minor nowadays is

usually played in half-bar phrases [sings, separating the aria's motivic fragments
in an extreme way] — but in our recording, we're back to the four-bar phrases
people used to play. We're not trying to romanticize it, but I don't think there's
anything wrong with a phrase being a few bars long."
I think that what an audience wants from a concert or a recording is to get

something larger than life, something more than going to work at nine and
coming home at five. I think that's the function of art in our society.^ It's relief

from the gray and the noise and the din. Now, there had been — necessarily so
a lot of careful and cautious p[aying in the ear[y-music movement. They were
trying to do something historically correct, and one doesn't want to bewrong
in something like history. But one result is that in the singing world right now,
and I think justifiably, there's more and more criticism of Baroque music

4. John Butt, in his Cambridge Music Handbook Bach: Mass in B Minor (Cambridge Uni-
versity Press, 1991), p. 61, points out that the aria's opening (ritornello) is binary —with "an-
tecedent" and "consequent" halves —and that those halves are each four bars long.
5. According to Lydia Goehr, demand has been widely made of Western
this only art for

about two hundred She

years. the "separability principle" —
calls it "separated com- art as

pletely from the world of the ordinary, mundane, and everyday" — and discusses emergence its

in her The Imaginary Museum of Musical Works (Oxford University Press, 1992), pp. 120-75.
In a sense, the "separability principle" seems always to have applied in the church setting, as
I suggested in discussing the meditative function of Gregorian chant, in the introduction to
Part One of this book. What Goehr describes is the emergence of the principle into the secu-
lar sphere with the nineteenth century's "sacralization" of art — to be discussed in the Nor-
rington interview. In this sense, there may be some continuity between what Thomas is try-

ing todo in concerts and what Bach did in church. As Thomas pointed out to me, there are
obvious and crucial discontinuities too: Bach's purposes involved encouraging the listener to
be a better Christian.
278 THE BAROQUE

singers. I think in Europe they've made some advances that we haven't yet here,
in that singers there are singing with more voice now. We still have a lot of
singers in the USA who sing stylistically very differently when they approach
music from before 1760. I was certainly guilty of that for many years. A decade
or so ago, when I started, I thought what conductors wanted was [he carica-
tures a very light-voiced tenor]. And I think they did want that. Now it does-
n't sit as well. But we're evolving, and getting to be more direct about it.

Of course, the biases are hard to get over. A couple of people criticized our
B Minor Mass for being very "romantic." The greatest compliment I got was
from one local musician, who left at intermission because he hated the tempo
—
of the opening Kyrie it was too slow. And I said, "Just because they did it
that way for the last eighty years doesn't mean it's bad." I personally don't
want that piece to be over too soon.

The way you articulated the fugue subject was full of the emotion of plead-
ing for mercy, and when you articulate that much, I suppose you need to give
it time.
It has those ascending pairs of notes. I wanted there to be a sort of effort

to get up to the top. We sort of elongated each note, "E-/e-i-son."

How do you approach issues of phrasing and articulation —aside from the
slurs Bach wrote?
There are two things that strike me about Bach's writing, whether in an aria
or a recitative. One is the rhetorical element, which I think our approach to
the Kyrie theme illustrates —though that's sometimes overdone nowadays. The
other is the harmonic element. From the first time I opened up a Bach score,

the vocal lines looked principally like arpeggiated harmonies, certainly in the
recitatives but also in the arias. This was always a clue to me about how to

sing Bach.
Imagine you're in a decent acoustic. And
if you have B minor for two beats

and then finally you get an At or some other dissonance, that's the first
a E| or
interesting bit after two beats. So what I ask people to do is to sing harmoni-
cally. For example [singing both]:^

(a) (b)

i!
Uj M
Be - ne -
F
die -
J^

tus
¥^^LU M^ ^
Be - ne - die - tus

The Ct is just a passing dissonance, and the B and D that follow are still
first

in the same B minor chord, so these three notes shouldn't be accented; it's the

6. Bach, Benedictus, from the Mass in B Minor: (a) as Jeffrey Thomas recommends singing
it, (b) as it is sometimes sung.
 TRIPLE COUNTERPOINT 279

Cl in the second bar that's the harmonic event. That's made clear in the first

way of singing it (a) —that Cl should get the first accent since the initial D.

This is obscured in the second way (b)/

On the other hand, I don't want to start reading into Bach all kinds of
things that aren't there. I don't know how often Bach counted the number of
notes or measures, and when he did, how much of that was in fun or seriously

intended. In some cases I'm sure it was deliberate; in others ... mean, come I

on, he was very clever. I don't want to read too much into music that's not on
the most immediate level, the level of the function it's trying to fulfill.

Regarding that function, there is of course an important distinction between

Bach's church and our concerts. In Bach's church, people were meant to par-
ticipate by taking the message to heart, not just by enjoying the music. His pur-
pose in setting the words to music was to frighten or inspire the congregation,

or whatever the text was supposed to do. He was being judged on how well
he conveyed the Lutheran message. That was his job. We don't ask the audi-
ence to subscribe to the words; the modern Bach audience is usually interested
in the aesthetic/artistic element. So what we're trying to do at ABS is take the
immediate content of the music and make it come alive for our audiences.

But do you use musicological evidence in preparing performances?

Yes, but I've always felt that musicology and performance make strange
bedfellows. Of course, their relationship in recent years has been very pro-
ductive. But there is a point at which each must take its own path. To make
a performance be about a work's stylistic aspects is to trivialize it.

Besides, musicology is one of those disciplines that can prove anything.

You can take opposite sides of an issue and find treatises that support both.
I don't want to get into the whole business of what's right and what's wrong

we're just trying to give really good performances. I'm not suggesting that
other groups aren't doing that either, or are trying to prove points, but we're
definitely not trying to.
Of course, there are certainly issues about which I've gone back and re-
searched things — but again, the results are rarely conclusive. Consider the fa-
mous memorandum that Bach sent to the Leipzig Town Council in 1730**

where he said that "it would be better if there were four subjects for each
voice." Joshua Rifkin argues that he wanted those singers so he'd have enough
to do motets when people got sick, and that he meant for his choruses to be
sung one to a part. We've done it both ways, one to a part and with a cho-

7. The C? Thomas accents is on the first beat of the bar. This exemphfies the Baroque doc-
trine of "good" beats and "bad" beats —the first beat is the best of the "good" beats. It also
concurs with the view that the doctrine harmonic practice, so that significant disso-
reflects

nances fall on strong beats. John Butt discusses this in Chapter 9.

8. See John Butt's section of this chapter, below, for a thorough discussion of this and of

Rifkin's hypothesis.
280 THE BAROQUE

rus. Ultimately, I'm not concerned about what Bach did, but about the artis-

tic results now.

How does your attitude apply to your continuo groups?'^

Well, there's more and more evidence of rather large bottom-octave config-
urations in that period in other parts of the world, and plenty of evidence for
it in Bach also, as Laurence Dreyfus has argued — not that, again, I try to fol-
low him exactly.^" I think nowadays people tend to under-balance the bass line
very often. For a recitative, we almost always try to have a cello and sixteen-
foot violone, and bassoon, and harpsichord, and organ; and yes, we even have
lutes playing.

Have the lutes caused people to walk out?

I gave Roger Norrington a copy of our first record (we were doing Berlioz's
Romeo et Juliette in Minneapolis). Sure enough, he played it that night in his
hotel room, because the next morning at rehearsal he said he enjoyed it very
much. Then he said, "Now tell me about this lute business. Are there histori-

cal reasons for that, or do you just like the way it sounds?" I said, "We-e-e-11,

you know, the famous lutenist Weiss visited Bach's sons for several months, and
there were certainly instruments around, and Kuhnau, Bach's predecessor in

Leipzig, is known to have used lutes in the continuo, and although there's noth-

ing definite to say that it was a regularly played continuo instrument, um . . .

yeah, I just Hke the way it sounds." Again, there are reasons to justify it, but
they aren't the ultimate criteria.

Searching for the Balance

Philippe Herreweghe

Edward Rothstein says that one of Philippe Herreweghe's Bach CDs "manages
to encompass the weightiness of the old [mainstream performances] and the
highly refined language of the new [historical ones.]"'* Reviewing another of
the Flemish conductor's Bach recordings, Gerald Hansen puts it differently: he
finds the delivery "romantic but ingratiating." He warns, however, "Purists, be-
ware."'^
In fact, some period-instrument experts cannot stand Herreweghe's work.
His mainstream appeal does not result, however, from any ignorance on his

9. See note six in the Alan Curtis interview for a definition of "continuo."
10. Dreyfus, Bach's Continuo Group (Cambridge, Mass.: Harvard University Press, 1987),
argues, among other things, that Bach sometimes used both harpsichord and organ in the con-
tinuo of his Leipzig cantatas. Dreyfus's conclusions are somewhat controversial, however.
11. Rothstein, "CDs in the Spirit."

12. Hansen, American Record Guide, March/April 1992, p. 20.

 TRIPLE COUNTERPOINT 281

part about Baroque performance practice. Herreweghe, a former conservatory

student in piano (he also has a degree in psychiatry), formed his chorus, the
Collegium Vocale of Ghent, in 1975; it attracted the attention of Gustav Leon-
hardt, who used it in recording a number of Bach cantatas. In the process, Her-
reweghe served a kind of apprenticeship with Leonhardt, whom he admires
enormously. Herreweghe is deeply concerned with the idea of rhetoric as the
key to Bach, but he uses it in a very different way from Leonhardt and Harnon-
court. His style could hardly differ more from theirs, as our discussion shows.

You once wrote that "for many years the definition of Baroque style was
shaped by harpsichordists and violinists who incited singers to imitate their
manner of playmg": a useful service, you thought, but causing "a certain loss
of melodic line."^^
I would only add that when people read eighteenth-century treatises about
the interpretation of music, they can read into them what they wish. When an
explanation about music is only in words, very often one can exaggerate, as
people in the eighteenth century did about, for example, articulation. And if

you're a harpsichordist, you might read it in a certain specific way that is dif-

ferent from how a singer might read it. I think this was the origin of a mod-
ern style of Bach performance. This style was necessary for getting us out of
the previous styles — first the very romantic one, and then later the very mo-
toric one. In this new style, though, many conductors were harpsichordists; and
while they brought very important things into the playing of this music, from
the beginning there was something in my opinion too edgy, too angular.
Of course, the approach has something useful in it, where you consider an
architecture based on the addition of small elements —that is something that is

special to Baroque music. My point is to find a balance between organizational

cells that are too small and the overly long line. I think I'm still searching for
that balance.
Also, I think that some instruments like voices, flutes, oboes, and even
strings tend not to sound as they should when played with this overly static

rhythmic approach. So you need a line, because I think that a line is a very im-

portant part of the architecture of the music, and also because without that line
voices do not sound as they should.

I think Bach tried to write something that was a musical enlargement of

how you speak the text. The verbal phrases tell you how you should phrase
musically. Of course there are also small rhythmic impulses that come from the
music. At the same time that it's a purely musical kind of architecture, though,
the text gives exactly the right approach. I think you can make Baroque music
with Baroque aesthetics and still pay attention to vowels and phrases. Now,

13. Booklet note to his recording of the St. John Passion, Harmonia Mundi 901264-65.
282 THE BAROQUE

that's a matter of debate. Some people say that my approach is too romantic.
But I'm convinced of it.

You've written that rhetoric is the only valid key to understanding the
Baroque aesthetic.^^

I always felt this by intuition when I worked on early music with other peo-
ple, but gradually I came to feel it was really central, and then I read as much
as I could about it. It's very obvious that rhetoric was central to Bach's man-
ner of composition and also interpretation. The system of rhetoric took differ-
ent forms over the ages and even over just two centuries of Western music; you
can find these forms in theoretical books from the period. But I don't think it's

just a theory: it's a musical reality. From the middle of the sixteenth century
until Bach, it gives the key to what should be done in the music, especially
music with a text.

For example, it can be useful in, for example, the first Kyrie from the B
Minor Mass to know that Bach himself wrote the first chorus exactly like a

classical speech. It helps your interpretation if you know that a certain section
is the confutatio, which is a modulation — in a speech, the confutatio is done
with more tension than when one is first exposing one's theme. ^^

Could you talk about the issue of hoys' versus women's voices in Bach?
Yes, at the moment I'm convinced that we should use authentic instruments,
because I find they are more appropriate to defining the sound and thus the
message of the music. For the voices, I'm convinced in theory that boys' voices
are the most appropriate for both soprano and alto arias in Bach. I could hear
that twenty years ago, when there were very good boys' choirs singing Bach.
Today, English boys' choirs are very good, but they cannot sing Bach well be-
cause the pronunciation of the German is so important. But in Germany, and
in Belgium (where the language is Flemish, which is pretty close to German),
the boys' choirs are simply disappearing, because the organization of society is

very different now

was twenty years ago. I think nowadays there are
than it

no boys who have the musical training that boys had in the time of Bach. The
boy singers before may have been very narrow, because they didn't do sports

14. Booklet note to his recording of the St. Matthew Passion, Harmonia Mundi
901155-57.
15. I wasn't able to confirm this with Herreweghe (he did not comment on this footnote
when he looked over this chapter), but he might analyze the Kyrie movement as follows: in-
troductio, bars 1-4; expositio, bars 5-47 (the orchestral ritornello and vocal exposition, all in
B minor); confutatio, bars 48-72 (the section where the voices continue while underneath
—
them the orchestra repeats the opening ritornello but in Ft minor, the "dominant" key); con-
firmatio, bars 81-101 (vocal exposition repeated in B minor); conclusio, bars 102-26 (open-
ing ritornello in the orchestra again, with voices continuing overhead but this time in B —
minor.) See also Joshua Rifkin's discussion of rhetoric in his interview, and John Butt's in
Chapter 9.
 TRIPLE COUNTERPOINT 283

and so on, but they did the music very well. And now we don't have boys from
the age of six or seven studying singing technique every day seriously, and
singing every day three hours in church, and also studying harmony and coun-
terpoint as some did in Bach's time.'" A decade ago we could do it with Leon- —
hardt and Harnoncourt, the last of the tradition was there — but it's more and
more difficult now. Of course, you still have boys with beautiful voices, but
they're not really trained. And there are other factors: in the time of Bach, boys
could sometimes sing soprano until age 16 or 17, Now, because of various
changes including diet, the body develops much earlier, so the voice breaks be-
fore they are as mature as in Bach's time.
So I prefer to do Bach with girls who have special training. The women
singing in Collegium Vocale do not have the normal vocal training for singing
Brahms and Strauss, and while that may be a pity for them, it's good for

Bach —they've developed their voices in the context of polyphony. When you
only sing polyphony up to Bach, and you don't sing Mozart and certainly not
Brahms or Puccini, then your voice develops in a certain direction. That is what
has happened with the girls I work with. Of course, it's different from what
Bach heard; but it's the best we can do, I think.

Joshua Rifkin argues that Bach didn't use a choir as we think of it, but had
a solo quartet sing the choruses. What is your view?
I'm not a musicologist, and he probably can make better arguments, though
I haven't studied them. From the subjective point of view, I think that using
more than four sopranos detracts, because very often there's a dialogue be-
tween, say, the sopranos and the flute, and if there are too many sopranos it
can no longer be a dialogue and the flute disappears. But when you have three
people singing instead of one, the individual disappears and becomes something
other than the individual. This for me is essential and is part of the emotion.
Going between the individual singer singing the arias and three sopranos
singing the choruses gives a very important emotional contrast.

You've advocated conducting later music as a way of improving one's con-

ducting of early music.

If you're not able to conduct at all, like many specialists who never con-
duct except in early music, you can still do a good, interesting B Minor Mass,
because in a way it's very simple from the conducting point of view. But I think
that at a certain point being able to conduct well is an advantage. The rehearsal

16.The leading book on this subject is John Butt, Musical Education and the Art of Per-
formance in the German Baroque (Cambridge University Press, 1994). It suggests that only
the finest boy singers in Bach's time studied harmony and counterpoint. (By the way, it can't
be assumed that Bach always used boys for alto lines; Bach's pool of alto singers in Leipzig
included not only boys but also several males in their early twenties. See p. 6 of Joshua Rifkin's
booklet note to his recording of Cantatas 80 and 147, L'Oiseau-Lyre 417 250.)
284 THE BAROQUE

can go quicker, so you can give more time to essential points rather than spend-
ing a lot of time trying to get everyone together. Another advantage is that even
for early music, when you have more than twenty musicians the beauty of the
sound can be determined by the gestures of the conductor.
On the other hand, I think a lot of conductors of Romantic music face a
danger in Bach and other Baroque music, because it's essential that this music
come from the musicians themselves. It's essential in the architecture of the

music that each musician has to be creative and make music himself. If you
conduct "too well" in a certain way, you kill that. So it's a paradox: for con-
ducting Bach, the ideal is to have a lot of technique and then to forget it and
certainly not to use it; otherwise you kill what is essential in the music. In Bach,

if something is not possible without a conductor, it's a sign that it's not a good
interpretation.

On Singing Bach Too Well

John Butt

Bach's pupil J. P. Kirnberger, a theorist and composer, suggested in 1771 that

students use the Italianate composer J. G. Graun as a model for writing

melodies. They should not use Bach, Kirnberger said, because his works are so
adventurous melodically and "require a very special execution that is exactly
suited to his style; for, otherwise, many of his works sound hardly bearable."'^
What is this special execution? Julianne Baird and William Christie have shown
us that the Italian style of singing was the mainstream of the Baroque; if Bach's
style was "very special," how did it differ from (or relate to) this mainstream?

And how might it have differed from (or related to) today's styles of Bach
singing? These questions came up in the course of my Chapter 9 interview with
John Butt (who is an experienced choral conductor). I asked about the same
Herreweghe quote that I read to Herreweghe.

Philippe Herreweghe writes that "for many years the definition of Baroque
style was shaped by harpsichordists and violinists who incited singers to imi-
tate their manner of playing": he thinks this a useful service, but believes it

caused "a certain loss of melodic line. " What do you think about this?
To answer that, note first that there's not a great deal about the theory and
physics of the voice in sixteenth- and seventeenth-century vocal treatises; but
there is an incredible amount about ornamentation and articulation. In virtu-

ally all the singing technique books from 1560 to well beyond the time of
Mozart, the emphasis over and over again is on ornamentation: how to sing a
trill, how to sing a messa di voce, how to sing runs that are really articulated.

17. John Butt, Bach Interpretation (Cambridge University Press, 1990), p. 24.
 TRIPLE COUNTERPOINT 285

And a very important point is that they were articulated in the throat —the
epiglottis and glottis acted like a percussive device letting the air through. This
is anathema to modern singing, so from that point of view, indeed, singers
might learn a lot today from harpsichordists.

Now, you said in your hook Bach Interpretation"* that in the Baroque
singing was the basis for instrumental articulation, not the other way around,
which seems to support Herreweghe's point. But you also said that singing in

Baroque Germany was different from singing now. What do we know about
Baroque singing?
That's a huge question. I've tried to answer it to some extent in one of my

books,'" at least in the context of German church music and the very limited
environment of choir schools; but I've also studied the more professional, Ital-
ianate styles of singing. From the late sixteenth century onwards, there was a
very professional environment for Italian singers, with academies that taught
them counterpoint all morning and vocal exercises all afternoon. From that
point of view, I think Herreweghe is right to suggest that there was a quality
to singing that's missing from a lot of early-music performance. Of course we're
not talking about the whole of Baroque music here — just about the Italian

mainstream of singing, which was a lot more professional than the amateur ef-

forts you often hear in early music.

How about the German style, though?

Today, Italian opera singing and Wagnerian German singing are somewhat
blended; it can be rather hard to tell the difference. But it's quite clear that well
into the nineteenth century there was a difference between German singing and
Italian singing. The Germans were always criticized throughout music history
as being incredibly "vulgar" singers, not being able to tie their words together,
declaiming, spitting and hissing and gutturalling, rather than actually singing.

How does that apply to Bach? Would he have preferred Italian singers like
those in Dresden?
He was always very complimentary about the Dresden singers; I think he
loved their capability because they could get around the notes so well. But
whether he would have liked their style of declamation is a different matter.
The music they would be singing, such as that by Johann Adolf Hasse, appar-
ently a great friend of Bach's, would have been a different style from his.

Does this relate to one of the points from your book Bach Interpretation
that when the little motivic figures are articulated, the polyphony is made more
clear}

18. Ibtd., pp. 9-34.

19. Butt, Musical Education and the Art of Performance in the German Baroque.
286 THE BAROQUE

Well, to me it doesn't makes sense to sing Bach either in the modern Ger-
man declamatory way, in full voice, or in the traditional "early-music" sound
of very light vocalization, often with hardly any sound at all and no edge to it;

while perhaps the best Italian way of Bach's time would not be suitable for
counterpoint. So in some ways, Leonhardt, and indeed Harnoncourt too, while
they might be miles away from what was done in Bach's performances, might
on the other hand come quite close to it: a mixture of rhetorical singing and
enunciation with Italian flexibility.

Your point about "vulgar" Germans "not being able to tie their words to-
gether" and "declaiming" reminds me that when Leonhardt conducted your
chorus in Bach's Magnificat he was criticized for having them break the words
up so much.
Exactly. From that point of view, you might think of what Leonhardt does
when he conducts a choir as being rather like bad Italian singing. In other
words, it has many of the principles of Italian singing but all cut up into bits,

into articulative fragments. In terms of what Herreweghe might do it would

seem almost vulgar. The Italians, by all accounts, were more slick. They still
had the same articulative power, particularly in runs, but there was always a
great emphasis on messa di voce, on being able to sustain notes and not chop-
ping things up. Even Schiitz said, regarding some of his vocal "sacred concerti,"
that the players must listen to Italian playing because Germans cut the music
up too much. There he was talking about violins rather than singing, but the
same was true of singers.
The Germans were continually trying to emulate Italian singing. One of the
first places this appears is in Praetorius, in the Syntagma Musicum of 1619: he
says, "I'm going to give you a little guide to the new Italian fashion of singing."
And you find Germans talking about the new Italian fashion of singing right
into the mid-eighteenth century. Perhaps there were several "new" Italian styles
of singing; or perhaps it always seemed new.

But in spite of that, the German style remained different in that they de-
tached the figures and words more?
components of Italianate ornamental fig-
Yes, they certainly catalogued the
uration more carefully, suggesting (though never very clearly) that each requires
a different method of articulation. Moreover, they were very concerned with
enunciation and the textual aspects.

What factors gave rise to the differentiation between German and Italian
singing styles?
Well, it has something to do with the central role of German music in
Lutheran worship. There was more music in the Roman Catholic church, but
it didn't play the same rhetorical role that it did in the Lutheran church. In the
 TRIPLE COUNTERPOINT 287

Lutheran church you might think of music as being almost of the same status
as preaching,-" whereas in the Itahan church the music is generally of the same
status as the architecture; it's part of the building. If you're talking about opera,
on the other hand, there are a few more things to think about, particularly the

forms used in opera, and the cultivation of a presentation geared towards

drama.

Does the difference in singing styles also reflect the difference between the
languages —the melodious, vowel-rich Italian and the more guttural, conso-
nant-rich German?
Yes, and by extension, those particular languages quite often reflect the cul-

ture, the way of thinking, as well. It's not a question just of the language, but
also of the temperament that creates the linguistic community that will use lan-
guage of a particular kind, both in terms of its grammar and of its sound.
There are many different cultural issues intertwined, but certainly if you just

take the raw sounds of Italian and German they produce very different effects
in singing.

How would you relate this distinction to modern Bach singing other than
Leonhardt/Harnoncourt —the American Bach Soloists (ABS), for example?
I would say the ABS make a good attempt at showing how trained singers
can bring out certain things in Bach's music. But on the other hand, like vir-
tually any group using professionally trained singers, in some sense they're
slightly too good.

What does "too good" mean?

Well, nothing can be too good from some points of view. The way the ABS,
indeed all professionals, perform would probably have surprised Bach, in that
everything is given its just deserts, everything is sung well; but I think there is

a calculation in Bach's writing that some things aren't going to be performed

as well as others. In fact, a point my colleague Richard Taruskin made,^' which
I think is a very strong one, is that Bach was quite aware that a lot of his
church music was hard to sing and play. / would say that that brings out a lot

of the light and shade of the music very well; Taruskin says that it brings out

a particular spiritual point, that humankind is flawed. Either way, I think it's

quite a good argument.

20. Unlike other Protestant reformers, Luther saw music not as undermining Scripture but
as "a bearer of the Word of God" —thus the rhetorical role Butt speaks about. See the entry
"Martin Luther," by Robin Leaver and Ann Bond, in The New Grove Dictionary of Music,
vol. 11, pp.365-71.
21. Taruskin, "Facing Up, Finally, to Bach's Dark Vision," in The New York Times, 27
January 1991, Sunday Arts and Leisure section, p. 1, reprinted in his Text and Act (New York:
Oxford University Press, 1995), pp. 307-15.
288 THE BAROQUE

But singing it thatway presupposes that the music is going to be used in

the same hturgical way that Bach used it, which is not the case. The audiences
going to the ABS are expecting professional-standard music, and it's hard to
explain to an audience that you're going to sing in a way that's slightly under
par in order to make it good! On the other hand, if you get together a group
of boys or university students who are not expected to sing well, and get them
to sing well or at least convincingly, that's one way of approaching this partic-
ular issue.

Does that connect to your idea that Harnoncourt and Leonhardt, who used
boys' choirs, may came close to the original style?
Well, they certainly bring something out that no one else does. Often it's in-
tensely ugly and you feel like kicking in the loudspeaker when you hear it

particularly Harnoncourt; not so much Leonhardt. But you gain something in

that terrible frustration. I can't stand some of the things I hear, but they're still

very relevant.
We're in an age where we expect things to be finished and cellophane-
packed. Some of the English performers, like John Eliot Gardiner, I think, are
examples of this in Bach: extremely refined performances, which are always
very lively but relatively safe. Jeffrey Thomas and theABS go well beyond that
in putting the guts back into the music. It's a very dynamic group, and there is
a sense of unexpectedness about some of it. I've played on many of the records
and am always struck by the grittiness of it.

What's your view on Joshua Rifkin's hypothesis that Bach meant for his
choral works to be sung with one singer per part?^^
I haven't looked at as many of the original parts as Rifkin has, but I have
looked at more than most people have, and I find that virtually everything
Rifkin says is very convincing. There is only one part for each choral Hne in
most Bach choral works; applying Occam's Razor, you have to invent more hy-
potheses to support the idea of multiple users of parts than if you assume one
voice per part.^^
Of course, other people argue that there's a lot of outside contextual evi-

22. See Rifkin's "Bach's Chorus: A Preliminary Report," The Musical Times 123 (1982),
pp. 747-51, and his shorter version, "Bach's 'Choruses' —Less Than They Seem.'" High Fi-
delity,September 1982, pp. 42-44.
23. Rifkin's logic can be expressed in the following syllogism. The major premise is that
in allbut nine of Bach's surviving choral works only one part was written out for each of the
choral lines; this premise is not seriously disputed among Bach scholars. Rifkin's minor
premise, though, is controversial: that the parts are copied in such a way that each one must

have been meant for one singer alone to use. If this premise is accepted, the conclusion is ob-
vious: Bach used a solo quartet to sing those great cantata choruses.
As Butt points out, nobody has entirely refuted the assertion that these parts were copied
so as to have been, in many works, difficult to share. Robert Marshall observes that Bach's
parts are larger than most modern parts which — are routinely shared —and that his actual no-
 TRIPLE COUNTERPOINT 289

dence of multiple choral voices/"* I've done a lot of work in that area, proba-
bly as much as Rifkin, and my inquiries suggest that there was a great deal of

variety in Bach's era.-'' Some people wanted multiple singers, some didn't. In

most cases each singer did indeed get his own part to sing from.-" What is cer-
tain is that there was a division between the concertists, the solo singers who
were the number one in each of their parts land sang throughout the piecel,
and the ripiemsts, the people that we would call the "chorus." That is crucial

evidence about the way choruses are put together. But it allows the argument
to go either way; it doesn't determine whether Bach had one per part or more.
The wider context would allow you to do either, but his original parts, with a
few exceptions, almost certainly suggest he used single voices.

Many people point to the 1730 memorandum to the Leipzig Town Coun-
cil where Bach insisted that each chorus needed at least three singers on each
part, and preferably four.

tation is "around twice as large" (Marshall, "Bach's Chorus: A Reply to Rifkin," The Musi-

cal Times 124 [January 1983], pp. 19-22; quote, p. 22). But Rifkin answers that it's not the
size of the parts that leads him to conclude that they would be difficult for soloists and cho-

rus members to share, but the shortage of cues about when the chorus should enter and exit.
He argues that such cues might have been almost essential in some works, considering that
surviving Bach works that use extra voices ("ripienists") tend to use them not to double
throughout a movement but to join in or drop out at specific places; he adds that rehearsal
time was limited. Marshall argues that these "ripieno" entries could have been cued by Bach
from the keyboard; and others have argued that the boys may have memorized the parts rather
than sight-read them, which might obviate the problem. Still, Rifkin would respond that Bach

did have time to add articulation and dynamic marks to many cantata scores, and that he
surely would have found choral cues valuable in at least some works; so his not writing them
suggests that they were not needed. See Marshall, "Bach's Chorus: A Reply to Rifkin"; Rifkin's
"Bach's Chorus: A Response to Marshall," The Musical Times 124 (March 1983), pp.
161-62; Marshall's "Bach's Choruses Reconstituted," High Fidelity, October 1982, pp. 64-66,
and Rifkin, "Bach's 'Choruses': The Record Cleared," High Fidelity, December 1982, pp.
58-59. See also George Stauffer's review of Butt's Bach: Mass in B Minor, Journal of Must-
cologtcal Research 13 (1993), pp. 257-72, and Rifkin's reply in the same journal, 14 (1995),

pp. 223-34.
24. E.g., the distinguished scholar Hans-Joachim Schulze noted that musical forces at the
Dresden court, with which Bach was directly familiar, were larger than those in Leipzig (see
Schulze, "Johann Sebastian Bach's Orchestra: Some Unanswered Questions," Early Music 17
[February 1989], pp. 3-15). Rifkin replies, however, that we can't assume that what was done
in Dresden was what Bach wanted in Leipzig; he adds that the Dresdeners still used one writ-

ten part per singerand player, which supports his crucial minor premise (see Rifkin's "More
(and Less) on Bach's Orchestra," Performance Practice Review 4 [Spring 1991], pp. 5-13).
Schulze also quotes Mattheson's and Scheibe's denunciations of one-per-part singing; as Butt
points out, however, the fact that both Mattheson and Scheibe felt compelled to denounce the
one-per-part chorus so strongly indicates that "it was indeed still an option at that time" (Butt,
Musical Education, p. 208, n. 55), as other evidence also indicates —
see Andrew Parrott,

"Bach's Chorus: A 'Brief yet Highly Necessary' Reappraisal," Early Music 24 (November
1996), p. 557.
25. This research is summarized on pp. 106-13 of Butt's Musical Education.
26. See ibid., p. Ill, for an exception.
290 THE BAROQUE

Well, Bach didn't say that was the number of people needed to sing in his
cantata choruses (the translation in The Bach Reader is misleading there). ^^ For
one thing, much of the memorandum is about double-chorus motets, usually
not by Bach, which they sang every Sunday. More important, the topic of the
memorandum is how many performers he'd need to operate a liturgical music
establishment over an entire year. The memorandum gives the number of peo-
ple who should be part of the Kantorei for each church.

And, says Rifkin, Bach, being a practical musician, would ask for some re-
serves in case someone had the flu when they had to sing a motet he men- —
tions frequent illness in that passage, and also observes that singers are some-
times needed to play instruments.
Right. If you were to ask Herreweghe or Leonhardt how many people were
going to be playing in their groups this year, the numbers they'd give wouldn't

all be playing in every concert. Now, if Bach had written, "For the chorus parts
in my cantatas I need twelve singers," then the memorandum would be strong
evidence.

I must say,though, that however convincing Rifkin's case may be, most of
my favorite Bach performances use choirs; perhaps that relates to your com-
ment [in Chapter 9] that Bach depends less than most composers on the
specifics of realization.
Exactly. Groups such as the ABS succeed through their imaginative use of
historical perspectives, not through a slavish devotion to greater or lesser
"facts." That way, they can be creative within the framework of historical in-
struments and techniques of performance. Indeed, "fact" is almost a dirty word
in contemporary intellectual thought. Historical performance, like "history" it-

self, should be always new.

27. In The Bach Reader, ed. Hans T

David and Arthur Mendel (New York: Norton, 1966),
pp. 120-24, the crucial passage at issue is rendered thus: "it would be still better if one . . .

could have 4 singers on each part and thus could perform every chorus with 16 persons";
Rifkin renders this "... one could have 4 subjects for each voice and thus 16 persons in each
choir." (The original German reads: "Wiewohln es noch be(?er, wenn der Coetus [student
body] so beschaffen ware, dai? mann zu ieder Stimme 4 subjecta nehmen, und also ieden Chor
mit 16. Persohnen bestellen konte."
One argument is that while the parts may suggest that Bach had only one singer per part,
the memorandum shows that he wanted four per part; as Butt points out, however, the doc-
ument does not unambiguously show that. We can never know for certain what Bach would
have considered ideal, but we have no evidence that clearly demonstrates that he wanted "four
singers on a part," or even, in most cases, that he wanted more than one per part.
Andrew Parrott raises an additional reason for doubting that the number refers to the per-
formance size of each vocal group: of the musicians in each "choir," Bach normally had to
assign at least three and sometimes several more to play instruments: Parrott, "Bach's Cho-
rus: A 'Brief yet Highly Necessary' Reappraisal," P- 570.
 TRIPLE COUNTERPOINT 291

SELECTED DISCOGRAPHY
Jeffrey Thomas's most widely acclaimed ABS recording is undoubtedly his B
Minor Mass (Koch 7194, 2 CDs). Lindsay Kemp praises it for its "fresh ideas":
"the opening bars lovingly sung instead of solemnly belted out; stabbed-out
string dissonances driving home the nails of the Crucifixus; smoothly tender
choral singing emphasizing the link between the more classically contrapuntal
choruses and old-style polyphony. some superbly musical instrumental
. . .

playing — listen to the flute and muted strings of the 'Domine Deus.'"^**
Two ABS recordings of early Bach cantatas have used the one-per-part ap-
proach. The of them (Koch 7164) received poor reviews because its coun-
first

tertenor soloistwas on bad form; but the second of them (Koch 7235; Can-
tatas Nos. 4, 131, and 182) was "warmly recommended" by Nicholas
Anderson for the "expressive warmth" of its singing and for being "stylistically
apposite and emotionally satisfying."-'* Thomas's recording of Cantata No.
198, which uses a chorus, is the one I praised in my first question to him (Koch
7163; the CD also contains Cantatas 8 and 156). Another starting point for
sampling the ABS might be its first release (Koch 7138), with solo Bach can-
tatas sung by four American singers, among them Julianne Baird and Thomas
himself.
Thomas's singing can also be heard as the Evangelist in Kenneth Slowik's
recording of the St. John Passion (Smithsonian ND 0381, 2 CDs). Teri Noel
Towe calls this "unequivocally . . . [St. John Pas-
the best of period instrument
sions] overall."'" Thomas Luekens Thomas, "entrusted with consid-
says that
erable freedom in forwarding the dramatic line, Igauges] the tempo fluctuations
deftly while, over a crisp, clearly felt pulse, he captures the particular feeling
of individual dramatic moments"; for example, in No. 12 his is "recitative

singing at its finest, a verbally incisive, tellingly shaped line that tapers affect-

ingly."^>

Teldec recorded the complete Bach sacred cantatas under Leonhardt and
Harnoncourt between 1971 and 1989; it has reissued the series on ten mid-
price 6-CD sets as well as on 1- and 2-CD sets. James Oestreich writes of its

"sustained novelistic prose and epic poetry," and says the "series was made all

the richer, of course, by the infusion of two separate and strong directorial per-

sonalities, often exasperating but always deeply involving."^^ An admiring

Nicholas Kenyon exemplifies critical response when he says that "on the whole,
there is more coherence in Leonhardt's quiet un-self-advertising performances,

28. Kemp, "Quarterly Retrospect," Gramophone 71 (September 1993), p. ^4.

29. Gramophone 73 (March 1996), pp. 78-79.
30. Towe, in Choral Music on Record, ed. Alan Blyth (Cambridge University Press, 1991),
p. 22.
31. Lueken, recording review, Historical Performance 3 (Fall 1990), p. 89.
32. Oestreich, "Why the Bach Cantatas? Because They're There," The New York Times,
3 November 1996, Sunday Arts and Leisure section, pp. 33-34.
292 THE BAROQUE

and more exuberance and a greater hit-or-miss failure rate in Harnoncourt's oc-
casionally brilliant offerings. "^^ A good one-disc sampler of both conductors is

Teldec 42615, featuring Cantatas Nos. 124-127. For those wanting a sampler
of Leonhardt alone, Sony Classics (SK 68265) recorded him in 1995 leading
three great Bach cantatas — Nos. 27, 34, and 41 — all of which had been as-
signed to Harnoncourt in the Teldec series.
Leonhardt brings insight and feeling to the St. Matthew Passion (Deutsche
Harmonia Mundi 7848-2-RC, 3 CDs). Like the cantata recordings just men-
tioned, it is sung entirely by male voices, and the boy choristers and soprano
soloists in this recording are excellent. I have to admit, though, that in pro-
found arias like "Aus Liebe" my heart is more thoroughly rent when the soloist
is a grown woman with the expressive depths of, say, Herreweghe's Barbara
Schlick. Leonhardt's St. Matthew avoids one of the problems of much early-
music performance, that of breathlessly fast tempi. Such tempi were perhaps a
necessary corrective to the ponderous ones chosen for Bach by, say. Otto Klem-
perer; but according to Malcolm Boyd, there may be indications of a "general
retreat" from the excessive speed often adopted since^'* —a trend he thinks "will
be welcomed by many." According to Wye J. Allan brook, in the opening cho-
rus of the St. Matthew the meter suggests "the slow progress of Christ into
Jerusalem riding on an ass (Matthew 21), or . . . the agonizing limp of his walk
to Calvary."^^ These images suggest "slow, limping rhythms," she says, so the
waltz tempos and "loose, swinging rhythms" found in many early-music ren-
ditions are inappropriate.^* Leonhardt, I think, gets this movement right. Leon-
hardt's Bach recordings show the kind of articulation —the chopping up of
words, etc. —that makes choral conductors wince, or at least raise their eyes
heavenward, but that for some listeners has come to represent a true Bachian
style. It does enhance transparency.
The choral directors I know have nothing but praise for Philippe Her-
reweghe's choral finesse. The same seems to be true of critics. In Edward Roth-
stein's New York Times survey of St. Matthew Passion recordings, Her-
reweghe's (Harmonia Mundi 901155) is the clear favorite: in Herreweghe's
hands, Rothstein says, "The Passion becomes a genuinely weird and moving
story, avoiding the innocent graciousness that once characterized the early-
music movement while retaining the gravitas of pre-authentic Bach."^^ Simi-

33. Kenyon, "Bach's Choral Works: A Discographic Survey, Part 2," Opus (February
1986), p. 54.
34. Boyd, "J. S. Bach: Two Choral Masterpieces," Early Music 11 (August 1994), p. 525
which praises Thomas's Mass in B Minor.
35. Allanbrook, "The Sleep of Sin: A Note on an Aria from Bach's St. Matthew Passion,^''
in Essays in Honor of Robert Bart, ed. Cary Stickney (Annapolis, Md.: St. John's Press, 1993),

p. 19. Her argument is part of an analysis of how Bach tries to illustrate texts in this work

musically.
36. Allanbrook, personal communication, 1996.
37. Rothstein, "CDs in the Spirit."
 TRIPLE COUNTERPOINT 293

larly high praise has greeted Herreweghe's Easter Oratorio (on Harmonia
Mundi 901513, with Cantata 66), motets (to Graham Sadler, "much the most
successful of those 1 have heard";'** Harmonia Mundi 901231), and, in fact,
most of his Bach recordings. I have been unable to find a negative review of
Herreweghe's Bach; the complaints I've heard have been spoken, from some (as

it happens) British early-music performers who believe that Leonhardt and

Harnouncourt take the right approach to articulation and that Herreweghe is

too smooth.
It may not surprise those performers that Herreweghe's forays into nine-
teenth-century music have been especially successful (though Paul Hillier is not
alone in admiring Herreweghe's recordings of pre-Bach composers like Schiitz).

An example of Herreweghe's nineeenth-century recordings is the Beethoven

Missa Solemnis (Harmonia Mundi 901557), a work that John Deathridge calls

"perversely difficult to perform." Deathridge, in a BBC Radio Three survey of

all available recordings (28 January 1996), calls Herreweghe's "by far the most
interesting," and backs up the judgment in detail. Michael Tanner, in Classic

CD, comes to the same conclusion: Herreweghe "has come nearer to solving
the problem [of performing the work] than anyone else, even Karajan or
Toscanini."'''

FOR FURTHER READING

Bach's choral works have been at the center of the revolution in Bach scholar-
ship that has been in progress since the 1950s. This revolution is described in

Christoph Wolff's Bach: Essays on His Life and Work (Cambridge, Mass: Har-
vard University Press, 1991), esp. chap. 1. The book also contains some help-
ful chapters on the choral works, as does Robert Marshall's The Music of Jo-
hann Sebastian Bach (New York: Schirmer 1989), Part 2. Marshall's Part 4
includes an interesting essay on the Teldec cantata series, entitled "'Authentic'
Performance: Musical Text, Performing Forces, Performance Style" (pp.
229-39). John Butt's Cambridge Handbook Bach: Mass in B Minor (Cam-
bridge University Press, 1991) is an especially good guide to form and compo-
sition in Bach's choral works.
An important and original discussion of the one-per-part controversy, by a
performer who favors Rifkin's position, is Andrew Parrott, "Bach's Chorus: A
'Brief yet Highly Necessary' Reappraisal," Early Music 24 (November 1996),
pp. 551-80. Parrott reviews the evidence thoroughly but concisely, and con-
cludes, "Rifkin's thesis deserves to be regarded as beyond reasonable doubt.
The burden of proof lies squarely with those who hold to the notion of an
'ideal' 12- (or 16-) strong Bach choir."

38. Sadler, Early Music 15 (May 1987), p. 303.

39. Tanner, "Herreweghe Triumphs in Beethoven," Classic CD, December 1995, p. 86.
 O?^ JV^ („^5^

CLASSICAL AND ROMANTIC

Few artists have ever described their own work as "classic"; it is a term re-

served for the dead. Mozart, Haydn, and Beethoven never dreamed that they
were writing in the "Classical style." Nobody else seems to have thought so ei-

ther, until the 1830s — much later in Beethoven's case. This problem here is

more than semantic' Robert Levin, Malcolm Bilson, and Roger Norrington,
though they don't try to ways in which we
reform our terminology, all suggest
enfeeble these composers by crowningthem with that particular laurel, and (to
use Norrington's image) mounting them on smooth marble pedestals.
The term "Romantic" also has problems, but at least it was in use when
the music we apply it to was written. Where it may confuse us is in the realm
of performance practice. Composers may not call themselves "classical," but,
as we'll see, some early-music performers describe their playing style that way.
Yet we'll also see that the assumption that "Classical" playing styles were strict

is, in various ways, questionable. It is also characteristic — but of us, and per-

haps not of history.

Nonetheless, in contrast to the pre-Baroque interviews, the following ones
rarely raise the idea that it's "impossible" to recover a historical performing
style. They do, however, spend a good deal of time on the historicist/main-
stream "turf wars." That's because these wars are fought mainly over com-
posers who were already part of the active concert repertory when the early-
music revival began —or so I wrote in the introduction, and you will find

evidence for that in these interviews. I also said that the turf wars seem to be

1. Wye J. Allanbrook sees the term as an essentially Romantic notion that continues to
mislead commentators on Haydn and Mozart. See the fifth lecture in her The Secular Com-
media (Berkeley: University of California Press, forthcommg).

295
296 CLASSICAL AND ROMANTIC

subsiding; you will find less evidence for that in Robert Levin's distaste for the
unhealthy alliance of competitions, record companies, and agents, in Malcolm
Bilson's reservations about modern style, or in John Eliot Gardiner's concerns
about overstandardized orchestral playing.
However valid these critiques may be, they raise the issue of differing tastes.
Monica Huggett told me that as a student she felt straitjacketed by the big
sound, lush tone, and constant vibrato of the violin school of Ivan Galamian
and Dorothy DeLayr but Itzhak Perlman, perhaps the most distinguished prod-
uct of that school, calls period string players' non-vibrato phrasing "sterile"
and "totally lacking in warmth." Robert Levin argues below that Mozart ben-
efits from being played in a historical style, but Perlman says, "I'm certain
Haydn and Mozart would have adored our modern approach to phrasing and
vibrato."^
Of course, we'll never know what they would have adored; and even if we
did, as I wrote in the introduction, some believe that we shouldn't care about
that so much as about what we adore. But Perlman does not make that argu-
ment. Like most mainstream musicians, he is just as concerned with the com-
poser's intention as, say, Bilson is (a fact that Bilson discusses in his interview).
Where the two differ is in their understanding of Mozart's intentions. Perlman
implies that those intentions transcend the accident of how people played the
violin at the time — but Bilson and Levin argue below that some details of the

era's playing style are crucial to Mozart's intentions.

Whichever side you take, it if Perlman prefers more
should be noted that
vibrato, few historical performers nowadays would insist that he suppress this
taste when, say, playing Brahms merely because Brahms might not have shared

it."* And of course, we can't know that Brahms wouldn't have adored the best

of Perlman's Brahms playing any more than we can know the opposite.
We do know, however, that the Galamian approach stokes Perlman's artis-
tic fires, while the historical-performance approach stokes those of Huggett and

Anner Bylsma (who can't bear too much vibrato in Brahms). The following dis-
cussions tell us a great deal about the music, and about musical life today and
in the past —
after all, it's the abundance of historical evidence that quiets the

objection that historical fidelity is not possible even in part — but in the end,

what matters most about this evidence may be that it serves my interviewees
as kindling.

2. Bernard D. Sherman, "Monica Huggett," Strings 10 (March/April 1996), pp. 54-61.

3. Ross Duffin, "Performance Practice: Que me veux-tu?" Early Music America 1 (1995),
p. 35.
4. Spare-vibrato playing sounded better to the ears of Brahms's close associate Joseph
Joachim, who said, among "A viohnist whose taste is refined and healthy
other statements,
will always recognize the steady tone as the rule" (in his Violinschule [with Andreas Moser;
Berlin, 1905], II, p. 94). Brahms might well have agreed.
 16
Restoring Ingredients
—^eOe^
Malcolm Bilson on

the Fortepiano

"[T]he time will come, 1 believe," wrote Andrew Porter in 1976, "when audi-
ences —and pianists —having once discovered the tone colors and clarity and
alertness of wooden-framed pianos with thin strings and buckskin-covered

hammers, will want to hear more of them.'" That time has yet to come. Even
in the early-music world, the fortepiano has fewer friends than we might ex-
pect. The harpsichordist Christophe Rousset speaks for many when he calls it
"an imperfect instrument, [which had] yet to evolve."^ Part of the problem is

that in a large modern concert hall the instrument becomes little more than a

tinkling symbol; but even in the small rooms it was designed for, its limited dy-
namic range and short tone life can seem desiccated to many who grew up with
a Steinway. Malcolm Bilson, who mastered the Steinway long ago, argues
below for the advantages of that short tone life.

In making a case for the older instrument, Bilson addresses a more basic

issue, still a touchy one for fortepianists: why might historical instruments mat-
ter at all? He also discusses changes in playing style, involving approaches to
articulation and phrasing that are diametrically opposed to what has become
the standard modern practice (though how recently it became "standard" is not
always recognized, as he also points out).
Even for the generally unconverted, Bilson offers a specific corrective. It's

often said not only that Beethoven was dissatisfied with the pianos of his day

1. Porter, "Pianists and Pianos," from the New Yorker, 23 February 1976, reprinted in

Music of Three Seasons: 1974-1977 (New York: Farrar, Straus and Giroux, 1978), pp.
299-300.
2. Stephen Pettitt, "Virtuosity with Heart," Gramophone 71 (September 1993), p. 16.

297
298 CLASSICAL AND ROMANTIC

(true, though perhaps exaggerated), but that in his mind he sometimes heard a
piano image "not unhke that of the modern Steinway concert grand. "^ Yet as
WiUiam S. Newman has shown, and as Bilson discusses below, Beethoven's
piano ideal differed profoundly from the modern one.** Beethoven would prob-

ably have been at least as unhappy with our pianos as he was with those of
his day. This is not to say we should abandon the concert grand (I'm not throw-

ing away my CDs of Schnabel, Goode, or Brendel, nor would Bilson say I
should^); but we shouldn't dismiss Beethoven on the fortepiano a priori.
In fact, my conversation with Bilson preceded a remarkable Beethoven
recital. In it, he proved to be so communicative a Beethoven player that he got
a large portion of the audience to laugh out loud at some of the musical jokes
in the Op. 33 Bagatelles, a response I had never encountered before at a

Beethoven concert. I interviewed Bilson at the site of the sponsor of the con-
cert, the Ira F. Brilliant Center for Beethoven Studies in San Jose; we were sur-

rounded by shelves packed with books, documents, manuscripts, and rare

Beethoveniana. The Poletti fortepiano used in his recital was in the room, and
at regular intervals Bilson walked over to it to make musical points.

My questions made Alfred Brendel — a pianist and thinker whom Bilson ad-
mires greatly —serve as Bilson's antagonist. A few months after I met with Bil-

son, I heard Brendel, in a radio interview with the pianist Sarah Cahill, say that
he now some early Beethoven sonatas are served better by
thinks that parts of
the fortepiano than by the modern grand. He said that the opening of Op. 7
is hard to play at a suitably fast tempo on a modern Steinway, but that the

tempo is easier to achieve on the fortepiano, with its incisive sound and shal-
low action. This is not to say that Brendel is a convert. He holds that early
piano music is rarely composed to exploit the sound of the fortepiano per se,

but uses it to suggest other instruments; he therefore thinks it unnecessary to

bother reviving the vintage piano. Bilson brought up this argument at the be-

ginning of our conversation.

Alfred Brendel says that when you play Beethoven on a modern piano you're
playing a transcription.^ Yet he also argues that the modern piano does more
justice to Beethoven.

3. Edward Greenfield, Robert Layton, and Ivan Marsh, The New Penguin Guide to Com-
pact Discs and Cassettes (London: Penguin, 1988), p. ix.

4. Newman demonstrated in 1971 that throughout Beethoven's life his piano ideal re-

mained the light, responsive Viennese fortepiano. This research, which has never been seri-

ously challenged, is summarized in chap. 3 of Newman's Beethoven on Beethoven (New York:

Schirmer 1988).
5. Though his older writings suggest a position less tolerant than the views presented in
this interview; see his "The Viennese Fortepiano of the Late 18th Century," Early Music 8

(1980), pp. 158-62, esp. pp. 161-62.

6. Brendel, Musical Thoughts and Afterthoughts (New York: Farrar, Straus and Giroux,
1976), p. 16.
 RESTORING INGREDIENTS 299

May I rather refer to Brendel's notes on the Mozart concertos," which I

think are really quite wonderful? In fact, I agree with virtually everything Bren-
del says, until he comes to the question of the instrument. He dismisses
Mozart's piano, and cites the Sonata in A Minor, K. 310, as an example. His
point, if I understand it correctly, is that pianos per se are not really so im-
portant anyway. The first movement, says Brendel, is symphonic (I'm not sure
I really agree with that, but let's buy it for the sake of argument); the second
movement imitates a voice, a cantabile, singing line; and the third movement
is clearly a wind divertimento. But the minute I accept Brendel's categories, it

seems clear to me that Mozart's piano does each of them better than a mod-
ern piano!

Howf
Well, if the first movement is symphonic, it does not seem natural to put
the most important part of the music into the background, which is what one

is forced to do on a modern piano. One cannot play those opening left-hand

chords full out on a modern piano, because it is simply too bombastic; and it
is virtually impossible to give them rhythmic pulse (strong, weak, strong,
weak —as required by the tutors of the eighteenth century), so what one
all late

hears —from everybody— melody with chords the background.

is a in

The second movement, says Brendel, is a vocal line, and I agree completely.
But in Mozart's 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 — but do
you know any performance or recording that slurs as Mozart asks? Look at

the beginning of the movement:**

Andante cantabile con espressione

The slur and the fp between the A and the F in the right hand reinforce the
"sighing" aspect of the little figure; and of course, the F must be released,
not carried over legato to the C — Mozart knows very well how to write three-
note slurs when he wants them. Have you ever heard anyone play this the
way it is so clearly written? On the contrary, one hears a "long, singing
line" —a much The modern piano is very good at achieving
later concept.

such a line, but very poor at achieving what I believe Mozart associated with
singing, namely declamation and inflection of the syllables of the text. (In
music without text, the tutors tell us, one should always inflect as if the text
were there; otherwise it is not natural. And Leopold Mozart tells us that a

7. Brendel, Music Sounded Out (New York: Farrar, Straus and Giroux, 1991), pp. 3-11.
8. Mozart, Sonata in A Minor, K. 310, second movement.
300 CLASSICAL AND ROMANTIC

group of notes under a slur should be played with a slight diminuendo^ —

point with expressive significance, I've come to feel.) The modern piano, with
its long tone life, has great difficulty playing the F sufficiently softer than the
A; it becomes inaudible due to the long singing of the A —and releasing the
F will invariably sound like a hiccup, because again the tone of the modern
piano wants to continue.

So the more rapid decay of sound on the fortepiano lets you convey that fp
on the top A, whereas the slow decay on the modern one prevents that?
Right. Then regarding the third movement, the wind band, I completely
agree with Brendel, but once again you have the difficulty that one cannot play
all the voices equally, due to the cross-stringing of the modern grand.

The bass strings being crossed under the other strings?

Yes. I'm beginning to think more and more that while cross-stringing has
some virtues, it has many more drawbacks. Do you know why they crossed

strings on the modern grand?

To save space?
Curiously, even someone from Steinway gave that answer; it's a widespread
belief. But after about 1800 the standard length for grands was eight feet; Stein-

way's first grands in the 1850s were straight-strung, and they were eight feet

long. When Steinway crossed the strings, they went to nine feet to get back the
extra length they had lost.

So length isn't the reason. What is?

To pull the bridges in toward the center of the soundboard, where the board
is most resonant. Think of a violin with the strings running right across the

center; on modern pianos, the soundboard is "crowned," as the violin belly is.

In addition to the cross-stringing, the grain of the wood on all modern pianos
runs diagonally from the keyboard treble to the tail bass. What that achieves

is to bring every tone to the center of the soundboard, and it gives the modern
piano an enormously concentrated tone, a wonderful virtue. But there are
drawbacks as well, compared to the old straight-strung, straight-grained con-
struction. The following passage, as I understand it, is unplayable on a mod-
ern piano:'"

9. first note of a group of "two, three, four, and even

Leopold Mozart wrote that the
more must be somewhat more strongly stressed, but the remainder slurred on
[slurred notes]
to it quite smoothly and more and more quietly." I quote it from the discussion of the slur in
Sandra Rosenblum's Performance Practices in Classic Piano Music (Bloomington: Indiana Uni-
versity Press, 1988), chap. 5, p. 159; see especially the subsection "Do All Slurs Indicate At-
tack and Release?" on pp. 172-83.
10. Mozart, Sonata in Bb, K. 333, opening.
 RESTORING INGREDIENTS 301

Allegro

There are two voices here, and the left-hand voice is, if anything, even more
interesting than the right; he could have written a simple Alberti bass, but he
didn't. Now, if we had a modern piano here, I simply could not give equal
weight to both voices.

Why not?
Because the tone is so concentrated that all the sounds slap at each other.
One can do it, but it sounds very bad: only a very poor pianist would do that.
And so, what you'll hear, invariably (listen to any recording), is a beautiful
singing right hand, with the left hand in the background. Most listeners are

used to hearing it that way, but to me it sounds like a vioHn-cello duo in which
the cello is put at the back of the stage!
Actually, the "slapping at each other," as I call it, this clashing of the sounds
together, is something jazz pianists use very much [plays a jazz progression].
There we are used to it; indeed, it is a major component of the jazz piano
sound. But we generally don't like it in "serious" music.

And so you lose the independence of voices. [Plays the beginning of the fi-

nale of K. 310.] There's a lot going on in the middle voices, but listen to any
recording — you'll hear the top voice — just as in K. 333. But on this fortepiano,
you can give each voice its due — all the strings are running parallel and the
grain of the wood is running parallel, so in a sense each note is independent,
whereas on a modern piano there's a conglomeration of sound.

How about Brendel's belief that the modern piano serves Beethoven better?
I think the modern piano serves Beethoven far worse than it does most of
the other Classical composers, because Beethoven was extremely interested in
the piano and really wrote quite differently for the piano than for strings or
for orchestra.

For example?
Well, one thing I consider enormously important in Beethoven is the
sforzandos. The modern piano doesn't have a sforzando, it only has loud notes.
Because, again, the tone life is so long, it doesn't have the punch in the shoul-

der that Beethoven would know.

Brendel discusses how difficult it is to play Beethoven's sforzando on a mod-

302 CLASSICAL AND ROMANTIC

em piano, and you suggest it's partly the instrument. You've also written^
about the way in which Beethoven's bass figuration depends on the fortepiano's
transparent bass.
Yes, left-hand transparency is an essential part of Beethoven's piano writ-
ing, and it is almost impossible on a modern grand, once again because of
cross-stringing.

To move to the next generation in Vienna, Brendel argues that most of

Schubert's piano music isn't conceived with the piano in mind, with the ex-
ception of the Impromptus. Do you think Schubert benefits from period pi-

anos?
I personally think that Schubert's piano music is very much tied to these in-

struments.

"^hy?
Well, one thing that is very important in Schubert but that one doesn't hear
much in modern performances is rhythmic inflection. Now, you know the \^an-
derer Fantasy, don't you? Sing me the opening bar.

[Interviewer sings: TUM! ta ta TUM! ta ta TUM! diddle-diddle-diddle turn

ta ta TUM TUM!]
That's terrific! Ask anyone who knows this piece, and they will sing just
what you sang, with strong accents on the downbeats (which are also marked
by Schubert in the score). Now, go to any recording, and what you will hear
is quite even beats. If you try to produce on a modern piano the inflection
you just sang, the after-beats sound weak and very peculiar. [Plays opening
of Wanderer with strong accents on the downbeats.] Anyone sitting down
at this fortepiano would, I believe, play it that way; but not on a modern
piano. All this is very much at the heart of what I think is important in
Schubert.
That applies not just to fortissimos, but also to pianos and pianissimos. Lis-
ten to any recording of the opening of the Bb Impromptu, Op. 142, No. 3.

Everyone plays the two eighth notes of the second beat evenly, or even a little

louder than the downbeat. Yet Schubert's articulation markings clearly indicate
the second beat as weaker than the first, an approach reinforced by all the tu-
tors of the time.
What is especially interesting here, I think, is that whereas a weaker second
beat is not natural to the tone of the modern piano, the four recordings I have
heard on fortepianos also crescendo across that second beat. "Pianos," says the
wag, "don't make music —people do," and there seems to be a lot of truth in

11. Bilson, "Keyboards," in Performance Practice: Music after 1600, ed. H. M. Brown and
S. Sadie (London: Macmillan, 1989, and New York: Norton, 1990), p. 232.
 RESTORING INGREDIENTS 303

that statement. The concept of the long, continuous Hne was drilled into all of

us in every conservatory in the world. We have to learn a new language; sym-

pathetic instruments won't do it all for you, but they can help.

The wag might ask another question: do these matters make a great per-
formance f
Of course they don't. A great performance is obviously a function of how
deeply one understands the music. But if you read Tiirk — Daniel Gottlob Tiirk's
Klavierschule is the most important keyboard tutor of the late eighteenth cen-
tury — his chapter on Performance is divided into two parts: Execution and Ex-
pression. I think that most musicians would agree with what Tiirk says about
expression, namely that what counts is depth of feeling, and how well one plays
one's instrument, and how well one can transmit one's inner understanding of
the music to the audience. But when one comes to execution (which embraces
just the matters we've been discussing, such as inflection or stress and release),
not only does one not hear these aspects in modern performance, but actually
very often one hears just the opposite.

Let's explore that, first regarding meter and metrical articulation. Schnabel
said that if he were rich enough he would have all his music printed without
bar lines. And Karajan said it should sound as if there were no bar lines at all.

On the other hand, the consensus seems to be that in Beethoven, Haydn, and
Mozart bar lines are very important.
Very important, and heavier down beats, which I was told by at least one
Schnabel student —possibly two—to always avoid. When I talk about this, one
of the pieces I like to demonstrate with is the first movement of K. 332, which
has slurs within each measure: '-

Allegro

Mozart has clearly set off each measure from the others with its own slur,
and that gives what I perceive as the lilt. When I was demonstrating this re-
cently, a Schnabel student sprang up onstage and said, "What you're saying
is completely wrong. Schnabel always taught us to accentuate the third beat,
in order to keep the long line going." There is a CD out now of Schnabel
playing K. 332, and you can hear that very clearly. Now, Schnabel was in a

way a wonderful Mozart player, but his basic aesthetic —thus his execution
is diametrically opposed to what any eighteenth-century musician would think
was normal.

12. Mozart Sonata in F, K. 332, first movement.

304 CLASSICAL AND ROMANTIC

Some people say that Schnabel was right — one scholar wrote that when
Mozart wanted continuous legato he would slur "each bar separately, but in-
tended no audible break between them."^^
Did Mozart ever want continuous legato in the sense that it is practiced
—
nowadays that is, without inflection? I believe that Mozart's slurs mean just

what we discussed before, a gradual tapering from first note to last (subject,

of course, to other rules of stress, such as high notes, dissonances, etc.).

"Diminuendo," by the way, might seem too strong what happens. a word for

Normally we use it to refer to going from forte to piano; that's not what I
mean. I mean diminuendo within piano, so that you end up weaker than you
started, albeit in the same overall dynamic. In any event, I think the reason
many modern-day musicians find it difficult to let go of the concept of con-
tinuous legato is that most, in order to "keep the tone going," do not make
this diminuendo under the slur, as in K. 332 or the Schubert Bt) Impromptu.
If one is going ahead full steam with a good, rich tone, an articulation break
will always seem intrusive and unnatural. If, however, the last note is softer,

a break is quite natural.

Genuine long legato slurs did, of course, exist at the time; the composers of
the London school used them, and they are found frequently in Beethoven.
They perhaps find their apogee in Chopin, who often wrote a single slur over
a page or more and is reported to have phrased that way. Mozart would doubt-
less have found his playing very original!

Let me reformulate the wag's question. Your point that Schnabel's continu-
ous legato in that phrase differs radically from Mozart's lilting slurs convinces
me—I'm thinking of Czerny saying that Beethoven didn't like Mozart's playing
"^'^
because it was "choppy, with no legato. But one could still argue that both

Robin Stowell, "Leopold Mozart Revisited," in Perspectives on Mozart Performance,

13.
ed. R. L. Todd and R Williams (Cambridge University Press, 1991), p. 138. Paul and Eva
Badura-Skoda say the same in their pioneering Interpreting Mozart on the Keyboard, trans.
Leo Black (New York: St. Martin's, 1962), pp. 54-55; they distinguish between "articulating
slurs" (the kind Bilson speaks about) and "legato slurs" (the kind Stowell mentions). How-
ever, in The Pianist as Orator (Ithaca, N. Y.: Cornell University Press, 1992), p. 110, George
Barth challenges Eva Badura-Skoda on this issue and shows that her long-line interpretation
of one-bar slurs is based on a misreading of Turk's chap. 6. Earth's arguments would apply
also to a passage in Leopold Mozart (chapter 1, section 3, paragraph 17, footnote) that has
been used to support the "legato slur" concept. By the mid-nineteenth century, however, the
endings of slurs may no longer have indicated breaks in articulation: see Nicholas Temperley,
"Berlioz and the Slur," Music and Letters 50 (1969), pp. 388-92. But Temperley, too, argues
that they did not indicate breaks even for Classical composers: see his Cambridge Handbook
Haydn: The Creation (Cambridge University Press, 1991), p. 116.
I side with Bilson on this issue, but I am at fault for not pursuing another question:

whether all of Mozart's slur marking should be understood as intentional, or whether some
might result from habit or convention.
14. Quoted in Rosenblum, Performance Practice, p. 23; see also her discussion on pp.
149-52 and her chap. 5. Czerny's report is widely but not universally assumed to be true.
 RESTORING INGREDIENTS 305

Schnabel's and Mozart's executions are valid and effective. In other words, the
wag might say that Mozart's way of playing it was not inherently the best or

ideal way. How would you answer that?

I would be the first to say that it's not necessary to be concerned with
Mozart's intentions; and it you're not, that's fine. If you are, though, I think
these slurs and other markings are perhaps even more important than the
notes — in some passages the exact notes weren't crucial to Mozart. Now,
Mozan wrote great notes, and you can play them in many different ways and
produce great music. I've heard wonderful Mozart from Rachmaninoff and
Schnabel (and Schnabel was very concerned with Mozart's intentions but un-
derstood them within a later aesthetic). I certainly don't claim that I or other
fortepianists play Mozart better than such artists. And I'm not a purist. I can,
if I like, choose to ignore a crescendo or slur mark sometimes; that's my right
as a performer. But my concern is that I think very few musicians today know
what these markings signify. Most haven't been taught how to read these
scores, which contain so much information. There are six Urtext editions out,
but I have heard no modern pianist try to make these slurs in K. i2>l audible.
This reflects neither their lack of interest in Mozart's intentions (pace the wag)
nor their instruments (which can sound these slurs) but the enormous change
in musical aesthetics across the nineteenth century.

To understand that change, let me begin by asking about influences on the

older aesthetic.
I think it has a lot to do with the prestige that poetry had at the time. In
every eighteenth-century source, music is compared to language, and there are
no languages I know of that don't have inflection. If you want to learn to speak
any language like a native, you have to learn not just the accent but also the
inflection. I think that's the most difficult thing when English actors try to do
American or American actors try to do English; the "can't" and "cahn't" part
is simple, but they inflect very differently from the way we do.

We say gar-age and they say gar-age.

Right; and if music is like speech, then inflection is very important in it, too,
and the rules of inflection are very clear in all these sources. There are things
the sources don't agree about, but on these inflection rules they all agree.

Stronger downbeats, stronger accents on dissonances — hold these notes longer

and those notes shorter. Consider how we say "Mother" —the second syllable
must be shorter and weaker. If you say "MO-THER" [giving the word two
equal stresses] you sound like a foreigner. But then of course this is the way
most modern singers sing. They sing (sings two long notes in operatic style]

"MO-THER" and they make it to the Metropolitan Opera!

This change, away from an aesthetic of music being like speech to the long.
306 CLASSICAL AND ROMANTIC

sustained line, is often attributed by early-music players to Wagner, above all,

with his "endless melody."

That's the generally accepted wisdom. But George Barth, in his fascinating
book The Pianist as Orator, accuses Beethoven's student Czerny of already
changing many of Beethoven's short slurs to longer ones/' For example, Cz-
erny puts slurs over the opening bars of the Pathetique Sonata, changing what
would clearly have been perceived as detached chords (the so-called French
Overture style) to the legato style we are accustomed to nowadays.

What aesthetic influences made Czerny do that?

Well, for one thing, the later the pianos get, the less susceptible they are to
doing this kind of inflection. In 1991 I played the Schumann Concerto with
John Eliot Gardiner on a replica of an 1830 Graf piano. A year later I had the
chance to do it again with the Cayuga Chamber Orchestra, in Ithaca, New
York, where I live, this time on an 1855 Bosendorfer, still a straight-strung Vi-
ennese piano. But the 1855 piano, with its much larger hammers, was a quite
a bit heavier than the Graf I was used to, and it was not available for me to

practice on, so I practiced on the Steinway in my studio at Cornell. It was cu-

rious how much less sensitive the 1855 Viennese piano was than the Graf, but
how far less sensitive the Steinway was than either of them!
Now, I realize that most of your readers will just about hit the ceiling —an
1830 piano is more sensitive than a Steinway? The Steinway (or Bosendorfer

or Bechstein) has such an enormous range of color; could an old piano com-
pete with that? But with all the color possibilities, fine nuances of inflections
are lost —the best modern pianists do virtually everything with color, and little

with articulation or inflection, which have come to be so important in my way

of hearing all this music.
For example, in the first statement of the Schumann's opening melody by
the piano, there's a sforzando on the high E that nobody plays. I realized, when
practicing on the Steinway, that I couldn't play it either, because there was no
way to get it to decay fast enough to make the proper inflection between it and
the following resolution on D (which should be much softer). It's quite easy for
anyone to sing that very expressively, by the way, but our standard modern pi-

anos won't deliver that kind of expression.

You know, both Brendel and Charles Rosen criticize the older pianos for
life; but when you think about it, the decay of tone (as it is
their shorter tone

so sweetly referred to in our language) is the only musical characteristic the

piano can call its own. The best composers are very sensitive to that, and know
how to turn it into a virtue (as we saw when we were talking about Beethoven's
sforzandi a while back).

So are you saying that part of the change in the aesthetic, to a more long-

15. Barth, The Pianist as Orator, pp. 94-95.

 RESTORING INGREDIENTS 307

line, less inflected style, came about as a result of pianos getting louder and
heavier — rather than as a cause?
It's partially true, although I cannot imagine that the "long-line" approach
is purely due to the larger tone of the later pianos;'*" that would surely be
imputing to pianos a bigger role than they deserve, although the piano had
certainly become, by the mid-nineteenth century, the main instrument for Eu-

ropean music. It must be remembered that English pianos of the late eigh-

teenth century already had a very long tone life, and, as I said, one sees very
long slurs in the works of the important London composers (Clementi, Cramer,
Dussek, et al.). This was the wave of the future; Vienna was conservative by
comparison.

So the modern piano has London and not Vienna?

its real roots in

That is certainly true to a great extent. The difference between the Vien-
nese-style pianos and the English-style pianos, and the two schools of playing
associated with them, continues until very late in the nineteenth century. The
English school is taken up in Paris and is well represented by the piano music
of Chopin and Liszt. The Viennese school goes on through Schubert and into
Schumann and, indeed, right up through Brahms to the end of his life. Com-
pare Schubert's Moment Musical No. 3 in F Minor to the B Minor Capriccio
of Brahms, Op. 76, No. 2. It's virtually the same kind of piece. Only, Brahms
was afraid the Viennese tradition of playing was getting lost, and so he indi-
cated small accents on the after-beats in the left hand; that wasn't necessary for
Schubert, but they are surely the same kind of inflection, where the "pah" of
the "oom-pah" is given a slight lift. And in the first piece of Op. 119, Brahms
still writes one slur over each bar —the same notation Mozart used in K. 332!
We know Brahms complained late in his life that players were no longer
that
observing the two-note slur, which he still seems to have considered a "sigh."'^
Does anyone play Op. 119, No. 1 that way, lifting at the end of each measure?

16. The enormous popularity of Italian bel canto opera at the time was perhaps another
factor; for a discussion of its influence on Chopin, see Charles Rosen, The Romantic Gener-
ation (Cambridge, Mass.: Harvard University Press, 1995), pp. 334-51. More significant may
have been the "newly systematic" use of the four-bar phrase as the basic time unit of music
(ibid., pp. 258-78). As Rosen says, this "gave a larger sense of motion to long works" (p.

278); it also made it important "to avoid giving a similar emphatic accent on the first bar of
every group, as one were accenting a downbeat" (p. 267).
if

17. to Joachim, in May 1879, that a slur over a pair of eighth notes indi-
Brahms wrote
cates that the second note is shortened, losing some of its value. Moreover, "To apply this to
larger note groups would mean an execution marked by liberty and delicacy, which never-
theless is appropriate most of the time." Quoted by Max Rudolf, The Grammar of Conduct-
mg, 3rd ed. (New York: Schirmer, 1994), p. 408, who notes that "this kind of phrasing calls
to mind eighteenth-century performance practices." The English pianist Florence May, after
studying with Brahms in 1871, noted that "he made very much of the well-known effect of
two notes slurred together, whether in a loud or soft tone, and I know from his insistence to
me on this point that the mark has a special significance in his music." May, The Life of Jo-
hannes Brahms (London: Arnold, 1905), p. 18.
308 CLASSICAL AND ROMANTIC

[Plays it.] At the slow speed Brahms asks for, it is quite easily realizable, even
on the richest modern piano.
It is illuminating to see that Chopin and Liszt (although they were by no

means the same) were doing in Paris something different from what Schumann
and Brahms (also not the same, of course) were doing in Germany, and that a
great deal of this has to do with piano aesthetics, so different in those two
places. Mendelssohn seems to be somewhere in the middle; both types of pi-
anos were important for him and play a role in his keyboard music.

That covers the differences among instruments. But to return to the rela-

tionship between aesthetics and understanding what the markings in the scores
were supposed to mean, I'd like to ask you about some details of "execution,"
beginning with articulation markings. For example, the "portato," the mark-
ing .^TT?.

I was taught that it was called the "portamento," which I no longer think it

should be called. I was also taught that it meant you should play the notes half
length, whereas I now know that half length is the normal length for notes that

are not otherwise marked. Instead, it seems that according to Tiirk, Leopold
Mozart, and others, notes played portato should be held as long as possible.
Suppose you have four notes descending. If these notes are not marked, then
they're to be played detached, at half their length. Though Leopold Mozart and
Tiirk qualify that: both say that the length of the note depends on the context
of the music. In a scherzo, unmarked notes would probably be shorter than
half length; in an adagio of a heavier nature, they would be longer, though still
detached from one another. Now, if you put a slur over those notes, they're not
just connected but, as we've seen, have an inherent diminuendo — it's as if you
come to the end of a bow-stroke, then start afresh with a new one."* However,
if you have portato, the slur with the dots under it, it's a continuous sign — it's

as if you either bow on each note or tongue on each note.

Again, you have to decide on what the context of the piece is. This is some-
thing I always try to get students to understand. The first thing you have to
understand is that whatever your teachers tell you is probably wrong —even if

it's me! We have what Beethoven gave us, but when Beethoven writes a por-
tato marking or a sforzando, do we really know what to do? Does a sforzando
mean to play longer? mean to play louder, and if so, how much? All
Does it

of these are really open questions. And it seems to me that any young person
who's starting out should doubt everything and try to make his own decisions

18. Bilson points out that theGerman word Bogen means both "slur" and "bowing." Re-
garding Monica Huggett pointed out to me that on Mozart's violin just changing the
this,

bow's direction creates an audible articulation; this, she said, reflects "both the old bows and
the gut strings." By contrast, "with a modern setup you have to make a kind of hard ar-
. . .

ticulation for it to register." Bernard D. Sherman, "Monica Huggett," Strings 10 (March/April

1996), pp. 54-61; quote, p. 58.
 RESTORING INGREDIENTS 309

about such things. And investigating instruments and how they react to such
markings as sforzando is very much part of this questioning.

So we decide just what contemporaries understood by "execution" based,

ultimately, on our own educated judgment?
Of course. You know, we do not have recordings of Beethoven or Schumann
or Chopin or Liszt or Bach or Mozart playing, but we do have recordings of
Prokofiev, Bartok, Rachmaninoff, and Elgar. Get a score of one of these com-
posers, and sit down and hsten to one of those recordings. It can be quite illu-
minating, but not as is often done, through asking, "How does Bartok play this

piece?" Rather, try it from the point of view of listening to what Bartok plays,

and then asking How did he write this down on the page? In other words, do
it backwards. And this might help understand what Beethoven might have heard
that would make him mark the page as he did. It gives you a fresh perspective.

one element of these recordings is often rubato. Let's talk about ru-
Well,
bato, as an example of execution. What about the contrametric rubato that
Mozart is said to have used, where the left hand kept steady time while the
right hand moved more freely?

I very much believe in the left hand not knowing what the right hand does,

and I try to do this in Mozart. In one of the Mozart concertos I recorded with
John Eliot Gardiner, in the middle section of the slow movement I made a great

was playing a beautiful, lilting accompani-

effort to realize this: the orchestra

ment, and I was wandering around and weaving in and out. But in the finished
product you don't hear much of it. I think that what happened was that when
they were editing the tapes they threw all those takes out because they weren't
together!

Will you tell me which concerto it is?

I'm sorry, I'd rather not. But, you know, Chopin talked a lot about this kind
of rubato, and I've been trying to do it in Chopin lately.

Isn't it difficult?

Chopin himself said it was difficult. He made his students practice the left

hand with two hands till they had it just right — because the accompaniments
must always be inflected as well — before he'd let them try to add the right hand.

/ don't know of many pianists who do much with that kind of rubato now.
Horszowski did: listen to his recording of the D]:> major Nocturne. Of course,
he was old; he was probably trained that way. Recent training shuns such things.

How about the agogic rubato — the usual kind — where both hands speed up
and slow down together. Do you use that in Mozart and Beethoven?
310 CLASSICAL AND ROMANTIC

I think the agogic rubato should be used, and I'm sure that I do. It's very
clearly shown in Earth's book that Beethoven did it.
I'm sure Mozart did, too. Now, there's a well-known quotation by Mozart
from a letter that so-and-so didn't play "im Takt."'^ That's usually translated
"in tempo," but that's not really the proper translation of Takt. Takt means
"time," means "beat," it means "measure," it means all kinds of different
it

things. I think what Mozart was saying, in effect, was that the music didn't
sound as if it was proceeding properly through time. I don't think that Mozart
was thinking that something should accommodate a metronome. The diffi-
culty with a metronome is that it doesn't distinguish between beats, bars, and
gestures. And all of the sources say that this beat is more important than that
beat, that this should be longer and that should be shorter, that this should

be heavier and that should be lighter.^" For instance, in the way you sang the
Wanderer, the two eighth notes were probably not exactly twice as fast as

the quarter notes. This belongs to a natural execution. Now, if you exagger-
ate that, and make the eighth notes conspicuously slow or fast, then of course
I'd say that you were playing out of time, because these things have to have
some relation to each other. As a basic rule, the sources tell us that the
long notes should be held longer and short notes played a little faster;^' this

is not done nowadays, but I think it makes perfect sense, as long as it is "im
Takt."

Again, the basic way to play —including the differing strengths and lengths

of equally notated notes — /5 not expression, but rather merely execution?

That's right, although of course expression and execution are always con-
nected. When I gave a lecture on this, I used a recording of Bartok playing a
piece called Evening in the Country, the first section of which is marked "with
rubato." He plays enormously freely —some eighth notes are more than twice
as long as other eighth notes. Then there's a rhythmic section marked "non-ru-
bato," but he plays the little figure -FJ faster than written. The music is strict,

but the little figure isn't. That's execution. He's not accelerating, he's not de-

celerating, but he inflects the rhythm. It's like playing a Viennese waltz; there's
a little hit to it, otherwise you're giving it the wrong execution.

And you say that people don't do such things nowadays?

One of the things that are missing in most modern playing is that there are
not enough ingredients in it. For example, I always tell pianists that they do a

19. Letter of 23-24 October 1777, mocking the playing of Nanette Stein, daughter of the
fortepiano maker: quoted in Rosenblum, Performance Practices, p. 23. Bilson draws a differ-
ent conclusion about Mozart's rubato than Rosenblum does, on pp. 383-84.
20. Tiirk, for example, wrote about this, concluding, "In poetry, no one has yet introduced
as desirable a meter which consists of nothing but one-syllable feet": discussed in Barth, The
Pianist as Orator, pp. 16-18.
21. This remained standard practice until after World War I, as shown by old recordings.
See Philip, Early Recordings and Musical Style, chap. 3.
 RESTORING INGREDIENTS 311

great deal with color and almost nothing with length. Even the greatest pianists
play virtually every note full length, connected to the next (unless they're play-

ing staccato). But length is an extremely important ingredient.

Another important ingredient is vibrato. String players even sixty years ago
used various different vibratos, or none, depending on the music. Today they
employ a single, constant vibrato. To me, it's not just a little bit crazy, it al-

most verges on lunacy. One hears a cellist in string quartet playing a low Bb as
—
an organ point and vibrating on it! This is something that I can never un-
derstand. They same way on everything. But unlike the older string
vibrate the
players, they rarely use another ingredient, portamento (sliding). Singers don't

slide much either nowadays. To have a word like "Liebe" [love] and not slide

between the two syllables is very hard for me to understand.

Another ingredient in piano playing which has disappeared, and which is

very important, and which you still hear in pianists early in this century, is not
playing all chords together but rather roiling some of them.

Bartok did that.

Bartok did all these things — rolling chords, playing rubato, not having the

hands together.-^ All of these are important ingredients that can be used and,

of course, abused.

Can't modern instruments execute in the old way?

Many modern pianos.
of the fortepiano techniques are difficult to do on
They're difficult to do on modern instruments, basically, because modern in-
all

struments are set up for long tones. But some could be done easily.
I'm teaching one day a week at Eastman this year, and I'm having a very
good time there. Very few of the students who play for me play on the old
piano I brought there —
never push them to. But one very talented young
I

woman who is playing Mozart is trying to, and I hope when she goes back to
the modern piano she'll bring something with her. But it's not because she sat
down to try the piano; it's because I said, "Look: look what's on the page
here." And she tried to play it on the Steinway and couldn't, and then she went
over to the smaller piano and found she could, and easily.
Because with a piano, once you've struck the tone, you can't really do any-
thing further. A violinist can learn to use his modernized violin and his Tourte
bow differently, and wind players can learn to do different things. But with the

piano, the one thing that's "set at the factory" is the aftersound; you can't do
anything about it, you can only adapt to it.

So does that imply that the piano repertoire most needs period instruments?
Well, when you say "period instruments," do you mean turn-of-the-century
Bechsteins with soft hammers for Debussy, and Steinways for Prokofiev?

22. Laszlo Somfai, Bela Bartok: Composition, Concepts, and Autograph Sources (Berke-
ley: University of California Press, 1996), pp. 279-95.
312 CLASSICAL AND ROMANTIC

That's an interesting question too. I hadn't meant that, but what do you
think?
I think that's true. People think of period instruments as meaning only old
ones, but it also means modern ones.

So what should a modern pianist do? A modern violinist might try to in-

corporate Baroque style, but do you have any specific suggestions for modern
pianists?
It's very difficult to discuss these things in the abstract. I realize that when
I learn these pieces, and then hear them in concert or hear records of people
playing them on modern pianos, I just read what's on the page so differently.
So, therefore, the overall interpretation comes out very differently as well. But

I think that what's being done with fortepianos might teach modern players to
play differently; I hope it will.

When I was a kid there would be a tough technical passage and I would
say, "This is not playable. I've been practicing this for three months and no-
body can play that!" Then I would get a record and hear somebody play it,

and I would be able to play it too — immediately! After three months of not
being able to play it and considering it unplayable. Many pianists talk about
imitating bassoons, and imitating voices, and imitating strings; now maybe
some of them can learn to imitate a piano!

SELECTED DISCOGRAPHY
Bilson was the first fortepianist to record the complete Mozart concertos, be-
tween 1983 and 1989. According to Joseph Kerman, "one can only admire his

responsiveness to the shades of expression called for in this amazing and amaz-
ingly diverse repertory. . . . He has to be the exemplary Mozart pianist for our
time."^^ Richard Taruskin particularly admires the concertos recorded after
1986 (Nos. 20-27, and 6 and 8); as the series proceeded, he says, Bilson dis-
"^"^
played "new poise and depth. Quoting an interviewer, Taruskin says Bilson
increasingly explores "'worlds of fortepiano color as yet undreamt of"; fur-

ther, he embellishes more, and his rubatos, "once virtually nonexistent, are be-
coming a trademark" (Bilson's comments above suggest that this may have
been in spite of the producers).
In the following year, 1990, Hungaroton recorded Bilson in the complete
Mozart sonatas. These performances take up from the high level on which the
concerto set left off; sample the audacious treatment of the left hand in the
opening of K. 310 (which Bilson discusses early in our interview) and of the

23. Kerman, "Mozart a la Mode," New York Review of Books, 18 May 1989, pp. 50-52;
quote, p. 52.
24. Taruskin, "A Mozart Wholly Ours," originally printed in Musical America (May 1990,
pp. 32-41), reprinted in Text and Act (New York: Oxford University Press, 1995), pp.
273-85; quote, pp. 283-84.
 RESTORING INGREDIENTS 313

harmonic progressions in the opening of the Fantasia, K. 475 —the hst could

go on at length. Bilson's rhythmic liberties reflect his concerns with speech-like

inflection, with slur markings, with the metrical hierarchy (see John Butt in

Chapter 9), and with the "psychological reaction" discussed in Robert Levin's

interview. In an essay about K. 475/457, Gretchen Wheelock praises Bilson's

recording for capturing "the enormous rhetorical range of the work," and for
exploiting "the expressive range of Ithe instrument! in a spirit of improvisatory
1 was reminded
freedom and discovery."^' Listening to the Fantasia and K. 310,
of what Nicholas Anderson wrote about Anner Bylsma's Bach suites:''' if these
two Mozart recordings had been released in 1965, they would have been de-
cried as self-indulgent and romantic.
Precisely such complaints greeted the first releases of Bilson's complete Schu-
bert series for Hungaroton (HCD 31587, featuring D. 537 and 959; and espe-
cially HCD 31586, featuring D. 850 and D. 568). Susan Kagan, a discerning
critic and respected pianist, objects to the "romantic" liberties Bilson takes; she

believes that Schubert should be played in a "more classical style."'" There is

evidence for and against her view; but as Bilson explains in his interview, and
as I hope emerges in later chapters, the equation of "Classical" with "strict"
may be a twentieth-century one. Other critics have found much to praise in these

Schubert releases. Nicholas A. Rast says of D. 959 that "Bilson's blend of spon-
taneity and distinctive contrasts of tonal colour in all movements winningly con-
veys both the music's potently dramatic use of motivic material and its large-
"-^^
scale psychological spans.
Bilson and some of his students are, as I write, recording the complete

Beethoven piano sonatas for Claves. The only Bilson Beethoven recording I've

heard is a set of the cello sonatas with Anner Bylsma on Elektra-Nonesuch

(79152 and 79236). Its music-making, says David Fanning, is "creative, spon-

taneous, and uninhibited," with "true Beethovenian drama."

''^
Those who dis-

miss the idea of Beethoven on the fortepiano should hear it; the period instru-
ments solve balance problems that modern pianos create in these works,
making it clear that Beethoven would have written the piano parts differently
for a modern grand. Fanning writes that Bilson's "first entry in the C major
Sonata sounds for all the world as if his partner is triple-stopping."

FOR FURTHER READING

Sandra R Rosenblum's Performance Practices in Classic Piano Music (Bloom-
ington: Indiana University Press, 1988) is a thorough, balanced exposition of

many of the issues discussed above. William S. Newman's Beethoven on

25. Wheelock, "Recovering Mozart's Fantasy," Early Music 24 (May 1996), p. 351.
26. And which I quoted to Bylsma in his interview.

27. Kagan, Fanfare 19 (May/June, 1996), p. 263.

28. Rast, Gramophone 74 (August 1996), p. 76.
29. Fanning, Gramophone 69 (April 1992), p. 93.
314 CLASSICAL AND ROMANTIC

Beethoven (New York: Schirmer, 1988) is an outstanding discussion of per-

formance practice in Beethoven's piano works. George Earth's The Pianist as
Orator (Ithaca, N. Y.: Cornell University Press, 1992) is an insightful analy-
sis of tempo flexibility and articulation in Beethoven, and of the eighteenth-
century rhetorical tradition in music, in which Beethoven was steeped. For a
subtle discussion of the "ingredients" lost in modern playing, and of how
they became lost, see Robert Philip's Early Recordings and Musical Style (Cam-
bridge University Press, 1992). Alfred Brendel's two books of essays {Musical
Thoughts and Afterthoughts and Music Sounded Out [New York: Farrar,
Straus and Giroux, 1976 and 1991]) contain fascinating insights into the Vi-
ennese Classical composers' piano works from one of their finest interpreters.
As Bilson mentions, another distinguished interpreter and important thinker,
Charles Rosen, has attacked the ideas behind the use of period pianos. A par-
ticularly subtle exposition is the first chapter of his The Romantic Genera-
tion (Cambridge, Mass.: Harvard University Press, 1995), which does not dis-

cuss the fortepiano in isolation, but mentions it in the context of a larger

discussion of the relationship between sound and music. His comment on
Beethoven's pianos is typically thought-provoking; he explains that Beethoven
worked with the gap between idea and realization, and that the strain this
puts on the listener's imagination is an essential element of his music. Rosen
then concludes, "The best argument for using the pianos of Beethoven's
time ... is not the aptness of the old instruments but their greater inadequacy
for realizing such an effect, and consequently the more dramatic effort re-

quired of the listener. The modern piano, however, is sufficiently inadequate

to convey Beethoven's intentions" (p. 3).

 17
Speaking Mozart's Lingo
— oOo^
gN

Robert Levin on Mozart

and Improvisation

Robert Levin became a fortepianist by accident. As a concert pianist and

Mozart scholar, he had met Malcolm Bilson at several conferences. Though he

had enjoyed their exchanges, he was astonished when Bilson asked him to par-
ticipate in recording Mozart's four-hand piano music. "But I don't play the
fortepiano," Levin said. "Ah," said Bilson, "You'll learn fast."
He did, though he still performs on the Steinway at least as often as on the
fortepiano. That kind of back-and-forth is becoming common among histori-

cal performers. What makes Levin uncommon is another element of accident,

one that he introduces into the Mozart works he performs: he doesn't play the
same notes at each performance but makes some of them up on the spot.
We now think of music in terms of a division of labor: the composer writes
it, the performer plays it. In earlier eras, as some previous interviews have in-

dicated, the labor was less divided. The performer was expected to contribute
a significant amount to the composing process; and, of course, the performer
often was the composer. In Mozart's time, genres like the opera aria and the
concerto still called for a good deal of performer input. In these works, Mozart
didn't write down all of the passagework, lead-ins, and cadenzas, but often
played (as he wrote to his father) "whatever occurs to me at the moment." This
leads to the paradox in the quest for "authentic" performance that Lawrence-
King spoke about: how can we apply the movement's original concern with fi-

delity when we're dealing with composers who composed their music anew at

every performance? you want to play the way Mozart played really the way
If

Mozart played you —had better learn to compose the way Mozart did, and be

able to do it "at the moment." As far as I know, Levin is the first modern per-

315
316 CLASSICAL AND ROMANTIC

former to have managed that. In an attitude that flouts standard modern prac-
tice, he follows period practice by not being faithful to the score. Levin even
improvises fantasias in Mozart's style on themes submitted by the audience —
feat that never fails to amaze.
Mozart's improvisations reflected an ethos that, as Levin points out, also
reached into what was then a relatively new institution, the concert hall. To
Mozart, playing a concerto was not an act of communion with a hallowed mas-
terpiece; was show biz. His letters show that he expected a kind of audience
it

response that would be found today at a jazz club but might get you thrown
out of some classical concert halls.
When did the ethos change in classical music to our current one —when, in

other words, did the music become "classical"? This question, too, features in
our discussion. To what Levin says would add only that the growing empha-
I

sis on sticking to the score went hand in hand with a growing concern with in-

tegrated musical structure' —since improvisation challenges the supremacy of

both. As I'll discuss in the "For Further Reading" section at the end of this
chapter, some musicologists have objected to improvising in Mozart concertos
because they believe the structure is too tight to allow for it.

It would be a mistake, by the way, to think that improvisation reduces the

need for rigor in musical training. On the contrary. Levin argues that to impro-
vise successfully one needs more rigorous training than is customary in mu-
far

sical education today. (He had the good fortune to have five years of study with
the legendary composition teacher Nadia Boulanger, who had earlier taught

Aaron Copland, Elliott Carter, Dinu Lipatti, and many other great musicians.)
Perhaps it's relevant that we spoke by telephone when Levin had just returned

to his alma mater. Harvard, to join its faculty. He hopes that renewed interest

in improvisation might foster a revival of higher standards of theory teaching

that is, a return to the tradition of teaching performers how to compose.

It's been said that the musical culture of Mozart's age was (ironically, given the
rigid class structure of his day) more like our popular than our "classical" cul-

ture.

There's no question that the boundaries between popular and —how should
one say it? — artistic culture, what's sometimes called "serious" music, were

1. Charles Rosen, The Romantic Generation (Cambridge, Mass: Harvard University Press,
1995), pp. 68-78, discusses the change to seeing musical works as abstract, independent aes-
thetic objects in the late eighteenth and early nineteenth century. Lydia Goehr argues for a far-
reaching change in the basic concept of music and art in this period; her work is discussed
briefly in the "For Further Reading" section of this chapter. See also Joseph Kerman's "A Few
Canonic Variations," Critical Inquiry 10 (1983), pp. 107-25, reprinted in his Write All This
Down (Berkeley: University of California Press, 1994), and Carl Dahlhaus's "The Metaphysics
of Instrumental Music," in his Nineteenth-Century Music, trans. J. Bradford Robinson (Berke-
ley: University of California Press, 1989), pp. 88-96.
 SPEAKING MOZART'S LINGO 317

much narrower in the past than they are now. In the late eighteenth century,

people were more comfortable with going back and forth. A man of the soil
like Joseph Haydn shows this most remarkably, but Mozart, who probably
identified more with the aristocracy than did Haydn, was certainly able to write

in the popular style when he chose to do so. It's seen in pieces like the Diver-
timento K. 287, where he quotes several folk tunes, and just as openly in The
Magic Flute.

Also, in our culture today it's the popular musicians, the vernacular musi-
cians, who command not only untrammeled adulation from the public but also
vast sums of money, and have to deal with meteoric rises and equally sudden
fallsfrom public grace. That's not unlike one aspect of Mozart's career, and
from that point of view your comparison is well taken. Mozart made fabulous
sums of money during the period when he was the darling of the Viennese and
Prague public, and he seemed to be making a comeback in 1791 before his life
was cut brutally short.

How about the culture of the concert hall itself? The modern concert, where
one is supposed to sit silent end of a piece, seems worlds apart from
until the

Mozart's concert hall, where people would applaud during the piece after a
striking passage, like a jazz audience.
Correct, and that reflects a concept of history that developed in the nine-
teenth century, butwas not at all a part of the concert scene when Mozart was
presenting his works to the public. The only kind of music that really absorbed
the public then was contemporary music. With the exception of some sacred
music, the Viennese didn't take much interest in anything more than two weeks
old. They did not want to hear the piano concerto that Mozart had written

three weeks or (heaven forfend) a year ago; they wanted to hear something that
was brand new. This kind of appetite seems remarkable in light of today's at-
titudes toward contemporary music, but the comparison is not entirely fair, be-

cause the musical language of the late eighteenth century was much more ac-

cessible to a well-educated listener. With this appetite for something current

comes a zest for it, and an audience that is seeking not to enjoy time-sanctioned
masterpieces but to be challenged, to be stimulated, to be astounded, to be con-
founded, to be overwhelmed with grief or ardor. That kind of public has an at-

titude much more like that of the public that goes to, as you said, a jazz or

rock concert —or to the movies.

With the growing attitude that music is a continuum with a glorious his-
torical tradition worth preserving came a more museum-like attitude toward

performance. The first of the great European orchestras, created in 1828 in

Paris, was called into being by Fran^ois-Antoine Habeneck for the purpose of

performing the symphonies of Beethoven —who had died the year before. This
shows that the bounds of interest were slowly shifting from a nearly exclusive

preoccupation with the present to this idea of a heritage. Along with the her-
318 CLASSICAL AND ROMANTIC

itage comes the sense of etiquette. And the sense of etiquette is a dangerous
one, because it can lead sooner or later to ossification, which it has, to a sub-
stitution of mores for content, which it has, and to a kind of obsession with
accuracy and responsibility, which it has —and not only among performers but
also, in some ways willfully and in other ways involuntarily, in the profession
of musicology. We have developed something which the Germans call Werk-
treue, faithfulness to the text, and also a sense that it's necessary in establish-
ing the text to strive to find the so-called Fassung letzter Hand, the composer's
definitive version.

But it's almost impossible to imagine a Fassung letzter Hand for Mozart.
He lived in an age of spontaneous performance, improvisation, embellishment,
and the inescapable demands that economics imposed on musical activities; all

this created performances that were constantly in flux. When Mozart revived
an opera and some singer couldn't do this or that passage, he revised the aria.
He added certain arias because a singer wanted more; and if it was a different
kind of voice, he changed or replaced extensively. These things were not seen
as beneath his dignity, because he lived in an age when composers still felt that
if the performers looked good, then the composer looked good.

As opposed to the mid-twentieth -century composer's idea of, "Who cares if

you listen?"^ When did the attitude start to change?
It changed in a critical way with Beethoven, who even said that his music
was not like the old music. Mozart's music was very often sight-read, which
was possible when people played in only one kind of musical style. Even or-
chestras seemed to have done this reasonably well, since Mozart did speak
about good and bad performances with orchestras although in terms of into- —
nation and ensemble we might perhaps consider what he considered bad per-
formances execrable and what he called good performances mediocre. Against
that background, it's very important to realize that Beethoven wrote music that
could not be sight-read. If Beethoven's music is sometimes awkward, it's not
because he didn't know how to write for the instrument, but because that was
what he wanted — and, for artistic reasons, had —to do. Berlioz is another ex-
ample of an astonishing innovator whose music certainly ought not to have
been sight-read if it was to make any kind of impact.

So the trends of our century —new music that's difficult,^ respectful treat-

2. "Who Cares If You Listen.'" This 1958 essay has been anthologized
Milton Babbitt,
often, including inContemporary Composers on Contemporary Music, ed. Elliot Schwartz
and Barney Childs (New York: Da Capo Press, 1978), pp. 244-50. The title was not Bab-
bitt's—he wanted to call it "The Composer as Specialist" —
but many people think it captures
an attitude of his era.

3. Rosen says in The Romantic Generation (p. 72) that the "impetus behind avant-garde
ideology [arose] at the very end of the eighteenth century. When the work of art is initially

rejected by the public, this provides its moral credentials; it demonstrates that the work was
not created for popularity or money, and justifies its success with posterity."
 SPEAKING MOZART'S LINGO 319

ment of a canon of works —had their seeds, more or less, with the coming of
the nineteenth century?
Of course, Mozart was considered difficult by his contemporaries; and some
early Romantics saw him as one of the first Romantics. But by and large his
stock fell at around 1830 or so, and rose again only in our own century."* Why
do you think this happened?
That is a fascinating thing. One of the reasons why Mozart did so badly in
the past was that he was always known as a precursor of Beethoven, and since
Beethoven had to be the greatest, Mozart could only be pretty good.
Beethoven's unforgettably obsessive rhythmic figures, a cellular way of getting
beneath the skin of the audience until he smashes atoms, is something we do
not regard as a Mozartian quality. Mozart's virtue was more the spinning out
of the line, with infinite rhythmic variety and flexibility, so that it always
sounds like the thing is being made up in front of you. Neither one of these is

better than the other — that's not the point — but previous generations seem to
have needed to rate people.
Mozart's "rising stock" also shows what happens when we begin to emerge
from the penumbra of this Romanticized view of things, which cast a very long
shadow. In it, Mozart continued to be regarded as a divine freak of nature
whose music, like a Hummel porcelain figurine, was always in exquisite taste
but on a somewhat limited scale, and whose claims to greatness had to be qual-
ified in light of the monumental and overwhelming achievements of subsequent

generations. Now, this is not to say that people like Furtwangler, Walter, Schn-
abel, Rubinstein, or Edwin Fischer didn't give transcendent Mozart perfor-
mances. One of the more brilliant of Taruskin's many contributions is his ob-
servation that the historical-performance movement reveals less about the
eighteenth century than it does about our age. This sudden fascination with the
re-institution of original bowings and so on is as much a phenomenon of the
twentieth century at its end as Furtwangler's performances were of the twenti-
eth century in its first half. And it's not possible to predict what fifty years from
now will qualify as representative performance of any of this literature, because
it's in the nature of things that very often we do not see our own prejudices

and our projections of our own values onto this music —you can call them lim-

itations if you wish to be negative. And I think that a composer such as Haydn
and Mozart might be astonished that everyone is so careful about doing it like

this or like that. Most likely they weren't careful about doing things; they just

did them.
But they did not have the problem that we have of two hundred inter-

vening years of music with performance styles and instruments so radically

different. To my ears, as long as Mozart was being played in the same per-
formance style as Schumann, his distinctive personality and its aesthetic un-

4. Leon Botstein, "Nineteenth-century Mozart," in On Mozart, ed. James Morris (Cam-

bridge University Press, 1994), pp. 204-26.
320 CLASSICAL AND ROMANTIC

derpinnings were likely to be washed over. Beethoven is for me a Classical

composer, but because his music was looking in a direction that the later

music represents, it didn't suffer in the same regard, except that the monu-
mentality led to a drastic disregard of Beethoven's tempos, and to sacrifices
in the liveliness of the surface.

Why do you think the historicist approach benefits Mozart's music?

Because in the case of Mozart, the baby was virtually thrown out with the
bath water. Those very sweet, long, legato lines served the purpose of showing
how tasteful and Olympian and perfectly balanced Mozart was. They did not
show how his music, like Haydn's, depends on a constant amazement, a per-
petual inconsistency with mercurial transformations from the flirtatious to the
grand, from the grand to the teasing, from the teasing to the beseeching to the
charming to the lyrical to the lamenting and back and forth — and often four
or five of these things within the space of eight bars. The music is this way
here we come back to the beginning of our conversation, about popular cul-
ture — because was designed to make an enormous impact on
it first hearing,
because there was not expected to be a second one.

Malcolm Bilson said this also reflected a new expressive need: "One essence

of the old 18th-century style is musical development [the "spinning out" of a

musical motif]; in the new style this is replaced by psychological reaction. What
comes in bar 2 comes there because it fits psychologically after bar 1, whether
or not it is motivically derived from that bar."^
Mozart was a dramatist. He was an opera composer first and fore-
Well,
most, and everything he wrote has to be understood in that regard. The con-
certos are a more polarized and evident example of that, but the dramatic prin-
ciple in his rhetoric is not missing from the piano sonatas or string quartets or

vocal music. And Mozart, being a man of the theatre, was constantly aware of
what is necessary to advance the plot, to entertain at the local level without
sacrificing the larger design. It is no coincidence that Mozart's music has such
an extraordinary variety of motivic material but that, in spite of the profligacy

of this richness, it never uses a theme without a hierarchical purpose. Archi-

tecturally, his is one of the most complex musics that Western culture has pro-
duced, rivaled only by a small number of works —one of which is Schoenberg's
First Chamber Symphony, Op. 9 — but not by Brahms or Haydn or Beethoven,
who were not temperamentally suited to this kind of composition.
Still, here again we can trace the boundaries between vernacular and art

music in a very effective way. From the moment the overture of Figaro begins
it's a smash hit, and after each act you go out singing the tunes, as you do at

a great Broadway show. Now, of course, the music is much more sophisticated

5. Malcolm Bilson, "Interpreting Mozart," Early Music 12 (November 1984), p. 520.

 SPEAKING MOZART'S LINGO 321

than Broadway in many respects; but in Figaro there is rarely an aria or en-

semble that doesn't advance the plot. And that kind of localized and long-term
development — which is also, by the way, reflected in Mozart's choice of keys
for successive numbers in an opera — is not the kind of thing that many other
composers were interested in. For instance, Mozart is the only composer in the

pantheon who took care in all of his mature operas to end in the key he began
in. This is good whereas not doing so is bad, but
not to say that doing so is

rather that Mozart's structural hierarchy and his dramatic control are not co-
incidences but are part of the same personality, and these things, along with
the extraordinary impulsiveness that Malcolm Bilson describes, all connect to
the need to grab the public immediately, to beguile them, to lure them, to se-
duce them, to charm them, as I said before, and leave them no room to let go.
The concertos, like the operas, were supposed to involve the combination of
theatricality and improvisation in terms of embellishments, and what they lack

in a specific plot they nonetheless make up for in the very careful delineation

of character that one can read underneath the musical surface.

How is character delineated in concertos

Tonality is a major factor. A piece in A major by Mozart could never be
confused with one in C major; if you know
his language well, you could tell

that therewas something wrong with a piece in A that was transposed to C.

Take the Piano Concerto No. 23 in A, K. 488, and play it in C major. It won't
sound right, because A major in Mozart is a key that's lyrical, sunny; we as-
sociate it with a special kind of radiance, a real iridescence often referred to as
autumnal because the clarinet works have such ineffable coloristic and expres-
sive feelings about them. On the other hand, C major and D major are much
more normal keys, D major being the standard trumpet and drum key and C
major being the majestic key.

Now, interestingly, none of the great composers used such a small range of
principal tonalities as Mozart did: there is not one piece Mozart ever wrote
whose principal tonality has more than four sharps or flats none. But on the —
other hand, the modulations within those tonalities in Mozart are more auda-
cious than those of just about any composer, with the possible exception of
Schubert. The conflict, the daring, the dislocation of such far-flung modulations
occur relative to a home turf that remains within bounds. It shows we're deal-

ing here with a remarkable kind of societal code; in Mozart these tonal wan-
derings have a sociological aspect to them. Thus, the delineation of character:
for Mozart, these tonalities represent ways of portraying people and types of
people. Heroes, anti-heroes, protagonists who are more delicate, or more vul-
nerable, assertive, grander — the decision to write a piece in a particular key is

already making a major statement. And because he was enormously conserva-

tive in his choice of keys, the way the artist represents the protagonists within
this frame is all the more riveting, because there's at least this external attempt
322 CLASSICAL AND ROMANTIC

to stay within these conventions. The straitjacketing of these conventions, the

pressures they bring to bear, can be all the more telling.

Does this choice of specific keys communicate anything to a listener who

doesn't have absolute pitch?
His audience was helped to a good sense of relative pitch by the use of
asymmetrical tunings, non-tempered tunings (on recordings I've used various
historical tunings). I won't say they're indispensable, because we're no longer
used to them, and a lot of people nowadays would think the instruments were
when Beethoven brings back a false reprise in the wrong
out of tune. But key,
people then knew it wasn't right because all the intervals sounded weird.

Even with these tunings, though, I wonder how much Mozart expected his
audiences to understand his system of key associations. After all, different peo-
ple had different associations.^ But one other way by which Classical com-
posers delineated character, as Leonard Ratner and Wye ]. Allanbrook empha-
size, is through reference to a large body of musical "topics" —musical features
(rhythmic, melodic, etc.) that by convention were associated in everyone's
minds with certain types of characters. The various dances each had well-
known associations to class and character; so did hunting horns, church styles,
and so forth.^ So, for example, K. 456 in Bh begins with four bars on a "mil-
itary" topic, but K. 595, which is also in B\>, begins with a "lyrical singing"
topic. Would you comment on the relationship between Mozart's key choices
and his use of topics?

6. Rita Steblin, History of Key Characteristics in the Eighteenth and Early Nineteenth Cen-
turies (Ann Arbor, Mich.: UMl Press, 1983) shows that this was a popular subject among
music theorists of the era, but that there was Httle consensus: different theorists' views of key
characteristics often conflicted.
7. See Allanbrook's Rhythmic Gesture in Mozart (University of Chicago Press, 1983), es-
pecially its introduction and first two chapters, and Ratner's Classic Music (New York:
Schirmer, 1980). If the idea of a repertoire of widely understood musical "topics" seems far-
fetched, consider our own musical culture. By alluding to rap, heavy metal, Sousa marches,
cool jazz, college fight songs, slow blues, show tunes. Dixieland, and numerous other musi-
cal topics, a composer can instantly make us think of the associated types of characters. Film
and TV composers allude to such topics particularly frequently.
The idea of a system of topical associations may remind us of what Page, Thornton, and
Hellauer say in this book about Gregorian modes, which may have communicated emotions
through a system of associations that people had internalized.
One question is whether the late-eighteenth-century topics communicate in that way
still

to us. To at least some extent, I think, Allanbrook is right in saying that some of them do:
we still recognize a descending chaconne bass in the minor as lamenting, and a gigue as light-
hearted. Some of this may reflect our having learned some "topical" conventions through our
exposure to them; but it is also possible that some musical topics relate to the thing they sig-
nify in a way depend on convention. For example, the "funeral march"
that doesn't entirely
(yet another topos) may imitate the slow, heavy way we move when we grieve; the gigue may
imitate the way we move when we're joyful.
 SPEAKING MOZART'S LINGO 323

Perhaps an appropriate analogy might be made between tonaHty in music

and locaHty in drama. In choosing to set a play or an opera in a specific place

the dramatist invokes the color, the mores, the dialect of its denizens. But the
choice of venue, whether exotic or familiar, scarcely limits the potential types
of characters. What it does is to put nobles, peasants, lovers, libertines, in-
genues, murderers or conspirators into a geographical context that provides an
overall frame. There are some kinds of activities or patterns of behavior that
are endemic to certain places. So it is with Mozart's choice of tonalities: B\>

major encompasses both lyrical and sprightly characters. But just as choosing
Paris as a setting excludes mountains or a seaport, Mozart's Bb major excludes
the majesty of trumpets and drums.
Even in his era, other composers did not equal Mozart's consistency and
conservatism in choosing tonal settings. And one could say that with the evo-
lution from Classical to Romantic music there is a motion away from these
kinds of sophistication, that delicacy of hearing, to —again — an ever more mon-
umental sort of style.

To help regain that delicacy of hearing, you've done research on Mozart's

style, and one of the methods you've used to study Mozart's language is sta-
tistical analysis: for example, in your reconstruction of the Symphonic Concer-

tante for Four Winds.^ Haven't some people found this a soulless, antiseptic

way to approach something as subtle and

alive as music?

Mozart specialists whose knowl-

That's exactly right, but even with certain
edge may exceed mine, I'll say, "Look
harmonic progression or thisat this

melodic figure; that's something which does not occur anywhere in Mozart."
And they'll answer back, "Sure, but he could have done that." And I'll say,
"Well, wait a minute; on the theoretical level he could have done anything, but
when you have close to a thousand pieces, including fragments, and something
never happens, then aren't you safer arguing that that thing lies outside of the
language?" If the idea is to try to define the language, you have to say, "Well,

they could have done that, but they didn't."

Weren't you criticized for being too objective by people who preferred the
idea of a composer as an intuitive genius?

The idea of being objective about these things flies in the face of why any-
one gets involved in art in the first place: they love the music, and they're very
subjective about it. If you get into a case of attribution, which is what the con-
tretemps with the Symphonie concertante for winds is about, nobody wanted

8. Levin's book \(/ho SfJrote the Mozart Four-Wind Concertante? (Stuyvesant, NY: Pen-
dragon Press, 1988) makes a convincing case that the Symphonie Concertante in Eb has come
down to us not as Mozart wrote it but in a nineteenth-century arrangement (K. 297b/Anh. C
14.0) for a slightly different group of instruments, in which the work is recomposed in a "di-
alect" later than Mozart's own.
324 CLASSICAL AND ROMANTIC

to be told on a statistical or analytical basis that a piece they hated was con-
forming to Mozart's plan, or that a piece they loved was not."" They wanted to
be left alone with their prejudices. I was continually being confronted by peo-
ple who couldn't understand how I could write a 500-page book about the
wind concertante and say nowhere in it from beginning to end whether I liked
the piece or couldn't stand it. But I wasn't interested in my own feelings about
the piece.
It seems to me that if you wish to be a linguist the first thing you do is doc-
ument. You go out in the field, you listen to people talk in a particular part of
northwestern Kentucky, and you listen to the color of their vowels and the as-
piration and pronunciation of their consonants. And then you say, "We found
this in this corner of the state, but if you go a little bit north, you begin to hear
the r's a little more like that." That's considered legitimate; but when you start

talking about Schubert that way, people get mad. But Schubert spoke, he had
a dialect. It was a sublime dialect, but once you want to find out what the di-

alect is, then speaking of it as being great or inspired or the product of genius
is not the point. The point is to try to decide what it is and what it does, and
to try to decide what its social fabric is, the kinds of conventions it observes
and the ways those conventions are violated, and whether any kind of code can
be drawn up that will account for the conventions and their violations. That
will never make you able to compose music of the greatness of that composer,
but it gives you a lot of insight. And if you want to improvise, to the extent
that you're capable of isolating the conventions and the exceptions you will be
able to fabricate music out of these mores and conventions that will sound to
most ears like a good replica of that music. It's as simple as that.
I mean, look what happened when they made Gone with the Wind: there
was such an enormous fracas because Scarlett was going to be played by an
Englishwoman. The elocution lessons and everything else that went on before
Vivien Leigh could take that part and not be the laughing-stock of everybody
—
south of the Mason-Dixon Line these were all considered absolutely neces-
sary. Well, why is it necessary in Hollywood and not in Carnegie Hall?

Perhaps it's that the aesthetic mode of film is realism, and all that period
detail helps us to suspend disbelief and forget that those are just actors up
there. Traditionally, people would say that instrumental music is more abstract
than film — its referents are less specific and "realistic, " so the musical charac-
ters you've described needn't look so exact in order for it to work. Besides,
people miss the musical subtleties more easily —most of us wouldn't notice if

one of Mozart's musical "characters" had a wrong accent.

Well, the funny thing about it is that it's like 'enry 'iggins. In some people's
cadenzas I can hear many things that are un-Mozartian; and, unlike someone

9. See Levin's exchange with Nancy Miller in Opus, August and November 1985.
 SPEAKING MOZART'S LINGO 325

who might have a hunch and be correct, I can even tell you why — but there
are undoubtedly sins that I'm committing that somebody with more knowledge
than I currently possess would point out immediately. And Mozart isn't around
to say, "Hey, you really got my lingo down!" or "Idiot! You don't understand
a thing!"

Let me ask you about an area where you might apply your analysis of the
language. Virgil Thomson criticized Schnabel for playing the passagework in

Beethoven too expressively, rather than treating it as neutral background ma-

terial;^'^' Paul Henry Lang made the same criticism of Leon Fleisher in Mozart.
How articulate should passagework be in Mozart and in Beethoven?
I'm working on the "Emperor" Concerto right now, and it's always re-

vealing to see the differences between Mozart and Beethoven in this respect.
Very often there are arpeggio passages in the "Emperor" that could be part
of an etude. They're relatively mechanical, and the middle-term scaffolding is
also an arpeggio. It's very rare to find something like that in Mozart. Mozart's
passagework is rarely devoid of melodic content. He's too much of a vocal,
operatic composer. Some pieces have more mechanical passagework than oth-
ers, but there usually is a cellular idea which fits into a middle- and long-
term shape. It doesn't mean that Beethoven is bad and Mozart is good. It

shows that Beethoven was explicitly interested in that mechanical element be-
cause he could get something from it. It's a much more overt kind of thing,
just like Beethoven's cellular repetition of rhythmic ideas. I think it can be an
irritant at times — it jars, it excites, it has a causticness, it has a willfulness
whereas there is always in Mozart (at a certain level) a perception of ele-
gance. Even when the music is decidedly angular, there's always an architec-
tural sense of harmony, of consistency of disclosure and rhetoric. And that's

why the passagework usually has a suave as well as a mechanical side to it.

There are exceptions, but they're few.

Regarding articulateness, also recall that if you open up the Breitkopf &
Hartel edition of Mozart (reprinted by Dover), every time there are sixteenth
notes for one or two bars there's a slur over them, and if there are more than
that it says "legato." Mozart never used the word "legato" in his scores in his

hfe; that was put in by Carl Reinecke, who edited the concertos. According to
Reinecke's values Mozart had to be balanced and beautiful and well-modu-
lated, and so he played it in the legato style. I play non-legato in Mozart be-
cause there is strong evidence that the premise of that period was that every-
thing is played non-legato unless marked to the contrary. I should point out
that this is an issue of clarity, definition, and balance. If you play legato on a
fortepiano with oodles of pedal, you'll never be heard. If you have an orches-

10. In a 1944 column, reprinted in A Virgil Thomson Reader (New York: E. P. Dutton,
1981), pp. 248-49.
326 CLASSICAL AND ROMANTIC

tra of any size, the sparkle, the crystalline speaking of the fortepiano will be
enhanced by non-legato. People who are of an open mind have said that they're
surprised at how they took to non-legato performance, because they found it

so lively in comparison to the other.

/ wonder what you think about the use of the continuo in Haydn and

Mozart. In Charles Rosen's view,^^ "in Baroque music the setting in relief of
the rate of change of the harmony [is] essential," because that's where the
motor impulse and energy come from —so the continuo playing is essential. But
in the Classical style, based not on those sequences of harmonies but
energy is

on larger phrases and modulations; so he concludes that the continuo's "em-

phasis of the harmonic rhythm is therefore not only unnecessary but positively
distracting" in Haydn and Mozart.
Charles Rosen is one of the most brilliant people of our age, and it's always
rewarding and provocative to read him. The distinction he makes between
those two styles is certainly relevant. But I wonder how Mozart or Haydn
would have responded to that comment. They might have said, "Oh, really?
Oh, well, maybe so!" This is not to denigrate what he says. The point that
Rosen really makes is that the music was evolving in new directions and the
composers didn't notice what that implied for performance.^^ This leads, he
says, to the question, Does the composer know how his piece is to sound? '^
Rosen seems to suggest that the composer might not, but I cannot imagine that
anyone would know better how his music is to sound than a composer like
Mozart, who was no mere abstract thinker: he had continual involvement in
performance (including leading the orchestra in opera and symphonic concerts).

But Rosen's viewpoint relates to something you've said in public several

times —that we can probably do better than the composer's contemporaries in

reconstructing the language because we're more familiar with it, having spent
so much time with it.

11. The Classical Style (New York: Norton, 1972), pp. 189-96 ("continuo" is defined in
the Alan Curtis interview, p. 135, n. 6). According to Neal Zaslaw, it appears that "under
many circumstances and in many repertories, the continuo instrument remained a part of con-
cert and opera performances": Zaslaw, Mozart's Symphonies (Oxford University Press, 1989),
p. 466. There is, however, heated controversy over whether they were used in non-theatre sym-
phony performances in late-eighteenth-century Continental Europe; see James Webster, "On
the Absence of Keyboard Continuo in Haydn's Symphonies," Early Music 18 (November
1990), pp. 599-608.
Rosen says that the use of continuo put Mozart and Haydn with "other performers
12.
of their day, whose idea of performance had not yet caught up with the radical changes of
style which had occurred since 1770, and for which Haydn and Mozart were so largely re-

sponsible." He says later, "there is no reason to assume that the composer or his contempo-
raries always knew with any certainty how best to make the listener aware of [the significance
of the music's meaning]." The Classical Style pp. 195-96. ,

13. Rosen asks, "does the composer know how his piece is to sound?" and eventually con-
cludes, "the composer's idea of his work is both precise and slightly fuzzy." Ibid., p. 195.
 SPEAKING MOZART'S LINGO 327

We've heard Mozart pieces a thousand times instead of once. And because
we're separated by a chasm of intervening styles, it's much easier for us to be

objective about things that people of that time would not have noticed at all.

Now, I play continuo because we know Mozart did it; he explicitly calls for it.

Still, we could say that it Mozart played along with the or-
doesn't matter that
chestra in tuttis of a concerto; we don't need to do that any more, so let's not
do it. And it doesn't matter that Mozart used the fortepiano, we have a Stein-
way, let's do it. It doesn't matter that Mozart's articulation was very much de-
signed around the instruments he had, which had consonants that spoke with
precision and a fast decay that produced a lively surface; we now have some-
thing which produces a smoother, more continuous surface, so let's do that.

You know — we now have cinderblock, so we don't need granite. I come back
to a culinary analogy: that's the cake that Mozart baked. That's the sauce he
concocted. And in that sauce, he says at the beginning of every single line of
every piano concerto, with exceptions that are as remarkable as they are im-
portant, that the piano is to play with the string basses when it's not soloing.
That's what he says. And in the early concertos the bass lines have continuo
figures. Admittedly the figures are mostly not in his hand, they're in the hand
of his father; but there's no chance — none—that his father didn't know what
his son had in mind; he was doing the mechanical stuff that could save his son
time. So there's no doubt that Wolfgang played continuo.
Again, you could say that he played continuo to keep the orchestra together;
we have a conductor, so we don't need that any more. Now, I wouldn't want
what I'm about to say to be misinterpreted, because I have had so many re-

warding experiences working with conductors, but in fact as soon as you have

a conductor you surrender the responsibility for the performance into the hands
of that conductor. When you play without a conductor and you have a con-
certmaster and a fortepiano (or Steinway) player, and they're seated in an inti-

mate circle around one another, they all listen because they have to make that
ensemble by themselves. The result is a performance that is likely to be much
tighter, much more active, and much more engaged than one with a conductor,
because there's collective responsibility. The pianist behaves like the timpanist,

keeping the orchestra rhythmically together in certain key sections, and the vi-

olinist leads in the melodic sphere —though sometimes there can be an overlap
of those functions. That also shows psychologically that Mozart's music is

never simply a conflict between a protagonist and the masses.

That brings us to Rosen's next point —that if the soloist plays continuo, it

reduces the effect of the soloist's entrances, which in his concertos are dramatic
events.
I'm rather surprised that a musician of Rosen's sophistication projected that
nineteenth-century conception upon Mozart. There are places where Mozart
flings down the gauntlet just like Beethoven: that occurs in the C Minor Piano
Concerto, K. 491, in the [first-movement] development, and it's thrilling when
328 CLASSICAL AND ROMANTIC

it happens. But how many times does it happen? Mozart has the piano ac-
companying the orchestra as often as the orchestra is accompanying the piano.
His solo-tutti relationships are much more sophisticated, involving symphonic,
concerto, and chamber-music elements. And the idea of a piano sitting in the

middle of the ensemble being an equal in the orchestral texture who at a given
pre-arranged signal rises up and becomes a personality in his own right is some-
thing I find enormously attractive. Also, I don't have to play "cold" when I get
to the first solo entry. (When one plays continuo with a Steinway, though, one
has to be careful not to play too loud.)

Rosen also says that in continuo playing the emphasis on the change of har-
mony is the only important thing —the doubling and the spacing of the har-
mony are secondary considerations.
Well, even in the Baroque era voice-leading and spacings were not secondary
but primary considerations —otherwise a text on thorough-bass could be a frac-

tion of the length of actual Baroque treatises on the subject: there is little art

in plunking out chords, but a great deal of finesse required to connect them
adroitly in a texture sensitive to the character prescribed by the composer, com-
plementing and deepening its meaning.
When playing continuo, I find it not very interesting to play just chords;
and I've been attacked for not playing just chords. But far from finding it dis-

turbing, I think it's wonderful for the audience to hear a roulade or a trill and
to know that there's something simmering in the pot. And I can't imagine the
world's greatest keyboard genius, with all we know about him, sitting at the
piano for 70 bars plunking chords. I suspect a rapscallion glint in the eye; he
was probably just jamming like a great Dixieland player.'"*

That brings us to the subject of improvisation, the most striking example

of which is that you don't write cadenzas out beforehand but improvise them
on the spot — the ultimate in getting away from the Fassung letzter Hand. Why?
Well,assume that most of these pieces were written for Mozart's own self-
I

expression, so to that extent whatever we know about his personal style ought
to illuminate what we do now. Otherwise I wouldn't improvise. Mozart's stu-

dents didn't improvise; they played prepared cadenzas, so I could do that. But
I don't find that nearly as stimulating, and the unanimous reaction of the au-
dience whenever I've played improvised cadenzas is that when the orchestra
stops after the 6-4 chord before the cadenza, the audience gets very quiet. For
the first time in most of their lives, they're at a classical concert where —despite
14. See Levin's "Instrumental Ornamentation, Improvisation and Cadenzas," in Perfor-
mance Practice: Music after 1600, ed. H. M. Brown and S. Sadie (London: Macmillan, 1989,
and New York: Norton, 1990), p. 288. But see also Tibor Szasz, "Beethoven's Basso Con-
tinuo," in Performing Beethoven, ed. Robin Stowell (Cambridge University Press, 1994), pp.
11-13, for a somewhat different viewpoint.
 SPEAKING Mozart's lingo 329

their familiarity with the piece —they don't know what's going to happen next.
On the other hand, it won't be a piled-on set of crashing dissonances, so they

can judge for themselves as well as anybody whether what happens succeeds
or fails. And that's a very hot seat for the performer. For me it's a place where
I use about as much adrenaline as I do throughout the rest of the concerto. It

takes a vast amount of concentration and coordination, and if the mind and
body are not perfectly synchronized there'll be a calamity.

The cadenzas are perhaps the example of your methods of master-

clearest

ing Mozart's language. You've written that the cadenza is "less a prolonged vir-
—
tuoso display than a decorated cadence" a way to create harmonic tension be-
fore the final resolution —and you mention that a Mozart cadenza has three
sections — well, I'm reading back to you what you've written.^
It doesn't mask the fact that there are Mozart cadenzas which don't do
those things. The cadenza in K. 488 uses a tiny snippet of material from one
of the least noticeable places in the concerto, just a little passagework, which
shows up in the beginning of the development; all of the big tunes are ne-
glected. That cadenza, revealingly enough, is not preserved on a separate piece
of paper, as was Mozart's usual practice, but is written into the autograph score
itself; that may suggest that this concerto is a unique case. But it's marvelous
to have that cadenza, because it shows that the harmonic principles are invio-

lable even though the question of how much motivic stuff you need isn't. You
can improvise a great cadenza that has just about nothing in it in terms of tunes
but keeps the harmonic juggling act going. What you can't do is write a ca-
denza that is harmonically stable but obediently reminds everyone that "and
then I played, and then I played ..."

You said earlier that you can hear things in other people's cadenzas that
aren't Mozartian; in an essay you give examples that have what you call "for-

eign accents" — stylistic features that are not in Mozart's language — in some
published embellishments of Mozart concertos?^
When you hear great Mozartians play, you can often hear a little Brahms,
a little Beethoven that creeps in. What's fascinating to me, though, is that mu-
sicians have never been concerned about this kind of temporal cleanness, the
kind of historicity we've been discussing in terms of linguistics and so on. It

has never been considered a sin to have Beethoven or Brahms present in a

15. In "Instrumental Ornamentation, Improvisation and Cadenzas," pp. 279 and 283,
which gives a technical analysis of the Mozart cadenza.
16. Ibid, pp. 277-78. That Mozart's music would have been embellished in, for example,
the returns of a many sparsely notated passages in concertos, has been
rondo theme or in

clearly demonstrated. See ibid., pp.269-79, and especially Levin's article "Improvised Em-
bellishment in Mozart's Keyboard Music," Early Music 20 (May 1992), pp. 221-33. See also
this chapter's "For Further Reading" section.
330 CLASSICAL AND ROMANTIC

Mozart concerto. When Brahms wrote cadenzas to the Mozart concertos they
sounded as much like Brahms as Uke Mozart, and the same principle applies
to Beethoven's or Clara Schumann's or Artur Schnabel's cadenzas. After all,

Beethoven was not concerned with this in his own music — his cadenza to his
Second Concerto seems to have been written around the time of the "Emperor"
and sounds like it, and it can't be played on the instrument for which the Sec-

ond Concerto was written. Beethoven was not interested in that kind of thing.
He'd probably say, "Well, why should I play on that old rattle-trap?"
Yet, as I was saying, could you imagine a Hollywood director doing a movie
that takes place in Chicago in 1930 with 1959 Chevrolets driving around? It's
interesting to me that this one so distinctly American idiom, the movies, is the
one place where these kinds of historical niceties are observed to the last T
but not in Beethoven and Schubert. We can say, "Well, it doesn't matter." Well,
it didn't in the past. Maybe it doesn't matter if the cadenza is completely dif-

ferent. But for me it has been an enormously revealing challenge to test one's
real ability to speak the language of the composer idiomatically. In the cadenza
you show that what you have been saying throughout the body of the concerto
is a personal utterance, whose legitimacy is proved by how idiomatically you

speak the language when you are no longer being fed the lines.

How do you translate the improvisation into the recording studio?

very difficult, because there has to be a primary version set down. Even
It's

in the main body of the concerto, away from the cadenzas, there are many pas-
sages that Mozart assumed would be embellished. And the beauty of those em-
bellishments lies in their spontaneity, perishability, and uniqueness. But when
you listen to that record fifteen times you may grow to like those embellish-
ments, and you may then not like some other ones, even by the same performer,
and even though those new embellishments might be, from the performer's (or
anyone else's) point of view, better than the ones on the record. So there is

something about recording which is antithetical to the freedom of improvisa-

tion. Nonetheless, in my first recordings with L'Oiseau-Lyre I recorded without
regard to the microphone, in the sense that each take had its own embellish-
ment, and the producer selected among them. And when it came to the ca-
denza, every time we had a take I improvised a different one. So a number of
improvised cadenzas were available for each place in this recording.

However, keeping in mind that a recording also has archival value, the peo-
ple at L'Oiseau-Lyre were worried about reviewers grouching that this record-

ing would be at a disadvantage to other performances because it does not use

Mozart's cadenzas. And I understand that; I have never claimed that impro-
vised cadenzas are going to be up to Mozart's standard. The idea is that some-
thing spontaneous sounds different from something that is not, and that the au-
dience benefits from that in a performance. And, in a sense, a recording that
you know to be of a live concert will always have an excitement that a studio
 SPEAKING MOZART'S LINGO 331

recording does not. Nonetheless, to be on the safe side, in a separate session I

recorded all of Mozart's own cadenzas.

So we can just program the CD player.

Exactly; the person who wishes to have all of the variants will presumably
have access to them. I have talked to the engineers about looking to the future
when a kind of random generator could select among a panoply of cadenzas. I

think we're on the verge of being able to do that. We decided that in doing what
we did we would be keeping abreast of subsequent technological developments;
we wanted to see how a new technology could be used to refresh an old per-
formance style. For once, new developments do not undercut old values.

On that point, Robert Philip's Early Recordings and Musical Style '^ argues
that recording itself changed playing styles, among other things fostering our
taste for precision.

Well, naturally, we rehearse, which they didn't do much; we use marked

bowings, which they did not do in that contemporary-oriented musical culture
of theirs. We much concerned with hygiene; you get a third take that
are very
is musically inspired but has five wrong notes, and most recording engineers

will insist that you fix them. So we do takes over and over again to get things

perfectly in tune, and then things are spliced together. I know an artist who
confessed to me that on one of his recordings there were 621 splices in about
50 minutes of music. Well, you can figure out how much spontaneity and how
much of an architectural arch is going to be found in something like that. We
live in an age that glorifies technical achievements of an Olympic sort above
everything else. Having done that, we reap the reward, which is that we now
get people who can play louder and faster than ever before, and better in tune,

but we do not have a generation of risk-takers; and this goes straight back to
improvisation. You rule out the idea that the performer is a creator, and turn
that person into a reproducer (which by the way did not need the great dan-
was already coming into place; the recordings merely cast
ger of recordings, but
the die in a more unequivocal way). Once that happens, the training of our
musicians becomes achievement-oriented.
And that's what happens today. I was the head of the theory department at
the Curtis Institute for five years and continued to teach theory for almost fif-

teen years, and I saw it very clearly. All of the students, with the smallest of
exceptions, fought tooth and nail the idea that they had to understand how the
musical language functioned. They wanted to practice and practice and do
nothing else so that they could win this and that competition and have a major

17. Cambridge University' Press, 1992. Neal Zaslaw argues, by the way, that we may un-
derestimate the technical standards attainable in Mozart's time {Mozart's Symphonies, pp.
504-06); but Philip's data strongly suggest that verbal reports of precise playing, like those
Zaslaw mentions, were made relative to standards that we would now consider low.
332 CLASSICAL AND ROMANTIC

career. It never occurred to them that once you stood in front of an audience
and began to speak and had nothing to say, it wouldn't be so good.
Now, they must be assuming that their instincts are powerful enough that
they would be able to do these things right. But instinct is a very tricky busi-

ness. It lies on a bedrock of cultural accretion, and that cultural accretion

comes about through being involved in the real events of the culture and its
—
evolution not on the skyline of mimicry. Now we have a generation of peo-
ple who have grown up learning music through recordings. A young aspiring
violinist listens 95 times to Heifetz playing the Sibelius concerto, and before
that person is even aware of it, out of veneration for Heifetz's indisputably
unique achievement he or she is using fingerings and bowings that Heifetz used,
absorbing these not by choice but automatically through mimicry. Then she or
he hears another marvelous artist play the Sibelius and doesn't like it because
Heifetz's ritard or portamento is missing, and another one appears somewhere
else. So we've become people who are less and less literate in our art —people
who often can't "speak" a who have no idea grammatically what
word of it,

a musical sentence is, who don't know that a parallel fifth is not something

that is grammatically acceptable in a work, who will change a passage to make

it more effective instrumentally and turn it into something grotesque and illit-

erate —even though they may be some of the most famous performers before
the public today. And I speak of experiences I've had as a member of the au-
dience.
That is something that fifty years ago would have been inconceivable.
George Szell and Artur Schnabel were composers; we may not know them as

composers, but that was how they were trained. There was no way that they
or a man like Wilhelm Furtwangler, who was also a composer, could possibly
be guilty of this kind of thing. They were architects within the music. Their
views of how the music goes may have been stylistically at odds with what we
now claim to know from the eighteenth century, on the basis of documents that
in some cases are really quite unequivocal; but nonetheless, the alleged sins of
that prior generation are as nothing compared to the lack of integrity of a mu-
sician performing in front of the public when the language has been absorbed
only through instrumental lessons and the habit of listening, rather than
through knowing what tension and release are and knowing how the music re-
ally functions in a palpable way.

7^5 like speaking a language phonetically, syllable by syllable, rather than

mastering it.

It's exactly like that. If one has any doubts about that, one only has to go
to the Neue Mozart- Ausgabe and look at the three sets of lessons we have from
Mozart as a theory teacher. We see there that regardless of what instincts and
intuitions he had, despite all of the fabled genius that we know and respect,

when musicians came into Mozart's care and he had to teach them, he taught
 SPEAKING MOZART'S LINGO 333

them with the same principles that are hated today by students who say that
theory is boring. He made them reahze figured basses, and harmonize melodies,
and write species counterpoint. He wrote the first half of a minuet and asked
them to write the second half. He made them write canons and various exer-
cises in free composition. We can see what the students wrote and what Mozart
corrected. And we see that then, as now, the idea was (as Renoir said), "First
become a good craftsman — this will not prevent you from becoming a genius."
The fact is that these things were regarded as matters of course by musicians
until recently.

This is a societal thing. We now live in an age in which people think you
don't have to play by the rules. Getting by is a creed. People think, "You don't
have to go to school —you can drop out and still become a football player or
supermodel or rock star and make millions of dollars." The old assumptions
about the work ethic and being scrupulous and honest and fair —those things,
if anyone watches television, are laughed at. In music it transposes to. If you
can play fast on your instrument and thunder your octaves, then it doesn't mat-
ter how ignorant you are —you're gonna get the jobs. So we have this unhealthy
alliance between concert managers, recording companies, and competitions.
And the result is that it's such a product/achievement-oriented thing that very
often the differences between the first-prize winner at competitions Y and Z,
or between the first- and fourth-prize winners, are tiny.

So in part, this reflects a change in the classical-music culture — to the per-

former being merely a reproducer — but also a weakness in the overall culture.

How do you relate these trends to your research and your work with impro-
visation?
One of the things that frustrate me the most, when I give lectures here and
there on improvisation, is that somebody comes up to me and would like me
to tell them the trick to doing, it in five sentences, so that they can go home
and do it immediately. That's our society at work. You don't have to sit down
and learn theory, you don't have to analyze music, you don't have to study
music; all you have to do is listen to five sentences and go out and start doing
it right away. 1 can only tell such people, "Too bad: it doesn't work that way."
I couldn't do it if I hadn't had those years studying with Nadia Boulanger, who
taught me to a fare-thee-well to listen and refine my hearing and my palate to
distinguish chord spacings and voice leadings and harmonic progressions and
structural articulations. I thank her for all of that. For years and years I tried
to teach this stuff with the same fervor because I think the survival of the art
depends on it.

Do you think that what you 're trying to do could give rise to increased in-
terest in learning music theory?
I would hope so. Look, in spite of all the pessimism I've voiced, there is no
334 CLASSICAL AND ROMANTIC

question that there are young people around who are doing marvelous things,
very imaginative things, who are less nailed to tradition and self-consciousness
than we ever were. Such people give me many grounds for hope. But I say to
the garden-variety conservatory student. You'd better learn how to do this, but
you'd also better learn what it means to do it; and it may mean that your mind-
less practicing ten hours a day is going to have to be curbed if you're going to
find out what it is you really need to do.

Regarding the direction of causality, do you think the decline in playing

standards has contributed in the decline of the size of the audience for classi-

cal music?
I would be tempted to make that kind of accusation, but it's not quite fair.

I would say that something about our concert life has become more and more
ritualistic and less and less healthy, and that invites desertion on the part of the

public. However, this must be qualified by the fact that in the United States
culture was transformed by the diaspora precipitated by the Nazi cataclysm in
Europe; the wholesale immigration of artists and their audience from Europe
to America in the 1930s changed the American cultural landscape. That gen-
eration is now passing from the scene, and their children, born in an American
environment, are shaped by the values of that society.
And in Europe, with the Americanization of culture there, the young peo-
ple, who drink Coca-Cola and eat Big Macs and listen to American pop music,
are as disengaged or disengaging from European art music as their American
counterparts. The danger is worldwide. When Richard Taruskin said in one of
his reviews that saving both vernacular and serious music was dependent on
finding some kind of bridge between the two cultures,^^ that song of Cassan-
dra is not limited in its poignancy or its urgency to our country. So there is

cause enough for alarm.

Now, in every age there have been incandescent performances by visionary
artists, and we have our share of those artists today, whom I believe in with-

out reservation, across the board —keyboard players and instrumentalists and
vocalists. Nevertheless, the overall pattern is distressing. We musicians have to
earn those dollars of philanthropic support. We have to earn those souls com-
ing and listening to us, do it by giving stuffy perfor-
and we're not going to

mances that are just warmed-over Backhaus or Heifetz or Szigeti.

SELECTED DISCOGRAPHY
Antony Hopkins called the cadenza "the orchestra's favorite part of a solo con-
certo." It has never been mine. A musicologist friend identified the reason
and at the same time the appeal of Levin's improvised cadenzas. In my (and
her) usual experience, cadenzas are often the parts that we hope will end

18. At the end of "A Mozart Wholly Ours," originally printed in Musical America, May
1990, pp. 32-41, reprinted in Text and Act, pp. 273-91.
 SPEAKING MOZART'S LINGO 335

quickly so that the real stuff can begin again; but if I knew I were hearing an
improvisation, it would more likely keep me on the edge of my seat.

Recording, of course, is antithetical to that sort of experience, but the first

discs in Levin's Mozart concerto series capture some of it (they include K. 271
and 414, L'Oiseau Lyre 443 328; and K. 413, 415, and 386, L'Oiseau Lyre 444
571). Reviewing K. 413, Stanley Sadie praises the performance's "hints of the
opera house in its characterization and its surprises," and calls Levin's playing
of the slow movement "very expressive and it brings out the latent rhetorical
quality in the piano line very effectively." As for the improvised cadenzas, he
'*'
writes, "Dare I say that in K413 his is as good as Mozart's?"
No critic has yet complained about the improvisations in Levin's Mozart,
but some have objected strenuously to improvised ornamentation in Levin's
Schubert sonata recording (Sony SK 53364).^° These critics are, I believe, mis-
taken when they call it historically unjustified. We have firm written evidence
that Schubert's favored singer embellished the songs in just this way when he
sang them (accompanied by the composer), and that was standard practice
this

at the time;^' it is inconceivable that Schubert himself would have felt more
constrained. Even these critics, however, sometimes praise Levin's Schubert
playing, and other critics have been unequivocal. For example, Nicholas Rast
says that in the A Minor Sonata, D. 537, Levin "delights in the varied tonal
characteristics of the 1825 Fritz fortepiano which he plays with a freer, more
flexible response to gesture [than Andras Schiff in the same work]."^^
Levin's recording of the last four Haydn piano trios (Sony SK 53120) — four
of Haydn's greatest works —with Anner Bylsma and Vera Beths is one of my
current favorite CDs. Sadie writes of Levin's "great vitality and delightful crisp-
ness," and the way he "puts across the intellectual force and the argumenta-
tive character of the music. "^' Sadie does find the slow movements of Nos. 42
and 43 too hard-pressed, but he praises the "eloquence and expansiveness of
Levin's playing" in No. 44 and especially delights in No. 45.
Levin is recording the complete Beethoven piano concertos with John Eliot
Gardiner on Archiv. The first release, the "Emperor," is distinguished by its

freedom and its vivid sense of rhetoric (Archiv 447 771). Erik Tarloff loves the
playing but says that to perform this concerto on the fortepiano "seems almost
perverse" —and concludes that had Levin "chosen to play his 'Emperor' on a
Steinway or a Bosendorfer, he might have given us one of the greatest record-
ings of the piece ever put on disc."^"* As a fortepiano fancier, I think Levin has

19. Sadie, Gramophone 73 (September 1995), p. 63.

20. Susan Kagan, Fanfare 19 (November/December 1995), p. 363; Nicholas Toller, "Schu-
bert in a New Light," Early Music 23 (August 1995), p. 524.
21. Walther Diirr, "Schubert and Johann Michael Vogl: A Reappraisal," 19th-century
Music 3 (1979-80), pp. 126-40.
22. Rast, Gramophone 74 (August 1996), p. 76.
23. Sadie, Gramophone 72 (June 1994), p. 70.
24. Tarloff, "Beethoven on Original Instruments," Slate (www.slate.com), 1 October, 1996.
336 CLASSICAL AND ROMANTIC

done exactly that. In it, by the way, Levin plays keyboard continuo, as all the
evidence suggests Beethoven would have.^^
Levin's completion of Mozart's unfinished Requiem has been recorded by
Martin Pearlman on Telarc 80410; Elliot Hurwitt places the performance
"among the best . . . now available."^* (Helmut Rilling has also recorded
Levin's version [Hanssler Classics 98979], but I failed to obtain a copy for re-

view.) When William Christie recorded the Requiem, he used the version pre-
pared by Franz Xaver Siissmayr, a student of Mozart, because it is "perhaps
the closest thing we'll ever have to truth . . . [Siissmayr] after all was there with
[the] dying composer."^^ But Levin argues that Siissmayr's completion suffers
from "grammatical and structural flaws that are utterly foreign to Mozart's

idiom. "^* This brings up a question raised in Levin's interview —whether we,
at 200 years' distance, might be able to gain a more idiomatic command of
Mozart's musical language than his near contemporaries did.
Fina[ly, Levin continues to play music of our own contemporaries: he is

recording the complete piano works of John Harbison and has made some re-

markable discs with the violist Kim Kashkashian on ECM.

FOR FURTHER READING

Music critics have had no qualms about Levin's improvised embellishments and
cadenzas in Mozart; but some musicologists have been skeptical. The debates
between those favoring such additions and those opposing them reflect, in cer-

tain ways, more basic aesthetic positions.

The strongest case for improvisation is made in Levin's own writings —the
best available introductions both to how ornamentation and improvisation
should be done in Mozart, and to why it is historically justified. The essay "In-
strumental Ornamentation, Improvisation and Cadenzas," in Performance
Practice: Music after 1600, ed. H. M. Brown and S. Sadie (London: Macmil-
lan, 1989, and New York: Norton, 1990), is especially useful, as is "Impro-
vised Embellishments in Mozart's Keyboard Music," Early Music 20 (May
1992), pp. 221-33.
The opposition is best represented by Frederick Neumann's encyclopedic
study. Ornamentation and Improvisation in Mozart (Princeton University Press,
1986). The cataloguing of historical data here is unmatched, and if you are se-
rious about this subject you must read this book; but be certain also to read
Levin's review of it in the Journal of the American Musicological Society 41

25. Szasz, "Beethoven's Basso Continuo."

26. Hurwitt, Fanfare 19 (January/February 1996), p. 266.
27. Christie, quoted in Joel Kasow, "An Interview with William Christie," Fanfare 19 (Jan-
uary/February 1996), pp. 72-74.
28. Levin, "The Editor's Perspective," booklet note to Pearlman's Mozart Requiem, Telarc
80410.
 SPEAKING Mozart's lingo 337

(August 1988), pp. 355-68. Neumann's conclusions are, as I implied, opposite

to Levin's —he would forbid improvisation and anything more than minimal or-
namentation. Another member of the opposition is Christoph Wolff (who, like

Levin, is a professor at Harvard). In "Cadenzas and Styles of Improvisation in

Mozart's Piano Concertos,"^'' Wolff argues that Mozart prepared his written
cadenzas for his own use, not his students', and that he wrote out many other
cadenzas for his own use that have since been lost. Some critics have com-

plained that Wolff cannot document either claim, and also that he offers no
concrete evidence for his explanation that Mozart's comment about playing
"whatever occurs to me at the moment" applied mainly to his early concer-
tos.^«

The evidence he provides, in fact, is his belief that Mozart's mature caden-
zas were too "motivically and metrically tightly controlled" to allow genuine
improvisation and, moreover, that they gave Mozart a chance to continue the
process of making "adjustments to a work." This plays into Richard Taruskin's
argument (which one can acknowledge regardless of which side one takes) that
both Neumann's and Wolff's arguments are attempts to defend a more basic
concept, which Levin also discusses in his interview Werktreue and behind —
it the concept of the unified, perfected musical work.^'
As Levin suggested in his interview, these concepts have not always been
common The evolution of the modern concept of the musical "work"
coin.
of musical pieces being integral works of art like paintings or sculptures is the —
subject of Lydia Goehr's fascinating book The Imaginary Museum of Musical
Works (Oxford University Press, 1992) and her related paper "Being True to
the Work," Journal of Aesthetics and Art Criticism 47 (1989), pp. 5-67. She
argues, to quote the latter (p. 5), that "the concept of a musical work first fully
emerged in classical music practice at the end of the eighteenth century and
that, since that time, it has been used pervasively in the world of music"; as
she often says, it began to "regulate musical practice" around then. One can
think of various examples of the work concept "regulating musical practice"
before 1800, and of numerous examples of later Western art music to which
the work concept is only partially relevant; but Goehr recognizes that such ex-

29. In Perspectives on Mozart Performance, ed. R. Larry Todd and Peter Williams (Cam-
bridge University Press, 1991), pp. 228-38
30. See Richard Maunder's review in Music and Letters 73 (November, 1992), p. 591, and
Taruskin's commentsText and Act, pp. 287-89. On the subject of Mozart and ornamen-
in
tation and improvisation, see also: Katalin Komlos, "'Ich praeludirte und spieite Variatonen':
Mozart the on Mozart Performance, pp. 27-54; Eva and Paul
Fortepianist," in Perspectives
Badura-Skoda, Interpreting Mozart on the Keyboard, trans. L. Black (London and New York,

St. Martin's Press, 1962), chaps. 8 and 11; Malcolm Bilson, "Some General Thoughts on Or-

namentation in Mozart's Keyboard Works," Ptano Quarterly 24 (1976), pp. 26-28; and
Henry Mishkin, "Incomplete Notation in Mozart's Piano Concertos," Musical Quarterly 61
(1975), pp. 345-59.
31. Taruskin, Text and Act, pp. 287-89.
338 CLASSICAL AND ROMANTIC

ceptions exist. Whether or not one accepts her ideas, it is unUkely that either
Werktreue or the work concept wholly regulated the performance (or creation)

of Mozart's concertos or operas^^ —although even if Wolff and Neumann's ad-

herents were to accept that, they could plausibly maintain that the concept
should have regulated the performances, or that they should now.
Wye J. Allanbrook's research on musical "topics" in Mozart sheds fasci-

nating light on the evocation of character that Levin speaks about. It's proba-
bly no exaggeration no more important book on Mozart's
to say that there is

musical language than her Rhythmic Gesture in Mozart (University of Chicago

Press, 1983). Her forthcoming The Secular Commedia, to be published by the
University of California Press, places topicality in a larger aesthetic framework.
Also, her mentor Leonard Ratner's Classic Music (New York: Schirmer, 1980)

is the groundbreaking study of the musical topics of this era.

More than other biographies I've seen, Maynard Solomon's Mozart (New
York: Harper Collins, 1995) gives the composer three convincing dimensions
and says interesting things about his music (notably in the chapter on the
Salzburg divertimenti). Solomon's Freudian interpretations are usually handled
tactfully, though at times they feel Procrustean or strained, and they are always
stimulating, even when one disagrees.

32. Charles Rosen, Sonata Forms (New York: Norton, 1980), pp. 9-10, argues that the
forms and techniques of composition that Haydn, Mozart, and Beethoven developed in cer-
tain genres led to a work of music existing as "an independent object" independent, that
. . . —
is, of text or virtuoso additions, or of primarily extramusical purposes. By contrast, the con-

certo and the aria still served "extramusical purposes."

 18
Taking Music Ojj the Pedestal
— NoOo ^

Roger Norrington
on Beethoven

Was Beethoven the first great Romantic or the last great Classic? If you believe

the movies, or the booklet notes for certain mass-market CDs, he was the ul-

timate Romantic, the rebel who burst the chains of Classicism. But if you be-

lieve most modern scholars, he was "the culminating composer of the Classi-
cal style." Yet recently, some Beethoven scholars, such as Maynard Solomon,
have argued that focusing entirely on Beethoven's "derivation from eighteenth-
century traditions" can oversimplify matters, by understating Beethoven's "rad-
ical modernism" and the "overlapping of Beethoven and Romanticism."'
Solomon concludes that "Beethoven's masterworks like his life arise out — —
of a perpetual tension between archaic sources and Utopian possibilities." Per-
haps it's this tension that allowed even Beethoven's disciples to put their own
spins on him. His students Carl Czerny and Ferdinand Ries described a
Beethoven who played "strictly in time"; yet some other musicians who knew
him reported a Beethoven who, to quote his self-styled Boswell, Anton
Schindler, played "without any constraint as to the rate of time." That may
sound like the birth of the Classical/Romantic dichotomy, but the issue gets
more complex: some argue that Czerny's "strict" Beethoven reflects the long
legato style of the nineteenth century, and that Schindler's apparently Roman-
tic Beethoven is rooted in the "speaking" style of the previous century.'

When "historically informed" players forged ahead to record the Beethoven

1. Maynard Solomon, Beethoven Essays (Cambridge, Mass.: Harvard University Press,

1988), p. X.

2. On this point see the "Postscript" at the end of this chapter.

339
340 CLASSICAL AND ROMANTIC

symphonies in the 1980s, they had no doubts that Beethoven should be played
Classically —which to them meant strictly. That they'd take this stand could
have been predicted by the least competent bookmaker in London. After all,

these pioneers —primarily Hanover Band and Christopher Hogwood's

the
Academy of Ancient —came
to Beethoven from an English early-music
Music
style characterized by energetic strictness. In addition, some of their spokesper-

sons expressed the view that in Beethoven's day, limited rehearsal time and the
lack of a conductor disallowed today's "wider variety of nuance and tempo
modification" in orchestral playing: the symphonies had therefore been given
"uncomplicated, rhythmical performances."^ In addition, some of the per-
formers claimed that such approaches allowed them to give performances that
were "accurately old."'*

Anyone could also have predicted that such claims would provoke Richard
Taruskin. Taruskin argued that these players, in propagating what he called the
legend of "Beethoven: Preserver of the Eighteenth-Century Tradition," were re-
ally making the music suit their own objective twentieth-century taste. Their
idea of Classical style, he said, is a historical fiction, fabricated by moderns as
a stick to beat the Romantics with. As for the musical results, he described one
of these musicians, Hogwood, as re-dedicating the Eroica "To Celebrate the
Memory of a Great Nebbish."^
At the time, Hogwood's goal was, apparently, to play music just as it was
played at the first performances. But when you apply that goal to Beethoven
symphonies, you run into its contradictions. In Beethoven's Vienna, everyone
agreed that orchestral standards had declined alarmingly;*^ to re-create them
would be to re-create something no one had liked at the time. To be true to

Viennese practice, the principal players would have to send substitutes to the
rehearsals and show up only for the performances. And to consider the result-

ing under-nuanced, under-rehearsed performances as reflecting the composer's

ideals is to ignore what we know about Beethoven's own conducting. Recall
Robert Levin's remark about how Beethoven was the first Viennese composer
to write orchestral music that "ought not to have been sight-read"; then con-
sider that when he rehearsed an orchestra he was "very particular" about try-
ing to get the players to realize "expression, the delicate nuances, the equable
distribution of light and shade as well as an effective tempo rubato."^ This

3. Clive Brown, notes to Hogwood's CD of Beethoven's first two symphonies, L'Oiseau-

Lyre 414 338.
4. Christopher Hogwood, "Hogwood's Beethoven," Gramophone 63 (March 1986), p.
1136.
5. "Beethoven: The New Antiquity," in Taruskin's Text and Act (New York: Oxford Uni-
versity Press, 1995); originally published in Opus, October 1987, pp. 31-41, 43, 63.
6. See Clive Brown, "The Orchestra in Beethoven's Vienna," Early Music 16 (February
1988), pp. 4-20, esp. pp. 4-6
7. According to an eyewitness, Ignaz von Seyfried; other accounts support him (see Brown,

"The Orchestra," p. 17).

 TAKING MUSIC OFF THE PEDESTAL 341

doesn't sound like a man with a strong preference for uncomplicated, rhyth-
mical performances.
In Taruskin's view, the only historically minded conductor to do something
honest and artful with Beethoven was the leader of the third British period-in-
struments Beethoven cycle to be recorded: Roger Norrington. Norrington, too,
claims Beethoven for the "Classical," but as our interview shows, he has hardly
lacked an interpretative point of view — a controversial one to this day, but an
influential one. Of the three cycles, only his has become a best-seller, has won
industry awards, and has affected the way many musicians and audiences — in-

cluding mainstream ones —approach Beethoven.

I spoke to Norrington in early 1994 when he was in San Francisco to con-

duct the San Francisco Chamber Symphony. Our interview took place on a
spring-like January morning at the mansion where he and his wife were house-
guests. Norrington couldn't have been further from the dictatorial maestro
stereotype. With his warm, open temperament, he has a gift for putting people

at ease. More importantly from an interviewer's standpoint, he is full of origi-

nal opinions, colorfully expressed —though he emphasizes that he is a per-

former, not a scholar. He shared his views generously and met my occasional
disagreements with his natural civility and humor. Lurking behind much of our
conversation was the question whether Beethoven was Classical or Romantic.

It's been said that you've influenced Beethoven playing more thoroughly than
anyone since Toscanini. A more recent ideal was Karajan, who was criticized
for being too smooth and refined. One thing he said was that one should not
be able to detect the presence of bar lines — it should be as if no bar lines are

there at all.

I couldn't disagree with Karajan more. Bar lines are terribly important in
Classical music —though, of course, not all are of the same importance. One of
the keys to determining their importance is the dance element, what's called the
"periodicity" of the music:** is it in four-bar phrases? two-bar phrases? three-

8. Norrington is referring to the grouping of measures into multi-bar phrases — imposing,

as Charles Rosen says, a "steady, slower beat over the beats of the individual bars." Four-bar
periodic phrasing "was already in frequent use in the early part of the eighteenth century," he
says, and by the last quarter of the eighteenth century "it dominated almost all composition."
In spite of the "slower" beat this imposes, "the music of the late eighteenth century actually
seems to move faster than that of the Baroque" —
he compares this to how in high gear the
motor of more slowly but the car moves faster. He later says, "For Beethoven
a car turns over
the four-bar rhythm takes on an even greater effect of motor energy than for the composers
of the previous generation, propelling the music forward; his deviations from it seem almost
always like an act of will": Rosen, The Romantic Generation (Cambridge, Mass; Harvard
University Press, 1995), p. 261; in a larger discussion of four-bar phrases, pp. 258-278. (In
his The Frontiers of Meaning (New York: Hill and Wang, 1994], p. 44, Rosen mentions
Beethoven's "insistent attempts to attack the rhythm of the bar line and to affirm it at the

same time.")
342 CLASSICAL AND ROMANTIC

bar phrases? And where are the irregular shapes, the five-bar and six-bar
phrases and so on?
Beethoven himself was totally aware of this, of course. For example, in the
middle of the Scherzo of the Ninth Symphony he suddenly writes, "Ritmo de
tre battute" and then "Ritmo di quattro battute": he expects you to beat in
three-bar periods, then go back to four later. Now, this relates to another ele-

ment of Beethoven — upsetting the expectations of bar lines. ^ But all his music
is written with reference to bar patterns.

And this is related to the dance element?

Absolutely. When I prepare a score that is an actual dance —the Prometheus
ballet music of Beethoven, or Les petits riens of Mozart, or the ballet in his

opera Idomeneo —the position becomes even clearer. My wife, Kay Lawrence,
is a choreographer of early dance, and we often work together. The lines I draw
in these scores to show phrase lengths are precisely where she sees the dance
changing direction —where the dancers go around one way and then the other
way, for instance. We're listening to the music, and she says, "there," "there,"
and "there." Her " there "s as a choreographer correspond exactly to my bar
patterns as a conductor.

And there are similar patterns even in the symphonies?

Yes.^° You don't actually dance to symphonies and concertos, of course, but
the structure of the music is related to dance music. Dance was the public's pri-

mary relationship to music of most kinds.

Would you say any movements in the Beethoven symphonies are in actual

dance style, other than scherzos?

You mean a first movement or a last movement. Well, let me see. We know
the Eroica finale theme is a contredanse, because he had already published it

as one. But many of his other movements have the flow and spring of poten-
tial dances.

What about other elements in Beethoven's thinking? For example, Leonard

Ratner identifies a number of "topics" in the Classical composer's "thesaurus":

Donald Francis Tovey points out that Beethoven is most hkely to upset or break bar
9.

patterns in his dance movements: "it is just where Beethoven's rhythms are most dance-like

that we encounter ambiguities and positive changes of stress the most vivid examples of
. . .

[this] are in Beethoven's scherzos. ... [in many places in Beethoven] the bar is still a typo-

graphical device rather than a constant rhythmic unit" (Beethoven [Oxford University Press,
1965 (1944)], pp. 69-72).
10. Leonard Ratner, in Classic Music (New York: Schirmer, 1980), writes that in music of
the Classical era that isn't an actual dance but uses dance elements, "the typical dance rhythms
are employed, but the length of sections does not conform to choreographic patterns of sym-
metry" (p. 18). Norrington said, when I sent him this passage, "That's what I'm sayingV
 TAKING MUSIC OFF THE PEDESTAL 343

one was dance styles, but there's also the "singing style," military and hunt
music, the learned style (what might be called the stile antico), and several oth-

ers.^^ And there's been a lot of discussion lately of how central the rhetorical
^~
tradition was to Beethoven's approach to music.

Of course, and you haven't mentioned dramatic declamation and recitative,

two more topics. The greater the music, the larger will be the frame of refer-
ence, and the more the ideas that will be clothed in passion. It is simply that
dancing was the strongest and most widely understood of these elements. Our
danger is that we don't think of Classical music as something that is useful any-

more. We think of it as awe-inspiring. We've sent it terribly up-market, put it

on a pedestal. People like Toscanini and Karajan very much put it on a pedestal.

It eventually became a replacement for religion, didn't it?

Yes, although I think it's been shown that both placing it on a pedestal and
exalting it like a religion go back to Beethoven's time, and that he played a key
part in spreading these ideas.^^ On the other hand, it's been said that the char-
acter of Beethoven comes across differently in your performances — in a way
that counters our mythic image of him.

A deeper implication of Norrington's pointis that the preoccupation of Baroque and es-

pecially Classicalcomposers with large-scale symmetry and "periodic" phrasing results from
the influence of dance music, whose prestige (and, therefore, influence on serious composi-
tion) increased owing to the prestige of the French court in the seventeenth and eighteenth
centuries. A clear discussion of the larger influence of dance on Baroque form can be found
in John Butt's Cambridge Handbook Bach: Mass B Minor (Cambridge University Press,
in

1991), chap. 6. Regarding the Classical style, see Charles Rosen, Sonata Forms (New York:
Norton, 1980), chap. 3, and Wye J. Allanbrook, Rhythmic Gesture in Mozart (Chicago: Uni-
versity of Chicago Press, 1983), chap. 2.
11. Ratner, Classic Music, pp. 9-24. He also lists the French Overture, the brilliant style,
the fantasia, so-called "Turkish music," the pastorale, and the Sturm und Drang style. How-
ever, he does say that "Dance topics saturate the concert and theater music of the classic style;

there is hardly a major work in this era that does not borrow heavily from the dance" (p.

18). Ratner details the dance elements that Classical composers used as "topics" (pp. 9-16).
12. George Earth's The Pianist as Orator (Ithaca, N.Y.: Cornell University Press, 1992) is
the most complete discussion of this topic. Other discussions include, for example, William S.

Newman, Beethoven on Beethoven (New York: Schirmer, 1988), chap. 6.

13. In this development — closely related to the emergence of the "work concept" discussed
in the Levin interview —Beethoven was a crucial transitional figure. Carl Dahlhaus, in Nine-
teenth-Century Music (Berkeley: University of California Press, 1989), pp. 94-95, and 183,
German discussions of the "religion of art" and of music as the
cites late-eighteenth-century

true revealer of religion; he also cites an E. T. A. Hoffman essay of 1814 (which Beethoven
almost certainly read) describing Beethoven's symphonies as the modern counterpart to Palest-
rina in revealing metaphysical truth. Dahlhaus also explains how the new bourgeois concert
audience sacralized art in a way the aristocracy had not; Beethoven not only suited this new
attitude, but helped catalyze it. An example of Beethoven placing music on a pedestal is his

remark, made to Bettina von Arnim, that world which does not intuitively feel
"I despise the

that music is a higher revelation than all wisdom and philosophy" (reported in her letter to
Goethe on 28 May, 1810).
344 CLASSICAL AND ROMANTIC

Yes, I badly wanted to restore the human side of his music, the vivacity, the
humor, the craziness, the elegance, the vulgarity —the exuberant range of ex-
pression which Beethoven's own contemporaries were already comparing to
Shakespeare. I felt that this "physical" side was essential both to his nature and
to his position as a survivor from the eighteenth century. What Beethoven and
Wagner meant as "mythic" were very different things.

In life, was plentifully human rather than wrapped in divine

too, Beethoven
mist. For instance, when he was writing the first eight symphonies he was quite
—
young he finished the Eighth when he was 42, and was still quite boisterous.
He was a bit shy of company because of his deafness, but when with friends
he was very good company, full of wit and repartee. If he had not lost his hear-
ing, he would have been very well, you know, he loved dancing and being
. . .

in company. He wasn't merely some sort of reclusive scholarly figure. He had

many sides. He was mortal.

It reminds me of one pianist's remark that Beethoven's music "is 85 percent

cheerful. " But what about his shaking his fist at fate?
Well, he might well have shaken his fist when he went deaf —who wouldn't?
We all shake our fists occasionally. And clearly he was a person of exceptional

musical integrity. He wasn't an ordinary person. But he did have ordinary

tastes —eating and drinking. We tend to forget that, because we idolize him. "He
couldn't possibly fart, could he?" We've got these reverential views.
But that's part of how I think about him — virility, energy, dance; a tremen-
dous brain for putting together this amazing music, and an incredible performer
on the piano. And an improviser, like all those great guys were — Beethoven's
improvising must have been staggering. And that's in the symphonies too.

Imp rovisation ?
Don't let me exaggerate. Classical music is not really improvisation. But im-
provisation lies near its creative heart. Beethoven could extemporize a sonata
form or a fugue. Any good performance, of course, wants to sound as if it's

being composed that moment, but a good development section must sound like
an improvisation —otherwise it seems too planned. I mean, if you think about
some nineteenth-century symphonies, the development sections are not terribly
convincing —they say it in the key of A, and then say it again in D, and then
again in G. You can tell what's going to happen a few bars on. With Beethoven,
you can never tell what's going to happen next.
A symphony may be a great structure, but it wants to sound as if it's being

—
made up particularly last movements. First movements are more about struc-

ture and argument — that's what the sonata form is, after all."'

14. —
The sonata form a way of organizing music, really, rather than a fixed form — is at

the heart of what musicians call the Classical style of Haydn, Mozart, and Beethoven. In a
 TAKING MUSIC OFF THE PEDESTAL 345

That brings up the topic of dramatic structure.

... the structure of drama, the architecture of feeUng. These were big
pieces. They were much the longest single movements that anyone was writing.
You don't get a single span of one tempo like that anywhere else, even in opera
finales. But they combined lightness and drama, think. 1

But Beethoven's concern for complex, tightly integrated structure, where

every note plays a part in the design — it did have a big impact. On this point

of unity and tight structure, most commentators have put great emphasis on,
for example, how he organized music dramatically, with a movement being
built around a dramatic conflict between two different keys (an approach he
got from Mozart and Haydn).
Sure. And of course the tonality generally fits the same bar patterns.

So you see them as integrated — the dance-based aspects and the tonal
drama.
Yes, they naturally go together: key and shape, speed and harmonic rhythm
[the rate at which important harmonies change]. The speed of the harmonic
rhythm is what I'm so constantly trying to elucidate in my performances: par-
ticularly in "slow" movements, because there, if you don't take care, the har-

monic rhythm moves unbearably slowly — a so-called Wagnerian idea, which

has been much in fashion since his time. Because conservatory-trained conduc-
tors were used to it in later music, they naturally made their Beethoven go
slowly. That's how they felt it. Slower for them was more "profound."

Wagner and were noted for conducting Beethoven slow movements

Liszt
more slowly than had been done before. Wagner said that an adagio can't be
too slow.
Yes, although he frequently complained that his own music was taken too
slowly, he always looked for drama and seemed to think that a slow movement
in Beethoven should be very slow, and the pauses should be very long; no
diminuendo, and a lot of sostenuto. He also made a big change of tempo for

sonata movement, the composer first establishes a home key (say, D major) and then creates
a dramatic conflict by moving the music into a second key (in this case, probably A major)
in a dramatized way. The two conflicting keys are often represented by contrasting themes
("subjects"). All of this takes place in the movement's opening section, which today is called
the "exposition." A second section (the "development") intensifies the sense of conflict; and
the third, final section ("recapitulation") resolves it, reconciling the conflicting material in the
home key.
Some historians say the sonata style grew out of symmetrical dance forms, expanding them
to a larger scale, turning their harmonic design into a drama, and increasing their range of
and variety. Charles Rosen's The Classical Style (New York: Norton, 1972,
internal contrast
esp. the chapter"The Coherence of the Musical Language") and his Sonata Forms are im-
portant books on the classical sonata style.
346 CLASSICAL AND ROMANTIC

the second subject. It could be very exciting. People liked it, and it became very
popular. Some people still like it!

I just find that it's not appropriate to the origins and gestures of the music.
And Beethoven's metronome marks bear that out. His metronome marks are
clearly in the Classical tradition. They're not "Wagnerian" at all, which is why
so few people follow them today.

Your fast tempos were the first thing everyone noticed about your
Beethoven set.

That was the scandal!

Although it wasn't really unprecedented — Toscanini, for example, often was

right up He was faster than you in the first movement of the Ninth. But
there.

the slow movements had often been a lot slower. What's your view now on
metronome markings?
I do them the same, I don't think I've changed. I suppose some people felt

I was just trying to stick to a rigid, outmoded code. But I just find them in-

spiring, you see. The only reason I do music in a particular way is because I'm

excited about it, not because it's morally "superior" or "politically correct." It

was crucial for me to take seriously any facts about Beethoven's music. But
since I had such an eighteenth-century training, I found his speeds as inspired
as they were comprehensible.
I must admit, I had a tussle with the Ninth. One does, because there are
one or two very strange tempos, and my performance may not be right —may
not be what he intended. It was an honest attempt to make those tempos work,
and they could be wrong. It's the first eight symphonies that I would swear by,
because he metronomized them together, in a block, seven or eight years after
he'd written the Eighth.'^ So he really knew those symphonies well. But when
he metronomized the Ninth he'd never heard it. He was guessing. He hadn't
had the experience, which everybody needs, of hearing the piece.

15. Arguments for the validity of Beethoven's metronome markings have shown that his
metronome was in good working order (Peter Stadlen, "Beethoven and the Metronome,"
Soundings 9 [1982], pp. 38ff.) and that he was consistent in the types of markings he gave
for similar movements (see note 20, below). One argument against their accuracy is that tempo
is influenced by acoustics and orchestral size, so that the tempo musicians "hear" when read-

ing a score is often different than the tempo that works in real performance. Further,
Beethoven, by the time he metronomized his symphonies, could barely hear real performances;
and he never had the opportunity to test the marked speeds in performance, something that
has often led composers to modify their scores. It's also argued that a composer's sense of
what tempo to take in a piece can vary with mood, age, and other factors; Beethoven himself
changed the metronome mark for the first movement of the Ninth from "108 or 120" for the
quarter note to a much slower (though still fast) 88. Recordings by composers like Stravin-
sky and Bartok of their own works support all these arguments —composers do not necessar-
ily follow their own metronome markings, nor are their tempos in a work consistent from one
year to the next.
All the same, it's reasonable to say that the metronome markings give a general sense of
Beethoven's tempos, and indicate that they were fast by modern standards, especially in the
 TAKING MUSIC OFF THE PEDESTAL 347

One of your metronome-based tempos that people had trouble with was in

the slow movement of the Ninth.

That one I'm absolutely sure of. The two that are really questionable are
the trio of the Scherzo, because he says "faster" and the metronome mark ap-
pears to say "slower," and the march in the finale. So those two are more foggy,
really.

It's been argued that he meant those to go twice as fast as the metronome
marks.^^
Yes. Ben Zander's recorded it that way, and it's exciting, but pretty wild. I

can't say I'm altogether convinced, but I'm not sure that I'm right either. Those
are two tempos I'd like to ask Beethoven about! The slow movement I would-
n't even bother to ask him about, because I'm sure it's right.

It's different in character from what we're used to.

He marks it "Adagio molto e cantabile." No doubt he meant it to be heard

in two main beats per bar; the feeling is "in two." If he had calculated a
metronome mark just for those two beats, it would have been "half note = 30"
that is, each of the two main beats would have been heard 30 times a minute.
But 30 was not on his metronome, so he marked it quarter note = 60. That's
how I would match the "Adagio" description with the metronome marking.
Secondly, in all his symphonies there are no "slow" movements, any more
—
than there are in Haydn or Mozart "adagio" is very rare in Haydn, and the
Italian word adagio doesn't even mean "slow" —
it means "easy." People at the

time usually referred not to "slow movements" but to Allegrettos, Andantes,

and so on.

The metronome marks for Mozart by Hummel, his student, and those for
Haydn and Mozart by Czerny, Beethoven's student, support you — the slow
movements aren't very slow.^"^ The same is true of the metronome markings Cz-
erny left for Beethoven's piano sonatas, and the markings Beethoven left for
^^
the " Hammer klavier. "

slow movements —
which is not to say that it's "wrong" to play them more slowly if we pre-
fer them that way.
16. Clive Brown, "Historical Performance, Metronome Marks and Tempo in Beethoven's
Symphonies," Early Music 19 (May 1991), pp. 247-58. Jonathan Del Mar makes a third sug-
gestion. He argues that the metronome marking half note =116 results from a mis-hearing
by Beethoven's scribe; Beethoven, he suggests, may have said "160" (the two words sound
alike inGerman as well as English). Another possibility is that Beethoven was giving the
metronome marking for the Scherzo (which is 116), not the trio. See Del Mar, "The Text of
the Ninth Symphony," Appendix 2 of Nicholas Cook's Cambridge Handbook Beethoven:
Symphony No. 9 (Cambridge University Press, 1993), pp. 110-17.
17. William Malloch, "Carl Czerny's Metronome Marks for Haydn and Mozart Sym-
phonies," Early Mustc 16 (February 1988), pp. 72-82.
18. Sandra Rosenblum, "Two Sets of Unexplored Metronome Marks for Beethoven's Piano
Sonatas," Early Music 16 (February 1988), pp. 59-71.
348 CLASSICAL AND ROMANTIC

Right. So, at 60, the Ninth's Adagio fits right into the Classical tra-

dition.

Thirdly, you didn't make

a mistake about 60. Even if you were deaf, you
knew was the tempo you were brought up with in the eighteenth cen-
60. It

tury. At home, you wouldn't hear the air-conditioning, or the fridge, or the

telephone, or television, but there was one mechanical noise you would hear,
and that was the pendulum of the grandfather clock. And they all went at
60. And if you were deaf, you could see the pendulum ticking at 60. I mean,
—
you just don't make a mistake about it 60's the one tempo that we can rely
on. If he'd put 48, that would be one thing; but 60, no.^^
Besides, it's a typical slow movement, and it sounds beautiful at 60, in
my view. If you think of the finale of the "Pastoral," it's that sort of world,
isn't it? The last movement of the "Pastoral" is also 60. So it doesn't sur-

prise me in the Ninth. Even some quite well-known Beethoven scholars say
they can't quite manage it. But I certainly don't find it difficult.

It's hard to adjust to if you're used to its being slower.

I suppose it /5 hard to adjust. It would be interesting to know how many
people have adjusted. On the other hand, my daughter, who hadn't heard it
any other way, heard a performance at the "old" tempo last year, and she
didn't recognize the piece! It was on the radio, and she said, "What the hell
is this music? It sounds vaguely familiar." Her contemporaries, the kids who
write to me from school, don't say, "You've changed my life": they say, "Why
do other people take these funny slow tempos?"
So it's a question of what you're used to. One tends to forget that new
people come out of school every year used to the alternative methods. It hap-
pens very quickly, this change of tastes. And as I said earlier, aside from the
two tempos about which I am unsure in the Ninth, in the other symphonies
all of the tempos are just wonderful speeds.

Some of movements are faster than we're used to, also.

the fast
That's right. first movement of the "Eroica" is 60, too. It's fantastic
The
at that speed. And when Beethoven was metronomizing, in some of the note-
books there are cross-references. He'll say, "Oh yes, that's allegretto, that's

90, like my trio in a certain other work." And you think, Yes, he's got a

19. Max Rudolf gives two other arguments for the approximate correctness of the
metronome mark movement. He says that the second part of the tempo marking, "e
for this
cantabile" ("and singing") was "a modifying afterthought." Says Rudolf, "Cantabile held a
special meaning for Beethoven, who once said, 'Good singing was my guide; I strove to write
as flowingly as possible.' When played at J = 40, as the Adagio has often been performed,
the melody is no longer singable in terms of human song. Moreover, the second subject of the
movement [Andante moderato] is marked J = 63, leaving no doubt that Beethoven felt little
difference in the pacing of the two themes." In his The Grammar of Conducting, 3rd ed. (New
York: Schirmer, 1994), p. 398.
 TAKING MUSIC OFF THE PEDESTAL 349

clear idea about these tempos.-" As opposed to, "Beethoven's a hopeless old
dotard, who doesn't know anything about his music. We know better." It's

an amazing viewpoint, isn't it? He died when he was —what—57? He was

younger than me, and we get instead [mimes a caricature of an old man on
his deathbed]. I mean, what is this?

That gets back to what you said about our reverential view of his char-
acter. Another aspect of that view is the notion that Beethoven was always
deadly serious. But together with Haydn, he was one of the greatest musical
humorists.
And how! I'm surprised at how much humor there is. For instance, peo-
ple don't expect to find humor in the "Eroica," but the last movement is hi-

larious, absolutely hilarious. It's as funny as any of the Haydn symphonies.

Some people think, "Oh, no, if it's heroic it couldn't be funny. Heroes don't
have a nice time." But to me, the hero is a person, and he's having a tremen-
dous time, in Heaven or wherever it's supposed to be, until the slow section
where he seems to be remembering the funeral march. We're in the middle
of the celebration, at the end of the battle, and we're remembering the peo-
ple who didn't make it —that seems to be what's going on. But then the humor
comes back again. It's hilarious in a very superior way. After all, humor does
not have to be silly. It can be revealing and inspiring.

Charles Rosen has written of instances of Beethoven's humor being based

not on a comic manner but on content.-^ Few, however, have applied the term
"hilarious" to that finale. Could you give an example of a joke from the
"Eroica" finale?
Some of the jokes, perhaps the kind Rosen's referring to, are very up-mar-
ket —expecting one key and getting another, or setting up a particular kind

of rhythm and then changing it. So the more you know about Classical style,

the more you enjoy that kind of high-table joke. But in the "Eroica" finale,
you also have this frenetic introduction, which sounds as if it's going to be
incredibly important and dramatic. And then you hear the pizzicato strings,
and you think, "What is this?" It's clearly absurd, but it's another 50 mea-
sures before you discover that it's the bass line of the tune, not the tune it-

20. This suppoaed by Rudolf Kolisch, Tempo and Character m Beethoven's Music,
is

reprinted in Musical Quarterly 77 (Spring and Summer 1993; originally published in Musical
Quarterly 29 [1943], pp. 169-87; 291-312). It shows that similar tempo marks and time sig-
natures in different works tend to get similar metronome marks. Better still is the summary
of Hermann Beck's research on this subject, in William S. Newman's Beethoven on Beethoven,
pp. 90-97.
21. Regarding the finale of Op. 101, which Beethoven tells the performer to play "with
decision,"Rosen says the humor is based not on manner but on "contrasts and surprises."
Rosen, notes to his 1971 recording. The Late Beethoven Sonatas (Columbia 30939-41), M
p. 3.
350 CLASSICAL AND ROMANTIC

self. And the tune is one Beethoven had already used three times in other
works:^^

oboes

H dolce
I
f" n r M ^ Mr g

So you can guess that he expected his audience to know it by then. They'd
heard it as piano variations, they'd seen it in a ballet, in the finale of
Prometheus [where it probably represented Bacchus], and he published it as a
dance to be used during Carnival. They'd danced to this damn thing before
hearing it in the "Eroica. " So suddenly they have this sort of Evita tune com-
ing up, and they knew it was a fun tune, because contredanses were fun dances.
They were not like aristocratic minuets; they were family dances, jolly middle-
class dances.

The "Eroica" middle-class? Surprise! But he was showing what a nice guy
the hero was. He wasn't stuffy. He was a man of the people, the way all he-
roes should be. All the great heroes, all great gentlemen in the eighteenth cen-
tury, like Washington, for example, had the "common touch." They didn't
make servants feel uncomfortable. They had a way of dealmg with their farm-
ers and their staff; people were admired for that. They weren't all up on huge
pedestals, you know.

It ties in with Beethoven's tearing up the dedication to Napoleon when he

heard he had crowned himself emperor.
Right, it's common man as well. And Beethoven, above all, would
for the
be somebody who would prefer that the hero be accessible. So in the finale
there's this jolly contredanse and all these larks. There are lots of jokes. The
fugues too are jokes, aren't they? Incredibly overcomplicated and hilarious. It's

very exciting — it's like skiing at high speed through a forest. You've got to think
fast and you've got to listen fast, and it's a lot of fun. So why shouldn't music
be fun? Mozart and Haydn showed how it could be done.

Maynard Solomon says that Beethoven's heroic works, for all the fright-
ening emotions they let into music, are not "conventionally tragic, let alone
death-haunted," because they usually close on "a note of joy, triumph, or
transcendence." He relates this ultimately to "the essential features of high
comedy. "^^
Beethoven wrote ten operas, didn't he Fidelio and the nine symphonies. I

22. Beethoven, Symphony No. 3 in E\>, Op. 55, "Eroica," fourth movement, bars 77-81.
23. Maynard Solomon, Beethoven (New York: Schirmer, 1977), p. 194.
 TAKING MUSIC OFF THE PEDESTAL 351

think they're all incredibly operatic. There are the comic operas, the Second,
the Fourth, and the Eighth, or the rescue operas, the Fifth and the Ninth. The
Seventh is more about dance — I think Wagner is right — but let's face it, dra-
matic dance. And the Sixth is the oddball, because it's so pictorial.
Every symphony is different from every other, amazingly different, but he
was a naturally dramatic composer. He had a hard time writing his first opera,

and he never got around to another. But he wrote very dramatic symphonies.
The Second Symphony finale is a hilarious comic-opera finale, the Fifth a
heroic-opera finale.
After all, symphony and opera came time. The time
to power at the same
when opera houses really opened to the Hamburg, Lon-
public, outside Italy,
don, and a few other places, was the mid-eighteenth century, just when the sym-
phony was coming up. The symphony is a naturally operatic animal. The two
genres seem to me to be totally allied. They share the same language, and that
language is Italian.

Rosen argues that the timing of the Classical language is that of Italian
wanted to ask you more about performance practice issues. When
opera.^'* I

you developed your interpretations, it obviously wasn't just a matter of using

period instruments and following the metronome markings. You wrote in

your CD booklet that orchestra size and pitch were not crucial to the era's
style — they varied from place to place, even in one —
town but that other
factors like bowing and phrasing were crucial. Would you describe some of
them?
Let's see. To begin with, they didn't seem to use much vibrato in orchestral

playing (though soloists did use —

some vibrato some more than others). That
gives you much cleaner textures, —
so you hear more it's more transparent when
there's no vibrato.

And that would aid transparency at the fast tempos.

At any tempo! So does the fact that the woodwinds are all different in sound
—
from each other they're less homogeneous than today, so that makes a dif-
ference. Returning to the strings, Spohr (Beethoven's contemporary, who knew
him and played under him in at least one early symphony, and wrote a book
on violin playing) didn't use spiccato at all [the technique, much used today, of
repeatedly bouncing the bow slightly off the string]: it clearly wasn't part of his

style. ^^ Beethoven may have intended some spiccato, but probably much less

than we are taught today. The basic bowstroke, even for shorter notes, was on

24. Rosen, The "The Origins of the Style."

Classical Style, chap. 3,
25. Spohr claimed that spiccato "went against the Classical tradition in German violin
playing." See Clive Brown, "Bowing Styles, Vibrato, and Portamento in Nineteenth-Century
Violin Playing," Journal of the Royal Musical Association, 113 (1988), p. 106.
352 CLASSICAL AND ROMANTIC

the string, not spiccato. So we have a definitely on-the-string style for quite a

lot of the time.

Then we use a style of bowing which derives ultimately from Baroque prac-
tice. We tend to use down bows for strong notes and up bows for weak ones,
more so than is done today. For instance, the beginning of each bar is a down-

bow normally,^*" unless there's a specific reason for it not to be.

Does that make the music more metrically accented?

It makes the music a bit more dancy. Then note length seems to me equally

important. We don't play staccato on notes unless Beethoven says to do so. We

don't just play all the short notes short, which is what modern orchestras tend
to do. We play the staccato pretty hard, because it's clear from the manuscripts
that he doesn't show dots, he shows daggers — really sharp-looking things.

Often they're reproduced in modern editions as dots, rather gentlemanly-look-

ing things (you know, put them in a club without raising eyebrows). But they

were clearly something much stronger.^^

In general, when there weren't dots or daggers on the notes there was a
smoother style of playing. The slur is the smoothest; the portato is smooth;^*
the notes with no markings are separate but long. If you had just one rehearsal
(the norm in those days) these small markings were crucial.

Are you saying that the articulation of notes in Beethoven tended actually
to be smoother?
Some smoother, some rougher: a big variety —and side by side.

How about the phrasing?

It was shorter than today, I think. But many short phrases can be beauti-
fully modulated into one long one. And the clarity and amount of the phras-
ing makehuge difference. There are two reasons for a different approach to
a
phrasing. The first is that Beethoven's musicians had a whole series of conven-
tions, which came from the Baroque (and which every musician was taught),

so that they knew where to play loud and soft without being told. They had

(known as the "rule of the down bow") is first reported in the 1590s,
26. This practice
and was dominant in eighteenth-century France, as Anner Bylsma observes in his
especially
interview. Although some important eighteenth-century Italian virtuosi —
Geminiani (1751)
—
and Tartini (1771) opposed it, and it was not applied rigidly in Italy as it had been in Lully's
orchestra, the principle continued to have relevance to orchestral playing in Beethoven's Vi-
enna and, for that matter, has relevance even to modern mainstream playing. See Robin Stow-
ell, Vtolin Technique and Performance Practice in the Late Eighteenth and Early Nineteenth

Centuries (Cambridge University Press, 1985), pp. 303f.

27. See Clive Brown, "Dots and Strokes in Late 18th- and 19th-century Music," Early
Music 21 (November 1993), pp. 601 and 607.
28. The portato marking is discussed in detail by Malcolm Bilson in his interview, p. 308.

Also see Clive Brown, "Dots and Strokes," p. 607.

 TAKING MUSIC OFF THE PEDESTAL 353

this whole system of what they called "good notes" and "bad notes. "^'' You
inflected everything; music is a language with inflections in it. People like
Leopold Mozart taught that right from the start you must stress one note and
lighten another. If you have two notes, one of them has to be louder than the
other. You have to decide. You can't just have uh-uh lillustrates two equal
stresses], which Stravinsky often expressly calls for — that's his style. With
Haydn, Mozart, and Beethoven that's impossible. So phrasing is absolutely
built into the way we play, and the lengthening and shortening of notes comes
into that, too, and accentuates it.

This system of "good" and "bad" notes reminds me of the Kodaly method
of music education in Hungary, where the first thing you sing is your own
name. Not duh-duh. It always has an inflection, like somebody calling, "MAH-
mee."

That example seems to reflect the Baroque/Classical idea that music is like

oration.
Right. But the second reason for that different type of phrasing takes us
back to the dance. Rhythm does not constitute dance; phrasing constitutes
dance. Phrasing what makes you want to dance. Rhythm uh -uh-uh -uh-uh
is —
isn't dance. Nowadays, we have consciously "primitive" dance rhythms but —
it isn't what they thought of as a dance. They had to have ONE-two-three,

ONE-two-three. If you play for that kind of dance, you have to propel the
music by phrasing. ONE-TWO-THREE doesn't make you want to dance. You
have to play ONE-two three. It elevates the first beat. They felt that in their
bones.
Don't forget, there wasn't one single musician in Vienna who didn't play in
a dance band for some part of the year. The court orchestra played at the court

balls for the whole of Carnival. The theatre orchestra played for balls at the

theatre and in the Redoutensaal. If you were a fiddler you played for a danc-

ing master. You couldn't be a musician and not play for dance. And if you play
for dance (as I was lucky enough to do for country dancing, when I was young)
you get to know how to move people to dance. When you reach the end of a
phrase you've got to take the music forward Isings wonderfully, with an irre-

producible lilt, an English folk-dance theme]. You've got to keep them amused,
to keep their feet going. It's the same in the concert hall —you keep their minds
and their feet amused. Keeping the sense of onward movement is done with
phrasing, not with tempo.

And this relates to making sense of Beethoven's tempos?

This relates to making sense of the tempos —the tempo's success or other-

29. Downbeats, for example, were "good" notes; second beats were "bad." See the inter-

view with John Butt, Chapter 9, for further explanation.

354 CLASSICAL AND ROMANTIC

wise is in the flavor of how you make the music move. When you get to a very
fast tempo, by showing the periodicity of that phrasing you can give it a poise.

Consider the "Eroica" opening: DUM de dum-dum, DUM de de dum [he ac-

cents the first beat of the first and third bars of the phrase]. That feels slow be-
cause it's a two-bar phrase:^°

Allegro con brio

Vcl.

Whereas DUM DUM

duh [accenting the first beat of each bar] will sound
de
more hurried. Mozart and Beethoven are constantly playing with that ambigu-
ity. Of course, it is particularly important in slow tempos. A slow movement

has lots of small periodicities that keep it going. If you play them too slowly,

those periodicities turn into a series of chains around your ankle. Even in the
Adagio that you heard on Monday, the opening of the Haydn Forty-ninth, I

said, "Don't think of six beats, just because there are six eighth notes in the

bar. Don't even think in three —think in one. Although it's Adagio, you can eas-
ily let it fall to bits." It's the phrasing of the slow stuff which helps you dance
to it, and stops it from being boring.

A related issue is rubato and tempo fluctuation. This has been a source of
controversy surrounding your recordings. What do you think about tempo fluc-

tuation in Beethoven?
In Beethoven I don't vary the tempos very much at all. I may be completely
wrong, and he may have varied it much more. I believe he may have played
more freely in solo sonatas, but not nearly so much in public, symphonic
works. It would have been a very difficult thing to do, anyway. People didn't
conduct much —there weren't virtuoso conductors around yet, in the modern
sense. And, then, fluctuation is not so relevant to a big public piece. A sonata
was to be played in a salon, so it's free; it's like an intimate conversation. But
when the king was addressing his people, he didn't slow up for the second sub-
ject. A symphony is just such a public occasion. It's a difference we no longer
observe today, because all of our music has become public, but it's an impor-
tant difference.
The main reason, though, is that for me tempo fluctuation just doesn't feel
necessary in Beethoven. As soon as I get to Mendelssohn, I feel it necessary to
vary the tempos. I feel the need to do it. That's guessing too, of course. We

30. Beethoven, Symphony No. 3 in Et, Op. 55, "Eroica," first movement, bars 3-6. No-
tice how Beethoven's slur marks support Norrington's approach: the two-bar slur discourages
an accent on the first beat of the second bar.
 TAKING MUSIC OFF THE PEDESTAL 355

know Mendelssohn didn't do it very much —Wagner thought he was a bit of a

prissy Kapellmeister, didn't he?" But it seems to me necessary. In the second

subject of the "Reformation" Symphony, for instance, I slow down quite a bit.

I just can't go straight on. But in Beethoven I feel a need, on the contrary, for
a kind of Classical uniformity, as in Haydn. Performance must always be in-

formation and feeling. The information is there, if you like, to help educate the
feeling.

There are reports of Beethoven slowing down and speeding up in his

playing, and some about his use of tempo rubato in conducting;''- but
Rosen argues that in orchestral pieces, the structure doesn't call for much
rubato inflection, reflecting performing conditions that made it hard to do
rubato.''''

Yes, one rehearsal was normal for a concert. Of course, I don't adhere

rigidly to each tempo. If you listen to the slow movements of the "Eroica" and
the Ninth, you'll find a lot of changes of tempo, but they are slight and (hope-
fully) well modulated.

Many of these issues reflect the fact that this was a transitional period, both

in performance practice and in society's views of the artist and the arts.

Yes. And Beethoven was full of new ideas. The fact that I keep talking about
him as an eighteenth-century composer doesn't contradict the fact that he was
looking forward. He was incredibly inventive.

'"''*
Solomon has emphasized his "radical modernism. Nonetheless, it seems
that you're trying to approach the music in terms of its historical antecedents

rather than its consequents.

31. Donald Mintz's article "Mendelssohn as Performer and Teacher," in The Mendelssohn
Companion, ed. Douglass Seaton (Westport, Conn.: Greenwood Press, forthcoming), argues
that Wagner's On Conducting, with its attacks on Mendelssohn, was "in the first instance an
anti-Semitic tract and only in the second a treatise about conducting." But while Mintz sup-
ports Norrington's view that extreme tempo variations within a movement came in only with
Wagner's conducting, he also shows that moderate tempo flexibility (relative to a single un-
derlying tempo) was part of nineteenth-century conducting style before Wagner
Interestingly, Joseph Wilhelm von Wasielewski, a conductor and a student of Mendelssohn,
wrote in 1883 that Mendelssohn "was exceptionally free in conducting his own work but very
strict in everything else" (David Fallows, personal communication, 1995). This speaks well
for Norrington's instincts with this music; it also points out "that composers tend to feel more
free to interpret their own works in their own way" (ibid.).

32. Beethoven's use of tempo rubato in conducting was reported by various people; I dis-

cuss this in detail in this chapter's Postscript.

The Classical Style, pp. 144-45.
33. Rosen,
Solomon, "Beethoven: Beyond Classicism," in The Beethoven Quartet Companion, ed.
34.
Robert Winter and Robert Martin (Berkeley: University of California Press, 1994), pp. 59-75.
356 CLASSICAL AND ROMANTIC

Yes, that's essential. And only by setting the eighteenth-century scene does
one most forcefully realize his modernity. People reacted strongly at the time,

too, of course. I read a lot of the contemporary criticism, for instance, which
is very revealing, because it mentions some things we wouldn't have thought
of, and it doesn't talk of other things at all. For instance, regarding the Ninth
Symphony, contemporary critics tended to talk one minute about key change
and the next minute about something quite fantastical, fairies-marching-
through-the-undergrowth stuff. They tended which
to think in literary imagery,
today you're not "supposed" to do.^' Back then was absolutely normal.
it

Berlioz did it to Beethoven's Fifth. He clearly thought that was how you lis-
tened to music: pictorially. So that's one thing that surprises us what they did —
put in.

One of the things that affected contemporaries was the roughness of

Beethoven's music. Solomon describes his heroic works as incorporating such
elements as "death, destructiveness, anxiety, and aggression, as terrors to be
transcended within the work of art "^^
itself. It was more violent than what they
were used to.

It is, I think we sometimes underplay Mozart and Fiaydn, too. I

though
don't think Mozart should ever sound rough, but he goes very near the edge.
Haydn comes close, too (in the opening of The Creation, or the drinking cho-
rus in The Seasons). Clearly, Beethoven did go over the boundary; they must
have enjoyed going very near the edge, but Beethoven went over it. So, yes, I

agree, it must have been more violent than usual. And they did think it was
wild. Schubert at first didn't like a lot of Beethoven he thought it was too —
rough. Weber said of the Beethoven Seventh (that repeated bit at the end of the
first movement) that Beethoven was ripe for the madhouse. So clearly there was

a feeling that he was too violent for words. Beethoven was wild, in any case.
He might not have been quite so wild if he hadn't been deaf, I suppose, and
fighting it —sometimes the myth is right. Railing against fate — that's what you
do if you're a musician and you lose your hearing. You either go under or you
fight it.

35. Some of these writings can be found in Nicholas Cook's Cambridge Handbook
Beethoven: Symphony No. 9 (Cambridge University Press, 1993). For example, the staccato
runs in the woodwinds in the Scherzo of the Ninth are compared to "Columbine tripping with
her Harlequin, who springs in bold leaps from one modulation ... to another" (p. 32). As
Norrington says, we tend to smile at such descriptions today; but Scott Burnham's Beethoven
Hero (Princeton University Press, 1995) argues for the "kinship of the 'highly technical struc-
and the picturesque stories that critics in the nineteenth
tural analyses' of the twentieth century
century made up works of music." (I quote Charles Rosen, "Beethoven's Tri-
to account for
umph," New York Review of Books, 21 September 1995, p. 54; he calls Burnham's demon-
stration "brilliant," and also argues that technical analysis of music is "fundamentally
metaphorical" [p. 53].)

36. Solomon, Beethoven, p. 194.

 TAKING MUSIC OFF THE PEDESTAL 357

Perhaps that relates to what you were saying before about his character.

Railing is a robust response.''^

That's good! Yes, it was Beethoven who was the hero of his own "Eroica."
That's why he could afford to tear up the dedication!
Anyway, the things critics wrote about music at the time are very important
to me. We can't always understand them, of course, because we don't have their
second-nature thoughts about it. But at least it helps to break the mold of what
our teachers taught about it, which is often very different. And then, having

used the old information to change your viewpoints, in the end you have to go
on your own instincts. Mine are instinctive performances. The information is
just information; however much background you discover about a piece, that
doesn't tell you how to play it.

Someone praised your Beethoven performances for not feeling pedantic.

Thank God! And if that's really true, it's because although they were fifty
percent informed, they were also fifty percent instinctive, organic. That's very
important to me. In the end you're doing the music for now, absolutely for
now. People get a bit confused about that. They think that because we're being
historically informed, we're trying to be historical or "authentic." I never use
the word "authentic." There is no such thing. I'm trying to get away from per-

formances that have seemingly rather irrelevant gesture, and replace them with
gesture that does seem likely to suit the music. But in the end we're doing it
for now.

The extreme case, though, might be whether you would flat-out contradict
the historical evidence, if it didn 't feel as artistically right to you as doing some-
thing anachronistic?
I guess the right answer is "Yes." I have to believe in what I'm doing, and
believe it at an instinctive level. But as I said before, I try to train the instincts

so that they will create gestures entirely in keeping with the mode of a partic-
ular epoch, while keeping an eye out for the unique and the extraordinary in

a particular piece.

/ was saying that Toscanini and, nowadays, people like Charles Mackerras

and David Zinman have done the symphonies at the fast tempos on modern

37. But, perhaps, not ultimately the healthiest. Maynard Solomon's enlightening essay
"The Quest "Beethoven had become strong enough
for Faith" concludes that in his late period
to set aside the armor of heroic self-sufficiency which had to some degree impoverished his
middle years. He found a new ability to call for help, to pray, to give thanks, to reveal weak-
ness, and even provisionally to accept his dependence upon an immaterial and unknowable
deity" (in his Beethoven Essays [Cambridge, Mass.: Harvard University Press, 1988], p. 229;

first published in Beethoven-} ahrbuch 10 [Jahrgang 1978/1981], ed. Martin Staehelin [Bonn:
Beethovenhaus, 1983], pp. 101-19).
358 CLASSICAL AND ROMANTIC

instruments. What is your view on using historical instruments versus modern

ones?
I just like old instruments very, very much. They teach me such a lot about
the music. When a modern orchestra is playing superbly, with control over vi-

brato, with an awareness of phrasing, and beautifully together, I must admit

that I do sometimes wonder why I bother with old instruments, because the ef-

fect is so good. But in the end I don't think that early music is about instru-
ments; it's about the music. You could do a historically informed performance
of a Bach suite on a steel band. It would be historically informed in the sense

that it had the right speed, the right phrasing, the right feel. It wouldn't be a
sound that Bach had heard before, but it could be historically appropriate. And
modern instruments are a lot nearer than a steel band, of course, and are in

fact quite similar to the old ones. All I have to do is to try to make their play-
ing relevant to the music.
I inhabit both worlds. About half my time is spent with early instruments,
the other half with modern ones. That's perhaps a little unusual. But I have a
very pragmatic approach to the whole thing. I'm a musician, not a scholar. I

like to play early music with old instruments because it sounds so good, not
because it's politically correct! It isn't a question of its being more "moral"
somehow —that it's immoral to play with vibrato, for instance. I take a more
hedonistic viewpoint. The sound of the old instruments is so beautiful; that of
the new ones so rich and strong.

SELECTED DISCOGRAPHY
Richard Osborne praises Norrington's keen "feel for the everyday dress and
furnishings of period Beethoven," and adds that Norrington's Beethoven sym-
phony cycle, "for all its several oddities, has something of the evocative power
of a Balzac novel"'** —an interesting comment in view of the priorities Nor-
rington discusses in this interview. If you want to sample the set, you might try

its "first and finest entry," as James Oestreich calls it, the Second and Eighth
Symphonies (EMI 47698). Oestreich praises their elan, and adds that "Nor-
rington's zippy, electric performances largely vindicate Beethoven's long-dis-
puted metronome markings. "^^ The other most-often-praised excerpt from the
series includes the First and Sixth Symphonies (EMI 49746). The remaining
performances in the cycle have been more controversial —with strong detrac-
tors and supporters, and few neutral parties.

Indeed, Norrington's conducting never fails to challenge preconceptions; he

may or may not please you, but he will never bore you. In other repertory (all

on EMI), some of the more highly praised CDs are his Rossini Overtures (which

38. In a review of Gardiner's Beethoven symphony cycle, in Gramophone 72 (November

1994), pp. 65-66.
39. "Beethoven Ever After," The New York Times, 1 March 1996, p. C27.
 TAKING MUSIC OFF THE PEDESTAL 359

Osborne called "perhaps the cheekiest, most shocking, most uproarious, and in
some ways the most revelatory"""' recording of these works; EMI 54091 ); Weber
symphonies and Konzertstucke (55348); and his Purcell Fairy Queen (55234),
which is well cast and "richly stylish," according to Eric Van Tassel though he —
thinks it has a few "excessively fast" tempos.*" Norrington's Brahms Ein
deutsches Requiem (EMI 54658) has been called "stirring and moving,""^
though it has detractors, some of whom object to its baroque-influenced phras-
ing and articulation and to its unusually quick tempos. (Contrary to what some
admirers have written, most of these tempos are substantially faster than the
original score's metronome marks —a fact that could be taken as evidence
against the charge that Norrington follows such markings slavishly.)
As these examples illustrate, faster-than-usual tempos have been a contro-

versial aspect of Norrington's style. Several of them figure in his widely praised

Don Giovanni (54859), though it also has a couple of unusually slow tempos.
It declares its iconoclasm from the opening Andante: the meter is alia breve
(two beats per bar), which Norrington says meant to Mozart a tempo "twice
as fast" as the same tempo marking with four per bar. Regarding its use in this
overture he differs from Wye J. Allanbrook, who sees this passage as an alia

cappella — i.e. slow — form of alia breve."*^ I can't tell you who is right, but while
I still prefer the slower tempos I'm used to, Norrington does make his approach
work.

FOR FURTHER READING

In addition to the Barth and Stowell books discussed in the Postscript below,
William S. Newman's Beethoven on Beethoven (New York: Schirmer, 1988)
provides a good starting point for investigating performance practice in

Beethoven, as do the relevant parts of Sandra Rosenblum's Performance Prac-

tice in Classic Piano Music (Bloomington: Indiana University Press, 1988). Max
Rudolf's The Grammar of Conducting, 3rd ed. (New York: Schirmer, 1994),
devotes seven wise, informative pages to Beethoven issues (pp. 397-403). Some
important bar-by-bar guides for conductors of the symphonies include Norman
Del Mar, Conducting Beethoven (Oxford University Press, 1992), and Heinrich
Schenker, Beethoven's Ninth Symphony, trans, and ed. John Rothgeb (New
Haven, Conn.: Yale University Press, 1992); both are very much worth ex-
ploring. Del Mar comments on historical-performance issues —rarely with ap-
proval, but always with intelligence.
Regarding Beethoven's music, Donald Tovey's Essays in Musical Analysis
(Oxford University Press, 1935-39), Beethoven (Oxford University Press,

40. Gramophone 68 (April 1991), p. 1842.

41. Van Tassel, "Collection: Purcell on Record," Gramophone 73 (November 1995), p. 65.

42. Lionel Salter, Gramophone 70 (April 1993), p. 106.

43. Rhythmic Gesture, p. 198.
360 CLASSICAL AND ROMANTIC

1944), and other writings contain illuminating discussions. The best introduc-
tion to Beethoven's musical context may well be Charles Rosen's The Classical
Style (New York: Norton, 1972); it also has a brilliant, original discussion of
Beethoven's style. Regarding his sociocultural context, some good introductory
discussions are the first four essays in The Beethoven Quartet Companion, ed.
Robert Winter and Robert Martin (Berkeley: University of California Press,
1994).
Norrington's remarks about Beethoven's jovial side are clearly meant as a
corrective, not a summation of his complex personality. Maynard Solomon's
Beethoven (New York: Schirmer, 1977) is one of the most successful psycho-
logical biographies of any composer, even for those of us who don't share
Solomon's Freudian perspective. Scott Burnham's Beethoven Hero (Princeton
University Press, 1995) is an especially important discussion of the heroic style

in Beethoven, how it has influenced our image of him, and how it relates to

the trends of thought of Beethoven's era.

Postscript: "Classical" and "Romantic"

Performance Practice in Beethoven

Did Beethoven play "Romantically" or in a strict "Classical" style? George

Barth's The Pianist as Orator'*'* (already mentioned by Malcolm Bilson) is one
of several recent publications that put a surprising twist on this question. Barth
attempts to reconcile the conflicting testimony about Beethoven's "strictness"
and his "freedom from constraint" in tempo, and concludes that the opposing
witnesses actually describe the same style —but that this becomes clear only

when you read them from the standpoint of late-eighteenth-century German

writers like C. P. E. Bach and D. G. Tiirk. These writers preferred a declama-
tory "speaking" style, which treated rhythm in music rather as we treat rhythm
in speech: it could emphasize an important note not only by playing it louder
(which is how modern performers do it) but also by holding it a little longer.

Also as in speaking, this style tended to pass a little hastily over unimportant
short notes, a practice modern musicians avoid. This was the style that
Beethoven grew up with, and plenty of written discussions of it have survived.
But how might it have sounded? We can hear what this style sounded like, says
Barth, in the Beethoven recordings of Bela Bartok.
To most modern critics that assertion is as disorienting as the news that
polyunsaturated fat is worse for your heart than other types. We had assumed
that Bartok's recordings, with their flexible rhythm, were anachronistically Ro-
mantic, and that the true historical style is found in stricter, low-fat recordings
hke Norrington's. The Bartok example would not, however, surprise another

44. Ithaca, N.Y.: Cornell University Press, 1992.

 TAKING MUSIC OFF THE PEDESTAL 361

author, Robert Philip. In his essay "Traditional Habits of Perfornmance in

Beethoven Recordings,'"" he argues that "in fundamental ways, musicians from
early in our century were closer to the traditions of Beethoven's day than we
are now" — and in "we" he includes Norrington, Gardiner, and Co. as well as
their mainstream colleagues. Philip does not say that the styles of early record-

ings are identical to Beethoven's, but he does indicate that they are closer with
regard to such seemingly un-Classical "ingredients" (to use Bilson's term) as
flexibility of tempo, non-literalness of rhythm, and use of portamento.
These assertions gain support from other writers. Clive Brown tells us that
by 1811 "a number of Viennese string players were using portamento liberally,
not only in their solo playing but also in the orchestra";**" its popularity was
due to the prestige of the French "RodeA^iotti" school of violin playing, which
involved "a highly expressive cantabile often involving the use of prominent
"'^
portamento.
As for tempo flexibility, although we have a good deal of evidence about
Beethoven's and even his predecessors' practices,"** the evidence is conflicting.

But the overall picture suggests that, as Philip puts it, "the idea of a constant
tempo in extended movements is an invention of the late twentieth century."
I've already quoted Ignaz von Seyfried, conductor at the Theater-an-der-Wien
(where various Beethoven premieres took place), on Beethoven's concern with
"an effective tempo rubato" when rehearsing. Schindler quotes a 21-bar ex-
cerpt from the Second Symphony detailing the rubato that he says Beethoven
used when leading it. He shows the composer speeding up in crescendos and
slowing down for soft passages —a practice common in early recordings."'*

Though Schindler is known to have forged some of his "evidence," the gist of
this account is affirmed by Ignaz Moscheles, who also heard Beethoven con-

45. In Performing Beethoven, ed. Robin Stowell (Cambridge University Press, 1994), pp.
195-204.
46. Brown, "The Orchestra in Beethoven's Vienna," p. 18. In 1811 Salieri wrote a letter
strongly attacking this trend in orchestras; he wrote a similar letter four years later, which sug-
gests that the first letter had had no effect.

47. Clive Brown, "Ferdinand David's Edition of Beethoven," in Performing Beethoven, p.

119; see also pp. 130, 132-36 in the same essay. But David Watkin's "Beethoven and the
Cello," pp. 112-15 in the same volume, says that although the treatises by cellists connected
with Beethoven recommend using portamento, they suggest fairly moderate use; he argues that

"the influence of the Viotti school was strongly felt by cellists during Beethoven's lifetime, but
it was (Bernhard) Romberg who inspired a school of cello playing," and his school used por-

tamento "sparingly." Watkin also reports that the c. 1825 treatise by Dotzauer, a pupil and
follower of Romberg, says that portamento should be used less in orchestral playing than in
—
solo playing though Salieri's letters make one wonder how many players took this advice.
48. See Brown, "The Orchestra in Beethoven's Vienna," pp. 17-18; and Barth, "The Pi-
anist as Orator," pp. 53-86.
volume is often marked by acceleration, and falling volume
49. In old recordings, rising
by retardation. Will Crutchfield notes that there's something natural about this: "A crescendo
is an expression of heightened energy; so is an acceleration. Is it so bizarre that they should

coincide?" ("Brahms by Those Who Knew Him," Opus, August 1986, p. 21).
362 CLASSICAL AND ROMANTIC

duct.'° Even Carl Czerny, who insisted that students play Beethoven to a
metronome, writes of the cello sonatas that "there occurs in almost every line

some notes or passages, where a small and often imperceptible relaxation or

acceleration of the movement is necessary."'' He also gives rules for when one
should slow down which, when applied to Beethoven, sound Schindler-like.^^
Finally, Beethoven himself writes in the score of his song Nord oder Sud that
the metronome mark applies only to the first few bars; after that, "feeling has

its own tempo."

Beethoven's large movements include contrasting themes and textures, so
it is at least conceivable that a movement's different themes might work best
at somewhat different tempos. Robert Martin, for example, tells of a string
quartet finding that the very fast metronome mark for Op. 95's first move-
ment works perfectly for the opening theme, but makes the second theme
sound "terribly rushed and out of character."'^ The group eventually decided
to compromise with a "somewhat slower loverall] tempo," because "most
modern quartets feel uncomfortable about changing tempos markedly within
a movement. Nowadays it seems objectionably self-indulgent to change
. . .

tempos (except very subtly) to accommodate the second theme." But would
it have seemed that way to Beethoven? As Martin's wording implies (and as
Taruskin and Barth argue), perhaps this has more to do with modern taste
than with history. In teaching the piano, Beethoven "rarely said anything"
about wrong notes, but if a student "lacked expression [in] the character of
the piece he became angry" ;^'* what flexible tempos emphasize is the differ-
tempo emphasizes their
ent sections' differing characters, while an unchanging
underlying structural unity. Of course, Beethoven's large movements are
amazingly unified, and of course he worked hard to unify them; and of
course, flexible tempos can distort his music when exaggerated or mishan-
dled (I think of some of the Russian "new subjectivist" pianists). But the best

of the old (and, for that matter, new) recordings show that conveying unity
doesn't require as extreme a suppression of diversity as we sometimes hear
^^
today.
A final characterful "ingredient" of old recordings is their speech-like
rhythm and phrasing. Philip shows that Bartok was not unique; early in our
century musicians often accented a note by holding it longer and by hastening

50. Schindler's excerpt and Moscheles's comment can be found in Taruskin, "Beethoven:
The New Antiquity," pp. 256-59.
5L Watkin, "Beethoven and the Cello," p. 112. See also Barth, pp. 74-86.
52. Barth, pp. 85-86.
53. Martin, "The Quartets in Performance: A Player's Perspective," in The Beethoven
Quartet Companion, p. 120.
54. According to Beethoven's student Ferdinand Ries; see Beethoven Remembered: The Bi-

ographical Notes of Franz Wegler and Ferdinand Ries, trans. Frederick Noonan (Arlington,
Va.: Great Ocean Publishers, 1987), pp. 82-83.
55. See Barth, The Pianist as Orator, p. 64.
 TAKING MUSIC OFF THE PEDESTAL 363

less important short notes — as we've seen was done in Mozart's and Beetho-
ven's day.

In replying to these arguments, Norrington could note that six of the sym-
phonies pre-date the portamento vogue that Brown describes — though evidence
suggests that Beethoven was on to the new French violin style by 1800. Ad-
dressing another topic, Norrington might point to testimony, such as that of
Ferdinand Ries,'^'' about Beethoven's keeping a steady tempo, as well as to ev-
idence that Beethoven's playing grew freer as he aged (that is, after many of
the symphonies were written). He could also note Schindler's comment, "That
orchestral music does not admit of such frequent changes of tempo [as piano
music] is an understood fact," and could mention Charles Rosen's view that
"even the late orchestral works like the Ninth Symphony clearly imply a per-
formance with few of the individual refinements of tone, accent, and tempo of
the sonatas and quartets."^^ And Norrington stands on unshakable ground
when he says that Wagnerian tempo fluctuations were more extreme than
Beethovenian ones. Besides, to return to a leitmotif of mine, how much diver-
sity of tempo is needed is ultimately a matter of taste —and Norrington says
that his preference for a unified tempo is based on his feeling. Lord Menuhin
says, "the dramatic build-up of Beethoven's rhythms has the inevitability of a
tidal wave. ... As soon as you change the pace you lose that inevitability."'**

Note that, like Norrington 's "feeling," this is the artistic judgment of a first-

rank Beethoven interpreter, rather than an appeal to historical practice.

In general, though, Philip believes that "period performers have got away
with the unspoken assumption that late-twentieth-century neatness and clean-
ness are somehow 'authentic' in Beethoven." That assumption, he concludes,
can't hold much longer. And doubting it may bring musical rewards. My reser-
vations about some early-instrument Beethoven may reflect the performances'
relative lack of interest in the ingredients Philip lists. It is these matters that can
make a fast tempo breathe and feel natural. They can also heighten expression,
by letting a great moment register or an important modulation tell. Some his-

torical performers seem to agree; many others may disagree. For myself, I be-
lieve that Barth and Philip et al. give historical performers valuable new infor-
mation to work with.

56. "[Beethoven] usually kept a very steady rhythm and only occasionally, indeed, very
rarely, speeded up the tempo somewhat. At times he restrained the tempo in his crescendo
with a ritardando, which had a beautiful and most striking effect" (translation from Barth, p.
54). Ries studied piano with Beethoven from 1801 to 1805.
57. Rosen, The Classical Style, p. 144.
58. Brian Hunt, "A New Challenge," Gramophone 73 (April 1996), p. 22.
 19
Reviving Idiosyncrasies
— oOo^
gN

John Eliot Gardiner on

Berlioz and Brahms

Since the 1960s, John Ehot Gardiner has spent much of his career conducting
two groups he founded in his native England, the English Baroque Soloists and
the Monteverdi Chorus. It has become obvious, though, that to categorize him
as an "English choral conductor" and "early-music specialist" w^ould be unfair.

Not only was he conducting opera at Sadler's Wells in 1969; he has performed
such composers as Bizet and Chabrier with the opera of Lyon (whose orches-
tra he founded), and he has more recently turned his attention to even later
music.From 1991 to 1994 he was Principal Conductor of the North German
Radio (NDR) Symphony Orchestra, with whom he recorded Rachmaninoff,
Mahler, Weill, and Britten, and he has recorded Lehar's Merry Widow with the
Vienna Philharmonic.
When I approached Gardiner about this book, he was at Tanglewood re-

hearsing the Boston Symphony Orchestra in Berlioz. Our topic was the use of
early instruments in Romantic music. Some observers have dismissed experi-
ments in this field as simple empire building by the historical-performance
movement. Yet Berlioz himself wrote that "at no period in the history of music
has there been greater mention made of instrumentation" than in his own.
These words appear in his treatise on orchestration —apparently only the sec-

ond (by just a few years) ever written on that subject. And Carl Dahlhaus notes
that "this 'emancipation of tone color,' initiated by Berlioz, freed tone color
from its subservient function of merely clarifying the melody, rhythm, harmony,
and counterpoint of a piece, and gave it an aesthetic raison d'etre and signifi-

364
 REVIVING IDIOSYNCRASIES 365

cance of its own."' If that is so, then a case could be made that period instru-
ments might be more significant in Berhoz or Liszt than in most earHer music.-
As we'll see, though, the case does not convince Gardiner.
It may seem odd that a book on "early music" should include Berlioz; yet
Berlioz might be considered a patriarch of period instruments. As Hugh Mac-
donald writes, Berlioz was among the first to put forth the view that when we
perform music of the past it "should be enshrined in its own period and not
brought up to date."' For example, Berlioz decried the "ravagers of works of
art""* who altered the orchestration of earlier composers. True, in reviving a
Gluck opera in 1859, he discreetly retouched the orchestration — but only, he
said, "solely in order to render it precisely as it was composed by Cluck."'' This
concern with the composer's instrumentation can be said to have led eventu-
ally to the use of period instruments in our century.
Brahms and Berlioz were in some ways polar opposites — Brahms the ex-
emplar of "absolute" music, Berlioz of literature-based music; Brahms, unjustly
derided for "gray" orchestration, Berlioz the unquestioned master of color. Yet
both were children of an era whose historical awareness was unprecedented.
According to Peter Burkholder, Brahms was the first composer to be "obsessed
with the musical past" and with his place in musical history; in our century,
such obsession became the norm." Brahms's obsession with the musical past
meant that he knew far more about early music than most of his predecessors.

He studied Schiitz, Bach, Palestrina, and many others in depth; he considered

their works to represent a lost golden age, and sometimes wrote in their long-

outmoded forms himself. The related feeling of nostalgia is sometimes under-

lined in modern performances of Brahms —
perhaps because (as I discussed in
this book's introduction) we share that feeling. Yet Gardiner and others who
are experimenting with historically informed Brahms performance often de-em-
phasize the nostalgic quality in favor of vigor; it's another example, perhaps,
of historical information being used to make old music sound new. Whether it

benefits the music is, of course, a matter of debate, on which Gardiner ex-
presses clear opinions.

1. Carl Dahlhaus, Nineteenth-Century Music, trans. J. Bradford Robinson (Berkeley: Uni-

versity of California Press, 1989), p. 243.
2. See forexample the review by Julian Rushton, Early Music 17 (November 1989), p. 623.
3.Macdonald, The New Grove Dictionary, vol. 2 (London: Macmillan, 1980), p. 601.
4. In Berlioz's "Instruments Added by Modern Composers to Scores of Old Masters," from

his A travers chant (Paris, 1862). A modern translation is The Art of Music and Other Es-

says, trans, and ed. Elizabeth Csicsery-Ronay (Bloomington: Indiana University Press, 1994);
the essay is on pp. 148—49. See also Jose A. Bowen, "Mendelssohn, Berlioz, and Wagner as
Conductors: The Origins of the Ideal of 'Fidelity to the Composer,'" Performance Practice Re-
view 6 (Spring 1993), pp. 77-88.
5. See Joel-Marie Fauquet, "Berlioz's Version of Cluck's Orphee," in Berlioz Studies, ed.
Peter Bloom (Cambridge University Press, 1992), pp. 189-253.
6. Burkholder, "Brahms and Twentieth-Century Classical Music," 19th-century Music 8/9
(1984), pp. 75-83.
366 CLASSICAL AND ROMANTIC

We began by discussing a trend that's affecting orchestras around the world:

standardization. Many orchestras are losing some of their region's traditional

orchestral sound and style. One factor behind this is jet-age maestros who, says
Nicholas McGegan, spend more time in the first-class lounges of airports than
in front of orchestras; another is the international distribution of recordings.
Yet technology is only part of the equation; attitude, as I think emerges in our
talk, is another. I hope Gardiner's attitude doesn't prove quixotic.

You recently founded the Orchestre Revolutionnaire et Romantique (ORR), re-

cruiting freelance period-instrument players in Europe and the USA, to focus
on nineteenth-century music. The skeptic might ask why it's needed, when this
music has been part of the repertoire since it was written.
I feel uncomfortable with the implication that one has to look to the out-

side to find a gap that requires filling, rather than respond to one's inner cre-
ative impulses. Although I enjoy working on nineteenth-century repertoire with
modern symphony orchestras (with whom perfectly valid and convincing in-

terpretations are of course possible), there is an added dimension to the work

we do in the ORR: it feels far more like a voyage of discovery than an act of
regurgitation.The coming into existence of such an orchestra is an antidote to
the monochrome sound quality that is fast becoming the norm nowadays in in-

ternational orchestral playing. It seems to me that this is due in the first place

to the phenomenon of a small posse of "star" conductors going around to a

small number of orchestras in different countries and creating a somewhat stan-
dardized sound. That's an exaggeration, of course, but there is some truth to
it; and it is a phenomenon reinforced by the major recording companies, whose

producers have been known to insist on replacement players for, say, wood-
wind soloists with pronounced national or regional timbres which do not con-
form to the accepted norm. This standardization process has tended to blur the
local, regional, and national characteristics of orchestras, and arrest the evolu-

tion of idiosyncratic performing styles.

For example, if you listen to recordings of Russian orchestras between the
world wars or even in the post-war years, but certainly way before perestroika
and glasnost, you get a very strong sense of their location and provenance, as

well as of the individual flavor of their music-making. When, for example, the

Leningrad Philharmonic with Mravinsky performed Tchaikovsky symphonies,

the style, the sound spectrum, and the emotional range were arresting and dis-
tinctive — utterly different from the same symphonies as played by, say, Kara an j

and the Berlin Philharmonic.

Today, even the Czech Philharmonic and the Dresden Staatskapelle, two of
the great Eastern European orchestras, seem to me to be under threat: their in-

tegrity and ability to continue to evolve and develop along their own lines are

in jeopardy because of insidious pressures to make them sound like all the other
 REVIVING IDIOSYNCRASIES 367

orchestras. But it's not a totally bleak picture, because you do of course get in-

dividual sounds still — the Vienna Philharmonic is immediately recognizable, I

would say, and particularly in their own special repertoire of the late nineteenth
century they are an inimitable living embodiment of a forgotten style of play-
ing. I recently recorded Lehar's Lustige Witwe with them and their playing was
utterly magical. It was also obvious to me that they are the period-instrument
orchestra for that repertoire!

/ recall reading that the Czech Philharmonic players themselves, after hear-
ing recordings by the Berlin or the Chicago, wondered whether international
listeners would be able to accept their style. So they felt pressured to modify it
when they recorded. The pressures toward standardization might be said to re-
sult largely from technology —
the jet planes that the conductors commute on
and the hi-fi systems the musicians listen on.
It's also partly to do with the instruments themselves. A lot of instruments

have national or regional characteristics, like the Viennese oboes and horns, or
the French woodwind instruments, in particular their bassoons and cornets.
The tendency now is towards a Germanization and, even more, an American-
ization of instruments. And I feel it's to be deplored, because we're not getting
the diversity we had. Modern French orchestras, for example, have generally
become rather Germanized, but that is perhaps more to do with the influence
of their principal conductors (predominantly non-French) and the repertoire
these conductors tend to favor.

A of the sounds in your recording of the Symphonic fantastique with the

lot

ORR remind one of the sounds that French orchestras produced until a gen-
eration ago —those orchestras' small tubas sounded more like your ophicleides
than large-bore modern tubas do; the bassoons and trombones were more nar-
row-bored, like yours; and so on. That fits with part of the justification you
gave for the ORR — to you put it, an antidote to the standardizing trend.
be, as

After all, early-music experiments often influence the mainstream: Beethoven

playing, for example, has been changing in the last five years or so.
It has. Period-instrument Beethoven has lost a layer of self-consciousness

and gained in technical fluency to the point where I can legitimately hope that,

say, our recording of the complete symphonies will be judged not in a sectar-

ian category ("the newest period-instrument recording") but on its own musi-
cal terms. As to whether the mainstream orchestras have been influenced by
period-instrument orchestras in this repertoire, I have my doubts. There is still

a lot of resistance, dismissive jibing, and ignorance. Yet the sort of cross-fertil-

ization that Harnoncourt, for one, has been able to achieve with a brilliant

modern-instruments orchestra like the Chamber Orchestra of Europe is an ex-

ample to mainstream orchestras of what can be achieved given the required will

and flexibility by the management and players alike. And it seems to me there
368 CLASSICAL AND ROMANTIC

are hopeful signs around. It wasn't all that long ago that I conducted the Phil-

harmonia Orchestra in London in an all-French program, and the principal bas-

soon asked me
minded if he and his colleagues played French bassoons in
if I

La mer. They'd never done so before in public, but they'd been working away
at these instruments privately and wanted to try them out. I think it's very en-
couraging when standard-instrument orchestral players take that sort of initia-

tive. To be able to choose according to the repertoire between French and Ger-
man —
bassoons systems
totally different with different fingerings and
embouchures — seems admirable to me. This is one of the most hopeful signs,

which might help to guarantee the survival of the standard symphony orches-
tra into the next century.

But yes, certainly, one of the raisons d'etre of the ORR is to combat the

tendency to make the whole repertoire sound the same. This is in- again an
sidious tendency, to homogenize the playing of music from Beethoven (or ear-
lier) to Richard Strauss in an all-purpose style —which is, for the most part,

the Zeitgeist of the 1920s and 30s, frozen in time but suited to the increas-
ingly clinical tastes of a technically obsessed age, for which gramophone record-
ings bear a heavy responsibility. And I think that one of the reasons for a
certain public disillusionment with symphonic music could be the rather rou-
tine, homogenized way of trotting out the same old pieces in the same old
style. That —together with some fairly unadventurous programming and an
old-fashioned formality of presentation —has made the concert hall somewhat
predictable and dull.

No doubt most conductors sense that there is an obligation to present a

work —of whatever period or style —as though for the first time, with a sense

of re-creativity, if not creativity. And in that context it helps enormously to try

to re-create and rediscover the idiosyncrasies and individuality of isolated and

individual composers, and of national styles evolving in a particular country.
That, in turn, up with the function for which the music was composed
is tied

and the kind of social context within which it was written. It leads you to ex-
amine not only the original source material, but the crucial matter of instru-
—
mentation the types of instruments, the style they were played in, in what pro-
portions, in what layout, and so on.

In this context, let me ask you more about Berlioz, whom you've recorded
with the ORR and the Opera de Lyon. He was, obviously, fascinated with the
orchestra.
It seems to me that Berlioz was a key figure, because if anyone was cen-
trally placed to influence the way the nineteenth-century orchestra developed,

it —
was he by his enthusiasm, imagination, and technical mastery although —
he was often mocked and despised (but not ignored) by his fellow composers.
Later on, of course, Mahler was a seminal figure. But both of them in their
 REVIVING IDIOSYNCRASIES 369

different ways (and not always different ways, obviously) celebrated the vir-

tuosity of the newly evolving orchestra, the kaleidoscopic colors that could
be obtained from the orchestra, and above all its expressive potential. Take
Romeo et Juliette, for example. It's significant that Berlioz decided that he

had no need to use singers for the character of either Romeo or Juliet. The
orchestra on its own was quite capable of expressing their separate charac-
ters and emotions. Berlioz actually says at one point that he relished the ex-
pressive potential of a symphony orchestra more than any composer before
him. I'm sure that's the case. He was using instruments that had not changed
much in form and sonority and shape since the Baroque, alongside the newest
instruments like the cornet a piston — and he was exploiting the tension that
exists whenever you juxtapose the new and the old. His was a very transi-
tional period: for example, string instruments were being converted by violin

makers like Vuillaume to be more powerful and to project into larger halls,

and therefore they needed to have their necks reset at a steeper angle to the

table and their bass bars made

and so on. So in experimenting with
thicker,

and was
against traditional usage, Berliozevery bit an avant-garde composer.
He took a great interest in all these developments and made sure that he was
au courant with the technical difficulties and capacities of all the orchestral
instruments. From his contact with people like Baillot and Paganini he learnt
the most up-to-date techniques and sonorities of modern violin playing, which
he then required his orchestral players to emulate and he gave those play- —
ers far more specific instructions as to what he expected of them technically

than any previous composer.

And then he brought much more fantasy and imagination to orchestral writ-
ing than anybody hitherto, including Beethoven: he seemed to rejoice in the dis-

tinct personalities of musical instruments — there's so much more dialogue and

dialectic within the Berlioz orchestra than in any of his contemporaries or his

predecessors that the instruments seem almost to be real people conversing, so

that with Berlioz's orchestra you're dealing with a far bigger organism put to
far greater expressive ends than ever before.

In the example you refer to, he has the tenor instruments represent the voice
of Romeo (cellos and violas, and horns in the tomb scene) and the violins and
high winds represent the voice of Juliet.
And remember how Berlioz justifies his treatment of the love duet in Romeo
et Juliette: "If the duets of love and despair are given to the orchestra, the rea-

sons are numerous and easy to comprehend. First, and this alone would be suf-
ficient, it is a symphony and not an opera. Second, since duets of this nature

7. The observation is from D. Kern Holoman's Berlioz (Cambridge, Mass.: Harvard Uni-
versity Press, 1989), which has a valuable analysis of Romeo et Juliette, especially the love
scene, on pp. 262-66.
370 CLASSICAL AND ROMANTIC

have been handled vocally a thousand times by the greatest masters, it was wise

as well as unusual to attempt another means of expression." And the follow-

ing observation is a crucial one: "It is also because the very sublimity of this
love made its depiction so dangerous to the musician that he had to give him-
self the imaginative latitude that the positive sense of the sung word would not
have given him, resorting instead to instrumental language, which is richer,

more varied, less precise, and by its very indefiniteness incomparably more
powerful in such a case."**

This is a tremendous manifesto for the powers of the nineteenth-century

orchestra. It's something that an opera composer like Verdi wouldn't have
agreed with —nor would many of his contemporaries but — Berlioz certainly
pushed the expressive powers of the orchestra far beyond the limits set by
his great hero, Beethoven. And I would have thought it a breakthrough that

Mahler would have admired. By the way, both Berlioz and Mahler wrote
wonderfully for voices — it's not that being "pro-orchestra" they were "anti-
voice."
So people mightsay. Fine, why can't a modern orchestra play them?^ Well,

of course, amodern orchestra can play them brilliantly; but the question is,
do we get the full force of the originality of these compositions when played
on standard modern instruments? There's so much to be gained and to be
learnt from trying to reconstruct Berlioz's orchestra. When we recorded the
Symphonie fantastique with the ORR in the hall of the old Conservatoire in
Paris, the site of its premiere, along with the thrill of assembling the instru-
ments known to or specified by Berlioz came the matter of their deployment.
We used the layout proposed by Frangois-Antoine Habeneck, the first con-
ductor of the Societe des Concerts du Conservatoire, who conducted the pre-
miere in 1830: a steeply raked series of steps for the orchestra, and a most
unusual layout with (for example) the harps out in front, behind the con-
ductor, and the double basses and cellos paired off or arranged in threes and
fours on a You get a totally different kind of sonority dryer,
huge staircase. —
more intimate, and far more vivid. '° Well, again, people can say, that's pos-
sible to do with a modern orchestra. Yes, perhaps, up to a point; but it is

surprising how conventionally and unadventurously music is often performed

in concert halls today.

8. From the Preface originally printed in the libretto for the 1839 performance.
9. Gardiner wrote about this question in the ORR's privately printed "Manifesto for a
New (Old) Orchestra": "I am not implying that two or three generations of great interpreters
have 'got wrong' up to now. That would be not merely impertinent but to deny the valid
it

creative tension which undoubtedly exists between the conception of a musical work and its
realization; in other words the way a composition can survive history and not merely toler-
ate but be enrichedby changes of instruments and styles of performance."
wrote at length about orchestral placement in his 1843/1855 Traite d' instru-
10. Berlioz
mentation, pp. 293-94 and 310. See Holoman's discussion in Berlioz, pp. 354-56.
 REVIVING IDIOSYNCRASIES 371

Various critics have been calling for years for a return to one of the things
you did, putting the first and second violins on opposite sides of the stage —
as most orchestras did until the Second World War —as an aid to transparency.
Right. That almost goes without saying for any music from eighteenth-
century concerti grossi onwards, where the vioUns are used antiphonally; and
several other arrangements are possible, too. I recently conducted Mahler's
Fourth on tour with the NDR Symphony Orchestra and we split not only the
violins left and right, but the cellos and basses as well, with just the violas
in the middle. For the slow movement, the individual strands of the string
lines were more beautifully separated and over a wider spatial spectrum than
I had ever heard before. That is just one example of what can be done with
a modern orchestra setup (though in this particular case I had no first-hand
evidence that Mahler himself deployed his strings this way; it just seemed nat-
ural!). But with a period-instrument band, in composers as diverse as Schu-
mann and Brahms and, above all, Berlioz, there is the potential for even
greater contrasts as a result of the instruments themselves. To return to Berlioz
as an example, there's the side-by-side juxtaposition of natural trumpets, which
had not evolved very much really since Monteverdi's or Bach's day, with brand-
new cornets a piston. Then you have dinosaur-like period instruments like the
serpent and the ophicleide, which Berlioz used in his early scores to give a
specifically raucous sound.

Julian Rushton writes that in the Dies irae episode in the Fantastique when
modern bass tubas replace the ophicleides it "turns a harsh parody into a
Falstaffian romp."^^ He also complains that modern bassoons sound too gen-
tlemanly and damp the work's fire somewhat.
Exactly. Overall, with period instruments in this repertoire you get a more
vivid and sharply defined palette of colors beyond that of
sym- a standard
phony orchestra, where modern cultivated sounds tend to assimilate and merge
rather than to retain distinctively separate strands of instrumental lines.

The example of the ophicleide might answer another objection from the
skeptic: vivid, specific, unstandardized tone colors may be fun, but are they
important musically? In a lot of earlier music, after all, tone color is clearly
not central to the conception —you can play the Fourth Brandenburg with a
harpsichord instead of a violin soloist (and in fact most of Bach is tran-
scribable), and of Baroque chamber pieces can be played on a range of
lots

different instruments. But Berlioz comes closer to using a specific tone color
as a central element. The ophicleides in the Dies irae section give an effect
that's part of the expression, and even a very similar instrument like the tuba

11. Rushton, The Musical Language of Berlioz (Cambridge University Press, 1983), p. 89.
Berlioz did sanction the substitution of a tuba later in life, but Gardiner, who calls this

tuba/bassoon pairing "ill-balanced," makes a strong musical case for the original.
372 CLASSICAL AND ROMANTIC

just can't re-create it.^^ So using original instruments might be far more sig-

nificant in Berlioz than in Bach.

Not quite. Both benefit from removing the film of later anachronistic in-
strumental tissue. But Bach's music is incredibly robust and seems to withstand
all manner of treatment: look how some of Glenn
incredibly illuminating
Gould's recordings have been. Bach even emerges recognizably on the Moog
synthesizer.'^

Another composer you've begun to explore with the ORR is Brahms, who
is usually thought of as hardly needing period performance practice.
It seems to me
more than any other composer from that pe-
that Brahms,
riod, has suffered from being interpreted
if not in a Wagnerian way, then by

applying the Wagnerian ideal of the endless, long-line melody, and the weighty
sound and texture of the Wagner orchestra.

Malcolm Bilson gives an example of the Wagnerizing of Brahms in one of

the late piano works, the Intermezzo Op. 119, No. 1. As he says, everyone now
overrides the two-note slurs — "sighs," as in Mozart^'* —and connects them into
a long Wagnerian line. Apparently Brahms protested against this trend as it

emerged in his own day.

12. Charles Rosen discusses

The Romantic Generation (Cambridge, Mass.: Harvard
this in

1, where he says, "orchestral color is not one of

University Press, 1995), especially in chap.
the fundamentals of form before the Romantic generation; tone color was applied like a ve-
neer to form, but did not create or shape it ... It was the Romantic generation that intro-
duced it directly into the initial stages of strict composition they altered the relationship
. . .

between the delight in sound and the delight in structure" (pp. 39-40).
As to how this applies to Berlioz, Leonard Ratner, in Romantic Music (New York:
Schirmer, 1992), contrasts Liszt and Berlioz: "Berlioz's point of departure is the
song. . . . around this firm basis of syntax, Berlioz's orchestra plays with sound to loosen
some of the edges, to digress, to displace momentarily, to isolate figures. . . . For Berlioz, the
point of reference is syntax, colored by sound; for Liszt, it is sound, put into order by syn-
tax" (pp. 216-17). Julian Rushton argues that "with the exception of a few outstandingly
coloured passages, it may be safely said that the substance of Berlioz's music is separable from
its instrumentation" (The Musical Language, p. 74). Holoman has shown that orchestration
didn't take place until late in Berlioz's compositional process. Thus I may overstate the im-
portance of period instruments in Berlioz's music. After all, as Gardiner himself says, Berlioz
works well with modern instruments. But the examples discussed in the interview show that
period instruments do make a big difference in at least some parts of Berlioz's music.
13. For further discussion of this, see the interview in Chapter 9 with John Butt. More
qualifications may be added to the previous note: Howard Mayer Brown, for one, has noted
that some Baroque music —
for example, much of the French Baroque depends more than —
Bach does on "nuances of sonority." See his essay "Pedantry or Liberation?" in Authenticity
and Early Music, ed. Nicholas Kenyon (Oxford University Press, 1988), pp. 29-30. And
Rosen, on p. 39 of The Romantic Generation, qualifies the remark quoted above by saying
that "Delight in instrumental color for its own sake is not new with Liszt." He finds a few
exceptional examples in which sound is of primary importance in pre-Romantic music, but
the examples mainly involve music that aims "at an illusion of improvisation," such as organ
toccatas and unmeasured preludes.
14. See the discussion in Bilson's interview.
 REVIVING IDIOSYNCRASIES 373

The same is true of his orchestral music. Take the second movement of the
First Symphony: how often, if ever, do you hear the slurred dotted rhythms of

the first-violin melody Ibars 7 and 8| played with the appropriate inflection?
Again, these are "sighs," and it is difficult for sighs to penetrate a caramelized
legato perpetuo:^"^

Andante sostenuto
Vn

The same is true of the great oboe melody in the slow movement of the Vio-
lin Concerto: are the devices of "circular breathing" which modern oboists
adopt really appropriate or necessary to bring out the pathos of that heavenly
inflected melody?
And as for Mahler and Wagner, I suspect that both are more sinned against
than sinning, in the sense that they had such a strong influence on the way that

the orchestra developed and the way conductors conducted that it spilled over

into Brahms interpretation. I am

Brahms should be de-Wagnerized.
sure that
And the effect of the instruments in my recording of the Requiem is, I think,
at least a step in the right direction, towards something that is not only more

transparent but more distinctive in its sound world even if the string sound is —
as yet too lightweight.
The orchestral music of Brahms lends itself to a weightier sonority than he
himself probably envisaged. It is more gratifying for musicians to play him in a

grandiose and syrupy way; but it seems to me that often this is wide of the mark.
To my mind, his music calls for a particular depth and breadth of sonority, but
not weighty or syrupy in the modern fashion, and a "breathing" sound capable
of subtle inflection, ebb and flow. It is a sonority quite different from, say, the

taut, almost abrasive sounds that Beethoven seems to have required of his play-
ers at full stretch or, again, from the kaleidoscopic switches of color that Berlioz
elicits from his orchestra. It takes a lot of skill on the part of the individual
player to control the broader, deeper sound required by Brahms without its ei-

ther becoming mannered, in a crypto-Baroque way, or else having all those won-
derful subtleties of inflection and cross-rhythm ironed out. Like all techniques,
it has to be worked at, once the appropriate sound concept has been deflned
and transmitted to the players in such a way that it fires their imaginations. I

feel we are only at the very beginning of our process of reappraisal with regard
to the orchestral music of Brahms; but the potential rewards here are immense.

15. Brahms, Symphony No. 1, second movement, opening.

374 CLASSICAL AND ROMANTIC

There is plenty of evidence of a sparing use of vibrato and of a more gen-

erous use of portamento in the performance styles of both Berlioz and
Brahms —though in different ways.^^ I would imagine that the ORR would
have little difficulty with limiting vibrato, coming as they do from Baroque
music, but great difficulty with portamento.
Actually, both practices are very difficult to introduce for purely expressive
ends, rather than merely to fulfill instructions, and are therefore hard to achieve
without self-consciousness.
Regarding vibrato, remember that in each generation there are those w^ho
advocate more vibrato and those who prefer less. It's a matter of taste. Along
these lines, some critics assume wrongly that the ORR and the English Baroque
Soloists don't use vibrato. They do, at varying rates and places, and if it does-
n't obtrude to the extent that it detaches itself from the expression then we've
achieved our goal.
The whole challenge consists in precisely this: finding the perfect meeting-
point of heart and mind, instinct and knowledge. But we should beware of in-
stinct as a bottleable commodity. It changes with habit, usage, and redefinition
of stylistic parameters.

SELECTED DISCOGRAPHY
Gardiner considers the juxtaposing of old and new instruments crucial to
Berlioz's conception of the Symphonic fantastique (Philips 434 402); Peter J.

Rabinowitz observes that the instruments that sounded "old" to Berlioz's au-

diences sound "new" to us, and vice versa. ^^ Still, Rabinowitz thinks that
"many of Berlioz's local linstrumental] effects come across exceptionally well"
with Gardiner's period instruments —though he believes this stems "more from
Gardiner's interpretive discernment" than from the instruments. Rabinowitz
praises Gardiner's responsiveness to detail and says that "many of the night-
mare qualities emerge clearly," though he feels there is a little too much "clear-

headed control." K. Robert Schwartz has no such reservations, and praises Gar-
diner for realizing "the full emotional range of the symphony."^^

16. According to Robin Stowell, Berlioz's contemporaries Habeneck and Baillot both pre-
scribed tasteful portamento "especially in slow movements and sustained melodies when a pas-
sage ascends or descends by step." They advised that the ascending stepwise portamento be
accompanied by crescendo, and descending portamento by diminuendo. (In Stowell, "Strings,"
Performance Practice: Music after 1600, ed. H. M. Brown and S. Sadie [London: Macmillan,
1989, and New York: Norton, 1990], p. 399). A more liberal use of portamento in Brahms
is demonstrated in the recordings by his close collaborator Joseph Joachim and by other evi-

dence. See Clive Brown, "Bowing Styles, Vibrato, and Portamento in Nineteenth-Century Vi-
olin Playing," Journal of the Royal Music Association 113 (1988), p. 117. Orchestral vibrato
appears to have been an infrequently used special effect before World War I: see Robert Philip,
Early Recordings and Musical Style (Cambridge University Press, 1992), chaps. 4 and 5.
17. Rabinowitz, Fanfare 17 (November/December 1993), pp. 185-86.
18. Schwartz, Classical Pulse, April/May 1994, p. 17.
 REVIVING IDIOSYNCRASIES 375

The Symphome is the most famihar of BerHoz's works, but Gardiner/ORR's

second BerHoz release is work the newly recov-
of the least famihar Berhoz —
ered Petite Messe Solennelle. Gardiner directed the modern premiere in Lon-
don. That performance, recorded by Philips (442 137), reveals the work to be
far more than juvenilia. Gardiner/ORR's third Berlioz CD, Harold in Italy (446

676), may be their best so far; Edward Greenfield, who finds it thrilling, praises
'*'
its "white heat" of intensity, attained, he says, without excessive speed.
As Gardiner notes, modern instruments work well in Berlioz — as is shown
by his recordings with the orchestra of the Opera de Lyon in pieces that reflect

a gentler side of the composer. The Berlioz scholar David Cairns concludes a
survey of recordings of L'enfance du Christ by calling Gardiner's "near to being
the answer to one's prayers"^" (though he dislikes a few tempos, such as the
very slow one chosen for the Shepherd's Chorus;'' Erato 45275, 2 CDs). The
Gardiner/Lyon Nuits d'ete (Erato 45517), in the original keys and with the
voice types specified, is outstanding.
Critics have been divided about the Gardiner/ORR Beethoven symphonies
(complete, Archiv 439 900-2AH5; also available singly). Erik Tarloff calls the
set "to my ears, the one to own,"^^ but Barrymore Laurence Scherer says the
performances "reveal Beethoven's architecture, his energy, his muscularity, but
they offer little of Beethoven's heart."'' Even Scherer likes the Seventh; Richard
Osborne calls it "glorious."^'* Osborne begins by praising the introduction's
"ideal blend of weight and anticipation," and has similar praise for the re-
mainder of the work. I can't recall a more nuanced performance of the Sixth's
"Scene by the Brook": listen, for example, to the perfectly judged rubato when
the bassoon makes the transition into the second subject in bar 32 (2:36-2:41
in the CD track). A little more of such nuance might have benefited the rest of
the "Pastoral" ("brisk efficiency," says Osborne), the "Eroica" (according to
Alex Ross, the performance mutes the "frenzied dissonances and funereal de-
spair"^^), the Fifth (though Ross considers this "the best in the set ... a
shock"), and the Ninth. In the first movements of Nos. 3, 5, and 9, Gardiner
builds up terrific climaxes; but in quieter passages, some people find Gardiner
not always as responsive.
In the ORR
Brahms Ein deutsches Requiem (Philips 432 140), Gardiner
avoids what he calls the "lugubrious and pompous" feeling of many perfor-
mances. His performance is, Lionel Salter writes, "notable for its intensity and

19. Greenfield, Gramophone 74 (August 1996), p. 44.

20. In Choral Music on Record, ed. Alan Blyth (Cambridge University Press, 1991), p. 185.
21. It is taken at little more than half of Berlioz's metronome mark.
22. Tarloff, "Beethoven on Original Instruments," Slate (www.slate.com), 1 October 1996.
23. Scherer, "Letting Beethoven Be Beethoven," The Wall Street journal, 21 February
1995, A15.
p.
24. Osborne, Gramophone 72 (November 1994), p. 66.
25. Ross, "Crossed Paths on the Rocky Road to 'Authenticity,'" The New York Times, 9
April 1995, Arts and Leisure Section, p. 44.
376 CLASSICAL AND ROMANTIC

fervour (not mere ferocity). "^^ Virginia Hancock, whose research has clarified
our understanding of what Brahms learned from early music, praises the
recording with many detailed observations.^^ Her criticisms are just as inter-
esting: some of them are of non legato that is "presumably intended to avoid
'Wagnerian sostenuto' in the choral parts, [but] sometimes create[s] an impres-
sion of over-fussy articulation." Hancock would also like a larger chorus
though the singing throughout strikes her as superb —and sometimes finds the
string section too small to balance the winds. Although it was a first attempt
atBrahms, what John Steane says of the sixth movement applies to the whole
"one is caught up in the power of it."^^
Most critics have agreed that Gardiner's Merry Widow (with the Vienna
Philharmonic; Deutsche Grammophon 439 911, 2 CDs) is "utterly magical": a
typical comment is that of Scherer, who says it "comes as close to perfection
"^"^
as I can imagine.

FOR FURTHER READING

Donald Tovey thought that Berlioz was a better writer than composer; he un-
derestimated the music, but not the prose, which D. Kern Holoman praises for
its "fetching combination of insight and wit." Berlioz's Memoirs are a literary
classic, demonstrating that of all the great composers, he would probably have
been the most entertaining dinner companion; David Cairns's translation is

published by Gollancz (London, 1969). The memoirs are no more trustworthy

than the average autobiography, of course; you might supplement them with
Holoman's excellent Berlioz (Cambridge, Mass.: Harvard University Press,
1989).
As Tovey's comment shows, Berlioz's musical style has sparked controversy
since its own time, mainly because of his idiosyncratic treatment of harmony
and counterpoint. Today his music has won general esteem, thanks in part to
some perceptive analyses. The most impressive is Julian Rushton's The Musical
Language of Berlioz (Cambridge University Press, 1983). Holoman's biography
also sheds a great deal of light on the music.
An Brahms performance is Will Crutch-
excellent discussion of historical
field's article "Brahms by Those Who Knew
Him," in the late, lamented mag-
azine Opus (August 1986, pp. 12-21 and 60). I discuss issues of early-music
Brahms performance in Early Music America, Spring 1997 and Early Music,
August, 1997. Robert Philip's book Early Recordings and Musical Style, to
which I've referred so often, is again an invaluable resource (Cambridge Uni-

26. Salter, Gramophone 68 (April 1991), p. 1881.

27. Hancock, "Brahms in Better Balance," Historical Performance 5 (Spring 1992), pp.
37-39.
28. Steane, Gramophone 69 (January 1992), pp. 81, 84.
29. Scherer, BBC Music Magazine, December 1995, p. 69.
 REVIVING IDIOSYNCRASIES 377

versity Press, 1992). The best current biography is Malcolm Macdonald's

Brahms, The Master Musicians (London: Dent, and New York: Schirmer,
1990). Good discussions of Brahms's music include Michael Musgrave's The
Music of Brahms (Oxford University Press, 1985) and, on a more advanced
level, Walter Frisch's Brahms and the Principle of Developing Variation (Berke-

ley: University of California Press, 1984). Norman del Mar's Conducting

Brahms (Oxford University Press, 1993) is a fascinating guide to conducting

the works for orchestra.

 20
Reinventing Wneels
— NoQO^
Joshua Rifkin on Interpretation
and Rhetoric

Joshua Rifkin's career as conductor, harpsichordist, and pianist has taken him
from Busnoys to Bach, from Schiitz to Stravinsky, and from Josquin to Scott

Jophn —with, along the way, the spoof The Baroque Beatles Book and some
lovely instrumental arrangements for Judy Collins. The Bach Ensemble, which
he founded in 1978, has recorded a number of Bach works and has toured
throughout the USA and Europe; he has also appeared as guest conductor and
keyboard soloist with many modern orchestras.
Rifkin is also a scholar, who has specialized in Renaissance and Baroque
music, with particular emphasis on Josquin and Bach. His controversial argu-
ment about the size of Bach's "choral" forces is discussed in Chapter 15. Our
conversation touched on more basic issues of historical performance, though,
going beyond the specifics of scholarship to the foundations of the whole en-
terprise.

How much "interpretation" and inflection should go into a performance of,

say, Bach or Mozart? Some early-music performers have advocated a great deal
of these, others far less.

There's a soft underbelly to a lot of what we do in historical performance,

in that there are profound discrepancies between our modern interpretative
practices and what we can determine about musical practice in the eighteenth
century and earlier (and somewhat later, for that matter). By calling attention

to these discrepancies, I don't mean necessarily to criticize what we do, but

merely to try to heighten some awareness and promote some reflection.

378
 REINVENTING WHEELS 379

It's pretty obvious and well known that "interpretation," as we have inher-
ited this idea in the performance of standard repertory in the twentieth century,

was foreign to was Nicholas Kenyon who

most earUer music-making. I think it

said that, by all the evidence we have, music-making in the eighteenth century
was more like what we would call "readings" than what we would call "in-
terpretations." Except for operas, we know that they were lucky to have two
rehearsals of a piece, or even one, and a rehearsal basically meant a read-
through. When I try to imagine how this all went, I think of the jingle session
a modern situation in which musicians come in to a performing space of some
sort, are handed a newly written piece of music, read it once or twice through,

play it more or less flawlessly with a sense of its basic stylistic assumptions,
and then go home. Of course, this notion is quite distant from the way we think
of performing the great masterpieces, which we imagine to require much more
profound insight born out of years of reflection.'

This much is easy and obvious enough, I think, but there are aspects that
are less easy and obvious. To get at these I would refer to an experience I had
a couple of months ago, when I recorded several of the "London" Symphonies
of Haydn.- I was dealing with an extremely good period-instrument orchestra,
very experienced, technically very capable; yet we all found this music exceed-
ingly difficult. I myself had underestimated its difficulty, not simply in terms of
the individual parts (particularly the violin parts) but also in terms of the en-
semble demands and even the directorial demands that they posed. In the
course of the sessions the producer and I had a conversation which some
led to
further thought. He asked, "What must this have sounded like in London at
its first performance."* Given the lack of rehearsal, what kind of effect could it

have made?" In fact, by all evidence, it made an absolutely stunning effect, and
people just loved it. The reviews were enthusiastic beyond measure.

In part that may reflect our much higher expectations today for technical
perfection.
Partly, but there's more to the issue than that. As I was thinking about the

1. A good discussion of this topic can be found in Donald Mintz's "Mendelssohn as Per-
former and Teacher," The Mendelssohn Companion, ed. Douglas Seaton (Westport, Conn.:
in

Greenwood Press, forthcoming). Mintz gives references to the most important German dis-
cussions of this issue, especially that of Hermann Danuser, in his Mustkalische Interpretation
(Laaber: Laaber Verlag, 1992), Section 1. It appears that musical "interpretation" in the mod-
ern sense began to be discussed only as recently as the 1840s.
It could be argued that the second half of the eighteenth century saw the publication of
numerous books on musical performance, and that these sometimes indicate something that
sounds a lot like "interpretation." These books, and the music of such treatise writers as C.
P. E. Bach, may well represent the first sproutings of the concept of interpretation. But it's

likely that concept postdates J. S. Bach, at any rate —which does not mean we should avoid
interpretation when we play his music.
2. Nos. 96, 97, and 99, with the Capella Coloniensis.
380 CLASSICAL AND ROMANTIC

Haydn premieres, I was recalling that had recently heard the premiere of the
I

new Partita for Orchestra of Elliot Carter. In the talk before it Carter praised
quite rightly —
the accomplishments of the Chicago Symphony Orchestra and
its conductor, Daniel Barenboim. But then Carter said that as his pieces live

and are performed and re-performed (he's one of the few contemporary com-
posers to have the good fortune to see his music develop a performing tradi-
tion), performers continue to find more in them, and find ways of playing not
just the notes but really the music —to find a personal slant on the pieces. He
has, at least in principle, clearly welcomed this. And I know from other cir-

cumstances that Carter has felt the performance of his music to improve regu-
larly. Every ten or twenty years most of his pieces get freshly recorded and, he
tends to feel, better recorded.
Well, this is where our Haydn question and what Carter was talking about
started to meet. There's no question that a lot of first performances are very
bad, but not all are that bad, and a lot of pieces have triumphed at their pre-
mieres, as did the Haydn symphonies and, in fact, the Carter Partita. But in a
sense, the quality of the first performance is not really at issue, because no mat-
ter how bad or how good it may be, subsequent performances are going to
change things. Subsequent performances are going to assume an increasing fa-

miliarity with the work and its language, and are thus going to be able to
achieve a naturalness that first performances cannot achieve. And it is precisely
on the basis of this, of really speaking the language, that one begins to develop
insights into the particular utterance of the language and begins to develop in-

terpretations as opposed to simple readings.

The dilemma I want to return to is that in the general practice of the eigh-
teenth century they didn't yet have a standard, regularly repeated repertoire,
and thus could rarely have reached the interpretative phase. We, by contrast,
can't avoid interpretation —given the increasing familiarity of the works, both
in our own performances and in our inherited traditions of performances and
composition, even if they are traditions from so-called mainstream practice.
Moreover, we may need interpretation in a way that the original audiences did
not, because we have heard the pieces so was hearing
often. Mozart's audience
the "Jupiter" Symphony for the first time, so a competent run-through was suf-
ficient for them; but we have heard it a hundred times, so a mere run-through
would bore us.

The example you gave of Carter comes from, our century, in which there
has been no central musical style that all composers use; in learning a new
piece, we often have to learn not only the piece, but also its unique style —what
you called its "language. " By contrast, in the eighteenth century composers had
a more unified style, or language. Would that common style mean that even
with limited rehearsal there could be not only more insightful performances but
also more interpretation?
 REINVENTING WHEELS 381

It would have improved certain things without question. 1 think in princi-

ple we can imagine that performers would, in general, understand the signifi-
cance of a lot of musical gestures, expressive devices, and so forth, in ways that
modern performers dealing with contemporary pieces might not. It's perhaps
akin to how jazz musicians today can take a chart and know how the written
lines are to be realized without being told — or, to mention it once again, the
jingle session. I think this can stand as an example of a larger sense of "speak-
ing the language." 1 think what familiarity with the language does make pos-
sible is giving a much better delivery right off the bat. But delivery and inter-

pretation are not exactly the same thing, although locating the boundary
between them is a difficult matter. I think we can probably imagine a good level
of delivery, even if the technical level sometimes may have been problematic,
as in the Haydn. I think in many instances this could have taken care of a lot
of the issues. With much of the repertory would have taken care of all the
it

issues; there are pieces that we generally accept are not the most profound
under the sun but that with the right performance can be quite wonderful. Well,
there you have delivery pure and simple, unclouded by these other issues. But
for the great pieces, the issue remains.
So the historical performer is facing a contradiction. The questions with
which he or she has to grapple are. How far does interpretation take us from
the original, how much of that is legitimate, how much is not legitimate, and
what means do we have for deciding? At the same time we have to reckon
with not only how far this is taking us from the original, but to what extent
we can reconcile all this with the supposed and, in part, actual foundations
of the use of "historical" instruments, practices, etc. There is an interesting
set of muddles.

One response to the dilemma you mentioned how much to interpret, how —
much that is related to the original practice and how much not has been to —
develop a common style of playing Baroque music. David Fuller describes it

as "a kind of generalized rubato intended to clarify the metre and highlight
important notes and events. . . . Its salient characteristic is an exaggerated
lengthening of the downbeat—sometimes every downbeat, no matter how
clear it may already be from the harmony or other factors of its context. Fur-
ther rhythmic distortions may emphasize dissonant or climactic notes, the-
matic entries, or any feature the performer fears his audience may miss. Some-
times all sense of the beat, far from being clarified, disappears in a fog of
nuance.
"It is true that liberties far beyond overdotting and inequality were culti-
vated in certain Baroque styles. The few mentions doubtless only hint at a
wider reality. . . .

"What is certain, however, is that the 'audible analysis' practised by mod-

ern players of Baroque music has nothing to do with notes inegales. . . .Any
382 CLASSICAL AND ROMANTIC

connection with old performance is purely speculative —which is not to say that
may "^
it not have corresponded exactly to the way some players performed.
Could you comment?
It's a very well-taken point. Let's turn it around for a second. Let's give this

phenomenon its most charitable explanation. We could say that v^e have
learned something about this music. It's wonderful that people know something
about what the harmonic points of a piece are, where the phrases are going,
and so forth. This is a level of knowledge that's not to be gainsaid. I remem-
ber coaching a chamber music masterclass in a serenade of Max Reger that is

basically tonal but is not as familiar in its syntax as other late Romantic pieces
might be. These were very capable performers, extremely gifted; any one of
them could render a dazzling Brahms fiddle concerto or its equivalent. But they
made hash of Reger's music, because they didn't know where the phrases were
falling, where the weight lay, and so forth. It had to be painstakingly pulled
out of them by consideration of questions How
are we going to make this
like.

palpable, and how are we going Now, in repertories where this

to project this?
has not been so self-evident — and even Bach may be considered as being such
a repertory —the fact that musicians have become sensitized to these things, so
that they really have a sense of how the language is working, is something to
celebrate.

What you've just described sounds like, essentially, good "delivery." Where
does interpretation fit into understanding this style that Fuller describes?
Well, the next question is, of course, What do you do with this knowledge
of how the language works? Here opinions may differ, and here again we're
entering this dimension where the historical evidence leaves us in the lurch. It's

a very subjective matter, because one person's violently stretched-out, meterless

performance is another person's weighting of the proper moments in time; one
person's well-proportioned performance is another person's dull, uninflected
reading. We have no way, really, of telling. And here is where we become very
personal, and have very little choice but to be so.
Although here again is where we might want to ask about objective foun-
dations, at least in the interest of keeping in touch with them. To take a sim-
ple, trivial enough question, does it make a difference whether the kind of thing
that Fuller describes is happening in a solo harpsichord piece or in a piece for
large ensemble? If it's happening in the latter instance, is it reconcilable with
what we know about the sources of the time —about the orchestral parts of the
time, with their minimal performance instructions, and about the performance
practices? Is it reasonable to assume that a bunch of even extremely skilled mu-

3. Fuller, "The Performer as Composer," in Performance Practice: Music after 1600, ed.

H. M. Brown and S. Sadie (London: Macmillan, 1989, and New York: Norton, 1990), pp.
138-39.
 REINVENTING WHEELS 383

sicians very familiar with the style would have sat down in real time in the
eighteenth century, with real eighteenth-century performance practices, and
produced the kinds of results that Fuller described? If not, does that mean it's

a bad thing to do so now?

That brings us back to the initial point of departure about interpretation.
These pieces have progressed and changed through performance, and through
the way that they gradually divulge secrets to performers, and the way that per-

formers gradually find new secrets to tease out of them. Where does one call
a halt to this process? I don't know the answer. In a way it's akin to what hap-
pens to folk music — you can never keep it in its pure state. I think I've tended
to become something of an extreme relativist on these issues, in principle. In

actual practice, I hatemany of these performances, but that's just my private

taste, and I try to do my own performances.

Behind many of these performances is a particular way of understanding

the concept of musical rhetoric, which looks upon it as involving a kind of mu-
sical lexicon. One major proponent said that in the Baroque era "a repertory
of formulaic expressions (musical figures) was available for portraying emo-
tions and for figures of speech; a vocabulary of musical possibilities, so to
speak." Through such means, music "was always expected to speak.'''* Could
you talk about the current use of this idea of rhetoric in Baroque performance?
Rhetoric is one of these areas where we have to clear out the stables a lit-
tle bit. Rhetoric should be a very simple matter. The first point is that the trea-

tises from the late fifteenth century to the eighteenth are addressed to com-

posers, not performers or analysts; they have to do mainly with composing

pieces of music, not usually with performing them. The second point is that
their significance has been misunderstood; indeed, it is a sign of misunder-

standing that they are thought to have such significance.

Rhetoric has, first of all, almost nothing to do with content and meaning.
The use of rhetorical terms in music theory was simply a way of labeling cer-

tain devices in compositions for which there was not a commonly accepted ter-

minology. This is obvious if one reads the treatises; So-called compositional the-
ory up until the late sixteenth century was in fact contrapuntal instruction,
concerned with the relationship of notes against each other, and with the basic
contrapuntal devices —canonic writing and the like. What is never addressed in

4. Nikolaus Harnoncourt, Baroque Music Today: Music as Speech, trans. Mary O'Neill
(Portland, Oregon: Amadeus, 1988), p. 119.
An influential source on the relationship of music to rhetoric is the article "Rhetoric and
Music," by George J. Buelovv, in The New Grove Dictionary. A skeptical voice is that of Peter
—
Williams see, for example, his discussion in The Organ Mustc of J. S. Bach, vol. 3 (Cam-
bridge University Press, 1984), pp. 69-72, and his essay "The Snares and Delusions of Musi-
cal Rhetoric," in Alte Musik/Praxis und Reflexion, ed. Peter Reidmeister and Veronika Gut-
man (Winterthur: Amadeus, 1983), pp. 230-40.
384 CLASSICAL AND ROMANTIC

the theoretical hterature of the time is how actual pieces behave, even in such
simple descriptive terms as "here there's a bit of imitation, here there's ho-
mophony, there the line rises and the line falls." Now, the first great musical
treatise tolook at rhetoric, Joachim Burmeister's Musica Poetica 11606J, un-
dertakes the praiseworthy enterprise of trying to come up with descriptive
terms for what actually happens on the surface of musical pieces. Lacking a
terminology of music, such as the technical terminology that we have, he sim-
ply goes to the sister discipline that already has a developed terminology for
describing surface phenomena in a performance medium, and that is rhetoric.
Rhetoric is not grammar, and it's not the basic tools of speech; the basic tools
of musical speech were already part of traditional compositional theory.
Rhetoric dealt with delivery, and with the shape of sentences in terms not of
grammatical parts but of whether you repeat words for emphasis and so forth.

It is a kind of taxonomic business, on the level of the sentence, and also in

terms of the parts of an oration, showing how B follows A and how C follows
B. First you have this theme, then you have that theme, then you have the de-
velopment. Burmeister, needing terms for this kind of thing in music, simply
appropriates the terms used in rhetoric. A line goes up? Well, in rhetoric you
have a term to describe a climactic situation. A line goes down? You have a
term to express this. The voices come together in what we would call ho-
mophony? Well, you could borrow rhetorical terms for that. Two significant

things about this are that it has only tangentially anything to do with mean-
ing, and that the terms themselves have absolutely no significance. Indeed, if

you look at the history of rhetoric in musical theory you discover that differ-
ent terms are used by different authors, and that even the same author will bor-

row different terms from rhetoric at different times trying to get to a closer

analogy. But there's absolutely no mystical significance to the terms themselves.

You could just as easily call them Ginger and Fred as anabasis, catabasis, or

any of them.
Now, let's go to the end of the rhetorical tradition and consider the most
famous example, in which Mattheson says a piece of instrumental music is like

a Klangrede, "oratory in sound."' This became the title of a famous book,

Musik als Klangrede, by a very famous conductor who cut his teeth in the
early-music world. '^
There the proposition is made (and it's very widely held
among modern performers of Baroque music) that somehow there is a funda-
mental difference between music before the French Revolution and music in
later eras. Later music is held to be just notes and their relationships, but early

5. Mattheson, Der vollkommene Capellmeister (Hamburg, 1736; facsimile, Kassel: Baren-

1954); English translation, Johann Mattheson's "Der vollkommene Capellmeister": A
reiter,

Revised Translation with Critical Commentary, trans. Ernest C. Harriss (Ann Arbor, Mich.:
UMI Research Press, 1981), p. 425.
6. Harnoncourt, Baroque Music Today: the original title is Musik als Klangrede.
 REINVENTING WHEELS 385

music —ah! that's an orationF Well, come on. You read Mattheson, and he's

saying, Look, a piece of music has a beginning, a middle, and an end. And
that's all he's saying. It shows that Mattheson was no dummy about music. If,

as Birnbaum says,** Bach could expound on the relationship of music and

rhetoric, it would also suggest that he understood that a piece of music had to
be coherent and well made. But let's not kid ourselves that it means anything
else. Again, none of this has anything to do with some different style of per-
formance. The kind of "rhetorical" performance that we have been blessed
with over the last twenty years —which sometimes milks every little gesture for
all it's worth, and finds deep meaning in rhetorical terms that really just de-

scribe standard musical phenomena — has no historical basis.

In some ways performers are not to blame, because they have been misled
by the scholars. There is a pernicious tradition of German scholarship that has
created the fiction of rhetoric and meaning in music —Hermann Unger, Arnold
Schering, Arnold Schmidt, and in our country Ursula Kirkendale"" —and it's

widespread. It certainly has created the notion of a kind of secret language. Per-
formers have turned to them as scholarly authorities. And let's face it, it's an
attractive idea. We are all attracted by secrets and hidden meanings, meanings
that are more profound than what has been accessible to all of us (and might
now be accessible only tosome of us). It's understandable; but, as Dorothy
Parker said, there's less there than meets the eye.
I think sensitive musicians, when they heard under-inflected, motoric play-
ing of Bach, intuited that "we're not getting at something; there is more to
this." Reading some treatises, and hearing things from some musicologists, they
found that way the rhetoric idea, even if it was misconstrued, reinforced
in a

and helped them come to grips with this thing that they were intuiting. And

7. Many do believe that music changes, around this time, from a mimetic ideal (art as a
mirror up to nature or to emotion) to an expressive one (art as expressing one's unique inner
feelings). This differs from but is not unrelated to the change Harnoncourt argues for.

8. In his second (1739) defense of Bach against the attacks of Scheibe (discussed in John
Butt's interview, Chapter 9), Bach's friend J. A. Birnbaum, a Leipzig rhetoric teacher, writes:
"(Bach] understands so thoroughly the parts and benefits which the composing of a piece of
music has in common with oratory that not only does one listen to him with a satisfying plea-
sure whenever he directs his profound conversation to the similarity and conformity between
the two, but one also admires the clever application of the same in music." Translation from
Peter Williams, The Organ Music of J. S. Bach, vol. 1 (Cambridge University Press, 1980), p.
91.
9. Kirkendale, "The Source for Bach's Musical Offering: The Institutio oratoria of Quin-
tilian,"Journal of the American Musicological Society 33 (1980), pp. 8-141, argues that Bach
structured the work to follow the parts of an oration as set out by Quintilian. Christoph Wolff
and Peter Williams deny her claim; see Wolff, Bach: Essays on His Life and Music (Cam-
bridge, Mass.: Harvard University Press, 1991), pp. 421-22, Williams, "Encounters with the
Chromatic Fourth," Musical Times 126 (1985), pp. 276ff, and Williams, "Snares and Delu-
sions," pp. 235-38.
386 CLASSICAL AND ROMANTIC

then perhaps they produced some ridiculous performances, but nevertheless

that served as a means to help understand the speech better.

Of course, while acknowledging the ridiculous performances,

some would
say produced a few great performances.
it also
Indeed, it may seem that I think no profound truths about the eighteenth
century were being discovered in this emphasis on rhetoric. On the contrary: if
one can understand rhetoric properly, not as it has been misapplied to musical
performance, then one can in fact find it rather meaningful —although basically
it should be telling us nothing that we do not already know. Rhetoric is sim-
ply effective speech —
good public speaking, if you will. I've referred often here
to the idea —
of knowing the language knowing how, say, Haydn's or Carter's
language functions. That is precisely what rhetoric is supposed to deal with
not with hidden meanings or anything like that, but with knowing, when I

want to make an effective statement about something, how to make it clear,

how to set it off from something else, which words carry a certain affective
weight in themselves, and so
what any native speaker knows, and
forth. It is

what any good actor knows, and what any good musical performer perform-
ing his or her native "language" of music knows. In that way any good per-
formance is rhetorical, be it Toscanini conducting Beethoven's Third, or
Horowitz playing the Liszt Sonata, or Stravinsky conducting his own music.
You listen to Stravinsky conducting his music and it sounds coherent, while
with a certain level of conductors it doesn't always. In that unchallenging sense,
rhetoric is very much at the basis of all that we do, but it's at the basis of what
any decent musician in any repertory does.
I have to stress that for the most part the significance of these things is

purely musical and syntactic. It has a semantic dimension, but not one you can
put your finger on. An unusual leap, a chromatic fourth, or a special chromatic
alteration — it's something that you have to understand has a certain meaning
because it's not the everyday plain occurrence in a piece. Because it is an elab-
oration effect, an unusual construction, it means that you have to be aware of
its unusualness, its relationship to the usual, and that you have to have some
understanding of how you might project this, how you might make this make
sense.

I'm reminded of a story I heard from Chris Krueger, the flautist in the Bach
Ensemble, about having coached in the Berg Chamber Concerto under Rudolf
Kolisch. He said that what was striking about Kolisch when he would sing
them examples from this piece is how highly inflected the singing was, in an
unmistakably Viennese fashion. It was instantly clear, when you heard Kolisch
sing it, what Berg had in mind with all these lovingly detailed markings of his,
which are so often just completely wrongly played today because they are
played without any understanding of what lies behind the notes. Berg's was a
Viennese dialect, and Kolisch spoke that language, so he inflected naturally in
 REINVENTING WHEELS 387

that style. Again, it's knowing where Max Reger's cadences fall, knowing where
he's interrupting and evading a cadence, knowing that you have to project both
the sense that it is going towards something and that it is evading it. All that

is linguistic understanding — all that is, in the true meaning of the term, rhetor-
ical: it's how to put the point across. But much of this other stuff — treating
each little pattern of a few notes as a meaningful rhetorical gesture whose
meaning is coded in a Latin term — is without foundation.
For that reason, I would prefer just for a while at least to avoid the word
"rhetoric" until it can be shorn of this extraneous baggage. At which point I'd

be very happy to use it once more, because, properly understood, it's something
that every good performance has and needs.

Let me pursue some qualifications. John Butt writes that rhetoric was an
aspect of German Baroque performance as well as composition, and in a way
from later styles of playing. In the performance sphere, he says,
that did differ
it haddo with using ornamentation to increase the eloquence and emotional
to
power of the presentation. But he agrees with you that the various ornamen-
tal elements had no specific meanings or connections to sung texts.^°
He writes elsewhere that the added ornaments, musically, were elaborations

of a fundamental structure —there was a hierarchy—and that these figures were

therefore given some form of delineation in performance. '
' How
do you re-
spond to the idea that in the earlier Baroque — the text-centered post-Mon-
teverdi style, which we still find echoing in early Bach —rhetorical performance
was from performance later, and that it involved some kind of artic-
different
ulation of smaller units for added effectf^^
If one qualifies the meaning of the word "rhetoric" carefully here, so that

it doesn't suggest the esoterica we've just discussed, I can agree with him. Nev-

ertheless, even when one says, "This is the elaboration of a simpler figure, and
therefore it must be delivered in that way," the question of course is howf If

you don't understand it at all, you will probably ride right over it. But in the

10. See Butt, Musical Education and the Art of Performance in the German Baroque (Cam-
bridge University Press, 1994), pp. 46-51. See also his interview in Chapter 9 of this book.
11. "When it is considered that Bernhard and Walther viewed the added figures as the
elaboration of a fundamental prima prattica structure — in other words there is a hierarchy of
diminution within the music —and further that such a style resembled a rhetoric, it is not un-
reasonable to infer that figuration in performance would have been given some form of de-
lineation" (Butt, Bach Interpretation [Cambridge University Press, 1990], p. 19).
12. David Schulenberg, "Musical Expression and Musical Rhetoric in the Keyboard Works
of J. S. Bach," in Johann Sebastian Bach: A Tercentenary Celebration, ed. Seymour Benstock
(Westport, Conn.: Greenwood Press, 1994), pp. 95-109, sees Bach's style changing in this re-

gard. He says that Bach's early cantatas, written in the language of Buxtehude, illustrate im-
portant words musically, declaiming them in a rhetorical way; Schulenberg also finds elements
of this approach in Bach's early instrumental music. In later Bach, though, he says that for-
mal architecture becomes more important than such rhetoric, and the response to text is less

detailed.
388 CLASSICAL AND ROMANTIC

first flush and joy of discovery, you may think that you really have to com-
municate your discovery and enthusiasm to the listener, i.e. pump it for all it's

worth. Again, real speech lies somewhere in between, even in highly rhetorical

much of the stuff in rhetoric is just making conscious to

actor's speech. Again,

us what good speakers and good musicians do. As musicians we are like
—
Moliere's Monsieur Jourdain we're constantly discovering that we're speaking
prose, or even poetry perhaps. Any halfway sensitive musician does this. In a

sense, the awareness of this can heighten your doing it. But I wonder if there's

an obligation to make it noticeable.

Even if there's no obligation, an artist may prefer the result.

Absolutely, and that of course is their prerogative. But it's another question
when people speak of what happened in the past. We can never know, of
course, but I like to think that the best performers in those days did it intu-
itively. I really do not was any conscious notion of this in
believe that there
performance. What theorists were saying about it, insofar as it had any appli-
cations to performance at all, was in a sense more descriptive: they were putting
a magnifying glass to what happens in a good performance, just as the science
of rhetoric was fundamentally putting a magnifying glass to what good orators
did.i^

George Barth writes that Beethoven^'* played much of his music in a

"declamatory" style, delivering the phrases like someone declaiming, rather

than someone singing a long, mellifluous line or maintaining a moto perpetuo.

He relates it to the more detailed, "speaking" articulation that many people see
in pre-nineteenth- century music. He believes that the speaking style lost some
footing in the course of the nineteenth century and has been neglected in much
Beethoven playing in recent decades. Do you have any comments?
It strikes me as, on the face of it, a well-taken point. In a sense, though (and
I'm sure George would be the first to agree with this), all good singing is

declamatory and good declamation sings. I suppose my personal idea of

all

good performance style, which is of course purely subjective, is always want-

ing somehow to gain the advantages of everything at once. Good singing
speaks; there's Callas. And if you don't speak well, you end up with Monty
Python's parody of a BBC announcer, with uninflected sentences starting and
stopping in mid-sentence. So, it's a question of emphasis, or of being aware of
a side of it of which we perhaps had not been aware —of developing a sense of
how intelligibility is a matter of seeing what the gestures are, what the moves
are. Very often these things have been lost.

13. Another consideration, as far as historical foundations go, may be nationaUty. Butt
suggests in Chapter 15 of this book that German singers articulated the figures more than did
Italians.

14. The Pianist as Orator (Ithaca, N.Y.: Cornell University Press, 1992).
 REINVENTING WHEELS 389

I'd add that it's not just in the performance of Beethoven. Where George
Barth and others might use the word "declamation," I word
tend to use the
"inflection." It is my sense that this "declamation" or "inflection" was more
characteristic of all musical performance before the Second World War. You lis-
ten to Kreisler or Furtwangler, and you hear highly inflected performances;

Schnabel's Beethoven always seemed to me a very speaking kind of playing. All

three are in many ways examples of what a lot of people in the so-called early-

music business have been trying to rediscover and recover. The music-making
of these pre-war artists is indeed speaking, saying things. In that sense I think
that we are reinventing a particular wheel —which I don't say as criticism, be-
cause wheels constantly roll out of sight and have to be reinvented. But maybe
what all this is about is that while we may need all the historical apparatus and
all the PR, in a sense we're just trying to plug into some home truths.

The problem, I think, is that we all like structural cohesion and continuity,
we all like detail, we all like declamatory speaking, we all like beautiful sounds,
we all like guts, —
we all like sensitivity but we can't do full justice to any one
of these elements without glossing over another one. In practice, what happens
is that each of us likes these elements in various proportions at various times;
and similar shifts take place, from decade to decade, in the fashions of musi-
cal reception. There are many ways of slicing it; every era will slice it differ-

ently, and so will every performer, each time he or she performs. I was just

reading some interview comments I made eight years ago. I was astonished to
see myself saying some of the things I said. Whether I have really changed my
beliefs or fashion has dragged me out of them I don't know; but one comes to

other places — thank God.

SELECTED DISCOGRAPHY
Rifkin's 1982 recording of the B Minor Mass (Nonesuch 79036, 2 CDs) was
the first to use the one-per-part approach he advocates. Critical responses vary
widely to this day. The disc won the 1983 Gramophone Award; its strongest
advocate, Teri Noel Towe, calls it "intensely powerful [and] revelatory," adding
that the participating musicians have "the guts to sing and play with vigour
and sensitivity, warmth and understanding."" By contrast, Nicholas Kenyon
says that Rifkin "just lets the music happen . . . and to my mind misses many
of the opportunities offered by one-to-a-part performance.""' Peter Williams,
occupying middle ground, agrees that "the singers simply sing: no rhetoric, no
showmanship," but he likes the result more, saying, "the chamber performance,
quiet and undemonstrative, brings out the wonderful inherent melodiousness of

15. Towe, "J. S. Bach: Mass in B Minor," in Choral Music on Record, ed. Alan Blyth
(Cambridge University Press, 1991), pp. 57-58.
16. Kenyon, "Bach's Choral Works: A Discographic Survey," Part 1, Opus, December
1985, pp. 14-17; quote, p. 16.
390 CLASSICAL AND ROMANTIC

Bach's music most beautifully." Still, he thinks the performance is undermined

by the instrumental work, which "in rejecting Harnoncourt's constant cresc-
dim effect . . . replaces it with nothing."'^ I will say only that the tempos,
which in 1983 often seemed too fast, today (after years of hearing early-music
Bach) seem reasonable throughout, with the exception of the speedy Sanctus.
According to John Butt, in Rifkin's 1985 recording of Cantatas 147 and 80
(L'Oiseau-Lyre 417 250-2) the advantages of one-per-part "are
manifold. . . . balance and ensemble are superb throughout."'** Butt also says
that the alto aria in 147 is "beautifully interpreted." Many feel that Rifkin's

finest Bach record is of the Miihlhausen cantatas numbers 106 and 131
(L'Oiseau Lyre 417 323).
Scott Joplin's revival in popularity was effectively launched by Rifkin's
Nonesuch recordings; in his latest contribution in this genre, a collection of rags

by Joplin contemporaries and tangos by Ernesto Nazareth {Rags and Tangos,

Decca 425 225), Malcolm Macdonald finds deep insights that are "aided and
abetted by [Rifkin's] exceedingly resourceful piano-playing."'^

FOR FURTHER READING

Richard Taruskin's seminal writings on authenticity and historical performance,
referred to so often in this book, are collected and updated in Text and Act
(New York: Oxford University Press, 1995). Another significant collection fea-
tures various authors: Authenticity and Early Music, ed. Nicholas Kenyon (Ox-
ford University Press, 1988). Peter Kivy's Authenticities (Ithaca, N.Y.: Cornell
University Press, 1995) is an important philosophical discussion of the topic.

17. Williams, Early Music 12 (February 1984), p. 139.

18. Butt, Early Music 15 (November 1987), pp. 575 and 577.
19. Macdonald, Gramophone 69 (April 1992), p. 120
 EPILOGUE

Nicholas McGegan described the current early-music scene to me in these

terms:

The sense of protest has gone out of it, which I think is potentially very

helpful. In other words, we're doing it because we en)oy it, not because we
don't enjoy what other people are doing. The other thing is that we are no
longer wrapping ourselves up in this cloak of being —unlike modern play-
ers
— "correct," which of course is something you can never be in the arts.

Also, once you get obsessed with style at the expense of content, things are
just as much out of whack as when people are obsessed with content at the
expense of style.

There's a lot of truth to that, and yet some of the interviews in this book (not
McGegan's) do convey a sense of protest and of being more "correct" than
mainstream players. In this and many other respects, the interviewees' motives,
creeds, and methods vary. The interviews bear out Michelle Dulak's comment,
"Never has historical performance been stronger than today, and never has it

been harder to say exactly what it is."'

Still, to facilitate discussion we can divide the interviewees into three broad
types. Placing an artist in any one camp may be (unintentionally) insulting —
apologize if so —and it usually oversimplifies, making what's ambiguous seem
definite; but it might be useful in picking out trends.
Artists of my first type uphold what we might call the central early-music
tradition: they adhere firmly to the ideal of trying to play music as it was played
in its own time. Their most explicit statement comes from someone whose
schedule didn't allow a full-length interview — Andrew Parrott, a British con-
ductor whose recordings have received deservedly warm reviews:

To argue that we need not concern ourselves with earlier performing con-
ventions simply because our ears are (necessarily) "modern" is, effectively,

1. "The Early Music Movement Circles Its Wagons Again," The New York Times, 11 June
1995, Sunday Arts and Leisure section, p. 25

391
392 EPILOGUE

a self-fulfilling prophecy: we can listen in only one way (a "modern" way),

so that is how we shall listen. Unfortunately, this is a popular viewpoint,

perhaps because most classical musicians are temperamentally more conser-

vative than they care to admit, but also because many react against any ap-
parent preoccupation with "rules" and against anything they perceive as
prescriptive or restrictive. I have never seen performance practice that way.
The more I learn —and there is so much more to discover than many begin
to imagine —the richer, the more fascinating, and, often, the easier perfor-

mance becomes. The challenge of absorbing new information, rather than

constricting and limiting the performer, acts as a stimulus to the creative

imagination and can also prove positively Hberating. It does not so much
dictate what not to do as offer us new ideas of what we might choose to

do. New possibilities emerge, even if old ones fall by the wayside. Surely it

is only the complacent, over-cautious or unimaginative musician who can-

not be bothered to rethink aspects of performing style in this way.

One could question that last sentence. We wouldn't consider a Furtwangler or

Beecham complacent, over-cautious, or unimaginative, but they probably did-
n't think about performance practice very much. They certainly didn't accept
the early-music movement's view of style as something you choose; what they
focused on was content. (This can be related to Robert Morgan's idea, cited in

this book's introduction, that musicians in the past believed themselves part of
a living tradition stretching back to Bach, and therefore felt free to play him in
their own current style.) One could also argue that for some early-music per-
formers —though not for Parrott —the "rules" have in fact been prescriptions,
which serve to reduce the need to make subjective decisions. All the same, Par-
rott's program is likely to stimulate, liberate, and give new possibilities — at least

when it is joined to Parrott's level of artistry. And his unapologetic conviction

does not bear on his artistic success.

Nonetheless, the interviews show that historical evidence doesn't inspire

everyone in the same way. The second type of artists I identify rejects, at least

partially, the core ideal of historical authenticity. Such artists aren't compla-
cent —they know their history and have rethought their styles. But unlike Type
Ones, they flout history openly when they prefer something else. We might as-

sume that these artists have been emboldened by the skeptics' battering at the

historicist ideal; on the whole, though, it seems to me that both Types One and
Two reach their positions not through the intellect but through temperament
or artistic need. Presumably Type Twos, unlike Parrott, believe themselves ei-

ther constrained or ill served by historical strictness. I think here of Peter

Phillips, who was arguing his viewpoint in 1978, and of Jeffrey Thomas and
others that I place in this group.

2. According to Taruskin, my Type Ones, the early-music mainstreamers, aren't simply try-
 EPILOGUE 393

The third type is often a subset of Type One, sharing the dedication to his-
tory that Parrott commends; but it uses history radically, to undermine a more
basic assumption, one that the first two groups share with the mainstream. This
assumption is Werktreue — fidelity to the work —
and, behind that, the concept
of the fixed, perfected work itself.^ In the introduction to Text and Act (p. 13),

Richard Taruskin complains that "The whole trouble with Early Music as a
'movement' ... is the way it has uncritically accepted the post-Romantic work-
concept and imposed it anachronistically on pre-Romantic repertories." What
draws his censure "is not the anachronism but the uncritical acceptance — and
the imposition." Whether you agree with him or not, the interviews show that
a growing segment of the "movement" neither accepts the work-concept "un-
critically" nor always imposes it, anachronistically or otherwise. I'm thinking
above all of improvisers like Andrew Lawrence-King, Julianne Baird, and
Robert Levin, who are being faithful to the historical practice of not being
faithful to the work. Baird speaks explicitly about changing the balance of
power between composer and performer, and restoring the performer's role in
the compositional act —that is, of returning to a prt-'Werktreue philosophy. She
has chosen a repertory that allows this, and so supports Dulak's view that
"early music" has come to mean, above all, a place where musicians are al-

lowed to experiment with approaches that are discouraged in the mainstream.

(I should add, though, that Levin's improvising is meant to fulfill Mozart's in-

tention.)
Another possible undermining of Werktreue is found among those artists,

like John Butt, McGegan, and Paul Hillier, who praise rougher, less perfect, less
polished playing —who might say, with Anner Bylsma, "To hell with the per-
fection of the highway tarmac!" Today's emphasis on flawless execution might
reflect(among other causes) the recording-studio ethos and, also, Werktreue —
being so true to the work that every one of its notes is audible and in just the
right place. While many musicians regard modern technical perfection as a
straightforward example of progress, others take a more Luddite view. I know
of a musicologist who prefers to hear Classical string quartets played by ama-

ing to play music as it was played when it was new. Taruskin would argue that Type Ones,

like Type Twos, also use history selectively, in order to produce a result they and their audi-
ences like. He argues, however, that this is not a conscious process: unlike Type Twos, they
usually won't violate what they have concluded was the historical practice, except when they
must (using mezzos or countertenors instead of castrati, for example) or in rare cases where
the historical practice is clearly harmful to the music. But unconsciously, he thinks. Type Ones
tend to deal with evidence according to their biases. Readers may decide whether the inter-
views support his hypothesis.
3. Lydia Goehr recognizes this: "More than any other movement currently existing within
the European tradition of classical music, the early music movement is perfectly positioned to
present itself not only as a 'different way of thinking about music,' but also as an alternative
to a performance practice governed by the work-concept." Goehr, The Imaginary Museum of
Musical \(/orks (Oxford University Press, 1992), p. 284.
394 EPILOGUE

teurs; the modern professional quartets have such perfect ensemble, he believes,
that they obscure the identity of the four parts in a way never imagined by the
composers. It sounds as if one person is playing all four instruments, rather
than a conversation of four distinct individuals.
In its movement was sometimes a caricature of
early days, the early-music
a poorly poUshed approach and was often derided for its amateurism; musi-
cians hadn't yet learned to play old instruments well, so they were often out of
tune and out of sync. It wasn't until they could play old instruments with all

the polish of the mainstream that they became accepted (and at the same time,
controversial —they could no longer be laughed off). The trend I see in my few
Type Three interviews is a turnaround: now they can play perfectly, but these
artists aren't sure it's always worth the price. Various performers today are
questioning the ideal of technical perfection, and the majority of them are in
the early-music sphere (mainstreamers who share such questions might be less
likely to get away with acting on them).'' This is logical on historical grounds:
although performers in past times could be highly virtuosic, the evidence indi-
cates that our concern with studio perfection didn't arise (usually) until some
decades into the twentieth century. Before then, for example, uniform orches-
tral bowing was rare, and what musicians call "ensemble" was looser (as we
hear in early recordings of quartets, for example). This historical argument has
sometimes been raised to excuse the amateurism of early-music performers of
the previous generation: that logic is questionable, but I don't think it moti-
vates the artists I mention. I think their concerns represent, once again, tem-
perament — in this case, a temperament that finds clockwork perfection an artis-

tic straitjacket.

Again, there are qualifications. Peter Phillips and Christopher Page are
hardly alone in the aesthetic pleasure they take in perfect ensemble; and per-
haps Page is right that this ideal was shared in the distant past. It seems to have
applied in the orchestras of Lully and Corelli as well, and perhaps elsewhere.
That, then, is my typology, although (as I said at the outset) it's far too
schematic. Leonhardt might seem a classic example of Type One, but not only
has he become less so in his statements, he is also known to have been a great
closet improviser all along. My Type Threes often make reference to jazz, but
so does William Christie, who is hard to classify. A single artist may fit all three
of the above types at various times; as I've said, the historical-performance
movement has never been harder to define.

4. I may this. When The New York Times mentioned "wrong notes and
be wrong about
hobbled which were by no means infrequent" in a recent Beriin Philharmonic Brahms
entries,
cycle in New York, it added that "they were the inevitable byproducts of risks being taken by
players however consummately skilled, a part of the overall exuberance of the perfor-
mances. ... In the way he marshals this excitement Abbado seems to have attached to
. . .

something more primal in the orchestra." James Oestreich, "Abbado, Making Berlin Philhar-
monic His Own," The New York Times, 9 October 1996, pp. Bl, B5.
 EPILOGUE 395

In the introduction I said that even when you believe that historical prac-

tice matters, you still have to decide whether each specific practice is essential

to performance, or not important, or even harmful. My "typology" could be

understood as three different angles on those decisions: Type Ones consider
more of historical practice essential than Type Twos do; Type Threes value a

different set of information. In any event, the three-part typology may drama-
tize the most obvious implication of this book: that to refer to historical per-
formers in the aggregate as "they" (as some critics still do) is too vague to be
useful.

My grouping also seems to call out for some crystal-ball gazing. I think
many people would agree with Robert Levin that the improvisatory approach
of some Type Threes is likely to prosper, simply because it offers so much ex-
citement to audiences and performers. Many would agree with
people, too,
those who see Type Two as representing an advance: when Harnoncourt de-
cides to conduct modern orchestras with only a few specific historical instru-
ments, it shows that for him historical performance has become not a recipe
but an option, which he can exercise for artistic reasons. Still others believe
that the Type One approach yields the most progress (Parrott is not the only
Type One whose performances are especially admired —some others are Rein-
hard Goebel and Sigiswald Kuijken, both of whom try to be uncompromising
about historical veracity). Parrott is not alone in fearing that the Type Two ap-
proach could lead to laziness. It would be easier to choose sides if one of the
three types were to produce consistently better artistic results than the others,
but none do; it all depends on the artist.

Part of my explanation of why Type Three will prosper —audience appeal

relates to another issue. Although marketers make claims about "Verdi as
still

Verdi would have heard it," few artists do anymore; more and more of them
seem preoccupied with the concerns that Anthony Rooley, Peter Phillips, and
others express about giving the audience a meaningful and even an enjoy- —
able — experience. For various reasons, some today*^ see this as a healthier di-
rection for performing artists than Romantic disdain for audience approval
an ideal sometimes at least claimed by classical musicians. But a less theoretical

factor may also bear on this change in orientation. Demand for classical music
has been shrinking, as younger audiences desert it for, essentially, American
popular culture.^ The reasons for the desertion are too complex to go into here.

5. See Taruskin's Hate Horowitz?" The New York Times, 28 Novem-

"Why Do They All
ber 1993, Sunday Arts and Leisure section, p. 31, for an example of this viewpoint.

6. Edward Rothstein ("The Tribulations of the Not-So-Living Arts," The New York Times,

18 February 1996, pp. El, E14) reports on a study commissioned by the National Endow-
ment for the Arts, "Age and Arts Participation with a Focus on the Baby Boom Cohort:
1982-1992," by Judith A. Balfe and Richard A. Peterson, which was based on interviews with
10,000 American adults. It shows a "massive shift in taste and tradition" over the generations
topop music and mass culture and away from the fine arts. This is hardly a unique finding.
396 EPILOGUE

but what is relevant is that it might especially affect early music, which has usu-
ally been even more marginal and specialized in its appeal than mainstream
classical music.

It's not surprising, then, that the rare early-music hits of the last few years
have often involved some form of popular culture. Several have involved the
dominant styles of our day —witness the Hilliard Ensemble's jazz collaboration.

Others have resulted from association with the dominant medium of our day,
—witness Jordi
film Savall's heartrending soundtrack to Tous les matins du
monde—though film tie-ins are hardly a new phenomenon (think of the 1940
cartoon Fantasia). The Hilliard and Savall examples arose from artistic con-
viction, but their lessons have not been lost on marketers. As Erato undertook
Ton Koopman's Bach cantata series in 1995, it tried to interest a Hollywood
studio in a film about Bach, featuring Koopman's playing. Polygram released a
pair of Gardiner's Beethoven symphonies to capitalize on the film Immortal
Beloved; the CD cover featured the following blurb:

You gotta have it! The full, incomparable thrill of Beethoven's most revo-
lutionary symphonies in white-hot performances. The Fifth, that cosmic tale

of tragedy leading to triumph, is the "Star Wars" of symphonic music. And

the Third is a swashbuckling thriller which for sheer passion, romance, and
gusto had to wait for Indiana Jones in "Raiders of the Lost Ark" to find its

visual counterpart.^

This may not seem much of an improvement over "Beethoven as Beethoven

imagined it," but the marketers know what works: this release sold well. We
can expect them to remember that.

We can also expect to see more performers trying to build bridges to mass
culture through their own presentations. I think of the ensemble Bimbetta,
which bills itself as "Five Babes Who Go for Baroque," and whose concerts
embed seventeenth-century music in hip, postmodern theatre. I've heard good
reports but haven't seen the group. I won't speculate on the general merits of
such approaches, beyond predicting that some will work and some will turn
the stomach — when
a young American maestro "bounds on stage dressed as
as
Superman or Mozart wig," which led Leon Botstein to say, "This is so
in a

horrendous it bears no description."^ (Of course, anything, including standard

concert etiquette, will turn the stomach of someone, somewhere.) Some believe
that such approaches are a necessity; the American Symphony Orchestra
League has, for example, suggested that orchestras play more pop and ethnic
music to attract larger and younger audiences (I think they used the adjective

7. DG Archiv 445 944.

8. Audrey Choi, "Modern Maestros Conduct Themselves in Offbeat Fashions: To Sell
Seats, They Will Dress Like Batman or Mozart; Riding in on an Elephant," The Wall Street
Journal, 8 January 1996, pp. 1, 4.
 EPILOGUE 397

"diverse"). I'm nor sure how one would apply thar particular recommendation
to early music, but the need is at least as real. The reconciliation of popular
and classical cultures that Robert Levin called for in his interview is happen-
ing in several areas today, and early-music crossover is one. Its successes need-
n't necessarily be hip ones, by the way: Anonymous 4 has managed, for ex-
ample, to touch large audiences deeply without making its concert presentation
even slightly pop. Still, the group's market success results not from its concerts,
memorable though they are, but from its recordings. These have clearly tapped
into popular trends (even if that doesn't reflect the artists' intentions) and have
reached the market through popular media. It's not for nothing that the New
Yorker called the group "the fab four of medieval music."
Mention of Anonymous 4 brings us to another forecasting question: what
will the few media hits do for the great majority of early-music performers?

It's possible that the hits may boost demand for early music as a commod-
ity. But it has also been argued that, in general, mass media and telecom-
munications tend to drive up demand only for the few market-preferred
"superstars," and that this tends to drive down demand (and fees) for the

non-superstars." It can't be assumed that the success of the media's chosen

few will rub off on the remaining body of worthy artists —the opposite is just

as likely. People who spend good money on Jordi Savall CDs and concerts
may be less inclined to come out to hear a local gambist, or at least to pay
a lot of money to do so. (And obviously, electronic media like CDs are a
major factor behind the shrinking of concert-hall audiences — music lovers stay
home rather than go to concerts.'" )

This brings us to another prognostication factor: the classical recording in-

dustry, which, by all accounts, is in financial straits. Sales of the standard reper-
tory have flattened, and even major orchestras are losing their recording con-
tracts." Early music's rise to popularity depended critically on recordings, so

the poor health of the record industry will be unwelcome news. Record com-
panies may become less willing to take chances with early-music recordings that
require large ensembles or that feature unusual repertory. They may also be less

9. This is the "superstar model" of income distribution, proposed by the University of

Chicago economist Sherwin Rosen in 1981; for an explanation, see Paul Krugman, Peddling
Prosperity (New York: Norton, 1994), p. 149. Rosen applies the model not only to enter-
tainers but also to top lawyers, business executives, and others. On this, see Robert Frank and
Philip Cook, The Winner-Take-All Society (New York: The Free Press, 1995).
"The NEA report notes
10. Rothstein writes, that 'video consumption" of classical
. . .

music is high for the same age groups that show declines in attendance at concerts. Record-
ings have also become more important as a replacement for the live experience" ("The Tribu-
lations"). He adds that this news is even worse than it may seem at first glance, because clas-
sical record buyers often use the music as background, rather than as something to listen to

with serious interest. This, as noted early in this book, is certainly true of currently popular
medieval music.
11. Allan Kozinn, "Strike in Philadelphia: What Stopped the Music," The New York
Times, 17 September 1996, p. CI.
398 EPILOGUE

willing to risk money on unknowns, even gifted ones, which would tilt the scale
even further toward the superstars.
However that plays out, it is fair to say that an artistic scene generally
thrives when those who want to practice the art can afford to do so. Economic
viability is not a sufficient condition for "an abundance of musical genius" to
emerge, but historical studies suggest it may be a necessary one.'^ A poet like
Wallace Stevens may have overcome conditions that wouldn't let him devote
time to writing poetry, but especially among performing artists, who need to
practice and rehearse together, such adverse conditions tend to reduce output
and thin the ranks. This brings us to another economic factor clouding the crys-
tal ball. The degree of non-box-office financial support available to artists
whether from private donors or from governments — is likely to affect any non-
popular art. Early music has usually had limited box-office appeal; even in its

era of origin, "private" patronage was typically paid for out of various forms
of hidden or direct taxation, whose revenues were spent by and for aristocrats
rather than the general population. We can't safely predict the extent of gov-
ernment support coming decades, but we can note that in France and the
in

Netherlands the generous government support, though probably not the key to
the thriving early-music scenes there, has probably been a non-trivial factor. It

has made it possible for hundreds of musicians to develop and pursue their art
full-time. We can also note that in the USA such support has been a whole
order of magnitude less per capita,'^ and that this may have been one reason
why so many leading US early-music artists have migrated to Europe,'** and

12. "Although suitable economic circumstances are in themselves hardly sufficient to elicit

an abundance of musical genius, they may constitute a necessary condition for that result."
William and Hilda Baumol, "On the Economics of Musical Composition in Mozart's Vienna,"
in OnMozart, ed. James Morris (Cambridge University Press, 1994), p. 87. The article dis-
cusses Elizabethan theatre, and music in the Hapsburg empire in the late eighteenth century.
There was high demand for theatre in Elizabethan London and for music in late-Hapsburg

states, and general wage levels were low in both eras; these factors combined to create a great
deal of attractive employment opportunity in the respective arts. "[SJurely it is plausible,"
write the Baumols, "that many of those entering the labor market would turn to careers for
which they thought themselves suitable in professions where there existed opportunities for
employment." Such reasoning might encourage modern youngsters to seek careers in pop
music, TV, or film (and of course in medicine and dentistry), but not in classical music.
13. The most recent international study, done by the London-based Policy Studies Institute,
found that in 1987 the Netherlands and France spent ten times as much per capita on the arts
as the US did (these figures sum arts spending from all levels of government, then divide the
total by population size). Similar studies from other sources have found similar ratios. Other
indirect factors —
welfare and educational benefits, government support for churches, universi-
ties, and classical-music radio —
tend to exaggerate the difference rather than diminish it; dif-
ferences in private philanthropy and corporate support do not alter it significantly.
14. For example, Benjamin Bagby, William Christie, Sarah Cunningham, Alan Curtis,
Bruce Dickie, Laurence Dreyfus, Jonathan Dunford, Elizabeth Gaver, Nancy Hadden, Sterling
Jones, the members of Project Ars Nova, Skip Sempe, Hopkinson Smith, Stephen Stubbs, Bar-
bara Thornton, Glen Wilson, and many others. Joshua Rifkin's home base is in Massachu-
setts, but most of his performing is done east of the Atlantic.
 EPILOGUE 399

why so few leading Europeans have migrated to the US (my book gives dis-

proportionate coverage to those few, for practical reasons). The demise of US

federal arts and humanities spending, and the possible reduction of tax incen-
tives for private donors, may therefore have implications for early music. Even
if these prove dire, the question remains of whether they should be a matter of
concern for US policy makers; 1 will not discuss that here. Obviously, if gov-
ernments pay people to do something (as those of the Netherlands or France
do for early music), more people will do it, and some of them will do it well;

but it may not be something a society values. Clearly the US citizenry, by and
large, does not value early music. Proving that it should do so is not as easy
as I would like, though would argue (were it less tangential to this book) that
I

the arts and humanities in general deserve the country's support.

We could try to read plenty of other tea leaves. Howard Gardner suspects
that "we are reverting to a period in which creative activity will be less in-
dividualistic and less iconoclastic, more communal and more continuous with
its past"'^ —an interesting prediction, but I'll refrain from speculating on it,

except to say that such a development might suit some forms of early music
well (and others poorly, such as that archetypal individualist/iconoclast,

Beethoven). Besides, I've yet to be convinced that it's happening, attractive

though it sounds. Bimbetta raises the issue of how early music might be af-

fected by the "postmodern condition." This book may or may not demon-
strate Brad Holland's remark that "if you're confused about [postmodernism],
that's probably because you're beginning to understand it";'" but I agree with

John Butt, who sides with "critics who are sceptical of postmodernism as an
ideal." '^ Butt adds, though, that the term "is certainly acceptable —indeed use-
ful — as a description of the condition we happen to be in" (I'd add that some
of its theoretical concerns are important ones). Still, Butt finds fault in see-
ing postmodernism "as the answer to all the evils of modernism, as the way
for the future, even as a happy utopia in which all differences will live side-

by-side in a pluralistic flux." To pursue his point, consider what happens

when you try to apply postmodernism to early music. If modernism implies
a disdain for one's audience, then those seeking to win popular audiences
might be considered postmodern — but Butt points out that when pleasing the
As for the European expatriates who appear in my book, two (Butt and Hillier) have uni-
versity positions, so don't rely on performing for their incomes (this is true of the few other
expatriates I —
can think of, who are, by the way, mostly British) and Butt will be moving back
to England at about the time this book is published. The third, McGegan, spends as much
time performing Europe as in the US. Many American artists who remain in the USA
in

—
such as Baird, Bilson, and Levin typically depend on university posts, not concerts, for their
living.

15. Gardner, "How Extraordinary Was Mozart?" in On Mozart, p. 50.

16. Holland, "Express Yourself: It's Later than You Think," The Atlantic, July 1996, pp.

66-68; quote, p. 66.

17. Butt, "Acting up a Text," Early Mustc 24 (May 1996), p. 327.
400 EPILOGUE

audience becomes a musician's overarching goal, it can create just as many

aesthetic dilemmas as disdaining the audience does. If by modernism we mean
what some call "reification" and "sacralization" of the work of art, then post-
modernism may be the term for those who embed music in hip theatre or
otherwise take it "off its pedestal"; but while Monteverdi sung in the con-
text of hip theatre may be fun and may reach new listeners, it may not nec-
essarily be a greater human experience than Beethoven played to a rapt au-

dience by Schnabel. And if by modernism we mean a preoccupation with form

and structure, then some of the improvising Type Threes could be classified
as postmodern (as well as pre-modern); but the achievement of a Beethoven
in using form for expressive ends may be truly great. A happy pluralism,
whether it's postmodern or something else, might be a good development
and it is safe to say that we're getting more of it today than we used to.

The historical-performance movement is the child of an unlikely union —that

of scholarship and art. It would bring book to a nice conclusion if I
this

could say that it is reaching adulthood. Pronouncements like that should be

made with caution, but a case can be made for this one. Of course, record-
ings from thirty years ago by David Munrow, Thomas Binkley, Michael Mor-
row, Leonhardt/Brueggen/Kuijken, and Harnoncourt preserve music-making
that is anything but immature. And there have been some dull and some
bizarre performances in recent years; and, as Laurence Dreyfus argues, com-
mercial success has led to institutionalization and to some formulaic, thought-
less playing.'** Still, there is a much larger pool of thoroughly accomplished
musicians in thefield now; their technical standards have risen markedly; they

indulge in mannerism and exaggeration; fewer "demonstration" perfor-

less

mances take place, whose main goal is to show that something can be done
rather than to make music. It can also be said that various groups of artists

are maturing. Rooky's discussion of his own artistic odyssey exemplifies one
such group: musicians of a more literalistic, polite background who have
learned to step out and become freer and more expressive. A mirror exam-
ple involves certain Continental musicians whose extremely inflected, some-
times overwrought playing has become more mature and integrated, and thus
even more exciting. It is also safe to say that the discourse about historical
performance is more mature and sophisticated. McGegan's view that the sense
of protest and correctness has gone out of it is, to a large extent, valid, even
among Type Ones.
Of course, there is no reason to assume that artistic movements like this one
necessarily evolve to ever higher states of maturity. Artists often do; movements
may or may not. As it happens, so far the movement has matured. There are

18. Dreyfus, in his section of "The Early Music Debate," Journal of Musicology 10 (Win-
ter 1992), pp. 114-17.
 EPILOGUE 401

a number of exciting artists at work today who have evolved wonderfully in

their own playing, and many of them have learned tremendous amounts from
their teachers' experiments or their own long experience. For this reason above
all, historical performance has never been stronger —or harder to define, partly
because it interacts with the mainstream in many ways, yet still maintains
so
the separateness of its niche. It wouldn't surprise me if in the future historical
performance becomes even harder to define and, not coincidentally, even
stronger.
 INDEX

Note: The index does not cover Arnold, Denis, 144

discographies, bibliographies or foot- Arts funding, government, 253, 398-99
notes except where they contain sub- Ars nova, 74
stantive discussions. Interviewees and Aubry, Pierre (duel with Jean Beck), 50
the ensembles associated with them are Augustine, Saint, 56 n. 5, 63
indexed only when they are discussed in Authenticity (historical), 9-20, 23, 26,
part of the book other than their own 29, 50, 78-79, 83-86, 98, 109,
chapters; their appearances in their own 165, 245, 254, 262, 267, 279-80,
chapters are not included in the index. 290, 319, 324, 329-30, 391-95;
Boldface indicates a musical example. and Aladdin's lamp, 221-22; and
composer's creative process,
Abbado, Claudio, 394 n. 174-77; and interpretation,
Abelard, Peter, 57 378-81; and Renaissance choral
Accademia Nazionale di Santa Cecilia, music, 102-3, 118-20
118 Avicenna, 57
Agricola,Johann Friedrich, 227, 229,
231-32, 237, 241 Bach, Carl Philipp Emanuel, 238, 360,
Alan of Lille, 57 379 n. 1

Allanbrook, Wye Jamison, 9, 292, Bach, Johann Sebastian 4, 11, 12, 15,
295 n., 323, 338, 359 117, 122, 129, 156, 157, 164,
119
Allegri, Gregorio, Miserere 165, 166, 221, 260, 378, 382;
Ambrosian chant, 27, 34-35 boy's choruses in, 282-83; choral
Angelopoulos, Lycourgos, 33, 38, 41 size, 279, 283, 288-90; French
Anglebert, Jean-Henri d', 261 and Italian styles influences on,

Anderson, June, 5 264; metaphysics and religion,

Anglican choral style, 97, 104, 113, 185-90; and the modern piano,
123, 245 195; and rhetoric, 182-83, 282,
Anonymous 4, 7, 24, 397 387, 387 n. 12; vibrato in, 236;
Anti-Semitism: in medieval texts, 46; Vivaldi's influence on, 216
Wagner's, 355 n. 31 Cantata 78 (BWV 78), 176, 190
Appoggiatura, 180 (defined, n. 9) Cantata 198 (BWV198), 276-77
Aquinas, Saint Thomas, 9 Clavieriihung III, 186
Aristotle, 63 Concerto in A Minor for Violin

Arnim, Bettina von, 343 n. 13 (BWV 1041), 178, 183

403
404 INDEX

Bach, Johann Sebastian (cont.) Beethoven, Ludwig van, 7, 14, 117,

Fughetta on "Diess sind die heil'gen 122, 157, 189, 212, 221, 317-20,
zehn Gebot" (BWV 679), 186 367-68; articulation and phrasing,
"Goldberg" Variations (BWV 988), 352; as Classical composer, 295,
176 339-41; dance in, 342; and the
Mass in B Minor (BWV 232), 178, fortepiano, 297-98, 301-2; on
180, 185, 186, 277, 278, 279, Handel, 243; humor in, 349;
282 metronome marks, 346-47; con-
Musical Offering (BWV 1079), trasted with Mozart, 318-19, 325;
184 385 n. 9
n. 19, orchestration, 351, 373; and
Passacaglia in C minor (BWV 582), rhetoric, 343, 360, 388, 389; com-
200 pared to Shakespeare, 344; tem-
Prelude in B Minor (BWV 544), 200 pos, 320, 345-49, 353-54;
Sonatas for Viola daGamba, 177 346-47 n. 15; tempo fluctuation
St. Matthew Passion (BWV 244), and rubato in, 354-55, 360-63
178, 185, 217 Piano Sonata No. 8, "Pathetique,"
Two-Part Invention No. 1 (BWV 306
772), 184 Piano Sonata No. 14, "Moonlight,"
Trio Sonata No 2 in c minor (BWV 214
526), 182 Piano Sonata No. 29, "Ham-
Backhaus, Wilhelm, 334 merklavier," 14, 14 n. 45., 347
Badura-Skoda, Eva and Paul, Second Symphony, 361
304 n. 13 Third Symphony, "Eroica," 348-50,
Bagby, Benjamin, 55 350, 354, 355, 386; "Eroica"
Baillot, Pierre, 369, 374 n. 16 compared to Indiana Jones, 396
Baird, Julianne, 171, 393 Fifth Symphony, 351, 396
Banzo, Eduardo Lopez, x Sixth Symphony, Pastoral, 348, 351
Bar groups in phrasing: 307, n. 16; Seventh Symphony, 351, 356
341-42 Ninth Symphony, 342, 346-48, 351,
Barenboim, Daniel, 7, 13, 380 355, 356 n. 35, 363
Barlines: freedom from in Renaissance bel canto, 233 n 16., 265
choral music, 105-6; importance Bellini, Vincenzo, 252
of in Baroque, 106, 307 Beneventan chant, 27
Baroque era, 183, 197, 197 n. 10, 209, Berg, Alban, 48, 386
228, 229 Bergeron, Katherine, 29 n., 38, 42
Barth, George, 304 n. 13, 306, Berhn Philharmonic Orchestra, 7, 123,
360-63, 388-89 366, 367, 394 n.
Bartlett, Clifford, 6 BerHoz, Hector, 16, 318, 356, chap. 19
Bartok, Bela, 156 n. 34, 251, 309-11, Bernhardt, Sarah, 245, 261
346 n.; Beethoven playing: 360, Bernstein, Leonard, 19, 20
362 Berry, Mary, 26, 29 n.
Bartoli, Cecilia, 5, 7 Biber, Heinrich, 163
Baumol, William and Hilda, 398 n. 12 Billboard, 3, 23, 44
Beck, Jean (duel with Pierre Aubry), 50 Bilson, Malcolm, 13, 320, 322, 360,
John 82
Beckett, 361, 372
Beecham, Sir Thomas, 6, 260, 392 Bimbetta, 396, 399
 INDEX 405

Binchois, Gilles, 100, 163 Byrd, William, 103-4, 112, 258

Binkley, Thomas, 71, 101, 400 Byzantine chant, influences on West,
Birnbaum, J. A., 385 n. 8 27, 33-34, 36
Bizet, George, 364
Boccaccio, Giovanni, 51 Caballe, Montserrat, 137, 138
Boccherini, Luigi, 12, 216, 218 Caccini, Giulio, 145
Bodky, Erwin, 6 n. 12, 174 Cahill, Sarah, 298
Boethius, 36, 66 Callas, Maria, 226, 227, 235, 388
Bordoni, Faustina, 239-40 Campion, Thomas, 239
Botstein, Leon, 396 Cardine, Dom Eugene, 49
Botticelli, Sandro, 129 n. 22 Carlos, Wendy, 6
Boulanger, Nadia, 316, 333 Carolingian empire, 26, 30
Boulez, Pierre, 5 n. 10, 250 Carruthers, Mary, 58
Bows and bowing, 218-19, 351-52 Carter, Elliot, 316, 380, 386
Bowers, Roger, 51 n. 8, 97-98, 101, Carter, Tim, 140 nn. 11, 12, 144
107 n. Casals, Pablo, 173, 207, 224, 260
Boy's choruses: in Renaissance music, Castrates, 244-45
97, 124; in Bach, 282-83 Cavalli, Francesco, 140
Brain-cell specialization, 92 Chance, Michael, 266
Brahms, Johannes, 178, 210, 211, 221, Charlemagne, 30, 31 n. 10
296, 320, 382; cadenzas for Charpentier, Marc-Antoine, 268,
Mozart, 329-30; phrasmg, 307-8, 271-72
372-73; orchestral sonority, 373 Chaucer, Geoffrey, 75, 82
Capriccio, Op. 76. No. 2, 307 Chicago Symphony Orchestra, 367,
Ein deutsches Requiem, 13 380
Intermezzo Op. 119, No. 1, 307-8, Chopin, Frederic Francois, 133-34,
372 258, 307-9
Symphony No. Op. 68, 373
1, Christie, William, 6, 171, 394
Brendel, Alfred, 6, 194, 298-302, 306 Cimarosa, Domenico, 140
Britten, Benjamin, 364 Clemens non Papa, 127, 128
Brown, Clive, 340 nn. 3, 6; 347 n. 16; Clerkes of Oxenford, 100
361 Collins, Judy, 378
Brown, Howard Mayer, 12, 72, 86, Conductors: in Baroque music,
128 220-21, 241, 253, 269, 283-84;
Brueggen, Frans, 193, 203, 400 in opera;228, 234 n. 19; in
Bryars, Gavin, 7 Mozart concertos, 327; in Renais-
Burdon, Eric, 133 sance music, 101-2; "star," 366,
Burmeister, Joachim, 384 367
Burnham, 356 n. 35
Scott, Consonance and dissonance, 91, 93,
Busch, Adolf, 221 179-80, 185
Busnoys, Antoine, 378 consonancias y redobles, 161
Butt, John, 16, 20, 387, 393, 399 Continuo, 162, 168-69, 269; defined,
Buxtehude, Dietrich, 164, 175, 135 n. 6; in Monteverdi, 135 n. 7;

387 n. 12 in Bach, 280; in Mozart, 326-28;

Bylsma, Anner, ix, 6, 12, 171, 193, in Beethoven, 336
203, 393 Copland, Aaron, 316
406 INDEX

Corelli, Arcangelo, 212, 214, 394 Dulak, Michelle, 6, 20, 391, 393
Corri, Domenico, 245-46, 248 Dunstaple, John, 100
Corsican polyphony, 39 Duport brothers, 212
Countertenors, 244-45, 266 Du Pre, Jacqueline, 207
Couperin, Francois, 193-94, 264 "Dutch" Baroque style, 171, 193-95,
Cross-stringing on pianos, 300-301 203^, 209
Crutchfield, Will, 225, 226, 235 n. 20, Dyer, Richard, 123
361 n. 49
Cuzzoni, Francesca, 245 Early Music, 142
Czech Philharmonic Orchestra, 366, "Early-music" Baroque singing style,

367 225-26, 245, 265-66, 277-78,

Czerny, Carl, 304, 306, 339, 347, 285-86
362 Edward,
Elgar, Sir 10, 309
English singing: in Bach, 282, 288; in
Da capo arias, 237, 244 Baroque music, 137, 265-66; ver-
Dahlhaus, Carl, 25 n. 1, 118 n. 3, me-
sus Continental singing in
343 n. 13, 364 dieval and Renaissance music,
Dallapiccola, Luigi, 208 41-42, 84-85, 99, 100-101, 113,
Dance: Baroque, 181, 251, 260; in 122-24; in Monteverdi 142-43,
Bach, 181; in Beethoven 342; in 147-49
Classical music, 342—43 n. 10; Re- Ensemble Clement Janequin, 113
naissance, 158 Ensemble Organum, 45
Danuser, Hermann, 5 n. 10, 379 n. 1 Ensemble Project Ars Nova, 72
Dean, Winton, 156 Erichson, Wolf, 6 n. 11, 172
Debussy, Claude, 190, 221, 311, 367 Esswood, Paul, 245
de la Croix, A. Pherotee, 260
de los Angeles, Victoria, 235 Fach system, 228
De Lay, Dorothy,296 Facial expressions, universals in, 90
Del Mar, Jonathan, 347 n. 16 Fallow^s, David, 101, 355 n. 31
Des Prez, Josquin. See Josquin Des Farinelli, 136
Prez Fenlon, Iain, 141, 143, 148 n.

D'Indy, Vincent, 136 Ficino, Marsilio, 152

Dissonance. See Consonance and disso- Fingering, early, 199

nance Fischer, Edwin, 319

Dixon, Graham, 130 Fischer-Dieskau, Dietrich, 137, 235

Dobszay, Laszlo, 41 Fitch, Fabrice, 41-42

Dowland, John, 149, 239 Fondation Royaumont, 30
Downbow, rule of the, 213, 352, Four-bar phrases. See Bar groups in
352 n. 26 phrasing
Dresden Staatskapelle, 366 Frederick the Great, 229, 240
Dreyfus, Laurence, 19, 21, 193, 195, Freni, Mirella, 226, 227
280, 400 Frescobaldi, Girolamo, 167, 216
Dronke, Peter, 58, 63, 66 Fretwork, 7
Dufay, Guillaume, 8, 9, 10, 21, Fuller, David, 262, 267, 381
100-101, 104, 163; choral bal- Furtwangler, Wilhelm, 14, 19, 207,
ance, 103 221 n., 319, 332, 389, 392
 INDEX 407

Gabrieli, Andrea, 129 Griffiths, Paul, 253

Gabrieli, Giovanni, 129, 216 Grocheio, Johannes de, 74
Galamian, Ivan, 221, 296 Guido d'Arezzo, 37
Garbarek, Jan, 7 Guildhall School of Music, 101, 266
Garcia, Manuel II, 233 n. 16, 234
Gardiner, John Eliot, 6, 7, 172, 288, Haar, James, 126 n., 162
306, 309, 396 Habeneck, Frangois-Antoine, 317,
Geminiani, Francesco, 210, 213, 214, 370
352 n. 26 Handel, George Frideric, 7, 15, 117,
Genetic determinism, 92 156, 173, 174, 212, 239-40, 260
Genius, concept in music, 208 n. 5 Hansen, Gerald, 280
German Baroque singing style, 285-87 Harding, fitienne, 32
Gesture, Baroque, 253, 270 Harmonia Mundi-USA, 44
Gesualdo, Carlo, 126, 127 Harnoncourt, Nikolaus, ix, 5, 7,

Ghezzi, Pier Leone, 216 17 n. 56, 171; on authenticity, 19;

Gibbons, Orlando, 129 Bach choral recordings, 180, 275,
Gimell, 160 286-88; Beethoven symphony
Glottal articulation, 231-34 recordings, 367; on Monteverdi
Gluck, Christoph Willibald, 129, 140, opera 135-37; on period instru-
243 ments, 12, 395; on rhetoric in pre-
Goebel, Reinhard, x, 171, 395 Romantic music, 383-85
Goehr, Lydia, 4, n. 5, 189 n., 277 n. 5, Harris, Jonathan, 23
337-38, 393 n. 3 Harrison, Frank LI., 51 n. 8

Goethe Johann Wolfgang von, 20, 57, Harrison, Lou, 173

173 Haskell, Harry, x
Gombert, Nicolas, 127 Hasse, Johann Adolph, 240, 285
Gone with the Wind, 324 Haydn, Joseph, 157, 295, 317, 320,
Gonzaga family, 146, 148, 151-52 349, 350, 356, 379-80, 386
"Good" and "bad" beats. See Metrical Hefling, Stephen, 263 n. 13
hierarchy Heifetz, Jascha, 332, 334
Gorecki, Henryk, 124 Heilbroner, Robert, 16
Gothic Voices, 54 Hellauer, Susan, 9, 26, 88,
Gould, Glenn, 5, 5 n. 10, 372 Hildegard von Bingen, 9, 15, 21, 48,
Gramophone, 7, 97, 141, 210, 257 51, 88, chap. 3
Grandi, Alessandro, 145 Ave Maria, O auctrix vite, 64
Graun, Johann Gottlieb, 284 O virtus sapentiae, 64
Graun, Karl Heinrich, 240, 269 Ordo Virtutum, 65
Greenberg, Noah, 71 Hiley, David, 26
Greenlee, Robert, 230 Hilliard Ensemble, 7, 396
Gregorian chant, 23-32, 39, 48, Hilliard, Nicholas, 101
60-65; as ritual or meditation, 24, Hillier, Paul, 7, 393
27, 38; rhythm, 49 n. 5; modes, Hindemith, Paul, 180, 225
60-62, 80-81; expressiveness in Hoffman, E.T.A., 343 n. 13
singing, 49, 79-81 Hogwood, Christopher, ix, 7, 194,
Gregory I, Pope, 25-26, 28 276, 340
Gregory II, Pope, 26 n. 4 Holland, Bernard, 4, 20 n. 67
408 INDEX

Hopkins, Antony, 334 Juilliard School of Music, 172, 276

Horowitz, Vladimir, 5, 260, 386 Jutta of Spondheim, 58
Huberman, Bronislaw, 221
Huelgas Ensemble, 113 Kagan, Susan, 313
Huggett, Monica, ix, 171, 296, 308 n. Karajan, Herbert von, 16, 303, 341,
Hummel, Johann Nepomuk, 347 343, 366
Hunt, Lorraine, 248 43
Keillor, Garrison,

Hunter, Chris, 97 n. 3 Hans, 207, 212

Keller, n. 13
Kenyon, Nicholas, 379
Ictus, 49 Kerman, Joseph, 4, 26, 197 n. 10,
Immortal Beloved, 396 275-76
Improvisation, 165-69, 246-47, Keyrouz, Sister Marie, 35, 41
328-333, 344 Khatchaturian, Aram, 221
Instruments, 4 n. 5, 12: in Berlioz, King's Chapel Choir, 123
364, 368-71; French orchestral, King's Singers, 149-50
367; in medieval music 65, 71-72, Kipling, Rudyard, 84, 86
163; in Monteverdi opera, Kirkby, Emma, 146, 148 n.
135-36; period, in Bach, 175, Kirkendale, Ursula, 385
177, 195, 282, 371-72; period, in Kirkpatrick, Ralph, 257
Beethoven, 351, 358; period, in Kirnberger, Johann Philipp, 188 n.,

Romantic music, 364-65, 284

368-372, 372 n. 12; orchestral, Kivy, Peter, 13
366-68; in Renaissance music, Knapp, Raymond, 6 n. 15
chap. 8 Knighton, Tess, 124
Intervals, 50-51; appeal of natural, 93; Kohsch, Rudolf, 349 n. 20, 386
in Milanese chant, 35 Koopman, Ton, 141, 194, 276, 396
Intonation, 51, 78, 83, 107-8 Kovacic, Ernst, 7
Isaac, Heinrich, 124 Kreisler, Fritz, 6, 210, 221, 389
Ison (drone) singing, 36-37 Krueger, Chris, 386
ItalianBaroque singing, 137-38, 143, Kuijken, brothers, 193, 203
230-41, 265; compared to Ger- Kuijken, Sigiswald, 395, 400
man singing, 285-87 Kurtzman, Jeffrey, 142, 145

Jacobs, Rene, 136 Landowska, Wanda, 194, 260

Jacques of Liege, 83 Lanier, Nicholas, 239
Jander, Owen, 233 n. 16 Lassus, Orlandus, 104, 127-28
Jazz, 35, 157, 246, 268, 317, 328, Lawrence, Kay, 342
394, 396 Lawrence-King, Andrew, 144 n. 20,
Jefferson, Thomas, 9 315, 393
Jerome of Moravia, 35, 40 Lehar, Franz, 365
Joachim, Joseph, 211, 296 n. 4, Leibniz, Gottfried, 187, 187 n.
307 n. 17 Leigh, Vivian, 324
Johnson, Robert, 239 Leningrad Philharmonic Orchestra,
Jolly, James, 7 366
Josquin Des Prez, 21, 73, 100-101, Leonhardt, Gustav, 17 n. 56, 134, 180,
111 n., 129, 378 276, 281, 286-88, 394, 400
 INDEX 409

Leonin, 74 Meter in Renaissance choral music,

Leppard, Raymond, 136 105
Levin, Robert, 7, 13, 393, 395, 397 Metrical hierarchy: in Baroque and
Lipatti, Dinu, 316 Bach, 179-81, 185, 196 n.,

Liszt, Franz, 307-8, 345, 372 nn. 12, 279 n. 7; in Beethoven, 353
13, 386 Metronome marks, 346—47, n. 15
Literary imagery in music analysis, Metropolitan Opera, 234, 305
356 n. 35 Microtones, 36
Lombardic rhythm, 214 Milan, Luis de, 161
Lowinsky, Edward, 73 n., 187 n., Milanese chant. See Ambrosian chant
188 n., 208 n. 5 Miller, Glenn, 217
Lubin, Steven, 9 Minter, Drew, 109, 245
Luduvico, 161, 166 Mintz, Donald, 355 n. 31, 379 n. 1

Lully, Jean-Baptiste, 213, 258, 261, Mizler Lorenz, 187 n., 188 n.

271-72, 352 n. 26, 394 Modal music, 62. See also "Gregorian
Lutheranism, 185-88, 286-87 Chant"
Lyon, Opera de, 364, 368 Moens-Haenen, Greta, 234-35 n. 20
Moliere, 268, 273, 388
Ma, Yo-Yo, 7, 207 Mondonville, Jean-Joseph Cassanea de,
Machaut, Guillaume de, 30, 57, 78, 82 260
Mackerras, Sir Charles, 6 n. 17, 7, 357 Monteverdi, Claudio, 8, 129, 168,
Macque, Giovanni de, 162 171, 230
Mahler, Gustav, x, 364, 368, 370, 371, Montpellier Codex, 45, 46
373 Montpellier Treatise, 36
Mara, Madame, 237 Moody, Ivan, 134
Marais, Marin, 214 Moore, Tom, 226
Marenzio, Luca, 145 Morgan, Michael, 6 n. 17
Marian worship, 45, 56 Morgan, Robert, 15, 17, 392
Marshall, Robert, 176, 184, Morrow, Michael, 82, 400
288-89 n. 24, 362 Moscheles, Ignaz, 361-62
Martin, Robert, 13 Mozarabic chant, 27
Mattheson, Johann, 384 Mozart, Leopold, 213, 299-300, 308
May, Florence, 307 n. 17 Mozart, Wolfgang Amadeus, 9, 10, 12,
McCreesh, Paul, 142 21, 24, 140, 158, 189, 260, 350,
McGegan, Nicholas 366, 391, 393, 356, 380; characteristics of keys,
400 321-323; as "classical," 295; con-
McKinnon, James 24 tinuo in, 326-28; on the fortepi-
Mead, Margaret, 89 ano, 299-301; dance in, 342; on
Medieval culture, world view, charac- Handel 243; improvised cadenzas
teristics, etc., 24, 38-41, 47, and ornaments in, 328-30,
56-58, 63, 66-67 336-38; legato in, 299, 303-305,
Mendelssohn, Felix, 308, 354-55, 325-26; rubato in 238, 309-10; as
355 n. 31 theory teacher 332-33; "topics"
Menuhin, Yehudi, Lord, 6 n. 17, 363 in, 322-23
messa di voce, 220, 231, 246, 284 Piano Concerto No. 24 in C minor
Metastasio, Pietro, 230 (K. 491), 328
410 INDEX

Mozart, Wolfgang Amadeus (cont.) Ordines Romani, 37

Piano Sonata in A (K. 310), 299 Organum, 39, 74, 75; defined, 37
Piano Sonata in F (K. 332), 303, "Original" instruments. See Instru-
305 ments
Piano Sonata in B^ (K. 333), 301 Ornamentation, 35, 214, 284: in plain-
Requiem (K. 626), completions, 336 chant, 29, 31-32, 35-36; in Re-
Symphonie Concertante in Bl> for naissance sacred music, 108-111;
Four Winds, 323-24 in Bach 183-85; in Baroque arias,
Die Zauberflote (K. 620), 260, 267, 237-38; in Handel, 247-48; in
317 French Baroque; 259; in Mozart,
Mravinsky, Yevgeny, 366 329 n. 16, 336-37
Mudarra, Alonso, 21, 161-62, 166, Overdotting, 213-14, 250-51, 268
167 "Oxbridge" style, 82-86, 100-102,
Muffat, Georg, 213, 269 113. See also English singing
Munrow, David, 71, 75, 400 Oxford Camerata, 100
Musica (instrumentalis, humana, mun-
dana), 65, 66, 188 n. Paderewski, Ignace Jan, 194
musica ficta, 101, 104 Paganini, Niccolo, 369
Musica Reservata, 82 Page, Christopher, 10, 24, 40 n. 23,
Mutter, Anne-Sophie, 207 51 n. 7, 97, 100, 394
Palestrina, Giovanni Pierluigi da, 21,
Nelson, Judith, ix 28, 106, 111-19, 124-131; orna-
Netherlands. See "Dutch Baroque mentation, 109; ensemble configu-
style" ration, 130
Neumann, Frederick, 263 n. 13, Missa Papae Marcelli, 111, 125,
336-38 126, 343
Neumes, 31, 59; defined, 31 n. 8 Parker, Dorothy, 385
New York Times, 43, 257, 394 n. Parrott, Andrew, 276, 391-93, 395
New Yorker, 397 Part, Arvo, 7, 44, 101, 124
Newberry Consort, 72 Patti, Adelina, 235 n. 20
Nilsson, Birgit, 137 Peres, Marcel, 18, 82, 88
Nord deutsches Rundfunk (NDR) Sin- Pergolesi, Giovanni Battista, 5
fonie-orchester, 364, 371 Peri, Jacopo, 145
Norman, Jessye, 225 Period instruments. See Instruments
Norrington, Roger, 6 n. 15, 14, 172, Perlman, Itzhak, 7, 296
276, 280 Perlman, Marc, 19 n. 63
Notation: medieval, 31-32; Renais- Perotin, 46, 74, 114
sance, 105-6 Petrucci, Ottaviano dei, 160, 163
notes inegales, 213-14, 262-64, 381 Philharmonia Orchestra, 368
Notre Dame polyphony, 39, 50 n. 6 Philip, Robert, 8, 250 n. 7, 251 n. 9,

310 n. 21, 331, 361-63, 392

Obrecht, Jacob, 76 Phillips, Peter, 101, 392, 394, 395
Ockeghem, Johannes, 100-101 Philpot, Margaret, 82
Old Roman chant, 27, 31-33, 36 Phobias, 91
Oral tradition and oral culture in Mid- Pietism, 186-88
dle Ages, 58-63, 61 n. 12 Pinnock, Trevor, 6 n. 15
 INDEX 411

Pius X, Pope, 26 Rhythmic modes, 39 n. 20, 50

Planchart, Alejandro Enrique, 121 n. Ries, Ferdinand, 14 n. 45, 339,
Plato, 58, 66 362 n. 54, 363
Portamento, 149, 212, 232-33, 311, Rifkin, Joshua, 11, 172, 279, 283,
361, 374 288-90
Porter, Andrew, 69, 297 Ritornello, (defined) 216
postmodernism, 93 n. 45, 399-400 RodeA^iotti school of violin playing,
post-structuralism, 77 361
Pothier, Dom Joseph, 49 Romantic love, 94
Praetorious, Michael, 286 Romberg, Bernhard, 212, 361 n. 47
Prepared learning, 91 Rooley, Anthony, 395, 400
Pro Cantione Antiqua, 100 Rosen, Charles, 21, 208, 356 n. 35; on
Pronunciation, period, 111-12, 139, musicology, 3; on Bach keyboard
254, 262 performance 11, 195, 202; on
Puccini, Giachomo, 236 continue in Classical music,
Purcell, Henry, 149, 263 326-28; on the emergence of mu-
sical works as independent objects,
Quantz, Johann Joachim, 214, 241 316 n. 1, 338 n.; on the fortepi-
ano, 306, 314; on four-bar
Rachmaninoff, Sergei, 5, 305, 309, 364 phrases, 307 n. 16, 341 n.; on the
Racine, Jean, 261 history of "genius," 208 n. 5; on
Rameau, Jean-PhiUppe, 12, 135, 260, humor in Beethoven, 349,
265 349 n. 21; on music ficta 104-5;
L'Impatience, 259 on opera seria characters
Ratner, Leonard, 322, 342, 342 n. 10,
253 n. 16; on tempo flexibility
343 n. 11
in Beethoven, 355, 363
Rattle, Sir Simon, 6 n. 17 Rosen, Sherwin, 397
Ravel, Maurice, 217 Rosenblum, Sandra, 300
Recording, 8-9, 200, 248-50, 330-32 Rossi, Luigi, 165
Reformation, 67 Rossini, Gioacchino, 147, 252
Reger, Max, 382, 387 Rothstein, Edward, 280, 395 n. 6
Registration of keyboard instruments, Rousseau, Jean-Jacques, 188, 208 n. 5,
200 266
Rembrandt van Rijn, 4 Rousset, Christophe, ix, 194, 195, 257,
Renaissance: musical characteristics of,
263, 297
73, 81, 98, 126, 144 n. 20, 162, Rubato, 209-10, 238, 313; in Baroque
183; "Twelfth-century" 56 performance, 238, 248, 269, 381;
Respighi, Ottorino, 136 in Beethoven 339, 355, 360-63, in
Rhetoric and "speaking" in music: in
Mozart, 238, 309-10
Bach, 182-83, 282, 387 n. 12; in
Rubinstein, Artur, 319
Baroque music, 182-83, 196, Rudolf, Max, 307 n. 17, 348 n. 19
219-21, 258-62 278, 383-89; m Rushton, Julian, 365 n. 2, 371,
Beethoven, 343, 360, 388, 389; in
372 n. 12
Classical music, 305
Rhythm, quantitative versus qualitative, Sacks, Oliver, 54
39-40, 93 Sadie, Stanley, 6 n. 15
412 INDEX

Salieri, Antonio, 361 nn. 46, 47 Somfai, Laszlo, 251

San Francisco Opera, 234, 276 Sonata, Classical, 344-45, n. 14
Savall, Jordi, ix, 7, 17 n. 56, 87, 396, spiccato, 351 n. 25
397 Spinoza, Baruch de, 187-88
Scarlatti, Alessandro, 136 Spohr, Ludwig, 351
Scarlatti,Domenico, 260 sprezzatura, 165
Scheibe, Johann Adolph, 188, 385 n. 8 Springfels, Mary, ix

Schering, Arnold, 385 Stadlen, Peter, 346 n. 20

Schiff, Andras, 258 Status, biochemistry and signahng of,

Schindler, Anton, 339, 361, 363 91

Schmidt, Arnold, 385 Stevens, Wallace, 398
Schnabel, Artur, 11 n. 33, 303-5, 319, Strauss, Johann II, 8

330, 332, 389, 400 Strauss, Richard, 368

Schoenberg, Arnold, 189, 320 Stravinsky, Igor, 17 n. 56, 180, 189,
Schola Cantorum Basiliensis, 55, 193, 190, 215, 346 n., 378, 386
194 Streisand, Barbra, 252
Schott, Howard, 194 Strings, metal versus gut, 210,
Schubert, Franz, 307, 324; and the 217-18
fortepiano, 302, 304 Strohm, Reinhard, 158 n.

"Wanderer" Fantasy, 302, 310 Summa Musice, 37, 81

Impromptu, Op. 142, No. 3, 302, Suspension, defined, 122 n.

304 Sutherland, Dame Joan, 7

Moment Musical No. 3, 307 Szell, George, 207, 332
Schulenberg, David, 387 n. 12 Szigeti, Josef, 207 n. 1, 221, 334
Schumann, Clara, 330
Schumann, Robert, 306, 319 Tactus, 121

Schijtz, Heinrich, 286, 378 Talbot, Michael, 217

Schwarzkopf, Elizabeth, 235 Talhs Scholars, 7, 98, 101
seconda prattica, 144 n. 20, 145 Tallis, Thomas, 73
Selfridge-Field, Eleanor, 216 Taruskin, Richard, 21, 26, 86-87, 258,

"Separability principle," p. 277 n. 5 334; on authenticity 17-19, 319,

Sequence (in Baroque and later music), 392-93; on Bach, 190, 287; on
217 composers' intentions, 13; on
Seyfried, Ignaz von, 340 n. 7; 361 early-music Beethoven, 340-41,

Sherr, Richard, 119 130

n., 362; on Mozart cadenzas, 337; on
Sibelius, Jan, 207, 332 performance of Renaissance
Silos, Benedictine monks of Santo masses, 120
Domingo de, 7, 23, 25, 27, 39 Tartini, Giuseppe, 352 n. 26
Singing. See under: English singing. Tavener, John 44, 124

Early Music singing; Itahan Taverner Choir, 100

Baroque singing; German Baroque Tchaikovsky, Piotr Ilich, 246, 366
singing Tebaldi, Renata, 225
Sixteen, the, 100 Telemann, Georg Philipp, 164, 269
Solesmes, 25-29, 49 Temperament: unequal temperament,
Solomon, Maynard, 339, 350, 355, 199-200, 322; Pythagorean, 83,
357 n. 37 107; meantone, 107
 INDEX 413

Tempo, 112, 122, 292; in Beethoven, van der Werf, Hendrik, 39 n. 20

320, 345-49; proportional, Van Tassel, Eric, 104 n.

121-22 Vartolo, Sergio, 125

"terza prattica" ("third practice"), Verdi, Gmseppe, 3, 129, 143, 147,
140 n. 12 156, 227-29, 246, 251-52, 370,
Text expression in music: in French 395
Baroque music 258-60, 264; in Vibrato, 104, 143, 149, 210-12, 296,
Gregorian chant 49, 79-81; in the 311; in Bach, 236; in Baroque
Italian Baroque 230; in Lutheran singing, 225-26, 234-36, 266-67;
Baroque church music, 286-87; in in Beethoven, 212, 351; in Berlioz,
medieval music, 62-63, 79, 81-82, 374; in Brahms, 211, 296, 374;
86-87; in Monteverdi and his era, studies on, 234-35 n. 20,
137, 140, 143^4; in Palestrina 236 n. 24
and the late Renaissance, 129; in Victoria, Tomas Luis de, 104,
seventeenth-century English music, 128-30
239 Vienna Philharmonic Orchestra, 7,
Thomas, Jeffrey, 392 364, 367, 376
Thomson, Virgil, 14 Viotti, Giovanni Battista, 361
Tiepolo, Giovanni Battista, 216 Vitry, Philippe de, 158
Timaeus, 66 Vivaldi, Antonio, 208, 214-17
Tomlinson, Gary 140 Concerto in G 215for Cello,
"Topics" in Classical music, 322-23, Cello Sonata in E minor, 216
342-43, 343 n. 11 von Stade, Frederica, 233
Toscanini, Arturo, 221 n., 341, 343, Vuillaume, Jean-Baptiste, 369
346, 357, 386
Tosi, Pier-Francesco, 227, 231-33, 237, Wagner, Richard, 48, 129, 236, 252,
238 345, 355 n. 31, 372, 373
Tourte bow, 218-19, 311 Walcha, Helmut, 197
Tous les matins du monde, 396 Walter, Bruno, 319
Tovey, Sir Donald Francis, 11, Walther, J. G., 185
342 n. 9 Wasielewski, Joseph Wilhelm von,
Trabaci, Giovanni, 162 355 n. 31
"Transhistorical humanness," 76, Weber, Carl Maria von, 356
88-95 Weber, Jerome F., 41
Transposition, 97, 106-7 Wegman, Rob C, 88-90, 94
Treitler, Leo, 20, 33 n. 11, 69 n. 12, Weill, Kurt, 364
62 Werktreue (fidelity to the work), 318,
Trill, 233 337-38, 393
Troubadors, 50 n. 6 Wilde, Oscar, 198 n.

Tuning. See Intonation, Temperament Williams, Peter, 264 n., 385 n. 9

Turk, Daniel Gottlob, 303, 308, 360 Wistreich, Richard, 142

Wolff, Christoph, 173, 336-38
Under Milk Wood, 1 Wolkenstein, Oswald von, 57
Unger, Hermann, 385 Women: In early-music movement, vii;

Universals, human, 89-92 represented in Hildegard's works,

Urtext, 229, 305 67-68; singing Bach, 282-83;
414 INDEX

Women {cont.) Wright, Robert, 92

singing Baroque music, 238,
244-45; singing medieval music, Youngren, William H., 207
51 n. 8, 52-53; singing Renais-
sance music, 97, 124; in Wagner, Zander, Ben, 347
252 Zaslaw, Neil, 8 n. 22, 326 n. 11,
"Work" concept, 4 n. 5, 189 n., 331 n. 17
337-38, 393 Zinman, David, 6, 276, 357
Wray, Alison, 139 n. 10 Zukerman, Pinchas, 7
Wright, Craig, 71, 74 Zwang, Gerard, 5
 |;ij:,[!JN IHJBLIC LIBR/UV

3 9999 03131 037 6

convey not only a devotion to the spirit

of period performance, but the joy of

discovery as they struggle to bring the
music most fully to life. Spurred on by
Shermans probing questions and
immense knowledge of the subject,
these conversations movingly docu-
ment the aspirations, growing pains,
and emerging maturity of one of the
most exciting movements in contem-
porary classical performance, allowing
each artist's personality and love for
his or her craft to shine through.

From medieval plainchant to

Brahms's orchestral works. Inside Early
Music takes readers — whether enthu-
siasts or detractors — behind the scenes
to provide a masterful portrait of early
music's controversies, challenges, and
rewards.

Bernard D. Sherman has written

about early music for Early Music,
Schwann/Opus, Fanfare, Historical
Performance, Early Music America,
Strings, Early Music News, and The
American Brahms Society Newsletter,
among other journals and magazines.
His essay on '"Authenticity in Musical
Performance" will appear in the forth-
coming Encyclopedia of Aesthetics
(Oxford University Press, 1998).
 adoa/ice azaUe wa,

inside early music

'Inside EarlyMusic is a fascinating book, and not just for readers with an interest
in Sherman's pointed interviews with opinionated people, and his
early music.
lucid introductions and postscripts, offer considerable insight into questions
about the nature of art, the nature of human nature, and some of the great
intellectual controversies of our time."
— Steven Pinker, Massachusetts Institute of Technology,
author of The Language Instinct and How the Mind Works
'I can't imagine a better book of its kind.... Readers will profit greatly and they
are addressed considerately and without condescension."
— Richard Taruskin, University of California, Berkeley,
author of Text and Act and Stravinsky and the Russian Traditions

'Bernard Sherman is one of the shrewdest, best informed, and most sensible
commentators on the early music scene. This book should be required read-
ing for anyone concerned with issues of 'historically informed' performance."
—
Walter Frisch, Columbia University,
author of Brahms and the Principle of Developing Variation and
Brahms: The Four Symphonies

'Sherman is no 'invisible' interviewer — he's

concerned with the historical
actively
record, with traditions, and with distinguishingbetween matters of personal
taste and objective judgment. He invites these artists to speak of their deep-
est musical convictions even as he challenges them with competing points of
view (and not only those they are accustomed to hearing). As a result, many
arguments that seemed merely polemical in the popular press are shown to
be subtle, urgent, and deep. The associated discographies highlight some of
the most exciting and moving performances now available, and there's a store-
house of wisdom in Sherman's discussions of 'further reading.' This is a
superb achievement, brilliantly done."
—
George Barth, Stanford University,
author of The Pianist as Orator

'An important contribution to our understanding of the early music phenomenon.

On topics ranging from Hildegard to Brahms, Sherman knows whom to ask and
what to ask them."
— Stewart Carter, Wake Forest University,
editor of A Performer's Guide to Seventeenth-Century Music and
The Historic Brass Society Journal, former editor of Historical Performance

90000

9 780195"097085