NOTES BECOME MUSIC

Notes Become Music: A Guidebook from the Viennese Piano Tradition addresses the
many unwritten nuances of dynamics, articulation and agogics as an expression
of fundamental principles of a common European musical language. It treats the
score as an incomplete musical shorthand that outlines the compositional and
interpretive imperatives implicit within it, drawing on historical records from
the eighteenth and nineteenth centuries and detailed comparisons of works to
underline the author’s presentation of Viennese tradition.
This book is not primarily concerned with questions of style or interpretation.
Rather, it explains the many facets of musical notation that were taken for
granted by composers who assumed a knowledge of the piano tradition of
their day. Notes Become Music informs not only those students in countries
where the central European music tradition is still unfamiliar, but also a younger
generation of Europeans who have grown up without a living connection to
their musical past.

Walter Fleischmann is Professor Emeritus of Piano at the University of Music

and Performing Arts, Vienna.
“Despite its modest length, Notes Become Music conveys fundamental insights into
how the works of the masters may be played both correctly and beautifully in
a way that can move the listener. The study is concerned with nuances of our
musical language that are often neglected or overlooked in piano tuition today.
Mr. Fleischmann’s insights are the fruit of decades of experience in piano teach-
ing and performing. They offer valuable ideas not just to students but also to
teachers and will help to curb the trend away from living musicianship towards
music as marketing and show-business.”
—Paul Badura-Skoda, pianist and conductor

“This book is valuable and stimulating in a variety of ways. In it the author

demonstrates many basic principles for the interpretation of piano music that
should not be neglected in a so-called free style of playing but which may also
serve to liberate players from what are for the most part acquired constraints.
This is particularly important for young players who are struggling to develop
their own artistic personalities and for those who perhaps have been guided by
their teachers into a sense of mistaken faithfulness to the original, or who have
been primarily formed by the ideas of the teachers themselves. In addition, the
explanations in this book may prevent many a piano student who is not yet
fully mature from attempting to take eccentric celebrities—whose playing should
on no account be imitated—as role models. Great piano music has lived for
centuries not only on the manuscript paper upon which it was notated by the
masters, but also in the changing styles of interpretation over the years. Walter
Fleischmann’s intriguing work points out limits that serious pianists should not
transgress, at the same time opening up a wealth of possibilities for legitimate
interpretation for which there are no traditional means of notation.”
—Alexander Jenner, Professor Emeritus,
University of Music and Performing Arts, Vienna
NOTES BECOME MUSIC
A Guidebook from the Viennese
Piano Tradition

Walter Fleischmann
UNIVERSITY OF MUSIC AND PERFORMING ARTS, VIENNA

TRANSLATED BY ROBERT J. CROW

English translation published 2020
by Routledge
52 Vanderbilt Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2020 Taylor & Francis
The right of Walter Fleischmann to be identified as author of this work
has been asserted by him in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission in
writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
Originally published as Aus Klavierspiel wird Musik: Prinzipien und
Gesetzmäßigkeiten der Musik und die Grenzen der musikalischen Notation: Ein
Ratgeber aus der Wiener Pianistentradition für Lehrende und Studierende
© STACCATO Verlag, 2017
Library of Congress Cataloging-in-Publication Data
Names: Fleischmann, Walter, author. | Crow, Robert J., translator.
Title: Notes become music : a guidebook from the Viennese piano
tradition / by Walter Fleischmann ; translated by Robert J. Crow.
Other titles: Aus Klavierspiel wird Musik. English
Description: New York ; London : Routledge, 2019. | Includes index.
Identifiers: LCCN 2019007002 (print) | LCCN 2019012044 (ebook) |
ISBN 9780429260261 (ebook) | ISBN 9780367202255 (hardback) |
ISBN 9780367202262 (pbk.)
Subjects: LCSH: Piano music—Interpretation (Phrasing, dynamics, etc.)
Classification: LCC MT235 (ebook) | LCC MT235 .F5813 2019 (print) |
DDC 786.2/1930943613—dc23
LC record available at https://lccn.loc.gov/2019007002
ISBN: 978-0-367-20225-5 (hbk)
ISBN: 978-0-367-20226-2 (pbk)
ISBN: 978-0-429-26026-1 (ebk)
Typeset in Bembo
by Apex CoVantage, LLC

Music notation: Caterina Fassetta, Vienna

CONTENTS

Translator’s Note vi
List of Musical Examples vii
Preface xiii
Acknowledgements xvi

1 Introduction: Musical Principles and the Limits of Notation 1

2 Dynamics 5

3 Articulation 16

4 Agogics 50

5 Pedal 57

6 Notation 72

7 Conclusion 89

Index of Quoted Composers and Works 92

Subject Index 97
TRANSLATOR’S NOTE

For the sake of simplicity and easier reading, categories of individuals are some-
times referred to in this translation using solely the masculine form. This form
is here taken to refer equally to women and men.
Robert J. Crow
MUSICAL EXAMPLES1

Streaming audio files can be accessed here:

www.members.aon.at/walter.fleischmann/book_audios_en.html

1.1 Robert Schumann, Sonata in F-sharp minor op. 11, first

movement, bars 74–84 4
2.1 Pattern of relative emphases, after Leopold Mozart 5
2.2 Franz Schubert, Impromptu in B-flat major op. posth. 142
D 935 no. 3 6
2.3 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 570,
first movement 6
2.4 Franz Schubert, Sonata in B-flat major D 960, third movement 7
2.5 Franz Schubert, Sonata in B-flat major D 960, third movement, Trio 7
2.6 Ludwig van Beethoven, Sonata in F minor op. 57, first movement,
bars 35–39 8
2.7 Graphic representation of crescendo 9
2.8 Johann Sebastian Bach, Das Wohltemperierte Klavier II, Prelude
in F minor, bars 20–24 9
2.9 Dynamic contour, four bars 10
2.10 Dynamic contour, crescendo 10
2.11 Dynamic contour, diminuendo 10
2.12 Dynamic contour, repeated note 11
2.13 Dynamic contour of phrase 11
2.14 Joseph Haydn, Sonata in E minor Hob. XVI/34, second
movement 12
2.15 Ludwig van Beethoven, Sonata in C minor op. 10 no. 1,
second movement 12
viii Musical Examples

2.16 Joseph Haydn, Sonata in E-flat major Hob. XVI/49, second

movement 13
3.1 Franz Liszt, Tarantella, bars 73–75 17
3.2 Ludwig van Beethoven, Sonata in E-flat major op. 31 no. 3,
third movement, Trio 18
3.3 End motive, quarter notes 18
3.4 End motive, eighth notes 18
3.5 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 570,
first movement, bars 11–12 19
3.6 Franz Schubert, Sonata in A minor op. 42 D 845, second
movement 19
3.7 Joseph Haydn, Sonata in E minor Hob. XVI/34, second
movement 20
3.8 Ludwig van Beethoven, Sonata in C Major op. 2 no. 3,
second movement 21
3.9 Johann Sebastian Bach, Das Wohltemperierte Klavier II,
Prelude in E major 21
3.10 Wolfgang Amadeus Mozart, Sonata in A minor KV 310,
second movement, bars 15–19 22
3.11 Wolfgang Amadeus Mozart, Sonata in A minor KV 310,
second movement, bars 64–65 22
3.12 Ludwig van Beethoven, Sonata in A-flat major op. 26, first
movement Var. IV 23
3.13 Ludwig van Beethoven, Sonata in A major op. 2 no. 2,
second movement 24
3.14 Franz Liszt, Valse oubliée 24
3.15 Franz Liszt, Valse oubliée, with exactly notated durations 25
3.16 Wolfgang Amadeus Mozart, Concerto in A major KV 414,
third movement 27
3.17 Franz Schubert, Rondo from Sonata in A major D 959 27
3.18 Ludwig van Beethoven, Sonata in E-flat major op. 7, second
movement, bars 20–21 28
3.19 Ludwig van Beethoven, Sonata in D minor op. 31 no. 2, first
movement, bars 159–169 28
3.20 Ludwig van Beethoven, Sonata in C-sharp minor op. 27 no. 2,
third movement 29
3.21 Sergei Rachmaninoff, Études tableaux op. 39 no. 6 29
3.22 Ludwig van Beethoven, Sonata in E-flat major op. 27 no. 1, first
movement 30
3.23 Ludwig van Beethoven, Sonata in E-flat major op. 27 no. 1,
fourth movement 30
3.24 Ludwig van Beethoven, Sonata in E major op. 109, second
movement, bars 9–25 31
 Musical Examples ix

3.25 Ludwig van Beethoven, Sonata in E-flat major op. 81a, second
movement, bars 13–14 31
3.26 Ludwig van Beethoven, Sonata in E-flat major op. 81a, second
movement, bars 29–30 31
3.27 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 333, first
movement 32
3.28 Wolfgang Amadeus Mozart, Sonata in G major KV 283, first
movement 33
3.29 Joseph Haydn, Sonata in C minor Hob. XVI/20, first movement,
bars 32–35 34
3.30 Joseph Haydn, Sonata in C minor Hob. XVI/20, first movement,
bars 47–51 35
3.31 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 333, third
movement, bar 23 35
3.32 Ludwig van Beethoven, Sonata in A major op. 2 no. 2, first
movement 36
3.33 Ludwig van Beethoven, Sonata in F major op. 10 no. 2, first
movement 36
3.34 Ludwig van Beethoven, Sonata in C major op. 53, first movement 37
3.35 Ludwig van Beethoven, Sonata in C minor op. 13, first movement
bars 51–79 37
3.36 Ludwig van Beethoven, Sonata in E major op. 14 no. 1, first
movement, bars 5–6 38
3.37 Ludwig van Beethoven, Sonata in G major op. 14 no. 2, first
movement 38
3.38 Ludwig van Beethoven, Sonata in E-flat major op. 27 no. 1,
first movement, bars 37–44 38
3.39 Sergei Prokofiev, Visions fugitives no. 5 39
3.40 Maurice Ravel, Sonatina, second movement, bars 15–16 39
3.41 Maurice Ravel, Sonatina, third movement, bars 4–8 39
3.42 Joseph Haydn, Sonata in E-flat major Hob. XVI/34, first movement 40
3.43 Joseph Haydn, Sonata in E-flat major Hob. XVI/52, first
movement, bars 27–29 40
3.44 Wolfgang Amadeus Mozart, Sonata in C minor KV 457, second
movement, bars 21–22 41
3.45 Frédéric Chopin, Scherzo in B-flat minor op. 31 41
3.46 Wolfgang Amadeus Mozart, Sonata in D major KV 576, third
movement 42
3.47 Joseph Haydn, Sonata in G major Hob. XVI/40, first movement,
bars 61–62 43
3.48 Robert Schumann, Kreisleriana, second piece, Intermezzo I 43
3.49 Robert Schumann, Sonata in G minor op. 22, third movement
Scherzo, bars 33–36 44
x Musical Examples

3.50 Ludwig van Beethoven, Sonata in A-flat major op. 110, third
movement, bars 9–12 44
3.51 Ludwig van Beethoven, Sonata in A-flat major op. 110, third
movement, bars 116–119 45
3.52 Ludwig van Beethoven, Sonata in A-flat major op. 110, first
movement, from bar 12 46
3.53 Robert Schumann, Sonata in G minor op. 22, first movement,
bars 93–96 47
3.54 Robert Schumann, Humoreske op. 20 47
3.55 Franz Liszt, Konzert-Etüde no. 3 (Il sospiro) from Drei
Konzert-Etüden 48
4.1 Johann Sebastian Bach, Toccata in E minor BWV 914, Adagio 51
4.2 Johann Sebastian Bach, Sarabande from Partita in B-flat major
BWV 825 52
4.3 Johann Sebastian Bach, Das Wohltemperierte Klavier I, Prelude
in E-flat minor 52
4.4 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 570, first
movement 53
4.5 Johann Sebastian Bach, Das Wohltemperierte Klavier II, Prelude
in F minor 54
4.6 Joseph Haydn, Sonata in C major Hob. XVI/48, second movement 55
4.7 Joseph Haydn, Sonata in D major Hob. XVI/19, third movement 55
4.8 Ludwig van Beethoven, Sonata in A major op. 101, first movement 56
5.1 Frank Martin, Prélude no. 1 from 8 Préludes pour le piano,
bars 35–39 58
5.2 Wolfgang Amadeus Mozart, Sonata in C major KV 545, first
movement 59
5.3 Ludwig van Beethoven, Piano Concerto no. 3 in C minor op. 37,
second movement 59
5.4 Frédéric Chopin, Nocturne in D-flat major op. 27 no. 2 60
5.5 Johannes Brahms, Intermezzo op. 116 no. 4, bars 15–19 60
5.6 Johannes Brahms, Intermezzo from Klavierstücke op. 119 no. 1,
bars 45–47 61
5.7 Johannes Brahms, Intermezzo op. 116 no. 4, bars 36–52 62
5.8 Johannes Brahms, Intermezzo op. 116 no. 4, last five bars 62
5.9 Frédéric Chopin, Allegro de Concert op. 46, bars 182–186 63
5.10 Frédéric Chopin, Rondeau op. 73, bars 103–108 63
5.11 Johannes Brahms, Sonata in F minor op. 5, second movement 64
5.12 Johannes Brahms, Ballade op. 10 no. 1, bars 27–37 64
5.13 Gabriel Fauré, Nocturne op. 37, bars 49–52 65
5.14 Gabriel Fauré Nocturne op. 74, bars 39–42 65
5.15 Robert Schumann, Sonata in F-sharp minor op. 11, fourth
movement, bars 94–101 66
 Musical Examples xi

5.16 Franz Schubert, Impromptu in B-flat major op. posth. 142

D 935 no. 3 66
5.17 Johann Sebastian Bach, Das Wohltemperierte Klavier I, Prelude
in C major 67
5.18 Johann Sebastian Bach, Das Wohltemperierte Klavier I, Prelude
in D minor, bar 2 67
5.19 Johann Sebastian Bach, Das Wohltemperierte Klavier I, Prelude
in C minor, bar 30 68
5.20 Sergei Rachmaninov, Prélude op. 3 no. 2 in C-sharp minor 69
5.21 Robert Schumann, Sonata in G minor op. 22, third movement,
bars 41–53 69
6.1 Frédéric Chopin, Sonata no. 2 in B-flat minor op. 35, fourth
movement, final bar 73
6.2 Franz Schubert, Sonata in B-flat major D 960, second movement 73
6.3 Johannes Brahms, Capriccio op. 76 no. 5, bar 113 to the end 73
6.4 Johannes Brahms, Capriccio op. 76 no. 5 Agitato, bars 71–72 74
6.5 Johannes Brahms, Capriccio op. 116 no. 7, final bars 74
6.6 Johannes Brahms, Rhapsodie op. 119 no. 4, final bars 75
6.7 Franz Schubert, Sonata in B-flat major D 960, first movement,
bars 27–30 75
6.8 Johannes Brahms, Capriccio op. 116 no. 3, final bars 76
6.9 Ludwig van Beethoven, Sonata in C major op. 2 no. 3, first
movement, from bar 222 76
6.10 Robert Schumann, Sonata in F-sharp minor op. 11, fourth
movement, bars 98–101 77
6.11 Hidden part writing, cadence 78
6.12 Hidden part writing, correct execution 78
6.13 Johann Sebastian Bach, Prelude from Partita in B-flat major
BWV 825 78
6.14 Johann Sebastian Bach, Menuett from Partita in B-flat major
BWV 825 78
6.15 Johann Sebastian Bach, Fugue from Toccata in E minor
BWV 914 79
6.16 Johann Sebastian Bach, Fugue from the Toccata in E minor
BWV 914, exact notation 80
6.17 Johann Sebastian Bach, Fugue from Toccata in E minor
BWV 914, inferior execution 80
6.18 François Couperin, Rondeau Les Bergeries, Deuxième livre
de Pièces de Clavecin 80
6.19 Johann Sebastian Bach’s simplified notation of Couperin’s
Rondeau Les Bergeries 80
6.20 Johann Cramer, Etude in C major 81
6.21 Johann Cramer, Etude in D major 81
xii Musical Examples

6.22 Ludwig van Beethoven, Sonata in D major op. 10 no. 3, first

movement from bar 31 82
6.23 Ludwig van Beethoven, Sonata in F-sharp major op. 78, first
movement from bar 19 with upbeat 82
6.24 Ludwig van Beethoven, Sonata in F-sharp major op. 78, first
movement from bar 19 with upbeat, cantilena 82
6.25 Wolfgang Amadeus Mozart, Sonata in B-flat major KV 333, first
movement, bars 18–20 82
6.26 Joseph Haydn, Sonata no. 53 in E minor Hob. XVI/34, first
movement, bars 19–23 83
6.27 Joseph Haydn, Sonata in D major no. 39 Hob. XVI/24, first
movement, bars 40–45 83
6.28 Johann Sebastian Bach, Sinfonia from Partita no. 2 in C minor,
BWV 826 84
6.29 Johann Sebastian Bach, Corrente from Partita in B-flat major
BWV 825 84
6.30 Frédéric Chopin, Polonaise-Fantaisie op. 61, from bar 254 85
6.31 Franz Schubert, first of Drei Klavierstücke D 946 85
6.32 Franz Schubert, Sonata in E-flat major op. posth. 122 D 568,
second movement, bars 43–46 85
6.33 Wolfgang Amadeus Mozart, Piano Concerto in B-flat major
KV 450, first movement, bars 76–77 86
6.34 Franz Schubert, Fantasie in F minor zu vier Händen
op. 103 D 940 86
6.35 Franz Schubert, Fantasie in F minor zu vier Händen,
bars 48–57 86
6.36 Frédéric Chopin, Nocturne op. 32 no. 2 87
6.37 Frédéric Chopin, Nocturne op. 32 no. 2, bars 35–36 87

Note
1. Where no bar numbers are cited, examples begin at the beginning of the movement/
work.
PREFACE

Readers may regard some matters discussed in this book simply as musical
common-sense. Such were indeed my own thoughts on first being approached
by colleagues to write the study. However, in the light of many years’ experience
both as a teacher and as a jury member in competitions, the need for such a
handbook has become abundantly clear. Although perhaps self-explanatory to a
musician steeped in the tradition from which this music springs, to an increas-
ingly large number of students—and indeed teachers—from both Europe and
further afield whose musical enculturation no longer reflects this tradition,
knowledge of the most basic musical assumptions behind our notation is often
sketchy. The result can be an obtuse, arbitrary, indeed absurd misinterpretation
of the musical text.
Nikolaus Harnoncourt uses the drastic example of the resolving dissonance
to point out how the loss of musical knowledge can lead to a complete misun-
derstanding of the written music:

The composer takes for granted that the musician will connect the dis-
sonance to its resolution as a matter of course. He could not take this for
granted today because we have become accustomed to play the notes, and
not the music they express. [. . .] In this way music can lose its meaning,
just as in language, if in any sentence we were to move a comma two
words to the left and the full stop two words to the right. If we now read
the text it no longer makes any sense. With music of the late Romantic
period the musician just has to play what is written. If he does this with
a Mozart symphony, where even the most elementary things are not
xiv Preface

written—because the musician of the time knew them anyway—the result

is a garbled stammering that makes no sense at all.
(Musik als Klangrede DTV/Bärenreiter Munich, 1985, p. 168 f )

There exist countless books on piano technique and questions of interpretation.

The subject of notation itself has been dealt with in numerous musicological
studies on eighteenth- and nineteenth-century music. However, this book is
intended to offer something which to my knowledge is currently not available:
namely a clearly readable, comprehensive and practical handbook of the most
important unwritten musical principles of European piano music of the period
as they refer most specifically to the needs of the on-going pianist.
Treating the score as an incomplete musical ‘shorthand’ rather than a complete
picture of the imagined sound world of the composer, it outlines the compositional
and interpretative imperatives that, though not immediately discernible from the
notation, are implicit within it. The study addresses the many unwritten nuances
of dynamics, articulation and agogics that underlie this music and are readily
overlooked in an all too literal or uninformed interpretative approach. This often
concerns details that are insufficiently differentiated or are not self-evident and
principles that are not recognised, and it is simply a matter of pointing out the
discrepancy between living music and its inadequate written record.
Over the centuries, different regions of Europe have emerged for a time as
leading musical centres of long-standing tradition—whether the England of
composers such as John Dowland, William Byrd, John Bull, Orlando Gibbons,
Peter Philips and Thomas Tomkins or the epoch of the Notre Dame School
with composers such as Léonin and Pérotin. Vienna was another such centre. A
large corpus of the Classical and Romantic piano repertoire was written in and
around this city. For most of the period under discussion, Vienna was regarded
as the major European musical centre. Composers such as Mozart, Beethoven,
Schubert, Brahms, Hugo Wolf and others passed on many of their assumptions and
attitudes to listeners, pupils and teachers, these in turn being passed on through
the generations even into the twentieth century. In this sense, knowledge of such
attitudes and assumptions could be taken for granted and was indeed ‘common-
sense’—a situation that unfortunately no longer applies today. It is within this
context that this book hopes to be of assistance, offering a systematic summary
of the basics of these unwritten principles—principles that are relevant not only
to Viennese music itself but that also inform the underlying assumptions of a
common European musical language.
The book is specifically intended for both students of piano and their teach-
ers. Though aimed at the more advanced pupil, its readership is by no means
restricted to those studying to become professional pianists. It is conceived not
as a work of musicological research, but as a practical guide addressed to all
serious-minded musicians, whether professional or amateur, who wish to deepen
 Preface xv

their understanding of the musical principles inherent within the notation of

European piano music.
The study is divided into the areas of dynamics, articulation and agogics, with
additional chapters on pedalling and aspects of notation such as hidden part-
writing, dotting and ‘technical notation’—notation indicating not the intended
sounding result but rather the action of the player. (A treatment of specialised
fields such as ornamentation in the works of Bach and Mozart would not only
be beyond the scope of this book but would also contradict the intentions of this
study, namely to illustrate problems of notation and to describe the framework
determined by fundamental musical principles for a naturally musical manner
of playing.) There is constant cross-referencing between the individual sections
to allow the reader to consider the various parameters as a whole. Within the
chapters, examples have been grouped together, sometimes according to the
technical point described, sometimes chronologically, as is most appropriate for
clarity of understanding. The book draws extensively on historical records of the
period, while over 100 musical examples—of which the most relevant are avail-
able to the reader as audio files under www.members.aon.at/walter.fleischmann/
book_audios_en.html—elucidate the points made.
It should be stressed that in outlining all these principles and their treatment
through numerous examples from the literature, I only intend to do justice to
the vitality and intelligibility of the music. The false impression might easily
arise that an interpretation can be simply determined by a set of rules and pat-
ent recipes. This was by no means my intention! On the contrary, it is a matter
of freeing the rendering of music both from a mistakenly literal view of the
musical text and from a dry, academic narrowness of approach and so to pave
the way for the performer to a natural manner of interpretation appropriate to
the emotional content of the music. I trust that the prudent reader and musician
will take this in the right way and will well understand that, beyond doctrinaire
coercion on the one hand and boundless arbitrariness on the other, there lies
a creatively inspired approach that does justice to both the content and the
significance of the masterpiece.
ACKNOWLEDGEMENTS

I owe a great debt of gratitude to Prof. em. Dr Paul Badura-Skoda for his valu-
able advice, and to the conductor and composer em. o. Univ. Prof. Wolfgang
Gabriel and the pianist Univ. Prof. Markus Prause, who have contributed sig-
nificantly to this book and without whose insistence and encouragement it
would never have materialised. I am also very grateful to Prof. Prause for his
critical review of the text. I would very much like to thank DDr. Gerhard
Klünger, who prepared the manuscript of the original German edition for pub-
lication. I am indebted to Dr Robert Jamieson Crow not only for the English
translation but also for his valuable editorial suggestions in the preparation of
this extended English edition.
1
INTRODUCTION
Musical Principles and the Limits
of Notation

Music is an affective experience. It is an emotional world imagined in all its

nuances by the composer, though he is unable to commit it to paper in its
entirety even with the most sophisticated notation. Precisely the many details
and subtle differences that express the character of the piece can only be inad-
equately hinted at in the notation. What is left is a mere representation of pitches,
an imperfect description of the length of notes and their interconnection (articu-
lation) and an under-differentiated indication of their loudness. Lacking the
breath that brings the whole to life, the musical score can convey no more than
a sterile succession of notes. Here is plainly a need for a clarification of those
unwritten principles of musical language to which humans are subject and by
which the composer is naturally also bound. These principles, lying within the
area of dynamics, articulation and tempo-variation (agogics), must be considered
if music’s shadow-picture in the notation is to be transformed into living music.
In this present book, it is only possible to illustrate principles, possibilities
and limitations and to correct mistakes and misunderstandings. However, such
considerations in no way amount to an interpretation. They simply provide a
general framework within which it is up to the pianist, through his imaginative
musical conception of a many-faceted, sensitive and spirited performance, to
bring the musical work to life.
Though the various aspects of a performance are here treated separately
according to the parameters of dynamics, articulation and agogics, they always
belong together and can only be regarded as a whole. Observing the principles
described here in no way reduces the possibility of different interpretations. On
the contrary, it helps the performer to a subtler depiction of the musical content.
Unfortunately, our notation fails to reflect such subtleties. Here is a summary
of its shortcomings:
2 Introduction

1. Dynamics are indicated by a series of stepwise gradations (ppp, pp, p, mp, mf, f,
etc.) which fail to represent subtler intermediary values. For the most part, only
a single dynamic range such as p(iano) or f(orte) is specified for one or more
phrases, without touching upon more subtle nuances within the phrase. A
crescendo or diminuendo may be indicated from time to time—but not always—
although the music itself is in constant need of dynamic differentiation.
Vertical dynamics within chords and in different voices are not notated at
all in piano writing up to the end of the nineteenth century (though such
indications are indeed present in orchestral and chamber-music scores), nor
is there a way of notating the difference between stronger and weaker beats.
Through the frequent presence of an incomplete bar at the beginning of a
piece—the upbeat or anacrusis—we only know that an emphasis immediately
after the bar line is implied.
2. Notation recognises only fixed durations that amount to half of their nearest
respective value—half note, quarter note, eighth note and so on, and their
respective dotted values which extend these durations by half. Apart from
this, the only possible way to represent more finely differentiated durations
is by connecting the various values with ligatures (ties). In addition, there are
triplets—the division of a duration into three equal, shorter values—as well as
quintuplets, sextuplets or any number of notes indicated in small print, which
are to be played during the duration of the prescribed value. Music itself,
however, continually displays a significantly subtler range of note-values,
demanding a shortening or lengthening of notes that would be too time-
consuming and complicated for a composer to notate and would also make
reading more difficult. It is left up to the interpreter of the music to arrive
intuitively at an appropriate grasp of such aspects of the work.
Progressive gradation from the shortest of touch to the sustaining of a note
for its full notated value (for example a quarter note) is represented in our
notation in only three steps (legato, portato, staccato)—staccatissimo is seldom
indicated.
3. Music is notated metrically—with mathematical precision—whereas a musi-
cal performance demands a constant, subtly differentiated deviation from this
exactness, even if no accelerando, ritardando or rubato is indicated. However, such
specifications, even with additional instructions such as poco, or molto, tell us
nothing of how, or to what extent, this is to be done. Neither is a lengthening
or shortening of dotted note-values, or an irregular distribution of quarter
notes within the bar—as for example in waltzes—recorded in our notation.
One only need listen to a (metrically exact) recording of a composition made
with a computer to realise how sterile and unmusical such metric and rhyth-
mic precision sounds.

The character of a piece of music, its musical expression, in fact those very
qualities that constitute the essence of the music, namely its emotional content,
 Introduction 3

can only be expressed through subtleties and nuances in its rendition, and
these can only be hinted at through notation in an inadequate way. It is precisely
at the point where expressive playing begins to take shape that our notation
fails. If we were to play a piece as we see it written, it would make no musical
sense at all. Such music would not move us. A potentially well-meaning attempt
to avoid all mistakes through a strictly literal adherence to the demands of the
notation would amount to a grave misunderstanding of what is meant by
authenticity, or faithfulness to the musical text. This does not imply a faithfulness
to the letter, but rather the endeavour to understand and accurately to re-create
the content and expression of a work.
Manners of delivery in music and speech are closely related. Their origins
lie equally within man’s spiritual and emotional being. They are based upon
man’s need for a degree of differentiation that shapes the progression of sounds
in music as in speech in accordance with the requirements of their meaning
and emotional content. This gives rise to certain principles that both musician
and speaker follow. We need offer only one example of the close relationship
between both these areas of expression:
Depending on its emotional content, speech may be whispered, delivered
more quietly or loudly, even to the extent of being shouted. It can sound
friendly, vehement, angry and so on. In music, the situation is no different. It is
an emotional experience with the same variety of expressive possibilities. It is
the art of the interpreter to reproduce this experience.
In addition, in music as in speech, we are familiar with differences of emphasis
and sound duration. Stresses within words and within a sentence correspond in
music to accentuation within a motive and within the phrase as a whole. Speech
offers another analogy to music in its short and long syllables, which may be
further elongated in moments of heightened emotion, for example in the expres-
sion of astonishment. This vital need for differentiation in speech is evidently
inherent in humans, and without it, language would be incomprehensible. We
recognise similar principles in music, such as the differentiation of strong and
weak beats, that clearly owe their origin to aspects of human physiology, such
as breathing in and out, systole and diastole phases of the heartbeat, right and
left steps in walking. Such principles are simply part of human nature, and it
is imperative that we take them into consideration, both in playing music and
in speaking. To disregard them in the interpretation of a piece of music has
nothing to do with the freedom of an individual ‘creative’ performance: it is
pure haphazardness.
Music consists of melody, harmony and rhythm. Depending on the demands of
the music, a pianist can at times bring out the rhythm, while at other times he
may give greater emphasis to the harmony or the melody. The three parameters
form an indivisible unity in the musical composition—one aspect or another is
simply accorded with greater weight. One very often hears a kind of playing
that is too much reduced to its rhythmic aspect: the performer pays too little
 4 Introduction

attention to what is going on melodically and harmonically and the music

becomes overly simplistic. Here is an example from a work of Schumann that
illustrates this point.1 In the first movement of his Sonata in F-sharp minor op.
11, bars 74–84, is this moving passage:

Ex. 1.1

One often hears the passage reduced to nothing but its rhythm, as in AUDIO
2
FILE 2.
Apart from tempo, the interpreter has at his disposal dynamics, articulation
and agogics to convey the emotional content of a composition. As in speech,
an essential natural vitality in the representation of musical content constantly
demands all three modes of expression.

Note
1. The musical examples are representative of countless passages that contain similar prob-
lems. For a better understanding of the explanations offered and to prevent possible
misunderstandings, a number of the musical examples have been recorded as accom-
panying audio files. They in no way represent the only possible interpretation.
2
DYNAMICS

In the same way that in speech each syllable is not equally weighted, music is
also characterised by moments of varied emphasis, namely the strong and weak
beats of the bar. The main emphasis occurs directly after the bar line. This is
most evident in the many pieces beginning with an incomplete bar (upbeat or
anacrusis) in which the first note is unaccented, and the intended emphasis is
placed after the bar line. In a 4/4 bar, the first quarter note is therefore generally
given the strongest emphasis and the third a little less, while the second and
fourth quarters receive the least accentuation (deviations from this rule can occur
as the result of shifts in harmonic tension). In his violin school,1 Leopold Mozart
describes a pattern of emphases within the 4/4 bar that corresponds to the fol-
lowing schema:

Ex. 2.1

In 3/4 and 2/4 time, only the first quarter note is emphasised, in 6/8 time the
first and fourth eighth notes. (Deviations from this pattern of emphases occur
with the hemiola, indicated by colouration in Renaissance music. In 3/4 time,
for example, two bars are combined and the emphasis falls not on the first of
the three quarter notes but on every second quarter. The same principle is true
for the hemiola in 3/8 and compound duple time.)
However, a still finer differentiation should be observed in a slow tempo by
playing the intermediary eighths or sixteenths even more quietly. In performance,
these differences appear as a fluid ebb and flow of dynamics and not as accents
that stand out from the rest of the notes.
 6 Dynamics

A further aspect of simple duple and quadruple time should be observed.

In so-called alla breve time (2/2), the underlying beat is no longer the quarter
note but rather the half. In a piece in alla breve time, the musical ductus has
a broader arch, owing to the reduction in points of emphasis. For example,
Schubert’s Impromptu in B-flat major (op. posth. 142 D 935 no. 3) loses its
graceful character if one accentuates both the first and third quarter note instead
of omitting the latter point of emphasis as the indication alle breve demands.2

Ex. 2.2

Such patterns of emphasis that, as in the recital of poetry, prevent a monotonous

‘rattling off ’ of the rhythm are also prevalent within the context of groups of
bars. Since musical units are made up of at least two bars, the first beat of the
second bar is less accentuated than that of the first bar. The pattern is repeated
in the next two-bar group, albeit somewhat quieter, unless for harmonic or
melodic reasons the phrase needs a crescendo. This results in the same pattern of
emphases over the four bars as is felt within a single bar. As we have mentioned,
these emphases presumably originate from the physiology of the human body.
In turn, these four-bar units make up the antecedent and subsequent sections
of a larger phrase. However, bars within a phrase may be grouped in other ways.
If the pattern of four-bar units changes over the course of a piece to groups of
three bars, Beethoven in his string quartets and his 9th symphony writes ‘Ritmo
di tre battute.’ As soon as the movement returns to the four-bar unit, he writes
‘Ritmo di quattro battute,’ that is to say, a rhythm of four bars, whereby the
points of emphasis described above are implied.
The beginning of the first movement of Mozart’s Piano Sonata in B-flat major
KV 570 is a good example of varied emphases within a phrase.

Ex. 2.3

3
 Dynamics 7

Here, again, the first note of the first bar is emphasised more strongly, the first
note of the third bar less so and that of the second and fourth bars even less.
If, on the other hand, the player wants a build-up towards the fifth bar, he will
begin the third bar louder than the first. The first and third bars should certainly
not begin at the same volume, as a return to the same dynamic gives the impres-
sion that the music is not moving forward and results in a mechanical ‘reeling
off ’ of the phrase. The two subsequent four-bar groups are also to be differenti-
ated dynamically. The differentiation within the bars should be made in the
same way as described above.
The opening theme of the Scherzo from Schubert’s Sonata in B-flat
major D 960 (Example 2.4) can be played with the same dynamic shape,
as far as this is possible in pianissimo . However, depending on one’s per-
sonal gusto—to use Mozart’s words—one might place the cantabile climax
on the third bar.

Ex. 2.4

A striking instance of this principle of dynamic differentiation occurs in the

Trio from the Scherzo of the same sonata: an eight-bar period made up of
antecedent and consequent phrases with an additional two-bar cadential exten-
sion. While the right hand employs a dynamic pattern similar to the above, the
left hand accentuates the weaker second, fourth and sixth bars with a fzp in
the manner of a syncopation played out against the dynamic emphases of the
phrase in the right hand (Example 2.5).

Ex. 2.5

5
 8 Dynamics

Beethoven’s Sonata op. 57 offers another illustration of the importance of observ-

ing stress patterns within the phrase as a whole. Here are bars 35–39 from the
first movement:

Ex. 2.6

This second section of the exposition, from the indication dolce in bar 35, consists
of an antecedent and consequent phrase, each two bars in length, beginning with
an upbeat. To avoid the monotonous scansion one so often hears, the player
should not only observe a difference between strong and weak beats; it is also
necessary to dynamically differentiate the first beats of adjacent bars that fall on
the same note, namely A-flat in the antecedent and C in the consequent phrase,
thereby giving a new turn to the sequence of events, as it were. The climax is
on the A-flat in the second bar of the antecedent phrase. In his sonatas, it is
striking how Beethoven frequently reduces the dynamic level suddenly at the
climax of a crescendo and indicates piano, as if he wants this climax to be reached
with not the loudest but rather the most beautiful sound.3 This characteristic
feature of Beethoven’s writing might be similarly achieved in the above passage
by playing the A-flat in the second bar of the phrase (bar 3 of the example)
softly, in accordance with this way of thinking.
Sometimes the course of a phrase after the first two-bar unit requires a cre-
scendo in the second two-bar unit that then leads from the fifth bar onwards to
the climax of the phrase, and then subsides calmly (see Example 2.14 by Haydn
and Example 2.15 by Beethoven on page 12).
The stricter phrase structure of the Classical period gives way in the music of
the Romantic, Impressionist and later periods to a freer, multiple-barred phrasing
that is often interposed with recitative or fantasy-like passages. This changes nothing
as far as the presence of strong and weak beats is concerned. In such music, too,
there is of course the need for correspondingly adjusted, differentiated dynamics.
If a crescendo is indicated, this cannot be achieved through a linear increase in
loudness from note to note: the strong and weak beats within the bar must still
be observed. In spite of the indicated crescendo, the note that immediately follows
 Dynamics 9

the stressed beat must first be played more softly; only then is the crescendo made
by increasing the loudness of each stressed beat. The intermediary weaker parts of
the bar must be played more quietly, unless one wants to give the impression of an
unmusical—because undifferentiated—chain of notes that makes no musical sense.
Equally, the diminuendo undergoes the same differentiation between strong
and weak beats.
A continuous increase in dynamic level from note to note on the piano
sounds like a knocking or hammering. In its uniformity, it distracts from what
is happening melodically and harmonically, reducing the music to its rhythm
alone. A fast run with a crescendo may serve as a case in point:

Ex. 2.7

Although a crescendo is indicated from the very beginning, one must still play a
diminuendo after the first sixteenth note and should only begin the crescendo from
about the fifth of the sixteenth notes. The graphic line above the passage illustrates
this approximately. This schematic illustration applies to every crescendo run, even if
no crescendo is indicated. In Examples 2.9–2.11 (page 10) this will be discussed at
greater length with special reference to slower passages of expressive significance.
Longer runs, figurations and groups of sixteenth or thirty-second notes joined
into groups of four, six or eight notes respectively are constantly occurring in
music in a large variety of metres. In such passages, we can often recognise figures
repeated on different degrees of the scale. It is not only essential to differentiate
these groupings of notes with respect to the strong and weak beats and their
subdivisions—that is to say, immediately playing softer after the first note—but
also to express the rise and fall of the motives dynamically. Here is a passage from
the Prelude in F minor from J. S. Bach’s Well-Tempered Clavier II, bars 20–24:

Ex. 2.8

Under no circumstances should one play this passage as on AUDIO FILE 8,

since repeating the motive at the same volume only directs attention to the
10 Dynamics

rhythmic repetition instead of bringing out the melodic and harmonic progres-
sion through dynamics (see also the explanations on page 54, Example 4.5
concerning the same prelude in the chapter on agogics).
A graphic representation of the dynamics in a sequence of four bars in 4/4
time would look something like this:

Ex. 2.9

If a bar consists not of four quarter notes but exclusively of sixteenths, the first
note of each four-note (or six-note) group creates the same dynamic contour
as illustrated above. The other sixteenth notes are correspondingly quieter, to
avoid a uniform hammering effect.
The musician may, however, be loath to see this rule forced into such a sche-
matic corset, even if it is valid in principle. We should again remind ourselves
of Leopold Mozart’s description of varied emphasis mentioned at the begin-
ning of the chapter. The character of a phrase frequently demands a deviation
from this general pattern anyway, for instance through sforzati, syncopations or
essential crescendi and diminuendi that are in such instances often, but not always,
indicated by the composer.
The crescendo, then, would look like this:

Ex. 2.10

and the diminuendo as follows:

Ex. 2.11

Again, this all sounds very schematic and not at all ‘artistically inspired,’ but it
is simply a law of vitality, necessitated by these differences within a bar and
within a succession of bars. The real musical artistry lies in the dynamically and
agogically finely nuanced connection of these main and subsidiary notes that
only then allows the music to breathe naturally. In its dynamic agility, even a
chain of sixteenth note runs can allow our feelings to resonate and can be
moving.
 Dynamics 11

Why is it so particularly important in piano playing to observe these dif-

ferences between strong and weak beats? The tone of the piano has the char-
acteristic quality of subsiding in volume immediately after the key has been
struck. If, at a slow tempo, we choose a flat, undifferentiated dynamic and play
all notes equally loud, each successive note will de facto sound like an accent,
due to the decrease in volume of the piano tone after each attack. A graphic
representation illustrates this:

Ex. 2.12

Now instead of shaping a motive or a phrase in a meaningfully vivid way, this

kind of playing sounds like an incoherent stringing together of hammered out
notes. If, however, one creates a dynamic arch by observing the stronger and
weaker beats, with an increase and decrease in loudness of these, as it were,
‘accented’ notes, the listener will be unaware of this deficit in piano tone. He
will experience the dynamic rise and fall of the accents in his mind as an increase
and decrease in musical tension.
Here, too, this tension is additionally supported through agogics by employing
a scarcely noticeable accelerando towards the climax of the phrase, followed by
a hardly perceptible ritardando. If this accelerando were to be recognised as such,
this would mean it was being applied too strongly. To put it another way: only
a sense of aiming towards the climax with a slight quickening of pace can do
justice to the musical content.
Strictly speaking, at every note the player is faced with the decision whether
to play the next note quieter or louder and by how much this should be done.
That all sounds very complicated, but it is only complicated to describe, not
to execute. Nonetheless, such playing demands great aural sensitivity from
the pianist.
As described above, even in fast runs a monotonous way of playing is
avoided through a rise and fall in the loudness of notes following the pattern
of strong and weak beats of the bar. Graphic representations of the dynamics
of musical themes are to be found in the book on Mozart by Paul and Eva
Badura-Skoda.4
The dynamic contour of a musically shaped phrase with a climax in the
middle might look something like this:

Ex. 2.13
 12 Dynamics

Here is an example of such dynamic shaping in a work by Haydn, in a phrase

with a different pattern of harmonic tension—his Sonata in E minor Hob.
XVI/34, second movement, Adagio:

Ex. 2.14

A two-bar motive begins in the first bar. It grows quieter towards the second
bar and ends on the dominant. In the third bar, the same motive appears one
degree higher with a corresponding increase in dynamic level. In the same way
that speech follows emotional content by giving more marked expression to the
repetition of a sentence or part of a sentence, in music such a repetition also
results in an intensification in the manner of playing. The climax of the phrase
is reached in bar 5. The sixth bar grows calmer once more, becoming somewhat
louder again in the seventh bar on the cadential six-four chord resolving on the
dominant. The phrase then ends in the eighth bar on the tonic.
The Adagio molto from Beethoven’s Sonata op. 10 no. 1 has a similar shape:

Ex. 2.15
10

Here, too, the second two-bar unit is louder. However, in the fifth bar Beethoven
starts a crescendo that leads to the climax on the subdominant in bar 6, closing
on the tonic in bar 9. With the fp on the subdominant chord, he wishes to
ensure that one starts the decrescendo much quieter in bar 7.
In the fifties of the last century, Paul Badura-Skoda and his wife Eva wrote
a book on Mozart (Mozart-Interpretation, Vienna 1957), which more recently
appeared in a revised English edition.5 Alongside many other examples, Paul
Badura-Skoda illustrates with an undulating line—indicating dynamic contour
of the first solo phrase of the first movement of the C minor Piano Concerto
 Dynamics 13

KV 491—how musical content gains in expression through a differentiation of

dynamics and an observation of the weak beats of the bar.
As the harpsichord was gradually superseded by the fortepiano from the
eighteenth century onwards, composers began, at first only very sporadically, to
notate dynamics more exactly. Earlier Baroque instrumental music was governed
by the rule that one should play forte so long as no other dynamic indication
was given. At first, one limited oneself to the indication p for quiet and f for
loud playing, and only gradually did a somewhat more nuanced notation develop.
However, not even these markings should be taken literally. Such indications allow
only for the tonal possibilities of the instruments of the time, which could not
produce the louder, more brilliant, even sharp-edged tone that is usual today.
Mozart was particularly fond of contrasting p with f as opposing dynamics. At
that time, dynamic transitions were still more seldom used. This practice origi-
nates in the harpsichord, which did not offer the possibility of a crescendo or
diminuendo, though the conception of the music, derived as it was from singing,
still contained inherently musical, natural dynamics.
In Haydn’s early works, a f can, under certain circumstances, indicate ff or
alternatively mf, i.e. mezzoforte, or even a crescendo. Only at a later period dur-
ing the gradual transition from the harpsichord to the modern pianoforte are
dynamics, unless otherwise indicated, to be regarded as fluid rather than abrupt.
The sforzati in Haydn can also mean various things. Depending on the sense
of the motive, the sforzato can be an accent within the range of the current
dynamic or it can emphasise the structure. It can, however, simply emphasise
the importance of a particular note. In other words, the sforzato can affect the
soft, delayed entry of a note and may not imply an accent at all. This is, of
course, above all true of slow, cantabile lyrical movements. In the Adagio from
Haydn’s Sonata in E-flat major Hob. XVI/49, we find two sforzati in the third
and seventh bars that require a soft expressive playing, even a slightly delayed
execution of the second sforzato towards the six-four chord.

Ex. 2.16
11

Why did Haydn find it necessary to write these sforzati at all? Without the
sforzati, which are in fact not loud accents, the pianist would tend only to
14 Dynamics

bring out the notes on the stronger beats expressively, that is on the first,
second and third quarter notes and to play the two intermediary notes, E-flat
and G, as weak beats, almost imperceptibly quieter. If Haydn intended all four
notes to be played in an expressive cantabile with a slight crescendo and the
equally agogically necessary delay towards the subsequent six-four chord, then
he could only indicate what he wanted in this way. Four bars later, the same
thing is repeated.
Only over the course of time did a more exact dynamic notation develop.
In works of Debussy one already finds the indication pppp for extremely quiet
playing. This tendency continues to develop up to the present day (for instance
in works of Ligeti).
In the works of Haydn, but also of Mozart, especially in his early works, it
is clearly not enough simply to observe the few dynamics indicated—markings
originally limited to p and f—where for whole pages of music no dynamic
markings of any kind occur at all. One is obliged to consider for oneself
what might be a suitably differentiated dynamic for the music’s character or
emotional content.
Incidentally, the same is true of articulation. In the Baroque era, this was left
mainly at the discretion of the performer, although indications as to the man-
ner of its application are to be found in the writings of Carl Philipp Emanuel
Bach and others.
A p not only means that we should play quietly, but at the same time that
we should consider the mental expression to which it refers. According to its
inherent quality, the mood of this piano can be cheerful, graceful, melancholy or
sad, depending on the musical and rhetorical intentions of the composer. Only
in the context of agogics and articulation will this p achieve the desired manner
of expression within the phrase.
Vertical dynamic differentiation is musically essential but, as we have already
mentioned, is not represented in our notation for the piano up to the end
of the nineteenth century, although in an orchestral or chamber-music score
such dynamics are indeed indicated with dynamic markings. Vertical dynamic
differentiation may imply playing accompanying figures more quietly, varying
dynamics within a chord for optimal tone quality—usually soprano and bass
louder, unless a middle voice plays a leading role in a musical progression—and
observing the dynamic differentiation of polyphonic voices. The subject will be
discussed further throughout the chapters that follow.
In this context, it is worth mentioning the terraced dynamics in orchestral
works of the Baroque period. Such dynamic differentiation by no means implies
that everything within each dynamic ‘terrace’ was played equally loudly or qui-
etly. Bach gives no indication of dynamics in his works for keyboard—dynamic
differentiation only became possible at the keyboard with the advent of the
fortepiano. It is therefore left to the pianist to decide what dynamic differen-
tiation to employ. Although alternating indications f and p do occur in Bach’s
 Dynamics 15

Italian Concerto, such indications merely point to the difference between tutti
and solo passages.
Not even the subtlest differentiation of dynamics alone can assure a perfor-
mance that does real justice to the musical expression implied by the notation.
Dynamics, agogics and articulation must first act together to facilitate a natural
manner of expression.

Notes
1. “The particularly dominating notes are as follows: in every bar the note which is struck
on the first quarter, [and] in 4/4 time the first note of the second half of the bar or
third quarter.” (Die sonderbar herrschenden Noten sind folgende: in jedem Tact die das erste
Vierttheil anschlagende Note [und] die erste Note des halben Tactes oder dritten Viertheils im
Vierviertheiltacte), Leopold Mozart Versuch einer gründlichen Violinschule, Johann Jacob
Lotter, Augsburg, 1756, 12/§9.
2. The indication alla breve originated in mensural notation and indicates double speed:
not the semi brevis but the brevis was to be followed as unit of measure or beat. Later
it changed to become also an indication of character.
3. With the typically Beethovenian characteristic of a piano after a crescendo, sometimes
the mistake is made of executing a decrescendo before the piano. In such cases the decre-
scendo is uncalled for: the composer wishes the climax of the phrase to be suddenly
and surprisingly different.
4. Eva and Paul Badura-Skoda, Interpreting Mozart, New York, London, Taylor & Francis
Ltd. 2nd edition, 2007.
5. Ibid.
3
ARTICULATION

Beyond the possibilities of dynamics, articulation—the way in which individual

notes are connected or separated and how long or short they are to be held in
relation to their notated value—offers further ways of shaping a piece of music.
Our notation knows only legato, staccato and portato, represented by a slur for
legato, dots for staccato or, if both are combined, with a slur drawn over the dots
for portato. Sometimes we find horizontal strokes above the notes indicating a
tenuto, or elongated portato. In addition, a staccatissimo is occasionally asked for.
(Vertical strokes in composers’ autographs up to and including Beethoven’s works
signify not the modern marcato, but a normal staccato. See below, page 45.)
Apart from the directions legato and staccatissimo, notation only indicates two
lengths for the execution of individual notes: staccato and portato. These indica-
tions do not do justice to the diversity of music and are nothing but a rough
notational shorthand. However, such directions are unfortunately often taken
at face value. In fact, music demands that a note be sustained for a far greater
variety of durations within its notated value before the next note enters at that
moment allotted to it by tempo and rhythm. These lengths may vary from very
short to very long, depending on the emotional content, the expression of the
motive or the phrase. In this sense, by playing notes longer or shorter, a staccato
or portato should be shaped with a great variability of expression.
Let us imagine the progression in time of a quarter note in a slow tempo: one
can play this note in about six different lengths, from very short to very long,
before the next note enters. The three shorter values would be described as stac-
cato, the three longer values as portato. It is up to the pianist to choose the right
one, since each of these differences in duration has a different expressive value
and thus a different emotional content. A constantly uniform staccato is to be
avoided if the playing is not to sound unintentionally comic, absurd or fatuous.
In practice, this means for example that an eighth note with a staccato dot is to
 Articulation 17

TABLE 3.1 Staccato, portato: comparative durations

A quarter note The following quarter note

|________________________ |________________________|
__ | extremely short touch
______ | short touch
___________ | held slightly longer
_______________ | held a little longer still
____________________ | long execution
_________________________ | a very long execution but still without a legato to the
next quarter note

be distinguished in its length from a quarter note with a staccato dot. To illustrate
this more clearly, let us imagine a series of quarter notes that are either given a
staccato dot or are required by the composer to be played portato. A single quarter
from this series may be played with different lengths. Represented graphically, a
quarter note appears within its notated duration as shown in Table 3.1.
The upper three differing lengths of attack correspond to various lengths of
staccato, while the lower three correspond to various lengths of portato. The varying
character of individual motives demands the employment of all these gradations.
Unfortunately, even some well-known pianists seem only aware of the short staccato.

Differences in Articulation Between Stressed and

Unstressed Beats
Inherent to music’s movement through time is the continuous repetition of a
specific cycle of pulses. It is, as it were, the heartbeat of music, of which the reflec-
tion is seen in the structure of the bar with its strong and weak beats. Dynamics
and articulation are interrelated. Notes on a weaker beat in a staccato or portato
should not only be played quieter but also somewhat shorter to express their
lightness. Otherwise they sound too cumbersome. It is like speech, where inter-
mediary syllables are dropped. Conversely, in a portato or staccato phrase a strong
beat requires a correspondingly longer execution of the note. The expression of
a graceful lightness, for example, cannot be expressed through dynamics alone. It
also requires a corresponding articulation. These varying emphases and lengths
illustrate once more the affinity between music and language. Such principles
cannot be ignored in a musical manner of playing, as here in Liszt’s Tarantella,
bars 73–75:

Ex. 3.1
12
 18 Articulation

One can hear how much like stuttering or barking it sounds if one fails to play
13 the stronger beats longer (AUDIO FILE 13).
The shortening of a weak beat, as in an upbeat for example, is often indicated
by a staccato dot, but not always. How short or long a staccato is to be played
depends on the character of the piece. This is the Trio of the Menuett from
Beethoven’s Sonata op. 31 no. 3:

Ex. 3.2
14

To do justice to the genial character of the phrase, the upbeat must be played
lighter and should be allowed to die away. The note should not be too short
or be abruptly cut off. This can be achieved through a careful lifting of the
fingers or through a slightly slower release of the pedal. In such situations I
prefer using the pedal, as this leaves more time to jump to the next chord and
helps to achieve a more relaxed and flexible technique. If the player follows
Beethoven’s suggestion of imagining a suitable text to such a passage,1 and plays
as if expressing something that is being sung rather than simply following the
instructions of the notation, the composer’s intention can hardly be missed. It
is principally the difference between simply playing music or experiencing it.
In a minuet, for example, the graceful character can only be achieved through
a shortening of the weaker beats. The principle of musical vitality forbids any
uniformity or mechanical repetition. Let us take the case of this following
common ending:

Ex. 3.3

In accordance with the explanations above, one would play the first note louder, the
second and third quieter and the first note of the next bar again somewhat louder,
if not as loud as in the first bar. At the same time, one plays the strong beats a bit
longer, the weak beats shorter. This is how it would look notated in 3/8 time:

Ex. 3.4
 Articulation 19

However, in 3/4 time, here in the first movement of his B major Sonata KV
570, bars 11–12, Mozart notates it in this way:

Ex. 3.5
17

In such a non legato phrase, the weaker beats are shortened: the strong beat is played
with a slight emphasis and, as mentioned above, is held slightly longer, while the first
rest is marginally abbreviated. This shortening of notes on weaker beats naturally
does not apply to legato passages or phrases with a sustained, solemn character such
as Example 3.13, page 24. Mozart wrote with a goose quill—an implement later
replaced by a metal dip-pen. With both tools, the beginning of each stroke or dot
was a tricky business. One often managed by writing a stroke instead of a dot, in
order not to interrupt the writing process. This is also the reason why sixteenth
notes were notated as eighths with a diagonal stroke, or thirty-seconds as sixteenths,
similarly crossed through. This could be done without lifting the pen.
In the 3/4 bar in the above example, the required shortening of the notes is
notated. In a 3/8 bar, it was obviously too complicated to write sixteenth notes
and sixteenth note rests. There are numerous similar four-note motives in 3/8
or 3/4 time of this kind, with slightly varied sequences of notes that must be
differentiated in the same way.
It is virtually impossible to notate the essential tiny differences in note-lengths
between strong and weak beats, and an attempt at exact notation would make
reading considerably more difficult. In the above example, in addition to dynam-
ics and articulation, agogics now also come into play, and the final note is slightly
delayed. One takes a fresh pedal on each note—longer or shorter depending on the
duration of the note—in order to avoid abruptly cutting off the tone (see Chapter
5, page 57). This simple passage illustrates the intimate correlation between dynam-
ics, articulation and agogics, akin to the interplay of these parameters in speech.
An example by Schubert: the second movement of the A minor Sonata op. 42
D 845:

Ex. 3.6
18

This example again shows how carefully and finely nuanced one must play the
staccati and the accents to do justice to this subtle Viennese style. It is a graceful,
 20 Articulation

genial theme with a dance-like character, similar to that of a minuet. Typically

for Schubert, the main theme E-F-D-E begins in the lower voice, as it were in
the ‘second violin,’ while the note G in the upper voice—the ‘first violin’—only
supplements the harmony and should be played pp. Only in bar 7 does the ‘first
violin’ take the lead. The upbeat is to be played lightly, that is shorter and more
quietly, but using the pedal to allow the sound to die away and to avoid the
ugly cut-off at the end of the note. Too short a staccato would be unfitting to
the amiable character of the piece. Schubert gives an accent to the first quarter
note F of the following bar to indicate even more clearly the dance-like swing
from the upbeat to the already emphasised first quarter note. Such accents by
Schubert are always to be played with care and a necessary sensitivity.
The second movement of Haydn’s Sonata in E minor Hob. XVI/34 offers
a further illustration of how staccato dots are to be thought of in relative terms:

Ex. 3.7

21

There are staccato eighth notes in the very first bar of the Adagio. Playing these too
short would amount to an absurd disruption of the calm, even-tempered character
of this peaceful movement. As we have said, music demands a great variety of
durations that our notation is incapable of depicting. As suits the piece, a player
must constantly differentiate between a short, medium or longer staccato touch.
One must bear in mind the instrument that Haydn had at his disposal when he
composed this piece. That was in 1781/82. The first dynamic markings that Haydn
used that point to the fortepiano, an instrument with a hammer mechanism, appear
in the Sonata No. 33 Hob. XVI/20, which he wrote in 1771. At that time, the
fortepianos still had a weak tone, and harpsichords remained in common use until
the end of the century. The hammerflügel, particularly the French, resonated a little
even through the light dampers. The sound of such instruments or the plucked
string of a harpsichord has quite a different tone than that of modern-day instru-
ments. Due to their light jacks and dampers, a staccato never sounded as short
and abruptly cut off as it does on the grand pianos of today, with their much
heavier dampers. One mustn’t forget that at this time a staccato dot often simply
meant ‘these notes are not to be joined.’ As we have said, music demands a most
subtle differentiation in note length that our notation is incapable of indicating
in sufficient variety. The length of the staccato stroke—whether short, medium or
long—must be adjusted in accordance with the character of the piece.
Haydn’s idiosyncrasy of almost always writing a staccato stroke above notes
before a rest is difficult to explain. Perhaps it has something to do with how
 Articulation 21

Haydn imagined the sound: in Haydn’s mind, it may have evoked the bell-like
sound produced by a springing movement of the arm on the harpsichord, dying
away meaningfully, as Leopold Mozart describes.2 Before 1800, the articulation
of the single note was ideally imagined to be analogous to that of a bell—with
the sound dying away. When this was not desired, and the composer wished
the note to sound in full loudness throughout its duration, one wrote tenuto or
sostenuto. Perhaps the words of Carl Philipp Emanuel Bach apply here:

Many things happen in music that you have to imagine, without actually
hearing them.3

Here is an example similar to the previous passage by Haydn, from Beethoven’s

Sonata in C major op. 2 no. 3, Adagio.

Ex. 3.8
31

Here, too, the staccati are intended as somewhat longer notes with a natural decay.
Similarly, the eighth notes should not be cut off but allowed to resonate a little
into the rests. To play the sixteenths short would in no way fit the character of
the piece. What is required here is a long staccato, which the notation cannot
represent, comparable to the resonating pizzicato of the cello. This is best achieved
by raising the pedal less quickly, whereby the release not only determines the
necessary duration of the note but also helps to avoid an ugly cut-off.
These examples show how varied staccato dots are to be played. At times it
is necessary not to lengthen but rather to shorten a note very slightly, as here in
J. S. Bach’s Well-Tempered Clavier II, Prelude in E major. Here we need to play
the light upbeat B to the second quarter slightly shorter, while allowing it to
resonate briefly with the pedal. If cut off abruptly by the dampers the motive
sounds sober, dry and mechanical.

Ex. 3.9
26

One might equally set a staccato dot on the B, producing the same effect, which,
as in previous examples, would be played somewhat longer. Either way we see
 22 Articulation

the correlation between dynamics and articulation: the difference between strong
and weak beats is expressed not just through loudness but also in the length or
shortness of a note. Some further examples serve to illustrate this point.
The execution of the following passage in the Andante of Mozart’s Sonata in
A minor KV 310 (bars 15–19) requires particular sensitivity.

Ex. 3.10
29

From bar 15, the note G is repeated eight times in staccato, the F nine times. Bar 68
presents the phrase again on a different degree. An identical repetition of the same
note would make little musical sense: it is difficult to imagine that such undifferenti-
ated repetition could be in character with Mozart’s witty, often exuberant liveliness
of spirit. This can be avoided by using varied dynamics for the strong and weak beats,
together with corresponding agogics. In this way, the full meaning of the passage is
revealed in all its beauty. It reminds us a little of Papageno’s singing in the opera.
The antecedent phrase begins with a C in the bass in bar 15. The first three
sixteenth notes are the upbeat to the second quarter. Even here, one should play
the middle of the three sixteenths a little heavier, perhaps with a short touch of
the pedal, in order to make the difference between this note and the other two
lighter sixteenths clear. Continuing to observe the different strong and weak
beats, the tension rises till we reach the last G, that (without staccato) may also be
played with a slight ritardando. The Fs that follow in the consequent phrase are to
be played in the same way, with gradually subsiding tension. The same thing is
then repeated in the left hand. AUDIO FILE 29 offers a possible interpretation.
After the middle section and a restatement of the opening theme, the passage
described above is repeated. However, before it enters, a D, repeated eight times,
appears as an upbeat to the next bar. This time, however, a slur is written above it,
indicating that it is to be played with a long portato, though it is to be played in
just as differentiated a manner as before and not too heavily. Here are bars 64–65:

Ex. 3.11
 Articulation 23

Another example is the Variation IV from Beethoven’s Sonata op. 26:

Ex. 3.12
30

As was customary in the Classical period, the tempo of the variations is the same
as that of the theme, unless the composer indicates otherwise. This is dictated by
the strictness of the form, which demands that the original theme remain recog-
nisable throughout the variations. Unfortunately, this principle is often misunder-
stood and the variation is played much too fast. The bass moves stepwise from
the tonic to the dominant, with an upbeat before each of the steps. The strong
beats, which deliver the harmonic progression, are to be played correspondingly
somewhat longer. In accordance with the graceful character of the piece, each of
the eighth notes on weak beats are to be held back a little—that is to say, played
more quietly—and should neither be cut off with a staccato nor abruptly curtailed
by a sudden release of the pedal but should rather be allowed to die away. If one
were to hold the left hand notes for their full value as if they had no staccato
markings—something that Beethoven consciously seeks to avoid in his notation—
the effect would be one of ‘dragging one’s feet’ instead of the graceful steps of a
dancer. A uniform staccato, on the other hand, would complete obscure the har-
monic progression. With the indication sempre stacc., Beethoven no doubt simply
wishes to point out the difference in articulation between the bass and the cantilena.
However, there can hardly be any doubt that Beethoven was fully aware of dif-
ferent lengths of staccato that unfortunately could not be expressed in the notation.
Once again, attention should be drawn to Example 3.7, which shows what subtlety
Beethoven employed in his notation when the notation allowed it.
If other composers, perhaps in the haste of composition, occasionally notate
with less precision, this does not alter the need for a proper execution by the per-
former. The same degree of differentiation in articulation as in Example 3.7 is to
be applied to a Liszt waltz (see Example 3.14 on page 24), though at the requisite
tempo these differences will be marginal. We will come to this example later.
In the Largo appassionato of the Sonata op. 2 no. 2 (Example 3.13), Beethoven
indicates sixteenth note staccati in the left hand. He evidently imagines an accom-
paniment in the bass like the plucked strings of a cello—a pizzicato. On the piano,
this effect can be achieved with the fingers or with the help of an extremely
careful use of the pedal. The rests between the sixteenths must still be observed,
although these will be shortened slightly by the reverberation of the pizzicato.
 24 Articulation

In this theme the F-sharp, played three times, aims towards the four-two chord
of the second bar (dominant-seventh, third inversion). One can imagine an increase
in dynamic similar to a vibrato on the cello that builds up towards the chord and
subsides after reaching the tonic in the second bar. Another possibility would be
to play the three chords with the same dynamic, closing with a quieter second bar.

Ex. 3.13

27

Liszt’s Valse Oubliée no. 1 offers an instance of highly differentiated articula-

tion, though all notes are notated in exactly the same way:

Ex. 3.14

28

The upbeats are to be played somewhat shorter than the quarter notes on
the strong beats, and the last quarter at the end of the phrase is accordingly to
be played a bit longer. That means that even with short staccati such as in this
example, one still should play in a nuanced fashion, as meaningful music demands.
The execution follows the same principle that Beethoven could notate more
exactly in his Sonata op. 2 no. 2, owing to its slower tempo (see Example 3.32,
page 36). The pedal is applied on each of the quarter notes on strong beats,
allowing it to die away and avoiding a cut-off in the sound, as explained earlier.
Of course, one can achieve this with the fingers alone, but with the pedal it is
not only easier but also improves the sound. Releasing the dampers allows other
strings to vibrate sympathetically, thus vastly enriching the sound spectrum by
allowing the overtones to be heard.
Dynamics go hand in hand with articulation. The first quarter note of the
second bar is necessarily quieter, as explained in the chapter on dynamics; the last
note of the antecedent phrase is, as the final note, again a little more accentuated,
though not so strongly as the very first strong beat. As a consequent phrase, the
following bars are played with the same overall dynamic differentiation, though
somewhat quieter. Because of the tempo, such differentiation will be very slight,
 Articulation 25

but it must be made, if one does not want to offer a performance such as that
produced by a computer—as is done with many a famous melody on the
telephone for the listener to ‘enjoy’ while he waits for his call to be answered.
It would be far too complicated to write out the necessary differentiation of
the articulation, and it would make reading more difficult. This attempt at an
exact notation of the Liszt waltz quoted above (Example 3.14, page 24) serves
to illustrate how such a precise notation would look:

Ex. 3.15

Countless examples of this sort might be quoted here, but what has been said
should suffice to clarify the problem. It should be equally unnecessary to point
out the difference in length between a staccato quarter note and a staccato eighth
note.
A very important clue as to how we may imagine piano tone is offered by
the characteristic qualities of a note that is sung. In the voice, a note undergoes
a transient build-up phase and a decay phase. We also find this phenomenon
in wind and string instruments. The sound neither begins nor ends abruptly. It
does not sound as if it is being cut off. With the piano, things are very differ-
ent: here the sound of the note is interrupted by the dampers in a manner that
does not correspond to its natural resonating process. With important notes,
chords or final notes, a good musician can achieve the effect of a fade-out by
grading the slow release of the pedal and so avoid the ‘cutting-off ’ of the sound
that so goes against our instinctive musical feeling. One wonders why good,
even well-known pianists evidently do not hear how they spoil an otherwise
musical performance with short notes that are quite unfitting to the character
of the phrase. The reason lies in the nature of the instrument itself. A string
player must shape the sound with both hands throughout the whole duration
of a note. If he doesn’t stop the string firmly enough with the fingers of his
left hand or continually maintain a certain pressure as he draws the bow across
the strings, the sound will no longer correspond to the tone he imagines. He
must listen to the note, keeping it under his control until the very end. The
same is true of wind players, who shape the sound with their breath. Both, like
26 Articulation

a singer, can influence the note while it sounds. The pianist, however, simply
strikes the note and has no more influence over it, even if he wanted to, and he
has become accustomed to pay no more attention to the way the end of a note
can be shaped, directing his attention only to the notes that follow. This is no
doubt the reason why Beethoven advised pianists to have a competent violinist,
wind player or singer perform the passage in question for them (see endnote 1
at the end of this chapter).
Owing to the small size and lightness of the jack, even on good harpsi-
chords the notes continue to sound slightly after the keys have been released.
This accounts for the otherwise inexplicable staccati in expressive themes as in
Examples 2.14 (Haydn, page 12) and 3.8 (Beethoven, page 21). In contrast, on
a modern piano the sound is abruptly interrupted by the considerably heavier
dampers. Therefore, in expressive passages in a slower tempo, in a portato or even
staccato, one should allow the notes to resonate for a shorter or longer time by
using pedal on each note, achieving the desired length necessary for a particular
expression by a careful release of the pedal. This effect may also be executed
with the fingers alone—albeit with somewhat greater difficulty. Our musical
sensibility and feeling dictate the necessary execution.
A musician will of course also wish to influence the onset of the note, too.
At times he may wish to begin the note gradually, at times suddenly in full
loudness. Every wind or string instrument is suited to this. Even the original
mechanical organ action allowed for a certain influencing of the start of a note.
This earlier type of action is often preferred today, as the sudden breaking in
of the note at full volume, usual on organs since the end of the nineteenth
century, goes against the players’ musical sensibility. Attempts to improve the
harmonium’s gradual, somewhat oily start to each note with clappers proved
unsuccessful—something which can perhaps account for the instruments failure
to achieve a greater popularity. On the piano we cannot influence the transient
build-up phase of the note. This can only be achieved by adjusting the hammer
mechanism itself—varying the regulation can allow for a harder or softer touch.
By altering the distance between sections of the keyboard and the double repeti-
tion mechanism exclusively used today, the technician can influence the effect of
the attack to a certain extent and thus the piano touch. This is the reason why
some famous pianists only give concerts on their own instruments.
Differences in tone-production between piano and orchestral instruments
can become unpleasantly conspicuous. In piano concertos—as in the example
below by Mozart—one can often hear in the staccato playing of the orchestra
how the sound of each chord continues to resonate slightly before decaying. If
the pianist—and this is something we hear all too often—also plays the same
theme staccato, but without pedal, the difference is immediately noticeable: the
staccato notes of the piano sound mechanical; the abrupt clipping of the notes
contradicts the physiological process of sound production in singing that string
players emulate. The phrase loses expressiveness and, with this, its meaning.
 Articulation 27

The following example from the Allegretto of W. A. Mozart’s Concerto in A

major KV 414 illustrates this:

Ex. 3.16
24

In this passage, the notes A on the accentuated beginnings of bars 2 and 3

sound like bells if one allows them to resonate into the rests with the pedal.
Played in this way, the motive sounds cheerful and charmingly graceful. If one
fails to let the A resonate, but instead cuts it off exactly at the end of its written
duration, the phrase becomes an uninspired, meaningless up and down movement
to A and back, and on top of this, the phrase is senselessly interrupted by the
abrupt hiatus in the sound. As so often, it is the sound that makes the music.
It should also be mentioned that at the piano entry, the upbeat to bar 88 in the
left hand is to be played staccato, exactly like the violas and cellos. Almost all
pianists inexplicably play legato here.
A further example is the first eighth note of the Rondo of the great A major
Sonata D 959 by Schubert—a note that is also played with pedal and resonates
into the rest to match the cheerful character of the piece and to avoid an inadver-
tently comic or coldly dismissive effect, depending on how short or loud one plays
the note. The eighth notes in the fourth bar are to be slurred together until the
C-sharp in the fifth bar, although the phrasing mark only extends until the D. This
corresponds to the customary Classical notation that can be found in the works of
Haydn, Mozart or Beethoven. Again, dynamic differentiation is also required—the
second bar is played quietly, while the third undergoes an increase in volume that
reaches its climax in the fourth bar (six-four chord with subsequent resolution):

Ex. 3.17
25

The excessively long reverberation time even of good harpsichords and fortepianos
was evidently considered a problem for the clarity of a harmonic progression. This
may account for the many seemingly inexplicable situations in which notes are
written overly short. When, in 1773, Johann Andreas Stein introduced his improve-
ments to the pianoforte mechanism, which allowed for a much better dampening
28 Articulation

of the sound, both Mozart and Beethoven praised the new instrument. The old
English and French pianos had much smaller and lighter dampers than Stein’s
instrument. Such dampers were only marginally more efficient than those of the
harpsichord. If at that time the main problem was the overlapping of harmonies,
today it is the abrupt short ‘hacking’ manner of some pianists. A cantabile phrase
is not compatible with a dry, excessively short cutting-off of the notes.
However, commensurate with real life, music evokes a wide diversity of emo-
tions, including the dismissive, the brusque, or passages of great drama that require
corresponding dynamics and articulation. When called for, the singer, string or
wind player is equally capable of adjusting his articulation. Such situations can
call for the shortest of staccato touches. The following passage in Beethoven’s
Sonata op. 7, for example, demands such staccato.
Ludwig van Beethoven, Sonata op. 7, second movement, bars 20–21:

Ex. 3.18

Here is a passage that demands a range of staccati from longer to very short from
the first movement of Beethoven’s Sonata in D minor op. 31 no. 2 bars 159–169:

Ex. 3.19

On its first appearance in bar 159–160, the motive calls for a staccato of medium
length, corresponding to the pp dynamic. Following a natural crescendo through
a louder, curter execution on its second appearance (163–164), the motive is
finally played ff with a staccatissimo (167–168). The player thus adjusts dynamics
and articulation to the character of the phrase or motive, rather than employing
a single undifferentiated staccato throughout.
 Articulation 29

Sometimes abrupt, impetuous motives can call for both staccato and pedal, as
here in the third movement of Beethoven’s Piano Sonata No. 8 op. 27 no. 2:

Ex. 3.20

Here Beethoven indicates staccato and pedal at the same time. With the two
staccati, he indicates the intensity and shortness with which the notes must be
played; at the same time, he wishes a fuller sound and so demands the use of pedal.
This is the beginning of Sergei Rachmaninoff ’s Études tableaux op. 39 no. 6:

Ex. 3.21

Here are two runs that end brusquely in a staccato with doubly notated accents:
though amounting to a vehement, aggressive ‘dismissal,’ they cannot be played
too short, as the runs end with a staccato quarter note, not an eighth.

Differentiation Between Staccato and Legato

As Haydn frequently writes no indications of articulation, the player must decide
what articulation is suitable for the phrase. Fast runs in works of Mozart and
Haydn, even when slurred, are to be played slightly non legato—leggiero, almost
30 Articulation

legato—in contrast to Beethoven, who makes an exact distinction between legato

and non legato in the notation of runs. An example of such differentiation is
found in the Allegro of the first movement of the Sonata op. 27 no. 1. Here are
the first eight bars:

Ex. 3.22

The arpeggiations at the beginning should not be slurred. Since the notes are
already very short—sixteenths—he does not notate staccato, though he does so
in the following eighth notes in the second bar, in order to keep them as short
as the sixteenths. The final note of the phrase is naturally longer. This point
will be discussed at greater length in the next sub-section. After a repetition of
the phrase on the dominant, two bars of legato follow, with a crescendo that
Beethoven concludes with a sudden piano and once again non legato. This varied
articulation is ignored by most pianists.
At the beginning of the fourth movement, Allegro vivace, of the same sonata,
the left hand plays alternately two bars non legato, two bars legato.

Ex. 3.23

In the last movement of the Hammerklavier Sonata op. 106, almost all the six-
teenth notes are non legato. After bars 153–183, which are marked with legato
slurs, Beethoven writes non legato in bar 184: a clear indication of how all the
other sixteenth notes should be played. Friedrich Gulda and Badura-Skoda both
make this differentiation in articulation.
 Articulation 31

Let us consider the following passage from the second movement of Beethoven’s
Sonata in E major op. 109, Prestissimo, bars 9–25:

Ex. 3.24

Here the player must make an exact distinction between non legato in bars 9–10,
13–14, 17–18, 21–22 and the intervening legato passages that Beethoven even
marks twice with the words legato. These clearly intended differentiations are
unfortunately disregarded by many pianists.
In the second movement of his Sonata op. 81a Les Adieux—which Beethoven
entitled Abwesenheit (absence)—the thirty-seconds in bars 13–14 and 29–30 are
to be played non legato. This articulation is in accordance with the sad character
of the passage: a rippling, mellifluous legato hardly portrays great suffering. Only
with the poco ritardando at the close do the notes necessarily become longer and
longer—which Beethoven implies with the slur over dots.
Bars 13–14:

Ex. 3.25

Bars 29–30:

Ex. 3.26
32 Articulation

In his early works, Haydn wrote fewer articulation marks. That does not mean
that everything is to be played portato or legato. It is up to the performer to
decide which sort of articulation seems suited to the piece, or indeed to the bar
or phrase in question. Hence runs without a slur are often intended as legato. It
has already been mentioned that such legato runs in Haydn and Mozart are to
be played leggiero.
This peculiarity of sometimes avoiding a slur over the bar line was customary
at the time of Haydn and Mozart and can still be found in works of Schubert.
In such cases, one must decide whether this is intended as an articulation to be
executed exactly as indicated, or whether the slur should indeed be continued
over the bar line. This is sometimes difficult to determine. The practice prob-
ably originated in the transfer of violin bowing marks to piano music. Reasons
of printing technique that are often cited are inconclusive, since it was quite
possible to engrave long slurs in the copper printing plates used at the time.
The slurs across the bar line almost always begin on an accentuated note. In
addition, one can see how the slurs join smaller and larger groups of notes together
to form a logical unit, thus perhaps making them more intelligible. The opening
theme of the first movement of Mozart’s Sonata KV 333 is just such a passage.

Ex. 3.27

Since this is a cantabile melody, today pianists frequently join the two legato
slurs into one,4 or even apply a single slur for the antecedent phrase of the
period and one for the consequent phrase. This tendency partly to ignore the
indicated articulation and to slur notes together into larger units is perhaps
to be explained by the influence of the long legato cantilenas of Romantic
music that have found their way into the playing of Classical themes. To what
extent one employs a more differentiated articulation is ultimately a question
of style and good taste. Following the original slurring, without exaggerating
the articulation, produces a more graceful effect. For instance, at the beginning
of his Sonata KV 570 (see Example 2.3 on page 6), Mozart writes a separate
slur over each bar of the four-bar phrase, whereas on its appearance in bars
101–115 he writes slurs over three bars.
As has been mentioned before, it was customary at this time to write shorter
slurs, assuming the practise of bowing notation for the violin where fewer notes can
be joined together. It was also common to avoid slurs across bar lines—although
 Articulation 33

the player naturally would not separate the final note from the theme. So, what
should one do in this case? The character of the theme would suggest articulat-
ing the phrase as Mozart notates in bars 101–115, ending each phrase on the
first note of the bar that follows the three slurred bars. However, should one not
wish the beginning of the sonata to conform to the later articulation, the slurs
must be separated very subtly to avoid chopping up the theme.
The following thoughts are intended as notes as to how such problematic
passages might be performed. Upbeats, especially leaps, were as a rule not slurred.
A typical example of this is the Sonata in G major KV 283 by Mozart, where
he does not slur the upbeat into the next bar but does write a slur across the
bar line above the two notes that form the close of the motive:

Ex. 3.28

This is a clear indication that Mozart, when he intended to, also wrote slurs
across the bar line. It is sometimes difficult to make the right decision. Notation
was often inconsistent.
During Haydn’s early life, the harpsichord was still the customary keyboard
instrument. Around 1763, Leopold Mozart acquired an instrument by Christian
Ernst Friderici for his children that had both a harpsichord manual and a manual
with a hammer mechanism. The capability of dynamic differentiation was very
limited on this instrument and on the clavichord. On the harpsichord, such dif-
ferentiation was impossible. As mentioned earlier, not until 1771/72 do the few
dynamic indications appear in Haydn’s piano works. The significant improvement
of Andreas Stein’s ‘Deutsche Pianofortemechanik’ did not appear until 1773.
Therefore, during this earlier period, composers used agogics and a highly dif-
ferentiated manner of articulation for the convincing presentation of a phrase.
Repeated accompanying figurations were slurred together into groups of
three, four or six notes. Naturally, this does not indicate articulation and one
must of course employ a continuous legato. If the last note of a run or of a
whole phrase ends in a new bar, it was not customary to draw the slur across
the bar line. Whether one employs a greater or lesser degree of strictness in the
articulation of problematic phrases is left to the discretion of the interpreter’s
judicious sensibility and feeling for style. There is most certainly not only one
possible way of doing things. A prescribed formula will not be of help here. A
more detailed and thorough treatment of these problems would go beyond the
scope of this book. In his book about Mozart, mentioned earlier, Paul Badura-
Skoda deals thoroughly with this subject.
 34 Articulation

Notes at the End of Motives and Phrases

Final notes on strong beats at the end of phrases or motives are often played far
too short. Such notes need time to fade naturally, as in the following example
from the first movement of J. Haydn’s Sonata in C minor Hob. XVI/20, bars
32–35 and 47–51. In this passage, figures of four thirty-second notes each end
with a staccato eighth that needs to be played appropriately longer than the
preceding notes:

Ex. 3.29
15

Haydn writes a vertical stroke above the final note of each of these five-note
groups. Here these strokes indicate nothing other than a staccato. With the goose
quill—at that time the only writing-implement available—one could not write
a dot and it became a line (see below, page 45).
Though Haydn likes to contrast p and f passages,5 the forte in Example 3.29
could indicate a crescendo. A crescendo, rather than an abrupt forte, seems better
to suit the phrase’s progression into the agitated breathlessness of bar 35, where
the two-bar unit is compressed into a single measure. The staccato eighth notes
are correspondingly longer than the thirty-seconds and are played with a pedal
fade. They also stand in places of harmonic importance—which is another reason
not to play them too short. Starting in the relative major, the first staccato eighth
note stands on the dominant of this key. The next eighth note leads back to the
relative major, then the same is repeated.
The phrase in Example 3.30 (bars 47 to 51) first presents the same harmonic
progression in B-flat minor, moving quickly to D-flat major in bar 51 via the
dominant of D-flat in bar 49. This shift from minor to major requires a cor-
responding increase in dynamics. As eighths, the end-notes are notated with four
times the duration of the preceding thirty-seconds: to play them with too short
a staccato ‘cut-off ’ would entirely distort this very appealing music.
Let us look at these phrases in more detail. In both the E-flat major
(Example 3.29) and the B-flat minor section (Example 3.30), the repeated
figures (thirty-second notes ending with an eighth) alternate harmonically
between the fifth and first degree of the key in question, whereby Haydn
 Articulation 35

always lets the tonic fall on the first quarter note—a strong beat. However,
since the tonic is usually felt as the resolution of the dominant and is thus
usually played quieter, one might, at least in the B-flat minor section, per-
form this as a variant, thereby playing the third quarter note louder than the
first. As dynamics came to be notated more precisely, such deviations from
the usual pattern of emphasis, if they were justified musically, were mostly
indicated by the composers.

Ex. 3.30
16

The commonly recurring mistake of an abrupt cut-off on the final note of a

section, motive or phrase often occurs when the right hand has to leap up after
the note to a higher register and so tears off the final note of the phrase with
a staccatissimo. In such cases, this can be alleviated by a short touch of pedal on
the final note.
In the Allegretto grazioso of Mozart’s B-flat major sonata KV 333 (Exam-
ple 3.31), a run of sixteenth notes in bar 23 ends in a staccato eighth note, which
is naturally to be played longer than a sixteenth. This is best achieved through
a short touch of the pedal, since this would otherwise mean playing the final
staccato much too short, due to the necessity of the quick leap downwards to the
major third, F-A, which would produce an undesirably comic effect.

Ex. 3.31

Such final notes or chords are the result, the résumé, of a musical process: a
cadence, a phrase or a larger section culminates here in its essential point, the
meaningful result of a musical development. It reaches its goal, its ending and
 36 Articulation

comes to rest. Such a final note is thus always to be played a little longer, as
befits its importance. Even the last note of a simple motive often needs to be
sustained slightly longer.
In the first movement of his Sonata op. 2 no. 2, Beethoven, who as far as
possible notates very precisely, indicates the variety of lengths to which notes
must be sustained to do full justice to musical logic and to ensure a vitality of
expression. Thus, the upbeat is a staccato eighth note, therefore shorter; the strong
beat is a staccato quarter note—somewhat longer; and the final note of the two-
bar motive is a full quarter note—very long. A lively sense of differentiation
forbids an equally short execution of all three notes:

Ex. 3.32

19

Consistent to his sense of musical logic, Beethoven himself must have imagined
these various durations in this way and could notate them in no other manner.
A similar example is to be found in the first movement of the Sonata in F
major op. 10 no. 2.

Ex. 3.33

At first, we see the same picture as in the previous example: the upbeat is a stac-
cato eighth, the downbeat a staccato quarter; after the upbeat triplets, again the
staccato quarter. Bars 5–12 are a cantilena, whose antecedent phrase in bar 8
closes with a full eighth while the consequent phrase in bar 12 closes the period
with a full quarter. Here again we see the logical relation of music to language.
A clause closes with a comma, the whole sentence with a full stop. In the same
way the antecedent phrase closes with a shorter, the consequent phrase with a
longer final note.
 Articulation 37

In the Waldstein Sonata op. 53, the motive in the third bar also ends with
a staccato eighth note, whereas the phrase closes in bar 4 with a staccato quarter.
This pattern is repeated in bars 7–8, 16–17 and in bars 20–21.

Ex. 3.34

Beethoven generally notates very precisely. What do other similar motives look
like in his works? Example 3.35 is taken from the first movement of his Sonata
op. 13, bar 51 onwards. While both antecedent phrases begin with staccato, they
close with a full-length quarter. The consequent phrases likewise close with a
full quarter after two staccato upbeats (bars 58–59/66–67 and 74–75/78–79). In
this case, the unbroken flow of the passage makes it impossible to close anteced-
ent and subsequent phrases with notes of a different durational value.

Ex. 3.35
38 Articulation

In the first movement of the Sonata op. 14 no. 1, bars 5 and 6, both hands play
alternate figurations consisting of eight sixteenths notes closing with a quarter:

Ex. 3.36

We see the same picture in the first movement of the Sonata op. 14 no. 2:

Ex. 3.37

Here, in bars 63–70, as at the start of the movement, there appear short and
longer motives, all ending with longer notes. Even the short four-note motive
repeated three times, each beginning with three sixteenths, always ends with a
staccato dot under the slur—in effect a portato—in any case longer than the
sixteenths. If a phrase consists of eighth notes, then Beethoven ends with a
quarter. The same occurs in the Sonata op. 27 no. 1 first movement, bars 38
and 40, where three staccato eighth notes close with staccato quarter notes:

Ex. 3.38

We find this logic in the motivic endings of a later era such as here in a work
of Sergei Prokofiev—his Visions fugitives no. 5—where in each bar sixteenths
or staccato eighths close with a full eighth.
 Articulation 39

S. Prokofiev, Visions fugitives No. 5:

Ex. 3.39

Or here in a work of Ravel, where, after nothing but portato notes in the right
hand (abbreviated sixteenths) the motive closes in the following bar with a
quarter slurred to a preceding full eighth note.
Maurice Ravel, Sonatina, second movement, bars 15–16:

Ex. 3.40

In the third movement of the sonatina, bars 4–8, we recognise the same logical
musical principle at work in Ravel’s imagination, even though the different
lengths of the notes are not audible, due to the pedal.

Ex. 3.41
 40 Articulation

The three-note motives in the first bars of Haydn’s Sonata No. 53 in E minor
Hoboken XVI/34, for example, make up a cadential phrase that leads via C
major to B major—the dominant—and is then reinforced by a threefold repeti-
tion of one of the motives. These final notes also appear on stressed beats and
are notated without staccato dots. A logical presentation of the whole phrase
requires that the harmonic progression be made clear. If one were to play these
final notes too short, the listener would not be able to understand the phrase
and all that is left would be the rhythm, which in no way could do justice to
what is going on musically.

Ex. 3.42

How could Haydn have written it differently? Notating these three eighth notes
in the right hand as staccato quarter notes would have made them too long and
would have sounded too ponderous. A slight lengthening of the eighth notes—with
a tied-over thirty-second note, for example—would have been much too compli-
cated and would only make reading difficult. The last note of the broken E minor
chord in the bass should be played correspondingly longer as well: something that
Haydn clearly indicated by omitting the vertical stroke (staccato) above the fourth
20
note but is unfortunately overlooked even in some urtext editions.
Haydn’s Sonata in E-flat major XVI/52 first movement, bars 27–29, and many
other passages in the same movement, follow the same principle:

Ex. 3.43

While in the left hand a full eighth follows two staccato eighths, in the right
hand we hear bell-like repetitions of a two-note motive consisting of a
 Articulation 41

thirty-second note upbeat to the sixteenth notes which form the melodic line.
To avoid losing the graceful, almost dance-like character of the phrase, these
sixteenth notes must not be shortened too much. On the contrary, these notes,
which constitute the melody, should resonate a little into the rests, aligning
them with the differing emphases of the left hand. The musical conception
of how a phrase best sounds—and Haydn no doubt had the optimal execution
in mind—does not always coincide exactly with its notation. Only forty-one
years later Schumann wrote similar phrases in his Sonata in F-sharp minor,
indicating with rests how the phrasing is to be imagined, while embedding
the passage with pedal, in accordance with the tonal ideal of the Romantic
era (see Examples 5.15 and 6.10).
Here is a passage from the Adagio of Mozart’s C minor Sonata KV 457, bars
21 and 22:

Ex. 3.44

Here a motive consisting of four thirty-seconds occurs twice, each time delib-
erately closing with a sixteenth note. Often enough, however, the composer
doesn’t think it necessary to indicate that the last note should be played longer,
either because this would be too complicated, or because the note finishes on
a strong beat and therefore should be played longer anyway. Nonetheless, as we
have explained, such final notes should be played longer if the passage is not to
sound comical. The question of how much longer will be determined by the
character and feeling of the phrase in question.
Perhaps the most blatant instance of the disregard for sustained closing notes
is the motive with which Chopin’s B-flat minor Scherzo op. 31 begins. The
motive closes with a quarter but is generally played today staccatissimo.

Ex. 3.45
 42 Articulation

We know from a pupil of Chopin that the composer attached great importance
to this passage and was never satisfied by the way his pupils played it. For him,
this passage was crucial to the whole scherzo.
Here is Wilhelm von Lenz, a Russian imperial official of Baltic descent
(1808–1883), writing about this motive:

Right from the start we struggled with a difficulty, because it was

never possible to play the two seemingly harmless groups of triplets
to Chopin’s satisfaction. “It must be a question,” Chopin said in his
lessons, and one could never play questioningly enough, never piano
enough, never sufficiently tombé [letting oneself fall]. Never with
enough meaning, as he used to say. “It should be a house of the dead”
he said once. I saw how Chopin spent much time with this bar in
his lessons, always coming back to this figure, at the start of the move-
ment and wherever it re-occurs. “It is the key to the whole work,” I
heard him say. Now this group of triplets is generally simply torn off,
or at best swallowed.6

The point could not be made more clearly.

The typical mistake of playing the last note of a phrase or motive much
too short and with an abrupt cut-off is often heard at the beginning of the
third movement of Mozart’s Sonata in D major KV 576 with its buoyant,
amiably graceful theme in an allegretto tempo. Although in the Bärenreiter
urtext edition the final notes of bars 2 and 4 and bars 10 and 12 are marked
staccato, as final notes they should not be played too short. In this context,
it should be mentioned that in the 1805 Offenbach edition of André, a
staccato is only written at the end of bar 2, not, however, in bars 10, 12,
29, 35 and 65f and many other analogous passages. Perhaps André is more
reliable.

Ex. 3.46

22

23
And certainly not as in AUDIO FILE 23.
 Articulation 43

J. Haydn, Sonata in G major Hob. XVI/40, first movement, bars 61–62:

Ex. 3.47

This is a typical example among countless in the literature where the player
is seduced into the easy habit of jumping from the last note of each motive,
allowing the hand to fall comfortably and relaxed into the next motive in
disregard of Haydn’s indication, quite unaware of the musical absurdity of
neglecting precisely those notes on the strong beats that constitute the har-
monic sense.
A lack of understanding on the part of the interpreter will often produce a
brusque clipping, even of melodic notes—when two notes are written beneath
a slur, players frequently abruptly cut off the second note. The following passage
from Schumann’s Kreisleriana op. 16 is an example among many:

Ex. 3.48

Here in the Intermezzo I of the second piece, the upper voice forms a
cheerful phrase that leads harmonically from B-flat to the dominant, F major.
Regrettably, one repeatedly hears how the flow of the melody is chopped to
pieces by these misplaced staccati on the second note of each slurred two-
note group, so that the musically significant notes, deliberately placed by the
composer on strong beats and without staccato, are replaced by a typically
clipped repetition of the two-note motive, accented sixteen times. Music
mutates into caricature.
A further example of the same problem with groups of two notes joined by
a slur is to be found in the Scherzo of Schumann’s G minor Sonata from bar 33.
 44 Articulation

Here, too, clipped eighth notes would allow this highly romantic passage to descend
inadvertently into absurd comedy:

Ex. 3.49

49

50 AUDIO FILE 50 represents a misreading of the composer’s intentions.

As has already been shown, composers repeatedly demonstrate the importance of
lengthening final notes by notating such notes with great precision. Often enough,
it proves necessary to follow our musical intuition and to lengthen a note a little
into the rest, at the same time avoiding an unnatural clipping of the note by a
careful release of the pedal. If one wanted to set this down explicitly in writing,
one would have to add on a sixteenth, or even just a thirty-second, tied to an
eighth or quarter note, shortening the rest by a sixteenth or thirty-second—which
would only make reading more difficult and would be too time-consuming for
the composer. One can see what this would look like in Example 3.15, page 25.
The composer expects an appropriate musical insight and feeling for the right
expression from the player, as these articulation marks are only intended as a regrettably
inadequate indication of the emotional content and are never to be taken literally.
Notation is just a shadow of the living music. One should not take it at face value.
Countless such examples can be found in the literature. The above must
suffice to illustrate the importance in playing longer closing notes, other than
when the character of a piece or phrase demands brevity as an expression of a
particular emotion, whether comic or brusque.
Great composers refer to the natural circumstances of the music and notate
these logically. The true musician among pianists will sense this logic intuitively
and play accordingly. Here a characteristic example by Beethoven, in which the
same phrase is articulated differently on its second appearance in accordance to
its different expressive content. It clearly demonstrates how greatly the shortening
of motivic end-notes can influence the expressivity of a phrase.
L. van Beethoven, Sonata in A-flat major op. 110, third movement, bars 9–12:

Ex. 3.50
 Articulation 45

While bars 9–12 are slurred without rests almost throughout, in bars 116–119
the articulation changes completely. Although the notes are the same (here in
G minor) in order to correspond to the indicated expression of exhaustion—
ermattet, klagend—the individual motives of only a few notes each end with
sixteenths or even thirty-seconds followed by a rest, as if expressing the breathless
and tired resignation by interrupting the legato cantilena:

Ex. 3.51

The passage reveals the expressive possibilities at the disposal of the sensitive
musician who is aware of how to shape a note throughout its shorter or longer
duration and at its close.
As we have said, our notation only allows for a stepwise gradation of note-
lengths and rests—whole note, half note, quarter note etc. Although intermediary
values can be represented by ties connecting different note-values, a note some-
times needs only to be lengthened very slightly: something that is not notated,
as this would be practically impossible, or too complicated to write and thus
would be more difficult to read. Its execution is left to the discretion of the
performer. While our notation possesses a sign for the shortening of a note—the
staccato dot—there is no sign for a slight lengthening of the note (the dot after
the note lengthens it by a full half of its value). In music, there are no strictly
defined borders, only fluid transitions from one durational value to the next.
These are randomly chosen passages from countless examples in the litera-
ture. They must suffice to show how great composers conceived the logic and
expression of motives and phrases. In written language, the sentence ends with
a dot, in speech with a pause; in music, the phrase ends with a longer note. It
is simply human nature.

The Different Meanings of Staccato Markings

One often finds vertical strokes above the notes in old manuscripts and early
printed editions. They can have different meanings. Above a longer note, such
as quarter in a slower tempo, the sign can stand for an accentuation. That is to
say, it has dynamic significance. In most cases, however, it is intended as an
46 Articulation

articulation mark. This meaning has its tradition: at a time when one still wrote
will quill feathers or a steel nib, it was virtually impossible to notate dots, so one
made do with lines—as the quill was lifted from the paper, the line became
wedge-shaped. Later printed editions either adopted the staccato-strokes or replaced
them with dots. The Henle editions of Mozart and Beethoven thus notate the
vertical strokes of the autograph as dots. The old urtext editions of Universal
Edition, on the other hand, retain the cuneiform strokes for the sonatas of
Beethoven.
Staccato markings may sometimes only indicate an emphasis and are not to be
considered as articulation signs. This is illustrated in the following passage from
the first movement of Beethoven’s Sonata op. 110 from bar 12.

Ex. 3.52

In the old Universal edition of this sonata, there are wedge-shaped markings over
the notes of the broken chords in the first movement from bar 12. In the Henle
edition, these have been replaced by modern staccato dots to avoid any misunder-
standing. In the autograph there are vertical strokes, which, as previously mentioned,
not only might mean staccato, but could also indicate an accentuation. In this case,
a slight accentuation seems to be the proper manner of execution.
In the above example, Beethoven indicates leggiermente, not legato. Thus, the
staccato dots signify a light accentuation that is not always applied in the same
way. On the tonic in bar 12, each eighth note is marked with a dot on the first
thirty-second, on every A-flat. However, in the next bar on the dominant, he
 Articulation 47

only places a staccato on the first thirty-second of each quarter note. In bars 14,
15 and 16, the same picture is repeated. From bar 17, a crescendo begins, and the
dots—undoubtedly intentionally—cease. He evidently wanted a regular increase
in volume without additional disrupting accents. In any case, the accompany-
ing chords in the left hand result in the usual slight differentiation between
strong and weak beats. Bearing in mind the exactness of Beethoven’s notation,
we should not assume he intends the staccato dots to be continued in the same
manner throughout bars 17 and 18 when he could have omitted them at the
repetition of the phrase in bars 14 and 15. As so often the case, he indicates a
subito piano at the climax of the crescendo. He evidently wishes the climax to be
not in the loudest, but in the most beautiful sound.
Another example is to be found in the first movement of Schumann’s Sonata
in G minor op. 22:

Ex. 3.53

In bars 93–96, Schumann places staccato dots on every note, at the same
time indicating the use of pedal. The legato produced by the pedal proves the
staccato dots are indeed not signs of articulation but must indeed have another
significance—and this can hardly be any other than that of an accentuation.
R. Schumann, Humoreske op. 20:

Ex. 3.54
48 Articulation

In the middle voice that fills in the harmony, there are staccato dots under
a slur, even though the passage is marked Pedal. Since these notes are placed
on the weaker beats in contrast to the previous example, they cannot indicate
accentuation. Why, then, this painstaking notation? The piece begins in piano.
With the articulation markings, Schumann evidently wishes to make clear that
the middle voice is to be played lightly, merely touching the notes, in order to
allow the moving phrase in the main voice to sound with a clearer cantabile.
He has divided the melody into a question and answer and this must be cor-
respondingly shaped both agogically and dynamically.
A very similar situation occurs in the following passage from Liszt’s Konzert-
Etüde No. 3 (Il sospiro) from Drei Konzert-Etüden:

Ex. 3.55

Here, too, the staccato eighth notes—in this case in the main voice—are an indi-
cation to delicately touch the keys while sustaining the pedal.
Very often, staccato dots serve a further function: they merely illustrate the
requisite arm or finger movement. This is referred to as ‘technical notation’7
(see the chapter on notation, page 72).

Notes
1. “In cantilenas, he referred to the method of well-trained singers, who neither do too
little nor too much. Furthermore, he sometimes advised that one set suitable words to
a difficult passage and sing them, or that one listen to such passages played by a well-
trained violinist or wind player.” Johann B. Cramer, 21 Etüden für Klavier nebst Fin-
gerübungen von Beethoven nach dem Handexemplar Ludwig van Beethovens, Universal Edition,
Vienna, 2013, ed. Hans Kann (o. J.).
2. “Like the sound of a bell, that, if strongly struck, gradually dies away.” (Wie der Klang
einer Glocke, wenn sie scharf angeschlagen wird, sich nach und nach verlieret.) Versuch einer
 Articulation 49

gründlichen Violinschule, Selbstverlag, Augsburg, 1756 I. Hauptstück, III. Abschnitt §19,

p. 44.
3. “Es kommen überhaupt bey der Musick viele Dinge vor, welche man sich einbilden
muss, ohne daß man sie würcklich höret.” Carl Philipp Emanuel Bach, Versuch über die
wahre Art das Clavier zu Spielen, Selbstverlag, Berlin, 1753, I. Theil, II. Hauptstück, III
Abtheilung §20, p. 78.
4. The slur above the last three eighth notes in the right hand in bar 1 appears in the
first edition. It is not in the original autograph.
5. In this context, reference should be made to the very instructive foreword by Christa
Landon in the Vienna urtext edition of Haydn’s sonatas.
6. Übersichtliche Beurtheilung der Pianoforte-Kompositionen von Chopin, Neue Berliner
Musikzeitung, Berlin, 1872.
7. The term technical notation comes from Josef Dichler, Der Weg zum künstlerischen Kla-
vierspielen, Doblinger Verlag, Vienna, 1948.
4
AGOGICS

The concept of agogics refers to all variations in tempo that deviate from a
metrically exact manner of playing. A ritardando or accelerando denotes that the
playing should slow down or speed up but does not specify by how much. A
rubato indicates irregular tempo in the playing, but it in no way tells us how
this is to be done. The same is true of the rhythmic irregularities in a waltz or
a polonaise. These are things that one cannot write down: a musical player must
feel such things or, through the teacher, acquire an understanding of them from
the musical tradition. Every phrase calls for agogics. It aims towards its climax,
subsequently subsiding with a hardly perceptible ritardando and diminuendo.
As living speech, music too has at its disposal the parameters of dynamics,
articulation and agogics. Without agogics, music does not breathe: it remains
mechanical, lifeless. Even a strict fugue has its agogic nuances, albeit very subtle
ones. It is like salt in the soup: if you notice it, then it is too much.
Beethoven’s sonatas very often give the indication a tempo after an espressivo.
Beethoven evidently intended that the expressive phrase be executed more slowly.
In the otherwise strict formal style of the Classical period—even in the
works of Mozart or Haydn—one should not be afraid to observe such tempo
adjustments to avoid playing passages of very different character in a mechanical
way. To maintain the tempo in all too strict a fashion would be a mistake—it is
not in our nature to do so, nor is it in the nature of living music. An example
would be the last movement of the Sonata in F major KV 332 of Mozart, where
the second section of the exposition from bar 50 cannot be played in the same
tempo, due to its different character. A nuance slower is enough. Too much
would cause the unity of the movement to fall apart. In music of the Romantic
era, the tempo is treated much more freely anyway.
We frequently encounter passages that require a freer treatment of tempo,
such as the recitative-like passages in the freer forms in the works of J. S. Bach.
 Agogics 51

Instrumental recitatives are also to be found in string quartets, for example by

Beethoven. As the name suggests, these are phrases that are to be played as if
recited in a cantabile manner—and this itself indicates the necessary timing: rhythm
and tempo correspond to an analogous, imaginary spoken text. If one assumes
too slow a speed, the recitative loses its liveliness and naturalness.
Such a recitative takes time to get going, as it were. It starts somewhat slower,
gets quicker and finishes again more slowly. It is as if it were set to words to be
declaimed in the natural tempo of speech. Here, as elsewhere, uniformity is fatal
to any vital interpretation of a piece of music. Hence, in the fourth movement
of his 9th Symphony for instance, Beethoven anticipates in the double basses
the recitative that the bass singer later actually sings (to Beethoven’s words!).
Beethoven’s sketches show that in bars 8 to 90 he originally wrote in the text
to the instrumental part (the text was deleted in the printed edition). When
faced with a recitative, the instrumentalist would do his best to imagine a text
to help him play the passage with rhetorical conviction. Even with a cantilena,
Beethoven advised underlaying it with suitable words (compare Beethoven’s
advice, as reported by Schindler, endnote 1, Chapter 3, page 48).
Here is an example of a recitative from J. S. Bach’s Adagio from the E minor
Toccata BWV 914. An Adagio in a work of Bach only means that the tempo
should be taken a little slower than the integer valor, the heartbeat of Baroque
music, in the same way that a presto merely signifies a somewhat faster tempo.

Ex. 4.1
32

The first five notes (E, F-sharp, G, D-sharp, E) are the carriers of the text, as it
were, while the sixteenths which follow in imitation of the rhythm in the lower
parts are only accompaniment. If this broken chord is played much too slowly,
as one often hears and with an agogically exaggerated expression as well, the
playing sounds ludicrous and doesn’t make the slightest sense. The recitativo
character refers only to the upper voice. (Here one should invent a text to the
 52 Agogics

playing!) After the opening broken chord, the sixteenth notes lead seamlessly
into the recitative. It demands a cantabile (and parlando) way of playing, as Bach
himself describes (see endnote 4). These recitative-like passages in Bach’s music
are to be played quite freely and with much expression.
The same goes for the Sarabande from the B-flat major Partita BWV 825
of J. S. Bach:

Ex. 4.2
33

This Sarabande is nearly always played too slowly. It is difficult to follow a

metronome indication here, due to the free agogic style of playing required, but
about 50 for the quarter note would be a good medium tempo. Right at the
start, the first motive requires slight over-dotting (see the chapter on notation,
page 83) as already described by Quantz.1 It is clear that here, too, a certain
declamatory style of playing is called for, an agogically freer rendering of tempo
and ductus, analogous to speech. To play the passage metrically as notated would
not match the sense of the music, which must make a declamatory statement
and should take its model from the vivacity and naturalness of human speech.
Another example from J. S. Bach’s Well-Tempered Clavier I, the Prelude in
E-flat minor:

Ex. 4.3
34
 Agogics 53

In his book,2 Paul Badura-Skoda attempted to transcribe a good, correct inter-

pretation of this prelude into notation. The upbeat eighth notes of the prelude
are shortened a little with a marginally late entry and the dotted note is slightly
over-dotted. This avoids an otherwise rather ponderous and sedate impression
and conveys a more festive character. The sixteenth notes in bars 9, 11 and 12
and in the succeeding bars up to the fifteenth bar are played as a recitative—more
freely and in more of a declamatory style.
As we have repeatedly said, to do full justice to the demands of a phrase in
our playing we need to make use of all three musical parameters of dynam-
ics, articulation and agogics. Only in combination can they ensure a plasticity
of delivery. Dynamics give the phrase its curvature, the flexibility of agogics
breathes life into it and a differentiated articulation avoids any uniformity. The
harpsichord—the keyboard instrument of the Baroque—lacked the facility of
dynamic differentiation. Harpsichord players were to a certain extent able to
compensate for the deficits of the instrument through a greater use of agogics.
A note could be emphasised by a slight delay in the attack.
Instrument makers endeavoured to develop and refine the harpsichord—an
instrument that was indispensable to musicians, being the only one upon which
one could play the full harmonic structure of a composition. Consequently, a
new instrument with a hammer mechanism emerged, which finally developed
into our modern pianoforte.3 This progress came at a price: the beautiful string
sound was lost, and the clarity and beauty of the bass notes was sacrificed.
There is a certain element of comedy in the fact that some pianists, in their
effort to faithfully render Baroque works on the piano, try to imitate precisely those
deficits of the harpsichord of which composers of that time were fully aware, namely
by playing the most beautiful of phrases and their sequences without any dynamic
differentiation and with an odd sounding, thoroughly coquettish staccatissimo.
In this context, the following passage from Mozart’s Sonata KV 570, quoted
in the chapter on articulation, is a fitting illustration:

Ex. 4.4
35

To summarise our findings so far, we should observe the following in this

example: dynamically, the first note of the first bar is to be more strongly
 54 Agogics

accentuated and the first note of the second bar somewhat less so, since two
bars nearly always form a unit in which the first receives greater emphasis than
the second. Regarding articulation, the first note is to be held slightly longer
than the subsequent two weaker beats and the last note, as a final note, is cer-
tainly to be held somewhat longer. For an effective close, agogics require a short
ritardando towards the last note after the first weak beat. In this simple four-note
motive, we see the necessary interrelation between the three parameters of
dynamics, articulation and agogics, which only in combination do justice to
musical expression—something that cannot be overemphasised.
Agogics also has an important part to play in the shaping of a phrase. After
the beginning of the phrase, there is a hardly perceptible increase in tempo. In
order to maintain tension, this should not decrease and should only ease off
again when an important part of the phrase, or its climax, is reached. Hand in
hand with this goes an increase and decrease in dynamic.
An exaggerated use of agogics can spoil a phrase. The extent of agogic
differentiation varies greatly depending on the musical epoch and the type of
phrase. Here a feeling for style and good taste must decide.
To return to the example of J. S. Bach from the Well-Tempered Clavier II:
the Prelude in F minor. Right at the start, a motive consisting of three notes is
repeated four times. It is an upbeat ‘sighing motive’ that must be correspondingly
dynamically shaped: the upbeat is to be played quieter, the following eighth note
on the strong beat somewhat louder, the third eighth note again quieter and
perhaps slowing down a little. The climax is in the third bar, and even if this
is expressed dynamically, we are still left with a feeling of indifferent aloofness
if the player does not achieve a sense of aiming towards the climax through
agogics. The fourfold repetition of the motive resembles words that must be
first put together into a sentence before they make sense.
Our emotional and spiritual empathy, the extent to which we are moved
(‘agogics’ comes from agere = to act, to move!), must—as in real life—be expressed
through agogics. It breathes life into the music.

Ex. 4.5

36

Here one can hear how it sounds without agogics. It becomes nothing but a
37
mechanical, stereotyped repetition.
Bach talks about the cantabile style of playing the clavier.4 His music requires
a speaking, declamatory treatment of the phrase. And this applies to later music,
 Agogics 55

too. Haydn’s special affinity for humorous expression is unmistakeable in many

of his sonatas, and no musical joke can be made without agogics—the result
would sound very humourless indeed. Let us take as an example the Rondo
movement from the C major Sonata Hob. XVI/48, with its six-bar phrase. It
is marked presto, though it should not be played too fast.

Ex. 4.6
38

The one and a half bar long cadence of this phrase demands a slight ritardando.
It is almost like a high-spirited dialogue that ends with a question in the first
phrase and only comes to a decisive close with the second phrase. And imme-
diately after the repeat sign, it carries on in a positively mischievous way. It calls
to mind a coquettish dialogue between two people that take turns in the thirds
of the two-bar phrases. However, literal pictures of this sort mostly come to
grief, because, as E. T. A. Hoffmann puts it, music begins where language ends.5
Another equally amusing theme that should be played with a sense of fun,
and must not be simply rattled off quickly, is the Finale of the Sonata in D
major Hob. XVI/19.

Ex. 4.7
39
 56 Agogics

Unfortunately, there are also instances of well-known pianists playing a sonata

or a concerto by Mozart, Haydn or Beethoven, or even Brahms, with a strictly
uncompromising, metrical exactness. The most beautiful dynamics are here to
no avail—the playing is still soulless. It is as if one was to stick relentlessly to
strict poetic metre in the scansion of every syllable of each spoken word. Alone,
the first bars of Beethoven’s Sonata op. 101 provide a telling example of how
agogics first bring music to life:

Ex. 4.8
40

Notes
1. Johann Joachim Quantz, Versuch einer Anweisung die Flöte traversiere zu spielen, V/§21
and XVII/§58 Johann Johann Friedrich Voß, Berlin, 1752; see also Paul Badura-Skoda,
Bach Interpretation, Laaber-Verlag, Regensburg, 1990.
2. Badura-Skoda, Bach-Interpretation, p. 225.
3. J. S. Bach himself took an interest in instrument making and its development all his
life. He even initiated decisive improvements in organ building. He also took an interest
in the newly developed hammerklavier of his day and suggested improvements. Towards
the end of his life, he was pleased to say that there were already hammerklavier that met
with his approval. It is fairly certain that Bach owned such an instrument. He also
preferred the clavichord for private playing: an instrument that, to a lesser degree, allows
dynamic differentiation for an expressive, cantabile style of playing.
4. “above all to achieve a singing manner of performance” (“am allermeisten aber eine
cantable Art im Spielen zu erlangen”), title page of the manuscript of his Inventions and
Sinfonias.
5. “And this is precisely the wonderful mystery of the musical art: that where our poor
speech runs dry, for music there opens up an inexhaustible well of means of expression!”
Ernst Theodor Amadeus Hoffmann, from Die Serapions-Brüder, vollständige Ausgabe,
Erstdruck der Sammlung: Berlin (Reimer) 1819–1821, Hoffman, Berlin, 2016, p. 80.
5
PEDAL

We attribute five different functions to the right pedal of the piano:

1. A legato can be achieved by using the right pedal if notes lie too far apart on
the keyboard or if chords are to be connected (legato pedal).
2. Notes or chords of the same harmony that are played consecutively can be
made to sound together (harmonic pedal).
3. A note sounds fuller and richer if all dampers are lifted by the pedal, allowing
related frequencies of other strings to sound in sympathetic vibration (tone
improvement).
4. Sometimes it is necessary to sustain the bass without holding higher notes
or chords. Because the lower bass notes require more time than higher notes
to be dampened, by rapidly changing the pedal—or lifting the pedal very
slightly—one can sustain the bass note while dampening the higher notes to
a certain extent (compromise pedal).1
5. As discussed in the chapter on articulation, notes that are cut off abruptly by
the dampers sound unnatural and thus not very musical. Here the correct
application of the pedal helps to allow the sounds to resonate a little and decay
naturally.

Urtext editions retain the old pedal indication: under legato phrases requir-
ing a pedal change due to a new harmony, one wrote Ped. at the start of the
phrase, then at the change of pedal a star * followed by the next Ped. marking.
In such cases, however, the pedal is to be released and pressed in one action on
the same note, so as not to interrupt the legato and not—as the old notation
would seem to indicate—a little later (see below, page 63). The modern pedal
notation gives us a more precise picture.
58 Pedal

Frank Martin, Prélude No. 1 from 8 Préludes pour le piano, bars 35–39:

Ex. 5.1

The horizontal lines indicate the duration; the vertical lines mark the point of
depression and release of the pedal. This makes it clear that in bars 35 and 36
the upbeats to the dotted quarters and the two eighth notes in bar 36 are to
be played portato, while in bar 37–39 the modern notation shows that all the
notes are to be played legato and indicates exactly on which note the pedal is
to be changed.
An alternative pedal notation for legato represents the short vertical lines as
connecting diagonals, thus allowing for an unbroken pedal line.
Precise and subtle use of the pedal is not a simple matter. Exact pedal indi-
cations by the composers are unfortunately seldom to be found in the works
themselves. When such indications are notated, they only appear sporadically
or in especially problematic places. Generally speaking, in music of the Classi-
cal era—in the works of Haydn, Mozart and Beethoven—not only each new
harmony but virtually each new melodic note requires a pedal change, unless
for reasons of greater clarity and transparency of timbre; for example, in Haydn
and Mozart, one chooses to do without pedal all together in order to reduce
the ‘thickness’ of the piano sound. In a dry acoustic, this can lead to an unap-
pealing, almost sterile tone.
If one doesn’t use the pedal at all—or changes pedal on each note of the
melody—one should sustain accompanying notes of the same harmony with the
fingers of the left hand so that the melodic note is not suddenly left sounding
on its own as a single note without the accompaniment. This produces a rather
unattractively bare sound and can result in ugly harmonic intervals where noth-
ing but an open fourth or fifth is left sounding. This sustaining of appropriate
harmonic notes is known as finger pedal.
 Pedal 59

W. A. Mozart, Sonata facile KV 545, first movement:

Ex. 5.2

The four eighth notes in the left hand are simply left down until the next group.
In his book on piano playing, Johann Nepomuk Hummel, a pupil of Mozart
and one of the most renowned piano virtuosos of his time, demanded a moderate
use of the pedal and complained that pianists used too much pedal. He refers
depreciatively to the pedal as a way of covering up poor playing.2 That, of course,
only applied to music of the Classical period. In the works of Haydn and Mozart,
one must find the right balance between a sort of playing which is too dry, too
sober and an excess of pedal that obscures the necessary transparency. With long
notes or cantilenas, the use of the pedal for tonal effect is advantageous as long
as the hall does not have an overly resonant acoustic. There are halls in which a
note can resonate for up to two or three seconds. Here a player must be careful
with his use of pedal. It goes without saying that under no circumstances may
the player spoil a specified articulation through his use of the pedal.
One should make a clear distinction between legato pedal and portato pedal.
Students never have any problem with pedalling after the note (syncopated
pedal), but the portato pedal, that is, a simultaneous application of the pedal
with each portato note without falling back into a legato, often presents almost
insurmountable difficulties.
Beethoven often indicates pedal throughout longer phrases, thus allowing
different harmonies to sound simultaneously. He evidently imagined a resonant
effect comparable to the reverberation produced by the acoustic qualities of a
church. An instance of this kind of pedal marking is to be found at the begin-
ning of the slow movement from the third piano concerto in C minor op. 37:

Ex. 5.3

Carl Czerny, a pupil of Beethoven, wrote as early as 1842 that one could no
longer play this passage on the piano of his day as Beethoven had played it in
1803, with the pedal marking indicated.3 The pianos of today are of course even
60 Pedal

stronger in volume, and one must use the pedal with correspondingly greater care.
The ear must decide how to come closest to the sound that Beethoven imagined.
Whereas with Haydn and Mozart—and even Beethoven—strictly changing
pedal on every melodic note is appropriate, such pedalling is no longer possible
in works of the Romantic period. Here it is often necessary to use the pedal to
sustain both the bass and the notes making up the harmony, while the melodic
notes progress above them, since a change of pedal and the ensuing cessation of
previously sounding harmonic notes would result in an ugly progression of intervals.
The problem becomes apparent in the following extract from Chopin’s D-flat
major Nocturne op. 27 no. 2.

Ex. 5.4

One need only try changing pedal on each note of the melody (as indicated with
arrows in the example) to hear how badly the resulting intervals sound on their
own. This would produce the intervals of a bare second, then a fourth, then an
octave and fifth, all sounding separately. However, sustaining the notes with the
pedal is in no way intrusive in this music and produces a beautiful chord.
Here is an example by Brahms—his Intermezzo op. 116 no. 4, bars 15
(upbeat) to 19:

Ex. 5.5

Here the lower part in the right hand is really the leading voice. Simply to
retake the pedal on each quarter is not advisable with today’s instruments. A better
solution is to prevent the C-sharp and B in the right hand in bar 16 and the A
and G-sharp in the next bar from sounding simultaneously by changing on each
eighth, which produces a better sounding inverted seventh chord. The pedal must
be changed again on the second and third quarter anyway. From the upbeat in bar
17, portato is indicated, which also requires portato pedal. The contrapuntal texture
of Brahms’ writing demands great subtlety in pedalling if a player is to bring out
the voice leading clearly. This is even more evident in the next example.
 Pedal 61

J. Brahms, Intermezzo from Klavierstücke op. 119 no. 1, bars 45–47:

Ex. 5.6

In this passage, the sustaining of the harmony and the release of the passing
dissonances is more complicated. Brahms has taken great pains to notate exactly
which notes are to be held. At the beginning of the passage, he wishes the third
beat of the bar to sound as a ninth chord. In the right hand, he sustains the F-sharp
of the cantilena almost until the end of the bar. He likewise exactly notates that
the D of the first triplet is also to be sustained till the end of the bar, whereas the
A-sharp, which does not belong to the chord, is to be released as soon as the first
note of the second triplet—the B in the left hand—sounds. Now this B and the
following G are again sustained, whereas the subsequent F-sharp he again notates
as a sixteenth note triplet so that the following ninth chord sounds cleanly. In the
second bar of the example only the B-sharp, and in the third bar only the E-sharp
are notated as sixteenth triplets: all other notes are sustained by ties.
In both bars, the pedal may not be taken until the third beat of the bar, when
the harmony is complete, or rather a moment later, as with ‘syncopated pedal,’
so that the dampers can dampen the previous note. Players who are capable of
a subtler pedalling will take the pedal from the start and retake the pedal with
a rapid change after the dissonant note. From this complex notation, we can see
exactly how Brahms imagined the sound he wanted, and we should not destroy
this by inexact pedalling.

Incomplete Pedal Notation

Another aspect of pedal notation in works of the Romantic period should be
mentioned. Very seldom do we find a pedal marking in the works of Schubert,
and such indications occur only sporadically in works of Schumann and Brahms.
Chopin also does not include pedal markings throughout. If, in a mistaken
attempt to be faithful to the original, one adheres to the exact pedalling notation
(or rather lack of it), this not only can lead to a poor, dry and sterile performance
but also goes against the ethos of the Romantic period and all its emotional
richness. Composers generally restrict themselves to the most essential pedal
indications and do not take the time to notate pedalling in all its subtleties.
Very often we find passages in the left hand, where one must use the pedal
to sustain the bass so that the cantilena does not appear to hang thinly in the air
above it. Here is an example by Brahms—his Intermezzo op. 116 no. 4, bars 36–52:
62 Pedal

Ex. 5.7

In bars 37–48 are passages where the player must sustain the arpeggiations with
the pedal. From bar 44 onwards, Brahms writes col ped, without any more precise
indication, expecting that the player is fully aware of where the pedal must be
changed. Why does Brahms only indicate pedal from bar 44, although pedal must
clearly be used in the previous bars? In bar 44 we see a chord in the right hand
with a durational value of a half, whereby one note of the chord is notated as a
dotted quarter, the reason being that this note ends exactly when the left hand
plays the same note and the right hand must release the key. Since Brahms evi-
dently fears that the player might perhaps use no pedal, for safety’s sake he indicates
its use. Bar 48 again bears the Ped. marking for the arpeggiation, which is now
slower and longer and, as a melodic close to the phrase, might perhaps have
mistakenly been played without the pedal. With the upbeat in bar 49, a phrase
begins in which the pedalling is unproblematic, so it is again left to the expertise
of the player. Only at the close of the Intermezzo, in the fifth and fourth bars
from the end, is a pedal sign notated on the first quarter of each bar:

Ex. 5.8

Here the middle voices of the chord appear before the bass note, so Brahms
plays safe and again indicates pedal. The composer wishes the initial chord and
the bass note to sound until the end of the bar. The notated rests are simply
‘technical notation’ (see next chapter). Because of the harmony change, a player
might be inclined to change pedal on the third quarter note. However, since the
 Pedal 63

bass note should sound until the end of the bar, the composer’s intention can
only be realised by a very rapid change or partial release of the pedal (half-pedal).
Such a release allows the bass note to remain while the notes of the chord that
no longer fit the harmony are sufficiently dampened.
In this example, one sees again the usual technical notation. Rests are notated
that are not adhered to.

Problems of Older Pedal Notations

The older way of notating pedal is inexact and, as it stands, at times incorrect.
To strictly adhere to the pedalling as notated can lead to unsatisfactory results.
Here are some examples that illustrate this.
F. Chopin, Allegro de Concert op. 46, bars 182–186:

Ex. 5.9

To take the fifth and sixth eighth note in each bar without pedal, as Chopin has
notated, would result in a sudden, quite inappropriate reduction in sound. This can
hardly be what is intended, after the full roar of the four eighth notes in the low
bass register played ff with pedal. Of course, the pedal is simply retaken. From the
end of bar 183 to bar 184, the pedal—here used as a legato pedal—clearly cannot
to be changed until the first eighth note of bar 184. However, here, too, the star
and the pedal sign are written too far apart. In the case of such perfunctory pedal
notation, it is up to the tonal sense of the pianist to reach an appropriate decision.
F. Chopin, Rondeau op. 73, bars 103–108:

Ex. 5.10
64 Pedal

Here the pedal is retaken after each bass note: as the new bass note is struck,
the pedal is released and depressed immediately afterwards on the same note. This
application of the pedal immediately after the beat is referred to as syncopated pedal
(German: Nachtreten). In bar 107, referring to the pedalling, Chopin writes simile,
though this too can be misunderstood. He obviously means the pedal to be changed
as before—that is to say, with each new harmony, in this case now on each new
eighth note—except for the first two eighth notes—that belong to the same harmony.
J. Brahms, Andante from the Sonata in F-minor op. 5:

Ex. 5.11

The example above illustrates why a composer often only notates pedal spo-
radically. For the first three bars Brahms writes no pedal markings at all: it is
anyway clear that the pedal must be changed with every note of the melody, so
as not to obscure the progression of the cantilena. In the fourth bar, however, he
is concerned that the player may continue to change pedal and thus interrupt
the sustaining of the bass, so he indicates pedal in this bar while once more
foregoing the pedal indication in the next bar.
J. Brahms, Ballade op. 10 number 1, bars 27–37:

Ex. 5.12

The upbeat to the Allegro is marked con Ped. and on each first quarter of
the second and third bars just Ped. The next Ped. indication appears at the ff
in bar 37. On our modern, louder instrument, such a pedalling produces an
 Pedal 65

unpleasant effect. The pedal may already be changed on the downbeat after the
upbeat without a loss of tonal fullness. Depending on the hall’s acoustics, one
may at least take the third quarter with half-pedal so as not to completely drown
the melodic steps of a second and to prevent the D major chord in the higher
register of the piano from being disturbed by the loud E. The dotted half note
chord in the descant unfortunately subsides somewhat, but it will be invigorated
a little on the third quarter by the overtones of the chord in the lower register.
The term ‘half-pedal’ refers either to a partial depression of the pedal, or an
exceptionally quick pedal change whereby higher notes are dampened while
notes in the lower register continue to resonate. In this case, however, it is
the upper chord that we want to sustain, while changing the bass notes. Here
the pianist is left with no choice but to use a very quick half-pedal to produce
an impression that at least to some extent corresponds to the desired musical
effect. As so often happens, our musical imagination sustains a sound that in
reality subsides on the piano. Following the further course of events, our imagi-
nation seamlessly connects everything into a coherent whole, unimpaired by the
deficiencies of the instrument.
We can see from these examples how necessary it is to sensitively adjust the
use of pedal according to articulation, harmony and style.
Gabriel Fauré, Nocturne op. 37, bars 49–52:

Ex. 5.13

In this example, the pedal release sign is evidently incorrectly placed, although
Fauré’s musical intention is clearly set out in the notation itself. In each bar,
the first quarter note of the melody is accompanied with a single eighth note
harp-like arpeggiation. The melody note should be audible on its own after
the arpeggiation: this is clearly represented by the notation of an eighth note
chord against the quarter note melody note. Thus, the pedal must be lifted on
the second eighth and not the second quarter of the bar.
Gabriel Fauré, Nocturne op. 74, bars 39–42:

Ex. 5.14
66 Pedal

Here too, in the Allegro section, the pedal must be changed exactly on the first
eighth of the next bar so as to produce a continuous progression in the bass from
C-sharp through G-sharp, G to F-sharp that must be clearly audible. To interrupt
each bar with a rest would chop the phrase into pieces. This is, incidentally, a good
example of what is referred to as ‘technical notation’: the notation of the shortened
portato eighths becomes invalid if they are sustained for a whole note by the pedal.
Such instances where rests are notated but cannot be adhered to because of
the implied use of pedal are not infrequent. There are several such places in
Schumann’s F-sharp minor Sonata op. 11—one of which is to be found in the
Finale from bar 85. Although there is no pedal marking, here the bass notes must
be sustained by the pedal, sometimes for the whole bar. This renders the rests
inaudible. This topic will be dealt with in the chapter on notation. Compare
the following example, bars 94–101:

Ex. 5.15

In the case of Schubert’s Impromptu op. 142 D 935 no. 3, the bass notes are
marked with a staccato dot:

Ex. 5.16

This is only a technical notation and indicates the movement of the hand. (More
about this in the next chapter.) These bass notes should be sustained with the
pedal almost until the fourth of the eighth notes. If one already releases the pedal
on the second and fourth quarters respectively, a six-four chord would be heard—if
only for a short time—that sounds ugly without the root or third in the bass.
Such passages can be found also in Schubert’s songs. Many examples of the same
nature, where the pedal must be sustained in spite of staccato dots, are to be found
in the chapter on notation. Situations where such basses are perhaps to be played
shorter arise where the expression demands an intentionally comic effect.
 Pedal 67

Pedalling in Works of J. S. Bach

It is sometimes said that the pedal should not be used at all when performing
Bach. On closer consideration, however, this view is not tenable. Apart from
purely homophonic passages such as those in dances, there are even passages that
cannot be played polyphonically on the single manual of our modern piano (see
Examples 6.15 and 6.16, the fugue from the E minor Toccata, pages 79–80 in
the chapter on notation). As described at the beginning of this chapter, we have
five possible types of pedalling at our disposal. There is no reason why one
should not use the pedal without altering the articulation either for tonal
improvement or for the natural fading of a note, as Leopold Mozart describes.
(See endnote 2, Chapter 3, page 48.) Likewise, it must be permitted to connect
notes with the pedal that otherwise could not be played legato. On the other
hand, to use the pedal to allow notes of a single harmony to sound simultane-
ously contradicts polyphonic voice leading. This would only be permissible in
such instances as the first prelude in C major from the first book of the Well-
Tempered Clavier, where Bach no doubt kept his fingers down.
J. S. Bach, Prelude in C major from the Well-Tempered Clavier I:

Ex. 5.17

There is, of course, no call for the so-called compromise pedal with Bach.
In any case, it is necessary to use pedal with great care in music of a poly-
phonic structure. In homophonic dances (contrapunctus simplex) the situation is
different, as in the Menuet from the Partita in B-flat major (see Example 6.14,
page 78 in the chapter on notation). Here one can take the pedal on each bass
note, playing a longer or shorter portato in accordance with the points of emphasis
in the bar—similar to a pizzicato accompaniment on the cello.
An example of where one must avoid too short a note within passage-work
can be found in the second bar of the Prelude in D minor from the first book
of the Well-Tempered Clavier.
J. S. Bach, Prelude in D minor, bar 2:

Ex. 5.18
68 Pedal

Here there is a leap from D to A. It is not possible to hold the D, and one must
leap up to the A at lightning speed, which results in a staccatissimo on the D. In
order to play the sixteenth triplets evenly, and to prevent the obtrusive hole
caused by the staccatissimo from interrupting the even flow of the triplets, one
can maintain musical fluency by briefly applying the pedal from D to A—but
no longer than that! The leap is not so intrusive on the harpsichord, though on
the piano the heavy dampers cause an unwelcome interruption in the even runs.
Another solution to this problem would be to play the D with the left hand,
which is however not so comfortable to execute. One is loath to divide a voice
between two hands unless it is unavoidable. Similar leaps are to be found in
many other works, as here in the C minor Prelude of the first volume, bar 30:

Ex. 5.19

In a cantilena, a richer tone can be achieved through the pedal, which allows
sympathetic vibration of other strings. This improvement in sound quality is advan-
tageous as long as the acoustic is not overly resonant (in some halls, a note can
reverberate for two to three seconds). Such tonal improvement can also be applied
to works of Bach. The harpsichord had a beautiful, very bright sound, rich in
higher frequencies. In comparison, the piano sounds more muffled, particularly in
its lower register. Here the change in tone colour achieved by the pedal can assist.

Left and Middle Pedals

The left pedal (una corda or soft pedal) usually produces a fainter, more muffled
sound. It is a bad habit to use it at every quiet passage, because piano is usually
written at cantabile, lyrical phrases that should not sound muffled or feeble, but
that on the contrary should sound particularly silvery and brilliant in tone.
One rather uses this pedal to obtain an alternative timbre, a different colour, so
it may even be applied in some forte passages. Unfortunately, the change in tone
produced by the pedal is not the same on all pianos. On some grand pianos it
tends to produce an unsatisfactory, somewhat nasal quality.
With the middle pedal (sostenuto pedal), with which better equipped pianos
are fitted, any note or chord can be sustained independently, while all other notes
remain unaffected by it. Thus, one can change the right pedal while the note/
chord continues to sound as long as the middle pedal is depressed.
We find an example of such pedalling in Rachmaninov’s Prelude op. 3 No. 2
in C-sharp minor:
 Pedal 69

Ex. 5.20

While the octaves on C-sharp in both hands in the second bar are sustained
with the middle pedal, the chords that follow may be changed with the right
pedal so as to tonally differentiate the melodic and harmonic progression. On
the third and fourth quarters of the third bar, without the middle pedal, the
octaves A and G-sharp are only sustained with the right pedal while the fol-
lowing chords are played. In the fourth bar, the C-sharp octaves must again be
held with the middle pedal to cleanly separate the following chords harmonically.
The ensuing bars are executed in a similar fashion.
If there is no middle pedal, one can achieve a similar effect by rapidly chang-
ing the pedal without releasing it entirely.
Here is another example, from the third movement of Schumann’s Sonata in
G minor op. 22, bars 41–53:

Ex. 5.21
70 Pedal

In bars 41, 45 and 49, the orchestral effect that Schumann often imagined in
his piano works can easily be achieved by using sostenuto pedal. One need only
think of his Symphonische Etüden op. 13. We shouldn’t forget his great enthusi-
asm for the pedalflügel, for which he composed several works. In his own words:

Quite frankly I put great faith in the idea that these instruments could
bring a new élan into piano music. Wonderful effects can be achieved
with this instrument.4

And in the Neue Zeitschrift für Musik he writes:

and should it come so far—as I believe it will—that the pianoforte can

be used with a pedal as on the organ, this will open up new vistas for
composers and liberate the instrument from the support of the orchestra,
giving it a greater richness of tone and independence of movement.5

In the above example, the passage can be effectively realised in various ways.
The sostenuto pedal can be applied on the first eighth of the bar and sustained
for two full bars while the right pedal is changed in accordance with harmonic
shifts. In the third and fourth bar, the right pedal is changed on every note
without using the sostenuto pedal. This pedalling is repeated in bars 45–48 and
49–52. Another solution that would emphasise the energetic octave steps from
G major to C minor would be to release the sostenuto pedal on the second
quarter of the second bar of each four-bar phrase. However, the passage can still
be effectively realised even without a sostenuto pedal. The right pedal is fully
sustained in each phrase until the second quarter note of the second bar, unless
an overly resonant acoustic mixes the harmonies so much that the expressively
emphatic second steps into the ninth chord can no longer be heard. In such a
case one should use half-pedal or skilfully change pedal very quickly in order
to sustain the bass.

Notes
1. Josef Dichler, Der Weg zum künstlerischen Klavierspielen, Doblinger Verlag, Vienna, 1948.
2. “Playing with an almost constant lifting of the dampers as a way of covering up an
impure sort of playing, in which notes are swallowed, has become so fashionable that
one often would not recognise a player if one were to hear him play without the
pedal.” Johann Nepomuk Hummel, Ausführliche Anweisung zum Pianofortespiel, Tobias
Haslinger, Vienna, 1828, Dritter Theill, zweiter Abschnitt, drittes Kapitel Vom Gebrauch
der Pedale, §1.
3. “Beethoven, who played this concerto in 1803, left the pedal down throughout the
whole theme, which worked very well on the weak-sounding pianos of the time,
especially when the una corda pedal had been added. Today, however, now that piano
sound has become much stronger, we would recommend re-taking the sustaining pedal
 Pedal 71

with each new significant change of harmony. This must be done without any notice-
able gap in the sound, as the whole theme must sound like a distant, sacred celestial
harmony.” Carl Czerny, Über den richtigen Vortrag der sämtlichen Beethoven´schen Klavierwerke,
chapter 3, § 27, Diabelli u. Comp., Vienna, 1842, ed. with comments by Paul Badura-
Skoda, Universal Edition Vienna, 1963.
4. Letter to the publisher Whistling, 6 May 1845.
5. Robert Schumann, Neue Zeitschrift für Musik, Leipzig, 4 January 1839, p. 1 f.
6
NOTATION

In the preceding chapters, I have attempted to demonstrate the limitations of

our system of musical notation. The following remarks are intended to point
out some aspects of the score that composers themselves omit for simplicity’s
sake. Such omissions were often just a matter of saving time in the writing
process or were the result of the writing utensils employed. Ignorance of these
simplified notational conventions can lead to a performance that misinterprets
the composer’s intentions.
One such simplification is a so-called technical notation. This differs from a
musical notation in that it omits certain details in the score and relies on the
player’s prior knowledge of harmony and melody for a correct rendition. This
form of notation was given this name because it simply notates the movement
of the hands. Musical notation specifies the exact duration of all notes (with the
reservations explained in the chapter on articulation), while a technical notation
indicates simply the actions that the pianist’s hands should execute.

‘Technical Notation’—Shortened Bass Notes

A note or chord in the bass is often notated with a short value because it must
be swiftly released in order to reach a chord or figuration in a higher register.
For musical reasons, however, both the bass and the notes above it must be
allowed to sound together by applying the pedal.
We find such technical notation in the last bar of the Finale from Chopin’s
Sonata in B-flat minor. The octave in the bass is written as a short note, followed
by a rest before the final chord. This is a six-four chord that cannot serve as a
closing harmony: naturally the bass note is to be sustained by the pedal, even
if Chopin had not written a pedal marking. Bass and subsequent chord should,
after all, sound together—they just cannot be played together.
 Notation 73

Ex. 6.1

Here is an example from the second movement of Schubert’s Sonata in B-flat

major D 960. The composer notates a staccato eighth note, a rest, a thirty-second,
then again a staccato eighth and at the same time demands the use of the pedal,
which results in all the notes being sustained almost until the end of the bar.

Ex. 6.2

41

Or Brahms’ Capriccio op. 76 no. 5, from bar 113 to the end:

Ex. 6.3

42

The final bars form a big crescendo in diverging octaves and chords to a fortissimo
in the penultimate bar. Playing the chords in the last two bars as written as
staccato eighths would make an unduly comic ending out of this very effective
conclusion to the piece. Of course, in the penultimate bar we pair the two eighth
notes of each harmony with the pedal, separating the tonic six-four on the first
half of the bar from the dominant on the second half with a brief, vigorous
74 Notation

portato, in keeping with the energetic character of the passage. The final chord
in the last bar, divided between a low bass note in octaves and the eight-note
chord in both hands above, is of course joined with the pedal.
The staccato dots in the bass in bars 71 and 72 of the same piece, in the agitato
after the double bar, are equally just ‘technical notation.’ The bass notes are again
connected to their subsequent chords with the pedal.

Ex. 6.4

Here are some more extracts from pieces by Brahms: firstly, the final bars of the
Capriccio op. 116 no. 7. The two staccato eighth notes are again here sustained
together with the previous bar until the end, or still better, one can change the
pedal on the last chord, so that this final chord closes with the root note D in
the soprano, rather than the third (F-sharp):

Ex. 6.5

Again, to make it perfectly clear: such staccato eighth notes do not indicate any
form of articulation. They are not to be played short. On the contrary, they are
to be sustained with the pedal. Composers had got into the habit of dispensing
with complicated, musically precise notation and notated just the movement of
the hands. There are countless such examples in the literature. Very often we
find broken chords as accompaniment whose bass notes must of course be held
with the pedal, as in the middle section of Brahms’ Intermezzo op. 118 no. 2,
where these low notes should be sustained—to a greater or lesser degree as fits
the harmonic progression. Here is a further example: the final three bars of
Brahms’ Rhapsody op. 119 no. 4.
 Notation 75

Ex. 6.6

Why does Brahms go to the trouble—here, as in the other places—of writing

eighth notes with eighth rests, and on top of that of marking them with staccato
dots, while it would have been much simpler just to write quarter notes, which
would have still been a ‘technical notation’? By writing short eighth notes in forte
and fortissimo respectively, Brahms surely wanted to imply an energetic, powerful
swinging action that can only be made with a very short attack as expressed by
his notation. It is not possible to retake the pedal for the final chord, as in this
period a movement cannot end on a sixth-three chord (first inversion).
In a continuous eighth note bass figuration, the initial note is frequently marked
staccato while subsequent notes are not. This is the case in the passage from Schubert’s
Impromptu in B-flat major quoted in Example 5.16 on page 66, where the bass
notes are clearly to be sustained with the pedal. The same situation occurs in the
following passage from Schubert’s Sonata in B-flat major D 960 (bars 27–30):

Ex. 6.7

In such cases, sometimes a pedal sign indicates that the staccato bass note must be
sustained. However, even when a pedal mark is missing, the bass is to be held. This
notation, which mainly established itself in music of the Romantic era, simply indicates
the movement of the left hand. The pianist leaps from the bass note so as to be able
to play the following notes or chords rhythmically in time. However, it would be
musically perverse actually to play these notes staccato, unless they occur in passages
with a coquettish, lightly hopping, dance-like character. The notation intends only
to show what is to be done technically—hence the term ‘technical notation.’
76 Notation

An instance of a comparatively rare form of ‘musical’ notation is to be found

in Brahms’ Capriccio op. 116 no. 3 (Example 6.8). In the two closing bars, we
suddenly see the notes extended by ties. Whereas Brahms writes the previous
bass notes C and C-sharp with ‘technical notation’—although they are sustained
in just the same way as the G at the end of the Capriccio—he only takes the
trouble to indicate this with a musical notation in the last two bars.
The reason is evident from the score: with the first two motives the left hand
must jump from the bass to the middle voices to complete the chord. However,
from the third bar of the example, it can play the three bass notes D–B-flat–A
in a sustained legato. Musically speaking, in the first two bars we have two chords
separated with a portato, each with its incisive bass, slightly detached with the
pedal, whereas in the last bars the bass notes form a legato phrase. For the sake
of a stronger cadential effect, only the last bass note G is slightly detached. As a
phrase ending—here the close of the whole piece—these legato half notes must
end with a longer note, in this case two whole notes joined with a tie. Even
though the effect would be the same, how unsatisfactory the score would look
if the final note in the bass was written in ‘technical notation’ as a quarter! Such
a musically absurd notation would not have occurred to Brahms.

Ex. 6.8

Here is a further instance of ‘musical’ notation. In the recapitulation of first

movement of Beethoven’s Sonata op. 2 no. 3, in the eighth note broken chords
from bar 222 onwards, we see just how complicated such a notation can be. In
the following triplet broken chords, Beethoven dispenses with a ‘musical nota-
tion.’ Here too, however, all the triplets are naturally meant to be sustained. A
‘musical’ notation would here have become much more complicated.

Ex. 6.9
 Notation 77

The previous Examples 3.3, 3.4 and 3.5 in the chapter on articulation (see
pages 18–19) show how vital it is to question the notation and not to rely on
the way music appears on the page. In the Finale of Schumann’s Sonata in F-sharp
minor op. 11, we find rests in various places that cannot be observed because
the bass note, here written only in ‘technical notation,’ must be sustained. Here
only musical intuition and a sense of tone can be of help in finding the right
manner of execution. Here are bars 98–101:

Ex. 6.10
48

Here every third sixteenth is a rest, but at the same time one should apply the
pedal, because the bass notes must be sustained—therefore the second sixteenth
sounds de facto as an eighth note. So why did Schumann not notate this ‘cor-
rectly’? He could have written the second sixteenth as an eighth, which would
even have been simpler. Through his manner of notation, the figure is recog-
nisable as a ‘sighing motive.’ The shorter notation of the second sixteenth forces
the player to lift the note, to play it shorter, since it anyway falls on a weak
part of the beat. This notation helps one to imagine a very delicate, melancholy
motive, whereas a notated eighth note would gain an inappropriate amount of
weight and suggest a ponderous ‘dragging the feet,’ as it were. The sustaining
of the pedal is not intrusive and avoids a dry rendering of the passage. Musical
intuition and a sense of tone must help us arrive at the proper execution of
similar passages.
In closing, another brief word about rests at the end of phrases: for simplic-
ity’s sake, composers often write an eighth note followed by an eighth rest
(or quarter note followed by quarter rest) where the final note needs to be
slightly prolonged and allowed to reverberate into the rest. To strictly cut off
the note at the start of the rest does not lead to a musical result. As we have
mentioned, our notation has no symbol for the slight lengthening of a note, only
the dot—which is just used in connection with a following note in a dotted
rhythm, not as a means of lengthening a single note. (The dot lengthens the
note by half its length, or, if adjusted to a triplet rhythm, by one triplet, as the
Example 6.30 on page 85, or in some cases a little longer, as in Example 6.31
on page 81. More on this subject in the section on dotted notes.)
 78 Notation

Hidden Part-Writing
Another simplified form of notation often occurs when two-part writing is
notated in one voice. Compare the following cadences (Examples 6.11 and 6.12):

Ex. 6.11

After the four-note arpeggiation of the harmony, there follow descending and
ascending leaps of a seventh, a descending leap of a fifth and an ascending leap
of an augmented fourth. These would not be beautiful, either musically or
vocally, were one not to recognise the implied two-part writing that, if musically
notated, would result in two separate voices, looking like this:

Ex. 6.12

Or the Prelude of the Partita in B-flat major BWV 825 by J. S. Bach:

Ex. 6.13

43

Here the notes C, D and E-flat must be sustained as an actual upper voice,
beginning on the second quarter with portato or legato notes of equal value against
the notes E-flat, F and G in the lower voice. The constantly repeated B-flat
between these notes is only an accompaniment completing the harmony and
should be played quietly so as not to interrupt the flow of the main ascending
notes that go to make up the melodic phrase.
Such a hidden upper voice is to be found in the Menuet I of the same Partita:

Ex. 6.14

44
 Notation 79

Here, too, it is very attractive to sustain the initial D and, in the second bar, to
play the D, E-flat and F together with the bass in notes of equal length with a
graceful, somewhat shorter portato as befits the style of the minuet.
The fugue from J. S. Bach’s E minor Toccata BWV 914 offers a further
instance of hidden part-writing:

Ex. 6.15

At the opening of this fugue, the first eight notes harmonically form a cadence
that serves to consolidate the position from which the bass line descends chro-
matically to the B in bar 4, leading to a repeat of the theme as the comes on the
dominant, and returning to the tonic in bar 9. From bar 5–8, this comes is
underlaid with a rhythmically articulated counterpoint as a bass. Following the
model of this new bass line, we can shape the opening theme (dux) in the same
way. The rhythm of the motive D/D/C-sharp in the counterpoint from the
second half of bar 5 to the first half of bar 6 finds its equivalent at the begin-
ning of the fugue in the notes E/E/D-sharp (beginning on the third beat of
the first bar). We will play these thematically more important notes louder than
the rest of the sixteenths. In the first half of the second and third bars, a leap
of a seventh is repeated four times. Here, too, the thematically important first
note is played loudly and sustained. Of course, because of the repeated notes,
this can only be done with the pedal. The redundant repetition of the seventh
interval, which serves to maintain the rhythmic energy of the toccata, must be
played quietly, to avoid it sounding like babbling. The following example shows
an exact notation of the method of execution outlined above (from bar 5 the
theme is of course played in the same way):
 80 Notation

Ex. 6.16

45

One might play the first notes of bars 2 and 3—D-sharp and C-sharp respectively—
as quarter notes (see Example 6.17), analogous to the equivalent notes in the left
hand in bars 6 and 7, but the uniform repetition of the sevenths is better masked
by sustaining the pedal for a half note as shown in the last example.

Ex. 6.17

One must understand that at the time it was customary to notate polyphony in
a simplified way, as the following example from the time of Bach shows. Cou-
perin notated an accompanying figure in the Rondeau, Les Bergeries, from his
Second livre de Pièces de Clavecin in the following manner:

Ex. 6.18

Bach included this piece in his Notebook for Anna Magdalena Bach and reproduces
the passage in this fashion:

Ex. 6.19

Bach, the master of polyphony, clearly did not intend to reduce two-part writing
to a single voice. He could count upon the general musical knowledge of the
 Notation 81

player to recognise the hidden counterpoint. Such notational simplifications were

common practice at that time.
Thus, Beethoven, in the following example of a study by J. B. Cramer1
that he used for teaching his nephew, expressly stipulates that the first of each
sixteenth note group be held until the fourth sixteenth note.

Ex. 6.20

In another Cramer study (Example. 6.21), Beethoven adds the comment: “The
melody lies almost continually on the third note of each group; the rhythmic
accent, however, is to be placed regularly on the first note. In order to join the
notes, the finger remains on these accented notes.”2

Ex. 6.21

One can imagine the number of places in his sonatas where is it musically
necessary to sustain notes in the same way. A single example must serve to
illustrate the point (Example 6.22): here, in the Sonata op. 10 no. 3, from bar
31 the eighth notes on the first and second beats (first and fifth eighth note)
form a hidden upper voice and must therefore be sustained, though not for a
full half note, but rather as a slightly longer quarter note, in accordance with
the cheerful character of the piece. On the other hand, the four eighth notes at
the end of the second, fourth and sixth bars are played with their normal value
as an upbeat. If one were to ignore this hidden principal voice, all that would
remain would be an evenly running succession of eighth notes like a study in
dexterity—though even in studies Beethoven demands that ‘accented’ notes (that
is, notes on strong beats) be sustained.
 82 Notation

Ex. 6.22

46

The sixteenth note passages in the first movement of Beethoven’s Sonata in

F-sharp major op. 78 likewise conceal a cheerful melody:

Ex. 6.23

From the upbeat to bar 19, the following melody results from by sustaining the
melodic notes with the fingers, or through use of the pedal.

Ex. 6.24

Similar passages are not only to be found in works of Beethoven. Here are a
few bars from W. A. Mozart’s Sonata in B-flat major KV 333 (first movement,
bars 18–20):

Ex. 6.25

The left hand plays the lower voice while the upper voice, moving parallel with
it, is carried by a sustaining of the first note of each sixteenth note group in
the right hand. We see the same picture in the following sonata of Haydn.
 Notation 83

J. Haydn, Sonata No. 53 Hob. XVI/34 first movement, bars 19–23:

Ex. 6.26

In bars 19–21, within the sixteenths of the right hand, the upper voice coincides
exactly with the notes of the lower voice in the left hand. In bars 22–23, the
situation is then reversed.
In the following example from a Haydn sonata, the execution is somewhat
more awkward.
J. Haydn, Sonata in D major No. 39 Hob. XVI/24, first movement, bars 40–45:

Ex. 6.27

Here and in many other passages in the sonata, the first voice lies in the sixteenth
notes. Here, however, the melodic note always follows an eighth after the second
voice—on the unstressed part of the beat. To make the melodic notes of the
upper voice more clearly audible without resorting to an offbeat rhythmic accent,
one leaves them quite unaccentuated while sustaining them as eighths.

Dotted Notes
Dotted notes are another instance of incomplete notation. Mathematically speak-
ing, the two notes stand in a ratio of three to one. However, in practice dotting
can signify ratios from 2:1 to 7:1. Double-dotting—that is, the extra dot after
the dot, or a dot after a rest—was still unknown in the Baroque era. It had yet
to be invented. Dotted rhythms can sometimes employ three or even four dots,
whereby each dot represents half the value of the previous dot. Triple dotting
occurs in works of Anton Bruckner and Richard Wagner, but also in piano
music, for instance in Frédéric Chopin’s Prélude op. 28 no. 3 in G major.
 84 Notation

We know from the literature that in a festive rhythm, such as that of the French
overture, over-dotting was customary. In the following example, this applies both
to the dotted eighth notes as well as the sixteenths and the sixteenth note rests, so
that all the notes after the longer notes of the dotted rhythm are equally short.
J. S. Bach, 2nd Partita in C minor BWV 826, Sinfonia:

Ex. 6.28

47

A slight over-dotting that cannot be notated is often to be found in rhythms of

a dance-like character.
In closing, as an example of the varied significance of dotted notes, a word
about the adjusting of dotted rhythms to triplets. If eighth note triplets are writ-
ten in one voice and dotted notes in another—that is, a dotted eighth note with
subsequent sixteenth note—then very often, but by no means always, this last
sixteenth is played with the third triplet. In the case of the dotted quarter, the
sixteenth coincides with the sixth triplet (Example 6.29). This form of notation
was customary in the Baroque era. Schubert almost always followed it. Editors
of urtext editions unfortunately often take it upon themselves to ‘correct’ the
alignment with mathematical precision. Such passages can be found also in the
works of Mozart and Beethoven.
Here is an example from a work by Bach, the Corrente from his B-flat
major Partita BWV 825:

Ex. 6.29

Here, too, the sixteenths in the left hand are played together with the third triplet
eighth note.
 Notation 85

Or in the final episode of Chopin’s Polonaise-Fantaisie op. 61, from bar 254:

Ex. 6.30

In all these instances, the sixteenth note should be played together with the third
triplet—as here at the opening of the first of Schubert’s Drei Klavierstücke D 946:

Ex. 6.31

The second movement of Schubert’s Sonata in E-flat major op. posth. 122 D 568
(bars 43–46) offers another interesting example:

Ex. 6.32

In the first bar, the thirty-seconds in the left hand are adjusted to the sextuplet
sixteenths in the right hand. In the second and third bars, there is a dotted
eighth that is held a little longer, so that the following sixteenth coincides with
the sixth sextuplet in the right hand. In this way, the shorter notes of the dotted
rhythm—both the sixteenths and the thirty-seconds—are exactly the same length
as the individual sextuplets.
In these bars and all other similar passages, the Henle urtext edition aligns the
dotted and triplet rhythms with mathematical correctness, although Schubert notates
86 Notation

musically: the sixteenths and thirty-second notes in the left hand are exactly aligned
with their respective triplet sixteenth notes in the right hand. Consistency would
demand that the first quarter of each bar be notated in the left hand as a double-dotted
eighth with subsequent thirty-second. Schubert here further simplifies the notation.
Here is a further example by Mozart, from his concerto in B-flat major, KV
450, first movement, bars 76–77:

Ex. 6.33

Schubert’s Fantasie for four hands op. 103 D 940 offers an example of a passage
where dotted rhythms should rather not be adjusted to the accompanying triplets.
The opening theme is in a dotted rhythm:

Ex. 6.34

A contrasting forte section begins in bar 48:

Ex. 6.35

The section opens with an extremely vigorous, rhythmic theme accentuated

with sforzati that closes in the fifth bar. One would like to maintain the dotted
 Notation 87

rhythm in the left hand. However, the triplets in the right hand make this sound
rather clumsy, unless one ‘over-dots’ the rhythm. Now there are no more sforzati
markings, which could be an indication that one should adjust the dotted rhythm
to the triplets after all—a milder rhythm as a variant would be perfectly accept-
able here. However, one should not be tempted into playing the opening theme
in this rhythm: to adjust the opening theme to a triplet rhythm would almost
make it sound like a waltz and would hardly suit its sad character—or the rhythm
of the contrasting theme in bar 48, which, with the repetition of the first theme’s
final two notes, emerges seamlessly from it.
Here is another example—from Chopin’s tenth Nocturne op. 32 no. 2—where
adjustment of a dotted rhythm to triplets would be out of place:

Ex. 6.36

In bar 8 on the second quarter note beat, there are triplets in the left hand,
while the right hand plays a dotted eighth note followed by a sixteenth. This
notation occurs frequently throughout the Nocturne, and in view of the slow
tempo, one might be inclined to choose the more peaceful triplet adjustment
rather than a true dotted rhythm. However, the appearance of a similar dotted
rhythm motive as a triplet in the following bar—demanding an even shorter
execution of the note after the dot—would seem to contradict such an execu-
tion. In addition, from bars 35 to 38, where he does want an alignment of the
two rhythms, Chopin adopts the modern notation of a triplet quarter followed
by triplet eighth. This suggests that, should he have wished the dotted rhythm
to be adjusted to triplets in bar 8 and similar passages, he could have used the
same modern notation. Bars 35–36:

Ex. 6.37

However, one cannot assume that a new notational convention is assumed from one
day to the next. It can well happen that a composer falls back into old habits.

***
88 Notation

Our notation can by no means claim absolute validity. It is not a set of instruc-
tions for the performance of a piece of music that is to be taken literally, but
rather a highly insubstantial and scarcely sufficient indication of how a composer
imagines the sound.
It remains the task of the interpreter, within the limits of musical style and taste,
to shape a work with imaginative musical fantasy in the spirit of the composer
in as vivid a manner as possible. In this way, in the hands of a gifted perform-
ing musician, a work will come into being of its own accord. Devoted to the
intuition of the composer and inspired by the sound, without intentional exercise
of will, in his hands the work will take shape out of the spirit of the music.

Notes
1. Johann B. Cramer, 21 Etüden für Klavier nebst Fingerübungen von Beethoven nach dem
Handexemplar Ludwig van Beethovens, Universal Edition, Vienna, 2013, ed. Hans Kann
(o. J.).
2. “Der Bindung wegen bleibt der Finger auf dieser akzentuierten Note liegen.” ibid.
7
CONCLUSION

In closing, I would like to recall some remarks with which this study begun, namely
those comments concerning the human need for expression in word and sound,
speech and music and, by extension, written language and musical notation. Just as
different countries in Europe have given rise to different languages, so they have
produced different musical styles, though the fundamental principles of this music
are the same. In this sense we can truly speak of a common musical language with
different regional dialects. The analogy with spoken language can be taken further:
even in a country with the same language we find subtle differences that have arisen
out of the mentality of the populations of various regions: differences in articulation,
in harder or softer consonants or slightly altered vowel-sounds, in short, a colouring
of the language with varied dialects that betray the origin of the speaker. The same
holds true for music, although the principles of the language may be the same. Slavic
countries, for example, tend towards a heavier, melancholy tone—a quality also
reflected in their folk music of a predominantly minor mode. The esprit and élégance
of the French surface in the works of Ravel and Debussy, the southern temperament
ignites the imagination of de Falla, Albéniz and Granados. A tendency to poetry
and expressive fire is reflected in the works of Schumann, and the sad melancholy
note sounds nowhere more plaintive than in the music of Schubert, while Mozart,
for all his dramatic artistry, shines in pure Apollonian beauty. And we shouldn’t
forget the highly personal characteristics of a Chopin or Liszt. Nevertheless, it is a
single language that can be understood by all who choose to learn it.
And how might we recognise a ‘Viennese tradition’ reflected in those com-
posers who were rooted here? It is a matter of certain characteristics, musical
idiosyncrasies, a colouration in dialect that was often to be found in the nature
of the people here or in the melodies of their folk music. Expressive, deeply
90 Conclusion

felt motives in a lower voice that go unnoticed under the apparent lead of the
melodic upper voices, or perhaps the subtle abandonment of a swinging move-
ment (as in the third movement of Schubert’s B major Sonata), aspects too often
overlooked and misunderstood: how could one ever write all this down? How
could a pianist or other musician recognise such things without a knowledge
of this tradition?
Just as the noble manner of the French, Vienna too gave rise to an equally
noble manner that must be learnt. One will not find it anywhere in writing.
And, nonetheless, even the North German Brahms ingeniously brought some
intangible aspect of the Viennese essence of his much-loved Johann Strauß to
life in his works.
Rather than succumbing to a self-indulgent capriciousness, we should
try with great sensitivity to engage with this music in all its imaginative
diversity—a music which expresses the distinctiveness of so many different
ways of life.

***

I would like to conclude with the words of Joachim Quantz, Franz Schubert
and Pablo Casals:

A speaker and a musician share one and the same intention, namely to
conquer the heart, to arouse or to calm the passions, one moment to arouse
a particular emotion in the listener, the next moment to arouse another
[. . .] to avoid uniformity, rather to let the sounds be heard now louder,
now quieter, now faster, now slower.
(J. J. Quantz)1

Some have assured me that the keys beneath my fingers become singing
voices, which, if it is true, pleases me greatly, for I cannot abide that accursed
hacking that is a quality even of most excellent pianists, for it delights
neither ear nor mind.
(Franz Schubert)2

Nature never stays at one level, there is a constant vibration. [. . .] Every

note has to have a different sonority, it is a song. [. . .] Variety is a great
word—in music as in everything; variety is a law of nature. Good music
has never monotony. If it is monotonous it is our own fault if we don’t
play it as it has to be played. We must give to a melody its natural life.
When the simple things and natural rules that are forgotten are put in the
music—then the music comes out!
(Pablo Casals)3
 Conclusion 91

Notes
1. Johann Joachim Quantz, Versuch einer Anweisung die Flöte traversiere zu spielen, Johann
Friedrich Voss, Berlin, 1752, pp. 100–101.
2. From a letter to his father and step mother on the 25th (28th?) of July 1825, Die
Dokumente seines Lebens, collected and commented by Otto Erich Deutsch, Bärenreiter,
Kassel etc., 1964, p. 299.
3. Quoted in: David Blum, Casals and the Art of Interpretation, University of California
Press, Berkeley, 1980, pp. 9 and 18.
INDEX OF QUOTED COMPOSERS
AND WORKS

Bach, Johann Sebastian

Das Wohltemperierte Klavier I, Prelude in C major, Example 5.17
Das Wohltemperierte Klavier I, Prelude in C minor, bar 30, Example 5.19
Das Wohltemperierte Klavier I, Prelude in D minor, bar 2, Example 5.18
Das Wohltemperierte Klavier I, Prelude in E-flat minor, Example 4.3/
Das Wohltemperierte Klavier II, Prelude in E major, Example 3.9/
Das Wohltemperierte Klavier II, Prelude in F minor, bars 20–24, Example
2.8/
Das Wohltemperierte Klavier II, Prelude in F minor, Example 4.5/
Prelude from Partita in B-flat major BWV 825, Example 6.13/
Corrente from Partita in B-flat major BWV 825, Example 6.29
Sarabande from Partita in B-flat major BWV 825, Example 4.2/
Menuet I from Partita in B-flat major BWV 825, Example 6.14/
Sinfonia from Partita no. 2 in C minor BWV 826, Example 6.28/
Fugue from Toccata in E minor BWV 914, Example 6.15
Toccata in E minor BWV 914, Adagio, Example 4.1/
Fugue from the Toccata in E minor BWV 914, exact notation, Example
6.16/
Fugue from Toccata in E minor BWV 914, inferior execution, Example
6.17
Simplified notation of Couperin’s Rondeau Les Bergeries, Example 6.19

Beethoven, Ludwig van

Sonata in A major op. 2 no. 2, first movement, Example 3.32/
Sonata in A major op. 2 no. 2, second movement, Example 3.13/
Sonata in C major op. 2 no. 3, first movement, from bar 222, Example 6.9
 Index of Quoted Composers and Works 93

Sonata in C major op. 2 no. 3, second movement, Example 3.8/

Sonata in E-flat major op. 7, second movement, bars 20–21, Example 3.18
Sonata in C minor op. 10 no. 1, second movement, Example 2.15/
Sonata in F major op. 10 no. 2, first movement, Example 3.33
Sonata in D major op. 10 no. 3, first movement from bar 31, Example 6.22/
Sonata in C minor op. 13, first movement, bars 51–79, Example 3.35
Sonata in E major op. 14 no. 1, first movement, bars 5–6, Example 3.36
Sonata in G major op. 14 no. 2, first movement, Example 3.37
Sonata in A-flat major op. 26, first movement Var. IV, Example 3.12/
Sonata in E-flat major op. 27 no.1, first movement, Example 3.22
Sonata in E-flat major op. 27 no.1, first movement, bars 37–44 Example
3.38
Sonata in E-flat major op. 27 no. 1, fourth movement, Example 3.23
Sonata in C-sharp minor op. 27 no. 2, third movement, Example 3.20
Sonata in D minor op. 31 no. 2, first movement, bars 159–169, Example 3.19
Sonata in E-flat major op. 31 no. 3, third movement, Trio, Example 3.2/
Sonata in C major op. 53, first movement, Example 3.34
Sonata in F minor op. 57, first movement, bars 35–39, Example 2.6/
Sonata in F-sharp major op. 78, first movement from bar 19 with upbeat,
Example 6.23
Sonata in F-sharp major op. 78, first movement from bar 19 with upbeat,
cantilena, Example 6.24
Sonata in E-flat major op. 81a, second movement, bars 13–14, Example 3.25
Sonata in E-flat major op. 81a, second movement, bars 29–30, Example 3.26
Sonata in A major op. 101, first movement, Example 4.8/
Sonata in E major op. 109, second movement, bars 9–25, Example 3.24
Sonata in A-flat major op. 110, first movement, from bar 12, Example 3.52
Sonata in A-flat major op. 110, third movement, bars 9–12, Example 3.50
Sonata in A-flat major op. 110, third movement, bars 116–119, Example 3.51
Piano Concerto no. 3 in C minor op. 37, second movement, Example 5.3

Brahms, Johannes
Sonata in F minor op. 5, second movement, Example 5.11
Ballade op. 10 no. 1, bars 27–37, Example 5.12
Capriccio op. 76 no. 5, bars 71–72, Example 6.4
Capriccio op. 76 no. 5, bar 113 to the end, Example 6.3/
Capriccio op. 116 no. 3, final bars, Example 6.8
Intermezzo op. 116 no. 4, bars 15–19, Example 5.5
Intermezzo op. 116 no. 4, bars 36–52, Example 5.7
Intermezzo op. 116 no. 4, last five bars, Example 5.8
Capriccio op. 116 no. 7, final bars, Example 6.5
Intermezzo from Klavierstücken op. 119 no. 1, bars 45–47, Example 5.6
Rhapsodie op. 119 no. 4, final bars, Example 6.6
94 Index of Quoted Composers and Works

Chopin, Frédéric
Nocturne in D-flat major op. 27 no. 2, Example 5.4
Scherzo in B-flat minor op. 31, Example 3.45
Nocturne op. 32 no. 2, Example 6.36
Nocturne op. 32 no. 2, bars 35–36, Example 6.37
Sonata no. 2 in B-flat minor op. 35, fourth movement, final bar, Example 6.1
Allegro de Concert op. 46, bars 182–186, Example 5.9
Polonaise-Fantaisie op. 61, from bar 254, Example 6.30
Rondeau op. 73, bars 103–108, Example 5.10

Couperin, François
Rondeau Les Bergeries, Deuxième livre de Pièces de Clavecin, Example 6.18
Johann Sebastian Bach’s simplified notation of Couperin’s Rondeau Les
Bergeries, Example 6.19

Cramer, Johann Baptist

Etude in C major, Example 6.20
Etude in D major, Example 6.21

Fauré, Gabriel
Nocturne op. 37, bars 49–52, Example 5.13
Nocturne op. 74, bars 39–42, Example 5.14

Haydn, Joseph
Sonata in D major Hob. XVI/19, third movement, Example 4.7/
Sonata in C minor Hob. XVI/20, first movement, bars 32–35, Example
3.29/
Sonata in C minor Hob. XVI/20, first movement, bars 47–51, Example
3.30/
Sonata in D major no. 39 Hob. XVI/24, first movement, bars 40–45, Example
6.27
Sonata in E minor Hob. XVI/34, first movement, Example 3.42/
Sonata in E minor Hob. XVI/34, first movement, bars 19–23, Example
6.26
Sonata in E minor Hob. XVI/34, second movement, Example 3.7/
Sonata in E minor Hob. XVI/34, second movement, Example 2.14/
Sonata in G major Hob. XVI/40, first movement, bars 61–62, Example 3.47
Sonata in C major Hob. XVI/48, second movement, Example 4.6/
Sonata in E-flat major Hob. XVI/49, second movement, Example 2.16/
Sonata in E-flat major Hob. XVI/52, first movement, bars 27–29, Example 3.43

Liszt, Franz
Tarantella, bars 73–75, Example 3.1/
Valse oubliée, Example 3.14/
 Index of Quoted Composers and Works 95

Valse oubliée, with exactly notated durations, Example 3.15

Konzert-Etüde no. 3 (Il sospiro) from Drei Konzert-Etüden, Example
3.55

Martin, Frank
Prélude no. 1 from 8 Préludes pour le piano, bars 35–39, Example 5.1

Mozart, Leopold
Pattern of relative emphases, after Leopold Mozart, Example 2.1

Mozart, Wolfgang Amadeus

Sonata in G major KV 283, first movement, Example 3.28
Sonata in A minor KV 310, second movement, bars 15–19, Example
3.10/
Sonata in A minor KV 310, second movement, bars 64–65, Example 3.11
Sonata in B-flat major KV 333, first movement, Example 3.27
Sonata in B-flat major KV 333, first movement, bars 18–20, Example 6.25
Sonata in B-flat major KV 333, third movement, bar 23, Example 3.31
Sonata in C minor KV 457, second movement, bars 21–22, Example 3.44
Sonata in C major KV 545, first movement, Example 5.2
Sonata in B-flat major KV 570, first movement, Example 4.4/
Sonata in B-flat major KV 570, first movement, Example 2.3/
Sonata in B-flat major KV 570, first movement, bars 11–12, Example 3.5
Sonata in D major KV 576, third movement, Example 3.46/
Piano Concerto in A major KV 414, third movement, Example 3.16/
Piano Concerto in B-flat major KV 450, first movement, bars 76–77, Example
6.33

Prokofiev, Sergei
Visions Fugitives op. 22 no. 5, Example 3.39

Rachmaninov, Sergei
Prélude op. 3 no. 2 in C-sharp minor, Example 5.20
Études tableaux op. 39 no. 6, Example 3.21

Ravel, Maurice
Sonatina, second movement, bars 15–16, Example 3.40
Sonatina, third movement, bars 4–8, Example 3.41

Schubert, Franz
Impromptu in B-flat major op. posth. 142 D 935 no. 3, Examples 2.2 and
5.16
Sonata in E-flat major op. posth. 122 D 568, second movement, bars 43–46,
Example 6.32
96 Index of Quoted Composers and Works

Sonata in A minor op. 42 D 845, second movement, Example 3.6/

Fantasie in F minor zu vier Händen op. 103 D 940, Example 6.34
Fantasie in F minor zu vier Händen, bars 48–57, Example 6.35
Impromptu in B-flat major op. posth. 142 D 935 no. 3, Examples 2.2 and
5.16
First of Drei Klavierstücke D 946, Example 6.31
Rondo from Sonata in A major D 959, Example 3.17/
Sonata in B-flat major D 960, second movement, Example 6.2/
Sonata in B-flat major D 960, third movement, Example 2.4/
Sonata in B-flat major D 960, third movement, Trio, Example 2.5/
Sonata in B-flat major D 960, first movement, bars 27–30, Example 6.7

Schumann, Robert
Sonata in F-sharp minor op. 11, first movement, bars 74–84, Example 1.1/
Sonata in F-sharp minor op. 11, fourth movement, bars 94–101, Example 5.15
Sonata in F-sharp minor op. 11, fourth movement, bars 98–101, Example
6.10/
Kreisleriana, second piece, Intermezzo I, Example 3.48
Humoreske op. 20, Example 3.54
Sonata in G minor op. 22, first movement, bars 93–96, Example 3.53
Sonata in G minor op. 22, third movement Scherzo, bars 33–36, Example
3.49/
Sonata in G minor op. 22, third movement, bars 41–53, Example 5.21
SUBJECT INDEX

Note: Names of composers and their works quoted in the musical examples are omitted
from the subject index - please see Index of Quoted Composers and Works/List of Examples

accented note 11, 81 cantabile 7, 13, 14, 28, 32, 48, 51, 52, 54,
accents 5, 11, 13, 19, 20, 29 56, 68
accompaniment 23, 58, 67, 74 cantilena 23, 32, 36, 45, 48, 51, 59, 61,
acoustics 65 64, 68
Albéniz, Isaac 89 Casals, Pablo 90
alla breve 6, 15 Classical period, Classical era 8, 23, 50,
anacrusis 2 58, 59
André, Johann 42 clavichord 33, 56
antecedent phrase 8, 22, 24, 32, 36, 37 comic, comic effect, humour 16, 27, 35,
arpeggiation 30, 62, 65, 78 44, 66, 73
attack 11, 17, 26, 53, 75 compound duple time 5
authenticity, faithfulness to the original 3 compromise pedal 57, 67
consequent phrase 7, 8, 22, 24, 32, 36, 37
Bach, Anna Magdalena 80, 81 contrapunctus simplex 67
Bach, C.P.E. 14, 21, 49 copper printing plates 32
Badura-Skoda, Eva 11, 12, 15 counterpoint 79, 81
Badura-Skoda, Paul 11, 12, 15, 30, 33, crescendo, diminuendo, relative dynamic
56, 71 gradation 8–10
Bärenreiter Edition 42 cut-off 20, 21, 24, 34, 35, 42
Baroque, Baroque era 13, 14, 51, 53, 83, 84 Czerny, Carl 59, 71
bass line 79
bass notes 53, 57, 65, 66, 72, 74, 75, 76, 77 dampers 20, 21, 24, 25, 26, 28, 57, 61,
bell, bell-like 21, 27, 40, 48 68, 70
broken chords 46, 74, 76 dance, dance-like 20, 41, 67, 75, 84
Bruckner, Anton 83 Debussy, Claude 14, 89
Bull, John xiv decay, decay phase 21, 25, 26, 57
declamatory, declamatory style 52,
cadence 35, 55, 78, 79 53, 54
cadential 6/4 chord 12, 13, 14, 27, 66, Deutsche Pianofortemechanik 33
72, 73 Dichler, Josef 49, 70
98 Subject Index

dip-pen, metal 19 marcato 16

dotted notes 83–87 mathematically precise notation 2, 25, 83,
double-dotting 83 84, 85, 86
double repetition mechanism 26 melodic notes 43, 60, 82, 83
Dowland, John xiv metal dip-pen 19
dux, comes 79 metre, metrical 2, 9, 50, 52, 56
dynamic contour 10–12 middle pedal 68–70
dynamic shaping 12 middle voices 62, 76
Mozart, Leopold 5, 10, 15, 21, 33, 67
emphases, patterns of emphases 5–7, 17, 41
end-notes, final notes 34–45 Neue Berliner Musikzeitung 49
Neue Zeitschrift für Musik 70
Falla, Manuel de 89 note-values 2, 45
final notes see end-notes Notre Dame School xiv
finger pedal 58
folk music 89 Offenbach edition 42
fortepiano 13, 14, 20, 27 organ 26, 56, 70
Friderici, Christian Ernst 33 over-dotting 52, 84
overlapping harmonies 28
Gibbons, Orlando xiv overtones 24, 65
goose quill 19, 34, 46
Granados, Enrique 89 parlando 52
graphic representation of dynamics 9–11 part-writing, hidden 78–83
Gulda, Friedrich 30 passing dissonances 61
pedalflügel 70
half-pedal 63, 65, 70 pedal notation, incomplete 61–63
hammerflügel 20 pedal notation, older 63–66
hammerklavier 56 Pérotin xiv
hammer mechanism 20, 26, 33, 53 Philips, Peter xiv
harmonic pedal 57 phrase shaping 11, 12, 16, 54
harmonic progression 10, 23, 27, 34, 40, pianoforte mechanism 27, 33
69, 74 pizzicato 21, 23, 67
harmonium 26 plucked string 20
Harnoncourt, Nikolaus xiii polyphony, polyphonic 14, 67, 80, 81
harpsichord 13, 20, 21, 26, 27, 28, 33, 68 portato pedal 59, 60
hemiola 5
Henle Edition 46, 85, 86 Quantz, Johann Joachim 52, 56, 90
Hoffmann, E. T. A. 55, 56 quill pen 9, 34, 36
homophony, homophonic 67
human physiology 3, 6 recitative, recitativo 8, 50, 51, 52, 53
Hummel, Johann Nepomuk 59, 70 Renaissance music 5
resolution of dissonance xiii, 27, 35
impressionist 8 rests 19, 21, 23, 27, 41, 61, 62, 66, 75, 77, 84
integer valor 51 reverberation 23, 27, 59
rhythmic alignment 84, 87
jack 20, 26 ‘ritmo di tre battute’,‘ritmo di quatro battute’ 6
joke, musical 55 Romantic period 8, 32, 41, 60, 61, 75
rubato 2, 50
Landon, Christa 49
legato pedal 57, 59, 63 scansion 8, 56
lengthening of notes 2, 40, 44, 45, 77 Schindler, Anton 51
Lenz, Wilhelm von 42 sforzato 13
Léonin xiv shortening of notes 18, 19, 44
Ligeti, György 14 sighing motive 54
logic, musical logic 12, 36, 38, 39, 40, 44, 45 singing 13, 22, 26, 56, 90
 Subject Index 99

slurs 30, 32, 33 tenuto 16, 21

soft pedal, una corda 68, 70 terraced dynamics 14
sound spectrum 24 Tomkins, Thomas xiv
speech, speech rhythm 3, 5, 12, 19, 51, triplet alignment 84–87
52, 89 triple dotting 83
staccato, portato: comparative durations 17
staccato markings, different meanings of una corda 68, 70
45–48 Universal Edition 46
Stein, Johann Andreas 27 upbeat 2, 5, 8, 18, 20, 21, 22, 24, 33, 36, 37,
Strauss, Johann 90 53, 54, 58
stress patterns, patterns of emphasis 5–8 urtext 40, 45, 49, 57, 84
strong beats, weak beats 3, 5, 8, 9, 11, 13,
14, 17–19, 22–24, 41, 43, 47, 54 vertical dynamics 2
subito piano 47 Viennese style, Viennese tradition 19,
sustaining of harmonic notes 58 89, 90
swing, swinging action 20, 75, 90 voice leading 60, 67
sympathetic vibration 24, 57, 68
syncopated pedal 59, 61, 64 Wagner, Richard 83
syncopation 7, 10 waltz 2, 50, 87
weak beats see strong beats, weak beats
taste, gusto 7, 32, 54, 88 wedge-shaped 46; see also staccato
technical notation, ‘musical’ notation 48, markings, different meanings of
61, 62, 63, 66, 72–77 Wolf, Hugo xiv