<0

ESSAT

"'ARO

ON THE TRUE ART

OF

PLAYING

KEYBOARD
INSTRUMENTS
By
Cari Philipp Emanuel Bach

Translated

and Edited

W I L L I A M J.

by

MITCHELL

ASSOCIATE PROFESSOR O F M U S I C , C O L U M B I A U N I V E R S I T Y

An engraving by F. C. Kruger, Sons, Berln, which appeared in Allgemeine Deutsche Bibliothek, Vol. 34, Berln and Stettin, 1778

W N O R T O N & COMPANY I N C New

Yyf^jiA

DEPARTAMENTO p
Ob
'ON

1 '

t-jg

7i

PREFACE

HE PRESENT edition of Bach's Versuch ber die wahre

Art das Clavier zu spielen is the first complete English
translation, having been preceded only by excerpts and
sc( tions which appear principally i n books by Dannreuther, Dolinctsch, and Arnold. I t is also the eleventh i n the complete series
o printings, reissues, and editions going back to the publication
i 111 1753 of the original edition of Part One. Published privately by
Bach, i t was set up by the court primer to Frederick the Great,
(iliristian Friedrich Henning. A second printing was made i n
1759 at Berlin by George Ludewig Winter, agair with the author
as pblisher. The same pair produced the first edition of Part Two
in 1 7 6 2 . Remaining copies of both parts along with manuscript
nipplements were sold to Engelhardt Benjamn Schwickert of
I .cipzig, who reissued the work i n 1780, the year of the transaction.
1 lie nature of this edition w i l l be discussed i n the Introduction
that follows. The first and only revisin of Part One was published
A 1 111 1787 along with an unaltered Part Two by Schwickert, who also
issuecTPart Two, revised, i n 1 7 9 7 . Thus, on the face of i t , the
Essay appeared four times during the eighteenth century.
I n 1 8 5 2 Gustav Schilling edited the Essay " i n the raiment and
iiter the needs of our time." But raiment and needs combined to
produce only a curious distortion of the original. I t was published
in Herzberg by Franz Mohr, and reappeared i n 1856, published by
1 anz Stage of Berlin.
I n our own century, Walter Niemann prepared an abridged
rdition, disrespectfully ignoring Bach's stern reproof of all "com(icndium writers." This publication, based on the editions of 1 7 5 9
nd 1 7 6 2 , first appeared i n 1906, published by C. F. Kahnt of
Leipzig. I t was reprinted i n 1 9 1 7 , 1920, and 1 9 2 5 . I t has not been

r It E l A C E

P II E l- A C E

possible to secure additional inormation about a projected new

Germn edition " w i t h manuscript supplements," which was announced several times during the last decade. I t is to be published
by Gustav Bosse of Regensburg.
The present translation combines the original and revised editions of the eighteenth century. A l l of Bach's alterations, additions, and footnotes have been incorporated into the main text,
but are identified i n the editor's notes. For the rest, the organization of material i n Part One follows the Germn text i n all details. Part Two, however, has been slightly altered i n order to
make i t more accessible to the reader. The original edition and
all subsequent ones consisted of forty-one seprate chapters, some
hardly a page i n length. The sequence of these suggested a larger
organization. Thus, without altering the order of these chapters,
but simply by demoting many of them to the rank of sections, the
total has been reduced to four. The first and last alone retain their
original status. But twenty successive chapters, following the first,
have been grouped under the heading Thorough Bass, and the
next nineteen under the heading Accompaniment, this being the
subject matter of these originally seprate chapters.

i>> I01111 .111 iniroduction to the more challenging earlier pieces.

I l i e 1 -l.il ivcly large format that would be required for a practici s . 1I111011 of all of these made it impossible to include them here.
I l i i w r v c - i , they are available i n modern editions, the
Probestcke
l u i v i i i K been published by Schott, and the Sonatine Nuove by
N i < |. Also, all but the first sonata of the Probestcke appear i n
v >111111c* 11 o Le Trsor des pianistes, Paris, 1 8 6 1 - 7 2 .
l i.uiilation and editing of the Versuch would have been a far
i i i m r difhcult, if not an impossible, task had i t not been for the
H< i i o o u s assistance of my colleagues, members of our library staff,
Hi.l my students. I am particularly indebted to Professor Erich
l l r i t / m a n n for much wise counsel, to Mrs. Susanne Morse, who
n l i i c d the manuscript with great care, to Miss Jane Paul, who paly trailed and captured many elusive details of information,
i . 1 to Mr. Wolf Franck of the Music Divisin of the New York
I'uI.Ik
Library, who brought to my attention several choice tems
(tul would otherwise have escaped me.

viii

The musical examples for Part One were originally engraved

and published under seprate cover. These have been placed i n
the text and numbered serially throughout. The examples for
Part T w o were originally printed i n with the text, but were unnumbered. For purposes of easy reference and maintenance of
order these also have been numbered i n continuation of the examples to Part One. I n preparing them for the present edition,
the Gclef which Bach used for the right-hand staff, after the dying fashion of his time, was discarded i n favor of the more familiar
G-clef.
As part of the illustrative material for Part One, Bach wrote
Achtzehn Probestcke

in sechs Sonaten. Bound originally with the

examples, they are an integral part of the entire work. W i t h respect

to technical and interpretative problems contained i n them, they
range from fairly easy to quite diffkult. A t least one of the pieces,
the free fantasa of the last Sonata, is among the finest of Bach's
works. I t achieved immediate and lasting fame i n the eighteenth
century. T o the revisin of 1 7 8 7 Bach added VI Sonatine Nuove.
Written i n his broad, later style, they seem to have been designed

IX

W.J.M.

COMTENTS

Vtrfntr

vii

/ nli odllt liori

PART
l'iiirword

ONE

to Part One

/til 1 oduttion
I M A I ' I I R

O N E .

iiiAi'ii'R

T W O .

27

to Part One

30

Fingering

41

Embellishments

79

(.I'.NERAL

79

T H E

87

A P P O G G I A T U R A

T H E

T R I L L

99

T H E

T U R N

112

T H E

M O R D E N T

T H E

C O M P O U N D

T H E

S L I D E

T H E

S N A P

T H E

E L A B O R A T I O N

B H A P T E R

127
A P P O G G I A T U R A

132
136
142

O F

F E R M A T E

143

Performance

T H R E E .

147

PART

TWO

t'oreiuord to Part Two

/ a 1 roduction
BHAPTER
<;IIAITER

169

to Part Two

F O U R .
F I V E .

Intervals

172
and Their

Thorough

Bass

Signatures

180
198

T H E

T R I A D

(i)

198

T H E

T R I A D

(il)

204

O F

T H E

C H O R D

T H E

C H O R D

T H E

T H E

D I M I N I S H E D

T R I A D

T H E

A U G M E N T E D

T R I A D

O F

T H E

S I X T H
S I X T H

(i)

20g

(il)

217
222
224

xi

xii

C O N T E N T S
T H E

S I X - F O U R

C H O R D

T H E

S I X - F O U R

C H O R D

T H E

F O U R - T H R E E

C H O R D

(i)

233

T H E

F O U R - T H R E E

C H O R D

(ll)

239

T H E

S I X - F I V E

C H O R D

T H E

C H O R D

C H O R D
O F

231

(i)

243

(ll)

248

T H E S E C O N D

O F T H E S E C O N D

T H E

F I V E - T W O

T H E

F I V E - F O U R - T W O

T H E

226

(ll)

C H O R D

T H E S I X - F I V E
T H E

(l)

C H O R D

T H E

C H O R D

T H E

S E V E N - S I X

252

(ll)

2 D O

C H O R D

T H R E E - T W O

T H E

(i)

262

C H O R D

264

C H O R D

264

O F T H E S E V E N T H
O F

(i)

T H E S E V E N T H

265

(ll)

274

C H O R D

283

T H E

S E V E N - F O U R

T H E

C H O R D

O F

T H E

C H O R D

O F T H E M A J O R

S E V E N T H

T H E

C H O R D

O F T H E N I N T H

(i)

T H E

C H O R D

O F T H E N I N T H

(ll)

N I N E - S I X

T H E

N I N E - F O U R

T H E

N I N E - S E V E N

T H E

F I V E - F O U R

six.

C H A P T E R
T H E

C H O R D

286

T H E M A J O R

T H E

S E V E N T H

(i)

T H E

(ll)

3O2

C H O R D

304

C H O R D

307

C H O R D

310

Accompaniment

313

F O R T H E L E F T

P O I N T

N O T E S

D O T T E D

S U D E

P R E C A U T I O N S

O F

A C C O M P A N I M E N T

T H A T

T H E

O F

A C C O M P A N I M E N T

F O R F I G U R E D

C H O R D S

BASSES

T H E I R

N O T E S

T H E M E S

F R E E

Bibliography

351

386

407
410
412

P R E C E D E

R E C I T A T I V E

S E V E N .

316

403

T O N E S

C H A P T E R

H A N D

384

R E F I N E M E N T S

P A S S I N G

BASS

irin by P. Haas

379

F E R M A T A

C H A N G I N G

3*

/ ri|

366

I M I T A T I O N

T H E

1 i f i l c r k k the Great with His Musicians

362

C A D E N C E S

N E E D

168

348
A P P O G G I A T U R A

P E R F O R M A N C E

S O M E

BASS

N O T E S

418
420
425
426

Improvisation
F A N T A S I A

t 11I1 I'.IK<- <> ibe First Edition of Essay on the True

|f| ,./ Playing Keyboard Instruments, Part I I

319

C O M P O U N D

D O T T E D

T H E

| lili I ' . I K C of the First Edition of Essay on the True

|f| / Playing Keyboard Instruments, Part I

322

T H E

C L O S I N G

Frontispiece

raving by F. C. Krger, Sons

313

A C C O M P A N I M E N T

O R G A N

T H E

|H

305

C H O R D

S Y N C O P A T E D

T H E

297

| 11,,1,,,,, hmanuel Bach

299

A P P O G G I A T U R A S

S O M E

293

U N I S O N

O N E - P A R T

ILL US TRA TIONS

430
4 3 O

446

xiii

WTRODUCTION

1 ()( >N A I T K R Emanuel Bach's death on December 14, 1788,

pliiis were formed to erect a commemorative monument i n
^ ' iii< Mii luir lisiare he in Hamburg. T o this project, which d i d '
nrc, the renowned poet and Bach admirer, Friedrich
I niilicli K lopslock, contributed an epitaph:
T a r r y not, imitators,
For you must b l u s h i f you r e m a i n .
Cari P h i l i p p E m a n u e l Bach,
Profoundest harmonist,
U n i t e d novelty a n d beauty;
W a s great
I n text-led strains
B u t greater y e t
I n b o l d , wordless m u s i c ;
Surpassed t h e i n v e n t o r o f k e y e d i n s t r u m e n t s ,
F o r he raised t h e a r t o f p e r f o r m a n c e
T h r o u g h teaching
A n d practice
T o its p e r f e c t i o n .

" m i iluiig less than inspired, it has valu as a catalogue of Bach's

i
pal adivines, and as a summary of the basis of his fame i n
||
11 Is of his contemporaries. Our chief concern here is with the
1 ii who "raised the art of performance, through teaching and
M ni 1 c\0 its perfection."
I hr most lamous pupil of Bach was his youngest brother, Johann
1 in 1.11 . n i , who studied with h i m during the four years he spent i n
1 11111.111 er his father's death. Another was the widely known Czech
1 H 1 |an Ladislav Dussek, who spent about a year at Hamburg
i n i , ' t wiih Bach, after having already come into prominence.
D11 ik was praised by both Haydn and Mendelssohn, and described

I N 7 R O I) U C T 1 O N

/ N T R O I) U C T I O N

in W . J. Tomasi hek's Autobiography as being "the first pianist who

placed his instrument sideways on the platform, i n which our pianoforte hroes now all follow him, though they have no very interesting profile to exhibit." I t is doubtful that he learned this as any
essential part of his study with Bach.
Other pupils were less well known, such as Nikolaus Joseph
Hllmandel, the Knigsberg organist Cari Gottlieb Richter, Friedrich Wilhelm Rust, and Cari Fasch, who alternated with and later
succeeded Bach as accompanist to Frederick the Great, after being
coached i n the musical idiosyncrasies of the flute-playing monarch.
For the rest, a good deal of Bach's teaching was directed to the
amateurs i n whom he had an enduring interest.
But i t would be a gross injustice to both h i m and his pupils to
l i m i t Bach's influence solely to those who studied directly with h i m .
His fame as the founder of a school was achieved much more significantly through the agency of his music and the Essay. The latter
was called by Haydn "the school of all schools." And Mozart,
Beethoven, and Clementi added their endorsements, speaking uniformly of Bach as one whose music must be studied, not simply
played. Beethoven, after hearing the*young Czerny perform i n 1801,
turned to the father and said, " T h e boy has talent; I shall take h i m
as my own student and teach him. Send h i m to me once a week. Be
sure to procure Emanuel Bach's instruction book on the True Art

i i > n ( )nc and manuscript revisions to the entire book, Bach asked
un rvidrntly received 180 louis d'or. T o dispel any false notion
i l . .1 ilir matter was urgent, Bach added shrewdly, " I am no more
i i U i g r d 10 sell than you are to buy." The transaction was com-

of Playing Keyboard

Instruments,

so that he may bring i t to his

next lesson." Czerny goes on to relate that Beethoven's method

followed the Essay closely and included the playing of the Probestcke. A l l of these men, especially Haydn, who discovered Bach
early i n life and never forgot h i m , can be called his pupils i n this
broader sense of the term.
The Essay became famous as an instruction book almost immediately and reached many students throughout the latter half of
the 18th century. No record is available of the number of copies that
were printed, but an idea of the rate of its distribution can be gained
by consulting a letter that Bach wrote to Engelhardt Benjamin
Schwickert, A p r i l 10,1780. I n i t he expressed his willingness to turn
over control of the work to this Leipzig publisher. He wrote i n part:
" I can now say with certainty that I still possess 260 copies of Part
One and 564 of Part T w o . " For these 824 copies, with the examples

le ( I

l he H24 copies, some going back 21 years, others 18, must rep11 r-ni only the lesser part of the total printed i n 1753, 1759, and
17(111. Bul more information is provided by the fact that Schwickert,
who most certainly was not inclined to destroy merchandise on
u l i i . I i Itach placed a retail price of 3 thalers an item, reissued the
Kny in 1780 by simply altering the title page of the acquired
1 opirs to make room for the ame of the new publisher. The title
|i.iK' of this so-called " T h i r d Edition" retained the revealing
/write Auflage" of the edition of 1759.

I he 260 copies of Part One must have been largely sold by 1787,
w l u n Schwickert issued the revised edition, and the 564 copies
I Part T w o had been, exhausted by 1797, when i t reappeared i n
irvision. This represents a yearly average sale of between 30 and
,|o < opics for each part, which, if extended backward to the original
yrurs of publication, would suggest that up to the appearance of
the revisions between 1000 and 1500 copies of each part had been
piintcd and sold. A modern publisher would sniff disdainfully at
mu li paltry figures, but i t should be kept i n mind that the reading
public in the i 8 t h century was far smaller than i t is today. Also,
inethods of printing and distribution were extremely modest. The
piihlisher Gschen later ( 1 7 8 7 - 9 0 ) printed 2000 copies of Goethe's
Werke, but could find no more than 602 subscribers to the series.
And sales of the individual volumes were even smaller. The Jena
I illeratur Zeitung, a very popular and widely read journal of the
time, achieved its success on issues of 2000 copies, according to a
letier written by Gschen to Wieland. For a work like the Essay,
designed for a very limited public, to reach our suggested, coniiei vatively estimated number of copies, is indeed impressive, when
compared with these other figures. I t must also be remembered that
the copying and borrowing of books were widespread practices at
this time. Henee 1000 to 1500 copies served many more than the
cquivalent number of readers.
Other indications of the spread of Bach's Essay, but with totally

INTRODUCTION

/ N T R O I) ll (', ION

unsatisfactory results, are round in an opon letter of protest, that the

author published i n the Hamburger

unpartheiischer

Correspond-

ent, 1773, No. 7. Dated January 11, 1773, it runs i n part:

I h a v e observed w i t h greatest satisfaction t h e change t h a t has come over the
w o r l d o f k e y b o a r d p l a y i n g since the p u b l i c a t i o n o f m y Essay. I can assert w i t h out

b o a s t i n g t h a t since its a p p e a r a n c e , t e a c h i n g a n d p l a y i n g h a v e i m p r o v e d .

And

yet I m u s t r e g r e t t h a t m y h i g h m o t i v e s h a v e i n n o c e n t l y g i v e n rise t o o d

a n d even worse b a r b a r i s m s . V a i n a n d selfish addlepates are n o l o n g e r satisfied

w i t h p l a y i n g t h e i r o w n f a b r i c a t i o n s c r e d i t a b l y a n d f o r c i n g t h e m o n t h e i r students. N o ! T h e y m u s t seek i m m o r t a l i t y t h r o u g h a u t h o r s h i p . As a r e s u l t so
m a n y s c h ool a n d t e x t books h a v e a p p e a r e d since m y Essay t h a t n o e n d o f t h e m
c a n be seen. I have been i n j u r e d m o s t b y those t h a t c o n t a i n s t o l e n passages
b o t h w i t h a n d w i t h o u t a c k n o w l e d g m e n t . P l a g i a r i z i n g is free t o a l l a n d I h a v e
n o t h i n g t o say against i t . B u t i t is m o s t h a r m f u l t o s t r i p p l a g i a r i z a t i o n s o f t h e i r
p r o p e r c o n t e x t s , a n d e x p l a i n o r a p p l y t h e m i n c o r r e c t l y . E v e r y o n e k n o w s the
d a m a g e t h a t can be d o n e b y a n i n c o r r e c t

fingering,

a wrong explanation and

a p p l i c a t i o n o f e m b e l l i s h m e n t s , a t h o r o u g h l y b a d c h o r d . I can assert w i t h o u t
anger, a n d i n t r u t h , t h a t every i n s t r u c t i o n b o o k t h a t I h a v e seen since the p u b l i c a t i o n o f m y Essay ( a n d I b e l i e v e I have seen t h e m a l l ) is filled w i t h erro r s . W h a t
I say can be p r o v e d i f necessary.

However, the Essay i n its uncorrupted form reached all parts of

the Continent. "Sale of my works is chiefly i n the North, i n Russia,
Livonia, Courland, Sweden, Denmark, Holstein, Hanover, Mecklenburg, i n Lauenburg, and Lbeck
wrote Bach i n the letter to Schwickert. But it also made its way southward. Czerny, for
one, procured his copy i n Vienna. Thus the seed of the Bach in
fluence was widely scattered. Some concept of the impact of his
music and Essay on the i 8 t h century can be gained from Mozart's
famous sweeping statement, as quoted by Rochlitz: "He is the
father, we are the children. Those of us who do anything right,
learned it from him. Whoever does not own to this is a scoundrel."
The evidence that is provided by such testimony from many
sources, and by the sale and spread of the Essay, makes it clear that
Emanuel Bach's contributions were no small part of the forces that
directed the leading musical activities of the time. I t should not be
necessary to seek for superficial thematic similarities between his
works and others' i n order to prove this. Mozart gave succinct expression to the relationship of the music of his generation to Bach's:
"We can no longer do as he did; but the way i n which he did it
places him beyond all others." I t takes a true student to make so
profound an observation. Rochlitz, the source of this as well as the

JIM 11

statement, is not always dcpendable; but i n this case he

111 he 11 usted, simply because h e lacked the insight necessary for
lili 1
111 lat ion of so penetrating an observation.
ii 1. n o t pointless to inquire after the teachers of this teacher.
1' 1 < b u les llurney i n his Present State of Music wrote of Bach:
I I . n v h e loimed his style, where he acquired all his taste and rei . i i . nirnl, would be difficult to trace; he certainly neither inherited
i.loptcd L h e m from his father . . ." But i n his History, he asK lia, " l l appears from Hasse's operas, where Emanuel Bach acijii i < i l his line vocal taste i n composing lessons, so different from the
. h \ laboured style of his father." Philipp Emanuel was inileed .1 gieat admirer of 77 caro Sassone, as the Italians called that
1 ivmed (omposer of operas, but it is certain that the Essay owes
h to Johann Sebastian. " I n composition and keyboard per[01 manee, I have never had any teacher but my father," we are told
n i the Autobiography. Repeatedly i n the Essay he mentions his inilrbtedness to his father.
Km a large part of the practical wisdom contained i n i t must
hive been gathered during the years that he spent at the court i n
l l e i l i n . Engaged informally i n 1738 by Frederick, the crown prince,
In was appointed to his position as chamber cembalist on the new
I Ing's accession i n 1740. We read i n Emanuel's Autobiography: " I
. . . had the honor to accompany alone at the harpsichord the first
Ihue solo that he played as king at Charlottenburg."
lUch absorbed much through his duties at the court. His presence
was t equired almost daily, for he played the accompaniments at the
lling'8 prvate concerts. These chamber concerts were held from 7
lo o P.M. except on Mondays and Fridays, when Frederick the
Great attended the opera. Punctuality was the king's rule i n all
.illaiis, henee the musicians found i t advisable to be assembled bebu e t h e required time. Precisely at 7 he would appear and sound
llie pitch. Earlier, the waiting musicians could hear h i m rehearsing
the more challenging passages of the evening.
There was not much variety over the years. Usually the main fare
c onsisted of about six concertos played by the monarch. Later this
number was reduced to three or four. Most of these were composed
by Johann Joachim Quantz, ilute virtuoso, whose playing had
uroused the young Frederick's interest i n that instrument. Frederick
had received ilute lessons from Quantz regularly since 1727, and the
ilni).',

I N T R O 1) 11 C T 1 O N

teacher wrote approxiniately 300 concertos for the exclusive use

of the king. Occasionally, Frederick played one of his own works.
Quantz and other instrumentalists played too, their performances
being varied with arias sung by the court singers.
Emanuel Bach's music was not popular at the court. Burney,
after his visit of 1773, made no mention of the performances of
Bach's music, but wrote: " T h e compositions of the two Grauns
and of Quantz, have been i n favour with his Prussian majesty for
more than forty years . . ." And later: " I t must be owned that
many of the passages, i n these pieces of M . Quantz are now become
od and common; but this does not prove their deficiency i n novelty,
when they were first composed, as some of them have been made
more than forty years."
Burney give us an intimate picture of one of the prvate concerts:
M . Q u a n t z b o r e n o o t h e r p a r t i n t h e p e r f o r m a n c e o f t h e concertos t o n i g h t , t h a n
t o g i v e t h e t i m e w i t h t h e m o t i o n o f h i s h a n d , a t t h e b e g i n n i n g o f each m o v e m e n t , e x c e p t n o w a n d t h e n t o cry o u t bravo! t o his r o y a l scholar, a t t h e e n d o f
solo p a r t s a n d closes; w h i c h seems t o be a p r i v i l e g e a l l o w e d n o o t h e r m u s i c i a n o f
the b a n d . T h e cadenees w h i c h h i s m a j e s t y m a d e w e r e g o o d , b u t v e r y l o n g a n d
s t u d i e d . I t is easy t o discover t h a t these concertos w e r e c o m p o s e d a t a t i m e w h e n
h e d i d n o t so f r e q u e n t l y r e q u i r e a n o p p o r t u n i t y o f b r e a t h i n g as a t p r e s e n t ; f o r i n
some o f t h e d i v i s i o n s , w h i c h w e r e v e r y l o n g a n d d i f f i c u l t , as w e l l as i n t h e closes,
he was o b l i g e d t o take h i s b r e a t h , c o n t r a r y t o r u l e , before t h e passages w e r e
finished.

There is no strong reason to believe that this concert which Burney

attended after Bach's departure for Hamburg was very different,
except i n unimportant details, from earlier ones.
From other contemporaries, chiefly Johann Friedrich Reichardt
and Cari Fasch, Bach's altrnate at the harpsichord and later Kapellmeister, we w i n more information. Fasch asserted that the king,
along with Bach and Franz Benda, was a great artist i n adagio playing, but that his rhythmic sense was not always dependable, especially i n rapid passages. As monarch he retained and exercised the
right to bring the ensemble into agreement with his wayward tempos
by beating time forcefully. A story goes that a royal admirer on one
such occasion exulted, "What r h y t h m ! " T o which Bach replied
dryly, "What rhythms!" The fruits of these daily experiences appear
throughout the Essay. They can be found i n the many details of
practical advice that Bach gives to his reader.
These two sources of Bach's artistic education, his father's instruc-

TNTROD

C TI OH

<
a n d the execution of his dutiei in the service of the king, were
iipplcinented by a third, his association with many of the leading
m u s i c a l figures of his day. At the court were two of the brothers
(.laun: Cari Heinrich, music director and celebrated composer of
/ he Death of Jess, and Johann Gottlieb, conductor of the royal
>i< hestra, composer, and eminent violinist. Quantz, already menlloned, was present as chamber musician. His Versuch einer Anin sung die Flote traversiere tu spielen (Berlin, 1752) was a spur
t o Bach. Five members of the Benda family played at the court, in1 luding Franz and, for a shorter time, Georg. Johann Friedrich
Armla, too, was there as court composer, "more corpulent than
| < > i i i c l l i , or than his relation Handel," according to Burney, who
v M i t r d him later.
I n Berlin was also the quarrelsome Johann Philipp Kirnberger,
Hite Agrcola a student of Johann Sebastian Bach. Engaged as vioI m i s i in the court from 1751, he left to become musical director to
tli< l'rincess Amalie i n 1758. He wrote several important theoret -ti works and contributed many of the musical articles to J. G.
Sul/cr's Allgemeine

Theorie der schnen Knste (Berlin, first edi-

t
1771-74), the remainder being written by his pupil Johann
A l n . i h a m Peter Schulz. On adding the ame of Friedrich Wilhelm
M . i i p u r g to this roster we have reassembled the group of writers
w h o made Berlin the hub of musical theory. Burney i n The Present
State of Music writes of Berlin:
I was i m p a t i e n t t o b e g i n m y m u s i c a l i n q u i r i e s i n a place . . . w h e r e b o t h t h e
Ihcory a n d practise o f m u s i c h a d been m o r e p r o f o u n d l y t r e a t e d t h a n elsewhere,
l<y professors o f great a n d a c k n o w l e d g e d a b i l i t i e s , w h o are s t i l l l i v i n g ; a n d w h o
I m v r p u b l i s h e d t h e r e s u l t o f t h e i r l o n g e x p e r i e n c e a n d s u p e r i o r s k i l l i n treatises
w h u l i are r e g a r d e d t h r o u g h o u t G e r m a n y as classical.

I l e proceeds to enumrate, not without errors, writings by Quantz,

B.u l i , Agrcola, Marpurg, Kirnberger, and Sulzer.
With the exception of Quantz's Versuch and two publications
h y Marpurg, all of the writings of these men appeared after Bach's
Fs.uty. Its influence is apparent i n many of them, just as it is i n still
I n t e r works such as Trk's Clavierschule (1789) or Milchmeyer's
Die wahre Art das Pianoforte zu spielen (1797). But trafne i n ideas

rnn in two directions; Bach met his associates frequently and ex1 lianged opinions with them. I n his Autobiography he wrote: " M y
Cumian duties never left me enough time to travel i n foreign

<v

/ N T H O 1) 11 C T 1 O N

countries. . . . This lack . . . would liavc been harmful to anyone i n my profession, had I not had the good fortune from my
youth on to hear at cise range the finest of all kinds of music and
to meet masters of the first rank, many of whom became my friends."
Bach absorbed much from these friendships, the results of which,
carefully evaluated and recast, appear throughout the Essay. Most
easy to discover are those points on which he and his contemporaries
disagreed. Although he rarely mentions anyone by ame, i t is clear
that he and Quantz were divided on several matters.
I n order to complete our sketch of the Essay's background we
must direct our attention to Bach's abiding interest i n the proper
instruction of the musical novice. A great deal of his music was
written for teaching purposes. The title of his best-known keyboard
works, the collections for Connoisseurs and Amateurs, indicates the
spread of his interests. I n addition, the Sonatas with varied reprises
and the pieces published i n Marpurg's and others' collections were
designed for the non-professional musician. I t is true that the Essay
was written for the student whose aim was complete mastery of the
keyboard. But the Introductions to both parts show clearly that
Bach was well aware of the general state of musical instruction, indeed tht he wrote with i t i n mind. Especially is this true of the Introduction to Part One, where he writes caustically of the pretentiousness of the average teacher, his abysmal ignorance and unmusicality.
A few contemporary documents give us information on keyboard
instruction at the time. The first, a continuation of the open letter
that has already been quoted, states Bach's views on teaching the
serious and the casual student:

I N T R O I) U C T I O N

For the second g r o u p , the a m a t e u r s , t h e re is i n d e e d n o i n s t r u c t i o n b o o k , i f

this r o u l d once be impressed u p o n t h e i r teachers. I n s t e a d , o n e s h o u l d p r o c e e d
U

I used t o , u n w i l l i n g l y b u t o u t o f necessity. Before each p e r i o d , I w r o t e o u t t h e

lesson t h a t I i n t e n d e d t o give a n d c o n c e r n e d m y s e l f o n l y w i t h t h e m o s t essential

p r i n c i p i e s . N i c e t i e s of, a n d p r o p e r c o n t e x t s f o r , e m b e l l i s h m e n t s , r e f i n e m e n t s o f
.i< < o m p a n i m e n t , t h e d i v i d e d a c c o m p a n i m e n t , etc. h a d t o be o m i t t e d ; t h e y w e r e
ROI needed. T h r o u g h o u t , t h e s t u d e n t was n o t a l l o w e d t o c o m m i t a si n gl e e r r o r
l i k c those t h a t are accepted as postulates i n m a n y books. I f t h e s t u d e n t was p r e -

ptred,

i t t u r n e d o u t t h a t t h e e n t i r e t r a n s c r i b e d lesson ( w i t h o u t e x a m p l e s a n d

the r u d i m e n t s , w h i c h were p r e s u p p o s e d since t h e y can be t a u g h t as w e l l b y a v i l l;ige schoolmaster as b y t h e greatest a r t i s t ) filled a b o u t a h a l f sheet o f p a p e r .

H e n e e , f o r purposes o f t h o r o u g h i n s t r u c t i o n t h e a b r i d g i n g o f a k e y b o a r d
h a u d b o o k , even w h e n i t is d o n e w i t h o u t e r r o r s , c l e a r l y does m o r e h a r m t h a n
g o o d . A l l o f t h e c o m p e n d i u m w r i t e r s t h a t I k n o w h a v e w r i t t e n , i n c e r t a i n re-

ipects,

t o o l i t t l e , i n o t h e r s , t o o m u c h , b u t i n a l l respeets, masses o f e r r o r s . W h a t

miserable nonsense can be f o u n d i n somel A n d t h i s is t h e reason: t o j u d g e f r o m

t h e i r bo o ks , t h e a u t h o r s h a v e n e v e r s t u d i e d c o m p o s i t i o n , w h i c h t h e y m u s t b y
means k n o w i n o r d e r t o c o n s t r u c t a n a c c o m p a n i m e n t . T h i s s t u d y is n o t
i n e r e l y o f the r u l e s o f c o m p o s i t i o n ; i t bears d i r e c t l y o n a n u n d e r s t a n d i n g o f
c o m p o s i t i o n . I n a w o r d , n o o n e can p u t his t r u s t i n a k e y b o a r d i n s t r u c t i o n b o o k ,
if the a u t h o r has n o t p r e v i o u s l y m a d e h i m s e l f k n o w n a n d p r o v e d h i m s e l f w o r t h y
to be c o n s i d e r e d a n a c c o m p l i s h e d composer t h r o u g h his g o o d c o m p o s i t i o n s .

I n Der critische Musicus an der Spree, the weekly that Marpurg

published i n 1749, there are two more letters, satiric i n nature and
probably written by the publisher, that refer to keyboard playing
and instruction. These give us more information than is revealed
in many textbooks of the time. The first, attributed to an anonymous young lady, appeared i n the issue of March 11, 1749. I t runs,
in part:
M y d e a r P a p a a c q u i r e d a n e x c e l l e n t i n s t r u m e n t a t a n a u c t i o n f o r 1 5 grosc h e n
a n d 6 p f e n n i g e . I a m i n s t r u c t e d o n i t b y a v e r y clever c o u n t r y o r g a n i s t f r o m a

T h o s e w h o assert t h a t m y Essay is t o o l o n g , say n o t h i n g a n d a t t h e same t i m e

near-by t o w n . W e l e t h i m v i s i t us a t h i s c o n v e n i e n c e every t w o weeks, a n d o n

reveal t h e i r gross i g n o r a n c e . I d i v i d e a l l k e y b o a r d p e r f o r m e r s i n t t w o g r o u p s .

each t r i p he gives m e a h a l f - h o u r lesson. H e is n o t e x p e n s i v e ; we p a y h i m r o u g h l y

I n t h e first are those f o r w h o m m u s i c is a g o a l , a n d i n the second, a l l a m a t e u r s

s o r 3 ducats a m o n t h , a n d each year m y d e a r M a m a gives h i m a b u s h e l o f

w h o seek t h o r o u g h i n s t r u c t i o n . M y Essay is i n t e n d e d f o r t h e first g r o u p ; n o

oats. Everi i f I h a d n o i n t e r e s t i n m u s i c , t h i s m a n w o u l d be t h e o n e t o c r a t e i t .

p a r a g r a p h is superfluous. I n fact i t w i l l be seen f r o m t h e s u p p l e m e n t s soon t o

H e is q u i t e u n a s s u m i n g , b u t f o r a m a n o f c o m m o n b l o o d he k n o w s v e r y w e l l

a p p e a r t h a t far f r o m h a v i n g said t o o m u c h , I have n o t yet said e n o u g h . T e a c h e r s

h o w t o get a l o n g w i t h p e o p l e . H e always sits o n m y l e f t w h e n I p l a y a n d n e v e r

m u s t k n o w e v e r y t h i n g t h a t a p p a r s i n m y Essay a n d be clever e n o u g h t o select

forgets t o b o w w i t h a few l i g h t steps after each lesson. I n o r d e r t o r e l i e v e m y

t h e m a n n e r a n d o r d e r o f i n s t r u c t i o n best a d a p t e d t o the s t u d e n t s t h a t they

m i n d o f unnecessary b o t h e r he m a r k s a l l notes w i t h l e t t e r s , a l t h o u g h I a m

teach. N i c e t i e s come last, as expressed i n o n e o f m y I n t r o d u c t i o n s .

already b e g i n n i n g t o recognize t h e c-clef o n t h e first a n d o t h e r l i n e s . H e can-

N o t h i n g f u n d a m e n t a l can be l e a r n e d w i t h o u t t i m e a n d p a t i e n c e . S t u d y o f

n o t bear t h e g-clef. I t was i n t r o d u c e d a b o r t i v e l y , he says, b y m u s i c a l f r e e t h i n k e r s ,

k e y b o a r d p e r f o r m a n c e is n o t a c o m p e n d i o u s afair, a n d d a r n o t be i f i t is t o

a n d h i s teacher's g r a n d f a t h e r was t h e i r s w o r n e n e m y . H e considers

be l e a r n e d t h o r o u g h l y . W h a t is t h e r e t o say a b o u t those false i n s t r u c t i o n b o o k s

a s m a l l m a t t e r w h i c h he leaves t o m y d i s c r e t i o n , a l t h o u g h he insists o n b a n i s h i n g

fingering

w h i c h i n t h e i r a lleged b r e v i t y are a l m o s t as l o n g as m i n e ?

the t h u m b , a n d o f t e n expresses a n n o y a n c e a t those w h o m a k e so m u c h use o f i t .

10

1 N

T R O I) U

CT

INTRODUCCION

1 O N

f u l master carries w i t h h i m a Jew's h a r p o r a p i p e w i t h w h i c h h e o f t e n ac-

linrsl and subtlest topics to the hroadcst and most basic. If it disaHirrs with Quantz, or echoes Couperin, it also lashes the local peilanlic music masters.
The Essay is first and foremost a practical book that was designed
I C M for discussion than for instruction. Its ancestry runs back
through works like Mattheson's General-Bass Schule, Heinichen's

c o m p a n i e s m e so t h a t , as he expresses i t , h e c a n g i v e m e a few ideas a b o u t c o n -

(eneral Bass, to Niedt's Musicalische

Because he has n o i n t e r e s t i n o r n a n u - n t s a n d does n o t w a n t t o delay m y progress

f o r t w o o r t h r e e years, he disregards a l l o f t h e m , asserting t h a t they h a m p e r
r a p i d p l a y i n g . A l s o , he assures m e t h a t I s h a l l soon b e g i n t o p l a y the latest arias,
since I a l r e a d y have u n d e r m y fingers a b o u t a h a l f dozen c h r a l e s , i n a d d i t i o n
t o t h e S m i t h y ' s C o u r a n t e , some p o p u l a r songs, a n d t w o P o l i s h dances. So I a m
p r e p a r e d f o r m o r e d i f f i c u l t pieces. I m u s t n o t f o r g e t t o t e l l y o u t h a t m y resource-

certos.

The second letter appeared i n the issue of May 13, 1749, and is
signed Musenhold. The body of i t describes a projected method of
financing an orchestra i n a small town by means of contributions and the novel imposition of fines for such transgressions as a
lady's premature wearing of a new coiffure or a husband's withholding of a small service from his spouse. The correspondent proceeds to describe the sorry personnel of the incumbent orchestra and
concludes as follows:
B a m b o o z l e r , o u r k e y b o a r d i s t , has fine h a n d s . B u t h e is u n a b l e t o p u t t h e m t o
g o o d use e x c e p t w h e n t h e governess, h i r e d b y t h e m o t h e r , excuses herself f o r a
m o m e n t , l e a v i n g h i m a l o n e w i t h h i s y o u n g l a d y s t u d e n t s . T h e mechanics o f
fingering

are c o m p l e t e l y u n k n o w n t o h i m . I n r i g h t - h a n d t r i l l s , h e uses o n l y t h e

second a n d t h i r d fingers, r e f u s i n g t o a l l o w t h e t h i r d a n d f o u r t h t o p l a y o n a n y acc o u n t . I n p l a y i n g t h r e e - p a r t c h o r d s i n t h e r i g h t h a n d i n w h i c h t h e lowest tones

l i e a f o u r t h a p a r t a n d t h e u p p e r a t h i r d ( p a r ex. d, g, b) he uses t h e second, t h i r d ,
a n d fifth fingers, even t h o u g h t h e m i d d l e t o n e m u s t be p l a y e d b y t h e f o u r t h . A n d
so, f r o m t h e b e g i n n i n g , h e r u i n s his s t u d e n t s ' h a n d s . A l s o , he is so b a d a t
thorough-bass t h a t h e k n o w s n e i t h e r t h e tones o r t h e t o n a l i t y o f t h e c h o r d o f
t h e a u g m e n t e d second. I n a c c o m p a n y i n g he is l i k e t h e l o w l i e s t c h r a l e p l a y e r ;
h e leaps a l l o v e r t h e k e y b o a r d f r o m o n e octave t o a n o t h e r w i t h h i s r i g h t h a n d ,
as i f t h e i d e n t i t y o f chords w e r e k n o w n t o h i m o n l y here a n d t h e r e . A n d a n o t h e r p r o o f t h a t he k n o w s n o t h i n g a b o u t h a r m o n y : n o t o n l y does he p l a y a l l
mistakes f r o m p o o r copies o f arias, he transcribes t h e m n o t e f o r n o t e i n h i s
students' copy books.

I f Bach was unacquainted with these two letters, published i n

Berlin by a musician who was well known to him, it made little difference, for this famous son and pupil of "od Bach of Leipzig," as
Marpurg called him, this celebrated keyboard player who knew and
practiced at its best the music of Berlin, who knew many of the coregnant musicians of his time, was also conversant with the common day-to-day practices. The content of the Essay provides us with
direct evidence, for the discussions that i t contains r u n from the

Handleitung,

the text on

Which his father's teaching was based. Also i n the background is

I' rancois Couperin's L'Art de toucher le clavecn. There is no trace
o the speculative temper of the Age of Reason that brought forth
I rssing's Laokoon, Sulzer's Allgemeine Theorie . . . , or earlier,
(liarles Batteux's Les Beaux-Arts reduits a une mme principe.

For

works that pronounced first principies and the governing laws of

esthetics, Bach had only the practitioner's scorn. "They dispense
their alms with a completely unhappy arbitrariness," he writes i n
the Essay.

Primarily the book seeks clarification and improvement of the

keyboardist's lot through a painstaking ordering and exposition of
the several factors that relate to the practice of his art. The author's
(|ualifications were eminently suited to the requirements. Of his
practical experience and wisdom we already know. His contemporaries set the highest store on his expressive playing. As a composer
he was the leading exponent of the Empfindsamkeit, the Germn
eounterpart of the style galant. Beyond this he had an enduring
interest i n all music, as well as highly developed critical faculties.
In his Autobiography he wrote:
I t is because I h a v e n e v e r l i k e d excessive u n i f o r m i t y i n c o m p o s i t i o n o r taste,
because I have h e a r d m a n y d i f f e r e n t k i n d s o f g o o d t h i n g s , because i t has always
been m y o p i n i n t h a t t h e g o o d s h o u l d be accepted regardless o f w h e r e i t m a y
be f o u n d , e v e n w h e n i t appears i n s m a l l d e t a i l s o f a piece; i t is because o f these
c o n s i d e r a t i o n s a n d t h e assistance p r o v i d e d b y a G o d - g i v e n n a t u r a l a b i l i t y t h a t
the v a r i e t y w h i c h is a t t r i b u t e d t o m y c o m p o s i t i o n s has arisen.

Another important qualification: he was a collector by nature. I n

his estte were over 300 portraits of famous men, mostly musicians,
which he had gathered together over the years. Many of these hung
in his home i n Hamburg, where Burney saw and remarked on
them. A n d without his careful preservation of many of his father's
scores, our knowledge of the Leipzig Bach's music would be far
poorer. Wide musical experience, catholic tastes and interests, dis-

I N T i O 1) U C T ION
crimination, the collettor's habita of acquisitiveness, all of these
factors contribute to the valu of the Essay and lend to it a unique
quality. But, finally, there can be found on more than one page a
sy, sometimes a caustic, wit. Johann Friedrich Doles, a school companion and one of Johann Sebastian's successors at the Thomasschule, once said, "Like many boys of active mind and body, he was
afflicted from childhood on with the malady of the roguish tease."
Symptomatic are his remarks on local teachers, Italian accompanists, the performance of incompletely marked scores.
Nowhere is Philipp Emanuel's indebtedness to his father more
clearly expressed than i n the chapter on fingering. The son worked
out the details, but the father fixed the basic principies. However,
it is clear from the reference to fingering as "a secret art, known and
practiced by very few," that the Bach family did not discover it, but
rather organized and elaborated its technique. Other facts can be
adduced to support this view.
Of the older fingering, it can be said that it lacked systematization. I t was conditioned by earlier musical styles and was characterized i n general by a sparing use of the thumb and fifth finger with
a consequent favoring of the middle fingers. For example, i n running passages the right hand often ascended and the left hand descended by repeatedly crossing the third finger over the fourth. As
the right hand descended the thrd finger repeatedly crossed the
second. The thumb carne into repeated use only i n wide stretches
and as the left hand ascended, a common fingering being
4,3,2,1,2,1,2,1. The differences from one school to another lay essentially in the amount of use allotted to the extreme fingers. I n
Girolamo Diruta's/Z Transilvano (1593?, 1597) these hapless members are almost completely banished. More kindly disposed toward
them were the English virginalists and Germans such as Elias
Nicolaus Ammerbach, i n whose Ein New Kunstlich Tabulaturbuch (1575) the fourth finger of the left hand crosses the thumb i n
stepwise ascent. I n Francois Couperin's L'Art de toucher le
clavecn (1716, 1717) the thumb is employed frequently i n wide
stretches, and i n running passages for the left hand, but i n the right
no more frequently than others had used it. Characteristic for the
French school at this time is the replacement of one finger by
another on an unrepeated, held tone, along with direct repetitions
of a single finger i n running passages. A palpable misprint ac-

,2

/ N T R O I) U C T ION

mus for the claim that LArt de toucher . . . foreshadows the

l r v v c T fingering.
A very mportant innovation of the new method was the turning
I I I K I C I ol the thumb i n running and arpeggiated passages. The older
hngei ing made use of the thumb in large stretches and runs, but i n
iln l.uter its sol function was to strike the key and remain inactive
wliilc the second or rarely the third and fourth vaulted it. Yet the
iiiincd thumb i n the Bach fingering must have been known and
nnployed by Domenico Scarlatti, for one, for the virtuoso passages
111 his sonatas could hardly have been delivered satisfactorily withiiul it. For corroborative evidence we can cali on Franz Antn
M . I K lielbeck, i n whose Die auf dem Clavier lehrende Caecilia
(Au^sburg, 1738) the turned thumb is called for repeatedly. This
I i< 1 is of special interest here, for Maichelbeck's own keyboard
woiks incorprate many of the bravura elements of the Scarlatti
nonatas.
I'urther, Marpurg's Die Kunst das Clavier zu spielen (1750/51)
rinploys the turned thumb as a basic technique i n the performance
ol scales. I t is quite possible, however, that Emanuel Bach had a part
in the working out of this feature of Marpurg's short treatise. Cerlainly Marpurg did not hesitate to pick plums from the Essay once
it liad appeared.
II, then, the new fingering was known to some, it remained a
closed book to the rank and file of teachers and students until
llach's systematic exposition appeared i n 1753. Marpurg's satiric
Irt t ers, quoted earlier, are clear enough proof of this, and also of the
l.ic 1 that the od fingering had outlasted its function. " W h o does
not know when a new epoch began for music i n general, and for its
tnost aecurate and finest performance i n particular . . . ," wrote
llach i n his Autobiography. The new style demanded a new deI ivery.
t Bach's fingering is the foundation of modern technique. Of the
older methods but few details remain i n his exhaustive exposition,
ittuh as the crossing of 3 over 4 i n the ascending right hand, but this
only as an alternative to the new method of turning the thumb. As
keyboard style developed, as the pianoforte with its different action
carne into its own, certain extensions of technique were required.
These were provided by Clementi, Czerny, J. B. Cramer, and many
others. I f Muzio Clementi is sometimes credited with introducing

I NT

RO

DUCTION

modern finger technique, we need only read his own acknowledgment of indebtedness to the Essay i n order to restore the proper
sequence: "Whatever I know about fingering and the new style, i n
short, whatever I understand of the pianoforte, I have learned from
this book."
The most extended contemporary review of the Essay appeared
i n the Bibliothek der schnen Wissenschaften und der freyen
Knste (Vol. 10, Pts. i , 2, 1763/4). I n i t Bach's work is ranked as the
equal of Quantz's Versuch, Leopold Mozart's Violinschule, and
Agricola's translation with additions of Tosi's Opinioni. The review
is laudatory on all counts save those that refer to the chapter on
embellishments, where several exceptions are taken to Bach's treatment and organization of material. Whether these differences of
opinin are entirely justified is less significant than the fact that they
indcate that Bach did not, as indeed he could not, represent all
practices of the eighteenth century. Taste and style are important
factors i n his treatment. Further, although Bach's ordering of his
material is clear and logical, it is obvious that a somewhat different
organization might have been undertaken. A l l i n all, the chapter
on embellishments contains a large but discerning selection of
ornaments from all styles.
Ornamentation at the time of the Essay was of two kinds. There
were first the optional elaborations which performers were expected
to interplate into the pieces they played. Ornamentation i n this
sense was a dying practice. Johann Sebasian Bach had already subscribed to the writing out of every note that was to be performed.
Philipp Emanuel, following his father's practice, treats free elaborar o n only briefly, i n connection with the performance of fermatas
and cadenzas.
The second kind was the stereotyped short embellishments, the
appoggiaturas, trills, turns, etc. T o these, Bach directed his full
attention. The task that he set for himself was a twofold one. First
he classified each type and designated a distinctive sign, notation, or
position for each subtype. For example, the section on the turn
includes the turn over a note, after a note, over a tie, over a dot,
the trilled turn, the snapped turn, and the ascending turn. I n all, he
cites seven types and twenty-four subtypes of ornaments exclusive
of slight varants. While the ends at which Bach arrived are, i n certain instances, peculiar to his own judgment i n these matters, the

INTRODUCCION

idra o sortingand classifying embellishments was not at all new, for

ihis was the subject matter of many books and prefatory notes of
the lime and earlier. Much more original and provocative was the
!><( ilying of the exact musical context that was suited to each ornai i i n i i . Here Bach attempted to assist the performer who must know
where to insert unspecified ornaments. For, if the practice of provid i ng free elaborations was approaching its end, the more modest
one of inserting short embellishments was still a vigorous art. Ceri.unly it met with Bach's approval, where the other did not.
I he chapter on embellishments is a difficult, but an inescapable
Iftd rewarding assignment for the musician who would discharge
ueditably his responsibilities to i8th-century music. Generally
l>e,iking, Bach's contemporariesand later composers did not accept
his advocacy of a seprate designation for each ornament. Instead,
ihey followed the practice of using a few signs to cover all cases,
when they did not write out the ornament completely. Today we
have come to believe that each of these signs represents a single, pat
foi mua. The often gruesome results of this misapprehension can be
heard from conservatory to concert hall. Bach's chapter is a primary,
1:01 rective source work. I n it we are provided with an opportunity
lo study i n detail the exact manner i n which these ornaments were
performed by one of the most precise and sensitive artists of his
peiiod.
In the third chapter of the Essay, Bach writes of performance.
On the evidence presented by his own keyboard music, it is doubtItiI that he possessed or sought the technical wizardry of Domenico
Scarlatti. His fame derived from other sources. I t was the heightened
rxptessiveness of his playing, the daring originality of his music that
impressed his listeners. Among those who heard him and remarked
on his performance were the poet Klopstock, the musicians and
mwsicographers Marpurg, Reichardt, and Burney. The last-named
wiote i n his Present State of Music: "His performance today conVinced me of what I had suggested before from his works; that he is
1101 only one of the greatest composers that ever existed, for keyed
Instruments, but the best player i n point of expression; for others,
perhaps, have had as rapid execution: however, he possesses every
ntyle; though he confines himself chiefly to the expressive." And the
niuhor, possibly Reichardt, of Musikalischer Almanach, Alethinopel (1782) writes similarly: " T o know Bach completely one must

/ N T RO I) U C II

INTRODUCTION

O N

iy

Throughout the Essay Bach distinguishes between the learned

and galant styles i n music. He set no high store on the former, although he wrote his share of polyphonic pieces and had a deep
admiration for his father's works. His predilection was for the galant
style, French i n derivation. Yet his own music and manner of performance were far different from the patterned forms, the restrained
elegance and grace of the rococ. His manner of delivery, like his
music, was replete with personal expressiveness, with song. This
view is clearly expressed in his Autobiography: " M y principal aim,
especially of late, has been directed toward playing and composing
as vocally as possible for the keyboard, despite its detective sustaining powers. This is no easy matter if the sound is not to be too thin
or the noble simplicity of melody ruined by excessive noise. . . .
I believe that music must, first and foremost, stir the heart. This
cannot be achieved through mere rattling, drumming, or arpeggiation, at least not by me."

h< (1 and joined notes, the exe< iition ol the vibrato and portato,
dolted notes, sustained and arpeggiated iones. And like the tempo
inhalo and dynamic shading, all of these matters are of importance
only as they advance the first aim of the performer, to seek and
intcrpret correctly the true expressive content of each piece that
he performs.
T o many it must seem strange that Philipp Emanuel, modernist
and eclectic of the eighteenth century, did not employ the theories
ol Ramean, in writing the chapters on intervals and thorough bass.
I le was not ignorant of the writings of the Clermont organist whose
Traite had appeared forty years before Part T w o of the Essay. I n deed, the Essay was written after the publication of all of Rameau's
theoretical works.
liach and his father were acquainted with Rameau's theory, which
has become the basis of most of the modern writings on harmony,
hut they disagreed with it. This was made known i n a letter to Kirnherger, cited i n his Kunst des reinen Satzes (Pt. I I , Sect. 3, p. 188):
"You may proclaim that my and my deceased father's basic principies are contrary to Rameau's." Extended consideration had been
given by the members of the Bach school to the new theories of the
lundamental bass, the suppositional root, the triad as the mother
o all chords, and the seventh as the origin of all dissonances. This
is apparent from the analyses i n the Rameau manner which can be
lound, according to Spitta, i n the definitive autographs of the Sarabande and two Menuets from Johann Sebastian Bach's D minor
Ki ench Suite, and i n Fischoff's autograph of the C minor Fugue and
I) minor Prelude of the Well-Tempered Clavier. Later, Kirnberger
analyzed the B minor Fugue of Book One and part of the A minor
l'relude of Book T w o with the avowed purpose of proving the
miperiority of his own analytic procedure over Rameau's. I n only
one respect can i t be said that Philipp Emanuel made use of any
(i)f the new principies. He speaks several times of chord inversin.
Hut this principie was known before the Traite was published, having made its appearance i n Andreas Werckmeister's Hodegus
curiosus (1687) and Godfrey Keller's Rules . . . (before 1700).

Thus i n the chapter on performance the points stressed are those

concerned with expressive playing, with correct interpretation. I t
is only after attention has been directed to these matters that Bach
turns to such technical details as the notation and performance of

Bach's rejection of Rameau can be traced largely to the fact that

the latter had pronounced a theory, whereas thorough bass was esscntially a practice. Certainly, as Bach presents his material, i t is
apparent that the pervasive problems were first tactile and then

hear the wealth of Iiis imagination, tlie prolound sentiment of his

heart, his constant enthusiasm as he improvises on his Silbermann
clavichord."
As the principal practitioner of the Empfindsamkeit, with its
emphasis on the feelings, the "affections," with the clavichord as its
best-loved instrument, Bach made technical mastery of the keyboard
only a contributory factor to the expressive end that he sought.
Music here was far removed from a decorative art, from abstract
patterns of sound; it was, above all else, a vehicle for the expression
of the emotions. Music must languish, it must startle, it must be
gay, i t must move boldly from one sentiment to another; these were
the requirements that had to be met by the composer. And the performer must understand the true content of each piece that he
played. He must transmit accurately and faithfully its expressive
nuances to an audience whose heart must be stirred. This was the
core of the aesthetic doctrine of the Berlin school. Its artistic parallels were the English sentimental novis and the romantic Germans of the literary movement that became known later as the
Sturm und Drang.

<l<t.i<

iH

I N T R O I) U C T

ION

artistic, but never speculative. Thus in organi/.ing the chords of

thorough bass, Bach follows an older principie. Chords, regardless
of their origin, are grouped according to the deinitive interval that
they contain. For example, all chords that contain sevenths are
treated successively. They are the chord of the seventh, the sevensix, the seven-four, and the seven-four-two chords. Although only
the first of these is a chord i n the Rameau sense, all are chords i n
Bach's sense. Each of them must be recognized from its signature
and played instantaneously. The student's task was to lcate at the
keyboard the deinitive interval and then to bring under his fingers
the various accompanying intervals. Identification of the root, real
or supposed, did not aid h i m i n his direct gauging of intervals above
a given bass tone. Moreover, i n thorough bass some chords were
closely associated, even though their roots were not identical. For
example, above certain bass tones the six-three and six-four-three
chords were regarded as interchangeable. Knowledge of the fact that
these chords had different roots would have deterred rather than
aided the student.
The greatest difficulty with the older system was caused by the
great increase i n the numbef and variety of chords that made
their appearance i n the course of the eighteenth century. Mattheson
referred scornfully to the thirty-two posted by Heinichen, and Usted
seventy, but overlooked six of the latter's chords. Bach has twenty,
but includes many others as subtypes, chromatic variants, and alternates. It.was this unwieldy bulk of chords that aided the spread of
Rameau's system, but i t is not pointless to note that the theory
gained unquestioned acceptance only after the period of the basso
continuo had passed. Bach's method, the one he inherited from his
father, was the only effective introduction to the musical practices
of his time.
The crucial difference between Rameau and Bach is most evident
i n those places where Philipp Emanuel explains the nature of
chords. Where Rameau's emphasis rests on the vertical origins of a
chord, Bach's rests on its behavior. Repeatedly he cites context,
voice leading, rhythmic and melodic manipulation as the critical
chord-shaping factors. Thus there are two kinds of six-four chords,
those that retard a following five-three, and those that retard a
following six-three. Where Rameau calis the two identical because
their roots are identical, Bach differentiates between them because

/ N T R O l) V C T ION

thcr behavior is different. The first attempt to reconcile these two

points of view, harmonic function and behavior, was made by Kirnlurger, whose works, despite certain obvious shortcomings, should
be examined by all. He distinguishes between essential and inesscntial chords, and makes the root a determinative factor of a succeision of chords rather than a single chord.
I n general, the chapters on intervals and thorough bass are concerned solely with the rudiments of accompaniment. Attention
is directed to chord construction, doubling, and spacing. This was
(lie groundwork that must be covered by every student accompanist.
But it was hardly enough to make a skilled practitioner of the
keyboardist. So, after treating the raw material, Bach turns i n the
chapter on accompaniment to refinements, stylistic matters, and
special problems of settings, such as the treatment of appoggiaturas,
passing tones, etc. He writes of the liberties that may be taken, of
the amount of freedom from four-part accompaniment that may be
indulged, of the ways i n which a realization might be made into an
active, essential part of a composition.
On only one final point is his thorough, detailed exposition less
than adequatehe did not include a complete piece with a fully
realized accompaniment. The examples themselves are highly informative and shed light on many particulars of construction, but
they are, by nature, isolated fragments. While it would have been
impossible to construct an accompaniment i n which all problematic
matters would find illustration, nevertheless a single complete
movement would have clarified our concept of the total shape of an
accompaniment, of its balances and parallelisms.
The extemporaneous realization of a figured bass is a dead art.
We have left behind us the period of the basso continuo and with
it all the unwritten law, the axioms, the things that were taken
for granted; i n a word, the spirit of the time. T o become convinced
o| this one need merely play through the effulgent nineteenthcentury tone poems that were added as accompaniments to eighteenth-century works; or the shy, halting harmony exercises that are
prevalent i n our own day. These latter reveal their timidity all the
more clearly through their small notation. Both types, i t should be
remembered, were painfully and studiously wrought, but they fail
completely to enter the creative milieu of the eighteenth century.
T o be sure there were bad, faltering accompaniments i n the eight-

I N T R O 1) U C I I O N

2O

I N T R O l) 11 C T

centh century too. We can rcad about them here and elsewherc.
But it is illuminating to read first-hand accounts of the accompaniments fashioned by one of the greatest improvisers of all time.
W r i t i n g of Johann Sebastian Bach, Johann Friedrich Daube expressed himself as follows i n 1756:
F o r the complete practice of thorough bass it is necessary to k n o w three
species: the simple or c o m m o n ; the n a t u r a l , or that w h i c h comes closest to the

O N

21

17 If. T h i s carefully ediled work contains a realization by K i r n b c r g e r of

|i|>

thr A n d a n t e from the Trio. T h e r e m a i n i n g movements of the T r i o a n d m i r r o r

ann

(pp. 59 ff.) have accompaniments by an u n k n o w n student of the eight-

r r n t h century. T h e s e accompaniments have i n the past been incorrectly atnibuted

(.

to K i r n b e r g e r

Schirmer, 1945,

C . F. H a n d e l , Werke,
II ni ti cmbalo

(cf.

H. T.

David, /.

S.

Bach's

Musical

Offering,

pp. 99 ff.).
V o l . 48, p. 115,

concertato,

|. S. B a c h , Werke,

character of a melody or a piece; the intricate or c o m p o u n d .

Adagio. T h e keyboard part is super-

but is i n the nature of an arpeggiated realization.

I I , 2, pp. 97 ff., A r i a , Chi in amore. T h i s is the most com-

T h e excellent B a c h possessed this third species n the highest degree; w h e n

|iU-x of the accompaniments listed hfe, but even i n its elaborated qualities it

he played, the p r i n c i p a l part h a d to shine. B y his exceedingly adroit accompani-

Miggcsts an extemporaneous realization of the " i n t r i c a t e " k i n d . T h e keyboard

ment he gave it life w h e n it h a d none. H e k n e w how to imitate it so cleverly

p a n is superscribed cmbalo

w i t h either the right h a n d or the left, a n d how to introduce a n unexpected

V. T . A r n o l d , The

obligo.

Art of Accompaniment

from

a Thorough

Bass, L o n d o n ,

counter-theme against it, that the listener w o u l d have sworn that everything

1931,

h a d been conscientiously written out. A t the same time, the regular accompani-

lioui textbooks by H e i n i c h e n , Mattheson, a n d G e m i n i a n i .

m e n t was very little curtailed. I n general his accompanying was like a

An accompanimit from a thorough bass demands more than a

arefully gathered knowledge of eighteenth-century idioms. I t re(|iiires i n additio a highly creative imagination. When these two
lactors are present much of the elusive spirit of a good setting can
he recaptured. As an example, Brahms' accompaniments to Handel's Duetti e Terzetti may be cited. They appear i n Handel's
Werke, Vol. 32, 2nd ed. Nos. Ib nd X to the end of the volume.
Nos. X V - X X were published i n Handel's Duette, Peters, No. 2070.
I t is a rare privilege to be invited into a composer's workshop
to look on as he fashions a model for us, as i n the chapter on improvisation. Partial glimpses of the creator at work are provided i n
letters scattered through the centuries; and many rare vistas are
opened up to the careful student of Beethoven's notebooks. But
aside from these and the final chapter of the Essay, our only recourse
is a vast desert of textbooks on the proper writing of inventions,
.K ademic fugues, sonatas, songs, etc. Their authors' compositions
heing at best of only minor significante, such books represent "but
sccondary sources for those who wish to know intimately of the
pioblems and processes of creation.
Burney i n his Present State of Music describes Philipp Emanuel's
improvising as follows: "After dinner, which was elegantly served,
and cheerfully eaten, I prevailed upon him to sit down again to
a clavichord, and he played with little intermission, t i l l nearly
eleven o'clock at night. During this time, he grew so animated and
possessed, that he looked like one inspired. His eyes were fixed,
his under lip fell, and drops of effervescence distilled from his coun-

concer-

tante part most carefully constructed a n d added as a c o m p a n i o n to the p r i n c i pal part so that at the appropriate time the u p p e r voice w o u l d shine. T h i s
right was given at times even to the bass, without slighting the p r i n c i p a l part.
Suffice it to say that anyone who missed h e a r i n g h i m missed a great deal.

in

Lorenz Mizler also listened to Bach's accompaniments. He wrote

1738:

W h o e v e r wishes truly to observe what delicacy i n thorough bass a n d very good

a c c o m p a n y i n g m e a n need only take the trouble to hear our Capellmeister B a c h
here, w h o accompanies every thorough bass to a solo so that one thinks it is a
piece of concerted music a n d as if the melody he plays i n the right h a n d were
written beforehand. I can give a l i v i n g testimony of this since I have h e a r d it
myself.

Because thorough-bass realizations were created extemporaneously and served only an immediate purpose, there was no need to
write them out. Nevertheless, a few have come down to us, some
avowed realizations, others that partake so much of the nature of
an accompaniment that they can be used to supplement Emanuel
Bach's discussion. As listed here they range from the simple,
through the natural, to the intricate, as classified by Daube:
P h i l i p p Spitta's / . S. Bach

(Novello, 1899, I I I , 388 ff.) contains a realization

by H . N . G e r b e r w i t h corrections by B a c h of a Sonata fot V i o l i n a n d Bass by

T . Albinoni.
G e o r g P h i l i p T e l e m a n n - , SingeM a x Seiffert, Barenreiter, 1935.

Spiel-

und

General

Bass Uebngen,

ed. by

T h i s volume contains several songs w i t h fully

realized accompaniments. I t was designed as a n instruction book.

Musical

Offering

by J . S. B a c h , prepared by H . T . D a v i d , G . Schirmer,

1944,

C h . I V . T h i s chapter contains suggestive, short, but complete

examples

22

/ N T R O I) U C T

ON

tenance." Reichardt was bewitched by Bach's communicative improvisations.

Signihcant i n Bach's exposition is the omnipresence of a ground
plan, regardless of whether the subject of discussion is the short
preliminary exercise, modulation, or the complete fantasia. The
improvisatory character of this type of composition is achieved not
by a meaningless wandering from key to key, but by an imaginative
manipulation of details that fit persuasively into a unified whole.
But the relation between execution and plan is bold and free.
Nowhere does the plan obtrude. Its function is to direct the general
course of the work, and this it accomplishes by remaining quietly
where it belongs, i n the background. And when necessary i t yields
to a free twist of the foreground. Under the conditions set by Bach
the sample piece could scarcely turn out to be one of his best works.
His avowed purpose is to show the student how to construct a free
fantasia. Limitations imposed by this aim were severe. Yet for all
its circumscribed, unassuming modesty, it breathes the same atmosphere as the famous final piece of the Probestcke, also a free
fantasia.
I n this chapter, as i n many parts of the chapter on thorough bass,
Bach presents himself as an analyst. His procedure is to discuss each
inflection with relation to its normal behavior. I t is instructive
to compare such a method with the present practice of chordnaming which is passed off almost everywhere as analysis. Where the
latter is mechanical and visual, Bach's approach is aural and artistic.
The requirements of such an approach are keen perceptive powers,
the ability to evalate musical processes, and a long experience i n
the art. Bach had all of these qualities, and having them, he could
never have regarded analysis as a search for chord roots and identification tags.
The Essay was Bach's only extended theoretical work. Aside from
it and certain illustrations that appeared i n Marpurg's Abhandlung
vori der Fuge, there was only one paper that carne to print, a Suggestion for the Constructing of Six Bars of Double Counterpoint i n
the Octave (Einfall . . .) which appeared i n Marpurg's HistorischCritische Beytrge (Vol. I I I , Pt. 2, pp. 167 ff.). I t is a work more
ingenious than useful. Other writings, some of which were planned
as supplements to the Essay, were concerned with thorough bass,
fingering, embellishments, modulation, and the free fantasia. They

I N T RO I) U C T I O N

23

irmained in manuscript (Wotquenne, Nos. 111, 256, 258). What|VCI merits can be found i n these other works, they contributed
little to their author's renown. None worked as did the Essay to
mtablish him as one who "raised the art of performance through
iraching and practice to its perfection."

FOREWORD

iifeer

ie

n>ai)re

TO PART ONE

3(vt

K
imb ad)ta<fpn $rt*c<etcfm in fed# onatm
etfutett.

rfter $&e.
>

SJfrrgung

t? Suctpri.

83er(in, 1759.
rtritcft (9 Qmgt nHmia, Mnttr.

EYBOARD I N S T R U M E N T S have many merits, but are

beset by just as many difficulties. Were i t necessary, their
excellence would be easy to prove, for i n them are combined
.ill the individual features of many other instruments. Full harinony, which requires three, four, or more other instruments, can
be expressed by the keyboard alone. A n d there are many similar
.idvantages. A t the same time, who is not aware of the many demands that are made upon i t ; how i t is considered insufficient for
the keyboardist merely to discharge the normal task of every ex(( utant, namely, to play i n accordance with the rules of good performance compositions written for his instrument? How, beyond
iliis, he must be able to improvise fantasas i n all styles, to work out
cxtemporaneously any requested setting after the strictest rules of
liarmony and melody; how he must be at home i n all keys and transpose instantly and faultlessly; and play everything at sight whether
dcsigned for his instrument or not; how he must have at his cominand a comprehensive knowledge of thorough bass which he must
play with discrimination, often departing from the notation, sometirnes i n many voices, again i n few, strictly as well as i n the galant
inanner, from both excessive and insufficient symbols, or unfigured
and incorrectly figured basses; how he must often extract this
1

Title page of the first edition of way on the True Art of Playing
Keyboard Instruments, Part I

1 Clavier. T h e meaning of the term "clavier" has suffered so many ramieations

that it seemed wise to avoid it entirely and use instead the more stable "keyboard
instruments." I n J . S. Bach's time, Clavier referred to all keyboard instruments. Henee
Das Wohltemperirte Clavier should read The Well-Tempered
Clavier rather than
Clavichord. But in the generation of C. P. E . Bach and later, Clavier often meant
mly clavichord. However, there can be no doubt of the inclusive meaning of the term
in the title of this Essay, despite the further confusin added by Burney when he
calis it (The Present Stale of Music, Vol. I I , p. 263) " A n Essay on the Art of Playing
llie Harpsichord." Translation of Clavier as "keyboard instruments" or simply "keyboard" led inescapably to "keyboardist" as the translation of Clavierist.
*7

2<v

o n /: w o

H )

r o

/* A R T

O N E

thorougli bass from large stores witli unligurcd or oven pausing

basses (when other voices serve as harmonk: l'undamcnt) and with
it reinforce the ensemble; and who knows how many other things?
A l l this must be done competently, often on an unfamiliar instrument which has not been tested to determine whether it is good or
bad, whether it is playable or not, i n which latter case extenuation
is but rarely granted. On the contrary, it can be expected that,
normally, improvisations w i l l be solicited without anyone's being
concerned whether the performer is i n the proper mood, and if he
is not, without any effort being made to crate or maintain the
proper disposition by providing a good instrument.
Notwithstanding these demands, the keyboard has always found
its admirers, as well it might. Its difficulties are not enough to discourage the study of an instrument whose superior charms are
ampie compensation for attendant time and trouble. Moreover, not
all amateurs feel obliged to fulfill all of the requirements. They
satisfy as many of them as they care to or as their innate talents
permit.
However, keyboard instruction could be improved i n certain
respects to the end that the truly good which is lacking i n so much
music, but particularly keyboard music, might thereby become
more widespread. The most accomplished performers, those whose
playing might prove instructive, are not to be found i n such numbers as might perhaps be imagined. A n d yet, study by listening, a
kind of tolerated larceny, is the more necessary i n music because,
even if ill-will were not so great i n mankind, many matters would
still present themselves which cannot be easily demonstrated, much
less written down, and would have to be acquired by ear alone.
I n presenting an introduction to keyboard playing it has not i n
the least been my intention to treat systematically all of the previously mentioned tasks and to show how they may be satisfactorily
discharged. Neither the art of improvising or thorough bass is
discussed here. These have long since been dealt with i n part i n
many excellent books. I t is my aim to show the performer how he
may play solos correctly and thereby gain the approbation of connoisseurs. He who has done his part i n this respect w i l l already have
2

2 Handsachen. Mattheson (Der vollkommene

Capellmeister) writes: ". . . Everything that is played on keyboard instruments falls into two classes, solos (Handsachen)
and general bass." However, cf. Pt. I I , Introduction, f % 12-15, where Bach's description of Handsachen would seem to limit the term to certain types of keyboard solos.

FOREWO

11 1)

T 0

P A Ii T

ON E

29

accomplished much, for his facility wiil help him to succeed far
inore easily in his remaining studies. The demands made of the
keyboard as compared with other instruments testify to its comprchensiveness and many capabilities; and it can be observed from
the history of music that those who have achieved renown i n the
world of music have usually excelled on our instrument.
I n all matters, I have had i n mind chiefly those teachers who have
lailed to instruct their students i n the true foundations of the art.
Amateurs who have been misled through false precepts can remedy
matters by themselves from my teachings, provided they have already played a great deal of music. Beginners, by the sanie means,
will easily attain a proficiency that they could hardly have believed
possible.
Those who expected a voluminous work from me are i n error.
1 believe I deserve more gratitude if through brief precepts I have
made practicable, easy, and agreeable many things that are quite
difficult i n the study of the keyboard.
I ask the forbearance of my readers for repeated mention of
divers truths, made necessary partly because matters on hand demanded it, partly to avoid frequent cross references, and finally,
because I feel that certain principies cannot be stated too often.
And perhaps some w i l l find themselves embarrassed by these truths,
although I wrote them without the slightest intention of malice.
Should the present work meet with the apprval of connoisseurs,
I might find therein the encouragement to continu with a few
supplements. The acclaim accorded this work by the musical public has given me the incentive to enlarge the present third edition
with textual additions and six new keyboard compositions i n fulfillment of the promise made i n the Forewords to the first and
second editions.
3

T h i s entence was added to the ed. of

|j * See Note 17, Introduction to Pt. I .
8

1787.

INTROD

INTRODUCTION

TO PART ONE

H E T R U E A R T of playing keyboard instruments depends

on three factors so closely related that no one of them can,
or indeed dar, exist without the others. They are: correct
fingering, good embellishments, and good performance.
2. Owing to ignorance of these factors and their consequent
absence from performance, keyboardists can be heard who after
torturous trouble have finally learned how to make their instrument sound loathsome to an enlightened listener. Their playing
lacks roundness, clarity, forthrightness, and i n their stead one hears
only hacking, thumping, and stumbling. A l l other instruments
have learned how to sing. The keyboard alone has been left behind, its sustained style obliged to make way for countless elabrate
figures. The truth of this is attested by the growing beliefs that to
play slowly or legato is wearisome, that tones can be neither slurred
or detached, that our instrument should be tolerated only as a
necessary evil i n accompaniment. As ungrounded and contradictory as these charges are, they are, nevertheless, positive reactions to the false art of playing the keyboard. I n view of the opinin
that the keyboard is unsuited to present styles, and the consequent
discouragement of many from studying it, I fear that the skill, already waning, which has been brought to us chiefly by great performers, w i l l suffer an even worse decline.
3. I n addition to the neglect of the three factors mentioned
above, students are taught the wrong position of the hand. A t least
their errors remain uncorrected. Thus, the last possibility of their
playing competently is removed, for it is easy to imagine the kind of
sounds produced by stiff, wire-strung fingers.
4. Most students are required to play their teacher's own works,
8

V C TI

O N

1 O

PART

O N E

)i

loi nowadays it secms to be scandalous not to compose. Good pieces

ly others which might be studied proitably are withheld under
i lie- pretext that they are obsolete or too difficult. Worst of all, there
|| a malicious prejudice against Frenen keyboard pieces. These
have always been good schooling, for this country is sharply disi inguished from others by its flowing and correct style. A l l necessary
finbellishments are clearly indicated, the left hand is not neglected,
nor is there any lack of held notes; and these are basic elements
ni the study of coherent performance. Our pedants can often play
nothing but their own fabrications; their abused, awkward fingers
tIrlver these stiffly; they can compose only what their hands can
subdue. Many are held i n high esteem who hardly know how to perlorm tied notes. Consequently there arise great quantities of miserable works and abominable students.
5. T o begin their studies, pupils are racked with vapid
Murkys and Gassenhauer i n which the left hand, its role reduced
lo a mere thumping, is rendered useless for its true employment.
Actually the left hand should be preferentially and intelligently
exercised i n order to attain a facility equal to that of the right,
which by the very nature of things is constantly active.
6. Should the student i n listening to other music acquire a more
discriminating taste, he is thenceforth revolted by the pieces he
must practice; and, convinced that all keyboard music is poor, he
seeks refuge i n arias which, when well set, and sung by reputable
voices, are suitable for the development of good taste and the study
of good performance but not for the development of the fingers.
7. Teachers feel that they must do violence to these arias and
transcribe them for the keyboard. Along with other disparities, the
left hand as usual is badly treated. A sluggish or even a drum bass
is assigned to i t which, even when it is suited to the character of the
piece, is more harmful than beneficial to the left hand.
8. As a result of all this the keyboardist loses the special asset,
1

1 An accompaniment consisting of broken octaves.

A Gassenhauer was a popular song or vaudeville in Bach's time and later. Cf.
Moser, Musiklexikon.
A similar protest was raised by F. W. Marpurg in Die Kunst das Clavier zu spielen:
" T h e arias which students are given serve to form their taste, but not to crate facility and dexterity, above all in the left hand."
* Trommel-Bass; a scornful reference to basses fashioned out of repeated tones.
They were an intermedate step between the older linear bass and the somewhat later
harmonic, catapulting bass.
2

?2

/N

7 R O I) U C I I O N

l O VA R T

ON E

/ N T R

possesscd by no other instrumcntalist, of keeping time easily and

enunciating its smallest fraction with exactness, an ability which he
acquires i n playing idiomatic keyboard music, for this comprises
more syncopations, short rests, and rapid dotted rhythms than any
other type of composition. O n our instrument such difficult elements are quite easy to master because one hand assists the other to
hold the beat, and this brings i n its train the spontaneous sharpening of the rhythmic sense. 1 know from experience that rapid
syncopations and, above all, short rests cause great ado among the
most rhythmically sure and accomplished of other instrumentalists.
A l l enter too late, even though other parts that enter just ahead of
them provide the same assistance as the keyboardist has i n his hands.
T o the latter these things are easy even when he omits the left hand
or accompanies with other instruments. Provided that he is certain
of the tempo, his entrance w i l l always be exactly right. Quantz i n
his Flute Method, page 113, even advocates a delayed entrance
(which goes to prove that a correct entrance is nearly impossible)
and thus takes the lesser of two evils.
5

9. The student i n playing the basses described above develops

a stiff left hand, for the amount of harm done by performing quick
repeated notes without a change of fingers is nearly incredible.
Many have suffered this injury as a result of industrious and prolonged study of thorough bass, wherein they have had to play such
notes with either hand but particularly with the left i n octaves.
9a.
I take this opportunity to express my thoughts on the per7

Remainder of paragraph from ed. of 1787.

T h e reference is to Johann Joachim Quantz (1697-1773), engaged as flute teacher
to Frederick the Great, and later as his Kammermusicus and court composer. H i s
encyclopedic work, Versuch einer Anweisung die Flote traversiere zu spielen (Berln,
1752), reads, in the passage that Bach mentions, " A t short rests which occur on downbeats i n place of principal notes, caution must be taken to avoid coming i n too soon
with the note that follows. For example, when the, first of four sixteenths is a rest,
the performer must wait half again as long as the valu of the rest, for the following
notes must be shorter than the first. T h i s applies as well to thirty-seconds." (Ch. 12,
f 12). A n abridged modera edition, prepared by Dr. Arnold Schering, was published
at Leipzig in 1906.
1 T h i s long paragraph appeared originally as a footnote and was retained as such
in all editions. At first glance it is puzzling i n both its length and somewhat heated
style, for the practice advocated here pf omitting certain repeated notes from the left
hand was not new. For example, Heinichen (General Bass, 1728) recommends not
only the expedient suggested, but also the changing of repeated notes into broken
octaves or into a right-hand Alberti figure against simple chords i n the left (cf.
Arnold, Art of Accompaniment
from a Thorough-Bass,
p. 774 ff.). Saint Lambert
makes similar suggestions (p. 196, l.c). However, the entire matter clears up on turn-

i)

uc

10

T ION

rA

R T

0N

formance of quick rcpetitions in the left hand for the benefit of

those who are charged with the task of playing thorough bass. T h e
device, an everyday occurrence in the present style, offers great risk
of stiffening and ruining the best of hands. This remark can stand
as a good argument against those who ask expressly that all notes
written for the left hand be performed. Certainly the right hand is
not required to accompany all notes, particularly when the bass
contains so common a device as the passing tone. The quick repetitions of whose hazards I speak are eighth notes i n rapid, and sixteenths i n more modrate tempos. Further, I assume that another
instrument is playing the bass with the keyboard. When i t plays
alone, these notes, like the tremolo, must be performed with alternating fingers. Although the consequent omission of the octave
w i l l detract from the sonority of the bass, this small defect is to be
preferred to other greater evils. W i t h an accompanying bass instrument i t is best to omit one, three, or five notes according to the
tempo and meter and strike the others i n octaves (or double octaves
with both hands i n a fortissimo), employing a heavy attack, somewhat sustained so that the strings will vbrate sufficiently and the
tones blend with each other. I n order not to confuse the ensemble,
the first bar may be played as written and notes left out from there
on. Another means that may be employed when every note must
be played is to strike the key alternately with each hand. However,
it has been my experience that, because the right hand usually
comes i n late, this expedient may upset the ensemble, a fact that has
strengthened my conviction that the keyboard is and must always
remain the guardin of the beat. I t is considered correct and indeed
8

10

ing to Quantz's Versuch, where we find the following statement (Ch. X V I I , Sect. I V ,
" O f the Keyboard Accompanist," f 32): " W i t h regard to allegro movements it is important that the accompanist . . . . possess the facility in his left hand to play
everything clearly and purely, . . . that when many eighth notes occur on a single
tone he play every one with his left hand, and avoid the practice of some who for
untimely reasons of convenience, strike one note and omit the following three or
even seven, especially in vocal works." When all factors are considered, i.e., the public a r o n dat^s of both Essays (Quantz 1752, Bach 1753), the cise association of both
men as fellow court musicians, etc., it becomes clear that Quantz is referring to Bach
and that Bach's paragraph is an elabrate rejoinder and defense of his own practice.
s See C h . I V , f f 68-78, and C h . V I , especially f 3, for a description of passing tones
in the eighteenth-century sense.
Cf. C h . V I , "Passing Tones," f f 4, 7-12.
. . . Wie die Schwrmer. T h e term was used chiefly with reference to the string
tremolo, although Marpurg applies it to repeated notes at the keyboard. T h e Italian
word was Bombo. Arnold translates literally, "resembling crackers" (Arnold, op. cit.,
p. 776).
1 0

?./

1 N T R O I) U C T 1 O N

7 O PA H T

ONE

advisable for the accompanist to rcpcat chords which are sustained

by the rest of the ensemble in order to rnaintain a clear indication
of the meter; the correetness and advisability of omitting notes
should be conceded for the same reason; the more so when such
omissions are compared with the hazards and impracticability of
literal performance. A n d this latter is really hazardous, for on other
instruments such notes are played with the tongue or the wrist; but
the keyboardist must express these rumblings with a rigid arm
when, due to octave doubling, he is unable to employ alternating
fingers. I n doing this the left hand grows stiff and incapable of performing the passage satisfactorily for two related reasons: The first,
because i n a prolonged contraction all muscles are employed; the
second, because most of the fingers are inactive. I t can be established
through experiment that the left hand and entire arm grow so
tired, twisted, and taut from thumping away at a drum bass that i t
is impossible to play anything active afterwards. This clinking
noise is impossible for another reason: Many of the drum basses that
are encountered today cannot be survived because of their sheer
length. I n all styles the other musicians have occasional rests; the
keyboard, however, is constantly at work often for as many as
three, four, or even more hours without respite. Assuming that one
were hardened to such labor, even the most dependable musician
would begin eventually to waver drowsily and unwittingly through
fatigue. The drum bass, i n most cases devoid of expression and
calling for little mental effort, can only annoy and weary a performer who, as a consequence, loses the inclination and ability to
perform stirring passages fittingly. Further, this injurious clinking
is contrary to the nature of the harpsichord as well as the pianoforte, for both instruments are thereby robbed of their natural
tone and clarity; the tangent of the harpsichord seldom reaets
quickly enough. The French, who understand the keyboard and
know that i t is capable of more than mere strumming, take pains
even today to inform the keyboardist that i n such passages not all
notes are to be played. Further, broad accented tones contribute to
the expression of basses that carry dots or dashes over the first of a
group of notes. Many cases arise where a clear, strong attack with
both hands is not only advisable but mandatory. The keyboard,
entrusted by our fathers with f u l l command, is i n the best position
11

i i Cf. C h . V I , "Performance," f 18.

INTRODUCTION

TO

PART

ONE

35

10 assist not only the other bass instruments but the entire ensemble
i 11 maintaining a uniform pace. And yet the best musician, fatigued,
m a y find it difficult to guide even his own pace at t i m e s , even though
he might be the master of his powers under normal conditions. Such
being the case with one performer, how much more important it is
that our expedient be employed i n an ensemble; the more so because t i m e is beaten today only i n larger C o m p o s i t i o n s . The tone of
the keyboard which, correctly placed, stands i n the center of the ensemble, can b e heard clearly by all. A n d I know that even diffuse,
elabrate compositions played by impromptu, average performers
can b e held together simply by its tone. If the first violinist stands
nears the keyboard as he should, disorder cannot easily spread. I n
arias, the singer's burden is lightened by our means when the tempo
changes precipitately, or when all parts scramble while the voice
alone has long notes or triplets which because of their divisin demand a clear beat. The less the bass is preoecupied with difficult,
involved runs, the more easily w i l l i t be able to rnaintain a steady
pace; the more i t is, the more frequent w i l l be the spectacle of compositions starting more vigorously than they end. Should someone
hasten or drag, he can be most readily corrected by the keyboardist,
for the others w i l l be too much concerned with their own figures
and syncopations to be of any assistance. Especially those parts
that employ the tempo rubato w i l l find herein a welcome, emphatic
beat. Finally, i t is easy (and often necessary) to make minor changes
of tempo by this means because exact perception w i l l not be hindered by the keyboard's excessive noise, and, i n addition, those performers located i n front of or beside the keyboard w i l l find i n the
simultaneous motion of both hands an inescapable, visual portrayal
of the beat.
12

10. Teachers try to make amends for a stiff left hand by teaching their students to favor the right and garnish adagio or expressive passages with a wealth of pretty little trills to the revulsin of
good tste. These are often interchanged with senile, pedantic embellishments and fumbling, inept runs i n the playing of which the
fingers seem to grow choleric.
11. Before we proceed to remedy these faults with wellgrounded instruction, something remains to be said about keyboard instruments. Of the many kinds, some of which remain little
12 a . Pate I V .

36

INTRODUCTION

TO

l'ART

ONE

known because of defects, others because they are not yet in general use, there are two which have been most widely acclaimed, the
harpsichord and the clavichord. The former is used in ensembles,
the latter alone. The more recent pianoforte, when it is sturdy and
well built, has many fine qualities, although its touch must be carefully worked out, a task which is not without difficulties. I t sounds
well by itself and i n small ensembles. Yet, I hold that a good
clavichord, except for its weaker tone, shares equally i n the attractiveness of the pianoforte and i n addition features the vibrato
and portato
which I produce by means of added pressure after
each stroke. I t is at the clavichord that a keyboardist may be most
exactly evaluated.
18

14

12. A good clavichord must have i n addition to a lasting, caressing tone, the proper number of keys, extending at the very least
from the great octave C to the three-lined e. The upper limit is
needed for the playing of scores written for other instruments. Composers like to venture into this high register because many instruments can reach it quite easily. The keys must be properly weighted
to help raise the fingers after ea'ch stroke. I n order that the strings
may be attacked as well as caressed and be capable of expressing
purely and clearly all degrees of forte and piano, they must be
resilient. T a u t strings keep the tone of a vibrato pur; yet they
should not be too taut or they w i l l sound strained and the performer
w i l l be unable to achieve any volume; on the other hand, if they are
too lose, they w i l l sound impure and unclear if they sound at
all. The keys must not fall too deep, and the pegs must be tightly
fitted so that the strings w i l l be capable of withstanding the full
forc of an attack and remain i n tune.
1 6

16

13.

A good harpsichord must have uniform quilling i n addition

Die Bebung und das Tragen der Tone. Cf. C h . I I I , f f 19-20, and the accompanying Note 17.
i * J . F. Reichardt wrote: "Not only does Bach play a slow, singing adagio with
the most touching expression (to the embarrassment of many instrumentalists who
could imtate the voice with far less difnculty on their own instruments), he sustains,
even in this tempo, a note six eighths long with all degrees of loudness, both in the
bass and the treble. But this is perhaps possible only on his very fine Silbermann
clavichord for which he has written sonatas in which long sustained notes occur.
" A n d it is the same with the extraordinary power which Bach can give to a passage: it is the utmost fortissimo. Another clavichord would go to pieces under it.
Likewise, his most delicate pianissimo would not sound at all on another clavichord."
is T h i s sentence, up to the semicolon, appears as a footnote in the ed. of 1787.
18 T h i s sentence appears as a footnote in the ed. of 1787.
l s

INTRODUCTION

TO

l'ART

ONE

37

10 .1 good tone and the proper tange. The tests of the quilling are
neat, facile execution of embellishnients, and an equal, quick
1 caction of each key as the thumbnail sweeps over the entire manual
with a light, uniform pressure. The action of the harpsichord must
not be too light and effeminate; the keys must not fall too deep; the
lingers must meet resistance from them and be raised again by the
jacks. On the other hand, they must not be too difficult to depress.
For the benefit of those whose instruments have less than the desirable range, I have so constructed my Lessons that they may be
played on a four-octave keyboard.
14. Both types of instrument must be tempered as follows: I n
tuning the fifths and fourths, testing minor and major thirds and
chords, take away from most of the fifths a barely noticeable amount
of their absolute purity. A l l twenty-four tonalities w i l l thus become
usable. The beats of fifths can be more easily heard by probing
fourths, an advantage that stems from the fact that the tones of the
latter lie closer together than fifths. I n practice, a keyboard so
tuned is the purest of all instruments, for others may be more
purely tuned but they cannot be purely played. The keyboard plays
equally i n tune i n all twenty-four tonalities and, mark well, with
full chords, notwithstanding that these, because of their ratios, reveal a very slight impurity. The new method of tuning marks a
great advance over the od, even though the latter was of such a
nature that a few tonalities were purer than those of many present
non-keyboard instruments, the impurity of which would be easier
to detect (and without a monochord) by listening harmonically
to each melodic tone. Their melodies often deceive us and do not
expose their impurity until it is greater than that of a badly tuned
keyboard.
17

15. Every keyboardist should own a good harpsichord and a

good clavichord to enable him to play all things interchangeably.
A good clavichordist makes an accomplished harpsichordist, but not
17 Probekcke. T h i s refers to the pieces written by Bach to illustrate the Essay.
They are available in the following modern editions:
(a) Sechs Sonaten fr Klavier, ed. by Erich Doflein, Edition Schott, Nos. 2353-54
(1935). These are the "18 Lessons in 6 Sonatas" which appeared with the first ed. of
the Essay. T h e Adagio of Sonata V appears in Zeitschrift fr Msik, no. 103, 1936,
Notenbeilage, and the Fantasa of Sonata V I can be found in Vierteljahrschrift
fr
Musikwissenschaft, No. 7, 1891.
(b) Kleine Stcke fr Klavier, ed. by Otto Vrieslander, Nagels Musik-Archiv, No. 65,
Hannover, 1930. Includes (Nos. 19-24) the VI Sonatine Nuove which were published
in connection with the third ed. of the Essay in 1787.

jS

INTRODUCTION

TO

l'ART

ONE

the reverse. The clavichord is needed fot the study of good performance, and the harpsichord to develop proper finger strength.
Those who play the clavichord exclusively encounter many difficulties when they turn to the harpsichord. I n an ensemble where
a harpsichord must be used rather than the soft-toned clavichord,
they w i l l play laboriously; and great exertion never produces the
proper keyboard effect. The clavichordist grows too much accustomed to caressing the keys; consequently, his wonted touch
being insufficient to oprate the jacks, he fails to bring out details
on the harpsichord. I n fact, finger strength may be lost eventually,
by playing only the clavichord. On the other hand, those who concntrate on the harpsichord grow accustomed to playing i n only one
color, and the varied touch which the competent clavichordist
brings to the harpsichord remains hidden from them. This may
sound strange, since one would think that all performers can express only one kind of tone on each harpsichord. T o test its truth
ask two people, one a good clavichordist, the other a harpsichordist,
to play on the latter's instrument the same piece containing varied
embellishments, and then decide whether both have produced the
same effect.
18

16. After mastering the requisite knowledge of keys, notes, rests,

rhythm, and so forth, students should be made to spend a good deal
of time practicing only the examples of fingering, slowly at first and
then more rapidly until i n due time good fingering, as difficult and
varied as it is at the keyboard, w i l l become so much a matter of
habit that it may be put out of mind.
17. Above all, practice i n unisn those examples i n which fingering is given for both hands, so that they w i l l become equally dexterous.
18. Then ply at the chapter on embellishments, practicing
them until they can be performed skillfully with proper facility.
Since this is an assignment on which a lifetime may well be spent
J . F. Reichardt wrote: "Bach's manner of playing would not have been devised
at all without the clavichord, and he devised it only for the clavichord. But he who
once masters this instrument plays the harpsichord quite differently from those who
never touch a clavichord. For him harpsichord compositions may be written which
under the hands of the mere harpsichordist become insipid, often unintelligible, and
disconnected."
Also: "Soul, expression, feeling, these things Bach gave first to the clavichord, and
the harpsichord could not receive the smallest degree of them save from the hand
of him who knew how to anmate the clavichord."
1 8

INTRODUCTION

T 0

PART

ONE

y}

(cnibellishments demand in part more technique and dexterity

than runs) the student should not be detained after his ability, depcnding on his aptness and age, is great enough to stand him i n
tnodest stead.
19. After this, proceed directly to the Lessons; play them first
without the ornaments, which should be practiced separately, and
then with them, according to the rules treated in the chapter on
performance. This must be done first at the clavichord and later
interchangeably with the harpsichord.
20. The whole approach to performance will be greatly aided
and simplified by the supplementary study of voice wherever possible and by listening closely to good singers.
21. I n order to become oriented at the keyboard and thus make
easier the acquisition of a necessary skill at sight reading, i t is a
good practice to play memorized pieces i n the dark.
22. I n notating the Lessons, I have scored everything that
seemed necessary; and I have played them many times with great
care so that not even the smallest detail would escape me. Therefore I believe that if everything is given careful attention, finger
dexterity as well as taste w i l l be improved to the point where other,
more difficult things may be studied.
23. I n order to avoid ambiguity I have written all triplets without the numeral three, and indicated the detached notes not with
strokes, but with dots, and abbreviations such as f., p . , and so forth,
without a period i n most places.
24. Because I wanted to publish a complete work illustrative
of fingering i n all keys, the use of embellishments, and all varieties
of expression, I could not prevent an increase i n the difficulty of the
Lessons. I considered it better to serve everyone rather than append a collection of very easy pieces and leave many things untouched. I hope that after clear preliminary instruction, the fingerings and performing directions, painstakingly added, w i l l make the
more difficult pieces a great deal easier I t is dangerous to delay the
student with too many easy things, for no progress can be achieved
in this manner. A few simple pieces at the beginning suffice, after
which the wise teacher w i l l do better to introduce his pupils
gradually to more challenging works. It is in accord with the art of
teaching and the reason asserted above that by this means the student w i l l be unaware of the increasing difficulty of his tasks. My

4<> 1 N T R O I) II (1 T I O N

7 O

P A R T

O N E

dcceased fatlier made many succcsslul cxpcriments of a similar nature. He introduced his pupila directly to his modcrately difficult
pieces. Therefore, no one need fear my Lessons.
25. Should some because of their facility be inclined to read
the Lessons at sight, I urge them first to study every smallest detail
with proper diligence.

CHAPTER

ONE

FINGERING

1
O A L A R G E extent the shape of an instrument determines
its fingering. I t would appear to be most arbitrary i n the
case of keyboard instruments, for the arrangement of the
keys is such that any one of them may be depressed by any finger.
2. For this and other reasons the study of fingering is a treacherous path along which many have erred. For one thing, there is only
one good system of keyboard fingering, and very few passages permit alternative fingerings. Again, every figure calis for its own distinctive fingering, which may require modification simply through
a change of context, and the comprehensiveness of the keyboard
creates an inexhaustible wealth of figures. Finally, the true method,
almost a secret art, has been known and practiced by very few.
3. This erring is the more considerable, the less one is aware
of it, for at the keyboard almost anything can be expressed even
with the wrong fingering, although with prodigious difficulty and
awkwardness. I n the case of other instruments the slightest incorrectness of fingering is usually betrayed by the downright impossibility of performing the notes. As a result, all manner of things
have been ascribed to what is believed to be the difficulty of the
instrument and the compositions written for i t .
4. From these remarks i t can be seen that correct employment
of the fingers is inseparably related to the whole art of performance.
More is lost through poor fingering than can be replaced by all conceivable artistry and good taste. Facility itself hinges on i t , for experience w i l l prove that an average performer with well-trained
fingers w i l l best the greatest musician who because of poor fingering
is forced to play, against his better judgment.
5. Because almost every figure requires its own, distinctive fin4

.2

FINGERING

gcring, presen t-da y musical thoughl, so radically difcrent from that

of the past, has devised a ncw method of exccution.
6. Our forefathers were more concerned with harmony than
melody and played in several parts most of the time. We shall soon
learn that i n this style the position of each finger is immediately apparent since most passages can be expressed i n only one way and
are variable to only a limited degree. Consequently, they are not so
treacherous as melodic passages with their far more capricious fingering. Furthermore, i n earlier times the keyboard was tuned differently and not all twenty-four keys were available as they are now.
Consequently, the variety of passages was not great.
7. Henee, today, much more than i n the past, no one can hope
to play well who does not use his fingers correctly. My deceased
father told me that i n his youth he used to hear great men who employed their thumbs only when large stretches made i t necessary.
Because he lived at a time when a gradual but striking change
in musical taste was taking place, he was obliged to devise a far more
comprehensive fingering and especially to enlarge the role of the
thumbs and use them as nature intended; for, among their other
good services, they must be employed chiefly i n the difficult tonalities. Hereby, they rose from their former uselessness to the rank
of principal finger.
8. Because this new fingering is such that everything can be
played easily with i t at the proper time, I shall expound i t here.
9. However, before proceeding to the actual use of the fingers,
I must mention certain points, some of which must be known i n advance of our study, others of which are so important that without
them even the best rules are futile.
10. The performer must sit at the middle of the keyboard so
that he may strike the highest as well as the lowest tones with equal
ease.
11. When the performer is i n the correct position with respect
to height his forearms are suspended slightly above the fingerboard.
12. I n playing, the fingers should be arched and the muscles
relaxed. T h e less these two conditions are satisfied, the more atten1

T h e few extant, notated fingerings attributed to J . S. Bach are largely in the

older style. They may be found in his Werke, 36.4, pp. 126, 224, 237. However, cf.
Spitta, Bach, Vol. I I , pp. 34-41; The Bach Reader (Norton, New York, 1945), pp.
223, 306-312; and Dolmetsch, The Interpretation of the Music of the XVIlth
and
XVIIIth
Centuries, pp. 412 ff.
1

F INGERI

NG

lion must be given to them. Stifncss hampers all movement, above

all the constantly required rapid extensin and contraction of the
hands. A l l stretches, the omission of certain fingers, even the indispensable crossing of the fingers and turning of the thumb demand
this elastic ability. Those who play with fat, extended fingers suffer from one principal disadvantage i n addition to awkwardness;
the fingers, because of their length, are too far removed from the
thumb, which should always remain as cise as possible to the hand.
As we shall see later, the principal finger is thereby robbed of all
possibility of performing its services, whence i t comes about that
those who seldom use the thumb play stiffiy, something that those
who use i t correctly can not do even willfully. For the latter, everything is easy. This can be observed immediately i n a performer:
If he understands the correct principies of fingering and has not
acquired the habit of making unnecessary gestures, he w i l l play the
most difficult things i n such a manner that the motion of his hands
w i l l be barely noticeable; moreover, everything w i l l sound as if i t
presented no obstacles to him. Conversely, those who do not understand these principies w i l l often play the easiest things with great
snorting, grimacing, and uncommon awkwardness.
13. Those who do not use the thumb let it hang to keep it out
of the way. Such a position makes even the most modrate span uncomfortable, for the fingers must stretch and stiffen i n order to encompass i t . Can anything be well executed this way? The thumbs
give the hand not only another digit, but the key to all fingering.
This principal finger performs another service i n that it keeps the
others supple, for they must remain arched as i t makes its entry
after one or another of them. Those passages which, without the"
thumb, must be pounced upon with stiff, tensed muscles, can be
played roundly, clearly, with a natural extensin, and a consequent
facility when it lends its assistance.
14. I t is evident that the muscles cannot remain relaxed or the
fingers arcl+ed i n leaping or stretching; and even the snap calis for
a momentary tensin. These are, however, the rarest cases and take
care of themselves by their very nature. A l l others follow the precepts of Paragraph 12. Especially children's hands, not yet fully
2

2 Das Schnellen. T h e nature and execution of this technical element of clavichord

playing are described i n C h . I , f 90; C h . I I , " T h e T r i l l , " % % 8, 36, and " T h e Snap,"
Un.

44

F I N G ERIN

FINGERING

grown, should be trained to stretcli as lar as possible, rather than

leap everywhere with the lingers bunched, as so ol'tcn happens. In
this manner it will be easy to strike the keys accurately, and the
hands will not readily depart from their proper position of swinging
horizontally over the keyboard, which they tend to lose i n leaps
by inclining to one or another side.
15. Pupils need not be alarmed when a passage must be tested
by their teachers i n order to ascertain its best fingering. Occasionally, doubtful cases arise which, even when they are played
correctly at the first reading, might require reflection before the
fingering can be recited to another person. Teachers are rarely
provided with a second instrument at which they might settle such
issues while accompanying their pupils. From all this it can be seen,
first, that i n spite of the endless variety of fingerings, a few good
principies are suficient to solve all problems; and second, that
through diligent practice, execution becomes so mechanical that,
eventually, a stage is reached where, without further concern, full
attention may be directed to the expression of more important matters.
16. While playing, always think ahead to the approaching
notes, for these often necessitate modification of a normal fingering.
17. The form of one hand being the reverse of the other, I have
found it advisable to illustrate the exceptional cases i h contrary
motion i n order to make them identically applicable to both hands.
Most of the examples that did not cali for inversin have been fingered for both hands so that they might be practiced i n unisn.
Every opportunity to practice i n this manner must be seized, as
recommended i n the Introduction. The clef signatures of each
example indcate the hand for which the fingering numeris are
intended. When numeris appear both above and below the notes,
those above refer to the right hand and those below to the left, regardless of the clef.
18. Having disposed of these preliminary points, all grounded
in Nature, let us now proceed to the school of fingering. Here, too,
3

As described in f f 86-92.
* T h e examples for the right hand appeared in the customary descant clef in the
original. In order to make them more accessible to modern readers, out of touch with
C-clefs, they have been transcribed in the familiar G-clef.
3

I shall buik upon Nature, for a natural fingering devoid of unnecessary strain and extensin is clcarly the best.
19. The shapes of our hand and the keyboard teach us how to
use our fingers. The former tells us that the three interior fingers
are longer than the little finger and the thumb. From the latter we
learn that certain keys are longer and lie lower than the others.
20. I shall follow the usual designations by indicating the
thumb with the numeral 1, the little finger with 5, the middle finger
with 3, that next to the thumb with 2, and that next to the little
finger with 4.
21. I shall cali the raised and recessed keys by the more usual
than correct ame of half tones.
22. I t follows directly from the statements of Paragraph 19 that
the black keys belong essentially to the three longest fingers. Henee,
the first principal rule: Black keys are seldom taken by the little
finger and only out of necessity by the thumb.
23. I have found it advisable because of the great variety of
passages to construct all types of examples; some i n one voice,
others i n several, some i n conjunct motion, and others i n disjunct.
24. The scales have been arsanged according to keys in the
first examples, which w i l l illustrate all twenty-four, ascending and
descending. Thereafter, the order of illustrations will be as follows:
Progressions i n several parts; spans and leaps, because these can be
more readily gauged after the study of progressions, or even traced
back to chords; and, finally, tied or held notes, a few licenses, exceptional cases, and certain expedients. The Lessons w i l l account
for the remainder. I n appending these with their continuous passages of all types, I believe I have served a greater purpose and
stimulated more interest in the difficult study of fingering than I
could have hoped for had I amassed quantities of fragmentary
examples, for these would have made the work overlong and unendurable.
25. Chrnge of fingers is the most important element i n our
study. Our five fingers can strike only five successive tones, but
5

s Halbentne. T h e "raised and recessed keys" of Bach's day were not universally
black, henee he could not use the convenient "black keys," which appears from this
point on as the translation of the term that he adopted with easily understandable
misgivings.
Die A bwechselung der Finger.
6

43

46

FINGER

F I N G

INC

there are two principal means whereby wc can extend their range
as much as required, both above and below. They are the turning
of the thumb and the crossing of the tingers.
26. Of the five fingers, the thumb alone is naturally adept at
turning under. Flexible and propitiously short, it is the only one to
be concerned with this technique, which is employed when the
fingers, playing i n their normal order, cannot encompass the range
of a passage.
27. Crossing over is a technique limited to the remaining fingers. I t occurs when a longer one vaults a shorter, including the
thumb, i n order to strike a tone that lies beyond the natural range
of the fingers. This device must be practiced u n t i l it is brought to
the point where the fingers w i l l not interlock.
28. These are to be avoided: T u r n i n g the thumb under the
little finger, crossing the second finger over the third, the third over
the second, the fourth over the fifth, and the fifth over the thumb.
29. The correct application of these two techniques can be
learned most readily from the patterns of scales. I n playing these
and runs based on them our precepts find their principal employment. I t is understood that i n the performance of scalewise runs
which begin or end differently from those illustrated here the performer must allocate his fingers so that they w i l l come out correctly
without his feeling obliged always to use the assigned finger on a
given tone.
30. Figure i represents the ascending scale of C major with
three fingerings for each hand. None of them is impracticable, although those i n which the third finger of the right hand crosses the
fourth, the second of the left hand crosses the thumb, and the
7

ERING

47

3 2

thumbs strike / are perhaps more usual than the others. Applications of each are shown i n Figure 2.
31. Figure 3 illustrates the descending scale of C major. Here,
too, there are three fingerings, all of which are good i n various
situations, as indicated in the examples of Figure 4, although aside
from these cases which require the specified fingerings, one may
turn up more often than the others.
Figure 4

Figure 3

'

..

\
12
11mi

12ij

Jy

*
3

3
4

} fi
f

r
^>
^ J 'ft EJ

f0

J
J

7 Das Untersetzen.
8 Das berschlagen.
Observe that one important technique of the older fingering is not ruled out, the
crossing of 3 over 4. It appears more than once in the fingerings of scales, and in
f 30 is among those that Bach expressly prefers.
8

32. I t can be seen i n Figures 2 and 4 that owing to the necessity of paying heed to the approching notes, the little finger is held
in reserve in stepwise passages and is used only at the beginning or
when a r u n happens to terminate exactly with it. This is illustrated
in the examples of scales where its use is specified. Elsewhere its
place is usually taken by the thumb. I n order to avoid confusin with regard to the little finger and to illustrate a more ex-

./S

FIN

l 1 N G K R 1 N G

tended change of lingers, 1 have led the scales hcyond the octave.
33. In Figure 5 we find A minor, ase enditig, with two fingerings
for each hand. The best are those directly above and below the
notes. Nevertheless, the others may be applied to good ends, as i l lustrated in Figure 6. However, since many more might be devised
were I inclined to construct the examples, and since those i n Figure
6 are not as natural as those that I have recommended, I include
them here more as a warning than as an endorsement, the more
so, because I know that they enjoy scattered popularity. Their defect is the assignment of d to the thumb despite the succeeding e
and two black keys, for the thumb is best used immediately before
black keys. I n any event, this fundamental rule should be observed:
The thumb of the right hand is brought i n after one or more black
keys in ascending, before them i n descending, and the left thumb
after in descending, and before i n ascending. Those who have this
rule i n their fingers will consider it unusual to commit the thumb
too soon before black keys.
Figure 5

f&

-2
1

Th ?

Figure 6

rk*'
4

\ -

m $

34. The descending scale of A minor is represented in Figure

7 with three fingerings. Because, as i n C major, there are no black
keys, all are good and practicable. Less usual than the others is that
in which the thumb takes d.
35. The ascending G major scale and its three fingerings appear
in Figure 8. Those marked with an asterisk are the least usual. The
middle one in the G-clef and the lowest i n the bass present an opportunity to state a new rule: Crossing the fingers, that is, passing
l

19 T h i s turns out to be the most usual modern fingering, at least for the right
hand. Perhaps Bach's reservations are concerned with the use of the fifth finger, which
does not appear in the alternative fingerings.

G F. R I N G

49

Figure 7

2
3

the second finger over the thumb or the third finger over the fourth,
is applied primarily to passages with no accidentis, where, if necessary, it may occur several times i n succession. Occasionally, it may
be used i n connection with a single black key in this manner: The
thumb or the fourth finger plays the tone immediately preceding
the black key, which is then struck by the second or the third finger,
an action easily performed by either, owing to their convenient
length. Thereupon, and i n accordance with the rule stated in Paragraph 33, the thumb takes its assigned place as a matter of course.
Example a of Figure 9 might stand for an exception to our rule
were it not more usual to execute the passage by turning the thumb
as i n Example b. I t is better i n such cases to cross the second finger
over the thumb than the third finger over the fourth. I n order to
illustrate the crossing of fingers in connection with black keys I
have written two octaves of this scale.
1

Figure 8

* i
w -f

#4
5

1
?

ra

ti

i
-4-

Figure 9

^
36. G major, descending, also with three fingerings, is illustrated i n Figure 10. That in which the thumb takes c is clearly the
least frequent; the farthest from the notes are the most dangerous;
but all may be used.

/<'/ N G E R I N G

F I N G E R I N G

5/

40. F major, descending, appears in Figure 14 with two fingerings in the G-clef and three in the bass, of which those directly
above and below the notes are the most usual. There is nothing irregular about the others. They are to be noted because there may be
need for them at times.

4-4-4.

Figure 14

37. E minor, ascending, has only one good fingering (Figure

11). I t is inadvisable to take the fourth step, a, instead of the fifth,
~b, with the thumb unless the succession demands i t . Contrary to the
rule of Paragraph 33, the thumb must avoid g when the ascending
scale ends on the octave, or there will be too few fingers to complete it. As we shall see later, this rule suffers a few exceptions,
which, however, do not i n the least reduce its valu i n the complete
school of fingering.
Figure 11

.
9

Figure 12

i j -

-3-

f
2

f
3

39. F major, ascending, has only one good fingering for the right
hand, as i n Figure 13, but three for the left, all of which are useful
in certain situations and should therefore be practiced.
Figure 13

if

-i

2"
4

3
2

1
1
4
3

2
- 2 - 1|
2 3
1
2
1
2 3
3

-3
3

r +-\

41. D minor, ascending, as shown i n Figure 15 has three fingerings for each hand, all of which are good and should be practiced,
although those farthest removed from the notes are somewhat less
usual than the others.

38. E minor, descending, is illustrated i n Figure 12 with two

fingerings for each hand, of which those directly above and below
the notes are the best.

I
0
1
1
f
r
ji
f

m1-4--4*j
1 1

-f- T

42. D minor, descending, appears i n Figure 16 with two fingerings for each hand. Those that lie farthest from the notes are poorer
because the black key, 6-flat, calis for the thumb on a.
Figure 16

A
f(T\

"Y
J

"

' ' h m
T

3
3

71
1

11

4
sm
w
i
l

1 ti
#

1
1

2
1
1
4

43. Only one fingering is possible for B-flat major, ascending

and descending, as illustrated i n Figure 17.
Figure 17

l< l N G E

ja

RING

44. G minor, ascending, in Figure iK lias two lingerings for

the right hand and three for the left. Those directly above the notes
and farthest below follow the rule of Paragraph 33. However, the
others will prove serviceable on occasion.
Figure 18

1 2
y}g-b-^

fh
mf
**

0
4
4
2

3
3

1
3
3 J*i
?
#
BE
fT
T

3
0t
3
1
mi
*i 3
m3 " T " ~
t
7~ 1
{ 3 8 - 1
2
4
3
4

45. G minor, descending, i n Figure 19 has only one fingering.

I t is understood that when a passage does not begin exactly as i l lustrated here, the appropriate finger must take the initial tone.

E I N G E R

N G

55

48. 11 minor, ascending, appears in Figure 22 with one fingering

for each hand. When a passage for the left hand begins below the
first step of the scale, the thumb replaces the fourth finger on b.
I n this connection note that the fingering for the higher octave
should be used to play all variants of the beginnings of scales. There
is an unavoidable departure i n the right hand from the rule of Paragraph 33. Those who have this rule i n their fingers must be careful
not to assign the thumb to d instead of e difficulty that makes the
scale rather treacherous.
Figure 22

Figure 19

46. D major, ascending, i n Figure 20 has only one fingering for

the right hand, but three for the left. According to the rule covering
the use of the thumb, and i n all passages that begin or end differently from the illustration, the fingering immediately below the
notes should be employed. However, the others are good, particularly i n the illustrated case, and should be practiced. The second
fingering in the bass demonstrates the merits of crossing the fingers
as discussed i n Paragraph 35.
Figure 20

1g

o
3

2
2

011
J

m
3

1 =1

47. D major, descending, i n Figure 21 has three fingerings for

the right hand and two for the left, all of which are useful.
Figure 21

9
w>

ni i

1 } ?^ *= =11

7 T y t
ff

P3

J -

49. B minor, descending, with one fingering is shown i n Figure

23. T o construct an altrnate setting for the right hand, begin with
the little finger, place the thumb on e, and the third finger on d i n
order to bring i n the thumb again on b. This setting, while it is not
incorrect and may be used, is good for only one octaVe. Extensin
of i t might easily prove confusing.
Figure 23

50. A major, ascending, is illustrated i n Figure 24 with one

fingering for the right hand and two for the left. That which stands
just below the notes agrees with the frequently cited rule and i n
most cases is more useful than the one below it, which, nevertheless,
is required at times.
Figure 24

51. A major, descending, i n Figure 25 has only one fingering. I t

is understood, as previously noted, that when the scale begins above

F I N ( Eli

y/

INC,

11

N C. 1:

RINC

55

the initial tone of the illustralion (he right hand takes a with the
thumb rather than the little finger. Also, when a passage for the left
hand in this key begins on the tonic degree, 1-2-3 should be substituted for 2-3-4.

Figure 28

Figure 25

55. C-sharp minor, ascending, has only one good fingering, as

illustrated i n Figure 29.
Figure 29
=F5=

52. F-sharp minor, ascending, with one fingering appears i n

Figure 26. The valu of the rule cited i n Paragraph 33 w i l l become
apparent in the forthcoming scales, for as they increase i n accidentis or black keys, they grow simpler, less treacherous, and
easier to learn.

56. B major, ascending and descending, and G-sharp minor,

descending, take only one fingering, which is shown i n Figure 30.
The latter, ascending, is different i n the size of its intervals but not
in the fingering, as we can see i n Figure 31.

Figure 26

Figure 30

i
53. F-sharp minor, descending, has, according to Figure 27,
one fingering i n common with A major. We learned i n Paragraph
50 that the additional setting for the latter i n the ascending left
hand is used only occasionally. As we proceed, note that descending
minor scales employ the same fingerings as major scales with the
same key signature or, i n the case of enharmonically equivalent
signatures, the same fingering as major scales whose tonic degrees
lie a minor third above those of minor scales.
11

Figure 27

54. E major i n Figure 28 has a simple fingering for both hands,

ascending as well as descending. C-sharp minor, descending, is the
same. Because anyone can determine the steps of descending minor
scales from the statement i n Paragraph 53, I shall omit their illustration as superfluous unless they have an exceptional fingering.
n Le. a minor third, not as notated, but at the keyboard.

S E

Figure 31

57. F-sharp major, ascending and descending, has one fingering

i n common with E-flat minor, descending, as illustrated i n Figure
32. The same is true of E-flat minor, ascending, except for the difference i n the size of its intervals and the notation (Figure 33). I n
the left hand there is an exception to the rule of Paragraph 33, according to which the thumb should take d rather than c.
Figure 32

11 n H

Figure 33

>

J6

FINGERING

58. I) llal or Cshaip major, its fingering applicable to ascending as well as descending scales, appears in Figure 34. B-flat minor,
descending, takes the same. B-flat minor, ascending, and its fingerings, of which there are two good ones for the left'hand, appear i n
Figure 35.

FINGERING

57

only to progressions within an octave. Note that as scales lose accidentis (which oceurs in the ascending minor before other scales)
the number of fingerings increases.
Figure 38

Figure 34

te

- ' ^
3

Figure 39

59. A-flat major i n Figure 36, ascending and descending, has

one fingering i n common with F minor, descending. The latter
with its ascending execution is shown i n Figure 37. The left hand
has two good fingerings, of which the one directly below the notes
is the better, although the other demonstrates anew the remarks of
Paragraphs 35 and 46.
Figure 36

Figure 37
f_

-93==

Figure 35

-ar -8-fcr

t "y |
p

0
4

' i

60. E-flat major is illustrated i n Figure 38. The fingering applies

to ascending and descending scales, as well as to C minor, descending. This latter scale, ascending, i n Figure 39 has two good settings
for each hand, of which the more removed from the notes apply

l \

II

m
2

61. From the study of these scales we learn that the thumb
is never placed on a black key, that it may be used after the second
finger, after the second and third fingers, or the second, third, and
fourth, but never after the fifth. Note that i n order to rnaintain a
uniform fingering i n playing through two or more octaves of a
scale with its seven degrees, the thumb is usually employed once
after the second and third finger, and again after the second, third,
and fourth. I n ascending with the right hand and descending with
the left this action of the thumb is called turning, a technique
which must be practiced until the principal finger has learned to
turn and take its note automatically. The performer who has
reached this point has gained the summit of fingering.
62. We learn, further, that a crossing oceurs when the second
finger, the second and third fingers, or the second, third, and fourth
pass over the thumb or when the third passes over the fourth. Later,
we shall find an exception, allowed under certain conditions, i n
which the fourth crosses the fifth. Also, a case w i l l arise i n the
study of embellishments i n which the third finger strikes a key
after the second. However, this striking must not be confused with
a crossing, which refers only to those cases where one finger crosses
another which is still depressing the key that i t has struck; i n the
former the initial finger leaves the key and the hand is shifted.
63. Finally, we learn that scales with few or no accidentis per12

13

12 See, in the present chapter, f 93.

i See C h . I I , " T h e T u r n , " f 30.

FIN

( F R I N (.

m i l the gicatcst varia!ion with rsped to fingering and the techniques of turning and crossing. I he others have only one execution. Henee, because the former have many fingerings, because both
techniques must be correctly applied to them without confusing one
with the other, and because a fingering, once it has been chosen,
must be retained in all registers, particularly with regard to the
thumb, the so-called easy keys are, i n fact, much more challenging
and elusive than the so-called difficult ones. These have only one
execution, i n which the thumb soon learns through practice to take
its tones efortlessly. These keys are called difficult because they
are never or, at best, rarely played or employed i n their own right.
As a result, their notation as well as the location of their tones remains unfamiliar. Once forbidding, when they were played without the thumb or the correct use of it, the difficult keys have become
inviting, thanks to the true study and employment of the fingers.
Thus, i n earlier times one of the great advantages of the keyboard,
the facility with which it can express all twenty-four tonalities, lay
hidden behind ignorance. While speaking of accidentis, I must
state my opinin concerning their employment. Our forerunners
followed the correct practice of placing an accidental before each
altered note which did not succeed itself directly. Today, one accidental is considered sufficient for several such notes. Accidental
signs must be used generously to clarify unexpected modulatory
shifts and their occasional resultant ambiguities.
14

F I N (' E R 1 N

cases and licenses will be reserved for discussion at the end of this
chapter.
66. We shall now treat progressions in parts. Leaps w i l l be
included in this discussion because under normal circumstances
they must be devised with a view to an unforced execution by
fingers of average length and are therefore fingered in the same
manner as part progressions. Should some find it more comfortable,
because of their longer fingers, to take chords, arpeggios, or
stretches with a fingering different from that recommended here,
they may do so, provided that the comfort is not imaginary. I have
stressed leaps and stretches i n a slow movement, the B-flat adagio,
in order to make them easier. Those who wish to practice them
rapidly by themselves may do so.
67. Adjacent tones, struck simultaneously, are taken by adjacent fingers. The preceding and following tones determine which
pair of fingers is to be used. Examples of such seconds are contained
in Figure 40. Observe that the thumb avoids black keys. I n the examples, notes without fingering numeris are to be played by the
finger assigned to the preceding note. Each clef appears only once
and remains i n forc until replaced by another.
15

64. Crossing and turning, the principal means of changing the

fingers, must be applied i n such a manner that the tones involved
in the change flow smoothly. I n keys with few or no accidentis the
crossing of the third finger over the fourth and the second over the
thumb is i n certain cases more practicable and better suited for the
attainment of unbroken continuity than other crossings or the turn.
W i t h regard to the latter, when a black key acts as the pivot the
thumb is conveniently provided with more room i n which to turn
than i n a succession of white keys. I n keys without accidentis crossing should cause no stumbling, but i n the others care must be exercised because of the black keys.
65. A l l runs must be approached i n the light of these scales
and the two techniques derived from them. Certain exceptional
1 4

Remainder of paragraph from ed. of

1787.

50

is T h e middle movement of Sonata I I . Cf. Pt. I, Introduction, Note 17.

/ N C E R I N G

6o

68. Ilrokcn scconds are played l>y altrnate fingers as illustrated in Figure ,| i . Allernation is better for this kind of passage,
nsually shirred, than a repeated (inger which causes an excessive
detaching of the notes. It should be noted here and more frequently
as we proceed that the thumb and second finger of the left hand are
used, generally, in those places where the right hand employs the
second and third fingers.
10

17

El

i
}

Figure 42

ti !
5

0 1

9*

i \

i 1

i? Das Fortsetzen eines Fingers.

| 3

3 !

ftl

II 1 "

'

II <te 9

li

'ff

ll

11

' l "

c.
TT05

-JF

.1

im

giif^is?
8

rrr 5 .... 1
*****
HtHW1

eiltft$*m
_,

r - f r

'

- 8 * Wv f
-~

T~4

ff *'|.........
T

e.

e.
t i

I T y3

11

3 4

Abwechseln.

sf
I
4

31

i "
2

~~

r - p 11
1

g2

r4
r g

f*m*t
4

i Das

\\

%
'ti

f ***

61

f
4

-Xt

quent than others. 1, 2, 3, being unnatural, are to be avoided.

RING

ifU!'f=F

5 5 4

9
69. Thirds are played by the fingers which are indicated i n the
several examples under Figure 42. Here, too, attention must be
directed to preceding and succeeding notes. The thumb and the
little finger do not play black notes, except when a contextual leap
makes it necessary. Because successive thirds are often encountered,
I have introduced several examples i n order clearly to indcate the
necessary finger changes. The little finger may also strike a black
note when the accompanying finger does likewise. Viewed i n this
light, the fingering for the right hand i n Example a is not as good as
that i n Example b or that for the left hand i n Example c. The
little finger is neither repeated directly or succeeded by another
(d). Normally, it is employed but once and then only on the extreme notes of a succession of thirds (e) unless single tones intervene, as i n Example /. Note that repeated thirds are played by repeated fingers, as i n the third and following examples. The same
applies to successive thirds i n fast tempos, like those i n Example g,
for a change of fingers is more difficult. I n conclusin, observe that
many fingerings are used on thirds, although some are more fre-

N CE

*-

_*_JJ

ui

* J

n
? f < i i fUr H
g'

62

F I

F 1 N (; /<;

NC.ER1NC,

.s

70. In slower lempos, broken thirds, singly o in succession, are

played in the same manner as the thirds desci ibcd in the preceding
paragraph. Several successive broken thirds in rapid tempos are
played with a pair of repeated fingers, s or 4, so long as black keys
do not intervene (Figure 43, Example a). When they do, the finger5

ing is changed, the thumb being withheld from them (b). 8 or 1 are
employed i n passages containing a sustained or an interpolated
tone (c). When necessary, the thumb may play black keys i n such
spans.

RING

71. Fourths are taken i n the manner of Figure 44. I n the examples written i n the G-clef the lowest notes are to be played by
the left hand, and i n the bass clef the uppermost notes by the right.
Broken fourths i n a slow tempo have the same fingering. A succession of these without black keys is executed by a pair of repeated
_

fingers, 4 or 2 (a). W i t h black keys, 4 may be used, but only once at

112

a time (b). Broken fourths may also be played 2, 3, 4, or 3, provided

that the succeeding notes cali for i t , as shown i n example c and
further.
Figure 44
1 4 1

3B

* C

11

*1
J
" i f l LL

2 5

| 22

2 5 2

1,

.f.

*^

4 1

b.

2 5

5
1

4 1

(,.

V INGERI

FINGERING

N V.

65

of Figure 47. When the thumb <Ktupies a black key it cannot be

CrOSSed by other lingers in the manner of Figure 48.
Figure

r-7-4--i

1

r~n

5 1

4 2

* *2

WH"
4

0
1
5

2
1

*
2
5

1
ni
II m
1
4

*1

a
2
-0

*2
5

IT

1 1

2
-U0
0

Tu

J) 11

JE-

&

it j'
j

1>

Figure 46
1

^=

72. Fifths and sixths may be taken i n three ways, as illustrated i n

Figure 45. Figure 46 indicates the execution of a series of sixths.
Broken sixths are played i n the manner discussed previously i n
the cases of thirds and fourths. I n stretches such as these the little
finger may be repeated directly; i n other words, it may be played
before the extremity of a passage has been reached.

-9

g):
J

Figure 45

vfl

i s i Mi

rT"H

fe. '

20

1m

"ai

Figure 48

rs^t

73. Sevenths and octaves are played l . Those who have long fingers and find it easy to take a seventh containing a black key with
2 l may do so. Beyond this, it is permissible i n playing these
large intervals to use the thumb as well as the little finger on black
keys without further ado.
74. Octaves that leap, particularly in the left hand, where they
appear most frequently, cali for the repeated thumb and little finger. Those who are not sufficiently trained to execute the octave
doublings of thorough bass can practice by playing any given bass
first with the thumb and then with the little finger. I n doing this,
progress w i l l be made not only i n a fundamental kind of finger
repettion, but also i n becoming familiar with the keyboard.
75. I n leaps of an octave, preceding or succeeding notes may
require the second finger to take the place of the thumb, or the
fourth finger to take the place of the fifth, as shown i n the examples
r

76. We shall now discuss the execution of three-toned chords.

Their fingering within the interval of a fourth is shown i n Figure
49. The additional tones of examples a and b cali for a special fingering.

77. Figure 50 illustrates the execution of three-toned chords

within the interval of a fifth. W i t h respect to example a, note that

66

/' / N C. E

lil

(,

y- /

in addition to the minor triad built from /, the same fingering applies to those from c, c-sharp, /-sharp, g, g-sharp, fr-flat, and b. And
with respect to Example b, the fingering of the major triad built
from d applies equally to those built from c-sharp, c-sharp, e,
g-sharp, a, t>-flat, and b. The longer third finger rather than the
fourth takes the third of major and minor triads, particularly when
it falls on a black key.
Figure 50

78. Three-toned chords within the interval of a sixth are taken

in the manner of Figure 51. Figure 52 illustrates the same chords i n
the interval of a seventh, and Figure 53 i n the interval of an octave.
Any finger may play black keys in wide stretches, as stated i n Paragraph 73, for a measure such as this is always better than an avoidFigure 51

A/

c; /-; ni

able strain. The execution of Figure 51, Example a, is comfortable

for some hands.
79. The fingering of four-toned chords appears i n the examples
of Figure 54. Example a represents such a chord within the interval of a fifth; Example b, within the interval of a sixth. Chords
built from the first degrees of the major keys Usted i n Paragraph 77
may be taken i n the manner of the illustration i n the bass clef under
Example b. The chords of Example c lie within the interval of a
seventh, and those of Example d within the interval of an octave.
The illustrations marked with double asterisks i n Example c are
for long fingers. The examples marked 1, 2, 3, 4 refer to the triads
of Figure 50, Examples a and b, now become four-toned chords; all
Figure 54
2
6
5
18

b.
1

3?fmH*- * i
? 1 v l r J.51
1

7?

67

Figure 53

5
2

Ir 4 H '

5
3
2

a.

1 l#

I tII 1i
h
1

5
1
2
m

II

Figure 52

1 1
2 3
5 5

1 1
5 5
is T h i s sentence and Figure 51, Example a, are from the ed. of

1787.

68

IINC.

E R I N C,

of the triads discussed in I'aiagraph 77 are lo be taken similarly

when a f o u r t h tone has been added to theni.
80. W h e n either of the outer parts of four-toned chords falls
o n a black key, i t is best to take a fingering i n w h i c h , i f possible,
either the t h u m b or the l i t t l e finger is n o t used. However, since
i t is n o t always comfortable to o m i t the l i t t l e finger, whence i t plays
black keys more often than the t h u m b , the p e r f o r m e r should be
g u i d e d by contexts and, n o t a l l fingers being alike, seek after an
unforced and n a t u r a l setting, especially i n spans. A slight discomf o r t being preferable to a greater one, i t is better to c o m m i t the l i t t l e
finger or the t h u m b to a black key than to o m i t t h e m and cause an
excessive, hazardous stretch. Performance of a succession of f u l l
chords is fcil itated by changing the fingers whenever possible.
8 1 . W h e n b o t h outer parts of four-toned chords fall on black
keys the t w o shortest fingers may be employed w i t h o u t f u r t h e r ado,
for w h e n b o t h play such keys the entire h a n d moves to the rear of
the keyboard, thereby r e m o v i n g the objections to their use.
82. Since arpeggiated and leaping passages are largely reducible
to chords, they should be played according to corresponding rules
of fingering and examined i n the l i g h t of particulars already discussed. T h e examples fashioned o u t of the chords i n Figure 55 w i l l
make m y m e a n i n g clearer to the reader.
Figure 55

\
_1_

FIN

('. E R I N C,

83. Good performance, as w e l l as preceding tones, calis for an

occasional slight change i n the fingering of b r o k e n chords. T h e
t h i r d finger is sometimes better t h a n the f o u r t h i n descending arpeggios, a l t h o u g h the latter is to be preferred w h e n the tones are
struck simultaneously (Figure 55, Example 1). W i t h regard to good
performance, the keyboardist cannot always expect the degree of
clarity f r o m a weaker finger that a stronger finger achieves q u i t e
readily, for c l a r i t y is w o n p r i n c i p a l l y t h r o u g h u n i f o r m pressure. I n
this respect, those w h o are left-handed possess no small advantage
o n o u r i n s t r u m e n t . I n Example 2, Figure 55, the t h i r d of the
arpeggio has been taken by the t h i r d finger because of the preceding/
84. H a v i n g learned the p r i m e importance of correct use of the
t h u m b i n conjunct as w e l l as disjunct, one-part as w e l l as f u l l y set
passages, we can understand more clearly the a m o u n t of h a r m done
by methods of keyboard i n s t r u c t i o n appearing even i n this day
w h i c h , w r o n g i n many of their teachings, are p a r t i c u l a r l y i n error
o n this p o i n t . One author dispenses entirely w i t h the t h u m b ;
another is even more antagonistic t o w a r d his studentsnot only
must they make a l l of t h e i r fingers clamber i n d i s c r i m i n a t e l y and
o u t of sequence over the entire keyboard, they must be able to do
this o n any one key alone. T h e first develops p u p i l s whose fingers
stumble, miss, and i n t e r l o c k ; the other's t i r e needlessly and u n seasonably, for their hands must be c o n t i n u a l l y twisted and distorted i n order to allow the t h u m b to take black keys w i t h o u t r h y m e
or reason, even i n tonalities w i t h many accidentis. Because of this
d i s t o r t i o n the r e m a i n i n g fingers lose t h e i r n a t u r a l p o s i t i o n and
must be forced i n t o use. T h e r e is, consequently, n o chance to loosen
or relax the muscles, and the fingers s t i f f e n .
19

20

iBach enjoyed this advantage, according to Reichardt (Selbst Biographie,

Allgemeine Musikzeitung, Jhr. 16, 1814, No. 2 ) .
20 T h i s paragraph indicates that the inadequacy of the older fingering was far from
common knowledge. For example, as eminent a musicographer as Tohann Mattheson,
who prided himself on his keyboard ability, describes a fingering in his Kleine General-Bass-Schule
(1735) in which the ascending right hand employs principally 3
over 4, and the descending, 3 over 2, the thumb being unemployed. (Cf. Spitta, Bnch,
I I , p. 38). Couperin's fingering (Cf. Vol. I of his Oeuvres, Paris, 1933) is discussed later
in f 88.

FINGER1NG

h INC

85. As wc have alrcady lcarncd, leaps and progressions i n parts

are easier to execute than scales. Sustained tones w i l l prove even
simpler, for, i t being mandatory to h o l d them strictly for their assigned length, there is rarely more than one way to p e r f o r m t h e m .
Greater l a t i t u d e is allowed here than w o u l d be otherwise advisable;
the repeated finger, the t h u m b on black keys, and other expedients
w h i c h w i l l be discussed later are a l l freely employed. Since i t is n o t
easy to err u n d e r such conditions, the few examples of Figure 56
should suffice.

NC

7i

Figure 57

yfe /M ^V - 1

figure 56

E III

trntta

Vb-^-r

5 .

= = # = *

JH
M I

r t

86. Figure 57 is the first of a few exceptional examples. I n i t

leaps appear i n w h i c h the second finger (a), the t h i r d finger (b), and
the f o u r t h (c), cross the t h u m b . I n Figure 58 we see the t h u m b used
i n b r o k e n passages. N o t e that this finger is u n i f o r m l y f o l l o w e d by
the f o u r t h , and the second by the fifth.

87. One of the most i m p o r t a n t licenses is the omission of cert a i n fingers f r o m stepwise successions. T h i s technique is shown i n
Figure 59, where, i n preparation for the approaching leaps, the
omissions i n the asterisked example are clearly better t h a n the
settings m a r k e d w i t h d o u b l e asterisks. Omissions are f r e q u e n t l y
called for i n the bass. T h e n a t u r a l flexibility of the t h u m b makes
Example 1, where three fingers are o m i t t e d , easier than Example 2,
where only t w o are o m i t t e d .

FIN

Figure 59
i

s
i

G Eli

Because o u r "forerunners rarely used the t h u m b , i t got i n the

Henee, they o f t e n f o u n d that they liad too many fingers. Gradually,
i t began to play a more active role, b u t traces of the od m e t h o d
survived and many were n o t enterprising enough always to set the
t h u m b o n appropriate tones. T o d a y , despite improvements i n the
use of the fingers, we find, at times, that we have too few of t h e m .

FINGERING

INC.

i ; ^
a

B r

4 4 ,,r77 i

88. I n the Lessons, w h e n t w o numeris appar i n succession over

a single note, the first of the assigned fingers is n o t released u n t i l
the second has a f r i v e d , f o r such notes are to be sounded only once
unless an embellishment intervenes. T h e successions of F i g u r e 6o,
E x a m p l e a, as w e l l as the performance of certain embellishments
cali f o r this replacing of a finger. Occasionally i t is needed i n order
to sustain the tones of an arpeggio (b). T h e flexibility of the t h u m b
malees i t w e l l suited f o r replacement. Because i t is n o t easy to employ this device s k i l l f u l l y i t is correctly restricted to relatively long
notes a n d cases of necessity. T h i s precaution should be heeded i n
the use o f a l l expedients w h i c h , partly by t h e i r nature, partly
by their unusualness, are a n d r e m a i n difficult. Pupils should n o t be
p e r m i t t e d to employ them except as a last resort or to avoid an even
greater difficulty. C o u p e r i n , w h o is otherwise so sound, calis f o r
replacement too f r e q u e n t l y and casually. U n d o u b t e d l y , the t h u m b ' s
correct use was n o t f u l l y k n o w n i n his t i m e , as suggested by some
of his fingered examples i n w h i c h he replaces fingers instead of
using the t h u m b or the repeated finger, b o t h of w h i c h are easier.

Figure 60

.a,

89. Because of this, i t is permissible, o n occasion, to use a finger

twice i n succession even w h e n the notes change. T h i s oceurs most
frequently a n d h a p p i l y i n m o v i n g f r o m a black key to an adjacent
w h i t e one. S l u r r e d notes can be w e l l expressed i n this manner.
(Figure 61). A simple device, i t may be employed for other purposes,
too, a n d i n faster tempos t h a n those suitable f o r replacements a n d
finger repetitions o n a single tone. Observe that i t may be used to
p e r f o r m detached as w e l l as slurred notes. T h e first of these uses appears near the b e g i n n i n g of the F-sharp m i n o r Lesson and the seco n d i n Figure 56. Beyond this, we have already learned f r o m Paragraph 88 that the repeated finger is more n a t u r a l , even o n successive
t i e d notes (where there m i g h t be a choice of expedients), t h a n the replaced finger.
Figure 61
2 2

21

2 i Francois Couperin in L'Art de toucher le clavecn, Pars, 1717.

1
90. Tones repeated at a modrate speed are played b y a single
finger, b u t a l t e r n a t i n g fingers are employed i n fast r e p e t i t i o n s .
23

22 Sonata I V , third movement, bars 6-7. Cf. Pt. I , Introduction, Note 17.
23 Die Wiederholungen. Marpurg's terin is "die Schwrmer."

74

FINGERING

FIN

O n l y t w o fingers should be used at a time. The l i t t l e finger is the

poorest because its weakness retards the snap, a <|iiick retraction
w h i c h oceurs w h e n a finger leaves a key as rapidly as possible so that
the succeeding finger may play its tone d i s t i n c t l y . T h i s k i n d of passage is most easily performed o n the clavichord.
9 1 . A l t e r n a t i n g fingers may be employed to advantage i n slower
repetitions by p l a y i n g the final note of the series w i t h the finger
that leads best to the f o l l o w i n g notes, as i l l u s t r a t e d i n Figure 62.
Such situations oceur often i n the left h a n d .
Figure 62

92. W h e n , i n keys w i t h many accidentis, there appear passages

not so extended as to r e q u i r e the n o r m a l succession of fingers after
the t h u m b has t u r n e d , i t is possible to take, instead, the finger that
has preceded the t h u m b . I n so d o i n g , the hand maintains a single
position, thereby a v o i d i n g the a w k w a r d shift caused by a n o r m a l ,
Figure 63

2 '

G F.

RING

75

b u t i n this case r a p i d , crossing. This rule applies solely to those cases

i n w h i c h a single tone follows that taken by the t h u m b . Should t w o
tones f o l l o w , the fingers are to be played i n their usual order. B o t h
types appear i n Figure 63. Some employ this device i n passages w i t h
two succeeding tones, as indicated by the uppermost numeris of
the t w o final examples. W h i l e this is n o t incorrect, I m a i n t a i n that
i t is mandatory to h o l d to the n o r m a l order except i n those few
cases where a m o d i f i c a t i o n helps to elimnate awkwardness.
93. I n the Lessons there are t w o places where, contrary to the
stated r u l e , the l i t t l e finger is used i n a one-part passage before the
l i m i t of the range has been reached. B o t h are q u o t e d i n Figure 64.
T h e first is j u s t i f i e d by the modrate speed of the notes. A crossing
such as this is to be employed o n l y w h e n the f o u r t h , a longer finger,
takes a black key by a slight twist of the h a n d w h i l e the l i t t l e finger
plays one of the adjacent w h i t e keys. I t should be employed o n l y
once at a t i m e . T h e second illustrates an unavoidable contraction
of the h a n d , facilitated by the l o n g note; otherwise the fingering
w o u l d be incorrect. Because of the r a p i d tempo a replacement o n
/ w o u l d probably be more difficult than a contraction. I n execution
the h a n d s h o u l d be t u r n e d slightly to the r i g h t . I n the same piece
replacements must be employed o n short notes preceding an embellishment i n order to avoid a precarious l e a p . T h i s w i l l be more
clearly understood after the discussion of embellishments.
2 4

25

Figure 64

_2 1

=^

^=sA1

V 1

1
J

94. I n three or more p a r t compositions where each voice expresses an i n d i v i d u a l l i n e there arise situations i n w h i c h the hands
must be interchanged i n order to p e r f o r m the notes correctly, even
t h o u g h o n l y one h a n d should play t h e m according to the n o t a t i o n .
95. F i n a l l y , i n order to p r o v i d e an o p p o r t u n i t y for b o t h hands
to practice simultaneously, I have appended i n Figure 66 t w o exercises i n treacherous keys w i t h one accidental, consisting, i n the first,
of a stepwise figure and, i n the second, of a m i x t u r e of leaps and
2* Sonata I , first movement, bars 18-19, and Sonata I I , third movement, bars 22-23.
Cf. Pt. I , Introduction, Note 17.
25 Sonata I I , third movement, bars 12, 14, 30, 32. Cf. Note 24.

11 N a n { i N c,

FINGERIN

Figure 65

Jr

a. \

f
-

X
2

1 2

-z-^3

T7

-2-|

1 2

J,

if^lr

1,

1 2

-z-irjr

steps. B o t h cali for t u r n i n g and crossing as w e l l as the use of the

l i t t l e finger.

3 4 3

g J f

#*1

ri

\9=43
3

rT

i !

2 3

1 2 3 2

, .

Tf J ' J J

3 r
3

T
ul 7
; 1 1

r*

2 3 4 a

h - 2 p
1
1

- J -

96. A t those places i n the Lessons where the performer m i g h t

be i n d o u b t or even err i n choosing the h a n d assigned to the
notes, I have t u r n e d the stems u p w a r d to indicate the r i g h t h a n d
a n d d o w n w a r d to indicate the left. W h e n , o w i n g to l i m i t a t i o n s of
space, a few notes i n the i n n e r parts lack tails, t h e i r nature and
l e n g t h must be ascertained f r o m the notes struck w i t h t h e m i n
other i n n e r parts or the bass. Since I have t r i e d t h r o u g h o u t to
l i g h t e n the beginner's tasks and have seized every o p p o r t u n i t y to
indicate t h r o u g h the notes the h a n d assigned to them, n o one
should be dismayed to find the vales of certain tones and the
conduct of parts notated u n c o n v e n t i o n a l l y at times. Despite this,
the t r a i n e d p e r f o r m e r w i l l have n o difficulty i n f o l l o w i n g the path
of each voice and d e t e r m i n i n g the note lengths. T h e occasions for
these remarks can be f o u n d i n the D major and A-flat Lessons.
26

97. A m o n g the Lessons there is one i n w h i c h the hands cross.

I t h o u g h t i t wise to illustrate a n a t u r a l use of this k i n d of j u g g l i n g ,
even t h o u g h i t has n o t been very m u c h employed of late. I have
27

28 Sonata I V , second movement, and Sonata V I , second movement. Cf. Note 24.
27 Sonata VI, first movement. Cf. Note 24.

/' I N G E R I N G

7*

indicated the notes that helong l o each hand by means of clef

signs. I t is also possible to do this w i t h writt.cn directions. Gompositions can be f o u n d i n which the cornposer calis for needless crossing. T h e performer should n o t feel obliged to i n d u l g e i n such i m p o s t u r i n g b u t should seek instead a more n a t u r a l execution. Nevertheless, the technique is n o t to be discarded, for i t helps to make
the keyboard a more comprehensive i n s t r u m e n t and opens u p new
possibilities of expression. However, the crossing must be so devised that the passage is either unplayable any other way or playable
only w i t h a difnculty that causes an ugly g a r b l i n g or even a dism e m b e r i n g of the parts. For the rest, i t is a v a i n tempest that can
b l i n d o n l y the ignorant, for the i n i t i a t e d k n o w clearly that, considered by itself, there is no challenge i n i t save its unusualness, and
this is soon overeme. A n d yet, i t is everyone's experience that excellent and also difficult pieces have been w r i t t e n w h i c h employ
crossed hands.
28

98. Comments o n the fingering of embellishments must be

w i t h h e l d u n t i l the symbols have been explained i n the f o l l o w i n g
chapter. W h e n the fingering of f u l l y w r i t t e n embellishments is
o m i t t e d i t must be ascertained f r o m the finger assigned to the p r i n cipal tone.
99. F i n a l l y , I refer m y reader to the Lessons, where continuous
examples of a l l types of fingering w i l l be f o u n d .
28 I n Bibliothek der Schonen Wissenschaften, Vol. lo (1763), Pt. I, p. 58, there appears the following corament on this paragraph: "About forty or fifty years have
passed since such sorcery became known in Germany. A keyboardist by the ame of
Sandoni is said to have started it in a little piece, and a great host of lesser keyboardists
tried to achieve eminence by way of the same kind of fraud."
Today, Pietro Giuseppe Sandoni (ca. 1680-1750) is remembered chiefly as the husband of the famous singer, Francesca Cuzzoni. He reached London about 1726, after
a highly successful European career at the keyboard. Domenico Scarlatti, whose use
of this technique needs no extended comment, seems to have preceded him in L o n don by some six years.
As for Bach, he had excellent models in at least two works of his father; the
Gigue of the B-flat major Partita, and the C minor Fantasa. Indeed, Cari Philipp's
first composition (1731, Wotquenne No. 111) made use of the device. I n his autobiography, mention of the work, which he engraved himself, is joined with the following
remark: " . . . a natural and at that time much exploited form of m a g i a " T o this
Nohl (Musikerbriefe, 1873) adds, parenthetically, "introduced chiefly by Domenico
Scarlatti."

CHAPTER

TWO

EMBELLISHMENTS
GENERAL

1
O N E disputes the need for embellishments. T h i s is
evident f r o m the great numbers of t h e m everywhere to be
f o u n d . T h e y are, i n fact, indispensable. Consider t h e i r
m a n y uses: T h e y connect and e n l i v e n tones and i m p a r t stress a n d
accent; they make music pleasing and awaken o u r cise a t t e n t i o n .
Expression is heightened by t h e m ; let a piece be sad, j o y f u l , or
otherwise, and they w i l l l e n d a fitting assistance. Embellishments
p r o v i d e o p p o r t u n i t i e s for fine performance as w e l l as m u c h of its
subject matter. T h e y i m p r o v e mediocre compositions. W i t h o u t
t h e m the best melody is empty and ineffective, the clearest conten
clouded.

2. I n view of t h e i r many commendable services, i t is u n f o r t u nate that there are also poor embellishments and that good ones are
sometimes used too f r e q u e n t l y a n d ineptly.
3. Because of this, i t has always been better for composers to
specify the proper embellishments unmistakably, instead of leavi n g t h e i r selection to the w h i m s of tasteless performers.
4. I n justice to the French i t must be said that they ntate t h e i r
ornaments w i t h painstaking aecuracy. A n d so do the masters of the
keyboard i n Germany, b u t w i t h o u t embellishing to excess. W h o
knows b u t that o u r m o d e r a t i o n w i t h respect to b o t h the n u m b e r
and kinds of ornaments is the influence w h i c h has led the French
to abandon t h e i r earlier practice of decorating almost every note,
to the d e t r i m e n t of clarity and noble simplicity?
5. I n summary: G o o d embellishments must be distinguished
f r o m bad, the good must be correctly p e r f o r m e d , and i n t r o d u c e d
moderately and
fittingly.
79

So

E M H E l. 1.1 S 11 M E N r s

EM

fi. Embellislimcnts may be d i v i d e d i n t o two groups: i n the first

are those which are indicated by conventional signs or a lew small
notes; i n the second are those w h i c h lack signs and consist of many
short notes.
7. I shall treat the latter g r o u p only briefly i n connection w i t h
the performance of fermate. T h e r e are several reasons for this. For
one t h i n g , their use is governed chiefly by taste; as a result, they are
too variable to classify. F u r t h e r , i n keyboard music they are usually
w r i t t e n o u t . I n any event, there is n o real need for t h e m , thanks
to the adequacy of the others. I shall discuss i n d e t a i l o n l y those
i n the first g r o u p , for most of t h e m have a l o n g and cise association
w i t h the keyboard and w i l l u n d o u b t e d l y always r e m a i n i n favor. T o
the accepted embellishments I have added a few new ones. A l l w i l l
be explained and t h e i r proper contexts specified insofar as i t is possible to do so. I shall fix t h e i r best f i n g e r i n g w h e n necessary and i n dicate t h e i r correct execution. I l l u s t r a t i o n s w i l l elucdate those
points that cannot be completely p u t i n t o words. I shall p o i n t o u t
certain incorrect or, at least, unclear signs so that they may be d i f ferentiated f r o t n the correct, and at the same t i m e I shall criticize
poor ornaments. F i n a l l y I shall refer m y readers to the Lessons and
hope t h r o u g h o u t to remove the false assumption, occasionally encountered, of the need for profuse keyboard o r n a m e n t a t i o n .
1

8. Nevertheless, those w h o are adept at i t may combine the

more elabrate embellishments w i t h ours. However, care must be
taken to use t h e m sparingly, at the correct places, and w i t h o u t dist u r b i n g the a f f e c t of a piece. I t is understood, for example, that
s

1 It was customary for the performer in earlier times to add his own embellishments
and elaborations freely. T h e practice was changing about 1750 to the modern method,
whereby the composer specifies every last detail and the performer, hopefully speaking, follows orders. Indicative of the widespread nature of the earlier practice is Bach's
Foreword to T w o Trios (Wotquenne No. 161), the first of which is programmatic.
He was anxious to have the first T r i o performed as written and in order to attain
this end (which would be taken for granted today) wrote: " I t would be best to play
the first T r i o as notated, without the addition of free ornaments." (Cf. Hans Mersmann in Bach Jahrbuch, 1917.) J . S. Bach won the censure of J . A. Scheibe because
of his practice of writing out all detail (Cf. The Bach Reader, Norton, New York,
1945, pp. 237 ff.). It should be added that Bach was a master at introducing ornaments of all kinds into other composers' works. Philipp Emanuel's position here is
undoubtedly influenced by his father's stand. T h e opening sentence of 8 suggests
indirectly a defensive reason for the stand taken by both. Evidently, not enough
performers were "adept at it."
2 T h a t is, there is no real need for them as free embellishments interpolated by
the performer.
3 T h e premise of the theory of the afleas was that music is capable of being more

BELL1

S H MEN

T S

81

the portrayal of s i m p l i c i t y or sadness suffers fewer ornaments than

other emotions. H e w h o observes such principies w i l l be j u d g e d perfect, for he w i l l k n o w how to pass s k i l l f u l l y f r o m the singing style
to the s t a r t l i n g and fiery ( i n w h i c h instruments surpass the voice)
and w i t h his constant changing rouse and h o l d the listener's attent i o n . W i t h these ornaments, the difference between voice and i n s t r u m e n t can be unhesitatingly e x p l o i t e d . For the rest, as l o n g as
embellishments are a p p l i e d w i t h discretion n o one need pause to
decide whether a played passage can or cannot be sung.
9. A b o v e a l l things, a p r o d i g a l use of embellishments must be
avoided." Regard t h e m as spices w h i c h may r u i n the best dish or
gewgaws w h i c h may deface the most perfect b u i l d i n g . Notes of n o
great m o m e n t and those sufficiently b r i l l i a n t by themselves should
r e m a i n free of t h e m , for embellishments serve o n l y to increase the
w e i g h t and i m p o r t of notes and to differentiate t h e m f r o m others.
Otherwise, I w o u l d c o m m i t the same error as orators w h o t r y to
place an impressive accent o n every w o r d ; everything w o u l d be
alike and consequently unclear.
than a mere pattern of sounds but is, rather, expressive of many passions. It was
therefore considered insufficient for a performer to play a piece solely in a technically
correct manner. He must "rouse and still the passions" by portraying the proper affect. All writers of the Berln School, Quantz, Marpurg, Sulzer, and Bach, were preoceupied with the theory of the arfeets. T h e term and discussion of its meaning recur
throughout the Essay. Cf. Paul Lang, Music in Western Civilization (Norton, New
York, 1941, pp. 434 ff.). For the rest, the following excerpt from Marpurg (Der Critischer Musicus an der Spree, Sept. 2, 1749) will serve for general purposes of orientation: " T h e rapidity with which the emotions change is common knowledge, for they
are nothing but motion and restlessness. All musical expression has as its basis an
affect or feeling. A philosopher who explains or demonstrates seeks to bring light to
our understanding, to bring clarity and order to it. But the orator, poet, musician
seek more to inflame than enlighten. With the philosopher there are combustible
materials which merely glow or give off a modest restrained warmth. Here, however,
there is but the distilled essence of this material, the finest of it, which gives off thousands of the most beautiful flames, but always with great speed, often with violence.
T h e musician must therefore play a thousand different roles; he must assume a thousand characters as dictated by the composer. T o what unusual undertakings the passions lead us! He who is fortnate, in any respect, to capture the enthusiasm that
makes great people of poets, orators, artists will know how precipitately and variously our soul reaets when it is abandoned to the emotions. A musician must therefore possess the greatest sensitivity and the happiest powers of divination to execute
correctly every piece that is placed before h i m . "
* Marpurg wrote in Der Critischer Musicus an der Spree (1750-51): " A special distinction of Berln music is that it makes very sparing use of manners and embellishments; but those that are used are the more select and the more finely and clearly
performed. T h e performances of the Grauns, Quantz, Benda, Bach, etc., are never
charactrized by masses of embellishments. Impressive, rhetorical, and moving qualities spring from entirely different things, which do not crate as much stir, but
touch the heart the more directly."

8a

E M i El.

1.1 Sil

M EN

T S

EM

10. W e shall sce presently that many passages allow for more
than one k i n d of embellishment. I n such cases, the art of v a r i a t i o n
may be used to advantage; introduce first a caressing ornament,
then a b r i l l i a n t one, or for a change, i f the passage permits, play the
notes d i r e c t l y as w r i t t e n b u t always i n furtherance of the true affect
and i n accordance w i t h the rules of good performance w h i c h w i l l be
treated later.
11.
I t is difficult to prescribe the correct context for every embellishment, for a l l composers are free to introduce t h e i r favorites
where they w i l l , so l o n g as good taste is n o t thereby assailed. Suffice
i t i f we instruct o u r reader t h r o u g h a few well-established precepts
and examples. A t least he w i l l learn that the nature of a passage can
n a r r o w his choice of ornaments. T h u s , w h i l e the performer w i l l
have n o need for concern i n those compositions where a l l embellishments are specified, where few or none are indicated he w i l l
k n o w h o w a n d where to insert t h e m i n the customary manner.
12. Because I have yet to find a f o r e r u n n e r w h o m i g h t have
b r o k e n a p a t h for me, no'one should criticize me for h o l d i n g that,
despite certain established cases, i t is conceivable that exceptions
can arise.
5

13. I n order to master this m a t e r i a l w i t h its many m i n u t i a e and

apply i t i n t e l l i g e n t l y , the ear must be t r a i n e d t h r o u g h constant
listening to good music. A b o v e a l l , to understand many things more
clearly, the p e r f o r m e r must possess a knowledge of t h o r o u g h bass.
I t is a matter of experience that those w h o are n o t w e l l g r o u n d e d i n
the study of h a r m o n y f u m b l e i n darkness w h e n they use embellishments and must thank their good f o r t u n e rather than insight w h e n
they are successful. W i t h this i n m i n d I shall add a bass to the i l lustrations whenever i t is necessary.
14. Singers and instrumentalists other than keyboardists w h o
wish to p e r f o r m w e l l need most of o u r short embellishments just as
m u c h as we do. However, o u r ways are m u c h more o r d e r l y t h a n
theirs, for keyboardists have given embellishments specific signs the
more exactly to indicate the detailed performance of t h e i r compositions.
15.

Because others have n o t shown such commendable

fore-

T h i s statement refers only to the problem of defining the contexts suited to each
ornament at the keyboard and not, as Dannreuther (Musical Ornamentation, Part I I ,
p. 5, footnote) believed, to the general description of ornaments and their signs.
See, for example, C h . I I , " T h e Appoggiatura," f 16.
5

BE L

L I S II MEN

T S

sight, b u t have t r i e d , rather, to indicate everything t h r o u g h only a

few signs, the study of o r n a m e n t a t i o n is m u c h more t a x i n g for t h e m
than i t is for the keyboardist. T h e i r signs have g r o w n ambiguous or,
indeed, incorrect, a c o n d i t i o n w h i c h even today causes many i m proprieties i n performance. For example, the m o r d e n t , one of the
most essential and widely used embellishments, is k n o w n by its
sign to few outside of keyboardists. I k n o w of one place i n a cert a i n piece w h i c h as a consequence is often r u i n e d . T h e passage, i f
i t is n o t to sound i n s i p i d , requires a l o n g m o r d e n t w h i c h , because
of the nature of the passage, n o one c o u l d presume, unless i t were
specified by its symbol. A n d yet, even i n the presence of its indicat i o n ( k n o w n o n l y to keyboardists), a t r i l l is p e r f o r m e d , o w i n g to
the a m b i g u i t y caused by a general lack of signs. W e shall realize
later i n s t u d y i n g the great difference between the t w o ornaments
how a w k w a r d i t is to substitute one for the other.
T

16. T h e French are especially careful i n setting the signs of t h e i r

embellishments. B u t u n f o r t u n a t e l y we have so far removed ourselves f r o m t h e i r music and t h e i r fine style of p l a y i n g that the exact
meaning of t h e i r embellishments is vanishing to the p o i n t where
signs once w e l l k n o w n are becoming unrecognizable even to keyboardists.
17. T h e tones of an embellishment adjust themselves to the accidentis of the key signature. Beyond this, we shall soon learn that
at times, preceding and succeeding tones or, m o r e frequently, m o d u lations r e q u i r e a d d i t i o n a l alterations. T h e t r a i n e d ear recognizes
such contexts immediately.

18. However, I have f o u n d i t advisable to f o l l o w the practice of

a d d i n g accidental signs to the symbols of ornaments i n order to assist the p e r f o r m e r . I n the Lessons, they w i l l be m e t singly and i n
pairs wherever they are r e q u i r e d .
19. A l l ornaments stand i n p r o p o r t i o n e d relationship to the
length of the p r i n c i p a l note, the tempo, and the affect of a piece.
I n those cases where a variety of embellishments is used and the perf o r m e r is n o t too restricted by the affect, note that the more tones
contained i n an ornament, the longer the p r i n c i p a l note must be, regardless of whether the source of this length is the note itself or the
tempo. T h e b r i l l i a n c e of an embellishment must n o t be d u l l e d by
excessive space f o l l o w i n g its execution. O n the contrary, the performer must avoid a too h u r r i e d performance, w h i c h blurs certain

84

E M li E L LI S II M EN
M li E L LI S H M E N T S

ornaments. T h i s is causee! moslly by tlie i n t r o d u c t i o n of embcllishments c o n i a i n i n g many tones and the excessive einbellisliment o
r a p i d notes.
Figure 67
Adagio
Andante
A l l e g r e t t o to Presto
7

20. Nevertheless, we shall soon learn that i t is permissible to i n troduce an embellishment w h i c h does n o t completely fill o u t the
w r i t t e n length of a long note. However, the last tone of the embellishment must n o t be released u n t i l the f o l l o w i n g note arrives, for
the p r i m a r y a i m of a l l embellishments is to connect notes.
21.
Henee, embellishments are better suited to slow or m o d rate t h a n to r a p i d tempos, a n d to l o n g rather t h a n short notes.
Observe especially that embellishments are best a p p l i e d to those
places where a melody is t a k i n g shape, as i t were, or where its pard a l , i f n o t complete, meaning or sense has been revealed. Henee,
w i t h regard to the latter case, they are f o u n d chiefly at half or f u l l
closes, caesurae, and jermate.
8

22. I n e x p l a i n i n g symbols and small notes, I shall w r i t e o u t the

correct lengths of the notes expressed by t h e m . Small notes have
been notated i n t h e i r real vales i n the Lessons.
23. A l l embellishments notated i n small notes p e r t a i n to the
f o l l o w i n g tone. T h e r e f o r e , w h i l e the preceding tone is never shortened, the f o l l o w i n g tone loses as m u c h of its length as the small notes
take f r o m i t . T h i s observation grows i n importance the more i t is
neglected a n d the less I was able to avoid a separation of such notes
f r o m their p r i n c i p a l tones i n the Lessons, o w i n g to the space taken
u p by the mass of fingering numeris, symbols of ornaments, and expression marks.
24. A c c o r d i n g to this r u l e the small notes rather t h a n the p r i n cipal tone are struck w i t h the bass and the other parts. T h e perf o r m e r should make t h e m glide i n t o the f o l l o w i n g tone. I t is w r o n g
i From the ed. of 1787.
* Remainder of paragraph from ed. of

1787.

T S

85

to j u m p roughly on the p r i n c i p a l tone, for this causes f u r t h e r awkwardness i n the i n t r o d u c t i o n as well as the execution of embellishments. I t m i g h t seem superfluous to repeat that the other voices
i n c l u d i n g the bass must be struck w i t h the i n i t i a l tone of an embellishment. Yet as often as this r u l e is cited, so o f t e n is i t v i o l a t e d .
0

Figure

68

10

25. Because o u r present taste, to w h i c h I t a l i a n bel canto has

c o n t r i b u t e d greatly, demands more than French ornaments alone, I
have had to accumulate the embellishments of several countries. T o
these I have added a few new ones. I blieve that that style of performance is the best, regardless of the i h s t r u m e n t , w h i c h a r t f u l l y
combines the corree tness a n d b r i l l i a n c e of French ornaments w i t h
the suavity of I t a l i a n singing. Germans are i n a good position to effect such a unin so long as they r e m a i n free of prejudices.
26. T h e r e f o r e i t may well be that some, predisposed to only one
taste, w i l l n o t be satisfied w i t h m y choice of ornaments. However, i t
is m y belief that i n music no one can offer reasonable c r i t i c i s m w h o
has n o t listened to a l l styles and cannot select the best f r o m each. I
agree w i t h a certain great rrian's opinin that, a l t h o u g h one taste
may be better t h a n another, each contains something good and none
is so perfect that i t w i l l n o t endure additions. I t is t h r o u g h such additions and refinements that we have progressed this far and w i l l
advance even f u r t h e r , b u t , certainly, n o t t h r o u g h a d d i c t i o n and restriction to o n l y one style. E v e r y t h i n g good must be p u t to use regardless of its origins.
1 1

27. Embellishments and their execution f o r m a large part of

good taste. T h e r e f o r e the performer must n o t be inconstant a n d ac9 Remainder of paragraph from ed. of 1787.
From the ed. of 1787.
11 His father's. Marpurg wrote in Der Critische Musicus an der Spree, Sept. 2, 1749,
" I n every type of music, in the music of every country there are bad and also good
things. T h i s is the opinin of od Bach of Leipzig who certainly counts for something
in music."
1 0

E M li El. 1.1 S II M E N T S

,V6

cept u n c r i t i c a l l y cvcry ncw r a n d o m ornament. or should lie be so

predisposed toward himself and his own taste that he is obstinatcly
u n w i l l i n g to accept anything strange. Certainly severe tests should
precede the acceptance of the new, for i t is possible that u n n a t u r a l
novelties m i g h t i n t i m e make good taste as rare as s k i l l . However,
w h i l e i t is wise n o t to be the first, one should also n o t be the last
to acknowledge new ornaments i n order not to f a l l o u t of style. Do
not t u r n against t h e m because they sound unattractive at first. T h e
new, as engaging as i t may be at times, very often repels us. T h i s
may indicate the presence of merits that w i l l prove more longlived than those qualities w h i c h at first are entirely too pleasant. As a r u l e we soon t i r e of such charms and they end by revolti n g us.
28; Just because most of the illustrations of embellishments are
w r i t t e n for the r i g h t hand, i t must n o t be assumed that I f o r b i d
these adornments to the left. I n fact I advise strongly that a l l ornaments be practiced by b o t h hands, the more so because this develops
a general f a c i l i t y and dexterity. W e shall see later that certain ornaments are frequently assigned to the bass. Moreover, a l l i m i t a t i o n s
must be exact to the smallest d e t a i l . Henee, the left h a n d should
practice ornaments u n t i l i t can i m i t a t e skillf u l l y , for ornaments that
lose their charm t h r o u g h poor execution are better o m i t t e d entirely.
29. I t w i l l be gathered f r o m the f o l l o w i n g pages that the explanations of ornaments w h i c h have been added to the second p a r t
of m y Sonatas are w r o n g . T h e publisher presumed to p r i n t these
under m y ame w i t h o u t my consent or knowledge. I am as guiltless
here as I was i n the case of the p u b l i c a t i o n of VI Sonates
nouveaux
per Cmbalo, 1751, listed under m y first a n d f a m i l y ames o n page
8 of this year's e d i t i o n of Lotter's Catalogus aller
musicalischen
Bcher.
I have n o t yet examined these Sonatas, b u t I am q u i t e
certain that either they are n o t m i n e at a l l or, at the most, are od,
badly copied things, stolen and then published, a very usual occurrence.
1 2

13

12 T h e Wrttemberg Sonatas (Nagels Archiv Nos. 21, 22) as published by J . Huffner

in Nuremberg, 1744.
i s An od firm of music printers founded about 1726 in Augsburg by Johann Jakob
Lotter. T h e works mentioned may be the six Sonatinas listed by Wotquenne (No. 64)
as compsed in Leipzig in 1734 and revised in Berln in 1744. O r , as suggested by
Bach, they may be works falsely ascribed to him. Johann Christian Bach, along with
many other well-known composers of the time, was a victim of this practice.

EM

li El.

1.1 S II M E N I S

T H E APPOGGIATURA

1. Appoggiaturas are among the most essential embellishments.

T h e y enhance h a r m o n y as w e l l as melody. T h e y heighten the attractiveness of the latter by j o i n i n g notes smoothly together a n d ,
i n the case of notes w h i c h m i g h t prove disagreeable because of their
length, by shortening t h e m w h i l e filling the ear w i t h sound. A t the
same t i m e they p r o l o n g others by occasionally repeating a preceding
tone, and musical experience attests to the agreeableness of w e l l contrived repetitions. Appoggiaturas m o d i f y chords w h i c h w o u l d
be too simple w i t h o u t t h e m . A l l syncopations and dissonances can
be traced back to t h e m . W h a t w o u l d harmony be w i t h o u t these elements?
2. Appoggiaturas are sometimes w r i t t e n i n large n o t a t i o n and
given a specified length i n a bar. A t other times they appear i n small
notation, and the large notes before w h i c h they stand r e t a i n t h e i r
length visually a l t h o u g h i n performance they always lose some of
i t to the ornament.
3. T h e l i t t l e that is n o t e w o r t h y about the former w i l l be reserved for the conclusin of this section. B o t h types ascend as w e l l
as descend to the p r i n c i p a l note.
4. I n execution some appoggiaturas vary i n l e n g t h ; others are
always r a p i d .
5. Because of their v a r i a b i l i t y , such appoggiaturas have been
notated of late i n their real length (Figure 69). P r i o r to this a l l
were w r i t t e n as eighths. A t that t i m e , appoggiaturas as diverse as
ours were n o t yet i n use. T o d a y , we c o u l d n o t do w i t h o u t the notat i o n of their real vales, for the rules covering their l e n g t h i n performance are insuflcient to cover a l l cases, since a l l types appear
before every k i n d of note.
2

6. W e can readily see i n the examples of Figure 69 that at times

appoggiaturas repeat the preceding note (a), at times they do n o t
(b), and that the f o l l o w i n g note may lie a step above or below, or i t
may be separated f r o m the ornament by a leap.
1 Der Vorschlag.
2 T h i s practice was generally adopted in the latter half of the eighteenth century.
T h e alternative (cf., par ex., Haydn) was to write the ornament in large notation,
as described in f 2. Henee the rules of length as discussed later in f u refer to a
dying practice, not always applicable even to the music of J . S. Bach. Cf. H . Schenker,
Ein Beitrag zur Ornamentik, pp. 26 ff.

ss

l.M

B E 1. 1, 1 S II M EN

T S

EMBELLISHMENTS

89

Figure 70

i
9

Figute

r r : ' us

JJ

11-

Ji 1

pnrr " y

J11

1 l^Ul.

/
71

11?

txx/r
ni

J^JJ

1
1

Ai

fo-H
7. W i t h regard to execution we learn f r o m this figure that appoggiaturas are louder t h a n the f o l l o w i n g tone, i n c l u d i n g any addit i o n a l embellishment, and that they are j o i n e d to i t i n the absence
as w e l l as the presence of a slur. B o t h of these points are i n accord
w i t h the purpose of appoggiaturas, w h i c h is to connect notes. T h e y
must be h e l d u n t i l released by the f o l l o w i n g tone so that b o t h are
smoothly j o i n e d . A n undecorated, l i g h t tone w h i c h follows an appoggiatura is called the relase.
8. Because the sign of the appoggiatura is universally k n o w n
(like that of the t r i l l ) i t is one of the few ornaments whose i n t r o d u c t i o n is usually notated. Nevertheless, since one cannot always depend o n this, i t is necessary to fix the proper contexts of the variable
appoggiatura, insofar as i t is possible to do so.
9. I n a d d i t i o n to the observations of Paragraph 6, the variable
appoggiatura i n d u p l e t i m e appears c o m m o n l y o n either the d o w n
beat (Figure 70, Example a) or the upbeat (b); b u t i n t r i p l e t i m e
only o n the downbeat (Figure 71) and always before a relatively
l o n g note. F u r t h e r , i t is f o u n d before cadential t r i l l s (Figure 72,
Example a); before half cadenees (b), caesurae (c), fermate (d), a n d
final tones w i t h (e) or w i t h o u t (/) a preceding t r i l l . W e learn f r o m
Example e that the ascending appoggiatura after a t r i l l is better t h a n
the descending; henee, the i l l u s t r a t i o n under g is poor. Slow d o t t e d
notes also take the variable appoggiatura (h). W h e n such notes have
tails, the tempo must be a suitable one.

rr

-i

i^.

I>JJ

fc)

fa

irtnr*

T*

J>J

J-

Jr r
r
JJ

iu

11

J>J 1

h la l ' l -ti

90

E MBELLISHM

EN TS

EM

10. The ascending variable appoggiatura is difficult to use except when it repeats the preceding tone; but the descending kind is
met i n all contexts.
11. The usual rule of duration for appoggiaturas is that they
take from a following tone of duple length one-half of its valu
(Figure 73, Example a), and two-thirds from one of triple length
(b). I n addition the examples of Figure 74 and their executions
should be carefully studied. Appoggiaturas which depart from this
rule of duration should be written as large notes. Errors i n execution, which distort melodies and often crate false chords, are
caused by inattentiveness and occasionally by distrust of the copyist's accuracy, for i n earlier times all appoggiaturas were notated invariably as eighths.

b.

7 3

J 1

- 4 _

F i

u r e

sr=
5

7 4

1p~ m

pj==q

0Lj!

Figure 76

Y.

inb^t

S v

k' 'Ul I

# = 1

(i ^

~<

0 r *

r r

m g = 5

PH=
1

r p

II
1

9'

13. I t is wholly natural that the unvariable short appoggiatura

should appear most frequently before quick notes (Figure 76, Example a). I t carries one, two, three, or more tails and is played so
rapidly that the following note loses scarcely any of its length. I t
also appears before repeated (b) as well as unrepeated (c) long notes.
Further, i t is found in caesurae before a rapid note (d), and insyncopated (e), tied (/), and slurred passages (g). I n all such cases, the
character of the notes remains unchanged. Example h with an
ascending appoggiatura is better when the ornament is played as
an eighth. For the rest, the short appoggiatura remains short even
when the examples are played slowly.

NTS

Figure 75

F l 8 l

HE L L I SU M E

11

3 In Chapter VI, "Appoggiaturas," ^4, Bach adds, "The shortest of these is never
more rapid than an eighth note in an allegretto."
i Remainder of paragraph from ed. of 1787.

fe!
d.

r -

The notation of the short appoggiatura as a small eighth note with a diagonal
stroke through the tan was not used by Bach, or indeed by the Viennese Classical
School. However, it did make its appearance in early nineteenth-century editions of
their works, notably those of Mozart published by Andr. While the older notation
gave rise to ambiguities (where variable and short appoggiaturas have the same
notation) the later notation, apart from those cases where editors used it indiscriminately for both the long and the short ornament, has the disadvantage of dulling
the performer's sensitivity to subtle variations of length in the short appoggiatura,
as described in the following ^| 14. A few short appoggiaturas, notated in the later
manner, appear in J . S. Bach's Werke, 36.4, pp. 10-11, 14. They are not authentic.
5

12. The examples under Figure 75 are frequent occurrences.

Their notation is not the most correct, since i n performance the
rests are filled i n . Dotted or longer notes should be written instead.

pr *f

||

b.
n

92

E MfELLISHMEN

TS

E M Ii E I. . SUMEN

T S

with all ornaments, the lempo must be a suitable one, for an excessive speed does not allow for embellishment. The asterisked example is intended to show us that an appoggiatura does not sound
well before a long note preceded by a much shorter one. Later, we
shall learn that an ornament which is better at filling out may be
introduced here.
10

C^!.j^r r'.'
|

,ntrp

^ffl

14. When these appoggiaturas fill i n the interval of a third, they

also are played quickly. However, i n an Adagio their expression
is more tender when (Example a, Figure 77) they are played as the
first eighth of a triplet rather than as sixteenths. The accurate divisin of triplets can be learned from Example b. For various reasons
the resolving tone of a melody must often be quitted abruptly.
When such a tone is an appoggiatura, it too must be played rapidly
(Example c). I n this example, the appoggiatura, which is present
only to complete the run, must be very short so that the principal
tone, c, which is the cause of the free execution and is therefore always especially important, loses little or nothing of its valu. Appoggiaturas before triplets must also be played quickly so that the
rhythm remains clear (d) and distinguishable from that of Example
e. When the appoggiatura forms an octave with the bass i t is played
rapidly because of the emptiness of the interval (/). On the other
hand, it is often prolonged when it forms a diminished octave (g).
I t remains short when it is substituted for a cadential t r i l l (h).
15. When a melody ascends a second and then returns_to either
a large note (Figure 78) or another appoggiatura (a), the middle tone
may be readily decorated with a short appoggiatura. I n Figure 79
there are many such passages containing notes of various lengths i n
duple and triple rhythms. We learn from Example a that a long appoggiatura may also be used here. I t is taken for granted that the
phrasing is normally legato i n such a context, since detached notes
must always be more simply performed and also because appoggiaturas are invariably joined to the following tone. Further, as
6

Figure 77

d.

z r r*

if

ir

This sentence appears as a footnote in the ed. of 1787.

7 Or, as Bach explains manipulated progressions in Part I I , the underlying progression is from the opening d" to the quarter note c", and the outline of the upper
voice is the two-part progression: d"c"
g'-f'-c.
This sentence from the ed. of 1787.

f rf

Figure 78

gjj
11

Figure 79

J |>JP1_LUS

U,
S

/T
J

9 Example h from the ed. of 1787.

1 0

See Figure 153, Example c.

s_

h.

cj.f

E M li E L L I S II M E N T S

EM

W-0

^4

S 11 M EN

TS

95

80, Example a). A t times the length is determined by the accompaniment, as i n Example b, where, if the appoggiaturas are played
as full quarter notes, the fifths struck against the bass w i l l sound
ugly. I n Example c, i f the appoggiatura is held beyond its written
length, it w i l l crate open fifths. Again, i n Figure 69, the appoggiatura i n the asterisked example must not be prolonged or the
seventh w i l l sound too harsh.

| , ^ i y ijp
r

BELLl

Figure 80
- a.

ir

>' |l>J

4" T
l;

Jh

<2.

m i'[ i . rmr rr uf i p
g

17. Henee, as with all embellishments, the introduction of an

appoggiatura must not corrupt the purity of voice leading. For this
reason the examples of Figure 81 had better not be put into practice. Thus it is best to ntate all appoggiaturas i n their real length.
Figure 81

I i I I i I LU l 'l 11
ii

1 "i

Ji

18. A profusin of appoggiaturas with their releases is particularly good i n affettuoso passages since the releases usually expire
pianissimo (Figure 82). I n other cases, however, they make a
11

>

11

4.

fl

1L
16. W i t h regard to the rule covering the length of appoggiaturas, there are a few situations i n which the ornament must be
extended beyond its normal length because of the affect. Thus it
may take up more than half the valu of the following tone (Figure

Figure 82

&

11 Cf. J. S. Bach, F rainor Prelude, Bk I I , W T C , and the Toccata of the E minor

Partita.

EM

E M li E I. L l S 11 M E N T S

melody insipid unlcss they are followed by livelicr embellishments

or are themselves supplemented by additional ornaments
19. When an appoggiatura is decorated, the following tone is
best performed plainly. Its simplicity will be happily complemented
by its usual piano execution. A n undecorated appoggiatura, however, leads well to a succeeding embellishment. This latter is i l lustrated i n Figure 83, Example a, and the former i n Example b.
20. The decorating of appoggiaturas leads us to other embellishments which will be explained later. Because these are often
written as small notes, it is better to write the appoggiatura i n such
cases as a large note with its length clearly notated (Figure 83, Example c). I n slow pieces the appoggiatura as well as the following
tone may be embellished on occasion (d).
21. However, appoggiaturas are often written i n large notation as a means of indicating that neither they or the following
tones are to be decorated (<?).
Figure 83

d.

ir

h.

a*

Figure 84

97

1 2

v=4r=
P f>

fe

22. Although the note following an appoggiatura relinquishes

part of its length, it does not lose any of its own embellishments
(Figure 84). On the other hand, embellishments which belong to
the appoggiatura should not be placed over the following tone.
They must always be written directly above the note to which they
pertain. I f they are to be performed between the appoggiatura and
the following tone the symbol must be placed between the two
notes (Figure 85).

TS

23. Descending appoggiaturas written i n large notation may be

decorated by another appoggiatura, long or short, when they repeat the preceding tone (Figure 86, Example a), or when they do
not lead into closing tones (Example b is therefore wrong). Ascending appoggiaturas i n large notation do not take another appoggiatura either from above or from below. (c). They may, however,
be followed by one (d).

r.

HE L L I Sfl MEN

11 J> ? J F ^ ^ = - -+
:

p" r r r r

'--

-^0-4

24. A few additional incorrect uses of the appoggiatura remain

to be considered. I t is wrong to place a descending appoggiatura
before the final tone of a cadenee when the final tone is preceded by
a t r i l l without an appoggiatura (Figure 72, Example g). However, a
t r i l l which is graced by this ornament may be followed by a
similarly graced final tone regardless of whether the final tone
stands lower (Figure 87, Example a) or higher (b) than the t r i l l .
Another error is the separation of an appoggiatura from its following tone either because the ornament is prematurely quitted or because it has usurped a portion of the preceding note's valu (Figure
88, Example a).
Figure 87

a.

ir

r 1?

ir

5te

12 Quantz and Bach are in complete disagreement over the performance of this

Figure 85
a

PPgg'

a t u r a

- The former writes (Versuch einer Anweisung

die Fite traversiere zu

spielen), "Two appoggiaturas are often found before a note, the first written as a
small note and the second as a large, measured one. They appear at caesurae. The
small note is short and placed on the preceding divided beat." Bach, of course, wanted
the small note to be played on the beat of the large appoggiatura. This difference

EMBELLISHMENTS

EM f EL 1.1 SU M EN

25? This latter dislocation is the origin of the repulsive unaccented appoggiatura, so extraordinarily popular, which is reserved,
unfortunately, for the most legato passages, such as those i n Figure
88, Example b. I f appoggiaturas should or must be used i n such
cases, the asterisked executions are more tolerable. Henee, the
remedy for unaccented appoggiaturas is to shift them ahead to the
next accent. Yet a good and frequent use of the unaccented appoggiatura is illustrated i n Figure 89, Example a. However, the last bar
is more fashionable than harmonious. Figure 89, Example b, is
to be avoided. I t illustrates those cases i n which a very short descending appoggiatura is inserted between an ascending one and its principal tone at a cadenee or i n a melody which does not descend immediately afterward.

TS

Figure 88

13

14

fiares up when Bach (f 25) calis the weak-beat appoggiatura the "hasslicher Nachschlag." Cf. note 13.
is Nachschlag. Bach's high-handed treatment of this ornament has brought censure from many sources. For example, the Bibliothek der schnen Wissenschaften, in
reviewing the Essay in 1763, wrote: "Marpurg [Anleitung zum Klavierspielen] gives
us better information on the Accent, dividing it into Vorschlag and Nachschlag, which
latter, Bach has actually neglected, treating it only superficially here and there." Later,
Dannreuther and especially Dolmetsch grew exercised over Bach's feelings and
pointed out, not without indignation, that this ornament had an honorable past
(cf. Dolmetsch, The Interpretation

of the Music of the XVIIth

and XVIIIth

Centuries,

pp. 148 ff.). Quantz (op. cit.) gives consideration to the Nachschlag or, as he calis it,
the durchgehende Vorschlag, an unaccented appoggiatura that filis in the interval
of a third in the manner of Figure 87, Example a, here. In France, where it had wide
use, it was known as "couler les tierces." Concerning its performance as a Vorschlag
(cf. ^1 14 here) Quantz writes: "This would be contrary to the French style from which
the ornament springs and the intentions of its authors, who won almost universal
acclaim for this device." Bach and Quantz are in obvious disagreement. The strong
adjectives used by the former in describing it here and in the sections on The Trill
(f 21), "The Turn" (f 29), also in Ch. III, f 27, indcate that he must have winced
more than once as court accompanist.
However, it must be pointed out in Bach's defense that he was not, as generally
believed, unconditionally opposed to the ornament. See, for example, Figure 89, example b, and, in the Lessons, Sonata IV, second movement, and third movement,
bars 14, 19, 46. A casual examination of the collections for Connoisseurs and Amateurs will reveal others. He always writes it in large notation except when it is incorporated into a larger ornament (cf. "The Slide"), because his basic rule for the
performance of ornaments written in small notes is that they must be played on the
beat of the following principal tone (cf. Ch. I I , ^ 23, 24). The essential reason for
his disapproval becomes apparent upon reading his description of the functions of the appoggiatura i(i % 1 of the present section, especially the last two sentences.
His immediate objections are directed to the excessive use of the ornament, its free
insertion by performers, and above all its use where the appoggiatura proper is
specified. All of us have suffered on this last score and henee should commiserate
with Bach to a degree.
M Remainder of paragraph from ed. of 1787.

P e
iTJT)^sf i.
ffi

f=H=i

9 SJJ

m. 0a

aa^

II

**

9 circir
Figure 89
^

1 II

ir

01

15

3% p r
#

ir

'

"P
r-jl

_ j 9 _ B*

26. Other embellishments which are written as small notes w i l l

be explained i n later sections.
THE

TRILL

1. Trills enliven melodies and are therefore indispensable. I n

earlier times they were introduced chiefly after an appoggiatura
(Figure 90, Example a) or on the repetition of a tone (b). The first
i Figure 89, Example b, from ed. of 1787.

E M i E L L 1 SU M E N T S

loo

EMBELLISHMENTS

101

is called the enclosed trill. Today they are used in both stepwise and
leaping passages, immediately at the beginning of a movement, i n
SUCCession, at cadenees, and, i n addition, on held tones (c), fermate
(b), and caesurae without (e) as well as with (/) an introductory appoggiatura. Thus, this embellishment has become versatile with the
passing of time.

6. A t times two short notes from below are appended. These are
called the suffix, and they serve to make a more rapid trill (Figure
92, Example a). The suffix is often written out (b) as well as indicated through an addition to the symbol (c). However, since the
long mordent has almost the same symbol, I think it better to retain
the m and avoid confusin.

Figure 90

Figure 92

a. <fr

b.

&

<tr

c.

ir
ur:

3E

ir

pj>r | i f i P g p i
2. However, i t is strongly recommended that the t r i l l be employed circumspectly i n affettuso passages.
3. The accomplished keyboardist has four trills; the normal,
ascending, descending, and half or short t r i l l .
4. Each has its distinctive sign i n keyboard pieces, although all
may be indicated by either the abbreviation tr. or a cross. The performer has no need to be unduly concerned about the proper location of the t r i l l , for its acknowledged symbols are almost always
notated.
5. The normal t r i l l has the sign of an m (Figure 91, Example
a), which is extended when it appears over long notes (>). Its execution is illustrated i n Example c. Since it always begins on the tone
above the principal note, it is superfluous to add a small note (d)
unless this note stands for an appoggiatura.
Figure 91
1

** ir

This is a rule of long standing. On the strength of Bach's inclusin of it and

the numerous examples that follow, it is safe to conclude that there is no form of
Bach trill that starts on the principal tone. The closest to one is the ribattuta, which is
illustrated in passing (\5 and Figure 105), although Bach does not give it its
common ame. Note, however, that when the trill proper enters in this example it
is the ascending kind and does not commence on the principal tone. The trill-like
long mordent starts on the principal tone, but it alternates with the lower neighbor
1

7. Trills are the most difncult embellishments, and not all performers are successful with them. They must be practiced industriously from the start. Above all, the finger strokes must be uniform
and rapid. A rapid t r i l l is always preferable .to a slow one. I n sad
pieces the t r i l l may be broadened slightly, but elsewhere its rapidity
contributes much to a melody. W i t h regard to the amount of pressure, the performer must be guided by the nature of the passage, be
it forte or piano, i n which the trill appears.
8. I n practicing the t r i l l , raise the fingers to an equal but not
an excessive height. T r i l l slowly at first and then more rapidly but
always evenly. The muscles must remain relaxed or the t r i l l w i l l
bleat or grow ragged. Many try to forc it. Never advance the speed
of a t r i l l beyond that pace at which it can be played evenly. This
precaution must be heeded i n practicing rapid as well as difncult
passages so that they may be performed with fitting lightness and
clarity. Through intelligent practice it is easy to achieve that which
can never be attained by excessive straining of the muscles. When
the upper tone of a t r i l l is given its final performance i t is snapped;
after the stroke the upper joint of the finger is sharply doubled and
drawn off and away from the key as quickly as possible.
3

9.

The t r i l l must be practiced diligently with all fingers so that

and henee has no bearing on the present point (cf. Figure 140, example a). However,
see Schenker, Ein Beitrag zur Ornamentik, p. 34, % 3, where a broader explanation
is evolved.
2 Nachschlag. Bach uses this term in a general rather than specific sense. It applies
to ornaments, parts of ornaments, or other notes that fall on divided, or fractional,
beats.
3 From meckern. Marpurg calis it chvroter.
* This sentence and the following were footnotes in the ed. of 1787.
5 Cf. Ch. I , Note 2.

ro

EMBELLISHMENTS

E M li E E El Sil M E N T S

they will bcconic strong and dexterous. However, let no one believe
that all of the lingers can be made to trill equally well. For one
thing, there are natural differences among them, and for another,
compositions usually offer more trills for certain fingers than for
others; henee these are unwittingly given more practice. Yet prolonged trills appear at times i n outer parts and preclude a choice of
fingers, most of them being engaged in performing the inner parts.
In additiori certain passages are extremely difficult to perform unless the little finger has learned to t r i l l rapidly, as illustrated i n
Figure 93.
Figure 93

10. No one can succeed without a minimum of two good trills i n

each hand: The second and third, and the third and fourth fingers of the right hand; and the thumb and second, and second and
third fingers of the left. I t is because of this normal fingering of
trills that the left thumb grows so age and along with the second
finger becomes about the most active of the left hand.
11. Some performers practice double trills i n thirds with one
hand. Various examples of these may be constructed from Figure
42. Such exercise, pursued as far as one wishes, is beneficial to the
fingers, but aside from this it is better not to employ double trills
unless they can be made to sound even and distinct, the two desiderata of good trills.
12. When the upper tone falls on a black key and the lower
on a white key it is ot incorrect to perform a t r i l l with the second
finger of the left hand crossed over the thumb as illustrated i n Figure 94. Also, some find it convenient to trill with the third and fifth
or the second and fourth fingers of the right hand when the action
of the keys is stiff.

Figure 94
,2121

13. Trills on long notes are played with a suffix regardless of a

subsequent stepwise descent or ascent. The suffix may also be added
to a t r i l l followed by a leap (Figure 95, Example a). When the
decorated notes are short, an ascent (b) is better after a suffix t r i l l
than a descent (c). Although i n quite slow tempi the trills of Example d may be suffixed (despite the fact that the rapid notes following the dot may be used as substitutes), i t can be seen that a
descending second is the least favorable for such an addition. I t is
not unconditionally necessary to suffix the ornament, provided that
the dotted notes are trilled for their full length^ The manner of performing the endings i n this example w i l l be taken up i n the following paragraph, which discusses dotted notes.
Figure 95

i f j f i'D 1 u TrJiTBjp

o This is in complete agreement with Couperin (L'Art de toucher le clavecn). The

thumb of the left hand played a more active role than that of the right in older
systems of fingering. For example, a widely used ascending fingering for the left hand
was: 4, 3, 2, 1, 2, 1, 2, 1, while the corresponding descent for the right hand was:
5, 4, 3, 2, 3, 2, 3, 2.

14. Dotted notes followed by a short ascent also alow for suffixed trills (Figure 95, Example e). However, instead of the usual
extremely rapid motion into the following note (/), when dotted
notes are trilled a very short separation must be made between the
last tone of the suffix and the following note (g). This separation
need only be long enough to show that the suffix and the following
note are two seprate elements. Its length is dependent on the

104

l {

E L 1 S 11 M EN

TS

EMBELL1

lempo; henee the execution of Example g is only approximately

suggested by the time valu of the last tone of the suffix, for the
note following a dot is always shorter i n execution than its notated
length (a point which will be treated i n Chapter I I I ) . The suffix
running directly into the following note i n Example h is, of course,
incorrect. The composer who wishes such an execution must cali
for it expressly.
15. The suffix must be played as rapidly as the trill proper. A
trill i n the right hand for the thumb and second finger is not favorable for a suffix because it can be added only by crossing the fingers,
which retards its pace. I n this manner the best trill will be brought
to a ruinous end.
16. The unsuffixed trill is best used i n descending successions
(Figure 96, Example a) and principally over short notes (b). The
suffix is omitted from successive trills (c) and from trills followed by
one or more short notes which are capable of replacing i t (d). I f
this substitution is made, the asterisked example must not be played
7

/)

b.

i Uf

1 r

* J

RI

______
JTT.

'

** 1

n j-aba II 1
*i

J
7 Cf.

Ch. III, f S3.

TS

/*'

F ~ g r 4 =

- 1

19. When accidental signs are not included with the symbols
of trills and suffixes the correct alterations may be arrived at by
considering the preceding tones (Figure 98, Example a) or the succeeding (b). Sometimes the ear alone or modulations will dctate the
necessary changes (c). While we are on this subject, it should be observed that neither trills or suffixes are allowed i n the interval of
an augmented second (d). Aside from the keyboard there is a constant need for the notation of accidentis attendant on trills; especially i n ripieno parts, for i n these it is difficult to perceive modulations with their rapid alteration. For this reason many set the
accidentis as appoggiaturas before the trill. This, however, is confusing i n that it suggests a holding of the initial tone rather than an
immediate t r i l l .
Figure 98

a.

*/**"

r - g g II - mP

j J J

J >

Figure 97

<m

S 11 MEN

in the slowest tempo. Furthcr, the suffix is not employed over

triplets (e). I t is always omitted from the last of those in Example
e, although it may be introduced into the first three, but only i n
very slow tempos.
17. The average ear can always tell whether the suffix should
be used. I have discussed it here only for the benefit of beginners
and because this is its proper place.
18. I n very rapid tempos the effect of a trill can be achieved
through the use of the appoggiatura (Figure 97). The last two short
notes are not an unsatisfactory substitute for the suffix.

>

lI
iilg J
i

M T f f

s Remainder of paragraph from ed. of 1787.

to

E M B E E LISHM

E N TS

20. Aniong the errors unwittingly causee! by trills we shall first

mention the following: Many burden the first notes i n the examples
of Figure 99 with a trill despite the presence of a slur. No matter
how enticing the appearance of such notes, they must not be trilled.
Why must the finest legato passages be ruined so often by inept
playing? Indeed, most errors are committed i n just such places.
Trills are introduced i n an attempt to rescue these passages from
oblivion. The pampered ear demands such treatment, being incapable of perceiving anything but a bustling noise. I t is apparent
that those guilty of these faults can neither think lyrically or
grant to each tone its proper weight and length. Tones w i l l sing
on the harpsichord as well as on the clavichord if they are not
detached Crom each other, although one instrument may be better
constructed for this purpose than another. The French are not well
acquainted with the clavichord, most of their compositions being
written for the harpsichord^ Yet their works are replete with held
and legato notes which are indicated copiously with slurs. Even
when the tempo is too slow or the instrument not good enough to
sustain tones properly, it is better to sacrifice a little of the clear
flow of a legato passage than to disrupt it with trills, for a correct
performance will be ampie compensation for the lack of sonority.
There are many things i n music which, not fully heard, must be
imagined. For example, i n concertos with full accompaniment,
the soloist always loses those passages that are accompanied fortissimo and those on which the tutti enters. Intelligent listeners replace such losses mentally, and it is primarily such listeners whom
we should seek to Dlease.
Figure 99

JL

0' m 0

>

21. There are other errors as ugly as they are frequent: The
appending of a limp suffix to a t r i l l as i n Figure 100; the addition
to the suffix of a short note, which can be justly included among the
worthless unaccented appoggiaturas (Figure 101); failure to give
trills their full length, which (excepting the short trill) must always agree with the valu of the note over which the symbol ap9

9Cf. Ch. II, "The Appoggiatura," Note 12.

EM

BEL

1.1 Sil M EN

TS

107

pears; plungng directly into a trill without playing a preceding appoggiatura or properly joining both ornaments; performing such
an impertinent t r i l l loudly i n a subdued, plaintive context; trilling
excessively under the delusion that every moderately long note must
bear a t r i l l . These are the pretty little trills mentioned i n Paragraph 10 of the Introduction.
Figure 100

Figure 101

Q ~r
22. The ascending t r i l l with its symbol and execution appears
in Figure 102. Because, aside from the keyboard, this symbol is not
widely known, it is often notated in the manner of the asterisked examples; or the general abbreviation tr. is written, the choice of
t r i l l being left to the discretion of the performer.
1 0

Figure 102

23. The ascending t r i l l requires a long note, for it comprises

many tones, including the normal suffix. A rapid suffix is written
out. W i t h regard to such details the performer should follow the
precepts, previously discussed, of the normal t r i l l .
24. The examples of Figure 103 are noteworthy. Example a
illustrates a suffix introduced after a tie; i n Example b the suffix may
be omitted because of the sixteenth; likewise i n c, because of the
thirty-seconds; however, i n a sufficiently slow tempo, or at a cadenee
or before a fermata (it being permissible to broaden ad libitum i n
the latter two), the suffix is included and the succeeding short notes
appended, the final one being played somewhat slower than the
other (d). I n my opinin, this widely used embellishment can be
best applied to Example c, although the last notes must be played
occasionally at other speeds. I n passing, note that i n minor mode
cadenees the t r i l l is sometimes played on the sixth above the bass
rather than the fifth.
10 Der Triller von unten.

EM

BEL

LI Sil M EN

TS

EMBELLISHMENTS

b.

4$.

109
Figure 1107

Figure 106

-CZL

26. I n successive leaps only the normal t r i l l may be used (Figure 108). I t is wrong to attempt to bring such a passage into bolder
relief by means of ascending or descending trills.
Figure 108
1 2

r
25. Thus the ascending t r i l l appears principally over long notes,
especially at cadenees, and before fermate. I n addition, however, i t
is found over a repeated note as i n Figure 104, Example a, i n conconjunct motion (b), and after a leap followed by an ascending or
descending progression (c). Over long notes of several bars, trills
that threaten to lapse can be revived by means of the ascending prefix, but i t must be interpolated without causing the slightest discontinuity. The ascending t r i l l is well fitted to this purpose of
resuscitation, for its insertion renews the fingers' strength. I t is possible to move through an octave with such a t r i l l , for its two short
introductory notes facilitate the fingering. Figure 105 illustrates
a manner of approaching it by means of a gradual acceleration, a
device frequently employed at cadenees. The ascending t r i l l may
also be used i n the course of modulatory changes, as i n Figure 106.
Figure 107 illustrates its application to caesurae.
11

Figure 104

-a.a.

b.

&rrET'r r
J

<T

a.

rr

"

27. The descending t r i l l is illustrated i n Figure 109 with its

correct symbol and execution. Apart from the keyboard i t is occasionally notated i n the manner of the asterisked example.
Figure 109

i i g

r rr r r
i

28. Because it contains more tones than any other t r i l l i t requires the longest notes. Henee both of the previously discussed
trills are better suited for a cadenee such as that of Figure 11 o than
the descending t r i l l . I n earlier times it was used widely, but today i t
is restriced largely to a repeated tone (Figure 111, Example a),
descending successions (b), and downward leaps of a third (c).
Figure 110
Allegretto
Figure 111
a.
W

r~

TTT
1 1

Known generally as the

ribattuta.

T>

1 2

r
-twr

r rr rr

Der Triller von oben. Cf. f 27.

E M li E L E I S H M E N T S
/ 1()
29. Because, as already mentioned, ornaments must not corrupt the purity of voice leading, it is better to employ either the
normal or the descending t r i l l i n Figure 112, for the ascending
t r i l l creates forbidden fifths.
Figure 112 **'
/U

T7*r f
30. The half or short t r i l l , which is distinguished from the
others by its acuteness and brevity, is notated for the keyboard i n
the manner of Figure 113. Included i n the figure is an illustration
of its execution. Despite the upper slur, which reaches from the
beginning to the end of the example, all notes are played except
the second g and the last f, each of which is tied to its preceding
tone by another slur which indicates that it must not be struck.
The large slur merely specifies the attendant phrasing.
13

1 4

E M E L L SUMEN

T S

in Paragraph 8, but with such exceeding speed that the individual

iones will be heard only with difficulty. Herein lies its acuteness,
which stands beyond comparison with the sharpest of other trills.
Like the short appoggiatura, it may appear over rapid notes, but i t
must be played with such speed that the listener will not feel that
the note to which it is applied has lost any of its length, but rather
that it has entered precisely at the proper moment. I t must not
sound as frightening as it looks fully written out. The short t r i l l
adds life and brilliance to a performance. I t is possible, when necessary, to omit any other ornament, even the other trills, and arrange
matters so that easier ornaments may be substituted for them. But
without the short t r i l l no one can play successfully. Even if all
other ornaments were correctly performed, no one could be happy
in the absence of this one.
33. Since the short t r i l l deniands great skill and speed i n execution, it is best performed by those fingers that t r i l l the best. Consequently, it is permissible, as illustrated i n Figure 114, to take liberties with the fingering of a passage and adopt unusual expedients i n
order to execute the ornament. Of course, this must be done so
skillfully that the performance as a whole does not suffer.
;

Figure 113
Figure 114 _,

31. The short t r i l l joins the preceding note to the decorated

one and therefore never appears over detached notes. I t represents
in miniature an enclosed, unsuffixed t r i l l , introduced by either an
appoggiatura or a principal note.
32. The half or short t r i l l is the least dispensable, the most attractive, but at the same time the most difncult embellishment.
Played incorrectly, either it cannot be heard at all or else it sounds
limp and ugly, which are attributes far from its true ones. Unlike
other ornaments, it cannot be demonstrated slowly to students. I t
must literally crackle. I n order to be truly effective the upper tone
must be snapped on its final appearance i n the manner described
13 Der halbe oder Prall-Triller. The symbol is that of our inverted mordent, but
cf. Ch. II, "The Snap," f i and note i .
1* In the original illustration (Figure 113 here) this "second g" was not tied to the
first, an oversight that has been perpetuated in most later (nineteenth- and twentiethcentury) reproductions.

34. The half or short t r i l l appears only i n a descending second

regardless of whether the interval is formed by an appoggiatura or
by large notes, as depicted i n Figure 115. I t is found over short
notes (Example a) or over those made short by a preceding appoggiatura (b). I n this latter respect, when it appears over a note extended by a fermata, the appoggiatura is held quite long and the
t r i l l is quickly snapped as the fingers withdraw from the keys (c).

l\
-

/ / 2

n /: 1.1.1 s n

A /; N

r s

.'55. In addition to its cniployment at cadenees and ermale it is

Eound in descending passages o three or more tones, as in Figure
i 6. I n this use it resembles the unsuffixed trill in a descending
succession and, like it, appears in passages where long notes are
followed by short ones (Figure 117).

EMBELLISHMEN

TS

side according to whether they pertain to the first or the third ^bone,
as illustrated in Example a.
Figure

118

!t A d a g i o

Modralo

Presto

Figure 116

Figure 117

36. W i t h regard to the execution of this trill, it must be pointed

out that it is almost insuperably difficult to play it lightly at the
pianoforte. Because the snap requires pressure, its performance
on this instrument increases the volume. Yet it is impossible to perform our trill without this characteristic element. Henee the performer is faced with a dilemma, worsened by the fact that the short
trill either by itself or combined with the turn often follows an
appoggiatura and therefore, according to the rules governing the
execution of appoggiaturas, must be played softly. T h e problem
arises i n all snaps, but particularly here, where it assumes its most
radical form. I doubt that the most intensive practice can lead to
complete control of the volume of the short trill at the pianoforte.
THE

TURN

1. T h e turn is an easy embellishment which makes melodies

both attractive and brilliant. Its symbol and execution are shown
in Figure 118. Leaps of an octave or other large intervals necessitate the use of four fingers i n order to perform it. W h e n such is
the case it is customary to place two fingering numeris over the
decorated note. C o n t e x t often requires the placing of one or two
accidentis over the symbol. T h e y appear on the right- or left-hand
2

1 Der Doppelschlag.
2 Remainder of paragraph from the ed. of 1787.

2. Because it is almost always performed rapidly I have had to

illustrate the vales of its notes in slow and rapid tempos. It is also
indicated by the symbol that appears i n the asterisked example. I
have chosen the other i n order to avoid an ambiguous placing of
fingering numeris.
3. T h e turn is employed i n slow as well as fast movements, and
over slurred as well as detached notes. I t does not appear to advantage over a very short note because the time demanded for the performance of its several tones may detract from the clarity of the
melody.
4. T h e turn is sometimes found alone, sometimes in combination with the short trill, and also after one or two thirty-seconds i n
small notation which are placed before a large note and differ from
the appoggiatura, as we shall see presently.
5. W h e n the turn alone is used, its symbol may appear either
directly over a note or after it, somewhat to the right.
6. I n the first case it is employed, as illustrated i n Figure 119,
in stepwise successions (a), leaps (b), caesurae (c), cadenees (d), fermate (e); immediately at the beginning (/), i n the middle (g), or at
the end after an appoggiatura (h); over a repeated tone (i), or a note
preceded by a repetition regardless of whether the note is reached
by a step (j) or a leap (k); without an appoggiatura, with one, over
one (l), after one, etc.
> E . F . Baumgart (Foreword to his edition of the Kenner und Liebhaber collections,
Breslau, 1863) expressed the opinin that in the Moderato example /-sharp should be
a sixteenth, the additional beam being an oversight. Cf. Figures 68, 128, and H . Schenker, op. cit., p. 45, note 1.
* Example a from ed. of 1787.

EM

ii4

li E l. 1.1 Sil

M E N

IS

EMBELLISHMENTS

y p s

^-~~~h? 1

j
fe

J)

would oceupy the entire duration of the note, the turn, much
shorter, would leave a part of the length unfilled.
11. While discussing this matter, I must point out an exception
in slow tempos where, because of the affect, a t r i l l may be replaced
by a soft turn, the last tone of which is held until the following note
enters. As illustrated i n Figure 120 this may oceur i n cadenees and
also after an ascending appoggiatura (a).

Figure 119

j-jj

Figure 120

-4-=1
i.

II

00
1

* m*m

r^J-

7. This lovely ornament is almost too obliging. I t fifs almost

everywhere and consequently is often abused. Many seem to believe that the sum and substance of the keyboardist's art consists
in introducing turns at every slightest instance. Henee its correct
use must be carefully investigated, for, despite its complaisance,
many apparent opportunities arise which are not actually suitable
to i t .
8. I n most cases the turn serves to add brilliance to notes. Henee,
passages which must be played undecorated and sustained because
of the affect are ruined by those who insert a turn because of the
length of the notes, i n ignorance of style and touch. Also, the error
of excessive use common to all ornaments applies to this one.
9. A general understanding of its correct use can be gained by
considering the turn a normal, suffixed t r i l l i n miniature.
10. I n most cases the turn is performed rapidly and its upper
tone is snapped i n the manner already described. Henee i t is wrong
to play it instead of the normal t r i l l on a long note. While the t r i l l

12. I t follows from its similarity to the suffixed t r i l l that the

turn prefers an ascending to a descending following tone. I t is easy
to move upward (but not downward) through an octave or even
further by means of a series of turns. Aside from the keyboard this
frequent use of successive turns is indicated i n the manner of Figure
121. The turn should not be applied to rapid, descending notes. I t
may replace the t r i l l i n those cases where the latter is difficult to
perform owing to the presence of another voice i n the same hand.
The substitution may be made only on a relatively short note, for
others cannot be completely filled i n by the turn (a).
s

F i

u r e

Presto

1 2 1 < i

'

'

13. Again, like the related t r i l l , our embellishment may be applied to leaping notes without further concern. Figure 122 illustrates its use i n both ascending and descending leaps.
Remainder of paragraph from ed. of 1787.
Example a from ed. of 1787.

n6

E M li E L L 1 S II MEN

Figure 122

a.

r >p r T

T S

EMBELLISHMENTS

fe

14. Although the turn is well used over a repeated tone, the following tone, a t least i n the case of rapid notes, should rise a
second. When i t descends, the compound appoggiatura is better
(Figure 123).
7

Figure 123

es:

17. Despite the musical worth of this ornament, its symbol is

little known apart from the keyboard. I t is often indicated by the
signs of the t r i l l or even the mordent, these two also being often
confused. I n Figure 126 there are many examples i n which the
turn is better and easier than the t r i l l . Examples a, g, p, and q are
the true home of the turn, for no other ornament can be applied
to them. Those marked j , k, l, and m, are as well suited to the t r i l l
as the turn i n rapid tempos. Note that i n these fragments the last
note repeats the middle one. I n example n an appoggiatura is occasionally appended to the turn i n slow tempos.
Figure 1 2 6

LLT

15. Further, the turn often appears over a long note which follows an appoggiatura as previously shown i n Figure 119, Examples
c, e, and /. Note that a turn over an appoggiatura (most of the repeated notes mentioned i n the preceding paragraph are appoggiaturas) will not suffer a decoration over the following tone (Figure 124). The exception to this oceurs before a fermata, where the
appoggiatura is lengthened beyond its normal valu. The last note
of the turn must then be held to crate a slight but not an awkward
break before the entry of the long mordent (a).
Figure 124

x,

a.

le

16. Appoggiaturas which do not repeat the preceding tone are

not turned (Figure 125), although the following resolution may
be (a).
Figure 125
a.

7 Matter between commas from ed. of 1787, footnote.

8>er Anschlag.

Cf. C h . I I , " T h e Compound Appoggiatura."

117

Example o from ed. of 1787.

n8

EMBELLISHMENTS

EMBELLISHMENTS

18. The lack of symbols aside from the keyboard often leads to
the setting of the trill's sign i n places where this ornament is i l l at
ease. Sometimes the speed of a piece makes i t impossible to execute.
Figure 126, Example o, illustrates such a case i n a passage typical
of Tartini, and many allegro movements. The performer should
play a turn here, for it is not merely acceptable but i n keeping with
the speed and the desired effect. I n other cases legato phrasing
makes for an awkward t r i l l . The last two examples are entitled
"Recit.," and differ from each other only i n the melodic endings.
Both cali expressly for a turn. I n the first of these-the last note of the
turn is not held i n the usual manner, i n order to imitate the
declamatory style of the voice. Since i t is impossible to set the sign
of a t r i l l here, the passage must be left to the discretion of the performer when other signs are lacking.
10

19. As illustrated earlier i n Figure 119, Example e, the turn may

appear over a fermata preceded by an ascending appoggiatura. I t is
never found, however, over a final note approaching i n a like manner (Figure 127). Yet it does appear i n both cases after a descending
appoggiatura (Example a and Figure 119, Example h).
Figure 127

21. The turn by itself may appear between a note or appoggiatura and the following tone in three situations: First, when the
note is fairly long, as i n Figure 129, Example a; second, over a tie
(fe); third, after a dotted note (c). These uses of the turn are very
frequent i n all kinds of music and cannot be clearly enough indicated without our distinctive symbol, although some cali for i t by
setting the t r i l l sign after a note. I n all three cases i t serves to fill
out notes.
22. The first case occurs i n all kinds of motions, but not very
well before a stepwise descent. When a performer wishes to avoid
a cadential t r i l l he may execute a turn after an appoggiatura which
ascends to the final note, as illustrated i n the asterisked example of
Figure 129. When such is the case a mordent should not be played
over the final note. The execution of all turns i n example a is
shown i n the last illustration of that group.
23. I n the second case, the tying note acquires a dot and the tied
note becomes the last tone of the turn. When the tempo is rapid the
11

Figure 1 2912

f ? L-Tf

rr

rr

20. Although the turn and the t r i l l are similar, there are two respects i n which they differ from each other. First, since the final
tones of the turn are played less rapidly than the preceding ones,
there is always a small space between them and the following tone.
Second, the turn occasionally lays aside its brilliance for a purposely broad execution in slow, expressive movements (Figure 128).
This kind of performance is also specified i n the manner of Example a.
Figure 128

00

ff-

fe

i a.

M=

=4
tr
V

f = p

1 J

Moderato
m. 0 m 1

II

es

rJ

= = = = =

r-Vi

* n

1.
6 0

12 Examples d, e, f, g, h from ed. of

1787.

1787.

Tj*

Allegro
~m~0 m -r10nt
m

1
-

'-U*'

" Remainder of paragraph from ed. of

10 T h i s sentence and the next from ed. of 1787, footnote.

119

m a-

E M liELLI

/90

1 ,

T S

I l i l i l

f. 4.

S H M EN

"

| i

sa

r F r

_^s

i,

r 1

execution
P"""l

usual indicacin

<? Allegretto

EM

h.

better r*"l

mdications
better

r-|

dot is omitted. Both executions are clearly written out under Example b. This use of the turn is frequent at cadenees.
24. I n the third case (Figure 129, Example c), two tones acquire
dots and the turn is placed between them as illustrated i n the notated execution of Example 2. The divisin of the tones is unvariable. This kind is often used when the tempo is so slow that the
dotted note, undecorated, sounds listless; it is also found i n caesurae
(Example 1) and before cadenees when a dotted note is followed
by a short one which anticipates a trilled tone (Example 2). Such
a turn is not employed after descending dotted notes of only modrate length. Example 3, set with this ornament, is the true home of
the turn, for a substituted t r i l l , whether it is placed over or after the
first note, is unconditionally wrong. Example 4 shows that the turn
may be placed over the second note as well as after the first. The
accompanying written-out divisin of the tones demonstrates conclusively that this employment requires a slow tempo. How is i t
possible to indcate a desired detaching of the final note of a turn
(d)? Either by notating a rest () or by placing a vertical stroke to
the right of the symbol and just above a second added dot. (/). A l though this new indication looks strange, i t is necessary, for all
means which lead to correct performance should be adopted, even
though they seem excessive. Occasionally a turn appears after a
relatively long note against which the bass has two or more notes or
13

i Remainder of paragraph from ed. of

1787.

BE

L L I S H MEN

TS

121

a rest, as in Example g. I n the interests of greater clarity, divide the

long note into two tied parts (h). The proper entrance of the turn
w i l l be thereby made apparent. Otherwise, many uninitiated performers might introduce the ornament too soon, thus necessitating
a slow performance of it i n order to fill out the note. This would be
unsatisfactory and contrary to the rule stated i n Paragraph 10.
25. Accidentis which oceur i n turns are brought about, as in
the case of trills, by preceding or following notes and modulations.
Like the t r i l l , a turn must not be used i n the interval of an augmented second (Figure 98, Example d).
26. The required snap i n the turn, at which the little finger is
not adept, requires an occasional slightly forced fingering (Figure 130).
Figure 130

27. The turn allies itself with the short t r i l l when its first two
notes are alternated with extreme rapidity by means of a snap. The
effect of the combined ornaments^an be most easily realized by
thinking of a short t r i l l with a suffix. This trilled turn introduces
a unique charm and brilliance to the keyboard. I t is a miniature but
lively, enclosed and suffixed t r i l l with which, however, i t must not
be interchanged, for there is as great a difference between the two
as there is between the short t r i l l or the turn and the normal t r i l l .
It has no distinctive symbol. I specify its use i n the manner of Figure 131, which also depiets its execution. W i t h regard to the long
slur over the second illustration, I refer the reader to the discussion
of the short t r i l l i n Paragraph 30 of the section on trills.
14

Fieure 1 3 1

15

28. The trilled turn oceurs either with or without a preceding

appoggiatura. However, like the short t r i l l it is used only i n a
descending second, the first note of which is drawn into the embeli * Der prallende
Doppelschlag.
T h e first two notes are not tied in the original illustration.
1 B

E MBELLI

722

S H

MENTS

EMBELLISHMENTS

lishment as illustrated in Figures 131 and 132. Inasmuch as the

trilled turn contains more notes than either of the ornaments which
comprise it, it filis out relatively long notes better than either one
alone. Consequently i t is better to use it instead of the short t r i l l i n
passages such as the first three of Figure 133. On the other hand, the
short trill alone is better i n the asterisked example when the tempo
is allegretto or faster. As a general rule it can be stated that neither
the simple or the trilled turn is good i n passages which are suitable
for an unsuffixed t r i l l . I n moderately fast movements the trilled
turn is often performed i n the manner of Example a. Such an execution is acceptable so long as i t does not crate bad voice leading with
the bass. Henee, while the example marked (1) is good, that marked
(2) is not. I t is used at half (3) and whole cadenees (4). Further, this
execution is more readily comprehended when, instead of setting
both symbols, the notation of Example a is adopted.
16

Figure 132

Figure 133

17

es

123

29. I n slow tempos when three notes descend, the second, over
which a trilled turn may appear, takes an appoggiatura, as does the
final note. This is illustrated i n undecorated form (a), with the ornaments (b), and with the execution written out (c), i n Figure 134.
The first appoggiatura is quite usual before slow notes, for i t helps
to fill them out. Moreover, i t is necessary here for the convenience
of the trilled turn, which must not enter before half the duration
of the principal note has passed, precisely the time taken up by the
appoggiatura. The second appoggiatura not only serves to shorten
the final tone, thereby bringing i t into agreement with the preceding; i t also satisfies the ascending tendeney of the turn which this
ornament has i n common with the suffixed t r i l l . There are three
reasons against playing the second appoggiatura ahead of its beat,
thus separating i t from its following principal tone and incorporating i t i n the turn: First, because it is a prefix and not a suffix;
second, because i n accordance with our explanation of the turn, its
final tone must never r u n directly into the following tone (the
appoggiatura i n this case) but must always delay a bit i n order to
avoid the fault of a t r i l l whose suffix acquires an additional t o n e ;
third, because the appoggiatura serves to divide the following tone
in a manner similar to the other parts of the passage. Here again we
see the amount of harm which can be caused by separating an appoggiatura from the beat of its following tone. T o avoid this error,
perform the body of the trilled turn according to the rule, so rapidly
that the final tone, c, may be made to sound like a simple sixteenth;
this will crate an adequate separation of the ornament from the
following appoggiatura. The illustrated execution of this passage
looks rather alarming. Indeed, were it written out according to the
Figure 134
18

J T3 W = j :

18

3.
i Cf. C h . I I , " T h e T r i l l , " f 21 and Figure
i Remainder of paragraph from ed. of
" Last six bars from ed. of 1787.

1787.

101.

is I n example c the second and third notes are not tied in the original illustrations.
Cf. Ch. I I , " T h e T r i l l , " f 30.

E M B E LLl

12./

S II M EN

T S

EMBELLISHMENTS

manner in which it must often be played over sixteenths i n an

adagio, with each note of the turn once again as fast, it would look
even worse. Nevertheless, the entire art of execution depends on
the ability to perform a rapid trilled turn, one whose execution
sounds natural and facile. Example d is different from the others,
but the performance of the last two notes is the same as i n the
other examples.
30. I n performing the simple turn or the suffixed t r i l l , at least
three fingers must always be employed. Because, beyond this, the
snap in these ornaments and particularly i n the trilled turn can be
well executed by only certain fingers, there often arise great difficulties of performance, i n the solution of which extreme expedients must be employed. Figure 135 illustrates a few such cases. I n
Example a, after e has been played by the second finger the hand
shifts slightly to the left and the third finger takes d. I t must not,
however, as incorrectly taught, strike its tone by crossing over the
second finger. I n Example b our compounded embellishment
forces the third finger to glide from a black key to the white one
below. The easiest fingerings are those in Example c. Nevertheless
it is advisable to practice the trilled turn with all fingers because
they w i l l thereby increase in strength and dexterity, and above all,
because we are not always i n a position to employ only the best
fingers in performing an ornament.
Figure 135
5

co

.,

14 3

Figure 136

125

&

co

I ^^J

l . I >lXj nf I
J

32. The short t r i l l and the related trilled turn provide unfailing tests of a harpsichord's quilling, for an instrument i n poor condition will be unable to enunciate them. Keyboardists must be
pitied who are robbed of these most essential and superior ornaments because of the poor repair of an instrument. Without them
most pieces can be but poorly performed.
33. When a turn is introduced over detached notes it gains
acuteness through the prefixing of a note whose pitch is the same
as the decorated one. I ntate this ornament, discussed i n no other
writings, by placing a small thirty-second before the turned note.
The thirty-second is unvariable regardless of the tempo or the valu
of the following note, for it is always played with a very rapid
stroke delivered by a stiff finger and immediately connected with
the following snapped note. This makes for a new kind of trilled
turn, which may well be called the snapped turn - to distinguish
it from the other. I t is better suited to rapid notes than the t r i l l , for
I feel that a t r i l l is at its best over a note whose valu allows for
generous alternation; otherwise another decoration should replace
it. Through the added small note the turn acquires a brilliance
equal to that of the trilled turn but applied to just the opposite
situation.
34. While the trilled turn may be introduced solely after a
descending slurred second, it is precisely this situation alone which
will not suffer a snapped turn. I n Figure 137 we find its symbol (a),
its execution (b), and a few of its characteristic uses (c). I t may appear at the beginning of a passage, i n the middle, before stepwise
motion or a leap, but not over a final tone, staccato or otherwise.
21

co

/V

3 2

5E3

$5

32

co

/V

2 3

31. Embellishments are not easily introduced into the bass unless they are expressly called for. Nevertheless the trilled turn may
be interpolated when opportunities such as those of Figure 136
present themselves.

20 Der geschnellter
Doppelschlag.
21 It will be seen from the illustration that Bach is discussing the full turn, which
is pressed into service by performers and editors far more frequently than it should
be. T h e indication used here was not widely adopted. It appears, for example, in
some of Haydn's early sonatas. In fact, Bach's heroic attempt to indcate by symbol
or notation all of the fine variants of the turn went for naught. Most of his contemporaries and later composers used the same sign (~ or tr.) for all types indiscriminately or they wrote out the ornament in both small and large notation. Such practices make the entire section with its many examples all the more important, for
it should serve to sharpen our sensitivity to the kind and amount of refinement that
reside in the turn.

126

E MBELLI

S H MEN

EMBELLISHMENTS

T S

I2J

It should be noted that aside from the keyboard the snapped turn
is indicated by the sign of a t r i l l and, even i n keyboard pieces, often
by the simple sign of the turn. I t may be introduced over the
second of a pair of slurred notes i n stepwise ascent, as i n Example
d. I n such a situation it replaces the ascending t r i l l or the ascending
t u r n . Example e shows the snapped turn over the first of a pair
of slurred notes i n stepwise descent. This use is justified by the preceding detached notes. When these are also slurred, as they may be
in a slow tempo, a simple turn or the compound appoggiatura is
better, as i n the asterisked example.

37. Finally, the turn may be preceded by two small thirtyseconds. These small notes are incorporated i n the ornament and
joined to it as rapidly as possible. The threefold beam is unvariable.
This variant, discussed here for the first time, represents a miniature ascending t r i l l for which i t may be substituted over short
notes. I t might be called an ascending t u r n . Its indication and
execution are illustrated i n Figure 139. When it is used instead of
an ascending t r i l l over the second of two slurred notes, a better effect can be achieved by tying the preceding note to the first note of
the turn as illustrated i n Example a.

Figure 137
a.
A

Figure 139,2S

foirHi^^

1. The mordent is an essential ornament which connects notes,

filis them out, and makes them brilliant. I t may be either long or
short. The symbol of the long mordent is shown i n Figure 140. Its
execution may be lengthened (a) if necessary, but the symbol remains the same. The short mordent and its execution are illustrated i n Example b.

2 2

23

24

26

27

THE

Allegro

Moderato

e.

forrar

35. This embellishment cannot be performed, or at least i t is

not easy to play with its essential briskness, when i t appears over a
note which must be taken by the thumb or by the fourth or the
little finger. The other fingers are much more adept at executing
it.
36. The snapped turn should not be confused with the simple
turn after a note. They differ from each other i n that the latter is
performed appreciably after the principal tone and is found after
slurred and sustained notes. I n order to differentiate the two more
clearly their execution is illustrated i n Figure 138.
Figure 138

m
22 Remainder of paragraph from ed. of 1787.
Cf. % 37-* Cf. C h . I I , " T h e Compound Appoggiatura."
25 Example d to end from ed. of 1787.
3 3

Figure 140
-4*-

I I

MORDENT

I I

2. Although it is customary to play the long mordent only over

long notes and the short over short notes, the symbol of the long
ornament is often found over quarters or eighths, depending on the
tempo, and that of the short mordent over notes of all vales and
lengths.
3. Example c of Figure 140 illustrates an unusual manner of
performing a very short mordent. Of the two tones struck simultaneously, only the upper one is held, the lower one being released
immediately. There is nothing wrong i n this execution, provided
1

26 Der Doppelschlag von unten.

27 Remainder of paragraph from ed. of 1787.
28 Example a from ed. of 1787.
1 Usually called the acciaccatura, but not by Bach. He reserves the term for auxiliary tones that are introduced into arpeggios. Cf. C h . I I I , ^ 26, C h . V I I , f 13.

/2,V

EM

i E I. L 1 S 11 M E N

li E L LI S II M EN

EM

TS

TS

that it is employed less liequently than the other mordents. I t is

used abruptly only, that is, i n unslurred passages.
4. The mordent is especially good in a stepwise or leaping ascent. I t seldom appears i n descending leaps and never i n descending
seconds. I t may be found at the beginning, middle, or end of a composition.
5. I t conneets slurred notes i n conjunct or disjunct motion,
with and without an appoggiatura (Figure 141). I n such passages i t
is employed most frequently over ascending steps and also occasionally after an appoggiatura, as i n the asterisked examples.
When the mordent is applied to an appoggiatura which is joined
to its principal tone by an ascending leap, the principal tone must
be long so that it can lend enough of its valu to make the mordent
full and impressive (a). This use of the mordent serves both to connect and to fill out notes. I t appears occasionally i n recitatives.

Figure 14 2

Figure 141

peated, and performed generally as illustrated i n the second set of

examples under the same lettering. This liberty must be indulged
circumspectly and out of necessity only. The expedient is wrong
when it distorts the composer's intentions. T o lessen the possibility
of such an error, strike a tone with due pressure and hold it. I n so
doing, one realizes that our instrument sustains tones longer than
generally believed. I n using mordents the performer must be careful not to destroy the beauty of a sustained tone. Henee, as with
other ornaments, he must not apply them to every long note or
overextend them. When mordents serve to fill out a note, a small
fraction of the original length must remain free of decoration, for
the most perfectly introduced moident sounds miserable when, like
the t r i l l , i t speeds directly into the following tone.
9. Mordents, chiefly the short ones, add brilliance to leaping,
detached notes. They are found over tones which i n relation to the
harmony are called definitive (Figure 143, Example a), over certain broken chords (b), and i n the middle parts of f u l l chords (c),
although the long mordent may also be employed when the notes
are long; further, they appear over detached dotted notes where the
dot is not held (d), and over notes followed by rests (). They also
oceur over longer notes preceded by short ones which rise by a
step (/) or a leap (g).

a.

6. When i t follows an appoggiatura, a mordent is played lightly

in accordance with the rule covering the performance of appoggiaturas.
7. The mordent is used to fill out sustained tones. Thus, as i l lustrated i n Figure 142, i t is found over tied (a), dotted (b), and syncopated notes. Syncopations may be fashioned out of a single tone
(c) or various tones (d). I n the case of the latter the mordent is best
used over the second tone of a single repetition (e). I t filis out syncopated notes and, i n addition, makes them brilliant.
8. W i t h reference to Examples a and b of Figure 142, it should
be noted that when the tempo is so slow that even a long mordent
w i l l not fill out the notes adequately, they may be shortened, re-

1 =1

b.

1-

ES

U '

1 J

4=i
w*

b
' *
'

01

^-9i-m

1p r

y>

p r

>

~m

r
m

~m

=E=P=

11 PJ

_
' *
1

2 Anschlagend. T h a t the term has a rhythmic rather than harmonio meaning is

clear from Quantz's term for the appoggiatura, anschlagender
Vorschlag.

EMBELLISHMENTS

EMBELLISHMENTS

190

10. Of all the embellishments, the mordent is most frequently

interpolated i n the bass by the performer, particularly over apex
notes reached by a step (h) or a leap (i), at cadenees and elsewhere,
especially when the following note lies an octave below (y).
Figure 143
- a.

12. Occasionally an unusual fingering must be taken i n order

to keep the best fingers free and ready to perform a mordent preceded by a short note as indicated i n Figure 144. Such a fingering
may be used only i n a modrate tempo. I t is justified by the detached
performance of the dotted note, owing to which the fourth finger
can strike the following tone i n order to leave both the thumb and
second finger i n good position to perform the embellishment. After
the third finger has struck its note i t sufices to shift the hand slightly
to the right. Undotted passages or faster dotted ones should be
played with a normal fingering.
13. The mordent, which, as we have already learned, is often
used to fill out long sustained tones, may be interpolated after a
t r i l l . However, it must be separated from the latter by dividing the
long note into two parts. Without this precaution i t would be wrong
to play the two embellishments i n direct succession, for ornaments
must never be crowded against each other. These remarks are
heeded i n the illustrated execution of Figure 145. The length of
such a mordent is determined by the tempo, which must not be
rapid, for if it is there w i l l be no need for such an expedient.
3

A.

i)UJj}l>lf I H j p
*

b. +

d.

dk

Figure 145
Adagio

11. I n the matter of accidentis this ornament adjusts itself to

circumstances i n the same manner as the t r i l l . Its brilliance is often
increased by raising the lower tone, as i n Figure 144.
Fieure 144

1 A *

C*~

4"

14. I t should be observed that the mordent is the opposite of

the short t r i l l . The latter may be used only over a descending step,
precisely the situation which is unsuited to the mordent. The one
element which they have i n common is that both may be applied to
the interval of a second; ascending i n the case of the mordent and descending i n the case of the short t r i l l . Both employments are clearly
illustrated i n Figure 146.
Figure 146

r r

Dots in certain contexts were performed as rests. T h e practice is described and

criticized in C h . I I I , \.

/} 2

E M i E LIA

Sil

M E N T S

E M RELEI

15. Wliilc discussing inordents, I must make mention of an arl)itrary dccoration often performed by singers in slovv movements
al the beginning and before fermat.e and rests. Characteristic passages and their execution are illustrated i n Figure 147. Since the
tones are identical with those of the mordent and the situation is
one that favors its use (except that as usually performed the ornament would be completed too soon) it may be regarded as a slow
mordent, which, however, has no use aside from these few cases.

149

T S

3 3

=fP^=^4jJ

S II M EN

suave expression

4. A dot often appears after the first small note of the second
type, but the first type is unvariable and appears only in more delibrate tempos when two notes are separated by an ascending leap.
A few characteristic passages are illustrated i n Figure 150.
THE

COMPOUND APPOGGIATURA

1. The compound appoggiatura may be applied to a note i n

two ways: First, the preceding tone is repeated and succeeded by the
step above the principal note; second, the tone below and then the
tone above are prefixed to i t .
2. Both types are clearly recognizable i n the illustrations of Figure 148.

Figure 150

Figure 148

3. The first type is less rapid than the second, but both are
played more softly than the principal tone (Figure 149). Melodies
grow i n attractiveness through the use of this ornament, which
serves to connect notes and, to a degree, ful them out.
2

1 Der Anschlag. T h e Bibliothek der schnen Wissenschaften says; "Marpurg calis

the Anschlag the Doppelvorschlag, which was its more appropriate earlier ame,
for it arises out of two Vorschlage." "Compound appoggiatura" is closer to Marpurg's
term than to Bach's. However, cf. ^1 3 and Note 2.
2 T h i s difference in dynamics between the Anschlag and the Vorschlag ("always
louder than the principal tone") may have been Bach's reason for differentiating
the ames of the two ornaments by means of prefixes. Only one sub-type of Anschlag
agrees with the usual manner of performing the appoggiatura (cf. f 10). Another
reason, however, is that Anschlag indicates unmistakably that the ornament is played
on the beat rather than ahead of it. Remember Bach's difficulties with the Vorschlag!
(Cf. C h . I I , " T h e Appoggiatura," Notes 12, 13).

5. Because the notes of the second type are performed rapidly

it is employed i n fast as well as slow tempos. Figure 151 illustrates a
context especially suited to the compound appoggiatura, for no
other ornament can be successfully applied to the passage. This example with its decoration is good so long as it is played no slower
than andante, although its speed may be considerably increased.
6. I n addition to the preceding example, this second type with
Figure 151

' Andante example from ed. of

1787.

& M liELLI

S H MEN

T S

EMBELLISHMENTS

its Icap of a third may be used in all of the illustrations of Figure

150. As shown in Figure 152, it is also found before notes isolated
by rests (a) and on the repetition of a tone followed by a descending
second (b). I n such a case it is better than the turn, just as the turn
is better before an ascending second, as i n the asterisked example.
I n fact, our decoration may not be used i n this latter context (h).
Further, the compound appoggiatura is better than the turn when,
in a slow tempo, it is placed between tones which stand an augmented second apart, for it softens the dissonant character of the interval (c). I t may also be placed within an ascending second (b), or
seventh (e), and before a descending appoggiatura (/). Thus, i n general, this ornament is better fitted to a melody that descends subsequently rather than one that ascends. Exceptions occur only when
the decorated note is repeated and when the tempo is slow (g).
4

Figure 152

ta
--

fia b J I J W J_1
b
1r>-

II j

#
'

r r
d.

|
1

<

A d

agio

11= )

jg=

Jh~

T
1

r
MI

1J

3J
wrong

J>J

11

. g
n

h t

* T h i s sentence from ed. of 1787.

Example h from ed. of 1787.

execution
notation
9. I t is easy to avoid errors i n performing this dotted ornament
once its origin is known. When a note stands one step above a preceding variable appoggiatura (Figure 154, Example a) and a short
appoggiatura is inserted between them (b), the first appoggiatura
acquires a dot and our embellishment is complete (c), on condition
that there is a subsequent descent of one or more notes.
Figure 154

=N=

7. The dotted compound appoggiatura is notated either as an

ascending appoggiatura or i n the manner of Figure 153. Its divisions
5

Figure 153

d.
K ^hY~h

TL

u3==

c.

135
being variable, they have all been caref ully expressed in the Lessons.
The following note loses as much of its valu as is needed for the
performance of the decoration. Example d illustrates its correct
and incorrect notation.
8. I t never appears i n rapid movements but is well used i n affettuoso passages. Correct uses of i t occur before the repetition of a
tone (Figure 153, Example a), or i n an ascending step (b), both of
which must be followed by a descent, comprising an appoggiatura
(b) or some other note (a). Example a often appears as a caesura i n
adagio movements. The asterisked example of Figure 79 is better
with this ornament than with an appoggiatura, owing to the long
/. Its execution is illustrated i n Example c of the present figure.

b.

v
10. I n performing this kind of compound appoggiatura observe that although the dotted tone is emphasized, the others are
played softly. The second note is connected as rapidly as possible
with the principal tone and all three are slurred.
T h i s sentence from ed. of 1787.
T Example d from ed. of 1787. I n the original of the example marked " r i g h t " the
ornament is notated
. T h i s differs from its notation in E x . a and b and in
Figs. 420-425, thus suggesting a most unfortunate misprint.

/ 76

E M li E E El Sil

MEN

EMBELLISHMEN

T S

TS

11. Figure 155 contains several examples with their execution.

In order to help the performer recognize the usual indication of
this ornament I have purposely retained its inadequate notation
as a simple appoggiatura. The slower the tempo and the more expressive the melody, the longer the dot must be held, as illustrated
in the example marked N.B.

'37
Occasionally the slide is indicated in the manner of Example a, and
frequently it will be found i n large notation (b).
4. The two-toned slide is distinguished from the three-toned
in that (1) it is always used i n a leap which i t helps to fill i n , as i n
Figure 156; the three-toned slide, as we shall see presently, performs
other duties i n addition to this one; (2) the two-toned slide is always
played rapidly (b), the three-toned is not.

Figure 155

Figure 156

^#

# r F 0

TTir-rJp

-"-iH

- E - i

* = t =

TlXf

(i)

m ~>-

a.

b.

5. Figure 157 illustrates the execution of the three-toned type.

Its pace is determined by the character of a movement and the
tempo. Inasmuch as there is no generally accepted symbol for this
ornament and, also, because its pattern is an exact inversin of the
t u r n , 1 find it more convenient to use the symbol of Example b than
to follow the occasionalpractice of writing out the notes i n small
notation. The eyes can more easily assimilate our indication of the
ornament, and it takes up less space.
2

j t -

J -#

^)

-yy{

J,J

m.

Figure 157

b.

a.

0 0
1

TJ t* i
THE

J
9
1

SLIDE

P m

=i=H

' L

1. The slide appears both with and without a dot. Its execution
is suggested by its ame. Melodies are made flowing through its
use.
2. The undotted slide consists of either two or three small notes
which are struck before a principal tone.
3. When i t consists of two notes they are notated as small thirtyseconds i n the manner of Figure 156. I n an alia breve they may also
appear i n the form of sixteenths, as i n the asterisked example.
1 Der

Schleifer.

V0T~0V

6. The three-toned slide is equally at home i n very rapid and

very slow tempos, i n flowing as well as highly expressive movements. Henee i t has two quite opposite employments. I n rapid
pieces it filis out notes and adds sheen. Further, it takes the place
of an unsuffixed, ascending t r i l l which cannot be used because of
the short note vales. Here it is always performed rapidly and, as
illustrated i n Figure 158, the following note may stand i n either
leaping or stepwise relation to it.
2 T h e Bibliothek der Schnen Wissenschaften remarks: "Marpurg divides the turn
more correctly into descending and ascending turns, the latter of which Bach seems
to consider more as a kind of slide. For the rest, however, Bach discusses this ornament with exceptional care." I n fact, Bach's ascending turn is discussed in the section
on that ornament. It is differentiated from the three-toned slide by its symbol and
its greater number of tones. Cf. Ch. I I , " T h e T u r n , " f 37, and in the present section,
f 93 T h i s symbol was not widely adopted. Usually the ornament appears fully written
out in small and large notation.

E M

'38
rigure O H

EMBELLISHMENTS

B E L L I S II M E N T S
I J i

<*>

7. I n its other use it is well fitted for the expression of sadness

in languid, adagio movements. Halting and subdued i n nature, its
performance should be highly expressive, and freed from slavish
dependence on note vales. Its most usual position is over a repeated tone as shown i n Figure 159, Example a-. I n addition i t may
appear after an ascending step or leap (b). I t can be seen that the
three-toned slide resembles a slow compound appoggiatura with its
interval of a third filled i n . Long notes may be expressively divided
and decorated by it in the manner of Example c.

3 9

9. The slide teaches us two things. First, in certain passages the

performer must aim more at an unarfected, subdued expressiveness
than at filling out notes. Therefore he should not always feel
obliged to select only profuse ornaments when decorating slow
notes, for if it were correct to do so, the slide would have to be replaced by the ascending turn, which resembles it. Secondly, and
conversely, it must not be concluded that the fewer the notes i n an
ornament the greater its expressiveness, for then it would follow
that the compound appoggiatura consisting of only two tones is
more expressive than the slide or its equivalent, the filled-in compound appoggiatura.
10. While the three-toned slide is effective i n portraying sadness, the dotted two-toned slide is equally effective i n awakening
more pleasurable feelings.
11. Its notation appears i n Figure 160. No other ornament is so
variable as this one i n its execution, which is determined by the
affect. Therefore, as with the dotted compound appoggiatura, I
have notated it and at times even specified its execution as clearly as
possible i n the Lessons.
4

Figure 159

*L

tznTTu

Figure 160

i J jy

1 1

8. Because the emotions are more stirred by dissonance than

consonance the slide is most frequently found over the former. I n
such cases, it appears over slow notes whose vales are incompletely
filled out on purpose, or completely filled out i n a halting manner.
It appears under similar circumstances i n allegro movements especially where there is a change from major to minor. The chords
which go particularly well with this ornament are the diminished
seventh, the augmented sixth when it contains a fifth, the sixth
with an augmented fourth and minor third, and other similar constructions. Since the behavior of all ornaments is determined largely
by their relation to the accompanying bass, it is easy to conclude,
that this one tends to move downward.

12. Several examples with their various executions appear i n

Figure 161. The asterisked divisin of the slide is better than the
following one because of the bass. Most of the examples present
contexts that are especially fitted to this ornament alone. Certainly,
as an unembellished performance w i l l readily disclose, there is a
need for additional tones, due i n some cases to the harshness of the
dissonances, i n others to the emptiness of the octave. Yet no other
ornament can be inserted as well as the slide. The following tone
usually descends, although, as illustrated i n Example x, the melody
may continu by repeating the final tone of the embellishment.
13. Remaining details of performance are illustrated i n Examples 1 and 2 of Figure 161. These show that although the dotted
tone i n the ornament is emphasized, the two succeeding tones are
* Bach provides helpful information on the normal and extended execution of this
ornament in C h . V I , " T h e Dotted Slide," f 4.
.

L 1 S ll MEN

EMBEL

142

TS

played softly. The dot under the principal tone, e, i n Example 1

informs us that the finger should be released before the termination of that note's written length; consequently, i n Example 2 this
dot has been changed to a rest after the corresponding c.
THE

SNAP

1. Figure 162 illustrates my unvariable notation of the short

mordent i n inversin, the upper tone of which is snapped, the
other tones being played with a stiff finger. Its execution suggests
that this ornament, not mentioned by other writers, might be called
the snap. I n its employment as well as its shape i t is the opposite
of the mordent, but its tones are identical with those of the short
trill.

EMBELLISH

MENTS

H3
descent. Undoubtedly this is because its second tone and the principal tont resemble an inverted suffix. Nevertheless it is different
from all trills i n that i t is never enclosed and never appears under
a slur.
4. I t must be assiduously practiced before it can be made to
sound as i t should. Because only the strongest, most dexterous fingers execute it effectively, i t is often necessary to play the following
tones with a finger that w i l l not interfere with the staccato character of the ornament, as illustrated i n Figure 163, Example a. I t is
often used at caesurae (b).

Figure 163
a.

Figure 162

THE

2. The snap is always played rapidly and appears only before

quick, detached notes, to which it imparts brilliance while serving
to fill them out.
3. I t is i n effect a miniature unsuffixed t r i l l . Unlike the sufnxed
t r i l l , which is best followed by an ascent, the snap is better before a
Der Schneller. Bach's term was not generally adopted. It is doubtful that universal
agreement will ever be reached on the execution and meaning of the English term
inverted mordent, a ame sometimes given to the ornament under discussion (cf.
Elson's Music Dictionary, "Mordent"). It is often played ahead of the beat when it
should be played on the beat, certainly in music of the eighteenth century. Further,
it is just as often made to consist erroneously of the lower auxiliary, its identity
being exchanged with the mordent. Unaccented inverted mordents can be found
among the examples of Figure 161, although they are recognizable as such neither by
their notation or by their ame (dotted slide). T h e Germans are just as badly off
in their attempt to find a ame, as revealed in the following remarle from the Bibliothek der Schnen
Wissenschaften (1763!): " I t seems more correct to us that
Marpurg should regard Bach's snap (Schneller) as a short trill (Pralltriller). But we
would not care to say that both are correct in stating that this brilliantly played
ornament is a kind of short mordent in inversin. T h i s opinin appears quite dubious
upon closer consideration of the nature of the trill and the mordent. Therefore, we
consider Marpurg's criticism well grounded when he condemns those keyboardists
who cali the Schneller a mordent." Bach's Pralltriller and Schneller are far from identical (cf. C h . I I , " T h e T r i l l , " 1 30). Because terminology is so completely confused, it
seemed advisable to take a neutral position and transate Schneller as directly as
possible.
1

2 Remainder of paragraph from ed. of

1787.

ELABORATION

OF

FERMATE

1. Although I have nu desire to discuss embellishments more

elabrate than those already treated, I find i t advisable to say something about them i n connection with fermate.
2. Fermate are often employed with good effect, for they awaken
unusual attentiveness. Their sign is a slur with a dot under it, which
denotes that a tone is to be held as long as required generally by the
nature of the composition.
3. A t times a note without the sign may be held for expressive
reasons. Aside from this, there are three places at which the fermata
appears: over the next to the last, the last, or the rest after the last
bass note. T o be used correctly the sign should be written at the beginning and again at the end of an elaborated fermata.
4. Fermate over rests occur most frequently i n allegro movements and are not embellished. The two other kinds are usually
found i n slow, affettuoso movements and must be embellished if
only to avoid artlessness. I n any event elabrate decoration is more
necessary here than i n other parts of movements.
1

1 P. F . TOS puts it as follows in the English translation (Observations on the Florid

Song, 1743) of his Opinioni: "Every A i r has (at least) three Cadenees, that are all
three final. Generally speaklng, the Study of the Singers of the present Times consists in terminating the Cadenee of the first part with an overflowing of Passages and
Divisions at Pleasure, and the Orchestre waits; in that of the second the Dose is en-

144

& M li

EE E

S II M EN

E M li E L L I SU MEN

T S

T S

'45

5. W i t h tliis in mind I have illustrated both types of fermate

with their elaborations i n Figure 164. A l l of the examples require a
slow or at most a modrate tempo. Since such elaborations must be
related to the affect of a movement, they can be successfully employed only when cise attention is paid to a composition's expressive aim. Other similar cases can be surmised through the figured
bass signatures.
Figure 1 6 4 ^

61

_A

f TT f

f t

1 ^
f" ? r 1
2.

1 m

6 5

1=

rf f

ir

f *f f
~sr

si/

creased, and the Orchestre grows tired; but on the last Cadenee, the Throat is set
a going, like a Weathercock in a Whirlwind, and the Orchestre yawns."

6. Those who lack the ability to introduce elaborations may

apply a long ascending t r i l l when necessary to an appoggiatura
which stands a step above a final tone (Figure 165, Example a).

I<f6

E M li E L L I SU MEN

T S

When the appoggiatura lies a step below, it should be played

simply and the final tone trilled (6). The same applies to a fermata
without an appoggiatura (c).
Figure 165

t m

CHAPTER

T H R E E

PERFORMANCE

m
r

EYBOARDISTS whose chief asset is mere technique are

clearly at a disadvantage. A performer may have the most
agile fingers, be competent at single and double trills, master the art of fingering, read skillfully at sight regardless of the key,
and transpose extemporaneously without the slightest difficulty;
play tenths, even twelfths, or runs, cross the hands i n every conceivable manner, and excel i n other related matters; and yet he may
be something less than a clear, pleasing, or stirring keyboardist.
More often than not, one meets technicians, nimble keyboardists by
profession, who possess all of these qualifications and indeed
astound us with their prowess without ever touching our sensibilities. They overwhelm our hearing without satisfying i t and stun the
mind without moving i t . I n writing this, I do not wish to discredit
the praiseworthy skill of reading at sight. A commendable ability,
I urge its practice on everyone. A mere technician, however, can
lay no claim to the rewards of those who sway i n gentle undulation
the ear rather than the eye, the heart rather than the ear, and lead
it where they w i l l . Of course i t is only rarely possible to reveal the
true content and affect of a piece on its first reading. Even the most
practiced orchestras often require more than one rehearsal of certain pieces which, to judge from the notes, are very easy. Most
technicians do nothing more than play the notes. A n d how the
continuity and flow of the melody suffer, even when the harmony
remains unmolested! I t is to the advantage of the keyboard that
dexterity can be developed beyond the limits of other instruments. But finger velocity must never be misused. I t should be reserved for those passages that cali for it, without advancing the
tempo of the piece as a whole. As proof that I do not disparage
47

/./.v

/ ' E liF O RMAN

CE

P E R F O IMAN

speed, or scorn its uselulness and indispcnsability, 1 p o i n t to the

Lessons i n G a n d F m i n o r and the runs n the C m i n o r Fantasa,
all of w h i c h must be played as r a p i d l y , b u t at the same t i m e as dist i n c t l y as possible. I n certain other countries there is a m a r k e d
tendeney to play adagios too fast a n d allegros too slow. T h e contradictions of such f a u l t y p l a y i n g need n o t be systematically stated.
A t the same t i m e i t must n o t be assumed that I condone those whose
u n w i e l d y fingers give us no choice b u t to slumber, whose cantabile
is a pretense w h i c h hides t h e i r i n a b i l i t y to e n l i v e n the i n s t r u m e n t ,
whose performance, thanks to their lazy fingers, deserves far greater
censure t h a n that addressed to shallow fleetness. A t least the technicians are subject to i m p r o v e m e n t ; their fire can be d a m p e d by expressly checking t h e i r speed. T h e opposite remedy is either n o t at
all or o n l y p a r t i a l l y applicable to the hypochondriac disposition
w h i c h is disclosed, to o u r greater misery, by flabby fingers. B o t h ,
however, p e r f o r m only mechanically; b u t a s t i r r i n g performance
depends o n an alert m i n d w h i c h is w i l l i n g to f o l l o w reasonable
precepts i n order to reveal the content of compositions.
1

2. W h a t comprises good performance? T h e a b i l i t y t h r o u g h singi n g or p l a y i n g to make the ear conscious of the t r u e content a n d
affect of a c o m p o s i t i o n . A n y passage can be so radically changed by
m o d i f y i n g its performance that i t w i l l be scarcely recognizable.
3. T h e subject matter of performance is the loudness a n d softness of tones, touch, the snap, legato a n d staccato execution, the
v i b r a t o , arpeggiation, the h o l d i n g of tones, the r e t a r d a n d aco d e r a n d o . Lack of these elements or i n e p t use of t h e m makes a
poor performance.
3

4. G o o d performance, then, oceurs w h e n one hears a l l notes

and their embellishments played i n correct t i m e w i t h fitting v o l ume produced by a touch w h i c h is related to the t r u e content of a
piece. H e r e i n lies the r o u n d e d , pur, flowing manner of p l a y i n g
w h i c h makes f o r c l a r i t y a n d expressiveness. W i t h these points i n
m i n d , however, i t is u r g e n t that the p e r f o r m e r test his i n s t r u m e n t
i n advance so that he may a v o i d either too heavy or too l i g h t an at1 Sonata I I , third movement, and Sonata V I , first movement. Cf. Pt. I , Introduction, Note 17.
2 Sonata V I , third movement. Cf. Pt. I , Introduction, Note 17, and C h . I I I , Note
11.

' Starcke und Schwache der Tone, ihr Druck, Schnellen, Ziehen, Stossen, Beben,
Brechen, Hallen, Schleppen und Fortgehen.

C E

tack. M a n y instrumenta do not produce a perfect, pur tone unless a

strong touch is ernployed; others must be played l i g h t l y or the volI I I I I C w i l l be excessive. I repeat these remarks, first made i n the I n t r o d u c t i o n , i n order to encourage a more musical way of portraying rage, anger, a n d other passions by means of h a r m o n i o a n d
melodic devices rather than by an exaggerated, heavy attack. I n
r a p i d passages every tone must be played w i t h a fitting pressure or
the effect w i l l be t u r g i d and chaotic. T h e snap, w h i c h was i n t r o duced d u r i n g the discussion of the t r i l l , is usually ernployed i n these
contexts. A w e l l - r o u n d e d manner of performance can be most
readily discerned f r o m the p l a y i n g of r a p i d pieces w h i c h c o n t a i n
alternating l i g h t and heavy runs of equal speed. Keyboardists are
often f o u n d whose ready fingers serve t h e m w e l l i n l o u d runs, b u t
desert t h e m t h r o u g h lack of c o n t r o l i n the soft ones, thereby maki n g f o r indistinetness. T h e y grow nervous, speed o n w a r d , a n d lose
c o n t r o l . I n the E-flat Lesson the broken chords must be played as
distinctly as the runs f o r b o t h hands. I n p e r f o r m i n g bars 24 to 3.4 of
this piece observe the remarks of Paragraph 16 i n the chapter o n
fingering.
T h u s , to facilitate the altrnate use of hands, employ
every slightest pause to b r i n g t h e m to the keys w h i c h are to be struck
i m m e d i a t e l y thereafter.
4

5. I n general the briskness of allegros is expressed by detached

notes a n d the tenderness of adagios by broad, slurred notes. T h e
p e r f o r m e r must keep i n m i n d that these characteristic features of
allegros and adagios are to be given consideration even w h e n a composition is n o t so m a r k e d , as w e l l as w h e n the p e r f o r m e r has n o t yet
gained an adequate understanding of the affect of a w o r k . I use the
expression, " i n general," advisedly, f o r I am w e l l aware that a l l
kinds of execution may appear i n any tempo.
6. T h e r e are many w h o play stickily, as i f they had glue between
their fingers. T h e i r touch is lethargic; they h o l d notes too l o n g .
Others, i n an attempt to correct this, leave the keys too soon, as i f
they b u r n e d . B o t h are w r o n g . M i d w a y between these extremes is
best. H e r e again I speak i n general, for every k i n d of touch has its
use.
7. T h e keyboard lacks the power to sustain l o n g notes and to
decrease or increase the v o l u m e of a tone or, to b o r r o w an apt ex* Remainder of paragraph from ed. of 1787.
5 Sonata V, first movement. Cf. Pt. I , Introduction, Note 17.

iy>

PERFORMANCE

prenion Erom painting, to shade. These conditions make it no small

task to give a singing performance of an adagio w i t l i o u t creating
too m u c h empty space and a consequent m o n o t o n y due to a lack
of sonority; or w i t h o u t m a k i n g a silJy caricature of i t t h r o u g h an
excessive use of r a p i d notes. However, singers a n d performers o n
instruments w h i c h are n o t defective i n this respect also do n o t dar
to deliver an undecorated l o n g note for fear of e l i c i t i n g o n l y bored
yawns. Moreover, the deficiencies of the keyboard can be concealed
under various expedients such as b r o k e n chords. Also, the ear accepts more movement f r o m the keyboard t h a n f r o m other instruments. Henee, satisfactory a n d successful examples of the art of
performance can be presented to all b u t those w h o bear a strong
prejudice against keyboard instruments. A golden mean is difficult
but n o t impossible to discover, p a r t i c u l a r l y i n view of the fact that
our most usual sustaining devices, such as the t r i l l a n d the m o r d e n t ,
are also well k n o w n to other instruments a n d the voice. Such embellishments must be f u l l and so p e r f o r m e d that the listener w i l l
believe that he is hearing only the o r i g i n a l note. T h i s requires a
freedom of performance that rules o u t everything slavish a n d
mechanical. Play f r o m the soul, n o t l i k e a t r a i n e d b i r d ! A keyboardist of such stamp deserves more praise than other musicias.
A n d these latter should be more censured t h a n keyboardists for
bizarre performance.
8. I n order to arrive at an understanding of the t r u e content
and affect of a piece, and, i n the absence of indications, to decide
on the correct manner of performance, be i t slurred, detached or
w h a t not, a h d f u r t h e r , to learn the precautions that must be heeded
i n i n t r o d u c i n g ornaments, i t is advisable that every o p p o r t u n i t y
be seized to listen to soloists and ensembles; the more so because
these details of beauty often depend o n extraneous factors. T h e
v o l u m e and t i m e valu of ornaments must be d e t e r m i n e d by the
affect. I n order to avoid vagueness, rests as w e l l as notes must be
given their exact valu except at fermate a n d cadenees. Yet certain
purposeful violations of the beat are often exceptionally b e a u t i f u l .
However, a d i s t i n c t i o n i n t h e i r use must be observed: I n solo performance a n d i n ensembles made u p of o n l y a few understanding
players, m a n i p u l a t i o n s are permissible w h i c h affect the tempo
itself; here, the g r o u p w i l l be less apt to go astray t h a n to become attentive to and adopt the change; b u t i n large ensembles made u p of

PERFORMANCE

motley players the m a n i p u l a t i o n s must be addressed to the bar

alone w i t h o u t t o u c h i n g o n the broader pace. W h e n a composer
ends a movement i n a foreign key he usually wants the f o l l o w i n g
movement to begin f o r t h w i t h . O t h e r reasons as w e l l may r e q u i r e
an u n i n t e r r u p t e d attack. I t is customary to indcate such a proced u r e by placing only one instead of the usual two-bar Unes at the
end of the movement.
6

9. A l l difficulties i n passage w o r k should be mastered t h r o u g h

repeated practice. Far more troublesome, i n fact, is a good performance of simple notes. These b r i n g f r e t f u l moments to many
who believe that keyboard instruments are easy to play. Regardless
of finger d e x t e r i t y , never undertake more t h a n can be kept u n d e r
c o n t r o l i n p u b l i c performance, where i t is seldom possible to relax
properly or even to m a i n t a i n a fitting disposition. A b i l i t y a n d disposition should be gauged by the most r a p i d and difficult parts i n
order to avoid an overexertion, w h i c h w i l l surely result i n a breakd o w n of the performance. Those passages w h i c h are troublesome
i n prvate and come off w e l l o n l y occasionally should be o m i t t e d
f r o m p u b l i c performance unless the performer finds himself i n a
p a r t i c u l a r l y favorable frame of m i n d . Also, the i n s t r u m e n t should
be tested beforehand w i t h t r i l l s a n d other ornaments. T h e r e are
two reasons for these several precautions: they w i l l assure an agreeable, flowing performance; they w i l l help to remove the anxious
m i e n w h i c h , far f r o m e n l i s t i n g the listener's sympathy, w i l l o n l y
annoy h i m .
10. T h e pace of a composition, w h i c h is usually indicated by
several w e l l - k n o w n I t a l i a n expressions, is based o n its general content as w e l l as o n the fastest notes a n d passages contained i n i t .
D u e consideration of these factors w i l l prevent an allegro f r o m bei n g rushed a n d an adagio f r o m b e i n g dragged.
11.
Every step must be taken to remove accompanying parts
f r o m the h a n d that performs the p r i n c i p a l melody so that i t may
be played w i t h a free, unhampered expression.
12. As a means of l e a r n i n g the essentials of good performance i t
is advisable to listen to accomplished musicias, as stated i n Paragraph 8. A b o v e a l l , lose n o o p p o r t u n i t y to hear artistic singing. I n
so d o i n g , the keyboardist w i l l learn to t h i n k i n terms of song. I n deed, i t is a good practice to sing i n s t r u m e n t a l melodies i n order
6

Remainder of paragraph from ed. of

1787.

PERFORMANCE

/' E R E O R M A N C E
to rcach an understanding o their correct performance. T h i s way of
learning is of lar greater valu t h a n the reading of v o l u m i n o u s
tomes or listening to learned discourses. I n these one meets such
terms as N a t u r e , Taste, Song, a n d M e l o d y , although their authors
are often incapable of p u t t i n g together as many as t w o n a t u r a l ,
tasteful, singing, melodic tones, f o r they dispense t h e i r alms a n d
endowments w i t h a completely unhappy arbitrariness.
7

13. A musician cannot move others unless he too is moved. H e

must of necessity feel a l l of the affeets that he hopes to arouse i n his
audience, f o r the revealing of his o w n h u m o r w i l l stimulate a l i k e
h u m o r i n the listener. I n languishing, sad passages, the performer
must languish and grow sad. T h u s w i l l the expression of the piece be
more clearly perceived by the audience. H e r e , however, the error of
a sluggish, dragging performance must be avoided, caused by an excess of affect a n d melancholy. S i m i l a r l y , i n lively, joyous passages,
the executant must again p u t himself i n t o the appropriate m o o d .
A n d so, constantly v a r y i n g the passions he w i l l barely q u i e t one
before he rouses another. A b o v e a l l , he must discharge this office i n
a piece w h i c h is h i g h l y expressive by nature, whether i t be by h i m
or someone else. I n the latter case he must make certain that he assumes the e m o t i o n w h i c h the composer i n t e n d e d i n w r i t i n g i t .
I t is p r i n c i p a l l y i n improvisations or fantasas that the keyboardist
can best master the feelings of his audience. Those w h o m a i n t a i n
that a l l of this can be accomplished w i t h o u t gesture w i l l retract
their words w h e n , o w i n g to t h e i r o w n insensibility, they find themselves obliged to sit l i k e a statue before t h e i r i n s t r u m e n t . U g l y
grimaces are, of course, i n a p p r o p r i a t e a n d h a r m f u l ; b u t fitting expressions help the listener t o understand o u r meaning. Those opposed to this stand are often incapable of d o i n g justice, despite
their technique, to t h e i r o w n otherwise w o r t h y compositions.
U n a b l e to b r i n g o u t the content of their works, they r e m a i n i g n o r a n t of i t . B u t let someone else play these, a person of delicate,
8

7 T w o specimens appeared serially in Marpurg's Der Critischer Musicus an der

Spree. Both were translations from the French. T h e first, Grandvall's Essay on Good
Taste in Music, started on June 3, 1749. Later, starting December 2, 1749, an Essay
on the Decline of Good Taste in Music by Bollioud de Mermet began. Both contain
terms similar to those mentioned here and are marked by " a completely unhappy arbitrariness."
T h i s sentence appeared as a footnote in the ed. of 1787.
Marpurg (op. cit., Sept. 9, 1749) i n covering similar material writes, " I know a
great composer [Bach?] on whose face one can see depicted everything that his music
expresses as he plays it at the keyboard."
8

lj}

sensitive insighl w h o knows the meaning of good performance, a n d

the composer. w i l l learn to his astonishment that there is more i n
his music than he had ever k n o w n or believed. G o o d performance
can, i n fact, i m p r o v e and gain praise for even an average composition.
14. I t can be seen f r o m the many affeets w h i c h music portrays,
that the accomplished musician must have special endowments a n d
be capable of e m p l o y i n g t h e m wisely. H e must carefully appraise
his audience, their a t t i t u d e t o w a r d the expressive content of his
p r o g r a m , the place itself, a n d other a d d i t i o n a l factors. N a t u r e has
wisely p r o v i d e d music w i t h every k i n d of appeal so that a l l m i g h t
share i n its enjoyment. I t thus becomes the d u t y of the performer
to satisfy to the best of his a b i l i t y every last k i n d of listener.
15. As stated earlier, i t is especially i n fantasas, those expressive n o t of memorized or plagiarized passages, b u t rather of true,
musical creativeness, that the keyboardist more than any other
executant can practice the declamatory style, and move audaciously
f r o m one affect to a n o t h e r . A short example is sketched i n the
final Lesson. As usually notated, c o m m o n t i m e is indicated b u t n o t
prescribed for the divisions of the entire piece. For this reason, bar
lines are always o m i t t e d . N o t e lengths are d e t e r m i n e d by the usual
superscribed moderato, a n d by the s u r r o u n d i n g vales. T r i p l e t s
can be recognized simply by t h e i r beam. U n b a r r e d free fantasas
seems especially adept at the expression of affeets, f o r each meter
carries a k i n d of compulsin w i t h i n itself. A t least i t can be seen i n
accompanied recitatives that tempo a n d meter must be f r e q u e n t l y
changed i n order to rouse a n d s t i l l the r a p i d l y a l t e r n a t i n g affeets.
Henee, the metric signature is i n many such cases more a c o n v e n t i o n
of n o t a t i o n than a b i n d i n g factor i n performance. I t is a distinct
m e r i t of the fantasa that, u n h a m p e r e d by such trappings, i t can accomplish the aims of the recitative at the keyboard w i t h complete,
unmeasured freedom.
10

11

16. Performers, as we have already learned, must t r y to capture

the true content of a composition and express its appropriate affeets.
Composers, therefore, act wisely w h o i n n o t a t i n g t h e i r works i n 10 For a more extended discussion of the free fantasa cf. C h . V I I .
11 T h i s sentence is from the ed. of 1787. T h e Lesson mentioned is the third movement of Sonata V I (cf. Pt. I , Introduction, Note 17). For a score set with a double text
that was added later by the poet and Bach admirer, Heinrich Wilhelm von Gerstenberg, see F. Chrysander i n Vierteljahrschrift fr Musikwissenschaft,
No. 7, 1891.

PERFORMANCE

elude terms, in a d d i l i o n to tempo indications, which lielp to elarify

the meaning o a piete. However, as w o r t h y as their intentions
m i g l i t be, they w o u l d n o t succeed i n p r e v e n t i n g a garbled performance i f they d i d not also add to the notes the usual signs and
marks relative to execution. W i t h regard to the first p o i n t , I hope
that I shall be f o r g i v e n for using a few unusual terms w h i c h , however, fitted the m e a n i n g that I wanted to express i n the Lessons. I
have attended to signs and marks w i t h lavish care, for I k n o w that
they are as m u c h needed by keyboardists as by other executants.
W h e n one voice is to be b r o u g h t o u t above the others i t carries
p e r t i n e n t markings. I n other cases such signs refer to a l l parts
played by a given h a n d . T h e appearance of these signs being better
k n o w n than t h e i r m e a n i n g and execution, I shall illustrate and
e x p l a i n the more usual of t h e m i n the f o l l o w i n g paragraphs.
17. Attack and touch are one and the same t h i n g . E v e r y t h i n g
depends o n their forc and d u r a t i o n . W h e n notes are to be detached
f r o m each other strokes or dots are placed above t h e m , as illustrated i n Figure 166. T h e latter i n d i c a t i o n has been used i n the
Lessons i n order to avoid a confusin of the strokes w i t h fingering
numeris. Notes are detached w i t h r e l a t i o n to: (1) t h e i r notated
length, that is, a half, quarter, or eighth of a bar; (2) the tempo,
fast or slow; and (3) the v o l u m e , forte or piano. Such tones are always h e l d for a l i t t l e less than half of their notated l e n g t h . I n general, detached notes appear mostly i n leaping passages a n d r a p i d
tempos.
Figure

%*

self-evident that all of the tones are to be played s i m i l a r l y u n t i l

another k i n d of m a r k intervenes. T h e slurred tones of b r o k e n
chords are h e l d i n the manner of Figure 168. T h i s applies to the E
major Lesson, where, i n a d d i t i o n to the fine effect of such an execut i o n , the advantages of this k i n d of i n d i c a t i o n are clearly evident. I n
the A-flat Lesson
the same execution has been notated i n the
French manner, w h e r e i n each tone of a chord stands for a seprate
voice. I have w r i t t e n i t this way so that the student m i g h t have an
o p p o r t u n i t y to become f a m i l i a r w i t h b o t h kinds of n o t a t i o n . Generally speaking, slurred notes appear mostly i n stepwise passages
and i n the slower or more modrate tempos. Passages i n w h i c h
passing notes or appoggiaturas are struck against a bass are played
legato i n a l l tempos even i n the absence of a slur (168 a). As illustrated i n the example marked N . B . , the same remark applies to
basses w h i c h are s i m i l a r l y devised. Note-against-note successions
may be either slurred or detached and therefore r e q u i r e express i n dications. Successions of thirds l i k e those i n Example b w o u l d be almost impossible to p e r f o r m i n a fast tempo. A t least, a r a p i d execut i o n m i g h t very easily cause stammering or a detaching of the notes
contrary to the composer's intentions. However, played i n the manner of the asterisked example w i t h the quarters h e l d for their f u l l
valu, the desired effect can be easily produced on the clavichord
as w e l l as the harpsichord. Likewise the succession i n Example c
may be played r a p i d l y i n the manner of Example d.
12

1 3

1 4

m \rrn

Figure 167

166

I '

18. Notes w h i c h are to be played legato must be h e l d for their

f u l l l e n g t h . A slur is placed above t h e m i n the manner of Figure
167. T h e slur applies to a l l of the notes i n c l u d e d under its trace.
Patterns of t w o and f o u r slurred notes are played w i t h a slight,
scarcely noticeable increase of pressure o n the first and t h i r d tones.
T h e same applies to the first tones of groups of three notes. I n other
cases only the first of the slurred notes is played i n this manner. I t is
a convenient custom to indcate by appropriate marks o n l y the first
few of prolonged successions of detached or legato notes, i t b e i n g

155

PERFORMANCE

in. g

Figure 1 6 8

15

sr.B.

1 Sonata I I I , third movement. Cf. Pt. I, Introduction, Note 17.

Sonata V I , second movement. Cf. Pt. I , Introduction, Note 17.
1* Remainder of paragraph from ed. of 1787.
is Examples a, b, c, d from ed. of 1787.
2

1 3

J i

Allegro

I T
y

* d

21. T h e notes of Figure 170 are played i n such a manner that

the first of each slur is slightly accented. Figure 171 is played similarly except that the last note of each slur is detached. T h e finger
must be raised immediately after i t has struck the key. T h e portato
and v i b r a t o of Figure 169 apply only to the clavichord; Figures
170 and 171 may be played o n b o t h the harpsichord and the clavichord, b u t
more effectively o n the latter. T h e execution of
Figures 170 and 171 must n o t be c o r r u p t e d i n t o that of Figure 171,
Example a, an error frequently c o m m i t t e d by beginners.

assai

757

PERFORMANCE

/' /< /{ F O R M A N C E

/ 6

2 0

T"'TI

Figure 170

19. T h e notes of Figure 169 are played legato, b u t each tone

is noticeably accented. T h e t e r m w h i c h refers to the performance of
notes that are b o t h slurred and d o t t e d is portato.
20. A long, affettuoso tone is p e r f o r m e d w i t h a v i b r a t o . T h e
finger that depresses and holds the key is gently shaken. T h e sign
of a v i b r a t o appears i n Example a. T h e
best effect is achieved
w h e n the finger w i t h h o l d s its shake u n t i l half the valu of the note
has passed.
16

17

18

1 9

Figure 169

Figure 171

22. Tones w h i c h are neither detached, connected, or f u l l y

held are sounded for half their valu, unless the abbreviation
Ten. (hold) is w r i t t e n over t h e m , i n w h i c h case they must be h e l d
f u l l y . Quarters and eighths i n modrate and slow tempos are usually
p e r f o r m e d i n this semidetached manner. T h e y must not be played
weakly, b u t w i t h fire and slight accentuation.
23. Short notes w h i c h f o l l o w dotted ones are always shorter i n
execution than their notated l e n g t h . Henee i t is superfluous to
place strokes or dots over t h e m . Figure 172 illustrates their execut i o n . T h e asterisked example shows us that occasionally the divisin
must agree w i t h the notated vales. Dots after l o n g notes or after
short ones i n slow tempos, and isolated dots, are a l l h e l d . However,
i n r a p i d tempos prolonged successions of dots are p e r f o r m e d as
rests, the apparent opposite demand of the n o t a t i o n n o t w i t h s t a n d i n g . A more aecurate n o t a t i o n w o u l d remove such a discrepaney.
L a c k i n g this, however, the content of a piece w i l l shed l i g h t o n the
details of its performance. Dots after short notes f o l l o w e d by
groups of shorter ones are h e l d f u l l y (Figure 173). W h e n
four
or more short notes f o l l o w a dot they are played w i t h dispatch,
there being so many of t h e m . T h e same applies to Example a and,
w h e n the tempo is n o t too slow, to Example b. Short notes, w h e n
they precede d o t t e d ones, are also played more r a p i d l y than t h e i r
21

is Das Tragen der Tone.

17 Die Bebung. T h i s and the preceding Tragen der Tone are clostly related. For
one thing, the true and only keyboard home of both is the clavichord. For another,
they are executed in the same manner, the difference being solely in the number of
times the key is pressed after the finger stroke. T h i s difference is indicated roughly
by the number of dots over each note. Franz Rigler (Anleitung zum Clavier, Vienna,
1779) explains the distinction by saying that the portato arises "when the key is
rather slowly rocked," and the vibrato, "when the tone is quite clearly rocked
(herausgewieget) according to the number of dots, and without repeating the finger
stroke." Daniel Gottlob Trk (Clavierschule, 1789) explains the portato as " a joining
of tones in such a manner that in progressing from one to another no break occurs.
At the clavichord [Claviere] this so-called Tragen is easy to express, for after striking
the key an additional pressure is exerted."
is Dr. Charles Burney in The Present State of Music in Germany, Vol. I I , p. 268, describes Bach's vibrato (Bebung) as follows: " I n the pathetic and slow movements, whenever he had a long note to express, he absolutely contrived to produce, from his instrument, a cry of sorrow and complaint, such as can only be effected upon the clavichord;
and perhaps by himself."
is T h i s sentence from ed. of

1787.

2 2

2 Remainder of sentence from ed. of 1787, footnote.

21 However, cf. C h . V I , "Performance," f 15.
22 Remainder of paragraph from ed. of 1787.

i 8

PERFORMANCE

PERFORMANCE

n o t a t i o n indicates. A l l of the short notes of Example c, even the

sixteenths, when the tempo is n o t too slow, f o l l o w this r u l e . I t w o u l d
be a better practice to add a beam to a l l of the notes. I t is only generally t r u e that the short notes described here should be played
r a p i d l y , for there are exceptions. T h e melodies i n w h i c h they appear should be carefully examined. Should ornaments of length,
such as the t r i l l or t u r n , appear over them, t h e i r performance must
be broader t h a n that of undecorated short notes. Likewise, i n sad
or expressive passages and i n slow tempos the exception is less accelerated t h a n i n other cases.

J'J- JI
-HT

J3

J- J

J-d

J'

Figure 174

II j

25. T h e p e r f o r m e r may break a l o n g tied note by r e s t r i k i n g the

key (Figure 175). Occasionally a short tie must be b r o k e n i n order
to clarify the leading of a voice (a). W h e n i n legato passages a voice
is assigned to a tone directly after i t has been taken by-a held note,
the h o l d should n o t be b r o k e n . Rather than vilate the legato, the
second tone's c l a i m should be denied, for such notes are o f t e n
w r i t t e n merely for the sake of n o t a t i o n (b). Should the t w o tones be
w e l l separated, the second must be struck, b u t i n such a manner
that the r i g h t h a n d w i l l regain the key before the left h a n d has released i t (c). I f the l o n g note is t r i l l e d i t must n o t be b r o k e n .
2 4

Figure 1 7 5

*J0 f*-

Figure 173 i

*59

JjJ

25

|J

J J ^ i V

%z

2:

J. n *

7^

' 0m

1 II

0'*

\ jrpl \

J-

Ai,
T

24. T h e first notes of Figure 174, b e i n g slurred, are n o t played

too r a p i d l y i n a modrate or slow tempo. I f they are, an excess of
u n f i l l e d space w i l l f o l l o w their execution. T h e first note is accented
by means of gentle pressure, b u t n o t by a sharp attack or a r a p i d relase.

f "

Tff*

26. T h e usual signs of arpeggiation a n d t h e i r execution appear

i n Figure 176. T h e asterisked example represents an arpeggio w i t h
an acciaccatura. T h e w o r d " a r p e g g i o " w r i t t e n over a l o n g note
calis f o r a c h o r d b r o k e n u p w a r d and d o w n w a r d several times.
Remainder of paragraph from ed. of
2" Kxamplcs a, b, c from ed. of 1787.
, i

-' Examplei a, l>, c, from cd. of 1787.

tfJ3

1787.

/' E R l< O R MAN

PERFORMANCE

C E

where a melody i n octaves is transposed three times against r a p i d

notes i n the left hand. Each transposition can be effectively perormed by gradually and gently accelerating and immediately thereafter r e t a r d i n g . I n affettuoso playing, the performer must avoid
frequent and excessive retards, w h i c h tend to make the tempo
drag. T h e affect itself readily leads to this fault. Henee every effort
must be made despite the beauty of detail to keep the tempo at the
end of a piece exactly the same as at the b e g i n n i n g , an extremely difficult assignment. T h e r e are many excellent musicians, b u t only a
few of w h o m i t can be said t r u t h f u l l y that i n the narrowest sense
they end a piece as they began i t . Passages i n a piece i n the major
mode w h i c h are repeated i n the m i n o r may be broadened somewhat
on their r e p e t i t i o n i n order to heighten the affect. O n entering a
jermata expressive of languidness, tenderness, or sadness, i t is customary to broaden slightly. T h i s brings us to the tempo r u b a t o . Its
i n d i c a t i o n is simply the presence of more or fewer notes than are
contained i n the n o r m a l divisin of the bar. A whole bar, p a r t of
one, or several bars may be, so to speak, distorted i n this manner.
T h e most difficult b u t most i m p o r t a n t task is to give a l l notes of the
same valu exactly the same d u r a t i o n . W h e n the execution is such
that one h a n d seems to play against the bar and the other strictly
w i t h i t , i t may be said that the performer is d o i n g everything that
can be r e q u i r e d of h i m . I t is only rarely that a l l parts are struck
simultaneously. T h e b e g i n n i n g of a caesura w h i c h terminates a
tempo r u b a t o may be d r a w n i n t o the m a n i p u l a t i o n , b u t the end,
as i n a l l endings of this tempo, must f i n d all parts together over the
bass. Slow notes and caressing or sad melodies are the best, and dissonant chords are better than consonant ones. Proper execution of
this tempo demands great c r i t i c a l faculties and a h i g h order of
sensibility. H e w h o possesses these w i l l n o t find i t difficult to fashion
a performance whose complete freedom w i l l show n o trace of
coercin, and he w i l l be able to maniplate any k i n d of passage.
However, practice alone w i l l be of no help here, for w i t h o u t a
fitting sensitivity, no a m o u n t of pains w i l l succeed i n c o n t r i v i n g a
correct rubato. As soon as the upper part begins slavishly to f o l l o w
the bar, the essence of the r u b a t o is lost, for then a l l other parts must
be played i n t i m e . Other instrumentalists and singers, w h e n they
are accompanied, can i n t r o d u c e the tempo m u c h more easily t h a n
2 9

27. W i t h the advent of an increased use of triplets i n c o m m o n

or 4/4 t i m e , as w e l l as i n 2/4 and 3/4, many pieces have appeared
w h i c h m i g h t be more conveniently w r i t t e n i n 12/8, 9/8, or 6/8. T h e
performance of other lengths against these notes is shown i n Figure
177. T h e unaccented appoggiatura, w h i c h is often disagreeable
and always difficult, can be avoided i n the ways i l l u s t r a t e d i n these
examples.
26

27

Figure 177

28. Figure 178 contains several examples i n w h i c h certain notes

and rests should be extended beyond t h e i r w r i t t e n l e n g t h , f o r affective reasons. I n places, I have w r i t t e n o u t these broadened vales;
elsewhere they are indicated by a small cross. Example a shows how
a r e t a r d may be a p p l i e d o p p o r t u n e l y to a melody w i t h t w o different
accompaniments. I n general the r e t a r d fits slow or m o r e modrate
tempos better t h a n very fast ones. T h e r e are m o r e examples i n the
opening allegro a n d the f o l l o w i n g adagio of the B M i n o r Sonata,
N o . 6, of m y second engraved w o r k ;
especially i n the adagio,
2 8

26 T h i s discrepant practice of eighteenth-century notation offers many problems

to the modern performer. See, for example, J . S. Bach's E minor Partita, Tempo di
Gavotta, and the E minor Fugue, Bk. I I , The Well-tempered
Clavier.
27 Cf. C h . I I , " T h e Appoggiatura," ^ 25 and Note 12.
28 T h e Wrttemberg Sonatas (Nagels Archiv, Nos. 21, 22). T h e y were preceded by
the first published works, the Prussian Sonatas, two years earlier, in 1742 (Nagels
Archiv Nos. 6, 15).

161

29 Remainder of paragraph from ed. of 1787.

,(,

p i<:

/'

RF O R M A N C /<;

the solo kcyboardist. T h e reason for this is the one just stated. I f
necessary, the solo keyboardist may alter the bass, b u t not the harmony. Most keyboard pieces contain r u b a t o passages. T h e divisin
and i n d i c a t i o n of these is about as satisfactory as can be expected.
H e w h o has mastered the tempo r u b a t o need n o t be fettered by the
numeris w h i c h d i v i d e notes i n t o groups of 5, 7, 11, etc. A c c o r d i n g
to his disposition b u t always w i t h appropriate freedom he may add
or o m i t notes.
80

R l< O I{ M A N C E

163

hand there must not be too m u c h restraint. I t is not possible to describe the contexts appropriate to the forte or piano because for
every case covered by even the best r u l e there w i l l be an exception.
T h e p a r t i c u l a r effect of these shadings depends o n the passage, its
context, and the composer, w h o may i n t r o d u c e either a forte or a
piano at a given place for equally c o n v i n c i n g reasons. I n fact, complete passages, i n c l u d i n g t h e i r consonances and dissonances, may
be m a r k e d first forte and, later, piano. T h i s is a customary procedure w i t h b o t h repetitions and sequences, particularly w h e n the
accompaniment is m o d i f i e d . B u t i n general i t can be said that dissonances are played l o u d l y and consonances softly, since the f o r m e r
rouse o u r emotions and the latter q u i e t them
(Figure 179, Example a). A n exceptional t u r n of a melody w h i c h is designed to crate a
v i o l e n t affect must be played l o u d l y . So-called deceptive progressions are also b r o u g h t o u t markedly to complement t h e i r f u n c t i o n
(b). A n o t e w o r t h y r u l e w h i c h is n o t w i t h o u t f o u n d a t i o n is that a l l
tones of a melody w h i c h lie outside the key may w e l l be emphasized
regardless of whether they f o r m consonances or dissonances and
those w h i c h lie w i t h i n the key may be effectively p e r f o r m e d piano,
again regardless of t h e i r consonance or dissonance (c). Because of
the brevity of E x a m p l e c, I have been obliged to crowd the forte a n d
piano indications. I k n o w that this constant changing f r o m l i g h t to
dark shadings is of n o valu, f o r i t leads to obscurity rather t h a n
clarity and i n the end turns a s t r i k i n g relationship i n t o an o r d i n a r y
one. A l t h o u g h each forte and piano i n the Lessons has been carefully
marked, i t is i m p o r t a n t to keep i n m i n d that certain ornaments as
discussed i n the chapter on embellishments are very m u c h charac3 2

29. P. means piano or soft; t w o or more of the letters standing

together denote greater softness. M . F . means mezzo forte or half
l o u d . F. means f o r t e ; to denote greater loudness t w o or m o r e of
the letters are placed together. I n order to c o n t r o l a l l shades f r o m
pianissimo to fortissimo the keys must be g r i p p e d firmly a n d w i t h
s t r e n g t h . However, they must n o t be flogged; b u t o n the other
31

so Le., it is easier for two performers to play in contrived disagreement than it is

for the two hands of a single performer.
31 Throughout this paragraph Bach is speaking of graded as well as terraced dy-

namics. T h e terms crescendo and diminuendo appear in his later compositions, but
only sparingly. Modern signs for graded changes were only evolving in his time. Cf.
R . E . M. Harding, Origins of Musical Time and Expression, Oxford Press, 19558,
C h . I V , and Harvard Dictionary of Music, Cambridge, Mass., 1945, "Expression," I I I .
Bach's older practice is the use of successive abbreviations, such as ff, f, p, pp, or a
more widely spaced ff., pp.
32 Bach writes here with reference to an elabrate theory of shading advanced by
Quantz (Versuch einer Anweisung die Flote traversiere zu spielen, 1752). It appears
in translation in Arnold, Art of Accompaniment from a Thoroue;h-Bass, pp. 407 ff.
T h e leading point of the theory is that the dynamic level at which chords are to be
6
played is determined by the kind of dissonances that they express. For example, 4*
6
6
2
is played mezzo-forte; 4f , forte;4v; fortissimo. Bach has many reservations; so many,
that he accepts the theory only in its broadest sense, relieved of all particulars. I n
this broadest sense it represents a common practice of the eighteenth century.

V E li /' O l\ A N C E

I(>l

PERFORMANCE

terized l)y dynanik: shadings. I I (lie Lessons are played on a harpsic h o r d w i t h two manuals, only one manual should be used t o play
detailed changes of forte and piano. I t is only when entire passages
are differentiated by contrasting shades that a transfer may be
made. T h i s p r o b l e m does n o t exist at the clavichord, for o n i t a l l
varieties of l o u d and soft can be expressed w i t h an almost u n r i v a l e d
clarity a n d p u r i t y . A l o u d , boisterous accompaniment must always
be balanced by a stronger melodic t o u c h .
33

34

Figure 179

a.

^
i>

f p

-1

*t

p / p f p

fe|K

#6

f p

Pf P

ff

P
P f7 P

i r n J Jf iJp J 7 Jp J f Jp JJ
f P
P

9 8
4 3

i(>

the general pace a n d difcrentiation of notes. T h e r e is always a

slight pause between statements i n two- and three-voice cadenzas
before a new voice enters. I n the Eessons I have indicated these
held endings w i t h whole notes instead of the more usual ties. T h e
w h i t e notes serve n o other purpose a n d are t o be h e l d u n t i l relieved by another note i n the same voice. Observe that i f another
note is assigned to the key oceupied by one of the whole notes, the
key must be r e l i n q u i s h e d , b u t reoecupied by the o r i g i n a l note after
the i n t e r v e n i n g voice has left. Should b o t h hands be engaged i n this
procedure, the o r i g i n a l hand must retake the key before i t has been
released finally by the other. I n this way c o n t i n u i t y of sound w i l l be
achieved w i t h o u t an a d d i t i o n a l attack. W i t h regard t o the l e n g t h
of the pauses indicated by w h i t e notes, imagine t w o or three persons i n conversation, each one of w h o m waits for the other to complete his statement before r e j o i n i n g w i t h his o w n . Played any other
way, the cadenza loses its distinguishing character and sounds more
l i k e a clearly measured and barred piece w i t h tied notes. A t t h e
same t i m e the pause is n o t observed w h e n the r e s o l u t i o n of a c h o r d
preceding the w h i t e note must be played by the other voices d i r e c t l y
on the w h i t e note's entrance.
3 6

37

31.
T h e F m a j o r Lesson is an i l l u s t r a t i o n of the present practice of varying extemporaneously the two reprises of an allegro.
T h e concept is excellent b u t m u c h abused. M y feelings are these:
N o t everything should be varied, for i f i t is the reprise w i l l become
a new piece. M a n y things, p a r t i c u l a r l y affettuoso or declamatory
passages, cannot be readily varied. Also, galant n o t a t i o n is so replete w i t h new expressions and twists that i t is seldom possible even
to comprehend i t immediately. A l l variations must relate to the
piece's affect, and they must always be at least as good as, i f n o t
better than, the o r i g i n a l . For e x a m p l e , many variants of melodies
i n t r o d u c e d by executants i n the belief that they h o n o r a piece,
actually oceurred t o the composer, w h o , however, selected a n d
wrote d o w n the o r i g i n a l because he considered i t the best of its
k i n d . Simple melodies can o f t e n be made i n t o elabrate ones a n d
3 8

30. Elaborated cadenees [cadenzas]

are l i k e improvisations.
I n keeping w i t h the substance of a piece they are p e r f o r m e d freely
i n an unmeasured manner. T h e notated lengths, therefore, at such
cadenees i n the Lessons are only approximate and represent merely
3 5

33 Johann F r . Cramer, in his Magazin der Musik (Vol. I , p. 1217), wrote, " A l l who
have heard Bach play the Clavichord must have been struck by the endless nuances of
shadow and light that he casts over his performance."
T h i s material is discussed at length i n C h . V I , "Performance," f f 5-1335 T h e common eighteenth-century term was "cadenee," which had several meanings, all but one of which have dropped out f use. T h e Italian word "cadenza,"
which carne to acquire a specific, and henee clearer, meaning, has been adopted here.
I n defense, a quotation from Quantz (op. cit., C h . X V , f i) revealing the varied
meanings of "cadenee" i n his own day, should suffice: " B y the word 'cadenee' I understand here neither the end or the interruption of a melody, much less the trill
which is called 'cadenee' by some Frenchmen. I shall treat only those elabrate embellishments which are furnished by a concertizing part out of free will and pleasure
at the conclusin of a piece over the penultimate bass note, namely the fifth of the
key of the piece." Bach's discussion of such cadenzas from the point of view of the
accompanist appears in C h . V I , "Closing Cadenees." T h e soloist's cadenzas are treated
in the last section of C h . I I .

39

38 Sonata I V , second movement, and Sonata V I , second movement. Cf. Pt. I , I n troduction, Note 17. A similar dialogue cadenza appears in J . S. Bach's Werke, 36.4,
p. 31, where, in the D major Toccata, the section marked con discrezione begins.
Quantz gives many examples (op. cit. X V . Haupstck, f f 19-31).
37 Cf. f 25 of the present chapter.
38 Sonata V , third movement. Cf. Pt. I , Introduction, Note 17.
3 T h i s sentence from ed. of 1787, footnote.

l>

E O R M A N C E

vice versa. A l l this mus be clone w i l h no sinall deliberacin. Consum attention mus be given to preceding and succeeding parts;
(here must be a visin of the whole piece so that the v a r i a t i o n w i l l
retain the o r i g i n a l contrasts of the b r i l l i a n t a n d the simple, the
fiery and the l a n g u i d , the sad and the j o y f u l , the vocal and the i n strumental. I n keyboard pieces the bass too may be m o d i f i e d so l o n g
as the h a r m o n y remains unchanged. Despite the present p o p u l a r i t y
of elabrate variations, i t is of first importance always to make cert a i n that the lineaments of a piece, by w h i c h its affect is recognized,
remain unobscured.
40

4" Of interest in this connection is Bach's Foreword to the first collection of

Sonatas with Varied Reprises (1760). H e writes: "Today varied reprises are indispensable, being expected of every performer. A friend of mine takes every last pain to
play pieces as written, purely and in accord with the rules of good performance. C a n
applause be rightfully denied him? Another, often driven by necessity, hides under
bold variations his inability to express the notes as written. Nevertheless, the public
holds him above the former. Performers want to vary every detail without stopping
to ask whether such variation is permitted by their ability and the construction of
the piece.
"Often it is simply the varying, especially when it is allied with long and much too
singularly decorated cadenzas, that elicits the loudest acclaim from the audience.
And what abuses of these two refinements arise! No longer is there patience enough
to play the first part of the piece as written; the long delay of the Bravos would be
unendurable. Often these untimely variations are contrary to the construction, the
affect, and the inner relationship of the ideasan unpleasant matter for many composers. Assuming that the performer is capable of varying properly, is he always in
the proper mood? Do not many new problems arise with unfamiliar works? Is not
the most important consideration in varying, that the performer do honor to the
piece? Must not the ideas that he introduces into the repetition be as good as the
original ones? Yet, regardless of these difficulties and abuses, good variation always
retains its valu." T w o later sets of pieces with varied reprises from 1766 and 1768
were published in 1914 by Universal Edition (No. 5395) under the title Klavierstcke.

FOREWORD TO PART
bcr

>te

iDQt)re

TWO

3frt

>as Slamer p fpteteii

H A S finally become m y pleasure to present the second part

of this Essay to m y friends a n d admirers. M y o r i g i n a l i n t e n t i o n

was to engrave the musical examples o n copper plates, a n d I

made a start w i t h a fantasia w h i c h has been appended to the last
pages. However, I changed m y m i n d later a n d chose the excellent
i n v e n t i o n of music p r i n t i n g so that illustrative matter m i g h t appear i n the text, thus e l i m i n a t i n g the tedious search for examples
1

in rocfclxm >ie &)re t>on bem 3lccompanemmt

un)

fcer

frecen

^antajte

in seprate tables. T h e most notable feature of this book is the

a t t e n t i o n given to artistic accompaniment, and i n this respect
it differs f r o m a l l previous manuals o n t h o r o u g h bass. T h e observations are n o t speculative b u t rest o n experience and w i s d o m . W i t h
no desire to boast, i t may be said that this experience can h a r d l y be

abge^antiflt roir.

9t6ft iner jtupfmafel.

2n

5Ser(egung

6(6 2(uctori.

r i v a l e d , for i t has g r o w n o u t of many years of association w i t h good

taste i n a musical e n v i r o n m e n t w h i c h c o u l d n o t be i m p r o v e d .
3

Cf. Figure 480.

2 T h e examples and Lessons to Part I were engraved and bound under seprate
cover. Bach's similar plans for Part I I were happily altered by the perfection of a
method of music printing by Johann Gottlob Breitkopf in 1755, consisting in the use
of minute fragments of type which were assembled to form musical characters. T h e
process was an improvement over sixteenth-century methods of type printing, which
had been abandoned. T h e new method proved eminently successful and led to mass
production of musical scores. Today, because of the troublesome nature of the process,
its use is limited mostly to short illustrations in books.
The friendly relations between Bach and J . G . Breitkopf are traced by H . von Hase
in the Bach Jahrbuch of 1911. Breitkopf, son Of Bernard Christoph Breitkopf, founder
of the firm of music publishers (Leipzig, 1719) ran the establishment successfully for
many years. T h e additional ame by which the modern firm is known was added
in 1795, when Gottfried Christof Hrtel joined it.
3 T h e Bibliothek der Schnen Wissenschaften observes: "Those who remember
that the author was first instructed by the greatest master in this element of practical
music; developed under him and already in Leipzig distinguished himself in the
happiest way; how later he became a member of the royal court at Berlin, where the
finest taste flourished under the direction of Graun, who died too early for the
good of music; where the art was practiced with extraordinary delicacy and true
sensitivity, especially because a monarch himself, a great conrioisseur of the art, par169
1

rtrucft

6e?

<org<

u&ereig

nter.

T i t l e page of the first edition of Essay on the True Art of

Keyboard Instruments,
Part I I

Playing

ijo

E O R E W O R I)

T O

r A It T

T W O

I have w r i t t e n the examples on one system only, i n order to keep

the size and cost of the w o r k w i t h i n reasonable bounds. T h e reader
must always f i n d the reason u n d e r l y i n g each example and must n o t
restrict his performance of i t to the notated register. W i t h regard to
the d i s t r i b u t i o n of tones, a l l necessary observations w i l l be f o u n d
i n the text. For purposes of i n s t r u c t i o n the niceties of accompaniment and the second part of each section should be treated last.
These should be preceded by the first principies of t h o r o u g h bass.
T h e short sections have no subdivisin.
T h e three-part examples are figured t h r o u g h o u t w i t h T e l e m a n n
bows, a sign w h i c h c o u l d well be used by a l l figurists who^wish to
indcate a three-voiced construction. I use a special sign, 4 3 , for
the six-four w i t h a suspended t h i r d w h i c h has a d o u b l e d sixth i n
four-part construction, i n order to differentiate i t f r o m the six-four
whose f o u r t h part duplicates the bass. I e x p l a i n these signs i n advance so that those w h o merely leaf t h r o u g h m y book w i l l n o t grow
c r i t i c a l or become dismayed by them, b u t rather w i l l understand
immediately that their a i m is clarification. T h e r e are t w o examples
w h i c h seem to contradict the r u l e w h i c h is stated i n Paragraph 4 of
the first section on the c h o r d of the major seventh. T h e first is the
second example of Figure 301, and the second is Figure 357, Example /. I n b o t h cases an ascending seventh is accompanied by the
figure 9 rather t h a n 2.1 have purposely i n c l u d e d these just as I have
f o u n d them, so that the performer m i g h t be apprised of t h e m . For,
although the i n d i c a t i o n is n o t as clear as m i n e , i t is used by some
figurists.
4

I f I had restricted the illustrations i n the first chapter to the matter at hand, I w o u l d have been obliged to o m i t certain essential observations or at least strip t h e m of their context, and many harm o n i c changes based o n preparation and resolution c o u l d n o t have
been treated. A n e x a m i n a t i o n of various i n t r o d u c t i o n s to t h o r o u g h
ticipated in it; those who remember all this will understand that only a virtuoso
like Bach could transform a stiff thorough bass into a fine accompaniment and
show the world of music how it must be played sensitively after the nature of the
piece."
* Telemannschen Bogen. A sign, 5, which indicated that the diminished triad
was to contain in realization only its three original tones without the sixth which was
customarily included. (cf. Chapter V, " T h e Diminished T r i a d , " f 3). I n its more
general sense it indicates a three-part accompaniment. As suggested by its ame, the
sign is said to have been introduced by the prolific composer, Georg Philip Telemann,
C . P . E . Bach's godfather and his predecessor at the Johanneum in Hamburg.
5 Cf. C h . V, " T h e Six-four Chord," I , f 9.

E O R E W O R i)

T O

V A R T

T WO

JI
7

bass w i l l reveal the inadvisability of w i t h h o l d i n g new progressions

u n t i l they have been exhaustively discussed. I have avoided this a n d
placed a greater trust i n the instructor. Various reasons have led me
to repeat certain examples a n d basic principies. For this, I beg the
reader's forbearance. Such excesses can do n o h a r m , for the i m portance of the matter justifies its r e p e t i t i o n , and the reader w i l l
gain the advantage of finding a l l things i n their proper order o n
l o o k i n g u p a passage.
I hope that this w o r k w i l l be as w a r m l y received as the first part,
and trust that i t w i l l prove as practicable f o r the student. A l t h o u g h
my various duties leave l i t t l e t i m e for w r i t i n g , i t is m y desire to
b r i n g o u t a few supplements, p a r t i c u l a r l y to the last chapter. I have
w i t h h e l d several examples and useful observations on improvisat i o n i n order to check the m o u n t i n g expenses of p u b l i c a t i o n . Perhaps these supplements w i l l appear w i t h those to the first book.

INTRODUCTION

INTRODUCTION

TO PART TWO

i
' H E O R G A N , harpsichord, pianoforte, and clavichord are
the keyboard instruments most c o m m o n l y used for accompaniment.
2. I t is u n f o r t u n a t e that the bowed clavier, H o h l f e l d ' s fine i n v e n t i o n , has n o t yet come i n t o general use. U n t i l i t does, its characteristics cannot be described i n detail. Certainly, i t w i l l prove a
useful accompanying i n s t r u m e n t .
3. T h e organ is indispensable i n c h u r c h music w i t h its fugues,
large choruses, a n d sustained style. I t provides splendor a n d m a i n tains order.
4. However, i n a l l recitatives a n d arias i n this style, especially
those i n w h i c h a simple accompaniment permits free v a r i a t i o n o n
the part of the singer, a harpsichord must be used. T h e emptiness
of a performance w i t h o u t this accompanying i n s t r u m e n t is, unfortunately, made apparent to us far too o f t e n .
1

5.

I t is also used for arias a n d recitatives i n chamber and theatri-

cal music.
6. T h e pianoforte a n d clavichord provide the best accompaniments i n performances that r e q u i r e the most elegant taste. Some
singers, however, prefer the support of the clavichord o r harpsic h o r d to the p i a n o f o r t e .
i j o h a n n Hohlfeld (1711-1771), originally an apprentice passementier, invented a
machine for the recording of improvisations and a Bogenklavier whose strings were
bowed rather than struck or plucked. T h i s was one of a long series of attempts, reaching as far back as sketches by Leonardo da Vinci, to produce a sustained tone and
graded dynamic changes at the keyboard. It was demonstratd at the Royal Court in
1753, on which occasion Bach played on it and praised it highly, as did Marpurg.
Bach wrote a sonata for the instrument and a song to its inventor on his death (Wotquenne, No. 65, Andantino, Hamburg, 1783, and No. 202c). Cf. C h . V I , "Performance,"

15-

178

TO

'ART

TWO

jyj

7. T h u s , no piece can be w e l l p e r f o r m e d w i t h o u t some l'orm of

keyboard accompaniment. Even i n heavily scored works, such as
operas p e r f o r m e d o u t of doors, where n o one w o u l d t h i n k that the
harpsichord c o u l d be heard, its absence can certainly be felt. A n d
f r o m a position above the performers a l l of its tones are clearly perceptible. I base these observations o n experiences w h i c h may be
d u p l i c a t e d by anyone.
8. Some soloists take only a viola or even a v i o l i n f o r accompanim e n t . T h i s can be condoned only i n cases of necessity, where good
keyboardists are n o t available, even t h o u g h i t creates many discrepancies. I f the bass is w e l l constructed, the solo becomes a duet;
i f i t is not, h o w d u l l i t sounds w i t h o u t h a r m o n y ! A certain I t a l i a n
master had n o reason t o i n t r o d u c e this k i n d of accompaniment.
W h a t confusin w h e n the parts cross! O r w h e n the melody is dist o r t e d i n order to a v o i d a crossing! As w r i t t e n by the composer, the
t w o parts r e m a i n cise to each other. H o w feeble the f u l l chords
of the p r i n c i p a l p a r t sound, u n s u p p o r t e d b y a real bass! A l l harm o n i c beauty is lost; and a great loss i t is i n affettuoso pieces.
2

9. T h e best accompaniment, one w h i c h is free of c r i t i c i s m , is a

keyboard i n s t r u m e n t and a cello.
10. As regards performers of t h o r o u g h bass, we are worse off
n o w t h a n we used t o be. T h e cause of this is the refinement of
m o d e r n music. N o one can be content any longer w i t h a n accompanist w h o merely reads a n d plays figures i n the manner of a b o r n
pedant, one w h o memorizes a l l of the rules a n d follows t h e m
mechanically. Something m o r e is r e q u i r e d .
11. T h e " s o m e t h i n g m o r e " provides the reason f o r this cont i n u a t i o n of m y Essay, and i t shall f u r n i s h the p r i n c i p a l m a t e r i a l
of its teachings. I a i m t o instruct those accompanists w h o , i n add i t i o n to l e a r n i n g rules, desire to f o l l o w the precepts of good taste.
12. I n order t o become a skilled p e r f o r m e r of t h o r o u g h bass,
due t i m e must first be given to the p l a y i n g of good solos.
13. G o o d solos are those that have well-constructed melodies
and correct h a r m o n y , a n d provide sufficient exercise for b o t h hands.
3

2 Bach is obviously not inveighing against unaccompanied duets, at which he himself tried his hand (cf. Nagels-Archiv No. 35) s early as 1752 (Wotquenne No. 141).
Rather, it is something like the Opus 4 of Emanuele Barbella (1704-1773), Six Solos
for a Violin and Bass or T w o Violins, published by R . Brenner. T h e score contains
a prefatory "Scale for teaching to play the Bass part on the V i o l i n , "
3 Handsachen. Cf. Pt. I , Foreword, Note 2.

7 4

I NT

RO

DUCTIO

TO

PART

TWO

14. I n p l a y i n g these the ear grows accustomed to good melody,

an i m p o r t a n t factor i n accompaniment, as we shall see presently.
15. T h e p e r f o r m e r also becomes f a m i l i a r w i t h a l l types of meter
and tempo along w i t h their characteristic passages. H e forms a
q u i t e useful acquaintance w i t h most of the problems of t h o r o u g h
bass, a n d acquires finger facility a n d practice i n sight reading. I n
brief, solos provide exercise f o r the eyes, ears, and fingers.
16. Especially recommended are constant listening to good
music and careful observing of good accompanists. T h i s w i l l cultvate the ear and teach i t to become attentive.
17. Such attentiveness w i l l p e r m i t no nuance to pass u n n o t i c e d .
Observe how musicians always listen to each other a n d m o d i f y their
performance so that the ensemble may reach the desired goal. A l e r t ness is r e q u i r e d of a l l performers, i n c l u d i n g the accompanist, regardless of how exact the figures may be.
18. O u r present taste has b r o u g h t about an entirely new use of
h a r m o n y . O u r melodies, embellishments, and manner of performance o f t e n cali for unusual chords. A t times they must be
played i n few parts, again, i n many. T h u s , the range of the accompanist's duties has greatly increased a n d the recognized rules
of t h o r o u g h bass, w h i c h must o f t e n be m o d i f i e d , are no longer sufficient.
19. A n accompanist must fit to each piece a correct performance
of its h a r m o n y i n the proper v o l u m e w i t h a suitable d i s t r i b u t i o n of
tones. H e must t r y to f o l l o w exactly the composer's intentions, and
to this end pay u n r e l e n t i n g a t t e n t i o n to the r i p i e n o parts. W h e n
there are n o m i d d l e parts to fill o u t the harmony, as i n solos and
trios, the keyboard accompaniment by itself must be constructed
i n accord w i t h the affect of the piece and the performance of the
other players so that the desires of b o t h the composer a n d the
executants w i l l be satisfied.
4

20. As i n r e a d i n g at sight, the accompanist must always l o o k

ahead to the approaching notes.
21.
I n p u r s u i t of the observations of paragraph 19, I shall give
* ". . . In der gehorigen Starke und Weite." T h i s is one of the essential, guiding
principies of Part I I , to which Bach refers repeatedly. A discussion of the means to
be used in achieving proper volume, i.e., by reducing or adding parts, by changing
from one manual to another, etc., will be found in C h . V I , "Performance," f f 5-13.
I n fact, the entire section is an extended treatment of each of the factors mentioned
in f 19 here.

INTRODUCTION

TO

PART

TWO

as clear and brief an exposition as possible of the usual rules and

their modifications, and make many observations o n accompanim e n t i n general, as w e l l as o n each progression. I shall be attentive
to means by w h i c h the various progressions may be recognized.
L u r k i n g errors a n d ways of overcoming t h e m w i l l be correctly
shown. I shall indicate the best d i s t r i b u t i o n of tones f o r certain
progressions a n d shall differentiate the essential, less essential, and
inessential intervals a n d t h e i r doublings.
22. T h i s latter is necessary because sometimes chords must be
t h i n , sometimes f u l l , and, w i t h reference to the n u m b e r of parts,
there are pieces that r e q u i r e a l l kinds of accompaniment.
23. A c c o m p a n i m e n t may be i n one, t w o , three, f o u r , or more
parts.
24. A u n i f o r m four- or more-voiced accompaniment is used i n
heavily scored music, pieces i n the learned style w h i c h feature
c o u n t e r p o i n t , figures, etc., a n d p r i n c i p a l l y i n works that consist of
music alone, i n w h i c h taste plays a m i n o r r o l e .
25. I n t r e a t i n g four-part accompaniment I shall discuss good
construction as w e l l as smooth progression of intervals. M a n y examples w i l l demnstrate that i n order to r e t a i n a good d i s t r i b u t i o n
of tones i t is better to allow t w o voices to take a unisn t h a n to
m a i n t a i n f o u r seprate, stiff parts w i t h t h e i r needless leaps a n d
a w k w a r d progressions. T h e r e w i l l also be examples to Ilstrate
the ways i n w h i c h the left h a n d must help the r i g h t to a v o i d these
defects, w h i c h are sometimes ascribed to accompanists' four-part
progressions.
5

26. Three- a n d fewer-voiced accompaniments are used i n delicate works where the taste, performance, or affect of a piece requires
a husbanding of harmonic resources. W e shall see presently that
such pieces often allow for delicate accompaniment o n l y .
T

s Marpurg (Der critische Musicus an der Spree, p. 216) writes, "Most Symphonies,
Fugues, and Trios are played in a manner that might be called the most usual kind
of musical expression. I include those Symphonies, etc., in which nothing magnificent,
noisy, playful, or very impassioned is sought. Neither the heart or the understanding
is touched to any great degree."
Telemann (Singe-Spiel- und Generalbass bungen, 1734, p. 9) writes: " W e have
not given place to the unisn, another consonance, because through its use one of
the four voices which we are using here is lost." T h i s expresses the view of the older
school, from which Bach departs, at least in this respect (cf. C h . V, " T h e Chord of
the Seventh," I , f 18, par ex.).
7 T h e learned style of f 24 and that described here, and identified elsewhere as
the galant style, are the two chief objects of Bach's discussion of thorough bass. T h e

tj6

l N T R O D UCTION

7 O

P A R T

T W O

27. I n the case of poor and a w k w a r d compositions i n w h i c h

there is often no clear m i d d l e voice at a l l , o w i n g to the ineptness of
the bass (out of w h i c h m i d d l e parts should flow) the keyboardist
must hide the errors as best he can i n a t h i n accompaniment. H e
must use chords sparingly and i n cases of necessity realize only one
figure. H e must seek refuge i n pauses, d i v i d e d beats, etc. W h e n his
accompaniment is n o t d u p l i c a t e d by other instruments and the
piece allows i t , the accompanist may m o d i f y the bass extemporaneously as a means of w i n n i n g correct, f l o w i n g m i d d l e parts,
just as he w o u l d change incorrect figures. A n d h o w often must this
be done!
8

28. One-part accompaniment consists of the notated bass alone

or its d u p l i c a t i o n by the r i g h t h a n d .
29. I n the first case, . s., tasto, or tasto solo appears over the
notes; i n the second, all' unisono or unisont*
Since the indications
are often lacking, I shall describe t h r o u g h examples and observations those places where such accompaniments are employed.
30. T h e p r i n c i p a l p a r t is that w h i c h performs the leading
melody of a piece i n w h i c h a l l other parts play a s u p p o r t i n g role,
as i n solos, concertos, arias, etc.
31.
T h e upper part is the highest played by the accompanist.
32. T h e student, i n receiving i n s t r u c t i o n , must first play each
example and then w r i t e i t o u t i n t w o staves. T h e ear and eye w i l l
thereby learn to distinguish clearly between the good and the
bad.
33. However, n o t h i n g must be taken f o r granted; b o t h w r i t t e n
and played versions must be j u d g e d . Every note must be j u s t i f i e d .
Objections should be raised w h i c h the student must answer by givi n g reasons why, for example, this or that note and no other must
be used.
34. I t is best to begin w i t h four-part accompaniment and es9

10

11

distinction is maintained throughout, as it was by most writers of the time, for these
were the two prevailing styles of the period. T h e distinction is elaborated by Arnold
in The Art of Accompaniment from a Thorough-Bass, p. 359, note 11, where the
terms "strict style" and "free style" are used in the same sense as our "learned" and
"galant" styles.
*Nachschldge. Cf. C h . I I , " T h e T r i l l , " Note 2.
Cf. C h . V I , "One-Part Accompaniment for the Left H a n d . "
10 Cf. C h . V I , " T h e Unisn."
" For a description of the manner of playing thorough-bass exercises see C h . V I ,
"Performance," \.

INTRODU

C T l 0

T 0

V A R T

T W O

177

tablish its foundations. Those w h o learn this style t h o r o u g h l y w i l l

find it easy to go o n to others.
35. Progressions must be practiced i n a l l d i s t r i b u t i o n s of
t o n e s so that they w i l l become k n o w n to the student. I n d o i n g
this, i t is clear that a w k w a r d progressions and poor d i s t r i b u t i o n s
w i l l arise. I n m e e t i n g these the student w i l l learn to differentiate
the good progressions and d i s t r i b u t i o n s f r o m the poor. However,
whenever i t is possible, the remedy for a w k w a r d voice leading
must be shown i m m e d i a t e l y .
36. A l t h o u g h the voice leading may prove a w k w a r d , i t must
n o t be incorrect. Proper preparation and resolution must be attended to, and f o r b i d d e n fifths and octaves must be strictly avoided.
37. I n studying the three tonal d i s t r i b u t i o n s of four-part chords
the student, as suggested i n paragraphs 25 and 35, should also learn
the use of f o u r expedients: first, the p l a y i n g of a m i d d l e voice by the
left h a n d ; second, the b r i n g i n g together of t w o voices o n the unisn;
t h i r d , the momentary a d d i n g of a fifth part to the r i g h t h a n d as a
means of averting fifths w h i l e a v o i d i n g a r e t u r n to a previous dist r i b u t i o n ; f o u r t h , the changing of a tonal d i s t r i b u t i o n by the repetit i o n of a c h o r d over a single bass note, as a means of r e g a i n i n g a
higher register w h e n the accompaniment goes too l o w . A l l f o u r expedients are n o t only allowed i n t h o r o u g h bass, b u t are often req u i r e d , as we shall soon learn.
38. Unavoidable a w k w a r d progressions, h i d d e n fifths and octaves, and certain permissible fifths against the bass must be placed
i n the m i d d l e parts. T h e upper p a r t must always sing and m a i n t a i n
a pur relationship w i t h the bass.
39. I n s t r u c t i o n should commence w i t h the easy progressions
a n d proceed systematically t h r o u g h the remainder. A b r i e f exercise
must illustrate every progression. Brevity does n o t t r y patience.
T h i s is an i m p o r t a n t consideration, for the student must n o t take
u p a new example u n t i l the od has been w e l l c o m m i t t e d to the
m i n d and hands. C o n t r a r i l y , n o t h i n g is gained by h o l d i n g u p the
progress of a student w i t h unnecessarily l o n g illustrations, for, f o l l o w i n g his mastery of the progressions, he should d i l i g e n t l y practice
the accompaniment of various, complete pieces. As preparation for
such study, short examples sufnce. T h r o u g h the practice of these
12

12 "In alien Lagen"; i.e., with each tone of the chord taking its turn in the upper
part. See f 37 of this section.

iy8

IN

T RO

DUCTI

ON

TO

PART

T W O

and the gradual e l i m i n a t i o n of errors, a satisfactory competence

w i l l eventual.ly be w o n .
40. T h e teacher should transpose these short examples i n a l l
tonal d i s t r i b u t i o n s to a l l keys, major and m i n o r , so that the student
w i l l become f a m i l i a r w i t h t h e m and their n o t a t i o n . Later, this task
should be carried o u t by the student alone.
41.
I consider i t better, i n transposing, to take keys at r a n d o m
rather than move stepwise f r o m one to another, for some students
l i k e to play and copy f r o m the untransposed example, depending
on their memory to make the necessary changes w i t h o u t g i v i n g the
matter any real t h o u g h t . T h e loss here is considerable. O n the
other hand, by t a k i n g keys at r a n d o m a student soon acquires
facility i n reading figures and simultaneously m a i n t a i n i n g a good
d i s t r i b u t i o n of parts. T h i s last factor assumes many forms and constantly offers o p p o r t u n i t i e s to make use of acceptable expedients as
a means of r e m a i n i n g i n a good register. I n a w o r d , he eventually
masters a l l intervals, regardless of where they l i e .
42. W h i l e studying transposition, the teacher must e x p l a i n key
signatures to his student and familiarize h i m w i t h t h e m . Ntate the
C major and A m i n o r scales and let h i m , using these as models,
w r i t e o u t a l l major and m i n o r scales. I t is hardly necessary for me
to p o i n t o u t that i t is customary to construct each scale i n stepwise
descent (c, b, a, g, f, etc.) and to correct w i t h accidentis those steps
that are too large or too small compared w i t h the m o d e l . By this
procedure, he w i l l soon memorize the n u m b e r and position of accidentis i n each scale, such as, for example, the n u m b e r of fats
i n D-flat, or sharps i n C-sharp. I f the keys are related by fifths and
fourths the gradual increase i n the n u m b e r of accidentis w i l l be
made apparent to h i m .
43. Such knowledge w i l l prove b o t h useful and indispensable.
Inescapable situations can very easily arise: T h e performer is suddenly r e q u i r e d to provide an accompaniment and is allowed to give
his part n o t h i n g more than a cursory e x a m i n a t i o n ; he is n o t given
t i m e enough to determine the key f r o m the final note; he can glance
only hastily at the key signature. W h a t an unhappy p o s i t i o n for
anyon w h o is aware of the rare services and heavy obligations of
ripienists, w h o knows a l l too w e l l that by rights a l l r i p i e n o parts
should be carefully studied i n advance to ensure a good p e r f o r m ance! Aside f r o m this matter of execution the part may c o n t a i n

IN

T RO

DUCTION

T O

PART

TWO

copyist's errors or, at least, i l l e g i b l e , ambiguous notes, unexpected

changes of meter, tempo, figures, keys, etc., w h i c h w o u l d r e q u i r e
preparation f r o m even the most experienced executant.
44. However, should there be suficient t i m e to look t h r o u g h
the part, examine first the key signature, w h i c h can be w r i t t e n i n
more than the one correct way described above. I n the past the
signature of D m i n o r rarely contained 6-flat, or C m i n o r , a-flat.
Some composers still w r i t e this way f r o m habit, love of the obsolete,
or perhaps, for other reasons. Others correctly h o l d to o n l y one
signature or none at a l l , especially i n h i g h l y chromatic works,
freely m o d u l a t i n g recitatives, etc., i n order to avoid frequent
changes, w h i c h m i g h t prove confusing to the performer. I n such
cases, many of the accidentis are n o t even i n c l u d e d i n the figured
bass, i t being assumed that the executant is conversant w i t h every
key.

C H A P T E R

INTERVALS

F O U R

AND

THEIR

SIGNATURES

1
4 L L composers who desire good accompaniments to their works
/ \t make certain that the bass is correctly and fully figj V ured. The rules which pertain to unfigured basses are often
wrong and, in addition, leave many questions unanswered.
2. When the upper part of solos is written in over the bass, or, in
larger pieces, when all parts are scored, the keyboardist can, of
course, fashion his accompaniment from an unfigured bass, provided he has had experience in composition. If, in addition, an exact
figuring is included, the accompaniment may prove to be good. By
good accompaniment, I mean only the very best. But aside from
this, I know that unfigured basses are often given to a certain keyboardist who is not always able to disentangle himself from the accompaniment.
3. With this in mind I shall discuss means by which an experienced accompanist will find it much easier to make a satisfactory
realization of unfigured basses. However, my principal concern will
be figured basses.
4. Pupils must learn the figures with dispatch. For this reason, I
am no defender of great masses of numeris. I oppose everything
that makes for unnecessary trouble and destroys incentive. At the
same time, no one can gain a comprehensive knowledge of thorough
bass or learn to accompany properly who does not have an exhaustive vocabulary of figures. Once the executant has triumphed
over his fear of signatures, he will be ready to direct his attention to
the refinements of accompaniment. These refinements require
morefiguresthan were formerly needed for the usual realization of
180

basses./ Can
present-day
disclose hisTintentions
all
N 7 the
E R
V A L S composer
AND
SIGNA
U R E S at 181
Wthout recourse to figures?
5. Teachers should let their students accompany pieces with
chromatic basses, for these require an ampie use of figures. With
this purpose in mind, I have assigned many of my father's basses to
students, without endangering their lives. Also, they do no harm
to thefingers.It' is good practice to use the correctly figured works
of many composers, so that the student may grow familiar with the
various types of signatures and chromatic changes. These should be
analyzed, once the student's understanding is adequate, for the insight which is gained hereby will later prove very useful. Such an
undertaking increases his knowledge offiguresby making it an indispensable part of the study.
6. Thorough bass would be an easier and move agreeable study,
if there were general agreement on the matter of figuring. The
main contribution to this cause must be made by accomplished
keyboardists who themselves can fashion a good accompaniment.
There are many excellent composers and musicians who, while
appreciative of a good accompaniment, might find it difficult to indicate their intentions with regard to the keyboard in an idiomatic
and requisite manner. The followirig are among the principal
points on which all should agree: Everything essential must be
indicated exactly; neither too many or too few figures should be
written; figures must be chosen with an eye to performance; they
must be correctly placed; where there are no figures, it must be
made known by some sign; all styles of accompaniment, especially
the three-, two-, and one-part, must be specified at the point where
each is to be used, etc.
7. The relationship of one tone to another is called an interval.
8. All signs in thorough bass that pertain to accompaniment
are called signatures.
9. All intervals are measured upward by steps from the bass and
are identified by the resultant numeral.
10. The most usual intervals are those in Figure 18o.
11. An interval retains its ame as long as its step is not changed,
regardless of changes in accidentis. Thus, all seconds are on the
second step, all thirds on the third, etc.
12. Differences in the size of intervals of the same step, whether

182

INTERVALS

AND

SIGNATURES

Figure 18JL

Thirds

Seconds

minor

major

augmented diminished

J |[J II J

minor

BF=IB
1

dimin ished per ect

y i

1 Rr IIJ 1 s
II IN1H ifi
1

^-

4*

diminished

g IiJ

H
II

Sevenths

major

II J

m e n

t e d diminished minor

II J

IIJ

H.
B-|
perfect

major

11 i

N i i uhs

Octaves

1J

augm ented

augme nted dimin ished perfect

minor

IIJ

Fifths

Sixths

diminished

major

I J

U11J

Fourths

H-i

INTERVALS

1\im
H^

=F=fc
H

augmented

minor

1 r

ii

major

1 JigiJi

brought about by accidentis or without them, are denoted by certain adjectives.

13. In order to recognize such variation, observe that the rela-

AND

SIGNATURES

183

tionship of adjacent keys is called a half tone. Two half tones make
a whole tone.
14. The minor second comprises a half tone, the major, a whole
tone, and the augmented, a tone and a half.
15. The diminished third comprises a whole tone, the minor, a
tone and a half, and the major, two whole tones.
16. The diminished fourth contains two whole tones; the perfect lies a half tone above the major third; the augmented comprises a whole tone more than the major third.
17. The diminished fifth lies a half tone above the perfect
fourth; the perfect comprises a whole tone more than the perfect
fourth; the augmented lies a half tone above the perfect.
18. The diminished sixth contains the same number of tones
as the perfect fifth; the minor sixth lies a half tone above the perfect
fifth; the major sixth lies a whole tone, and the augmented sixth a
tone and a half above the perfect fifth.
19. The diminished seventh contains a half tone more than the
minor sixth; the minor seventh lies one tone below the octave; the
major, a half tone below.
20. The diminished octave is a half tone lower than the perfect;
the perfect contains five whole and two half tones; the augmented
lies a half tone above the perfect.
21. In practice the minor ninth and the minor second, the
major ninth and major second express the same tones. Actually an
octave separates each ninth from its parallel second.
22. Primes, tenths, elevenths, and twelfths are nothing more
than octaves, thirds, fourths, and fifths. They are denoted by the
numeris, 1, 10, 11, and 12, and appear often in galant notation and
three-part accompaniment. Suchfiguresare used to specify the exact
progression of voices, as in Figure 181. It can be seen that the succession 1-2 or 2-1 is more natural and easier to read than 8-2 or 2-8
(Example a). The same applies to 10, 11, and 12 (b). Successions of
these numeris are written as a general rule only before or after a
simple 7-8-9 (c). Also, by this means it is possible clearly to determine whether two voices should progress in thirds or sixths (d), a
point not always to be decided on arbitrarily in a fine accompaniment.
23. The unisn in its narrow sense occurs when two or more
voices occupy the same key. Henee, it cannot very well be called an

184

INTE

li V A L S AND

S I (', N A T R E S

Figure 181

8 4
1 2

5
3

10 11 11)
9 8

9 8 7
7 6 5f

b. 8

- 21

'
*=r-1
,f_^jH lf

-^r-

6 !>7
C. 4 5

8,9
6W

"

10 11 ,12
5 6 7
56 ka'. 3 4 2

12 11 10
10 9 8

i 9 8 1,7
1,7

6 5

1f=
8 9 10 111211 - 10
3 4 5 6 7 - - 8

r r r r ' r r r r *r ? T T
interval. Very often it denotes an octave, a meaning which we shall
later discuss in detail. Some use the term "unisn" instead of
"prime," designating it also with the numeral i .
24. Intervals express the same tones and retain their ames in
all octaves.
25. The second and ninth express the same tones but otherwise
differ greatly from each other, as we shall soon learn.
26. With regard to quality, intervals follow the construction of
the staff. They are thus affected by key signatures without any
further indication in the figured bass. For example, if the key signature contains an /-sharp, the sixth above a is not f, but /-sharp, and
this is indicated simply by 6 over the a.
27. However, all chromatic alterations, aside from those included in the key signature, must be expressly indicated.
28. An interval is said to be naturally major, etc., when it agrees
with the key signature, and artificially major, etc., when it is altered by an interpolated accidental.
29. A stroke through a figure or a sharp next to it raises an interval a half tone, as in Figure 182.
1

INTERVALS

AND

SIGNATURES

185

The use of strokes is generally known to Germans and customary

with them. The Italians also use them; it is only the French who
cause confusin by departing from the practice. In Leclair's figured
basses one finds both natural major and artificial minor intervals
indicated by a stroke.
30. A fat sign before a figure or after it lowers an interval a half
tone, as in Figure 183.
Figure 183

JJ

J J

-fw1

1
1-6

If
6V-4

31. A natural sign before a figure or after it restores an interval

to its natural size. It is hardly necessary for me to remark that a
natural sign lowers intervals in sharp keys and raises them in fat
keys, as in Figure 184.
Figure 184

32. Two strokes, two sharps, or a single sharp drawn before a

figure or after it raise an interval a whole tone, as in Figure 185.
Figure 185

$,7

fi

Figure 182
5

8 #8

1 Cf. C h . V , " T h e Chord of the N i n t h , " f 5, and " T h e Chord of the Major Seventh"

The use of two sharps is rare and not clear.

33. Two fats or one large fat before a figure or after it lower an
interval a whole tone, as in Figure 186.

iS6

INTERVALO

Figure 186

J
r

AND

SIGNA

INTE

T U R ES

The large flat, despite its convenience, is not yet widely used.
34. The combinations of natural-flat and natural-sharp which
follow a double alteration and restore an interval to its normal size
are not as frequently met as would be required by an exact notation. But since they do appear occasionally I mention them here i n
order to forestall the performer's alarm on meeting them.
35. I t should not be considered strange that some write fats or
strokes where naturals should appear. T h e double meaning of the
natural sign, which raises and also lowers tones, is responsible for
this practice, as i n Figure 187.
Figure

187
\instead o f t)

m r
t

instead o f 1)7

& instead of 5 t)

iW'-

T& instead of 6l)

R V A L S A NI)

SIGN

AT

U RE S

187

37. Strokes, and fats and naturals which are drawn before the
figures, are the easiest to read and, when figures appear i n cise
succession, elimnate all uncertainty over the figures to which they
pertain.
38. Unless an accidental is canceled i t contines to be effective.
39. This rule also applies to figures which appear over repeated
tones each one of which expresses a chord. The first figure is effective until another appears, as i n Figure 189.
Figure 189
6

^ rr r r r r i
1

I n this example the sixth is struck on each of the first four notes,
after which i t is replaced by the fifth.
40. Figures that are placed directly over a note are realized immediately; but when they are placed to the right they are realized
after the note has been struck, although they pertain to i t and are
measured from it, as i n Figure 190.
Figure 190

-4L

&\instead o f 6tj

I t is customary to indcate the diminished fifth and minor and

diminished sevenths by means of a flat.
36. The third may be indicated simply through the accidentis
that alter or restore its normal size, as i n Figure 188.
2

Figure 188

2 Even when it is naturally diminished, a common practice of thorough bass.

41. I t is not good to place figures under the notes, tor this position should be reserved for piano and forte signs. But at times i t is
not possible to write them elsewhere; for example, when two voices,
one for the cello, the other for the keyboard, are written on one
staff.
42. When the subject of a fugue is given to the bass, play the
notes as written, omitting chords until figures appear. The same
rule applies to short passages where the right hand plays an obbligato accompaniment, which is usually written i n small notation,
as i n Figure 191.
43. Figures that appear over the dots that lengthen notes are
realized at the point where the dot takes effect, although they pertain to the preceding note.

iSS

INTER

VALS

A N I) SI (N

ATURES

Figure 191
7

INTER

Figure 193

U RE S

189

47. If two successivefiguresappear over a note of triple length

or two unequal lengths, which is the same thing, the first figure receives two-thirds or the larger part of the valu and the second the
remaining third or smaller part, as in Figure 197.
Figure 197

ir

t.

SE

JV^

A trained musician can very easily determine which of these two

cases applies to a given rest by examining its context.
46. Figures that are placed after a note are realized on divided
beats according to its length. If it is duple, the figure or figures are
played on the second half of its length, as in Figure 194.
Figure 194

MU

SI GNAT

45. Figures over long rests are also played on the rest, but pertain to the preceding note, as in Figure 193.

S AND

If three successive figures appear over such a note, the first figure,
directly over the note, is given half of its length and the others take
ecpial parts of the remaining half, as in Figure 196.
Figure 196

44. Figures that appear over short rests are played on the rest
but pertain to the following note, as in Figure 192.
Figure 192

VAL

If two successive figures appear over a bass note of duple length,

eachfigureis given half of the note's valu, as in Figure 195.
Figure 195

If there are three successive figures, each is given a third of the

valu, as in Figure 198.
Figure 198
V

"iT

'7

48. These are the usual divisions; any departure from them
must be expressly indicated, as in Figure 199.
Figure 199

r i- V
<2

|7

6 -

7
6

In both examples the appoggiatura calis for a modificaton of the

rules. The dash which often serves to indcate the prolonging of a

io
9

INTERVALS

AND

SIGNATURES

figure clearly expresses the desired divisin. Some omit the dash and
seprate the last figure from the others. This is not a reliable practice, for it may prove ambiguous. Often the performer cannot be
certain whether the composer or the copyist made the grouping, as
in Figure 200.
Figure 200

,I

II
9

1 ,1

|J

In this particular case another sign is lacking, as we shall see

presently.
49. In the following examples the figures are realized in equal
lengths, as in Figure 201.
Figure 201
6 4

SIGNATURES

a degree, that they are never allowed to be played in succession. Out

of this arises the well-known first principal rule of harmony: Two
octaves or perfect fifths in a pair of voices may never be played in
parallel motion, either by leap or by step. Violations are called
simply "fifths" or "octaves," as in Figure 202.

1
[7

AND

Figure 202

INTERVALS

41

I | - I|
f

Parallel motion occurs when two or more parts move in the same
direction upward or downward (Figure 203, Example a), and contrary motion, when they move in opposite directions by leap or
step (b):
Figure 203

II

50. Because the position of figures is so important, the composer as well as the copyist must be careful to leave enough space
in the score to be able to write them in their proper place, especially when there are a great many slurs and other signs relative to
performance.
51. Intervals are either consonant or dissonant.
52. A consonant interval is one that may be played without
preparation (that is, without being present in the preceding chord),
that may be doubled, and quitted by ascending or descending leap
or step.
53. This may be done with minor and major thirds, perfect
fths, minor and major sixths and perfect octaves. Consequently,
these are the consonant intervals.
54. In this connection, observe that the octave and fifth are
called perfect, first, because they cannot be altered and remain consonant (as soon as they are made smaller or larger they become
dissonant); second, because, struck once, they satisfy the ear to such

55. It is obvious that the parallel octaves rule .does not apply
to those places where good reason prompts a composer to lead two
voices in unisn. It does apply to chord progressions.
56. Thirds and sixths are called imperfect consonances, because
both forms of them, major and minor, sound well; the ear will accept successions of both intervals.
57. Basically the remaining intervals can not be treated in the
manner of consonances as described in paragraph 52. Henee, they
are dissonant.
58. The basic characteristics of dissonances are suggested by
their ame, which expresses the fact that they sound bad. From this
it follows that they may be used only under certain conditions.
Their natural harshness must be mollified by preparation and
resolution; that is, the dissonant tone must be played, previously,
as a consonance and it must succeed to a consonance. By itself, a
3

T h u s the perfect fourth, which has shifted its allegiance from consonance to dissonance many times in theoretical writings, becomes a dissonance here. Cf. C h . V ,
"The Six-Four Chord," I , f 7.

i2
9

INTERVALS

AND

INTERVALS

SIGNATURES

dissonant tone is sufficiently disagreeable; henee it is wrong to

double it; moreover, because it must be resolved, doubling would
induce forbidden octaves.
59. As a means of gaining a clear understanding of the use of
dissonances, observe the preparation in the first bar of each of the
examples of Figure 204 and, in the second bar, the resolution, which
calis for descending or ascending stepwise motion.

Figure 206
" 1

J.

AND

4
4

SI GN AT
6

UR E S
6

'r

A !l _I

9
6,

flJ

9 3

4 3

- i r f B-f

rj r
A

V
5b

8
-

6
4

(
5

Figure 204

Such a relationship is called a delayed resolution

(retardatio).
65. Occasionally, the right hand does not wait for the arrival
of the bass that resolves its dissonance but expresses its resolution
prematurely (Figure 207, Example a); the bass also does this at
times (b).
6

Vi

60. Resolution is a constant requirement of all dissonances,

but not preparation. Later, however, we shall discuss two cases
where even resolution may be omitted.*
61. All dissonances may be struck unprepared over a stationary
or repeated bass. The lack of motion in the bass precludes preparation, but at the same time compensates for it.
62. There are many other ways of introducing unprepared dissonances.
63. An accidental which lowers a dissonance does not disrupt
its preparation (Figure 205). This follows from the discussion of
paragraph 11.

Figure 207
(5

a.

y J)|J1 i r r f l
-^

64. A moving bass often causes dissonances to resol ve to dissonances (Figure 206, Example a) or to remain stationary (b). Eventually, however, there must be a resolution to a consonance.
* Cf. f 77-79.

b.

b.2

3
6

Figure 208
4+

. 11
11 r l r f,

"4pm

Both are known as an anticipated resolution

(anticipati).
66. When a tone in the bass is interchanged with one in the
right hand prior to the resolution of a dissonance, a transfer of
chordal tones is said to oceur, as in Figure 208.
6

Figure 205

a.3

-6
4
3

67. When the bass strikes the tone to which a dissonance in the
right hand should resolve, a transferred resolution is said to oc8

5 Eine Aufhaltung der Auflisung.

8 Eine Vorausnahme der Harmonie. T h u s , if the chord arrives before its bass or
the bass before its chord, regardless of rhythmic position, an anticipation takes place.
T h i s gives the term a broader meaning than it has today, when anticipations and
suspensions are distinguished by rhythmic position. T h u s only the examples under
b are analyzed as anticipations today, while those under a are analyzed as bass suspensions or appoggiaturas. T h i s point comes up repeatedly when Bach discusses the
rhythmic dislocation of chordal tones.
i Eine Verwechselung der
6 Eine Verwechselung der

Harmonie.
Auflsung.

9 4

INTERVALS

AND

SIGNATURES

INTERVALS

AND

SIGNATURES

ip

cur. The dissonant tone is freed by this action of the bass, which
satises the need for resolution, as in Figure 209.

70. In the case of an extended succession of passing tones the

chord may be repeated, as in Figure 211.

Figure 209

Figure 211
6

5E

5\>

Our aim here is merely to introduce the accompanist to these liberties; we leave it to the composer to employ them with discrimination.
68. It is rare that each tone of a succession of rapid notes takes
a seprate chord. Such unaccompanied notes are said to pass.
69. A single passing tone is not indicated by a sign, but several
in succession are denoted by a dash which extends to the point
where the right hand resumes. Passing tones appear in all styles,
meters, and tempos. Sometimes half of the notes are passing (Figure
210, Example a); sometirrjes fewer (b); but often in rapid tempos
with short notes most of them are passing (c).
9

i-

71. In certain cases which will be treated later, the intervals may
be said to pass. This happens in three ways: (1) Over a stationary
bass (Figure 212).
Figure 212
8 \fl

(2) Over a moving bass with a stationary right hand (Figure 213).
Figure 213

Figure 210

(3) When both hands move (Figure 214).

Figure 214

Sie gehen durch; later, f 69, die durchgehende Noten. These terms are general
in meaning and apply to all types of connecting or lling tones regardless of whether
they proceed by step, leap, or simply repetition (cf. % 72). T h e passing tone in its
narrow meaning is Der Durchgang, standing for a stepwise transition from one tone
to another (cf. f 73). It can be seen that the terms overlap, since both are concerned
with stepwise motion, the first partially, the second exclusively. We have no ready
English equivalent for the broader concept unless we extend the connotation of our
term passing tone. T h i s proved to be the most advisable thing to do here, especially
since Bach uses both terms loosely. For the accompanist, it is important to distinguish
between principal tones that require chords, passing tones which are not accompanied,

72. In rapid drum basses the playing of which may cause a

stiffening of the wrist, notes are occasionally passed over, or
omitted. An extended discussion of this expedient will be found
in the Introduction to Part I , Paragraph 9a.
and those over which chords are played. T h i s matter is discussed in C h . V I , "Passing
Tones" (cf. f 3) and "Changing Notes." Chordal by-products of horizontal motion
are called passing chords. Cf. Arnold, The Art of Accompaniment from a ThoroughBass, C h . X V I I I .

i6

INTERV

AL S A N I) S I G N A T U R S

/ N T E R V ALS

73. The term "passing tone" (transitus), refers in its narrow

sense to a stepwise bass.
74. When the accompaniment is played with its proper bass
on the long part of a bar, the passing tone is called regular (transitus
regularis). With notes of equal valu, the first, third, etc., are long
according to the meter and the second, fourth, etc., short (Figure
Figure 215
7
5

SI GNAT

UR E S

197

Figure 218

r r 1i r r ' ^gf
79. The same freedom of treatment applies to dissonances
which become consonances through an enharmonic change as in
Figure 219.
Figure 219
i>7

75. When the accompaniment is played on the long pulse ahead

of its proper bass, which falls on the short pulse, the passing tone is
irregular (transitus irregularis) and is known as a changing note
(Figure 216).

AND

6l>

7b %

80. On the other hand, we shall learn later that consonances

sometimes lose their freedom and require the same preparation and
resolution as dissonances.
10

Figure 216
7
5
2

7
4
2

76. When the composer prefers not to figure changing notes he

may set figures over only the succeeding tones, or place over the
changing notes an oblique stroke, a circle, a semicircle, or an m ,
which, on occasion, may be lengthened (Figure 217).
Figure 217
L

5 " /
2

3- 6 5 " 4

6 ~ 5

The oblique stroke under Example 2 is the best.

77. Changing notes also occur as anticipations, illustrations of
which appear in Figure 207, Example a.
78. The dissonances which result from the introduction of both
types of passing tone do not always require resolution, even when
they are prepared (Figure 218).

10 T h i s is a reference to an o d dilemma of thorough bass. A l l intervals were computed solely from the bass, and their consonance or dissonance was determined by
this relationship. T h u s , in the six-five chord, the second formed by the fifth against
the sixth is not mentioned as an interval. Instead Bach writes: "Die Quint wird wie
eine Dissonanz gebraucht; sie lsst sich . . . von der sexte binden und gehet allezeit
herunter." (The fifth is treated as a dissonance; it is restricted by the sixth and always
progresses by stepwise descent). T h u s does a consonance lose its freedom. T h i s statement recurs in all cases where chords contain a tone which is consonant with the bass,
but dissonant in its relation to another upper part. Cf. in C h . V : " T h e Four-Three
Chord," I , f 4; " T h e Six-Five Chord," I , f 4; " T h e Seven-Six Chord," f 5; " T h e
Chord of the Major Seventh," I I , f 2; " T h e Five-Four Chord," f 4. T h e fourth in this
last chord is a dissonance, but a mild one, and is not always obliged to descend in other
contexts.

THOROUGH

CHAPTER

FIVE

THOROUGH BASS

TRIAD,

1
HE most perfect consonant chord, that with which most
works begin and all end, is the triad.
2. It consists of a ground tone, fifth, and third.
3. When the octave is added it becomes the common chord, the
fifth of which must be perfect. It is only the third that is variable,
appearing as either major or minor.
4. The chord is major when its third is major, and minor when
its third its minor.
5. The unnatural triad contains either a diminished or an augmented fifth.
6. In the first case it is called the diminished triad, in the second,
the augmented.
7. These chords, which contain dissonances, will be discussed
on the completion of our study of consonant chords.
8. The tones of the common chord may be distributed in three
ways: with either the fifth, octave, or third in the upper voice (Figure 220).

Figure 220

1 1

i Cf. Arnold, Art of Accompaniment

from a Thorough-Bass,
198

pp. 498-505.

99

9. The common chord is played when there appears over a bass

which is not passing, either nothing at all, an isolated accidental,
or 8, 5, 3, each by itself, or combined with one or both of the other
numeris.
10. Since the fifth is always perfect, it is played as such without
any indication (Figure 221).
Figure 221

#
THE

BASS

Vr B r
11. According to circumstances, the octave of the bass may be
omitted and either the third or the fifth doubled.
12. However, when the third becomes major by chromatic alteration it is not doubled.
13. In three-part accompaniment the octave of the bass is
omitted, although a resolution or the melody of the principal part
may require that the fifth be omitted instead.
14. In two-part accompaniment, other things being equal, only
the third is played.
15. In order to recognize the common chord from its notation,
observe that its tones fall on adjacent lines or adjacent spaces.
16. When I strike two notes which are separated by three keys,
I play the major third; but separated by two keys, the minor
third.
17. Contrary motion makes the best and safest accompaniments,
especially when triads are employed. With it, open and hidden
fifths and octaves can be avoided.
18. Hidden fifths and octaves become apparent when two voices
which move in similar motion are filled in, thereby creating open
fifths and octaves (Figure 222).
2

2 Throughout, Bach distinguishes between a doubling by the right hand of the

bass note and a doubling within the right hand. T h e first has been translated as
"duplication" or "the octave of the bass" (Bach writes simply die Octave) and the
other as "doubling" (Bach, die
Verdoppelung).

T II O R o u C II

zoo

r II O R O II C, II

li A s s

Figure 222

f f ' r" r r
11

T^TJ

19. These are more permissible between inner voices or an

inner voice and the bass than between the upper voice and the bass,
for these parts must always be constructed with a view to absolute
purity and good melody. Since such progressions crate impure relationships, they are bad.
20. Nevertheless, the following hidden fifths may appear in the
outer as well as inner parts (Figure 223).
Figure 223

Jj-

21. Two open fifths of different quality may be played in succession.

22. In any pair of voices a perfect fifth may descend to a diminished fifth (Figure 224).
Figure 224

li A s s

201

fifth to a perfect (b) because of the descending tendency of the

diminished fifth.
Figure 226

:i

/ . i

r r

11

r r

Both progressions belong to inner parts.

24. The right hand should not play above the two-lined f, unless the bass is written very high, or the bass clef is replaced by a
higher one; or unless a special effect is intended, as, for example,
when a passage is varied on its repetition by a change of register, etc.
25. The right hand should not play below the upper half of
the unlined octave, unless conditions the reverse of those mentioned
in the preceding paragraph are present.
26. For purposes of instruction, students may exceed this range
in order to practice successions in various distributions and thereby
become familiar with all registers.
27. At first, however, it is customary to restrict the right hand
to the confines of the descant clef, and the bass to the bass clef.
28. The foundations of accompaniment can be best established
when students are required to learn thoroughly all twenty-four
common chords. This should occur gradually. The chords in their
three distributions should be played up and down the entire keyboard. At first it is sufficient to let this be done slowly, but later the
speed must be regularly increased so that finally the hands will
develop the ability to strike any common chord without hesitancy.
29. To begin with, only a few chords should be assigned and
reassigned until the requisite knowledge and facility have been
acquired.
30. All lessons must be related to each other so that od material
will be constantly reviewed and not forgotten.
3

But a diminished fifth may succeed to a perfect fifth only out of

necessity, and preferably not in the outer parts (Figure 225).
Figure 225

s Known as Bassetto or, in Germn, Bassett, Bassetgen. Cf. Arnold, op. cit., pp.

23. In ascending motion the progression of a perfect fifth to a

diminished fifth (Figure 226, Example a) is better than a diminished

373 ff* T h a t is, the C-clef on the first line.

5 Le., major and minor triads on each degree of the chromatic scale.

THORO
U G H li A S S
31. Here, as in other exercises, the student must be asked numberless questins about intervals so that he will develop the ability
to recognize them automatically without reflection. This suggestion
is based"&T*experience. There are many who by virtue of long practice and a good ear realize chords and figures, and even accompany
entire pieces without any knowledge, intervals as well as rules being
little known to them. Useful and essential as a good ear is, it will
prove undependable and harmful when the executant depends on it
alone without exercising his mind.
32. Chords succeed to each other in the most direct manner.
This should be observed in all accompaniments.
33. Henee, when the bass rises a third, retain the intervals common to both tones and take anew only the fifth over the second tone
(Figure 227).
202

Figure 227

THOROUGH

35. When the bass ascends a half step and both chords have
major thirds, the fifth and third of the first chord move to either
an octave or unisn. Henee, the second chord has a doubled third
and no octave (Figure 230).
Figure 230

w r - l

F9^Fi
B =

And when the bass falls a third, only the octave of the second tone
must be sought (Figure 228).
Figure 228

BA S S

Figure 229

1 1

11

1m

Played in reverse, the first chord must have a doubled third and no
octave (Figure 231).
Figure 231

34. But when the bass rises or falls a second, all upper voices
must move contrary to it (Figure 229).
Cf. C h . V I , "Accompaniment," f 14.

If this precaution is not taken, one of the voices will express an

augmented second, which is to be avoided (Figure 232).

II O R

O U G II li A S S

Figure 232

T 11 O R O U G II

li A SS

205

eney lo ascend unless it is hindered from doing so by a prepared

dissonance or a poor doubling (Figure 234).

36. In a final cadenee the fifth must never appear in the upper
voice. The octave is the best interval when it can be reached; next
best is the third; but the closing note of the principal part must not
lie below this third.
37. When the hands come too cise to each other, or the right
hand moves too low, the chord may be shifted to a higher register
by repeating it over a single bass note, provided there is sufficient
time; if there is not, a new voice my be added on top, and the
lowest one over the bass relinquished. These expedients may be
used, first, only in an emergeney, for I believe that under normal
conditions the accompanist should restrict himself to four parts and
not increase their number; second, only with consonances, for dissonances impose limitations on the accompaniment.
THE

TRIAD,

4. Because of this tendeney, the major third of the upper voice

in the first chord of Figure 235 must not descend in contrary motion to the fifth of the second chord.
Figure 235

II

1. The student should be urged to use contrary motion even

when it is not required. In his exercises all kinds of treacherous progressions should be introduced so that he may learn clearly how to
avoid them. For this purpose it is good practice to write out realizations.
2. But when it becomes apparent that he is fully aware of lurking errors, he may be shown those successions in which parallel
motion is sometimes preferable to contrary, as in Figure 233.
Figure 233

A lesser evil, hidden octaves (Figure 236), should be chosen at a

cadenee in preference to the unnatural progression of Figure 235.
Figure 236

m
3. It can be seen in Figure 233 that it is good to lead the upper
voice in parallel thirds with the bass. The major third has a tend-

5. Most disposed to an ascent is the chromatic major third (Figure 237, Example a). Therefore when the octave of such a chord

THOROUGH

2C)6

T H0 R 0 VGH BASS

BASS

moves to the seventh, a voice must be added to the next chord so

that it will be complete (b). If, however, the fifth of the major chord
leaps to the seventh, the expedient is not needed (c).
Figure 237
a.

3E

b.

1 s

TTHP

i f iffifirJif

i.t

207

8. Normally, the common chord does not require a signature.

There are times, however, when it is necessary to write the numeral
or numeris which denote its intervals. The reasons are: dissonances struck and resolved into our chord over a stationary tone
(Figure 240, Example a); a dissonance which follows our chord, the
bass being held (&); a chordal change over a single tone (c); a tone
in the bass which might otherwise be considered passing (d). In all
of these cases the entire chord is played.
Figure 240
9 8

9 8
43

4 3

8 7
5 4
3 2

6. In four-part accompaniment the major third in an inner

part does not require such cise attention. It may progress by a
descending leap (Figure 238).

6 5
43

,
8 t7

8 7,
5 5l>

b.

JF3C

Figure 238
5

- i J J rf ;i
MI
f r
\
1

8 7
j

8 7
.

IIP

9. But there are times when a series of threes is written over

a rapid passing bass as a means of informing the accompanist that
the right hand is to play only parallel thirds with the bass (Figure
7

241).
Figure 241
3

'7. In three-part accompaniment, however, the major third in

the middle part ascends, regardless of the resultant incomplete
triad (Figure 239).
10. Exercises on common chords must not go beyond simple
modulations. Otherwise the ear will grow confused in its attempt to
take in all twenty-four keys at a time. It is far better to hold extravagances in reserve and concntrate on natural chord successions.
Besides, when short studies are transposed to the various keys, all
chords put in an appearance anyway. Further, in transposing, the

Figure 239

J
Z

H\

t Cf. C h . V I , "Some Refinements of Accompaniment," f | 6-8,

thirds without indication is discussed.

where the use of

908

T 11 O R O 11 G U

7' / / O R O II (. II

li A SS

Student leams why some iones are w r i t t e n w i t l i sharps or fats and

yet retain the same relative sound (Figure 242).
Figure 242

J II
94#

- r

9H

,j

\S

&"

i
r r

rTS
1

- ,

-n

**

tf

THE

ir ?
4

11. The following short examples will perhaps suffice to illustrate the meaning of Paragraph 10. The numeris denote the best
intervals for the upper voice (Figure 243).
Figure 243
3
8 5
8

3
8 5 3
5
*n
*r-n

3 5 3 5

=(=F

- m

8 5 3

ff%=

-J

-IM

i
3 5 ^ 8

\> j

y 1*
-q:
* 0 J -*

8 5 8

$ 8 5

0: r J1 *] 1 i

J 1 J

J1

5 8 5 3

41

1 1f f f Pin* 1

Cf. Arnold, op. ci't., p. 503, Note 15.

11

3 5 3
1

Divided accoinpaniment is used either out of necessity or to

i
~ ==iti
F gal
H

njti

1 i.

8 3

"IT*f

II

2<HJ

achieve elegance. Everything that the accompanist must know about

this technique will be illustrated at a later time. Divided accompaniment occurs when the left hand realizes some of thefigureswithout
increasing the normal number of parts. Chords are spread out and
often made more attractive by this means. Occasionally the resolution of a dissonance requires it.
13. Our earlier discussion of primes, tenths, and twelfths applies to divided accompaniment also.

j J- rrh

I.J 1 . J

],

12.

li A SS

CHORD

OF T H E SIXTH,

1. The chord of the sixth that contains a major or minor sixth

consists entirely of consonances, namely, the sixth, third, and
octave.
2. The usual signature of the chord is 6 alone, but at times the
other intervals are specified for various reasons.
3. In figuring the bass, required accidentis must not be overlooked.
4. The third below the bass note becomes the sixth above it, and
the triad that is built from this third or sixth provides the tones
of the chord of the sixth.
5. The chord of the sixth containing the octave of the bass is
rare, being used on single bass notes marked 6, or out of necessity,
when required by dissonances, etc. More frequently, the third or
sixth is doubled and the octave of the bass omitted.
6. In these doublings, which may be in the unisn as well as the
octave, none of the figures is unrealized. The tones of the common
chord (Figure 244, Example a) which comprises the position of the
sixth are present in all of its doublings.
Figure 244

01

=N=
I

y
rTTTTf
But by means of this position many errors can be circumvented and
good voice leading maintained, as we shall see later.
Cf. C h . I V , | f 22-23.
1 Cf. Arnold, op. cit., pp. 516-533.

no

THOROUGH

li A ss

7. The following rules of doubling should be observed:

( i ) When a diatonic sixth ?.nd third are major, either interval
may be doubled (Figure 245).
Figure 245
6

y f i 1
(2) When a diatonic or chromatically raised sixth is major, but
the third is minor, the sixth is not doubled (Figure 246).

T H0 R0

li A S S
111
oceurs when the principal part moves piano in notes that lie a sixth
above the bass while the accompanist plays thirds (Figure 249).
U (' 11

Figure 249

L_

Jd
r r r r 14

1 j
1

10. When the bass of the position of the sixth moves through
successive steps or thirds, the doubling must be alternated with a
duplicated bass in order to avoid octaves (Figure 250).

Figure 246
6

y r 1J 1
(3) When a chromatically raised sixth and third are major,
either may be doubled. Also when only the third of this chord is
raised chromatically, it may be doubled (Figure 247).
Figure 247
...

t,

8
(4) When the bass is raised chromatically it is not duplicated
(Figure 248, Example a) unless the sixth is similarly raised (b):
Figure 248
a. 6

8. Three-part accompaniment comprises only the third and

sixth.
9. In two-part accompaniment one of the intervals must be
omitted; henee the position cannot be used readily. A typical case

II

I T l T | 3 ' ? II f
f

6 6

6 6 6 6

gj J l
6 6 6 6

6 6 6 6

6 6 6 6

6 6 6 6

6 6 6 6

Although it is more urgent to attend to the doubling when the bass

moves by step, when it moves by leaps an alternated doubling helps
to crate a better upper part.
11. When the tempo is rapid, such successions are best expressed
in three voices, in which case only one distribution of tones is possible, for in the other the fourths become fifths. In short, the sixth
should always lie in the upper voice. Even in four-part accompaniment this is the safest and, for melodic purposes, the best distribution.
12. When the octave of the bass is taken, it is best not to place
it in the upper voice.
2 However, see C h . V I , "One-part accompaniment for the left hand,"

3, 4.

Til

2/2

13.

O R O II C, II

T h e unmelodic progressions marked with an asterisk can be

avoided by d o u b l i n g " (Figure 251).

Figure 251

M I

i \ U i i i \

rr rr rr rf rr rr
6*

14. W h e n 6 succeeds to 5, the part that expresses the s i x t h

progresses to the fifth w h i l e the other parts are h e l d . T h i s progression often appears i n succession. A l l three dispositions of the posit i o n of the s i x t h may be employed here, p r o v i d e d that the rules of
d o u b l i n g are heeded. O n r e d i s t r i b u t i n g the tones of the f o l l o w i n g
examples, i t w i l l be f o u n d that a u n i s n d o u b l i n g must be used occasionally. I n a few cases where the t h i r d is d o u b l e d , one of the
d o u b l e d tones progresses to the fifth w h i l e the sixth is held. By this
means leaps, otherwise unavoidable o n single appearances of the
Figure 252

T6 T5 T

(; // li A SS
and a u n i l o r m disposition

li A SS

setting.
16. W h e n the accompaniment is i n four parts, errors are easily
avoided by d o u b l i n g , since the progression consists entirely of consonances. Best are the examples i n w h i c h the two kinds of d o u b l i n g
are alternated. A n exception to such a u n i f o r m procedure occurs o n
the appearance of the d i m i n i s h e d fifth (Figure 253, Example a).
Further, the leap of an augmented f o u r t h should be avoided (b). A
leaping accompaniment w i t h or w i t h o u t a d o u b l i n g i n the r i g h t
hand is acceptable, b u t not always attractive (c). Example d illustrates a d i v i d e d accompaniment.
Figure 253

5 6
6 5

65

65

65

J J

hk

6 5

65

r
5

h
5

m 'I I ','MI J M J M '

r r

6 5

? ' j

r r
5

i$h-*L

i V J i

r r f v r r

m
6 5

65

65

65

6 5 6 5

65

m k s B S

piogrcssion, can be e l i m i n a t e d
maintaitied (Figure 252).
15. W h e n 5 6 appears over a bass note, the c o m m o n c h o r d is
slruck and its fifth moves to the sixth w h i l e the other voices are
held. W h e n this progression appears i n succession, a three-part
accompaniment is the easiest. Also i t is the best fitted for the accompaniment of r a p i d notes i n pieces w h i c h do not cali for a f u l l

r
6 5

21

T 11 o 11 o u

65

3 I.e., instead of a duplication of the bass. Cf. C h . V, " T h e T r i a d , " Note 1.

65

T H O i O VGII

BASS

TH OROUGH

B A SS

2/5

ful to distingus!) the one sctting from the other. A sign of differcntiation would be helpful here, for the indication by itself is
ofi.cn ambiguous. The figures are found over basses which, if not
sel in three parts, might take for the fourth part a third (Figure
255, Example a) or a fourth (b); and at times any additional interval would sound extremely harsh.

18. In four-part settings, the signature 6 often indicates the

resolution of a preceding dissonance (Figure 256, Example a) and
also specifies the exact course of a part (b). But inasmuch as this
latter use applies also to three-part realizations, and no sign of differentiation is mployed, the best advice that can be given to an
accompanist is to listen and judge.
Figure 256
6 5
a.

The diminished fifth seems to abandon its descending tendency in

Figure 253, Examples a, c, and d. But a closer examination of the
voice leading discloses its resolution (Figure 254).
Figure 254

^ i"

rT
7T

1-U &

r
5

pi

51

'

< 17. Occasionally, 6 will be found in the galant style, where it

calis for a three-part realization. Since the same signature is also
mployed in four-part accompaniment, the performer must be care-

9 8 8 7
4 3 6 5

J > U

-1

19. The presence of compound signatures before and after 6 is

usually a sign of three-part accompaniment. If a Telemann bow

9 8
7 6

were placed abo ve the signature ( 6 ), chord and context would be

immediately recognizable as three-part.
20. When the sixth appears with a diminished octave, no additional interval is played. The octave descends and can most frequently be regarded as a retardation of the next tone. The examples
' 7

of Figure 257 are noteworthy. In the last one, 5 precedes an embel\

lishing 6.

2l6

II

o nO

U V, I I

li A

T I I O li O II (l I I

s s

Figure 257

li A

SS

T H E CHORD O F T H E S I X T H , II

I?

6 5 *

5 *

1. T h e accompanist must remember that i t is most necessary

to look ahead i n the case of those chords that p e r m i t more than
one k i n d of realization. H e w i l l n o t always enjoy a free choice b u t
must k n o w how to adjust his c h o r d to those that follow.
2. W h e n the m i n o r sixth and chromatic major t h i r d app^ar at
65
cadenees instead of , i t is best to take the octave of the bass w i t h
tlie sixth (Figure 260, Example a). Moreover, the octave is r e q u i r e d
when i t prepares (b) or resolves (c) a dissonance. I n the last example
the octave serves to e l i m i n a t e a leap. H e r e again the signature
should i n c l u d e a T e l e m a n n bow ( 6 ) as a w a r n i n g .
4

W,f,r ,if,r i
1lJ

J
6

h'8
6

7
5

ll

7
5

7 l|8
5 6

6
4

21.
T h e augmented sixth is a dissonance w h i c h appears w i t h
preparation (Figure 258, Example a) or w i t h o u t i t (b) and always
ascends. T h e r e q u i r e d accidental is i n c l u d e d i n the signature. I f n o
other figure appears, the t h i r d is added i n three-part accompanim e n t and d o u b l e d i n four.
Figure 258
T

Wm

i A

r-r*-

3J
44-

22. T h e d i m i n i s h e d sixth, a dissonance, is rare. I t demands a

distinct type of admirer. Those w h o use i t , prepare i t a n d resolve
i t by stepwise descent. I t sounds passably w e l l w h e n accompanied
solely by the m i n o r t h i r d . T h e r e q u i r e d accidental must n o t be
o m i t t e d f r o m its signature (Figure 259).
Figure 259

6t*

jt

5
4

J f

6
5 6

U f i f

9 8 9 8
7 6 5 6 7 6

11 r

"fr"

ijr' 'rJ'j'
y

3. W h e n a bass note m a r k e d 6 ascends a step to one m a r k e d 5

i t is best, w h e n possible, to add the octave of the bass to the s i x t h ,
for i t makes the smoothest voice leading (Figure 261, E x a m p l e a).
I f the t h i r d is d o u b l e d , one of the voices must leap (b). Composers
may have v a l i d reasons for i n t r o d u c i n g leaps i n t o inner parts, b u t
accompanists avoid them for equally v a l i d ones. A d o u b l e d sixth i n
o u r progression can easily cause fifths (c). T w o parts must leap i n
order to avoid them (d). I repeat, add the octave of the bass when
possible, for there are times w h e n only the sixth or t h i r d can be
doubled. A chromatically raised bass (which is n o t to be duplicated)
may necessitate a d o u b l i n g of the t h i r d (e). Correct r e s o l u t i o n of
dissonances may r e q u i r e a d o u b l e d sixth, as i n the case of the
seventh and augmented fifth of Example / .

/ non o v c, ti n A

TllOHOin.ll

SS

BASS

Figure 264

"^

4. W h e n a bass expresses several sixths, m o v i n g by stepwise

ascent or descent interspersed w i t h unaccompanied tones, cise att e n t i o n must s t i l l be given to d o u b l i n g i n four-part realizations
(Figure 262).

5. I n certain d i s t r i b u t i o n s even contrary m o t i o n cannot be

counted u p o n to avoid fifths, as illustrated i n Figure 263. Such
errors can be corrected t h r o u g h d o u b l i n g (a). B u t i n b, contrary
m o t i o n is good i n a l l d i s t r i b u t i o n s of tones except the case illustrated i n c.
Figure 263

44- 6 -6-

'jt

'#

'#

trtftr i*rf r t fV r r r f - 1
6

- 6 6

i
le

11 L
l^e
5 4

- 6 6

J.M^-A"^
p |
'

6. D o u b l i n g i n the u n i s n is often preferable to that i n the

octave, for i t helps m a i n t a i n cise position and makes a better upper voice, as i l l u s t r a t e d i n F i g u r e 264.

7. I f the accompanist does n o t look ahead as he should, errors

w i l l be avoided only by good fortune. I n the first of the f o l l o w i n g
ixamples, the octave of the bass must be retaken over the unaccompanied note i n order to prepare the f o l l o w i n g seventh (Figure
2 (5, Example a). Such basses are convenient to the accompanist, for
ihey give h i m an o p p o r t u n i t y to decide beforehand how each c h o r d
shall be realized. A t the same t i m e the expedient of Example a never
adds to the attractiveness of a setting. I n Example b the t h i r d of the
chord of the major sixth must be d o u b l e d unless recourse is taken
to a d i v i d e d accompaniment f o l l o w i n g a d u p l i c a t i o n of the bass.
T h e reason is that the f o u r t h i n the second c h o r d must r e m a i n
stationary. For a similar reason the sixth of the six-four c h o r d i n the
asterisked example must be d o u b l e d or, as an alternative (c), the
tone a must enter as an eighth note i n order to prepare the followi n g seventh.

6 6

7 6

f cr

11 1H
le \
5

6
4

1
6

8. I n the first of the examples of Figure 266 we see the use of a

varied d o u b l i n g as a means of a v o i d i n g octaves. I n the second example the same means must be more extensively employed for the
same reason. A consistent d i s t r i b u t i o n of tones is thereby m a i n tained and unnecessary leaps are avoided.

2 20

T II O R O U G II

li A SS

TU

Figure 266
%

M
I

6,

7 -6~

riTFf M r r f K

O H O U C, I I

li A

221

SS

11. W h e n Figure 250 is reali/.cd in lliree parts, the r i g h t hand

iniisl. not be widely separated froni the bass or the o u r t h s w i l l become too pronounced. B u t beyond this the accompanist need n o t
l'ccl uneasy about them. T h e y may ascend and descend by step or
leap, for the i n t e r v a l is created by upper parts and does n o t p e r t a i n
lo the bass. T h e only caution that must be heeded is to see that they
do not become fifths by i n v e r s i n .
121 W h e n the t h i r d or sixth of a chord is chromatically altered
by a s h r t note i n the p r i n c i p a l part over an u n f i g u f d bass, the accompanist may ignore the change and r e t a i n the chord he has already realized, even w h e n the tempo is slow (Figure 269).

9. T h e major sixth accompanied by the m i n o r t h i r d tends to

ascend. For this reason the second of the f o l l o w i n g realizations is
preferable to the first. T h i s precaution is most essential w h e n the
sixth lies i n the upper part.
6

Figure 267
18. A t times a succeeding chord or progression necessitates a
five-part realization of the c h o r d of the sixth (Figure 270).
Figur 270
6

10. T h e u n i s n d o u b l i n g may be employed more freely t h a n

octave doublings. I t may be used o n a chromatic tone as a means, for
example, of a v o i d i n g a leap (Figure 268). Inasmuch as composers
w r i t e inner parts i n this manner, even t h o u g h b o t h tones of the
u n i s n d o u b l i n g can be heard, the keyboardist is certainly justified
i n availing himself of the resource, for o n o u r i n s t r u m e n t we strike
and hear only one tone i n such a d o u b l i n g .

r
14. I t has been stated repeatedly that the augmented second
must n o t be used i n an accompaniment. Yet, since this progression,
w h i c h is as good as the d i m i n i s h e d t h i r d , is often used as a melodic
refinement, there are times w h e n the accompanist cannot be taken
to task for e m p l o y i n g i t . I n fact, i n such cases, an attempt to e l i m nate i t w o u l d distort the setting (Figure 271, E x a m p l e a). B u t aside
f r o m such a s i t u a t i o n i t is correctly avoided.
4

Figure 268

w
7
5

*". . . Which is as good as the diminished third" inserted in ed. of 1797. Since the
accompanying Figure 270 included, in the original, examples of the diminished third,
it would seem reasonable to conclude that the clause had been overlooked.

T // O R O U ('. I I

222

li A SS

ir

Figure 271

# rrrr
.

6l>

r f
6

r r i L

k
t

6^1

r*

T*~1*

O R O II (', I I

TU

6
4

7
*

4. T h e d i m i n i s h e d fifth is a dissonance which is i n t r o d u c e d w i t h

preparation (Figure 272, Example a) or w i t h o u t i t (fe). I t resolves
by stepwise descent.
Figure 272
a.

r e

6
6
5\ 5

5t

):

^=

TH

Ef

t_r
6

ff

l Ir
*>

J
1 -

r
*

11

rr
6l>

4+

T H E DIMINISHED T R I A D

1.

J \H ii
f*i
f

T h e d i m i n i s h e d t r i a d contains a m i n o r t h i r d and an octave

i n a d d i t i o n to the d i m i n i s h e d fifth. I n three-part accompaniment

the octave is o m i t t e d .
2.

I t is denoted by either n o figure or the usual signature of the

d i m i n i s h e d fifth (5b). I n sharp keys a n a t u r a l sign (5^) sometimes replaces the flat. Occasionally the r e m a i n i n g figures are i n c l u d e d i n
the signature.
3. For the sake of convenience the sign of the d i m i n i s h e d fifth
is often placed over a bass w h i c h is to be realized as a six-five chord.
Henee, voice leading alone must be the judge of whether the d i m i n ished t r i a d or the six-five c h o r d is intended. For the first m e a n i n g
Kapellmeister T e l e m a n n wisely places a bow over the 5 ('5 ).
W h e n necessary the accidental is i n c l u d e d (5^ ) . By this means confusin is ended, and novices w h o lack a penetrating knowledge of
voice leading are spared embarrassment.

22)

li A SS

T-

5 6

T#

1 J r, J 1 J1J j ii J 1 J J1J
R

rs
6 5t4

3-

r 1" r
1

RR

rg

toa

11

4;

5t>
3

7 6

^
5

5. T h e d i m i n i s h e d fifth appears more often w i t h other figures

than i t does w i t h the octave and t h i r d , as we shall see later. Its t r i a d
sounds w e l l i n three parts b u t rather empty i n four. W h e n the t h i r d
is d o u b l e d i n place of a d u p l i c a t e d bass, a l l of the upper parts f o r m
consonances w i t h each other, thus m a k i n g the c h o r d more acceptable. I t sounds poorest w h e n the octave of the bass is i n the upper
part. T h e p r i n c i p a l responsibility of the cautious accompanist is
the disposition of the chord rather than its d o u b l i n g , for a choice
of the latter is occasionally restricted by the r e q u i r e d resolution of a
dissonance (Figure 273).
Figure 273
6
5

9 8

1 Cf. Arnold, op. cit., pp. 506-510.

2 Cf. Pt. II, Foreword, Note 4.

6. W h e n the bass of this c h o r d is chromatically raised, the octave

is o m i t t e d and the t h i r d d o u b l e d (Figure 274, Example a). T h e same
d o u b l i n g is often used i n order to m a i n t a i n good voice leading and
to avoid a w k w a r d leaps (fe).

2 2./

T II O R O U C II

T I I O R O II C U

li A S S

Figure 274

00

'

Figure 276

wrong
5

225

limes use it as a melodic relinement i n place of the perfect fifth

(Figure 276, Example a). More often i t appears before a retarded
sixth (b). O n occasion a chromatic change calis for its e m p l o y m e n t
w i t h o u t its being indicated (c).

a.

BASS

" J.

7. I n the m i n o r mode, the chord on the second step contains

a d i m i n i s h e d fifth regardless of whether the octave or the major
sixth is also present. I n Figure 275, w h e n the bass is unfigured b u t
the p r i n c i p a l voice is w r i t t e n i n , the figures w h i c h have been placed
under the bass are best because of the succession. A leap may be
m d e to an unprepared d i m i n i s h e d fifth (a). As part of a t r i a d this
dissonance is treated more freely than i n other relationships.

4. I t does n o h a r m to the t r i a d to o m i t the octave and double the

t h i r d , for this establishes consonant relationships among the upper
parts (Figure 277).
Figure 277

Figure 275
Adagio

1^ T^-

44- -5- 6

a.

1*

^ J
}

f r

1
1

T H E A U G M E N T E D TRIAD

* r r

1. I n four-part accompaniment, the augmented t r i a d consists

of a major t h i r d , an octave, and the augmented fifth. I n three-part
accompaniment the octave is o m i t t e d .
2. Its bass carries the i n d i c a t i o n of the augmented fifth ( 5 , 510;
or the altered fifth and the appropriate r e m a i n i n g intervals.
3. T h e augmented fifth is a dissonance w h i c h is n o t easily i n t r o duced unprepared. I t resolves by stepwise ascent. Composers some1 Cf. Arnold, op. cit., pp. 512-513.

5. Inasmuch as the augmented fifth is most frequently used as a

refinement, i t is better fitted to three- than to four-part accompaniment. T h e latter is n o r m a l l y used w h e n the signature includes additional figures.
6. A slow chromatic progression that features the augmented
fifth is accompanied i n three parts. Such half steps i n the p r i n c i p a l
part are n o t easily adaptable to r a p i d tempos; b u t w h e n they oceur
the accompanist omits them (Figure 278).
Figure 278

J
1
r - r

Adagio

|6

i5

J> , J M I
sS
T
-5-

16

-5-

'6

T II O II O U C, II

22f>

li A S S

T H E SIX-FOUR C H O R D ,

r 11 o no

1.

I n four-part accompaniment the six-four chord contains an

octave i n a d d i t i o n to the intervals w h i c h give i t its a m e . I n three
parts the octave is o m i t t e d .
2. T h e signature, 4, is a l l that is r e q u i r e d to denote the chord.
3. T h e m i n o r and major sixth and a l l three kinds of f o u r t h may
be expressed by i t . Henee i t contains only one dissonance, the
f o u r t h . T h e qualities of the intervals are recognizable f r o m the key
signature and the a c c i d e n t i s added to the signature of the chord.
4. T h e d i m i n i s h e d f o u r t h requires preparation (Figure 279,
Example a), b u t the perfect and augmented fourths may be i n t r o duced either w i t h or w i t h o u t i t (b). Because the six-four chord w i t h
the augmented f o u r t h offers the fewest useful examples, I have had
to cali o n the chord of the second, w i t h w h i c h this interval is most
frequently used, i n order to illustrate its characteristic behavior.
T h e d i m i n i s h e d and perfect fourths resolve by stepwise descent, b u t
the augmented f o u r t h by stepwise ascent over a descending bass.

Figure 280 g
6
4

ucu

227

IASS

m
i n i r r r i p r m
U
3

6
4

ir
6

|7

i
6

Figure 279

.1..

II

I.

II

I.

II .1

rr T T T T TTr'r'rr"
43

43

6 4 3

46
2

8. T h e perfect f o u r t h may be accompanied by either the major

or m i n o r sixth. I t may resolve directly to a five-three c h o r d (Figure
281, Example a), although i t is n o t always r e q u i r e d to do so. T h e
bass may r e m a i n stationary or progress, for the succeeding figures
are often different from the expected ones, and this sometimes retards b u t never disrupts the resolution (b).
Figure 281
65
a 4 3
Y

6 4 6
2

6 4
b. 4
2
II
H
H

6 t>7
4
5

65
4 - 3

p
II
1|

6t 5
4 3

5. Those w h o k n o w the t r i a d that is b u i l t o n the f o u r t h above

the bass w i l l recognize the tones of the six-four chord.
6. W e shall learn soon that the sixth, a consonance, may be
d o u b l e d w h e n there is good reason to do so. Even though the octave
of the bass is o m i t t e d w h e n this takes place, n o tone of the chord is
thereby lost.
7. O f the three kinds of f o u r t h , the perfect is the least dissonant
i n this chord. Nevertheless i t must be resolved, unless i t is used
as a passing interval. I n this latter case i t may be d o u b l e d i f necessary and i f p e r m i t t e d by the preceding signatures. T h e examples of
Figure 280 illustrate the passing f o u r t h .
1 Cf. Arnold, op. cit., pp. 536-540.
2 This sentence appears in the original ed. as a footnote.

A/ *
/

8 7 6 5 4 - 3

6 7 4 - 3
1 2 1

8 7 1>7
6 5 4 - 3

5 6
3 4

^ w
4
2

11

^6

9. W h e n the t h i r d of a chord of the sixth is retarded by a f o u r t h ,

the progression, a delicate one, is best expressed i n three parts. I f a
four-part accompaniment is r e q u i r e d , the octave is o m i t t e d and the
sixth d o u b l e d . Examples a and c of Figure 282 illustrate the occurrence of this chord before a six-five c h o r d w i t h a d i m i n i s h e d

; //

2<V

T II o no u
li A ss
l i l t h . A l l three kinds Of Eourth and the two consonani sixtlis may be
employed, provided that the fourths are prepared and move by
Stepwise descent. T h e progression is m u c h used i n our present,
agreeable style, b u t i t never contains the octave. H o w essential
therefore is some i n d i c a t i o n that m i g h t make i t recognizable to the
6
u n i n i t i a t e d ! W e shall choose the sign 4.
10. W h e n the f o u r t h is d i m i n i s h e d , the sixth is m i n o r (Figure
282, Example a); w h e n i t is augmented, the sixth is major (b); b u t
when i t is perfect, the sixth may be either q u a l i t y (c) as we have already learned. W i t h regard to the asterisked example, i t does n o t
occur freely unless the bass first ascends and then descends. I n the
4
6
last t w o examples the best d i s t r i b u t i o n is that w h i c h follows 3 and 5
i n open p o s i t i o n .
A
A
Figure 282
6 5 \
6 a.
43
4 3
3

43

5 l>

6
4

4 3

44

5i>
3

4 3

4-

4 3

44 3

O RO

IJV.II

HASS

Li

6
5i> 4

5b
3

6
51 4

5
3

5t>

6 5 b
5i 4 S
G

T h e 4 i n the examples of F i g u r e 283 is a passing chord. T h e underl y i n g relationships are those of Figure 284.
Figure 284
6

b7

-fe
12.

5 \> -

7 6

5 \>

5 \-

r| s

L=i*

1
-.

W h e n the sixth is major i n this six-four c h o r d and the re6

tarded t h i r d is m i n o r , the four-part accompaniment is 4 (Figure

4-

6
43 5

285).

Figure 285
9

r ir

44 3

11. W h e n this k i n d of six-four chord follows a d i m i n i s h e d fifth

over a stationary bass, a three-part accompaniment is best. I f a
f o u r t h part is r e q u i r e d , the sixth is d o u b l e d and the octave of the
bass o m i t t e d (Figure 283).
Perhaps because the initial chord sounds more sonorous when the upper parts
lie a seventh rather than a second apart. At least, neither disposition present any
challenging problems of voice leading.
3

Figure 283 A
l> 7
6
5t> 4

w
A

TU

#
/

\>
Y
6

m
m
r
4

m
r

rn

s ~ m
m
r
m
r
5

KI

" pm
1

13. W h e n the augmented f o u r t h is used i n a passing relationship, the bass is not always r e q u i r e d to progress by stepwise descent
(Figure 286, Example a). T h e second example takes only a threepart setting. I n b the augmented f o u r t h over / anticipates its n o r m a l
entrance, w h i c h is o n the f o l l o w i n g e i g h t h , c, where i t functions
as a passing tone m o v i n g to a major sixth. I n the last example tlie
sixth above / may be d o u b l e d w h e n the t h i r d above b lies o n top.
T h i s is the best disposition here.

r no

o v c u aA ss

i j > i y ,H;.J
i

8 7 - 6
6 5 - 4 +

- 7
5

144
2
6
4

1. I n the examples of Figure 289, i t is perhaps better to transfer

the resolution of the d i m i n i s h e d fifth to the bass of the six-four
c h o r d and double the s i x t h t h a n lead the voices i n a strictly correct
manner w i t h the d i m i n i s h e d fifth resolving to the octave of the
second note i n the bass. T h e latter progression always makes the
six-four c h o r d sound ugly. For this reason the figuring of Example b
is better than that of a.
Figure 289
1>7

a. 5b

r r rr

' r

6
4

44

63 -4

Figure 287

6 5
4 3

4, 3

5
4 *

6
4

5
3

15. W h e n the bass of a six-three chord ascends a step to a sixfour chord, the bass may be duplicated or the other intervals
d o u b l e d unless the t h i r d lies o n top. W h e n i t does lie o n top, this
t h i r d should be d o u b l e d i n either the octave or u n i s n (Figure 288,
Example a). I f this is not done, even opposite m o t i o n cannot prevent fifths (b). W h e n the sixth is major and the t h i r d m i n o r i n this
progression, the octave of the bass may be taken, provided that the
sixth, i f possible, b u t not the t h i r d , is placed i n the upper part (c).

a.

5b

9 8

6
4

a.

Ai

a.

w aij

JiPig

16
4

16

'6
4

5
3

b.

b.

Vi

4,

5b

I J If p

b.

5b

IJ |

9 8
6 4 3

2. I n Figure 290 the d o u b l e d sixth of the six-four c h o r d i n a

four-part setting must be released immediately o n the entrance of
the d i m i n i s h e d t r i a d :
Figure 290

3. T h e augmented f o u r t h i n the six-four c h o r d sounds rather

empty i n four-part settings b u t is i m p r o v e d by the a d d i t i o n of a
second or a t h i r d . T h i s chord, the signature of w h i c h must i n c l u d e
the s i x t h along w i t h the f o u r t h , may have a duplicated bass or a
d o u b l e d sixth. T h e latter construction is used n o t o n l y because i t
sounds w e l l w i t h its upper parts consonant w i t h each other, b u t also
Figure 291

Wrong

Figure 288

16

14. W h e n the bass of a c o m m o n chord or the six-three c h o r d

descends one step to a six-four chord, the d o u b l i n g of the first c h o r d
must be such as w i l l avoid octaves (Figure 287).

S S

T H E SIX-FOUR CHORD, I I

Figure 286

15
6
3 4 4

li A

T H O R O If C, H

M
6
4

better

6 6

T u o n o u r,

2 ?2

II

ti A ss

T I I O li O U (1 I I

because i t is sometimes required (aside from those cases where 4 is

specified) i n order to n i a i n t a i n flowing parts and avoid errors (Figure 291.)
4.

T h e six-four c h o r d w i t h a perfect f o u r t h sometimes arises o u t

SS

Figure 294

" r r ?

4t

THE FOUR-THREE CHORD/

1.

r c. c
44

U
5.

16 ^

4jf q5

1)5

6
4

A six-four c h o r d w h i c h is created by the action of an ascend'IT

i n g changing note after6, is realized only i n three parts (Figure 293).

Figure 293

33
?6
6 4

6.

I n Figure 294 i t is w r o n g to denote the c h o r d of the s i x t h o n

the second note as either 4 3 or 4. T h e f o u r t h o n the last sixteenth

2
serves only as a refinement w h i c h passes decoratively to the appoggiatura before the final note. T h e u n d e r l y i n g progression is illus-

2??

trated under a. I n passing, observe that the octave above /-sharp

rather than the fifth should be taken as a f o u r t h part because of
the preceding c.

of a retardation of b 5, i n w h i c h case i t is realized i n three parts. T h e

perfect and augmented fourths must n o t be confused w i t h each
other, especial ly w h e n the former is indicated by a c c i d e n t i s w h i c h
n o r m a l l y p e r t a i n to the augmented f o u r t h , as i n the f o l l o w i n g examples (Figure 292).
Figure 292

li A

T h i s c h o r d consists of a t h i r d , f o u r t h , and sixth.

4

2.

Its signature is 3. T h e eye is more accustomed to this indica3

d o n than i t is to the occasional 4. T h e sixth is i n c l u d e d i n the signature w h e n i t is chromatically altered (Figure 295, Example a), resolves a dissonance (b), or moves to a passing tone over a h e l d
bass (c).
7 6l?
6 7
Figure 295
5b4
4 6
-'
6
3 5
6.
3
- 6
22
g

3. T h e intervals that appear i n the c h o r d are the m i n o r , major,

and augmented sixth, the perfect and augmented f o u r t h , and the
m i n o r and major t h i r d .
4.
T h e exceptional features of the c h o r d are that the t h i r d is
treated as a dissonance and the f o u r t h enjoys more freedom than
usual. T h e former is usually restricted by the latter, and always
resolves by stepwise descent. T h e f o u r t h remains stationary or
ascends. These progressions w i l l be clearly i l l u s t r a t e d and discussed
i n detail w h e n we examine a l l types of four-three chords, an undert a k i n g w h i c h is necessitated by the great v a r i a t i o n i n the behavior
of b o t h fourths, the stationary and the ascending.
2

5.

W h e n the c h o r d consists of a major sixth, perfect f o u r t h , and

1 Cf. Arnold, op. ext., pp. 628-646.

Le., except when the fourth is omitted (cf. \, par ex.).
2

2^4

T 11 ORO U G 11

liASS

m i n o r t h i r d , either the f o u r t h or the t h i r d must be prepared. Most

frequently i t is the t h i r d , w h i c h then moves by stepwise descent.
T h e f o u r t h remains stationary. T h i s progression also appears over a
4

tied bass, and i t is sometimes denoted by a simple 6 rather than 3.

T h e bass progresses by stepwise ascent or descent. I n ascent i t proceeds to a c h o r d of the sixth, i n descent to a c o m m o n chord. Those
w h o k n o w the tones of the six-four c h o r d can easily find the fourthree chord by o m i t t i n g the octave of the bass and replacing i t w i t h
the t h i r d (Figure 296).
Figure 296
4
6,
^

|-1

6.

1 r

-6.
4

4
3

nature 3 expressly. I n the second example the chord of the sixth is

4

35

7. I n three-part accompaniment one tone of the chord is lost.

Yet certain relinements are met o n occasion w h i c h are n o t suitable
to four parts. For example, the expression may r e q u i r e a soft performance w h i c h an accompanist m i g h t not be able to achieve o n a
resonant i n s t r u m e n t unless he used a t h i n setting, etc. I n such a
case one interval must be o m i t t e d . I n Figure 296 the f o u r t h may be
left out, b u t i n Figure 297 a four-part realization is presupposed.
8. W h e n the c h o r d consists of a major sixth, augmented f o u r t h ,
and major t h i r d , either the f o u r t h or the t h i r d must be prepared.
T h e latter descends afterwards w h i l e the f o u r t h remains stationary
or ascends. T h e bass may be tied or not, and i t progresses by step4

4f

T h e rather unusual examples of Figure 297 r e q u i r e the sig-

clearly better than 3:

Figure 297

O R O (7 G 11 li A SS

wise ascent or descent. H e r e the signature 3 or 3 is r e q u i r e d more

-6-

II

than i n Figure 296, for a simple 6 or 4 as the signature can easily

cause c o n f u s i n . T h e disposition w h i c h sounds best is that i n w h i c h
the f o u r t h and t h i r d are separated from each other. Example a of
Figure 298 may express fifths i f the t h i r d of the preceding c h o r d of
the sixth is d o u b l e d . Such b e i n g the case, they can be avoided by
placing the f o u r t h on top (b). I n three-part accompaniment the
sixth of the four-three chord may be o m i t t e d except i n Example a.
T h e figures that stand below the notes have n o bearing o n those
w h i c h stand above.
6
Figure 298
44
44
4
6
3
3

1 I~=*J

44
3

'

|p* | |4 | 1
1

|F = j =

- > >

J Ji
t

4
3

11

6
4
JL

'

6 5
4

a.

4
3

9. * W h e n the c h o r d contains a m i n o r sixth, perfect f o u r t h , and

m i n o r t h i r d , either the f o u r t h or the t h i r d must be prepared. T h e
3

This sentence appeared in the original ecl. as a footnote.

* In all eighteenth-century editions this paragraph was misnumbered 8. In correcting it, all following paragraph numbers have had to be increased by one.
8

2}6

T II O R O U G II

/* A

SS

T II O R OUC,

{QUrth remains sttionary and the t h i r d descends. M e piogression

may be i n t r o d u c e d over a tied or u n t i e d bass which descends afterwards. Example a of Figure 299 w i l l be f o u n d occasionally, alt h o u g h i t is n o t especially good. T h e major sixth makes a better

Figure 300
6

Figure 299

-Ai

5 6b 4
5b

6
b>

rN
LL\I

^ 4
5&3[

11.

"Jal

6
m

p *IH

in

SS

237

pp

44 7 6
3 -5- 6

44
3

a.

p
# 3 6

bass may r e m a i n o n a single tone as i n the organ pon, or i t may

progress. T h i s c h o r d sounds best w h e n the t h i r d a n d f o u r t h are

4 b7 6

5,

b7
6b

li A

4-1

44

progression (b). T h e signature of the c h o r d is 3, b u t w h e n the s i x t h

is lowered i t is i n c l u d e d w i t h its accidental. I n the second a n d t h i r d
examples there is only one good disposition. T h e sixth must l i e o n
top, because of the succeeding six-five c h o r d . O t h e r realizations cont a i n fifths. I n three parts the f o u r t h is o m i t t e d f r o m a a n d b:

II

5b-

1S
i

separated f r o m each other. Its signature is 3. W h e n the bass is held,

the c h o r d is realized i n f o u r parts; otherwise the f o u r t h is o m i t t e d .
A c h q r d o f the sixth is better t h a n the four-three c h o r d i n the last
t w o examples of F i g u r e 301.
Figure 301
6

10 9

4 - 3
3 2
3

6 7
4 -

J J

4
3

10. W h e n the c h o r d contains a major s i x t h , augmented f o u r t h ,

and m i n o r t h i r d , the f o u r t h or the t h i r d is usually prepared. I n
Example a of F i g u r e 300 b o t h enter freely over a passing note whose
i n i t i a t i n g tone has been elidd (b). T h e t h i r d resolves b y stepwise
descent, the f o u r t h "by ascent. T h e bass may, b u t need n o t be, t i e d ;
. 4 + 4

4t-

subsequently i t descends by a step. T h e signature is s' j>3, or \3

Those w h o k n o w the tons of the c h o r d of the second w i t h a raised
f o u r t h can lcate this four-three c h o r d q u i t e easily by t a k i n g a
t h i r d above the bass instead of the second. Except i n the asterisked
example, the sixth is o m i t t e d f r o m three-part realizations.
11. W h e n the chrd contains a m a j o r sixth, perfect f o u r t h , a n d
major t h i r d , either the f o u r t h or the t h i r d is prepared. T h e t h i r d
resolves by stepwise descent; the f o u r t h remains sttionary. T h e

6.

4
3
= =

-fr5b

4 ^
=

^ ^ >J

6
6

= = :

=F-m

m-yt

tri
3

II

t
3

12. W h e n the c h o r d contains an augmented s i x t h , augmented

f o u r t h , a n d m a j o r t h i r d , p r e p a r a t i o n of the sixth is o p t i o n a l , b u t i t is
r e q u i r e d of either the f o u r t h or t h i r d . T h e t h i r d progresses subseq u e n t l y by stepwise descent. T h e f o u r t h remains sttionary or
ascends. T h e bass may, b u t need n o t be, tied. I n either case its subsequent m o t i o n is stepwise descent i n parallel m o t i o n w i t h the
t h i r d . M a n y denote this c h o r d ambiguously w i t h only a s i x t h a n d

2fS

T II O Ii()

UGH

T H O H O U (i II

BASS

its accidental. The beit signature expreSSM all of the intervals. I n

threc-part accompaniment the f o u r t h is o m i t t e d (Figure 30a).

Figure 302 *
&

A
3
^

w[ p

~64
3

* 8

6 5
4 #

K r.

A
44
6
3
.
n(
w 1 i t U

-4

6~64

4 f}

3
11 F

?p1

6 8
4 7.

A S . 4 j

5
tt

6
4

y J r r r r^nTT

^64
3

11

Figure 3<34

y
8

-41

Iti

7 8
5 6
3 3

w rong

} ni f t
'tr' i

s 11
- f M i
\: .

-6'6
4

* r

44
9 3

< V

M
4

]e
1

= t =

rr

r rr

1
5

6
4

|6
5

'6
5

wr ong

J L = I =

r-

11 !

Hr1
1

-t

8
6
443

le

a.

II

H^-

<

]f

wrong

a.

4#

ture. I n Example a, *enters prematurely as the result of an elisin.

A c t u a l l y , the n i n t h , seventh, and f o u r t h should be resolved first, as
i n Example c. O u r c h o r d thus turns o u t to be merely a passing
Figure 303

FOUR-THREE CHORD,

1. T o the chord of the m a j o r sixth and m i n o r t h i r d , the perfect

f o u r t h is sometimes added w i t h o u t express i n d i c a t i o n (Figure 304,
Example ) i n order to a v o i d the errors of aa, to gain a u n i f o r m
disposition (b) w i t h o u t the leaps of c, or to crate a good u p p e r
part (d).

13. Occasionally an octave must be taken i n a d d i t i o n to the

other intervals, n o t merely f o r the sake of a f u l l setting, b u t more
i n order to resolve a preceding dissonance (Figure 303, Example a),
or to prepare a succeeding one (b). T h i s b e i n g the case, i t is best to
place a l l f o u r intervals i n the signature a n d thus preclude conjec-

8
y -6.
5 44
4 3

239

chord, w h i c h accounts for the lcaping rather than stepwise movem e n t of the bass.
THE

78

liASS

11

:, 1

IB

r r

II
II

*
-

1 i-

T I TT
w

2. I n Example a of Figure 305 the six-four-three c h o r d sounds

very w e l l against the l o n g appoggiatura. T h e succeeding figures are
i n most cases already present i n the first c h o r d . T h e upper voice

240

T 11 O R O U G H

li A S S

THORO

contines m e l o d u ally i n thirds w i t h the bass. Kxample b, set simply, suffers neitlier a d o u b l e d t h i r d above d, or an ascent of the
t h i r d to g, for this t h i r d , f, is also the seventh above the second
eighth, g, and must therefore be p r o p e r l y resolved (c). T h e dissonances of Example d are conveniently prepared by the four-three
chord:
^
poor
Figure 305

^ '

'6
6

|7

~"

'-*s^

'6
5

17

UGH

liASS

241

t h o r o u g h bass that a l l o w for variants and yet cannot always be

realized o p t i o n a l l y make i t a l l the more necessary to look ahead and
listen carefully.
Figure 306

2
e

/ -

r 1 m -i
3.

I n the examples of Figure 306 the c h o r d of the s i x t h is taken

over each d. T h e four-three c h o r d sounds ugly against the repeated

6

cjjr

4. Some consider i t sufficient to signify 4 after 5 i n the f o l l o w i n g

example, since this specifies the progressing parts (Figure 307, Example a). However, an inexperienced accompanist m i g h t add the
octave of the bass to the six-four ch ord according to the r u l e of cons t r u c t i o n of this c h o r d rather t h a n h o l d the t h i r d . T h e figuring of b
is clearer a n d m o r e correct, despite the fact that the eye must sean
an a d d i t i o n a l figure.

a (a). A chromatic a l t e r a t i o n sometimes prevens the t a k i n g of 4

(b). A g a i n , passing notes may r e q u i r e the simple c h o r d of the s i x t h ,

as i l l u s t r a t e d here by the / (c). O t h e r factors w h i c h oblige the ac6

companist to play an i n d i c a t e d c h o r d of the s i x t h rather t h a n 4 are:

the succeeding figures, w h i c h may be m o r e easily realized by h o l d i n g or repeating the c h o r d of the sixth (d); the succeeding notes,
w h i c h may prevent a proper r e s o l u t i o n of the t h i r d of 4 (e); errors

3
4

of voice leading w h i c h are caused by 3 (/). Those progressions i n

5.

F i g u r e 308 is exceptional, a n d its accompaniment can give

4

nse to several errors. T h e t h i r d of the first 3 does n o t resolve b u t

remains sttionary to become a f o u r t h , f o r the c h o r d is to be re-

242

THOROUGH

IASS

TH

garded as passing. T h e augmented f o u r t h , however, ascends charac4

teristically. T h e second s bchaves n o r m a l l y . I n order to realize the

first signature completely a n d make i t possible for the augmented
f o u r t h to ascend, the second of the chord should be d o u b l e d .

6.

I n F i g u r e 309 a unisn d o u b l i n g of the sixth (a) makes a

better progression than a d o u b l e d t h i r d

(b):

ASS
presupposes

an

ac-

Figure 311

LL ' C

9. A l t h o u g h the earlier statements concerning the three-part

realization of o u r c h o r d are generally applicable, the accompanist
must be attentive to the p r i n c i p a l part, for at times the tones perf o r m e d by i t may be o m i t t e d f r o m a l i g h t accompaniment (Figure 312).
Figure 312
piano

Figure 309

44*^

ccled i n the 3 , for the augmented f o u r t h

companying major s i x t h .

Pe'

O R O V C, II

4+

gil
44

J'

g j .

61

, J .

rr T

Figure 310
6

pal

*r, r f r r f r

m
6

7. O n e of the best uses of the augmented f o u r t h w i t h the augm e n t e d s i x t h is the f o l l o w i n g . I n other cases this latter i n t e r v a l
usually sounds better w i t h the fifth or a d o u b l e d t h i r d . T h e b,
t h r o u g h its presence i n most of the chords, provides an i n t e n d e d

.
obstinacy w h i c h heightens the effectiveness of 4 (Figure 310).

= |J

. J

r r 1f

g JJJi nf ifrnrrnf
r r r

r r

8. I n Figure 311 the melody of the p r i n c i p a l part, a n d to a

greater degree the preparation of the t h i r d , necessitates a v i o l a t i o n
of the r u l e p r o h i b i t i n g the progression of a d i m i n i s h e d t h i r d . T h e
preceding chromatic a l t e r a t i o n of the s i x t h is automatically can5

5 Changed in ed. of 1797 from the earlier "augmented second."

THE

i.

SIX-FIVE

CHORD,

T h i s c h o r d consists of a s i x t h , f i f t h , a n d t h i r d .

2. Its signature is 5 , o r 5b alone w h e n the fifth is d i m i n i s h e d .

T h e t h i r d , w h e n i t is chromatically altered, is i n c l u d e d w i t h its
accidental. Also, accidentis w h i c h p e r t a i n to the s i x t h a n d fifth
must n o t be o m i t t e d f r o m the signature.
3.

I n this c h o r d there are three sixths, augmented, m a j o r , a n d

i Cf. Arnold, op. cit. pp. 602-626.

244

T II O RO

U G II

Ii A S S

T II O R () V ; //

m i n o r ; two fifths, the d i m i n i s h e d and perfect; and t w o thirds, the

major and m i n o r .
4. T h e fifth is treated as a dissonance. I t is usually restricted by
the sixth, and always progresses by stepwise descent.
5. T h e perfect fifth does n o t readily occur w i t h o u t preparation
(Figure 313, Example a); b u t the d i m i n i s h e d fifth may l i e i n the
preceding c h o r d or enter freely (fe). However, w h e n this latter interval is taken unprepared, the sixth is usually present i n the precedi n g c h o r d . As the fifth, especially the d i m i n i s h e d fifth, resolves, the
bass n o r m a l l y ascends one step. Example c shows us that o n occasion
the bass may r e m a i n sttionary or leap u p w a r d and d o w n w a r d , i n
w h i c h event the resolution of the fifth is often retarded. I n the last
example under c there is an interchange of chordal tones and an
elisin, as i l l u s t r a t e d i n d.

6
5

Figure 313 6
5
a. 6
-f
V:
m

a.
1

= M

>

*F=t

<

6
5

ff=+
m

5b
0

6
5

c.

11

-75

s.
5b

t u

4=

a.

i
1

5 1

b.

5b

6
5

c.

=t

> 6

'
>

jt
m i

6
5
1 p

= F

c.

5b
,si

4
2

5b
P

'" r

=** -J1

7
5

d.

4 3

4 jt

'r., r
4 #

6 5

fi

-6.
5b

1i

H
1

rr

'44.
3

7. A t times the octave of the bass must be taken as a fifth part

because of the resolution (a) or preparation (fe) of a dissonance
(Figure 314).
Figure 314

S i

2
6-6-

6 5
4 |

g 8"

4-1.

1i

6. Those w h o k n o w the progression 6 5 or 5 6 can easily lcate

the tones of o u r c h o r d by o m i t t i n g the octave of the bass, p l a y i n g
the adjacent figures simultaneously alng w i t h the t h i r d , a n d att e n d i n g to preparation.

5b 5b

5s
5 b

1>b

*JF=

-64
3

6 5
4 }t r.

6
5

7
5

.1

5
5

>C

f=Ff=

-K-

b.

6
5

6
5

c. 6

1=1

2 I.e., except when the sixth is omitted, when the fifth is diminished (in which
case it is already a dissonance, henee, restricted anyway), or when the chord is a
"passing chord" (cf. f 10 here).

"6,

6
5

6 5
4 tt

b.

6 b9 8
/
, 5 7 *-5,b7|6-'
)t
>

8. W h e n the signature 5 4 appears over a sttionary bass, the

t h i r d is o m i t t e d and the octave of the bass taken as a f o u r t h part, for

2J

TII
O R O U
II i A S S
the u n d e r l y i n g construction is simply a six-four c h o r d w i t h its
f o u r t h retarded by a fifth. I n such a case the fifth is perfect a n d prepared (Figure 315, Example a). B u t should the fifth succeed n o t to
T b u t to some other signature (b) or should the bass move s i m u l -

7*7/

0U

G II

1u

17

6
5

lk

1<

51'

TT

i fe 1 j j

J 1 =\

;j

fe W-

4 - 3

**

9-

jjt

1
f

(>

t
7

6
5

s Cf. Part I I , Foreword, Note 4.

6
5

44
3

H"H*7

76 6 T J

jt

^ F i
M I
a

*t t
5

7
5

5 r

c .
4 jt

7 6 4 4

I
6 t|

5b

a.

8 7 6 4 - t l i r

||

p b7 6 6
~t~\>7

-o-

9. W h e n the c h o r d contains an augmented sixth i t always takes

a perfect fifth a n d a major t h i r d as its r e m a i n i n g intervals. T h e
fifth is usually prepared (Figure 316, Example a); b u t i t a n d the
sixth too may be taken freely w h e n the bass is sttionary (b). N o r mally, the sixth should enter one eighth later, as illustrated i n c.
Subsequently the sixth progresses by stepwise ascent, and the fifth,
m o m e n t a r i l y sttionary, by stepwise descent:

b.

Figure 316
a.
-

87

6-6.

8 7

Figure 315

Isb

J s ^ j J

^^^^^^

SS

taneously w i t h the fifth (c), the usual 5 accompaniment is to be

taken. For the benefit of the inexperienced accompanist, the first
of these six-five chords (they of ten appear over organ points) should
be distinguished f r o m the n o r m a l construction by means of a Telem a n n b o w . T h e last example u n d e r a is n o t e w o r t h y because of the
d i v i d e d accompaniment and the d o u b l e d s i x t h :

c>i

li A

8
5

h
1

-. 6 6
5
5

-6-

' t t i h
I 1

r, f 7

F*
5

II

10. Since three-part accompaniment necessitates the omission

of one interval, i t should n o t be employed w i t h o u t good reason. B u t

2./.V

T II O R O II G II

HA

T II O H O (I (, II

SS

when i t is r e q u i r e d , t h e o m i l l c d intctval mus be dccided u p u n caref u l l y . T h e t h i r d , the perfect fifth, or the sixth, especially w h e n the
last-mentioned is accompanied by the d i m i n i s h e d fifth, may be
o m i t t e d according to circumstances. W h e n the chord is used as a
passing construction, the fifth does n o t resolve. I n such a case the
t h i r d serves n o good end and may be o m i t t e d i n favor of the sixth

SS

Figure 318

5b

* 6

5b

1h

and fifth. I n the examples of Figure 317, the chords that preceed 5
and provide its sttionary fifth are set u n i f o r m l y i n three parts. Inasm u c h * as we have made use of the T e l e m a n n bow i n other situations i n order to distinguish a three- f r o m a four-part realization, i t
may be i n c l u d e d here i n the signature of a six-five c h o r d whose
t h i r d is to be o m i t t e d .

li A

1 V

6
5

5b

5b

6 * 6
9

~~6-
3 5b 6

6 5b

2. Those w h o set Example a o Figure 319 w i t h its unprepared

perfect fifths (an occasional b u t undesirable progression) must defend i t as an elisin of the succession 6 5 (b) or as a m a n i p u l a t i o n
of 8 7 (c). Over the bass g of bar one, and /, bar t w o , the sixth must
be separated f r o m its preparation as a means of avoiding fifths (d).
T h e fifths by contrary m o t i o n over a and c i n the first bar cannot
be, or indeed need they be, circumvented. Example e is even worse
than a:
Figure 319
a.

H 1i
J

5
c 8 7

THE

SIX-FIVE

CHORD,

II

1. T h e examples of F i g u r e 318 show us h o w the sixth, as w e l l as

its accompanying d i m i n i s h e d fifth, may enter unprepared (a). T h e
u n d e r l y i n g progression w i t h o u t elisions is illustrated i n b. T h e
fifth of the first bass note must n o t be placed o n t o p , for the
d i m i n i s h e d fifth progression belongs' i n the i n n e r parts.
* T h i s sentence appeared in the original ed. as a footnote. For Telemann bow, cf.
Part I I , Foreword, Note 4.

I J J
5

8 7
3Z

"

ii
55b
9

6 5

6 5

r r f r1

wrong

3. T h e d i m i n i s h e d f i f t h may enter freely; and even w h e n i t can

be prepared, i t may be separated f r o m its preparatory interval for
certain justifiable reasons. T h e y are: the r e s o l u t i o n of a dissonance
(Figure 320, Example a); the r e t e n t i o n of a convenient spacing a n d
the c o n t i n u a t i o n of a good melody (b); the avoidance of a poor
relationship between the outer parts (c). I n the absence of these

250

TU

O R O V C, II

li A

SS

TU

O R O II C II

li A

SS

25/

factors, howevcr, strict observance must be made of the rules w h i c h

cali for the preparation and resolution of a dissonance i n the voice
i n w h i c h the dissonance appears.
Figure 320

m
7

6
5b

5b

5b

poor

5^

5b

4. A t times i t is better to o m i t the sixth and double the t h i r d

i n a chord c o n t a i n i n g a d i m i n i s h e d fifth, even t h o u g h the sixth
w o u l d n o t clash w i t h chromatic changes. T h e reasons are: T h e
sixth, taken freely, may add a discordant element to the resolution
of a preceding dissonance (Figure 321, Example a); the d o u b l e d
t h i r d may help to m a i n t a i n a good melody (b) or to avoid errors. (c).
Figure 321

'6

not good

u
4

6
5

poor

9
4

'6
5b

5. As a r u l e , all chromatic changes that p e r t a i n to the t h i r d must

appear i n the signature, as i n the case of a l l figures. Nevertheless
there are times when the accidental is o m i t t e d , under the assumpt i o n that the q u a l i t y of the t h i r d w i l l be k n o w n f r o m its context. I n
Figure 322, i f a d i m i n i s h e d t h i r d is to be played i t must be expressly
specified by a fat over the bass:
Figure 3 22
g

*r r r
*y

a . = =

b5

k
m

11

Figure 323 should please the devotees of strange chords. I t

,
X
6
6
gives passing expression i n a slow tempo to 5 , 4 , and 4+ ,
6.

7b
5b

tne latter t w o i n open position.

T II o n O U C, II

2^2

If A

SS

THORUGR

Figure 323

U
9

9
5

6
4
3_

6
44
3

6
4

5
*
*

7. I n a slow tempo w h i c h requires a soft performance the sixth

may be o m i t t e d f r o m the six-five chord i n the first example of Figure
324 a n d the t h i r d f r o m the same chord i n the second example. I n
the t h i r d example the t h i r d may be o m i t t e d f r o m b o t h the six-five
and seven-five chords i n order to provide the p r i n c i p a l part w i t h
sufficient freedom a n d q u i e t to express its slow notes i n accord w i t h
the desired affect.

n A

ss

4. T h e dissonance is i n the bass and may enter as either a tiecT

note (Figure 325, Example a) or a passing tone (b), b u t i t always
resolves by stepwise descent. Henee, the octave of the bass is never
taken by the r i g h t hand, even i n an inner part, a l t l i o u g h i t may be
by the left for re-enforcement. T h e second itself is treated as a consonance; i t may enter freely, r e m a i n sttionary, or leap; a n d i t may
be d o u b l e d .
Figure 325
a.

5. W h e n the chord contains a major second, major sixth, a n d

perfect f o u r t h , the last-named interval may subsequently ascend,
descend, r e m a i n sttionary, or leap d o w n w a r d (Figure 326, Example a). I t enjoys the same freedom w h e n i t is associated w i t h tire
major second a n d m i n o r sixth (b); or the m i n o r second and m i n o r
sixth (c):
Figure 326
a.

5
THE

1.
2.

CHORD

OF

T H E SECOND,

jt

'7

T h i s chord consists of a second, f o u r t h , a n d sixth.

Its signatures are: 2, 4+-, 4!) ( i n those cases where the n a t u r a l

b,

sign raises a tone), f , a n d 4.

3. T h e chord may c o n t a i n the major or m i n o r sixth, the augmented or perfect f o u r t h , the major, m i n o r , or augmented second.
1 Cf. Arnold, op. cit., pp. 648-672.

k5

U
4
3

(6)

4bJ ,4k

b.

le

T II O R O U C> II

- T*1

i-

"T

h=f
4

0;

i-

2 b 6
b

44 6

ffg

'2

Mi id ;J liaj
g
-de fl
1
j J i.
t 4 i- 0

"V r

i j . i

'

15b

44

"^2|> bl7 -6
5b -

H=rJ| i d4

5b

-M

f KE

1
b

r J

i~il-.jp |

'2b

^y.|-

'

2b >7

Ir r

Figure 527
a.

m||_
^ .p.
2
6

IT V -

w
.

9
6

'6

-A

:trX
2

ni
2

1
b

5b

Ui 1 I i]T
ir r e1

It 1

11

" r~
8
6

J| j J J

f
^
f
r
1?

9: W j 1 j

a.

+1

a.

2b >6b 5

44-

BASS

6. W h e n the augmented f o u r t h is associated w i t h the m a j o r

second a n d major s i x t h , i t may r e m a i n sttionary, or ascend (Figure
327, Example a); this also holds w h e n i t appears w i t h the augmented
second and m a j o r sixth (b). A l t h o u g h i t descends m o m e n t a r i l y
i n a passing r e l a t i o n s h i p i n the last example, i t ascends i m m e diately thereafter:

... Z

2b

11 CU

HrfHhl

f
<

1 H F

riIORO

^=4

j.

M
*

v-

n A s s

l.r

256

T II O R O U V, II

li A S S

T II O R O U ('. II

li A SS

257

8. T h e chord of tlic second is easy to ind. Its Iones orm the

t r i a d that lies a second above the bass.
9. Since this c h o r d is f o u n d e d o n a t r i a d , the accompanist
must be careful to avoid fifths w h e n i t is preceded by a t r i a d or b y a
c h o r d w h i c h comprises one (Figure 329).
Figure 329
I wrong I
Iwrongl

"4 b
2

44-3

7. I n order to w i n f o u r i n d i v i d u a l tones i n f o u r - p a r t accompaniment, the augmented f o u r t h must sometimes leap d o w n w a r d . I n

fact, I see n o other means of a t t a i n i n g such an end, should this progression be p r o h i b i t e d . A t the same t i m e , the parts progress m u c h
more smoothly a n d this i n t e r v a l behaves m u c h more characteristically w h e n i t progresses by stepwise ascent w h i l e unisn a n d octave
doublings are alternated. F u r t h e r , by this means a l l dispositions of
the c h o r d may be employed, whereas i n the other case the leap
f r o m the augmented f o u r t h must be assigned to an i n n e r part i f
the progression is to be acceptable (Figure 328).

10. For purposes of re-enforcement, a fifth part may be taken

at times w h i c h doubles the major or m i n o r second. Moreover, this
provides a means of disguising the poor d o w n w a r d leap f r o m the
augmented f o u r t h . Such a d o u b l i n g may n o t be a p p l i e d to the augmented second o r to the major second w h e n i t is associated w i t h
the m i n o r s i x t h . (Figure 330).
Figure 330

'44

not good

poor

wrong

'-6

1 i \yj
1 i1
t
better

44-

11. I n a three-part accompaniment the chord loses an i n t e r v a l .

Henee i t should n o t be used freely unless there is sufficient reason,
i n w h i c h case the s i x t h is o m i t t e d .
12. So that the eye w i l l n o t be overwhelmed by a mass of
numeris, certain intervals are to be taken f o r granted i n the absence of specific indications: T h e major sixth accompanies a specified augmented f o u r t h (Figure 331, Example a); the m i n o r sixth
accompanies a specified m i n o r second (>); the augmented f o u r t h
accompanies the augmented second (c); a n d the major second a n d
major sixth accompany a f o u r t h w h i c h is augmented by means of a
d o u b l e sharp (d):

2 5

TIIOROUCll

liASS

T II O R O II ('. II

li A SS

259

polatcd over the second tone. The chord is created by accompanied

Figure 331
a. 4 4

a.

b.

44

b.

2b

yt'i.J i r S i r B

appoggiaturasr henee its bass does n o t descend, f o r 4 is only a decora-

t i o n of the u n d e r l y i n g t r i a d . T h e second and f o u r t h move to a t h i r d

and the sixth moves to a fifth. I n a l i g h t accompaniment the f o u r t h
(a) or at times the sixth (b) may be o m i t t e d (Figure 333).
2

c.

c.

)i
J
^

Figure 333

II*):,
n y i>

11

p J^
s >

5=1

441-

' a

13. I n Figure 332, however, the raised sixth must be expressly

indicated. I f i t is not, an accompanist inexperienced i n the ways of
chromaticism, far f r o m b e i n g provided w i t h a convenient reduct i o n i n the n u m b e r of figures, w i l l be caused no end of difficulty
and embarrassment. I n the last example the sixth may be replaced
by a d o u b l e d second i n a n t i c i p a t i o n of the approaching t r i a d .
Figure 332

4
2

1 ^ - =

r i"
9

5
3

i'
lol.l

it

fe
4

T^4H
6

5
3

b>7
5b

> *
5

4
2

II

I7
5

(9^pH

5
3

11

JcJ

"8

'6

5
3

N 1 Jlli !
- r r * r r
6

{ =4 ~T|
II 1

ntH

>:

>6.
4

5
3

4
2

5l>

7 ^ - 5
4
2

'

15. Occasionally 2 w i l l be f o u n d over a sttionary or a repeated

bass. I t is to be realized i n three parts w i t h o u t any a d d i t i o n a l
tones. T h e intervals are n o m o r e i n need of r e s o l u t i o n t h a n the
bass, for, as passing notes, they may ascend or descend. T h e preceding and f o l l o w i n g signatures are also realized i n three parts most
of the t i m e . Such an accompaniment usually duplicates other perf o r m i n g parts. Once i n a w h i l e , however, one of these parts sustains
the octave or fifth of the u n d e r l y i n g c h o r d . I f i t is the fifth, a major
4

seventh may be added to 2 (Figure 334, Example a). As a w a r n i n g , a

4

T e l e m a n n b o w may be placed over 2 .

8

14.

Occasionally w h e n the bass of 5 ascends one step to become

4

the bass of a m i n o r t r i a d , the figures,

or 4 w i l l be f o u n d inter-

2 T h e sixth is present in both examples under b. Evidently, "sixth" is a misprint

6
for "second," since this interval is omitted from the illustrations. Bach's 4 signature
verifies this assumption.
s Cf. Pt. I I , Foreword, Note 4.

2O

r i o n o 11 c, II

n A

T 11 O li O II C 11

ss

li A

SS

261

Figure 336

wrong

wrong

frf r
'6

THE

CHORD

O F T H E SECOND,

II

1. T h e second of the successive 2 chords i n the f o l l o w i n g

example arrives one eighth too soon as the result of an elisin w h i c h
is illustrated i n Example a of Figure 335. A perfect f o u r t h may precede the augmented f o u r t h i n the same k i n d of progression (b).

ir c r ^+"^T 4 ir r
2. I t is w r o n g to w r i t e the signature of an augmented f o u r t h
alone i n the f o l l o w i n g examples. T h e c h o r d of the second, w h i c h is
specified by this i n d i c a t i o n , cannot be realized i n the first example
of Figure 336 because of the preceding g-sharp, or i n the second
because of the necessity of resolving the preceding dissonances. T h e
six-four must be taken i n b o t h cases. I n the n o t a t i n g of parts i t is
sometimes forgotten i n haste that the signature M- is an abbreviated
sign of an entire six-four-two chord rather than the six-four.

1-17

9
7

8
6
41,

t r T
^

4t|

r
, l

3. W h e n the fifth and t h i r d of 5b are retarded by a slow twopart appoggiatura and this rather discordant embellishment must
4

5b

be played by the accompanist, 2 and 3 are indicated over the bass,

the sixth of the f o u r - t w o chord being o m i t t e d f r o m the accompanim e n t as w e l l as the signature (Figure 337, Example a). W h e n the
second of the c h o r d of the second is retarded by a slow appoggiatura o n the augmented octave, the accompanist takes the f o u r t h
alone i n a l i g h t accompaniment and does n o t add the second and
sixth u n t i l the second enters i n the p r i n c i p a l part. However, since
the augmented octave is more f e a r f u l to the eye t h a n to the ear
(which is n o t unpleasantly deceived by its resolution i n a slow tempo
accompanied by the f o u r t h ) , i t may be i n c l u d e d i n the signature
and played by the accompanist (b). Also, the execution under c does
not sound bad w i t h the t h i r d , as an appoggiatura, m o v i n g to the
f o u r t h . Those whose ears are oversensitive can w i t h h o l d the r i g h t
hand's accompaniment f r o m the appoggiatura i n b o t h cases. I n any
event, w h e n the accompaniment is to be very l i g h t , the performance
of these refinements should be left to the p r i n c i p a l part. T h e augmented octave is a dissonance that resolves u p w a r d , a n d i t is used
only as an appoggiatura.

262

T II O R O V (', II

Figure 337
a.

6):

li/SS
Figure 339

j |*J:

rrr

T II O R O 11 C II f A S S

Id
r

5b
2 3

4 5b
2 3

b.

i4k

4.

T"'" 6

l ~ l
.

#8
4

f.
1

*3

J)

1 i
i
1
g v , . . ^
r
3

-1-

.2

6
4

4. Figure 339 shows clearly that 2 results f r o m the premature

entrance of the tones of the f o l l o w i n g c h o r d of the s i x t h . W h e n
the second of the c h o r d is d o u b l e d the c h o r d of the s i x t h becomes
S
6
6, a n d w h e n the fifth is d o u b l e d , 3.

^||
6

I n order to avoid octaves, the t h i r d or fifth of the last c h o r d

5. T h e five-two c h o r d always sounds empty regardless of

whether i t is realized i n three or f o u r parts. I t is made sonorous by
its resolution. Rare i n the galant style, i t is more f r e q u e n t i n learned
works and i n company w i t h syncopations. Consequently i t is
realized i n f o u r parts.
6. Inasmuch as one of the tones, b u t n o t the bass, must be
d o u b l e d , care must be taken to avoid octaves w h e n a preceding
c h o r d also contains a d o u b l i n g . D o u b l i n g s must be alternated i n
such a case (Figure 340).
2

must be d o u b l e d i n F i g u r e 338.
Figure 338

-^4

1 hju

1 r 1 i m i J j ij
THE

FIVE-TWO

CHORD.

2
7. T h e five-two c h o r d w i t h an augmented fifth sometimes results f r o m the action of an i r r e g u l a r passing tone or changing note

1. T h i s c h o r d consists of a second and a fifth. For a f o u r t h part

either of the intervals may be d o u b l e d .

(Figure 341).
Figure 341

2. Its signature is 2 . T h e second is m a j o r a n d the fifth perfect.

3. As i n a l l chords of the second, the bass is dissonant; i t must
be prepared, a n d i t resolves by stepwise descent (Figure 339).
1 Cf. Arnold, op. cit., pp. 708-709.

&

tus i

I.e., octave and unisn doublings in the right hand.

1.
2.

n i

o n o

THE

FIVE-FOUR-TWO

IIASS

l i a n

CHORD

T n o n o n a 11

265

n A ss

T h e c h o r d consists of the intervals w h i c h give i t its ame.

T

Its signature is 4. T h e second is major, a n d the f i f t h a n d

2
f o u r t h are perfect.
3. Here too, the bass is tied a n d resolves by stepwise descent,
for i t is the bass that is dissonant. T h e fifth or the f o u r t h must also
lie i n the preceding chord. By means of this chord the upper parts of
the six-five c h o r d w i t h a d i m i n i s h e d fifth are anticipated (Figure

34?)-

4
2

4
2

4. Because i t appears o n l y i n works that requir a f u l l accompaniment, i t is always realized i n f o u r parts; a l l the more so,
because none of the intervals may be o m i t t e d .
5. T h e c h o r d can be located by p l a y i n g the six-five c h o r d o n the
tone that lies one step below the w r i t t e n bass note.

THE

THREE-TWO

CHORD

4. T h e three-two chord is always realized i n f o u r parts, since i t

is expressly indicated i n order that n o i n t e r v a l w i l l be o m i t t e d . T h e
tones of this c h o r d can be located by p l a y i n g a t r i a d a n d replacing
the octave of the bass w i t h a second.
5. T h r o u g h the action of an irregular passing tone this c h o r d
anticipates the upper parts of a four-three chord w i t h a major
second (a) or a m i n o r t h i r d (b) (Figure 344).

1. T h i s c h o r d consists of a m i n o r second, major t h i r d , a n d perfect fifth.

3

2. Its signature is 2 w i t h 2 lowered by an accidental; a n d w h e n

the t h i r d is chromatically raised i t is represented i n the signature
solely by an accidental.
3. Once again, the bass is dissonant; i t is tied, a n d resolves by
stepwise descent (Figure 343).

THE

CHORD

OF T H E SEVENTH,

1 Cf. Arnold, op. cit., pp. 711-712. T h i s chord is not to be confused with the FiveFour Chord, to which a seprate section is devoted.
2 Cf. Arnold, op. cit., pp. 713-714.

1. T h e chord of the seventh exists i n three forms, consisting of:

a seventh, fifth, a n d t h i r d ; a seventh, t h i r d , a n d octave; a seventh
and a d o u b l e d t h i r d .
2. T h e signatures are 7 and 5 . Accidentis must n o t be forgotten, especially w h e n the t h i r d becomes major or m i n o r t h r o u g h
chromatic alteration.
3. C o m p r i s i n g this chord are: D i m i n i s h e d , m i n o r , a n d major
sevenths; augmented, perfect, a n d d i m i n i s h e d fifths; major a n d
m i n o r thirds; the octave.
4. T h e seventh is a dissonance w h i c h appears w i t h preparation
(Figure 345, Example a) and w i t h o u t i t (b). I t progresses by stepwise
descent. T h e ascending major seventh w i l l be given seprate treatm e n t later. I n o u r present study this seventh, l i k e the others, resolves d o w n w a r d . T h e passing seventh alone may r e m a i n sta1 Cf. Arnold, op. cit., pp. 542-599.
2 Or, more accurately, the seventh formed by a passing bass.

T II O R O II (l II

266

li A

SS

THO

tionary on occasion (<:). But w l i c n i t enters altor the bass i t too

desc:ends (o).
Figure 345

b.

6
4

5i

5
3

8,!>7

5. T h e seventh is the same as the second below the bass, and

the chord of the seventh w i t h a fifth contains the t r i a d o n the t h i r d
above the bass.
6. T h e fifth and octave are o m i t t e d i n three-part accompaniment, b u t the t h i r d must always be represented except i n the galant
style.
7. Use of one or another of the three forms of the chord is n o t
always o p t i o n a l . C e r t a i n great difhculties arise i n connection w i t h
their employment, as we shall see later. I t w o u l d be a simple task
b u t of great benefit to practiced as w e l l as inexperienced accompanists i f the fifth and octave were i n c l u d e d i n the signature
whenever they were to be played. T h e i r presence w o u l d offer n o t h i n g new to the eyes, for they are of ten f o u n d i n the signature anyway.
T h e most essential consideration w i t h regard to the construction of
the chord is that the seventh must appear a n d resolve i n the voice
i n w h i c h i t is prepared.

R 0

(' li

liASS

267

chromatically, even the augmented fifth may be taken i n a 7 6 progression, and w i t h o u t being indicated, p a r t i c u l a r l y w h e n i t stems
f r o m a preceding unresolved augmented f o u r t h . T h e d i m i n i s h e d
fifth appears occasionally i n this progression a n d may also be taken
w i t h o u t i n d i c a t i o n o n c o n d i t i o n that i t resolve properly. Examples
of a l l of these variants w i l l illustrate m y meaning.
11.
I n Example a of Figure 346 the octave of the bass may be
taken as w e l l as the d o u b l e d t h i r d . I n the case of the f o r m e r , the
t h i r d i n the r i g h t h a n d moves against the left. T h e disposition w i t h
the f i f t h of the first c h o r d o n top is the poorest; the octave o n t o p
is the best. Should the t h i r d be d o u b l e d i n the octave, b o t h hands
progress i n parallel m o t i o n . I n Example b o n l y the d o u b l e d t h i r d
is possible because of the r u l e against the augmented second.
N e i t h e r Example a or b can be realized w i t h the fifth because the
aseen t of the first tone i n the bass w o u l d crate fifths. I n c, the fifth
being perfect, i t may be played i f necessary, a l t h o u g h the other accompaniments are preferable. I n d the fifth cannot be i n c l u d e d because i t disagrees chromatically w i t h the augmented s i x t h over the
f o l l o w i n g tone. T h e octave must be taken, since, according to Paragraph 8, a d o u b l e d t h i r d is n o t permissible i n this c h o r d . I n e the
Figure 346
*

b.

7 6

* J 1r
7

=
F

76

II < ) !
II /

&

76

c.

p
P

s 1 H H y
- - f

II'
d

f.
7

8. T h e major t h i r d must n o t be d o u b l e d w h e n i t appears w i t h

the m i n o r seventh, regardless of whether i t comes by its q u a l i t y
n a t u r a l l y or chromatically.
9. T h e seventh may resolve o n either its o w n bass or a di fferent
one. B o t h resolutions appear singly a n d also i n succession.
10. A single appearance of 7 6 is better realized w i t h a d o u b l e d
t h i r d , or the octave, than w i t h a fifth. B u t w h e n the latter is perfect
and n o t contrary to a chromatic context, i t too may be taken, prov i d e d that care is exercised to avoid fifths. I n fact, w h e n i t fits

7 6

1 r ' r 1r 1r1

^lUf 1
r r r

M * - l

\r

u
*

"6
5

J
'

268

T II O RO

U ( II

liASS

Til

O l O U(

II

li A

SS

269

fifth is d i m i n i s h e d , but cannot be resolved. T h i s is a p o i n t which

must be carelully observed, for i t mak.es the reali/.ation of the fifth
dangerous. Henee, either of the two r e m a i n i n g forms of the chord
must be employed. Successions of sixths must be placed i n the upper
part i n order to avoid errors or a w k w a r d voice leading. I n f, only
the octave of the bass is good, for the other constructions introduce
a w k w a r d and bad progressions.

thein." Consequently the Other two Corma of the chord are safer
here (b). Example c is n o t e w o r t h y : T h e d o u b l e d t h i r d is not to be
used here according to Paragraph 8, and the octave of the bass does
not go well w i t h the f o l l o w i n g g-sharp. T h e r e f o r e only the f i f t h is
possible. I n d the easiest accompaniment is the octave of the bass,
b u t the f i f t h may also be used. A d o u b l e d t h i r d is r u l e d o u t . A
unisn d o u b l i n g may be used to good ends i n this example.

12. I n Figure 347, Example a, a l l three forms of the chord may

be used. T h e d i m i n i s h e d fifth is allowed i n this progression because
i t can move to a f o u r t h o n the entrance of the raised sixth. I n b
the fifth must be augmented, i f i t is to be taken at a l l , a n d i t moves
to a unisn d o u b l i n g on the sixth. Since this f i f t h is indicated as
i n f r e q u e n t l y as the d i m i n i s h e d fifth, w h o can tell whether a composer wants i t to be used? O r d i n a r i l y , a dissonant interval w h i c h
clashes w i t h an already dissonant chord is n o t realized w i t h o u t
its being indicated. O f course i t is another.matter w h e n the fifths
are expressly called for. B u t u n b i d d e n fifths steal i n t h r o u g h the use
of accompaniments that r e q u i r e the maintenance of f o u r seprate
parts. Henee, a unisn d o u b l i n g removes any necessity of realizing

13. I n Figure 348, Example a, the perfect (first example) and

augmented (second example) fifths are r e q u i r e d because of the preceding augmented f o u r t h . T h e r e s o l u t i o n of the f o u r t h causes b o t h
fifths to ascend. W h e n the f o u r t h of a c h o r d of the second is perfect, i t is customary to indcate the d i m i n i s h e d fifth i n a succeeding
c h o r d of the seventh (b). I f the sixth w h i c h accompanies the f o u r t h
Figure 348
7 6
44
5b 76

Figure 347
4*
J-

a. 6,

7-6.

te

r
6

r
8
'7

I_I
5

= H =

"T
5

Ilijjl j
j i

M
p-3i

r
1

g r 1 r 1 r
d.

d. b7 6
5b-

d.

6
5

7 6

r
6 b

*3
g t

in " Hif r h * - l j J

'_

[J

JJ

l|

ij

b7 6

s Cf. Pt. I I , Introduction, Note 6.

15-

5b

6 U

7 6

270

II

O It O U (i II li A s s

r no

no

in:

27/

HASS

is m i n o r i t is usually followed by a d i m i n i s h e d seventh w i t h a

d i m i n i s h e d fifth (c). I n d the octave must be taken because of the
preceding signature. Example e calis for a three-part accompaniment. T h e chromatic course of the p r i n c i p a l voice does n o t allow
readily for a f u l l realization. T h e resolutions must take place
neither sooner or later t h a n reqired. I n order to prepare the
seventh i n Example f, a fifth voice must be taken over the first tone.

T h e c h o r d of the seventh may be ^- or L, a n d the seventh and

augmented f i f t h resolve i n t o a unisn.

14. Successive bass tones i n ascent or descent, each expressing
a 7 6 progression, are frequent oceurrences. I n descent, a three-part
accompaniment is easiest, a n d i t is preferable i n passages w h i c h
d o n o t need f u l l chords. A four-part setting is preserved f r o m errors by a voice leading w h i c h employs alternately a l l forms of the
chords of the sixth and seventh w i t h a n d w i t h o u t d o u b l i n g . " Everyt h i n g that has been previously discussed a n d need n o t be repeated
here must be g i v e n constant consideration. I n so many words,
preparation, resolution, a n d d o u b l i n g must agre w i t h the rules.
15. Figure 349 may be accompanied i n several ways. T h e best
are those i n w h i c h the d o u b l i n g is constantly varied (a). W h e r e i t is
too u n i f o r m , that is, where the bass is d u p l i c a t e d , a n d the t h i r d or
sixth are constantly d o u b l e d , errors can be c o m m i t t e d easily and
the octaves or fifths may become too p r o m i n e n t . T h u s , Example b
is poor because there are too many fifths. Moreover, n o t a l l are
perfect; the d i m i n i s h e d fifth is n o t resolved. T h e octaves i n the u p per voice of c make the accompaniment ugly; a n d the thirds are
dangerous, for they can easily lead to a v i o l a t i o n of the r u l e of
Paragraph 8. T h e realization i n d is poor because of b o t h the thirds
and the fifths. Example e is worthless o n several counts: T h e d i m i n ished fifth is unresolved; worse, the fifths lie i n the upper p a r t ;
eventually the realization must lead to an incorrect d o u b l i n g . I n
short, i t is w h o l l y bad. Example / is acceptable so l o n g as the m i n o r
t h i r d does n o t appear w i t h a major sixth. Example g is poor o n account of the unresolved d i m i n i s h e d fifth a n d the octaves i n the u p per p a r t :

16. Ascending basses w h i c h express

cannot be easily accompanied i n other
h a n d moves against the left. T h e octave
the seventh and the sixth. Example a is
than b (Figure 350).
Figure 350

Jfl
J - l H|-f
r r r
"i
f

76

76

76

76

T h e examples from b to the end appeared originally in the text in the form of
signatures. I have realized them in order to make them more immediately comprehensible.
5

* T h a t is, a doubling within the right hand in alternation with the octave of the
bass.

successive 7 6 progressions
than four parts. T h e r i g h t
and t h i r d accompany b o t h
the best, a n d more n a t u r a l

272

O li O U V,

II

II

li A SS

V / / o / o

17. Isolated sevenths w h i c h resolve as the bass progresses are

realized i n most cases w i t h a f u l l accompaniment w h i c h includes
the fifth. Special cases of this type of progression w i l l be treated i n
the second section. W h e n the f i f t h is d i m i n i s h e d , i t too must resolve (Figure 351).
Figure 351

i 1 11 1 111 1 J 1 J 1

f ir ' T i

4L_ 1 ^ " r77-

*
b.
\ ~ \
Ii | y

*-

'

9
7
4

1r

=
7

Y
7

20. A passing seventh w h i c h enters after its bass is best w h e n

i t comes f r o m the octave (Figure 354, Example a). However, since
the o r i g i n a t i n g c h o r d is consonant, i t is n o t incorrect to leap to the
seventh i n order to change the disposition of tones (b). B u t this
freedom is w i t h h e l d w h e n 8 7 is f o l l o w e d by other signatures over
the same bass (c). A l t h o u g h this is n o t its proper place, o u r discussion suggests a case where the octave, f o l l o w e d by a seventh, is accompanied by other dissonances and must itself be prepared a n d resolved. O f course the octave here moves to the seventh (d):

b:

a.

1-

||

4
3

alternated. T h i s makes the safest a n d best accompaniment. D o u b l e d

thirds can easily lead to a v i o l a t i o n of the d o u b l i n g r u l e . Nevertheless, I have constructed an example (b) i n w h i c h this d o u b l i n g is
acceptable (Figure 352).
Figure 352
a.

' H0

a. %

is taken w i t h the seventh. I n a four-part realization 5 and 7 are

-t

1 : ; j
C

a.

]_ I

18. Successive bass tones w h i c h express sevenths that resolve o n

the f o l l o w i n g c h o r d progress by ascending fourths or descending
fifths. W h e n a three-part accompaniment is r e q u i r e d , only the t h i r d

27j

jf

11 <-; / / / { / i ss

I!). T h e lifth is no! easily included w i t h the passing seventh i n

Figure 353. Moreover, a d o u b l e d t h i r d or an octave sounds better
(<x). W h e n the f o u r t h is prepared and the t h i r d can resolve by stepwise descent, b o t h intervals may be realized w i t h the seventh.
Example b does n o t really belong here b u t rather w i t h the discussion of r a p i d passing notes over w h i c h the r i g h t h a n d holds the
i u i t i a l c h o r d . Henee, i n practice i t is n o t figured.

rf "
7

|7

ft
i 7

1-ti*

- h

-9j-

-4

m
b

I I

u r

'

Cf. C h . I V , U | 68-72, C h . V I , "Passing Tones" and "Changing Notes.'

T II O i O U C, II

O
I
-o6 7
8
4 - 5
3 2
3

SS

r iio

l>7

8 7

m <-m
1

li A

8 7 b7
4 - 3
3
-

THE

CHORD

OF T H E SEVENTH,

II

1. A t cadenees, and also elsewhere, i f the bass ascends a step

or a f o u r t h or descends a fifth, a chord of the seventh may be taken
over the first note w i t h o u t i n d i c a t i o n , p r o v i d e d that the f o l l o w i n g
c h o r d is a t r i a d . T h e fifth is also taken w i t h the seventh here (Fig355 Example a), and i f the r i g h t hand threatens to go too l o w
the octave should be h e l d w h i l e the fifth moves to the seventh (b).
By this means a good upper part can be retained and the final
c h o r d w i l l be complete. W h e n the bass ascends a step, the t h i r d of
the second chord must be d o u b l e d occasionally (c).
u

Figure 355
a.

a.

r" rr y
2. T h e examples of Figure 356 r e q u i r e the octave of the bass
w i t h the seventh. I n a, fifths are created by the fifth w h e n i t is i n cluded w i t h a seventh that lies i n the upper part. Henee the octave
replaces i t i n this disposition of the c h o r d . I n the other dispositions the fifth may be used. T h e fifths of Example b caused by the

11 o v c, 11 HASS

275

m o t i o n of the p r i n c i p a l part, a to g, against a m i d d l e voice, d to c,

are avoided by h o l d i n g the octave over f r o m the six-four chord
and leading the sixth to the seventh, thus o m i t t i n g the fifth. Perhaps this observation seems far-fetched; yet, i n a slow tempo, delicately performed, such fifths can be heard. Moreover, w h e n the
p r i n c i p a l part is w r i t t e n i n over the bass, the accompanist is obliged
to avoid them. I n c. the d i m i n i s h e d f i f t h must be prepared by the
octave of the c h o r d of the seventh. T h e r e is o n l y one disposition for
this progression. I n d, b o t h the s i x t h and the d i m i n i s h e d fifth are
prepared w h e n the octave is taken over g. T h e major t h i r d above
this tone moves nicely u p w a r d i n thirds w i t h the bass. I n e f o r b i d den fifths are e l i m i n a t e d by t a k i n g the octave of the bass over e.
I n f, where the retardation of a resolution (as explained i n ff) creates
successive sevenths, the first seventh must be accompanied by the
octave; i f i t is n o t there w i l l be fifths. I n g, the preparation of the
second seventh requires the octave w i t h the first one. I n h, there
w o u l d be no place for the t h i r d over c i f the f i f t h were taken i n the
preceding chord of the seventh. I n i the chromatic major t h i r d
over a can move u p w a r d naturally, and the sixth of the succeeding
chord of the second w i l l be prepared, i f the octave is taken over a.
I n /, i f the fifth is played i t w i l l crate octaves. I n k, the tones of a
chord are interchanged, as illustrated i n kk. T h e second seventh
seems to be a resolution of the first, b u t i n fact i t is n o t h i n g more
than a decorative detail of the upper voice, rendered negligible by
the interchange of parts. T h e octave is taken w i t h the first seventh
i n order n o t to obstruct this figuration and also to prepare the
second of the c h o r d o n c. I n / the d i m i n i s h e d seventh must resolve
to the octave of the f o l l o w i n g tone. I n m the octave is taken w i t h
the seventh i n order to avoid fifths, and to prepare the tones of the
n i n t h c h o r d over e. I n n the octave again eliminates a f a u l t y progression and places i n the hand the tones of the four-three c h o r d

7
over b. I n o, i f 5 is expressly called for, the octave must be taken as a
f i f t h voice over / i n order to prepare the f o l l o w i n g d i m i n i s h e d
seventh. I n p, the octave is r e q u i r e d w i t h the seventh i n order to
prepare the f o l l o w i n g fifth. I n q the octave is better t h a n the fifth,
for i t eliminates octaves against the d i v i d e d beats i n the bass by
creating contrary m o t i o n . I n r the octave eliminates an u n m e l o d i c
progression w h i c h w o u l d be i n t r o d u c e d by the fifth.

2J6

11 C

T II O li O

V // O li O U C, II

li A SS

II

Figure 356

n.

II

j3

7
It

fc>: p

*-

7
tt

-f1

H H

7
5

6
5

' 'm

7
5

7 6 7

tt

6
5

9%

6 5
4 tt

5b

tt

f. 6

- t t

- -

=tJ=
7

7
=

- Y -

it

<

II

1J
| J

Z3

rr
-f,

II

3. T h e examples of Figure 357 r e q u i r e a fifth w i t h the seventh.

I n Example a i t must be taken because of the f o l l o w i n g six-five
Figure 357

'jt

-64
3 o.

277

good

wrong

fcl!

6
5

li A SS

: = h = _ p

__d

JL_

65

<*

>: -

-p1

-&'

7
6 5
4 3
2 3

d.

!>7

r-

not good

*
6
4

5
3
3

9 8
7 6
6

; s
/tt

ti

4 3

9 8

9 8
4, 3

27<V

T I I O 11 O U C, I I

li A S S

chord. I n b the d i m i n i s h e d fifth accompanies both the m i n o r and

d i m i n i s h e d sevenths w i t h o u t i n d i c a t i o n . T h e disposition w i t h
the fifth of the first chord o n top is worthless. I n c octaves are created
if the octave is taken over the first c, since the octave must be taken
over the f o l l o w i n g c i n order to prepare the ensuing seventh. I t is
this last factor that causes the error unless the g-sharp leaps to c (ce)
w h i l e the resolution of the first seventh is transferred to the bass. I t is
better to take the fifth and lead i t u p to c. T h e disposition w i t h the
seventh over e i n the upper voice is n o t to be used. I n d, the fifth is
r e q u i r e d i n order to prepare the f o u r t h and thus b r i n g to complet i o n the tones that appear over the f o l l o w i n g tone. I n e, the seventh
over c must be prepared by the fifth of the preceding chord. Over
the last chord the t h i r d is played as a fifth part. I t may enter as the
octave of the preceding e. I n /, preparation of the n i n t h over the
final tone is p r o v i d e d by the fifth of the preceding c h o r d of the
seventh. I n the first i l l u s t r a t i o n under / an octave may be taken
over e as a fifth part i n order subsequently to w i n a complete t r i a d
over a. Finally, the second and f o u r t h examples of Figure 355 show
us that the fifth is better than the octave w h e n the bass of the chord
of the seventh ascends one step to a t r i a d .
4. T h e examples of Figure 358 are n o t e w o r t h y for t h e i r signatures as w e l l as their realizations. I n a, a fifth part enters i n the
second bar. As soon as i t has p e r f o r m e d its office i t may be dropped,
a t t e n t i o n being directed solely to a n o r m a l preparation and resolut i o n . I n b, care must be taken to avoid octaves and fifths against the
passing notes i n the bass. T h e a d d i t i o n a l illustrations show how
easily they can be e l i m i n a t e d by an alternation of octave and unisn
doublings. I n c, the seventh over the changing note / ascends, for
this first seven-five chord is only an a n t i c i p a t i o n of the six-four
c h o r d that belongs to the f o l l o w i n g e. I n d the fifth of the seventh
c h o r d cannot be taken i n an u n d i v i d e d accompaniment w i t h o u t
causing an error. Since the major t h i r d may n o t be d o u b l e d , the
octave of the bass must be taken. Otherwise o n l y a three-part realizat i o n or a d i v i d e d accompaniment can be employed. N o r m a l l y , the
n i n t h should resolve over c, thus p r e p a r i n g the f o l l o w i n g seventh.
T h e nature of the retardation is made apparent i n Example dd.
T h e realization of Example d i n an u n d i v i d e d accompaniment is
best when the n i n t h lies i n the upper part, and m u c h easier w h e n a
seventh accompanies the n i n t h . I n e a l l forms of the c h o r d of the

T II O li O II ('. II

HA

SS

270

seventh may be used, provided that the fifth over e does not progress
to the fifth over / as illustrated i n the first realization of this example. T h e correct accompaniment of Example / is shown i n the
first of each pair of illustrations. I n short, the octave is taken w i t h
the first seventh, and the fifth w i t h the second i n order to prepare
the last one. Should this preparation be overlooked u n t i l the c h o r d
has changed, the r i g h t h a n d may play t w o chords over the second
seventh i f its length, as here, allows, and take the proper accompaniment o n the second c h o r d . I n using this acceptable expedient care
must be taken n o t to d i s r u p t any preparation. I n g the first seventh
takes a d o u b l e d t h i r d and the second a t h i r d and a f o u r t h . These
latter intervals are played so that the four-three c h o r d w i l l be complete o n the r e s o l u t i o n of the seventh to the major sixth. T h e fifth
cannot be taken w i t h the first seventh because of the f o l l o w i n g
sixth, c-sharp. or can the octave be taken, since i t w o u l d cause
fifths i n the other parts (gg). T h e progression is illustrated i n x
w i t h o u t the retarded resolution. I n h the t h i r d of the seventh must
be d o u b l e d , for the chromatically raised bass may not be d u p l i c a t e d ;
or can the fifth be taken. Example i contains t w o extraordinary
examples w h i c h I have come u p o n . I n t r u t h , they should be figured
i n the manner of ii. As they stand i n i, errors cannot be avoided
w i t h o u t resorting to a fifth part or the illustrated d i v i d e d accompaniment. I n / care must be taken to avoid unmelodic and i n correct progressions. T h e illustrated dispositions are good; i n the
r e m a i n i n g one, fifths are struck over the second and t h i r d bass notes.
I n k, too, only t w o dispositions may be used; the t h i r d w i t h the
octave on top over c leads to errors. I n l, an alternated d o u b l i n g is
r e q u i r e d . I n m, the chromatic m i n o r t h i r d is taken w i t h o u t indicat i o n i n a four-part realization. A n unaltered d i m i n i s h e d t h i r d above
an altered bass must be indicated. T h i s i n t e r v a l is n o t unsuited
to chromatic contexts.
8
Figure 358

t:

c
o

c
o

->
t- ia
c - ta

t-

mi

XX:

ta

km

co

to t a -

ta
t-

oo
t-

00

ftfc

;:

2(92

///

o no

u o

T II O

ss

l{

O II (', II

A SS

li

2Vj

T
THE

SEVEN-SIX

CHORD

1. T h i s c h o r d exists i n two forms, consisting of a seventh, sixth,

and t h i r d , or a seventh, sixth, and f o u r t h .
2

5. I n delicate accompaniments the t h i r d is o m i t t e d f r o m m i n o r

and d i m i n i s h e d sevenths, especially w h e n i t must be chromatically
raised (Figure 359, Example a). I n such a case, some composers prefer to double the d i m i n i s h e d f i f t h i n the belief that i t is better than
the chromatic m i n o r t h i r d . T h e r e m a i n i n g signatures i n these examples are also realized i n three parts.
Figure 359

1,7

4 3

5b

j i i

6 5
3

k
"

3a

OL

1
^
5

$
6

7
5

6
44-

YL

7
6

"1?
4

1
V ~ ^ ~
5
H

y *
g

6 5
4 3

6. W h e n a cadential bass note, expressing 5 b, is raised and

ascends a half step, the t h i r d of the last c h o r d is d o u b l e d and the
octave o m i t t e d i n the interests of a more melodic part w r i t i n g (Figure 360).

2. Its signature i n the first case is 6 5. T h e t h i r d is indicated

only w h e n i t is chromatically altered. Chromatic alterations of the
r e m a i n i n g intervals must n o t be o m i t t e d f r o m the signature.
3. W h e n the f o u r t h is present instead of the t h i r d i t must be i n dicated. Usually i t is f o l l o w e d by a three under the succeeding
five ( 6 5 ) .
43

4. C o m p r i s i n g the chord are the m i n o r seventh, the major or

m i n o r sixth, and the major t h i r d , w h i c h may be replaced by the
perfect f o u r t h .
5. T h e seventh may enter freely, and i t remains stationary; the
sixth is restricted i n the manner of a dissonance by the seventh and
is therefore present i n the preceding chord. I t resolves d o w n w a r d to
the fifth. W h e n the f o u r t h is present i t too must lie i n the preceding
chord; i t resolves to a t h i r d w i t h the s i x t h . T h e bass may be held or
enter freely (Figure 361).
6. I t can be seen i n Figure 361 that the u n d e r l y i n g relationship
is a chord of the seventh whose fifth and t h i r d are retarded by a sixth
and f o u r t h . T h e best disposition is that w i t h the sixth o n top and
the seventh i n the lower m i d d l e part. However, the accompanist is
notalways free to take this d i s t r i b u t i o n , for its realization is dependent o n the requisite preparation.
1 Cf. Arnold, op. cit., pp. 701-703.
2 T h i s chord will be recognized as that which has come to be known as the dominant thirteenth. Thorough-bass writers knew that the sixth was merely a replacement
of the fifth. Later the sixth simply graced the fifth, both intervals being present simultaneously. Harmonic theorists explained the chord by accumulating enough thirds
to reach to the thirteenth. And having erected this exegetic skyscraper, they promptly
forbade their students to use it unless the fifth, ninth, and eleventh stories were first
demolished. T h e chord is still known by the ame of this perilously unsound
structure.

T II O li O
Figure 361

11 C

7 ( 5
4 3

7
6 5
i

7 5 6 6 5
3 4 4 3

1i

7
6 5

Til

li A SS

II

6 7
34 6 5
i
1o

2(V

Figure 362
1

7
5 6 6 5
jt 4

7
6
2

6
4

(5
3
6 _
5 5

5 6 7 #465
4

li A SS

other q u i t f l f a m i l i a r intervals (b), which have never been criticized.

I ara l i t t l e i n favor of strnngc intervals; yet, study of various
w r i t i n g s on accompaniment has convinced me that ugliness often

TV
6 5
4 jt

O l{ O II C, II

6
5

I'J'I'I',

6
5

5
3

T6
5

4
2

3
8 t>7

5b

g 0

i J|iJJ|.^iiJ ffl 1
*

h
9 8 ,"7
7 6 5b

6 5
6 5
4 $

Kv

K
s

-4^^^

6 b - 5 b 4 #
.

.
h

r K *

s T h e diminished octave happens to have been one of Bach's favored intervals.

It recurs throughout his works, as pointed out by Marpurg. Cf. C h . V, " T h e Chord
of the Second," I I , f 3, where Bach also expresses approval of the augmented octave.
His defense of these intervals may have been prompted by Heinichen's paraphrase of
a hoary couplet:
"Octava deficiens & superflua
Sunt do Diaboli in Msica."

Der Ceneral-Bass, p. 101.

i>7

rrwrr

'

7. Three-part accompaniment of the chord is rare. I t is n o t

easily realized i n the galant style, b u t w h e n i t does appear, f o u r
parts should be employed unless a l i g h t performance necessitates
omission of the t h i r d .
8. Instead of a d d i n g a second section, we shall conclude o u r
discussion here w i t h f o u r noteworthy examples (Figure 362). I n the
first, an accessory second replaces the f o u r t h i n the seven-six chord
and moves to a t h i r d . T h i s progression, l i k e others that occur over
organ points, is best understood w h e n separated f r o m the bass. T h e
essential relationship is i l l u s t r a t e d i n a. I n b the d i m i n i s h e d
sevenths appear i n company w i t h a m i n o r sixth and m i n o r t h i r d .
T h e u n d e r l y i n g relationship is i l l u s t r a t e d i n c, where we find that
the m a n i p u l a t i o n consists of a fifth retarded by a sixth. T h e progression sounds ugly i n a l l dispositions. Even w h e n the sixth is placed
o n top i t is n o t m u c h i m p r o v e d . Henee I prefer the progression i n
Example d. I t is interesting that the d i m i n i s h e d octave, once
r o u n d l y denounced, should be so clearly better here (d) t h a n these

5
4
/

-^-=

-7

#f

4^
7

ffe

r
[7

286

T II O li O

11 C

II

TU

li A SS

results p r i m a r i l y f r o m iiniisual combinaliuns o usual sounds. I n

the t h i r d example (e) the d i m i n i s h e d seventh appears i n company
w i t h the d i m i n i s h e d sixth and m i n o r t h i r d . T h e u n d e r l y i n g progression is i l l u s t r a t e d i n f, and shows us that the ifth is retarded by
a s i x t h . W h e n this i n t e r v a l is o n top, the progression (e) is not too
bad. B u t other distributions r e q u i r e ears w h i c h are as extraordinary
as the example. A three-part accompaniment (g) is more acceptable.
I n the t h i r d bar of the final example (h) the resolution of the
f o u r t h is retarded u n t i l the fifth has descended a chromatic half
step. T h i s progression is good w h e n the bass is h e l d before and
after, w h e n the tempo is rather broad, and the major t h i r d , g-sharp,
does n o t appear immediately p r i o r to the entrance of the retarded
f o u r t h , as i t does i n the last i l l u s t r a t i o n of Figure 361. T h e last example (Figure 362, Example h) makes use of the organ p o i n t , w h i c h
w i l l be treated separately later. B u t i n order to restore the confidence of those who are overwhelmed by the mass of numeris, we
state that o r d i n a r i l y the r i g h t hand does n o t play such passages. T h e
signatures are therefore o m i t t e d , and tasto solo is w r i t t e n over the
bass. I n our present study they serve to indicate the voice leading
and chord changes.

THE

SEVEN-FOUR CHORD

1. T h i s c h o r d really belongs i n the discussion of appoggiaturas,

still ahead of us. However, since i t can be f o u n d i n pieces i n w h i c h
appoggiaturas are n o t represented i n other signatures, we shall give
i t special consideration here.
2. I t appears over bass tones that are n o r m a l l y accompanied
by the chord of the seventh or the six-four c h o r d .
3. W h e n this chord replaces the c h o r d of the seventh i t takes
either of t w o forms, consisting of a seventh, fifth, and f o u r t h , or a
2

seventh, octave, and f o u r t h . I n b o t h cases its signature is 4. Here,

as elsewhere, i t w o u l d be b u t a small task to i n c l u d e the t h i r d figure.
T h i s w o u l d make i t easier for the beginner and eliminate its confusin w i t h the chord of the major seventh, w h i c h , as we shall see, is
similarly indicated at times.
1 Cf. Arnold, op. cit., pp. 703-708.
2 T h e seven-four chord as a manipulated chord of the seventh is discussed in
^
_
the final f ^ 14-15 being directed to the seven-four chord as a manipulated
six-four chord.
3

1 3 >

O l{ O II (', II

li A SS

287

4. Compt ising this chord are major, m i n o r , and d i m i n i s h e d

sevenths; augmentcd, p e r l c t t , and d i m i n i s h e d fifths; d i m i n i s h e d ,
perfect, and augmented fourths.
5. As we have already learned, the u n d e r l y i n g relationship is
the chord of the seventh. T h e only difference is that the t h i r d is
retarded by the f o u r t h . B o t h the seventh a n d the f o u r t h , or at least
one of these intervals, should be prepared. B o t h resolve by stepwise
descent; even the augmented f o u r t h , for i t appears only as an inessential appoggiatura rather t h a n a d e f i n i t i v e chord tone. T h e
seventh and f o u r t h seldom resolve simultaneously; usually one
follows.the other. T h e bass behaves as i n the chord of the seventh.
6. Inclusin of the fifth or the octave hangs on the same considerations that govern their appearance i n the u n d e r l y i n g chord
of the seventh.
7. T h e seven-four chord is o n l y rarely used w h e n the seventh
resolves to a s i x t h over a stationary bass; and even more rarely,
here, do the seventh and f o u r t h resolve simultaneously. For one
t h i n g , i t sounds bad, and f o r another i t can be dangerous, dependent
o n the d i s t r i b u t i o n of tones, to have appoggiaturas i n t w o voices
a f o u r t h apart. I use the expression appoggiaturas advisedly, f o r dissonances, p a r t i c u l a r l y those that resolve over a stationary bass, are
i n essence n o t h i n g more than decorations.
I n Figure 363 the seventh and f o u r t h resolve simultaneously.
7
Example bb is preferable to b; 4 may always be taken w h e n the
8.

seventh moves to a major sixth, and the f o u r t h , i f i t progressed,

w o u l d move to a m i n o r t h i r d (bb). I n c and ce the augmented fifth
must be expressly indicated. Example d is best i n three-part accompaniment, and e is not p a r t i c u l a r l y good inasmuch as the best
disposition, w i t h the augmented f o u r t h a n d m a j o r t h i r d widely
separated, cannot be played because of fifths.

288

T II O R O

"4i J

< ) : ^ s f ( ~ -

'

6
4

"JJ

U ('. II
c.

li

Mil
6

ji;

r 1

f-JLfT

3
|74 6

144

^6.

5
wrong
i
'7

T5 *
5

1)

li

289

ss

aa.

I- . J J

i4

7
4

11 a 11

o no

Figure 365

9. I n Figure 364, where the f o u r t h resolves before the seventh

(a) and the seventh before the f o u r t h (b), three-part accompaniments should be used. A f o u r t h part is rather forced i n a and i m possible w i t h o u t errors i n b.
Figure 364
a.

1i

an incomplelc one (aa). I ! the lilil is taken, the f o u r t h is best placed

on top. Occasionally, owing to a required preparation of a succeedi n g interval, the octave must be taken (b). Here the fifth may be
included as an a d d i t i o n a l part i f i t fits. I n Example c the t h i r d (1) or
the f i f t h (2) should be d o u b l e d . I f this is not d o n e , ' i t is better to
seprate the f o u r t h f r o m its preparation (3) than keep b o t h i n the
same part (4). T h e reason centers a r o u n d the progression of the
bass f r o m a to g-sharp w h i l e the m i d d l e part moves f r o m a to /.
W h e n the latter m o t i o n is filled i n , i t creates a cross r e l a t i o n . A l t h o u g h this consideration is no longer as i m p o r t a n t as i t used to be,
the progression can be avoided very easily. Certainly n o one w i l l
deny that the execution of the unelaborated f o r m of this progres-

-9
7
-65b, 4
3

^ j y

'
J

A SS

J-F
L

h-Jh 4
>1

m1

m
7
4

5b

7
5b
4

3
5.

7
4

b7
5b
4
10. I n Figure 365 the f o u r t h resolves directly to a t h i r d w h i l e
the seventh awaits the f o l l o w i n g bass note. W h e n the accompanist
is free to take either the fifth or the octave w i t h an unprepared
seventh, i t is better to take the fifth because i t creates a complete
c h o r d on the resolution of the f o u r t h (a), whereas the octave makes

5b

5b

T II O ROU G II

po

ASS

li

sion w i t h contrary m o t i o n (5) is better than its cxecution w i t h similar m o t i o n (6). Except i n the galant style, only 1 and 2 should be
used.
11. W h e n the seventh is prepared, choice of a f o u r t h part is
m u c h more l i m i t e d . I n Figure 366 there are b u t few examples
where the fifth or the octave is o p t i o n a l . T h e progression i n the
asterisked example sounds best i n the notated d i s t r i b u t i o n of parts.

TU o no a c u HASS

Figure 367

Figure 36C

'JUMO,
\
=9=k

4 3

ong

11

i r "
4

6
3

i ]
r
4 = 1
7
4

11

-Y

13. I n the example of a passing seventh i n Figure 368, an inner

part moves i n thirds w i t h the bass, and the seventh and f o u r t h re7

'4

1.
4 3

i - i

u
- 4 1 ?
r f
f
7 1?" 7 F
43
3
43
4 77
7 7
4343
4
i 1343
S: J "
1 r 14 = 1 4
P
r

1 p* 1i 4 =
7 u
6
7

4 . 3

01

ji3
r~ j 1

t-3-

\HIf-

ffti

11 ?

21)1

11

m a m stationary. T h e signature, 4 , used by some, is n o t clear

r
7
4

11

enough; 4 is better. A d o u b l e d t h i r d , or a t h i r d and an octave, is

3

better than the i l l u s t r a t i o n .

12. W h e n successive sevenths appear over a bass that leaps by

fourths and fifths, altrnate sevenths are often accompanied by
4 3. T h e progression may be realized i n f o u r parts i f necessary (a).
B u t , should 4 3 appear w i t h a l l of the sevenths, the accompanist may
w i t h clear conscience l i m i t the setting to three parts. T h i s succession occurs i n the galant style only. Its realization is shown i n b
and its bass, figured, i n bb. I n fact, the seven-four chord is usually
given a three-part setting, for the c h o r d does not readily appear i n
the learned style (Figure 367).

Figure 368

14. W h e n o u r chord appears i n place of the six-four chord, i t is

because the s i x t h is retarded by the seventh. T h i s latter i n t e r v a l
w i t h the f o u r t h and octave f o r m the conten of the chord. T h e

292

Til

li o a

; //

li A SS

seventh is usually m i n o r , b u t the f o u r t h is always p i r l e c t . Both i n tervals resolve d o w n w a r d , the seventh before the ourth. The signa. . .
.
. 76
ture of this progression is 4 .
15. Figure 369 provides more detailed i n f o r m a t i o n about this
progression. I n a the bass remains stationary, a n d neither the
seventh or the f o u r t h is prepared. I n b the seventh is prepared b u t
7
8
not the f o u r t h . Over the first note, 5 or 7 may be taken. I n c, b o t h
6
intervals are prepared. 3 should be taken over the first note. A
6
d o u b l e d t h i r d w i t h the s i x t h (ce) creates h i d d e n octaves, and 8 direct
3

O II O

Til

(I

C II

29 J

li A SS

i
iJ J " j J
l " ' rE = M = s j

"

9-

u
7

m 1

1|23

'8

7 6 .

d.

II f
It1

16

7
6 /
4 - 6

4 -

<r<

G
=

6L

'7

wrons
| ce. j
J

wrongj |

-6 f

c.

octaves. I n d the seventh is prepared. T h e t r i a d over the first note

8

may be realized as 5 , 3 , or 5 . T h e t r i a d w i t h the octave is excellent b u t n o t w i t h the f i f t h o n top, f o r this d i s t r i b u t i o n causes errors.
I n e b o t h dissonances, seventh a n d f o u r t h , are prepared. Over the
7
3
,
first tone 5 or 7 may be taken. Because the octave is r e q u i r e d over
3

o. 6
* 5

7 6
4 -

ggp

AL

h.

7 6

4 -

1J If-t-g

the second tone, i t cannot be taken over the first. I n / the seventh

w i l l be prepared i f the first c h o r d is realized as 6 . W h e n the t h i r d

3

of the chord of the sixth lies o n t o p , fifths are unavoidable. I n g,

b o t h dissonances are prepared. However, i n order to prepare the
seventh, a n octave must be taken as a fifth p a r t over the first bass
note. T h e f o u r t h a n d seventh are also prepared i n h. T h e fifth is
taken w i t h 7 6 over the second bass note. I n a d i v i d e d accompanim e n t the d o u b l e d t h i r d may be played, as illustrated i n the final
example. T h e octave cannot be taken wiyh this 7 6 progression
w i t h o u t causing errors.
Figure 369
a,

tr-ri

r
1

76
4 -

5
3

7 6
4 -

-f-

-9-

-9-

7 6 5
4 - 3

1f

J^V-
i

11 1 |

THE

1.

1
r-

T-J

1
w

CHORD OF

1
1

THE

MAJOR

L = ? F = *

r r

SEVENTH,

N o r m a l l y , this c h o r d consists of a major seventh,

perfect

f o u r t h , a n d major second.
7

2.

Its most usual signature i n f o u r - p a r t accompaniment is 4

1 Cf. Arnold, op. cit., pp. 674-682.

T H O R O U G II

294

li A S S

T II O R O V. II

w i t h accidentis as r c q u i r c d . Confusin is caused by those who expect a four-part realization b u t o m i t 2 f r o m the signature or specify
only 7.
3. T h i s c h o r d appears as a passing relationship over a stationary
bass, and also as the retardation of a t r i a d f o l l o w i n g a m o v i n g bass.
I n the first instance, a l l three intervals are taken freely and ascend
(Figure 370, Example a). I n the second, the seventh and second
must lie i n the preceding c h o r d ; the f o u r t h may (b), b u t need n o t
(c), be prepared. T h e second and seventh ascend; the f o u r t h descends. W h e n the f o u r t h i n Example a lies o n top, i t too descends.
Figure 370
a.

Ai1

l
1< Us
-

8
3

fl

7
4
2

8
3

%w

r
7
5

-**
M

T .
2

IT4
2

8
3

T h e signature 7 is often used instead of our 4 . I n the course

4
2
of o u r study we shall see that certain forms of the c h o r d may be i n dicated by either one. T h e d i s t i n c t i o n to be observed here is that
the major seventh w h e n accompanied by a n i n t h always resolves by
stepwise descent, b u t i n o u r c h o r d the seventh and the second always resolve by stepwise ascent. T h e second, appearing as i t does
over a stationary bass, and being used consequently as a passing
tone or a retardation, enjoys the same r i g h t here as i t does i n similar
cases; namely, the r i g h t to ascend.
7
5. T o lcate 4 , play the t r i a d o n the seventh above the bass.
2
6. I n three-part accompaniment the second or the f o u r t h is
o m i t t e d . W h e n i t is necessary to specify one of these settings the
. 78
78
signature is 2 3 or 4 3. Care must be taken to observe the indicated
resolution of the seventh i n the latter signature i n order to avoid a
realization of the seven-four c h o r d instead of the i n t e n d e d threepart seven-four-two c h o r d .
7. O u r chord is occasionally realized i n five parts. T h e addi4.

li A SS

295

tional i n t c r v a l is either a sixth, major or m i n o r , or a perfect fifth.

T h e bass may r e m a i n stationary or move.
8. Either sixth may, b u t need not, be prepared. B o t h progress
d o w n w a r d to the fifth, thereby m a k i n g a complete t r i a d o n the
resolution of the c h o r d . T h e second is sometimes o m i t t e d i f f o u r
parts are preferred. T h i s is most frequent w h e n a bass note w i t h the
4
signature 6 or 3 descends stepwise to the tone over w h i c h our c h o r d
appears. Should the s i x t h over the first bass note be augmented, the
i n t e r v a l of the second cannot be taken w i t h the next chord, for i t
lacks preparation.
9. T h e examples i n Figure 371 w i l l make my meaning clearer.
A n exact i n d i c a t i o n of the intervals is especially needed here. I n a
the second may, b u t need not, be played, depending on its being
specified. I n the t h i r d and f o u r t h illustrations under a the sixth
moves to a fifth w h i l e the seventh and f o u r t h r e m a i n stationary.
T h e first and t h i r d examples, i l l u s t r a t i n g the major sixth, sound
good o n l y i n the notated d i s t r i b u t i o n of parts. I n b and c five-part
Figure 371

a.

T II O II O U G II

II A S S

T II O li O II C, II

1
f

16- I*
5 6 65
4

i m \^ i Y ir 1
6 5
4
#

6 5
* I

8
5

"r>
5

7 8
6 5
4 I

|7
6
3 55

8
5
jt 8

7 8
6 5
4

4
33

6 7 8
4 5 5
4 3
2

-O-

5 6
3 4

1
5
4
2

IT

l
5
4
2
THE

10. W h e n the fifth is taken as an a d d i t i o n a l part, i t remains

stationary. I t may, b u t need not, be present i n the preceding
c h o r d . By means of this i n t e r v a l the f o l l o w i n g t r i a d is made complete and, even w h e n the second is o m i t t e d , f o u r correct parts are
retained, as i l l u s t r a t e d i n the last three examples of Figure 372.
H e r e again cise a t t e n t i o n must be given to the r e s o l u t i o n of the
seventh i n order to forestall confusin w i t h the seven-four c h o r d ,
whose signature is the same as ours. T h e f o u r t h a n d fifth examples
are sometimes indicated w i t h 9 instead of 2.
Figure 372

5
3

te

1.

settings of o u r c h o r d are c o n t i n u e d f r o m the o p e n i n g c h o r d . I n b

a l l of the tones of o u r chord lie i n the preceding c h o r d ; i n c o n l y
the s i x t h must be sought out. I n i and 2 the second joins the t h r o n g
as a detail of decorative motions i n the m i d d l e parts; henee i t s h o u l d
be o m i t t e d i n i t i a l l y . I n 3 the second is o p t i o n a l . T h e raised seventh
resolves before the entrance of the six-four c h o r d . I n 4, 5, a n d 6 the
second is o m i t t e d for the reason stated i n the preceding paragraph.

mu

8 6 5
4 - 3

7
4 5
4
2

8
-6>
4*
3

l
5
4

8
3
3

-6-

CHORD

H A SS

7
5
4

8
3

OF THE

11 1 .
*

5
4
MAJOR

IIL^J -

-5IM^f
1

8
8

-6.

5
4

8
3

S E V E N T H , II

T h e ascending major seventh may n o t be prepared as the

octave of the preceding bass note. Henee Figure 373 is incorrect:

Figure 373

> r r
1

'

wrong

2. W h e n the seventh is retarded by an octave the other parts

are not affected b u t enter w i t h the bass note. T h e octave becomes
a dissonance, b e i n g restricted by the second above the bass. Henee
i t resolves by stepwise descent to the seventh. I n the signature of this
relationship 8 a n d 7 stand adjacent to each other, and the remaini n g figures that enter w i t h 8 are placed below i t . I n Figure 374,
Example a, the second as w e l l as the seventh is retarded, b u t by a

T II O R O

2i)8

1/

GH

II

ASS

t h i r d . Henee, this latter interval, like the octave, assumcs the characteristics of a dissonance. I n b only the second is retarded, again by a
t h i r d . T h i s t h i r d may be d o u b l e d i n the preceding t r i a d (c). T h e
notated d i s t r i b u t i o n of parts is the best i n a l l of the examples.
3. W h e n the f o u r t h , l y i n g i n the p r i n c i p a l part, is retarded by a
7
fifth, the accompanist plays, o n the entrance of the bass note, j, 4,
7
2
or simply 2, according to the need for a f u l l or t h i n accompanim e n t (Figure 375).

Figure 375

Puf

As

(5

a.
i 3

w
7
2

6
4

99

i
1

'8r

-p-

9
7

8
6

11

8
>

iN-r-^
1 1

7
2

es

4. Figure 376 is best accompanied i n three parts. I f a f o u r t h part

should or must be added, the fifth rather t h a n the f o u r t h is taken.
T h e setting w o u l d be made ugly by a d d i n g the f o u r t h to the several
appoggiaturas already present. T h e fifth, on the other hand, throws
the notated appoggiaturas of the p r i n c i p a l part i n t o bolder relief
and makes the succeeding t r i a d complete. A fifth part cannot be
used to attain completeness, for the progression does n o t assimilte
f o u r parts very w e l l , let alone five.

rfl

'r

BASS

5. T h e apparent d o w n w a r d m o t i o n of the major seventh i n

Example a of Figure 377 is the result of an ellipsis. T h e complete
9
relationship is i l l u s t r a t e d i n b. I n Example c the resolution of 7 is
7
differentiated f r o m that of 2. A l t h o u g h no detail of resolution
98
w o u l d be lacking i f i n the last bar 7 8 were taken instead of o u r
43
c h o r d , n o one cart deny that the i l l u s t r a t e d progression is closer to
the sense of the passage.
Figure 377

b.
-

H OROUGH

* m

'

7
3

7
4
2

8
3
3

Figure 376
THE

CHORD

OF THE NINTH,

1. T h i s chord consists of a n i n t h , fifth, a n d t h i r d .

2. Its signature is 9 8 w h e n the n i n t h resolves o n a stationary
bass, b u t simply 9 w h e n the bass progresses. Accidentis are n o m o r e
to be overlooked here t h a n i n other signatures.
2

8
3

8
3

5
4

1 Cf. Arnold, op. cit., pp. 693-694.

2 Henee this chord is not to be confused with the present-day " n i n t h chord," which
Bach discusses later as the nine-seven chord.

T II O It O U ( II

00

li A S S

3. C o m p r i s i n g the chord of the n i n t h are major and m i n o r

nintlis, augmented, peTect, and d i m i n i s h e d fifths, major and m i n o r
thirds.
4. T h e n i n t h is a dissonance w h i c h must always be prepared. I t
resolves by stepwise descent (Figure 378).

T I I O It O II (', 11
Figure 379
6
- 5
a

/ .

Ai

Figure 378

i:

1, '
jf

9
8
^ 6 1 8

6
5

9
5

9 8-6
'

r.

II

9 8 6
5 6 5

\nm

8
*

f
X

*(1

6
5

f
1

v
Vi 6
5b -

r e m a i n u g l y . W h e n the best disposition cannot be taken i t i? better

to o m i t the sixth f r o m the six-five c h o r d a n d d o u b l e the t h i r d i n 3

*V

9 5

' r 1 r f r ir
p

TTf

(g).
8. I n Figure 380 w i t h its a l t e r n a t i n g n i n t h a n d six-five chords
the only disposition free of errors is that i n w h i c h the n i n t h is placed
i n the lower m i d d l e part. T h e fifths that oceur i n the other two dispositions, n o matter how ardently they may be defended, are a n d

9 8
&- 6

f f

'

6. T o realize the chord of the n i n t h , take the t r i a d above the

bass b u t strike the n i n t h instead of the octave. Those w h o k n o w
the 2 chord k n o w the c h o r d of the n i n t h .
7. T h e m a j o r n i n t h may be accompanied by either the perfect
or the augmented fifth. W i t h the perfect h f t h the t h i r d may be
major (Figure 379, Example a) or m i n o r (b); w i t h the augmented
f i f t h i t is always major. T h i s latter fifth lies i n the preceding c h o r d
and resolves w i t h the n i n t h or by itself (c). T h e m i n o r n i n t h may be
accompanied by either the perfect or the d i m i n i s h e d fifth. W i t h the
perfect fifth there may be either a major (d) or a m i n o r t h i r d (e). I n
the latter case the fifth sometimes ascends to the sixth o n the
resolution of the n i n t h (e). A l t h o u g h the d i m i n i s h e d fifth may
be taken freely (/), i t is better w h e n i t lies i n the preceding chord

9 8
-S- 6 ' -6
-6
:
1 . -r
0

r r r P P

5. T h e n i n t h has the same position o n the staff as the second b u t

is clearly distinguishable f r o m i t i n its accompaniment, preparat i o n , a n d resolution. I n the case of the second, the dissonance lies i n
the bass, w h i c h must be prepared and resolved; b u t i n the n i n t h the
dissonance lies i n the upper tone, w h i c h must be prepared and resolved. Differences i n the accompaniment of these t w o dissonances
have already been noted and w i l l be enlarged u p o n i n this a n d the
next section.

liASS

J ,J J J j

11

-m7

T h e object of this remark is Marpurg (Cf. Arnold, op. cit., pp. 401 ff.), and
possibly Cari Heinrich Graun, whose support of greater tolerance i n these matters is
cited by Marpurg.
3

T II O R O U (, H BASS

T I O R O U G H li A S S

o2

stead (a). Otherwise the realization i n b, e m p l o y i n g a d i v i d c d accompaniment, should be noted and used whenever possible.
9. T h e fifth is o m i t t e d f r o m three-part realizations. Because one
i n t e r v a l is thus lost, the accompanist must exercise the same care i n
using this k i n d of accompaniment as we have f o u n d necessary i n
other s i m i l a r cases.
THE

CHORD O F T H E NINTH, II

1. T h e n i n t h is and remains a n i n t h even w h e n i t is placed

directly adjacent to the bass. T h i s relationship is often unavoidable.
For example, composers f r e q u e n t l y meet i t w h o w r i t e obbligato
parts for bass instruments. A double bass is best fitted here t o give
the lowest voice its proper gravity. O f course, aside f r o m such a case
Figure 381
d

J)

1 13

, ,

8 7 6

S 8

li a.>'
j ** _4_j
>

1 m

II

J)

Aiih

"^4 &==

^ 1
9

J>J.

"t

II2

.. j > , j

ie*

- 4
etc

r
6

*~T
6
5

3. T h e n i n t h may n o t be prepared as a n octave over the preceding bass. Henee Figure 382, Example a, is w r o n g . R e s o l u t i o n
of the n i n t h t o the octave is the cause of this r u l e . Henee, w h e n the
resolution is n o t t o an octave, the r u l e may be ignored. Formerly,
musicians wrote thoughtlessly i n the manner of Example b. These
octaves o n the after beat sound n o better than the octave preparat i o n and resolution of the n i n t h . Proof that this r u l e was introduced
because of the resolution rather than the preparation is provided
by the fact that other dissonances may be prepared o n the octave
(c). Despite this, octave preparation of the n i n t h is never attractive.
I t must be avoided: (1) i n the outer parts; (2) i n t h i n settings; (3)
except for c o n t r a p u n t a l reasons. T h e bass must always be changed
o n the resolution, i f this use of the n i n t h is t o be allowed. I believe
that the direct fifths of Example d (which some defend by c l a i m i n g
that they are covered, i n the notated d i s t r i b u t i o n ) sound worse than
the after-beat octaves of Example a. However, b o t h are poor.
5

a. and b.

d.

as this, i t is always better to place the n i n t h nine degrees above the

bass.
2. Examples a and b of Figure 381 cali for a d i v i d e d accompaniment w h e n tney are realized i n f o u r parts. I n the first the n i n t h , as
a passing i n t e r v a l , is n o t resolved (a) and i n the second its resolut i o n is retarded (>). W h e n the accompaniment is n o t d i v i d e d the
t h i r d chord is realized i n three parts (c). I n d a d o u b l e d t h i r d or
sixth i n the c h o r d of the s i x t h is the best setting, for i t eliminates
large leaps and prepares the d i m i n i s h e d fifth i n the c h o r d of the
n i n t h . I n e three parts are safest. Should a f o u r t h p a r t be used, the
sixth must appear o n top over the first note (/). T h e r e m a i n i n g t w o
dispositions crate fifths.
4

*. h -
y
1-

~
<fc

303

r
7

5b

=J=
4=f

-T*

W H

Figure 382
A 5 6

8 b.

J-'i^i-jU-

11J*

* Remainder of paragraph from ed. of 1797.

His father must be included i n this stricture. Cf. J . S. Bach, Choralgesange
5

(Br. u. H.) No. 209, Jess Meine Zuversicht, bar 1, bass and tenor.
Cf. Arnold, op. cit., pp. 397-406.
7 Examples from b to end from ed. of 1797.

O R O U (',

V //

y>4

11 li A SS

j.

5b
THE

1.

f r

CHORD

3
1

T h i s c h o r d consists of a n i n t h , sixth, and t h i r d .

9

2. Its signature is 6 w i t h appropriate accidentis. T h e resolut i o n of the n i n t h leads to a chord of the s i x t h w i t h the bass d u p l i cated. Henee, those f a m i l i a r w i t h this latter c h o r d can easily find the
nine-six.
3. T h e three intervals that comprise the chord may be major
or m i n o r , as i l l u s t r a t e d i n Figure 383. T h e disposition w i t h the
n i n t h on top is generally best. T h e three examples that bear the
letter a sound rather poor even i n this disposition. A n i m p r o v e d
progression follows each example.
Figure

38?

--f

1,1

9
6

9
6

a.

9
8 tp !>7
6b
5i>

8
6

J I

I J 1 p *p

better

6
4

5
tt

6
4

5
jt

8 Vi
6b 5b

1 f fHrJ4j*-*p

7
5

9
6
t

8 - 7
5
% ~

I I I| II i

9
%

better

384

a.
Gf.

'

1i

s
4

8
3

7
5

6
5

9
4

8 8
- b7
3

5b

4. T h e n i n t h may be major or m i n o r ; the fifth, augmented, perfect, or d i m i n i s h e d ; b u t the f o u r t h , as we shall see i n the f o l l o w i n g
examples, is always perfect. I t is better to prepare the d i m i n i s h e d
f i f t h than strike i t freely. T h e augmented fifth, too, must lie i n the
preceding chord (Figure 385).
Figure

385
9 8
6
5

"r'r r "i

(>

better

Figure

4=^

9
5

C:HORD

Its signature is 4 , w i t h accidentis w h e n necessary. W h e n

g
b o t h dissonances resolve over the o r i g i n a l bass note, 3 is placed to
the r i g h t of the signature.
3. Since b o t h the n i n t h and the f o u r t h must be prepared, i t is
only the r e m a i n i n g interval that must be located. I n order easily
to i d e n t i f y the tones, take the six-five c h o r d on the second below the
bass note, for i t contains the same tones as our c h o r d and often precedes i t . I t w i l l also be recognized by those w h o k n o w the five-fourtwo chord. B o t h dissonances resolve simultaneously i n most cases
(a); b u t on occasion successively (b) (Figure 384).

j 1 e
1

ss

/< //

T h i s cliord consists of a n i n t h , fifth, and f o u r t h .

0
9
5

NINE-FOUR

THE

1.
2.

NINE-SIX

o li o u ; //

//

9 8 7
4

98

4)1

5 6
4 3

9 8
5b

6
5

4 3

98
4 3

y M Ulf If ir UN IJ IJ i
5.

T h e sixth must be i n c l u d e d i n the signature ( 6

) i f i t is to be

taken i n place of the fifth. I t may be major or m i n o r . T h e tones of

this c h o r d may be i d e n t i f i e d by p l a y i n g the c h o r d of the second o n
1 C f . A r n o l d , op. cit., p p . 6 9 8 - 6 9 9 .

1 C f . A r n o l d , op. cit., p p . 6 9 7 - 6 9 8 .

o6

T II O R O U G H t A S S

TU

the g i v c n bass note. O l t e n the sixth nioves to the fifth w h i l e the

n i n t h and f o u r t h resolve. Such being the tase, only two d i s t r i b u tions of the tones are practicable, for the r e m a i n i n g one creates
fifths. T h e last three examples of Figure 386 illustrate this progression:
Figure 386
b7

11

9
6

8
- TS

9 8
6 -

[I

JJ

^ 6 5
'5
4

9 8
6 -

if"

J
8

|J

r
9
7

9
6

b7
5b

MhJ-J

8
5
3

7
5

9
6

4=J

&

JEBN

r
6

b.

a.

DI

6. I n the galant style the f o u r t h is sometimes played w i t h o u t

preparation (Figure 387, Example a). T h i s unprepared f o u r t h may
even be augmented (b). Such fourths are b r o u g h t about by the i n t r o d u c t i o n of appoggiaturas that are p r o v i d e d w i t h a three-part accompaniment. T h e first example is better t h a n the second.

6
5 66 5

9 8

4 3

4
4!

r
'5

T H E

NINE-SEVEN

CHORD

1. . T h i s c h o r d consists of a n i n t h , seventh, a n d t h i r d .
9

2.

Its signature is 7, w i t h accidentis as r e q u i r e d . W h e n b o t h

8

dissonances resolve over the o r i g i n a l bass note, 6 follows the o r i g i n a l

signature.
3. T h e n i n t h as w e l l as the seventh must be prepared. B o t h dissonances resolve by stepwise descent, simultaneously i n most cases
(a), b u t at times successively (b) (Figure 389).
Figure 389

Figure 387

a.

a.

|JJ

HASS

/(;//

a,

5
7
5

n o

Figure 388

distribution
9
6

r r
'6 5
'6 5
4 3

9
9

8
8

f
|6

7. T h e examples of Figure 388 are also realized i n three parts.

I n the second i l l u s t r a t i o n u n d e r a the n i n t h as w e l l as the f o u r t h
seems to lack preparation. I n b, however, i t can be seen that the
opposite is true, as soon as the appoggiaturas are removed. T h e accompaniment to b o t h of these illustrations is the same as the realization i n a.

8 7

9
7

8
6

b. ,

7
%

9
7

b.

8
7

b7
5b

8
-

4. As illustrated i n Figure 390, the three intervals that comprise

the c h o r d may be major or m i n o r .
5. Occasionally the octave of the preceding c h o r d must be taken
as a fifth part i n order to prepare the seventh. Such being the case,
the fifth of o u r c h o r d should also be played, for i t too w i l l lie i n the
1 T h e e q u i v a l e n t of t h e p r e s e n t - d a y " n i n t h c h o r d . " I t s h o u l d be n o t e d t h a t B a c h ' s
c h o r d resolves u n i f o r m l y o n a s t a t i o n a r y bass. C f . A r n o l d , op. cit.,

pp. 699-701.

T I I O li O II C, I I
n g u r e 3 90

M
J

V:

1)
(

l>7

8
r 6

8
6

5b

i A S S

TU

8 7
7 6 r

II '
H

5
4

9 8 7
7 6 5

*^~\ H*

6 5

II y

9
7

8
6

preparatory c h o r d . I t may be d i m i n i s h e d , perfect, or augmented

(Figure 391).

O It O II C. I I

li A SS

309

the seventh, as illustratcd i n the last lour examples. I n the first t w o

examples of Figure 39a the tones of the irst chord should be dist r i b u t e d i n such a manner as to avoid the seventh i n the upper
part.
7. Example a of Figure 393, similar to several others w h i c h have
already been illustrated, is realized i n three parts. T h e f o u r t h part
is regained on the chord of the second. A three-part setting is also
used i n b and c, where the suspended seventh and n i n t h move upw a r d before resolving. I n d the chord of the. seventh enters prematurely, as explained i n e. I n / o u r c h o r d , unresolved, results f r o m the
m o t i o n of a passing bass. T h i s k i n d of passage w i t h varied figuration
can be f o u n d i n heavily scored, noisy pieces, such as symphonies,
etc. (g).
Figure :>93

a.

=H= J J i IIII
f

r tf

tf

6. I f the f o u r t h is to be taken instead of the t h i r d of this chord,

i t must be expressly indicated i n the signature. Since this tone, too,
w i l l be present i n the preceding c h o r d , the entire construction, i n c l u d i n g the fifth as an a d d i t i o n a l part, w i l l lie under the fingers. T h e
f o u r t h may be perfect, d i m i n i s h e d , or augmented; and as stated
above, i t is prepared. A fifth part is also taken i n order to prepare
Figure 392
b7,

9 8
7 6 . 7

9
7

8
6

7
5b

b.

- c

10

8' 7 6
5
4
6
2

5 b

--

d.

Hf=S

>

'- T. Uf f 11
i

*F=f
v ,

'

e.

II *

I
6

~ ~ i

9
7

r ~ r

|II 1=f H

10

rr

'r r r
T

6 1
3

87
76 5

/
6

Ltt '

1 1

117

T I I O H o u r. I I

3 io

T H E

1.

FIVE-FOUR

If A

CHORD

ss

T I I O H () II (, I I

T h e five-four chord consists of a f o u r t h , lilil, and octave.

5

2. Its signature is 4 3 or 4 3 w h e n the f o u r t h resolves to the t h i r d

over a c o m m o n bass note. B u t w h e n the resolution occurs over a
5

m o v i n g bass, 4 or 4 is sufficient. I n the first case, 3 is often replaced by an accidental w h i c h specifies the size of the t h i r d . T h e
accidental must be separated f r o m the preceding 4 i n order to i n dcate clearly that i t refers, n o t to this n u m e r a l , b u t to 3.
3. T h e perfect and d i m i n i s h e d fifth, the perfect f o u r t h , and the
octave are the intervals w h i c h may appear i n the five-four chord.
4. T h e f o u r t h is always prepared and resolves by stepwise descent. T h e fifth, w h i c h restricts this dissonance, is n o t always present
i n the preceding chord, being struck freely at times, even w h e n i t
is d i m i n i s h e d (Figure 394).
Figure 394

3LL
4

**4

6*

ti

43
5,

4, 3

- f H h *

-rHhd

11

5. I n order to lcate the chord, play the t r i a d o n any bass note,

b u t substitute a f o u r t h for the t h i r d . T h i s procedure makes i t easy
to learn the d i s t r i b u t i o n of tones and the resolution of the f o u r t h .
6. T o avoid fifths i n Example a of Figure 395, the octave of the
five-four chord must be o m i t t e d and the fifth d o u b l e d instead. N o
interval is thereby lost. T h i s step is n o t necessary af ter the r e m a i n i n g
dispositions of the preceding chord of the seventh. I n b the t h i r d

1
7

11

h
1r

-i

BASS

3"

J * 'J

1_

"9*

times enters unprepared (along w i t h the fifth) i n the f o r m of an

appoggiatura w h i c h cannot be o m i t t e d f r o m the accompaniment
unless i t is replaced by a rest. I n Example a, Figure 396, the perfect
f o u r t h may enter by stepwise or leaping m o t i o n . B u t i n b the aug2

mented f o u r t h is approached o n l y by step, and 4 must be expressly

called for. T h e notated d i s t r i b u t i o n of tones is the most acceptable.
I n other cases the augmented f o u r t h may be o m i t t e d f r o m the acc o m p a n i m e n t and replaced by a quarter rest (c). I n d a l l forms of
the chord of the s i x t h may be used, and the perfect f o u r t h w h i c h
follows may be approached by step or leap. B u t the realizations of
dd must be avoided.
Figure 396
b.

i
6

5
4

dd'

wrong

11

(5

wrong

of the chord that precedes 4 must be d o u b l e d . I f this cannot be

done, the accompaniment must be d i v i d e d (bb).
7. I n the galant style the perfect or augmented f o u r t h some1 C f . A r n o l d , op.

cit.,

pp. 694-698.

8. I f for adequate reasons a three-part realization is r e q u i r e d ,

the octave can be most readily o m i t t e d .
2 F o r a n e x t e n d e d d i s c u s s i o n of t h e a c c o m p a n i m e n t to a p p o g g i a t u r a s , cf. C h . V I ,
"Appoggiaturas."

C H A P T E R

S I X

CCOMPAMIMEMT

T H E

UNISON

1
H E octave is i n c l u d e d i n the meaning of the t e r m unisn.
T h u s w h e n parts progress either i n real unisons or i n octaves, they are said to move i n unisn (aW unisono),
even
w h e n the figuration of one of the parts is different f r o m that of the
other (Figure 397).

Figure 397

xn m

2. T h e r e is no need to c o m m e n d this technique, w h i c h attains

its beauty t h r o u g h the omission of harmony, for the many examples
of i t to be f o u n d i n the works of good composers provide a dependable testimony.
3. Yet i t is surprising that some composers do n o t always specify
a unisn accompaniment i n scoring the bass. Figures w i l l be f o u n d
where they are n o t to be realized. T h e results can only be u n h a p p y .
I m a g i n e a s i t u a t i o n : A composer works industriously over a piece,
lavishing o n i t every last resource of melody and harmony. A t a
certain p o i n t he feels that his audience must be roused w i t h somet h i n g different. H e searches enthusiastically f o r a passage whose
splendor and majesty shall be pronounced and s t r i k i n g . H e decides
to discard the beauty of h a r m o n y for a w h i l e ; the passage shall be
played i n unisn; i t alone is to occupy the thoughts and actions of
1

C f . A r n o l d , The

Art

of Accompaniment

from
313

a Thorough-Bass,

C h . I I I , Sect. 12.

C C O M PA N 1 M E N T

the perormers. T h i s is i o l l o w e d by a rcsumpon of liarmony, etc.

H e completes the work and i t is p e r f o r m e d . His pleasant expectations of the i n t e n d e d execution of the passage are shattered by the
accompanist w h o , at the keyboard, prepares and resolves the i n d i cated intervals as carefully and regularly as possible. A t another
t i m e this w o u l d be m u c h to his credit, b u t n o w i t is only a source
of annoyance. Fortunately, for h i m , the composer realizes that he
overlooked something i n w r i t i n g o u t the bass part a n d is overjoyed
w h e n the accompanist, displeased at his i n a p p r o p r i a t e accompaniment, abandons chords and reinforces the passage i n accordance
w i t h the first r u l e of accompaniment as stated i n Paragraph 19 of
the I n t r o d u c t i o n : A n accompanist must fit to each piece a correct
performance of its h a r m o n y i n the proper v o l u m e .
4. I n order to clarify this r u l e , we shall discuss t w o cases i n
w h i c h the accompanist is obliged to play i n unisn. Accompanim e n t i n the unisn occurs w h e n the bass is played i n octaves w i t h
b o t h hands.
5. T h e first case relates to passages w h i c h are w r i t t e n for one
part. W h e n a l l performers play i n unisn i t is only n a t u r a l that the
accompanist too should f o l l o w the unisons and give u p his chords.
Such passages usually carry the i n d i c a t i o n , unisoni or
all'unisono.
6. T h e r e is a special case w h i c h departs somewhat f r o m the
preceding. I t occurs w h e n the r i p i e n o parts are i n unisn w i t h the
bass w h i l e the p r i n c i p a l part has a l o n g held note or a different
melody. T h e r i p i e n o parts must be observed carefully to learn
whether their b r o k e n chords give expression to the essential intervals, especially the dissonances and resolutions of the u n d e r l y i n g
h a r m o n y . I f they do, the accompanist also should play i n unisn
(Figure 398, Example a). B u t i f the accompaniment to the p r i n c i p a l
part is simple a n d n o t only calis for harmony, b u t attains its affect
t h r o u g h the use of i t , a chordal setting must be chosen (b). Especially needed here is an exact i n d i c a t i o n of the r e q u i r e d accompaniment, f o r free choice demands an insight w h i c h is capable of decidi n g whether a chordal accompaniment helps or hinders the p r i n c i pal part. F u r t h e r m o r e , according to circumstances the case discussed
here m i g h t take either type of accompaniment.
2

2 I.e., o n e n o t e for e a c h h a n d .
s W i t h r e g a r d to t h e t r e a t m e n t of d i s s o n a n c e s i n a unisn a c c o m p a n i m e n t , see C h .
V I , " S o m e P r e c a u t i o n s " o f A c c o m p a n i m e n t , " f 3 , w h e r e B a c h takes a freer v i e w of
t h e n e e d for r e s o l u t i o n s .

// C C O M I ' A N

1 M E NT

375

Figure 398
Adagio

uns.

7. For special reasons, composers sometimes place a melody i n

the bass and accompany i t i n unisons i n the n a r r o w sensethat is,
w i t h o u t any octave d o u b l i n g either above or below. W h e n the bass
is thus to be played o n l y i n the notated register, the accompanist,
his r i g h t h a n d silent, plays such misleading unisons i n one part
alone w i t h the left hand. T o be played i n a similar manner are
those melodies w h i c h , w h i l e n o t always b r i l l i a n t , are of s t r i k i n g expressiveness and appear o n occasion i n the lowest part alone. T h e y
should n o t be h i d d e n beneath a harmonic accompaniment or
raised i n register by octave doublings. T h e composer who contrives
such a studied effect must desgnate i t w i t h great accuracy or his
plans may fall short of f u l l realization i n performance.
4

8. T h e second case that calis for unisn accompaniment concerns a l l b r i l l i a n t passages for the lowest part i n the setting of w h i c h
the composer has a special purpose i n m i n d . T h e y may be fashioned
o u t of leaps, runs, b r o k e n chords, successive trills, and countless
other figures. F r o m our p o i n t of view, such passages must stand o u t
clearly, and this is achieved less by a chordal t h a n a unisn accompaniment. I t is n o t yet a general practice to desgnate this case,
unisoni or all'unisono;
henee, the manner of support is left to the
discretion of an understanding accompanist. Experience has proved
to me the effectiveness of a unisn accompaniment i n such passages.
9. B r i l l i a n t basses are usually p r o v i d e d w i t h a chordal acc o m p a n i m e n t only i n two-part pieces such as a solo or solo aria.
10. T e r m i n a t i o n of a unisn accompaniment is indicated by
figures placed over the bass at the p o i n t where a chordal setting is
resumed. S hould the first note express a t r i a d that is indicated
o r d i n a r i l y w i t h o u t any figures, at least one of its intervals must be
designated.
* O n e of the m o s t s t r i k i n g l y e x p r e s s i v e e x a m p l e s , w r i t t e n l a t e r b y B a c h , is t h e
o p e n i n g of Die Auferstehung
und Himmelfahrt
Jesu ( W o t q u e n n e N o . 240).

ji6

A C C O M TA
ONE-PART

ACCOMPANIMENT

1M E N T
F O R

T H E

L E F T

A
11AND

1. I n this k i n d of accompaniment, w h i c h is indicated by t.s.,

tasto, or tasto solo, the left h a n d alone plays the bass w i t h o u t octave
d u p l i c a t i o n . I n certain passages i t is as m u c h needed as the unisn
accompaniment w h i c h we have just discussed; and performance w i l l
suffer as m u c h f r o m an incorrect i n d i c a t i o n of i t as w i l l the unisn
accompaniment.
2. Italians do n o t use either of these accompaniments. Perhaps
they believe that only chords can be played o n our instruments and
consider i t u n f i t t e d for the accompaniment of the most b e a u t i f u l ,
affettuoso passages. T h e y do n o t care to have the t i n k l i n g sounds of
their keyboardists i n such places; the more so because i t is k n o w n
that they can play scarcely any c h o r d w i t h o u t r o l l i n g i t . Henee i n
I t a l i a n works, delicate passages usually carry the d i r e c t i o n sema
cmbalo over the bass as a k i n d of w a r n i n g . W h o l e arias sometimes
have this i n d i c a t i o n , w h i c h is laughable to the singers of that country w h e n they are shown the words i n their o w n scores.
3. W e use the tasto solo to great advantage i n suitable passages;
for example, w h e n the bass and p r i n c i p a l part move i n thirds and
sixths w i t h n o a d d i t i o n a l voices. T h e piece may be for t w o or more
parts. Should the bass be m a r k e d piano, and the thirds or sixths
lie cise together, thus p r e c l u d i n g octave d u p l i c a t i o n , n o other acc o m p a n i m e n t is as n a t u r a l as ours. T h e double basses are silent

C C 0 MPA

NIM

E N

317

while the other bass instrumenta play softly i n the notated register
along w i t h tire keyboard. The examples of Figure 399 are typical.
4. O n the other hand, w h e n such passages are d o u b l e d and the
thirds or sixths are widely separated, the accompaniment i n unisn
or all'unisono may be used, the bass being d o u b l e d . I f this p a r t does
not go too low, the d o u b l i n g should be i n the lower rather than the
upper octave. Such settings can be f o u n d i n symphonies and concertos. T h e first and second v i o l i n s play together w h i l e the violas
and basses move i n unisn (Figure 400).
Figure 400
Allegro

, j m m n mn,

rgn^

5. A t half and whole cadenees where the p r i n c i p a l p a r t has an

appoggiatura whose relase is piano, as discussed i n the first p a r t of
this Essay, an accompanist at the harpsichord plays o n l y the bass;
at the clavichord or pianoforte, however, b o t h the appoggiatura
and the relase may be played, b u t execution must be adj usted to
the v o l u m e and l e n g t h of the ornament i n the p r i n c i p a l part, so
that i t w i l l r e t a i n the freedom to p e r f o r m its notes i n accordance
w i t h the affect. O n the pianoforte an alternative is to p e r f o r m o n l y
the bass under the appoggiatura as l o u d l y as necessary and then the
relase q u i t e softly w i t h the r i g h t hand.
2

6. Tasto solo is also used on a bass over w h i c h a melody is set

i n a low register w i t h n o accompaniment above i t . B u t w h e n there
is an accompaniment by several instruments, also l o w , figures may
be placed over the bass, w h i c h an understanding accompanist w h o
is aware of the construction of the piece w i l l p e r f o r m o n l y i n the
low register. However, because the j u d g m e n t of thorough-bass
players, many of w h o m are dilettantes, is n o t always t r u s t w o r t h y ,
i t is better and safer i n this case also to indcate .5. over the bass
and dispense w i t h the h a r m o n y of the keyboard than to endure an
accompaniment w h i c h cries o u t above the other instruments and
i C f . A r n o l d , op.

cit.,

C h . I I I , 12.
2

C f . C h a p t e r I I , " T h e A p p o g g i a t u r a , " f 7.

7./.V

C C O M P A N 1 M I: N T

r u i n s the passage. T h i s k i n d o i setting occurs in concertos for lowpitched instrumenta, arias for low voices, etc.
7. T h e f o l l o w i n g a d d i t i o n a l instantes o tls k i n d of accomp a n i m e n t should be noted (Figure 401). I n a, where the p r i n c i p a l
voice starts i n actual unisn w i t h the bass, the first note should be
played t.s. I n b, the r i g h t hand is silent at those places where t.s. appears, even w h e n there are figures. Performance i n a slow tempo
w o u l d suffer here were the accompanist to anticipate the change of
harmony of the p r i n c i p a l part.

C C 0 M P A N 1M E N T

319

weak that only in this manner can a proper balance be reached. B u t

i t is always better and more appropriate to the nature of tasto solo
w h e n such an expedient is not employed, for i t is i n respect to
d o u b l i n g that tasto solo is distinguished f r o m unisono.
9. T h e r e s u m p t i o n of a chordal accompaniment after t.s. is i n dicated by the reappearance of figures, just as i n the case of unisono.

T H E

ORGAN

POINT

1. T h e organ p o i n t or point d'orgue occurs w h e n various harm o n i c changes, orten i n v o l v i n g tied notes, are made over a h e l d or
repeated bass note.
2. I t appears generally i n learned things, especially fugues, near
the end over the d o m i n a n t or over the final note. Occasionally i t
w i l l be f o u n d i n the course of a piece over the d o m i n a n t or tonic
of a key reached by m o d u l a t i o n . I n the first case, composers often
introduce a l l manner of contrapuntal devices i n stretto.
3. T h e organ p o i n t may be i n three or more parts. T h e harm o n y is usually complete even w i t h o u t the bass, w h i c h however,
adds a final, appropriate gravity. I n order to comprehend or e x p l a i n
the chords and the unusual combinations of intervals the bass
should be disregarded. W h e n this is done, the strange signatures
t u r n o u t to be indications of n o t h i n g more than the ordinary progressions of t h o r o u g h bass.
4. I t is not easy to figure organ points, so they are usually set
tasto solo. Those w h o do figure t h e m must accept the fact that they
w i l l be played tasto solo anyway. T h e reason for this can be
ascribed not only to a justifiable simplification of the accompanist's
tasks b u t often to the impossibility of reading the figures. Assumi n g that the r i g h t hand could accompany all organ points, gratitude
w o u l d never compnsate for the expended anxiety and t r o u b l e .

8. T h e bass i n tasto solo must never be d o u b l e d i n the octave

by the left hand, unless the passage is so l o u d and the i n s t r u m e n t so

5. T o play the organ p o i n t tasto solo removes the necessity of

scanning unusual signatures and successions of t o w e r i n g figures.
Parts are often constructed i n such a manner that one crosses the
other. T h i s m i g h t oblige the accompanist to cross parts, w h i c h is
n o t allowed i n t h o r o u g h bass, since many errors m i g h t be thereby
excused w i t h o u t satisfying the ear. I n such a case, therefore, the entire organ p o i n t must be played i n d i v i d e d accompaniment to en1 C f . A r n o l d , op.

cit.,

Ch. X X .

jao

A C C O M > A N M E N I

A C C O M

1' A N I M E N T

321

sure correct preparation and resolution, and pievenl the r i g h t hand

f r o m descending too far. T h i s is an excessive demand. Sometimes
the chord changes are so r a p i d that they can scarcely be b r o u g h t out
even when the accompanist tries to realize them.
6. T h e examples of Figure 402 w i l l suffice to illustrate these remarks. Figures have been i n c l u d e d i n order to provide an understanding of the chords. A setting w i t h o u t the organ p o i n t follows
each example.
Figure 402

M u ^ r i i T r
^

4
t

4 7 6 2 5
5

i i n r r

'6
4

32:

8
4

f
-

l>7

^
6
4

5
4

7
3

~
8

,2

^
6
5

6 7 2

7
5
4
.6 -

6 5P5 4

!>7
6 j7
- 4 4
3
3
2
6

5 2 5 5,

b7
5

81>7

r i

- r i r r
4

7 6
4 I,

"f

5 3
4 4

1 J 1 M;-

8
3

26

4 ^

76

7 6 44 t K

A^A

1 *T~^9.
:

Lj

" gm
= M = F

4f 6

#4 2 5

-F-T-H
^5-1

6 15

Allegro

j H
1 = 1 ==1 U

le

l>?

6
4

5
3

6
5
3
2l> 8

f
6b
4
2

5 6 7 6 5 8 7 6
j } 4 5 4 $ 6 5 4

5
#

9 8 7 6 8 7 6
7 6 5 4 6 5 4

5
#

J22

ACCOM

P A N I
APPOGGIATURAS

MENT

ACCOMPANIMEN
F.gure 403

3. Appoggiaturas r e t a r d the chords w h i c h are called for by the

bass. I t is c o m m o n knowledge that, according to the rules of good
performance, the ornament is emphasized and its relase played
l i g h t l y . Consequently, i t is d o u b l y w r o n g to o m i t i n d i c a t i o n of
t h e m f r o m the signature, for w i t h o u t some clue the accompaniment i n most cases can only be poor. T h e chord retarded by an
appoggiatura takes o n a q u i t e different appearance t h r o u g h an
exact i n d i c a t i o n . B u t , although knowledge of the usual signatures
w i l l n o t suffice and the accompanist must learn to recognize strange
combinations of numeris, he w i l l soon grow accustomed to t h e m .
I n scores where the p r i n c i p a l part is n o t notated i n the c o n t i n u o
part these signatures are indispensable, for the presence of an appoggiatura can n o t be guessed. A n d even w h e n the p r i n c i p a l p a n
w i t h its ornaments is present, how can the numeris be m o d i f i e d i n
performance i f the figures contain no reference to the appoggiaturas? H o w can m i d d l e parts, i f they are r e q u i r e d , be realized?
4. I n Chapter V m u c h material was covered relative to appoggiaturas. T h i s w i l l not be repeated here. O u r present discussion begins w i t h the long, variable appoggiatura. T h e shortest of
these is never more r a p i d than an eighth note i n an allegretto.
5. W h e n a bass note is figured w i t h o u t reference to an accompanying appoggiatura, and this appoggiatura w i t h its relase
complements the given numeris or is identical w i t h one of t h e m ,
the accompaniment need n o t be m o d i f i e d even i f i t is i n f o u r parts.
T h e examples of Figure 403 are constructed i n this manner.
p p . 422 ff.

323

1. I t w o u l d be superfluous to repeat the discussion of the appoggiatura w h i c h is contained i n the rst part of this Essay. I assume
that the reader has read this material carefully, for i t is inseparable
f r o m the present remarks.
2. I t is rare that an accompaniment can be constructed w i t h o u t
reference to appoggiaturas, for they are i n most cases an integral
part of i t . T h e y appear most f r e q u e n t l y i n pieces where taste rules,
for they are one of its outstanding refinements. Such pieces r e q u i r e
a delicate accompaniment w h i c h aims to b r i n g the appoggiatura
i n t o relief rather than to obscure or destroy i t .

1 C f . A r n o l d , op. cit.,

I'I'I
'7

I V ' P ' . i i ' ^ i 't

8
4

6
6

i AS

' I . '6[

6 '7
6

fll

g ( |
7

ij! 1 j. gpil
4 3

6
5

6
4

uJJ

5
3

" r r r r
4=^

A C C O M VA N l M E N T

ACCOMPANIMKNT

3*4

6. B u t i f the appoggiatura is n o t related to the harmony of the

relase and therefore uitters f r o m the intervals of the bass note over
w h i c h i t appears, the accompanist should play the o r n a m e n t a n d as
many intervals f r o m the indicated signature as r e q u i r e d by its l o u d ness and its suggested harmony. W h e n i t is played softly a n d w i t h
great affect, its length dependent o n the caprice of the p r i n c i p a l
part, the accompanist should o m i t the ornament and play one or at
most t w o s u p p o r t i n g parts. T h i s occurs most f r e q u e n t l y w i t h
chromatically raised appoggiaturas. T w o - p a r t appoggiaturas are
played i n the accompaniment, m a k i n g a total of three parts. Some
appoggiaturas suffer n o accompaniment at a l l . As a corollary to
these remarks i t should be observed that the greater the affect of a
piece, the more delicate must its accompaniment be. Such delicacy
is concerned w i t h the selection, entrance, r e d u c t i o n , a n d omission
of chords. M y m e a n i n g w i l l be illustrated by examples of a l l
characteristic situations.
1?
I n Figure 404 the illustrated delicate accompaniment is
better t h a n the complete c h o r d of the n i n t h .
Figure 404
\>1 6

5t
2 3 7-6.

4 3

J MLM ,

8. Figure 405 illustrates the three kinds of second, used as

ascending appoggiaturas. Even t h o u g h they are n o t always played
by the accompanist they must be indicated i n the signature. Unless
they are reckoned as n i n t h s , t h e i r signature is usually 2 3. A p p r o priate accidentis a n d the r e m a i n i n g figures must be i n c l u d e d . O n e
n u m e r a l above the 2 indicates a three-part accompaniment. I n the
examples that f o l l o w t h r o u g h o u t this section, the first i l l u s t r a t i o n
is figured w i t h o u t reference to the appoggiatura; b u t i n the second,
4

w h i c h is the accompaniment t o the first, the signatures are com5

p u t e d correctly. I n a, 3 may be taken i n the second bar, b u t i n the

f o u r t h bar, accompaniment of the augmented second is momentarily replaced by an eighth rest after w h i c h only the fifth is taken.
2 F r o m t h e e d . o f 1797.
F r o m t h e e d . o f 1797.
* T h a t i s , m a j o r , m i n o r , a n d a u g m e n t e d seconds m e a s u r e d f r o m t h e bass.

325

I n b, only the seventh is rcali/.cd, b u t i n the two-part appoggiatura

of bb the second as well as the seventh is played. I n c, the seventh or
b o t h the seventh and the second may be taken, since b o t h tones appear i n the preceding c h o r d . T h e same applies to d, where either 6
or 2 may be realized. I n e, the second is treated as a n i n t h , that is, i t
moves to an octave. I n a slow tempo the appoggiatura i n / may, of
course, be realized; b u t elsewhere i t is replaced by a quarter rest
and the seventh alone is played. T h e appoggiaturas a n d their re4

leases are accompanied i n g. A slur must be placed over ^ t o i n d i cate omission of the sixth. I n h, a n eighth rest w o u l d be too short,
were the accompanist to o m i t the ornament; henee i t should be
played, especially since i t is present i n the preceding c h o r d . I n i, the
second cannot be treated as a n i n t h because of the f o l l o w i n g /-sharp
5

i n the bass; however, the i n t e r v a l may be o m i t t e d a n d 3 alone

played. T o avoid fifths, the sixth must be o m i t t e d f r o m the first
/-sharp. For reasons of affect, the p r i n c i p a l part i n a slow tempo freq u e n t l y retards o n the a, carrying i t over to the next bar. T h e accompanist should n o t do likewise b u t continu to play i n strict
tempo. I n j , the second may be treated as a n i n t h i n a slow tempo;
b u t elsewhere i t should be o m i t t e d and the t r i a d played directly o n
each bass note. W h e r e many chromatic appoggiaturas appear, as i n
k and kk, the h a r m o n y must be t h i n and interspersed w i t h rests i n
order to b r i n g the ornaments i n t o relief and avoid ugly chords. I n
l the appoggiatura is played, since i t appears i n the preceding chord,
and the fifth is added to i t . A l t h o u g h the seconds i n m may be
realized, the illustrated accompaniment is better f o r a soft performance. Moreover, the rests w i l l help to clarify the appoggiaturas
and remove any b l u r r i n g of t h e i r o u t l i n e . I n n, where the c h r o m a t i cally raised o r n a m e n t concurs w i t h a change of bass, the sixth alone
should be taken. A n y of three accompaniments may be used f o r o,
comprising: (1) the fifth alone; (2) the fifth a n d augmented second;
(3) these t w o intervals a n d the octave. Choice is governed by the
r e q u i r e d loudness or softness of the accompaniment. I n p, the first
second is treated as a n i n t h , b u t the f o l l o w i n g one is accompanied
by the sixth alone, the t h i r d b e i n g struck later. Over the bass note
c o n l y the fifth a n d n i n t h are realized, a n d over the f o l l o w i n g
/-sharp, the d i m i n i s h e d fifth a n d t h i r d . T h e f o u r t h over c a n d its
resolution are o m i t t e d , so that the p r i n c i p a l part may p e r f o r m t h e

)26

C C O M PA NIM

E N T

A C C

resolution w i t h complete reedom. This detail is one of the refinements that are reserved for the p r i n c i p a l part. I t should be observed
here that accompaniments must be so contrived that they clarify or
at least do not obscure the various refinements of melodies whether
these consist of chromatic intervals, retarded and anticipated resolutions, or, above a l l , syncopations, especially i n slow pieces of an
affective nature. C l a r i t y is attained t h r o u g h rests, and obscurity can
be avoided by t h i n n i n g the chords. W e r e all refinements realized o n
the keyboard, listeners w o u l d n o t be able to t e l l whether i t was
being played as an accompanying or a solo i n s t r u m e n t . I n q, the
t r i a d is retarded by a c h o r d of the major seventh w h i c h expresses a
descending appoggiatura o n the second above the bass. W e have
already seen several such examples. T h i s retardation is o n l y occasionally good, b u t bad taste makes constant use of i t .
Figure 405

g" ' r f
1

I76

I7

4f

44 '6

17
"6I
2
3

56

^ T

6 6
5

7
2

6 6
3 5

r'r
7

'

5b

--3

7 6

5b

Andante

T
a

fe 2r -3

rf f

56

F- ? r
2 3

327

rr Tr y i r T r * r r

0 M PA NI M E N T

r'
4

r
4

r'r r r r'r

"C

6
5

5b 6
3 5b

r'rr

4 3

-2- 3 5b

TTTT

frrrTf r T

0 0
6

9 8 9 8

9 8

^5,

7
5b

&

A C

J2$

C O M PA

ACCOMPANIMENT

EN T

NIM

3*9

3 T 6

4r

&

17
-2"

L J

6
4

6 5 7 5
4
# j|
-2-

H tt

jjj,j-,i.jji,j,,Hr v
j

6 5b 4

1
6 "6s

6 ^ 6

-2-3

. I

fe:

tf

ai v m -

6 -

-T 3

5 -2-3

i l i1

7
5

6
2 3

7
5

J ^E

9
4

5b

r r Tr
9
8

4
2

9. Other appoggiaturas, i n a d d i t i o n to those o n the second, cali

for consideration. I n the examples of Figure 406 the c h o r d of the
seventh is retarded by this ornament. I n a, the appoggiatura may
be i n c l u d e d i n the accompaniment as indicated by the figures above
the staff, w h i c h refer to the bass notes. B u t the setting that follows
may also be used. T h i s accompaniment, w h i c h holds the d i m i n i s h e d
fifth and t h i r d t h r o u g h o u t the bar, provides the p r i n c i p a l part w i t h
freedom to p e r f o r m its appoggiatura w i t h the appropriate affect.
6

I J J l.jjfl'W

l j J

-9-*

5b

'7
5b

Si
r

T h e same remarks apply to aa. I n b, 4 may be taken over the first

7

bass note and 5 over the second, or the appoggiatura and relase may
be o m i t t e d and, i f necessary, o n l y 4 3 played. T h e two-part appoggiaturas of c are accompanied as illustrated. I n a delicate setting the
7

6
-2-

6
4

5
3

8
5

5 - 6
3

Adagio

f
-

6
4

'P'r
5
3

7
5

appoggiaturas are replaced by a quarter rest and 5 b is played afterwards. Examples d and dd are alike, the difference between t h e m
b e i n g o n l y that dd contains a two-part appoggiatura. T h e accompaniments to b o t h are almost i d e n t i c a l . Rests are n o t used i n the
accompaniment to dd, because the notes of the o r i g i n a l are slow and
legato; b u t i n the more r a p i d tempo of d they are effective. Example e and its accompaniment are i d e n t i c a l .
10. I n the examples of Figure 407 the c h o r d of the second is
4

retarded by appoggiaturas. I n a the 3 is played over the e i g h t h rest

A C C O M P A N

33"
Figure 406
Andante 6

1 M li N T

ACCOMPANIMENT

7 6

s i x t h and second r e m a i n stationary. I n c, % is played, after w h i c h the

7

t h i r d moves t o a second w h i l e the other tones are h e l d . I n d, 4 is

2
played, f o l l o w e d by a sixth over the h e l d f o u r t h and second. I f the
seventh lies i n an inner part, the appoggiatura must be replaced by
a q u a r t e r rest. I n e only the seventh and fifth are taken and succeeded by the c h o r d of the second. T o indicate omission of the

J J

/, after w h i c h the fifth moves to the augmented f o u r t h w h i l e the

7
5b

6.
and ollowed by the major t r i a d on e. I n b, 5 is taken over the second

b7
5b

t h i r d , a slur can be placed over 5 . I n f, the preceding 5 is retained

and f o l l o w e d by a c h o r d of the second. I n g, the major s i x t h is
o m i t t e d because of the preceding m i n o r sixth, thus leaving o n l y

the fifth and second (2 ). T h e fifth succeeds to the augmented f o u r t h .

I n h, i t is best to double the t h i r d of the first c h o r d and lead the
4

6 7
4 3
Allegretto
o

7
5b

6
4

}}

"5

7
5b

b7
5b

6 5
4 i

J
8
6

7
5

6 5
4 It

W<
# W

5b

,J|J
6

44

7
5

*
6 5
4 | JJ

b7
5b

44

rttp-S

^
44 5b
1

6 5
#t

kd
?
6

7
5

44-

7
5

lower t h i r d to the f o u r t h of the succeeding 3 . I n i the c o m m o n

c h o r d or t r i a d i n three parts is played because of the preceding
three-part setting. Example / m i g h t very wel be accompanied i n
four parts by p l a y i n g the appoggiatura; b u t i n a delicate accompaniment, the p r i n c i p a l part must n o t be hampered at a fermata f r o m
resolving the ornament freely, according to the affect. Furthermore, the accompanist i n p l a y i n g the ornament exposes himself to
the risk of p e r f o r m i n g the relase before or after the p r i n c i p a l
part. W e have already seen i n the first part of this Essay that the
affect at fermate calis for great l i b e r t y of execution and that, consequently, appoggiaturas are often shortened t h r o u g h the i n t r o d u c t i o n of elaborations and decorations or lengthened and h e l d w i t h o u t f u r t h e r m a n i p u l a t i o n . I n b o t h events the accompanist, as a
precautionary measure, should l i m i t himself to a three-part accomp a n i m e n t or strike the bass note alone and play the c h o r d of the
second afterwards. I n k, where the same example appears w i t h twopart appoggiaturas, the bass alone is taken and f o l l o w e d by a slow
u p w a r d arpeggiation of the c h o r d of the second. I n l, the example
6

and its accompaniment are identical, b u t 5 may be taken as an accompaniment to the appoggiatura.
11. I n the examples of Figure 408 the c h o r d of the sixth is retarded by appoggiaturas. I n a, a f o u r - p a r t realization may duplicate

A C C O M /' A N I M l: N T

taz

k.

/7\

3 b

0-0-05
3

111

r r r r

4i

2T

44

--JX.

<7\

. S

c f r
6
5

/.j.

441

5 6
3 4

5
3

5
3

6 5
4 3

the bass of the t r i a d over e or, better, double its t h i r d . I n a threepart setting o n l y the fifth and t h i r d are taken; b u t i f only one p a r t
is played by the r i g h t h a n d i t should be the t h i r d , subsequently
held. I n aa w i t h its allegretto a n d p i a n o indications, the accompanist may employ either of the illustrated settings. I f the passage
is to be played l o u d l y , the appoggiatura a n d its relase may be i n cluded i n the first accompaniment. I n b only three parts at most
should be employed, f o r the undecorated chords r e q u i r e n o more.
I f the accompaniment must be soft the bass should be accompanied
by thirds alone; b u t the register of the upper p a r t must be watched
carefully i n order to prevent the fourths w h i c h i t forms against the
p r i n c i p a l p a r t f r o m becoming fifths. As indicated i n Example c a n d
its illustrated accompaniment, i t is a matter of opinin whether
a m n i m u m of three parts or f o u r are to be employed. T h r e e parts
are correct for d; b u t i f a delicate accompaniment is decided u p o n
for reasons similar to those addressed to Example j , Figure 407, the
t h i r d alone, subsequently held, should be realized over g-sharp. T h e
accompaniments to e and / are identical w i t h the originis. A very

A C C O MPAN

134

/ ; C o MPANIM

l M E N T

EN

*.l

l o u d accompaniment must be r e q u i r e d bclore a [ o u r t l i part is taken.

I n g the illustrated accompaniment may be taken i f the accompanist

decides against 3 as discussed i n Paragraph 5 .

Figure 408

Allegretto

9
7

i i
1

44
2>

8
6

44
3

8
3

4
44
3

'6
g- 6

44
3

12.

44
44
3

fe

t.s.

I n the examples of Figure 409 the t r i a d is retarded by ap7

44

5-6 2

'5 6|

- 4 i , JJ
16

, i J i ^ J Ji

i n b only 2 is realized, f o l l o w e d by 3 . Either of the illustrated accompaniments to c may be realized. B o t h have been discussed w i t h
their signatures i n Chapter V . T h e r e are five accompaniments to
d, of w h i c h the last t w o are the most delicate. T h e y have been p u r posely assembled here, even t h o u g h each one has already appeared
separately. N o t h i n g should be struck against the chromatic appoggiatura of Example e; henee the r i g h t hand, its parts replaced by a
quarter rest, does n o t play.
5

r T T f O rr

poggiaturas. I n a, 4 is taken above c and succeeded by the t r i a d . B u t

Figure 409

r,.
6

rrx,uf.r

II

C f , ir. [ r

44

*
' J
B - 4

|1

II

&

d.
j-

-H-n l

II
'6

3
9

C f . sections o n 4 a n d 4 c h o r d s .

Ti

// C C O M

17

1' A

A C C O M /' A N I M E N T

N I M E N T

le
6 -5
5

>
'99 8
6 5
4 3

4
2

Ir

3
3

te2

337

sixth taken according to circunistanccs. T h e appoggiatura in b is

best replaced by a rest. I n c selection of an accompaniment is dictated by the required loudness or softness of the setting. T h e first
of the illustrated accompaniments is best in the natated distribution of tones. T h e accompaniment to d is identical with the example itself.

5s
3

14. I n the examples of Figure 411 the chord of the major

seventh is retarded by appoggiaturas. A desire for orderliness has
8

rr

4
2

13. I n the examples of Figure 410 the six-five chord is retarded

by appoggiaturas. I n a, the ornament may be played or only the
Figure 410
a.

realized and followed by the seventh while the fourth and second
are retained. If it is decided not to play the appoggiatura (since it
Figure 411

-X

jj
7
4
3

r ' r ' i . r ' r ^ ^

F'i

> 4X

7
2

8
3

5b

r- r 1
a

-frpr
r i"r r " r
L

<>>!

brought about the reappearance of certain examples. I n a 4 is

7
4
2

8
5
3

ir. r f
4
2

5
3

d.

-*
i

6
5

It

IT

IT

r7
4
2

8
3

f.J.

4
2

5
3

ACCOMPANIMENT

ACCOMPANIMENT
.

339

forms an eirlpty octave), 2, but nothing less, must be taken. I n b,

7

4 alone may be played and held; but if the ornament is included it

must lie on top. A three- or four-part accompaniment may be taken
in c. If the latter is chosen, the notated distribution of tones is
best. T h e most appropriate accompaniment to all of the examples
under d is the appended one with its quarter rest.
15. I n the examples of Figure 412 the six-four chord is retarded
by appoggiaturas. I n a the accompanist may play the fourth and
fifth together or the fifth and the sixth and, in the latter case, resolve the fifth by stepwise descent. If he takes the fourth and fifth
the fifth must lie on top. T h e fourth alone is realized in a soft accompaniment. Example b and its accompaniment are identical. If
the chromatic appoggiatura i n c is played, it can be supported quite
well by the fourth. Example d and its accompaniment are identical.
I n e a triad accompanies the appoggiatura and is succeeded by a
chord of the sixth. T h e same accompaniment is applied to the twopart appoggiaturas of ee. I n / a complete chord of the seventh may
7

be played against the appoggiatura; also 3 or even the seventh alone.

Considerations of execution and affect should govern the choice
of a complete or incomplete setting. W h e n this example appears
with two-part appoggiaturas (ff), the realization should be similar
to the example itself or to the appended accompaniment. Examples
g and h are similar to / and ff.
Figure 412

6 5
4 3

5 6 5
3 4 3

6
4

- 5
- 3

ee. K
0

1.1' u
-

Hr

3 *

'

5
3

i1

' P

6b
4

b7
5b

b7
5b

7
5

i - i
-0000-

r
5
3

6
4

b7
5b

p ""

6
4

=Ft=l

fj)|Jj

L1LI P

b7
5b

11 3

-9

UXI

9-0-0

0 0 w

1^

6 b
4

l 90 -1

_?

^5-| 4

r-

P :

>0 0

^//if-'/ipVui/iJifi

4 3

6
4

5
3

p ' ' ' P

P

6
4

7
5

6
4

PP

h.
-#06
4

5
3

A C C O M r A N I M /; N T

/ i c ; O A/ /' A N 1 M E N T

f,

^4
_

7
5

6
4

7
5

6
4

Figure 413

.11

r.

44-

IFT- 11 j

1 iiJ1

11 J

1: T
5

aI - i

le

'r
1 J

r,
4-1
3

r
6

r
?
3

1-

j i

F=
6

a
r
44
3

6 * 1
6

4-1-

11 * a

1 jrj

^ 4 II J
II j | J 1
r L' r r. r t
3

su

16. T h e examples of Figure 413, excepting the last, illustrate

the four-three c h o r d retarded by appoggiaturas. I n a the accompanist may choose three or f o u r parts. I n the first of the i l l u s t r a t e d
accompaniments, the notated d i s t r i b u t i o n of tones is best. T h e sixfive chord should be retained i n b, after w h i c h the fifth succeeds
to the f o u r t h . B u t the appoggiatura may be o m i t t e d as i l l u s t r a t e d
i n the slightly varied example (bb). I n c the c h o r d of the seventh
is taken and its t h i r d h e l d w h i l e the seventh a n d fifth descend to
the sixth and raised f o u r t h . I n d, the accompanist may select the req u i r e d realization f r o m those i l l u s t r a t e d i n f o u r , three, and t w o
parts. A nine-four c h o r d is retarded by an appoggiatura i n e. Since
the ornament disagrees completely w i t h the figures i t is replaced
by a rest.

l - S U g !

34'

i i \\^~ i ^II 11 j 1 4 J 1 J=jj

Ti. t r r
r
1 j - i iliJ-ii
f.
"r Pr
p
f
1

>^

1 = 1

1 y

JI.

i -

r; r p r
11

"3*"}:

r^ T r
4

II

17. W h e n the p r i n c i p a l part of solos or other pieces r e q u i r i n g

a delicate accompaniment has many appoggiaturas i n a slow tempo,
the accompanist, i n order to avoid an obscuring of the melody,
should n o t play a l l of the ornaments. Those that cannot be readily
o m i t t e d should be m o d i f i e d by the i n t r o d u c t i o n of p a r d a l rests as
a means of d i f f e r e n t i a t i n g the accompaniment f r o m the solo. A
m o m e n t a r y w i t h h o l d i n g of the accompaniment gives the soloist an
o p p o r t u n i t y to i n t r o d u c e the appoggiaturas alone. T h i s m o d i f i cation, b r o u g h t about by rests, is increased i n effectiveness w h e n
the bass maintains a u n i f o r m pattern t h r o u g h o u t the passage. T h e
beauty and charm of appoggiaturas are most clearly perceptible
w h e n p e r f o r m e d i n this manner. Composers are w e l l aware of the
effectiveness of such an execution and often place rests i n the bass
at the entrance of appoggiaturas. B u t even w h e n they do n o t appear

A C C O M VA

N I M E N

in the left h a n d tlu-y may I)e interpolated

/'

A C C 0 M I' ANIMEN

in i h c right. Rests are

good i n F i g u r e \
Figure 414

l .

i J
4

m '
7
4

6
-

7
4

6
-

:-7,,,*,^,.*7,,,jr
7
4

ilu un

7
3

MU

V
4

6
5

gcr'mr

J
4 3

> i

7
4

A
6 6
4 5!>

4b 3

34)

18. I n the examples undcr a (Figure 415), w h i c l i are met occ a s i o n a l l y , the lirst o each p a i r ol eighth notes i n the bass should
be lengthened by the a d d i t i o n of a dot, as shown i n the illustrations w h i c h f o l l o w each passage. A literal performance of the appoggiaturas w o u l d prove that the example is w r o n g , its errors being
caused by ignorance or absent-mindedness o n the part of the composer. I f the ornaments were w r i t t e n o u t i n their correct vales, such
errors w o u l d never arise. Performed as notated, the ornaments
clash i n t o l e r a b l y w i t h the bass and i n so d o i n g lose their essential
charm. Even rests f o r the d u r a t i o n of the appoggiatura are i n the
m a i n n o t very h e l p f u l , for the r i g h t hand must re-enter o n the dissonant relase, since b o t h appoggiatura and relase disagree w i t h
the m o v i n g bass. T h e examples suggest no m i d d l e parts or, at most,
no n a t u r a l or good m i d d l e parts. T h i s is an unmistakable sign of
a poor or poorly conceived piece. Those w h o wish to t h i n k correctly
about composition must give simultaneous consideration to melody
and harmo ny. I t w o u l d be difficult to find examples that present
so many ready opportunities to w r i t e fifths. However, the a d d i t i o n
of dots to the bass makes the signatures and the accompaniment
n a t u r a l and simple. I n those cases where only one acceptable accomp a n i m e n t can be realized i t has been appended, b u t i t must never lie
above the p r i n c i p a l part. A t times the accompanist w i l l find himself
i n a situation where n o t h i n g may be altered. I f i n such a situation he
finds i t impossible to fashion an accompaniment, he must resort to
tasto solo. I n b, i f an accompaniment were realized f r o m the figures,
w h i c h u n f o r t u n a t e l y are f o u n d far too often, i t w o u l d sound exceedingly ugly. I n the appended accompaniment the correct figures
are given. I n the bass of c the first eighth note of each bar can be
easily replaced by a rest as a means of a v o i d i n g the miserable accented fifths." Passages l i k e this can be f o u n d i n l i g h t , present-day
I t a l i a n works. Experienced accompanists w h o can and dar make
m i n o r extemporaneous corrections i n a composition should receive
f u l l credit for their deeds, b u t this should n o t lessen the composer's
responsibility for such blemishes. I n o u r example i t is advisable that
b o t h hands pause for the d u r a t i o n of the appoggiatura. I n d the acc o m p a n i m e n t is similar to the o r i g i n a l . T h e dissonances are passing
and the r i g h t h a n d should be an exact d u p l i c a t i o n of the m o t i o n
> Perhaps Bach's o w n direct fifths of similar type are misprints! Cf. Prussian
Sonatas, no. i , last movt., bars 5 - 8 8 , 75-76 (Nagels A r c h i v ho. 6).

I < < O M VA N 1 M E N T

J44

/ c; r; o

ol (he p r i n c i p a l p a n . A composcr coulcl only he madc unliappy by a

I t r i c t resolution of the dissonances or the slightesl alteration of the
sonority or m o t i o n of the parts. T h e last two bass notes carry f o u r
parts. I n e the o r i g i n a l may be duplicated or the r i g h t h a n d may
rest. I n / a n d ff the accompaniment must n o t cross the p r i n c i p a l
part.

-4

M v A NI M EN

345

9 80 5 6 6

r f ^ . V .

>

Figure 415

6 6

6 6

- 4
JJ,

4 3-6. 7 6 6
5t

6 5l>

"5

_==

"p

p
6

9 8 6 6

5
3

4 3

J *r

&

-91

gj-^

7
5
r

8
6

6
4

5 6
3 4

rJ

7
5

6^6

5
3

j iVJiiJflJBHiJi

m t ^ ^

Z~

6 5
9 8 7 2 3 6

e*^ti

u f i

6 6

5!

J J. .
5 5

6 6

7 6 6 7 6 6 7 6

p T

tr=f

11

4
2

IK

9
7

V
10

8
6

^1

6
4

5
3

r
2

6
4

fl

A C C OMPAN1

M / N T

A C C O M r A NI M E N T
Figure

b.

a.
-

6
4

*.

*00 00

S 6

6 4

t4r
7

<aa

i .

10
7

> "p 'P

6

ft

S
6

/y.
5

|
6

IS

'7

|4

II

a
|5 6

5b

'

*t *

f*

tus

~
6

7
6

5l>

i .
H

hjn
15

H-p^

I?

9
7

" " T T T ^

>4r

6
4

0000

347

416

-6-

1
4

|5

|4

2
b7

19. T h e short, invariable appoggiatura is n o t realized. I n fact

i t calis for no m o d i f i c a t i o n of the accompaniment. However, a few
examples (Figure 416) i n a slow tempo are illustrated here w h i c h
cali for certain precautions. I n a over the second g-sharp o n l y the
d i m i n i s h e d fifth and t h i r d may be played, n o t the seven-five c h o r d
or the c h o r d of the s i x t h . I n b and c rests are good. T h e y b r i n g
about a r h y t h m i c m o d i f i c a t i o n and help to clarify the appoggiatura
i n b. I n c they are r e q u i r e d as a means of a v o i d i n g ugly accented
fifths between the t h i r d above the bass and the p r i n c i p a l part. I n
d also, the rests lessen the harshness of the chords w h i c h result f r o m
the successive appoggiaturas. I n dd, w i t h its two-part appoggiaturas,
the r i g h t h a n d should n o t play at a l l , for i t is better to o m i t chords

T Note that none of these is w r i t t e n i n small n o t a t i o n .

r r r r

p
5b

348

( :

( :

O M PA

A C C O M P A N I M E N T

N 1M E N T

349

solo. E x a m p l e d i n a r a p i d l e m p o w i t l i o u t the s u p p o r t i n g thirds is

a c c o m p a n i e d as i l l u s t r a t e d i n e.
Figure 417

FTTTT

6
5

98

JW>J JM
than to strike disagreeable ones. Rests are needed i n e f o r the reasons stated i n b.
20. Unless an appoggiatura i n the bass is set w i t h its o w n signature, i t is accompanied by the c h o r d w h i c h belongs to the relase.
SYNCOPATED

NOTES

1. C h o r d a l tones are either anticipated or retarded by syncopations.

2. Slow syncopations w h i c h anticipate chordal tones r e q u i r e n o
modification i n the accompaniment. T h e accompanist strikes the
indicated c h o r d over its proper bass note (Figure 417, Example a).
But w h e n a chordal tone is retarded by syncopation the accompaniment is fashioned i n the manner discussed i n connection w i t h appoggiaturas. T h e r e t a r d i n g tone may be played or o m i t t e d , the
harmony may be reduced to those figures w h i c h agree w i t h the ret a r d i n g and retarded tones (b), rests may be inserted (c), or the
syncopations may be played i n t h e i r entirety (d). I n c the r i g h t
h a n d must leave the keys i m m e d i a t e l y on the entrance of d-sharp
i n the p r i n c i p a l part. I f Example d is slow a n d supported by thirds
(dd) the three-part accompaniment is identical w i t h dd. B u t i n other
than a slow or, at most, modrate tempo i t is accompanied tasto
2

1 Cf. A r n o l d , op. cit., p p . 431 ff.

2 Die Rckungen.
T h e term has no unequivocal English parallel. Syncopations
is satisfactory provided that the term is stripped of the narrow meaning that i t derives
f r o m its use i n strict counterpoint, and understood only i n the sense of r h y t h m i c a l l y
shifted notes. Arnold's translation, " D r i v i n g Notes," is an od English term. Cf.
A r n o l d , op. cit., p p . 127 ff., Note g.

m
6

M>J J J

frrrrrrr

rrrr

M TT TTrTTt rrf
5
fe

43
98

3 2
4 b-

1.

5t| 6
7 6 '

9 8 7
7 6 5

'

Is

l rf r

6-5
4 - 3

3. R a p i d syncopations are never played as such by the accompanist b u t are supported by chords c o n t a i n i n g anticipations or retardations according to the n a t u r e of the m a n i p u l a t i o n . Regardless

A C C O M I' A N 1 M /<, N T

A C C O M I' A N I M E N T

of wliether they lie i n the p r i n c i p a l part or iti the bass, the accompanist plays i n an even r h y t h m as, for example, i n Figui e.418, where
each of his chords has the valu of a quarter note. T h u s w h e n the
bass is syncopated the r i g h t h a n d holds to the r h y t h m of the bar (a).
Figure

418

jLjqi

a.

pr p 1 r
J I

6
5
5 b 2

4. T h e accompaniment to chromatic syncopations must be delicate i n order to b r i n g t h e m i n t o relief and avoid ugly clashes. I n
Figure 419, Example a, the t r i a d w i t h o u t the octave d u p l i c a t i o n is
taken. I n b the accompaniment consists of a nice i m i t a t i o n of the
chromatic tones (bb). Should a f u l l e r setting be r e q u i r e d , the intervals of the p r i n c i p a l part may be i n c l u d e d as indicated i n the signatures of bb. A l l of these examples presuppose a slow or, at most, a
modrate tempo.
Figure 419 k

mn 1
-JT TI
pr^ ^p 1
r
^1

i>

y
b7

1|

5b

_
1/

bb. 1
=
#

DOTTIil) COMI'OUNI) AIM'OGCIATURA

1. T h i s section cannot be read proitably unless the reader is

f a m i l i a r w i t h the earlier discussion w h i c h appears i n Part I of this
Essay. I f he is not, m u c h of the present treatment w i l l be incomprehensible and most of i t incorrectly understood. Once the nature
of the ornament is k n o w n i t can be seen that i t has an i m p o r t a n t
bearing o n h a r m o n y .
2

Allegro

THE

93*

i!

jn

5-6

~JTp7- 1

FJ
5

T
5>
3

5
2

^
3

"5- 3 - 2 -

2. T h e d o t t e d c o m p o u n d appoggiatura appears only i n pieces

that are dependent o n taste and affect, i n w h i c h the accompaniment
must be especially delicate. T h e proper chord of a bass note is more
retarded by this ornament than by the appoggiatura, for i n execut i o n the p r i n c i p a l tone of the p r i n c i p a l part does n o t enter u n t i l the
last short note of the ornament has been played. W i t h respect to
loudness and softness, the performance of o u r embellishment is the
same as that of the simple appoggiatura; the retarded p r i n c i p a l
tone is played softly and the r e t a r d i n g tone l o u d l y . I t w o u l d seem
that these factors w o u l d certainly lead to an exact figuring of this
o r i g i n a l l y vocal embellishment. B u t u n f o r t u n a t e l y the same comp l a i n t must be raised here as i n the case of the appoggiatura. U p to
now, figurists have n o t treated the ornament w i t h due cre.
3. I n accompanying a bass note over w h i c h a dotted c o m p o u n d
appoggiatura appears, the same expedients are necessary as were
illustrated i n connection w i t h the appoggiatura. T h e indicated harmony may be changed, reduced, or at times even o m i t t e d . W h e n the
ornament appears i n succession w i t h only a few chords, rests are n o t
always good except i n the slowest tempo, for the numerous d i v i d e d
beats of the r i g h t h a n d can easily d i s t u r b the sustained melody.
4. Chords that are struck o n d i v i d e d beats usually enter o n the
second half of the bass note. B u t i f the latter is of great length, its
second half is subdivided and the delayed c h o r d enters o n the last
quarter of its total length.
5. I n the examples of Figure 420 a second above the bass is retarded by o u r ornament. A g a i n i n this section the usual figures
w h i c h appear w i t h each example have been c o m p u t e d w i t h o u t
reference to the ornament; b u t the correct signatures appear i n the
illustrations of the accompaniments. I n a the r i g h t hand pauses o n
1 Cf. A r n o l d , op. cit., p p . 433 ff.
2 Cf. Ch. I I , " T h e Compound Appoggiatura," f 7 ff.

A C C O M l' A N I M E N

A C C O M I' A N I M E N T

/'

the nrst quarter over b, and the six-live chord eniets later. T h e correct divisin or the o r n a m e n t can be seen i n a and its usual notat i o n i n x. Example b has the same accompaniment. I n c the seventh
alone is played o n the entrance of b i n te bass, and the t h i r d
comes i n later. I n d the seventh and d i m i n i s h e d fifth are played first
and the six-five chord is taken afterwards. I n e, i f a t h i n accompanim e n t is called for, the seventh alone is played first and the t n i r d
afterwards. B u t i f a f u l l e r c h o r d is r e q u i r e d , the second is struck
w i t h the seventh. These remarks apply to a l l similar cases. I n f, the
d i m i n i s h e d f i f t h is played alone, the second being i n c l u d e d i f necessary; the s i x t h is o m i t t e d . I n g the r i g h t h a n d pauses and then plays
the c h o r d of the s i x t h . I n h the sixth is played and, i f necessary, the
second. I n i the chord of the n i n t h and ii the nine-four c h o r d are
struck and f o l l o w e d by their usual resolutions. I n a l i g h t accomp a n i m e n t b o t h chords may be o m i t t e d and triads played o n the
second quarter. I n / and k the fifth, alone or i n company w i t h the
second, may be taken. I n a d d i t i o n , k may be accompanied by a passi n g c h o r d of the second f o l l o w e d by a t r i a d i n three or f o u r parts,
whichever is appropriate (x). I n
after an eighth rest, the fifth
enters alone f o l l o w e d by the t h i r d . I n m there are three possible accompaniments: T h e seventh may be played alone and succeeded by
the augmented s i x t h and the t h i r d ; or the second may be struck w i t h
the seventh; or 3 may be taken after an eighth rest. A l l three are
good depending o n the r e q u i r e d v o l u m e of the performance and
6

the affect. I n n i t is best to pause and then play 3. I n o the s i x t h

and the augmented f o u r t h are taken and the t h i r d is o m i t t e d . I n p
rests i n the upper part clarify the ornament. T h e seventh may enter
immediately. I n q the fifth, alone or w i t h the t h i r d , may be struck,
depending o n circumstances. I n r the seventh and f o u r t h are played
a n d succeeded by their resolutions.

(X)

r
I

16

16

'7 6
5b-

r'

r r t
5b

7 6
5b

fi

.2-3

m 1?

>/J|^J

'5

5b

5 b

I
t.s.

6
4

'7 6

T rrrr

-2-3

16

11

16

J rH
6
Ifi

T9
'o

16
JSf

8
fl

-T

r r r : r r
56

'6
5b

9 8 8
4
3

J IJ I >

I -PJ
f

5
R
3

5 6

6
c
4

5
3

T
6

554

A C COMPANIM

/ ; C O M I' A N I M E N 7

E N T

I ^ 4

JCZ*

r
-

7J|J^.J i i j | ^ J ^ J

r rr

ii

r r r f

/.i.

rr
7

7 - 6

6
44
44

/\-w

I
4

8
3

Tfr

77 - 6 .
4
3

6. T h e r e are other retardations i n a d d i t i o n to those o n the

second. I n the examples of Figure 421 the chord of the seventh is
delayed by o u r ornament. E i t h e r one of the t w o accompaniments
under a may be selected. T h e first may be realized i n f o u r parts.
Example aa, w h i c h consists of r a p i d notes and does n o t portray any
affect, may be accompanied i n f o u r parts. I n Example b the openi n g tone of the ornament increases the harshness of a simultaneously
realized d i m i n i s h e d seventh. Henee, i t is better to i n t r o d u c e an
b7

=g

eighth rest a n d strike 5 \ afterwards i n three parts. I n c the illustrated accompaniment may be used or, i n its place, an eighth rest

'6

re rr U.

r r n r
2- 3

i g i

rr

J,

% | i

T I

755

f o l l o w e d by 5 b w i t h o u t the t h i r d . T h e last t w o examples u n d e r d

sound ugly, a l t h o u g h they w i l l be encountered. O u r ornament,
here, defeats its purpose, w h i c h is to p rovi de a pleasant, caressing
q u a l i t y . T h e passage sounds better w i t h o u t i t , and i f there is need
for an embellishment at a l l i t should be the r a p i d c o m p o u n d appoggiatura. I t is best to accompany our o r n a m e n t w i t h an eighth
b7

rest f o l l o w e d by 5 \ i n three parts. As for the first example (d), the

t r i a d may also be played i n three parts. Example e is not good, for
the ornament makes the passage sound l i k e a d u l l succession of
sixths and almost obliterates the seventh f r o m w h i c h the progression gains its attractiveness. T h e t h i r d should be taken on the entrance of the ornament, and succeeded by the sixth and t h i r d o n the
last quarter. Example /, w i t h its preceding d i m i n i s h e d fifth, is
better. A three-part five-three chord is played on the first quarter
and f o l l o w e d by the six-five chord. Either one of the accompani-

A C ; O M /* A N I M /< N 7

57

ments to g may be useil; or as an altcrnative, pause on the lirst

7

eighth and on the second take r> or ;i i n three parts. T h e reason for
the omission of an i n t e r v a l f r o m the chord of the seventh is the l i g h t
manner i n w h i c h the last note of the ornament and the p r i n c i p a l
tone are p e r f o r m e d .
7. I n Figure 422, Example a, the c h o r d of the second is retarded
by o u r ornament, w h i c h o w i n g to its length makes the passage rather
unattractive. A shorter c o m p o u n d appoggiatura w o u l d be better.
T h e t r i a d is taken o n the / and f o l l o w e d by the chord of the second,
b o t h preferably i n three parts. I n b, c, and d the chord of the s i x t h
is retarded. I n b the t r i a d is played and f o l l o w e d by the c h o r d of
the s i x t h . O r the t h i r d alone may be taken over f, followed by the
sixth. I n c i f the accompanist does not wish to use an eighth rest,
he may play either of the illustrated accompaniments. T h e accomp a n i m e n t to d is l i k e that to b. T h e examples under dd, i n w h i c h a
three-part eight-six c h o r d is retarded, are a l l accompanied i n the
manner of the appended i l l u s t r a t i o n . I n the r e m a i n i n g examples
7 8

the t r i a d is retarded. I n e the accompaniment may be 2 3 i n three

parts, or an eighth rest f o l l o w e d by a t r i a d . I n / the accompanist
7 8.

takes 4 3 i n three parts, or pauses for an eighth before p l a y i n g the

t r i a d . I n g the chord of the n i n t h and its resolution are played. B u t
i n h the nine-four c h o r d is r e q u i r e d , f o l l o w e d by its resolution. I n i,
T

5 and then 3 are played.

Figure 422

? r r r f\H| rrifirl
6

6
4

6
6

54
5 42

6
6

5 6 6

Ah4
6

6 '6
5

5 6 ^

>4

le

J5<V

A C C O Ai

P /l

e; c o

/V / A i fi JV 7
Figure 42}

prrpp^ i
r

'5

r i mi T

r
-

7 78
5

7
5

t.s.

6
4

-6.
5b

r r r f r p-m
1

17 8
2 3

'6 5 b

8
3
8

lf

Ijt

-^

5
3

5
3

L r ' T r "r'r p T r

F F 'r-'ij 'r
1i 4M

159

5 b

- - 7
7
5

gEg

A/ /*,/ N I A /; N T

5
3

Ai.

5
3

8. I n Figure 423, Examples a and b, the six-five c h o r d is retarded

by o u r ornament. A n y of the appended accompaniments to each
example may be used. I n the r e m a i n i n g examples the c h o r d of the
major seventh is retarded. T h e r a p i d c o m p o u n d appoggiatura is
better than o u r dotted k i n d i n c. T h e latter sounds empty because
of the l o n g retardation o n the octave. I t is u p to the accompaniment (either of the appended ones may be used) to restore the i n -

6
4

le"

8
5
3

11

5
3
--

- -

r r
#

5
3

ni J

Ir 'IT
4

y.^y
*4r

-6- - 6
4 5b
3
-

i
4

J J 11 r
rr r r r r

5b

r4

1-

11

^=F

ACCOMPAN!MENT

6o

A C C O M

tended relationships. 1 he accompaninients to d and e are i d e n t i c a l :

they may be i n three or f o u r parts. I n / the four-part chord of the
major seventh is taken o n the entrance of the ornament. I n order
to seprate the sixth f r o m the seventh, the p r i n c i p a l part must lie
above the accompaniment. I n g and h the r a p i d c o m p o u n d appoggiatura w o u l d be better than o u r dotted k i n d . I n g the ear demands
a c h o r d directly o n the entrance of the ornament, b u t i n h a quarter
rest is i n order.
9.
I n the examples of Figure 424 the six-four chord is retarded
by o u r ornament. I n a the accompanist may choose any of three accompaniments. T h e first t w o may be realized i n three or f o u r parts,
but the t h i r d , w h i c h is the lightest, should be left i n t w o . T h e first
cannot be used readily i n any b u t the illustrated disposition. I f an
Figure 424

'5

f
4 - 3

N T

361

o p t i o n a l ; over the note c, 4 or 4 may be taken. T h e same applies to

c. I n b and c the nine-four c h o r d and i n d the nine-seven are retarded. T h e accompaniment to the final example may be i n three
or f o u r parts. A n eighth rest is i n t r o d u c e d w i t h the ornament, a n d
succeeded by nine-seven.
425

9,

r r r"'' i r " ?'

i Ai
1
1

/',

10. I n Figure 425, Examples a and b, the five-four c h o r d is retarded. Example a is n o t good, for o u r l o n g ornament destroys the
beauty of the i n t e n d e d dissonance. A short c o m p o u n d appoggiatura
w o u l d be better. T h e accompaniment must r e m a i n i n the notated
disposition; otherwise an eighth rest must be i n t r o d u c e d over g,
f o l l o w e d by a complete five-four chord. I n b the accompaniment is

' r l ^ l / l f l K ' l .
5

I N I M

eighth rest over the second r; of Kxample b is not used, the appended
accompaniment should be taken. The rest is r e q u i r e d i n c. T h e acc o m p a n i m e n t to d must r e m a i n i n the notated disposition; otherwise, an eighth rest accompanies the entrance of the ornament. A
rest is r e q u i r e d i n e and /. B u t i f i n d, e, and / the t r i a d on c replaces
the i n i t i a l c h o r d on /-sharp, the illustrated accompaniments are to
be retained.

Figure

1'

'4

'3

r4

'5

d.

F F
1

t.s.

r"
5

\i

4 3

if

^iJ

16

15

6
4

5
3

7
5

6 5
4 3

r~ r r *r r r
'7
5

H? f
T

|7
5

~m

'7
5

6
4

'

rr
6
4

9
4

t.s.

5
3

8
3

rr 1 r
g

'9

68

75

=p

"*

6
6

5
5
t

16

1v
r^f
t.s.

"
9
7

7
5

)2

ACCOMPANIMENT

A C C OM P A N IM E N T

3<)

Figure 426
THE

DOTTED

SLIDE

mm i

1. A l l that has been said i n the first t w o paragraphs of the preceding section about p r i o r knowledge of the d o t t e d c o m p o u n d appoggiatura as discussed i n the first part of this Essay, the i m p o r t a n t
bearing that the ornament has o n harmony, a n d the consequent
need f o r a clear i n d i c a t i o n of its presence, applies equally to the
dotted slide.
2

2. A l t h o u g h the d o t t e d slide does n o t appear as o f t e n as the t w o

ornaments that have already been discussed, chords are sometimes
more retarded by i t t h a n by the others. T h e affect w i t h w h i c h the
slide is sometimes p e r f o r m e d requires that the i n i t i a l tone be h e l d
beyond its usual d u r a t i o n , w i t h the result that the entrance of the
succeeding c h o r d must be delayed by half its notated l e n g t h a n d
the preceding c h o r d correspondingly p r o l o n g e d . I n F i g u r e 161
there are several examples that have various executions. I n the f o l l o w i n g examples, the bearing of these variants o n the accompanim e n t w i l l be illustrated.
3. I n the f o l l o w i n g cases the accompanist realizes the prescribed
signatures, even t h o u g h they are c o m p u t e d w i t h o u t reference to the
ornament. B u t certain executions of the o r n a m e n t r e q u i r e slight
changes i n the accompaniment. These have been noted i n the examples. I n F i g u r e 426, Example a, the accompaniment is n o t
changed regardless of the l e n g t h of the slide. B u t i n b the accomp a n i m e n t w h i c h is appended to each example should be f o l l o w e d i f
the first note of the ornament is h e l d t h r o u g h part of the succeeding
bass note. A l t h o u g h b(x) w i l l be encountered occasionally, the prolonged f o u r t h i n the slide does n o t sound w e l l . I f o r n a m e n t and
p r i n c i p a l tone together have a valu no greater t h a n a quarter note,
or i f the d o t t e d d is retained over the last bass note i n the bar, the
effect may be acceptable. B u t i t is n o t good w h e n the divisin is
such that the / i n the bass a n d p r i n c i p a l part are played s i m u l taneously. T h e empty octave after an i n s i p i d f o u r t h makes an awkw a r d progression. Regardless of the execution of this example, the
accompanist takes the sixth a n d t h i r d i n three parts or the f o u r t h
and t h i r d similarly, as illustrated. T o do justice to b(y) the e m p t y
octave should n o t be h e l d too l o n g ; b u t i f i t is, a n d the / enters over
c i n the bass, 4 3 is played instead of the t r i a d .
1 Cf. Ch. V I , " T h e D o t t e d C o m p o u n d A p p o g g i a t u r a , " N o t e 1.
2 Cf. Ch. I I , " T h e Slide,"
10 ff.

6
5

4 8

a.
wm

4
6 - 4 6
2 5

4
2

b.

"i'iu'i; ! "
1

'6
6

4 3

rr r rJ J J f l l f l f l f l U l J
6
4

5
3

11

7
4
2

r rr
4
2

5
3

7
4
2

8
5
3

8
5
3

7
4
2

J I
-

r
7
5

8
5
3

)6

(1

C O M V A N IM E N T

h.

1 i -=t=

U- 7

m
y

8
4 3

S-

T=F=\ -H
J

"

V:

r r"
'9
4

8
3

T=|= i

ii

-s

9
4

6
5

11

le le

- 3

43

4
3

'(

6
i

&m
= F

7 6

played, because of the sixth 011 which the slide commences. B u t ,

since only the t r i a d and no other chord its the bass, the accompanist must take the one i n l e i v a l , aside f r o m the octave, w h i c h the
c h o r d of the sixth and the t r i a d have i n c o m m o n , namely, the t h i r d ;
or he may play the bass alone. W h e n a piece begins i n this manner,
i t is best n o t to play the bass to the ornament. I f the slide i n e is of
n o r m a l l e n g t h , that is, i f g i n the p r i n c i p a l part enters o n the last
b i n the bass, the complete six-five chord should be repeated o n the
last quarter. B u t i f the i n i t i a l tone of the o r n a m e n t is retained over
this b, the accompaniment must be fashioned i n the manner of the
appended i l l u s t r a t i o n . B o t h examples under / take the accompanim e n t w h i c h has been placed between t h e m . I n g the six-four c h o r d
must enter o n the second quarter of b o t h examples. I n h, the fourthree c h o r d is played and i n i, a three-part 4 3 . I f , i n this last example, the first note of the slide is h e l d beyond its n o r m a l l e n g t h ,
the t h i r d should be played o n the last eighth and the sixth repeated at the same p o i n t . T h i s is shown i n the final i l l u s t r a t i o n .
Figure 427

n'

363

1 ,lili 1
1r r r

f
4
2

16

P.

le
1

^ H
H fi -

ij

a -1 1
rj F-t c r 1 f
*fc

"1

7 8
4 3
2

*(*>!_

A C C O M I' A N I M E N I

*
4

9 8
4 3

4. I n the examples of F i g u r e 427 the accompaniment w o u l d

sound ugly i f i t were constructed i n agreement w i t h the usual
signatures; henee the figures must be changed as i n d i c a t e d i n the
illustrations appended to each example. T h e usual signatures appear w i t h the examples. A d d i t i o n a l modifications, w h i c h are made
necessary by an e x t r a o r d i n a r i l y slow performance of the ornament,
are shown i n the last of each series of i l l u s t r a t i o n s . I n b the slide
mollifies the harshnes of a simultaneously realized seventh. Even
t h o u g h this tone appears i n the preceding c h o r d , i t is w e l l to o m i t
i t f r o m the accompaniment to the o r n a m e n t a n d to play only the
d i m i n i s h e d fifth a n d the t h i r d t h r o u g h o u t the bar. T h e examples
under c have i d e n t i c a l accompaniments. I n d the t r i a d cannot be

S-b

6 5

66

ACCO

1JU

r *r

A C C O M P A N I M EN

MPANIMEN7

1 r

r i

967

an accompaniment. This l a t t t i lias inore eletnents w h i c h are concerned w i t h the rules o! good performance t h a n solos have, for an
accompanist is responsible for more than a correct realization of a
bass; he must make i n t e l l i g e n t adjustments w i t h respect to the
v o l u m e and register of chords. As stated i n Paragraph 1 9 of the I n t r o d u c t i o n to Part T w o , i t is r e q u i r e d of the accompanist that he
fit to each piece a correct performance of its h a r m o n y i n the proper
v o l u m e a n d w i t h a suitable d i s t r i b u t i o n of tones.

r, r cr cccc ce Ct
5

8
6

'-7
-5

2. T h e fewer the parts i n a piece, the finer must be its accomp a n i m e n t . Henee, a solo or an aria provides the best o p p o r t u n i t y to
judge an accompanist. H e must take great pains to catch i n his acc o m p a n i m e n t a l l of the nuances of the p r i n c i p a l part. Indeed, i t is
difficult to say whether accompanist or soloist deserves greater
credit. T h e latter may have taken a l o n g t i m e to prepare his piece,
w h i c h , after the present fashion, he himself must compose. Nevertheless, he cannot c o u n t o n the applause of his audience, f o r i t is
o n l y t h r o u g h a good accompaniment that his performance w i l l be
b r o u g h t to l i f e . O n the other hand, the accompanist is usually
given m u c h less t i m e ; he is a l l o w e d only a cursory e x a m i n a t i o n of
the piece, b u t must nevertheless support a n d enhance extemporaneously a l l the beauty o n w h i c h so m u c h t i m e and care have
been expended by the p r i n c i p a l p e r f o r m e r . Yet the soloist takes a l l
bravos to himself and gives n o credit to his accompanist. B u t he is
r i g h t , f o r he knows that i g n o r a n t custom direets these bravos to h i m
alone.
2

3. G r a t u i t o u s passage w o r k and b u s t l i n g noise do n o t constitute

the beauties of accompaniment. I n fact, they can easily do h a r m to
the p r i n c i p a l p a r t by r o b b i n g i t of its freedom to i n t r o d u c e variations i n t o repetitions a n d elsewhere. T h e accompanist w i l l achieve
eminence a n d attract the a t t e n t i o n of i n t e l l i g e n t listeners by lett i n g t h e m hear an u n a d o r n e d steadiness and noble simplicity i n a
f l o w i n g accompaniment w h i c h does n o t interfere w i t h the b r i l l i a n c e
PERFORMANCE

1. I t is w r o n g t o believe that the rules of good performance

p e r t a i n o n l y to the p l a y i n g of solos. T h e m a t e r i a l o n performance
w h i c h is contained i n the first p a r t of this Essay (to w h i c h the reader
is referred) must also be observed i n certain respects i n fashioning
1

i Handsachen.

Cf. Pt. I , Foreword, Note 2.

2 Franrois Couperin i n L'Art de toucher le clavecn (pp. 44-45) expresses himself

just as b i t t e r l y . " I f i t were a question of choosing to b r i n g to perfection either the
p l a y i n g of pieces or accompaniments, I suspect that vanity w o u l d lead me to prefer
pieces. I a d m i t that n o t h i n g is more d i v e r t i n g by itself, or brings us more i n t o
the company of others t h a n to be a good accompanist. B u t what injustice! W e are
the last to be praised at concerts. O n such occasions the accompanist is regarded merely
as the f o u n d a t i o n of an edifice w h i c h , a l t h o u g h i t supports everything, is almost
never mentioned. O n the other h a n d , those w h o excel i n pieces revel i n the a t t e n t i o n
and applause of the audience."

368

ACCOMPANIMENT

of the p r i n c i p a l par. 1 le need lee 1 no anxiety over 11 is being forgotten i f he is not constanily j o i n i n g i n tlie t u m u l t . N o l A n understandi n g listener does not easily miss a n y t h i n g . I n his soul's perception
melody and harmony are inseparable. Yet, should the o p p o r t u n i t y
arise and the nature of a piece p e r m i t i t , w h e n the p r i n c i p a l part
pauses or performs p l a i n notes the accompanist may open the d r a f t
o n his damped fire. B u t this demands great a b i l i t y and an understanding of the t r u e content of a piece. I n t r u t h , i t suffices to realize
an accompaniment w h i c h does n o t h i n g more than meet the requirements, whether expressly called for or not, of Paragraph 19
of the I n t r o d u c t i o n to Part I I . T o this end we shall proceed to relate
these precepts, along w i t h questions of pur style, to the construct i o n of a fine accompaniment.
4. I t is sometimes necessary and n o t really i m p r o p e r for the
accompanist to discuss a piece w i t h the performer of the p r i n c i p a l
part before its performance and let h i m decide o n the liberties that
are to be taken i n the accompaniment. Some w a n t the accompanist
to be greatly restricted; others not. Since opinin varies so greatly
and i t is u p to the p r i n c i p a l p a r t to decide, the safest procedure is
to seek a p r e l i m i n a r y understanding.
5. W e shall open the subject of performance by discussing
v o l u m e . O f a l l the instruments that are used i n the p l a y i n g of
t h o r o u g h bass the single-manual harpsichord is the most p e r p l e x i n g
w i t h regard to forte and p i a n o . T o make amends for the imperfect i o n of the i n s t r u m e n t i n this respect the n u m b e r of parts must be
increased or reduced. B u t care must be exercised to i n c l u d e a l l
necessary tones and avoid incorrect doublings. Some resort to a
h i g h l y detached touch i n order to express a piano, b u t the performance suffers tremendously by this; and even the most detached
staccato performance requires pressure. I t is better to reduce the
v o l u m e by using the r i g h t h a n d less f r e q u e n t l y over passing tones.
T h e fine i n v e n t i o n of o u r celebrated H o l e f e l d w h i c h makes i t possible to increase or decrease the registration by means of pedis,
w h i l e playing, has made the harpsichord, p a r t i c u l a r l y the singlem a n u a l k i n d , a m u c h - i m p r o v e d i n s t r u m e n t , and, f o r t u n a t e l y ,
3

s M a r p u r g (Der Critischer Musicus an der Spree, 26 August, 1749) writes, "Clever

artists . . . know how to deceive the ear at the harpsichord i n such a manner that we
believe that we hear soft and l o u d tones, a l t h o u g h the quills deliver a l l w i t h almost
equal forc."
Si'c for H o h l f e l d . Cf. Part I I , I n t r o d u c t i o n , f 2 a n d Note 1.

A C C O M P A N I M E N T

369

elirninated all dillicullies connecled w i t h the performance of a

piano, l only all harpsichords were s i m i l a r l y constructed as a
t r i b u t e to good tastel
6. B u t aside f r o m this i n v e n t i o n , the clavichord and pianoforte
enjoy great advantages over the harpsichord and organ because of
the many ways i n w h i c h t h e i r v o l u m e can be gradually changed.
T h e pedal o n the last-named i n s t r u m e n t does commendable service
w h e n the bass is not too r a p i d ; and i t can be made more penetrati n g by means of a sixteen-foot registration. However, rather than
mutlate the melody of the bass, the pedal should be o m i t t e d w h e n
n o t a l l of its notes can be played by the feet, and the lowest p a r t
played solely by the left h a n d .
7. T h e r e are certain general rules that govern the performance
of forte and p i a n o o n the organ and the two-manual harpsichord:
fortissimo and forte are played o n the louder m a n u a l . Fortissimo
may be attained by d u p l i c a t i n g i n the left h a n d a l l tones of consonant chords, and the consonant tones of dissonant chords w h e n
the nature of the bass makes i t possible to d o so. T h e l o w register
must be avoided, the d o u b l e d tones b e i n g placed cise to the r i g h t
h a n d i n such a manner that the notes of b o t h hands a d j o i n , leaving
no i n t e r v e n i n g space. Otherwise, the r u m b l i n g l o w notes w i l l crate
a miserable b l u r . A simple octave d o u b l i n g of the bass by the left
h a n d also has a penetrating effect; i t is indispensable w h e n the
notes are n o t very r a p i d a n d are easily played, b u t yet express a
well-defined theme w i t h a f a i r l y w i d e range. These octave doublings
are very good for i m i t a t i o n s w h i c h are to be l o u d l y p e r f o r m e d or
for the entrance of fugal subjeets. B u t w h e n a subject or any passage of significance contains lively figuration w h i c h cannot be easily
executed by one h a n d i n octaves, at least the p r i n c i p a l tones should
be d o u b l e d a n d the others played simply (Figure 428). T h e r i g h t
h a n d contines w i t h its chords, w h i c h cannot be readily o m i t t e d
f r o m c o n t r a p u n t a l works. I n a mezzo forte the left h a n d may play
the bass as w r i t t e n o n the louder m a n u a l w h i l e the r i g h t accompanies o n the softer. I n a piano b o t h hands use the softer m a n u a l .
A pianissimo can be expressed o n this m a n u a l , b u t w i t h reduced
parts. I n order to practice these precepts the ear must p r o v i d e constant assistance, for indications are n o t always exact; moreover,
matters of t o n a l v o l u m e depend o n the desires of the p e r f o r m e r
of the p r i n c i p a l part.

A (COMPAIA!

A C C O M P A N I M I:N T

E N T
Figure 429

JTT1

fia

[7T

8. A n accompanist must be careful to observe whether the h i g h

a n d l o w registers o the singer and instrumentalist w h o m he accompanies are equally l o u d , and whether his tones are just as clear
f r o m a distance as near by. I f they are not, he must m o d i f y his playi n g i n order n o t to cover the weak tones w i t h a l o u d accompanim e n t . For example, i t is c o m m o n knowledge that the upper tones of
the transverse ute are b r i l l i a n t , b u t the lower tones not, a l t h o u g h
otherwise its tones are u n i f o r m .
9. A forte i n a t u t t i passage is to be differentiated f r o m a forte
that accompanies a soloist. T h e latter must be accurately proportioned to the strength of the p r i n c i p a l part, b u t the f o r m e r can, of
course, be m u c h louder.
10. M o d u l a t o r y changes are announced by a reinforced accomp a n i m e n t . T h u s , for example, i n a fortissimo, a f u l l c h o r d is taken
by b o t h hands and arpeggiated r a p i d l y upwards, the bass w i t h its
octave and the tones i n the r i g h t h a n d being retained (Figure 429,
Example ). W h e n certain passages are repeated i n transposition,
d o u b l e o n l y the p r i n c i p a l bass notes i n the interests of greater
clarity (b). B u t i f such passages are so fashioned that they can be
played t h r o u g h o u t i n octaves, differentiate the p r i n c i p a l notes
t h r o u g h the use of reinforced chords, played, perhaps, by b o t h
hands. T h e notes referred to are those i n c over w h i c h a stroke has
been placed. Aside f r o m these cases, chords struck d u r i n g short
rests achieve a distinct weight and greatly assist the other performers, for i t is generally acknowledged that, except for the
keyboard, such rests present
considerable
difficulties (d).
T h i s last suggestion applies to a l l passages i n w h i c h short rests
appear.
11. T h e first note after a fermata
or general pause is struck
l o u d l y . Even i f the note is m a r k e d piano, a certain weight is
given to i t by means of a moderately strong attack. Such l i b e r t y is
especially needed w h e n the bass alone breaks the silence. I t is better

?7'

Allegro

b.

r r r

r 'r r r

to play one note somewhat louder than indicated and thereby ret a i n order i n the ensemble, t h a n to observe a l l indications w i t h
exaggerated exactness and f a i l to give an indispensable signal to
the others. T h e r e is a possibility that the resultant confusin m i g h t
r u i n a considerable p a r t of a composition, i n t o w h i c h the composer may have i n t r o d u c e d especially b e a u t i f u l effects. I n such
places the i n s t r u m e n t that first resumes is the leader, even i f i t
should be the viola.
12. Notes that i n t r o d u c e closing cadenees are played l o u d l y regardless of whether they carry express indications. By this means
the p r i n c i p a l part is i n f o r m e d that an elaborated cadenee is expected, for w h i c h the accompanist w i l l halt. T h i s signal is especially
necessary i n allegros, for elaborated cadenees are more usual i n
adagios. I n the latter the p r i n c i p a l p a r t o f t e n plays the note before
the cadenee w i t h a retarded forte so that the accompanist may k n o w
that the cadenee is going to be elaborated.
13. W h e n the p r i n c i p a l p a r t has a l o n g h e l d note w h i c h , accordi n g to the rules of good performance, should commence pianissimo,
grow by degrees to a fortissimo, and r e t u r n s i m i l a r l y to a pianissimo, the accompanist must f o l l o w w i t h the greatest exactness.
Every means available to h i m must be employed to a t t a i n a forte
6

5 A nuance that was generally callee! messa di

voce.

?72

//

C C O M VA

A C C O M V A N I M E N T

N l M E N I

373

and j)iano. Ili.s inuc.isc and dccrease must coincide w i t h that of

the p r i n c i p a l part; n o t h i n g inore, n o t l i i n g less.
14. W i t h regard to staccato a n d legato, i t should be observed
that i n p l a y i n g chords over a bass w h i c h is n o t m a r k e d staccato, i t is
not obligatory to use a fresh attack on every note. Tones that already lie i n a preceding c h o r d and can be carried over to a followi n g one are h e l d . Such an execution, w h e n associated w i t h owing
pfogressions i n the best d i s t r i b u t i o n , gives the accompaniment a
singing effect. I t is indispensable w i t h legato notes. For these good
reasons, most of the examples i n the present book should be taught
i n this manner, so that the student w i l l be i n t r o d u c e d early to a
sustained delivery and be preserved f r o m the hacked t h o r o u g h
bass w h i c h is as f r e q u e n t as i t is ugly. I f the tempo is very slow and
the i n s t r u m e n t so unusually poor that h e l d notes do n o t last l o n g
enough, a fresh attack is, of course, i n order. B u t this expedient
does n o t apply to the organ.

placing them between notes. It is stiange that tlu-y have been overlooked despite the fact that 0111 present elegant taste must make the
need for them q u i t e obvious. I l o w m u c h unevenness i n resolutions
can be caused by their absence! A n d h o w the performance and
character of pieces sufferl H o w ceaselessly attentive must be the
ear that w i l l p e r m i t n o error! Figure 431 w i l l illstrate m y meaning.

15. T h e sixteenths i n the examples of Figure 430 sound i n s i p i d

i n an adagio i f dots are n o t placed between t h e m . I t is advisable to
correct this f a u l t i n performance. Because proper exactness is o f t e n
lacking i n the n o t a t i o n of d o t t e d notes, a general r u l e of p e r f o r m ance has been established w h i c h , however, suffers many exceptions.
A c c o r d i n g to this r u l e , the notes w h i c h f o l l o w the dots are to be
played i n the most r a p i d manner; a n d o f t e n they s h o u l d be. B u t
sometimes notes i n other parts, w i t h w h i c h these must enter, are so
d i v i d e d that a m o d i f i c a t i o n of the r u l e is r e q u i r e d . A g a i n , a suave
affect, w h i c h w i l l n o t survive the essentially defiant character of
d o t t e d notes, obliges the p e r f o r m e r slightly to shorten the d o t t e d
note. Henee, i f only one k i n d of execution is adopted as the basic
p r i n c i p i e of performance, the other k i n d s w i l l be lost.

17. W h e n many repeated s l u r r e d notes appear i n a slow tempo

and the lower octave is to be taken w i t h t h e m , the d o u b l i n g s h o u l d
oceur o n only the first a n d t h i r d notes, or, i n a t r i p l e t , o n only the
first. F u r t h e r m o r e , these tones i n the lower octave should be h e l d
(Figure 432, Example a ) . B u t i f such repeated notes are to be played

Figure 430

+.

+-2

Figure 431
Adagio

"C.
4.

5
3

C
4- 5
2 . 3

6^
4

7
3

a.

"
i

i.

'fffff

'r r "

cJrrlrruficLjLfr^^ccJ

16. A m o n g the signatures n o t yet i n use are dots. T h e r e are just

as many grounds f o r p l a c i n g t h e m between figures as there are f o r

\ = = =#==)

9 = 4

Dots in signatures were used much earlier by Johann Staden (Kurzer una einfltiger Bericht,
1626). T h e i r use is described in Arnold, op. cit
p. 105. Neither
Staden's or Bach's practice was widely followed.
6

-J'

-J

# J

--

L U J

374

A C COMPAIA!

E N T

A C C O M P A N I M E N I

staccato a n d rapidly so that they sound loudly (l>), the execution i n

c may be employed However, the passage nnrsl not last long, for i f
i t does the w r k t s w i l l grow stiff and exhausted. Such being the case,
i t is better to play the notes i n the manner of other d r u m basses as
described i n the I n t r o d u c t i o n to Part I , Paragraph 9 a .
18. I n a concert or any heavily scored piece, w h e n the bass a n d
r i p i e n o parts h o l d a tone w h i l e the p r i n c i p a l p a r t contines w i t h
its o w n m o t i o n , even v a r y i n g i t at times w i t h syncopations, i t is
wise f o r the accompanist to m a i n t a i n the beat a n d guide the other
performers by p l a y i n g a c h o r d w i t h the r i g h t h a n d o n the divisions
of the bar even t h o u g h the h a r m o n y does n o t change. I f the bass
alone has the held note, the accompanist may repeat solely the bass
note just as i t dies out. B u t this must n o t be done "against the beat,"
as the expression goes. I n a d u p l e bar the repetitions may oceur at
the b e g i n n i n g and i n the m i d d l e , according to its divisions a n d the
pace I n a t r i p l e bar only the downbeat is played. B u t i f i n the
course of a passage a forte appears after a piano, the accompanist
should give u p the prescribed divisin and, observing closely, play
the bass i n the left h a n d a n d the c h o r d i n the r i g h t d i r e c t l y o n the
entrance of the forte i n a fortissimo b o t h hands take a f u l l c h o r d .
H e r e again, because of a lack of signs, the entrance of l o u d a n d soft
cannot be accurately indicated i n the bass or the figures.
7

19. W h e n the bass a n d several other parts p e r f o r m t h e i r notes

pizzicato, the accompanist pauses, leaving the passage to the cellos
and d o u b l e basses. B u t i f only the bass is pizzicato the accompanist
plays staccato, e m p l o y i n g his left hand alone, unless the composer
has w i t h good reason placed figures over the notes, i n w h i c h case the
r i g h t h a n d adds its chords, w h i c h are also played staccato. W h e n
pizzicato a n d coll'arco
are interchanged after b u t a few notes, the
one must be sharply distinguished f r o m the other, regardless of
whether this is achieved by way of a pause, a detached tasto solo or
unisoni,
or a staccato performance of the chords. I f d u r i n g this req u i r e d detaching the p r i n c i p a l part has appoggiaturas, w h i c h u n d e r
n o r m a l c o n d i t i o m w o u l d be played i n the accompaniment, they
should be o m i t t e d and only the r e m a i n i n g appropriate figures
7 See also Arnold bp. ctt pp 774 ff., where Heinichen's views are stated and
Saint Lambert s mentioned. T h e latter's opinions appear on p. 196 of that book. Bach
is in agreement with both

175

realized. A legato appoggiatura and relase do not go well i n a

staccato passage.
20. I n slow or modrate lempos, caesurae are usually extended
beyond their n o r m a l l e n g t h , especially when the rests and notes i n
the bass are the same as those i n the other parts, or i n the p r i n c i p a l
part i n the case of a solo. Great pains must be taken to achieve a u n i f o r m performance and prevent anyone's c o m i n g i n before or after
the others. T h i s applies to jermate, cadenees, etc., as w e l l as caesurae.
I t is customary to drag a b i t a n d depart somewhat f r o m a strict observance of the bar, for the note before the rest as w e l l as the rest i t self is extended beyond its notated l e n g t h . Aside f r o m the u n i f o r m i t y w h i c h this manner of execution achieves, the passage acquires an impressiveness w h i c h places i t i n relief (Figure 433).
Figure 433

i -

11

p r *p
1

21.
Closing t r i l l s are o f t e n extended, regardless of the tempo.
B u t i f a piece has reprises, the extensin of t r i l l a n d accompanying
bass note takes place only at the end of the final r e p e t i t i o n . By this
means weight is added to the conclusin a n d the audience is made
to feel that the piece is about over. T h i s k i n d of cise, however,
despite its good uses, cannot be i n t r o d u c e d i n t o a l l contexts. Henee
the accompanist must be extremely w a t c h f u l , especially because
some closing t r i l l s m u s t be played strictly i n tempo due to either
the b r i l l i a n t or the reflective character of a passage (Figure 434,
Example a). I t is understood, moreover, that the accompanist does
not h o l d back w h e n the t r i l l appears over a m o v i n g bass (b). B u t i f

A C C O MPA

37*

A C C O M P A N 1M E N T

N I M E N T

the last of these bass notes is the f i l t l i of the key, it should be held
u n t i l i t is observed that the p r i n c i p a l part or the other executants
are ready to conclude their t r i l l (c). T h e same procedure is to be
f o l l o w e d w h e n solely the fifth of the key is repeated i n the upper or
lower octave after the entrance o f the t r i l l ( d ) . B u t i f a piece ends
w i t h o u t a closing t r i l l , i t should be played i n tempo w i t h o u t h o l d i n g back (e).
Figure 434
Andante
Allegro

Pf mr f r r r

Adagio

I* I'

"

" 1

r-

d.

Figure 435
"

>y

P
1

377

i ip ce
2

3 3

17

44

3
3

23. I n accompanying a solo o n a l o w i n s t r u m e n t (i.e. a bassoon,

cello, etc.), or an aria f o r a low voice (i.e., a tenor or bass), cise att e n t i o n must be given to the range of the melody so that the chords
w i l l n o t be placed too h i g h above the p r i n c i p a l part. O r d i n a r i l y
the r i g h t h a n d should n o t venture above the one-lined octave. I f i t
is necessary to take the chords q u i t e l o w , they should be reduced,
for f u l l chords played low lose t h e i r clarity.
24. I f a piece for a l o w p r i n c i p a l p a r t has r i p i e n o parts, t h e i r
range must be carefully observed a n d the keyboard accompaniment
placed i n the same register. T h e melody of the p r i n c i p a l p a r t must
not be obscured by m i d d l e parts w h i c h l i e above i t . Henee composers occasionally place the m i d d l e parts first i n the l o w register
i n the interests of a good setting and v a r i a t i o n f r o m the n o r m , a n d
later, i n the r i t o r n e l l o , restore them t e l l i n g l y to t h e i r usual upper
register. T h e accompaniment must be fashioned i n c o n f o r m i t y w i t h
such a procedure. I n a f o l l o w i n g section of this Essay a t t e n t i o n w i l l
be directed to those cases where different registers are e x p l o i t e d i n
the interests of r e f i n e m e n t . A t present i t w i l l suffice to r e m i n d the
accompanist of a p o i n t w h i c h we have stressed repeatedly, namely,
that care must be taken to construct a good melody i n the u p p e r
part. A n d i n connection w i t h this, the best d i s t r i b u t i o n s are to be
taken at a l l possible times. Should i t become necessary to ascend
above the p r i n c i p a l part, tones must be placed i n the u p p e r part
w h i c h move i n thirds or sixths w i t h lower m i d d l e parts (a), the
p r i n c i p a l part (b), or the bass (c) (Figure 436).
11

r T ^ r f
r
r B m i i ^T^m i
r
r rr
1

22.* W h e n the p r i n c i p a l part moves i n tenths w i t h a m i d d l e

part, the accompanist plays thirds i n the lower register rather t h a n
tenths. T h i s octave d o u b l i n g of the p r i n c i p a l part is better t h a n a
similar d o u b l i n g of the m i d d l e p a r t (Figure 435).
9

From the ed. of 1797.

Which would happen if the thirds were played in the higher octave.
8

25. A l t h o u g h i t is poor constantly to duplicate the melody of

the p r i n c i p a l part i n the accompaniment, there are times at the beg i n n i n g of r a p i d pieces w h e n i t is necessary a n d henee permissible,
12

10 From the ed. of 1797.

11 Cf. C h . V I , "Some Refinements of Accompaniment," f f 9, 10.
12 Cf. J . S. Bach's correction of a figured bass accompaniment by N . Gerber in
Philipp Spitta, J. S. Bach, Vol. I I I , Appendix, beginning of second and third movements. However, the beginning of the last movement, allegro, which is similarly
constructed, has a four-part chord.

A C C O M I' A N 1 M E N T

379

bass of a consonant chord leaps an octave the register may be more

safely changed t h r o u g h opposite rnotion than i n other leaps (Figure
437, Example a ) . F u r t h e r , progressions w h i c h c o n t a i n dissonances
on d i v i d e d beats and unprepared dissonances are also convenient,
for they leave the accompanist free to select the register (b).
Figure 437
a.

especially w h e n they are i n t w o parts. B o t h hands thereby pick u p

the tempo i n the same m a n n e r , and the audience w i l l n o t miss
any part of the b e g i n n i n g , for i t w i l l be u n i f o r m and orderly. T h i s
is an expedient w h i c h may be employed by weak musicians, whether
accompanists or leaders, at any p o i n t i n a composition as a means
of r e t r i e v i n g a u n i f o r m beat, once i t has been lost.
ia

26. I f , as a result of many downward-resolving dissonances, the

r i g h t h a n d descends too l o w , the accompanist must make use of a l l
of the means w h i c h have been described i n this Essay i n order to
r e t u r n gradually to a h i g h register. T h e y are to be employed especially over long bass notes, o n the appearance of consonant chords
and repetitions of t h e m , over passing tones i n the bass, etc. Sometimes, w h e n the p r i n c i p a l part is n o t notated i n the c o n t i n u o part,
this procedure is a r e q u i r e m e n t of good p l a n n i n g , for p r i n c i p a l
parts and others too, b u t n o t the c o n t i n u o , may leap a b r u p t l y f r o m
the l o w to a h i g h register. I t can be seen f r o m these cases and others
(which, a l t h o u g h they defy description, are soon discovered) that
the accompanist must practice d i l i g e n t l y i n order finally to achieve
mastery of the registers. I mean here m o r e t h a n the mere a b i l i t y to
realize intervals f o r t h w i t h i n any register; I refer to the artistry
w i t h w h i c h chords can be used i n order to reach a r e q u i r e d register,
be i t what i t may. I n this connection, a few situations await illustrat i o n w h i c h expedite a change of register. For example, w h e n the
is Presumably they play in unisn.

5l

i
4

- 6

8.U7

6,

I J

i I r i r i i r i'

5j

27. I n accompaniments, just as m u c h as i n solos, a constant

p l a y i n g o n the surface of the keys must be avoided; rather they
should be depressed w i t h definite forc. T h i s w i l l n o t occur unless
the hands are raised somewhat. P r o v i d e d this is n o t done i n the
manner of a woodchopper, the raised hands are n o t only n o t w r o n g ,
b u t necessary and good, i n that they p r o v i d e a simple way of i n d i cating the tempo to the other performers a n d make i t possible to
strike the keys w i t h proper weight so that the tones w i l l sound
clear, i n accord w i t h the rules of good performance.
CLOSING

1.

CADENCES

M y readers w i l l have learned f r o m Part I of this Essay that

closing cadenees appear b o t h w i t h a n d w i t h o u t elaborations. I n

2

Quantz furnishes a historical note on the cadenza (Versuch einer Anweisung

die
Fite traversiere
zu spiefon, C h . X V , . ^[ 2): " T h e best Information that can be given
on the origin of the cadenza is that some years before the cise of the last century
[the i7th] and during the first ten years of this one, the performer would add a short
passage over a moving bass, and append a good trill, at the conclusin of a concertizing
part. But between roughly the years 1710 and 1716 tht present type of cadenza, during
which the bass pauses, became fashionable. T h e fermata or retard ad libitum in the
middle of a piece may well be of earlier origin." T o this it should be added that the
cadenza, as discussed both by Bach and by Quantz, is different from the later Mozartean type in that it is not compounded of the thematic material of the main body
of the movement which it graces.
2 Cf. C h . I I , " T h e Elaboration of Fermate."
1

f80

A C C O M VA

N I M EN

the present sed ion we shall instruct the accompanist i n the treatment of b o t h kinds.
2.
O n the entrance of an elaborated cadenee, the accompanist,
regardless of whether a fermata
appears over the bass, holds the
six-four c h o r d for a w h i l e and then pauses u n t i l the p r i n c i p a l part,
at the end of its cadenza, plays a t r i l l or some other figure w h i c h
requires r e s o l u t i o n of the c h o r d . A t this p o i n t the t r i a d is struck
at the keyboard, the seventh being taken as a fifth part. F r o m
adagio m o l t o to andante the six-four c h o r d and the succeeding
t r i a d are arpeggiated u p w a r d either slowly or rather r a p i d l y acc o r d i n g to the requirements of tempo and affect.
3. W h e n the bass of a piece i n more t h a n t w o parts has a rest
after the b e g i n n i n g of an elaborated cadenee, the accompanist
strikes the t r i a d o n the d o m i n a n t at the conclusin of the cadenza,
regardless of whether i t is announced by a t r i l l , or l a c k i n g this,
some other figure, or a pianissimo, and then pauses again i f other
rests f o l l o w .
4. A t times the bass enters i m m e d i a t e l y after the conclusin of
a cadenza or a cadenee prolonged simply by means of an extended
t r i l l . T h e entrance must be made w i t h firmness and an assured res u m p t i o n of the tempo as soon as i t is observed that the t r i l l i n the
p r i n c i p a l part has been sufficiently extended and i f c o n t i n u e d may
grow weak. T h e tones w h i c h the bass plays must be p e r f o r m e d
strongly and l o u d l y , even i n the absence of an i n d i c a t i o n , so that
the other performers w i l l grasp the restoration of the n o r m a l tempo.
I f such bass notes are m a r k e d piano (a case w h i c h arises o n l y
rarely) at least the first of the tones w h i c h precede the approaching
bar should be struck l o u d l y , or some m o t i o n of the body should
be made as a means of i n d i c a t i n g the divisin of the bar (Figure
438).

Figure 438

5. Occasionally the p r i n c i p a l performer feels d i s i n c l i n e d to

elabrate a cadenee, despite the presence of a fermata over the bass.
Such being the case, he motions w i t h his head or body to i n f o r m
his accompanist. T h e latter, h a v i n g observed this, substitutes for

A C C O M V A N 1 M E N T

381

(he long, sustained bass note ,1 series ol short notes similar to the
previous ones, as a means o m a i n t a i n i n g good order and l e t t i n g the
other performers hear the u n i n t e n upted c o n t i n u a t i o n of the tempo
(Figure 439).
Figure 439

11 1

i i

va
4

6
4

6. W h e n , as i n Figure 440, a composer, u n m i n d f u l of elaborations, allows his bass to continu its m o t i o n t h r o u g h a closing
cadenee, the accompanist holds the first g and repeats i t o n the
t r i l l , after w h i c h he begins the f o l l o w i n g bar. T h i s case, w h i c h often
oceurs i n allegros, requires an attentive ear.
3

Figure 440

5*
6

6
4

5
3

7. I n an a n d a n t i n o and allegretto, the six-four c h o r d and its

succeeding t r i a d are arpeggiated u p w a r d q u i c k l y and h e l d . I n allegros, however, the six-four c h o r d and its bass are o f t e n q u i t t e d
a b r u p t l y before the cadenza. T h e p r i n c i p a l p e r f o r m e r is thus
granted the freedom i n fiery pieces to begin his elaborations after a
very short h o l d and to i n t r o d u c e many r a p i d and other notes that
are somewhat related to the preceding chord. T h i s need n o t always
be the case, a l t h o u g h the six-four c h o r d should be kept as m u c h i n
m i n d as possible at the b e g i n n i n g of elaborations. I f too n a r r o w
l i m i t s were placed o n the elaborated cadenee, the abuse w o u l d be
more unbearable t h a n i t is already, for even n o w we must p a t i e n t l y
endure m u c h that cannot be remedied. Aside f r o m this case, i n
s T h i s is the earlier type of cadenza described by Quantz (op. ext., Note 1). Bach's
suggested modification is a means of bringing it up to date.

8i

ACCOMPANIM

A C C O M I' A N 1 M E N T

E N 1

w h i c h the p r i n c i p a l performer commences his elaborations i m mediately after the six-four c h o r d , he usually holds the fermata f o r
a w h i l e i n order n o t to be too greatly restricted by the reverberati n g six-four c h o r d , a n d begins his cadenza only after the sound of
the keyboard has almost e x p i r e d . T h i s execution is good, f u r t h e r ,
because the listener w i l l be p r o p e r l y prepared, the preceding sixf o u r c h o r d h a v i n g been w e l l impressed u p o n his ears.
8. I t is m o r e custom t h a n musical law that leads the c o n c l u d i n g
t r i l l of cadenzas to be played on the fifth above the bass, or o n the
sixth, occasionally, i n the m i n o r mode. Because the accompanist
awaits this t r i l l and strikes his t r i a d d i r e c t l y o n its entrance, he
must be extremely careful, i n the case of cadenzas that are fashioned
out of a series of t r i l l s , to a v o i d c o m i n g i n w i t h his chord o n the
entrance of a long t r i l l on the t h i r d . Such a t r i l l is usually a clear
sign that the cadenza is n o t at a l l ended, a n d i f he plays the t r i a d
p r e m a t u r e l y there is a danger that many subsequent tones w i l l appear that disagree w i t h i t . A competent p r i n c i p a l p e r f o r m e r w i l l
make every effort i n such a s i t u a t i o n to shorten and conclude his
part so that n o one w i l l hear ugly sounds. B u t accompanists should
not provoke such a change. Should i t be a p e r f o r m e r s pleasure to
conclude his cadenza w i t h a t r i l l o n the t h i r d , i t must also be his
pleasure to w a i t for an accompanist w h o does n o t play his t r i a d
immediately, b u t listens to the t r i l l f o r a m o m e n t to make certain
that the cadenza is g o i n g t o be ended by i t . Some p r i n c i p a l performers take satisfaction o u t of p l a y i n g a l o n g t r i l l on the fifth, leadi n g the accompanist to enter w i t h his r e s o l u t i o n of the six-four
chord, after w h i c h they continu w i t h elaborations w h i c h very
often do n o t harmonize w i t h the r e s o l u t i o n . T h e accompanist
should c o n t a i n himself i n the face of so b o l d a stroke, i n the assurance that n o justifiable c r i t i c i s m can be directed at h i m . A n d he
should gainsay his leader neither his pleasure or the credit for an
effective v a r i a n t .
9. I n the examples of F i g u r e 441, w h i c h are f o u n d at times, the
t r i a d , i n a r a p i d tempo, is struck over the first note and retained
u n t i l the last appears, at w h i c h p o i n t the h o l d occurs. T h e intermediate notes are played w i t h o u t any change i n the r i g h t hand,
despite the signatures (a). I n a slow tempo the signatures are m o d i fied as indicated i n b a n d the h o l d s.tarts over d.*
T h e reason for the modificador! is that the first b in each example is extended

>9)

Figure 441

r r r " r r r r r r rrr
r

7
5

7
5

65
43

11
65
43

10. H a l f cadenees, i n w h i c h there appears over the next to last

note of a piece 7 6, or 7 6$, are n o longer as f r e q u e n t as they used
to be. T h e chord of the seventh is h e l d u n t i l the p r i n c i p a l performer plays the r e s o l u t i o n , w h i c h is usually a l o n g t r i l l o n the
sixth or t h i r d over the bass, o f t e n i n the wake of a few embellishments. T h e accompanist t h e n plays and sustains a c h o r d of the
sixth w h i c h may be arpeggiated i n a slow tempo (Figure 442).
Figure 442
6

jJ I
6

I .1

II

11.
W h e n an aria or other piece i n the m a j o r mode changes to
m i n o r i n the second section a n d a da capo follows, the final c h o r d of
Figure 443

7kn

Andante

\7

6> 5

\
4

beyond itsiwritten length arid thereby delays the entrance of a until the last quarter
of the bar. Further, since this is a closing cadenee, the a is trilled beyond its written
length, henee the hold starts on the last quarter. Cf. Arnold, op. ext., pp. 289 and 293,
where this deviation is related to eighteenth-century practice.

jfy

A CC

O A i PAI

A <; C O

M E N T

the second section ai'ter the cadenza must be major even i n the absence o an i n d i c a t i o n . S i m i l a r l y , care must be excrcised i n pieces
i n the m a j o r mode, for composers sometimes approach the cadenza
by way of the m i n o r of the o r i g i n a l tonality. A l t h o u g h the m a j o r
sixth a n d seventh become m i n o r , the last chord, f o l l o w i n g the
cadenza, must be major (Figure 4 4 3 ) .
5

A/

P A N

/ A/

E N I

,V

other ernbellishinenis are I I I I K K I I K <<l, tasto solo becomes indispensable. If a l o r i e appears al 11 imclei the held bass, the r i g h t hand
may strike its c h o r d , sharply delached or very r a p i d l y arpeggiated,
against the bass note.
Figure 445
a.

12. I n F i g u r e 444 the signatures have been constructed w i t h o u t

regard to an extended cadenee. T h e accompanist does n o t play the
6 5

prescribed progression, b u t realizes 4 3 over g instead of 4 3, once

he has observed that the p r i n c i p a l part w i l l be prolonged, whether
w i t h or w i t h o u t elaborations. A l l similar cases should be realized
i n this manner, regardless of their signatures.
Figure 444

4 3

THE

3. W h e n the p r i n c i p a l p a r t broadens before entering a fermata,

the accompanist must d o likewise. T h e notes i n question are
m a r k e d w i t h crosses i n F i g u r e 446. Should the p r i n c i p a l p e r f o r m e r
h o l d the a and t h e n i n t r o d u c e embellishments, the accompanist
contines u p to the /-sharp a n d remains o n i t and its respective
c h o r d u n t i l he observes that the appoggiatura over the f o l l o w i n g
bass note has appeared, w h e r e u p o n he plays g w i t h the left h a n d
alone a n d at the conclusin of the embellishments repeats i t under
its arpeggiated t r i a d .
Figure 446

FERMATA

mm

1. W e k n o w f r o m Part I of this Essay that fermate

are perf o r m e d i n various ways. N o t h i n g remains here b u t to show h o w
the accompaniments are constructed.
2. T h e p r i n c i p a l p a r t may approach a fermata
t h r o u g h a leap
or an appoggiatura. I n b o t h cases the accompanist strikes the h e l d
bass note w i t h his left h a n d alone and, at the end of the
fermata,
repeats i t , arpeggiating the appropriate h a r m o n y above i t . T h e
p r i n c i p a l p a r t is thereby granted a l l possible freedom to execute
the h o l d i n any desired manner. T h u s , i n Example a, F i g u r e 445,
an artistic increase and decrease i n the v o l u m e of the t r i l l s or h e l d
notes w i l l n o t be obscured by the sound of a chord, a n d i n b there
w i l l be n o interference w i t h the execution of the appoggiatura. I f
2

5 I.e., the six-four chord and the seventh of the chord of the seventh on /-sharp.
r T h i s is the "retard ad libitum"
mentioned by Quantz (cf. C h . V I , "Closing Cadenees," Note 1). T h e present section treatS cadenzas that oceur in the course of a
piece and is thereby distinguished from the preceding section, which discusses only
final cadenzas.
2 Cf. C h . I I , " T h e Elaboration of Fermate."

4. W i t h regard to h o l d i n g a n d c o n t i n u i n g i n tempo at
fermate,
the accompanist must f o l l o w exactly the precepts that have been
stated i n Part I of this Essay.? T h e p r i m a r y concern is that the p l a i n
notes as w e l l as the embellishments of the p r i n c i p a l part, w h i c h
precede the fermata, harmonize p r o p e r l y a n d enter simultaneously
w i t h the bass a n d its chords.
5. Fermate w i t h o u t appoggiaturas or elaborations, and passages
i n w h i c h the h o l d is placed over a succeeding rest, are realized
f o r t h w i t h and p l a i n l y .
6. Figure 447, s i m i l a r to F i g u r e 433, is sometimes p e r f o r m e d
i n the manner of a fermata,
even t h o u g h none is i n d i c a t e d . For
s Cf. Note 2 above.

j86

A C C O M I A N I M I: N I

A C C O M I' A N 1 M /<. N T

reasons of affect, (lie pt inc i pal part, d e v i a t i n g from a strict beat,

moves q u i t e slowly f r o m the appoggiatura c i n t o b. T h i s note is
held, the sixteenth rest is extended, and only therealter is the tempo
resumed. T h e accompanist must f o l l o w this closely and, above a l l
else, a v o i d a premature resolution of the fifth over the t h i r d f , for
this i n t e r v a l is the same as the appoggiatura. T h e resolution appears
on the d i v i d e d beat as part of a slowly arpeggiated c h o r d of the
second, i m m e d i a t e l y after the p r i n c i p a l p a r t has taken b.

he accompanies to w i n <oveled honor, even though his powers may

at times o u t s t r i p theirs. H e exhibits this modesty especially t o w a r d
amateur performers. Far f r o m overshadowing them, he allows t h e m
to predomnate. Moreover, he always puts himself i n agreement
w i t h the aims of performer a n d composer; and he seeks to advance
and support these aims. H e employs every possible nuance of performance and accompaniment, p r o v i d e d i t is r e q u i r e d by the conten of a piece. B u t i n using these nuances he takes every precaution
to see that no one w i l l be hampered by t h e m . W i t h this i n m i n d he
is n o t lavish w i t h his artistry, b u t indulges i t sparingly a n d o n l y
w h e n i t creates a good effect. H e is never afflicted w i t h overbearing
wisdom, a n d he never forgets that he is an accompanist, n o t a
soloist. H e knows that a good accompaniment brings a piece to l i f e ;
that, o n the other h a n d , the best performer suffers immeasurably
f r o m an i n e p t accompaniment, for a l l of his nuances w i l l be r u i n e d
by i t a n d , even worse, his appropriate disposition w i l l be destroyed.
I n short, a discreet accompanist must have a fine musical soul,
w h i c h comprises great understanding and good w i l l .

Figure 447 , - 3
Adagio

JjJ-l ^ jj

a r\
9

SOME

66 -4
2 -

'B

REFINEMENTS

OF

ACCOMPANIMENT

j8y

1. I t is w o r t h repeating that the refinements of accompaniment

need n o t consist of i n a p p r o p r i a t e r a p i d figuration a n d embellishments, w i t h w h i c h some accompanists distort the melody of the bass.
W e have already shown q u i t e different ways by w h i c h the accompanist may w i n acclaim, a n d we shall continu here i n a similar
vein.

4. T o accompany w i t h discretion means also to make adjustments to others' errors a n d to give way before t h e m . T h i s may oceur
out of politeness or necessity, as, for example, i n the usual performance of a large piece whose numerous players are n o t of u n i f o r m a b i l i t y . T h e best leader must give way i n such a s i t u a t i o n , a n d
so must the accompanist.

2. M u c h w i l l appear here of w h i c h the accompanist s h o u l d n o t

avail himself u n t i l his insight has progressed to the p o i n t where
he knows precisely w h e n and where refinements may be i n t r o d u c e d .
H e should n o t ascend higher t h a n his wings w i l l carry h i m .
3. T h e most usual expression that is used to describe a' good
accompanist is, " H e accompanies w i t h d i s c r e t i o n . " Such praise has
an inclusive m e a n i n g ; i t amounts to saying that the accompanist
can discrimnate a n d henee fashion his setting according to the nat u r e of a piece, the n u m b e r of its parts, the other performers, especially the p r i n c i p a l performer, the instruments, voices, place,
audience, etc. W i t h extreme modesty he tries to assist those w h o m

5. T o accompany w i t h discretion means moreover to make

modifications i n accord w i t h certain liberties that are taken at
times by the p r i n c i p a l performer, w h o , w i t h o u t "its b e i n g actually
r e q u i r e d , may depart somewhat f r o m the w r i t t e n notes i n i n t r o d u c i n g embellishments and variants. A k n o w i n g p r i n c i p a l perf o r m e r w i l l do this w h e n he knows that he has an able accompanist,
and thus abandon himself w i t h complete freedom to the affect of a
piece. Such liberties spring n o t f r o m f a l t e r i n g u n c e r t a i n t y b u t
f r o m a r a t i o n a l sovereignty and p e r t a i n only to details w h i c h exact
f r o m the accompanist n o t h i n g b u t attentiveness. I n Figure 448,
Example a, the p r i n c i p a l performer i n i n t r o d u c i n g embellishments
sometimes substitutes one of the i l l u s t r a t e d series of signatures f o r
the other. T h e accompanist must m o d i f y his h a r m o n y accordingly.
I n a d d i t i o n to such substitutions, the accompanist must be attentive a n d give way w h e n embellishments, i n t r o d u c e d i n t o the p r i n -

i Von gewissen Zierlichkeiten

des Accompagnements.
Much valuable supplementary material is contained in C h . I V of Arnold, op. cit. I n addition to generous
excerpts from Bach, Arnold includes material from Heirtichen (with whom Bach disagrees in certain respeets), Geminiani, and Mattheson, and a reference to Saint L a m b e n as probably the first to discuss such matters as are here treated.

j88

C C O M I> A N I M

/: a ; /

A C C O M I' A N I M 1: N T

cipal part, lead to a latcr entrance of chords than actually denoted

by the signatures (b):

98$

j JJ J i J

Figure 448

6. O u t s t a n d i n g among the refinements of accompaniment is

parallel m o t i o n i n thirds w i t h the bass. T h e r i g h t h a n d here never
restriets itself to a u n i f o r m l y f u l l setting. A consistent f o u r - p a r t
accompaniment is rare, o c e u r r i n g o n l y i n connection w i t h slow
notes (Figure 449, Example a), for thirds i n a f o u r - p a r t setting cann o t be clearly b r o u g h t o u t at a r a p i d tempo. Preferable are threepart and, i n most cases, two-part accompaniments, the latter consisting simply of a bass p a r t d u p l i c a t e d i n the t h i r d above. F r o m the
examples of Figure 449, i t can be seen that a stepwise bass and one
that leaps i n thirds are most convenient for this k i n d of setting. T h e
accompanist often plays t h i r d s i n order to avoid a bad progression.
Some transitional passages can also be accompanied by thirds alone
(b). T h e second accompaniment under b must be changed w h e n i t
appears i n the m i n o r mode (c):
2

Figure 449

* r

>6

i i 1

-4f
W^u \ 1 r M
P

m P

7
5

"mh"Ha, - f - f
r

mfl

2 T h i s paragraph and those that follow, through *| 8, treat in detail a kind of

refinement which is discussed in a textbook used by J . S . Bach, F. E . Niedt's A U I calische Handleitung,
1700 (cf. Arnold, op. cit., pp. 213 ff., especially pp. 229-230).
J . S . Bach's copy of the work is reprinted in The Bach Reader, W . W . Norton and
Co., 1945. Examples of thirds used in imitative passages appear in Figur 458 here.

"

6
4

A C C O M /' A N I M E N T

ACCOMPANIMENT

391

hand against the bass are best when the p r i n c i p a l part has a held
note (Figure 450, Example ), a repeated note (>), slower notes (c),
or notes w h i c h are at least again as last as those i n the bass ( d ) . I n the
last case the precautions which are always f o l l o w e d where thirds
are used must be d o u b l e d i n order to avoid an ugly clash (e) or forb i d d e n progressions (/).

7. W h e n the bass of a two-part piece is so constructed that i t

m i g h t be accompanied i n thirds by the r i g h t hand, b u t the p r i n c i p a l
part has these thirds or some other i n t e r v a l w h i c h moves i n the
same r h y t h m as the bass, the chords should be played simply a n d the
thirds abandoned. I n the first case the accompaniment w o u l d duplcate the notes of the p r i n c i p a l part, a n d i n the second case the
melody of b o t h p r i n c i p a l p a r t a n d bass w o u l d be obscured by an
i n t e r p o l a t e d new m o t i o n . Consequently, parallel thirds i n the r i g h t
Figure 450
a.
-o-

pm
y

* f f f

l'i

LJ

' t / m
6

6l>

8. Sometimes sixths may intersperse w i t h successions of thirds

(Figure 451, E x a m p l e a ) . M a n y errors can be e l i m i n a t e d by such an
interchanging of the tones of the m i d d l e voice w i t h those of a passi n g bass (b). I n c w i t h its presto, the first note should n o t be weighted
d o w n w i t h a f u l l c h o r d ; rather the accompaniment should be constructed i n the manner of the i l l u s t r a t i o n that succeeds the example.
I t is easy to p e r f o r m , a n d its speed makes i t sound f u l l e r t h a n i t
really is. T h e same applies to d. I n e, where the bass leaps d o w n a
t h i r d f r o m a note w i t h the signature 6, a n d t h e n back again, a l l
three bass notes may be set w i t h thirds, or the t h i r d over the first
note may be h e l d a n d 6 3 6 taken i n the m i d d l e part. T h e illustrations u n d e r / are n o t e w o r t h y i n that various r h y t h m s , suspensions,
and h e l d notes appear i n the p r i n c i p a l part. I n g, a u n i f o r m fourpart setting over the r a p i d bass wOuld be difncult to p e r f o r m ; consequently the setting should agree w i t h the succeeding i l l u s t r a t i o n ,
w h i c h , moreover, can be conquered by average fingers. I t can be
seen f r o m this i l l u s t r a t i o n h o w easy such r a p i d basses can be made
by the use of thirds a n d sixths and, above a l l , h e l d tones. These latter are good f o r several reasons; they connect chords, p r o m o t e the
singing style and, at the same time, are easier and less of a risk t h a n
repeated attacks, w h i c h , i n a four-part setting at a r a p i d pace, are almost impossible a n d ineffective. E x a m p l e h illustrates the use of
thirds i n a series of chords of the seventh. Since the accompaniment
is i n f o u r parts, the tempo must n o t be very r a p i d . I n i the contrary
m o t i o n of the m i d d l e part circumvents octaves; henee i t is unnecessary to leap to the fifth of the second chord. I n j , thirds may be played
o n every bass note. Because of the ch'romatic change, m o t i o n i n
thirds cannot be used i n k; henee, p l a i n chords are played or use is
made of opposite m o t i o n to the bass, starting f r o m the first t h i r d .
U n d e r l a n d m a series of illustrations appear i n w h i c h the r i g h t
h a n d moves i n a graceful relationship w i t h the bass. Several other
examples can be deduced f r o m those cited i n n, i n w h i c h the r i g h t
hand plays lower thirds or sixths t h r o u g h o u t , as illustrated. T h i s

A CO

394

OMPANIMENT

"1

r; ; o

,J5.

m. 0 F

A N I

A /<; A/ T

395

'Irf ' i

Presto

Presto

Ti

V
0 1
i
i
- I

J - \

>
g+
?
t K i

a.

fl

i !

1* i
r

=ra
1
1

f l :

7 6 5b7

7 6 5

7 6 5

ji

88

II

ra.
1

. 1 -A'-Jra.

r | *

Adagio

AM

i > g f l

'iriLii''j
4

ib Jli
6

ra.

166

Adagio

Andante

AL

j6

A C C O M PA

1M E N T

Andante

Andante

n.

r' r

r f
5

* f

\5

16

194

9. A n elegant accompaniment that is n o t restricted to a u n i f o r m n u m b e r of parts may express certain leaps at times i n the r i g h t
hand. These often provide variety. T h e most frequent o p p o r t u n i ties to i n d u l g e this l i b e r t y occur i n passages w h i c h allow for i m i t a t i o n (Figure 452, Example a) or w h i c h c o n t a i n h e l d notes (b) or repeated figuration w i t h (c) and w i t h o u t transposition ( d ) . I n the case
of these last t w o examples, the justifiable demand of the ear for
v a r i a t i o n , w h i c h is caused by the excessive u n i f o r m i t y of the figurat i o n , can be met very easily and w i t h great freedom by an understanding accompanist. I t can be observed generally that those passages w h i c h contain only slight changes i n themselves are most
adaptable to v a r i a t i o n i n the realization of the chords. B u t even
though pieces w h i c h contain such passages can be aided by an elegant, free accompaniment, caution must be exercised to avoid excessive and u n t i m e l y employment of such nuances.
Figure 452

a.

^fc l.
0

&
6

'

1 J^JTII

'

r
1

r.

A C i,' O M P A N 1 M E N T

ACCOMPANIMENT

399

Figure 453

i?

T u .
- 5b

# "T

-6-6

16
5

r*r

'9
4

8
3

4
2

Crr rr Crr trr

65

65

6 5

Mi

d.

6 5

_ j J

65

i
)!

5 5

5 '

| J-

ri

r
'
f
r
t f 1
JJ
7

10. D i v i d e d accompaniment, w h i c h can be mastered by p l a y i n g

good keyboard music, is o f t e n one of the best refinements. I n the
preceding chapter a t t e n t i o n was directed to the occasional necessity of using i t . Aside f r o m such cases, i t is w e l l k n o w n that open
h a r m o n y is o f t e n extremely effective as a contrast to cise h a r m o n y .
For instance, we can see i n Figure 453, Example a, that the usual
construction of the chords sounds disagreeable because of the excessive u n i f o r m i t y of the passage, and, henee, that i t is better to use
another d i s t r i b u t i o n of the c h o r d of the second, b u t best to take i t i n
a d i v i d e d accompaniment (b). A repeated passage can be made attractive by the a l t e r n a t i o n of u n d i v i d e d a n d d i v i d e d accompaniments (c). I n d the sixths i n the r i g h t h a n d come t h r o u g h better
and the melodic progression is made clearer w h e n the l o w e r m i d d l e
part, w h i c h expresses n o melody b u t simply completes the setting, is
played by the left h a n d i n a r h y t h m s i m i l a r to that of the bass.

d.

17

n
g

frr
7

'7

"

9 9

5
3

11. T h e filling o u t of slow notes is one of the refinements of accompaniment. W h e n the tempo is slow, turns may be inserted over
the dots i n F i g u r e 454, Example a. W e r e this o r n a m e n t played i n
the bass also, the effect w o u l d be unclear. Because the tone of a
harpsichord does n o t always last l o n g enough, a n d slow or sustained
notes usually sound a b i t empty, an accompaniment that filis o u t

4<><>

A C C O AI PANIA

the d o t t e d notes may be ehosen for /; i n a slow lempo. T h i s example

illustrates a connecting passage d u r i n g which the p r i n c i p a l part
pauses, thus r e q u i r i n g the accompanist to find something that w i l l
prevent a feeling of emptiness. B u t w h e n the p r i n c i p a l part leads
i n t o the f o l l o w i n g section, p l a y i n g thirds or such against the bass,
the accompaniment should r e m a i n simple. Aside f r o m this, howFigure 454

A C (,' O M f A N I Ai E N T

E N I

401

ever, transitional passages ptovide e n l i c i n g challenges to an accompanist's inventiveness. Itui his i n v e n t i o n must be i n accord
w i t h the affect and conten o a piece. So m u c h the better i f part of
a preceding phrase can be r e i n t r o d u c e d , even i f this requires a
m o d i f i c a t i o n of the bass and a revisin of the transition. R a t i o n a l
sovereignty must be granted to the accompanist i n this case, so l o n g
as the p r i n c i p a l part is not thereby hampered. I n c either of the
appended accompaniments may be employed to fill o u t , b u t the
tempo of the second must be slower t h a n that of the first. T h e p r i n cipal part may have a h e l d note or a rest. Should i t be desired to let
the p r i n c i p a l part stand o u t i n d (whence the accompanist w o u l d
avoid a d u p l i c a t i o n of the tones w h i c h enter after the bass), the acc o m p a n i m e n t may be either of the settings u n d e r 1. B u t i f the p r i n cipal p e r f o r m e r varies this example by h o l d i n g the t h i r d above the
bass t h r o u g h o u t the bar (2), w i t h or w i t h o u t a t r i l l , the accompanim e n t u n d e r 3 should be chosen.
3

12. I n conclusin, we shall examine some signatures that are

used f o r purposes of i n s t r u c t i o n i n refinements (Figure 455). T h e y
are f o u n d singly and at times i n c o m b i n a t i o n . T h e y are realized
after the entrance of the bass or o n passing tones and illustrate a
decorative progression of one or more parts. T h e passing dissonances that are created by t h e m do n o t r e q u i r e r e s o l u t i o n . T h e
r e m a i n i n g tones of the preceding c h o r d are held. A l l of the followi n g examples take a four-part accompaniment except the last four,
w h i c h are realized i n three parts. T h e signatures that are based o n
the decorative progressions appear below the notes; the usual signatures have been placed above. Great care must be taken i n the use of
these h a r m o n i c nuances i n order to avoid a h a m p e r i n g or obscuring
of the p r i n c i p a l part.
Figure 455

\ *

6.

17

5b

7 7

|l 2

'5

3 4

T h i s sentence appeared in the first ed. as a footnote.

Mi
H

4 |6
1

A C C O M r A N 1 M E N

IMITATION

1. I m i t a t i o n is often used i n passages w h i c h are varied o n their

r e p e t i t i o n . T h e accompanist must participate i n the v a r i a t i o n so

A C C O M I' A N l M E N I

// C C O M r .1 N I M E N T

that his i m i t a d o r ] w i l l remain clear and lose nonc ol its attractiveness. T h e accompaniment must be patterned as closely as possible
after the leading part (Figure 456).

he must deny himself the delighls ol variation and play the notes
simply. I n Figure \ the bass leads the i m i t a t i o n .
3. I f the accompanist has an incompetent leader w h o precedes
h i m w i t h inept or even w r o n g variations, he must choose the safest
way o u t and, again, play the notes only as w r i t t e o . H e thus frees
himself of c o m p l i c i t y , k n o w i n g w e l l that i t suffices to hear a poor
v a r i a t i o n b u t once.

Figure 456
Andante

ir

2. I n these i m i t a t i o n s leader and f o l l o w e r must stand i n cise

r a p p o r t and be f a m i l i a r w i t h each other's powers and inventiveness. Otherwise m u c h can be spoiled i n performance. Especially i m p o r t a n t is this to an accompanist w h o must lead an i m i t a t i o n , for he must k n o w h o w dependable his f o l l o w e r is i n the ways
of v a r i a t i o n . I f he lacks f u l l confidence i n the a b i l i t y of the latter
Figure 457
Andante

405

4. B u t i f the fellow-performer is sufficiently able and i n t e l l i g e n t ,

the accompanist may rouse h i m , as always by a "good accompaniment, b u t especially by his varied i m i t a t i o n s . A n d his inventive
leading or correct i m i t a t i n g may spark the fire that puts the other
i n a disposition w h i c h he may have lacked previously. However, the
accompanist w h o leads i n a v a r i a t i o n must a l l o w his f o l l o w e r sufficient freedom afterwards to imtate correctly. T h e b r i l l i a n t and
the p l a i n must be j u d i c i o u s l y alternated, the procedure b e i n g i n
general as indicated i n the last paragraph of Part I of this Essay. T h e
accompanist w h o leads an i m i t a t i o n must pay cise a t t e n t i o n to the
k i n d of notes, i f any, that appear simultaneously i n the p r i n c i p a l
part, so that he may fashion a v a r i a t i o n w h i c h is sufficiently d i f ferent f r o m i t . A n d at the conclusin of his v a r i a t i o n he must ret u r n to a simple accompaniment, so that the p r i n c i p a l part's i m i tation, especially i f i t is made of many notes, w i l l be distinct. I t is
just as w r o n g for b o t h to be noisy at the same t i m e as i t is for b o t h
to f a l l asleep. Henee, they should n o t play o u t of either ignorance
or malice. I n the first case they w i l l r u i n each other's p a r t u n i n tentionally; i n the second, i n t e n t i o n a l l y .
5. T h e keyboardist w h o plays i n an ensemble w i t h d u p l i c a t i n g
bass instruments must w i t h h o l d his v a r y i n g i f he is u n c e r t a i n that
the others w i l l f o l l o w h i m .
6.
I n some passages w h i c h may be accompanied i n thirds, i t is
possible to have the m i d d l e part particpate i n a varied i m i t a t i o n
w h i l e i t moves i n thirds w i t h the bass (Figure 458, Example a ) .
W h e n the p r i n c i p a l performer abandons the simple setting of b
1

Note that the preceding paragraphs have dealt exclusively with an imitation
specified by the composer, the point of the discussion being the ways in which accompanist and soloist can work hand in hand to vary the prescribed relationships.
In flf 6-7, however, the discussion centers on the ways in which middle and upper
parts may be constructed so that they form voluntary imitations of the principal part.
Figure 452 provides an additional example. Heinichen's discussion and illustration of
such techniques appear in Arnold, op. cit., pp. 383 ff.
1

4o6

ACCOMPANIMENT

i n favor of c, neither the bass or the i n n e r part can imtate exactly.

I t suffices i n such a case to contrive an i m i t a t i o n whose figuration is
r h y t h m i c a l l y . the same as that of the leading part w h i l e the bass retains its o r i g i n a l , or slightly varied, tones.

A C C O M P A N I M E N T

4"7

figure 439"

Figure 458

SOME

7. Those w h o have a good knowledge of voice leading may also

substitute f o r the usual accompaniment a m i d d l e part i n decorative i m i t a t i o n of the p r i n c i p a l part. Passages w h i c h express successive chords of the seventh or six-five chords over a leaping bass
l e n d themselves best to this purpose. T h e tempo, however, must
n o t be very r a p i d , for i f i t is there w i l l be a loss of clarity (Figure
459)-

PRECAUTIONS

OF

ACCOMPANIMENT

1. Because most of this subject matter has already been treated

i n various appropriate places, there r e m a i n only a few a d d i t i o n a l
observations o n certain cases w h i c h cali for caution o n the part of
the accompanist.
2 T h e meaning of these examples is not immdiately apparent. Note that each
long slur endoses three identical bass parts, but that each upper part is different
from the others. T h e first two bars under each slur represent a bass and principal
part as written. T h e next two show the soloist's variation, and the last two the
accompanist's imitative accompaniment. I n the last two bars under each slur, the first
f in the upper part is the cued-in initial tone of the soloist. Henee it is not to be
played by the accompanist. T h e figure was apparently misunderstood by Arnold (cf.
Arnold, op. cit., p. 387).

A C CO

.f<>S

MPAN

A ; C O AI r A N I A E N T

I M li N T

2. W h e n the p r i n c i p a l part i n Figure \(o, Example a, lies above

the accompaniment, 7 (i should be taken over b instead of 6 alone
i n order to avoid fifths. I n b o m i t the sixth above e and play instead

ing t r i l l . T h e accompanist t l i e t c l o i e ceases at the same t i m e as the

p r i n c i p a l performer."
Figure 460

better

9 8

9 f o l l o w e d by 8 over the d o t . I n c the sixth over a is also o m i t t e d , a

3

b.

d o u b l e d t h i r d and fifth b e i n g taken i n its place. I n b o t h examples,

b and c, fifths w o u l d be created i f the accompaniment as d i s t r i b u t e d
i n the example f o l l o w e d the signatures. A similar example appears
i n F i g u r e 380. A l l progressions that c o n t a i n successive fo urths are
dangerous w h e n the d i s t r i b u t i o n of tones is changed. O f course,
the best a n d safest d i s t r i b u t i o n should be taken whenever possible;
b u t at times this cannot be done. Such b e i n g the case, the accompanist should t r y to remove the danger and reach another d i s t r i b u t i o n . T h i s may be done by d o u b l i n g a consonant tone i n order to
crate a fifth part, or by repeating the c h o r d i f the bass note is l o n g
or comprises one or more passing tones. B u t i f none of these expedients can be employed, the signature that causes the t r o u b l e must
be o m i t t e d . I n d i t is easy to repeat and change the disposition of
the c h o r d over the figuration of the bass, i f r e q u i r e d , as a means
of averting an error. I n e the second accompaniment is preferable
to the first w h e n the tones of the p r i n c i p a l part are n o t d u p l i c a t e d
i n the u p p e r part. I f the first accompaniment lies above the p r i n cipal part, w i t h w h i c h i t moves i n parallel m o t i o n , there w i l l be
fifths. I n /, where dissonances are resolved by the b r o k e n bass, the
accompaniment must be m o d i f i e d i n order to avoid ugly octaves o n
the d i v i d e d beats. Henee the dissonances i n the signature may be
o m i t t e d w i t h o u t scruple. I n g the extensin is o m i t t e d w h i c h is
o f t e n f o u n d i n the bass at the end of pieces after a p r o l o n g e d clos1

' rr

fr'rf

-E

Ms- aN

1
r

-4r#K-^

Ai
7
6

4 6

6
6

pr

fd.

1 J 11
frr
1

111

J l l l

. 1

1 It is to be gathered from this statement that the context of these examples is

such that the accompanist finds himself obliged to play the opening chords in the
notated distribution. T h e Bibliothek
der schnen
Wissenschaften
. . . (1763), unwilling to make this assumption, writes of example b: "Would not this example be
just as melodic if the octave of the first bass note were played by the right hand in
the uppermost part, to be followed by c, the sixth of the bass e, and then b over g?
T h e other two parts which are the upper parts of the example would retain their
places as middle parts. T h e harmony would be fuller and more agreeable without
an unnecessary doubling of the third [over e]." I n fact the same holds true when the
first note in the top part is /. Obviously, Bach knew this.
2 There will also be fifths if the second accompaniment lies above the principal
part, for they are caused by the accompanying part that moves from b to a, present
in both illustrations. I n any event, J . S. Bach was not quite so fastidious as his son
in the matter of superficial fifths. See, for example, the fifths between bass and soprano
in the chrale, Jesu, der du selbst so wohl, bar 15, caused by the action of a changing
note in the bass.

3.
sonus

Resolution of a dissonance can be disregarded w h e n uniis prescribed, since the unisn must be played by b o t h hands.

3 T h i s refers to the arpeggio in sixteenth notes. It is to be assumed that the

reason for this abridgment is the omission of the customary trill or other elaboration
in the principal part.

io

ACCOMPA

A C C O M P A N 1 M E N T

N I M E N T

A practiced ear w i l l replace the suppressed resolution w i t h an

imaginary one.
4. I t is rather usual, though not exactly necessary, that the beginnings of pieces should sound confused. I n the interests of orderliness a n d a precise start, i t is customary for even the most experienced performers to show one another the k i n d of notes each
has i n the o p e n i n g bars of a piece. B u t because there is o f t e n n o opp o r t u n i t y to examine parts for this purpose, a n d because a single
piece may contain many kinds of tempo, i t w o u l d be good to ntate
at least the b e g i n n i n g of the p r i n c i p a l p a r t i n small notes over the
bass. T h i s precaution m i g h t also be extended to passages w h i c h
succeed general rests a n d fermate,
especially w h e n the bass does
not resume simultaneously w i t h the p r i n c i p a l part.
THE

NEED

FOR FIGURED

BASSES

1. W e learned i n the I n t r o d u c t i o n to Part I I that even w h e n

a bass is figured as i t should be, a good accompaniment comprises
many a d d i t i o n a l factors. T h i s alone exposes the ridiculousness of
the d e m a n d that accompaniments be realized f r o m u n f i g u r e d basses,
and makes evident the i m p o s s i b i l i t y of fashioning even a passable
accompaniment. I n recent times there has been a m a r k e d tendency
to ntate short, essential ornaments a n d signs that p e r t a i n to good
performance. I f o n l y there were a corresponding decrease i n u n figured basses! I f only keyboardists were less w i l l i n g to do everyt h i n g demanded of t h e m ! O t h e r ripienists can c o m p l a i n w h e n they
are given an incorrectly w r i t t e n part; b u t accompanists must be
satisfied w i t h a part that is either u n f i g u r e d or so sparsely figured
that the signatures appear o n l y over those notes whose chords are
self-evident. I n short: I t is u n j u s t to expect an accompanist to learn
t h o r o u g h bass b o t h w i t h a n d w i t h o u t figures.
i T h i s section expresses strong disagreement with attempts by many theorists to
formlate rules for the reading of unfigured basses. Many of these writings were
superficial, transitory affairs that were forgotten almost as soon as they appeared, but
others carne from the principal theorists of the thorough-bass period, starting with
Viadana and continuing later with Francesco Gasparini (L'Armnico
pratico al cmbalo, 1708), Rameau (Traite
de l'harmonie reduite ses principes
naturels,
172a),
and Heinichen (Der General Bass, 1728). These are discussed by Arnold in his Art
of Accompaniment
from a Thorough-Bass,
pp. 65 ff. and 265 ff. T o the list may be
added the dissident Bach on his own admission (f 2 here) and reluctant promise (Ch.
I V , "Intervals and their Signatures," f f 2-3), largely unfulfilled. Marpurg (op. cit.,
April 29, 1749) reached conclusions similar to Bach's in answering an inquiry from
a young lady.

411

2.
Some have gone to gtrat t r o u b l e to systematize the realizat i o n of u n f i g u r e d basses, and I cannot deny that I have u n d e r t a k e n
s i m i l a r experiments. B u t the more I have t h o u g h t about i t , the
richer have I f o u n d harmonic usages. A n d these are increasing to
such an extent, what w i t h the fineness of o u r tastes, that i t is impossible to formlate h a r d and fast rules w h i c h w i l l at once shackle
free creations a n d enable one to surmise the o p t i o n a l twists of a
composer to w h o m b o u n t i f u l n a t u r e has granted a glimpse of the
i n e x h a u s t i b i l i t y of the art. Even g r a n t i n g that some f o r m u l a t i n g is
possible, are we to rack o u r memories i n l e a r n i n g rules w h i c h by
their nature must be numerous a n d n o t always valid? A n d h a v i n g
finally learned the given rules, are we t h e n to squander endless t i m e
and energy o n the mastering of exceptions? Even i f we d i d a l l of
this, the results w o u l d be of only small valu, f o r the ablest musiran can err w h e n presented w i t h o n l y one alternative, let alone
several.
3. A n d so i t remains i r r e f u t a b l e that a correctly figured bass is
an indispensable a d j u n c t to the good performance of a piece'. T h e
composer w h o wants his works to be played as w e l l as possible must
take every step to reach this end. H i s n o t a t i o n must e x p l a i n everyt h i n g w i t h such c l a r i t y that each detail w i l l be understood. T h i s is
the very least t h a t can be demanded, f o r we have stated repeatedly
that exact i n d i c a t i o n of an accompaniment includes something
m o r e t h a n signatures. W e have even shown that there is s t i l l a lack
of signs f o r certain situations. H e r e we have clear p r o o f that an acc o m p a n i m e n t made f r o m n o indications at a l l can only be poor. W e
already have signatures; let us use t h e m a n d t o r t u r e neither ourselves i n the f o r m u l a t i o n of inadequate rules, or o u r students i n
the l e a r n i n g of t h e m . Those w h o are too i n d o l e n t or i g n o r a n t to
figure their basses as r e q u i r e d f o r a good effect s h o u l d let an able
accompanist do i t f o r t h e m .
4. O f course signatures are n o t needed f o r every smallest detail,
and a figured bass should n o t be made i n t o a solo. Nevertheless,
n o t h i n g that is necessary a n d essential should be overlooked. M a n y
are too sparing i n t h e i r use of figures because they w a n t to a v o i d
an o v e r t a x i n g of the accompanist's eyes. B u t an experienced accompanist can easily sean basses w h i c h c o n t a i n m o r e indications
t h a n are usually given, for, l o n g before his study of accompaniment,
he h a d to read i n t w o staves c o n t a i n i n g many notes, accidentis, a n d

42

A C C O M /' A N l M E N T

ACCOMPANIMENT

other superimposed characters. W h i c h is easici l o read, this latter

w i t h its web of attendant difliculties, or three and, at the most, four
figures, one over the other, w h i c h he must learn anyway i n studyi n g t h o r o u g h bass, w h i c h arise constantly i n active accompanying,
a n d w h i c h , consequently, cannot be as fearf ul as many an i n d o l e n t
accompanist seems to believe?

413

8. Passing iones appeai i n slrpwisc and leaping contexts, a n d

are often rather r a p i d . W h e n ihcy occur singly they are not i n d i cated. O f the stepwise and leaping notes i n Example a, F i g u r e 463,
those o n the d i v i d e d beats are passing. T h i s r u l e is acceptable:
Stepwise passing tones may not be f o l l o w e d by a leap, a n d the
octave of a disjunct passing tone must lie previously i n the r i g h t
h a n d . T h u s , each note i n b is accompanied, f o r these conditions
are n o t present. W h e n a bass note, instead of m o v i n g d o w n a
second ( c ) , leaps to the seventh above ( d ) , the f o l l o w i n g tone may
be passing, even t h o u g h its octave is n o t previously played. Leaps
of an octave are n o t regarded here as leaps b u t as repetitions. W h e n
passing tones appear i n succession, the use of a dash is j u s t as necessary as w h e n slow notes pass (e).
2

PASSING TONES

1. I t is as necessary to indcate passing tones i n most cases as i t

is to indcate figures. B u t since figurists do n o t proceed w i t h adequate care, the accompanist must learn to pick o u t passing tones
(Figure 4 6 1 , Example a) t h r o u g h constant practice and attentive
listening. T h e y can be surmised at times f r o m a preceding c h o r d
w h i c h contains the bass notes that f o l l o w i t (b), and f r o m a r e q u i r e d
preparation a n d r e s o l u t i o n (a a n d c ) .
Figure 461
a.

7
b.

asJJ

c.

OJJM 1

5l>

Figure 463
6

6
6 5

rf

6
5

>r (VT-llr II rrTr f ' H - ^ ^ J C - T c J " II

r

11

2. I n d e ed , there are rules f o r the r e c o g n i t i o n of passing tones,

b u t they are n o t always dependable. Because these rules d o n o t provide a sol i d f o u n d a t i o n , because passing tones cannot always be surmised, and, finally, because i t is generally acknowledged that there
are far more poor t h a n good accompanists, i t w o u l d be safest to provide exact indications; i n fact, too many rather t h a n too few. A
good accompanist is n o t confused by an a d d i t i o n a l dash, a n d beginners are m u c h helped by i t . I t is a t r i b u t e to the French to assert that
they desgnate passing tones w i t h great diligence. Usually, they
employ a diagonal stroke (Figure 462).
Figure 462

g JJ I

1 Von durchgehenden
Noten. Cf. C h . I V , "Intervals and their Signatures," f f 69 ff.
and Note 9. A n exhaustive study of the accompaniment to passing tones in the bass
appears in Arnold, op. cit., Chs. X V I I I and X I X . C h , X V I I I is concerned especially
with Heinichen's rules to which Bach is alluding at the beginning of f 2 here.

c.

d.

6 e.

"ME

rjrij^ri^igtfcri ^ JJ
j

4. T h e f o l l o w i n g bass notes usually take a c h o r d apiece; half

notes i n an alia breve a n d a fast three-two whose most r a p i d notes
are e i g h t h ; quarter notes i n a slow three-two, i n the so-called simple
meters f r o m allegretto ( w i t h n o t h i n g faster t h a n thirty-seconds) to
T h e term "passing tone," i n the general sense in which it is used here, includes
the neighboring note, complete and incomplete. T h e incomplete neighbor (chappe)
or the incomplete double neighbor would require an extensin of the rule here
stated, for it is precisely a "passing tone" that is followed by leap. Examples from
Heinichen are quoted by Arnold in the reference given in Note i here. For the rest,
it is of interest to note that the neighbor appears only once in the present section
(Figure 464, example c) without its being mentioned in the text. However, it is clear
that Bach included this element among his "passing tones." For an example see Figure 280, first illustration, and the accompanying text.
s I n other words, the leap must be to a tone in the prevailing chord.
* Bach included this clause because he had just stated that the octave of a disjunct
passing tone must lie previously in the right hand. T h e case under discussion.here is,
of course, not to be regarded as a type of disjunct passing tone, but as a special kind
of conjunct passing tone.
2

A C C O M I' A N I M I; N T

ACCOMPANIMENT

presto, a n d i n three-four and six-four meters i n r a p i d tempos;

eighth notes i n a f o u r - f o u r meter f r o m adagio to allegretto, and i n a
slow three-eight, six-eight, nine-eight, twelve-eight, three-four, a n d
six-four. W h e n these latter meters appear i n a r a p i d tempo every
g r o u p of three eighth or quarter notes has its o w n accompaniment.
5. Notes that seem to take an accompaniment, b u t are i n fact
passing, must, of a l l notes, carry a dash. Notes that l o o k l i k e passi n g tones, b u t are i n fact accompanied, must be figured. A incorrect i n d i c a t i o n of the f o r m e r is a m u c h greater risk t h a n the latter.
6. T h e use of thirds alone to accompany passing tones has been
discussed i n the section "Some Refinements of A c c o m p a n i m e n t . "
7. Various a d d i t i o n a l remarks r e m a i n to be made o n the accomp a n i m e n t to repeated tones. These remarks w i l l be based o n tempos
as they are p e r f o r m e d here, where adagio is far slower and allegro
far faster t h a n is customary elsewhere.
8. F r o m the slowest tempo to largo, quarter notes a n d greater
are played by b o t h hands a n d h e l d f u l l y . E i g h t h notes are played
s i m i l a r l y b u t h e l d f o r only half of t h e i r length. A l l sixteenths are
played by the left h a n d and, i n the absence o a staccato sign, are
f u l l y h e l d . T h e r i g h t h a n d accompanies these sixteenths a n d faster
notes w i t h eighths w h i c h are h e l d for half of t h e i r l e n g t h , prov i d e d that the expression does n o t r e q u i r e a different execution.
W h e n the bass has a c o n t i n u a l flow or, at least, great numbers of
thirty-seconds or faster notes, the left h a n d may o m i t one or m o r e
notes, p r o v i d e d that there is an accompanying bass i n s t r u m e n t . I f
there is none, the keyboardist alone must s u b m i t to the tortures of
this t r e m b l i n g m o t i o n . T h e r i g h t h a n d comes i n o n only the first
note of every t r i p l e t of one or t w o beams. T h e same procedure holds
for each g r o u p of three e i g h t h notes or the equivalent i n threeeight, six-eight, nine-eight, a n d twelve-eight.
5

9. F r o m larghetto a n d andante to allegro the r i g h t h a n d plays

f u l l y h e l d quarters w h e n the bass expresses quarters, eighths, faster
notes, a n d triplets. Longer notes are f u l l y h e l d by b o t h hands.
10. I n a siciliana, be i t fast or slow, quarter notes a n d longer
are played a n d h e l d by b o t h hands. T h e single eighths w h i c h f o l low the quarters are also accompanied by the r i g h t h a n d . I n a l l
M n f f 6-8.
I.e., i n Berln.
i As discussed at length in Pt. I , Introduction, f ga.

i$

other cases, regardless of the ( onsiriu t i o n of the bass, the r i g h t hand

plays only once for each g r o u p ol l i n c e eighth notes or the equivalent.
11.
F r o m allegro assai to prest issimo the r i g h t h a n d plays either
f u l l y h e l d half bars or hall-held quarters to an eighth-note bass.
Quarters are struck by both hands and h e l d for half their l e n g t h ,
and longer notes are f u l l y held. For the rest, I refer the reader to
Paragraph ga of the I n t r o d u c t i o n to Part I .
12. These remarks h o l d o n l y i n so far as a change is n o t called
for by signatures or signs of performance.
13. W h e n t r a n s i t i o n a l passages cannot be accompanied by
thirds or other o r n a m e n t a l relationships, as discussed i n the section
"Some Refinements of A c c o m p a n i m e n t , " they should be allowed
to pass. Extensions at the end of a piece are to be treated s i m i l a r l y .
14. T h e examples of Figure 464 w i l l serve to conclude this sect i o n . I n a, an unusual k i n d of expression (which may be revealed
by the content of a piece or the construction of the r i p i e n o parts)
sometimes requires a c h o r d o n each note instead of the usual omission of a seprate c h o r d f r o m the short notes. T h i s oceurs f r e q u e n t l y
i n accompanied recitatives. I n b, the composer w h o wants a c h o r d
to be struck o n the note e because of the expression, must place the
signature 6 over i t . I n c, the passing tones must be accompanied i n
order to avoid errors. One p a r t may interchange its tones w i t h
those of the bass, or the entire c h o r d may be repeated. I n d, preparat i o n of the seventh above /-sharp requires an accompaniment to the
passing e i n the f o r m of a r e d i s t r i b u t e d c h o r d . I n e, the second c h o r d
of the bar is repeated over the passing bass as a means of a v o i d i n g
too great a descent a n d averting fifths. T h e r e p e t i t i o n i n / accomplishes several ends: i t avoids fifths w h e n the t h i r d of the preceding

a is d o u b l e d ; i t avoids octaves w h e n 3 is taken over this a; a n d i t

6

helps to r e t a i n the preceding register. I n g, where i t is presupposed

that the notated d o u b l i n g of the sixth, or a d o u b l e d t h i r d , is req u i r e d over the first f , the c h o r d of the sixth must be repeated, u n d o u b l e d , over the passing tones. T h i s enables the g-sharp to be
accompanied w i t h o u t an i m p u r e or a w k w a r d progression, a n d w i t h out descending to a lower d i s t r i b u t i o n . I n I t a l i a n bel canto, singers
* T h e octaves would come from the strong beat progression, a to g, bass and middle voice.

416

A CCO

MPA

// C C O Af /' ANIMEN

NI M I : N l

announce the e n d of a h o l d by rising and then l a l l i n g a half or

whole step according t o the s i t u a t i o n , w i t h o u t any slightest trace of
an i n d i c a t i o n . Example h illustrates the n o t a t i o n , and i the execut i o n of such a h o l d , a l t h o u g h the latter is occasionally w r i t t e n o u t .
T h i s refinement is unaccompanied, the ch or d being struck b u t once
i n order t o preserve the clarity of the p r i n c i p a l part's rise a n d f a l l .
T w o such cases are illustrated i n / w i t h t h e i r accompaniments. T h e
figuring
of these examples, as i n k, is incorrect. T h e most experienced accompanist is Hable t o err i f the composer does n o t
figure w i t h sufficient aecuracy; especially w h e n he fails t o indcate
passing tones i n passages w h i c h seem to c o n t a i n a r e s o l u t i o n ('); a n d
similarly, w h e n he fails t o figure those notes w h i c h look l i k e passing
tones b u t r e q u i r e their o w n chords (m). T h e accompanist i n such
cases w i l l be free f r o m blame i f , i n a d d i t i o n , the p r i n c i p a l p a r t is
n o t notated over the bass part. T h i s was stated earlier i n Paragraph
5. A good precautionary use of figures over passing tones is illustrated i n n. T h e average accompanist is thereby l e d unmistakably
to recognize the necessity of a l t e r i n g his d o u b l i n g i n order to avoid
errors. I n o, a dash is needed as a means of i n f o r m i n g the accompanist that instead of repeating the foregoing c h o r d he must play
the t r i a d of the f o l l o w i n g tone o n the rest.
Figure 464

, 5

a.

ey

- A *

i 11
f

H
6

ftl II

0J

IT t

i -

4
3

'

til

II

f
4
3

11 ^

J-j-ip

4-

^ 4
- rrr'r rr' rr
ei

4
5

=r=H1

6 5

/ -

o.

7 6

1s -

r r r ' r r r ' r f4 ^

nrfcf
1

It

11

m m

6 5

* *

5l>

II J ^ ^ q

F*h

J6
6

-rl

r " u n f "f r r rr r r r r

11
II t

Err" t
"-

TT-*

4*7

Allegro

rr

i-

II

4iS

A C C O M P A N I M E N T

ACCOMPANIMENT
CHORDS

THAT

PRECEDE

THEIR

BASS N O T E S

1. I t is often necessary to strike chords over short rests i n advance of their bass notes, as a means of r e t a i n i n g order a n d w i n n i n g
variety.
2. Some figurists f o l l o w the commendable practice of i n d i cating such chords by placing over the rest the signature or dash
that pertains to the f o l l o w i n g bass note. I t w o u l d be excellent i f
everyone adopted such an exact means of i n d i c a t i o n , for i t w o u l d
l i g h t e n the tasks of many accompanists.
3. I n the absence of p r o p e r signs, t w o observations can be made:
first, the rests discussed i n this section are n o t greater t h a n a sixteenth rest i n an allegretto; second, parts that enter o n the rest
must agree w i t h the tones of the anticipated c h o r d . T h e examples
that f o l l o w w i l l clarify m y meaning.
4. I n order to catch the beat w i t h certainty, a beginner is allowed to strike the C major t r i a d over the rest i n F i g u r e 465, Example a, b u t an experienced accompanist w i l l let the rest and the
first c pass, and wait u n t i l e appears before p l a y i n g a c h o r d of the
sixth i n the r i g h t h a n d . I n b, there is no alternative to t a k i n g a
c h o r d over the rest, unless half the bar is a l l o w e d to pass unaccompanied. I n a fast tempo this means of establishing the beat is as
m u c h needed by the p r i n c i p a l part as by the accompanist. T h e
r i g h t h a n d may enter after the rest only w h e n the tempo is n o faster
than andante, for otherwise confusin of beats m i g h t be caused by
i t . I n c, regardless of the tempo, the c h o r d may n o t be struck before
the entrance of the first bass note, since the / i n the p r i n c i p a l part
does n o t harmonize w i t h i t . I n d, the chords must be played i n an
eighth-note r h y t h m , even i n a slow tempo, i n view of the stationary
p r i n c i p a l part and the syncopated bass. B u t the first eighth of the
bar may be allowed to pass i n order n o t to obscure the usual soft
b e g i n n i n g of h e l d notes. I n e, a c h o r d over the rest is indispensable, especially w h e n this example appears i n a piece f o r a
large, heavily d o u b l e d orchestra i n w h i c h a l l instruments enter o n
the r a p i d notes. T h i s s i t u a t i o n is c o m m o n l y f o u n d i n operas, where
i t occurs i n dramatic, accompanied recitatives, w i t h singers w h o ,
because of the constant, vigorous action, may be d e c l a i m i n g u p stage, downstage, at the sides, or i n the center, w i t h a d d i t i o n a l
1 Von dem

Vorschlagen

mit der rechten

Hand.

419

noises to boot. H e r e the accompanist must take the lead and, o n

the short rest, give his ene w i t h as heavy an attack as possible. I n f ,
where the notes w h i c h follow the rest are n o t as r a p i d as those i n the
previous example (although again a l l parts enter i n unisn f o l l o w i n g a general pause), the c h o r d is n o t played. I n g w i t h its d o t t e d
notes the accompanist again does n o t play his c h o r d i n advance, f o r
there is an appoggiatura i n the p r i n c i p a l part w h i c h resolves o n l y
after the first short note i n the bass has been played. A n t i c i p a t o r y
chords are n o t employed over bombastic basses i n the French manner, f o r their use w o u l d r o b the passage of its resolute nature (h).
Example i illustrates a case i n w h i c h the composer has placed a
certain stress on the p r i n c i p a l part, a n d consequently wishes to have
this part alone i n t r o d u c e each h a r m o n i c change. T h e accompanim e n t is appended to the example. I n ; the appoggiatura w i l l n o t
suffer a t r i a d o n the rest; henee, i t is best to w i t h h o l d the c h o r d u n t i l
the second eighth of the bar appears. I n k, chords are played over
the dots and t i e d notes.
Figure 465
Allejretto
a.

Presto

b.

2
5b
Presto

6 1,7
4 5b

Allegro
legro

e.

C J ' T

cur

A C C O MP

A N I M E N T

A C C O M I' A N I M E N T

421

progres.sions wcie considercd loo p l a i n . But today, thanks to o u r

i n t e l l i g e n l lasic, exc:ej)tioiial liai nionies are i n t r o d u c e d i n t o the
recitative only rarely, and tlien w i t h suflicient m o t i v a t i o n . I n sett i n g his chords to this present k i n d , the accompanist need no longer
sweat so profusely. Nevertheless, an exact figuring is s t i l l r e q u i r e d ,
even w h e n the p r i n c i p a l part is notated over the bass.
1

ki

tr
ms.

2. Some recitatives, i n w h i c h the bass and perhaps other instruments express a definite theme or a continuous m o t i o n w h i c h does
not particpate i n the singer's pauses, must be p e r f o r m e d strictly i n
t i m e for the sake of good order. Others are declaimed now slowly,
now r a p i d l y according to the conten, regardless of the meter, even
t h o u g h t h e i r n o t a t i o n be barred. I n b o t h cases, especially the latter,
an accompanist must be w a t c h f u l . H e must listen constantly t o the
p r i n c i p a l performer, and w h e n there is action, watch h i m as w e l l , so
that his accompaniment w i l l always be ready; he must never desert
the singer.

Allegro

crtrf
6 6

t;

y JI

- 6
5

THE

RECITATIVE

1. N o t so l o n g ago, recitatives used to be crammed w i t h endless

chords, resolutions, and enharmonic changes. A special k i n d of
beauty was sought i n these harmonio extravagances, w i t h o u t there
b e i n g the slightest excuse for t h e i r employment. N a t u r a l h a r m o n i o

3. W h e n the declamation is r a p i d , the chords must be ready i n stantly, especially at pauses i n the p r i n c i p a l part where the c h o r d
precedes a f o l l o w i n g entrance. A t the t e r m i n a t i o n of a chord, its
successor must be struck w i t h dispatch. T h u s the singer w i l l n o t
be hampered i n his affects or t h e i r requisite fast execution, for he
w i l l always k n o w i n good t i m e the course and construction of the
h a r m o n y . W e r e i t necessary to choose between t w o evils, i t w o u l d
be preferable to hasten rather than to delay. Indeed, the better is
always better. A r p e g g i a t i o n must always be w i t h h e l d f r o m r a p i d
declamation, especially w h e n there are f r e q u e n t chordal changes.
For one t h i n g , there is no t i m e for i t , and even i f there were, i t
m i g h t very easily lead accompanist, singer, and audience i n t o confusin. F u r t h e r m o r e , arpeggiation is n o t r e q u i r e d here, for it finds
its n a t u r a l e m p l o y m e n t i n q u i t e different situations, i n slow recitatives and sustained chords. I n such cases i t serves to r e m i n d the
singer that he is to r e m a i n i n a given chord, and prevens h i m f r o m
l o s i n g the p i t c h because of the length of the c h o r d , or f r o m assumi n g that the c h o r d has changed. These fiery recitatives o f t e n occur
i n operas where the orchestra has a w i d e range w i t h Dasses p l a y i n g
divisi, w h i l e the singer declaims upstage, far removed f r o m his ac1 T h i s refers to the type of recitative which is discussed in Heinichen, op. cit., Pt.
I I , C h . I I I , pp. 769 ff. He writes: " I t is generally known that the recitative, unlike
all other styles, has no regular key, but rather casts its tones quite irregularly, moving
abruptly and without order forwards and backwards to the most remote keys."

22

A C CO

MPANIM

A C C O MPANIM

/'. N I

companiment. Such being the case, the first harpsichordist, w h e n

there are t w o , does n o t await the t e r m i n a l i o n o the singer's cadenees, b u t strikes o n the final syllable the chord w h i c h should
r i g h t l y be played later. T h i s is done so that the r e m a i n i n g basses or
other instruments w i l l be prepared to enter o n t i m e .
4. T h e pace w i t h w h i c h a c h o r d is arpeggiated depends o n the
t e m p o and content of a recitative. T h e slower and more affettuoso
the latter is, the slower the arpeggiation. Recitatives w i t h sustained
accompanying instruments are w e l l adapted to arpeggiation. B u t as
soon as the accompaniment shifts f r o m sustained to short, detached
notes, the accompanist must play detached, resolute chords, u n arpeggiated, and f u l l y grasped by b o t h hands. Even i f the score
expresses t i e d w h i t e notes, the sharply detached execution is retained. A heavy attack is most necessary i n the theater w i t h its
memorized recitatives, because of distance. O f course, the accompanist must also play q u i t e softly at times i n the theater, b u t more
so i n the c h u r c h or the saln, where noisy, f u r i o u s recitatives are
not q u i t e at home. I n recitatives, as m u c h as elsewhere, chords
must be expressed i n proper v o l u m e .
5. I n recitatives w i t h sustained accompanying instruments, the
organ holds only the bass, the chords being q u i t t e d soon after they
are struck. Organs are seldom p u r e l y t u n e d , w i t h the result that
h e l d chords, w h i c h are often chromatic i n such recitatives, w o u l d
sound ugly and disagree w i t h the other accompanying instruments.
I t is often difficult i n such a case to make an orchestrawhich
need n o t be the most wretchedsound i n p i t c h . A r p e g g i a t i o n is
not employed at the organ. Other keyboard instruments d o n o t use
ornaments or refinements, aside f r o m arpeggiation, i n the accompaniments to recitatives.
6. I n intermezzos and comic operas w i t h m u c h noisy action,
and i n other works for the theater where the action often oceurs
backstage, constant or f r e q u e n t arpeggiation must be resorted to,
so that the singer and accompanist w i l l hear each other clearly at a l l
times. W h e n the sense of the words or an i n t e r v e n i n g action delays
the entrance of the singer after the preparatory c h o r d has been
played, the accompanist must repeat the chord, b r o k e n slowly upw a r d , u n t i l he observes that the d e c l a i m i n g has resumed. Unless i t is
urgently r e q u i r e d , neither too l i t t l e or too m u c h u n f i l l e d space
should be allowed i n the accompaniment. W h e n recitatives are ac-

/<; N T

4*3

companied in detached lashion l>y instruments other than those that

p e r f o r m the Iwss part, incidental harmonic modifications, such as
8 [>7 or 6 b>5, must be played softly or o m i t t e d by the keyboardist
w h e n they are indicated solely over the bass part and, as is often
the case, appear i n succession. T h u s , the p r i n c i p a l part w i l l n o t be
overaccompanied and the other instrumentalists w i l l hear the
singer more clearly, w i t h the result that they w i l l be able to direct
a t t e n t i o n to t h e i r subsequent entrance. T h e harpsichord, w h i c h
sounds l o u d to those cise to i t , especially w h e n i t has a penetrating
tone, can easily d i s r u p t orderliness. A n d sometimes this suppression
of tones o n the part of the keyboardist adds to the impressiveness of
words that the composer wishes the singer to recite w h i l e the instruments, for good reasons, r e m a i n silent. W h e n there is vigorous act i o n upstage, this precaution is even more essential, for the singer's
tones w i l l often pass unheard over the orchestra p i t , w h i c h is r i g h t l y
constructed o n a lower level t h a n the parterre.
7. W h e n a singer departs f r o m the w r i t t e n notes, i t is better to
strike a f u l l c h o r d repeatedly than to play isolated tones. I n recitatives, correct h a r m o n y is the p r i m a r y factor; henee singers should
not be expected to sing only the w r i t t e n notes and no others, especially i n i n d i f f e r e n t passages. I t suffices i f they declaim w i t h i n the
confines of the proper c h o r d . O f course, a single tone may be
struck i n the case of a remote m o d u l a t i o n . I f the singer is of sufficient a b i l i t y , there is no need for a l a r m w h e n he chooses to sing
Example a of Figure 466 i n the manner of Example 1 or 2. Causes
for such changes may be a desire to f i n d a convenient register, or
Figure 466

* J>J>
6

o-

424

A ; C

// C C O M I' A N I M E N T

A N I M E N I

simply forgetfulncss. I n m e m o r i z i n g their parts, singers often confuse the many similar patterns of recitatives, for (hey are more i m pressed by the u n d e r l y i n g harmony than by the melody. I w o u l d
be less apt to forgive an accompanist w h o hesitated over a modificat i o n of a, t h a n I w o u l d one w h o was startled by b, a case that arises
occasionally, where signatures m i g h t be lacking, the tempo m i g h t
be r a p i d , a n d half of the passage, perhaps d i r e c t l y after the beginn i n g , m i g h t be w r i t t e n o n a new stave.
8. O n the last arpeggiation of a preparatory c h o r d i t is wise
to place i n the upper part the singer's i n i t i a l tone. T h u s placed i t
w i l l be most clearly heard and thereby ease the singer's task. Rather
t h a n abandon such an expedient, i t w o u l d be better to tolrate cert a i n irregularities w h e n they cannot be avoided, such as d i s r u p t i n g
the p r e p a r a t i o n of a dissonance, or placing its r e s o l u t i o n i n the
w r o n g part, the a i m b e i n g simply to reach the r e q u i r e d d i s t r i b u t i o n q u i c k l y . However, i t is o f t e n easy to do this by means of a r a p i d
arpeggio w i t h o u t i n d u l g i n g i n such liberties.
9. W h e n , i n a recitative w i t h accompanying instruments, the
bass enters ahead of the other performers after a cadenee or pause,
the keyboardist must strike his c h o r d a n d bass note strictly o n t i m e ,
w i t h a sure, f u l l attack, especially w h e n the orchestra is large (Figu r e 467, Example ). However, i f a l l the instruments attack simultaneously, the keyboardist does n o t anticpate, b u t signis w i t h his
head or body i n good t i m e so that a l l w i l l enter together (b). I n
Example c, a six-four c h o r d is r e q u i r e d over the first bass note,

425

preferably w i t h the octave on l o p . A l (lie rest, the seventh and l i f t h

of the same bass note are played. I'inally, 1 refer my reader to Figure 465, Kxample c, and ihe acooiupanying text. M a n y similar examples can be deduced f r o m this one.
2

CHANGING

NOTES

1. T h e m e a n i n g of changing notes or i r r e g u l a r passing tones

has been stated i n Chapter I V , Paragraphs 73 to 78. T h e i r indicat i o n is most essential, for beginners i n t h o r o u g h bass cannot easily
surmise t h e i r presence.
2. Some figurists place signatures over the note against w h i c h
the c h o r d is struck; others over the f o l l o w i n g note. T h e first p r o cedure is n o t bad, especially w h e n the signatures are f a m i l i a r ones
(Figure 468, E x a m p l e a), and a m b i g u i t y can be thereby avoided
(b). B u t aside f r o m these considerations changing notes that are
i d e n t i f i e d by means of an o b l i q u e stroke make for simpler signatures, and the accompanist is spared his hesitaney over the unusual
successions w h i c h are o f t e n expressed by the other m e t h o d of i n d i cation. Nevertheless, i t is advisable that the student of t h o r o u g h
bass become completely f a m i l i a r w i t h the figures, f o r b o t h methods
of indication/are s t i l l i n use.
Figure 468

5
a.

4
2

7
a.

a.

6
5 # 5

g
5

/ 3

y JTJ"JIIJ jgppgjggig

instead o f
3. I r r e g u l a r passing tones are to be regarded as appoggiaturas
that have been w r i t t e n o u t a n d g i v e n an exact l e n g t h . Against
these appoggiaturas the r i g h t h a n d plays the c h o r d that pertains to
the f o l l o w i n g bass note. Henee i f a tone that is n o r m a l l y consonant
forms a dissonance against the changing note w i t h w h i c h i t is struck,
i t retains its o r i g i n a l freedom and character. I t may be d o u b l e d
(Figure 469, E x a m p l e a), a n d i t requires neither p r e p a r a t i o n or
r e s o l u t i o n (b). S i m i l a r l y , dissonances do n o t forsake t h e i r basic
ways w h e n they are m o m e n t a r i l y made i n t o consonances b y the act i o n of a changing note i n the bass (c).
2 Cf., par ex., J . S. Bach's Werke, 11.2, pp. 164 ff.
See C h . V I , "Passing Notes," Note 1.
1

426

ACCOMPANIMENT

ACCOMPANIMENT

4 J

I '
/

'

I |{ I
1

i i

b.

~ 3

TT

4. I n the presence of changing notes a t r i a d cannot be denoted

by the absence of a signature; at least one of its figures must be
posted (Figure 470, E x a m p l e a). W h e n an o b l i q u e stroke is succeeded by chords w h i c h r e q u i r e d o u b l i n g (b), or may be taken i n
more than one way (c), the accompanist must make provisions i n
advance f o r correct construction, especially w h e n there is the poss i b i l i t y of c o m m i t t i n g an error (d).
Figure 470
. a .

I 5 b. V ,

.6

6 c. /r- /

98 r.

lent examplcs of such t h r m m which may w e l l serve as perfect

models. Such a theme should have a manly bearing w h i c h o n occasion may express tones f r o m other parts of its proper chords, by
b r e a k i n g to t h e m or by other means of melodic elaboration; b u t
w i t h o u t sacrifice of its essential nature. Its lineaments must n o t be
extravagant. Cadenees and caesurae must be basslike; at least, the
chords of the latter should sound n a t u r a l . Appoggiaturas should be
i n t r o d u c e d i n t o such a bass w i t h great caution, i n order n o t to dist u r b the flow of the harmony. Moreover, these and other melodic
complements are better left to the p r i n c i p a l part, for this latter
w o u l d be entirely too m u c h restricted, i f the bass were made to
share equally i n a l l garnishments. Rather, the bass should be accompanied by chords that express numerous, effective suspensions
w h i c h allow for the construction of a singing p r i n c i p a l part. Especially recommended are progressions w h i c h p e r m i t the use of many
seventh, five-four, six-five, and n i n t h chords, as i l l u s t r a t e d by the
p l a i n basses of Figure 471.
Figure

471

7.

7. 6 7 6

76

4. 3 4 3

r-^L*

43

-9*i
-A
-

BASS

427

6
9, 8 5
r.

9 8 5
G

6
1
1

11

Ffj

6
5 5

THEMES

1. Good bass themes, evolving n a t u r a l l y , are among the master

touches of composition. T h e famous Kapellrneister
T e l e m a n n and
G r a u n , along w i t h m y deceased father, have given us m a n y excel1

1 Of the three G r a u n brothers, two, Johann Gottlieb and Cari Heinrich, were active at the court of Frederick the Great. T h e former was Konzertmeister and the

latter, to whom Bach is referring, Kapellrneister. Since Bach refers to him as still
alive, or at least since his death in 1759 is not mentioned, it would seem that the
present section was written before this year.

428

A C C O M

A N

M E N I

2. Composers c o m i n i t t w o kinds ol error in < o n s t r u c t i n g bass

themes. A t times they want a shower of b e a u t i l u l song, a l t h o u g h i t
is i n a p p r o p r i a t e ; they proteed to w r i t e a linc melody, w h i c h , however, is h i g h and lacking i n bass progressions, and w h i c h itself allows for the a d d i t i o n of a good bass. A n experienced accompanist,
instead of seeking an upper melody to this k i n d , m i g h t find i t
easier to give the theme to the r i g h t h a n d and, i n the left, improvise
a bass that provides the proper h a r m o n i z a t i o n . T h e other k i n d of
error concerns themes w h i c h are too d r y . Here, the composer wishes
to avoid the faults m e n t i o n e d above and to provide every opport u n i t y for the elaboration of the p r i n c i p a l p a r t ; consequently he
writes a good, f o r t h r i g h t , simple, b u t inexpressive bass. However,
this latter k i n d has i n its favor the fact that i t permits an a d r o i t acc o m p a n i m e n t , whereas the f o r m e r very often defies the superimposition of chords.
3. Bass themes are p e r f o r m e d either i n unisn by a l l i n s t r u ments or by bass instruments alone. I n the first case the accompanist
omits chords and plays the w r i t t e n notes i n octaves w i t h b o t h hands.
B u t should the composer place signatures over the bass advisedly,
they must be realized. T h e reason is that the suspensions w h i c h may
be thereby i n t r o d u c e d need to be heard, for they w i l l n o t only n o t
obscure the theme b u t make i t more l u c i d . Some themes are so
constructed that an understanding listener is o n l y half satisfied i n
the absence of an accompaniment, for i n his i n n e r comprehension
the h a r m o n y is inseparable f r o m the tones that he hears. I n such
a case the organ provides the best accompaniment, n o t only because
of the suspensions b u t also because of its penetrating v o l u m e . T h e
second case m e n t i o n e d above, w h i c h is f o u n d i n vocal and i n s t r u m e n t a l pieces, requires a chordal accompaniment.
4. T h e r e are t w o accompaniments to bass themes, and they provide an able accompanist w i t h excellent o p p o r t u n i t i e s to display
his s k i l l . Those w h o possess an adequate knowledge of composition
a n d whose p r o p i t i o u s i n v e n t i v e faculties are tempered by good
j u d g m e n t may fashion an a d d i t i o n a l melody to be played by the
r i g h t h a n d instead of the usual accompaniment w h e n the p r i n c i p a l
part pauses or performs p l a i n , sustained notes. T h i s melody must
agree w i t h the content and affect of a piece and never hamper the
p r i n c i p a l part.
5.

B u t those w h o lack the a b i l i t y to do this should adhere to the

// C C O M I' A N I M E N T

429

prescribed harmony and p c r f o r m it in keeping w i t h the rules of

good performance, their attention being directed to the selection
of the best progressions and d i s t r i b u t i o n s and the construction of a
singing upper part.

IMPROVISATION

C H A P T E R

S E V E N

IMPROVISATION
THE

FREE

FANTASIA

1
F A N T A S I A is said to be free w h e n i t is unmeasured and
moves t h r o u g h more keys than is customary i n other pieces,
w h i c h are composed or improvised i n meter.
2. These latter r e q u i r e a comprehensive knowledge of composition, whereas the f o r m e r requires o n l y a t h o r o u g h understanding
of h a r m o n y nd acquaintance w i t h a few rules of construction. B o t h
cali for n a t u r a l talent, espedally the a b i l i t y to improvise. I t is q u i t e
possible for a person to have studied composition w i t h good success
and to have t u r n e d his pen to fine ends w i t h o u t his h a v i n g any g i f t
for i m p r o v i s a t i o n . B u t , o n the other hand, a good f u t u r e i n compos i t i o n can be assuredly predicted for anyone w h o can improvise,
p r o v i d e d that he-writes profusely and does n o t start too late.
3. A free fantasia consists of varied harmonic progressions w h i c h
can be expressed i n a l l manner of figuration and motives. A key i n
w h i c h to begin and end must be established. A l t h o u g h n o bar Unes
are employed, the ear demands a definite relationship i n the succession and d u r a t i o n of the chords themselves, as we shall see later,
and the eye, a relationship i n the lengths of notes so that the piece
may be notated. T h e r e f o r e , i t is usually assumed that such fantasas
are i n a f o u r - f o u r meter; and the tempo is indicated by the words
w h i c h are placed above the b e g i n n i n g . W e have already learned of
the fine effect created by fantasas i n Chapter I I I of Part I of this
Essay, to w h i c h I refer m y reader.
2

4.

Especial care must be exercised i n i m p r o v i s i n g at the harpsi-

1 A detailed study of this chapter and an analysis of the appended Fantasa (Figure
480) appear in Heinrich Schenkers Das Meisterwerk in der Msik, Drei Masken Verlag, Mnchen 1925, Vol. I, p. 11 ff.
2 In
15.
430

c h o r d and ihc o i g a n ; at the l o i i n e r , i n order to avoid p l a y i n g i n a

single color; at the latter i n order to sustain constantly and h o l d
chromatic progressions i n check, A t least, they should n o t be i n t r o duced sequentially, for the t u n i n g of the organ is very rarely tempered. T h e best instruments for o u r purpose are the clavichord
and pianoforte. B o t h can and must be w e l l t u n e d . T h e u n d a m p e d
register of the pianoforte is the most pleasing nd, once the perf o r m e r learns to observe the necessary precautions i n the face of its
reverberations, the most d e l i g h t f u l for i m p r o v i s a t i o n .
5. T h e r e are occasions w h e n an accompanist must extemporize
before the b e g i n n i n g of a piece. Because such an i m p r o v i s a t i o n is
to be regarded as a prelude w h i c h prepares the listener for the conten of the piece that follows, i t is more restricted t h a n the fantasa,
f r o m w h i c h n o t h i n g more is r e q u i r e d than a display of the keyboardist's s k i l l . T h e construction of the f o r m e r is d e t e r m i n e d by
the nature of the piece w h i c h i t prefaces; and the content or affect
of this piece becomes the m a t e r i a l o u t of w h i c h the prelude is
fashioned. B u t i n a fantasa the performer is completely free, there
b e i n g n o attendant restrictions.
6. W h e n o n l y l i t t l e t i m e is available for the display of craftsmanship, the performer should n o t wander i n t o too remote keys,
for the performance must soon come to an end. Moreover, the p r i n cipal key must n o t be lef t too q u i c k l y at the b e g i n n i n g or regained
too late at the end. A t the start the p r i n c i p a l key must p r e v a i l for
some t i m e so that the listener w i l l be unmistakably oriented. A n d
again before the cise i t must be w e l l prolonged as a means of prep a r i n g the listener for the end of the fantasa and impressing the
tonality u p o n his memory.
7. F o l l o w i n g are the briefest and most n a t u r a l means of w h i c h
a keyboardist, p a r t i c u l a r l y one of l i m i t e d a b i l i t y , may avail himself
i n e x t e m p o r i z i n g : W i t h due c a u t i o n he fashions his bass o u t of the
ascending and descending scale of the prescribed key, w i t h a variety
of figured bass signatures (Figure 472, Example a); he may interplate a few half steps (b), arrange the scale i n or o u t of its n o r m a l
sequence (c), and p e r f o r m the resultant progressions i n b r o k e n or
8

a T h e upper signatures of the first ascending and descending scales in major and
minor agree with the older Regola deW Ottava, which was used by i7th- and i8thcentury theorists to instruct beginners in the proper chord for each step. It was
adopted with minor variants by Rameau, Heinichen, and Mattheson, among others.
It formed the first step of instruction in the reading of unfigured basses, and was

4 3

/ M ' RO

I M r R O V ISA T 1 O N

sustaincd slylc at a suitablc pare. A tonic oigan p o i n l is convenient

for establishing the tonality al. the beginning and end (d). T h e
d o m i n a n t organ p o i n t can also be introdneed elfec tively before the
cise (e).
Figure 472
a.
6 6

98989898986
76767676765

5b

14

765656

7 6
5b-

jt

44

'5 4 6

6 98 7
b5- 4b 3

'5b '9 8 9 8 7 * 5b
4b3 4 3

64
26

r r r gr
^

56 7 6 '

6 5
9.8 4# 6

i I I I ili
7

6,

5 * 6 5b 56

9 8 '9 8 * 5 >
7 6 76 5b

*
6
41
3

76

541-6

7*

6 *

4
5465b2 6

m,

76

6
56 55

5* 6
5b

6
5

H 6

r'r

7 6 *4 jj 7 *

i
5b

9 8
*
*
4,3
4,3 7i6l>
7 6b 5b
5b 5b
5t> ; 5b

IIJ J J
6
5

b7
5

5b

.6,4

6.2

6 5

46

53

b7
5

56 5b

-0 0.

5b

b7
5

7*6

7 65b9 8 *
5 43 765b 5b

9 8

7*

6b5
4 3

5b-

6
b?
6 5 5.6 5

6
5

7,*

7*

6
6 5 43

6
5

J J

'5 6

#4

98 98
7*7 6

62

7676

76 6

7 6 '7 6 '7 4 6

5b

6
5,2

7 6 5.6

r r rr i

5 46
32

5 4 *
65
3 2 5b 4 3 41

7*

5 * 6 5b

56 7 6 5 * 4 3 6

7*

S 6 4+6 2 6

7-87
6
7 6 5b- 6 5 9 8 *6t|5

ION

'76 5b 9 8 9 8 7 6 5b

546
6

fl.

* 6b5
5b 4 3 6

6 5l>98

' | | I T f I I L-pFFF^I
1

y ISA

rr"

87
87 *9L8 76
Jt

recommended by Heinichen as the basis of improvised preludes. T h e great variety

of Bach's other signatures and his failure to mention the Regola, which he must have
known, indcate that he placed little weight on it. Cf. E . Borrel, Tribune de St. Gervais, X X I , p. 175.

54
3

4
b
b7
5

4
b

'

b7
b7
5

*
6
# 44. 6 6b5 5b 6 5

4#

5
3

r r1 rr ^ r t g
f

5 6 4 8W 6 - 5
3 4 2 3 5 5 4 3

6 4 5
8 7 4 2 4 3

7
4

b7

6
4

7
6b 8
4 5

7
6b
4 b7 6b4
8 2 3 4 2

-b7
5 5
-

IMPROVISATION

IMPROVISATION

8. W h e n the performer is allowed adequatr l i m e lo liave attent i o n directed to his work, he may modulate to remoler keys. B u t
f o r m a l closing cadenees are n o t always r e q u i r e d ; thcy are employed
at the end and once i n the m i d d l e . I t suffices i the leading tone
(semitonium
modi) of the various keys lies i n the bass or some other
part, for this tone is the p i v o t and token of a l l n a t u r a l m o d u l a t i o n .
W h e n i t lies i n the bass, the seventh chord, the c h o r d of the sixth, or
the six-five c h o r d is taken above i t (Figure 473, Example a); i t may
also be f o u n d i n chords w h i c h are inversions of these (fe). I t is one
of the beauties of i m p r o v i s a t i o n to feign m o d u l a t i o n to a new key
t h r o u g h a f o r m a l cadenee and then move off i n another d i r e c t i o n .
T h i s and other r a t i o n a l deceptions make a fantasa attractive; b u t
they must n o t be excessively used, or n a t u r a l relationships w i l l become hopelessly b u r i e d beneath t h e m .

r e m a i n i n g keys are the most distan!; any of them may be i n c l u d e d

i n a free fantasa even thougli iliey stand i n v a r y i n g distantes f r o m
the tonal center. T h i s may be seen f r o m an e x a m i n a t i o n of the wellk n o w n Circle of Keys. B u t i n a free fantasa, the performer should
11

Figure 474

Figure 473

a.

a.
6

ti

11J

6
5b

3415 443

b.
.65

b.
- I

1J J

6b b7

-fr

2j

7
|
b7
5

6
V .

9. I n a free fantasa m o d u l a t i o n may be made to closely related,

remote, and a l l other keys. Strange and profuse modulations are
n o t recommended i n pieces p e r f o r m e d i n strict measure, b u t a fantasa w i t h excursions to only the next related keys w o u l d sound too
p l a i n . F r o m a major key the acknowledged closely related keys are
o n the fifth degree w i t h a major t h i r d and o n the s i x t h w i t h a m i n o r
t h i r d . A n d f r o m m i n o r keys m o d u l a t i o n is made chiefly to the t h i r d
degree w i t h a major t h i r d , and the fifth w i t h a m i n o r t h i r d . B u t the
remote keys i n major are o n the second and t h i r d degrees, b o t h cont a i n i n g m i n o r triads, and o n the f o u r t h w i t h a major t r i a d . T h e
* Die Verkehrung jener Accorde. Inversin here has a looser meaning than it had
in Rameau's systematic use of the term.

00

6 - b?6
4 5b 5-4Jg 1 . .

"

7
#

4
2

b7
5

56
b 6

7
#
1
i

9
5* 7

51r6

6 b7
4 5

43

,b7
5

6 * b76 9 7
4 5b 5 4 5

65
4#

s T h e Circle of Keys was invented by Heinichen, on his own testimony, after hearing from Kuhnau about Kircher's method of moving through keys by fourths or

4}6

i M r a o v i s A r i o N

/ M i' / <> I I S A T I O N

feel no urther o b l i g a t i o n n i i lie* circle, for il vvoiild be w r o n g i n

this k i n d of piece to make a eyelic excursin t l i i o u g h all twentyf o u r keys. 1 shall leave it to the prvate study ol my leader to practice m o d u l a t i o n to the closely related keys by means of a s k i l l f u l
a t t a i n i n g of their leading tones, and shall illustrate here, i n the
interests of b r e v i t y , a few particular ways to approach these keys
gradually (Figure 474). W e o w n immediately to the possibility of
there b e i n g many other ways to accomplish these ends; after the
i n i t i a l bass note, any other may be taken be i t what i t may. W e are
stopped f r o m a t t e m p t i n g a clear proof of this statement by the threat
of diffuseness.
10. T h e examples of Figure 475 illustrate slightly circuitous
ways of m o d u l a t i n g f r o m a m a j o r key to the distant keys w h i c h
were m e n t i o n e d i n the preceding paragraph. T h e cise relationship
of A m i n o r to C major relieves us of the repetitious task of f u r n i s h i n g similar examples f o r the m i n o r mode. W h e n i t is desired to
reach distant keys conclusively instead of simply passing t h r o u g h
t h e m , i t is n o t sufficient merely to reach for the leading tone i n the
belief that once i t is f o u n d the goal w i l l have been attained a n d that
f u r t h e r ends may then be sought immediately. T h e ear, i n order n o t
to be disagreeably startled, must be prepared for the new key by
means of intermedate h a r m o n i c progressions. T h e r e are keyboardists w h o understand chromaticism and can e x p l a i n t h e i r progressions, b u t few w h o k n o w h o w to employ i t agreeably, relieved of its
Figure 475
5> * *

It *+" - l tt,

]f'

11*
d

It
hn

It

A*

6
44-

5 b 8 t7
1

uU

5b-

il J U
1

!Hb7

UH n n

76 5

4$ 4 g H

4i

&*7

4 tt 2|. 6
<)!

6b5 .
4tf

b7.

b7
J

41-

4
2b b

6b 5 b

k.

6b
4b

b7
5
3

5b

5b

2b

>
44

5b

JJ

44 6

" ,W
;!J J J

jt

7
fr
4 j

Hl
i

-#*

5b
gal

* 5
b 5b 41 *
l

5 b 6b 5

'

4 B

1 1

3 fe

fe

fifths. Mattheson was scornful of Heinichen's Circle and offered an improved construction. Both are reproduced in Arriold, op. cit., pp. 268 and 277.

* 1# ' - n i

*<w d-

TV- d>r>
5b 4 3 b

1UJ^J

b7

1H = l

'

J n J i r j

4b 3

44 8

pt=i

b7

4121-

6
5.6 5

^5-

417

4 g

b7, 9 8 6 b 5 ,
5b 4 b4 3b

te

7 b7
7 6 5 5b 6 &
M 4 4 jt 2 jj 4 g

rrrip
j

4 3

1 M P R O VI S A T I O N

IMPROVISATION

crudeness. I t should be observed generally, b u l p a r l i c u l a r l y i n the

f o l l o w i n g examples, that the progressions which introduce remote
m o d u l a t i o n s f r o m an established key must be playee! more broadly
than those of other modulations. By transposing these and the preceding examples, and c o m b i n i n g them, a facility i n m o d u l a t i o n w i l l
eventually be attained.
11. As a means of reaching the most distant keys more q u i c k l y
and w i t h agreeable suddenness n o c h o r d is more convenient and
f r u i t f u l t h a n the seventh c h o r d w i t h a d i m i n i s h e d seventh and fifth,
for by i n v e r t i n g i t and changing i t enharmonically, a great many
chordal transformations can be attained. A n d w h e n there is added to
this a l l the h a r m o n i c artistry and rare progressions of the preceding
chapters, w h a t an endless vista of harmonic variety unfolds before
us! Does i t s t i l l seem difficult to move wherever we will? H a r d l y , for
we need o n l y decide how circuitous or direct o u r r o u t e must be.
T h e r e are o n l y three of these chords of the d i m i n i s h e d seventh w i t h
t h e i r three superimposed m i n o r thirds, for the f o u r t h c h o r d is a
r e p e t i t i o n of the first, as i l l u s t r a t e d i n Example a, Figure 476. I t
w o u l d take too l o n g to demnstrate a l l of the o p p o r t u n i t i e s afforded
by this c h o r d to guide h a r m o n y i n any conceivable d i r e c t i o n . T h e
possibilities of e x p e r i m e n t a t i o n w h i c h are suggested u n d e r b must
suffice for the present. W e repeat that such chromatic progressions
are to be played o n l y occasionally, w i t h artistry, and broadly.
F

*
**

b7
5

4* b7
>

5 b 2j.

-'NMJij

b7

6 b

b7
S.b

>
5^

I||J

l7 6l
5 4

4 4

21.

|b

t-
5 4

ll|J

5 b l>

b 7 | x

b7
5b

jt

1||J

"^

6 b 6b
b 4

5b 6

j J1 Jy IIJ |J J lJ||J inJ l''p,a

r

12. T h e beauty of variety is made evident i n the fantasia. A

diversified figuration and a l l attributes of good performance must
be employed. T h e ear tires of u n r e l i e v e d passage w o r k , sustained

i V )

chords, 01 broken chords. Ily themselves Muy neither stir or still

the passions; and i t is for ihese pinposes that the fantasia is exceptionally well suited. Broken chords must not progress too r a p i d l y
or unevenly (Figure 477, Kxample a). Occasional exceptions to this
precept may be i n t r o d u c e d w i l h good effect i n t o chromatic progressions. T h e p e r f o r m e r must n o t break his chords constantly i n a
single color. B o t h hands may progress f r o m the l o w to the h i g h
register, or the left h a n d may do this alone w h i l e the r i g h t remains
i n its o w n register. T h i s k i n d of execution is good on the harpsichord, for o u t of i t there comes an agreeable alterntion of devised
forte and piano. Those w h o are capable w i l l d o w e l l w h e n they depart f r o m a too n a t u r a l use of h a r m o n y to i n t r o d u c e an occasional
deception; b u t i f t h e i r attainments are insufficient for the purpose,
they must enhance by means of a varied and fine execution of a l l
manner of figuration those harmonies w h i c h sound p l a i n w h e n perf o r m e d i n the usual style. Most dissonances may be d o u b l e d i n the
left h a n d . T h e ear w i l l accept the resultant octaves i n f u l l h a r m o n y ;
fifths, however, must be avoided. T h e f o u r t h , w h e n i t appears i n
company w i t h the fifth and n i n t h , and the n i n t h at a l l times are
not doubled.
Figure 477
6

'

*
7

b 7
2

wrong

i>7,

b?
5b

right

13. A l l chords may be b r o k e n i n many ways and expressed i n

r a p i d or slow figuration. B r o k e n chords i n w h i c h p r i n c i p a l as w e l l
as certain n e i g h b o r i n g tones are repeated (Figure 478, Example )
are especially attractive, for they are more varied than a simple
arpeggio where the tones are played successively just as they l i e
under the hands. I n the interests of elegance the major (b) or m i n o r
(c) second may be struck and q u i t t e d below each tone of a b r o k e n
t r i a d or a relationship based o n a t r i a d . T h i s is called " b r e a k i n g
w i t h acciaccature."
I n runs, the n o r m a l tones of chords are filled i n .
These runs may pursue a direct course t h r o u g h one or more octaves
u p w a r d and d o w n w a r d . B u t an agreeable variety arises o u t of repetitions (d) and the insertion of f o r e i g n tones (e). R u n s w h i c h contain
6

Le., at a fixed dynamic level.

44

I M r ItO

1 M r H O V I S A T I O N

IIS

A I I O N

II'

many half stcps rccjuirc a modrate: speed. A l l maniiei of groupings

may be alternated i n the course of runs (/). T h e triad and its i n versions may be expressed by the same r u n , and also the seventh
c h o r d and its inversions. A t times the augmenled sccond is avoided
i n chords w h i c h contain that i n t e r v a l (g); b u t i n certain igurations
i t is acceptable (h). I m i t a t i o n s i n parallel and contrary m o t i o n can
be very w e l l i n t r o d u c e d i n t o various parts (i). T h e chromatic chords
w h i c h were discussed i n Paragraph 11 are best fitted to slow figurat i o n and the expression of p r o f o u n d feeling, as we can see i n the final
movement of the last Lesson i n Part I of this Essay.
7

14.

I n order to p r o v i d e m y reader, t h r o u g h continuous ex-

i<*>0

00

J O

ampies of a l l k i n d s , w i t h a clear and useful conception of the construction of a free fantasia, 1 refer h i m to the Lesson m e n t i o n e d i n
the preceding paragraph, and Figure 480. B o t h are free fantasas;
the first is interspersed w i t h m u c h chromaticism, w h i l e the second
consists largely of n a t u r a l and usual relationships. T h e f r a m e w o r k
8

Si*gf-

i Sonata V I , third movement. Cf. Pt. I , Introduction, Note 17.

s T h i s paragraph will serve to illustrate Bach's views on musical analysis as described i n a letter to a friend, dated from Hamburg, Oct. 15, 1777 (cf. Bitter, Cari
Philipp Emanuel Bach und Wilkelm Friedemann Bach, und deten Brder, Vol. I, p.
348): " I n my opinin, in instructing amateurs, several things could be omitted that
many musicians do not, indeed, need not know. A most important element, analysis,
is lacking. T r u e masterpieces should be taken from all styles of composition, and the
amateur should be shown the beauty, daring, and novelty in them. Also, he should
be shown how insignificnt the piece would be if these things were lacking. Further,
he should be shown how eFrprs, pitfalls, have been avoided, and especially how far a
work departs from ordinary ways, how venturesome it can be, etc."

442

IMPROVISATION

I MP

of the latter, i n the f o r m of a igured bass, may be l'ound i n Figure

479. T h e note vales have been w r i t t e n as accuratcly as can be expected. I n performance each c h o r d is arpeggiated twice. W h e n the
second arpeggio is to be taken i n a different register by either the
r i g h t or the left hand, the change is indicated i n the fantasia. T h e
tones of the slow, f u l l y g r i p p e d chords, w h i c h are played as arpeggios, are a l l of equal d u r a t i o n , even t h o u g h restrictions of space
have necessitated the superposing of w h i t e a n d black notes i n the
interests of greater l e g i b i l i t y . A t the b e g i n n i n g and end (1) of the
sketch (Figure 479) we find l o n g extensions on the tonic harmony.
A t 2 there is a m o d u l a t i o n to the fifth o n w h i c h the performer remains f o r some t i m e u n t i l at x he moves toward E m i n o r . T h e three
tones at 3, j o i n e d by a slur, elucdate the t r a n s i t i o n to the r e p e t i t i o n
of the c h o r d of the second w h i c h is regained by means of an interchange of chordal tones. T h i s t r a n s i t i o n is p e r f o r m e d i n slow
figuration,
the bass b e i n g purposely o m i t t e d f r o m the piece as perf o r m e d . T h e change f r o m the seventh c h o r d o n b to the f o l l o w i n g
c h o r d of the second on b-flat is an ellipsis, for n o r m a l l y the six-four
c h o r d o n b or the t r i a d o n c w o u l d precede the chord of the second.
T h e c h o r d at 4 seems to p o i n t t o w a r d D m i n o r , b u t the m i n o r t r i a d
is o m i t t e d a n d instead the c h o r d of the second (5) w i t h an augm e n t e d f o u r t h is played o n c as i f the plan were to move o n to the
G major c h o r d . Instead, the G m i n o r c h o r d is played at 6, to be f o l lowed largely by dissonant relationships leading back to the p r i n cipal tonality, o n w h i c h the fantasia ends over an organ p o i n t .
Figure 479
Allegro

y t i f t H 0

OHhi

" 8

HIm

43

ir np-r,

7
5

2,

6b 6
2

1
1\-m
I
f
o
j

V
e

(x.)
(3.)
.1

<c

| j, ^

6
4

2
L

c L_m>
(4.)

(5.)

H O V 1 S ATI

O N

li 1 li I I O (. R A l> II Y

447

Cramer, I I . "Die Violonccll Koinposiiioncn P. E. Bachs," in Allgemeine

Musik Zeitung, Jahrg. 57, pp. J J I ( II., Berln, 1930.
Cramer, H . "Einiges zu P. K. Bachs Kammermusik," i n
Musik Zeitung, Jahrg. 57, pp. 519 f., Berlin, 1930.

Daffner, H . "Die Entwickelung des Klavierkonzerts bis Mozart," i n

Publikationen der Internationalen
Musikgesellschaft,
Beihefte, zweite
Folge, Heft I V , pp. 24 ff., Leipzig, 1906.

BIBLIOGRAPHY

Dannreuther, E. Musical Ornamentation,

Aldrich, Putnam. The Principal Agrments of the Seventeenth

and
Eighteenth
Centuries, Diss. unpublished, Harvard University, 1942.
Allgemeine musikalische Zeitung, Leipzig, 1798-1848. Numerous a n i eles and notices pertaining to C. P. E. Bach, especially i n the early
volumes.
"Anmerkungen ber einige der besten musikalischen Schriften von der
Ausfhrungskunst insonderheit H r n . C. P. E. Bachs Versuch ber
die wahre A r t , das Clavier zu spielen," i n Bibliothek der schnen
Wissenschaften und der freyen Knste, V o l . 10, Pts. 1 and 2, Leipzig,
1763-64.

]j A r n o l d , F. T . The
London, 1931.

Art of Accompaniment

from a

Thorough-Bass,

Bach, C. P. E. " E i n f a l l einen doppelten Contrapunct i n der Octave von

sechs Tacten zu machen ohne die Regeln davon zu wissen," i n
Historisch-Kritische
Beytrge, F. W . Marpurg, ed., V o l . I I I , Pt. I ,
pp. 167 ff.
Bitter, C. H . Cari Philipp Emanuel und Wilhelm
und deren Brder, 2 vols., Berln, 1868.

Allgemeine

Friedemann

Bach,

London [n.d.]

Daymond, E. R. "Cari Philipp Emanuel Bach," i n Proceedings

Musical Association, 33ra Session, London, 1907.
*\/t Dolmetsch, A. The Interpretation
X VHIth Centuries, London [n.d.]

of the

of the Music of the XVIIth

and

Essner, W . "ber die Kunst Cari Philipp Emanuel Bachs," i n Zeitschrift

fr Musik, N o . 103, pp, 922 ff., 1936.
European Magazine, London, October 1784. (Contains a sharp attack
on Bach and his music.)
Flueler, M . Die norddeutsche
Berlin, 1908.

Sinfonie

zur Zeit Friedrichs

des

Grossen,

Hamburger unpartheiischer Correspondent, Hamburg. (Contains a n i eles and notices pertaining to C. P. E. Bach during the period of his
Hamburg residence, 1767-88. Cf. N o . 7, January 11, 1773.)
Hase, H . von, "Cari Philipp Emanuel Bach u n d Joh. Gottl. I m .
Breitkopf," i n Bach Jahrbuch,
1911.
Hoffmann, H . Die Norddeutsche Triosonate
und C. Ph. E. Bach, Kiel, 1927.

des Kreises um J. G. Graun

Burney, Charles. A General History of Music, V o l . I V , London, 1789.

Jalowetz, H . "Beethovens Jugendwerke i n Ihrem melodischen Beziehungen zu Mozart, Haydn, u n d Ph. E. Bach," i n Sammelbnde
der
internationalen Musikgesellschaft,
Jahrg. 12, 1910-11, pp. 417 ff.

Burney, C. The Present State of Music in Germany, The

and United Provinces, London, 1773.

Kaswiner, Salomn. Die Unterrichts-Praxis

1450-1 y50, Diss. Vienna, 1930.

Cherbuliez, A. E. " C . P. E. Bach," i n Allgemeine

Zrich, Neujahr Blatt, 1940.

Netherlands

Musikgesellschaft

in

Chrysander, F. "Eine Klavierphantasie von Karl Philipp Emanuel

Bach," i n Vierteljahrschrift
fr Musikwissenschaft,
N o . 7, 1891.
Clercx, S. " L a forme d u Rondo chez Cari Philipp Emanuel Bach," i n
Revue de musicologie, V o l . X V I , pp. 148 ff., Pars, 1935.
Cramer, C. F. See Magazin der

Musik.
446

Kothe, W . Friedrich

der Grosse ais Musiker,

fr

Tasteninstrumente,

Braunsberg, 1869.

Magazin der Musik, ed. C. F. Cramer, Jg. 1-2, Hamburg,

Jg. 3, Copenhagen, 1789. Numerous references to Bach.

1783-87;

Mersmann, H . " E i n Programmtrio K. P. E. Bachs," i n Bach

Jahrbuch,

i9i7Miesner, H . "Aus der Umwelt Ph. E. Bachs," i n Bach Jahrbuch,

1937.

4 4

RI.IOC,

Miesner, H . "Ungedruckte
Zeitschrift

RA

l'll

li I II I I O (. R A I' II Y

Briefe von Philipp F.iir.imicl Bach," i n

X I V , pp. 224 11.

fr Musikwissenschaft,

Miesner, H . P. E. Bach in Hamburg

Leipzig, 1929.

Miesner, H . "Ph. E. Bachs musikalischer Nachlass," i n Bach

Jahrbuch,

1938, 1939, to be cont.

Monthly

Musical

Record,

Monthly

Musical

Record,

V o l . 12, No. 142, pp. 222 ff., London, 1882.

V o l . 20, N o . 239, p p . 242 ff., London, 1890.

Mller, F. "Cari Philipp Emanuel Bach," i n Zeitschrift

fr

Musik,

No. 105, pp. 1323 ff., 1938.

MuSicians, London, 1867. Translation
Briefe, Leipzig, 1867.

Opel, R. "ber Beziehungen Beethovens zu Mozart und zu Ph. Em.

Reichardt, J. F. Briefe

V, 1922-23, pp. 30 ff.

fr Musikwissenschaft,
eines aufmerksamen

Reisenden

die Musik

betref-

fend, Pt. 1, Frankfurt and Leipzig, 1774, Pt. I I , Frankfurt and Breslau,
1776.
Reichardt, J. F. [?]. Musikalischer

Almanach

(Alethinopel)

Reichardt, J. F. Musikalischer

Almanach,

Reichardt, J. F. Musikalisches

Kunstmagazin,

[n.p.], 1782.

Berlin, 1796.

Riemann, Hugo. "Die Shne Bachs," i n Prludien

Leipzig, 1901.
Rochlitz, F. " K a r l Philipp Emanuel Bach," i n Fr Freunde
kunst, Vol. I V , pp. 177 ff., 3 r d ed., Leipzig, 1868.
^ - R o l l a n d , R. A Musical

Tour

through

the Land

der Ton-

of the Past, New York,

1922.

Salzer, F. "ber die Bedeutung der Ornamente i n Philipp Emanuel

Bachs Klavierwerke," i n Zeitschrift
fr Musikwissenschaft,
XII,
1929-30.

Jfc Schenker, H . Ein Beitrag

zur Ornamentik,

Universal Edition No. 812,

Vienna, 1908.
Schenker, H . "Die Kunst der Improvisation," i n Das Meisterwerk
Musik,

in der

V o l . I , Munich, 1925.

Schering, A . "Cari Philipp Emanuel Bach u n d das 'redende Prinzip'

i n der Musik," i n Jahrbuch
Schmid, E. F. C. P. E. Bach

Sincero, D . " L a Sonata d i F. E. Bach," i n Rivista

pp. 677ff.,1898.

der Musikbibliothek
und seine Kammermusik,

Peters,

V o l . 45, 1938.

Kassel, 1931.

fr

Musikwis-

derneuen

musicale,

liach-

V o l . 5,

Steglich, R. " K . P. E. Bach u n d der Dresdener Kreuzkantor Gottfried

August Homilius i m Musikleben ihrer Zeit," i n Bach Jahrbuch,
1915.
Klavier

Sonaten

zur Zeit Friedrichs

des

Grossen,

Diss. Berlin, 1930.

Terry, C. S. Origin

of the Family

of Bach Musicians,

Thouret, G. Friedrich
zig, 1898.

der Grosse

Thouret, G. Friedrichs

des Grossen

ais Musikfreund
Verhaltnis

Oxford, 1929.
und Musiker,

zur Musik,

Leip-

Berlin, 1895.

Torrefraneo, Fausto. Le origini

italiane
del romanticismo
musicale,
T u r i n , 1930. Pp. 388-433, La fortuna di P. E. Bach, appear also i n
Rivista musicale, V o l . 25, Nos. 3 - 4 .

Vrieslander, O. Cari Philipp

Vol. I I I ,

und Studien,

Wicii, 1932.

llaydiis,

Schneider, M . "Bach-Uikunden," i n Verfjentlichungen

gesellschaft,
Jahrg. 17, Heft 9, Leipzig, i g i 7 [ ? ] .

U l d a l l , H . Das Klavierkonzert

Berlin, 1782, 1791.

Schmid, K. F. "laydn und Ph. F i n . Bach," i n Zeitschrift.

senschaft, X I V , pp, 299 ff., 1931-32.

Stilz, Ernst. Die Berliner

N o h l , L . Letters of Distinguished
by Lady Wallace of Musiker
Bach," i n Zeitschrift

Schmid, I . . I' He Vorfahrrn

.f. ,,

der Berliner
Emanuel

Vrieslander, O . "Cari Philipp

Emanuel
neuer Musik, Cologne, 1925.

Wien-Claudi, Hertha. Zum Liedschaffen

1928.

Schule,

Bach,
Bach

Friedrich

M u n i c h , 1923.
i n Von

ais Theoretiker,"

C. P. E. Bach,

Wotquenne, A l f r e d . Catalogue
thmatique
Philippe
Emmanuel
Bach, Leipzig, 1905.

Zelter, Karl Friedrich. Karl

Leipzig, 1928.

des oeuvres

Christian

Fasch,

Reichenberg,
de

Charles

Berlin, 1801.