(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "J.S. Bach Volume 1"

v.l 



Sch.we.itacr 
Jof 1 '. Bach = 





SEP 



OCT291987 



I- I 



-"; M *, * ,, 

NOV J, 1975 
MA > FEB 10 ikij 

MAI f.a^lti 3 
Y ^ 9 



'Li '; J! ./tV,' 



MAI:;APR'J 



WAJ APR 1 7 



197H 



MAR 



780.92 BllBsc 

v.l 

Schweitzer, Albert, 

187S-1965. 
J.S. Bach. 

[19351 



J. S. BACH 



ALSO BY ALBERT SCHWEITZER 
THE PHILOSOPHY OF CIVILIZATION 

I. THE DECAY AND THE RESTORATION Of CtVH,t7ATlON 
II. ClVlMJiATION AND KTHIC8 

THE MYSTERY OF THE KINGDOM OF GOD 
THE QUEST OP THE HISTORICAL JESUS 

PAUL AND HIS INTERPRETERS 
THE MYSTICISM OF PAUL THE APOSTLE 
ON THE EDGE OF THE PRIMEVAL FOREST 
and MORE FROM THE PRIMEVAL FOREST 

GOETHE 

INDIAN THOUGHT AND ITS DEVELOPMENT 
THE PROBLEM OF PEACE IN THE WORLD OF TO-DAY 

ADAM CHARLES BLACK 



J. S. BACH 



BY 

ALBERT SCHWEITZER 

DR. THEOL., DR. MED. DR. PHIL., STRASBOURG 
HON, MIT8. DOC, EDINBURGH. HON. D.0., OXON. HON. LL.D., ST. ANDREWS 



TRANSLATED BT 

ERNEST NEWMAN 
PREFACE BY C. M. WIDOR 



VOLUME I 

WITH THREE PLATES 



NEW YORK 
THE MACMILLAN COMPANY 



First published, in French, fa i95 

Evkfgtd version published in Gtrmtn, in jrpotf 

English translation front the Ctrmtm Kdition, with alimtim and additions 

by the author, published in ign by lirtitkotf **4 HwM t Lmdon 

Re- issued in igA$ by A. * C, Black, /.& London 
Rented jp^5, x$3> W4$> *94?> '949, *95> WS* M * W5 



Made in Great Jlritain 
Printed ly Low <S- lirydon .(Printtws) !M*, l 



TO 

FRAU MATHILDE SCHWEITZER 

IN GRATEFUL REMEMBRANCE 



TRANSLATOR'S FOREWORD. 

Within the last few years Bach research has made a 
notable advance. Among the books that have contributed 
to this progress, that of Dr, Albert Schweitzer takes a 
leading place. It is equally valuable on the xsthetic and 
the practical sides; its convincing demonstration of the 
pictorial bent of Bach's mind must necessarily lead to a 
reconsideration not only of the older view of Bach as a 
mainly "abstract" musician, but of the aesthetics of music 
in general; while the chapters on the right manner of per- 
forming Bach's works throw many a new light on this 
obscure subject. Most of all are correct ideas on this latter 
point invaluable now, when Bach is beginning, as one 
hopes, to win his due popularity among not only musicians 
but music lovers as a whole, 

The present translation has been made from the German 
version of Dr, Schweitzer's book (1908), which is itself 
a greatly expanded version of a French original published 
in 1905, The text, however, has been largely altered 
and added to at Dr. Schweitzer's request. The English 
edition is thus fuller and more correct even than the German. 

Like most other translators I have found it convenient - 
and indeed necessary to preserve the word "clavier" 
to cover all the seventeenth and eighteenth century instru- 
ments the harpsichord, clavichord, clavicembalo, &c* 
of the type now represented by the pianoforte. 

For the benefit of the English reader I have given all 
the references to Spitta's "Life of Bach" in the correspond- 
ing pages of the English edition of that book, published 
by Messrs. Novello & Co. 

The index to the German original of Dr. Schweitzer's 
book being hardly adequate, I have prepared a fuller one 
of my own, which I hope will increase the usefulness of 
the volumes. 

ERNEST NEWMAN 



PREFACE TO THE GERMAN EDITION (1%8.) 

In the autumn of 1893 a young Alsatian presented himself 
to me and asked if he could play something on the organ to 
me, "Play what?" I asked. "Bach, of course", was his reply. 

In the following years he returned regularly for longer 
or shorter periods, in order to "habilitate" himself as 
they used to say in Bach's day in organ playing -under 
my guidance, 

One day in 1899, when we were going through the chorale 
preludes, I confessed to him that a good deal in these 
compositions was enigmatic to me, "Bach's musical logic 
in the preludes and fugues", I said, "is quite simple and 
clear; but it becomes cloudy as soon as he takes up a chorale 
melody. Why these sometimes almost excessively abrupt 
antitheses of feeling? Why does he add contrapuntal 
motives to a chorale melody that have often no relation 
to the mood of the melody? Why all these incomprehen- 
sible things in the plan and the working-out of these fan- 
tasias ? The more I study them the less I understand them ." 

"Naturally," said my pupil, "many things in the chorales 
must seem obscure to you, for the reason that they are 
only explicable by the texts pertaining to them." 

I showed him the movements that had ptmlecl me the 
most; he translated the poems into French for me iroiu 
memory. The mysteries were all solved. During the 
next few afternoons we played through the whole of the 
chorale preludes. While Schweitzer for he was the 
pupil explained them to me one after the other, 1 made 
the acquaintance of a Bach of whose existence I had pre- 
viously had only the dimmest suspicion, In a flash it 
became clear to me that the cantor, of St. Thomas's was 
much more than an incomparable contrapuntist to whom 
I had formerly looked up as one gazes up at a colossal 
statue, and that his work exhibits an unparalleled desire, 
and capacity for expressing poetic ideas and for bringing 
word and tone into unity. 



Preface to the German Edition (1908.) IX 

I asked Schweitzer to write a little essay upon the 
chorale preludes for the benefit of FrencH organists, and at 
the same time to enlighten us as to the nature of the German 
chorale and the German church music of Bach's epoch, 
as we knew too little of them to enter thoroughly into 
the spirit of the cantor's music. 

He set to work at this. A few months afterwards he 
wrote to me that it was necessary to include the cantatas 
and Passions in his essay, since the vocal works explained 
the chorale works, and vice versd. " Your essay", I re- 
plied, "will simply be so much the more valuable to us." 

The remarks upon the chorale and the church service 
in Bach's time grew into an epitome of the history of 
Protestant church music; the observations upon the nature 
of Bach's musical expression became a chapter upon 
"Bach's tone-speech"; a short literary portrait of the com- 
poser was seen to be desirable; then there came chapters 
on the practical performance of Bach's works; and so the 
essay upon the chorale preludes grew, in the space of six 
years, into a complete book upon Bach. The author sent 
me each chapter as it was written. When I wrote a pre- 
face to the book in Venice, on 20th October 1904, it was 
with the joyous feeling that the work would open up for 
us a free road to Bach. 

Now, as I sketch the preface to the German edition, 
I cannot rid myself of a certain feeling of embarrassment, 
Is it not presumption for me, a Frenchman, to draw the 
attention of Germans to a work upon Bach? 

I may partly plead in excuse that in a limited sense 
I am the joint originator of the book. It was at my request 
that Schweitzer undertook the work; it was I who induced 
him to persevere with it when the difficulties of the under- 
taking increased and began to look, at times, almost in- 
surmountable. 

I believe, therefore, that it is not only my right but my 
duty to prepare the way for this book in Germany if 
that be necessary since it seems to belong to a special 
category in the German literature of the arts. I rank it 



X Preface to the German Edition (1908.) 

among tlic works the significance of which consists in the 
fact that while they arc founded on a thorough professional 
knowledge, they treat their subject from the standpoint 
not of a single art but of art and science in general, 
Schweitzer is a philosopher through and through^ as is 
shewn by his work on Kant; at the same*, time he. is a 
theologian with a profound historical faculty, as may be 
seen from his well-known and comprehensive studios in 
the life of Jesus and in the literature of that subject; more- 
over lie is an exceptionally good organist, one* of the 
most skilful and experienced players that airy conductor 
could desire to have at the organ during the performance 
of a Bach cantata or Passion. 

The not unreasonable complaint is sometimes heard 
that our sestheticians are so seldom executive, artists also, 
and therefore cannot view things from the standpoint; of 
the musician. There is no community o : f fooling between 
the philosophy of art and creative and executive art. For 
this reason works by practical men who arc at the same 
time conversant with philosophical aesthetic are- always an 
event in the literature of music. To read Schweitzer's 
Bach is not only to get to know the composer and his woi k, 
but to penetrate also into the essence of music in general, 

the u art per se". It is a book with "horizons". Who 
could have supposed that a study of the great master of 
the "Zopf * epoch would throw a light on the modern - 
even the most modern problems of music, as is clone 
in the three chapters "Poetic and Pictorial Music.'*, 
"Word and Tone in Bach", and "Bach's Musical Language" 

with which Schweitzer prefaces his discussion of the 
cantatas and the Passions? 

An introductory note by a Frenchman to a German 
book on Bach may further show that we on this skit* of 
the Vosges have also some rights in the composer. We 
have won them by the veneration we have felt for him. 
Our Bach worship does not date from yesterday. For a 
generation now our organists have been almost exclusively 
occupied with Bach; he is the master who has revealed 



Preface to the German Edition (1908.) XI 

afresh to us the true art of the sacred Instrument. People 
speak of a new French organ school: it is founded on 
Bach, It was a curious dispensation of Providence that 
at the very time we were being led to Bach by the Belgian 
Lernmens who had become acquainted with the classical 
organ-art through old Hesse, of Breslau there arose an 
organ-builder after Bach's own heart, who gave us organs 
that made us the envy of Bach enthusiasts in every land. 
Cavaill6-ColFs instruments have revealed to us the beauty 
of the master's preludes and fugues; with these organs 
Bach has made his entry into our cathedrals and churches. 

If he has not as yet taken his due place in our public 
concert life, that is to be accounted for on purely external 
grounds. Our public is enthusiastic for Bach, our singers 
and instrumentalists not less so. I myself have had proof 
of this during my ten years' conduct orship of the "Con~ 
cordia", when we performed many cantatas, the Magnificat 
and the St. Matthew Passion. 

There are German artists whose works we admire, while 
at the same time we know that they will never be domes- 
ticated among us. When we try to appropriate them to 
ourselves, we feel that a certain something remains that, 
how shall I say it? does not come to us from the soul. 
We never have this feeling with Bach; there seem to be 
bonds of affinity between his art and ours. 

The correctness of tins feeling is confirmed by the interest 
and the admiration that Bach himself evinced for contem- 
porary French art. What he thought of Couperin and 
the others is shewn by the copies that have come down 
to us in his handwriting and that of his pupils. His first 
biographer, Forkel, whose information came from Bach's 
own sons, expressly says that the composer thought a 
great deal of the old French organists, - whose works 
have now at last been rescued from oblivion. And Zelter 
himself, the old Bach enthusiast and grumbler, rather 
wrathfiilly proves to his friend Goethe that his idol did 
not escape the influence of the French, "especially of 
Couperin", His works seem to Zelter to be covered with 



XII Preface to the German Edition (1908,) 

a kind of elegant "tinsel" that must be attributed to foreign 
art. He would have liked to skim it off and show the true 
German Bach underneath. 

The present-day German admirers of Bach do not, share 
Zelter's view of the "tinsel". The finish, the elegance 
and the formal charm of Bach's work do not strike them 
as a disavowal of the German spirit. We, however, who 
aim at form and plastic clearness in every art, find our- 
selves again in Bach, And when Schweitzer, without 
being at all one-sided, again and again insists upon the 
"pictorial" as the fundamental tendency in Bach*s music, 
he only makes it clear to us what it is that attracts us to it. 
The time will come when Bach will be one of the most 
popular of composers in France, not merely because 
we can discover in him traces of French influence and of 
our own sense of form, but because Bach is on the whole 
the most universal of artists. What speaks through his 
works is pure religious emotion; and this is one and the 
same in all men, in spite of the national and religious par- 
titions in which we are born and bred. It is the emotion 
of the infinite and the exalted, for which words are always 
an inadequate expression, and that can find proper utter- 
ance only in art. For me, Bach is the greatest of preachers, 
His cantatas and Passions tune the soul to a state in which 
we can grasp the truth and oneness of things, and rise 
above everything that is paltry, everything that divides us. 

By thus conquering artistic and religious mankind, 
Bach fulfils a mission to our time, which will never rise 
above the barriers that the past has erected unless the 
great souls of- the past come to its aid. We are made one 
by what we admire in common, revere in common, com- 
prehend in common. 

Paris, aoth October 1907, 

CHARLES MARIE WXDGR* 



CONTENTS, 
Volume I 

Chapter L The Roots of Bach's Art .... 14 
Subjective and Objective Art i 4. 

Chapter II. The Origin of the Chorale Texts . 414 

The Reformation and the mediaeval sacred song 
4 7, The first hymn-book 710, Protestant poets 
of the hymns n 14. 

Chapter III. The Origin of the Chorale Melodies 1424 

Borrowings from the Middle Ages and new crea- 
tions 14 17, Borrowings from the secular songs 
17 21. The end of the creative period 22 -24. 

Chapter IV. The Chorale in the Church Service 2440 

The organ and congregational singing at the time of 
the Reformation 24 26. The choir and the congrega- 
tional chorale 2732. Osiander and Hassler 33* The 
organ undertakes the leading of the congregational 
singing 3438, Congregational singing in Bach's 
time 38- 40, 

Chapter V. The Chorale Prelude before Bach, 4050 

Samuel Scheldt 40 42. Pachelbol, Bdhm, Reinken, 
Buxtehude 42 46* Bach and his forerunners 46 50. 

Chapter VI, The Cantata and the Passion before 

Bach 5096 

The old church music 5156, Schutz. The in- 
fluence of Italian art on the German church service 
5660. The question of the text-form. 8 trophic 
song and madrigal 6164. The achievements of 
Schutz 6568. The tendencies of the new church 
music in the period after Schutz 69 72, The chief 
representatives of church music in the i/th century 
73~""-7S- The Lubeck AUndmusiken 76- 77, The 
cantatas of the northern school 78- 79. The new 
cantata 798 x . The development of the older Passion 
8284. The German opera and its significance for 
church music 8491. Neumcistcr and Salerno Franck 
91 92, The new form of Passion music 92 -96, 



XIV Contents. 

Chapter VII. From Eisenach to Leipzig , . , 97112 

Bach's ancestors 9798. Childhood and boy- 
hood 99100. Arnstadt and Muhlhausen (1704 to 
1707) 100104, Weimar (17081717) and Cdtheii 
(17171723) 104107. Journey to Hamburg (1:720) 
and appointment as cantor at. St. Thomas's 109 >i 12, 

Chapter VIII. Bach in Leipzig 112150 

The cantor's duties 112 <i 14. The financial posi- 
tion 115116, The conditions at St. Thomas's school 
116119. The struggle over the University service 
119122. Choir and orchestra 123-' 125. Music 
in the Leipzig service 126- 132. The first conflict 
with the Council 132138, Application for the, title 
of Court Composer 138140, The struggle with the 
rector 140- 143, Bach's position in the musical world 
of Leipzig 143144, Bach's children and their fate 
144- 150. 

Chapter IX, Appearance, Nature and Character 151170 

Bach's friendliness and modesty 151- 152, His 
attitude towards other artists 1531 57. His economy 
and hospitality 157158, Emmanuel inherits his 
father's economic spirit 158160. Portraits of Bach 
160- 162. The discovery of the skeleton 162164. 
Bach's artistic personality 164 r 66. His religion 
i 66 170. 

Chapter X. Artistic Journeys, Critics and Friends 1703:87 

Journeys in the pro-Leipzig period 170174. 
Journeys in the Leipzig period 174178. Matthcoon 
and Bach 178179. Scheibe's criticism 179- 183. 
Panegyrists in prose and verse 183186. Acquaint- 
ances and friends i86 187. 

Chapter XL The Artist and Teacher .... 187- 222 

Bach's general culture 1 87-- 189. The Mizler 
Musical Society 1891:92, Bach studies and ar- 
ranges the works of other men 192195, Bach's own 
imagination stimulated by the work of others 1 95 - 1 </>, 
His employment of other men's themes i97-ic)8, 
Bach and the organ construction of his epoch 
198- 200. Clavichord, clavicembalo, and pianoforte 
200- 203, The lute-clavier and the viola pompoaa 
204-^-206* Bach's clavier touch and violin playing 
206- 208. Improvisation, registration, conducting 
209210. The composer at work 211214. Bach's 
pupils 214*- 215. His method of teaching composi- 
tion 216220. The work of his pupils 220 >%2jt< 



Contents, XV 

Chapter XII. Death, and Resurrection .... 222 265 

Illness and death 222 224. Obituary notices 
225 -226. Why he was forgotten 227 235. The 
revival 235, Forkel and Rochlitz 235 240. Zelter 
and Goethe 240 241. The revival of the St. Mat- 
thew Passion 241 244. The results of the victory 
245. 246. Mosewius 246247, Obstacles to the 
proper appreciation of Bach 248 249. History of 
the Bachgesellschaft edition 249 254. Spitta's 
biography 254 -256. Liszt and Wagner 257 258. 
Bach in France, England and Italy 259- 261. Bach 
and the present day 261 265. 

Chapter XIII. The Organ Works 265 294 

Dates of composition, Youthful works 266 26i>. 
Preludes and fugues of the Weimar period 269275. 
Preludes and fugues of the Leipzig period 276- 277. 
Smaller preludes, organ sonatas, and passacaglia 

278281. Early chorale preludes 281282. The 
Qrgelbtichlein 283 -288. The chorale preludes on the 
catechism hymns 288290, The eighteen chorales 
290292. Paralipomena 293 294. 

Chapter XIV. The Performance of the Organ 

Works 294 320 

The Bach organ and the modern organ 294" 295. 
Registration 296 303, The natural architecture of 
the preludes and fugues 303304. Changes of manual 
305311. Tempo, phrasing and ornamentation 
311 -3 17. Organ and clavier fugues 3 1 7 -3 19. Tran- 
scriptions 319- 320. 

Chapter XV. The Clavier Works 320344 

Publication of the Klaviev&bwng 321325. llic 
French and English Suites 325- 328. The Little 

Preludes, Inventions and Sinfom'as 328, 331, Origin 
of the Well-tempered Clavichord 331 335. The 

autographs of the Well-tempered Clavichord 335336. 
The spirit of the Well-tempered Clavichord 336339. 

Separate preludes, fantasias, sonatas and toccatas 
339343^ Capriccios 343344. 

Chapter XVI. The Performance of the Clavier 

Works 345383 

The ornaments 345352. Cembalo or modem 
pianoforte 35 2- 35 5. The dynamic nuances 3 5 5- 364, 
The phrasing 365375. The accentuation of Bach's 
themes 3753.80, The tempo 381382, Epilogue 
383- 



XVI Contents, 

Chapter XVIL Chamber and Orchestral Works 384 -417 

The suites and sonatas for solo violin 384388, 
Polyphonic violin playing in Bach's time 388393, 

The suites for cello solo 393-" 394. The sonatas for 
clavier and violin and their performance 394401, 
Sonatas for gamba and for flute 401 -402, Or- 
chestral overtures 402- 403. The Brandenburg con- 

c^tos^aud^^ their performance 403- 410. A The 'clavier 
concertos 41 141 if," "The concertos" for three or four 
claviers 414 415. The violin concertos 415417, 

Chapter XVIII. The Musical Offering and the 

Art of Fugue 417 -428 

Origin- and character of the Musical Of foxing 
417- 421. The canons 421423. The origin of the 
Art of Fugue 423425. The fate of the Art: of Fugue 
425- 426. The musical quality of the Art of Fugue 
426428. 



CHAPTER I. 
THE ROOTS OF BACH'S ART. 

Some artists are subjective, some objective, The art 
of the former has its source in their personality; their 
work is almost independent of the epoch in which they live. 
A law unto themselves, they place themselves in opposition 
to their epoch and originate new forms for the expression 
of their ideas. Of this type was Richard Wagner. 

Bach belongs to the order of objective artists. These 
are wholly of their own time, and work only with the forms 
and the ideas that their time proffers them. They exercise 
no criticism upon the media of artistic expression that they 
find lying ready to their hand, and feel no inner compulsion 
to open out new paths. Their art not coming solely from 
the stimulus of their outer experience, we need not seek 
the roots of their work in the fortunes of its creator. In 
them the artistic personality exists independently of the 
human, the latter remaining in the background as if it 
were something almost accidental. Bach's works would have 
been the same even if his existence had run quite another 
course. Did we know more of his life than is now the case* 
and were we in possession of all the letters he had ever 
written, we should still be no better informed as to the 
inward sources of his works than we are now. 

The art of the objective artist is not impersonal, but 
superpersonal. It is as if he felt only one impulse, to 
express again what he already finds in existence, but to 
express it definitively, in unique perfection. It is not he 
who lives, it is the spirit of the time that lives in him. 
All the artistic endeavours, desires, creations, aspirations 
and errors of his own and of previous generations are 
concentrated and worked out to their conclusion in him, 

Schweitzer, Bach. I 



2 The Roots of Bach's Art, 

In this respect the greatest German musician has his 
analogue only in the greatest of German philosophers. 
Kant's work has the same impersonal character. lie is 
merely the brain in which the philosophical ideas and 
problems of his day come to fruition. Moreover he uses 
unconcernedly the scholastic forms and terminology of the 
time, just as Bach took up the musical forms offered to 
him by his epoch without examining them. 

Bach, indeed, is clearly not a single but a universal 
personality. He profited by the musical development of 
three or four generations. When we pursue the history oC 
this family, which occupies so unique a position in the art- 
life of Germany, we have the feeling that everything that 
is happening there must culminate in something consum- 
mate. We feel it to be a matter of course that some day 
a Bach shall come in whom all those, other Baohs shall 
find a posthumous existence, one in whom the fragment of 
German music that has been embodied in this family shall 
find its completion. Johann Sebastian Bach, to speak 
the language of Kant is a historical postulate. 

Whatever path we traverse through the poetry and tlu* 
music of the Middle Ages, we are always led to him. 

The grandest creations of the chorale from the twelfth 
to the eighteenth century adorn his cantatas and Passions. 
Handel and the others make no use of the superb treasures 
of chorale-melody. They want to be free of the past. Bach 
feels otherwise; he makes the chorale the foundation of 
his work. 

If we pursue, again, the history of the harmonisation of 
the chorale, we are once more led up to him. What the 
masters of polyphonic music, Eccard, Praetorius and 
the others strove after, he accomplishes. They could 
harmonise the melody only; his music at the same time 
reproduces the text. 

So it is, again, with the chorale preludes and the chorale 
fantasias. Pachelbel, Bohna and Buxtehude, the masters 
in this field, created the forms. But it was not given to 



Subjective and Objective Art. 3 

them to quicken the forms with the spirit. If all their 
struggles towards the ideal were not to be in vain, a greater 
man had to come, who should make his chorale fantasias 
musical poems. 

Out of the motet, under the influence of Italian and 
French instrumental music, came the cantata. From 
Schiitz onwards, for a whole century, the sacred concert 
struggles for its free and independent place in the church. 
People feel that this new music is cutting the ground of 
the old church service from under their feet. It forces 
itself further and further out of the frame of the service, 
aiming at becoming an independent religious drama, and 
aspiring towards a form like that of the opera. The ora- 
torio is being prepared. At this juncture Bach appears, 
and creates cantatas that endure. A generation later it 
would have been too late. As regards their form, his can- 
tatas do not differ from the hundreds upon hundreds of 
others written at that time, and now forgotten, They 
have the same external defects; they live, however, by 
their spirit. Out of the ardent will-to-create of generations 
that could not themselves give birth to anything durable, 
there has come for once a will equal to the ideal that hovered 
before the two previous generations, and that triumphs in 
spite of all the errors of its epoch, purely by the grandeur 
of its thought. 

At the end of the seventeenth century the musical Pas 
sion-drama demands admission into the church. The 
contest rages, for and against* Bach puts an end to it by 
writing two Passions which, on their poetical and formal 
sides, derive wholly from the typical works of that time, 
but are transfigured and made immortal by the spirit that 
breathes through them. 

Bach is thus a terminal point. Nothing comes from him; 
everything merely leads up to him. To give his true bio- 
graphy is to exhibit the nature and the unfolding of German 
art, that cornes to completion in him and is exhausted in 
him, to comprehend it in all its strivings and its failures. 



4 II. The Origin of the Texts of the Chorales, 

This genius was not an individual, but a collective soul. 
Centuries and generations have laboured at this work, 
before the grandeur of which we halt in veneration, To 
anyone who has gone through the history of this epoch and 
knows what the end of it was, it is the history of that cul- 
minating spirit, as it was before it objoctivatcd itself in a 
single personality. 



CHAPTER II. 
THE ORIGIN OF THE TEXTS OF THE CHORALES*; 

BIBLIOGRAPHY. 

PHHJPP WACKERNAGEL: Das dcutsche Kirchcnlied von def &lte$fon 

Zeit bis zu Anfcmg de$ XVILJahrhwiderts. 5 vols, 18641877. 
WILHELM BAUMKER: Das kctthotische deutsche Kirchcnlipd in $ehwn 

Singweisen, 3 vols, 1^83, 1886, 1891, (The second volume is 

a revised edition ol Scvcrin Holsters' Das deut$ch& katholische 

Kitchenlied. } 
ALBERT KNAPF: Evangclischcy Liedevschatx* 2 vols, ist ed. 1837; 

3rd cd. Stuttgart, 1865. (Contains 3130 hyms from all periods*) 
WILH. FRIEDR. FISCHER: KwchGnliederlexihon, 2 parts* GotUu* 

1878. Supplemental volume 1886. 
HOFFMANN VON FALLERSLEBEN : GescMchte dcs dcutschan 

licdes Ms auf Luther $ Zeit, 1854. 3rd od. 1861. 
FRIEDRICH ZELLE; Das fittest* hithenschcnansgcsanghuch( 

Enchiridion 1524), Gottitigcn 1903. 
FRIBDRICH SPITTAJ Bin feste Burg t$t unscr Gait. Die Li&dev Luthtrs 

in ihrer B&deutung jti? das evan^elische RivchmluuL Gdttlngcn 

1905. Studien xu Luth&rs Lied&rn. Gdttingon 1007* 
ED. EM. KOCH; GescMchte de$ Kirchenlieds und Kirchcngts&ngs 

der christfichen, insbesondere der deutschen evangelischtn JKiwfo* 

8 vols. 3rd ed. Stuttgart, 18661877, 
E. WOLF: Das deutsche Kirchenlied des XV L und XV XL Jdbr* 

hunderts, Stuttgart 1894* 
PHXUPP DIETZ: Die Restaumtion d$ wangtttschen JKirchenKtds* 

Marburg, 1903. 

It was the custom in the Catholic church, in the earliest 
times, for the congregation to take a direct part in the 
singing during the service; to it belonged the doxologie, 

* The chorale preludes oi Bach are cited in the well-known 
Peters edition ol the organ works. 



The Reformation and the Mediaeval Hymn, 5 

the Amens, the Kyries and the hymns. At the end of the 
sixth century and the beginning of the seventh, however, 
this privilege of the faithful, which had been secured by 
Ambrose, was taken from them by the Gregorian reform, 
which substituted the singing of the priests for that of the 
congregation. 

In Germany, however, this reform was not adopted in 
its entirety. The people still preserved a few of their 
privileges, especially in the Easter service, when they joined 
in the Kyrie and the Alleluia. The result was that it 
became the custom to insert German verses among the lines 
of the liturgy in these places. In this way the German 
sacred song gained admission into the religious service under 
cover of the Kyrie and the Alleluia. Throughout a long 
period of time these ejaculations formed the obbligato verse- 
ending to every hymn sung in the church. Hence these 
songs were called "Kirlcisen" (i. e. Kyrie songs.)- 

The oldest Easter-hymn dates back as far as the twc?lftli 
century. It runs thus: 

Christ 1st crstanden 

Von der Harder alle* 

DCS sollen wir alle froh sein, 

Christ soil unser Trost sein, 

Kyrioleis. 

Hallcluja, Halleluja, Halleluja. 
Des sollen wir alle froh sein, 
Christ soil unser Trost scin, 
Kyrioleis. 

Tbe Mystery Plays that had such a vogue in the four- 
teenth a,nd fifteenth centuries also helped the German hymn 
to conquer the church. The mixed Latin-German Christ- 
mas cradle-songs have a quite uncommon charm. The 
poetry of them is of the most primitive kind imaginable, 
The words are put together loss with regard to the sense 
than to the sound and the rocking rhythm; yet the bright 
Christmas enchantment that surrounds them affects us no 
less than it did the generations that have vanished. 



6 II. The Origin of the Texts of the Chorales, 

In Bach's organ chorales there arc two of these old 
Christmas songs : 

In dulci jubilo, 

Nun singet und sekl frob. 

Unsers Herzcns Wonne 

Liegt In praesepio, 

Und leuchtet als die Sonne 

Matris in gremio. 

Alpha et 0, Alpha ct O, (V, Nr. 35,) 

Puer natus in Bethlehem, 

In Bethlehem, 
Unde gaudet Jerusalem, 

Jerusalem. 
Halle, Hallcl, 

Ein Kind geborn 211 Bethlehem, 

Zu Bethlehem, 
Des frcuet sich Jerusalem. 

Halle, Hallcl. 

Cognovit bos et asinus, 

Asians, 
Quod Puer erat Dominus, 

Dominus. 
Halle, HalleL 

Das Ochslein und clas Eselein, 

Eselein, 

Erkannten Gott den Herren sein, 
Halle, Halld. (y> ^ ^ 

In time, translated Latin hymns came to bo admitted 
into German sacred poetry; the Credo, the Pater uostor, 
the ten commandments, the Seven Last Words, and va- 
rious Psalms, in metrical paraphrases, were also incor- 
porated in it, 

When the Reformation of the sixteenth century threw 
the doors of the churches open to German poetry, it was 
under no necessity to set to work to compose appropriate 
hymns, but could choose what suited it from the treasures 
of the fourteenth and fifteenth centuries, Luther, with 
that wonderful artistic feeling for diction that even Nietzsche 



The first Hymn-book. 7 

had to acknowledge in him, undertook to revise tlie old 
possessions for the new Church, and to alter and improve 
them as might be required. At the same time he himself 
continued the work of the Middle Ages by re-fashioning 
Latin hymns, psalms, liturgical chants and biblical frag- 
ments into hymns for the German service. 

The German Reformation had this advantage over the 
French, that it found a spiritual song already existing in 
the popular tongue, and therefore a ground upon which it 
could build; but its great good fortune was that it possessed, 
in Luther, a man who would not permit the old wood to be 
cut down, recognizing with sure prescience that the new song 
must grow up in the shade of the old. On the other hand 
the sacred folk-song withered away in the Romanesque 
countries, because it had no root in the Middle Ages, and 
had to exist as best it could upon the Psalter, as it does to 
the present day. 

At the first glance it may seem incomprehensible that 
Calvin, by making the Psalter the hymn-book of the people, 
should from the very beginning condemn his church to 
infertility. He obeyed the instinct of the Romanesque 
spirit, and pronounced the judgment that was decreed 
on the French Reformation even before il came into being. 
Later on, German chorales and English hymns were borrow- 
ed to be added to the Psalter. 

The first German hymn-book, the so-called Erfurt En- 
chiridion, appeared in 1524, and was probably compiled by 
Luther's friend Justus Jonas, It was issued simultaneously, 
curiously enough,- by Tratobulsch at the imprimcrie of 
the Dyeing Tub, and in Malcr's imprimo.rie of the Black 
Horn. The sole surviving copy of the Malcr issue was 
destroyed by fire in 1:870 at the bombardment of StraD- 
burg. Fortunately it had been reproduced in facsimile in 
1848. The Trutebulsch hymn-book has bem recently 
brought out in a new edition. * 

* Fricdrich Zcllc, Das alU&te luthcrische Haus-Gesangbuch (F&fbe- 

f ass-Enchiridion 1324) ; Gdttingen, 1903. In bis masterly Introduction 



8 II. The Origin of the Texts of the Chorales. 

In accordance with the practice of the time, this first 
hymn-book was shamelessly pirated everywhere, among other 
places at Nuremberg, where the printer Hans Hergott so zea- 
lously pirated Luther's writings from the very beginning 
that Luther, on the 26th September 1525, had to petition 
the town -council "to forbid the Hergotlein to pirate",* 

Among the twenty-six songs of the Erfurt Enchiridion 
are eight German translations of psalms, including u Au<5 
tieffer not schrey ich zu dir", a series of hymns done into 
German,** the two mediaeval Easter hymns "Christ lag yn 
todes banden" and "Jhesus Christ unser Heyland, dex den 
Tod iiberwand", the old hymn upon the ten command- 
ments, three hymns by Paul Speratus, including the well- 
known "Es ist das heyl uns kommen her", and some of 
Luther's hymns, including the "New lied von den zweem 
Mertererern Christi zu Brussel". "Ein feste Burg" does 
not appear in this hymn-book. *** 

Besides Luther and Paul Speratus (1484 1551) there 
may also be mentioned, as the earliest writers of sacred 
poems, Nicolaus Decius (died in 1541) and Nicolaus Sel- 



the editor gives a survey of the various Lutheran hymn-books that 
appeared in Luther's life-time. The title of this first small Ixymn- 
book runs thus: Eyn Enchiridion oder Handbttchlein eywM fe 
Christen fast nutzlich bey sich zu hdbn %u stettcr Hbung 
geystlicher Gesenge und Psalmen Rechtschaffm und h&nsttich 
teutsM. 3:524. Below, on the title-page : Mit diesen und 
Gesenge sollt man by Hick die yung&n Kinder aufferteitften* 

* Zellc, p. 23. 

** Veni redemptor gentium jV*w kom d&r Hey den htyl&nd. 
Veni sanct spiritm Kom heyliger Geyst herr Gott* 
A solis ortus cardine * Chrystum wir sotten lob&n schon* 
Veni creator = Kom Gott schepfer heyli^er Gey$t. 
Grates nunc omnes reddamus Gelobt seystu J&su Christ. 
As well as the sequence: 

Media in vita ^Mytten wir im leben s&ynd, and John Muss's hymn i 
Jesus Christus nostra salus Jesus Christus unser Htifand, 

der von uns den Zorn Gottes wand. 

*** For the latest research into the much-debated date of origin 
of this hymn see Friedrich Spitta, "Ein f&st& Burg ist unser GoW'i 
Die Lieder Luthers in ihrer Bedeutung fur das evangelische Kirchenlicd; 
Gottiogen, 1905. 



Protestant hymn writers. g 

aekker (1530 1592), The last hymn-book to appear in 
Luther's life-time was published in 1545 in Leipzig by 
Valentin Babst; this, in its numerous reprints and pirated 
editions, remained the standard for all evangelical hymn- 
books until the end of the sixteenth century. 

In Bach's organ chorales are found such of these oldest 
hymns as were also included in the later hymn-books: * 

A. SACRED SONGS OF THE MIDDLE AGES: 

(1) Easter Hymns, 

Christ ist erstanden (V, No, 4). 

Christ lag in Todesbanden ffi, No, 5 j VI, Nos. 15 and 16; cantata 

No. 4). 
Jesus Christus unser Heiland, der den Tod (V, No. 33). 

(2) Christmas Hymns. 
In dulci fubilo (V, No. 35). 

Puer natus in Bethlehem (V, No. 46). 

(3) "Improvements" of mediaeval song-paraphrases. 

Da Jesus an dem Krauze stund (The Seven Last Words. V, No. 9). 
Dies sind di& heilgen zehn Gebot (V, No, 12; VI, Nos, 19 and 20), 
Vater unser im flimmelreich (V Nos. 47 and 48; VII, Nos. 52 

and 53). 
Wir glauben all an einen Gott (VII, Nos. 60, 61 and 62). 

(4) Hymns translated from the Latin. 

&r Tag der ist so freudonreich (Dies est laetitiae. V, No. n). 
Christum wir soll&n Men schon (A solids ortus car dine. V, Nos. 6 

and 7). 

Erstanden ist der h&ilg& Christ ( Surrexit Christus hodi& V, No. 14), 
Herr Gott dich Men wir (Te Deum laudamus. VI, No. 26). 
Kowm Gott Schdpfer, heiliger Gcist ( Veni creator spiritus* VII f 

No. 35). 
Komm heilger Geist, Herre Gott (Veni sancte spiritus. VII, 

Nos, 36 and 37). 
Nun komm der Heiden H&iland ( Veni redemptor gentium. V, Nos, 42 

and 43 j VII, Nos. 45, 46 and 47; cantatas Nos. 61 and 62). 

B. HYMNS BY LUTHER. 
(i) Translations. 

Jesus Christus unscr Heiland, der den Zorn Gottes (Jesus Christus 
nostra salus; Hymn of John Huss, Passion hymn. VI, Nos. 30, 
31, 32 and 33), 

Gelobct seist du Jesus Christ ( Grates nunc omnes reddamus; Christ" 
mas hymn, V, Nos. 17 and 18). 



* In the following list the Roman figures indicate the numbers 
of the volumes in the Peters Edition of Bach's organ works. 



10 II. The Origin of the Texts of the Chorales. 

(2) Biblical paraphrases, 

A us tiefer Not schrei ich xu dir (Psalm 130: D& profundis* VI, 

Nos. 13 and 14; cantata No. 38). 

Bin feste Burg (Psalm 46. VI, No. 22 ; cantata No. So), 
Mit Fried und Freud ich fahr dahin (The Song of Simeon, Luko II* 

V, No. 41). 

(3) Original Hymns, 
Christ uns&r mm Jordan kam (Baptismal hymn. VI, Nos. 17 

and 18; cantata No. 7), 
Vom Himmel hoch da homm ich her (V, No, 49, and pp, 92- *ioi ; 

VII, Nos. 54 and 55). 
Vom Himmel kani der Engel schar (V, No. 50). 



C. TRANSLATIONS AND PARAPHRASES FROM VARIOUS 

AUTHORS, 

Attein Gott in der Hdh sei Ehr (Gloria in 6xcelsi$ f by Nicolaua 
Deems, [died. 1541]. VI, Nos, 3 ~u). 

Christe du Lamm Gottes (The Agnus Dei in its simple form. 

V, No. 3). 
O Lamm Gottes unschuldig (The Agnus Dei expanded into throe 

verses, by Nicolaus Decius, V No. 44; VII, No. 48). 
An Wasserflussen Babylon (Psalm 137, Super flumina, by WolC- 

gang Dachstein. VI, Nos. laa and isb). 
Christ der du bist der helle Tag (Christe qiti lux es ct dies, V, 

pp. 60 ft, Partita). 
In dich hab* ich gehoffct Herr (Psalm 31, In te Domino $perctvi 9 

by Adam Reissner, [died. 1562]. VI, No. 34). 
Meine Seele erhebt den Herrn (The Magnificat VII, NOB. 41 

and 42; cantata No. 10). 

Kyrie, Gott Vater (Kyrie Jons bonitatis. VII, Nos. 39*1 and 4oa), 
Christe, aller Welt Trost (ChrisU unite Dei Patris. VII, Nos* 39!) 

and 4ob). 
Kyrie, Gott heiliger Geist (Kyrie ignis divine, VII, Nos, 390 

and 400). 

The really creative period of the hymn begins at the end 
of the sixteenth century. The whole of German poetry is 
impelled upon the religious path. While France, under a 
monarchy conscious of its own goal, is developing into a 
strong national state, in which there springs up a brilliant 
literature, fostered by an art-loving court, Germany is on 
the way to complete ruin. The nation as such disappears, 
and with it that national feeling without which, no true 
literature is possible. When the country relapsed into 
barbarism during the Thirty Years' War, the only thing 



Protestant hymn writers. II 

of the soul that survived was religion, IB its bosom poetry 
took refuge. Thus Germany, in its bitterest need, created 
a religious poetry to which no thing in the world can compare, 
and before which even the splendour of the Psalter pales. 

The hymns of that time are a mirror of contemporary 
events. When the plague ravages, in 1613, the eastern 
parts of Germany, Valerius Hcrberger sings his joyous dirge 
< Valet will ich dir geben, du arge falsche Welt" (VII. 
Nos. 50, 51);* Martin Rinkart's (1586 1649) c ' Nun danket 
alle Gott" (VII. No. 43), is composed while the bells are 
ringing out the conclusion of peace in 1648. 

These hymn-writers are by no means talents of the first 
order, Nevertheless the sincerity of devout feeling and the 
grave beauty of a diction formed by a constant reading of 
the Bible keep the average of the songs fairly high. Per- 
haps all these poets wrote too much. It happens, too, 
with the sacred poem as with the lyric: in one inspired 
song the poet, become for the moment a genius, will express 
magically what in other songs he could only stammer out* 
And this one song will live, Johann Rist (1607 1667) 
composed six hundred and fifty-eight songs; of these five 
or six survived in the hymn-books ** 

Among these hymn-writers were two mystics, Philipp 
Nicolai (1556- 1698) and Johann Franck (1618 1677), 
To those Bach felt himself particularly drawn, for they, 
like himself, were steeped in the atmosphere of the Song 
of Songs. He wrote a cantata on Nicolai's "Wie schon 
leuchtct der Morgenstcrn" (No. i), and another on his 
*'Wachet auf, ruft uns die Stimmc" (No, 140), also basing 

* "Valet wilt ich dir gcbcn"; Ein andechtiges Gebet, dawiit die 
Mv&n$*eli$che BUyg&rschaft zu Frauenstadt Anno 1615 im Herbbt, 
Gott dem Herrn das Hertz erweichet hat, daft er seine schar/fe Zorn~ 
ruthc, untcr welcher bey xweytausend Menschcn schlaffen sind gawgen, 
in Gnaden hat ni&der^elegt, So wol ein trdstlicher Gesang, darinnen ein 
frommes Hertz dieser Welt Valet giebt, Beyd&s gestellet durch Valerium 
Herbergerum* Prcdigcrn beym Kripplein Ghristi* Leipzig, 1614. 

** We may mention also Paul Flemming (1609 1640), Johann 
Hcermann (15851647), and Simon Bach (1605 1659). 



12 II. The Origin of the Texts of the Chorales, 

an organ chorale on the latter (VII. No. 57). He treated 
Franck's "J esu meine Freude" in a motet and in two 
organ chorales (V. No. 31 and VI. No. 29). The communion 
hymn of the same poet, "Schmiicke dich, o lie-be Socle", in- 
spired him to a cantata (No. 180) and to that splendid 
chorale fantasia (VII. No, 49) that sent Schumann into 
ecstasy when he heard Mendelssohn play it on the organ. 
But even in that epoch there are premonitions of decline. 
Subjectivity of feeling and a didactic point of view invade 
religious poetry, and deprive it of that naive, simple ob- 
jectivity that alone can create true congregational songs 
for the church service. At the commencement of the period 
of decay, when feeling and diction are already becoming 
super-subtilised, there appears on the scene, as if to check 
the decline, the king of hymn -writers, Paul Gerhardt 
(1607 1676). His actions show that he was an adherent 
of the Lutheran scholastic, that brought about the real 
Reformation with such startling rapidity. The reformed 
Electoral Prince Friedrich Wilhelm had required the Berlin 
preachers to sign a declaration by which they pledged them- 
selves, for the sake of peace, to treat with moderation the doc- 
trinal differences between the Reformed and the Lutheran 
churches. Paul Gerhardt, in spite of friendly advances 
from the Prince, could not be induced either to sign the 
declaration or to make a verbal promise, and consequently 
had to relinquish his office. The gentle-hearted man, it is 
true, had never employed in the pulpit the violent kind of 
polemic to which the Electoral Prince wanted to put an 
end; he regarded, however, the promise that was demanded 
of him as a sort of treachery to the faith of his fathers.* 
Of his hundred and twenty hymns, more than twenty have 
found^a place in the hymn-books. They breathe a vigorous, 

* Among tlie plentiful literature on the subject of Paul Gorhardt 
which appeared in his commemoration year (1907), Paul Wernles* 
Religionsgeschichtliche Volksbftchw (Halle), calls for special mention. 
We know little more than is narrated above of the life of the poet, 
It is noteworthy how unequal Gerhardt's work is ; he often uses other 
poems as models. 



Sources of the Chorale Texts. 13 

simple piety, and are expressed in a popular diction of 
excellent quality. Even in the life-time of the poet some 
of them came into use in the church, while in Bach's time 
many had become public property. Bach was an admirer 
of Gerhardt, and repeatedly employed verses from his 
hymns in Ids cantatas. In the St. Matthew Passion he 
makes use of five verses from "0 Haupt voll Blut und 
Wunden" and one from "Befiehl du deine Wege".* 

The really creative period of the church song, however, 
had come to an end by Bach's time. Pietism did indeed 
produce some spiritual poetry; but for Bach's work, so far 
as the chorale strophes are concerned, this is of little im- 
portance.** He seized upon the copious treasures of the 
past that lay to his hand in the hymn-books. The following 
figures will give an idea of the increase of the riches at his 
disposal: the little Erfurt hymn-book of 1524 contained 
twenty-six songs; that of Babst, in the first edition, a hun- 
dred and one; Criiger's (that was in use in Berlin for almost 
a century), in its first edition (1640) two hundred and fifty, 
and in its forty-fourth edition (1736) thirteen hundred; the 
Liineburg (1686), two thousand; the Leipzig (1697) over 
five thousand, 

We know from the inventory that has been preserved 
that the eight volumes of the Leipzig hymn-book were in 
the possession of Bach. *** What became of the volumes, 
the leaves of which he must so often have turned over, is 
not known. 



* St. Matthew Passion, Nos, 21, 23, 63 (2 verses), and 72 ; No. 53. 
The hymn M O Haupt voll Blut und Wunden 11 is derived from 
St, Bernard of Clairvaux's "Salve caput cruentatum", 

** The hymn-book of Johaxm Anastasius Freylinghaiisen was 
the first to include pictistic poems ( Geistr&iches Gesangbuch t den 
K$m alter und neuer Lied* enthaltend; Halle, Part. I 1704; Part II 
1714). It was, indeed, the most widely circulated of all the hymn- 
books of the eighteenth century, and its contents increased, in the 
many editions it went through, from six hundred and eighty numbers 
to more than fifteen hundred, 

*** Spitta, II, 278, and III, 267. The Ml title of this hymn-book 
runs: A nd&chtiger Se&kn g&istliches Brand- und Gante-Offw, das 



14 III, The Origin of the Melodies of the Chorales. 

It was unfortunate for Bach's work that the old chorale 
took so prominent a place in it ; for this reason it was in- 
cluded in the censure which Rationalism, in the name of 
purified taste, pronounced upon the church hymn of the 
past. For the second half of the eighteenth century Bach's 
cantatas and Passions did not exist; they had gone into 
exile with the old church hymn. Only after the reaction 
instituted by Ernest Merits Arndt (17691860), Max von 
Schenkendorf (1783 1817) and Philipp Spitta (18011859), 
against the neglect of the hymn-book, had once more 
brought the old poems into repute, were the conditions 
established under which a new epoch could again compre- 
hend the old master and the piety that gave birth to his 
works. Thus it is no accident that it was the son of the 
poet of "Psalter and Harp" who made it his life-task to 
reveal Bach to the world, 



CHAPTER III. 
THE ORIGIN OF THE MELODIES OF THE CHORALES, 

BIBLIOGRAPHY, 
A, KdSTLiN; Luther ah der Vater des evangelischen Kfochengesangs. 

( Sammlung musikalischer Vortrtige und Aufsdtze); Breitkopf imd 

Hartel, Leipzig, 1881. 
PH, WOLFRXJM: Die Entstehung und erste Rntwichtung des dmUchm 

evangelischen Kirchenlieds in musihaliscker J&enehung, Leip- 
zig, 1890. 
JOHANNES ZAHN: Die Melodien der dentschen cvangelischon Kircfan- 

lieder aus den Qmllen geschdpft und mit^eteilt Gtttcrsloh, 

1889 1893, 6 vols. 
FRIEDRICH ZELLE: Das dlteste luthsrisch Haus-Gesan^buch. Gdt- 

tingen, 1903. 
KtfMMERLE; Enzyklopddie der evangelischen Kwchenmusik, Giiters- 

loh, 1886. 



ist ein vollstdndiges Gesangbuch in acht untwschi&dlichm Teilen, 
Leipzig, 1697. The Vollstdndige und vermehrtv LeipM^er Oesan^ 
buch of L. R Werner (1733), containing 856 songs, must also bo 
mentioned in connection with Bach. 



Borrowings from the Middle Ages, 15 

Luther acted with regard to the melodies on the same 
principles as he had done with regard to the text ; he took 
whatever old melody suited his purpose and "improved 46 
it, only the improvement was often more drastic in the 
case of the tune than in that of the words, for it was his 
first care to sec that the melodies were singable and easily 
grasped. 

In 1524, the crucial year for German church-music 
Conrad Rtipff and Jolxann Walther,* two eminent mu- 
sicians, were for three weeks Luther's guests, acting as his 
"housc-precentory" (Kantorei im Hatise). Kostlin, in his 
essay on Liither als der Vater des evangdischen Kirchen- 
gesanges, depicts the trio at their work.** 'While Walther 
and Rupff sat at the table, bending over the music sheets 
with pen in hand, Father Luther walked up and down the 
room, trying on the fife the tunes that poured from his 
memory and his imagination to ally themselves with the 
poems he had discovered, until he had made the verse- 
melody a rhythmically finished, well-rounded, strong and 
compact whole." 

Thus the sacred songs of the Middle Ages preserved 
their own melodies, and the Latin hymns were translated 
in such a way that the new wojrds fitted the old melodies, 
just as in the Enchiridion of 1524. Often, indeed, the poem 
was so constructed as to fit the "tone" of some sacred song 
that was already well known.*** 



* Johann Walthcr was born in Thuringia in 1496; from about 
1523 he belonged to the Torgau precentory of the Electoral Prince. 
When, in 1 5 30, lack of funds compelled the prince to give it up, 
Walther, encouraged by Luther, organised a prcccntory maintained 
by the corporation. After the battle of Muhlberg (1547), which 
made him the sovereign of a new territory, Moritz of Saxony founded 
a Kapelle in Dresden, of which he appointed Walther the head. 
He presided over this until 1554. He then returned to Torgau, 
where he died in 1570. 

** Kdstlm, p, 306, 

*** The summary given on pp. 9 and 10 of the provenance of the 
texts of the old songs applies also to th melodies. 



16 III* The Origin of the Melodies of the Chorales. 

Since we rarely know the history of a melody before it 
became attached to a hymn, the name of which it hence- 
forth bears, it is difficult to decide which melodies were 
adopted and which composed by the musicians of the 
Reformation. In any case we must not under-cstimate the 
number of the latter. Johann Walther in particular seems 
to have employed a rich inventive gift in the service of 
religion. To what extent Luther himself was a composer 
of melodies cannot be determined. Contemporary testi- 
monies, on the strength of which a number of tunes are 
attributed to him, are much too vague to prove anything 
positively. The melody of c< Ein feste Burg", that may 
with certainty be attributed to him, is woven out of Gre- 
gorian reminiscences. The recognition of this fact deprives 
the melody of none of its beauty and Luther of none of th'c 
credit for it; it really takes considerable talent to create 
an organic unity out of fragments,"* 

In his melody to the German Gloria ("Allein Gott in der 
Hoh sei Ehr"), Nicolaus Decius openly makes use of the 
"Et in terra pax" from the "Gloria paschalis". There was 
nothing strange in this to the men who had been brought 
up in the singing-schools of the catholic Church; it would 
indeed be surprising if they had thought it so. We may 
recall the fact that the mediaeval hymn in its turn derives 
from the Gregorian use, 

Nicolaus Hermann, cantor of Joachimsthal, in Bohemia, 
who was both poet and musician, wrote some very good 
chorale melodies . To him we owe "Lobt Gott ihr Christen 
allzugleich" and "Erschienen ist der herrlich Tag",** On 
the whole the number of musicians who wrote melodies for 
the church was not large, not because at that time there 
were no musicians capable of the work, but rather because 

* On this question cf. W. Baumker's article in the Monatsh&fU 
fur Musikgeschichte* 1880. For a counterblast to the excessive stress 
laid by Baumker on the external derivation of the melody see 
H. A. Kostlin's essay Luther als der Vater d&s evang&Uschm Kirch&n- 
ge$angs($ammlungmu$ikali$cher Vortrftge, Leipzig, 1881, pp, 313 fl). 

** Bach. V, Nos. 40 and 15, 



Borrowings from Secular Songs, 17 

their services were not called for. For a new melody to 
become a true folk-melody, of the kind that would gain 
immediate acceptance everywhere, was a difficult process, 
requiring a long period of time. It was much more natural 
to impress existing melodies into the service of the Church, 
sacred melodies at first, and then, when these did not 
suffice, secular ones* The Reformed Church made the most 
abundant xxse of this latter source *. 

Among the church tunes there are as few indigenous 
melodies as there are among those of the people. ; all have 
had an external origin. The learned August Gevaert has 
expressed the opinion that the oldest Catholic church- 
music was transplanted into the church from the pagan 
streets **. 

For the Reformation it was a question of much more 
than acquiring serviceable melodies. While it brought the 
folk-song into religion, it wished to elevate secular art in 
general. That the object was conversion rather than 
simple borrowing is shewn by the title of a collection that; 
appeared at Frankfort in 1571 ; Gassenhaucr, Renter- und 
Bergliedlein, christlich, mor&liter and sittlich verdndert, damit 
die bdse und drgerlichc Weise unnutze und schampare Lied- 
lein auf G&s$&n t Fcldern und in Hdusern zu singen mit def 
Zeit abgchcn mochtc, wenn man geistiga gutc, nutzc Texte und 
Worte darunter haben mdchte. ("Street songs, cavalier songs, 
mountain songs, transformed into Christian and moral 
songs, for the abolishing in course of time of the bad and 
vexatious practice of singing idle and shameful songs in 
the streets, in fields, and at home, by substi toting for them 
good, sacred, honest words.") 

Believing, as he said, that "the devil does not need all 
the good times for liirnself ', Luther formed MR Christmas- 

* For the secular originals sec P. M. Bdhme, AUdeutsohes Lied&r- 
huch: Volhsliedcr der D&utschen nach Wort und Weis& aus dom XII. 
bis xwil XVII. Jahrhundert. Leipzig* 1877* 

** August Gcvacrt, Dcr Ursfintng des v&mischen Kirohen- 

gesaugs, a paper read before the Belgian Academy of Arts, 2/th Oct, 
1889, German translation by H. Rictnarm, Leipzig, 1891, 

SchwcitK$r, B*,ch, 2 



t8 III. The Origin of the Melodies oC the Chorales. 

hymn "Vorn Himmel hoch da konim ich her" out of the, 
melody of the riddle-song "Ich komm aus frcmdcn Landcn 
h er " i n which the singer propounds a riddle and takes 

her garland from the maiden who cannot solve it*. After- 
wards, however, he had to let the devil have the melody 
back again, for even after its conversion it haunted every 
dancing-place and every tavern, In 1551 Walt her ejected 
it from the hymn-book, replacing it by the tune to which 
Luther's Christmas hymn is sung to this day **, 

Reversions of this kind, however, were the exception. 
The majority of the melodies that were dignified by ad- 
mission into the church were able to maintain themselves 
in their new station, and may justly feel aggrieved that in 
all these centuries the tooth of time has not been able to 
make away with the scanty documentary evidence of their 
secular origin. It would have been difficult to detect the 
secular element in them, for age confers on all music 
a dignity that gives it a touch of religious elevation, A 
mystic bond embraces and unites antiquity and religion; 
one clever writer maintains, not without reason, that we 
could mislead all the purists of church music by putting 
before them an old secular motet with an accompanying 
sacred text. 

To give a few examples, Heinrich Isaak's melody to 
"Inspruck, ich muB dich lassen," became the, chorale '*() 
Welt ich muB dich lassen" *** and the foot-soldiers' song at 
the battle of Pavia, the "Pavter-tone" became the 



* See Fr. Zelle, Das MtesU luthcnsche Hau$- Gesangbuch, Gdttirigen, 
1903, pp. 48 50, where the melody is given in the form iu which it 
was first used as a church melody (in King's Ge$anglntch t 1531). 
** Bohme was the first to conjecture that the ground of tho 
ejectment of the first melody was its profane power of resistance. See 
Zelle p. 49. The new melody the one now current (Bach V, No* 49 
and pp, 92 fl) is found in a Leipzig hymn-book as early as 1559. 

*** See Bach's St. Matthew Passwn t chorales Nos, 16 and 44, 
Today this melody is usually denoted by the song "Nun ruhcn oUo 
Walder". It first appears as a chorale tune in the Nuremberg 
hymn-book of 1569. 



Borrowings from Secular Songs, 19 

chorale "Durch Adams Fall 1st ganz verderbt" (Bach V, 
No. 13) ; the chorale melody "Von Gott will ich nicht lassen" 
(Bach VII, No. 56), is derived from the love-song "Einmal 
tat ich spazieren" ; the chorale "Ich hab mein Sach Gott 
heimgestellt" (Bach VI, No. 28) borrows its melody from 
another love-song, <6 Es gibt auf Erd kein schwerer Lcid"; 
the melody "Helft mir Gottes Giite preisen" (Bach V, 
No. 21) had been one of the Table-songs (1572) of Joachim 
Magdeburg. 

In 1601 Hans Leo Hassler (1564 1612) published at 
Nureniburg a Lustgarten neuer teutsoher Gesange, Palletti, 
Galliarden and Intraden mit vier, fiinf und acht Stimnicn, 
("Pleasure garden of new German songs, balletti, galliards, 
and intrades, for four, five, and eight voices"). Twelve 
years later, one of these melodies, the love-song "Mein 
G'nrat ist mir verwirret von einer Jungfrau zart" -appear- 
ed as a chorale tune to the funerary hymn "Herzlich thut 
mich verlangen" (Bach V, No. 27), while at a later date, 
allied with Paul Gerhardt's poem "O Haupt voll Blut und 
Wunden", it became the most important melody of Baches 
St. Matthew Passion. 

Any foreign melody that had charm and beauty was 
stopped at the frontier and pressed into the service of the 
evangelical service. This was the lot of the melody of "In 
dir ist Freude" (Bach V, No. 34), which came from Italy 
in 1591 with the Balletti o Giovanni Gastoldi. So again 
with the little French song "II me suffit de tous mes maux", 
which appeared in 1529 among the Trente et quatre chansons 
musicales of the celebrated Parisian music-engraver Pierre 
Attaignant; it was utilised for the hymn u Was mein Gott 
will, das g'scheh allzeit" *. One wonders if Bach had any 



* It was, however, quite a tragic love-song, as will appear from 
the opening verse: 

II me stiffist do tons mes nuxulx, puis qu'ils rn'ont Iivr6 d 

la mort. 

J'ay cndur6 peiixe et travaulx, tant de clouleur et decomfort, 
Que faut-il que je fasse pour cstrc on votrc grace? 
De douletir inoni coaur est si naort s'il ne voit votre face, 



20 III. The Origin of the Melodies of the Chorales. 

suspicion of this fact when he harmonised the splendid 
melody for the St. Matthew Passion! * 
Other French folk-times came at a later dale into the 

German chorale by way of the Huguenot Psalter, As the 
Calvinist church found no sacred folk-songs already in 
existence, it was compelled to borrow even more largely 
than the German Church. O. Doucn has shown, in his 
interesting work on Clement Maroi ct h Psauticr Hugue- 
not**, the process by which the melodies were compiled for 
the Psalter. Even Calvin had to laugh, for the only 
time in his life when he saw the most frivolous tunes 
walking along, chastely and devoutly, hand in hand with 
the lofty poems of David and Solomon, 

The Huguenot Psalter appeared in its definitive form in 
1562. As early as 1565 Ambrosius Lobwasser, Professor of 
Law in Konigsberg, published a German version of the 
Psalms, adapting them to the hundred and twenty-five 
melodies of the French work, These tunes thus became 
known, and were at once incorporated in the German 
chorale books. The splendid melody of *'Wenn wir in 
hochsten Notcn sind 1 ' (Bach VII, No. 58), comes from the 
Huguenot Psalter; it was probably a French folk-song 
originally ***. 

Only the shameless curiosity that characterises our 
boasted historical sense can rejoice at these discoveries, 
The musician dors not trouble himself about them, and 
forgets them as soon as they are told to him; for they tell 
him no more than what he already knew by instinct 
that all true and deeply-felt music, whether secular or sacred, 
has its home on the heights where art and religion dwell. 

Happy are the chorales of whose origin nothing is known! 
This was the good fortune of the melodies to NicolaFs 

* St. ^Matthew Passion* No. 31, 

** Etude historique, Iitt6raire musicale et bibliographique ; 
2 vote, 1878 and 1879. 

/''** Hiilipp WoUrum gives a comprehensive r6smm& of the origin 
of the songs in his Die Entstetmng und erste Entwickluti^ ties deutsche* 
cvangelischen Kirohentieds in musikalischer Beziehun%; Leipzig, 1890, 



The End of the Creative Epoch. 21 

"Wie sciioa leuclitet der Morgenstern" and "Wachet auf, 
ruft uns die Stimme" (Bach VII, No. 57). Both songs 
appear for the first time in 1598 in the appendix to a treatise 
on the glories of the future life. 

When the treasures of melody to be drawn upon were 
at last exhausted, there carne the epoch of the composer. 
The copious spiritual poetry of the seventeenth century 
called them to the work. There was scarcely an evangelical 
musician of that time who did not compose melodies for 
the church. Almost all the masters whose names adorn 
the history of polyphonic choral music call for mention 
also in the history of the origin of the chorale melodies. 
.And it was with the melodies as with the poems: while a 
great many of a composer's times were destined to perish, 
he managed to breathe the breath of eternal life into 
a few of them, or at any rate into one, which will shine 
in imperishable beauty in our hymn-books as long as the 
evangelical hymn exists. The most notable of these writers 
is Johann Criiger (1598 1662), of St. Nicholas's church 
in Berlin, who devoted his art to the poems of Paul Ger- 
hardt and Johann Franck. The most beautiful of his 
melodies, such as "Jesu mcinc Frcudc", "Schmiicke dich, 
o liebe Seele", u Nun danket alle Gott", occupy a place 
of honour in Bach's work; the very first chorale in the 
St. Matthew Passion, "Herzliebster Jcsu" is by Criiger. 

The spirit, however, which dominated music about the 
beginning of the eighteenth century made it incapable 
of developing the true church- time any further. German 
music got out of touch with German song, and fell further 
and further under the influence of the more '"artistic" 
Italian melody. It could no longer achieve that nai"vet 
which, ever since the Middle Ages, had endowed it with 
those splendid, unique times. Moreover the secular music 
that was then flourishing in the towns and at the courts 
lured it on to new problems, and it could no longer find its 
sole satisfaction in a self-denying co-operation with religious 
poetry. 



22 III. The Origin of the Melodies of the Chorales. 

When Bach came on the scene, the great epoch of chorale 
creation was at an end, like that of the sacred poem, Sacred 
melodies were indeed still written; but they were songs of 
the aria type, not true congregational hymns; an inde- 
finable air of subjectivity pervaded them. 

In this matter Bach too was subject to the laws of his 
epoch. When in 1736 Schemelli, cantor at the castle of 
Zeitz, published through Breitkopf a large hymn-book, con- 
taining nine hundred and fifty-four numbers, he approached 
the famous cantor of St. Thomas's church to beg his 
co-operation. Bach undertook, we are told in the preface 
to the book, not only to revise the figured basses, but to 
compose melodies for the hymns that lacked them. Since 
in this hymn-book, as in the others, the names of the com- 
posers are not given, we cannot be perfectly sure which 
or how many tunes are by Bach. Those, however, which 
we can ascribe to him. with some certainty, since they 
cannot be traced to an earlier date, are rather sacred arias 
than chorales. This description applies, however, only to 
their character, not their beauty; for their peculiar love- 
liness comes from the fact that they are the work of an 
artist brought up on the German chorale, writing under 
the influence of the formally perfect Italian melodic form, 
Any one who has been thrilled by the strains of u Komm 
siiBer Tod" or "Liebster Herr Jesu" knows how unspeakably 
grand these melodies are*. We must not attempt, however, 
to sing them as congregational songs, or to arrange them for 
four voices, for then they wither at once, like the water-lily 
that has been torn from, its home, The heaviness and dull- 
ness that settle on them nowadays when they arc treated 
as chorales, show clearly that tl},ey are not true chorales. 

After Bach, the bonds between the chorale and the 
sacred song are completely broken. The melodies that 
Emmanuel Bach, Johann Joachim Quantss, Johann Adam 
Hiller and Beethoven wrote, in artistic rivalry, to Gellert's 

* The best known edition of these melodies is that of Zahn, 
V ientndzivanzig geistlicke Lieder fiir eine SingsUmm (Gtitersloh). 



Tlie End of the Creative Epoch. 23 

poems only show what a distance separated them all from 
the chorale. 

In the epoch of Rationalism, it is true, the melodies were 
not diluted to the same extent as the text; but there was 
still a hard struggle until the old melodies were again 
rehabilitated everywhere, and were no longer jostled in the 
chorale books by the characterless tunes of the later 
epoch. Now that this has been achieved, the dispute today 
is as to whether we shall retain the old chorales with the 
uniform note-values in which we have received them from 
the eighteenth century, or whether we should restore to them 
their original rhythmic variety, A definite decision, in- 
deed, is hardly possible. Each "pro" that can be adduce^ 
from historical, artistic, or practical considerations is at 
once opposed by a "contra" of equal force in its way. 
Bach is concerned in this controversy to the extent that 
those who advocate the uniform polished form of chorales 
can plead that, although the opposite tradition had a 
powerful following all round him, he felt no artistic com- 
pulsion to revert to the old rhythmic form of the chorale, 
and so there is no cogent objection, from the purely musical 
point of view, against the chorale as we have received it 
from his hands. Against the enthusiasts for the rhythmic 
melodies the old master can plead as St. Paul once did 
against the Corinthians who knew all things so much better, 
that he too thinks he is possessed by the spirit. 

Here are the three typical forms of "Ein feste Burg" : * 

i) Original form of Luther's chorale. 








==r^^ 



* Friedrich Zello discusses the history of this melody in his 
'Studien tiber 'Ein J?e$te Burg' " (Gartner, Berlin, 18951897). 



24 IV. The Chorale in the Church Service, 

2) Luther's chorale in a hymn-book of 1570 (See Wolf rum, p. 216) 













3) Luther's chorale in the form used by Bach. 










CHAPTER IV. 
THE CHORALE IN THE CHURCH SERVICE. 

BIBLIOGRAPHY. 

GEORG RIETSCHBL: Die Aufgabe d&r Or gel im CottesdienstQ bis in 

das XVIII, Jahrhwndwt; Leipzig, 1892. 
R. v, LILIENCRON: Uber den Chorgesang in der evangelischen Kirch; 

. Berlin, 1881. 
A, G. RITTER: Zur Geschichte d&s Qygehpieh, voynehmlich d&s 

deuischen* im XVI, bis zum Anfang des XVIIL Jahrhund&rts; 

Leipzig, 1 88 1. Vol. I. Text. Vol. II, Musical examples, 

See also the works mentioned in the bibliographies to chapters II 
and III. 

How was the congregational song introduced into the 
church service at the time of the Reformation? It is 
usual to look upon the question as very simple, and to 
suppose that the people had little by little come to sing the 
melody while the organ played it. Did the sacred in- 
strument really teach the congregation in this way? 

We may read through all Luther's writings without 
finding a single place where he speaks of the organ as the 



Organ and Congregational Singing in Reformation Times. 25 

instrument accompanying the congregational singing *. 
Moreover lie, the admirer of true church music of every 
kind, gives no directions as to how the organ is to cooperate 
in the service. It is really incredible, however, that in the 
few places where he mentions the organ at all, he speaks 
of it not enthusiastically but almost scornfully! He does 
not look upon it as necessary or even desirable in the 
evangelical service, but at most tolerates it where he finds 
it already. 

His contemporaries shared his view. We need not be 
astonished that the Reformed Church dealt drastically 
with the organs and banished them from the churches. In 
the Lutheran and even in the Catholic churches at that 
time it fared almost the same. It had always had, indeed, 
its adversaries. No less a person than St. Thomas Aquinas 
had declared war on it, not regarding organ music, or 
indeed instrumental music in general, as calculated to 
stimulate devotion. In the sixteenth century, however, 
complaints against it arose on all sides, and the Council of 
Trent, (1545 1563), which dealt with all the doubtful 
questions relating to the church and its service, was com- 
pelled to enact severe regulations against the erroneous and 
too prevalent employment of the organ in worship. Catho- 
lics and Protestants alike at that time imposed on it a 
term of penance, in order that it might alter its ungodly 
nature, in default of which the Church would excommuni- 
cate it* 

It had fully merited this disgrace, The character of the 
tasks allotted to it may be seen from the Ct&remoniale 
Episcoporum issued by Pope Clement VIII in the year 

* The following remarks are in the mam a repetition of the 
views of Geo. Rietschel, expressed in his masterly essay on Di& 
Aufgabe der Or gel im GottesdiensU bis in das XVIII. Jahrhundert 
(Leipzig 1892). Rietschel is the first to have thrown a light on this 
question, since, instead of spinning theories, he lets the documents, 
church regulations, prefaces to hymn-books, sermons at the dedi- 
cations of organs, and funeral orations on organists speate for 
themselves. 



26 IV. The Chorale in the Church Service. 

1600. The organ preluclised in order to give the tone to 
the priest, or the choir. It further gave out the liturgical 
songs and hymns in alternation with the choir, one verse 
being sung and the next played on the organ. It was never 
used, however, to accompany the choir. The primitive 
structure of the organs of that time quite forbade this; 
their heavy keys did not permit of polyphonic playing, 
while their crude, untempered tuning made it as a rule 
impossible to play on them in more than one or two keys. 

Since therefore they could not cooperate, the choir and 
the organ functioned in turns. When the organ had com- 
pleted its verse, the text, in accordance with the above- 
mentioned regulations of the Pope, was either recited 
loudly by a chorister, or else sung, which latter was re- 
commended as the better course *. 

With the organ employed in this independent way, 
abuses could not fail to creep in, As the organist was 
unable to play polyphonically on his instrument, lie was 
tempted to amuse himself with quick running passages in 
his preambles to the verses or during the course of these. 
Still worse was it when he indulged in well-known secular 
songs, which seems to have been a wide-spread practice. 
In 1548 an organist in Strassburg was dismissed from his 
post for having played French and Italian songs during 
the offertory **. 

At a later date the organ unwarrantably deprived the 
choir of many of the hymns, taking almost everything upon 
itself, The extent to which this had become prevalent 
appears from an incident that happened to Luther, which 



* C&remoniale Episcoforwn; Papst Clemens F//I. x 600. Cap* 38: 
Be Organo, Orgaaista et Musicis sen cantoribus et nonna per ^os 
servanda in divinis, "Sed advertendum writ, ut quandocuncquc 
per organum figuratur aliquid cantari sen responded altematim 
versiculis Hymnorum ant Ganticorum, ab aliquo de chore intelligibili 
voce proniantietur id quod ab organo respondendum est, Et laudabilo 

esset, ut aHquis cantor conjimctim cum organo voce clara idem 

** .._ -L__^I_ i> 



cantaret." 

** Rietschel, p. 41 



The Choir and Congregational Singing. 27 

he tells in his best style in the Table Talk: "When I was 
a young monk in Erfurt", he says, "and had to make the 
rounds of the villages, I came to a certain village and cel- 
ebrated mass there, When I had dressed myself and 
stepped before the altar in my fine attire, the clerk began 
to strike the Kyrie deison and the Patrem on the lute. 
I could with difficulty keep from laughing, for I was not 
used to such an organ ; I had to make my Gloria in excelsis 
conform to his Kyrie" 

It seemed so much a matter of course at that time to 
substitute the organ for the choir in the liturgy that this 
clerk, in default of an organ, simply had recourse to the 
lute! 

In the Evangelical -church the rdle of the organ had for 
a long time now been the same as in the Catholic church. 
It preambled to the hymns of the priest and the choir and 
alternated with the latter; only now the congregational 
song is merely an addendum, to which the organ preambles 
atid wherewith it alternates. In Wittenberg it preambled 
to almost all the vocal pieces, whether of priest, choir or 
people, and shared with the choir in the rendering of the 
Kyrie , the Gloria and the Agnus Dei. We learn this from 
Wolfgang Musculus, who in 1536 attended the Concordia 
conferences at Wittenberg, and described the singing at 
the service in the Wittenberg parish church on the fifth 
Sunday after Easter*, 

This explains the curious injunction which we find in 
the church ordinances of the fifteenth and sixteenth cen- 
turies, namely that the organ "shall strike into the song in 
the churches". It means that certain verses are to be 
played by the organist alone, the congregation being silent. 
At the same time the caution is given that this must not 
happen too often, but at the most two or three thaes in 
the one hymn. It is so laid down in the "Strassburg Church- 
ordinance" of 1598,** and, in exactly the same way, in the 

* Rietschel, p. 13; Luther's Tischrecfan, ed. Brlanger, p, 399. 
** Rietschel, p. %\ ff. 



28 IV. The Chorale in the Church Service, 

''Nuremberg Congregation ordinance" of 1600, At first, 
and for another three generations at least, there was no 
question of the organ accompanying the congregational 
singing. 

How did the choir stand with regard to the congregational 
chorale? Did it take the place of the organ, guiding and 
supporting the song of the people? A glance at the earliest 
hymn books appointed for the service shows us that this 
solution also did not occur to Luther, 

The above -mentioned* Erfurt Enchiridion of Justus 
Jonas was a hymn-book not for the church but for the 
home, as, indeed, its title expressly indicates. The melody 
alone was noted over the poem, so that the father of the 
household could give it out to the children and the servants. 
The Strassburg reformer Catharina Zell hoped that u a poor 
mother should go to sleep, and, if at midnight the crying 
child had to be rocked, sing it a song of heavenly things;" 
this would be the right kind of lullaby, and would please 
God more than all the lullabies played on the organ in the 
Catholic church **, 

The Church chorale book published at Wittenberg in 1524 
by Luther and Walther, while the Enchiridion was being 
printed at Erfurt, makes no reference whatever to congre- 
gational singing. It merely consists, in fact, of the vocal 
parts of chorales written in four and five parts, and the co- 
Speration of the faithful is barred at the outset by the 
fact that the chorale melody lies in the tenor, not in the 
soprano***. These vocal parts, which were probably 
engraved by Luther's friend, the painter and wood engraver 
Lucas Cranach are those of chorale motets sung by the 
choir, and therefore having a cantus firmus, as was custom- 
ary in the religious and secular music of that time, 

* See p. 7. 

** Catharina Zell, in the preface to her Gesangb&chMn of 1534, 
See Rietschel, p. 26, 

*** The title of this edition of the vocal parts runs; Geystlicbt 

Gesanck-Bttchtein: Wittenberg, 1524. It contains: thirty-eight 
hymns. 



The Choir and Congregational Singing. Q 

Luther was not only a reformer tut an artist. The 
logical outcome of bis reforming ideas would have been a 
remodelling of the church service on the lines of the simple 
home service, in which case the congregational chorale 
would have been the only music used in the church. This, 
indeed, is the line we find him pursuing in his first drastic 
treatise on the service*. But, as in most men of genius, 
there was a fatal side to his greatness that prevented him 
from thinking out his ideas to their logical conclusion, and 
made him endow a thing and its antithesis with equal life. 
He was an admirer of the contrapuntal music of the Nether- 
lands school. He regarded artistic music as one of the 
most perfect manifestations of the Deity. "When natural 
music is heightened and polished by art", he said once, 
"there man first beholds and can with great wonder exa- 
mine to a certain extent, (for it cannot be wholly seized qr 
understood) the great end perfect wisdom of God in His 
marvellous work of music, in which this is most singular 
and indeed astonishing, that one man sings a simple tune 
or tenor (as musicians call it), together with which three, 
four or five voices also sing, which as it were play and skip 
delightedly round this simple tune or tenor, and wonder- 
fully grace and adorn the said tune with manifold devices 
and sounds, performing as it were a heavenly dance, so that 
those who at all understand it and are moved by it must 
be greatly amazed, and believe that there is nothing more 
extraordinary in the world than such a song adorned with 
many voices." The wonders of contrapuntal polyphony 
have never been so admirably described before or since**. 

His favorite composers were Josquin des Pr6s (1450- 1521), 
the court musician to Louis XII, of France, and Heinrich 



* Ordnung des Gottesdienstes in der Gemeine; 1523. 
** The passage is to be found in the so-called Lobwde Luthers 
auf die Musik, which, however, - as H. Holstein showed in 1883 - 
is simply a preface which he orginally wrote in Latin to Johann 

Waltlier's Lob und Preis der Mniwlischtvi Kiwst Miisica (and ed, 

1564). Sec Riftschcl, p. 36, 



30 IV, The Chorale in the Church Service. 

Isaak's pupil Ludwig Senfl (died 1550), who was success- 
ively in the service of the courts of Vienna and Munich, 
His remark upon Josquin is well-known; "He is the master 
of the notes ; they have to do as he wills ; other composers 
have to do as the notes will," On one occasion, when 
a motet of SenfFs was being performed in his house, lie 
called out: "I could not write such a motet if I were to 
tear myself to pieces, just as he, for his part, could not 
preach a sermon like me" *. 

The musician in Luther could not tolerate the banishment 
of choir and art-song from the church, as many people 
desired, or the restriction of the choir to leading the con- 
gregational singing. "And I am not of the opinion" he 
says in the preface to Walther's chorale parts of 1524, 
"that on account of the Gospel all the arts should be 
crushed out of existence, as some over-religious people 
pretend, but I would willingly see all the arts, especially 
music, in the service of Him who has given and created 
them" **. 

A licence was thus granted to the art in the Lutheran 
service ; it took its place in the ritual as a free and independ- 
ent power. All the phases of the development of music 
in general are to be clearly seen in the Lutheran service. 
Finally, when the motet, under the influence of Italian 
art, was transformed into the cantata, bringing not only 
instrumental music but an undisguised opera-style into 
the church, the service actually came to be interrupted by 
a sacred concert, which was looked upon as its culminating 
point, It was at this juncture that Bach came on the 
scene. On the covers of his scores he writes, not "cantata 1 *, 
but "concerto". 

Thus had Luther not been an artist, Bach would never 
have been able to write his sacred concert -music for 
church purposes and as part of the church service. Would 

* Luther's Tischreden, eci Irmischer, B. 62. 
** Friedrich Zellc, Das dlteste luthensohe 
Gdttingen, 1903, p, 10, 



The Choir and Congregational Singing. 31 

he nevertheless have written it in any case? What would 
he have done had he been born in Zurich or in Geneva? 

At first, then, the congregational chorale was not sup- 
ported either by the organ or by the choir, but sung umsono 
without accompaniment, precisely as in the Catholic church 
at the end of the Middle Ages. 

We must not over-estimate the number of the congrega- 
tional chorales that were sung during a service. Where a 
choir existed, the congregation took little part in the sing- 
ing, being restricted to the Credo, sung between the 
reading of the Gospel and the sermon and perhaps a 
communion hymn. In Wittenberg so it appears from 
the account given by Musculus, the congregation as a 
rule did not sing, but left even the chorales to the choir. 
In other places, Erfurt, for example, it was customary 
for the people to sing alternately with the choir between 
the Epistle and the Gospel, in such a way that the choir 
sang the sequence and the people joined in with a German 
chorale appropriate to the time of the year. Five or six 
chorales in the year sufficed for this, since the same chorale 
was used on each Sunday during that particular period, 

In the churches that had t no choir, more importance 
attached to the congregational singing, since in that case 
the Kyrie, the Gloria and the Agnus Dei were sung in the 
corresponding German chorales. But here again, as a 
rule, fifteen or at most twenty chorales, which had been 
laid down, once for all, for their particular Sundays, suf- 
ficed for the whole year. 

On closer inspection we get the impression that the 
congregational singing, instead of gaining ground, was in 
the course of the sixteenth century driven back by the 
art-singing and by the organ, the pretensions of the latter 
increasing everywhere, in spite of all ordinances *. 

There was thus good cause for the attempt that was 
made, at the end of the first century of the Reformation, 



Rietschel, p. 49. 



32 IV, The Chorale in the Church Service, 

not indeed by a musician but by a priest to improve the 
position of the chorale. In 1586 the Wiirtemberg court 
preacher Lucas Osiander published his FUnffzig geistliche 
Liedef und Psalmen, mil mer Stimmen auf kontrapunkt- 
weise, fttf die Kirchen und Schulen wi IdMicfien PHrstcntumb 
Wiirtemberg, also gesetzet, dass eine gantm christliche Gemcin 
durchaus mitsingen kann" ("Fifty sacred songs and psalms, 
for the churches and schools in the worshipful principality 
of Wiirtemberg, set contrapuntally in four parts in such a 
way that the whole Christian congregation can always join 
in them"). This was the first real chorale book in our 
sense, except that it was written for the choir instead of 
for the organ. Tbe fact that Osiander relics only on the 
choir, not on the organ, for the leading of the congregational 
singing, proves that the instrument in his time had no 
concern whatever with the latter *, 

In his preface he expresses his confidence that lie has 
made things easier by removing the melody from the 
tenor to the soprano, and thinks that when the laity 
recognise the tune they will joyfully take part in it**, 

Was not his confidence misplaced? It was indeed only a 
half-measure, a false compromise between polyphony and 
melody. If he wanted polyphony, he should have allowed 
the whole congregation to sing in chorus in four parts, 
as was the custom later in Switzerland; on the other hand, 
if he wished to do without polyphony, he should have lot 
the choir sing in unison, acting, as it were, as precentor, 
somewhat in the way the village cantors in his day led the. 
chorale without choir or organ, simply by the unison singing 
of the school children. His desire, however, was to reconcile 
artistic singing and popular singing, and instead of a 
solution he achieved only an unstable compromise, For 



* Friedrich Zelle, Das erste evangdische Chovalbwh 
1:586; Berlin, 1903, 

** The fact that among the thirty-eight songs in Walther'w 
vocal parts of 1524 -two have the cantus firmus in the soprano doe 
not, of course, imply that the congregation was expected to take, 
part in these two, 



Osiander and Hassler. 33 

what support could the harmonies of a choir - and the 
choirs at that time were very weak In numbers give 
to a cantus firmus sung by a mass of people? 

Hans Leo Hassler also tried to make a forward step in 
this direction, and published, besides his splendid Cantioncs 
sacrae and Sacri concertus (for performance by the choir 
only), his Kircfaengesang, Psalnwn und geistliche Lieder auf 
Me gemein&n MeloMen mit vicr Stimmen simpliciter gesetzt, 
which, according to the preface, were so constructed that 
the ordinary man could sing them in the Christian assembly 
to figurate music.* 

It would be wrong, however, to suppose that all the 
masters of church music who, in the sixteenth and seven- 
teenth centuries, removed the melody to the soprano part, 
were imitators of Osiander, and that it was for purely 
practical reasons that they abandoned the earlier system. 
The real reason is quite different, and must be sought in 
the fact that in the meantime German church music had 
shaken off the influence of the purely contrapuntal music 
of the Netherlands school, and had fallen under that of the 
Italians, in which the melodic style began to dominate the 
contrapuntal. Melchior Vulpius,** Seth Calvisius,*** Michael 

* Hans Leo Hassler was born in Nuremberg in 1564, The 
Fugger family sent him, when lie was twenty, to Venice, to study 
music with the masters living there. From 1601 1608 he was 
organist and choir-master in his native town; in 1608 he was called 
by the Electoral Prince to Dresden, His constitution, however, 
was already almost ruined by consumption. He died the 8th June 
1612 in Frankfort, whither h had accompanied his master to the 
assembly of princes. 

** Melchior Vulpius was born in 1560. In 1600 he became Cantor 
at Weimar; his death in 161:6 was a great loss to art. Pars prima 
cantionum sacrarum cum VI, VII, VIII, et pluribus vocibus; Jena, 
1602. Kivch&ng&sdng und geistliche Lieder; Erfurt, 1603. A St. Matthew 
Passion of his was also published at Erfurt. 

*** Seth Calvisius, born in 1556, was equally famous in his day 
as philologist, mathematician and musician- He was one of the 
predecessors of Bach at St. Thomas's church in Leipzig, He died 
in 1615. Kivchenges&nge und geistliche Li&dev ZV, Lutheri und andwev 

frommer Christen mit viw Stimmen contrapunktweis richtig 

ge$6tst; Leipzig 1597. 

Schwc i txer, Bach, * 



34 



IV, The Chorale in the Church Service. 



Praetorius,* and Joliann Eccard** thus follow in their 
admirable music not so much the lead of the Wurtemberg 
Court preacher as the trend of the art itself, 

It was a pure accident that through this change in 
polyphonic art the possibility was opened to the congre- 
gation to join in the cantus firmus with the choir. How 
far it availed itself of it we do not know, for in the history 
of art, as a rule, we never get to know the things that would 
be of practical interest to us, for these, being looked upon 
as matters of daily custom, are not recorded* The fact that 
at this epoch the term * 'chorale" begins to be applied to the 
melodies sung by the congregation throws no light on the 
question,*** unless we regard it as proving that by this time 
the melodies of the church song had ceased to be congre- 
gational property and had become the property of the choir. 

In any case the composers themselves, in spite of the 
fine practical suggestions as to congregational singing that 
they put forward in their prefaces, thought only of the choir 
when composing, as is shown by their counterpoint, which, 
with all its simplicity, becomes richer and more and more 
in the style of the motet.f For us these chorale pieces. 



* Michael Praetorius (15711621) was Kapellmeister to the 
Duke of Brunswick. Musae Sionae: Geistliclw Concertgw&ng* 
die filrnembsten Teutsche Psalmen und Luder, ww sie in dw 
lichen JKivche gesungcn werden wit VIII und XII Stimtnen 
(1605 1610, 9 vols., containing 12/14 hymns), 

** Johannes Eccard, born in 1553, studied at Munich with 
Orlando Lasso, Alter he had been for some time Kapellmeister 
to the Fugger family at Augsburg, he entered in 1585 the service 
of the Duke of Prussia at K6nigsberg, where he at once began the 
collection and harmonisation of the melodies in vogue in Prussia* 
His great work, Geistliche Lwd&r auf den Choral, oder die. tfcbrauch" 
liche Kirchenmelodie g&richt&t und filnfstimmig gMtutt (Kdnigsberg 
1597 & 1598) is the result of these labours. In 1608 the Electoral 
Prince Joachim Friedrich summoned him to Berlin. He died in, 16 1 a, 
*** Until then the only term in use had been "sacred song 1 '* , 

t Reference may be made to the preface of Eccard to the 
hymns he published in 1597, and to that of Michael Praetorius to 
the Musae Sionae . Only on thing is clear from their remarks, 
that so far as they are concerned they are merely experiuwnttag, 
See also Rietschel, pp. 54 ft 



The Organ and the Choir, 35 

with their singularly beautiful blending of Italian and 
German art, are choral works pure and simple, and the 
idea of trying again the experiment of letting the con- 
gregation join in them would not occur to us. But if only 
we could hear them even as choral works! When will Ilie 
time come when these treasures are exhibited each Sunday 
in our church services? 

The attempts to have the singing of the congregation 
led by the choir were made abotit the end of the sixteenth 
century and in the first decade of the seventeenth. By 
the middle of the seventeenth century the question is 
settled by the organ assuming this r61e. In 1650 appears 
the Tablature-book of Samuel Scheldt, with a hundred 
chorale harmonisations intended for the accompaniment of 
the congregational singing.* 

This was no thought-out experiment, but a solution 
arising out of the facts, i. e, the progress of organ-building. 
The sacred instrument had in the meantime been made 
more practically fitted for polyphonic playing, and endowed 
with such fulness of tone that it overwhelmed the small 
and weak choirs of that time. Whereas hitherto it had 
accompanied the choir, which supported the singing of 
the congregation, its powerful tone now made it possible 
for it to assume the lead. But again we cannot be sure of 
the date at which the organ began to support the choir 
in the chorale, or when it began to cooperate with the 
choir in general. This was certainly not the case before 
the beginning of the seventeenth century, Vulpius, 
Praetorius, Eccard and the others appear to know nothing 
of it. But as early as 1627 Johanii Hermann Schein, 
Cantor of St, Thomas's church in Leipzig, adds a figured 



* Tabfaturbuch too geistltcher Licder und Psafaien Dcctons 
Martini Lutheri und and&rer $>QttsUgr Mdnn&r, fUr die Herren 
Organistm, mit d&y chvistlichen Kirch&n und Gemeine auf dr Orget, 
desgleichen auch zu XXause, %u $piefan und $u singen. A uf atte Feste 
und Sonntage durchs gantze Jahr. Mit vitr Stimmen componiwt 
von Samuel ScMdt, (GSrlitz, 1650), Scheldt (1587 1654) was 
organist at Halle, He is the real father of German organ music, 

3* 



36 IV, The Chorale hi the Church Service. 

bass, intended for "organists, instrumentalists, and 
lutenists" to the four, five, and six-part chorale pieces 
for the choir in his Cantionale of that year; and this most 
probably points to a joint performance by choir and 
organ.* 

We must not, however, conceive the organ accompani- 
ment to the chorale, as it was practised in the second 
half of the seventeenth century, as a supplanting of the 
choir by the organ in the chorale, The choir, even in 
Bach's time, cooperated in the chorale as in earlier times, 
polyphonically indeed although the organ took the 
lead, as it were a kind of second and stronger choir 
without words. 

This transference of vocal polyphony to the organ 
by means of chorale accompaniment was of cardinal signi- 
ficance to the art of organ music* The chorale was the 
teacher of the organists, leading them from the false and 
fruitless virtuosity of the keyboard to the true, simple 
organ style. From this moment German organ music severs 
itself from that of Italy, France, and the Netherlands, 
and, always under the control of the chorale, pursues the 
path along which, in the course of two generations, it was 
to arrive at perfection, Scheldt, already in possession of 
the true organ style derived from the chorale, sees that 
his life-work consists in combating the "colored** organ 
style of the school of the Dutchman Swcelinck.** 

It is an illustration of how an idea is, in the end, always 
stronger than circumstances, Organ music did not come 

* Bietsch!, pp. 57 and 53. In 1637 Theophilus Stadc, organist of 
St. Loreixz's in Nuremberg, brought out a new edition of Hassler's 
hymns, with a preface in which he dedicates them to his "dear 
and faithful colleagues who, by means of the organ, maintain, the 
congregation m the right tune, height and depth/' This shows 
that at that time, in Nuremberg", the organ participated to some 
extent in the chorale, How it did so, however, cannot b gathered 
either from the preface or from the book itself* But the mere 
employment of the organ with the choir is an interesting fact. 

** Jan Pieters Sweelinck ( 1 540 1 62 1 ) was organist at Amsterdam , 
All the North German organists came under his influence, 



The Organ and the Choir. 37 

to perfection in Paris or in Venice, where everything 
seemed to be in its favour, but among the poor cantors 
and schoolmasters of an impoverished country, as the 
Germany of the two generations after the Thirty Years' 
War was. How small Frescobaldi, the organist of St. Peter's 
in Rome, whose fame among his contemporaries was so 
great, seems beside a Samuel Scheidt, whose name was 
unknown on the other side of the Alps I * 

From the moment when organ, choir and congregation 
together gave out the chorale, it was inevitable that the 
antiphonal method, under which the organ alone performed 
certain of the verses, should sooner or later fall into disuse, 
But of the perfection of these independent organ renderings 
at that time we may judge from Scheldt's Tablatwa nova, 
published in 1624, It consists for the most part of a 
species of variations upon the chorales most generally 
used, the number of variations corresponding to the 
number of verses of the song, and upon the hymns of 
the various seasons of the church year, which at that time 
were still sung in Halle in Latin, and not, as in other places, 
in German. In addition there are liturgical pieces, such 
as the Kyrie, Gloria, Magnijicat, and the Psalmus sub 
cowmunione "Jesus Christus unser Hciland", which are all 
treated in the same way.** 

* At the same time the technical powers of tlxe Roman organists 
were in some respects really extraordinary. It is only lately, when the 
works as well as the names of these men have become known to us, 
that we have been able to estimate these powers. Special reference 
may be made to Alex. Gttilmant's Archives des Mattres da I'orgm, 
which up to the present have contained the compositions of Titelouxe, 
A. Raison, Roberday, L. Marchand, C16rambaxalt, du Mage, d'Aquin 
Gigault, Grigny, F. Coupcrin, Boyvin and Dandricu. A study of 
these works gives one the impression that Bach knew more of them, 
and was more influenced by these composers, than is generally 
supposed, These publications, that supply one of the most important 
chapters in the history of organ, music, should be available in every 
library. Very many of these old pieces are still suitable for per- 
formance, 

** Scheldt's TaUatwa nova is in three parts, The first two 
contain the chorales, in which each verse is made the subject of a 



38 IV, The Chdrale in the Church Service, 

The Celli Tablature that appeared twenty-three years 
earlier is on the same lines, except that it also contains 
the complete " catechism songs". 

How long the custom, testified to in all contemporary 
tablatures, of rendering vocal pieces on the organ alone, 
still lasted after the process of decay had once set in, can 
no longer be ascertained. When we consider the extremely 
numerous arrangements by Bach of the chorale "Allein 
Gott in der Hdh sei Ehr", we are inclined to think that 
even down to his day there persisted, tinder certain circum- 
stances, the practice testified to by Scheldt, of the organ 
responding to the Gloria intoned by the priest at the 
altar. 

As to the position of the congregational singing in Bach's 
time, we have only conjecture to go upon. One thing at 
any rate had been achieved, the number of the hymns 
affiliated to the service had considerably increased. Each 
Gospel had one or more of these allotted to it, so that the 
same ones were always sung on a particular Sunday. They 
were called the Cantica de tempo? e; in the hymn -books 
they formed the first class and were arranged according to 
the Sundaj^s of the ecclesiastical year. The cantor selected 
them himself without consulting anyone else. In our day, 
on the contrary, the hymns are always selected by the 
clergyman, to tally with the spirit of his sermon, 



separate musical treatment, together with variations o the same 
kind upon secular songs, such as the can-Mo Belgica "Wehe, Windgen, 
wehe" (twelve verses) the cantio Gallicd u Est-ce Mars" (ton verses), 
and the German song "Also geht's also steht's" (seven verses), 
The third part, with the Kyri@s> Glorias, Magnificats in the various 
tones, and hymns, is meant to serve as a liturgical annual for organists. 
The hymns are: 

Hymnus de adventu: Veni redtmptor, 

Hymmts de nativit&U; A Solus ortus carding. 

Hymnus tempore quadvagesimali: Christe qui tux $s @t dies. 

Hymnus de resurrection: Vita sanctorum* decus angtlorum* 

Hymnus de sancto spintui Veni creator spiritus, 

Hymmis de sancta Trinitate; Lu%> bmta Trinitas. Cr$do 

(Choralis in Basso), 
Psalmus sub communione; "Jesus Christits unscr Holland". 



Congregational Singing in Bach's Time. 39 

This use of the Cantica de tempore helps us to understand 
how the organists of the time of Pachelbel and Bach came 
to write cycles of chorale preludes for each Sunday of the 
ecclesiastical year. 

Whether the congregation took possession of all these 
hymns and took an active and hearty part in the singing 
of them is, however, another question. It is well known 
that Mattheson and the famous Hamburg musicians thought 
nothing at all of the congregational chorale, and in general 
refused to recognise singing of this kind as music. From 
this we may conclude that it did not occupy a prominent 
place in their churches, and that they, for their part, did 
nothing to encourage it. It must have been the same in 
other towns that had celebrated choirs. The cantata 
that sacred concert intercalated in the service absorbed 
all the interest, and the art-song, as at the beginning of 
the Reformation, had once more triumphed. 

We do not know whether things were better in this 
respect in Leipzig than in other towns. The truth is that 
no remark of Bach's has come down to us to show that, 
in contradistinction to his contemporaries, he felt any 
particular interest in congregational singing. In his 
Passions, at any rate, he does not desire its co-operation, 
in spite of the splendid r61e that he assigns to the chorale 
in those works. If is highly probable that in Bach's time 
the singing of the Leipzig congregations was not so good 
as is commonly supposed. 

Not until the concert style of music was banished from 
the service, in the generation after Bach, and the town 
choirs that had been allotted to the churches ceased to 
exist, did congregational singing become th& characteristic 
and sole service-music of the Protestant church. In the 
epoch of rationalism and pietism the ideal was realised 
which the Reformation had indeed perceived, but, for 
conservative and artistic reasons, had not pursued* However 
barbarously rationalism behaved towards the old hymn, 
it did good work for congregational singing, Its ultimate 



40 V. The Chorale Preludes before Bach. 

aim, of course, was to substitute a new kind of hymn for 
the old, the diction and the ideas of which had by then 
become so antiquated as to unfit it for use as a real congre- 
gational hymn. 

Whether the problem has been really solved by allowing 
the organ to support the congregational singing is doubtful. 
The method has, established itself, because it is practical, 
But the ideal is not congregational singing of this kind, 
directed by, and dependent on, the organ; the true ideal 
is free and confident unaccompanied singing, as in the 
congregational singing of the Middle Ages and of the first 
Reformation period. Perhaps that complete and unfettered 
cooperation of organ, choir and worshippers was, in its 
way, an ideal, towards which we shall some day aspire 
more than we do now. 



CHAPTER V, 
THE CHORALE PRELUDES BEFORE BACH, 

BIBLIOGRAPHY. 

PHILIP? SPITTA: /. 5. Bach. I, 96 etc. 

A. G. RITTBR: Zur Geschichte des Q?gelspi&ls p vwnehmHeh d&s 
deutschen, im XIV. bis zum Anfang des XV 111, Jahrhundefts; 

Leipzig, 1884; 2 vols, 
CARL VON WINTBRPELD : Der va<ngeli$che Kitchengcsang; vol. 11, 

1845- 

FRANZ COMMER: Musica sacra. Botc and Bock, Berlin, 
M, STRAUSS: Alte M&istey des Orgelspwfa; Peters, Leipzig 1904, 

In proportion as the independent performance of the ver- 
ses of the songs and hymns on the organ fell into disuse 
through the logic of events, preludising on the chorale 
became of importance. How this preludising was carried 
out until the time of Scheldt we do not know, since no 
compositions of this kind have been preserved, None even 
of Scheldt's have come down to us. But from his time to 
that of Bach the most notable German masters of the 
organ dedicate their powers not to free composition but 



Samuel Scheldt. 41 

to the chorale prelude, or, as they said in those days, 
to "preambling to the chorales". 

The three great masters in this field are Pachelbel, 
Bdhm and Buxtehude. We cannot, indeed, say that in 
technique they created anything new beyond Scheldt. In 
this respect, indeed, organ music has on the whole not gone 
beyond the Halle master even to the present day; nor can 
we imagine how it is ever to be done. Scheldt was one 
of those men whose penetrating intelligence, when a new 
world opens out before them, darts f through it at once 
from end to end, with the clearness and swiftness of light. 
In his polyphonic chorale verses for the organ he saw himself 
faced by the problem of making the melody stand out 
clearly in performance, with a special tone-colour, not 
only in the soprano but also in the alto, tenor and bass; 
and at a glance he surveyed all possible solutions, and 
embodied his knowledge in those famous pieces which, 
speaking generally, contain everything that we can imagine 
in the way of employing the manuals and pedals in con- 
formity with their special qualities.* 



* See the well-known remarks at the end of the third part of 
the Tdblatwa nova (1624): "If it is a Bicimum" [i, e. in two parts] 
"and the chorale a discant, the chorale is to be played with the 
right hand on the tipper manual or 'work', while the left hand 
plays the two parts on the Rtiokpositiv. If the chorale is a discant 
with four parts, the chorale is played on the Rttckpositiv with the 
right hand, the alto and tenor on the upper manual or 'work* with 
the left hand, and the bass with the pedal, If the chorale is a tenor, 
the chorale is played on the Rikckpositiv with the left hand, the 
other parts on the upper manual or 'work 1 with the right hand, and 
the bass with the pedal. The alto can also, in special circumstances, 
be played with four parts on the Rttchpositiv, but the discant must 
be taken on the upper manual with the right hand, and the tenor 
and the bass on lie pedal in two parts at the same time, but it 
must be particularly arranged so that the tenor does not go higher 
than c, since the d is seldom found on the pedals, and they must 
also not be set widely apart from, each other, only an octave, or 
fifth, or third, for otherwise we cannot fully span such intervals 
with the feet. 

"NB. But this is the finest and most suitable style of all, to 
play the alto on the pedal ; the knack and dexterity however are in 



42 V. The Chorale Preludes before Bach. 

In the third decade of the seventeenth century he 
speaks of the playing of two obbligato parts on the pedal 
as quite a matter of course, and thinks that every organ 
should have a four-feet pedal stop, so that the organist 
may be able, under any circumstances, to play a middle 
part with the feet! 

The path was now traced out for the organ players, 
and, what in the first place was almost more important, 
for the organ builders.* They had only to go forward. 
Middle German and North German organ music outdis- 
tanced that of the Romans and the Southern Germans 
almost at a bound. Spitta rightly says that what was 
regarded in the south as a test of the highest virtuosity 
looks almost like an elementary exercise in comparison 
with the compositions of the northern organists*** 

Thus as regards technique there were no new conquests 
for Pachelbel, Bohm, Buxtehude and their epoch to strive 
after. It was reserved for them, however, to create the 
various forms of the chorale prelude. 

Pachelbel's conception of it is almost the grandest ; *** 
he conceives the chorale prelude as a chorale fugue. Each 

the registering and the colouring in the organ, and knowing how to 
make good use of four or eight-feet tones. An eight-foot tono must 
always be on the positive, and a four-feet tono on the pedal/' 

* On the development of the art of organ building see Otto 
Wangemann, Ge$chicht& der Orgel und Aw Orgtitbaukunst, Demmin 
1880,- and Die Or gel, ihve Geschichte und ihr Bau 9 $rd ed, Leipzig 

1895- 

** Uber J, S. Bach ( Sammlung musikalischef VorM$&; Leipzig, 

1*79). 
*** Johann Pachelbel was born in 1653* From 1674 to 1677 he 

was assistant organist at St Stephen's in Vienna, Later on w 
find him in Eisenach, Erfurt - where h remained twelve years 
Stuttgart and Gotha. In 1695 he became organist at St. Sebald'a, 
in his native town of Nuremberg. He died in 1706. His works are: 

AM Chordle zum Praambulwren, published by Chris-dan Weigel, 
Nuremberg, 1 69 3 . 

Tabulaturbuch geistlichw Gesdnge D. Martini Lutheri und an 
dever gottseliger Mdnner sambt beyg&fttgten Chorat-Fugm durchs 
gantze Jahr. Allen Liebhabern des Claviers componieret von Johann 
Pachelben, Organisten xu St. Sebald in Nttrnbwg, 1704; containing 



Paehdbel, B6bm, Rein ken, Buxtehuda. 43 

separate phrase of the melody is worked out in a fughetta- 
like prelude, at the conclusion of which it appears as a 
cantus firmus. The chorale prelude as a whole thus consists 
simply of separate fugues, which are held together by the 
fact that their themes, taken in succession, form the melody 
of the chorale. It is the form of the chorale prelude as we 
have it, for example, in the two great arrangements by 
Bach of Aus tiefer Not (VI. Nos. 13 and 14). 

This style was very largely cultivated at that time, 
Pachelbel exercised an influence upon central Germany 
that cannot easily be over-estimated. He was not a genius; 
he was not even always clever, and his art is not free from 
a certain stiffness and formality. He had, however, a real 
sense of the dignity of the organ, and communicated it to 
his pupils. That was his greatest service. We must 
remember that in Scheidt, as in Frescobaldi, a secular 
conception of organ music as an art exists side by side with 
the religious conception of it. Even the Tablatura nova 
bears traces of this dualism, for it impartially gives varia- 
tions upon secular songs and upon chorales. With Pachelbel 
this impartiality disappears for ever from German organ 
music. 

If the average level of German organ playing in Bach's 
time was higher than any that has been attained since, 
it was entirely owing to Pachelbel, As a type of that 
generation of organists we may mention Johann Gottfried 
Walther, Bach's colleague at Weimar, whom Mattheson, 
the Hamburg author, called "Pachelbel the second".* He 

1 60 harmonisations of melodies, and 80 small chorale preludes. 
(The manuscript is in the Grand-ducal Library at Weimar.) 

Examples of Pachelbel's chorales will be found in Ritter and 
Commer. A portion of his work has appeared in the Denhm&ler 
d&r Tonkunst in Oesterreich, VIII, Jahrg. Bd, II. 

* Johann Gottfried Walther, born in 1684, was intended for the 
study of law, but at an early age showed such rich musical gifts 
that in 1702 he was made organist at Weimar, In 1707 he became 
organist at the Town Church at Weimar Bach being at that time 
at the Court in which post he remained until his death (174^)- 
He is best known as the author of the Musihcdische L&xihon (Leipzig, 



44 V. The Chorale Preludes before Bach. 

composed, as Mattheson tells us in his Critica musica (1725), 
a complete year ol chorale preludes of the Pachelbel type. 
Two of them, by an error of transmission, were included 
among Bach's works. * This is a testimony to the art of 
them. They are, indeed, written in a correct organ-style, 
that at times shows considerable richness of invention. 
The chorale preludes of Johann Christoph Bach (16421703, 
organist at Eisenach, and Johann Michael Bach (16481694), 
organist at Gehren, (the uncles of John Sebastian), also 
give us an idea of the thorough capabilities of the organists 
of PachelbeFs generation.** 

With all its excellencies, however, the Pachelbel form 
of the chorale prelude labours under the grave artistic 
defect of incoherence. The chorale melody the bond 
that should hold together the separate fughettas, 
cannot really give them intrinsic unity. In the last resort 
they amount to no more than a string of fragments, 

The chorale prelude of the Pachelbel type is, indeed, 
really conceived on choral lines. When the words arc 
added to the melody, the two together weld together the 
separate fugal movements into an effective whole. In the 
chorale choruses in Bach's cantatas that are constructed 
on the model of Pachelbel's chorale preludes. in the can- 
tata Ein Feste Burg for example, the impression of 
homogeneity is very strong. In Pachelbel's chorale pre- 
ludes, where the melody is deprived of the words, this is not 

1732), which contains very valuable articles upon the music and 
musicians of his day. His great collection of chorale preludes by 
all kinds of composers is also important. More than thirty chorale 
preludes of Buxtehude's have been preserved for us solely in Walther'a 
transcriptions, 

* Bach. VI. Nos. 24 and 28, (See Spitta I, 382, II, 37,) 
** Forty-four of Johann Christoph Bach's chorale preludes, and 
seventy-two of Johann Michael's, have come down to us. Wo may 
also mention Friedrich Wilhelm Zachau (died 1714), of the Lieb- 
frauenkirche in Halle, the teacher of Handel, and jfohann Kuhnau 
(1667 1722), Bach's predecessor in the cantorship of St. Thomas's 
Church, Leipzig. Tunder, Buxtehude's predecessor in Lttbeck, ap- 
pears also to have done important work in this field, but unfor- 
tunately little of his organ work has been preserved* 



Pachelbel, Bdhm, Rcinken, Boxtehude. 45 

the case; each fughetta has an independent existence. This 
was perceived by the composers of the time, who con- 
sequently felt themselves justified in elevating the fughcttas 
upon the first lines of the chorale to the status of in- 
dependent compositions. It may be taken as certain that 
a number of the surviving chorale fughettas of Paclielbel 
upon the first lines of the chorale originally belonged to 
complete chorale preludes, and were first separated from 
the whole by the copyist, for reasons of a practical nature.* 

Bohm,** the Liineburg master, has an entirely different 
conception of the chorale prelude. He is under the influence 
of the "coloristic" style of the school of Sweelinck. His 
favorite method is to break the chorale melody up into 
luxuriant coloratura, and to keep this rich and flowing 
paraphrase moving about over a simple harmonic ac- 
companiment of a more or less free nature. He has nothing 
of Pachelbel's formal dignity; his works are all life and 
movement. He makes use of the ba$so ostinato, i. e, the 
continual repetition of a characteristic motive in the pedal, 
a means with which Bach, in his chorale preludes, after- 
wards attained such striking effects. Bach's arrangement 
of Nun kommt der Heiden >Heiland (VIL No, 45) will give 
some idea of the style of Bdhmu 

Only parenthetic mention can be made of the Hamburg 
organist Johann Adam Reinken, who has attained a 
certain fame in the history of the chorale prelude by means 
of two works of appalling length, Heinrich Scheidemann, 
a pupil of Sweelinck, organist at St Catharine's church in 
Hamburg, had been his teacher. Reinken succeeded to 



* This is certainly the case with, a number of the small chorale 
preludes' from th Tablature Book of 1704. See Eitner, Monats- 
hefte filr Musihg$cMchte 1874, and Bitter, G&schiMe d&s Or gel- 
spiels I, 151, 

** Georg Bdhm, born in 1661, became organist in 1698 of St. 
John's Church in Liineburg, which post he filled until his death 
in 1733, Eighteen of his chorales have been preserved. Examples 
will be found in Commer, Ritter, A, W. Gottschalg's collection, 
and Straube's Alt* Meister des Orgelspiels (Peters, 1904). 



46 V, The Chorale Preludes before Bach, 

the latter's post in 1664, and occupied it fifty-eight years, 
until his death in 1722* He was very proud of his two 
lengthy chorales, The first, on "Es ist gewisslich an der 
Zeit", contained two hundred and thirty- two bars; the 
other, on "An Wasserfliissen Babylons", ran to three hun- 
dred and thirty- five. He had the latter engraved on copper. 
The melody, as in Bohm, moves about in color ature of all 
kinds, while the accompaniment, which is more in PachelbeFs 
style, is constructed out of the motives of the separate 
lines of the melody. He uses the double pedal a good 
deal. 

Great as is the technique and the virtuosic ability shewn 
in these two works, from the musical point of view they 
are unsatisfactory. Everything is calculated merely for 
outward effect. The melody is tormented to death, and 
the heater feasts on the dexterity of the torture.* 

Incomparably higher stands the Lxibeck organist Dietrich 
Buxtehude (1637 1707), who in 1668 succeeded the famous 
Franz Tunder at St. Mary's Church, He is the greatest 
organist between Scheldt and Bach, and may, indeed, be 
regarded as the real creator of the German organ toccata,** 
His chorale preludes are chorale fantasias of the most 
varied kinds, from the simplest to the most ingenious. 
In the simple ones the melody goes its way quietly, just 
embellished here and there with a few ornaments, and 
accompanied by interesting and always ingenious harmonies. 
Bach's chorale preludes upon Herzlich thut mich verlmgen 
(V. No. 27) and Liebster Jesu wir sind hier (V. No. 36), are 
written in the style of these simple fantasias of Buxtehude, 

* An analysis of the chorale prelude "An WasserMssen Baby- 
Ions" is given by Hitter. All the same, Reinken is a notable 
artist, 

** The organ toccata derives ultimately from Claudio Merulo 
(i53 2 1604), the great master of the Venetian organ school, which 
in turn developed out of the school of the Netherlands. The toc- 
cata was afterwards brought by Frescobaldi to the highest per- 
fection it ever reached in Italy, In Georg Muffat'd celebrated 
Apparatus musico-orgamsticus (1690) we see the art of which he 
is the last great representative coming to a standstill, 



Bach and his Predecessors. 47 

and give a good idea of his work, except that the copy 
rather idealises the original, since the fervour with which 
Bach fills these little fantasias, is not, as a rule, found in 
those of Buxtehude in anything like the same degree.* In 
the large chorale preludes Buxtehude tears the melody in 
pieces, throws the fragments into the flood of a brilliant, 
animated fantasia, and sends them scudding along, one 
in the soprano, another in the alto, another in the tenor, 
another in the bass, according as his fancy suggests. Bach's 
chorale prelude on Ein feste Burg (VI. No. 22) is wholly 
conceived in the spirit of this virtuosic style using the 
term in its good sense of Buxtehude.** 

Such are the forms of the chorale prelude created by the 
masters of the end of the seventeenth century. From the 
formal standpoint they performed their task to the full, 
since they worked out rigorously all the possible types 
of the species. There are three of these. In the first, the 
whole prelude is constructed out of the motives of the 
melody, in which case the latter is not altered in any way, 
but runs through the whole as a cantus firmus. This is 
the "motivistic" method of PachelbeL In the second, the 
melody is broken up into arabesques, that climb and 
wind like a flowering creeper about a simple harmonic 
stem. This is the "coloristic" method of Bohm. In the 
third the melody forms the core of a free fantasia, as in 
the chorale fantasias of Buxtehude. 

All other imaginable kinds of chorale prelude are only 
intermediate forms between these three main types; we 
may, for example, in a Pachelbel chorale-fugue, lightly 



* Buxtehude's organ works were edited by Philipp Spitta in 
two volumes (Breitkopf and Hartel, 1876 1877). The first con- 
tains the miscellaneous compositions, the second the chorale preludes^ 
A new edition has been prepared by Soiffert. The small chorale 
fantasias number thirty-two, 

** Together with Buxtehude may be mentioned the Husum 
organist Nicolaus Bruhns (1665 1697) a richly endowed artist 
who died young. His chorale prelude on *'Nian teomm der Heiden 
Holland" is given in Commer. 



48 V. The Chorale Preludes before Bach. 

colour and ornament the cantus firmus, or weave motives 
of the melody into the harmonies that support the chorale 
arabesque in the Bohm style, or, lastly, derive the themes 
of the Buxtehude fantasia more or less freely from the 
melody of the chorale. 

Bach found these main types and the intermediate forms 
already in existence. He created no new ones; even 
Brahms and Reger, modern as they are, have not done so, 
for. it is quite impossible. The only difference between 
Bach and his predecessors is that he did what they could 
not made something more than form of them. 

The more we try to see into the development of things, 
in any field whatever, the more we become conscious that 
to each epoch there are set certain limits of knowledge, 
before which it has to come to a halt, and always at the 
very moment when it was apparently bound to advance 
to a higher and definitive knowledge that seemed just 
within its grasp. The real history of progress in physics, 
philosophy, and religion, and more especially in psychology, 
is the history of incomprehensible cessations, of conceptions 
that were unattainable by a given epoch, in spite of all 
that happened to lead it up to them, of the thoughts it 
did not think, not because it could not, but because there 
was some mysterious command upon it not to. In the 
same way, the true history of art is the history of invisible, 
insuperable barriers, which only fall when the due time 
comes, without anyone understanding why this happens 
exactly when it does, and not just as well earlier or later* 
Thus it is incomprehensible that the masters who created 
the types of the chorale prelude did not recognise that 
they were no more than forms, and felt no necessity to 
give life to the form By breathing into it the poetic spirit 
that was associated with the melody. They could not see 
that the chorale prelude, really to answer to its title, must 
be born not only out of the melody but out of the text. 
As the pre-Bachian masters of the chorale movement 
harmonised only the melody, not the text, the inspiration 



Bach and his Predecessors, 49 

of their chorale preludes is a purely musical one, owing 
nothing to poetry. No matter how ingenious their ideas 
may be, they never flow from the text. 

In Buxtehude everything is interesting. Many chorale 
preludes are full of real feeling; in one or two of them, 
indeed, the text is to some extent reflected in the music, * ~ 
in the prelude upon "Durch Adams Fall ist ganz vcrderbt", 
for example, where he more than once suggests the "fall" 
by figures in the bass. On closer investigation, however, 
we discover that these reminiscences of the text are more 
or less accidental, and that lie was as little concerned as the 
others deliberately to take the poetry as his starting-point. 

Thus all they did was really only pioneer work. Perhaps 
we should not know that it was only such, if the greater 
spirit had not come after them, who, almost before he had 
ceased to be their apprentice, comprehended, with the 
intuition of genius, that the true chorale prelude must 
bring out the poetry that gives the melody its name, and 
prepare the hearer not only for the melody but also for the 
contents, a spirit, too, who had the secret of making 
tones speak. 

In no other art does the perfect consign the imperfect 
to oblivion so thoroughly as it does in music. Early 
painting retains its own artistic charm for all time. It 
deals with nature, with reality, and renders it, no matter 
how awkwardly, with a primitive truth that makes so direct 
an appeal to the spectator of all epochs that he himself 
looks at the scene with the child-like eyes of those early 
artists. Music, however, does not depict the external 
universe, but is the image of an invisible world, which 
can only be expressed in eternal tones by those who see it 
in its whole perfection and can reproduce it as they have 
seen it. Anything less than this pales and fades in the 
course of time, even to unrecognisability. It may indeed 
be of historical interest, as the record of an aspiration 
towards a goal; but it has lost the power of giving direct 
artistic satisfaction. 

Schweitzer, Bach. A 



<0 VI. The Cantata and the Passion before Bach, 

This is the experience of everyone who has been affected 
by the chorale preludes of Buxtehude and the other old 
masters. At first he is amazed at the artistic treasures he 
has discovered; when, however, he goes further into them, 
a more sober mood comes oyer him. He realises that he 
has been looking at them comparatively, i. t\, with the 
historical understanding and has appraised them with 
that acquired, idealising justice which is indeed appropriate 
to the investigator, but must be wholly foreign to the 
artist, the criterion of art being absolute and immediate. 

Thus the chorale preludes of the composers before Bach 
are finally, for the modern admirer who wishes to do them 
justice, and even more than justice, no more than what 
they are in themselves, forms that they created for 
the greater master who was to come after them, so that 
he might find them when he needed them, and make living 
things of them. 



CHAPTER VL 
THE CANTATA AND THE PASSION BEFORE BACH. 

BIBLIOGRAPHY. 

PHXUPP SPITTA: /. S Bach, Vol. L 1873. 

C. VON WiNTERFEtD: Der evang&Uscke Rirch0n$@sang im 

Jahrhwidwt; Vol. 3. Leipzig, 1847. 
R. FRBIHERR VON LruENcRON; Liturgi$Gh*mmik& 

dev evangetischm Gottesdi&nsie von 1323 1700; ScWeswlg, 1893, 
OTTO KADE: Die fittest* Passionskomposition bis MUM Jahr i6jx; 

Giitersloh, 1893. 

C, H. BITTER; Beitvdge suv Geschichte cUs Qratorium$; Berlin, 1892, 
JOSEPH SITTARP: JKomp&ndium dw G&schicht d&r 

Stuttgart, 1 88 1. 
FRANZ M. BOHME: Die GescMohte des Oratorium$ fU 

fawn und fasslich dargestettt; and d. Giitersloh, 1887, 
OTTO WANGEMAKN: Geschichte des Qtatoriums; Bemmin, 1881, 
PHILIPS? SPITTA: tteinrich SchtiW Leben und Werhe 

Uche Aufsdixe); Berlin, 1894, pp* i <5o. 

SPITTA: Die Passionen notch den Wf JBvangtlfon von 

rich Schtttz; Leipzig^ 1886. 



The Older Church Music. 51 

PHILIPP SPITTA: Die Anfdnge madrigalischev Dichtkunst in Deutsch- 
land (Musikgeschichtliche Aufsatze); Berlin, 1894, pp.63 76* 

C, STIEHL: Die Organisten an der St. Marienkivche und die Abend- 
musiken m Lubeck; Leipzig, 1886. 

ARREY VON DOMMER: Elemente der Musik; 1862, 

WILHELM LANGHANS : Die Geschicht& dcr Musik des X VII., X VIII, 

und XI X. Jahrhunderts; Leipzig, 1882. Vol.1. 
(The works of the composers mentioned in this chapter have 

almost all been published in the Denkmater der Tonkunst,) 

In the history of the cantata there are two questions to 
be considered, a liturgical one and a musical one. 

How did it come about that in the evangelical church 
service a sacred concert should be inserted between the 
reading of the Gospel and the sermon, that is to say, 
precisely in the place where one would least expect a musical 
interruption of this kind? This is the liturgical question. 
The musical question is concerned with the evolution of 
the old purely vocal motet into the cantata of Bach's 
time, with its arias, recitatives, and rich instrumental 
accompaniment . 

In order to understand how the cantata won its place in 
the church service we must begin at the new arrangement 
of worship in the Reformation epoch. Luther did not 
banish the Mass from the service, but retained it, cutting 
out only the offertorium, the essentially Catholic act 
of sacrifice, and substituting the sermon for it.* 

This alteration did not in any way affect the musical 
structure of the church service, since the great choral 
portions of the Mass, the Kyrie> Gloria, Credo, Sanctus 
with Benedicts, and Agmts Dei, still figured in the 
Protestant worship. in the same place as in the Catholic. 



* See the three famous documents, Von der Ovdnung dcs Gottes~ 
dunstes in dev Gem&ine (1523), Formula Missae at Communionis 
pro eccksia Witteribergensi (1523), and Deutsche M&ss und Ordnung 
des GottesdiensUs (1526). On the German Masses before Luther 
see Julius Smend, Di& Evangelischm d&utschen Mess&n bis m Luthcrs 
Deutscher Me$$@, Gdttingen, 1896. The present account relies 
largely on, the exhaustive study of Liliencron, Liturgi$chmu$i~ 
kalische OesohiMe d&r evangelischen Gottesdienste von 15^3 1700 
(Schleswig, 1893), 



52 VI. The Cantata and the Passion before Bach. 

It is true that they could be replaced by German hymns; 
the Kyrie by "Kyric Gott Vatcr in Ewigkcit",* the Gloria 
by "Allein Gott in der Hoh sei Ehr", the Credo by "Wir 
glaubcn all an einen Gott", the Sanctus by Luther's a J csa i a 
dem Prophetcn das geschah", the. Agnus Dei by **0 Lamm 
Gottcs unschuldig". But in the churches that had choirs 
this did not happen, since Luther himself had thought it 
desirable to retain the Latin choral song, at all events at 
first. In this he was partly influenced by the consideration 
that the Latin song would be a salutary exercise in that 
language for the young. 

From the time of Luther to that of Bach, these great 
musical pieces were common to both Protestant and Catholic 
services. The Protestant cantors composed Masses exactly 
like the Catholics, and the Mass-movements of the Italian 
masters were performed in the Protestant churches without 
anyone thinking it strange* Collections of Masses wore 
published, in which both Protestant and Catholic composers 
were represented. Bach himself copied out a number of 
Italian church-compositions, the copies have come 
down to us not because he had no better way of employing 
his time, but because they were to be performed on Sundays 
at St. Thomas's Church. Thus the distinction between 
Protestant and Catholic church-music, of which we hear so 
much, had not made its appearance at that epoch. 

The service in the churches that followed the Lutheran 
observance was accordingly constituted in this fashion; 
Introit: Kyrie: Gloria: Epistle; Gradual: Gospel; Credo 
(Nicene Creed); Sermon; Communion with Sanctus* Bent* 
dictus and Agnus Dei, This sequence was in the main 
observed everywhere, however much the arrangements of 
the service for Luther had thought the time was unripe 
for imposing a definitive uniformity on it, might deviate 
in detail. The Kyrie, Gloria, Credo* Sjanctus, Benedicts 
and Agnus Dei were always the same for every Sunday* 

* Bach VII, No, 39 (a, b, c) and No, 40 (a, b, c). 



The Older Church Music. 53 

The Introit and the Gradual, however, changed each week, 
since the words of these had to be appropriate to each 
particular Sunday. The Offertorium> which again has to be 
in keeping with the day, had fallen into disuse and been 
supplanted by the German sermon. 

What was inevitable under the circumstances now 
happened : the German hymns, in keeping with the German 
sermon, aimed at expressing the character of each Sunday. 
In the Introit, that Latin antiphonal song between the 
priest and the choir, the hymns could find no place of 
entry. In the Gradual., however, between the Epistle and 
the Gospel, where from the earliest time the verses of a 
German hymn suited to the time of the year alternated 
with the Latin hymn, they could now be sung by the 
choir or by the congregation. How strong the tendency 
was to give the German hymns the imprint of the ecclesias- 
tical season is shewn by the fact that in the course of the 
second half of the sixteenth century each Sunday had 
allotted to it once for all its two or three hymns. A hymn- 
book published in 1566 is entitled Geistliche Lieder nach 
Ordnung der Jahreszeit ausgeteiU ("Spiritual Songs distrib- 
ttted according to the Order of the Season").* 

A still greater significance^ than that of these German 
Gradual hymns, which led from the Epistle into the Gospel, 
became attached to the hymns which were inserted between 
the reading of the Gospel and the Credo on the one side and 
the sermon on the other. They naturally had the closest 
bearing on the Gospel, and were thus specially in keeping 
with the particular Sunday, And since these hymns were 
given to the choir, the congregation singing, between 
the Gospel and the sermon, the hymn "Wir glauben all' 
an einen Gott" nothing stood in the way of writing more 



* See also page 38. The earliest hymn-books shew no distribu- 
tive plan at all; while this arrangement of the hymns according to 
the "order of the season'* corresponds to that of the Missal, which, 
as we know, falls into two parts, Ordinarium and 
de T&mpore et de Sanctis. 



54 VI. The Cantata and the Passion before Bach. 

and more new hymns touching on the Gospel for the day, 
for performance at this point. 

Thus by the side of the German sermon on the Gospel 
for a given Sunday there sprang up a kind of parhelion, 
in the form of a sermon in music. Whatever docs not come 
within the range of its rays is lost in shadow. Musicians 
suddenly became conscious of a greater task before them 
than for ever writing fresh music to the statutory hymns 
of the Mass; there were new poems on the Gospel to be 
set to music year by year. The effect of this freer church 
music on them was to make them practically indifferent 
to the statutory musical portion of the service. The 
same Kyrie or Gloria could be sung every Sunday, so long 
as the motets bearing on the sermon were new and expres- 
sive. So it came about that even in the churches where 
the art had its due place, the Mass was given wholly in 
figurate music only in rare cases, on high Feast-days, 
Ordinarily they were satisfied with "musicising" the Kyrie 
and the Gloria. The remainder was indeed retained in 
the service, but was musically starved, the whole strength 
of the artistic feeling being poured into the sermon-motets* 
Thus the Protestant artists turned away from the old god 
to serve the new one, who was more interesting and promised 
them more reward. They took more pleasure in composing 
new motet -texts than in turning into tone again and 
again, with deadly contempt, the woefully unmusical 
Nicene Creed, They would rather write a whole year's 
sermon-music than one complete Mass. Bach, in fact, 
composed five yearly series and only one complete Mass; 
and when he needed the music for a Mass, he borrowed 
it from the Italians, or from his own cantatas,* He did 
not borrow his cantatas from other composers, although, 
as a matter of fact he could just as well have made use 
of material of this kind as of the Mass-fragments that he 
borrowed. He preferred to compose his own cantatas, 

* See Bach's four short Masses (B. G., Year VIII). 



The Gospel Music. 55 

which was new and therefore always interesting work 
and to borrow the Mass-movements that he needed; for 
the composition of Latin texts had little interest for him. 
In this he was only following the instinct of Protestant 
church-music since the middle of the sixteenth century. 

The first important cycle of Gospel settings appeared 
in 1542. It was Martin Agricola's Sangbuchlein alter 
Sonntagsevangelien; cine kurtze deutsche Segen-Music mit 
sampt den Evangelien durchs gantze Jahr auf alle Sonntage. 
Afterwards Nicolaus Hermann* and Homerus Herpol** 
did valuable work in this field. Texts were provided by 
the preachers Bartholomaus Ringwalt *** and Johann 
Heermannf the poet of "0 Gott, du frommer Gott" and 
"Herzliebster Jesii, was hast du verbrochen". The latter 
entitled his work: Anddchtige Rirchseuffzer, oder Evange-* 
Usche Schliessgldckhin, in den Safft und Kern aller ge- 
wohnlichen Sonntags- und wrnehmsten Fest Evangelien 
Reimweis gegossen und damit seine Predigten beschlossen hat 
Johannes Heermann. 

This ecclesiastical art was free, bound by no tradition 
and cramped by no convention. The task it had set itself, 
of expounding the Gospel in music, was so great and so 
admirable that all the progress in music the whole world 
over seemed appointed only to bring German evangelical 
music nearer to its goal. So Protestant church music 
from the beginning of the seventeenth century to that of 



* Nicolaus Hermann was cantor in Joachimsthal. ^Die Sonn- 
tagsevangelien ilfter das gantz& Jahr, in Gesange verf asset, 1560. 

** Homerus Herpol: Novwn et insane opus musicum, in quo 
textus evangeliorum totius anni, vero ritui ecclesiae correspondens, 
quinque vocum modulamine singulari ind^stria et gravitate expri- 
mitur; Freiburg i. B,, 1555- 

*** Bartbolomius Ringxvalt; The Gospels for each Sunday and 
Feast day, "diirchs gairtee Jahr ncbcn etzlichen BusspsalmeE in 
Roim und Gesangweise vertieret". 2nd ed. 1581, 

t Johann Heermann was a poet of suffering. In his whole 
life he could not recollect having had one completely healthy day. 
As a pastor in Silesia he saw the desolation wrought by the Thirty 
Years' War there. 



5O VI. The Cantata and the Passion before Bach. 

'the eighteenth consciously and deliberately gave itself up 
to all the influences of both religions and secular music, 
from whatever source, without shrinking from or fearing 
anything new, animated only by a holy impulse to creation, 
The Landgrave Moritz of Hesse-Casscl must have been 
moved by a singular presentiment when he came to Mar- 
burg, in 1609, to urge his former chapel-boy, now a law 
student, Heinrich Schutz, to go to Venice with a stipend 
of a hundred thalers for two years, to study with the 
masters living there. In the person of this youth, German 
art itself crossed the Alps, Instead of two years he re- 
mained 'four. His teacher was Giovanni Gabrieli, who had 
such an affection for him that on his dying bed he be- 
queathed a ring to him. Gabrieli died in 1613; Schtitz 
accompanied him to the grave before returning home, His 
other teacher was Monteverde, the creator of the old Italian 
opera, by whose instruction he benefited again in 1628 
when he spent another year in Venice. These two masters 
between them gave German art, which was now sitting at 
their feet in the person of Schtitz, just what it needed for 
its renaissance. From Giovanni Gabrieli it learned a new 
polyphony. While Germany was still under the influence 
of the animated but slender counterpoint of the Nether- 
lands school, and lacked the power to develop it further 
unaided, the three great Venetian masters Andrea Gabrieli 
(1510 1586), his nephew Giovanni Gabrieli (1557- 1613), 
andClaudio Merulo (15321604) liad evolved a style which 
was at once bolder and more singable than that of the 
northern school.* The polyphony is transfigured by its 
melodic quality. Each separate voice really sings, is a 
musical personality. This new style was developed simul- 
taneously in organ music and choral music. At the same 
time instrumental music entered upon a quite new stage; 

* The founder of the Venetian School was Adrian 'Willaort 
(14801562), a pupil of Josquin, His successors at St Mark's 
were Cyprian de Rorc from Mechlin (1516 1565), and Giuseppe 
Zarlino (1517 *59o) who was especially eminent as a theorist, 



Italian Influence on German Church Music, 57 

it began to be independent. Giovanni Gabriel! not only 
used his small orchestra to support tlie choir, but gave 
it short independent preludes to play. The gift of Monte- 
verde, * the first great opera composer, to German art was 
even more precious; he implanted the dramatic sense in it. 
This oldest Italian opera must be absolved from the 
censure that Wagner pronounced upon the later one. It 
was not a loose collection of arias, but really what it claimed 
t be a dramma per musica* Monteverde, of all com- 
posers, perhaps has the closest affinity with Wagner; there 
is some truth in Guido Adler's remark that the creator of 
The Ring slumld really be regarded as a representative of 
the Renaissance, and more particularly of the Renaissance 
opera.** The creators of the stilo rappresentativo, as the 
new kind of music was called, had the same ideals as 
Wagner. For them, as for him, music was 1 not an end-in- 
itself, but served only to express the drama, and it was 
their desire that the orchestra should be invisible.*** Their 
melody, however, took the form of dramatic declamation, 
the inherent expressiveness of which can still move the 
modern hearer most profoundly , Monte verde' s lament of 
Ariadne (Lamento d'Arrianna) may be cited as an example. 



* Qaudio Monteverde (1567 1643) lived from 1590 at the court 
of the Duke of Mantua; in 1613 he became Giovanni Gabrieli's 
successor at St. Mark's, Venice, which post he retained until his 
death, His first opera, Or/eo, was produced in 1607. The Venice 
opera house was not founded until 1637. The creators of the Italian 
opera were the Florentines Giulio Caccini and Jacopo Peri* They 
jointly wrote, for the wedding-feast of Henry IV. and Marie de 
Medicis, a dramma par music entitled Eundice ; it was produced 
on 6th October 1600, which date may reasonably be regarded as 
the official birthday of the opera. 

** Guido Adler, Richard Wagner; Vorlesungen, gehatten <m der 
TJniversitdt Wi*n; Leipzig, 1904, 

*** The score of Orfw has survived in its entirety. Monteverde's 
orchestra consisted of two clavicembali, two orgjarn de legno, two 
contrabassi da viola, ten viole di bra25so one arpia doppia, two 
violini piccioli alia francese, two chitarroni, three bassi da garnba, 
four tromboni, two cometti, one flautino, one clarino, and three 
trombe sordine. See Wilhelm Langhans, Geschichte for Musik de* 
XVI '!., XVHL t and XIX. JahMunderts; Leipzig, 1882, I, 92, 



j5 vi. The Cantata and the Passion before Bach, 

This mighty Renaissance art made its entry through 
Schiitz into the German church, We can scarcely realise 
the enthusiasm with which it was greeted on this side of 
the Alps. What questions and answers there must have 
been when Schiitz and Michael Praetorius who, without 
having himself studied in Italy, looked to that country 
for the regeneration of German art, met in Dresden in 
the latter part of the summer of 1614, where they had to 
provide the music for a christening in the family of the 
Electoral Prince! 

It was Schutz's destiny, during years of restless wandering, 
to carry the new art from court to court, even as far as 
Copenhagen. It is true that in 1617 he was appointed 
Kapellmeister to the Electoral Prince at Dresden, which 
post he occupied for fifty-five years, until his death. But 
from the beginning of the thirties the Kapelle existed 
mostly only in name, the miseries of the Thirty Years' War 
having compelled the court to retrench to the utmost* 
In 1639 the number of the musicians had fallen, from, 
thirty-six to ten.* Salaries were paid when there happened 
to be any money; in the meanwhile Schiite and his sub- 
ordinates had to shift as best they could. More than once 
we find him resident for a considerable time at the court 
of Copenhagen, in the service of the Danish Crown Prince, 
the son-in-law of the Electoral Prince of Saxony* Other 
princely courts also offered him temporary shelter* It 
was in these sad times that he wrote his finest works, for 
which, however, it was often years before he could find an 
engraver. Probably much of it that only existed in manu- 
script has been lost to us, ** We owe the preservation 

* Philipp Spitta, Heinnch Schttte* Leben und WevM, in Mu$ik*> 
geschichtliche Aufsdtee, Berlin, 1894, p. 24. 

** Spitta (SMte, p, 37) says: "A quantity of valuable music 
perished by fire in Dresden in 1760 and in Copenhagen in 1794; 
in Gera too, the great fire of 1780, which reduced all the churches 
to ashes, probably destroyed the numerous compositions of Schutz 
which were preserved there," Perhaps it is only because of this 
accident of destruction that we possess no organ works by SchiHx 



Schiitz. 39 

of the Seven Last Words to the Cassel library; other works 
were preserved at WolfenbiitteL 

Even when the war was over and Schiitz settled down 
again definitively in Dresden, he could not succeed in 
re-organising the Kapelle, All the personal sacrifices he 
had made, in order to keep together through the period of 
misery at least a nucleus of young musicians, in expectation 
of a better time, seemed to have been in vain. When the 
new Electoral Prince, George II,, who took a greater 
interest in music than his father had done, came to the 
throne in 1656, the Italians whom he favoured threw the 
old man of seventy into the shade. In deep dejection he 
cursed the day when he devoted himself to music and 
entered the service of princes. He would willingly have 
gone to some other large art-loving town he thought 
indeed of Hamburg if the distress of the Prince's musicians 
and the infirmities of age had not retained him in Dresden. 
But the art that he cursed, as Jeremiah did his prophetic 
calling, sustained him. The old man wrote four great 
Biblical "Histories", a Christmas History (1664) which 
has been almost entirely lost, a St. John Passion (1665), 
a St. Matthew Passion (1666) and a St. Luke Passion. He 
passed away gently in the afternoon of the 6th November 
1673, during the singing of the friends who surrounded his 
couch. His pupil Christoph Bernhard, cantor at St. Jacobf s 
in Hamburg, had, at his request, his own strength 
being insufficient for the work, sent him his funeral 
text, the passage from the Psalms "Deine Rechte sind 
mein Lied in meinem Hause", arranged as a five-part 
motet, for which Schiitz; had thanked him.* 



The music to his Daphne, the first German opera, is also lost. Only 
the text has been preserved ; it is a poem by Opite, founded on the 
Italian text of Ottavio Rinuccini, The opera was produced in Torgau 
in 1627, on the occasion of the marriage of the eldest daughter of the 
Electoral Prince to the Landgrave George II. of Hesse-Darmstadt. 
* The chief works of Schiitz are: Psalmen Davids samt ettichen 
Motetten und Conowten (1619); Historic dw fvdhlichen und sieg* 
Aufer&tehung unswes tinxigtn Erldsers und Seligmachets 



DO VI. The Cantata and the Passion, before Bach. 

To our surprise we do not find among Schiitz's works 
just what we should have most expected to find complete 
yearly series of compositions on the Gospels for each 
Sunday. If he did not write any it was because the only 
kind of text which, as an Italian of the new school, ho 
could use for this kind of work, did not appear in -Germany 
until he was an old man. The texts for the Gospel music 
were strophic hymns, and were in no way distinguishable 
from the texts of the congregational hymns. The only way 
they could be composed was as sacred songs or motets 
with a chorale-like cantus firwvus. Italian music, on the 
other hand, demanded a poem of much freer construction 
the madrigal.* 

The music of a Montcverde had been a re-birth from 
musical declamation. Bein^ in essence the negation of 
song, it could not employ any so^g-text that was compressed 
into an artificial or monotonous verse-metre; what it 
demanded was a free rhymed prose, in which rhyme and 
metre existed only as servants of the music. With this 
aversion to rhyming lines of the same length, that broke 
up the musical tissue in an obviously unnatural way, music 
entered the path that was to lead it, after a couple of 
centuries of wandering, to the style of Wagner. 

A fundamental problem now presents itself which, in 
the last resort, runs through the whole history of music. 



Jesu Christi (1623); Cantiones sacra& (1625); Beckers gerrimte 
men (1628); Symphomae $acm& f Part I (1629); Kleine 
Concerto, Part II (1639); Die sieben Wovt& %m$&y&$ Erl6$ws und 
Seligmachevs Jesu Chnsti 9 so Ev am Stamm d$ keiligen JKwutxes 
gesptochen* gantx beweglich gasetzt (1645); Sympkonw sacra , Part II 
(1647); Geistlicht Chovmusik (1648); Symphonic sacrae, Part III 
(1650); Ziu6lf gnstliche Gesfag* (1657); Joti&nnespassion (1665); 
Matthduspassion (1666); Lukaspassion (1666?). According to Spitta 
the St. Mark Passion is not by Schtitz. See Spitta's complete 
edition of his works (Breitkopf and H&rtel). 

* On this point see Spitta, Bach, Vol. I; also his Du Anfdng 
madrigalischer Dichtung in DeutsMand (Mu$ib$e$chichtlich$ Auf$&&80 9 
Berlin, 1894), pp. 62 ff. The madrigal is Italian in origin and denotes 
originally a pastoral poem, the name being obviously connected 
with mandr<i f a flock, (Spitta, p. 63.) 



The Problem of the Texts. The Madrigal. 6l 

If the opera begins to decline immediately after Monteverde, 
ultimately to degenerate into a loosely~knit, undrainatic 
string of arias, the fault lay not with music but with poetry, 
which offered it neither an adequate matter nor an adequate 
form, but went on its own way, merely dropping now and 
then something towards which the languishing art ccmld 
stoop as St. Peter did towards the cherries.* 

The history of sacred music until Bach is likewise exclu- 
sively a history of the musical texts, and so it forms the 
true pendant to the history of the opera, which is equi- 
valent to saying that the history is a tragic one, 

JThe madrigal , says Caspar Ziegler** in his treatise of 
1653 upon this text-form, is a short, epigrammatic poem, 
in which the culminating effect ("Konklusion") resides in 
the last two rhymes or even in the last line alone. The 
preceding lines may be of any number, and each may be 
just as long as the poet likes to make it, but usually of 
seven or eleven syllables, some rhymed, some unrhymed. 
"A madrigal must not be constrained in form; it will often 

be more like ordinary speech than a poem I must 

however mention in conclusion that no single Genus 
carminis in the German tongue suits music better than 
a madrigal. For the union of the two gives the best possible 
results; and since the words can be set so finely in their 
natural construction, the harmony also is so much better 
and more agreeable,"*** 

According to Ziegler, these madrigals were sung in the 
stylo recitatiw* When a great many of them are strung 
together, it is desirable, he thinks, to let an arietta or an 
aria of several stanzas "run between", so as to get the 



* The reference is to Goethe's poem Legends vom Hufeism [Tr.], 
** Caspar Ziegler, born at Leipzig in 1621, was a theologian, and 
jurist ; he was professor in Wittenberg at the time of his death (1690), 
The title of his treatise rims thus - Von d&n Madrigatm, inev 
schdn&n und xiw Musik bequemsten Avt Vers, wie sie nach dw 
Italwner Maniev in unswer d&utschen Sprache amxuarbeiUn, n&- 
"benst etlichen Ex&mpdn (Leipzig, 1653; 2ia< ^ edition 1685). 
*** See the complete citation in Spitta, pp, 65 and 66* 



6s VI. The Cantata and the Passion before Bach. 

needful variety. With Ziegler, however, recitative does 
not mean the bare recitativo secco* of the later Italian 
opera or of the Bach Passions, but the dramatic melodic 
recitative of Monte verde, resembling our arioso* In the 
same way his arietta or aria is far removed from the later 
formal da ca<po aria, also used by Bach; it denotes simply 
a melodic piece of a more pronounced song-like structure 
than the arioso.** 

Thus for Caspar Ziegler the ideal text is one that permits 
the musician to pass by means of a continuous "affektvoll" 
(passionate) as the term then was declamation from 
melodic recitative to pure melody and back again. He 
thus lays down the whole ideal of declamatory music 
There was lacking only the poet to realise the ideal and 
to give the musicians texts of this kind for their Gospel 
music. In that critical moment, however, German literature 
was not sufficiently advanced, and later, when it perhaps 
might have found it possible, music and poetry had drifted 
so far away from each other that each had lost sight of 
the ideal of co-operation. 

How admirably suited the madrigal style is to music 
may be seen from Bach's St. Matthew Passion. The texts 
of the ariosos that precede the great arias arc in the madrigal 
manner. They consist of a succession of freely arranged 
verses which are only preliminaries leading up to the 
ultimate "Konklusion". For example; 

"Mein Jesus schweigt zu falschen Liigcn stille, 
Um uns damit zu zeigen, 
Dass sein erbarmungsvoller Wille 
Fiir uns zum Leiden sei geneigt, 
Und dass wir in der gleichen Pein 
Ihm sollen ahnlich sein 

und in Verfolgung stille scliweigen/ 1 



* Secco really means "dry", in contradiction to 
the animated melodic declamation of Monteverd with instrumental 
accompaniment. 

** The da capo aria consists of a main section and a subordinate 
section, after which the main section is repeated, 



The Problem of the Texts. The Madrigal, 63 

or again: 

"Er hat uns alien wohlgetan. 

Den Blinden gab er das Gesicht; 

Die Lahmen macht er gehend; 

Er sagt tins seines Vaters Wort; 

Er trieb die Teufel fort; 

Betriibte hat er aufgerieht; 

Er nahm die Sunder auf und an. . . 

Sonst hat mein Jesus nichts getan/* * 

These are the attempts at a musical-poetic style that we 
meet with here and there in Bach's work. But where else 
has the master's art such freedom of speech, where else is 
it so incomparable? 

The most characteristic German madrigal of the seven- 
teenth century is the "Hirtenlust" of the poet-musician 
Johahn Hermann Schein, a friend of Schiitz. This work 
appeared in 1624; curiously enough Ziegler does not 
mention it, although it represents the ideal of the decla- 
matory song for which he is pleading. Schein also left a 
collection of sacred madrigals with the title Israels Brilnn- 
lein (1623).** 

Schiitz received Ziegler's treatise immediately after its 
publication, and he sent a friendly letter to the author, 
who was a relation of his. In this he heartily wishes him, 
in the name of music, good luck in his exertions for the 
German "madrigal". "The German composers", he says, 
"who until now have repeatedly tried to set to music in 
good style the beautiful inventions of the new poetry of 

* Of the same type are the ariosos t( Du lieber Heiland du, 
wemx deine Jiinger thdricht streiten" (No. 9), "Wiewohl mein, Herz 
in Tr&nen schwimmt, dass Jesus von uns Absohied nimxnt" (No. 18), 
"Der Holland fallt vor seinem Vater nieder (No. 28), "Erbarm 
es Gott! Hier steht der Heiland angebundenl" (No. 60), "J& frei- 
lich, will unser Fleisch und Blut zum Kreuz gezwungen sein (No. 65 ), 
"Ach Golgatha, -unsel'ges Golgatha!" (No. 69), and "Am Abend, 
da es kuhle war, ward Adams Fallen offenbar" (No. 74). 

** Professor A. Prufer of Leipzig has opened my eyes to the 
significance of Schein. See his collected edition of Schein's works 
(Breitkopf and Hartel, vol. 3, 1907). 



64 VI. The Cantata and the Passion before Bach. 

to-day, have yet always lamented that the genus Poeseos 
which best suits the making of a skilful composition, 
namely the madrigal has hitherto not been seized iipon 
by them, but neglected".* 

Schiitz thus did not feel himself to be in a position to 
devote himself to the strophic form of text, ** With him 
begins the estrangement between the art-song and the 
chorale, which hitherto had been harmoniously united in 
the chorale motet. For Schute, in contrast with the 
composers of his time, the harmonisation of the chorale 
has no interest whatever. Nor was he greatly concerned 
about congregational singing. It is quite a mistake to 
intersperse his Passions, when we perform them nowadays, 
with chorale verses, or even with congregational hymns. 
Schiitz himself never thought of doing so, 

In the course of time, this estrangement developed into 
a bitter war, Mattheson, the famous Hamburg contem- 
porary of Bach, wears his pen to the stump*** in proving 
again and again that the true church music must get rid 
of the chorale in particular and the strophic song in general^ 
since the strophe interrupts the musical development, 
and in general is to be regarded as the "maladie de la 
m61odie", the French play upon words being Mattheson*s 
own. 



* See Spitta, p. 73. In the original the citation forms a sub- 
ordinate sentence, 

** This does not moan that n did not occasionally compose 
strophic songs. He did so, for example, in the Aria d$ vita fugaci* 
tate, written on the death of his sister-in-law (1625), in which ho 
treated the chorale "Ich hab main Sach Gott heimgostellt". (See 
the two arrangements in his Collected Works, XII, No. 3 and VI, 
No, 24.) He also set to music in 1628 the rhymed Psalms ol Cor- 
nelius Becker, in the form of sacred songs for chorus. (Collected 
Works, XVI.) We meet with strophic texts here and there in other 
works of his. Professor Prufer of Leipzig, the highest authority 
on the German song of that epoch, has shewn me the significance 
of the German secular madrigals of SchUte (vol. XV of the complete 
edition of his works). They are in the concerto style, independent 
instrumental parts being added to the figured bass. 

*** Cyitica mmica, 1732, 



The Problem of the Texts. 65 

The problem, then, that gave such trouble to all the 
composers of German church music before Bach, and 
indeed to him as well, already faced Schiitz in its acutest 
form. Jhe new music cannot work with the old-style 
strophic songs upon the Gospels, for these are quite un- 
dramatic. On the other hand, the poetry of the time is 
unable to give music the dramatic texts in madrigal form 
into which it could throw itself and gratify its new-born 
dramatic instincts. The dramatic-musical representation 
of the Gospel for the day remained an ideal of the future, 
which Bach and his contemporaries tried to realise in their 
cantatas. Schiitz, therefore, disregarding the Gospel cycle 
and contemporary Gospel poems, has direct recourse to 
the Bible itself, plunging into its treasures in the hope of 
finding the dramatic texts which he cannot get from the 
poetasters of his time. He composes Psalms, detached 
verses from the Bible, whole dramatic fragments. Where 
the Bible does not of itself provide him with something 
musically dramatic, he makes it by arranging the finest 
passages in dialogue form. The Pharisee and the publican 
go into the temple; a prophet exhorts his people; King 
David laments over his son Absalom; above the prostrate 
Paul is heard the voice calling from above, until the 
questions die away in the sky; the Saviour on the cross 
speaks the Seven Last Words. Where is there, indeed, 
such a German requiem as the funerary music woven by 
Schiitz out of Biblical texts and songs? * Thus his texts 
have escaped the doom that has overtaken the poetry of 
his time. The music has not, as with Bach, to cover with 
its own splendour' the nakedness of the words; here it is 
only the artistic setting that brings out more fully the 
brilliance of the precious stones of the text* 

* This music was composed at ihe death of Prince Heinrich 
Postumus von Reuss, Schilta's sovereign, and performed at his 
funeral on 4th February 1636* When the Prince felt himself to 
be near his end, he had a coffin made, the lid and sides of which 
h ordered to be covered with his favorite Biblical hymns. Schdtx 
compiled his work from these. See Spitta, p. 17. 

SchwoUxr, Bach. - 



66 VI, The Cantata and the Passion before Bach, 

If Schiitz, in so far as he goes back for his texts almost 
exclusively to the words of the Bible, is anything but an 
innovator, on the other hand, from the musical point of 
view he must really be regarded as a revolutionary. In 
German church music, (J-abrieli's manner of employing 
several choruses in order to get massive dramatic effects 
was a revolution, upon which Schiitz discourses in the 
preface to his Psalmen Davids (1619). So too was the 
independent employment of the orchestra in the works 
Schiitz wrote after his second Italian sojourn, when he 
became acquainted with Monteverde's music. Another 
revolution was the introduction of recitative-like solo 
songs. In order to grasp the significance of this innovation, 
we must realise that hitherto, in Protestant church music, 
the Biblical words had always been rendered by a single 
voice in the psalmody, or as it was then called, the "collect 
tone" i, e. the monotonous recitation, not yet divided into 
bars, which had been customary in the Catholic church, and 
which Luther and Walther had taken over for their own 
church service with a few trifling modifications It was 
in this old style that the Evangelist still recited in the 
Historic von der frdhlichen und siegreichen Auferstehung 
of 1623.* But from the moment when Schtite became 
acquainted with Monteverde's recitative4ike arioso, he 
employs this, and does not shrink from setting in this 
manner even the Seven Words of Jesus on the Cross. 

All this is revolutionary. In the last resort, however, 
it is merely the visible expression of the essentially revolu- 
tionary thing in Schiitz's art, the ideal he had set himself 
of making music characteristic and pathetic. He may 
renounce choral effects, instrumental accompaniment, even 
dramatic recitative, as he does in his last works, the 
Passions, where the choir sings unaccompanied and the 



* In the preface to this work, Schiitz tells us the manner in which 
he wishes this recitative to be accompanied on the organ or other 
instruments. 



Schiitz's Achievement. 67 

story of the Passion is recited to the "collect" tone; but 
his art remains the same.* 

The new form, startling as it is, and great as was the 
transformation it wrought, is only the servant of the new 
spirit. And Schutz's art is primitive art, but of such a 
kind that it cannot be surpassed by any later art, precisely 
because it is not form but spirit. As we find it hard to 
part from the first spring days and pass into the season 
of full unfolding and ripening, so we tear ourselves al- 
most regretfully away from this primitive art, with its buds 
full of the coming wealth of ideas and forms, in order to 
see what ultimately became of it. In art, as in everything 
else, is not all unfolding and ripening a kind of withering, 
since in the full bloom we no longer have truth and reality 
appealing, to us with that mysterious directness that is 
more magically eloquent than even perfection itself? For 
primitive art of this kind, the later product, perfect though 
it be, is not a heightening of something less perfect, but 
merely the revelation of all that was latent in the primitive 
organism. 

Bach did not know Schiitz *s music, or if he did, he did 
not greatly value it. He copied out the works of all kinds 
of previous and contemporary masters; of Schutz not a 
line has come down to us in Bach's handwriting. As 
Spitta says, Schiitz's relation to him cam only be conceived 
as an ideal one.** He does not stand on Schiitz's shoulders, 
but is unconsciously nourished by his work, just as in 
nature a new vegetation draws its sustenance from the 
invisible but still active forces of its buried predecessor. *** 



* Spitta well says, that Schutz in the "collect" tones of his 
Passions, "has written the most expressive recitative of his time" 
(p. 52). 

** Spitta, Schiite, p. 59. 

*** See also Spitta' s admirable lecture on Handel, Bach and Schiite, 
delivered in 1885, and published in 1892, in Breitkopf and Hartel's 
Sammlung mitsihali&chw VorMge, The article on Schittz in Walther's 
Musiklexikon of 1732 (p. 559) is very interesting. It shows that 
Schutz was indeed a celebrity in the eyes of Bach's contemporaries, 

5* 



68 VI. The Cantata and the Passion before Bach. 

-With Schiitz, "concert-music" made its way into the 
church, and transformed the motet into the cantata. The 
new form is met with under various names; it is called 
indifferently Motetta, Concerto, Symphonia, or Dialogue, 
The term "cantata", in the usual meaning of the word, 
does not come till later; even in Bach's time this term was 
almost exclusively employed to designate the solo cantata. 
He himself entitles one of his first cantatas, the "Rats- 
wechsel" cantata for Mxihlhausen (1708, No. 71), a "mo- 
tetta". In the church ordinances, the place for the per- 
formance of the cantata is indicated simply by the words 
"Hernach wird musiziert" ("here follows music").* 



but that they had no idea of his quality and importance. Schiitaj 
remained a long time forgotten. Attention was first drawn to 
his work, and the significance of it shewn, by Winterfeld in his 
GesMchte de$ evangelischen Kirchengesangs (1845), although he did 
the work much less than justice, estimating the past by the Barrow 
ideal he had before him of Protestant church-music, which he found 
in Eccard and the masters of that time, i. e., in the represen- 
tatives of the pure undramatic vocal style, more fully than in 
Schiitz. In Winterfeld's opinion Schutz begins the epoch of de- 
cline, which continues to Bach. This theory of Winterfeld, how- 
ever, is not quite so narrow and false as it is generally represented 
to be. There is a good deal of truth in it, for Italian art in the end 
did indeed lead German church music into the wrong path. There 
is nothing of this, however, in Schutz. At a later date, Philipp 
Spitta became the advocate of Schutz, and erected a living monu- 
ment to him in his edition of his works. 

Karl Riedel, of Leipzig, was one of the first to perform. Schutz '& 
works, with the famous choir that he founded in 1854. He did 
not do the best thing possible, however, by compiling a new Passion 
out of those of Schiitz, instead of producing each of these in its orig- 
inal form. 

* A definite study of the musical history of the cantata is still 
lacking. It cannot be written until the numerous compositions 
that bear upon the question are sifted out and the most valuable 
published. Whether we shall then be able to shew a connected 
evolution, is as yet doubtful. 

The question of the continual Italian influence on the German 
cantata is also a difficult one to settle. Ludovico Grossi da Viadana 
(1564 1645), Schutz's older contemporary, influenced him and 
the German masters very strongly. He was the first to lay down 
the principle that vocal writing should be built on the foundation 
of the figured bass. His celebrated Canto Concetti ecclesiastic* 



Ideate of Church Music in the Generation after Schtitz, 69 

But if the name varies, the fact is indisputable; in place 
of the sermon motet it was now permissible to perform 
choral works with soli and orchestra. This innovation had 
come about without a single voice being raised in opposition. 
This was possible because it was an epoch of living ideals. 
In the German towns of that time, large and small, we find 
ideals such as have animated no other citizen communities 
since the time of the ancient Greeks. Fatal as it was 
from the political standpoint that religion had become 
an affair of States and communes, yet by this very means 
those ancient conditions were renewed in which the citizen 
community regarded it as its highest civic duty to look 
to the artistic form of its own religious service. The service 
is the concern not of the church but of the town. It is 
not the consistory that engages the cantor and appoints 
the singers and the instrumentalists for the church, but 
the town council and the citizens. The reputation and 
the credit of the town are involved in having an artistic 



were published at Venice in 1602 ff. In Germany, Michael Prae- 
torius in particular cultivated and developed the style of Viadana. 
Praetorius demands the harmonic figuring of the bass, which Via- 
dana had not yet supplied in his compositions. On the path thus 
indicated by the organ harmonies, the voices could now move 
much more freely than before. From that time the pure a capetta 
style passed more and more out of use. Even in Bach's time, no 
choral work is performed without the organ. 

At the same time, however, composers necessarily began to lose 
more and more the feeling for the pure vocal style. Even with 
Gabrieli, Viadana, Schiitz and Praetorius the vocal writing has 
already an instrumental tinge ; later on it became* quite instrumen- 
tal. The pure vocal style did not exist for Bach and his contem- 
poraries. 

The cantatas and oratorios of Giacomo Carissinii (1604- 1674, 
Maestro di capella at Rome) had no particular influence on German 
church music, which at that time was developing wholly under 
the stimulus it had received from Schutz and Praetorius. To Caris- 
simi's pupil Agostino Steffani, who was Kapellmeister in Hanover 
from 1685, Handel owed some valuable stimuli. Some of Carissimi's 
oratorios Jephthah, Behhazxar, The Judgment of Solomon, and 
Jonah, were edited in 1869 by Handel's biographer, Chrysander. 
The manuscripts of other oratorios of his are in the Bibliotheque 
National at Paris. 



70 VI. The Cantata and the Passion before Bach. 

cultus. When Quistoph Bernhard, Schiitz's favorite pupil, 
went to Hamburg in 1663, to be cantor and musical director 
at the Johanneum, "the leading people of the town", so 
Mattheson tells us, "went as far as Bergedorf to meet him 
with six coaches, a distance of two miles". Johann 
Rudolf Ahle (1625 1673) was both cantor and burgo- 
master at Miihlhausen in Thuringia. 

The town musicians were chiefly intended to assist in 
the church music. The educational institutions of the 
town had also to lend their aid to the art. The Latin 
schools furnished the choirs. Every boy with a good 
voice entered the school, and was maintained by the town 
during the whole period of his education. If he lost his 
voice at puberty, he had meanwhile learned an instru- 
ment and could now play in the orchestra. If he had 
any artistic talent, he could safely continue his studies 
at the university, with the certainty of being able to live 
by his art. Telemann, the Hamburg master, secured an 
important position for himself in Leipzig even when a 
student. 

One result of this close connection between the educa- 
tional system and music was that the level of culture 
among the musicians of that time was higher than it has 
ever been since*. If we examine the careers of the artists 
of that time we see that almost all of them decided to 
take up music during or after their university course. 
German jurisprudence may be proud of having provided 

* We very often find them holding the highest positions as 
teachers. The Landgrave of Cassel had settled upon Schiitz as 
tutor to his children, and unwillingly parted with bim to Saxony. 
In 1674 the Electoral Prince of Saxony asked the Hamburg town 
council to let him have Christoph Bernhard back to be preceptor 
to his beloved grand-child, as well "as Vice-Kapellmeister. Mat- 
theson, in Hamburg, was the tutor of the children of Johann von 
Wichs and those of the English ambassador, whose legation secre- 
tary he afterwards became. The employment of the musician of 
that epoch in all kinds of offices requiring a sound general and 
university education can - be proved by many other interesting 
examples. 



Ideals of Church Music in the Generation after Schiitz. 71 

almost all the best musicians of the seventeenth century 
and the early part of the eighteenth. It can claim Schiitz, 
Walther, Mattheson, Handel, Kuhnau, Emmanuel Bach, 
and many other distinguished names. 

"Whether a composer necessario must have studied" 
is the question put by one Johann Beerens in an essay 
published in 1719; he answered it with a decisive affirma- 
tive. The real quality of their culture is seen when we 
examine their literary productions, from the Syntagma 
Musicutn of Michael Praetorius* to the works of Em- 
manuel Bach, Gerber, Adlung, Marpurg and the rest. 
On the other hand, this close connection between art and 
education resulted in every educated person having some 
knowledge of music, and those who owed their schooling 
to it remained true to it, to whatever position of dignity 
they rose. This general diffusion of artistic culture ex- 
plains the interest, quite incomprehensible from the 
standpoint of the present day, that was taken in the 
music of the church. To the Protestant towns of that 
time the artistic church service was what the theatre was 
to the Greek community - the centre of art and religion. 
For the rulers, again, great and small, the religious 
service was a concern of state. Many of them, indeed, 
when they were looking out for a good Kapellmeister 
or cantor, disregarded the injunction as to * 'unyoking, 
extorting, or alienating" as Luther puts it in his ex- 
planation of the tenth commandment, and did any- 
thing but admonish the person on whom they had cast 
covetous eyes to remain where he was and do his duty. 
A good musician in those days had his value as a political 
commodity, The desire to stand well with the Elector 
of Saxony was certainly not the least consideration that 
induced the Landgrave of Hesse-Cassel to yield to the 

* Michael Praetorius (i57i-~i62o), was Kapellmeister at Wolfen- 
bftttel. His Syntagma Musicum (Part I, History of Music; Part It, 
History of Instruments; Part III, Practical Instruction) appeared 
1615 1619. 



72 VI. The Cantata and the Passion before Bach. 

continual solicitations of Johann George of Saxony, and 
let him have the young Schiitz. 

The misery of the Thirty Years' War broke up many 
of these unions of souls. The princes who had fostered 
art in obedience to an ideal made the greatest sacrifices 
and did all they possibly could to make the service the 
last to feel the retrenchments necessitated by the events of 
the time. The Prince of Liegnitz was one of these. Jo- 
hann George of Saxony, however, who had ostentatiously 
carried his Kapelle" about with him everywhere, practised 
his first economies in connection with his musicians. 

Unfortunately it is for the most part only in account 
papers, yellow with age, and the dreary protocols of town 
councils, that we can read the story of the efforts made 
by the German communities of that epoch for religious 
music; and most of the documents still slumber in the 
archives. * It thus happens that we have more informa- 
tion upon the material side of the case than upon the ar- 
tistic. From the transactions of the Miihlhausen "Society" 
we may learn all that happened at the annual banquet, 
the Conmmum musicale, what was served at table and 
what it cost. From the council records one would think 
that this formal, solid feasting was the main feature of the 
affair. 

Surveying the situation as a whole, we must acknow- 
ledge that the conditions were singularly favourable for 
the coming of great church music. Yet this epoch created 
nothing great and durable. Not that it was lacking in 
creative force or the joy of creation. Never was so much 



* See Otto Taubert, Die Pflege der Musik in Torgau 
1868); Philipp Spitta, Die musikalische Sozietdt und das *' Convivium 
Musical*" zu Miihlhausen im XVII. Jahrhundert (Musikgeschicht- 
tiche Aufsatee, Berlin, 1894, pp. 77 85); Joseph Sittard, Geschicbte 
des Musik- und Konzertwesens in Hamburg vom XlV.Jahrhundert 
bis auf die Gegenwart (Altona-Leipzig, 1890). 

The registers give us a good idea of the relations of the Ham- 
burg town-musicians to the upper pastry-cook under whose orders 
they were at weddings and banquets. 



Church Composers of the i^th Century. 73 

music composed as then, Every cantor made it his pride 
to supply a cantata of his own for each Sunday and feast- 
day. No one was looked upon as a musician unless he 
could compose. Men of the most mediocre talent had a 
complete mastery of the technique of composition. They 
laid the foundation of it by copying out voice-parts, and 
afterwards a thorough practical education made them 
capable of producing serviceable music at quite an early 
age. The system of instruction in that epoch cannot at 
all be compared with ours, which is almost wholly directed 
to making performers; it was at once more practical and 
had a far loftier ideal. From the standpoint of education, 
printed music is a Danaus-gift, since it enables students 
to escape the elementary instruction given by copying 
music. 

Although the conditions were so favourable, that epoch 
created nothing durable. Imagine what a corresponding 
epoch would have produced in painting under similar cir- 
cumstances! In music we have merely names deserving 
of reverence and respect, but not immortal works. And 
even if all the cantatas of that time that are still sleeping 
in sacristies and church vaults, awaiting their discoverer, 
should be brought into the light of day, we should be no 
richer in really great art. The way from Schutz to Bach 
goes over hills, not over mountains. 

The problem of the text is still unsolved. To some 
extent the form of the strophic song holds its ground; 
some composers experiment with the madrigal form; 
others go back to the Bible and the old church hymn. But 
no definitive form is evolved. 

The most important representative of the transitional 
cantata is Andreas Hammerschmidt (1611 1675), organist 
of St. John's Church in Zittau, whose Musikalische Andach- 
ten and Mwsikalische Gespr&che were universally admired *. 

* MusihaliscHe Andachten, in five Parts, 1638 1653. I* Geist- 
ttche Konzerte, Freiberg, 1638. II. Geistliche Madrigalien, Freiberg, 
1641. Ill, GeisUiche Symphonies Freiberg, 1642. IV. Gtisttich* 



74 VI. The Cantata and the Passion before Bach. 

Johann Rudolf Able* (16251673) of Miihlhausen, and 
Wolfgang Karl Briegel (16261712) of Darmstadt, were 
also held in high esteem. In comparison with Schiitz, their 
music may be called almost conservative. 

The art of Johann Christoph Bach, the court and town 
organist at Eisenach (1642 1703) is much bolder. Johann 
Sebastian, his nephew, thought a great deal of him, and 
produced his Michaelmas cantata "Es erhub sich ein 
Streit" (Rev. XII, 712) in Leipzig. This has come down 
to us. The work of the uncle foreshadows the art of the 
nephew. The cantata is written in twenty-two real parts, 
and contains some astonishing harmonic audacities. Jo- 
hann Christoph's mastery of polyphony was so great that 
he never played on the organ or the clavier in fewer than 
five real parts. Philipp Emmanuel Bach inherited from 
his father, along with the copies of his great-uncle's works, 
his father's admiration for him. When Forkel, the first 
biographer of Bach, visited Philipp Emmanuel in Ham- 
burg, the latter played him some of the compositions of 
his ancestor. "I still have a very lively recollection", 
writes the biographer later, "how pleasantly the old man 
smiled at me during the most remarkable and boldest 
passages, when he gave me the pleasure of letting me hear 
some of these old works in Hamburg" **. 

Johann Michael, the brother of Johann Christoph, or- 
ganist and town-clerk of Gehren, was less important. 



Motetten und Konzevte, Freiberg, 1646. V, Geistliche Chormusik 
auj Madrigaltnanier, Freiberg, 1653. Dialogi oder Gespr&che zwisch&n 
Gott und einer gldubigen Se&le, in two Parts, Dresden, 1645 and 
1646. Musikalische Gesprdche ttber die Evangelia, in two Parts, 
Dresden, 1655 and 1656. 

* Geistliche Dialoge mit zwei, drei, vie? und mehr Stimmen, 
Erfurt, 1648. Also collections of Sacred Songs, "Konzerte", 
Motets, "Andachten", &c. His son and successor, Johann George 
AMe, was of less importance. 

** Forkel, Vber Johann Sebastian Backs Leben, Kunst und 
Kunstwerhe, Leipzig, 1802, p. 2. Max Schneider, in the Bachfahr-* 
buck for 1907, pp. 101 177, gives a full thematic index to the com- 
positions of the older generations of the Bach family. 



Church Composers of the i/th Century. 75 

Nevertheless Johann Sebastian copied out several of his 
motets, one of which was for a long time erroneously 
regarded as his own work. 

In the north, where the church music at first did not 
ally itself with contemporary poetry, it mostly followed the 
lines of Schiitz, deriving its texts simply from the Bible 
and the hymn-books, and throwing the whole weight of 
the dramatic expression upon the music alone. As the 
north had suffered least during the Thirty Years* War, 
it was able to do much more for art than was pos- 
sible elsewhere. Nuremberg, in the south, ceased to be 
an artistic centre. For the next two generations Dresden 
and Leipzig allowed Hamburg and Lubeck to take the 
lead of them. Hamburg in particular was regarded as 
the land of promise for musicians. The old Schiitz would 
have liked to settle there. When Bach was looking round 
for a definitive post he set his hopes on Hamburg; and it 
is practically by an accident that he did not go there, 
if indeed we can call by that term what was really the 
successful intrigue of a mediocrity, backed by money, 
against a man of ability. 

In Hamburg we find the two most important pupils 
of Schiitz, Matthias Weckmann (1621 1674) and 
Christoph Bernhard (1627 1692)*. In Lubeck was Franz 
Tunder (1614- 1667), a pupil of Frescobaldi, and the pre- 
decessor of Buxtehude, who, in accordance with the cus- 
tom of the time, married the organist's office in the person 
of the daughter of its previous occupier. His successor 
Schiefferdecker did the same thing**. 

* Max Seiffert, Matthias Weckmann und das "Collegium Mu~ 
sicwn" in Hamburg (Sammelbdnde der Intevnationahn Musih- 
gcscllschaft, 1900 1901, pp. 76 132). This article gives one of 
the best pictures of the artistic life of that epoch. 

** On 4th May 1706 Buxtehude, at that time sixty-nine years 
old, petitioned the authorities that after his death his post might 
be given to one of his daughters, for whom he had a good "sub- 
jectum" in view. His request was granted. See C. Stiehl, Die Or- 
gamsten an dev St. Mariewkirche ttnd die Abendmusiken zu 
Leipzig, 1886. 



76 VI. The Cantata and the Passion before Bach. 

When and how the celebrated Abendmusiken (evening 
performances of music) originated in the Liibeck Marien- 
kirche seems to be no longer ascertainable with certainty. 
Mention is first made of them in the protocol book of 
the Marienkirche for 1673, according to which "any one 
who henceforth shall be appointed and engaged as mu- 
sician to the council" must assist at the organ at the five 
A bendmusiken gratis*. The organist himself had to pay 
the assistant musicians, whom he engaged out of the 
musicians* guild. For this he received his douceur from 
the leading citizens. 

In Buxtehude's time it was customary for the organist 
to present to the "high patrons" the printed text-books 
of the Abendmusiken, as we learn from a copy inscribed 
by him in 1700 to a certain Herr "Dietr. Wulfrath" **, If 
the collection did not cover the expenses, the town council 
helped. It also looked after the maintenance of order. 
At the performances in commemoration of the Emperor 
Leopold I., on 2nd December 1705 Bach was staying 
in Liibeck at the time and was present the pressure 
of the crowd was so great that "two corporals and 
eighteen privates" were necessary. As a rule the "council 
house guard" sufficed***. 



* For further particulars see the above cited work of Stiehl. 
** The custom of having printed programme-books appears to 
have existed in Liibeck since at least 1677. When an enquiry was 
made, in the fifties of the eighteenth century, for a complete col- 
lection of the text-books of the Abendmusiken for the Liibeck public 
library, some one offered the whole collection from 16771757. 
As it is not in the town library now, it is doubtful whether it was 
acquired. See Stiehl, p. 7. On the whole it appears to have been 
a fairly common custom at that time to give the audience the printed 
text of a cantata. Johann Daniel Gumprecht, in the preface to 
his Sabbatsgedanken (1695) speaks of it as a recognized demand 
of the time. 

*** Stiehl, p. 8. In the weekly record of the Marienkirche of 1700 
we read: "Also on this day, by the grace of God, the Abendmusiken 
customary in the church from old time; but further, at the desire 
of the council, a congratulatory poem on the weU-being of the town 
of Lfibeck was printed and was by me publicly presented in a populous 



The Mbeck "Abendrnusiken". y*j 

When the Liibeck cantor Ruetz, a contemporary of 
Bach, enquired in 1753 among the old people how the 
Alendmusiken were begun, he was told that in old times 
the organist at St. Mary's used to play something on the 
organ for the citizens before they went to the bourse, 
and that the Abendmusiken developed gradually out of 
these performances. This conjecture, which has been 
repeated in all the histories of music, flattering as it is 
for the old Liibeck speculators, has no daim to authen- 
ticity. It does not explain why the performances took 
place on Sunday, and in the cold season of the year *. The 
music was given, that is, between four and five o'clock 
on the Sundays between Martinmas and Christmas, at 
the conclusion of the afternoon service, with the exception 
of the first Sunday in Advent ; thus five concerts were given 
each season. It is much more probable that the perform- 
ances originated in an Advent festival, although it 
must be set against this view that in conformity with the 
Gospel selection in the old church, which dealt with the 
last judgment, Advent Was regarded as a time of peni- 
tence and mourning, during which all music, even the 
organ, had to cease. In Leipzig and most other places 
no cantatas were given on these Sundays. 

The five cantatas constituted a single whole. The 
titles have come down to us of three cycles of the kind 
composed by Buxtehude. One was called "The Wedding 
of the Lamb"; the second "Heavenly Delight of the Soul 

assembly in a complete Musica-, and in order to prevent any tumult, 
in and about the church, the council house guard had to be in at- 
tendance, for which they received as usual 6 marks/ 1 

* In Bach's time Ruetz complains that "the Abendmusiken 
were given in such an inclement and raw season, namely in the 
middle of winter, that after one had passed three hours (i. e. at 
the afternoon service) in the cold, one had also to freeze for a fourth. 
The horrid noise of mischievous youths, and the unruly running, 
racing and brawling behind the choir, destroy almost all the charm 
the music might have had; not to mention the iniquities and un- 
godlinesses that are committed under cover of the obscurity and 
the low lights". 



78 VI. The Cantata and the Passion before Bach. 

upon Earth over the Birth of our Saviour Jesus Christ 
and His becoming Man"; the third "The most Terrible 
and most Joyful of all Things, namely the End of Time 
and the Beginning of Eternity, exhibited in Dialogue". 
All three titles indicate Advent. Unhappily both texts 
and music are lost. 

When Stiehl, in 1885, was writing upon Tunder in Eitner's 
Monatshefte fiir MusikgescUchte, and, in the following year, 
upon the organists of the Marienkirche, he lamented the 
fact that we possess hardly any of the choral works of 
the northern masters of that time, those of Liibeck or 
those of Hamburg, except some twenty church cantatas 
of Buxtehude*. But when travelling in Sweden, in 1889, 
he discovered that the famous library at Upsala had pre- 
served a great number of these church compositions. It 
possesses the musical collection of the Diiben family, in 
which, at that time, the office of Hofkapellmeister at 
Stockholm had been hereditary for three generations, up 
to Karl Gustaf Diiben, who occupied it from 1719. The 
Diibens kept up a brisk correspondence with the two 
North German musical centres, often visited their friends 
among the composers, and copied out whatever parti- 
cularly pleased them. From their cantatas we first learn 
what Tunder, Weckmann and Bernhard could do*?; 
Buxtehude also we now know better***. 

One feature that is common to all of them is the fre- 
quently brilliant treatment of the orchestra. They aim 
at colour effects, especially Weckmann and Buxtehude. 
The latter employs, in the cantata Ihr lieben Christen 



* There are twenty church cantatas in manuscript in the Liibeck 
library, and two in the Berlin library. The latter two, Dixit 
Dominus and Nun freut euch ihr Fyommen appeared in the Monats- 
hefte fiir Musikgeschicfrte. 

** See Stiehl, Die Famttie Duben und die Buxtehudeschen Manu- 
skripte auf der Bibliothek zu Upsala (Eitner's Monatshefte filr Musik- 
geschichte, 1889). 

*** There are a hundred manuscripts of Buxtehude at Upsala, 
and eighteen of Tunder. 



The New Cantata. 



79 



freuet euch, three violins, two violas, three cornets, three 
trombones, two trumpets, bassoon, contrabass and organ 
ontinuo. The wind predominate, and are combined with 
the organ in all kinds of ways*. A composition of the 
year 1697, for the consecration of the new altar at the 
Marienkirche requires three choruses, kettle-drums and 
trumpets. Tunder is almost more important than Buxte- 
hude. His cantata on Sin* feste Burg is a powerful and 
spirited work. 

Already, however, the influence of the instrumental 
style on vocal works begins to be visible. We feel that 
we are coming to the epoch when the pure vocal style is 
no longer the concern of any German composer. With Weck- 
mann we often fancy we are reading Bach, so instrumental 
is his writing for the voices**. 

At the same time we must not form an exaggerated 
idea of the means that these instrumentally-minded com- 
posers had at their disposal. At Liibeck, difficulties arose 
from the character of the church. Under the most la- 
vourable circumstances Buxtehude could accommodate 
forty choristers on the six rows of seats running by the 
side of the great organ. Since it was hardly possible for 
him to get each voice doubled, the chorus, according to 
our views, must have been completely smothered by the 
instrumental mass. In Hamburg it was no better. The 
cantor at St. Peter's, for example, had at his disposal in 
1730 seven singers, against whom there were seventeen 



* In this respect the northern composers do not follow the lead 
of Schiitz, who had made some steps towards emancipating him- 
self from the old preference for the wind, and giving more promin- 
ence to the strings. 

** In the Marienkirche there seems to have always been a strong 
preference for instrumental effects. It was the duty of the ^violinists 
and lutenists to "assist" with music (i. e. solo numbers) on the 
organ when, the consuls, the members of the council, or the prin- 
cipals of the church took communion. In 1659 the lutenist was 
enjoined to 'let himself be heard on the organ several times per 
month in addition to the Feast Pays", These orders were still in 
force in 1737. See Stiehl, p. 12, 



80 VI. The Cantata and the Passion before Bach. 

instrumentalists 4 , without counting the three trumpets and 
the kettle-drum*. 

In their method oi employing the chorale, these com- 
posers hark back to the epoch before Schutz. It. plays 
a great part in their works, whereas he had been disposed 
to put it aside. They write whole cantatas upon chorales, 
and so create the type of the chorale cantata, which, later 
on, alternately attracts and repels Bach again and again, 
Strictly speaking, this form of cantata is a mere medley, 
since a strophic song is not in place in the text of a work 
with solo pieces. The whole species invites the criticism 
that a new patch does not suit an old garment. But the 
chorale is put to marvellous uses in the cantatas constructed . 
out of passages from the Bible. Buxtefrude's texts are 
often quite impressive in their harmony of Biblical pas- 
sages and chorale strophes, especially when the chorale 
threads its way through the whole like a central idea. 
The symbolism of the wordless, purely instrumental chorale 
nielody, with which Bach was afterwards to express his 
deepest thoughts, is also already perfected. 

Thus the problem of the text seems to be settled at 
last. Contemporary poetry abandons the attempt to 
create a free form, and the musician constructs the cantata 
for himseli out of verses from the Bible and strophes from 
the hynrn-books, which he contrasts with each other or builds 
up in a dramatic sequence. One could wish that the artists 
of that time had realised that this was the only solution, 
that the treasures of the Bible and of the German hymn- 
books were inexhaustible and will last out as long as music 
and the world themselves; and that music therefore had 
everything it needed within its own hands, and had no 
need of poets for its texts. 

But just at this moment there springs up again the 
ideal of the co-operation of jjoetry and,, jwsjb in 



* Joseph Sittard, Geschichte der Musih und des $m 

tn Hamburg (Altoma-Leipzig, 1890), p, 40, Sittard gives also the 
equipment of the other churches. 



The New Cantata. Si 

^a dramatic representation of religious ideas. It bewitches 
"Protestant church music, luring it onward with the charm 
of a great ideal, helped by the insufficiency of the German 
poetry of the time; and the course of music became an 
uncertain wandering. The fatal thing, .fo that Bach, as 
the child of his time, had to take part in this wandering, 
and kept groping after tlie true form of the cantata his 
whole life long. Inspired by the idea of the religious 
drama, the poorest poetry was blind to its own incom- 
petence. 

The new cantata differs from the old both textually 
and musically, textually, for the most part, in that it 
dispenses with Bible passages and chorale verses, and 
trusts entirely to free poetry. This free text, however, is 
laid out on the plan of the contemporary Italian opera, 
which has nothing in common with Monteverde's music- 
drama, but consists of da capo arias and recitatives that 
come near to speech. In this new art, which is under the 
Neapolitan influence, the melodic and the declamatory 
are no longer blended as they were in Monteverde's arioso, 
but separate into unrnelodic recitative and undeclarnatory 
song*. Nothing is left of that strong musical-dramatic feel- 
ing that we find in the drammaper musica of the Renaissance. 

The dramatic is now wholly transformed into the reflec- 
tive, i. e. into the aria. As this has no longer to con- 
sist of music that conforms with the action, it becomes 
purely formal, while the recitative, the carrier of the 
action, renounces all melodic form. Thus the new re- 
citative and the new aria only resulted from the inability 
of the operatic art of that time to create real dramatic 
music. As sopjias a mind appears that can conceive and 
express music and action in one artistic idea, it revolts 
against ' this divorce of song and declamation. This was 
the case with Gluck and with Wagner, who both, like 
Monteverde, had in view the true drammct per musica. 

* This change becomes evident in the operas and cantatas of 
Alessandro Scarlatti (1659 1725), 

Schweitzer, Bach, (J 



8z VI, The Cantata and the Passion before Bach, 

At the end of the eighteenth century, however, the two 
new forms that had sprung out of the decline of the Italian 
musical stage were received with royal honours by the 
German composers of evangelical church music. Any- 
one who did not welcome them was regarded as a scorner 
of the true musical gospel. They pushed choral music 
into the background; and church music came to consist 
almost exclusively of solo songs. 

This time the Passion music, which hitherto had gone 
on its way unmolested, also became embroiled in the 
revolution; henceforward its venerable antiquity was to 
be no protection against innovation. Till then it had 
really had no history. As far back as the fourth century, 
according to tradition, the story of the Passion according 
to St. Matthew was recited on Palm Sunday, and that 
according to St. Luke on the Wednesday of Passion Week, 
In the eighth and ninth centuries the Passion according 
to St, Mark was allotted to Tuesday in the Holy Week, 
and the Passion according to St. John to Good Friday*. 
Already in the thirteenth century Durandus desires the 
recital to be in dramatic form. Only the words of the 
Evangelist are to be delivered in the Gospel-tone, i. e., as 
psalmody, while the gentle words of Jesus and the cries 
of the unbelieving people are to be rendered characterist- 
ically**. The psalmody method of delivery lasts until 
the end of the fifteenth century. The Passion is recited 
like every other passage from the Gospel. At the be- 
ginning of the sixteenth century the composers of the 
Netherlands undertake, for the first time, to set the story 



* See, on the whole subject, Otto Kade's thorough and interest- 
ing work Die alteste Passions-Komposition bis zumjahv x6$x. Giiters- 
loh, 1893. 

** Non legitur iota passio sub tono EvcLngelii, sed cantits verbovum 
Christi Aidcius moderatu-r. Evangelistae verba in tono Bvangeln pro~ 
ferantur, verba vero impussimorum Judaeonttn damose et cum aspen- 
tate vocantuy. Whether by this is meant a kind of recitation with 
the roles distributed, as Kade thinks, cannot be decided. We only 
know what Durandus wants, but not how it was done. 



History of the Passion. 83 

of the Passion to music, The first musical Passion is by 
Jacobus Obrecht (born 1450), and dates from 1505; Luther's 
friend, Johann Walther, copied it out twice. When it 
was published by Georg Rhaw, in 1538, Melanchthon con- 
tributed a preface to it. In the epoch that followed, Pas- 
sion music was written by both Catholic and Protestant 
composers, as in the case of the various portions of the 
Mass; nor is any distinction made between German and 
Latin Passions. 

The numerous Passions of that time fall, according 
to their form, into two groups, motet Passions and 
dramatic Passions. In the former, the whole text, in- 
cluding the words of Jesus, is rendered by the choir; in 
the latter, the words of the Evangelist and the speeches 
of Jesus are recited by one person in the old Gospel-tone, 
and only the cries of the people are set polyphonically, 
though the chorus renders also the words of Pilate, of 
the false witnesses, and of the malefactors. 

The dramatic Passion naturally triumphed over the un- 
dramatic motet Passion*. The St. Matthew Passion of 
Johann Walther is the first German composition in this 
style; according to tradition, it was performed on Palm 
Sunday in 1530. His St. John Passion, which was given 
on Good Friday, has survived to our day. It was sung 
yearly inZittau from 1609 to 1816, in a Czech translation. 

Schiitz retains the dramatic Passion as he had received 
it. He dispenses with instrumental means of every kind, 
making the Evangelist psalmodise in the old collect tone, 
and makes no use, in the Passions, of the declamatory 
arioso that he elsewhere employs for the solo renderings 
of Bible passages. No aria, no chorale interrupts the action. 

The severe beauty of this old Passion form, trans- 
figixred by Schiitz's art, is unique of its kind. It reminds 

* Among the most prominent of the composers of Latin Pas- 
sions are Claudia von Sermisy (1534), Orlandus Lassus with 
four Passions dating from 1575 ft, and William Byrd (1607). 
Their works belong to the dramatic species. 

6* 



&J VI. The Cantata and the Passion before Bach. 

us of the affecting representations of the Passion by the 
realistic par iters of the Netherlands . Its profound effect 
when the singer taking the part of the Evangelist knows 
how to enter into the old style* can almost make us 
believe that we are listening to one of Bach's Passions. 

The lack of the pathetic and the contemplative, however, 
necessarily discredited the old dramatic Passion and the 
older church music in general, in an epoch that took rhe- 
torical reflection for the essence of drama. 

The new movement begins with the founding of the 
Hamburg opera. When, in 1678**, Gerhard Schott, the 
Licentiate Liitjens and the organist Reinken cooperated 
in the building of a theatre, they had in their minds not 
so much a secular as a religious opera. Theile's Adam and 
Eva was the first opera (Singspiel) to be given on the 
stage***; then came Michael und David (1679), DicMakka- 
bdisdic Mutter und ihre sieben Sohne (1679), Esther (1680), 
Chris! i Gclurt (1681), and Kain und Abel, oder der ver- 
zweifdte Brudermorder (1682). 

It is, as a matter of fact, hard to find anything religious 
in the trivial and absurd texts of these operas; that age, 
however, thought otherwise. The clergy took a prom- 
inent part in supporting the undertaking.. One clergy- 
man, Heinrich Elmenhorst, wrote the texts of several 
Singspiele, and from the pulpit recommended the faithful 
to go to the opera. 

* Only those who have heard Friedriclx Spitta as the Evan- 
gelist in a Schutz Passion can appreciate fully the beauty of the 
Collect tone in its dramatic transfiguration, 

** The Dresden opera was founded in 1662 by Carlo Pallavicini. 
*** The opening took place on 2nd January 1678. The Ml title 
of the first opera runs thus: Atom und Eva, Der erschaf/ene ge~ 
fallem und aufgmcktote Mens&h. In einem Singspiel dawestetlt . 
Theile was a pupil of Schtitz and the teacher of Buxtehude The 
opera is lost. On the beginning of the opera in Germany see Her- 
mann Kretzschmar, Das erste Jahrhundwt dw dmtschen Qper 
(Sammelbande dev Intevnationahn Mitsikgesellschaft, loor loo- 
pp. 270 293). ' ' ' 



The German Opera and Religious Music 85 

When, in course of time, the theatre had forfeited the 
good opinion of many serious-minded citizens* by reason 
of the coarseness and tastelessness that had flaunted them- 
selves on the stage, Elmenhorst, in \\isDramatologia antictua- 
hodierna, tried to rehabilitate the institution. He argues 
that this theatre is simply the ancient Greek drama ap- 
plied to Christian ends. As the Greeks represented on 
the stage, for religious edification, the" stories of their 
gods and heroes, so it is a necessity to Christendom to 
see^ the Biblical stories unfolded before the eye in animated 
action. The theological faculties of Rostock and Witten- 
berg, being consulted on the matter, declared themselves 
in principle for the religious opera. 

All this, however, could not arrest its decline**; and in 
the" course of time the religious drama disappeared from 
the stage. But the ideal that is upheld in Elmenhorst's 
work still lived and dominated the artistic life of Ham- 
burg in the epoch the most brilliant for Hamburg 
that followed, when Keiser, Mattheson, Handel and Tele- 
mann lived and worked there, Reinhard Keiser came to 
Hamburg in 1694. Handel lived there from 1703 to 1705. 
Mattheson, a Hamburger by birth, belonged to the ope.ra 
from childhood; his true work, however, did not begin 
until 1705, when he abandoned the career of singer and 
actor. Telemann was invited to Hamburg in 



* In his Thcatromachia* oder die Werke dor Finst&rnis (1682) 
Anton Reiser, the pastor of St. Jacob's, numbers the opera among 
the works of the devil, In the same year Magister Ranch replied 
to him in his Theatrophonia ZUY Verteidigung*der christlichen* vor- 
nehmlich ab&r der musikalischen Opera. 

** From about 1730 the opera fell into deeper and deeper decay. 
In 1740 an Italian troupe look possession of it; in 1750 the opera- 
house, with the whole of its fittings, was sold by public auction to 
the highest bidder. 

*** Reinhard Keiser (or Kaiser) was bora in 1673 ; he received 
his general education at the St. Thomas school in Leipzig, and 
later at the university of that town. From 1697 onwards he de- 
voted his energies to the Hamburg opera house, writing for it also 
religious operas, among which may be mentioned Die ubw die Liebe 
triumphievende WeisJteit, oder Salom, in ewew Singsptcl auf dem 



86 VI. The Cantata and the Passion before Bach, 

What position these artists occupied, whether they 
held office in the theatre or in the church, mattered no- 
thing, since they all wrote impartially for both church and 
opera. From 1715 to 1728 Mattheson was cantor of the 
cathedral, in which post he was able to do a good deal 
for the new church music. When his increasing deafness 
forced him to resign, Reiser, who until then had officiated 



grossen hamburgischen Schauplatze dargestellt (1703), and D&r ge~ 
stuvzle und wieder erhohte Nebukadnezar, Konig von Babylon unter 
dem grossen Prophcten Daniel (1704). Keiser really had talent 
and a rich and lively imagination. His frivolous way of living 
prevented him from exercising the influence he might have done, 
He died in Hamburg in 1739. 

Johann Mattheson was born in Hamburg in 1681, He is of im- 
portance not so much for his purely musical work as for his books. 
He is the first literary champion of "modern music". His eighty- 
four works arc most valuable documents for the musical history 
of the time. Without the Syntagma musicum (1614- 1620) of 
Michael Praetorms, Walther's M-usiktexikon (1732) and Mattheson's 
works we should know very little about the music of the seven- 
teenth and early eighteenth centuries. 

When Handel went to Hamburg, Mattheson constituted him- 
self his instructor and protector, and in this capacity seems to have 
made himself a bit of a burden. 

Mattheson particularly champions the new form of church 
mrusic in his Der musikalische Patriot (1728). His Grosse General- 
bassschute (1731) and Der vollkotnmene Kapellmeister "(1739) were 
also of great importance. The Grundlage einer Ehrenp forte (1740) 
contains a very valuable collection of autobiographical articles 
on all the notable musicians of the time. 

His main calling was that of secretary to the English embassy 
in Hamburg. He died in 1764. 

Georg Philipp Telemann, although less talented than Keiser and 
Mattheson, was more highly regarded than they by his German 
contemporaries. Born at Magdeburg in 1681, he went to Leipzig 
in 1701 to study law, and remained there three years. He became 
organist at the New Church. His foundation of the Collegium 
Musicum was of great importance for the musical life of Leipzig. 
This society was chiefly recruited from student circles. After having 
been successively Kapellmeister at Sorau, Eisenach, and at the 
Church of St. Francis and St. Catharine at Frankfort-on-the-Main, 
he went to Hamburg in 1721 as cantor at the Johanneum. He 
also wrote a good deal for the opera. His models were the French. 
Telemann's works he enumerates them in Mattheson's Ehren- 
pforte are legion. He died in 1767. 



The German Opera and Religious Music. 87 

only at the opera, became his successor. Telemann had 
been called to Hamburg to be choir-master and cantor 
at the Johanneum; but he was also engaged by the theatre 
to provide operas at a yearly salary of three hundred 
thalers. Handel, who had gone to Hamburg mainly for 
the opera, also wrote music for the church*. 

These composers gave up the effort to consider the Ham- 
burg theatre, in accordance with the intentions of its 
founders, as primarily a home of religious art, and to main- 
tain it on that level; but in exchange they tried to import 
the religious opera into the church service, not troubling 
themselves as to whether it had any appropriateness to 
the ecclesiastical season or to the Gospel for the day. To 
what Gospel are the" Theatrical Soliloquies" of Telemann 
appropriate, such as Der verkaufte Joseph, Der von 
Zedekia geschlagene Micha, Der von seinem Volke verjolgte 
Dcwid, Der sterbende Simson, or Der versenkte Jonas? They 
even wrote oratorios in two parts, intended for performance 
before and after the sermon, choosing for them any drama- 
tic scene that took their fancy. Not that they were any 
less pious or less in sympathy with the church than their 
predecessors had been. On the contrary, they believed 
that this free, self-existent church music was the only true 
music, and that credit was due to them for having set 
it free from its Babylonian bondage. They aimed at 
creating a Protestant, i. e,, a subjective and dramatic 
art. All they really achieved, however, was the introduc- 
tion into church music of the emotional theatrical style 
with its orchestral painting of situations. It finally be- 
came a question whether the operatic secco recitative and 



* The first of his operas to be performed was Almira (1704); 
Nero followed in 1705. Flovindo and Daphne were already finished 
when he left Hamburg, but were not given until 1708. Of his church 
compositions of this time may be mentioned a cantata on the chorale 
"Ach Herr, mich armen Sunder", an oratorio in two parts, for 
St. John's Day, Die Erldsung des Volkes Gottes aus Aegypten", 
and a Passion written in 1704, to a text by the Hamburg opera- 
librettist Postel, 



88 VI. The Cantata an<3. the Passion before Bach. 

the formal three-section da capo aria should not also be 
admitted into the music of the church. 

Naturally these composers could do nothing with the 
choir-boys whom they had at their disposal in the churches 1 . 
To the choir, indeed, they attached little importance, as 
their music mainly consisted of solo numbers. For their 
arias, however, they needed accomplished singers, male 
and female. Until then, women had not been allowed to 
sing- either in the Protestant or the Catholic church, ex- 
cept with the rest of the congregation in the chorale. The 
treble and alto parts in the choruses and in the soli were 
taken by boys. Mattheson pours his scorn on the church 
choristers of that epoch who had to do duty as soloists. 
He describes the treble "with a feeble falsetto, singing 
like a toothless old woman;" the alto, "with a voice like 
a calf"; the tenor, "who brays like a hoarse jackass"; the 
bass, "who rumbles out the eight-feet G in the depths, 
like a cock-chafer in an empty boot, in a way that would 
waken a sleeping hare hardly thirty paces off, while he 
howls the four-feet G like an Indian lion". Is it possible 
so runs his perpetual refrain upon this question to 
make music at all with the singers provided by the church ? 
Before condemning the new church music, people should 
provide the composers with the performers necessary for 
its adequate rendering. 

Nor was he satisfied 'with mere platonic discussion. 
Soon after he became cantor of the Cathedral, he succeeded 
in introducing female singers; and in 1716 some ladies 
named Rischmiiller, Schwarz and Schober made the 
church resound with their runs and trills to the glory 
of God. 

Mattheson plumed himself upon this achievement all 
his life. "I was the very first", he wrote in his General- 
basssctiule, "to employ three or four female singers in the 
ordinary church music, before and after the sermon; though 
with what difficulty and vexatiousness, and against what 
opposition, cannot be described. -At first I was implore^ 



Tlie German Opera and Religious Music. 89 

not to bring any women into the choir; in the end they 
could not have enough of them/' 

After the death of Telemann, in 1767, the female sin- 
gers seem to have disappeared again from the churches. 
The example of the Hamburgers had not found many 
imitators. We know that at Liibeck, which modelled its 
church music upon that of Hamburg immediately after 
Buxtehttde's death, no female singers took place in the 
Abendmusiken until 1733 at the earliest. In Leipzig they 
never did so*. 

The new style of church music did not triumph without 
a struggle. In 1726 there appeared an attack on it by 
a certain Joachim Meyer of Qottingen, who was both a 
musician and Doctor juris**. Mattheson made a very clever 
reply. He brings forward in defence of theatrical church 
music the same argument that Elmenhorst had used to 
vindicate the religious theatre. For him, too, the church 
art for which he is fighting is the antique tragedy trans- 
planted into Christendom. Every religious festival, he 
says, is theatrical, since the Biblical stories aiid the religious 
ideas they contain are represented in some way or other, 
The most perfect representation, however, is art***. 



* Buxtehude's son-ia-law and successor, Schiefferdecker, had 
previously been at the Hamburg opera. During his twenty-five 
years of office (1707 -1732) h<? did not give a single work of his 
father-in-law at the Abendmusiken, but wrote each year an oratorio 
in five parts in the Hamburg style. The texts have survived. His 
successor Johann Paul Kuntzen also came from Hamburg. He 
too employed female singers in the Abendtnusiken. 

** Unvorgreiftiche Gedanken uber die neulich eingerissene thea- 
trattsche Kirchenmusih und von den dann Usher ublichen Kantaten, 
mit Vergleichung der Mvsih voriger Zeiten xwr Verbesserung der 
unsrigen vorge$telkt> 1726. 

*** Der neue Gdttwgische, aber vul schlechter als die alUn Lace- 
ddmonuchen urteilende Ephorus, wegen der Kirchenmusik eines an- 
dem belehrt (1727). Joachim Meyer replied in his J>ev antnassliche 
Hamburger 'criticus sine cris? (1728). See also Mattheson's Der 
musihaUsch* Patriot (1728). The Berlin cantor Freudenberg, under 
the pseudonym of Innocentius Frankenberg, took the side of the 
innovators in his GwcchU Wagschal (1729), 



go VI. The Cantata and "the Passion before Bach. 

The most important and stubborn opponent of the 
new style was Johann Kuhnau (1660 1722), Bach's pre- 
decessor at St. Thomas's, Leipzig, whose talents and 
capabilities were everywhere recognised, and whose famous 
dramatic clavier sonatas on Biblical histories made it 
impossible to call him a reactionary*. He threw him- 
self into the fight just when the n#w tendency began to 
attract attention. When he printed his cantata texts for 
the ecclesiastical year 1709 1710, he accompanied them 
with a preface, in which he laid down his own ideas of the 
true style for church music, in contrast with the theatrical 
style**. 

His views did not suit the taste of all the Leipzigers, 
particularly of the circle that had formed round Tele- 
mann. In his last years he had to yield to the pressure 
of the time, and write a Passion in the new style, which 
was performed in 1721, After his death it was proposed 
to make Telemann, who was the antipodes of Kuhnau, 
his successor. 

From about 1700, all opposition to the new church 
music was in vain. The educated people, clerical as well 
as lay, had before their eyes the antique ideal of the religious 
drama, which Elmenhorst and Mattheson had advocated 
so cleverly and with such real enthusiasm. Neither they 
nor the musicians themselves were conscious how far poetry 
and music were from, attaining this ideal, 

We, for our part, cannot place ourselves at the point 
of view of an epoch which revelled in the confident joy 
of creation 5 and went on its way without being crushed 

* Kuhnau had written an opera, which, however, had been a 
failure. See Richard Miinnich's careful study Kuknaus Leben, in 
the Sammelbdnde der Intevnationalen Musikgesellschaft* 1901 1902, 
pp. 4/3527. 

** See Richter, Eine Abhandlung Johann Kuhnaus, in Eitner's 
Monatshefte jur MusikgescMchte> 1902. At the beginning of each 
ecclesiastical year, Kuhnau issued in printed form the texts of the 
cantata cycle that was to be produced. Upon the fragments of 
these publications that have come down to us from the years 
1707 1721, see Richter, loc* cit. 



Neumeistex and Franck. gi 

or made self-reflective by any great tradition either of 
classical dramatic poetry or classical music, but in un- 
moved simplicity prized its own imperfect work as being 
the realisation of the ideal it was aiming at, and thought 
that all it needed to produce classical art was to pursue 
an antique ideal. 

From our standpoint we must always judge that age 
unjustly, for we lack the self-confident creative force that 
was at once its strength and its weakness. We feel our- 
selves to be epigones. The tribute paid by these men to 
antiquity, however, was merely a rhetorical one; they 
were convinced that with them the last great artistic epoch 
had arrived. Feeling themselves thus free of the juris- 
diction of antiquity, they fell victims to every deception 
and illusion, and finally obeyed the voice that counselled 
them, if they would make church music wholly new, to 
cease taking their texts from the Bible and the old 
congregational hymn. 

The fate of the cantata was decided by the appearance 
of the first yearly cycle of Erdmann Neumeister's Kirchen- 
andachten*. This first cycle was composed for the court 
church in Weissenfels, and was set to music by Philipp 
Krieger (1649 1725), the court-Kapellmeister there. Neu- 
meister's "madrigal cantatas" have nothing in common 
with the earlier attempts at the madrigal in German 
religious poetry, but are imitations of the Italian opera 
texts, the author himself declaring ? in his preface, that 
for him a cantata is only a fragment of an opera. He 
makes each of them consist of four arias and four recita- 
tives. He discards altogether Biblical passages and verses 
from the hymn-books, nor are there any choruses. 

In the succeeding cycles he makes some concessions. 
The second (1708) again gives the chorus its due; in the 
third (1711) a modest place is again granted to Biblical 



* Erdmann Neumeister, born in 1671, was court deacon at 
Weissenfels and afterwards at Sorau. In 17 1 5 he became the minister 
of St. Jacob's, Hamburg. 



92 VI. The Cantata and the Passion before Bach. 

passages and hymn verses by the side of arias and recita- 
tives. The "modern" cantata was henceforth cultivated 
in this infelicitous compromise. In 1716 Tilgner issued 
in one volume all the texts of Neumeister that had appeared 
during the five~ previous years- The composers at once 
seized upon them. Telemann, who at that time was not 
yet in Hamburg but in Eisenach, had set the third and 
fourth cycles to music on their first appearance in 1711 and 
1714 respectively. Bach also composed music to Neu- 
meister's texts. He knew the author personally, and admired 
him as a genuine poet. Neumeister, on his side, valued Bach 
very highly and would gladly have had him in Hamburg. 

Neumeister had a rival in Salerno Franck (born in 1659) 
of Weimar, who was secretary to the Consistory in Arn- 
stadt from 1691 to 1697, where he published his Madri- 
galische Seelen-Lust ilber das heilige Leiden unseres Er- 
losers (1697). Bach's Passions show traces of the influence 
of this poem. At a later date Franck became personally 
acquainted with Bach, , when the latter was stationed 
at Weimar and supplied him with some cantata texts. 
The poems of Franck that appeared after 1711 are in- 
fluenced by Neumeister. He harks back, however, much 
more than the latter to the old madrigalesque religious 
poetry, and, if not so clever, is more profound than the 
man he had taken for his model. 

In the future it became a recognised part of the busi- 
ness of a poet to write cantata texts. To do so it was*not 
necessary for him to have any personal connection with 
sacred poetry; he could insert a yearly cycle of texts for 
church pieces in the middle of some gallant or satirical 
publication. Thus for a long time the sources of a number 
of cantatas by Bach, that are connected textually, mu- 
sically, and chronologically, were unknown mntil Philipp 
Spitta found them in the collected poems of Marianne 
von Ziegler, a contemporary Leipzig poetess*. 

* Spitta, Marianne von Ziegler und Jdh. $el>. Bqch, in 2ur 
Mitsik. Berlin, 1892. 



The New Form of the Passion, 



93 



The new cantata brought with it the new Passion. The 
first theatrical Passion was produced in Hamburg in 
1704, during the Monday and Wednesday vespers of Holy 
Week. The text was by Christian Friedrich Hunold, a 
writer of opera libretti living in Hamburg from 1700 to 
1706, who did not enjoy the best of reputations. In the 
literary world he was known by the name of Menantes*. 
Reiser supplied the music. 

The whole Passion was now represented as a dramatic 
action. The place of the Biblical story of the Passion 
was taken by a versified text that connected the separate 
scenes. It is noteworthy that in this Passion we already 
have the "Daughter of Zion", whom we shall meet with 
again in Bach, 

The indignation was universal, not so much on account 
of the new departure itself as on account of the extremely 
wretched theatrical style. In the same year Handel also 
wrote a Passion to a text by' Postel, another Hamburg 
P era P oet which was performed without winning any 
success. 

In 1712 appeared the Passion poem of the Hamburg 
town councillor Barthold Heinrich Brookes. In the main 
he retains the plan and the constituents of Reiser's Passion. 
He makes use of free recitative and da capo arias, admits 
the Daughter of Zion, and replaces the Gospel narrative 
by a versified recital of the Passion, keeping more closely, 
however, to the Biblical wording than Hunold-Menantes. 
The only really new feature was the insertion of chorale 
strophes**; for the rest he did nothing more than discard 

* Spitta, Bach und Christian Friedrich Hunold, in Musika- 
lische Aufsatze, Berlin, 1904, pp. 89 ft In Hamburg, Hunold was 
known as an obscene litterateur; in 1706 he had to leave the town 
on account of a licentious novel. He went in 1708 to Halle, where 
he gave lectures on poetry and jurisprudence. He knew Bach, 
and supplied him with texts for secular cantatas, . 

** The Passions of Schiitz contain no chorale strophes. So 
far as we know, the Brandenburg-Prussian Kapellmeister Johann 
Sebastiani was tbe first to use chorale verses in his Passions. Das 



94 VI. The Cantata and the Passion before Bach, 

some of the theatrical elements of Hunold's Passion, and 
to purge the diction of its worst impurities. 

And this text became the classical one for the Passion I 
Keiser set it to music in 1712, Handel and Telemann in 
1716, and Mattlieson in 1718. The last-named's Passion 
was performed during the service on Palm Sunday in 1718. 
Telemann produced his in Frankfort. "It was excellently 
performed", he himself tells us, "in the chief church on 
several special week-days, in the presence of various great 
people -and an indescribable throng of hearers, for the 
good of the orphan institution, I may record the singular 
fact that guards were posted at the church doors, who 
allowed no one to enter who had not a printed copy of 
the Passion, and that most of the members of the ministry 
were at the altar in their robes of office. Moreover, this 
Passion has been heard in many German churches and 
concert-halls."* 

When Handel set Brockes's text to music, he had devel- 
oped far beyond the other Hamburg composers. Later 
on he made use of this music in other works. We possess 
the work, however, only in a copy made by Bach and 
his wife. The manuscript comprises sixty pages; the first 
twenty-three are by Bach, the remainder by Anna Magda- 
lena, his second wife. It is thus probable that Bach gave 
a performance of Handel's Passion in Leipzig* 

How are we to explain the enthusiasm of composers 
for Brockes's Passion poem? It was certainly not the dis- 
tinction of its language that attracted them, for the verse 
is positively vulgar. The scourging is described in the 
following recitative: 

Leyden und Sterben unsevs Herm und Heylandes Jesu Christi, in 
eim veciHerende Harmonie von funf singenden und seeks spielenden 
SUmmen, -nebst dent Basso continue* ge$etzet> wonnnen mv Erweckung 
mehrerw Devotion unterschiedlicbe Verse aus denen gewdhnttchen 
KiYchen-Liedevn mil bingejuhret und dem Texte accomodiret worden; 
Komgsberg, 1672. 

* Wilhehn Langhans, Geschichte der Mmik im AT//., XV 'III. 
und XlX.Jahrhundert, 1882, p, 430 fl 



The New Form of the Passion. 95 

"Drauf zerrten die Knegknecht* ihn herein 
Und riefen ihre Wut mehr anzuflammen 
Die ganze Schar zusammen. 
Die banden ihn an einen Stein 
Und geiBelten den zarten Riicken 
Mit nagelvollen Stricken." 

After Peter's denial there follows this aria: 

"Heul, du Scham der Menschenkinder, 
Winsle, wilder Siindenknecht ! 
Tranenwasser ist zu schlecht; 
Weine Blut, verstockter Sunder!" 

At the same time the text had some dramatic vitality, 
and it was extremely rich in opportunities for musical 
painting. The inflated diction did not at all repel the 
composers, who took it to be the characteristic of a poetic 
sensibility that was akin to music. Compared with Hunold- 
Menantes's Passion poem and the libretti of the Hamburg 
operas, the style must indeed have seemed positively pure. 

Literary Germany before the time of Gottsched and 
Lessing possessed nothing of our modern sense of verbal 
values. The connection with music was really pernicious 
to the poetry of the time, for it was being perpetually 
worked up into an Over-exuberance of feeling, in which 
it aimed at drastic images and pathetic expression with- 
out any other thought than that of making its ideas as 
strong and flowery as possible. Not until it parted com- 
pany from music, and was placed on a footing of its own 
by Gottsched and Lessing, did it find itself again. 

Bach, however, lived in the decadent epoch when music 
and poetry led each other astray, an epoch of excessive scrib- 
bling, of superficial art, in which even men of real talent like 
Keiser went to ruin, an epoch which seemed fated to be 
impotent to create anything of durable value*. Whereas 
at other times and in other places the great artist has 
been only one star among others, whose light, if less brilliant 

* The verdict of Eitner, the best judge of that epoch, upon it is 
eqxially severe. See his Kantaten aits dem Ende des XVI I, undAn- 
fangdes XVIII. J ahrhundevts (Monqkshefte fur Musikgeschichte, 1884). 



g6 VI. The Cantata and the Passion before Bach. 

than his, lie nevertheless did not extinguish, Bach is sur- 
rounded by mere will-of-the-wisps, which his epoch and 
he with it mistook for stars. Of the innumerable can- 
tatas that were written and admired at that time, his alone 
have survived their own day, and even these exhibit, 
both in their form and in their texts, traces of the dead 
world from which they have come. There is no stronger 
testimony to the greatness of Bach than the fact that in 
an epoch of error, and sharing .its errors, he nevertheless 
wrote imperishable works. We have finally, however, the 
sad consciousness that he was only great enough to save 
himself, but not his epoch as well that he did not hurl 
himself against it and strive to lead it back from this 
stilted poetry and the empty forms of Italian recitative 
and the da capo aria, to the true, simple, and really dram- 
atic church music. 

To this clear perception, which was to a certain degree 
attainable by a Kuhnau, Bach never arrived. He was in 
fact not the beginning of a new epoch, but the end of an 
old one, in which the knowledge and the errors of suc- 
cessive centuries found expression for the last time, as if 
seeking salvation together by genius. Since Bach held his 
peace, and, though inwardly opposed to his epoch, never- 
theless went its way with it, it was inevitable that his works 
should be thrown into the general grave with those of 
his contemporaries, there to await their resurrection. 

If the talents succumb to the errors of their time, 
what matters? But when the men of genius are ensnared 
in them, centuries have to suffer for it. The very great- 
ness of Aristotle, held Greek natural philosophy back 
when it was already on the path that would have led it 
to the discoveries of Galileo and Copernicus. Bach, with 
an easy consciousness of his own strength, burdened him- 
self with the Italian forms and formulas, and so retarded 
the progress of German religious music along the path that 
would have led it, even at that time, to an art such as 
Wagner was afterwards to realise in drama. 



Bach's Ancestors. 97 

CHAPTER VII. 
FROM EISENACH TO LEIPZIG. 

The Bach family can be traced back in Thuringia as 
far as the beginning of the Reformation. In the family 
chronicle begun by Johann Sebastian and continued by his 
son Philipp Emmanuel, the baker Veit Bach is named as 
the progenitor of the line to which the composer belonged*. 
Forkel, in his biography of Bach (1802), expressed the 
opinion that this Veit Bach came from Hungary. The 
truth is, however, that he had emigrated there from Thu- 
ringia, returning when the Germans in Hungary began to 
suffer during the counter-Reformation. He settled in 
Wechmar, near Goth a. When he went into the mill to 
grind his corn, he would take his guitar with him and 
play, regardless of the racket around him. 

One of his grand-children, Heinrich, settled at Arn- 
stadt**. His sons Johann Christoph (died 1703) and Jo- 
hann Michael (died 1694) were especially prominent mem- 
bers of the race. Johann Christoph was organist in Eise- 
nach***; Johann Michael was organist and town clerk at 
Gehren. 

The members of this huge family of musicians had, as 
Forkel says, "a very great attachment to each other". 
"Since it was impossible to live all together in one place, 
they made a point of seeing each other at least once a year, 

* This family chronicle is in fifty-three sections ; each gives 
a short biographical notice of a male member of the family. The 
important document passed from Philipp Emanuel to Forkel, the 
first biographer of Bach, from him to the Hamburg music-teacher 
PSlchau, and from him to the Berlin Royal Library. The genealog- 
ical tree which Forkel received from Bach's son along with the 
family chronicle has been lost. See Spitta's Bach, preface. 

** The history of Bach's ancestors has been told in masterly 
fashion by Spitta in the first volume of his biography. For con- 
fused tradition he has substituted facts based on documents. 

*** See p. 74 as to his Michaelmas cantata Es erhub sich em 
Streit. 

Schweitzer, Bach. * 



gS VII. From Eisenach to Leipzig, 

and appointed a certain day at which they were all to be 
present at a chosen place. Even when the family had 
greatly increased, and had spread itself abroad beyond 
Trruringia, in various parts of Upper and Lower Saxony 
and in France, they continued their yearly foregatherings. 
The rendezvous was generally Erfurt, Eisenach, or Arn- 
stadt. The manner in which they passed the time during 
the meeting was wholly musical. As the company con- 
sisted of cantors, organists and town musicians, all con- 
nected in some way with the church, and as it was the 
general custom at that time to commence all things with 
religion, the first thing they did when they met together 
was to sing a chorale. From this devout beginning they 
passed to jests, which often contrasted very strongly with 
the chorale. That is to say, they improvised folk-songs 
together, (some of which were comic, some even indecent), 
in such a way that the various impromptu parts made a 
kind of harmony, though the words were different in each 
voice. They called this kind of extempore counterpoint 
quodlibct, and could not only laugh heartily at it them- 
selves, but it aroused also an equally hearty and uncon- 
trollable laughter in every one who heard them" * 

Bach's grandfather, Christoph Bach (died 1661), was a 
son of Hans Caspar Bach and a grandchild of Veit Bach; 
Bach's father, Johann Ambrosius, lived first in Erfurt 
and afterwards, from about 1671, in Eisenach, He had 
a twin brother, Johann Christoph, court and town mu- 
sician at Arnstadt, who was so like him that even their 
respective wives could only distinguish them by their 
clothes. They had the most tender affection for each 
other; speech* sentiments, the style of their music, their 
methods of performance, all were alike. When one 
was ill, so was the other. They died within a short time 
of each other. They were the admiration of all who saw 
them**. 

* Forkel, pp. 3 and 4, 
** Forkel, p, 4. 



Early Years, QQ 

Bach's mother, Elisabeth, was a Lammerhirt by birtlu 
Her father was a furrier at Erfurt. 

Johann Sebastian was born on the 2ist March 1685, 
at Eisenach. His mother died nine years later, leaving 
her husband with four children, of whom Johann Sebastian 
was the youngest. Not long afterwards, at the beginning 
of 1695, the father also died; he had married for the second 
time a little while before. Bach was thus left an orphan 
at the age of ten. The eldest of the brothers, Johann 
Christoph (born in 1671), took the two youngest, Johann 
Jakob and Johann Sebastian, with him to Ohrdruf, where 
he was organist. They attended the gymnasium there, 
and their brother instructed them in music. Johann Se- 
bastian was too zealous for his teacher. He asked the 
latter for a volume containing clavier pieces by Froberger, 
Kerl, Pachelbel and others. Being refused it, he dragged 
it with his tiny hands through the latticed door of the 
cupboard in which it was kept, and copied it out on moon- 
lit nights. In six months the copy was complete. The 
brother heard of it, however, and took the copy from 
him*. 

In 1700, as Johann Christoph's family was always in- 
creasing, Bach had to thihk of finding a shelter elsewhere. 
His good soprano voice secured him a place in the school 
of the convent of St. Michael in Liineburg, with his friend 
Erdmann. He soon lost his voice, but was retained because 
he was useful as a violinist in the orchestra. We do not 
know whether he received lessons from the organ virtuoso 
Bohm. He at any rate heard him play, though Bohm was 
not at St. Michael's church but at St. John's. It is a fact 
of great consequence that in the choir to which he 
belonged he had opportunities of becoming acquainted with 
the best specimens of German phurch music. The catalogue 
of the well-stocked musical library of the gymnasium has 

* These anecdotes appear in the Nekrohg (obituary notice) of 
Bach written by his pupil Agricola and Philipp Emmanuel, that 
appeared in Mizler's Musikalische Bibliothek in 1754. 



IOO VII. From Eisenach to Leipzig. 

been preserved; it includes works by Italian as well as 
German composers*. 

From Liineburg Bacli went more tlian once to Hamburg 
to hear the famous Reinken, and, no doubt, to see the 
opera**. He also went to Celle. The court band there 
consisted largely of French, players. Duke Georg Wilhelm 
of Brunswick, having married a Huguenot lady, Desmier 
d'Olbreuse, surrounded himself with a French court ; even 
the Court organist, Charles Gaudou, was a Frenchman***. 
We do not know who procured for Bach the entree to the 
court concerts. It is possible that he was employed as 
assistant violinist, for after leaving the gymnasium in 1703, 
at the age of eighteen, he procured an engagement in the 
band of Duke Johann Ernst at Weimarf. He stayed 
there only a few months, however, going to Arnstadt in 
1704 as organist at the New Church. The instrument had 
just been builtff. At the Franciscan church the organist 



* W. Jtmghans, Johann Sebastian Bach als SMler der Parti- 
kulavschule zu St. Michaelis in Ltinebuig ( Gymnasialpvogrammr 
Liineburg, 1870). Here are given also the wages of the choristers, 
derived from the old accounts. Their earnings were considerable 
for that time. This explains the strong attraction that the institu- 
tion had for the youth of the surrounding country. Whoever joined 
it was at any rate sure of a living, and could even lay a little by. 
On the career of Georg Bohra, and his relations with the Bach 
family, see the thorough and interesting article of Richard Buch- 
mayer in the Bachjahrbuch for 1908, pp. 105 122. 

** Once on the way back from Jiamburg to Liineburg, Bach, 
hungry and penniless, was standing in front of an inn. The win- 
dows opened, and some herring-heads were thrown into the street. 
He picked them up and found in each of them a Danish ducat, 
This anecdote, without which no biography of Bach is complete, is 
told by Marpurg in his Legenden eintger Musikheiligen, Cologne, 1786. 
*** Interesting details on the musical conditions at the Celle 
court are given by Andre Pirro in his /. 5. Back (Alcan, Paris, 
1906), p. 26 ff, 

f This was not the Court band. Johann Ernst was the younger 
brother of the reigning Duke. See Spitta I, 220 ff. 

ft Spitta (I, 224, 225) gives the specification as follows: Ober- 
werlc (Upper Manual): i. Principal (i.e., diapason) 8ft., 2. Viola 
da gamba 8 ft., 3. Quintaton 16 ft., 4. Gedackt 8 ft., 5. Quint 6 ft,, 
6. Octave 4 ft, 7. Mixture 4 ranks; 8. Gemshorn 8ft, 9. Cymbal 



Arnstadt (1704 1706). 101 

was Chris toph Herthum. He had married a woman of 
the Bach family, and besides being an organist was a 
Count's clerk of the kitchen. 

In Arnstadt Bach laid the foundation of his mastery 
of the organ. As his office claimed him only three trnes 
a week, he had a good deal of time to himself. In October 
1705 he was granted leave of absence for four weeks, in 
order to go to Lubecfc and hear the great organist Buxte- 
hude. He was thus present at the great memorial per- 
formance on 2nd December 1705, on the occasion of the 
death of Leopold I. We do not know whether he learned 
from Buxtehude only by hearing him play, or whether he 
had lessons from him. In any case the attraction of the 
master was so strong for him that he quite forgot the 
necessity of returning to Arnstadt. He stayed over Christ- 
mas and the New Year in Liibeck, and did not get back 
^o Arnstadt until the middle of February, 1706. 

On the 2 1st of that month he was summoned before the 
Consistory, to justify his having exceeded his holiday. The 
proceedings of the meeting are still preserved*. Bach 
did not condescend to make any excuses, but said he 
thought that his deputy would have filled the office in 
such a way that no complaint would have been possible. 
The Consistory availed itself of the opportunity to remon- 
strate with him on his extravagant way of accompanying 
the chorales ; and further reproached him with not attend- 
ing to the choir of scholars, and with having performed 
so little figurate music. 



i ft. 2 ranks 5 to. Trumpet 8ft,, n. Tremulant; 12. Cymbelstern. 
Brust-positiv (Choir): i. Principal 4ft., 2. Lieblich gedackt 8ft., 
3. Spitz flute 4 It., Quint 3 ft., 5. Sesquialtera ; 6. Nachthorn 4ft., 
7. Mixture i ft. 2 ranks. Pedal Organ: i. Principal 8 ft. 2. Sub- 
bass 1 6 ft., 3. Posaurie i6ft, 4. Flute 4ft,, 5. Cornet 2ft. 

The church itself had been finished in 1683. It was intended 
to replace the church of St. Boniface, that was destroyed during 
the great Arnstadt fire of 1581. See Weissgerber, Johann Sebastian 
Bach in Arnstadt (Gymnasialpy&gramm, Arnstadt, 1904), 

* They are in the archives of the Principality of Sondershausen, 
See Spitta I, 315 ff. 



102 VII. From Eisenach to Leipzig. 

We must not regard the church authorities as being 
lacking in a sense of the genius of the young organist. 
Their complaints were wholly justified. Bach had been 
unable to do anything with the choir, thus already reveal- 
ing that lack of talent for organisation that later on was 
to make his situation in Leipzig so difficult.' In this he 
differed radically from Schiitz, who always knew how to 
get the best results out of the material at his disposal, 
and to train his forces gradually up to the highest possible 
efficiency. Bach, on the contrary, was no pedagogue; 
he could not even maintain discipline. If things did not 
go as he wanted, he flew into a temper, thereby only 
making the matter worse lost heart, and let things go 
as they chose. He was on very bad terms with the singers, 
and with the scholar who led the choir. ' Before his Lubeck 
journey there had been a fracas between himself and one 
of the scholars, Geyersbach. The latter, having had an 
injurious epithet applied to him by Bach, had set upon 
him with a stick in the street. Bach had drawn his sword. 
Fortunately other scholars had thrown themselves be- 
tween them and separated them*. The affair had gone 
before the Consistory, where it was proved that Bach had 
really used the offensive epithet in question. 

At the February sitting he was required to say definitely 
whether he would attend to the choir or not; eight days 
were given him for reflection. In November he had not 
yet replied. He was again summoned to a sitting, on the 
eleventh, when he promised to give his reply in writing. 
Whether he ever did so we do not know. 

At this last sitting he was reproached with having re- 
cently "made music" in the church with a "stranger 
maiden", without having received permission to do so. 
He made the excuse that he had spoken about it to the 
clergyman, Magister Uthe. It need hardly be said that 
this "music-making" was a private and week-day matter, 



* Andr6 Pirro, /. S. Bach, p. 38. 



Miihlhausen (17061707). 103 

and that the "stranger maiden" did not take part in the 
Sunday service. This was not permitted under any cir- 
cumstances ; at that time women were not allowed to sing 
in church even in Hamburg. 

His position had become untenable. Just then, on 2nd 
December 1706, the organist of the church of St. Blasius 
in Miihlhausen, Johann Georg Able, died. In the spring 
of 1707 Bach received an invitation to give a trial per- 
formance on the organ of this free imperial town, which 
was artistic in its sympathies. On the 15 th June he 
received the appointment ; on the 29th he gave up the keys 
of the Arnstadt organ at the council house, and left to his 
cousin Ernst, son of Johann Christoph Bach, his father's 
twin brother, the five gulden of his salary which were 
still unpaid*. His emoluments in Miihlhausen consisted 
of 85 gulden, 3 coombs of corn, 2 cords of wood and 
6 trusses of brushwood, both to be delivered at his door 
and 3 pounds of fish per annum**. 

On the I7th October of the same year Bach married 
his cousin Maria Barbara Bach, daughter of Job ami Michael 
Bach, the organist and clerk at Gehren. They were 
married at Dornheim, near Arnstadt, by Johann Lorenz 
Stauber, who was connected with the Bach family by 
ties of friendship and kinship. This Maria Barbara Bach 
was presumably the "stranger maiden" with whom Bach 
had "made music" in the church at Arnstadt. Her mother, 
the daughter of Wedermann, the town-clerk at Arnstadt, 
had an unmarried sister in that town, Regina, whom 
she may have been visiting. Soon afterwards, in 1708, 
this aunt married Stauber, the clergyman at, Dornheim, 
who had lost his wife a year before. 

When Bach entered upon his post at Miihlhausen, the 
musical conditions of the town were in a woeful state of 
decay. The town was living on its past reputation. The 



* This cousin had probably acted as his deputy during his ab- 
sence in Liibeck. 
** Spitta I, 337. 



104 VI1 - From Eisenach to Leipzig. 

congregation was split in two by a disagreement between 
the orthodox party and the pietists, by which art did not 
profit much. A fortnight before the appointment of the 
new organist, a fire had reduced to ashes a great part of 
the town, and that the richest and most beautiful. It 
can easily be understood that just then the burghers had 
something else to think about than the reorganisation of 
the church music. They thought they had done their 
part when they had engaged an artist at an exceptionally 
good salary, He, however, had no capacity for reorganisa- 
tion. A year after his appointment he applied for his 
release. In his petition he frankly acknowledges that he 
is going because he does not see any immediate improve- 
ment in the musical conditions*. They parted, however, 
on good terms with each other. Bach was still to super- 
intend the work on the organ, the renovation of which 
had been undertaken according to his plans. 

In his new post Bach had nothing to do with a choir; 
he went to Weimar as Court organist and chamber musician 
to the reigning Duke Wilhelm Ernst**, one of the most 
distinguished and cultured princes of his time, and thor- 
oughly devoted to art. When Bach entered his service 
the Duke was in his forty-sixth year***. In the religious 
struggles he was on the side of orthodoxy, and took care 
that his people had the pure doctrine. He had married 
a Princess of Jena, but they soon separated. 

The court band numbered about twenty members. 
Many of them as was usual at that time, also acted 
as footmen, cooks or huntsmen. On special occasions 
they waited on their master in Hungarian costume; so 
that Bach also must have donned this dress f. The organ 
in the castle church was not large, but it is evident from 



* Spitta I, 373, 374. 

** On the Weimar epoch see Paul von Bojanowski, Das Weimar 
J. S. Backs, Weimar, 1903. 

*** See Spitta's sympathetic portrait of him (I, 375 fl). 
t Spitta, I, 378. 



Weimar (1708 1717). 105 

the specification of it that has been preserved that it must 
have had a uniformly fine tone*. Bach must occasionally 
have been incommoded by the fact that the pitch of the 
organ was the cornet- tone, i. e., a minor third above the 
ordinary pitch (Kammerton). 

In the town church there was a considerably larger 
organ. It was played by Johann Gottfried Walther, the 
subsequent author of the first German musical lexicon**. 
On his mother's side she was a Lammerhirt by birth, 
he was related to Bach. The two men seem to have formed 
a sincere friendship with each other, though we do not 
know whether they still had much intercourse after Bach 
had left Weimar. Spitta is of opinion that later on a cer- 
tain estrangement sprang up between them, since Walther 
in his Musiklexikon (1732) devotes only a moderately short 
article to Bach. The inference is rot conclusive. Walther's 
articles are confined to an enumeration of printed works. 
The section on Handel is still shorter than that on Bach. 

Bach's salary to begin with was 156 gulden; it increased 
until in 1713 it was 225 gulden. In the following year it was 
probably again increased, Bach having been then advanced 
to the position of Konzertmeister. From this time he had to 
provide cantatas for the church service. The Kapellmeister 
was Johann Samuel Drese, who was already well on in 
his sixties ; his son Johann Wilhelm acted as his deputy. 

We do not know whether Bach had any close personal 
relations with the prince his master. Probably not, for 



* It had two manuals and a well-equipped pedal. Upper Manual : 
i. Principal 8 ft., 2. Quintaton i6ft, 3. Gemshorn 8 ft., 4. Gedackt 
8 ft., 5. Quintaton 4 ft., 6. Octave 4 ft, 7. Mixture 6 ft., 8. Cymbel 

3 ranks; 9. Glockenspiel. Lower Manual: i. Principal 8 ft., 2. Viola 
di gamba 8 ft, 3. Gedackt 8 ft., 4. Trumpet 8 ft., 5. Small Gedackt 

4 ft., 6. Octave 4 ft., 7. Waldflote 2 ft, 8. Sesquialtera. Pedal Organ: 
i. Great "Untersatz" (support) 32 ft, 2. Sub-Bass 16 ft, 3. Bass- 
Trombone 1 6 ft, 4. Violin-Bass 16 ft, 5. Principal-Bass 8 ft., 6. Trum- 
pet-Bass 8 ft., 7. Cornet-Bass 4 ft. Spitta I, 380. The organ was 
placed under a cupola in the third gallery. See the picture of the 
church in von Bojanowski. 

** On Walther see p. 43 of the present volume. 



106 VII. From Eisenach to Leipzig. 

in that case it would be inexplicable how he came to .be 
passed over when, in 1716, a successor had to be appointed 
to the deceased Kapellmeister. At first an attempt was 
made to secure Telemann, who was at that time in Frank- 
fort. When he declined the offer, the post was given to 
Drese's son. He was a musician of little account, his sole 
claim being that he had always acted for his father during 
his last years. 

After this event, Bach's only thought was how to get 
away->from Weimar as quickly as possible. When Prince 
Leopold of Anhalt-Cothen offered him the post of Kapell- 
meister at his Court he snatched at it eagerly, which he 
certainly would not have done had it not been a case of 
finding another situation at any cost. The office, indeed, 
had little that was attractive for a man with the objects 
that Bach had in view. The Cothen Court belonged to 
the reformed church; there was consequently no church 
music. The church of the castle contained a small organ 
of inferior quality; that of the reformed town-church was 
rather larger. Bach was merely the director of his master's 
chamber music. 

In his haste to leave Weimar he seems to have demanded 
Ms immediate release in a rather peremptory way. The 
Duke, who did not approve of behaviour of this kind, 
had the refractory Court organist arrested on the 2nd No- 
vember, and kept under arrest until the 2nd December*. 

He took up his new position at Christmas 1717**. If 
it was not wholly satisfactory from the artistic stand- 



* See von Bojanowski, p. 63. He quotes from Borrnann's re- 
cords: "On 6th November the late concertmeister and organist 
Bach was arrested in the justice's room on account of his obstinately 
insisting upon his resignation being accepted at once, and was 
finally set free on the 2nd December and notified that he had been 
ungraciously allowed to resign.* 1 

** For information about Bach in Cothen see Rudolf Bunge, 
Bachjahrbuch, 1905, pp. 1447. One is shocked at the history 
of the band and the penury of the poor musicians after the death 
of Prince Leopold. See Bunge, p, 34. 



Weimar (1708 1717). 107 

point, in another respect it was extremely agreeable. The 
Prince was young he was not yet twenty-five and 
had had a sound musical education. He had travelled in 
Italy, taking with him Johann David Heinichen (1683 
1729) one of the most notable musical theoreticians of 
the time, to initiate him into Italian art. In the orchestra, 
which was not very large, the Prince himself seems to have 
played the violin. He also possessed a well-trained bass 
voice. He was well qualified to appreciate the worth of 
his new Kapellmeister. He was proud of him, and took 
him with him on all his journeys. In time a cordial friend- 
ship grew up between the two men, which lasted even after 
Bach had left Cothen. 

The six years that he passed in this small capital were 
the most pleasant in Bach's whole career, He had time 
for composition, and there was no unpleasantness of any 
kind to mar his joy in his work. During the Cothen 
epoch, however, there befel the most serious misfortune 
he had yet known. Returning with the Prince from Carls- 
bad in July 1720, he found that his wife had died suddenly 
in his absence, and had been buried on the 7th July. All 
her husband could do was to make a pious pilgrimage to 
the grave of her who for thirteen years had been the faith- 
ful and devoted sharer of his lot. Of the seven children 
that Maria Barbara had born to him, four were living at 
the time of her death; the eldest, a daughter named Katha- 
rina Dorothea, was twelve years old; Wilhelm Friedemann 
was ten; then came Philipp Emmanuel and his brother 
Johann Gottfried Bernhard, who was about a year younger 
than he*. 

A year and a half later Bach found a new life-partner 
in Anna Magdalena Wiilken, the daughter of Johann 
Caspar Wiilken, Court and field trumpeter at Weissenfels. 



* In 1703 the parents had lost two twins shortly after their 
birth; another son, born I5th November 1718 in Weimar, and 
named after his godfather Prince Leopold August, died on 28th 
September 1719. See Spitta II, 8. 



Io8 VII. From Eisenach to Leipzig. 

The marriage was solemnised on 3rd December 1721 ; the 
bridegroom was thirty-six, the bride twenty-one. This 
marriage was a thoroughly happy one in every respect. 
Anna Magdalena was not only a careful housewife, who 
behaved with the utmost kindness to Bach's motherless 
children; she was also an artist, who could enter intel- 
ligently into her husband's work. She had a good and 
well-trained soprano voice. Her husband made it Ms 
care to develop her musical faculties. We still possess 
two Kl&vierbuchlein [Little Books for the Clavier] von 
Anna Magdalena Bach, the first belonging to 1722, the 
second, which is in a fine green leather binding 
bearing the date 1725. The first contains twenty-four easy 
pieces for the clavier; the second consists of preludes, 
suites, chorales, and sacred and secular songs. Bach also 
instructed his wife in the art of playing from figured basses. 
At the end of the Klamerbtichlein of 1725 we find "some 
highly important rules of General Basso" recorded in his 
handwriting. His scholar richly repaid him for the pains 
he took with her, for she was useful to him in copying 
music. A number of the finest works of Bach have come 
down to us in her writing. In the course of years her 
script became so much like that of her husband that it is 
difficult to distinguish one from the other. For a long 
time, for instance, the score of the cantata heU'ges Geist- 
uncL Wasserbad (No. 165) was regarded as an autograph 
of Bach's, until Spitta proved it to be a copy made by 
Anna Magdalena*. How many hours must she have had 
to steal from her household duties when the week was 
drawing to a close, and the parts of the new cantata were 
not yet copied! 

She also taught the children to do this kind of work. 
In the second oboe part in the cantata Iht, die ihr euch 
von Christo nennet (No. 164), the headings, the key-signa- 

* Spitta II, 690. In the Bachjahrbuch for 1906, pp. 134, 135, 
Johannes Schreyer questions the theory that the two note-books 
were written for Anna Magdalena. 



Journey to Hamburg (1720). 109 

tures and the bar-lines are in her writing, but not the notes, 
which are clumsy and awkwardly connected, A small 
monogram at the end of the part, in which an attempt is 
made to intertwine the three initials W. F. B., shows the 
copyist to have been Wilhelm Friedemann Bach. The 
cantata probably belongs to the year 1724; the boy at that 
time was fourteen years old; it was his first fair copy. We 
can see him sitting at the table: the sunlight plays on the 
floor; the mother, busily flitting to and fro, supervises his 
work. He has just written 77 Fine at the end. But it is not 
done well enough for her; she writes the words again in 
her large, easy characters. There is a footstep on the 
stairs; the door opens; the father has come home. 

His growing boys now made it imperative for Bach to 
think of looking out for another post. In Cothen he could 
not procure for them the education that they needed. He 
himself longed to be back at his organ, and lamented his 
virtual severance from church music. Hamburg attracted 
him. Although the opera there had for some time now 
lost a good deal of its earlier glory, the town was still one 
of the musical centres of Germany. Here Mattheson sat 
in judgment upon artists and their works; here raged 
the war between the new and the old church music; here 
were the most splendid Organs; here lived Erdmann Neu- 
meister, the celebrated librettist of church cantatas. 

It so happened that the organist's post at St. Jacob's 
church became vacant in September 1720, by the death 
of its quondam occupier Heinrich Friese. A few weeks 
afterwards, Bach went to Hamburg and performed on the 
organ of St. Catharine's church before Reinken, who 
was then nearly a hundred years old and a select com- 
pany*. The story is well known of how the old master 
of the organ went up to the younger one, who had just 
improvised for half an hour on the chorale "An Wasser- 
fliissen Babylon", and complimented him with, the words: 

* Forkel (p. 8) gives the date of the journey wrongly as 1722. 



IIO VII. From Eisenach to Leipzig. 

"I thought this art had perished, but I see that it still lives 
oninyou." The praise was all the more flattering inasmuch 
as Reinken himself had treated the same melody at length 
in a chorale prelude, of which he was not a little proud*. 

Bach was exempted from giving a trial exhibition for 
the post at St. Jacob's. We may be sure that Neumeister, 
who was the clergyman of the church, strenuously urged 
his election. His candidature failed, however. The choice, 
which was made on the iQth December, fell on a certain 
Johann Joachim Heitmann. The church accounts let us 
see wherein consisted his superiority over Bach in the eyes 
of the authorities of St. Jacob's. On 6th January 1721 
he paid into the church treasury the sum of 4000 marks in 
acknowledgment of his election. The fact that he expended 
so much to secure the post leads us to surmise that it must 
have had some very lucrative perquisites attached to it. 

Neumeister was indignant, and gave vent to his vexa- 
tion in a sermon. Speaking at Christmas of the angels 
who made music at the birth of Christ, he added that their 
art would certainly have availed them nothing in Hamburg; 
he really believed, he said, that if one of the angels of Beth- 
lehem, who could play divinely, were to come down from 
heaven and try to become organist at St. Jacob's, but had 
no money, he would simply have to fly back again**. 

Whether Mattheson exerted himself on behalf of Bach 
in this affair is not known. That Bach did not settle at 
Hamburg can only be regarded as a misfortune. The 
position offered much fewer difficulties and occasions for 
mortification than the one he afterwards accepted in Leip- 
zig. On the other hand we must not forget that in Ham- 
burg he would have had to dispense almost entirely with 
the chorus in 'his church music, since there was no choir 



* Ante, pp. 45, 46, 

** For further details as to the election see Spitta II, 19 ff. We 
owe the passage from Neumeister's sermoa to Mattheson, who 
refers to the episode in 1728 in his Musikalische Patriot (p. 316), 
without, however, mentioning Bach directly by name. 



Cantor at St. Thomas's (1723). m 

there. And what encouragement for his creative work 
would he have found among a committee that set finance 
above art? 

A year and a half later, in June 1722, the post of cantor 
at St. Thomas's church in Leipzig became vacant. The 
council was looking for a worthy successor to Kuhnau. 
It did not, however, think of Bach in the first place, but 
entered into negociations with Telemann, who was at 
that time regarded as the leading German composer, and 
was favorably remembered by the Leipzigers from his 
student days (1701 1704)*. The negociations were broken 
off because Telemann could not obtain his release from 
Hamburg, where he had but recently (1721) been appointed 
director of the town music. After him, the Darmstadt 
Kapellmeister Graupner, a capable pupil of Kuhnau, was 
most thought of in connection with the post. Bach did 
not apply until towards the end of the year. He delayed 
it so long because he found it hard to leave his agreeable 
situation with his cultivated prince, and to surrender the 
position of Kapellmeister for that of a simple cantor, to 
be under the orders of a school rector and to teach choir 
boys. He finally succeeded in making the resolution to 
sacrifice his leisure and his pride to his children. "At 
first it was not altogether agreeable to me", he writes 
some years later to his friend Erdmann, u to change the 
position of Kapellmeister for that of cantor. Conse- 
quently I delayed my resolution for a quarter of a year; 
nevertheless this post was so favorably described to me 
that finally, especially as my sons appeared to be inclined 
to study, I ventured upon it in the jiame of the Most High 
and went to Leipzig, passed my trial, and at once set about 
the removal." 



* See p. 86, The story of the election is told by Richter in 
his interesting essay Die Wahl J. S. Backs zum JKantor tier Thomas- 
schule im Jahre 1723, in the Bachjahrbuch for 1905, pp. 48 67. 
See also Kleefeld, Bach und Graupner al$ Bewerber urn das Leip- 
ziger Thomaskantorat, in Peters Jahrbuch t p. 70. 



113 VIII* Bach in Leipzig. 

He was not exempted from the trial, which had not 
been waived even in the case of Telemann. He produced 
the cantata Jesus nahm zu sich die Zwolfe on Quinqua- 
gesima Sunday, 7th February 1723, As Graupner could 
not -get his discharge from the Darmstadt court, and the 
other competitors could not measure themselves against 
Bach, he was elected unanimously, 

It has recently become the fashion to reproach the Leipzig 
council with having only taken up with Bach after it had 
vainly tried to secure the * Shallow" Telemann and the 
insignificant Graupner. The reproach is unjustified. Both 
the other men were well known in Leipzig, and had a repu- 
tation with their contemporaries that Bach did not yet 
possess. It is too much to expect of a committee that it 
should anticipate the judgment of posterity. The town 
council's business was to find a musician of recognised 
ability to succeed Kuhnau, and not to be influenced by 
any other consideration. It therefore finally selected 
Bach. The choice honoured both the judges and the com- 
petitor; for Bach certainly felt it an honour to become 
the successor of Kuhnau. 

His nomination was notified to him on 5th May 1723; 
and on Monday, the sist of the same month, he was in- 
stalled in his new office. He took up his quarters in the 
cantor's house in the left wing of the St. Thomas's school 
buildings. His wanderings were at an end. 



CHAPTER VIII. 
BACH IN LEIPZIG. 

One cannot help feeling a little sadness when one reads 
the agreement that Bach had to sign upon his nomina- 
tion*. He was not to leave Leipzig without the permission 
of the burgomaster, and he undertook, for the sake of 

* See Spitta II, 185 ff. 



n Jf l^m**s *v/i, 2 <2)i 




St. 'I'liomas's Church and School in 1723 

The illustration shows the old school (erected in 1553), in which Bach worked ; 

it was extended and heightened in 173-2 
(i) St. Thomas's Church, (is) St. Thomas's School. (3) The Stone Cistern. 



The Duties of the Cantor, 113 

economy, to instruct the boys not only in singing but in 
instrumental playing, so that they might also be available 
in the church orchestra. It was further part of his duty 
to accompany the choir of school boys who sang the motet 
or chorale at funerals. At the less important burials, in 
which only a portion of the St. Thomas's choir took part, 
one of the senior scholars would often act as his deputy. 
But how often must he have walked abstractedly with 
his scholars at the head of the procession, in wind and rain, 
turning over his next cantata in his mind! 

Before his definite installation he had to undergo, as 
was the custom of the time, an examination of his religious 
belief, out of which he came satisfactorily. He also had 
to sign the Concordia Formula, for without signing this 
no one could hold an appointment in Saxony*. 

It was not the best of omens that at this formal induc- 
tion into his office, on Monday, the 3ist May, the represen- 
tatives of the consistory and those of the council almost 
came to blows, the delegates of the council being of the 
opinion that some remarks that Licentiat Weisse, who 
had come in lieu of Superintendent Deyling, had ad- 
dressed to Bach in the name of the consistory, claimed for 
the church committee a right at the ceremony that did 
not belong to it. The affair afterwards led to a long dis- 
cussion between the council and the consistory. Thus on 
the very first day there broke out that rivalry between 
the two bodies that controlled the cantor, which was 



* The Concordia Formula is the last symbolical document of 
Lutheranism. It was drawn up at the end of the eighth decade 
of the 1 6th century at the instance of the Electoral Prince of Saxony, 
and was meant to unite the whole of the Lutheran established 
churches of Germany. The majority accepted it. This confession 
put an end to the theological controversies that had agitated the 
church of the time, and condemned not only Calvinism, but also 
the milder tendency in Lutheranism that came from Melanchthon. 
In a later day, pietism and the "Aufklarang" lowered the authority 
of the Concordia Formula. Its political and juristic authority 
broke down, with the rest of the old Germany, during the Napole- 
onic wars. 

Schweitzer, Bach. 8 



VIII. Bach in Leipzig, 

afterwards to re-appear so frequently. We cannot say that 
Bach suffered from this tension. It ministered admirably 
to his own need for independence, for he played the con- 
sistory off against the council and the council against the 
consistory, and meanwhile did what he liked. 

His work at St. Thomas's school was not vexatious. 
Besides the rector there were seven teachers. The cantor 
came immediately after the subrector, thus being third 
in rank, and, like the rector, he had to teach only three 
hours a day. Besides taking the upper classes in singing, 
the cantor had to instruct the third class in Latin. Bach 
had expressly declared his willingness to take these classes, 
and he appears to have felt a certain pride in doing so, 
while Telemann had asked to be absolved from non-musical 
teaching, Afterwards, however, this work did not please 
him, and he asked his colleague Magister Petzold to act 
as his deputy in return for a enumeration of fifty thalers 
per annum. The council gave its consent to this. 

Singing classes were held on the three first days of the 
week, at nine o'clock and twelve o'clock. On Thursday 
the cantor was quite free; on Friday he taught a singing 
class at twelve o'clock. On Saturday afternoon, after 
vespers, while private confession was going on below, the 
rehearsals for the cantatas took place. Bach seems not 
to have taken the instruction in singing very seriously. 
The reproach is always being made against him that he 
left it almost entirely to the older scholars, instead of at- 
tending to it himself. To what extent he was really lax 
in his views of his duties we cannot now judge; but the 
charges that were made against him can hardly have been 
without some justification. 

But even if he fulfilled all his duties most scrupulously 
he would be occupied in the school no more than two or 
three hours each day. Every four weeks he had to inspect 
the school. This was the duty of the four principal teach- 
ers, who during this week had to submit to the general 
regulations of the school, and perhaps even sleep in it. 



Bach's Financial Position, 



115 



We know that in his later years, when it was his turn to 
do the inspection, Bach was exceedingly irregular in his 
attendance at devotions and at grace. In any case he 
was not overworked, and had plenty of leisure for com- 
position. 

He gives his salary, in the letter to Erdmann, as about 
seven hundred thalers*. Of these, indeed, only some- 
thing over a hundred thalers was fixed; this was augmented 
by school-fees and receipts from legacies, of which he re- 
ceived a share. The main part of his income, however, 
came from the wedding and funeral ceremonies, from 
which he always received something according to the 
"class". The highest payment for weddings was two 
thalers, for funerals one thaler fifteen groschen. 

The income of the cantor was thus rather variable. 
Kuhnau had complained that many wealthy people, in 
order to avoid expense, were married in a neighbouring 
village church, or dispensed with music of any kind, vocal 
or instrumental, at their funeral. Occasionally even the 
weather had something to do with it* In 1729, as Bach 
himself says in the letter to Erdmann, the air was so 
healthy that he was a hundred thalers to the bad on the 
ordinary funeral receipts. That year which was a 
happy one for posterity, for it gave us the St. Matthew 
Passion, left a bad memory with the creator of that 
work, for during it the people of Leipzig would not die in 
sufficiently large numbers! 

Another part of the emoluments consisted of the share 
the cantor received of the monies which the alumni col- 
lected by singing at the houses at Michaelmas and the 
New Year. Compositions for special occasions appear to 
have been fairly well paid for. Leipzig was certainly not 
an Eldorado for musicians. The burghers, like the council, 
thought a good deal of art, but spent nothing on it. Mari- 
anne von Ziegler, the intelligent widow of an officer, and 



* The letter will be given in full later on. 

8* 



1 1 6 VIII. Bach in Leipzig. 

a crowned poetess, at whose house a good deal of music 
was performed, says in a letter of this date: "The remunera- 
tion that the musicians get for their trouble here is as a 
rule poor, and they often have to be thankful if, in return 
for several hours of musical service, they get a lean bone 
to pick. How then can such people live, since no one has 
any care for them or gives them any help?"* The foreign 
musicians in Dresden and at other princely courts were 
differently situated. 

On the whole, however, Bach's income, if we take into 
consideration the value of money at that time, cannot 
have been a poor one. He brought up his large family 
honestly, gave his children a good education, was profuse 
and cordial in his hospitality, and at his death left not 
only a rich collection of first-rate musical instruments, but 
also a not insignificant sum of money. His household 
property, judging from the inventory that was drawn up 
when it came to be divided, was that of a well-to-do 
burgher**. Anna Magdalena was certainly an excellent 
housewife. Bach himself was a good man of business, and 
did not treat money matters as an unimportant part of 
life. One even has the impression that he frequently put 
them very decidedly in the foreground. 

The position of the cantor would thus have been from 
every point of view desirable if better conditions had pre- 
vailed in St. Thomas's school at the time* When Bach 
joined the institution, however, it was in an advanced 
state of decline. This had not come about all in a day. 
For years previously the rector and the council had seen 
clearly what was going on, and they had exchanged many 
documents on the subject, enacting new ordinances and 
reviving old ones. No effectual remedy, however, had been 

* See Philipp Spitta, Marianne von Ziegler und Joh. Seb. Bach, 
in Zur Musik, Berlin, 1892, pp. 93 119. The composer used the 
works of this poetess, (who was closely connected with, the Gottsclied 
circle) for the texts of eight of his finest cantatas. 

** The inventory of the property Bach left is in the district 
court at Leipzig. It is given in Spitta, III, 351 fl 



State of St. Thomas's School. 117 

found. Kuhnau in his latter days had suffered a good 
deal from the bad state of affairs, which were the result 
of several causes. The rooms in which the scholars were 
lodged were far too few in number, and left a good deal 
to be desired from the point of view of health. The school 
was therefore a centre of contagion. The scholars lived 
almost in dirt; Kuhnau, in one of his memorials, speaks 
of certain scholars as scabious. It was consequently not 
surprising that the burghers sent their children elsewhere. 
In the three lower classes, that had formerly contained a 
hundred and twenty scholars, there were in 1717 only fifty- 
three*. The charitable portion was always completely 
occupied, for the free admissions were in much request. 
These young people, however, were hard to keep in order. 
Discipline had quite disappeared. The rector at that time, 
Heinrich Ernesti, who had presided over the institution 
since 1684, was a man without any energy. 

If the place was to be reformed, the singing in the streets 
would first have to be abolished. This was impossible, 
however, since the rector and the two head teachers made 
a not inconsiderable profit out of the collections, and the 
scholars themselves were dependent upon these earnings. 

How art fared in such an establishment can easily be 
imagined. Kuhnau's memorials to the town council paint 
a truly wretched picture**. The voices of the young scholars 
were ruined by their going about singing in storm and 
rain before they had had proper training. As these rounds 
were mostly made at the New Year, it was impossible to 
rehearse with any thoroughness the cantatas for that 
season. Formerly boys with good voices had been engaged 
as supernumerary alumni, From false considerations of 
economy this practice had been in the course of time given 
up, to the great detriment of the music. Previously, also, 
the cantor had had at his disposal certain small allow- 
ances by means of which he could ensure the cooperation 

* Spitta II, 199 ff. 
** See Spitta III, 303305 ; Five Memorials of Kuhnau. 



II 8 VIII. Bach in Leipzig. 

of musical students for the orchestra and male voices for 
the choir. These also had been gradually taken from him. 
The majority of the students consequently stayed away, 
and the cantor had to do the best he could with the eight 
town-pipers and his own boys. Seeing no way out of the 
difficulty, Kuhnau begs that something from the receipts 
of the alms-bag may be devoted to the choir, so that he 
may provide the voluntary helpers with some entertain- 
ment, and give them a dinner once a year. "This" he 
observes at the end of the memorial, "since it concerns 
the advancement of church music and the glory of God, 
would be as much, perhaps even more, a causa pict than 
if the money were given to beggars, who would not do the 
same good with it." 

The town council turned a deaf ear to these appeals. 
Speaking generally, the cantor had not control of so much 
as a pfennig. To get a board with nails in it fixed in the 
church, on which the violins could be hung, he had to ad- 
dress a petition to the council. He also had to beg for 
some violin cases, in order that the instruments might not 
receive more damage than was necessary when being taken 
from one church to another. The impression given us by 
all these petitions is that Kuhnau had no standing at all 
as regards the council. 

The situation was particularly critical from the fact that 
Kuhnau was quite uninterested in the newer musical life 
in Leipzig. The new art was antipathetic to him ; he hated 
everything connected with the opera or operatic music*. 
Moreover he not only saw his best choristers and soloists 
enticed from the school by the opera before they had fin- 



* He had composed an opera, which, however, had been a com- 
plete failure. No doubt this was the origin of his violent hatred 
of the newer music. See Richard Munnich's able article, Kuhnaus 
Leben, in the Sammelbande der Internationalen Musihges&tts&haft, 
IQOI 1902, pp. 473 527. Kuhnau's rough satire on the Italian 
musicians and their German imitators, Dev musikalische Quack- 
salber (1700), has recently been re-issued by Benndorf (Behr, Berlin, 
1900). 



The Question of the University Service. 119 

ished their time, but found that the more music-loving of 
the students even those who did not ask for remunera- 
tion turned their backs on the official church music 
and hankered after an art that offered more satisfaction 
to the children of the epoch. The foundation of Tele- 
mann's singing union at the beginning of the eighteenth 
century had been a severe blow for the cantor of St. Tho- 
mas's. Still worse was to come. Though he was the director 
of the services at all the churches in Leipzig, one of them 
made arrangements to dispense with him, and to supplant 
the other churches in the favour of the citizens and the 
council by admitting the modern style of music. This 
was the New Church, at which Telemann was organist in 
1704, His successors continued the singing union he had 
founded, and so drew the students to them. 

Telemann's performances on feast days and at fair times 
had caused some sensation. The council had been generoxis 
with its help. In vain did Kuhnau assert his rights as 
director of the church music in Leipzig. He did not suc- 
ceed in bringing the New Church under his jurisdiction 
again, in spite of his warnings to the council as to the 
dangers of theatrical music in the church service, - When 
on one occasion he refused to place the St Thomas 
singers at the disposal of the New Church for a perform- 
ance of the Passion, he was compelled to do so by his 
superiors. 

He almost, indeed, lost another church. Until 1710 the 
university church of St. Paul had been used only for the 
academic services at the three great feasts, at the com- 
memoration services for the Reformation, and once a 
quarter, The cantor of St. Thomas's had been in charge 
of the music, for which he had been paid by the university. 
In 1710 the university arranged a service in this church 
for each Sunday, and the question arose as to who should 
have the musical direction of this. A law student, Joh. 
Friedr. Fasch, who later on became Kapellmeister in 
Zerbst, who had founded a second Collegium musicum by 



I2O VIII. Bach in Leipzig. 

the side of that of Telemann, applied for the post*. Kuhnau 
could only save this "new service", as it was called, for 
the cantor of St. Thomas's by declaring his readiness to 
attend to the music for it gratis, and to be content with 
the payment he already received for the academic festival 
services. 

This was the condition of things when Bach entered 
upon his office. There was no hope of any improvement 
in the state of the school so long as the old rector, with 
whom, for the rest, he was on very good terms, was 
still alive. To accomplish anything at all he had to set 
his hope on the university, and try to interest the academic 
world in his artistic undertakings. The prime essential 
was to get the university service in his hands, and to restore 
the full authority of the cantor of St, Thomas's as supreme 
director of the whole of the Leipzig church music. During 
.the long interregnum a certain Corner, an arrogant but 
insignificant musician, who had formerly been organist at 
St. Paul's, and now held that office at St. Nicholas's 
had offered to look after the music for the university church. 
If Telemann had succeeded Kuhnau in Leipzig, he would 
have received the post of director of the academic church 
music, since he had done the university the honour to 
ask expressly for it, as if he were not aware that the post be- 
longed by prescriptive right to the cantor of St. Thomas's, 
Bach appears to have omitted to do this ; and the univer- 
sity authorities now definitively assigned the old and the 
new service to Gorner, just to show "that the academy is 
not bound to accept the town cantor every time"**. This 
was on the 3rd April 1723, three weeks before Bach 
was appointed by the council. 



* The son of this Fasch, Karl Friedrich Christian (1736 1800) 
founded in 1792 the Berlin Singakademie, which in course of time 
revived Bach's St. Matthew Passion. 

** Spitta II, 210 fl B. F. Richter published the results of some 
new research into this matter in the Monatskefte fttr Musikgeschichta 
for 1901, p, looff. 



The Question of the University Service, 121 

Scarcely was the new cantor settled when he plunged into 
the fight over his position. On 28th September 1723 he 
petitioned the university for the twelve thalers that had 
always been allowed for the "old service" they were 
paid out of endowments and that Kuhnau had regularly 
received. He met with a refusal, but appears to have pro- 
vided some festival compositions for which he received 
payment. The affair went on for two years. On 3rd No- 
vember 1725 Bach addressed a written appeal direct to 
the King, who had the matter looked into at once, and 
demanded a report from the university, to which Bach 
afterwards replied. The end of all this discussion was that 
the university was ordered to pay the cantor the twelve 
thalers, both for the past and in the future, since they were 
provided by old endowments; Gorner retained the "new 
service", for which he was specially paid; Bach provided 
the compositions for the feast days; for academic festival 
functions the university appointed now one, now the other. 
Bach had thus won only half a victory. Gorner was the 
director of the university music, though as organist of 
St. Nicholas's he was Bach's subordinate*. 

This energetic action did not win much sympathy for 
the master in the academic world. Even the members of 
the council were perhaps not altogether pleased to find 
that they had a cantor who dared, at the first opportunity, 
to make a direct appeal to the sovereign, trusting to his 
good connections in Court circles. It is certain that the 
university gave the preference to Gorner when commis- 
sioning compositions for academic festival occasions. When 
a certain Herr von Kirchbach, acting on instructions from 
the Court, commissioned Bach to write the funeral ode for 
Queen Christiane Eberhardine, who had died on 7th Sep- 
tember, the university tried to get the bearer of the order 

* Johann Gottlieb Gorner was born in 3697, at Penig in Saxony. 
In May 1712 he entered the St. Thomas school as an alumnus, be- 
came organist of St. Paul's in 1716, organist of St. Nicholas's in 
and of St. Thomas's in 1729. He died X5th February 1778. 



122 VIII. Bach in Leipzig. 

from the Court for the "ovation" which was to take 
place at St. Paul's to countermand the commission that 
had been given to Bach, and transfer it to Corner, and 
sent word to von Kirchbach that "Bach would not be ad- 
mitted", When von Kirchbach offered twelve thalers to 
the university music director as compensation for not 
getting the commission, the university authorities gave 
way, but demanded that Bach should sign a declaration 
admitting that the permission to produce his music at this 
ceremony in St. Paul's was to be regarded as a favour for 
one occasion only, carrying with it no similar rights for the 
future, and that he should not undertake such commissions 
again without the permission of the university*. Bach 
naturally would not sign this document* The university 
registrar, who called on him at eleven in the morning of the 
nth October to get his signature, had to go away at mid- 
day without having effected his purpose. 

In 1729 Gomer was promoted to the organistship at 
St. Thomas's, but still retained his post as university mu- 
sical director. Bach must have been more than once 
nettled by this presumptuous person. It is related that 
once, at the rehearsal of a cantata, he flew into such a 
passion with the organist, who was always going wrong 
in the accompaniments, that lie tore off his wig and threw 
it at the man's head, telling him that he would have done 
better to have been a cobbler. If the anecdote is true, 
it is most likely to have been Gorner who was the object 
of this singular missile. 

In time the relations of the two men seem to have im- 
proved. When Bach's property was divided after his 
death, Gorner was appointed guardian of the four children 
who were underage**, which would hardly be explicable 

* The document is still preserved; it is couched in terms that 
are truly humiliating for Bach. See Kichter, Joh. S&b, Bach und 
die Umvwsitat zu Leipzig, in the Monatshefte fiir Musikgeschichte, 
1901, pp. 150 ff. 
** Spitta III, 357. 



The Choir and the Orchestra, 123 

unless his latter relations with Bach had been fairly 
good. 

Until the year 1729 nothing unusual occured between 
the composer and the council. Bach accommodated him- 
self to the situation as well as he could, and did his duty 
in the school and in the church. As the New Church and 
St. Paul's had by now become almost independent, he had 
for the most part only St. Thomas's and St. Nicholas's to 
attend to. 

The forces at his disposal were slight. St. Thomas's 
had a total of fifty-five alumni. Out of these, four choirs 
had to be formed: one for St. Thomas's, one for St. Nicho- 
las's, one for the New Church, and one for St. Peter's. 
Bach naturally left to the last two the choirs composed of 
the indifferent and bad singers; at a pinch they could be 
used in the chorales and motets, but were no use in solo 
music, which, indeed, was not essential in these two churches. 

Even in the best circumstances he could not get more than 
three voices to a part in his choirs. As a rule he had to be 
thankful if he could get the two principal choirs at this 
strength, that is, composed of twelve singers each. And 
small as his vocal forces were, there had to be deducted 
from them the singers whom he needed in the orchestra. 
The council gave him only eight town-musicians. No more 
students could be found to help; the gaps therefore had 
to be filled with scholars. In ordinary cases he had an 
orchestra of about eighteen to twenty, two or three 
first violins, as many seconds*, two violas, two violon- 
cellos, one "violon" or contrabass, two or three oboes, ac- 
cording to requirements, and one or two bassoons, flutes 
and trumpets. 



* Kuhnau, in one of his memorials, asked for four to each string 
part. Copious details of the musical conditions at St. Thomas's 
school during Bach's time are given by B. F. Richter in his article 
on Stadtpfeifev und Alumnen dw Thomasschule in Leipzig %u Backs 
Zeit, in the Bachjahrbuch for 1907, pp. 32 78. He gives the names 
of the singers and instrumentalists whom Bach had at his disposal 
during his cantorate. 



124 VIII. Bach in Leipzig. 

Deductions of course had to be made for those who were 
sick or hoarse, those whose musical education was not suf- 
ficiently advanced to permit of their entering the choir, 
and those who were generally good for nothing. At the 
end of the school year, when the scholars had had a whole 
year's training, he would be able to bring together the 
sixteen singers and twenty instrumentalists, and even per- 
form the Passion with a double chorus and a double or- 
chestra, as in the case of the St. Matthew Passion, for ex- 
ample, though it is still doubtful whether, at the first 
performance of this work, he had three or four voices to 
each part. But how he managed after Easter, at the be- 
ginning of the new school year, remains a mystery. Cer- 
tainly many a cantata is orchestrated as it is simply be- 
c'ause at that particular time Bach had only those instru- 
ments at his disposal. 

The chorus and the orchestra were divided into solo 
and ripieno groups*. The soloists sang the arias and re- 
citatives, and also sang with the choir. Special soloists 
were not employed. Kuhnau had urged the council to 
engage two extra soloists, who should be exempt from all 
other duties, especially that of singing in the streets. His 
request, however, was not granted. The arias and recita- 
tives of the St. Matthew Passion were thus sung by school- 
boys. We must not rate their performances either too 
high or too low. The technique of singing was at that 
time a more general possession than it is now. Colorature 
and trills were practised even in the elementary stage of 
instruction, and anyone who possessed the least natural 
aptitude for singing could soon acquire a certain, though 
may be a superficial, facility. Unfortunately we have no 
means whatever of learning how Bach taught singing. 
Possibly his scholars sajig the arias better, from a tech- 
nical point of view, than we might expect ; into the spirit 
of them, however, they could hardly penetrate. They had 



* Ripieno (Italian for "full"), is the antithesis of solo. 



Music in tie Leipzig Service, 125 

not the necessary time for this. We must remember that 
they had to perform a new aria an& a new recitative each 
Sunday, to say nothing of the Feast days. 

The orchestral ripieno played only in the choruses and 
in the tutti passages of the arias; the solo singers were 
accompanied by the solo group of instruments alone. This 
consisted chiefly of strings, there being very few of the wind 
to each part. Here again we can form no conception of 
how the musical mechanics for the town-musicians 
were nothing else or the scholars, who, in addition 
to their other occupations, had to learn an instrument 
from an older scholar in a few months managed with 
the oboe, flute and trumpet parts, that offer such great 
difficulties even to the expert wind players of the present 
day. We can only suppose that the art of wind-instrument 
playing stood on a higher general level then than we can 
now imagine. It is always difficult to form even relatively 
accurate ideas upon the state of executive art in a partic- 
ular period of the past. These are purely practical matters 
upon which we possess no documents, and upon which 
documents in any case could tell us nothing. Certain arts 
and aptitudes assuredly die with particular generations, 
and never again re-appear in quite the same way. 

At the head of each of the four choirs was a prefect. 
These posts were much sought after, the prefects taking a 
special share of the receipts from the street singing and 
the other odd sources, so that they could lay by, while at 
school, something substantial for their later student period. 
The cantor conducted the choir that performed the can- 
tata on a given Sunday; the duty was shared on alter- 
nate Sundays by St. Thomas's and St. Nicholas's. If the 
cantor conducted the cantata at St. Thomas's, the prelect 
of the other choir conducted the motet at St. Nicholas's, 
and vice versd. 

This rotation could on no account be deviated from 
as regards either the cantatas or the Passions. It happen- 
ed once that Bach wished to produce the Passion at 



126 VIII, Bach in Leipzig. 

St. Thomas's, where the space was more adequate 
though for that year it was the turn of St. Nicholas's. The 
printed programmes, in which St. Thomas's was mentioned 
as the place of performance, were actually in the hands 
of the public. It mattered nothing. The town council 
would not permit the cantor to do as he chose, and the 
latter had to yield. 

Of Bach as a conductor nothing is known. At that 
time it seems to have been customary to conduct church 
music with a sheet of music rolled up like a baton. On 
the title page of Walther's musical lexicon, the conductor, 
a roll of music in each hand, stands behind the organist 
and by the side of the contrabassist. Many, again, con- 
ducted from the clavicembalo; others led with the violin. 
There was no question of conducting in the modern sense, 
i. e., a communication and translation of musical inten- 
tions; it was simply a matter of keeping the players and 
singers in correct time. Later on Bach had the Rxick- 
positiv of the instrument at St. Thomas's arranged so that 
he could play on this independently of the great organ. 
He undoubtedly had the manual fitted up in this way for 
convenience in conducting. Sitting there, he could survey 
both choir and orchestra, and in the more difficult solo 
numbers could himself play from the figured basses. 

Cantatas were sung each Sunday, with the exception of 
the last three in Advent and the six of Lent; in addition 
there were the three Feasts of the Virgin, the New Year, 
Epiphany, Ascension, the Feast of St. John, Michaelmas, 
and the Reformation Feast, in all, fifty-nine cantatas 
every year. If Bach really wrote five complete yearly 
cycles of cantatas, as the obituary notice says and Forkel 
also affirms, then the total number comes to two hundred 
and ninety-five. About one hundred, then, must be re- 
garded as lost, since we possess only a hundred and ninety. 

The service in both the principal churches began at 
seven o'clock. The. organ prelude was followed by the 
motet; then came the Introit; after this the Kyrie, that 



Music in the Leipzig Service. 127 

was sung once in German in the hymn "Kyrie Gott 
Vater in Ewigkeit" and once in Latin. The Gloria 
was intoned from the altar, and answered either by the 
choir with et in terra pax, or by the congregation with 
"Allein Gott in der Hoh' sei Ehr" (To God alone on high 
be praise), the German version of the Gloria. After the 
collect, the Epistle was read, or rather sung in the old 
psalmody. This was followed by a congregational hymn, 
whereupon the Gospel was chanted by the priest, who also 
intoned the Credo. Then the organist began to preludise, 
keeping mainly in the keys which the instruments needed 
for tuning. At a sign from the cantor he ceased, and the 
cantata began, at the end of which the hymn "Wir glauben 
all an einen Gott" (We all believe in one God) was sung. 

The cantata lasted on an average about twenty minutes. 
In the summer the cantor did not need to keep to this 
time so precisely as in the winter, when the cold made 
them take care that the already long service was not made 
too long. It was no light matter to stay in the cold church 
three or four hours, which was the time the service lasted. 
In St. Nicholas's Church the choristers maintained a coal 
fire ; at St. Thomas's they went out during the sermon 
and warmed themselves in the school. They did not, 
however, escape the sermon, for while there they had to 
read one, the rector usually being present, as Kuhnau, 
to whom we owe this interesting information, says in one 
of his memorials to the council. There was no fear of their 
miscalculating the time, since the sermon, according to 
rule, had to last exactly an hour, from eight to nine o'clock. 

The sermon was followed by a prayer and the blessing, 
and then a congregational hymn led into the second part 
of the service, the communion celebration. German hymns 
were usually sung during the communion. The choir at 
St. Thomas's was as a rule no longer at its full strength 
at this stage, as the alumni had to prepare the table in 
the school for the meal at eleven o'clock. Kuhnau even 
found it sufficient for the prefect alone to remain behind 



128 VIII. Bach in Leipzig. 

in the church to lead the communion hymns*. The cantor 
usually left immediately after the cantata. Thus Kuhnau 
had been able to offer to take charge of the music at the 
university service, since he could get there from St. Tho- 
mas's in time for the cantata. Bach must often have 
remained to play the organ during the communion, there 
being plenty of scope here for preludising and impro- 
vising, as is proved by his chorale preludes upon com- 
munion hymns. In the first part of the service the organist 
mostly had his opportunity during the opening prelude, 
and afterwards before the congregational hymn that came 
between the Epistle and the Gospel. The chorale "Allein 
Gott in der Hoh' sei Ehr" seems also to have been pre- 
faced by a performance on the organ, for otherwise it is 
hard to explain why Bach wrote more preludes upon this 
melody than upon any other. 

At a quarter to twelve there, was a short service with a 
sermon, at which the choir had not to assist. Vespers began 
at a quarter past one with a motet. After various prayers 
and congregational hymns came a sermon, as a rule upon 
the Epistle; this was followed by the German Magnificat. 
At the end "Nun danket alle Gott" (Now thank we all 
our God) was sung. 

On the three last Sundays in Advent, and in Lent, 
no cantatas were given, the organ was silenced**, and the 
motets were discontinued. Instead of these, the Nicene 



* See Kuhnau's memorial of i8th Dec. 1717: "The scholars 
have a coal fire in the St. Nicholas's church. At St. Thomas's they 
go out and read the sermon, the Herr Rector generally being with 
them. Ours, however, cannot have this or indeed endure it. Thus 
during communion the choir is not intact, many of them having 
to go out to attend to the arrangements for eating. Nor are they 
necessary; for it has always been found to be better that the prefect 
alone should start the hymns and verses and sing them with the 
congregation." 

** This means that even the congregational hymns were not 
accompanied by the organ, but were led by the choir alone. On 
ordinary Sundays, too, the sermon hymn was sung without organ 
accompaniment. 



Music in the Leipzig Service, 129 

Creed was sung by the choir in Latin; and after the Epistle 
the Litany, the hymn of intercession of the ancient church, 
was sung, the congregation joining in. On these Sundays 
the Kyrie seems to have been rendered in the concerted 
musical form. We know this to have been the case on the 
first Sunday in Advent, 

The services on Feast days were quite overloaded with 
figurate music, On the first two days the three great 
Feasts were each celebrated for three days cantatas 
were given at vespers as well*. The St. Thomas choir 
gave the cantata it had performed at its own church in 
the morning at St. Nicholas's in the afternoon, the choir 
of the latter church singing at St. Thomas's, in the after- 
noon, the cantata that had been given at St. Nicholas's 
in the morning. On the naming of the first day of the 
Feast, Bach always conducted in St. Nicholas's, that church 
deriving a certain pre-eminence from the fact that the 
superintendent, Salomo Deyling, was minister there. On the 
third day, music was given in one of the churches only. 

Further, on Feast days the old hymns were sung at the 
commencement, before the organ prelude. During the 
communion service the Sanctus was sung, and at vespers, 
after the sermon, the Magnificat. The first Sunday in 
Advent and the Annunciation, even though they came 
during Lent, were celebrated as Feast Sundays, with organ 
and figurate music. 

As is generally known, Bach has noted down the order 
of the service for a certain first Sunday in Advent on the 
cover of the cantata Nun komm der Heidcn Heiland (No,6i). 
It is headed, "Order of divine service in Leipzig on the 
morning of the first Sunday in Advent 5 ', and runs thus: 

"The Prelude. Motet, Prelude to the Kyrie, which 
is accompanied throughout. Intoning at the altar. Read- 
ing of the Epistle. The Litany sung. Prelude to the 



* Also at New Year, Epiphany, Ascension, Trinity and Annun- 
ciation. 

Schweitrcr, Bach. 9 



130 ^ ^ IIL Bach in Leipzig. 

chorale. Reading of the Gospel. Prelude to the principal 
music*. The Creed to be sung. The sermon. After the 
sermon, as usual, some verses of a hymn to be sung, Vcrba 
institutionis. Then preludes and singing of chorales alter- 
nately until the end of the communion et sic porro" 

At the evening service on Good Friday the Passion was 
performed. If it were in two parts, the first was sung 
before, the second after, the sermon; if in one part only, 
it came before the sermon. This was the more rational, 
as the sermon was preached on the burial of Christ; while 
the two-part Passion took up the story again, after the 
sermon, with the trial and condemnation. When Bach 
came to Leipzig, the Passion performances at vespers were 
still quite new, dating only from 1721 . In that year Kuhnau 
had had to bow to the taste of the times by writing a 
Passion in the modern concert style, so that St. Thomas's 
should not lag too far behind the New Church, Previously 
Leipzig had known only the old motet Passion in the 
a capella style, which took the place of the reading of the 
Gospel at the principal service. As the first vesper Passion 
was produced at St. Thomas's in 1721, and the services 
were held in alternate years at each of the two principal 
churches, it is easy to discover in which church Bach's 
Passions were given in any particular year. The first, the 
St. John Passion of 1724, would necessarily be performed 
at St. Nicholas's. 

From 1766 onwards the Passions were transferred to 
the morning service; at a later date they were abolished 
altogether. 

Latin still prevailed in the Leipzig service in Bach's 
day, the Feast hymns being sung and the Epistle and 
Gospel read in that language. We cannot exactly deter- 
mine, however, the proportions of Latin and German in 
the service at that time. From 1702 onward, the council 
fought for a purely German service; at first it seems to 



* I. e. the cantata. 



Music in the Leipzig Service. 131 

liave little support, though it was able to effect a few 
changes at once*. 

There was no specific Leipzig hymn-book; the congrega- 
tion was supposed to know the hymns allotted to each 
Sunday. Any one who wanted to refer to them made use 
of the Dresden hymn-book**, which was what the cho- 
risters used. We know from the inventory of Bach's effects 
that he himself possessed the collection, in eight volumes, 
of Anddchtiger Seelen geistliches Brand- und Gantz-Opfer 
(Leipzig 1697), to which he must have turned when look- 
ing for good chorale verses for his cantatas. 

It was the ancient custom for the cantor to choose the 
sermon hymns. Bach had to decide upon the hymns for 
the whole of the Leipzig churches. The choice offered 
was not large; tradition had allotted certain hymns to 
each Gospel, and the hymn-books were arranged accord- 
ingly. In Leipzig the churches followed in this matter the 
Dresden hymn-book. 

Now it happened that in 1727 the afternoon preacher 
at St. Nicholas's, Magister Gaudlitz, desired to choose the 
hymns for his own sermons. He asked the permission of 
the Consistory and of the cantor to do this, and obtained 
it from both. This arrangement lasted a year. Then Bach 
took it into his head to ignore entirely the wishes of the 
preacher, and to select the hymns again from the Dresden 
hymn-book, Gaudlitz complained to the Consistory, who 
requested the cantor to keep to the previous arrange- 
ment. Bach, however, although the matter was one that 
concerned the church service alone, saw fit to call up 



* Thus the Te Deum, which at matins at St. Nicholas's was given 
out alternately by the organist and the choir, was replaced by the 
German version, "Heir Gott, dich loben wir.'* The wonderful 
harmonisation of this chorale verse by verse, (included in the Peters 
Edition among the chorale preludes, VI, 26), was therefore written 
by Bach for matins. This arrangement would also be used at the 
Reformation Feast, where it was customary to sing the Te Deum. 
It has come down to us in a copy made by Forkel. 
** First issued in 1694. 

9 



1 32 VIII. Bach in Leipzig. 

the Council against the Consistory and the poor Magister. 
He made a written report to the Council on the matter, 
and enlarged upon all the possible dangers of the innova- 
tion. How far this helped him again to "his rights" against 
the Consistory we cannot discover. The affair does not 
show Bach in a particularly favourable light. If he had 
wanted to guard his rights he should have done so at the 
beginning; if afterwards he desired to annul the agree- 
ment, he should have taken another course than the one 
he did*. 

In 1729 a serious conflict broke out between Bach and 
the Council**. In Easter of this year, some foundation 
scholars were received into the school whom Bach had 
found, on examination, to be quite unmusical; several who, 
he said, were particularly competent, had been rejected; 
while some of the candidates appear not to have presented 
themselves to him for examination at all. In the same 
year Bach assumed the direction of the Telemann Society. 
This brought about once more the old and desirable situa- 
tion which the Council had so long wished for, the stu- 
dents were again at the disposal of the cantor of St. Tho- 
mas's for his church performances. But the members of 
the Council, on their part, should now have restored to the 
cantor the old stipends for the students who were regularly 
employed in the choir and the orchestra. This they omitted 
to do; nor did they offer to fill up the foundation scholar- 
ships still unallotted. 

Naturally there was now no hope for good music in 
the churches for the year 1730. The poor condition of the 
choir must indeed have been noticeable in 1729, at the 
first performance of the St. Matthew Passion*, while at 
the celebration of the two-hundredth anniversary of the 
Augsburg Confession, on 25th, 26th, and 27th June 1730, 
it appears to have been unmistakeable. 

* On this matter see Spitta IT, 231 fl". 
** Foi fnll details see Spitta II, 239 ff, 



The First Conflict with ibe Council. 133 

The Council laid the blame on Bach; he laid it on the 
Council. When, in the summer of 1730, the Council was 
electing a new rector Ernesti haying died on the i6th 
October 1729, one of the members expressed the hope 
that they would fare better in this selection than they 
had done in that of the cantor. The reproach was general 
that Bach did not take enough trouble with the choir and 
with the singing lessons. 

There may have been some truth in this; Bach had 
really lost heart. He was no organiser; whatever he under- 
took to do, he did it with the impetuosity of genius. But 
if those around him were not carried away by his enthu- 
siasm, he was powerless. He knew nothing of the means 
by which a slower and more methodical mind would have 
gained its ends bit by bit. He could not even maintain 
discipline; such authority as he could count upon was 
merely that of genius and of the man in pursuit of an 
ideal. This, however, made no impression on the scholars. 
Of the pedagogue's authority, which alone could have kept 
them in order, he had nothing; and the passionate wrath 
by which he let himself be carried away, every now and 
then, made it still harder for him to get discipline. Dis- 
order was therefore rife in the choir. More than once he 
had to call in the aid of the rector to procure obedience. 
Ernesti was very well disposed towards him, and sup- 
ported him so far as his ill-health permitted; but his second 
successor, also named Ernesti in the end deserted 
Bach altogether. 

Other matters were discussed at the Council meeting on 
the 2nd August 1730, where the rancour against the cantor 
broke out*. Magister Petzold, who acted as his deputy in 
the teaching of Latin, had done his work badly. Bach, 
without the knowledge of the Council, had sent a scholar 
from the choir into the country probably to assist at 
some musical festivities; he himself had gone away with- 



See SpHta II, 243 ff. 



134 VIII. Bach in Leipzig 

out permission. "Not only does the cantor do nothing, 

but he will not give any explanation We must 

therefore stop it once for all" said one of the councillors. 
Another, the syndic Job, declared the cantor to be "incor- 
rigible". 

As a matter of fact they were not so angry with him for 
his errors or omissions in the performance of his duties, 
as for the disrespect he showed for the authority of the 
Council. He was, in fact, no cantor, but the Herr Kapell- 
meister of the courts of Cothen and Weissenfels, who had 
taken service with the Leipzig Council, but wanted to 
make his subordinate position something different from 
what it really was. The indifference he exhibited to- 
wards the Council in great things and in small provoked 
them to an attempt to break the pride of this man, who 
too often ignored, generally without reason the 
deference appropriate to his position. 

Bach, for his part, was in the right when he energetically 
repelled the reproach that he was answerable for the bad 
state of the church music. In a memorial dated 23rd Au- 
gust 1730 he details the constitution of the choir, and 
states that of the total number of alumni "seventeen are 
competent, twenty not yet fully capable, and seventeen 
incapable". Discretion forbids him to say anything as to 
the quality or the musical attainments of the eight town 
musicians. Nevertheless it should be considered that some 
are emeriti, and some in no such exercitio as they should 
be. For example, he always has to recruit the second 
violins, the violas, the violoncelli and the violon (i. e,, the 
contrabass) from among the scholars. Had the councillors 
really asked themselves how he was to manage on Feast 
days, when cantatas had to be produced simultaneously 
in the two churches? The new art demands much better 
performers than "the old style of music, which no longer 
sounds well in our ears". Therefore they ought to in- 
crease, rather than diminish, the money allotted to the 
students who assisted, Everywhere it is the custom to 



The First Conflict with the Council. 135 

underpay German artists, and to "abandon them to the 
cares of getting a livelihood", which cares prevent many 
of them from going on with their musical training*. The 
conclusion is "therefore simple"; if the subventions are 
to be continually withheld, he does not know "how he is 
to improve the music'*. 

The language of the memorial is that of an indictment. 
It is signed "Job. Seb. Bach, Director Musices", without 
any of the usual submissive formulae. It was something 
different from the memorials that the Council was accus- 
tomed to receive from its cantor. Bach's predecessor, in 
similar circumstances, used to sign himself "Your Magnifi- 
cence's, Right Noble's and Most Wise's dutiful and most 
obedient Johann Kuhnau, cantor at St. Thomas's school". 

While Bach was thus justifying himself to the Council, 
the latter had already taken action against him. It was 
at first proposed to remove him to a lower class in the 
school, where, instead of teaching Latin, he would have 
had to give elementary instruction, for which he would 
not have been allowed a deputy. On reconsideration, 
however, they appointed on their own account a capable 
deputy for the Latin lessons, whom Bach, of course had 
to pay as he had done the other. But it was resolved to 
reduce his emoluments so far as that was practicable. His 
salary and perquisites of course could not be touched. 
But there were legacies and endowments which the Council 



* The passage in the memorial runs: "It is moreover some- 
what strange that it is expected of German musicians that they 

should be capable of at once performing ex tempore every 

kind of music, whether Italian or French, English or Polish, as 
those virtuosi can before whom it may be placed and who have 
studied it a long time previously, and, indeed, almost learned it 
by heart, and wao quod notandum are very well paid, so that their 
trouble and industry are richly rewarded; but all this is not con- 
sidered, and these people are left to take care of themselves, so 
that many a one, in the need of finding a livelihood, cannot think 
of attaining proficiency, much less of distinguishing himself. To 
show this by one example, we need only go to Dresden, and see 
how the musicians are paid by the King " 



136 V1II. Bach in Leipzig. 

could distribute among the teachers as it thought fit. Little 
as he had had from these in the past, he was to receive 
nothing at all in the future. In 1730 there were two hundred 
and seventy thalers to distribute; the subrector received 
a hundred and thirty thalers, the third teacher a hundred, 
and Bach nothing. 

The bitterness of his soul may be seen from a letter he 
wrote on the 28th October of this year to Erdmann, his 
former fellow-student at Luneburg, begging him to help 
him to find a new appointment. This letter, one of the 
few by Bach that we possess, was packed with other old 
papers in an old chest after the death of its addressee, 
sealed, and sent to Moscow. There it wandered into the 
State archives, whence it was rescued by Spitta's friend 
0. von Riesemann, of Reval, who had been entrusted by 
Bach's biographer with the search for whatever might be 
left of Erdmann's papers*. 

The letter runs thus: 

4 'Honoured Sir, 

"Your Excellency will excuse an old and faithful servant for 
taking the liberty to trouble you with this letter. Nearly four years 
have now flown by since your Excellency honoured me with a gra- 
cious answer to the letter I sent you, but as I remember that you 
graciously wished me to give you some news of my vicissitudes, 
I shall now most obediently proceed to do so. From my youth up 
my history has been well known to you, until the change which toolc 
me as Kapellmeister to Cothen. There lived there a gracious Prince, 
who both loved and understood music, and with whom I thought 
to live the rest of my days. It so happened, however, that his Se- 
rene Highness married a Princess of Berenburg, and then it seemed 
as if the musical inclination of the said Prince had grown a little 
lukewarm, while at the same time the new Princess appeared to b 
an amusement to him; so God willed it that I should be called to 
this place as Director Musices and cantor at St. Thomas's school. 
At first it was not wholly agreeable to me to become a cantor after 
having been a Kapellmeister, on which account I delayed making 
a decision for a quarter of a year ; however, this post was described 
to me in such favourable terms that finally especially as my sons 
seemed inclined towards study, I ventured u pon it in the name of 

* Erdmann had become the Russian agent in Dantzig. Spitta 
gives further details of the lucky discovery in the preface to the 
first volume of his biography, p, V ff . 



The First Conflict with the Council. 137 

the Most High, and betook myself to Leipzig, passed my examina- 
tion, and then made the move. Here, by God's will, I am to this 
day. But now, since I find ( i ) that the appointment here is not nearly 
so considerable as I was led to understand, (2) that it has been de- 
prived of many perquisites, (3) that the town is very dear to live in, 
and (4) that the authorities are strange people, with little devotion 
to music, so that I have to endure almost constant vexation, envy, 
and persecution, I feel compelled to seek, with the Almighty's aid, 
my fortune elsewhere. Should your Excellency know of, or be able 
to find, a suitable appointment in your town for an old and faithful 
servant, I humbly beg you to give me your gracious recommenda- 
tion thereto ; on my part I will not fail, by using my best diligence, 
to give satisfaction and justify your kind recommendation and inter- 
cession. My position here is worth about 700 thalers, and when 
there are rather more funerals than usual the perquisites increase 
proportionately; but if the air is healthy the fees decrease, last 
year, for example, being more than 100 thalers below the average 
from funerals, In Thuringia I can make 400 thalers go further 
than twice as many here, on account of the excessive cost of living. 
And now I must tell you a little about my domestic circumstances. 
I am married for the second time, my first wife having died in Cothen. 
Of the first marriage, three sons and a daughter are still living, 
whom your Excellency saw in Weimar, as you may be graciously 
pleased to remember. Of the second marriage, one son and two 
daughters are living. My eldest son is Studiosus Juris, the other two 
are one in the first and the other in the second class, and the eldest 
daughter is still unmarried. The children of the other marriage 
are still little, the eldest, a boy, being six years old. They are one 
and all born musicians, and I can assure you that I can already 
form a concert, vocal and instrumental, with my family, especially 
as my wife sings a good soprano, and my eldest daughter joins in, 
quite well. I should almost overstep the bounds of politeness by 
troubling your Excellency with any more, so I hasten to conclude 
with all devoted respects, and remain your Excellency's life-long 
most obedient and humble servant, 

Joh. Seb. Bach,* 
Leipzig, 28th October 1730. 

For a letter to an old comrade of the Liineburg days, 
the epistle is, according to our notions, couched in too 
submissive a tone. Perhaps the two were no longer on 
terms of friendship Erdmann had been in Saxony in 
1725 without visiting Bach so that the letter was really 
one from the musician to the great gentleman, asking for 
his kind protection. This he probably found bitter enough, 



* Spitta II, 253, 254. 



138 VIII. Bach in Leipzig. 

Happily liis condition was not so bad as he imagined 
it to be. He was known to the new rector, Johann Matthias 
Gesner, who had been subrector of the Weimar Gym- 
nasium at the time when Bach was stationed at the Court 
there. Moreover he was warmly interested in music. As 
he was a capable teacher and an excellent organiser, he 
soon succeeded in bringing some degree of order into the 
school. The Council prized him very highly, and when 
he interested himself in Bach, the Council abandoned the 
measures it had taken against the cantor. In 1732 Bach 
was again admitted to a share in any money that was to 
be distributed. Gesner had previously had him definitely 
released from the necessity of teaching Latin. 

It has recently become the habit to paint the malignity 
of the Leipzig Council in the blackest colours; Bach is 
spoken of as being "degraded" by his superiors. The 
author of a recent biography goes to the length of saying 
that "so far as it was possible for the Council to paralyse 
the creative faculty of the genius, it honestly tried to do 
so"*. That is not the question. On the contrary, never 
did the Council spend more on music than at the very time 
when it was most at loggerheads with Bach**. The way 
in which the members of the Council tried to impress on 
him the inferiority of his position was of course unhand- 
some; but we must not forget that Bach had done What 
he could to irritate them. 

He did not regard the friendship of the rector as a suf- 
ficient guarantee against future Pleasures. In spite of his 
titles as Kapellmeister at Cothen and %t Weissenfels, for 
the' members of the Council and for the university he was 
merely the cantor. Of what avail was it for him to sign 
himself not cantor but Director Musices? Only one thing 
could win for him the respect of the Leipzig authorities, 



* Wolfram, Joh. Seb. Bach, Berlin, 1905. 

** On the expenditure at that time for church music see Spitta 
II, 252. Spitta does full justice to the Council. 



Bach seeks the Title of Court Composer. 139 

to become attached in some way to the Court of the so- 
vereign. He therefore conceived the plan of petitioning in 
Dresden for the title of Court composer. It was not the 
mania for a title that urged him to this, but the struggle 
for dignity. This is clear from the document in which he 
solicits the grant. As every one knows, the document 
took the form of a dedicatory epistle accompanying the 
parts of the Kyrie and Gloria of the B minor mass, which 
he presented to the young Electoral Prince*. It is dated 
27th July 1733, and runs thus: 

"In the deepest devotion I lay before your Kingly Majesty the 
accompanying trifling work [proof] of the science I have attained 
in music, with the very humble petition that you will graciously 
regard it not according to the poorness of the composition, but 
according to your world-renowned clemency, and deign to take me 
under your most powerful protection. For some years now, and up 
to the present time, I have had the direction of the music in the 
two chief churches in Leipzig; but I have innocently had to suffer 
at different times, from one vexatious cause and another, a diminu- 
tion of the fees attached to this office, which might be withheld 
altogether unless your Kingly Majesty would be gracious enough to 
confer on me a Praedicate of your Court Kapelle, and would issue 
a command in the proper quarter for the granting of a patent to 
that effect. Such a gracious granting of my humble petition will 
bind me in infinite veneration, and I offer myself in the most 
dutiful obedience, whenever your Kingly Majesty may graciously 
desire it, for the composition of church music or music for the 
orchestra, to show my indefatigable diligence, and to dedicate my 
whole powers to your services, remaining in constant fidelity your 
Kingly Majesty's most humble and obedient servant, 

Johann Sebastian Bach."** 

He had to wait three years for the desired distinction. 
Not that he was regarded in any but a favourable light; 
but the Prince had other cares. The troubles that had 
broken out in Poland demanded his presence in his other 
kingdom, where he remained from November 1734 to 
August 1736. After his return, Bach's well-wishers seem to 



* He appears to have gone to Dresden himself for the purpose. 
The dedication is addressed from there. 
** See Spitta III, 38. 



I^o VIII. Bach in Leipzig. 

have reminded him of the matter;* and Bach received 
his nomination as Court composer on igth November 
1736. It came just at the right time to help him in a new 
struggle with his superiors. 

In 1734 Gesner accepted a professorship at Gottingen. 
The Council itself was to blame for losing so soon this 
strong and capable man; it had refused him permission 
to accept, in addition to his rectorship, a professorship at 
the Leipzig university, though his immediate predecessor 
had been allowed to do so. 

Ernesti, the subrector, was appointed in his place. He 
tried to carry the reforms of Gesner still further, without, 
however, possessing the fine humanity of his predecessor. 
At first he and Bach got on excellently together; the rector 
twice acted as godfather to children of the cantor. In 
1736, however, Ernesti ordered the head prefect Gottlieb 
Theodor Krause, who, in a fit of anger, had punished 
some choristers for behaving unbecomingly at a wedding 
ceremony to be dismissed and degraded by a public 
castigation. Bach interceded for his prefect, and wanted 
to take the whole responsibility on himself, but in vain. To 
escape the punishment, Krause, who would shortly have 
gone to the university, absconded from the school. His 
place was given to the second prefect another Krause, 
Johann Gottlob. Bach did not think very highly of him. 
The question of making him a prefect had arisen a year 
before, and Bach had then told the rector that Krause 
was "a dissolute dog". Being in a good humour, how- 
ever, Ernesti and he were discussing the matter on 
their way home from a good wedding-feast he declared 
himself agreeable to Ernesti's nomination, nor had he any 
objection to make when Krause was promoted from fourth 
to third prefect, and later to second. He even accepted 
him for a time as first prefect, although he was angry with 

* On 27th September 1736 Bach had drawn up a second peti- 
tion, which lie seems to have delivered to the Prince during his 
visit to Leipzig. Spitta III, 8. 



The Conflict with the Rector. 141 

the rector for Ms severity towards the other Krause. Some 
weeks after, however, he removed him to the second prefect- 
ship and promoted the third prefect to the first place. He 
notified this to the rector, who made no objection. When 
Krause complained to Ernesti, the latter referred him to 
Bach, who allowed himself to be carried away into telling 
Krause that he had put him back into the second prefect- 
ship because the rector had advanced him to the first on 
his own authority, and he would now show the rector who 
was master there. Krause immediately reported this to 
Ernesti; and when the latter asked Bach for an explana- 
tion, he repeated it to his face. Thus, for a matter quite 
insignificant in itself, he imprudently raised the question 
of the right of appointment of the prefects, upon which the 
school laws threw no clear light. 

The rector demanded categorically the re-instatement of 
Krause as first prefect. Bach appears at first to have re- 
gretted his foolish impetuosity, and to have complied. 
But on one of the following Sundays, when Krause was 
preparing to conduct the motet, he turned him out in 
the middle of the hymn. At vespers the prefect again 
appeared in his place in obedience to the orders of the 
rector, who had at the same time forbidden the scholars 
to obey any prefect whatever appointed by Bach. The 
latter again turned Krause away. On the following Sunday, 
the igth August, the same scenes were enacted. The 
scholars did not know whether they were to obey the 
rector or the cantor. The second prefect, Kiittler, was 
sent away by Bach on the Sunday evening from the school 
meal, for having obeyed Ernesti instead of him. 

The offended master addressed himself to the Council, 
after having first tried, but without success, his old tactics 
of taking shelter behind the Consistory. This time he 
managed it badly, and the Consistory would not take his 
part. The memorials in which the cantor and the rector 
fought each other are preserved for us in the archives of 
the Council. The miserable business dragged on for two 



142 VIII. Bach in Leipzig. 

years. Bach's memorials show him to have been blinded 
by the fury of his wrath. Ernesti keeps cool and behaves 
as m asterof the situation; too adroit always to act honour- 
ably, he turns the mistakes of his antagonist to his own ad- 
vantage*. He does not stop at the vilest slander. It is 
not difficult to imagine what the state of discipline of the 
choir was all this time. 

The church authorities, even .those who had always 
been favourably disposed towards Bach, were angry with 
him for having tried to drag them into the dispute. Even 
his protector, Superintendent Deyling, was offended with 
him, and made him aware of it. 

The Council avoided taking energetic measures. Krause 
was to leave the school at Easter 1737, and the conflict 
would thus come to a natural end. For Bach, however, this 
did not at all settle the matter. He wanted to have the 
question of right decided whether the rector was entitled 
to interfere in matters concerning the prefects, and to 
force Ernesti to make public amends to him, so as to restore 
his authority among the scholars. As in the meantime 
he had become a Court composer, he made a direct appeal 
to the King, who immediately demanded a report. In 
February 1738 the affair was still unsettled; at Easter the 
King and Queen came to Leipzig, and Bach performed an 
Abend-Musik in the open air in honour of the royal couple. 
This is lost; but we learn from contemporary accounts 
that it made a very good impression. It would appear 
that the King then settled the dispute in Bach's favour, 
since from this time the records of the Council are silent 
upon it. Bach, however, had won nothing. Ernesti re- 
mained rector and created difficulties for him whenever 
he could. The other teachers sided with the head of the 
school. 

This affair of St. Thomas's was typical of what went 
on everywhere in the schools of that time. It was an 

* The documents relating to the affair are given in Spitta III, 
307 **. 



Bach's Position in the Leipzig Musical World. 143 

epoch in which the constitutions of schools were being 
reorganised. People were beginning to prosecute studies 
for study's sake, with the result that music was no longer 
allowed to occupy so much space and time in the work of 
the school. It was being squeezed out; the choirs of board- 
ers had outlived their time, like the old choirs of scholars 
in general. A new epoch was beginning. 

It was a misfortune that Bach's cantorate came during 
this time of transition. Henceforth the alumni of St. 
Thomas's fell into two categories those who were there 
for study, and those who were there to render the music. 
The cantor had a grudge against the first, the rector against 
the second. Ernesti was an enemy of music. When he 
met a student practising an instrument, he would ask 
him sarcastically: "Are you also going to be an ale-house 
fiddler?" Bach, on his part, abhorred the pupils who 
were bent merely on acquiring general knowledge and 
only casually took up music, as the pastor Joh, Friedr. 
Kohler, himself at one time a student at St. Thomas's, 
tells us in his history of the Leipzig school system*. 

The Krause affair had deprived Bach of his authority 
over both teachers and pupils. Henceforth he did his 
work at St. Thomas's as a stranger. Whether his rela- 
tions with Superintendent Deyling ever resumed their 
early cordiality we do not know. The Council embarrassed 
him no more. Once it even confirmed some expenditure 
that the cantor had made without previously asking the 
permission of the Council. This indicates a friendly dis- 
position towards him. The prompt interference of the 
monarch on behalf of his Court composer had instilled 
respect into his employers; and henceforth they avoided 
conflict. 

At the end of the thirties or the beginning of the forties, 
the date cannot be accurately determined Bach gave 



* Spitta III, ii, 12. The work exists only in manuscript in 
the Royal Library at Dresden: Historic* Scholarum Lipsiensiwn 
cottecta a Joh. Friedr, Kohlevo, pastor e Tauchemi 9 1776*!. 



VIII. Bach in Leipzig. 

up the direction of the Telemaim singing society, and so 
withdrew from public musical life. About that time, in 
1741, a concert society was founded, with a wealthy merch- 
ant named Zehmisch at its head, out of which the Gewaiid- 
haus concerts afterwards grew. Bach was not connected 
with it, at his own wish, not that of the founders of the 
society. He had no desire to come out of his retreat, feeling 
that the new generation and he no longer understood each 
other. 

Speaking generally, he does not seem to have been 
closely connected with any Leipzig circle. He did not belong 
to that of the poetess Marianne von Ziegler, who received 
many musicians at her house; at any rate he is never 
mentioned in her letters. He came into contact with Gott- 
sched in the autumn of 1727, when the latter was commis- 
sioned by Herr von Kiirchbach to write the text for the 
funeral ode in commemoration of Queen Christiane Eber- 
hardine. But later on, in 1736, when Frau Gottsched 
wished to study musical composition, he did not offer to 
teach her himself, but recommended his pupil Johann Lud- 
wig Krebs, who became sincerely enthusiastic over the 
talent and the charm of his pupil. 

Bach was very intimate with the librettist Christian 
Friedrich Henrici, who wrote under the pseudonym of 
Picander. He was a post-office official, who wrote satires 
and humorous verses in order to attract attention and to 
find a patron who would help him to a better appointment. 
Later on, indeed, he was promoted to a collectorship of 
land and drink taxes. Every one was astounded when, 
in 1724, he turned to religious poetry, and published a 
yearly cycle of cantata texts. He continued, however, 
witH the utmost unconcern, to print the most vulgar and 
unpleasant effusions. People wondered how Bach could 
feel attracted to so coarse and unsympathetic a man. 

He had many domestic troubles to bear in Leipzig. 
Of the thirteen children that Anna Magdalena bore to him, 
seven died; at his own death, of his twenty children only 



Bach's Children. 145 

nine were living five sons and four daughters*. The 
eldest son of Anna Magdalena, Gottfried Heinrich, was 
of weak intellect. At the division of the estate he was 
represented by a guardian. After the death of his father, 
his brother-in-law Altnikol took him with him to Nauru- 
burg, where he lived until 1763. Emmanuel was of opinion 
that he had some musical genius, which, however, re- 
mained undeveloped. Legends gathered at an early date 
about this son. Rochlitz, an admirer of Bach, tells us in 
his Filr Freunde der Tonkunst (1832)** that the master 
had a son named David, who, when he improvised at the 
piano in his own way, often moved the father to tears. 
No son of this name is mentioned in any of the genealogies 
that have come down to us. 

Bach's happiest time, in spite of all external discomforts, 
was that of the years just before and after the St. Matthew 
Passion, when he still had all his children round him. 
Friedemann and Emmanuel were already capable mu- 
sicians and a source of delight to him. It was at this time 
that he could hold the domestic concerts of which he speaks 
in the letter of 1730 to Erdmarm***. In the same letter 
he says that Friedemann is studying law. Emmanuel, 
after leaving school, took up the same pursuit. This does 

* Two sons, (Wilhelm Friedemann and Carl Philipp Emmanuel) 
and one daughter, (Catharina Dorothea), were of the first marriage ; 
the third son of Maria Barbara, named Bernhard, died at Jena in 
1738. Of Anna Magdalena's children there survived him: Gott- 
fried Heinrich (1724 1763), Elisabeth Juliane Friederike (born 
1726, year of death not known), Johann Christoph Friedrich 
(1732 1795), Johann Christian (1735 1782), Johanna Caroline 
(i737 1781), and Regine Susanna (1742 1809). 

The following children of Anna Magdalena all died at Leipzig: 
Christiane Sophie Henriette (born 1723, died 29th June 1726); 
Christian Gottlieb (born 1725, died 2ist September 1728); Ernestus 
Andreas (died ist November 1727, soon after birth); Regine Jo- 
hanna (born 1728, died 2 5th April 1733); Christiane Benedicta 
(died 4th January 1730, soon after birth); Christiane Dorothea 
(born 1731, died 3ist August 1732); Johann August Abraham 
(died 6th November 1733, shortly after birth). 

** IV, 278 ft 
*** See p. 136. 

Schweitzer, Bach. 1O 



146 VIII. Bach in Leipzig. 

not imply, however, that Bach intended his sons to be 
anything else but professional musicians. A university 
course of some kind was at that time part of the complete 
education of the artist. Other pupils of Bach also went 
in for an academic course of study, though firmly resolved 
to adopt music as their profession. In many musical posi- 
tions, in fact, an education of this kind was essential. 

Wilhelm Friedemann studied three years; then he ap- 
plied for the post of organist at St. Sophia's Church in 
Dresden, to which he was appointed in 1733, having come 
out the best at the examination performance on the 22nd 
June. He remained there thirteen years* In 1746 he 
went to St. Mary's Cfyurch at Halle, where Handel's old 
teacher, Zachau, had at one time been organist. The ap- 
pointment of his favorite son to this famous post must 
have been very gratifying to Bach. Even at that time 
Wilhelm Friedema^n shewed signs of the disorderly tem- 
perament that later on quite mastered him and brought 
Mm to adversity. The father could have had no presenti- 
ment of the tragic finish of this musical career that had 
so ideal a beginning, though in Ms last years he certainly 
had some anxiety about his son, who had begun to be 
addicted to drink, and quite neglected his art. Bach may 
have learned of the misuse his favorite son had made of 
one of his own Passions. Being commissioned to write 
some music for a university festival at Halle, at a fee of 
one hundred thalers, Friedemann, being too indolent to 
compose music of his own, simply adapted to his text the 
music of one of Ms father's Passions. The work was actu- 
ally performed, but the fraud was exposed by a cantor 
from the neighbourhood of Leipzig who chanced to be 
present. The indignation was so great that Friedemann 
was not paid the fee that had been agreed upon*. 



* On Bach's sons see C. H. Bitter's Cavl Philipp Emmanml und 
Wilhelm Ffiedemann Bach und deren Bvtider, Berlin, 1868. The 
story of Friedemann's plagiarism from his father is told by the 
famous musical theorist Marpurg (j/*8 1795), an admirer of 



Bach's Children. 147 

Emmanuel caused his father nothing but joy. He sent 
him to study at Frankfort- on- the-Oder, where the young 
artist founded a Collegium musicum among his fellow- 
students. In 1738, as he was preparing to accompany a 
young and distinguished Livonian on his travels, Frederick 
the Great, at that time Crown Prince, called him to him 
at Ruppin, and in 1740 appointed him his clavier accom- 
panist. In his autobiography, Bach's son boasts that "he 
had the honour of accompanying on the piano, quite alone, 
at Charlottenburg, the first flute solo that Frederick played 
after becoming King". Soon after his appointment he 
married the daughter of a Berlin wine dealer, and Johann 
Sebastian Bach became a grandfather. "My son in Berlin 
has already two male heirs; the first was born about the 
time when, unhappily, the Prussians invaded us; the other 
is about fourteen days old", he says in 1748 in the post- 
script to a letter to his cousin Elias Bach of Schwein- 
furt*. 

In the Kapelle of Frederick the Great there were many 
admirers and pupils of Sebastian Bach. The royal virtuoso 
greatly prized his accompanist, though there was no such 
dose relation between them as there was between himself 



Bach, in his Legenden einiger Musikheiligen (1786), pp. 60 63. 
He mentions no names, but relates the story in such a way that 
every one can guess whom he means. 

In 1764 Friedemann gave up his post in Halle, left his wife 
and Ms little daughter, and became a vagabond. For some time 
(1771 -1774) h stayed in Brunswick; later on he went to Berlin, 
where he procured some pupils. Mendelssohn's grandmother was 
taught by him. His ever-increasing asperity made it hard for people 
to interest themselves in him. One of his acquaintances wrote at 
a later time: "Friends of art and of the name of Bach have literally 
raised him more than once out of the very dust, lodged him de- 
cently, and provided him with the necessaries of life. They could 
never keep him in order, however, for any length of time. His wil- 
fnlness, his vulgar pride and his strong passion for drink caused 
him always to relapse into misery". He died ist July 1784, at the 
age of seventy-four. In the following year, Handel's Messiah was 
given in Berlin. The widow of Bach's favorite son received a grant 
from the receipts of the concert. See Bitter, p. 267. 

* The eldest grandson was born on the joth November 1745. 



14$ VIII. Bach in Leipzig. 

and Quantz. According to Zelter, Emmanuel was of too 
independent a nature as an artist to agree with all the 
musical opinions of the King, who \vou\d brook no con- 
tradiction in these matters. All the same he stayed 
twenty-seven years in the King's service. When Telemann, 
the musical director at the Johanneum in Hamburg, died 
in 1767, after having held the office for forty-six years, 
Emmanuel succeeded him. This event, of course, the father 
did not live to see*. 

J.G.Bernhard, the third son of Bach's first marriage, 
became organist at Miililhatisen in 1735, when scarcely 
twenty years old, not at St. Blasius's, however, where 
his father had once officiated, but at St. Mary's. In the 
spring of 1737 he became organist at St. Jacob's Church 
in Sangershausen. He quitted that town in a year, leaving 
behind him considerable debts, and died soon afterwards 
at Jena**. 

Johann Christoph Friedrich, born in 1732, a man of a 
quiet, pleasant disposition, became during the life-time of 
Ms father chamber musician to the Count von der Lippe 
at Biickeburg***. 



* In Hamburg, Emmanuel became acquainted with Klopstock 
and Reimams. Every artist passing through the town visited 
Ms house. He died in 1788. It was proposed to erect a memorial 
to him in St. Michael's Church, with an inscription "by the author 
of DM Messias. It got no further, however, than the inscription, 
which we still have. His youngest son, Sebastian, was a painter, 

to his father's horror, so Forkel tells us. He died at Rome, barely 

twenty-six years old, during the life-time of his father. 

** We still have the two letters that Bach wrote in recommen- 
dation of his son to Sangershausen; also the two he sent to Herr 
and Frau Klemm (with whom J. G Bernhard had lodged), in response 
to their request for a payment of his debts. In one of these we can 
see the deeply-stricken father; the other is short and business-like- 
See S. Schmidt, Vier aufgefundene OriginalMefe von J. S. Bach, 
in the Zeitechrift der IntemaHonaUn Musikgesellschaft III, 1901 
1902, pp. 35 iff. 

*** He remained in this post until his death, in 1795. His son, 
Wilhelm Friedrkh Ernst, born in 1759, later on became cembalist 



Bach's Children. 149 

The delight of Bach as he grew old was his youngest 
son, Johann Christian, whom he was able to teach up to 
his fifteenth year. His affection for him was so great 
that during his life-time he gave him three of his finest 
claviers, which so angered the sons of the first marriage 
that they tried to dispute the gift. 

Bach certainly could not have anticipated that the 
fame of this youngest born of his would at first completely 
overshadow his own. There is something fabulous about 
the life of Johann Christian; it has the charm of some 
bright romance. At the age of fifteen, after the division 
of the estate, he went to Berlin, to Emmanuel. He was 
soon seized, however, by a longing for Italy. He went 
there in 1754 the first of the Bachs who went to the 
home of art beyond the Alps for his musical culture and 
at once found rich patrons in Milan. To perfect his musical 
education he became a pupil of Padre Martini. For a 
time he lived in Naples, where he soon became one of the 
most celebrated of opera composers. After his conversion 
to the Roman Catholic faith, about the end of the fifties, 
he was appointed organist of Milan cathedral. The docu- 
ment relating to "Signor Giovanni Bacchi" is still in exis- 
tence. In 1762 he went to London, where he had been 
commissioned to write an opera. It was produced at the 
King's Theatre in the Haymarket in the presence of the 
whole court, with extraordinary success. The composer 
was appointed master of music to the queen, and soon 
became one of the favorite music teachers of the higher 
aristocracy; they used to charge half-a-guinea a lesson, 
and kept a horse and carriage in order to get to all their 
pupils in good time. In 1767 he married a London operatic 
star, Cecilia GrassL The Electoral Prince of the Palatinate 
sent for him to Mannheim for the production of an opera 
there. In 1779 he was living in Paris, where again he had 



to Queen Louise of Prussia and music teacher to her children. He 
was the sole direct descendant of the great cantor who was present 
at the unveiling of the Leipzig monument in 1843. He died in 1845. 



150 VIII. Bach in Leipzig. 

been commissioned to write an opera, for the manuscript 
of which he was to receive ten thousand francs*. 

Would the embittered cantor, when he taught his young- 
est son in the school house at Leipzig, have dared to 
cherish such lofty dreams for him? Did he ever imagine 
that life would pour into the lap of this child of fortune 
all the good things it had denied to himself? 

On 2oth January 1749, Elisabeth Juliane Friederike 
Bach (born 1726), became the wife of Altnikol, the devoted 
pupil who, a little while before, on the master's recommen- 
dation, had received the appointment of organist in Naum- 
burg. Bach rejoiced greatly at this marriage. 

Two little girls were playing in the room while he sat 
correcting and making dear copies of his fugues Jo- 
hanna Karoline, aged twelve, and the six-year-old Regine 
Susanna. The mother went to and fro and tended her 
husband, who took no care of his failing eyes, but obstinately 
made them serve him as long as there was a gleam of light. 
He did not know that he had only a few more months to 
live; the wife and the two children did not know what 
misery was in store for them. 



* The "London Bach" died ist January 1782. Opinions upon 
him were sharply divided from the beginning, Rochlitz regarded 
him merely as a musician who sacrificed his artistic ambition to 
the applause of fine ladies. His numberless works are, in truth, 
merely fashionable compositions. On the other hand we must not 
forget that the man with whom Mozart, during his residence in 
London in 1764 and 1765, studied for a year and a quarter, and 
of whom he thought highly all his life; was at any rate an artist, 
His melodic invention is not always banal. Whether he was really 
t&e man of pleasure he was depicted as being by many of his con- 
temporaries and by his first biographer Bitter, is an open question. 
Max Schwarz, in an exhaustive and singularly able study, has re- 
cently attempted a vindication of him artistically and morally. 
(Johann Christian Bach, in the Sammelbdnde der InUrnationalen 
Musikgesettschaft II, 1900 1901 ; pp. 401 454) It is to be wished 
that the other sons of Bach could find modern biographers of the 
same kind. The details given above are derived from this study. 



Bach's Amiability and Modesty. 151 

CHAPTER IX. 
APPEARANCE, NATURE, AND CHARACTER. 

In the conflicts that agitated his life and embittered 
his soul, Bach does not always appear in a sympathetic 
light. His irritability and his stubborn belief that he was 
always in the right can neither be excused nor glozed over. 
Least of all can we find excuse for the fact that at first he 
would be too easy-going, would always remember too 
late what he called his rights, and then, in his blind rage, 
would make a great affair out of what was merely a trifle. 

Such was Bach in his relations with people whom he 
suspected of a desire to encroach upon his freedom. The 
real Bach, however, was quite another being; all testi- 
monies agree that in ordinary intercourse he was the most 
amiable and modest of men. He was, above all, upright 
and incapable of any injustice. His impartiality was well 
known, It was particularly evident in the judgments 
he was so often asked to give upon organ matters. In 
these affairs he was dreaded for his strictness, for nothing 
escaped his sharp eye. Whether it was an examination 
of candidates for an organist's post, or the scrutiny of a 
newly-erected organ, in either case he was so conscien- 
tious and impartial that, as Forkel says, the number of 
his friends was seldom increased by it*. It even made 
enemies for him. When, for example, the young Scheibe 
was a candidate with Gorner and others, in 1729, for the 
post of organist at St. Thomas's, it availed him nothing 
that his father, the organ builder, was a friend of Bach. 
The latter declared for Gorner, with whom he had had so 
many tussles, and whose arrogant nature could not be sym- 
pathetic to him, Scheibe bore them both a life-long grudge 
for it. To his indignation we owe a highly interesting 



* Forkel, p. 22. 



152 IX. Appearance, Nature, and Character* 

criticism of Bacli in the Hamburg Critische Musikus of 1737. 
The master was hurt by it, but his relations with the elder 
Scheibe remained unaffected; he still expressed himself 
very favourably upon his work after the article, as he had 
done before. 

Bach was more than impartial: he was benevolent. If, 
when trying an organ, he found that the sum agreed on 
was too small in comparison with the good work that had 
been done, so that the builder would make only a little 
profit or perhaps an actual loss, he did not hesitate to 
recommend the congregation in question to make an 
additional payment, his suggestion being often adopted*. 
The organ examiners of our own day would do well to 
follow his example in this regard, and to impress upon 
congregations the necessity for paying prices for organs 
on which art and the builders can subsist, instead of 
accepting only the lowest tender. 

If Bach could do any one a service, he never refused. 
When his pupils were trying for a situation, he exerted 
himself most warmly on their behalf. In circumstances 
like these he did not mind writing to the church authorities 
in the most submissive terms**. 

Along with this kindness there went an agreeable 
modesty. The man who faced his superiors with a pride 
that must necessarily have offended them, let no one else 
feel his superiority. His modesty was not the hypocritical 
and conceited thing " in which celebrities often love to 
drape themselves in order to bulk still larger in the eyes 
of the world, but the sane and healthy modesty that 



* Forkel, p. 23. 

** See the four letters he wrote in 1726 to the Council at Plauen 
recommending as cantor Georg Gottfried Wagner, a former student 
of theology and philosophy at St Thomas's who had studied music 
with him. He describes him as "well versed in humanioribus and 
in musitis", "free and unmarried", and of quite "honest" way 
of life. We owe these letters to Wilhelm Fischer, who discovered 
them in the archives of the Council of Plauen. They will be found 
in the Neue Zeitsohrift fur Musik, 1901, pp. 484 and 485. 



His Relations with Other Artists. 153 

comes from the simple consciousness of one's own worth. 
He always preserved his dignity even when writing to 
kings, The petitions he addressed to his sovereign are 
couched in the submissive formulae of the epoch; but 
behind these formulae, prescribed by custom, a resolute 
pride is evident, We can read between the lines: "I, 
J. S. Bach, have the right to demand this of my prince 19 . 
The document he sent to Frederick the Great with the 
Musikalisches Opfer is in a somewhat different tone. He 
writes to him as his equal, in spite of the respect that he 
pays to his royal dignity. Divested of its fine courtliness, 
the letter runs: "Johann Sebastian Bach regards it as the 
greatest honour to add something to the fame of Frederick 
the Great by publishing a work based on a theme of his 
invention." 

He criticised the work of his pupils severely, but gave 
praise wherever he could; upon other musicians he never 
passed an opinion. Even his triumphs over others he was 
unwilling to hear discussed. Forkel tells us that he never 
mentioned voluntarily his musical contest with Marchand*. 
The full details of that victory are well known. Marchand, 
(1669 1732), court organist to the King at Versailles, 
and titular organist at several churches in Paris, had 
fallen into disfavour with his royal master in 1717, and 
had betaken himself to Germany**- At the Dresden 
court his elegant style of playing made so good an im- 
pression that the King promised him an appointment. 
The idea of measuring the Frenchman and Bach against 
each other in artistic rivalry came from Volumier, the 
leader of the Court band, According to Forkel, Bach 
was summoned expressly for this purpose to Dresden by 
a message from the King. It is more probable, however, 
that he had gone to the Court to hear the famous artist 
and to learn from him, and that, being there, it occurred 



* Forkel, p. 45. 
** Later on he returned to Paris. 



154 IX. Appearance, Nature, and Character. 

to his friends among the Dresden musicians to give Mar- 
chand whose overbearing and vain-glorious character 
must have made him unpopular a dangerous antagonist 
in the person of the simple Weimar conductor. Bach in- 
formed Marchand by letter that he was prepared to perform 
any musical task that he chose to set him, if he, on his 
side, would accept the same obligation. The whole com- 
pany took the liveliest interest in the contest, which was 
to take place in the house of the minister, Count Flem- 
ming. The invited audience, the referees, and Bach were 
there at the appointed time, but not Marchand. When 
they sent for him, they learned that he had left early in 
the morning in post haste. Bach accordingly had to play 
alone, which he did to the admiration of all. It is a curious 
fact that he received from the Court neither a gift nor an 
order for this affair. Forkel affirms that the King had 
intended him to receive a hundred louis d'or, but that 
they never reached him. They were probably intercepted 
by Court officials*. 

When asked how he had managed to bring his art to 
such perfection, Bach usually answered: "I have had to 
work hard; anyone who will work equally hard will be 
able to do as much"**. 

Even in his dealings with pretentious artists lie did not 
lose his amiability, and did not let it appear that he had 
seen through their vanity. One day, seemingly about 
1730, the Brunswick organist Heinrich Lorenz Hurle- 
busch visited him, not to hear Bach, but for Bach to hear 
Mm play the clavichord. Bach, says Forkel ***, received 



* Forkel, p, 8. The Necrology also says that Bach was cheated 
out of the money intended for him (p. 164). 

** Forkel, p. 45. 

*** Forkel, p. 46. Hurlebusch was organist at three churches 
in Brunswick at the same time, and had his son as assistant. Wal- 
ther's verdict on him in the Musikalische Lexikon (1732) is interest- 
ing: "He handles the organ very charmingly, excels in French 
suites, has an excellent judicium and ingenium* and is very popular 
in society and a paragon of politeness." Hurlebusch appeared 



His Relations with Other Artists. 155 

him in a friendly and courteous way, and listened with 
patience to his playing, which was quite ordinary. When 
Hurlebusch was leaving, he gave to Bach's eldest sons a 
printed collection of his sonatas, exhorting them to study 
them diligently (they who had already studied such very 
different things !) ; whereupon Bach only smiled, and did not 
behave any the less graciously to his visitor. Forkel 
dwells especially on Bach's modesty. The composer's sons, 
from whom Forkel got his information, took care that 
this trait in the character of their father was properly 
emphasised. They wanted to give a dementi to the wild 
stories that were current about him, as he himself had 
tried to do when living. Forkel expressly contradicts the 
legend that Bach would sometimes go into a chiirch dis- 
guised as a poor village schoolmaster and ask the organist 
to be allowed to play a chorale, merely to enjoy the aston- 
ishment his playing created among the company, or to 
hear the organist say it must either be Bach or the devil *. 
The friendly modesty of Bach's attitude towards all 
artists was a matter of common knowledge to his contem- 
poraries. We find a corroboration of it in a dedication 
addressed to him. Georg Andreas Sorge, "court and town 
organist to the Count of Reuss and Plau at Lobenstein", 
was impelled, although he was not Bach's pupil, to dedicate 
to the "prince of all clavichord arid organ players" some 
quite insignificant clavier pieces of his own; and in the 
dedication he commends him for the fact that "the great 
musical virtue that your Excellency possesses is embel- 
lished with the excellent virtue of affability and unfeigned 
love of your neighbour."** 



several times in Hamburg. He performed there on sth February 
1722, i Sth December 1727, and nth February 1728. See Sittard, 
Geschichte des Musik- und Konzertwesens in Hamburg, pp. 69, 70. 
* Forkel, pp. 45 and 46. The story of Bach and the village or- 
ganist is to be found in Marpurg's Legenden einiger Musihheilige'n 
(1786), pp. 98100). 

** Schletterer, Eine Widmung an Joh. Seb. Bach, in Eitnor's 
Monatshefte fur MusihgeschichU, 1879. 



156 IX. Appearance, Nature, and Character. 

His attitude towards Handel, indeed, shews how Bach 
admired whatever he thought great, without a touch of 
personal vanity. It was not Ms fault that he and his great 
contemporary never met. Handel came from England 
three times to visit his native town of Halle. The first 
time was in 1719, when Bach was living in Cothen, only 
four miles from Halle. Bach set out at once to visit the 
famous artist; but when he arrived Handel had just left. 
When the latter came a second time to his native town, 
in 1729, Bach was in Leipzig, but ill. He sent his eldest 
son Wilhehn Friedemann with a most courteous invitation 
to Handel to visit him in Leipzig. Handel regretted that 
he could not come. At the time of Handel's third stay in 
Halle, Bach was dead. He regretted all his life not having 
known Handel. His longing to meet him certainly did not 
come from the desire to pit himself against him. In Ger- 
many such a contest was indeed desired, for comparisons 
between the two men were always being made. It was 
universally admitted that Bach would be the victor on 
the organ. Bach's wish, however, was not to compete 
with him for pre-eminence, but to learn from him. How 
highly he valued him is seen from the fact that, assisted 
by Anna Magdalena, he made a manuscript copy of a 
Passion by Handel, which points to the fact that he also 
performed it. 

The copies that he made of other men's music are, on 
the whole, the finest testimony to his modesty. Long after 
the time when he could regard himself as any one's pupil, 
he made copies of Palestrina, Frescobaldi, Lotti, Caldara, 
Ludwig and Bernhard Bach, Telemann, Reiser, Grigny, 
Dieupart and many others. Sometimes we ask ourselves 
how it was that his critical sense did not stop him every 
now and then in his copying. It seems incomprehensible 
to us that he could bring himself to copy out whole can- 
tatas by Telemann. But these men were acknowledged 
masters: he respected them and was desirous of spreading 
their works. Which of the contemporary composers 



His Economy and HospitaHt\ T . 157 

troubled to make a copy of the St, Matthew Passion, with 
the view of preserving that work for posterity? 

Bach took thoroughly to heart the injunction to be al- 
ways hospitable. "Any lover of art, stranger or fellow- 
countryman", says Forkel, u could visit his house and be 
sure of meeting with a friendly reception. These sociable 
virtues, together with his, great artistic fame, caused his 
house to be rarely free from visitors"*. The members of 
the numerous and wide-branching Bach family who hap- 
pened to be in Leipzig for their studies were always heartily 
welcomed by him**. His cousin Johann Elias Bach, 
cantor at Schweinfurt, who in 1739 had stayed a long time 
in Leipzig, still remembered gratefully in 1748 the sociable 
friendliness shown him in the house adjoining St. Thomas's 
Church, and felt himself bound to send his famous relation 
a small cask of new wine. When it arrived it was two- 
thirds empty, and contained no more than six quarts. 
Bach tells the sender this on the and November 1748, 
gives him an account showing how much the present has 
cost him, and adds the request not to let his kindness put 
him to such expense again. The conclusion of the letter 
runs: "Although my worthy cousin is good enough to 
offer to send me some more of the same liquor, I must 
decline on account of the excessive expenses here; for the 
freight was 16 gr., the delivery 2 gr., the inspector 2 gr. 
the town excise 5 gr. 3 pf. and the general excise 3 gr., so 
that my good cousin can calculate for himself that it cost 
me nearly 5 gr. a measure, which is much too expensive 
for a present".*** 

This letter is at the same time a testimony to Bach's 
sense of economy in household affairs, that is so strongly 
noticeable in other things. He was very particular in 
money matters. During his struggle with Corner over the 
university church, he put the financial question in the 



* Forkel, p. 45. 
** Spitta III, 273. 
*** See the letter in Spitta 272, 273. 



158 IX. Appearance, Nature, and Character. 

forefront. In the letter to Erdmann he cannot help show- 
ing his indignation over the healthy year 1729, when the 
Leipzigers took so little pleasure in dying that the burial 
fees brought the cantor a hundred thalers less than usual. 
He tells his cousin EHas Bach, of Schweinfurt, who had 
asked him for a copy of the "Prussian Fugue", that it is 
out of print at the moment, but that he may inquire 
again in a few months, and remit the necessary thaler 
at the same time*. 

In all these cases there is indeed nothing more than a 
certain frankness in the treatment of money matters, 
which in the case of a man with so large a family is partly 
natural. That Bach was not avaricious is proved by the 
hospitality he dispensed. 

His business sense, however, seems not to have been un- 
known to his fellow-townsmen. Rector Ernesti the younger 
takes advantage of this in his fight with the cantor, and 
ventures to assert, in a document addressed to the Council, 
that Bach is not insusceptible to money when making re- 
commendation for admission 'among the alumni, and that 
many times an old specie thaler had made a soloist of one 
who was no soloist before **. The responsibility for this 
slander must be borne by the man who dared to utter it. 

The economical sense of the father came out very strongly 
in Emmanuel, so strongly as to throw a certain shadow 
over his artistic nature. As early as 1756 the way in which 
he announces that he is prepared to sell the plates of the 
Aft of Fugue at any decent price, makes an unpleasant 
impression on us***. When, in 1785, G. F. G. Schwencke, 
a pupil of Emmanuel Bach and of Kirnberger, was trying 
for the post of organist at St. Nicholas's in Hamburg, he 
was unsuccessful in spite of Ms splendid playing at the 



* See the letter in Spitta HI, 271, 272. 

** Ernesti's memorial of isth September 1736. See Spitta II, 
904 (German edition only). 

*** The announcement is given by Bitter, Cart Philipp Emmanuel 
Bach tend Wilhelm Friedemann Bach (1868), pp. 171 and 172. 



Emmanuel inherits his Father's Sense of Economy. 159 

examination; the appointment went to the son of the de- 
ceased organist, Lambo. Schwencke thus refers to the 
matter in a letter: "If Herr Lambo, who for the most part 
played miserably, worked his theme out well, it was pro- 
bably because he had previously studied it, and therefore 
probable that Bach had been bribed. He was avaricious 
enough for this."* Here again the responsibility for the 
calumny must lie with the man who uttered it. It is certain, 
however, that Emmanuel had the reputation of being avari- 
cious. One of his friends, Reichardt, writing on him, after 
his death, in the Mmikalmanach for 1796, said that" Even 
towards young artists who came to him full of the desire 
of learning, he was in the highest degree mercenary" **. 
A letter of his written in June 1777, while his son was 
hovering between life and death in Rome, is characteristic 
of him. "My poor son in Rome", he says, "has been down 
for five months with a very painful illness, and is not yet 
quite out of danger. O God, how my heart suffers! Three 
months ago I sent him fifty ducats, and in another fort- 
night I shall have to pay another two hundred thalers for 
doctors and surgeons."*** On the other hand this man, 
who in the midst of his anguish over his son stOl has an 
eye to his money, was as hospitable as his father had been. 
He had also inherited the family feeling. As we have seen, 
at the division of the property he was greatly irritated 
with the fifteen-year-old Jofhann Christian because the boy 
maintained that the father had made him a present of 
three pedal daviersf; nevertheless he afterwards took 
charge of him and brought him up. 

In one thing only did the family spirit forsake him: he 
did not take Ms stepmother in her hour of poverty, and 
allowed her, two years after the death of her husband 



* Joseph Sittard, GescUchte des Musik- und Komertmsens in 
Hambwg, 1890, p. 52. 

** Bitter, E. wnd Fr. Bach I, 173- 
*** Bitter, I, 346. 
f See Spitta III, 351 360. 



l6o IX. Appearance, Nature, and Character. 

(1752) to beg for alms from the Council that he had so 
proudly withstood, and finally let her die in receipt of poor 
relief, on 2yth February 1760. Even if he felt no special 
sympathy for her, and was himself not in flourishing cir- 
cumstances, he owed it to the honour of his father to save 
Magdalena Bach from want. Thus Bach's economical 
sense became meanness in his second son. Friedemann, 
the firstborn, inherited his father's obstinacy of spirit, 
and was ruined by it* 

In the portraits in which Bach's physiognomy has been 
preserved for us we can read a good deal about the nature 
and the bearing of the man. Until about twelve years 
ago, virtually only two original portraits of the master 
were known. One was in the possession of the firm of 
Peters, the musical publishers; it had been the property 
of Philipp Emmanuel, whose daughter sold it in 1828 to 
Greuter, a flute virtuoso and Conservatoire Inspector at 
Leipzig; the other belongs to the St. Thomas's school, to 
which it was presented in 1809 by August Eberhard Miiller, 
the successor of Hiller in the cantorate. To hang it up in 
one of the school-rooms was no doubt natural, but not the 
best thing for the picture, for Bach had to submit in effigie 
to the humours of the later Thomaners, and more than 
once served as target for missiles of all kinds. Both pic- 
tures are signed with the same name Hausmann 
which is a little astonishing, since they show notable 
differences in execution; both have suffered not a little 
from being painted over at a later date*. 



* On the question of the portraits, see Professor Wilhelm His's 
Anatomische Forschungen tiber Johann Sebastian Backs Gebeine 
und Antlite nebst Betnerkungen uber dessen BUder, Leipzig, 1895. 
He thinks it not impossible that the Peters portrait may be a free 
and not very capable copy of that in the Thomas school. Professor 
His discusses the non-authentic Bach portraits and the provenance 
of the various engravings. See also the brief and lucid essay of 
the respected librarian of the Peters Musikbibliothek, Dr. Emil 
Vogel, on the portraits of Bach (Jahrbuch der Musikbibliothek 
Peters, No. 3, 1897, P- 13 18.) 



Portraits of Bach. l6l 

The portrait in the St. Thomas school is perhaps the 
one that Bach had painted when he joined the Mizler 
Society, the statutes of which ordained that a new mem- 
ber must contribute "his portrait well painted on canvas" 
to the library of the society; 'for in this portrait Bach holds 
in his hand the "Canon triplex a 6 voc." which he sub- 
mitted to the society as his qualifying work. As Bach 
joined the society in the summer of 1747, this portrait 
would depict him in his latest years. 

A third authentic portrait of Bach used to be at Erfurt 
in possession of the organist Kittel, the last pupil of Bach ; 
it probably belonged at one time to the ducal family of 
Weissenfels. After Kittel's death, in 1809, it was, in ac- 
cordance with his wishes, hung up on the organ. During 
the Napoleonic wars, when the church was used as a hos- 
pital, it disappeared from the edifice with other valuable 
paintings. The French soldiers no doubt sold old Bach 
to the marine-store dealer for a few glasses of brandy. 

The well-known Bach portrait by C. F. Rr, Liszewski 
in the Joachimsthal Gymnasium in Berlin was not painted 
until 1772, twenty-two years after Bach's death. It is 
interesting because it is clearly not derived from either 
the Peters or the St. Thomas school portrait, and so pre- 
supposes another original. It shows Bach de face, sitting 
at a table with some music paper, as if about to run through, 
on the adjacent piano, some composition that he has just 
finished. 

An entertaining story of a Bach portrait is thus told by 
Zelter in a letter to Goethe: 

"Kirnberger had in his room a portrait of his master Sebastian 
Bach, that I have always admired, hanging over the piano, between 
two windows. A wealthy Leipzig linen merchant, who had of old 
seen Kirnberger, when he was a Thomaner, singing at his father's 
door, comes to Berlin, and resolves to honour the now famous Kirn- 
berger with a visit. Scarcely has the Leipziger sat down when he 
cries out: 'Eh 1 Good Lord ! I see you have our cantor Bach hanging 
there ; we have him also in Leipzig, in the St. Thomas school. He 
was a rough fellow; if the vain fool hasn't had himself painted in a 
splendid velvet coat I 1 Kirnberger quietly gets up, goes behind the 

Schweitzer, Bach. II 



1 62 IX, Appearance, Nature, and Character. 

man's chair, and, taking hold of his visitor with both hands, calls 
out, first softly, then crescendo, 'Out, dog! out, dog!' My Leip- 
ziger, in a mortal fright, runs for his hat and stick, opens the door 
as fast as he can, and bolts into the street. Kirnberger now has the 
picture taken down and cleaned, the chair of the Philistine washed, 
and the picture, covered with a cloth, again put back into its place. 
When anyone asked him what the cloth was for, he answered, 'Never 
mind! there's something behind it!' This was the origin of the 
report that Kirnberger had lost his reason*." 

It was always a matter of regret that, having neither a 
death-mask nor a skull of Bach, it was impossible to model 
a reasonably true bust of him. His grave was unknown. 
It was only known that he was buried in St. John's church- 
yard, and, as the sexton's receipt shows, in an oaken 
coffin. There was a tradition that the grave was on the 
south side of the church, six paces from the door. The 
churchyard had long been converted into a public place 
when, in 1894, after the dismantling of the old church, 
excavations for the extension of the foundations of the 
new church were begun at the spot where Bach's bones 
should be resting. Here there were discovered, on 22nd 
October 1894, three oaken coffins**. One contained the 
bones of a young woman, another a skeleton with the skull 
in pieces, the third the bones of an "elderly man, not very 
large, but well-built". The skull exhibited at the first 
glance the characteristic features that one would haVe ex- 



* The letter is dated 24th January 1829. BriefwechseZ zwischen 
Goethe und Zelter, ed, Reclarn III, 107. 

** Dr. G. Wustmann's' researches into the written and verbal 
traditions as to Bach's burial-place revealed the great importance 
of the information that he had been placed in an oaken coffin. In 
the year of his death, out of 1400 persons who were buried outside 
the enclosed burial ground, only twelve were interred in oaken 
coffins. Professor Wilhelm His acted as anatomical expert. See 
his report to the Leipzig Town Council, Johann Sebastian Bach: 
Forschungen uber dessen Grabstatte, Gebeine und Antlite, (F. C. W. 
Vbgel, Leipzig). An extract from this report is given in the Musi- 
kalische Wochenblatt (Jahrgang XXVI, 1895), PP- 339 anc l 34- 
See also the Allgemeine Musikzeitung for 1895, pp. 384 ff. The 
detailed results of the anatomical inquiry are given by His in the 
brochure mentioned on page 160. 



The Discovery of the Skeleton. 163 

pected from the pictures of Bach's head prominent 
lower jaw, high forehead, deep-set eye-sockets, and marked 
nasal angle. The identity of the skull with that of the 
cantor of St. Thomas's is thus as good as certain, more 
certain than in the case of the Schiller skull, for example. 

Among the interesting peculiarities of Bach's skull may 
be mentioned the extraordinary toughness of the bone of 
the temple that encloses the inner organ of hearing, and 
the quite remarkable largeness of the fenestra rotunda*. 
The plaster cast shows that the two upper flexures of the 
temple, in which the musical faculty has of late been sup- 
posed to be located, were not extraordinarily developed 
in him**. 

A Leipzig sculptor, Seffner, then tried to model the 
features upon a cast of the skull, after copious researches 
had been made into the relation of the fleshy parts of the 
face to the bony parts in elderly people, in order to settle 
the course of the line of the skin over the line of the 
bones ***. The bust thus obtained shewed not only a sur- 
prising similarity to both of the Bach portraits, but even 
surpassed them in vivacious and characteristic expression, 

Recently Professor Fritz Volbach, of Mainz, has dis- 
covered yet another portrait of Bach. It is a realistic 



* The bones in which the ear of Beethoven was embedded 
could not be compared with it, as they had been sawn out of the 
skull,, the intention being to preserve them in the Vienna Museum of 
Pathological Anatomy, whence, however, they have disappeared. 1 1 is 
supposed that one of the attendants sold them to an English doctor. 
** See the article of the Strassburg anatomist Prof. Schwalbe, 
alte und neue Phrtnologie (Korrespondenzblatt der deutschen 
Gesellschajt fur Antropohgie, Ethnologie und Uvgeschichte, XXXVII. 
Jahrgang, Nos. 9 11, 1906). According to this theory, Schubert 
must be counted among the unmusical musicians, since the two 
upper flexures of his temple are even smaller than those of Kant, 
who had so great a contempt for music. The capacity of Bach's 
skull was 1479.5 ccm. ; the height of the body, judging from the 
length of the bones, would be 166.8 cm, 

** * On the method of these measurements, and on earlier attempts 
of the same kind, see His's Anatomische Forschungen iiber J. S. 
Bachs Gebeine, usw., pp. 24 32. 

ii* 



1 64 IX * Appearance, Nature, and Character. 

piece of work, showing the face of a man who has tasted 
of the bitterness of life. There is something fascinating 
in the harsh expression of these features, which are painted 
full face. Round the tightly compressed lips run the hard 
lines of an inflexible obstinacy. It is thus that the cantor 
of St. Thomas's may have looked in his last years as 
he entered the school where some new vexation or another 
was awaiting him *. In the two other portraits the severity 
is softened by a touch of easy good nature. Even the 
short-sighted eyes look out upon the world from their 
half-closed lids with a certain friendliness, that is not even 
negated by the heavy eyebrows arched above them. The 
face cannot be called beautiful; the nose is too massive 
for that, and the underjaw too prominent. How sharply 
this projected may be estimated from the fact th?t the 
front teeth of the lower jaw are level with those of the 
upper, instead of closing within these. In the attempt 
to mitigate this peculiarity somewhat, the Hausmann por- 
traits cease to be characteristic. 

The longer we contemplate it, the more enigmatic be- 
comes the expression of the master's face. How did this 
ordinary visage become transformed into tl>at of the 
artist? What was it like when Bach was absorbed in the 
world of music? Was there reflected in it then the wonder- 
ful serenity that shines through his art? 

In the last resort the whole man is for the most part an 
enigma, for to our eyes the outer man differs so much from 
the inner that neither seems to have any part in the other. 
In the case of Bach, more than in that of any other genius, 
the man as he looked and behaved was only the opaque 
envelope destined to lodge the artistic soul within. In 
Beethoven, the inner man seizes upon the outer man, up- 
roots him from his normal life, agitates Mm and inflames 
him, until the inner light pierces through him and finally 
consumes him. Not so with Bach. His is rather a case 

* The discoverer of the painting thinks, with some reason, that 
this is the Erfurt picture. (In a letter to the present writer.) 




J. S. BACH 

From the portrait discovered by Dr. Fritz Volbach 



His Artistic Personality. I^c 

of dualism ; his artistic vicissitudes and creations go on side 
by side with the normal and almost commonplace tenor 
of his work-a-day existence, without mixing with or mak- 
ing any impression on this. 

Bach fought for his everyday life, but not for the recogni- 
tion of his art and of his works. In this respect he is very 
different from Beethoven and Wagner, and in general from 
what we understand by an "artist". 

The recognition that the world gave to the master of 
the organ and the clavier, really only the external and 
contemporary side of his artistic activity, he took as 
a matter of course. He did not ask the world for the re- 
cognition of that part of his work that was not of his own 
age, and in which his deepest emotions found expression. 
It did not even occur to him that he should or could ex- 
pect this from his epoch. He did nothing to make his 
cantatas and Passions known, and nothing to preserve 
them. It is not his fault if they have survived to our day. 

A modern student of Bach has said, d propos of some 
of the later chorale cantatas, in which the expert in Bach's 
scores notices a certain weakening of invention, that his 
whole work can be understood only as a mighty struggle 
for recognition, in which fight he was finally crippled*. 
Bach was certainly crippled at that time, not however in 
the struggle for recognition, but in the struggle for good 
cantata texts, in which he was finally thrown back again 
upon the chorale cantata, and, in a kind of fit of despera- 
tion, distorted chorale strophes into arias**. But this 
phenomenon has nothing to do with Bach's artistic life. 

The unique thing about him is precisely the fact that 
he made no effort to win recognition for his greatest works, 
and did not summon the world to make acquaintance 

* See Bernhard Friedrich Richter's interesting essay: Die Wahl 
J. 5. Backs zum Kantor der Thomasschule im Jahr 1723, in the 
Bachfahrbuch for 1905, pp. 48 67. See especially pp. 49 and 67 * 
"Bach wanted to compel the world to recognise him." 

** For a more detailed examination of the later chorale can- 
tatas, see the analysis of them in chapter XXXIV. 



1 66 IX. Appearance, Nature and Character. 

with them. Hence the kind of consecration that rests 
upon his works. We feel an unaffected charm in his can- 
tatas such as we do not meet with in other art-works. 
The grey volumes of the old Bachgesellschaft speak a 
moving language. They discourse to us of something that 
will be imperishable simply because it is big and true, 
something that was written not in the hope of recognition 
but because it had to come out of him. Bach's cantatas 
and Passions are not only children of the muse, but also 
children of leisure *, in the honourable and profound sense 
that this word had in the old days, when it signified the 
hours of a man's life that he employed for himself and 
himself alone. 

Bach himself was not conscious of the extraordinary 
greatness of his work. He was aware only of his admitted 
mastery of the organ and clavier and counterpoint. But 
he never dreamt that his works alone, not those of the men 
all round him, would remain visible to the coming genera- 
tions. If it is one of the signs of the great creative artist, 
born before his time, that he waits for "his day", and 
wears himself out in the waiting, then was Bach neither 
great nor born before his time. No one was less conscious 
than he that his work was ahead of his epoch. In this 
respect he stands, perhaps, highest among all creative 
artists; his immense strength functioned without self -con- 
sciousness, like the forces of nature; and for this reason 
it is as cosmic and copious as these. 

Nor did Bach reflect whether the Thomaners could per- 
form his works properly, or whether the congregation under- 
stood them. He had put all his devotion into them, and 
God at any rate certainly understood them. The S. D. G. 
(SoK Deo Gloria, "to God alone be praise") and the J. J. 
(Jem juva, "Help me, Jesus 1") with which he garnishes 
his scores, are for him no formulas, but the Credo that 



* The play upon words in the German cannot be reproduced 
in English, "sind nicht nur Kinder der Muse, sondern auch. Kinder 
der Musse". [Tr.] 



Bach's Religion. 167 

runs through all his work. Music is an act of worship 
with Bach. His artistic activity and his personality are 
both based on his piety. If he is to be understood from 
any standpoint at all, it is from this. For him, art was 
religion, and so had no concern with the world or with 
wordly success. It was an end in itself. Bach includes 
religion in the definition of art in general. All great art, 
even secular, is in itself religious in his eyes; for him the 
tones do not perish, but ascend to God like praise too deep 
for utterance. 

"Figured bass", he says in the rules and principles of 
accompaniment that he gave his pupils*, "is the most per- 
fect foundation of music. It is executed with both hands 
in such a manner that the left hand plays the notes that 
are written, while the right adds consonances and dis- 
sonances thereto, making an agreeable harmony for the 
glory of God and the justifiable gratification of the soul. 
Like all music, the figured bass should have no other end 
and aim than the glory of God and the recreation of the 
soul ; where this is not kept in mind there is no true music, 
but only an infernal clamour and ranting." 

The OrgelbuMein (Little Organ Book) the collection 
of small chorale preludes that Bach put together in Cothen, 
is adorned with the following dictum: 

Dem hochsten Gott allein zu Ehren, 
Dem Nachsten draus sich zu belehren. 

("For the glory of the most high God, and for the instruc- 
tion of my neighbour.") 

Lastly, musical education also belonged to the sphere 
of religion; and so Bach wrote in Friedemann's Klavier- 
buahlein (Little Clavier Book), over the first piano pieces 
he gave to his eldest son, "In Nomine Jesu". 

At the same time he recognised that there was a species 
of art whose only purpose was entertainment. He did 



* They are preserved in a copy dating from 1738. See Spitta 
HI, 3 i 7 ff. 



i68 IX. Appearance, Nature, and Character. 

not rank it highly, as we see from his somewhat satirical 
reference to the little songs of the Dresden opera when he 
asked Friedemann to accompany him there. All the same, 
when he was in the mood he could shake "little songs" 
that bordered on the burlesque out of his sleeve, as if he 
had to give himself up to heartfelt laughter now and then 
in order to get back again to proper seriousness. 

His culture was not merely serious but religious. In the 
inventory of the property he left we find a large number 
of theological works, among them a complete edition of 
Luther's writings, Tauler's sermons, and Arnd's Wahres 
Christentum. Polemical literature is well represented, and it 
enables us to see that Bach's views were strictly Lutheran. 
In Cothen he would not permit his children to attend the 
reformed school, but had them taught in the newly founded 
Lutheran school*. 

To pietism also he was sharply opposed**. When he was 
in Miihlhausen there was a struggle between an orthodox 
and a pietist divine there ; he took the part of the represen- 
tative of rigid Lutheranism, Georg Christian Eilmar, who 
appears to us in anything but a sympathetic light in his 
controversy with his older colleague Frohne. He must 
have had close personal relations with Eilmar, for he asked 
him to be godfather to his first child***. 



* See K. Bungc, /. S. Bach in Cothen, in the Bachfahrbuch for 
1905, p. 28. 

** Pietism derives from the Alsatian Philipp Jakob Spener 
(born in Rappoltsweiler in 1635), wno successively filled high clerical 
positions in Frankfort-on-the-Main, Dresden, and Berlin. New 
paths were opened by his Pia desideria, oder he-rzliches Verlangen 
nach gottgefdlliger Bessenmg dev wahren evangelischen Kivche (1675), 
in which, he insisted on subjective devotion and a thorough ab- 
sorption in the Bible. When Spener died, in 1705, Germany was in 
conflict on these points. Pietism did not, indeed, wish to attack the 
established dogmas of the church ; but through the significance that it 
attaches to personal conviction it did, as a matter of fact depreciate 
these. It was essentially a Reformation within the Reformation. 
The Protestantism of today is in part the product of pietism. 

*** Upon the religious troubles in Miihlhausen, and Bach's at- 
titude towards Eilmar, see Spitta I, 358 ff. 



Bach's Religion. 169 

As for the real points at issue in that epoch, Bach was 
as little conscious of these as his contemporaries were. 
Pietism was unsympathetic to him as a disintegrating 
innovation. He was dogmatically opposed to the repre- 
sentatives of orthodoxy. The submissive humility which 
the disciples of Spener affected was antipathetic to him, 
In addition, pietism was fundamentally inimical to art of 
any kind in worship, and was especially set against the 
concert style of the church music. The musical perform- 
ances of the Passion were a particular abomination to it; 
it wished the service to be adorned only with simple con- 
gregational hymns. So every cantor necessarily hated the 
pietists, and Bach took it particularly ill of them that 
they dragged his religious and artistic ideals in the dust. 
Nevertheless we possess no utterance of his, written or 
verbal, against the new sect. 

For all that, his own works exhibit visible traces of 
pietism; the texts of the cantatas and Passions are strongly 
influenced by it, as indeed the whole of the religious poetry 
of the early eighteenth century is. It can be seen in the 
reflections and the sentimental attitude with which Bach's 
librettists were so conversant. Thus the opponent of 
pietism invested with his music poetry filled with the 
breath of pietism, and so made it immortal. 

In the last resort, however, Bach's real religion was 
not orthodox Lutheranism, but mysticism. In his inner- 
most essence he belongs to the history of German mysti- 
cism. This robust man, who seems to be in the thick of 
life with his family and his work, and whose mouth seems 
to express something like comfortable joy in life, was in- 
wardly dead to the world. His whole thought was trans- 
figured by a wonderful, serene longing for death. Again 
and again, whenever the text affords the least pretext for 
it, he gives voice to this longing in his music; and no- 
where is his speech so moving as in the cantatas in which 
he discourses on the release from the body of this death. 
The Epiphany and certain bass cantatas are the revelation 



170 X. Artistic Journeys, Critics and Friends. 

of his most intimate religious feelings. Sometimes it is 
a sorrowful and weary longing that the music expresses; 
at others, a glad, serene desire, finding voice in one of 
those hilling cradle-songs that only he could write; then 
again a passionate, ecstatic longing, that calls death to 
it jubilantly, and goes forth in rapture to meet it. As we 
listen to arias like "ScKLummert ein ihr miiden Augen"*, 
"Ach schlage doch bald, sel'ge Stunde"**, or the simple 
melody "Komm, siisser Tod", we feel that we are in the 
presence of a musician who is not merely bent on render- 
ing into tone the thoughts of his text, but has seized upon 
the words and made them his own, breathing into them 
something of himself that was yearning for expression. 

This is Bach's religion as it appears in the cantatas. It 
transfigured his life. The existence that, considered from 
the outside, seems all conflict and struggle and bitterness, 
was in truth tranquil and -serene. 



CHAPTER X. 

ARTISTIC JOURNEYS, CRITICS AND FRIENDS. 

Bach loved travelling. As a young man it attracted 
him because of his desire to learn from every master. 
Later, when he himself was a master, he felt the necessity 
from time to time of getting away from the narrowness of 
his environment, and realising himself in freedom else- 
where. It appears that he undertook these artistic jour- 
neys early, as a rule in the autumn; though unfortunately 
our information about his travels is very scanty. 
. Bach's first master, the Prince of Weimar, appears to 
have been generous in the matter of furlough. In 1709 
Bach, with his friend Walther, opened the organ at Miihl- 
hausen, performing for that purpose the chorale prelude 

* From the cantata Ich hdbe genug (No. 82). 
** From the cantata Christus, der ist main Leben (No. 95). 



Early Journeys. 171 

upon "Bin feste Burg" (Peters VI, 22), if indeed he did 
not compose it expressly for this occasion in order to show 
the Council the full wealth of tone of the renovated instrument. 

Of the artistic journeys of the following years we know 
nothing. In 1713 or 1714 he performed before the Court 
at Cassel, at the invitation of the Hereditary Prince, the 
future King of Sweden. When the Prince heard Bach 
play a pedal solo, he was so enraptured at the dexterity 
of it that he drew a ring from his finger and placed it on 
that of Bach. The anecdote is told by Constantin Beller- 
mann, Rector of Minden, in a pamphlet issued in 1743- 
He does not give the date of the journey, and thinks that 
at that time Bach was already in Leipzig. As a matter of 
fact he did go on one occasion from Leipzig to Cassel to 
examine a renovated organ; but this was about 1732. By 
that time the Hereditary Prince was King of Sweden, and 
we know positively that he was not in Cassel. The journey 
during which Bach played before the Hereditary Prince 
must be dated about 1713 or 1714. In those years the 
Prince spent some time at home, while before then he was 
almost constantly in camp, being commander in the Span- 
ish War of Succession. At the end of 1714 he went to 
Sweden, where he married, in 1715, Ulrike Elenore, the 
sister of Charles XII*. 

To the same period belongs a journey in which Bach 
touched upon Halle. By the death of Zachau (i4th August 
1712), the post of organist at the Liebfrauenkirche there 
had become vacant. It was left unfilled for a time, a new 
organ, with thirty-six stops, being in course of construc- 
tion. After Bach had tested so much of the instrument 
as was already playable, he presented himself to the church 
authorities and said he would have no objection to becom- 
ing Zachau's successor. He was asked to write at once a 

* Bellermann thus confuses the two journeys to Cassel; Spitta 
does the same (Spitta I, 515, 634, 635). The matter was cleared 
up by Scherer, in hisjoh, Seb. Bachs Aufentbalt in Kassel, in Eitner's 
Monatshtfte fur MusikgescUchU, 1893. 



172 X. Artistic Journeys, Critics and Friends. 

cantata as a test piece, which he did. When he returned 
to Weimar, he made closer enquiries as to the emoluments 
of the post, and finding that they were lower than what 
he already had he broke off the negociations at the last 
moment. The Halle people, who thought they had al- 
ready secured him, were very displeased. They took it 
ill of him that he had kept them in suspense in this way, 
and maintained that he had only begun negociations with 
them in order to extort an increase in salary from the 
Duke. Bach did not allow this to pass; he refuted the 
imputations in a courteous but firm letter of I4th March 
1714. Later on the Halle people were reconciled to him, 
and in 1716 they invited him, with Kuhnau and Rolle of 
QuedHnburg, to the testing of the now completed organ. 
We still possess a letter in which Bach answered this in- 
vitation. It is addressed to his friend Herr August Becker, 
Lizentiat Juris, who had acted also as negociator between 
him and the Council in the affair of the appointment*. 
It runs thus: 

"Most Noble 

and Particularly Highly Honoured Sir, 

I am greatly obliged by your honour's very particular and most 
gracious confidence, and by that of the whole of the most honoured 
Collegium; and as it is always the greatest pleasure to me to give 
your worship mv most willing service, I shall now more than ever 
strive to serve your worship well, and according to my best ability 
give satisfaction in the examine you ask of me. I beg you therefore 
to communicate this my resolution without delay to the most hon- 
oured Collegium, and at the same time to give them my most humble 
compliments and assure them of my dutiful respects for their special 
confidence. 

I also acknowledge with humble gratitude all the trouble your 
honour has taken for me both now and formerly, and I assure you 
that I shall always feel the greatest pleasure, as long as I live, in 
subscribing myself, most honoured Sir, 

Your most obedient servant 
Joh. Seb. Bach 

Concerbneister. " 



* On the Halle affair see Spitta I, sisff., and M. Seiffert's 
/. S. Bach 1716 in Halle, in the Sammelbdnde der Intemationalen 
Musikgesellschaft, VI, 1905. Zachau had been Handel's teacher. 



Early Journeys. 173 

The opening of the organ took place on the 3rd May. 
In conjunction with it the Council gave the "deputies" 
a dinner, the menu of which we can reconstruct from 
the receipts. It included: "Baffallamote" (i. e. t&uf d la 
mode), pike, gammon of bacon, peas, potatoes, spinach 
with little sausages, boiled pumpkins, fritters, preserved 
lemon-peel, preserved cherries, warm asparagus salad, 
cabbage salad, radishes, fresh butter and roast veal. The 
cost of the whole was eleven thalers twelve groschen; the 
drink came to fifteen thalers twelve groschen. Servants 
were also placed at the disposal of the "deputies"*. 

Bach seems to have visited Leipzig for the first time 
in 1714. On the inner cover of the cantate Nun komm 
d&r Heiden Heiland (No. 61), which certainly belongs to 
the year 1714, he has noted down the "Order of the morn- 
ing service in Leipzig for the first Sunday in Advent". 
The most natural assumption is that he performed this 
work in Leipzig on the first Sunday in Advent in 1714, 
also officiating at the organ during the service. We must 
not forget, however, that this is a pure hypothesis, for it 
is also possible that Bach wrote this cantata in 1714 for 
Weimar, but did not produce it until later, perhaps on the 
first Sunday in Advent 1722, when applying in person for 
the cantorate of St, Thomas's **. 

In the autumn of 1717 he went to Dresden, where the 
Marchand incident happened; he also appears to have been 
about this time in Meiningen, where his distant and con- 
siderably older cousin Johann Ludwig Bach was Kapell- 
meister. Bach must have thought very highly of his com- 
positions, for he made copies of many of them***. 

We know positively that Bach was in Leipzig on the 



* See Seiffert's article already referred to. 

* * This explanation is thought the more probable one by Bern- 
hard Friedrich Richter, in his interesting essay Die Wahl Job. Seb. 
Backs zum Kantov der Thomasschule im JaJw 1723, in the Bach- 
jahrbuch for 1905, p. 48 ff. 

*** An appreciation of Johann Ludwig Bach's compositions will 
be found in Spitta I, 574 ff. 



X. Artistic Journeys, Critics, and Friends. 

i6th December 1717, on which day he had been invited 
by the University to test the new organ in St. Paul's 
church. His verdict was extremely laudatory; and from 
that time Scheibe, the builder of the instrument, had a 
reputation everywhere as one of the best organ builders, 
whereas previously he was hardly known. 

When Bach undertook this journey, he had not long 
been released from his Weimar captivity, and had just 
settled in Cothen. In his new position he had more opport- 
unities for travelling than in the old one, since it was part 
of Ms duties to accompany Duke Leopold everywhere. 
These duties he would probably not be unwilling to fulfil. 
He spent, for instance, part of the summer of 1720 with 
his master in Karlsbad. It was on his return from this 
journey that he found in his home, instead of his beloved 
wife, only his motherless children. In the autumn of the 
same year he undertook the journey to Hamburg that 
gained him the admiration of Reiiiken and the art-lovers 
there. 

Even when he was cantor at St. Thomas's he made a 
point of escaping every year from the narrow circle of 
Leipzig and its artistic conditions. According to agree- 
ment, he should have applied for leave to the Burgomaster 
each time. This, however, he frequently did not do, con- 
tenting himself with providing a reliable deputy and 
telling the rector. If he had not a capable first prefect 
at the time, the cantata would be conducted by the or- 
ganist of the New Church, who, as soon as it was over, 
would get back to his own organ in time to fulfil his duties 
there*. 

So long as Duke Leopold was alive he died the igth 
November 1728 Bach often went to Cothen, to produce 
one or other of his compositions on festival occasions. 
At the entombment of his former master, in 1729, he con- 

* Ernesti mentioned this in one of his memorials to the council 
on the subject of the Krause case, See Spitta II, 904 (German 
ed.), and English ed. Ill, 230. 



Journeys during the Leipzig Period. 175 

tributed a large piece of funerary music for two choirs, 
which he appears to have taken from the St. Matthew 
Passion, upon which he was then engaged. 

As shortly before his Leipzig appointment he received 
the praedicate of Weissenfels Court composer, he also had 
duties towards this Court, that from time to time called 
for his presence there. 

He went to Dresden frequently. He appears to have 
been summoned to the Court there shortly before 1725 for 
some occasion or other *; he also went there now and then 
to hear the opera. He was accompanied on these occa- 
sions by his favorite Friedemann. Some days previously 
he would say to him "Friedemann, shall we go again to 
hear- the pretty little Dresden songs?"** He was present 
at the first performance, on I3th September 1731, of 
Hasse's opera Cleofide, in which the composer's wife Faustina 
appeared. On the following day he performed on the organ 
in St. Sophia's Church before the whole Kapelle and 
many connoisseurs. The Dresden musicians thought very 
highly of him. Hasse and his wife were much attached 
to him, and more than once visited him in Leipzig***. 

Bach probably visited Dresden still more frequently 
after Friedemann' s appointment as organist there in 1733. 
Moreover, his appointment as Court composer made it his 
duty to keep in tbuch with the musical life of Dresden. 
He received the decree in the last days of November 1736; 
on the first December he performed from two till four 
o'clock pn the new Silbermann organ in the Frauenkirche 
at Dresden f. 

While the reigning Duke Wilhelni Ernst was still alive 
he did not return to Weimar, as we can readily under- 
stand. When, however, in 1728, Ernest August, the youn- 
ger Duke, succeeded to power, he certainly went there 



* Spitta III, 223. 
** Forkel, p. 48. 
*** Forkel, p. 48. 
f Spitta III, 226. 



176 X. Artistic Journeys, Critics and Friends. 

now and then, for he was on terms of cordial friendship 
with this artistic and intelligent prince. 

In 1727 he performed for the second time in Hamburg, 
perhaps at the invitation of Telemann, who thought a very 
great deal of him.. About the same time he seems to have 
visited Erfurt. 

Probably most of his journeys were for the purpose of 
testing new organs. Interesting information upon a trip 
of this nature is given in a document in the municipal 
Board of Works in Cassel. Bach had been invited there 
in 1732 to try the renovated organ at St. Martin's, which 
had been in the repairers' hands for two years. For this 
he received fifty thalers remuneration and twenty-six 
thalers for travelling expenses. In addition the Council 
paid the expenses of the Herr Kapellmeister and his wife 
at their lodgings, two thalers for the "porters" who carried 
Herr Bach twenty-five paces, and a thaler for the man- 
servant who waited on him during the eight days of his 
sojourn*. These things were evidently done more splen- 
didly in those days than they are now. 

At the end of July and the* beginning of August 1736, 
during the conflict with Ernesti over the appointment of 
the first prefect, Bach was absent from Leipzig for a fort- 
night. We do not know where he went. His last journey 
may be described in Forkel's words, who narrates it ap- 
proximately in the terms in which he had it from Friede- 
mann: 

"The fame of Joliann Sebastian's all-surpassing art was so wide- 
spread that even the King very often heard it spoken of and extolled. 
He was consequently desirous to hear for himself so great an artist, 
and make his acquaintance. At first he quietly hinted to Bach's 
son his desire that Ms father should some day come to Potsdam; 
then he began to ask definitely why the father did not come. The 
son could not but communicate these expressions of the King to 
his father, who, however, was mostly overwhelmed with too many 
occupations to attend to them at once. But as the King's remarks 
were repeated by the son in several letters, he at length, in 1747, 

* Scherer, /. 5. Backs Aufenthalt in Cassel, in Eitner's Monats- 
hefte fur Musikgeschichte, 1893. 



Journeys during the Leipzig Period. 177 

arranged to make the journey, accompanied by his eldest son, 
Wilhelm Friedemann. At this time the King had a chamber con- 
cert every evening, at which he himself played some concerto on 
the flute. One evening, just as he had got his flute ready, and his 
musicians were assembled, an officer brought him the list of stran- 
gers who had arrived. He looked over the paper with his flute in 
his hand, immediately turned round to the musicians, and said with 
some agitation, "Gentlemen, old Bach has come!" The flute was 
thereupon laid aside, and old Bach, who had gone to his son's quar- 
ters, was at once commanded to come to the castle. I had the story 
from Wilhelm Friedemann, who accompanied his father, and I 
must say that to this very day I recall with delight the way in which 
he told it to me. In those days it was still customary to make rather 
long-winded compliments. The first appearance of Johann Sebastian 
Bach before so great a king, who did not even allow him time to 
exchange his travelling clothes for the black coat of the cantor, 
must thus necessarily have led to many excuses being made. I 
will not here give specimens of these excuses, but merely remark 
that Wilhelm Friedemann gave them in the style of formal dialogue 
between the King and his self-excusing father. 

But more interesting than all this is the fact that for this evening 
the King gave up his flute concerto, and invited the so-called "old 
Bach* 1 to try his Silbermann forte-pianos, of which he had several 
in different rooms of the castle. The musicians accompanied them 
from room to room, and Bach had to try all the pianos and im- 
provise upon them. After he had done this for some time, he asked 
the King to give him a fugue subject, for him to work out impromptu. 
The King was astonished at the erudite way in which his theme 
was developed extempore, and, apparently in order to see how far 
such an art could be carried, expressed a wish to hear also a fugue 
in six parts. As, however, it is not every theme that is suited for 
this kind of polyphony, Bach chose one himself, and developed it 
immediately, to the great admiration of all present, in the same 
brilliant and learned way as he had formerly done the theme of the 
King. Then the King wanted to hear him on the organ. He ac- 
cordingly took him on the following days to aU the organs t o be 
found in Potsdam, as he had taken Him previously to all the Silber- 
mann forte-pianos. After his return to Leipzig he worked out in 
three and six parts the theme the King had given him, added various 
clever canonic manipulations of it, had it engraved on copper with 
the title of Musikalisches Opf&r (Musical Offering) and dedicated 
it to its inventor. This was Bach's last journey."* 

Bach's artistic journey's made him, at an early date, 
famous throughout Germany. After his victory over 
Marchand in 1717 he was one of the celebrities of .the 
fatherland. The German musicians were proud to be able 

* Forkel, pp. 9, 10. 

Schweitzer, Bach. 12 



178 X. Artistic Journeys, Critics and Friends. 

to oppose a master of their own race to the French and 
Italian virtuosi. Let German musicians affect^ Italian ways 
if they would, in order to win a cheap renown*, let the very 
existence of a German art be denied; the fact remained 
that there was such a thing, and that it had publicly tri- 
umphed over the other. 

Bach thus had no need to fight for recognition. Only 
the virtuoso, it is true, won fame; the composer of the 
cantatas and Passions had small share in this recognition. 
No one, not even his enemies, ventured to deny that he was 
the prince of clavichord players and the king of organists ; 
but no one, even among his best friends., had a suspicion 
of the real greatness of the composer. 

As a composer he was actually censured by the two 
leading critics of the time, Mattheson and Scheibe. Mat- 
theson** examines the cantata Ich "hatte- viel Bekummernis 
(No. 21), which Bach probably had performed during his 
visit to Hamburg in 1720, and finds the declamation im- 
perfect, in that Bach at the commencement makes the 
chorus repeat three times the detached words "Ich, ich, 
ich", before it gives out the whole phrase, while in other 
places he has split up the phrases in a way that contra- 
dicts the sense. So far as the threefold repetition of the 
"I" is concerned the criticism is not wholly unjustified. 
It is curious that the great Hamburg art-oracle should 
have taken just this cantata, which is not impeccable in 
its declamation, as the basis -for his verdict on the most 
perfect master of musical declamation 1 Had he taken the 
trouble to acquaint himself with other vocal compositions 
of Bach he would have been undeceived, and would cer- 
tainly have enlisted him as an ally in his justifiable war 
against the careless musical declamation of the epoch. 



* Kuhnau's Der musikalische Quacksalbe* (1700), shows us the 
feeling of the German musicans towards foreigners and those who 
assumed foreign manners.. The work has recently been issued in 
a new edition (Behr, Berlin 1900). 
** Critica musica, II, 368. 



Mattheson and Bach. 179 

In truth, however, Mattheson had no particular desire 
to become acquainted with Bach's works. He had hailed 
him in 1717, in his Das beschiitzte Or Chester, as a rising 
star*, and had asked him for an autobiographical con- 
tribution to the Ehrenp forte, which he was then planning; 
Bach did not send it to him, having no liking for things 
of that kind. During his sojourn in Hamburg, Bach pro- 
bably did not think it necessary to court the protection 
of the famous critic, which would hardly increase Mat- 
theson's good will towards him, for he could not endure 
independent characters, and only felt sympathetic to- 
wards people who sought and glorified his authority. So 
on the few occasions on which he mentions Bach, his tone 
is one of recognition but at the same time of indifference, 

Scheibe's criticism of Bach is much more interesting. 
It is true that he also was not unbiassed, for he bore Bach 
a grudge for having rejected him when he applied for the 
organist's post at St. Thomas's in 1729. His criticism ap- 
peared in the Kritische Musikus for 1737, a journal that 
Scheibe edited from 1737 to 1740 in Hamburg, where he 
had taken up his abode. An anonymous "Friend", who 
is instructed by him in the art of criticising according to the 
true rules of reason, sends him in a travel-letter a miscel- 
lany of good and bad upon the musicians of several towns 
that he has just visited. The people concerned are indeed 
not mentioned by name, but they are so clearly indicated 
that they are easily recognised. The passage in which 
Bach recognised himself runs thus: 

"Herr ... is the most eminent musician in ... He is a wonderful 
artist on the clavichord and the organ, and so far he has met with 
only one with whom he can contend for superiority. I have several 
times heard this great man play. His dexterity is astonishing, 
and one can hardly comprehend how it is possible for him to cross 
his fingers and his feet so remarkably and so nimbly, or to make 



* "I have seen some pieces by the celebrated Weimar organist, 
Herr Johann Sebastian Bach, both for the church and instrumental, 
that are certainly written in such a way as to make us esteem the 
man highly." See Spitta II, 21, etc. 

i a* 



i8o X. Artistic Journeys, Critics and Friends. 

the widest leaps with them, without once striking a wrong note or 
distorting his body, no matter how quick his movements are. 

"This great man would be the admiration of all nations if he 
had more amenity, and if his works were not made unnatural by 
their turgid and confused character, and their beauty obscured by 
too much art. As he judges by his own fingers, his pieces are ex- 
tremely hard to play; he expec'ts the singers and the instrumenta- 
lists to do with their throats and their instruments what he can play 
on the clavier. This, however, is impossible. All graces, all little 
embellishments, and everything that one understands by style in 
playing, he writes out in the exact notes, which not only deprives his 
pieces of the beauty of harmony, but makes the melody absolutely 
indistinct. All the parts work together and with the same diffi- 
culty, so that we cannot distinguish any leading voice. In short, 
he is in music what formerly Herr von Lohenstein was in poetry. 
Turgidity has led them both from the natural to the artificial, and 
from the lofty to the obscure; and in each of them one wonders at 
the painful labour of it all, that nevertheless comes to nothing, 
since it is at variance with reason/** 

It would be a mistake to see in this criticism only an 
outburst of the personal resentment that Scheibe har- 
boured against Bach. Scheibe regarded himself and 
to a certain extent with justice as the literary cham- 
pion of a new music. He makes war on the contemporary 
art which, by its imitation of the Italian style and its 
deliberate artificiality, has departed from the true ideal, 
and has no longer any poetic value. As a pupil of Gott- 
sched, he believes himself called to preach the return to 
simple nature in music, to declare war on the aria and all 
erudite formulae, to deride perpetually the unnatural- 
ness of the Italian opera and especially its German imita- 
tion, and to uphold as the ideal that form of art in which 
the words are not a mere pretext for the music, but word 
and tone unite in perfect unity. The opera of the future 
is for him a real musical drama. Certain chapters in which 
he discusses stage music are really excellent; much of what 
he says might have been expressed in the same way, word 



* Dey kritische Musikus, No. 6; Tuesday, I4th May 1737, pp. 4$ 
and 47. Scheibe himself was, of course, the author of the * 'Letter". 
In March 1738 the first year's numbers of the journal appeared in 
volume form with an index. The second volume, extending from 
March 1739 to March 1740, appeared in 1740. 



Scheibe's Criticism. l8l 

for word, by Wagner. It is therefore not to be wondered 
at that Scheibe's views exercised so deep an influence on 
Gluck*. 

It was precisely these theories of his, indeed, that made 
it impossible for him to do justice to Bach. The latter's 
astounding contrapuntal techniqtie made him seem to 
Scheibe the chief representative of artificial music. Per- 
haps, however, the critic of the Gottsched circle bore him 
the worst grudge for showing no interest whatever in the 
aspirations towards the art of the future, and was not 
in the least particular as to the poetic quality of his texts. 
Scheibe wanted his criticism to strike the most eminent 
of the class of musicians who persisted in being musicians 
and nothing more. Of the fact that this learned music 
had within it a poetry of its own, Scheibe had as little 
intuition as the sons and the friends of the master. 

Bach was incensed to the highest degree by this criti- 
cism, and asked his friend Magister Birnbaum, a professor 
of rhetoric at the Leipzig University, to take up the pen 
for him. The latter gladly di^ so, and first of all published 
in January 1738 an anonymous article, which, however, 
was not very effective. Scheibe easily countered it by 
shewing up the dilettante quality of the musical judgments 
of the anonymous writer, whose name was not unknown 
to him**. From this reply we learn that the "Herr 

* See especially pp. 177 208 of the first volume, and pp. i ff. of 
the second. On Scheibe see the interesting article of Eugen Reichel, 
"Gottsched und Johann Adolf Scheibe", in the Sammelbande der 
Iniernatwnalen Musikgesellschaft, 1900 1901, pp. 654 668. 

Johann Adolf Scheibe was born in Leipzig in 1708. He went to 
Hamburg in 1735; from there he went to Kulmbach as Kapell- 
meister in 1740; from 1744 he conducted the Court orchestra at 
Copenhagen. He lost this post in 1749* ai *d thereafter held no per- 
manent appointment. After a life of erratic wandering he died in 
Copenhagen in 1776. 

** Scheibe's reply is to be found in the Kritische Musikus of Tues- 
day, the i8th February 1738 (I, 203 ff.)> and in a supplement to 
this number: Beantwoytung der unparteiischen Anwerkwigen itber 
eine bedenkliche Sidle in dem sechsten Hauptstuck des kritischw 
Musikus. Ausgejertigt von Johann Adolf Scheibe. Hainburg, 1738 



1 82 X. Artistic Journeys, Critics and Friends. 

Hofcompositeur" himself distributed the vindication of 
himself to his friends and acquaintances, with no small 
gratification, on the 8th January of this year. 

One unpleasant feature of the controversy is the way 
in which Scheibe always adverts to the fact that Bach is 
not in a position himself to take up the pen against him, 
and how, at every opportunity, he represents him as a 
musician who lacks the necessary general education. With 
exquisite humour he sketches an imaginary letter of Bach's, 
in which the master shows the philosophers and quill- 
drivers to the door of the temple of art. Scheibe probably 
inwove into the letter some authentic sayings of Bach that 
had remained in his memory from an earlier time, as 
when he makes him say : "I have always been of the opinion 
that a musician has enough to do simply with his art, 
without wasting his time on long-winded books and learned 
philosophical discussions."* 

Against the reproach that the master was lacking in the 
general artistic culture that is indispensable to musicians, 
Birnbaum vindicated the offended cantor in an article of 
March 1738, signed with his name. From this we learn 
something of Bach the aesthetician : "Bach knows so per- 
fectly", the Magister says, "the analogies between the 
working-out of a musical piece and the art of rhetoric, 
that people not only listen to him with satisfaction and 
delight when he expounds lucidly the resemblances and 
correspondences of the two, but admire also the skilful 
application of them in his works." To this Scheibe has 
nothing that is very rational to say**. 



* Kritischer Musikus of 2nd April 1739, II, 34 36. Bach's 
style seems to be not badly imitated in the letter. 

** Birnbaum's article is entitled: M. Johann Abraham Birn- 
baums Verfeidigung seiner unparteiischen Anmerkungen, uber eine 
bedenkliche Sulle in dcm sechsten Stuck des kritischen Musikus, 
wider Johann Adolf Scheibens Beantwortung derselben. ("Magister 
Johann Abraham Birnbaum's vindication of his impartial remarks 
upon a questionable passage in the sixth number of the Kritischer 
Musikus, against Johann Adolf Scheibe' s reply to the same.") 



Scheibe's Criticism. 183 

All the same, Bach's "prickly" critic the adjective 
often occurs in the controversy regards him as one of 
the greatest composers of pure music. In the number for 
22nd December 1739 Scheibe is lavish in his praise of the 
Italian Concerto*. As a cantata composer, to be sure, 
he ranks him below Telemann and Gratm**, 

On the whole it may be said that Scheibe's criticism did 
harm to the author himself, but brought good to the object 
of his attack, since its offensive tone everywhere stimulated 
sympathy for Bach. Later on Scheibe also appears to have 
recognised that he had not gone about the affair in the 
right way. In the preface to the second edition of the 
Kritischer Musikus we can see the glimmerings of some- 
thing like an apology. 

In spite of its incivility, Scheibe's criticism is really the 
most interesting of contemporary criticism of Bach. The 
other deliverances upon him run on general lines of ad- 
miration and amazement and rhetorical analogies from 
ancient mythology. But we learn nothing from them of 
what we should most like to know, how the characteristic 
quality of Bach's art affected his contemporaries. We 
would gladly exchange all these panegyrics for a single 
sentence of some one who, at the first performance of the 
Stf. Matthew Passion, had an intuition of the real spirit 
of Bach's music. 

Bach was probably most pleased with the monument 
that the friendship and affection of his former Rector, 
Gesner, erected to him under the cover of a Latin note to 



Scheibe replied in the number of $oth June 1739 (II, pp. 141 to 144). 
In the second edition, of the Kritischer Musikus (1745), he printed 
also both Birnbaum's articles. The above citation follows the first 
edition. On the whole affair see Spitta III, 252 ff. 

* Kritischer Musikus II, 342, where the clavier pieces alone are 
discussed: "One of the best of published pieces is a clavier concerto 
of the famous Bachj it is in the key of F, and written in the best 
style of this genre. This clavier concerto is to be regarded as a 
perfect model of a well-worked-out concerto for one instrument." 
** In the supplement to an article by Mattheson. See Spitta 
III, 252. 



184 X. Artistic Journeys, Critics and Friends. 

the Institutions variae of Quintilian, an edition of which 
he edited in 1738. At the end of a passage referring to 
the artist who, while singing, accompanied himself on the 
cithara and beat time with his feet, he adds this note : 

"All this, my Fabius, you would think quite trifling if you could 
rise from the dead and see Bach whom I mention because not 
long ago he was my colleague at St. Thomas's school in Leipzig; 
how with both hands, and using aH his fingers, he plays the clavier* 
that contains in itself the tones of many citharae, or that instru- 
ment of instruments, whose innumerable pipes are animated by 
bellows ; how he flies over the keys this way with both hands, and 
that way with his nimble feet, and, unaided, calls forth a plurality 
of quite different passages that yet harmonise with each other; 
could you, I say, see how, while he achieves what a number of your 
citharists and a thousand of your flute-players together could not 
achieve, he not only sings one melody, like a man who has nothing 
more to do but sing to the cithara, but attends to everything at 
once, and keeps thirty or forty musicians in order, one by a nod, 
another by stamping time with his foot, and a third with a warning 
finger, and joins in with his own voice now in a high part, now in 
a lower one, and again in a middle one; and how he alone, when 
they are all working together at their loudest, although he has 
the hardest task of all yet at once notices when and where some- 
thing is wrong, and keeps them all together, and watches every- 
thing, and if there is any hesitation restores certainty; how rhythm 
is in every limb of him ; how his quick ear grasps every harmony, 
and how he himself reproduces each voice within the small compass 
of his own. In general I am a great admirer of antiquity, but I 
believe . that my friend Bach, and whoever may be at all like him, 
comprises in himself many men like Orpheus and twenty singers 
like Arion*/' 

The note prohably achieved the purpose its author had 
in view, the mollifying of the cantor who had been so 
nettled by Scheibe's sarcasms. If, by the way, the picture 



* Quintilian, Institutiones oratoriae ad I, 12, 3: "Haec omnia, 
Fabi, paucissima esse diceres, si videre ab inferis excitato contin- 
geret Bachium, ut hoc potissimum utar, quod meus non ita pridem 
in Thomano Lipsiensi collega fuit: manu utraque et digitis om- 
nibus tractantem vel polycordum nostrum multas unum citharas 
complexum, vel organum illud organomm, cujus infinitae numero 

tibiae follibus animantur Maximus alioquin antiquitatis fautor, 

multos unum Orpheos et viginti Arionas complexum Bachium 
rneum, et si quis illi similis sit forte arbitror." A German trans- 
lation is given by Spitta II, 89, 90 (German edition). The passage 
is cited by J. A. Hiller in his biography of Bach (1784). 



Eulogies of Bach. 185 

is not a purely rhetorical one, it points to the fact that 
Bach conducted his cantatas from the organ. 

Bach was also celebrated in verse, although this honour 
fell to him less frequently than it afterwards did to his 
two eldest sons. He possessed at Hamburg a poetical ac- 
quaintance, Herr Friedrich Hudemann, Doctor of Law, 
whom he had distinguished in 1727 by dedicating a very 
learned canon to him. Hudemann, in return, addressed 
the following verses to Bach in a collection of poems, 
Proben einiger Gedichte, which he published at Hamburg 
in 1732: 

"Wenn vor gar langer Zeit des Orpheus Harfen-Klang 

Wie er die Menschen traf, sich auch in Tiere drang, 

So muB es, grofier Bach, weit schoner dir gelingen: 

Es kann nur deine Kunst verniinftge Seelen zwingen. 

Und dieses trifft gewiB mit der Erfahrung ein: 

Oft sieht man Sterbliche den Tieren ahnlich sein, 

Wenn ihr zu bidder Geist nicht dein Verdienst erreichet, 

Und in der Urteilskraft dem dummen Viehe gleichet, 

Kaum treibst du deinen Schall an mein geschaftig Ohr, 

So tonet, wie mich deucht, der ganze Musenchor. 

Ein Orgelgriff von dir muB selbst der Keid bescharnen, 

Und jedem Lasterer die Schlangenzunge lahmen. 

Apollo hat dich langst des Lorbeers wert geschatzt, 

Und deines Namens Ruhm in Marmor eingeatzt. 

Du aber kannst allein durch die beseelten Saiten 

Dir die Unsterblichkeit, vollkommner Bach, bereiten." * 

Contemporary works of musical biography tell us hardly 
anything about Bach. Since in spite of two requests he 
could not make up his mind to send Mattheson an auto- 
biographical contribution for the Ehrenpforte, the latter 
was offended and passed him over in it without mention**. 

* Quoted by Spitta II, 478 (German edition). 
** The Ehren-pforte was to contain only autobiographical notices. 
Handel, who had also left Mattheson's enquiry unanswered, was 
nevertheless dealt with in the work in an article by the compiler 
himself. This honour probably fell to him because he had onc 
been Mattheson's "pupil". Bach had not had that good fortune. 
The Ehrenp forte appeared in 1740. 



1 86 X. Artistic Journeys, Critics and Friends. 

Walther, in his musical lexicon of 1732, enumerates 
only dates in Bach's career, and the work that had appeared 
in print, as he did in the case of every contemporary 
musician. He cannot refrain from characterising the clavier 
pieces i. e. the six suites of the first part of the Klavier- 
iibung as excellent. At the end he observes that even 
the letters BACH are melodic in their sequence a 
remark that owed its origin to the Leipzig Herr Bach*. 

The enumeration of the names of all who felt friend- 
ship and respect for him would make a fairly long list. 
The members of the Dresden Kapelle, and those of the 
Berlin Kapelle, to which his son belonged, looked upon 
him as one of themselves. Of the musicians, the two most 
famous of that time, Hasse and Telemann, were most 
cordially attached to him; all the Bachs honoured him as 
the head of the great old family; his pupils were devoted 
to him, and lost no opportunity of showing him their devo- 
tion, and of demonstrating their justifiable pride in their 
teacher; the leading members of society in Dresden set up 
for his patrons; the cultivated Livonian Freiherr von 
Kayserling, who from 1733 to 1745 was Russian am- 
bassador at the Dresden Court, befriended him whenever 
he could**; princes such as Duke Leopold of Cothen, 
Duke Ernst August of Weimar, and Duke Christian of 
Weissenfels, treated him as a friend. 

When all is said, however, all these were only good ac- 
quaintances. Bach does not appear to have had a real 
friend so closely bound up with him as to have a part in 
his deepest thoughts and experiences. His intimates were 
his wife and his two eldest sons; to others he did not reveal 
the whole friendliness and joviality of his nature. He 



* Walther's Musikalisches Lexihon, Leipzig, 1732. The article 
on Bach commences on p. 64 and comprises 40 half-lines. The 
whole volume contains 659 closely printed pages. 

** It was probably he whom Bach had to thank for his appoint- 
ment as Court composer. It was through him that he received the 
patent of the ipth November 1736. 



Friends and Acquaintances. 187 

kept them all at a certain distance from him. For this 
reason we know nothing of Bach's real inner nature. No 
one has been able to hand down to us a remark in which 
he revealed anything of his inmost soul. Not even his 
sons could tell Forkel anything of this kind. 



CHAPTER XL 
THE ARTIST AND TEACHER. 

In his criticism of Bach, Scheibe in one place goes so 
far as to say that the Herr Hofcompositeur is lacking in 
the general culture that we ought to expect from a great 
composer*. What truth is there in this reproach? 

As regards culture, Bach was not the inferior of any of 
the musicians of his epoch. The Latin schools at Ohrdruf 
and Liineburg which he had attended enjoyed a first-rate 
reputation; and we now know that he went through the 
classes from the beginning. That he did not afterwards go 
to the university was entirely due to the prime necessity of 
earning his daily bread. He was well versed in Latin, or 
he would not have been able to offer to give the instruc- 
tion in this language that it devolved upon him to give in 



* In the Beantwortung der un-parteiischen Anmerkungen fiber 
eine bedenkliche Stette in dem sechsten Stuck des kritischen Mu- 
sikus, Scheibe's reply to Birnbaum's first article, which ap- 
peared as a supplement to the first year's volume of the Kritischev 
Musikus, in 1738. On page 22 he says: "the prime cause of this 
defect (i. e. Bach's) deserves further consideration. This great man 
is not particularly well up in the sciences that are especially required 
of a learned composer. How can one be quite without blemishes in 
his musical work who has not, by knowledge of the world, qualified 
himself to investigate and understand the forces of nature and 
reason? How can one achieve all the benefits that come from the 
acquisition of good taste, who has barely concerned himself with 
critical observations, enquiries, and rules that are so necessary not 
only in rhetoric and poetry but in music that without them one 
cannot possibly be moving and expressive, principally because the 
attributes of good and bad style in writing, both in general and in 
particular, proceed almost entirely from these." 



l88 XI. The Artist and Teacher. 

St. Thomas's school. The foreign words in his letters are 
always used in their correct sense. This shews that he 
also knew French, which is again clear from the dedication 
of the Brandenburg concertos, which is very elegantly 
expressed in that tongue. The addresses of his letters 
were usually written in French. He signed his name some- 
times in German, sometimes in French, and sometimes in 
Italian, He was conversant with rhetoric as it was then 
taught, as Birnbaum expressly testifies in the second of 
his vindications. The fact, again, that men like Gesner 
and Birnbaum found his conversation interesting, speaks 
for his culture; so does the thorough, scholarly education 
that he gave his sons. 

Unfortunately Emmanuel and Friedemann have made 
it impossible for us to know much about the reading of 
their father, since before the division of the property they 
placed on one side Ms collection of old mathematical and 
musical historical works, which therefore do not figure in 
the inventory. However, the list of Bach's theological 
books is alone enough to testify to the activity of his mind. 
It is interesting to know that he had a translation of Jo- 
sephus's History of the Jews. We can imagine him, after 
the day's work, seeking recreation in the classical work of 
the favorite and friend of Vespasian ! 

Scheibe was thus mistaken. And yet, looked at from 
his own standpoint, a good deal of his reproach is not un- 
founded, only that he expressed it clumsily. Bach was 
self-taught, and as such had an aversion to all learned 
theories. Clavier-playing, organ-playing, harmony, com- 
position, he had learnt them all by himself; his sole 
teachers had been untiring work and incessant experiment. 

To a man who had made the fundamental rules of art 
his own in this manner, many theories that were interesting 
or new for others were a matter of indifference, for he had 
been to the roots of things. Now Bach lived in the 
epoch when it was thought that the perfect art could be 
discovered by aesthetic reasoning, while others, again, 



Bach's General Culture. 189 

thought that salvation for music lay in mathematical 
speculations upon the numbers that underlie intervals. 
To all these endeavours Bach opposed a robust indifference. 
"Our Bach", says the Necrology, "did not engage at all in 
deep theoretical considerations of music, but was all the 
stronger in the practice of it."* 

Perhaps he let this indifference become too visible. It 
is certain that it was generally known how little he troubled 
about the mathematical basis of the fundamental laws of 
harmony. Not only Mattheson** but Scheibe testifies to 
this, "Let any one", says the latter, "ask a great Bach, 
who has perfect command of all artifices of art, and whose 
astounding works one cannot see or hear without sur- 
prise, whether, in the attainment of this great skill and 
dexterity, he even once thought of the mathematical rela- 
tions of the tones, and whether he once consulted mathe- 
matics in the construction of so many musical artifices."*** 

"Two fifths and two octaves must not follow each other, 
since that is not only a mtium, but it sounds ill;" so ran 
the third rule of thorough-bass that Bach dictated to his 
pupilsf. "It is not only a mtium but it sounds ill;" 
we imagine we can see him as he goes up and down listen- 
ing to the scratching of the -quill, and laughing merrily to 
himself ! 

He took so little interest in the activities of the scientific 
musical society that was constituted at Leipzig under his 
eyes, that at first he had no thought of becoming a member. 
The fact is all the more striking as Lorenz Christoph Mizler 
(1711 1778) the founder of the society, had been his pupil 



* Mizler's Musikalischt Bibliothek, 1754; vol. IV, Part I, p. 173* 
** Ehrenpforte, p. 231, note, referring to the fact that Mizler 
claimed to be a pupil of Bach. 

*** Kritischer Musikus, 1739, II, 355. Scheibe is here praising 
Bach. 

t See Bach's rules for thorough-bass in the copy that has come 
to us from Peter Kellner (Spitta III, 315 ff.)- Peter KeUner does 
not seem to have been a direct pupil of the master. See the preface 
to B. G. XXVII 1 . 



1 90 XI. The Artist and Teacher. 

for the clavier and composition, and had dedicated to him 
in 1734, together with three other celebrities, his doctoral 
dissertation "Quod nrasica ars sit pars eruditionis philo- 
sophicae" "Music is part of a philosophical education"*. 
The "Sozietat der musikalischen Wissenschaften" was 
founded by him in 1738. He was of opinion that as a result 
of its inquiries a "new period in music" was going to open; 
and he had a medal struck, on which, according to his 
own description, "a naked child flies towards the dawn, 
on its head a bright star, in its right hand a burning torch 
inverted, beside which a swallow flies, denoting the dawn- 
ing of day in music*'**. 

There was a good deal that was fantastic in the activities 
of this Society. One cannot, however, but respect it 
when one looks through Mizler's journal, the Musikalische 
BMiothek, the organ of the Society. It testifies to a solid 
and learned spirit, and contains plenty of interesting matter. 
The reader was kept admirably informed upon everything 
that "happened in the sphere of music and of the adjoin- 
ing sciences" ***. 

Telemann belonged to the Society from 1740; Handel 
was made an honorary member in 1745; Bach, however, 
in spite of all the importunities of Mizler, could not decide 
to apply for admission. He finally did so in 1747, sub- 
mitting as his specimen composition some canonic varia- 
tions upon the Christmas hymn "Vom Himmel hoch da 
kornm ich her"f, and the "Canon triplex a 6 voc.", which 
afterwards appeared in the Musikalische Bibliothek-ft. 

* In 1736 Mizler gave lectures on mathematics, philosophy and 
music at the Leipzig University, In 1743 he went to Warsaw, 
where he later was made a Court counsellor. His journal, the Musika- 
lische Bibtiothek, appeared from 1736 1754. 

** MMer's Musikcdische Bibliothek, Vol. IV, Part I. Leipzig, 
*754 P- 105, 106. 

*** See, for example, in Vol. IV, Part I (p. 48 68) the thorough 
anatomical description of the ear for musicians. The articles on 
general aesthetics are of less value. 

f Peter's ed, of the organ chorale preludes, V, 92 102. 
tt V54> Vol. IV, Part I, Supplement. 



The Mizler Musical Society. 191 

Thereupon he was admitted as fourteenth member, in 

June 1747- 

Among other things the Society occupied itself with the 
question of cantata texts, and cherished the design of 
issuing a pattern cycle*. It first contented itself with 
laying down fundamental principles for the writing of 
texts of this kind, and published the result of a conference 
upon this question, according to which the church cantatas 
should be shorter in winter than in summer, and consist 
of only three hundred and fifty bars, lasting about twenty- 
five minutes; in the good season eight to ten minutes could 
be added, and the "music" could extend to four hundred 
bars, 

The normal cantata should be laid out as follows: 
(i) chorale chorus or chorus upon a short passage from the 
Bible; (2) recitative of twelve or twenty lines in length; 
(3) an aria-arioso or a fugued chorale; (4) recitative; 
(5) aria; (6) chorale or fugue. 

A warning is given against "too ardent and too emphatic 
poetry", since there is a risk, "if violent passions are ex- 
pressed in church", of making a ridiculous effect. The 
librettist must not only have the gift of writing ably and 
edifyingly, but must also have some understanding of 
music. The da capo form is admitted for the arias; for the 
recitative, rhymes are desirable**. 

Bach does not seem to have participated in these de- 
liberations, for the cantata type that was put forward as 
the norm is quite different from his. The warning, indeed, 
that nothing passionate should be expressed in church 
seems rather to be directed against him than to emanate 
from Mm. 



* 3754. Vol. IV, Part I, p. 104. "The Society is still working 
at the church cycle, and it will be some years yet before this can 
be finished." It was never issued. 

** 1754, Vol. IV, Part I, p. 108 in; from the report, of the 
proceedings of the Society in 1746 and the following years, 5th 
,, Packet". 



192 XI. The Artist and Teacher. 

Posterity owes a debt of gratitude to the Mizler Society. 
In its journal for 1754 it gave a Necrology of Bach, that 
contains the earliest biographical material relating to him. 
Or should we rattier say that the Society is indebted to 
Bach; for but for him who would remember this old asso- 
ciation today? 

The self-taught Bach thus belonged to no school. No 
preconceived opinions guided him in his studies. His 
authorities were all acknowledged masters, the old as well 
as the new. As often as the distances, his means and his 
leisure permitted, he went to hear contemporary celebrities 
and to learn what he could by observation of them. He 
copied out the works of others. In conformity with the 
old tradition of the Bachs, he did not extend his journeys 
beyond the German frontier. None the less was he thor- 
oughly acquainted with Italian and French art. Among 
the French, Couperin chiefly attracted him. In Weimar he 
occupied himself especially with the Italians, Frescobaldi 
(15831644), Legrenzi (Lotti'sr teacher), (16251690), 
Vivaldi (died 1743)* Albinoni (16741745), and Corelli 
(1653 1713). Of the German davecinists he had a par- 
ticular regard for Froberger*, According to Forkel, he 
studied the contemporary French organ composers with 
the greatest assiduity**. 

We have evidence of Bach's occupation with other masters 
in his transcriptions of their works, sixteen davier con- 
certos, four organ concertos, and a violin concerto arranged 
for four daviers***. Misled by the inexact headings of the 
copies of these arrangements, musicians at one time thought 
they were all works by Vivaldi, whose violin concertos, at the 
time Bach was in Weimar, astonished the whole musical 



* M. Jacob Adlung, Anleitung jsur musikalischen Gelahrtheit, 
1758, p. 711: "Froberger was highly thought of by the late Bach 
of Leipzig, although he was already somewhat old." 
** Forkel, p. 24. 

*** <pk e ci av i er COBCer tos are in B. G. XLII; the organ concertos 
in Vol. XXXVIII; the concerto for four claviers in Vol. XLIII. 



Bach's Transcriptions of Other Composers' Works. 193 

world by the novelty of their style. Vivaldi was especially 
well known in Germany from his having been Concert- 
meister at the Darmstadt Court until 1713. Later research, 
assisted by the discovery of the originals of the Vivaldi 
concertos, has shewn that not all these transcriptions are 
based on works by Vivaldi*. From him are derived the 
clavier concertos Nos. i, 2, 4, 5, 7, 9 and - *4; the organ 
concertos Nos. 2 and 3, and the concerto for four claviers; 
clavier concerto No. 3 is an arrangement of an oboe con- 
certo of the Venetian Benedetto Marcello (16861739); 
No. 14 is founded on a violin concerto by Telemann; Nos. 
T II and 16 are derived from violin concertos of Duke Jo- 
hann Ernst of Weimar, who was a pupil of Walther and a 
friend of Bach. This young prince, a nephew of the reign- 
ing Duke, died at Frankfort-on-the-Main in 1715, barely 
nineteen years old, after long and severe suffering. Bach's 
transcriptions are "like a greeting sent into eternity to the 
friend who had been torn from him". 

Johann Ernst's concertos were until recently regarded as 
lost. It was only known that Telemann had edited six of 
them, and that Mattheson, in his great "General Bass 
School", had spoken of them in terms of high praise. "To 
find independent princes", he there says, "who compose 
music that can be performed, is indeed something that 
does not happen every day". In 1903 the Telemann edition 
was found by the active Bach investigator Sobering in the 
Grand Ducal library. It has a preface in French by the 
editor, that bears the date of ist February 1718. A second 
collection was to have followed, but it never appeared. 
The clavier concerto No. 13, the first movement of which 
re-appears in the first organ concerto, is also probably 
founded on a work by the young prince as is shown by 
some remarks on the manuscript copy that we possess. 
The original, however, has not yet been discovered. 



* In the B. G. edition (Vol. XLH) all sixteen clavier concertos 
still figure as transcriptions from Vivaldi. 

Schweitzer, Bach. 13 



I 9 4 



XI. The Artist and Teacher. 



The original of the clavier concertos Nos. 6, 8, 10, 12 
and 15, and the organ concerto No. 4, are as yet unknown. 
They may come to light when the treasures of the old 
chamber music become better known*. When this hap- 
pens, it is possible that this or that "original" work of 
Bach will prove to be ,a transcription. Spitta believed the 
three clavier sonatas to be especially characteristic of 
Bach's style, and even arranged them chronologically, 
placing the one in C major among the works of the com- 
poser's maturity: afterwards he had to recognise them as 
the work of Johann Adam Reinken, whom one was not 
exactly accustomed to regard as a striking composer. They 
were originally suites for three stringed instruments, pub- 
lished in the Hortus wusicus of the Hamburg composer**. 

What was Bach's object in making these transcriptions? 
It was formerly thought that he did this work simply for 
his own instruction. That may be so, to some degree, in 
the case of Vivaldi. The Prince of Weimar, however, was 
not an acknowledged master. Then did Bacli wish to make 



* Spitta (I, 411 ft) is best acquainted with the Vivaldi concerto 
that forms the basis of Bach's clavichord concerto No. 2. For later 
research into the question of Bach's arrangements see: (a) Paul 
Count Waldersee, Antonio Vivaldis Violinhonzerte unter beson- 
derev Beruchsichtigung der von J. S. Bach bearbeiteten, in the Viertel- 
jahrschrift fur Musikwissenschaft, 1885, p. 356 380. In this the 
editions are cited of the Vivaldi works concerned. Various collect- 
ions of concertos for four, two and one solo violin, with accom- 
paniment for string orchestra and cembalo, appeared in London 
and Amsterdam. It is interesting to note that of the twelve con- 
certos that appeared as op. 8, the first six are pure programme 
music. The first four are accompanied by illustrative poems (So- 
netto dimostrativo ), and bear the titles "La Primavera", "L'estate", 
"L'Automno", "L'Inverno" ; the fifth and sixth are entitled "La 
Tempesta di Mare" and "II Piacere" respectively, (b) Arnold 
Schering : Zur Bach-ForscJnmg, in the Sammelbdnde der Internatio- 
nalen Musikgesettschaft, 1902 1903, IV, 234 243, and 1903 1904, 
v 5 6 5 570- The author shews that Bach transcribed Telemann, 
Duke Johann Ernst of Weimar, and Benedetto MarceUo, and makes 
conjectures as to the derivation of the concertos the originals of 
which are still unknown. 

** Spitta, Bachtana: No. 3, Umarbeitung fvemder Originate in his 
Musikgeschichtliche Aujsdtze, 1894, p. in 120, 



Explanation of the Transcriptions 195 

these chamber music works more generally known by ar- 
ranging them for a single instrument? This was not the 
reason; for he did not concern himself in the least to tran- 
scribe them as they were in the original, but treated them 
all alike with the utmost freedom, no matter who the com- 
poser might be. Where he thinks the basses inexpressive, 
he substitutes others for them; he adds new or interesting 
middle parts almost throughout; he even transforms the 
upper voice completely when it occurs to him to do so. Not 
even the plan and the development of the works are re- 
spected. Sometimes he goes his own way immediately 
after the first bars., then follows the original for a little 
while, again branches off, returns once more, omits some- 
thing here, inserts something there, without troubling him- 
self whether his transcription becomes only half or twice 
as long as the originals*. He does not learn from the orig- 
inals, but, with his masterly corrections, rather sits in 
judgment on them though this was certainly not his 
intention. 

It is really quite inconceivable that Bach, now in the 
epoch of his first mastery, with a copious fund of themes 
and motives, should need to lean upon the ideas often 
commonplace of others. That his Weimar friend Jo- 
hann Gottfried Walther should have delighted in making 
transcriptions, and cultivated the practice assiduously, is 
not astonishing; his was not at all a creative mind. But 
that Bach, whose inventive power is simply beyond com- 
prehension, should have indulged in the practice, is merely 
part of the fact incapable of psychological explana- 
tion that whenever he could he went to external stimuli 
and examples for his own creations. This was the case not only 
in his youthful period, but also in his latest creative epoch. 
He liked other people's music in the most uncritical way, 
simply because it stimulated his own creative activity. In 



* Detailed analyses of the method of the transcriptions are given 
in the above-cited essays of Waldersee, Schermg, and Spitta. 

'3* 



ig6 XI. The Artist and Teacher. 

certain cases it was an actual necessity to him. His con- 
temporary Magister Pitschel, of Leipzig, tells us that be- 
fore improvising he generally played, from the score, a 
work by some other man, as if he first had to set the machine 
of his invention going by artificial means. This fact was 
a matter of common knowledge. "You know", writes 
Magister Pitschel to his friend, "that the famous man who 
in our town enjoys the greatest reputation for music and the 
admiration of all connoisseurs, cannot, they say, ravishpeoplc 
with his own combinations of tones, until he has played 
something from a score to set his imagination in motion" *. 

Forkel also tells us something of this suggestive action 
of other men's music on Bach. He says that if a single 
bass part, often badly figured, were set before him, he 
would amuse himself by immediately playing a complete 
trio or quartet from it; if he were in a gay humour and 
fully conscious of his power, he would instantly extem- 
porise to three obbligato voices a fourth of his own, thus 
turning a trio into a quartet. 

So that Bach transcribed the Vivaldi and other con- 
certos not to make them more accessible to the public at 
large, nor to learn from them, but simply because this 
was his way and it gave him pleasure. Nevertheless it is 
certain that he derived some profit from Vivaldi. He 
learned from him clarity and design in the structure of a 
work. " Through the Italian he won freedom from the Nor- 
thern masters and their ingenious, intricate style. There is 
preparing that great synthesis of the North German art of 
ideas and the Latin art of form that traverses Bach's work 
in the most varied phases, till finally, in the organ works 
of the later period, the art of Buxtehude and Pachelbel 
again emerges, purified, transfigured, and more profound, 
and close the circle. 



* Spitta III, 263. The passage occurs in the journal of the Gott- 
sched party, Belustigung des Verstandes und Witzes, (Vol. I, Leipzig, 
1741), in Pitschel's letter to a friend "on visiting public religious 



Back's Use of Other Men's Themes. 197 

Bach further learned from Vivaldi the perfect violin tech- 
nique, the art of writing "singably". The violin art of the 
North, which he had hitherto known, was in many respects 
more brilliant and splendid; but it had not the same know- 
ledge of how to utilise the natural advantages of the instru- 
ment. It is interesting to observe that Bach transfers violin 
music foT the clavier and the organ, and tries to get the 
effect of the strings on the keyed instruments. It shows, 
what we can also gather from his worEs, that for him there 
was really only one style, that naturally suggested by 
the phrasing of the stringed instrument and that all other 
styles are for him only modifications of this basic style, 

At other times Bach borrows from his original only the 
theme, which he then works out quite independently. We 
know that the theme of the C minor organ fugue (Peters 




IV, 6) is derived from Legrenzi (1625 1690), and that 
of the small one in B minor (Peters IV, 8) from Corelli 



(1653 1713). Albinoni (1674 1745) gave him the themes 
for two clavier fugues*. 

We cannot say, of course, how many themes of his own 
owe their origin to the effect of other men's ideas, There 
are certainly more of these, however, than we should at 
first imagine. Who would think that the consummate 
theme of the great G minor organ fugue 




* Spitta I, 425, 426. One of the fugues is in A major (Peters' 
ed. of the Pianoforte Works I, Part 13, No. 10; B. G. XXXVI, 
p. 173), the other in B minor (Peter's ed. of the Pianoforte Works, 
Part 3, No. 5 ; B. G. XXXVI, p. 178). 



198 



XI. The Artist and Teacher. 



was generated in Bach's imagination by another musician? 
We know, however, now that Reinken's Hortus nmsicus 
has been made accessible, that it was he who set Bach's 
invention on that path. The theme in question is found 
in the fifth Suite* 




Thus many a great idea, that in other men merely existed 
obscurely in an insignificant form, had to come to Bach^ 
to be endowed with the life that really belonged to it**. 

In Leipzig, where he was mainly occupied with vocal 
compositions, he concerned himself more with the masters 
of Italian vocal music Palestrina (1515 *594)> Lotti 
(1667 1740), and Caldara (16701736). To the end of 
his days, he was, Hke all the great self-taught men, very 
receptive, and sure that he could always learn something 
from others. He showed the liveliest interest in every- 
thing that appeared in his own sphere. Kant was not 
more anxious to learn what was going on in European 
literature than Bach was to get hold of the works of con- 
temporary composers. 

He had also the open mind of the self-taught man for 
inventions. He was not interested in scientific and aesthetic 
theories upon music; whatever related to practice, how- 
ever, seemed to him even if it were concerned with the 
smallest detail important enough "to be worthy of his 

* Spitta, Bachiana : No. 3, in Miisikgeschichtliche Aufsatze, 1894, 
pp. 1 1 8, 1 19. A clavier fugue of Bach's in B flat major (Peters' supple- 
ment to the Pianoforte Works I, n) is also derived from the Hortus 
musicus. 

** The fundamental idea of the Italian Concerto is found in the 
final movement of a "symphony" in Georg Muf fat's Florilegium 
primum (1695). See Arnold Schering's Zur Bachforschung, in the 
Sammelbdnde der Intern atio-nalew Mnsikgesettsckaft, 1902 1903, p. 243. 



Bach and Contemporary Organ-Building. 199 

serious attention. He was particularly interested in instru- 
ment-making. As one of the leading experts of his epoch, 
he witnessed the transition from the old to the modern 
instruments, though he saw only the beginnings of the 
new era, and still clung to the old with some tenacity. 

What he thought upon the progress of organ and 
clavier making we know from his pupil Agricola, who 
in 1759 succeeded Graun as leader of the Royal Kapelle 
in Berlin. Lorenz Albrecht showed him the manuscript 
of the Musica Mechanica Organoedi of Jacob Adlung, with 
the publication of which he had been entrusted after the 
death of the author, and asked him to indicate, in the 
notes, Bach's views upon clavier and organ -building, so 
far as they differed from what was said there, which 
Agricola did. Where Agricola is silent, we may assume that 
Adlung's opinions are in agreement with those of Bach*. 

We also know from Forkel that Bach gave his special 
attention to the improvement of the organ wind-chest and 
bellows. "The first thing he did in examining an organ was 
to draw all the sounding stops and then play the instru- 
ment, getting as much tone as was possible out of it. He 
used to say, in jest, that he must first of all know whether 
the instrument had good lungs"**. He liked good reed- 
work, and an organ could hardly have too much of it to 
please him; he did not object, for example, to the sixteen 
reeds on the organ of St. Catharine's Church in Hamburg. 



* Jacob Adlung was teacher of the Gymnasium at Erfurt, and 
organist at the Rats- und Prediger-Kirche there. His first important 
work, Anleitung zur musikalischen Gelahrtheit (1758) was published 
by himself; the second, the famous Musica Mechanica Orgqnoedi 
was left by him at his death (1762) in manuscript, but ready for 
printing. It appeared in 1768. Not only does it exhibit an ad- 
mirable knowledge of physical science and great practical experience, 
but also a fine aesthetic feeling, as shewn in his judgments. He 
lost many valuable manuscripts in a fire. At the beginning of the 
Musica Mechanica he gives the story of his own life. The index to 
the Musica Mechanica is not a complete record of the passages in 
which Bach is mentioned. 
** Forkel, p. 23. 



20O XI. The Artist and Teacher. 

Reinken, who voiced it himself, also thought its disposi- 
tion excellent *. On the same organ Bach was surprised 
by the clear and precise speaking of a 32-feet principal and 
the i6-feet pedal trombone**. 

In order to move easily and surely from one manual to 
another, Bach lilfed them to be as close together as pos- 
sible. The keys were to be short and narrow, so as to be 
spanned with ease; he liked the semitone keys which 
were white on the old organs to be narrow at the top 
and well rounded. He also appears, so we gather from 
Agricola to have favoured the old, narrow pedals, and to 
have insisted that the pedal-board should lie in a correct 
and natural position with regard to the manuals***. 

In general it may be said that organ-building in Bach's * 
time had attained to a beauty and richness of tone par- 
ticularly through the technical improvements of Andreas 
Silbermann of Strassburg (1678 1734) and Gottfried 
Silbermann of Freiberg (1683 1753), which have not 
been surpassed since. A certain finality had been reached 
in this department. It was not so, however, in the sphere 
of clavier manufacture. Here three instruments, two 
old and one new, struggled with each other for pre- 
dominance. The oldest was the clavichord, in which the 
strings were set in vibration by a rather primitive piece 
of mechanism, a "tangent" striking them from below. 
Nevertheless at the beginning of the eighteenth century 
some Important improvements had been made in this 
instrument. Its advantage consisted in the fact that it 
permitted dynamic nuances, since the strength of the tone, 
as on our modern pianoforte, could be regulated by the 
touch; it had, however, the great disadvantage of not pro- 
ducing a very loud tonef . 



* Adlung I, 66 and 187. 
** Adlung I, 288. 
*** Adlung II, 23, 24. 

f The clavichord is described in Adlung II, 135 ff. As a rule it 
had two strings to each note (p. 144). 



Clavichord, Clavicembalo, and Pianoforte. 2OI 

In the clavicembalo also called clavecin, cembalo, or, 
in general, "Fliigel", the tone was made by the strings 
being plucked, or, as Adhing expresses it, "f lipped' ', by a 
quill or metal pin *. It had a clear and penetrating tone, 
which, however, soon ceased. Nuances and a singing style 
were therefore both impossible upon it. In order to get 
at least two degrees of tone, in Bach's time clavicembali 
were built with two keyboards, one manual being for forte, 
the other for piano ; further improvements were a pedal- 
board, also provided with strings, and a manual coupler 
by means of which the lower manual could be made to 
sound simultaneously a higher octave. It was for clavi- 
cembali of this kind that the Goldberg Variations, the 
Italian Concerto, and the so-called organ sonatas of Bach 
were written. Speaking generally, all organ works could 
be played on these instruments. This explains why Bach 
published a collection of long chorale preludes as the third 
part of the Klavieriibung**. 

There also sprang up at this time the Hammerclavier 
"piano e forte*' which was destined to combine the ad- 
vantages of the clavichord and those of the clavicembalo, 
and to supplant them both. Our modern pianoforte is 
derived from it. The self-releasing hammer, by which the 
hammer-clavier first became possible, was invented about 
the same time in Italy by the Florentine mstrument-rxiaker 
Cristofori (before 1711), and in Germany by Gottlieb 



* The clavicembalo is described in Adhing II, 102 115. The 
spinet was a small clavicembalo; the term "Epinette 11 comes from 
"epine", suggesting the "thorn" or quill that plucked the strings. 
** In Bach's time the general name of "clavier" was given to 
all keyed instruments, including the organ. We still possess a cem- 
balo that belonged to Bach, It passed from. Friedemann to Count 
Voss in Berlin, and is now in the Royal collection of instruments 
there. Rust describes it in the preface to Vol. IX of the B. G. The 
tonal effect of the clavicembalo wDl be discussed in greater detail 
in the chapter on Bach's clavier works and the manner of perform- 
ing them. 

The best history of clavier construction is to be found in the 
supplement to Weitzmann's Geschichte des Klaviev spiels. 



2O2 XI. The Artist and Teacher. 

Schroter, organist at the Hauptkirche in Nordhausen (1717). 
Gottfried Silbermann, the Freiberg organ builder, tried to 
improve upon the invention and to interest his friend Bach 
in it. The master's attitude towards the new instrument 
may be given in the words of Agricola, who is our sole 
testimony in the matter: 

"Herr Gottfried Silbermann," he says in a long note to Adlung's 
Musica MecHanica*, "had at first made two of these instruments. 
The late Kapellmeister Joh. Sebastian Bach had seen and played 
upon one of these. He had praised the tone of it, indeed wondered 
at it, but had objected that it was too weak in the upper part, and 
that it was much too hard to play**. Herr Silbermann, who could 
not endure to have his work blamed, took this very ill. For a long 
time he was angry with Herr Bach for it. Nevertheless his con- 
science told him that Herr Bach was not altogether wrong. So he 
thought it best to his great credit be it said not to make any 
more instruments like these, but to try all the harder to correct the 
defects that Herr Jo. S. Bach had pointed out. He worked at this 
for many years. And that this was the true reason for this delay 
I doubt so much less, as I have heard it candidly confessed by Herr 
Silbermann himself. At last, when Herr Silbermann had really 
hit upon many improvements, especially with regard to the action, 

he again sold one to the Prince of Rudolstadt Shortly 

afterwards the King of Prussia ordered one of these instruments, 
and, as he was greatly pleased with it, he ordered several more of 
Herr Silbermann. In all these instruments, those who, like my- 
self, had seen one of the two earlier ones, were especially able to see 
how industriously Herr Silbermann must have worked at improving 
them. Herr Silbermann had also had the praiseworthy ambition 
to show one of these new instruments to the late Herr Kapellmeister 
Bach, and to get him to try it, and he was warmly praised for it." 

These were the Silbermann forte-pianos, on which Bach, 
according to Forkel, had to play before Frederick the 
Great at Potsdam. The King had acquired fifteen of 
them. u They now all lie about useless, in various corners 
of the royal castle", says Forkel in his biography of Bach 
(1802)***. This passage came into the mind of Robert 
Eitner, the able musical historian, when he was delayed 



* n, 116. 

** According to a fairly probable calculation, Silbermann made 
both these hammer-claviers about 1733. 
*** p. 10. 



Clavichord, Clavicembalo, and Pianoforte. 203 

in Potsdam by the rain one afternoon in 1873. He went 
into the castle, and found one of the Silberniann instru- 
ments in Frederick the Great's room*. We thus still 
possess one of the claviers upon which old Bach played 
to old Fritz. 

At that time, however, he could not have found very 
much pleasure in the new instrument. At any rate he did 
not trouble to acquire one for himself. Five clavicembali 
are specified in the list of the property left at his death, 
among them the three with pedals which he had presented 
to his youngest son**. 

He thus clung to the older, in fact to the very oldest 
instrument, the clavichord. This was his favorite. The 
clavicembalo was too soulless for him, although he ap- 
preciated, and knew how to turn to account, the effect to 
be had by changing from one keyboard to the other. He 
played it in public. For himself, for private musical enter- 
tainment, and for practice, as Forkel says, he used only 
the clavichord. "He found it most apt for the expression 
of his finest thoughts, and did not believe that such variety 
of nuance in the tone could be got on any "Fliigel" or 
pianoforte as on this instrument, that was indeed poor in 
tone, but extraordinarily flexible in detail." *** Mattheson 
also preferred the clavichord to the clavicembalo, on account 
of its "delicacy", as Adlung expresses it. The latter is 
also of opinion that "manieren" (embellishments) cannot 
be expressed on any other instrument so well as on the 
clavichord", and that "none sings so sweetly and so long".t 



* See Ein altes Pianoforte, in Eitner's Monatshefte fur Musik- 
geschichte, 1873. Eitner there gives an exact description of the 
instruments. 

** Spitta III, 358, 359. 
*** Forkel, p. 17. 

t Adlung, Musicamechanicall, 144 and 152. It is curious that 
in the instruments left by Bach not a single clavichord is enumerated, 
Karl Nef, in his interesting essay Klavizimbel und Klavichord, thinks 
we must conclude from this that it is not certain that Bach pre- 
ferred the clavichord, as Forkel would have us believe. See the 
Jahrbuch der Musikbibliothek Peters, 1903, p. 29. 



204 



XI. The Artist and Teacher. 



In that epoch, every artist was still to some extent an 
instrument maker, and every instrument maker to some 
extent an artist. It was at least expected of every 
capable player that he should b able to keep his instru- 
ment in repair, including the tuning and, in the case of 
the clavicembalo, the renewing of defective quills. Bach 
was expert in both of these. "No one", says Forkel, 
"could ever quill his Tliigel' to his satisfaction; he always 
did it himself. He also tuned both his Tliigel' and 
his clavichord, and was so expert at this work that it 
never took him more than a quarter of an hour*." He 
was, indeed, not so expert as Adlung, who built claviers in 
his leisure hours, until he lost all his costly woods in a 
fire. He invented, however, a lute-clavier, which he got 
the organ builder Zacharias Hildebrand to make for him. 
Agricola gives a description of the instrument, which he 
himself had seen, in Adlung's work**. What Bach meant 
to do with this clavier is not quite clear. It could only be of 
use for playing his lute compositions on a keyed instrument. 

The invention of the viola pomposa, on the other hand, 
which Bach had built by the Leipzig court instrument 
maker Hoffmann, had some practical value. It had five 
strings, tuned C', G, d, a, e, thus occupying a middle place 
between the violas and the cellos ; it was held like a violin. 



* Forkel, p. 17. 

** In Adltrag's Musica mechanica II, 139. "The author oi these 
remarks remembers to have seen and heard in Leipzig, about the 
year 1740* a lute-clavicembalo devised by Herr Johann Sebastian 
Bach, and constructed by Herr Zacharias Hildebrand ; it was slightly 
smaller than the ordinary clavicembalo, but in other respects was 
made like one. It had two catgut strings to each note and a so- 
called 'little octave* of brass strings. It is true that under ordinary 
circumstances (that is, when only one stop was drawn) it sounded 
more like the theorbo than the lute. But when the so-called lute 
stop (i. e. the damping of the metal strings) was drawn with the 
cornet stop, then lutenists by profession might almost be deceived 
by it." 

According to Rudolf Bunge (Johann Sebastian Backs Kapelle 
zu Cothen, in the Bachjahrbuch for 1905, p. 29), Bach had a lute- 
clavicembalo made for him by a master joiner when he was in Cothen. 



The Viola Pomposa. 205 

Bach had the difficult and hastily figured basses in the 
accompaniments of the arias played on it. Our informa- 
tion as to this instrument comes from Ernst Ludwig Gerber 
(1746 1819), who served under three ruling princes of 
Schwarzburg-Sondershausen as jurist and court organist, 
and occupied his spare time with the compilation of a 
historical-biographical dictionary of musicians, that was 
intended to complete Walther's work*. His father, who 
was also jurist and court organist to the princes of Schwarz- 
burg-Sondershausen, had been Bach's pupil in Leipzig from 
1724 to 1727, and had then heard the viola pomposa used 
as a substitute for the violoncello. How greatly Bach 
was pleased with this instrument may be guessed from 
the fact that to the five sonatas for violoncello he added 
another for viola pomposa. When he had this instrument 
constructed, whether earlier in Cothen or later in Leipzig, 
we cannot now be sure. He also planned a pedal-glocken- 
spiel, that was to have been added to the organ at St. 
Blasius's Church in Miihlhausen. Whether this happened 
is doubtful, as Bach left that town before the completion 
of the organ repairs which he was superintending**. In later 
times he set no value on trivialities of this kind in organs. 
It was during his life-time, again, that the oboe d'amore 
was invented, and the flute traversiere, which supplanted 
the old flute a bee, was improved. The story that at Cothen 
he amused himself with inventing a musical clock is due to 
the imagination of one of his biographers***. 



* Historisch-biogva-phisches Lexikon der Tonkunstter, Part I, 
A M, 1790; Part II, N Z, 1792; Breitkopf, Leipzig. The accounts 
of the viola pomposa are in Vol. I, columns 90 and 491 fl The second 
edition of the work appeared as Neues historisch-'biogvaphisches 
Lexikon der Tonkunstler, in four volumes, (I and II, 1812 ; III, 1813; 
IV, 1814; Kuhnel, Leipzig). For the musical history of the eighteenth 
century, Gerber's works are of priceless value. On the viola pom- 
posa see also Spitta II, 69, III, 227, 250. 
** Spitta II, 370. 

*** Bitter, Jolu Seb. Bach I, 161. He thinks that a musical 
clock in the castle at Nienburg a. S. was at one time in the castle 
at Cothen, and was made by Bach. 



2o6 XI. The Artist and Teacher. 

An innovation that Bach introduced in fingering was of 
permanent value. On the organ, as well as on the clavier, 
he desired an absolute and delicate legato style. Such a 
method of playing, however, was impossible without a 
reform of technique. At that time everyone placed his 
fingers just as it occurred to him to do; the thumb was not 
used at all, or only in cases of necessity. "My late father", 
says Emmanuel in his Versuch uber die wahre Art das 
Klavier zu spielen, "told me that in Ms youth he had heard 
great men who never used the thumb except when it was 
necessary to make big stretches. As he lived in an epoch 
in which there came about gradually a most remarkable 
change in musical taste, he found it necessary to think 
out for himself a much more thorough use of the fingers, 
and especially of the thumb, which, besides performing 
other good services, is quite indispensable in the difficult 
keys, where it must be used as nature intends."* 

About the same time as Bach, Francois Couperin (1668 
to 1733) in Paris hit upon the new fingering, and published 
it- in his book Uart de toucher le clavecin (1717). As Bach 
was deeply interested in Couperin's work, it was thought 
that he had got his fingering from him, against which 
theory, however, Emmanuel rightly protested to Forkel. 
He points out how much more radical the reform of his 
father proved than that of Couperin**. 

We must be careful, however, not to identify Bach's 
fingering with that of modern times. It was richer than 
ours, since it represents the transition from an old method 
to a new one, just as the richness of his harmonies is due 
to the fact that he writes in major and minor and yet 
keeps within the sphere of the old modes. Thus of the old 
'fingering he retained the simple passing of one finger over 
another, while adopting the new method of employing the 
thumb. 



* Section I, "On Fingering", p. 15. 

** Forkel, p. 16. Examples of Bach's fingering are to be seen 
in Friedemann's KlavierbiicMein. 



Bach's Touch and Fingering. 207 

In time this free passing of one finger over another fell 
more and more into disuse, and fingering came to be 
based simply on the combinations made possible by the 
passing under of the thumb. Emmanuel occupies a middle 
position between his father and the modern method. He 
still retains, in principle, the crossing of the fingers, but 
discards the passing of the second over the third finger, 
the third over the second, and the fourth over the fifth. He 
regards the passing of the thumb under the fifth finger as 
an abomination. On the whole he cannot do much with the 
little finger; so that Bach evidently did not use this much*. 

Every player of Bach's works, unless he wishes to 
lose himself in complicated substitutions has frequently 
to pass one finger over another, especially the third 
over the fourth and the fourth over the fifth ; so that Bach's 
music itself instructs us as to Ms fingering, and indeed 
compels us in some measure to adopt it. 

Strongly incurved and loose fingers and loose wrists 
were also part of Bach's method of playing. The fingers 
rested directly on the keys. He himself played with so 
slight and easy a motion of the fingers that one could 
hardly notice it. Only the front joints of the fingers were 
in motion; the hand preserved its rounded form even in 
tht most difficult passages ;, the fingers were only slightly 
raised above the keys. 

His touch was very complex. He aimed chiefly at a 
singing tone. To this end he did not merely let the key, 
after pressing it down, come to rest and then ascend, but 
raised it by a gradual drawing-back of the finger-tips to- 
wards the inner flat of the hand, so as to give the string 
the proper time to vibrate and die away. By this means 
the tone was not only prolonged, but also made more 
beautiful, so that even on ail instrument so poor in tone 
as the clavichord he could play cantabile and legato**. 



* See Emmanuel's fingering of the scales in the Ver$uch uber 
die wahre Aft das Klavier zu spielen (1759) I, 21 31. 
** Forkel (p. 12), discusses Bach's touch in detail. 



2O8 XI. The Artist and Teacher. 

This style of playing applies, of course, only to the 
clavichord, where finger and string stood in intimate 
relation to each other, and not to the clavicembalo. 
Bach's touch was therefore absolutely modern, for the 
latest theories upon the "singing tone" on the pianoforte 
agree in recognising it to depend not only on the striking 
of the keys, but also, to a great extent, on the regulation 
of their ascent. Bach is also modern in this respect, that 
he would have spoken not of a "touch" but only of a con- 
scious transmission of strength and pressure to the keys. 
Sebastian Erard's invention of the double escapement 
(1823), which made possible endlessly varied gradations in 
the re-striking of a tone, would thus have been joyfully 
welcomed by Bach; the mechanism of the modern piano- 
forte is, in fact, mostly only the fulfilment of his boldest 
dreams. He would not have altogether agreed with the 
size of our keys and with the strength of pressure that 
they demand. 

We have no definite information as to his organ 
pedalling. We may suppose that he did not use the 
heel much, the shortness of the pedals in those days 
not allowing this. On the other hand he would take 
advantage of the narrowness of the pedals of that time, 
which he himself favoured, to glide from one to the other 
with the toes. 

What astonished his hearers, besides the plasticity and 
clearness of his playing, was his calmness during perform- 
ance. Scheibe and Forkel refer to this as to something 
quite exceptional. Our pianists, and more especially our 
organists, would do well to take him as their model in 
this respect. 

Of Bach as a violin player we unfortunately know little. 
This much is certain, that he had thoroughly studied the 
technique of the instrument, or he would not have written 
the sonatas for solo violin in the way he did. The ties 
that he indicates for the strings in his scores also reveal 
the master. 



Bach's Improvisation, 209 

We do not exactly know how the violin was played in 
Bach's time*. It seems to be more and more certain that 
at that time the hairs of the violin bow were fastened 
without screws to the wood, and were stretched by the 
thumb of the right hand tighter or looser as one wanted. 
"Springing" bowing was thus out of the question, whereas 
double stopping was made possible to an extent which we 
today, with our bows stretched as they are, cannot form 
any idea. The chord sequences of the Chaconne, which 
we laboriously accomplish now by throwing the bow back 
on the lower strings, presented no difficulties in those days, 
when the player relaxed the bow for a moment so that it 
could curve over the strings. We can hardly imagine now 
the special tone-effects made possible by this technique. 
For echo effects the player suddenly loosened the hairs, 
thus obtaining a murmuring, ethereal tone. 

In chamber music Bach played for preference the viola. 
"With this instrument he was as it were in the centre of 
the harmony, where he could hear and enjoy to the utmost 
what was going on on both sides of him**". 

His mastery in improvisation was well known. He was 
often begged by friends to perform on the organ, and he 
generally complied with willingness. He probably pre- 
ferred the organ of the University church for these per- 
formances, since it answered better to his wishes than that 
of St. Thomas's, although even it did not wholly satisfy him. 
He himself always regretted "not to have had a really 
large and really fine organ at hand for his regular use"***. 

If he improvised for as long as two hours together, the 
theme remained the same from the beginning to the end. 
First of all he made a prelude and a fugue out of it on the 



* The following is based on Arnold Schering's article Verschwun- 
deneTraditionen desBachzeitalters, in the Bachjahrbuckfor 1904, p. 1 13. 
** Forkel, p. 46. 

*** See the Necrology in Mizler's Musikalische Bibliothek, IV", 
Part I, 1754, p. 172. It is nevertheless questionable whether he 
had the organ of the University church at his free disposal at any 
time, since the service there was not under his charge. 

Schweitzer, Bn^h. 14. 



210 XI. The Artist and Teacher, 

full organ. Then he showed his skill in registration in a 
trio or a four-part movement; then, as a rule, came a chorale 
prelude. Finally he developed a new fugue upon the old 
theme. So at least we are told by Forkel, who also says 
that, according to Emmanuel, the organ compositions that 
we have give no adequate idea of the magnificence of 
Bach's improvisation on the organ*. 

Organ-builders and organists were appalled when they 
saw him registering. u They thought that such a combina- 
tion of stops could not possibly sound well together, but 
wondered greatly when they perceived that the organ now 
sounded at its best, the effect merely being a slightly strange 
and unusual one, which could not be achieved by their 
own style of registration"**. This refers, however, only to 
the style of Bach's preliminary organ-registration for his 
prelude or fugue, not to a constant change of registration 
during the performance, after the manner that is today 
thought best for his organ works. Bach was as ignorant 
of this practice as of modern orchestration. 

It gave him particular pleasure, when improvising, to go 
into all possible keys, and to move about even in the most 
distant ones in such a way that his hearers did not observe 
it, but thought he had only modulated within the inner 
circle of a single key***. 

"In conducting", says the Necrology, "he was very 
accurate, and extremely sure in the tempo, which he gen- 
erally took very briskly." Forkel also lays stress on his 
lively tempi. This however is to be understood relatively, 
with reference to the way of taking tempi in those days. 

It seems strange, at first sight, that Bach, when impro- 
vising, should have confined himself to a single theme. 

* Forkel, pp. 18 and 22. 
** Forkel, p. 20. 

*** Forkel, p. 17. Nevertheless the "harmonic labyrinth" that 
has come down to us with Bach's name attached to it is probably 
not by him. See Spitta II, 43. This curious piece is given in the 
B. G. XXXVIII, p. 225. In the preface the possibility is men- 
tioned that it may be by Heinichen, the celebrated musical theorist. 



His Method of Working. 211 

One would have expected that this richly inventive genius 
would have been so overwhelmed by the weight of the 
ideas that thronged within him that he could not have 
kept to one theme alone, but would necessarily have given 
voice to them all. It was not so. Everything points to 
the fact that Bach did not invent easily, but slowly and 
with difficulty. We must not be misled by the fact that 
we possess hardly any sketches or draughts of his, and 
that the scores of his cantatas give the impression of having 
being thrown off in one breath, so quickly, in fact, that 
the flying pen could scarcely keep pace with the ideas. 
The working-out and elaboration of the themes may in- 
deed not have cost him very much time, since it often 
happens with him that the whole piece, with all its devel- 
opments, is already implicit in the theme, and evolves out 
of it with a certain aesthetic-mathematical necessity, quite 
irrespective of the formal element in the development of 
a chorale chorus, a fugue or a da capo aria. 

We can, however, form only a faint idea of the long and 
arduous mental work that is presupposed in the develop- 
ment of a characteristic theme of this kind, according to 
its own mysterious inner laws, into a masterly piece of 
music* If Bach, as is generally supposed, had shaken 
themes out of his sleeve after the manner of genius, it 
would be incomprehensible how they all come to be so 
extraordinarily rich and char ct eristic. In all his works 
there is not one that is banal. But neither do we get the 
impression in any one of these melodies, quivering as they 
are with inner life, of effortless invention ; and the deeper we 
penetrate into Bach, the stronger does this feeling become*. 

Bach thus worked like the mathematician, who sees the 
whole of a problem at once, and has only to realise it in 
definite values. His way of working, as Spitta says, was 
consequently quite different from that of Beethoven. The 



* The judgment of the Necrology upon Bach's themes runs 
thus: "His melodies were indeed singular, but always diversely 
rich in invention and like tho!>e of no other composer", (p. 170). 

H" 



212 XI. The Artist and Teacher. 

latter experimented with his thoughts. In each case the 
explanation must be sought in the nature of the music it- 
self. With Beethoven the work is developed by means of 
"episodes" that are independent of the theme. These do 
not occur in Bach; with him everything that "happens" 
is simply an emanation from the theme. 

It was characteristic of his method of creation that he 
generally wrote in quick succession a string of works of 
the same kind, as if he wanted to pursue to its conclusion, 
in all its possible manifestations, some image that was 
floating before his eyes, and as if he had to exhaust the 
taste of this before he could pass on to something else. 
Those works of his that have an inner relationship to each 
other almost always originated about the same time. 

He seldom took up a work a second time without funda- 
mentally transforming it; to make a copy of anything was 
to improve upon it. Yet these corrections, however radical 
they might be, were as a rule only in the details, the plan 
of the piece being affected by them only in exceptional 
cases. Among the more thorough-going revisions may be 
mentioned that of the St. John Passion. Towards the end 
of his life he appears to have intended a revision of all his 
instrumental works. Death overtook him while he was 
revising the chorale preludes. 

In Bach's documents we see the same mathematical 
habit of inind that we find in his compositions. Be it a 
petition to the Electoral Prince, or an estimate for the 
renovation of an organ, or a memorial to the Council, 
we are always struck by the same logical clearness and 
plasticity. Not a word too many, not a word too few. It 
is an aesthetic pleasure to read them. But wherever he 
wants to express something emotional, speech fails him 
completely. The passages in the texts which we are pro- 
bably justified in attributing to him are so confused and 
awkward, so nonsensical, in fact, that we ask ourselves 
how he could pen such things, to say nothing of the 
Saxon dialect which he uses in his verses. We must 



Architectonic Quality of his Music. 213 

remember, however, that this curious change almost in- 
variably occurs whenever the mathematical mind branches 
out into the domain of imaginative thought, whether in 
poetry or philosophical speculation. 

Perhaps we can still better characterise Bach's mind as 
architectonic. The powerful aesthetic impression given by 
his works comes from the harmony of the whole structure, 
in which all the copious and animated details fit quite 
naturally. Bach's music is the perfected Gothic of the art. 
The further he advances in his fugues, the simpler and 
grander become the lines. The great organ fugue in G minor 
is among the most perfect in this respect . It is extremely rich 
in the most striking and most interesting evolutions of detail; 
but not one of them diverts the interest to itself; the only 
purpose of them all seems to be to throw into relief the fun- 
damental simplicity, vitality, and lucidity of the structure. 

In Bach's music, much more than in that of any other 
composer, the plastic outline of the whole is the result of 
the optical effect of the details; it requires, in order to 
become visible, a synthetic activity of the hearer's aes- 
thetic imagination. Even to the best musician, at a first 
hearing, a Bach fugue seems chaos; while even to the or- 
dinary musician this chaos becomes clear after repeated 
hearings, when the great lucid lines come out. 

An anecdote that has survived shows that in practical 
architecture also Bach was gifted with unusual insight. 
"When he was in Berlin", says Forkel, "he was shown the 
new opera house. At the first glance he detected every- 
thing that was excellent or faulty in the non-musical por- 
tions of the building, which others had discovered only by 
experience. He was taken into the great dining room; he 
went into the gallery that runs round it, looked at the 
ceiling, and said, without more ado, that the architect had 
here accomplished, perhaps unconsciously, and without 
any one else being aware of it, a piece of jugglery. That 
is to say, when some one spoke quite softly against the 
wall in one corner of the gallery of the oblong room, 



214 



an(i Teacher. 



anyone who stood above the arch in the other corner, with 
his face turned to the wall, could hear the words quite 
distinctly, though they were not audible to any one else 
in the whole room, either in the middle or at any other 
spot. This effect came from the form and position of the 
arches fixed to th'e ceiling, the peculiar quality of which 
he detected at the first glance*." 

A man with this open and lucid understanding of every- 
thing that concerned his art must have been a good teacher. 
That he did not achieve anything with the pupils at, St. 
Thomas's was due to special external circumstances and 
difficulties, and to his own inability to preserve discipline. 
The member of the Council who declared, at the sitting 
after his death, that "Herr Bach was certainly a great 
musician, but no schoolmaster", was quite right. Kant, 
looking back upon his own years as a private tutor, said of 
himself that there had never been a worse teacher with 
better intentions. Of Bach it could be said that no one 
had ever been a worse schoolmaster with greater talent 
for teaching. But to those who were willing to learn from 
him he was an excellent guide. 

We know the names of some of his pupils. In Weimar 
and Cothen there were Johann Martin Schubart, who suc- 
ceeded him in the organist's post at Weimar in 1717; Jo- 
hann Caspar Vogler, who in 1735 won a brilliant victory 
over ten other candidates at an examination performance 
in the Markt-Kirche at Hanover; Johann Tobias Krebs, 
organist at Buttstedt, who left Walther in order to study 
with Bach; Johann Caspar Ziegler, later at St. Ulrich's 
Church in Halle; Bernhard Bach, his nephew from Ohr- 
druf, who afterwards succeeded his father, and to whom 
we are indebted for many valuable copies of Bach's organ 
works. Among those of the Leipzig period may be men- 
tioned: Johann Nicolaus Gerber, afterwards organist and 
jurist at Sojidershausen; Samuel Anton Bach, the son of 



* Forkel, pp. 20, 21. 



Bach's Pupils. 2 IS 

Ludwig Bach of Meiningen, afterwards Court organist 
there; Johann Ernst Bach, the son of Johann Bernhard of 
Eisenach; Johann Elias Bach, cantor at Schweinfurt, who, 
though well on in years, came to study with his relative 
in 1739; Johann Ludwig Krebs, the son of the former 
Weimar pupil, who studied with Bach from 1726 to 1737, 
first as a foundation scholar at St. Thomas's, then as a 
student, and whom the master thought most highly of after 
his two sons, he was afterwards organist at Zwickau, 
then at Zeitz, and finally at Altenburg, where he died in 
1780; Johann Schneider, organist at St. Nicholas's in Leip- 
zig, Bach's faithful assistant; Georg Friedrich Einicke, 
afterwards cantor at Frankenhausen; Johann Friedrich 
Agricola, (related on his mother's side to Handel), who in 
1741 became organist at Berlin, and Royal Kapellmeister 
after Graun's death in 1759; Johann Friedrich Doles, 
Bach's second successor in the cantorate ; Gottfried August 
Homilius, afterwards cantor at the Kreuzschule at Dres- 
den; Johann Philipp Kirnberger, whom Gerber brought 
to Bach), who became Court musician to the Princess 
Amalia of Prussia and died in 1783; Rudolph Straube, 
who afterwards went to England; Christoph Transchel, 
afterwards an esteemed teacher of the pianoforte in Dresden 
(died 1800); Johann Theophilus Goldberg, the clavecinist 
of Count Kayserling, for whom Bach wrote the celebrated 
Variations ; Johann Christoph Altnikol, who in 1749 became 
Bach's son-in-law; Johann Christian Kittel, who was 
eighteen years ,of age at Bach's death, and who, as organist 
at Erfurt, carried on the Bach tradition into the nine- 
teenth cerjtury (he died in 1809); Johann Gottfried Miithel, 
afterwards organist at Riga, who came to Bach when the 
latter was already in his last illness*. 



* For full information upon, the labours and the fortunes of Bach's 
pupils, see Spitta I, 522 fl, II, 47 ff., Ill, n6fl, 239 fl, 262. In 
the above list, those Thomaners are not included who, without hav- 
ing enjoyed Bach's special tuition, afterwards claimed him as their 
teacher. See Spitta III, 248 ff. 



2i6 ' XI. The Artist and Teacher. 

The chapter on Bach's work as a teacher is one of the 
most interesting in ForkePs book. Emmanuel and Friede- 
mann must have given him copious information on the 
subject. "The first thing Bach did was to teach his pupils 
his own special style of touch. To this end, for several 
months they Jiad to practise nothing but separate exer- 
cises for all the fingers of both hands, with constant regard 
to this clear and neat touch! These exercises were pre- 
scribed to everyone for several months; Bach's convic- 
tion was that they should be continued for at least six to 
twelve months. If it happened, however, that any one's 
patience was becoming exhausted after a few months, he 
was pleased to write little connected pieces, in which these 
exercises were embodied."* This was the origin of the 
Little Preludes for beginners, and of the Inventions, which 
Bach, according to Forkel, composed during the hours of 
instruction. 

When the pupils had acquired the sense of touch, he 
gave them moderately difficult exercises. This is clear 
from the arrangement of Friedemann's Klavierbuchlein, in 
which, according to our ideas, the difficulties increase rather 
rapidly. He familiarised them from the beginning with the 
"manieren", i. e., the embellishments. In Friedemann's 
Klavierbilchlein they are all noted and explained on the 
first page, so that we would almost believe that the teacher 
had used them for the first finger exercises. This authen- 
tic information as to how Bach himself performed the em- 
bellishments is of the utmost practical significance for us. 

Preludising on the part of the teacher played a great 
part in Bach's method of instruction. Gerber, who was 
his pupil from 1724 1727, had the first part of the Well- 
tempered Clavichord played for him by his master no less 
than three times, and reckoned those the happiest hours 
of his life when Bach, under the pretext of being unable to 
find any pleasure in "informing", sat down at one of his 



* Forkel, p. 38. 



His Method of teaching Composition. 217 

excellent instruments and transformed the hours into 
minutes*. 

After the pupils had made some progress in the tech- 
nique of the instrument, the lessons in composition began. 
A musical education devoted merely to the technique of 
performance, such as is mostly the rule nowadays, un- 
fortunately, was not usual then, and least of all with Bach. 
The pieces that he made his pupils play served at the same 
time as examples in composition, as the title-pages of the 
Inventions and of the Orgelbilchlein expressly declared**. 

"Bach began teaching composition 53 , says Forkel, "not 
with dry counterpoint that led nowhere, as was the way 
with other music teachers of his time; still less did he de- 
tain his pupils with calculation of the tone-relations, which 
in his opinion, did not concern composers so much as pure 
theorists and instrument makers. He went immediatelv 
to pure four-part thorough-bass, laying much stress on the 
setting-out of the voices, since in this way the conception 
of the pure progression of the harmony was made most 



* Gerber's son tells us this ia his Tonkunstierlexikon I, col. 490 ff . 

** The title-page of the Inventions is as follows: "Aufrichtige 
Anleitung, wormit denen Liebhabern des Claviers, besonders aber 
denen Lehrbegierigen, eine deutiiche Art gezeiget wird nicht alleine 
mit zwei Stimmen reine spielen zu lernen, sondern auch bei weiteren 
Progressen mit dreien obligaten Partien richtig und wohl zu ver- 
fahren, anbei auch zugleich gute inventiones nicht allein zu be- 
kommen, sondern auch selbige wohl durchzufiihren, am allermeisten 
aber eine cantable Art im Spielen zu erlangen und darneben einen 
starken Vorschmack von der Composition zu bekommen." 1723. 
( "An honest guide, by which the lovers of the clavier, and especially 
those desirous of learning, are shewn a clear way not only to play 
neatly in two parts, but also, in further progress, to play correctly 
and well in three obbligato parts; and at the same tune not only 
to acquire good ideas but also to work them out well, and above 
aU to acquire a cantabile style of playing and withal a strong fore- 
taste for composition/') 

The twelve small preludes and fugues are for organ instruction 
what the Inventions were for the clavier. Then followed the chorale 
preludes of the OrgelbucUein. The title-page of this runs as follows: 
"Orgelbuchlein worinne einem anfahenden Organisten Anleitung 
gegeben wird auf allerhand Art einen Choral durchzufuhren, anbei 
auch im Pedalstudio sich zu habilitieren ; indem in solchen darinnen 



2i8 XI. The Artist and Teacher. 

intelligible. From there he went to the chorale. In these 
exercises he himself set the original bass, and made his 
pupils add only the alto and the tenor. Gradually he let 
them make the bass also. Everywhere he insisted not 
only on the utmost purity of the harmony in itself, but 
also on naturalness of progression and a melodic flow of all 
the separate voices. 5 * 

He thus taught harmony and counterpoint simul- 
taneously, and both practically. The filling-in of the figured 
basses was to result in a genuine four-part piece, in which 
every voice was to be 'interesting. We are in a position to 
form a clear idea of his manner of teaching. The elements 
of general-bass playing are described in his own hand in 
the Klavierbuchlein of Anna Magdalena Bach (1725);* in 
addition there is a manuscript that gives in full detail the 
"Rules and principles for the four-part playing of thorough- 
bass for his musical scholars"**; thirdly, we possess a 
violin sonata by Albinoni, with the bass realised by Gerber, 
and revised by Bach***. 



befindlichen Choralen das Pedal ganz obligat traktiret wird. Dem 
hochsten Gott allein zu Ehren, dem Nachsten draus sich zu belehren." 
("A Little Organ Book, wherein a guide is given to the beginning 
organist for the working-out of a chorale in every kind of way, also 
for perfecting himself in the study of the pedal, inasmuch as in the 
chorales to be found in it the pedal is treated as quite obbligato. To 
the honour of the Lord Most High, that my neighbour may be 
taught thereby.") This work also belongs to the Cothen period, 
* They will be found in Spitta III, 347 ff. 
** "Des Koniglichen Hofcompositeurs etc. Herrn Johann Se- 
bastian Bach zu Leipzig Vorschriften und Grundsatze zum vier- 
stirnmigen Spielen des Generalbass oder Accompagnement fur seine 
Scholaren in der Music. 1738." ("Rules and Principles for playing 
thorough-bass or accompaniment in four parts, made for his musical 
scholars by Herr Johann Sebastian Bach of Leipzig, Royal Court 
Composer etc.") With many musical examples. This treatise was 
dictated by Bach. It is given in Spitta III, 3 1 5 ff . 

*** This sonata will be found in Spitta III, 388 ff. This is a most 
valuable document, as it gives us a typical specimen of Bach's way 
of realising the figured bass. No one should venture to accom- 
pany a cantata of Bach/s without first having studied thoroughly 
this exercise and his corrections. It cannot be denied that a detail 
here and there in the accompaniment is somewhat odd. 



His View of Composition. 219 

One is astounded at the clearness and force of Bach's 
method of teaching, to judge from these evidences. "To 
avoid a succession of fifths or octaves", he says in one of 
the earliest lessons, "it is an old rule that the hands must 
always go against each other, so that when the left ascends, 
the right descends. 5 '* 

His pupils were not allowed to bring him compositions 
of their own until they had fully mastered four-part writ- 
ing. Instruction in free composition began with the two- 
part fugue. Composition at the clavier was strictly for- 
bidden. Bach called this kind of composer a "clavier 
knight" ; those who, when improvising, had to follow their 
fingers instead of their ideas, he called "clavier hussars"**. 

A confused leading of the voices was forbidden; Bach 
permitted only obbligato, i. e., strictly independent, parts. 
He disputed the right of existence of harmonic inner parts 
"fallen from heaven". For a slovenly leading of the voices 
he had the expression "puddling". 

Every piece of music, he told his pupils, is a conversation 
between the separate voices, that represent the characters. 
If one of them has nothing pertinent to say, it may keep 
silence for a while, until it can again enter quite natur- 
ally into the conversation. But none must break in with 
an interjection that is meaningless and has no reason to 
be there. 

So long as they respected in this way the personalities of 
the voices and made the most of them, his pupils could 
permit themselves what liberties they liked. As regards 
intervals and melodic and harmonic sequences he let them 
do whatever they would or could, so long as it had some 
meaning and served to express an idea. Every modulation 
had to grow out of the preceding idea, Sudden audacities 
that were only meant to surprise the hearer were inad- 
missible. 



* Regula 2 of the "Vorschriften und Grundsatze". 
** Forkel, p. 23. 



220 XI. The Artist and Teacher. 

So long as they were with him his pupils were to learn, 
in addition to his own compositions, nothing but classical 
works. Bach knew only one way to get on in music, the 
way he himself had followed, to go to school to every 
true master, and to learn from his works. 

In spite of this excellent instruction, not one of his pupils 
became a great composer, not even Friedemann or 
Emmanuel. It is difficult to form a just opinion upon the 
creative works of Bach's sons; there is a danger, now of 
underrating, now of overrating them. Emmanuel's great 
significance in the development of the pianoforte sonata 
is evident from the fact that Haydn, by his own confession, 
was impelled along a quite new path by the study of his 
works; many of Emmanuel's sonatas have even today 
lost nothing of their moving beauty. But at the same 
time there is so much in his works that is utterly insig- 
nificant that we can finally rank him only among the 
interesting talents. Friedemann had more genius; many 
things in his vocal works remind us of his father's style. 
His creative faculty, however, withered all too soon. His 
father shewed what he thought of his clever organ concerto 
in D minor by doing it the honour to copy it with his own 
hand. This interesting autograph is in the Berlin library. 
This, however, is practically all that Bach's sons achieved 
in organ composition. Emmanuel's seven organ sonatas, 
only one of which has an obbligato pedal part, are really 
a negation of the organ art of Johann Sebastian*. 

* Analyses of the compositions of Emmanuel and Friedemann 
will be found in Bitter's work on Bach's sons (2 vols. 1868), Em- 
manuel's best-known church oratorio, Die Israelites in der Wuste, 
dates from the year 1775. He also set to music Klopstock's Morgen- 
gesang am Schopfungsfeste (1784) and Ramler's Auferstehung and 
Himmelfahrt Jesu (1787). Towards the end of his life he wrote 
two Passions, the first according to St. Matthew (1787), the second 
according to St. Luke (1788). He was most widely known by his 
melodies to the sacred songs of contemporary poets. He wrote 
a .great many "characteristic pieces" for pianoforte "La Buch- 
holtz", "La StahT, "La Complaisante", "Les langueurs tendres", 
"La Capricieuse", etc. We have nineteen orchestral symphonies 



Emmanuel and Friedemann as Musicians. 221 

As executive artists they were unique in their epoch. 
Dr. Burney, who visited Emmanuel during his continental 
tour of 1772, tells us in his letters that he was greatly im- 
pressed by his piano playing. When improvising, he sat 
before the instrument motionless, as if absent-minded, and 
with a look of ecstacy on his face*. Friedemann was still 
able to astonish and transport his hearers when, at the 
age of sixty-four, and greatly weakened by an unsteady 
life and by drink, he gave an organ recital in Berlin. 

They did not think very highly of each other's creative 
work. Friedemann used to say of Emmanuel that he s "had 
done some pretty little things"**. Emmanuel did not 
express a particularly favorable opinion of Haydn. 

The most important organ composer from Bach's school 
was Johann Ludwig Krebs, of whom the facetious dilet- 
tanti of the time said that "in a brook (Bach) there was 
only one crab (Krebs) caught"***. His organ chorales are 
worthy imitations of those of Bach. 

The most important productions that have come down 
to us from the school of Bach are really two works of in- 
struction, in which we have a reflection of his genius, 
Philipp Emmanuel's Versuch uber die wahre Art das 



of his, and forty-seven piano concertos with orchestral accompani- 
ment; sonatas he composed by the hundred. He published a great 
number of these in volumes of six at a time. He had, unlike his 
father, the satisfaction of seeing most of his works published. Hans 
von Biilow brought out a new edition of six of his sonatas (Peters) ; 
Leuckart issued six volumes. A complete catalogue of his works 
has been compiled by Alfred Wotquenne, the librarian of the Brus- 
sels Conservatoire (Breitkopf and Hartel, 1905). Little of Friede- 
mann's music was printed. His organ concerto (published by Pe- 
ters) should be known to every organist ; suites and piano concertos, 
edited by Riemann, have been published by Steingraber. 

* The same thing is said by J. F. Reichardt, in his Brief 'e eines 
aufmerksamen Reisenden, die Musik treffend, 1776, Part II, p. 15. 

** There is an old anecdote of a meeting of the brothers at which 
Emmanuel recognised Friedemann by his playing (he was going 
about with a troupe of travelling musicians), Leipziger Musika- 
lische Zeitung, 17991800, II, Jahrg. p, 83Off. 

*** Forkel, p. 43. 



222 XII. Death and Resurrection. 

Klavier zu spielen*, and Kirnberger's Die Kunst des reinen 
Satzes in der Musik*** 

But what, in the last resort, does it matter to a genius 
what his direct pupils do? He really becomes an instruc- 
tor, in the true sense of the word, when his mouth has long 
been closed and his works begin to speak. When Bach 
referred his pupils again and again to the works of the 
masters, as containing the sole instruction that was of any 
use, he could not foresee that only to a later epoch would 
he himself be revealed as one of the true teachers. 

It is said that Brahms awaited with impatience each 
new volume of the Bachgesellschaft. When once he had 
it in his hand, he put aside everything in order to read 
through it, "for", he said, "in old Bach there is always 
something astonishing, and, what is the main thing, there 
is always something to be learned from him." When a 
new volume of the Handel edition came, he placed it on 
the shelf, and said: "It is certainly interesting; as soon as 
I have time I will look through it.' 1 



CHAPTER XII. 
DEATH AND RESURRECTION. 

Bach enjoyed all his life the best of health. He seems 
never to have been seriously ill. In the summer of 1729 
we learn this accidentally he had an indisposition that 
came at an inconvenient time for him, since it prevented 
him from visiting Handel, who was at that time in Halle. 

Yet the state of his eyes was always rather unsatisfactory. 
Bach was extremely short-sighted. He never spared his 
eyes. In his youth he sat up, as we learn from the Necrol- 

* Vol. I, Berlin, 1759; Vol. II, Berlin, 1762. 
** Part I, 1774; Part II, sections i 3, 1776 1779. The work 
is not complete; the conclusion of the chapters on fugue in par- 
ticular is lacking. 



His last Illness. 223 

ogy and from Forkel, whole nights through, copying music; 
the demands he made on his visual powers in later times 
were hardly less . They must thus have steadily deteriorated . 
This is probably one of the main reasons for the slackening 
in his productivity from about 1740 onwards. 

At the end he was attacked by a painful malady of the 
eyes. "He was operated upon, partly from the desire to 
go on serving God and his neighbour with his still active 
mental and physical powers, partly on the advice of some 
of his friends, who placed much faith in an oculist lately 
arrived in Leipzig. Yet this operation, although it had to 
be repeated, turned out very badly. He not only lost his 
sight, but his otherwise exceedingly good bodily health 
was quite undermined by it, and by some mischievous 
medicines and other treatment, so that for a full half-year 
afterwards he was almost continually ailing*." 

During Ms illness he went on with the revision of his 
great chorale fantasias, which had already occupied him 
for some time. The manuscript part of the property left 
by Emmanuel tells a story of suffering. In the second 
version of the chorale "Jesus Christus unser Heiland" ap- 
pears the handwriting of Altnikol, who had become Bach's 
son-in-law in 1749**. Then we meet again with Bach's 
clear characters. He even found the strength to make a 
new and improved fair copy of the canonic variations on 
C4 Vom Himmel hoch da komm ich her", which he had had 
engraved and published in 1747, on joining the Mizler 
Society. 

He appears to have passed his last days wholly in a 
darkened room. When he felt death drawing nigh, he dic- 
tated to Altnikol a chorale fantasia on the melody "Wenn 



* Necrology, p. 167. In addition to this we have only Forkel's 
information as to Bach's illness and death (pp, 10 and n.) Ac- 
cording to this, the physician who performed the operation was an 
Englishman. 

** B. G. XXV. Jahrg., 2. Lieferung, pp. 140 and 142. The col- 
lection at which Bach was working in his last days comprises eighteen 
chorales. See Rust's preface, pp. 20 and 21. 



224 ^**- Death and Resurrection. 

wir in hochsten Noten sind", but told him to head it with 
the beginning of the hymn "Vor deinen Thron tret' ich 
allhier", that is sung to the same melody. In the manu- 
script we can see all the pauses that the sick man had to 
permit himself; the drying ink becomes more watery from 
day to day; the notes written in the twilight, with the 
windows closely curtained, can hardly be deciphered*. 

In the dark chamber, with the shades of death already 
falling round him, the master made this work, that is 
unique even among his creations. The contrapuntal art 
that it reveals is so perfect that no description can give 
any idea of it. Each segment of the melody is treated in 
a fugue, in which the inversion of the subject figures each 
tune as the counter-subject. Moreover the flow of the 
parts is so easy that after the second line we are no longer 
conscious of the art, but are wholly enthralled by the spirit 
that finds voice in these G major harmonies. The tumult 
of the world no longer penetrated through the curtained 
windows. The harmonies of the spheres were already 
echoing round the dying master. So there is no sorrow 
in the music; the tranquil quavers move along on the other 
side of all human passion ; over "the whole thing gleams 
the word "Transfiguration". 

Bach's eyes all at once seemed to improve. On waking 
one morning he could again see quite well, and could endure 
the light. A few hours later he had a stroke. 'This was 
followed," says the Necrology, "by a violent fever, in 
which, in spite of all possible care on the part of two of 
the ablest physicians in Leipzig, he gently and blissfully 
passed away, on the 28th July 1750, at a quarter to nine 
in the evening, in the sixty-sixth year of his age." 

The burial took place at St. John's cemetery on the 
morning of Friday, the 3ist, the second day of humiliation 
in Saxony. 



* The end of the manuscript is missing. It comprises in all 
twenty-five and a half bars. Fortunately the chorale was included 
in the first edition of the Art of Fugue, so that we possess it complete. 



Posthumous Praise. 225 

Bacli was universally lamented. Magister Abraham 
Kriegel, his colleague at St. Thomas's school, eulogised 
him in an obituary notice*. Telemann, the famous mu- 
sician, eulogised him in the following sonnet: 

"Lasst Welschland immer viel von Virtuosen sagen, 
Die durch die Klingekunst sich dort beruhmt gemacht: 
Auf deutschem Boden sind sie gleichfalls zu erfragen, 
Wo man des Beifalls sie nicht minder fahig acht't. 
Erblichner Bach! Dir hat allein dein Orgelschlagen 
Das edle Vorzugswort des * e GroBen" langst gebracht; 
Und was fur Kunst dein Kiel auis Notenblatt getragen, 
Das ward mit hocnster Lust, auch oft mit Neid, betracht't. 
So schlaf! Dein Name bleibt vom Untergange frei: 
Die Schiiler deiner Zucht, und ihrer Schiller Reih*, 
Bereiten fiir dein Haupt des Nachruhms Ehrenkrone; 
Auch deiner Kinder Hand setzt ihren Schmuck daran; 
Doch was insonderheit dich schatzbar machen kann, 
Das zeiget uns Berlin in einem wiirdgen Sohne." ** 

Emmanuel and Agricola were commissioned to write the 
Necrology which the Mizler Society wished to devote to 
its member. It appeared in 1754.*** This Necrology con- 
tains the best-known anecdotes, the fate of the music 
books in which the boy had copied out music at night, 
the contest with Marchand, the masterly organ perform- 
ance before Reinken, the visit to Frederick the Great. It 
also gives the first list of the printed and unprinted works. 

The deceased members of the Musical Society were, in 
addition, commemorated in a "Singgedicht". The com- 



* N"Atzhch& Nachvichten von denen Bemuhungen d$r Gelehrten 
andern Begebenheiten* in Leipzig,, 1750, p. 680, See Spitta III, 
276. 

** Quoted in Marpurg's Historisch-kritische Beitrage zuv Auf- 
nahme der Musik, Vol. I, 1754 1755, p. 561. 

*** Musikalische Bibliothek, Vol. IV, Part I, p. I29fi: Memorial 
oi three deceased members of the Society of Musical Sciences; i, 
Georg Heinrich Biimler, Kapellmeister in Brandenburg- Anspach ; 
2, Gottfried Heinrich Stolzel, Kapellmeister at Saxe-Gotha; p. 158. 
"The third and last is the right noble Herr Job. Seb. Bach, the world- 
famous organist, Court composer to the King of Poland and 
Elector of Saxony, and musical director in Leipzig." 

Schweitzer, Bach. ie 



226 XII. Death and Resurrection. 

position of the one for Bach was entrusted to a Dr. Georg 
Wenzky*. It is not exactly a masterpiece. First of all 
the Muses were invoked 

Chorus: "Dampft, Musen, euer Saitenspiel! 

Brecht ab, brecht ab die Freudenlieder! 
Steckt dem Vergniigen itzt ein Ziel: 
Und singt zum Trost betriibter Briider. 
Hort was fur Klagen Leipzig singt. 

Es wird euch storen: 

Doch miiBt ihrs horen." 

Then "Leipzig" appears and announces in a recitative 

"Der groBe Bach, der unsre Stadt, 
Ja der Europens weite Reiche 
Erhob, und wenig seiner Starke hat, 
1st ... leider! eine Leiche." 

After the "composers" and "friends of music" have 
given vent to their grief in rhyme, the members of the 
Musical Society, as the true inities, finally break out into 
a song of lamentation in the form of a two-part aria. At 
the end the "glorified one" speaks. He comforts the friends 
with the reflection that the musical conditions in heaven 
are better than those in Leipzig, whereupon the chorus 
brings the "Singgedicht" to an end. 

Mattheson, the leading critic of that time, made an end, 
after Bach's death, of the secret envy he had cherished 
towards him all his life long, and wrote in warm praise 
of the Art of Fugue, which appeared in 1751. u joh. Seb. 
Bach's so-called Art of Fugue", he writes in the same year, 
"a practical and splendid (praktisches und prachtiges) 
work of seventy folio copperplates, will astonish all the 
French and Italian fugue-writers, provided they can under- 
stand it, I will not say, be able to play it. How would it 
be then if every German and every foreigner should venture 
his louis d'or on this treasure? Germany is and remains, 
without doubt, the true land of organ music and fugue**." 



* It comes immediately after the Necrology, p, 173, 
** Spitta, III, 204. 



His Greatness not recognised. 227 

The same note had been struck by the celebrated Berlin 
musical theorist Friedrich Wilhelm Marpurg (1718 1795) 
in the preface that he wrote to the Art of Fugue at Em- 
manuel's request, although he had been a pupil not of 
Bach, but of Rameau in Paris' 1 '. 

It would be a great mistake, however, to imagine that 
Bach was at that time regarded as one of the leading 
German composers. It was the organist who was famous; 
the theoretician of the fugue was admired; but the com- 
poser of the Passions and cantatas was only incidentally 
mentioned. In the same volume of the Mizler Bibliothek 
that contains the Necrology, there is a list of the artists 
who constituted the glory of German music; they come in 
the following order Hasse, Handel, Telemann, the two 
Grauns, Stolzel, Bach, Pisendel, Quanz and Biimler**. 

Bach's fame made hardly any advance during the 
eighteenth century. Johann Adam Hiller, in his Lebens- 
beschreibung beriihmter Musikgelehrter (1784) allotted him 
only a few superficial pages, which, moreover, dealt only 
with the "coryphaeus of all organists"; and Gerber, in his 
Tonkunstlerlexikon, does not take the slightest trouble to 
appreciate the composer as well as the virtuoso***. 

For all that we must not be unjust to those who did not 
recognise his greatness at that time. They were not to 
blame; they could not do otherwise. In the first place we 
must take into consideration the artistic ideal of the men 
of that epoch; they were too simple to rank the art of the 



* This preface is given in the B. G. edition of the Art of Fugue, 
XXV, i, 1875, P- * 5- Kirnberger and Marpurg could not endure 
each other. 

** Musikalische Bibhothek> Vol. IV, Part I, 1754, p. 107, In 
the Kritischer Musikus of 5th May 1739 (page 80), Scheibe places 
Bach in the fifth place, after Fux, Hasse, Handel, and Telemann. 
*** J. A. Hiller (1728 1804) was cantor at St. Thomas's from 
1789 to i8or. His Lebensbeschreibungen beruhmter Musikgelehrter 
und Tonkunstler neuerer Zeit appeared in Leipzig in 1784. Bach 
is discussed on pp. 9 23. 

Ernst Ludwig Gerber's Neues Tonkunstlerhxikon appeared in 
2 vols., in 1790 1792. 

IS* 



228 XII. Death and Resurrection. 

previous generation as highly as that of their own. They 
were convinced that music was always advancing, and as 
their own art was later than the old art, it must necessarily 
come nearer to the ideal. That epoch could not resign it- 
self to regarding mere performers of other men's work as 
artists; if a man wished to appear before the public, it 
should be with works of his own. So implicitly was this 
principle accepted, that many people did not hesitate to 
perform the works of others under their own name. It 
was only when musicians began to recognise that other 
men's living thoughts were better than their own dead 
ones, and to be content with being purely executive artists, 
that the past ceased to be regarded as surpassed by the 
present. This however was not until the end of the 
eighteenth century and beginning of the nineteenth; but 
until it happened, Bach's day could not possibly come. 

Nor must we forget that even during the master's life- 
time art had taken a path that led it far away from can- 
tatas and Passions. People were weary of fugues and of 
pieces constructed of obbligato parts, and longed for a 
music that should be spontaneous feeling and nothing 
else. The concept of Nature which, in the epoch of grow- 
ing rationalism, transformed philosophy and poetry, as- 
serted itself also in music. Answering as they did to the 
needs of the epoch, the emotional compositions of the day, 
with their * 'tender and pathetic expression", insignificant 
as they were in themselves, appealed to thoughtful artists as 
being nearer to truth than the music of the epoch of rigid rule. 

That Bach's art in its own way was also true to nature, 
and that in his strict polyphony a volcanic emotion and 
thought were embedded, like substances petrified in lava 
this the men of the expiring eighteenth century could not 
see, \There has never been a movement so lacking in the 
historical sense as the rationalism of the eighteenth cen- 
turyj The art of the past, in every department, it regarded 
as mere affectation. Everything old was necessarily anti- 
quated, at least in its form. If they were to appreciate its 



His Greatness not recognised. 229 

utterance, it must be expressed in asimpler and more natural 
way. In this spirit they restored so much of the ancient buil- 
dings as they allowed to remain ; in this spirit did the BacK 
admirers of that epoch, among them his own sons a'nd 
Zelter, undertake revisions of his works that are among 
the most barbarous things of their kind in the whole world. 

Zelter, in particular, discovered French powder in Baclr* 
pigtail. "Old Bach," he writes in one place, "is, with all 
his originality, a son of his country and of his epoch, and 
did not escape the influence of the French, especially of 
Oouperin. People try to be agreeable, and so write what 
cannot last. This alien product, however, we can skim 
off him like a thin froth, bringing out the real worth of 
what lies immediately underneath, I have arranged in 
this way many of his church pieces for myself alone, and 
my heart tells me that old Bach nods approvingly at me, 
like the good Haydn: 'Yes, yes, thatis just what I wanted'."* 

"Bach's sons were the children of their epoch, and never 
understood their father; it was only from piety that they 
looked at him with child-like admiration.' 3 This remark of 
Eitner,** severe as it may be, is true. The London Bach 
was even lacking in piety. He spoke of his father as "the 
old perruque". 

In the eyes of the public and the critics of the end of the 
eighteenth century the great composer of the Bach family 
was Emmanuel. No one stood so much in the way of his 
father's fame as he. Burney (1726 1814), the celebrated 



* Zelter to Goethe, 5th. April 1827. See Goethes und Z alters 
Briefwechsel, Reclam's ed., vol. II, pp. 467 and 468. On the 22nd 
April Goethe asks his musical friend to tell him more precisely 
what he means by French froth, and how he would skim it off from 
the German basic element (II, 472). He received no definite answer 
from. Zelter on the point. 

Karl Friedrich Zelter (1758 1832) was the conductor of the 
Berlin Singakademie. 

** Monatshefte fur MusikgeschicUe, 1885: Vber Wilhelm Fnede- 
mann Bach. Nevertheless Zelter teUs us that Philipp Emmanuel 
said with regard to his father, "We are all children beside him". 
See the Briefwechsel Goethes und Zelter s, Reclam ed. II, 517. 



230 XU. >eath and Resurrection. 

English critic, who visited him on his second continental 
tour, in 1772, extols him as one of the greatest composers 
for keyed instruments that ever lived, and is of opinion 
that he is not only more learned than his father, hut "is 
far before him in variety of modulation". He makes only 
casual mention of the fact that Emmanuel had showed 
him the two volumes of the Well-tempered Clavichord; 
they were compositions that the deceased Herr Kapell- 
meister had written "on purpose for him when he was a 
boy". Burney was several days in Hamburg, and spent 
almost the whole time with Emmanuel; the latter, how- 
ever, did not play him a single note of his father's music*. 

In a conversation with his visitor, Emmanuel made 
merry over the composers who affected canons, and said "it 
was ever a certain proof to him of a total want of genius in 
any one that was fond of such wretched studies and un- 
meaning productions". On the other hand he praised 
Hasse, "the greatest cheat in the world", who 3 in his com- 
positions, without considering the obbligato leading of the 
parts, made such divine effects "as one could never ex- 
pect from a crowded score". This points to a quite new 
conception of orchestral composition, which was after- 
wards realised in the Beethovenian symphony; but at the 
same time it shows the son's complete lack of understand- 
ing of the character of his father's scores. 

For Reichardt, the greatest critical authority of that 
epoch, old Bach ranks considerably below Handel, though 
both are reproached with clinging to old forms**. The only 
man who ventured to place Bach above Handel at that 
time was an enthusiastic anonymous author of an article 



* Bxirney, The Present State of Music in Germany, etc.> 1773, 

II, 124ft 

** Job. Friedr. Reichardt (1752 1814), Kapellmeister to the 
Prussian Court, of which post he was deprived in 1792 on account 
of his sympathy with the French Revolution, had also a great name 
as a composer. He discusses Bach in the Musikalisches Kunst- 
magazin {2 vols. 1782 1791), Vol. I, p. 196. See also Bnefe ernes 
aujmerksamen Reisenden, 2 Parts, 1774 1776. 



His Greatness not recognised. 231 

in the Allgemeine deutsche Bibliothek upon Bach's piano 
and organ works*. 

Bach's case was greatly prejudiced by the German 
Handel-worship that dated from the first Berlin perform- 
ance of the Messiah under Hiller (igth May 1786), especially 
as Hiller "became cantor at St. Thomas's, Leipzig, in 1789, 
where for ten years he worked for Handel and his own 
teacher Hasse, as if a Johann Sebastian Bach had never 
existed. When he had completed his own collection of 
motets, and was desirous of doing still more to increase 
the supply of good church music, it did not occur to him 
to publish the few cantatas of Bach that lay in the musical 
library of the cantorate, but only to issue the finest pieces 
from Hasse's Italian operas, with a German church text 
added to them. The clergy of the town took the greatest 
interest in this undertaking**. 

We learn from Zelter that Hiller tried "to inspire the 
Thomaners with abhorrence of the crudities" of Bach***. 

The only cantor of St. Thomas's of the second half of the 
eighteenth century who did anything for Bach was Doles, 
and he only half-heartedly. Although he had actually been 
Bach's pupil, he made a rule for himself, when writing con- 
trapuntally, "to observe its proper limits, and at the same 
time not to forget delicate and affecting melody", in which 
he took Hasse and Gratin as his modelsf. Nevertheless, 



* Attgemeine deutsche Bibliothek, edited by Nicolai, Vol. 81. 
This article is mentioned in Richard Hohenemser's interesting 
study, Welche Einflusse hatte die Wiedevbelebuwg der dlteven Musik 
im XIX. Jahrhundert auf die deutschen Komponisten? Leipzig, 1900. 
** The plan was not carried out, the hundred subscribers it re- 
ceived being insufficient. See Lampadius, Die Kantoren der TAo- 
masschule zu Leipzig, 1902, p. 50. 

*** Briefwechsel zwischen Goethe und Zelter, Reclam's ed., Vol. II, 
p. 507 (19-01 August 1827). 

f See Richter's essay on Dole's autobiography, in the Monats- 
hefte fur Musikgeschichte, 1893. Bach's immediate successors were 
Gottiob Harrer (1750 1755), who had received his musical educa- 
tion in Italy, Johann Friedrich Doles (1756 1789), and Johann 
Adam Hiller (1789 1801). 



232 XII. Death and Resurrection. 

as we know from Rochlitz, who was a Thomaner under 
him from time to time he performed works by Bach, 
among them certain motets and Passions*. It was 
through Doles that Mozart, who revered and loved him, 
made the acquaintance of Bach's motet Singet dem Herrn 
ein neues Lied. Rochlitz, who was present at that per- 
formance, gives the following account of it: 

"Mozart knew Bach more by hearsay than from his works ; at 
any rate he was quite ignorant of his motets, which had never been 
printed, Scarcely had the choir sung a few bars when he started 
up; a few bars more, and he called out: 'What is that?' And now 
his whole soul seemed to be in his ears. When the performance was 
over, he called out joyfully, 'That is indeed something from which 
we can learn!' He was told that this school, at which Sebastian 
Bach had been cantor, possessed a complete collection of his motets, 
and treasured them as sacred relics. 'That's right! that's fine I' 
he said. Xet me see them.' As there were no scores of these works, 
he got them to bring him the separate parts; and now it was a joy 
to the silent observers to see how eagerly Mozart distributed the 
parts around him, in both hands, on his knees, on the nearest chairs, 
and, forgetting everything else, did not rise until he had carefully 
read through everything that was there of Bach's. He begged and 
obtained a copy for himself, which he valued very highly**," 

The one who seemed in the best position to produce 
Bach's cantatas was Emmanuel, who was church music 
director at Hamburg. So far as we know, however, he 
performed only a few cantatas and sections of the B minor 
Mass. In any case he would not have been able to do 
much for his father's works, even if he had wished, for his 
chorus and orchestra were in a sorry state. Burney laments 
that a piece of church music which he heard in St. Cathe- 
rine's it was one of Emmanuel's own should have 
been done so very badly, "and that the congregation should 
have listened to it so inattentively" ***. Nothing remain- 
ed in Hamburg at that time of the old enthusiasm for 



* Rochlitz, Fur Freunde der Tonhunst, II, 2ioff., Ill, 364. 
** Fur Freunde der Tonhunst, II, 212, 213 note. Rochlitz had 
already published this anecdote previously in the first volume of 
the Leipzig Musikalische Zeitung. 

*** Burney, The Present State of Music in Germany, etc., II, 249. 



His Greatness not recognised. 233 

church music. "You should have come here fifty years ago/* 
said Emmanuel sadly to his visitor. It is evident, too, 
from the discussions upon the reform of church music 
after his death, that people had taken his productions of 
his father's works in ill part. At any rate the pastors were 
forced to defend him against the reproach that "many 
old compositions had been used, and with them also an 
old and often unedifying text". His apologists excused 
him on the ground "that it could not be otherwise with 
the excessive quantity of music" *. How little interest was 
felt in religious art is shown by the fact that the college 
of sexagenarians abolished the regular Sunday music from 
motives of economy, retaining only the music of the six 
feast days. 

Conditions were no better in the other German towns. 
The regular music was abolished; the cantorate choirs were 
mostly extinct; voluntary mixed choirs scarcely existed 
as yet, and were not permitted in the church service **v 
Bach's cantatas were really impossible in the churches by 
reason not only of their music but of their old-orthodox 
texts. We must bear all this in mind before we speak of 
the lack of understanding of the musicians who let Bach's 
works fall into oblivion. 

Religious music was possible only outside the church 
service, and even this had to be of the kind in which the 
effect depended more on the choral masses than on the 
text. For this reason Handel, the oratorio composer, 
triumphed over Bach, the master of the cantata. The 
reaction against rationalistic spiritual poetry had to be far 



* On tliis and the following subject see Joseph Sittard's 6?^- 
schichte des Musik~ und KonzBVtwesens in Hamburg ; 1890, p. 47 ff. 
** The Berlin Singakaclemie was founded in 1791; the other 
singing societies did not originate till after 1800. The Frankfort 
C&cilienverein, for example, which did so much for Bach, dates 
from 1 8 1 8. At first, however, it occupied itself only with Handel, 
most of whoso works it performed Alexanders Feast in 1820, 
Judas Maccabeus in 1821, Samson in 1822, the Hundredth Psalm 
in 1823, The Messiah in 1824, and Israel in Egypt in 1827, 



234 "XII* Death and Resurrection. 

advanced before the public could again tolerate a Bach 
text*. 

Even if anyone, in despite of the epoch, had wished to 
give Bach's works, he could hardly have done so, for the 
simple reason that they were nowhere to be had. The 
possessors of the five yearly series of cantatas were Em- 
manuel and Friedemann, who had divided the treasure 
among them. Those in Friedemann's hands were soon 
dispersed**. Emmanuel took more care of his. To pub- 
lish them, however, was impossible, on account of the 
expense; nor would he have found many purchasers. His 
unfortunate experience with the Art of Fugue, of which, 
by the autumn of 1756, only thirty copies had been sold 
was not encouraging. So he confined himself to lending 
the scores of the cantatas to the few people who were in- 
terested in them, for inspection or for copying, for which 
they had to pay him a fee, not even Forkel, his friend, 
being except ed. After his death his wife continued the 
business; when she died, in 1795, Anna Carolina, the sole 
surviving grand-daughter of Johann Sebastian Bach, ap- 
pended to the obituary notice in the papers an, announce- 
ment that she would continue with the utmost attentive- 
ness the business hitherto carried on by her late mother 
with the music of her late father and grandfather***. 

In Leipzig there were the parts of the motets which 
belonged to the school three Passions, and some cantatas. 
These were probably the works that Bach's widow offered 
to the Council in 1752, when she applied for relief f. 
The works of Bach in the possession of Amalie, the sister 



* Even Zelter was repelled by the Bach- texts ; on the other hand 
lie thought the libretto of The Messiah admirable. See the Bne/- 
wechsel mit Goethe, Reclames ed. II, 259. 

** Fortunately a great part of them c^me into the possession of 
Count Voss, of Berlin. 

*** Hamb. Kowesp. 1795, No. 122. 

f According to Rochlitz, there were twenty-six chorale cantatas 
at the St. Thomas school in the time of Doles. Fttr Freunde der 
Tonkunst, III, 364. 



The Turning of the Tide. 235 

of Frederick the Great and a pupil of Kirnberger, were 
temporarily withdrawn from publicity, and were known 
only to a few intimate friends. After her death, in 1787, 
her collection went to the Library of the Joachimsthal 
Gymnasium in Berlin*. 

The piano and organ works were hardly more widely 
diffused. Of those that hacl been engraved, there had al- 
ways been so few copies that they were scarcely better 
known than those which, in Bach's lifetime, circulated 
only in manuscript. It is hardly credible how little was 
known of Bach by those who spoke admiringly of him. 
It had always been so. Marpurg's celebrated 'Treatise 
on fugue, according to the principles and examples of the 
best German and foreign masters" **, gives us the im- 
pression that, apart from the Art of Fugue, he had not seen 
many fugues of Bach; yet he refers enthusiastically to him 
in -his preface. The chorale preludes, again, he scarcely 
knew, judging by the way in which he speaks of them***. 
The Well-tempered Clavichord was perhaps the most widely 
known work. But at the end of the eighteenth century 
it seemed, on the whole, as if Bach were for ever dead. 

At the very beginning of the nineteenth century, how- 
ever, there was felt the breath of the spirit that was to 
wake him to immortal life in his works. In 1802 Forkel's 
biography appeared; this marks the turning of the tide. 

Johann Nicolaus Forkel (17491818) was the University 
musical director at Gottingen. He was also a musical 
historian, engaged on a general history of music, that was 
to extend from the foundation of the world to his own 
time. As he feared that he might die before he got as far 
as Bach, and he thought it imperative that he should 

* Robert Eitner, Katalog der Musihaliensammlung des Joachim- 
thcdschen Gymnasiums m Berlin, 1884. 

** 2 vols., Berlin, 1753 and 1754. The first volume is dedicated 
to Telemann, the second to Friedemann and Emmanuel Bach, 
*** Marpurg, Hist.-knt. B&itr&ge zur Aufnahme der Musik, 5 vols., 

I7 ,j4 1772. See the article on chorale accompaniment, Vol. TV, 

p. 192 ft 



236 XII. Death and Resurrection. 

preserve for the world what he had learned of the master 
from his two sons, he decided to anticipate and publish the 
chapters on Bach, especially as the "Bureau de Musique" 
of Hoffmeister and Kiihnel in Leipzig was planning an 
edition of Bach's works*. His biography was to prepare 
for and justify that undertaking. 

The significance of this work of sixty-nine pages does 
not lie in what it actually says, although it contains 
plenty of interesting things, nor chiefly in the fact that 
it made the world, for the first time, acquainted with Bach 
and his art, but in the conquering enthusiasm that ani- 
mates it. Forkel appealed to the national sentiment. "The 
works that Johann Sebastian Bach has left us," he says at 
the beginning of his preface, "are a priceless national herit- 
age, of a kind that no other race possesses." And again, 
"the preservation of the memory of this great man is not 
merely a concern of art; it is a concern of the nation". 
The book closes with the words : "and this man, the greatest 
musical poet and the greatest musical rhetorician that 
has ever existed, and probably that ever will exist, was 
a German. Be proud of him, oh Fatherland, be proud of 
him, but also be worthy of him!" 

Zelter goes astray when he imagines that Forkel had 
written a life of Bach without knowing more about it than 



* He did rightly: at his death the Allgemeine Geschichte d&r 
Musik had only arrived as far as the i6th century (2 vols., 1788 
to 1801). Zelter, who could not endure him, writes to Goethe in 
1825: "Forkel was both Doctor of Philosophy and Doctor of Music, 
but all Ms life came into immediate contact with neither the one 
nor the other, and made a bad end. He began a history of music, 
and ended just where a history is possible for us." Briefwechsel 
Goethes und Zelters, Reclam ed. II, 358. This verdict is unjust, 
The complete title of this first biography runs : Vber Johann Se- 
bastian Backs Leben, Kunst und Kunstwerke. Fur patnotische Vw- 
ehrer echter musikalischer Kunst. Leipzig: Hoffmeister und Kiihnel, 
Bureau de Musique 1802. The book is dedicated to Freiherr von 
Swieten (1734 1803) who, besides being an admirer of Bach, was 
a friend of Haydn and of Mozart and a patron of Beethoven, He 
was director of the Royal Library at Berlin, and also chairman of 
the Court Commission of Education there. 



Forkel and Rochlitz. 237 

the whole world knew already*. No one before him had 
understood as he did the greatness of the Leipzig master. 
It is true that he treats of the creator of the cantatas and 
the Passions very briefly; but this was for the simple 
reason that Forkel had seen only a few of these works. 

If Forkel was the first Bach biographer, Rochlitz was 
the first Bach aesthetician**. He tells in really thrilling 
style the story of how he came to Bach. The fact that as 
a boy he had sung in Bach's motets and Passions at 
St. Thomas's had merely made him "scared" of the master 
and his works. As a young man he feels himself attracted 
to him by a vague enthusiasm, and he studies the four- 
part chorale movements in the cantatas that Emmanuel 
had published. Light does not dawn on him, however, 
for in the absence of the text in this edition he does not 
understand what Bach is aiming at. From these he goes, 
not knowing of the existence of the Inventions, to the 
Well-tempered Clavichord, Against the pieces in this col- 
lection that please him he makes a mark. At first there 
are very few of these; with further acquaintance several 
more are added, then still more, till at last "in the first 
Part about half, and in the second Part perhaps two-tliirds, 
have their mark in the margin". Then he ventures on the 
vocal compositions, and now' Father Bach appears to him 
as * e the Albrecht Durer of German music", since he * 'chiefly 



* Bviefwechsel Goethes und Zelters, Reclam's ed. II, 358. 
** Johann Friedrich Rochlitz (1769 1842) was editor of the 
Att$emein& musikalische Zsitung t founded by Breitkopf and Hartel 
in 1 798. Everyone knows how he fought with his pen for Beethoven's 
symphonies in Germany. The various articles that he wrote upon 
Bach during the second and third decades of the nineteenth century 
are collected in his chief work Fttr Preunde der Tonhunst, 4 vols., 
Leipzig; Vols I and II, 1830 (2nd ed,); Vol. Ill, 1830 (ist ed.); 
Vol, IV, 1832 (ist eel.), See, in Vol. II, "Geschmach an Sebastian 
Bach's Kompositionen, besonders filr Klavier; Brief an einen Freund" 
(pp, 203 229); in Vol. Ill, "Joh. Seb. Kantate, 'Ein* feste Burg ist 
unser Got?" written in 1822, to make known that work, which 
had just been published by Breitkopf (pp. 361 381); in Vol, IV, 
" Uber S. Backs gtasse Passionsmusih wack dem Evangelisten Jo- 
hannes" (pp. 397 448). 



238 XII. Death and Resurrection. 

attains great expression by the profound development and 
inexhaustible combination of simple ideas". In this he 
contrasts Bach with the "moderns" of his own time, as one 
of the old masters whom one discovers when one seeks the 
way that leads from the art that "pleases" to that which 
"contents". "But", he confesses, "we must, to be sure, 
first of all get used to these old masters; it depends on them 
as well as on us." 

His analysis of the St. John Passion and the cantata 
Ein* jeste Burg are masterpieces of aesthetic criticism. The 
finest thing about them is their immediate freshness. He 
pronounces judgments towards which he has slowly and 
painfully worked his way, and which he is almost afraid 
to set forth, for even to himself they seem too astonishing. 
He ventures to set Bach above Handel, for his parts al- 
ways move so independently, and yet work together with 
such marvellous unity as is hardly ever attained by other 
composers. If Handel is more splendid, Bach is truer. If 
the one is a Diirer, the other is a Rubens. It is rarely, he 
goes on to say, that Bach pleases immediately, rarely that 
he works directly on the emotions ; he addresses himself to 
the "active, inflammable, and penetrating" representative 
reason. This gives the hearer the satisfaction that comes 
from a perception of the truth. And the first Bach ses- 
thetician says of the recitatives of the St. John Passion, 
"This truth, this sincerity, this clear delineation of char- 
acters and events merely by tones and rhythms, this art 
that is apparently simple and hidden, and yet is so rich, 
deep and manifest, who has ever exhibited this 
precisely this more perfectly? Who can even imagine 
it being exhibited more perfectly?"* 

* Of the history of Protestant church music Rochlitz and his 
contemporaries know nothing. They have no idea of what lies on 
the other side of the rationalistic epoch, Rochlitz does not know 
that there was a musical Passion in the church before Bach. He 
thinks that Bach and his superintendent Deyling were the first to 
invent the plan of a Passion of the type realised in the St. John. He 
is ignorant, too, of the previous history of the cantata, for he re- 



Rochlitz. 339 

Bach's time is therefore bound to come. Rochlitz does 
not feel that it is now close; he rather believes that it will 
be delayed. He remarks that the first enthusiasm of 1800, 
"when the rolling wheel of fate for a brief moment brought 
the spoke of the revered Father Sebastian Bach to the 
highest point", is stifled. The projected edition of Bach's 
works did not come to anything, and many people no longer 
saw any practical object in publishing his complete works. 
However he recommended it, apart from any practical ob- 
ject, to the future, "since the revolution of things brings 
to the top again, after a shorter or longer interval, all the 
main manifestations of the greater human spirits". 

This perception does- honour to the artist Rochlitz; 
equally honorable to Rochlitz the man is his care of Bach's 
last child. When he learned that Regina Susanna, who 
had been eight years old at the death of her father, was in 
want, he published an appeal in the May number of the 
Musikalische Zeitung for 1800, which runs thus; 

"Rarely hav I taken up the pen with so much alacrity as now; 
for scarcely ever could I, in confidence in the goodness of men, be 
so firmly convinced of doing something useful with it as now, The 
family of the Bachs has become extinct with the exception of one 
daughter of the great Sebastian. And this daughter, now very old, 
this daughter is starving. Very few p6ople know it; for she cannot, 
no, she must not, shall not, beg! She shall not do so; for surely 
people will listen to this appeal for her support; surely there are still 
good men who will consider, not me how could I hope for that? 
~ but a fitting occasion to see that the last twig of so fruitful a stem 
does not perish without cam If everyone who has learned some- 
thing from the Bachs would give only the smallest trifle, how com- 
fortably and free from care would the good woman be able to spend 
her last years!" 

As one of the first who had learnt from- Bach, Beethoven 
sent his contribution; a year later he gave Brcitkopf and 
Hartel a work to publish for the same good object; there 
were also other donors, so that enough was got together 

gards the chorale cantata as the most original creation of Bach, 
who based his cantatas upon chorale melodies in order that the 
congregation, who did not understand his music, might at least 
have some part in the well-known, all-pervasive tunes, 



240 XII. Death and Resurrection. 

to set the remainder of the life of Bach's daughter free from 
galling care. 

Beethoven had made the acquaintance of Bach through 
his Bonn teacher, Christian Gottlob Neefe (1748 1798). 
When still a boy he studied the Well-tempered Clavichord, 
which in after years he used to call his musical Bible. It 
was he who said: "Nicht Bach! Meer sollte er heissen"*. 
At the end of his life he planned an overture on the name 
of Bach**. 

Zelter won two friends for Bach, Goethe and the 
youthful Mendelssohn. He himself had had to struggle to 
understand Bach, and had only gradually comprehended 
that 3 as he writes to his friend on the 9th June 1827, Bach 
was "a poet of the first order". He closes this letter, which 
branches out into quite a dissertation on Bach, with the 
remark: "When everything has been weighed that can be 
said against him, this Leipzig cantor is a sign from God, 
clear, yet inexplicable"***; to which he does not forget to 
add proudly that he might say to him : 

"Du hast mir Arbeit gemacht, 

Ich liabe dich wieder ans Licht gebracht," 

(Thou hast given me work; I have brought thee to the 
light) f. 

Goethe listens willingly. Zelter sends him the Well- 
tempered Clavichord, Schiitz, the organist at Betka, plays 
to him from it; whereupon something of the greatness of 
the old master dawns upon him. On 2ist June 1827 he 
writes, "I expressed it to myself as if the eternal harmony 
were communing with itself, as might have happened in 
God's bosom shortly before the creation of the world. It 
was thus that my inner depth were stirred, and I seemed 
neither to possess nor to need ears, still less eyes, or any 

* "He should be called Sea (Meer), not Brook (Bach)." [Tr,] 
** See Arthur Priifer, Sebastian Bach und die Tonkunst de$ 
XlX.Jahrhunderts. Leipzig, 1902, pp. 10 and n, 
*** "Klar, doch unerklarbar." [Tr.] 
t Goethes und Zelters Briefwechsel, Reclames ed. II, 481 ff. 



Zelter and Goethe. 241 

other sense*. When the young Mendelssohn was staying 
with him in May 1830, he had to play to Goethe a good 
deal of Bach. He remarked of the overture to an orchestral 
suite in D major, which his guest played on the piano 
for him, "the opening is so pompous and important that 
one can really see a file of trim people going down a great 
staircase**." 

Zelter always regretted that his friend could not attend 
any of the performances of the motets in the Singaka- 
demie. "Could I," he writes on 7th September 1827, "let 
you hear, some happy day, one of Sebastian Bach's 
motets, you would feel yourself at the centre of the 
world, as a man like you ought to be. I hear the works 
for the many hundredth time, and am not finished with 
them yet, and never will be***." 

The best work that Zelter did for Bach was when he 
prevailed upon himself to retire in favour of his pupil 
Mendelssohn, and allow him to perform the St, Matthew 
Passion with the chorus of the Singakademief, This 



* Ibid. II, 495, 

** Mendelssohn quotes this saying in a letter of 22nd June 1 830 
irom Munich to his teacher. He says that he has also played for 
Goethe the Inventions and several things from the Well-tempered 
Clavichord. See his Reisebnefe aus d&n Jahren 1830 1832, Leipzig, 
1869, p. 17, Goethe, in his Diaries, records that Mendelssohn played 
him works by old and new masters, He says nothing of the im- 
pression Bach made on him* 

*** Ibid, II, 517. See also III, 457 and 45 8. Zelter seems also to 
have performed some of the cantatas privately. He particularly 
admired Brich dem Hungrigtn d&in Brot, Ihr werdet weinen und 
heulen, Jesus nahm m sick die Zwdlfe, and Unset Mund sei voll 
LachenSy in which he is astonished at Bach's "saintly simplicity" 
and the "apostolical irony 11 of the musical interpretation of the 
text, by reason of which "it often expresses something else than 
what the words say, 1 ' (II, 482), In general, however, Bach's texts 
were an abomination to him. For this reason he did not contem- 
plate a public performance of the cantatas and the Passions. 

t Zelter had directed the Singakadernie since 1800, It had 
been founded in 1791 by Karl Friedrich Fasch, who in 1756 had 
been appointed second clavicembalist at the court of Frederick the 
Great. The father of this Fasch, Johann Friedrich (1688 1758). 
Court Kapellmeister at Zerbst, had been Bach's competitor for the 

Schweitzer, Bach. I<5 



242 



XII. Death and Resurrection. 



had not been easy for him. He was on the point of dis- 
missing with a surly reply the two "y un people" 
Eduard Devrient accompanied Mendelssohn on the difficult 
errand who had disturbed him at his work on that Jan- 
uary morning in 1829. Mendelssohn already had his hand 
on the door to go; the old man growled something about 
young cubs who thought themselves capable of everything; 
hut Devrient, from whom the whole plan had come, did 
not lose heart, and at last brought him round. 

The success of the work, so far as the singers were con- 
cerned, was decided after the first rehearsals. When 
the two friends set out to engage the soloists, they re- 
marked, in front of the opera-house, how wonderful it was 
that the Passion should again come to light exactly a hun- 
dred years after its first performance under Bach, and 
that "an actor and a young Jew" should be accountable 
for it. 

The performance took place on the nth March. The 
chorus numbered about four hundred; the orchestra was 
mostly composed of dilettanti of the Philharmonic Society, 
with leaders of the strings and the wind drawn from 
the Royal band. Sturmer sang the Evangelist's music, 
Devrient that of Jesus; Bader was the Peter, Busolt the 
High Priest and Pilate, Weppler the Judas ; the soprano 
and alto soli were taken by ladies named Schatzel, 
Milder, and Tiirrschmiedt. All gave their services free, 
and relinquished their right to free tickets. The copying- 
out of the parts had been undertaken by Rietz with his 
brother and his brother-in-law; they too refused any 



cantorship. He had previously, when a student, founded a choral 
society in Leipzig. 

The whole history of the memorable performance is given by 
Eduard Devrient in his Meine Evinnerungen an Felix Mendelssohn* 
Bartholdy, 2nd ed., Leipzig, 1872, pp. 48 68. See also a letter 
from Fanny Mendelssohn of 22nd March 1829, given in Hensel's 
Die Familie Mendelssohn, Berlin, 1879, I, 205 210. 

It should also be mentioned that Thibaud had been performing 
Bach with his Heidelberg choral society since 1825. See the preface 
to Vol. XLVI of the B. G. edition, p. 301. 



Revival of the St. Matthew Passion. 243 

honorarium. Fanny Mendelssohn was angry with Spontini 
for having accepted two free tickets. 

Mendelssohn, who was then just twenty years old, con- 
ducted the whole excellently, although it was the first time 
he had stood before a large orchestra and chorus*. In 
accordance with the tradition of the Singakademie he con- 
ducted from the piano, his face turned sideways to the 
audience, so that he had the first choir at his back. To 
humour Devrient he beat time only in the intermezzi and 
the difficult passages, for the rest letting his hand hang 
quietly by the side. 

The audience was transported, not only by the work 
but also by the fine dynamics of the choir, which were 
something unusual in those days. Not less powerful was 
the religious impression made by Bach's music. "The 
crowded hall looked like a church," writes Fanny Mendels- 
sohn. "Every one was filled with the most solemn devo- 
tion; one heard only an occasional involuntary ejacula- 
tion that sprang from deep emotion.'* 

On the aist March, Bach*s birthday, the work was re- 
peated**, Spontini had wanted to prevent a further per- 
formance; but Mendelssohn and Devrient had gone direct 
to the Crown Prince, whose orders the all-powerful ruler 
of the Berlin opera had to obey. The enthusiasm was, 



* The work had been severely "cut" for this performance. The 
majority of the arias were omitted; of others, only the orchestral 
introductions were given ; in the part ol the Evangelist everything 
was left out that did not relate to the Passion. The recitative 
"And the veil of the temple was rent" had been orchestrated by 
Mendelssohn, 

* The proceeds of both performances were devoted to the en- 
dowment of two sewing-schools for poor children. A third per- 
formance was conducted by Zelter himself, Mendelssohn meanwhile 
having gone to England. Thenceforth the Si, Matthew Passion 
was given almost every year in the Singakademie in Holy Week. By 
its side, however, Graun's Tod Je$u maintained itself pertinaciously. 
It was performed almost every year at the same time. Not until 
about the middle of the century was it ousted by Bach's work, 
See Zwr G&schichte dev Singakademie in Berlin; am jttnfgigsten 
Jahrestag ihrw $ti/tung t Berlin, 1843. 

16* 



244 



Deatl1 and Resurrection. 



if possible, even greater than before. Mendelssohn, how- 
ever, was not quite satisfied with the performance; the 
chorus and the orchestra had indeed done excellently, but 
in the soli there had been errors made that put him out 
of humour. 

On that evening a select company of admirers of Bach 
was invited to supper at Zelter's, who was now quite recon- 
ciled with the undertaking. Frau Eduard Devrient sat 
next to a man who seemed to her very affected, being con- 
tinually anxious lest" her wide lace-trimmed sleeve should 
touch the plate. "Do tell me who is the stupid fellow next 
to me," she said softly to Mendelssohn, who sat close by 
her. He held his serviette for a, moment before his mouth, 
and whispered: "The stupid fellow next to you is the famous 
philosopher Hegel*." 

Hegel took the warmest interest in Bach, and took the 
opportunity to refer in his Aesthetic to the master "whose 
grand, truly Protestant, pithy yet learned genius we have 
only lately learned to value again properly." Hegel saw 
in Bach's music the genuine Raphael-like beauty, in that 
it had progressed from the "merely melodic to the charac- 
teristic", though "the melodic remains justified as the sus- 
taining and uniting soul**." In March 1829, while he 
was conducting the rehearsals for the St. Matthew Passion, 
Mendelssohn went to Hegel's lectures on aesthetics, which 
were then dealing with music***. 

For Schopenhauer, who attributed so great a sig- 
nificance to music, Bach did not exist; he did not fit in 
with the philosopher's definition of the nature of music. 

In the early years of the third decade the Si. Matthew 
Passion was produced in a great number of German towns, 
among them Frankfort, Breslau, Konigsberg, Dresden, and 



* Therese Devrient, Jugenderinnwungen, Stuttgart, 1905, p. 309. 
** Hegel, Aesthetik* Part 3 ; Vol. X of his collected works (1838). 
For Bach, see p. 208. 

*** See Zelter's letter of 22nd March 1829 to Goethe, in Goethes 
und Zelters Briefwechsel, III, 124 and 127. 



Results of the Victory. 245 

Cassel. Leipzig did not hear it until 1841, when Mendels- 
sohn was working there*. 

The St. John Passion, which was performed for the first 
time on the 2ist February 1833, in the Berlin Singaka- 
demie, had not the same rapid success. 

The glory of having revived the B minor Mass belongs 
to Schelble (17891837), the founder of the Frankfort 
Caecilienverein. He had performed the Credo as early as 
1828, but nobody had taken any notice of it. In 1831 he 
followed it up with the Kyrie and the Gloria. The Berlin 
Singakademie gave the first part in 1834, and the whole 
work, much curtailed, however, in 1835 ** Schelble 
did not live to hear the performance of the Christmas 
Oratorio that he had projected; it was not given until 1858. 

Considered as a whole, Mendelssohn's victory hardly 
went further than the St. Matthew Passion. The fact that 
the piano and organ works now interested the public more 
was the bye-product of this victory; we must remember, 
in this connection, what Mertdelssohn had done for Bach 
by public performances of these works. The programmes 
of his organ concerts were devoted almost exclusively to 
Bach. It was he who initiated Schumann into the beauty 
of the chorale fantasias. 

Of their favorite chorale prelude, "Schmiicke dich, o Hebe 
Seele", Schumann writes; "Round the cantus firmus hung 
golden garlands of leaves, and it was full of such beatitude 
that you yourself" (i e. Mendelssohn, who had played it 
to him) Confessed to me that if life were to deprive you of 
hope and faith, this one chorale would bring it all back 
again to you***/' 



* See the announcement o this performance in Schumann's 
Neu& Zeitschrift fttr Musih, 1841, No, 25. On 23rd April 1843, on 
the occasion of the unveiling of the Bach monument at St. Thomas's 
school, Mendelssohn gave the Rathswahl cantata of 1723, Preise 
Jerusalem. 

** Zur GeschichU <Ur Sin$<tkad$mie, Berlin, 1843. 
*** Schumann, Musih und Musiker I, 153. A letter of Mendels- 
sohn of 1 4th November 1840, in which he describes to his sister 



246 XII. Death and Resurrection. 

The cantatas still remained forgotten. Until 1843, the 
Singakademie had produced only one of them Perhaps 
it would have been otherwise if Mendelssohn, as he had 
hoped, had become Zelter's successor. Things were rather 
better in Leipzig, where, since the cantorate of August 
Eberhard Mtiller (1801 1810), and more especially under 
his successor. Johann Gottfried Schicht (1810 1823), 
Bach began to be honoured again at St. Thomas's. Men- 
delssohn, when in Leipzig, did a great deal to introduce 
Bach into the concert room. The real Bach epoch for the 
St. Thomas choir began during the cantorate of Merits 
Hauptmann (1842 1868). 

At Frankfort, Schelble produced the Actus tragicus 
(1833) and the cantata Liebster Gott, wann werd ich sterben 
(1843), as if he had a presentiment of the early death 
that was to call him away from his ideal work*. 

In Breslau, Johann Theodor Mosewius (1788 1858) 
performed the cantatas Bin* feste Burg (1835), Gottes Zeit 
(1836), Sei Lob und Ehr (1837), Wer nur den lieben Gott 
Idsst walten (1839), and the first two parts of the Christ- 
mas Oratorio, with the Singakademie founded by him in 
1825. He also published at the same time, in order to make 
these works better known, an essay on /. 5. Bach in semen 
Kirchenkantaten und Choralgesdngen**. Mosewius is the 



Fanny his method of playing the arpeggios in the Chromatic Fan- 
tasia, throws a characteristic light on his comprehension of Bach* 
In the upper part he accents and sustains a melodic note (Brief* 
aus dtn Jahren 1833 *%47* 5th ed. Leipzig, 1865, p. 241). Ac- 
cording to Kretzschinar (B. G. Vol. XLVI, Preface, p. 29) his most 
noteworthy remarks upon Bach are to be found in the still un- 
published letters to Franz Hauser. He particularly liked the can- 
tatas Liebster Gott, wann werd' ich sterben, Christ unser Herr sum 
Jordan kam, Also hat Gott die Weltgeliebt, and/^w, der du meine Seele, 
* On Friday evenings, among friends, he performed a number 
of Bach's cantatas. On his way to Paris, in the autumn of 1831, 
Mendelssohn heard him give the Actus tragicus, the Magnificat, 
and the B minor Mass ,- he writes to Zelter about them. See Hensel's 
Die Fcvmilie Mendelssohn, I, 333. 

** Published in 1845 ; at the end of it there is a long list of Bach 
cantatas. We also owe to Mosewius the first musical- aesthetic 



Mosewius. 



247 



first great Bach aesthetician after Rochlitz, whose ideas 
he resumes. He always dwells upon Bach's penetrating 
musical treatment of the text as the characteristic feature 
of his art. At the same time he recognises the pictorial 
elements in his style, and shows that Bach almost always 
gives a figurative turn to the spiritual meaning of the words, 
Bethinks that the key to Bach's development is to be found 
in his passion for pictorial illustration. As he says 
"Bach represents standing and moving, resting and hurry- 
ing, elevation and depression, with a naivet6 almost char- 
acteristic of the first beginning of art, Without abandoning 
this minute detail-painting in his later works, his method 
now becomes, as it were, transfigured. His thought, 
vision, and emotion have remained unchanged, but in the 
later works the tone-painting is not so isolated; it is part 
and parcel of the melodic form that constitutes the basis 
of his movements, and his genius provided him with themes 
that contain, in their germ, all the possibilities of expres- 
sion that the movement will afterwards require,"* 

In spite of its pictorial character, Mosewius regards 
Bach's music as genuine church music. "Precisely in 
Sebastian Bach," he says, in one place, "we can clearly 
recognise that not this or that style alone can lay claim 
to the title of a church style, but that only a soul filled 
with the holiest and highest can speak the language that 
can bring the most exalted things home to us, and that 
discards the mean and the unworthy,"** He therefore 
concludes that the cantatas are well suited for the church 
service, and would have them performed at the end of 
the sermon. 

Mosewius was the last to criticise Bach impartially; 
after that the composer is drawn into a conflict of opinions 

analysis of the St. Matthew Passion (Berlin, 1852), He gives an 
account of his first productions of Bach's works in bis Geschichte 
der Bre$laui$chn Singahafcemi (1850), 

* P. 7< 
** P, 10. 



248 XII. Death and Resurrection. 

with which he had no concern, and more than a genera- 
tion had to elapse before men could again contemplate 
and criticise him purely as he was. 

The discussion Was mainly upon the question of true 
and false church music. The reformation in church music, 
which, in the middle of the nineteenth century, was every- 
where victorious over the inartistic ideal of pietism and 
rationalism, was not favorable to Bach. It harked back 
to the epoch before him, and condemned his cantatas, in 
common with the whole of the church music of the eight- 
eenth century, as theatrical art, not calculated for the edi- 
fication of the faithful. Carl von Winterfeldt undertook 
the execution of the sentence in his work on Der evangelischc 
Kirchengesang und sein Verhaltnis zur Kunst des Ton- 
satzes*. For him the true church style is that of an Eccard, 
that aims at the objective, not the subjective, expression 
of feeling. Bach, however, in spite of the ample piety 
th^t is evident in his works, is not a church composer, 
since his imagination always runs away with him, his art 
is incomprehensible by the multitude, and he aims at 
being dramatic. "Even the extraordinary impression he 
makes on the souls of his hearers, and the means by which 
he effects this, exclude the wonderful work of Bach from 
the church, which is a place of worship."** Winterfeldt 
did not say this without weighing his words, for he revered 
Bach. He finally consoles himself with the thought that 
he can at any rate exempt the organ works from the 



* 3 vols, 18431847. For Bach, see III, 256 428. Carl von 
Winterfeldt (17841852) was the pioneer of scientific research 
into the history of church music. He has the great merit of having 
depicted Bach, who till then was regarded as an isolated phenom- 
enon* in his relation to the work and the ideals of his epoch, 
and of having thrown some light on the history of the cantatas 
and Passions. But it was a fairly long* time before musicians got 
rid of the old unhistorical ideas upon the church music of Bach's 
time, as is abundantly clear from the remarks of Wagner, for ex- 
ample, upon Bach. 

** Hegel, in his Aesthetic, repeatedly protests against this narrow 
conception of church music. 



The Publication of Bach's Works. 



249 



condemnation, not seeing as indeed Rochlitz and Mo- 
sewius before him had failed to see that the chorale 
fantasias are as pictorial in their conception as the cantatas. 

Thus the church doors were closed to Bach. The church 
choirs refused to do for the cantatas what the oratorio 
societies had done in the concert room for the Passions; 
while it did not occur to these societies to bring to light 
the treasures hidden in the cantatas, these being works 
with nothing of the oratorio about them, and not being 
long enough to fill a programme. 

Even greater obstacles were placed in the way of the 
resuscitation of Bach's music by the controversy upon 
modern and classical art that sprang up around the work 
of Wagner. The consequent neglect of Bach was almost 
the least effect of this controversy, and in itself, indeed, 
was quite natural, for nowhere has the present such a 
right to be its own arbiter as in art. What did him 
harm, however, was the narrow definition of the classical 
that was put forward in opposition to the Wagnerian style 
and to Wagner's manner of interpreting Beethoven. The 
conservative party maintained that true classical music 
should concern itself only with perfect form and the ex- 
pression of indefinite feeling, and prove its true greatness 
by avoiding drastic tone-painting and far-reaching poetic 
pretensions. Bach was an old musician; therefore he was 
a classical musician; therefore he could not have thought 
otherwise than as one was entitled to assume the classical 
masters thought ; thus he was a witness against Wagner. 
This thoughtless and polemical attitude was accountable 
for people not trying to find the real Bach, and this 
just at the time when his works were at last made ac- 
cessible to the world. 

The earlier history of the publication of Bach's works 
is an unpleasant story. None of the hopes were fulfilled 
that had been built on the scheme for the publication of 
the complete works, at the beginning of the nineteenth 
century, by Hoffmeister and Kiihnel (afterwards Peters). 



250 XII. Death and Resurrection. 

Two other publishers, Simrock and Nageli, had had similar 
plans, but their promises were unredeemed. They could 
only bring out the works for which there was a market, 
i. e., pianoforte and instrumental compositions*. The dif- 
ficulties in the way of issuing the cantatas were realised 
by Breitkopf and Hartel when, in 1821, they published 
Bin* feste Burg at iVs thalers per copy. In 1829 Zelter 
writes to Goethe that they regarded the work as a "drug"**. 
Such was the fate of the first cantata of Bach's that was 
offered to the German public. More success fell to the lot 
of the six motets which the same firm had published in 
1803 at the instance of Schicht, who afterwards became 
a St. Thomas cantor. The publication of the Magnificat 
(Simrock 1811, in Eflat major instead of D major) went 
almost unnoticed***. 

After the Berlin performance of the St. Matthew Passion, 
circumstances seemed to improve; in 1830 the score of 
this work was published by Schlesinger; in the same year 
Simrock printed six cantatas Nimm von uns Herr; Herr 
deine Augen; Ihr werdet weinen und heulen; Du Hirte 



* On the whole subject see Hermann Kretzschmar's account 
of the Bachgesellschaft, in his preface to vol. XL VI of the great 
Bach Edition. 

The Well-tempered Clavichord was first issued by Simrock of 
Bonn, in 1800; then came editions by Hoffmeister and Kiihnel 
(Peters) and Nageli. It was soon pirated in Paris. In England 
F. A. Kollmann, an admirer of Bach, had issued an edition of this 
work as early as 1799. About the same time appeared also the 
Inventions and the suites, pieces for the violin, and separate organ 
works. Peters followed most tenaciously the plan ,pf publishing 
the "CEuvres completes", by which, however, was understood 
from the beginning only the instrumental works. It was of the great- 
est significance that this firm brought out, in the middle of the 
forties, the complete organ works in a critically correct edition 
by Griepenkerl and Roitsch, by means of which Bach became widely 
known among organists. 

** Briefwechsel Goethes und Zelters, III, 99. 
*** In 1818 Nageli invited subscriptions for the B minor Mass, 
but without result. The Kyrie and the Gloria appeared in 1833 ; 
the second part was issued by Simrock' in 1845. Simrock had pub- 
lished the small Mass in A major in 1818. 



Foundation of the Bachgeseilschaft, 2 "5 1 

Israels; Hen, geJie nicht ins Gericht; and Gottes Zeit: in 
1831 Trautwcin brought out the St. John Passion. Then 
the movement was once more checked*. 

Thenceforward no doubt could exist among musicians 
that if it were left to the publishers alone the complete 
Bach would never appear, but that the work would have 
to be taken in hand by the community of Bach lovers. 
Schelble wrote to this effect to Mendelssohn's friend, the 
singer Franz Hauser (1794 1870), who had a large collec- 
tion of Bach axitographs and copies. In 1837 Schumann, 
who had done so much for Bach with his pen, inquired in 
the Neue Zeitschrift filr Musik whether "it would not be 
an opportune and useful thing if the German nation were 
to resolve upon the publication of the complete works of 
Bach"; and he referred to two letters of Beethoven that 
had just been made public, in which the composer had 
congratulated the publisher Hoffmeister on his projected 
edition of Bach**. When, in 1843, the formation of the 
English Handel Society was announced, Schumann re- 
marked in his journal that the time was no longer distant 
"when the plan of a complete Bach edition might be laid 
before the public,"*** In July 1850 the Bachgesellschaft 
came into being. At its head were Moritz Hauptmann, 
(then cantor at St. Thomas's), Otto Jahn (the biographer 
of Mozart, and professor of archaeology at Leipzig), Karl 
Ferdinand Becker (professor of the organ at the Leipzig 



* The Peasant Cantata and the Coffee Cantata were published 
in 1837 at the instance o{ Dehn. Between 1840 and 1850 appeared 
seven other cantatas Nimm, was dein ist; Himwclskdnig, sei 
millkontmcn; ftarmhermgcs XXerze; and Siehe zu> 9 dass deine Gottes- 
/urcht> published by Trautwcin (1843); Warum betrUbst dn dioh; 
Wachct auf; and Also hat Gott die Wdt gdiebt, published by Broil- 
kopf and H artel (1847)^ as supplement to the third volume ot 
Wiatcrfclclt'H Evangclischcr JKwchcngesang. 

** See Schumann's Sohriftcn ubev Miwik und jMfatsiher, II, 103, 
104. The two letters of "Beethoven appeared in the same volume 
of the Ncuc Zcitsohrift /rt/ Musik in which Sclwmann agitated Cor 
the "Bach edition, 

*** Ntuc Zcitschrift /ur Mitsik* XIX, 87. 



252 XII. Death and Resurrection. 

Conservatoire) and Schumann. The printing and the 
financial arrangements were undertaken by Breitkopf and 
Hartel. 

From the commencement the undertaking had the 
greatest difficulties to contend against. There really 
should have been some years of preparatory labour, in 
order to sift the material and to draw up a clear plan for 
the edition *. It was feared, however, that public interest 
might slacken if something were not issued at once; con- 
sequently a start was made with what happened to be 
ready. In this way an element of disorder crept in that 
was never afterwards mastered. The editors lived from 
hand to mouth. Thus the B minor Mass and the French 



* When Rochlitz was writing about Bach, he expressed the 
fear that most of the cantatas must be regarded as irretrievably 
lost. He did not know that Berlin, where the influence of Kira- 
berger and Marpurg had lasted for a long time, had become the 
rallying point for Bach manuscripts. This fact was of the utmost 
significance for the publication of Bach's works. The chief private 
collectors of Bach manuscripts, besides Forkel and Hauser, were 
Chr. Fr. Schwenke (1767 1822), the successor of Philipp Emmanuel 
Bach at Hamburg; J. G. Schicht (1753 1823), cantor of St. Tho- 
mas's at Leipzig; and Georg Polchau (1773 18 3^) who bought 
Philipp Emmanuel's library. Polchau was librarian to the Sing- 
akademie from 1833 1836. The Royal Library at Berlin pur- 
chased his private collection in 1841 for 8000 thalers; it acquired 
also the library of the Singakademie, with its rich stores of Bach, 
The largest private collection of Bach manuscripts and copies at 
present is that of the singer Joseph Hauser, in Karlsruhe. It is 
not impossible that there are still some hitherto unpublished 
works of Bach in private hands in England, 

The progress of the Bach cult in the forties may be measured by 
the rise in the value of manuscripts, which formerly had been so 
low that in 1824, at the sale of Schwenke's effects, the autograph 
of the Magnificat could be bought for seven Hamburg marks. 
Valuable preliminary material for an edition of Bach's works 
existed in the catalogue of Bachiana of the Royal Library in Berlin, 
published in 1845 by the custodian of the musical department of 
the library, S. W. Dehn (17991858), in his journal Caecilia, and 
the complete list of Bach's works, comprehending 672 numbers, 
which Hauser began in the thirties. Mendelssohn took the warmest 
interest in this work of his friend, and helped him by sending Mm f 
in February 1834, a list of all the Bach compositions then in Berlin. 
This catalogue was not printed. 



History of the Bachgesellschaft Edition. 253 

and English Suites were published without any reference to 
the oldest versions for the Mass, the autograph formerly 
in Nageli's possession, now in the Royal Library in Berlin 
which of course necessitated new and corrected editions. 

At first the members of the Committee had thought 
that voluntary labour would be sufficient to see the edition 
through; but it soon became evident that the task de- 
manded the whole strength and the whole time of some one. 
From the ninth year onward the work devolved upon 
Wilhelm Rust (1822 1892), the grandson of the well- 
known composer Friedrich Wilhelm Rust of Dresden 
(1739 1796), who, with Emmanuel Bach, plays an im- 
portant part in the history of the pianoforte sonata. He 
superintended the issue with ideal devotion from the ninth 
until the twenty-eighth year. The prefaces which he con- 
tributed to the separate volumes are sometimes masterly, 
They deal not only with critical and historical matters, 
but with purely practical questions relating to the manner 
of performing Bach's music. Towards the end he fell off; 
the task exceeded the powers of one man; the responsibility 
also was too great for a single person to bear. In 1882 
he surrended the editorship, in order to be more equal to 
his new duties as cantor at St. Thomas's, to which post he 
had been appointed in 1880*. In his place came new 
forces Dorffel, Count Waldersee, Naumann, Wiillner. 
These completed the difficult work in accordance with a 
pre-arranged scheme. 

On the 27th January 1900 the completed final volume 
year XLVIth was laid before the Committee, on 
which not one of the original founders of the Bachgesell- 
schaft was represented. 



* Between Hauptmann and Rust, the office had been held by 
Ernst Friedrich Bduard Richter (1868 1879). Rust's retirement 
was partly duo to some criticisms, not wholly unjustified, by Spitta, 
(who in 1880 had finished his biography of Bach), in a memorial 
to the Committee of the Bachgesellschaft In 1888 Rust left the 
Committee also, The valuable matter of his prefaces sometimes 
suffers by his rather obscure and incoherent style, 



254 XII. Death and Resurrection. 

To the very end the work had been carried on in face 
of an apathetic public. The number of subscribers, of 
whom there were three hundred and fifty at the end of 
the first year, did not increase; without the enthusiastic 
labours of Franz Liszt and Hauser, who exerted them- 
selves to fill up the gaps as they arose, the number would 
not even have remained at that. The financial situation 
was always so bad that the question of the continuance 
of the Society was raised time after time. Only a few artists 
realised the magnitude of the undertaking of the Bach- 
gesellschaft. Among them was Brahms, who used to say 
that the two greatest events during his lifetime were the 
founding of the German Empire and the completion of 
the Bach edition. The church choirs, whose help had been 
counted on, did absolutely nothing. 

It must, however, be admitted that the method of 
publication was the most unpractical that could have been 
devised. People had to subscribe for the whole edition, 
and pay for each volume in advance. Single volumes 
could not be bought separately. The Society thus threw 
away the good business they might have done with separate 
issues that were in general request such as the Passions 
and the piano and organ works; and the general public 
was not brought into touch with the weighty undertaking. 
When at last, in 1869, it was resolved to sell the volumes 
separately, at thirty marks per volume, i. e. double the 
subscription price it was too late. The press practically 
ignored the work. The history of the publication of Bach's 
works is thus a repetition of the history of his own life. 

The undertaking of the Bachgesellschaft was supple- 
mented by the work of one man, Spitta's Life of Bach, 
of which the first volume appeared in 1874, the second in 
1880*. For the first time the world had a real biography 

* Philipp Spitta (18411894), son of the poet of Psalter und 
Harfe, was originally a philologist. In 1875 he became professor 
of musical history at the university of Berlin. His book was trans- 
lated into English in 1899. 



Spitta's Biography, 255 

of the master. Before ForkePs book, writers on the sub- 
ject had simply reproduced the Necrology with more or 
less of their own fantasy; after Forkel, they were content 
to gather together the existing information. No one had 
issued a work adequate to the historical questions involved; 
Bitter (1813 1885), afterwards Prussian minister of finance, 
had indeed set himself this aim in his book on Bach (1865), 
without succeeding in getting beyond dilettantism*. 
Spitta's work is really a unique performance among artis- 
tic biographies. It is rarely that the first scientific inves- 
tigator of an art epoch leaves so little for his successors to 
do as in this case. He not only awoke Bach to new life 
but vivified the whole world in which Bach had worked. 
It was, indeed, not a work for the average reader, nor a 
book for musicians to refer to casually. For this it was 
too exclusively scientific and not always simple and clear 
enough in its plan, the author having worked too closely 
together the story of Bach's life, an analysis of his works, 
and an account of contemporary art. Only those who 
had the time and the enthusiasm to follow him in his 
tortuous path could really appreciate the depths and the 
many beauties of the work, ft was predestined to serve 
as a storehouse of material for writers of Bach biographies 
of the popular kind. 

Nor was it wholly satisfactory to musicians on the aes- 
thetic side. It contains, indeed, many admirable ana- 
lyses, couched in poetic language; there are some which 
no one who reads them can ever forget. But the aesthetic 
view-point is too subordinate to the historical, and owing 
to the plan of the book the essential artistic quality of 
Bach's art is never presented "as a whole. Ren de R6cy 
has formulated this reproach most clearly in the Remw 



* Spitta discusses his predecessors in the preface to his first 
volume, pp. VII if, C. L* Hilgenleldt's /. S. Bach (Leipzig, 1850) 
also deserves mention. A second edition of Bitter's Bach, in two 
volumes, appeared m 1880, It: must be admitted that this book 
did a great deal to make Bach better known. 



256 XII. Death and Resurrection. 

des deux Mondes for 1885. Spitta saw that Bach-aesthetic 
was no longer so simple as in the time of Mosewius; he is 
too obviously bent on holding up the cantor of St. Thomas's, 
as representative of pure music, as an exemplar to the 
erring artists of his own day. The historical inquiry had 
prejudiced the aesthetic. So it was, again, with the sub- 
sequent biographies that were based on Spitta. They do 
not complete him on the artistic side in the desired way, 
and they are too much under his influence in another 
respect, following him in the plan of mixing up the bio- 
graphy with analyses of the works. There is not the slight- 
est reason for this with Bach. In the case of no other artist 
has the external course of his life so little to do with the 
origin of his works, or is what we know of his life so in- 
significant, and, as regards his personal experiences, so 
uninteresting. Still we are bound to recognise how much 
these popular biographies of Bach have done for him** 
Those whom, according to Hauptmann, the promoters 
of the collected edition of Bach's works had principally 
in view, actually profited by it least. Though correct 
scores of the cantatas were/ from the very beginning, 
issued at the rate of something like ten each year, per- 
formances of them were hardly more frequent than before. 



* A list of them will be found in the index to Bach literature 
given by Schneider in the Bachjahrbuch for 1905. Among them 
may be mentioned Reissmann's (1880), Otto Gumprecht's (1883), 
William Cart's (in French, 1885), R. Batka's (Reclam, 1893), and 
H. Earth's (1902). Philipp Wolfrum (1906) deals for the most 
part only with the instrumental works, and for the cantatas promises 
a future volume a limitation of scope that should have been 
indicated in the title of the book. As it is, it resembles an American 
high road, that comes to a sudden end in the middle of a field. 
The motto of the book runs thus, "Only so far as history serves 
life will we serve it," which every artist will echo from his heart. 
The latest biography of Bach is the French onebyPirro (1906). 
The real progress of historical research is less evident in all these 
books than in a number of striking essays that have appeared in 
various journals, which have been cited in the preceding pages of 
the present work. The "Bachjahrbucher' 1 have latterly become the 
rallying point for essays of this kind. 



Liszt and Wagner. 257 

Matters did not improve in this respect until the Bach 
societies sprang up in various towns, In Vienna, whither 
the St. Matthew Passion had penetrated in 1862, Brahms, 
as conductor of the local Singverein, exerted himself on 
behalf of the cantatas. In our own time Robert Franz 
fought for them with his pen.* The Passions, after about 
1860, were taken up in most towns. 

For the piano compositions Franz Liszt continued the 
work that Mendelssohn had begun, and, by brilliant 
transcriptions of the organ compositions, especially of the 
G minor and A minor fugues, forced Bach as an organ 
composer on the public attention. The Peters edition 
carried the preludes, fugues, and chorale preludes into 
every church about the middle of the forties. A landmark 
in the victorious course of Bach was afforded by the in- 
auguration of the Eisenach memorial in 1885, where the 
reverence of the artists who assembled round Liszt found 
public expression, 

It is well known that Wagner was an admirer of Bach. 
He regarded him as the great teacher of Beethoven, who 
cut himself loose from Haydn as the youth developed into 
the man. He has thus expressed, in his essay "What is 
German?" the significance of Bach for German spiritual 
life "If we would comprehend the wonderful originality, 
strength and significance of the German mind in one incom- 
parably eloquent image, we must look keenly and discern- 
ingly at the appearance, otherwise almost inexplicably 
mysterious, of the musical marvel Sebastian Bach. He is 
the history of the inner life of the German mind during 
the awful century when the German people was utterly 
extinguished. Look at this head, hidden in its absurd 
French full-bottomed wig, look at this master, a miserable 
cantor and organist in little Thuringian towns whose 
names we hardly know now, wearing himself out in poor 



* See Neue Zeitschrift filr Musik, vol. 47, pp. 49 52 (1857) 
' 'Remarks upon Bach's cantatas, occasioned "by a performance 
of several of them by the Halle Singakademi/ 1 

Schweitzer, Bach, 17 



258 XII. Death and Resurrection, 

situations, always so little considered that it needed a 
whole century after his death to rescue his works from 
oblivion; even in his music taking up with an art -form 
which externally was the complete likeness of his epoch, 
dry, stiff, pedantic, like perruques and pigtails in notes; 
and see now the world the incomprehensibly great Se- 
bastian built up out of these elements ! To these creations 
I only refer briefly, since it is impossible to characterise 
by any comparison whatever their wealth, their grandeur, 
and their all-embracing significance."* 

Unfortunately Wagner nowhere discusses the nature of 
Bach's art more thoroughly, or fixes his aesthetic im- 
pressions of it. Nor must it be overlooked that, for Wagner, , 
Bach's cantatas hardly ranked as true church music, of 
which his own ideal was pure choral song with occasional 
organ accompaniment. He regards the addition of instru- 
ments as the beginning of the decline of this branch of 
the art, which explains why Wagner often speaks of 
Bach's motets, but hardly mentions the cantatas**. 

But it was in his works, rather than his words, that he 
prepared the way for Bach. From them the world learned 
again to look for a profound inner relation between word 
and tone in the musical setting of poetry. The outcome 
of Wagner's art was a revolution of the whole musical con- 
sciousness. The hearer became exigent. Henceforth only 
the truly characteristic in music could satisfy him, only 
the truly dramatic could move him. Thus a whole mass 
of music sank slowly into the abyss of oblivion; and by 
the side of the music drama of Wagner the dramatic 
religious music of Bach came out in a clear light. Warring 
as Wagner did against "the beautiful" in music, he was 
at the same time, though unconsciously, fighting for Bach, 
whose "pithy and often most poetical conception of the 
text" in the motets astonished him. It is only now, after 

* Gesammelte Werke, X, 65, 66. The essay was written in 1865. 
** He discusses the Passions in vol. I/p. 169 if. of his Gesam- 
Werke. 



Bach in France, England and Italy, 259 

the strife is over, that we can see the importance of the vic- 
tory. The magnitude of the change of view makes it 
wholly incomprehensible to us how the post-Beethovenian 
epoch could remain insensible to the greatness of Bach, 
and how even those who planned the great Bach edition 
could make a distinction between the "pleasing" and the 
"unpleasing" works. 

Bach numbered from the beginning many admirers 
among French musicians, as was shown by their ardent 
co-operation in the subscription for the collected edition. 
Among them were Gounod, whose understanding of 
Bach must not be estimated merely by his dubious ar- 
rangement of the C major prelude, and the older school of 
French organists, Saint-Sae'ns must indubitably be reckon- 
ed among the best Bach connoisseurs ; and the same may 
be said of Gabriel Faure". Guilmant, Widor and Gigout, 
the creators of modern French organ music, derive di- 
rectly from Bach. The violinist Charles Bouvet, with his 
little Bach Society, worked hard for the instrumental 
works. The Paris performance of the St. Matthew Passion 
by the Concordia Society under Widor, in 1885, was a 
decisive point in the public recognition of Bach. The Schola 
Cantorum, under Vincent d'Indy and Bord&s, gave ex- 
cellent performances of some of the cantatas; this, again, 
was the special object of the Paris Bach Society, begun 
under the auspices of Faure" and Widor, the conductor 
of which, Gustave Bret, is chiefly bent on organising 
a capable choir perhaps the most difficult of undertak- 
ings in Paris, The lack of rapced clioirs is, generally speak- 
ing, the greatest obstruction to the diffusion of Bach's 
works in France; many years may still elapse before a 
change is made in this respect and the Bach of the can- 
tatas becomes more widely known, Excellent Bach trans- 
lations have been made by, among others, Maurice Bouchor 
and Madame Henrietta Fuchs*. A uniform translation of 



* See also Gevaort's French edition of the St. Matthew Passion 
(published by Lemome, Paris and Brussels). 

17* 



26o XII. Death and Resurrection. 

the whole of the cantatas is being prepared by G. Bret, 
the conductor of the Bach Society. 

In France, as a matter of fact, we can best realise how 
Wagner had prepared the way for Bach. The enthusiasm 
for Bach sprang up when the Wagner enthusiasm, that had 
finally become little more than a fashion, had spent itself, 
and the conviction arose that there was something alien 
to French artistic feeling in Wagner's union of poetry and 
music. It was precisely at this stage that Bach came on 
the scene. Every year it becomes more obvious how 
largely his system of musical characterisation coincides 
with French artistic perceptions. His characterisation is 
wholly plastic; whereas the monumental formlessness of 
Wagner sets an ever-widening gulf between him and the 
French genius. Even French military music has come 
under the influence of Bach. In order to make the chorale 
preludes accessible to the people, M. Th. Barnier, chef dc 
musique of the 57th regiment of infantry in Bordeaux, 
has arranged them for military band, and plays them at 
promenade concerts. 

England has this advantage over France, that it pos- 
sesses exceptionally fine choruses. The future will show 
how the contest there between Bach and Handel will end. 
That the musical saint of the English is regarded on the 
continent as decidedly inferior to Bach cannot be denied; 
and the general experience is that cultivation of the can- 
tatas of the one leads to something like injustice to the 
oratprios of the other, which, a couple of decades ago, 
completely dominated the musical world. Mendelssohn 
in this respect shares the fate of Handel*. 

In Belgium the indefatigable Gevaert fought Bach's 
battle with great success. 



* For information as to the progress of Bach's music in England 
see Grove's Dictionary of Music and Musicians, arts. "Bach Society", 
"Bach Choir", and the appendix (in the fifth edition) to the article 
on Johann Sebastian Bach. See also the Musical Times, Sept. 
to Dec. 1896. [Translator's note.] 



Bach's Position Today. 261 

Bach triumphed also in Rome. At first many of his works 
were produced privately by Hcrr von Keudell, Fran Pro- 
fessor Helbig, and Frau Dr, Mengarini. Alessandro Costa 
rehearsed the B minor Mass with a small chorus recruited 
from these circles. In the spring of 1889 the whole of 
higher Roman society was invited to a performance of the 
Mass in the oratory in the Via Belsiana. In his Trionfo 
della Morte (1894), Gabriele d'Annunzio gives a picture 
of the public at this memorable performance and the im- 
pression it made. The Roman Bach Society dates from 
the year 1895*. 

All this, however, constitutes only the external history 
of Bach's victory; to estimate its true magnitude we must 
look in the scores of the composers of the nineteenth cen- 
tury. Since Mendelssohn, every composer of any signific- 
ance has been to school to Bach, not as a pedantic teacher, 
but to one who impels them to strive after the truest and 
clearest expression, and to achieve impressiveness not by 
the wealth of the means they employ but by the pregnancy 
of their themes. In Wagner, the spirit of Bach is most 
evident in the score of the Meistersinger. An interesting 
Bach-renaissance is visible in the consummate polyphony 
of Max Reger* How Bach will influence modern orchestral 
composition cannot yet be seen; only this much is clear, 
that he will lead us back to a certain simplicity, and will 
develop in a quite extraordinary way the sense of form of 
future generations**. 

As regards the present-day esteem of Bach, we must 
beware of taking all verbal enthusiasm for reality. Since 
it ceased to be a risk, and became a recommendation, to 
swear by Bach, lip-service has been plentiful. Much of 
what is said of him represents no personal experience at all, 
but is a mere echo of the experience of others. How far 
have we really got? 



* 17 Trionfo della Morte, pp. 41 fl 

* * Interesting opinions of leading musicians upon Bach may be seen 
in the replies to queries in Die Musik, Year V, Part I, 1905 1906. 



XII. Death and Resurrection. 

He is clearly influencing domestic music, This is beyond 
dispute. The Inventions, the Suites and the Well-tempered 
Clavichord have become the property of the people. What 
the average amateur of the present day lacks in theoretical 
musical education is supplied for him by these works of 
Bach, from which he unconsciously imbibes certain prin- 
ciples of thematic formation, of part writing, of modula- 
tion and of construction, and from which he acquires a 
certain unconscious critical faculty, that protects him 
against inferior art. 

As regards our public music the conditions are not so 
satisfactory. To expect to hear the complete Bach in our 
concert rooms would be to experience many disappoint- 
ments. Our pianoforte virtuosi give us transcriptions of 
the organ works rather than original piano compositions, 
on what grounds is not apparent. Why must it always be 
the A minor prelude and fugue that is given to the public? 
Even in Liszt's arrangement they are merely makeshifts 
on the piano. Where can we hear, except rarely, per- 
formances of the suites, the WM-tewpered Clavichord, the 
Italian Concerto, the Chromatic Fantasia, the piano con- 
certo in A minor, the C major concerto for two pianos? 
Where are the Brandenburg orchestral concertos and the 
orchestral suites securely fixed in our programmes? Where 
are Bach's secular cantatas regularly given? Statistics 
of Bach performances in our concert programmes would 
bring some curious facts to light, and would show that there 
are not too many towns where the auditor can really get 
to know Bach. 

The church cantatas stand in a category of their own. Even 
where the Passions are regularly given, there are certain 
difficulties in the way of producing the cantatas. It looks 
as if their title were against them. Many conductors who 
have vowed themselves to the service of Bach think it in- 
judicious to place cantatas on their programmes, with 
the exception of the one or two that have become classics. 
They do not strike these people as sufficiently decorative. 



Bach's Position Today. 263 

When, however, one of them takes it into his head to devote 
a whole evening to the cantatas, he has to invoke all the 
muses and all the saints against his choral committee, 
who are afraid that the programme will not draw, or that 
it offers too little variety, One often believes oneself back 
in the days it was in 1858 when so excellent an in- 
stitution as the Hamburg Bach Society, terrified lest there 
might be too much Bach in the programme, arranged 
it in this way; An eight-part motet by Bach; Chopin's 
Berceuse; the u Hall of Song" scene from Tannhduser, 
(in a piano arrangement); Bach's chorale fe$u meine 



Latterly the question has been sharply debated whether 
the cantatas should be given elsewhere than in church. 
A paper read at the second Bach Festival (1904) took for 
its motto : "The church works of Bach for the church"* 1 ". 
The demand seems at bottom just, and yet is false. Bach's 
cantatas today could only in quite exceptional cases be 
given in the course of the church service, and it is neither 
to be expected nor to be wished that the service should be 
so altered as to restore the old Leipzig conditions. The 
independent position of music in the ritual at that time 
was, as a matter of fact, something quite unnatural in 
itself, and only explicable by peculiar historical circum- 
stances. The evolution of things has led to a separation 
between the church service and art that is good for both 
of them, we have the service on one side and on the other 
the sacred concert, or whatever name people may prefer 
to call it by, The ideal for the present day is really a 
sacred concert, composed of three or four Bach cantatas, 
selected for the appropriateness of their text to the eccle- 
siastical season. Pure Bach services of this kind should 
have the preference over the liturgical celebrations that 
are grouped round a Bach cantata. Yet here again opinions 



* Se Joseph Sittard's GeschiMe des Musik- und 
wesens in Hamburg, Leipzig, 1890, 
** Bachjahrbuch 1904, p. 25. 



264 XII. Death and Resurrection. 

differ *. It should be observed, however, that under these 
circumstances the church must no longer be regarded as 
a sacred place. If the church, for any reason, is not avail- 
able, the performances can be transferred to the concert 
room without their religious character being affected there- 
by. How can Bach help it if churches are often so built 
today that no chorus and no orchestra can be placed in 
them, or only in such a way that the chorus sings into the 
backs of the audience? The great point is that Bach, like 
every lofty religious mind, belongs not to the church but 
to religious humanity, and that any room becomes a church 
in which his sacred works are performed and listened 
to with devotion. It follows that everything that might 
disturb the audience must be avoided, and that a single 
cantata in a programme of other works is no use. Either 
a cantata evening or no cantata at all! Wherever such 
cantata evenings have been ventured upon, their success 
has shown that all the fears for them, real or imaginary, 
were groundless. 

In comparison with the cantatas, everything else that 
Bach has done appears as hardly more than a supplement. 
So long as the public has heard only the Passions, the 
Mass and the Christmas Oratorio, we cannot say that the 
whole Bach is ours again. He is not yet known and will 
not be until then. History and criticism have done almost 
all for him that can be done. It is time for the aesthetic 
to take the place of the historical, time to try to comprehend 
the nature of Bach's art in its whole depth and its rich 
multifariousness**. The necessity also becomes more and 

* Waldemar Voigt, of Gottingen, (see the Bachjahrbuch for 
1904, pp. 41 and 42), was forced by external circumstances to 
recognise the "ideal" Bach services as the best. At St. Wilhelm's 
in Strassburg, under the conductorship of Professor Miinch, several 
musical festivals of this kind, consisting only of Bach's cantatas, 
have been held annually for more than twenty years, 

** See Kretzschmar's preface to Vol. XLVI of the Bachgesell- 
schaft edition. In the introduction to the Bachjahrbuch for 1905, 
Arnold Sobering speaks earnestly on this subject to all true lovers 
of Bacli. 



Bach Festivals. 265 

more urgent for more exact investigation into the musical 
practice of Bach's time. From this quarter much interest- 
ing light is yet to be thrown. We need it. The deeper we 
go into the question of how Bach should be performed, 
the more complicated it becomes, It breaks up into a 
number of questions of detail, which can be solved only by 
historical gleanings and by ever-repeated practical trials. 
For the elucidation of these questions, as well as for 
the spreading of a wider general knowledge of Bach, much 
is to be hoped for from the triennial Bach Festivals planned 
by the Bachgesellschaft. The hope seems justifiable. The 
four that have already been held have done much to 
stimulate interest*. It is certain, however, that Bach 
Festivals, and everything else that we can do ad gloriam 
Bachi, are not what are finally needed most, but the quiet, 
modest work of thousands of unknown men, who go to 
Bach for nothing more than their own inner satisfaction, 
and love to communicate these riches to their neighbours. 
Only to people like these will he truly reveal himself. 



CHAPTER XIIL 
THE ORGAN WORKS. 

Bachge&eUschaft Edition. 

Vol. XV, Sonatas, The great Preludes and Fugues. Toc- 
catas. Passacaglia. 

- XXXVIII 1 . Preludes, Fugues, and Fantasias. Eight 

small Preludes and Fugues. 

- XXXVIDX Concertos after Vivaldi, 

- XXVX The OvgdMtohlti*. The ScMbler Chorales. The 

Eighteen Chorales, 

- Ill, The Preludes upon the Catechism Hymns. 

- XL. Detached Chorale Preludes and the Chorale Varia- 

tions. 



* Berlin 1901, Leipzig 1904, Eisenach 1907, Duisburg 1910. 
The new Bachgesellschaft was founded on the 2/th January 1900, 
on which date the old Society was dissolved, having fulfilled its 
task, the publication of Bach's works. The Bach Museum at 
Eisenach is* the property of the new Gesellschaft. 



266 XIII. The Organ Works. 

Peters Edition. 

Vol. I. Sonatas. Passacaglia. Pastorale. 

- II, III and IV, Preludes and Fugues. 

- V. Small Chorale Preludes and Chorale Variations. 

- VI and VII. Large Chorale Preludes. 

- VIII. Concertos after Vivaldi. Eight small Preludes and 

Fugues. 

- IX. Supplement. 

Of all his preludes and fugues for the organ, only the 
prelude and fugue in E flat, in the third part of the Klavier- 
ubung, was published by Bach. Everything else of this 
order has come down to us in manuscript, either in auto- 
graphs, comprising about a third, or in copies, or even 
in copies of copies. It is really wonderful that under these 
circumstances more should not have been lost. We pos- 
sess the F minor prelude 




and its fugue, for example, only in a copy made by a pupil 
of Kittel I We owe the preservation of the great C minor 
fantasia 



to Krebs, who, as he notes at the end of it, made a copy 
of it on the loth January 1751, a few months after the 
master's death. This manuscript had almost fallen into 
the hands of a shopkeeper for use as waste-paper. Bene- 
ficent fate willed that it should be rescued by Reichardt, 
court organist in Altenburg. The Russian pianist Palschau, 
of Petersburg, deserves mention as a kind of unfaithful 
steward among copyists; he undertook to improve the 
Dorian toccata, and to this end behaved towards it, as 
Rust says, like a Russian censor. 

The majority of the organ compositions belong to the 
Weimar and pre- Weimar periods; at Cothen and in the 



Chronological Evidences. 267 

first Leipzig period Bach appears only occasionally to have 
written for this instrument. Afterwards, however, about 
1735, the first love revives in him, and he writes the gigantic 
organ works of his latest and maturest period. In the same 
period he sifted and revised his earlier compositions, during 
which occupation he was overtaken by death. 

It is only in exceptional cases that we can determine 
precisely the date of composition of the separate preludes 
and fugues. Spitta thinks that the great G major prelude 
with its fugue (Peters II, No, 2; B. G. XV, No. n) belongs 
to 1724 or 1725, and the C major prelude (Peters II, No. i; 
B. G. XV, No. 15) to 1730. He bases this opinion on the fact 
that the water mark of the paper used for the autograph is 
the same as that in the scores of the cantatas of those years. 

As a rule we have to rely on internal chronological 
evidences. Fortunately they are fairly clear. For every 
discriminating player the preludes and fugues, on closer 
acquaintance, fall into four groups : the works in which Bach 
is still under the influence of contemporary masters, those 
in which his independent mastery is evident, the consum- 
mate compositions of the Weimar period, and the final works. 

There are about a dozen preludes and fugues in which Bach 
reveals himself as a gifted pupil of Frescobaldi and Buxte- 
hude*. The storm and stress of the whole of the early 

* Prelude and fugue in G major (Peters IV, No. 2, C minor 
(Peters IV, No. 5), C major (Peters III, No. 7), G minor (Peters III, 
No. 5 ; B. G. XV, No* 5), A minor (Peters III, No. 9), toccata and 
fugue in E major (B. G* XV, p. 276), fugue in C minor (Peters IV, 
No. 9), prelude in A minor (Peters IV, No. 13), fugue in C minor 
on a theme by Legrenzi (Peters IV, No. 6), fugue in B minor on a 
theme by Corelli (Peters IV, No. 8), fantasia in G major (Peters 
IV, No. u) carwona (Peters IV, No, io) prelude and fugue in 
D minor (Peters III, No. 4), and pastoral in F major (Peters I, 
No. 86). The four last-named works rank much above the others, 
and already have a certain perfection. The theme of the D minor 
fugue is identical with that on which the fugue in the first sonata 
for violin solo (G minor) is constructed, the priority belonging to 
the violin fugue. The charming pastorale was one of Mendelssohn's 
favorite pieces. In the Peters edition some little piano pieces are 
appended to it, with which it has no connection. 



268 XIII. The Organ Works. 

organ art comes to life again in these works; the preludes 
have a kind of dramatic excitement, and are somewhat 
spasmodic and lacking in unity; the fugues are often con- 
fused; but the proportions on which the works are laid 
out give us the feeling that the future promises something 
great. 

Bach owed his development not only to his perpetually 
improving organ technique, but before all to the study of 
Legrenzi, CoreUi and Vivaldi, whose music was just then 
becoming known in Germany. Here he learned what 
neither Buxtehude nor Frescobaldi had been able to teach 
him clearness and plasticity of musical structure*. In 
the C minor fugue (Peters IV, No. 6 ; B. G. XXXVJII, No. 14) 
upon a theme of Legrenzi 






and in that in B minor (Peters IV, No. 8; B, G. XXXVIII, 
No. 19), in which Corelli's fugue on 






iqj 11 .y gj. I r 

^ -V ^ , 

is expanded from thirty-nine bars to one hundred, we see 
his effort to realise a new ideal, and to design in simpler, 
broader lines. In the canzona in D minor (Peters IV, 
No. 10 ; E.G. XXXVIII, No. 20), he has entered the 
world of beautiful forms, which he never leaves again. In 
the G major Fantasia (Peters IV, No. u; B. G. XXXVIII, 
No. 10) there is a lengthy five-part section of tranquil and 
finished polyphony, surrounded by quick, and brilliant yet 
at the same time simplified northern passage-work. Thus 
did Bach win his freedom from Buxtehude by means of 

* On Bach as a pupil of the Italians, and his adaptation of 
Vivaldi's violin concertos for organ, see p. 192. These arrange- 
ments are in Vol. VIII of the Peters edition, and Vol. XXXVIII * 
of the B. G. edition. 



The Weimar Preludes and Fugues, 269 

the Italians, and was enabled to bring to glorious reality 
the ideals that for two generations had agitated German 
organ music. 

We can only surmise in what order the works came in 
which he rises to independent mastery*. An important 
point in Bach's work is no doubt indicated by the small 
G minor fugue (Peters IV, No, 7; B. G. XXXVIII, No. 18) 
and the well-known D minor toccata and fugue (Peters IV, 
No. 4; B. G. XV, p. 267). So vigorously and broadly laid- 
out a theme as that of the G minor fugue is not to be met 
with in previous organ music, to say nothing of the 
rapid and weighty development of the fugue, in which 
there is hardly a trace of the ordinary fugal phraseology. 
Only the second pedal passage, that seems to belong to 
the older world, is inexpressive. 

In the D minor toccata and fugue, the strong and ardent 
spirit has finally realised the laws of form. A single dram- 
atic ground-thought unites the daring passage work of the 
toccata, that seems to pile up like wave on wave; and in 
the fugue the intercalated passages in broken chords only 
serve to make the climaoc aU the more powerful. 

The peculiar charm of these works comes from their 
spontaneous freshness of invention. They affect the 
hearer almost more powerfully than any other of Bach's 
organ works, and to play them is always to experience 
something of what the master himself must have felt when, 
for the first time, he exploited the full possibilities of the 
organ with regard to wealth of tone and variety of com- 
bination. 

For this reason the- wonderful pathos of the prelude 
and fugue in D major (Peters IV, No, 3; B, G. XV, No. 2) 
and the toccata and fugue in C major (Peters III, No. 8; 

* We may cite the fugue in G minor (Peters IV, No, 7), a small 
fantasia in C minor (Peters IV, No. 12), which Mendelssohn par- 
ticularly liked to play; the toccatas and fugues in D mmor (Peters 
IV, No, 4; B. G. XV, p. 267) and C major (Peters III, No, 8; B. G. 
XV, p. 253); the preludes and fugues in D major (Peters IV, No. 3; 
B. G. XV, No, a) and C major (Peters IV, No. i ; B, G, XV, No, i), 



270 XIII. The Organ Works. 

B. G. XV, p. 253) is as potent today as it ever was. Per- 
haps we are able to appreciate these works even more than 
our ancestors, for the great music of the nineteenth cent- 
ury has certainly had one result it has given those who 
have been nourished upon it a clear criterion for distinguish- 
ing between true and false pathos, and a doubled sense of 
enjoyment of the true, which is so rare. 

In the brilliant and dashing fugues belonging to these 
preludes, and again the one belonging to the majestic 
C major prelude (Peters IV, No. i; B. G. XV, No. i) there 
is an extraordinary display of virtuosity. We must not 
judge them too strictly by the rule that a fugue should 
be good in the first part, better in the middle, and ex- 
cellent in the last, for the two fugues in C major, at any 
rate, fall off in quality somewhat towards the end. 

Between these masterpieces of his youth and his real 
masterpieces Bach himself has drawn a clearly perceptible 
line; the former he left as they were; at the others he 
worked incessantly until he had given them their definite 
form. Thus it happens that for these works it is not the 
oldest copies, often not even the autographs, that are the 
most valuable, but the manuscripts that embody the work 
in its latest form. The extant autograph of the prelude 
and fugue in A major (Peters II, No. 3; B. G. XV, No. 6), 
for example, has no practical value, since it gives us merely 
an earlier, imperfect form of the work in 3 /a time in- 
stead of 3 /4* It should be noted, too, that in the manu- 
scripts the indication of certain preludes as fantasias or 
toccatas is not uniform. Sometimes they are not even 
entitled "prelude", but vaguely "pi&ce d'orgue", There 
are occasional Italian titles*. 



* The Weimar (and Cothen?) masterpieces for organ, some 
existing in later revisions, are the prelude and fugue in F minor 
(Peters II, No. 5 ; B. G, XV, No. 4), D minor (toccata, Peters III, 
No. 3; E.G. XV, No. 8), F major (toccata, Peters III, No, 2; 
B. G, XV, No. 10), C major (Peters II, No, i ; B. G. XV, No, 15), 
G major (Peters II, No. 2) B. G. XV, No. n), C minor (fantasia. 
Peters III, No. 6; B. G. XV, No. 7), C minor (Peters II, No, 6; 



The Weimar Preludes and Fugues, 



271 



Bach worked longest at the fugues in A minor (Peters II, 
No. 8); B. G. XV, No. 13) and G minor (Peters II, No. 4; 
B. G. XV, No. 12). The original form of the A minor 
subject is found in a three-part clavier fugue, in this shape* 




The elements of the later theme are already all there, but 
the great and simple melodic line that the musician is striv- 
ing after is as yet hidden by accessories, and is too ab- 
breviated to be effective. Only after long labour did it 
attain the calm plasticity of the finished theme, with its 
mixture of playfulness and strength in the semiquavers, 




The plan of the fugue and its main incidents are already 
prefigured in the first form. 

The A minor has also undergone revision. There is a 
copy of it J, P. KeUner's in which its essential chrom- 
atic line does not come out clearly through the majestic 
tread of the opening, 



B. G. XV, No. 16), A major (Peters II, No. 3; B. G. XV, No. 6), 
E minor (Peters III, No. 10; B. G. XV, No. 3), G minor (fantasia, 
Peters II, No, 4; B. G. XV, No. 12), A minor (Peters II, No. 8; 
B. G, XV, No. 13). 

* B. G. Ill, p. 334; Peters ed. of the Clavier Works, Part 4. 
See Reinhard Oppel (Bonn) Die grouse A moll Fuge fur Or gel und 
ihre Vorlage, in the Bachjahrbuch, 1906, pp. 74 78. 



272 XIII. The Organ Works. 

Definite form: 




Original form : 




In the second edition of Ids great Generalbassschule 
Mattheson tells us that at an organ examination in 1725 
he gave the candidates the following theme to develop 
extempore 




He does not mention Bach's name, but says he knows well 
to whom the theme belongs and who was the first to work 
it out artistically. The most natural assumption is that 
he had heard the fugue during Bach's Hamburg journey 
of 1720. It would thus be one of the works that the master 
played to Reinken at that time, perhaps out of compliment 
to him, the theme of the fugue being, in fact, borrowed 
from Reinken*. 

In Bach's version the theme is much simpler and 
more elegant. It runs thus 




* See p. 197. 



The Weixnar Preludes and Fugues. 273 

How is this difference to be explained? Ts the form in 
which Mattheson quotes it the original one, which Bach 
has altered and improved? Or has Mattheson remem- 
bered it wrongly and misquoted it? The probability is 
that the theme never existed save in its present perfect 
form. Mattheson knew it, but he could not give it to 
his candidate in this shape, because according to the 
rules of fugue it was incorrect. It is laid down in the 
rules that a fugue theme shall not extend over an oc- 
tave. The Hamburg examiner therefore thought it ne- 
cessary to alter Bach's theme in order to bring it into 
conformity with the eternal laws of the art.* 

The fugues in B minor and *G minor are virtuoso works, 
like those in C major and D major, not, however, like 
those, merely brilliant streams of notes, but perfect archi- 
tectonic creations, late -Gothics in music. As in that 
mediaeval form of architecture, the luxuriant detail of 
the pierced work only serves to unify and vitalise the 
simple, boldly- flung lines, and to exhibit power in its ut- 
most flexibility. The A minor fugue is the simpler and 
clearer in construction; that in G minor surpasses it, how- 
ever, in richness of imagination, 

In general, however, in the Weimar fugues virtuosity 
becomes less and less prominent. The themes become 
compact, simple, unadorned, almost- severe; in the working- 
out there is no longer any thought of effect. On the border- 
line stands the G major fugue (Peters II, No. 2; B. G. XV, 
No. n), the theme of which, in a minor form, however, 
is used again in the first chorus of the cantata Ich hatte 
viel Bekiimmernis (No, 21). The themes of the others no 
longer proceed by way of rapid passage work, but are built 
up of massive notes. This similarity groups them into a 
category of their own. They are the fugues in C major 
(Peters II, No. i; B. G, XV, No. 15), C minor (Peters II, 
No, 6; B. G. XV, No. 16), C minor (Peters III, No. 6; 

* The author is indebted for this explanation to Herr 
H. Keller, of Stuttgart, 

Schweitzer, Bach, 18 



274 



XIII, The Organ Works. 



E.G. XV, No. 7), F minor (Peters II, No. 5; B. G. XV, 
No, 4), F major (Peters III, No. 2; B. G. XV, No. 10), 
D minor (Peters III, No. 3; B. G. XV, No.<8), and A major 
(Peters II, No. 3; B. G. XV, No. 6). Their lack of showy 
effects accounts for these works not being so popular with 
players and audiences as the A minor and G minor fugues. 
But one has only to live with them to prize them more 
highly than those, even if at first sight they have not the 
same fascination. They represent the pure sublime, not, 
as before, the sublime in guise of the pathetic. The C minor 
fugue 







and the F minor fugue 



are so tremendously tragic precisely because they have 
divested themselves of every shred of passion, and ex- 
press only great sorrow and deep longing. The theme of 
the D minor fugue is indescribably suggestive of tranquil 
power; it throws out its limbs like an arch of mighty 
stones 



%/ 2P ^ ^TP ' ^ 



Those who still maintain that Bach's fugues are too 
elaborately wrought for church use are apparently ignorant 
of this one, or do not feel the Palestrina-like character 
of its style, or perceive that all these themes are really 
embodiments of religious ideas. An organist who recognised 
their true character once declared that he could no longer 
hear them without imagining a secret superscription to 
each of them. The C minor Fugue 



is 



The Weimar Preludes and Fugues, 275 

with its grand victory over its chromatic counter-theme, 
seemed to him the symbol of confident, faith. Over the 
sunny and vivacious A major fugue 







he would write u the gladsomeness of faith". It is worth 
noting that the same theme, slightly modified, dominates 
the orchestral introduction to the cantata Tritt auf die 
Glaubensbahn (No, 152), The curious step-rhythm that 
runs through it should be noted. 

The question of how many of the preludes and fugues 
originated together is difficult to decide, The preludes 
and fugues in A major, D minor (toccata), C major and 
G major seem to have sprung from the same ideas. This is 
also the case with the prelude and fugue in E minor (Peters 
III, No. 10; B, G. XV, No, 3), which are unique in their 
brevity and concision. In the A minor fugue the inner 
kinship of its therne with the motives of the prelude is so 
obvious that it seems to rise from the foaming prelude 
like Venus from the waves of the sea. 

On the other hand, it looks as if the two C minor preludes 
and the toccata in F major belonged to a later period than 
their respective fugues; Bach has apparently substituted 
these preludes for earlier ones with which he was dissatis- 
fied. If this be the case, he has nevertheless conceived the 
substituted preludes in the spirit of the fugues, preserving 
in each case an inner community between the two pieces, 
He always conceived preludes and fugues in pairs. If we 
find isolated specimens of each, they are works rejected 
by him at some revision or other. 

We may probably assume two dates at which Bach 
revised his organ works. The first fell in the period when 
Friedemann and Emmanuel were in a position to play 
their father's compositions; this would account for the 
fact that the preludes and fugues in G major and C major, 

1 8* 



276 XIII. The Organ Works. 

in their definitive form, are written on paper of 1725 and 
1730. A later and much more drastic revision probably 
took place towards the end of his Leipzig period, when 
Bach, having practically ceased to write cantatas, began 
to feel a new interest in his organ works. It is possible 
that he had the idea of making a complete collection of 
his preludes and fugues, as well as of the larger chorales 
he had revised. That after his death the basis for such a 
collection existed is proved by the fact that in a number 
of manuscript copies the fugues and preludes in A minor 
and C major and the pair in C minor (Peters II, }Sk). 6; 
B. G. XV, No. 16), with the three last Leipzig works 
the preludes and fugues in C major (Peters II, No. 7; 
B. G. XV, No. 17), B minor (Peters II, No. 10; B. G. XV, 
No. 14) and E minor (Peters II, No. 9; B. G. XV, No. 18) 

are grouped together as the "six great preludes and 
fugues". 

Besides these three great Leipzig works there is a fourth, 

the prelude and triple fugue in E flat major (Peters III, 
No. i; B. G. Ill, pp. 173 and 254) which embody the great 
chorales that appeared in 1739 as the third part of the 
Klavierubung. The preludes and fugues in C major, 
B minor and E minor, however, belong to a later date. 

The works of this period, apart from the preludes and 
fugues in C major, show a return to the style of Buxte- 
hude. They are not constructed on a single unified idea, 
like those of Bach's middle epoch, but are based on the 
dramatic opposition of different themes. Nevertheless the 
power and vastness of the design give the works an air of 
grandeur that is very different from the dramatic rest- 
lessness of Buxtehude and Frescobaldi. The old German 
organ style thus receives its final transfiguration in the 
symphonic works of Bach's old age, just as his last organ 
chorale, "Wenn wir in hochsten Ndten sind", brings the 
Pachelbel style of chorale treatment to perfection. 

It is the symphonic character of the latest works that 
makes it probable that the two preludes in C minor, which 



The Leipzig Preludes and Fugues, 377 

are conceived in the same spirit, belong to the same epoch, 
although they are grouped with fugues of the Weimar 
period. In the F major toccata and the C major prelude 
(Peters II, No. 7; B. G. XV, No. 17), there is a return to 
the virtuoso style, now, however, raised to a higher dignity 
and simplicity. In each case a single idea is worked out 
in complete accordance with its own nature. The C major 
prelude reminds us strongly of the first chorus in the can- 
tata Sie warden aus Saba alle kommen (No. 65). 

The prelude in E flat major, that introduces the greater 
chorales, symbolises godlike majesty. The triple fugue 
at the end of them is a symbol of the Trinity. The same 
theme recurs in three connected fugues, but each time 
with another personality. The first fugue is calm and 
majestic, with an absolutely uniform movement through- 
out; in the second, the theme seems to be disguised, 
and is only occasionally recognisable in its true shape, as 
if to suggest the divine assumption of an earthly form; 
in the third, it is transformed into rushing semiquavers, 
as if the Pentecostal wind were coining roaring from 
heaven. 

Perhaps the most striking thing among these Leipzig 
works is the flowery arabesque of the B minor prelude. 
The E minor prelude and fugue are so mighty in design, 
and have so much harshness blended with their power, 
that the hearer can only grasp them after several hearings. 
Time is needed, again, before one can feel at home in the 
quiet world of the B minor and C major fugues. It is not 
less certain, that only by degrees do we find our footing in 
the majestic monotony of the F major toccata and the 
C major prelude. This does not imply, however, that these, 
works should not be performed in church and at sacred 
concerts, but rather that they cannot be played often 
enough. Nor are they too long for the church service. The 
few minutes that they take can be found somehow in the 
liturgy. This is not too much to expect; to play the works 
with "cuts" is criminal. 



278 XIII. The Organ Works, 

The Eight little preludes and fugues (Peters VIII, B. G. 
XXXVIII), and the organ sonata (Peters I, B. G. XV) 
were written for the instruction of the two eldest sons. 
Any student with a fairly good piano technique can take 
up these works at once "in order to perfect [habilitate] 
himself in pedal-study", as they said in Bach's time. He 
will reach his- goal quicker and better than with modern 
organ- schools, against which while fully admitting their 
merits the reproach always holds good that they keep 
the student too long at the elements, and are too pedagogic 
in plan. Bach, on the other hand, loved to place his pupil 
at once in the midst of difficulties. 

In the strict sense of the term, it is wrong to speak of 
Bach's "organ sonatas 9 '. The two manuscripts in Which 
they have come down to us one from Friedemann's 
possession, the other, from Emmanuel's prove that they 
are really works for the clavicembalo with two manuals 
and pedal. This instrument was at that time in common 
use. It was excellently adapted for playing in three real 
parts, which accounts for the sonatas being in strict trio- 
form. This does not imply that Bach never played them 
on the organ also. He intended the last movement of the 
E minor sonata to come between the prelude and the fugue 
in G major (Peters II, No. 2; B. G. XV, No. n) and the 
largo from the C major sonata to come between the prelude 
and the fugue in G minor (Peters III, No. 5; B, G. XV, 
No. 5). 

If these sonatas were written for Friedemann, they must 
have originated towards the end of the second decade of 
the century. Some sections of them are indeed older, the 
first movement of the D minor sonata, for example, that 
occurs among the variants of the first part of the Well- 
tempered Clavichord. This collection was finished in 1722. 
An older version of the adagio and of the vivace of the 
sonata in E minor, for oboe d'amore, gamba, and continuo, 
figures as the introduction to the second part of the can- 
tata Die Himmel erzdhkn (No. 76), which certainly belongs 



The Small Preludes, Organ Sonatas, and Passacaglia. 279 

to the year 1723. According to Spitta, the whole of the 
organ sonatas were in existence as early as 1727; in 1733 
Friedemann went to Dresden as organist. 

"We cannot say enough of the beauty of these sonatas", 
writes ForkeL For the connoisseur, indeed, there is hardly 
a purer aesthetic delight than to pursue these three contra- 
puntal lines so free and yet so bound by the laws of 
beauty through their delightful intertwinings, to say 
nothing of the perfection of the themes. The dreamy sub- 
ject of the adagio of the D minor sonata 



made even Bach himself its captive ; he took it up again 
later and made it into a trio, full of longing, for clavier, 
flute and violin*. 

Forkel tells us that Friedemann owed his consummate 
technique to these sonatas, which is quite credible**. To 
this very day they are the Gradus ad Parnassum for every 
organist. Whoever has studied them thoroughly will find 
scarcely a single difficulty, in the old or even in modern 
organ music that he has not met with there and learned 
how to overcome; and before all he will have attained that 
absolute precision that is the chief essential for good organ - 
playing, since in this complicated trio-playing the slightest 
unevenness in touch is heard with appalling clearness. 

It is noteworthy that Friedernann's manuscript of the 
sonatas, on which the Peters edition is founded, contains 
many more embellishments than Bach's own autograph 
copy, which Emmanuel possessed. This belongs to a later 



* A dagio of the concerto in A minor, lor flute, violin and clavier, 
with accompaniment of two violins, viola and continuo. B. G. 
XVII, No. 8. 

** Forkel, p, 60, Forkel speaks of several other separate trios 
that he knew. Among the pieces of this kind may be mentioned 
the pastorale in F major (Peters I, p. 86 ; B. G, XXXVIII, No. 22), 
and a trio in D minor (Peters IV, No. 14; B. G. XXXVIII, No. 23), 



28o XIII. The Organ Works. 

date than Friedemann's copy. We see from this how 
Bach became more and more sparing with ornaments; it 
is, indeed, one of the reproaches levelled against him by 
his critic Scheibe. The edition of the Bachgesellschaft 
follows the autograph. Both manuscripts are now in the 
Berlin Library. 

The passacaglia (Peters I, p. 75; B. G. XV, p. 289) was 
also written in the first place for the cembalo with pedal, 
and later arranged for the organ. As a matter of fact its 
polyphonic structure fits it so thoroughly for the organ 
that we can hardly understand nowadays how anyone 
could have ventured to play it on a stringed instrument. 
On the other hand there is no organ work that makes 
such demands as this in the matter of registration. Each 
of the twenty sections constructed on the repeated bass 
theme must have its characteristic tone-colour, and yet, 
if disconnectedness is to be avoided, no colour must be too 
sharply differentiated from its predecessor or its successor, 

Passacaglia, in French passecaille properly denotes 
an old Spanish dance. Musicians understood by it a 
piece constructed on a recurring bass theme; in the cia- 
cona in French chaconne which is also developed 
like a string of pearls, it was permissible to introduce the 
theme in any voice. In Bach's work the theme appears 
several times in the upper part, so that it is not a pure 
passacaglia, but partakes of the nature of both the passa- 
caglia and .the chaconne. 

The work was conceived under the influence of Buxte- 
hude, whose organ compositions in this genre are of con- 
siderable significance. It therefore seems somewhat strange 
that there are not a number of youthful works by Bach 
in this style. He saw clearly, however, that on the whole 
the incoherency of this kind of work was not suitable to 
the greatest organ music, and he ventured upon the ex- 
periment only with this colossal theme. He follows his 
teacher, again, in grouping the passacaglia with a fugue. 
Buxtehude, however, placed this at the beginning, while 



Early Chorale Preludes. 381 

Bach, with more reason, place.s it at the end, where it has 
the effect of rising to a climax. But in the last resort no 
comparison of Bach's passacaglia with any of Buxtehude's 
works of the same kind is possible, for the pupil puts into 
his a dramatic life that was beyond the power of the 
master. 

Both the external and the internal evidence point to 
the later Weimar period as the date of origin of this work. 
The autograph was still known to exist about the middle 
of the nineteenth century. Since then it has disappeared 
completely, like that of the prelude and fugue in B minor, 
that must be somewhere in Scotland. Both autographs 
were used for the Peters edition. 

The chorale preludes which he thought worth preserving 
were grouped by Bach in five collections, containing al- 
together about ninety of these works. They are, the 
OrgelbttMein, begun in Weimar and written out in a fair 
copy at C6then; the chorales published in 1739 as the 
third part of the Klavieriibung; six chorales published by 
Schubler, of Zella, about 1747; the canonic variations on 
the Christmas hymn "Vom Himmel hoch da komm ich 
her", published about the same time by Baltbasar Schmidt 
of Nuremberg*, which Bach afterwards submitted to the 
Mizler Society on joining its ranks; and the collection of 
eighteen great chorales, during the revision of which he 
was overtaken by death. 

Another fifty or so chorale preludes mostly youthful 
works have come down to us** in copies made by 
pupils and friends. Some of these e. g. that for doxible 



* The approximate date of this edition is determinable by the 
fact that Emmanuel's clavier concerto in D major, which was issued 
by the same firm in 1745, bears the publication number 27, while 
the father's chorale variations are numbered 28. See Spitta, III, 
294. Kretzschmar thinks it possible that these variations had al- 
ready appeared in 1723. (See his preface to Vol. XLVI of the Bach 
Edition, p. 21). Internal reasons make this date hardly possible. 
** B. G. XL, p. I 102. For chorale works of which Bach's 
authorship is doubtful, see p. 167 ft 



282 XIII. The Organ Works. 

pedal on "An Wasserfllissen Babylons' (Peters VI, No. I2a) 
would certainly have been included by Bach in the col- 
lection had he got that far with the revision. 

It is not clear what aim he had in view in publishing 
the six chorales issued by Schiibler. They are only ar- 
rangements of three-part chorale arias from the cantatas, 
that have nothing in common with his other chorale pre- 
ludes, and do not even go particularly well on the organ *. 
He already had in his portfolio dozens of splendid chorales 
ready for engraving. Why did he pass these over and 
issue mere transcriptions? 

The chorale partitas upon "Christ, der du bist der helle 
Tag", "0 Gott, du frommer Gott", and u Sei gegriisset, 
Jesu giitig" (Peters V, p. 60 91; B. G. XL, p. 107 123)**, 
in which the number of variations corresponds to that of 
the verses in the respective hymns, are certainly works 
of his earliest youth, as is evident from the awkward har- 
monisation of the chorale melody and the optional use of 
the pedal. When he wrote these works Bach had not 
yet found himself; he was still a pupil of Bohm. Where 
and when he composed them, whether at Liineburg or 
at Arnstadt, cannot be determined. In any case they 
are clever student's exercises, which it is impossible to 
play without being delighted with the power of original 
thematic figuration which they exhibit. Bach seems later 
on to have revised the third partita, as is fairly clear from 
the improved harmonisation and the obbligato use of the 
pedal in the last variations. The five-part final variation 
is a masterpiece. Afterwards Bach no longer writes chorale- 
variations. He was probably led away from the genre 

* For the Schuhler chorales see the B.,G. XXV, p. 23 ff,, and 
Peters VI, No. 2, VII, Nos. 38, 42, 57, 59, and 63. The chorale 
prelude on "Wachet auf, ruft uns die Stimrae" comes from the 
cantata of that name (No. 140). 

** It may be left undecided whether the partitas upon "Ach, 
was soil ich Sunder machen" and "Allein Gott in der H6h sei Ehr" 
(B. G. XL, p. 189 207) are by Bach. It is not impossible. By 
"partita" here is to be understood a suite of variations. 



The OvgelbticMew. 283 

by purely artistic considerations, though we must not 
forget that in Weimar and Leipzig he no longer had any 
practical use for such works, which were only serviceable in 
places where, according to the old custom, the organ 
"struck in between the singing" and worked out inde- 
pendently every other verse of the chorale, the singing 
ceasing*. 

When, at the end of his career, Bach once more returned 
to this form and wrote the variations on "Vom Himmel 
hoch da komm ich her", his only purpose was to pack into 
a single chorale the complete art of canon; and in the last 
variation he could not deny himself the pleasure of intro- 
ducing all four lines of the melody simultaneously in the 
last three bars. If this work already she-yvs the tendency 
to abstract thought that was characteristic of his last years, 
there is, for all that, a good deal of emotion in these chorale 
arrangements. They are full of Christmas joyousness and 
cheeriness. The first variation is of a truly bewitching 
beauty of tone. It is an interesting fact that besides the 
engraved copy we possess an original manuscript, that is 
of later date, and gives the definitive version. It is evident, 
then, that even in his printed compositions Bach always 
found something to improve. 

In the true chorale prelude, Bach appears to have culti* 
vated chiefly the forms of Pachelbel, Bohm, Buxtehude, 
and Reinken. Towards the end of the Weimar period, how- 
ever, he becomes independent of his masters and produces 
a type of his own the chorale prelude of the Orgelbiich- 
lein. In this the melody is used as a cantus firmus, unal- 
tered and uninterrupted, usually in the uppermost voice ; 
round it plays an independently conceived motive, not 
derived from any of the lines of the melody, but prompted 
by the text of the chorale, and embodying the poetic idea 
that Bach regarded as characteristic for music and expres- 
sible in musical terms, Thus in the chorale preludes of 

* See p, 27, 



284 XIII. The Organ Works. 

the Orgelbiichlein the melody and the text are both repre- 
sented at the same time, the cantus firmus being poetically 
illustrated by means of the characteristic motive*. 

Here Bach has realised the ideal of the chorale prelude. 
The method is the most simple imaginable and at the 
same time the most perfect. Nowhere is the Diirer-like 
character of his musical style so evident as in these small 
chorale preludes. Simply by the precision and charac- 
teristic quality of the line of the contrapuntal motive he 
expresses all that has to be said, and so makes clear the 
relation of the music to the text whose title it bears. 

The OrgelbuMein is thus not only of significance in the 
history of the development of the chorale prelude, but is 
one of the greatest achievements in music. Never before 
had any one expressed the texts in pure tone in this way; 
no one afterwards undertook to do so with such simple 
means. At the same time the essence of Bach's art comes 
clearly into view for the first time in this work. He is not 
satisfied with formal perfection and fulness of sound, 
otherwise he would have continued to work with the forms 
and formulae of his teachers in the chorale prelude. He 
aims at more than this; he aspires after the plastic ex- 
pression of ideas, and so creates a tone-speech of his own. 
The elements of such a speech already exist in the Or gel 
buchlein: the characteristic motives of the various chorales 
correspond to many of Bach's later emotional and pictorial 
tone-symbols. The Orgelbuchlein is thus the lexicon of 
Bach's musical speech. This must be our starting-point 
if we would understand what he is striving to express in 
the themes of the cantatas and the Passions. Until the 
significance of the Orgelbuchlein was perceived, the funda- 
mental character of Bach's art remained, almost down to 
the present day 3 obscure and disputable. 



* The form of the chorale prelude in the Orgelbtichlein is new 
only in so far as the independent motive illustrates the text. Re- 
garded from the purely formal standpoint, it is a variety of the 
Buxtehude genre of small chorale fantasia. 



The OrgelbttMein. 285 

The title, indeed, does not indicate the universal sig- 
nificance of this collection. It runs thus: "The Little 
Organ Book, wherein instruction is given to a beginning 
organist to work out a chorale in every style, also to perfect 
himself in the study of the pedal, the pedal being treated 
quite obbligato throughout in the chorales herein con- 
tained. To the honour of the Lord Most High, and that 
my neighbour may be taught thereby. Autore Joanne 
Sebast. Bach. p. t. Capellae Magistro S. P. R. Anhaltini- 
Cothinensis." 

The autograph is now in the Royal Library at Berlin. 
It contains ninety-two leaves, and is bound in paste-board 
with leather back and corners. At the head of each page 
Bach wrote the title of the chorale that was to appear 
thereon, so that if the composition extended beyond the 
page he had to paste an extra strip of paper below, or make 
use of the tablature. All these chorales were written in 
Weimar. Afterwards, in Cothen, he made a careful fail- 
copy oi them. There still exists a Weimar autograph of 
the Orgelbttchlein, that once belonged to Mendelssohn. It 
lacks the pages containing the first twelve chorales. On 
the cover is a no'te to the effect that the owner had cut 
out three more leaves two for his bride's album, and 
one for Clara Schumann*. 

The order of the chorales is that of their succession in 
the church year. This is easily understood when we re- 
member that at that time each Sunday had its own special 
hymns, allotted to it once for all, and that other organists 
of the epoch e. g. B. Walther of Weimar also wrote 
similar yearly cycles of chorale preludes. In the details 
of their grouping, however, especially with regard to the 
chorales appropriated to feast days, each individual was 
naturally allowed a certain amount of liberty. Bach made 



* For further details of this autograph of the Orgelbttchlein 
see Spitta II, 986 (German ed.). It originally contained eight 
chorales fewer than the C6then one. It is not cited in the B. G. 
edition. 



286 XIII, The Organ Works. 

the most ingenious use of this freedom. He disposed of 
the, chorales in such a way that the Christmas ones formed 
a miniature Christmas oratorio, those of the Passion time 
a Passion, and those of Easter an Easter oratorio *. He 
aimed also at other effects of contrast. The chorale "Das 



* The following is the original disposition of the chorales for 
the various feasts: 

CHRISTMAS ORATORIO 

Introduction: Gottes Sohn 1st kommen. V, No. 19. 
Herr Christ der einzig Gottessohn. V, No. 22. 
Lob sei dem allmachtigen Gott. V, No. 38. 
The Manger; Puer natus in Bethlehem. V, No, 46, 
Gelobet seist du, Jesus Christ. V, No. n, 
Der Tag, der ist so freudenreich. V, No. n. 
The Appearance of the Angel; Vom Himmel hoch da komm ich 

her. V, No. 49. 

Vom Himmel kam der Engel Schar. V, No. 50. 
The Adoration "before the Manger; In dulci jubilo (mediaeval sacred 

cradle song). V, No. 35. 

Lobt Gott ihr Christen allzugleich. V, No. 40. 
The Mystical Adoration: Jesu meine Freude, V, No. 31. 

Christum wir soUen loben schon (canto fermo in alto). V, No. 6. 
Final hymns; Wir Christenleut haji jetzund Freud. V, No. 55, 
Helft mir Gottes Giite preisen. V, No, 21. 

THE PASSION 
Introduction: O Lamm Gottes. V, No. 44. 

Christe du Lamm Gottes. V, No. 3. 

The Seven Last Words; Da Jesus an dern Kreuze stund. V, No. 9. 
Jesus' death: O Mensch bewein dem 1 Sunde gross. V, No. 45. 
Song of Thanksgiving: Wir danken dir, Herr Jesn Christ. V*, No. 56, 
Meditation; Hilf Gott, dass mirs gelinge. V, No. 29 

(The chorales on the descent from the cross, the laying in the 
tomb, and the departure from the tomb, are not worked out.) 

EASTER ORATORIO 
Easter Morning: Christ lag in Todesbanden. V, No. 5. 

Jesus Christus unser Heiland, der den Tod iiberwand, V,No. 32. 
Christ ist erstanden. V, No. 4. 

The Announecment of the Resurrection: Erstanden ist der heilge 
Christ. (The text of this chorale represents a dialogue 
between the woman and the angel at the grave.) V, 
No. 14. 

Erschienen ist der herrlich Tag. V, No. 15. 
Heut triumphieret Gottes Sohn. V, No. 28. 
(In the B. G. edition the Orgelbuchlein will be found in Vol. 
XXV 2 . 



The OrgelbtiMHw. 287 

alte Jahr vergangen 1st" (V, No. 20) is a sorrowful medita- 
tion in the twilight as the last evening draws to its close; 
it is followed by the jubilant song "In dir ist Freude" 
(V, No. 34) that is filled with the light of the new day, 
Of the two chorales relating to the presentation in the 
temple and Simeon's hymn of praise, the first, "Hit Fried* 
und Freud' ich fahr dahin" (V, No. 41), depicts a joyous 
longing for death, and the other, "Herr Gott, nun schleuss 
den Himmel auf ' (V, No. 24), a sorrowful longing. The 
sombre hymn on original sin, "Durch Adams Fall" (V, 
No. 13), is followed at once by the hymn of salvation in 
Christ, "Es ist das Heil uns kommen her" (V, No. 16). 

The Orgelbttchlein is barely one-third finished. The 
Cothen autograph is planned out for a hundred and sixty- 
nine chorales ; forty- five of these are complete ; for the others 
we have only the white pages. What is the explanation 
of this? Did the Leipzig appointment come just at this 
time and prevent the continuation of the work? In this 
case, why did not Bach resume it when later on he turned 
his attention again to the chorale prelude? The aban- 
donment of the collection in its incomplete state must 
have been due to some inward reason. Speaking generally, 
it is the grouped chorales relating to . the various feast- 
times that are finished, and, of the others, those of which 
the strong pictorial or characteristic quality seemed to 
make .them specially suitable for music. The texts of 
the numbers not completed lack these musical qualities. 
No characteristic theme could be evolved from them ; they 
could only be developed as pure music, not in their poetic 
or pictorial aspects. All the chorales of this collection, 
however, were to be little tone-pictures; and as circum- 
stances made this plan impossible, Bach preferred to leave 
the collection unfinished. How strictly he adhered to the 
characteristic type he had in his mind for the QrgelbuMein 
can be seen from the fact that he did not include beautiful 
chorales like "Herzlich tut mich verlangen" (V, No. 27) 
and "Liebster Jesu wir sind hier" (V, No. 36), which were 



288 XIIl. The Organ Works, 

quite suitable as regards their size and were certainly in 
existence at that time, simply because they were not 
constructed on a characteristic motive. 

When Griepenkerl edited the OrgdbUchlein for Peters, 
about the middle of the forties, he unfortunately altered 
the original order, in which each chorale is in a position 
that explains it, and adopted instead a merely alphabetical 
arrangement, besides inserting "smaller" chorale preludes 
and chorale fughettas that did not form part of the collec- 
tion. We get the correct OrgelbuMein by eliminating 
from the fifth volume of Peters' edition of the organ works 
Nos, 7, 18, 20, 23, 26, 27, 36, 39, 43, 47, 52, and 53, and 
arranging the remainder in this order Nos. 42, 19, 22, 
38, 46, 17, n, 49, 50, 35, 40, 31, 6, 55, 21, 10, 34, 41, 24, 
44, 3* 8, 9, 45, 56, 29, 5, 32, 4, 14, 15, 28, 25, 37, 12/48, 13, 
*6, 3<>, 33 5*> 54> 2 > ** Between Nos. 28 and 25, again, 
must be inserted the first part of the chorale "Komm Gott 
Schopfer, heiliger Geist" (VII, No. 35), which, although 
Spitta (I, 611) does not think so, was originally part of 
the Orgelbuchlein. The second verse was not added until 
later. Spitta is of opinion that the treatment of the pedal 
in the first part is not sufficiently obbligato to authorise 
our regarding the work as composed for the Orgelbilch- 
lein\ but every organist will testify that the obbligato 
character of the pedal part is shown by the fact that it is 
much more difficult to play than it looks ; it is by no means 
easy always to strike these simple notes on the weakest 
part of the bar. 

The plan of the second collection of chorale preludes is 
explicable, like the first, from the order of the old hymn 
books. The Orgelbuchlein deals with the cantica de tern- 
pore, i. e., the hymns grouped according to their order in 
the church year; the other collection, that appeared in 
1739 as the third part of the Klavierttbung, deals with the 



* Organists who possess only the Peters edition can recon- 
struct the original in this way. 



The Catechism Hymns. 289 

catechism hymns'* 1 . By these was understood, at that 
time, a small collection of classical hymns on the main 
points of the Christian doctrine, that were included in every 
hymn-book. The arrangement was the same as in the 
Lutheran catechism. The core of them was formed by 
Luther's hymns: "Dies sind die heiligen zelm Gebot", 
**Wir glauben all an einen Gott*", "Vater unser im Himmel- 
reich", "Christ, unser Herr, zum Jordan karn", "Jesus 
Christus unser Heiland" (the communion hymn), and 
"Aus tiefer Not schrei ich zu dir" (the confessional hymn). 
Bach chose this catechism in the form of Lutheran hymns 
for musical treatment. In order to have the dogma com- 
plete, he prefaced these five chief hymns with the Kyrie 
and Gloria to the Holy Trinity from the Leipzig service, 
i. e., the three hymns "Kyrie Gott Vater", ' 'Kyrie Gott 
Sohn", "Kyrie Gott heiliger Geist", and the hymn to the 
Trinity, "Allein Gott in der Hoh sei Ehr", this last of 
course, in three versions. 

Luther, however, had written a greater and a smaller 
catechism. In the former he demonstrates the essence of 
the faith; in the latter he addresses himself to the children. 
Bach, the musical father of the Lutheran church, feels it 
incumbent on him to do likewise; he gives us a larger and 
a smaller arrangement of each chorale, with the exception 
of "Allein Gott in der Hoh sei Ehr". The larger chorales 
are dominated by a sublime musical symbolism, aiming 
simply at illustrating the central idea of the dogma con- 
tained in the words; the smaller ones are of bewitching 
simplicity, The whole collection is introduced by the 
majestic E flat major prelude, and ended by the triple 
fugue in the same key. 

One would have thought this conception at any rate 
interesting enough to be respected in the various editions. 
This, however, has never been done except in the original 



* B. G. Ill, p. 170 260. The duets on pp. 242 253, how- 
ever, have nothing to do with the third part of the Klaviertibung, 
They were accidentally included during the engraving. 

Schweitzer, Bach. ig 



XIII. The Organ Works. 

edition of the Bachgesellschaft. Even Naumann, in the 
practical edition he brought out for Breitkopf and Hartel, 
mixes these works up with the others without any regard 
to their special quality and their inner connection. The 
Peters edition of these catechism chorales can be recon- 
structed thus 

Introduction: Prelude in E flat major, III, No. i. 

Trinity: "Kyrie". (Large version, VII, Nos. 39a, b, c.; small ver- 
sion, VII, Nos. 4oa, b, c.) 

"Allein Gott in der H6h sei Ehr' J . (Large version VI, Nos. 
5, 6, and 10.) 

The Ten Commandments. "Dies sind die heilgen zehn Gebot", 
(Large version, VI, No. 19; small version, No. 20). 

Faith; "Wir glauben all 1 an einen Gott". (Large version, VII, 
No. 60 ; small version, VII, No. 61.) 

The Lord's Prayer: "Vater unser im Himmelreich". (Large ver- 
sion, VII, No. 52; small version, V, No. 47.) 

Baptism: "Christ unser Herr zum Jordan kam". (Large version, 
VI, No. 17; small version, VI, No. 18.) 

Penitence: "Aus tiefer Not schrei ich zu dir". (Large version, 
VI, No. 13; small version, VI, No. 14.) 

The Lord's Supper: "Jesus Christus unser Heiland, der von uns". 
(Large version, VI, No. 30; small version, VI, No. 33.) 

Conclusion: Triple fugue in E flat major, III, No. i. 

Bach was not correct in placing the penitence between 
the baptism and the communion, and it is impossible to 
say why he did so. It should really come last among 
these doctrinal pieces. 

These chorales were probably all composed at the same 
time, expressly for this collection, towards the end of the 
thirties. This was certainly the case with the larger ver- 
sions; in the case of the smaller ones we cannot be sure 
whether they formed an earlier collection. 

It is otherwise with the last collection, the "Eighteen 
Chorales"*. It contains, for the most part, compositions 
of the Weimar period, which Bach at the end of his life 
revised and partly rewrote. Rust, indeed, in the preface 
to volume XXV of the B, G. edition, maintains against 

* B. G. XXV, p. 79 ff . 



The Eighteen Chorales. 291 

Spitta that they belong to the Leipzig period; but this 
is hardly probable. They are plainly masterpieces that 
Bach wrote while still more or less dependent on the 
forms, laid down by Buxtehude, Bohm and Pachelbel. 
It contains no chorales of the type of those in the Orgel- 
buchlein. 

How Bach has polished these works is evident from 
fifteen older versions that have come down to us*. The 
autograph of the "Eighteen Chorales" is in the Berlin 
library; it belonged at one time to Philipp Emmanuel. 
The last chorale, u Wenn wir in hochsten Noten sind", is 
incomplete in the autograph, and must have been com- 
pleted from the Art of Fugue, in which it appeared as 
Bach's last work**. Here again it is unfortunate that, 
regardless of Bach's last wishes, these revised chorales are 
always mixed up in order to get a purely alphabetical 
arrangement, although, it is true, this collection is not 
like the two others, governed by a definite sequence of 
ideas. The authentic order can be restored in the Peters 
edition thus: 

(1) Komm, heilger Geist. VII, No, 36. 

(2) Alio mode, VII, No. 37. 

(3) An Wasserfliissen Babylon, VI, No, i2b. 

(4) Schmiicke dich, o Hebe Seele, VII, No. 49. 

(5) Herr Jesu Christ, dich zu uns wend 1 , VI, No. 27. 

(6) O Lamm Gottes unschuldig. VII, No, 48. 

(7) Nun danket alle Gott, VII, No. 43. 

(8) Von Gott will ich nicht lassen. VII, No, 56. 

(9) Nun komrn dor Heiden Heiland, VII, No. 45, 
(10) , Alio modo (Trio), VII, No. 46. 

(u) Alio modo. VII, No. 47. 

(12) AJlein Gott in der Ho"h sei Ehr, VI, No, 9. 

(13) Alio modo, VI, No. 8, 

(14) Alio modo (Trio). VI, No. 7, 

(15) Jesus Christus unser Hciland. VI, No. 31. 

(16) Alio modo. VI, No. 32, 

(17) Komm, Gott Schdpfer, hoiligcr Geist, VII, No, 35. 

(18) Wenn wir in hdchsten Ndten sind (Vor deinen Thron 
tret' ich allhier). VII, No. 58. 



* See B. G. XXV a, pp. 151189. 

** With regard to Bach's last chorale prelude see pp. 223, 224. 

19* 



XIII. The Organ Works. 

The trio upon "Nun komm der Heiden Heiland" (VII, 
No. 46) makes so strange an impression on us that it seems 
like a transcription of a movement from a cantata. Strictly 
in the old Pachelbel style is the angular arrangement of 
"Nun danket alle Gott" (VII, No. 43), that charms both 
player and hearer more and more as their familiarity with 
it increases. The chorale "Allein Gott in der Hoh sei Ehr" 
(VI, No. 9), is purely in the style of Bohm; to many people 
it seems rather youthful. In the prelude on "Nun kornm 
der Heiden Heiland" (VII, No. 45), that is laid out on the 
same plan, the arabesque-like contour of the melody seems 
much more mature and perfect. It is full of a dreamy 
expectancy. 

We see the style of Bohm perfected and idealised again 
in the chorale "An Wasserfliissen Babylon" (VI, No. i2b), 
in which the melody is given to the tenor. We are re- 
minded of Buxtehude by the arrangement of "Jesus 
Christus unser Heiland*' (VI, No. 32), the brilliant and 
animated "Komm, heilger Geist, HerreGott" (VII, No. 36), 
the "Gott Schopfer, heiliger Geist" (VII, No. 35), and the 
"Von Gott will ich nicht lassen" (VII, No. 56). 

The most important works in this collection, however, 
do not conform to any strict type. They arc fantasias 
planned on broad lines, with free thematic use of one or 
more of the lines of the melody of the chorale. Bach has 
welded the forms into a new unity, through which the 
older outlines are only visible as through a fine blue mist. 
This chorale form might fairly be called the mystic. The 
chorale themes become veiled, the melodic line more free, 
as if everything external had been lost, and only the general 
mood, the fundamental emotional idea, were being ex- 
pressed. In this style the chorales "Allein Gott in der Hoh 
sei Ehr" (VI, No. 8), "Komm, heilger Geist, Herre Gott" 
(VII, No. 37), and "Schmiicke dich, o Hebe Seek" (VII. 
No. 49) form a category of their own, Mendelssohn was 
so affected by the mood-painting in the last named 
chorale that he told Schumann "if life were to deprive 



Paralipomena. 293 

him of hope and faith, this one chorale would bring them 

back"*. 

The triple chorale on "0 Lamm Gottes" (VII, No. 48) 
and the arrangement of "Jesus Christus unser Heiland" 
(VI, No. 31) represent the ideas more in their dramatic 

aspect, so much so that one is almost tempted to agree 

with Rust against Spitta, and date these works from the 
Leipzig period. 

It is difficult to agree with Spitta's division of Bach's 
chorale arrangements into the three categories of pure 
chorale preludes, organ chorales, and chorale fantasias. 
It is more reasonable to group them according to the style 
of treatment in the fugued style of Pachelbel, the 
"coloristic" style of B6hrn and Reinken, or free fantasias 
in the style of Buxtehude, There is further the type of 
the Orgelbuchkin, in which the characteristic motive 
illustrates the uninterrupted cantus /tVwws, and finally 
the great chorales, that offer a perfected synthesis' of all 
the forms. 

Besides the arrangement of "An Wasserflussen Babylon" 
for double pedal (VI, No. I2a) there were several ex- 
cellent and interesting chorales not included by Bach in 
any collection, among which may be mentioned the fan- 
tasia on "Bin' feste Burg 9 * (VI, No, 22), the sublime fugue 
on the Magnificat (VII, No. 41), the joyous trio on "Nun 
freut euch, lieben Christen g'mein" (VII, No. 44), and the 
expressive chorale "Erbarm* dich mein, o Herre Gotf* 
(B. G. XL, p, 60), which in its kind the melody is sup- 
ported by evenly flowing quavers, is unique among 
Bach's chorale preludes, 

u Christ lag in Todesbanden }> (VI, No. 15). "Jesu meine 
Freude" (VI, No. 29), a Vom Himmel hoch, da komm ich 
Jier" (VII, Nos, 54 and 55), and "Wir glauben all' an einen 
Gott" (VII, No. 62) are all admittedly youthful works, 

* Schumann's Musik und Musiker, Reclam's edition, I, i$3- 
See also p. 245 of the present volume. 



294 XIV. The Performance of the Organ Works. 

in which we can follow Bach's earliest development. It 
is not clear why the harmonisation of the chorale "Herr 
Gott, dich loben wir" (VI, No. 26), that is meant as an 
accompaniment to the hymn, should figure among the 
chorale preludes in all editions, even in that of Naumann. 
What is needed is a cheap edition of the chorale preludes 
in their original form, distinguishing the collections planned 
by Bach himself from the detached chorales that have 
come down to us, the latter being freely grouped according 
to their style and their value. With an alphabetical index 
any one of them could be found in a moment. It is also 
desirable that the texts should be printed along with the 
chorales, many of them having by this time disappeared 
from our hymn-books*. 



CHAPTER XIV. 
THE PERFORMANCE OF THE ORGAN WORKS. 

How did Bach play his organ works, and how should 
they be played? This practical question is much more 
important than the historical and aesthetic question. 
Upon the performance it depends whether these works 
can really be brought home to the hearers, or whether they 
are simply to be admired in a kind of respectful wonder, 
their beauty being taken on trust rather than actually felt. 
This certainly happens frequently at the present day. 

Indications for performance are scarce in Bach's works. 
Once or twice, as in the D minor toccata (Peters III, 
No. 3), the changes of manual are indicated; in the Qrgel- 
buchlein we are expressly told which pieces are to be played 
on two keyboards; in the Schxibler chorale trios we are 
told whether eight or four or sixteen feet stops are to be 
drawn (Peters VII, Nos. 38, 57, 59, 63). Thjat is practically 

* Chapter XXII of the present volume is devoted to the elucida- 
tion of the chorale preludes. 



The Bach. Organ and the Modern Organ. 295 

all. We know, from Walther's copy, Bach's manner of 
playing the chorale prelude u Ein 5 festc Burg" (Peters VI, 
No. 22) at the opening of the new organ at Muhlhausen; 
his unfriendly critic Schcibe informs us that his manner 
at the organ was extraordinarily quiet; Forkel says that 
he astonished other organists by the audacity of his tone- 
combinations*. Otherwise he was distinguished from his 
contemporaries only by his consistent pursuit of the prin- 
ciple of legato playing. He had no experience of the 
Venetian shutter swell, which was introduced about that 
time in England, where Handel took great interest in the 
invention. In Germany the opposition to this so-called 
trifling lasted a long time. When Bum'ey, more than 
twenty years after Bach's death, heard the Berlin organs, 
he was astonished to find that not one possessed a swell. 
It no more occurred to Bach than to the rest of his con- 
temporaries that some day organs would be fitted with 
combination stops, adjustable combinations, and all the 
rest of the apparatus of the modern organs, especially the 
so-called "concert'* organs, 

How do we stand now with regard to the performance of 
Bach's works on the modern organ? We have achieved 
infinite possibilities in registration, the power of gradual 
variation from pianissimo to fortissimo, and, by means of 
the swells, a certain power of tone nuance. But we have 
lost the old tone of the organ that Bach wrote for; and, 
since the tone is the chief thing, it must be said that the 
modern organ is not so suitable for Bach playing as is 
generally supposed**. 



* "His method of registration was so extraordinary that many 
organ builders and organists were appalled when they saw him, 
It seemed to them that such a combination of stops could not pos- 
sibly sound well; but they wondered greatly when they observed 
that the organ sounded at its best, though the effect was of an 
unusual kind that could never 1 have been produced by their own 
style of registration.'* (Forkel, p. 20.) 

** On the organ of Bach's day see Pirro's UQvgue de J, S 
Bach (Paris, 1895); Albert Schweitzer's Deutsche und franxdsische 



296 XIV. The Performance of the Organ Works. 

Our registers are all voiced too loudly or too softly. If 
we pull out the whole of the diapasons and the mixtures, 
or add the reeds, we get a force of tone that in the end 
becomes positively unbearable. The lighter manuals are 
weak in comparison with the great organ; they usually 
lack the necessary mixtures. Our pedals are coarse and 
clumsy and also poor in mixtures, as well as in four-feet 
stops. The trouble comes principally from the change in 
the disposition of the organs, the relation between diapa- 
sons and mixtures having been altered, wholly to the de- 
triment of the latter; but also from the unnaturally strong 
bellows of the modern organ. In our passion for strength 
of tone we have forgotten beauty and richness of tone, 
which depend upon the harmonious blending of ideally 
voiced stops. The older organs are becoming scarcer and 
scarcer. There are many organists today who have never 
heard Bach played on the kind of organ the composer 
had in view when he wrote. The day is not far distant 
when the last of our beautiful Silbermann organs will be 
replaced, or renovated beyond recognition; and then the 
Bach organ also will be one of the unknown things of the 
past, like certain orchestral instruments that he uses in 
his scores. 

If we play Bach on an old and well-preserved Silber- 
mann organ, both players and hearers are as little con- 
scious as the master himself was of the need for frequent 
changes of register, for on such an ipstrument the diapasons 
and mixtures give a forte so rich, intense, full-coloured, 
and yet in no wise fatiguing, that we can, if need be, pre- 



Orgelbaukunst und Orgelkunst (Leipzig, 1906); J. W. Enschede^s 
Moderne Of gals en Sachs Orgelmuziek (in Caecilia, Amsterdam, 
April and May 1907); O. Dienel's Die Stellung der moderns Of gel 
zu S. Backs Ovgelmusik (a lecture, Berlin, 1890); H. Reimann's 
Cber den Vortrag dev Orgelkompositionen Backs (Musikalischo 
R&ckblicke, 1900). Mention may also be made of Isidor Mayr- 
hofer's Back- Studien, Vol.1, Orgetwerke (Leipzig, 1901), This 
work gives aesthetic analyses of all Bach's organ compositions, 
but does not touch the question of performance. 



Registration. 297 

serve it unchanged throughout a prelude or a fugue*. On 
such an organ, moreover, both the inner parts and the 
pedal come out clearly, whereas on the modern organ the 
inner parts are confused, and the pedal, by reason of its 
deficiency in four-feet stops and mixtures, and its inferiority 
in weight to the enormous masses of tone above it, cannot, 
even at its most brutal, throw out a clear line. And all 
this on account of the too heavy voicing of our registers i 
The organs of forty years ago, that are voiced with the 
normal pressure, for the simple reason that at that time 
the electric bellows was unknown, and the wind was con- 
sequently sparingly used are better Bach organs than 
the modern ones. What a joy it is, for example, to play 
Bach on the beautiful Walker organs built between about 
1860 and 1875! 

As a rule Bach kept to the characteristic registration 
with which he began, getting variety and gradation in his 
playing by transitions from one manual to another**. It 
is noteworthy, however, that he played a great many organ 
pieces throughout on the great organ without any change 
whatever of manuals and without any gradation of tone, 

* 10 Alsace there are still some organs of this kind. I confess that 
it was these that first put me out of humour with the ultra-modern 
interpretations of Bach's organ works, and made me long for organs 
on which one could play the preludes and fugues in their natural, 
lofty simplicity, Widor afterwards confirmed me in these views. 

** The representatives of the old German organ school, that 
still preserved some traditions from the Bach epoch, played Bach's 
preludes and fugues throughout on the great organ with diapasons 
and mixtures. The pedal was sometimes strengthened by reeds. 
In the last three decades of the nineteenth century this method of 
performance came to be looked upon as stiff and pedantic. The 
struggle ended unfavorably to the old tradition, mainly through 
the fact that this method is really impossible on the brutal organs 
with the modern voicing. During the last few years a reaction has 
set in against the false modernisation of the performance of the 
organ works; the aim is to restore to favour what is legitimate 
in the old simple style, The Belgian and French organists 
Lemmens, Guilmant, Widor, Gigout always played Bach ac- 
cording to the principles of the old German school. Their organs 
were voiced more lightly and clearly than the German ones. 



2Q8 XIV. The Performance of the Organ Works. 

the essence of them being the evolution of a single idea, 
free from any dramatic suggestions. This is especially the 
case with the works in which the pedal is employed un- 
interruptedly throughout, for example, the two preludes 
in Cmajor (Peters II, Nos.i and 7) and A major (II, No, 3) 
respectively, the majority of the chorales of the Orgelbuch- 
lein, and both the larger and smaller ones ia the f ugal style 
of Pachelbel. Here any variation of tone-colour or alterna- 
tion of strong and weak would destroy the ideal unity of 
the work. The organ sonatas in trio-form, again, are most 
effective when the tone-colour that has been found to be 
the best for each of the three obbligato parts is maintained 
throughout. 

As regards the choice of tone-colours, it need only be 
said that these are sufficiently Bachian when they suit the 
character of the work. We must not grudge even months 
of trouble in order to find the right registration. It is still 
disputable what is meant by the expression "organo pleno" 
that often figures at the head of a work*. It practically 
amounts to this, that in passages of this kind Bach desires 
the main strength of the organ, at any rate diapasons and 
mixtures. On present-day organs, however, this must be 
done with discretion, the tutti of the diapasons and 
mixtures, thanks to the disagreeably strong quality of the 
latter, hardly corresponding to what Bach had in his mind. 

Should we also use the reeds in Bach's preludes and 
fugues? His objection to those usually found in modern 
organs would have been that they are too blaring and 
that they obscure the polyphony. On the other hand it 
is probable that he added reeds to his diapasons and mix- 
tures, for he cultivates the metallic tone in the orchestra 
as well. His pedal timbre was really based on reeds. More- 

* As bearing on the history of the expression, it may be noted 
that the Allemannic people still distinguish between the "half 11 
and the "whole" organ. Thus "pro organo pleno" may well have 
meant originally "for full organ", the reference being to the quality, 
not to the stops. In point of fact, however, on the old organs the 
forte was used for preludes and fugues. 



Registration. 299 

over we know how highly he valued good reeds in aa or- 
gan *. What we have to do in future is to restore the old 
delicate and beautiful reeds, that just add a lustre to the 
diapasons without overwhelming them as ours do **. Un- 
til then we must manage with compromises, and use dia- 
pasons, mixtures, and reeds with sufficient discretion to 
get something like the old quality of tone. 

It is interesting to note that Bach's contemporaries 
complained that the Silbermann brothers voiced their 
organs too softly, in order to get beauty of tone. Bach 
evidently did not think so, 

The effect that can be made with a fine full fortissimo 
combining all the timbres may be seen in the little prelude 
in E minor (Peters III, No* 10). If we play it right through 
without any change, we realise at once that this is how 
Bach conceived it, and that to play it with any variation 
of colour or of force is to destroy its dramatic majesty. 

We play the chorale preludes of the Orgdbticblein and 
many others too softly, again, because we do not make 
sufficient use of beautiful mixtures on the secondary 
manuals, which would not only sound weU in themselves 
but would permit the use also of one or two reeds. We 
are thus thrown back, as a rule, on the characterless tone- 
colour of some eight-feet diapasons which particularly 
obscure the polyphonic writing, (the four-feet and two- 
feet are also generally voiced too strongly) and we 
try to make up in sentimentality for what we have lost in 
richness and quality of tone. It is obviously wrong, for we 
lose the simple effect of the cantus firmus. 



* See p, 199, 

** Even the French reeds, in spite ol their beauty, are unfitted 
for Bach playing. On the other hand, the diapasons and mixtures 
of the Cavaill6-Coll organs seem made for it, this builder having 
been particularly anxious to avoid abnormally strong and "solid" 
voicing. On the organs at St, Sulpice and Notre-Dame, Bach's 
fugues come out with extraordinary clearness. One of the finest 
Bach organs in existence is the one, rich in mixtures, that adorns 
the Cavaill6Coll aUUev in Paris (15 Avenue du Maine), 



3OO XIV. The Performance of the Organ Works. 

We should carefully consider which chorales are written 
for two and which for one manual, and not plume our- 
selves on our cleverness when we play the latter on two 
manuals*. Bach's own intentions can always be gathered 
from the style of writing. A part that he intended for one 
manual cannot be played upon two without seriously 
marring the grouping and leading of the voices. Conversely, 
a work conceived for two manuals is written in such a way 
that each part lies smoothly and clearly on its own manual, 
down to the smallest detail. This principle can be applied 
to all the organ works. 

If the cantus firmus is broken up into coloratura, as in 
the chorale preludes in Bohm's manner, it often comes out 
to particular advantage with an oboe or clarinet colouring. 
Wonderful effects of blending can be obtained by using a 
small mixture in the swell and adding diapasons and an 
oboe colouring to it. These chorale preludes in Bohm's 
style should be played the most delicately of all. 

The pedal should not be too heavy, and, at any rate in 
the chorale preludes, uncoupled wherever possible. It is 
most effective with its own stops. Frequently only an eight- 
feet tone should be employed for example in the chorale 
preludes "O Lamm Gottes" (Peters V, No. 44) and "Gottes 
Sohn ist kommen" (Peters V, No. 19). At other times 
wen only a four-feet tone is suitable; e.g. in "In dulci 
Jubilo" (Peters V, No. 35)**. Where the double pedal is 



* Most organists succumb to this temptation, e.g., in the little 
chorale prelude "O Lamm Gottes" (Peters V, No. 44). 

** In "Vom Himmel hoch" again (Peters V, No. 50), the pure 
eight-feet pedal is decidedly better, as it does not cover up the semi- 
quaver figure of the third voice, that often goes very low. It is 
a mistaken idea that "In dulci Jubilo" requires a pedal with F sharp. 
Bach writes it thus in order to make the tenor position of the part 
clear. It is played an octave lower and with a four-feet* Bach 
would be delighted to know that the latest pedals of Cavaill6-Coll f 
that go to G, have the natural F sharp; but he would have ex- 
pressed himself pretty forcibly with regard to the numerous organ 
builders who think that for the ' 'ordinary" organ a pedal to D is 
enough. 



Registration. 301 

prescribed throughout, it goes without saying that only 
eight-feet stops should be employed, with the four-feet for 
stronger effects, This rule is frequently disregarded. The 
case is different when the double pedal is used at the end 
of a work e. g. in the D major prelude (Peters IV, No. 3); 
here the sixteen-feet is to be maintained, though it must 
be admitted that the fortissimo of a modern pedal of this 
kind is far from charming. 

The organist should not worry either himself or his 
hearers too much with the working-out of a canonic pas- 
sage. The piece is not there for the sake of the canon, but 
the canon for the sake of the piece, especially in the canons 
of the Orgelbilchtein, If we hear properly the melody of 
the cantus firmus, the other parts can be so far kept in the 
background that the uninitiated need not even suspect 
there is a canon in progress. 

In chorales to be played on two manuals, experience 
teaches us that as a rule it is better to let the string char- 
acter prevail in the left hand and a flute colour in the upper 
parts, this colour being freeirom harshness there, but muddy 
in the lower register. 

Special difficulties are offered by the two great chorales 
with double pedal, "An Was^erfliissen Babylon*' (Peters 
VI, No. I2a) and "Aus tiefer Not schrei ich zu dir" (Peters 
VI, No. 13). For the first the following registration is re- 
commended strings in the pedal, in the left hand flutes, 
in the right hand strings, all soft eight-feet stops*. The 
chorale prelude "Aus tiefer Not" is very effective when the 
whole eight and four-feet register diapasons, mixtures, 
and reeds, is used in, the pedal ; in this way the mixtures 
and reeds which are often missing can be got from the 
second and third manuals by coupling, and we can play the 
four upper parts on the great organ with the full eight and 
four-feet diapasons, even adding a good mixture later on. 

* Many organists erroneously imagine that they are bound to 
use sixteen-feet stops in pieces with double pedal. What then 
becomes of the fourths, thirds, and seconds in the lower octave I 



3O2 XIV. The Performance of the Organ Works. 

This solves the problem of bringing out the cantus ftrtnus 
clearly in the upper pedal part. Under certain circum- 
stances it is as well to omit the eight-feet trumpet, and 
to employ only the four-feet stops for the reed timbre. In 
truth, however, we can only play these double pedal parts 
quite legato either on the old narrow Bach pedals or the 
curved French and English pedals with their circular arrange- 
ment. It is impossible on the flat and excessively broad 
pedal keyboard that is regarded as the only correct thing 
in Germany. 

The foregoing remarks apply to the works that seem to 
call for neither a change, of manuals nor a change of colour. 
As a rule, however, Bach goes on the presupposition that 
we shall play his works with the variety suggested by 
their contents and their style of treatment. He gives no 
indications on the point, simply because the works carry 
their own indications. Leading parts on the grand organ, 
subsidiary parts, (generally recognisable by the omission 
of the pedal) on the supplementary manuals, this is 
what he expects from the player. This can be seen from 
the D minor toccata (Peters III, No. 3), in which he has 
specified the changes of manual, probably for a pupil. 

In the chorale on "Ein' feste Burg" (Peters VI, No. 22) 
we can reconstruct Bach's registration from Walther's 
copy. His indications evidently refer to the Miihlhausen 
organ, the renovation of which Bach superintended, and 
which he opened, in all its new glory, in the autumn of 
1709, perhaps at the Reformation feast*. First of all, 
bars i 20, Bach kept the right hand on the second manual, 
employing, among others, the sesquialtera, and the left 
hand on the first manual, defining the tone-quality of this 
by the fagotto 16". Bars 20 24 he played, on the third 

* For the specification of the Miihlhausen organ see Spitta I, 
355. The registration marks in Walther's copy (Spitta I, 394, 395) 
are a little confused, owing to the fact that he transferred the 
registration to his Weimar organ, that had only two manuals. 
The indications for the Miihlhausen organ can, however, be made 
out without difficulty. 



The Natural Architecture of the Preludes and Fugues. 303 

manual, drawing the soft pedal stops, especially the new 
sub-bass. During this he strengthened somewhat the other 
manuals, and came back to them in bars 24 32, in which 
the assistant probably Walther, took advantage 
of the short pause in the bass in bar 24 to draw the full 
pedal. In bars 32 37 Bach returned to the third manual, 
the assistant shutting off the strong pedal registers. The 
finale, from bar 37, was played on the great organ with all 
the stops, The registration of the piece could not be simpler 
or more effective. These two examples show us how in- 
geniously Bach managed the changes of manual. 

The first thing, therefore, is always to look for the simple 
architectural lines of the work. The registration that brings 
these out is the right one; any other, no matter how in- 
genious it may be, is less good, in that it obscures the real 
configuration of the work. We must keep to the principle 
that every fugue and every prelude is to begin and end on 
the great organ. It is quite wrong to give out a fugal theme 
piano or pianissimo, and let eadh voice, as it enters, take 
it up more loudly. The theme, whether joyous or sad, 
must always be given out with a certain fullness of tone, 
leaving the cumulative effect to come from the entries of 
the different voices. It is painful to hear themes that 
should enter proudly, like those of the A minor and G minor 
fugues, given out softly on the third manual in a way that 
quite obscures their real character all for the sake of 
the precious crescendo. In many a fugue the whole archi- 
tectural effect is sacrificed to the desire to render the theme 
always fully audible, to which end it and the other 
voices are transferred from the great organ to one of the 
others. This is unpermissible. Now and again we hear 
a Bach fugue played in such a way that it tapers off at the 
end in the most beautiful pianissimo. 

This modernisation is partly the product of our present- 
day way of looking at music. If our organists wish to prove 
themselves modern musicians, it can only be by trans- 
ferring the modern orchestral style to these works, They 



304 XIV, The Performance of the Organ Works. 

forget that Bach's own orchestral style was the ancient one, 
not the modern. The effects he aimed at on the organ 
are the same that he aspires after in the Brandenburg con- 
certos. The organist, therefore, would be well advised to 
study these works thoroughly, in order to penetrate to 
the secret of Bach's style and to realise that with him it 
is a question not so much of a gradual cumulation of effect 
as of the lucid opposition and combination of two or three 
bodies of tone. For this reason the modern swell really 
does our organists a disservice, in that it is always tempting 
them to indulge in these gradual crescendi. The true cu- 
mulative effects in Bach are made by the entry at definite 
moments of two or three new tone-masses, and the de- 
crescendo by their departure. 

On the other hand the constitution of our organs, that 
are incapable of the real Bach forte, and in which the 
polyphonic writing does not come out clearly, makes us 
have recourse to artificial effects instead of natural ones, 
and we try to make Bach interesting by variations of tone 
and of colour and by an over-insistence on the theme. Here 
also, until we begin building ideal Bach organs again, we 
must resort to a wise compromise. This does not mean 
that the gradual rise and fall of tone, effected on a 
small scale by means of the Venetian shutter swell and 
on a large scale by means of the cylinder, is always wrong 
in Bach. Archaistic tendences should not be tolerated 
in music. Bach would have been the last to set his face 
against new methods. Many passages e, g. the con- 
clusion of the A minor fugue really demand an increase 
in the forte itself. And how happy Bach would have been 
could he have got a finer piano on his third manual by 
shutting off some of the wind, as is possible by means of 
the Venetian shutter swell! To refuse to make use of this 
device in the great episode in the A minor fugue beginning 
at the fifty-first bar, employing first a decrescendo, then 
a crescendo, is to be false to Bach. Only the present-day 
organist must make use of the device in such a way as not 



Changes of Manual. 305 

to disturb the original architecture of the fugue, and be 
sure that the various episodes of the work come out simply 
and clearly. Within these limits he may do what he thinks 
necessary. If this principle is generally recognised, there 
will be an end of much of the modern pretentious virtuosity 
in the performance of Bach's organ works, and people 
will come back from the art that merely stimulates interest 
to the art that satisfies. And then the hearers will realise 
that Bach's organ works are not complicated, but ex- 
tremely simple. 

Organists should particularly avoid the sudden de- 
crescendo in the cadences that has gradually become the 
fashion under the seductive influence of the cylinder swell. 
It is to be hoped also that some day the practice will 
cease of employing the cylinder swell at the beginning of 
the F major toccata, instead of starting with a good forte 
and leaving the crescendo to the dramatic unfolding of the 
canon. For the rest, this toccata is one of the works that 
are most effective when played simply with various nu- 
ances of the one forte. 

The works differ greatly with regard to the changes of 
manual they require. In many these changes amount 
merely to an occasional bar or two on the subsidiary man- 
uals; we may even doubt whether Bach went to these 
manuals in the bars where the pedal ceases, e. g. in the 
C major fugue (Peters II, No. i). Usually, however, the 
changes are so important that we cannot be in doubt as 
to where the intermezzo begins on the secondary manual, 
and where it ends. In a number of fugues the change 
comes exactly in the middle, so that they appear to be 
triform both in structure and performance: (i) the 
first exposition on the great organ; (2) the intermezzo; 
(3) a second exposition extending to the final cadence. 
Of this kind are the fugues in G major (II, No. 2), C major 
(II, No. 7), F major (III, No. 2), A minor (II, No. 8), 
G minor (II, No, 4) and B minor (II, No, 10). The inter- 
mezzo on the subsidiary manuals begins each time at the 

Schweitzer, Bach. 2O 



306 XIV. The Performance of the Organ Works. 

place where the pedal ceases, (or shortly afterwards) and 
ends at the re-entry of the pedal (or shortly before). In- 
stead of one big crescendo, lasting from the beginning to 
the end, as we moderns conceive the fugue, the fugue as 
Bach conceived it consisted of two equipollent main sec- 
tions, with a subsidiary section between them. 

To destroy the character of the intermezzo is to destroy 
the Bach fugue. The most striking fugues in respect of 
this simple structure are the three just mentioned, in 
A minor, G minor and B minor. Here we clearly realise 
the necessity of a diminuendo up to a certain point in the 
intermezzo. The theme retires to some extent into the 
innermost and uppermost parts of the organ, there to evolve 
slowly until the time comes for the re-entry of the pedal, 
signalised by a return to the tone-colour that prevailed 
at the commencement of the intermezzo, when the pedal 
ceased. How the organist manages this, how he passes 
from one manual to another, how he introduces mixtures 
and reeds into the diapasons and takes them out again, 
and brings out this architecture even down to the smallest 
detail, that is his affair. He will have to be ruled by 
the disposition of his organ. The chief thing is for him 
to recognise the plan of the fugue and bring this out, 
not a fantastic plan of his own invention*. 



* The following is a suggestion for the registration of the A minor 
fugue: The manuals are coupled; stops drawn, diapasons 8 and 4 
and mixtures of manuals II and III. Begin the iugue on the great 
organ. On the first beat of bar 44 add the mixtures of the first 
manual. In bar 51 the left hand remains on the first manual, 
from which the mixtures are taken off, and the right hand goes 
to the second; in the second half of bar 59 it is followed by 
the left ; during the sustained F in bar 60 the right hand goes im- 
perceptibly to the third manual; in bar 63 the left follows it; at 
the same time the mixtures of the third manual are taken off ; the 
swell box slowly closes until bar 70 ; in bar 7 1 the right hand goes 
to the second manual, while on the third, with the swell box closed, 
the mixtures are again introduced ; the swell box is opened slowly ; 
in one of the following bars the left hand goes imperceptibly to 
the second manual; in the second half of bar 78 it gives out the 
theme on the first manual, the mixtures of the second entering at 



Changes of Manual. 307 

Other fugues exhibit two or more changes. There are 
two in the following fugues: A major (Peters II, No. 3, 
bars 59 87 and 121 146); F minor (II, No. 5, bars 43 64 
and 96 120); C minor (II, No, 6, bars 59 94 and 118 
143); C minor (III, No. 6, bars 27 50 and 58 67); 
E minor (III, No. 10, bars 15 19 and 27 33), We meet 
with more than two intermezzi in the great E minor fugue 
(Peters II, No. 9), that has, indeed, more the character of 
a fantasia. It can be properly played only on an organ 
whose subsidiary manuals are so supplied with mixtures 
that they do not contrast too markedly with the great 
organ. The sections dominated by the main theme and the 
quaver figures are to be played wholly on the great organ, 
and the passages with the semiquaver figures on the sub- 
ordinate manuals. The charm depends each time upon 
the immediate entry of the chief theme and the quaver 
movement. Here one doubly regrets that Bach has not 
recorded the change of manuals. An excellent plan is to 
play the great semiquaver figures, in which the pedal 
shares, on the first manual, without its own mixtures, 
but adding those of the other manuals by means of the 
couplers, and then, at each return of the main theme, 
bring in the mixtures and finally the reeds of the great 
organ. 

The prelude in E flat major (III, No, i), which is 
similarly constructed, must be played in the same way. 
The triple fugue appended to it is most effective when 
we play the first part with the full diapasons of the great 
organ, perhaps with delicate mixtures of the other 
manuals coupled to them , the second part on the sub- 
sidiary manuals with all the mixtures, and in the third 



the same time ; in bar 88 the mixtures ol the iirst manual are added, 
to which the right hand goes in, bar 9 1 ; on the sustained E in, bar 
94 the reeds of the third manual are added, on the first beat of bar 
113 those of the second, from bar 132 onwards those of the first, 
and whatever else is available for a fortissimo* The G minor fugue 
could be registered on the same lines. 



308 XIV. The Performance of the Organ Works, 

part return to the first manual, which has meanwhile 
been increased to fortissimo. 

In a number of preludes in which the pedal is used 
throughout, it is as well to work on a basis of varied fortes, 
relieved one against the other by the intensity of the 
mixture-tone obtained by adding the coupled manuals to 
the full diapasons. For the G minor fantasia (II, No. 4), 
we would propose at the commencement, the three 
coupled manuals, diapasons, mixtures and reeds; in bars 
9 14 retain only the diapasons; from bar 14 onwards 
add gradually, each time on the strong beat of the bar, 
first of all the mixtures, then the reeds in the order III, 
II, I; in bar 25 take them all off again, so that bars 25 31 
are played only with the diapason tone-colour, (of course 
on the great organ); in bar 31 introduce into the diapason 
mass first of all the mixtures and reeds on the third manual, 
two bars later those on the second, and two bars further 
on those on the first, until the fortissimo is reached, which 
is then retained to the end. This method, by which the 
player always keeps to the same manual, is perhaps less 
interesting than many of the modern virtuoso methods ;r 
it has, however, the advantage of presenting the work 
to the hearer in all its grand simplicity. 

This gradated forte can be employed with the same 
good effect in the preludes in C minor (II, No. 6) and' 
E minor (II, No, 9), only that here, in the section without 
pedal, there are episodes that need to be played on the 
subsidiary manuals. 

The fantasia in C minor (III, No. 6) is the despair of 
every organist. It is almost impossible to reproduce its 
ideal beauty in material tone. After every attempt we 
come back to the simplest method, that consists in be- 
ginning with a flexible diapason basis, introducing, at the 
transition from bar ri to bar 12, almost all the diapasons 
and mixtures of the subsidiary manuals, returning in 
bar 21 to the first diapason colour, maintaining this until 
bar 32, and then gradually introducing, till the end is 



Changes of Manual* 309 

reached, the whole of the diapasons of the full organ and 
the mixtures of the subsidiary manuals. 

One of the works that suffers most at the hands of or- 
ganists is the B minor prelude (II, No. 10), although its 
structure is the simplest possible. We should begin on 
the great organ; at bar 17 go to the subsidiary manuals; 
in bar 27 the right hand returns to the great organ, followed 
by the left hand in bar 28; both remain there until bar 43, 
and then move to the other manuals until bar 50, when 
they again return to the first manual; from bars 56 60 
keep to the great organ, but retain only the diapasons; 
during bar 60 introduce mixtures and afterwards reeds 
under the cover of the chromatic passage; maintain this 
on the great organ until bar 69, when we again take off 
the mixtures and the reeds and continue with the dia- 
pasons; in bar 73 we come back for the last time to the 
subsidiary manuals, which we dexterously bring up to 
fortissimo, and return in the course of bars 78 and 79 to 
the great organ with all its stops drawn. 

In the passacaglia it is very effective to give out the 
theme with the complete diapasons of tKe organ coupled 
to the pedal, and then to begin pianissimo on the third 
manual and to draw more stops at each new variation. 
In bar 73 we may perhaps go to the first, manual; in the 
four variations that follow we may introduce by degrees 
the whole of the diapasons and mixtures, and finally some 
reeds. In bar 105 we go to the second manual, and after- 
wards to the third, taking off the mixtures and reeds; 
from bar 114 onwards we close the swell box slowly; the 
arpeggio passages are played with fine eight, four, and two- 
feet registers of the third manual. On the last beat of 
bar 129 we revert to the diapasons and mixtures of the 
first manual; in the following variations we add reeds and 
mixtures of the other manuals; finally we add also the reeds 
of the first manual. 

To ascertain where the change of manual can be made, 
and the way in which it can best be effected, we must 



3IO XIV. The Performance of the Organ Works. 

endeavour by continual study of the work to discover 
the ground principles on which it is constructed. The 
great art consists in going back with both hands to the 
great organ at the moment of the pedal entry. From the 
structure of the passage we must try to infer whether we 
must go to the first manual with one hand after the other, 
in which case it is always best to begin with the left, 
since this can enter almost imperceptibly in the lower 
part or whether Bach demands a decided contrast, 
for which we must bring both hands simultaneously to 
the great organ. His own playing must have been charac- 
terised by extraordinary refinement, since he expressly 
desires that the manuals shall lie quite close to each other, 
so that he may easily pass from one to the other*. There 
are many critical passages in which sustained notes or 
harmonies have to be taken with perfect smoothness of 
transition on another manual. 

For changes of manual in accordance with Bach's in- 
tentions a certain homogeneous tone-colour must unite the 
three manuals. On the organs of today, that have scarcely 
any mixtures on the subsidiary manuals, this is often dif- 
ficult. It is also regrettable that on our organs the three 
manuals no longer represent , three different and sharply- 
characterised qualities of tone. Hence the main effect of 
the change of manuals and of the coupling and uncoupling 
of them is lost. This causes the organ virtuosi of today to 
renounce the most natural means of effect and have recourse 
to the cylinder swell, that finally becomes so monotonous. 

If the vital question in Bach playing is that of the 
coupling and uncoupling of the manuals, and of the entries 
and exits of different tone-groups, it must be said that 
the plan of our modern organs does not lend itself greatly 
to this. The couplers are worked by the pressing of knobs, 
which means that the player cannot make full use of them, 
since in Bach playing neither hand can be dispensed with. 

* See p. 200. 



Tempo and Phrasing. 

Further, our combination stops and adjustable combinations 
usually work in such a way that they interfere with the 
existing registration instead of reinforcing it; and they 
often have the further disadvantage of not acting separate- 
ly on each manual, but on the whole organ. The console 
ought to be so arranged that the couplers can be worked 
either by the hand or the foot, the two mechanisms, 
of course, being connected. The player could then draw 
the coupler with his hand and release it with his foot, 
and vice versd, or he could employ only the hand or only 
the foot, as suited him best. The collective stops and 
the adjustable combinations should be arranged on a double 
principle, so that at the will of the organist they could 
either suspend or supplement the drawn stops. It would 
be an advantage if they also could be worked both by the 
hand and the foot. This would imply another much-needed 
simplification of our organs*. 

It looks as if Bach's works were destined not only to 
instruct the organist but to reform the organ builders of 
the present day, to emancipate us from the folly of the 
inventor and lead us back from the complicated to the 
simple, from the strong-toned organ to the organ of rich 
and beautiful tone. 

The more we play Bach's organ works, the slower we 
take the tempi. Every organist has this experience. The 
lines must stand out in calm plasticity. There must be 
time also to bring out their dovetailing and juxtaposition. 
At the first impression of obscurity and confusion, the 
whole effect of the organ piece is gone. 

If so many organists imagine that they play Bach * 'in- 
terestingly" by taking him fast, this is because they have 
not mastered the art of playing plastically, so as to give 



* The console sketched above is realised in the organ of the new 
church at Strassburg-Kronenburg, built by the Alsatian firm of 
Dalstein and Harpfer (Bolchen) ; the stops of this organ are voiced 
in the old Silbermann style. The organ at St. Nicholas's in Strass- 
burg is also built on this plan, as a Bach organ. 



312 XIV. The Performance of the Organ Works. 

vitality to the work by bringing out its detail clearly. It 
is quite a mistaken idea that what Bach chiefly wants is 
a monotonous smoothness. He certainly favored the 
legato style. But his legato is not a mere levelling; it is 
alive. It must be filled by a fine phrasing which the hearer 
need not perceive as such, but of which he is conscious 
as a captivating lucidity in the playing. Within the 
legato, the separate tones must be grouped into living 
phrases. This" intimate style of phrasing breaks up the 
stiffness of the organ tone. The effect should be as if what 
is impossible on the organ had become possible, that 
is to say, that some notes have a heavy and others a light 
touch. That is the ideal to be aimed at. 

In the old days, when the absolutely uniform legato 
obtainable by passing under the thumb was not known, 
so that only a few of the notes were played legato while 
the others were detached owing to the displacement of 
the hand, players had a feeling for the artistic grouping 
of notes within a legato that we have lost, but of which 
we can form a rough idea by observing how, at that time, 
runs were divided between the two hands. Even in the 
introduction to the C major toccata (Peters III, No. 8), 
or that of the E minor prelude (Peters III, No. 10), there 
is revealed a whole world of interesting legato combina- 
tions. Many organists, indeed, have no idea that this 
division between the two hands indicates Bach's phrasing; 
they are even proud of themselves when they play these 
passages with one hand, or in octaves with both hands, 
making one monotonous scale passage of them. If we 
follow the principle indicated by Bach's manner of writing 
his phrases, we see that he usually conceives four con- 
secutive notes as grouped in such a way that the first is 
detached from the others by an imperceptible break, and 
belongs rather to the previous group than to the one that 
follows. Thus not 

HI JTJ3 but 



Tempo and Phrasing. 



313 



In this way there is no sense of monotony in the legato. 
When we apply this principle we are surprised with what 
clearness and animation the passages come out. Conse- 
quently we must play thus 

Prelude in A minor (Peters II, No. 8): 



SE 



ssgi|^| 




Toccata in I) minor (Peters III, No. 3): 




One of the most instructive examples in this regard is 
the passage-work in the subsidiary section of the B minor 
prelude. Grouped as a scale it is always unsatisfactory; 
it only acquires life and form when we play it thus: 




The fresh and healthy prelude in C major, again, (Peters 
IV, No. i) with the fine pedal solo, only loses its stiffness 
when we phrase it on this principle. We have the same 
experience with the other C major prelude (Peters II, 
No. i). This phrasing, however, must never be obtrusively 
noticeable within the general legato. Its effect must be 
merely that of an agitation of the main contour, To bring 
out the phrasing by means of slurs and breath-signs is muck 
too clumsy a method; it should really be done by delicate 
and inexpressible means. 



314 



XIV. The Performance of the Organ Works. 



It is doubtful whether, for the sake of variety, we should 
now and then play whole quaver or semiquaver passages 
staccato. 

The phrasing of the fugue themes is still in dispute, 
though we are gradually getting further away from the 
extravagances of the earlier virtuosi, when they wrought 
such violence on Bach in the first joy of their virtuosity. 
A phrasing is fundamentally wrong that is not simple, 
and cannot be maintained throughout the whole piece, 
especially whenever the theme enters. Therefore any 
notes that interrupt the normal flow and are denoted by 
characteristic leaps, must to some extent be taken out 
of the group and stand by themselves. Thus : 

Fugue in A minor (Peters II, No. 8): 




Fugue in G minor (Peters II, No. 4); 




Toccata in F major III (Peters No. 2): 






The theme of the E flat triple fugue is interesting, as the 
transformations it undergoes necessitate changes in its 
phrasing: 

First fugue: 






Tempo, Phrasing and Ornamentation. 
Second fugue: 



315 



f 



Third fugue: 



Only when phrased in this way do the themes of the 
second and third fugues become perfectly clear to the hearer, 
which is impossible if the notes are grouped evenly. 

When the same note is struck repeatedly it should be 
sustained for only half its time-value, with a pause for the 
remainder. The repeated notes are thrown into relief by 
the preceding and following slurs. Thus: 

Fugue in G major (Peters II, No. 2): 







Fugue in C minor (Peters III, No, 6). 




This rule holds good not only for the phrasing of the 
themes, but for the treatment of the repeated notes in 
general. It cannot be observed too strictly. Equally 
weighty is another rule. If a repeated note occurs on 
the paper by reason of the same note entering in another 
voice, the note must be held without repetition; it must 
not be heard as two notes because one voice takes it over 
from the other. 

In successive chords, the repeated notes are to be de- 
tached, and those moving in intervals are to be legato. 
IB this way the leading of the voices is brought out with 
extraordinary clearness* If one tries to translate these 
rules into practice, the simplest pieces become difficult. 



XIV. The Performance of the Organ Works. 

In order to realise the difficulties and the effect of play- 
ing in this strict style, the succession of chords in the middle 
section of the little E minor prelude (Peters III, No. 10) 
should be studied. 

It is upon this plastic style of playing, not on ingenious 
registration and virtuosity, that the effect of Bach's organ 
works depends. But everything rests on seeming trifles, 
and we organists are much too modest and too indulgent to 
ourselves in this respect; moreover so few of us acquire 
a technique really adequate to Bach's demands. Many 
do not even acquire absolute precision of touch. 

Ornaments occur relatively seldom in the organ works, 
yet frequently enough for there to be plenty of opportun- 
ities for sins of thoughtlessness*. It is always forgotten 
that the Bach trill, as Emmanuel expressly informs us, 
does not begin on the main note but on the secondary 
note, and, when it is long, always has a final turn. 

Thus the theme of the F minor fugue (Peters II, No. 5) 
should be played either: l 




Too rapid trills are to be avoided on the organ. 

In the opening mordents of the Dorian toccata (Peters 
IV, No. 4), the E minor fugue (Peters III, No. 10), and 
the E flat prelude (Peters III, No. i), the ornaments must 
be played with a whole-tone interval. The grace-notes 
in the pedal part of the B minor prelude (Peters II, No. 10) 
are to be played as quavers; Rust thought he recollected 



* The following remarks should be supplemented by the 
detailed examination of Bach's ornamentation in the chapter on 
]the performance of the clavier works. 



Organ and Clavier Fugues, Transcriptions, 317 

that in the autograph, that unfortunately has disappeared, 
they were so noted. It is best, however, to separate the 
two notes, thus: 






There remains the question of the organ works for 
practical use. It is entirely a question of usage. Who- 
ever uses Naumann's instructive collected edition (Breit- 
kopf and Hartel) or Schreyer's collection (Hofmeister, 
Leipzig) to mention these alone certainly gets a 
good deal of practical information, and no one should 
neglect to acquaint himself with the brilliant and pro- 
found observations and suggestions contained in these 
and similar works*. 

For daily use, however, editions of this kind arc not 
the best. The multitude of fingering, phrasing marks, 
slurs, tempi marks, and suggestions for registering give 
the works an overloaded look, and what is really essential 
remains after all unsaid. What is needed is not these very 
practical editions so much as thorough separate studies 
of registration, manual changes, and the like. Here almost 
everything still remains to be done**. 

An organist of the right kind will not take refuge in a 
practical edition, but will use an original edition, and enter 
in it his own observations and experiments. 

On the structure of Bach's organ themes, again, and 
on the architecture of the works in themselves and in 
relation to his clavier music, scarcely anything has been 



* In Schreyer's edition the phrasing is especially interesting. 
In Naumann's suggestions we recognise the thoughtful practical 
organist, One only regrets that he does not give his principal rules 
and hints in prefaces and notes. A re-issue of them in book form 
would be welcome, 

** An interesting theme, for example, would be a demonstra- 
tion of the possibilities of the natural manual changes in the B minor 
prelude (Peters II, No, 10). A special subject for investigation 
would be the note on which the change is to be made each time. 



318 XIV. The Performance of the Organ Worfcs. 

published as yet that goes to the root of the subject. The 
rhythm of the themes of the organ fugues, it may be re- 
marked, is much simpler than that of the clavier fugues. 
A few quite elementary syncopations apart, scarcely an 
accent falls on the weak part of a bar; the main accent 
always falls on the strong beat. Bach sees quite clearly 
that any other than this natural accent is impossible on 
the organ. For clavier and for orchestra he writes much 
more freely. Thus the object of transcribing clavier fugues 
for the organ is incomprehensible. No one who really 
understands the nature of Bach's organ works can listen 
to transcriptions of this kind. 

Further, the structure of the works is quite different. 
In the organ music Bach works upon much broader and 
simpler lines than in the clavier works. We seek in vain 
in the organ fugues for the subjective life, so rich in sur- 
prises, of the clavier fugues. The former are meant to work 
on a certain inner faculty of conception rather than on 
the immediate feeling, and to exhibit an idea in lofty sim- 
plicity. For this reason Bach's clavier works sound rest- 
less on the organ. A satisfactory registration cannot be 
discovered for any of them. The difference between the 
clavier and the organ styles, in fact, cannot be better 
realised than by placing Bach's organ fugues by the side 
of his clavier fugues, and studying the musical architecture 
of both in detail. 

There is, to be sure, one work of Bach's for the organ 
that stands on the border-line of the style he has laid 
down for his organ works in general, I,t is the A major 
fugue (Peters II, No. 3): 




Every organist can convince himself of this. If we play 
the theme legato, without the articulation that gives 
character to it, it goes very lamely. If we play it as it is 



Organ and Clavier Fugues. Transcriptions. 310 

intended to be played, accenting the syncopations by 
cutting them short, then, no matter how perfectly we 
play the piece, it has a notably restless effect; this is in- 
creased by the fact that the structural interest lies rather 
in the detail than in the whole, to say nothing of the further 
fact that sequences of thirds in the bass and a conclusion 
like that of this fugue are to be found nowhere else in the 
organ works of Bach, 

The interesting point is that we can prove that the 
theme of this exceptional fugue was originally conceived 
not for the organ but for the orchestra; in its primary form 
it was written for the instrumental introduction to the 
cantata Tritt auf die Glctubensbahn (No. 151). 

There is more to be said for the transcription of organ 
works for the piano than for the reverse proceeding, since 
the piano, as Liszt said, is to music what engraving is to 
painting; it serves to multiply and disseminate works of 
art. When masters like Liszt, Saint-Safins, Busoni, Reger, 
Philipp, d* Albert, Vianna da Motta, and Ansorge undertake 
to arrange Bach's organ works for the pianoforte, the 
intelligent player has not, only the advantage of learning 
works from which he would otherwise be barred, but the 
aesthetic pleasure of finding organ effects cleverly realised 
on the piano. Bach, who was himself passionately devoted 
to the art of transcription, would have been delighted 
with the pianoforte apostles of his organ gospel. 

There is danger, however, in going to excess. These 
transcriptions, even when they are made with the utmost 
art, cannot in the long ran give complete satisfaction. 
The organ themes lose something on the pianoforte; the 
simple plan of the works has to be replaced by an artificial 
one, since the various degrees of strength in the organ 
tone cannot be reproduced even on the modern pianoforte. 
When this perception grows, men will some day discover 
what Bach himself experienced, it will look back on the 
age that delighted in transcriptions as on something long 
passed away, and its joy will not be in the transcription 



32O XV. The Clavier Works. 

itself but in the education it afforded*. We must not 
allow these artistic transcriptions, that often surpass the 
powers of the average player, to make us forget the old 
German domestic resource of playing the organ works from 
the original in arrangements for four hands, one player 
taking the manual parts, the other the pedal parts in 
octaves. 



CHAPTER XV. 
THE CLAVIER WORKS. 

Bachgesellschaft Edition. 

Vol. III. Inventions and Symphonies: Klavierubung (Par- 
titas: Italian Concerto: Duets: Goldberg Variations); 
Toccatas in F sharp minor and C minor ; Fugue in A minor. 

- XIII 2 . English and French Suites. (As this volume was 
not critically accurate, the suites were re-issued in Vol. XLV*.) 

- XIV. The Well-tempered Clavichord. 

- XXXVI. Suites, Toccatas, Preludes, Fugues, Fantasias, 
Little Preludes. 

- XLIL Sonatas: Sixteen Concertos after Vivaldi. 

- XLV 1 . Preludes, Fugues, and other pieces. Doubtful 
Works. Supplementary volume. 

The two fugues for two pianos, from the Art of Fugue, are con- 
tained in Vol. XXV. The Klavierbuchlein for Wilhelm Friede- 
raann Bach is given in Vol. XLV 1 ; Vol. XVIII 2 contains the two 
note-books of Anna Magdalena (1722 and 1725). 

Peters Edition, 

Parts I and II, The Well-tempered Clavichord. Ill, Sonatas 
and single Suites. IV, Fantasias, Toccatas, Preludes and 
Fugues. V and VI, Partitas; Italian Concerto; Goldberg 
Variations. VII, Little Preludes; Two-Part and Three- 
Part Inventions; French Suites. VIII, English Suites. 

IX, Toccatas, Preludes, Fantasias, Fugues, Little Preludes. 

X, Vivaldi Concertos. XIII, Additional. Supplement I. 
Series I, Doubtful Works. Pieces from Anna Magdalena's 
Klavierbiichlein. The two fugues for two pianos from the 
Art of Fugue are given in Part XI. 



* Friedrich Spiro holds the same views upon what is perishable 
and what is permanent in the epoch of transcriptions. See Bach 
und seine Transcriptoren, in the Neue Zeitschrift fiir Musik, 1904, 
pp. 680 ff. 



The Publication of the Klavier&bung, 321 

The excellent Bischof f edition of the piano works calls for special 
mention (7 vols, Steingraber edition). The instructive popular 
edition issued by Ricordi with the cooperation of B. and S. Cesi, 
Longo, Marciano, Mtigellini and Philipp is also very interesting, 
though unfortunately it gives only a selection, not the complete 
works. 

The piano concertos will be discussed among the instrumental 
works. 

The clavier works, like those for the organ, mostly date 
from the Weimar and Cdthen periods. Bach, however, 
published only the great works of the Leipzig period, 
six large partitas, the Italian Concerto, four duets, and 
the Goldberg variations. 

Partitas, not suites, though this is what they really 
are in form was the name he gave to the works in 
the style of his predecessor Kuhnau, who had issued, in 
1689 and 1695, two collections of Klat)iertibungen> each 
containing seven Clavichord "Partitas"*, 

The first partita appeared in 1726; it was the first com- 
position that Bach published**. He was at that time 
forty-one years old. Each following year saw the birth 
of a new partita of his. When six of them had appeared, 
he united them under the title of Klavierttbung (Part I), 
again in imitation of his predecessor. "Ubung" means 
here, of course, not so much a work for student's practice 
as one for diversion***. 

If Bach's sons informed Forkel correctly, the work 
made a great sensation in the musical world, "Such ex- 
cellent clavichord compositions had never before been 



* See Spitta III, 155. 

** The Mxihlhausen Ratswahl (Council Election) cantata Gott 
i$t mein KdHig (No. 71), of the same year, had received the honour 
of engraving, not at Bach's instance but at that of the Council. 

*** The title of the volume is JKlavieritbung besUhend in Pr&ludien, 
Allemand&n, JKouranUn, Sarabanden* Giguen, Menuetten und an- 
dern Galanterien* den Li&bhab&rn zur Gemtitsergotxung verfertigt 
Erster Teil, Im Verfag des Autors, 1751. (Clavichord exercises, 
consisting of Preludes, Allemandes, Courantes, Sarabandes, Gigues, 
Minuets and other Galanteries, composed for the mental recreation 
of art-lovers. First Part, Published by the author, 1731). 

Schweitzer, Bach. 21 



322 



XV. The Clavier Works. 



seen or heard. Any one who learned to play a few pieces 
out of them well could make a great success with them"*. 
The second part of the Klavierubung, consisting of the 
Italian Concerto and the B minor partita, was published 
in Nuremberg in 1735 by Christoph Weigel. Even Scheibe 
could not help paying a tribute of admiration to the Italian 
Concerto**, It is interesting to note that Bach got the 
idea of the work from a sinfonia in Muffat's Florilegium 
primum (1695). The similarity of the themes is too strik- 
ing to be explained by mere chance : 



Muff at: 




In 1739 the third part of the KlavierMung appeared. 
It was intended to contain only organ works, the preludes 
on the catechism hymns; the four clavichord duets got 
in by mistake. These organ pieces could be played, of 
course, on the two-manual pedal-clavicembalo, which 
was very popular at that time. How these great works 
were received by organists is not recorded. 

The fourth part of the Klavierilbung was also published 
in Nuremberg, not by Weigel, however, but by Balthasar 
Schmidt, who was also Emmanuel's publisher. It con- 
tained the Goldberg Variations. Goldberg was the clave- 
cinist of Count Kayserling, a patron of Bach, who acted 
as Russian envoy at the Dresden Court. It was he who 
procured Bach the appointment of Court Composer; at 
any rate the diploma came through his hands. Goldberg 
was a pupil of Friedemann, who was in Dresden at that 
time. When he went to Leipzig with his master, which, 



* Forkel, p. 50. 
** See p. 183. 



The Publication of the Klavievtibung. 323 

according to Forkel, often happened, he visited Bach to 
learn what he could from him. Forkel gives the following 
account of the origin of the variations: 

"Count Kayserling fell very ill and could not sleep at night, 
Goldberg, who lived with him, had on these occasions to spend 
the night in an adjoining room, so as to be able to play to him 
when sleepless. Once the Count said that he would like Bach to 
write some clavichord pieces for Goldberg, of a quiet and at the 
same time cheerful character, that would brighten him up a little 
on his sleepless nights. Bach thought the best thing for the pur- 
pose would be some variations, a form which h had previously 
thought rather little of, by reason of the persistence of the same 
basic harmony throughout The Count afterwards al- 
ways called them Ms variations. He could not hear them often 
enough, and for a long time, whenever h had a sleepless night, 
it was "Dear Goldberg, play me one of my variations. 1 ' Bach 
was perhaps never so well rewarded for any of his works as for 
this, The Count gave him a golden goblet, containing 100 louis 
d'or"*. 

That Bach had no particular fondness for the variation 
form may be gathered from the fact that apart from the 
Goldberg set, the only variations he wrote were the youth- 
ful Ana variata alia maniera italiana (B. G. XXXVI, 
pp, 203 208). In his organ music also he soon ceased to 
write variations on chorale melodies. 

The theme of the Goldberg variations is found in the 
KlavierbJlchkin of Anna Magdalena Bach (1723). It is 
the Sarabande that follows the song u Bist du bei mir". 
It had been in existence at least ten years before he thought 
of writing variations on it, 

The variations, however, are founded less on the theme 
itself than on its bass. Over this Bach's imagination plays 
freely, and the work is in reality more a passacaglia worked 
out in chiaroscuro than a series of variations, 

It is impossible to take to the work at a first hearing. 
We have to get to know it, and to understand the music 
of Bach's last period, in which the interest resides not so 
much in the charm of this or that melodic part, as in the 



* Forkel, p. 54. 

21* 



XV. The Clavier Works. 



free and masterly working out of the ideas. When once 
we arrive at this standpoint, we can savour the gentle, 
consoling cheerfulness that gives such warmth to these 
seemingly artificial pieces, In the last variation the cheer- 
fulness becomes laughter of the merriest kind. Two folk- 
songs disport themselves in it : 



Kraut und Rit - ben 



ha - ben mich ver - trie - ben, 




HattmeinMut-ter Fleischge-kocht,so 



ich Ifln-ger blie-ben. 






Ich bin solangnichtbei dirge west; Ruck her, Ruck her, Ruck her. 

Thus Bach in his old age returns to the quodlibet with which 
his ancestors used to enjoy themselves so hugely in their 
great family gatherings, 

Of all Bach's works this comes the closest to the modern 
pianoforte style. If their authorship were not known, 
any one would take the penultimate and anti-penultimate 
variations, even from the mere look of them on paper, to 
be works of Beethoven's last period. 

The Goldberg Variations, the Italian Concerto, and the 
accompanying partita are written for the clavicembalo 
with two keyboards, Even without a positive statement 
to this effect any one would soon realise it in performance, 
in the difficulty, for one keyboard, of the passages in which 
the hands become entangled. 

It is strange that Bach did not think of publishing some 
of his other clavier works. The Well-tempered Clavichord 
indeed, was out of the question; it was too large. A copy, 
according to the prices of that day, would have cost at least 
ten or fifteen thalers, Why, however, were the French 



The French and English Suites. 325 

and the English suites not published? Perhaps because 
they did not strike him as sufficiently difficult aad in- 
genious. As he could permit himself- the trouble and 
luxury of an engraved issue only in a very limited degree, 
he preferred to expend them on works that would win 
him honour and recognition among professional musi- 
cians and connoisseurs. As compositions in those days, 
however, were valued less for their aesthetic qualities than 
for their ingenuity, it would not have benefited Bach to 
have published these simple suites. 

Nevertheless it is a mistake to suppose that his other 
clavier works were not widely diffused, Tjhey were ob- , 
tainable in manuscript copies. After 1720, indeed, there 
was scarcely a good German musician anywhere who did 
not possess at least one work of J. S Bach. As early as 
1717, in his book Das leschtttzte Orchester* Mattheson 
reckons the * 'celebrated Weimar organist, Herr Joh. Se- 
bastian Bach", among the leading composers, on the 
ground of some works of his that he had seen. 

Besides the seven partitas that appeared in the Klavier- 
ttbung, Bach wrote fifteen other suites, the six French, 
the six English, and three smaller suites, that may have 
been sketches for the French*. It is not * known how the 
French and English suites acquired these names. Even 
Forkel could give no precise information on the point**. 
He conjectures that the former were so called because they 
are written in the French style, and the latter "because 

* The French and English suites are in B. G. XIII a ; as the 
edition of 1863 was not based on the autographs, these works were 
re-issued in 1895 in Vol. XL V 1 , The three youthful suites (in 
A minor, E flat major, and F major) are in VoL XXXVI (1866). 
In the same volume are some fragments of suites and separate 
dance-pieces. The suite in E major (XLII, pp. 16 ft.) is an arrange- 
ment of some instrumental work. The authorship of the suite in 
B flat major (XLII, pp. 213 ff) and the "saraband e con partita" 
(XLII, pp. 221 fl) is doubtful. The passacagHa in D minor (XLII, 
pp. 234 ff) is certainly not by Bach, It is by Christian Friedrich 
Witt (d. 1715). 
** Forkel, p. 56. 



336 XV. The Clavier Works. 

the composer wrote them for an Englishman of quality/' 
which latter was certainly not the case. At a later date 
an unsuccessful attempt was made to give the partitas 
the title of "German suites". 

The French suites figure, though not quite complete, 
in the first KlavierbUchlein of Anna Magdalena Bach 
(1722). There is also an autograph of them with the in- 
scription: "Sex Suiten pur le Clavesin compossee par Mos: 
J. S. Bach." The title of the autograph of the English 
suites is also in French, but written more correctly. There 
is a very valuable copy of the two collections of suites in 
Gerber's hand-writing, made between 1725 and 1726, 
when he was a pupil of Bach. 

The French suites can be proved to be not later in date 
than the Cothen period. The English suites also, in all 
likelihood, belong to the same time, though all the manu- 
scripts and the copies that we have of them fall within 
the first Leipzig period. In the first year of his work at 
St. Thomas's, Bach had to write a new cantata for almost 
every Sunday, so that he could have had little time for 
other works. 

The suite owes "its origin to the pipers of the seventeenth 
century, who used to string together various national 
dances. The German- clavichord players adopted the form 
from them and developed it. The rule was that it should 
consist of at least four pieces, the allemande, the cou- 
rante, the sarabande, and the gigue. The alleniande is 
in easy 4 /4 time, with a quaver or semiquaver up-beat; 
the courante or corrente is in 3 /2 time, and is characterised 
by its uninterrupted sequences of equal notes; the sara- 
bande is a grave Spanish dance, also in 3 / 2 time, the heavy 
notes of which are surrounded by coquettish embellish- 
ments ; the gigue as a rule goes evenly and rapidly, and may 
be in all kinds of triple rhythms. It gets its name from 
the gigue (ham, or gammon) the satirical French name 
for the older violins; thus a gigue really means a fiddler's 
dance. 



The French and English Suites. 327 

There was no reason for refusing admission to the suite 
to other dance forms that cropped up later. The French, 
especially Marchand and Couperin, made a point of in- 
troducing aU possible dances. Their suites contain the 
gavotte, in 2 /2 time, with a half-beat up-take; the minuet, 
in simple triple rhythm; the passepied, a Breton dance 
similar to the minuet, which, under Louis XIV,, made its 
way into the French ballet; the bounce, in quick */4 time, 
an angular dance originating in Auvergne, The French 
also incorporated into their suites the rondeau, the rigau- 
don, the polonaise, and even independent movements in 
no particular dance-form. 

Bach takes all these rich suite-forms over from his 
French models, but preserves moderation where they run 
to extremes*. He follows tradition in placing the dances 
that were not originally part of the suite between the sara- 
bande and the gigue, so that the latter forms the con- 
clusion. He generally places the extraneous movements 
at the beginning. Thus the English suites open with 
preludes, and the great partitas in the Klavierilbung with 
preludes, symphonies, fantasias, , overtures, preambles and 
toccatas; the French suites, however, begin at once with 
the allemande. 

Naturally some of these dances were somewhat altered 
in the clavier suite. The gigue, for example, which runs 
to considerable length in the suite, in its original dance* 
form consisted merely of two eight-bar phrases with repeats* 
The Italian composers as a rule retained only the metre 
and rhythm of the various dances, without troubling to 
preserve their essential character. The French were more 
scrupulous in this respect, and made a point of pursuing 
to its conclusion the rhythmical characteristic of each 
dance-form**. Bach goes still further; he always vitalises 



* On the relation of Bach to the contemporary suite see Spitta 
II, 84 ft A full history of the suite will be found in Weitzmann's 
GesMchte der Klaviermusik* 3rd edition by Max Seiffert I, 91 ft 
** See Spitta II, 84 if. 



XV. The Clavier Works. 

the form, and gives each of the principal dance-forms a 
definite musical personality. For him the allemande 
represents vigorous but easy motion; the courante repre- 
sents a measured haste, in which dignity and elegance go 
side by side; the sarabande represents a grave and majestic 
walk; in the gigue, the freest of all forms, the motion is 
quite fancy-free. He thus raises the suite-form to the 
plane of the highest art, while at the same time he pre- 
serves its primitive character as a collection of dance- 
pieces. 

As with the organ music, so among the clavier works 
are a number which Bach wrote as teaching pieces for his 
sons and pupils. His clavier school consisted of the preludes 
for beginners, the two-part and three-part Inventions, 
and the Well-tempered Clavichord. 

Of the preludes for beginners we have altogether eighteen.* 
Seven are in Friedemann's Klavierbiichlein; six more are 
contained in an old manuscript with the title, Six PrSludes 
a Pusage des Commengants composes par Jean S&bastien 
Bach: they were published for the first time by Forkel**. 
The others have been handed down to us through pupils. 

Even in these little works the overwhelming greatness 
of Bach is revealed. He merely meant to write a few simple 
exercises; what he actually wrote were compositions that 
no one forgets who has once played them, and to which 
the adult returns with ever new delight. Particularly 
captivating are the prelude in C minor (p. 119), with its 
dreamy arpeggio semiquavers; the clear-cut prelude in 
D major (p. 131); and the jubilant one in E major (p. 132), 
the intoxication of which must have been one of the richest 
youthful experiences of any one who thoroughly grasped it. 

The title of the principal autograph of the Inventions 
and symphonies runs thus: "An honest guide, wherewith 
lovers of the clavier, and especially those anxious to learn, 

* B. G. XXXVI (1886), pp. 118127. 
** See p. 54 of his book. These were the only six that Forkel 

knew. 



The Little Preludes, Inventions and Sinfonias, 329 

are shown a clear method not only how to learn to play 
neatly in two parts, but further to play correctly and well 
in three obbligato parts; and at the same time not only 
to acquire good inventiones [ideas] but to work them out 
well; but above all to attain a cantabile style of playing, 
and in addition to get a strong taste for composition. 
Written by Joh. Seb. Bach, Hochf. Anhalt-Cdthen Kapell- 
meister. Anno Christi 1723," 

Besides this autograph we possess two others, of an 
earlier date. Friedemann's Klavterbtichlein, begun in 1720, 
contains most of these compositions, but the title is dif- 
ferent; instead of "Invention" we have "praeambutum", 
and instead of "symphony" we have "fantasia"*. When 
Bach copied them out again he altered the arrangement of 
the pieces ; they still follow each other in the order of their 
keys, but to each invention he added the corresponding 
symphony, which is justified by the fact that the two-part 
and the three-part pieces, as is shown by certain thematic 
similarities, were, as a rule, written at the same time. In 
the definitive autograph he again distinguished, on didactic 
grounds, between two-part and three-part pieces, Here 
again the final work is a strict selection made by Bach 
himself from a large number of works of the same kind, 
relieved by a number of smaller pieces, splinters, as it 
were, from the workshop**. 

Bach's inability to settle upon the title was due to the 
absolute novelty of this kind of work. He abandons the 
binary song-form that was customary for small clavier 
pieces, and which he himself had used in the six preludes 
for beginners, and creates a new form of his own, marked 
by no external divisions, but giving free play to the natural 
development of the musical idea. The result was that 
his invention and manner of working became less melodic 
than thematic and "motival". The same principle would 
necessarily have led from the da capo aria to the freer 

* SeelaG. XLV, pp. 2x3!! 
** Compare Spitta II, 58 ff. 



330 XV. The Clavier Works. 

song-form. Here, however, he never quite won his freedom, 
though he often seems on the way to it. 

The title * 'Invention" for a clavier piece seems not to 
have been devised by Bach himself, as was formerly thought, 
but to have been derived from some unknown composer 
whose works he copied out at that time for his sons *. He 
might just as well have called all the pieces simply "pre- 
ludes" ; but he thought this title too general and not suf- 
ficiently characteristic for the strict contrapuntal working- 
out that he had in view, 

The Inventions are written not for the clavicembalo 
but for the clavichord, which at that time was called 
simply the "clavier". It was only on this instrument 
that the "cantabile style" of playing was possible that 
Bach had chiefly in his mind when writing them. Thus 
in the history of pianoforte playing the Inventions and 
symphonies are a protest against the dulcimer-like tinkling 
that was the vogue at that time and not only at that 
time. We feel in every bar of these pieces that the idea 
at the root of them has been the singing and modulatory 
capacity of the instrument. 

When Bach penned the title and expressed the hope 
that these pieces would give players a strong taste for 
composition, he could not anticipate how amply his wish 
would be realised. If the average modern musician, in 
spite of his possessing less theoretical knowledge of the 
technique of composition than those of Bach's day, at 
any rate has a clearer sense of the distinction between 
true and false art, it is primarily due to these little works 
of Bach. The child who has once practised them, no 
matter how mechanically, has acquired a perception of 
part- writing that he will never lose. He will always in- 
stinctively look for the same masterly weaving of the voices 
in every other piece of music, and feel the poverty of the 
music where this is lacking. And any one who has studied 



* Spitta II, 659, 660. 



Origin of the Well-tempered Clavichord. 331 

the pieces thoroughly, in their formal and aesthetic aspects, 
under a capable teacher, has henceforth a criterion of true 
art, whether he himself becomes a composer or, as in the 
majority of cases, simply an executant. In any case 
Bach's title, with its evident desire that clavier study 
should not be an end in itself but an introduction to com- 
position, is of significance to piano teachers today. 

These outwardly similar compositions fall into certain 
particular types, which are distinguished from each other 
according as the development of the piece is purely formal 
or conditioned by an imperative dramatic idea. The first 
type is represented by the well-known F major Invention, 
the other by the E minor and F minor symphonies. Closely 
considered, however, each of these works is a masterpiece 
sui generis, with no exact analogue among the others. Only 
an infinitely fertile mind could venture to write thirty 
little pieces of the same style and the same compass, and, 
without the least effort, make each of them absolutely 
different from the rest, In face of this inconceivable 
fertility it seems almost a superfluous question whether 
any other of the great composers has had an inventive 
faculty so infinite as Bach's. 

The two Parts of the Well-tempered Clavichord belong 
to widely separated periods*. The first was finished in 
1722, as appears from the dating of the autograph by 
Bach himself; the second was compiled in 1744, as we 
learn from the Hamburg organist Schwenke, who in 1781 
made a copy of it from an autograph (now lost) belonging 
to Emmanuel, the title-page of which bore the date 1744. 

In Friedernann's Klavierbttchlein of 1720 are found 
eleven preludes from the First Part, among them the one 
in C major. Bach's revisions of this and three others (in 
C minor, D minor, and E minor) made it probable that 
the majority of the pieces of the Well-tempered Clavier 
did not achieve their present perfection at the first stroke, 

* B. G. XIV (18166). 



332 XV. The Clavier Works. 

but were continually worked over by the composer with a 
view to giving them a form that would satisfy him. 

Gerber, in his Dictionary, says that Bach composed the 
First Part of the Well-tempered Clavichord at a place 
where time hung heavily on his hands and no musical 
instrument was available. There may be some truth in 
this. Gerber* s father had been Bach's pupil in the early 
Leipzig years, so that the tradition may quite well be based 
on some remark of Bach's, especially as we know that 
Gerber was studying the Well-tempered Clavichord at that 
time, and Bach himself played it to him thrice*. Bach 
may well have been in such a situation during some journey 
with Prince Leopold of Cothen, when the small portable 
clavier that figures in the list of the Court instruments 
would be left behind **. The tradition is at any rate correct 
to this extent, that the majority of pieces in the Well- 
tempered Clavichord were written in a relatively short time. 
This manner of production was indeed characteristic of 
Bach. The Second Part was written after he had practically 
finished with cantata .writing. 

A number of preludes and fugues, however, existed for 
some time before Bach conceived the idea of a collection. 
This holds geod for the Second Part no less than for the 
First. In both there are pieces which, in their original 
form, really go back almost to the composer's earliest 
years. Any one thoroughly conversant with Bach will 
gradually discover for himself which pieces belong to this 
category. He will at once see, for example, that of the 
preludes of the First Part, those in C minor and B flat 
major do not show the same maturity as most of the others. 
That the A minor fugue from the same Part is a youthful 

* The tradition oi the origin of the First Part of the Well- 
tempered Clavichord is given on p. 90 of the first volume of the 
Dictionary, and the account of Bach's teaching of Gerher the elder 
on pp. 490 fl of the same volume. 

** As a rule the prince took with him, when travelling, a com- 
plete sextet of his chamber musicians. See Bunge, /. S. Backs 
KapelU zu Coihen, in the &ach~Jahrbuch for 1905, pp, 27 and 42. 



Origin of the Well-tempered Clavichord. 333 

work is shewn not only by a certain thematic looseness 
and lack of design, but also by the fact that it is evidently 
written for the pedal clavicembalo, The final note in the 
bass, prolonged through five bars, cannot be sustained by 
the hands alone, but needs the pedal, as is often the case 
in the early works. Otherwise the Well^tempered Clam- 
chord, like the Inventions and the symphonies, is designed 
primarily for the clavichord, not for the clavicembalo. 
Bach himself does not appear to have called the 1744 col- 
lection the Second Part of the Well-tempered Clavichord, 
but simply "Twenty-four new preludes and fugues". 

He inscribed the work completed in Cothen the Well- 
tempered Clavichord by way of celebrating a victory that 
gave the musical world of that day a satisfaction which 
we can easily comprehend. On the old keyed instruments 
it had become impossible to play in all keys, since the 
fifths and thirds were tuned naturally, according to the 
absolute intervals given by the divisions of the string. 
By this method each separate key was made quite true; 
the others, however, were more or less out of tune, the 
thirds and fifths that were right for their own key not agree- 
ing among each other. So a plan had to be found for tuning 
fifths and thirds not absolutely but relatively, to "tem- 
per" them in such a way that though not quite true in any 
one key they would be bearable in all. The question had 
really become acute in the sixteenth century, when the 
new custom arose of allotting a separate string to each note 
on the clavichord; previously the same string had been 
used for several notes, the tangents dividing the string 
into the proper length for the desired tone. The organ 
also imperatively demanded a tempered tuning. 

The question occupied the attention of the Italians 
Giuseppe Zarlino (1558) and Pietro Aron (1529)*. At a 
later date the Halberstadt organ builder Andreas Werk- 
rneister (1645 1706) hit upon a method of tuning that 

* See the B. G. XIV, p. 25 of the preface; and the chapter 
on this question inWeitzmann-Seiffert's Geschichte desKlavier spiels. 



334 x ^ T he Clavier Works. 

still holds good in principle. He divided the octave into 
twelve equal semitones, none of which was quite true. 
His treatise on Musical Temperament appeared in 1691. 
The problem was solved; henceforth composers could 
write in all keys. A fairly long time elapsed, however, 
before all the keys hitherto avoided came into practical 
use. The celebrated theoretician Heinichen, in his treatise 
on thorough-bass, published in 1728, i. e. six years after 
the origin of Bach's work confessed that people seldom 
wrote in B major and A flat major, and practically never 
in F sharp major and C sharp major*; which shows that 
he did not know Bach's collection of preludes and fugues. 

At one time it looked as if Bach were to be deprived 
of the honour of having written the first Well-tempered 
Clavichord. In 1880 there came to light a manuscript of 
one Bernhard Christian Weber, organist at Tennstedt, 
bearing a very similar title to that of Bach's work, and 
with the date "1689" in red crayon. The excitement, 
however, was soon allayed by WilhelrnTappert's demonstra- 
tion that the manuscript was not the work of a forerunner 
but of a mediocre imitator of the middle of the eighteenth 
century. If any one can be regarded as a forerunner of 
Bach it is Mattheson, who, in his Organistenprobe (1719), 
in the article on thorough-bass, advocated the employment 
of all keys and gave two examples, a difficult one and an 
easy one, in each**. When this work appeared, however, 
Bach was already engaged upon the Well-tempered Clavichord. 

Bach found an imitator in his admirer Georg Andreas 
Sorge (1703 1788) the Lobenstein organist, who also 
wrote preludes and fugues in all the twenty-four keys, 

* See Spitta II, 162. To understand the problem of the twelve 
equal semitone system we must remember that the octaves had 
all to be kept true, while the intervals constituting them had to 
be tempered. 

** See Wilhelm Tappert's Das Wokltewperierie Klavier, in 
Eitner's Monatshefte fur Musikgeschichte, 1899, pp. 123 fl, for an 
examination of all the pre-Bachian works that claim to make use 
of all the twenty-four keys. 



Autographs of the Well-temperad Clavichord. 335 

and issued the work in 1738 through Bach's publisher, 
Balthasar Schmidt of Nuremberg*. 

For a long time it was taken for granted that the auto- 
graph of the Second Part of the Well-tempered Clavichord 
was lost. In the middle of the eighteen -nineties Sir George 
Grove and Ebenezer Prout announced that it still existed 
and had just passed from private hands into the possession 
of the British Museum. It had come to England through 
Muzio dementi, who had acquired it in some unknown 
way; after his death it had been bought by a Mr. Emett, 
at whose house Mendelssohn saw it in 1842 and recognised 
it as a genuine autograph; his daughter sold it to her friend 
Miss Eliza Wesley, who bequeathed it to the British Mu- 
seum on her death in 1895, It is not the original autograph, 
but a copy carefully made by Bach himself, in which he 
had continued each prelude and fugue on a separate sheet 
in such a way that he did not need to turn over when 
playing it. Unfortunately three of these leaves are lost**. 

We have several autographs of the First Part of the 
Well-tempered Clavichord, For each of his two elder sons 
Bach himself wrote out the work with the utmost care. 
Friedemann gave his copy Jo Muller, the cathedral organist 
at Brunswick, with whom he sometimes stayed after he 
had given up his post at Halle; it is now in the Berlin 
Royal Library. Emmanuel's copy was sold in 1802 by 
his daughter to the Ziirich publisher Nageli, and is presum- 
ably still in private hands in that city***. Another auto- 



* Spitta II, 671 (German edition). 

** For a history and description of this autograph see 0. Taub- 
mann's Ein Autograph de$ zweitert Teils von Backs WohlteMperiertert 
Rlaviev, in the Allgemeine Musihxeitung for 1896, (which gives 
the English axithorities), and Alfred Dftrffel's remarks in the preface 
to the B. G. XLV (1895), pp. 6872. 

*** This autograph was not available lor the B, G, edition of the 
Well-tempered Clavichord, See Spitta II, 665 ff . In 1 885 it belonged 
to Herr Stadtrat Hagenbuch, president of the Allgememe Muaik- 
gesellschaft of Ziirich. Ddrffel, who had the opportunity of ex- 
amining it, regarded it as a fairly rough copy. See his preface to 
B, G. XLV, pp. 65 ff. 



336 XV. The Clavier Works, 

graph of the year 1722 was once in the possession of a 
Herr Volkmann, of Pesth, where, in the middle of the 
eighteen-forties, it was involved in an inundation of the 
Danube, traces of which it still bears. It is called, after 
a later owner, the Wagener autograph*. 

There was one copyist in particular who thought it his 
duty to improve Bach, and tried to do so in every prelude 
and fugue in both Parts. His chief concern was to rid 
them of all unnecessary complexity, and to give them 
the form that Bach himself would have conferred on them 
had he lived in another epoch than that of the "Zopf * 
(pigtail) and had a more refined taste. He transforms, 
for example, the D major fugue of the First Part in this 
way 




Forkel had a number of the preludes and fugues from 
each part in a still more drastically curtailed form. It 
seems almost incredible that he, and after him the Bach 
biographers Hilgenfeldt and Bitter, should have regarded 
these versions as the authentic ones, and energetically main- 
tained their superiority over those generally accepted **. 
The children of Bach's muse slink along in the most mis- 
erable shapes, they are merely skin and bones; and the 

* The Imperial Library at Vienna possesses no autograph of 
the Well-tempered Clavichord, as might be -supposed from a some- 
what inaccurate article by Bunge in the Bachjahrbuch for 1905 
<P 32). 

** Both these redactions are now in the Berlin Royal Library. 
The Forkel form of the C major prelude from the First Part is given 
in the supplement to B, G. XIV. It contains only twenty-four 
bars instead of thirty-five. It should further be noted that bar 24 
(G in the bass), as it is given in most editions, is an error. It is an 
interpolation derived from Schwenke's copy of 1781, which is un- 
trustworthy in other respects. 



Editions of the Well^tBmpered Clavichord. 337 

man who knew Bach's sons, who had heard them play, 
and in whom one might presuppose some "breath of the 
Bach spirit, fell a victim to this clumsy deception ! Even 
Zelter had to try his hand at simplifying Bach, So great 
was the aesthetic suspicion with which people at that 
time regarded the works of the musical rococo period, 
even when they bore the name of Bach! 

The title of the First Part runs thus in the autograph ; 
"The Well-tempered Clavier, or preludes and fugues 
in all tones and semitones, both with the tertiam majorem 
or Ut, Re, Mi, and the tertiam minorem or Re, Mi, Fa. 
For the profit and use of young musicians desirous of know- 
ledge, as also of those who are already skilled in this studio^ 
especially by way of pastime; set out and composed by 
Johann Sebastian Bach, Kapellmeister to the Grand Duke 
of Anhalt-Cothen and Director of his chamber music. 
Anno 1722." 

The first edition of the Well-tempered Clavichord was 
issued by the Englishman Kollrnann in 1799*; in the 
following year the work was issued simultaneously by 
Nageli of Zurich and Simrock of Bonn, In the latter 
edition the Second Part comes before the first **. The 
first Peters edition appeared in 1801. Breitkopf and Hartel 
did not bring it out until 1819. The contemporary Paris 
and London editions are only pirated reprints of the Nageli 
version. The first satisfactory text was that of KroU in 
the B. G. edition (XIV, 1866)***. 

* The circle of Kollmann and Wesley was warmly enthusiastic 
lor Bach. Wesley called him simply 'The Man". Bach-recitals 
were given, and a subscription opened for a complete English 
edition of his works. See Kretzschmar*s preface to B. G. XXVI, 
p. 24. , In 1 81 2 KoEmann published "An. Analysis of S, Bach's 
Preludes and Fugues 11 in the Quarterly Musical Register. 

** The second Peters edition (1837) was edited by Czerny, and 
the third (1862/3) by KroU. 

*** Among other editions that of Bischoff, published by Stein- 
gr^ber, calls for special mention, The same firm also published 
Stade's thoughtful and searching analyses of the fugues of the 

Schweitzer, Bach. 22 



33& XV, The Clavier Works. 

The Well-tempered Clavichord is one of those works by 
which we can measure the progress of artistic culture from 
one generation to another. When Rochlitz met with these 
preludes and fugues at the beginning of the nineteenth 
century, only a few of them really appealed to him. He 
placed a tick against these, and was astonished to find 
how the number of these ticks increased as he played the 
works*. If some one had told this first of Bach prophets 
that in another hundred years every musically -minded 
man would have regarded each piece in the collection as 
perfectly easy to comprehend, he would hardly have 
believed it. 

The fact that the work today has become common 
property may console us for the other fact that an analysis 
of it is almost as impossible as it is to depict a wood by 
enumerating the trees aud describing their appearance. 
We can only repeat again and again take them and 
play them and penetrate into this world for yourself. 
Aesthetic elucidation of any kind must necessarily be super- 
ficial here. What so fascinates us in the work is not the 
form or the build of the piece, but the world-view that is 
mirrored in it. It is not so much that we enjoy the Well- 
tempered Clavichord as that we are edified by it. Joy, 
sorrow, tears, lamentation, laughter to all these it gives 
voice, but in such a way that we are transported from the 
world of unrest to a world of peace, and see reality in a 
new way, as if we were sitting by a mountain lake and 

Well-tempered Clavichord. We may mention also Carl von Bruyck's 
Technische und asthetische Analysen des Wohltempeyierten Klaviers 
(Breitkopf and Hartel, 2nd Edition 1889), Jadassohn's Erlaute- 
rungen zu ausgewdhlten Fugen aus J. S. Backs Wohltempeyierten 
Kla-vier, (the Supplement to his Lehrbuch des Kanons und der 
Fuge, Leipzig, 1888); Hugo Riemann's Analyse des Wohltempe- 
rierten Klaviers und dev Kunst der Fuge (Leipzig, 1890, 1891, 1894); 
E. von Stockhausen's Die harmonische Grundlage von zwolf Fugen 
aus J. S. Backs Wohltemperiertem Klavier (Leipzig, W, Weber) ; 
and Wie studiert man J. S. Backs Wohltemperiertes Klavier?, in 
the Neue Musikzeitung, 1904. 
* See ante, p. 237. 



Detached Preludes and Fugues. 339 

contemplating hills and woods and clouds in the tranquil 
and fathomless water. 

Nowhere so well as in the Welt-tempered Clavichord are 
we made to realise that art was Bach's religion. He does 
not depict natural soul-states, like Beethoven in his sonatas, 
no striving and struggling towards a goal, but the reality 
of life felt by a spirit always conscious of being superior 
to life, a spirit in which the most contradictory emotions, 
wildest grief and exuberant cheerfulness, are simply phases 
of a fundamental superiority of soul. It is this that gives 
the same transfigured air to the sorrow-laden E flat minor 
prelude of the First Part and the care-free, volatile prelude 
in G major in the Second Part. Whoever has once felt 
this wonderful tranquility has comprehended the mys- 
terious spirit that has here expressed all it knew and felt 
of life in the secret language of tone, and will render Bach 
the thanks we render only to the great souls to whom it 
is given to reconcile men with life and bring them peace* 

Half a dozen connected preludes and fugues and a dozen 
isolated fugues remained over after the compilation of 
the Well-tempered Clavichord, Bach apparently not think- 
ing them important enough to be included in that collec- 
tion*. Two of them, in A major and B minor respectively, 
are based on themes by Albinoni**. 

Other preludes arid fugues were too large and too self- 
subsistent for it to be possible to include them in a collec- 
tion. Of this kind are the fantasia and fugue in A minor*** 
and the prelude and fugue in the same keyf, which are 
among the grandest things in piana literature, , Later on, 
Bach rearranged the prelude and the fugue, with con- 
summate technique, as a concerto for flute, violin and 



* These preludes and fugues will be found in B. G. XXXVI ; 
No. i? is not by J, S. Bach, but by Jolmnn Christoph Bach of 
Eisenach. 

** B. G. XXXVI, pp. 173 if. and 178 if, 
*** B. G. XXXVI, p. 8 1 ft 
t B. G. XXXVI, p. 91 11 



340 



XV. The Clavier Works. 



clavier with orchestral accompaniment, adding, by way of 
adagio, an expanded version of the middle movement of 
the third organ sonata*. The prelude begins thus 




and the fugue thus 




He notates the later, in the orchestral concerto, in V* 
time, and combines it with a broad and free tuttL This 
A minor prelude and fugue probably belong, in their 
original form, to the Cothen period; they certainly existed 
as early as 1725**. It was perhaps rearranged as an 
orchestral concerto at the beginning of the thirties, when 
Bach conducted the performances of the Telemann Musical 
Society and had need of concerted pieces. 

The fugue in A minor, with its introduction of arpeggio 
chords, would perhaps be better described as a fugal fan- 
tasia. The spirited and brilliant work contains no fewer 
than one hundred and ninety-eight bars***. 

The Chromatic Fantasia and Fugue f was from the first 
one of the most popular of Bach's clavier works, as is proved 
by the many manuscript copies of it that we possess, dating 
from his own and from later times. The first is found 
in a volume bearing the date 1730. The work, however, 
is much older than this; it probably dates from the same 
epoch 1720 as the great G minor organ fantasia. 
It has a kind of inner affinity with this work; not only 



* The concerto will be found in B. G. XVII, p. 223 ff. 
** This date appears on a manuscript copy of Peter Kellner's, 
*** B. G. Ill, p. 334 ff. See also ante, p. 271, and Oppel in the 
Bachjahrbuch for 1906, pp. 74 78. 
t B. G. XXXVI (1886), p. 71 ff. 



Fantasias and Sonatas. 341 

does the same peculiar fire burn, in each of them, but in 
both the recitative style is carried over into an instru- 
mental medium. 

In the C minor fantasia Bach employs the Neapolitan 
clavier style that had been founded by Alessandro and 
Domenico Scarlatti, one of the main effects of which con- 
sisted in the crossing of the hands*. Bach had already 
made use of this effect in the gigue in the B flat major suite 
of the First Part of the Klamerubung. The C minor fantasia 
may belong to the same period as the Italian concerto; 
perhaps it was written somewhat later, about the end of 
the thirties. The autograph which we possess points to 
this period. A fugue was appended to this fantasia, but 
the autograph unfortunately gives only the first forty- 
seven bars of this; they are rich in promise**. This does 
not imply that Bach left it unfinished. It certainly lay 
before him complete when he made the copy we now have 
of the C minor fantasia ; only he did not get as far as the 
copying of the whole fugue. By this accident it is lost to 
us. This is doubly regrettable because, judging by the 
commencement of it, it was a very individual fugue, more 
in the form of a fantasia. The theme is constructed on 
one of those chromatic sequences that we so often meet 
with in Bach's fugues. It runs thus 



Four larger works of Bach for the clavier, in several 
movements, have come down to us in the form of clavier 
sonatas. Two of these, however, must be deducted, 
those in A minor and C major Spitta having discovered in 
1881 that they are only arrangements and amplifications 



* B. G. XXXVI, p. 145 fl Some other fantasias, mostly 
youthful works, are also given there, among them one in A minor, 
the style of which is particularly interesting (p. 138 ff). 

** The fragment will be found in B. G. XXXVI, p. 238. 



342 



XV. The Clavier Works. 



of instrumental pieces from Adam Reinken's Hortus 
musicus*. 

The D minor sonata is only a transcription for the clavier 
of the second sonata for solo violin**. The sonata in D major 
is therefore the only original composition for the clavier***. 
In this, however, there is hardly any originality, the youth- 
ful composer haying written it under the influence of 
Kuhnau's clavier sonatas. In the final fugue he amuses 
himself by imitating the cackling of a hen, working out 
this theme 







and a counter theme 

A^r-^T^-r 



in m^rry if not very witty style. A note in Italian tells 
us what the piece is meant to represent. 

Seven clavier compositions in several movements are 
designated "toccatas*'. They might with equal propriety 
have been called sonatas. At that time every piece in 
several movements for a keyed instrument could be called 
a toccata, without the title implying anything as to its 
special form. Of these seven toccatas, five those in 
D major, D minor, E minor, G minor and G major 

* These two sonatas are given in B. G. XLII, pp. 29 ff , and 42 f f. 
In the first edition of his book, Spitta thought 'they were particularly 
characteristic of Bach. The essay in which he gives the real facts 
will be found in his MusikgeschiMliche Aufs&tze (Berlin, 1894) 
p. in ff. Reinken's Hortus musicus has since then been brought 
out by Riemsdijk. See also ante, p. 194; [and Spitta's Bach I, 
429, 430- TR,] 

** B. G. XLII, p. 3 ff. The sonata for solo violin (B. G. XXVII*, 
p. 24 ff.) is in A minor. 

*** B. G. XXXVI, p. 19 ff. There is another piec in A minor 
(XLV, p. 1 68 ff.) called a sonata. It may be an arrangement of 
some orchestral piece. 



Toccatas and Capriccios, 343 

belong to the first Weimar period*; the other two in 
F sharp minor and C minor seem to have been composed 
somewhat later**. As a whole the G minor toccata is the 
most interesting. The melancholy adagios of the D minor 
and the G major toccatas are touching in their simplicity. 
In the two later toccatas the total impression is some- 
what weakened by the lack of finish in the design. In 
the F sharp minor toccata we find already the descending 
chromatic theme that later dominates the Crucifixus of 
the B minor Mass. It is here in the following form 




Quite $m generis is a "capriccio" in B flat major, pro- 
bably written in Arnstadt in honour of Bach's second 
brother Johann Jacob***, who had enlisted as oboist in 
the Swedish guard when Charles XII was in Poland in 
1704, The nineteen-year old Johann Sebastian may have 
written the Capriccio sopra la lontananza del suo fratetta 
dilettissimo ("Capriccio on the departure of his beloved 
brother") for the family leave-taking. It begins with an 
arioso, inscribed "Cajoleries of his friends, who try to deter 
him from his journey"; then comes an andante, meant 
to be "a representation of the diverse accidents that may 
befall him in foreign lands"; the "general lamentation of 
his friends" is depicted in a passacaglia-like adagissimo 
on a descending chromatic theme suggesting that of the 
Crucifixus in the B minor Mass ; in the following move- 
ment "the friends, seeing that it cannot be otherwise, come 



* B. G. XXXVI, p. 26 ff. The commencement of the toccata 
in D major reminds us of the organ work in the same key (Peters 
IV, No. 3). The toccata in A major (B. G, XLII, p. 243 ff.) is not 
by Bach, but by Purcell, 

** B. G. Ill, pp. 311 ff. and 322 ff. 

*** B. G. XJ&XV1, pp. 190 196. On the aesthetic significance of 
this work see the chapter on "Word "and Tone in Bach". 



344 xv - The Clavier Works. 

and say farewell"; thereupon comes the "Aria of the Pos-* 
tilion"; and a "fugue in imitation of the postilion's horn- 
call" ends the delightful work, that follows in the footsteps 
of Kuhnau's Biblical histories represented in clavier sonatas, 
published four years earlier (1700). 

A capriccio in E major, that probably belongs to the 
same period, is not so interesting. It may have been written 
in honour of his eldest brother, Johann Christoph Bach of 
Ohrdruf, by whom the composer was brought up*. 

Among the clavier pieces are some that Bach designed 
also for the lute, and were even, perhaps, written in the 
first place for that instrument**. This is the case with the 
little prelude in C minor 




Recent researches show that the prelude in E flat major 
(B. G. XLV, p. 141), the suite in E minor (B.. G. XLV, 
p. 149 ft), that in E major (B. G. XLII, p. 16 ft), and that 
in C minor (B. G. XLV, p. 156 ff.) are also clavier arrange- 
ments of compositions for the lute ; the fugue of the G minor 
sonata for solo violin and the Suite discardable for violon- 
cello solo have also come down to us in lute tablature. 
The three Bach partitas for lute mentioned in Breitkopf s 
catalogue of 1761 are thus not lost, as was almost uni- 
versally supposed. We may therefore probably answer 
in the affirmative the question whether Bach himself 
played the lute. 



* B. G. XXXVI, pp. 197212. 

** The following remarks are based on Wilhelm Tappert's Se- 
bastian Backs Komposition&n fur die Laute> Berlin, 1901 (reprinted 
from the Redende Kiinste, 6th year, Nos. 36 40). In this essay 
the author condemns the editors of the B. G. edition for not having 
issued the lute compositions separately. His reproof, however, is 
only partly justified. 



Bach's Ornamentation. 



345 



CHAPTER XVI. 

THE PERFORMANCE OF THE CLAVIER WORKS. 

The ordinary player of Bach finds the ornamentation 
one of the greatest, difficulties. It is a book with seven 
seals to him. In reality the question is by no means so 
complicated as it appears at first sight*. 

We must start from the explanations which Bach him- 
self gives on the third page of Friedemann's Klavierbuch- 
lein (1720), under the heading "Explanation of divers 
signs, showing how to play certain ornaments neatly". 
Bach elucidates each sign by writing out in full the manner 
of playing it, thus 



Trill. 



Mordent Trill and Mordent. Cadence. Double-Cadence. 



( 


i _., jj-.p... . 


_! , 




.,, nrr ,.. llr - L |l.u.|"-,-^,,,r ~ - - r- -r- 


1 ., .. .,., - 


li- 
ft 


iem. Double-Cadence 
and Mordent. 

f f f f f.fiiari^ 


Idem. 


*lj&= 







* On the question of Bach's ornamentation see the following: 
Rust, Preface to B. G. VII ; Franz Kroll, Preface to the Well- 
tempered Clavichord, B. G. XIV; Edward Dannreuther, Musical 
Ornamentation* I, pp, 161 210; H. Ehrlich, Die Ornamentik in 
J. S. Backs Klavierw&vhen (Steingraber) dealing mostly with the 
suites ; H. Schenker* Ein Beitrag zur Ornamentik (Universal Edition, 
Vienna) ; Klee, Die Ornamentik der klassischen Klaviermusih (Breit- 
kopf and Hrtel)j Germer, Die musikalische Ornamentik (Hug, 
"Leipzig, 3rd ed. 1899), See also Bischoffs remarks in his edition 
of Bach's clavier works (Steingraber), and the excellent elucida- 
tions of the ornaments in Ricordi's edition. 



346 



XVI. The Performance of the Clavier Works. 



Accent Accent Accent and Accent and Trill. Idem, 
ascending, falling. Mordent. 




In addition there is the essay on "Manieren" (ornaments) 
in Carl Philipp Emmanuel Bach ? s' Versuch fiber die wahre 
Art das Klavier zu spielen ("Essay on the true way of play- 
ing the clavier")*. 

The main rules to be observed are the following : 

1. Bach indicates the trill simply by the signs t, tr-, 
^-, AV, without specifying every time the particular 
manner or duration of it. As a rule it occupies the 
whole or the greater part of the note-value. 

2. The trill begins, as a rule, with the upper accessory note, 
and only in exceptional cases with the principal note. 
In long trills it is desirable first of all to linger a 
moment on the principal note, and then begin the 
trill with the adjoining note, especially where a move- 
ment or a theme see, e. g., the F sharp major fugue 
in the Second Part of the Well-tempered Clavichord 
commences with a trill, or when the upper note has 
just been struck. 

3. The Bach trill is further distinguished from the 
modern trill by the fact that it must be played much 
more slowly. It is spoiled by being taken quickly. 
We must bear in mind that the sign /w over a quaver 
signifies nothing more than that it must be decom- 
posed into two pairs of easy demi-semiquavers; in the 
same way a crotchet, if the tempo be somewhat fast, 
will be simply split up into two pairs of semiquavers. 



* Berlin 17531762. Part I, 2nd edition, 1759, pp. 45 100. 
Daniel Gottlob Turk's KlavierscMe (1789) is also very valuable 
in this connection (2nd edition, Leipzig-Halle, 1802, pp. 232 369). 



The Ornaments. 347 

The ornament is best realised when we play it with 

almost an exaggerated deliberation. 
4. If the succeeding note is a descending second, then 

the sign *v, as a rule, indicates not an ordinary trill 

but a Pralttriller* This must be carefully observed. 
Bach writes the trill with a Nachschlag (after-beat) thus 
AV ~K., conceiving it as a trill with a mordent. The down- 
ward and upward Vorschlag (preliminary grace-note) are 
denoted by crooks of a similar kind but reverse direction 
<** and <**-. Trills with Vorschlag and Nachschlag (double 
cadence and mordent) have both signs, thus, c** or ***. 
The manner of performance is explained by Bach himself 
in the above-mentioned examples in Friedemann's Klavier- 
bilchlein. According to Emmanuel, long trills must al- 
ways have a Nachschlag ; but this is dispensed with when 
several trills follow each other. The sign <* before a des- 
cending second thus signifies a Pralltriller, i. e. a broken 
trill; it must be played much faster than the ordinary trill. 
The final note of it must, to use Emmanuel's expression, 
be "filliped"; by which he means that the key must be 
struck quickly, and then jerked up again by an equally 
rapid drawing inwards of the point of the finger, which 
gives the note a very marked accent, Thus: 

Partita IV: Aria, 



m 



In the case of a practised player, Emmanuel recommends 
a prolongation of this Pralltrittey by one or two extra notes. 

His view of the Pmlltriller is that it is simply a very 
rapid trill, of longer or shorter duration, which is suddenly 
cut short on the staccato main note in such a way that the 
whole purpose of the trill seemed to be merely the throwing 
of a weightier accent on this note. 

The mordent also, denoted thus <*, is an interrupted 
trill, in which it is less a matter of the number of notes 



348 XVI. The Performance of the Clavier Works. 

trilled than of throwing the accent on the main note thaf 
cuts the trill short. Unlike the PralUritter, it is not limited 
to any particular situation. It takes the next note below, 
and is thus, as it were, the inverse of the Pralltrilkr. Both, 
in Emmanuel's phrase, "glide into the second, the mordent 
above, the Pralltrilkr below". 

We may distinguish two chief forms of the mordent, 
a short and a long. The latter has generally two con- 
stituent notes. It can also be represented by the prolonged 
sign +*~. Mordents have a preference for the major second. 
French Suite: Sarabande in E flat major. 






The turn in Bach's music, denoted thus *, must as a 
rule be played in four equal notes, in accordance with the 
note in Friedemann's book. A longer duration may, 
however, be given to the principal note if the tempo is not 
too quick to allow of this. Thus 



i 




The Vorschlage (appoggiature), indicated by slurs or 
small notes, are sometimes long, sometimes short. In 
each case, however, the accent falls on them, not on the 
principal note. The latter is tied with them and struck 
lightly: Emmanuel calls, this the Abzug (pulling off). 

If the VonMag is long, it takes half the value of the 
following note, if equal division is possible; if not, two- 
thirds fall to the Vorschlag, thus 



This rule, however, is not to be rigorously adhered to, 
but must be interpreted with reasonable regard to the re- 
quirements of the rhythm at the moment, The Vorschlag 



The Ornaments, 



349 



before a long note is generally long, and that before a 
short or passing note is generally short. Yet here again 
everything depends upon the position and the significance 
of the note. According to Emmanuel, a Vorschlag that 
takes the interval of a third is always short, even before 
a long note. The examples are very instructive in which 
the long and the short Vorschldge are met with in the same 
work, as in the sinfonia in E flat major and the sarabande 
from the partita in G major, Here the VorscMdge on the 
second and third beats of the bar are best rendered short, 
and those on the first beat long. Thus: 

Sinfonia in Eflat. 



To be played: 

Rt- * 




Partita V; Sarabande, 




The more we study the nature of the Bach Vorschlag, 
the more we see that the actual note-values are a matter 
of indifference, and that the real questions are the weight 
and the energy of the accents. As he always writes the 
accent as an abbreviation of the note, such passages as 
these 



j. ff rf rr 



35O XVI. The Performance of the Clavier Works. 

only attest that the first note must have a strong accent, 
while the second must be tied to it and have a diminuendo. 
The time-values of the notes must be as if written thus 






Conversely, quavers or semiquavers grouped in pairs must 
always be played in such a way that the second is only a 
kind of after-breath of the first, and sustained for only 
a fraction of its time-value. The tie is thus simply an 
accent-mark. 

The Nachschlag, indicated by a crook appended to the 
note or by small notes, is always short, and must merge 
into the note that follows. Thus when Bach writes 



r- 



it must be played as if written th,us 




When, as in the courante of the first partita, he writes 



""3"* 



m 



JT3 JJ] 



this is merely the old, inexact method of notation for 
^""~" ' " ~*<r 



L-jn-ML^p= 

JET f" Q -F P'v? f^yi^3: 

I --p / ^Lp I ^i,L 



All through his work we find him using the dot in the old 
way, that was more summary than exact. 

These explanations of Bach's "manieren" can only be 
regarded as general rules for the average case. If the 
ornaments accumulate, we soon exhaust the art of 



The Ornaments. 351 

explanation, and are thrown back upon our sense of natural 
euphony as the last authority. This is also Philipp Em- 
manuel's view. After pursuing the casuistry of the render- 
ing of ornaments into its final subtleties, in the end he 
submits the decision to artistic feeling, and so disowns 
the scholasticism at which he has been so seriously labour- 
ing all the time. Any one who has made himself familiar 
with the fundamental principles of Bach's ornamentation 
will find, by a little reflection, a satisfactory solution of 
the difficult rhythmical and tonal problems involved. He 
must always bear in mind that the reason for the orna- 
ments being indicated by signs, instead of written out in 
notes, is that he may have a certain freedom in the em- 
bellishment of the design. If we can get out of some of 
the beaten tracks of modern piano playing, and acquire 
a sense of the formal principles underlying the old orna- 
mentation, we shall have achieved a good deal. The 
question of whether we shall ever agree upon an interpreta- 
tion of the numerous ornaments in the aria ,of the Goldberg 
variations will then be nearer a satisfactory answer*. 

Closely considered, the whole system of ornamentation 
of that epoch indicates the. partial surrender of the com- 
poser to the virtuoso, who wanted to make his own effect 
by adding freely to the music. During the last century and 
a half we have slowly moved away from this conception 
of the rdle of the executant artist. The first to set his face 
against it was Bach himself. He could not combat his 
critic Scheibe's reproach that he left nothing to the player, 
but wrote out in full what had been formerly indicated by 
signs**. The reproach is just. As a matter of fact Bach 
banishes ornaments from his music. In the Well-tempered 
Clavichord we practically never meet with them; even a 
bravura piece like the Italian Concerto is almost destitute 
of them. In his vocal music, again, they are almost wholly 



* Dannreuther (I, 202 204) suggests one interpretation of 
the aria. 

** See ante, p. 180. 



352 XVI, The Performance of the Clavier Works. 

dispensed with. His music certainly seemed bare to his 
contemporaries. Only in the galant genre, the suite, does 
he allot the ornaments a role which, however, is very 
trifling in comparison with what was customary at the 
time. There is thus a certain irony in the fact that the 
small amount of ornamentation in his work should give 
such, trouble to the average player of today. This is, how- 
ever, his own fault, not Bach's. Any one who will devote 
four or five hours to getting a clear idea of the main points 
here discussed will find that the works have lost their 
terrors for him, and in the end he will find these "manieren" 
a source of pleasure. 

The question whether our modern piano is the right 
piano for Bach does not as yet occupy the general public 
very seriously, as it can form no conception of the instru- 
ments Bach used. In more expert circles, however, the 
question is already being debated with some heat. 

What would be Bach's attitude towards the modern 
piano? Exactly the same as towards the modern organ. 
He would hail with enthusiasm the perfection of its me- 
chanism, but not be particularly enchanted with its tone 
qualities. Sebastian Erard's invention in 1823 of the repeti- 
tion action made possible on the hammer-clavier the finely 
gradated touch that caused Bach to prefer the weak clavi- 
chord to the full-toned cembalo. Henceforth, however, 
it was towards greatly increased strength of tone that at- 
tempts to perfect the piano were directed. The more 
powerful the tone became, the duller became the timbre, 
so that the piano of today no longer suggests in any way 
the tone of the instrument of Bach's time, the result 
of substituting an iron resonance for the dear and bright 
wood resonance. 

The duller the timbre of an instrument, the less suitable 
is it for polyphonic playing, in which each voice must 
stand out clearly from the others, and be always perceptible 
by the hearer without effort. How unsatisfactory our piano 
is in music with bold obbligato parts, like those of Bach, 



Cembalo or Modern Pianoforte? 353 

only becomes evident after we have heard a few preludes 
and fugues on a good clavichord or clavicembalo. The 
clavichord is a string quartet in miniature; every detail 
comes out lucidly on it. On the cembalo every melodic 
line is quite clear, the plucked tone having a much sharper 
quality than that of the modern piano. 

Whether, however, it is advisable to posit a return to 
the old instruments as imperative for the true enjoyment 
of Bach's clavier music is doubtful. The clavichord is put 
out of court at once, for we could never again accustom 
ourselves to so weak a tone. It is otherwise with the cem- 
balo. The charm of its sparkling and rustling tone is not 
so easily resisted, and the variety of tone that it permits, 
by means of change of keyboards, coupling, uncoupling, 
and octave coupling, almost makes us forget that no nuance 
of tone is possible on it. The bass part too, comes out 
more clearly and beautifully on the cembalo than on any 
other instrument. Any one who has heard Frau Wanda 
Landowska play the Italian Concerto on her wonderful 
Pleyel clavecin finds it hard to understand how it could 
ever again be played on a modern piano*. 

But even in a quite small concert room the cembalo 
fanatic's enthusiasm has to contend against the unfortunate 
fact that at a distance of about twenty-five feet the tone 
that is so rustling when heard at close quarters becomes 
somewhat feeble and tremulous. Nor do all Bach's works 
sound equally well on the cembalo. It is best for pieces of 
unbroken and uniform motion, those that, like the C major 
prelude of the Second Part of the Well-tempered Clavichord, 
consist of arpeggio-like broken chords, and particularly 
those in pure two-part form, like the A minor prelude 
from the Second Part of the Well-tempered Clavichord. 
On the other hand the abruptness of its tone and its inability 



* See also Richard Buchmayer's article on Cembato oder Piano- 
forte, in the Bachjahrbuch for 1908, pp. 64 93. In Germany Hirl 
(Berlin) and Rehbock (Duisburg) "particularly devote themselves 
to making cembalos. 

So hw citzer, Bach. 23 



354 XVI - The Performance of the Clavier Works. 

to sustain a note make it unsuitable for music that demands 
a singing quality of tone. The reconstructed cembalo 
thus seems better fitted for Bach's music in private circles 
and for certain of his works than to impress the general 
public with the clavier compositions as a whole. None the 
less all lovers of Bach owe their gratitude to the scholars, 
artists and instrument makers who have again brought the 
cembalo in honour among us; and it is to be hoped that 
the instrument will be more used for Bach playing. On 
the other hand we must not imagine that the cry of "back 
to the cembalo" solves the problem of which instrument 
we should play Bach upon. For the moment we can only 
be sure that the modern piano is not so close in character 
to the one Bach dreamed of as Spitta and his contem- 
poraries thought it was. This is evident from the peculiar 
demands made by Bach's works, and a certain toning down 
is already beginning to be noticeable in the modern piano. 
We are gradually realising that the excessively strong 
and blunted tone of our grand pianos may be necessary 
for a large concert room, while in a small music room in 
a house it is too deafening to be pleasing, and that we must 
combine our consummate mechanism with a fabric of such 
a kind that the tone shall again be bright and clear and 
metallic. When this has been properly realised, and we have 
again reproduced in perfection the type of the table piano 
of 1830, as we have done with the cembalo, then we shall 
be much nearer the solution of the question which piano 
we ought to play Bach upon at any rate at home. Until 
then, any one who would rather play Bach beautifully than 
powerfully must be content with a well-restored table 
piano of 1830 or 1840, 

But the problem will not be wholly solved even by a 
reform of the domestic piano, for the reason that Bach had 
two instruments in his mind's eye. The modern piano is 
suitable for the music of the clavichord type; while the 
pieces calculated for the cembalo come out in their true 
beauty only in the silvery tone of that instrument. 



Dynamic Nuances, 355 

As regards the interpretation of the clavier works, opinions 
are gradually become clearer. When Liszt and Billow, 
in the middle of the nineteenth century, undertook to 
show the public again the living Bach, they had to fight 
a tradition that made stiffness, pedantry, and absence of 
temperament the true requisites for Bach playing. We can 
therefore easily understand their falling into the other 
extreme, and thinking that if Bach was to speak intelligibly 
to us he must be born again in the spirit of modern virtu- 
osity. This was the origin of those Bach arrangements and 
Bach * interpretations" that aimed at modern effects, in 
which the aim was not so much to bring out the laws of the 
works themselves as to make Bach talk like a modern. 
Later on Biilow himself went astray in his Bach editions 
that of the Chromatic Fantasia was a typical case 
trying to give the public a simpler idea of Bach's works 
than the composer had. What he was dimly striving after 
came into clearer consciousness in a later and more re- 
flective generation of pianists, that had learned to look 
at Bach's work as a whole. Two typical representatives 
of this new school are Busoni and Vianna da Motta. These, 
and the others who see eye to eye with them, do not aim 
so much in their Bach playing at ingenious dynamic nu- 
ances or striking effects as on making the broad plastic 
lines of the work speak for themselves. 

Bach is an organist rather than a "klavierist" ; his music 
is more architectonic than * Sentimental". That is to say, 
his feelings express themselves in a kind of acoustic design. 
As in his organ works, so in his clavier works, there is little 
of the imperceptible merging of 'piano into forte and vice 
versd. A certain strength of tone dominates a whole period, 
and is followed by another period of a contrasted intensity 
of tone. "Bach's music is always more or less ma- 
jestic*', says Vianna da Motta*. "It always rises in broad 



* See his article on Zur Pflege Bachscher JKtawierwerke, in 
the Neue Zeitschrift ftlr Musik, 1904, p. 678 ft The article dis- 
cusses the problem of a new style of Bach playing, 

23* 



356 XVI. The Performance of the Clavier Works, 

terraces, like the primitive Assyrian temples." It is this 
structure that we must look for if we wish to understand 
a given work thoroughly; otherwise we shall always come 
to it with an arbitrary conception of our own, and try to 
force it into conformity with this. 

We must begin with the study of the clavier works in 
which Bach himself has indicated forte and piano. These 
are the Italian Concerto, the Chromatic Fantasia, if 
we can rely on the copies here and the last partita, in 
B minor. These shew us on what very simple lines Bach con- 
ceived the structure of a work. The contrasting periods are 
very long. A change only occurs when there comes a salient 
episode of a different kind. Even in the echo piece at the 
end of the partita the contrasts are not lavishly employed. 

This very partita, however, shews us that in many works 
Bach did not calculate on any variation of tone-strength. 
After having made use of the two keyboards in the over- 
ture, he adopts only one colour for all the other numbers. 
The courante, the sarabande and the gigue are taken forte 
throughout; likewise the first gavottes, passepieds and 
bourrees ; to each of these three dances a second is appended 
that is to be played on the piano manual. This implies 
that the dance movements in the other suites are also to 
be rendered in a uniform colour. To play them with 
emotional transitions from forte to piano and vice versd 
is to destroy their character. 

On the other hand it is obvious that the majority of the 
long introductions to the English suites and the partitas 
demand two tone-colours, and are built on the same plan 
as the overture to the last partita. In the prelude to the 
English suite in G minor 




the piano begins at bar 33 and continues to bar 67. Then 
comes a forte passage, lasting to bar 99; from this point 



Dynamic Nuances. 357 

the hands play for some bars on different keyboards, the 
one that has the flying quaver motive derived from the 
main theme being each -time on the forte manual; from bar 
109 to bar 125 both are on the main manual; the succeeding 
piano passage lasts until bar 161, where the right hand 
goes to the forte manual, while the left stays on the soft 
manual; in bar 185 this hand also goes to the forte. 

It is important that in the pieces noted by Bach with 
forte and piano we should be quite clear how he goes from 
one tone-tint to the other sometimes with both hands 
at once, sometimes with one following the other. In this 
connection the organ works should again be studied, which, 
being laid out on larger and simpler lines, exhibit the 
"terraces" much more clearly. Nor should the Branden- 
burg concertos be forgotten; they are most illuminative as 
to the structure of Bach's works and the employment of 
different degrees of strength of tone. We must consider 
too that in the orchestral suites called "overtures" 
(B. G. XXXI 1 ) Bach indicates changes only for the intro- 
ductions and the free movements, but not for the dances. 
This accords with the principles exhibited in the last 
clavier partita. 

With the information thus acquired we have some sort 
of a guide as to the dynamic plan underlying the preludes 
and fugues of the Well-tempered Clavichord, Here again 
there are a number of pieces, among both the preludes and 
the fugues, that are intended to be played in one colour 
throughout. Where the character of the architecture 
shows no logical necessity for change, where, consequently, 
alterations of forte and piano seem arbitrary, it is better' 
to maintain an agreeable and flexible forte, for example, 
the preludes in C sharp minor, D minor and E major from 
the First Part. In the fugues we shall still more frequently 
have the conviction that they are not planned for dynamic 
variety. 

In the pieces that employ two degrees of strength of 
tone, these degrees are either distributed simultaneously 



XVI. The Performance of the Clavier Works, 

between two voices or they alternate. To the first order 
belong the preludes in D minor, A minor and B minor 
from the Second Part ; here the proper dynamic is to bring 
to the front each time the voice that has the main theme, 
while the other voice remains more in shadow, as it were. 
Many preludes are laid out in such a way that the one 
hand always plays piano, the other always forte, as it is 
prescribed in the middle movement of the Italian Concerto. 
A typical specimen of this is the three-part prelude in 
F sharp minor from the Second Part, which is written in 
such a way that the left hand plays the lower voices, and 
the right hand the upper voice alone. To play now forte, 
now piano here would be as wrong as to do so against 
Bach's express specifications in the middle movement of 
the Italian Concerto, 

The clue to the distribution of light and shade in the 
A flat major and A major preludes from the First Part is 
given by the entry of the theme. Wherever it appears, 
whether in one part alone or in the ensemble, it must be 
given out forte; the rest must be kept piano. 

In the F minor prelude of the Second Part, again, that 
is so often maltreated 






the variations of forte and piano are unequivocally de- 
termined, for the first section at any rate, by the entry 
of the main theme; in the second section the matter is less 
clear, theme and episode not being so clearly marked off 
from each other. The most natural way of playing it 
would be thus bars i 4*/2 f^te; 4*/2 8*7* piano ; 
8l /2 i6V 2 forte; i6y 2 -- so 1 /* piano; 2oV 2 ~32Va forte\ 
32 x /2 40 piano; 40 */ 46 Vs, right hand forte, left piano; 
46 V 2 52 V 2 vice versd; 52 Va S^A piano; then forte to 
the end. 

The prelude in C major from the First Part seems to be 
calculated on an echo effect of the kind obtainable on the 



Dynamic Nuances. 



359 



two-manual cembalo. The first half of the bar should thus 
be played forte, the second half piano. In this way it 
sounds exquisite on the cembalo. The prelude has then the 
effect of a cheerful reverie, instead of the pathetic effect we 
are apt to read into it when we try to bring out the mys- 
terious melody that seems to hover above it. If we play 
it pathetically on our pianos, with a big and continuous 
crescendo, the effect is unsatisfactory no matter whether 
this finishes in a fortissimo, or, contrary to Bach's clear 
intentions, dies away to a pianissimo (if indeed it has not 
done this several times already). The one thing we can 
be sure of with regard to the modern way of interpreting 
this prelude is that there has never yet been a pianist who 
has succeeded in playing it to the satisfaction of another. 
The D major fugue in the First Part, which is more 
on the lines of a fantasia 



makes its natural effect when we play the main theme 
the upward-striving demisemiquaver figure forte each 
time it occurs, contrasting it with the calm, downward- 
moving semiquaver figure, which must be played piano. 
In this way the powerful final climax becomes particularly 
effective*. 

In the D minor prelude from the Second Part, again 




* The plan may be detailed thus: bars i 2, forU; 3 piano; 
4 and 5 right hand forte, left piano; 6 9*/4 both hands forte; 
9 s /4 io> the downward semiquaver answer, piano; io l /4 forte; 
the semiquaver answer again piano; n 16 both hands forte; 
1 3 ip the first crotchet each time strong, the three others weak; 
20 forte in both hands; 21 first crotchet forte, the others weak; 
22 to the end, forte in both hands. 



360 XVI. The Performance of the Clavier Works, 

the descending answer to the ascending fanfare-like theme 
should contrast with this "by being taken piano. If we are 
to play this prelude in accordance with the sense of it we 
must conceive it as scored for Bach's orchestra. On closer 
examination we shall discover only a few quite short sec- 
tions that are to be given in the piano tint. These short 
piano interludes are very characteristic of Bach, as we see 
from many examples among the organ and orchestral 
works. Their very brevity makes them doubly effective*. 

In the majority of pieces, however, these episodes are 
to be recognised only from their structural plan, and not 
as contrasts inherent in the theme itself. It is safest to 
be guided by the cadences and the incidents of the poly- 
phony. If the segment is preceded by an important ca- 
dence, or if one or more voices are resting, then it is a fair 
assumption that the passage in question is to be played 
piano. The E major fugue from the Second Part may 
serve as an example; bars I 22 forte\ 23 34 piano; 35 to 
the end forte. The Eflat major fugue from the same Part 
is similarly laid out: bars I 30 forte; 30 58 piano; but 
with the theme given out forte-, 59 to the end forte. The 
C sharp minor fugue from the First Part calls for a similar 
treatment. For increases of intensity within the forte, 
Bach relies only secondarily on dynamic means; he builds 
up the effect by the fulness of his tone and the way in 
which the voices co-operate. 

The evidence for the inner necessity of a change to 
piano is not always so clear as this; as a rule we have to 
be satisfied with a certain amount of probability, for 
example, when, in the F major fugue from the Second Part, 
we begin the piano at bar 29 and let it cease in the left 



* The following plan seems the most natural: bars t 
forte; 29/431/12 piano; 3 2 /i2~~ 41/4 forte-, 4*/4 SVn piano \ 
5V 4 16 forte; 17181/12 forte; 181/12 2* Vaa piano, the left hand 
playing bar 19 forte; siVia S^Vi* forte; 321/12 4iVia piano; 
4i*/i2 4* 1 /* A"'*; 42 1 /r 43 1 /ia piano; 43 1 /isr~~44 l 7* forte} 
44 1 1 1 45 Via piano; then forte to the end. 



Dynamic Nuances. 361 

hand in the second half of bar 66, in the middle voice in the 
right hand in bar 70, and in the upper voice in the second 
half of bar 73; or when, in the G minor fugue of the First 
Part, we end the forte at bar 12 y 4 , take the following bars 
piano, only bringing out the theme forte wherever it occurs, 
and let the final tutti begin at bar 28; or when, in the 
C major fugue from the First Part, we begin the piano at 
bar 14 and let it continue until bar 24; or when, in the 
D major fugue from the Second Part, we play forte until 
bar 16, piano until %7 l /z, and then bring in the final tutti. 

With many preludes and fugues, however, our efforts 
are in vain; we can discover in them no dynamic plan 
that could not equally well be replaced by Another, and 
our only recourse is to bring out the theme and avoid marked 
contrasts. As an example we may cite the F minor fugue 
from the Second Part. In the G minor fugue from the 
Second Part it is clear enough that the final tutti begins 
with bar 67 ; but it is not so clear how far back the previous 
piano extends. One is tempted to begin it in bar 40. 

It is a false modernisation to let a cadence that ends 
a forte section die away in a diminuendo, so as to lead over 
into the following piano ; or to make a crescendo at the end 
of a piano section in order to glide imperceptibly into the 
forte. In this way we level the terraces on which every- 
thing depends in these works, and destroy the 'plastic out- 
lines of the piece. The Bach cadence is something solid, 
and must be given out in the normal tone-strength of the 
section which it concludes. It is this stark antithesis that 
gives the music its charm, as can be seen in the Branden- 
burg concertos, The tutti voices suddenly cease, and the 
concertino of the solo voices, after having entered on the last 
chord of the tutti t remains suspended, as it were, in the air. 

It is another false modernisation to begin or end pian- 
issimo. Just as in a pianoforte piece that Bach himself 
has marked, or in a Brandenburg concerto, or in an organ 
work, we have to begin and end in a forte, we must so begin 
and end in the pieces of the Well-tempered Clavichord. This 



362 XVI. The Performance of the Clavier Works. 

rule will not at first be acquiesced in by players who think 
in the modern pianistic way. It will strike them as pedan- 
tic. Nevertheless, the longer we play Bach, the more we 
revolt against artificialities and clevernesses of every kind, 
and feel that the simplest way is the only right one. 

Some day we shall even dare to begin and end the Eflat 
minor and B flat minor preludes from the First Part in a 
beautiful but full tone-colour, and to conceive them largely 
and pathetically, instead of delicately and sentimentally 
as we do now. In the Eflat minor prelude all the chords, 
whether they lie above or below, should be taken piano, 
while everything that pertains to the eloquent theme, in- 
cluding the cadences, should be taken for te. In the B flat 
minor prelude we should begin with a vigorously-shaded 
forte, take bars 13 and 14 piano, then begin again with a 
quiet forte, that works up until three bars before the end, 
the final cadence being taken in a quiet forte, 

A further fault, that is likewise grounded in the pianistic 
feeling of our time, consists in sacrificing everything in a 
fugue to the working-out of the theme, and, as soon as 
this enters, clapping the hand as it were over the mouth 
of the other obbligato voices, the result being that the 
auditor hears the theme but not the fugue. In episodical 
passages it is natural to play the theme, whenever possible, 
on a separate manual, as it were ; but in the logical opening 
and final structure of the fugue all the obbligato voices 
have equal rights, and the theme can only be considered 
as primus inter pares. 

From the foregoing it is evident that on the piano, as 
on the organ, it is wrong to begin a theme pianissimo and 
then work it up through a continuous crescendo to a final 
fortissimo, as if we were first presenting it as a cub, and 
then showing it in all the stages of its development until 
it becomes a lion. Every Bach theme, whether it expresses 
joy or grief, has a touch of the sublime about it, and it 
must be delivered in this style from the beginning. It 
must be confessed, though, that it is difficult for us modern 



Dynamic Nuances, 363 

musicians to emancipate ourselves from the idea of work- 
ing Bach gradually up to a final climax. There are still 
many artists who thihk that the essence of the fugue in 
general, and of the Bach fugue in particular, consists in 
a gradual piling-up from piano to fortissimo. 

The laws here laid down for the plastic performance 
of Bach's music are not derived from traditions, but are 
grounded in the works themselves. They attest that nothing 
is gained by besprinkling his compositions with pianissimo, 
j>iano, mezzo forte, forte, fortissimo t crescendo and Aecrescendo, 
as if they were written for the accordion, but that we must 
aim either at the working-out of a broad dynamic plan, 
or decide to play the work with a uniform strength of tone. 
This view can seem pedantic only to those who think that 
it elevates monotony to an artistic principle. The opposite 
is really the case. When the one quality of tone is dis- 
tributed over a segment, large or small, it must be richly 
shaded in detail, but in such a way that these shadings do 
not overstep the limits of that particular degree of tone. 
Bach loved the clavichord so much precisely because he 
could produce this shading of detail on it; and the magic 
of his playing for his contemporaries came from this vivacity 
of detail. We must therefore distinguish in his music 
between an architectonic dynamic, that aims at bringing 
out the great lines of the piece, and the dynamic of detail 
that accompanies it, the object of which is to give life to 
these lines. The latter could almost be styled declamatory 
dynamic, being somewhat analogous to the cadential 
element in musical speech. 

Bach's music is Gothic. Just as in Gothic architecture 
the great plan develops out of the simple motive, but 
enfolds itself in the richest detail instead of in rigid line, 
and only makes its effect when every detail is truly vital, 
so does the impression a Bach work makes on the hearer 
depend on the player communicating to him the massive 
outline and the details together, both equally clear and 
equally full of life. 



364 XVI. The Performance of the Clavier Works. 

11 in place of this dual architectonic dynamic we put 
the uniform modern emotional dynamic, such as we meet 
with in Beethoven's works, we have simply done our best 
to make Bach unintelligible, by confounding the great and 
the small nuances in one*. 

Therefore to play the preludes in E flat minor and B flat 
minor from the First Part with a uniform strength of tone 
does not mean playing them monotonously, but declaim- 
ing them with simple and impressive pathos, avoiding 
sentimental effects, and trying to get all possible richness 
by the perfect shading of the normal forte. Within our 
Bach forte we must get the relief of a piano and a forte, 
and a similar shading in the piano. Our mezzoforte is un- 
known to Bach; the sign mf is nowhere to be found in his 
music. He severs the link that binds piano to forte. On 
the other hand, as is proved by his markings in the scores 
of the cantatas, he distinguishes a pianissimo within the 
piano. A mezzoforte makes a Bach piece uninteresting. 
This holds good not only for the piano and the organ 
works but for the orchestral concertos and the cantatas. 
If an orchestral piece does not make its effect, we may 
be sure that in half the cases this is the fault of a lack of 
character in the quality of tone employed. Unfortunately 
our pianists and instrumentalists can get a mezzoforte^ but 
neither a piano nor a forte that is rich and capable of grada- 
tion. For this reason it is no unreasonable demand, curious 
as it may seem, that whoever wishes to interpret Bach must 
first of all learn how to play forte and piano. 

* This is the case in the Czerny edition of the Well-tempered 
Clavichord (1837). We wonder whether the editor himself used to 
execute the numerous cnscendi and decrescendi that he has sprinkled 
over the text. It is interesting to know that he justifies his mark- 
ings by the practice of Beethoven, by whom he often heard the 
preludes and fugues played. The Steingraber edition and the 
Ricordi edition are much more sparing with nuances. In principle, 
however, their standpoint is that of Czerny, their system of dy- 
namics not springing from the natural plan of the work, but being 
grafted on to it. The player will never feel these nuances to be 
"necessary", in the true artistic sense, but merely adventitious. 



Phrasing. 365 

In Bach, however, the really essential thing is not so 
much the dynamic shading as the phrasing and accentua- 
tion. We soon realise this when we look at one of those 
precious orchestral scores which he has furnished with 
indications for performance. Seiffert, in his fine essay 
on Praktische Bearbeitungen Bachscher Komfyositionen* 
justly contrasts him in this respect with Handel. "In 
Handel", he says, "the care for dynamic effects is always 
uppermost; for the phrasing he is satisfied with incidental 
hints, Bach gives extremely little in the way of dynamic 
suggestion; but he all the more carefully phrases the or- 
chestral parts," Seiffert's explanation of this difference 
that Handel had at his disposal trained orchestral players, 
while Bach had only town musicians, is not con- 
vincing. The difference in the two methods is rooted in 
the dissimilarity of the men's music, Bach's works demand 
a characteristic and subtle phrasing of the themes and of 
the parts, on which the effect of the whole almost entirely 
depends. Handel's themes and passages, on the contrary, 
run more on the cust&mary lines, and demand no such 
individuality of phrasing. 

In general we may lay down the principle that in Bach 
every theme and every phrase must be delivered as if we 
were playing it on a bowed instrument .| This holds good 
for the pianoforte not less than for the wood-wind instru- 
ments. We play the preludes and fugues of the Well- 
tempered Clavichord in accordance with Bach's intentions 
when we try to render them as if we were employing not a 
keyed instrument but a quartet or a quintet. The ideal 
must be to link the notes in such a way that they do not 
seem to be struck one after the other, but as if several 
bows were being simultaneously drawn over the strings. 

Bach's omission to record the phrasing and grouping 
in his clavier works is due partly to the fact that this was 
not customary at that time, the "executive artist*' in our 



* In the Bachjahrbuch for 1904, p. 59, 



366 XVI. The Performance of the Clavier Works. 

sense of the term being as yet unknown; and partly to the 
fact that the players he had in view were almost always 
his sons and his pupils, who were familiar with his 1 prin- 
ciples. In the works for clavier with other instruments 
there are often indications of the phrasing in the clavier 
part. Bach thinking it necessary that the clavier player 
should phrase precisely like the instrumentalists. Look, for 
example, at many movements in the violin sonatas, and, 
among the clavier concertos, at the one in A major (No. 4). 
The greatest importance, however, attaches to the dots 
and ties in the orchestral parts of the Brandenburg con- 
certos and of certain cantatas. A study of these will con- 
vince any one of the right way to phrase the clavier works. 
Legato playing, that is regarded as the characteristic of 
the Bach school, does not, as we observed in connection 
with the organ works, mean a uniform style of delivery, 
but implies an endless variety in the tieing and grouping 
of single notes of equal value. Four semiquavers are for 
Bach not four semiquavers, but the raw material for quite 
varied shapes, according to how he groups them : thus 
MIH} CIMMIHIM CiCiiC2 CZCZCiH E?3^22 

*?T? or <?T3 r T?' or ^T? r JT-T3- 

The last system of connection, that is almost universal 
elsewhere, is always subordinated by him to the other 
and more characteristic systems. He mostly groups the 
notes in twos when the second and third semiquavers have 
the same note in common. In these cases he always uses 
short ties, as if he were afraid of being misunderstood. Thus 
on the very first page of the Italian concerto 




The same grouping is also found in certain cases where 
none of the notes is repeated; but it is observable that 
here, as a rule, the melody moves in seconds. In the 
B minor fugue from the First Part, Bach indicates this 



Phrasing, 



367 



grouping by twos, as it did not seem to him quite self- 
evident 




In this connection it cannot be sufficiently insisted on 
that the second of the tied notes should be like a mere breath, 
A Bach connoisseur like Gevaert advocates writing such 
passages in this way * 



J! t ft * J! R I 



This accentual rule naturally holds good only for the 
normal case in which the first of the two tied notes coin- 
cides with the stronger beat. In the opposite case the second 
gets the strong accent, and the first counts as a preliminary. 

Prelude 20 (Second Part). 
~Q ' 



1 



The rhythm in which the fourth note is detached from 
the first three appears in the presto of the Italian Concerto, 
where Bach in addition places a dot over the last note, so 
as to ensure it being properly detached from the others 




The combination most frequently met with, however, is 
that in which the last three notes are tied and separated 
from the first. Here again Bach makes sure of the freedom 
of the first by means of a dot. Thus at the commence- 
ment of the Italian Concerto 




* In his vocal score of the Si. Matthew Passion he uses this 
method lor the final chorus of the First Part. 



368 



XVI. The Performance of the Clavier Works. 



In a longer succession of tied notes, again, the first must 
be detached from the others. Bach phrases the presto of 
the Italian Concerto sometimes thus 




The dot only signifies that the note does not belong to the 
same group as those that follow; it is, however, closely 
bound to the preceding ones. In our notation this is best 
expressed by including the detached note in the same tie 
as the previous notes, thus 

Fugue in B flat (Second Part). 




Fugue in G major (Second Part)/ 




In the groups in triple time, again, the first beat is very 
often the end of a tied series, not the commencement of 
a new one, Bach's phrasing of the theme of the A major 
fugue from the First Part would therefore be this 




Even in pure triplets it is in certain cases better to 
separate the second and third notes from the first as if 
taking a light breath. The eighth bar of the B flat major 
prelude from the Second Part should consequently be 
played thus 



The Phrasing. 369 

This phrasing is so frequent in Bach because his themes 
and figuration, even where the bar-divisions do not indicate 
it to the eye, have generally an up-beat character, beginning 
with an unaccented notfe that leads into an accented one. 
It is a great mistake to play successive notes in his music 
with equal values, in the style of Czerny's "School of Veloc- 
ity' % or Clementi, or Cramer. There the essence of the 
legato phrasing is that we begin with an accented note 
and play the remainder as nearly as possible with the same 
weight. Bach's legato, however, is much less pianistic 
and much more vital. The large tie in his music embraces 
numberless smaller ones, that gather the notes into sub- 
ordinate groups. Neither in his piano runs nor in his 
violin runs does he desire equivalence in the notes. Each 
has its relative value, namely that belonging to it from 
its place in the tie. 

This can be explained on historical grotmds. The mono- 
tonous legato that obsesses our pianoforte schools could 
only come after the passing-under of the thumb had be- 
come recognised as the cardinal principle in playing. A 
legato of this kind was impossible so long as people 
simply passed the other fingers over and under each other, 
or merely moved the hand as a whole, and fingered scale 
passages 3, 4, 3, 4, 3, 4, or 5, 4, 3, 2, 2, i, as Bach tells his 
son to do in the Klavierubung. Ties of the second kind 
grew out of the style of fingering. For this reason Bach's 
legato, compared with the usual pianistic legato, is as much 
richer and more varied as his fingering was richer and more 
diverse than ours. We must not imagine we are playing 
Bach with the right legato when we have worked out on 
paper all the passings-under of the thumb and the com- 
plicated interchanges of the fingers, and so made sure that 
every note will be properly sustained even to the last 
hundred- thousandth of its value. The correct tieing and 
fingering are those which bring out all the variety intended 
by Bach in the phrasing and the accents, i. e. those which 
"tell". 

Schweitzer, Bach, 24 



370 



XVI. The Performance of the Clavier Works. 



The Bach staccato coincides only in rare cases with our 
light modern staccato. It is not so much a key pizzicato 
as the short and heavy stroke of a bow. Its effect is there- 
fore to accentuate the note rather than to lighten it. It 
would be better indicated in our notation by a short stroke 
than by a dot. 

Long quaver or semiquaver passages in staccato are 
rarely met with in Bach; when he separates the notes it 
is only by way of a transient interruption of the legato. 
Perhaps in passages like this * 




the semiquavers between the two ties are to be played 
staccato, and this time lightly, not heavily. Certain indica- 
tions in the orchestral parts make it probable that in the 
weakest beats of the 6 /s an< ^ similar times, when the motion 
is quite uniform, a light staccato is often intended to break 
the legato. In general, however, the rule holds good that 
Bach's staccato does not run a uniform course, and that 
it is not light but heavy. An interesting case where he 
makes use of a staccato to accentuate a beat is found in 
the final allegro of the clavier concerto in A major: 




fin* 






The transition here is in every respect so abrupt, especially 
as the orchestra enters on the trill, that the brake must 
to some extent be put on the triplets. To play them with 
a light staccato would be to run the risk of negating Bach's 
intentions. 



* From the Siciliano of the fourth sonata for clavier and violin, 
bar 9, The ties are Bach's own. 



The Phrasing. 371 

The only rhythm that must always be given in staccato 
notes is that which Bach uses to express solemnity 
M J d Ji We must detach the second note from the 
first, let it press forward towards the one that follows, and 
play it rather too heavily than too lightly, in order to 
preserve the impression of a somewhat formal solemnity, 
of the kind that Bach wishes to suggest in many of his 
sarabands and gigues. Thus the bass of the F sharp major 
prelude from the Second Part should be played thus 



Nor should we be afraid to play whole pieces, such as 
the G minor prelude from the Second Part, if they are 
wholly in this rhythm, without a single really legato 

note. ^ ^^ 

The rhythms J f4 and J"JJ> and the variants of 

these, demand a certain emphasis on the short notes, as 
if there were a fear of their passing unobserved. The shorter 
they are in actual time-value, the more careful must we be 
to make them tell. In this way we can detach them some- 
what from the main note and play them with the heavy 
staccato. 

For example 

Well-tempered Clavichord. Prelude 17 (Second Part). 




Have we any means of knowing whether groups of notes 
are to be played staccato? In the first place, the staccato 
is required when the sequence is in characteristic or widely- 
separated intervals. Under this first rule come all the 
themes that have anything of a "springy" motion. The 



372 



XVI. The Performance of the Clavier Works. 



best-known example is perhaps the theme of the F major 
Invention 




The theme of the chorale fantasia Jesus Christus unser 
Heiland, again, from the Third Part of the KtavierUbung, 
must be played staccato 




The theme of the A minor fugue from the Second Part 
of the Well-tempered Clavichord must be given thus 




It seemed self-evident to Bach that the crotchets in this 
theme could only be rendered in this way. Some players, 
however, might have been in doubt as to whether to play 
the quavers legato or staccato. Bach therefore settled 
the matter by adding dots to the notes. He could not 
foresee that by so doing he would tempt the players of 
later generations to play the crotchets legato, as they have 
no dots, and to contrast the staccato quavers with them. 
The second rule for the employment of the staccato in 
Bach might be formulated somewhat thus that any 
note interrupting a uniform motion at once steps out of 
the general tie embracing the preceding notes. The ex- 
amples of this are innumerable; a typical one is the theme 
of the D minor fugue from the First Part, in which Bach 
himself marks the staccato point throughout the whole 
piece 



The Phrasing. 



373 



Here indeed it becomes perfectly clear that the principle of 
the Bach phrasing is derived from an ideal bowing. This 
general rule, however, has one exception. The interrupting 
passage is only to be played staccato if it moves in a zig-zag 
line. If it consistently ascends or descends in uniform 
and dose-lying intervals, it must be played legato. 

There are thus two fundamental conceptions from which 
the whole of Bach's themes have sprung. The first is 
that of the differentiated ties; the second is that of the 
staccato as a rhythmic interruption of the legato. There- 
fore the phrasing of the themes should not be left to the 
caprice of genius, but must be deduced from the ground- 
rules of legato and staccato in Bach. Every player can 
acquaint himself with the typical combinations that arise 
from the differentiation of the ties and from the concur- 
rence of legato and staccato, and can thus learn to bring 
out the various groups of which a Bach theme is composed. 

A few examples from the Well-tempered Clavichord may 
serve to elucidate these points, 

Themes consisting purely of differentiated ties : 
Prelude 20 (Part II). 




374 XVI - Tiie Performance of the Clavier Works, 
Prelude 2 (Part II). 




Themes combining legato and staccato : 
Fugue 2 (Part -I). 






Fugue 21 (Part I). 



1/1 1 JJJJI 

s ~\ " 



Fugue 16 (Part I). 




Fugue i (Part I). 



Fugue 15 (Part I). 




Fugue ii (Part I), 






Fugue 5 (Part I). 






The Accentuation. 375 

Fugue 12 (Fart I). 

TTT^f- 



Fugue 7 (Part II). 




Closely bound up with the question of phrasing is that 
of accentuation; it is, indeed, solved by the right solution 
of the other. It has already been remarked, in connection 
with ties, that Bach's themes are mostly conceived as be- 
ginning with a large up-beat. The unaccented notes do 
not follow, but lead up to, an accented note. Therefore 
to play Bach rhythmically means accenting not the down- 
beat but the emphatic beat. With him, more than with 
any other composer, the bar-divisions are only external 
divisions of the themes, the real metre of which cannot as 
a rule be represented in simple time-species. The first 
to express this clearly was Rudolf Westphal, in his metrical 
study of the fugues in common time in the Well-tempered 
Clavichord, in which he proves again and again that those 
who regard the bar-lines in Bach's music as the borders 
of the rhythmic factors are bound to play him unrhythm- 
ically*. 

In a Bach theme everything urges forward to a principal 
accent. Till this comes aE is restless, chaotic; when it 
arrives the tension relaxes, and at one stroke all that went 
before becomes dear, we understood why the notes 
had these intervals and these values. The chaos becomes 
order, the restlessness becomes peace. The theme lies 
before the hearer like a good coin, with the milling fresh 



* See Westphal's Die C Takt-Fugen des Wohltemperierten 
JKlavurs, in the Mmihalisches Wocheriblatt for 1883, pp. 237 ff. The 
justice of his views upon Bach is unassailable, even if we do not 
share his general opinions upon musical metre. 



376 



XVI, The Performance of the Clavier Works. 



and sharp. If, however, this emphasis is lacking, and in 
place of the note that is an integral part of the rhythm of 
the theme we accent strongly a note that belongs only to 
the bar-rhythm, the unity and totality of the impression are 
destroyed ; all he has in his hand is a coin of several pieces 
soldered together, that keep falling asunder. This of 
course does not mean that in Bach the thematic accent 
and the bar-accent never coincide. Cases of this kind, 
however, are more or less accidental. 

The superior vitality and lucidity given to Bach's music 
by throwing the main accent on the characteristic notes 
can be seen at once by any one who makes a practical trial 
of it, accenting according to the nature of the theme, in- 
stead of taking the phrases with the usual rise and fall. 
The final diminuendo, with all its shadings, that we intro- 
duce into everything out of pure habit, is one of the direst 
enemies of stylistic Bach playing. 

The point can be best elucidated by reference to the 
theme of the E flat minor prelude. It is usually played 
thus 




It should really go thus 

IT) 




If we play it in this way, the prelude, couched as it is in 
dialogue form, acquires a much finer inner coherence. 

To these indications for the acceitfuation of Bach's 
melodies one is tempted to add another rule, that may 
seem extrinsic and imperfect, but which does good service 
in the majority of cases. According to this rule, we should 
primarily emphasise the notes at the end of a melodic line 



The Accentuation. 



377 



running in a certain direction, whether this is carried through 
in one piece or breaks off now and then and continues in 
separate segments. In the latter case the end notes of each 
segment form preliminary accents, that are meant to lead 
up to and culminate in the last and principal accent. It is 
just these themes built up in periods that are most often 
misunderstood. The following is a specimen of correct 
accentuation 



Well-tempered Clavichord, 



Fugue ii (Part II). 
I 




Here, as a rule, we do indeed hear the first two accents; 
the third, however, on the F, that ought to crown the others, 
is lost, and the theme runs about in the fugue like the 
poor ghost that carries its head under its arm. 

Fugue 17 (Part II). 

J 




This case is similar to the foregoing. The average player 
resists the seductive appeal of the third beat for the first 
accent; he cannot, however, resist the first beat of the 
second bar, he respects the traditional rights of a note 
that occupies so important a position, and gives the A flat 
the accent that should be given to the D flat, without 
noticing that the succeeding semiquavers are no more 
than an ideal retardation of the passage of the D flat into 
the C. 

In the second place, a main accent must be given to a 
note that interrupts a melodic line more or less suddenly, 
whether the note be syncopated or not. As a matter of 
fact, in the majority of cases it will happen either on a 
striking interval or on a syncopation. This sudden holding- 
back is a characteristic feature of a number of Bach's 
themes* If we efface it, by emphasising the strong beats, 



378 XVI. The Performance of the Clavier Works. 

we only torture the outline of the theme; but if we bring it 
out by proper accentuation, we reveal the great natural 
line that embraces this angular curve. We should there- 
fore always emphasise with confidence any conspicuous 
interval or syncopation. 

This holds good even in matters of detail. In the B minor 
prelude of the First Part of the Well-tempered Clavichord 
the natural bar-accent seems demanded by the regular 
quaver-movement in the bass. If, however, we accent 
thus 




J J| 






it will immediately be found that this rhythmical antagon- 
ism between the upper parts and the even bass is the very 
life of the prelude. It is unnecessary to say that here the 
emphasised notes are to be detached from the preceding 
ones. 

The following are types of themes that call for this syn- 
copated accentuation 
Fugue 19 (Part II). 




That the main accent here must be given to the last 
note, which on the first thought one would hardly venture 
to do, is proved by the further course of the piece, and 
especially its end. The two syncopations are thus only 
preliminary accents, leading up to the third. 

Fugue 16 (Part II). 



Here the method of accentuation may be thought too 
venturesome; but with repeated playing it will finally 



The Accentuation. 



379 



seem more right than the more obvious accentuation on 
the first beat. A particular result of this syncopated 
accentuation is to give life to the otherwise monotonous 
seven quavers on the same note. 

Fugue 8 (Part I), 







S J>^ 



This theme clearly shows how little significance the bar- 
divisions in Bach have for the thematic emphasis. That 
the consistent syncopated accentuation here is the only 
right one is shewn by the composer himself at the finish. 
As if he had sufficiently played with the performer and the 
hearer he brings in the theme in augmentation in such 
a way that the syncopations fall on the strong beats 




In the animated second subject of the Italian Concerto, 
again, we can see how the syncopation deprives the neigh- 
bouring strong beat of its accent. It is usually accented 
thus 




This, however, does not bring out the essential prin- 
ciple of its structure. The correct way of phrasing is that 
in which the two syncopations act as preliminary accents, 
leading up to the final accent, thus 



380 XVI. The Performance of the Clavier Works. 

It may be objected that accentuation on these principles 
is too harsh and rough to be correct. It is, however, the 
only accentuation that does justice to the original character 
of Bach's themes. The hearer will not feel anything dis- 
agreeable in the sharp characterisation they thus receive; 
he will only feel the truth and vitality of it. In any case 
the fugues will now give him more enjoyment than before, 
as he will grasp the rhythmic essence of them, instead of 
regarding them, in the customary way, as mere sequences 
of intervals. And it is quite certain that the cultivated 
hearer also pursues a theme in all its developments not as 
a sequence of intervals, but primarily as a succession of 
certain characteristic accents, that bring with them the 
idea of the intervals associated with them* 

Reference has been made, in the chapter on the organ 
works, to the profound distinction between Bach's organ 
themes and his clavier themes. The former are almost 
always based on the principle of the accentuation of the 
strong beat, since it is impossible to bring out any other 
accent on the organ. It is even impossible to make one 
note stand out above the others. The possibility of this 
on the piano, if only in a limited way, is made use of 
by Bach to the utmost permissible limits. Bowed in- 
struments, again, can individualise the tone much more 
even than the piano; the themes of Bach's instrumental 
works are much freer and bolder than those of the clavier 
works. 

Whatever force we give to the accent, the phrasing must 
be kept discreet and unobtrusive. In the last resort the whole 
phrase, with its varieties of ties and the staccati between 
them, must be embraced as it were in one large tie, which 
permits of diversity without restlessness. The hearer 
really should not be conscious of the phrasing as such; 
he must only be conscious of it as a self-evident and 
vital illumination of the whole, down to the smallest detail, 
so that he is himself surprised at the ease with which he 
grasps this complicated polyphony. 



The Tempo. 381 

There is little to be said with regard to the tempo in 
Bach's clavier works, The better any one plays Bach, the 
more slowly he can take the music; the worse he plays 
him, the faster he must take it. Good playing implies 
fine phrasing and accentuation in every detail in every 
voice. This of itself sets certain technical bounds to speed. 
On the other hand, in playing of the right kind the hearer, 
even if the tempo is not quick in itself, has the feeling of 
it being quite fast enough, for the reason that at any 
quicker pace he could not grasp the detail. It should 
never be forgotten what a complicated process it really 
means for any musician, even for one who is not 
listening to it for. the first time, to follow one of 
Bach's polyphonic works properly. Of course if we are 
careless as to our phrasing and accentuation, and so oblit- 
erate the greater part of the detail, we can play faster 
with impunity, so as to give the music another interest 
of a kind. In general, however, the maxim holds good 
that the vivacity of a Bach piece depends not on the 
tempo but on the phrasing and the accentuation, 'in 
this sense every one may strive to play him with plenty 
of temperament. 

The tempo marks, where they exist, should not be inter- 
preted in a modern sense. Bach's adagio, grave, and lento 
are not so slow as ours, nor his presto so fast; therefore 
we are easily betrayed into making his slow movements 
too long-drawn and of hurrying his fast ones. The circle 
of possible tempi in his music is a relatively narrow one*. 
The question is really one of varied nuances on either side 
of a moderato. The presto of the Italian Concerto is, as 
a rule, played twice as fast as it should be. Nobody will 
make this mistake who tries to realise the complicated 
system of ties that Bach has indicated in the movement. 

It goes without saying that the alia breve sign in Bach 



* For this reason it is hard to see the use of the tmauthentic 
tempo marks that are sometimes inserted in the works. Metronome 
marks should be moderately used, and relegated to foot-notes. 



382 XVI. The Performance of the Clavier Works. 

implies nothing as regards the tempo, it does not double 
the speed of the four-crotchet beat. 

The ideal edition of Bach's clavier works would be a 
critically correct reprint of the original, in which the text 
is not adorned with ties and dynamic indications. It is 
hard to say how much mischief has been done by many of 
the old editions and arrangements. It would be some- 
thing to establish the principle that the editor should 
indicate all additions of his own, SQ that the player who 
has not the original edition at hand may know what is and 
what is not by Bach, and can use his own judgment. An 
inquiry among the piano-playing public would show that 
the majority have no idea that the ties and dynamic indica- 
tions in their scores are not Bach's own. 

The best thing for our Bach editors to do would be to 
follow Busoni's example, and publish, not arrangements, 
but their "interpretations" of the works. Busoni's edition 
of the Chromatic Fantasia (published by Simrock) is one 
of the finest achievements of this kind, that fascinates 
even those who cannot always see eye to eye with Busoni 
on the question of the permissible limits of modernisation 
of Bach's music. It is to be hoped that Billow's arrange- 
ment of this work is nowadays relegated to a bygone epoch 
of Bach interpretation; he himself would wish it to be so, 
were he alive now. 

"Instructive" editions of Bach's clavier works have this 
difficulty to contend with that in half of them the text 
becomes unreadable by reason of the number of ties, staccato 
dots, and accents that are needed in order to give vitality 
to the phrasing. Perhaps future editors will adopt the 
plan of expressing their views in prefaces and foot-notes 
instead of in the text. After all, the essential things are 
only the suggestion of the great dynamic lines and the 
phrasing and accenting of the themes; the details of the 
development of the work follow of themselves frorri these. 
Another advantage of editions of this kind would be that 
they could discuss various alternative phrasings, instead 



Epilogue. 383 

of peremptorily thrusting one upon the player. It cannot 
be denied that many editions of piano music do not induce 
self-reliance, but the lack of it. This does more harm in 
the case of Bach's music than in any other. No one can 
play it satisfactorily who is not conscious of the essential 
principles of its musical structure. 

Perhaps even the copious fingerings that adorn our 
Bach editions are not so beneficial as is generally supposed, 
for they relieve the player of the trouble and the profit 
of working them out for himself. 

The rules for playing Bach's clavier works here elaborated 
are open to many objections. The view that the phrasing 
of the preludes and fugues of the Well-tempered* Clavichord, 
is primarily to be learned from Bach's own phrasing of 
orchestral parts will perhaps be thought too drastic; the 
rules as to accentuation may appear to lay too much stress 
on characterisation; it will be objected that these dynamic 
rules insist too strongly on the architectonic side of the 
music. All this, however, does not in any way invalidate 
the rules. We may even regard it as a positive necessity 
to formulate general principles for performing Bach's 
clavier works; and it may be that these will coincide with 
the views of a number of our Bach pianists, who are also 
in revolt against excessive and unintelligent modernisa- 
tion of the works, and believe that more is to be hoped 
for from a careful artistic enquiry than from self-satisfied 
caprice. 

The greater or less extent to which these views are adopted, 
however, is of little consequence. They will have fulfilled 
their object if they prompt players to more thorough 
reflection and a more careful examination of the works. 
In any case, no reader can be so conscious as the author 
that the framing of rules is more unsatisfactory and more 
imperfect in art than anywhere else. 



384 XVII. Chamber and Orchestral Works. 

CHAPTER XVII. 
CHAMBER AND ORCHESTRAL WORKS. 

Bachgesellschaft Edition. 

Vol. XXVII. Six- sonatas for violin solo ; six suites for cello solo. 

- IX 1 . Three sonatas for clavier and flute ; one suite and 

six sonatas for clavier and violin; three sonatas for 
clavier and gamba ; sonata for flute, violin and continue ; 
sonata for two violins and continuo. 

- XLV. Four inventions for violin and clavier. 

- XLIII 1 . Three sonatas for flute and continuo. Concerto 

for four claviers with orchestral accompaniment (Vivaldi). 

- XVII. Seven concertos for clavier and orchestra ; concerto 

for clavier, flute, violin and orchestra. 

- XXI 1 . Three concertos for two claviers and orchestra. 

- XXXI 8 . Two concertos for three claviers and orchestra. 

- XXXI 1 . Overtures for orchestra. 

- XIX. Brandenburg concertos. 

- XXI 1 . Two concertos for violin and orchestra; concerto 

for two violins and orchestra. 

Petws Edition. 

Series II. Parts i 13. Works for clavier with orchestra. 

- III. Parts i 8. Works for violin and flute. 

- IV. Parts i and 2. Works for cello and gamba. 

- VI. Works for orchestra. 

Bach seems to have cultivated violin playing from 
childhood; when he left the gymnasium at Liineburg he 
was an accomplished violinist and could take a place as 
such in the orchestra of Johann Ernst, the brother of the 
reigning Duke of Weimar*. Nor did he neglect the stringed 
instruments in later life. In chamber music he played 
by preference the viola, for in this way he found himself 
as it were in the centre of the web of tone**. 

We have no direct information as to the extent of his 
proficiency on these instruments. We may be sure, how- 
ever, that he had a thorough practical knowledge of the 
technique of the stringed instruments, otherwise it 

* See p. 100. 
** See p. 209. 



Suites and Sonatas for Solo Violin. 385 

would have been impossible for him to take such unique 
advantage of all the effects that can be obtained on them t 
as he does in the polyphonic works for violin, gamba and 
cello solo. 

Nor can he quite forget that he is a violinist in the works 
written for keyed instruments; the violinist is observable 
on every page. The characteristic of Bach's piano and 
organ style is precisely this, that he demand? from the 
keyed instruments the same aptitudes for phrasing and 
modulation as the strings. At bottom he conceived every- 
thing for an ideal instrument, that had all the keyed instru- 
ment's possibilities of polyphonic playing, and all the 
bowed instrument's capacities for phrasing. This is how 
he came to write polyphonically even for a single instru- 
ment. 

Polyphonic playing on the violin had long been custom- 
ary in Germany. Bruhns (1666 1697) of Husum, Buxte- 
hude's pupil, used to perform in this way, at the same 
time playing the bass on th organ pedals *. The Italians 
were far behind the Northerners in this art. 

Although Bach wrote in reality three partitas (suites) 
and three sonatas, it is customary to speak, for brevity's 
sake, of his six sonatas for solo violin. As might be guessed 
from our knowledge of his way of composing, these works 
were written within a short time of each other. They 
belong to the Cothen epoch; the oldest autograph we 
have of them probably dates from about 1720. ^ 

The history of this autograph is told by Polchau, the 
gallant manuscript-hunter of the early days of Bach en- 
thusiasm, in a short note on the front page of it: "I found 
this excellent work, in Bach's own handwriting, in Peters- 
burg in 1814, among a lot of old papers, destined for the 
butter shop, that had belonged to Palschau, the pianist.** 
At a later date it passed, with the other manuscript trea- 



* On polyphonic violin-playing in the eighteenth century see 
Spitta II, 79 ff. 

Schweitzer, Bach. 25 



386 XVIL Chamber and Orchestral Works. 

sures of the finder, into the possession of the Berlin Royal 
Library. 

Polchau was in error, however, in imagining it to be an 
autograph of Bach's, It is from the hand of Anna Magda- 
lena, whose handwriting even at that time was deceptively 
like that of her husband. While she was making the copy 
she was watching over one of the boys perhaps Friede- 
mann, who used a free page for exercises of his own 
in making notes, his father having written some examples 
for him. There are no blank sheets, for the leaves are 
filled on the same principles as the English autograph of 
the Second Part of the Well-tempered Clavichord, in 
order to avoid the necessity of turning over, each work is 
made to occupy one or other side of a page only, 

In the later Leipzig period Anna Magdalena made a 
new copy of the sonatas for solo violin. She made them 
up in one volume with those for violoncello, with the title : 
"Pars I. Violino Solo, Senza Basso, compos^e par Sr 
Jean Seb. Bach. Pars II. Violoncello Solo, Senza Basso, 
compos^e par Sr. J. S. Bach. Maitre de la ChapeUe et 
Directeur de la .Musique & Leipsic. Ecrite par Madame 
Bachen, Son Epouse." Bach, it will be observed, is not 
called Court composer; the copy* must therefore be dated 
earlier than 1736. 

JN The sonatas for solo violin were first printed ia 1802, 
by Simrock of Bonn. In 1854 Robert Schumann edited 
them for Breitkopf and Hartel, adding a pianoforte accom- 
paniment to them. This was following in the footsteps 



* This manuscript also belonged to P61ch.au, at whose death 
it went to the Royal Library in Berlin. These two "autographs" 
are very interesting, as they exhibit a change in Bach's method 
of notation that is also met with elsewhere, and helps us somewhat 
to date his works. In the older manuscripts, a sharp is usually 
cancelled by the j? sign, rarely by the natural; later from about 
the first Leip2ig period, the natural is exclusively used for this 
purpose. 

These sonatas for violin solo exist also in a copy made by Joh. 
Peter Kellner in 1726. 



The Suites and Sonatas for Solo Violin. 387 

of Mendelssohn., who had adopted the same procedure 
in 1847 with the Chaconne from the second partita*. To 
us it is incomprehensible that two such great artists could 
believe that they were thus carrying out Bach's intentions. 

The sonatas and partitas are so arranged that each 
sonata is following by a partita. In both of them we 
hardly know what to admire most the richness of the 
invention, or the daring of the polyphony that is given 
to the violin. The more we read, hear and play them, 
the greater our astonishment becomes. 

The Chaconne that concludes the second partita has 
always been regarded as the classical piece for solo violin, 
and justly, since both the theme and its development 
are consummately adapted to the genius of the instrument. 
Out of a single theme Bach conjures up a whole world. We 
seem to hear sorrow contending with pain, till at last they 
blend in a mood of profound resignation-. 

It is very instructive to compare the Chaconne with the 
Passacaglia for organ, which is also in reality a chaconne**. 
For the organ, Bach takes a theme that is accented only 
on the strong beats of the bar, knowing well that the least 
syncopation would give the whole work a restlessness 
that would make it unbearable on the organ. On the 
violin, however, that permits of every kind of accent, 
cross-accentuation makes for superior force and vivacity. 
He therefore employs here a quite unusual amount of 
syncopation ; the instrument, free from, the hindrance of 
an accompaniment, shall for once realise all its powers 



* He published his arrangement first through Ewer and Co., 
London, and afterwards (in 1849) through Breitkopf and Hartel. 
The Leipzig Conservatoire edition, with David's fingering and 
bowing, was issued by Kistner in 1843. See Dorffel's erudite pre- 
face to B. G. xxvm. 

** See p. 280. The chaconne and passacaglia are derived from 
old dance forms, and are characterised by the fact that they are 
developed out of an ever-recurring theme of eight bars in 3 / 4 time. 
In the chaconne this theme may appear in all the parts; in the 
passacaglia it is confined to the bass. 

25* 



XVII. Chamber and Orchestral Works. 



in perfect unrestraint. It is interesting to Jook at the two 
themes, side by side, that between them incarnate the 
organ and the violin music of Bach : 

Chaconne for violin. 




Passacaglia for organ. 



Notice how, in the Chaconne, Bach alternates between 
polyphonic and monophonic writing, so as to give the hearer 
relief, and to heighten the effect of the polyphony by the 
monophony interspersed among it, His music as a whole 
is full of these fine calculations of effect. 

Every one who has heard these sonatas must have 
realised how sadly his material enjoyment of them falls 
below his ideal enjoyment. There are many passages in, 
them that the best player cannot render without a certain 
harshness. The arpeggio harmonies sometimes make a par- 
ticularly bad effect, even in the finest playing. Polyphonic 
arpeggio playing is and must be an impossibility. There 
is thus some justification for the question whether Bach, 
in these sonatas, has not overstepped the bounds of artistic 
possibility. If it be so, he has for once acted against his 
own principles, for everywhere else he has been careful to 
set an instrument only such tasks as it can solve with 
satisfaction to the ear. 

Recent research seems to show that the traditions of 
Bach's own day can throw some light on this. In an 
interesting article by Arnold Schering, one of the most 
assiduous of the Bach students of our time, some passages 
from old works are cited that make it probable that the 



Polyphonic Violin Playing in Bach's Time. 389 

old arched bow, with which the tension was effected not 
by means of a screw but by the pressure of the thumb, 
was still in use in Germany in Bach's time*. The flat, 
mechanically stretched Italian bow, the predecessor of 
that of today, was indeed known in Germany from the be- 
ginning of the eighteenth century, but displaced the older 
one very slowly. 

The German violinist of Bach's day could thus stretch 
the hairs tighter or relax them as he liked. Chords that 
the virtuosi of today can only play with difficulty and 
without any beauty of effect by throwing the bow back 
on the lower strings, gave him no trouble at all; he simply 
loosened the hairs a little, so that they curved over the 
strings. This accounts for the fact that the Germans cul 
tivated polyphonic playing on the violin**, while it was 
almost unknown to the Italians. In Italy, the straight 
bow with its purely mechanical tension had already es- 
tablished itself by the end of the seventeenth century. 
This bow permits polyphonic playing only to a limited 
extent, since it allows of no further tension during the per- 
formance, for if the hairs are relaxed they do not arch over 
the strings, but fall against the stick. With the old German 
bow, on the other hand, the stick of which was not straight, 
there was a fair space between the hairs and the bow. 



* See the "Bachheft" of the Neue Zeitschvift fur Musik, 1904, 
p.675 ff, Verschwundene Traditionen des Bachzeitalters. The 
article appeared in an expanded form in the Bachjahrbuch for 1904, 
pp. 104 115. The chief passage cited by him runs thus: "As 
regards the small and medium-sized violins, the Germans hold the 
bow like the Lullists, pressing the hairs together with the thumb, 

the other fingers resting on the back of the bow ; . . . while 

the Italians leave the hairs untouched, as also the gambists and 
others do in the bass, the fingers lying between the wood and the 
hair" (Georg Muffat's preface to the Florilegium secundum, 1698). 
Caspar Majer testifies to the same thing in his Neu eroffneter Mttsik- 
Saal (Nuremberg, 1741). Schering's theory has already been 
mentioned on p. 209. 

** Interesting, examples of violin chords are given in Joh. Jak. 
Walther's Hortus chelicus, 1694. See Schering's article in the Neue 
Zeitschrift fttr Musik, 1904, p. 677. 



390 XVII. Chamber and Orchestral Works. 

The last representative of chord playing on the violin 
was the Norwegian, Ole Bull (1810 1880). His bridge 
was quite flat; he had his bows made in such a way that 
the stick stood at a considerable distance from the hairs*. 
It is interesting to know that he always maintained that 
this method was no new invention of his own, but a return 
to the true violin method of the past. It is quite possible 
that in Scandinavia the traditions of the seventeenth 
century had been retained, along with the old bows, down 
to Ole Bull's time. 

' In his sonatas for violin solo Bach has thus demanded 
of the instrument nothing impossible or even unsatisfactory 
per se, but only what seems so with our excessively arched 
bridges and our flat Italian bows. To play these sonatas 
as he did, we need only to file down the arching of the bridge 
so as to bring the strings almost level, and to use a bow 
so shaped from nut to point that the hairs can curve to- 
wards the stick without touching it**. Still better is a bow 
with slightly curved stick. In this way violinists will be 
able once more to play Bach in a correct style. Any one 
who has heard the chords of the Chaconne played without 
any restlessness, and without arpeggios, can no longer 



* His stick was bent downwards considerably at the point. 
Schering does not mention Ole Bull. A special study of the play- 
ing of this notable virtuoso should have some bearing on the 
question under discussion. Corelli's bow was stretched mechani- 
cally, but it was not so flat as ours. 

** It is to be hoped that the instrument makers will soon provide 
us with serviceable bows of the old type. Till then the player must 
make shift with an old bow, inserting a couple of pieces of wood 
between the stick and the hairs at top and bottom. A bow recon- 
structed in this primitive way will do quite well to experiment 
with. I owe to my friend E. Hahnemann the opportunity of 
hearing the Chaconne and other works of Bach for solo violin 
played in the old way by an eminent violinist with a bow thus 
provisionally arranged. The flat bridge is less disadvantageous 
than one would think, as in Bach's music we have not to play in 
the high positions, where the string is pressed down so deeply that 
there is a danger of the bow touching two strings at once, Ole 
Bull, according to Spohr, used his A and D strings only in the 
lower position. 



Polyphonic Violin Playing in Bach's Time. 391 

doubt that this is the only correct and, from the artistic 
standpoint, satisfactory way of playing it. Of course the 
slackly-stretched bow demands a different technique from 
the usual one. The pure "springing" bowing is impossible*. 
But Ihe tone also undergoes a change ; it acquires a curious 
softness. If we play the chords with the hairs of the bow 
relaxed, we get an almost organ-like tone, somewhat like 
that of a soft salicional. To get an idea of this tone, un- 
screw the hairs of an ordinary bow, place the stick under 
the violin, lay the hairs over the strings, and fasten them 
again to the stick. If we move this reversed bow, we ob- 
tain the organ-like ethereal tone that the relaxed bow 
produced. 

When we have imitated the old bow as best we can, 
and given the hairs their utmost tension by thumb pressure 
in the monophonic passages, we still do not get the power- 
ful tone of the modern mechanically-stretched bow. We 
purchase beauty of tone, that is to say, at the expense of 
some loss of strength. The reproach was always being 
levelled against Ole Bull that his tone was weak. 

It is a question whether the modern public would ac- 
custom itself to this weak tone. In large concert rooms 
it will scarcely be possible to play the sonatas for solo 
violin in the old way, as the tone would not "carry" suf- 
ficiently. In chamber music performances, on the other 
hand, the proper style of rendering should suit admirably. 
If we have once heard the Chaconne in this way we can- 
not afterwards endure it in any other. Thus the result 
would be that the works for solo violin would disappear 
from the programmes of the larger concerts, and be restored 
to the chamber music to which they really belong. 

The movements of the sonatas for the solo violin that 
exist also for the piano or organ are arrangements of violin 



* Schering conjectures that the echo passages that play a large 
part in Bach's orchestral scores were rendered by all the string players 
sxiddenly relaxing the hairs of their bows. It would be interesting 
to experiment with a string orchestra playing with the old bows. 



392 XVII. Chamber and Orchestral Works. 

originals*. It is doubtful whether they are all effective in 
this form. They show, however, how completely the violin 
method of phrasing ranked with Bach as the universal 
method, to which the keyed instruments had to try to 
conform. When we read the prelude of the third partita 
we find it impossible to believe that Bach could have 
entertained the idea of asking the organ to perform these 
repeated semiquavers, the proper articulation of which 
is possible only on a Bowed instrument yet this is what 
he actually does in the instrumental prelude to the "Rats- 
wahl" cantata Wiv danken dir, Gott, wir danken Air (No, 29)**. 
He could only venture to do so because he himself played 
the part on the Rilckpositiv on the organ at St. Thomas's. 
The fugue of the first sonata may have been originally 
conceived for the organ. Its theme is derived from the 
first movement of the "Veni Sancte Spiritus", and runs 
thus 



Mattheson quotes this theme in his Grosse Generalbass- 
schule, and sketches a development of it which resembles 
Bach's at many points, especially in the employment of 
a chromatic counter -subject. As he quotes in the same 
place the theme of the great organ fugue in G minor with- 
out mentioning Bach's name, it may be taken for granted 
that he had also the fugue of the third sonata before him. 
Did Bach play it in Hamburg, about 1720, as an organ 
fugue? Assuredly not in the present form, for the structure 
of the fugue it reminds us of the Chaconne has no- 



* The fugue from the first sonata (G minor) appears as an organ 
fugue in D minor (Peters III, No, 4). The second sonata (A minor) 
exists as a clavier sonata in "D minor. The first movement of the 
third sonata (C major) is arranged for the clavier in G major. It 
is very interesting to see how Bach has here transformed and vivi- 
fied the bass part. (B. G. XLII, p. 27 ff.) 
** The cantata was written in 1731. 



The Suites for Cello Solo. 393 

thing whatever in common with those for the organ, but 
is wholly designed for the violin. Here again, therefore, 
the violin form must be the original one*. 

In two places Mattheson speaks of the fugue of the 
A minor sonata 



ft 



with unreserved admiration**. 

The six sonatas for cello solo also belong to the Cothen 
period. They are as perfect in their own way as the works 
for violin solo. Chord-playing, of course, is not used to 
anything like the same extent; nor does Bach even employ 
a simple kind of two-part polyphony. This is to be ex- 
plained by the fact that the Germans played the "big 
violin" with the non-relaxable bow. 

In quality these suites remind us of the French. The 
last but one is described as "Suitte discordable", and 




requires the tuning of the A string down to G, thus 

f 

For the last, Bach requires a five-stringed instrument, 




with an E string above the A 9 A -~ ^ There may 

__ 

have been five-stringed cellos at that time. It is more 
probable, however, that Bach wrote the suite for the 

* On this question see Spitta II, 82 ff . ; he supposes that Mat- 
theson he^ard Bach, play the fugue on the organ, as the former 
knew it in 1727, at a time when he had probably not seen the so- 
natas and suites for violin. But why not? Why should not B$ch 
have taken these violin works to Hamburg with him in 1720? 
** Kern Melodischer Wissenschaft (1737)1 P- H7? && Vo ^" 
kommene Kapellmeister (1739), p. 3^9- He quotes the theme in- 
correctly each time (Spitta II, 79). 



394 XVII. Chamber and Orchestral Works. 

"yip|j| pomposa" that he invented, and that was strung 
in this way. We know from Gerber, who was the son of 
one of Bach's pupils, that Bach employed this instrument 
in the orchestra in the early Leipzig years*. c The stiff 
way", says the lexicographer, "in which the violoncello 
was played in Bach's time compelled him to invent, for the 
animated basses in his works, the so-called viola pomposa, 
which was a little longer and deeper than a viola, and was 
tuned like a violoncello, with a fifth string, e, and was laid 
on the arm; on this convenient instrument very high and 
rapid passages were easier." 

In .considering the works for a solo instrument with 
clavier, we must remember the distinction of that day 
between the obbligato clavier and the accompanying cla- 
vier. In a sonata with obbligato clavier, the latter plays the 
,chief role, several obbligato parts being given to it, while 
the solo instrument has only one. In Bach we do not 
get a "sonata for violin and clavier", or "sonata for flute 
and clavier", but "sonata for clavier and violin" and 
"sonata for clavier and flute". A light is thrown on the 
way of looking at the matter in that epoch by the fact 
that a work for clavier and violin, if the polyphony was 
in three parts, was called a trio; ihey counted, that is to 
say, not the instruments but the obbligato parts. By 
sonata for violin and clavier Bach means a composition 
in which the clavier only supplies the bass and the figured 
bass. Even Zelter uses the terms in this sense. 
*** Of Bach's works for obbligato clavier and violin we 
possess a suite in A major** and six sonatas***; for violin 
with accompanying clavier he wrote a sonata, a fugue f 
and four Inventions ff. 

The suite for clavier and violin is not on the same level 



* See p* 205. 
** B. G. IX, p. 43 ff. 
*** B. G, IX, p.69ff. 

t B, G. XLIIIi, p. 31 ff. 
ff B. G. XLVi, p. 172 ff. 



The Sonatas for Clavier and Violin. 



395 



as the six sonatas; it is probably an earlier work. Bach 
incessantly improved the sonatas, as is shown by the 
various copies of them. The last version is represented 
by a copy of Altnikol's. The manuscript that belonged 
to Emmanuel is overloaded with ornaments. In the same 
way, Friedemann, when transcribing his father's organ 
sonatas, embellished them according to the taste of the 
epoch*. This shows us how much reliance is to be placed 
on the ornaments in any work of Bach's that we possess 
only in copies, even though these are by his sons. 

These sonatas were written at Cothen, How small, by the 
side of them, seem the works of Corelli and the other Italian 
violin composers, at whose feet Bach had sat in Weimar 1 

Bach's sonatas, like Beethoven's, depict soul-states and 
inner experiences, but with force in the place of passion. 
Whether he is sunk in sorrow or in mystical dreams, Bach 
always recovers himself in a compact fugal finale. 

Sorrow predominates; we could almost imagine that 
Bach wrote these works under the impression of the loss 
of his first wife. The Siciliano of the fourth sonata is con- 
structed on a theme that closely resembles that of the aria 
"Erbarme dich" in the St. Matthew Passion, and a sob 
runs through both of them 

Siciliano from the fourth sonata. 




Aria "Erbarme dich" from the St. Matthew Pass-ion. 




* For an account of the provenance of these sonatas see Rust's 
fine preface to B, G. IX. It is to be regretted that many autographs 
were not available for the able editor of the volume of instrumental 
sonatas until after it had been printed, so that many of Bach's 
interesting phrasing-marks are not given. The Inventions were 
only discovered a few years ago. They do not exhibit the same 
mastery as the sonatas, 



396 XVII, Chamber and Orchestral Works. 

This movement has of course nothing in common with 
the "Siciliano" of earlier composers except its motion 
and its rhythm. It has nothing* of the lyrical pastoral 
mood; it is filled with the deepest pathos throughout. 
It should be played in this spirit, the violin part, as in the 
corresponding aria in the St. Matthew Passion, being 
heavily rather than lightly accented. The third and sixth 
quavers should be brought out with some force, and op- 
posed, by means of a certain weight of emphasis, to the 
following strong beat. If Bach had not conceived it thus 
he would not have aimed at greater intensity, towards 
the end, by throwing the accent back on to the weak 
beat by means of syncopation 







Many of our violinists play these sonatas in a * 'sentimental'* 
style instead of the "expressive" Bach style. An ex- 
perienced critic in the Neue Zeitschrift fur Musik rightly 
complains that in the practical edition of the new Bach- 
gesellschaft the player will be misled into beginning the 
proud and energetic themes of the allegro movements in 
a soft piano, in order to get the usual gradual crescendo 
and will do this even in the theme of the last sonata* 







That this is intended to suggest impetuous motion is shewn 
by the fact that Bach uses it again, only in fuller form, 
in the secular cantata Weichet nur, betrttUe Schatten (B. G, 
XI 2 ), to illustrate the words "Phoebus flies with swift 
horses". 



* See the Neue Zeitschrift fur Musik for 1904, pp. 686, 687, 
He also complains with justice that a great number of ties and 
staccato signs have been inserted in the text. 



The Performance of the Sonatas for Clavier and Violin. 307 

Of what use are all the pianos, mezzo-fortes and fortes 
that the majority of our players and many editors import 
into these sonatas? The musician has yet to be born 
who can convince another where the -piano or the forte 
or the crescendo or decrescendo should begin and end in 
the various movements. Of course the obvious echo 
effects and the contrast intended between a pianoforte 
theme in one part and the tutti theme in another admit 
of no question. But what nuances are we to put, for ex- 
ample, into the first movement of the B minor sonata 
or that of the E major? Why should we play now softly, 
now loudly, in the andante of the first sonata? As this 
alternation is not an essential part of the structure of the 
works, its only effect must be to create an impression of 
aimlessness and unrest. What should we say if any one 
were to paint over fine old steel engravings in modern 
tints, under the pretext of heightening their effect? Yet 
there is no such general outcry against the variety of colours 
that are imposed upon Bach's works. 

In general, each movement of the violin sonatas should 
be played with a uniform strength of tone. This does 
not mean that they are to be played monotonously. De- 
clamatory nuances, contained within the range of the 
general tone-quality, must bring out the detail in the same 
animated way as in the case of the pianoforte works. 
But from the architectonic standpoint there should be 
no insistence, in this music, on variations of tint that 
will affect the broad plastic lines of it; our modern emo- 
tional dynamics would obliterate the plan of the work 
and give a false idea of the co-operation of the three ob- 
bligato voices*. 

We should try to play these sonatas with an eye only 
to the animated and plastic working-out of the detail, 
and to a kind of broad declamation, leaving the variety 
to corne from the vivacious interplay of the parts. The 



* On the principles of the Bachian dynamics see p. 355. 



398 



XVII. Chamber and Orchestral Works. 



direct impression the pieces make in this way will prove 
the Tightness of the principle. 

As regards the tempo, the andante movements are as 
a rule taken rather too slowly, and the allegros much too 
fast*. We seldom meet with a player who plays, for ex- 
ample, the theme of the final allegro in the third sonata 
in such a way that the hearer receives the impression of 
proud strength that should be given by this sequence of 
intervals built up on a basis of moving semiquavers 




Even the best performance, however, is not wholly 
satisfactory. This is the fault of our pianoforte, the dull 
timbre of which does not blend with that of the violin. 
It is hopeless to try to blend a modern grand piano and 
a stringed instrument; as Wagner, says, the timbre of the 
piano and that of the violin are incompatible. In a Bach 
clavier sonata this is unpleasantly evident, for the com- 
poser has calculated on the co-operation of absolutely 
homogeneous obbligato parts. This was obtainable in 
his day, the cembalo producing the pure tone of a string 
vibrating on a wood resonance. It was only slightly 
brightet than that of the violin owing to its coming from 
a metal string. When the cembalo part is played on a 
modern piano the ensemble of the equal homogeneous parts 
is destroyed; we only hear a solo with pianoforte accom- 
paniment. 

Even if, in the future, we should go back to a domestic 
instrument that substitutes a pure, silvery timbre for the 
thick and brutal tone of our huge grands and uprights, the 



* On the general question of tempo see p. 381. 



The Performance of the Sonatas for Clavier and Violin. 399 

difficulties with regard to the performance of the sonatas 
for piano and violin would be only half surmounted. In the 
clavier parts of these works, Bach counts on the simultaneous 
doubling of a tone in several octaves, which is possible 
only^on the cembalo. A good cembalo, like that of Friede- 
mann's acquired by Count Voss, had four strings to each 
note, two giving the ground tone, one the lower octave, 
and one the upper octave. According to the couplers he 
used, the player could strike simply the two ground-tone 
strings, or these and that of the upper octave, or all four 
together. The tone was small but very rich. In this 
respect our modern piano cannot compare for a moment 
with the sparkling, dashing cembalo. Moreover the player 
could bring out the theme more pointedly by doubling it 
in the upper octave, or if the theme was in the bass 
in the lower octave. We can thus imagine how different 
the adagio of the first sonata, that of the third, and the 
largo of the fifth, would sound then from what they do on 
a keyed instrument of today, on which it is impossible to 
play in three octaves at once. As a matter of fact the 
adagio of the fifth sonata, with the double stopping in the 
violin and the arpeggio demisemiquavers in the davier, 
is positively unpleasant when performed nowadays. The 
necesssary vapor ousness and at the same time definite- 
ness of the tone can never be got on our grand piano, and 
the double stopping in the violin is never quite beautiful, 
for it can only be properly done with the loosened bow. 
Only in this way can these two violin parts receive the 
delicate, organ-like tone they need, with the silvery tone 
of the clavier playing round it*. Nature fortunately is 
compassionate, and lets us believe that we hear music 
only with the ears, while in truth we take it in at the same 
time through the eyes, and correct our hearing accordingly. 
We enjoy these works because the eye's perception of the 



* It is an interesting fact that in the original version of this 
work the piano has semiquavers, not demisemiquavers. See 
B. G. IX, 250 it 



4OO XVII. Chamber and Orchestral Works, 

beauty of them on the paper, and the mind's conception 
of the noble 1 counterpoint, permit the ear to believe that 
the works sound well in performance; and the delusion 
continues even if both players indulge in senseless alterna- 
tions of pianissimo and fortissimo. But any one who has 
the misfortune to hear only with the ears is bound to admit 
that Bach's sonatas for clavier and violin imperatively 
demand the cembalo. This was insisted upon by Rust 
in his preface to the original edition (1860)*. He could 
not foresee that forty years later the demand would be 
still more peremptory, and that an attempt would be made 
to meet it by the construction of the new cembali. We 
would not deprive those who have only a modern piano 
of their pleasure in these sonatas; only we must be quite 
clear as to what an ideal performance of the works would 
sound like, and not seek this ideal in a false modernisation 
of them. 

We must not omit to mention the tradition that in per- 
forming sonatas of this kind the cembalo bass was helped 
by a stringed instrument. We shall, in fact, find that a 
discreet violoncello does good service in these works of 
Bach, especially where the theme has to be brought out in 
the lowest voice. This can be tested, for example, in the 
largo of the F minor sonata. An old manuscript partly 
autograph of the sonatas categorically recommends the 
use of an optional gamba to strengthen the bass. It bears 
the title: "Sei Suonate & Cembalo certato d Violino Solo 
col Basso per Viola da Gamba accompagnato se piace, 
Composte da Giov. Sebast. Bach," 

The chords to be struck between the obbligato voices 
are only rarely figured. They must be added, however, in 
every place where the contrapuntal web is thin for 
example, when only the violin part and the bass are at 
work, or when the clavier alone is playing in two parts. 
If the movement begins with only a bass note, it goes with- 

* B. G, IX. 



Other Instrumental Sonatas. 401 

out saying that the whole chord is to be struck, but in 
such a way as if it were being sounded by another clavier, 
not by the one that is playing the obbligato parts. In this 
discreet way the harmonic basis of the whole can be in- 
dicated throughout, even when three or more voices are 
going together. But to handle the piano part in this 
way requires exceptional skill, and a knowledge of the 
peculiarities of Bach's figuring, otherwise the harmony will 
not be correctly filled in. Kirnberger appears to have em- 
ployed two claviers when playing the sonatas; one played 
the obbligato parts, the other strengthened the bass and 
supplied the harmonies*, 

We also possess the following instrumental sonatas by 
Bach : three wonderful ones for clavier and gamba (B. G, 
IX, p. 175 ff.); three for obbligato clavier and flute (B. G. 
IX, p. 3 ff.); three for flute with clavier accompaniment 
(B. G. XLIII 1 , p. sff.); one for two violins with clavier 
accompaniment (B. G. IX, p. 231 ff.); one for two flutes 
with clavier accompaniment (B. G. IX, p. 26off.) which 
Bach afterwards re-wrote as the first sonata for clavier 
and gamba, though it sounds better for two flutes**. 

The end of the first movement of the third sonata for 
clavier and flute (A minor) is lacking. Bach wrote it on 
the same sheet as one of the concertos for two claviers 
with orchestral accompaniment, employing, in his usual 
economical way, the three vacant staves at the bottom 
of each page. From six of these leaves the lower part has 
been cut away, so that we lack some fifty bars of the 
sonata***. The autograph was already in this mutilated 
state when von Winterfeld bought it for a few groschen 
from an antiquary in Breslau. 

The sonata for flute, violin and accompanying clavier 



* See Rust's remarks on the existence of two clavier parts in 
Kirnberger's copy of the sonatas (B, G. IX, Preface, p. 17). 

** A sonata for violin and clavier in G minor (B. G. IX, p, 274), 
if it be genuine, is a youthful work. 

*** The fragment is given in B. G. IX, 245 ff. 



4O2 XVII. Chamber and Orchestral Works, 



(G major)* is written for the "violino discordato" ; Bach de- 
sires the player to tune the two upper strings a tone lower, 
jy j and consequently writes the part out 
_ jjm jp r a tone higher. As there is nothing in 
u *^ * the violin part that could not equally 
well be played with the strings tuned 



I 



in the ordinary way, Bach's purpose in prescribing the 
peculiar tuning can only have been to get a softer timbre, 
that would blend better with that of the flute. 

Of Bach's orchestral works probably scarcely any have 
been lost; we possess four large suites** and six con- 
certos***. 

It cannot be settled whether the suites were written in 
Cothen or in Leipzig. In any case Bach performed them 
not only before the Prince of Cothen but also in the Tele- 
mann Musical Society at Leipzig, which he conducted from 
1729 to 1736. He calls these works overtures, not suites 
or partitas, this being the customary name, at that time 
for an orchestral suite in which the introduction played 
the chief part. They are, however, just as much real par- 
titas as those in the Klavieriibung, except that the old 
dances, the allemande, the courante and the sarabande, 
retire in favour of the newer and freer movements, 

The introductions are monumental movements, all con- 
structed on the plan of the French overture, They begin 
with a stately section; to this succeeds a long and brilliant 
allegro; at the end the slow section returns. When Men- 
delssohn, in 1830, played to the old Goethe, on the piano, 
the overture of the first of the two suites in D major, 
the poet thought he saw a number of well-dressed people 
walking in stately fashion down a great staircase f. In 
1838 Mendelssohn succeeded in getting them performed 
by the orchestra at the Gewandhaus, Leipzig. It was the 



* B. G. IX, aai ff, 
** B.C. XXXI* (i B8 1). 
*** B. G. XIX (1868), 
| See ante, p. 241. 



The Brandenburg Concertos. 403 

first performance of any of these splendid works since 
Bach's death*, 

In the dance melodies of these suites a fragment of a 
vanished world of grace and elegance has been preserved 
for us. They are the ideal musical picture of the rococo 
period, Their charm resides in the perfection of their 
blending of strength and grace. 

The celebrated "aria" is found in the first D major 
overture. 

The six concertos are known as the "Brandenburg*', 
having been written for the Margrave Christian Ludwig 
(1677 1734). He was the youngest son of the great 
Electoral Prince by the latter's second marriage; he was 
passionately devoted to music, and maintained an excellent 
orchestra. He made Bach's acquaintance about 1719, 
perhaps at the Meiningen court, with which he had 
relations through his sister, or perhaps in Karlsbad, 
when Bach accompanied Prince Leopold there. En- 
chanted by Bach's playing, he asked him to compose 
some works for his orchestra. Bach fulfilled this wish 
and sent him two years later these six concertos with the 
following dedication : 

A son Altesse Royalle, Monseigneur Cr&tien Louis, Marggraf 
de Brandebourg. 

Monseigneur 

Comme j'eus ily aune couple** d'annees, le bonheur de me faire 
entendre a Votre Altesse Royalle, en vertu de ses ordres, 8c que 
je remarquai alors, qu'Elle prennoit quelque plaisir aux petits 
talents que le Ciel m'a donnes pour la Musique, et qu'en prennant 
Conge de Votre Altesse Royalle, Elle voulut bien me faire 1'honneur 
de me commander de Lui envoyer quelques pieces de ma Com- 
position: j'ai done selon ses tres gracieux ordres, pris la libertfc 
de rendre de mes tres-humbles devoirs a Votre Altesse Royalle, 



* CThey were published, but only in part, by Peters in 1853. 
The firm did not venture to print the second I) major overture 
until 1881, not being sure till then, in default of an autograph, 
that the work was really Bach's! 

** This can only mean two years, It is inconceivable that Bach 
should allow a longer time to elapse before he carried out the wishes 
of the Prince. 



404 XVII. Chamber and Orchestral Works. 

par Ics presents Concerts, que j'ai accomodes a plusieurs Instru- 
ments; La priant tres-humblernent de ne vouloir pas juger leur 
imperfection, a la rigueur du gout fin et delicat, que tout le monde 
scait qu'Elle a pour les pieces musicales; mais de tirer plutot en 
benigne Consideration, le profond respet, < la tres-hurnble obeis- 
sance que je tache & Lui t6moigner par 1&. Pour le reste, Mon- 
seigneur, je supplie tres humblement Votre Altesse Royalle, d'avoir 
la bontfe de continuer ses bonnes graces envers moi f et d'etre per- 
suadee que je n'ai rien tant & coeur, que de pouvoir etre employ6 
en des occasions plus dignes d'Elle et de son service, moi qui suis 
avec une zele sans pareil 

Monseigneur De Votre Altesse Royalle 

Le tres humble & tres obeissant serviteur 
Jean Sebastien Bach. 

Coethen, d. 24 Mar (Mars? Mai?) 
1721. 

(To his Royal Highness, Monseigneur Crdtien Louis, Margrave 
of Brandenburg. 

Monseigneur, 

As I had the honour of playing before Your Royal Highness a 
couple of years ago, and as I observed that You took some pleasure 
in the small talent that heaven has given me for music, and in 
taking leave of Your Royal Highness You honoured me with a 
command to send You some pieces of my composition, I now, accord- 
ing to Your gracious orders, take the liberty of presenting my 
very humble respect to Your Royal Highness, with the present 
concertos, which I have written for several instruments, humbly 
praying You not to judge their imperfection by the severity of 
the fine and delicate taste that every one knows You to have for 
music, but rather to consider benignly the profound respect andi 
the very humble obedience to which they are meant to testify. 
For the rest, Monseigneur, I very humbly beg Your Royal High- 
ness to have the goodness to continue Your good graces towards 
me, and to be convinced that I have nothing so much at heart as 
the wish to be employed in matters more worthy of You and Your 
service, for with zeal unequalled 
Monseigneur, 

I am 
Your Royal Highness's most humble and most obedient servant 

Jean Sebastian Bach. 
Coethen,, 24 Mar (March? May?) 
1721.) 

How the Prince received this gift, and how he rewarded 
Bach for it, we do not know. When he died, these concertos, 
together with the rest of his large musical collection, were 
inventoried and valued. They do not, however, figure in 



The Brandenburg Concertos. 405 

the inventory under the composer's name, like those of 
Vivaldi and other Italians, but are included in two lots, 
one containing seventy-seven concertos by various writers, 
the other a hundred. Each of these concertos was valued 
,, at four groschen*. T^w$ m the y$&r 1734 the six Branden- 
burg concertos were worth twenty-four groschen. & a 
lat^r date the autograpK score which Bach ha3 sent to the 
Margrave came into the possession of Kirnbergef, who left 
it to his pupil Princess Amalie of Prussia. SJieJbegueathed 
it to the library of the Joachimsthal Gymnasium, \yhence 
it afterwards came to the Royal Library in Berlin. f In 
elegance and cleanness this autograph surpasses even 
the famous score of the St. Matthew Passion. jjtic bar- 
lines are drawn throughout with a ruler**. 

In spite of all this care, the score contains an error. In 
the eleventh bar of the fifth concerto the semiquavers of 
the viola descend in fifths with those of the obfyligato 
qembalo* It is interesting to see that the error is due to a 
correction made by Bach in the fair copy. He noticed, 
that is to say, that the viola in the original version 
which we have in the orchestral parts, that have been 
accidentally preserved ascended in hidden octaves with 
the solo violin. He at once erased this sequence, which he 
had already written in the fair copy, and inserted in place 
of it the descending semiquavers, without observing that 
in this way he was falling out of the frying-gan into the 
fire***, This is presumably the passage referred to by 
Zelter in a passage on Mendelssohn in one of his letters to 
Goethe. "In the score of a splendid concerto by Sebastian 
Bach," he says, "the lynx eye of my Felix, when he was ten 
years old, detected six consecutive fifths, which I perhaps 



* Spitta discovered this inventory in the Royal domestic 
archives in Berlin. See his Bach, II, 129. 

** %hj5^cw^^ t?y Peters, in ijSjp..^ 

*** See Rust% pirelace' to' B, G XIX, "j^ifT '"Hie 'liachge- 
seUschaft editors have elected to print the older version as the 
lesser evil. 



406 XVII. Chamber and Orchestral Works. 

would never have discovered, as I do not bother about 
these things in large works, and the passage is in six 
parts*." 

The Brandenburg concertos are the purest products of 
Bach's polyphonic style. Neither on the organ nor on the 
clavier could he have worked out the architecture of a 
movement with, such vitality; the orchestra alone permits 
him absolute freedom in the leading and grouping of the 
otTfiTgato voices. When we said, in another connection 
that Bach's mode of expression is to be conceived as a 
plastic one, and deduced from this certain principles for 
performance, there was a danger of our being misunderstood, 
as if our object were to try to re-introduce the old, stiff 
way of playing his music. But one has only to go through 
these scores, in which Bach has marked all the nuances 
with the utmost care, to realise that the plastic pursuit of 
the musical idea is not in the least formal, but alive from 
beginning to end. Bach takes up the ground-idea of the 
old concerto, which develops the work out of the alternation 
of a larger body of tone the tutti and a smaller one 
the concertino. Only with him the formal principle becomes 
a living one. It is not now a question merely of the alterna- 
tion of the tutti and the concertino ; the various tone-groups 
interpenetrate and react on each other, separate from each 
other, unite again, and all with an incomprehensible artistic 
inevitability. The concerto is really the evolution and the 
vicissitudes of the theme. We really seem to see before 

* Briefwechsel zwiscken Zelter und Goethe, ed, Reclam II, 394; 
letter of 25 May 1826. There is another error in the hundred and 
eighty-second bar (B. G. XIX, p. 1 20, bar 4). The last three quavers 
of the first flute descend in octaves with the bass. Spiro's ingenious 
and unquestionably correct "Bach-conjecture" is well known; he 
regards the error as simply a clerical, one and proposes to play 
the$e three quavers a third higher. (See the S&mmelbdnde der 
Internationalen Musikgeselhchaft, 1900 1901, pp. 651 -653.) It 
may be less well known that the mistake must be laid not at Bach's 
door but at that of the editors, as Professor Ludwig, of Strassburg, 
kindly informs me. See the somewhat confused correction in the 
following year's issue, ~~ B. G. XXXVIII, p. 45, 



Performance of the BrandenbuRg-MToncertos. 407 

us what the philosophy of all ages conceives as the funda- 
mental mystery of things, that self-unfolding of the 
idea in which it creates its own opposite in order to over- 
come it, creates another, which again it overcomes, and so 
on and on until it finally returns to itself, having mean- 
while traversed the whole of existence. We have the same 
impression of incomprehensible necessity and mysterious 
contentment when we pursue the theme of one of these 
concertos, from its entry in the tutti, through its enigmatic 
struggle with its opposite, to the moment when it enters 
into possession of itself again in the final tutti, 

In Bach we often have not one but several groups of 
solq_ instruments, that are played off against each other 
m the development of the movement. The wind instru- 
ments are used with the audacity of genius. In the first 
concerto Bach employs, besides the strings, a wind-ensemble 
consisting of two horns, three oboes and bassoon; in the 
second, flute, oboe, trumpet and violin are used as a kind 
of solo quartet against the body of the strings ; in the third 
he aims at no contrast of timbres, but employs three string 
Jxios, all constituted in the same way; in the fourth con- 
certo the concertino consists of one violin and two flutes; 
in the fifth it consists of clavier, flute and violin; in the 
sixth, Bach employs only the timbre effects to be had from 
the strings, two violas, two gambas, and cello, 
ft The study of Bach's nuances in these works is a con- 
tinual source of delight. They are all so simple, and yet 
so full and rich. Observe, for example, how, in the first 
movement of the fourth concerto, from the twenty-seventh 
bar before the end, the piano comes down in a wavy line 
from top to bottom, following the line of the forte theme, 
that winds downwards and lies, as it were, in violent con- 
vulsions on the ground, till suddenly a bold forte of the 
whole orchestra puts an end to the unrest that began with 
the first entry of this subsidiary theme in the violins, and 
leads into the victorious conclusion. 

Many conductors, indeed, are still of opinion that Bach 



408 XV1I. Chamber and , Orchestral Works. 

ought to be corrected here and there; they think that 
the nuances should not be sharply defined against each 
other, but should merge into each other by a crescendo or 
diminuendo. This, of course, quite destroys the terrace- 
like plan of Bach's music to which Vianna da Motta has 
called attention. There are also conductors who try to 
get a better effect by making a final tutti dribble out to 
a pianissimo. 

The same rules hold good for the tempo here as in the 
organ and clavier works; the better the playing, the slower 
the tempo can be, -because when the hearer perceives all 
the expressive detail a quite moderate tempo has the effect 
on him of a quick one, while in the faster tempo he could 
scarcely grasp the rich polyphony. 

Are the Brandenburg concertos suitable for our concert 
halls ? No one can doubt this who has heard one of them 
say under Steinbach, and observed the effect on the 
audience. These works should become popular possessions 
in the same sense as the Beethoven symphonies are. 
Spiro finely says, in a glowing article in which he affirms 
the right of the modern public to Bach's orchestral works, 
that these concertos in reality are not concertos but sym- 
phonies*. It is to be hoped that the overtures also will 
come into their own before long. Our instrumentalists 
would profit greatly by going to school to Bach. There 
are no insuperable difficulties in the way of performance. 
It is not for the best that the flutes d bee in the fourth con- 
certo should have to be replaced by our traverse flutes, 
but the total effect really does not suffer. Viola players 
who can also play the gamba will probably be found before 
long in nearly every orchestra, so that the sixth concerto 
may some day be freed from its Babylonian captivity. 
For the second concerto the instrument-makers Alexandra 
Brothers, of Mainz, have made a small F trumpet, on which 



* Bach und seine Tfanshriptoren, in the-N><? Zeitschnfi 
Jtfusik, 1904, p. 680 fl 



Performance of the Brandenburg Concertos, 409 

it is possible for any good trumpeter, with a little practice, 
to play the original Bach part, so that for the future it 
will not be necessary to modify the part or give it to the 
clarinet.. Mendelssohn had recourse to the latter device 
when he produced the first D major overture; and it is 
retained in David's edition of the work for the Gewand- 
haus concerts. The small Quartgeige, giving the four-feet 
tone, for which Bach writes in the first concerto, must also 
find cultivators again. 

Too large an orchestra is rather a disadvantage to 
these concertos, as it destroys the natural proportion 
between the solo instruments and the tutti. We are on 
the very border-line between chamber music and orchestral 
music. The wood-wind in the tutti must of course be in- 
creased in proportion to the strings. The accompanying 
clavier must not be omitted, even when the orchestra is 
a large one. In a small room a cembalo can be employed; 
in a large one it is best to have a good table piano, or a 
small Erard grand of the old style. A modern grand is 
too dull harmonically. For the "concertising" piano, how- 
ever, we should always employ a modern grand, as here it 
must play the part of a solo voice; the harmonies should 
be given on another piano. It is most desirable also that 
the basses in the tutti be accompanied in octaves by a 
piano, as this will bring them out more clearly than ex- 
aggerated emphasis in the double basses. It will always 
be noticed that to force the tone anywhere in these works 
is to spoil the effect. 

When once the Brandenburg concertos and the over- 
tures have established themselves in the concert room, 
the question will arise as to how far some of the preludes 
to the cantatas can win the same rights of naturalisation. 
It is already settled, in principle, by the fact that Badi 
himself does not hesitate to transfer movements from his 
overtures and concertos into his cantatas. He used the 
introduction to the first Brandenburg concerto as the 
"sinfonia" for the cantata Falsche Welt, Mr trau ich nicht 



4io 



XVII. Chamber and Orchestral Works. 



(No. 52)*, and the first movement of the third concerto 
as the prelude to the cantata Ich licbe den Hochsten von 
ganzem Gemilte (No. 174) ; but as a piece for strings alone 
did not seem to him rich enough for the church, especially 
in a 'festal cantata on Whit Monday, he added in the latter 
case three obbligato oboes and two horns, without altering 
the original composition. This would be regarded as an 
astounding technical feat, had he not eclipsed it by another. 
He made the splendid chorus of the Christmas cantata, 
Unser Mund sei voll Lachens (No. no) by simply adding 
vocal parts to the allegro of the second overture in D major 
(B. G, XXXI, p. 66 fi). One would almost suppose he 
had written the overture and the cantata together; the 
allegro theme of the overture is so characteristic a musical 
representation of laughter that it seems to have been 
prompted by the text of the cantata 




A cantata prelude particularly fitted for performance 
as an independent orchestral piece is the wonderful or- 
chestral mood-picture from Am Abend aber desselbigen 
Sabbats (No. 42), in which Bach paints the silence and the 
peace in which the slowly descending twilight envelops 
the earth**. 



* It is given in the latter form as a supplement to B, G. XXXI 1 , 
p. 96 ff. The small Quartgeige is omitted. The whole of the first 
Brandenburg concerto has come down to us in a somewhat shortened 
form, in a copy of 1760, also as a "sinfonia'% consisting only of 
Allegro, Adagio, Minuet, Trio I, and Trio II. See Ddrffel's intro- 
duction to B. G. XXXJ>, p. 19. 

** Others are: the vigorous C major symphony from the Easter 
cantata Der Himmel lacht, die Erde jubilieret (No. 31); the "sonata" 
of the cantata Himmelskdnig, sei willkommen (No. 182); the intro- 
duction to the cantatas Tritt auf die Glaubensbahn (No. 152) and 



The Clavier Concertos. 411 

The arrangement of Bach's clavier and organ works for 
orchestra must be regarded as superfluous and imperfect, 
even when they are as sensitively done as Raff's arrange* 
ment of the G minor English suite. ^ 

It is misleading to speak of Bach's clavier concertos and 
violin concertos, as these works have nothing in common 
with the modern concerto, in which the role of the orchestra 
is largely that of an accompanist. With Bach it is only 
a matter of giving a specially brilliant obbligato part to 
the solo instrument. If this happens to be the cembalo 
it may also play the bass in order to employ both hands. 

That Bach thought in the first place of the obbligato 
part, and only secondarily of the instrument that was to 
play it, is evident from the fact that the majority of the 
seven clavier concertos are not primarily planned for the 
clavier. Almost all are arrangements; six probably come 
from violin concertos*. 



Gleichwie dev Regen and Schnee (No. 18); and the orchestral chorale 
fantasia on Was Gott tut, das ist wohlgetan, from the cantata Die 
Elenden sollen essen (No. 75). On a smaller scale are the splendid 
preludes to Gottes Zeit (No. 106), Nach dir, Herr, verlanget mich 
(No. 150), Ich hatte viel Bekiimmernis (No. 21) and Ich sUh mit 
einem Fuss im Grabe (No. 156). 

* The concertos will be found in B. G. XVII (1867). No. r 
(D minor) is taken from the lost violin concerto. The first two 
movements are used again in the jubilee cantata, Wir mussen 
{lurch viel Trubsal (No. 146). Bach uses the first movement, en- 
riched with three (in part) obbligato oboes, as the introduction to 
the cantata. On the second movement, the adagio, he super-im- 
poses the main chorus, an achievement that can be set by the 
side of the transformation of the D major overture into the opening 
chorus of the Christmas cantata Unset Mund ist voll Lachens 
(No. 1 10). 

No. 2 (E major). The introduction and the Siciliano of this 
concerto are used again in the cantata Gott soil attain mein Herze 
Jidben (No. 169); the finale serves as introduction to the cantata 
Ich geh' und suche mit Verlangen (No. 49). This concerto may 
have been originally conceived for the clavier, 

No, 3 (D major) is a transformation of the violin concerto in 
E major (B. G. XXI*, No. 2). 

No. 4 (A major) seems also from its facture to have been derived 
from a violin work, although it is somewhat more pianistic in form 
than the others. The original is unknown. 



412 . XVII Chamber and Orchestral Works. 

Bach needed clavier concertos when he directed the 
Telemann Society. The arrangements are often made with 
quite incredible haste and carelessness; either time was 
pressing, or he felt no interest in what he was doing. Violin 
effects to which he could easily have given a pianistic turn 
are not re-modelled at all; later on he improves them here 
and there in the score, but leaves them as they are in 
the clavier part*. The reason for this was that he himself 
played the cembalo part, and did as he pleased with the 
notes before him, making a new part out of them, 

We are under no special obligation to incorporate these 
transcriptions in our concert programmes. It is otherwise 
with the triple concerto for clavier, flute and violin in 
A minor, which we are accustomed to regard as the Bach 
clavier concerto. No audience, surely, could help being 
carried away by this work even at a first hearing. It 
would certainly occupy the first place in the repertory of 
every earnest pianist if the co-operation of the two other 
solo instruments and the whole style of the work did not 
demand more rehearsals than the majority of concert 
directors are in the habit of allowing to a piano con- 
certo, 

As is well known, this concerto has grown out of the 
clavier prelude and fugue in A minor. On comparing the 
sketch with the masterly expansion of it we seem to share 
the pride that Bach must have felt when he saw the new 

No. 5 (F minor) is an arrangement of a lost violin concerto in 
G minor, as is shown beyond question by the nature of the work. 

No. 6 (F major) is, identical with the fourth Brandenburg con- 
certo (B. G. XIX, p, 85 ft, G major), 

No. 7 (G minor) is the A minor violin concerto (B. G. XXI 1 , 
No. 2). In the last six empty staves of the G minor concerto we 
find the commencement of a second D minor concerto for piano, 
that has come down to us complete as the introduction to the can- 
tata Geist ttnd Seel& wird verwirret (No. 35). 

For further information see Rust's preface to B, G. XVII. 
* Rust gives the oldest originals of the arrangements in B. G, 
XVII, p. 273 ff. The first transcription of the ID minor concerto 
is particularly interesting, 



The Concertos for Two Claviers* 413 

work arise in all its majesty out of the old. The middle 
movement is taken from the organ sonata in D minor*. 

Of the three concertos for two claviers and orchestra, 
two the first and the third are arrangements of con- 
certos for two violins**. The original of the first, in C minor, 
no longer exists; the third, also in C minor, is identical 
with the D minor concerto for two violins***. How Bach 
could venture to transfer the two cantabile violin parts 
in the largo of this work to the cembalo, with its abrupt 
tone, must be left to himself to answer. Had he not done 
it himself, we should be protesting in his name today 
against so un-Bach-like a transcription. This is not the 
only case in which he makes it hard for his prophets to 
go forth in his name against the evil transcribers. 

The one original concerto, however No. 2, in C major 
compensates us for all our disappointed expectations in 
the two others, if we can speak of disappointment in 
connection with Bach. The fact that it was originally 
conceived for two claviers is shown at once not only by 
the rich writing for the two solo parts, (in the third section 
of the splendid fugue they are in three parts throughout,) 
but also by the subordinate position given to the orchestra. 
It is not an orchestral concerto with two soli cembali, 
but a concerto for two claviers with orchestral accom- 
paniment. Perhaps, indeed, the first movement existed 
at one time without instrumental accompaniment. Certain 



* The concerto is given in B. G. XVII, p. 223, where it is styled 
the eighth clavier concerto. On the question of its relation to the 
prelude and fugue in A minor and the middle movement of the 
third organ sonata, see ante p. 339. The fifth Brandenburg con- 
certo (D major), that is on the same lines, might be called the second 
of Bach's original concertos for clavier. For both there exist, be- 
sides the part for the solo cembalo, one for the accompanying 
cembalo, which adds the harmonies in the tutii passages. But even 
for the ordinary clavier concertos Bach made use of two claviers, 
as is shown by the fourth concerto (A major), for which he wrote 
out with his own hand $he part for the accompanying clavier, 

** B. G/XXI* (1871), 
*** B. G. XXI 1 , p, 41. 



414 XVH. Chamber and Orchestral Works. 

indications go to show that this was added later, and that 
Bach wrote it out at first not in score but in parts. Other- 
wise we cannot explain how it happens that in two places 
of this first allegro, bars 83 and 108 the orchestra 
enters with the major third, while the clavier parts main- 
tain the minor third, which grows logically out of what 
has gone before, and do not make it major until the 
following crochet. Bach would certainly have noticed this 
error had he had the clavier and orchestral parts before 
him in the score. The curious thing is that the mistake 
was not noticed in performance, and at once corrected 
in the clavier parts*. 

An accompanying piano is not necessary here, the two 
solo claviers themselves supplying the most essential 
harmonies. The cembalo accompagnato is here really the 
orchestra, consisting of a simple string quartet, which in 
reality only plays a figured bass that has a good deal of 
rhythmical interest. In a performance in a small room it can 
be quite well replaced by a third piano. An ordinarily good 
player could easily play the part direct from the orchestral 
score. We could even arrange for the two pianos all that 
is really indispensable in the orchestral accompaniment**. 

The two concertos for three claviers (B. G. XXXI 3 ) are 
constructed on the principle that underlies the original 
works for two claviers. In them also the string quartet 
retires into the background. For the most part it only 
supplies the harmonies, and aims at supporting and throw- 
ing into relief the leading part in the ensemble of the three 

* The observation is Rust's ; see his preface to B. G, XXI a , p. 8, 
The error is of course corrected in the B. G. edition. Rust's hypo- 
thesis that Bach resorted to this concerto form from the desir 
to omit the third (accompanying) clavier from the other concertos 
for two claviers, in which, however, it is a necessity cannot 
be regarded as proved. 

** This had been already observed by Forkel (p, 58). He also 
mentions that Pachelbel had written a toccata for two claviers. 
Spitta conjectures that Bach knew Coirperin's allemande for two 
pianos. No works of importance for two claviers belonging to that 
epoch seem to be known, except those of Bach, 



The Violin Concertos, 415 

claviers. The second concerto it is not agreed whether 
the original key is C major or D major is planned on 
larger lines than the first (D minor), and the orchestra 
plays a more important part in it. In the adagio there 
are even tutti passages in which the three claviers merely 
accompany the orchestra. The tonal and rhythmical 
effects that Bach has achieved with three claviers are 
indescribable. At every hearing of these works we stand 
amazed before the mystery of so incredible a power of 
invention and combination. 

An old tradition has it that Bach wrote these two con- 
certos in order to play them with his two eldest sons. If 
this be true, they must date from about 1730 1733. 

The concerto for four claviers (A minor) is based upon 
a Vivaldi concerto for four violins*. 

Of the violin concertos we possess only the half those 
left by Philipp Emmanuel; those belonging to Friedemann 
are probably lost for ever. We have three concertos for 
violin and orchestra (A minor, E major and G major)**; 
an important fragment of an allegro movement from a 
work for violin and large orchestra (D major)***; and a 
concerto for two violins with simple string orchestra 
(D minor) f. Of lost violin concertos at least three two 
for one violin and one for two violins have come 
down to us in clarier arrangementsff. 

* B. G. XLIII l , p. 7 1 ff. The Italian original is printed after 
it. From the middle of the nineteenth century it has been regarded 
as lost, though Hilgenfeld had seen it. 

** A minor and E majorj in B. G. XXI*, pp. 3 ff. and 21 ff. ; as 
clavier concertos (G minor and D major) in B. G. XVII, pp. 199 ff. 
and 8 1 fl The G major concerto figures as the fourth of the 
Brandenburgs in B. G. XIX, p. 85 ff. ; as a clavier concerto (F major) 
in B. G. XVII, p, i53ff. 

"** B. G. XXIS p . 65 ff. The end has been added by another 
hand. 

t B. G. XXI 1 , p. 41 ff. ; as a concerto for two claviers (C minor) 
in B. G. XXI*, p. 83 ff. 

ft The clavier concerto in D minor (B. G. XVII, p. 3 ff . ) cor* 
responds to a violin concerto in D minor ; that in F minor (B. G. 
XVII, p. 135 ff.) to a violin concerto in G minor; the concerto for 



416 XVJI. Chamber and Orchestral Works. 

The concertos for violin and orchestra that have survived 
are among the works of Bach to which it is useless to em- 
ploy the method of analysis; we must put them in the 
category of which Forkel briefly and eloquently observes: 
"One can never say enough of their beauty." The A minor 
and E major concertos are beginning to win a place in our 
concert halls. Modern audiences are enthralled by the 
two adagio movements, in which the violin moves about 
over a basso ostinato. We involuntarily associate them 
with the idea of Fate, The beauty of the A minor con- 
certo is severe, that of the E major full of an unconquerable 
joy of life, that sings its song of triumph in the first and 
last movements. 

The concerto for two violins, in D minor, is perhaps 
more widely known still. It can be played at home, as 
its orchestral part can be easily transcribed for the piano. 
Every amateur should know the wonderful peace of the 
largo ma non tanto in F major. 

The concerto in E major was regularly given in the 
Berlin Singakadernie even in Zelter*s time. This Bach- 
improver for such he is shewn to be by his revision of 
the parts and the marks of expression he has added 
thought it necessary to have more alternations of solo 
and tutti than Bach had indicated. Emmanuel seems to 
have .performed this concerto in Hamburgj otherwise he 
would not have had the parts copied so carefully. 

Forkel has a notable passage to the effect that Bach 
had instrumental soli played during the communion, and 
wrote most of his own for this purpose*. Of the violin 

two claviers in C minor (B. G. XXI*, p, 3 ft) points to the existence 
of a concerto for two violins in the same key* This list, however, 
must comprise only a few of the lost compositions for violin, and 
orchestra. A re-arrangement of these clavier concertos for violin 
is desirable. 

* Forkel, p. 60. The theory is only mentioned here to be 
scouted. It is a pure conjecture of Forkel's, who confesses, in his 
chapter on the instrumental works, that he knows hardly anything 
of them, and finds it more interesting to lament as lost a genus of 
Bach composition that he himself cannot imagine. 



The Musical Offering, 417 

concertos, however, he could have used only the largo 
from the concerto for two violins. 

In modern performances of the two concertos for single 
violin the orchestra is generally too large. This becomes 
unpleasantly noticeable when the basso ostinato in the 
middle movements is played in an intolerably heavy style 
by half a dozen contrabasses and twice as many cellos. 
The accompanying piano is usually omitted, without regard 
for those hearers who are conscious of gaps in the passages 
where only the violin and the bass are playing. Ysaye 
plays these concertos in captivating style, even though at 
times he modernises them too much ; but his habit of having 
the general bass performed on a harmonium is inexplicable 
either on historical, or logical, or musical grounds. 



CHAPTER XVIII. 
THE MUSICAL OFFERING AND THE ART OF FUGUE. 

The Musical Offering: B. G. XXXI* ; Peters Ed. of the Clavier 

Works, Part XII. 
The Art of Fugue: B. G. XXV*; Peters Ed. of the Clavier Works, 

Part XI. 

Bach wrote The Musical Offering on his return from 
Potsdam in 1747. He had been received by the King on 
May 7th ; on July 7th he sent him his gift. The Composition 
and the engraving of it had therefore been a matter of less 
than two months. Nor did the engraver even live in 
Leipzig. It was Schiibler, of Zella, through whom Bach had 
already published several clavier works and six organ 
chorales. 

The copy with the dedication passed from the possession 
of Princess Amalie into that of the Joachimsthaler Gym- 
nasium, and is now in the Royal Library at Berlin. Bach's 
dedication runs thus: 

Schweitzer, Bach. 27 



4i8 XVIII. The Musical Offering and the Art of Fugue. 

Allergnadigster Konig 

Ew. Majestat weylie hiermit in tieffster Unterthanigkeit ein 
Musicalisches Opfer, dessen edelster Theil von Deroselben hoher 
Hand selbst herruhrt. Mit einem enrfurchtsvollen Vergniigen 
erinnere ich mich annoch der ganz besonderen Koniglichen Gnade, 
da vor einiger Zeit, bey meiner Anwesenheit in Potsdam, Ew. Ma- 
testat selbst, ein Thema zu einer Fuge auf dem Clavier mir vor- 
zuspielen geruheten, und zugleich allergnadgist auferlegten, solches 
alsobald in Deroselben hochsten Gegenwart auszufiihren. Ew. Ma- 
jestat Befehl zu gehorsamen, war meine unterthanigste Schuldig- 
keit. Ich bemerkte aber gar bald, dass wegen Mangels nothiger 
Vorbereitung, die Ausfiihrung nicht also gerathen wollte, als es ein 
so treffliches Thema erforderte. Ich fassete demnach den Ent- 
scliluss, und machte mich sogleich anheischig, dieses recht Konig- 
liche Thema vollkommen auszuarbeiten, und sodann der Welt 
bekannt zu machen, Dieser Vorsatz ist mmmehro nach Vermogen 
bewerkstelliget worden, und er hat kerne andere als nur diese un- 
tadelhafte Absicht, den Ruhm eines Monarchen, ob gleich nur in 
einem kleinen Puncte, zu verherrlichen, dessen Grosse und Starke, 
gleich wie in alien Kriegs- und Friedens-Wissenschaften, also auch 
besonders in der Musik, jedermann bewundern und verehren muss. 
Ich erkuhne mich dieses unterthanigste Bitten hinzuzuf iigen : 
Ew. Majestat geruhen gegenwartige wenige Arbeit mit einer gna- 
digen Aufnahme zu wurdigen, und Deroselben allerhochste Konig- 
liche Gnade noch fernerweit zu gonnen 

Ew. Majestat 

allerunterthanigst gehorsamsten Knechte, 
Leipzig, den 7. Julii. 

1747, dem Verfasser. 

("Most gracious Kingi 

To Your Majesty I dedicate herewith, in deepest submissiveness, 
a Musical Offering, the noblest part of which comes from your 
own exalted hand. It is with respectful pleasure that I remember 
still the very special royal favour with which, on my recent visit 
to Potsdam, Your Majesty Yourself deigned to give me on the 
clavier a theme for a fugue, and most graciously imposed on 
me the command to develop it at once in Your Majesty's ex- 
alted presence. It was my most humble duty to obey Your Ma- 
jesty's command. I soon observed however, that owing to the lack 
of necessary preparation, the working-out was not as successful as 
so excellent a theme demanded. Therefore I resolved, and set to 
work immediately, to work out fully this truly royal theme, and 
then make it known to the world. This undertaking has now been 
accomplished to the best of my ability, and it has no other object 
than the irreproachable one of exalting, if even in only a small 
degree, the fame of a Monarch whose greatness and power in all 
the arts both of war and of peace, but especially in music, every- 



The Musical Offering. 

one must admire and honour. I make bold to add this most humble 
request, that Your Majesty will deign to honour this small work 
with Your gracious acceptance, and continue to bestow Your 
most exalted kindly favour on Your Majesty's most humble and 
obedient servant, 

The Author." 
Leipzig, 7th July 1747.) 

Along with this dedication, however, he sent only the 
first third of the work, as far as the six-part ricercare*; 
the two remaining parts he probably sent to the King by 
the hands of his son. 

Five leaves in brown leather binding with gold tooling 
form the bulk of the portion sent first. The paper is of 
uncommon fineness and strength. The dedication occupies 
two leaves; then follows the three-part ricercare aud a 
canon. Afterwards comes a separate large folio-sheet with 
the Canon perfietuus, five Canoncs diversi, and the Fuga 
canonica in Epidiapente, all upon the Thema vegium. 

The ricercare is rather a fugal three-part fantasia than 
a fugue, and contains many surprising things**. We 
ask, for instance, what is the meaning of the triplet passage 
that enters unmotived at the thirty-first bar, especially 
as it is at once abandoned, and twice again emerges only 
to disappear at once. Why did not Bach open his work 
with a larger and stricter piano fugue upon the royal 
theme? The most natural explanation is that he wished 
to keep to the improvisation that he had made before the 
monarch. The "complete working-out" of which the 
dedication speaks must consequently be taken to mean, 
with regard to the first piece, only a fundamental revision 
of the fugue actually played on the 7th May. We thus 
possess one of Bach's improvisations, written down by his 



* Ricercare was originally the term for all imitative and fugal 
pieces, even when they were very freely developed. Etymologically 
the word signifies a piece of music in which we have to "seek" 
something namely, the theme. In Bach's day it signified a fugue 
worked out with particular ingenuity. 

** It is inscribed: Regis Jussu Cantio et Reliqua Canonica Avte 
Resoluta an acrostic on the word "ricercare". 

27* 



43O XVIII. The Musical Offering and the Art of Fugue, 

own hand. That this is so is further suggested by the 
unusual freedom and animation of the fugue *. It is re- 
grettable that this splendid work, not being included 
among the clavier compositions, is almost unknown to the 
majority of players. 
The royal theme runs thus 








Frederick had desired Bach to extemporise a six-part 
fugue for him. Bach did so, not, however, on the King's 
theme but on one of his own, giving as his excuse that it 
was not every theme that lent itself to six-part treatment. 
Afterwards he made it a point of honour to work out the 
King's theme also in six parts. Thus originated the ricer- 
care, which, with two supplementary canons, constituted 
the second consignment* This is not so luxuriously got 
up as the first; it consists of four ordinary leaves, held 
together by a pin. 

The six-part ricercare is Bach's richest piece of fugal 
writing. In order to allow of a more comprehensive view 
of it, he wrote it out in score on six staves. It is, however, 
as playable on the piano as any of the fugues of the Well' 
tempered Clavichord**. 

From the technical standpoint the work is unique; but 
we seek in vain in it for the inspiration and the poetry 
that make the fugues of the Well-tempered Clavichord so 
beautiful. No matter how often we play it, it affords no 
lasting satisfaction. It is a product of Bach's last creative 
period, in which the contrapuntal technique, though not 

* Observe, for example, the almost wanton play he makes with 
the diminution of the theme, 

** In the autograph that Bach sent to the engraver, the fugue 
is compressed into two staves. In this form, which is more con- 
venient for the player, it is given as a supplement to B. G* XXXI 
p. 45 *f- 



The Canons. 



421 



actually an aim in itself, nevertheless plays the leading 
part, the invention taking a subordinate place. 

The Musical Offering ends with a sonata for flute, violin, 
and accompanying clavier, to which there is added a 
canon fierpetuus. It is in four movements, largo, allegro, 
andante, allegro. In the largo the royal theme is merely 
suggested; in the fugued allegro it is used as the cantus 
firmus; the andante harks back to motives from the three- 
part ricercare; the royal theme forms the basis of the 
finale in this form 




Bach thus wrote two sonatas for flute, violin and ac- 
companying clavier; one in the Weimar or the Cothen 
period*, the other three years before his death. The 
difference between the two works is enormous. The first 
belongs to his naive period, when he was solely intent 
on beauty of sound. When we listen to it we seem to be 
wandering by a woodland brook, over meadows that the 
morning dew has studded with diamonds. The later 
sonata transports us to great mountain heights, where 
vegetation ceases, and peaks, rising one above the other, 
stand out in sharp outlines against the blue sky. 'The 
beauty of the trio-sonata of the Musical Offering is of this 
quality. It is profound and severe, without any of the 
gracious charm that distinguishes the work of the youth- 
ful period**. 

We have a manuscript copy of the Musical Offering in 
which the figured bass of the clavier part is written out by 



* In G major: largo, vivace, adagio, presto. B. G, IX, p. 221 ft 
** The first re-issue of the Musical Offering was made by Breit- 
kopf and Hartel in 1832; Peters brought it out in 1866. 



422 XVIII. The Musical Offering and the Art of Fugue. 

Kirnberger *. This work by a pupil of Bach is invaluable 
as showing us how simply and correctly the composer 
wished the figured bass to be worked out. 

The Musical Offering contains in all ten canons, in- 
cluding the fuga canonica at the end of the first part. They 
are not canons in the ordinary sense of the word, aiming 
at a definite musical effect, but clever musical charades, 
of the kind that the musicians of that time were fond of 
propounding to each other. The solutions of the first six 
canons are given by Kirnberger in his Kunst des reinen 
S#&0s**. In two of them, the fourth and the fifth 
Bach aims at a certain musical symbolism. Over the 
fourth, in which the theme is treated in augmentation in 
contrary motion, he writes: "Notulis crescentibus crescat 
Fortuna Regis", ("May the good fortune of the King 
increase like that of the note-values")- The fifth, a circle 
canon ascending through the scale, is inscribed: "Ascen- 
denteque Modulatione ascendat Gloria Regis", ("And as 
the modulation ascends, so may it be with the glory of 
the King"). 

It was the custom to indicate the way to the solution 
of the canon by showing the notes of the theme on which 
the other parts had to enter. In the two canons that 
precede the sonata, Bach omits this hint, "Quaerendo 
invenietis" ("Seek, and ye shall find!") runs his inscrip- 
tion. The first is in two parts, the second in four. While 
the latter is 'clear enough, the former permits of several 
solutions, which were put forward by Agricola, Kirnberger, 
and the Freiburg cantor Fischer the latter in the Att~ 
gemeine musikalische Zeitung of 1806***, 

Besides these we have five other canons by Bach. One 
is given by Marpurg in his Abhandlung von der Fuge\ 
another, belonging to the year 1713, was inscribed to an 



* It is reproduced in B, G, XXXI 2, p. 52 ff. 
** Vol. II, p. 45 ff. ; reproduced in B. G. XXXI*, p, 41 ff. See 
also Spitta III, 195. 

*** B. G. XXXI 2, pp. 12 and 13 (Preface) and 49. Spitta III, 195. 



Origin of the Art of Fugue. 423 

unknown person, probably his Weimar colleague and friend 
Walther; he paid the same honour, during his Hamburg 
journey of 1727, to an amateur of that town, a "Monsieur 
Houdemann", who returned the compliment with a poem 
on Bach*; another canon, also given in Marpurg's Ab- 
handlung von de? Fuge, was inscribed, so Spitta conjectures, 
to Schmidt, the organist at Zella; the fifth canon is found 
on Hausmann's picture of Bach, belonging to St. Thomas's 
school**. 

While engaged on the Musical Offering, Bach resolved 
tp_carry out systematically a plan which he had he^Sder- 
taken somewliat ^ unsystemafically7 to writeTa complete 
wort on a single theme. The new work was to be a practical 
illustration of the art of fugue. 

Tt is an error to say he 30 not complete the Art of Fugue. 
He died before the engraving was completed; hence the 
work has come down to us in a seemingly incomplete form. 
During the last weeks of Bach's life none of the elder sons 
was with his father. After his death they went on with 
the engraving in ignorance of what his plan had been. 
The plates were prepared by Schiibler, of Zella, to whom 
Bach had also entrusted the engraving of the Musical 
Offering***. Perhaps Bach had originally intended to etch 
the work himself on copper ; three pages of the autograph, 
written in such a way that they could be reproduced di- 
rectly on the plate, point to this intention. 

How little Schiibler and the sons were acquainted with 
Bach's design is evident from the fact that they paid no 
regard to a list of errors, carefully made by Bach, that 
has fortunately come down to us. They were not even 
clear as to the arrangement of the ^ 



* See p. 185. . 

** These canons and their solutions are given in B. G, XLV 1 

(1895), pp. 131 138. See also Spitta I, 387; III, 228, 229, 265. 

*** Four copies of the original edition still exist, as well as an 

autograph of Bach's of an earlier date. The work was first issued 

afresh by Nageli, of Zurich, and afterwards by Peters, of Leipzig. 



424 XVIII. The Musical Offering and the Art of Fugue. 

inserted a simple variant as a new piece, fugue No. 14 
is identical with fugue No. 10, except that it lacks the 
first twenty-two bars. Rust thinks that the whole .style 
of the edition indicates that none of the elder sons had 
anything to do with it. That is not so; they merely 
attended to it in a hurry. 

Among Bach's papers was found also a large fugue 
upon three themes, at which he had worked until the 
last, without finishing it. Emmanuel and Friedemann 
thought it had been intended for the Art of Fugue, and 
printed it there, unfinished as it was. In order, however, 
that the work might not end in this incomplete way, they 
added the organ chorale "Wenn wir in hochsten NSten 
sind", which Bach had dictated to AltnikoL No one can 
say whether it had really been his intention to end the 
Art of Fugue with these two works. In a sense they belong 
to it, in another sense not. They have nothing to do with 
the specimen fugues, for they are not based on the same 
subject. On the other hand, they are so skilfully worked 
out, see, for example, Bach's constant manipulation 
of the theme in inversion in the organ chorale, that he 
may well have written them with a view to their forming 
an appendix to the Art of Fugue. 

The three themes of the unfinished fugue run thus : 







III. B 



A C 



|J 



The Fate of the Art of Fugue. 425 

The three separate fugues on these themes are com- 
pleted; Bach is just about to combine them at the point 
where the manuscript breaks off. 

The theme of the last fugue spells Ba(:h j s w name. In the 
Weimar days Bach had remarked to his colleague Walther 
upon the peculiarity of the four letters of his name, as 
accounting for the musical aptitudes of the Bach family. 
Walther mentions this at the end of the meagre little 
article that he devotes to his former friend in his musical 
dictionary (1732), and expressly says that the "remarque" 
came from Herr Kapellmeister Bach himself*. This makes 
it all the more curious that Bach should have waited until 
the last year of his life before making a fugue on this in- 
teresting theme, Friedemann, when questioned by Forkel 
upon this point, said positively that his father had never 
written any fugue but this upon the family name**. The 
various fugues on BACH, that claim to have been com- 
posed by Johann Sebastian, therefore cannot be his. There 
are four of these. One of them, if not like Bach, is not 
uninteresting. Spitta tries to preserve the ascription of 
at least two of these fugues. They are not printed in the 
B. G, edition, however, even among the doubtful works; 
only the themes are given***. 

The theme B A C H is a favorite one with the moderns. 
LiszF~and ScEuiann^av? w wriffen fugues upon it. ' In 
Reger's music we fancy we can often detect it even where 
it is not expressly indicated. Nor must we forget Barblan's 
accomplished organ passacaglia on B A C EL 

The Art of Fugue was published some months after Bach's 
death, at the price of four thalers. It had, however, no sale; 
then, at Emmanuel's request, Marpurg (1718 1795) wrote 
a preface to it; and the work was re-issued with a new cover 



* See also p. 186. 

** Forkel told this to Griepenkerl, by whom it was communi- 
cated to Roit&sch, from whom the tradition comes. 

*** Spitta III, 206, 207. For the themes of the four apocryphal 
fugues oa B A C H, see B. G. XLII (1893), Preface, p. 34. 



426 XVIII . The Musical Offering and the Art of Fugue. 

and the recommendation of the celebrated theoretician 
at the Leipzig Easter fair of 1752 . Its worth was recognised ; 
Mattheson praised it warmly*; but still it did not sell. 
In 1756 Emmanuel had sold barely thirty copies. The 
hundred and thirty thalers received did not cover the cost; 
and the disappointed son sold the plates of his father's 
last work for the value of the metal. Such was the fate of 
the Art of Fugue. 

In his biography Forkel says indignantly, "If a work 
of this kind, by so exceedingly famous a man as Bach, 
had appeared anywhere but in Germany, perhaps ten fine 
editions would have been taken up out of pure patriotism. 
In Germany there were not sold eyj\. enough copies to pay 
f 5I^1re ^Cdppet-plates and the engraving**." 

Perhaps Forkel's exasperation with his countrymen 
carries him too far. It was not the fault of individuals, 
but of the epoch, that the great cantor's work had no 
success. Music had struck into new paths, that led it 
away from the fugal style, and those who were still in- 
terested in it were ixot fugue masters tut fugue school- 
masters, and incompetent to understand the true Bach, 
however much they swore by him. We get this impression 
even from Marpurg's preface, which partly consists of only 
a moderately clever polemic against the new tendency 
that refuses to recognise the fugue as the vital cornerstone 
of music***. 

The^Art of Fugue consist 1 pf Jourteen fugues and four, 
canons "6n"ffie"tEeme 




The theme cannot strictly be called interesting; it is 
not a stroke of genius, but has plainly been made with 



* For his remarks upon it see p. 226. 
** P- 53- 
*** See B. G. XXV*, prefatory notes, pp. 15 and 16. 



Contents of the Art of Fugue. 427 

an eye to its manifold "workaW$&ess" and capacity for 
jjfl^sTonT'' Nevertheless it grows upon us after repeated 
hearings. It introduces us to a still and serious world, 
deserted and rigid, without colour, without light, without 
motion; it does not gladden, does not distract; yet we 
cannot break away from it. 

We get the same impression from the first four fugues, 
that deal with the theme itself and its inversion. With, 
the fifth fugue, however, the monotony of the theme is 
broken. The regular pace of the first four notes becomes 
more varied rhythmically; the theme acquires a grave 



c I r (! r i ^ r M ^[ 



EE 



Jfe, .eighth fugue onwards it becomes more and 
more animated, until, in the eleventh, it assumes the 
following form 



We do not know which to wonder at most 
that all these combinations could be devised by one mind, 
or that, in spite of the ingenuity of it all, the parts always 
flow along as naturally and freely as if the way were not 
prescribed for them by this or that purely technical ne- 
cessity. 

work being ?J 



Bach writes the fugues out in score, and calls them "counter- 
points". 

The last four fugues are grouped in pairs, each of a pair 
being note for note an exact inversion of the other, as 
if we were reading it in a mirror. They are in three parts; 
the negative stands immediately under the positive. Here 



428 XVIII. The Musical Offering and the Art of Fugue. 

again Bach soars playfully above every technical difficulty. 
The pieces are bright and animated from beginning to end, 
as if it were a pure accident that one of them happened to 
be the reflection of the other. 

Bach himself must have felt the purest pride in them. 
He arranged the last pair for two claviers, adding a fourth 
obbligato part so that both instruments should be fully 
occupied. In this form the last two pieces of the Art of 
Fugue were given as a supplement, and when the work 
was republished in the nineteenth century this part of it 
was fastened on by the pianists and soon became the most 
popular of all. The theme runs thus 




114403