Beethoven Sketchbooks in the British Museum

Author(s): Joseph Kerman

Source: Proceedings of the Royal Musical Association , 1966 - 1967, 93rd Sess. (1966 -
1967), pp. 77-96
Published by: Taylor & Francis, Ltd. on behalf of the Royal Musical Association

Stable URL: https://www.jstor.org/stable/765901

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Taylor & Francis, Ltd. and Royal Musical Association are collaborating with JSTOR to digitize,
preserve and extend access to Proceedings of the Royal Musical Association

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
 Beethoven Sketchbooks in the
British Museum

JOSEPH KERMAN

THE Department of Manuscripts at the British Museum own

an appreciable number of Beethoven letters and also of musical
manuscripts from his hand. These comprise holographs of
finished compositions, sketchbooks and sketch sheets, corrected
printer's copy, and the like. At one period, from around 187
to 1895, the Museum would appear to have been in the marke
for Beethoven material, in a small but discriminating way; and
its holdings have been augmented by occasional gifts an
bequests, from the time of Vincent Novello (1843) to E. H. W
Meyerstein (1952). By comparison with the Beethoven archives
of Berlin, Bonn, Vienna, and Paris, the Museum collection
certainly counts as modest, but it counts as large by comparison
with any other Beethoven collections of today, public o
private. And while certain of the items are slight, others are of
major importance. The great treasure is the so-called Kafk
Sketchbook, a large assembly of miscellaneous material all
from the I78os and 1790s, including not only sketches, whic
are especially rare in this early period, but also the holographs
of ten actual compositions, most of them unica.
The British Museum autographs are duly described in th
Museum catalogues and in the Beethoven literature, though
not quite completely on either side, and in any case the refer-
ences are widely scattered. For convenience, a list of the sketch
books and other sketches is appended to this paper (pp. 94-96
None of them has been studied adequately, though atten
tion was drawn to the Museum sketchbooks as long ago as
1892, in a serial article for the Musical Times by J. S. Shedlock.
Shedlock devoted his major attention to the Kafka Sketchboo
-as was reasonable enough; he said little or nothing abou
the other manuscripts, though he studied them and put his
notes at the disposal of Augustus Hughes-Hughes for th
preparation of the very impressive, very inconvenient
Catalogue of Manuscript Music in the British Museum, I9o6-9
The information supplied by Shedlock and Hughes-Hughes
1 j. S. Shedlock, 'Beethoven's Sketch Books', Musical Times, xxxiii (1892),
331, 394, 461, 523, 589, 649 et seq.
2 Information about the Kafka Sketchbook, for example, has to be asembled
from nineteen separate entries.

77

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
can be added to, and bears some correction, but it forms a
most helpful basis for the study of this material.
In the following, I shall not in fact speak of the Kafka
Sketchbook, for reasons that I hope will not seem too contra-
dictory. On the one hand, as appears from the references on
p. 94, so much of its contents has been discussed that a
general idea of its scope is not hard to come by. On the other
hand, the next round of discussion of its contents would
require more time than the present occasion affords. This
paper, then, will be confined to the least-known of the Museum
sketchbooks, Egerton MS 2795, and to what should be the
best-known one, Additional MS 31776, a sketchbook for the
Sixth Symphony and the Trios, Op. 70. This has recently been
made available in a complete transcription by Dr. Dagmar von
Busch-Weise, as the third volume in the edition of Beethoven's
sketchbooks projected by the Beethoven Archive at Bonn.3
With the help of this transcription, one can proceed at once to
a study of the compositional process for many of the move-
ments of the symphony and the trios; not all aspects of this
process can be analyzed, of course, but some aspects can be
analyzed in considerable detail. For reasons that will be
apparent, with the Egerton MS the investigation has to be
less analytical in nature than descriptive and historical.
Egerton 2795 is a small pocket sketchbook of the kind that
Beethoven carried around on his sorties into the countryside
and taverns around Vienna. Such books were sometimes made
by folding over a small number (eight in this case) of the
ordinary wide sheets, thus producing a not-too-thick booklet
of about 8 by 5 inches, which would go into the ample pockets
of those days. The original centre sheet is still continuous, and
stitch holes can be seen; and although the other sheets have
been cut for binding, they appear to have been kept in the

SBeethoven. Ein Skizzenbuch zur Pastoralsymphonie Op. 68 und zu den Trios

Op. 70, I und 2. Vollstandige, mit einer Einleitung und Anmerkungen versehene
Ausgabe, ed. Dagmar Weise (Ver6ffentlichungen des Beethovenhauses in Bonn.
Neue Folge, general editor, Joseph Schmidt-G6rg. Erste Reihe. Skizzen und
Entwurfe: Erste kritische Ausgabe [III]), Bonn, 2 vols., 1961. I say 'should
be the best known', but in fact the lack of interest in this publication has
been extraordinary, both in and out of England.
It is true that certain features of the publication have the effect of
militating against easy use. These features are discussed with great care
and understanding by Lewis Lockwood in The Musical Quarterly, liii
(1967), 128-136, and I should certainly like to associate myself with his
views and suggestions. Professor Lockwood's review is, I believe, the first
in English to call attention to the great importance of this sketchbook
edition.

78

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
right order4 and there is no reason to think that any pages
have been lost. The pages look very much like those illustrated
in facsimile from a pocket sketchbook of I819-20, published
by Professor Schmidt-G6rg as the first volume of the Beethoven
Archive sketchbook edition.
Egerton dates from the summer of 1825 and transmits
studies for the Quartet in B flat, Op. 130. The sketches, which
tend to be short, are done in pencil, which makes them dismally
hard to read.5 Serious work was carried on at home in large
books marked with ink, and in fact the large book used simul-
taneously with Egerton can be identified with certainty. Before
coming to this point, however, it will be necessary to review
something of the history of the B flat Quartet, and also that of
the immediately preceding composition, the Quartet in A
minor, Op. 132.
As it happens, important parts of this history were unknown
to Nottebohm, whose comprehensive survey of the sketches
made in the years around I870 is still the best available. But
three relevant sketchbooks escaped Nottebohm's dragnet. The
most impressive of them, a full-sized book of 80 pages, was
described in 1905 by its then owner, the Spanish critic Cecilio
de Roda.* In round figures, about 35 pages at the beginning
of this book are devoted to the later movements of the A minor
Quartet, from the slow movement on. Then there are about
5 pages each to the first two movements of the B flat Quartet
(Adagio ma non troppo-Allegro and Presto, B flat minor), 15
pages to each of the slow movements (Andante con moto, ma non
troppo, D flat major, and the Cavatina, E flat), and a few pages
at the end to the fugal finale. None of these various movements
arrives at its final state. In particular, the first movement of
the B flat Quartet, in the course of the comparatively small
number of sketches preserved here, remains in a primitive
stage. There is no trace yet of the second subject, there is an
unfamiliar sextuplet-semiquaver figure, and-strangely!-the
Allegro is conceived in 3/4 time rather than in the 4/4 of the
final version. On the other hand, both slow movements work
their way slowly, but distinctly, from very remote beginnings
towards their final versions.
It is known that the first two movements of the A minor

4 In spite of what might be suspected from an irregular alternate foliation

on the bottom right-hand corners of the sheets. Several sketches run from
verso to recto across a page-opening.
sA few unidentified notes in ink on ff. Iov-i ir.
6 Cecilio de Roda, 'Un Quaderno di Autografi di Beethoven del 1825',
Rivista Musicale Italiana, xii (1905), 63-108, 592-622, 734-767-

79

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
Quartet were written before Beethoven's serious illness in April
and May 1825, that the others-from the Heiliger Dankgesang
einer Genesenen an die Gottheit, in den lydischen Tonart on-were
written later, and that the quartet as a whole was finished by
around the end of July.' The de Roda sketchbook would
therefore appear to have been used between May and July or
August. Work proceeded simultaneously on the B flat Quartet,
but in May this was still in the beginning stages. As early as
in March, Beethoven had written to Charles Neate in London
that the B flat Quartet was nearly ready, but this must be
taken (as doubtless Neate took it) with a large grain of salt.
A number of pocket sketchbooks have been preserved from
the time of the de Roda book. One of these, at the Glinka
Museum, Moscow, was described in several articles by M.
Ivanov-Boretsky at the time of the Beethoven centennial. This
book consists of two parts, one devoted to the three last move-
ments of the A minor Quartet, and the other to the first move-
ment (only) of the B flat. The B flat movement, as Boretsky
showed, is a little further advanced than in de Roda. Although
work is apparently still restricted to the Adagio introduction
and to the exposition, the second subject is now present, the
sextuplet figure has been dropped, and sketches appear in both
3/4 and 4/4 time.
A second pocket sketchbook is Egerton. By the time this was
in use, the A minor Quartet was already written. This can be
inferred from the fact that the top of an inner page contains
four brief references to its first and last movements-not
sketches, but notations of the sort that would be made in
touching up a holograph :,a

7 See Kinsky-Halm and Thayer-Forbes, p. 957-

a Beethoven-Zentenarfeier, Internationaler Musik-historischer Kongress, Referate,
Vienna, 1927, pp. 88-90o; Melos, vii (1928), 407-414; Muzykalnoye
Obrazovanie (Musikalische Bildung), Moscow, Nos. 1/2 (1927), 9-91. These
publications are very scarce, and I am greatly indebted to Professor
Gerald Abraham for lending me his copy of the last one--which includes
a complete transcription of the sketchbook with Boretsky's attempted
identification of all the material, line by line.
' The words 'Allegro appassionato' written on f. i Ir presumably also refer
to the Quartet in A minor.
At bar x8 of the first movement, the referent for two of these notations,
the holograph of Op. 132 was indeed erased and altered. I was able to
consult the Bonn Beethoven Archive photograph of this holograph-the
MS itself has disappeared-as well as other material, referred to in notes
15 and 16, thanks to the courtesy of the Director, Professor Joseph
Schmidt-G6rg, and Dr. Hans Schmidt.
On the transcriptions, see the note on p. 81.
8o

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
 *Ex. 1

., ,t
f. 12 . jd15I-lot
.,
A~ d.,
F I d _

far:

1000 c dm dk .ban
i4. b 11 .31 S.3-
310-319.3S9.3"

(The British Museum sketchbook for the Sixth Symphony also

includes a page of notations of this sort, referring to the Cell
Sonata in A, Op. 69.) One is bound to consider the possibil
that these notations were entered later than the time of the
main sketching, even though the hand-writing looks no
different. However, Beethoven can hardly have used the little
Egerton book for more than a few weeks; that he would have
left a single recto page blank within it, and then gone back to
a book of this kind, seems altogether unlikely.
The actual sketches in Egerton refer to the first movement
of the B flat Quartet, to the following Presto, and to the two
slow movements, the Andante and the Cavatina.1O The book
opens with short, scattered studies for all the sections of the first
movement-the Adagio, the Allegro exposition, development,
recapitulation, coda. These are now approaching their final
form. Since the writing is uncommonly hard to read, even for
Beethoven, the following sample transcriptions are conjectural
at many points:

* Note on the Transcriptions. Material not in the source is shown by means

of square brackets or dotted lines. Sometimes Beethoven appears to have
paid little attention to specifying the exact pitches that he must have meant;
when an interpretation has been made in such cases, the literal MS reading
is given above the note, as a letter. Readings that are seriously in doubt are
indicated by question marks, but the attempt has not been made to list all
the arguable alternative readings.

10 First movement: Ir--3v (except 2r/I-2), 5r/6-12, 5V/3-12, 8r/I-2, 15r/

I I-12.

Second movement: 4v/I-4, 8r/8-9v/5, 9r/4-12.

Third movement: 5v/I-2, 5V/5, and perhaps 4v/4-I r
7v/3-8.
Fifth movement: 7v/9-12, 8r/x2, gv-i-or/4, xov/8-12, I Iv/3-9, 12v/7-
13v, 14r/IIx-16v (except 15v/I1-12).
81

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
Ex. 2

r.3'p.I@

cf.- bu frbs". . ' . 1

f.27-9

AlS
I P.P.. ... ..pF FIPC

I cf..
i6I ,,,,, 1,,,
I I .- .t
"i W :;L!, I?- ; I. I L
W_.!krc q.ri
4L hibII)

[I~b IllrlYcf. ben 123 -1314

i-q ,, cf I 136-14.
V110cf. b s5r r 161

. ./ l _;l, JL.. ..J . .. .. ..J

- ,.i. ii --a .~
'" . "'4 "-t_ " F" ." - ;-
PL mn

-: ... - ,L+ _ _,

The few sketche

the other hand,
anywhere as far
the Egerton sket
of the final scor
the earlier of th
on folios 4v-5r
page 48 (de Rod
barcarola'). Egert
danza tedesca or t
Now, since the t
82

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
the time-span of de Roda, it allows us to pinpoint certain
details of Entstehungsgeschichte that are necessarily obscure on
the evidence of the larger book alone. This is, perhaps,
Egerton's chief interest. First of all, Egerton narrows down the
period in which serious work on the B flat Quartet can be seen
to have been begun. Nottebohm could only tell that work on
the quartet was proceeding 'at the earliest in March 1825' ;1
he knew the letter to Neate and he knew later sketchbooks for
the quartet but he did not know de Roda. The de Roda
sketchbook shows that little thought had been given to the
piece before May. Kinsky-Halm errs on the other side, or
perhaps it is on the side of caution, by stating merely that the
quartet 'was worked out (ausgearbeitet) in August ... right
after the completion of the A minor Quartet'. But by August
the first and probably the second movements were already
close to completion. Secondly, Egerton provides evidence as
to the relative time of composition of certain of the movements.
Nottebohm observed that work on the four last movements
proceeded simultaneously, or at least overlapped in time. The
de Roda book, with its sketches for the first movement that
are less advanced than those for the Andante and the Cavatina,
tempts one to extend Nottebohm's concept of overlapping
composition to all six of the movements. However, Egerton
fixes a point in time at which the first two movements were
nearly finished and the later ones were scarcely begun. It is
astonishing to see that at this stage Beethoven had not even
decided on the keys. The D flat Andante is sketched in G,
the E flat Cavatina in D and D flat.
Much has been said about the 'organic unity' of all six
movements of the Quartet in B flat, and especially about the
role of the great fugal finale in this respect. If such unity exists,
it was something that Beethoven groped his way towards
during the process of composition. We cannot say, with
Romain Rolland and others, that the fugue 'must' have been
central to his conception from the start. There is a celebrated
piece of misleading evidence here: the fugue subject has been
spotted next to sketches for the first movement of the A minor
Quartet, back at the beginning of the year."12 Whether the
subject was actually written at that time, or entered into the
sketchbook later, is not established. But certainly the fugal
finale was not developed at that time, or at any other time early
in the composition of the quartet. Instead numerous other
x1 Beethoveniana, ii, 2.
12 Ibid., 550-551.

83

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
possibilities were tried for a concluding movement: de Roda
records six, the Moscow sketchbook has at least one, and there
are two more in Egerton (labelled 'finale', f. I2v, and 'Fin',
f. 7r). None of them has any suggestions of fugal treatment or
any signs of a certain four-note configuration. In July, as we
have seen, when the first two movements were nearly ready,
Beethoven still had only the vaguest notion as to how the rest
of the quartet would look. Yet a few weeks later, on 29 August,
he wrote to both Karl Holz and his nephew that the piece
would be done within a space of days-all six movements of
it, including the 5oo-bar fugue! By then, obviously, the con-
ception was quite clear, even if the estimate was sanguine
(once again: even though it was made to intimates, rather
than to a prospective customer such as Neate). The piece was
completed by November or possibly somewhat earlier. 13 By the
end of August, when the letters were written, both the Egerton
and de Roda sketchbooks had doubtless been filled and put
aside.
For the rest, Egerton contains the draft of a melody-
evidently destined for a canon-on a text from one of Beeth-
oven's favourite books:

F. Ior:
Alle gewaltsame That misfiillt ja den [seligen] G6ttern;
Tugend ehren sie nur und Gerechtigkeit unter den Menschen.
These lines are among many specially marked in Beethoven's
copy of the Odyssey (XIV, 83-84).1" The sketchbook also
contains a disquieting amount of unidentified material, some
of which might well be traceable by means of a thorough
comparison with other sketchbooks. One interesting sketch
Hughes-Hughes suggested might refer to the Quartet in C
sharp minor, Op. 131, composed in late 1825 and I826:b
L3 ([k

odw .dio do mo i.bnucb na de duaur deo dr

r~rt
~do7--7f~P
4.7W
Irl

--9

M q A .6s O

Aft , *Wr ,es ,, ,/, PI& .a/ AVb ; ,,

84

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
This could indeed be an early notion for Op. 13 --what else
could it be ?-but we can scarcely speak with certainty about
a sketch that so little resembles the final work or any studies
that have been definitely related to it.15
However this may be, the C sharp minor sketch was of
more than passing interest to the composer, who copied it out
almost verbatim onto larger worksheets, just as he had done
with the G minor barcarole. It appears after another C sharp
minor idea on one of a set of sketch sheets for the Quartet in
B flat at the Library of the Gesellschaft der Musikfreunde,
Vienna. * These sheets are otherwise devoted almost entirely
to late drafts, in score, for the first movement and the Andante.
Those for the first movement are presumably approximately
contemporary with Egerton. Those for the Andante, since they
are on separate sheets, may be presumed to be later.

To turn from Egerton to the Sixth Symphony sketchbook,

Add. 31776, is to move into a very different situation. The
difference is first and most obviously that between workbooks
of two kinds-between a big book kept for steady work at
home and a small pad used to record ideas on the run. But it
is also the difference between the assurance of Beethoven's
compositional impetus in I8o8 and the relatively uncertain,
experimental course that he was following in 1825. The earlier
book includes few unidentified sketches, or serious diversions,
or even temporary wrong turnings; it includes little, in fact,

'3 All that is really known about this comes from a conversation-book entry
transcribed by Thayer and dated by him in November. Beethoven wrote
'Title for the Quartet'; a strange hand ('des Neffens?', asks Kinsky) wrote
out a suitable formal inscription; and then Beethoven initialled it. From
this Kinsky-Halm infers that the piece was 'ended' in November. Thayer-
Forbes infers that it was 'not completed before November'. Thayer-
Riemann infers that in November the piece was ready to be sent to the
copyist-or had already been sent.
14 See Ludwig Nohl, Beethovens Brevier, Leipzig, 1870, p. 25. On pp. 19-20,
footnote, Nohl draws attention to another project to write a canon on
words from the Odyssey.
11 See Joachim von Hecker, Untersuchungen an den Skizzen zum Streichquartet
cis-moll op. 131 von Beethoven, unpub. diss., Freiburg, 1956. He is inclined
to leave the Egerton sketch out of consideration for Op. 131. An origin
for the sketch may perhaps be traced early in the year, at the top of a
sheet which also has sketches for the finale of the Quartet in E flat,
Op. 127: see the facsimile opposite p. 0o8 of the Sotheby auction catalogue
for i x June, 1963.
16 Unnumbered, unpaginated. The only printed description of these sheets
is apparently Mandyczewski's laconic 'Skizzen zum Bdur-Quartett
op. 1 30. 34 Bl.' in the Zusatz-Band zur Geschichte der K. K. Gesellschaft der
Musikfreunde in Wien: Sammlungen und Statuten, Vienna, 19I 2, p. 88.

85

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
besides dense, almost monolithic studies for a small number of
movements. The Andante. of the Sixth Symphony is sketched
on as many as 36 pages of large, sixteen-stave paper, including
about 20 large drafts covering sections of the size of a complete
exposition, a development plus a recapitulation, etc. This does
not take account of the 31 sheets that are known to have been
cut out of the book.17 And there must have been more sketches
yet, for the book preserves neither the very earliest stages of
the evolution of the Andante nor the very latest ones.
Sketch material of this extent seems to cry out for investiga-
tion in terms of the mapping-out of the large proportions of
the work. While there is much to be learned from an evolution-
ary study of details, such as thematic shapes, it is obvious
that a serious investigation of Beethoven's compositional
process must also take account of the major structural outlines,
large goals, balances between sections, and so on. The classic
study on this level is Nottebohm's monograph on the 'Eroica'
sketches; Nottebohm's comprehensive knowledge of the
sketches was matched by his flexibility and imagination in
dealing with them from every point of view. All other studies, by
Nottebohm and later scholars, have paid more attention to
details than to large-scale structural considerations.
Nottebohm outlined some of the difficulties involved in
trying to manipulate a large set of sketches with a view to
tracing the evolution of Beethoven's musical ideas. First of all,
and most troublesome, the chronology is very uncertain. Since
he was always working on several movements at once, Beet-
hoven would leave blank pages to be filled on another occa-
sion,18 and he would often go back to old sketches and correct
or extend them at some later time. Secondly, even if the
chronology could be worked out with complete certainty, the
sketches do not fall into a 'logical' sequence that makes for
happy analysis. Beethoven would repeatedly abandon one
perfectly good idea in order to try out others--only to back-
track, later, to the old idea that was safely recorded somewhere
in the sketchbook. For him, sketching was as much trial and
error-akin, perhaps, to improvization-as methodical, step-
by-step refining. From a third difficulty, that of deciphering a
scrawl-cum-shorthand intended for no one's eyes but Beeth-
oven's, we are in this case saved by the formidable expertise
17 All but one have been preserved, and have been identified by Josef
Braunstein, Beethovens Leonore-Ouvertdren, Leipzig, 1927, PP- 38-39, and
more completely by Weise, Ein Skizzenbuch zur Pastoralsymphonie ...,
pp. 1o and 12.
Is See footnote 2o.

86

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
of Dr. von Busch-Weise. However, there remain constant
questions of interpretation, of deciding what follows what and
how-sometimes even within a single line. Examination of the
manuscript itself does not always provide answers.
With these difficulties stated by way of caution, I wish now
to examine one segment of Beethoven's work on the Andante
of the Sixth Symphony, that devoted to the development
section. It is perhaps unnecessary to add a further caution
with respect to the personal equation. One studies the sketches
to learn something about Beethoven's music, but what one
sees in them is bound to be coloured by one's own preconcep-
tions.
For the purposes of the following discussion, let us briefly
recall the main features of the development section as we now
know it.c The exposition is quitted in a 4-bar passage modulat-
ing from F to G major. Then the development proper centres
on three roughly parallel episodes starting in G major, E flat
major, and G flat major. For each of these key-areas, the basic
material comes from the first theme of the exposition, which is
less 'developed' than decorated in a colouristic way. Thus in
the G major and E flat major episodes, four bars of first-theme
material run into six bars of semiquaver figuration for solo
flute, oboe, and clarinet, culminating in long trills. These trills
are resolved by an important canonic motif-a sort of rota-
which had served in the exposition to make the main cadences,
but which now serves to make the modulations. The G flat
major episode is shorter, moving rather firmly through C flat
(B) to a 5-bar retransition passage over a pedal F.
As might be supposed, Beethoven did not start sketching the
development section in any detail until he had done a good
deal of work on the exposition. But characteristic development
sonorities seem to have been firmly in his ear from the start.
The first page-opening that is given over to miscellaneous jot-
tings for the Andante includes some not-unfamiliar-looking
semiquaver figures and trills for 'clarinetti . . . solo' in E flat
and 'flaut' in G. (The sketches for the Pastoral Symphony in
general, Paul Mies has observed, are exceptionally rich in
indications of instruments, dynamics, and phrasing.)"9 Also on
this page is a single 7-note upward semiquaver arpeggio, which
turns up in the symphony as a colourful decoration to the main
theme, first in the G major and E flat major development
episodes and then in the recapitulation. According to
Schindler's famous story, Beethoven told him that this arpeggio
19 Die Musikforschung, xvii (1964), 333-

87

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
was meant to imitate the song of the yellow-hammer. Tovey,
who called this the 'giraffe-throated yellow-hammer', might
have enjoyed seeing the fabulous monster being hatched in G
major-but in the bass clef (f. 7v, line 3)-
Another aspect of the development section concerned
Beethoven from the start, namely the initial modulation out
of the exposition. Besides sketches leading from F to G, as in
the final version (f. I2r/2, 9, I I), he also tried sketches leading
in various other directions (ostensibly to D flat major, I Ir/5,
I Iv/ io; D major, Ior/I-2; D minor, I5r/I I1-12; E flat major,
I2V/I2). This matter was nearly, but not quite, settled in his
mind by the time he came to plan the development section as
a whole. The first of the large sketches (designated as Sketch A)
occurs relatively early within the portion of the sketchbook
devoted to the Andante. On successive folios towards the end
of the portion, seven further large sketches occur (Sketches B
to H).O2
The first attempt, while certainly primitive in detail, already
transmits certain features of the completed symphony-which
is not to say that these features hold fast throughout all the
sketching. Thus Sketch A has the initial modulating passage
proceeding (in 3 bars) from F to G. However, it seems to be
G minor rather than G major. In this key, semiquaver figures
ensue, after which the mode does change from minor to major.
Whereupon the music moves directly to the pedal F, and to
material foreshadowing the retransition of the final version.

2s Sketch A: 13r/Io--II, 13v/9-13.

Sketch B is a composite: g9v/I-6 repeats and expands upon I9r/I1I-r4.
Sketch C: 20o/8-13, 2Tr/I.
Sketch D: 21v/2-8.
Sketch E: 23Av/I-12.
Sketch F: 23Av/x3-x6, 24r/4-8, 25r/9.
Sketch G: 22v/5-12.
Sketch H: 23r/4-8.
It must be concluded that Sketches G and H were written after E and
F, even though they come earlier in the sketchbook. This is indicated by
the relative state of the recapitulation and coda as well as the develop-
ment.

'23A' designates a sheet that was cut out of the book and that
forms pages 151-152 of the sketch miscellany Landsberg io T 78
footnote 17). The sheet appears to belong between folios 23 and 24,
sketches read continuously (with the aid of signs) from the end of 23
x6 to the beginning ot 24r/5, and from the end of 23Av/I2 to the be
ning of 24r/I x. The sign at the beginning of 24r/I I ( x) has been mi
in the published transcription as-P
I am grateful to the Stiftung Preussischer Kulturbesitz, Depo
Staatsbibliothek, Tiibingen, Dr. Wilhelm Virneisel, Director,
microfilm of Landsberg io and for permission to print a transcri
of page 152 made from the film (Appendix, p. 94).
88

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
 It would appear, then, that at the early stages Beethoven
conceived a very simple one-key development section. The
key, G, remained as the initial key in the final version of the
development; this much was clear enough from the start, as
was also a general concept of the sonorities. However, neither
Theme I nor the canonic motif from the exposition figures in
the original conception. On f. 17v/Io-I I and 12 there are two
small sketches refining the initial modulating passage, but they
give no indication that Theme i is to follow. And Sketch B, a
number of pages further on in the sketchbook, merely elaborates
Sketch A without expanding its basic plan. The modulating
passage still leads not to Theme i but immediately to semi-
quaver figuration, though now this comes at once in the major
mode and is beginning to approach its final thematic aspect-
it includes Schindler's yellow-hammer, the characteristic trill,
and the canonic motif. The retransition still follows directly
after the G major section, though it is longer than before.
Even as a beginning stage, this seems suspiciously simple-
minded for a large-scale Beethoven sonata movement. He must
have had something else in mind, as is indeed suggested by the
recapitulations which, significantly, are appended to both
Sketches A and B. The recapitulation of Sketch A includes an
emphatic digression to the subdominant at the bridge passage.
What happens to the recapitulation of Sketch B, after a few
bars, is unfortunately hard to fathom: either Beethoven tried
a cryptic new draft for the development, or else he tried digres-
sions within the recapitulation extending to an impressive array
of flat keys.21 There are other recapitulation sketches involving
such modulations." On this evidence, it seems clear that the
simple one-key development section was to have been com-
pensated by striking modulatory digressions within the
recapitulation.
This makes a certain sense. However, it did not satisfy
Beethoven for long-perhaps because a recapitulation with
such digressions did not strike him as 'pastoral' enough. His
next efforts were directed to expanding the development itself
so as to provide further modulations after G major and further
episodes in other keys.
Sketch C takes a new tack. The initial modulation from F to
G is abandoned, temporarily, in favour of a grinding modula-
tion through an upward chromatic scale landing in B major.
Instead of solo figuration, trill, and canonic motif, four bars
21 F. 19/9-I15
22 Ff. 23r (Sketch H) and 27r.

89

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
of Theme I appear at this point. Another upward chromatic
modulation arrives at D flat major. Theme I appears again.
Another modulation prepares the retransition.
Since this was obviously terrible, in Sketch D Beethoven
went back to expand more directly upon Sketch B. Through
the G major episode the same ground is covered, with the
interesting difference that the canonic motif now occurs not
only after the figurative passage, but also before it. It looks as
though he felt a need for something thematic immediately
after the modulation out of the exposition, but hesitated to
bring Theme i. After the G major episode, episodes following
E flat major and in E flat minor leading to G flat. (Or seem
to follow: again there is something unclear about the sequence
of the sketch.) A repeated-note motif is introduced:

FjL 4

v Wdia

This motif plays a curious role in the final version of the

Andante, in that after contributing two bars to the bridge
passage of the exposition, it does not recur in the recapitulation
or anywhere else in the piece. We will not be surprised to
observe that in the course of certain planning stages, Beethoven
considered using it more widely.
Sketch E (see p. 94) may be regarded as a coarse
bodying-out of ideas from Sketch D. Evidently Beethoven now
wanted the three key-areas G, E flat, and B to stand out very
boldly, and so he furnished each of them somewhat mechani-
cally with material from Theme i. He hardly gave a thought
to the transitions-the facing recto page, however, bears the
following notation for the ultimate modulation:
Ex. 5

Theme I seems to have had the effect of crowding out both

the characteristic trill and the repeated-note motif; but
Beethoven reintroduced them during the next sketch, Sketch F,
entered directly below Sketch E (with an oder) and not far
removed from it in general plan. The development was getting
very long. Sketch G, which is again similar, is shorter. Here
Beethoven was sufficiently attracted by the repeated-note
90

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
motif to employ it as the means of making the modulation out
of the exposition.
Sketches D, E, F, and G all progress towards the key-plan
of the final development section, and also towards its domina-
tion by first-theme material-something that was no part of
Beethoven's original conception. Otherwise they are frequently
puzzling in detail, especially as to the transitions. With the
benefit of hindsight, we may be inclined to picture the com-
poser working with the right ingredients, over and over again,
without being able to find just the right recipe. Perhaps some
impatience on his own part is reflected in the next attempt,
Sketch H.
This is instructive from several points of view. For once, its
chronology can be spoken about with some confidence: it
must have followed Sketch G directly, at a single sitting, for
the two appear on facing pages in a strange spidery script
indicative of an unusually fine pen. And Sketch H backtracks
dramatically and clearly-all the way back to Sketch B with
its plan of a one-key development section and compensating
digressions within the recapitulation. The recapitulation now
moves at leisure through E flat, G flat, and C flat; the
development itself, on the other hand, moves from G major to
the F pedal, as in the earliest large sketches, A and B. At the
same time, there is much refinement of detail. The initial
modulating passage drops the repeated-note motif in favour of
the exact harmonic and motivic scheme of the final version.
The flute figuration begins to sound quite familiar. The trill
lasts for a full semibreve instead of a minim, as in some of the
earlier sketches.
Sketch H appears to be the latest preserved for the develop-
ment section. Beethoven still had to make up his mind that the
one-key development section was an impossibility. Once this
decision was made, however, there was perhaps not much more
work to do, for almost all the elements of the final development
section had been worked out on one page or another of the
sketchbook. The same can be said for the relevant section of
the recapitulation: Sketch F runs into a recapitulation with
a bridge section closely approaching the final version in its
extreme conciseness and its inclination in the dominant direc-

tion. (No trace yet, however, of the admirable bass G in bar 97-)
This is of course the opposite idea from the recapitulation with
flat-key digressions of Sketch H.
In summary: eight large sketches taken together with some
smaller ones allow us to trace Beethoven's work on the large
91

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
proportions of the development section in considerable detail.
This was one of those situations in which themes were less
important to him initially than other aspects of the composi-
tion. The first clues were essentially sonorous, not thematic.
He marked down figurative and instrumental ideas which are
close enough to the final version, and he arrived at once at the
key of G which likewise figures in the completed movement.
In the early stages, indeed, the development section was to
have been held to a single G major plateau, and the simplicity
of this scheme was to have been compensated by modulatory
digressions within the recapitulation. It was only when
Beethoven revised this scheme to a more conventional one-a
development section with several episodes in several keys and
a straightforward recapitulation-that the thematic substance
of the final version began to emerge. He first settled on the
canonic motif from the exposition; then he settled on the main
theme; then he tried (and ultimately rejected) the repeated-
note bridge motif. The sequence of keys for the development
episodes was arrived at fairly soon, though not without some
detours. One surprising detour, near the end of the sketching,
involved shunting the flat keys to the recapitulation and re-
instating the simple one-key development.
Analyses of this kind can be made with regard to other
sections of the movement-and other movements, and other
works-with greater or lesser success, depending on the amount
and nature of the sketch material that happens to have been
preserved. To speak only of the Andante, we have already had
occasion to draw on information about the recapitulation
sketches in analyzing those for the development section. We
need to know about the evolution of the exposition and the
coda too. To give one obvious example, the thematic material
used to make the modulation out of the exposition varies in
the sketches, even though the key sequence is generally the
same. To follow Beethoven's thought here, we really need to
know where the cadence theme of the exposition stood at the
time of each of the development sketches.
The difficulty of assembling all the necessary material in
chronological sequence should not be minimized. Even with
the development sketches discussed above, in which the general
outline of the chronology seems safe enough, the details of
backing and filling may very well have been carried out in a
more complex way than has been suggested. To establish
chronologies for the sketches of other sections of the movement,
and then to co-ordinate them all, presents problems of a much
92

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
higher order; uncertainties will always remain. Nevertheless,
the effort seems to me to be worth making. Much tact is
needed in the interpretation of sketch material, but studies of
this sort allow us to come closer to an understanding of the
compositional process than any other branches of musical
scholarship that come easily to mind. And while an under-
standing of the compositional process is not equivalent to
insight into the work of art as such, it is, once again, a closer
route to such insight than is provided by most of our other
scholarly activities.

During the lecture, sketches were played on the piano at a and b, and at e
the development section of the Andante of the Sixth Symphony was played
on the gramophone.

93

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
 APPENDIX
Sketch E

AL
1 -1 t: O- 4 - ? ?4L
S..;.

p. -- -I 2.A I P " o.

_. + L ___ _ .. ........ ...-, M-W , ---+- --_ :_

cord

O- 'urt
--d
? dI
IX3 1?
adcr Gm--cy----mQ"--

,:- ._ -x-;. ---- - ~ -.=-.--- ----- __

Sketch F____ ___ .C _. [me 24R1.1.

J. [c~[see 24RM

A CHECK-LIST OF BEETHOVEN SKETCHES

IN THE BRITISH MUSEUM

Hess = Willy Hess, Verzeichnis der nicht in der Gesamtausgabe verdff

Werke Ludwig van Beethovens, Leipzig, 1957-
N i, ii = Nottebohm, Beethoveniana, Leipzig, I872, 1887.
i. Add. 2980', if. 39-162. Autograph Miscellany fr
I780s and I790s. Purchased from J. j. Kafka, r875; the so-
Kafka Sketchbook.
94

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
 Includes complete or fragmentary holographs of the following composi-
tions: Op. 71, WoO 32, 43/1, 52/2 (see Beethoven. Supplemente zur Gesamtaus-
gabe, ed. Willy Hess, V [1962], p. 85), WoO Io9, 117, I 19, Hess 13, 48, 64
(see A. E. F. Dickinson, 'Beethoven's Early Fugal Style', Musical Times,
xcvi [I955], 75-79).
On the sketches and drafts, see N i, I, 51, N ii, 21-59, 64-73, 228-229,
356-363, 508-512, 515-516, 536-538, 561-567, 574-575; J. S. Shedlock,
'Beethoven's Sketch Books', Musical Times, xxxiii (1892), 33', 394, 461, 523,
589, 649 et seq.; Hess 79, 298; Hans Boettcher, Beethoven als Liederkomponist,
Augsburg, 1928, appendix; Joseph Schmidt-G6rg, 'Ein neuer Fund in den
SkizzenbUichern Beethovens: die Lamentationen des Propheten Jeremias',
Beethoven-Jahrbuch, iii (1957-58), 107- I l.

2. Add. 31766. Sketchbook for the Sixth Symphony and the

Trios, Op. 70. 59 ff- x6 staves. I8o8. Purchased from Julian
Marshall, r88o-81.
Described and. transcribed in Ein Skizzenbuch zur Pastoralsymphonie ...,
ed. Dagmar Weise, Bonn, 1961 (see footnote 3, above).

3. Add. 29801, if. 2-37. I6 staves. Sketchbook for Die Ruinen

von Athens and Kinig Stephan. 18 11. Bound and purchased with
No. i, above.
See N I, 39, N ii, I4, I38.

4. Egerton 2795. Pocket Sketchbook for the Quartet in B flat,

Op. 130. i6 ff. 12 staves. 1825. Purchased from Liepmannssohn,
I895.
See above.

5. Add. 29997. Autograph Miscellany from 1799-1826, w

much material from i815-i7. 40 ff. Purchased from J. N. Ka
1876.
Hughes-Hughes' identification of the works sketched can be extended to
the following: Op. 18/4(?), 58 (cadenzas), 73, 98 (see N ii, 334), 102/2(?),
107/10, 108/24, 117, I23, 13I, WoO 75, I49, I56/6, I68(?), I87, and un-
finished works discussed in N ii, 345, 350, 577- Also included are copies of
Mozart's 'additional accompaniments' to Messiah, a rejected sheet from the
holograph of Op. 132, contrapuntal exercises, and unidentified matter.

6. Add. 47852, ff. 13-14. io staves. Draft of the first 28 bars

of Der Liebende, WoO 138. 1809. Bequest of E. H. 1". ?AIeyerstein,
1952.
Bound with this, as f. 2, is a two-stave snippet from the top of a sketch
sheet containing a few notations. Meyerstein connected these with the
finale of Op. 59/1.

7. Add. 14396, f. 30. 20 staves. Sketches for the Sonata in

B flat, Op. io6, first and second movements. I818. Gift of
Vincent Novello, 1843.

95

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms
8. Egerton 2327. Drafts of the themes for twelve of the Folksong
Variations for Piano and Flute, Op. I05 and 107 (Op. I05/1-2,
4-6, 107/1-2, 4-5, 8-io), plus brief sketches for certain varia-
tions. In all, 6 ff. i6 staves. 1817-18. Purchased from L. Bihn,
1873.

9. Add. 38070, ff. 51-52. IO staves. Sketch for the Quartet in

C sharp minor, Op. 131, Andante. 1826. Bequest of Alfred
Morton, 1913.

96

This content downloaded from

193.144.2.38 on Sun, 17 Dec 2023 11:38:42 +00:00
All use subject to https://about.jstor.org/terms