Beethoven Violin Sonatas Performing Practice

BEETHOVEN
Sonatas
for Pianoforte and Violin
Performing Practice Commentary
by
Clive Brown
(University of Music and Performing Arts, Vienna)
and
Neal Peres Da Costa
(Sydney Conservatorium of Music, University of Sydney)
Bärenreiter Kassel · Basel · London · New York · Praha

BA 9014 | BA 9015
© 2020 by Bärenreiter-Verlag Karl Vötterle GmbH & Co. KG, Kassel
All rights reserved
This edition is subject to copyright; any use for other than strictly private purposes
is expressly prohibited and requires prior authorisation from Bärenreiter-Verlag.
This applies in particular to the use of texts or text excerpts in publications of any kind.
BA 9014: ISMN 979-0-006-53296-4 | BA 9015: ISMN 979-0-006-53297-1
CONTENTS
Piano Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1. Arpeggiation and asynchrony . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2. Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3. Pedals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Editions of Beethoven Sonatas for Piano and Violin by 19th-Century Musicians . . . . . . . . . . . . . . . 10
1. Ferdinand David . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2. Jean Delphin Alard and Louis Diémer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3. Friedrich Grützmacher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4. Edmund Singer and Wilhelm Speidel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5. Friedrich Hermann and Carl Reinecke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
6. Adolph Brodsky and Max Vogrich . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
7. Joseph Joachim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
8. Arnold Rosé . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
9. Carl Halir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
10. Fritz Kreisler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
11. Leopold Auer and Rudolph Ganz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
12. Arthur Seybold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Metronome Marks and Tempo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Commentary
Sonata op. 12, no. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Allegro con brio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Thema con Variazioni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Rondo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Sonata op. 12, no. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Allegro vivace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Andante più tosto Allegretto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Allegro piacévole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Sonata op. 12, no. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Allegro con spirito . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Adagio con molt’ espressione . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Rondo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Sonata op. 23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Presto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Andante scherzoso, più Allegretto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Allegro molto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Sonata op. 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Allegro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Adagio molto espressivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Scherzo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Rondo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Sonata op. 30, no. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Allegro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Adagio molto espressivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Allegretto con Variazioni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Sonata op. 30, no. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Allegro con brio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Adagio cantabile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Scherzo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Allegro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Sonata op. 30, no. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Allegro assai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Tempo di Minuetto, ma molto moderato e grazioso . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Allegro vivace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Sonata op. 47 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Adagio sostenuto – Presto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Andante con Variazioni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Presto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Sonata op. 96 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Allegro moderato . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Adagio espressivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Scherzo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Poco Allegretto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
PER FOR MING PR ACTICE COMMENTARY
This Commentary provides performers with evidence- For each movement of the ten Sonatas for Piano and
based perspectives on Beethoven’s notational precon- Violin, the metronome markings given in the sources
ceptions and what he expected his notation to convey described below (pp. 15 –17) are evaluated in relation
to the musicians of his own time. It also invites seri- to Beethoven’s tempo conventions and his own met-
ous engagement with expressive practices that were ronome marks for other works.
integral to the composer’s sound world, which were
progressively outlawed during the Modernist revo-
lution of the first half of the 20th century. More re-
PI A NO TECH NIQUES
cent performances of Beethoven’s music, both in the
modern and historically-informed arenas, have been (Neal Peres Da Costa)
largely shaped by the artistic agendas that emerged This section supplements the information given in the
from that ideologically-motivated purge of inherited edition, in “Reading between the lines” (5/c/ii),1 with
traditions of expressive performance. Some, or many information about pianos from Beethoven’s time, and
of the suggestions for interpretation and execution in provides a more detailed consideration of the ways in
the following commentary, which are firmly based on which 19th-century pianists employed arpeggiation,
historical evidence, will be seen to go against the tide asynchrony, and pedalling.
of current aesthetic and practical conceptions, which
are still firmly rooted in demonstrably unhistorical
ideals of studious faithfulness to the notation. Much 1. Arpeggiation and asynchrony
of the information presented here, directly challenges 1.a Arpeggiation
these ingrained misconceptions, which are kept alive Abundant written sources from Beethoven’s time
by the exigencies of such activities as orchestral audi- make it clear that particularly, but not exclusively in
tions, competitions, and commercial recording. music of an expressive character, chords not marked
The choice of period or modern instruments will staccato or not very short were expected almost always
dictate specific possibilities for colour, timbre, articu- to be arpeggiated, normally from lowest to highest
lation, and balance; but, regardless of whether the so- note. Frequent arpeggiation is evident on the piano
natas are played on modern or historical instruments, rolls of pianists born in the first half of the 19th cen-
it is the revival of un-notated, though historically-veri tury, and most notably in the piano playing of the old-
fiable expressive practices from the late 18th and 19th est German pianist, the much-revered Carl Reinecke
centuries, such as rhythmic and tempo flexibility, piano (1824 –1910), born when Beethoven was still alive.2 Swift
arpeggiation and asynchrony, portamento, and orna- arpeggiation seems to have been the default mode as
mental, rather than continuous vibrato effects, that of- prescribed, for example, by pianists closely associated
fers the greatest challenge to performer and listener. with Vienna including Johann Nepomuk Hummel and
The creative potential of these long-neglected expres- Sigismund Thalberg, the latter explained that the chords
sive resources – already beginning to be rediscovered that support (carry) the melody notes must always be
and exploited by increasing numbers of adventurous arpeggiated “but very tight and almost together, with
young and scholarly-orientated professional musi- the melody note more strongly than the other notes”.3
cians – promises to revitalise this magnificent music,
1 References to “Reading between the lines of Beethoven’s nota-
which, although still central to the mainstream clas- tion / Beethovens Notation – zwischen den Zeilen gelesen” (in BA
sical repertoire, is in danger of becoming stale from 9014 and BA 9015) are given in the following manner: (5/c/ii). This
formulaic repetition. The very fact that these practices indicates Part 5, Section c, Sub-section 2.
2 See Neal Peres Da Costa: Off the Record. Performing Practices in
are not specified in the notation allows great scope for Romantic Piano Playing (Oxford, 2012), pp. 159–165.
individual artistry. Practical experimentation, draw- 3 Sigismund Thalberg: L’Art du chant appliqué au piano, op. 70,
ing upon the evidence presented and discussed here, 1 er série (Paris: Heugel & Cie., [c. 1853] unpaginated 2). “Les ac-
cordes qui porteront un chant à la note supérieure devront tou-
offers the possibility to illuminate and inspire fresh jours s’arpéger, mais très serrés, presque plaqués, et la note de
interpretations of Beethoven’s chamber music. chant plus appuyée que les autres notes de l’accord”.
© 2020 by Bärenreiter-Verlag, Kassel Piano Techniques ▪5

In L’Art du chant, one of Thalberg’s aims was to help chord unarpeggiated while arpeggiating very
pianists to produce on the piano the illusion of sing- swiftly the left-hand chord.9 Thalberg seemingly
ing with its “sustained and prolonged sounds, but notates this type of effect in bars 7–8 of his ar-
also that of [its] swelling sounds.” 4 rangement of Mozart’s “Lacrymosa” using the
sign to denote unarpeggiated chords in the right
Varying speeds of arpeggiation in certain circum- hand, against arpeggiated chords marked thus 
stances are prescribed by Philip Anthony Corri, J ohann in the left hand.10
Baptist Cramer, and Carl Czerny in places where mood iv) Arpeggiation might be used when individual parts
or texture require it (3/d/i; 5/c/ii). in a musical structure (such as in polyphony or
Arpeggiation was used: canon) require delineation.11 While Czerny stated
▪ to give special expression to melody notes by delay- that arpeggiation should not generally be used in
ing them; fugues, there is nothing to suggest that he would
▪ to create rhythmic and temporal variation through have taken issue with slight separation of fugue
agogic accentuation; voices (at their point of entry) from the accom
▪ to effect varying degrees of softness and brilliance; panying texture.
▪ to ensure structural delineation. v) Where one-part ends, and another simultaneously
Some of the intricacies of arpeggiation that trained begins, arpeggiation can be used to make sepa-
pianists of the era employed (often probably uncon- ration.12
sciously) may be gleaned from a few sources connected vi) When chords are to be accented:
with Beethoven or his era. These include the prac- ▫ a “special rhythmical effect” (agogic accen-
tices of influential pianist-pedagogues such as Carl tuation), for example when the beginning of
Reinecke, Thalberg, and Theodor Leschetizky – who a bar is marked sf, can be created when the
studied with Czerny.5 lowest left-hand note coincides exactly with
i) When the texture is an expressive song or aria in the beat, while the upper left-hand note/s is/
a soft dynamic in which the melody is accompa- are struck together with the right-hand chord
nied by chords, frequent arpeggiation creates ex- played unarpeggiated, which causes “an ex-
pressive delay of the melody note (which itself is tremely slight retardation.” 13
given tonal emphasis), bringing it out of the tex- ▫ “a slight [swift] arpeggio” will “soften the
ture while also creating the illusion of sustained, hardness of touch apt to arise” from the ac-
prolonged and even swelled sounds. This effect cent for chords marked ff or sfz.14
is evident in bars 2–3 of Thalberg’s arrangement ▫ “a very short [swift] arpeggio” will mitigate
of Mozart’s “Lacrymosa” in imitation of the sus- “all unpleasantness of effect without weaken-
tained sounds of a choir,6 and also in bars 104 –109 ing its character” for “very abrupt disson
and 145 –151 of his arrangement of Beethoven’s song ances” marked sfz.15
Adelaide.7 Additionally, Leschetizky’s and Reinecke’s piano
ii) When the texture is chords in both hands a “tender rolls confirm multiple means of arpeggiation in-
or delicate effect” can be attained by arpeggiat- cluding:
ing the right-hand chord (presumably in varying ▫ arpeggiated main beats and unarpeggiated
speeds according to expressive purpose) while the weak beats
left-hand chord is unarpeggiated.8 ▫ arpeggiated left-hand chord together with
iii) An energetic effect, which is not hard (for example un-arpeggiated right-hand chord
for sff) can be achieved by playing the right-hand
9 Ibid., p. 72.
10 Thalberg: ibid., p. 1.
4 Ibid., unpaginated 1. “à produire l’illusion des sons soutenus et 11 Brée: Die Grundlage der Methode Leschetizky, p. 72
prolongés, mais encore celle de sons enflés.” 12 Ibid.
5 See Peres Da Costa: Off the Record, pp. 129–139. 13 Ibid.
6 Thalberg: ibid., p. 1. 14 Otto Klauwell: Der Vortrag in der Musik. Versuch einer systema-
7 See Thalberg: L’Art du chant appliqué au piano (Boston, n.d.), tischen Begründung desselben zunächst rücksichtlich des Klavierspiels
pp. 9–10. (1883), trans. as On Musical Execution: an Attempt at a Systematic
8 Malwine Brée: Die Grundlage der Methode Leschetizky (1902), trans. Exposition of the same Primarily with Reference to Piano-Playing (New
Theodor H. Baker as The Groundwork of the Leschetizky Method (New York, 1890), pp. 110–115. Otto Klauwell studied with Reinecke.
York, 1902), p. 71. 15 Ibid.
6 ▪ Piano Techniques © 2020 by Bärenreiter-Verlag, Kassel

▫ arpeggiated right-hand chord together with tion of the notes of the melody a long time after those
un-arpeggiated left-hand chord of the bass [have been sounded] and thus producing,
▫ unarpeggiated right-hand chord played after from the beginning to the end of a composition, the
arpeggiated left-hand chord effect of continuous syncopations.” 17
▫ arpeggiated right-hand chord played after In theory, asynchrony was applied in order to give
unarpeggiated left-hand chord melody notes “more relief [prominence] and a softer
▫ unarpeggiated right-hand chord played after effect”,18
unarpeggiated left-hand chord ▪ on the first beat of each bar
See also Neal Peres Da Costa, “Carl Reinecke’s ▪ at the start of each period or phrase
Performance of his Arrangement of the Second ▪ on important notes
Movement from Mozart’s Piano Concerto K. 488. ▪ on strong beats
Some Thoughts on Style and the Hidden Messages In reality, however, Leschetizky, Reinecke and other
in Musical Notation,” in: Rund um Beethoven. In- 19th-century pianists employed asynchrony much
terpretationsforschung heute, ed. Thomas Gartmann more often (for example on every note in a sequence
and Daniel Allenbach (= Musikforschung der Hoch- of poignant melody-notes), in ways not described in
schule der Künste Bern 14), pp. 140ff.). written sources, and to an extent (in terms of delay be-
tween melody note and accompaniment) that hardly
1.b Asynchrony equates with the modern meaning of the term imper-
Written evidence also reveals that certain melody- ceptible.19
notes were expected to be played in an asynchronous
manner “imperceptibly” after or (as an exception) be-
fore the accompanying bass note or chord. This expres- 2. Instruments
sive practice was introduced into French 17th-century 2.a Viennese
keyboard playing (harpsichord, virginals, spinet and In Viennese aristocratic settings or concert venues, per-
clavichord) in imitation of lute practice, and appears formances of Beethoven’s works with piano would al-
to have quickly gained popularity. By the end of the 17th most certainly have been heard on a Viennese-action
century it was notated as an ornament called suspen- grand piano. These were straight-strung, wooden-
sion, which François Couperin prescribed for instances framed pianos, with leather covered hammers, Austro-
where “bowed instruments would increase [swell] their German “bouncing” action or Prellmechanik (allowing
tone” and which on the harpsichord “by a contrary the hammer to fly at high velocity to the string), and
effect, seems to produce this desired result.” 16 shallow key dip which required a very light touch mak-
By the middle of the 18th century some French coming rapid passagework easy. Such pianos are charac-
posers notated suspension on practically every note in teristically light and transparent in sound (compared
music of a slow, gracious, tender or expressive charac- with modern grand pianos) and have distinct tonal
ter, a seemingly telling indication of the frequency of ranges from bass to treble (due to stringing) that allow
its use. Other countries adopted suspension and it un- melody and accompaniment to stand out distinctly
doubtedly continued as a normal and valued expres- and naturally. In domestic settings, Viennese square
sive device by harpsichordists (in the last three or so or other types of pianos would also have been used.
decades of the 18th century) who started to play pianos Between 1797 and 1815, the period encompassing the
of various emerging types. During Beethoven’s life- composition of the Sonatas for Piano and Violin, Vien-
time, and until at least the mid-19th century it is like- nese-action grand pianos were normally triple strung
ly that asynchrony was used with increasing rather
17 Thalberg: L’Art du chant, p. 3. “Il sera indispensable d’éviter,
than decreasing frequency, a fact supported by Thal- avec exagération le frappement des notes de chant longtemps après
berg’s description in c. 1853: “It will be indispensable celles de la basse, et de produire ainsi, d’un bout à l’autre d’un
to avoid, in playing, the ridiculous habit and in bad morceau, des effets de syncopes continues. Dans une mélodie lente
écrite en notes de longue durée, il est d’un bon effet, surtout au
taste, of withholding with exaggeration the produc- premier temps de chaque mesure ou en commençant chaque pé-
riode de phrase, d’attaquer le chant après la basse, mais seulement
16 François Couperin: L’Art de toucher le clavecin (Paris, 1716), p. 16. avec un retard presque imperceptible.”
“en sorte que dans Les occasions ou les instrumens à archet enflent 18 Brée: The Groundwork, p. 73. See also Thalberg: L’Art du chant,
leur sons, La Suspension de ceux du clavecin semble, (par un èffet unpaginated 2.
[sic] contraire) retracer à L’oreille La chose souhaitée.” 19 See Peres Da Costa: Off the Record, pp. 51–72.

throughout, though sometimes the bass end might be cantly higher string tension that supported a character-
double strung, and were known for their efficient damp- istically rounder, and more resonant sound. English-
ing. In Beethoven’s early years the compass would action pianos were heavier in touch than Viennese-
have been just over 5 octaves, by his death up to six action pianos with a deeper key dip, which made the
and a half octaves. Throughout his lifetime Beethoven execution of fast passage work more challenging than
greatly admired Viennese pianos, and in particular on Viennese pianos. Beethoven’s Broadwood had a
those of Nanette Streicher and Conrad Graf. compass of six octaves, was triple strung throughout
and fitted with two pedals; on the left a soft pedal
2.b French (una corda shift) and on the right a damper pedal di-
From 1803 Beethoven also became well acquainted with vided in two to dampen separately the treble (right
Sébastien Érard’s pianos (he already knew Haydn’s side) and bass (left side). For an informative report on
Érard), having been gifted one by the Paris maker. This the difference between Viennese- and English-action
piano had an English-style action (see below) with- pianos see also Johann Nepomuk Hummel.21
out double escapement with which Beethoven was evi-
dently initially very happy, despite reportedly finding
it somewhat peculiar and difficult to play due to its 3. Pedals
heavy action. It had a range of five and a half octaves, 3.a Soft pedal
was triple strung throughout and significantly more By the late 18th century, grand pianos began to be fit-
resonant than contemporary Viennese grand pianos. ted with a keyboard shift mechanism engaged by a
It was also novel in having four pedals: an una cor- pedal so that, depending on the level of pedal depres-
da shift, a damper, a lute and moderator. The Érard sion, the hammers could conveniently strike alterna-
inspired the composition of various piano works in- tively one string (una corda), two strings (due corde)
cluding the ‘Waldstein’ Piano Sonata op. 53. Beetho- or three strings (tre corde), thereby effecting substan-
ven relinquished the Érard in 1810, remarking that it tial modification of dynamic and timbre. Beethoven
was no longer of use to him. Given that all but the last will have expected pp (particularly if accompanied by
of the Sonatas for Piano and Violin were written and espressivo) and the rare instances of ppp in his music
performed by 1803, an Érard piano of this era would (see above) to be enhanced with the una corda, a practice
make an historically appropriate choice.20 already recommended by Johann Peter Milchmeyer
in 1797 as “excellent” and suggesting “a very distant
2.c English music, or the answer of an echo.” 22
By all accounts, Beethoven cherished the English grand For mp, dolce, mf, and for louder dynamics, Beetho-
action piano that was gifted to him by Thomas Broad- ven will have expected the use of the due corde and tre
wood in 1817, and which inspired many late compo- corde settings providing effective contrast. He some-
sitions including the ‘Hammerklavier’ Piano Sonata times specified its use, marking, for example, Mit einer
op. 106. The Broadwood arrived well after the compo- Saite, Sul una corda at the beginning of the 3rd move-
sition of the last of the Sonatas for Piano and Violin ment of his Piano Sonata op. 101 after which (at the
(op. 96 composed between 1812 and 1815), neverthe- end of bar 20) he instructs the player gradually to
less, considering Beethoven’s popularity in England, shift to more strings, Nach und nach mehrere Saiten. In
the use of a Broadwood piano would also be histori- the slow movement of his Piano Sonata op. 106 Beet-
cally appropriate. This piano had the English-action hoven specifies the use of the una corda and the shift
or Stoßmechanik (push mechanism) in which a jack was back to tre corde many times, stipulating in bars 76–77
made to engage directly with the hammer butt, caus- that the player should shift from one to three strings
ing the hammer to be pushed forcefully to the string. 21 Johann Nepomuk Hummel: Ausführliche theoretisch-praktische
It was fitted with a double escapement allowing quick Anweisung zum Piano-Forte-Spiel, 3 parts (Vienna, 1828), part 3,
and efficient note repetition, and had a much sturdier pp. 438– 439.
22 Cited from Robert Rhein: Johann Peter Milchmeyer’s “Die wahre
construction than Viennese pianos allowing signifi- Art das Pianoforte zu spielen”: An annotated translation (Ann Arbor,
1993), p. 156. “Die Veränderung, wo der Hammer nur eine Saite
20 For a fuller discussion about Beethoven’s Érard see Tilman anschlägt, nimmt sich sehr gut aus, und stellt, wenn man mit zu-
Skowroneck: “Beethoven’s Erard Piano: Its Influence on His Com- gemachtem Deckel spielt, eine weit entfernte Musik, oder die Ant-
positions and on Viennese Fortepiano Building,” in: Early Music, wort des Echos vor.” Johann Peter Milchmeyer: Die wahre Art das
vol. 30, No. 4 (2002), pp. 522–538. Pianoforte zu spielen (Dresden, 1797), p. 65.
8 ▪ Piano Techniques © 2020 by Bärenreiter-Verlag, Kassel

gradually (poco a poco due ed allora tutte le corde). The late 18th and first half of the 19th centuries, was based
una corda, also sometimes referred to in German as on a rhythmic style effected by “damping with the
mit Verschiebung (with a shift of the action), produces change of harmony or at the bar line”.26 Czerny’s ad-
an ethereal sound world difficult to replicate on later vice on damper pedal use supports this idea; he states
pianos that do not have an una corda shift capability. that the damper should be changed very rapidly and
“must take place strictly with the first note of each
3.b Moderator chord”.27 Further evidence comes from Moscheles, who
Additionally, late-eighteenth- and-early nineteenth- advised that “At each new harmony the Pedal must
century pianos were often fitted with a single (some- be carefully taken off, and again used at the begin-
times also a double) moderator operated by pedals, ning on a new one.” 28 Certainly, the placement of
which could be used to modify the piano sound by damper pedal indications in printed piano music of
engaging a piece of cloth (usually felt) between the the early 19th century tend to support the predomi-
hammers and strings to be used “for pianissimo pas- nance of rhythmic pedalling, but as the 19th century
sages, for echo effects, or for contrast with fortissimo progressed, it appears to have been increasingly su-
sections”.23 Beethoven would also undoubtedly have perseded by a syncopated style in which the damper
expected the use of the moderator to enhance music pedal was changed after the new note or chord had
designated with dolce, espressivo and other expressive sounded, in line with the ideal of legato or connected
tone and tempo modifying expressions or where the style in piano playing.29 Nevertheless, rhythmic ped-
compositional style called for special effect. alling continued to be employed as late as the early
20th century as exemplified in the playing of Reine
3.c Sustaining (damper) pedal cke and other 19th-century pianists.
By Beethoven’s era sustaining or damper pedals were Thalberg’s advice about the use of the pedals in L’Art
already very popular. According to Milchmeyer they du chant reflects attitudes that developed in the first
were used to “create the most beautiful but also the half of the 19th century and are therefore pertinent to
most dreadful modification, depending on whether pedalling in music of Beethoven and his era: “The use
they are employed with taste or poorly, for in the lat- of two pedals [the soft and the sustaining] (together
ter case all notes sound mingled together, and cause or separately), is indispensable, to ensure the proper
such intolerably bad noise that one would like to plug [full] effect in performance, to sustain similar har-
one’s ears.” 24 We know from Czerny that Beethoven monies, and produce, by their judicious use, the illu-
used the damper pedal much more than he notated sion of prolonged and sustained sounds. Often for these
it.25 Beethoven will have expected its use as an aid to particular effects, the pedals must only be used after
legato but also, from time to time, for special blurring sounding the long notes of a melody [a seemingly
effects, for example in the first movement of the Piano clear description of syncopated pedalling], but it is
Sonata op. 27 no. 2 where he specifies the dampers to difficult from here on to designate precisely these par-
be continually raised throughout - senza sordino (most ticular cases, because they depend mainly on the sen-
successful on an early 19th-century Viennese style in- timents and sensations to be expressed, rather than in
strument), and multiple instances in the Piano Sonata following fixed rules. In using the pedals, which play
op. 31 no. 2, particularly in bars 143–148 and 153–158 such an important role in the performance of piano
(4/a/ii). It appears that damper pedal technique, in the music, one should take great care never to mix dis-
23 Sandra Rosenblum: “Pedaling the Piano, A Brief Survey from 26 Elfrieda Hiebert: “Listening to the Piano Pedal: Acoustics and
the Eighteenth Century to the Present,” in: Performance Practice Pedagogy in late Nineteenth-Century Contexts,” in: Osiris, vol. 28,
Review, vol. 6 no. 2 (1993), p. 163. no. 1 “Music, Sound, and the Laboratory from 1750–1980” (January
24 Rhein: Johann Peter Milchmeyer’s “Die wahre Art das Pianoforte 2013), pp. 232–253, here p. 234.
zu spielen”, p. 144. “Uiber die Dämpfer ist viel zu erinnern, sie ma- 27 Carl Czerny: Pianoforte School (London, 1839), vol. 3, pp. 59–63,
chen sie schönste aber auch die abscheulichste Veränderung, je here p. 62.
nachdem sie mit Geschmack oder übel angewendet werden, denn 28 Ignaz Moscheles: 24 Etudes op. 70, “Etude no. 9” (Leipzig: [1827]),
im lezten Falle klingen alle Töne unter einander, und verursachen p. 44. “vor einer neuen Harmonie sind die Saiten präcis zu dämp-
so unerträglichen Uibellaut, daß man sich die Ohren verstopfen fen, und mit dem Anfang derselben das Pedal wieder zu heben.”
möchte.” Milchmeyer, Die wahre Art das Pianoforte zu spielen, p. 59. 29 See Neal Peres Da Costa: “Performing Practice in Piano Play-
25 For a detailed discussion about this see Leonardo Miucci: ing,” in: Clive Brown, Neal Peres Da Costa, and Kate Bennett Wads-
“Beethoven’s Pianoforte Damper Pedalling: a Case of Double No- worth: Performance Practices in Johannes Brahms’ Chamber Music
tational Style,” in: Early Music, vol. 47 (August 2019), pp. 371–392. (Kassel, 2015, BA 9600).

similar harmonies, thus producing disagreeable dis-
sonances. There are pianists whose use of the pedals 1. Ferdinand David
is abused in this way, or rather who employ them (Leipzig, Peters, 1868 plate numbers 4899, 4926); revised
with so little logic, that with them the sense of hear- edition (Leipzig, Peters, c. 1885, plate number 6531).
ing is perverted and that they have lost consciousness See CH ASE http://mhm.hud.ac.uk/chase/view/edition/
of pure harmony.” 30 264/ for the complexities of the various versions of
David’s edition.
From 1823 to 1824, Ferdinand David (1810–1873) stud-
ied in Kassel with Spohr (who was friendly with Beet-
hoven during his time as second Kapellmeister at the
EDI T IONS OF BEET HOV EN SONATAS Theater an der Wien between 1812 and 1815). David be-
FOR PI A NO A N D V IOLI N came Mendelssohn’s close friend, then his colleague
BY 19T H-CEN T U RY M USICI A NS as leader of the Gewandhaus Orchestra from 1835 and
violin professor at the Leipzig Conservatorium from
(Clive Brown)
its foundation in 1843.
The performing editions by 19th-century musicians In 1863 David published his Violinschule, which was
that have been consulted during preparation of the the first important treatise to attempt an assimila-
Performing Practice Commentary, in addition to edition of newer French bowing practices into a German
tions and writings about performance from Beetho- context. There is no evidence, however, that he envis-
ven’s own time, are listed below, in order of publica- aged the application of those practices in his editions
tion. Where there was a separate editor for piano, the of Viennese Classical chamber music, only in certain
violinist editor is given first. An edition (Wiener Neu- virtuoso repertoires and in contemporary music.
stadt, Wedl, 1883) by Jacob Dont (1815–1888), whose fa- His bowing instructions in the edition seem largely
ther, the cellist Joseph Valentin Dont, his first teacher, to indicate a style similar to Spohr’s, with the execu-
was a close colleague of Beethoven, has very unfor- tion of short separate notes in the middle and upper
tunately been impossible to locate, despite years of half of the bow, mostly on the string, but perhaps oc-
searching. casionally with something similar to what he calls a
The following notes locate each of the editors in his- “hopping” [hüpfend] bowstroke in his Violinschule, for
torical context. passages of separate, rapid notes, where the elasticity
of the stick is utilised, but without the bow-hair leav-
ing the string.31 His fingering is not comprehensive
and seems to take much for granted, such as the use
of open strings and harmonics, which can often be
inferred from the left-hand position.
The evidence of his surviving personal copies, with
comprehensive handwritten markings, indicates that
in his own performances he probably used more so-
phisticated fingering and varied bowing than the
30 Thalberg, L’Art du chant [Heugel], unpaginated 2. “L’usage des rather basic technical guidance in the edition.32 The
deux pédales (ensemble ou séparément) est indispensable pour
donner de l’ampleur à l’exécution, soutenir les harmonies sem-
piano part is essentially that of the Breitkopf & Härtel
blables et produire, par leur emploi judicieux, l’illusion des sons Vollständige Ausgabe, which he and Carl Reinecke had
prolongés et enflés. Souvent, pour ces effets particuliers, il ne faut helped to edit.
les employer qu’après l’attaque des longues notes de chant; mais il
nous serait difficile ici de préciser les cas généraux, attendu qu’ils
The unattributed revision of David’s edition, issued
tiennent en partie plutôt au sentiment et aux sensations qu’aux in the decade after his death, made many changes to
règles fixes que nous formulerons dans notre méthode. On devra bowing, slurs, staccato marks and fingering. The bow-
dans l’emploi des pédales, qui jouent un rôle si important dans
l’exécution, apporter le plus grand soin à ne jamais mêler les har- ing changes result in the upper part of the bow being
monies dissemblables et à produire ainsi de désagréables disso- used somewhat less, and fewer instances of slurred
nances. Il est des pianistes qui font des pédales un tel abus, ou
plutôt qui les emploient avec si peu de logique, que chez eux le 31 Ferdinand David: Violinschule (Leipzig, [1863]), vol. 2, p. 38.
sens di l’ouïe en est perverti et qu’ils ont perdu la conscience d’une 32 See Clive Brown: “Ferdinand David as editor” http://mhm.hud.
harmonie pure.” ac.uk/chase/article/ferdinand-david-as-editor-clive-brown/.
10 ▪ Editions of Beethoven Sonatas © 2020 by Bärenreiter-Verlag, Kassel

staccato or short notes being hooked into a bowstroke. a paragraph headed ‘Of separate bowstrokes’ (Du Dé-
It is unlikely that they were David’s own revisions; taché) states: “Whenever there are no [staccato] dots,
they seem to reflect the practices of a younger gene the notes are to be executed with the bow on the string.
ration. This version of David’s edition, with its unat- On the contrary with dots, it will be in the middle of
tributed revisions, seems to have formed the basis for the bow, what is called sautillé”.34 On the other hand,
Joachim’s, Halir’s, and Seybold’s markings. the application of the instructions in his General
David’s violin parts (original and anonymously re- Comments to his Beethoven editions is questionable,
vised) can be accessed on C H A SE : http://mhm.hud. since in op. 30 no. 2 he marked both dots and vertical
ac.uk/chase/view/edition/264/ strokes for staccato which have no relationship with
The piano and violin parts of the original edi- the articulation marks in the Breitkopf & Härtel Voll-
tion are also accessible on I M SLP : https://imslp.org/ ständige […] Ausgabe, even adding them on notes that
wiki/Sonaten_f%C3%BCr_Pianoforte_und_Violine_ have no staccato marks in any source; and in other
(Beethoven%2C_Ludwig_van) sonatas he frequently fails to extend Beethoven’s stac-
cato marks at the beginning of a passage to the fol-
lowing notes. It is also clear from his use of  and 
2. Jean Delphin Alard / Louis Diémer that some passages with staccato marks cannot have
(Paris: Heugel, c. 1868) been envisaged with an elastic bowstroke.
Jean Delphin Alard (1815–1888) studied at the Paris Alard’s editions can be accessed on Gallica at https://
Conservatoire from 1827 with François Antoine Habe- catalogue.bnf.fr/ark:/12148/cb42839780b
neck (1781–1849). He became Pierre Baillot’s successor
as professor at the Conservatoire in 1843 and pub-
lished his École du violon, Méthode complète et progres- 3. Friedrich Grützmacher
sive à l’usage du Conservatoire in 1844. Louis Diémer (arranged for cello) (Leipzig, Breitkopf & Härtel, 1874)
(1843–1919) studied at the Paris Conservatoire and be- Friedrich Grützmacher (1832–1903) became a protégé
came a professor there in 1888. His piano parts include of Ferdinand David in 1848, and shortly afterwards
detailed fingering. Their edition, part of Alard’s much his colleague in the Leipzig Gewandhaus Orchestra,
larger Les maîtres classiques du violon, is one of the few at the Conservatorium and in the Gewandhaus String
to contain metronome marks. For most movements, Quartet. He moved to Dresden in 1860 as first cellist of
these are distinctly slower than those provided in the orchestra and, later, became a teacher at the Dres-
sources with a Viennese connection. An increasing den Conservatorium. He was also a much-travelled
use of vibrato was already associated with French and soloist, considered by some to be the Joachim of the
Franco-Belgian violinists from the 1840s onwards, but cello.35
the numerous open strings and natural harmonics Wilhelm Joseph von Wasielewski (1822–1896), who
marked in Alard’s edition demonstrate beyond doubt knew him personally, observed: “In Grützmacher’s
that, like Charles de Bériot (1802–1870), he still envis- playing were happily united the endowments of a fin-
aged a fundamentally pure, vibrato-free sound as the ished mastery of complex technical difficulties and a
normal tone of the instrument, with nuance and col- delicate manner of expression, more especially in the
our provided by varied bow pressure and occasional rendering of Cantilena. He is not only a virtuoso of
expressive vibrato; in fact, his École teaches portamen- the first rank, but also an excellent interpreter of clas-
to, but does not teach vibrato at all.33 sical chamber music. For this latter qualification the
Alard’s approach to bowing, however, is fundamen- foundation had already been laid by a careful musical
tally different from David’s and the earlier German vio education under his father’s roof, to which Friedrich
linists. In the General Comments (Observations géné- Schneider [1786–1853] had substantially contributed.” 36
rales) preceding each volume of Les Maîtres classiques,
34 “Tous les fois qu’il n’aura pas de points, il s’exécutera l’archet
33 See Charles de Bériot: Méthode de violon (Paris, [1858]), especially sur la corde. Au contraire avec des points, ce sera du milieu de
part 3, pp. 220ff. Also Clive Brown: “Singing and string playing in l’archet, ce qu’on nomme sautillé.”
comparison: instructions for the technical and artistic employment 35 See Kate Bennett Wadsworth: ‘Precisely marked in the tradition
of portamento and vibrato in Charles de Bériot’s Méthode de violon,” of the composer’: the performing editions of Friedrich Grützmacher, PhD
in: Zwischen schöpferischer Individualität und künstlerischer Selbstver- diss. University of Leeds, 2017, p. 56.
leugnung, ed. Claudio Bacciagaluppi, Roman Brotbeck and Anselm 36 Cited from Wilhelm Joseph von Wasielewski: The Violoncello
Gerhard (Schliengen, 2009), pp. 83–108. and its History, trans. Isobella S. E. Stigand (New York, 1894), p. 127.
© 2020 by Bärenreiter-Verlag, Kassel Editions of Beethoven Sonatas ▪ 11

Grützmacher’s edition includes an extensively adapt-  Hinaufstrich
ed piano part and is interesting for its regular mark-  Herunterstrich
ing of turns at the end of trills where Beethoven did Fr. Frosch
}
not bother to notate them. The cello parts can be ac- Sp. Spitze des Bogens.
cessed on C H A SE : http://mhm.hud.ac.uk/chase/view/ M. Mitte
edition/1575/ ⁄ Rutschen (gleiten).
restez in der Lage bleiben.
4. Edmund Singer / Wilhelm Speidel

’

Zeichen für
eine kurze Pause.
(Stuttgart, Cotta, 1887) – Gehalten.
Edmund Singer (1831–1912), after initial studies in Bu- I.a E-
}
dapest, was, like Dont and Joachim, a pupil of Joseph II.a A-
a
Saite.
Böhm in Vienna, after which he also spent a year at III. D-
the Paris Conservatoire (1845), which was perhaps IV.a G.
decisive, especially for his approach to bowing. De-
spite having studied with Böhm at the same time as The edition is particularly informative for its marking
Joachim, their editions suggest a quite different style of of portamento fingering and the portamento symbol
playing. Singer succeeded Joachim as Liszt’s Konzert ⁄ , which specifies sliding in places where it is not
meister in Weimar in 1856 and from 1861 to 1902 worked obvious from the fingering (especially between bow-
in Stuttgart as Hofkonzertmeister and taught at the strokes). Singer’s use of portamento may have been
Conservatorium. Wilhelm Speidel (1826–1899) studied more extensive than that of most violinists of Beetho-
in Munich from 1842 with Ignaz Lachner (1807–1895), a ven’s time, perhaps representing one end of a spec-
member of the Schubert circle. He taught at the Stutt- trum, while the more restrained employment of ex
gart Conservatorium from 1857. Their edition, which pressive shifts in David’s and Joachim’s editions rep-
was published as “edited and precisely marked up es- resents the opposite end. His fingerings indicate very
pecially for use in music conservatories” (“Insbeson- restrained use of vibrato. Speidel’s piano parts are ex-
dere zum Gebrauch in Konservatorien für Musik re- tensively fingered.
vidiert und genau bezeichnet”), provides metronome Singer’s violin parts can be accessed on C H A SE :
marks, which generally reflect a somewhat slower con- http://mhm.hud.ac.uk/chase/view/edition/635/
cept of Beethoven performance than the metronome
marks of Moscheles and Czerny. The edition also has
many added performance instructions, explanations 5. Friedrich Hermann / Carl Reinecke
of ornaments, marks of expression and touch, includ- (Leipzig, Breitkopf & Härtel, 1890)
ing tranquillo, espressivo, leggiero and so on. Not all of Friedrich Hermann (1828–1907), a pupil of Ferdinand
these will have a connection with Beethoven’s time, David from 1843, was also his colleague in Leipzig in
but they provide fascinating insights into a mid-19th- the Gewandhaus Orchestra, the Gewandhaus cham-
century approach to the sonatas, which differs very ber ensembles, and from 1848 a professor at the Con-
substantially from 20th- and early-21st-century main- servatorium. His Violin-Schule (Leipzig, Peters, c. 1879)
stream practice. is clearly in the classic German tradition, with its em-
The violin parts are annotated with the following phasis on firm, on-string bowstrokes; the numerous
symbols: exercises contain examples of figurations and rhythms
with separate and slurred notes, which all correspond
“In dem Spiel Grützmachers waren die Vorzüge vollendeter Be-
herrschung der kompliziertesten technischen Schwierigkeiten und
with types used by Beethoven. Like Spohr and Da-
feinsinniger Ausdrucksweise, namentlich auch bezüglich des Kan- vid, he gives clear instructions showing that except
tilenenvortrages, in glücklicher Weise miteinander vereinigt. Er for longer notes and slurred figures, the bow is used
war indessen nicht nur ein Virtuos ersten Ranges, sondern auch
ein vorzüglicher Interpret der klassischen Kammermusik. Zur mostly in the upper half. He devotes only a single ex-
letzteren Eigenschaft wurde der Grund schon im elterlichen Hause ercise near the end of the method (No. 128) to “Exer-
durch eine sorgsame musikalische Erziehung gelegt, bei welcher cises for the springing and throwing way of bowing.”
Friedrich Schneider wesentlich mitgewirkt hatte.” Wilhelm Joseph
von Wasielewski: Das Violoncell und seine Geschichte, 2nd edition, [“Uebungen für Stricharten mit springendem und mit
ed. Waldemar von Wasielewski (Leipzig, 1911), pp. 140f. werfendem Bogen”]

Carl Reinecke (1824 –1910), as well as being a noted Moscheles and Reinecke. After extensive concert tours
composer and conductor, was one of the most respect- he moved to New York in the late 1880s.
ed classical pianists in Germany. He was encouraged Brodsky’s fingering is quite traditional, still with
by Mendelssohn and, particularly admired by Schu- many open strings and harmonics. Vogrich, like sev-
mann. His perceived authority as a faithful preserver eral other editors provides additional indications of
of Classical traditions in Mozart and Beethoven per- legato and quite detailed fingering.
formance was considerable. From 1860 he worked in Brodsky’s violin parts can be accessed on C H A SE :
Leipzig as conductor of the Gewandhaus Orchestra http://mhm.hud.ac.uk/chase/view/edition/110/
and professor of composition and piano at the Con-
servatorium; his long-term association there with Mo
scheles may have given him particular insights into 7. Joseph Joachim (Leipzig, Peters, 1901)
early-19th-century Viennese performing practices. He Following preliminary studies in Budapest, Joseph
was also one of the editors of the Breitkopf & Härtel Joachim (1831–1907) went to Vienna in 1839 and, after
Vollständige Ausgabe of Beethoven’s works. His playing brief study with Georg Hellmesberger Sr (1800–1873),
of Classical repertoire is preserved on piano rolls.37 became a pupil of Joseph Böhm. In 1843, he was men-
Hermann’s marking of the violin part is practical, tored by Mendelssohn in Leipzig and took consulta-
with limited expressive fingering. Reinecke provides tion lessons with Ferdinand David, subsequently be-
extensive fingerings and amplification of legato mark- coming his colleague in the Gewandhaus orchestra
ings. Their edition is also notable for its addition of and chamber ensembles. His subsequent career took
turns at the end of trills where Beethoven did not no- him to Weimar, where he worked with Liszt, to Han-
tate them. nover, and then to Berlin, where he held a leading po-
Hermann’s violin parts can be accessed on C H A SE : sition in the Hochschule and in the musical life of the
http://mhm.hud.ac.uk/chase/view/edition/458/ city, especially as leader of the Joachim Quartet, until
The score and parts can be accessed on I M SLP : his death. From the time of his London performance
https://imslp.org/wiki/Sonaten_f%C3%BCr_Pianoforte of Beethoven’s Violin Concerto under Mendelssohn in
_und_Violine_(Beethoven%2C_Ludwig_van) 1844, he also became an almost annual visitor to Lon-
don and had his own quartet there. His reputation as
a masterful performer of Beethoven was unrivalled.
6. Adolph Brodsky / Max Vogrich Hugo Heermann recalled: “I felt deeply that through
(New York, Schirmer, 1894) Joachim I was truly initiated into the ultimate secrets
Adolph Brodsky (1851–1929), born in Russia, studied at of chamber music playing – one really heard the mur-
the Vienna Conservatorium with Joseph Hellmesber murings of the source when Joachim with his quartet
ger Sr (1828–1893) from 1860 and was 2nd violin in the played Haydn, Mozart, or Beethoven. As well as his
Hellmesberger Quartet for a few years before moving brilliant gifts he also had the great good fortune to
to Moscow in 1873. From 1883 to 1891 he was a lead- have spent three of the youthful years in which his
ing figure in Leipzig where he formed the Brodsky whole artistry was developed in the most intimate
Quartet. Elisabeth von Herzogenberg, however, criti- contact in Vienna with that Josef [sic] Böhm, who with
cised him to Brahms in 1885 for his excessive use of Beethoven had, from the manuscripts, given the com-
vibrato and portamento in Brahms’ Violin Concerto.38 poser’s last quartets their first transformation into
The later part of his career was spent in New York sound. I grasped ever more clearly and consciously
(1891–1895), where he edited the Beethoven Sonatas, the task that German violinists had to solve in the
and Manchester (1895–1929). meaningful shaping of the chamber compositions of
Max Vogrich (1852–1916), born in the Austrian Em- the mighty great masters, that inexhaustible musical
pire, studied at the Leipzig Conservatorium under world, which in those days had so few listeners.” 39
37 See Sebastian Bausch: “Die Idee des Componisten ins Leben ru- 39 Hugo Heermann: Meine Lebenserinnerungen (Leipzig, 1935), ed.
fen” – Personalstil und Geschichtsbewusstsein in der akademischen Inter- Günther Emig (Niederstetten 2014, Kindle ebook Location 372. “Zu-
pretationspraxis des 19. Jahrhunderts im Umfeld des Leipziger Konserva- tiefst empfand ich es, daß ich durch Joachim in Wahrheit in die
toriums, PhD Diss Universität Bern (forthcoming). letzten Geheimnisse des kammermusikalischen Musizierens ein-
38 Clive Brown: “String Performing Practice,” in: Brown, Peres Da geführt wurde – man hörte wirklich die Quelle rauschen, wenn Jo-
Costa, Bennett Wadsworth: Performance Practices in Johannes Brahms’ achim mit seinem Quartett Haydn, Mozart oder Beethoven spielte.
Chamber Music (Kassel, 2015), p. 11. Joachim hatte eben neben seiner genialen Begabung noch das große
© 2020 by Bärenreiter-Verlag, Kassel Editions of Beethoven Sonatas ▪ 13

Joachim’s edition, however, is very reticently mark- klassischen Kompositionen, kein Springbogen zur An-
ed and can give little impression of how he performed wendung gebracht werden dürfe”] (Moser, classified
the music; a leading feature of his playing was spon- this among “violinistic habits and prejudices” [“gei-
taneity, and until late in life he was a very reluctant gerische Gewohnheiten und Vorurteile”]), he report-
editor.40 In this edition, he seems even more concern edly asked Mendelssohn whether he could employ
ed to present a clean text with minimal bowing and it in that repertoire and received the pragmatic advice
fingering than David, many of whose markings he to use it “if it is suitable for the specific place and
adopted, than to preserve an interpretation. As with sounds well.” 43.
David’s edition, the piano part is essentially that of Joachim’s edition in its original state is currently
the Breitkopf & Härtel Vollständige, kritisch durchgese- unavailable online. Some rather questionable altera-
hene überall berechtigte Ausgabe. Joachim’s five 1903 re- tions were made after Joachim’s death by Moser, and
cordings reveal a very striking disparity between the this version of the edition, which is the most com-
musical notation and his performance of it, especially monly circulated one, is referred to in the Commen-
with regard to rhythmic freedom. The biography of tary below as Joachim-revised.
Joachim by his colleague and former student Andreas The violin parts of the revised edition are accessible
Moser, and the Joachim and Moser Violinschule of 1905 on C H A SE : http://mhm.hud.ac.uk/chase/view/edition/
indicate that although Joachim deplored the bowing 499/
practices of later Franco-Belgian players, his own ap-
proach to bowing reflected the influence of the French
and Franco-Belgian practices that had been developed 8. Arnold Rosé
and gradually disseminated in the 1830 and 1840s.41 In (Wien, Universal, 1901)
particular, his use of springing bowstrokes in the mid- Arnold Rosé’s (1863–1946) career was largely centred
dle and lower half of the bow, which was not taught in Vienna, where he studied with Carl Heissler (1823–
in the classic Viotti School treatises, seems to have 1878), a pupil of Georg Hellmesberger Sr. He was cele
owed much to that influence.42 With Böhm, Joachim brated as a soloist and especially as a chamber music
had studied ‘brilliant’ French repertoire, virtuoso mu- player, as leader of the Rosé String Quartet. He was
sic by Heinrich Wilhelm Ernst and Josef Mayseder, also, for many years, concertmaster of the Vienna Phil-
and the caprices of Paganini, cultivating a formidable harmonic Orchestra until he was compelled, as a Jew,
technique that gave him, by the age of 14, the ability to flee to London after Hitler’s annexation of Austria.
to tackle the most difficult violin music of the day. Rosé’s recordings reveal a very different style of play-
Aware of the prevalent view among German violinists ing from those of Joachim and Joachim’s pupils Marie
in the 1830s and 1840s that “in Classical compositions Soldat (1863–1955) and Karl Klingler (1879–1971); in his
no springing bowstroke should be used” [“dass in solo recordings he used much more vibrato, though in
recordings with the Rosé Quartet, vibrato is less promi
Glück gehabt, drei sein ganzes Künstlertum aufbauende Jugend- nent. His expectations for the use of vibrato are sug-
jahre im innigsten musikalischen Kontakt mit seinem Lehrer erlebt gested in his edition by avoidance of the open strings
zu haben, jenem Josef Böhm in Wien, der mit Beethoven dessen
and natural harmonics that appear in most of the edi-
letzte Quartette aus dem Manuskript erstmals zum Erklingen ge-
bracht hatte. Immer klarer und bewußter erfaßte ich, welche Auf- tions by older violinists. His edition, however, still in-
gabe es gerade für den deutschen Geiger zu lösen galt in der ver- dicates considerable use of portamento.
ständnisgetragenen Gestaltung der Kammermusikwerke der ge-
waltigen Großmeister, jener unerschöpflichen musikalischen Welt,
Rosé’s violin parts can be accessed on CH ASE : http://
die sich in der damaligen Zeit noch so wenig Hörern wirklich er- mhm.hud.ac.uk/chase/view/edition/585/
schlossen hatte.”
40 See Clive Brown: “Joseph Joachim as editor” http://mhm.hud.
ac.uk/chase/article/joseph-joachim-as-editor-clive-brown/.
41 See Clive Brown: “The decline of the 19th-century School of 9. Carl Halir
Violin Playing,” http://mhm.hud.ac.uk/chase/article/the-decline-of- (Braunschweig, Litolff, 1905)
the-19th-century-german-school-of-violin-playing-clive-brown/.
42 For an extensive and stimulating appraisal of Joachim’s aesthet-
Carl Halir (Karel Halíř, 1859–1909) studied with An-
ics see Johannes Gebauer: “Der ‘Klassikervortrag’. Joseph Joachims tonín Bennewitz (1833–1926) in Prague and Joseph Jo-
Bach- und Beethovenvortrag und die Interpretationspraxis des achim in Berlin. He became second violin in Joachim’s
19. Jahrhunderts” (= Veröffentlichungen des Beethoven-Hauses Bonn,
Reihe I V, Schriften zur Beethoven-Forschung, ed. Christine Sie 43 Andreas Moser, Joseph Joachim. Ein Lebensbild (Berlin, 1898), p. 45.
gert), expected 2020. “wenn es für die betreffende Stelle passt und gut klingt”

Berlin quartet in 1897, but he also had a successful in- was also a cellist and conductor, studied piano first
dependent career as soloist and chamber music player. with Robert Freund (1852–1936) who had studied with
His edition follows closely in the David, Joachim tradi- Moscheles.
tion, but he provides many more additional markings, Auer’s violin parts can be accessed on C H A SE :
particularly to supplement articulation and dynam- http://mhm.hud.ac.uk/chase/view/work/307/
ics. In this respect, his edition complements Joachim’s
with information about the kinds of practices typical of
Joachim’s close colleagues. It is notable that Halir used 12. Arthur Seybold
unconventional slurrings in the piano parts, which (Hamburg, Benjamin, 1919)
he also edited, resembling the method developed in Arthur Seybold (1868–1948) studied at the Hamburg
Hugo Riemann’s (1849–1919) phrasing editions. Conservatorium. He joined the Meiningen Orchestra
Halir’s violin parts can be accessed at: http://mhm. in 1890. Thereafter he concentrated on violin teach-
hud.ac.uk/chase/view/edition/394/ ing and published a method entitled Das neue System
in 1913. His edition relies very heavily on the anony-
mously-revised version of David’s edition, following
10. Fritz Kreisler it in almost every detail. He adds many natural har-
(Mainz and London, Schott / Augener, 1911) monics, however, that are only implied in the David
Fritz Kreisler (1875–1962) studied first at the Vienna editions, revealing a very conservative German ap-
Conservatorium from 1882 to 1885 with Joseph Hell proach to vibrato.
mesberger Jr (1855–1907) before going to the Paris Con- Seybold’s edition can be accessed on I M SLP (in-
servatoire, where he studied with Lambert Massart cluded on the locations for individual sonatas).
(1811–1892), a pupil of Rodolphe Kreutzer, until 1887.
It was there that he developed his characteristic ‘con-
tinuous’ vibrato, which he recalled as an intensifica-
MET RONOME M A R K S A N D TEM PO
tion of practices developed by Henryk Wieniawski
and Eugène Ysäye.44 His edition shares with Rosé’s a Beethoven left no metronome marks for his Sonatas
general, but not total avoidance of open strings and for Piano and Violin. Here, and in the online Perform-
natural harmonics. His 1935 recordings of the sonatas ing Practice Commentary, at the beginning of each
reveal that he still used some of the traditional har- movement, tempos in analogous movements of works
monics. His tempos were generally much slower than for which the composer gave metronome marks are
those supplied by Czerny and Moscheles. considered.46 A list of metronome marks from 19th-
Kreisler’s violin parts can be accessed at: http://mhm. century sources of the violin sonatas is also included,
hud.ac.uk/chase/view/edition/525/ in which they are identified as follows:
Haslinger
11. Leopold Auer / Rudolph Ganz Sämmtliche Werke von L. van Beethoven (Mit Angaben der
(New York, Fischer, 1917) Tempobezeichnungen nach Mälzl's Metronom) (Vienna,
Leopold Auer (1845–1930) studied with Jacob Dont in Tobias Haslinger, 1828–).
Vienna from 1856. In 1861 he attended Alard’s classes Metronome marks intended to preserve a reliable
in Paris. It seems only to have been two years spent performing tradition, were included in this ultimate-
with Joachim in Hannover that, in his own words, ly incomplete collected edition of Beethoven’s works,
“opened before my eyes horizons of that greater art which was initiated in the year after Beethoven’s death.
of which until then I had lived in ignorance.” 45 It is They were to be provided, according to the prospectus
not surprising therefore that his edition reveals many for the edition, by Beethoven’s close colleagues Carl
similarities with Joachim’s; but it is more detailed in
46 For a more extensive consideration of the early metronome
its bowing and fingering, and contains additional per- marks in editions associated with Ignaz Moscheles and Carl
formance instructions. Rudolph Ganz (1877–1972), who Czerny see Marten Noorduin: Beethoven’s Tempo Indications, Diss.
University of Manchester, 2016, pp. 36– 46. Also Marten Noorduin:
“Re-examining Czerny’s and Moscheles’s Metronome Marks for
44 Louis Paul Lockner: Fritz Kreisler (London, 1951), p. 19. Beethoven’s Piano Sonatas,” in: Nineteenth-Century Music Review,
45 Leopold Auer: My Long Life in Music (New York, 1923), p. 63. vol. 15 (2017), pp. 209–235.
© 2020 by Bärenreiter-Verlag, Kassel Metronome Marks and Tempo ▪ 15

Czerny, Ignaz Schuppanzigh, and Carl Holz (1798– could be accessed for this publication; these metro-
1858). No information was given about who supplied nome marks are therefore given here from a second-
the marks for which pieces. Moscheles, however, in ary source.50
his English translation of Schindler’s Beethoven biog-
raphy stated, in relation to his metronome marks for Czerny-Vortrag
Beethoven’s piano sonatas, “the tempi that I have ven- Carl Czerny: Die Kunst des Vortrags der ältern und neuen
tured to give differ very slightly from those affixed Claviercompositionen oder Die Fortschritte bis zur neues
to Haslinger’s Vienna edition, by Carl Czerny, whom ten Zeit. Supplement oder 4te Theil zur grossen Pianoforte-
I consider a competent authority in the matter.” 47 It Schule (Vienna, A. Diabelli, [1846]).
is not clear whether Moscheles referred only to the Metronome marks are given for most of Beethoven’s
sonatas for piano solo, which will surely have been instrumental works that include the piano. Gustav
supplied by Czerny, or also to the sonatas with an Nottebohm, who knew Czerny, commented: “These in-
accompanying instrument, for which there may possi- dications, though not of authentic validity, can still lay
bly have been input from the violinists Schuppanzigh claim to a certain trustworthiness, especially for those
(who died, however, in 1830) and Holz. A later issue works, which we know Czerny either heard played by
of the edition includes some changes to metronome Beethoven or studied under his instruction. Czerny
marks, which are identified below as Haslinger 2. claims (on page 35 and 121 [in the fourth volume of his
Piano School op. 500]) that he tried to represent the
Moscheles-Cramer tempo that Beethoven himself took to the best of his
Moscheles (ed.): Beethoven’s Works. Complete Edition memory. Anyone who knew Czerny personally, who
(London, J. B. Cramer, Addison & Beale, [1834 –1839]).48 had the opportunity to observe his nature, which was
Moscheles wrote: “I hope I may be permitted to above all directed towards the practical, will believe
state, that in superintending for Messrs. Cramer & Co him capable of impressing firmly on his memory a
the new edition of his works, and in metronomizing tempo that he had heard, and will have noticed the
the several compositions, I have not merely listened certainty that he had in such outwardly tangible mu-
to my own musical feelings, but have been guided sical matters.” 51 As discussed in the commentaries,
by my recollections of what I gathered from Beetho- however, some of his markings seem to represent a later
ven’s own playing, and that of the Baroness Ertmann, opinion, conditioned by changing performance condi-
whom I have heard perform many of his works in his tions.
presence, and to his entire satisfaction, at the musi-
cal meetings [at Czerny’s] alluded to by Schindler in Czerny-Simrock
this work (p. 73) and Mr. Zmeskall’s. In some of the Sonates pour le Piano, composes … par Louis van Beet
quick movements, I have purposely refrained from hoven. Edition revue, corrigée, metronomisée et doigtée par
giving way to that rapidity of piano-forte execution, Ch. Czerny (Bonn, Simrock [1856–1868].
so largely developed at the present time.” 49 The last A few metronome marks for the sonatas with violin
comment is curious, since Moscheles’ tempos are, in were also included in this edition.
general, the fastest of all.
Alard/Diémer
Moscheles-Meyer École classique concertante[.] Oeuvres complètes de Haydn[,]
Further editions (probably based on the London edi- Mozart[,] Beethoven (Paris, Heugel [c. 1868–1870]).
tion), primarily of Beethoven’s piano sonatas, were pub- The metronome marks in this edition are generally
lished with metronome marks attributed to Moscheles. much slower than those from German sources.
According to the research literature, the publishing
house G. M. Meyer in Braunschweig printed, among
others, an edition of the Violin Sonata op. 23. No copy
47 Anton Schindler, trans. and ed. Ignaz Moscheles: Life of Beetho-

ven (Boston, n.d.), p. 145 f.n. 50 See Herbert Seifert: “Czernys und Moscheles’ Metronomisie
48 Alan Tyson: “Moscheles and His ‘Complete Edition’ of Beetho- rungen von Beethovens Werken für Klavier,” in: Studien zur Musik-
ven,” in: The Music Review xxv (1964), pp. 136–141. wissenschaft 34 (1983), pp. 66 and 77.
49 Schindler, trans. and ed. Moscheles: Life of Beethoven, p. 145 f.n. 51 Gustav Nottebohm: Beethoveniana (Leipzig, 1872), p. 136.
16 ▪ Metronome Marks and Tempo © 2020 by Bärenreiter-Verlag, Kassel

Speidel/Singer
ABBREV I AT IONS
Sonaten für Pianoforte und Violine von Ludwig van Beet-
hoven. Insbesondere zum Gebrauch in Konservatorien für AG (Alte Gesamtausgabe/old collected edition) Voll-
Musik revidiert und genau bezeichnet von Wilhelm Speidel ständige, kritisch durchgesehene überall berechtigte Aus-
und Edmund Singer […] (Stuttgart, J. G. Cotta, 1887) gabe (Leipzig, Breitkopf und Härtel, 1862–1865)
Many of the markings in this edition, especially for BW Beethoven Werke, Abteilung V, ed. Sieghard Bran
faster movements, are distinctly slower than Haslin- denburg (München, Henle, 1974)
ger, Moscheles, and Czerny. Vl Violin
Pno Piano
Kreisler/Rupp rh Piano right hand
These are derived from the recordings made by Kreisler lh Piano left hand
with pianist Franz Rupp in 1935. They show some strik-
ing similarities to those in the Alard/Diémer edition. Bar numbers and positions in the bar are indicated
as follows: 25ix indicates bar 25 note 9. Grace-notes
*** and other notes in small type are not counted in the
For each movement of the ten Sonatas for Piano and numbering.
Violin, the metronome markings given in the above References to “Reading between the lines of Beet-
sources are evaluated in relation to Beethoven’s tempo hoven’s notation/Beethovens Notation – zwischen den
conventions and his own metronome marks for other Zeilen gelesen” (in the edition) are given in the fol-
works. The quotations that immediately follow the lowing manner: (5/c/ii). This indicates Part 5, Section c,
list of historical metronome marks for each movement Sub-section 2. References to “Piano Techniques” (avail-
come from Carl Czerny: Die Kunst des Vortrags. Piano able only in this online Commentary) are similarly
forte-Schule op. 500, vol. 4 (Vienna, [1846]), pp. 77ff., given as follows: (PT: 1/a/ii).
from which the Czerny-Vortrag metronome marks are
also taken.
© 2020 by Bärenreiter-Verlag, Kassel Abbreviations ▪ 17

COMMENTARY
SONATAS OPUS 12 Moscheles-Cramer  = 92

Czerny-Vortrag  = 88
The ambiguity of slurring, staccato, and occasionally Czerny-Simrock  = 88
dynamics in the first edition, especially in the separate Alard/Diémer  = 152
violin part, indicates that all performances (apart, per- Speidel/Singer  = 76
haps, from those taking place with the participation Kreisler/Rupp  = c. 69–72
of the composer or under his supervision) must have Czerny advised: “The impetus always powerful and
depended to a considerable extent on the instincts and decided. In the second part [i.e. after the double bar]
choices of the performers, who would certainly have the eighths very legato and crescendo when they rise.
needed to interpret the notation to a much greater The following passage, deriving from the first bars of
extent than we are accustomed to do today. Only the opening theme, very light and short. The triplet
with the appearance of AG in the 1860s was there a accompaniment in both hands very soft, legato but
more consistent, though not always source-critically clear and speaking. The whole very lively, merry and
dependable text. In the case of sonatas for which the brilliant.” 52
sole source was the first edition, the editors were 1– 4. Vl: It is likely that many violinists of Beethoven’s
obliged to make choices based on their own musical time would have taken each of the broken-chord
judgment, and even where Beethoven’s autographs, or figures that follow the opening chord , analogous
corrected copies exist, his intention for the notation is with instructions in 18th- and 19th-century sources
often indeterminable. There is, therefore, no entirely to use  repeatedly for a succession of chords, even
reliable text. This tells us much about the composer’s where these follow one another without a break. At
expectations for the notation and for its execution. anything like the tempos indicated by Czerny and
His priorities were: 1) pitches, note lengths, and tem- Moscheles, which will surely have been intended to
po terms; 2) dynamics; 3) articulation (legato, portato, apply to the opening even if a more relaxed tempo
staccato). The latter, however, was only partially and were taken in some later passages, there will have
often very ambiguously indicated. Even when Beet- been little difference between these figures and the
hoven corrected copyists’ scores or proofs, it is evi- rapidly arpeggiated opening chord, especially be-
dent that he frequently overlooked mistakes. cause such upbeat figures were generally played
somewhat later and more rapidly than notated. Da-
vid marked  on each figure, but in David-revised
Opus 12, No. 1 the  signs were removed. Brodsky marked all the
Allegro con brio ascending figures  and the descending ones ; oth-
Tempo er early editors took them alternately  and .
Allegro con brio, as used by Beethoven in works for 1– 4. Pno: Pianists of Beethoven’s time would almost
which he gave metronome marks, indicates a very rap- certainly have given brilliance to the opening chord
id tempo, just short of Allegro molto. Analogous Allegro in 1 (and other similar instances) by making a tight
con brio movements containing a similar quantity of arpeggiation (notes almost together) from lowest note
16th-notes, for which Beethoven supplied metronome to highest note, holding down the chord for its full
marks, are the Second Symphony op. 36/i ( = 100), and length. Singer and Speidel mark tenuto (5/c/ii). Beet-
the String Quartet op. 95/i ( = 92). Moscheles’ mark- hoven would have expected the notes in the broken
ing is probably very close to what Beethoven would chord figures here and elsewhere throughout the
have given for this movement; his and Czerny’s mark- movement to be overheld (4/a/ii).
ing are significantly faster than has become conven- 52 “Die Bewegung stets kräftig und entschieden. Im 2ten Theile
tional during the past century and a half. die Achteln sehr legato und beim Aufsteigen anschwellend. Die
The op. 12 Sonatas were not included in the incom- nachfolgende, aus den ersten Takten des Thema gebildete P assage
leicht und kurz. Die Triolenbegleitung der beiden Hände sehr sanft,
plete 1828 – c. 1834 Haslinger collected edition. legato aber deutlich und sprechend. Das Ganze sehr lebhaft, heiter
und brilliant.”
18 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

5i. Pno: The position of p in the first edition is ques- vious one was; an analogous use of staccato marks
tionable. Since Vl is clearly marked with p on ii, it on long notes where ties are present can be seen in
seems unlikely that Beethoven expected a subito p the final movement of Mozart’s Symphony No. 41
in Pno. b. 84ff. On the other hand, the staccato mark may
5iii. Vl: Violinists of Beethoven’s time will most likely have implied a light accent, so that the bow change
have remained in 1st position here, but the use of is clearly audible. It evidently troubled 19th-century
harmonics in such contexts is also well documented editors: David, Joachim, Auer, Halir, Rosé, Seybold
in early-19th-century sources and would have been replace the dot with a tenuto line, Singer with a line
a plausible choice at a time, when little or no vibrato over dot, Kreisler omitted the staccato mark alto-
was expected. Alard marks a harmonic. Most edi- gether and played a full-length note in his 1935 re-
tors made no specific recommendation for finger- cording.
ing. 12. Vl: The short slurs, necessitated by the repeated
5ii–iii, 7ii–iii, 142ii–iii, 144ii–iii. Vl: At this date Beet- note, may imply a nuanced performance with greater
hoven may have assumed a slurred execution with- agogic accentuation of the first note, with the others
out the necessity of marking it. This is strongly im- slightly hurried (2/a).
plied by the notation of the equivalent figure in both 14, 16, 18, 151, 153 and 155. Pno: Beethoven probably
Vl and Pno in bb. 106–120, where the upbeat is given intended legato for the descending arpeggio figures.
as a grace-note, which, even without a marked slur, Diémer, Reinecke, Halir, and Vogrich add slurs. Halir,
was always expected to be slurred to its following however, adds dots under the slurs.
note (4/c/i). The use of slurring in many passages 15, 152. Pno: The first note of the grace-note turn align-
where none is marked, was certainly envisaged in ed with the bass (5/a).
Viennese string music of this period. None of the 17–18. Vl: The one-bar slurs of the 1st edition probably
editors supplement Beethoven’s notation, but in the represent Beethoven’s notational intention, for the
absence of staccato, and by analogy with the related sake of an effective crescendo, and execution of the
figure in Pno, which has a slur (13, 15, 17, 154, though staccato note on 19i in the middle of the bow. AG
missing in 150, 152), a legato execution was surely and many later editions, including BW, however,
expected. print a two-bar slur by analogy with the preceding
5–11, 13–18, 27–31, 142–148. Pno, Vl: Beethoven’s slurs passage.
coincide with the phrase units, but they were evi- 19. rh: The turn should probably start on the upper
dently conceived as indicating a continuous, seam- auxiliary and be left as late as possible so as to be
less legato, since at 142–148 Pno has a single long appended to ii (5/b/i)
slur. Nevertheless, Beethoven would scarcely have 21, 23, 25. Vl: Although AG (following the 1st edition)
expected a skilled performer to play the successive has a slur over all eight 16ths, which probably cor-
8th-notes rhythmically exactly, but to add subtle in- rectly represents Beethoven’s intention for the no-
equality, perhaps lingering on harmonically signifi- tation, David’s edition (followed by Joachim, Auer,
cant notes. The early piano editors including Reine Halir, Rosé, and Kreisler) marks the first note stac-
cke, Speidel, Vogrich, and Ganz mark a continuous cato and begins the slur on the second 16th; this is
slur from 5–11; AG and most editors mark continu- followed in some piano parts.
ous Pno slurring from 6–11. Beethoven’s Pno slurs lh: Given that Beethoven slurred the correspond-
might indicate over-legato touch and sustaining of ing violin figurations, it is curious that he did not
notes that belong to a single harmony (4/a/ii). slur the lh, though he surely expected legato. Some
5–20, 30, 33, 34, 37, 38, 71, 77–81, 100f, 142–157, 158, early editors add slurs from the second 16th, match-
159, 225f. Pno: All chords (apart from very short or ing their violinist colleagues’ treatment of the slurs.
marked staccato) could be swiftly arpeggiated (5/c/ 21–24. Pno: Beethoven probably expected the descend-
iii; PT: 1/a). ing broken chords to be played legato, overholding
10ii, 147ii. Vl: The purpose of the staccato mark here as many notes as possible to create ample resonance
is debatable (4/c/ii). A significant shortening of the (4/a/ii). Some early editors added slurs. Speidel marks
note in this context seems improbable. It may per- 21ii–22xvi leggiero but also indicates sustaining ped-
haps have been intended merely to emphasise that al throughout.
this note was not tied across the barline, as the pre- 26. Vl: David, followed by all except Alard and Brodsky,
© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 19

marks separate bows for the 16ths. There is no tex- solutions. Speidel marks espressivo which may indi-
tual justification for this in the sources. Although cate both a slower tempo and agogic accentuation
such liberties, for the sake of an effective delivery, of the highest notes. Halir adds tenuto lines to his
were typical of the period, as they would have been unusual slurring patterns (perhaps suggesting ago-
in the early 19th century. Separate bows, however, gics) on 51 iii, 52 i and vii, and 54 i. Ganz marks the
seem unlikely at anything like the rapid tempo passage non troppo legato.
marked by Moscheles and Czerny, which was al- 50i–ii. Vl: David and Joachim give no fingering and
most certainly close to what Beethoven envisaged. presumably envisaged 1–0 on the A-string; a like-
28iii–viii. Vl: Some editors indicate half position, a ly choice for early-19th-century performers. Many
probable choice for an early-19th-century violinist; might also have gone to 3rd position on 49viii, as
others go to 3rd position. do several of the editors.
31i. Vl: Early-19th-century violinists would probably 51i, 176i. Vl: Alard and Halir mark a harmonic, others
have used the open E-string. All the editions ex- seem to have expected 1st position. Both were dis-
cept Auer mark 0, which is completely unproblem- tinct possibilities for early-19th-century performers.
atic with a gut string, but would need more careful 51–53, 176–178. Vl: In this passage the sfs are surely
management with a metal string. within a piano dynamic; at this period in his output
35, 39. Pno: Asynchrony in which rh ii is played slight- Beethoven did not yet use sfp (as he did for instance
ly after lh would give the melody note special prom- in the Scherzo of op. 96) to specify accents within
inence (5/c/ii; PT: 1/b). piano; he generally used sfp only where a previously
42i. Vl: A harmonic would probably have been used forte dynamic was followed by piano. This was rec-
by most violinists of Beethoven’s time, as marked or ognised by 19th-century violinists: Singer provides
implied in almost all the editions. a footnote: “The whole passage light and p, the sf
43–50, 168–175. Pno, Vl: In addition to legato, Beetho- not too strong”.53 Hermann marks sfp instead of sf,
ven’s slurs probably imply a nuanced performance Auer adds p before the first sf, and Halir not only
with the first note of each slur slightly lengthened marks p but also changes the sfs to > .
and the following notes accelerated somewhat to 55–57. Vl: It is probable that a bowing in the upper
make up the time (2/a). Halir added tenuto lines on half was envisaged. All the editors except Brodsky,
Pno 43 iii, 44 i and v, 45 i and v and 46 i perhaps who marks  for 51i, seem to have expected , but
indicating agogic nuance, but, curiously, he did not it is clear from David’s and Singer’s marking of up-
mark them in the corresponding passage in Vl 47–50. bow staccato in 55f after the slurs on the triplets (re-
Given the change of character and the fact that the moved in David revised), that they began that pas-
accompaniment (43– 44 and 47– 48) is momentarily sage near the point of the bow. Singer also marks five
silent, the tempo might be slightly slackened here. successive up-bows from 53iii–54iv. At the tempos
Interestingly, Speidel marked a continuous slur from suggested by Moscheles and Czerny, most early-19th-
43ii– 45viii but also added espressivo in 43 (as Singer century violinists would probably have executed the
did in Vl at 47), perhaps implying both a nuanced passage with short, on-string strokes in the upper
style (with occasional agogic accents) and a slightly half of the bow.
broader tempo. 56iv–vi. Vl: Most violinists in Beethoven’s time and
47. Pno: Given Beethoven’s p, the chord might be ar- throughout the 19th century would have executed
peggiated somewhat more slowly and gently than all three notes with the 1st finger; a procedure con-
1i (5/c/ii). demned in the 20th century by Carl Flesch.
49. Pno: The portato signifies that the chords should 58–70, 183–195. Pno: In addition to legato, the slurred
be played with only a slight separation almost le- broken-chord figures imply overholding (of those
gato and invites very slight arpeggiation from low- notes that form stable harmonies) to create reso-
est to highest note as recommended by Moscheles nance (4/a/ii); this might also be aided by judicious
in his Studies op. 70 (1827) (4/b/i). use of the sustaining pedal (PT: 3/c). Beethoven’s
50–54. Pno: Beethoven’s articulation in Vl 54 –57, sug- two-bar slur groupings (retained by most early edi-
gests a similar pattern here, though it is likely that tors) do not necessarily have implications for nuance
he expected the passage to be predominantly le-
gato. The early editors give a range of stimulating 53 “Die ganze Stelle leicht und p, die sf nicht zu stark.”

or phrasing. AG, Diémer, Reinecke, Halir, Vogrich as well as < > with the apex on the first of
indicated legato, marking various lengths of slurs. each slurred pair in Pno 71–73 and 78 and paral-
59, 63, 65, 67, 69, 184, 188, 190, 192, 194. Vl: The 16ths lel passages. In Vl 75–77 Halir marks tenuto lines
immediately before the barline have neither a stac- under slurs for the three 8th-notes preceding the
cato mark nor a slur to the following note. Like the slurred pairs of quarter-notes (also in parallel pas-
upbeat 16ths in 5ii–iii, 7 etc., it is very likely that vio- sages) perhaps implying broadening. Rosé marks Vl
linists of the time would have slurred them. Beet- iii–v in 75–77 with tenuto lines, perhaps also imply-
hoven’s expectation that this was obvious may be ing broadening of tempo. Kreisler and Rupp (1935)
strengthened by the curious notation employed in make a distinct slowing-down in the passage from
Pno in 65 where, for notational convenience, he in- here until 83/208, as indicated by Speidel/Singer.
dicates a grace-note after the barline, rather than a 71ii–73i, 196ii–198i. Vl: Several editors envisaged por-
16th before it. In any case, the 16ths were probably tamento effects, most notably Singer, who indicat-
expected to be played lightly, and later than their ed them in both places on the rising figure. Many
notated position, like pre-beat grace-notes. marked a harmonic g1 in 197. These would not have
61, 186. Vl: In such contexts, where slurring was en- been unusual fingerings in Beethoven’s Vienna (5/c/
visaged, it was often unmarked. The turn itself, of iii).
course, requires a slur, but Beethoven surely did 75ii–iv, i–iii in 76 and 77, 200ii–iv, i–iii in 201 and
not expect it to be separated from the main notes, 202. Vl: Early-19th-century violinists might either
which it connects. AG merely adds a slur over the have used separate bows or slurred staccato/portato.
grace-notes themselves (but see the Critical Report David, Alard, Joachim mark slurs over dots; Halir
for 186). The 19th-century violinist editors, appar- and Rosé mark tenuto lines under slurs.
ently for reasons of convenient bow division, indi- 77i. Vl: A harmonic would be a very likely option.
cate a bow change after the turn. Alard, Brodsky, Halir, Kreisler mark one.
64, 66, 189, 191. Vl: Harmonics may well have been 79–81, 204 –206. Pno: Beethoven surely expected the
used here by early 19th-century violinists; they legato 8th-note figurations to be played in a nu-
are marked by Alard, Halir, Seybold. David and anced fashion, perhaps lingering on the first note
Joachim, who provide no fingering, very likely en- (dissonant with the underlying harmony) of each
visaged them. In 64 and 189 it could be effective to sighing figure (2/a): rh 79 and 80 iv–v and vi–vii, 81
begin the note as a harmonic and then to increase iv–v and vi–vii, 82 i–ii, iii–iv, v–vi. Halir changes
finger pressure, perhaps applying an accelerating Beethoven’s slurs to bring out the sighs, giving occa-
vibrato to support the crescendo (5/c/iv). sional tenuto lines that suggest agogic accentuation.
63–70, 184 –193. Vl, Pno: The 16th-note upbeats in both Ganz marks < > , suggesting agogic nuance
instruments and the grace-note in Pno in 65 are on rh iv–vii in 79 and 80, and espr. in 81. Beetho-
clearly intended to have the same effect. Since the ven may well also have expected a gradual return
grace-note was conventionally expected to be slur to tempo with cresc. (if the tempo had slackened
red to its main-note, Beethoven probably expected in the previous passage). Speidel marks poco a poco
all these upbeats to be slurred. in tempo at 79 to compensate for his earlier etwas
71–73, 75–78, 196–198, 200–203. Pno, Vl: The expres- zurückhaltend at 71.
sive character of the slurred pairs of quarter-notes 81f, 206f. Vl: Beethoven’s intention was surely for a
suggests a nuanced treatment, with the first note connected legato (see Critical Report for 82 and 207),
stronger and longer and the second shortened and with or without bow change. All editors except
softer (2/a). A slight slackening of tempo would cer- Alard and Hermann chose to change bow after 82ii/
tainly be in keeping with Beethoven’s decresc. in 72, 207ii to sustain an effective forte.
76, 197 and 201. Speidel marks a tenuto on rh ii in 83i, 208i. Vl: The change of dynamic might well be
71–73 and 78, and Vl ii in 75–77 implying a some- enhanced by a harmonic. Alard marks one.
what lengthened note, as well as etwas zurückhaltend 83f, 208f. Pno, Vl: The trill was probably expected to
(somewhat slowing down) at Pno 71 and Vl 75, and commence on the upper auxiliary note, since the pre-
smorzando (dying away) in Pno 78. Halir suggests vious passage ends with the trill’s main note (5/b/ii).
a nuanced performance with tenuto lines implying Ganz marks an upper auxiliary with a dotted line
agogic lengthening on the first of each slurred pair, aligning it with the bass note. Reinecke, Speidel,

however, mark 2–3 fingering indicating a main-note and Brodsky. For the performance of < > see
start, a practice that became more common in the 3/b/v. Speidel and Singer mark espressivo and Halir
second half of the 19th century. The trill in Vl at 85 adds many more < > in the Vl part.
and 210 was probably also expected to begin with 110ii, 114ii, 118ii, 122ii. Vl: Although Pno consistently
the upper note. has grace-notes in 108, 112, 116, 120, Vl has one only
87f. Vl: For the significance of the staccato marks see in 106. Whether through Beethoven’s, a copyist’s, or
note to 10ii above; here too the 19th-century editions an engraver’s oversight, it is very likely that they
contain similar editorial markings. were omitted in error. In any case, the addition of
87–90. Pno: According to Czerny, chords that are short a very rapid grace-note in those circumstances is
should not be arpeggiated, though it is likely that he effective, and it would have been well within the
did not consider very tight (almost imperceptible) remit of a late-18th-century performer to add one.
spreading of notes, which would enhance energy, to 114ii, 118ii. Vl: A harmonic, as marked by Alard, Auer,
be arpeggiation (5/c/ii). Beethoven may not have ex- Brodsky, would have been a plausible late-18th-cen-
pected these chords to sound extremely short, but tury choice here.
simply separated from each other and perhaps ac- 126–136. Vl: The slurring in the carelessly-engraved
cented in the way that a violinist of Beethoven’s era violin part of the 1st edition may well be inaccurate.
would employ a sharp, on-string stroke. Speidel and Alard and Joachim make no changes; but it seems
Halir mark each chord in 87–88 with staccato dots highly unlikely they played it thus in practice, since
and tenuto lines, and Rosé adds > in both rh and it would be difficult to avoid inappropriate accen-
Vl on 87 ii and 88 i and ii. tuation on the staccato 8th-notes. All other editors
91–92, 216–217. Pno: Beethoven surely expected lega- suggest changes. David, Auer, Seybold, Kreisler ar-
to, perhaps with a slight quickening of pace in line ticulate the staccato note in the same bow as the fol-
with his cresc. and the music’s character. Diémer, lowing slur; Hermann, Singer, Brodsky, Halir, Rosé
Reinecke, and Vogrich add slurs. include the staccato note in the preceding slur. The
93, 95, 97. Pno: The first note of the grace-note arpeg- latter, which may be close to Beethoven’s intended
gio should probably be aligned with the lh chord, slurring, seems a very credible solution, since Pno
which might also be tightly arpeggiated enhanc- consistently slurs across the barline except in 130,
ing the effect of sf while avoiding harshness (5/c/ii, 134. For the performance of < > see 3/b/v.
PT: 1/a). 157i. Vl: Almost all editors take this note with the
94, 96, 219, 221. Vl: Almost all editors change bow 3rd finger. A 3rd-finger harmonic was marked by
after the tied note, but at the rapid speed almost Alard.
certainly envisaged by Beethoven, this would be 163–166. Pno: All octaves not marked staccato might be
unnecessary. tightly broken to enhance energy, a practice notice-
96f. Vl: No editor before Kreisler suggests half posi- able in sources such as Cipriani Potter’s mid-19th-
tion; the others either marked or assumed sliding century editions of piano sonatas by Mozart and
the 1st finger for the d # and g #, in the typical 18th- Beethoven (5/c/ii).
and 19th-century manner. 164f. Vl: The majority of editors remain in 1st position
99, 223f. Vl: The quarter-notes with staccato marks using an open D-string; Alard, Singer, Halir, how-
were certainly not intended to be played too short ever, mark the passage in 3rd position, after a 4th-
(surely not with a thrown stroke near the frog), mere- finger harmonic on the last note of the preceding bar.
ly with a powerful slightly separated détaché (indi- 171f. Vl: Most violinists of Beethoven’s time are likely
cated here in the edited part with tenuto lines and to have favoured 1st position; Alard, Singer, Brod-
staccato marks). sky, Rosé mark the passage on the A-string.
107f, 115f, 119f, 123f. Vl: Although the slurs in the
separate violin part of the 1st edition never cross Thema con Variazioni
the barline, the equivalent slurs in the more care- Andante con moto
fully marked piano part do, and it seems unlikely Tempo
that Beethoven intended a difference. AG extend- For the 2/4 Andante con moto of the String Quartet op. 18
ed these slurs across the barline and this was fol- no. 3 Beethoven gave the metronome mark  = 92, but
lowed by all the 19th-century editors except Alard that movement has a considerable number of 32nd

notes, which may explain the somewhat faster tempo expressive inequality of rhythm (4/b/i). Halir marks
suggested by Moscheles and Czerny for the theme of a tenuto line on 3i suggesting agogic accentuation.
this movement which contains no 32nds, though there For 4i two alternatives offer stylish possibilities:
are many in Variation 2. Although Kreisler plays the 1) play the bottom lh note on the beat and other
whole movement at an almost constant tempo, there lh and rh notes unarpeggiated slightly after, or
is no reason to think that Beethoven would have ex- 2) swiftly arpeggiate both hands from bottom note
pected this; Variation 2 might be a little slower and to top note (5/c/ii and PT: 1/a), 4iii unarpeggiated
Variation 3 somewhat faster. For Variation 3, in fact, and softer than 4i, and 4iv gently arpeggiated. Ad-
Moscheles gives a considerably faster tempo, while ditionally, a slight crescendo through 3 and a cor-
Rosé marks poco più mosso, and both indicate tempo 1o responding diminuendo through 4 (as marked by
for Variation 4. Speidel and Halir) is musically effective. This could
Theme Var. 3 be enhanced with subtle tempo inflection. Beetho-
Moscheles-Cramer  = 108  = 138 ven may have intended 4 to continue with portato
Czerny-Vortrag  = 108 articulation as suggested by Ganz.
Czerny-Simrock  = 104 In bars 5–8 each downbeat can be played with rh
Alard/Diémer  = 92 slightly after lh, perhaps with the longest delay at
Speidel/Singer  = 96 the height of the phrase on 7i, thereby enhancing
Kreisler/Rupp  = 92 the expression of the major 7th interval (PT: 1/a).
Czerny advised that “The theme should be moderate- In 5 and 6, rh iv could be played slightly after lh
ly slow and the beautiful melody nicely expressive.” 54 (4/b/i). Additionally, 5i–7ii might be played slightly
1–8. Pno: The phrase structure of the theme is 2+2+4 crescendo and with forward momentum, and 7iii-8
bars. Although Beethoven marked an initial p, he dying away and easing momentum. Speidel marks
undoubtedly required the pianist to make varying poco cresc. at the end of 5 and > in the second
shades of dynamics, which would at times be en- half of 7. Similarly, Halir marks < from the end
hanced by gentle ebb and flow of tempo and flex- of 5 to 7i, then 7ii f followed by > until the end
ible placement of all notes. In addition, a sophisti- of 8. The final chord in 8 unarpeggiated.
cated application of arpeggiation and asynchrony 8–16. Vl: The natural instinct of 19th-century violin-
would result in a ‘beautiful’ and stylish interpreta- ists was to take upbeats ; this, however, brings out
tion; an over-strict adherence to the notation would the sf in 10 on , the portato in 11 , and the cadential
undobtedly have seemed inartistic to Beethoven appoggiatura in 12 . The traditional ‘rule of down
and his contemporaries, as would an exaggeratedly bow’ on strong beats was already challenged by Tar-
mannered one. tini and other 18th-century violinists. As comment-
Given the dolce character of the theme (Halir marks ed in 1798, in an announcement of the 15th reprint-
rh dolce), a moderately fast arpeggiation of 1i and ing of J. B. Cartier’s edition of Corelli’s Sonatas op. 5:
swifter arpeggiation of 2i (or perhaps a swift ar- “it is a mistake to believe that it is always necessary
peggiation of lh with rh notes played together and to use an up-bow on an upbeat and a down-bow
slightly after lh) is appropriate for an sf accent with- when playing the strong beat of the bar.” 55 Three
out causing harshness of tone (PT: 1/a). This will editors (Alard, Brodsky, Halir) chose to begin .
naturally give rise to a rhythmically flexible render- More significant is the division between those who
ing of important melody notes. Halir marks a tenuto sought to retain the colour of particular strings, with
line on rh 1i possibly indicating agogic lengthening. the concomitant use of portamento, and those who
A crescendo through 1 and diminuendo through 2 is preferred a simpler, more direct character, by remain-
indicated by Speidel and Halir; and rh 2ii–iii could ing in 1st position throughout. The former includes
be altered to a dotted figure (long/short). To render Alard, Singer, Auer, Brodsky, Halir, Rosé, Kreisler;
the portato chords in 3 expressively, each rh chord the latter David, Joachim, who mark no fingering,
might be played very slightly arpeggiated, or the rh and Hermann, Seybold, who mark some open strings.
chord might be played slightly after the correspond-
ing lh note, which will give rise to a gentle long/short 55 Quoted in Lionel de La Laurencie: L’école française de violon,
de Lully à Viotti; études d’histoire et d’esthétique (Paris, 1923) vol. I I ,
54 “Das Thema mässig langsam, und die schöne Melodie wohl p. 314. “C’est une erreur de croire qu’il faille toujours pousser l’ar-
ausgedrückt.” chet au levé, ou le tirer en frappant la mesure.”

The older mainstream German players, therefore, 24 –32. Vl: In the second part of the theme some edi-
seem to have favoured a simpler treatment of the tors (David, Joachim, Auer) maintain 1st position
melody in this respect, though whether in their until 29; most, however, favour the A- and D-strings
own practice they used this straightforward finger throughout, with Singer even explicitly marking a
ing is indeterminable. Some late-18th-century vio- portamento between bow strokes from 27iv–28i. All
linists were already making considerable use of ex- shift 1–1 from 29ii–iii, except Rosé, who shifts 3–3
pressive portamento (5/c/iii). from iii–iv. A harmonic is marked on 31v by all ex-
The earlier editors evidently expected little or no cept David and Joachim, who probably considered
vibrato; Alard, Singer, Brodsky mark a harmonic on it obvious.
the sf in 10 and Alard, Singer also in 12 (5/c/iv). 27f. Vl: The portamento between the bowstrokes is a
8–16, 23, 29–32. Pno: In addition to legato, the slurs in plausible late-18th-/early-19th-century vocal effect,
rh invite overholding of notes belonging to a single intensifying the legato and the crescendo. If used, it
harmony (4/a/ii). To enhance texture and expression, should probably be left late and made fairly rapidly.
the lh octaves might be swiftly arpeggiated accord-
ing to individual taste, and asynchrony between rh Var. 1
and lh would be musically effective for example on Czerny suggested that one should play the first vari-
first beats of measures and at moments of strong dis- ation “with feeling, but not slower.” 56
sonance (PT 1/b). The 16th-notes in rh will gain ex- Pno: A singing execution is apt for this variation.
pressive effect through rhythmic inflection, linger- Reinecke marks dolce; Speidel marks cantabile. While
ing on notes of harmonic or hierarchical importance Beethoven’s overall dynamic mark is p, he undoubt-
and moving more quickly through less important edly expected subtle dynamic inflection to enhance
ones. An exact rendition of the notated rhythms the character and undulation of the melody. Speidel’s,
would undoubtedly have been considered “correct”, Vogrich’s, and Halir’s additional dynamic marks may
but rigid and unsophisticated to musicians of Beet- provide inspiration. In the same way as the main
hoven’s era (2/b). theme, passages of smaller-value notes should be
17–18, 25–26. Pno: Chords marked staccato should rhythmically inflected. Halir marks tenuto lines on
generally be unarpeggiated (5/c/ii); chords marked the first of each group of demisemiquavers in 46
portato could be very slightly arpeggiated (4/b/i). and 47 presumably signifying agogic inflection. All
19–25, 27–28. Pno: In a similar manner to bar 4 of the chords in lh should be arpeggiated, those marked sf
theme, the chords could be arpeggiated in various in 37 and 38 could be swiftly so, while others more
ways. slowly according to mood and textural needs (5/c/
21–22, 29–30. Pno: In addition to dynamic nuance, the ii). Asynchrony between important melody notes
double hairpin < > might signify a slight has- and bass would undoubtedly have been a require-
tening towards and lingering at the apex (22i and ment of a sweet or singing style in Beethoven’s era
30i) and a return to tempo afterwards (3/b/v). (3/d/i; 5/c/ii; PT: 1/b).
24 –25. Pno: The turn sign over the barline should surely Vl: Singer’s additional dynamics suggest the kind
be realised lightly, with an upper-note start (5/b/1). of natural inflections with which a sensitive 19th-
Reinecke writes out the effect as a quintuplet turn century violinist would have supported the piano’s
starting on the main note, but no doubt expected a elaboration of the melody.
rhythmically flexible rendering. 34. Pno: The first note of the grace-note turn should
20, 28. Pno, Vl: The rinf /rf is probably Beethoven’s in- surely be aligned with lh (5/b/i).
tention, despite Pno having sf in the 1st edition at 28 34 –35, 36–37. Pno: The turn, over the bar line, should
(see Critical Report). Following immediately from be realised in the same way as 24 –25. The > in
cresc., it was probably intended as an instruction 34 rh iii–iv and vii–viii perhaps invites asynchrony
to intensify the crescendo all the way to the third between melody and bass on iii and vii.
quarter-note beat, which, however, as the resolution 39. Pno: The slurred 16th-note pairs could be inflected,
of an appoggiatura, was probably expected to be the first longer than the second (2/b). The short trill
somewhat quieter (3/b/ii). AG, however, changed signs might be realised as a Schneller comprising
rinf./rf to sf in both places, and this was followed in
all the performing editions. 56 “mit Gefühl, aber nicht langsamer.”

three notes in the order main/upper auxiliary/main rh played slightly after lh would add much to the
(5/a/iv). expression with a softening effect that both con-
40, 48. Pno: It is self-evident that the first note in rh be trasts with the sf on rhii and enhances the >
played with more emphasis than the second: Spei- (5/c/ii; PT: 1/b).
del marks the first with >; Halir with a tenuto line. 64. Pno: The final chord unarpeggiated (5/c/ii).
41. Pno: Beethoven’s sf invites both a sharp accent and
a slight delay of rh ii after the corresponding note Var. 3
in lh (PT: 1/b). Czerny advised: “In the 3rd Variation every second
44. Pno: The grace-note in rh indicates a lower-auxil- bar ff and with Pedal. The whole bright and hard.” 58
iary start to the trill which should be aligned with Czerny’s word “grell”, translated here as “bright”
the bass (5/b/ii). could also mean strident, grating, or harsh.
45– 47. Pno: The first of each group of four slurred Given Czerny’s description it is conceivable that
32nd-notes in rh might be given slight emphasis Beethoven expected a somewhat faster tempo. This
and agogic nuance to bring out the expressive dis- would correspond with Moscheles’ additional metro-
sonance (2/a). In 46 and 47 Halir marks the first of nome mark and Rosé’s poco più mosso.
each group with a tenuto line. 64f. Vl:  for the upbeat is virtually inevitable here
and is marked by all except Alard, who probably
Var. 2 considered it too obvious to indicate.
Here Czerny suggested: “The 2nd Variation light, pi- 65f. Pno: All broken chords not marked with staccato
ano, short staccato in the bass and everything appro- dots could be executed with notes overheld to cre-
priately corresponding with the brilliant violin vari- ate resonance, which may be aided by subtle use
ation.” 57 of the sustaining pedal (4/a/ii). Diémer and Speidel
As Czerny advises, all Pno bass notes (octaves) mark- add slurs to all broken chords.
ed with dots should be played as short staccato, and 65, 67. Vl: All remain in 1st position, as Beethoven
therefore probably not arpeggiated. In rh the slurred probably expected, except Alard and Singer, who
broken chords can be overheld as far as possible to introduce shifts for tonal and expressive purposes.
enhance resonance and to obviate the need for sus- 66f. Pno: The first note of the grace-note slide should
taining pedal (4/a/ii). be aligned with the bass, but as far as possible the
51. Vl: Several later editors (Hermann, Brodsky, Halir, main note should receive the accent (5/a/iv). Beetho-
Kreisler) mark a harmonic for the e3. Alard, David ven may have intended a decrescendo to p through
and Joachim, who mark no fingering, probably con- the bar as suggested by Halir, who marks > , in-
sidered this obvious, as many violinists of Beetho- stead of a subito p in the following bar.
ven’s time would have done. 69f. Pno: The effect of the lh octaves marked sf can
52i–ii. Vl: Many editors, from David and Alard on- be particularly enlivened by very quick and almost
wards, mark 4 – 4, which was a common expressive imperceptible arpeggiation (5/c/ii).
fingering in the early 19th century. 71f. Pno: For the syncopated figure on rh ii marked
55xi–xii. Vl: All except Alard and Brodsky, who break sf (which appears as an octave in 71 but as single
the slur, change the finger between these two notes, note in later occurrences), slight asynchrony with
all 2– 4 except Singer (2–3); this too would have been the corresponding bass will produce heightened
a quite normal expressive fingering in Beethoven’s energy (5/c/ii).
time. Examples can be seen in Rode’s 14th Caprice 72f, 74f, 88f. Vl: Two approaches to the execution of the
(b. 1) and Kreutzer’s 10th Étude; it is also taught in crescendo half-notes are illustrated by the two ear
Spohr’s Violinschule (p. 175f.). liest editions. David, evidently executing the 16th-
58i. Vl: All except Auer, Halir, and Kreisler employ note triplets in the upper half of the bow, includes
the open string. a slurred staccato  on the last two notes to permit 
58–60. Pno: The lh octaves not marked staccato might from near the point for the subito p crescendo; Alard
be swiftly arpeggiated (5/c/ii). In 60, asynchrony with and others, evidently playing the 16ths lower down
57 “Die 2te Variation leicht, piano, den Bass kurz abgestossen, und 58 “In der 3ten Variation jeder zweite Takt ff und mit Pedal. Das
alles mit der brillanten Violin-Variation wohl übereinstimmend.” Ganze grell und hart.”

the bow, execute the half-note  and retake for the also have encouraged the violinist to use higher posi-
next bar, but with an unavoidably longer break. tions on the lower strings, and harmonics, as indicat-
Only Seybold adopts David’s solution. Kreisler’s re- ed by many of the 19th-century editors. If the tempo
cording contains breaks of almost an 8th-note. of the previous variation was faster, there would
73f. Pno: Whether Beethoven expected the sustaining probably have been a return to the opening tempo, as
pedal to be employed to make the rh octaves sound indicated by Rosé, or perhaps an even slower tempo.
legato is unclear. Of the early editors, only Ganz Beethoven will have expected a sophisticated and in-
marks pedal throughout such bars. If the choice is dividual dynamic shaping beyond his more general
not to use sustaining pedal, which might accord markings. Halir’s suggestions may provide inspira-
with Czerny’s description, then gentle (moderately tion.
slow) arpeggiation of the rh octaves will to create a 97–100. Vl: Alard, Hermann, Singer use the D-string
sustaining effect. and with a harmonic on 98ii; David, Joachim, Sey-
76f. Pno: The rh chord will be greatly energised by a bold mark nothing. Rosé, Kreisler evidently expect-
very quick, strident arpeggiation (5/c/ii). ed vibrato.
77f. Pno: The rh chords could be sharply accented and 101i–ii. Vl: A harmonic is marked by Alard, Singer.
played without arpeggiation. However, very slight 103, 111, 135. Pno, Vl: In addition to a dynamic effect,
asynchrony with rh played after lh will enhance the the < > might indicate a slight lingering at
gruffness of the sf (5/c/ii). the apex (3/b/v).
85f, 93f. Vl: The staccato triplet 16ths naturally suggest 104ii, 112ii. Vl: Alard, Singer, Joachim, Auer, Seybold
a springing bowstroke to modern players, but it mark a harmonic, which involves sliding the 4th
seems very unlikely that any early violinist would finger from the d #3. It is implied in David and Her-
have considered such a stroke here. Auer and Halir mann. Only the later violinists Brodsky, Halir, Rosé,
even mark tenuto lines throughout these bars. Start- Kreisler, who tended to favour a lower finger for the
ing with a short stroke near the point, the violinist d #, perhaps to facilitate vibrato, employed 4th or 5th
would broaden the bowstroke until using a sub- position and a stopped e3.
stantial proportion of the bow by the end of the 104, 108, 112, 116, 124. Pno: The chords on rh ii unar-
crescendo. peggiated (3/c/ii).
90, 92. Vl: Only Rosé offers a fingering that avoids 113f. Pno: All rh chords marked sf can be further ener-
shifting, or using the same finger for pairs of ad- gised with a very swift arpeggiation (5/c/ii; PT: 1/a).
jacent semitones; he marks the passage to start in 117–119. Vl: The majority of editors mark a harmonic
2nd position, contracting to 1st on iv. Others start for the d2 in 119 (Alard and Singer also for the one
in 3rd and either provide no guidance thereafter, in 117). Singer’s fingering is typically rich in porta-
or offer various solutions involving the use of the mento implications with the last three notes of 117
same finger consecutively. It is possible that a few taken with the 3rd finger and 119vi–vii also 3–3.
violinists of Beethoven’s time might have employed 121, 122. Vl: The portamento suggestions in the edited
Rosé’s fingering; such fingerings are rarely encoun- violin part of this edition, sensitively and not too
tered in music of that period, however, although heavily executed, would give the phrase an expres-
Spohr occasionally indicates similarly sophisticated sive vocal character, entirely characteristic of violin
fingering in the chromatic passages that abound in playing around 1800.
his compositions. 123. Vl: The majority of editors take a2 on the A-string,
Alard and Singer with a harmonic, but not David,
Var. 4 Hermann, Joachim, who remain in 1st position. From
Czerny recommended: “The 4th Variation gentle and 123i–ii, Singer, using the 3rd finger for both notes,
calm, the bass significant.” 59 marks a portamento line between the separate bow-
Beethoven’s dolce may have suggested a fast, light strokes.
bow to the violinist and to the pianist gentle arpeggia- 125f. Pno: All chords could be played with the gentle
tion of all rh chords unless marked staccato. It may arpeggiation accorded the opening theme (5/c/ii).
126i. Vl: A harmonic is marked by all except the three
youngest editors and David, who probably regard-
59 “Die 4te Variation sanft, und ruhig, der Bass bedeutsam.” ed a harmonic as obvious.

126, 128. Pno: Beethoven's slanted lines through the performed very spiritedly and brilliantly, remains true
chords indicates arpeggiation from lowest to high- to this jocular character. Only the middle section in
est notes. F major is to be performed gently and calmly.” 60
128–130, 134. Vl: The bowing/phrasing of the triplet Beethoven will have expected skilled performers to
figures is textually problematic. In the 1st edition the add many subtle dynamic shadings beyond those he
equivalent slurs in Pno in 129, 130, 132, 133 are clearly prescribed. Singer/Speidel’s and Halir’s added dynam-
from i to iii and only in 135 from i to ii, while in Vl ics, accents, and tenuto marks give some understand-
they are from i to ii in 128, 129, 134 and i to iii in 129. ing of what 19th-century musicians expected to hear
For a performer in Beethoven’s time, the slurring in in performance.
the edition would merely have indicated a legato 0, 8. Pno, Vl: The absence of staccato on the upbeat 8th
beginning, and whether the bowing extended to ii here, and elsewhere in the movement, indicates that
or iii would have been the performer’s decision. The Beethoven expected it to be smoothly connected to
editors of AG chose to regularise all these slurs from the following downbeat, which seems to have been
i–ii, which was also adopted in BW. Interestingly, the normal assumption for upbeats at that time un-
since he edited AG, David slurs from i–iii on all less an articulation was marked. A smooth connec-
these figures in the violin part of his Peters edition is also indicated by the ornamented upbeats in
tion, although the piano part corresponds with AG; Pno at bars 2 and 4.
Singer, Brodsky, Halir, Seybold followed David’s ex- 1–8, 51–59, 126–134, 192–207, 211–212. Pno: The jocu-
ample. larity of the theme can be specially emphasised by
131. Vl: The apex of <> would be a classic place for the applying an asynchronous style (rh after lh) for ex-
use of expressive vibrato, perhaps combined with ample at 1i (PT: 1/b), where Speidel marks > , 2ii, 3i
agogic accentuation (3/b/v); the portamento suggest- and iv, 4ii and similar places. Such a treatment can
ed in the edited violin part, quite distinctly executed provide a great variety of accentuation and colour
through a slow shift of the finger, would have been (especially when the theme is repeated). In 8 the
a characteristic expressive gesture around 1800. first chord should receive the accent (as indicated
135. Pno, Vl: Here too the <> may invite not only dy- by Speidel with >) with swift arpeggiation, the sec-
namic nuance but also lingering as well as Vl vi- ond gently resolved and unarpeggiated (5/c/ii). This
brato and Pno asynchrony at the apex. type of cadential accentuation is recommended in
all similar places throughout the movement. In 2
Rondo rh iii the turn should comprise four notes – upper
Allegro auxiliary-main-lower auxiliary-main as annotated
Tempo by Speidel, Vogrich, Ganz, and Halir.
Beethoven’s metronome marks for the 6/8 Allegro first 4. Pno: Beethoven’s use of   (prallender Doppelschlag)
movement of the String Quartet op. 18 no. 5 and the first in rhiv, rather than tr, is presumably an instruction
movement, Vivace, of the Seventh Symphony op. 92 to begin the trill figure on the note rather than from
are both . = 104, and it seems very probable that he the upper auxiliary (5/b/i). Given the tempo of the
expected a similar speed for this movement. His slow- movement Reinecke, Speidel, and Halir annotate a
er metronome mark for the 6/8 Allegro of the String simple but stylish realisation comprising main – up-
Quartet op. 59 no. 2, . = 92, is explained by the more per auxiliary – main – lower auxiliary, while Vogrich
frequent and complex 16th-note patterns. recommends the same turn as in 2.
Moscheles-Cramer . = 108 8–16, 59–63, 119–126. Vl: Performance in the upper
Czerny-Vortrag . = 112 part of the bow was surely envisaged for this theme,
Czerny-Simrock . = 104 probably with the sfs executed  at the point and the
Alard/Diémer . = 104 staccato 8ths detached, but not very short. It is evi-
Speidel/Singer . = 100 dent from the bowing patterns in the 19th-century
Kreisler/Rupp . = 92 60 “In diesem neckischen Thema liegt das Eigenthümliche in dem
Czerny observed: “The individuality of this playful rfz [sic] jedes zweiten Taktes, welches besonders markirt und hu-
theme lies in the rfz [sic] in every second bar, which moristisch hervortreten muss. Das ganze, sehr lebhaft und bril-
liant vorzutragende Rondo bleibt diesem scherzhaften Character
must come out in a particularly marked and humor- treu. Nur der Mittelsatz (in F dur) ist sanft cantabile und ruhig vor-
ous manner. The whole of this Rondo, which is to be zutragen.”

editions that this theme was played in quite dif- Ganz marks tenuto under all lh notes coinciding
ferent ways by later violinists. David and Joachim, with quarter-notes in rh.
probably most closely representing a classic Ger- 28–36, 145–154. Pno: To give a special emphatic nu-
man tradition (also followed by Seybold), mark all ance to rh iii marked sf, pianists of Beethoven’s era
three sfs , almost certainly envisaging a fouetté-style may well have applied arpeggiation or asynchrony
stroke (6, p. X X X I I ). Hermann, Singer, Brodsky, Rosé or both (5/c/ii).
mark bowings which might either have been ex- 37–38, 155–156, 184 –185, 209–210, 213–214. rh: Beetho-
ecuted in the upper half of the bow or further to- ven probably expected legato, with overholding of
wards the frog. Kreisler, however, clearly used an notes to create resonance, which might have been
on-string bowing, corresponding with the markings further enhanced with the use of sustaining pedal
in his edition, in his 1935 recording. (4/a/ii). Most of the editors mark slurs of varying
9–16, 119–126, 209–210, 213–214, 219–229. Pno: Pianists lengths.
of Beethoven’s time may well have played some or 40– 43, 158–161. Pno: The dolce invites both a more
most chords with swift arpeggiation even in a fast relaxed tempo and a moderately slow arpeggiation
movement such as this. Certainly, swift arpeggia- perhaps most stylishly applied to i in each bar, but
tion of the chords marked sf gives them a particu- not necessarily limited to this (3/d/i; 5/c/ii).
larly fiery expression that is not achievable when 40– 43, 158–161. Vl: Some editors bow the passage
the notes are played strictly together (5/c/ii). The essentially as it stands (e.g. Alard, Singer); others
chord at rh 16iii unarpeggiated. combine staccato and slur in a single bowstroke in
18f, 136f. Vl: A diminuendo was often associated with a variety of patterns (e.g. David, Auer). Singer is ex-
descending scales in Beethoven’s time and may well plicit about beginning in the upper half of the bow
be appropriate here, allowing a less sudden p in 19, (Bogen oben).
137, where a fast, light bowstroke can effectively take 44 – 46, 162–164. Vl: David, Hermann, Halir mark slurs
the bow into the upper half. from i–iii and iv–vi over Beethoven’s slur and stac-
20. Pno: The first chord could be slightly accented and cato mark. They probably envisaged a bowstroke in
with swift arpeggiation, the second gently resolved the upper half, with light separation of the staccato
and unarpeggiated. note. This bowing was clearly not to everyone’s taste;
21, 111, 135, 139, 191. Pno: Arpeggiation of the chord the added slurs were removed in David-revised and
is highly appropriate. Singer specifically marked that the passage should
23. Vl: Although 2nd position would be a rational be played in the lower half (Bogen unten), to facili-
fingering here, none of the editors except Kreisler tate which he marked a slur over 43iv–v with an in-
suggest it. Most mark neither fingering nor bowing. struction to use the whole bow (G[anzer]. Bo[gen].).
26–36. Vl: Violinists of Beethoven’s time would surely 44 – 46, 71–74, 162–164. Pno: Overholding of the single-
have utilised open strings for e2 and a1, and har- harmony broken chords will create resonance, and
monics for the e3 here. Alard, Singer, Seybold mark could be enhanced by the use of the sustaining pedal
a harmonic for the e3, David, Joachim, Hermann, (4/a/ii). Speidel and Ganz provide sustaining pedal
who mark no change of position, probably took it marks.
for granted; most younger editors mark stopped 44 – 47, 162–165. Vl: David, Alard, Joachim, Seybold in-
notes. Early-19th-century violinists would probably dicate no shift from 1st position; others provide ex-
have bowed predominantly in the middle and up- pressive fingerings. Either approach might plausibly
per half of the bow, with short strokes for the piano have been taken by an early-19th-century violinist.
and well-extended strokes for the forte. Most would 45ii, 163ii. Vl: Kreisler, in his 1935 recording, employs
probably have used     in 28, 32, 34, 36, taking the traditional harmonic, included in his edition,
the sf near the point, as do most of the editors; only which was marked by Alard, Hermann, Singer,
Alard, Auer, Halir, Kreisler take the repeated notes Brodsky, Halir, and likely employed by David and
in 28 etc. with separate bows. Joachim, who mark no fingering here.
27–36. Pno: Beethoven may have expected a predomi- 47, 165, 183. Pno: The sf invites either swift arpeggia-
nantly non-legato touch in lh. Of all the editors only tion or an asynchronous performance (5/c/ii).
Diémer marks slurs over every half bar. In each bar, 48. Vl: A portamento shift up the A-string, 2– 4, might
Speidel marks staccato dots in lh i, vii and xi, while well have been used by a violinist around 1800, al-

though many would probably have chosen to remain 102, 106, 110. Vl: Singer marks ii–v  with slurred stac-
in 1st position. Alard, Singer, Halir, Rosé, Kreisler cato, a common bowing for many violinists of Beet-
specify the shift; others provide no guidance. hoven’s time.
48–51, 166–170. Pno, Vl: The < > invites linger- 127–132. Vl: The 8th-notes with staccato marks were
ing at the apex, perhaps with expressive vibrato in surely not intended to be played too short (certainly
Vl and either arpeggiation or asynchrony in Pno not with a thrown stroke near the frog), merely with
(3/b/v). The turn should comprise four notes: upper a slightly separated détaché (indicated here in the
auxiliary-main-lower auxiliary-main as annotated edited part with tenuto lines and staccato marks) in
by Reinecke, Rosé, Halir. the middle part of the bow.
60–64. rh: It is possible that Beethoven expected a 166. Vl: Most editors shift from a 1st to 3rd or 4th finger
non-legato touch here, or at least not an overheld (Hermann to a harmonic); David, Joachim give no
legato for the broken chords, though Reinecke, Dié guidance, but very probably envisaged a harmonic.
mer and Halir mark slurs of varying length and Spohr instructed that in shifting to a harmonic,
Speidel marks legato. the finger might come down fractionally early and
65, 67, 69. Vl: Violinists using the upper half of the slide into it.
bow would execute the sf , as marked by David, 170–174. Pno: In addition to legato and depending on
Alard, Joachim, Singer, Seybold. Others, using the the resonant qualities of the piano being used, the
lower half of the bow, mark  for the sf. slurred quavers in rh could be overheld, obviating
66, 68. Pno: Beethoven surely expected an upper-aux- the need for sustaining pedal (4/a/ii).
iliary start to the trills, to emphasise the sf (5/b/ii). 174ii–177iii. Vl: David marks no shifts; most, includ-
The effect of the sf could also be enhanced by starting Alard, Hermann, Joachim, Singer, Auer, Brod-
ing the trill slightly earlier than notated, a technique sky, Seybold, Kreisler shift to remain on the D-string
recommended to produce rapidity and brilliance in in 174f; Alard, Hermann, Singer, Brodsky, Halir, Sey-
mid-19th-century sources, and occasionally employ bold also retain the D-string in 175f; Alard, Singer,
ed by some of the oldest musicians on record such Auer also shift to remain on the G-string from
as Saint-Saëns and Adelina Patti.61 176iv–177. Fingerings of this kind are not unthink-
73. Vl: Although the 1st edition and AG have the ver- able in Beethoven’s time, but would probably have
sion in the present edition, David, Joachim, Auer, Ha- been exceptional.
lir, Kreisler replace note ii with a rest to match b. 71. 175f. Pno: The dissonant octaves at rh i would receive
73f. Pno, Vl: The < and the character of the music a special expressiveness by being played slightly af-
invites an accelerando (3/b/v). ter the bass (5/c/ii; PT: 1/b).
77–90. Vl: Slurring in the unreliable 1st edition is in- 195ii–vi, 196ii–iv. Vl: Slurred staccato, a typical 18th-
consistent here and the editors all diverge from it. century bowing in such circumstances, is marked
David, Hermann, Joachim, Brodsky supply no fin- by most editors.
gering, suggesting that they expected 1st position 200ii. Vl: A harmonic would almost certainly have
throughout as many violinists of Beethoven’s time been used by early performers.
probably did. Alard and Singer keep the whole pas- 206ff. Pno, Vl: Speidel/Singer propose several tempo
sage on the A-string with a harmonic on 79iii and modifications in the final section of the movement:
a shift to 5th position on 86iii. Younger editors also in 206 poco ritenuto, in 209 in tempo animato, in 217
retain the A-string, but without the harmonic. etwas zögernd (somewhat held back), and, from the
77–110. Pno: Beethoven’s dolce perhaps signals a calm- cresc. in 227, accelerando al tempo 1. There is no rea-
er atmosphere and tempo, with overholding of bro- son to believe that such flexibility would have been
ken chords in rh (4/a/ii), arpeggiation of lh octaves, alien to Beethoven or his contemporaries (1/b).
and frequent asynchrony between melody and bass 209f, 213f. Vl: Accompanying a fortepiano of 1798, there
particularly from 93 onwards and especially for the would be no dynamic problem executing Beetho-
sfp in rh 103 and 107 (5/c/ii). The trill in rh 97, pre- ven’s slur; but all the editors, envisaging perform-
ceded by the note above would surely start with ance with a much more powerful piano, divide the
the main note. slur into three, except Hermann, who changes bow
on 209/213iii.
61 See Peres Da Costa: “Off the Record,” pp. 200–201 and 219. 209f. Pno: Beethoven may have refrained from slur-

ring the rh notes to indicate that he wanted legato A major issue for the violinist in this movement is
but not overheld legato (4/a). The cresc at 213 invites the style of bowing Beethoven might have envisaged,
a hastening of pace. which will have reflected the bowing practices of his
211–212, 215–217. Pno: Swiftly arpeggiating the rh oc- contemporaries. It has become customary to play the
taves (or occasionally playing them in an asynchro- figure that first appears in 10–12 and the passage from
nous fashion after the bass) would increase their 27 to 60, 144ff. predominantly in the lower half of the
singing quality and give special emphasis to those bow, with a very short percussive springing staccato
marked sf (5/c/ii; PT: 1/a and PT: 1/b). on many of the separate 8th-notes. This was undoubt-
212, 216. Vl: A diminuendo in the second half of the edly not how any early-19th-century violinist would
bar would probably be appropriate here. have played it, nor, as the annotated editions indicate,
217, 221. Vl: Harmonics in one or both bars would have how most of those of the next couple of generations
been a probable choice by early violinists. David, approached it. The bowing in p would probably have
Hermann, Joachim, probably assuming harmonics, been fast and light, in the upper half, without any at-
give no fingering in this passage, but harmonics tempt to stop the bow, the staccato being achieved by
are marked by Alard, Singer, Auer, Brodsky, Halir, the liveliness of the bow change.
Seybold. Beethoven would likely have expected trained mu-
sicians to add many more dynamics and accents than
he took the trouble to mark. Some inspiration for what
Opus 12, No. 2 might be appropriate can be seen in the extra dynam-
Allegro vivace ic markings given by Singer/Speidel, Auer/Ganz, and
Tempo particularly Halir. Given the overall character of this
For Beethoven’s metronome marks for 6/8 Allegros see movement, it seems unlikely that Beethoven expected
the introductory comments on the last movement of any extreme tempo modification although some of the
op. 12 no. 1. Beethoven left a metronome mark for only cresc. and decresc., as well as certain compositional fea-
one 6/8 Allegro vivace, the second section of the cantata tures, might elicit a slight increase or decrease of pace.
Meeresstille und glückliche Fahrt op. 112: .  = 138. In that 1–8, 124 –131, 204 –211. Vl: This accompaniment figure
movement there are many fewer 16ths, most of which is given in the editions in two ways. David, Sing-
are tremolo effects. The more complex passages of 16ths er, Auer, Halir, Rosé, Seybold mark ii–iii and v–vi
in op. 12 no. 2 certainly indicate a slower tempo. Even slurred staccato; the others leave it as separate stac-
if the first movement of op. 12 no. 2 had been marked cato notes. The first method suggests an on-string
merely Allegro, its somewhat fewer 16ths than the staccato in the upper middle of the bow, the second
Rondo of op. 12 no. 1 would already suggest a slightly a stroke more in the middle, but the long crescendo
faster tempo than that movement, and the addition of lead-in to the recapitulation from 119–123 marked
vivace indicates a more rapid pace. It seems possible in a single up-bow by David, Rosé, Seybold would
that Beethoven might even have allotted a somewhat take the bow to the frog. Several editors change
faster metronome mark than Moscheles, perhaps 120. bow during the lead-in. In fact, although all edi-
Moscheles-Cramer . = 116 tions follow the 1st edition p on 124i, it seems prob-
Czerny-Vortrag . = 108 able that this is an engraving error, since Pno is fp
Czerny-Simrock . = 108 on 124i; fp in the violin part would facilitate a rapid
Alard/Diémer . = 104 bowstroke towards the upper half.
Speidel/Singer . = 104 1, 5, 9 etc. lh: All chords in lh should probably be
Kreisler/Rupp . = 88–92 swiftly arpeggiated from lowest to highest notes
Czerny characterised this Allegro vivace briefly with (5/c/ii). This would also help produce the lightness
the comment “The whole movement should be per- and merriment recommended for this movement by
formed lightly, quickly and merrily. Only the closing Czerny.
melody of the first part [68/184ff] is to be played serious- 1ff. Pno: The first of each of the slurred 8th-note pairs
ly and in a measured manner, but always in tempo.” 62 (a central feature of the movement) might be gent
ly accented and held longer than notated, as in-
62 “Der ganze Satz mit Leichtigkeit, schnell und heiter vorzutra-
gen. Nur die Schlussmelodie des ersten Theils ist ernst und gemes- structed by L. Mozart (2/a), which will bring out the
sen, doch stets im Tempo zu spielen.” coquettish character of Beethoven’s writing. Halir

marks these with > . In 3– 4 and 7–8 and similar sive on-beats separated by rests, as demonstrated
places (for example 110–113) occasional asynchrony by Georg Simon Löhlein.63
between rh and lh would provide stylish textural 17ii, 18ii. Vl: A harmonic is marked by most editors.
variety (adding to the coquettishness). David and Joachim, who mark nothing, almost cer-
8–9. Pno: In rh, the 8th-note upbeat to 9 should prob- tainly took it for granted (it is marked in Joachim-
ably be smoothly connected to the first 16th in 10 revised in 17).
which was standard practice in Beethoven’s Vienna. 17f. Vl: As in Pno, the first of each slurred 8th-note
Halir slurs 8 rhiii to 9i. Alard, Speidel and Ganz pair, which Halir marks with > , could be gently
mark 8 rhiii with a staccato dot but this appears to accented and held slightly longer (2/a).
be a later 19th-century practice. 21i. Vl: Alard, Singer mark a harmonic; David, Jo
11, 102, 104 –107 etc. rh: The grace-note in this figure achim, still in 3rd position, probably envisaged it.
could be played together with iv and immediately 27– 45, 144 –161. Vl: Bowings in the annotated editions
released, as a classic acciaccatura (5/a/iii). again indicate a general use of the middle and up-
11–12f. Pno: According to Czerny all chords “consist- per half.
ing of very short notes,“ which in an example he 27–29f. Pno: An asynchronous style with rh slight-
shows as staccato 8th-notes, should be unarpeggiat- ly after lh, employed occasionally on main beats,
ed (5/c/ii). would add much to the expression of the first note
11, 15, 104 –107, 134, 138 etc. Pno: Halir marks tenuto of the rh slurred 8th-notes pairs. (5/c/ii; PT: 1/b).
lines in both hands on the first and fourth 8th-notes 27–29. Vl: Many editions mark     and Halir, Rosé,
perhaps indicating slight accentuation. Seybold substitute a tenuto line for the staccato on
12–16, 25–27, 135–139. Vl: In the 1st edition 12ii–iii, the quarter-notes, which surely represents the ex-
25i–ii, 135ii–iii are consistently slurred as pairs, with ecution expected by the earlier editors.
a new slur beginning in the following bar. This may 34i. Vl: Singer marks a harmonic.
well represent a deliberate bowing instruction on 31–35, 147–151. Pno: Beethoven probably intended rh
Beethoven’s part, perhaps a practical response to to be legato, and overholding of notes in each bro-
executing the following long slur forte. AG, how- ken chord in 34 –35 is historically appropriate to
ever, printed a single slur from the beginning of heighten resonance (4/a/2). In 34 and 35, the sf could
the forte in all three places. All the violinist-editors be enhanced by asynchronising rh and lh (5/c/ii;
follow the version in AG; all except Singer in 25 be- PT: 1/b). Given the character of the music, Beetho-
gin . David, Alard, Joachim, Rosé, Kreisler, Sey- ven may have expected the lh to be staccato. Speidel
bold retain the long slur, thus ending close to the marks sempre staccato and Ganz staccato.
point, where they presumably executed the follow- 30ii–vi. Vl: Several editors take all or some as slurred
ing figure. Hermann, Auer, Brodsky, Halir take  staccato: Singer ii–vi, David, Auer, Halir iv–vi, sug-
on 14i/26i/136i and again on the staccato 8th-note at gesting that they kept the preceding bars near the
the end of these bars, which suggests that they may point, but expected the next bar to be played closer
have executed the following bar more towards the to the middle.
middle of the bow. Singer certainly ended nearer to 36–60, 153–176. Pno: lh could be staccato throughout.
the frog in 27, because he took the whole of 25f in . The turn in 40 and 41 should start on the upper
13–14, 25–26. Pno, Vl: Beethoven probably expected a note (5/b/i) as annotated by Reinecke and Ganz. The
crescendo. Halir marks 13 rhi mf and < in both grace-note in rh 43 could be interpreted as a short
Pno and Vl in 14. appoggiatura aligned with the bass and given slight
16. lh: It is likely that Beethoven intended continuing emphasis (5/a/ii).
staccato or at least non-legato and perhaps diminu- 37, 39, 153, 155. Vl: Some editors take ii–iii and v–vi ,
endo. Speidel and Diémer mark staccato dots. Halir others with alternate bows.
marks > . But Ganz marks sustaining pedal i–iv. 37– 45v. Vl: David marks no fingering, but most edi-
17f, 21f, 88f, 92f. Vl: Alard marks  on 17i, probably tors remain in 1st position throughout, most prob-
assuming  for each of the slurred pairs. The em- ably using the open A-string in 37 (0 is marked by
ployment of  for successive off-beats after rests
was already a feature of violin playing in the 18th 63 Georg Simon Löhlein: Anweisung zum Violinspielen (Leipzig/
century, and was also sometimes used for succes- Züllichau, 1774), pp. 86f.

Joachim, Auer, Kreisler, Seybold) and it is also mark- pianist played, this would have been the instinct of
ed by Seybold in 40– 43. most early-19th-century violinists, rather than tak-
40f, 156f. Vl: David’s bowing with slurred staccato ing the unaccented 8th-note in a separate bow. Ganz
from iii–vi, certainly performed near the point, is suggests 64 – 65 to be una corda and without sustain-
typical for Beethoven’s time. As in AG, however, he ing pedal, with tre corde in 66.
and other editors omit a slur on iv–v, which Beetho- 68–86, 184 –202. Vl: When a melody is repeated in this
ven seems to have intended (see Critical Report). manner it is almost certain that a distinctly differ-
42f. Vl: Some use  on i–iii and  on iv–vi, others take ent treatment of the repetition was envisaged. It is
alternate bows. quite likely that Beethoven conceived the repetition
44 – 46. rh: A mixture of asynchrony and arpeggiation from 76/192 with portamento in mind (5/c/iii). Two
will help soften the texture and enhance the feeling portamento shifts are common to all the editions
of legato (5/c/ii). Within the general p dynamic, it except Rosé’s and Kreisler’s: 76ii is taken with the
would have been natural for pianists of Beethoven’s 4th finger sliding to a harmonic on 77i and from
era to give dynamic emphasis to the cadential 6/4 at 81i–ii there is a shift from 3rd to 1st finger. Singer,
the beginning of 45. Halir marks < > with the Halir, Seybold mark a harmonic on 82ii; David,
apex at the beginning of 45. Alard, Hermann, Joachim almost certainly took it
48i, 52i, 57ii. Vl: Most editors evidently expected the for granted; Auer, Brodsky mark fingering that re-
open E-string here; Brodsky marks it explicitly in 48 quires a stopped note.
and 57; Singer, Halir, Kreisler mark stopped notes. Pno: Beethoven probably expected pianists to play
50–51. rh: The octaves should probably not be arpeg with occasional asynchrony between the hands for
giated, but to achieve a specially emphatic sf they important notes (5/c/ii; PT: 1/b), which together with
might be played slightly after the accompaniment the Vl portamento would create a variety of tex-
(PT: 1/a and PT: 1/b). ture and accent. Halir marks tenuto lines on Pno
54 – 60, 170–176. Pno: To give particular expression to 70i, 72i, 73i, 74i and 75i which might be particularly
this beautiful harmonic cycle, the rh octaves could apt places for such a treatment. This could similarly
be arpeggiated and or played asynchronously after be applied to Pno 76–86 though here the rh octaves
the accompaniment; the latter would be apt for the might also occasionally be swiftly arpeggiated. In
chords marked sf. In 60, a special emphatic nuance 82 and 83 the sf could be specially enhanced by
can be achieved for the sf by playing the lower lh applying asynchrony, playing the rh octaves very
note on the beat and the upper lh note with the rh slightly after lh. Additionally, Beethoven might have
chord unarpeggiated and slightly after (PT: 1/a). expected a slightly relaxed tempo here which would
58. Vl: A harmonic is marked by all editors except not contradict Czerny’s edict that this section should
David and Hermann, who probably took it for “be played seriously and in a measured manner,
granted, and Rosé and Kreisler, who certainly did but always in tempo.” Nevertheless, the cresc. at 76
not. together with the character of the writing would
61. Pno, Vl: Slightly lengthening the silence that fol- naturally inspire a quickening of pace towards the
lows the sf chord in 60 would heighten the rhetori- climax at 84 –85, which the decresc. in 86 might
cal effect, which accounts of Beethoven’s playing counteract.
suggest he would have expected. 70, 72f, 186, 188f. Pno, Vl: AG adds staccato marks
62–65, 178–181. rh, Vl: Despite the slurs over only the on iii in each of these bars, for which there is no
four 16ths in the 1st edition, the piano parts of all evidence in the source, and this was followed in all
the 19th-century editions follow the piano part in the editions. It seems much more likely that Beet-
AG, in which slurs extend to the 8th-note (although hoven expected a very smooth legato from 70–74ii,
in the violin part of AG, the slur is only over i–iv). 186–190ii.
In practice, of course, the pianist can hardly make 81–85, 197–201. Vl: Hooked bowings are used in these
a distinction between the two notations at the rapid bars by all editors except Auer, who uses them only
tempo of this movement, except by giving an accent in 81–83.
to the 8th-note. In Vl, only Alard follows AG in re- 84f, 100f. Vl: David, Singer, Halir, Rosé, Seybold mark
taining the four-note slur; all the other editors slur the 8th-notes with tenuto lines, evidently regarding
to the 8th-note. Perhaps, responding to what the Beethoven’s staccato marks more as emphatic than

significantly shortening, which is probably what he 228ii, 232i, 236ii, 242ii. Vl: Many editors mark or im-
expected. ply a harmonic.
88. rh: Beethoven’s use of a small-note appoggiatura 234 –238. Pno: Occasional asynchrony between rh and
here is curious, given the preceding figures and the lh, perhaps on the first note under each slur, would
Vl figure, it is unlikely that he expected something intensify the sighing quality of the passage (5/c/ii;
rhythmically different. All the editors, following AG, PT: 1/b).
change the grace-note to a normal 8th-note. 234 –245. Pno, Vl: Beethoven surely expected much
102–107. Pno: The effect of sf might occasionally be more dynamic shading than he indicated. The early
enhanced by slight delay of rh after lh. editors give an interesting range of options: Auer/
114 –117. Vl: David, Singer mark each slurred pair ; Ganz mark diminuendo from 241 and pp at 243; Ha-
this would accord with the practice of using  suc- lir marks dim. at 242 and pp at 242; Singer/Speidel
cessively in such contexts, as exemplified by Löhlein marks poco crescendo across the barline from 240–
in 1774 (see note to 17f., 21f., 88f., 92f.). 241, dim. from the middle of 242 and pp at 244; Rosé
115–117. rh: Slight arpeggiation of the octaves could marks decresc. at 239 and pp at 242. Such changes in
be used to enhance legato (5/c/ii). dynamics may also have been expected to elicit a
161. Vl: The double stop, rather than the two 8ths, as slight relaxation of tempo. The first of each double-
in b. 45, may have been intended to facilitate the fol- note chords in 239–244 could be swiftly arpeggiated.
lowing leap. In any case the player might well have
played the chord with a rapid arpeggiation, which Andante più tosto Allegretto
would have given virtually the same effect. Tempo
162–170. lh: Perhaps to add character and reinforce the Beethoven evidently employed più tosto (he also used
bass, Speidel marks i and iv with > . it in op. 5 no. 2) in the sense ‘or rather’ and may have
184 –196. Vl: David and Joachim mark no fingering. derived it from Haydn’s usage, for instance in his
All the others remain on the G-string until 188ii. String Quartet op. 76 no. 2 and his Symphony no. 103.
All except Rosé and Kreisler shift with a 4th finger Beethoven’s treatment of 2/4 sometimes suggests that
to the harmonic d2. he was really thinking in terms of 4/8, a time signature
199–202. Vl: Alard, Singer, Seybold specifically mark he neglected, and sometimes as a genuine 2/4, which
a harmonic e3. David almost certainly took it for makes his intended tempos problematic. Metronome
granted, and Joachim may also have done so. Her- marks for this movement range more widely than for
mann and most of the younger editors mark a shift many others. Perhaps the closest analogy here is with
to 5th position on 198iii, therefore a stopped e3. the 2/4 Allegretto of the Seventh Symphony, for which
Alard, Singer also mark a harmonic on 202ii, where Beethoven gave  = 76. Czerny, unusually, suggests a
all the others except David, Joachim, mark 1. faster tempo than Moscheles, and later suggestions
212ff. Vl: Singer marks all slurred pairs from 212 to are very much broader. Anything slower than Mo
the end of the movement . scheles’ tempo is undoubtedly contrary to Beethoven’s
212–221. Pno: Occasional asynchrony between rh and conception of this movement. His expectation for a
lh would intensify the expression (5/c/ii; PT: 1/b). rapid tempo is suggested also by the occurrence of
216–220. Vl: In the 1st edition the pairs of slurred 8ths two-bar slurs in the violin part.
in the violin part are on the beat; in AG, David’s, Moscheles-Cramer  = 138
Singers’, Rosé’s editions they continue the rhythmic Czerny-Vortrag  = 76
placement of 212–215. Brodsky has the same place- Czerny-Simrock  = 80
ment in 216f, but on the beat from 218. The motiva- Alard/Diémer  = 96
tion for these changes (a lost source?) is unknown. Speidel/Singer  = 104
222f. rh: Beethoven probably expected legato here Kreisler/Rupp  = c. 92
(some early editors slur), but even at a fast tempo Czerny described the movement very briefly as
such as this he would not necessarily have expected “Somewhat melancholy, though not drawn-out, but
passages of 16th-notes to be played absolutely even- with much expression.” 64
ly (2/b). Some agogic nuance would certainly help
delineate interesting shapes and obviate monotony. 64 “Etwas schwermüthig, doch nicht gedehnt, aber mit vielem
Speidel marks > on 222 i and 223 vii. Ausdruck.”

Trained musicians of Beethoven’s day would cer- by which he indicated in an example a very slight
tainly have added many more dynamics and accents shortening of each. (4/b/i).
than Beethoven notated. Halir’s additions in Vl and 7f. Pno: To create a suitably ethereal effect where Beet-
Pno offer an example of 19th-century practice in this hoven marked pp, the use of the una corda shift or
respect. some form of moderator if available will be invalu-
1f, 68f. Pno: All chords ought probably to be arpeg able (PT: 3/a and PT: 3/b).
giated (5/c/ii). By this means, the “somewhat melan- 9–14, 77ff. Vl: All mark the same fingering (given in
choly” mood described by Czerny can be brought the present edition). The 4 – 4 shift in 11 was certain-
out and subtle shadings of expression achieved ly meant to elicit an expressive portamento.
through varied speeds (in the moderate range) of ar- 15f, 31f. Vl: Most finger 2– 4-o, a very plausible ear-
peggiation. In this respect, the placement of certain ly-19th-century fingering; only younger violinists
notes in arpeggiation sooner or later than expected Brodsky, Rosé, Kreisler avoid the harmonic.
(at the discretion of the pianist) can produce poign- 34, 36, 43, 45, 47. Vl: Downward portamento finger-
antly poetic effects. Those marked fp or sf might be ings are given by all except Alard, Hermann, Brod-
arpeggiated more swiftly and which would render sky, who mark one only in 47. Many violinists of
them more energised. Halir marks the movement Beethoven’s time would surely have seen this pas-
dolce ed espressivo. Speidel marks the chords in 1 and sage as an invitation to retain the tone colour of
9 with staccato dots and tenuto lines. According the D-string with the use of expressive portamento,
to Czerny, all chords (4, 8, etc.) “consisting of very while others may have preferred the simplicity of
short notes,” which in an example he shows as stac- 1st position throughout. Some might have chosen
cato 8th-notes, should be unarpeggiated (5/c/ii). only to employ portamento fingering from 42.
2, 10, 78, 112. rh and Vl: Beethoven may well have ex- 34 – 63. Pno, Vl: Beethoven’s dolce gives license for a
pected the rhythms to be more flexible than notated, slightly broader tempo as well as agogic inflection
perhaps lingering on certain notes due to their dis- of important 16th-notes, bending rhythms in subtle
sonant quality (2 rh i is a good example), and mov- and varying ways (1/b; 2/b). An exact rendering of
ing more swiftly through others, thus producing a the notation here would have seemed alien to the
quasi improvised effect through the interpolation ideals of ‘beautiful’ performance in Beethoven’s era.
of a range of rhythms from triplets to over-dotted All lh chords (apart from 40 lh ii, 48 lh ii) could
figures (2/b and c). Beethoven’s notation in 112 is be gently arpeggiated in varied speeds within a
rather interesting in providing a different rhythmic moderate range which would significantly add to
option. the feeling of dolce and help fill out the texture
3, 11, 18, 26 etc. Pno, Vl: The <> signs are very hap- (5/c/ii). In 47, the rh slurred 8th-note pairs should
hazardly reproduced in the 1st edition, but it seems certainly be given a lilting execution (long/short)
clear that Beethoven will have centred them upon which would bring out their sighing quality (2/a).
the strongest note in each phrase, which he expect- In 53–55 and 61–63 the <> invites slight ac-
ed to receive a particular kind of emphasis, less per- celeration towards and lingering at the apex with a
cussive than sf, and perhaps agogic. For the violin- return to tempo afterwards.
ist <> would probably have encouraged special bow 50ii–iii, 59ii–iii. Vl: 3–3 is marked by Alard, Singer,
pressure and bow speed, perhaps together with Auer, Rosé.
vibrato, which, however, would only be effective 51ii, 60ii. Vl: Alard, Singer mark a harmonic.
here if used sparingly and sensitively elsewhere. 61. Pno: Halir alters rh vii to G adding in a footnote:
For the pianist it would almost certainly have en- “Original […] altered after the third Bar of the pre-
couraged arpeggiation and/or asynchrony of the ceding Violin part, as in Beethoven’s first creative
hands. (5/c/ii) period the compass of the piano went only from
6, 14. Pno: All chords marked portato might be given low F to the f above the third ledger line.”
special expression, as Moscheles recommends, with 64. Pno: The beginning of the grace-note turn should
swift arpeggiation of each chord “giving them the be aligned with the bass (5/a), as indicated by Ganz,
same length of time as a dot under a slur requires”,65 but the accent could still be given to the main note
65 “und mit derselben Geltung angeben, welche das Staccato un- as marked with a > by Speidel and Halir.
ter einem Bindungszeichen erfordert.” 72. Vl: All editors mark 4 – 4.

75i–ii. Vl: A harmonic, followed by 2nd finger is mark- Allegro piacévole
ed by Alard, Joachim, Singer (with a portamento Tempo
sign between i and ii), Auer, Brodsky, Halir, Seybold. Beethoven gave seventeen metronome marks for Alle-
David almost certainly took the harmonic for grant- gros in 3/4, but evaluating them is complicated by the
ed. Joachim’s Violinschule explains downward porta- distinctions associated with minuets, scherzos, and
mento from a harmonic: “The finger, which at first other movements in that metre. His expected tempo
is only softly laid on the string, has to be pressed for this movement is also problematic because of his
down with increasing strength on the string while unusual use of the term piacévole, which he may have
sliding.” 66 intended to refer only to character, or to have acted as
88–91. Vl: Beethoven’s slurring is impracticable to play a moderating factor on speed. Perhaps the most help-
in a single bow even at Czerny’s speed, but the ex- ful analogy among the movements for which Beetho-
pected legato can be achieved by very smooth bow ven gave metronome marks is the first movement of
changes. the Eighth Symphony op. 93; this is designated Allegro
90f. rh: The grace-note could be treated as an acciac- vivace e con brio and marked . = 69; its fastest notes
catura, i.e. played together with iii and immediately thematically are 8ths, with 16ths occurring only as
released (5/a/iii). tremolo in the strings. In the Allegro piacévole of this
94, 95. Pno: Asynchronous performance with rh played sonata, the fastest notes are triplet 8ths and that, to-
very slightly after lh would produce an especially gether with the tempo term, suggests a slower pace
expressive sf (5/c/ii; PT: 1/b). than the symphony. From that point of view, Mosche-
100–107. Pno: Beethoven probably intended lh to be les’ marking seems distinctly too fast. Czerny’s may
legato. Overholding the notes in each broken chord be closer to what Beethoven would have written for
to create resonance and obviate the use of sustain- this movement. This is a rare case in which some
ing pedal is a viable option (4/a/ii). In 101, 103, 105 later tempos are faster than the earlier ones. Czerny-
playing rh slightly after lh would add much to the Vortrag gives the tempo term as Allegretto piacévole,
expression of sf. Additionally the rh octaves in 101 but in view of his metronome mark that may simply
and 105 might be broken. be an error.
122ii–iii. Vl: The expressive portamento fingering 3–3 Moscheles-Cramer . = 76
is marked by Hermann. Others shift 3–3 on iii–iv. Czerny-Vortrag . = 66
126f. rh: Beethoven may have intended a slight slow- Czerny-Simrock –
ing down to bring the movement to a poetic close. Alard/Diémer . = 63
Halir marks dim. e rit across the barline in Pno 126f Speidel/Singer . = 69
and in Vl 127. Kreisler/Rupp . = c. 80
127. Pno, Vl: rinf., in 127, following from cresc. in 124, Czerny instructed that the movement should be “In a
suggests a strongly intensified treatment of crescen- lively tempo, but with contented calmness and light-
do over the last three notes of the bar (3/b/ii). Auer’s hearted humour.” 67
suggestion of a shift from 2 to 3 on v–vi, a type of Beethoven will have expected a range of dynamic
expressive fingering also found in early-19th-centu- nuance which was impossible to indicate in the no-
ry sources, provides a particularly effective inten- tation. Halir’s and Ganz’s extra dynamic annotations
sification, executed with a distinct slide of the 2nd provide insights into what Beethoven might have ex-
finger under a very firm bowstroke, in conjunction pected.
with Pno’s leap to f 3. Speidel marks > on rhiii per- 1ff, 83ff. Pno: Beethoven’s dolce invites the use of both
haps suggesting agogic intensification. asynchrony and noticeable arpeggiation according
128–129. Pno: The octaves might be swiftly but gen- to taste and circumstance (5/c/ii). Given the time sig-
tly arpeggiated which would help achieve a very nature and the lively tempo, half-note chords might
peaceful pp (5/c/ii). be arpeggiated while quarter-note chords are unar-
peggiated, but this need not be consistently fol-
lowed. All slurred passages imply overheld legato
66 Joseph Joachim and Andreas Moser: Violinschule (Berlin, 1905), touch where possible (4/a/ii).
vol. 2, p. 93. “wird der zunächst nur sanft aufgelegte Finger wäh-
rend des Gleitens allmählich immer stärker auf die Saite niederge- 67 “In lebhafter Bewegung, aber mit vergnügter Ruhe und heite-
drückt werden müssen.” rem Humor.”

1iv–2iif, 9iv–10iif. rh, Vl: The characterisation of the 54. Pno: Ganz marks espressivo.
theme will be greatly enhanced by the 18th- and 54ff, 120ff, 152ff, 283ff. Pno: Beethoven surely expect
19th-century practice of accenting syncopated notes ed the broken chords to be played legato in rh and
as if they were strong beats (probably gently in view lh respectively and with overholding of notes be-
of Beethoven’s dolce). longing to single harmonies. Alard adds slurs; Spei-
2ii, 3ii, 10ii, 11ii. rh, Vl: Here and in subsequent ap- del and Halir mark legato. In 120 Ganz marks poco
pearances of this theme, AG gives a staccato over legato in rh.
the slur, but this is not in the 1st edition. All the 55f. Vl: The a2 is given as a harmonic by all the older
editions except Alard, who may have consulted the editors and some of the younger ones (only Brod-
1st edition, and Seybold include the staccato. sky, Rosé, Kreisler mark a stopped note). This and
5–8, 21–25. Pno: In addition to dynamic nuance the similar instances confirm that very sparing use of
< > might inspire a slight increase of mo- vibrato, even in melodic passages with longer note
mentum to and lingering at the apex; perhaps also values, which is indicated by documentary evidence
broader arpeggiation on 7i/22ii (3/b/v). from Beethoven’s time, was still expected in this
8. Vl: The majority begin . Alard, Joachim, Singer, repertoire in the second half of the century. All ex-
Auer . All begin with the open string. cept Hermann, Brodsky, Rosé, Kreisler follow with
17–18i, 100–101i. Vl: All except Alard, Kreisler employ a downward portamento shift to 3 in 56.
the same fingering, 2–3. 57, 65, 286, 294. Pno, Vl: The grace-note should be
21–25, 88–91, 104 –107, 234 –237, 250–253, 286–289. Pno: played as close to the beat as possible.
The < > could signify increased momentum 58f. Vl: Most mark or imply a shift from 3– 4, to create
to and lingering at the apex. The sf in 24 would gain a characteristic portamento that supports Beetho-
energy from playing rh very slightly after lh (5/c/ii). ven’s <> ; only Singer, Auer, Halir, Kreisler
23, 106. Pno: The trill, preceded by the note above, mark 1 in 58.
should almost certainly start on the main note (5/b/ 61–64. Vl: All take this phrase on the G-string, with
ii). an unavoidable portamento shift from 4 –2 in 63f,
24ff, 108ff. Pno: All chords in lh not marked stacca- creating a mirror image of the phrase in 58–60.
to could be tightly arpeggiated (5/c/ii). All broken 79, 81, 308, 310. Pno, Vl: For rhetorical effect these bars
chords could be played with notes overheld as far might be somewhat extended.
as possible. 120–122. Vl: All except Brodsky, Rosé mark a harmonic
24, 25. Vl, Pno In a footnote annotation Speidel ex- on 121iii. All except Brodsky indicate a portamento
plains poco f = mf. shift between 122i–ii. Such strong portamento tradi-
34 –36, 263–265. Pno: Slurs over two equal-value notes tions in this theme may well go back to Beethoven’s
imply emphasis and inequality (2/a). The sf chords time (5/c/iii). In 120 Singer marks cantabile in Vl.
could receive special emphasis by swift asynchrony 120ff. Pno: In addition to strong accentuation, the sf
of the hands, rh before lh. Halir marks the first of might occasionally be enhanced by asynchrony,
each with tenuto lines. playing rh very slightly after lh (5/c/ii; PT: 1/b).
37. Vl: There are basically three fingerings for this 121–127, 151–157. Vl: David, Singer, Auer, Halir, Rosé,
arpeggio: all mark iv with 0 and v with 1; most re- Seybold replace the staccato dots with tenuto lines,
main in 3rd position for vi, but David, Halir mark probably indicating a broad, light, but well-connect
1 again on vi (though David-revised removes this ed bowstroke to produce a dolce effect. Beethoven’s
fingering); vii is given as a harmonic with 4th fin- staccato marks in such places were surely not in-
ger by Alard, Kreisler, Seybold, with 3rd finger by tended to signify a significant shortening of the
Joachim, Singer; and as a stopped 4th finger by note-value, merely separate bows.
Hermann, Auer, Brodsky, Rosé. 122. Vl: The portamento from a harmonic, given in
38–39ff. Pno: All lh octaves might be tightly arpeg the edited violin part, was traditional, marked by
giated (5/c/ii). David, Alard, Hermann, Joachim, Singer, Halir etc.
42– 4 4, 49–50, 271–273, 278–279. Pno: In addition to For its execution see the note to the second move-
agogic accentuation of the first note of each of the ment of this sonata b. 75.
slurred-pairs, asynchrony with rh before lh would 136. Pno, Vl: Ganz marks dolce espressivo in Pno. Sin
enhance expression by creating an agogic effect. ger marks espressivo in Vl.

144ff. Pno: The first of each slurred-pair in rh might Czerny observed: “This sonata is significantly grand-
be both slightly lengthened (2/a) and tightly arpeg er than the previous two, and written in a noble, bril-
giated (5/c/ii). liant, but also weighty style. The tempo is a moderate
151ii. Vl: All except Rosé mark a 4th finger harmonic. Allegro since it contains many passages in faster note-
156. Vl: All make an expressive shift 4 –2 except Alard, values, which, however, are to be played very fast and
Rosé, who extend back from 4 to 1. with bravoura.” 68
172–174. Vl: Singer, Auer, Brodsky, Kreisler go up the 0f, 4f, 105f. Pno: The descending broken chords should
G string, a possible but unusual fingering for Beet- surely be slurred across the barline (as indicated by
hoven’s time. All other editors remain in 3rd posi- all but AG, Diémer, and Joachim) and played with
tion. notes overheld (4/a/ii).
184 –187, 194 –197, 230–237, 254 –261, 312ff. Pno: Tight 1, 3, 104, 106. Pno: The notes in the rising broken
arpeggiation of these chords (apart from those chords could be overheld and or played with sus-
marked staccato) would be appropriate (5/c/ii). taining pedal as suggested by Speidel and Ganz.
193–195. Vl: Alard, Singer, Halir begin in 2nd posi- 1–9. Vl: Most editors envisage 1st position, which
tion and mark a harmonic on 194iii, with 4 again would surely have been the fingering of choice in
on 195ii, an elegant fingering that might well have Beethoven’s time.
been used in Beethoven’s time. 2, 4, 105, 107. Vl: The notation  instead of . is typical
206–216. rh: All chords would gain much energy by in early-19th-century notation where another part
being tightly arpeggiated. has a resolving appoggiatura, probably partly to
284f. Vl: All except Kreisler mark a portamento shift alert the player to the necessary expression (in this
from 4 to 2 (Singer marks espressivo from 282iii), case, gentle accentuation followed by diminuendo)
Kreisler descends to 1, minimising the portamento. and partly to warn against shortening the note and
292f. Vl: All finger this 3–1, before returning to 3rd stopping before the resolution of the dissonance.
position in 294, except Joachim, Auer, whose edi- 2f, 105f. Vl: All but AG, Alard, Joachim slur across
tions omit 1 in 293; but whether they really expected the barline.
a fifth across the strings is questionable. 2, 4, 105, 107, 162, 163, 165. Pno: the chords supporting
296f. Vl: All except Kreisler finger 4 – 4 – 4, all except poignant melody-notes could be swiftly arpeggiat-
Rosé with a harmonic on 297i. Auer, Brodsky, how- ed and the melody-notes given special emphasis by
ever, break the slur at the end of 296, perhaps to being delayed. It is self evident that rhi should be
avoid portamento. stronger than rhii, the latter of which is the reso-
lution. Speidel marks > and Halir tenuto lines on
2 rhi (5/c/ii). Additionally, Halir marks < through
Opus 12, No. 3 1 and > from 2 rh i–ii. In 165, to give special ago-
Allegro con spirito gic emphasis to the sf, the lowest note in lh could
Tempo be played on the beat with the remaining lh and rh
Beethoven gave no metronome marks for Allegros in notes played unarpeggiated and very slightly later
 meter with numerous triplet 16ths, but the musical (PT: 1/a).
material has a close resemblance to the 2/4 final move- 4f 107f. Vl: AG and all editors slur across the barline.
ment of the String Quartet op. 18 no. 1, marked simply 5–6, 108–109, 166–168. Pno: rh i might be played with
Allegro, which also contains many normal and triplet special emphasis (Speidel marks tenuto lines) and
16ths, for which he gave the tempo  = 60. This is close perhaps by playing the rh asynchronously after the
to Czerny’s and Moscheles’ suggestions for the tempo lh (5/c/ii); rh ii–v could be played diminuendo di-
of this movement and seems a very likely figure for minuendo (as marked by Halir and Rosé). Given the
Beethoven to have chosen. half-notes in lh, it is likely that Beethoven expected
Moscheles-Cramer   = 126
Czerny-Vortrag   = 116
Czerny-Simrock  = 120 68 “Diese Sonate ist bedeutend grossartiger, als beide vorherge-
Alard/Diémer  = 108 henden, und in einem edlen, brillanten, aber auch schwereren S tyle
geschrieben. Das Tempo ist ein gemässigtes Allegro da v
iele Passa-
Speidel/Singer  = 116 gen in einer schnellern Notengattung darin vorkommen, welche
Kreisler/Rupp  = 104 jedoch sehr geläufig und mit Bravour vorzutragen sind.”

a fairly sustained approach for the accompaniment 15ff. rh: Probably matching the articulation of Vl in
perhaps akin to portato, as notated by Speidel 13ff; Speidel and Ganz mark staccato.
(though not in 108–109 and 166–168). 15–17. Vl: Some editors retain the 1st edition’s slur-
7–8, 110–111. Pno: Beethoven’s < > is likely to ring. David and many others mark a single slur in
have signified subtle tempo fluctuation hastening 16 (matching the 1st edition slurring in 116); all ex-
towards and lingering at the apex, and returning cept Hermann continue it to 17i and begin a new
to tempo afterwards, in addition to dynamic shad- slur on 17ii.
ing (3/b/v). He may have intended either legato or 18, 20. Pno: Very swift arpeggiation in lh will help
non-legato touch in lh: Diémer adds slurs; Reinecke produce the fiery energy required to support the
a long slur; Speidel marks staccato; Ganz indicates rambunctious rh figures (5/c/ii; PT: 1/a), which could
quazi staccato (presumably meant also in 110–111). be further enhanced with sustaining pedal as mark-
8, 111. Vl: The bowing instructions here are particu- ed by Ganz in 18, and Speidel in 20. In a footnote
larly revealing. David envisaged the 16ths played Speidel offers a solution to the difficulty of the rh
close to the point, because he slurs the first of them figure suggesting that rhvi and xiii be taken by lh
into the down-bow on 8i (slur removed in David- stating: “This passage can also be divided between
revised). Singer indicates a slurred staccato ad lib., the hands in the following manner.” 69
which would have been started at the point and re- 19ii–v, 21ii–v. Vl: Most editors mark successive up-
mained in the upper half. Halir marks M (middle), bows.
perhaps with short strokes on the string, or with a 19, 21. Pno: The broken chords overheld and possibly
hopping (hüpfender) bowstroke. with sustaining pedal as marked by Speidel.
9–13, 112–113. Pno: Beethoven likely expected the lh 22i–ii. Vl: Although these two notes are taken in sep-
broken chords to be legato and overheld (4/a/ii): arate bows in all the editions, the editors mark a
Diémer slurs each broken chord; Speidel marks a change of position from 3rd to 1st and in some cases
continuous slur; Halir marks legato; Ganz marks at least, almost certainly expected an audible connec-
sustaining pedal every half bar (presumably meant tion between the bowstrokes (the 4th finger sliding
also in 112–113). In 9, 10, 11, 112 and 113, rh i might down the E-string as far as b 2), which is essential
be asynchronised playing it slightly later than the to create a true legato connection. In Grützmacher’s
lh (5/c/ii). cello version, where the two notes are taken on the
11. Vl: The slur was printed from i–v in AG, but BW A-string, he specifically marks gliss. between the
chose to take it to vi, although this in not at all clear bowstrokes here.
in the 1st edition. Most of the editors, beginning 22, 122. Pno: Given the ff and the climactic nature of
the bar , evidently played the 8ths near the point, the writing, both over-legato where possible (4/a/ii),
and thus beginning  on vi. Singer, however, marks with sustaining pedal as indicated by Ganz (pre-
slurred staccato on vi–ix, to arrive  on 12i. sumably meant also in 122) will be a great support.
12ff, 114ff. rh: All 16th-note passage work with slurs The octaves in lh could also be arpeggiated swiftly
could be played with over-legato touch where pos- (5/c/ii).
sible (4/a). 23–25. Vl: Almost all editors mark the passage to be
13ff. Vl: Although all the editions leave these notes played on the A-string, which for the sake of tone
with separate bows, a typical 18th- and early-19th- colour will probably have been the preference of a
century treatment of such passages of repeated ac- violinist around 1800. All the editors who use the
companiment notes would have been to execute A-string repeat their fingering in 23 and 25 (either
them in groups of four to a bow, with an articulation to 3 on iii or the less expressive 2 on iv) rather than
somewhere between staccato and portato, accord- varying it for expressive reasons, which might have
ing to the musical context (probably here a semi- been expected of an early-19th-century violinist. The
portato articulation). Some violinists of Beethoven’s grace-note in 23, as the resolution of the cadence
time may well have chosen this type of bowstroke, from 22, will certainly have been envisaged as oc-
while others would have played gently articulated curring on the beat, but very short.
separate bows in the upper half of the bow. In the
absence of staccato marks, they would certainly not 69 “Diese Stelle kann auch folgendermassen unter beide Hände
have been played with a very short articulation. vertheilt werden.”

23–25. Pno, Vl: Given the change of character and their edition. This was followed by all subsequent
sustained melodic writing, Beethoven may have ex- editors and is given editorially in BW. It is possi-
pected a fairly smooth lh, as indicated by Reinecke, ble that Beethoven assumed the pattern would be
who marks slurs over non-repeated notes, and Ganz, continued, but it is unusual for him not to mark it
who marks quasi legato. Speidel marks staccato though throughout. It may be noteworthy that there are no
he may simply have meant separation of some sort. slurs nor staccato marks in the piano part, where
In 23–24 rh, the octaves might be gently arpeggiated this figure occurs in bb. 29–35, 125–131, and he may
which would give them a soft expressive effect (5/c/ have intended a distinction between the initial pat-
ii). Ganz marks these with tenuto lines, but Speidel tern and later parts of the passage. In any case,
with staccato. with this kind of accompaniment figure musicians
26. rh: Both Speidel and Ganz thought of rh i, ii, and of Beethoven’s time would have had no compunc-
iii as requiring special articulation. Speidel marked tion about altering the bowing patterns to suit their
tenuto lines, Ganz tenuto lines with dots. own ease of playing, nor is the composer likely to
27–28. Pno: Overholding in rh would help achieve a have objected if the delivery was effective. David,
special resonance for the subito p enhanced by swift Singer, Auer, Halir, Rosé made changes to the pat-
but gentle arpeggiation of lh octaves (4a; 5/c/ii). Rei- tern given in AG.
necke slurs across the barline in rh. 37– 43, 132–138. Pno: In 37 rh i–iv might be played
28xi–xvi, 124xi–xvi. Vl: Hermann, Singer mark slurred with flexible placement of the notes which Speidel
staccato, the latter continuing it until 29iv/125iv. marks grazioso and Halir with portato. For height-
29i–iv, 125i–iv. Vl: David has separate bows, begin- ened expression any of the rh notes might be played
ning . Most, however, mark slurred staccato , but, asynchronously, slightly after the bass, with the lh
like Halir, who has tenuto lines under the slur, may octaves very quickly arpeggiated (5/c/ii). The trill
have envisaged a stroke closer to portato. Hermann in rh 38 and 134 should start with the upper aux-
adds dolce, Singer grazioso; Grützmacher removes the iliary note (5/b/ii). In addition to asynchrony in 39
staccato entirely and marks espress. Such markings and 135, the rh, marked portato, might be played
perhaps support an expressively flexible placement with unequal slightly lilting (long-short) rhythms
of the notes. (5/c/ii; 2/b). In 41 and 42 Speidel marks tenuto lines
29–35, 125–131. Pno: Given the lh slurs, Beethoven on rh i. Beethoven probably expected the lh notes
may have expected the rh to be played legato with throughout to be played with varying articulation
or without overholding. Diémer marks slurs every according to character. Speidel, for example, marks
half bar while Speidel marks the same articulation most staccato except in 41 and 42 which he marks
pattern that Beethoven gives to Vl in 37–38 explain- portato. In 42 and 43, the 16th-notes in rh might be
ing in a footnote that “Weak players should be al- played somewhat unequally, lingering on the most
lowed a legato here.” 70 important, including the first of the slurred pairs in
33. Vl: This is a classic situation for portamento, and 43 (2/a). Halir’s tenuto markings in 43 rh v, ix and
all the editors mark 3–1. xiii provide some inspiration in this respect.
35v, 131v. Vl: This movement, being in E flat, gives 44 – 49, 140–145. Pno: All the sextuplet 16th-notes
few opportunities for the use of natural harmon- might be played with as much overholding as pos-
ics, but Alard, uniquely among the editors, employs sible (4/a/ii). What articulation Beethoven intended
one here. for the lh is not clear. Only Ganz and Speidel indi-
36, 132. Pno: Overholding (4/a/ii), with use of the sus- cate staccato from 44. In 47 Speidel and Rosé modify
taining pedal, as marked by Speidel (though not in Beethoven’s notes in both rh and lh, pointing out,
132) would be appropriate here. in a footnote, the analagous material in bar 142. In
37– 43, 133–138. Vl: The editors of AG evidently de- 48 and 144 the sf in lh could be given special ener-
cided that the pattern established in 37f, 133 and gy and accent by a swift arpeggiation of the octave
repeated in the first half-bars of 41f (but not 137f) (5/c/ii). Likewise, in 49 the rh octaves might receive
was meant to apply to all the separate notes in this swift arpeggiation to energise them.
passage, and marked it thus, without comment in 45– 49, 141–145. Vl: To produce a convincing forte with
Beethoven’s slurs is challenging, and it seems unlike-
70 “Schwächern Spielern sei hier ein legato gestattet.” ly that Schuppanzigh or his contemporaries treated

them as bowing instructions. All editors except Her- playing the upper lh note together with rh notes
mann retain the single slur throughout 45/141, some unarpeggiated a little later (PT: 1/a); Speidel marks
also in 47/143, but most change bow in 47/143 and risoluto in 75 and presumably meant it to apply to
49/145. Those who mark no bow changes at all (Da- 77; on 75 and 77 rh ii a special energised but not
vid, Rosé, Kreisler) may well have made some in harsh accent will be created with a swift arpeg-
practice. David often divided long printed slurs in gio. In rh, all slurred broken chords and all slurred
his personal copies. passage work could have notes overheld wherever
50f, 146f. Vl: Most editors mark the 8ths to be played possible (4/a/ii). From 64 –74 Speidel gives several
, a very plausible 18th-century bowing, indicated sustaining pedal marks that would certainly help
by Löhlein in 1774 (see note to op. 12/2/I bb. 114 –117). aid resonance and the feeling of con fuoco which he
50–55, 146–151. Pno: What articulation Beethoven ex- marks at 64.
pected for the quarter-notes is uncertain, though 70f. Vl: All except David, who marks  on 70 (removed
staccato may be appropriate given the markings in in David-revised), evidently expected  for the sf, al-
Vl, in which case the lh should be unarpeggiated. most certainly fouetté.
In line with Vl, many editors mark staccato on 50 75, 77. Vl: Only Hermann, Auer, Brodsky mark these
and 146 rh i and lh i. Speidel marks staccato on all slurs to be divided.
quarter-notes in both rh and lh in 50–53 and 146– 82i, 83i. Vl: David marks  here, surely envisaging fou-
149; Ganz marks lh staccato throughout (presum- etté, which would be a very effective bowing within
ably meant also in 146–151) contrasting with the rh the tradition of late-18th-century violin playing; but
in which all quarter-notes are marked with tenuto this was removed in David-revised.
lines to be enhanced with sustaining pedal (pre- 84 –94. Vl: The off-beat 8ths are predominantly indi-
sumably meant also in 146–151). In 50–51 and 146–147, cated to be played , as they would almost certainly
Beethoven may have expected rh iv to be accented have been in Beethoven’s time. Those editors who
(in line with Vl) as marked by Halir with > . Each did not explicitly mark it probably regarded it as
grace-note acciaccatura in rh could be played to- self-evident. Some or all 8ths in 89–91 are marked
gether with the note to which it is appended and with slurred staccato (David, Singer, Auer, Halir,
lifted immediately after (5/a/ii-iii). Rosé, Seybold).
56–57, 152–153. Pno: Many later editors added the up- 85ff. Pno: All chords not marked staccato might be
per octave at 56ii and 57i, which was not available swiftly arpeggiated (5/c/ii).
on Beethoven’s piano. Czerny argued that adding 94 –95. rh: Occasional swift arpeggiation of the chords
the higher octave where it would have exceeded perhaps on the downbeats would be appropriate. In
the range of the piano available to Beethoven at the addition to dynamic nuance, Beethoven’s < >
time of composition should be avoided (5/c/i), but might imply a slight hastening towards and a slow-
this is unlikely to have deterred Beethoven’s con- ing down after the apex (3/b/v).
temporaries from doing so, and might well have 96–103. Pno: It seems likely that Beethoven intended
been the composer’s choice. Swift arpeggiation of a special ethereal sound world for this section, per-
chords not marked staccato and especially those haps with a significantly slower tempo and/or the
marked sf would help produce the requisite energy use of a tone modifying pedal such as the una cor-
and accent while mitigating harshness (5/c/ii and da or moderator (if available) (PT: 3/a and PT: 3/b).
PT: 1/a/vi). He may well also have expected arpeggiation and
59–63, 155–159. rh: Beethoven’s > might elicit asyn- asynchrony of the rh melody-notes and slight modi-
chrony with the melody-note played very slightly fication of tone and tempo where he marked <
after the accompaniment (5/c/ii). In 62–63 and 157– > in Vl, presumably also expected in Pno, in
158 in lh, Beethoven’s pp might be made more ef- 101–102. Speidel marks tranquillo e molto espressivo at
fective with use of the una corda pedal as suggested 96, legato under lh in 97, smorzando at 103, and then
by Ganz. Tempo 1 at 104; he also adds < > in 99–100. Ha-
64 – 80. Pno: All chords in lh not marked staccato might lir marks legato at 6 and ritardando halfway through
be swiftly arpeggiated (5/c/ii). In 75 and 77, the sf on 103. Ganz marks p espressivo at 96.
i might be made most effective with an agogic effect 97–102. Vl: All except Alard and Joachim, who remain
by playing the lowest note in lh with the beat and in 3rd position, follow David in executing this mel-

ody on the D-string. Singer adds molto espressivo, Speidel/Singer  = 63
Auer espress, Grützmacher p ma espress. Kreisler/Rupp  = c. 60
98. Vl: The absence of a turn in Vl, in this unison pas- Czerny instructed: “The melody with the greatest
sage between Vl and rh, in the 1st edition is prob- possible expression, which is to be achieved through a
ably an error. It is conceivable, though unlikely, that beautiful touch and tone. The left hand must hold the
it was left out because the a flat would have made 8th-notes to their full value. The bass accompaniment
it awkward to play in 1st position. None of the 19th- (from the 9th bar onwards) extremely light and short.
century editors suggest 1st position, because all in- From the 23rd to the 38th bar, all leggatissimo and with
clude it. expression that matches the violin’s melody. (The 37th
110. rh: The grace-note should be played on the beat and 38th bars with pedal.) For the following theme,
with the main note following very closely (5/a/ii). the pedal must also be used at each quarter bar. The
119–121. lh: All chords could be arpeggiated moder- final bars ritardando.” 71
ately quickly which would help fill out the texture Czerny’s use of the term legatissimo is noteworthy.
(PT: 1/a). For the piano, it may imply overholding of the tones.
125iii–132. Vl: Alard, Halir mark all this passage to For the violin it suggests the most perfect legato con-
be played on the A-string; Joachim, Singer, Auer, nection between bowstrokes, which, during the whole
Kreisler on the A-string from 127. of the 19th century, implied a significant amount of ex-
133. Vl: See note to 37ff. pressive portamento. Many opportunities for expres-
165. Vl: Most leave this bar unmarked, but evidently sive shifting on the violin present themselves in this
expecting the 16ths to be played in the upper half movement; in view of Beethoven’s reported liking for
of the bow, as they would surely have been in portamento (4/a/iii), and his performing with Rodolphe
Beethoven’s time. Singer specifically instructs Spitze Kreutzer, who certainly employed it, portamento may
(point); Auer, Halir replace the staccato dots with well be implicit in the notation in many places where
tenuto lines, while Brodsky marks a slur over the legato passages require or invite shifting.
staccato from 165ii–166i. In addition to Beethoven’s notated dynamic and
168–172. lh: Presumably, Beethoven intended this to be accent marks, he would undoubtedly have expected
played legato and overheld where possible (4/a/ii). trained instrumentalists to inflect the music, in the
Diémer, Speidel, and Halir give various slurrings. way that singers would use to enhance emotive words,
Ganz marks poco legato and various sustaining ped- taking note of its melodic and harmonic contours and
al indications. its expressive qualities (3/a). In this respect, violinists’
171. Vl: All the editors except Hermann, Brodsky ex- and pianists’ markings help us to understand the ex-
tend the slur to xiii. pressive practices of 19th-century musicians.
1–8, 19–22, 39– 45. Pno: In 1 and 39 the grace-note should
Adagio con molt’ espressione be aligned with the beat (Ganz shows alignment
Tempo with a broken dotted line), with the main note fol-
Beethoven gave  = 84 for 3/4 Adagios in the Second lowing closely after (5/a/ii). In accordance with the
and Fourth Symphonies; these have more melodic expressive and singing quality of the theme (Spei-
32nd-notes than this movement, in which the majority del indicates cantabile, Halir dolce in 1 and probably
of the 32nds are in arpeggiated accompaniment pat- intended it at 39), all chords in lh might be arpeg
terns. For the Adagio cantabile of op. 18 no. 2, which has giated with varying speeds within a moderately
much more florid ornamental figurations, including slow range depending on melodic and harmonic
one in 64th-notes, he gave  = 72. In this movement, context (5/c/ii), except perhaps 4, 6, and 19–22 where,
with its cantabile melodic lines in long note values,
in both piano and violin, Beethoven would probably
have given a somewhat faster marking, perhaps in the 71 “Die Melodie mit dem möglichsten Ausdruck, der durch schö-
nen Anschlag und Ton hervorzubringen ist. Die linke Hand muss
region of Czerny’s in the Simrock edition. die Achtelaccorde nach ihrem Werthe halten. Die Bassbegleitung
Moscheles-Cramer  = 80 (vom 9ten Takte an) äusserst leicht und kurz. Vom 23sten bis zum
Czerny-Vortrag  = 80 38sten Takte alles leggatissimo und mit dem Ausdruck, der dem
Gesange der Violine entspricht. (Den 37sten und 38sten Takt mit
Czerny-Simrock  = 92 Pedal.) Bei dem nachfolgenden Thema ist bei jeder Taktviertel das
Alard/Diémer  = 50 Pedal ebenfalls zu benützen. Die Schlusstakte ritardando.”

following Moscheles’ advice, all the chords marked stroke through the stem. The portato is marked by
portato could be swiftly arpeggiated (4/b/i) and David, Joachim, Auer, Halir (with tenuto lines un-
held for the same length as notes with dots under der the slur), Rosé, Seybold, Kreisler. It would have
slurs, by which he intended a slight separation. All been executed by all trained 19th-century violinists
chords should be held their full length, as advised by with very little separation, more by pressure than
Czerny (Ganz indicates tenuto lines in 1 presumably by stopping the movement of the bow.
expecting tenuto to carry on appropriately). From 8ix–9. Vl: All the editions remain on the A-string go-
39– 45, frequent asynchrony with rh placed slightly ing from 2 on 7ix to 4 on 8i, undoubtedly with a
after lh (or exceptionally before) would be highly delicate portamento between the bowstrokes. In 9,
appropriate. In 2 and 42 the turn should be from fingerings divide between 4 –2–2 and 4 – 4 –2, the ma-
the upper note and quite late, after lh iii as notated jority choosing the former.
by Reinecke (5/b/i). In 6, the trills, preceded by a 9v–10i. Vl: Here the turn requires either a change of
note at the same pitch would certainly start from position or, what many editors indicate, an execu-
the upper note, and in 21, preceded by the note tion of the concluding two notes of the turn entirely
above, from the main note (5/b/ii). Where Beetho- with the 1st finger.
ven marked slurs, overholding would undoubtedly 9–15i. Pno: All chords in rh could be gently arpeggiat-
have been expected (4/a) and he probably also ex- ed and held for their full length (Speidel marks por-
pected judicious use of the sustaining pedal (Spei- tato; Ganz tenuto) with varying speeds within a mod-
del and Ganz provide various sustaining pedal erately slow range (5/c/ii). In lh the slurred broken
marks that might be helpful in this regard). The chords could be overheld to create resonance (4/a/
unison passage in the second half of 22 can be ren- ii), perhaps aided with judicious use of sustaining
dered highly expressive by employing a lilting style pedal as indicated by Speidel and Ganz.
(inequality) enhanced by occasional asynchrony be- 11iii. Vl: Many violinists around 1800 would probably
tween rh and lh. In 44 rh Beethoven’s syncopated have assumed a harmonic. It is explicitly marked
rhythms ought to be played as flexibly as possi- by Alard, Singer, Halir; other editors may have as-
ble, as if uttering passionate words. As advised by sumed it to avoid the awkward fifth across the string
Czerny, sustaining pedal could be used, especially with a single finger.
on a Viennese-action instrument of that era, on eve- 12. Vl: The turn is realised in a footnote by Singer as
ry quarter-note beat. From 39– 45 Speidel marks lh five 32nds beginning on d2, but by Halir as four
staccato with various sustaining pedal indications 32nds beginning on e2. Beethoven will probably
akin to Czerny’s; Ganz, too, marks some sustain- have expected the turn to begin with e2 and per-
ing pedal but far less than Czerny, and he indicates haps with the first three notes played rapidly.
tenuto in 39 lh, though it is not clear if this applies 13i–ii. Vl: Some editors remain in 2nd position, but
only to lh i or all lh notes in the passage. most shift from 1st or 2nd position to the c3 in 3rd
2, 40. Pno: The different slurring in these two bars sure- position, for the sake of an expressive portamento.
ly indicates no difference in execution; in both cases 14vii–viii. Vl: The vocal effect described by Spohr of
Beethoven would have expected legato through- changing the finger on the repetition of the same
out though perhaps with rh iv a little less in volume note (see note to op. 12/1/ii b. 55) is indicated by Her-
than rh i. The difference may have resulted from mann, Singer, and Brodsky.
Beethoven’s inconsistent notation or a copying or 15i. Vn: In the 1st edition, the appoggiatura is notated
engraving error. as e , which means either a 16th-note at this date,
/
3, 5, 41, 43. Pno: The absence of slurs on both occur- or potentially a grace-note (short appoggiatura). AG
rences of the theme in Pno, especially because of the and most editions notate it as an 8th-note appoggia-
slurs in Vl in 11, is curious. Beethoven surely expected tura, which is probably Beethoven’s intention, since
legato (Reinecke, Ganz, and Halir mark slurs), but his autographs do not use the notation e ; Hermann,
/
perhaps did not want to encourage overheld legato however, gives a full size 16th-note slurred to the g2;
on the double-dotted notes. Auer and Kreisler reproduce the notation of the 1st
6. Vl: The portato in 4 was surely envisaged here too, edition, but with a slur to the main note; Halir gives
though it is missing from the 1st edition where ii–v a full size 8th-note, slurred to the g2.
are given in abbreviated form as a half-note with 15–16. rh: The sf could be specially enhanced by an

asynchronous style playing the rh slightly after lh Auer, Brodsky), probably intensified the legato by
(5/c/ii; PT: 1/b). an audible connection between bowstrokes where
17–18. Pno: The expressive portato and slurred-pair shift and bow change coincide. This would be par-
patterns invite subtle rhythmic inflection and per- ticularly valuable for connecting the upbeat to 23,
haps even a broader tempo (4/b/i; 2/a; 3/d/ii). Rigid- using the fingering 2–2, by sliding the finger during
ity of rhythm and tempo here would contravene the initial bowstroke and changing bow simultane-
the norms of 19th-century ‘beautiful’ performance ously with the finger’s arrival at 23i.
(Ganz marks espressivo). 23–38. Pno: All broken chords in rh could be overheld
21. rh: The trill almost certainly starting on the upper (4/a/ii), and more resonance created through use of
auxiliary. the sustaining pedal as indicated by Speidel and
20–22. Vn: The portato marked in the 1st edition in Ganz. All octaves in lh might be slightly arpeggiat-
18f is surely expected to continue here as assumed ed, but especially those marked sf in 34, 35 and 37
in most editions. (5/c/ii). Beethoven’s perdendosi in 25 and 29 (see fol-
22v–38. Vn: The original slurring in this passage is lowing note) and the generally pp dynamic invites
challenging at anything less than the fastest met- use of the una corda or moderator if available (PT:
ronome suggestions; even Alard, however, with the 3/a and PT: 3/b). Speidel suggests una corda from 23
slowest tempo, retained the original slurring (al- with tre corde in 35. Ganz gives una corda in 23 and
though this does not necessarily mean that he used 27, and tre corde in 25 and the third quarter beat
it in performance), while Joachim, Singer, Auer, of 28. At 26 Speidel puts Beethoven’s pp in brackets
Brodsky, Kreisler indicate some changes of bow. and adds ppp with a footnote: “The dynamics that
Most take the upbeat in 22 and 26 . Singer marks are sometimes bracketed in this edition are Beetho-
espress. in 23, and at 34 gives a footnote referring ven’s, the added ones are those of the editors”.73 For
to 34 –37: “This wonderful, deeply felt passage re- Speidel’s footnote at 37 see previous note.
quires, in our opinion, a continual cresc. embued 25, 29. Pno, Vl: Perdendosi here evidently has the mean-
with passionate warmth up to the sf >”.72 See also ing, as Muzio Clementi instructed, “extinguishing
the note on legatissimo in the introductory para- gradually the sound, ’till it be almost lost”.74 Per-
graph to this movement. haps, like calando, it may also have been expected
There can be no doubt that violinists of Beethoven’s to elicit a slight relaxation of tempo.
time would have utilised portamento in this pas- 45. rh: The trills start with the lower auxiliary note
sage, which the composer surely expected. The edi- which should be aligned with the bass as shown by
tions offer various expressive fingerings. Between Ganz with dotted line notation (5/b/ii).
23iii–iv 4 – 4 is indicated by Singer, Brodsky, Halir; 46f. Vl: Tenuto lines under slurs are marked by Singer,
David, Alard, Hermann, Joachim, Seybold shift be- Halir.
tween iv–v, with the less intense 2–1. In 27 Singer, 46– 47. Pno: The chords on i might be very gently and
Brodsky make the same shift on the D-string; oth- swiftly arpeggiated (5/c/ii). The sf chords on iii could
ers remain in 3rd position on the D- and A-strings. either be swiftly arpeggiated or asynchronised with
Some shift expressively in 24, 28, while others re- rh slightly after lh (PT: 1/a and PT: 1/b). Ganz gives
main in position. In 25, 27, some make an expres- sustaining pedal marks that assist the resonance of
sive shift from i–ii. In 31, all make an expressive the dotted figures.
shift either between i–ii (Singer, Rosé, Kreisler) or 48. rh: The turn from the upper note (5/b/i).
ii–iii (David, Alard, Hermann, Auer, Brodsky, Ha- 48f. Vl: On 48ii, all except Alard begin , evidently
lir, Seybold) and in 33 all but Joachim and Brodsky at the point of the bow, executing the sf on . All
shift i–ii (most) or ii–iii (Hermann). At 34i–ii, 35i–ii, but Joachim, Kreisler mark slurred staccato/portato
36i–ii all make shifts, all with 4 –2, 4 –2, 3–1, except in 49; some , some , some in several bows. Halir
Kreisler, whose use of more intense and continu- marks tenuto lines under slurs.
ous vibrato probably caused him to mark 3 on 34i
and 35i. Those who take additional bows (Joachim, 73 “Die zuweilen in dieser Ausgabe vorkommenden eingeklam-
merten Vortragsbezeichnungen sind Beethoven’sche, die beigefüg-
72 “Diese wunderbare, tieferregte Stelle erfordert, nach unserer ten solche der Herausgeber”
Ansicht, ein mit leidenschaftlicher Wärme empfundenes, anhal- 74 Clementi: Introduction to the Art of Playing on the Piano Forte,
tendes cresc. bis zum sf>”. vol. 1, p. 14.

50–59. Pno: All chords except 53 rh ii could be gently elicits lingering and perhaps also asynchrony, play-
arpeggiated (5/c/ii). In 51 Ganz marks una corda for ing rh ii after lh. In 70 rh, the particularly poign-
Beethoven’s pp with tre corde in 52, which would be ant alternation of double-notes in 6ths (which Ganz
appropriate. Beethoven’s < > in 52 invites both marks espressivo) requires rhythmic flexibility (a lilt-
arpeggiation and lingering at the apex (3/b/v). In ing style) (2/b), aided by subtle asynchrony, with rh
53 and 54 the rh octaves marked portato could be xiii–xvi given maximum expression through asyn-
swiftly arpeggiated with the notes held almost full chrony and/or arpeggiation, lingering at the apex
length (4/b/i), and the lh chords marked sf might of the < > and slowing down as suggested by
similarly be arpeggiated or asynchronised by play- Speidel, who marks slentando. In 71, the chord on i
ing them slightly before rh (5/c/ii; PT: 1/a and PT: 1/b). might be somewhat lengthened which Halir marks
For maximum contrast at 58 the una corda or mod- with a tenuto line and played slightly louder than
erator (if available) might be used. In 59, the chords its resolution.
would produce the required strong energy by being 63f. Vl: The editors who mark fingering envisage por-
very tightly arpeggiated and accented (Ganz marks tamento, either between 64 i–ii (4 – 4) or ii–iii (2–2).
marcato). Ganz gives sustaining pedal marks, for ex- 66–68. Vl: Most editors divide Beethoven’s slur. Some
ample to assist the resonance of the dotted figura- editors evidently remain in 1st position, using an
tions in lh 53–55, but advises senza pedal in 51. extension for c3. Others begin in 3rd position and
52. Vl: Editors suggest a variety of fingerings. The shift on 67 i–ii or ii–iii. Singer, who marks molto es-
expressive fingering beginning with a portamento pressivo, shifts 3–3 on i–ii in both bars. All indicate a
3–3 and a harmonic on a2, which is given above the change of bow except Joachim, who, however, surely
notes in the edited violin part of the present edition, made one in practice. If 66f is taken in 1st position,
is suggested by Alard. David, Joachim, Auer begin it could be effective to make portamento between
1–3, with more reticent portamento and probably the bowstrokes, as in 52, by sliding the 1st finger
also used the harmonic, though they do not mark back towards f natural just before the bow change.
it. Singer, Brodsky, Halir begin 1–2 and execute an-
other portamento playing the last two notes 3–3. Rondo
Whether the editors who retain the 1st edition slur Allegro molto
over the whole bar actually played this in one bow Tempo
is questionable. Alard, at his specified tempo, could The closest analogy to this movement among those
only have done so if he played very quietly; David, for which Beethoven gave metronome marks is the
in his personal copies, often breaks up longer slurs, Allegro molto quasi presto of op. 18/2/iv, to which he gave
even in his own editions; and Joachim was noted  = 92, but in that movement there are many fewer
for his variability in performance. 16ths, and they are also in less complex patterns. Com-
56. Vl: Those who used slurred staccato/portato in 49 parison of this movement with the very similar third
do so also here in various ways. movement of op. 30 no. 3, marked Allegro vivace, for
60–61. Pno: The lh octaves could be gently arpeggiat- which Czerny and Moscheles give faster metronome
ed (5/c/ii), while overholding and over legato in rh marks, is interesting.
would produce beautiful resonant effects (4/a/ii). Moscheles-Cramer  = 152
62–71. Pno: The appoggiaturas to the trills, marked Czerny-Vortrag  = 72
by Beethoven, should be aligned with the bass as Czerny-Simrock  = 144
shown by Ganz with dotted line notation (5/b/ii). Alard/Diémer  = 112
All lh chords might be arpeggiated apart from 71ii Speidel/Singer  = 132
(5/c/ii). In 65–66, the rh broken chords could be Kreisler/Rupp  = 132
overheld (4/a/ii). From 62–68, Speidel and Ganz sug- Czerny writes briefly: “Very lively and with all the
gest sustaining pedal patterns that are suitable for vigour of powerful, assertive, and brilliant playing.” 75
the passage. In 66–67, Beethoven’s < > elicits 1–8, 87–94, 163–171. Pno: Even at this fast tempo and
a slight increase of momentum towards and linger- very much in line with Czerny’s description, all
ing at the apex with a slowing down afterwards chords, apart from those marked staccato and 4 rh ii,
(3/b/v). In 68–69, the rh portato chords could be 75 “Sehr lebhaft, und mit allem Feuer eines kräftigen, entschiede-
swiftly arpeggiated (4/b/i), and Beethoven’s < > nen und brillanten Spiels.”

might be arpeggiated very tightly or swiftly (5/c/ii), 24vi. Vl: The open string is marked by Joachim, Sin
to achieve the effect on the ear that Samuel Wesley ger, Auer, Halir, Kreisler, and was probably as-
described in 1829 (using “a harsh Military Meta- sumed by others.
phor”) as “a Running Fire”.76 In 1, 2, 5, and 6 a special 24ff. Pno: Asynchrony in which important melody
agogic accent can be created by playing the lowest notes in rh are sounded very slightly after lh would
note of lh on the beat and playing the remaining lh help bring them into relief, especially useful for
and rh notes together unarpeggiated and immedi- those marked sf (PT: 1/b). Here, too, it is uncertain
ately after (PT: 1/a). Speidel marks tenuto above 1 rh whether Beethoven intended legato or non-legato in
i and 2 rh i. lh. Diémer marks slurs until 38. Speidel adds stac-
8–16. Vl: The bowing of this theme is treated differ- cato and non legato from 24 to halfway through 30,
ently by the various editors. Alard, Hermann, Kreis- followed by slurs until 38. From 27–32i Halir marks
ler take all the upbeat 8ths with separate bows, sug- tenuto lines on lh i, iii, v and vii perhaps indicating
gesting the use of the middle and upper half of the over-legato. Ganz marks sustaining pedal through
bow; David’s bowing (also given in Rosé, Seybold), each bar from 24 –30 but in 32 marks ben articulato
with   on 8i–ii, 9ii–iii, 12ii–iii, 13ii–iii, but sepa- without sustaining pedal.
rate bows on 10ii–iii, suggests performance near the 25, 29. rh: Overholding is appropriate (4/a/ii).
point, which may also be implied by Halir’s use of 39ii–iii (43 ii–iii, 182 ii–iii, 186 ii–iii). Vl: Editors di-
down-bow slurred staccato on 11i–ii, but up-bow vide between those who take the notes with alter-
staccato for the other upbeat pairs, while Joachim, nate bows (Alard, Joachim, Auer, Kreisler) and the
Singer, Auer, with slurred staccato only in 9f, 12f and rest, who mark  .
separate bows in 8, 11 may imply execution closer 40–51, 187ff. Pno: Swift arpeggiation or asynchrony at
to the middle of the bow. 40, 44, 48i, 49i, and 51 would provide the requisite
9–15, 78–85, 170–172. rh: It is not certain whether Beet- energy and accent without excessive harshness (PT;
hoven expected legato or non-legato. Given the fast 1/a and PT: 1/b). In 41– 43 and 45– 47, the rh broken
tempo and the type of figuration, legato without octaves might most characterfully be played non-
overholding is possible and perhaps appropriate (4/ legato. Speidel marks them leggiero.
a/ii). Diémer slurs each bar; Speidel marks non le- 41f (45f, 184f, 188f). Vl: Joachim-revised, Auer mark
gato. In 169–171 many editors mark slurs. The < > 41ii-iv with slurred staccato; Singer, Brodsky mark
in 171–172 might elicit a slight agogic accentuation slurred staccato from 41ii– 42iv.
and asynchrony at the apex in addition to dynamic 51ii–58 (194ii–201). Vl: Bowing patterns suggest dif-
nuance. ferent approaches to the bowstroke. Those who take
16ff, 174ff. Pno: All chords not short nor marked stac- the sf in 55  (Alard, Joachim, Singer, Auer, Halir,
cato might be swiftly arpeggiated (5/c/ii). Rosé) evidently play the passage towards the mid-
20, 178. Pno: Asynchrony at i would help to mark the dle of the bow (Singer in fact marks leggiero, a typi-
rising 7th interval in rh (PT: 1/b). cal indication for a sautillé bowstroke). Those taking
18, 22. Pno, Vl: The trill was certainly expected to the sf , probably fouetté (David, Hermann, Seybold,
begin with the upper auxiliary (5b/ii). Depending Kreisler), clearly favoured execution near the point.
on the tempo, a turn from the upper note might be 52–60i, 194ff. Pno: Beethoven surely expected the rh
substituted. broken chords to be overheld to create as much res-
20i. Vl: The point of the bow is suggested by David’s onance as possible (4/a/i). Diémer and Speidel marks
 here. Execution nearer the middle may be implied slurs, Ganz gives sustaining pedal indications. Ar-
by the other editors’ use of  or no bowing indica- peggiation and or asynchrony will help enhance the
tion. sf in 55i and soften the arrival at 59i.
23i–24i. Vl: The harmonic with 4th finger, to avoid 59–67i, 203ff. Pno: At 60i, 62i and 64i, asynchrony will
an awkward 5th across the strings, may well have be expressively effective. The chord marked sf at rh
been regarded as self-evident by many violinists. 63 would gain much energy with a swift arpeggia-
Singer gives an individual fingering for 22iii–iv, 23i: tion (PT: 1/a). 67i could be softened with arpeggia-
2–02–3. tion.
76 Philip Olleson, ed., The Letters of Samuel Wesley: Professional and 67–78, 210ff. Pno: Beethoven probably expected the
Social Correspondence, 1797–1837 (Oxford, 2001), p. 439. broken chords to be played legato with overholding,

though he may also have permitted a mixture of le- 109–116, 131–138. Vl, Pno: Although the 1st edition has
gato and staccato articulations. From 67–72, Diémer no staccato marks in Vl, Halir and Seybold added
offers a mixture, Speidel a similar way, though at 75 them. Others may well have expected the kind of
he marks non legato, but in a footnote he explains “soft spiccato” recommended by Rostal,79 perhaps
that legato is permitted from 67 and again at 75. Rei- similar to David’s “hopping bowstroke” [hüpfender
necke and Halir slur. Strich] executed in the upper middle of the bow
60i, 62i, 64, 66i. Vl: In practice the staccato mark (only without the hair leaving the string. This kind of
in 60 in the 1st edition) signifies merely that the slur bowstroke, however, seems very unlikely to have
begins on ii, not that the note should be shortened. been envisaged by Beethoven, who may even have
86. rh: Presumably legato as given by many of the edi- expected the smoothest kind of portato, perhaps
tors. slurred four notes to a bowstroke or perhaps very
87–93, 163–169. Vl: The unmarked 16ths of the 1st edi- soft separate bows, executed close to the point. The
tion are ambiguous. In such circumstances it is very absence of staccato is paralleled in the repeated notes
unusual for Beethoven to leave the notes unmarked of Pno lh throughout these bars, which accompa-
with neither slurs nor staccato; the absence of staccato ny a passage of 16ths under a single, long slur in
marks makes it unlikely that he expected each note rh, perhaps with over-legato touch (4/a/ii). For lh,
to receive a separate bow. Most violinists of Beetho- however, Speidel marks staccato, and Ganz staccato
ven’s time would probably have played them either dolce. Use of the una corda (as suggested by Ganz)
slurred throughout in whatever patterns were most or a moderator (if available) is appropriate (PT: 3/b).
convenient or with a mixture of slurs and separate 117–123, 139–145, 154 –156. Vl: Most editors employ
bows (6, p. X X X I I ). Alard, Hermann, Brodsky leave slurred staccato on iii–iv in piano and many also in
the passage as in the 1st edition; all the other editors forte. Slurred staccato in such circumstances would
slur i–ii in 87–92/163–168 and i–ii, v–vi in 93/169 ex- also have been a very typical late-18th-century bow-
cept Auer, who adds many more slurs. Max Rostal, stroke executed between the point and middle of the
evidently assuming separate bows, favoured a “light bow, as described by Reichardt in 1776 (6, p. X X X I I ).
jumpy bow (sautillé)”,77 but this seems definitely to 139–146, 153–157. rh: As much over-legato as possible
be ruled out by the absence of staccato marks. (4/a/ii).
95–105, 118–128, 147–150. Pno: The nature of the slur 157–162. Pno: At lh 157i, 159i and 160i, arpeggiation is
red pairs together with the sf elicits an agogic accen- appropriate. Given the decrescendo, the < > in
tuation that might best be achieved by asynchrony 159–160 might elicit an immediate broadening.
between the hands or else playing the lowest note 218–225. Pno: All the rh broken chords overheld. Rei-
in lh on the beat with the remaining notes in lh and necke marks sempre legato, Diémer slurs, Ganz sus-
rh slight afterwards and unarpeggiated (PT: 1/a and taining pedal, but Speidel marks non legato.
PT: 1/b). 219–221. Vl: Most editors hook the bowing bar by bar,
97f, 101f. Vl: Most editors employ slurred staccato. a typical practice of Beethoven’s time.
Those who take separate bows almost certainly ex- 222f. Vl: Most mark  for the off-beat 8th-notes and
pected the 8th-notes to be played in a sharply ac- the others surely assumed them, as would many
cented manner (martelé) near the point. violinists around 1800.
105–107. Vl: Most editors probably envisaged martelé 229f. Vl: Most begin  in 228 and slur from 229v–230ii.
near the point here, and this was also recommend- 230–233. rh: Probably all legato as marked by Rei-
ed by Rostal, as well as in 97, 101.78 necke and Diémer and with overholding (4/a/ii).
108, 130. Pno, Vl: Beethoven may have expected the 245–248. Vl: Most mark the 8th-notes  , though in
chords to be played staccato. Reinecke/Hermann, 248 David marks them  .
Halir, and Ganz/Auer mark staccato, Speidel stac- 246–248. Pno: The effect of sf might be enhanced ei-
cato with > . ther by asynchrony between rh and lh or else swift
arpeggiation of rh (5/c/ii; PT: 1/a and PT: 1/b).
77 Max Rostal, Beethoven: The Sonatas for Piano and Violin. Thoughts 258, 262. rh: Probably legato as marked by many of
on Their Interpretation (London, 1985), p. 66; trans. Anna M. Rosen- the editors, and overheld (4/a/ii).
berg and Horace D. Rosenberg from Beethoven. Die Sonaten für Kla-
vier und Violine (München, 1981).
78 Ibid., p. 66 79 Ibid.

259–273. rh: All legato with overholding (4/a/ii). Rei- Czerny’s comments on this movement make it even
necke marks sempre legato; other editors mark slurs. more surprising that his recommended tempo is so
Ganz gives sustaining pedal indications. At 266, moderate. He wrote: “Extremely fast and never drag-
Speidel changes dynamic to pp and marks una corda ging. Although serious in character this movement must
with tre corde at 275. rather be played lightly than passionately, since the
266ii–275i. Vl: Only some editors mark fingering. interest already lies in the continually rapid tempo.” 82
Some suggest remaining in 4th position throughout. For this sonata, early-19th-century markings, from
David, Seybold provide an interesting and effective a copy of the violin part (Österreichische Nationalbib-
fingering that keeps the whole passage on the E- liothek shelf mark: SH. Beethoven.121, plate number
string, sliding back to 1st position with 4th finger be- M.1124 c. 1808) will be referred to as Anon.
tween 268i–ii and 272i–ii and returning via 2 on 270ii. 1, 3, 70, 702 etc. Pno, Vl: The two-note ornament (Slide,
274, 278. Pno: Swift arpeggiation would be appropri- Schleifer) was surely conceived as occurring with,
ate (5/c/ii). not before the bass note (5/b/i). There were divergent
opinions about its accentuation in Beethoven’s time:
some considered the accent to fall on the ornament,
others on the main note. Where fp is marked, as
SONATA OPUS 23
here, it seems probable that Beethoven envisaged
Presto a vigorous, but not sharply accented attack on all
Tempo three notes, dying away after the beginning of the
All 19th-century suggestions for tempo in this move- main note. In practice, when the short notes are
ment seem very slow when compared with Beetho- played as rapidly as possible (which is surely what
ven’s extraordinarily rapid, and perhaps even unreal- was expected), the accent is in any case perceived
istic marking . = 96 ( . = 192) for the 6/8 Presto finale as occurring on the main note. In a style of per-
of his String Quartet op. 18 no. 3, which contains still formance where vertical togetherness in the mod-
more unremitting 8th-notes than this movement. Why ern sense was not expected, this is unproblematic.
such steady tempos should have been selected, even In 1 and 3, Speidel marks rh i ten., perhaps implying
by Beethoven’s contemporaries, is hard to explain; an accent. Ganz marks dotted lines connecting the
Beethoven himself, for the second part of his cantata first note of the ornament to the bass.
Meeresstille und glückliche Fahrt op. 112, which is in 6/8, 1–12i, 72–75, 84 –91, 136–143 and 152–159. Pno: Over-
marked Allegro vivace, and even includes passages holding of notes in the broken chords would add
of 16th-notes, gives . = 138, which is precisely the much to the character, energy, and resonance (4/a/ii).
same metronome mark Haslinger and Moscheles sug- 5–10, 59f, 168f, 173–175. Pno: Swift asynchrony with
gest for this movement. It seems almost certain that rh slightly after lh of the notes marked sf would
Beethoven’s use of Presto for the first movement of op. enhance the effect (5/c/ii; PT: 1/b). The grace-note in
23 implies something much faster, probably at least 8 (171) played together with the main note and re-
. = 88–96, or, as in the case of some of his other very leased immediately after (5/a/iii).
fast movements, as fast as it can practicably be played. 5ii. Vl: It is likely that most violinists around 1800
In fact, Max Rostal evidently recognised this problem would have taken the sf , following the natural
and specified an exceptionally wide range ( . = 132–176) bowing sequence. Most of the 19th-century editors
in his list of suggested tempos,80 though he did not mark this, evidently performing the passage between
explain his reasoning in his discussion of the sonata.81 the point and middle of the bow until 11. Joachim,
Haslinger . = 138 Auer however mark  here.
Moscheles-Cramer . = 138 10iv. Vl: Singer marks a harmonic with 4, followed by
Moscheles-Meyer . = 138 2 on the E-string. A fingering of this kind would
Czerny-Vortrag . = 132 not have been uncommon in the early 19th century,
Alard/Diémer . = 132 when vibrato was sparingly employed.
Speidel/Singer . = 132
Kreisler/Rupp . = 132 82 “Äusserst schnell und nirgends gedehnt. Obwohl von ernstem
Character, muss dieser Satz doch mehr leicht als leidenschaftlich
80 Ibid., p. 31f. gespielt werden, da das Interesse schon in der rasch fortlaufenden
81 Ibid., pp. 68–70. Bewegung liegt.”
© 2020 by Bärenreiter-Verlag, Kassel Opus 23 ▪ 47

12 etc. Pno: Here and in similar places throughout the est note of the lh on the beat and the other notes
movement swift arpeggiation of i would produce in lh and rh following closely and unarpeggiated
the requisite energy following the cresc., and swift would create a special sf with an agogic effect (PT:
asynchrony with rh after lh could be used on ii to 1/a).
create special emphasis for the beginning of the slur 52i. Vl: The sf can be achieved just as well with  or ;
(5/c/ii; PT: 1/b). the 19th-century editors are almost equally divided.
12ii–16i. Vl: Most editors mark  on 14ii, probably start- 53–54. lh: Swift arpeggiation of the octaves would
ing the preceding f close to the frog, playing 13 help to create fiery energy.
in the middle of the bow with a stroke to the point 58–61, 210–213. Pno: Given the ff dynamic, the down-
on 14i; then executing 15 near the point. This would beats in 58–60 would gain much effect by special
be a very plausible bowing for Beethoven’s time. arpeggiation: perhaps in 58 a swift spread and in
Singer, however, marks  on 14ii, 15ii, 15, iv, evi- 59 and 60 a similar type of spread to 52i above. 61iv
dently taking them close to the frog. could also be swiftly arpeggiated to give emphasis
14ii–19iv. Pno: Swift arpeggiation of chords not mark to the beginning of the slur in rh.
ed staccato would create a softer edge (5/c/ii; PT: 1/a). 62–68, 214 –220. Pno, Vl: All chords marked sf in rh
21f. Vl: David’s and Halir’s use of the middle of the could be specially emphasised with a quick arpeg
bow for the f is indicated by their slurred staccato giation (PT: 1/a). The reiterated octaves in lh could
over 21iv–vi taking the bow towards the frog for be given energy by a quick, almost imperceptible
the sf on 22i. arpeggiation. The first notes of the slurred figures
22–29. Pno: The use of asynchrony at 22i for the sf in Pno and Vl in 62 and 64 might be given spe-
and various downbeats, and arpeggiation at lh i in cial emphasis, though not as much as sf, in order
24 –28 will be an aid to the softening of expression to mark their unusual placement as recommended
(5/c/ii; PT: 1/a and PT: 1/b). by C. Ph. E. Bach, Daniel Gottlob Türk, and others
24 –29. Vl: Alard, Joachim, Brodsky give the bowing in the second half of the 18th century (2b). In both
of the 1st edition exactly. All the others, including Pno and Vl, Halir contravenes this by marking <
Anon, adapt the slurring to achieve more convenient before the sf in 63 and 65 and > after. In Pno and
bow distribution. If Beethoven’s slurring was intend- Vl, Speidel/Singer and Halir mark > after the sf
ed to indicate phrasing, rather than simply legato in 65 and 66. Speidel/Singer mark < after the sf
(which is by no means certain), this can be achieved in 67, but Halir marks another > .
with longer slurs, but slightly greater emphasis on 72ii–76. Vl: This figure seems to have been problem-
25i and 26iv than would otherwise be the case. atic for many 19th-century editors, who made vari-
29–50. Pno, Vl: This passage may invite a slight re- ous changes, although it is perfectly easily executed
duction in tempo, with an accelerando back to the as written, starting  at the point on 72ii and taking
original speed from the cresc. in 46 to the f in 50. the fp in 74  in the middle of the bow.
30– 45i, 182–197i. Pno: The overlapping (polyphonic) 76–82. lh: The lh chords could be given rambunctious
thematic material invites the use of arpeggiation energy, particularly 76 and 80, by swift arpeggia-
to create delineation, for example, where one part tion (5/c/ii). Ganz gives sustaining pedal indications
ends and another begins, perhaps best applied at at 80i and 84i.
such places as the end of 31, 41 and 42 as well as 76–83. Pno, Vl: Even at presto tempo the first note of
other places (PT: 1/a). the slurred pairs could be made a special feature by
33–37. Vl: The long slur is unproblematic, especially at giving them emphasis both by dynamic and length
the very fast tempo Beethoven undoubtedly envis- (2/a). In Pno Speidel marks tenuto lines on 76 and
aged. Most 19th-century editors retain it. 80 rh i; Halir marks tenuto lines on 78 and 82 rh i.
45– 48. Pno: Speidel marks > on lh: 45ii, 46iii, and 47iii, 78. Vl: David begins in 3rd position and extends the
and rh: 46i, 47i, and 48i. 4th finger for the e 3, before bringing it back for the
48. Vl: AG prints the slur, questionably, from ii, and d3, a perfectly effective and secure fingering, typi-
this is followed by later editors. cal of the early 19th century, which, however, was
49–57, 197–209. Pno: Asynchrony (with rh after lh) at removed in David-revised. Most, however, includ-
downbeats and other places (where possible) would ing Anon, begin in 4th position, coming back to 3rd
be stylistically appropriate. At 52i playing the low- with the 4th finger.
48 ▪ Opus 23 © 2020 by Bärenreiter-Verlag, Kassel

78–83. Vl: Many later 19th-century and most 20th- clear that the crescendo from 92 goes no further than
century violinists would execute the 8th-notes in f in 95.
the lower half of the bow, allowing the two-bar slur 94, 100, 106. rh: A swift arpeggiation of the octave
from 84 to be taken . This would not have been the marked sf would produce an energised accent with
natural choice for a violinist of Beethoven’s time, harshness.
who would instinctively execute fast notes of this 94 – 95, 100–101. Pno: Ganz gives sustaining pedal
kind in the upper half of the bow. It is possible to marks presumbly to give special resonance that
use the middle of the bow and still play  in 84 from would heighten the effect of these accented notes.
middle to point, but David, Hermann, and Halir, 95ii, 101ii, 107ii. Pno: Either asynchrony with lh slight-
clearly executing the 8th-notes nearer to the point, ly after rh or arpeggiation are appropriate here for a
all take 84 . This bowing was removed in David- gentler accent than at rh ii in 94, 101, and 107, which
revised. Speidel marks with > .
79vi. Vl: An open string is marked or evidently as- 97, 103, 109. Pno: To delineate between the different
sumed by all editors. figures in rh and lh, asynchrony (with lh slightly
83iv–vi. Vl, Pno: Singer, Halir mark > rather than after rh) at rh ii would be effective (5/c/ii; PT: 1/a).
expecting a subito p. Halir and Speidel mark tenuto lines on rhi 97 and
84i. Vl: The open A-string (indicated by Alard, Brod- 103 rh i, and Halir marks another on 109 rh i.
sky) was evidently assumed by the others. 110. Pno: Asynchrony (with rh hand slightly after lh)
84 –90. Pno: Beethoven’s > signs might elicit asynchro- would produce an appropriately abrupt energy to
ny with rh slightly after lh, which would help to launch the highly energetic ff section that follows.
bring out these successive pedal notes (5/c/ii). 111–118. lh: A fast and especially accented arpeggia-
84 –91. Vl: This passage was surely conceived by Beet- tion with the chords held full length is appropriate
hoven with vocal portamento in mind. From 85i–ii here both for abrupt energy and textural support
and 87i–ii Anon marks shifts (2–3 and 4 –1), which (5/c/ii). Halir marks tenuto lines on rh i in 113, 115,
are also given by all the editors. In 84 Singer adds and 117 presumably to create a slight agogic accent
the instruction cantabile; Auer and Rosé espressivo. which would not be inappropriate here.
After an open A-string on 88i, Anon and all editors 120. Vl: The fp should almost certainly be on i, not ii.
except Hermann, Brodsky, Kreisler mark a harmon- Despite Beethoven’s propensity for the unexpected,
ic on 89ii (Singer with 2nd finger and Rosé with 3rd) its alignment with the fp in Pno rh and lh seems
followed by 3rd finger on 89iii. Most remain on the more likely; and from a musical point of view, too,
A-string in 90f, fingering 2–2– 4 – 4, and stay on the a subito p on ii seems more characteristic. Its mis-
A-string until 93. Some violinists of Beethoven’s placement could easily have resulted from cramped
time may, like Anon, have remained in 3rd position writing in the autograph, or from a simple copy-
until 91iii, descending to 1 on 91iv. ing or engraving error. Beethoven’s self-confessed
84 –110. Pno, Vl: The whole of this passage may invite fallibility in proof-reading and numerous verifiable
a slight reduction in tempo, with an accelerando instances of uncorrected mistakes and ambiguities
back to the original speed in the final bars, espe- in the sources make error perfectly plausible. Only
cially with the cresc. Alard, however, considered it a mistake and placed
92. rh: A swift arpeggiation would help fill out the it on i.
texture and give a slightly dramatic surge (5/c/ii). 120–124. Pno: For immediate contrast, asynchrony
92–110. Pno, Vl: The dynamics in the 1st edition are (with lh slight after rh) or arpeggiation wherever
problematic. Strictly followed, they indicate an 18- possible will have a noticeably softening effect (5/c/
bar crescendo from p to ff. It seems highly likely, ii).
however, that one or more dynamics were omitted 121f. Vl: Anon continues the slur to 121iv.
through oversight, either by Beethoven, by a copy- 122–128. Vl: Few editors left the 1st edition slurring
ist, or during engraving. The Singer/Speidel edition entirely unrevised; several extend slurs across bar-
addresses this problem in the piano part, adding lines.
many additional dynamic inflections. Halir simply 129–131. Vl: All editors use a hooked bowing for the
adds f on 94ii in the violin part. Others make no ad- 8th-notes, as was surely envisaged.
dition. In view of Beethoven’s cresc. in 108, it seems 132–163. Pno, Vl: Singer/Speidel add allargando in 132

and poco rit. in 134 with in tempo at 135iii. Ganz obvious; Singer, Auer, however, who take the pre-
marks sostenuto in 132. A relaxation of tempo in 132f ceding bar up the D-string, mark 2.
may have been implied by Beethoven’s > or, in 198–214i. Vl: See note to 46–62i
any case, by the approach to a fermata. The follow- 214ii–220. Vl: See note to 62ii–68
ing passage may well have been taken at a slightly 224. Vl: Anon, Alard, Singer, Auer, Brodsky mark a
slower tempo than that of the movement as a whole, harmonic, which early violinists might well have
with a gradual accelerando back to Tempo 1 during used in this context, while others remained in 1st
the cresc. from 156. In 132 and 134, asynchrony in position.
Pno with an especially expressive delay (more than 227iv–v. Vl: Anon and all except Alard, Hermann,
usual) is appropriate to enhance this highly expres- Brodsky mark the staccato and slur in a single up-
sive moment. Pianists of Beethoven’s era might have bow, taking a new  in 228. Only Hermann, Brod-
felt inspired to ornament the notes in 134 –135 per- sky retain the 1st-edition slur across the barline to
haps with a simple flourish. 228.
134iii–143. Vl: Singer, Brodsky, Rosé remain on the 233f. Vl: Singer, Brodsky, Rosé go up the D-string.
D-string. 242i–ii. Vl: Alard, Singer employ open A-string and
135. Vl: Anon and all editors (the latter following AG) harmonic.
extend the slur to 135iii. 247–262. Pno, Vl: The rests might be extended for rhe-
136, 137, 141, 142, 225, 226iv–v. Vl: Anon and all edi- torical effect, for which Beethoven was particularly
tors except Kreisler hook these notes in a single bow. admired. Gentle arpeggiation of the final three
136–162, 224 –243. Pno: To delineate between rh and chords in Pno would be appropriate and effective
lh figurations, frequent asynchrony (with lh slight- for the fading away. Several editors mark these
ly after rh) would be effective and appropriate to chords staccato.
create subtle dynamic and agogic accents (5/c/ii; PT:
1/a). Halir gives various > and tenuto lines which Andante scherzoso, più Allegretto
may have implications for asynchrony. Tempo
155–159, 235–237iv–v. Vl: All editors except Kreisler See the notes to op. 12/2/ii for comment on Beetho-
hook these notes in a single bow. ven’s treatment of 2/4 Andante/Allegretto tempos. For
164. Vl: The arpeggio instruction in the 1st edition, not this movement a faster tempo might be suggested by
included in any later editions (see Critical Report), his use of the qualifier, scherzoso, though this is more
was perhaps Beethoven’s shorthand for a spread likely to characterise the performance style, perhaps
chord in which the a2–e2 reflect the grace-notes in indicating shorter execution of staccato notes than
other occurrences of this figure. As a three-note might otherwise have been the case (see note to b. 40).
chord, it might otherwise have been played almost For the 2/4 Allegretto scherzando of the Eighth Sym-
exactly together. phony, Beethoven gives  = 88, but in that movement,
173iv. Vl: Anon, Alard, Singer slide the 4th finger to much of the thematic movement involves 32nd-notes,
a harmonic. suggesting a relationship in which the 8th-note rather
176–179. Vl: Beethoven’s slurring here is impractical than the quarter-note is seen as the basic time unit
and neither Anon nor the editors leave the slurring (perhaps really conceived as 4/8). That Beethoven’s
exactly as in the 1st edition. faster Andantes could be as fast as his Allegrettos
176–181. lh: Arpeggiation of the octaves would help is demonstrated by his metronome mark for the 3/8
fill out the texture and give character to these pedal Andante scherzoso quasi allegretto in the String Quar-
points (5/c/ii). tet op. 18 no. 4 and the 3/8 Allegretto vivace e sempre
181ii–189i. Vl: The majority take the passage in 1st po- scherzando in the String Quartet op. 59 no. 1, to both
sition with an open D-string in 184, which would of which he gave the same metronome mark of . = 56.
probably have been the choice of most violinists in The Haslinger tempo is probably close to what Beet-
Beethoven’s time. In 184f, however, Singer marks hoven would have given.
4 – o and several editors take the entire passage on Haslinger  = 92
the G-string. Moscheles-Cramer  = 84
194. Vl: Alard, Seybold mark o, which most other Moscheles-Meyer  = 92
editors, utilising 1st position, evidently regarded as Czerny-Vortrag  = 92

Alard/Diémer  = 138 16. Pno, Vl: Speidel/Singer mark espressivo perhaps re-
Speidel/Singer  = 72 sponding to the cessation of staccato marks on the
Kreisler/Rupp  = c. 92 second note under the slurs, although it is unlikely
Czerny describes the movement as “A merry, charm- that the composer intended a difference, merely as-
ing joke, to be performed with humour and tender- suming the continuation of the previous notation.
ness. The tempo always lively. The fugal passage well 30ii–31, 153ii–154. Vl: All earlier editors except Jo
marked. The last 12 bars of the first part very gentle achim mark 30ii/153ii as a harmonic followed by 2
but also not drawn out.” 83 on 31i/154i.
Beethoven will have expected trained players to 32–33iff. lh: Beethoven would have expected legato
add many more dynamic nuances than he marked. across the bar line (4/a/ii), similarly for Vl 36–37.
A sense of what 19th-century musicians would have Speidel marks 33i > , Singer the same on 37i.
added might be gained from studying the annota- 33, 37, 40, 44ff. Pno, Vl: The trill in this figure, pre-
tions of Speidel/Singer, Halir, and Auer/Ganz. ceded by the note above, certainly starts from the
Throughout this movement all chords, apart from main note (5/b/ii).
those marked staccato or forming resolutions, might 36. lh: The grace-note played together with the main
be arpeggiated at a generally moderate speed, though note and raised immediately after (5/a/ii and iii).
with variation according to melodic and harmonic 38–39, 164 –165. Pno: The 16ths presumably to continue
context (5/c/ii). staccato. Many of the editors mark them thus.
0–16 etc. Pno, Vl: The slurred pairs of 8th-notes will 39i. Vl: All indicate 3rd finger on 39i.
have signified slight strengthening and perhaps even 40. Vl: leggiero is marked by Hermann, Seybold, sug-
lengthening of the first note under the slur, which gesting that they envisaged the kind of hüpfenden
would enhance the feeling of scherzoso. Bogenstrich described in David’s Violinschule, where
8–30. Vl: David, Hermann, Rosé indicate  successive- the elasticity of the bow-stick is brought into play,
ly until  on 23i, before resuming  until 30ii; Alard using a short stroke between the middle and two
provides only  on 8i and  on 23i, but probably also thirds of the way towards the point, but without the
expected repeated  this may have been executed bow-hair leaving the string. Probably other players
just below the middle of the bow, beginning each  envisaged this style of bowstroke, perhaps also in
in the same place, or perhaps close to the point, us- Beethoven’s time.
ing very little bow and starting again at the point at 44i. Vl: Most editors evidently expected a 4th-finger
12ii, 17i, or even, perhaps, returning to the point for trill on 44ii.
each of the slurred pairs. The successive use of  for 44 – 45. rh: Presumably to continue staccato. Many of
off-beats after rests was already established in the the editors mark them thus.
18th century, and probably came naturally to early- 46– 49, 166–169. Pno: Asynchrony (with rh slightly after
19th-century violinists (see note to op. 12 no. 3 bb. 17f., lh) at various points of melodic/harmonic interest
21f., 88f., 92f.). Singer, in contrast, marks everything would enhance the expressive quality of the music
, surely playing in the lower half of the bow. Some and particularly for the sf (5/c/ii; PT: 1/b).
editors do not mark any bow directions. David-re- 54 – 55i, 58–59i, 174 –175i, 178–179i. lh: Some or all of
vised omits all the repeated . the chords could be gently arpeggiated (5/c/ii). Spei-
Most editors retain the dynamics of the 1st edition, del marks these portato.
but Singer adds espressivo in 25 and includes an edi- 59–76, 179–188. Pno: In 59–64 the first lh chord un-
torial cresc. in 27, which is also added by Auer in der the slur could be arpeggiated, the second unar-
28. Halir, typically, offers a more complex dynamic peggiated, those marked sf with perhaps a swifter
scheme. arpeggiation than the others (5/c/ii). For maximum
14ii. Vl: A harmonic is marked by Alard, Hermann, expression the rh melody notes at 60i and 62i could
Singer, Halir. be played asynchronously with lh slightly after rh.
68i. Pno: Perhaps best unarpeggiated.
64, 184. Vl: AG slurs from ii, although the slur is
83 “Ein heit’rer, lieblicher Scherz, mit Humor und Zartheit vorzutra- clearly from iii in the 1st edition, and all the editors
gen. Die Bewegung stets lebhaft. Den fugirten Satz wohl markirt.
Die letzten 12 Takte des ersten Theils sehr sanft, aber auch nicht adopt this.
gedehnt.” 67i. Vl: A harmonic (marked by Alard, Singer, Auer,

Halir, Kreisler) is likely to have been an obvious tions. Alard, Joachim, Singer, Auer, Kreisler begin
choice for early 19th-century violinists.  and evidently play 125ii–127vi in the middle of
70–72, 190–192. rh: 70i could be swiftly arpeggiated the bow, either with alternate bows on the sepa-
to enhance the sf, but 70ii unarpeggiated. 71i and ii rate 16th-notes in 126 (Alard, Kreisler) or with  
and 72i could be arpeggiated (5/c/ii). (Joachim, Singer Auer); Singer specifically marks
72–75, 192–195. Pno, Vl: Beethoven’s cresc. might elicit ‘middle’ [Mitte] in 126 and from 129ii; the others all
a slight hastening to and lingering on the sf in 74, begin  and play the separate 16th-notes in 126  ;
followed by a return to tempo in 75. This would David, Halir, Rosé, Seybold are evidently towards
give the passage a more emphatic character while the point of the bow in 126, since they mark 127ii–
also increasing the contrast with the similar mate- vi slurred staccato to move back down the bow;
rial in 68–71, 188–191. Hermann, Brodsky probably remained closer to the
74 –75, 194 –195. Pno: rh i and lh ii in 74 could be swift- middle, since they leave 127 with separate bows.
ly arpeggiated to enhance the sf, while the last beat 127, 134 and 138. Pno, Vl: The slurred pairs could
of the bar is unarpeggiated. All chords in 75 could be nuanced by lengthening and playing the first
be arpeggiated to soften them. stronger than the second, which is shortened some-
76–79, 196–199. Pno: The rh broken chords overheld what (2a).
to create maximum resonance and the lh chords ar- 131f, 135f. rh: The trills starting on the upper auxil-
peggiated fairly slowly to fill out the texture. Speidel iary.
and Ganz indicate sustaining pedal through each 139f. Pno, Vl: Speidel marks espressivo on the final 8th
bar. of 139 and Singer on 140i.
77, 79, 81, 83, 197, 199, 201, 203. Pno, Vl: The notes 140–144. Vl: David, Singer, Halir, Rosé, Seybold mark
marked portato should probably be played almost  slurred staccato for the 16ths. Singer also marks
connected. At 197 Singer marks Vl dolce. espressivo in 140.
82i, 198ii. Vl: A harmonic is marked by Alard, Her- 146ii. Vl: Alard, Singer, Brodsky mark a harmonic;
mann, Singer, Seybold; David and Joachim may David, Hermann, Joachim probably considered it
have regarded it as self-evident in 3rd position. self-evident.
82f, 85f, 202f, 204f. Pno: The missing staccato marks 147vi. Vl: Singer marks espressivo.
in the first edition were supplied in AG and all the 159. Vl: Seybold leggiero
editions. Curiously they are missing in the exactly 171. Vl: Alard, Singer, Brodsky, Seybold mark a har-
equivalent places on both occurrences of this pas- monic (after 1st position in 170); other editors using
sage. 1st position in 170 (David, Hermann, Joachim) prob-
84 – 85, 204 –205. Pno, Vl: The grace-notes on the beat ably assumed it.
with the main note following swiftly after (5/a/ii). 190ii. Vl: Auer, Halir mark a harmonic
86–87, 206–207. Pno, Vl: The final three chords per- 196i. Vl: Alard, Singer, Halir mark a harmonic.
haps played long but separated as if portato. Singer 198i. Vl: See note to 82i
marks Vl in 86 and 206 portato. The Pno chords
might be gently arpeggiated. Allegro molto
87ii–90, 111ii–122, 147ii–152. Vl: A similar pattern of Tempo
 and  in the various editions occurs in these pas- Beethoven left metronome marks for two Allegro molto
sages as in 8–30. ? movements with many 8ths: he gave  = 152 for the
94f. Pno, Vl: Singer/Speidel mark slentando. fourth movement of the Second Symphony op. 36, and
98–103i. Pno: In this passage asynchrony and/or ar-  = 84 for the finale of the String Quartet op. 59 no. 3.
peggiation, especially for notes marked sf, would be For the Allegro con brio first movement of the String
an aid to expression. Quartet op. 18 no. 6, he marked  = 80. The Haslin-
115–123. Pno, Vl: The writing here might elicit a freer ger and Moscheles tempos are surely closest to what
approach to rhythm and tempo, giving the passage Beethoven would have given.
a more improvised, coquettish, or perhaps hesitant Haslinger  = 76
feeling. Moscheles-Cramer  = 160
123ii–131v. Vl: A diversity of practices for this some- Moscheles-Meyer  = 76
what awkward passage is demonstrated by the edi- Czerny-Vortrag  = 138

Alard/Diémer  = 138 In addition to his own dynamic markings, Beetho-
Speidel/Singer  = 144 ven will have expected trained musicians to add
Kreisler/Rupp  = c. 144 many more to achieve a ‘beautiful’ interpretation.
Czerny comments tersely: “To be performed just as Halir’s copious added dynamics might offer some
light and fast as the first movement.” 84 inspiration in this respect.
0ii–1f, 9ii–10f, 53ii–54f, 61ii–62f, 93ii–94f, 101ii–102f, 0–8, 53–61, 93–113, 203–222. Pno: Overheld legato
203ii–204f, 211ii–212f, 303ii–304f, 311ii–312f. Pno, would be appropriate here to create a special reso-
Vl: In AG the slur begins from the upbeat in all these nance, obviating the need for sustaining pedal (4/a/
places, except 53 and 311, although it is always from ii). Pianists of Beethoven’s era would likely have
the following note in the 1st edition (and in Vl there used asynchrony at important points, including the
is even a staccato mark on 303ii, where the figure beginnings of bars; and for the sf in 6, asynchrony
appears at a different pitch in a somewhat different would soften any potential harshness (5/c/ii and PT:
context). All the violin editors, follow the bowing 1/b).
in AG, beginning . The absence of a staccato mark 9–24, 62–73, 304 –311. Pno: At the swift tempo, Beetho-
in both Vl and Pno, on all but the final, modified ven may have expected the rh figurations in bars
appearance of the theme in Vl, certainly suggests to be played legato but without overholding which
that Beethoven did not expect a sharply detached may explain why he did not mark slurs (4/a). It is
upbeat except in that final appearance. According also possible that a non-legato articulation would
to the general theory of the period, an upbeat was have been acceptable to him (Speidel and Halir
connected smoothly to the following downbeat un- mark leggiero). He would surely have expected over-
less specifically marked to the contrary by staccato holding of the broken chords in 15, 16 and 19. From
or a rest (the vocal parallel is obvious; there are few 20–24 it is likely that Beethoven expected legato
contexts in which the words encourage a staccato without overholding (Alard marks slurs within each
upbeat). Beethoven’s slurring, if it was consciously bar, Speidel a long slur over the passage). In 13
conceived to indicate something, may have been and 309 playing the lh note before rh would help
intended to convey to the performer that although achieve a special sf. In 304 Speidel marks lh marcato.
a legato connection was expected, the accentuation Asynchrony at various points might also be intro-
should fall on the strong beat, not the upbeat; this duced to aid expression.
may also have implications for the frequent places 25– 43, 248ii–266. Pno: For the downward broken
in the movement where a slur actually begins on chords under slurs in 25–27 etc. overholding is
the upbeat. From a technical point of view, slurring apt, perhaps also with sustaining pedal as marked
the upbeat across the barline makes almost no aural by Ganz. At 25i, 29i and 33i etc. the fp might be
difference (except if a shift to 3rd position is made). enhanced with asynchrony. In 27iv etc. the notes
A violinist could, of course take the upbeat with tied across the barline in rh should receive a gen-
 at the point of the bow. (See note to op. 12/1/ii tle emphasis as was advised for such instances by
bb. 8–16.) C. Ph. E. Bach and other writers in the late 18th and
In the first four appearances of the theme Alard, early 19th centuries, which could be enhanced with
Singer, Auer, Brodsky, Hermann, Halir, Kreisler, a swift arpeggiation. The same type of emphasis
Seybold make a portamento shift from 1– 4 on the could be applied to the tied notes in 36– 41 etc. From
first two notes, returning to 1st position via an open 36– 41 Halir marks the tied notes with > (which he
A-string in the second bar. This treatment evidently had already marked in Vl on the tied notes from 27
became traditional in the second half of the 19th onwards). In 39– 41 lh the first chord under each slur
century. Anon, however, does not apply this fin- might be arpeggiated, the second not (5/c/ii).
gering, probably executing the passage in 1st posi- 43– 48. Pno: The rh slurred figures with overholding.
tion as David and Joachim apparently envisaged. It is possible that Beethoven expected the last note
As a variant, however, portamento between e1 and under each slur to be shortened somewhat; many of
c2 might perhaps be considered on a later statement the editors mark staccato (2/a). In lh the first of each
of the theme. slurred pair might be arpeggiated, the second not.
84 “Eben so leicht und schnell wie der erste Satz vorzutragen; 49–53. Pno, Vl: Speidel/Singer mark espress. in 49 (Vl)
jedoch leidenschaftlicher.” and 51 (Pno) and lento in 51 (Vn) and 53 (Pno). Spei-

del also instructs in a footnote in the piano part David, Joachim, Brodsky, Seybold provide no finger-
that the piano’s f 2 should not be played until after ing, apparently expecting 1st position with an open
the completion of the violin’s ornament.85 Whether E-string in 115. With a gut string and in the con-
Beethoven expected the small notes to be executed text of only ornamental vibrato, this is a very plau-
ad lib. and broadly, as suggested by Speidel/Singer, sible fingering for Beethoven’s time; with a metal
or to be played as a rapid ornament (like a kind of E-string, even without continuous vibrato, it will
trilled rising appoggiatura ending in a turn), is un- require more refined bow speed and pressure not
knowable. The latter, however, seems a distinct pos- to make it obtrusive. The others utilise the A-string,
sibility since the standard teaching on the execution with implications for portamento. Here too, as with
of small-note ornaments was that they should be the principal theme, a different fingering might be
played rapidly even in slow tempo, and with the considered in 145ff.
beat, not before it (therefore with the piano’s f 2). In 113–144, 276–283. Pno: In this context many of the
view of the improvisatory tendencies of early-19th- whole- and half-note chords would have been ex-
century performers, it is not inconceivable that, for pected to be arpeggiated perhaps at varying speeds
instance, the pianist might have added more notes according to context which also accords with a can-
to the ‘trill’ so that it was longer than the violinist’s tabile style suggested for the Vl (see note above).
ornament. In Pno the effect of the rh note marked sf Ganz marks espressivo in 122. Pianists would cer-
could be enhanced by playing it slightly later than tainly have applied asynchrony at many places in
the lh. the passage according to context and need (5/c/ii).
74 –93. Pno, Vl: Speidel/Singer mark poco tranquillo in 121 rh iv – 122 rh i surely legato.
74, poco allargando in 77, a tempo in 82, poco allargando 118–121. Pno, Vl: The <> in 118–121 is likely to
in 85, poco più lento in 90. Perhaps this reflects the have elicited a subtle increase of momentum to-
kind of tempo freedom that was typical of Beet- wards and lengthening at the apex (with asyn-
hoven’s time; the question is to what extent such chrony in the Pno and perhaps vibrato in Vl) and a
tempo flexibility was expected to occur. In Pno all return to tempo afterwards (3/b/v).
chords marked staccato should probably be unar- 135, 167. Vl: The fingering with a harmonic is marked
peggiated as advised by Czerny or (5/c/ii), or at most by Singer.
very tightly arpeggiated. The chords in 81, 88 and 145–190. Pno: All the triplet quarter-notes with as
92 could be arpeggiated very slowly and poetically, much overholding/over legato as possible. Some or
though at varied speeds according to context, and all of the half-note chords (including octaves) could
perhaps in a different order than the usual lowest be arpeggiated according to context. Ganz marks
to highest notes. espressivo at 154. The trill in 176 with a main-note
Vl: This is a very likely situation in which the pairs of start. The sfp in 190 could be enhanced with arpeg
staccato quarter-notes will have been played in the giation (PT: 1/a). On the down beat of 179 and 183,
middle portion of the bow, with a very short stroke asynchrony would give a special colouring to the
from the forearm that allows the bow-hair slightly dissonance (5/c/ii).
to leave the string, but with a somewhat longer 191, 193, 195. Pno, Vl: That Beethoven envisaged an
on-string stroke during the cresc. Auer marks the upper auxiliary start to the trills is almost certain
quarters with staccato dots in 74 –77, 82–85, and with in this context (5/b/ii).
tenuto lines in the other bars; Rosé only adds tenuto 198–201. Pno, Vl: The reason for Beethoven’s notation
lines in 80, 87, 91. Hermann, uniquely, marks  , per- here is unclear. Dots under slurs always indicate
haps to give a more natural lift after the first note. portato in his notation, i.e. an articulation interme-
113ii–121, 145ii–153. Vl: Singer, Halir mark cantabile, diate between staccato and legato, but it is very unu-
Auer tranquillo. Here, too, the implied tempo nu- sual in his writing for there to be rests between the
ances may reflect traditional practice. 113i–114i etc. notes under the slur. Perhaps here he wanted to
would surely have been expected to be played emphasise the larger phrase: a phrase of four notes
legato. All the editors break the long slur either af- rather than four separate, unconnected notes. Simi-
ter four bars or two bars. lar notation was more extensively used by Brahms.
223–248. Pno: Beethoven probably expected the 8th-
85 “Das f wird erst auf die halbe Note der Violine eingesetzt” note broken chords in rh and lh to be played legato

with overholding where possible, but legato without arpeggiated which would help to achieve the req-
overholding or non-legato is also a possibility. Spei- uisite softness (5/c/ii; PT: 1/a) Ganz marks dolce and
del marks non legato at 223 and 231 but also gives gives sustaining pedal indications through each bar
sustaining pedal indications through each bar. Ganz until 302.
marks non legato and sustaining pedal indications 292ii–iii, 294ii–iii. Vl: Singer employs a fingering typ-
from 231. Rosé marks non ligato from 223. From 224 – ical for him, sliding the 2nd finger to the harmonic
230 the quarter-note chords in lh could be arpeg a2 in 292 and the 4th finger to the harmonic in 294.
giated swiftly and with gruff accentuation. Speidel 303. Vl: The use of the open E-string here would prob-
marks these staccato with arpeggio signs and > . ably have been almost universal in Beethoven’s time.
Presuming that Beethoven expected all the chords It is marked by all the editors except Rosé.
in 231–246 rh to be staccato, these should be unar- 306. Vl: Alard, Hermann, Singer use an extended 4th-
peggiated (5/c/ii) or at most very tightly broken finger harmonic for the e3. Anon and most of the
(perhaps best reserved for the chords marked sf ), editors shift the 2nd finger from b2 to c3. Both fin-
which would give them the requisite fire. The dra- gerings are plausible for the early 19th century.
matic chord marked sf in 247 could be played in 311–323i. Pno: Beethoven will surely have expected
various ways: a fast arpeggiation from lowest note the rh octave at the end of 311 to be connected to the
to highest note; playing the lowest note on the beat following octave as marked by Speidel, Rosé, and
with the other notes in lh and rh very slightly after- Halir, although Diémer marks 311 rh ii staccato (see
wards and unarpeggiated; arpeggiating the chord Beethoven’s slur in lh from 16–17). To achieve legato
upwards and downwards several times to sustain through the octaves a combination of legato finger-
the sound through the chord’s length; a combina- ing (alternating 5, 4 in the upper notes as marked
tion of all of these (5/c/ii; PT: 1/a and PT: 1/b). by a few editors) and sustaining pedal as marked
231–246. Vl: Most violinist of Beethoven’s time would by Ganz will be helpful. The octaves marked sf in
probably have remained in 1st position and used 316, 317, 320 and 321 could be enhanced with swift
separate bows throughout, as Anon apparently did. arpeggiation or asynchrony (5/c/ii).
Few editors provide fingering, presumably assum- 312–323i. Vl: 1st position is expected by all editors ex-
ing execution in 1st position; Hermann simply marks cept Halir, who shifts to 3rd position on 119vi and
0 on 231iii. Alard, Singer, however, suggest 3rd posi- back to 1st on 321viii. As in 231ff, Auer adds some
tion in 237, 241, 245, returning to 1st in the follow- slurs.
ing bars. Not until Kreisler does an editor suggest 323–332. Pno: The slurred rh figures (and lh figures in
2nd position in 237, 241, 245. Auer alone adds oc- 329 and 330) with overholding. From 324 –329 the lh
casional slurs, which some violinists of Beethoven’s chords might all be swiftly arpeggiated.
time may have employed in a passage of this kind, 327f. Vl: All editors except Alard, Rosé mark an open
although Beethoven’s staccato marks in 231 might E-string; all except Rosé, Seybold explicitly mark,
suggest that he envisaged separate bows. or obviously assume a harmonic e3 with 4th finger
257–266. Vl: One of Anon’s only two markings in this in both bars.
movement is fingering here; the ascent in 258f is 329iv. Vl: All except Alard, who remains in 3rd po-
accomplished by a shift from 4 on b 2 to 1 on c #3 sition until the end of 330, descend via the open
followed by a 4th-finger extension for g3; the 4th- E-string.
finger is also used for f 3 and the rest of the descent
executed by 3 on c3 and 2 on c2. Most of the editors
move to 3rd position with 2–2 from 257 to 258i, but
SONATA OPUS 24
Singer, Brodsky begin already in 3rd position, and
Kreisler, Seybold begin 1–2. This sonata may serve as a case study for some issues
268–275. Pno, Vl: Speidel/Singer again add poco tran- that occur throughout Beethoven’s works, at least un-
quillo in 268 followed by allargando in 272 and a tem- til his last decade.
po after the fermata. The chord in Pno 274 might be In bars 1–24 and 124 –148, Beethoven provided only
given a fairly slow arpeggiation (5/c/ii). a general piano dynamic with two short crescendos re-
284 –303. Pno: The rh figures with overholding. The turning to p and a final cresc. to f. There is no rea-
lh octaves from 284 –294 could be swiftly but gently son to imagine that he expected a constant dynamic

level where nothing else was marked: a fine early- Beethoven’s Quintet op. 16 in 1816). Although many of
19th-century performer was expected to understand Czerny’s embellishments in his arrangements of op. 47
how to nuance the volume within a general dynamic conflict with his assertion in 1846 that, in performing
level; the composer marked only the major effects that Beethoven’s works, “the player ought absolutely not
might not necessarily be obvious from the shape of to permit himself any change to the composition, any
the melodic line. Most later editors have been content addition, any abbreviation”,86 this contrast between
to leave such refinements, which are nevertheless es- earlier practice and later opinion undoubtedly reflects
sential for a beautiful performance, to the executants. the “changed taste of the times” to which Czerny also
The editions by Speidel/Singer, Halir, Rosé, and to a referred.87
lesser extent Ganz/Auer, however, provided sugges- Improvised ornamentation was still practised in the
tions, evidently with less experienced performers in 1840s, but was increasingly regarded as inappropri-
mind, for shaping the melodic line dynamically. Ex- ate. This is nicely illustrated by Berlioz’s complaints
perienced musicians will have been expected to apply about the otherwise excellent first oboist of the Dres-
more subtle, or indeed contrasting nuances, ideally den Court Orchestra in 1842, who, in performing the
treating the melody differently on the repeat of the ‘Scène aux champs’ in the Symphonie Fantastique, re-
exposition from the first time. For the violinist, this vealed “an old style and a mania for making trills and
might have included changing the fingering to use mordents”.88
the tone colour of different strings, varied vibrato or In op. 24, a sensitive early-19th-century violinist might
portamento (see, for instance, Singer’s treatment of the perhaps have made the following types of embellish-
opening bar, referred to in the note to 1i–ii, 861); for ment during the repetition of the exposition of the
the pianist, in addition to tone colour, a varied use of first movement:
arpeggiation and asynchrony, or for either performer,
varied dynamics and a different modification of the b. 3
notated rhythms. j œ œ
Whether Beethoven might have envisaged, or in- œ œ œ œ œ œj œ œ
deed encouraged additions or changes to the nota-
tion on the repetition of a section or melody, remains bb. 3–6
mj Ω œœ
speculative; but in the aesthetics of the time in which œ.
œœ ˙ œœ œ Œ œ. œ œ œ œ œœœ ˙ œ
œœœ
Beethoven grew up, skilful, proportionate, and appro-
priate variation of a melody on its subsequent appear-
ances was certainly regarded as a sign of superior mu- bb. 7–8
œ œ œj ˙ mœ œ œ
sicianship. Beethoven is known to have approved of œ. œ œ# œ œ œ œ œ œ œ
Bridgetower’s improvised embellishment of the first
fermata in the Presto of the first movement of op. 47 or
(see the Commentary on that sonata). Furthermore, œœ œœœ ˙
despite Beethoven’s admonishing of his pupil Carl œ. œ œ# œ œ
Czerny for inappropriate embellishment in a perform-
ance of his Piano and Wind Quintet op. 16, Czerny b. 29
added a substantial number of ornamental additions
œ # œ œ œ œ œ nœΩ
Ÿ
and changes in his 1823 solo piano arrangement of b ˙
or with an accenting ornament
the second movement of Beethoven’s ‘Kreutzer’ So-
on the first note instead of, or as well as the trill.
nata op. 47, and his 1825 piano duet arrangements of
the whole sonata. These offer valuable insights into
the kinds of embellishment that were evidently re-
garded as acceptable by a musician in Beethoven’s
close circle, whose musicianship he is known to have 86 Czerny, Die Kunst des Vortrags, p. 34. “darf der Spieler sich
valued (see the Commentary on op. 47 below for de- durchaus keine Änderung der Composition[,] keinen Zusatz, keine
tails of Czerny’s added ornaments, and “Reading Abkürzung erlauben.”
87 Ibid., “durch den veränderten Zeitgeschmack”.
between the lines” (5/c/i) for Beethoven’s criticism of 88 Hector Berlioz, Mémoires, vol. 2 (Paris, 1870), p. 276. “un vieux
Czerny’s excessive embellishment to the piano part of style, et une manie de faire des trilles et des mordants”.

Op. 24 ed. Wilhelm Speidel and Edmund Singer

Op. 24 ed. Halir

b. 157 b. 8
œ j
& œœ. nœ œ œbœœ œ œœ ‰ Œ
œ bœ œ œ
b ‰ œ œ œ œ bœ nœ œ
J [ º]
 j
cresc. f
œ
? œ˙ œ œ nœ œ ‰ Œ
Or on repetitions of the principal theme of the Rondo: J
˙ œ mœ # œ œ m
b. 65 or b. 133 or b. 125
T
˙ œ œ# œ œ
#œ œ œ ˙
&
œ œ œ œ
b. 69 or b. 137
œ œ n œ œ œ œ œ œ œ œΩ œΩ
Kr
 ? œœœœ
œœ œœ
bb. 70–73, bb. 138–141, or b. 205

bb. 126–127
œΩ œΩ œ œ
œ œΩ œΩ œ œ œ# œ œ œ œ œ mœ œ œΩ œΩ
r
#œ œ œ ˙ œ œ# œ œ
œœœ œœ œ œ œ œ œ# œœ œ œ œ
b &

6
5
? œœœ œ
œ œœ œœœœœœ œ
Piano variants in the principal theme of the first œ œ
movement might include:
bb. 13–14
Allegro
œœ
& œ. œj œ œ œ œ. œ œ œ Œ Tempo
 ? w
w w
w
Beethoven assigned metronome marks to several Al-
legro movements in  metre. The fastest at  = 88 is
op. 59/1/i, in which the smallest note values are tri-
plet 8ths, the slowest at  = 80 is the storm movement
b. 16 of the Sixth Symphony, which has some notes short-
er than 16ths; the finale of the Fifth Symphony and
œ œœœœœ
& Œ the first movement of the String Quartet op. 74 are
 ? w
w
both marked  = 84; the latter has a similar compo-
nent of slurred 16ths to this movement and it seems
very likely that Beethoven would have allotted it a
similar number. There is no evidence that Beethoven
b. 18 ever changed his mind about the tempo of this move-
œ. œ œ œj bœ œ œjœ ment, but with the sonata’s continuing popularity, the
œ
& first movement’s lyricism seems to have been increas-
 ? ˙
˙
˙
˙
ingly cultivated at the expense of its drive. This may
already have been happening by the time it was re-
issued in the Haslinger collected edition; even  = 76
seems anomalous for a  Allegro.
Or in the Rondo: Haslinger  = 76
bb. 5–7 Haslinger 2  = 66
Moscheles-Cramer  = 144
œ œ œ œ œ œ mœ œ œ œ bœ œ
& œ œ# œ œ œ œ œ œ œ œ œ œ nœ œ œ œ Czerny Vortrag  = 132
 œ œ œ œ œ œ  = 112
Alard/Diémer
? œ œ œ œ œ œ bœ œ œ œ œ œ œœ œ œœ œ œœ œ Speidel/Singer  = 126–138
Kreisler/Rupp  = 108–112

Speidel/Singer add (non tanto) after Allegro; this and 25, 29, 31, 149. Pno: The staccato chords unarpeggiated.
the last movement of the same sonata are the only 26–27, 150–151. Pno: Slight asynchrony at 26i etc.
movements in their edition for which they do not give could be effective in enhancing a sense of drama.
a single number, perhaps reflecting a degree of am- The whole passage with overholding (4/a/ii).
bivalence that was still associated with the tempo of 30, 154, 156. Pno: The chord on lh i could be swiftly
those movements in the later 19th century. arpeggiated (5/c/ii). The rh with overholding and
Czerny writes: “One of Beethoven’s loveliest and perhaps judicious use of sustaining pedal as indi-
most melodious sonatas, which draws upon the beau- cated by Ganz.
tiful performance style of both players in every sense. 29, 31. Vl: Beethoven surely expected some kind of
The tempo is a calm Allegro, which, however, here expressive nuance in these bars. Singer adds <>
and there (e.g. bars 26 and 27) does not exclude a more on i, possibly encouraging vibrato and/or some de-
animated tempo. Similarly, bars 38 and 39 are a bit gree of lingering. Halir and Rosé add > through
livelier and should also be pedalled. The conclusion of the whole bar.
the first part, as well as all the brilliant passages, with 33i, 157i. Pno: Asynchrony could give much expres-
fire.” 89 The expression “der schöne Vortrag”, literally sive significance to rh i.
“the beautiful performance style”, has specific impli- 34f, 158f. Pno: Why Beethoven did not mark slurs over
cations that are not immediately obvious to modern the descending chromatic scales is unknowable. Per-
readers. See “Reading between the lines”. haps he felt that a proper legato with overlapping
1i–ii, 861. Vl: Alard marks a harmonic on i, with 4th from one note to the next (4/a/ii) was difficult or
finger again on ii in both places. Singer begins in impossible and assumed pianists would in any case
1st position, but in the 1st-time bar gives the same connect the notes as well as possible (see Critical
fingering as Alard. Report for the apparent deletion of slurs in the 1st
1–9, 134 –136. Pno: Overholding of the broken chords edition). Reinecke, Diémer and Halir mark slurs. At
for resonance is essential (4/a/ii), perhaps aided by 34i, 35i, 36i and iii and 37i etc., asynchrony would
judicious use of sustaining pedal. In 1–6 Beethoven certainly be appropriate to increase dramatic effect.
probably intended legato in lh: Halir marks slurs; 35. Vl: Most mark or evidently envisage , but  is
Speidel marks each note with staccato and tenuto marked by David (removed in David-revised), Ha-
lines possibly indicating something like portato. lir, Rosé. This is a very effective bowstroke if deliv-
3v–vi. Vl: Most editors give the fingering 4 –3, which ered firmly (almost, but not quite fouetté).
would probably have been the fingering of choice 38f, 42f, 54f, 56f, 90f, 94f, 162f, 166f, 178f, 182f. Pno,
in Beethoven’s time. Vl: The predominant convention in 18th- and 19th-
5f. Vl: David, Hermann, Joachim and others remain century treatises supports the notion that, except
in 1st position, evidently with an open E-string on in a few specific circumstances, grace-notes should
6ii (specified by Brodsky); others shift to 5v, some be conceived as occurring on the beat (5/a/iv). For
up the A-string. pianists this would surely have meant aligning the
11–24, 124 –132, 144 –148. Pno: Various of the lh chords first grace-note with the lower octave in rh in 54f,
in 11–19 might be arpeggiated (5/c/ii; PT: 1/a). Asyn- 58f etc. Later in the 19th century this was changing.
chrony could be applied to important melody notes Pre-beat performance was certainly envisaged by
especially but not exclusively at the beginnings of Speidel, who gives a footnote about the grace-notes
bars or on main beats (PT: 1/b). Speidel’s and Ha- in b. 54: “grace-notes [Nachschlag] to be played be-
lir’s various tenuto and accent marks might provide fore the 3rd quarter-note.” 90 The term “Nachschlag”
guidance as to places suitable for the application of normally refers to the two-note ornament at the end
asynchrony. of a trill, confusingly referred to as a turn in Eng-
lish. The ornament in question would be better de-
scribed as a “Schleifer” (slide). No comment of this
89 “Eine der lieblichsten und melodiereichsten Sonaten Beetho-
vens, die den schönen Vortrag beider Spieler in jedem Sinne in An- kind occurs in the violin part at 38, however, per-
spruch nimmt. Das Tempo ist ein ruhiges Allegro, welches jedoch haps because by the 1880s Singer already consid-
hie und da, (z. B. Takt 26 und 27) eine belebtere Bewegung nicht ered that it would be obvious to violinists. Pre-beat
ausschliesst. Eben so sind die Takte 38 und 39 etwas lebhafter und
auch mit Pedal zu nehmen. Der Schluss des ersten Theils, so wie
alle brillanten Passagen mit Feuer.” 90 “Nachschlag, vor dem 3. Viertel zu spielen.”

performance in this passage became standard for Joachim, Auer, Kreisler mark alternating bows.
20th-century pianists and string players. Either approach would be characteristic of the early
Beethoven’s sf here may suggest that the attack on 19th century. In view of Beethoven’s notation in 54f
the first of the small notes is accented, but the con- etc. it seems clear that the staccato quarter-notes here
cept that the accent fell primarily on the main note were not intended to be particularly short. Singer’s
was widespread, and the faster the grace-notes are notation, however, is curious: he gives not only a
performed, the more the accent is perceived to oc- slur over the staccato quarter-notes, but also marks
cur on the main note. each with  above the slur, perhaps intending to in-
38– 45, 162–169. Pno: Some of the chords (perhaps just dicate a lifted bowstroke.
the metrically stronger ones) could be played with 46, 48, 50, 63, 65, 67, 170ff. Vl: Six editions mark all or
very swift arpeggiation which would maximise dra- some of the repeated 8th-notes as up-bow slurred
matic effect (5/c/ii). Sustaining pedal, for example staccato, several editors begin it from ii (David-
indicated by Speidel (throughout) and Ganz (only revised, Hermann, Rosé, Seybold), others from iv
in 39– 40 but presumably to be applied in parallel (David, Halir); these editors surely envisaged a clas-
situations) might also be used for extra resonance sic firm staccato starting near the point of the bow
to enhance this special rocketing effect in the mu- and ending before the middle. Among the ones who
sic. The sf at 40i and fps at 42i and 44i etc. might be take only some of the 8th-notes with successive up-
enhanced with slight asynchrony. bows, Alard has a unique approach, evidently begin-
38–63, 162–187. Pno, Vl: Beethoven’s dynamic mark- ning the 16ths , he takes the last two 8th-notes  to
ings in this passage are apparently incomplete. It is arrive  for the sf; this, combined with a  from 51i–
exceptional for him to indicate crescendo for one in- 52ii may suggest that he bowed the repeated notes
strument while another remains in piano throughout, in the middle or even lower half of the bow. Auer
or for one instrument to remain piano over several having executed the quarter-notes in 44f with sepa-
bars while another is forte. Some 19th-century edi- rate bowstrokes, probably in the upper half, takes
tors added dynamics; others retained this passage 46ii–v , vi  and vii–ix again ; in conjunction with
as it appears in the sources, leaving it to the players his dynamics, this suggests that he moved towards
to find a satisfactory solution. In Joachim’s edition, the lower half of the bow for the sf in the following
the piano score remains unaltered, but f is marked bar. The editors who play all the repeated 8th-notes
in the separate violin part at 63 (which is a logi- in 46 etc. with separate bows probably expected to
cal presumption from the cresc. in 61), though not at execute them in the upper-middle of the bow, or
187; the piano part, however, remained theoretically perhaps, as Singer’s dynamics suggest (see below)
within piano. Auer, Rosé, Kreisler also mark this f. increasing the bow length with each note and mov-
Halir and Speidel/Singer suggest more detailed but ing progressively towards the frog.
differing dynamic contours: the former envisaged In 51–53, the five editors who used slurred staccato
a scheme in which the crescendos, fortes, and pianos on all the last five 8ths in 46 etc., and Brodsky, make
are matched in both parts, the latter retained a ba- two bow changes, enabling them to arrive on , to-
sic piano dynamic throughout, with many < sf > wards the point for 54i; after another  on ii, they
markings. Either approach may plausibly represent surely envisaged crisp martelé strokes. David (but not
Beethoven’s expectations. When performers were David-revised) also marks 55ii–iv etc. in up-bow, pre-
playing from separate parts, without knowing the sumably in connection with the crescendo but prob-
dynamic markings in the other, however, the for- ably keeping the bow close to the point; this con-
mer scheme is more likely to have arisen through a trasts with Alard, Joachim, Kreisler, whose bowing
musically-motivated response to the other part. The in 51–53 takes them towards the frog, and with Brod-
present edition leaves it to performers to find their sky, who marks  successively from 54ii–55iv. With
own solution, which certainly need not be identi- the gradual marginalisation of martelé and firm stac-
cal on each appearance of this material, or in each cato in the 20th century, the latter practices have be-
performance. come predominant in performance of this passage.
40iii– 42i ,44iii– 46i, 164iii–166i, 168iii–170i. Vl: For To what extent any of these approaches would have
the quarter-notes in 44ii– 45iv, as in 40f, the major- been typical of the early 19th century is difficult to
ity of editors indicate slurred staccato; only Alard, determine. The use of firm slurred staccato for the

8ths, however, reflects practices going back at least but Beethoven’s deletion of slurs in the autograph
to Leopold Mozart. (see Critical Report) suggests that he wanted to dis-
46–54, 62–68, 170–177. Pno: The lh octaves, particular- courage overholding. Diémer marks both bars with
ly whole and half-notes, might be swiftly but gently slurs, presumably assuming the use of sustaining
arpeggiated which would fill out texture and create pedal at least from 68i–ii. Speidel marks portato from
drama (5/c/ii, PT: 1/a). At 48i and 50i etc., slight asyn- 68ii etc. At 68/192i, asynchrony would help achieve
chrony would enhance the sf effect (PT: 1/b). the sudden p (PT: 1/b).
51, 67, 175. Pno, Vl: Beethoven’s rinf. is probably not 70–85, 194 –209. Pno, Vl: Beethoven marked particular
an immediate accent, but rather an instruction for syncopated chords in lh (75ii and 76ii etc.) sf, but this
a short but powerful crescendo.91 Speidel/Singer ac- would not have precluded other syncopated notes
knowledge this by adding < after rinf. in both in both lh and rh from being emphasised, as rec-
parts in 51; in 67 and 191 in Vl only; in 175 in Pno ommended in late-18th- and early-19th-century trea-
only. In 51 Beethoven originally wrote cres. but tises. Speidel marks all syncopated chords (includ-
changed it in his autograph to the more powerful ing those with sf ) with staccato and tenuto lines.
rinf., which he used in subsequent occurrences of Beethoven did not mark lh 74/198ii sf (presumably
this idea. in connection with the p in Vl), but several of the
b. 51f: editors add sf in both places. Arpeggiation of chords
not marked staccato, including half-note octaves and
chords in both hands could be effective (5/c/ii; PT:
1/a). The use of sustaining pedal from 70–76 as indi-
cated by Speidel and Ganz might also enhance tex-
ture. The 16th-note scales in both Pno and Vl will
surely have been expected to be played with dynam-
ic shaping, normally ascending with crescendo and
53, 69. Pno, Vl: The grace-note turn should probably descending with decrescendo as marked by Halir
be left as late as possible and connected to the main and Ganz. Beethoven’s <> in 84 –85/208–209
note (5/b/i). invites a hastening of pace to the apex, and agogic
54 – 61, 90–97, 178–185. Pno: For the performance of accent (perhaps with vibrato on 85/209i in Vl and
the grace-notes see above (38f). In 56, 60 etc., a slight asynchrony or arpeggiation in Pno), with a return
asynchrony at sf would be dramatically effective to tempo afterwards.
(PT: 1/b). Judicious use of the sustaining pedal as in- 79, 81, 83. Vl: The trills should certainly begin from
dicated by Ganz would also aid in creating drama, the upper auxiliary in this context.
particularly for sf. There are several approaches to bowing. Alard, Da-
62–70, 186–191. Pno: All the longer-value rh octaves vid, Kreisler start , take  for the 16th, and then
not marked staccato might be swiftly but gently ar- continue  towards the point; Hermann, Joachim,
peggiated. At 63i, 65i and 67i etc., asynchrony would Auer hook the 16th into the preceding long note;
enhance the dissonance. David-revised, Halir, Rosé, Seybold hook in 79, but
63, 65, 67. Vl: The editors use the same bowing as in not in the other two bars; Singer, Brodsky use the
46ff. hooked bowing in 79, 81 but continue the slur that
68f. Vl: All the editors who employ slurred staccato begins on the trill to the end of the following bar.
in 63–67, probably having executed it in the upper 84f. Vl: David, Alard, Singer, Joachim retain Beetho-
half of the bow, change bow on 69ii or 68iii (Her- ven’s two-bar slur; Hermann and all later editors
mann). Alard, Joachim, Kreisler, obviously closer to (including David-revised) take a new bow in 85.
the frog, retain Beethoven’s slurring. 862f. Vl: Singer takes this in 4th position with a har-
68–69, 192–193. Pno: It is uncertain what Beethoven monic on 87i. Many begin in 4th or 5th and descend
expected in terms of articulation in lh. Legato from to 3rd on the second c #3.
68/192 lh ii (with or without overholding) is possible, 862, 210. Pno: A swift arpeggiation of the chord is apt
here to emphasise the drama created by the sudden
91 See Clive Brown: Classical and Romantic Performing Practice (Ox- modulation (5/c/ii; PT 1/a).
ford, 1999), pp. 87–92. 89, 213, 215. Pno: Asynchrony would be an histori-

cally appropriate, expressive practice for the notes replace staccato marks with tenuto lines (3/c/vi). In
marked portato (4/b/i). fact, with the crescendo, the separate notes should
90–96. Vl: All editors execute this passage like 54ff. probably be played with increasing length, either
98–115. Pno: The nature of Beethoven’s writing here with forward drive, which is typical for crescendo,
might elicit a fiery performance style perhaps with a or perhaps holding back the tempo. Beethoven’s pu-
slightly faster tempo. Speidel marks con fuoco. Over- pil Ferdinand Ries noted that occasionally Beetho-
holding of notes under slurs was expected (4/a/ii) ven “held the tempo back in his crescendo with
for resonance, which could also be enhanced with ritardando, which made a very beautiful and ex-
sustaining pedal as suggested by Ganz. The first tremely striking effect.” 92
notes under each slur, especially those marked sf, 155iv–v. Vl: Performance on the A-string with a por-
could be energised without causing harshness by tamento fingering is indicated by Hermann, Auer.
playing them asynchronously (5/c/ii; PT: 1/b). 157. Vl: All remain in 1st position except Hermann,
98ff. Vl: Beethoven’s clear slur from ii–iv in A, as also Singer, Rosé, Kreisler, who stay on the A-string
in E, may indicate that, thinking as a string player, with inevitable portamento between iii and iv.
he considered a   bowing for the following triplets, 168iii–183. Vl: See note to 44iii–60.
which allows the off-beat sfs to be executed effec- 187ff. Vl: See note to 63, 65, 67.
tively . Such ‘reverse’ bowings were particularly 192f. Vl: See note to 68f.
characteristic of players influenced by contempo- 216, 217. Pno: To enhance the sf, asynchrony might be
rary French practice, with which Beethoven was cer- applied (5/c/ii; PT: 1/b).
tainly familiar. AG, however, ignored this slurring 219–227. Pno, Vl: For successive equal-value notes in
(which was also ignored in BW) and none of the a legato context, an unequal style of performance
19th-century editors marked it; some used a retaken was expected (2/b). In Pno, occasional asynchrony
 for the sf, others took it  and tucked in another . would heighten expressivity, perhaps at the begin-
In a masterclass I (CB) gave in Bydgoszcz (Poland) ning of each of Beethoven’s slurs and particularly
in 2018, however, a student, with no knowledge for all syncopated notes in the last three bars.
of the sources, executed the reverse bowing very 222–226i. Vl: Most editors remain in 1st position.
effectively in this passage. Alard, Singer, however, mark the whole passage in
100–101 etc. Pno: Despite the lack of slurs, it is probable 3rd position, Alard specifying a harmonic on 224ii
that Beethoven expected the broken chords to be and Singer (restez in 222) surely assuming one.
played smoothly and with some overholding where 228–230. Pno: Arpeggiation of the chords in both
possible. In the editions, perhaps surprisingly, Rei- hands will have a softening effect and help achieve
necke alone adds slurs; but Speidel and Ganz pro- decrescendo (5/c/ii; PT: 1/a).
vide sustaining pedal indications. 231. Pno: The double trill will gain much brilliance
112–115. lh: The chords marked sf could be swiftly with the use of the sustaining pedal.
arpeggiated, with sustaining pedal as indicated by 232–239. Pno: The slurred triplets in rh with overhold-
Speidel and Ganz (5/c/ii; PT: 1/a). ing (4/a/ii). From 232–237 asynchrony would help
117–119. rh: Some or all of the double-note chords might distinguish the lh thematic material from the rh
be swiftly but gently arpeggiated to enhance their accompaniment (PT: 1/a). At 237, perhaps a marked
softness. asynchrony for the sf (PT: 1/b). At 232 the octave at
134ff. Vl: Only Singer, Halir suggest any change in lh i could be swiftly arpeggiated (5/c/ii; PT: 1/a).
the treatment of the melody. Singer adds molto es- 238f. Vl: David, Joachim, Halir, Rosé, Kreisler, Sey-
pressivo at 136iii, suggesting D-string ad lib., and in bold, marking  from 238iii–239v, indicate perform-
138 adds sempre cresc. ance of the following triplet passage near the point
137–143. Pno: The expressive writing here invites asyn- of the bow; Alard’s  from 239iv–v indicates per-
chrony at downbeats (especially those marked sf ) formance of the triplets in the middle or lower half;
and at half bars in 137, 139 and 142–143 (PT: 1/b). In others (including Joachim-revised), changing to 
137 Speidel marks molto espressivo.
œ. œ. -œ 142f.
-œ -œ Vl:
-œ These -œ. -œ. œnotes
-œ. -œ. staccato œ œ(marked Ω Ω -œΩ inŒ the
œ œΩ œΩ with
œœ
92 Franz Gerhard Wegeler and Ferdinand Ries: Biographische No-
edited violin part of the present edition, should cer Ó tizen über Ludwig van Beethoven (Coblenz, 1838), p. 106. “[Mitunter]
hielt er in seinem crescendo mit ritardando das Tempo zurück, wel-
tainly not be played too short. Singer, Auer, Rosé ches einen sehr schönen und höchst auffallenden Effekt machte.”
∑ ∑ ∑ ∑ ∑ ∑
∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑
on 239ii may have executed the triplets more to- chord. This clearly applies to pianos of the 1840s
wards the middle of the bow, although they too and later (Speidel marks sustaining pedal through
could have played them near the point. each bar). On the type of piano that Beethoven had
240–244. Pno: From 241–244, asynchrony at the begin- at his disposal at this period (PT: 2), it is likely that
ning of each bar will delineate the separate voices he used some form of overholding of notes in the
(PT: 1/b). The chord at 240 rh i could be arpeggiated arpeggiated lh broken chords (4/a/ii) to create reso-
swiftly and gently to effect p. nance in addition to or instead of sustaining pedal.
This would of course extend to the rh in 9–16, 37ff.
Adagio molto espressivo An expressive rendering of the accompaniment fig-
Tempo ure would certainly have required a slightly une-
See above (op. 12/3/ii) for Beethoven’s markings in 3/4 qual style, using agogic accentuation (lingering) on
Adagios, which range from  = 72 to  = 84. In this important notes or beats at the expense of others.
movement there are many fewer melodic passages of From 9–16, when the accompaniment is in unison
32nd-notes than in those movements, which suggests octaves, this agogic accentuation could be aided by
a more flowing tempo. Beethoven’s use of two-bar applying asynchrony to important beats. In 17 the
slurs in the violin part in 43f., 46f., 52f. may indicate last four 16ths could be made very expressive by
that he conceived a tempo closer to those of the 1828 lingering on the first and perhaps even the third.
Haslinger edition and Moscheles, perhaps even a little Playing the notes equally would have been consid-
faster, since even at those tempos such long slurs are ered ‘correct’, perhaps sufficient for learners, but in-
impracticable. sufficient for an artistically sophisticated and ‘beau-
Haslinger  = 92 tiful’ performance. From 10–16 where Vl takes over
Moscheles-Cramer  = 96 the cantilena, pianists might consider using the una
Czerny-Vortrag  = 84 corda shift pedal or perhaps even the moderator if
Czerny-Simrock  = 88 available (Speidel marks pp) as a special colour (PT:
Alard/Diémer  = 72 3/a and PT: 3/b). In 37, the sudden shift to B flat mi-
Speidel/Singer  = 44 nor might inspire the use of a double moderator if
Kreisler/Rupp  = 72–76 available and certainly a special tonal colouring un-
Czerny makes no direct comment about the speed of til the crescendo at 46. 49 might start with the una
the movement, writing: “Everywhere, where the ac- corda shift, with change at 50 to the due corde, and
companying figure [ex. bb. 1–5] comes in the bass, 51 tre corde, again if available (PT: 3/a).
or in both hands, the pedal is to be taken for each 2–9, 26, 30ff, 54ff. Pno: Pianists of Beethoven’s era
change of chord. The following passage (from the would certainly have used asynchrony to height-
30th bar) [ex. bb. 30–34] must be extremely delicate, en the expression of the cantilena (which Speidel
the decoration light and even, and everything will be marks cantabile), delaying (or less frequently an-
performed in tempo. A heavenly tranquillity predom- ticipating) melody notes with varying time lapses
inates in this Adagio, which must be characterized by to give emphasis, colour or enhance texture (5/c/ii;
the tenderest expression and harmonious effect.” 93 PT: 1/b). In 8 the rh notes marked portato would
Speidel/Singer and Halir add dynamic nuances and almost certainly have required asynchrony (4/b/i)
articulation marks that, while reflecting their own In 2, 4 and 5 etc., the rh cantilena may encourage
practice, might provide inspiration for making subtle rhythmic flexibility. For example, the 16th-notes
dynamic nuances not marked by Beethoven but un- might be played in a lilting fashion (long/short) but
doubtedly expected. varied. Beethoven notated this at the end of 7, but
1–17, 29ff., 58ff. Pno: Czerny states that the sustain- there is no reason to believe that he would have had
ing pedal is to be used and changed at each new any issue with pianists of his day making similar
rhythmic nuances in places where he did not notate
93 “Überall, wo im Bass, oder in beiden Händen, die hier beglei-
tende Figur vorkommt [Ex. bb. 1–5], ist für jeden Accordwechsel them, indeed he would surely have expected it (2/c).
das Pedal zu nehmen. Folgende Stelle (vom 30sten Takt an) [Ex. 3i–ii. Vl: Alard, Hermann, Singer shift 2–3; the others
bb. 30–34] muss äusserst delikat, die Verzierung leicht und gleich, remain in position.
und alles im Tempo vortragen werden. Eine heilige Ruhe herrscht
in diesem Adagio, die durch den zartesten Anschlag und durch 4, 12, 40. Pno, Vl: The trills were surely expected to
harmonischen Effekt charakterisiert werden muss.” begin from the upper auxiliary because they are

preceded by a note at the same pitch (5/b/ii). They 10. Vl: Hermann adds dolce, Singer cantabile, and Halir
were undoubtedly expected to conclude with a mp.
turn. This is marked, however, only by Reinecke/ 14. Vl: To shape the 64th-note fioritura expressively,
Hermann, Singer (Speidel does not mark one in 4, the violinist might begin it early in order to take
perhaps through oversight). more time. Freedom of this kind was typical of Jo
7f, 15f. Pno, Vl: In the autograph, in 8, Beethoven ini- achim’s tempo rubato, and would surely have been
tially wrote the same rhythms for the ornamented characteristic of many performers in Beethoven’s
figure, but with a turn sign over the dot instead of time (see also the note to 30–36 below).
the small notes in 7; he then deleted the figure in 8 15–17. Vl: All remain on the D-string.
and rewrote it in its present form. 17–27, 54ff, 63ff, 70ff. Pno: Beethoven’s pp may suggest
use of the una corda pedal or moderator. In Beet-
hoven’s era pianists would almost certainly have
arpeggiated all chords, in this kind of context, ex-
When he wrote down the repetition of this mate- pressively (whether indicated or not, unless very
rial in the violin part in 15f, he made an identical short in value or marked staccato), generally with a
differentiation in the notation of the ornaments. Ac- fast spread, reserving slower spreads for more ex-
cording to C. Ph. E. Bach the two different notations pressive moments or for the creation of more lavish
would stand for essentially the same execution,94 texture. Beethoven’s arpeggio sign in 23 might in-
dicate a slower, more poignant spread (which Spei-
del marks tenuto). In 18 and 20 the > might elicit a
(2) #
### .T œ ≥ ≥
œ. œ œ œ. œ œ# œ œ. œ œ. œ œ
poignant (moderately slow) arpeggiation of the lh
œ
with the rh chord unarpeggiated and played signif-
but despite Beethoven’s respect for C. Ph. E. Bach’s icantly later. While for the sf in 19 and 21, the lowest
authority, his repetition of the same notational dif- note in lh might be played on the beat with the other
ference in 15f. suggests that the distinction was notes in lh and rh played shortly afterwards and
intentional. The notation in 7, 15, and his initial unarpeggiated (PT: 1/a). This would certainly dis-
notation in 8, accords closely with the so-called tinguish between the two types of accent and cre-
Haydn ornament, and the initial notation with the ate individual agogic effects. In 21 the slurred pairs
turn sign in 8 suggests an analogy with Haydn’s in rh could be rendered highly expressive with the
notation at the beginning of the Andante of his first under each slur lengthened and strengthened,
String Quartet op. 77 no. 2, where he first wrote while the others are correspondingly shortened and
the ornament figure with three small notes then softened (2/a). The dissonances at 22i (Speidel marks
replaced these by a turn sign. By his notation in 7, tenuto on rhi) and 27i would benefit expressively
therefore, Beethoven may have envisaged the kind from either arpeggiation or asynchrony; variation
of rapid turn on the sixth 8th-note beat of the bar is the key to successful delivery of these rhetorical
associated with the execution of small notes ‘on the moments. In 24, Speidel marks all chords portato,
beat’, and by his final notation in 8, 16 he may have in 25 and 26 also the rh chords. Ganz marks the rh
envisaged the type of realisation of a turn figure chords 25 and 26 tenuto.
shown by C. Ph. E. Bach, or the kind of broader turn 19ii–iv. Vl: All except Alard, Kreisler shift 2–2, which
figure illustrated in Viennese treatises by Starke and would almost certainly have been the choice of most
Swoboda; the essential difference is that the second early-19th-century violinists.
form of notation leaves open the possibility for a 20iii–x. Vl: David, Alard, Hermann, Joachim evident-
more leisurely final note, a 16th rather than a 32nd.95 ly remain in 3rd position, crossing to the E-string;
Brodsky begins with an open E-string and remains
in 1st position; the others, including David-revised,
94 Carl Philipp Emanuel Bach: Versuch über die wahre Art das Cla-
vier zu spielen, 3rd edition, (Leipzig, 1787), vol. 1, 2tes Hauptstück,
go up the A-string, shifting to 5th position on viii;
4te Abteilung ’Von dem Doppelschlage’, § 24, pp. 66f (Eng. Trans. Kreisler also moves to 3 on 21i, presumably to fa-
by W. J. Mitchell (London, 1974), pp. 119f). cilitate vibrato.
95 Friedrich Starke: Wiener Pianoforte Schule op. 108 (Vienna, 1819),
p. 18; August Swoboda: Allgemeine Theorie der Tonkunst (Vienna, 23f. Vl: The fingering indicated by most editors here
1826), p. 51. is one that seems highly plausible for a violinist of

Beethoven’s time. They shift to 3rd position on v ings and additionally marks it with connecting lines
and mark a harmonic on x. The exceptions are Alard, between 39i–ii (2–2), 41iii– 42i (1– 4) between bow-
Rosé, who shift to 6th position and take a stopped strokes, and 44i–ii (4 –2), where he presumably en-
a2 before coming back to 1 on 24ii, and Brodsky, who visaged a more prominent portamento than the nor-
apparently remains in 1st position. Singer marks mal audible connection during slurred shifts, which
espressivo. Grützmacher also marks a harmonic, but he indicates between 42ii–iii (3–3) and 43i–ii (1–1).
also gliss. between 24i–ii. To produce an effective portamento with a string
30–36. Pno: Czerny’s advice “the decoration light and crossing in 44, using the 1–2 fingering in the edit
even, and everything will be performed in tempo”, ed violin part of the present edition, the 1st finger
taken at face value, might give a false impression, should descend all the way to b b1 followed by an
encouraging a ‘correct’ but static performance. In 30, immediate string change to f1 on the D-string. The
asynchrony applied to rh i and some of the reitera- same shift at 42ii–iii is given by Hermann, David-
tions of D will add expression and need not interrupt revised, Halir (who marks no other portamento
the sense of evenness. In 32 a slight broadening of here) and Seybold. David, and Joachim give no por-
the tempo will help accommodate the complex fior- tamento fingering, although David, on the basis of
itura ornament and would probably not have con- his manuscript annotation in his own editions, al-
travened the notion of in tempo in Beethoven’s era. most certainly employed it, and Joachim will surely
Czerny expressly advises, giving several examples, at least have expected 4 –2 down the D-string in 44.
that complex fioriture should be allowed to take 43f. Vl: All the editors take two bows.
more time than the notation suggests.96 In a foot- 43– 44, 52–53. Pno, Vl: Beethoven’s <> might elicit
note, Speidel offers a suggestion for grouping the a hastening towards and lingering at the apex, with
notes of the ornament as follows: vibrato in Vl, noticeable asynchrony in Pno, and a
œ œ œ œœœœœœ relaxation of tempo after (3/b/v).
œœœœœœœœ œ
46, 50. Pno, Vl: The crescendi might encourage hasten-
4 4 4 6
ing.
This is a classic situation to apply tempo rubato of 40iii. Vl: The editors are evidently divided on whether
the kind described by Mozart in a letter of 1777 (1/b), the turn after the trill should be with a-natural or
and still apparent in the playing of the oldest gen- a-flat. An open A-string is obviously intended by
eration of 19th-century pianists who recorded, in those who remain in 1st position: David, Hermann
which the left hand remains more or less in time, (who specifies a), Brodsky, Halir, Seybold. The oth-
while the right is free.97 The ornament might start ers, who go up the D-string, probably expected a;
earlier and/or finish later than notated. The same Singer, Auer specify it.
type of flexibility might be applied to the fioritura 46f. Vl: Only Brodsky leaves the two-bar slur unbro-
in 34, which might even be allowed to spill over ken; all the others take two bows. David, Alard,
into 35, a practice heard in Saint-Saëns rendition of Seybold mark 4 in 46 and a harmonic on 47i; Grütz-
the slow movement from Beethoven’s Sonata op. 31 macher (notating the passage an octave down for
no. 1 preserved on a Welte piano roll (rec. 1905). In the cello) also marks a harmonic on 47i, but addi-
34 and 35 each grace-note turn in rh should prob- tionally marks gliss. over the bow change from 46,
ably be left as late as possible. In 36 the slurred pairs which some violinists may also have envisaged. Ha-
of 16ths in rh might be made to sound as pleading lir, Rosé employ the D-string, others the A-string.
as possible by lengthening and strengthening the Rosé alone specifies portamento between 47i–ii.
first of each and shortening the second (2/a). 52f. Vl: All the editors take two bows. No portamento
38– 45. Vl: The editors divide between those who mark fingering is given from 52i–ii except by Singer, Rosé,
none or few shifts and those who employ a con- who also mark a connecting line. An open A-string
siderable amount of portamento. Singer marks con is marked on 53ii by Hermann, Joachim, David-
molto espressione at 38, but indicates only a single revised (it is implicit in David), Halir, Seybold, Kreis-
obviously portamento fingering, from 44i–ii. Rosé ler; some or all of these may have expected a degree
specifies the largest number of portamento finger- of portamento to the open string, as is made explicit
96 Czerny: Pianoforte-Schule op. 500, vol. 3, p. 33ff in Grützmacher’s cello version where he adds gliss.
97 See Peres Da Costa: Off the Record, pp. 232–233. to the open A-string. Portamento from a stopped

note to an open string in similar circumstances is 8th-notes of the Trio, Beethoven himself might have
explicitly required by Spohr in the first bar of his given it a metronome tempo between 96 and 104.
annotated version of the Adagio from Rode’s Sev- Haslinger . = 80
enth Violin Concerto.98 Moscheles-Cramer . = 92
54 –73. Pno: All chords (perhaps apart from resolving Czerny-Vortrag . = 80
chords on i in 64, 66, 68 and 70) might be arpeggiated Czerny-Simrock . = 76
at varying speeds according to context, and perhaps Alard/Diémer . = 76
slower on half-notes than quarter-notes. In 73 the Speidel/Singer . = 84
chord could be played very slowly and poignantly Kreisler/Rupp . = c. 80
to fill out the texture (5/c/ii). At 57 rh i (which Ha- Czerny writes: “With merry capriciousness. In the Trio
lir marks with a tenuto line), the dissonance might the crescendo well marked and the forte strengthened
be enhanced by asynchrony and a slight lingering. by the pedal. Everything very lively.” 99
The first note of the grace-note turn in 57 should be 0– 43. Pno: Czerny makes it clear in his Pianoforte-Schule
aligned with the bass, as indicated by Ganz (5/b/i). that short chords and those marked staccato ought
54 – 60. Pno, Vl: The 32nd-note figures could be played not to be arpeggiated. Very tight arpeggiation, how-
with subtly flexible rhythms, perhaps lingering in- ever, of the type referred to by Thalberg as presque
creasingly on the first note of each to enhance the plaqué (almost together), may have been normal. In
dissonance (2/b). any case, given the very fast tempo even tight ar-
61. Pno: The chord at i could be arpeggiated swiftly. peggiation may be impracticable, depending on the
The 32nd-note slurred pairs could be played un- type of piano being used (5/c/ii). Speidel marks leg-
equally for expressive effect (2/a). giero in 1. Ganz provides various sustaining pedal
71. Vl: If the violinist wanted to produce an echo of indications that might be of interest but almost cer-
the portamento in 70, the 2nd finger on note 5 could tainly reflects later 19th-century practice.
be pulled back somewhat before playing the open 26–27. Pno: The chords could be arpeggiated very
string. swiftly.
72. Vl: Singer marks perdendosi. 8–16, 21–27. Vl: The style of bowing this passage seems
to have changed significantly over time. Today it is
Scherzo generally performed in the middle or lower half of
Allegro molto the bow with the pairs of staccato notes in 10–12,
Tempo 14 –16, 22–25 played either   with an off-string stroke,
Beethoven’s own tempo markings for Allegro molto or with successive up-bows around the middle or
or Allegro molto e vivace 3/4 scherzos and scherzo/ slightly above the middle of the bow. Some young-
minuets are very fast; to all of them he gave numbers er 19th-century violinists may well have envisaged
over . = 100. Even the Allegro scherzos, except where this. Alard, Joachim provide no bowing instruc-
there are many 8ths or 16ths, are marked between 96 tions; Hermann, Auer simply mark 8i, 9ii, 9iii   
and 100. The faster scherzos often contain passages followed by separate bows; Halir marks the same,
of successive 8ths; the Allegro molto Scherzo of op. 18 followed by a succession of up-bows. Singer, Brod-
no. 1, and the Allegro molto e vivace Minuetto of the sky, Kreisler have    followed by successive up-
First Symphony op. 21, for instance, which both have bows, and Singer’s added instruction leggiero in 8
slurred 8ths in the Trio, are marked . = 112 and 108 may suggest a light off-string stroke. The bowing
respectively. In op. 24, the passages of 8ths in the Trio given by David, Seybold, however, has different
are marked staccato for both violin and piano, which implications: they mark    in 8–9, followed by 
may suggest a somewhat slower pace. Nevertheless,  in 10,   in 11 and   in 12; this is most plausibly
the Haslinger and Czerny markings, as well as those interpreted as a slurred staccato near the point and
by younger editors, seem substantially too slow for can be very effective.
Allegro molto. Moscheles’ marking is more persuasive, 18, 20. Vl: Open strings are marked by Alard, David-
but even this may be slower than the tempo conceived revised, Singer, Auer, Brodsky, Halir, Rosé (20 only),
by the composer. Perhaps, considering the detached
99 “Mit dem heitersten Muthwillen. Im Trio das crescendo genau
98 Spohr: Violinschule (Vienna, [1833]), p. 209. markirt, und das forte durch das Pedal verstärkt. Alles sehr lebhaft.”

Seybold, Kreisler and were surely regarded as too Haslinger  = 84
obvious to mark by David, Joachim. Moscheles-Cramer  = 92
19. Vl: A harmonic is marked by Alard, Brodsky, Ha- Czerny-Vortrag  = 76
lir, Seybold and was surely regarded as obvious by Czerny-Simrock  = 76
David, Hermann, Joachim, Auer. Alard/Diémer  = 152
28ff. Vl: The performance of the 8th-notes with an Speidel/Singer  = 69–72
elastic bowstroke in the middle or upper middle of Kreisler/Rupp  = c. 76
the bow without the hair leaving the string, as de- Czerny writes: “Of a similar melodious character to
scribed by Spohr and David (6, p. X X X I I I ) is more the first movement. The triplet middle section (in D
or less inevitable in this passage. minor) should be performed in a lively and marked
28–34. lh: It is not clear whether Beethoven expected manner, just as, in general, many passages also re-
staccato, as in rh, or perhaps non-legato which is quire fiery, brilliant playing.” 100
marked by Speidel. 1, 8. Pno, Vl: Speidel/Singer mark the principal theme
40– 43. Pno: The downbeats of each bar might be ar- grazioso; Reinecke/Hermann dolce; Ganz marks es-
peggiated swiftly or asynchrony might be applied pressivo at 1 but not 8.
(5/c/ii) 1–8i, 8–18. Pno, Vl: It is unlikely that a pianist or vio-
1–27 da capo. Pno, Vl: The normal Viennese practice linist of Beethoven’s time would have played the
at this period was undoubtedly to play repeats on rondo theme similarly on each of its three identical
the da capo unless otherwise directed. Beethoven appearances. Beethoven himself varied it substan-
sometimes indicated a scherzo with da capo senza tially on its fourth and final appearance (bb. 188ff).
repetitione. In this case the need for the observance Early-19th-century performers would probably have
of the repeat is obvious, since the first part has a made small modifications on the second and third
written-out repetition of the theme. By the late 19th statements of the theme, through rhythmical, orna-
century the convention of omitting repeats during mental, or dynamic variants (see the introduction to
a da capo was sufficiently established for Speidel/ the sonata above).
Singer to include a footnote instructing the player 1–18, 56–73i, 112–131. Pno: Pianists of Beethoven’s era
to observe it here. would certainly have used overholding of notes in
the broken chord accompaniment to support the
Rondo beautifully expressive melody (4/a/ii). They might
Allegro ma non troppo also have used sustaining pedal as an aid to this,
Tempo though, interestingly, Speidel gives no suggestions,
Beethoven specified relatively few metronome tempos and Ganz only in 17. For the melody in 1–8, pianists
for Allegro movements in  , and none for Allegro ma would have employed asynchrony frequently to
non troppo; the short Allegro ma non tanto section in colour important melody notes (5/c/ii), and would
the last movement of the Ninth Symphony, which he surely have altered the nuance and rhythms of notes
marked  = 120 provides little guidance, since it con- such as slurred pairs (2/b) and successive notes of
tains nothing faster than 8ths. For the Allegro first equal value in 5–6 (2/b) to create lilting or perhaps
movement of the String Quartet op. 18 no. 3, which even dotted rhythms. Interestingly, Halir removes
has a similar range of note values to this Rondo, Beet- the slurs from Beethoven’s slurred-pairs and marks
hoven also gave  = 120. To the Allegro con brio in the the notes alternatively with tenuto lines and stac-
first movement of the Septet he gave a slower tempo cato. In 7–8 the double-note chords might well have
of  = 96, presumably because it contained several pas- been swiftly arpeggiated. From 6–7 the < >
sages of 16th-notes. In this context Moscheles’ metro- might elicit an increase of momentum towards
nome mark for the Rondo in op. 24 seems rather fast, and a lingering at the apex, with a return to tempo
though not implausible; the Haslinger marking of 84 afterwards (3/b/v). This will very probably have
may be closer to Beethoven’s conception, but the slower caused non-alignment between melody and accom-
ones seem likely to represent a later tendency to reduce
the faster tempos, especially where the music here, as 100 “Von gleichem melodischen Character wie der erste Satz. Der
Triolen-Mittelsatz (in D-moll) ist lebhaft markirt vorzutragen, so
also in the first movement of this sonata, could be inter- wie überhaupt auch viele Stellen ein feuriges, brillantes Spiel er-
preted as predominantly lyrical rather than energetic. fordern.”

paniment. From 9–18 overholding of broken chords sustaining pedal indications and marks all chords
will provide resonance and the overall expression in rh with tenuto lines. From 34 –37 the notes marked
might be enhanced by swift arpeggiation of some sf in rh might be slightly delayed, which would cause
or all chords in both rh and lh. The sf in 15 and 17 an agogic effect, with the following note shortened
might inspire particularly noticeable arpeggiation somewhat (PT: 1/b).
or asynchrony (PT: 1/a and PT: 1/b). 20f, 24f 28f, 32f, 143f, 147f, 151f, 155f. Vl, rh: The trills
8–18i. Vl: The editions reveal a number of different ap- were certainly expected to begin from the upper
proaches to bowing and fingering the rondo theme. auxiliary, because they are preceded by a note at
There is no suggestion in the editions to treat any of the same pitch (5b/ii).
its first three identical appearances differently, but 23i, 31i. Vl: On 23i, Alard, Singer, Halir, Kreisler mark
in practice, it is unlikely that skilled 19th-century o (which some of the others probably considered
violinists treated it the same on each occasion. obvious); all of these except Kreisler also mark it
9, 11. David, Hermann, Joachim-revised, Auer, Halir, on 31i.
Rosé mark a slur from i–iv, over Beethoven’s slurred 27iii–xii, 150iii–xii. Vl: Singer marks a slur over the
pairs and, on v–vi, a slur over the staccato marks staccato dots with the recommendation stacc. ad lib.,
(Halir, Rosé replace the staccato marks with tenuto referring to the firm up-bow staccato.
lines, which they also employ on all the staccato notes 28f, 151f. Vl: All editors mark a hooked bowing for
in the theme). David-revised, Joachim do not con- the dotted figure.
tain these additional bowing instructions although 38–55, 161–188. Pno: All the octaves in lh might be ar-
they too are unlikely to have shortened the staccato peggiated swiftly and held full length (Ganz marks
notes significantly. tenuto lines on the octaves in 39 and 40). Beetho-
13. Vl: Hermann, Singer replace Beethoven’s whole- ven may well have expected the broken octaves in
bar slur with slurs from i–iv, v–viii. 40– 42 etc. to be played full length, though both Spei-
15i, 17i. Vl: These sf notes are taken  near the point by del and Ganz mark them staccato. In 38– 40 and
David, Alard, Joachim, Rosé, Seybold, probably en- 42– 44 the slurred figures with as much over legato
visaging fouetté; others, including Joachim-revised, as possible (4/a/ii), and chords arpeggiated (5/c/ii).
contrive to arrive , though Hermann does so only At 42i asynchrony would be particularly effective
for 15i. for the sf, and from 48–55 as much over legato and
15, 17. Vl: Several fingerings are proposed for this fig- overholding as possible in both hands. In 49 and 51
ure: David, Seybold mark a shift to 3rd position on a slight emphasis of the first note under the slurs
ii and remain there; Hermann, Brodsky, Rosé mark in rh would make these figures particularly expres-
the same shift but on 15v–iv mark 2–3; Alard, Halir sive. The trills in 180 probably have a main note
also shift to 3 on ii; Halir marks 1 on v in both bars, start because they are preceded by the note above.
Alard only in 17. Joachim, Auer, Kreisler remain in 40– 47. Vl: Singer marks bowing instructions, specify-
5th position, marking a 4th-finger extension on vi ing the lower half from 40ii, the upper half from
and 2 on vii; Singer keeps the whole figure on the 42iii (beginning  ), the middle from 44ii and the
E-string shifting 4 – 4 on i–ii; Alard, Hermann, Sin lower half from 46ii.
ger, Brodsky, Halir mark 0 on 17vii. 48iii–iv, 181iii–iv. Vl: All mark 2–2.
19–38i, 142–161i. Pno: All chords not marked staccato 55, 188. Vl: Singer marks ’ immediately before the
might be arpeggiated, generally swiftly, apart per- subito p.
haps from thoses in 19, 23, 27 and 31 which could be 63–73i. Vl: See notes for 8–18
spread more slowly to fill out the texture. Ganz marks 73–97. Vl: None of the editors retain the slurring of
19 and 27 tenuto. Important rh melody notes such as the sources exactly. None include the questionable
at 23i, 24i, and 34i might be expressively delayed as slur from 74vii–viii.
part of the arpeggiated lh or through asynchrony 73ii–105i. Pno: Any octaves not marked staccato might
(5/c/ii). The melodic sequence in 23 would have been be swiftly arpeggiated, making an expressive con-
expected to be played with rhythmic inequality ac- trast with those marked staccato (Ganz marks tenuto
cording to taste (2/b). In 20, 21, and 22 Beethoven at 74 rh but sustaining pedal on each half bar in this
probably expected full length quarter-notes though passage). Additionally, the syncopated rh octaves
Speidel marks all but 22i staccato. Ganz provides might each receive emphasis (Halir marks each >).

In 81 and 97 Beethoven probably expected legato for 120–124. Pno, Vl: Speidel/Singer mark poco a poco strin-
the triplet 8th-notes, though non-legato, or staccato gendo with a tempo in 124.
(as marked by Rosé and Halir in Vl) might be ap- 131–140. Vl: See notes for 8–18.
propriate. From 90–93 the beginnings of each slur 131ii, 196ii. Vl: Singer marks + and a footnote explain-
in rh might be given a fairly strong emphasis (Halir ing that it signifies left-hand pizzicato.
marks each with >) and legato fingering (alternating 132–144i. Pno: All broken chord textures with over-
4, 5 as marked by Ganz) used to connect the octaves. holding (4/a/ii). Speidel and Ganz give occasional
73iii–vii, 89iii–vii. Vl: Singer again marks a slur and sustaining pedal indications as an aid to resonance.
stacc. ad lib. Some or all chords might be swiftly arpeggiated
86ii. Vl: Alard marks a harmonic to descend from 7th and asynchrony applied according to taste. At 138
to 3rd position. lh i swift arpeggiation would mitigate harshness
88iii. Vl: Marked as a harmonic by Alard, Hermann, for the sf, while at 140 i asynchrony would enhance
Singer, Brodsky, Seybold. David, Joachim probably the sf. (5/c/ii; PT: 1/a and PT: 1/b).
regarded it as obvious. 167ii–171ii. Vl: All mark the passage without porta-
89i. Vl: Marked as a harmonic by Singer, Seybold. mento shifts, except between 170iii–171i, where Da-
91f, 99f. Pno: The rinf. will probably apply to the notes vid-revised, Singer, Halir, Rosé, Seybold, Kreisler
in both hands and perhaps signify an emphasis and mark one.
crescendo through the following notes (3/b/ii). 175ii–180. Vl: This passage was almost certainly in-
98ii–iii. Vl: Marked 4 – 4 by Alard, Hermann, Singer, tended to be played with a martelé bowstroke.101
Halir. Some of those who mark nothing may have The majority mark no fingering; of those who do,
regarded 4 – 4 as obvious. Joachim evidently expected 1st position from 176–
99. Vl: The rinf. probably signifies an emphasis and 177iv, thereafter 1– 4 except where open strings oc-
crescendo through the following notes. cur; Halir takes most pairs 1– 4, but 176v–viii in 3rd
106–108. Pno: Slow arpeggiation of the lh chords in position with a 4th-finger extension on viii; Kreisler
106 and 108 is appropriate to fill out the texture marks 1st position from 176iii–177iv then 1– 4, until
and create softness (Speidel marks these tenuto), 179v, where 1st position is again resumed. Leaping
but the lh chord in 107 being a resolution might be over an intermediate string was a practised tech-
played unarpeggiated (5/c/ii). Ganz marks sustain- nique in Beethoven’s time, which was much used by
ing pedal for all three chords. The rh melody notes Viotti in his concertos and supplied for practice in
106i and 108i could be significantly delayed which Fiorillo’s Études (no. 30), Kreutzer Études (nos. 6, 16),
would make them particularly expressive and help and Rode Caprices (nos. 4, 15, 21). The feasibility of
achieve softness. this depends somewhat on tempo; at the lively speed
109–111. Pno: Beethoven may well have expected this probably envisaged by the composer, 1– 4, where
passage to sound improvised perhaps with the possible, may be preferable.
8th-notes starting slower and increasing in tempo 196iii. Vl: Alard, Joachim Kreisler mark , a very con-
through the triplets and 16ths. He may also have venient bowing after the pizzicato, while other edi-
envisaged a slightly under tempo start for the re- tors retain the , used on other occurrences of the
turn of the theme in 112 marked by Speidel as poco main theme; to facilitate the latter, Singer specifies
tranquillo (see note below). It is also possible that, left-hand pizzicato on 196ii, as in 131, and on iii marks
on pianos of Beethoven’s time, use of the sustain- not only  but also the instruction Bogen oben (upper
ing pedal throughout might have been envisaged, part of the bow).
which would create a very special effect especially 197i–iv, 199i–iv, 201i–202iv. Vl: Beethoven surely ex-
if the una corda was also used from the pp in 107. pected his dotted notation to stand for what would
3
117f, 119f. Vl: All indicate 117 as D-string and 119f as later have been notated as   . This modern triplet
A-string. A harmonic is specified in 117 by Alard, notation was scarcely used at this date; and even
Hermann, Singer (who also marks espressivo and towards the end of the 19th century there are clear
<> peaking at the barline), Auer, Brodsky, Ha- instances of dotted notation standing for triplets (2/
lir, Seybold, and may have been regarded as obvi- c/iii).
ous by others; Seybold also marks o on 118ii. In 120 101 See the example from the 1803 Paris Conservatoire Méthode in
all shift 4 –2 except Kreisler, who marks 1 on ii. section 6 of “Reading between the lines” p. X X X I .

Those who start  in 196 either take each of the dot- unmarked or added staccato marks. Auer, however,
ted pairs with a hooked bowing (David, Brodsky, added slurs on 223 viii–ix and xi–xii. Perhaps slurs
Halir, Rosé, Seybold), or just the first pair hooked, over each group of three or more notes was envis-
followed by   (Joachim-revised, Auer); Alard, Jo- aged in 222, or a variety of slurs reflecting the pre-
achim, starting  in 196 take all the dotted pairs  ; vious patterns. A possible solution is suggested in
Kreisler marks an unusual mixture of hooked and the edited violin part of the present edition.
separate bows. 224ii–228i. Pno: Beethoven may have expected this
189–205. Pno: Beethoven will undoubtedly have ex- hymn-like interlude to be in a broader tempo which
pected the lh broken chords from 189–196 to be le- Ganz marks espressivo. The portato articulations and
gato with overholding (4/a/i). Some of the editors the chordal texture signal expressive arpeggiation
mark slurs and sustaining pedal. For the rh slurred in varying speeds (5/c/ii).
pairs, both dynamic and rhythmic nuance would be 228–243. Pno: Overholding of all broken chords with
an historically appropriate option and would be so- judicious use of sustaining pedal (see Speidel and
phisticated (2/a). From 197, as much overholding as Ganz) was probably envisaged, and all chords, par-
possible with perhaps some use of sustaining pedal ticularly the lh octaves marked sf, would gain ex-
(Speidel and Ganz offer the same marks here). Oc- pression from being swiftly arpeggiated (5/c/ii; PT:
casional arpeggiation of chords according to taste is 1/a). If the tempo is broadened from 224ii, it might
an option (5/c/ii). In 203 and 206 the sf might elicit start to return to tempo from the middle of 232
asynchrony (PT: 1/b). which Speidel marks appassionato, with animato in
205. Pno, Vl: The absence of > after the sf, in contrast 236.
to all earlier statements of this figure, is probably 229. Vl: Almost all take the passage from 228–232 with-
not the result of oversight; the forte continuation to out audible shifts, beginning in 3rd position and
212 is musically persuasive. moving to 5th on 229iv. Alard, however, shifts to
206–224i. Pno: The sfs would gain special energy from 5th on 229iii, and Singer, marking (espressivo), shifts
either asynchrony or arpeggiation (PT: 1/a and PT: 1–1 on 229i–ii and explicitly indicates portamento
1/b). Both techniques could be employed according with a slanting line.
to taste. All the broken chords with or without slurs 231iii, 234, iii, iv, 235iv. Vl: Beethoven marks no turns
overheld. In 209ff assimilation of the dotted figures after the trills, but he surely expected one in 231
in lh to match the rh triplets would have been nor- and 235 and at least after the second trilled note in
mal practice (2/c/iii). 234. It is probable that those who mark a 1st finger
213. Vl: All descend 2–2 on iii–iv except Singer, who on 234iii (David, Singer, Bordsky, Halir, Seybold)
marks the more expressive 4 – 4 shift, Rosé, who takes did not envisage a turn there; those who marked
the figure in 4th position without shifting, and Kreis- 2–2 (all the rest except Kreisler) may have done so.
ler, who remains in 3rd position. Singer and Grützmacher indicate turns after 231iii,
215–217. Pno, Vl: The figures with 8th-note followed 234iv, and 235iv, Hermann only in 231 and 235.
by two 16th-notes would almost certainly have been 232iii–233ii. Vl: A portato bowing (dots under slur)
assimilated to the triplet rhythms, leaving the 16ths is marked on 232iii–v by all except Alard, Joachim
late. (though it is portato in Joachim-revised), Kreisler,
221–223. Vl: The slur over the last three notes may and Singer; the latter not only adds the instruction
possibly be an engraver’s error, but it may well re- appassionato, but also instructs B. u. (Bogen unten:
flect the changed pattern of notes here and in the the lower part of the bow) and adds tenuto lines
following two bars, where there are neither slurs on 232iii–v and a portamento line across the bow
nor staccato marks. Other passages of accompani- change between 232v–233i, fingered 2–2.
ment figures, for instance in op. 12, are also left 236. Pno, Vl: Speidel/Singer mark animato.
ambiguous. In this legato context, especially in 223,
Beethoven seems unlikely to have envisaged sepa-
rate bows throughout. The editors of AG changed
221x–xii to agree with the previous slurring pattern
and left the next two bars unmarked. Later editions
adopted this reading in 221 and most either left 222f

3–7, 10ff, 152ff, 159ff etc. Pno: The entry of the con-
SONATAS OPUS 30
trapuntal line in 3 rh and 11 lh i, and other points of
Opus 30, No. 1 interest such as 5 and 6 might be given special at-
Allegro tention and expression with the use of asynchrony
Tempo (5/c/ii and PT: 1/a); and overholding wherever possi-
There is one analogous 3/4 Allegro for which Beetho- ble would create a delicate resonance without need
ven supplied a metronome mark: the first movement for sustaining-pedal, though sustaining pedal might
of the String Quartet op. 18 no. 1, to which he gave also be judicously employed.
.  = 54. In that movement there is a similar proportion 8. Pno: Given the fp on i, the chord might be arpeggiat-
of 16ths, partly legato and partly detached. Moscheles’ ed with the lowest note in the lh played first, with
suggested speed is close to Beethoven’s tempo for that the beat, and the other notes in lh and rh played
movement; Alard’s, unusually, is also rapid; tempos together very slightly afterwards, which would also
suggested by other 19th-century musicians are slower, produce a slight agogic effect (PT: 1/a).
slightly (Haslinger/Czerny) or substantially (Speidel/ 10–13. Vl: All the editions except Alard’s divide Beet-
Singer) than Beethoven might have been expected to hoven’s four-bar slur.
indicate. 14f. Vl: Contrary to the sources, AG gives separate
Haslinger  = 144 slurs in each bar and all the editions follow this.
Moscheles  = 160 17iii, 166iii. rh, Vl: Reinecke/Hermann, Grützmacher,
Czerny Vortrag  = 144 Speidel/Singer include a turn (Nachschlag) to the
Alard/Diémer  = 152 trill (5/b/ ii). At 17 Speidel specifically illustrates in a
Speidel/Singer  = 138 footnote that the trill should start on the main note:
Kreisler  = c. 130 ##
& # œœœœœ œ
7
Czerny comments: “Of calm, gently grave character. #œ

More speaking than sentimental. The triplet passages
from the 40th [sic] bar on must be played very lightly Halir gives a different solution for the trill:
staccato and strictly in tempo.” 102 ##
& # œœœœœœ
3
Beethoven’s dynamic, accent and articulation marks #œ

throughout this sonata represent only the minimum
of what he and musicians of his era expected to hear Reinecke’s fingering implies this too. Czerny, how-
in performance, leaving the finer matters of expres- ever, was clear that in these circumstances it should
sion to the sagacity and sensitivity of the performer. have both an upper-note start and a turn.
The dynamics, accents, articulation and other marks 19. Pno: A swift arpeggiation of the chord on i would
added by Speidel/Singer, Halir, Rosé, and Auer/Ganz, have an energising effect (PT: 1/a).
though representing a later 19th-century view will 22–23, 26, 171–172, 175. Pno, Vl: In Beethoven’s era,
nevertheless be instructive of the types of expression slurred-pairs of 8th-notes would almost certainly
expected in Beethoven’s era. have been played both with dynamic and rhyth-
1. Vl: Alard marks V. mic inflection, the first stronger and longer than
1, 19, 27 etc. Pno: A swift arpeggiation of the chord the second (2/a). Such inflection might also apply
on i would have an energising effect (5/c/ii) whether to 8th-notes in bar 22 and similar places.
in piano or forte dynamic and could apply to dou- 24i, 173i. Vl: Alard, Hermann, Singer, Brodsky mark a
ble-note chords such as in 22 and 23 lh and other harmonic, Singer with 2 (following 2 on the previ-
chords in both rh and lh. ous note); probably David, Joachim regarded a har-
2–3, 9–10. lh: Pianists of Beethoven’s time would prob- monic (with 4) in this context as obvious.
ably have overheld the notes in 2 (4/a/ii). Curiously, 31ii–32i. Vl: Alard, Hermann, Singer, Auer, Brodsky
Beethoven’s slur in 9–10 is different from 2–3, but mark a shift up the A-string from 2–3.
would surely not have implied a different phrasing, 31–32, 180–181. Pno: The sf could be made more effec-
as acknowledged by all editors. tive either with a very swift arpeggiation across lh
and rh, or by playing rh slightly after lh (PT: 1/a).
102 “Von ruhigem, sanft ernsten Character. Mehr sprechend als
sentimental. Die Triolenpassagen vom 40sten [sic] Takte an muss The lh chords swiftly arpeggiated to enhance en-
sehr leicht staccato und streng im Tempo gespielt werden.” ergy (5/c/ii).

34 – 40, 95–98, 102–105, 110–113, 187–193, 234 –238. Pno: ing of 52i and 54i would make the dissonant har-
On important or accented notes, asynchrony with monies more expressive.
rh slightly after lh (or perhaps exceptionally before 50f, 203f. Pno: The grace-notes here are almost certain-
in a few cases) would be stylish (5/c/ii; PT: 1/b), and ly intended to be played as acciaccaturas, as taught
accord with the expressive quality of the passage in Junghanss’ Pianoforte-Schule (Vienna, c. 1820), si-
which Halir marks dolce. In a footnote Speidel ad- multaneously with the main note and quickly re-
vises that the sf in 38 rhiii and 39 rhiv “are valid leased (5/a/iii).
only for the right hand”.103 Overholding of slurred 51iiff, 55ivff, 63, 67, 114 –117, 134, 140, 204iiff, 208ivff,
notes would create a delicate resonance. It is inter- 216, 220. Pno, Vl: Beethoven’s intention for the ar-
esting to note that Beethoven marks dolce at 95 but ticulation of the 16th-notes here is unclear. In the
not at 34, perhaps indicating that he wished for a piano part the absence of both slur and staccato
variation of effect perhaps with the use of a moder- suggests a default legato with overholding, but in
ator or una corda shift. In 110–113, the lh double-note the violin part, the few staccato marks following
chords (except 113i) could be arpeggiated according the slur are typical where the composer wants to
to taste, but particularly those marked sf. show that slurring should not continue. It is plau-
35i, 36i, 43i, 44i. Pno, Vl: Although Beethoven does sible, however, that many violinists of Beethoven’s
not provide turns for the trills here, he does so in time, might have mixed slurred and separate notes
the recapitulation (188, 196) where an accidental is in a variety of patterns, just as pianists might have
required; and where the figure comes in the devel- varied the accentuation. The markings of 19th-cen-
opment (96f) he provides a turn in both bars. AG tury editors reveal varied understanding of this no-
prints the turns here, providing a b # in 35/43; this is tation. Speidel marks non legato at 51, but provides
followed in the editions. Beethoven surely intended a slur from 55ii–58xii; at 55 Singer marks “Mitte”
the b #, although in such circumstances he normally (middle of the bow) and leggiero, probably imply-
indicated an accidental. Czerny’s arrangements of ing a stroke equivalent to David’s hüpfender Strich,
op. 47 (see below) confirm the likelihood that these perhaps changing gradually to a short détaché at the
trills, preceded by the note above, were expected to crescendo. It is possible that Singer’s suggested bow-
begin with the main note. stroke might have been used here by some of the
41– 48. Pno: The triplets that Czerny instructed “must violinist of Beethoven’s time, although by the time
be played staccato very lightly” have slurs added by he composed this sonata, that kind of bowing, con-
Reinecke, Diémer, Speidel, who also gives sustain- demned as unworthy of serious music by Spohr and
ing-pedal marks through each bar. Romberg, was rapidly going out of fashion, and a
43f. Vl: David, Hermann, Joachim, Singer, Auer, Halir, short detaché in the upper half seems more likely
Rosé, Seybold mark the first note of the turn after to have been the general choice (6, p. X X X I I ). In con-
the trill in 43 to be played with 1st finger, remaining trast to Speidel, Reinecke marks 51–55i, 63, 67, etc.
in 2nd position with 4 on ii; Kreisler gives the more with a legato slur and similarly at 204ff; Reinecke
modern fingering with the trill and turn executed also adds a slur from 208ii–212i, but this is missing,
in 1st position followed by 3 on ii. David, uniquely, probably through oversight, at 55ff. At 204ff Speidel
marks 1–1 for the turn in 44. again writes non legato but leggiero is absent from
46– 48, 99ff, 199ff. Vl: David, Hermann, Singer, Brod- the violin part (probably through oversight).
sky, Halir, Rosé take some or all the staccato 8th- 60. Vl: Slurred staccato is marked by David  iii–vi,
notes with slurred staccato (removed in David- Singer  ii–vi, Brodsky  i–vi Halir, Rosé, Seybold,
revised). Singer additionally marks leggiero. and David-revised  iv–vi.
49–54, 203ff. Pno: Some or all of the lh chords could 61–62, 132–133, 142–148, 241–215. rh: Moderately slow
be swiftly arpeggiated which would create subtle arpeggiation would fill out the texture and enhance
energy and aid tension and dynamic (5/c/ii), though the effect of piano (at 142ff the effect of pp).
Speidel marks these staccato. 49 rhii might be played 60vi. Vl: A harmonic is given by Alard, Hermann,
after the bass to enhance the sf. A slight lengthen- Singer, Halir. David, Joachim, Auer almost certainly
regarded it as obvious.
103 “Die Sforzati bei a) u. b) haben blos für die rechte Hand Gül- 61, 214. Vl: A harmonic is marked by Alard, Hermann,
tigkeit” Singer, Auer, Brodsky, Seybold. Rosé probably did

not intend one, because he marked <> (also in 62, 65f, 138–141. rh: All chords could be swiftly arpeggiated
214ff), probably envisaging vibrato in these places (5/c/ii).
(although at an earlier period <> is sometimes mark- 142–147. Vl: Beethoven’s tenute over these bars may
ed on a harmonic or open string). be not only a reminder to hold the note for its full
70–72, 76–79, 223ff, 229ff. Pno: It would be effective to length, but also an indication that it should be given
arpeggiate the third beat in each bar (across lh and extra emphasis, as indicated by the Viennese musi-
rh) with the subsequent down-beat unarpeggiated cian August Swoboda (2/c/vi).
except at 79 and 232–233 where a moderately slow 149f. Pno, Vl: The chords in rh with very swift ar-
arpeggiation of lh would help fill out the texture. peggiation as an aid to achieving a rapid cresc. Spei-
74 –75, 227–228. Pno: The downbeat of 74 arpeggiated, del/Singer add poco rit. in 149 and in tempo in 150.
but 75 unarpeggiated, would produce the effect of 166iii. rh, Vl: See note to 17iii.
light and shade. 182–184. Pno: Occasional asynchrony between rh and
81–82. Pno: Spreading the chords in lh and or apply- lh would provide a contrasting texture for the p and
ing asynchrony between lh and rh would enhance be an aid to the sf (5/c/ii; PT: 1/b). In a footnote, Spei-
dramatically the effect of crescendo. del explains that the sf applies only to the rh.104
83–86 (second time bar). lh: Although Beethoven tied 194ff. Pno: See note to 41.
the A in the bass in 83–84 and 85–86, it was custom- 195–201. Vl: Fingering in this passage provides an
ary to repeat notes where needed if the sound had instructive indicator of the limited expectation for
decayed too much. Depending on the piano being vibrato except as an expressive gesture, or reinforce-
used, re-striking these notes and also arpeggiating ment of accents, and it also shows that it was not
each chord would be an aid to filling out the tex- universally regarded as an obligatory adjunct to the
ture. execution of sf. An open E-string is marked on 195i
84, 88–92. Pno: The slurred broken chords with all by all the editors except Alard, who probably re-
notes overheld would enhance resonance and is an garded it as obvious; as an alternative, Singer of-
alternative to using sustaining-pedal. (4/a/i). In 88– fered the expressive fingering 4 –o 4 as his principal
92, Speidel marks rh portato which seems apt for fingering (the open string on 195i is marked only
the expressive quality of the writing. in the violin part in the piano score). On 197ii a
85. rh: Beethoven probably expected the trill to com- harmonic is marked on the sf by Alard, Hermann,
mence with the upper auxiliary (5/b/ii) and on the Singer (with 2), Brodsky, Seybold; David, since he
beat as marked by Ganz with a grace-note with dot- marks no alternative, surely regarded it as obvious.
ted lines connecting it to lh. Speidel and Vogrich Joachim’s and Auer’s fingering makes it clear that
give fingering which implies a main note start. they intended a stopped note here. For the sf on
87. Vl: Singer marks a harmonic, Halir an extended 201ii, only Alard, Hermann, Seybold mark a har-
4th finger on the G-string. monic, but David would very plausibly have played
116–117. Pno: To create special accentuation, the rh one.
octaves marked sf might be played asynchronously 207ii–208i. Vl: Here too a harmonic is indicated by
after the bass, and the octave itself very swiftly ar- Alard, Hermann, Singer, Brodsky, Seybold (only
peggiated. on 207i with 4 on 207ii). David, like Joachim, Auer,
118–130. Pno: It is unclear what articulation Beethoven marks 3, but this was removed in David-revised.
expected for this passage. Speidel marks each note 208ii–210. Vl: If harmonics are chosen, the hand will
staccato but also indicates sustaining-pedal with a already be in 3rd position in 207 (with an extended
change of pedal every three bars. 4th finger); if not, having established the hand in
119vi–vii, 123vi–vii. Vl: All except Alard, Brodsky 3rd position in 208, it is most effective (especially
mark the typical 19th-century 1–1 fingering. with a 19th-century violin hold) if the thumb re-
131, 135. Vl, Pno: Some of the annotated editions as- mains immobile while the fingers merely extend
sumed that a decrescendo, as in other occurrences back to 2nd position for the g #2 and up to 4th for e3.
of this material, had been mistakenly omitted here, 248. Pno, Vl: Speidel/Singer add slentando.
and added one. Since Beethoven did not include
one in either Vl at 131 or Pno at 135, it seems un-
likely that it was omitted in error. 104 “Die Sforzati bei a) b) u. c) gelten nur für die rechte Hand”

247–248. rh: Moderately slow arpeggiation of the that Beethoven regarded them as standing for a 2 : 1
chords particularly in 248 will help achieve the pp rhythm rather than a 3:1 rhythm, though this certain-
and fill out the texture. ly does not mean that he intended the rhythm to be
249i. Vl: All the editors mark a harmonic (Singer exact or unvaried; skilled and experienced perform-
with 2). ers will have been expected sometimes to modify
it, as they would all rhythms to a greater or lesser
Adagio, molto espressivo extent, according to context. Max Rostal’s comment:
Tempo “Rhythmically, there are delicate problems to be
Beethoven’s treatment of tempo in 2/4 meter is far from overcome here, as for instance the absolute precision
straightforward and attempts to explain it are inevit- of the 32nd-notes in theme and accompaniment”
105
ably complex and inconclusive. The tempos supplied nicely demonstrates the predominant 20th-century
for this movement by 19th-century editors are very misconception that the composer expected such
varied. Unusually, Moscheles indicates a significantly rhythmical details to be delivered precisely as writ-
slower tempo than Haslinger, Czerny, Speidel/Singer. ten.107
Moscheles’ tempo relates well to the metre/tempo term/ 1, 27, 64. Vl: Various approaches to fingering the begin-
note values relationship that seems broadly to lie be- ning of the theme result in quite different charac-
hind Beethoven’s tempo conception. The faster metro- terisation. A harmonic is marked in David-revised,
nome marks may represent a tradition of performance Halir, Seybold. David has only 4 (in the revised im-
that was already current in Beethoven’s later years, pression o has been added above it). Others start
but do not necessarily correspond with the composer’s either with a firm 4 on the A-string (Rosé), in 1st
original expectations. On the other hand, the relative- position on the E-string (Hermann, Joachim, Auer,
ly slow harmonic change and the lilting quality of the Brodsky, Kreisler), or 3rd position on the E-string
accompaniment to the principle theme, may encour- (Alard, Singer). It would be effective to use different
age a somewhat more flowing tempo than Moscheles fingering on successive appearances of the theme,
gave. Comparison of the metronome tempos given for to vary tone colour and characterisation. Beethoven
this Adagio and the 2/4 Adagio espressivo of op. 96 is will surely have expected trained instrumentalists
interesting. of his day to vary the opening music on its repeat at
Haslinger  = 76 26 using dynamic shading, rhythmic freedom and
Moscheles  = 63 perhaps some simple ornaments (5b and c; see also
Czerny Vortrag  = 72 Sonata op. 24 in this Commentary).
Alard/Diémer  = 58 1ff, 27– 42. Pno: In addition to modifications of strict
&rhythm Œ Ó above), pianists of Beethoven’s
∑ œ™ œ (see‰ note
Speidel/Singer  = 69
Kreisler  = c. 72 era would certainly have employed asynchrony and
Czerny provided the following recommendations: arpeggiation in a slow movement such as this as a
“The 32nds in the accompaniment sharp and distinct. &matter of course∑ to heighten
∑ ∑ the ∑expression
∑ ∑ im- ∑
of ∑
The whole thing in a very singing style and with feel- portant moments (both melodic and harmonic) and
ing. The ornament in the 50th bar very light, delicate, 19 to vary the texture (5/c/ii). This might be seen to be
and in tempo, very little smorzando. The character of &the ∑equivalent

∑ ∑ the∑ violinist’s
of ∑ ∑ of∑ subtle∑ and ∑
use ∑
106
this movement is gentle, almost ballad-like.” varied vibrato and portamento. In 1 for example,
1ff, 27ff. Pno, Vl: A major consideration in this move- 29 the double-note chords in rh might be arpeggiated
ment is the performers’ treatment ‰Œ Ó

& of ∑the œ™ œ. Beet- ∑ but
&swiftly ∑ gently,
∑ particularly
∑ ∑ the∑ chord ∑ marked
∑ sf ∑ ∑
hoven and his contemporaries routinely used this (PT: 1/a and b). In 8 (and 71), the chords in lh would
notation for a spectrum of long-short rhythms, from 39 sound particularly soft and expressive with gentle
&
2:1 through to 6:1 or even more over-dotted ∑
rhythms ∑ &arpeggiation.
∑ ∑ ∑ ∑ ∑And ∑ ∑9, when
∑ in ∑ ∑ the∑ ∑ piano∑ takes∑ over ∑ ∑
(2/c/i–iii). In this movement, it seems most likely the melody, the use of frequent asynchrony will
19 49 produce a special expressiveness that cannot be
&
105 See Noorduin: Beethoven’s Tempo Indications, ∑
pp. ∑
98–105. ∑ ∑ ∑ ∑ ∑ ∑with∑ all
&achieved ∑ ∑ played
∑ notes ∑ ∑ ∑ exactly
∑ ∑as notated.
∑ ∑ ∑
106 “Die 32steln in dem Accompagnement scharf und bestimmt. Das In 10, the notes marked portato elicit asynchronous
sten
Ganze sehr singbar und mit Empfindung. Die Verzierung im 50
Takte sehr leicht, delikat, und im Tempo,29sehr wenig smorzando. 59
∑ ∑
&Balladenartig.”
Der Character dieses Satzes is sanft, beinahe ∑ ∑ & ∑ ∑ ∑Beethoven,
107 Rostal: ∑ ∑ ∑p. 94.∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑
39 69
© 2020 by Bärenreiter-Verlag, Kassel
& ∑ ∑ ∑ ∑ &∑ ∑∑ ∑ ∑ ∑ ∑ ∑Opus
∑ 30, No. ∑1 ▪ 75 ∑
49 75
& ∑ ∑ ∑ ∑ &∑ ∑∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑
performance as described by Louis Adam and oth- the bass as indicated by Ganz with dotted line nota-
ers (4/b/i). Ganz marks con Pedale at 1 lh. tion. Beethoven most likely expected legato through-
5i–ii, 31i–ii, 68i–ii. Vl: All mark a shift to 2 up the out the bar.
A-string except Alard, Joachim (who mark no shift), 26. rh: The lh chord, which Ganz marked tenuto, could
Kreisler (shift to 3); Singer changes bow between i be arpeggiated slowly. The trill could start either
and ii but marks a portamento line across the bow from above or below in this instance, but was almost
change. certainly not expected to start from the main note.
6, 23. Pno, Vl: The descending 16th-notes most likely Although Beethoven provided an Eingang, there is
legato and could be played unequally which would no reason to suppose that he expected the same or-
give them an expressive lilt (2/b). nament on successive performances. Pianists of his
9–16, 21–25, 35– 41, 87–90, 96. Pno: Asynchrony with rh time were expected to improvise their own Eingänge
slightly after lh for important melody notes would (5/c/i). Speidel gives the following realisation of the
be an aid to expressiveness and indispensable for a trill and Eingang:
singing style. Speidel marks rh cantabile at 9, 35 and
# œ# œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ# œ œ œ œ œ
72. At 83 rhii Speidel marks espressivo. Ganz marks &# ˙
espressivo at 9 and 21. In lh, overholding as much pp 10 10 10 poco rit.
as possible would mitigate the need for sustaining-

pedal, though Ganz gives sustaining-pedal indica- 42– 49. Pno: In 42f, asynchrony is apt to achieve a mys-
tions. In 15 Speidel gives this realisation of the trill terious pp. From 44 – 49 arpeggiation with varied
in rh: speeds and dynamics will produce contrasts.
# œ 50. rh: Given the complexity of the ornamentation in
& # œœœœœœœœ this bar, a slight broadening of the tempo would
be appropriate as recommended by Czerny109 and
13–15, 39– 41, 76–78. rh: The octaves might occasion- allow for a quasi-improvisational effect which Beet-
ally be arpeggiated, a practice indicated in annota- hoven surely intended.
tions to an 1848 edition of Mozart’s piano sonatas 51–59, 64 –70. Pno: The broken chord figurations could
by Cipriani Potter (who knew Beethoven and very be overheld to create resonance with or without sus-
probably heard him play the piano).108 taining pedal (4/a/ii). In 56, rhii–iv should probably
17. Vl: All shift up the D-string between i and ii; Da- be assimilated with the last three sextuplets in lh as
vid, Alard, Halir, Seybold to 3 followed by a har- was customary in Beethoven’s era (2/c/iii). In 51–55,
monic on iv; Joachim, Hermann, Auer to 3 followed lh chords particularly on main beats might be ar-
by a stopped 4; the others shift to 2 followed by 4 peggiated. In 64 and 66, the chords in lh marked sf
on iv, except Kreisler, who marks 3 on iv, probably could be swiftly arpeggiated (PT: 1/a).
for the sake of vibrato. 57f. Vl: Beethoven, as an active string player, must
19ff, 44ff, 87ff. lh: The octaves might occasionally be surely have been aware that his slur required the
arpeggiated on longer note values, which could be violinist to go up the D-string for 58ii, which is
used to enhance the effect of either softness, loud- what is marked in all the editions.
ness, or accent. 57, 59. Pno: Asynchrony at i with rh after lh would
20v–vi. Vl: AG supplies the missing slur from iv–x enhance the effect of sf guarding against harshness
and all the editors indicate a shift up the A-string, while creating and agogic accent (PT: 1/a). The grace-
which would certainly have been expected to be notes should be sounded simultaneously with the
audible. first main note (aligned with the bass) and released
20. rh: The chords might be arpeggiated in varying quickly, so that the triplet rhythm is preserved (5/
speeds according to expressive aims; the first two a/ii).
very swiftly, particularly the second, the third more 58. rh: The first note of the slide/Schleifer should be
slowly to give softness for the decresc. (5/c/ii). aligned with the bass as indicated by Ganz with
25. rh: The grace-note should probably be aligned with dotted-line notation (5/a/iv).
60–61. Pno: The chord in 60 could be arpeggiated very
108 See: Cipriani Potter, “Recollections of Beethoven with Remarks
on His Style”, in: The Musical Times and Singing Class Circular, De-
cember 1861, vol. 10 no. 226, p. 152. 109 Czerny: Pianoforte-Schule (c. 1839) vol. 3, pp. 33–34.

deliberately and softly, while for dramatic contrast 82i–ii, 83i–ii. Vl: A rising appoggiatura figure of this
the chord in 61 played rather abruptly sf, the lowest kind was described by Domenico Corri as swelling
note in lh with the beat and the other lh and rh into the resolution,111 whereas a falling appoggia-
notes unarpeggiated and slightly afterwards (5/c/ii; tura was to be performed more strongly than the
PT: 1/a). note of resolution.
62ii–iii. Vl: Here David, Halir mark an expressive 85. Pno: The grace-note was probably expected to
downwards shift 2–3; all others remain in 3rd posi- align with the bass (5/a/ii), as indicated by Ganz
tion until 63i. In David-revised, the 3 is removed with dotted line notation.
and 2 added on 63i. 89vi. Vl: A harmonic is marked by Singer, Kreisler
62. Pno: Any or all of the octaves might be slightly (who also marks one on iv).
broken (arpeggiated). 90. rh: Many of the editors mark the turn on iii, rather
63f. Vl: All evidently expected an open A-string on than as Beethoven did between iii and iv. Reinecke
the upbeat to the theme. David, in contrast to the realises the turn according to Beethoven’s placement:
opening of the movement, also marks a harmonic # œ œ œ œ4 œ œ œ œ. œ
for the a2, which he presumably retained until he &#
indicates 4 on 65ii; Seybold gives the same finger- sf
5
? ##
ing. Singer, Halir also mark the harmonic in 63, œ œ œ
which is retained until being replaced by 1 on the œ œ œ
E-string at 64ii. Rosé explicitly marks the open string
in 63 and connects it with a portamento line to 4 on However, Halir gives a different realisation:
the following note. ## œ œ œœœœ œ. œ
63. Pno, Vl: Beethoven’s expectations for the begin- &
ning of the trill are unspecified and unclear. A vio- 3
linist of his time might easily have chosen a main 91–93. Vl: All except David, Rosé, Kreisler mark a har-
note start, while a pianist began with the upper note monic (David, almost certainly regarded it as obvi-
(neither seeing the other’s part). Later in the century ous and it is included in David-revised). A fingering
a main-note start for both would have been most is marked on the upbeat in 91 only by David, Her-
likely. The first note of the inverted double turn in mann, Brodsky (2), Kreisler (1); other editors, having
Pno and Vl should probably align with the lowest been in 1st position immediately before, mark noth-
note of the lh chord on the beat as advised by Spei- ing, apparently assuming an open string. It is prob-
del.110 The lh chord might be arpeggiated slowly. able that many of these violinists envisaged a hint
64 –70. Pno: The rh broken chords with overholding of portamento into the a2 (as marked in 63 by Rosé).
(4/a/ii); the lh chords arpeggiated according to ex- This might be accomplished by putting the first fin-
pressive aims (5/c/ii). ger at the nut and sliding it up almost imperceptibly
69–74. Pno, Vl: The dotted notation was surely intend- from the open string before allowing the 4th finger
ed to stand for triplet notation here. to come down onto the string just before reaching
72–77. lh: The figurations, though not marked legato, the harmonic; see Spohr’s instructions for rising into
can be played with overholding of notes wherever a harmonic.112
possible to create resonance, with or without the 91–95. rh: The double-note chords gently arpeggiated
sustaining-pedal (4/a/ii). Alard, Speidel mark slurs; would enhance create a softened effect to the feel-
Ganz marks poco legato. ing of pp (5/c/ii). Ganz marks this una corda. Speidel
81–86. Pno: The precise degree of long-short for the marks poco marcato in lh.
continuous dotted rhythms would have been the 95i–ii. Vl: None of the editors give a fingering, all ap-
choice of the performer, and a certain degree of parently envisaging a descent from 3rd to 1st posi-
variation was well within the remit of ‘beautiful’ tion within the slur.
performance in Beethoven’s era. Many, in this con- 97–101. Pno: For the slurred duplet 8th-note figures,
text, would surely have chosen a flexible, but basi- arpeggiating the first but not the second, which is
cally triplet rhythm.
111 Domenico Corri: A Select Collection of the Most Admired Songs,
110 “Die erste Note der Verzierung wird auf den Bassaccord ge- Duets etc. (Edinburgh, c. 1782), vol. 1, p. 8.
spielt.” 112 Louis Spohr: Violinschule (Vienna, [1833]), p. 121.

marked staccato, would accord with Czerny’s guide- but not legato. The 4th variation determined. The 5th
lines (5/c/ii). Speidel marks the first of each duplet variation is calm, measured, and in an old-fashioned
in both rh and lh with tenuto lines and the second style. The finale pretty lively ( . = 88) and cheerful.” 114
staccato.
101ii. Vl: Alard, Hermann mark a harmonic. Theme
101–105. Pno: All the chords could be gently arpeggiat- 1ff. Pno: Beethoven’s dolce invites both arpeggiation
ed, the chord in 105 lh fairly slowly, which would and asynchrony according to taste and expressive
enhance the decres. aims, particularly when the melody is in the piano
102ii. Vl: All except Joachim, Singer, Auer, Kreisler part (3/d/i). Overholding of the broken chords would
mark a harmonic. Singer adds espress. perhaps sug- be an aid to achieving dolce and creating resonance
gesting vibrato. (4/a/ii). Ganz gives a general con Pedale. The use of
103ii. Vl: A harmonic is marked by all except Her- a moderator if available would also help to achieve
mann, who may simply have omitted it in error. a dolce tone (PT: 3/b).
104i, 104ii. Vl: For 104i all either explicitly indicate 19i, 19iii.. Vl: As detailed in the Critical Report, the ap-
or tacitly imply a continuation of the fingering on poggiatura at the beginning of the bar is distinctly
103ii; and all except David, Joachim, Halir, Kreisler written as an 8th-note in Beethoven’s autograph
mark a harmonic on 104ii. and the 1st edition, but AG and all the 19th-century
editions print both this and the 16th-note appoggia-
Allegretto con Variazioni tura before iii as grace-notes. It seems clear that the
Tempo first is a long appoggiatura and should be played
Alla breve allegrettos are very rare in Beethoven’s out- expressively, approximately as an 8th-note (5/a/i).
put, but they seem generally to have functioned rather 22ii. Vl: A harmonic is marked by Alard, Hermann,
like allegrettos in 2/4, although with half-note values. Singer, Halir, Seybold; a same-finger shift is made
The longer, ‘heavier’ note values in alla breve move- to the harmonic by Hermann (4 – 4) and Singer (2–2);
ments nevertheless suggest a slightly slower tempo; Rosé also shifts from i–ii (2–3).
for slower tempos (adagio to andante) this effect seems 23, 31. Vl, Pno: The reverse dotted rhythms were al-
to have been minimal, but to increase proportionally most certainly not expected to be played strictly as
with the faster tempos.113 The presence of many tri- written; the accent was probably expected to fall on
plets in Variation 3 may also suggest a slightly slower the dotted note with the 16th played very rapidly on
speed than if there were only normal 8ths, although the beat (though, in practice, this often produces a
a discreet amount of tempo change between varia- pre-beat effect, probably very similar to the effect
tions would have been perfectly normal. The tempos that can be heard in the trio section of the minuet
suggested by the 19th-century editors seem broadly in Mozart’s String Quartet K. 421 as performed in
plausible; those of the Haslinger edition and Czerny the Klingler Quartet’s recording c. 1911).
are perhaps the most persuasive. 27. Pno: The first grace-note should be aligned with
Allegretto Allegro, ma non tanto the bass (as indicated by Ganz with dotted line no-
Haslinger  = 84 . = 92 tation) and could be held for as long as an 8th-note.
Moscheles  = 144 . = 92 The second grace-note should be played together
Czerny Vortrag  = 84 .=88 with the main note, aligned with the bass and im-
Alard/Diémer  = 144 – mediately released (5/a/iii).
Speidel/Singer  = 72 . = 88
Kreisler  = c. 80 Var. 1
Czerny comments: ”Alla breve, a rather fast allegro 33ff. Pno, Vl: Speidel/Singer suggest a slightly faster
[sic], but do not rush. The first variation is brilliant- tempo (Un poco più mosso) for this variation.
ly marked. The 2nd variation gently legato. The 3rd
variation very powerful and the bass with bravura, 114 “Alla breve, also ein ziemlich schnelles Allegro, doch nicht
übereilt. Die erste Variation brillant, markirt. Die 2te Variation sanft
legato. Die 3te Variation sehr kräftig und der Bass mit Bravour, aber
113 See Johann Abraham Peter Schulz: “Vortrag”, in: Johann Georg nicht legato. Die 4te Variation entschlossen. Die 5te Variation ruhig,
Sulzer (ed.): Allgemeine Theorie der schönen Künste, revised 2nd edi- gemessen, und alterthümlich. Das Finale ziemlich lebhaft ( . = 88)
tion (Leipzig, 1792– 4), vol. 4, pp. 707–709. und heiter.”

33ff. rh: Overholding of the slurred duplet 8ths would Var. 3
produce resonance mitigating the need for sus Pno, Vl: The intended dynamics in this variation are
taining-pedal (4/a/ii). However, Ganz gives exten- unclear. A p beginning might be assumed from the
sive sustaining pedal indications. end of the previous variation and the cresc. in 69, but
36. Vl: Singer marks leggiero, usually implying execu- no return to p is marked in the 1st- or 2nd-time bars or
tion with an elastic bow stroke in the upper middle elsewhere (a p in Pno in b. 75, which appears in previ-
of the bow. ous editions, stemmed from a misreading of Beetho-
II
41 ff. Vl: In order to play the long notes, violinists ven’s autograph – see Critical Report). 19th-century
who preferred to play the triplets in the upper half editions provide various dynamic schemes. A decresc.
of the bow required a strategy. David, who evident- to the opening dynamic in 72II seems plausible and
ly envisaged that style of bowing, simply marks a has generally been added; but Beethoven was evident-
change of bow in the middle of the long notes in ly happy to leave dynamic shading to the performers
12 and 15 (this was altered in David-revised). Sing- here.
er marks slurred staccato on 41IIviii–xii, 48Iviii–x, 65ff. Pno, Vl: Speidel/Singer mark this variation to be
which conveniently takes him towards the frog for played in the tempo of the opening theme (Tempo
the following long note; Auer marks this only on des Thema’s). Some or all the notes marked sf in rh
48Iviii–x. Alard, Brodsky apparently envisaged the might be played asynchronously, slightly after the
triplets in the lower half of the bow. The bowing bass which would give them special agogic empha-
given in Hermann, Joachim, David-revised, Halir, sis while also mitigating harshness.
Rosé, Seybold suggests the lower half in 41II, 48I, 65ff. lh: Beethoven may have expected legato except
and the upper half in 44, ending in  for the long in 76 and 80I (4/a/ii). Speidel and Halir mark slur-
note in 45f. ring patterns throughout. Ganz marks poco legato
43, 47, 48II. rh: Beethoven presumably expected legato perhaps suggesting a non-legato touch which is also
as indicated by Reinecke, Diémer (though Diémer plausible.
and Halir mark staccato in 48II). In 43 Speidel, Ha- 67. rh: The trill should probably be from the main note
lir adopt Beethoven’s articulation in Vl 40II. In 47 because it is preceded by the note above; Speidel
Speidel marks a slur in rh i–xi with staccato on xi and Reinecke give a main-note-start fingering (3– 4).
and xii. 71. Pno, Vl: The trills from the main note as in 67. Rei-
46ii. Vl: A harmonic is marked by Alard, Singer, Her- necke’s fingering indicates a main-note start.
mann. 75. Pno: Speidel writes out the trill in a footnote, start-
ing on the main note and ending with a turn thus:
### œ œ œ œ œ # œ œ
Var. 2 3 1
49ff. Pno, Vl: Speidel/Singer suggest a slower tempo &

marking molto tranquillo. The 8th-notes might be
played unequally according to taste and expressive but in the light of Czerny’s practice, Beethoven would
aims (2/b). surely have expected an upper-auxiliary start.
49–53. Vl: Some indicate 1st position throughout for 79. Pno: Here the trill will certainly have started on
these bars (David), or all 1st position apart from the main note, but its conclusion is not indicated.
49iii–v in 3rd position, returning via the open A- Reinecke notates it thus:
string on vi (Brodsky, Halir, Seybold); others retain Ÿ j. œ
the D-string either until 52vii (Joachim) or 53vii
### œ # œœ
&
(Alard, Singer, Auer, Rosé).
50ii. Vl: A harmonic is marked by Alard, Singer. and Speidel gives it in a footnote thus:
51ii–iv. Vl: Singer uses the 3rd finger for all three œ œ . œ
### œ œ œ# œ œ
3 3
notes with a harmonic on iii.

&
57ff. rh: All chords might be arpeggiated swiftly but 1
gently (5/c/ii). Speidel marks rh portato.

61viii. Vl: A harmonic is marked by Alard, Hermann, Var. 4
Singer, Brodsky. 81ff. Vl: All previous editions, including BW, failed
to see that Beethoven deleted staccato marks on the

chords in his autograph. They should probably be 131. Vl: The 16ths were almost certainly expected to
arpeggiated tightly but with the top notes sustained be played somewhat shorter than their written val-
for full value. Singer marks all . ue and very lightly. Hermann marks leggiero.
82ff. Pno: Chords could be arpeggiated in varying 132i–ii, 134i–ii, 136i–ii. Vl: The falling fifth is fingered
speeds, though generally fairly tightly. Asynchrony by all editors with a shift down a single string,
could be used effectively in 96I 96II (5/c/ii and PT: which in the practice of the period would not have
1/a and 1/b). Since Beethoven deleted staccato on the been executed inaudibly. Beethoven will surely have
chords in Vl (see 81ff above), it seems likely that the been aware of this. All shift 2nd to 1st in 36, which is
staccato in Pno was an oversight and these could in any case logical and in 38, 40 some finger it with
therefore also be arpeggiated (5/c/ii). the shortest possible shift of a single position, 3rd
Successions of 8th- and 16th-notes not marked stac- to 2nd (Alard, Hermann, Kreisler). All the others,
cato might be played in a lilting fashion slightly certainly aware of the portamento implications, shift
unequally (2/b). two positions, 3rd to 1st, in 38, 40.
137. Vl: David marks  on 137i and an additional slur
Var. 5 over all, strongly suggesting that he began the dot-
Speidel/Singer, perhaps influenced by Czerny, mark ted figure in the upper half of the bow. This addi-
this variation “Calm and even”.115 tional slur was removed in David-revised, suggest-
97, 113. Pno: Speidel designates the bass poco marcato. ing performance of the figure in the middle or low-
99. Pno: The trill was most certainly expected to begin er half of the bow, obviously broadening the stroke
with the main note. towards the frog during the crescendo to take the
105ff. Pno: From 105–112 the contrapuntal nature of the longer slur  in the following bars. David, Singer
writing might preclude the use of arpeggiation and shift up the A-string, all others remain in 1st position.
asynchrony, as advised by Czerny (5/c/ii). But from 137v–vi. Vl: In the violin part in the score of Singer’s
113 onwards, where the character of the writing edition these are marked 0– o4 , presumably to begin
changes, asynchrony and arpeggiation would bring the next bar as marked, with 3 on the appoggiatura.
much expression particularly to the chord in both In the separate violin part, however, the fingering
hands from 125 onwards. For the chords marked sf in 137 is absent, presumably in error, since the 3 in
in 126 and 128 a swift arpeggiation would mitigate 138 is still there.
any harshness of tone (PT: 1/a/v). The chord in 97, 138. Vl: The appoggiatura is surely meant to be played
marked arpeggio by Beethoven, could be arpeggiat- with similar rhythm to the 16ths in the previous bar.
ed fairly slowly and with poignant placement of in- 140, 142. Vl: The majority of editors shift up the E-
dividual notes, or it might even be arpeggiated up- string. Only David remains in position, crossing from
wards and downwards in an improvised fashion as A- to E-string; this was changed in David-revised.
indicated in some sources.116 144i. Vl: The open E-string is marked by Alard, Her-
106–111. Vl: Alard, Joachim, Singer, Auer, Kreisler re- mann, Halir, Seybold.
main entirely on the G-string. Fingering is absent in 145–151. Pno: Beethoven’s gradual decrescendo to ppp
David, but David-revised adds the G-string finger- suggests the use of the shift pedal (if available) mov-
ing in 10–12, which is also used by Hermann, Halir, ing from tre corde to due corde to una corda (PT:
Seybold. 3/a). Speidel marks poco rit. in 151.
115, 123. Pno: For the performance of the grace-notes
see note to 27 above. In 115 Ganz marks a dotted Var. 6 Allegro, ma non tanto
line from the first grace-note to the first note in Vl. 151–155. Pno: Beethoven’s dolce invites the use of asyn-
127–147. Pno, Vl: For the execution of reverse dotted chrony and arpeggiation (particularly useful for sf ),
figures see note to 23 in the Theme. as well as a tone modifier such as a moderator (if
128ii–iii. Vl: A 3–3 fingering, with clear portamento available), perhaps applied only within softer dy-
implications is given by all except David, Hermann; namics not for crescendo or sf (3/d/i; PT: 3/b).
it was added in David-revised. 154v, 158i, iv, 161v, 166i, iv. Pno, Vl: Speidel/Singer il-
lustrate the execution of the trills in footnotes to the
115 “Ruhig u. gleichmässig” score and violin part, all starting on the note and
116 see Peres Da Costa: Off the Record, pp. 115–117. concluding with a turn. Reinecke/Hermann include

a turn in Pno and Vl. All seem to assume a main- 189–195, 205–217. Pno: Overholding would be essen-
note start; in the light of Czerny’s practice, however, tial to creating special resonance. Ganz gives sug-
it seems almost certain that Beethoven envisaged gestions for sustaining-pedal use.
an upper-auxiliary beginning, probably rendering 190–194. Vl: Singer marks harmonics in 190f and 1 in
the trills essentially as four-note turns. 194.
159–163. Vl: Some editors envisaged the theme en- 200. Pno: Beethoven’s pp for this repeated material
tirely in 1st position, or only changed to 2nd for the might well be achieved with the use of the una corda
trill in 162; others retained the D-string throughout pedal (PT: 3/a).
(Alard, Joachim, Auer, Rosé, Kreisler). 199–205. Vl: Singer remains on the D-string through-
159ff. rh: Presumably Beethoven intended the 16ths out with a harmonic in 205 and 1 in 206; Alard, Her-
to be legato in which case as much overholding as mann, Seybold remain on the D-string until 203iii
possible would mitigate the need for sustaining- with a harmonic on 202i.
pedal (4/a/ii). Speidel, Diémer, Halir add slurs. 220ff. Pno: In 220, the grace-note might be aligned
160. Pno: Presumably Beethoven intended Pno to be with the beat and realised as a 16th-note (5/a/i). From
dolce to match the Vl. Halir marks a > for the here to the end, asynchrony would work particu-
last two 8th-note beats of 159 and p at 160. larly well on the first of the slurred duplet 8th-notes
164iv. Vl: Beethoven surely did not expect the staccato in 224 etc. (5/c/ii; PT: 1/b). Ganz gives many sugges-
mark significantly to shorten the quarter-note. Da- tions for sustaining-pedal use.
vid, Halir, Rosé, Seybold mark it with a tenuto line; 222f. Vl: Singer marks 1 on 222i, o/3 on 223i and 4 on
Auer slurs smoothly from 164iv to v. 223iii.
166, 176, 180, 197, 219 etc. Vl: Editors divide between 222ff. lh: Presumably Beethoven intended the 16ths to
hooking ii and vi into the slur (David, Joachim, Auer, be legato as marked by Reinecke, Speidel, Diémer.
Brodsky, Halir, Rosé, Seybold, Kreisler) or taking 226f. Vl: Singer, Halir, Kreisler mark 3 on 226iii, a har-
them with a separate bow (Alard, Hermann, Singer). monic on 226vi, and 3 on the A-string on 227i.
The latter execution has the potential to bring the 234i, iii. Vl: David, Auer, Halir, Rosé, Seybold mark
bow somewhat off the string and to be executed more tenuto lines. Singer removes the staccato marks.
towards the frog. The use of one or the other method 234, 235. rh: Beethoven presumably expected legato
seems to have been somewhat contentious; both with overholding (4/a/ii). Reinecke, Speidel, Diémer
David-revised and Joachim-revised reverse some of mark slurs. Halir marks a combination of slurs and
the bowings of the originals. In 75ff, for instance, staccato.
everything is hooked in David, but all these bow-
ings were removed in the revision; in Joachim-re-
vised, some hooked bowings are added where there Opus 30, No. 2
were none in the original. These changes suggest Allegro con brio
that a greater use of the lower half of the bow and Tempo
of bowstrokes more characteristic of the 20th cen- Among movements to which Beethoven himself gave
tury, increased during the later 19th century. metronome marks, the one most closely analogous
168, 172, 174. rh: The slurred duplet 16ths could be to this is the Allegro con brio first movement of the
played slightly unequally (2/a). String Quartet op. 95, to which he gave  = 92, but in
169–171, 175–182. rh: Beethoven most likely assumed the C minor Sonata there are more complicated bow-
legato as marked by Reinecke, Speidel, Diémer, Ha- ing patterns on some of the 16ths, which may suggest
lir (4/a/ii). In 176ff Ganz gives suggestions for sus- that he would have allotted it a somewhat steadier
taining-pedal use. tempo, closer to those he gave for straightforward Al-
177. Vl: Singer marks leggiero. legros in  metre. It seems unlikely that Beethoven im-
181i–ii. Vl: Alard, Singer shift to a harmonic with 4th agined a tempo much less than the one suggested by
finger. Moscheles. Czerny’s comment (below) that “the tempo
183ff. Pno: Asynchrony would be very effective for should not be taken too quickly”, probably represents
important notes for example 184i and 185i (5/c/ii and a later interpretation of the movement, reflecting
PT: 1/b). Ganz gives suggestions for sustaining-pedal his general statement about performing Beethoven’s
usage. music, that “even the spiritual conception has acquired

a different validity through the changed taste of the poser’s music to life and to manifest an artistically
times”.117 ‘beautiful’ interpretation. Several of the early editors
Haslinger  = 152 (including Reinecke, Speidel, Halir and Ganz) added
Moscheles  = 88 marks of expression (including articulation, dynam-
Czerny Vortrag  = 144 ics and accents, and pedalling). While their additions
Alard/Diémer  = 132 undoubtedly reflect later 19th-century attitudes, these
Speidel/Singer  = 144 preserve practices that would surely have been heard
Kreisler  = c. 126 in the first half of the 19th century.
Czerny, having just discussed op. 30 no. 3 (he desig- 1–5, 125–138. Pno: Czerny’s description suggests p for
nates that sonata no. 2 and the C minor no. 3) wrote: the initial dynamic, which all the editors give. Slight
“This sonata too is among his greatest, and all the asynchrony for 1i, 3i, and 5i would add to the dra-
seriousness that dominates it must be grasped by the matic impact. At 125, very swift arpeggiation of the
player. The quiet but significant beginning increases chord on i would help to give it the energy it needs
to full power in the 23rd bar, whereupon the march- (5/c/ii; PT: 1/a). In 131 and 133, the < > invite both
like middle movement enters in E flat major pp, in agogic nuance on ii and perhaps slower, poignant
which the 8th-notes are played as staccato as possi- arpeggiation (3/b/v). All other chords could be swift-
ble, but they have swell later at the entrance of the ly arpeggiated (5/c/ii) (see below 137–138).
bass and the rising of the melody. In the later passage, 6–8. Pno: As explained by Louis Adam in his Méthode
the fingering is as follows: [Czerny’s example is given de piano (1804), for notes marked portato as in 6, a
below. See note to 62f]. At the end of the whole move- little delay of the upper note (i.e. asynchrony) adds
ment (in the last 19 bars), the octaves must be played “not a little” to the expression (PT: 1/b). Pianists of
in an extremely legato and surging manner, the whole Beethoven’s era would most likely have played the
passage must always increase, and finally the last 10 notes in this bar unequally to emphasise the chro-
bars with the greatest force, using the pedal, to end matic movement, perhaps also, reacting to the cresc.,
with a stormy conclusion. The character of this piece with a slightly forward momentum. Halir marks a
is military, and the tempo should not be taken too tenuto line on rh i. In 7 and 8, swift arpeggiation
quickly, for the movement must always be magnifi- of the chords would almost certainly have been ex-
cent, heroic, just as the passages are calculated for pected (5/c/ii). Given the rhetorical character of the
bravura and brilliant clarity.” 118 music, the chord in 9 might be elongated slightly.
Beethoven’s markings (meticulous as they may ap- Speidel marks ten.
pear) represent only the indications of expression that 9. Vl: All editors before Rosé start in 3rd position and
he absolutely required. Musicians of his era and the remain there until 12. Rosé goes back to 1 for 10i;
later 19th century understood that much more needed Kreisler starts in 4th position, perhaps for the sake
to be added in the way of expression to bring the com- of vibrato with the 3rd finger on the first note.
9–15, 131–145. Pno: Beethoven probably expected legato
117 Czerny: Die Kunst des Vortrags, p. 34. “selbst die geistige Auf- with overholding wherever possible (4/a/ii). Reine
fassung erhält durch den veränderten Zeitgeschmack eine and’re
cke, Speidel, Halir mark slurs. In addition, Speidel
Geltung”.
118 “Auch diese Sonate gehört unter seine grössten, und muss marks sustaining-pedal through each bar including
von dem Spieler mit all dem Ernste aufgefasst werden, der in ihr in 16.
herrscht. Der ruhige, aber bedeutende Anfang steigert sich bis zur
vollen Kraft im 23sten Takt, worauf der marsch-artige Mittelsatz in
17–22, 147–152. Pno: Swift arpeggiation of all chords
Es-dur pp eintritt, in welchem die Achtelnoten möglichst staccato would add much to the drama, especially for the sf
zu spielen sind, aber später beim Eintritt des Basses und beim Stei- in 19 (5/c/ii, PT: 1/a). In 20 (150), along with Vl, the
gen der Melodie anschwellen müssen. In der spätern Passage ist
der Fingersatz wie folgt zu nehmen: [Czerny’s example is given
8th-notes might be played unequally. In 22 (152), the
below. See note to 62f]. Am Schlusse des ganzen Tonstückes (in trill, preceded by the note above, was probably ex-
den letzten 19 Takten) müssen die Octaven äusserst legato und an- pected to start on the main note (5/b/ii).
schwellend gespielt werden, die ganze Passage muss sich immer
steigern, und endlich die letzten 10 Takte mit aller Kraft, mit dem
19i–ii, 149i–ii. Vl: Alard, David, Hermann, Joachim,
Pedal, und stürmisch das Ganze beschliessen. Der Character dieses Halir, Seybold, Kreisler remain in 1st position; Singer,
Tonstückes ist militärisch, und das Tempo darf nicht allzuschnell Auer, Brodsky shift 1–1 and remain on the G-string
genommen werden, denn die Bewegung muss immer grossartig,
heroisch sein, so wie auch die Passagen auf Bravour und brillante until 23i. Singer’s portamento intention here is made
Deutlichkeit berechnet sind.” clear by a slanting line.

23, 27–28, 72–73, 153–154, 157–161, 205–207, 218, 220, this passage is discussed by Moser in volume 3 of
253, 254. Pno: According to Czerny in his Pianoforte- the Joachim and Moser Violinschule, where he sug-
Schule, “all chords consisting of very short notes” gests: “In Beethoven’s C minor sonata, the performer
should unless expressly marked by the composer can choose between martelé and springing bow-
be unarpeggiated.119 But this might not have pre- strokes at the relevant passage; with both, the com-
cluded extremely swift arpeggiation, the type that poser’s intentions can be equally well realized.” 120
Thalberg in L’Art du chant (1853) described as presque (He illustrates bb. 38f.) Whether Joachim played this
plaqué, or almost together (5/c/ii). passage in both ways on different occasions is un-
25–26, 154 –156. Pno: All chords might be swiftly but clear. Moser’s posthumous revisions to Joachim’s edi-
gently arpeggiated (5/c/ii). The trill in 25 (155) was tion may be calculated to facilitate a spiccato bowing,
surely expected to start on the upper auxiliary (5/b/ whereas the markings in the original apparently
ii), though Reinecke’s and Speidel’s fingering indi- keep the bow in the upper half. The sautillé/spiccato
cate a main note start. The < > in 26 (156) option seems likely to have been a later, French-in-
suggests an agogic accent at the apex perhaps with spired approach, which would have been unfamil-
asynchrony between the rh and lh (3/b/v). iar to Beethoven; he may more likely have imagined
28ii–35, 99ff, 160ff, 223ff. Vl: The dotted figures are martelé since he had played with Kreutzer, but he
predominantly executed with hooked bowing; some may also, simply have expected a detached, but not
editors indicating occasional separate bows, but Jo necessarily sharply-detached bowstroke.
achim and Kreisler mark no hooked bowing. There 36–37. lh: Speidel, Ganz indicate poco marcato.
is evidence of two entirely different bowing styles. 39, 172–173. Pno: The grace-notes played as acciacca-
Alard, probably representing a specifically French turas, simultaneously with the main note and re-
practice, instructs “du talon de l’archet” (at the frog leased quickly (5/a/iii), as explained in Junghanss’
of the bow) in 28. Violinists in Austro-German tra- Pianoforte-Schule.121
ditions seem to have favoured the upper half of the 43, 45, 176, 178. rh: Overholding and arpeggiation
bow. Singer certainly bows near the point, because would add much to the character, particularly the
he begins   in 28, marking Sp[itze] (point) above sudden p in 43.
28iii, and also takes 33iv-vii  , but at 34ii after the 46, 48, 179, 181. rh: Beethoven may have expected
up-bow dotted half-note, he marks “Mitte” (middle). legato (4/a/ii). Reinecke, Halir mark slurs.
David, Hermann, Halir, Seybold take every dotted 47, 49, 180, 182. Pno: The notes marked sf could be
figure with hooked bowing, beginning  in 28; Halir given special emphasis with asynchrony, playing
specifically marks Sp[itze] in 28, and all four surely them very slightly after the bass.
expected the use of the upper part of the bow with 47–53. Vl: David’s bowing (changed in David-revised)
a relaxed, but well-separated stroke, entirely from shows unmistakably that he executed the 16ths to-
the elbow with a loose wrist. Joachim and Kreisler wards the point of the bow, for he marked the first
begin  on 28ii, evidently at the point of the bow two notes with dots under a slur, thus taking the
(Joachim-revised, however, has many hooked bow- third note ; the ‘backwards’ bowing continues un-
ings). Auer’s and Brodsky’s bowing is certainly cal- til 53viii, correcting itself through the omission of
culated to remain close to the point. The dotted fig- the slur on 53v–vi. This kind of bowing allows a
ures are similarly treated in later occurrences of this very broad and powerful stroke in the upper half of
material. the bow, corresponding with the type described in
36– 42. Vl: The division between Alard and the others the Paris Conservatoire 1803 Méthode and its deriva-
is also apparent here; according to his own general tives, and in Spohr’s Violinschule.
instructions at the beginning of the edition, Alard in- 54 – 56, 185–189. Pno: To delineate between the over-
tends the staccato marks to indicate a ‘sautillé’ (spring-
ing) bowstroke in the middle of the bow. All the
120 Joseph Joachim and Andreas Moser: Violinschule (Berlin, 1905),
others except Singer, who marked “Mitte” in 34, ap- vol. 3, p. 12. “In der c moll-Sonate von Beethoven kann der Vor-
parently envisaged a detached bowing in the upper tragende an der betreffenden Stelle zwischen hämmernden und
half of the bow. The option of martelé or spiccato for springenden Strichen wählen; mit beiden lassen sich die Absichten
des Komponisten gleich gut verwirklichen.”
121 Johann Christian Gottlieb Junghanss: Theoretisch-practische
119 Czerny: Pianoforte-Schule op. 500 (1839), p. 55. Pianoforte-Schule [new edition] (Vienna, 1823), p. 26.

lapping textures, playing rh slightly after lh at ably reflecting an earlier tradition, have the fewest
54 rh iv, and so on, would be effective (5/c/ii; PT: shifts, with Auer, Brodsky, Kreisler the next fewest.
1/a/iv). 84. Vl: The fingering in the edited violin part of the
58–59, 1919–192. Pno: A slight arpeggiation would present edition presupposes that the 1st finger slides
render the octaves in lh more powerful than play- from a #2 to c #2 and remains there throughout the bar.
ing the notes together (5/c/ii). The trill in 58 (191) rh 97–98. rh: Given the martial character of the music,
should certainly start from the upper auxiliary (5/b/ the grace-notes should probably be played as ac-
ii), though Reinecke, Speidel, Ganz give fingering ciaccaturas (5/a/iii).
for a main note start. 99–102. Pno: Probably expected to be played with sus
60–61, 193–194. Pno: Beethoven likely expected legato taining pedal which Ganz gives every half bar.
with overholding in lh (4/a/ii), indicated with slurs 99ff. Vl: See note to 28ii–35.
by only one of the early editors – Speidel –, though 113–124. Pno: Beethoven notated a slur only in 113,
Ganz gives sustaining-pedal indications. The ex- but undoubtedly intended legato to continue (4/a/
pressive quality of the rh melody invites frequent ii). Reinecke gives slurring patterns from 113–118,
asynchrony and agogic nuance. Ganz marks espres- but not after, perhaps an oversight. Speidel marks a
sivo and < > in 61 v–x, while Halir marks slur from 113–114, but afterwards staccato dots in 115
tenuto lines in 60 rhi and v and 61 rhi and iii. The on i, v, ix, and xiii in rh and lh and presumably to
trill in 61 (194) was likely expected to start on the continue. Ganz slurs 114 –115 and 117–118, with inter-
upper auxiliary note and have a turned ending mittent sustaining pedal marks throughout.
as marked by Halir, Ganz (5/b/ii), though Speidel, 114i. Vl: David, Alard, Brodsky mark  for the sf; Jo
Ganz, Vogrich give main note start fingerings. achim, David-revised, Auer, Halir, Seybold mark ,
62–64, 195–197. rh: Apart from notes marked staccato, probably envisaging a fouetté bowing. Singer also
Beethoven probably expected legato (4/a/ii), as mark- marks , but the sf is absent, presumably resulting
ed by Reinecke, Speidel. Ganz however marks non from a printer’s error, since in the score it appears
legato in 62 (195). erroneously in the piano part.
68–71, 201–204. Pno: A slight arpeggiation would ren- 129. Pno, Vl: Speidel/Singer, Auer mark ’ at the end
der the octaves in lh energetic and thus powerful of the bar.
(see 58–59 above), especially in 69 (202) for sf. Spei- 131f. Vl: Singer offers a curious fingering, 1–o–1, the
del marks a tenuto line and staccato on i and ii in harmonic at the peak of <> , where perhaps oth-
lh 68 and 70. In 69 and 70, rh was likely expected er violinists would have used an expressive vibrato.
to be legato as marked by Reinecke, Speidel, Ha- 137i–138ii. Vl: All except Brodsky use the same fin-
lir. In 71 (204), swift arpeggiation of lh would pro- ger for the a 2 and g2, all mark 4 – 4 except Kreisler
duce an energised effect without harshness the sf (3–3) and Singer. The latter, who uses 1–1 (on the E-
(PT: 1/a). string), marks the passage appassionato e largamente,
75–91, 208–217. Pno: The rh broken chords were prob- adds a portamento line between 138i–ii and a ’ after
ably expected to be legato (as marked by Speidel 138iii before taking 139i with 4 on the A-string.
with slurs in 75ff and sempre legato in 208ff, and Ha- 137–138. Pno: Speidel marks appassionato e largamente,
lir with legato) and overheld (4/a/ii). Given the fiery > in 137 rh and lh ii, and a ’ after 138iii. Ganz marks
nature of the writing, Beethoven would surely have <> on 137ii.
expected the use of the sustaining-pedal, in which 149i–ii. Vl: See note to 19i–ii.
respect Ganz’s indications could be helpful. 160ff. Vl: See note to 28ii–35.
76–91. Vl: At 76 Singer marks molto espressivo, Auer 209. Vl: Singer again marks molto espressivo and Auer
espress. They, along with most other editors, employ espress.
portamento where possible in this passage. Alard in- 209–215. Vl: All except David mark this passage to al-
dicates A-string for the first four bars. Singer, Auer, ternate between 1st and 3rd position on the A-string.
Brodsky, Rosé, Seybold, Kreisler go up the A-string David marks no shifts, perhaps envisaging the pas-
on 78iv and shift down for 79ii. Alard, Singer, Halir, sage in 1st position with an extended 4th finger for
Rosé, Seybold shift under the slur between 84ii–iii the f 2; the shifts were added in David-revised. Either
and iii–iv, and all include a shift between 85i–ii. This approach would have been plausible for an early-
pattern is repeated in 90f. David, Joachim, prob- 19th-century violinist.

216. Pno, Vl: Speidel/Singer mark con fuoco, followed precisely brought out. The entry into the theme (from
by Tempo 1 at 221vi. the 49th bar) is to be performed with the utmost atten-
229–230. Pno: All chords could be swiftly arpeggiated. tion, very legato, and somewhat held back. The runs
230–236. Pno: Beethoven may have expected legato in starting from the 60th bar with the utmost lightness
lh, but unusually none of the editors mark slurs (4/a/ and clarity, although at first legato and pianissimo,
ii). Various of the notes in the rh might be played and not sluggish, since during them the violin plays
asynchronously with the lh (5/c/ii). In 234 –236, the the theme. The conclusion must float by quietly and
chords in rh might all be swiftly arpeggiated, which leggierissimamente.” 122
would give them energy. 0, 8. Pno, Vl: Speidel/Singer mark molto espressivo;
235. Vl: Rosé replaces each staccato dot with >. Halir dolce espressivo.
236. lh: Speidel marks leggiero. Ganz marks un poco 0i–ii, 2i–ii, 8i–ii etc. Pno, Vl: The upbeat quarter-notes
agitato. marked with dots under a slur by Beethoven are cer-
248, 250, 251. Vl: The sf can be very effectively achieved tainly to be played portato, with all the expressive
by playing the 16th-note upbeat with a fouetté at- characteristics that this implies (4/b). Halir marks
tack, very short and late, almost on the beat, like a them consistently with tenuto lines over or under
tightly broken octave. dots, and under a slur.
252. lh: Ganz adds that he (the Editor) “plays unbro- 0–15, 20–23, 28–32, 52–60, 88–91. Pno: 19th-century
ken octaves in the left hand”. sources make it clear that a passage such as this
within a cantabile context requires frequent arpeg
Adagio cantabile giation (5/c/ii). Pianists would have applied arpeg
Tempo giation in varying degrees according to taste, some
The only relevant adagio in C for which Beethoven on all chords, others generally on all main beats,
provided a metronome mark is the introductory Poco perhaps not on anacrusis figures such as 0, the sec-
adagio in the first movement of the String Quartet ond half of 2 etc. But, here the portato notation on
op. 74, for which he gave  = 60. In that movement the several anacrusis figures in rh (2, 4, 8, 10, 12) must
fastest notes are 16ths, but in this sonata there is a also be considered in light of Moscheles’ advice in
section of 32nd-note runs; however as Czerny’s com- his Studies for the Piano Forte op. 70, Bk 1 (1827) that
ments (below) indicate, this is more in the nature of double-notes and chords “should be struck very
an ornamental accompaniment to the principal theme slightly in the Arpeggio manner, giving them the
and therefore perhaps not particularly relevant to the same length of time as a dot under a slur requires.”
overall tempo. A tempo somewhere in the region In these cases, this might indicate that rh is arpeg
of those given in Haslinger, Czerny, and Moscheles giated quickly and gently while the lh remains un-
seems plausible. The other metronome marks suggest arpeggiated. In 3 the sf might be given a special en-
a conflicting later 19th-century conception of the mu- ergy without sounding hard by playing the bottom
sic, which seems to misunderstand the implications note in lh with the beat while the other notes in lh
of the time signature, which both editions give as , and rh are played slight afterwards and together, a
as do several other 19th-century editions (but not AG, practice that would create a slight agogic accent (PT:
David, nor Joachim). 1/a/v). Resolutions such as 4ii etc. would probably
Haslinger  = 66 have been unarpeggiated.
Moscheles  = 63;
Czerny Vortrag  = 60
Alard/Diémer  = 96 122 “Da das Tempo alla breve ist, so wird dieses Adagio als ein
mässiges Andante vorgetragen, aber mit all dem gefühlvollen Aus-
Speidel/Singer  = 48 druck, zu dem die so edle und schöne Melodie jeden bessern Spie-
Kreisler  = c. 52 ler ohnehin schon begeistern muss. Die Sechzehnteln (vom 33sten
Czerny explains: “Since the tempo is alla breve, this Takt) müssen sehr delikat und klar abgestossen und dabei jedes
cresc: genau hervorgehoben werden. Der Eintritt in das Thema
adagio will be performed as a moderate andante, but (vom 49sten Takt an) ist mit grösster Aufmerksamkeit, sehr legato,
with all the soulful expression that such a noble and und etwas ritardiert vorzutragen. Die vom 60sten Takte anfangen-
beautiful melody must anyway inspire in any more ac- den Läufe mit möglichster Leichtigkeit und Deutlichkeit, obwohl
anfangs legato und pianissimo, und ja nicht schleppend, da während
complished player. The 16ths (from the 33rd bar) must dem die Violine das Thema vorträgt. Der Schluss muss ganz leise
be very delicate and clearly articulated, with each cresc: und leggierissimamente verschweben.”

7, 15, 59, 75, 67, 83, 85. Pno, Vl: For the turn, Speidel/ 17. Vl: Singer adds espressivo; Halir, Rosé mark <> on i.
Singer give a footnote: 19i–ii. Vl: All except David indicate a shift 1–2; it was
bb added in David-revised and may have been an in-
& b b œ œ œ œnœ œ œ œ ,
5
advertent omission in the original; it would certain

ly have been typical of David and in David’s per-
which is probably close to what early-19th-century sonal copy of the sonatas 2 had been added on ii,
musicians would have played. But musicians of though not in his hand.
Beethoven’s time, perhaps including the composer, 19vii–viii. Vl: Singer shifts here (1–2).
may well have used a more elaborate ornament on 23, 31, 75, 83, 92. Pno, Vl: The trill in this figure could
some later occurrences of this figure. The appog- be realised in many ways with either upper or lower
giatura before note v, written by Beethoven as an auxiliary note starts. Skilled musicians would prob-
8th-note, may have been intended as a short appog- ably not have realised it in quite the same way on
giatura, conventionally notated as an 8th-note with every occurrence. Potential treatments might include
a slanting line through the stem, as marked in AG; the following:
b b œ œŸ œ œ. œ
but it might equally have been interpreted as a long r
appoggiatura (5/a/i and 5/a/ii) and, on later occur- b
& b œ
rences, have suggested a somewhat more elaborate
b b œ Ÿ œœ
ornament, perhaps with a mordent (Schneller), a r
three note turn from below, or even a trill on v. & b b œ œ œ œ. œ

8–15. Vl: Violinists suggest several expressive strate-
Ÿ
gies. All mark the opening on the D-string, all start- bbbb œ œ œ œ. œ
& œ œœ
ing 4 except Kreisler, who, evidently for the sake
of vibrato, begins with 3; David, Halir, Rosé shift
to 2 on 9v, allowing them to make a delicate vo- 24 –28, 36. Pno: The 16th-notes in 24 could be played
cal-style portamento between 10i–ii (3–1), this was slightly unequally or liltingly (2/b). From the second
changed in David-revised to a 1–2 shift between 9v– half of 24 to the first half of 28 as much over hold-
vii, which is also used by Joachim and others. ing as possible (4/a/ii) and with occasional agogics
Alard’s fingering allows a more pronounced shift on important notes or beats to break up the regu-
between 10i–ii (2–1). larity.
A fast light bowstroke, employing the whole length 28. Vl: In the autograph, the slur extends to iv only,
of the bow for each stroke, can produce a beauti- and the portato marking is omitted; the 1st edition,
ful transparent quality of tone. A slower, weightier however, has the present slurring and articulation.
bowstroke, made predominantly in the middle third A separate bowstroke could be executed as if it were
of the bow, will elicit a more intensely expressive portato.
sound. These effects might both be used in this 31f. Vl: Alard marks this passage to stay entirely on
movement on different repetitions of the theme. the D-string.
11vii–viii. Vl: Alard, David, Singer, Brodsky, Halir, 32. Pno: The < > invites both agogic lengthening
Rosé, Seybold employ a portamento shift 3–3; the at the apex and arpeggiation or asynchrony (3/b/v).
others remain in position. 33i–ii, 35i–ii. Vl: Singer, Auer (33 only) take both notes
13ii–iii. Vl: All shift 1–2 between bowstrokes. Some, with 2nd finger and Singer marks a portamento line.
perhaps all, will have made an audible connection, This expressive fingering is a very plausible 19th-
more or less obvious, for the sake of legato. century effect, typical of manuscript alterations in
16–20, 68–72. Pno: In 16 rh, the 16th-notes marked porta- David’s personal copies, which many editors might
to might be played unequally which would increase have used, but were probably reluctant to mark on
their expressive quality, a practice heard on many the basis that it could encourage tasteless excess. It
early recordings of 19th-century musicians (2b). From is given below the stave in the edited violin part.
the half bar of 16 (68), all main notes and others 33– 40. Vl: All editors divide Beethoven’s long slurs.
deemed important in the rh melody might be played 36iii–iv. Vl: If the previous notes have been played in
asynchronously slightly after the lh (5/c/ii; PT: 1/b). 6th position, it would be very effective to take the
At 68 Speidel marks cantabile. repeated e s with a completely smooth bowstroke,

producing the articulation merely by the substitu- Czerny, in his 1820s arrangements of op. 47 for piano
tion of the 2nd for the 1st finger.123 and piano duet, however, indicated the use of up-
41– 48. Pno: All chords in rh might be gently arpeg per appoggiaturas and turns, where not supplied
giated at a moderately fast speed (5/c/ii and PT: 1/a). by the composer, in some rising chains of trills,
At 41 Speidel gives cantabile and marks all rh chords in contrast to his teaching in his Pianoforte-Schule
with tenuto lines. Ganz marks these chords ten. op. 500 (1839). (See commentary on op. 47 below, es-
41– 46. Vl: Most mark  for the beginning of each group pecially the note to 164 –167, 183–185 of the second
of 16ths, which would certainly have been played movement.)
near the point of the bow. David and Hermann mark 71. Pno, Vl: The slurred duplet 16ths might be played
, clearly indicating execution in the upper half of with the first note longer and stronger than the sec-
the bow. Alard, with his very slow tempo, marks ond (2/a).
slurred staccato. Halir too employs slurred articula- 74. Pno: The chord marked sf could either be arpeggiat-
tion but changes the dots to –. . ed normally, or by playing the bottom note in lh
48–52. Pno: As much overholding as possible would with the beat while the other notes in lh and rh are
be appropriate throughout. From 50 the 16th-notes played slightly afterwards and together, a practice
could be played in lilting fashion, unequally with that would create an agogic accent (PT: 1/a/v).
agogic nuance on important notes (2/b). From bar 52 76iv–viii. Vl: Slurred articulation is added as on 60iv–
the tempo might be somewhat held back (presum- viii, but this time with  expected.
ably very slightly slower or steady) until the entry 76–78. Pno: Beethoven most likely expected legato and
of the theme half way through 52. Speidel/Singer overholding for the broken chords (4/a/ii), which
mark calando (decresc.) with a tempo at the half bar. Reinecke, Speidel mark with slurs, and Ganz marks
60ii–iii. Vl: All except Brodsky mark a shift 2– 4. armonioso.
60iv–viii. Vl: All except Singer, Brodsky give slurred 78iv–79i. Vl: See note to 70iv–71i.
articulation (Auer only iv–vi, and Halir with _ un- 78–79. Pno: The lh octaves might be swiftly but gently
der the slur) and mark, or expect  on vi. arpeggiated which would help achieve a legato sing-
60–68, 80–83, 107–112. Pno: As Czerny advised, the ing style (5/c/ii).
32nd-notes played legato but with lightness and clar- 80f. Vl: All the editors go to 3rd position, as in other
ity (perhaps not overheld), and not sluggish. In 60 statements of this theme, either in 80 or on 81i. For
and 80, Beethoven’s sempre leggieremente may imply variety’s sake, however, and to create a calmer feel-
something less smooth than a real legato, akin to ing, an early-19th-century violinist might well have
non legato. In 107 Beethoven did not specify sempre chosen to remain in 1st position.
leggieremente, but this might be assumed, as the note 84. Pno: The dolce invites arpeggiation and/or asyn-
patterns are similar to the previous passages. (4/a/ii). chrony, as well as the use of a moderator if available
Pianists of Beethoven’s era would nevertheless have (3/d/i and PT: 3/b).
added agogic nuances to such a passage to avoid 85. Pno, Vl: From here, in Vl, Singer starts to replace
monotony and to bring out important notes and to staccato dots with –, suggesting a more connected
enhance phrase shapes (2/b). In 64 – 65, the rh chords portato than at the beginning of the movement. In
could be swiftly arpeggiated (5/c/ii). Likewise, the 86 and 95, where Pno and Vl play repeated notes
lh octaves in 67–69. In 69, the grace-notes played as together, Pno retains the original dots while the Vl
acciaccaturas, simultaneously with the main note has –; this reflects a different understanding of por-
and released quickly (5/a/iii). In 107–112, all accom- tato notation by pianists and violinists, which was
panying chords might be swiftly arpeggiated. discussed in detail in Brahms’ correspondence with
70iv–71i. rh: After the trilled notes, Speidel adds grace- Joachim.124
notes at the same pitch, to indicate that the trills 85i–ii, 94i–ii. Pno, Vl: Speidel provides the following
should be played without the normal turn (these execution of the trill at 85 in a footnote, and Singer
are not present where the Vl plays this figure in 78f): repeats the same as rh in a footnote in Vl at 94:
b Ÿ~~~~~~~~~~
& b bb œ ÆJ nœ ÆJ
œ œ
cresc. 124 See Clive Brown: “Joachim’s violin playing and the perform-
ance of Brahms’s string music,” in: Performing Brahms, ed. Michael
123 See Spohr: Violinschule, p. 175. Musgrave and Bernard Sherman (Cambridge, 2003), pp. 52–54.

œ
b œœœœœœœœœœœœœ
of the later 19th-century than of Beethoven’s time,
& b bb when there is no clear evidence for this practice.
 b
& b bb œœ œ
3
œ
.
Spohr, who stresses the expressive value of using
the bow near the bridge for strong effects and over
the fingerboard for delicate ones, might well have
moved towards the bridge during the cresc. and
Reinecke gave a similar treatment of the ending. gone directly to the fingerboard for the pp.
b œŸ. œj œ œ œ
&b bb
Scherzo
Beethoven, however, would almost certainly have Allegro
expected an upper-note start to the trill. Tempo
86. Pno: The chords marked portato might be ar- Beethoven’s metronome marks for Allegro scherzos,
peggiated slightly, as recommended by Moscheles taking into account the occurrence of triplet 8ths in
(see note 0–15 above), and with una corda pedal (PT: this movement, suggest a tempo perhaps closer to the
3/a). Haslinger/Czerny marking than to Moscheles’. The lat-
88i–ii. Vl: A harmonic is marked by Alard, David, er metronome marks, especially Alard/Diémer, seem
Singer, Halir, Seybold. somewhat slower than Beethoven might have imag-
90, 99, 100. Pno: For the slurred quarter-note to 8th- ined for such a movement.
note figures, the first could be swiftly arpeggiated, Haslinger . = 76
the second not (5/c/ii). Moscheles . = 84
93–96. Pno: All chords could be swiftly arpeggiated Czerny Vortrag . = 76
(5/c/ii), the chords marked portato in 95 especially Alard/Diémer   = 184
so (see note 0–15 above). Speidel/Singer . = 72
98iii. Vl: Alard, Singer, Halir take the figure in 3rd Kreisler . = c. 60
position with a harmonic on iii. Czerny writes: “Lively, very humorous and cheerfully
98. Pno: Asynchrony (with rh slightly after lh) on main joking. In the Trio, the right hand legato, and the bass
notes would almost certainly have been expected brought out emphatically, since it proceeds canoni-
(5/c/ii). cally with the violin.” 125
101–103. Pno: The octaves in lh could be gently ar- Beethoven, in later life, may have felt that the char-
peggiated which would soften the edges and give acter of this movement was inappropriate for its con-
a special expressivity. Speidel marks molto espress. text in an otherwise serious and intense sonata. Anton
The grace-notes in 102 rh as acciaccaturas (see note Schindler claimed that, in connection with a projected
60–68 above). Curiously, in 103, Reinecke, Speidel, edition of his collected works, he seriously considered
Ganz remove Beethoven’s cresc., and insert < > to “whether, in order to achieve greater unity, some of
match the Vl. the four-movement sonatas of earlier times, in which
103–107. Pno: All the chords apart from 104i and 105i the four-movement structure was only a matter of ac-
might be arpeggiated in a moderate speed. In 104, cepted custom, should be made into three-movement
the < > invites agogic lengthening. pieces.” 126 And he went on to say that Beethoven was
105–106. Pno: Reinecke, Ganz inserts < > in the “definite, however, that the Scherzo Allegro should be
second half of each bar to match Vl. removed from the highly-pathetic Sonata in C minor
113–114. Pno, Vl: These final chords would almost cer- with Violin, op. 30, as conflicting with the character of
tainly be arpeggiated by the pianist (5/c/ii). What the whole. He was always against this movement and
Beethoven intended in writing a cresc. in 113 is not
clear for the Pno. Certainly, a crescendo effect could 125 “Lebhaft, sehr humoristisch und heiter scherzend. Im Trio die
rechte Hand legato, und den Bass mit Nachdruck herausgehoben,
be achieved through an arpeggiation but not if the da er mit der Violine kanonisch fortschreitet.”
chord were to be played with notes together. Ganz 126 Anton Schindler: Biographie von Ludwig van Beethoven. Dritte,
removes cres. and replaces it with < > . Speidel/ neu bearbeitete und vermehrte Auflage (Münster, 1860), vol. 2, p. 215.
“ob nicht zu Erzielung größerer Einheit einige der viersätzigen
Singer mark ’ before the last chord. This kind of Sonaten aus früherer Zeit, in welcher die Vielsätzigkeit nur ange-
Luftpausa may well have been more characteristic nommener Brauch war, in dreisätzige umzuwandeln seyn.”

wanted to leave it out.” 127 Whether, in view of Schind the first – a swift arpeggiation to mitigate harsh-
ler's known distortions of the truth, this represents ness; the second – playing the lowest note in the
Beethoven's or Schindler's opinion is questionable. lh with the beat and the other notes in lh and rh
1ff. rh: The grace-notes would be best played as ac- together and slightly after which would create a
ciaccaturas, simultaneously with the main note and slight agogic accent.
released quickly (5/a/iii). Czerny advised that chords 31–34. Vl: This was probably executed by most with
of short value (which he marked staccato in his mu- a very short, elastic bowstroke in the middle of the
sical examples) should not be arpeggiated, but this bow (Singer marks “M” on 35ii), returning to the
might not have precluded very tight arpeggiation point on 36ii (Hermann, Brodsky, Halir mark slur
(see note to 23 in the first movement of this sonata). red staccato on 36i–iii).
2, 17, 22, 36, 47. Pno: The first note in rh might be 42. Pno: The double-note chords in rh and lh might
played slightly after the lh (5/c/ii). both be swiftly arpeggiated and, if they were played
4, 12, 38. Pno, Vl: The first note of the grace-note slide unequally, certainly a choice in 19th-century pian-
should be aligned with the chord in the lh (which ism (2/a), the second chord might in reality coincide
itself might be swiftly arpeggiated) (5/a/iv). with the last note in Vl.
7, 15, 17, 41, 45, 47. rh, Vl: The trill should surely have 48–50. Vl: It is probable that most violinists of Spohr’s
an upper auxiliary start (though Speidel gives main and David’s generations would have begun near the
note start fingering). At the Scherzo tempo it can frog, executed 49iv–vi in the middle and then con-
only be played as a four-note turn, unless begun tinued to the point on 50i.
an 8th early, when six notes could be played (5/b/ii). 48–50, 58–61. Pno, Vl: The < > might elicit a
8–18. Vl:  and  markings suggest that the majority slight increase of momentum towards the apex and
of editors executed separate 8ths and quarters near lingering there before returning to tempo (3/b/v).
the point (Halir marks “Sp[itze]” at the beginning) 48ff. Pno: As advised by Czerny the rh should be le-
or, when f, with a longer bowstroke in the middle of gato. Speidel marks slurs in the first two bars then
the bow, although some may have envisaged a mar- legato.
telé near the point in 16f, 46f. Singer marks “M[itte]” 53, 54 etc. lh: Any or all of the notes (octaves) marked
in 8, but he differs from the others by marking not sf might be played asynchronously, lh before rh, as
only a slur over the staccato in 9 and 11, but also  a means of creating emphasis without harshness
over each note, which may be his way of indicat- (5/c/ii; PT: 1/b).
ing a thrown stroke near the middle; from 13i–14iii, 69–72. Vl: Alard, David, Joachim, Auer, Seybold, who
however, Singer marks a normal slurred staccato, left the text exactly as in the original (in Joachim-re-
suggesting perhaps that 8iii was to begin near the vised, slurred staccato was added in 69f), may con-
point. The editors differ over hooked   bowing ceivably have expected the use of the middle with
and separate ,  bowing for the dotted figures. a sautillé, but some of them probably envisaged it
22–25, 26. Vl: In 22–25 all editors arrive at  on 24ii near the point. Bowing in the upper half is clear
and surely intended everything to be played close from Halir’s markings. Hermann, also with the
to the point; Singer marks “Sp[itze]” on 22i. In 26 original text in 69f, surely expected these bars near
Alard, Auer, Kreisler mark the whole bar as slurred the point, because from 71iii–72vi he marked slurred
staccato. staccato. Singer, who marked slurred staccato on
27–32. Vl: Beethoven’s autograph has p only in 27, but 69iv–vi, 70iv–v, also marked slurred staccato with
in the 1st edition it has been added also in 28, prob- the instruction staccato ad lib. in 71f; the same slurred
ably to clarify that the violin is intended to return staccato bowings were marked by Brodsky, Kreisler,
to p after each sf. but without ad lib.
29–32. Pno: For these chords marked sf, pianists of The da capo should certainly include a repetition of
Beethoven’s era might have applied two practices, the second half.
127 Ibid., pp. 215–216. “Mit Bestimmtheit hatte er sich aber nur für Allegro
Entfernung des Scherzo Allegro aus der hochpathethischen Sonate Tempo
C moll mit Violine, Op. 30, als mit dem Charakter des Ganzen im
Widerspruch, erklärt. Gegen diesen Satz war er stets und rieth ihn Beethoven supplied metronome marks for several alla
aus vorstehendem Grunde wegzulassen.” breve allegros. These range from  = 120 for the first

movement of the String Quartet op. 18 no. 3, which 19. rh: It is probable that Beethoven expected the trill to
has triplet 8ths as its fastest notes, to  = 152 for the start from the upper-auxiliary note (or possible lower-
last movement of op. 18 no. 5, which has normal 8ths, auxiliary note which would produce an inverted
mostly in uncomplicated patterns; between these come turn start) (5/b/ii). Reinecke, Speidel give main-note
the last movement of op. 18 no. 4 with 8ths in more in- start fingerings, but Ganz notates a grace-note up-
tricate patterns at  = 132 (  = 66), and the first move- per-auxiliary start.
ment of the Piano Sonata op. 106. At  = 138. The final 29ff, 191ff. lh: Some or all of the chords could be ar-
Allegro of op. 30 no. 2 corresponds most closely with peggiated swiftly (5/c/ii).
the last two, suggesting that perhaps Moscheles’ mark- 35–39, 197–100. rh, Vl: In a passage such as this, Beet-
ing is rather fast, but perhaps not implausible. hoven is likely to have expected a certain degree of
Allegro Presto agogic nuance for important notes on main beats
Haslinger  = 144 and notes marked sf (2/b).
Moscheles  = 152  = 88 37f, 199f. Vl: Beethoven surely envisaged the obser-
Czerny Vortrag  = 132 vance of his long slur, beginning , and therefore
Alard/Diémer  = 132 the execution of the following quarter-notes at the
Speidel/Singer  = 144 point of the bow. All editors observe this bowing,
Kreisler  = c.120 starting , except Alard, Auer, who change to  at
Czerny explains: “The wild, excited, stormy humour 38i, and Singer, who retains the long bowing, but
that characterizes this piece of music must be ex- marks the beginning  (probably envisaging a fou-
pressed, at a rapid tempo, alla breve, by appropriate, etté for the sf ) and hooks in 39ii. All three of these
vigorously determined performance. Only the middle violinists clearly envisaged the quarter-notes at the
section (from the 40th bar on) must be very light, and frog of the bow.
be staccato in both hands. The final Presto as stormy 39–54, 201ff. Vl: Alard, Singer, Auer evidently remain
as possible.” 128 near the frog and take the long slur in 53 ; Her-
0 etc. Pno: Beethoven’s slurred duplets certainly im- mann, Halir, Kreisler evidently move to the frog
ply that he did not want four equal value 8th-notes, during this passage, probably at the cresc. in 51 and
but the first of each duplet stronger and perhaps also take 53 . The others stay in the upper half and
slightly longer than the second (2/a). take 53 . The same procedures take place at 199ff.
2ff, 93ff, 166ff, 257ff, 313ff. Pno: Czerny advised that 39–56, 201ff. lh: Beethoven probably expected legato
chords marked staccato should not be arpeggiated, with over holding as far as possible (4/a/ii), but Spei-
but this may not have precluded very tight arpeg del marks legiermente at 39 and puts staccato marks
giation (see note 23 in the first movement of this on alternate 8th-notes from 39– 42, 47–50, 201–204,
sonata). 209–213.
3, 10 etc. Vl: David, Brodsky mark slurred staccato , 42, 46, 60, 222. rh: Given that the main note of the trill
probably starting near the point. All the others take is sounded immediately before it, the trill would al-
separate bows, probably at the frog. most certainly have been expected to start on the
15ff, 107ff, 179ff. Pno: Chords, particularly on main upper auxiliary, but given the tempo, might be re-
beats, could be arpeggiated swiftly but in varying alised as a turn starting on the upper-auxiliary note
speeds according to harmonic importance (5/c/ii). (5/b/ii). Speidel, Ganz however, give main-note start
17, 113, 181. Pno: The grace-note compound appoggia- fingering.
tura in rh should be aligned with the chord in the 53. rh: Beethoven most likely intended slurs as indi-
lh (which itself might be arpeggiated) and could be cated in 54.
made quite long according to the desired rhetorical 55–56, 217–218. rh: Beethoven probably expected lega-
effect (5/a/i). to with overholding (4/a/ii). Reinecke, Speidel marks
slurs. The lh octaves and chords might be arpeggiat-
128 “Der wild aufgeregte stürmische Humor, der dieses Tonstück
ed swiftly (5/c/ii).
characterisirt, muss sich im raschen Tempo alla breve durch ent- 63–65, 227. Pno: Legato with overholding would prob-
sprechenden, kräftig entschlossenen Vortrag kund geben. Nur der ably have been expected here. In 63 (225) Vl and 65
Mittelsatz (vom 40sten Takte an) muss sehr leicht, und in beiden
Händen staccato, vorgetragen werden. Das Schluss-Presto so stür- (227) rh and Vl, the trills should probably start on
misch wie möglich.” the main note, because the upper auxiliary is pre-

viously sounded, and given the tempo, they might early editors, Reinecke opts for legato apart from
consist of 5 notes including the turned ending (5/b/ 144 –147 (possibly to match Vl), Halir for slurred stac-
ii). cato, while Ganz marks ben articulato. At 142, Ganz
66–71, 228–234. Pno: All chords in lh could be ar- marks marcato. Throughout the passage until 165,
peggiated swiftly (5/c/ii). From 68–71 (230–233), le- there are places here and there where arpeggiation
gato with overholding was probably intended for and/or asynchrony (for example 136i, 142i, 144i, and
the descending arpeggio figures in rh (4/a/ii). Rei- all octaves marked sf ) might be applied to create
necke, Speidel mark them with slurs. energy and fire (5/c/ii), despite the contrapuntal na-
72–79, 244 –241. Pno: The 8th-note broken figures ture of the writing.
should probably be legato with overholding (4/a/ii). 142ff. Vl, rh: Beethoven marks no staccato here or in
At 76i, Speidel writes in a footnote: “This chord can the similar passage beginning in Pno at 134. Most
be placed under both hands for enrichment as given violinist editors added staccato marks; only David,
[in the example].” 129 Hermann, Brodsky, Halir leave the notes without
œœj
5 markings (David-revised adds staccato marks); Auer
? bb œœ
3
marks them –. Beethoven’s intention may either
1
b œ
J have been for a distinct non-legato execution, or for
Speidel marks legato in 234 rh and 238 lh, but not legato in the piano and détaché in violin (as given
in 72 and 76. in Reinecke/Hermann). It is quite likely that some,
82i–ii. Vl: Alard marks 4 on i and a harmonic on ii; perhaps all the violinist editors regarded the stac-
Singer 0 on i and a harmonic on ii, but returns to cato marks not as shortening the notes, but merely
1 on the D-string at 83ii. All the others stay in 1st indicating vigorous separate bows.
position. 146ii. Vl: Beethoven’s autograph and 1st edition clear-
106ii. Vl: Singer adds cantabile. ly give a  , as does AG. David, Hermann, Singer,
107ff. Pno: See note to 15ff above. Here, Beethoven’s Brodsky, however, mark a #; presumably they be-
dolce perhaps invites slower and more noticeable lieved the natural sign was an error in view of the
arpeggiation of all chords as recommended by P. A. semitone on every other occurrence of this figure.
Corri (5/c/ii),130 as well as the use of a moderator if 149ff. lh: Ganz marks pesante.
available until at least 130 (PT: 3/b). In 110 lh Beetho- 150ff. rh: Beethoven may well have expected legato
ven surely intended legato with overholding (4/a/ii). and overholding for the descending figures (4/a/ii).
Reinecke, Speidel, Halir mark slurs. In 127, the trill Reinecke is the only editor to mark these with slurs.
in rh should probably be from the upper auxiliary, 163. Pno, Vl: Speidel/Singer mark allargando, with a
as the main note is previously sounded (5/b/ii). tempo at 165.
130–131. Pno, Vl: Beethoven surely intended the scales 183. rh: The trill probably from the upper auxiliary
to be non-legato, the piano articulation reflecting (5/b/ii). Ganz marks an upper-auxiliary grace-note.
the violin’s broad détaché. Singer marks the Vl with 184, 186, 188. rh: Beethoven probably expected legato
staccato dots in 130 and Halir marks the Pno with (4/a/ii). Reinecke, Speidel mark slurs. Halir marks
slurs and dots in 130–131. 184 with slurred staccato, and 186 and 188 staccato.
131iii, 132i. Vl: A harmonic is marked by Alard. 234, 238. Pno: Speidel marks legato in rh at 234 and in
131. Pno, Vl: The trill might effectively start either lh at 238.
from the main note or from the upper- or lower- 225. Vl: Beethoven’s inconsistent notation of the trill
auxiliary note (5/b/ii). Speidel, Vogrich, Ganz give was surely not intended to indicate anything differ-
main-note start fingering. ent than if he had notated the turn with small notes
134ff. Pno: It is unclear what articulation Beethoven as at the equivalent place in 63.
expected for the 8th-notes, perhaps non legato giv- 266–281. Pno, Vl: At 266, where Beethoven wrote con
en the nature and intensity of the writing. Of the espressione, Speidel/Singer add poco tranquillo and at
280 smorz[ando]. For pianists of Beethoven’s era the
129 “Dieser Accord kann zur Erleichterung, wie vorstehend unter term con espressione would almost certainly have
beide Hände vertheilt werden.” invited noticeable arpeggiation (5/c/ii). Ganz marks
130 Philip Anthony Corri: L’anima di musica, an original treatise
upon piano playing, in which musical expression & style are reduced to tenuto lines on all chords in 267–272 rh.
system […] (London, 1810), pp. 76f. 282–291. Pno: Asynchrony and/or arpeggiation of the

half-note downbeats would be effective (5/c/ii). The humour. In particular, the 12 last bars of the first part
8th-notes in 286–287 possibly legato (4/a/ii). Reine and the next 10 of the second section [81–102] should
cke marks slurs in the piano part. be played in a bustling [the German word ‘rauschend’
292–311. Pno: Beethoven possibly expected legato has no exact equivalent in English, it conveys both
throughout (4/a/ii), though none of the early edi- noise and movement, for instance in a fast-flowing
tors marked slurs, apart from Reinecke, who added mountain stream] and exciting manner.” 133
them from 300–311. 1–2. Vl: Alard marks  at the beginning of the bar,
324 –325. Pno: Beethoven might have expected non- evidently intending the 8ths in 2 to be played at
legato or staccato here given the staccato in Vl. All the point of the bow. David achieves the same end
the editors mark staccato dots. by taking 2i–ii slurred staccato . David’s bowing
was removed in David-revised and all later editors
begin  in 1 and take the passage as it comes, evi-
Opus 30, No. 3 dently more in the middle of the bow.
Allegro assai 2ff. Vl: Here and elsewhere throughout this move-
Tempo ment, the staccato marks on 8th-notes were probably
The term assai is problematic, with the potential mean- intended mostly to indicate well-articulated sepa-
ing of ‘very’ or ‘rather’. Here all the metronome marks rate bows rather than significantly shortened notes
provided by 19th-century musicians suggest that they (see Bernhard Romberg’s comment in 4/c).
understood it in the former sense. This does not seem 3ff. Pno: Chords, particularly on main beats might be
to have been Beethoven’s understanding of the term: swiftly arpeggiated (5/c/ii). In rh, the slurs invite
the contexts in which he employed it, and also his use overholding (4/a/ii).
of the German equivalent ‘ziemlich’ on several occa- 4ff. Pno: According to Czerny in his Pianoforte-Schule
sions, strongly suggests that he intended it to convey op. 500 (1839), “all chords consisting of very short
‘rather’, ’fairly’, or ’enough’.131 It seems very likely that notes” should, unless expressly marked by the com-
Czerny and Moscheles both misinterpreted Beetho- poser, be unarpeggiated. But this might not have
ven’s use of assai as indicating ’very’.132 In this case, precluded extremely swift arpeggiation, the type
unusually, the metronome mark suggested in the two that Thalberg in L’Art du chant (1853) described as
later editions may be closer to Beethoven’s conception, presque plaqué (almost together).
but even these may be somewhat too fast for Allegro 9ff, 125ff. Pno: Beethoven’s dolce perhaps invites slow-
assai. Beethoven’s fastest metronome mark for a plain er and more noticeable arpeggiation of all chords
Allegro in 6/8 is . = 104, for the first movement of the as recommended by P. A. Corri in L’Anima di musica
String Quartet op. 18 no. 5, which has relatively few (1810) (5/c/ii), as well as the use of a moderator if
16ths; the first movement of op. 59 no. 2, which has available.
a larger number of 16ths, comparable with the first 13ff, 128ff. rh: The slurs invite overholding (4/a/ii).
movement of this sonata, though in rather more intri- 16ff. Vl: The figure with two slurred 8th-notes followed
cate bowing patterns, is given . = 84. Perhaps a tempo by a separate staccato 8th-note recurs frequently in
between . = 88–96, would be closer to Beethoven’s con- this movement. David almost invariably hooks the
ception of this movement. staccato note into the bow that takes the slur, but
Haslinger . = 112 these hooked bowings were removed in David-re-
Moscheles . = 112 vised. Singer hooks here, but not later; Halir hooks
Czerny Vortrag . = 112 frequently; Rosé sometimes; Seybold very occasion-
Alard/Diémer . = 104 ally; Alard, Hermann, Joachim, Kreisler never (al-
Speidel/Singer . = 104 though some are added in Joachim-revised). Differ-
Kreisler . = c. 96–100 ent violin playing traditions seems to be particu-
Czerny (who designates this sonata op. 30 no 2) writes: larly evident here.
“One of Beethoven’s most lively, humorous and bril-
liant sonatas, if it is performed with the proper fire and 133 “Eine der lebhaftesten, launigsten und brillantesten Sonaten
Beethoven’s, wenn sie mit dem gehörigen Feuer und Humor vorge-
131 Steward Deas: “Beethoven’s ‘Allegro assai’,” in: Music and Let- tragen wird. Besonders sind die 12 letzten Takte des ersten Theils
ters 31 (1950), pp. 333ff. und die 10 nächstfolgenden des 2ten Theils rauschend und aufge-
132 Noorduin: Beethoven’s Tempo Indications, p. 203. regt zu spielen.”

16, 18ff. lh: What articulation Beethoven intended for 35, 39, 144ff. Pno: The pp invite una corda pedal (PT:
the octave leaps and repeated notes is not clear, but 3/a). Arpeggiation of the chords was probably ex-
it may well have been predominantly legato (or at pected (5/c/ii).
least long notes rather than short notes). On lh i, 36, 40, 144, 148. Vl: Here it is very likely that Beetho-
Reinecke, Speidel, Diémer, Halir insert a quarter- ven merely wanted to indicate separate bows by
note stem to create overlap. Halir also adds articula- his staccato marks, but with no significant degree
tion to indicate longer notes: of shortening.
. . -œ œ œ 42– 46, 150–154. Vl: Alard, Hermann alone retain Beet-
? # ‰ œ œ œ œ œ œ ‰ œœ ‰ œœ
hoven’s long slurs without any indication of chang-
 ?# ‰
œ
œ. œ. -œ œ œ œ œ œ. œ. # œ. œ.
œ #œ
ing the bow. Singer marks ’ after 42/150, but it is
clear that all the others expected a legato connec-
tion here. There are various approaches to fingering:
David marks nothing, perhaps expecting 1st posi-
20iii–iv, 22iv–v. Vl: Singer fingers these differently: in tion with an extension for the c3; some evidently
20, 0–1 and in 22 he marks a portamento 2– 2o. Halir envisaged portamento effects.
marks a harmonic in both bars. 43– 48, 151–155. Pno: The texture with overlapping of
20ff. Pno: The ascending broken chords in rh should parts here lends itself to frequent application of
probably be legato with overholding (4/a/ii). Rei- asynchrony with rh slightly after lh (5/c/ii; PT: 1/b).
necke, Speidel, Diémer, Halir mark slurs. Speidel, 49. Vl: Singer suggests the following execution of the
Ganz indicate sustaining-pedal. For lh, Speidel marks trill:
staccato. # œ œ .
26–27. rh: The slurs invite overholding (4/a/ii). & œ œ# œ œ œ
27iv–32. Vl: The editions reveal various approaches
5
to this passage. Alard appears to expect  on the but his main-note start in this context is certainly
first of each pair of 8ths, therefore the separate 16ths anachronistic (5/b/ii).
beginning ; Joachim marks no bowing in 27 or 29, 50ff. lh: Speidel marks staccato, while Ganz gives sus-
but on 31ii gives . Singer also begins  on the 8ths, taining-pedal through each bar. Application of asyn-
but suggests staccato ad lib. for the separate 16ths. chrony would be appropriate in this passage on im-
David, Halir beginning  in 28, apparently take the portant notes (main beats) (5/c/ii).
16ths as they come, but in 29, 31 execute the 8ths  . 51, 159. rh: The grace-note should probably be played
28, 29, 30, 31, 49, 136f, 157. Pno, Vl: For the trill there as an acciaccatura, simultaneously with the main
is probably only time for a turn from the upper note note and released quickly (5/a/iii), as explained in
or perhaps, by means of shortening the preceding Junghanss’ Pianoforte-Schule (Vienna, c. 1820).
note, an additional reiteration of the upper auxil- 53vii–ix. Vl: All except Alard, Joachim mark slurred
iary and main note can be included (as illustrated staccato, but it is added in Joachim-revised.
in the footnote in the edited violin part). This kind 53ff. rh: Beethoven probably intended legato with over-
of rhythmic treatment to increase the brilliance of a holding (4/a/ii). Speidel marks non legato, but Dié-
trill is illustrated in Spohr’s performing version of mer marks slurs over each bar.
Rode’s 7th Violin Concerto (5/b/ii). 57ff, 165ff. Pno: Beethoven probably intended legato
29, 31–32. rh: The scalic runs should probably be lega- for the continuous 16ths. Diémer marks lh (and rh
to. Reinecke, Diémer mark slurs, but Speidel, Halir from 61) with slurs.
mark staccato apart from i–ii, for which they retain 65–66, 173–174. lh: Swift arpeggiation of chords is ap-
a slur. propriate (5/c/ii).
32. Vl: All except Rosé (who goes from 1 on i to 4th po- 67i, ii. Vl: Singer marks harmonics.
sition on ii) either mark or obviously expect 3rd fin- 67–78, 175–186. Pno: Asynchrony would help to make
ger in 1st position on 32i and a harmonic on ii; Sin poignant moments such as 68i and 69i etc. expres-
ger, Auer, Brodsky, Halir, Seybold specifically mark sive (5/c/ii; PT: 1/b).
the harmonic with 3rd finger. A harmonic with 4th 69ff. Vl: Most envisage beginning in 3rd position, some
or, less plausibly 3rd finger, would have been a going to 1st and some remaining centred in 3rd. Alard
natural fingering for violinists of Beethoven’s time. marks a harmonic on 70ii, 74ii, 76ii, Halir on 70ii.

77f. Vl: Younger violinists (from Auer onwards) go ma molto moderato were added subsequently, and final
onto the G string either from 77i or iii; this would ly in a different ink e grazioso. The addition of molto
not have been implausible for an early-19th-century moderato was evidently intended to reduce the speed;
violinist. whether grazioso was intended to reduce it further,
79–80. Pno: It is unclear what Beethoven expected in rather than characterise it, is debatable. The Haslinger
terms of articulation. Speidel marks lh staccato. marking seems very plausible; Moscheles’ suggestion
Several editors mark 80 rh with a slur. is probably somewhat too fast for molto moderato. The
81ff. Pno: Given the articulation in Vl, Beethoven may later markings are surely too slow, since they entirely
have expected non-legato. Speidel marks lh with lose the feeling of a minuet.
staccato patterns, Halir with combination of tenuto Haslinger  = 92
lines and staccato, but Diémer with slurs. Moscheles  = 112
103ff. Pno: It is not clear what articulation Beethoven Czerny Vortrag No metronome mark given
expected in rh. Speidel marks the rh with staccato Alard/Diémer  = 84
dots. Beethoven may have expected the use of the Speidel/Singer  = 84
una corda in 103 for the pp (PT: 3/a). Ganz marks una Kreisler  = c. 76
corda. In 104, he marks senza Pedale at the beginning Czerny writes: “With artless grace and tender feeling
of the bar and misterioso where the lh enters. throughout, but not dragged. The sf on the bass notes
107. Vl: Singer adds leggiero. (from bar 51 onwards) must be very noticeably accented,
151iii–iv, 153iii–iv. Vl: David, Joachim, Auer, Halir, so that the 2nd quarter-note of the bar is powerfully
Rosé, Seybold mark the portamento shift 4 – 4 in both staccatoed, while all the rest remains piano.” 134
places. The triplet figurations in Pno 9–15 and the assimi-
161ff. rh: It is not clear what articulation Beethoven lation to triplets that was almost certainly envisaged
expected here. Diémer marks slurs on each bar. in 80–85 makes it likely that all dotted figures in this
188. rh: It is not clear what articulation Beethoven ex- movement were expected to be played with a relaxed
pected here. Diémer marks a slurs over the bar. rhythm probably varying flexibly between 4 : 1 and 3 : 1
189–200. lh: Given the articulation and character of and perhaps occasionally over-dotted, for instance in
Vl, Beethoven may well have expected non-legato, the passage from 22–27 and similar places.
perhaps in combination with legato. Diémer marks 1–8. Vl: Singer, Auer, Rosé, Kreisler provide more ‘ex-
each bar with a slur. Halir marks a combination of pressive’ fingerings for this accompaniment part,
staccato and tenuto lines from 189–193. but these seem unlikely to have been envisaged by
the composer.
Tempo di Minuetto, ma molto moderato e grazioso 1–8, 30–37, 91–98, 120–127, 178–181. Pno: The texture
Tempo here permits several opportunities for expressive
Beethoven made a clear distinction between minuet/ asynchrony with the rh after the lh (5/c/ii).
scherzo movements, to which he gave a metronome 3, 11, 32, 53 etc. Pno, Vl: The trill in 3 etc. was prob-
mark for the dotted half-note, or expected a dotted ably envisaged as starting from the upper auxiliary,
half-note pulse, and those he designated Tempo di but Speidel and Ganz give main-note start finger-
Minuetto or a similar term (opp. 20, 22, 31 no. 3, 34 ings. These trills are surely intended to be followed
var. 4, 49 no. 2, 54, 59 no. 3, 93 etc.) for which he gave by a turn (5/b/ii).
a quarter-note metronome mark, or clearly expected 6, 14, 35, 56 etc. Pno, Vl: In Beethoven’s autograph
a quarter-note pulse at around the same tempo. The the turn in 6, where an accidental is required, was
latter are much rarer in his output than one-in-a-bar written out, though not in the violin part in 14. In
scherzos and scherzo/minuets. The only slow minu- 3, 11, and equivalent places, where no accidental is
ets for which he gave metronome marks are the third needed, Beethoven evidently considered it unneces-
movements of the Septet op. 20 and of the 8th Sym- sary to specify the turn. The turns in 3, 11, and all
phony op. 93, to which he gave  = 120 and 126 respec-
tively. In the autograph, this movement was originally 134 “Durchaus mit naiver Anmuth, und zarter Empfindung, aber
designated Andante, then Tempo di Minuett (at which ja nicht schleppend vorzutragen. Die sf in den Bassnoten (vom
51sten Takt an) müssen sehr auffallend markirt werden, so dass
point he probably had a tempo of around  = 120–126 die 2te Takt-Viertel kräftig abgestossen wird, während alles Andre
in mind); the o on Minuetto and the qualifying term piano bleibt.”

occurrences of these figures, as well as those in 6, String Quartet op. 130, but this is the only time (as
14 etc. are written out by Reinecke/Hermann and far as I (CB) have discovered) that it was included in
Speidel/Singer. a printed edition. The notation was almost certainly
4, 73–74. rh: Beethoven’s duplet slurrings imply a nu- derived from the 1787 edition of C. Ph. E. Bach’s Ver-
anced treatment with the first of each stronger and such, where it was used to indicate that the value of
perhaps a little longer than the second. This is sug- the dot should be treated as a rest (2/c/iv). Beethoven’s
gested by Halir’s marking of the first with a tenuto use of it here raises the question why it should im-
line, the second with a staccato dot in 4. ply any different performance than a staccato mark
5, 13. Pno, Vl: The grace-note before ii undoubtedly over the dotted note. Given that the staccato mark
stands for a full-length 16th (to be aligned with the may be taken to shorten the note over which it is
bass note), which also means that in performance it placed, although, of course, it need not do so (4/c), its
might well be given greater length at the expense positioning over the dot of prolongation may have
of the following notes (5/a/i). Speidel adds in a foot- been intended to guard against too short an execu-
note “langer Vorschlag” (long appoggiatura). Rei- tion of the first note of the figure. Reinecke/Her-
necke, Halir write it out as a normal 16th-note. Ganz mann treat the figure rather inconsistently, some-
aligns the grace-note with lh iii using dotted line times with no staccato, sometimes with – on the first
notation. note and sometimes with a staccato dot. Speidel/
6. rh: The first note of the grace-note turn aligned Singer consistently notate it in the score (in both Pno
with lh i (5/b/i) which Ganz shows with dotted line and Vl) with –. , but in the separate violin part only
notation. with –; and Halir also marks – instead of a staccato
7. rh: The turn should start from the upper auxiliary mark on the dotted note. This may be close to re-
and be left as late as possible and joined on to the flecting Beethoven’s intention; though it is unlikely
following notes (5/b/i). that their use of this notation has any reference to
9–16i, 51–58i, 99–106i, 141–148i. Vl: All except Alard, Beethoven’s original, of which they were probably
Brodsky mark this to be played entirely on the D- unaware since AG printed the staccato over the first
string. This might well have been done by a violin- note of the figure.
ist of Beethoven’s time, but a more cultivated player 19ff. Pno, Vl: The grace-note should be aligned with
would probably have given the four-times repeated the bass (as indicated by Ganz with dotted line no-
theme a varied treatment, perhaps, for instance us- tation) and played as a 16th-note (5/a/i).
ing the D- and A-strings at first and reserving more 38f. Pno, Vl: Beethoven’s inconsistent slurring here in-
expressive fingering and tone colour for a later oc- dicates that he did not not expect a phrasing break
currence. In the edited violin part, greater use of before the third beat of the bar, but legato throughout.
portamento is suggested on the final complete state- 61–63, 69–71, 151–153, 159–161. Vl, rh: Beethoven’s in-
ment of the theme. tentions for the placement of the staccato marks in
9–15, 51–57, 75–77, 99–105, 141–147. Pno: All slurs im- these places is unclear (see Critical Report).
ply overholding (4/a/ii). 66–74, 157–160. Pno: All slurs imply overholding (4/a/
17–28, 37– 49, 51–58, 107–118, 127–139, 162–174, 181–187, ii). All important notes in rh might be played slightly
189–196. Pno: All chords not marked staccato could after (exceptionally before) the lh to create expres-
be arpeggiated, particularly those on main or im- sive nuance (5/c/ii). In 69, the grace-note should be
portant beats (5/c/ii). aligned with the lh as indicate by Ganz with dotted
19, 22–27, 39, 44 – 48, 109, 113–118, 134 –138. Pno, Vl: line notation (5/a/i). Beethoven may have intended
The autograph and 1st edition are clear about a no- to write dolce at 67 as he did for the same music in
tation of the dotted figures with a staccato mark Vl from the end of 148 and Pno in 157. Some editors
above the dot of prolongation rather than the first add it. For the Pno, the term dolce elicits the use of
note of the figure. AG, however, ignores this nota- a moderator if available (PT: 3/b).
tion and most editors of the sonatas follow the read- 78. rh, Vl: AG and almost all editions (but not Alard)
ing in AG. This is not the only occasion on which place p on note ii, although it is clearly on i in the
Beethoven used this notation; it can be found as autograph and 1st edition.
early as the autograph of the solo part of the Piano 79iv–85ix. Vl: David marks a bowing slur over each
Concerto op. 19 and as late as the Cavatina in the group of three. Halir marks this only in 79–81.

thers take a separate bow for the third note of
O 0ff. Pno: leggiermente probably implies a delicate, non-
each group. legato touch, except where Beethoven marks slurs,
79–86. Pno: The pp could be achieved by using the equivalent in effect, as far as possible, to the violin-
una corda (PT: 3/a). The octaves marked portato in 79 ist’s separate bows and slurs. Ganz adds (non legato);
might be slightly arpeggiated but “giving them the Reinecke, however, marks slurs from 0i–5i and 5ii–8i.
same length of time as a dot under a slur requires” 1–8. lh: The drone octaves might be tightly arpeggiat-
as recommended by Moscheles in his Studies for the ed, which would give them a special energy (5/c/ii).
Piano Forte op. 70, Bk 1 (1827). In 80, the grace-note 4f. Vl: The majority of editors begin in 4 with   and
should be played as an 8th-note and aligned with take 5iv , Joachim, however, begins   in 4, all
the lh (5/a/i). At 81 lh iii, Speidel gives an alternative: evidently in the upper half of the bow. Hermann,
?b
bb œ however, begins   and takes 5iv , perhaps more
œ towards the middle of the bow.
œ
5, 7, 13, 15, 25, 27, 33, 35 etc. Pno, Vl: At a true Beet
80, 82–85. Pno, Vl: The dotted rhythms here were hovenian Allegro vivace tempo, the trills in this
surely intended to represent a 2 : 1 ratio. In 82 and 83 figure will probably be played as four-note turns,
the < > might inspire a slight increase of mo- since they must surely start with the upper auxil-
mentum towards and lingering at the apex, before iary. But in 61, Speidel gives the following:
a return to tempo (3/b/v).
# 3 #œ
& œ# œ œ# œ œ œ

90f. Pno, Vl: Speidel/Singer mark poco rit. in 90, fol-
1
lowed by a tempo in 91. 5

188–191. Pno, Vl: Speidel/Singer mark poco allargando
in 188 followed by a tempo in 191. 10f, 23, 46f, 81f, 151f. Pno: The grace-notes as acciacca-
turas (5/a/iii), played simultaneously with the main
Allegro vivace note and released quickly as explained in Jung
Tempo hanss’ Pianoforte-Schule (Vienna, c. 1820).
See the note to the third movement of op. 12 no. 3. Al- 12ff, 15ff. Vl: Most of the editors take the bowing as it
though that 2/4 movement, containing the same note comes, starting either  (the majority) or . Beetho-
values in similar proportions, is marked Allegro molto, ven’s leggiermente marking may have encouraged
Czerny and Moscheles give it slightly slower metro- violinists beginning  to employ a light stroke in the
nome marks. Beethoven gave  = 92 for the Allegro molto upper middle of the bow, with some elasticity of
quasi presto of op. 18/2/iv, which, however, has many the bow-stick, similar to Baillot’s detaché legère and
fewer 16ths in less intricate patterns. David’s hopping (hüpfender) bowstroke. Those who
Haslinger  = 76 begin , however, will probably have played with a
Moscheles  = 160 light, but firmer stroke, closer to the point.
Czerny Vortrag  = 76 20. lh: The chord on i could be swiftly arpeggiated
Alard/Diémer  = 132 (5/c/ii).
Speidel/Singer  = 132 21–32. Pno, Vl: Beethoven’s repeated use of the term
Kreisler  = c. 140 ten. here is probably not only an instruction to sus-
Czerny writes: “The right hand very light and not le- tain the note, but also to emphasise it, with a degree
gato, while the left one makes the octaves piano but of force less than sf. In 21–24 this may also have in-
sounding like a bell. This finale surpasses even the cluded vibrato for the violinist; in 29–32 many may
first movement in vitality, cheerful caprice, and bril- have chosen to use the open A-string, which could
liant effect. In particular, the powerful passages are also produce the kind of ‘vibrations’ (Schwingun-
to be played with a mischievous capriciousness, and gen) referred to by Swoboda (3/c/vi).
properly reinforced by the pedal.” 135 26ii. Vl: All editors except Brodsky mark a harmonic,
135 “Die rechte Hand äusserst leicht und nicht legato, während
Hermann, Singer, Rosé with 3, which may also have
die linke die Octaven piano aber Glockenartig ertönen lässt. Dieses been envisaged by others who mark only o.
Finale überbiethet noch den ersten Satz an Lebendigkeit, munt’rer 56–60. rh: The slurred broken chords with overhold-
Laune, und brillanter Wirkung. Besonders sind die kräftigen Stel-
len mit muthwilliger Laune zu spielen, und gehörig durch das ing (4/a/ii).
Pedal zu verstärken.” 60ii–65. Vl: Most of the editors begin ; Alard, Singer,

Brodsky, Kreisler, however, begin , which, playing alternatively a delicate vibrato, which Joachim ap-
the passage as notated, without adding slurs, allows parently used in dolce passages.136
a more fluid string-crossing movement and will 145ff. Pno: Here, like the opening, leggiermente is al-
have been played closer to the point (Singer marks most certainly intended.
Spitze). All except Auer leave the passage without 175f. Pno, Vl: Speidel/Singer mark allarg. with a tempo
slurs. In a number of other instances, however, Beet- after the fermata.
hoven seems to have expected passages of a similar 176. Pno: Very swift arpeggiation of the chord is ap-
kind, in which he marked neither slurs nor staccato propriate (5/c/ii). It is quite possible that Beethoven
marks, to be played with a mixture of slurred and would have considered it appropriate for either pia-
separate notes in a manner that suited the instru- nist or violinist to make a short Eingang, or, more
ment and the individual player. Auer gives a very likely, an elaborated arpeggiation of the fermata.
effective, ‘violinistic’ solution (given below the stave 177. Pno: The pp may suggest use of the una corda (PT:
in the edited violin part of the present edition). This 3/a).
type of treatment may well have been adopted by 180–184i. Vl: Various fingerings are proposed for this
earlier editors, even though their respect for the no- awkward passage. Alard gives none, apparently re-
tation inhibited them from presenting it in print. maining in 3rd position; all the rest except Brodsky,
David’s personal copies of classical works contain Halir mark a beginning in 3rd position, Hermann
many such added slurring patterns It is quite pos- remaining throughout; Joachim, Singer, Kreisler
sible that Auer derived his version of this passage also stay in position but mark 1–1 on 181vi–vii and
from his teacher, Joachim, who was noted for such 183vi–vii David shifts to 4th position on 181ii, 183ii,
freedoms in performance. returning to 3rd on viii; Auer, Rosé, Seybold begin
66–71. Pno: The chord at 66 rh might be arpeggiated like David, but remain in 4th, using 1–1 on 183i–ii;
swiftly (5/c/ii). In lh, one might either overhold notes Kreisler uses 1–1–1 on 181v–vii, 183 v–vii; Halir moves
in in the broken chords or else play them non-legato from 1st to 3rd position on 180iv, 182iv and back to
in the same way as the theme. Reinecke marks the 1st on 181v, 183v; Brodsky remains in 4th position
passage with slurs. throughout, apparently crossing to the A-string for
91–101. Pno: All chords, but especially those marked 181i, 183i. Hermann, Auer, Halir modify the slur-
sf might be swiftly arpeggiated to produce accent ring, changing 181 and 183 from i–ii to i–iii.
without harshness (5/c/ii). In lh, legato may well have 181–187. Pno: The dolce invites the use of the modera-
been intended. Reinecke marks slurs. At 91, Ganz tor if available (PT: 3/b), and noticeable arpeggiation
marks non legato. (perhaps moderately slowly) of all chords (3/d/i).
92f. Vl: The grace-notes were probably expected to be 184. rh: Beethoven may have intended staccato for these
played very short and sharp, like the pianist’s ac- notes imitating the Vl in 180. Ganz marks quasi stacc.
ciaccaturas (5/a/c). 194f. Vl: Most begin ; David, however, begins , prob-
107ii. Vl: All except Brodsky either mark or envisage a ably at the point of the bow and taking the first sf
harmonic (most with 3); David, Rosé do not indicate with a fouetté bowstroke.
it, but since they remain in 3rd position it would 194 –221. Pno: All chords might be swiftly arpeggiated
have been virtually inevitable. (5/c/ii).
113–129. Pno: This passage might have the articula- 210. rh: It is unclear what articulation Beethoven in-
tion pattern marked at the beginning of the move- tended here. Ganz marks non legato.
ment. Reinecke marks slurs in lh but curiously not 215–217. Vl: Some editors, who mark nothing (David,
in rh, perhaps through oversight. Alard, Hermann, Brodsky, Halir, Rosé) may have
131–135. Pno, Vl: For the piano, the dolce at 133 in- envisage the use of the 4th finger on the D-string for
vites the use of the moderator if available (PT: 3/b), the a #1 or perhaps assumed the typical 19th-century
with noticeable arpeggiation in lh (perhaps moder- use of successive 1st fingers; Joachim, Auer, Kreisler
ately slowly) (3/d/i). In Pno, Speidel moves dolce to explicitly mark 1–1 on 216 and 217iii–iv. None sug-
131 and marks sustaining-pedal through each bar. gest half position for 216i–iv, 217i–iv.
Ganz moves dolce to the middle of 132. For the vio-
linist, the dolce may have encouraged the very light, 136 See 3/d/i and also Brown, Peres Da Costa, Bennett Wadsworth:
fast bowstroke suggested by August Wilhelmj, or Performance Practices in Johannes Brahms’ Chamber Music, p. 14.

218. Pno, Vl: The cresc. and the character of the mu- for the Presto fourth movement of op. 59 no. 2, this is
sic encourages an accelerando. Speidel/Singer mark a likely choice for the Presto in op. 47. A tempo faster
string[endo]. than  = 88 for the Adagio, perhaps significantly so,
seems quite likely. A tradition of a faster tempo than
suggested by any of the editors may perhaps have
existed, since Marjorie Hayward and Una Bourne, in
SONATA OPUS 47
their 1918 recording, very convincingly take a tempo
Adagio sostenuto – Presto of about  = 66 for the Adagio; this may, of course,
Tempo have been influenced to some extent by the limited
There are only four Adagios in 3/4 for which Beetho- duration of shellac discs, but they would surely not
ven gave metronome marks. In all he provided an 8th- have played it vastly faster than normal.
note number. To two of these (the second movement of Adagio Presto
the String Quartet op. 18 no. 2 and the opening Adagio Haslinger  = 80  = 160
of the Septet op. 20) he gave 72 and to the others (the Moscheles  = 84  = 80
slow introduction to the first movement of the Sec- Czerny Vortrag  = 72  = 144
ond Symphony op. 36 and the second movement of Alard/Diémer  = 69  = 144
the Fourth Symphony op. 60) he gave 84; but all these Speidel/Singer  = 42  = 152
movements or sections contain a significant number Kreisler  = c.80  = 138–152
of 16ths and 32nds. The introductory Adagio sostenuto Czerny’s account is particularly interesting because he
to this movement is much broader in its character, evidently had knowledge of the premiere, which oc-
with mainly quarter-notes and a few 8ths in the first curred shortly after he became Beethoven’s pupil, and
12 bars, giving the feeling of a pulse in three rather because he arranged the sonata for piano duet during
than six. Despite the 16ths in the section leading to the Beethoven’s lifetime. He writes at length: “Only this
Presto, this suggests that Beethoven would have allot- colossal sonata, which had become the most celebrated,
ted a metronome mark significantly faster than those was able to outdo the preceding ones in grandeur, as
he supplied for other 3/4 adagio movements. it is extremely brilliant, difficult to interpret for both
Beethoven’s metronome marks for Allegro and Presto instruments, and written in a concerto-like, highly ef-
movements with the time signature C, are very fast in- fective style. For the pianist, this sonata can hardly be
deed. For the Allegro fourth movements of the String said to be particularly difficult in terms of passages,
Quartets op. 18 no. 4 and op. 18 no. 5 he gave  = 66 because (with one exception) everything falls into a
and  = 76 respectively; for the Allegro con brio first pianistic style, but its requirements for endurance,
movement of op. 18 no. 6 and the Allegro vivace of the strength, and maintenance of its stormy, wildly excited
Fourth Symphony  = 80; for the Allegro vivace fourth character always demands a virtuoso, if it is to be
movement of the 8th Symphony and the Allegro molto presented worthily. The Introduction (Adagio) is per-
finale of the String Quartet op. 59 no. 3  = 84; and formed majestically with expression. The theme of the
for the Presto finale of op. 59 no. 2,  = 88. It seems Presto very marked, and the embellishment of the sec-
likely therefore that even the fastest of the metronome ond fermata extremely fast and powerful, with pedal.
marks suggested by the 19th-century editors are slower From there the stormy motion begins, which proceeds
than the ones Beethoven would have given. clearly, at first slightly, but always increasing, to the
The 19th-century editors give tempos for the Presto calm, melodious middle section, which is to be played
that are exactly or approximately four times as fast in tempo and only ritardando eight bars before the
as those they suggest for the Adagio, presumably be- fermata. Here again the former speed and especially
cause of the anticipation of the quarter-note upbeat the following passage with maximum energy: [Ex. bb.
to the Presto that emerges in 16ths in the last six bars 144 –148 etc.] In the second part the following passage
of the Adagio. The slower than expected tempo sug- is to be practised diligently: [Ex. bb. 230–238] because
gested by Speidel/Singer may perhaps be explained it has to be played very strongly, fluently, precisely,
by their indication poco ritenuto in the last two bars and brilliantly. Towards the end of the movement, the
of the Adagio. A relationship between the two tempos effect has to be increased more and more.” 137
seems plausible, but there is nothing to suggest that 137 “Nur diese, vorzugsweise berühmt gewordene kolossale So-
Beethoven envisaged it. Since Beethoven gave  = 88 nate konnte die Vorhergehende an Grösse überbiethen, da sie für

Important sources for understanding performing prac- Chord playing
tice issues in this sonata are Carl Czerny’s arrange- Neither of the more substantial violin methods pub-
ments lished in Vienna during Beethoven’s lifetime give in-
▪ Of the second movement for piano solo: Variations structions for the bowing of 3- and 4-part chords.138
brillantes tirées de l’Oeuvre 47 / de Louis van Beethoven. The subject is also ignored by less substantial ones.
arrangées pour le Piano-Forte seul par Charles Czerny. There seems, however, to have been a widespread,
Vienne Cappi et Diabelli [c. 1823]. though not universal convention in the late 18th and
▪ Of the whole sonata for piano duet: Grand duo bril- 19th centuries that three- and four-part chords were
lant pour le Piano Forte à quatre mains, arrangé d’après la taken  unless the contrary was specified. Johann
Sonate de L. van Beethoven, Oeuv 47, par Charles Czerny. Friedrich Reichardt stated this as a general rule in
Vienne chez Ant. Diabelli et Comp. [c. 1825] (http://digital. 1776; illustrating a succession of 3-part chords fol-
bib-bvb.de/view/bvbmets/viewer.0.6.4.jsp?folder_id lowing a 4-part chord, he instructs: “Thus every
=0&dvs=1592401851840~474&pid=3483564&locale=de chord must be played down-bow, because otherwise
&usePid1=true&usePid2=true) it sounds too sharp, as if torn out. Even in the case
And, to a lesser extent of an upbeat chord, it must be played down, but the
▪ Sonata per il Pianoforte ed un Violoncello obligato […] bow taken off and the following first note of the
da L. van Beethoven Op. 47. La parte del Violoncello tra bar played down again.” 139 Georg Simon Löhlein,
scritta da C. Czerny. Bonn, Simrock [n. d. apparently however, was more comfortable with the use of up-
published after 1850, but perhaps made earlier] bow, even for 4-part chords. In his Violinschule he
▪ Edited by Cipriani Potter: A Grand Sonata or Con- explains in a footnote, having illustrated a 3-part
certante for Piano Forte and Violin, dedicated to chord , followed by a 4-part chord : “With these
Rodolph Kreutzer by Louis van Beethoven op. 47 harmonic chords, one must set the bow down fully,
London, R. Mills [n. d. c. 1860] close to the frog, and draw it strongly, with a short,
1–11. Vl: Beethoven’s employment of successive 4- and round movement, so that the notes are round and
3-part chords in this manner is unique in his violin clear. You can do it both with the up- and down-
writing. It is possible that he was inspired to use bow; but the latter is more advantageous, because
multiple stopping in this manner by Bridgetower, the down-stroke makes it stronger, and the upper
who, according to Samuel Wesley, was a masterly note, as the main note of the melody, is heard the
performer of Bach’s Sonatas and Partitas for solo longest.” 140 Spohr explains, in characteristic detail,
violin (see Introduction p. V). The possibility that his conception of multiple stopping, which allows
Bridgetower played some of Bach’s solo violin mu- alternate bows, but only for long-held chords and
sic for Beethoven is intriguing.
138 Violin Schule oder Anweisung die Violine zu spielen von Leopold
Mozart. Neue umgearbeitete und vermehrte Ausgabe (very extensively
revised and rewritten, anonymously, by Johann Conrad Wilhelm
Petiscus) (Leipzig, [1804]; unauthorised reprint by Cappi in Vienna
in 1806). We are grateful to Axel Beer for this information. For
beide Instrumente äusserst brillant, bedeutend schwierig, und in fuller bibliographic details see Axel Beer: Das Leipziger Bureau de
einem concert-artigen, höchst wirkungsvollen Style geschrieben Musique (Hoffmeister & Kühnel, A. Kühnel). Geschichte und Verlags-
ist. Für den Pianisten ist diese Sonate in Bezug auf die Passagen produktion (1800–1814) (München/Salzburg, 2020), p. 355 (print in
kaum besonders schwierig zu nennen, da, (bis auf eine Ausnahme) preparation). Joseph von Blumenthal: Kurzgefasste theoretisch-prak-
alles sehr claviermässig in die Hand fällt, aber die Ausdauer, die tische Violin Schule (Vienna, 1811).
Kraft, und das Festhalten des stürmischen, wild aufgeregten Cha- 139 Johann Friedrich Reichardt: Ueber die Pflichten des Ripien-Vio
racters derselben erfordern immerdar einen bedeutenden Virtuosen, linisten (Berlin/Leipzig, 1776), pp. 12–13. “So muss jeder einzelne Ac-
wenn sie würdig vorgetragen werden soll. Die Introduction, (Ada- cord herunter gestrichen werden, weil er sonst zu scharf, wie ab-
gio) ist majestätisch, mit Ausdruck vorgetragen. Das Thema des gerissen klingt. Selbst bey dem Fall, wo der Auftakt ein Accord ist,
Presto sehr markirt, und die Passage der zweiten Haltung äusserst muß dieser hinunter gestrichen, aber der Bogen abgesetzt und die
schnell und kräftig, mit Pedal. Von da beginnt die stürmische Be folgende erste Note des Takts wieder hinunter gestrichen werden.”
wegung, welche deutlich, anfangs leicht, aber immer sich steigernd 140 Löhlein: Anweisung zum Violinspielen, p. 56. “Bey diesen harmo-
bis zu dem ruhigen, melodiösen Mittelsatz fortwährt, der im Tempo nischen Accorden muß man den Bogen voll, nämlich nahe beym
und erst 8 Takte vor der Haltung ritardando zu spielen ist. Hierauf Frosche ansetzen, und ihn stark, mit einer kurzen, runden Bewe-
wieder die frühere Bewegung und vorzüglich die folgende Stelle gung ausziehen, damit die Töne rund und deutlich ausfallen. Man
mit aller Energie: [Ex. mm. 144 –148 etc.] Im 2ten Theile ist folgende kann sie mit dem Aufstriche sowohl als mit dem Niederstriche
Stelle wohl zu üben: [Ex. mm. 230–238] da sie sehr kräftig, ge- machen; doch hat bey dem letzten der Niederstrich den Vorzug,
läufig, deutlich und brilliant gespielt werden muss. Gegen den weil er dadurch kräftiger, und die oberste, als die Hauptnote der
Schluss des Tonstückes ist die Wirkung immer mehr zu steigern.” Melodie, am längsten gehöret wird.”

perhaps only when the  chord is 3-part (he pre- Pierre Baillot teaches that 3-note chords can and
sents no examples of consecutive 4-part chords). In should be played simultaneously by executing them
the introduction to Ex. 58 (Minuet 3/4) he states: further away from the bridge, where the strings are
On the four-note chord of the first bar, the bow is slacker.143
placed firmly on the two lowest strings close to the The practice of playing most successive chords 
frog, then pulled down with a strong pressure onto continued at least until the middle of the 19th cen-
the two highest and now evenly continues on them tury with many violinists. The Viennese violinist
to the point. Although the two lowest notes are writ- Joseph Hellmesberger senior, for instance, marks re-
peated  at the opening of J. S. Bach’s Chaconne in
ten as quarters, the bow must not linger on them, and
their length should not exceed a 16th.
his 1865 edition of the Solo Sonatas and Partitas, and
The second bar [see Ex.] is played like the first, but in
the Edmund Singer, also Viennese-trained, marked
up-bow, the third again in down-bow.
b 3 ˙.. ˙˙. ˙ œœ2  successively three times in bb. 3f, 8f, and 10f of his
& b 4 ˙œ œ œ̇ edition of this sonata. Spohr’s pupil Ferdinand Da-
œ
.
vid in 1843 and other subsequent editors, however,
ff 4
Likewise, the first four bars of the second part [see Ex.] marked many weak-beat chords , although in his
are alternately taken in the down- and up-bows. The edition of the ‘Kreutzer’ Sonata David indicated
chords in quarters, however, in the fifth and the fol- successive  in 8f and 10f.
lowing bars are all attacked down-bow, close to the Beethoven’s notation of the opening bar is idealistic;
frog, with strong pressure of the bow and broad hair
as written it is impossible to play with a smooth
and as far as possible executed simultaneously and the
bow replaced anew for each one. But the strokes must legato. What he expected, and what he might have
not be too short, otherwise the chords would become heard from Bridgetower and other early performers
sharp and dry.141 of the work can only be surmised. The editions by
˙. ˙. ˙. ˙. 19th-century violinists propose a range of solutions.
& .. n˙œ. ˙. ˙.
œœ ˙œ. All take  for the first chord and  for the next two,
ff œ
œ
except Alard, who, despite suggesting the slowest
tempo, leaves the slur over three notes. Alard and
In contrast to Spohr, David’s pupil, Friedrich Her- Hermann suggest the most obvious fingering for
mann, specifies spreading chords from the bottom the second two chords, 31 31 , which is also the most
note up, rather than 2 and 2.142 likely to have been used by violinists in Beetho-
>œ ven’s time; if the player puts the 1st finger across
œœ
>
œ œœ
œœ
œœ œœ the A- and E-strings and the 3rd finger across the
≥ ≥ ≥ ≥ D- and A-strings for the first chord, the hand can be
# 4 œœ
Moderato œ4 œ
œœ smoothly shifted to 3rd position without replacing
∏∏∏∏∏∏∏∏∏∏
& 4œ œœ œœ œœœ
∏∏∏∏∏∏∏∏∏
∏∏∏∏∏∏∏
∏∏∏∏∏∏∏
œ œ the fingers, and then just as smoothly shifted down

f
for the following chord. David provides a fingering
utilising the harmonic on ii that, while facilitating
a smooth connection between ii–iii, is awkward, be-
141 Spohr: Violinschule, p. 147. “Bey dem vierstimmigen Accord des cause on i and ii the 3rd finger is used on different
ersten Taktes wird der Bogen dicht am Frosch fest auf die bey- strings, making a break inevitable; but this finger-
den tiefsten Seiten gesetzt, dann mit einem starken Druck auf die
beyden höchsten herabgerissen und nun ruhig auf diesen bis zur
ing continued in use as late as Seybold’s edition.
Spitze fortgezogen. Obgleich die beiden tiefsten Noten als Viertel Singer, still using the harmonic on ii, proposed a
geschrieben sind, so darf der Bogen doch nicht auf ihnen verwei- less awkward version, 24 , which also has the virtue
len und ihre Dauer höchstens die eines Sechzehntels betragen.
Der zweite Takt wird wie die erste, doch im Aufstrich, der Dritte
of allowing the 2nd finger to slide up the A-string.
wieder im Herabstrich gespielt. Eben so werden die vier ersten Joachim is the earliest editor to suggest a fingering,
Takte des zweiten Theils abwechselnd im Herab- und Auf-Strich involving an extension of the 2nd finger to d2 fol-
genommen. Die Accorde in Vierteln aber, im fünften und den fol-
genden Takten werden alle im Herabstrich, dicht am Frosch, mit
lowed by 2nd position on iii, which enables maxi-
starkem Druck des Bogens und breit liegenden Haaren und mög- mum legato without audible sliding, and this was
lichst zugleicherklingend herabgerissen und der Bogen bey jedem also given by Auer and Kreisler. Auer gives Singer’s
von neuem angesetzt. Doch dürfen die Striche nicht zu kurz seyn,
weil sonst die Accorde scharf und trocken werden würden.”
142 Friedrich Hermann: Violinschule (Leipzig, 1879), p. 100. 143 Pierre Baillot: L’art du violon (Paris, [1835]), p. 85.

fingering as an alternative (without specifying the to 3rd position for ii, which allows a smooth shift of
harmonic). Each of these fingerings gives a very dif- the 1st finger on the D- and G-strings, and this was
ferent character to the opening bars. also given by Auer, Brodsky, Halir. Kreisler shifts
For ease of reference: to 2nd position.
3, 8–11. Vl: In the editions, Alard marks no bowing
Alard, Hermann except  on 10i; this may suggest that he expected
all chords to be taken as they come, but since that
# # 3 œ œœ œ1 1 ### 3 œœ œœ3 œœ3 3
1 1
would mean that the chord on 4i would come , this

1
& # 4 œœ œ œœ & 4œ œ
œ
œ p œ p is unlikely; more probably, he took  for granted and
wanted the single note on 10i also to be  for empha-
f f
David (Seybold) sis. Singer meticulously marks  on every note and

o chord from 3–11. David marks 3iii , but begins  on
### 3 œœ œœ4 œœ3
3 1
& 4œ œœ 8i, 10i and  on 9i, 11i, as do Auer, Brodsky, Rosé,

œ ≤ Seybold. In David’s personal copy (http://mhm.hud.
ac.uk/chase/view/pdf/298/1/) however, a pencil  has
f p
Rosé been written over the printed , suggesting that he

may have changed his mind. Joachim marks   on
### œœ 34 œœ≤ œ3
1
& œœ œ œœ 3ii, iii and the same as David in 8–11. Halir alone
-œ. marks -Ω
œ. œ. œ. œ. -œ -œ -œ -œ -œ. -œ. -œ. œ œ œ onœ 8i–ii,
œΩ œΩ 10i–ii
œœ œ Œ -œ -œ).
Ω Ω (with
p
&
f
5–13. Pno: Arpeggiation of chords at varying speeds
Singer according to context (including harmonic and bar
hierarchy, accentuation, and melodic expressivity
## œ œ œ œ
& # œœ œ24 &
œ œ and so on), would almost ∑ certainly
∑ have
∑ been∑ ex- ∑
œ p0 13 pected (5/c/ii). Cipriani Potter adds arpeggio signs
f
in his edition:
Joachim (Brodsky,∑Kreisler)
∑ ∑ ∑ ∑ ∑ ∑
& œœ n∑œ ∑ ∑. . ∑. ∑ ∑ ∑
## œœ œœ œœ œœ œœ.. œj œ œ œ œœ nœœj ‰ Œ Œ nœœ. œœ.
p
# # œ œœ œœ œ &# Û
∏∏∏∏∏
nœ œ œ œ # œœ œ
1 2
œ œ
1

& # œœ
2
∏∏∏∏∏∏∏∏∏∏∏∏∏
4
∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏
œ 4 fp
œ ≤ œ œ j œcres:œ
cres:
nn œ œ œ# œœ∑
œ œ̇ œ nn œœ. œ ‰ Œ Œ n œ œ
∑ ∑ ∑ ∑ ∑ ∑ ? ###∑ Û nœ
& œ ∑œ nœ ˙œ ∑ œ ∑ ∑ ∑ ∑
f p
.
œ œ
∏∏∏∏∏
Auer
## œ j
sfz
≤1
& # œ. # œ ‰ Œ Œ # œœ n œœ œ œ œ œ
∏∏∏∏∏∏∏∏∏∏∏∏
### œœ œœ2 œœ4 3

& ∑ ∑ ∑ ∑ ∑ œœ.∑ ∑ pœ cres:
œ n œœ∑ œœ n œœ n œœ∑ ∑ ∑ ∑
2 1
∏∏∏∏∏∏∏∏∏∏∏∏∏∏
& œ 24 1 œœ ? ###
œ p 3
fp
‰ Œ Œ
œ œ nœ œ # œ n œ
.
f
nœ # œ nœ
Halir & ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑
But these should not be understood as definitive, or
### œœ œœ3 œ4 3
2 1
1
& œ œ excluding arpeggiation where he did not mark it.

œœ
≤ -œ Potter’s markings are surely an indication of where
f p
he believed that Beethoven’s conception of ‘beauti-
The second and third bars, too, present obstacles ful’ performance absolutely required arpeggiation.
to a smooth legato between chords. All editors ex- Potter undoubtedly expected pianists to arpeggiate
cept Alard take a new bow after 2i. David, Alard, at other times according to taste and expressive pur-
Joachim, Singer, Rosé, Seybold mark no fingering pose. In 7 and 9, the portato articulation may be seen
here, suggesting that they envisaged 1st position as an explicit indiction of arpeggiation as instructed
throughout, which inevitably breaks the legato be- by Moscheles in his Studies for the Piano Forte op. 70,
tween 2iii and 3i. Hermann, who was already in his Bk 1 (1827), where he advises that double-notes
60s when his edition was published, is the oldest of and chords marked portato “should be struck very
the editors to suggest a different possibility, moving slightly in the Arpeggio manner, giving them the

same length of time as a dot under a slur requires.” Execution.146 Potter marks the chords in 27 and 35
In a footnote Speidel explains that the three chords with arpeggio signs.
in bar 5 “can only be effectively bound by extremely 24, 33. Pno, Vl: Speidel/Singer mark allargando here,
subtle treatment of the pedal”.144 anticipating Beethoven’s rallentando.
11–13. Vl: Some editors take two bows, some three for 27. Vl: It was here that Bridgetower made his fer-
Beethoven’s slur. mata embellishment in the premiere, apparently to
13–15. Vl: David marks all these figures , but from Beethoven’s delight. The account given in Thayer,147
13iv the  signs were removed in David-revised. is problematic, however, because it quotes Bridge
13–18. Pno: For the slurred duplet 16th-note figures, tower’s note as:
when the texture is chordal the first of each might at the repetition of the first part of the Presto, I imitated
be swiftly arpeggiated, the second not (5/c/ii). He the flight, at the 18th bar [of the Presto], of the pianoforte
would probably have expected the slurring to indi- of this movement thus:
cate a displacement of the metrical accent. œ œ
œœ œœœ œ œœJ
1 ma volta.
œ œœ
18ii, 36ii. Vl: David marks 0, but this was removed in
& œ œœœ œœ œœ
œ œœ jŒ
David-revised; it is not altered, however, in his an- œ œœœœœœ œ
u
notated personal copy. All the other editors envis- œœœœ œ“” œ œ
œœ œœ œœ j Œ
da
œœ œœ
2 volta.
age 3rd position. The open E-string would be very & œ œ œ œ œœ œ œ
plausible for a violinist of Beethoven’s time. œ œ œœ œœœœ uœ
18ii–27. Vl: The bowing marked by David, Alard, Sin
ger, Joachim-revised (Joachim’s original bowing is But Bridgetower wrote two different versions, one
ambiguous), Brodsky, Halir certainly indicates a for the first fermata in the exposition, therefore be-
martelé bowstroke near the point. That is by far the fore the piano’s embellished fermata, and a differ-
most likely bowstroke for Bridgetower and other vi- ent one for the repeat. If, however, as seems likely,
olinists of Beethoven’s time to have used, although Beethoven’s spontaneous embrace of the violinist
some may have tended more towards a détaché bow- occurred in rehearsal, when Bridgetower ‘imitated’
stroke in the upper half. David ensures  on the his ‘flight’. This would therefore have been during
whole-note in 25, by hooking 24iii–iv (this however the repeat. In that case Bridgetower’s music exam-
was removed in David-revised); Joachim-revised, ples would refer to what he did in the performance.
Brodsky hook 19ii into the  from 18ii and begin However, Bridgetower’s use of the word ‘imitated’,
20–24 . The bowing envisaged by those who mark may be misleading; he might well have taken the in-
nothing is impossible to determine; some may have itiative before hearing Beethoven embellish the fer-
started from 19ii with a tight martelé and broadened mata in 36, since embellishment of a fermata would
the bowstroke during the cresc. to move down the have been second nature to a violinist like Bridge-
bow for  in 25, while others may possibly have em- tower.
ployed an elastic bowstroke in the middle and lower The fragmentary autograph (in the Beethoven Haus,
half. Auer certainly played the staccato near the frog, Bonn MS NE 86), from which Beethoven almost cer-
since he marks  from 18ii to 19ii. Max Rostal still tainly played at the premiere, contains no embel-
recommended martelé.145 lishment of the fermatas in 27 or 36, so both players
19–35. Pno: All chords could be arpeggiated very presumably elaborated them spontaneously.
swiftly, particularly those marked sfp and sf, which The fact that Beethoven did not include a violin em-
will help to mitigate harshness in sound (5/c/ii; PT: bellishment of the fermata at 27 in the Stichvorlage
1/a/v). Beethoven marked various chords with ar- or edition, need not indicate that he did not want
peggio signs in this movement which might serve one; it would be entirely in the spirit of the time to
as exemplars for the addition of arpeggios in other leave it to the individual player. He provides none
places. Similar cases were examined by Otto Klau- in his Violin Concerto, nor does he provide one in
well, a former pupil of Reinecke, in his On Musical the second movement of Op. 47, where Czerny in-
cludes a different cadenza in each of his two piano
144 “Diese drei Accorde können nur durch höchst subtile Behand- 146 Klauwell: Der Vortrag in der Musik, pp. 110–115.
lung des Pedals wirksam gebunden werden” 147 Alexander Wheelock Thayer: Beethovens Leben, 3rd edn. (Leip-
145 Rostal: Beethoven, p. 137. zig, 1922), vol. 2, p. 392 f.n.

arrangements of the movement (see below). It is un-
likely, too, that Beethoven would have made his own fouetté; David marks  on 73ii and 77ii, but in his
embellishment identically both times. In a case of personal copy, crosses out his printed , presum-
this kind, strict observance of the printed text is ably envisaging fouetté like Joachim. Singer’s bow-
surely the last thing the composer would have ex- ing also brings him to  here.
pected or wanted, and it would certainly not be con- 75ii–80i, 80ii–81i. Vl: Slurs from the trill (surely be-
trary to his conception for a violinist to include an ginning from the upper auxiliary, 5/b/ii) to the fol-
arpeggiated elaboration of the C major chord. Had lowing main note are added by all except Alard,
Beethoven definitively not wanted an embellishment who slurs only to the end of the turn; Singer, Halir,
he would have needed to prevent it by some such Brodsky, however, offer a slur only to the end of the
term as come sta.148 turn as an alternative, and in 80 Singer marks a slur
Bridgetower, referring to his embellishment of the to the turn without alternative, specifying a separate
fermata, also stated that Beethoven “held the open bow for 81i. In the autograph fragment, however,
pedal during this flight, the chord of C as at the Beethoven wrote unambiguous slurs across the bar-
ninth bar [of the allegro (i.e. b. 27)].” This was pre- line in all these places except 80ii–81i which comes
sumably to allow the piano to resonate during the on a new page where also the continuation of the
violinist’s arpeggios and might well be employed slur from 79 is missing although it stretched well
with either a period or modern piano. into the margin on the previous page.
36ii– 4 4. Vl: David makes changes to the printed bow- 81–88, 402– 409. Vl: Beethoven’s unusual notation here
ing in his personal copy of his edition; most notably, leaves his expectations for the bowing ambiguous.
he adds slurs over the staccato in 42– 44 (the printed Without an autograph source that served as model
one in 44 in David-revised was absent from David’s for the Stichvorlage, it is difficult to determine his
original), clearly keeping the bow in the upper half. intention. While it is unusual for Beethoven to leave
36. Pno: Although Beethoven notated the embellish- out slurs in a string part where he envisaged them,
ment, pianists of his generation might well have he did not always specify the slurring meticulously
elaborated their own. Speidel marks veloce. Ganz in music of a virtuoso kind, as is very clear not only
marks brilliante. from the Violin Concerto op. 61, but also from the
37–89, 117ff, 156ff, 366ff. Pno: This is surely legato passages at 188f and 210–225 in this movement, where
probably with overholding (4/a/ii). Reinecke marks slurring seems indicative rather than prescriptive.
legato. Diémer marks slurs, though not on the fig- The occurrence of staccato marks regularly on i and
ure-types in 47. Speidel marks slurs from 37– 46, but v at 81ff and 402ff could either signify an accent
in 47 leggieramente (which for Beethoven probably (separation at this tempo is impossible) or may well
meant non-legato) with staccato on the main notes be merely an indication not to include these in a
continued for all similar figure-types. In 47 and slur. It is conceivable that he envisaged, or at least
similar Halir marks rh ii–v with a slur and v and left open the possibility of slurs on ii–iv, vi–viii here.
vii with tenuto line over a staccato dot. In 48, Halir All the editors except Auer, however, present the
marks i, iii, iv, vii with tenuto line over a staccato passage essentially as it is given in AG. Auer slurs
dot. Throughout this passage asynchrony between ii–iv and vi–viii in the first five bars, and i–iv, v–viii
rh and lh might occasionally be employed to help in the following three bars; it is quite possible that
bring out important moments (5/c/ii; PT: 1/b). In 156 he derived this from Joachim’s practice (despite Jo
rh Speidel marks slurs and in 158 legato. achim’s edition adding no slurs), because he cer-
45ii–iv, 49ii–iv, 55ii–iv, 59ii–iv. Vl: Here and in equiv- tainly studied the classics with him. Singer, who re-
alent places a slur is added by Singer (in brackets as tains separate bows, begins  on 81i and bows near
an alternative to staccato), Auer, Brodsky. Whether the point, allowing  on the long slur in 89f/410f;
this was an innovation, or a legacy of earlier prac- Alard, Hermann enable  at 89/410 by slurring v–vi
tice is uncertain. in the preceding bar. Others presumably played the
73ii, 77ii. Vl: Most editor begin  on 73ii and retake  8th-notes nearer the middle of the bow or moved
for 77ii. Joachim, however, begins  on 73ii, probably further towards the frog in the final bars.
148 For various implications of the fermata sign in Beethoven’s There are basically two approaches to fingering. The
lifetime see Brown: Classical and Romantic, pp. 589–598. majority seem to have envisaged a backwards ex-

tension of the 1st finger for the d #1, taking the e2 and been shared by many later musicians, and it has
d #2 alternatively with 3rd and 4th fingers from 81v– become customary to play a tone below the main
85i. From 81v–85iv, however, Joachim, Auer, Kreis- note (this is already heard in the 1918 recording by
ler finger each group of four notes 4 –3–2–1. Marjorie Hayward and Una Bourne and the 1940
89f. Vl: In David’s personal copy of his edition, he live recording of Béla Bartók and Josef Szigeti, al-
marks  in 90, deleting the printed  in 91, and makes though Bronisław Huberman and Ignaz Friedman
the same change at 410f by deleting the slur (com- recorded in 1925 play a semitone).
parison with 401f suggests that he had forgotten to 101–104. Vl: Singer, Auer, Brodsky shift to 4 on the
remove the slur across the barline from 80–81). G-string in either 102 or 103.
90. Vl: David marks the open string, a very plausible 107. Pno, Vl: Czerny states that it is only here that a ri-
choice for the early 19th century; this was removed tardando should begin. He does not mark one here,
in David-revised. All the younger editors except Sey- however, in his 1825 piano duet arrangement. But in
bold, who gives nothing, begin from 89 in 3rd posi- the parallel passage in the recapitulation he marks
tion. one at 433, three bars before Beethoven’s Adagio.
91–115, 412– 421. Pno, Vl: Beethoven’s dolce might in- 107–116. Vl: It seems highly likely that early performers
spire a very slightly broader tempo and the use of of this music would have employed harmonics for
the moderator if available (PT: 3/b). It also invites some or all the e3s in this passage. Only David (who
noticeable arpeggiation, perhaps at moderately fast probably regarded harmonics as obvious), Joachim
speeds, though of course varied according harmonic (who may also have taken them for granted in this
importance (3/d/i and 5/c/ii). Potter marks sostenuto context), Seybold (following David), Rosé, Kreisler
at 91 (412) which indicates that he thought the tempo (both of whom mark the passage in 6th position
should be slightly broadened here. Speidel/Singer from 107, presumably for the sake of vibrato) do not
mark poco più tranquillo (Singer already from 410 in mark them. Auer, Joachim-revised mark them only
the later passage). These markings may be precur- in 110, with 3 in 114. Brodsky, Halir mark 110i, 114i
sors of a later 20th-century tendency to take this to be played with 2nd finger as a harmonic on the
theme much more slowly than the prevailing tempo A-string. The harmonic with 4 is also marked or
of the movement, which would certainly not have clearly intended in 115 by Alard, Hermann, Singer,
been Beethoven’s intention. probably by David, and possibly by Joachim, fol-
In his personal copy David divides Beethoven’s slurs lowed by 4 in 116. Singer, Auer, Joachim-revised also
as do Hermann, Alard, Brodsky, Halir, Kreisler. mark open strings on 109ii, 113ii. A similar proce-
All keep the first passage from 91 on the D-string dure is followed on the A-string at 428ff.
and from 412 on the A-string. No portamento is 109, 113. Vl: In his 1825 piano duet arrangement, Czerny
marked in 91–100 by Joachim, Singer, Auer, Kreisler, writes out a turn in 109 and a trill with upper-aux-
who give the whole passage in 2nd position. Others iliary beginning and turn in 113. These would be
shift several times, notably 95f and 99f; at 412– 421, potential ornaments for the violinist to add on the
all except Alard, Rosé, Seybold leave the passage in repeat of the exposition.
1st position throughout. 116ii–117i. Vl: An open string followed by a harmonic is
95, 416. Vl: There can be little doubt that Beethoven given by David, Alard, Singer, Auer, Brodsky, Halir,
intended a semitone below the main note for the Seybold. Nothing is marked by Hermann, Joachim,
turn here and all the editions except Halir print but the harmonic is added in Joachim-revised. This
this. Speidel/Singer, however, give a footnote: “Here, treatment would surely have come naturally to most
and similarly in the later parallel passage, we have violinists of Beethoven’s time and may even have
to suggest that the player use a simple # instead of been envisaged by the composer.

the , which might sound a bit too hard for some 119ff, 439ff. rh: The arpeggio figures probably require
ears.” 149 Halir specifically marks # at 95 and gives legato with overholding (4/a/ii). Speidel marks slurs.
no accidental at 416. Their feeling seems to have 120ii–121. Vl: Singer, Auer, Brodsky shift up the A-
string.
122ii–124i. Vl: Alard, Singer, Auer, Halir, Brodsky mark
149 “Wir müssen es hier, wie bei der späteren Parallelstelle, dem

Spieler anheimstellen, anstatt des , welches manchem Ohre etwas this phrase to be played on the G-string, Alard,
zu hart klingen dürfte, ein einfaches # anzuwenden.” Auer, Brodsky, Joachim-revised with harmonic g1.

Joachim, however, marks it with 1 on 123i, therefore the lower octave grace-note, but only in 221 where
on the D-string. Either reading might have been an accidental A b is essential.
employed by an early-19th-century player. ? Ÿ
r
128ii–130ii. Vl: A harmonic is marked by Alard, Sin rbœ w œœ bw
bw
ger, Seybold and those who mark nothing (David, œ
Joachim) may well have regarded it as obvious. He also marks the upper auxiliary on the similar
Auer, Brodsky, Halir, Kreisler, however, certainly trill in 472 and 484 (in the latter only on the trill
envisaged a stopped note. None of the editors mark in secondo, not the one in primo). The grace-note
three successive down-bows for the sfs. Some start would almost certainly have been aligned with the
, some  ; David changed his mind, altering the lh (5/a).
printed  to  in his personal copy. The up-bow sfs 153iii, 165iii. Pno, Vl: Speidel/Singer point out the me-
were probably played fouetté. lodic discrepancy here with the parallel passage at
132–134, 144ff, 465ff. Pno: The octaves in rh might be 474iii, 486iii and enclose the # in brackets, giving
best brought out through the application of asyn- the following footnote: “In the parallel passage of
chrony or arpeggiation (5/c/ii), which would miti- this splendid motive, the step of a minor second
gate the hardness that might otherwise arise. has become a major one; whether with the Master’s
140. Pno: Ganz in a foot note explains: intention, we must leave open. Nevertheless, in our
œ .
. opinion, the major second is much better suited to
? œ œ nœ # œ
*) The editor plays unbroken octaves:
œ nœ. # œ etc.
its grandeur than the minor one, and so we would
like to recommend the superior characteristics of
.
141–176. Pno, Vl: Speidel/Singer mark allargando at 141, the former to performers.” 151
then at 143 a tempo and, in Pno, “forceful/heavy” 160. Vl: A harmonic is specifically marked by Singer,
(wuchtig), which is then marked in Vl on the upbeat Seybold; David, Joachim, who mark no change from
to 156, where Pno has marc[ato] molto; this evidently 3rd position, surely considered it obvious; Joachim-
implies a holding back of tempo, for at 168 a foot- revised, however, marks a shift to 5th position on
note states: “These 4 bars involuntarily drive back 159iv.
to the original tempo.” 150 And they mark a tempo at 163, 484. Vl: Czerny notates an upper note start and
172. The same procedure is followed from 463– 497. a turn following the trill in his 1825 piano duet ar-
The 8th-note up-beats in 145ff should certainly be rangement, and in his arrangement for cello and
left late and perhaps sometimes even played almost piano, Czerny also adds the upper auxiliary to the
like grace-notes before the beat, but powerfully. In trill in the cello part.
such contexts, up-beat figures and short notes fol- 166ii. Vl: David, Joachim, Singer, Auer, Halir, Seybold,
lowing a dotted note were conventionally played in Kreisler mark 0.
this manner in Beethoven’s time. 172. Pno: Ganz explains: “The editor again plays two
144. Vl: Alard, Hermann, Singer, Halir mark 0; prob- bars of unbroken octaves in the left hand.”
ably all the others regarded it as obvious. 178ff, 499ff, 583ff. Pno: According to Czerny in his
149, 153, 165ff, 469ff. Pno: For such slurred duplet 8th- Pianoforte-Schule op. 500 (1839), “all chords consist-
notes, it was characteristic throughout the 19th cen- ing of very short notes” should, unless expressly
tury to play the first note stronger and longer than marked by the composer, be unarpeggiated. But this
the second (2/a). might not have precluded extremely swift arpeggia-
151, 163, 217, 221, 472, 484. Pno, Vl: In these bars in the tion, the type that Thalberg in L’Art du chant (1853)
piano part, Beethoven wrote the bottom octave (fol- described as presque plaqué, or almost together (5/c/ii).
lowing on from the octaves in the preceding bars) 188v. Vl: sf added in AG and all the editions.
as a grace-note to allow the performance of a trill in 188v–189viii. Vl: These notes, left by Beethoven (or
the upper part. He marked no upper auxiliary, but the copyist) with neither slurs nor staccato were re-
almost certainly expected one. Czerny, in his 1825 151 “Bei der Parallelstelle dieses herrlichen Motivs ist aus dem
piano duet arrangement marks one, in addition to kleinen Secundenschritt ein grosser geworden; ob mit Absicht des
Meisters, müssen wir dahingestellt sein lassen. Gleichwohl ent-
spricht unseres Dafürhaltens die grosse Secunde ungleich besser
150 “Diese 4 Takte drängen unwillkürlich nach dem ursprüngli- der Grossartigkeit desselben als die kleine, und so möchten wir
chen Tempo zurück.” den Ausführenden die Leitereigenheit der ersteren anempfehlen.”

garded by all editors as requiring some form of start to the trill only in 221, where a b (omitted in
slurring. David, Alard, Singer, Halir, Rosé, Seybold the sources) is required, but it was surely also envis-
merely continue the pattern established on 188i–iv. aged in the other bars, where no accidentals were
Hermann slurs all in pairs (a very plausible early- needed.
19th-century solution), while Joachim, Auer, Brod- 224iii–viii. Vl: In David’s personal copy he replaced
sky use various patterns of slurring in groups of the separate notes with a slur. In David-revised, this
four across the beat, a bowing that was also em- bar was given staccato marks on iii–iv, vii–viii and
ployed by Beethoven’s contemporaries; there is an a slur on v–vi.
example, for instance, in Franz Clement’s Violin Con- 226ff. Pno: It is uncertain what Beethoven expected
certo in D of 1805 (1st movement, bb. 277f). here in the way of articulation, but it is likely to
189–191. Pno, Vl: Speidel/Singer mark allargando and have been legato. Speidel, Diémer marks slurs. Vog-
’ after 191. rich marks legato. But Ganz marks non legato.
190. Pno: This chord would almost certainly have been 228ii–iii, 232ii–iii, 240ii–iii. Vl: Early-19th-century vio-
arpeggiated either in the normal fashion (5/c/ii) or linists may well have minimised shifting by play-
perhaps by playing the lowest note in the lh on the ing some or all of these semitones successively with
beat and the other notes in lh and rh together and the 4th finger. In 240 Auer marks 4 – 4; and it is im-
slightly after (PT: 1/a/v). Potter marks an arpeggio plied in David, Joachim, who mark no change of
sign. position. Neither David, Joachim, not Auer mark
192. Pno: This chord could be arpeggiated gently and anything in 228, suggesting that here too they may
fairly slowly (5/c/ii and PT: 1/a). have expected 4 – 4. Most, however, mark 3– 4 in 232.
210–225. Vl: Beethoven left this passage with only a 234. lh: Ganz marks marc.
few sporadic slurs and staccato marks, but he cer- 245f. Vl: David, Alard, Halir, Rosé, Seybold mark 0–o
tainly did not expect a violinist to play the passage (the harmonic with 4) and 4 again on 246i and ii.
as it stands. As in the Violin Concerto, he seems This is a typical early-19th-century practice, involv-
to have recognised that it was appropriate to leave ing minimum changing of the left-hand position.
it to the player to find a suitable way of deliver- To execute it effectively it is probably best to keep
ing ‘technical’ passages of this kind effectively. The the whole hand in a firm shape and move it as if
19th-century editors provide various solutions, de- from 4th to 5th position, but without shifting the
riving from their own traditions and preferences. position of the thumb. Other editors give 0–3– 4.
Only Alard preserves Beethoven’s slurring in 211 258–269. Pno: All chords might be swiftly arpeggiated
and extends it in the next two bars; he also follows (5/c/ii). In 266, Speidel marks poco marc.
the original in 214f and then mixes slurred pairs 270. Pno, Vl: Speidel/Singer mark poco animato.
with separately bowed notes. Singer’s approach is 274 –277. Vl: The majority of editors remain in 1st posi-
similar, but he never slurs more than two notes and tion, evidently using 4 for 275i (marked by Joachim).
uses more slurs than separate notes; Hermann’s so- This would most likely have been the choice of early-
lution contains element of both these approaches. 19th-century violinists. Different fingerings to avoid
An anonymous violinist, who provided slurring and the diminished 5th are suggested by Singer (4th posi-
staccato for the edition of this sonata in Ludwig van tion in 274 and 2nd in 275) and Auer (1st position in
Beethoven Erste vollständige Gesammtausgabe unter Re- 274 and 2nd in 275). Another alternative fingering is
vision von Franz Liszt (Wolfenbüttel, L. Holle [c. 1860]), suggested in the edited violin part.
marked alternating two-note slurs and staccato in 274 –277. Pno: Ganz marks senza Pedale.
all bars except those where notes iii–iv were re- 274viii, 283viii. Vl: In his personal copy, David
peated three times, where slurs on i–ii and vii–viii changed these notes respectively from a 2 to f2 and
are marked (215 was misprinted in the violin part, c2 to a1, presumably to avoid the diminished 5th with
but is correct in the piano score). Other editors fol- the following note! Such minor changes for tech-
low David in employing a significant number of nical convenience would probably not have been
four-note slurs across the beat, similar to those in uncommon in Beethoven’s time, but would already
188f. Beethoven would probably have been perfectly have been frowned upon later in the century.
happy with any of these solutions, well executed. 288ii, 292ii. Vl: David, Hermann, Joachim, Auer take
217, 221, 224f. lh: Czerny marks an upper auxiliary the , surely envisaging a fouetté bowstroke.

293i, 300i. Vl: Alard, Singer, Brodsky, Seybold mark corde to una corda to enhance the cresc., would be
harmonics. Others who indicate no change from effective (PT: 3/a).
3rd position probably regarded a harmonic as obvi- 313–324. Pno: See 310 above. In 324, Potter marks an
ous. arpeggio sign.
294. Pno: Ganz suggests the playing of unbroken oc- 325–362. Pno: The arpeggiation of various chords
taves. would be appropriate (5/c/ii). In 326, 333–334 and
295. Vl: Alard, Hermann, Brodsky, Halir indicate 1st 354 –355 Potter marks arpeggio signs.
position with open E-string. 320f. Vl: Some stay in 1st position others go to 5th
300–303. Pno: Overholding would have been expected position on vi.
(4/a/ii). 321v–322i. Vl: Editors suggest various shifts some us-
300ii–304i. Vl: David, Alard, Joachim, Auer, Halir, Rosé, ing 4 – 4 for an expressive portamento on 321vi–vii
Seybold mark the passage in 4th position without (David, Singer, Auer, Halir, Rosé, Seybold), some
shifts. Hermann, Singer, Brodsky, Kreisler take the avoiding portamento by staying in position until
opportunity to introduce expressive portamento, 322i (Alard, Hermann, Joachim, Kreisler).
using the open E-string and shifting afterwards, 364ii, 365i. Vl: Alard, marks a harmonic in both bars,
Kreisler uses the David fingering in 300–302i and Hermann, Brodsky only in 365.
the portamento fingering only from 302 (though in 402– 409. Vl: See note to 81.
his 1935 recording he makes the portamento both 410– 421. Vl: The melody lies easily in 1st position,
times). Singer marks the passage appassionato and and most editors offer no fingering suggestions; but
uses a particularly prominent shift from 1–2 in 301, Alard, Brodsky, Rosé suggest a variety of expressive
303 (a fingering also given by Brodsky) and em- fingerings, and Singer marks it to be played entirely
phasises the portamento with a connecting line. It on the D-string.
is likely that, for the repetition of the motif from 428– 435. Vl: See note to 107–116. On this repetition of
300ii–302i in 302ii–304i, Bridgetower and other fine the theme, Czerny, in his 1825 piano duet arrange-
violinists of Beethoven’s time would have used a ment, replaces the violin line with a four-bar trill,
different fingering, or at least a different charac- followed by a four and a half-bar trill on a3, begin-
terisation of the phrase, probably more expressive ning the first with the upper auxiliary and with a
(with the fingering above the stave in the edited turn at the end of each trill.
violin part of the present edition, the shift from 1– 4 433. Pno, Vl: Czerny marks ritardando in his piano
might be made lightly in 301 and more intensely in duet arrangement.
303). It is also unlikely that, in practice, like Kreisler, 436. Vl: Czerny gives a simple embellishment at the
the other 19th-century editors would always have end of his long trill:
played the fingering exactly as supplied in their
˙. #œ œ œ œ. œ.
. 3
editions. The modern player may like to select any &
effective combination of the various fingering pos- w Adagio
sibilities.
& w
cresc:
304 –311. Pno: In 314, where Vl has the similar mate-

rial, the rising arpeggio figures are unslurred and, A violinist might have played something similar,
just as in Pno, slurs are only marked on the final one for instance:
and a half bars. This strongly suggests that Beetho- 12 U
ven did not envisage a continuation of the legato w œ œ #œ œ œ œ œ n˙ .
&
marked in 300–303, but a non legato articulation of
the notes (as far as that is possible at the rapid tem-
po) until the marked slurs in 310–311. Reinecke, Spei- 443f. Vl: David added an effective fingering in his
del, Diémer, continue with slurs. But Ganz marks personal copy, remaining in 6th position and tak-
non legato. ing the a2 in 444 as a harmonic with 2nd finger.
307i–ii. Vl: A portamento is indicated by the finger- Hermann and Singer also cross to the A-string, but
ing of all the editors except Brodsky. in 5th position. Others descend to 3rd on v.
310–313. Pno: Overholding (4/a/ii), with swift arpeg 463ff. Pno, Vl: See note to 141, 143, 155ii.
giation and perhaps with a shift of colour from tre 484. Vl: See note to 163.

489i, 491i. Vl: Alard, Hermann, Singer Brodsky, Halir longer slurs 4, 4, 2, 2 bars ending on 171i, Singer,
mark a shift to 5th position on 489i; all the other Kreisler mark two 6-bar slurs.
editors mark no change from 3rd position, leaving 571. Vl, Pno: Czerny marks rallent. here in his arrange-
it likely that they assumed a harmonic e3 with 4th ments for cello and piano and piano duet. Speidel
finger marks tranquillo.
489. Pno: Speidel advises in a footnote: “These 4 bars 574 – 576. Vl: All editors mark this with 2 in 574 and 2
involuntarily push back to the original tempo”.152 in 576, evidently expecting a portamento shift. All
493i. Vl: David, Singer, Brodsky, Seybold print an open mark or assume  except Auer.
E-string instead of Beethoven’s e3! Joachim gives e3 575–581. Pno: All chords might be arpeggiated. Potter
as a harmonic, but the o was removed by Moser in marks arpeggio signs in 579–581.
Joachim-revised. 584iv–v, 588iv–v, 592iv–v. Vl: All the editors mark 4 – 4
514 – 517. Pno: All the chords marked sf might be ar- in this figure.
peggiated swiftly to produce a fiery energy as well 585ff. Pno: It is not clear what Beethoven expected here
as to mitigate harshness (PT: 1/a/v). in terms of articulation. Legato seems probable but
518–533. Pno: Undoubtedly legato. Diémer marks slurs. non-legato would also work. Reinecke marks legato.
Speidel marks leggiero. Ganz marks non legato. He 594. Vl: Singer, Auer take this bar entirely on the G
also marks una corda and senza Pedale in 518 with string; this is also marked in Joachim-revised, but
tre corde in 530. At 518 Vogrich offers an alternative: not in Joachim’s original edition. This fingering
would surely have been an unlikely choice for a vio-
? œbœ# œ œnœ œ œ œ
linist in Beethoven’s Vienna.
oder:
 ? bœ# œ œ œ œ œ
œ nœ Andante con Variazioni
Tempo
546iii–549ii. Vl: Alard marks 0 on 546iii, o on 546iv The editors show rare unanimity in their suggestions
and 4 on 549ii, evidently intending the harmonic on for the tempo of the theme. However, in the light of
all the e3s. The use of the harmonic in this manner Beethoven’s own metronome marks for the compar
would probably have been a first choice for many able 2/4 Andante cantabile in the String Quartet op. 18
violinists of Beethoven’s time. Some of the editors, no. 5, also a set of variations, to which he gave  = 100,
particularly the older ones, who mark no alterna- it seems likely that the 19th-century metronome marks
tive fingering on, or immediately before or after for the Andante in op. 47 are somewhat slower than
these notes (David, Joachim, Auer, Brodsky, Rosé, he envisaged; perhaps closer to poco adagio. For the
Kreisler) may have intended the same as Alard, al- Poco adagio at the end of Variation 5 in the quartet he
though the younger ones, especially, are likely to gave  = 88. Overall, the range of note values in the
have assumed the continuation of 1– 4 for the oc- quartet, which has two variations with a substantial
taves, as marked by Singer and Halir. number of 32nd-notes, is similar to that in the sonata,
538–546. Pno: Beethoven may have intended staccato although in in Variation 4 in the sonata there are sex-
to continue. Speidel marks sempre staccato. tuplet 32nds and a few fioriture in 64ths. Since Beet-
547. Pno, Vl: Speidel/Singer mark largamente, with a hoven did not indicate a slower speed for Variation
tempo at 559. 4 in the sonata, it might suggest a somewhat slower
547–570. Pno: Arpeggiation to produce energy and fire overall tempo for the whole movement than in the
as well as to fill out the texture would be appropri- quartet. On the other hand, in the context of a sonata
ate (5/c/ii). Ganz marks espress. Potter marks arpeg- “written in a very concertante style”,153 Beethoven
gio signs in 547i, 548i and 549i. may well have intended Variation 4 to have precisely
559–570. Vl: AG and most editions slur 559ii–viii (a the virtuoso character that is warned against by Spei-
few follow the 1st edition in slurring from i). In the del/Singer (see below). It is probable, however, that
following bars AG and most editions give slurs over Beethoven expected each variation to have its own
i–viii. David, Brodsky, Halir, Rosé, Seybold mark fundamental tempo, not significantly faster or slower
than the theme, but adapted to its own note values
152 “Diese 4 Takte drängen unwillkürlich nach dem ursprüngli-
chen Tempo zurück”. 153 “scritta in un stilo molto concertante”

and musical characteristics. The metronome marks 1ff. Pno: Arpeggiation of all chords not marked stac-
suggested by Speidel/Singer show that this approach cato is a distinct possibility and stylish (5/c/ii).
was not unfamiliar to musicians born around the time Potter marks arpeggio signs on 4i, 8 rh ii, 12 rh i,
of Beethoven’s death, and it would accord well with 21ii, 31 rh i, 35 rh i, 54 rh i, again not to be taken as
late-18th- and early-19th-century writings about tem- absolutely prescriptive but suggestive and perhaps
po. What is curious about the metronome marks allot- to be used as a model for further application of ar-
ted by Speidel/Singer is that they recommend a faster peggiation. Slight arpeggiation for the portato in
tempo for Variation 4 than for the theme. Had Beet- 2–3 is appropriate as recommended by Moscheles
hoven himself provided a metronome mark for the in his Studies for the Piano Forte op. 70, Bk 1 (1827),
theme, it would most likely have been in the region where he advises that double-notes and chords
of 96–100. The edition by Cipriani Potter, who knew marked portato “should be struck very slightly in
Bridgetower, as well as Beethoven, gives 96 for this the Arpeggio manner, giving them the same length
movement. of time as a dot under a slur requires” (5/c/ii and
Var. 1 Var. 2 Var. 3 Var. 4 see note 5–13 in the first movement above). Halir
Haslinger  = 88 changes dots with slurs to tenuto lines with slurs
Moscheles  = 88 perhaps to encourage notes as long as possible.
Czerny Vortrag  = 88 1, 2. Pno: The notation suggests a smooth connection
Alard/Diémer  = 88 between ii and iii; the portato begins only in 2 be-
Speidel/Singer  = 84 - 96 92 88 tween the final note and the beginning of the next
Kreisler  = c. 92–96 c. 100 c. 108 c. 92 c. 80 bar.
Czerny advised: ”Everything that can create a sing- 5ff, 13ff, 32ff, 51ff. Pno, Vl: This passage comes four
ing, expressive, but not sluggish performance, must times. In 13 only is there a crescendo instruction
be used to render the beautiful theme appropriately. and only at 8 and 34 is there a > , solely in Pno.
The successive trills in the second part are perfectly Whether Beethoven simply overlooked discrepan-
legato, crescendo, and are to be performed distinctly cies or whether he specifically wanted different
with the 3rd and 4th fingers (whereupon the thumb treatments of the dynamics is indeterminable, but
comes on the 2nd of the small notes). The first varia he will certainly not have wanted an identical treat-
tion a little livelier, well-marked, and the triplets in ment on each occasion. Taken literally, the score sug-
both hands staccato. The 2nd Variation very light gests a sf within piano in 6/33 and a diminuendo in
and piano, and following the violin in all its nuances. 7/34; at 14 a sf within forte and a forte continuation to
The 3rd Variation extremely legato and with serious 16; at 52 a sf within piano and a piano continuation.
expression, but animated, otherwise it would seem In practice, performers will decide for themselves
dragged. The 4th variation with the tenderest delicacy how they want to shape these phrases.
and the ornaments light and smooth at the tempo of 7, 15, 34, 53. Pno, Vl: Speidel/Singer illustrate the trill as:
the theme. Take care with the pedal everywhere, as it 9
is essential.” 154 &b œœ œœœœ œœœ

1. Pno: Speidel/Singer add sostenuto e molto cantabile,
which corresponds with their slightly slower met- Reinecke/Hermann and Grützmacher also add a
ronome mark. turned ending. In 7 Ganz explains in a footnote:
154 “Alles, was ein gesangreicher, ausdrucksvoller, aber ja nicht The editor plays: & b œ œ œ œ œ œ œ œ œ œ œ œ5 œ
schleppender Vortrag bewirken kann, muss angewendet werden, 3
1 3
1 3
1 3
1 2 1 2 1
um das schöne Thema entsprechend auszuführen. Die Kettentril-

ler im 2ten Theile sind streng gebunden, crescendo, und mit dem
And in 15 Ganz explains:
3ten und 4ten Finger, (worauf der Daumen auf die 2te kleine Note 3 3 3 4
kommt,) deutlich vorzutragen. Die erste Variation ein wenig beleb- ? b œ1 œ œ1 œ œ œ œ œ œ2 œ1 œ2 œ œ
ter, wohl markirt, und die Triolen in beiden Händen staccato. Die 2te
Variation sehr leicht und piano abgestossen, und in allen Schattie-
rungen der Violine folgend. Die 3te Variation äussert [sic] legato und
mit ernstem Ausdruck, aber belebt, da sie sonst gedehnt erschie- They illustrate, however, a main note start, which
nen würde. Die 4te Variation mit der zartesten Delikatesse und die
Verzierungen leicht und gerundet, im Tempo des Thema. Das Pedal contradicts Czerny’s instructions in his Pianoforte-
überall wohl beachtet, da es wesentlich ist.” Schule (vol. 1, p. 131), where he states unequivocally

that where a trill is preceded by a note at the same then marks a harmonic on 14iii, presumably coming
pitch it must begin with the upper note. This rule back to 4th position, and then, since he writes a slur
seems to have remained constant from his practice from 15i–iii, changing the finger on the e2 from 1–2;
in Beethoven’s lifetime (but see below for circum- these markings, which are given in the edited vio-
stances in which his treatment of trills changed after lin part of the present edition, are repeated at 33f.
the 1820s). Usually, Czerny did not trouble to notate Each of the editors provide essentially the same fin-
an upper appoggiatura in such circumstances. He gering for all three appearances of this material.
notates the expected turn consistently in all three An imaginative violinist of Beethoven’s time or the
arrangements, but provides no appoggiatura for the later 19th century may well have varied the finger-
trill beginning in his arrangement for piano solo or ing on repetition of the thematic material to provide
for cello. In his piano duet arrangement, however, contrasting expressive effects.
perhaps as a reflex action, he included an initial ap- 10f. Pno, Vl: There is a discrepancy between the phras-
poggiatura, but only in the secondo part at b. 53. ing of the piano and violin parts in the sources: Pno
And in the solo arrangement, a fingering, 4 –3, dem- has portato from the last 8th of 10 to the end of 11
onstrates an upper-auxiliary beginning at b. 34 (see here and also at 12f, while Vl has a smooth slur from
Table 1). ii–iii in 10 and portato from i–iv in 11. The portato
Elsewere he probably regarded it as obvious because dot was added in AG and all editors except Alard,
the preceding note was at the same pitch. Brodsky added one here; Hermann also starts the
8–16, 27ff, 47ff. Vl: All editors except Hermann continue portato slur on 10iii and Halir marks the portato
on the A-string until 14i; some go to the E-string on with –. All these additional portato dots (between
14ii others remain until 15iv; Alard remains until slur and note), however, imply separation between
16. Joachim, Singer, Auer, Brodsky, Halir, Kreisler ii and iii, which is not indicated by the portato in
go to 5th position in 12, returning to 3rd position Pno, which only begins with separation after the
with an expressive portamento on 13ii. David, Sey- final 8th of 10. See note to 1, 2 above.
bold take the passage in 4th position, returning to 11iii–12i. Vl: David, followed by Seybold gives three
3rd on 13ii. A harmonic is marked on 13i by Alard successive notes with the 2nd finger.
(with an extension from 3rd position) and Singer. 12, 31. Vl: Auer marks ’ at the end of the bar.
Hermann starts the movement in 3rd position, but 13ff. Vl: See note to 5ff, 13ff, 32ff, 51ff.
goes to 1st on 9ii and remains there, with 0 on 11iii, 15vi–16i. Vl: Singer shifts 2–2 up the A-string, chang-
presumably until the trill (though he marks no fin- ing bow, but marking a connecting portamento line.
gering there). David, having gone to the E-string at 24 –26, 43– 45. Pno, Vl: Czerny, in his piano duet ar-
14ii, changes this in his personal copy to a rather rangement, gives these passages more or less ex-
elegant fingering, going to 5th position for 14ii, he actly as they appear in the 1st edition.
1st edition piano part, bb. 7ff Czerny’s solo arrangement, bb. 7ff Czerny’s duet arrangement, bb. 53ff
Primo
Ÿ
œ.
Ÿ nœ œ œ œœ œ œ œ œ œœœœ œœj “” Ÿ œ.
& b œ œ œ œ œ œ œœœ œœ ‰
j & b œœJ œ œ
J œœ œ œ œ#œ œ œ œœœ œœ
&b œ

œ œ

œ J

j Ÿ J
? b œj œœ œœ œ œ œœ œ œ œ œ
œœ œ œ œŸ#œ œ œ œœœ œœ
? œœ œœ œœ œœ œ œΩ œΩ œΩ ‰ œ &b œ
.
b J œ. œ œ. œJ œ
œ . J
Solo arrangement b. 34 Secondo

Ÿ
œ
nœ œ œœ 4œœ3œ œ œ œ œœ
&b œ
œ œ rœ œ
? b œ œ œ œ #œ œ œ œ œœœœ œ

J

1 Ÿ
? œœ œ œ œ œ ?b œ œ œ œ œ œ œ
b œ œ œ œ. œ œ.
Table 1 œ œ œ œ .

In his arrangement for piano solo, however, he gave 49. lh: The grace-note would probably have been ex-
fingerings, which demonstrate that he expected all pected to align with the beat. Speidel gives the fol-
the trills to begin with the upper note in 24f and lowing realisation of the trill:
43f, except in 26/45, where he gave a grace-note pre-
?b
ceding the first trill, to show that these trills, fol- j
œœœœœœœœœœœœœœ œ
lowing a note pitched a tone higher, were to begin
œ
from the main note. But Czerny almost certainly expected an upper ap-
poggiatura, as in the first movement at b. 221.
–
œ2 .. œ1 œ1 Ÿ Ÿ Ÿ œŸ œ Ÿœ œ œ œŸœ œ œ2 .. œ1 œj œŸ œ œ2 .. œ Ÿœ œ œ.. œ
5
4
35 43 43
n 32 4 3 3 2 4 3 5 4 4 3 5
2 1 2 4
& b nœ œ œ œ#œ œ œ#œ œ nœ œ œ nœ œ œ nœ œ œ œ œ œ

2 1 2 1 1 2 3
Var. 1
1 2
œ œ

55ff. Vl: Hermann, Singer, Auer, Halir, Rosé indicate
j œ œ œj j sf j
sf sf p sf
slurred staccato/portato on some or all groups. Au-
? œœ ˙ œ œ œ œœ. œ œ œ
œ œ
œ
œ œ
b J œ er’s pupil Efrem Zimbalist uses this bowing in his
J
1926 filmed performance with Harold Bauer.
54 – 56, 68–69, 74 –77. Pno: The octaves and chords (es-
In 26, 45 the turns to the trills (absent in 26 in the pecially those marked sf) might be arpeggiated or
sources) are also present in Czerny’s piano duet ar- played asynchronously with the bass, which would
rangement. Corresponding with later practice, Rei- help energise while mitigating any hardness (5/c/ii;
necke, Speidel, Diémer, Ganz, Vogrich give main- PT: 1/a and PT: 1/b). In 55 rh iii and 59 rh iii marked
note start fingering. sf, Speidel remarks: “The accents at a) and b) can
38i–ii. Vl: All editors except David, whose fingering is only apply to the right hand, as with other similar
evidently incomplete here (the omission was recti- places”.155 Curiously, Potter removed Beethoven’s trill
fied in David-revised), shift 1– 4; Singer emphasises signs on 54 rh iii and 55 rh iii and replaced them with
the portamento with a connecting line. > . He evidently saw this as a viable alternative.
43f. Vl: Two basic approaches to the execution of the 54 – 56, 74f. In his solo arrangement Czerny did not
trills are given in the editions. The majority shift notate turns to the trills, but they are present in
predominantly during the turns (David, Joachim, his piano duet arrangement. He did not mark an
Auer, Halir, Rose, Sybold, Kreisler); Alard, Singer, upper-auxiliary start to the trills in 54 –56 in any of
Brodsky apparently expect the shift to occur after his arrangements, but it is probable that they were
the turn. envisaged. In 74f, in his piano duet arrangement, he
47– 49. Vl: In his solo piano arrangement Czerny no- marks both upper auxiliary and turn.
tates the appoggiatura in 47, 48 as a full-length 16th- Secondo
note, but it is omitted entirely in 49. Ÿ
r j r Ÿj
œ. œ œ œ œœ j œ œ œœ
& b œœœJ œœœ œ
œJ
œœ œœ
œ œ œJœ
œœj œœj œœ.

œœ œœ œœ œ œœ œœ. œœœ. œœœ
& b œœJ œ œJ œ
.
œJ œJ œ œ œ œ
p

? b œœ œ œ œ œ œ œ œ œ œœ œ œ œœ œ œ œ œ œ œ œ œ œœ œ œ
........... .
? b œj j j j Ÿ œ ‰ J ‰ J
œ œ œ œ œ j œ. œ œ œœ œœœ
˙ ˙ œ
In 54, Ganz explains in a footnote:
œ1 œ3 œ1 œ3 œ œ œ œ3 œ2 œ1.
In his piano duet arrangement they are given as The editor plays: &b J
full-size 16th-notes in all three bars.
“” But this is surely anachronistic.
bœœ œœ œ. œ œ
œ œ œj œœj œ œ œ œœ œœ 57–62. Pno: Speidel marks poco string. In the second
. . .
œ œ. œœ œ œœ
& b œ œ œ. œœ œœJ œœ œœ œJ œ œJ œœ œœ œœ œ

J J half of 57, with poco rit. at the end of 58 and a tempo
at the beginning of 59. In 59, he also marks con molto
fcres:
œœ œ œ. œ œ
p fz
œœ j j
cres:
fz
œ œ œ. œœ œ œœ œœœ œœ
. . .
& b œ œ œ. œœ œœ œœ œœœ œœ œ œœ œœœ
œJ œ J œJ œ œ œ espressione. He marks poco rit. at the end of 61 and a
J J tempo at the end of 62.
155 “Die Accents bei a) u. b) können natürlicherweise, wie bei der
Wiederholungsstelle, nur für die rechte Hand Geltung haben”.

59ff. rh: The short trill sign would almost certainly 100–102, 105–107. Czerny in his 1825 piano duet ar-
have been interpreted as a Schneller or snap (5/a/iv), rangement gives the following reading of 100–102:
a quick ornament starting with the main note (on n œ œ œ œ. œbœ œ. œ. # œœ œ. œ. # œnœ nœ mœ . . . . . T. >œ . . Ÿ
.
œ œœœ# œœœœœœœœ nœœœ œœœœ . . nœœœœ
the beat) moving to the upper auxiliary and then &b
..
back to the main note. Speidel gives the following dim calando p smorz
realisation:
1 4 He thus replaces Beethoven’s cresc. in 100 with dim.,
œ œ œ nœ œ œ œ œ
4
3 2 1 3
œ.
2 1
indicates tempo relaxation, and in 101 presumably
&b ≈
quite extreme holding back that allows the interpo-
lation of a turn before the last four notes.
73–74. Pno, Vl: Speidel/Singer mark allargando in the He had marked these bars similarly, but differently
second half of 73 with a tempo at the beginning of in his earlier piano solo arrangement:
74. n œ. œ. œ œ. œ. bœ. œ. œ. # œ. œ. œ. œ. # œ. nœ. œ. nœ. œœ œœœœœ . . nœœœŸœ œœnœœœ
. . #œœœœ . . œjœœœœ
.
74f. Pno: See note to 54 –56, 74f &b

..
œ œ œ œ œ œ œ œ œj œœ œœ. œœ œœ
4 5 smorz: dol: pp
œœ. œœ œ œœ œ œœ œ œœ œ œ œ
p
œ
5
b
.
& œœ œœ ? b œ. œ œ. œ. œ œ œ̇
R
° *
74 –75. Pno: Curiously, Potter removes the trill on rh iii In 105–107 Czerny suggests the following in his solo
and replaces it with ten., again apparently a viable arrangement:
alternative. The chords on rh ii in 74 and 75 might . . . . . . . . mnœmœœœm œœœœœœœœœ
# œœœœœœœœœœœ
. . . . . . . . ŸœœŸœœŸ m3 2 1. œ2. œ1. œ2. œ1. œ2.
œ œœœ
œ # œœ
be arpeggiated swiftly to give energy without hard-
&b

ness to the sound (PT: 1/a/vi)
œœ. œœ.
? b œ. œœ œœ. œœ. œ œœ œ
p
œœ. œœ.
smorz: cres:
76–78. Pno: Speidel marks poco stringendo in the sec- œœ.
.
ond half of 76, poco rit. in the second half of 77, and a .œ œ œ œ œ œ œ œ
. . . . . .
tempo at the beginning of 78. He also marks 78 with
con molto espressione.
80. Pno, Vl: Speidel/Singer mark poco riten. in 80. In his duet arrangement, he marked embellishments
but no dynamic or tempo changes.
Var. 2
œ. œ. m
œ œm
œ œm
œœ m
œ œ. œ. œ œ œ œ œ œ m
œ œm
œ œm
œ œ œ œ œ . œ. œ. œ. œ.
œ
81ff. Pno: Given Beethoven’s marking leggiermente in
&b œ
.
Vl, which ordinarily indicates non legato, it is prob-
able that he intended non legato or staccato in Pno.
Reinecke, Diémer, Ganz, Rosé, Vogrich mark stac- Such freedoms would not have been untypical at
cato in lh. Speidel marks sempre staccato in rh and lh. the time, and Czerny’s versions here might well be
83, 85. Pno, Vl: Speidel/Singer add instructions that are taken by the violinist as a guide to ornamenting the
still reflected in modern performances. They mark repetition in this variation.
a) at the beginning of the Variation and  in 85, Similar tempo modification is included in the Spei
suggesting in a footnote: “On the repeat begin hesi- del/Singer edition, with allargando in 100, etwas zö
tantly and gradually go back into tempo at .” 156 gernd (somewhat holding back) in 101 and from 102i
92f. David, in his personal copy of his own edition, to the beginning of 103 nach und nach-----in tempo
marks an alternative bowing, perhaps intended for (gradually in tempo).
the repetition of this section. Combined with Czerny’s markings, this may sug-
gest a tradition of tempo modification that goes back
to Beethoven’s own time and is still reflected in the
earliest 20th-century recordings and many modern
performances.
108xvi–109viii. David, Alard, Auer, Brodsky, Halir,
156 “Bei der Wiederholung etwas zögernd zu beginnen, und dann Seybold mark the harmonic.
nach und nach in’s Tempo bei Zeichen .”

Var. 3 of virtuosity. Therefore, beware of too rapid a tempo
108ff. Pno: The mood of this variation and the general and, despite the dazzling trills and fioriture, strive to
style of writing supports the use of arpeggiation as be faithful to the spirit of Beethoven’s work.” 157
often as deemed necessary for expressive purposes Curiously, they indicate a tempo slightly faster than
(5/c/ii), certainly at important moments and to sup- that of the theme, although slower than the two pre-
port legato and enhance phrasing. For the chords ceding variations.
on 112i and 120i, one might consider playing the Beethoven’s expectations for the execution of the
bottom note of lh with the beat with the other notes numerous trills in this variation are impossible to de-
in lh and rh together and slightly afterwards (PT: termine. In the 20th century it became customary to
1/a/v). Potter marks 112i with an arpeggio sign. In execute all the trills from the main note and without
127, Speidel marks the rh octaves as portato articu- turns, except where an alternative was marked. Clearly,
lation. Ganz marks molto legato e con Pedale at the however, Beethoven was not punctilious about mark-
beginning of the variation. ing the places where he might have expected an up-
110–112. Pno, Vl: Speidel/Singer mark etwas drängend per-note start or a turned ending. In 140, for instance,
(somewhat forward-pressing) in 110 and a tempo in the turn is marked, but it is not given at the equivalent
112. places in Vl in 148 and Pno in 167. Where the addition
113–115, 132–134. Pno, Vl: Beethoven’s espressivo may of a turn seems to have been envisaged as obvious,
have elicited a variety of responses from the per- according to the conventions of Beethoven’s time, it
formers of his time. Czerny teaches that espressivo has been suggested editorially. Reference is also made
almost always implies some broadening of tempo to Czerny’s arrangements, and trill beginnings have
(3/d/2). For a pianist it might also encourage some also been added editorially in contexts where Czerny
rhythmic flexibility and perhaps some breaking of supplied them. Possibilities in other cases are consid-
the chords. For a violinist, in addition to rhythmic ered in the following notes.
flexibility, it might encourage vibrato (on the fp) and 135ff, 151ff. Pno: Beethoven’s dolce (to which Speidel
perhaps portamento, especially with the crescendo. adds e graziosamente in 134) inspires the use of the
In 114/133, portamento is indicated by Alard (fin- moderator pedal if available (PT: 3/b) as well as ex-
gering 1–1), Auer, Brodsky (3–3 on the D-string). At pressive asynchrony and arpeggiation, and these
115/134 Speidel/Singer mark smorz[ando], and from two expressive devices might be used throughout
115v–116i/134v–135i Singer marks a portamento line the movement for example at 194 –196, 203–206 rh
between bowstrokes and shifts to 2 on the A-string. and 219–232 (5/c/ii). The 32nd-note broken chords in
Of course, since each section is repeated, it is proba- lh undoubtedly legato and with overholding (4/a/ii).
ble that accomplished pianists and violinists of Beet- Speidel, Diémer marks slurs. Speidel also adds in lh
hoven’s time would not have played these passages to change the sustaining-pedal “with the changes
quite the same on each occurrence, the violinist per- of harmony”.158 Much of the rh figurations will have
haps changing fingering. In the edited violin part, also been expected to be legato. Reinecke occasion-
alternative fingerings are given, and variants might ally adds slurs. Speidel, Diémer mark many slurs
be tried during the repeats. throughout.
120. Vl: Many editors suggest a portamento shift from 136–137, 163–164. rh: The trill sign at rh i should prob-
i–ii returning to 1st position on iv; Alard, Hermann, ably be interpreted as a Schneller (see note to 54ff).
Singer, Auer, Halir, Kreisler with a harmonic on ii, The turn in 136 and 229 should be left as late as pos-
Rosé with a stopped 4th finger. sible and joined the following main note, perhaps
127–131. Pno, Vl: Speidel/Singer mark allargando in the as shown by Ganz in 136:
second half of 127, followed by ’ at the end of the
bar, a tempo in 20, etwas drängend (somewhat for-
ward-pressing) in 129, and a tempo in 131.
157 “Der Reiz vorstehender Variation wird entschieden geschmä-
lert, wenn dieselbe, wie heut zu Tage so oft geschieht, zu einem
Var. 4 kühlen Virtuosenstückchen herabgewürdigt wird. Man hüte sich
Speidel/Singer give a footnote: “The charm of the pre- deshalb vor einer zu raschen Temponahme und bestrebt sich trotz
der blendenden Triller und Fiorituren dem Geist des Beethoven-
sent variation is decisively diminished when, as we schen Werkes gerecht zu werden.”
have so often seen it today, it is reduced to a cold piece 158 “Ped bei Harmonieveränderung”

mœ 2 3
>1 4 3 Also by Reinecke:
œ œ œœœœ œ œ Ÿ~~~~~~~~~~~~~~~
&b
.
# œœ œœ œ n œ œ œ
j j2 5

œ
&b
.
&b œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ 123
3 3
° ° °
At 164, Ganz marks espressivo. Ganz uses dotted-line notation to align the grace-
139f. Pno: The chain of trills should probably start on note with the bass, but omits the trill ending:
the upper-auxiliary note with no turned ending ac- Ÿ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# œœ œ œ5 n œ œ œ 1
j
cording to Czerny’s rule (5/b/ii). But Reinecke’s fin- œ
&b
.

2 341313
gerings for the trills in 139 show upper note starts,
œœœ œ œ œœœ œ
3 3
though in 140 a main note start. The turned end- &b œ œ œ œ œ œœ

ing at the end of the chain is given by Beethoven.
The question is whether he expected it on each trill,
which is possible in an ascending chain (though 143iii. Vl: Singer suggests left-hand pizz. Followed on
Czerny says it should be marked if required). In iv by . On iv, David, Joachim (but not Joachim-
139 Czerny gives an upper auxiliary start both in revised), Halir, Rosé, Seybold, Kreisler mark , the
his solo arrangement and his duet arrangement; in others .
the latter he does not mark upper notes to the two 143iv–v. Vl: All shift 2–3.
following trills. 144iv–v. Vl: All except Joachim, Kreisler shift back;
Duet arrangement: those shifting 3–3 (David, Singer, Brodsky, Seybold),
r Ÿ~~~~~~~~
œŸ
before going to 2nd position on 9iv, evidently ex-
#œ œœ œ œœ œ œ # œŸ J #œ œ
&b J pected a more prominent audible connection than

the others, who shift 3–2.
œ
&b œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ 147–149. Vl: Although Czerny does not add the upper
œ œ auxiliary to the first trill in 147 in his duet arrange-
ment, he had marked it in the identical figure in
In the former the fingering indicates them: 139. In his piano solo arrangement, the fingerings
Solo arrangement: indicate upper-auxiliary starts to all the trills except
Ÿœ 149i, which is preceded by the note above, where
# œ œ œ œ œ œ œœŸ # œŸ
232 32
j
4 43
#œ œ
&b J J the fingering explicitly indicates a main-note start.
 Ÿ5 4~~~~~~~~~~~~~~~~~~~~~ Ÿ5 j2 3 2 Ÿ4 j3
œŸ œ œ5 œ œ nœ 1 4
6
œ œœ # œœ œ œœ
&b œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œœ
#œ œ
& b œR21 ≈ œ œ
J

1
œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ
?b œ œ œ œ œ œ œ œ œ œ œ œ œ
141. Pno: The trill ending here is given by Czerny.
In his duet arrangement:
Ÿ 343 œŸ # œ2 œ3 œ5 œ4 n œ œ œ1 œ3 nœ
& b œJ
2 1
“” Ÿ #œ œ
& b #œ œ œ œ œ œ nœ œ œ
r

œ .
œ œ œ œ œ œ œ œ# œ œ œ œ œ 9 œ œ
.
?b œ œ
6
&
Somewhat differently in his solo arrangement:

Ÿ~~~~~~~~~~~~~~~~~~ In 148 Czerny supplies the turn, which Beethoven
œ. œ œ œ œ n œ6 œ œ
2 3 4
#œ œ œ
œ
loco
added in the parallel passage at 140, but omitted
&b ‰ ≈R ≈R
here. In 149 a similar ending to the turn was evi-
1 3
dently expected as at 141. Czerny supplies it in modi-
fied form in his solo piano arrangement.
152f. Vl: The dolce may best be achieved by a fast, light
bowstroke between the lower half and the point.
164. Vl: Singer adds espressivo.

164 –167, 183–185. Pno, Vl: Czerny, reflecting Beetho- Solo arrangement, bb. 183–186:
ven’s notation in bars 24 –25 in the theme, supplies mœ ^ œŸ œŸ œŸ œŸ œŸ “”
œœ Ÿ
turns to all the trills except 165i/184i in his 1825 duet J # œŸ œŸ # œŸ‹œ œ œJ
&b J J
arrangement, but the 164 –167 passage includes the sempre più dolce pp
accidentals on the first note of the turns only in
166, where he marks the same as in 185. Probably Duet arrangement:
the accidentals were also intended in 164f as in 183f, Ÿ
œœ œ . œ Ÿœ nœ œ œŸ#œ œ œŸ œŸ#œ œ œŸ #œ œ œŸœ œ #œJŸ #œ œ œŸ œ œ #œJŸ‹œ œ œJ #œ œ
but overlooked; on the other hand, it is likely that &b œ J
sometimes a semitone and sometimes a tone would pp
have been used, according to the taste of the per-

former, where nothing was specified. The version 165i, 184i. Pno, Vl: With the exception of this one, all
in the 1823 solo arrangement at 145–148, where up- the notes of this chain of trills in Czerny’s piano
per-auxiliary starts to all the trills are indicated by duet arrangement, on both occurrences, are con-
the fingering, is a substitute for Beethoven’s fiori- cluded with a turn. It’s omission here seems musi-
ture in the violin part, which Czerny included in cally persuasive, and may well be deliberate.
his duet arrangement, but altered here for the sake 174. Vl: In his 1823 piano solo arrangement Czerny, on
of variety and playability. Czerny’s fingering in his the final 8th-note beat of the bar, adds an initial c #,
1823 solo piano arrangement shows upper-auxiliary replacing Beethoven's 32nd-note rest, and a grace-
starts on the trills at 164iv, 165i, 165iii, 166iii, 167i. note:
The fingering for 165ii clearly indicates a main-note œ œj œ œ
& b #œ œ œ
start, presumably because this trill is preceded by
the note above (although this does not always pre- 6
clude Czerny from specifying an upper-note start, 175. Vl: Czerny’s piano solo arrangement gives a low-
for instance in bb. 74, and 146, where his fingering er auxiliary start from the semitone below to the
shows an upper-auxiliary start). first trill. A start from a tone below or from the up-
Solo arrangement, bb. 145–148: per auxiliary would also be stylistically plausible, a
main-note beginning probably less so.
Ÿ Ÿ Ÿ Ÿ Ÿj Ÿ~~~~~~~ 5Ÿj2 3 2 Ÿj œŸ
54 43
j Ÿj
54 43 54 54 54
Solo arrangement:
œœ2 # œ œ #œ œ
2 1
œœ œ œ œ œ # œ
& b œJ 2 œ 1 œ 2 œ 1 œ # œœ21 œ1 ≈ œ 1 œ œ Ÿ~~~~ Ÿ Ÿ

J b # œ œ œ œ œ œ œ nœ# œ œ
&
Duet arrangement, bb. 145–148: Duet arrangement:

Ÿœ œ œŸ Ÿ
nœ #œ œ œ.
“”
m nœ œ nœ œ œ &b
#œ
b œ œ œ œ œ œ# œ œ œ œ œ# œ œ œ œ œ# œ œ œ œ œ# œ œ œ œ œ œ œ R
&
Ÿ Both Czerny’s arrangements omit a turn to the
# œ œ œ œ œ œ œŸ # œŸ œJ
<“>
& b J second trill, but probably through oversight, since
6 in the duet arrangement, a turn is included on all
three trills when he gives the same figure in b. 156
Solo arrangement, bb. 164 –167: (where the passage does not, however, occur in this
œj œ.. œ œ œ2 j1
3
Ÿj Ÿ Ÿ~~~~~~ Ÿj Ÿ form in Beethoven’s original or Czerny’s solo piano
œ ## œœj 21 œœ 31 #4œ2 3 2 œJ
5
œŸ œŸ œŸ œ arrangement).
&b J J J J R ≈Œ
21 32 1 2 Ÿ 31 21
Duet arrangement, b. 156

cres:
Ÿœ Ÿ Ÿ
# œ œ œœ œ nœ# œ œ œ.
Duet arrangement: &b R
Ÿœ œ Ÿœ Ÿœ œ Ÿœ œ œ Ÿœ œ œ # œŸ #œ œ Ÿœ Ÿ
œ œ # œ ‹œ œ
b
& Jœ œ J J Whether Beethoven would have envisaged a main-
note start to the second and third trills is uncertain,
but Czerny gives no indication of an upper auxil-

œ
iary beginning in any of his arrangements. &b œ œ œœ # œœ
œœ œœ œœ.... œ
œ..
∏∏∏∏∏
179ii. Vl: A harmonic is marked by Hermann, Auer. a tempo
181. Vl: No ties or slurs appear in the sources. Perhaps
the trill sign, extending to ii in the 1st edition, im- and his duet arrangement:
plied a tie. In Czerny’s solo piano arrangement and Ÿ
U T j œ œœœ.. œ œ œ U
œ œœ
his arrangement for cello and piano, a tie from i–ii & b ‰ œ œ œ œ.. œ œ œ œ. . J
is marked. p dol espress: fz
181i. Vl: In his arrangement for piano duet, Czerny
omits an upper auxiliary before the trill, but surely While Beethoven may not have expected this kind
through oversight, since he supplies one here in the of ornamentation from the violinist, Czerny’s nota-
solo arrangement. An upper auxiliary start or, alter- tion, resembling the notation used in singing trea-
natively, one from below would both be stylistically tises to indicate portamento, certainly suggests a
plausible; a main-note start would probably have portamento connection between the d1 and b 1 in
been less likely at that time. 193 and from the g1 to the g2 in 195, which would
181–189. Vl: David’s changes to bowing and dynam- be stylistically normal at Beethoven’s time. A por-
ics in his personal copy are particularly interesting tamento from the open D-string in 193 would be
(see CHASE http://mhm.hud.ac.uk/chase/view/pdf/ very plausible.159 Bridgetower, with his Esterháza
298/1/#page). and Viotti School connections, might well have exe-
183–185. Vl: Czerny adds the turns in his piano duet cuted a portamento of this kind; he might also have
arrangement. The turn at 185iii, with f 3, is given made a hint of portamento up the D-string with the
in Czerny’s arrangement for cello and piano and is first finger, before stopping the g2 with the 3rd fin-
also is given by Singer. In Czerny’s solo arrange- ger in 4th position, as suggested by Hermann and
ment, when this material first appears in 164 –167, he Auer, or even with the 2nd finger on g2.160
also gives fingerings that indicate upper-auxiliary 194. Pno: Speidel marks both hands with portato.
beginnings to the trills except on 165ii (184ii), which 196. Vl: Czerny embellishes the fermatas. In his solo
is preceded by the note above. See note to 164 –167. arrangement, he elaborates the first fermata, but in
188f. Pno, Vl: Speidel/Singer mark poco riten. In 53 and the duet arrangement, he omits the first fermata sign
a tempo at the half bar in 54. and makes a cadenza (Eingang) after the fermata
190–191. Pno: Cipriani Potter adds Sostenuto in 190 and with the trill.
arpeggio markings for both chords:
Sostenuto Solo arrangement:
& b œœœ œœœ œ˙. j “”
œ œ

œ œ œ œ œ œ nœ œ
∏∏∏∏∏
œU
œ
∏∏∏∏
œœ Allegro.
œ œ nœ œ œ nœ œ
&b œ nœ œ
˙˙
dolce espress:
? b œœ œ œ œ

œœ œ
œ ˙
∏∏∏∏∏
sf °
∏∏∏∏∏
? b nU œ œœœ
nœ
œ nœ
While Speidel marks arpeggios on 190i and 191i: n œ ff œ
- - -52 - œœœœ
loco
nœ œ œ œ œ œU
Ÿ
j œ. . . U
œ œ œ œ œ œ œ. œ. œ. . œ.
& b œœ œ &b J œ
bœœœ œ˙. œ
nœ
œœ
 
∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏∏
∏∏∏∏∏∏∏ ∏∏∏∏
U P
sf p
U
dolce ed espress.
? b œœ œœ ˙˙ ? ‰ ‰
œ œ ˙ b
These two examples strongly support the fact that 159 See David: Violinschule, vol. 2, p. 33, where, illustrating rising
terms such as dolce, espressivo and sostenuto were intervals from an open string with a small note between them, he
provides a footnote: “Put down the first finger behind the nut and
linked closely with the practice of piano arpeggia- draw it up to the small note / Bei diesen Stellen setze man den
tion. ersten Finger hinter den Sattel und ziehe ihn bis zur kleinen Note
193–195. Vl: Czerny gives the following ornamented herauf”.
160 For information on portamento in Haydn’s circle see Clive
versions of the violin part in his piano solo arrange- Brown: “Haydn’s Musical Legacy: Reception and Performing Prac-
ment: tice,” in: Eisenstädter Haydn-Berichte, vol. 12 (2020), pp. 239–274.

Duet arrangement: 203f. Pno, Vl: The dotted figures were almost certain-
œ œnœ ly expected to be assimilated to the triplet rhythm
U
œ œœ œœœ œ œ# œ “”
&b J nœ œ# œ œ œ œ œ œ œ# œ œ œ œ# œ œ œnœnœ# œ (1/c/iii).
Presto: 205ff. Pno: Broken chords in lh and rh could be over-
<“>
œ œ œ # œ œ bœ n œ œ n œ œ # œ œ
œœ# œœ œ œ# œœ n œ # œ œ
held (4/a/ii).
nœœ# œ œ œbœnœœ
& b œ 205-207. rh: The octaves might be gently arpeggiated
and or played asynchronously with the bass which
nU
œ
<“>
b n œ œ# œ œ œ œ# œ œ# œ œ œ# œ œ# œ . œ nœnœ œnœ œ œ U
œ.
œ
would produce a gentle expressivity (5/c/ii).
& œ# œ œ# œ 206. lh: Potter marks legato.
rallen.
207i–208iv. Vl: This phrase can easily be played in
1st position, which all apparently envisage, except
Whether Beethoven envisaged an improvised caden Alard, Singer, who begin in 5th and shift 4 – 4 on the
za for the violinist is indeterminable, but it seems first two notes, and Auer, who begins in 3rd and
very likely. It would seem a bold step for Czerny shifts 2–2 on 73ii–iii.
to notate one in Beethoven’s lifetime if the com- 213. Vl: Alard marks 0, Halir 2; none of the others
poser had not envisaged it. Violinists’ versions of specify a fingering.
the kinds of cadenza written by Czerny might be 214ff. rh: The broken chords undoubtedly legato and
something like the following. overheld (4/a/ii). Reinecke, Speidel, Diémer mark
From the first fermata: slurs.
215f. Vl: A shift with the 1st finger from 3rd to 5th
U œ≥ œœ
3
œ œ œ œ œ œ œ3 œ œnœ
1
œ position at the end of 215, followed by substitution

œnœ œ œ œ
7
&b œnœ
3 1
sf
œ œ œnœ œ
œ œ nœ œ œ of the 2nd finger on 216i descending to 1st finger
œ œ œ œ œ4 œ œ œ4 on 216ii (as marked by Alard) would be an effective
œ œ n œ œ n œ4 œ œ U
4
m≤
# œnŸœ œ œ -- fingering, enabling a rapid expressive portamento
&b to enhance the crescendo to sf and a gentle porta-
mento descent from 216i–ii.
217i–ii. Vl: All shift down the A-string 3–2 except
From the trilled fermata:
Alard, who shifts 2–1.
≤
Uœ œ œ œ Ÿœ œ œ œ n œ œ2 œ œ 0 218iv. Vl: Czerny gives the upper auxiliary and turn in
&b nœ œ œ
œn œ œ œ œ œ œ œ œ# œ œ œ œ# œ his 1823 piano arrangement, and a turn follows the
sf œ œ trill in his duet arrangement. Hermann also adds
≥ the turn after the trill. In Czerny’s duet arrange-
œ œ œ# œ œ
0 2 1
& œ œ nœ # œ œ œ # œ œ
b
1
bœ nœ œ ment he replaces the g1 immediately preceding the

œ bœ n œ bœ n œ œ # œ œ trill with a1 and marks no auxiliary to the trill.
U ≤
œ# œ œ# œ œ œ # œ œ# œ œ n œ --
m Whether his substitution of a1 for g1 indicates that
& b # œ œ# œ œ bœ nœ he had discovered that the g1 was a mistake in the
Stichvorlage and 1st edition is unverifiable.
197, 200. Vl: The upper-auxiliary beginning to the trill 223. Vl: In this legato context, and the absence of stac-
in 197 and the turn in 200 are given by Czerny in cato marks, it seems unlikely that separate bows
his piano solo and his duet arrangement. The up- were envisaged. Only one of the 19th-century edi-
per auxiliary in 200 is specified only in the duet tors (Auer), however, added slurs over each group
arrangement. In 197 a trill beginning from below of three, although all added slurs in the previous
would also have been a plausible option. bar. Several, including David, added staccato marks.
202i. Vl: A harmonic is marked by Alard, Hermann, Separate bows, portato, or legato over each group of
Singer, Auer, Brodsky, Halir. three might possibly have been employed.
202x–xii, 203x–xii. Pno, Vl: These groups of three notes 225. Pno, Vl: Czerny, in his arrangements, gives an
are slurred in Czerny’s piano duet arrangement. upper-auxiliary start to the trills.
203xii. Vl: A harmonic is marked by Alard, Singer, 227. Pno, Vl: Speidel/Singer include a footnote: “One
Auer, Halir; David and others in 3rd position prob- counts 6 8ths on the fermata in the violin part and
ably considered it obvious. 4 on that of the piano part, and makes the trill with-

out a turn.” 161 It seems very unlikely that Beethoven [Interestingly, however, Czerny does not mark these
expected a trill without a turn; pianists and violin- dynamics in his 1825 piano duet arrangement, in which
ists of the time would certainly have played one. he otherwise very freely changes Beethoven’s dynam-
Czerny, in his piano arrangements, notated a turn ics, or in his arrangement for cello and piano.] The
after the second trilled note, in 1823 with f 2 and later passage in 2/4 time at the same tempo as every-
d #2, but in 1825 without accidentals. Reinecke also thing else, so that its quarter-note takes just as long
gives a turn with f 2 for the piano but Hermann as would a dotted quarter. The twice-recurring little
does not provide a turn in the violin part. Adagio at the end of the piece is not at all dragged,
231f. Vl: It could be effective either to make ii–iii more but as expressive as possible. The conclusion noisy
expressive the first time or the second time with and prestissimo.” 162
a portamento. A sensitive violinist would probably 1–18. Vl: The 19th-century editors have a range of
not have used the same portamento fingering twice. strategies for the bowing. At 1–3ii, all but two of the
234ix. Vl: Halir gives a harmonic with 3. 19th-century editors begin  and execute each half
bar  ; Hermann, Rosé begin  then hook each half
Finale bar    . It seems very likely that Beethoven had
Presto the contre coup d’archet in mind when he devised his
Tempo theme. This kind of distinctly accented bowing, ex-
The only 6/8 Presto for which Beethoven gave a metro- ecuted at the point of the bow, which is closely re-
nome mark is the final movement of the String Quar- lated to martelé, was certainly a characteristic of the
tet op. 18 no. 3; he gave it . = 96, which is on the verge French style that Beethoven would have experienced
of impossibility. It seems therefore that, although he in the playing of Rodolphe Kreutzer a few years
could have marked the movement Prestissimo, Beet- earlier, and, it is marked by the majority of the 19th-
hoven was essentially instructing performers to play century editors. Bridgetower, strongly influenced by
the Presto as fast as possible. The earliest marking for Viotti, would probably have used it, and it was sure-
the Presto in op. 47 comes very close to that speed; the ly current among younger Viennese violinists.
others are a little slower. From 3iii to 5, David, Hermann, Halir, Rosé, Sey-
Haslinger . = 92 bold use   on the second half of each bar arriving
Moscheles . = 88  in 6, where they take iii, iv  ; Joachim, Kreisler
Czerny Vortrag . = 88 take separate bows on the second half of each bar,
Alard/Diémer . = 176 while Singer, Auer, Brodsky take separate bows in
Speidel/Singer . = 176 3 but   in 4, 5, arriving  in 6 and taking iii, iv 
Kreisler  = c.88 . Alard takes only 5ii–iii  , apparently envisaging
Czerny writes: “Very fast, just as brilliant and fiery the 8ths from 6iii  , therefore 7–10 beginning ;
as the first movement, but much more cheerful. All this is consistent on each entry of the theme and
the 8ths staccato, where the opposite is not expressly suggests that he played the separate 8ths close to
specified. The middle melody with the following ex- the point of the bow, while the others may have
pression, piquant and humorous: played more towards the upper middle of the bow.
r Singer marks leggiero at 6ii and M[itte] (middle of the
# # œ. œ. œ.
. . . . # œ. ˙.
œ. œ. œJ. œ. œJ œ œJ œ J
œ #œ
. œ. œ. œ.
J J
&#
bow), presumably envisaging something like Ferdi-
J J

nand David’s hüpfender Strich where the elasticity
p
œ. œ
sf
œ
pp dol.
? ### œ. ‰ œ ‰ Œ ‰œ ‰ Œ ‰œ ‰ Œ ‰œ ‰ Œ ‰œ ‰ of the bow-stick comes into play, but without the
œ. œ hair leaving the string.
r
### œ œJ œ # œJ œ œJ œ # œJ ˙.
#œ . . . . . . . .
& 162 “Sehr schnell, eben so brillant und feurig wie der erste Satz,

aber viel munt’rer. Alle Achteln staccato, wo nicht ausdrücklich das
œ.
? ### Œ ‰ œ ‰ œ ‰ œ ‰ œ ‰ Œ ‰ Gegentheil vorgezeichnet ist. Die Mittelmelodie mit folgendem Aus-
œ. œ.
œ druck, pickant und humoristisch: [Ex.] Die spät’re Stelle im 2/4 Takt
im selben Tempo, wie alles, so dass da eine Viertelnote eben so
lange daure, wie sonst eine Viertel mit Punkt. Das zweimal wie-
derkehrende kleine Adagio am Schlusse des Tonstücks durchaus
161 “Man zähle auf die Fermate der Violinstimme 6, auf die der nicht schleppend, aber so ausdrucksvoll wie möglich. Der Schluss
Klavierstimme 4 Achtel und mache den Triller ohne Nachschlag.” lärmend und Prestissimo.”

1, 230. Pno: Very swift arpeggiation of this opening 127–129, 135–137, 404 – 406, 412– 414. Pno, Vl: In Cipri
chord will give it requisite fire and energy as rec- ani Potter’s edition, tenuto lines have been added
ommended by 19th-century musicians such as Sam- over the staccato marks. The same notation is used
uel Wesley and Charles de Bériot (5/c/ii). by Halir, while Auer simply changed the staccato
14ff, 62–77, 459– 466, 493– 496, 505–524. Pno: Presum- marks to tenuto lines. This almost certainly reflects
ably Beethoven expected legato for the sequence of Beethoven’s intention that the half-notes should be
continuous 8th-notes (4/a/ii). But the editors do not separated and given special weight, but not very
mark slurs and Speidel marks leggierissimo in 14 which short.
suggests non legato. In 62 Ganz marks leggiero and 127–135. Pno, Vl: Contrary to Czerny’s comment in
at 453 non legato. Von dem Vortrage, Speidel/Singer add poco tranquillo
22. Vl: Singer marks M[itte]. at 127, a tempo at 134, poco tranquillo at 135.
31ii–iv. Vl: Alard, Brodsky mark a harmonic on ii with 127ff, 404ff, 489ff. Pno: According to Czerny in his
4 on iii; Hermann, Halir 4 – 4 on ii, iii; others give Pianoforte-Schule op. 500 (1839), “all chords consist-
4 –3–3 here. ing of very short notes” should, unless expressly
36–51. Pno: Undoubtedly, the arpeggios would have marked by the composer, be unarpeggiated. But this
been expected to be played legato with overhold- might not have precluded extremely swift arpeggia-
ing (4/a/ii). Reinecke, Speidel, Vogrich marks slurs; tion, the type that Thalberg in L’Art du chant (1853)
Ganz marks non legato but marks slurs in 50–51. All described as presque plaqué, or almost together. In
chords in lh marked sf might be swiftly arpeggiat- 130–131, the portato articulation might signal slight
ed which would produce the requisite fire without arpeggiation but with the notes given “the same
hardness of sound (PT: 1/a/v). length of time as a dot under a slur requires” as rec-
60–62, 337–339. Pno, Vl: Speidel/Singer give poco ommended by Moscheles in his Studies for the Piano
allargando in 60/337 with a tempo at 62/339. Forte op. 70, Bk 1 (1827). All other chords might be
62–69. Vl: See Carl Czerny’s instruction (above) for the swiftly arpeggiated.
dynamic treatment of this theme when it comes in 130f, 407f. Vl: A harmonic is marked by Alard; Auer
the piano part at b. 78. marks one only on 111ii.
64 –94, 188–212. lh: Any of the octaves might be swift- 142f, 419f. Vl: A harmonic is marked by Alard, Singer,
ly arpeggiated (5/c/ii). Auer, Brodsky. It was surely envisaged, at least as a
65. Vl: None of the editors mark a harmonic, but those possibility, by some of the others who mark noth-
who indicate no change from 3rd position, probably ing. Alard does not mark the harmonic at 419, prob-
regarded it as obvious. ably through oversight, since he marks it at 407.
78–85. Pno: See Carl Czerny’s instruction (above) for 148f, 424f. Pno, Vl: Since the start is from below the
the dynamic treatment of this theme. trilled note, which indicates an upper auxiliary on
82, 83 etc. rh: The grace-note, as an acciaccatura (5/a/ the strong beats, in the sources at 148, it is evident
iii), played simultaneously with the main note and that Beethoven would either have expected an up-
released quickly, is explained in Junghanss’ Piano- per-auxiliary start to the trills in Vl, or perhaps, in
forte-Schule (Vienna, c. 1820). imitation of the piano, a lower-auxiliary beginning
103–125. Pno: In this texture, occasional or even fre- to the first of the violin’s trills. The same was sure-
quent asynchrony would be very appropriate (5/c/ ly expected at 424f, where no trill beginnings are
ii). In 122–125, legato was presumably intended. Rei- marked in the sources. Ganz aligns the grace-note
necke, Speidel, Diémer marks slurs. with the bass using dotted line notation.
115. Vl: A harmonic is marked by Alard, Singer, Brod- 150–151. lh: Presumably legato as marked by Reine
sky. Others who indicate no change from 3rd posi- cke, Diémer, Vogrich.
tion probably regarded it as obvious, or at least a 151, 428. Vl, Pno: The turn to the trill is supplied by
perfectly normal alternative to a stopped extension. Czerny in his piano duet arrangement. Hermann,
126ii. Pno, Vl: Speidel/Singer give a footnote: “Make a Singer, Halir also include it.
very short pause”.163 156–169. Vl: As at the beginning, different 19th-century
editors took different approaches. Alard, David, Her-
mann, Brodsky, Halir, Rosé use hooked bowings to
163 “Eine ganz kurze Pause machen” bring out all, or all but one of the sfs on . Joachim,

Singer, Auer, Kreisler use fewer hooked bowings 255–266. Pno: With as much overholding as possible
and execute all or most sfs , presumably fouetté. (4/a/ii). Ganz marks armonioso and una corda with tre
161–165. Vl: David, Brodsky, Halir, Seybold, who gen- corde in 291.
erally employ the     bowing, change here to  267ii–283. Vl: Brodsky marks all the e2s 0. Joachim,
  , presumably to allow greater bow length for Singer, Auer, Seybold, Kreisler mark those in 271–
the crescendo. 275 and 280–283 with 0, Hermann marks them only
174 –178, 451– 454. Pno: Presumably legato as marked in 271f, Halir just in 271, but presumably intend then
by Reinecke, Speidel, Halir, but Ganz marks non to continue. David probably considered it too obvi-
legato. ous to mark after 253 and probably, like Brodsky,
180–181, 214 –217. Pno: The sf might be enhanced with expected all to be played on the open E-string.
slight asynchrony with rh after lh (5/c/ii; PT: 1/b). 291–292, 299–300. Pno: the octaves perhaps non legato
182–185. Pno: if the Vl were to play short notes on the to accord with Vl.
string, the Pno might adopt a similar articulation. 293–298, 301–312. Pno: Again, presumably legato (4/a/
186–201. Pno: One might expect this to be generally ii), but none of the editors mark it so.
legato. Reinecke marks slurs only in 192i–193 and 321–322. lh: Presumably legato with overholding (4/a/
200i–201, Speidel 192iv–193 and 200iv–205. Potter ii) as marked by Reinecke, Speidel.
marks a slur in 186i–vi, presumably to be continued. 335f. Vl: An open E-string for the sf is already marked
Diémer marks slurs from 186 onwards. In 182, Ganz in the Stichvorlage, perhaps reflecting the perform-
marks martellato. ance with Bridgetower.
192, 200. Vl: Among the editors, only Hermann and 399– 402. Pno: Presumably legato as marked by Rei-
Brodsky contrive to play the sf  (by breaking the necke, Speidel, Diémer
slur and taking  for the following note); the others 403. Pno, Vl: Speidel/Singer mark a Luftpause ’ after ii.
take the sf , again surely fouetté. 407f. Vl: See note to 130f.
207–213. Pno, Vl: Beethoven’s inclusion within the slur 419f. Vl: See note to 142f.
of the notes repeated at the same pitch is so consist- 424f. Pno, Vl: See note to 148f.
ent in Vl and Pno that it is surely deliberate. By this 427. Pno, Vl: See note to 151.
notation he almost certainly wanted to signify the 427– 428. lh: Presumably legato as marked by Reine
most connected type of portato, probably achieved cke, Vogrich.
on the piano by the substitution of one finger for an- 428. See note to 151.
other on the same key. A violinist might also have 443f. Vl: Those who remain in 6th position (David,
used a similar technique. This is described by Spohr Alard, Brodsky, Seybold) may well have envisaged
as an imitation of vocal practice; after discussing vi- a harmonic with 2 on 444i. David marks this in blue
brato (Tremolo) he writes: “By changing the finger crayon in his personal copy.
on a note, a vocal effect is also imitated, namely the 479– 482. rh: Presumably legato as marked by Reinecke.
separation of two sounds, sung at the same pitch 479. lh: Ganz marks marc.
in one breath while pronouncing a new syllable.” 164 491, 500. Pno, Vl: Czerny adds ritard in his piano duet
None of the 19th-century editors marks a fingering arrangement. In 491, Singer adds espressivo.
of this kind, although it might easily be applied (as 492. Pno, Vl: Speidel/Singer mark a Luftpause at the
suggested by the fingering above the notes in the end of the bar.
edited violin part of the present edition); all mark a 497. Vl: Singer marks espressivo.
change of bow between the notes at the same pitch. 500. Pno, Vl: Speidel/Singer mark a Luftpause at the end
253i–ii. Vl: David, Halir, Rosé, Seybold, Kreisler mark of the bar and leggiero from the middle of the bar.
o–0; Brodsky o– 4. 501. Vl: Singer marks M[itte] (middle of the bow).
517–521. Vl: Alard marks a harmonic at 517i, after 3rd
position from 215iv–516vi; whether he envisaged a
stopped note from 517ii is unclear, but on 522ii he
164 Spohr: Violinschule, p. 175. “Durch das Wechseln der Finger auf marks 4. David, Seybold both take 516 in 3rd posi-
einem Ton wird ebenfalls etwas, dem Gesange angehörendes nach- tion and may well have envisaged a harmonic, but
geahmt, nämlich das, durch das Aussprechen einer neuen Sylbe
bewirkte Trennen zweier, auf derselben Klangstufe befindlichen in his personal copy David marks 3 on 516iv, obvi-
und in einem Athem gesungener Töne.” ously intending a stopped note on 517i.

521, 527, 531, 533. Vl: Alard marks o on all but 521, Haslinger  = 138
Seybold on all but 533; David, Hermann, Joachim Moscheles  = 160
give no guidance, but, remaining in 3rd position, Czerny Vortrag  = 132
may have regarded a harmonic as obvious. All the Alard/Diémer  = 120
others clearly use a stopped note in either 4th or Speidel/Singer  = 126
5th position. Kreisler/Rupp  = 120–126
521. Pno: Presumably all the arpeggio figures legato The general adoption of a very broad tempo by 20th-
with overholding (4/a/ii). Reinecke, Speidel mark century performers is indicated by Kreisler’s 1935 re-
slurs from 525 onwards. Diémer only in 525–526 cording, Adolf Busch and Rudolf Serkin at c. 112, and
and 529–530. Rostal’s suggested  = 112–126.
524. Pno, Vl: The turns to the trills are supplied by Both Moscheles and Czerny may have had the op-
Czerny in his piano duet arrangement. portunity to hear an early performance of the sonata,
525. Vl: Singer instructs: “remain at the point of the but whether Czerny’s comment on it in 1846 reflects
bow”.165 his own musical instinct or a genuine tradition of per-
535. Vl: Alard marks a harmonic on i and open string formance is impossible to know. It seems, however,
on ii. The open string is also marked by Hermann, at odds with Beethoven’s tempo conventions. This is
Auer, Brodsky, Joachim-revised, the three latter with perhaps another instance of his statement that “even
4 on 536i. the spiritual conception has acquired a different va-
lidity through the changed taste of the times.” 166
He stated: “This piece, written in a calmly noble,
melodious, but also humorous character, must be per-
SONATA OPUS 96
formed with delicacy and feeling in a most moderate
Allegro moderato tempo (almost Tempo di Menuetto), since it ought not to
Tempo be played brilliantly or with any attempt at bravura.
Beethoven gave no metronome marks for allegros with The middle melodies with grace and delicacy, the pas-
a slowing qualifying term. Allegros in 3/4 to which he sages in thirds, clearly and legato.” 167
gave metronome marks fall into two distinct catego- 0, 1, 2, 6, 7 etc. Pno, Vl: The execution of the trill, which
ries. Some of those that are essentially scherzos have performs such an important role in the thematic
significantly faster metronome marks than those that material of this movement, has given rise to much
are not; in the Third Symphony, for example, the first speculation. In this case, Beethoven does not specify
movement (Allegro con brio) is given . = 60 and the how the trills should begin, or how they should end.
third movement (Allegro vivace) . = 116. For the first The most important matters to consider are his nota-
movement of the String Quartet op. 18 no. 1 (Allegro tion of trills in general and his expectations for the
con brio) he gave . = 54; the difference from the Sym- understanding of his notation by contemporaries.
phony may be explained by the nature of the 16th- Since a turned ending was overwhelmingly under-
note passages. In the Sonata op. 96, apart from a few stood to constitute a normal element of the trill in
bars near the end of the movement, the fastest notes the circumstances in which it occurs in this sonata,
are triplet 8ths, which may imply that the moderato Beethoven, if he positively required a trill without
qualifier was intended to prevent a tempo even faster turn, would surely have needed to indicated this
than that of the Third Symphony first movement. In somehow; one possibility would, of course, have been
the light of Beethoven’s practice, it seems possible that a verbal instruction, for instance ‘ohne Nachschlag’,
Moscheles’ suggestion is closest to Beethoven’s expec- another would have been to use the wavy line trill
tation, although the lyrical character of the movement notation instead of tr., which for Clementi, Czerny
may lend support to the Haslinger marking, which,
however, is very significantly faster than the conven- 166 Czerny: Die Kunst des Vortrags, p. 34.
167 “Dieser in einem ruhig edlen, melodiösen, aber auch humo-
tional 20th-century tempo for this movement. In any ristischen Character geschriebene Tonsatz muss mit Zartheit und
case, there is no reason to think that Beethoven envis- Gefühl in einem gemässigsten Tempo (beinahe Tempo di Menuetto)
aged a rigid adherence to the opening tempo. vorgetragen werden, da er weder brillant, noch mit irgend einem
Aufwand von Bravour gespielt werden darf. Die Mittelmelodien
mit Anmuth und Delikatesse, die Terzenpassagen deutlich und
165 “an der Spitze bleiben” legato.”

and others indicated a trill without turned ending; probably thought that, since, especially in earlier
the wavy line, though, could easily have conveyed works, he frequently marked a turn after a trill, the
other meanings during the first decades of the cen- fact that he did not do so here was a positive injunc-
tury. Without any such indication, Beethoven must tion not to play one. In 1931 the pianist Ernst Denhof
have known that tr, whether followed by a notated wrote: “As is well known, the principal subject in
turn or not, would signify a turned ending to the the first movement of this work begins on the third
performer. beat with a short trill, for which Beethoven marked
The two earliest editions, David’s and Alard’s, as no final turn. As, however, he generally wrote very
well as the much later ones by Joachim, Brodsky, exactly, especially in his later works, to which this
Halir, Seybold, and Kreisler, give no guidance about sonata belongs, it is evident that he did not want a
the performance of the trill. Alard, though he did final turn. Both ways, with and without final turn,
not mark it, almost certainly took a turned trill for have their advocates, and as the passage is repeated
granted, since in his École de violon he instructed: 27 times, it is necessary that both players be agreed
“The trill should always have an ending, it is not upon the point beforehand. In my experience the
complete without it.” 168 And in the alternate cello majority of professionals play it without – as I did
version by Auguste Franchomme that accompanied myself.” 171
Alard’s edition, the turn is marked in the first bar. Already, in 1887, Speidel/Singer had added a foot-
David, too, undoubtedly assumed it; in his Violin note to the first trill: “all the trills in this movement
schule he states: “As a rule, one begins the trill on are to be executed without a turn.” 172 In 1917, Auer
the lower note and makes a turn.” 169 Only in chains gave a footnote showing this realisation of the trill
of trills does he admit the omission of the turn. from the main note and without turn:
# œ œ œ
œœœœ -
David’s Leipzig colleagues, Reinecke/Hermann and
& œ
Grützmacher, add a turn after the trill throughout J
the first movement of op. 96. They probably felt the
need to do so because by the time their editions ap- Another of Joachim’s pupils, Ossip Schnirlin also
peared in the 1890s the propriety of a turn, where instructed in 1925 that the trill should be without
Beethoven had not specifically marked it, was al- turn (ohne Nachschlag).173 Max Rostal in 1981 still
ready becoming an issue. It is quite possible that asserted that such ornaments should be used only
an early-19th-century tradition of trill performance where expressly prescribed by the composer, but
was passed on to these Leipzig musicians not only acknowledged that Adolf Busch and Rudolf Serkin
through Ferdinand David’s teacher, Spohr, but also had added a turn in their 1950 recording.174
through Ignaz Moscheles (their colleague at the Leip- Joachim, the acclaimed master of Beethoven per-
zig Conservatorium between 1846 and 1870). Grütz- formance in the second half of the 19th century, left
macher, in a letter of 24 December 1876 to Dr. Max the trill sign without explanation in his 1901 edition.
Abraham of Peters Edition, about his edition of This is the case also in Joachim-revised. In a much
Mendelssohn’s cello sonatas, wrote that he had de- later reprint however (of which I [CB] received a
rived information about performing practices from new copy as a school prize in 1965), there is a foot-
Moscheles, David, and Julius Rietz.170 note, ‘siehe Vorwort’, in the piano score, referring to
The growing reverence for Beethoven’s text, as op- a Forword (Vorwort) which appears never to have
posed to the messages that had until then been read been printed, probably referring to the execution of
between its lines, caused later 19th- and 20th-century the trill.
musicians to regard his notation as, in many re- Reports of Joachim’s practice are contradictory, sug-
spects, much more exact than it actually was: it was gesting that he sometimes played the trill with a
turn and sometimes without. Jelly d’Aranyi, writing
168 École du violon Méthode complète et progressive à l’usage du Con-

servatoire (Paris, 1844), p. 30. “La Cadence doit toujours être termi- 171 The Scotsman, June 27, 1931, p. 18.
née, elle ne serai pas complète sans cela.” 172 “Sämtliche Triller in diesem Satz sind ohne Nachschlag aus-
169 David: Violinschule, vol. 2, p. 42. “In der Regel fangt man den zuführen.”
Triller mit dem unteren Tone an und macht einen Nachschlag.” 173 Ossip Schnirlin: Der neue Weg zur Beherrschung der gesamten Vio-
170 Quoted in Bennett Wadsworth: ‘Precisely marked in the tradition linliteratur (Mainz/Leipzig: Schott, 1925), vol. 3, p. 42.
of the composer’, p. 122. 174 Rostal: Beethoven, p. 169.

in 1927 stated: “I have played it with several pia- ly be heard, and it is in such cases nothing more
nists who have often played it with my great un- than a strong initiation of the trill.” 178
Ω
œt. . œœœ ‰ œ. œœœ ‰ œ. œ œ œ œΩ œΩ œΩ Œ
cle, Joachim – half of them say he ended the trill, t. t.
the others say he did not, so he obviously played it & b c

J J
both ways.” 175 But a preference for the turn is sug-
gested by a number of sources. Joachim was one of It may also be significant that the violin part of
the initiators of Adolf Beyschlag’s Die Ornamentik the op. 96 sonata was intended for performance by
der Musik, in which Beyschlag, in his discussion of Pierre Rode, whose own compositions and the Paris
Beethoven’s trills, gives this op. 96 trill, as Fig. 69, Conservatoire Méthode de violon, of which he was a
with a turn, and writes: “In our opinion, all the co-author, make it clear that such trills begin with
trills from Figs. 62 to 70 and many others require a the upper auxiliary and have a turned ending.
turn, although B. did not notate it.“ 176 The strong- The weight of evidence, therefore, indicates that
est evidence for Joachim’s practice, however, is his Beethoven expected his tr to represent a trill begin-
1905 Violinschule, where a footnote in the section on ning with a very short upper auxiliary and a turned
ornaments refers to “the much-disputed theme of ending. Some of his contemporaries may have been
the Beethoven G major Sonata for piano and vio- inclined to begin the trill from the upper auxiliary
lin” followed by the explanation that the trill must and some from the main note, but it seems highly
have “ a proper turn if the theme is not to forfeit its unlikely that any would even have considered end-
natural charm,” and a music example:177 ing it without a turn.
#3 Ÿ œœœ Spohr’s advice for the early beginning of a trill in
& 4œ Ausführung: œ œ œœœœœœœ œ œ œ
(played) Rode’s 7th Violin Concerto is probably also relevant
p
9 here; it is very probable that early-19th-century vio-
The matter of the trill beginning is equally unclear. linists and pianists would have had no compunc-
By the late 19th century, following the instructions tion about beginning the trill slightly early in order
for a main-note start in the treatises of Hummel to achieve a couple more alternations. (For this and
(1828), Spohr (1833), and Czerny (1839), there was a general discussion of the notation and execution
no apparent doubt that it began on the main note, of trills in Beethoven’s music see 5/b/ii.) See also the
despite the vast majority of earlier 19th-century note to b. 140f
sources that specified an upper-auxiliary start as 0. Vl: Few editors mark a bow direction for the open-
standard in such circumstances. In view of the ing trill, perhaps assuming , which is only marked
overwhelming predominance of upper-auxiliary be- by Seybold. Brodsky and Singer, however, mark .
ginnings to trills specified in Carl Czerny’s arrange- 2. Vl: Most editors mark  here, but they break Beetho-
ments of Beethoven’s ‘Kreutzer’ Sonata, made dur- ven’s slur from 4i to 6iii, perhaps imagining a slower
ing the composer’s lifetime, his earlier practice (as tempo than seems likely to have been Beethoven’s
Beethoven’s pupil) is quite clearly at odds with his intention. The long slur, however, is perfectly feasi-
later teaching. It is extremely likely that Beethoven ble at a moderately fast speed in a piano dynamic.
expected the trills to start from the upper auxiliary, 2. Pno: In a footnote Ganz shows “Various readings of
in the manner described by Leopold Mozart, when, the trill used by the author”:
after giving examples where the trill begins with a (a) - œ œ œ (b) - œ œ œ
long appoggiatura, he wrote: “If, however, a pas- ?# œ œ œ œ œ ?# œ œ œ œ œ œ œ
sage begins with a trill, the appoggiatura will hard- 1 3 1 3 2 1 3 1 3 1 3 2
>œ œ œ
(c) œ œ œ
175 Jelly d’Aranyi, in: Music and Letters, 8 (1927), p. 191.
? # œ œ œ œœœ
176 Adolf Beyschlag: Die Ornamentik der Musik (Leipzig, 1907), 2nd 1 3 1 3 2 1 2
ed., 1908, p. 219 (Fig. 69 on p. 218). “Unserer Ansicht nach verlangen

alle Triller von Fig. 62 bis 70 und viele andere einen Nachschlag, 2f. Pno: Throughout this movement all chords, par-
obschon B. denselben nicht ausnotiert hat.” For further contextu-
alisation of Joachim’s attitude see Gebauer: Der „Klassikervortrag“,
ticularly but not exclusively those of harmonic im-
pp. 379ff.
177 Joachim and Moser: Violinschule, vol. 1, p. 164. “das vielum- 178 Leopold Mozart: Versuch einer gründlichen Violinschule (Augs-
strittene Thema der Beethovenschen G dur-Sonate für Klavier und burg, 1756), S. 223 “Wenn aber eine Passage mit einem Triller an-
Violine Op. 96” – “einen regelrechten Nachschlag, wenn das The- fangt, so wird der Vorschlag kaum gehört, und er ist in solchem
ma seine natürliche Anmuth nicht einbüssen soll”. Falle nichts denn ein starker Anstoß des Trillers.”

portance or on main beats, might be arpeggiated 48. Pno: In the autograph, Beethoven also wrote the
perhaps with variation of speeds so that, for exam- rh part below the stave in pencil and added the
ple, half-note chords are slower than quarter-note instruction dimin.; this does not appear in the edi-
chords (5/c/ii). tions. It would, in any case be a natural dynamic
10ff. Pno, Vl: Beethoven would almost certainly have shaping for this phrase, which would have been in-
expected some inequality of rhythm in such a pas- stinctively applied by any sensitive musician.
sage of continuous 8th-notes. Speidel/Singer provide 49iii. Vl: Singer, Seybold mark a harmonic.
the interesting instruction molto tranquillo ed equale. 49–53, 188–192. Vl: A variety of different bowing
This indicates that they did not expect equality of strategies are suggested. Alard, using a typically
rhythm as a matter of course (2/b). In the Pno, over- French approach, begins  (though at 49 he forgets to
holding would have been expected and normal (4/a/ mark it) and takes everything with alternate bows.
ii). This is also given by Joachim, Singer, Kreisler. Da-
17v. Vl: Alard, Singer mark a harmonic; it is implied vid, Brodsky, Halir, Rosé, Seybold hook all the dot-
in David, Hermann, Seybold, who move to 1 on 17iv ted figures beginning . The others use a mixture.
(Singer evidently expected a slide from 4 on 17iv). Singer specifies Fr[osch] in 50, Sp[itze] in 51.
Joachim and other editors move to 4th position on 49–54iii. rh: Although no slurring is marked from
17iv. 49–53, it seems most likely that Beethoven expect-
20, 21, 23, 24, 28. Vl: David marks slurred staccato on ed legato. The addition of a slur on 54i–iii in the
the repeated 8ths, but this was removed in David- Vienna first edition may suggest a correction in a
revised. It would, however, have been a very typical lost copied source, but the motivation for adding
bowing in Beethoven’s time and quite likely used a slur only in 54 is unclear, unless it was only at
by Rode. Slurred staccato could be delivered with this point that Beethoven wanted overheld legato
various degrees of separation, from an essentially (but such a level of detail seems rather unlikely, and
portato articulation to a martelé style staccato. there is no slur at the equivalent place in 193). In any
20ff. Pno, Vl: The slurred duplet 8th-notes would case, the musical context seems to demand overheld
surely have been nuanced with the first note longer legato despite the absence of slurs. Beethoven’s use
and stronger than the second, a practice explained of slurs on the runs in 33, 35, and 37– 40 argues
by L. Mozart and Quantz and still to be heard in against his use of slurring to indicate overheld legato
the early 20th century as evidenced in recordings (Milchmeyer’s gebundenes Spiel) as distinct from nei-
(2/a). ther slurs nor staccato for normal legato (4/a/ii). Spei-
24ii–iii. Vl: Alard, Singer mark a harmonic on ii (Sin- del adds slurs, but many other editors, including
ger with 2) and 1 on iii. Reinecke and Joachim retain the text of AG, which
41– 46, 180–185. Vl: Singer, Auer marked slurred stac- adds no slurs here.
cato for the 2nd and 3rd notes of each triplet. 54iii. Vl: Alard, Seybold mark a harmonic; it is im-
41–54, 180–193. Pno, Vl: It is unclear whether Beetho- plied in David.
ven expected a more or less 3 : 1 dotted rhythm, an 55–57. Vl: It is unlikely that Beethoven expected a
over-dotted rhythm against the triplets, or simply sharp staccato separation here rather than a distinct
used this notation to indicate a 2 : 1 ratio. Since he phrasing. David, Rosé hook in the separate 8ths,
never used modern triplet notation for these kinds clearly in the upper part of the bow, all the others
of unequal patterns, his rhythmic expectations can except Halir, Seybold (whose bowing brings them
only be determined from the context. Here it seems to a down bow in 55) take the slurred pairs with .
very likely that he intended triplet rhythms. In any Singer specifies Sp[itze].
case, he will surely not have expected metrically 63, 202. rh: The trill probably with a main-note start.
strict performance. The third movement of the Piano 58ix. Vl: David, Singer, Brodsky, Halir, Seybold, Kreis-
Trio op. 97, where both dotted 8th–16th and 8th–8th ler mark a harmonic.
could both be read as synonymous with 2 : 1, pre- 58. Pno: Speidel marks all chords with portato articu-
sents similar problems (see 2/c /iii). lation.
46. Pno: A swift arpeggiation of the chord marked sf 59. Pno: Halir marks dolce.
would enhance the accent while mitigating harsh- 67iii. Vl: A shift up the G-string to a harmonic is given
ness (PT: 1/a/v). by Alard, Singer (with 3), Auer (with 3); Brodsky

marks the harmonic but also takes the previous note could be equally effective to use a long, fast, light
in 3rd position. bowstroke between the first and third quarters of
68–70, 207–209. Pno: Beethoven’s notation for the trills the bow, or a shorter, slower, more intense stroke,
(with quarter-notes) is unusual. Reinecke’s finger- focused on the middle of the bow.
ing (with main note start) shows that he did not 115–120. Vl: David, Alard, Joachim, Brodsky, Halir,
consider the trill to be continuous giving 1–3–2 for Seybold remain in 1st position almost certainly en-
each note in rh. visaging an open E-string on 119iii and in 120 (Brod-
71, 210. Pno, Vl: Speidel/Singer mark ’ at the end of sky specifically marks it); others shift to 2 in 120 after
the bar. 0 on 119iii; Singer goes to 5th position on the A-
72. Pno: If the decision is made to observe dolce (pre- string at 119ii and back to 3rd on 119iii; Rosé moves
sent when this material returns at 211), then the use to 4th position on 117ii.
of a moderator if available would be appropriate 116. Pno: To Beethoven’s sempre p, Ganz adds e legato.
(PT: 3/b), as would the use of noticeable arpeggia- 123ii, iii, 124. Vl: Harmonics are marked by Alard, Her-
tion and asynchrony (5/c/ii). mann, Joachim, Singer, Auer, Kreisler. Auer speci-
72i. Vl: Kreisler marks a harmonic. A fingering of this fies the harmonic also in 124, and the others, who
kind would be typical of David and probably of mark nothing there, evidently intended it. Kreisler
many early-19th-century violinists, but David does marks 1 in 124. Halir, Brodsky mark a harmonic
not mark it here, evidently regarding it as obvious, only in 123ii, followed by 1. David almost certainly
since he makes no change from 3rd position. expected harmonics.
75, 84 – 86. Pno: The portato articulation inspires slight 127ii. Vl: The harmonic is marked here by Alard, Sin
arpeggiation, but with the notes given “the same ger, Brodsky, Halir, Seybold, Kreisler. David, Her-
length of time as a dot under a slur requires” as rec- mann, Joachim probably took it for granted.
ommended by Moscheles in his Studies for the Piano 139. Pno: Ganz explains “The editor plays the right
Forte op. 70, Bk 1 (1827). hand under the left”:
84ff, 188ff, 223ff. Pno: The triplet figurations should 7
œ œ œ
# œœ œ œœ œ œœ œ œœ Œ Œ
4
2
surely be legato as marked by Speidel, Diémer. In
84, Ganz marks espressivo in rh. &
p 2
89i, 90i, 91i, 91iii, 92iii, 93iii. Vl: Rosé marks these
4
notes <>, probably envisaging a soft vibrato accent. 139f. Pno, Vl: Beethoven surely expected a turn at the
91iii–95i, 230iii–234i. Vl: It is perfectly feasible to play end of these trills. C. Ph. E. Bach specifically stated:
this passage in a single bowstroke, especially at tem- “A trill without following notes […] always has a
pos suggested by the earliest metronome marks and turned ending.” 179 Grützmacher and Reinecke/Her-
Beethoven’s own tempo preferences. Brodsky is the mann mark the turn on the trill here. Joachim evi-
oldest of the 19th-century editors to suggest a bow dently used it, as an anecdote recorded by Ernst
change, probably reflecting the growing tendency Denhof demonstrates. Following on from the ex-
to take this movement more slowly than Beethoven tract quoted above (note to 0, 1, 2 etc.) Denhof states
and his contemporaries envisaged. that Joachim agreed to play all the trills without
91iii–95i. Vl: Although none of the 19th-century edi- turn at a concert they gave together in 1904, but he
tors suggest it, it would not be inconceivable that a adds that in their rehearsal, “at the end of the devel-
violinist of Beethoven’s time might have played this opment before the re-entrance of the theme where
passage in 1st position with an open A-string. the trill occurs four times, without the notes e, d, b,
98–115. Pno: In this texture, much expression can be which follow in other instances, Dr Joachim inter-
created through asynchrony of important melody rupted, and after a moment’s thought, remarked, ‘I
notes (octaves) in rh, played slightly after the lh (5/c/ almost think a final turn should be made at this
ii). place!’ ” 180
108. Pno: Ganz marks smorz.
109ii. Vl: Singer marks espress. 179 Carl Philipp Emanuel Bach: Versuch über die wahre Art das Cla-
109ii–114i. Vl: Most accomplished violinists of Beet- vier zu spielen, 3rd edn. (Leipzig, 1787), vol. I, p. 54. “Ein Triller
ohne folgende Noten […] hat allezeit einen Nachschlag.” This final
hoven’s time would have created the tone colour sentence to §. 13 was an addition in the 1787 3rd edition.
with the bow, probably with little if any vibrato. It 180 The Scotsman, June 27, 1931, p. 18.

159f. Vl: David marks slurred staccato, removed in 1ff. Pno: In a slow movement of this character notice-
David-revised. able arpeggiation (with varying speeds) and asyn-
196vii–197ix. Vl: Singer marks espressivo. Hermann, chrony would have been expected and normal (5/c/
Singer mark the passage on the D-string, with a ii). Speidel adds cantabile. He also changes the slur-
harmonic on 197vi. ring in the middle voice, adding shorter phrasing
196. Pno: Staccato is marked by Speidel, Diémer, Halir. slurs with a note: “The 16th-note accompaniment in
260–261. Pno: Surely legato, as marked with slurs by the right hand pp”.182 Ganz adds Con Pedale.
Reinecke, Speidel, Diémer, Halir, Vogrich. 2, 17, 18, 39. Pno: Beethoven’s < > invites linger-
262, 265. Pno: A trill from the upper auxiliary is indi- ing and perhaps noticeable arpeggiation at the apex
cated by Ganz in 262 rh and 265 lh. (3/b/v). Most editors place the apex on or near to the
267. Pno, Vl: Reinecke/Hermann give the grace-note, second 8th-note (although Beethoven’s autograph
suggested editorially in the present edition, at the clearly locates it on the 3rd 8th-note). Halir changes
end of the long trill. this so that the hairpin extends across 1 and 2 with
the apex on 2i.
Adagio espressivo 7–8, 44. Pno: Beethoven’s < > invites an increase
Tempo of momentum towards and lingering at the apex
Beethoven’s treatment of Adagios in 2/4 is considered (perhaps noticeable arpeggiation at the apex), with
in the introduction to the Adagio molto espressivo of a return to tempo afterwards (3/b/v).
op. 30 no. 1 above. Despite the similar tempo term and 9, 38. Vl: Beethoven used the terms sotto voce and mezza
the employment of similar note values (although the voce very rarely. His treatment of them in the Al-
op. 96 adagio has a few more fioriture of notes faster legretto ma non troppo in his String Quartet op. 95,
than 32nds), most editors give very different metro- however, provides a valuable clue to their impli-
nome marks for those two movements. Moscheles, cations at this period of his life. At the beginning
however, is consistent and plausible, giving the same of the Allegretto the cello, playing alone is marked
marking here as in op. 30 no. 1, where his marking was mezza voce, but when the other instruments come
slower than all the others except Alard; here he is fast- in, in bar 5, the 1st violin which has the melody, is
est. The great distance between the Haslinger/Czerny marked mezza voce, while the other three are mark
Vortrag numbers in these two movements cannot be ed p, implying that mezza voce implies a dynamic
explained by anything in the notation, which suggests level somewhat more than p. At bar 112, where the
that one or both of these markings failed to represent opening material returns, the cello solo is marked
Beethoven’s envisaged tempo. For ease of reference, sotto voce, following immediately from a pp dy-
the numbers for op. 30 no. 1 are given in a second namic, and when the other instruments enter, they
column. are all marked dolce. Perhaps these terms also im-
op. 96 op. 30/1 ply a special quality of tone, sotto voce, more veiled
Haslinger  = 56  = 76 than mezza voce, but still with rather more carrying
Moscheles  = 63  = 63 power than pp.
Czerny Vortrag  = 58  = 72 9–11. Vl: Only Alard retains Beethoven’s slurring, but
Alard/Diémer   = 76  = 58 gives no guidance as to its execution. All the other
Speidel/Singer  = 40  = 69 editors divide it. None suggest the very plausible
Kreisler  = c. 56  = c. 72 treatment of the repeated notes under a slur, which
Czerny writes tersely: “Calm, serious, and with all ex- would be to keep the bow moving steadily while
pression that agrees with this character. The scherzo changing the finger. This practice is described by
follows directly.” 181 Spohr as an imitation of vocal practice; after dis-
Moscheles’ tempo seems very plausible, but at what- cussing vibrato (Tremolo) he writes: “By changing
ever basic tempo is selected, a considerable degree the fingers on a note a vocal effect is also imitated,
of freedom, though always centred around the basic namely the separation of two sounds, sung at the
tempo, was probably envisaged. same pitch in one breath while pronouncing a new
syllable.” 183 This was surely used also by Rode and
181 “Ruhig, ernst, und mit allem Ausdruck, der mit diesem Cha- 182 “Die Sechzehntelbegleitung der rechten Hand pp”.
racter übereinstimmt. Das Scherzo folgt zusammenhängend.” 183 Spohr: Violinschule, p. 175. “Durch das Wechseln der Finger

other contemporaneous violinists, and may well have Whether any of these practices had a direct tradi-
been in Beethoven’s mind when he slurred these tion going back to Vienna and Beethoven’s expec-
repeated notes together. A fingering of this kind is tations for the performance of this kind of theme
suggested in the edited violin part of the present is uncertain. David’s experience, his background as
edition. a Spohr student, and his tendency to enhance his
11vi–18. Vl: The significance of Beethoven’s espressivo personal copies of his own editions with additional
instruction is thought-provoking. All begin in 3rd portamento fingering (for instance in a sonata at-
position and none of the editors except Rosé sug- tributed to Geminiani in David’s Hohe Schule des Vio-
gest any unnecessary position changes until 16, linspiels http://mhm.hud.ac.uk/chase/view/pdf/178/2/
where all shift back to 1 for the c2 and stay in 2nd #page) may suggest that the ‘purity’ of approach
position until shifting back to 3rd on either 18ii or here owed more to a later 19th-century veneration
iii, the latter, marked by David, Hermann, Joachim, for the sublimity of Beethoven’s music than to prac-
Auer, Halir, Rosé, Seybold, suggesting a delicate tices with which the composer himself would have
portamento connection. Except for Rosé, therefore, been familiar. It is arguable that many of the editors
portamento was not an obvious response to ex- felt an obligation to present the notation in a neu-
pressiveness in this passage. According to Marion tral guise, which, in their own performances, they
Ranken, Joachim’s typical approach to espressivo might not have observed. David’s radical alterations
in Beethoven involved “an intense and pure tone, in personal copies of his own editions, and Joachim’s
which can be produced without any vibrato what- propensity for playing spontaneously, may suggest
ever”; this was achieved by the use of a slow-mov- that their editions do not come close to reflecting
ing bow (usually in the middle) and was used in what they would have done in practice. Given the
all dynamics from pp to ff. She added that “Joachim widespread 19th-century admiration for the vocal
very generally used this sort of tone in deep and qualities of violin playing, it may be plausible that
intense passages, such as those which occur so of- subtle and delicate portamento might, in fact, reflect
ten in Beethoven.” 184 A little later, she commented: early-19th-century ideals for a melody of this char-
“no one who listened appreciatively to his playing acter. Spohr’s fingering for the Adagio of Rode’s 7th
will ever forget the stillness and grand simplicity Concerto, undoubtedly reflecting his recollections
of the way he so often played slow themes of Beet- of Rode’s playing in 1803, depends significantly on
hoven, allowing himself not one single slide when portamento for its expression; David’s own edition
avoidable or one hint of vibrato, but remaining una- of that concerto, based, according to his own state-
bashed in the low positions, using fingerings such ment in a footnote, on Rode’s teaching of it, as com-
as would probably be chosen for a child in its first municated to him by the composer’s pupil Eduard
lessons.” 185 Rietz, has exactly the same fingering as Spohr’s.
David’s edition differs radically from the others, in- 12. Pno: Speidel adds in a footnote: “b) The pedal
cluding David-revised, in his use of separate bows may only be held until the sixteenth rest in order
in 12 and 15f. Doubtless the effectiveness of his per- to avoid any harmonic impurity compared to the
formance depended upon his ability to make vir- violin part”.186
tually undetectable bow changes. Perhaps he used b
a slow bowstroke, remaining in the middle of the &b b œ œ
bow, similar to the one described in his pupil b
&b b ≈ œ œ ≈ œ œ

Joachim’s performance of expressive passages. œ œ œ œ
? bb ≈ œœ. œœ. ≈ nœœ. œœ.
b)
b
°* ° *
auf einem Ton wird ebenfalls etwas, dem Gesange angehörendes
nachgeahmt, nämlich das, durch das Aussprechen einer neuen 19. Vl: Alard, Hermann, Kreisler are the only edi-
Sylbe bewirkte Trennen zweier, auf derselben Klangstufe befindli-
chen und in einem Athem gesungener Töne”.
tors to leave Beethoven’s whole-bar slur unbroken,
184 M[arion] [Bruce] R[anken]: Some Points of Violin Playing and though whether in practice these violinists (espe-
Musical Performance as learnt in the Hochschule für Musik (Joachim
School) in Berlin during the time I was a Student there, 1902–1909 (Ed- 186 “Das Pedal darf durchauss nur bis zur Sechzehntelpause ge-
inburgh, privately printed, 1939), pp. 12f. halten werden, um jede harmonische Unreinheit, der Violinstimme
185 Ibid, p. 16. gegenüber, zu vermeiden”

cially Alard, at his exceptionally slow tempo) would as a “violinistic virtue that cannot be sufficiently
have played it in a single bow is questionable. At praised”: the ability and practice “to achieve an im-
Moscheles’ tempo, however, it is certainly feasible. perceptible bow change at the frog.” 188
Singer changes bow on an off-beat (ix) to enhance Perhaps surprisingly, none of the editors retain the
the connection and provides a footnote “The bow long slur, or merely mark bow changes above it;
change to be made as un-noticeable as possible.” 187 they give shorter slurs. The approaches to bow divi-
He also continues the slur to 20i. David, followed by sion here, which occur in David-revised and Alard,
Rosé and Seybold, makes two changes (on ii and x), reflect the two distinct approaches that dominate
perhaps taking Beethoven’s molto dolce to imply the the later editions. David-revised (followed by Brod-
kind of quasi-flautando recommended by his pupil sky, Rosé, Seybold), marks changes of bow on the
August Wilhelmj for the achievement of dolce (3/d/i). half bar until 30, whereas Alard (followed by all
Since dolce seems also to have had connotations of a the others except David) regularly changes at the
somewhat broader tempo, it might also imply that beginning of the bar. David’s original edition, as
the 64th-notes should be played in a very unhur- elsewhere, contains many more bow changes: he
ried manner. provides no slurs except from 27ii–28i (evidently for
All begin in 3rd position. Joachim, Singer, Auer, the sake of portamento).
Kreisler remain until they shift to 5th position on Joachim offers the simplest fingering: 3rd position
xvi, and they then shift back on 20i; Halir goes back to 29 and 1st from 30i; Hermann, Singer mark the
to 1st position on vii before continuing the bar en- whole theme on the A-string with inevitable porta-
tirely on the D-string. The others shift to 1st posi- mento from 27i–ii; others begin in 3rd position and
tion on vii (David, Rosé, Seybold) on x and then go to 5th on 28i, all except Auer with a connecting
to 3rd on xxii, certainly with an audible, if delicate bowstroke from 27ii, hence portamento. Some por-
connection. tamento is also likely to have occurred where posi-
20. Vl: Only Alard leaves Beethoven’s slur intact. Most tion changes occur between bowstrokes, to enhance
change bow at the half bar; David (followed by the legato. The expressive choices are clearly indi-
Seybold) takes four bows, again suggesting a very vidual. All except Joachim include a portamento
fast light bowstroke. David also uses the greatest fingering between 31i–ii.
number of position changes with portamento im- 32–35. Vl: The fioriture were surely expected to be
plications (again followed by Seybold), with audible played very freely and (especially at Moscheles’ met-
shifts from iv–v (2–2) and vi–vii (4 –3); most others ronome mark) in a more leisurely tempo. In 35 the
use an inaudible fingering (1–2) for vi–vii. basic tempo would probably have been gradually
22, 24. Vl: The predominant fingering in 22, marked re-established.
or implied by all except Brodsky, Kreisler, is from 35–36i. Vl: All remain on the A-string. David, Alard,
1st or 2nd position to 4 on 22ii and 4 again on 23i. Hermann, Singer stay in 3rd position and use a 4th
Brodsky, Kreisler go to 5th position on 22ii, presum- finger extension for 36i; the others go to 4th posi-
ably for the sake of a better vibrato in 23. On 24i–ii tion on xxix.
all shift from 1st to 3rd position. A degree of porta- 37. Pno, Vl: Speidel/Singer mark ’ at the end of the bar.
mento is absolutely necessary to make the smooth- 37ii. Pno: Speidel explains in a footnote: “The outer
est possible connection, as Beethoven will surely voices move here in [parallel] octaves. Was there an
have envisaged. error in the original copy? The octaves could easily
25–32. rh: Undoubtedly legato. Reinecke, Diémer mark be avoided by changing the way the bass is oper-
slurs. Speidel, Ganz mark sempre legato. ated, as follows“:189
26–31. Vl: Beethoven’s continuous slur here, simply ? bb œ
signifying legato, suggests that he was unwilling to
b nœ œ œ
specify bow changes, perhaps because this might
188 Joachim und Moser: Violinschule, vol. 3, p. 18. “nicht genug zu
lead to a break in the legato. The player’s task here, rühmenden geigerischen Tugend” – “das Wechseln des Bogens am
therefore, is to utilise what Joachim’s colleague Frosch unmerklich zu bewerkstelligen.”
Andreas Moser described in their joint Violinschule 189 “Die äussern Stimmen schreiten hier in Octaven einher. Sollte
hier nicht ein Fehler in der Originalcopie zu Grunde liegen? Die
Octaven wären durch eine andere Führung des Basses leicht zu
187 “Den Bogenwechsel möglichst unbemerklich zu vollziehen” vermeiden, wie folgt: [Ex.]”

The octaves are, however, clear in the autograph 3rd position throughout, some shift to 3 on the final
and all the sources. note of the bar. Only Kreisler shifts back by a semi-
38. Vl: mezza voce: see note to 9, 38 above. tone to 2nd position on vii.
38– 46. Vl: What Beethoven may have intended by
semplice is unclear. At this date it might still have Scherzo
functioned as a warning not to embellish the me- Allegro
lodic line. Tempo
Here too Beethoven’s long legato slurs are divided. Beethoven’s tempo conception for scherzos is dis-
As elsewhere, David (but not David-revised) has the cussed above in relation to op. 24 and op. 30 no. 2,
most bow changes, talking 38– 44 with two bows where many of the editorial suggestions seem much
per bar, and three in 45. Again, this may be his re- slower than Beethoven is likely to have envisaged.
sponse to mezza voce with a fast, light, almost whis- Here, the Haslinger tempo seems very plausible,
pering bowstroke. Others mostly mark one bow per matching Beethoven’s own metronome mark for the
bar or occasionally two. third movement of the Fifth Symphony, where the Trio
Most remain on the D-string until 45iv; Brodsky in- contains many 8ths. Since the 8ths in the Trio of this
dicates G-string for the whole passage; Auer, Halir, scherzo are slurred rather than non-legato, as in the
Kreisler go to the G-string in 42. Symphony, the tempo might perhaps be expected to
48. rh: Speidel adds cantabile at the end of the bar. be a little faster, as in the third movement of the
Ganz adds espr. String Quartet op. 18 no. 3, for which Beethoven gave
49i. Vl: Auer marks a harmonic, which would have 100. Speidel/Singer add ma non troppo in parentheses.
been a very plausible fingering in the early 19th-cen- Haslinger . = 96
tury. Others who mark nothing (David, Hermann, Moscheles . = 80
Joachim, Singer, Brodsky, Seybold) probably envis- Czerny Vortrag . = 80
aged it as a possibility. Alard/Diémer . = 72
54i–ii. Vl: All shift 1–1. David omits the slur, which is, Speidel/Singer . = 72
however, inserted in David-revised. Kreisler . = c. 56
55. Vl: David has no slur (one was inserted in David- Czerny described the movement as “Also serious, but
revised). lively and very humorously accented, since the witty
58–61. Vl: Beethoven’s slurring is again broken up in effect is especially in the sfp of the 3rd quarter-note
different ways. All except Rosé, Seybold, Kreisler beat. The Trio gentle and legato, but just as fast.” 191
omit the slur in 58, which is absent in AG. Alard, 0–32. Pno: The chords marked sfp might be arpeggiat-
Joachim, Kreisler change at the beginning of the bar. ed swiftly while the following note to which it is
David slurs pairs across the beat from 59ii–62i and slurred is not (5/c/ii; PT: 1/a/v). Other chords (par-
others follow him exactly or in part. Singer marks ticularly on quarter-note down-beats) might also be
a portamento line over a bow change between 59i– arpeggiated. Shorter value chords could probably
ii. Various portamento fingerings are marked; the be played with all notes together as advised by
largest number are given by Singer (who seems, Czerny.
however, to have forgotten to mark 2 on 61i). Vl: The editions provide little evidence of how indi-
61f. Pno: Speidel offers the following alternative from vidual violinists envisaged the style of bowing for
the middle of bar 61 “for those with small hands”:190 this movement, which would in any case have been
5 affected by the tempo they chose to take. Alard,
j 2
j Joachim, David-revised, Brodsky, Kreisler leave the
? bb œ œ œ œ œ œ œœœ œ œ œœ.. ≈
notes as Beethoven wrote them without any bow-

b ≈ J
? bb ≈ œ j ing instructions except  on the first note; Rosé,
œ œ. ≈
b œ Seybold also leave the notes with no other bow-
ing instructions than  at the beginning, but add
œ œ.
62f. Vl: All seem to envisage the passage on the A- staccato marks. Hermann, Singer, in bars where the
and D-strings until the last note. Some mark it in
191 “Auch ernst, aber lebhaft und sehr humoristisch markirt, da
die launige Wirkung besonders in dem sfp der 3ten Taktviertel liegt.
190 “Für kleinere Hände” Das Trio sanft und legato, aber eben so schnell.”

second and third beats are 8ths, mark a slur over Allegretto con Variazioni of the String Quartet op. 74,
these notes, to which they also add staccato marks; which he marked  = 100. In the variations of that
in these places, David, Auer, Halir mark a slur not movement, two have a substantial number of 16ths.
over two, but over three notes (the two 8ths and the This movement has a similar range of note values
first beat of the next bar); Auer, Halir also add stac- before the change of tempo, as with other variation
cato marks on all three notes. In addition, Halir in- movements, a certain flexibility of tempo was surely
structs that the slurred notes should be performed expected. The designation Poco allegretto probably in-
in the middle of the bow; but this need not have dicates a somewhat slower tempo than a plain Alle-
been the bowstroke envisaged by David or Auer. gretto, but Czerny’s 1846 marking, significantly slow-
It is quite likely that David, especially, envisaged er than the one in the Haslinger edition, may reflect
a stroke close to the point of the bow, similar to later practice. Haslinger’s and Moscheles’ tempo are
the slurred staccato. That kind of bowstroke would probably closer to Beethoven’s expectations.
probably have come naturally to many violinists of Only three of the 19th-century editions provide met-
Beethoven’s time, and is envisaged with the bowing ronome marks for the other sections. For the Adagio
marked above the notes in the edited violin part of they range very widely, from  = 72 to 56. Among move-
the present edition. The bowing suggested below ments with Beethoven’s metronome marks is the 6/8
the notes, played in the middle or lower-middle of Adagio sostenuto of the Piano Sonata op. 106, to which
the bow, represents a bowing style that may well he gave  = 92. Like the Adagio espressivo in this sona-
have been more typical of the later 19th century ta it contains 32nd-notes, which become increasingly
than of Beethoven’s time. insistent throughout the movement, but it does not
25iii. Vl: A harmonic is marked by David, Singer, Ha- have the 32nd-note triplets/sextuplets that are present
lir, Rosé, Seybold. Joachim, Auer mark 1. in this movement, which might suggest a somewhat
33. Pno: Beethoven’s dolce invites the use of the mod- slower pulse, but not necessarily as slow as 72, almost
erator if available (PT: 3/a), as well as noticeable certainly not as slow as Moscheles’ 63 and undoubt-
arpeggiation, long and sonorous wherever possi- edly not as slow as Speidel/Singer’s 56.
ble (5/c/ii). Frequent asynchrony between rh and lh Beethoven gave a few metronome marks for fast
would also be stylish. movements in 2/4 (see the commentaries above on the
35–39, 49–58. lh: Undoubtedly to be played legato, as last movements of op. 12 no. 3 and op. 30 no. 3). He
marked by several of the editors. supplied the following metronome marks for plain
43, 59, 63, 75, 77, 79. rh: The grace-note as an acciac- Allegros; for the central section of the second move-
catura, (5/a/iii), played simultaneously with the main ment of the String Quartet op. 18 no. 2, which contains
note and released quickly as explained in Jung- many 16ths he gave  = 69, for the final movement of
hanss’ Pianoforte-Schule (Vienna, c. 1820). op. 18 no. 1, which has triplet 16ths, he gave  = 60
56iii–64ii, 68iii–81ii. Vl: David, Halir, Rosé, Seybold and for the first movement of op. 18 no. 2, which has
consistently mark a slur over the tied note and the 32nds, he gave  = 96. For the Allegro ma non troppo first
one following, but Seybold omits the staccato (per- movement of the Sixth Symphony, which has fewer
haps a printing error); Auer marks it only from 56iii– fast-moving notes, he gave  = 66 (the Allegro ma non
58ii, but probably expected it to continue. The others troppo finale of the Fourth Symphony has  = 80, but
mark no slurs. this anomalous metronome mark is surely either a
115. Pno: Speidel writes in a footnote: “Start the coda misprint,193 or, more improbably, represents a change
a bit calmer”.192 of mind about the tempo term).
123. Pno, Vl: Speidel/Singer advise poco a poco string. Poco Allegro
from the end of 123. allegretto
Adagio Presto
Poco Allegretto espressivo
Tempo Haslinger  = 120  = 72  = 152
The closest parallel for this movement among the ones Czerny Vortrag  = 100
for which Beethoven gave metronome marks is the
193 Clive Brown: “Historical performance, metronome marks
and tempo in Beethoven’s Symphonies”, in: Early Music 19/2 (1991),
192 “Die Coda etwas ruhiger beginnen” pp. 247–258, here p. 249.

Moscheles  = 116  = 63  = 76  = 160 haps vibrato or portamento in Vl at the apex (3/b/v).
Alard/Diémer  = 80 On the repetition of each half, it is quite plausible
Speidel/Singer  = 108  = 56  = 138 that Beethoven and his contemporaries might have
Kreisler/Rupp  = c. 104 –108 added a Schneller or trill in some of these places.
Czerny suggests: “This theme must be performed (at 48 etc. Pno: The ornament sign is likely to have been
very moderate tempo) extremely delicately and with interpreted as a Schneller (5/a/iv).
taste. The following variations a little livelier, and ac- 57–64. Pno: The chords on main beats might be ar-
cented. The Adagio very slow and fantasy-like, the peggiated swiftly (5/c/ii).
following variation, as well as the ending, vivacious 81, 101. Pno: Speidel adds poco marcato.
and powerful.” 194 89, 105. Vl: Singer adds poco marcato.
1. Pno, Vl: Speidel/Singer add e grazioso after Beetho- 90ff. lh: Undoubtedly legato as marked by all the edi-
ven’s dolce. tors.
1, 3, 5, 80, 115 etc. Pno: Beethoven’s dolce invites use 97. Pno, Vl: Espressivo was evidently expected to en-
of the moderator if available (PT: 3/a) as well as courage a noticeably broader tempo, since Beetho-
frequent arpeggiation and asynchrony (5/c/ii). The ven followed it with a tempo. Reinecke adds p cresc.
slurred duplet 16ths might be nuanced with the first with > in lh v–viii.
longer and stronger than the second (2/a). 99. Pno, Vl: All the editors adopt the inauthentic un
9, 11, 13, 25, 27, 29. Vl: Alard, Hermann, David-revised, poco ritenuto of AG, while Speidel/Singer introduce
Brodsky, Seybold, Kreisler take the slurs with sepa- a second un poco ritenuto in 106 and a second a tempo
rate bows. (AG reproduces b. 13 the same as the other in 109 (see Critical Report).
similar figures and all editors follow that reading). 101. Pno: Speidel marks poco marcato.
David, Joachim, Singer, Auer, Halir, Rosé add an- 113. Pno: Speidel marks energico.
other slur over the two slurred pairs. David, un- 145–163. Vl: Rode, noted for his expressive use of
like the others, also includes the preceding quarter- portamento, would certainly have been expected to
note, marking it with – , in his added slur, and in 27, employ it in this section and all the early editors
29 adds a bowing slur over the whole bar. Similar provide fingerings that give ample opportunities
procedures are followed when this material returns for its use.
later in the movement. 148v–xvi. Vl: David, Halir, Seybold remain in 1st posi-
10, 12, 14f, 26, 28, 30f. Vl: David adds a bowing slur tion; David marks 1–1 on xi–xii, but this, strangely,
i–iii; this was removed in David-revised. was removed in David-revised without an alterna-
10ii. Vl: A harmonic is marked by Alard, Hermann, tive suggestion being added, and neither Halir nor
Singer, Auer, Halir, Brodsky, Seybold. David may Seybold mark it. All the others go up the D-string,
well have used one. Singer’s, Auer’s, Seybold’s fin- with Alard, Hermann, Singer (with 2), Brodsky
gering indicates portamento from the previous note. marking a harmonic on xiv, followed by 4 on xv;
15ii. Vl: A harmonic is marked by Alard, Hermann, Joachim probably expected a harmonic since he is
Singer, Seybold. Singer’s fingering indicates porta- in 3rd position on the preceding note
mento from the previous note. 149ii. Vl: All take this trilled note with 1st finger; Her-
26i–ii, 28i–ii. Vl: An audible connection is indicated mann, Singer, Grützmacher, Halir add the conclud-
by the fingering in all the editions. ing turn, Singer, Halir with the fingering 1–1. In this
31ii. Vl: A harmonic is marked by Alard, Singer, Auer, context an upper-auxiliary start would have been
Seybold. Singer’s, Auer’s, Seybold’s fingering indi- obvious to Beethoven’s contemporaries.
cates portamento from the previous note. 149v–xvii. Vl: Various strategies are adopted for fin-
33ff. Pno, Vl: Beethoven’s < > invites lingering gering this passage: 1st position from vii (David,
and perhaps noticeable arpeggiation in Pno and per- Halir, Seybold); 1st position to xvi (Brodsky, Kreis-
ler); 3rd position with marked or implied harmonic
on xiv (Alard, Hermann, Joachim); 3rd and 5th posi-
194 “Dieses Thema muss, (bei sehr gemässigtem Tempo) äusserst tion with 3–3 at xiv–xv (Singer, Auer).
delikat und mit Geschmack vorgetragen werden. Die nachfolgen- 150i–ii. Vl: All except Alard, Hermann, Joachim, Brod-
den Variationen etwas belebter, und markirt. Das Adagio sehr lang-
sam und fantasie-artig, die hierauf folgende Variation lebhaft und sky, Halir give a portamento fingering here, and
kräftig, so wie auch den Schluss.” Singer, Rosé emphasis it with a slanting line.

151i–ii. Vl: A portamento fingering is marked by Sin- 204. Vl: All the editors remained in 1st position, evi-
ger (2–o2 ); Rosé, with the fingering 0–2 marks a por- dently expecting 3 on 205i, but a shift to 2 on 204vii
tamento line; some or all of those who also mark would be a potential fingering to facilitate the 5th
0–1 (David, Joachim, Singer (as an alternative), Auer, across the string with 1st finger.
Brodsky, Halir, Seybold, Kreisler) will have expected 217. Pno, Vl: Speidel/Singer add molto tranquillo. Ganz
a portamento from the open string. Alard marks o marks una corda and misterioso.
on ii. 226iv. Vl: A harmonic is marked by Alard, Singer
152v–vi. Vl: Singer marks a 4 – 4 shift. (with 2), Halir.
153, 171ff. Pno: It is not clear what Beethoven intended 228iii. Vl: A harmonic is marked by Alard, Singer.
with the < > , perhaps a very slight agogic accent. 235i. Vl: A harmonic is marked by Hermann, Singer,
156. rh: The trill is evidently intended to continue un- Auer, Halir, Kreisler.
broken but with the auxiliary changing to c natural 236ii. Vl: All the editors either mark or seem to envis-
at the fermata. While Beethoven provided a florid age an open D-string.
ornament at the fermata, he would not necessarily 245–252. lh: This is presumably legato, as marked by
have expected trained pianists to play the same one Speidel, Diémer, Vogrich. Ganz, however, marks non
each time. legato.
157i–ii. Vl: Beethoven’s slur from 157i–ii can only 248iii–iv. Vl: Only Singer, Auer, who mark a shift to
have been conceived as a portamento connection, 3rd position, apparently envisaged playing Beetho-
achieved by going up the A-string to c3 in the pre- ven’s notation literally; the others all apparently re-
vious bar and sliding the finger back to 1st posi- main in 1st position, making it impossible to sustain
tion before playing f #1. Most of the editors indicate the a1, unless it is repeated by substituting it with a
this and those that do not (Alard, David, Brodsky, 4th finger above the c1.
Seybold), evidently considered it so obvious that a 253–260. rh: This is presumably legato as marked by
fingering was not needed. Speidel, Diémer.
160–161. Pno, Vl: Speidel/Singer mark espress. for the 261. Pno, Vl: Speidel/Singer mark con fuoco.
32nd-note fioriture. 264. Vl: In order to play the separate bows in 265–267
163. Vl: All mark the obvious fingering in the second with as broad a stroke as possible in the upper half,
half of the bar with 1–3 on xvi–xvii and 4 – 4 on xx–xxi. it may be preferred to change to  on v rather than i,
170–173. Pno, Vl: Speidel/Singer, unaware of the miss- as suggested above the stave in the edited violin
ing p (see Critical Report) add the instruction drän- part of the present edition.
gend (pressing forward), and Singer marks a porta- 265–267. Pno, Vl: AG, presumably on the basis that
mento fingering from i–ii. At 171 they add a tempo the absence of staccato marks suggested a continu-
and at 173 poco rit. Delicate portamento, as sug- ation of the slurring from 261–264, added slurs and
gested by the fingering in the edited violin part of these are reproduced by all the editors. Beethoven,
the present edition, and agogic accentuation, can however, clearly wanted separate bows in Vl and
enhance the effect of the violin’s subito p and give an equivalent articulation in Pno (see Critical Re-
characteristic expression to the <> in 172; a har- port). Beethoven surely envisaged that the pianist
monic is marked on ii by Alard, Hermann, Singer, would play these 16ths with great energy and as
Halir, Seybold, and all except Alard, Halir mark it non-legato as possible.
with the same finger as the preceding note. This 268–270. Vl: All editors except Hermann divide Beet-
was probably also expected by David. The pianist hoven’s impracticable slur into two or three sepa-
would almost certainly have employed arpeggia- rate bows, some on the strong beat (Alard, Joachim,
tion as well as agogic accent at the apex of the <> Kreisler) and some on the weak beat (David, Singer,
markings in 171–173. Auer, Brodsky, Halir, Rosé, Seybold).
174, 190. rh: Presumably legato. Reinecke marks sem- 274vi. Vl: All editors go to 1 here as in the London 1st
pre legato. Ganz marks poco legato. edition (see Critical Report).
203vi. Vl: All editors took 203i with 2nd finger and 275–287. Pno: Some or all of the chords could be
all except Rosé (who marked 3 on v) evidently as- swiftly arpeggiated.
sumed a harmonic on vi; even Kreisler explicitly 277i–ii. Vl: All except Hermann, Brodsky give 2–3,
marked the harmonic. with an inevitable portamento.

278. rh: Speidel marks cantabile. 285i–ii. Vl: Singer, Auer specifically indicate a 3rd
281. Pno: Speidel marks sempre dim. finger slide to the harmonic. Probably these effects
283. Vl: Singer marks sempre dim. were envisaged as possibilities by those who mark
285. Pno: Speidel marks slentando. nothing.
283ii, 285ii–287ii. Vl: Alard, Singer, Auer, Halir (only
285–287), Kreisler (only 285f) mark harmonics.

A Selection of Editions for Violin
B Ä R E N R E I T E R U R T E X T
Wolfgang Amadeus Mozart

Complete Works for Violin and
Piano in Two Volumes
Volume 1: BA 5761 / Volume 2: BA 5762
Concertos for Violin and Orchestra

- in B-flat maj K. 207 (No. 1).
BA 4863-90
- in D maj K. 211 (No. 2). BA 4864-90
- in G maj K. 216 (No. 3). BA 4866-90
- in D maj K. 218 (No. 4). BA 4865-90
- in A maj K. 219 (No. 5). BA 4712-90
Sinfonia concertante in E-flat maj
for Violin, Viola and Orchestra K. 364
BA 4900-90
Niccolò Paganini
from: Franck, Sonate op. 6 · BA 9425 Capricci op. 1 and 24 Contradanze
Inglesi (First edition) for solo violin
BA 9424
Johann Sebastian Bach Concerto in D maj for Violin and
Orchestra op. 77. BA 9049-90
Three Sonatas and three Partitas Maurice Ravel
BWV 1001-1006 for solo violin. BA 5116 Arcangelo Corelli Tzigane. Rhapsody for violin and
Six Sonatas BWV 1014-1019 Sonatas for Violin and Bc op. 5 orchestra. BA 8849-90
for violin and obbligato harpsichord Volume 1: BA 9455 / Volume 2: BA 9456
- Edition in two volumes Camille Saint-Saëns
Volume 1: BA 5118 / Volume 2: BA 5119 Antonín Dvořák Havanaise for violin and piano op. 83
- Edition in one volume. BA 5240 Concerto in A min for Violin and BA 9426
Orchestra op. 53. BA 10422-90
Concerto in A min BWV 1041
for violin, strings and bc. BA 5189-90 Franz Schubert
César Franck
Concerto in E maj BWV 1042 Complete Works for Violin and Piano
Sonate / Andantino quietoso op. 6 /
for violin, strings and bc. BA 5190-90 - Fantasia in C maj D 934 op. post. 159
Mélancolie for piano and violin
BA 5620
Concerto in D min BWV 1043 for two BA 9425
violins, strings and bc. BA 5188-90 - Rondo in B min D 895 op. 70
Georg Friedrich Händel BA 5618
Ludwig van Beethoven Complete Works for Violin and Bc - Sonata in A maj D 574 op. 162
HWV 358, 359a, 361, 364, 371, 368, 370, BA 5605
Concerto in D maj for Violin and
Orchestra op. 61. BA 9019-90 372, 373, 408, 412. BA 4226 - Three Sonatas “Sonatinas” D 384,
385, 408 op. post. 173. BA 5606
Romances in F maj and G maj Felix Mendelssohn Bartholdy
for Violin and Orchestra op. 50, op. 40 Rondo in A maj for Violin and
BA 9026-90 Concerto in E min for Violin and Orchestra D 438. BA 5653, BA 5653-90
Orchestra op. 64
- Piano reduction of version 2. BA 9099-90
Johannes Brahms - Piano reduction of version 1. Georg Philipp Telemann
Works for Violin and Piano BA 9099-92 Twelve Fantasias for Violin without
- Sonata in G maj op. 78. BA 9431 Bass. BA 2972
- Sonata in A maj op. 100. BA 9432 Sonatas for Violin and Piano
- Sonata in D min op. 108. BA 9433 BA 9066
- Sonatas in F min and E-flat maj
Antonio Vivaldi
after op. 120. BA 10911 The Four Seasons for violino
- Sonata Movement in C min from the principale, two violins, viola and bc
F.A.E. Sonata WoO 2. BA 10908 BA 6994-90
A 31R · 2007
Bärenreiter www.baerenreiter.com

Beethoven Violin Sonatas Performing Practice

Uploaded by

Copyright:

Available Formats

Beethoven Violin Sonatas Performing Practice

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Beethoven Violin Sonatas Performing Practice

Uploaded by

Copyright:

Available Formats

BEETHOVEN

Performing Practice Commentary

Bärenreiter Kassel · Basel · London · New York · Praha

Sonata op. 12, no. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Sonata op. 12, no. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Sonata op. 30, no. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Sonata op. 30, no. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Sonata op. 30, no. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Sonata op. 96 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

© 2020 by Bärenreiter-Verlag, Kassel Piano Techniques ▪5

6 ▪ Piano Techniques © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Piano Techniques ▪7

8 ▪ Piano Techniques © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Piano Techniques ▪9

10 ▪ Editions of Beethoven Sonatas © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Editions of Beethoven Sonatas ▪ 11

4. Edmund Singer / Wilhelm Speidel

12 ▪ Editions of Beethoven Sonatas © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Editions of Beethoven Sonatas ▪ 13

14 ▪ Editions of Beethoven Sonatas © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Metronome Marks and Tempo ▪ 15

47 Anton Schindler, trans. and ed. Ignaz Moscheles: Life of Beetho-

16 ▪ Metronome Marks and Tempo © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Abbreviations ▪ 17

SONATAS OPUS 12 Moscheles-Cramer  = 92

18 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 19

20 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 21

22 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 23

24 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 25

26 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 27

28 ▪ Opus 12, No. 1 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 1 ▪ 29

30 ▪ Opus 12, No. 2 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 2 ▪ 31

32 ▪ Opus 12, No. 2 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 2 ▪ 33

34 ▪ Opus 12, No. 2 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 2 ▪ 35

36 ▪ Opus 12, No. 2 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 3 ▪ 37

38 ▪ Opus 12, No. 3 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 3 ▪ 39

40 ▪ Opus 12, No. 3 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 3 ▪ 41

42 ▪ Opus 12, No. 3 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 3 ▪ 43

44 ▪ Opus 12, No. 3 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 12, No. 3 ▪ 45

46 ▪ Opus 12, No. 3 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 23 ▪ 47

48 ▪ Opus 23 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 23 ▪ 49

50 ▪ Opus 23 © 2020 by Bärenreiter-Verlag, Kassel

© 2020 by Bärenreiter-Verlag, Kassel Opus 23 ▪ 51

52 ▪ Opus 23 © 2020 by Bärenreiter-Verlag, Kassel

SONATAS OPUS 12 Moscheles-Cramer  = 92

advertent omission in the original; it would certain