See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/277715943

Fabian, D. (2005). Towards a Performance History of Bach’s Sonatas and Partitas

for Solo Violin: Preliminary Investigations. In Vikárius, L. & Lampert, V.(eds.)
Essays in Honor of...

Chapter · January 2005

CITATIONS READS
12 6,148

1 author:

Dorottya Fabian
UNSW Sydney
41 PUBLICATIONS   358 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Dorottya Fabian on 05 June 2015.

The user has requested enhancement of the downloaded file.

 Published in Essays in Honor of László Somfai: Studies in the Sources and the Interpretation of Music
Lanham, Maryland: Scarecrow Press, 2005: 87-108

Towards a Performance History of Bach’s Sonatas and Partitas for Solo Violin:
Preliminary Investigations

Dorottya Fabian (UNSW)

Nowadays Bach’s Six Sonatas and Partitas for solo violin (BWV 1001-1006) are regarded as
touchstones of any violinist’s technical and musical maturity. This has not always been the case.
In this paper I sketch the performance and reception history of these compositions reporting on
initial observations of written documents, various editions and sound recordings. The material at
hand is large (see number of editions and sound recordings available in Appendices 1-2).
Therefore the present discussion is limited to mapping out major trends and highlighting
significant issues. 1

Origin, Reception, Editions

Apart from surviving manuscript sources, among which Bach’s autograph dated 1720 is the most
precious, our first record of the pieces comes from the middle of the eighteenth century. 2 C.P.E.
Bach mentions them in his obituary of his father, published in the last volume of Lorenz
Christoph Mizler’s Musikalische Bibliothek in 1754. He stresses that the works demonstrate how
well Johann Sebastian understood the possibilities of string instruments and states that a “great
violinist” once told him “that he had seen nothing more perfect for learning to be a good violinist,
and could suggest nothing better to anyone eager to learn, than the said violin solos without
bass”. 3 Whether because of CPE Bach’s authority or because the pieces are so different from
Italian baroque violin music, the notion of regarding Bach's Solos as indispensable for perfecting
the violinist’s polyphonic technique — rather than as supreme compositions in their own right —

1
This project has been supported by a University of New South Wales Goldstar Award for research development. I
am also grateful to the Faculty of Arts and Social Sciences for a 6 months sabbatical that enabled me to collect data
in European archives and to Eric Sowey for his comments on an earlier draft.
Selected short audio examples of discussed recordings can be accessed from:
http://empa.arts.unsw.edu.au/research/em/bachsoloviolin.html
2
For a list of sources consult Kritischer Bericht (ed. Günter Hausswald), Neue Bach-Ausgabe VI/i (Kassel:
Bärenreiter, 1958).
 2

remained strong for a long time. This is evidenced in the subtitle of Ferdinand David’s 1842-3
edition which reads: “For use at the Leipzig Conservatoire …”. 4

Since about the middle of the twentieth century, however, the opinion has been strengthened
which holds that the Solo designation in the title indicates concert pieces.5 Bach researchers have
also been pointing to contemporary virtuoso violinists active in Dresden and Weimar, such as
Johann Paul von Westhoff (1656-1705), Johann Georg Pisendel (1687-1755), and Jean Baptiste
Volumier (ca. 1670-1728), and to similarly polyphonic compositions for violin by Heinrich Ignaz
Franz von Biber (1644-1704) and the Dresden court violinist, Johann Jacob Walther (ca. 1650-
1717) or even Pisendel himself. From Bach's immediate circle, Franz Benda (1709-1786) who
worked in the Prussian court of Frederick the Great, should also be mentioned. Benda was fond
of playing the pieces and introduced them to his pupils. Among the best of these were Friedrich
Wilhem Rust (1739-1796) and Johann Peter Salomon (1745-1815) who “played a major part in
keeping the Bach tradition alive”. 6 Contemporary records inform us, for instance, that Salomon
performed the pieces on several occasions. He also had a key role in disseminating the works
outside Germany, especially in Paris and London at the turn of the 18th century. It is likely that
his performances influenced the inclusion of the Fugue movement from the C Major Sonata into
Jean Baptiste Cartier’s L’Art du violon published in 1798. Apparently Cartier obtained a copy of
the score from Pierre Gaviniés (1728-1800), who may have acquired a copy from Salomon while
the latter was in Paris on his way to London from Berlin in 1780-81. 7 This publication, together
with Simrock’s 1801-2 edition of the three sonatas represent the first printed versions of the
pieces. The first complete publication was prepared by Ferdinand David (Leipzig: Kistner,
1843); the Bach-Ausgabe of the Bach-Gesellschaft brought out the works in 1879 (ed. Alfred
Dörffel).

3
Hans T. David & Arthur Mendel (eds.): The New Bach Reader revised and enlarged by Christoph Wolff (New
York: Norton, 1998), p. 397.
4
Cited in Walter Kolneder: Das Buch der Violine (Zürich: Atlantis, 1993). Eng. trans. by Reinhard G. Pauly as: The
Amadeus Book of the Violin (Portland, Oregon: Amadeus Press, 1998), p. 311.
5
Eduard Melkus: ‘Gedanken zur Interpretationsgeschichte der Chaconne für Violine solo von J. S. Bach’
Österreichische Musikzeitschrift 40 (1985) 100-108, esp. p. 103.
6
Kolneder: The Amadeus Book of the Violin p. 373
7
Hubert Unverricht: ‘Spieltraditionen von Joh. Seb. Bachs unbegleiteten Sonaten und Partiten für Violine allein’
Bachfest (55) der Neuen Bachgesellschaft (Mainz, 1980) pp. 176-184.
 3

While in the eighteenth-century the general opinion was that it would be impossible to add an
accompanying part to the works, during the nineteenth-century this changed. Then it was
believed that the public needed an aid to facilitate their understanding of the music and saw the
solution in providing accompaniment to the Solos. The result of this was the publication of
innumerable transcriptions as well as newly composed piano accompaniment from the 1840s
until the turn of the century. 8 The most famous of these are the accompaniments of Mendelssohn
(1840) and Schumann (1854) and the transcriptions for piano alone, especially of the Chaconne
from the D minor Partita, by Brahms and Busoni. The Chaconne became so popular that there
even exists an arrangement for orchestra and violin by August Wilhelmj (1845-1908), the first
Bayreuth concertmaster and a pupil of Ferdinand David. 9 Brahms captured its awe-inspiring
power in a letter to Clara Schumann, written in June 1877:

The Chaconne is for me one of the most wonderful, incomprehensible pieces of music. On
a single staff, on a small instrument the man writes a whole world of the deepest thoughts
and the most powerful feelings. If I were to imagine how I might have made, conceived the
piece, I know for certain that the overwhelming excitement and awe would have driven me
mad. 10

This kind of reverence has since been shared by many generations, elevating the status of the
Solos from being studies in mastering polyphonic violin technique to being regarded as the
artistic pinnacles of the instrument’s repertoire.

The stream of modern editions started with that of Joachim and Moser in 1908. Here, just as in
David’s 1843 edition, Bach's autograph score is reprinted in modern notation under the edited
staff, as an alternative. Violinists writing in the first half of the twentieth century or later all

8
For an account of 19th century publications and transcriptions of the Solos see Zay David Sevier: ‘Bach's Solo
Violin Sonatas and Partitas: the First Century and a Half’ Bach (Journal of the Riemenschneider Bach Institute) 12
/2-3 (1981) pp. 11-19; 21-29. Many individual movements that were recorded during the first half of the twentieth
century were also done with piano accompaniment (e.g. Kreisler).
9
For an account of the Chaconne transcriptions see Georg Feder: ‘Geschichte der Bearbeitungen von Bachs
Chaconne’ in Martin Geck (ed.): Bach-Interpretationen – Festschrift Walter Blankenburg zum 65. Geburtstag
(Göttingen: Vandenhoeck & Ruprecht, 1969), pp. 168-189
10
Styra Avins (ed.): Johannes Brahms: Life and Letters (New York: OUP, 1997), p. 515
 4

stress the importance of studying the original markings as well as the editorial suggestions. 11
This insistence allowed for such an editorial practice to become expected standard (e.g. Marteau,
Busch, Flesch, Havemann, Rostal) whilst Bach’s fair copy itself has also been reprinted in
various facsimile editions (refer to Appendix 1 for detail). 12

Close study of the various editions enables one to gain an insight into changing interpretative
approaches to the Solos. Apart from editorial slurs and other phrasing or articulation marks,
fingering and bowing indications provide further information. Copies of scores used by
particular artists, for instance Joseph Szigeti, are especially telling because they reflect a single
artist’s view of the piece and reactions to editorial suggestions. 13 Out of the many early 20th
century editions, Adolph Busch’s (1919) is practically devoid of markings. This might reflect the
severe and literalistic approach, which became typical in the 1930s and beyond. This ‘objective
style’ that is often linked to Stravinsky’s neo-classical ideology and referred to as Neue
Sachlichkeit can be observed on many recordings as well, even beyond the 1960s. Carl Flesch’s
(1930) edition reflects more the late romantic tradition, in spite of providing a reprint of the
autograph in modern notation. Phrasing marks are generally useful but fingering promotes the
use of high positions. Rhythm is often altered in an attempt to indicate performed values in
polyphonic section. Jean Campeil’s (1959) edition is the first to include the facsimile in an
appendix. Although it seems cluttered with editorial additions, these often contribute to stylish
solutions, for instance in regard to rhythmic flexibility and inequality. In his survey of editions
Stowell deems Campeil’s version “pioneering in its reverence to 18th century conventions … and
texts”. More recent editorial approaches and their implications for performance are also noted by

11
Leopold Auer: Violin Playing as I Teach it (New York: Dover, 1980; first published by Frederick A. Stokes Co.,
1921); Carl Flesch: The Art of Violin Playing (New York: Carl Fischer, Inc., 1930 English text by F. Martens);
Joseph Szigeti: Szigeti on the Violin (New York: Dover, 1979; first published by Cassell & Co., 1969).
12
A good summary of the strengths and weaknesses of available editions is provided by Robin Stowell: ‘Building a
Library: Bach’s Violin Sonatas and Partitas’ Musical Times 128 (1987), pp. 250-56. See also Joel Lester: Bach’s
Works for Solo Violin (New York – Oxford: OUP, 1999) and Richard Efrati: Treatise on the Execution and
Interpretation of the Sonatas and Partitas for Solo Violin and the Suites for Solo Cello by Johann Sebastian Bach
(Zurich: Atlantis, 1979).
13
The research library of the Liszt Academy of Music in Budapest has a rich collection of these; a telling set
documenting aspects of the Hungarian violin school fostered by Hubay around the beginning of the twentieth-
century. In his cited book Szigeti mentions the importance of studying old editions and earlier violinists’ markings
in them because “[w]e are so poor in documentary evidence about performance traditions … that scraps of
information … are of great value”. Devoting attention to such minutiae reveals “how little one can depend on Urtext
without oral tradition supplementing it.” (Szigeti on the Violin pp. 202-203).
 5

Stowell. He reports, for example, that the fingering in Szeryng’s edition (1981) reflects a “stylish
preference for the lower positions” whereas Rostal’s (1982) promotes an “indiscriminate and
unstylish use of harmonics, in spite of being “based largely on 18th-century principles”. The
edition of Babitz is controversial: Although it is “intended to capture 18th-century ideals on
articulation …, aims for the speaking performance … and advocates lower positions”, it is
“invariably mislead[ing], heavy with annotations [and] numerous misprints in the preface”. 14
Field’s dissertation gives an overview of certain other trends regarding interpreting Bach’s Solos
that can be deduced from the various editions. 15 However, no written score can be as telling
about performance style as a sound recording. Before these are discussed it might be useful to
briefly recall the performance history of the pieces prior to the advent of recording technology
and independent of editions.

Early performance history

As mentioned earlier, during the 18th century the works were better known among German
violinists than among other European musicians. Their historic appearance on the international
concert platform in the 19th century is traceable to Mendelssohn’s role in reviving Bach's
compositions in general. Sevier informs us (see note 7) that Mendelssohn asked David to play
the Chaconne during the 1839/40 season of the Gewandhaus Orchestra in Leipzig. However,
David only obliged when Mendelssohn agreed to accompany him on the piano. The performance
took place on February 8, 1840 and David offered the Preludio from the E Major Partita as an
encore, also with accompaniment.

All sources seem to agree that it was Joachim who first performed the pieces unaccompanied and
not only in Germany but in London as well. The reception was, however, not always very
positive, indicating the level of difficulty both for player and listener in the performance of these
works. Bernard Shaw reports in The Star on 28 February 1890:

14
Stowell: Building a Library pp. 255-56
15
Elizabeth Field: Performing Solo Bach: An Examination of the Evolution of Performance traditions of Bach’s
Unaccompanied Violin Sonatas from 1802 to the Present DMA diss., Cornell University, 1999.
 6

[Joachim] played Bach's Sonata in C at the Bach Choir Concert at St James’s Hall on Tuesday.
The second movement of that work is a fugue some three or four hundred bars long. Of course
you cannot really play a fugue in three continuous parts on the violin; but by dint of double
stopping and dodging from one part to another, you can evoke a hideous ghost of a fugue that
will pass current if guaranteed by Bach and Joachim. That was what happened on Tuesday.
Joachim scraped away frantically, making a sound after which an attempt to grate a nutmeg
effectively on a boot sole would have been as the strain of an Eolian (sic) harp. The notes
which were musical enough to have any discernible pitch at all were mostly out of tune. It was
horrible – damnable! Had he been an unknown player, introducing an unknown composer, he
would not have escaped with his life. 16

Joachim also made the first sound recording of the works. In 1903, just four years before his
death at the age of 72, he recorded the Adagio from the G minor Sonata and the Bourrée from the
B minor Partita. Although the recording quality is obviously limiting, we can glean something
of his style, tone and intonation. Joachim provides a rhythmically defined Bourrée with strong
down-beats and some slight long-short unevenness in quaver pairs. Other patterns are grouped so
as to emphasize the  pulse and up-beat phrasing. The Adagio is moderately free: ornamental
figures are grouped; bowing is light and varied, resulting in a clearly articulated rather than
sustained or ‘on-the-string’ legato. Only the abruptly broken accented chords and multiple stops
arrest the flow of the improvisatory feel even though they are executed so as to fit the melodic
line.

Joachim’s violin tone sounds rather thin and straight. Spectrogram analysis confirms that his
vibrato is narrow and selectively used. For instance, in the Bourrée, vibrato is perceptible only
on the minims (e.g. g” in bar 62, and less clearly on a” in bar 17). 17 In the Adagio there is
hardly any audible vibrato on the high g” preceding the f#” trill in bar 4 (see Fig. 1a), although
its use may be observed on other eighth notes (e.g. high g” in bar 12). The different delivery
could be related to whether the note is a single pitch or part of a multiple stop; the latter tends to
be played straighter. Joachim also uses portamento sparingly and to good musical effect: about
four times in the Bourrée (e.g. bar 18 a” to g”, b.27 g” to b”, b.48 second e’” to f#”, b.50 second
d’” to e#”) to reinforce rhythm and melodic grouping and a few times in the Adagio (e.g. last

16
Bernard Shaw: Shaw’s Music – The complete musical criticism in three volumes (London: The Bodley Head,
1981), Vol. 1, pp. 933-4
17
Notes are named according to Helmholtz’s system, where middle c is c’.
 7

two thirty-second notes]in the third beats of bars 17-18) to heighten the melodic expression
before the end of a phrase.

Vibrato and Portamento

Both of these devices (i.e. vibrato and portamento) are important enough to warrant further
discussion. During the 17th and 18th centuries vibrato was generally regarded as a species of
ornament to be employed with discretion, mostly to decorate longer notes. However, just as the
singing voice has a natural vibrato, it came to be recognised that a rich, sonorous tone on string
instruments could be created by vibration. Thus, by 1751, Geminiani was advocating the use of
continuous vibrato for the sake of a singing tone. Much ink has been spilt over the appropriate
use and type of vibrato in various repertoires and geographic regions and whether continuous
vibrato was really only a 20th century invention and vogue. Some argue that the central German
tradition represented by Spohr and David in the 19th century was upheld by Joachim and his
pupils who used vibrato selectively, whilst the Franco-Belgian school was more influenced by
Southern European traditions and Ysaÿe and Sarasate both employed continuous vibrato. Others
highlight Leopold Auer’s role as a teacher and point to his pupils, particularly Heifetz and Elman
as key violinists of the early 20th century who promoted a tone quality based on vibrato. 18 There
are also those who subscribe to Carl Flesch’s claim that it was Fritz Kreisler who introduced a
unique sound by employing vibrato even in fast running passages. 19 Katz, on the other hand,
argues that “the copious, continuous vibrato be understood as a response to the exigencies of
sound recording” and that “Kreisler was simply held up as a model when other violinists began to
use a prominent vibrato” in order to “compensate for the limitations and liabilities of early
recording equipment”, to “obscure imperfect intonation” that are more exposed “on record than
in a live setting” and to “offer a greater sense of the performer’s presence on record”. 20

18
Auer himself advocates the sparing use of vibrato even if he allows for an entire phrase—not just a single note—to
be expressively enhanced or highlighted by it (Auer: Violin Playing pp. 22-25).
19
See e.g. Carl Flesch jnr.: And do you also play the violin? (Toccata Press, 1990); Kolneder: The Amadeus Book of
the Violin; Robin Stowell (ed.): The Cambridge Companion to the Violin (Cambridge: CUP, 1992).
20
Mark Katz: ‘Aesthetics out of exigency: violin vibrato and the phonograph’ in Hans-Joachim Braun (ed.) Music
and Technology in the Twentieth Century (Baltimore—London: The John Hopkins University Press, 2000) pp. 174-
185, this citation on p. 179.
 8

Listening to sound recordings seems to support Katz’s point, at least in relation to performing
baroque music. Ysaÿe did not record any as far as I could ascertain and only a very fast
rendering of the E Major Preludio demonstrates Sarasate’s Bach interpretation on which vibrato
practice seems similar to that of Joachim in the Bourrée. All other early Bach (and Handel,
Corelli or Veraccini etc.) recordings I have heard confirm the continuous presence of vibrato.
The difference lies in the speed and width of it. A few spectrogram images (Fig. 1a-d) can serve
to illustrate the vibrato of various artists. Huberman’s and Szigeti’s are, for instance, wide, slow
and regular; Milstein’s, Heifetz’s and Busch’s narrower but still slow and regular; Menuhin’s,
Grumiaux’s, Perlman’s, Ricci’s, Hahn’s and Szenthelyi’s faster than these whilst Suk’s
Zehetmair’s and Edinger’s less regular and not continuous. Violinists playing on period
instrument use vibrato selectively to ornament certain notes.

In Figure 1c. (Huberman) one can see an upward straight line. This is a portamento. Hauck
claims that the 19th-century development of music being regarded as a shaped expression of
feeling brought along the emphasis on portamento rather than vibrato as the primary means of
this heartfelt and individual expression. 21 Looking at violin instruction literature up to and
including that of Carl Flesch’s in 1939, this view can be supported by many statements. These
emphasize that portamento should be used only for heightened expression, when the performer
feels overwhelmed. It becomes objectionable, however, when “it is executed in a languishing
manner, and used continually”.22 Sound recordings testify that around the turn of the century
there was a much more liberal use of portamento than from the 1940s onwards. It is likely,
argues Hauck, that as continuous and intense vibrato became the norm, portamento was gradually
abandoned and regarded as an overtly emotional means unsuitable for expression in a more
technologically and rationally oriented age. 23 Among the recordings of Bach's Solos portamento

21
Werner Hauck: Das Vibrato auf der Violine (Cologne, 1971), Eng. Trans., Kitty Rokos as Vibrato on the violin
(London: 1975)
22
Auer: Violin Playing p. 24. He compares the sound of the ‘languishing manner’ to the mewing of a cat and states
that the portamento should only be used “when the melody is descending, save for certain very exceptional cases of
ascending melody”. Although other sources are less categorical in this regard it is noteworthy that whilst descending
portamento is the norm on early recordings, the next generation of violinists mostly perform them in an ascending
context (e.g. Huberman).
23
This trend can be noted by comparing the playing of two turn-of-the-century violinists. Huberman (1882-1947) is
about 25 years younger than Hubay (1858-1937) and his use of portamento is extremely moderate by his elder’s
 9

is employed more liberally by Huberman, Heifetz, Enesco24 and Telmányi up to the 1950s, and
also, if somewhat surprisingly, by Ricci and Perlman in the late 1980s.

Fig. 1a: Joachim (1903): G minor Adagio, b. 4: trill Fig. 1b: Busch (1929): D minor Sarabanda, b. 6
on f#” but no obvious vibrato on previous g” (both showing slow, regular, not too wide vibrato
eighth-note value)

standard. Hubay’s 1929 recording of the Air from Bach's Suite No. 3 in D Major BWV 1068 (Mat. CV 713 HMV
AN418, issued e.g. on ‘La Storia Discografica del Violino 1: La Scuola Ungherese Libro 1’ IDIS 276) has a
portamento in practically every bar, mostly in descending motion. Huberman uses portamento much more sparingly
and more commonly in rising motion. More significantly for Hauck’s argument, this excerpt seems to contradict the
view that performers exchanged portamento for vibrato as the primary expressive device, for Hubay uses both and
not just moderately. Portamenti are, however, more prominent than the vibrato which is not used, for instance, on
the long opening note, nor on the minim a” in bar 3.
24
The Rumanian composer-pianist and violinist Georges Enescu preferred spelling his name as Enesco whilst living
in Paris.
 10

Fig. 1c: Huberman (1942): D minor Sarabanda, b. 6: Fig. 1d: Huggett (1995): D minor Sarabanda, b. 6:
c#”—b” sixteenths with portamento in-between, notes 1-5. Vibrato is used to ornament the long a”
followed by dotted eighth a” showing wide vibrato around the middle of its duration

Sound Recordings

Returning now to the recording history of the Solos in general, the following trends can be noted:
Initially only individual movements were recorded, often interpreted as showy bravura pieces.
The first complete work was recorded by Busch (1929, D minor Partita) and the first complete
set by Menuhin (1934-1936) and Enesco (1940). During this time and up to about the 1980s the
general approach could be best described as “reverential” with an emphasis on the serious, on
sustaining the polyphonic lines, and making the works sound grand. This could be seen also
from the fact that the Sonatas were more regularly performed whilst often only movements were
recorded from the Partitas. 25 Although the D minor Partita might seem an exception, even in
this case there is a clear predominance of individual movements, notably the Ciaccona and the
Sarabanda. The recordings testify that violinists seemed to be concerned to show their technical
mastery, especially when playing double, triple and quadruple stops and other polyphonic
 11

sections. There was a tendency to play each note or chord according to the written notation, often
causing rather harsh or whipping bow strokes and a loss of clarity in part playing. The difficulties
involved even led to the concoction of a theory that Bach and his circle used a special curved
bow that enabled the simultaneous playing of more than two strings.26 Towards the mid 1970s,
in the wake of the early music movement, certain musicians started to realise the opportunities
afforded by the application of original 18th-century playing techniques and the interpretative
information encoded in Bach's original articulation and bowing markings preserved in the
autograph fair copy. The increasing awareness of historical practices and experiences with
playing on period instruments led to a new trend in interpretation that utilized the characteristic
short articulation of the baroque bow, placed emphasis on rhythmic grouping and pulse, and did
not strive for sustaining polyphonic lines. Whilst the first such recording by Sergio Luca from
1976-77 is not well known, 27 the escalation of available recordings that use a period violin and
bow has been considerable, especially since the mid-1990s (cf. Appendix 2). Furthermore, the
playing of artists like Christian Tetzlaff and Thomas Zehetmair who use modern violins (and
bows) is also audibly inspired by historical performance practice. 28 All together, there have been

25
Szigeti clearly must have favoured the G and A minor Sonatas because he marked the score of these but not the
others and he also recorded them in the 1930s and 1940s whilst the complete set only in 1955.
26
Albert Schweitzer was among the most prominent proponents of this idea and encouraged bow makers and
violinists to experiment with an appropriate construction and potential playing technique (Schweitzer, Albert:
‘Reconstructing the Bach Violin Bow’ Musical America 70 [1950], pp. 5-13 and Idem: ‘Der für Bachs Werke für
Violine solo erforderte Geigenbogen’ in Matthaei, Karl, ed. Bach-Gedenkschrift 1950 [Zurich, 1950], pp. 75-83; see
also Joachim, Henry: ‘Bach Solo Violin Sonatas and the Modern Violinist’ Musical Times March 1931, pp. 221-
222). During the 1940s-1950s period Emil Telmányi became a proponent of the ‘VeGa’ bow so named after its
maker, VEsterGArd. Apart from having a curved stick, the bow’s hair could also be relaxed so as to play all four
strings at once if necessary. However, as Telmányi explained, the strain on the player’s thumb that manipulated the
hair tension eventually proved too much, preventing the regular (or complete) performance of the pieces (see Liner
notes to his recording from 1955 and Telmányi, Emil: ‘Some Problems in Bach’s Unaccompanied Violin Music’
Musical Times January 1955, pp. 14-18.). The historical existence of this hypothetical Bach-bow has since been
unanimously rejected by the entire community of researchers and performers of the baroque violin repertoire (for a
major rebuttal see David Boyden: The History of Violin Playing from its Origins to 1761 and its Relationship to the
Violin and Violin Music (London: OUP, 1965). Nevertheless, the German violinist Rudolf Gähler has recently
(1996) recorded the complete set using such a bow and even published a book in defence of it: Der Rundbogen für
die Violine - ein Phantom? (Regensburg: ConBrio, 1997).
27
For instance, it is not mentioned in a recently published and otherwise most comprehensive book on the history of
Bach Interpretation: Martin Elste: Meilensteine der Bach-Interpretation 1750-2000 (Stuttgart/Weimar & Kassel:
Metzler-Bärenreiter, 2000)
28
How historical performance practice gained ground among solo violinists can be gleaned from Robin Stowell’s
book: The Early Violin and Viola – A Practical Guide (Cambridge: CUP, 2002). The book also provides a good
summary of the differences between 18th-century and ‘modern’ violin techniques. Apart from studying original
treatises, further information on historical practices can be gathered from Stowell’s other book: Violin Technique and
Performance Practice in the Late Eighteenth and early Nineteenth Centuries (Cambridge: CUP, 1985).
 12

more than 40 complete sets and many single works or movements recorded since Joachim’s
historic 1903 recordings. Except for Misha Elman, practically all the famous violinists made
recordings of the Solos; some more than once (see Appendix 2 for a detailed list).29

Performance Practice

Turning to specific issues of performing the works, the problem of how to render polyphonic
sections might be discussed first. Two matters to highlight are the playing of chords and the
sounding length of notes relative to their written value. Generally, four-note chords are broken as
two plus two or one plus three in modern violin playing. However, historical sources indicate
that arpeggiation was the convention during the baroque period. At the same time Melkus warns
that the Lyra da braccia—the original instrument for which most early German baroque
polyphonic music was composed—played full chords and that arpeggiation seems to be more a
pre-classical ideal when polyphony went out of fashion. 30 Unverricht further notes that
arpeggiation does not have to be from bottom up, but should rather take into account the melodic
pitch and return to it as the ‘final’ pitch of the chord in order for the linear aspects of the music to
remain clear for the listener. He refers to what he calls zurückschlagenden bow-stroke that was
recommended by Karol Józef Lipińsky (1790-1861) to Joachim in relation to playing quadruple
stops in these works. Lipińsky, a representative of the Viotti-Spohr violin school, apparently
witnessed this bowing in Salomon’s interpretations of Bach’s Solos early in the 19th century. 31
This zurückschlagenden Bogen probably means a manner of bowing where the player returns to

29
Carl Flesch did not record any of the works but wrote about them in his The Art of Violin Playing. His edition also
reflects his interpretative view of the Solos. Besides, he was probably more significant as a violin pedagogue than as
a soloist. Other successful artists of the first half of the 20th century (e.g. Alma Moodie) had fewer opportunities to
make recordings. Furthermore, even in concerts it was more likely for them to perform concertos or short bravura
pieces with piano accompaniment than to play Bach’s Solos or fill their program entirely with chamber sonatas (for
more on recital programs at the time see, for instance, Flesch: The Art of Violin Playing pp.115-124 or Szigeti:
Szigeti on the Violin pp. 9-18).
30
Eduard Melkus: ‘Gedanken zur Interpretationsgeschichte’ p. 101
31
Unverricht: ‘Spieltraditionen von Joh. Seb. Bachs unbegleiteten Sonaten’ p. 177. He borrowed this information
from Andreas Moser: ‘Zu Johann Sebastian Bachs Sonaten und Partiten für Violine allein’ Bach Jahrbuch 17 (1920)
p. 42
 13

the main melodic note at the end of the arpeggio. 32 It would be particularly useful when the
melody is in the middle or lower voice, as it often happens in the D minor Ciaccona but also in
the G minor Adagio, and elsewhere. Without it the melodic line is easily blurred because the
chord’s top note is emphasized (held) instead of the melodically important pitch (in the middle or
the bass). Nevertheless, 20th-century violinists do not seem to use it, not even baroque
specialists. Among recordings from the 1930-70s, it is common to hear the opening bars of the G
minor Sonata as notated in Ex. 1.a (e.g. Menuhin, Heifetz, Szeryng ‘68, Suk), whereas Enesco,
Szeryng ’65, Milstein ’55, Wallfisch, Szenthelyi, Zehetmair, Tetzlaff, Huggett, Kuijken, and
Luca perform it more like in Ex. 1b, which is more in line with the implications (Ex. 1c) of
Bach’s own notation (Ex. 1d). 33 In the interpretations of Szigeti, Grumiaux, Milstein ‘75, and
Ehnes the melodic notes of g’ (bar 1), c” and b’ flat (bar 2) are somewhat covered by the
sustained higher pitches whilst in Podger’s the swell she plays on the chords affect the perception
of the melodic notes.

32
I’ve asked many baroque violinists about this term and this seems to be the most reasonable meaning, even if it is
speculative. I am grateful to Marc Strümper and his friends in Vienna for the clarification. [Since publication Ferenc
Rados has kindly drawn my attention to another source: Meyer-Sichting, G. (ed.), Georg Kuhlenkampf: Geigerische
Betrachtungen (Regensburg: Gustav Bosse Verlag, 1952). On p. 52 there is an account of Joachim’s use of
zurückschlagenden Bogen and Mendelssohn’s tirade against it: "Wie instinktsicher ein Künstler wie Mendelssohn
hinsichtlich der Darstellungsart der Chaconne war, zeigt am aufschlussreichsten seine Begegnung mit dem jungen
Joseph Joachim. Der damals 16jaehrige war der erste Geiger, der nach Pisendel den Mut fand, die Bachschen
Solosonaten in ihrer Originalfassung öffentlich vorzutragen. Joachim spielte Mendelssohn die Chaconne vor - und
zwar spielte er die oben erwaehnte Stelle (=the first 2 motives of the first variation) mit zurückgeschlagenem Bogen.
Mendelssohn aber rief ihm zu: "Wie können Sie sich auf bloss auf solch verkünstelte Manöver einlassen? Spielen Sie
die Bachschen Sachen getrost in ihrer bisheriger gesunden Weise und lassen Sie sich bei allem sagen: intelligente,
wirklich musikalische Menschen hören nicht nur mit dem ausseren Ohr, sondern auch mit dem inneren und wissen
infolgedessen stets, wo ein Motiv herkommt und wohin es geht; für die unmusikalischen ist ohnehin kein Kraut
gewachsen - die mögen die betreffenden Stellen nach irgend eines Anderen Ratschlag ausführen." Joachim, einer
der grössten, geistig gestaltenden Geiger, ist diesem Ratschlag Mendelssohns sein Leben lang treu geblieben."]
33
Example 1b is taken from Max Rostal: ‘Zur Interpretation der Violinsonaten J. S. Bachs’ Bach Jahrbuch 59
(1973), pp. 72-78. The citation is from pp. 72-3
 14

Ex. 1a: Common performance solution to b. 1-2, G minor Adagio. c” (bar 1) and f#” (bar 2) are held over the
melodically more important g’ and c” respectively.

Ex. 1b: Alternative performance option reflecting more closely the basic melody (3rd beat of b.1 and down-beat of
b.2).

Ex. 1c: Basic melody (à la Corelli)

Ex. 1d: Bach's original notation

Apart from straight chords, these works are also rich in long polyphonic passages. Many
commentators mention that the notation provides a visual overview of the compositions’ structure
that was not meant to be executed literally. In other words, Bach notates the ideal length of
notes, but in practice these may not sound for their full value. Editors sometimes indicate the
sounding value making the score look cluttered. Proper interpretation depends on analysis of the
 15

context and developing suitable bowing techniques, which may be assisted by an awareness of
the characteristics of various baroque bows.

Bowing

The baroque bow is shorter than the one developed by the Tourte family and currently in use. Its
weight distribution naturally distinguishes between up-stroke and down-stroke, highlighting the
metrical divisions of bars and other units. Its shorter length makes legato playing of extended
sections impractical; its peculiar characteristic of sounding louder in the middle makes dynamic
nuances easy to produce. The basic difference between an early and a modern bow-stroke can be
clearly seen in the spectral view of sound files. The contrast between the typically even and
sustained stroke of Menuhin (Fig. 2a) or Szigeti (Fig. 2b), for instance, and Luca’s (Fig 2c) and
Wallfisch’s (Fig. 2d) “period” stroke with its soft beginning and rapid decay is obvious but can
be imitated as demonstrated by Szenthelyi (Fig. 2e) who uses a modern bow and who consulted
with László Somfai on matters of historical performance. 34

34
Liner Notes to Szenthelyi’s recording of the complete set (Hungarton HCD 32071-72), p. 5
 Published in Essays in Honor of László Somfai: Studies in the Sources and the Interpretation of Music
Lanham, Maryland: Scarecrow Press, 2005: 87-108

Fig. 2a: Menuhin 1934: D minor Sarabanda, b. 1-2 Fig. 2d: Wallfisch 1997: D minor Sarabanda, b. 1-2,
showing the sustained bowing style illustrating the characteristic arch-shape of period
bow-strokes.

Fig. 2b: Szigeti 1955: D minor Sarabanda, b. 1-2 Fig. 2e: Szenthelyi 2001: D minor Sarabanda, b. 1-2;
showing the sustained bowing style with some decay bowing imitates baroque practice with more graded
at the end of the long notes resulting in clearer beginnings and ends of notes. Note also the limited
articulation and rhythm use of vibrato (e.g. end of first chord).

Fig. 2c: Luca 1977: D minor Sarabanda, b. 1-2

showing typical strokes of a baroque bow with
gradual beginning and ending of notes.
 Published in Essays in Honor of László Somfai: Studies in the Sources and the Interpretation of Music
Lanham, Maryland: Scarecrow Press, 2005: 87-108

What is perhaps less well known is that already during the baroque there existed a difference
between the short French bowing style and longer Italian style that was more suitable for melodic
or cantabile interpretation. It is hard to know for sure towards which style Bach was inclined.
Although the French orchestral style was much admired at the time, Italian virtuosos were also in
high demand at German courts where Bach was employed while composing the pieces. It is
perhaps noteworthy, in this context, that Bach labelled the movements of Partitas I and II in
Italian rather than in French and wrote embellishments of the Italian kind for the Adagio and
Grave of Sonatas I and II. Considering the differences between bow types should also be related
to differences among the national aesthetic ideals. Although French, Italian and German baroque
sources all drew analogies between music and speech, there are fundamental differences between
what they actually meant – mostly because the specific focus of attention differed from country to
country. For the Italians a singing melody was paramount. The French refined the art of
distinguishing between long and short syllables. The Germans favoured a more general approach
that emphasized rhetorical expression based on minute differences between stressed and
unstressed notes, as associated with poetic metre and achieved through the subtle differences
between up- and down-strokes. 35

In contrast, modern violin technique strives for precise regularity and the greatest possible
evenness between up and down bows, so as to project a continuous melodic line – an ideal that
came to the fore in the 19th century. The near permanent use of the portato-stroke, when hardly
any difference remains between legato and détaché, serves this ideal. 36 Interestingly, however,
this bow-stroke seems to have been developed more in the middle and second half of the 20th
century than directly in relation to 19th century compositions. Although sustained legato playing
is common on Bach recordings from the 1920s-1930s, these present a greater variety of bow
strokes, especially in the dance movements, than those from the 1950s to 1980s period. Heifetz
(1935) and Szigeti (1934), for instance, produce seamless legatos for particular phrases, yet they
also play closely articulated groups, utilize light staccato, marcato, thrown bow-strokes;

35
Melkus: Gedanken zur Interpretationsgeschichte pp. 103-104
36
Ibid.
 18

differentiate between using just the tip of the bow, playing at the nut, or making long strokes.
Apart from Grumiaux, and to a lesser extent Szeryng, violinists of the next generation and
beyond tend to have a more uniform approach to bowing. Szigeti noticed this decline in right-
hand technique in favour of a focus on the left-hand when he remarked: “At a time like ours
when the average violinist’s equipment is at an unprecedented high level … it may sound
paradoxical when I say that there seems to be—at least to me!—a regression in the use of the
bow for effects that are far beyond the ordinary.” 37 The ever-present portato-stroke limits the
range of musical character that a performer can project making the Solos sound more
homogeneous and reinforcing their timeless rather than style-specific qualities. Nevertheless
when executed with perfection, as on Hilary Hahn’s recording (1997), the beauty and
seamlessness of the violin tone is quite wondrous.

If the striving for evenness and continuous legato created a decline in ‘articulated’ bowing, too
much emphasis on differentiating between up and down strokes carries potential problems as
well. The exaggerated articulation of the smallest units, too many stresses and too much dynamic
nuancing can quickly lead to mannerism; tone quality suffers and the flow of the music is
disturbed as can be observed in certain movements on the recordings of van Deal, Wallfisch and
Huggett, among others. The messa di voce is an expressive device; it is better to use it only
selectively. At the same time, Bach’s polyphony is only perceptible when the hierarchy between
the voices is clearly articulated through different shades of tone and by carefully weighing up the
relative length or importance of notes. The perception of structure and harmonic implications in
movements like the Preludio of the E Major Partita or the Allegro Assai of the C Major Sonata
can also be much enhanced when the hidden or implied polyphony is brought to the fore through
rhythmical stress, rubato and a feeling of improvisation 38 (compare, for instance Mintz with
Zehetmair in the Allegro assai or Sarasate or Ehnes with Podger in the Preludio). The baroque
violin and bow allows for the exploitation of low positions and the use of the sonorous open gut
strings. Players can easily skip strings because they use the middle of the bow much more than

37
Szigeti: Szigeti on the Violin, p. 197
38
‘Jaap Schroeder Discusses Bach’s Works for unaccompanied Violin’ – edited transcript of the discussion held at
Queens College, City University of New York, January 9, 1977. Journal of the Violin Society of America 3 (1977),
pp. 7-32. Lester (Bach’s Works for Solo Violin) also argues for a more structural—harmonic approach to
performance that takes into consideration the implications of metre.
 19

the upper half, which is what is typical of modern bowing. These period techniques as well as
the judicious use of rhythmic flexibility and metric stresses contribute substantially to the vitality
of rhythm, clarity of harmonic-polyphonic detail and variety of musical character witnessed on
recordings of Luca, Zehetmair, Tetzlaff, Huggett, van Deal, Wallfisch, Podger and Szenthelyi.
Portato-bowing, terraced dynamics and long-range crescendos and diminuendos create an
essentially different effect: more even tone, projection of large-scale structure, authoritative
command of intonation, and a generally homogenous delivery of the various movement types
(e.g. Milstein, Kremer, Perlman, Ehnes, Poulet, Ricci, Hahn).

Dance Movements

The above already implies that bowing plays a crucial part in projecting the character of the
dance movements. Here it needs to be informed by an awareness of the historical meaning of
time signatures and their significance for the distribution of metric stresses within the bar. 39
Familiarity with characteristic dance steps and patterns also fosters the effective rendering of the
Partitas. 40

Take the Loure from the E Major Partita, for example. It is often interpreted as a somewhat sad
or lyrical “romance with exaggerated sentiment” when in reality the loure is a “dance of Spanish
origin with a certain amount of temperament and pronounced stresses on the strong beat” but not
on the third or last (sixth) beats. 41 The tempo, therefore, should be fairly fast and the articulation
should recall the hopping character of the dance. Apart from Zehetmair, only historically
informed violinists perform a loure. Even Tetzlaff plays it slow and lyrical, although with
perceptible pulse. Others, in particular Hahn, Suk, Milstein, Szigeti, Enesco and Menuhin play
the movement either too slowly, or too legato, with little sense of metre and focusing on the
melody. There is no emphasis of the characteristic rhythmic groups; in fact accents are often

39
Geroge Houle: Meter in Music (Bloomington: Indiana UP, 1987)
40
A useful source here is Meredith Little & Natalie Jenne: Dance and the Music of J.S. Bach. Expanded edition.
(Bloomington & Indianapolis: Indiana UP, 2001).
41
Schroeder discusses Bach, p.19
 20

misplaced due to a trill or double stop. Similarly, the performances of the first Minuet in this
partita generally sound like a heavy, over-accented, and melodically oriented rendering of triple
stops (e.g. Milstein ’50, Szeryng, Grumiaux, Menuhin ’57 etc.) rather than a dance in 3/4 (e.g.
Heifetz, Zehetmair, Tetzlaff, Wallfisch, etc.). In the latter case the pairs of quavers are sometimes
rendered with a long–short lilt or dotting (Tetzlaff, Podger, van Dael, Kuijken).

The Sarabanda of the D minor Partita also warrants a few words. The saraband is a slow, stately
dance in triple metre with an accent on the second beat. Bach’s scoring fits the pattern perfectly:
rhythm, harmony, texture, written out ornaments all support the saraband character. Recordings
from the first half of the century play it legato, in a sustained style, rhythmically even and literal.
Even lay listeners labelled these versions ‘romantic’. The saraband rhythmic pattern and
ornaments are decipherable in Grumiaux’s version but he plays very fast and matter-of-factly. In
contrast, Luca’s airy phrasing and grouped figures clearly project the triple metre with accent on
the second beat and provide an expressive performance without a hint of romanticism. 42

Tempo

It is no coincidence that, so far, it is mostly bowing that has been discussed, whilst the
importance of choosing the “right” tempo has hardly been mentioned. Without wanting to
deprecate the significance of tempo in musical performance, it seems important to stress that the
emphasis researchers have put on the issue of tempo is not always warranted. First of all, there
are the statements of Harnoncourt and Schroeder which claim that more clearly articulated music
often sounds faster than an under-articulated one. Then there is the growing evidence that,
contrary to general assumptions promoted by certain publications, 43 performance tempo has not
changed as radically as other aspects of the interpretative vocabulary. 44 This is borne out by

42
Dorottya Fabian & Emery Schubert: ‘Is there only one way of being expressive in musical performance? - Lessons
from listeners’ reactions to performances of J. S. Bach’s music’ in C. Stevens, D. Burnham, G. McPherson, E.
Schubert, J. Renwick (Eds.) Proceedings of the 7th International Conference on Music Perception and Cognition,
Sydney, 17-21 July 2002 (Adelaide: Causal Productions, 2002 CD-ROM)
43
e.g. Richard Taruskin: Text and Act (New York: Oxford, 1995)
44
Dorottya Fabian: Bach Performance Practice 1945-1975 (Aldershot: Ashgate, 2003) pp. 97-104.
 21

examination of the recordings under scrutiny here as well. An earlier study examined twenty
versions of the D minor Partita and found no significant trends for speeding up, not even among
historically informed performances. 45 The analysis of the other recordings supports this result to
a large extent (see Fig. 3a-b). Although the tempi are obviously varied, more recently issued
versions are not necessarily faster than earlier ones and most fit within a standard range of tempi.
Statistically speaking, in the Sonatas only five to ten performances deviate from this norm by
more than  1 STDEV. 46 Extreme deviations (i.e. STDEV greater than  2) are rare indeed: apart
from Heifetz (2.86) and Telmányi (-2.25) in the A minor Fuga, not more than one in any of the
movements. Only Zehetmair’s A minor Andante represents an even more extreme tempo choice
(STDEV 3.75): his MM quarter note = 48 is way faster than average (MM = 30) or even the next
fastest (Luca & Milstein 1975; both at MM = 35). Broadly speaking, tempo choice seems to
fluctuate more in the Partitas, especially in terms of degree. There are quite a few with more than
 2 STDEV and three with about  3 (Zehetmair in the E Major Gavotte en Rondeau [3.01] and
D minor Corrente [3.02]; Goldstein in D minor Sarabanda [-3.01]). What is also noteworthy is
that apart from the extremes, overall trends seem to be similar in earlier and more recent times
(e.g. D minor Corrente MM quarter note = 113 in 1929 and MM = 112 in 2001 with average MM
= 115). A slight trend towards faster tempi can be observed in the D minor Partita’s Sarabanda
movement (MM quarter note = 29 in the earliest and MM = 45 in the latest, average MM being
39), but the fast extremes of the Allemanda in earlier decades seem to have slowed since 1985
(see Fig. 3a). The Giga of the same work is also taken slower by current players (cf. Busch,
1929: MM dotted quarter note = 84; Szenthelyi, 2001: MM = 73). Finally, when one compares
performances on period instruments with other interpretations the picture remains fairly mixed: in
some cases the former are faster than the latter, in others it is the other way round, and quite often
there is no difference. In the Sonatas there is a trend for historically informed violinists to play
the fugues and final movements slower whilst the opening movements faster than other violinists.
Such generalisations are harder to make regarding the Partitas, except that the B minor

45
Richard Pulley: ‘A Statistical Analysis of Tempi in Bach’s D minor Partita’ in C. Stevens, D. Burnham, G.
McPherson, E. Schubert, J. Renwick (Eds.) Proceedings of the 7th International Conference on Music Perception
and Cognition, Sydney, 17-21 July 2002 (Adelaide: Causal Productions, 2002 CD-ROM)
46
By referring to the number of standard deviations from the mean (STDEV) information regarding the extremity of
tempo choice can be obtained. A STDV between  1 and  2 means that the performance falls within the tempo
choice of approximately 65 to 95 per cent of all performances. A negative number means a tempo that is slower than
standard.
 22

Allemanda, the D minor Corrente and Ciaccona and the E Major Menuet II are rendered faster by
period instrument violinists but the B minor Corrente and E major Preludio slower (see Fig. 3a-b
for tempo choices in D minor Partita).

Tempo can, of course, be approached from the point of view of flexibility. How steady it is in
certain interpretations or how much it fluctuates and when (or for what purpose) in others?
During the last ten years or so researchers of musical performance have experimented with
various techniques to account for tempo flexibilities. 47 In the present context the so-called tempo
mapping method could be used as the material at hand is pre-recorded violin music rather than
especially in-putted MIDI or disklavier piano renderings. A preliminary overview indicates that
fast and fugue movements have a steadier tempo than dance and slow movements. Tempo
fluctuates in a somewhat angular manner in those renditions where the performer articulates the
music in greater detail (e.g. D minor Allemanda by Tetzlaff, E Major Preludio by Huggett, A
minor Fuga by Wallfisch). By “angular” I mean tempo differences that are either pronounced
(i.e. a quasi sudden arrest or rush ahead) or closely linked to rhythmic grouping and therefore
locally nuanced. In “phrased” performances of slower type movements tempo fluctuation is
smooth and “long-range”; it conforms to the typical custom of slight slowing down for phrase
ends (e.g. E Major Loure by Hahn, Milstein 1950s; A minor Andante by Enesco, Ricci, D minor
Sarabanda by Huberman, Telmányi, Kagan; G minor Siciliano by Ehnes, Menuhin in 1957). It
seems that a detailed comparative tempo-mapping study of these sections would provide essential
data regarding the interpretative differences between “traditional” (or “mainstream”) and
historically informed approaches; or what I like to call “phrased” versus “articulated”
performances.

47
See, for instance, David Epstein: Shaping Time (New York: Schirmer Books, 1995); José Bowen: ‘Tempo,
Duration, and Flexibility: Techniques in the Analysis of Performance’ Journal of Musicological Research Vol. 16
(1996), pp. 111-156; John Rink (ed.): The Practice of Performance: Studies in Musical Interpretation (Cambridge:
CUP, 1995)
 MM values MM values

0
10
20
30
40
50
60
70
80
90
100
0
20
40
60
80
100
120
140
160
180
Busch Busch
Busch Busch
Busch Busch
Menuhin Menuhin
Heifetz Heifetz
Enesco Enesco
Francescatti Francescatti
Huberman Huberman
Heifetz Heifetz
Telmanyi Telmanyi
Milstein Milstein
Szigeti Szigeti
Menuhin Menuhin
Grumiaux Grumiaux
Szeryng Szeryng
Szeryng Szeryng
Suk Suk
Milstein Milstein
Goldstein Goldstein
Luca Luca
Kagan

violinists
Kagan

violinists
Kremer Kremer
Kuijken Kuijken
Mintz Mintz
Zehetmair Zehetmair
Perlman Perlman
Ricci Ricci
Edinger Edinger
Tetzlaff Tetzlaff
Huggett Huggett
Poulet Poulet
van Dael van Dael
Hahn Hahn
Podger Podger
Wallfisch
Fig. 3b: Tempo choices in D minor Giga and Ciaccona arranged in chronological order (1929 – 2002)

Wallfisch
Gahler Gahler
Paraschkevov Paraschkevov
MM Giga

Ehnes Ehnes
MM Corrente

Szenthelyi
MM Ciaccona
MM Allemanda

Szenthelyi
MM Sarabanda
Fig. 3a: Tempo choices in D minor Allemanda, Corrente and Sarabanda arranged in chronological order (1928-2002)
23
 24

Ornamentation

Finally, a few words about ornamentation are needed. The earlier discussion of the G minor
Adagio’s opening bars and of the D minor Sarabanda has already implied that the many small
note values and rhythmically intricate looking groups in Bach’s scores reflect the Italian manner
of delivering slow movements. Therefore a “gestural” rather than metronomic performance is
more appropriate. Keeping a constant beat while playing with a degree of liberty and freedom
projects a performance in which melodic embellishments sound improvised at the spur of the
moment; an historical characteristic of baroque performance practice and the prerogative of all
accomplished baroque performers. At the face of it, Bach takes away the performer’s right to
such extempore ornamentation by notating every single small note. But this practice of his does
not mean that everything he writes has identical compositional significance and needs to be
played with equal importance. On the contrary: modern performers must realise that Bach’s
scores represent written out copies and render decorative notes and patterns accordingly.
Distinguishing between ornamental groups and principal melodic pitches requires subtle control
of tone and rhythmic rubato or local freedom within a fairly strict beat. These means create a
different type of expression from that generated by emphasis on large-scale phrasing and long-
range dynamic- or tempo fluctuations (compare, for instance, Heifetz, Suk, Menuhin and others
with Wallfisch, Luca, van Deal et al). 48

But of course ornamentation in baroque music also means the addition of trills and other graces.
These, however, do not make a huge difference in overall performance style and most violinists
play those indicated by the score some adding a few extra (e.g. Podger in Minuet I of the E Major
Partita). Recently a few historically informed performers ventured to add more elaborate figures
and groups of melodic notes as well. Luca’s embellishments in the Andante of the A minor
Sonata make his rendering especially beautiful. Transcriptions below show Monica Huggett’s

48
This point is discussed at greater length in Fabian: Bach Performance Practice (esp. pp. 159-167). Many recent
sources on historical performance practice emphasize the role of articulation and metre in creating inflected or
“embellished” performances. See, for instance, Colin Lawson & Robin Stowell: The Historical Performance of
Music: An Introduction (Cambridge: CUP, 2000) or Bernard Sherman: Inside Early Music (New York – Oxford,
1997)
 25

(Ex. 2a) and Rachel Podger’s (Ex. 2b) virtuosic complete re-writes of the final rondo statement in
the E Major Partita’s Gavotte movement.

Ex. 2a: bars 94-100, E Major Partita: Gavotte en Rondeau as performed by Huggett (1995)

Ex. 2b: bars 94-100, E Major Partita: Gavotte en Rondeau as performed by Podger (1998-9)
 26

Conclusions

Summing up these necessarily cursory observations the most obvious thing to note is the richness
of the material at hand: over fifty editions and more than forty complete recordings, not to
mention the numerous transcriptions and arrangements and the extended scholarly and
pedagogical literature on the works. Studying them does not simply provide further support for
theories regarding general trends in music performance and violin playing but, importantly,
expands and refines our understanding of these trends.

Appendix 1: A Selective List of Important Twentieth-Century Editions

Year Editor Place Publisher

1908 J. Joachim / A. Moser Berlin Bote und Bock
1915 L. Capet Paris Sénart
1917 L. Auer New York Fischer
1919 A. Busch Bonn Simrock
1921 J. Hubay Wien Universal-Edition
1922 H. Marteau Leipzig Steingräber
1922 E. Herrmann New York Schirmer
1925 B. Eldering Mainz Schott
1930 C. Flesch Leipzig Peters
1934 J. Hambourg London Oxford UP
1934 E. Polo Milano Ricordi
1935 J. Garcin Paris Salabert
1940 G. Havemann Berlin Bote und Bock
1950 Facsimile (W.M. Luther) Kassel Bärenreiter
1958 Facsimile (G. Hausswald) Leipzig un Wiesbaden Insel-Verlag
1958 NBA; G. Hausswald Kassel Bärenreiter
1959 J. Campeil (with facsimile) Paris Heugel & Cie
1971 I. Galamian (with facsimile) New York International
1972 S. Babitz Los Angeles Early Music Laboratory
1981 H. Szeryng Mainz Schott
1982 M. Rostal Leipzig Peters
1987 K. Rönnau (W. Schneiderhan) München G. Henle Verlag (Urtext)
 Published in Essays in Honor of László Somfai: Studies in the Sources and the Interpretation of Music
Lanham, Maryland: Scarecrow Press, 2005: 87-108

Appendix 2: List of Studied Sound Recordings 49

Performer Rec. Date Issue Date Label ID Comment

Busch, Adolf 1928 1997 Appian Publications & Recordings Dm Sarabanda
5543
Busch, Adolf 1929 1991 / 1994 EMI Japanese edition TOCE 6781-97 Dm
Busch, Adolf 1934 1983 Danacord DACO 136 Gm
Edinger, Christiane 1991 1992 Naxos 8.55057 complete
Ehnes, James 1999 2000 Analekta Fleurs de lys FL 23147-8 complete
Enesco, Georges 1940 1999 Instituto Discografico Italiano IDIS complete
328/29
Francescatti, Zino 1958 1991 Biddulph ? Bm
Francescatti, Zino 1940? Columbia/SONY P14194 Dm, EM
Gähler, Rudolf 1998 1999 Aretenova Classics 74321 67501-2 complete
Goldstein, Mikhail c.1967- *Mitra CD 16239 (ADD) Dm
1980
Grumiaux, Arthus 1961 1993 Philips Duo 438736-2 complete
Hahn, Hilary 1997 SONY SK 62793 Dm, EM, CM
Heifetz, Jasha 1925 1992 EMI References (mono) CDH 7 64494 Em Minuet I-II
2
Heifetz, Jasha 1935 1992 EMI References (mono) CDH 7 64494 Gm, CM, Dm
2
Heifetz, Jasha 1952 1994 BMG Classics (RCA) 09026 61748-2 complete
Henkel, Susanna c.1995+ The Spot Records 28869-3 EM
Huberman, Bronislaw 1942 1997 Arbiter 105 Dm
Huggett, Monica 1995 1997 Virgin Veritas 5452052 complete; baroque violin
Joachim, Joseph 1903 Opal CD 9851 Bm Bourree, Gm Adagio
Kagan, Oleg 1979 Live Classics LCC 171 Bm, Dm
Kremer, Gidon 1980 circa *Philips 416 235-1 complete
Kreisler, Fritz 1910-1946 1995 BMG 0902-61649-2 EM Preludio & Gavotte; with
piano

49
All studied recordings were CD issues except when marked with * in the Label column. Those were long playing (vinyl) records accessed in sound
archives.
 28
View publication stats

Kuijken, Sigiswald 1981 1983 Deutsche Harmonia Mundi GD77043 complete; baroque violin
Kuijken, Sigiswald 2000 Deutsche Harmonia Mundi complete; baroque violin
05472775272
Luca, Sergio 1977 Nonsuch complete; baroque violin
Menuhin, Yehudi 1934-36 1989 EMI Reference (mono) CHS 763035 2 complete
Menuhin, Yehudi 1957 1993 EMI Classics 5 69249 2 complete
Milstein, Nathan 1955 1966 EMI ZDMB 6479323 complete
Milstein, Nathan 1973 1975 DG Originals DG 457701-2 complete
Mintz, Shlomo 1983-84 1984 Deutshce Grammophon DG 413810-2 complete
Paraschkevov, 1998 *Telos Studios TLS 007 Gm, Am, Dm
Vesselin
Perlman, Itzak 1986 1988 EMI Classics 7 49483 2 complete
Podger, Rachel 1997-99 1999 Channel Classics CCS 12198 complete
Poulet, Gerard 1996 Arion 268296 complete
Ricci, Ruggiero 1988 1992 “Ricci: Celebrating Six Decades on Concert; Am, Dm
1991 Stage” One-Eleven URS 92033 Concert; Bm
Sarasate, Pablo 1904 Opal CD 9851 EM Preludio
Schroeder, Jaap 1984-1985 1987 / 1990 Smithsonian Institution ND-0382 complete; baroque violin
Suk, Joseph 1971 1999 EMI ClassicsDouble fforte 5 73644 2 complete
Szenthelyi, Miklós 2001 2002 Hungaroton HCD 32071-72 complete
Szeryng, Henryk 1965 1991 SONY (AAD mono) 01-046721-10 complete
Szeryng, Henryk 1967 1968 (R1996) DG Double 453 004-2 complete
Szigeti, Joseph 1908 1991 Biddulph LAB 043 EM Preludio
Szigeti, Joseph 1941 1993 Biddulph LAB 070 Bm Bourrée
Szigeti, Joseph 1946, 1949 1993 Music and Arts CD4774 Concert; Gm, Am, EM
Szigeti, Joseph 1927-1933 1999 Biddulph LAB 153 Gm, Am, Bm Bourree, EM
Gavotte
Szigeti, Joseph 1955-56 1991 Vanguard Classics OVC 8021/22 complete
Telmanyi, Emil 1953-54 1984 Danacord DACO 147-149 complete
Tetzlaff, Christian 1993 1994 Virgin Veritas 545089-2 complete
van Dael, Lucy 1996 1999 Naxos 8.554423 complete; baroque violin
Wallfisch, Elizabeth 1996 1997 Hyperion dyad CDD 22009 complete; baroque violin
Zehetmair, Thomas 1983 1992 (R1998) Teldec 903176138-2 complete