Bartok, Beethoven and The Sonata For Two Pianos and Percussion

Journal of the Royal Musical Association, Vol. 137, no.
2, 307349
Bartók, Beethoven and the Sonata for

Two Pianos and Percussion
MICHAEL RUSS
ON 1 May 1892, the young Hungarian composer Béla Bartók gave his first public
concert. Among the works he performed were his own composition A Duna folyása
(The Course of the Danube), BB1/DD20a, and a Beethoven work, the first movement
of the ‘Waldstein’ Sonata in C, op. 53, written in 18034.1 The subsequent
importance of the music of Beethoven for Bartók has often been noted; several
commentators have referred to Bartók using Beethoven as a model or to quotations
from Beethoven in Bartók’s music.2 A specific connection between the ‘Waldstein’
and the first movement of Bartók’s Sonata for Two Pianos and Percussion, written in
1937, when Bartók was 56, is noted in a recent essay by Richard Cohn, and Paul
Wilson observes that the first movement of the Sonata for Two Pianos and
Percussion utilizes a ‘Beethovenian sonata form’.3 However, the relationship between
these two works has never been subjected to detailed scrutiny, and the influence of
Beethoven on Bartók’s work is just one of several that have been remarked upon,
ranging from Bach and Liszt in the shape of the opening idea, to Stravinsky and
Markevitch, both of whom produced works including multiple pianos and
percussion.4 Bartók’s interest in combining piano and percussion is evident in
earlier works the First and Second Piano Concertos (1926 and 19301) but in
E-mail: [email protected]
1
Malcolm Gillies, ‘Bartók, Béla’, The New Grove Dictionary of Music and Musicians, 29 vols. (London,
2001), ii, 787818 (p. 788). Otto Gombosi, in an unpublished biography, challenges this and states
that the work performed was Beethoven’s Sonata in C, op. 2 no. 3; see Benjamin Suchoff, Béla Bartók:
Life and Work (Lanham, MD, 2001), 268, n. 18.
2
See, for example, John A. Meyer, ‘Beethoven and Bartók: A Structural Parallel’, Music Review, 31
(1970), 31521, and Mark A. Radice, ‘Bartók’s Parodies of Beethoven’, ibid., 42 (1981), 25260.
3
Richard Cohn, ‘PitchTime Analogies and Transformations in Bartók’s Sonata for Two Pianos and
Percussion’, Music Theory and Mathematics: Chords, Collections, and Transformations, ed. Jack Moser
Douthett, Martha M. Hyde, Charles J. Smith and John L. Clough (Rochester, NY, 2008), 4971 (p.
69, n. 1); Paul Wilson, The Music of Béla Bartók (New Haven, CT, and London, 1992), 141.
4
Roy Howat notes that Bartók entitled an early draft ‘Sonate pour 2 Pianos et Percussion’, as if to align
the work with Debussy or the Stravinsky of Les noces. Roy Howat, ‘Masterworks II’, The Bartók
Companion, ed. Malcolm Gillies (London, 1993), 31530 (p. 329, n. 2). Another possible influence
was Igor Markevitch’s alternative or rehearsal scoring for two pianos and percussion of his ballet
L’envol d’Icare (1932). It is not clear, however, whether Bartók studied this work in the two-piano
version. Clive Bennett, ‘Icare’, Tempo, new series, 1334 (1980), 4451 (p. 44).
ISSN 0269-0403 print/ISSN 1471-6933 online

# The Royal Musical Association
http://dx.doi.org/10.1080/02690403.2012.717470
http://www.tandfonline.com
308 MICHAEL RUSS
this piece, which was conceived initially as a work for one piano and percussion,
Bartók felt that two pianos would balance the percussion more effectively:
I already had the intention years ago to compose a work for piano and percussion.
Gradually the conviction grew stronger in me that one piano would not be in satisfactory
balance in relation to the often rather penetrating timbre of the percussion instruments.
The plan was therefore altered to the extent that two pianos instead of one would oppose
the percussion instruments.5
Bartók’s and Beethoven’s sonatas are of roughly the same duration. Performances of
the Beethoven sonata typically take around 25 minutes; Bartók’s performance
timings on the published score indicate that his work should last 24? 34ƒ, and the
recorded performance given by Bartók and his wife has a duration of 25? 55ƒ.6 Both
works are in (or, in Bartók’s case, on) C; both have a three-movement structure with
a weighty first movement in sonata form and a lengthy rapid finale. In the final
version of the ‘Waldstein’ Beethoven replaced the central movement (which became
the ‘Andante favorit’) with a much briefer affair serving as an introduction to the
finale. Bartók’s central movement is much more extensive: it is an example of
Bartók’s ‘night music’, and its Impressionistic sonorities leave it closer to Debussy
than to Beethoven. Each work has a lengthy finale in 2/4. In Bartók this is a brighter,
more diatonic movement in comparison with the darker, chromatic first movement.
The closest relationships exist between the first movements of the Sonata for Two
Pianos and Percussion and the ‘Waldstein’, and only these movements will be
discussed in the analysis below. Both works have principal themes that reiterate C
major (see Examples 1ab), and the opening subjects are then followed by ideas that
emphasize thirds. The most obvious similarities, however, are between the secondary
themes (see Examples 2ab), which have a focus on E and very similar melodic
shapes. In the recapitulation both composers begin their secondary themes on A.
There are a number of more subtle connections. For example, Bartók’s juxtaposition
of the leading note with its flattened counterpart, as in the passages that lead to the
exposition and recapitulation, may link to the opening of the ‘Waldstein’ where,
after the descent of the bass to Bh, the opening theme is transposed down a tone to
begin on a triad of B =. The opening of the ‘Waldstein’ moves briefly to the ‘sharp’
side of C (bars 23) and then to the ‘flat’ side (bars 57). Such symmetrical
balancing of sharp and flat is at the heart of Bartók’s harmonic vocabulary.
Beethoven’s contrasts of major and minor (as in the move from a chord of F major to
F minor in bars 78, or the shift in the recapitulation of the second subject from A
major to A minor) are fused by Bartók into major/minor tetrachords, examples of
5
Béla Bartók, ‘About the Sonata for Two Pianos and Percussion’, Essays, ed. and trans. Benjamin
Suchoff (Lincoln, NE, and London, 1992), 41718 (p. 417).
6
Recording by Béla Bartók and Ditta Bartók-Pasztory (broadcast performance recorded by Columbia
in 1940), available in Naxos Classical Archives, 9.80806.
BARTÓK, BEETHOVEN AND THE SONATA FOR TWO PIANOS AND PERCUSSION 309
Example 1a. Beethoven, ‘Waldstein’ Sonata, op. 53, first movement, opening, principal theme.
Allegro con brio
cresc.
10
12
decresc.
310 MICHAEL RUSS
Example 1b. Bartók, Sonata for Two Pianos and Percussion, first movement, principal theme, bars
329. # Copyright 1942 Hawkes and Son (London) Ltd. Reproduced by permission of Boosey and
Hawkes Music Publishers Ltd.
8
32 Allegro molto ( = 132)
Piano I
Piano II
Timp.
Perc. I
8
35
( sim. )
( sim. )
8
38
Example 2a. Beethoven, ‘Waldstein’ Sonata, op. 53, first movement, bars 2542 including secondary
theme (bars 3541).
25
27
cresc.
29
31 decresc.
decresc.
34
cresc.
dolce e molto ligato
38
cresc.
42
3
3 3
3
312 MICHAEL RUSS
Example 2b. Bartók, Sonata for Two Pianos and Percussion, first movement, secondary theme, bars
8490. # Copyright 1942 Hawkes and Son (London) Ltd. Reproduced by permission of Boosey and
Hawkes Music Publishers Ltd.
Un poco più tranquillo ( = 104)

84
dolce
Piano I
Piano II
Timp.
Perc. I
86
Example 2b (continued)
88
Tempo I
90
314 MICHAEL RUSS
which appear in several places (notably the opening of the coda, in bar 105).
Rhythmically, Beethoven’s clear juxtaposition of duple and triple is expanded by
Bartók into a structural argument spanning the work.
Some years ago, Malcolm Gillies wrote: ‘A continuing challenge in Bartók studies
remains for these borrowings and influences from other composers or schools to be
probed [. . .]. For too long has the concern with folk influence deflected attention
from this other, equal partner in the formation and continuing renewal of his style.’7
Bartók himself discussed folk music extensively, but made few detailed comments
about his own works, rarely going beyond formal outlines. Nor did he say a great deal
about the work of his contemporaries (although a need to position himself relative
to Schoenberg and Stravinsky is evident in some of his writings). As Gillies notes,
Bartók may have found it more acceptable to admit a folk influence than to admit the
influence of another composer. Whereas folk music belongs to no one individual and
can be used and adapted without stigma, the borrowing of art music could be seen
as weakness or immaturity. This view is part and parcel of the ‘Western Romantic
notion of genius’.8
This article will explore the limited but significant connections between the
‘Waldstein’ and the first movement of Bartók’s Sonata. Inevitably, some connections
will be more speculative than others and will seem to relate more broadly to Bartók’s
dialogue with principles of high Classical sonata construction. Rather than suggesting
a recomposition of Beethoven’s work, Bartók’s references to the ‘Waldstein’ may be
intended as metaphors for progressive, Beethovenian, high Classical structure, for
works that challenge the principles of construction in their time as Bartók’s music did
in his. Flowing on from this, it is relatively easy to take the view that borrowings from
Beethoven and to a lesser extent from Bach, and Bartók’s engagement with Classical
form, confirm a familiar and generous view of his relationship with the past in which
he attempted to make a synthesis of old and new. However, detailed consideration of
the way in which he transforms borrowed materials and remakes sonata structure
suggests a more anxious relationship with the past, and that Bartók concealed deep
anxiety about his place in musical modernism.
Bartók and Beethoven

Scholars have drawn a number of parallels between the music of Bartók and
Beethoven. In 1928, writing in answer to a question from Edwin von der Nüll,
Bartók acknowledged the importance of Beethoven’s influence on the early part of
his career:
7
Malcolm Gillies, ‘Bartók and his Music in the 1990s’, The Bartók Companion, ed. Gillies, 317 (p.
15).
8
Ibid., 14. See also Malcolm Gillies, ‘Stylistic Integrity and Influence in Bartók’s Works: The Case of
Szymanowski’, International Journal of Musicology, 1 (1992), 13960.
Why do I make so little use of counterpoint? [. . .] In my youth my ideal of beauty was not
so much the art of Bach or of Mozart as that of Beethoven. Recently it has changed
somewhat: in recent years I have considerably occupied myself with music before Bach,
and I believe that traces of this are to be noticed in the Piano Concerto and the Nine Little
Piano Pieces.9
Many of the direct connections with Beethoven relate to the earliest works of Bartók.
Günter Weiss-Aigner draws attention to several references to Beethoven’s ‘Pathé-
tique’ Sonata, op. 13, and Fifth Symphony, op. 67, in Bartók’s early Piano Sonata in
G minor (1894), and to the ‘Spring’ Sonata, op. 24, in Bartók’s Violin Sonata in
C minor (1895).10 But such references to predecessors are the stuff of many
composers’ formative years, when they might use earlier music to shape their
development. As Bartók’s career advanced it was, as János Kárpáti argues, ‘creative
principles and compositional methods’ that Bartók took from Beethoven.11 For
example, the integration of a free approach to counterpoint within large-scale
structures has a precedent in Beethoven’s String Quartet in C> minor, op. 131, in
that both works begin contrapuntally.12 But there is some evidence of more direct
appropriation of materials. For example, the three-quaver reiterations and falling
minor third at the opening of the second movement of the First Piano Concerto
(1926) recall Beethoven’s Fifth Symphony. (The first 68 bars of this movement,
written for only piano and percussion, also look forward to the Sonata for Two
Pianos and Percussion.)
The overt influence of Beethoven is evident in several other late Bartók works. The
most frequently commented upon is the link between the ‘Heiliger Dankgesang’
from Beethoven’s String Quartet op. 132 and the slow movement (‘Adagio
religioso’) of Bartók’s Third Piano Concerto. The fact that both were written
when their respective composers felt they were recovering from illness has not been
lost on commentators. Kárpáti notes that the ‘structure of the dialogue [. . .]
definitely alludes to the ‘‘Heiliger Dankgesang’’ movement of Beethoven’s String
Quartet in A minor, op. 132’.13 Both Beethoven and Bartók open quietly and
imitatively with successively lower voices entering at a distance of two crotchets
9
Edwin von der Nüll, Béla Bartók: Ein Betrag zur Morphologie der neuen Musik (Halle, 1930), 108,
quoted and trans. in Halsey Stevens, The Life and Music of Béla Bartók (3rd edn, Oxford, 1993),
2312.
10
Günter Weiss-Aigner, ‘Youthful Piano Works’ and ‘Youthful Chamber Works’, The Bartók
Companion, ed. Gillies, 1019 and 21525. Between 1899 and 1901 Bartók orchestrated the first
194 bars of Beethoven’s ‘Pathétique’, one of a number of exercises (BB19) he submitted to Koessler.
See László Somfai, Béla Bartók: Composition, Concepts, and Autograph Sources (Berkeley, CA, and
London, 1996), 300. I am grateful to Amanda Bayley for pointing this out.
11
János Kárpáti, Bartók’s Chamber Music (Stuyvesant, NY, 1994), 23.
12
Ibid., 400.
13
János Kárpáti, ‘The First Two Piano Concertos’, The Bartók Companion, ed. Gillies, 498514 (p.
513).
316 MICHAEL RUSS
working down through the strings (five parts in Bartók’s case, but with divided
violins and no double bass; Bartók also introduces a low clarinet to his texture). An
archaic spirit occupies these movements; the writing in both is diatonic, and
Beethoven makes a striking use of the Lydian mode. The older composer’s six-bar
phrases unfold one after the other, each introduced by a point of imitation. Bartók
begins with a more continuous 15-bar passage, but thereafter the phrases get shorter
and are separated by the piano soloist, who begins with an unmistakable reference to
the first four notes of Beethoven’s ‘Heiliger Dankgesang’.
John Meyer traces Bartók’s dialogue of strings and piano back to the brief central
movement of Beethoven’s Fourth Piano Concerto. While the dotted rhythms and
unison string writing of this concerto are distant from the opening of the Bartók
movement just discussed, the solemn homophonic piano writing is not; nor is the
principle of interweaving strings and piano in such a way that the phrases of each
could be connected if the phrases of the other were to be removed. Meyer
demonstrates that Beethoven employed broadly similar pitch shapes in both strings
and piano, as does Bartók (the first four notes in the piano melody in bars 1617 are
anticipated in inversion in the strings in bar 12). Meyer points out that the opening
of the slow movement of Bartók’s Second Piano Concerto contrasts archaic strings
with piano in a similar way (although the piano writing there is more elaborate and
improvisatory).14
Mark Radice refers to the Adagio of the Third Piano Concerto as a ‘parody’ of the
Beethoven passage, but uses this term without any sense of the sarcastic or comedic
imitation that is sometimes found in Bartók. At face value, it seems that Bartók was
simply drawing a parallel between his own circumstances and those of Beethoven,
and making perhaps his final tribute to that composer. However, in his extensive
analysis of this passage Joseph Straus offers a more Bloomian interpretation. He
argues that Bartók extracted pitch-class sets from Beethoven that have motivic and
structural implications beyond the passage in question, as though he is offering a
‘radical commentary. He depicts himself as one who does fully what Beethoven can
only partially do.’15 Such an argument, questionably perhaps, turns pitch-class sets
into a resource actively deployed by composers who recognized such a manner of
writing as inherently progressive. Radice also draws a number of parallels between
the unison entry in bar 14 of Bartók’s Sixth Quartet and the opening of Beethoven’s
Die grosse Fuge, op. 133. But again he does not speculate on the nature of, or
motivation for, the apparent influence.16 The instances of borrowing outlined here
are rare in Bartók’s output. His borrowing from the ‘Waldstein’ is without precedent
in making a direct reference to Beethoven in addition to having movement-spanning
structural parallels.
14
Meyer, ‘Beethoven and Bartók’, 31516.
15
Joseph Straus, Remaking the Past: Musical Modernism and the Influence of the Tonal Tradition
(Cambridge, MA, 1990), 139.
16
Radice, ‘Bartók’s Parodies’, 2559.
Bartók and sonata theory

In a statement that summarizes the characteristics that could well have attracted
Bartók to the ‘Waldstein’ as an emblem for progressive high Classical form,
Hepokoski and Darcy describe its first movement as a
famously deformational structure composed in the high-middle period of a composer

obsessed with conceptually ‘difficult’ modifications of standard sonata practice. The
‘Waldstein’ is an illustration of an intellectualist or highly sophisticated deviation from the
norm or at best of a lower-level default not of the norm itself.17
Beethoven’s sonata was composed on the cusp of the Classical and Romantic eras.
Bartók referred to the latter’s ‘excesses’ in an oft-quoted statement made in 1931,
just six years before the composition of the Sonata for Two Pianos and Percussion:
At the beginning of the twentieth century there was a turning point in the history of
modern music. The excesses of the Romanticists began to be unbearable for many. There
were composers who felt: ‘this road does not lead us anywhere; there is no other solution
but a complete break with the nineteenth century’.18
Folk music provided one new road; returning to the methods and procedures of
music current at the turn of the eighteenth and nineteenth centuries another.
Despite the impression given by his own minimal and rather formulaic notes on
his works, Bartók’s engagement with sonata concerned more than just the static
arrangement of themes and tonal centres. As James Webster puts it, understanding
sonata structures is a twofold process. He refers to Kurt Westphal, ‘who in the 1930s
distinguished between ‘‘Form’’ and ‘‘Formung’’: between form-as-shape (balance,
symmetry, proportions, architecture), and form-as-process (the dynamic develop-
ment of musical ideas through time)’.19 Bartók’s own descriptions focus on Form
rather than Formung. His notes were possibly influenced by Formenlehre from the
mid-nineteenth and early twentieth centuries that would undoubtedly have been a
part of his Germanic, Riemann-influenced musical education. They often provide
just an outline of the structure and comment on one or two details. In describing the
Fourth Quartet’s first movement, Bartók tells us that the work is in the ‘three
sections of sonata form: exposition, development, recapitulation’ and then provides
the bar numbers of these sections along with subdivisions (‘main theme’, ‘transitory
passage’, ‘secondary theme’, ‘closing theme’).20 He provides no information on tonal
17
James Hepokoski and Warren Darcy, Elements of Sonata Theory (Oxford, 2006), 235.
18
Bartók, ‘The Influence of Peasant Music on Modern Music’, Essays, ed. Suchoff, 3404 (p. 340).
19
James Webster, ‘Formenlehre in Theory and Practice’, Musical Form, Forms and Formenlehre: Three
Methodological Reflections, ed. William Caplin, James Hepokoski and James Webster (Leuven, 2009),
12339 (p. 123). Here and elsewhere below, italics are present in the original text unless otherwise
indicated.
20
Bartók, ‘Structure of the Fourth Quartet’, Essays, ed. Suchoff, 41213 (p. 412).
318 MICHAEL RUSS
centres, the nature of the themes or the character of the motives, and none on how they
contribute to the compelling dynamism of this quartet. In discussing the use of sonata
form in the first movement of the Fifth Quartet, Bartók draws the reader’s attention to
the sequence of tonal centres, which forms a whole-tone scale, and to the reversed
ordering of the recapitulation, thus creating (prior to the coda) an overall symmetrical
design. Interestingly, in a footnote Bartók provides an alternative, suggesting that the
recapitulation and coda could begin a little earlier in each case than is shown in his
diagram.21 From these descriptions it is easy to conclude, erroneously, that Bartók
regarded sonata form as a shell, a convenient framework to use in conveying outline
structure to audience and performers. Lászlo Somfai observes that Bartók deleted some
technical detail from programme notes, deliberately writing them in a ‘stereotyped’
form avoiding ‘exhibitionism’.22
The note on the Sonata for Two Pianos and Percussion makes clear that Classical
models were at the front of his mind when composing the work, and reads almost as
though Bartók were describing a late eighteenth- or early nineteenth-century sonata
form. Apart from discussion of the unusual instrumentation and the references to
ostinato, the work’s modernist elements are marginalized:
The formal structure of the work can be related as follows. The first movement begins
with a slow introduction in which a motive of the Allegro movement is foreshadowed.
The Allegro movement itself is in C and is in sonata form. In the exposition the main
theme group is announced, consisting of two themes (of which the second has already
been alluded to in the introduction), after which follows the secondary (contrasting)
theme. Out of this a codetta develops on rather broad lines, at the end of which a brief
reference to the contrasting theme occurs by way of conclusion. The development, after a
short transition of superimposed layers of fourths, consists essentially of three parts. The
first one, in E, employs the second theme of the main theme group as an ostinato motive,
over which the first theme of the main group proceeds in the form of imitatively treated
interpolations. The second part is in the nature of a short interlude, after which the first
part with the ostinato in G sharp and inverted is repeated in a much altered form. In
the recapitulation there is no proper closing section; it is replaced by a rather extensive
coda (with a fugato beginning) built on the closing theme.23
As Hepokoski and Darcy explain, once a composer invokes the word ‘sonata’ they
are in dialogue with ‘an intricate web of interrelated norms as an ongoing action in
time’.24 Although primarily focused on the music of the late eighteenth century,
their recent extensive work on sonata theory provides a framework for the
exploration of Bartók’s return to Beethovenian form in his Sonata. Their work
points to the need for sonata structures to be directed towards particular key events,
21
Bartók, ‘Analysis of the Fifth String Quartet’, Essays, ed. Suchoff, 41415 (p. 414).
22
Somfai, Béla Bartók, 13.
23
Bartók, ‘About the Sonata for Two Pianos and Percussion’, 417.
24
Hepokoski and Darcy, Elements of Sonata Theory, 11.
Figure 1. Beethoven, ‘Waldstein’ Sonata, op. 53, and Bartók, Sonata for Two Pianos and Percussion:
formal plans of first movements.
for example the ‘medial caesura’ (MC) and ‘essential expositional closure’ (EEC).
These events remain important in Bartók’s music, where there are firm tonal centres,
but the capacity to generate ‘dynamic development’ towards key events is often
achieved through careful control of parameters other than functional tonality. Close
analysis informed by Hepokoski and Darcy’s recent theory confirms that Bartók was,
consciously or unconsciously, engaging with, challenging and revitalizing some of
the very same compositional issues as Beethoven.
Formal structures
Figure 1 provides an outline structure of the Beethoven and Bartók first movements.
Bartók adds a slow introduction of 31 bars that tends to enhance the movement’s
sense of gravity and significance as does its ‘creation music’ opening, slowly emerging
from the depths. The theme, speeded up, is significant later in the movement, and
the latter part of the introduction introduces an important rhythm of three crotchets
and two quavers (referred to as ‘a’ below) that foreshadows the primary theme of the
Allegro. Both movements have short transitions, but Bartók’s exposition is expanded
in relation to Beethoven’s, mainly by the addition of a substantial codetta. Each
composer has a lengthy development section. Bartók compresses the recapitulation,
but he may have taken a lead from Beethoven in having a substantial coda of 117
bars (Beethoven’s 51-bar coda was extensive for his time).
Hepokoski and Darcy note that Beethoven’s first subject is ‘underdetermined’. In
other words, the tonic is understood but ‘not secured with an authentic cadence’,
320 MICHAEL RUSS
making the transition essentially a brief (eight-bar) ‘dissolving consequent’ to the

opening subject’s antecedent.25 The opening subject of Bartók’s Allegro is also brief
and similarly ‘underdetermined’. Although it consists of four two-bar phrases based
on the main theme, in combination they end on an imperfect cadence as the timpani
falls to G with the dominant chord minus the third above. This, and the rapid
tempo, endows bars 3240 with an antecedent character. The theme gives way to the
second main theme, which as Bartók states is loosely related to the material of the
introduction, presented in a two-part sequence. Rather than feeling like a grand
consequent to the opening antecedent, this passage has a connective role, the central
part of a small ternary structure where dynamic and register build towards the return
of the opening subject, focused initially on A. This little form is open-ended, and
there is a parallel with Beethoven in the way in which the final section ‘dissolves’ into
a short transition at bar 69. Beethoven’s main theme group is less extensive and lacks
Bartók’s ternary structure, but does have its own particular drama at its heart when,
in bars 1314, the momentum of the music is interrupted by the tonic minor
arpeggio falling to a pause on G.
Both Beethoven and Bartók have secondary themes in or on E, with very definite
rhythmic similarities (Examples 2ab). Hepokoski and Darcy describe the writing of
the secondary theme in the mediant as a ‘deformation’ investigated by Beethoven
around 1800.26 Beethoven’s subject has a headnote of G>; Bartók’s subject begins on
E, with a subsidiary focus on G>. Both composers’ secondary themes are well defined
and stand out clearly from the music around them as brief, quiet lyric interludes.
Their contour and phrase structure is almost identical, and both composers balance
the appearance of the mediant in the exposition with a recapitulation of the
secondary theme in/on the submediant, A. In the recapitulation Beethoven then
moves to C via A minor, Bartók to C via F>. Neither composer employs his
secondary theme in the development.
One of the most fascinating connections between the two works is their structural
juxtaposition of duple and triple rhythms, something Bartók accentuates. His
codetta and coda are both firmly in the compound triple metre of the 9/8 time
signature, but his main subject, with its ‘a’ rhythm, when removed from the notated
time signature, suggests a 4/4 metre (a subject to which I return below). In the
‘Waldstein’ Beethoven completes his second subject with a ‘long excursus in triplets’
after both primary and secondary subjects employ only duple divisions of both beat
and bar.27 Despite the cadence at bar 50 in Beethoven’s sonata, Hepokoski and
Darcy regard this ‘excursus’ as prolonging the secondary theme group to bar 74 and
the ‘trill cadence’ there (the EEC). Bartók completes his secondary theme with a
25
26
Ibid., 119.
27
Ibid., 152.
cadence whose final chord is preceded by five bars of trills; the lead into the
development in bar 175 is also preceded by trills.
In Bartók’s own analysis, the first movement has a standard two-part exposition
with a codetta ‘on rather broad lines’ (see quotation above). In Hepokoski and
Darcy’s theory, the presence of, approach to and nature of the medial caesura (MC)
in such expositions is of particular significance. The medial caesura normally appears
immediately before the secondary theme [S]; Hepokoski and Darcy go so far as to
say: ‘If there is no MC, there is no S.’28 Beethoven’s exposition is particularly
significant in this respect. Once the repeat of the primary theme dissolves into the
transition at bar 23, an energy-gaining module in semiquavers affirms the dominant
of the secondary theme. In late eighteenth-century sonata forms this dominant
preparation would be followed, typically, by three hammer-blow chords and a
moment of silence.29 Beethoven expands the hammer blows to a four-bar passage of
E:V53, 64, 53 arpeggios that gain momentum in bars 278, not least through the
threefold use of szforzando, before collapsing downward in bars 2930. The medial
caesura belongs on the last crotchet beat of bar 30, where Beethoven could have
written a single crotchet rest and simply begun the secondary theme in bar 31 (see
Example 2a). Instead, he covers this silence by reversing the downward direction of
the semiquaver arpeggio. Then, rather than proceeding directly to the secondary
theme, he inserts a four-bar link passage of leaping octave quavers with a decrescendo
to piano, draining the impetus of the transition.
Bartók learnt much from Beethoven about the ebb, flow and articulation of sonata
form. He uses caesuras freely as a means of formal articulation, often corresponding
to timing marks in his scores, as is the case with the medial caesura at the end of bar
83. The caesura comes at the end of a diminuendo during which, from bar 80, a
reiterated D pedal is joined by some quiet dissonant dyads spelling out the
symmetrical tetrachord 4-9: [1,2,7,8] so familiar in Bartók’s music and often referred
to as chord ‘Z’. The pedal D, as much by virtue of location as of harmonic function,
has the character of a dominant preparing for the approaching secondary theme.
Beneath it there are timpani glissandos between G> and D. As in Beethoven, the
point of caesura is not entirely silent; on the final beat of bar 83, nearly all activity
ceases except for a single reiterated D in the second piano and a timpani glissando.
There is no equivalent to Beethoven’s linking passage, and the music proceeds
directly to the secondary theme. In a conventional VI move to a new key, the
dominant pedal is absorbed as the fifth of the new tonic. Here Bartók’s G> and D are
both absorbed into the dominant-seventh chord on E in third inversion that begins
the secondary theme at bar 84; the union of this chord (set 4-27: [2,4,8,11]) and the
‘Z’ tetrachord that precedes it is the octatonic hexachord 6-Z29: [1,2,4,7,8,11];
the dominant seventh normally appears before a new tonic, and the aggregate set of
28
Ibid., 117.
29
Hepokoski and Darcy cite Mozart’s Piano Sonata in D, K.284, I, bar 21 as an example. Ibid., 304.
322 MICHAEL RUSS
the V7/I progression is a diatonic hexachord (6-Z47). Bartók’s octatonicism links

him to Stravinsky and other early twentieth-century modernists, but the underlying
structural principle of a short open-ended exposition of the main theme group
dissolving into a short transition, leading to a clear dominant preparation and medial
caesura, clearly owes much to his Classical predecessors.
Hepokoski and Darcy define a point of essential expositional closure at the ‘first
satisfactory perfect authentic cadence that proceeds onward to differing material’.30
In the ‘Waldstein’ it is relatively easy to place this moment at bar 50, immediately
before the ‘long excursus in triplets’ referred to above.31 In the Sonata for Two
Pianos and Percussion, the falling of the timpani through a fifth in bars 989,
coupled with semitonal voice-leading, suggests that Bartók’s equivalent of the
essential expositional closure is bar 99, leaving the conclusion of the secondary theme
on the movement’s ‘true’ dominant, G, itself preceded by a long pedal on its
dominant, D. Bartók’s equivalent to Beethoven’s ‘excursus’, the extended codetta
with its compound triple rhythm, then ensues.
The codetta builds in energy and dynamic with overlapping imitative lines and
contrary motion in the two pianos and timpani glissandos. It concludes with a
dramatic caesura at the end of bar 160 followed by the return to the secondary
theme, which now lacks a proper closure and is dovetailed into the passage of fourths
that leads into the development (as referred to in Bartók’s note). The origin of this
return to the secondary theme lies not in Beethoven’s exposition, but in the extensive
coda of the ‘Waldstein’, where Beethoven returns to his secondary theme at bar 284.
Beethoven’s coda is almost a second recapitulation, beginning in D= major (an
expanded Neapolitan), which builds in a longer and more dramatic way to two
paused dominant sevenths replacing the medial caesura.
Bartók’s three-part development will be discussed later in this article, as it is
distant from Beethovenian explorations of sonata except for the establishment of
pedals, the use of sequence and the final poco allargando that lead to the
recapitulation. The most obvious link with Beethoven in Bartók’s recapitulation is
the initial return of the secondary theme on A. Beethoven quickly takes his secondary
theme via A minor to C and the point of ‘essential structural closure’ at bar 211.32
Bartók’s recapitulation of the primary theme is a compressed affair. After an entirely
recomposed transition, he returns to his reworked secondary theme. The original 15
bars are now expanded to 35, with statements of the theme on A and F> before
reaching C at bar 309, where the theme is presented in a largely whole-tone form.
While the closing passage at bar 326 in the recapitulation almost exactly parallels the
equivalent passage at bar 99 in the exposition, it lacks the fall of a fifth in the
timpani, so the sense of closure expected here is weakened. The recapitulation ends
30
31
While bar 50 is, as they point out, a strong candidate, Hepokoski and Darcy prefer to delay its onset
until bar 74. Ibid., 152.
32
See ibid., 232, for a discussion of essential structural closure.
by fading to a caesura in bar 331 in the two pianos, bridged by a side-drum roll and
crescendo announcing the extensive coda. Early strong closure may have suggested
that the coda lies beyond the essential part of the movement whereas it is, in fact,
integral to it. Again the origins may lie in Beethoven’s relatively weak (in Hepokoski
and Darcy’s view at least) expositional and essential structural closures. Bartók and
Beethoven both end their codas with final, powerful statements of the primary
theme.
Synthesis
Straus, whose focus is entirely on the music of the first half of the twentieth century,
describes three models of musical influence: influence as immaturity, influence as
generosity and influence as anxiety.33 The first category clearly relates to the work of
Bartók’s apprentice years, and the second might seem appropriate to describe the
relationships with Beethoven just discussed. This view of influence echoes the critical
writings of T. S. Eliot and his description of the poet as a ‘catalyst’: a benign presence
allowing the coming together of the elements from which the work is made, elements
(in Bartók’s case modernism, folk music and the sonata tradition) that would not
otherwise fuse. Bartók’s foregrounding of Beethoven in the Sonata might be seen to
resonate with Eliot’s view that ‘not only the best, but the most individual parts of the
work may be those in which the dead poets, his ancestors, assert their immortality
most vigorously’.34 Eliot’s view was that the individual artist should labour to gain
the historical sense, which we may call nearly indispensable to anyone who would
continue to be a poet beyond his twenty-fifth year; and the historical sense involves a
perception, not only of the pastness of the past, but of its presence; the historical sense
compels a man to write not merely with his own generation in his bones, but with a
feeling that the whole of the literature of Europe from Homer and within it the whole of
the literature of his own country has a simultaneous existence and composes a
simultaneous order.35
This resonates with Bartók, even if the musical equivalent of ‘the literature of his own
country’ was folk music. But Eliot’s casting of the creator in the role of catalyst carries
with it the corollary depersonalization, a quality less easy to square with the Hungarian
composer: ‘What happens is the continual surrender of himself as he is at the moment
to something which is more valuable. The progress of the artist is a continual self-
sacrifice, a continual extinction of personality.’36 Eliot’s characterization of the artist
33
Straus, Remaking the Past.
34
T. S. Eliot, ‘Tradition and the Individual Talent’ (1919), Selected Essays: 19171932 (London,
1932), 311 (p. 4).
35
Ibid., 4.
36
Ibid., 67.
324 MICHAEL RUSS
as generous towards his predecessors might seem to fit well with the view that
‘synthesis’ is at the heart of Bartók’s aesthetic. The use of this word in Bartók studies is
often traced back to a statement made by Bartók to Serge Moreux in 1939, not long
after the completion of the Sonata for Two Pianos and Percussion:
Kodály and I wanted to make a synthesis of East and West [. . .]. Because of our race, and
because of the geographical position of our country which is at once the extreme point of
the East and the defensive bastion of the West, we felt this was a task we were well fitted to
undertake. But it was your Debussy, whose music had just begun to reach us, who showed
us the path we must follow. And that, in itself, was a curious phenomenon when one
recalls that at the time, so many French musicians were still held in thrall by the prestige of
Wagner [. . .]. Debussy’s great service to music was to reawaken among all musicians an
awareness of harmony and its possibilities. In that, he was just as important as Beethoven,
who revealed to us the meaning of progressive form, and as Bach, who showed us the
transcendent significance of counterpoint [. . .]. Now what I am always asking myself is
this: is it possible to make a synthesis of these three great masters, a living synthesis that
will be valid for our own time?37
Moreux, like others after him, subsequently oversimplifies Bartók’s aesthetic to

‘attempting to synthesize the modal and the tonal, impressionism and German
classicism’,38 whereas Bartók speaks broadly of ‘awareness’ of the ‘possibilities’ of
harmony, the ‘meaning of progressive form’ and the ‘transcendent significance of
counterpoint’. In the 1960s, Kárpáti embraced and developed the idea of synthesis,
writing that ‘studies of [. . . Bartók’s] style and musical language have amply shown
the elements he availed himself of in the rigorous discipline of Bach’s counterpoint
on the one hand and in the refined technique of Beethoven’s last quartets on the
other’. To this Kárpáti adds not only a second synthesis: ‘that of the major trends of
his own age’ which he defines as Schoenberg’s expressionism, atonality and
developing variation, and Stravinsky’s renewal of old forms (together with a touch of
primitivism) but also a third: ‘folk music and the learned traditions’.39 Later, in an
analysis of the Sonata for Two Pianos and Percussion, Kárpáti continues to place
synthesis at the core of his approach, finding ‘in the Sonata a kind of model
synthesis. [. . .] the struggle between three-movement and four-movement forms
results in a peerless structure’.40 Such views invoke synthesis as a defence against
those who would denigrate Bartók as a folklorist or would (like Adorno and Boulez)
view him as not quite the equal of Schoenberg and Stravinsky. The major
37
Bartók, private interview with Serge Moreux, 1939, reported in Moreux, Béla Bartók (London,
1953), 92.
38
Ibid.
39
János Kárpáti, ‘Bartók, Schoenberg, Stravinsky’, Bartók Studies, ed. Todd Crow (Detroit, MI, 1976),
938 (pp. 94, 98). (The original article dates from the 1960s: János Kárpáti, ‘Bartók, Schoenberg,
Stravinsky’, New Hungarian Quarterly, 10 (1966), 21116.)
40
Kárpáti, Bartók’s Chamber Music, 400.
instrumental works of Bartók’s final period are often used as examples of the high
point of this synthesis in their combination of Classical formal rigour, counterpoint,
modernist harmony and folk-derived material.
The Sonata for Two Pianos and Percussion makes reference to Bach as well as to
Beethoven. The shape of the opening subject (see Example 3a) is closely related to the
subject of Bach’s Fugue in C> minor from the first book of Das wohltemperierte Klavier
(a thematic shape that also echoes the BACH motive), and Bach might be more loosely
associated with the movement’s contrapuntal aspects. The harmony is modernist, but
not Debussian in character, although the more Impressionistic aspects of the second
movement do connect more directly with that composer. In addition to the three
composers named by Bartók, there is a possible link with Liszt, who was influential on
Bartók’s early work. The first movement, in Ferenc Bonis’s view, begins with a
‘paraphrase’ of a theme from Liszt’s ‘La leggierezza’, the second of the Trois études de
concert (18459).41 As shown in Example 3b, there are some similarities to Bartók’s
opening theme in shape, rhythm and doubling in sixths (Bartók’s theme is doubled in
sixths from bar 8). Bonis’s choice of the word ‘paraphrase’ implies a deliberate, if
modified, borrowing from his countryman. The borrowing is from a work Bartók had
‘studied and also performed’ and, if not coincidental, is perhaps a simple reference to
his own musical upbringing as a pupil of István Thomán, himself a pupil of Liszt.42
Kárpáti doubts the significance of this quotation, rightly questioning, despite the
similarities, quite why Bartók would quote from this particular work in this context,43
and the relationship with Liszt will not be pursued further here.
Identifying thematic and structural connections with other composers is far from
demonstrating that synthesis or fusion has taken place. At times, the presence of a
satisfactory synthesis may simply be asserted. Lendvai’s flamboyant description of the
Sonata for Two Pianos and Percussion is so broad-brushed as to lack any real
meaning: ‘[it is] one of the most universal syntheses in the life-work of Bartók. In
organising the form he went so far that the material and content, symbol and
programme, means and meaning have become virtually inseparable from one
another.’44 Elliot Antokoletz, also speaking of the Sonata, describes it as ‘an exemplar
of the composer’s lifelong evolution toward synthesis of divergent folk and art music
sources’, with Classical formal procedures providing an important ‘structural control’:
in the 1930s, a greater Classical concern is manifested in the meticulous design of his
complex structures and in the clarity, distinction, and balance of the formal components
41
Ferenc Bonis, ‘Quotations in Bartók’s Music: A Contribution to Bartók’s Psychology of
Composition’, Studia musicologica Academiae Scientiarum Hungaricae, 5 (1963), 35582 (p. 361).
The passage identified by Bonis begins in bar 27, where Liszt has modified the theme from its first
appearance in bar 11.
42
Barbara Nissman, Bartók and the Piano (Lanham, MD, and Oxford, 2002), 135.
43
44
Ernő Lendvai, The Workshop of Bartók and Kodály (Budapest, 1983), 319.
326 MICHAEL RUSS
Example 3. Bartók, Sonata for Two Pianos and Percussion, first movement. Properties of the opening
idea.
(a) Bartók, Sonata for Two Pianos and Percussion, first movement, bar 5 (Piano I).
B A C H
1 4 1 5 1 3 5 1
6 6
(b) (i) Liszt, Trois études de concert (1845−9),‘La leggierezza’, bar 27;
(ii) Bartók, Sonata for Two Pianos and Percussion, first movement, bar 5 (Piano II).
(i) espressivo
(ii)
(c) Kárpáti’s analysis of theme: (i) compound nature; (ii) incremental pattern of intervals.
(i)
(ii)
4 5 6
[3+1] [4+1] [5+1]
on all levels of the music. The result is a new sense of rigorous structural control over the
enormous dynamic energy concentrated within the compositional framework. It is in such
works as the Sonata that we can observe an ideal relation between the highly synthesized
musical language that Bartók had begun to evolve from the first decades of the century
and the Classical forms that were being revitalized during the more conservative era of the
1930s and 1940s.45
How Bartók ‘juxtaposed, transformed and eventually synthesized both chromatic

and modal sources’ through the use of various scales, symmetries and interval cycles
is important to understanding this ‘highly synthesized musical language’, as
Antokoletz demonstrates in a number of writings.46 The synthesis of ‘folk and art
music sources’ requires yet another form of synthesis since, as Judit Frigyesi points
out, these two sources are ‘at first glance contradictory’, one having a communal
spirit and the other representing the ‘personal feeling of the artist’.47
Synthesis carries with it the idea that the synthesized product is not the equal of
the original materials. Kárpáti, drawing on Adorno in The Philosophy of the New
Music, argues that while seeking to reconcile Schoenberg and Stravinsky might be an
‘advantage’, the music of Bartók might be ‘depreciated by the compromise’.48 Gillies
remarks:
History has, however, been reluctant to place the synthesist, however creative, on the same
level as the more single-minded iconoclast, however destructive. And with Bartók there are
other concerning issues, anyway: How fundamentally original was he? Did he overstep the
limits of ‘good taste’ in synthesising others’ materials?49
Arriving at a positive assessment of these issues is usually achieved analytically, by

demonstrating the organic, unified nature of Bartók’s music. Organicism is a key
value in the music of the late nineteenth and early twentieth centuries and the
analytical strategies that flow from it, strategies that rescue Bartók from possible
criticism from commentators moulded by the ideas of Schoenberg and Adorno.
Kárpáti, in his discussion of the synthesis of Schoenberg and Stravinsky in Bartók,
tellingly refers to an ‘organic synthesis’ ‘Bartók merely used them in forming his
own creative system.’50
Bartók’s music encourages an ‘organic’ approach. For example, the opening idea of
the Sonata for Two Pianos and Percussion is often regarded as a source from which
45
Elliot Antokoletz, ‘Organic Expansion and Classical Structure in Bartók’s Sonata for Two Pianos and
Percussion’, Bartók Perspectives, ed. Elliot Antokoletz, Victoria Fischer and Benjamin Suchoff
(Oxford, 2000), 7794 (p. 77).
46
Ibid.
47
Judit Frigyesi, Bela Bartók and Turn of the Century Budapest (Berkeley, CA, and London, 1998), 121.
48
Kárpáti, ‘Bartók, Schoenberg, Stravinsky’, 97.
49
Malcolm Gillies, ‘The Canonisation of Béla Bartók’, Bartók Perspectives, ed. Antokoletz, Fischer and
Suchoff, 289302 (p. 295).
50
Kárpáti, ‘Bartók, Schoenberg, Stravinsky’, 98 (my italics).
328 MICHAEL RUSS
many subsequent elements grow. Not only does it come back in various
transformations, it also contains all the interval classes except 2 (Example 3a). Each
interval within the opening motive becomes a point of focus somewhere in the
work.51 Interval classes 1, 5 and 6 are prominent in the harmonic organization:
semitonal voice-leading combined with both doublings and bass motion by fifths and
tritones is evident in many places throughout the movement (see, for example, the
final cadence). Interval classes 4 and 3 are contrasted in the opening motive. Interval
class 4 is an important interval of doubling in the first movement: for example, in the
second and third imitative statements of the introductory subject (bars 6 and 12) and
at the presentation of the second subject (bar 84). It appears as a reiterated melodic
(minor-sixth) interval in the coda, which also foregrounds arpeggiated augmented
triads built entirely from this interval class in bar 377. The earlier, thematically
related, codetta reiterates interval class 3, the major sixth. This long-term dialogue of
interval classes 4 and 3 crystallizes in the prominent major/minor tetrachord (417),
notably at the beginning of the codetta (bar 101) and the conclusion of the coda (bar
417). Kárpáti points to the compound nature of the opening theme (Example 3c(i))
and to a process of logical expansion within the series of intervals (Example 3c(ii)),
itself a form of localized organic growth. Scholarly studies of motivic working, scales,
symmetries and interval cycles, even of quasi-Schenkerian voice-leading in Bartók’s
music, do yield impressive results and lend credence to his style being ‘highly
synthesized’. But such analyses inevitably diminish the sense of tension, of discomfort,
between the restored and the progressive, the borrowed and the original.52
Straus, speaking of two of Bartók’s contemporaries, observes that the generous
view of influence ‘obscures, or ignores, the deep ambivalence felt by artists
contemplating the past. It has no place for Stravinsky’s concern about being crushed
by strong pincers or Schoenberg’s distress at being forced to tread a harder road.’53
To some extent all composers have to work out their own position, their own balance
of tradition and innovation. For Bartók, the problem was his own position at ‘an
angle’, to use Stravinsky’s phrase,54 to the European tradition, and his feelings of
marginality in relation to, and alienation from, the musical equivalent of the
European literary tradition Eliot described. As noted above, the musical ‘literature of
his own country’ was folk music (although influences in Bartók’s late concert works
tend to come from a more generalized, less country-specific view of folk music).
Bartók was taught in the German tradition, primarily by the German-born Koessler,
who ‘barely spoke Hungarian’.55 But, while he respected the German musical
51
An extraordinary combination of all the opening intervals with a particular emphasis on ic 1 and ic 5,
some vertically and some horizontally, is heard in the finale beginning at bar 315.
52
See Kárpáti’s discussion of Adorno. Kárpáti, ‘Bartók, Schoenberg, Stravinsky’, 978.
53
Straus, Remaking the Past, 11.
54
Igor Stravinsky, in Stravinsky and Robert Craft, Dialogues and a Diary (Garden City, NY, 1963), 14.
55
Lynn Hooker, ‘The Political and Cultural Climate of Hungary at the Turn of the Century’, The
Cambridge Companion to Bartók, ed. Amanda Bayley (Cambridge, 2001), 723 (p. 17).
tradition of which Beethoven was the high point, Bartók both disliked the German
(and Jewish) hold on intellectual life in Budapest and felt alienated from the
urbanized folk music he heard there.56
Bartók would also have railed against any sense of subjugation to tradition and
against anything that caused the depersonalization of his work. In 1909 the young
composer wrote to Márta, soon to be his wife:
I strongly believe and profess that every true art is produced through the influence of
impressions we gather within ourselves from the outer world of ‘experiences’. He who
paints a landscape only to paint a landscape, or writes a symphony just to write a
symphony, is at best nothing but a craftsman. I am unable to imagine products of art
otherwise than as manifestations of the creator’s boundless enthusiasm, regret, fury,
revenge, distorting ridicule, or sarcasm. In the past I did not believe, until I experienced it
myself, that man’s works designate the events, the guiding passions of his life more exactly
than his biography.57
The focus here is very much on his life and passions, and the identification of
qualities such as ‘regret, fury, revenge, distorting ridicule, or sarcasm’ is highly
significant. He goes on to write:
It is strange that in music the basis of motivation has so far been only enthusiasm, love,
sorrow, or, at most despair that is only the so-called lofty feelings. It is only in our times
that there is place for the painting of the feeling of vengeance, the grotesque, and the
sarcastic. For this reason the music of today could be called realist because, unlike the
idealism of the previous eras, it extends with honesty to all real human emotions without
excluding any.58
These remarks were made by Bartók in his late twenties, a time when he encountered
disappointments in both music and love, an unsettled phase in his life that was to
continue for more than a decade. Arguably, such views had mellowed by the time of
the Sonata for Two Pianos and Percussion, but Bartók’s music continues to
demonstrate boundless energy and to retain at least some of the mixture of the
combative and the sarcastic that he describes. The letter says nothing about directing
any of these strong feelings towards his predecessors, nor does he say that they arise
from any feelings of marginalization or alienation, but the letter does suggest that
Bartók’s engagement with any borrowed material is unlikely to be a neutral affair.
56
Leon Botstein, ‘Out of Hungary: Bartók, Modernism, and the Cultural Politics of Twentieth-
Century Music’, Bartók and his World, ed. Peter Laki (Princeton, NJ, 1995), 362.
57
Letter to Márta and Hermina Zeigler (1909), in Bartók, Béla családi levelei, ed. Béla Bartók, Jr
(Budapest, 1981), 1878, trans. and quoted in Tibor Tallián, Bartók, Béla (Budapest, 1981, Eng.
trans. 1988), 88, and in Somfai, Béla Bartók, 11.
58
Ibid., trans. and quoted in Frigyesi, Béla Bartók, 120. (This passage is not included in the extract
quoted by Somfai, Béla Bartók, 11.)
330 MICHAEL RUSS
Influence as anxiety
Not least because of the composer’s own use of the term, ‘synthesis’ will remain a
valuable concept in Bartók studies. However, there is a need to examine alternative
interpretations that are less comfortable and which do not undervalue the tensions
that may exist when using borrowed materials or powerful structural principles from
the past. Understanding of the emotional impact of Bartók’s music may be enhanced
by proceeding from an observation that borrowed elements do not always fuse
willingly into a new organic whole, but offer varying degrees of resistance and at
times jar against each other (a jarring intensified by a dissonant language and, in the
Sonata for Two Pianos and Percussion, by the percussive nature of the work).
We should also keep in mind Bartók’s interest in the grotesque, a significant aspect
of his makeup. As Julie Brown has aptly argued, a precise definition of the musical
grotesque is perhaps elusive, but hybridity and bodily elements together with the
perceived debasing of music to more earthy and folk-like forms are core to our
understanding of this phenomenon.59 Bartók’s Sonata for Two Pianos and
Percussion is clearly a hybrid of late eighteenth-century sonata form and early
twentieth-century modernist harmony. It also has recourse at points to more earthy,
bodily and physical sounds, most obviously in the finale, where musical raspberries
and antics of various sorts seem to occur. In the first movement, the pounding
chords associated with the first subject have a strongly physical character.
Furthermore, the addition of extensive percussion, the most basic and physical of
musical instruments, which for the most part are less able to participate in the
intricacies of the post-tonal pitch language, might be interpreted as a grotesque
addition to the piano duet’s engagement with the sophisticated dialogue between old
and new formal and harmonic principles in this work.
In the remaining part of this article the discussion shifts towards the third of
Straus’s models of influence, ‘influence as anxiety’, which has its origins in the work
of the literary critic Harold Bloom.60 However, the temptation to invoke a set of
‘revisionary ratios’, whether the elaborate ones of Bloom or their musically orientated
reinterpretation in the work of Straus, will be resisted until the conclusion. Although
a less extravagant course will be charted, the intention is nonetheless to demonstrate
that Bartók is moving beyond the simple borrowing of materials and erecting a
defence against his predecessors as a means of creating his own compositional space.
In this context the actual treatment of the ‘Waldstein’ metaphors (and the reference
59
Julie Brown, Bartók and the Grotesque, RMA Monographs, 16 (Aldershot, 2007). On p. 25, Brown
quotes the same passage from Bartók’s letter to Márta that is quoted from Frigyesi above; however,
Brown uses the translation made by Gillies and Gombocz (Malcolm Gillies and Adrienne Gombocz,
Bartók Letters: The Musical Mind, which remains unpublished). In this translation the word
‘hyperbole’ is used in place of ‘grotesque’. Nevertheless, Brown’s study makes the significance of
grotesquery abundantly clear.
60
Harold Bloom, The Anxiety of Influence: A Theory of Poetry (New York and Oxford, 1973; 2nd edn,
1997); idem, A Map of Misreading (New York and Oxford, 1975; 2nd edn, 2003).
to Bach’s C> minor Fugue) becomes highly significant, because if Bartók is

consciously revising the ideas of his predecessors, this constitutes what might be
called ‘wilful misreading’. The extent to which Bartók knowingly misreads
Beethoven (and Bach) in the Sonata cannot be precisely defined without further
documentary evidence,61 but the thematic and structural parallels already identified
do make it seem likely that he did so. The approach taken here differs from a strictly
Bloomian one since in Bloom’s theory the influences which cause the poet most
anxiety are generally concealed. This makes it particularly problematic to apply the
theory to music, since in musical scores, even more than in poetry, it is difficult to
demonstrate influence that does not have some trace in the text.62 In wilfully
misreading Beethoven (and Bach), Bartók typically parodies the original, which may
be mocked, exaggerated and possibly endowed with an element of the grotesque. He
also demonstrates a more generalized ‘anxiety of style’, a less conscious misreading of
aspects of Classical style in which key structural elements may be undermined and
marginalized and/or mocked and exaggerated.
Bartók emphasizes his reference to the four-note subject of Bach’s Fugue in C>
minor in the introduction by retaining Bach’s notation of the interval between the
second and third notes as a diminished fourth. Bach’s fugue is lengthy and solemn;
the subject is terse, but flowers into long-breathed lines. Although Bartók’s opening
statement seems initially to expand the idea to three bars, once the quasi-fugal entries
get under way in bars 5 and 7, entries come after two quavers as though a stretto has
begun, and the ‘fugue’ is interrupted after a single bar by cymbal clashes, tremolo
clusters and glissandos; these interruptions seem to have no function other than to
counter, even to mock, the solemnity of the subject as well as interrupting the logical
symmetrical pattern of entries that is beginning to unfold. Entries of the subject in
bars 4, 8 and 12 move round the cycle of fifths, but each is answered by its tritone,
forming the cycle F>, C, G, D=, A=, D.63 In traditional fugal structure, the principle
of tonal answer at the fifth allowed the preservation of key, but the simultaneous
employment of entries at the tritone neutralizes that process, as does the employment
of ‘real’ rather than tonal answer. In the second and third pairs of entries (bars 8 and
12) each line is doubled at ic 4 (in the continuation at bar 13, this is replaced by ic
6). The result is to marginalize any sense of harmonic progression and replace it with
a sequence of non-tonal aggregate tetrachords that will play a significant role
61
Except, perhaps, the orchestration of the ‘Pathétique’ referred to in note 10.
62
Vikárius has argued that misreading of predecessors and contemporaries may combine with other
defences such as the use of folk music: ‘The allusion to folk music is a defence against great models,
possibly primarily against Beethoven.’ László Vikárius, Modell és inspiració Bartók zenei gondolk-
odásában: A hatás jelenségének értelmezéséhez (Model and Inspiration in Bartók’s Musical Thinking:
Toward the Interpretation of the Phenomenon of Influence) (Pecs, l999), 206; trans. and quoted by
Klára Móricz, ‘The Anxiety of Influence and the Comfort of Style’ (review of Vikárius, Modell és
inspiració ), Studia musicologica Academiae Scientiarum Hungaricae, 40 (1999), 4734.
63
The set formed by the last four pitch classes is 4-9: [1,2,7,8], the same set identified as significant at
the end of the transition (bars 803).
332 MICHAEL RUSS
elsewhere in the work.64 The interruptions in bars 6 and 10 might be seen as

grotesque, and the marginalization of tonal voice-leading in the canons at the
beginning suggests a wilful misreading of Bach’s fugue. Brown, writing of Bartók’s
Third String Quartet (1927), relates how if Bartók could
identify as dissimulative ‘empty’ sequences and fugata in Liszt, how much more might one
find such traditional, learned techniques in a non-tonal context? There the previously
cerebral musical activity of canon formation is emptied of all intellectuality [. . .]. Double
canon by inversion at some prescribed interval is no longer the height of musical
intellectuality but an empty shell.65
The passage from Bartók’s Third String Quartet that Brown analyses here is the
‘Seconda Parte’, in which ‘ironic and burlesque effects are juxtaposed. Towards the
end [. . .] a long canonic and fugato section emerges, at the end of which the last
canon audibly implodes into glissandi [and] apparent mockery.’66
Bartók produced an infamous parody of Shostakovich’s Seventh Symphony in his
Concerto for Orchestra. In that work, parody is used with sarcastic intent; Bartók’s
parody of the secondary theme of the ‘Waldstein’ in the Sonata, however, is ironic
and mocking rather than sarcastic (see Example 2b). Beethoven’s theme is an
elegantly harmonized idea in E major; the phrase structure is an entirely regular 44
(see Example 2a). The whole eight-bar sentence then comes back with a counter-
point in triplets before the more energetic business of the movement gets under way
with the second part of the subject group. Bartók’s melody begins on E and clearly
follows the contours of Beethoven’s. However, the latter’s material, epitomizing
calmness, order and compositional craft, is twisted and distorted by Bartók.
Beethoven’s functional harmony and voice-leading is emasculated by the use of
parallel first-inversion triads and mocked by the insertion of falling-fifth acciacca-
turas. Bartók further misreads Beethoven’s subject by placing it in a folk-based
‘Bulgarian’ rhythm which divides the 9/8 bars irregularly. Beethoven’s
rhythm becomes . On the face of it, the only difference in the first
bar is the small addition of a dot, but Bartók also makes the second bar almost
mirror the first and sets it against an echo of the a rhythm in the right hand of Piano
II. It is ironic that Bartók chooses Beethoven’s interlude of pure and regular classical
craftsmanship as a moment to demonstrate his rhythmic inventiveness and lack of
concern with Classical harmonic progression. After a brief interlude, the theme is
reworked. Beethoven’s triplet counterpoint is replaced by trills, a device often
associated with the mocking and ironic in music.
64
In bar 9 the sequence is: 4-17: [1,2,5,t] and [0,1,4,9]; 2-4: [4,8]; 4-21: [3,5,7,9] and [0,2,8,t]; 3-12:
[3,7,e]; 4-7: [2,3,6,7]; 4-19: [1,2,6,9]. In bar 12, the same sequence is transposed.
65
Brown, Bartók and the Grotesque, 145. Brown is referring to Bartók’s essay ‘Liszt’s Music and Today’s
Public’, Essays, ed. Suchoff, 4514, where he discusses the ‘irony’ of the fugato in Liszt’s Piano
Sonata in B minor (p. 453).
66
Brown, Bartók and the Grotesque, 141.
Bartók inverts the opening harmonic third, EG>, of Beethoven’s theme and
unfolds it as a downward melodic sixth projected across the first two bars (thus also
expanding the initial descending fifth to a minor sixth). Beethoven’s third is a
possible source of the ambiguity between Bartók’s initial tonal centre with its
accompanying chord (dominant seventh on E) and the scale of G> minor, which then
seems to govern the melody over the next two bars. The initial descending melodic
hexachord is diatonic in both B major and G> minor, but the FÜ leading note and FÜ
major triad in bar 86 suggest G> minor. The melody then shifts towards G (FÜ)
minor (bars 8990) before pausing on the leading note (F>) of G for four bars. F> is
harmonized with its own major triad over a pedal D (resulting in pitch-class set 4-19,
one of the harmonies heard beneath the opening idea in the introduction). This
moment in the music has many of the characteristics of a mid-point imperfect
cadence. After a brief caesura, the melody sets off on A=, and Piano I trills on this
note seem to confirm a G>/A= centricity. As mentioned earlier, at the end of the
secondary theme the timpani pedal falls from D to G. Above this is a triad of B
major, resulting in a rhyming transposition of the chord heard in bars 914 and the
impression of a functionally weak perfect cadence in G.67
Bartók is playing a game of fifths and thirds here, showing how he can build
harmonic features latent in Beethoven into a sophisticated structure that counter-
points old and new. In Classical sonata structure the pitch classes one and two fifths
above the tonic have specific functions as V and V of V. Beethoven modified this by
substituting E for G and B for D, each a minor third lower than ‘normal’, eventually
putting things right at the end of the exposition. But Bartók in his secondary theme
plays on both the old and the new simultaneously, and extends the cycles. So
(literally) over the motion from D to G in the timpani, the local tonal centre E
becomes associated with G>, G with B and D with F>. Furthermore, G>/A= takes on a
particularly significant role not only as the mediant of the mediant, two major-third
steps from C, but also as the tritone, two minor-third steps from D, held as a pedal
throughout with both notes being subsumed into the opening E7 chord. Figure 2
shows these relations using an extract from a Riemannian Tonnetz.
At the end of the exposition, rather than following Beethoven and directing the
music to the true dominant, G, Bartók states the secondary theme again in E in
Piano I, and this tonal centre is emphasized by the extended trill that runs from bar
166. The second piano initially echoes the theme on A, a fifth lower.
Arguably, the main theme of Bartók’s Allegro is unrelated to the ‘Waldstein’ (see
Examples 1ab). However, the repetitions of C major in both works do provide a
link, and it is possible that Bartók is both intensifying and reversing aspects of
Beethoven’s theme. The pianissimo C/E thirds at the opening of the ‘Waldstein’
67
Kárpáti regards the D as the dominant of G, but in the context of the brief suggestion of G minor in
the melody in bars 8890; he also casts it in the simultaneous role of ‘mistuned’ dominant of G>.
334 MICHAEL RUSS
Figure 2. Tonal centres mapped on an extract of a Tonnetz.
become heavily scored full triads over five octaves marked fortissimo in Bartók. There
are also some connections in rhythmic and motivic shape: Beethoven moves gently
up through a third at the end of his phrase, while Bartók moves forcefully down
through a third; and Beethoven’s attack-point rhythm of repeated quavers is
interrupted by two semiquavers immediately before the final note of the phrase,
whereas Bartók’s repeated crotchets are interrupted by two quavers before his final
note. However, Beethoven’s gently tapping theme fits comfortably and unambigu-
ously within his 4/4 metre; Bartók’s does not.
Anxiety of style
In more general terms, Bartók completes and exceeds Beethoven by the provision of
both a substantial introduction and a complete, modernist, second movement plus a
finale that is more organically wedded to the whole than Beethoven’s. His
instrumentation obviously exceeds Beethoven’s, realizing (and intensifying) an
implication inherent in the percussive elements in the ‘Waldstein’. However, similar
comparisons could be made with many Beethoven sonatas and what is being
described is a misreading of high Classical sonata style. In the discussion that follows,
the first movement of the ‘Waldstein’ remains a key text, but comparisons are now
made with a generalized view of the high Classical sonata, rather than a specific
influence of the Beethoven work, reflecting anxiety of style.
Bartók’s own descriptions of sonata form focus on the architectural shell rather
than the tonally dramatic and teleological. He describes it simply in terms of
exposition, development, recapitulation and coda, and points out symmetrical
arrangements of thematic materials and tonal centres rather than suggesting any
parallels in his music to Beethovenian tonal drama. Despite the structural parallels
outlined in the earlier part of this article, in the Sonata for Two Pianos and
Percussion the tonal drama that arises in Beethoven from the employment of clearly
functional harmony is marginalized and replaced with a momentum gained by the
manipulation of other parameters that takes the listener to goals arranged, at least in
part, symmetrically. As shown in Figure 1, the main thematic group, secondary
theme and development are all small, symmetrical ternary forms. The ternary
structure of the first subject group is obvious; the second subject’s return after the
material Bartók identifies in his note as ‘codetta’ completes the second ternary; and
the development has the three-part structure described by Bartók. Furthermore, the
recapitulation can be thought of as having three parts: the altered return of materials
from the exposition (the first subject is greatly condensed); the extensive coda; and
the return of the work’s main theme at the end.
By nesting a 3333 symmetrical pattern within the larger structure, Bartók
creates a hybrid, a symmetrical arrangement of thematic material within a structure
that remains Beethovenian and asymmetrical in the importance it places on the point
of recapitulation, roughly two thirds of the way through the movement. As it
happens, this moment falls at the precise golden mean of the entire first movement,
but only if calculated in bars or beats. If the golden mean is calculated using Bartók’s
timing marks, then it is some ten bars earlier.68 The wider significance of golden
mean proportions has been discussed elsewhere and the discrepancy between the bar/
beat and temporal calculations is indicative of the lack of credibility of these
calculations in Bartók analysis.69 As here, the way in which time elapses in a musical
work, or is perceived to pass, cannot be adequately accounted for in these
calculations and, as Somfai has pointed out, there is absolutely no evidence of
Bartók consciously working out golden mean and Fibonacci proportions in any of
his works.70 Nevertheless, in a much more general sense, it is hard not to see Bartók’s
location of the recapitulation at the precise golden mean of the first movement (and
very approximately at the temporal golden mean of the first-movement Allegro
only)71 as evidence of an almost architectural interest in formal proportion, even if
the exactitude of compliance with golden mean proportions is intuitive rather than
calculated. After ten performances of the original version, Bartók reduced the length
of the retransition by five bars, but the reasons may have been as much about
replacing an ‘explicitly pianissimo climax [. . . which was] beautiful music, but
appeared somewhat unprepared’ as about conformance with the golden mean.
Bartók found it difficult to ‘lead from [. . .] solitude to the loud recapitulation of the
main theme in a natural way’.72
‘Symmetricization’ is one of Straus’s ‘strategies for reinterpretation’ in which
‘traditionally goal-oriented harmonic progressions and musical forms [. . .] are made
68
In the first movement the recapitulation begins at bar 274 at the golden section (443
bars 0.618 273.77). The temporal golden mean (12? 10ƒ 0.618 7? 31ƒ) is roughly cotermin-
ous with bar 264.
69
See Ernő Lendvai, Béla Bartók: An Analysis of his Music (London, 1971), 1826; idem, The Workshop
of Bartók and Kodály, 3246; and Howat, ‘Masterworks II’, 3202.
70
Somfai, Béla Bartók, 81. Somfai points out that Bartók’s papers contain many calculations but none
of them related to golden mean/Fibonacci proportions.
71
The temporal golden mean of the first-movement Allegro falls at 5? 55ƒ, shortly after bar 274, which
using Bartók’s timings occurs 5? 40ƒ from the start of the Allegro.
72
Somfai, Béla Bartók, 198.
336 MICHAEL RUSS
inversionally or retrograde-symmetrical, and are thus immobilised’ as the later

composer seeks to defend himself against his predecessor in Straus’s (?mis-)reading of
Bloom’s theory of the anxiety of influence.73 Symmetry’s power to immobilize is
even more evident in Bartók’s harmony than in his form. The symmetries inherent
in Bartók’s pitch system have been much discussed and include the use of a variety of
symmetrical cells, the symmetrical deployment of tonal centres,74 and the use of
symmetrical scales including the hexatonic and octatonic.75 But, just as Bartók’s
formal language rests on a juxtaposition of the symmetrical and asymmetrical, so
does his pitch organization, as encapsulated in his frequent juxtaposition of fifth and
diminished fifth and his interest in cycles of thirds discussed above. The only tonal
centres Bartók refers to in his programme note are C, E and G>; the addition of G>
extends Beethoven’s mediant relation to a complete cyclic (ic 4) division of the
octave suggesting, in Bloomian terms, that his predecessor did not go far enough.
Bartók does not locate E and G> in the secondary theme but in the development,
suggesting perhaps that their use in the exposition anticipates their significance in the
larger structure. Bartók extends other symmetries that are latent in Beethoven. As
described above, the way in which Bartók’s introduction moves round the cycle of
fifths with each entry counterpointed by its diminished fifth may have its origin in
the way Beethoven’s opening bars allude, after C, to G major and to F major (then
minor). There are many other places in Bartók’s work where tonal centres are
arranged symmetrically for example, the first subject has a strong emphasis on the
minor-third cycle C, E=, F> and A, as does the recapitulation of the second subject,
where initially A= and F> are substitutes for C.
The emphasis on E and G> in the development reverses traditional assumptions
about the nature of the development as an area of tonal flux. Furthermore, rather
than experiencing a process of motivic development, the opening motive is frozen
into an ostinato (albeit against some development of the principal subject). The
structure of the development, Bartók’s emphasis in his programme note on tonal
centres within it, and the highlighting of thematic connections between introduction
and development rather than between exposition and recapitulation suggest that
important aspects of sonata structure are being turned inside out. Similarly, the coda
considerably exceeds the dimensions of the compressed recapitulation and, as will be
described a little later, has an important function in completing the core business of
73
Straus, Remaking the Past, 17.
74
See, for example, Elliot Antokoletz, The Music of Béla Bartók: A Study of Tonality and Progression in
Twentieth-Century Music (Berkeley, CA, and London, 1984); Jonathan Bernard, ‘Space and
Symmetry in Bartók’, Journal of Music Theory, 30 (1986), 185201; Lendvai, Béla Bartók ; and idem,
The Workshop of Bartók and Kodály.
75
Richard Cohn, ‘Bartók’s Octatonic Strategies: A Motivic Approach’, Journal of the American
Musicological Society, 44 (1991), 262300; Lendvai, Béla Bartók ; idem, The Workshop of Bartók and
Kodály; Michael Russ, ‘Functions, Scales, Abstract Systems and Contextual Hierarchies in the Music
of Bartók’, Music and Letters, 75 (1994), 40125; Wilson, The Music of Béla Bartók.
the movement. Codas, by definition, normally sit outside what Hepokoski and
Darcy term ‘generic sonata form proper’; they form part of ‘parageneric spaces (or
not-sonata-space)’.76 However, Bartók’s enhanced role for the coda may itself have
its origin in the ‘Waldstein’, since it is only at the end of the coda that Beethoven
restates his secondary theme in the tonic, completing what was unfinished in the
recapitulation.
The ostinato that forms the second idea in Bartók’s primary thematic group may
have its origins in the reiterated thirds in bar 14 of Beethoven’s sonata. In Bartók the
repeated thirds are part of an octatonic ostinato in what Straus might term a
‘symmetricization’ of the Beethoven passage. The octatonic ostinato is transposed
upwards by a fourth in bar 50, effecting a change of collection (see Example 4).77
However, beneath the ostinato are long timpani pedals on dominant and then tonic;
this rising fourth coincides with the transposition of the ostinato, suggesting an
expanded perfect cadence. But the VI move in the timpani is not supported
functionally by the octatonic harmonies, and the timpani pedals are not members of
the octatonic collections above them, creating a subtle metaphor for old/new, inside/
outside. (In a slight alteration, and as if to emphasize C’s ‘outside’ status, the second
piano retains this pitch class as its lowest voice.) It is as though the timpani stand
metonymically for the ‘key of C’, even though the harmony above seems to resist.
Transpositions of octatonic and other material through fourths or cycles of fourths
are prominent elsewhere in the first movement. In bars 6972 the transition begins
with the transposition of an octatonic pentachord through an eight-note cycle of
fifths; in bars 17594 the link into the development is built on chordal fourths
transposed by fourths; in bars 26473 a chromatic idea is transposed through a
complete (12-note) cycle of fifths to reach the recapitulation at bar 274; and in bars
33253 the coda begins fugally with entries moving four steps round the cycle of
fifths. Transposition by fifth, voice-leading by step and the use of pedals, as well as
the echoes of dominanttonic motion in the timpani, are all attributes of tonality
used metonymically in a movement which, while it has clear points of tonal focus,
contains not a single unambiguous functional progression.
In the ‘Waldstein’, quavers and semiquavers are the most prominent rhythmic
values, and the attack-point rhythm is often a continuous flow of one of these (most
often the latter). At key points this flow is interrupted by the steadier, but still duple,
rhythms of the secondary theme or by extensive passages of triplet quavers.
Significantly, the duple and triple divisions of the crotchet are not mixed.78 In the
Sonata for Two Pianos and Percussion, Bartók appears not only to intensify the
debate between duple and triple rhythmic types but also, in a turning inside out of
musical priorities, to make rhythm the equal of pitch in generating and subsequently
76
77
See Wilson, The Music of Béla Bartók, 145, Example 6.4.
78
There are, however, some very rare appearances of triplet semiquavers at bars 14650 and 275.
338 MICHAEL RUSS
Example 4. Bartók, Sonata for Two Pianos and Percussion, bars 4750. # Copyright 1942 Hawkes
and Son (London) Ltd. Reproduced by permission of Boosey and Hawkes Music Publishers Ltd.
47
Piano I
Piano II
Timp.
Perc. I
S.D. c.c.
S.D. s.c.
Perc. II
B.D.
49
T5
T5
T5
Octatonic collections:
C, D, E , F, F , A , A, B T5 F, (G), A , (B ), B, D , D, E
resolving long-term structural tension. At the end of the introduction the rhythmic
figure earlier called ‘a’ emerges; it is reiterated in several positions against the 9/8
metre (see Example 5). It is then used prominently as the rhythm of the main Allegro
theme, which repeats it four times. a has eight quaver beats (22211) and
sits more naturally in a 4/4 metre than it does in the notated 9/8. The second
rhythmic type is the straightforward compound symmetrical 9/8 of the work’s time
signature, most prominent in the first part of the introduction, codetta and coda,
where the bars are divided equally (333). In a 9/8 metre, a rhythm such as a
will, to use Cohn’s term, always appear to be ‘open’, since the closest multiples of 2
are one short or one in excess of 9.79 He reads these different rhythmic types
allegorically:
Downstream from Bratislava the open mode is natural, the closed mode stodgy and
imperial. The first movement can be read as a map of Bartók’s music-cultural psyche, or as
an allegory of the cultural position of Hungary in the first years of the twentieth century.80
In a Beethoven sonata, rhythmic argument is subsidiary to tonal; harmony and

tonality create forward momentum in a way that they are unable to in Bartók. Bartók
has a clearly organized set of tonal centres, but creating a strong sense of direction
between them depends on building energy using rhythm in combination with
dynamic and texture. But there is a deeper and more significant reversal of structural
priorities here. Cohn, referring to the work of Leong, points out that
in the course of the movement, rhythmic patterns generated by open intervals become
displaced by, or progress to, those generated by closed patterns. These progressions occur
on multiple structural levels, but undergo a reversal in the final measures of the coda. Her
[Leong’s] study demonstrates among many other things, the malleability of the sonata-
form framework. Historically used to show contrast, progression, transformation and
modulation in the tonal and thematic domain, it can show similar processes in the metric
domain as well.81
The debate between a and 9/8 takes a number of forms as the movement progresses,
and the two rhythmic types coexist in different degrees of tension. For example, the
relationship between them is more stable in the second statement of the main
subject in bar 61 than it is in the first at bar 32. In the transition at
bar 69, two variants of the a motive occur ( and ),
the second melding into a 9/8 rhythm: . These and similar
derivatives are found in other transitional sections and in the central section of the
79
Cohn, ‘PitchTime Analogies’, 51.
80
Ibid., 52.
81
Ibid., 51. His reference is to Daphne Leong, ‘A Theory of Time-Spaces for the Analysis of
Twentieth-Century Music: Applications to the Music of Béla Bartók’ (Ph.D. dissertation, Eastman
School of Music, 1999).
340 MICHAEL RUSS
Example 5. Bartók, Sonata for Two Pianos and Percussion, ‘a’ rhythm in bars 2934. # Copyright
1942 Hawkes and Son (London) Ltd. Reproduced by permission of Boosey and Hawkes Music
Publishers Ltd.
(più agitato) al
[1] 2 + 2 + 2 + 1 + 1 [1] 2 + 2 + 2 + 1 + 1 [1] 2 + 2 + 2 + 1 + 1
29
Piano I ( cresc. )
❉ ❉
Piano II ( cresc. )
[1] 2 + 2 + 2 + 1 + 1 [1] 2 + 2 + 2 + 1+1
❉ ❉
2 + 2 + 2 + 2 [1] 2 + 2 + 2 + 2 [1] 2 + 2 + 2 + 2 [1]
Timp.
Perc. I
( cresc. ) 2' 35''
[1] 2 + 2 + 2 + 1 + 1
8
32 Allegro molto ( = 132)
[1] 2 + 2 + 2 + 2 2 + 2 + 2 + 2 [1] [1] 2 + 2 + 2 + 2
α rhythm
2 + 2 + 2 + 1 + 1
development; their existence between the two rhythmic types reflects their
transitional function.
At the beginning of the recapitulation, the a rhythm and the main theme of the
work are alluded to only briefly, and 9/8 dominates. That 9/8 has prevailed is
subsequently confirmed by the extensive coda. When a returns at bar 423, its
repeated crotchets are augmented to dotted crotchets. This version of a is then heard
as a canon on A and E =, in which a is reconciled with 9/8.82 Only in the last two
bars of the movement does a make one last effort to reassert its original form. The
coda, a formal section traditionally considered subsidiary to the principal sections,
assumes a more significant structural role in the Sonata for Two Pianos and
Percussion as the seat of the resolution of movement-spanning rhythmic processes,
suggesting a reversal of the roles of recapitulation and coda.
So far it has been demonstrated that Bartók, while engaging with many of the
basic tenets of sonata form and modelling ideas on the ‘Waldstein’, is not always
generous in his approach. He parodies Beethoven and turns on their head some of
the key principles of high Classical sonata style. At times Bartók’s approach seems
almost to mock aspects of that style. For example, the most intense point in any
Beethovenian sonata form comes just before the recapitulation, the moment of
recapitulation itself being a powerful release. In both the first movement and the
finale of the Sonata for Two Pianos and Percussion, the lead-up to the recapitulation
is intense. From bar 264 of the former, overlapping statements of a chromatic idea
using variants of the a rhythm are transposed through the cycle of fifths, rising in
register and dynamic and, just before the recapitulation, doubled in tritones. Beneath
are timpani pedals falling from E = to B =, the flattened leading note, which at the
very last moment gives way to the ‘real’ dominant, G. But on arrival at bar 274 the
main subject is curtailed, represented only by fragments. The second piano’s written-
out C/B trill (doubled in fifths) turns the instrument into a giant cymbal which,
together with the xylophone and the timpani trills and glissandos, seems to mock the
whole event (see Example 6).
Other major points of articulation in the first movement combine elements of the
old and new into hybrid structures. Kárpáti points to the way in which the cadence at
the end of the introduction (bars 313, but building from bar 23) retains a vestige of
tonal function while also proceeding to the destination chord of C from collections
of semitonal neighbours (see Example 5).83 From bar 26, this collection takes the
form of rising arpeggios outlining repeatedly the diminished seventh B, D, F, A =,
supported by the timpani reiterating BF. This diminished seventh suggests
resolution to C. However, Bartók adds a B = at the top of each arpeggio, the
82
As Cohn points out, the two quavers are placed in such a way that the second quaver falls on the
weak triplet quaver of a triplet group, an attack point not sounded in the foregoing coda, and its
activation corresponds with the demise of octatonic writing in the movement. Cohn, ‘PitchTime
Analogies’, 545.
83
342 MICHAEL RUSS
Example 6. Bartók, Sonata for Two Pianos and Percussion, first movement, bars 2715. #
Copyright 1942 Hawkes and Son (London) Ltd. Reproduced by permission of Boosey and Hawkes
Music Publishers Ltd.
G
poco allarg. al
B A 8
271
Piano I più
Piano II più
Timp.
Perc. I
Un poco maestoso ( = 112)

C
274
Piano I
8 8
Piano II
Timp.
Perc. I
Xyl.
Perc. II
flattened leading note seeming to make reference to more archaic forms of cadence.
Bartók also makes reference to the new. In bar 31 he alters the timpani F to F>,
which, as well as being a semitonal neighbour to G in the C major chord of bar 33, is
the symmetrical divider of the C octave. As though to emphasize this point, he
begins the Allegro with reiterated C/F> dyads in the timpani before the onset of the C
major triads.
The powerful final cadence at the end of the first movement (see Example 7) has
many of the qualities of a perfect cadence, not least the dominanttonic move in the
timpani, the rising BC motion in Piano I, and the contrary motion between the
two pianos. However, the leading note is doubled by its fifth, F>, and is
counterpointed by a falling D = C motion in Piano II. D = is itself doubled with
its fifth, A =. This doubling in fifths, the arrival on a final chord without third, and
the emphasis on line over functional harmony take us back to an almost medieval
view of cadence. Sheer force of emphasis is also important; the dynamic marking is
fortissimo. Such cadences seem to contain an element of parody or even travesty:
elements of the tonal cadence are retained, but the cadence is distorted in a way that
appears to mock the original and revert to more primitive forms of musical motion.
Antokoletz provides a logical explanation of the final cadence in terms of interval
cycles. For him the cadence expands out from tritones to fifths, and he demonstrates
how the cycle of fifths spanning D = to F> that is spelled out by the antepenultimate
and penultimate chords lies symmetrically within the final CG fifth.84 However, it
could equally be argued that the expansion of the doubling interval to the fifth in
both hands of both pianos is to allow a linear convergence by semitones onto the
final fifth; the logic of interval cycles seems to undervalue these powerful forces.
The points of articulation just described suggest Bartók is returning to more
primitive forms of cadence, where the power of line is more significant than
harmonic function, and at the same time embedding modernist, symmetrical pitch
relations. Bartók referred to Bach demonstrating the ‘transcendent significance of
counterpoint’, and counterpoint has great significance in this work. But, as was
noted in the discussion of the introduction, despite the logical sequence of entries,
the harmonic structure is very much the result of doubling and aggregation.
Similarly, towards the end of the work, quasi-fugal writing is found at the beginning
of the coda, where a short succession of entries moves round the cycle of fifths. But
the writing in this extensive section is entirely freed from the considerations of
consonance, dissonance and harmonic progression that make Bach’s counterpoint
such a towering achievement. Recognizable harmonic phenomena appear in the coda
from time to time, not least the augmented triads that combine to form hexatonic
scales (bar 378) and the prominent octatonic writing towards the end of the coda
(bar 413). The direction of these passages, however, arises from control of register, of
melodic ‘shape’ (most obviously in the creation of ‘wedge’ shapes in bars 40612), of
84
Antokoletz, ‘Organic Expansion and Classical Structure’, 84.
344 MICHAEL RUSS
Example 7. Bartók, Sonata for Two Pianos and Percussion, first movement, closing cadence.
# Copyright 1942 Hawkes and Son (London) Ltd. Reproduced by permission of Boosey and Hawkes
Music Publishers Ltd.
440
Piano I più
Piano II più
Timp.
Perc. I
poco allarg.
8
442
più C
B
G
F
A
G
D
C
the balancing of ideas that repeat with those that move forward (particularly in the
middle of the coda after bar 377), and of textural thickness (particularly doublings in
thirds and octaves). The sheer velocity of this music and the use of imitation,
suggesting that the two pianos are chasing one another, are also important in creating
coherence, substituting for the presence of an underlying tonal and harmonic
scheme. After setting us up for a complex fugue, Bartók’s parody of this style is short-
lived. Far from having ‘transcendent significance’, Bartók’s use of counterpoint
seems to return it to a more primitive, physical state.
Conclusion
Rosen once described Bartók’s use of sonata form in the first movement of his Fifth
Quartet, written just before the Sonata for Two Pianos and Percussion, as ‘not so
much a new version of sonata form as a brilliant twentieth-century metaphor for
sonata form’.85 One musical domain (form in Bartók’s Fifth Quartet, first
movement) is, therefore, a metaphor for another (sonata form). In other words,
the two forms are analogous but they are fundamentally different in kind. On the
face of it, this is an attractive idea, encapsulating a feeling that Bartók’s use of sonata
is not authentic. However, Rosen’s application of the trope is difficult to sustain
because the domains are insufficiently distinct and the connections between them go
beyond analogy.86
In the Sonata for Two Pianos and Percussion, the specific references to the
‘Waldstein’ and its formal construction (rather than sonata form in general) are
metaphorical; the ‘Waldstein’ stands as a metonym for Beethoven’s progressive use of
sonata. This gives the trope more validity than in Rosen’s usage, since the
‘Waldstein’ can theoretically be replaced by a number of other sonatas. As shown
in Figure 3, the analogies between Bartók’s form and sonata form in general are
separated from the particular relationship between two works. In this, the most
obvious interpretation of this work, the borrowings from the ‘Waldstein’ are there
because that work is a metaphor for progressive form in its time as Bartók’s was in
his, thus confirming Bartók’s words to Moreux.
However, as has been demonstrated, some aspects of this music speak of a less
generous influence, of Bartók at the very least attempting to assert with the spirit of a
modernist his own authority, originality and progressiveness, perhaps wilfully
misreading his predecessors. Bartók, like so many composers of the last century, was
anxious about his place in the Western canon; for Bloom, the route to the canon is
85
Charles Rosen, Sonata Forms (New York and London 1980), 408.
86
The most powerful metaphors are between domains that are quite different in kind, for example: ‘A
tsunami swept through the financial markets.’ Furthermore, as figures of speech, it is possible to
substitute one metaphor for another, for example: ‘The financial markets came off the rails
dramatically.’ The impact of the application of metaphor is retained, but there is a world of
difference between a tidal wave and a rail crash.
346 MICHAEL RUSS
‘Waldstein’
Bartók’s references to
‘Waldstein’ is a METONYM
‘Waldstein’ are
for progressive high Classical
METAPHORS
structure.
for progressive structure
(which he exceeds).
Sonata for Two Pianos High Classical sonata

and Percussion Bartók’s structure is
ANALOGOUS to high
Classical structure.
Figure 3. Metaphor, metonym and analogy in Bartók’s Sonata for Two Pianos and Percussion.
through ‘misreading and overcoming other strong poems’ or, in this case,
compositions.87 In such a reading, rather than embracing Beethoven as a model
and acting as a detached catalyst permitting a synthesis of old and new, Bartók is
heard to employ a series of defences against the older master’s work.
Bloom’s ‘revisionary ratios’ offer a way to draw together observations about the
nature of influence in this work, even if the ratios themselves are slippery and raise as
many questions as they answer.88 For example, in Bloom’s ratio of ‘Clinamen’ the
successor ‘swerves’ from the predecessor: ‘the precursor poem went accurately up to a
certain point, but then should have swerved precisely in the direction the new poem
moves’.89 Images of presence and absence are associated with this ratio, as is the
rhetorical trope of irony. The music of Beethoven is only a partial presence in
Bartók’s Sonata for Two Pianos and Percussion; the later composer often ‘swerves’ in
new directions and the earlier composer disappears. There are a number of instances
of irony, notably Bartók’s misreading of Beethoven’s second subject. Clinamen is
associated with the Freudian psychic defence of reaction formation in which, rather
than moving away from the source of anxiety, the successor begins to copy and to
emulate the predecessor, as Bartók might be heard to do in his Sonata.
Clinamen (a ‘Limitation’ on the Map of Misreading) is paired with ‘Tessera’ (a
‘Representation’), which literally means shards (most often associated with small
pieces of coloured stone or glass used to make a mosaic). Bloom defines Tessera as
‘completion and antithesis [. . .]. A poet antithetically ‘‘completes’’ his precursor, by
so reading the parent poem as to retain its terms but to mean them in another sense,
87
Kevin Korsyn, ‘Towards a New Poetics of Musical Influence’, Music Analysis, 10 (1991), 372 (p.
28).
88
The revisionary ratios are summarized succinctly as a map in Bloom, A Map of Misreading, 84.
89
Bloom, The Anxiety of Influence, 14.
as though the precursor had failed to go far enough.’90 The ratio is associated with
the trope of synecdoche. Bartók’s modernist language clearly takes tonality in new
directions, the old being reflected simply through its neutralized component parts
employed synecdochically for the old order: triads, falling fifths at cadences and so
forth. ‘Shards’ detached from Beethoven’s sonata may be treated antithetically; for
example, the quiet C/E thirds at the beginning of Beethoven’s sonata become
pounding C major triads in the main subject of the first movement (and also at the
beginning of the finale) of Bartók’s, exemplifying the psychic defence of reversal into
the opposite.
One problem with this kind of discussion is that it is equally possible to apply the
two remaining pairs of ratios to the work, and it is difficult to decide whether all
ratios should be in force or whether a single ratio or pair of ratios should
predominate. Bloom’s theory is not an analytical or interpretative method, so the
results can be judged only on the insights they bring. ‘Kenosis’ is associated with
images of fullness and emptiness, and might well be associated with emptying the
Classical tonal language of many of its features. On the other hand, the addition of
an extra piano and a battery of percussion, and the sheer complexity and virtuosity of
the music, seem to signal that Bartók is exaggerating the basic terms of Beethoven’s
sonata and invoking the ratio of ‘Daemonization’ with its associated trope of
hyperbole. The first movement’s introduction, moving as it does from one pole of
the circle of fifths (F>) and the deepest notes of the piano to the opposite pole (C)
and the highest notes at the beginning of the Allegro, are also evidence of this trope.
Bloom associates Daemonization with the psychic defence of repression. His
structural control over Beethovenian materials perhaps helps Bartók to repress
feelings of marginalization.
These observations take apparent similarities between musical works and build
Bloomian arguments around them. Such arguments follow the traditional musical
road of influence manifesting itself in some kind of transformed borrowed material.
But the strongest influences on a composer need not manifest themselves in audible
connections between specific works. As Taruskin points out, Bloom would have
rejected the mere demonstration of similarity as ‘weak’ and, like the identification of
plagiarism, little better than source study.91 For Bloom, ‘criticism is the art of
knowing the hidden roads that go from poem to poem’,92 and the most significant
relationships may not be the most obvious: ‘No strong poem merely alludes to
90
Ibid.
91
Richard Taruskin, ‘Revising Revision’ (review of Korsyn, ‘Towards a New Poetics of Musical
Influence’, and Straus, Remaking the Past), Journal of the American Musicological Society, 46 (1993),
11438 (p. 117).
92
Bloom, The Anxiety of Influence, 96.
348 MICHAEL RUSS
another, and what look like overt allusions and even echoes in strong poems are
disguises for darker relationships.’93 Taruskin summarizes this elegantly:
The nature of Bloom’s theory makes unconsciousness of influence an implicit certainty,

virtually a requirement. As he describes it, an artist’s creative personality is something
feral: remorseless, aggressive, wholly selfish all id, one is tempted to observe. So like the
id it is encased in more acceptable modes of awareness that ‘conceal’ the artist’s true
attitudes and appetites ‘in the codes of civilization’ through repression, sublimation, the
whole panoply of Freudian defensive mechanisms. An artist’s own testimony, it follows, is
the least reliable indicator as to his real creative forebears, the ones whose work he has
perversely reinscribed, except insofar as disavowal, betokening anxiety, may arouse
suspicion (‘for poets rightly idealize their activity; and all poets, weak and strong, agree in
denying any share in the anxiety of influence’).94
Bartók’s words to Moreux and his analysis that reads like a programme note for a
Classical sonata may, like the metaphorical references to a high Classical culture
outside or other than Bartók’s own, deliberately deflect us from more concealed
roads to Bartók from composers other than Beethoven.95 Bloom’s ratio ‘Askesis’
brings together the trope of metaphor, and images of inside/outside. ‘Askesis’ means
severe self-discipline, something reflected in Bartók’s avoidance, both in his note and
in his music, of any ostentatious demonstration of modernism. The extremes found
in works such as the Improvisations, op. 20, or the Third and Fourth String Quartets
are not generally evident here. Furthermore, Bartók’s energy and enthusiasm for
Hungarian folk music has been sublimated into part of his general compositional
language (sublimation is the psychic defence linked with Askesis). What may well be
suggested by this is that the hidden roads go not between Beethoven and Bartók, but
between Bartók and more immediate sources of anxiety: Schoenberg and Stravinsky
(Kárpáti’s second ‘synthesis’). This is not to say that defining Beethoven as an ‘other’
in this work makes Bartók’s relationship with him any less anxious at the very least,
the parody of Beethoven’s second subject speaks against such a view; but Beethoven
(as well as Bach) is also (along with Bartók’s folk-enriched modernism) a defence
against Schoenberg and Stravinsky. Bach and Beethoven become steps on the way
from the origins of music in the folk to Bartók’s own art, which, far from just
synthesizing strands of modernism, fights for a space in its own right. Direct
borrowings from Schoenberg and Stravinsky are not evident, although Les noces is
an obvious precedent, and the motive of a semitone and a third that begins
Bartók’s Sonata for Two Pianos and Percussion in the guise of a reference to Bach is
93
Harold Bloom, ‘The Breaking of Form’, Deconstruction and Criticism, ed. Bloom et al. (New York,
1979), 131 (p. 15).
94
Taruskin, ‘Revising Revision’, 11718.
95
Korsyn, ‘Towards a New Poetics of Musical Influence’, 54. This view echoes Korsyn’s Bloomian
analysis of the anxious influence of Chopin’s Berceuse op. 57 on Brahms’s Romanze op. 18 no. 5, in
which the music of Chopin is a metaphor for otherness in Brahms.
all-pervasive in Schoenberg’s (and Webern’s) atonal music. Indeed, the absence of

specific Stravinskian and Schoenbergian references may be just the point.
ABSTRACT
The first movement of Bartók’s Sonata for Two Pianos and Percussion (1937) has been noted
for its Beethovenian use of sonata form and shares a number of features with the ‘Waldstein’
Sonata. This article examines the relationship between these two composers and these two
works, starting with an examination of Bartók’s use of sonata form. It attempts to establish
whether Bartók borrowed gratefully from his predecessor in an attempt to ‘synthesize’ old
and new, folk and art, or whether he misread Beethoven in the anxious process of carving a
place for himself in the musical canon.
Copyright of Journal of the Royal Musical Association is the property of Routledge and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.

Bartok, Beethoven and The Sonata For Two Pianos and Percussion

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Bartok, Beethoven and The Sonata For Two Pianos and Percussion

Uploaded by

Document Information

Original Title

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

Bartok, Beethoven and The Sonata For Two Pianos and Percussion

Uploaded by

Copyright:

Available Formats

Journal of the Royal Musical Association, Vol. 137, no.

Bartók, Beethoven and the Sonata for

ISSN 0269-0403 print/ISSN 1471-6933 online

Allegro con brio

Un poco più tranquillo ( = 104)

Bartók and Beethoven

Bartók and sonata theory

famously deformational structure composed in the high-middle period of a composer

making the transition essentially a brief (eight-bar) ‘dissolving consequent’ to the

the V7/I progression is a diatonic hexachord (6-Z47). Bartók’s octatonicism links

Moreux, like others after him, subsequently oversimplifies Bartók’s aesthetic to

How Bartók ‘juxtaposed, transformed and eventually synthesized both chromatic

Arriving at a positive assessment of these issues is usually achieved analytically, by

to Bach’s C> minor Fugue) becomes highly significant, because if Bartók is

elsewhere in the work.64 The interruptions in bars 6 and 10 might be seen as

Figure 2. Tonal centres mapped on an extract of a Tonnetz.

inversionally or retrograde-symmetrical, and are thus immobilised’ as the later

In a Beethoven sonata, rhythmic argument is subsidiary to tonal; harmony and

[1] 2 + 2 + 2 + 2 2 + 2 + 2 + 2 [1] [1] 2 + 2 + 2 + 2

Un poco maestoso ( = 112)

Sonata for Two Pianos High Classical sonata

The nature of Bloom’s theory makes unconsciousness of influence an implicit certainty,

all-pervasive in Schoenberg’s (and Webern’s) atonal music. Indeed, the absence of

You might also like