Tesis Sobre Franck

TONALITY, FUNCTIONALITY AND
BEETHOVENIAN FORM IN THE LATE
INSTRUMENTAL WORKS OF CÉSAR FRANCK
Rachel Mary Swindells
A thesis submitted for the degree of Doctor of Philosophy at
the University of Otago, Dunedin, New Zealand
29.9.11
i
Abstract
After Beethoven’s compositional achievements, particularly those of his late
period, nineteenth-century composers drew on his ideas in many different ways.
The harmonic language of music darkened considerably in the nineteenth
century, with strong chromaticism becoming prevalent and Wagner’s influence
being felt throughout the Western Musical World. One way many nineteenth-
century compositions were harmonically organised was with axial tonality, a
system which privileges major-third relationships due to their functional
equivalence. Composers from Beethoven onward were familiar with such
organisational principles, and the instrumental works of César Franck make an
illuminating case study in how axial tonality operates in a deeply chromatic
context. This thesis examines how chromatic harmony might operate within the
functional principles of tonality despite an enriched harmonic language, and what
changes were necessary to formal models as a result. Seminal late-Beethovenian
works are examined and their internal procedures deduced: this is discussed in
relation to immanent and transcendent dominants which operate differently in
each formal prototype. Works of César Franck are then examined, to deduce their
harmonic language and formal procedures. Though some of Franck’s works
display allegiance to Beethoven’s formal models, others have their own peculiar
inner workings. These findings are discussed in relation to axial tonality, the
importance of rhythm, and cyclic form, a technique closely associated with
Franck.
ii
Acknowledgements
I extend my sincere thanks to my supervisor Dr. Graeme Downes, for his support,
patience, and perseverance. Without him, this thesis would not exist. I also
acknowledge Prof. John Drummond for additional supervision. Many thanks also
go to Dr. Elman Poole and the community at Knox College, Dunedin, for financial
support and provision of an excellent work environment by way of the Elman
Poole Fellowship, which I was fortunate to receive during 2009 – 2010.
Thanks also must go to Andrew Wilshere and Katrina Yelavich for help with
translations, and Margaret Finney for proof-reading.
There are many other people who have contributed significantly to this work, in
some way or other, over the past three and a half years; there is not space to
acknowledge them all here. However, I must mention the University of Otago
staff and students, particularly Dr. Alan Davison.
I dedicate this thesis to the memory of my late father, David Swindells, who (with
my mother) taught me to love music.
iii
Table of Contents
Front Matter i
1 César Franck 1
2 Historical Documentation and Theory of Third-Relatives in Music 34
3 The Eroica Symphony and the Problem of the Finale 73
4 Late Beethovenian Models 97
5 Issues of Nineteenth Century Composition, Le chasseur maudit and Psyché 125
6 The Mono-Axial model and the Quintet in F Minor 158
7 String Quartet 191
8 Symphony in D Minor 214
9 Proto-Progressive Tonality and the Violin Sonata 243
10 Conclusions 268
Bibliography 282
iv
List of Figures
Chapter Two
Figure 2.1 Hauptmann’s diatonic key-schema
Figure 2.2 Riemann’s early Tonnetz
Figure 2.3 Downes’ Major Third Axis System
Figure 2.4 Downes’ dominant/subdominant functions
Figure 2.5 Dominant voice-leading efficiency (adapted from Downes)
Figure 2.6 Subdominant voice-leading efficiency (adapted from Downes)
Figure 2.7 Supertonic voice-leading efficiency (adapted from Downes)
Figure 2.8 Cohn’s “Table of Tonal Relations”
Figure 2.9 Cohn’s hyper-hexatonic system
Figure 2.10 The eight third relations from a major tonic as expressed in
Riemann: root-interval system (from Kopp)
Figure 2.11 Harrison – disassembly of a mediant
Figure 2.12 Harrison – disassembly of a diminished seventh chord
Figure 2.13 Pitch-class set [4, 5, 8, 9, 12, 13]
Chapter Three
Figure 3.1 Beethoven, Symphony No. 3/iii (Op. 55), bars 1 – 7
Figure 3.2 Beethoven, Symphony No. 3/iv (Op. 55). Adapted from A. Peter
Brown
Figure 3.3 Schubert. Symphony No. 9 (D 944). Adapted from Downes
Chapter Four
Figure 4.1 Beethoven, String Quartet (Op. 127), I.
Figure 4.2 Beethoven, String Quartet (Op. 127), I, bars 7 – 14.
v
Figure 4.3 Beethoven, String Quartet (Op. 127), II.
Figure 4.4 Beethoven, String Quartet (Op. 127), III.
Figure 4.5 Beethoven, String Quartet (Op. 127), IV.
Figure 4.6 Beethoven, Cello Sonata (Op. 102), Part I
Figure 4.7 Beethoven, Cello Sonata (Op. 102), Part II
Chapter Five
Figure 5.1 Brahms, Symphony No. 2/iii (Op. 73), bars 1 – 14
Figure 5.2 Franck, Le chasseur maudit
Figure 5.3 Franck, Sommeil De Psyché (Sketch 1)
Figure 5.4 Franck, Psyché Enlevée par les Zéphirs (Sketch 2)
Figure 5.5 Franck, Psyché Enlevée par les Zéphirs (Sketch 2) bars 9 – 10
Figure 5.6 Franck, Les Jardins D’éros (Sketch 3)
Figure 5.7 Franck, Les Jardins D’éros (Sketch 3) reh. C+4 – 7
Figure 5.8 Franck, Psyché et éros (Sketch 4)
Figure 5.9 Franck, Sommeil De Psyché (Sketch 1), bars 8 – 9 (strings)
Figure 5.10 Franck, Psyché et éros (Sketch 4), reh. O+3 – 10
Figure 5.11 Franck, Psyché et éros (Sketch 4), reh. S+24 – 29
Chapter Six
Figure 6.1 Beethoven, Hammerklavier Sonata (Op. 106), bars 1 – 5
Figure 6.2 Beethoven, Hammerklavier Sonata (Op.106), adapted from
Downes
Figure 6.3 Brahms, Piano Sonata (Op. 1), bars 1 – 4
Figure 6.4 Brahms, Piano Sonata (Op. 1)
Figure 6.5 Mendelssohn, Piano Sonata (Op. 106), bars 1 – 5
Figure 6.6 Franck, Final (Op. 21), bars 28 – 32
Figure 6.7 Franck, Final (Op. 21)

vi
Figure 6.8 Franck, Piano Quintet, I
Figure 6.9 Franck, Piano Quintet, I, bars 135 – 144
Figure 6.10 Franck, Piano Quintet, II
Figure 6.11 Franck, Piano Quintet, II, bars 9 – 12
Figure 6.12 Franck, Piano Quintet, III
Figure 6.13 Liszt, La Lugubre Gondola No. 1, bars 12 – 15
Chapter Seven
Figure 7.1 Franck, String Quartet I
Figure 7.2 Franck, String Quartet, I, reh. N+4 – 7
Figure 7.3 Franck, String Quartet, I, reh. C+20 – 21; German sixth acting as
a dominant
Figure 7.4 Franck, String Quartet, II
Figure 7.5 Franck, String Quartet, III
Figure 7.6 Beethoven, String Quartet Op. 130 (Cavatina) bars 38 – 39
Figure 7.7 Franck, String Quartet, III, bars 186 – 187
Figure 7.8 Franck, String Quartet, IV
Figure 7.9 Comparison of recordings
Chapter Eight
Figure 8.1 Franck, Symphony in D minor, I
Figure 8.2 Franck, Symphony in D minor, II
Figure 8.3 Franck, Symphony in D minor, III
Figure 8.4 Franck, Symphony in D minor, III, reh. K+9 – 12
Chapter Nine
Figure 9.1 Franck, Violin Sonata, I
Figure 9.2 Franck, Violin Sonata, II

vii
Figure 9.3 Franck, Violin Sonata, III, bars 12 – 13
Figure 9.4 Franck, Violin Sonata, III
Figure 9.5 Franck, Violin Sonata, III, rhythmic analysis
Figure 9.6 Franck, Violin Sonata, III, bars 65 – 68
Figure 9.7 Beethoven, String Quartet Op. 59 No. 1/1, bars 1 – 4
Figure 9.8 Franck, Violin Sonata, IV, bars 1 – 4
Figure 9.9 Franck, Violin Sonata, IV
Figure 9.10 Franck, Violin Sonata, IV, rhythmic analysis
Figure 9.11 Franck, Violin Sonata, main themes
Chapter Ten
Figure 10.1 Franck, thematic examples
viii
Chapter One
César Franck
While César Franck’s music is the primary focus of this study, his background is
also important. By surveying his upbringing, significant compositions, musical
influences and personality, subtle threads emerge; our perception of Franck,
composer and artist, becomes clearer. This chapter is built around a chronology of
Franck’s life and work before an analysis of his persona. It draws from the
important twentieth-century biographers and includes a brief survey of Franck’s
significant works, justifying which of these I have selected for closer analysis. The
review concludes by identifying issues of Franck reception – including historical
conditions around 1870 – and discussing the issue of musical influences.
Franck Biography
César Auguste Jean Guillaume Hubert Franck – a “misfortunate” name as Liszt
described it1 – was born on December 10th 1822 in Liège, Belgium. He had one
younger brother, Joseph, a violinist. As Franck showed propensity for music from
an early age, his father, Nicolas Joseph, decided that César would have a career as
a virtuoso pianist. After study and prizes from the Liège Conservatoire, his father
organised a Parisian concert debut in 1835. The concert was not a success, but the
family had by then moved to Paris, where his father persevered with the virtuoso
dream. Though César was at first refused entry to the Paris Conservatoire because
of his Belgian nationality (Liszt had also been refused on the basis of his
Hungarian nationality a decade earlier), 2

Franck studied harmony and
1Adrian Williams, ed. Franz Liszt: Selected Letters (Oxford: Oxford University Press, 1998), 227.
2John Trevitt and Joël Marie Fauquet, "César Franck," in Grove Music Online. Oxford Music
Online.
1
counterpoint privately with Reicha, who had studied with Beethoven and taught
Liszt and Berlioz, and is a possible catalyst of transmission between these
composers.3 At the age of 14, in April 1837, Franck met Liszt for the first time; the
occasion was a concert where César played a Fantasy by Hummel. 4 This
relationship would prove influential for the young Franck, who – as discussed
below – received feedback on his early compositions from Liszt. After entering the
Paris Conservatoire in October of that year after his father secured naturalisation
papers, Franck won prizes in piano, organ and counterpoint; yet in 1842 his father
withdrew him before he could contest the coveted Prix de Rome, organising a
Belgian concert tour so that Franck could concentrate on his virtuoso career
instead.5
When not on tour during the next few years Franck stayed with his family in
Liége.6 He began composing and in 1843 his first compositions, Trios concertants
Op. 1 for Piano, Violin and Cello, were published. Liszt, Meyerbeer, Halévy and
Chopin purchased copies, among others. Franck and Liszt had met a second time –
probably in Brussels, in 18427 – and it was on Liszt’s advice that Franck exchanged
the finales of his Trio Op.1 No.3 and his Trio Op.2 No.4.8
Sadly enough for his father, Franck’s virtuoso career never eventuated: his final
Parisian concert tour in 1844 suffered from a poor critical reception. One critic
3 Davies points out that Franck was only 13 and 14 when he studied with Reicha, and warns
us about placing a disproportionate amount of weight on the relationship. Laurence Davies,
César Franck and His Circle (Boston: Houghton Mifflin Co. , 1970), 48. However, we know that
composers can be influenced substantially by their early teachers, and J.C. Bach’s influence on
Mozart is surely perceptible. Cliff Eisen et al., "Mozart," in Grove Music Online. Oxford Music
Online. (2010). Though it is difficult to establish either way, Reicha certainly remains a
possible connection between Beethoven and Franck.
4 Williams, ed. Franz Liszt: Selected Letters, 966.
5 Trevitt and Fauquet, "César Franck," 178.
6 Léon Vallas, César Franck (George G. Harrap & Co. Ltd., 1951), 47.
7 Ibid.
8 Trevitt and Fauquet, "César Franck."
2
compared Franck to Vaucanson’s mechanical puppet, one of the city’s attractions;
others complained about the “standard of César’s execution.” 9 Nevertheless,
Franck returned to Paris later that year and by 1846 was making a living through
teaching privately. He continued to compose, and after the Trios came Franck’s
first large-scale work, the biblical oratorio Ruth, which premièred in 1846. The
oratorio brings about further evidence of an association with Liszt, in the form of a
letter from Liszt to Ary Scheffer supporting Ruth. “You have had confirmation . . .
of the opinion I gave you of [Franck’s] oratorio Ruth”, writes Liszt, “and the great
master [Meyerbeer’s] sincere approbation seems to me to make the matter
conclusive. . . . among the young people who sweat blood and water to set down
some ideas on cheap music paper, I know not three who are his equal.” 10 Yet
despite Liszt’s support and the recent success of Félicien David’s Le désert – an
oratorio with similar subject material – Franck’s Ruth received a poor reception
until its revision many years later. This put further stress on Franck’s relationship
with his father, and Franck formally left his parents’ home in the summer of 1846. 11
In 1847 he obtained an organist’s post at Notre Dame de Lorette: the menial role of
organiste accompagnateur or choir accompanist.
In 1848 Franck married one of his pupils, and in 1851 he became organist at St
Jean-St François, which possessed an organ built by Cavaillé-Coll. In 1858 he
bettered this by his appointment to Ste Clotilde, which possessed a fine new organ,
again from the Cavaillé-Coll company. It was on this organ that Liszt heard Franck
play in 1866. On this occasion Franck played from his Six Pièces for organ, a set that
represents one of the most significant contributions to the French organ repertory
in over a century. “How could I ever forget the composer of those Trios?” Liszt
reputedly asked, to which Franck replied: “I fancy I have done better things
9 Davies, César Franck and His Circle, 52.

10 Williams, ed. Franz Liszt: Selected Letters, 227. Williams’ source is Rivista musicale italiana, 39
(Turin, 1932).
3
since”. 12 From these anecdotes it seems that Franck was merely one of many
nineteenth-century musicians whom Liszt helped.13 Yet this conclusion would not
necessarily negate the formal similarities in the works of these two men, which I
will discuss below.
All of Franck’s organ works, including the Six Pièces of this period, are influenced
by the capabilities of Cavaillé-Coll’s instruments. Franck worked stoically at Ste
Clotilde for the next decade, continuing to teach and play the organ, with little
time for composition. Trevitt writes: “It was during this creatively fallow period
that he was unwittingly laying the foundations of a remarkable phenomenon of
19th-century [sic] French culture: the cluster of pupil-disciples”14, though it should
be noted that Liszt and Wagner attracted similar “clusters”. Franck’s pupils
included Duparc and Coquard, to whom he taught piano at the Jesuit College in
the rue Vaurigard. In 1870, however, the Franco-Prussian war disrupted life;
during the war and the ensuing Commune, the Conservatoire closed and
advanced music education in Paris ground to a halt. The psychological effects of
the Franco-Prussian war and the Commune had a profound effect on French
musical life in the coming decades. When the Conservatoire reopened in 1872
François Benoist retired from his post of organ professor. Franck was appointed to
succeed him, and this necessitated applying for French citizenship.15 The Société
Nationale de Musique formed after the war, with Franck becoming President in
1880. 16 Aside from this, Franck’s organ class throughout the 1870s and 1880s
gained the reputation of being composition-focused. His pupils were unusually
devoted, forming the “band à Franck.” Vincent d’Indy, who joined the organ class

13 Williams, ed. Franz Liszt: Selected Letters, 966.
15 Rollin Smith, Playing the Organ Works of César Franck (New York: Pendragon Press, 1997), 29.
4
in 1872 and became Franck’s most devoted disciple, led this group.17 It was only in
the 1870’s that Franck began to hit his compositional stride; the best loved and well
known Franck compositions are those that date between 1875 and 1890. Compared
to the output from his earlier years the list is surprisingly long:
Rédemption (oratorio – revised – symphonic poem, several versions) 1874;
Les béatitudes (oratorio) 1869 – 1879;
Les Eolides (symphonic poem) 1875-6;
Trois Pièces (organ) 1878;
Quintette (piano quintet) 1879;
Rebecca (oratorio) 1880-1;
Le chasseur maudit (symphonic poem) 1882;
Les Djinns (symphonic poem) 1884;
Prélude, choral et fugue (piano) 1884;
Hulda (opera) 1879-85;
Variations symphoniques (piano/orchestra) 1885;
Sonate (violin and piano) 1886;
Prélude, aria et final (piano) 1887;
Psyché (symphonic poem with chorus) 1887-8;
Symphonie, 1886-8;
Quatuor (string quartet), 1889;
Trois chorals (organ), 1890; and
Ghiselle (opera) 1888-90.
Franck died in November 1890 from complications resulting from pleurisy. 18
The following larger segments of this chapter present the challenges of researching
César Franck. These include: Franck’s personality and the biased nature of his first
17 Ibid. Accessed 23.7.09

18 Vallas, César Franck, 234.
5
biographers; issues of Franck’s nationality and what implications these may have,
including the effect of the Franco-Prussian war on the French musical psyche; and
an overview and hypotheses of Franck’s most important musical influences.
Franck’s Biographers
This section examines two important themes in Franck biography: his personality,
and the changing reception of his compositions.
Franck’s personality: an “aura of sanctity?”
Vincent d’Indy, a pupil from Franck’s organ class at the Conservatoire, wrote the
first account of Franck’s life and work in his 1906 biography César Franck19. D’Indy
provides background to Franck’s childhood and upbringing, and includes brief
analyses of his important works. However, the biography also contains a number
of inaccuracies, one of the most significant being d’Indy’s claim that Franck began
composing his Symphony before he heard Saint-Saëns’ Symphony in C minor.20 By
claiming this, d’Indy was in a strong position to argue Franck’s originality in the
symphonic genre and ensure Franck’s Symphony compared favourably with Saint-
Saëns’. Recent research suggests, however, that Franck had not even begun his
own Symphony before hearing Saint-Saëns’. 21 However, d’Indy is certainly not
alone in his adoration of Franck; indeed, this is a theme that runs through the
writings of all Franck’s pupils.
Franck’s other pupils discussed their teacher’s life and work in articles and
correspondence. 22 An excerpt translated from Guy Ropartz’s book “A propos de
19 Vincent d'Indy, César Franck: A Translation from the French of Vincent d'Indy with an
Introduction by Rosa Newmarch (London: John Lane the Bodley Head, 1909).
20 Ibid., 173.
21 A. Peter Brown and Brian Hart, The Symphonic Repertoire Volume III Part B: The European
Symphony from ca. 1800 to ca. 1930: Great Britain, Russia, and France (Bloomington and
Indianapolis Indiana University Press, 2008), 593.
22 Ibid., 590.
6
quelques symphonies modernes” indicates how devoted Franck’s pupils were to
their Master:
Among the multitude of contemporary composers, M. César Franck appears to be a
man from another time. They are skeptics, he is a believer; they are occupied with
themselves, he works in silence; they seek glory, he awaits it; others make
concessions and compromises – the basest even, the shameful steps towards easy
fame – while he works quietly and firmly at what he has to do, without hesitation,
without weakness, without calculation; he has given us the most beautiful example
of artistic probity one could encounter.23
If we hold this description to be (in the least partially) accurate, it suggests that
Franck considered himself a serious composer of serious music. This may seem
unimportant, but when contemplating Franck’s influences it may be timely to
consider that Franck may have wished to contribute to or emulate works within
the “canon”. Discussing Franck’s Symphony, Ropartz considers that it contains
“true philosophical thoughts.”24 Louis Vierne, a blind pupil, wrote about Franck
with great warmth in his Mes Souvenirs:25

Of Franck I had made almost a religion, combining passionate admiration, filial
affection and profound respect. I experienced with intense joy, with which,
however, there was also mingled a certain mysterious awe, the almost magnetic
fascination which emanated from that man at the same time, so simple, so natural,
so truly good.
In 1905 Albert Mahaut, another of Franck’s blind pupils, published a pamphlet:
César Franck, Professeur d’Orgue.26 He writes about how he came to be in Franck’s
organ class, the wonderful warm resonance of Franck’s voice, and confirms much
of what other pupils say of Franck’s personality: “The Maître was remarkably kind
23 Guy Ropartz, “A propos de quelques symphonies modernes,” Notations artistiques, Paris:

Lemerre, 1891, 181, quoted in ibid., 592.
24 Ibid., 610.
25 Louis Vierne, Mes Souvenirs, (Paris: Les Amis de l’Orgue, 1970), Translated by Esther E.
Jones as “Memoirs of Louis Vierne: His Life and Contacts with Famous Men,” The Diapason,
(Sept. 1938 – Sept. 1939), quoted in Rollin Smith, Toward an Authentic Interpretation of the
Organ Works of César Franck (New York: Pendragon Press, 1983), 42.
26 Translated and reprinted in ibid., 154.
7
toward all of his students” and, later, “His combined ardor and serenity overcame
us and left an indelible imprint upon us.”27 Another pupil, Gabriel Pierné, in an
interview with Pierre de Lapommeraye, describes Franck as being “very serious –
he didn’t have a sense of humour.”28 Yet Pierné also describes Franck as “truly a
beautiful soul, with an inexhaustible goodness, and a bit of naïveté.”29
Charles Tournimere was 19 years old when he studied with Franck in 1889. He
subsequently published a biography of his teacher in 1930. Tournimere was an
organist, and consequently focuses on analysing sections on Franck’s organ works.
Tournimere uses over-ardent language, as does d’Indy in his biography; the
exposition of Franck’s Fantasie Op. 16 is “white as snow” and Franck becomes the
“Fra Angelico of Sound.” 30 Tournimere’s volume is useful for organists as it
contains performance instructions, but like other pupils his concern to depict
Franck as a saintly figure betrays the lack of objectivity in his writing.
So why were Franck’s pupils so devoted? John Trevitt suggests it was because of
Franck’s seriousness, suggesting this quality “stood in marked contrast to the
superficiality of the Opéra-dominated establishment [the Conservatoire.]” 31
Perhaps it was this atmosphere that motivated Duparc when he wrote about
Franck’s Symphony: “I would even say that a success at the Conservatoire would
have disturbed me a bit – I will always refuse to admit that those people have the
right to understand the things we love at first hearing.”32
27 Ibid., 155.
28 Ibid., 159.
29 Ibid., 161.
30 Ibid., 78 – 79.
32 “Lettres d’Henri Duparc à Ernest Chausson,” Revue de musicology, December 1956, 130,
quoted in Joël-Marie Fauquet, César Franck, Paris: Fayard, 1999, quoted in Brown and Hart,
The Symphonic Repertoire Volume III Part B: The European Symphony from ca. 1800 to ca. 1930:
Great Britain, Russia, and France, 592.
8
Franck comes across as serious indeed when compared to contemporary organists.
Fenner Douglass considers Franck’s disposition to be “diametrically opposed” to
that of famed Parisian organist Lefébure-Wély, and sees this reflected in the styles
of music in which each organist preferred to improvise and compose. Lefébure-
Wély became famous for his “thunderstorm” improvisations and “showpieces”
that proved popular with the public; Franck’s improvisations and compositions
reflected a more serious, measured style, and in many reviews of his concerts he is
described as “severe”, as well as being reprimanded for playing too long. 33
Anecdotally at least, the musical styles seem to mirror each organist’s personality.
This seems to have contributed to the reason that Franck was not awarded several
prestigious organist posts in the 1850s and 60s,34 the most significant of these being
the position at St-Sulpice left vacant by Lefébure-Wély’s death in 1869. With
Franck and Widor being the two standout candidates, the middle-aged Franck was
overlooked in favor of the young virtuoso Widor. Douglass describes this as “the
most important organ position in France”, and although Franck succeeded Benoist
at the Conservatoire a few years later, he was bitterly disappointed at missing out
on the St-Sulpice post.
Norman Demuth provides short analyses of all of Franck’s main works, as well as
biography and reception details, in his 1949 book César Franck. He devotes a
chapter to examining the “aura of sanctity” surrounding Franck’s contemporary
perception, declaring that “it [was] d’Indy, whose book so far has been the only
one accessible in English, who has given us this [aura of sanctity] view and,
knowing no better, we have taken what he said as gospel.” 35 But Demuth also
explores why Franck’s circle promulgated such ideas, suggesting it was due to
33 John Caldwell, et al, "Keyboard Music," in Grove Music Online. Oxford Music Online, and
Smith, Toward an Authentic Interpretation of the Organ Works of César Franck, 15.
34 Fenner Douglass, Cavaillé-Coll and the French Romantic Tradition (New Haven and London:
Yale University Press, 1999), 108.

35 Norman Demuth, César Franck (London: Dennis Dobson Ltd., 1949), 205 – 08.
9
Franck’s personality. From the accounts of his pupils detailed above, we know
Franck was a friendly, happy, gentle man; deeply religious and devoted to his
pupils.36 Further anecdotes are easy to find: Ropartz stated “I never heard him say
anything unkind about anyone”, whilst Debussy referred to him as “this intensely
good man”. 37 Demuth agrees with Trevitt, arguing that such accounts became
legendary because such personality traits contrasted strongly with conventions at
that time, as “Franck was surrounded with flippancies and artificiality [at the
Conservatoire].” However, Demuth’s assessment is not difficult to agree with.
Given what we know of the atmosphere cultivated by the Paris Conservatoire,
Franck must have seemed saintly in comparison; 38 Trevitt notes that it was an:
“Opéra-dominated establishment” which contrasted strikingly with Franck’s
serious personality.
Other writers also discuss Franck’s nature and references to him as a “Pater
Seraphicus.” In the small 1932 volume French Piano Music by Alfred Cortot, the
author addresses both the “Franck Myth” and the subject of mysticism:
Not that I subscribe for my part to the favourite legend of a mystic César Franck, a
sort of Pater Seraphicus lost in dreams of heaven, illumined with the ecstatic
contemplative fervour which removed him from the pricks and realities of life. The
nobility of Franck’s life, reflected perfectly in the beauty of his work, was precisely
that he neither shook off reality nor despised the affairs of humanity, but rose
above them.39
Writing later in the century, Lawrence Davies refutes it further:
The obstacles which stand in the way of portraying César Franck the man are
already somewhat notorious. They refer not only to his retiring disposition, but to
the unduly apologetic dispositions of his admirers. It is a fact that those who have
36 Ibid., 205.
37 Quoted in ibid.
38 Ibid., 209.
39 Alfred Cortot, French Piano Music, trans. Hilda Andrews (London: Oxford University Press,
1932), 49 – 50.
10
written about him most vividly also happen to be those least likely to have resorted
to impartiality.40
Davies is also pragmatic about other issues entangled with Franck biography, as
“in being obliged to rebut the slanders of the composer’s enemies, [biographers]
also have to be on their guard against over-dramatising a man who in essence was
a singularly un-dramatic figure.”41 Perhaps symptomatic of this, Harvey Grace has
observed that Franck’s reputation suffered – in England at least – because there
was a “reaction against the uncritical adulation of his devotees in France.” 42 It
seems this reaction has taken much of the century to dissipate – though whether it
has completely is up for contention.
Many of Franck’s early biographers are problematic as (secondary) sources as their
emotionalised language indeed marks them as less than objective. Yet, as Davies
mentions, they are the people who have written about him most vividly, and for
many biographical aspects they are our best sources. Perhaps in the epoch his
pupils needed a heroic figure, and Franck’s supportive nature lent him to this role.
We know from contemporary accounts that although Franck was usually an
understated man, he showed great enthusiasm when discussing music.43 Though it
may appear that Franck has been “beatified” by his pupils, the writers of his
history, he must have been a special man to invoke such a following. (Though he
possessed a very different personality to Franck, Liszt also attracted a large cluster
of pupil-disciples.44) We can realise, from these accounts, that Franck was a warm,
gentle and devoted person, and, with this measured analysis, consider what effect
this may have had on his musical style and reception.

41 Ibid., 5.
42 Harvey Grace, "Franck, César Auguste," in Grove's Dictionary of Music and Musicians, ed. H.
C. Colles (New York: The MacMillan Company, 1927), 303.

43 Smith, Toward an Authentic Interpretation of the Organ Works of César Franck, 154.
44 Arthur Hedley, "Liszt the Pianist and Teacher," in Franz Liszt: The Man and His Music, ed.
Alan Walker (London: Barrie & Jenkins 1970), 32.

11
Style and reception
With a few exceptions, Franck’s works received a poor reception during his
lifetime, although for differing reasons. Ruth, Franck’s first oratorio and large-scale
work, was compared unfavourably (and unfairly) to Rossini’s operas45; Liszt stated
that the Prélude, Choral et Fugue was “uncouth and tiresome to play”; 46 and
similarly, Prélude Aria et Final was described by one reviewer as “long and
tedious.”47 Many of Franck’s works suffered from disastrous premières, including
that of the Variations Symphoniques, where the piano and orchestra were completely
out of sync for the entire finale,48 but perhaps worse was the premiere of Franck’s
Piano Quintet, in which the pianist Saint-Saëns sight-read the work then stormed
offstage when Franck attempted to dedicate the score to him. In some of these
instances it is fair to say Franck was unlucky; in others, the reviews reflect
attributes we recognise in Franck’s music today.
The String Quartet, however, was one of the few of Franck’s works that had a
successful premiere, greeted as it was with “unmixed applause.” 49 The Violin
Sonata gradually grew in popularity, promoted by its dedicatee Eugene Ysaÿe, and
continues to be one of Franck’s most popular works today. Furthermore, Franck’s
organ works were largely well received in his lifetime, though their audience may
have differed slightly from that for Franck’s later chamber works and orchestral
pieces.50 Brian Hart describes Franck’s Symphony as “unquestionably the most
controversial French symphony composed after Berlioz’s time.” 51 His view is that
some of this controversy was due to Franck’s heritage, which I discuss more fully
below.

46 Ibid., 221.
47 Ibid., 222.
48 Ibid., 224.
49 Ibid., 235.
50 Smith, Playing the Organ Works of César Franck, 37.
51 Brown and Hart, The Symphonic Repertoire Volume III Part B: The European Symphony from ca.
1800 to ca. 1930: Great Britain, Russia, and France, 590.

12
Hart discusses the Symphony’s reception, with numerous references to
contemporary reviews, and illustrates the marked differences in opinion between
Franck’s sympathisers and critics. Though he concedes that Franck’s Symphony is
not faultless, he surmises that some of the complaints about the Symphony arise
from the way it is performed: “elements that appear in isolation as structural or
stylistic flaws cohere to powerful effect . . . Franck is often blamed for the sins of
the conductor.”52 On the whole, however, Franck received little public acclaim for
the majority of his works. Whilst some works grew in acclaim after Franck’s death,
others have almost fallen out of the repertoire entirely. Why this has happened is
one question this thesis hopes to answer, but we must first survey Franck’s
twentieth century reception.
In 1923 Harvey Grace published several articles in the journal Musical Times
focusing on Franck’s organ works. 53 In the eloquent narrative style of the day,
Grace analyses all of Franck’s organ output, and is critical about Franck’s rich
chromaticism and “mere meandering”, particularly in Franck’s Trois Chorals. 54
Conversely, he considers the Six Pièces to be “direct and diatonic” – “better” is
inferred – which is not the general consensus nowadays.55 Grace credits Wagner
with influencing Franck’s chromatic idiom, and Beethoven is also considered an
influence in matters of form: “The Chorals are merely fine examples of the large
variation form of Beethoven’s last period . . . in regard to form and methods of
development [Franck] looked rather to Bonn than to Leipzig” (a reference to
52 Ibid., 611.
53 These articles were published collectively in 1948 as Harvey Grace, The Organ Works of César
Franck (London: Novello and Company Ltd., 1948).
54 Ibid., 18.
55 Trevitt considers the Chorals “immortal” in the most recent Oxford Music Online article,
and they are among Franck’s most played organ works for recitals, particularly the Third
Choral. The Six Pièces, in comparison, are not performed nearly so regularly.
13
Beethoven).56 Though limited to the organ works, Grace’s views give us some idea
of the reception of Franck’s organ works in early twentieth-century England.
However an entry in Grove’s Dictionary – 3rd Edition (1927) demonstrates how
Franck’s reception has changed over the years. Harvey Grace is the third edition
author,57 and writes that in the piano works “we are . . . frequently reminded of
Beethoven, Wagner [and] Liszt” 58 (curious, as Grace had not earlier considered
Liszt an important influence on Franck’s Organ works). The form of Franck’s
Symphony is described as a “monologue rather than the sonata” 59 (which is
considered to the Symphony’s detriment). Grace’s summation of Franck as a man
of contradictions is one of the best in the literature:

. . . a Belgian who founded a French school, yet whose own music is Belgian and
German rather than French; who was a classicist in form and an impressionist in
harmony; and whose whole career was a curious blend of drudgery and idealism,
hack-work and inspiration. 60
In his 1951 book French Music, Martin Cooper is critical of much of Franck’s
output.61 Cooper describes Franck’s main weakness as “his failure to subject his
works to a sufficiently high critical standard.” 62 He also describes a “lack of
emotional restraint” in the Quintet, which Cooper considers removes the work
from the French tradition, thus: “betraying the Germanic affinities of his genius.” 63
Cooper is equally critical of the chromaticism characteristic of much of Franck’s
output, reserving his praise for Les Eolides, the Violin Sonata and the String
56 Grace, The Organ Works of César Franck, 28.

57 ———, "Franck, César Auguste."
58 Ibid., 298.
59 Ibid., 299.
60 Ibid., 303, fn.
61 Martin Cooper, French Music: From the Death of Berlioz to the Death of Fauré (Oxford:
University Press, 1951).

62 Ibid., 29.
63 Ibid., 31.
14
Quartet. Cooper sees these works as imbued with the spirits of Liszt and
Beethoven. Despite his critique, Cooper considers that Franck has had quite an
effect on the development of French music:

[Franck’s] mature passion for Beethoven led him to conceive of all music as
primarily an expression of soul, a philosophy of life, in his case half-Christian and
half-romantic but wholly emotional. It is in the light of these extreme opposites
that we must view . . . the subsequent reaction against his music, which had such a
large place in determining the direction of French musical development.64
Cooper considers this effect largely due to Franck’s pupils and the compositional
paths they trod following Franck’s death – as well as, perhaps, their fervent
devotion to their teacher and mentor: “In all [of Franck’s pupils] there lived on,
under varying forms, something of Franck’s idealism, something of his careful,
loving and slightly gauche craftsmanship.”65
Davies, however, reminds us that much of the criticism that Franck endured in his
life was not without reason, and urges us to not forget it:
The early testimonials he received – like the reviews of his first concerts – should
not be dismissed as so much insensitive verbiage. They provide a valuable clue to
some of the composer’s cardinal weaknesses, and tell us, quite as clearly as any
sociological analysis, why he met with such stiff opposition. Throughout his life
Franck was subjected to moderate – and frequently immoderate – criticism, not just
in his capacity as an executant, but in everything that he did. That some of this was
motivated by spite is obvious. But it would be disingenuous to imply that a fair
proportion did not reflect real deficiencies on his part.66
Franck seems to have been ambivalent to this criticism. Accounts of the first
performance of Franck’s Symphony report that he seemed unperturbed by the
64 Ibid., 47.
65 Ibid., 55.
15
reception – his pupils regarded it to be poor – instead commenting on how pleased
he was with the sound of the work.67
Few people have written about Franck without agenda: to sanctify him or to berate
the sanctifiers. It is easy to put too much weight on biographical aspects of a
composer, but it is the aim of this chapter to set the scene for the later analysis
rather than provide the basis of it. For now, we might explain the polarity of views
regarding Franck and his works by the wider context in which he lived and
worked.
It is now pertinent to examine the issue of Franck’s nationality, and how it might
inform arguments within this thesis.
Franck’s Position in French Music
The importance of nationality depends on how much weight we place on Franck as
a defender of the French tradition, and it is a thorny issue. Franck did spend
almost his entire working life in Paris, and applied for French nationality when he
was appointed professor at the Conservatoire. Rollin Smith describes him as a
“Parisian musician in that, from the age of sixteen, he lived in Paris. But he was not
a French musician and this has always been a source of annoyance to the natives.”68
“Is it really French,” Jean-Aubry asked, “this mysticism, this readiness to take
everything seriously?”69 Yet the circle of pupils he attracted later in life represents
some of the strongest defenders and representatives of the French tradition. Even if
we do consider Franck a more “Germanic” influence than French, Lawrence
Davies is correct in observing that “the dominance of German culture was hardly
69 Jean-Aubry, quoted in Davies, César Franck and His Circle, 205.
16
something that resulted from the work of a few accidental expatriates.” 70
Ultimately, Franck was from a mixed background and drew a great deal on great
Teutonic composers; yet he worked in Paris for nearly all of his life, influencing an
entire generation of French composers. Disagreements regarding his prime
nationality continue to this day, and become even more important when
considering the Franco-Prussian war and the Commune.
Few Parisians were unaffected by the War. Many of Franck’s pupils played a part
in defending Paris, and Franck himself volunteered, despite being almost fifty
years old.71 Yet beyond the physical effects, the impact the War and the Commune
had on the French cultural spirit was immeasurable and dramatic. “Its impact was
so drastic and purifying,” wrote Davies, “as to transform the entire fabric of French
culture, proposing a new set of goals for both society and the artist.” 72 Richard
Anthony Leonard reiterates by stating: “There is no mystery about the after-effect
of the Franco-Prussian War of 1870. Following that humiliating defeat the entire
French nation underwent a catharsis. . . [and] Revolt against the existing order was
nowhere more strongly motivated than in the arts.”73
It is in light of these historical conditions that we must consider Franck’s
nationality; born in Liége to a German mother and a father from Gemmenich, 74 a
town in eastern Belgium close to the modern border with Germany and the
Netherlands.75 At the time, the Prince-Bishopric of Liege was part of the United
Kingdom of the Netherlands, formed by the Congress of Vienna in 1815. In August
1830 the Belgian Revolution occurred, and in July 1831 Belgium became an
independent nation. From these changing political arrangements, Franck was born
70 Ibid.
71 Ibid., 116.
72 Ibid., 117.
73 Ibid., 119.
17
Dutch, became Belgian, and later, in 1871, applied for French citizenship. French
was his first language, but he continued to speak German to his mother until she
died. 76 This would not necessarily be important if Franck had lived in the modern
European Union, but in the period 1870 to 1890 it was important indeed because of
the intense nationalism and rivalry between France and Germany.
The Société Nationale formed in February 1871, one month after the armistice
ending the Franco-Prussian War.77 Its aim was to: “encourage the composition and
performance of ‘serious’ French music.”78 Franck attended the first meeting, as did
Saint-Saëns, Fauré, Bussine, Castillon and Dubois, among others.79 The founding
charter promotes the composition of works that “reveal elevated and artistic
aspirations,” and the revival of concert music in France would indeed owe much to
the Société. 80 There exist mixed views, however, on whether this promotion of
French music was actually anti-German. One Saint-Saëns biographer describes the
Société as “a stick with which to beat what [Saint-Saëns] saw as the continuing
German influence on French musical life”, 81 but this position seems to be
overstated and contrasts strongly with the views of other members of the group
who idolised Wagner, Liszt and Schumann.82 At the same time, outside the realms
of the Société Nationale, there existed a small but “highly vocal group of musical
conservatives” who proclaimed that “to detest Prussia was to love France.” 83 They

77 Michael Strasser, "The Société Nationale and Its Adversaries: The Musical Politics of
'L'invasion Germanique' in the 1870's," 19th-Century Music 24, no. 3 (2001): 225.
78 Ibid.
79 Jean Gallois, "Chausson, Ernest." Grove Music Online. Oxford Music Online.

81 Strasser, "The Société Nationale and Its Adversaries: The Musical Politics of 'L'invasion
Germanique' in the 1870's," 226.

82 Ibid.
83 Ibid., 241.
18
professed that this new “music of the future”84 – which influenced most Société
members – was “as threatening to France’s artistic heritage as German
industrialisation was to her security.”85 These conservative and nationalist ideas
were often professed by Gaston Escudier in his journal L’Art musical; in 1872, when
the Société was beginning to attract attention, a scathing front page article labeled
the Société a “fervent cult for Richard Wagner and for neo-German music,” and
more viciously suggested that if a member “commits, even by chance, the error of
writing a truly melodic phrase, [he] will be forthwith expelled. . . . Death to
melody! Death to genius!”86 Such attacks persisted throughout the decade; in 1875
a critic reported on a recent Société concert: “We were saturated with music more
or less of the future; that is to say; nebulous chromaticism was feted with
idolatry.”87
However, the Société viewed the “music of the future” as positive and forward-
looking. In an 1876 review of Wagner’s Ring at Bayreuth, Saint-Saëns praised the
immensity of Wagner’s accomplishment and noted that it received an enthusiastic
reception. He then labeled those who protested the music had “no melody” or was
“difficult to understand” as having their heads buried in the sand, and stated that:
“the advance of art was impossible to stop and that Wagner’s operas were being
played everywhere, not only in Germany, but in Russia, England, Italy, and
America.”88 He also explained later that it was the duty of French composers to
“study new techniques”; despite Wagner’s bad press in same quarters, the Société
composers were intrigued by his innovations, as well as by the music of Schumann
84 This phrase is used to describe the music – usually new and chromatic – of Liszt and
Schumann, usually with negative overtones. It was also applied to Wagner’s music by his
opponents, though Wagner tried to distance himself from it.
85 Strasser, "The Société Nationale and Its Adversaries: The Musical Politics of 'L'invasion
Germanique' in the 1870's," 241.

86 Ibid., 246.
87 Ibid., 248.
88 Ibid., 242.
19
and Liszt.89 The Société members professed they did not wish to exclude the music
of other composers, but rather promote a more serious musical style that was
“worthy of a great nation”, which in itself suggests the presence of competing
nationalisms.90
The criticisms from Escudier continued, however, and by 1880 Saint-Saëns felt
compelled to defend the Société for its perceived intolerance toward those outside
their circle. “The members come together on one point only,” said Saint-Saëns: “the
cult of serious music, and separate on all others.”91 Strasser notes a great deal of
diversity in the Société’s concert programs, but the critics’ problems seem to be with
“seriousness” of music, aiming to elevate rather than amuse, and, to some, this
seemed a very “German” attitude to take.92 By the 1880s the Société seemed to have
become less of a forum for new music and more of an instrument “for the renewal
of French society through the promotion of ‘serious’ music”: in other words, a
quasi-political vehicle rather than an encouraging compositional forum. 93
However, although the composers of the Société wished to create a distinctive
French style, they suffered criticism from conservatives for drawing on German
techniques and ideals. Such conservatives labeled the Société as a haven for those
who professed Germanic sympathies, and this view was to intensify in the coming
years.
Despite Saint-Saëns’ 1880 defence, by 1886 both he and Bussine, the two principal
founders of the Société, had resigned. Strasser explains:

[Their resignation] revolved precisely around the growing influences of Wagner’s
most fervent admirers, many of whom were also pupils of César Franck.
89 Ibid., 238.
90 Ibid., 236.
91 Ibid., 250.
92 Ibid., 251.
93 Ibid.
20
Composers . . . who had once defended the controversial German master and
insisted that French musicians should welcome fresh ideas from any quarter, came
to fear the impact that Wagner’s music was having on both the public and young
French composers. . . Franck’s young disciples eventually caused many of the
Societe’s less adventurous members to leave the organization.94
Although Franck was elected President of the Société Nationale that same year, it is
difficult to tell how much of this fervent Wagnerian attitude can be attributed to
him. It seems more likely that this attitude came from his enthusiastic disciples –
though it seems unlikely that they would have followed this path without Franck’s
tacit approval.95 Nevertheless, following Franck’s election to the Société Nationale
Presidency, the Société concerts allowed German works to be performed alongside
French.
There seems to be no “correct” answer to Franck’s nationality; indeed, he seems
most cosmopolitan. Despite working in Paris and influencing a generation of
composers who were French, his musical influences were, in the broadest sense,
predominantly German. Perhaps attempting to delineate any further is counter-
productive – we know Franck’s background – and this thesis hopes to illuminate
that which might be considered novel and that which might, in a technical or
constructive sense, be considered derived from the Germanic tradition of the time.
The following section introduces Franck’s influences and style, as well as the
works that will prove most fruitful for detailed analysis.
94 Ibid.
21
Influences and Style
“Liszt has the richest musical imagination of our time. His works whether for
piano or for orchestra, are a mine of melodic and harmonic treasures.”
– César Franck96
Liszt was an enormous influence on Franck from the first Trios concertants that he
purchased and suggested changes to, as previously discussed. These Trios
foreshadow the cyclic procedures that were to become commonplace in Franck’s
later works. However, it is important to differentiate between Franck’s cyclicism
and other formal innovations that were gaining popularity during this period, the
most famous being those of Berlioz’ Symphonie Fantastique. In this work, Berlioz
uses an obsessive idea – an idée fixe – in all five movements of his programmatic
Symphony. However, this technique differs from cyclic procedures in that it is
present in all the movements and used obsessively, its rhythm and mood altered
each time to fit the programmatic context.97 In a more straightforward cyclic form,
a composer recalls a theme or themes from preceding movements in the finale.
This differs from a theme being reused and reworked in all movements, and such
restatements are most often used in non-programmatic works. Furthermore,
cyclicism usually only becomes apparent in the finale, as opposed to appearing
throughout the entire work (though there are exceptions to this).
With Beethoven’s recall of themes in the Ninth symphony (1824), Schubert’s
“Wanderer” Fantasy (1823) and Schumann’s Piano Quintet (1842), by 1843 there
were certainly precedents for cyclic procedures.98 Despite this, d’Indy suggests that
Franck’s Trios initiated Liszt’s interest in cyclicism:

How this young pupil of the Paris Conservatoire came to conceive the idea of
constructing an important work upon the basis of a single theme, concurrent with
other melodies which also reappear in the course of the work, thus creating a

97 Hugh Macdonald, "Idée Fixe" in Grove Music Online. Oxford Music Online.
98 Robert et al. Winter, "Schubert, Franz," in Grove Music Online. Oxford Music Online.
22
musical cycle – a form which Liszt alone foresaw without ever arriving at a perfect
development of it.99
Although Liszt developed cyclic and thematic procedures in a far more
sophisticated manner than Franck, it is unlikely we will ever know which
composer first conceptualised cyclic form; it may even have been neither Franck
nor Liszt, but someone else entirely.100
An obvious similarity between the two composers is their genres of choice. Both
wrote solo piano and organ music and a number of Symphonic poems (though
Liszt wrote his primarily during the 1850s.) One such poem of Franck’s, Le chasseur
maudit, makes a fascinating tonal study, and a useful precedent for later works as it
employs “active” key relationships – those of minor thirds – which contrast with
the subdominant axial emphasis in the first section, related by major thirds. Le
chasseur maudit implies an awareness of tonal relationships as agents of drama; a
function that Liszt (among others) exploited in harmonic schemes of many of his
works. Franck’s Prélude choral et fugue for piano demands formal comparison with
Liszt’s epic B minor Sonata. Trevitt notes that the Prélude, choral et fugue’s
“chromatic generating motif (later becoming the fugue subject) further emphasises
[Franck’s] underlying relationship to the music of Liszt.”101 He also discusses how
Liszt influenced Franck’s musical idiom, particularly in the piano works.102 Both
99 D'Indy, César Franck: A Translation from the French of Vincent d'Indy with an Introduction by
Rosa Newmarch, 88.
100 Trevitt and Fauquet, "César Franck." Liszt dedicated his B minor Sonata to Schumann as an
acknowledgement of Schumann dedicating his Fantasy in C to Liszt. The Fantasy, from 1836, may
be a link between Schubert’s Wanderer Fantasy and Liszt’s B minor Sonata. Although Schubert’s
work is not as explicitly cyclical, it is obvious Liszt was aware of Schubert and the cyclicism of his
E-flat Piano Trio. It is difficult to pin the genesis of cyclic form, but it was surely a German
preoccupation via Beethoven to Schubert and Schumann before Liszt and Franck. However, this
transmission was not necessarily a seamless one, as Dahlhaus argues that nineteenth century
composers did not fully understand the cyclic processes in Beethoven’s works. Carl Dahlhaus,
Between Romanticism and Modernism: Four Studies in the Music of the Later Nineteenth Century
(Berkeley: University of California Press, 1980), 41.
102 Ibid.
23
works unify a multi-movement sonata into a continuous musical cycle; Franck
making use of cyclic form (here defined as the recall of themes unaltered) and Liszt
of thematic transformation (here defined as recall of themes that are recognisable
but altered). Moreover, both works use an axial tonal scheme, which I will discuss
in more detail in Chapter Two.
Wagner must rank highly among the other important influences on Franck’s
musical style. By 1873 Franck had heard Lohengrin, Tannhäuser and Die Walküre,
though Duparc described Franck’s reaction to the last Opera as: “tepid”.103 In 1874
Franck heard the Prélude to Tristan for the first time, and James Briscoe describes
how “Franck . . . treasured his copy of [Tristan und Isolde] without seeing fit to
remove the word ‘poison’ he had early scrawled on the title page.” 104 Though he
never made the pilgrimage to Bayreuth, Wagner’s harmonic language
undoubtedly affected him. Some works after this date bear aspects of Wagner’s
expanded tonality, including Les Eolides, the Prélude, Choral et Fugue and a
quotation in the String Quartet, though the extent to which the chromaticism might
have been learned from other sources is difficult to tell.
Yet, Franck’s music does seem to have become more chromatic as he matured;
Trevitt notes that “a very rich strain of chromaticism is indeed a consistent feature
of his mature works.”105 However, indications of rich chromaticism are also present
in works that pre-date 1874, when Franck first encountered Wagner’s prelude to
Tristan. 106 We must remember, however, that Franck did not make the type of
harmonic advances we attribute to Wagner, Brahms, Liszt or Debussy. According
to Davies, Franck:

104 James R. Briscoe, "Debussy, Franck, and the 'Idea of Sacrifice'," Revue belge de Musicologie /
Belgisch Tijdschrift voor Muziekwetenschap 45, no. César Franck et son temps (1991): 37.
106 Leon Platinga, Romantic Music. A History of Musical Style in Nineteenth-Century Europe (New
York: W.W. Norton and Company, 1984), 443.

24
lacked culture and that sort of intellectual vision of his art which modern critics
rate as indispensable.107 One would have thought [that Wagner and Franck] had
much in common, especially in view of their chromatic leaning. But it is quite likely
that Franck never really grasped what Wagner was trying to do with the language
of music.108
This charge demands exploration. Though there is truth in the statement “Wagner
was to explore the association of much more distantly related harmonies [than
Franck did]”, this does not mean that Franck’s harmonic idiom is immature or
underdeveloped – instead, when examined closely, it presents a different internal
logic from that valued and utilised by Wagner. This is not least because Franck
wrote predominantly absolute music – he did attempt several operas, but these
were not his prime focus (Vallas suggests they were composed to appease his wife,
who encouraged Franck to pursue any venture she thought might be commercially
successful109) and the harmonic requirements of a Music-Drama composer and that
of a writer of symphonic poems and chamber music may well be different. Both
composers may be building with similar materials, but the inner frameworks are
incomparable. This is why Franck compares more easily to Liszt; their genres of
choice are more homogenous.
It is also important to consider exactly what the term “chromaticism” entails.
Though in a simple sense the term refers to any note outside a given key, by the
late nineteenth century some out-of-key chords became so common that they were
virtually considered diatonic.110 Yet we must somehow also differentiate between
largely diatonic works that use chromatic sonorities and works that are more
strongly influenced by a chromatic background structure – especially as Franck
wrote works representative of both categories. Cohn discusses the appropriateness

108 Laurence Davies, Paths to Modern Music (London: Barrie and Jenkins, 1971), 14.
110 Robert Cameron Cook, "Transformational Approaches to Romantic Harmony and the Late
Works of César Franck" (University of Chicago, 2001), 120.

25
of such terms and suggests that: “‘triadic post-tonality’ may be the more
appropriate term for the chromatic music of the late nineteenth century.” 111
“Expanded tonality” is also useful in its imagery, conveying an open-ended
diatonicism grown to include non-traditional sonorities. As we cannot consider
Franck an atonal composer, either term seems appropriate, though “chromaticism”
is such a prevalent term in the literature that it is bound to continue to turn up in
this thesis. I outline the analytical terms of reference in Chapter Two.
Finally, we must also consider the influence of the Viennese masters, in particular
Beethoven. One of Franck’s earliest works, Final from Six Pièces for organ, presents
this affect very simply: the opening manual parts are uncannily similar to
Beethoven’s Hammerklavier Sonata, Op. 106; furthermore, both are in the key of B-
flat. The similarity is mainly from the plagal voice-leading that permeates the
opening phrase of each work. Though the works develop differently, Beethoven’s
Hammerklavier makes a fascinating study within the axial-tonal paradigm, and is a
work we will return to.112
The String Quartet is similarly influenced: unsurprisingly, like others after
Beethoven, Franck regarded the genre with “awe and superstition” and, as in the
case of Fauré, the presence of only one exemplar in this genre suggests its
importance.113 D’Indy reports seeing scores of Bach, Schubert and Beethoven on
Franck’s workbench (and it must be noted that this is one of the instances of
specific influence that we can point to in Franck).114 Similarities with Beethoven’s
E-flat major Quartet Op. 127, in particular, are present; indeed, there is a note
111 Richard Cohn, "Introduction to Neo-Riemannian Theory: A Survey and a Historical

Perspective," Journal of Music Theory 42, no. 2 (1998): 168.
112 Beethoven’s Hammerklavier seems to have made a similar impression on the young Brahms:
the opening of his first Piano Sonata in C Major, Op. 1 also bears a strong likeness to the
Hammerklavier. I discuss this extensively in Chapter Five.
114 Ibid.
26
scrawled in the margin of the finale: “A new phrase needed here; see the E flat
quartet of Beethoven, Op. 127.”115 Furthermore, in the first movement, Franck’s use
of a slow “Lied” theme in conjunction with an Allegro is similar to Beethoven’s
alternation of Maestoso and Allegro in the first movement of the E-flat Quartet Op.
127.116 Franck was also fond of the “Beethoven’s Ninth” technique of quoting the
themes from each movement at the start of the finale, and he does this in the String
Quartet as well as in the Grande Pièce Symphonique. Though such techniques may
seem contrived to our modern ears, this does not negate the very real influence
Beethoven and his predecessors had on Franck. This may have been part of
Franck’s charm, as it were: Trevitt suggests that,

To his pupils, Franck communicated both the Beethoven idealism inherent in the
cultivation of the strict genres of symphony, quartet and sonata and the harmonic
innovations of late Romanticism. This double allegiance to the Viennese tradition
on the one hand, and to Liszt and Wagner on the other, was undoubtedly
responsible for the self-indulgent massiveness which characterizes many Franckist
works.117
This “massiveness” may be characteristic, but it also proves problematic,
particularly when its effect seems “staged” as Dahlhaus suggests.118 I examine the
Symphony in more detail in Chapter Eight.
As well as the important influences of composers, it is apt to note some stylistic
features of Franck’s that appeared from an early age (that we might describe as
“influences of personality”) and others that may have developed because of his
instrument, the organ. One of Franck’s first professors, Zimmerman, commented
that Franck possessed an “over-ardent manner of expression”.119 Alongside this,
Davies suggests that Franck’s intelligence “did not seem equal to his musical

118 Carl Dahlhaus, Nineteenth Century Music (Berkley: University of California, 1989), 276.
27
gifts.”120 “He lacked self-criticism”, Davies explains, and he is often described as
the “organ-loft composer”. An examiner of Franck’s in an early test described his
improvisation as “a little vague and monotonous.” 121 Similarly, an early reviewer
described the harmonies of his improvisation as “reminiscent of Fingal’s cave”, but
lamented that Franck “did not end soon enough.” 122 Indeed, part of Liszt’s
suggestion to reverse the finales of two of Franck’s early Trios was because he
thought the finale of the third was too long and constituted a work within itself. 123
Though it is desirable to write off these descriptions as youthful immaturity, it
seems that Franck had a penchant for long improvisations his entire life. An
account by Pierné, a pupil of Franck’s and his successor at St-Clotilde, describes
the length of Franck’s improvisations:

The curé of Sainte-Clotilde had first installed a little bell in the blower room. When
this little bell rang, it meant: “Monsieur Franck, by order of the curé, stop!” But M.
Franck, absorbed, did not hear the little bell. Then they put in a louder electric bell.
That one Le Père Franck heard (he couldn’t do otherwise) but then he would
exclaim, “I never have time to return to the correct key.” And, unruffled, he
modulated, following the rules, in order to come back to the original key . . . the
good curé of Sainte-Clotilde, after having opened and closed his breviary several
times, always hoping to hear the final chord, got up and went himself to the
maîtrise to give two or three imperious rings. Then he confirmed his order by
sending a chorister up to the organ. But Franck didn’t stop.124
This story effectively illustrates the difficulty Franck had in completing his
improvisations in a timely fashion, a difficulty that some argue is also inherent in
many of his compositions. Despite this problem, however, improvisation can also
lead to inspiration, and may lend itself positively to his work. Emory Fanning
120 Ibid., 44.

122 Ibid., 25.
123 Vincent d'Indy, "La Première Manière De César Franck" Revue belge de Musicologie / Belgisch
Tijdschrift voor Muziekwetenschap 4, no. 5 (1923): 6.

124 Smith, Toward an Authentic Interpretation of the Organ Works of César Franck, 160.
28
writes that improvisation “accounts for the character of his melodic extension and
development, thematic expansion in which an incidental fragment or motive is
made the constituent of a new phrase.”125 Trevitt also discusses how Franck creates
thematic material in this way: “Often Franck developed complex phrase structures
using a kind of mosaic of variants of one or two germinal motifs, a technique
which again underlines his indebtedness to Liszt.”126 I discuss these features more
fully later in the thesis, but at this point it is pertinent to look at some of Franck’s
broader compositional characteristics.
There are several features of Franck’s architectural principles that require
discussion: cyclic form; the blending of movements to create unity; and his
unusual tonal schemes.
Cyclic form is a technique commonly associated with Franck, referring to the
quotations from earlier movements in the finales of the Symphony, String Quartet
and other works. However, recent scholarship has begun to grapple with how we
might actually define cyclic form and related techniques including thematic
transformation, ideé fixe, leitmotif, and thematic reminiscence. Walter Frisch
quotes Hugh Macdonald’s New Grove article: “cyclic form”, in the “strictest sense”
is most often used to describe “works where thematic links bind more than one
movement; it is not properly applied to mere thematic resemblances.” 127 The
“Thematic Transformation” entry by the same author describes it as: “A term used
to define the process of modifying a theme so that in a new context it is different
but yet manifestly made of the same elements.” 128 Yet, Frisch points out that
“propriety is hard to establish in other contexts: one analyst’s ‘mere thematic
125 E.M. Fanning, "The 19th Century French Organ of Cavaillé-Coll and Organ Works of César
Franck" (PhD Diss., Boston University, 1964), 48.
127 Walter Frisch, "The Snake Bites Its Tail: Cyclic Processes in Brahms's Third String Quartet,
Op. 67," The Journal of Musicology 22, no. 1 (2005): 155.

128 Hugh Macdonald, "Transformation, Thematic," in Grove Music Online. Oxford Music Online.
29
resemblance’ may be another’s ‘thematic link.’”129 Furthermore, at what point do
such instances constitute a cyclic form? In his 2006 thesis on Mendelssohn’s
instrumental music, Benedict Taylor suggests that “‘Cyclic form’ . . . is a vague and
ambiguous term which incorporates many differing definitions.” 130 He observes
that in previous literature “any work in which music from a previous movement is
recalled and reused in a later one can be termed “cyclic”. Thus, in itself, thematic
affinity between movements of a work – even possibly an ongoing development
and resolution – should strictly speaking not really warrant the epithet “cyclic”.”131
His eventual definition, which he uses to examine Mendelssohn’s music in the later
chapters of his thesis, is as follows:

Cyclicism, with regard to musical form, entails at the very least close thematic
affinity between movements or, more properly, explicit recall of music from one in
another. These two types can be designated transformative and recalling cyclicism.
To insist on a division between these two would be problematic, since the
boundaries can be blurred between what constitutes close allusion and literal
reprise. Moreover, many works which recall past movements tend to feature such
thematic interconnections; cyclicism would be seen as a continuum of possibilities
between these two types.132
Many of Franck’s works display instances of both transformative and recalling
cyclicism, from the Op. 1 Trios concertants to the piano works, Violin Sonata, String
Quartet, Symphony and organ Chorals. 133 The motivation behind the idea is
understandable: such techniques are one way to create unity in a large work. In
some of Franck’s pieces the effect is convincing, or at least continued presence in
the repertoire would seem to suggest, such as in the Violin Sonata. Here, themes
from the first movement are parodied in the second and recur in the third and
129 Frisch, "The Snake Bites Its Tail: Cyclic Processes in Brahms's Third String Quartet, Op. 67,"
155.
130 Benedict Taylor, "Cyclic Forms in the Instrumental Music of Felix Mendelssohn: Time,
Memory and Musical History" (PhD Diss., University of Cambridge, 2006), 9.

131 Ibid., 8.
132 Ibid., 9.
30
fourth, but do not sound contrived. Elsewhere, such as in the Symphony, the recall
of previous themes in the last movement sounds awkward: the haunting Cor
Anglais melody from the second movement is paraded oddly by the brass in the
third, whilst the return of the first theme in the same movement requires a clumsy
tempo shift.134 Perhaps it was this that motivated Camille Bellaigue to write: “what
[Franck] takes to be unity and cohesion could well be nothing but aridity and
poverty.”135 It is important to note, however, that Franck’s cyclicism is not thematic
transformation by another name. The latter technique, associated with Liszt, is
more sophisticated, describing a theme that undergoes change in the course of a
work. The effect is similar to that of cyclic recall, but with the advantage of being
able to be altered providing its identity is preserved, meaning a programme – such
as “darkness to light” – can be underscored by changes in the theme, such as the
oft-cited Third Symphony of Saint-Saëns.
Another formal principle common in Franck’s works is the amalgamation of
several movements into one, or the creation of one movement with several sections
of varying character. Examples include Grande Pièce Symphonique, the solo piano
works, the symphonic poems, Variations symphoniques and the Trois Chorals. Such
unification is not peculiar to Franck: many of Liszt’s Symphonic poems also do this
and the B minor Sonata is an oft-cited example; in the realm of “absolute”
Symphonic music Schumann uses the technique in his fourth Symphony136, and
Saint-Saëns unified movements 1 – 2 and 3 – 4 of his Third Symphony. The second
movement of Franck’s Symphony is particularly innovative, however: featuring a
“telescoped” Scherzo framed by two slow sections of equal length. This suggests a
concern for form that persevered well into Franck’s career.
135 Ibid., 609.
136 Taylor, "Cyclic Forms in the Instrumental Music of Felix Mendelssohn: Time, Memory and
Musical History", 12.

31
The final aspect of Franck’s style that I consider now is his tonal architecture. This
is the foremost focus of this thesis; here, I present several ideas, which I will
expound in later chapters. Consider the following comments made about Franck’s
Symphony in D minor by a reviewer in Le Temps: “M. Franck sins by an excess of
enthusiasm: in wanting to do too much, he has made his work dull and has abused
chromatic progressions to a singular degree.”137 Furthermore, Ambroise Thomas,
from the Conservatoire, supposedly asked: “What is a symphony in D minor
whose first theme is in D-flat at the ninth measure, C-flat at the tenth, F-sharp
minor at the twenty first, C minor at the twenty sixth, E-flat at the thirty ninth and
F-minor at the forty ninth?” 138 Though these keys are temporary or sequential
modulations rather than significant centers, Thomas’ grievance – the incessant
chromatic modulation within Franck’s Symphony – remains relevant, and I
consider this in Chapter Eight.
It is interesting that Schubert, Liszt and others employ similar tonal discursiveness
without attracting the same level of criticism. There are a number of possible
reasons why Franck’s reception has been so mixed: perhaps his music was
misunderstood (in which case analysis might inform us better about internal
structures therein); or perhaps such a mixed reception is symptomatic of the time,
as Franck’s structure may be also. Further study will explore both possibilities.
What is most important is that something in Franck’s tonality is unusual; there is
more than chromaticism at work, and there is more than chromaticism to be
concerned with. The analysis, Chapters Five to Nine, reveals a deeper coherence
behind the discursiveness in the music of Franck and others. The task of this thesis
is to see whether a similar approach to Frank’s work reveals something similar or
profoundly different.
138 Ibid.
32
It seems inconclusive to state that Franck “did not understand what Wagner was
doing”139 or “A very rich strain of chromaticism is indeed a consistent feature of his
mature works.”140 Why this chromaticism, and why this tonal discursiveness? Only
closer analysis of Franck and his predecessors will tell if there is something deeper
beneath the surface. I examine six works from Franck’s late period in detail: Le
chasseur maudit and Psyché (Symphonic Poems), the Quintet, the String Quartet, the
Symphony and the Violin Sonata. I will refer to other works peripherally as
necessary. These works demonstrate an internal logic of tonal architecture,
elucidating Franck’s internal structures and considerations.
Before this analysis can begin, however, it is necessary to trace the origin of the
ideas about tonal organisation that may be relevant to Franck. The next chapter –
part two of the literature review – considers literature regarding axial tonality.
139 Davies, Paths to Modern Music, 14.

33
Chapter Two
Historical Documentation and Theory of
Third-Relatives in Music
This section of the literature review is primarily concerned with Riemann and
Neo-Riemann analysis. Cohn writes that “Neo-Riemann theory arose in
response to analytical problems posed by chromatic music that is triadic but not
altogether tonally unified.” 141 He quotes Rothstein’s analysis of Wagner’s
music:
Some phrases . . . are not true phrases at all from a tonal point of view
because they do not contain a single coherent tonal motion. Such phrases
. . . must be said to cross the hazy line separating tonality from post-
tonality (or whatever one wishes to call triadic but post-tonal practice).142
Although some of Franck’s music is purely diatonic (such as some of the Six
Pieces), some works seem to cohere to other principles or not at all – to “cross
the hazy line”, as it were. Sections from the Symphony, String Quartet and
Symphonic Poem Psyché belong to this category. This is the type of music that
Cohn is concerned with when he asks “if this music is not fully coherent
according to the principles of diatonic tonality, by what other principles might
it cohere?”143 Riemann, Neo-Riemann and related types of analysis offer insight
into ways that such coherence might be described. These types of analysis are
concerned with how systems that divide the octave equally might form
coherent modulatory patterns; within this, the intervals of major and minor
141 Richard Cohn, "Introduction to Neo-Riemannian Theory: A Survey and a Historical

Perspective," Journal of Music Theory 42, no. 2 (1998): 167.
142 Ibid., 168.
143 Ibid., 169.
34
thirds become significant in their ability to divide an octave equally. Many
writers examined here are interested in properties of major-third related keys in
particular and their potential for functional equivalence.
This literature review is primarily concerned with two questions. Firstly, what
evidence is there to suggest that major-third relatives can be functionally
equivalent, in terms of both chord to chord motion and large-scale key
relationships? Secondly, to what extent can such progressions or structures be
incorporated into a framework that is still largely governed by diatonic
principles? To answer these questions, this chapter begins with a survey of
Hugo Riemann’s analytical insights, as he was one of the first – and arguably
most significant – to articulate harmonic function theory. As function is so
integral to our analysis, it is necessary to examine Riemann’s work first. His
writings draw on the work of Rameau, Reicha, Weber, A.B. Marx and
Hauptmann, among others. From these he draws a dualist theory which
privileges third-relationships, and in the past three decades his ideas have
piqued the interest of modern analysts. The most important of these will be
examined in this chapter: Daniel Harrison, Richard Cohn, Graeme Downes and
David Kopp all contribute to the elucidation, updating and expanded
application of Riemann’s ideas. The synthesis of these ideas will form my
analytical method, to be used in my analysis of musical works in subsequent
chapters. The aim is a representation of tonal events that accurately describes
the listener’s experience and the composer’s intention. As Lewin writes: “The
task of the analyst is ‘merely’ to point out things in the piece that strike him as
characteristic and important (where by ‘things’ one includes complex
relationships), and to arrange his presentation in a way that will stimulate the
musical imagination of his audience.”144
David Lewin, "Behind the Beyond: A Response to Edward T. Cone," Perspectives of New
144
Music 7, no. 2 (1969).

35
Key
Hugo Riemann (1849 – 1919)
One of Riemann’s first writings was “Musikalische Logik” (1874), an early
treatise and adaption of his doctoral dissertation. Despite not addressing
harmonic function specifically, Riemann treats Hauptmann’s ideas in a way
that might be considered prophetic. Riemann describes Hauptmann as having
“well understood the unity of chords within a key” but criticises him for not
addressing “the different meaning of these chords to each other, about their
logical meaning in musical composition.”145 Though the word “function” is not
used, Harrison points out that “logical meaning” might be considered a
synonym to Funktionsbezeichnung (which can be translated as “function
indicator” 146 ), describing the idea as a “constant red thread” throughout
Riemann’s career,147 and signifying his ideas about harmonic function may have
been brewing as early as 1872. In Skizze einer neuen Methode der Harmonielehre
(1880), dedicated to Franz Liszt, 148 Riemann presented the idea of dualism:
major and minor being mirror images of one another, generating mirror image
systems. By this Riemann is referring to the conception of the dominant triad
being built above the tonic triad, and the subdominant triad inverted below,
like so:
Figure 2.1: Hauptmann’s diagram of keys149
F–a–C –e –G –b–D
Subdominant Tonic Dominant
145 Daniel Harrison, Harmonic Function in Chromatic Music (Chicago: University of Chicago
Press, 1994), 266.
146 Translation by K. Yelavich, 2011.
147 Harrison, Harmonic Function in Chromatic Music, 266.
148 Ibid., 274.
149 David Kopp, Chromatic Transformations in Nineteenth-Century Music (Cambridge: University
Press, 2002), 54.

36
This is a borrowing of Hauptmann’s diagram from the treatise Die Natur der
Harmonik und der Metrik of 1853. 150 Riemann discusses the importance of
“directed intervals between . . . roots” when defining a key: “In a work: Key is
nothing greater that the meaning of a chord in relation to its tonic. The key is
left as soon as this meaning changes, which can happen without the
introduction of a chromatic tone. The modern name for this notion of key is:
Tonalität.”151 But as we might expect, not all root-interval progressions are equal
in power and clarity. Riemann outlines the progressions which he considers the
most intelligible; that is, those that present “direct opposition” without
endangering the key:152 these include fifth relations, third relations, and some
step progressions. So whilst Riemann does not describe mediants as tonic or
dominant substitutes, he does state (in Kopp’s words) that, “like the dominants,
chromatic mediants are a separate type of progression which may support
meaningful key relation to the tonic.” 153 In supporting this meaningful key
relation, Riemann suggests that chromatic chords should not be considered
borrowings from other keys, but as related to the Tonalität.154
Some of these ideas begin to change in Riemann’s next treatise, the
Systematische Modulationslehre published in 1887. Here, Riemann is beginning to
think more in terms of harmonic function and a cadential model of harmony.
He shortens the list of keys that enjoy “close relation” to the tonic: now, in C,
the keys would be the minors C, A and E; and the major keys G, F, E and A-flat
majors (notice how this list includes both major third relations to the tonic). 155
This illustrates Riemann’s belief that the two most important relations to a tonic
were a fifth and a major third, further reinforced by the inclusion of a grid
150 Ibid., 52.

151 Ibid., 72.
152 Ibid., 73.
153 Ibid., 75.
154 Ibid., 78.
155 Ibid.
37
representing fifth relations horizontally, major third relations vertically and
minor third relations diagonally.156
Figure 2. 2: Riemann’s early Tonnetz157
Riemann also discusses how chromatic mediants might best be incorporated
into a musical work. Parallels between what Riemann discusses here and the
works analysed later in this thesis are striking, and pose the question whether a
major third related key can sustain direct harmonic motion. Riemann writes:
This question is to be answered with an unconditional yes . . . The
principal meaning of third relations at its clearest and most
indisputable stands out when, in cyclic forms (sonatas, symphonies), a
free-standing middle section in a third-related key is set against [the
tonic], or also when a second theme of a movement enters in such a
key. If E and A-flat major are understandable and in excellent relation
to (or rather in between) C major, then there is only a single
explanation, that a connection by third relations is to be understood
between the two tonics. Marx also puzzled over why it is that E major
and A-flat major seem less foreign than D major and B-flat major.158
156 Ibid., 76 – 78.

157 Ibid., 77.
158 Ibid., 81.
38
Riemann’s writings will be discussed further in the section regarding chords.
For now, more recent writing on third-related keys will be examined.
Other analysts
Graeme Downes’ 1994 thesis on Progressive Tonality in Mahler’s music asks
why “composers favoured some tonal areas over others despite the equal
availability afforded by an increasingly chromatic harmonic language.”159 For
example, ♭III is a common second tonal area in minor key works, as it seems to
articulate the expected opposition of sonata form. However, the opposite
relationship (I—vi) is much less common, which suggests that composers
simply found this relationship less useful. Stating that “if we find that several
keys relate to the tonic in the same way then we can justify the notion of an
axial system”, Downes demonstrates which voice-leading steps are necessary to
define a tonic and how those with similar characteristics interrelate. This
concept relates specifically to Riemann’s ideas about which chords were
capable of conveying “tonic”, and there is certainly much overlap between the
systems. Downes refers to this as an “axis system” and outlines it thus, relative
to the tonic of C:
Figure 2.3: Downes’ Major Third Axis System160

Subdominant Tonic Dominant Supertonic
IVα A Iα E Vα B IIα F sharp
IV F I C V G II D
IVβ D flat Iβ A flat Vβ E flat IIβ B flat
This table can be considered one of both chords and keys; as Downes states,
159 Graeme Downes, "An Axial System of Tonality Applied to Progressive Tonality in the
Works of Gustav Mahler and Nineteenth Century Antecedents" (PhD Diss., University of
Otago, 1994), 15.
160 Ibid., 17.
39
Tovey describes tonality as being a “. . . long-range view of harmony”.
It naturally follows that each level is governed by the same rules, so
that, for example, a chord of the dominant seeks resolution in the same
way that any passage “on” or “in” the dominant key seeks resolution to
the tonic key.161
However, to justify this capacity for functional equivalence, we must break
down the tonicisation process into constituent elements: that is, voice-leading
elements which tonicise. The common denominator is that of a major third –
each major third above the tonic is labelled Iα, and the major third below the
tonic Iβ. Downes describes how chords which share efficiency of voice leading
to the tonic, dominant and subdominant can relate to and substitute for one
another.162 If the triads in the table are observed vertically, we can see that such
chords are those related by major third. He explains that in this context, for a
chord to function as a dominant, it must either feature a leading note (with the
resolution by upward rising semitone) or a perfect wholetone (with the
resolution by an upward rising whole tone).163 Downes describes dominant and
subdominant functions in this way:
Figure 2.4: Downes’ dominant/subdominant functions
Dominant functions
Leading note – resolution by upward rising semitone
Perfect wholetone – resolution by upward rising whole tone
Subdominant functions
plagal leading note – resolution by falling semitone
plagal wholetone – resolution by falling tone164
161 Ibid., 14.

162 Ibid., 18.
163 Ibid.
164 Ibid.
40
Therefore if we are in the key of C, B can function as a dominant with its
upward rising semitone from B – C and E-flat can function as a dominant with
its upward rising wholetone from B-flat – C. Similarly, chords with a
subdominant function feature a plagal leading note (resolution by falling
semitone) or a plagal wholetone (resolution by falling tone). If substitution
occurs, we can label such sonorities as “axial dominants” and “axial
subdominants,” and, similarly, such progressions as “axial progressions.” 165
Such progressions on the small scale also hold true when extended works
modulate to larger areas. However, this must be qualified: despite being able to
“read” axial relationships into a large amount of music, there are, of course,
other ways in which keys relate which are sometimes more pertinent. For
example, some analyses in this thesis feature wholetone relationships, though
axial “readings” may be possible. The question we must ask ourselves is
whether we hear a substitute dominant or subdominant, or an extension of the
dominant or subdominant area. Often the answer is yes, but if it is not – or it is
unclear – then we must seek alternate explanations. So, enlightening though the
axial system is, we must beware of applying it to works that may be operating
under different frameworks.
Nevertheless, Downes correctly states that the compositional aims of
nineteenth-century composers “enforce[d] the avoidance of the cadential
structures that dominated the Classical period.”166 Works which might be said
to avoid traditional cadential structures – or where V – I cadences are avoided
in favour of other cadential motion – certainly include the first movement of
Franck’s Violin Sonata, Beethoven’s E-flat Major Quartet Op. 127, the first
movement of the Hammerklavier Sonata, and the first movement of Brahms’
third Symphony. In these and similar cases, linear concerns became more
165 Ibid.
166 Ibid., 20.
41
important: “cadences less emphatic than the traditional perfect cadence had the
advantage of being able to define the tonic without disrupting continuity.”167
What is most significant about this proposed axis system is the premise that
third-related keys will exhibit “a degree of neutrality or functional equivalence
based on the premise that tonality is a deeper structural level application of the
harmonic laws of the foreground.”168
Figure 2.5: Dominant voice-leading efficiency (adapted from Downes169)
Here I illustrate the Dominant voice leadings, and I have added VLE, which
refers to voice-leading efficiency, as illustrated by Cohn in “Maximally Smooth
Cycles”.170 Downes writes: “The most important tonic defining motion is the
leading note, which is common to all axis members hence justifying the axis
status of these keys on melodic grounds” 171 , and further, “Naturally many
(though not all) of these progressions are rare . . . as actual cadences. But they
167 Ibid., 21.

168 Ibid.
169 Ibid.
170 Richard Cohn, "Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of Late-
Romantic Triadic Progressions," Music Analysis 15, no. 1 (1996): 23.

171 Downes, "An Axial System of Tonality Applied to Progressive Tonality in the Works of
Gustav Mahler and Nineteenth Century Antecedents", 21.

42
do appear at the deeper structural levels.”172 It follows that structural dominant-
axis progressions are those which utilise leading-note and other dominant
voice-leadings; subdominant-axis structural progressions are those which
utilise subdominant voice-leadings.
Figure 2.6: Subdominant voice-leading efficiency (adapted from Downes173)
Now we will examine the final partitioned group, the “supertonic axis”. This
may be a misleading label for this axis group, as “supertonic” implies its
specific scale-defined role: dominant preparation. Though key areas or chords
from this group may be utilised in this way, it is also possible for them to
function as subdominant of the subdominant, or as wholetone dividers
between two tonic axis members. 174
172 Ibid., 22.

173 Ibid., 23.
174 Ibid., 24.
43
Figure 2.7: Supertonic voice-leading efficiency (adapted from Downes175)
According to Downes, this can lead to a “more stable tonic axis member” 176 and
hence may be a desirable effect. In the analyses which feature in the following
chapters, ‘WT1’ is used to refer to the same wholetone scale as the tonic when
such descriptions are appropriate. When key areas are instead acting as
supertonics or otherwise, they will also be notated as such; it is important to
note the different ways keys and chords in this axis can function. Downes’
work is compelling in its argument for the ability of an augmented triad to
function as the chromatically enhanced dominant to any tonic axis key. This
idea will have implications for many of the analyses of later chapters.
It is helpful to examine Downes’ research simultaneously with arguments by
Richard Cohn. Cohn identifies the same partitioning system: though the detail
and labelling is different to the eye, the keys are grouped together identically.
Cohn’s diagram features a map of keys grouped according to their voice
leading efficiency, with transitions between neighbouring keys made by one
semitone shift. These key groups are labelled Northern (C – E – A flat), Western
(G – B – E flat), Eastern (D flat – F - A), and Southern (B flat – D - F sharp). Each
175 Ibid., 25.

176 Ibid., 24.
44
of these constitutes a “hexatonic system”, and the grouping of the four as a
whole is described as a “hyper-hexatonic system”.177 Hexatonic refers to the set-
class 6 – 20, which in its prime form is [0, 1, 4, 5, 8, 9], or to give a pitch-class
example, [C, C#, E, F, G#, A].178 This set-class can be transposed three further
times, creating the four hexatonic sets. This concept is examined further later in
the chapter, where I link it with Downes’ ideas.
Cohn discusses how complementary regions possess a “capacity [for] motions
between the cycles to form coherent modulatory patterns.”179 In his article “As
Wonderful as Star Clusters”, which discusses Schubert’s Piano Sonata D.960 in
B-flat, Cohn writes that:
theorists recognized voice-leading efficiency as an alternative basis for
harmonic relations already during Schubert’s lifetime and throughout
the nineteenth century, but were reluctant to explore its systematic
implications, primarily because to do so would have required them to
relinquish their deep prior commitment to acoustic theory and tonal
centricity, in favour of a less hierarchic, more networked conception of
harmonic relations.180
Like Downes, Cohn demonstrates that hexatonic systems are reconcilable with
traditional functional principles. This diagram, taken from the same article,
shows the keys in each hexatonic system horizontally, demonstrating the
efficient voice-leading between triads which are a semitone away from one
another enharmonically. The vertical columns represent the traditional view of
tonality, presenting chords related by perfect fourths and fifths.
177 Cohn, "Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of Late-Romantic
Triadic Progressions," 23.
178 Ibid., 18.
179 Ibid., 17.
180 Richard Cohn, "As Wonderful as Star Clusters: Instruments for Gazing at Tonality in
Schubert," Nineteenth Century Music 22, no. 3 (1999): 217.

45
Figure 2.8: Cohn’s “Table of Tonal Relations”181
E♭+ E♭- B+ B- G+ G- E♭+
B♭+ B♭- G♭ + F♯ - D+ D- B♭+
F+ F- D♭+ C♯ - A+ A- F+
C+ C- A♭+ A♭- E+ E- C+
Cohn further discusses how chords within a hexatonic system can be mobilised
for the smoothness or edginess of their harmonic progressions. He describes
three ways in which chords within an axis relate to one another, and the
relative smoothness of each transition:
(1) adjacent harmonies, such as the relation of E flat major to E flat minor,
involve a single semitonal displacement; (2) modally matched harmonies,
such as the relation of E flat major to G major, involve dual semitonal
displacements in contrary motion; (3) polar harmonies, such as E flat
major and B minor, have no common tones and so simultaneously
displace all three voices semitonally.182
Such terms can be applied to transitions between hexatonic systems, as well as
within them.
Cohn is particularly interested in the relationship between polar harmonies
within an axis, discussing them in his article “Maximally Smooth Cycles”. 183
This focus is worth noting, as tonal poles are an important feature of Franck’s
Quintet. Cohn describes the movement between two tonal poles as “uncanny”,
and states that the juxtaposition of tonal poles has the ability to “annihilate
181 Cohn, "As Wonderful as Star Clusters: Instruments for Gazing at Tonality in Schubert,"
217.
182 Ibid., 215.

46
tonality”.184 This idea will be explored both later in this chapter and in relation
to Franck’s Quintet. It is also worth noting that pitch-class sets are discussed
later, though pitch-class set analysis itself will not be employed. Such analysis
was developed to identify structural components within atonal works, and
though Franck’s compositions are deeply chromatic, it is my belief that they
still require an analysis based in a tonal framework. Rather than analysing from
a pitch-class set framework, pitch-classes are a taken into consideration when
analysing the factors that contribute to coherence in Franck’s works, and also
how they might temporarily annihilate tonality. Background to this approach is
covered by Allen Forte.185
Cohn’s diagram showing the relationship of the four hexatonic systems –
described as the “hyper-hexatonic system” – shows each system with what
Downes would label the “subdominant axis” on its left, the “dominant” on its
right, with the “supertonic” opposite. As he explains, they are arranged in this
way to denote voice-leading efficiency. Within a hexatonic axis the number of
semitone steps between triads ranges from 0 (enharmonic equivalents, such as
F-sharp major and G-flat major) to 3 (hexatonic poles such as F-sharp major to
D minor). The reader may also note that two augmented triads can be formed.
For each neighbouring hexatonic system the steps range from 2 – 4, and for
opposite hexatonic systems the number of steps ranges from 5 – 6. Cohn
describes how “each hexatonic system is directly connected to the two
neighbouring systems, but . . . not directly connected to the system opposite.” 186
Opposite systems are used in some works, such as Franck’s Psyché, but such
works’ coherence may be more determined by shared wholetone scales or other
forces rather than relationships of an axial nature. These are examples of
184 Ibid.
185 Allen Forte, "Pitch-Class Set Analysis Today," Music Analysis 4, no. 1/2 (1984).

47
compositions which adhere to principles other than axial-hexatonic ones,
despite their apparent use of axial-hexatonic systems.
Figure 2.9: Cohn’s hyper-hexatonic system187 ‘
This diagram also effectively illustrates the pc sets inherent within the
complementary augmented triads; this will be discussed more extensively in
relation to pc set use in Franck’s Quintet later in the thesis.
In the article “Weitzmann's Regions, My Cycles, and Douthett's Dancing
Cubes”188, Cohn draws a close relationship between a primary hexatonic system
187 Ibid., 24.

48
and its “subdominant” or “Eastern” system. The relationship is explored with
regard to the augmented triad, say, C–E–G-sharp, and the triads that can be
created by displacing one semitone of the triad. From this exercise the triads C
major, A minor, E major, C-sharp minor, A-flat major and F minor can all be
generated. In this article, Cohn has drawn on the work of Weitzmann (who
predates Riemann) and his work on the augmented triad. 189 Though Cohn
explores the potential for this cycle more thoroughly (and in graphic form),
what is important for our purposes is the close relationship between one
hexatonic system and its “subdominant” neighbour, and we must consider the
possible implications of such close key-relationships.
Another writer concerned with the functional equivalence of major-third
related keys is David Kopp. Cited earlier in this chapter for his explanatory
work on Riemann, Kopp’s own thoughts on function are demonstrated in a
number of analyses. In contrast to Cohn, Kopp analyses the effect of mediant
progressions over a larger scale (possibly comparable to the “middle-
ground”). 190 He concludes that mediant relations within a system/tonic axis
may prolong the tonic. Though he does not examine the possibility of
complementary systems, he does suggest that:
given repeated hearing, third-circles can indeed convey a heard sense of
directed departure from and return to the tonic. The distance covered is
greater; the pace of change is more marked; the chromatic content
challenges the ear. But the circle is short, direct connections are present
between each element, and there is no reason to assume that the tonic
must be reintroduced by its dominant to be heard as the tonic once
again. In fact, after two identical chromatic third relations, the likely
188 Richard Cohn, "Weitzmann's Regions, My Cycles, and Douthett's Dancing Cubes," Music
Theory Spectrum 22, no. 1 (2000).
189 Ibid.
190 Kopp, Chromatic Transformations in Nineteenth-Century Music, 229.
49
continuation is one more of the same – which provides the return to the
tonic. Also a memory of the tonic may well linger to confirm arrival.191
Kopp is convinced that a tonal system is capable of supporting chromatic
relationships alongside diatonic ones, and that functionality and scale degree
can be separated. He states that: “the tonal system can support chromatic
relationships along with diatonic ones without loss of the sense of tonic . . .
harmonic relationships are holistic entities whose essence may surpass the sum
of their parts.”192 This idea is significant, as the “functions” many writers are so
concerned about can vary according to context. He finishes by saying “the
language of common-tone relations and transformations represents first and
foremost a way of thinking directly about chromatic harmony in music.” 193 This
new way of thinking, which concerns all the authors discussed in this chapter,
requires descriptive tools that can recognise chromatic entities communicating
diatonic function; tools that are descriptive rather than prescriptive; and tools
that identify the relationships between major-third related keys.
Before moving onto chord-specific literature, it is apt to consider one final
theorist: Robert C. Cook, a pupil of Cohn’s. Cook’s work is significant as his
thesis uses Franck’s work to demonstrate his theories of transformative
processes. It should be noted, though, that Cook’s ideas regarding
interpretations of chromaticism and approaches to analysis differ significantly
to my own. His view of the music is “from a transformational outlook, in which
triads relate through networks of chromatic voice leading, rather than tonal
hierarchies”,194 and he also states that “it is precisely the failure of traditional
analytical methodologies to address their underlying prescriptive, regulatory,
191 Ibid.
192 Ibid., 235.
193 Ibid.
Works of César Franck" (PhD Diss., University of Chicago, 2001), xiv.

50
hierarchical conceptual foundations that renders their analyses of chromatic
music unsatisfying”. 195

Though Cook’s work is more focused on
transformational theory rather than our concerns of sonata form and its relation
to harmony, his work reinforces one of the broader themes of this analysis: that
is, that Franck is experimenting with the functions of chromatic harmony, with
mixed success. In the article “Parsimony and Extravagance”, Cook takes
interest in passages which feature tonal poles and the harmonic events that
surround them: that is, “when a musical event or series of events encourages
intuitions of coherence while at the same time this coherence resists ascription
to the influence of a diatonic tonic.” 196 Parsimonious voice-leading is defined as
“[voice-leading which retains] two common tones when moving between
triads”197 and extravagance defined as voice-leading where “each of the three
triadic voices moves by semitone and only by semitone; no common tones are
permitted.”198 Such voice-leading effects are discussed later in the analysis of
Franck’s Quintet.
In his thesis, Cook writes that the aspects of chromatic music that are important
are the aural effect of particular progressions, motivic and thematic character,
and participation in a contextually-defined syntax: not, he notes explicitly, “the
extent to which they support a diatonic tonal center”. 199 I agree with Cook
inasmuch that any analytical method(s) should be self-reflective. Yet, as a
listener, I approach Franck from a diatonic framework because this is how I
experience most of his music: the fact that he adheres to traditional notions of
concentric tonality makes it common sense to do so, and Franck seems to see
himself working within a diatonic convention. Yes, Franck’s music is chromatic,
195 Ibid.
196 Ibid., 110.
197 Ibid., 116.
198 Ibid.
199 Ibid., 5.
51
to be sure; but if I perceive any hierarchy of sonority in his music, should I not
at least consider the prospect of a diatonic centre? Furthermore, is not the
“aural effect of particular progressions” likely to reference some sort of musical
hierarchy, whether it be part of a diatonic tonal framework or otherwise? I
suggest that interpreting Franck’s work in a diatonic context that is able to
accommodate Franck’s rich chromaticism is likely to be the best approach.
However, Cook’s analyses of Franck do sometimes make astute observations,
and some of these will be included in my own analyses. Interestingly, Cook
only examines singular movements of the Quintet, Quartet and Symphony –
not the whole works. In this sense it feels as though his analysis is focussed
more closely on isolated movements, whilst I consider the relationship between
individual movements and the whole work to be important.
It is now important to consider triads more closely, as the relationships between
chords at a foreground level have implications for keys on a larger scale.
Chord – triads (i)
Let us return briefly to Riemann. If Systematische Modulationslehre emphasised
the special properties of chromatic mediant relations, Riemann’s later treatises
treated this idea more cautiously. In the Katechismus der Musik, Riemann began
to examine function more specifically. Each chord and progression is addressed
in relation to the tonic, and designated as possessing either as tonic,
overdominant (V) or underdominant (IV) function. 200 However, despite the
prescriptive nature of explanation, Kopp does not consider this to be a “fully
worked out” functional theory. Vereinfachte Harmonielehre (1893) is a more
indicative treatise, as here Riemann retains the notion that there are three
“pillars of harmony”, as discussed above, and describes each of these as

52
embodying a function. He describes how all diatonic chords other than the
tonic, dominant and subdominant contain mixtures of notes which belong to
one of these primary triads, and that the tonic, dominant and subdominant
“constitute the three main tonal functions, or harmonic essences.” 201 Each mixed
triad has a “single functional identity”, which might be described as the
“predominant functional identity”, and is usually that of the majority element.
Relationships are described through transformational modifications of each
primary chord: the adjectives Parallele, Variante and Leittonwechsel (leading-note
change) describe the relationship between the primary chords and those with
shared intervals. 202 The parallel describes triads with common major thirds
(such as C major/A minor); the variant describes those that share a perfect fifth
(such as C major/C minor) and the leading-tone change describes those chords
that share a minor third (such as C minor/A-flat major).203 Context does play a
role however, as of course all major and minor triads are made up of both a
major and minor third. Hence an E minor chord could be functionally identified
as a tonic Leittonwechsel or a dominant parallel, depending on its context, as it
shares two common notes with the tonic and two different common notes with
the dominant. Similarly, the VI triad in major may be a tonic parallel or
subdominant Leittonwechsel, depending on its context.204
201 Brian and Alexander Rehding Hyer, "Riemann, Hugo," in Grove Music Online. Oxford Music
Online.
202 Ibid.
203 Ibid.
53
Figure 2.10: The eight third relations from a major tonic as expressed in
Riemann: root-interval system (from Kopp205)
As well as this discussion of functional equivalency, Riemann asks: whether a
mediant might convincingly precede the tonic in a final cadence. If it can, then
mediants may be considered functional substitutes for the dominant or
subdominant (see footnote 68). In Kopp’s words “[Riemann] asserts that major-
third mediants unquestionably can do so, and that minor-third mediants
probably can.”206 If this evaluation holds true, it sets an important precedent for
this thesis; yet, we must remember that despite the theoretical potential for a
third-relative to precede the tonic in a final cadence, there are few (if any)
musical examples to suggest the effect would be compositionally desirable. As
Schoenberg accounts in his “directions for better progressions” chapter:

Looking for a substitute for V, we shall consider the suitability of III. This
degree has two chord tones in common with I, and that is a shortcoming here.
But it has the leading tone, thus excluding F major, and its root progression is
relatively strong (third downwards). It ought then to be suitable, anyway. Yet,
it is not commonly used; hence, we shall not use it much either, but shall
remember why we do not: chiefly because it is not commonly used. That
means, it could be used. Its effect would very likely be weaker; above all,
though, it would be unfamiliar.207

206 Ibid., 92.
207 Arnold Schoenberg, Theory of Harmony, trans. Roy E. Carter (Berkeley, Los Angeles:
University of California Press, 1978), 134.

54
This illustrates both the pervasiveness of harmonic convention and the
importance of remembering that theories are well and good, but it is existing
music that we are most interested in.208
Despite the evidence thus far that Riemann gives for the functional significance
of mediants at a foreground level – which seems limited to common tones and
some suggestion of similar voice-leading possibilities – it is important to note
that in his 1890 treatise Katechismus der Harmonielehre, Riemann “moderated his
advocacy” of chromatic mediants, as he had difficulty accounting for them
within his theory of function:209
The chromatic mediants more readily resolve to other chords than
directly back to their tonics. These are idealised mediating chords
which regularise the chromatic tones by preserving them or by
continuation through passing motion. They are not meant to be taken
as examples of the regular, properly notated behaviour of Terzschritte
[minor third relatives] in music.210
Yet Riemann’s earlier theories discuss the inherent potential for mediant chords
to be functionally equivalent. Whether or not they must always behave this
way is another question entirely; what is important at this vantage is that they
invoked such curiosity in Riemann, and as such they deserve to be considered
more closely. Even in the passage above, Riemann grants mediants a special
208 There are certainly instances in which juxtaposition of chords that share an axis may be
desirable, but whether the effect is “cadential” is another question. Such an instance might
include the awakening of Brünnhilde in the final scene of Siegfried, which juxtaposes the
chords of C and E (which also have associative significance). Patrick Phillip McCreless,
"Wagner's Siegfried: Its Drama, History, and Music," (Ann Abor, Michigan: UMI Research
Press, an imprint of University Microfilms International, 1982), 211. Furthermore, the last four
bars of Liszt’s Symphonic Poem Les Préludes utilise a I – iii – I cadence, but this is superseded
by a final perfect cadence. This I – iii – I movement might be interpreted as a foreground
manifestation of iii’s significance: yet despite this, I – iii – I remains insufficient to end the
work conclusively.
210 Ibid.
55
quality. To answer the question posed at the beginning of this chapter, Riemann
does seem – at one time or other – to have been intrigued by the potential for
the functional equivalence of major thirds.
Riemann’s critics and problems with his theory
One aspect of Riemann’s analytical technique is that the meaning (function) and
direction (trajectory) of a chord may be described, but not together; yet both
these aspects are required to build a complete tonal picture. Later or Neo-
Riemann theory – from the 1980s to the present day – is more concerned with
structural coherence than trajectory, representing a “break from [the]
teleological paradigm” 211 represented by Schenkerian analysis. Neo-Riemann
analyses such as those by Lewin and Cohn are usually concerned with
presenting networks of harmonic relationships that are “directional but not
determined.” 212 Therefore neither an old nor new Riemann analysis – nor a
teleological Schenkerian one – is capable of illustrating both the function and
the trajectory of a given chord, primarily because music does not behave so
prescriptively that one can determine how chords will behave one to the next;
certainly not in music composed after 1800. As a functional and directional
analysis is indeed desirable, complementary analytical tools will be necessary.
It is important, however, to consider the shortcomings of Riemann’s theories.
Dahlhaus, for instance, has criticised Riemann for not clarifying whether the
tonic dominant and subdominant are “terms for chordal scale degrees or for
functions.” 213 Harrison elucidates this problem, reminding us that “Riemann
essentially equated function with primary triad. Tonic function equals tonic
meaning equals tonic triad. But what now does meaning mean? And what other
211 Ibid., 137.

212 Ibid.
56
than itself could tonic possibly mean?” 214 Harrison suggests that functional
energy is found within scale degrees, as Cook describes: “the tonal meaning of
triad relations is carried by the voice leading between individual scale degrees
of complementary tonal charge (e.g. 7 – 1 or ♭6 – 5).”215 Brian Hyer also points
out related issues with Riemann’s theories, in particular his conceptualisation of
tonic: “Because of the gravitational distortion a tonic introduces into tonal
space, and because [Riemann] thought of the dominant as a concrete triad
rather than an abstract relation, Riemann was never able to appreciate the
transformational (or chromatic) potential of his harmonic theories.”216 The same
applies for the other triads and functional categories Riemann discusses:
“rather than regarding [parallel, relative and leittonwechsel] as names for
relations between harmonies, he used them as names for the harmonies
themselves.” 217
Whilst Hyer acknowledges that “certain harmonic
configurations do seem to insist on an intrinsic relation or affiliation with a
referential tonic”, he finds the consideration of the dominant as a “being” rather
than a “doing” problematic.218
It is worth considering why certain harmonic applications were rarely used by
composers, whilst other configurations were employed regularly. It seems
reasonable to think that curiosity alone would have led to experimentation
using different keys as tonicising roles. As mentioned previously, III is often
used as a tonicising agent in minor keys, but the major key inverse, I – vi, is
rarely used.219 This leads us to consider that perhaps vi could be considered
214 Ibid.
215 Cook, "Transformational Approaches to Romantic Harmony and the Late Works of César
Franck", 9.
216 Brian Hyer, "Reimag(in)Ing Riemann," Journal of Music Theory 39, no. 1 (1995): 109.
217 Ibid., 128.
218 Ibid.
219 Exceptions here perhaps include Brahms’ B major Piano Trio and the first movement of the
Hammerklavier.
57
more of a “being” than a “doing”, if compositional statistics hold true. This
consideration may also be applied to the supertonic axis; aside from specific
well-documented instances (which are often parodistic), the supertonic axis is
rarely used as a tonicising agent or a “doing” in other words.220 Exceptions of
this include the second movements of Beethoven’s Ninth Symphony and his
Op. 70 Piano Trio. Both instances use the flat-seventh scale degree as an
oppositional tonal centre (IIβ), the second instance a sonata without
development.
Coherence
Harrison recognises that lack of coherence (or lack of a semblance of coherence)
is problematic in the extremely chromatic music such as Reger’s that typifies
the end of the nineteenth century. Analysis of such works became difficult,
sometimes to the point of near impossibility. In an article discussing Reger’s
“Atonal Expressionism”, specifically looking at his Fantasie uber “Wachet auf!
Ruft uns die Stimme”, Harrison states that:
The chord string for each repetition is slightly different each time,
however, and the tonal relations of each chord are, at best, neighbourly
on voice-leading grounds, but not at all representative of a governing
tonal center. They are what Ernst Kurth called “Absolute Progressions,”
with each chord seemingly self-sufficient and tonally referential only to
itself.221
In “Harmonic Function in Chromatic Music”, however, Harrison states that
“the umbilical relationship between certain chords and harmonic tonality was,
by the late nineteenth century, discovered to be withered and unnecessary.” 222
220 Exceptions which use the supertonic axis include the Scherzo of the Ninth Symphony and
the Ghost trio, both of which use ♭VII.
221 Daniel Harrison, "Max Reger Introduces Atonal Expressionism," The Musical Quarterly 87,
no. 4 (2005): 668.

58
Though this statement is arguably more applicable to Schoenberg than earlier
composers, if it does hold true and composers were still concerned with tonicity
as a means of coherence, then the palate that nineteenth century composers
enjoyed must be considered functional as well as non-functional. Whether
every composer managed to differentiate between the two is another question,
and Reger’s meagre reception history suggests not – it is important to
remember that the nineteenth century was a time of great experimentation, not
all of which is worthy of canonisation. Though Reger quipped that “any chord
can follow another chord”, music that is composed on this basis and still wishes
to retain any functional principles is unlikely to enjoy a celebrated reception
history.223 “Riemann’s harmonic theory acknowledged this fact through its idea
of harmonic function, which asserts that other structures besides, for example,
the tonic triad are capable of communicating ‘tonicness’.” 224 What these “other
structures” are is vital to answer our opening questions: whether there is
evidence to suggest that major-third relatives can be functionally equivalent
and the extent to which such relatives be incorporated into a framework that is
still largely governed by diatonic principles.225
Harrison’s analytical method is similar to Riemann’s in that it consists of
dissolving chords into their scale-degree components. Within triads, SD 1s are
described as “bases”, SD 3s as “agents”, and 5s as “associates.” 226 Yet Harrison
shows how these labels need to be considered in their harmonic context: using a
tonic 6/4 chord as an example, the base of dominant function overrides the
root/base tonic function, and hence gives a dominant “discharge”. In this,
223 Ibid., 1.
224 Ibid., 11.
225 Downes argues that this is the case in Mahler’s music, but whether the same is applicable
to Reger, or even Franck, is arguable.

59
Harrison emphasises the importance of inversions and scoring in analysis. 227
(We must remember that harmonic ambiguity may be a quality desirable to a
composer, and it is quite possible for, say, a 6/4 chord to have both tonic and
dominant function.) Harrison concurs with Riemann’s delegation of I, IV and V
being primary triads, functioning in the ways we are familiar with, and also
examines secondary triads, explaining that they are functionally mixed
structures that are able to “communicate more than one function.” 228 For
example, E minor may be v/v/v/V in the context of C (i.e. it may participate in
the cycle of fifths) but when it functions as iii the effect – its expressive aura – is
very different. Kurth discusses this with the idea of the “absolute progression”
and “absolute effect,” arguing that a chord’s effect depends on its relationship
to the chord before it. It can be argued that I-iii-vi-ii-V-I is expressively a
different order to I-iii-I: minor chords may reveal their potential morbidity
when cut free of cycle of fifths function. Kurth states that:
This last phenomenon, the luminous effect of an individual
progression, the appeal resulting from the collision of two chords, is
concealed in simple tonal music, [is] more suppressed and
inconspicuous in the overall effect of the harmonic context, being more
absorbed in the tonal structure of the total complex. . . . in other words .
. . the relation of a harmony to the tonic is more significant than the
relationship in the progression of two successive chords individually.229
However, as well as a chord’s relationship to a referential tonic and its
relationship to the harmony immediately preceding it, there is the
consideration of a chord’s “sonic appeal . . . by its absolute effect.” 230 In
straightforward tonal music this absolute effect is masked by relative
227 Ibid.
228 Ibid., 60.
229 Lee A. Rothfarb, Ernst Kurth: Selected Writings, Cambridge: Studies in Music Theory and
Analysis (Cambridge: Cambridge University Press, 1991), 119. It is also arguable that such
movement will also affect the chord appearing after it, depending, of course, on context.
230 Ibid.
60
relationships, but it is possible – and indeed more prevalent in chromatic music
– for the more local effects to be stressed:231
When a harmonic style reaches the point of highlighting the effect
inherent in the unmediated collision of two chords as a specific effect in
itself, we then have a phenomenon that, in its content and significance,
lies on the same evolutionary path as the third evolutionary stage, the
highlighting of an individual chordal, sonic quality as such . . . [it
becomes a] preliminary step in the escalation toward the absolute
individual effect.232
As for when to label an effect as “absolute”, Kurth suggests there is an
“absolute progression effect” when such a progression is “thrust outward – i.e.,
in relief against the surrounding context.” He suggests this is (at least in part)
due to the Romantic period’s “delight in the sonic appeal itself.”233
Returning to Harrison, in regard to mediants, he writes that their function
depends largely on their inversion and structure: if SD 3 is the root-base its
function is “strongly tonic”, whilst SD 5 and 7 (agent-associate) are “strongly
dominant.” 234 In a similar fashion, Harrison describes how the diminished
seventh can convey both dominant and subdominant functions, depending on
which tones are prominent and how the inversion is manipulated.235
231 Ibid., 120.

232 Ibid.
233 Ibid., 121.
235 Ibid.
61
Figure 2.11: Harrison – disassembly of a mediant236
Figure 2.12: Harrison – disassembly of a diminished seventh chord237
We might also note that such chords can be used as rootless German sixths or
dominant minor ninths. In practical application, it is most likely that the
dominant agent and associate will aurally outweigh the subdominant base and
agent, but the subdominant function is present nonetheless.
Many of Harrison’s ideas are useful to the methods used here, particularly the
importance he grants to individual scale degrees and the roles they play in the
determining of harmonic movement and perception of tonicity. His ideas about
mediants reinforce those of Riemann, who wrote about their ability to convey a
sense of tonicity despite being secondary triads. However, one concern with
Harrison’s theory lies with his concentration on harmony, which he justifies by
describing how the New German School did not find parallel fifths problematic
236 Ibid., 61.

237 Ibid., 65.
62
because their primary concern was harmony, not voice-leading. Yet it is
precisely voice-leading properties that encourage us to hear tonicity – within
cadences and elsewhere – and hence they must remain a vital consideration,
along with harmony, for this form of analysis. Salzer reinforces this concept:
In separating harmony from counterpoint, we shall find that harmonic
progressions contain a contrapuntal feature, the simultaneous motion
of several voices. The converse does not hold true, however, because
contrapuntal progressions can be studied in pure form without any
interference from harmonic considerations. This appears quite logical if
we realize that harmonic progressions came to be used only after great
contrapuntal experience had been acquired during the course of several
centuries.238
Additionally, the various interpretations of secondary triads depend on voice-
leading properties to make their “meaning” known. Because of this, both
harmony and voice-leading will be considered in this thesis. With this in mind,
we must also consider chords larger than triads – those with added sixths and
sevenths – in isolation.
Chords – Added Sixths and Sevenths (ii)
Harrison is also concerned with the role of the augmented sixth chord. He
describes how the chord resolves in the same way as the diminished fifth or
seventh, concluding that all three chords share “at least some” inherent “tonal
propulsion.” 239 As we know, diminished fifths and sevenths occur naturally
between scale degrees, playing “position finding” roles (that help define a sense
of tonicity) therein. Schenker describes these intervals as “univalent.”240
238 Felix Salzer, Structural Hearing: Tonal Coherence in Music (New York: Dover Publications,
1952), 51.
239 Daniel Harrison, "Supplement to the Theory of Augmented-Sixth Chords," Music Theory
Spectrum 17, no. 2 (1995): 172.

240 Ibid., 173.
63
Figure 2.12: Harrison – intervallic resolutions241
Conversely, augmented sixths do not appear
naturally between scale degrees, and this compels
Harrison to describe them as “nonvalent.” 242 This
definition suggests the augmented sixth’s potential
for position-finding function is very different.
Harrison argues for any chord possessing an
augmented sixth interval to be considered as such
(not just the ethnic varieties) and also prescribes
predominant (dominant preparation), dominant and
subdominant qualities on all these augmented sixths, depending on which scale
degrees are involved and how they are used. The potential for the augmented
sixth to act as a dominant is of particular interest as augmented sixths are often
present in Franck’s music, and the implications of these will be explored further
in the following chapters. The idea of an augmented sixth chord substituting for
the dominant is also present in Schoenberg, who gives such a progression as an
example in Theory of Harmony. However, Schoenberg seems to consider the
progression coherent because it is an “abbreviation” of a cadential “cliché”,
rather than because augmented sixths contain special voice-leading properties –
similar to the abbreviation of II or ii resolving directly to I without the
intervening V. 243 Nevertheless, his assertion that “such abbreviations can in
general be undertaken only with progressions that have a definite function,
hence, primarily in cadences” ties in with Harrison’s ideas.244
241 Ibid., 171.

242 Ibid., 173.
243 Schoenberg, Theory of Harmony, 359.
244 Ibid., 360.
64
Adrian Child’s work on seventh chords is also useful for this discussion. 245
Childs notes that “while the analytical insights [of Neo-Riemann analysis] have
proved rich and stimulating, a fundamental problem has also arisen: the
composers whose works seem best suited for Neo-Riemann analysis rarely
limited their harmonic vocabulary to simple triads.” 246 Indeed, no analyst in this
chapter has yet discussed the voice-leading nature(s) of seventh chords. Childs
suggests that a new model (to add to Neo-Riemann analysis) is needed, one
that incorporates seventh chords, particularly dominant and half-diminished
sevenths. He then examines extracts by Wagner and Chopin. Both passages use
chains of seventh chords, and Childs demonstrates how sevenths contribute to
the efficacy of the voice-leading. A straightforward Neo-Riemann analysis
would most likely focus only on the triads, but Childs argues for a new model
which incorporates sevenths into analysis. 247 His model has potential for
describing different types of seventh chord transformations, but he uses only a
few, short examples to illustrate the potential for analytical insight.
Nevertheless, Childs reminds us that when discussing music of this type,
sevenths must be awarded the special consideration they deserve, particularly
in passages which are resistant to purely triadic analysis or which involve
chains of sevenths. However, it is worth noting that the roles of seventh and
seventh-less dominant could not be effectively reversed in their respective
contexts. According to Downes’ axial theory, the addition of a seventh to a
perfect cadence “combines a subdominant function to the prevailing dominant
functions already present.” 248 This is not to say that V7 – I is a weaker
progression than V – I: rather that the effect is a different one.
245 Adrian P Childs, "Moving Beyond Neo-Riemannian Triads: Exploring a Transformational

Model for Seventh Chords," Journal of Music Theory 42, no. 2 (1998).
246 Ibid., 181.
247 Ibid., 182.

65
Matthew Bribitzer-Stull also contributes to these ideas with an article on what
he calls “The A-flat-C-E Complex.”249 Bribitzer-Stull is primarily concerned with
the effect an augmented triad can have on the background structure of a work.
He suggests that although two keys within the A-flat-C-E axis can be
incorporated into a diatonic framework and read as mixture, when all three are
present and are not “anchored by a diatonic Stufe” it is possible for the
background structure to be disturbed.250 It is less the chord itself and more what
it leads to that determines a chord’s role in this context. He also notes that
disrupting the diatonic Stufe is a significant step towards atonality because of
its symmetry, and uses Liszt’s “Nuages gris” as an example.251
Some Methodological Issues
At this point it is worth examining what an axial-hexatonic system is able to
accomplish. Many writers examined here seem to agree on the ability of axial-
hexatonic systems to convey functional equivalence, or at least “tonal
neutrality”: the ability for music to modulate without sounding as if it has
disturbed tonal function.252 But what are the differences between these writers,
and what parameters are appropriate for this study?
249 Matthew Bribitzer-Stull, "The A♭-C-E Complex: The Origin and Function of Chromatic
Major Third Collections in Nineteenth-Century Music," Music Theory Spectrum 28, no. 2
(2006).
250 Ibid., 171. Stufe is a Schenkerian term meaning “a triad whose root is located on a scale of
perfect fifths emanating from a tonic triad.” Robert Snarrenberg, "Schenker, Heinrich," in
Grove Music Online. Oxford Music Online.
251 Bribitzer-Stull, "The A♭-C-E Complex: The Origin and Function of Chromatic Major Third
Collections in Nineteenth-Century Music," 177.
252 Charles Rosen, The Classical Style (London: Faber and Faber Ltd., 2005), 272. Rosen notes
that in Beethoven’s ninth Symphony, “Beethoven replays the main theme on B-flat major, the
submediant (which he will again use as a dominant), but in such a way that it still sounds as
if the tonality had not left the tonic D minor.” [My emphasis.]
66
Downes, Cohn and other writers state or imply tonal neutrality as the basic
function of axial relatives, though some writers are more overt in this than
others. In analyzing Schubert’s piano sonatas, for example, Downes states that:
For Schubert, the tonal interval of the major third clearly had the
capacity for what could almost be called passive modulation (my
emphasis.) This property was exploited to increase the breadth and
scale of some of his later sonata form works, without jeopardizing the
set functions of the various structural sections.253
Similarly, Bribitzer-Stull notes that in [the A-flat-C-E complex], chromatic thirds
could be “strung” from and to the tonic. This happens in Chopin’s Polonaise,
Op. 53, in which the tonic A-flat is prolonged (my emphasis) first by its upper
third, C, and later by its lower third, E.254 Kopp notes similar prolongations in
the works he examines, including Schubert’s fourth Symphony and Chausson’s
Piano Trio No. II.255
For Cohn, in contrast, hexatonic modulation is perhaps more ethereally
suggestive than functionally prolongational. Regarding the effect of modulation
within a hexatonic system, Cohn suggests that motion between hexatonic poles
is “frequently affiliated, by both composers and listeners, with an ethos of
uncanniness.” 256 Examples from Wagner’s Parsifal and Mahler’s second
Symphony are used for illustration; both extracts concerned with the spiritual –
in the first, the removal of Kundry’s soul from her body, and the second dealing
with “resurrection”. Cohn considers the juxtaposition of hexatonic poles as
especially significant, as the juxtaposition “resists interpretation in terms of
Downes, "An Axial System of Tonality Applied to Progressive Tonality in the Works of
253

254 Bribitzer-Stull, "The A♭-C-E Complex: The Origin and Function of Chromatic Major Third
Collections in Nineteenth-Century Music," 177.

67
diatonic tonality.”257 This may be true, but despite the triads III and ♭VI not
being diatonic, they are often incorporated into largely diatonic compositions
all the same, suggesting they are desirable tonal areas for other reasons.
Perhaps one of these reasons is the parsimonious voice-leading they enable, or
functional equivalence. Earlier, I mentioned Cohn’s concern with the ability of
tonal poles to “annihilate tonality”. Though this might be temporarily true in a
piece of music, usually it will only be for a specific case within a wider axial
structure, as hierarchy can also be established rhythmically using phrase
structure.258
Figure 2.13: Pitch-class set [4, 5, 8, 9, 12, 13]
Nonetheless, we can illustrate this
relationship by creating a pitch-class set
using the tones of D-flat and A minor: G-
sharp – A – C – C-sharp – E – F.259 Using these notes we can create pairs of
triads that have two common tones, one common tone, or no common tones,
constituting three levels of triadic potential. This construction is reinforced by
the wholetone nature of the fundamental bass between two major third related
chords within a pitch class set: there is no way of telling whether chords have
two, one or no notes in common in the absence of a non wholetone bass, or
which has a dominant or subdominant function one relative to the other. I
surmise, then, that polar relationships represent the non-plus-ultra of
ambivalence within the larger framework of tripartite axial ambivalence, but do
not annihilate tonality outright. However, they come close to doing so, because
such a pc set almost amounts to a triple Schreckensfanfare – the tonality-
257 Ibid.
258 If phrasing and rhythm lack regularity, it may well be possible that tonal poles can
annihilate tonality. However, in the works that both I and Cohn examine, this is rarely (if
ever) the case.
259 This pc set is used because it is the same one Franck uses in his Quintet, analysed in
Chapter Six.
68
obscuring opening to the finale of Beethoven’s Ninth Symphony. The
Schreckensfanfare has the tones A-B-flat-D-F, which constitute both a major-
seventh chord and a double tonic complex of B-flat/D minor (the opposing keys
of the Ninth Symphony). The pc set here can comprise three Schreckenschords –
A-C-E-F, C-sharp-E-G-sharp-A and F-A-flat-C-D-flat – and the ability to
befuddle tonality is certainly present.
However, the fourth level of triadic potential – further than those chords which
have no common tones – is represented by two augmented triads (here, G-
sharp-C-E and A-C-sharp-F). Cohn does not discuss this potential in relation to
pc sets and polar relationships. We know that these triads can – on a localised
scale – annihilate tonality. Take, for example, the locus classicus of augmented
triads: the opening of Liszt’s Faust Symphony. This opening famously
comprises four augmented triads comprising two hexatonic pc sets. Each acts as
a subdominant of the chord which follows as the notes fall by semitone.
Tonality is indeed annihilated in this instance: we have no way of interpreting
the chords as anything but subordinate to the voice-leading processes of chords
which precede them.
As well as differing in descriptions of their systems, Downes and Cohn also
differ in terms of the musical works they choose to illustrate their ideas.
Downes considers Beethoven and Schubert the most important instigators of
the axial system, discussing Beethoven in detail and considering works of Liszt
and Brahms before finally examining Mahler, the focus of his thesis. Cohn,
however, is primarily concerned with the work of Schubert and later romantic
composers such as Liszt, Franck, Wagner and Mahler. Though the analysis
behind Cohn’s work is impressively detailed and at times extremely complex in
regard to illustrating voice-leading relationships, he seems to be relatively
unconcerned with precedents to a third-related key system. When Cohn

69
mentions precedents to romantic examples of hexatonic progressions in Mozart,
Beethoven and even Bach, he states that in these examples “the triads in the
hexatonic system are lightly tonicised using the standard resources of diatonic
tonality.”260 He then concludes:
A related way to constrain the domain of hexatonic analysis would be
to acknowledge that hexatonic elements might infiltrate compositions
that otherwise operate according to the principles of diatonic tonality,
but to limit the application to elements of these compositions that fail
the standard test of diatonic coherence.261
Though Cohn admits this approach may be limited as hexatonic and diatonic
analyses should not be mutually exclusive, given that he restricts his analysis to
examples from Schubert onward we might reasonably assume he considers
works prior to Schubert “diatonically coherent.” This brings up an important
point – what exactly is diatonic coherence, or lack thereof? Could it be
considered the need to explain a key area or progression according to chromatic
principles, regardless of whether the chromatic principles are in the fore or
middle ground of a given work? With this definition I would describe
Beethoven’s Hammerklavier Op. 106 as “lacking diatonic coherence” – how are
we to explain the third movement’s key of F-sharp minor, when the tonic of the
Sonata is B-flat? These keys are on completely opposite sides of the circle of
fifths, our traditional roadmap of diatonic key relations.
Perhaps works prior to Schubert require a new evaluation of their diatonic
coherence based on axial-hexatonic relationships, and perhaps Cohn has not
fully considered the significance of hexatonic progressions in Beethoven’s
work, which – particularly in the case of the Hammerklavier – become more
261 Ibid.
70
logical when analysed in an axial context. Furthermore, Cohn seems to regard
the main effect of hexatonic progressions to be an evocation of the “uncanny”
or an allusion to undertones of magic, and does not expound the idea that such
progressions may play a crucial role in expanding and developing sections to
unprecedented lengths whilst maintaining tonal neutrality. 262 To put it an
alternate way, Schubert’s tonal practices (and he is but one example) seem to
indicate the functional as much as the expressive potential of axial systems, as
well as the use of axial systems to evoke “heavenly length”263, otherwise known
as monumentality.
Theoretical integration and Methodology
Both Downes and Cohn, our predominant analysts, have much to offer an
examination of Franck’s work. Axial analysis is predominantly focused on
background relationships and how they are integrated into other facets of
composition, from macro to micro events. As such, key tables will be included in
my analyses to demonstrate works’ broad harmonic outlines and how they
adhere to axial relationships (as they often do). However, Cohn’s discussions of
pitch-class sets also have much to offer. Cohn’s perspective reminds us that we
experience music in a temporal sense, and that as such foreground events are
crucial to our experience and understanding of all music. As an analyst, one
wishes to represent what is actually heard or experienced, not construct a
discussion which bears little relation to the music itself. In Franck’s Quintet
(discussed fully in Chapter Six) the tonal poles (which Cohn discusses) form a pc
set which is integral to understanding the unity in the movement overall. Cohn’s
suggestion that the tonal poles “annihilate tonality” forms a springboard from
which tonal hierarchies – amalgamated with axial theory – are discussed. Hence,
263 Berthold Hoeckner, Programming the Absolute: Nineteenth-Century German Music and the
Hermeneutics of the Moment (Princeton: Princeton University Press, 2002), 63.

71
Cohn’s (and other theorists’) ideas will be used for discussion in the body of the
text following the axial key tables. Through this, I aim to demonstrate how these
complementary theories can work together. Before the analysis of Franck’s works
can begin, however, it is apt to consider the influence Beethovenian models may
have had on Franck’s compositional processes.
72
Chapter Three
The Eroica Symphony and the Problem of the Finale
This chapter examines Beethoven’s Eroica Symphony and the Problem of the
Finale. Before I discuss this, however, a disclaimer and justification is
appropriate, to explain why Beethoven is considered such an influential figure in
this study. His prominence needs to be justified both in terms of his influence on
Franck and use of axial-type procedures generally. Firstly, there is evidence that
Franck was familiar with at least some (if not all) of the Beethoven works
examined here; those we can be certain of include the String Quartet Op. 127 and
the Ninth Symphony. It is also likely that Franck was familiar with most (if not
all) the late-Beethoven String Quartets, as they regularly featured on the Société
Nationale’s concert programs. 264

Furthermore, Franck’s comments on the
composition of String Quartets imply a preoccupation with the significance of the
genre (and hence what had gone before). Franck was convinced: “(and several of
his students garnered his opinion on this point) that to risk writing a string
quartet was a perilous enterprise that one could attempt only later in life.”265
It is also important to note that axial-type procedures did not necessarily begin
with Beethoven. There are instances of Haydn and Dussek (and others) using
third-related keys as secondary key areas in sonata form works, including (but
not restricted to) Haydn’s String Quartet Op. 20/4 and Dussek’s Piano Sonata No.
26 Op. 70. Stewart-MacDonald suggests such major-third digressions,
particularly those to the “flat side” (frequently Iβ) strengthen a tonic return when
264 Michael Strasser, "Art as Teaching: Vincent d’Indy and Programming at the Société
Nationale after 1886 " in Vincent d’Indy et Son Temps (Sprimont: Mardaga, 2006), 140.
265 Joël Marie Fauquet, César Franck (Paris: Fayard, 1999), 657.
73
it occurs, as flat-side exploitations evoke a “strategic withdrawal of energy”.266 It
seems certain that these composers pre-dating Beethoven were aware of the
potential inherent in axial-type modulation. I argue, however, that whilst these
earlier composers did use axial tonality in this way, their composition outcomes
are not comparable to Beethoven, who not only used axial tonality but fully
realised its potential as a manipulative agent of tonal drama. So, while these
earlier classical composers had begun to use such sonorities, the potential of this
type of tonal system was not fully realised until later. The present chapter and
the next discuss some of Beethoven’s late period works more fully, illustrating
the importance of such works in their respective genres. Franck’s knowledge of
these seminal Beethoven works makes an examination of Beethoven’s oeuvre not
only desirable, but mandatory.
Beethoven’s late period was one of experimentation, where he often departed
from the conventional four-movement model usually used for symphonic and
chamber works. This is not to say that his earlier works were not also formally
innovative, but in his last period the preoccupation with form became more
acute; he seems to have sought more varied and radical solutions to the finale
problem. End-weighted works such as the Ninth Symphony and the String
Quartet Op. 133 with its Große Fuge suggest by their size and demeanour that
Beethoven was more conscious than ever of the problem of the finale. The works
examined in this chapter see Beethoven adopting axial tonal schemes to effect
more radical solutions to the problem of how to distribute weight across a multi-
movement work, and through examining them we can discern several trends
which could potentially influence those composers who followed him.
Rohan H. Stewart-MacDonald, New Perspectives on the Keyboard Sonatas of Muzio Clementi

266
(Bologna, Italy: Ut Orpheus Edizioni, 2006), 83.

74
One of the main concerns of this last period was the distribution of weight –
emotional or durational in some works, but particularly tonal weight. Hence we
often encounter Beethoven reserving the structural dominant until late in a multi-
movement work. How and when Beethoven does this requires terminology that
accurately describes the aural experience and compositional idea. This thesis
posits the terms “immanent dominant” and “transcendent dominant.” 267
“Immanent” is used here to describe the strongest positioning of dominant
within a single movement, such as the structural dominant in the first movement
of the Eroica. The first movement affirms the tonic clearly, and the following
movements are proportional in relation to each other by the standards of the
classical period, despite the Eroica’s enlarged proportions. “Transcendent
dominant” refers to an overarching dominant of the entire work, such as works
in which the central movements posit a structural dominant (“transcendent”) of
the work overall. Often works in which the immanent dominant is withheld 268
feature moments in which the dominant is briefly touched on, such as at
significant structural points. I highlight instances of this in the following analysis,
in this and later chapters. “Transcendent” may seem problematic as a term, but
taking the verb “transcend” we can understand its meaning as “be or go beyond
the limits of (something abstract, typically a field or division).” 269 By contrast
“immanent” is defined as “existing or operating within; inherent”. 270 In the
examples we will examine, we will see that Beethoven and others alter the
weighting of multi-movement works by altering the placement and shaping of
dominants both within single movements and works as a whole. As such, we can
267 I am aware of the use of these terms in a theological sense, but I consider these terms the
best fit for the meaning I am trying to convey. Hence the above discussion might be
considered a reshaping for a musicological audience, as the terms convey the sense of
significance within the concepts discussed.
268 Such as Beethoven’s Hammerklavier Sonata Op. 106 or the Ninth Symphony.
269 "Transcendent", in The New Oxford Dictionary, ed. Judy Perasall (Oxford: Clarendon Press,
1998).
270 "Immanent", in The New Oxford Dictionary, ed. Judy Perasall (Oxford: Clarendon Press,
1998).
75
consider “immanent” and “transcendent” to be appropriate terms for describing
how the dominant operates within any given work.
The Eroica Model
With these terms in mind, we begin by examining the Eroica model. Despite its
inherent problems, the Classical trope of the regular sonata form movement as
expanded by Beethoven significantly influenced some composers, particularly
those writing shortly after his death. Here we are able to distinguish between
those who followed immediately after Beethoven, and the “second age of the
symphony”. 271 Dahlhaus discusses this at length, and notes that in Oper und
Drama Wagner declared the death of the symphony, “viewing the post-
Beethovenian efforts as a mere epilogue with nothing substantially new to
say.”272 Yet, French Nationalism also plays a part here: the establishment of the
Societe Nationale in 1871, with its slogan ars gallica:

sought refuge amidst the French catastrophe [and] was almost insolubly
ambiguous. In order to found a national symphonic and chamber music culture
in opposition to their exclusive orientation in opera, French composers had to
appropriate and recast a specifically German tradition – namely, Viennese
classicism, a tradition upheld in Paris by Saint-Saëns himself.273
It is unlikely to have escaped these composers – Franck among them – that if they
were to say anything new, they must draw inspiration from the late Beethoven
works, rather than the middle period ones which were largely responsible for the
“mere epilogue” Wagner refers to. As the works examined in this thesis are
almost exclusively composed post-1870, we are dealing with the “second age of
the symphony”, and with that, the influence that such a second age could have
on other genres. Because of this, we must examine what occurred in the period
272 Ibid.
273 Ibid., 263.
76
between Beethoven’s death and the revival of the Symphony in the 1870s; hence
the examination of the Eroica model.
Beethoven must have realised – or so it would seem from his late-period
experimentation with form – the problem that arose from his otherwise
successful expansion of classical form, and he attempted to create alternatives in
the models we explore later in this chapter. Whilst it is understandable that some
composers took the great middle-period works as exemplars (such as Schubert, a
representation of whose work is examined shortly), many realised the problems
this would create and were influenced instead by the formal innovations of
Beethoven’s later works. Others attempted to turn away from Beethoven
altogether.
It is relevant to examine the Eroica itself together with a work which uses that
model: a regular sonata form movement by Schubert that possesses the Eroica’s
famously expanded proportions. The results of this analysis may be useful to
compare with those which follow different models, and will be examined in later
chapters.
At this point it is important to note the usefulness of analysing works in relation
to “models”. I suggest that, for each of the works discussed as examples of a
model, the model is not a blueprint or monumental work that lends itself to
apologetic parodying. Rather, these chapters and their models are grouped
according to the guiding idea that gives the overall impression of the work,
particularly in regard to immanent and transcendent dominants. Schubert’s
Ninth Symphony, for example, is not a misunderstood Eroica Symphony, but
rather shares important structural and formal characteristics with the Eroica
model. In line with this, each sub-chapter will include a summary of the defining
features of each model, followed by analysis and discussion.

77
Beethoven’s Eroica
It is worth examining the Eroica Symphony to ascertain just what features are
significant to our model, though the main features will already be known to us.
Although the Eroica was a landmark work in its time, it still fulfils Classical
expectations in terms of Sonata form, though expanded in all parameters (and
not only the first movement, but those that follow). The first movement is 846
bars long if the expository repeat is observed; which, it seems, was Beethoven’s
final intention, and justifiable in that the numerous ideas require re-hearing for
comprehension.274 That the first movement is a thoroughly satisfying statement in
itself is almost inarguable, but just how did Beethoven achieve such a statement?
A. Peter Brown writes:

In this first movement, Beethoven mustered all the resources at his disposal to
present a large and thoroughly coherent statement. Carefully weighed
articulations lend coherence to a number of discrete thematic ideas, continuations
are built from developmental expansions, and rhythmic and harmonic
dissonances are appropriately resolved. Even the opening hammerstrokes are
recalled within the exposition, development, and recapitulation, and at the
conclusion; this cliché, like every other gesture in this Allegro con brio, becomes
part of an organically convincing essay.275
An oft-cited example of resolution is the C-sharp in the seventh bar of the theme,
which clouds the tonality from the beginning and demands resolution. This is
eventually achieved in the recapitulation, where C-sharp gains new meaning
when enharmonically reinterpreted as a D-flat.276 Beethoven exploits the “pun” of
the German sixth and dominant seventh, as in the recapitulation the D-flat
resolves downward to C, as opposed to up to D as occurs in bar seven. This is a
wonderful resolution of Beethoven’s earlier audacity.
274 A. Peter Brown, The Symphonic Repertoire: The First Golden Age of the Viennese Symphony:
Haydn, Mozart, Beethoven and Schubert vol. II (Bloomington & Indianapolis: Indiana University
Press, 2002), 461.
275 Ibid., 468.
276 Charles Rosen, The Classical Style (London: Faber and Faber Ltd., 2005), 393.
78
Any discussion of this movement cannot be complete without touching on just
how much the “heroic” aspect captured the imagination of contemporary
listeners and continues to capture our imagination today. Scott Burnham’s
Beethoven Hero discusses this at length. He writes: “Beethoven’s first movement is
. . . expressive of an almost universally accessible psychological process: a
dangerous yet necessary exploration of some unconscious aspect of the psyche is
followed by a tremendous sense of reintegration and affirmation.” 277 Yet listening
to Beethoven’s late works sometimes gives us another sense entirely – of
something dangerous but necessary being wilfully avoided, which we might also
describe as dramatisation of inaction (or blockage) in the late works as opposed to
the dramatisation of action in movements like the first movement of the Eroica.
Given that one of the focuses of this thesis is the distribution of symphonic
weight, the movements following the first of the Eroica must be examined: what
can possibly seem fulfilling after such an affirming psychological process as the
first movement? Perhaps because the projecting of programmes onto the first
movement has been so popular, the “Marcia funebre” has been considered “a
logical sequence to the ‘heroic’ character of the Allegro con brio.” 278 This is also a
large movement, a rondo in C minor. What the first movement emphasised in
rhythmic displacement, the second emphasises in metric regularity; a complete
rhythmic contrast.
Some of the keys used in the development of the first movement are expounded
in the first part of the second – C, E-flat and F, whilst G minor is also used later.279
The second episode begins with a double fugue which increases the tension
277 Scott Burnham, Beethoven Hero (Princeton, New Jersey: Princeton University Press, 1995),
23.
278 Brown, The Symphonic Repertoire: The First Golden Age of the Viennese Symphony: Haydn,
Mozart, Beethoven and Schubert, 469.

279 Ibid.
79
significantly whilst simultaneously retaining tonal flatness (which is partially
why fugue as a form is so useful; see the discussion of fugue in the analysis of
Op. 131 later in this chapter).280 A. Peter Brown writes:

The tonal plan interlocks the funeral march with the first movement, where the
key of C became a well-articulated tertiary area. In the funeral march, E-flat
major occurs twice (mm. 17, 181) and once in the second episode (m. 135) in a
tonally solid context. This cyclic coherence is not the last of its sort to be found in
the symphonies of Beethoven (e.g., Nos. 5 and 7), but it certainly is a convincing
occurrence.281
As with all music, the contrast between unity and diversity is an ongoing
balancing act, but this coherence suggests Beethoven’s preoccupation with
overarching tonal unity was in place relatively early.
The Scherzo, back in the tonic E-flat, is initially metrically and tonally
ambiguous. Yet whilst the first, second and final movements of the Eroica are
considerably expanded compared to both Classical norms and to Beethoven’s
earlier symphonic counterparts, the Scherzo is only slightly longer. What this
movement lacks in length, however, it makes up for in energy. Rhythmically, this
movement reactivates the 2/4 3/4 opposition that permeates the first movement
so strongly.
Figure 3.1: Beethoven Symphony No. 3/iii (Op. 55), bars 1 – 7
280 Ibid.
281 Ibid.
80
Beethoven groups the crotchets in pairs at the opening and only reveals his triple
time at bar five. The Alla breve at bar 381 (the downward arpeggiating minims,
previously minim-crotchet) is the logical reason why Beethoven composes out
the Scherzo as opposed to employing a Da Capo repeat; it “corrects” the
rhythmic argument that begins the movement. The short length of the Scherzo
also contrasts with the longer finale, which is more summative.
The finale is of particular significance for those of us who are concerned with
symphonic balance. Here, Beethoven combines variation form with the Sonata
Principle; the effect is a “musical struggle between melody and bass”: 282 an
introduction, strophic variations and coda bookend developmental variations. In
this though, it seems a mere echo of sonata, in that expository and recapitulatory
events are bookending developmental ones but eschewing the tonal opposition
of sonata. As Rosen writes:

a finale demanded a simpler and less complex form than an opening movement:
that is why it is generally a rondo, or a set of variations. . . . If it is a ‘sonata,’ then
it is necessarily a squarer and simpler version of that form; . . . but , in any case,
the thematic material of a finale is always rhythmically squarer than that of the
first movement, the cadences heavily emphasised, the phrases well-defined, and
the first theme completely rounded off before any harmonic movement can take
place.283
The Eroica finale certainly fits this description; it also embodies a grounding of
tension of the work overall. The following table outlines the structure more fully
and illustrates the keys used in each section.
282 Ibid., 475.

283 Rosen, The Classical Style, 275.
81
Figure 3.2: Beethoven, Symphony No. 3/iv (Op. 55). Adapted from A. Peter
Brown284
Symphony No. 3, Fourth Movement, Opus 55
Introduction: G minor to E-flat
Part One: Exposition (Strophic)
Bass (m.12) [I]
Var. 1 (m.44)
Var.2 (m.62)
Var.3 (m.78) = Bass + Theme
Transition to C minor (m.115) [ivα]
Part Two: Development (Non-Strophic)
Var. 4 (m.117): Fugato on Bass, C minor [ivα]
Var. 5 (m.175): Recapitulation of Theme and Var., D major [Vα]
Var. 6 (m.211): March, Ostinato on Bass, G minor [iα]
Var. 7 (m.258): Recapitulation of Theme and Var., C major [IVα]
Var. 8 (m. 276): Fugato in inversion of Bass and Theme, E-flat major
Part Three: Recapitulation (Strophic) [I]
Poco Andante
Var. 9 (m. 349): Theme only/Bass gone
Var. 10 (m.381): Theme in Bass
Closing (m.398): Modulatory
Coda
Brown, The Symphonic Repertoire: The First Golden Age of the Viennese Symphony: Haydn,
284
Mozart, Beethoven and Schubert, 461. ‘To’ is used where there were originally arrows.
82
Introduction (m. 431)
Fanfare Transition of Theme (m. 435)
My notes in square brackets interpret the keys of the Eroica finale in an axial
context, showing a prevalence of subdominant axis members. This often occurs
in second movements following a weighty sonata form first movement, but here
occurs in the finale: why this may be is examined later.
Note in particular variation seven: its harmonic progression is uncannily similar
to that of the first movement coda. In both cases the key “behaves itself” by
progressing through the cycle of fifths to E-flat. A similar allusion can be drawn
between the opening of the first movement’s development and the finale’s fourth
variation: C minor proceeds through D minor and G minor, before the dissonant
climax (F major7 in first inversion – bars 276 – 279) and the remote E minor are
reached midpoint in the development. In this sense, the finale seems to be a
compression of the important key relationships of the first movement: a
remembering of the dangerous then safe expeditions from the same source – C.
Christopher Lewis believes the finale of the Eroica to be one of the most
“extraordinary” symphonic finales: the others being Beethoven’s Ninth and
Brahms’ Fourth.285 Lewis suggests this is largely because the movements are in
variation form – sonata form’s “artistic antithesis” – but that another type of
duality is grafted onto the variation scheme to evoke drama: in the case of the
Eroica, Lewis believes this duality to be thematic. Yet this does not effectively
encapsulate the other aspects of the movement which also contribute to this
“extraordinariness.” Brown writes: “Beethoven captures the heroic character [of
the variation theme] at the end with a summary of its various facets and a mighty
Christopher Orlo Lewis, Tonal Coherence in Mahler's Ninth Symphony (Ann Arbor,
285
Michigan: UMI Research Press, 1984), 102.

83
transformation of the Prometheus melody.”286 Yet the power also lies in the unity
of the symphony as a whole, and some of the final loose ends of unification are
left until the finale. Brown writes:

The unity of the “Eroica” Symphony goes far beyond its topic and principal
melodic material. The openings of each movement unfold in an extraordinary
manner. The primary, secondary and tertiary keys of each movement are
manipulated to fit into a larger cyclic complex. Previous to the “Eroica,” unity of
key was a given; it now becomes an organic concept. The individual movements
are stretched over wider expanses by avoiding complete closure and by delaying
the realizations of earlier implications. It is this inevitability of Beethoven’s ideas
in the “Eroica” and their extensions that set a new concept for symphonic
composition.287
However, Lewis is correct that even an “extraordinary” symphonic finale (such
as the examples he mentions) is not simply a re-conception of the symphony in
which the high point is moved to the end: the first movement is still the
weightiest because of the sheer tonicising force of sonata form – the Eroica, and
other works which use this model, posit the immanent dominant in the first
movement. As this immanent dominant has occurred, no transcendent dominant
action is necessary. Furthermore, the success of a finale of this type also relies on
the acceptance of the status quo: that is, the Eroica finale accepts the tonic – the
tonicisation of which has occurred in the first sonata form movement – and does
not attempt to dramatise the process again. The movements cited by Lewis
modulate little (or not at all in the case of Brahms’ fourth) and utilise tonic
together with more neutral subdominant axis members when they do. The
employment of variation form reasserts the notion that a finale is a resolution of
an entire work, due to its “flat” tonal state. However, Beethoven’s Ninth is a
special example: it is a work in which the finale cannot be discussed without a
287 Ibid.
84
full understanding of the compositional procedures going on in the earlier
movements. However, the problem that Beethoven was grappling with in the
Ninth was one of symphonic weight.
The Distribution of Symphonic Weight
When discussing the Eroica, it becomes evident that one of the features that
preoccupies analysts is the sheer size and impact of the work. Monumentality
becomes a crucial issue for nineteenth century composers, particularly those of
symphonic works, with the seeds of the issue evident even in Mozart.288 With the
will to monumentality grew the problem of the finale (itself a subset of the
Doctrine of Originality as espoused by Dahlhaus).289 However, by the second age
of the Symphony, composers felt the finale problem so strongly that many turned
to one movement form, such as Liszt with his Symphonic poems. Wagner stated
that if he were to compose symphonically after Parsifal, then the works must be
in one movement because of the obstacle of the finale.290 The crux of the finale
“problem” is that sonata form is such a strong tonicising force in itself that any
following movements lack tonal drama in comparison.
Classical sonata form multi-movement works had always been first movement
heavy, but the balance issue only arose when first-movement sonata form was
expanded to the lengths which we see in Mozart (see the following footnote) and
in Beethoven’s symphonies and chamber works. Discussing Mozart, Rosen
explains: “The problem of the finale is naturally one of weight, of sufficient
seriousness and dignity to balance the opening movement, but there would be no
problem at all if it were not for the classical conception of the finale as a
288 See, for example, Mozart’s C major and G minor Quintets and their discussion by Rosen.
Rosen, The Classical Style, 274.
289 Dahlhaus, Nineteenth Century Music, 25 – 27.
290 Lewis, Tonal Coherence in Mahler's Ninth Symphony, 101.
85
resolution of the entire work.”291 Furthermore, “the limit of dramatic complexity
in a classical finale is reached with Mozart’s G minor Symphony: despairing and
impassioned, it is also rhythmically one of the simplest and squarest pieces that
Mozart ever wrote.”292
As we have discussed, many of Beethoven’s symphonic and chamber works offer
“solutions” to the problem. However, Rosen mentions that Beethoven’s coda to
the finale of the F minor Quartet Op. 95 “has often seemed irrelevant, if not
positively frivolous.”293 Rosen’s response to this finale says less about the actual
finale itself than the size and complexity of the movements which it follows. The
finale problem manifests itself most strongly in monumental, emotionally intense
or tragic works. A classical conception of the finale as resolving the tensions of
the work whilst attaining metrical regularity and major modality is simply not
enough to follow the emotional power and complexity that Beethoven was able
to wring out of a sonata form first movement, such as the Eroica, and Kerman
touches on this when he compares Beethoven’s String Quartet Op. 59 No. 1 to the
Eroica:
These are exactly the two main masterpieces of Beethoven, I think, that wrestle
most seriously with the problem of the individuation of single movements. One
feels after the first movement of the Eroica much as one feels after the first or
second movement of the quartet – that so exhaustive an experience has been
conveyed, and in so many facets, as to make any further communication
superfluous. . . the later movements are set awkwardly for the total impression.294
It is important to note, however, that the correlative of the finale problem is the
conception of the symphony as a unit, as opposed to the eighteenth-century
notion of a group of pieces in related keys following a predictable format.
291 Rosen, The Classical Style, 274 – 75.

292 Ibid., 275.
293 Ibid., 274.
294 Joseph Kerman, The Beethoven Quartets (Oxford: Oxford University Press, 1966), 115.
86
Eighteenth-century concerts would regularly break symphonies into various
movements; this is now almost unthinkable nowadays, but it was the norm at
that time.295 Yet, it is likely that by the end of their compositional careers both
Haydn and Mozart were beginning to conceptualise a more teleological
symphonic model, and it is this that Beethoven draws on; more importantly
though, it is this model that mediates our hearing today. As Scott Burnham
writes:
The discomfort of earlier critics [of the Eroica finale] and the belaboured
justifications of later critics are both occasioned by the attitude that a finale must
somehow resolve the entire work. The case of the reception of the Eroica finale
shows the depth of our attachment to the end-orientation model. This way of
understanding Beethoven’s musical process clearly arises from the strong
instances of several of the first movements and from the four-movement design
of the Fifth Symphony.296
In lay terms, this was a problem Beethoven created, and which Beethoven – on
the evidence of his later works – took it upon himself to solve.
The centering of the dramatic weight in a first movement may seem to account
for an unsatisfactory finale, such as in some reviews of the Eroica as mentioned
above. However, Kerman is flattering in his reasoning: regarding Beethoven’s
String Quartet Op. 59 No. 1, he states that by this point in his career Beethoven
was composing “individual movements so terrific that [he] cannot always face
them one to the next.”297 He suggests that the reason for this is the unprecedented
length and breadth of the Eroica, which “marked a . . . radical deepening of the
terms of the problem, which was bound to delay its solution.” Regardless of
details, however, the concept of any sonata form finale is problematic simply if
296 Burnham, Beethoven Hero, 60.
297 Kerman, The Beethoven Quartets, 341.
87
tonicisation has occurred in the first movement. Even if the issue is one of
symphonic weight as opposed to sonata form specifically, front-weighting can be
an issue: Michael Talbot notes that front-weighting works best when “the general
dimensions are small and the disparities not too extreme. Once a listener
becomes consciously aware that a first movement is long, expectations for the
length of the rest are raised.”298 Talbot considers that in (for example) the opening
movements of Bach’s “English” Suites “whilst magnificent in their own right, do
nevertheless tend to overshadow and thereby devalue the rest of the cycle
[emphasis mine].”299 The dissatisfaction we may sometimes sense in a Classical
finale may be because the first movement, however brilliant, has effectively
devalued the rest of the work. (It might rightly be asked whether the works were
conceived as a set, or intended to be played individually, but considering the
pieces as a unit, the observation remains.)
In most cases we can sense which movement possesses symphonic weight
aurally, but it is important to define exactly what we mean by this. Talbot notes
that weight includes aspects of length (potentially in bar numbers but more
importantly in time), instrumentation (in fullness etc.) and tone (or seriousness).
Yet these factors may all be mediated by other elements, and the only
consistently significant feature of movements which possess symphonic weight is
“complexity of form.”300 “The number of significant events in a movement – its
rhythmic harmonic and tonal changes, as well as the thematic and textural
pattern woven by the composer – has relevance for its weight.”301 The attainment
and fleshing out of the structural dominant ties in with these “significant events”,
298 Michael Talbot, The Finale in Western Instrumental Music (New York: Oxford University
Press, 2001), 47.
299 Ibid.
300 Ibid., 45.
301 Ibid.
88
in which case, we might consider the immanent dominant as one of the most
significant events.
However, this is not to suggest that it is impossible to write a fulfilling finale after
an extended sonata form first movement: rather, that expanded dimensions in
previous movements mean a rethinking will be required particularly in the
extended symphonic works of the nineteenth century.
A summary of the Eroica Model’s characteristics:
 Regular sonata form works in three or four movements;
 Structural Dominant (immanent dominant) established and prolonged in
first movement;
 Central movements in tonic major/minor or subdominant major/minor;
 Axial relationships may be present, but are used to embellish/expand tonic
and dominant key areas rather than delay dominant. No transcendent
dominant necessary (though Beethoven experimented with one in the
second Symphony);
 There may be an emphasis on the subdominant, or a
transcendent/immanent subdominant (in the case of the Eroica’s second
movement, there is an immanent subdominant);
 Finale in Rondo/Variation/Modified Sonata form;
 Overall impression is of “Classically” shaped work, formally (first
movement the “heaviest”, finale conclusive but formally “lighter” or
“squarer”).
89
Schubert’s Ninth Symphony
Schumann discovered Schubert’s “Great” C-Major Symphony and famously
reviewed the work in 1840, twelve years after Schubert’s death. 302 Claiming that
“this symphony . . . had an effect among us like no other after the ones by
Beethoven”303, Schumann’s review notes particularly the “heavenly length” of the
Symphony, which is largely achieved through axial expansion of traditional tonal
areas.304 The following analysis by Downes demonstrates this (my modifications
in square brackets):
Figure 3.3: Schubert, Symphony No. 9 (D 944). Adapted from Downes305

[1 – 77 C (with E minor and A-flat) I (with iα and Iβ)]
1 – 133 C I
134 – 173 e – V of G iα – V of V
174 – 189 G V
190 – 227 E-flat-e – V of G Vβ – iα – V/V
228 – 25 G V
254 – 35 development section
356 – 39 C I
440 – 491 c i
492 – 509 C I
510 – 557 A-flat – A – V of C Iβ – ivα – V of I
558 – 685 C I
As mentioned earlier, Downes summarises Schubert’s technique:

For Schubert, the tonal interval of the major third clearly had the capacity for
what could almost be called passive modulation. This property was exploited to
302 Berthold Hoeckner, Programming the Absolute: Nineteenth-Century German Music and the
Hermeneutics of the Moment (Princeton: Princeton University Press, 2002), 63.
303 Ibid., 71.
304 Ibid., 63.
Otago, 1994), 30.
90
increase the breadth and scale of some of his later sonata-form works, without
jeopardizing the set functions of the various structural sections. In other words,
axial modulation allowed the expressive and constructive use of chromaticism at
the deeper structural levels, without jeopardizing the ability of tonality to
articulate form.306
The ability of tonality to articulate form is perhaps one of the best descriptions
possible of sonata form. This movement is an excellent example of expansion via
axial relatives whilst retaining a sense of proportion and traditional key
relationships. E minor (iα) and E-flat (Vβ) are used extensively in the exposition,
whilst A-flat (Iβ) and A minor are used in the recapitulation. Downes notes that
although the use of E and E-flat in the exposition can be interpreted as
Schenkerian third-related dividers, “they exhibit the ability of axial modulation
to instigate structural expansion.”307 Downes also notes that E minor prolongs the
tonic, C, whilst E-flat prolongs the dominant, G. At the end of the first tonal area,
the music lingers on G, the dominant.308 Schubert then falls to E-flat major (Vβ).
The introduction of new thematic material confirms this as a second subject area,
setting up the expected sonata form duality: this is established, and the
immanent dominant is attained. A similar process occurs in Schubert’s C major
Quintet, and the effect of such prolongation is crucial, as Downes explains:

The E-flat episode emphasises the axial substitution at the foreground level as G
major repeatedly threatens to take over. The overall breadth of this section and
indeed the movement as a whole is largely made possible by this axis
substitution and the subsequent inter-axial dialogue between E-flat and G before
the true dominant G is established.309
Though Schubert famously experimented with wrong key recapitulations, the
analysis of this movement suggests that he was primarily concerned with
expansive and colouristic qualities afforded by axial modulation. The consistency
306 Ibid., 32.

307 Ibid., 30.
308 Ibid.
309 Ibid.
91
of Schubert’s major-third relative use suggests that these keys had properties that
prolonged tonal areas as he desired.
However, two aspects are absolutely crucial to this analysis: 1) the structural
“immanent” dominant is attained and substantiated, as in the Eroica model, and
2) the movement is proportional in relation to itself. For example, the enormous
Andante introduction is balanced by the large coda: “As in Beethoven’s
Symphony No. 5/1, the coda provides a climax to the movement; but here, it
balances and intensifies the structure of the slow introduction, not of the
development section.”310 I will now examine how this might be balanced by later
movements.
The second movement is an Andante con moto in A minor. It is in a type of
modified sonata form, which Brown describes as “modified Sonata Rondo.”311 Its
exposition section modulates from A minor to F major: further evidence for axial
tonality being used to expand traditional tonal areas. The development-type
section is more remote, modulating from A minor to D-sharp/E-flat and then to
B-flat (i – IV/IVβ – IVβ). These are tonal centers that are “flat” – they do not
provide (and do not need to provide) any dominant. Instead, interest is sustained
through the sudden shifts of dynamic. The recapitulation begins in A major then
turns to A minor.
The Scherzo is an Allegro Vivace in C major, styled as another modified sonata
rondo. 312 The development begins in A-flat (Iβ), and other modulations are
axially related. What is particularly interesting, however, is the reception of the
later two movements of this Symphony, which will be touched on after my
311 Ibid.
312 Ibid., 637.
92
analysis of the finale. Although Schubert’s ending is in the realm of the
“apotheosis” finale (like the finale to Beethoven’s Seventh Symphony), Brown
suggests it is still well handled, though he is reserved in his adulation. He states
that this finale is sustained by “rhythmic drive, thematic power, and long-range
tonal control.”313 Again, this is a movement in sonata form, though modified: the
recapitulation separates tonal and thematic return: “The recapitulation and coda
depart radically from the expected tonal thrust, even if we are willing to accept
beginning the recapitulation outside the tonic key as within the range of
Schubertian normalcy.”314
A contemporary review of the Symphony’s first performance hinted at the
inherent problems of Classical form and the continuation of the Eroica model into
the mid-to-late nineteenth century: “Overall, the first two movements appeared
to be generally the most interesting.”315 This suggests (as readers will now be
familiar) a lack of tonal drama following the sonata form of the first movement
and lyricism of the second, although the reviewer has his own thoughts:
Conceivably, this [impression] could have resulted from the fact that the
listener’s attention was not yet exhausted by the all-too-long duration of each
movement. Still, it seemed to us that the first movements did indeed deserve
preference over the last, since the Scherzo was all too repetitive, and the final
movement resembled an opera finale without words.316
Further to this, Brown writes “This pace and the prolonged length [of the finale]
cause the grand pauses and cessations of movement to be concessions to
exhaustion, not only for the imaginary dances but also for the violinists who,
313 Ibid., 638.

314 Ibid., 639.
315 Ibid., 631.
316 Anonymous review in Allgemeine Musikalische Zeitung (Leipzig) XLI/13 (March 27, 1839),
cols. 256 – 57, quoted in ibid., 631.

93
during the nineteenth century and today, find its repetitive figuration
straining.”317
Evidently, form and symphonic weight were core issues that concerned
nineteenth-century composers. The increased emphasis on the finale – and
consequently less on the opening movements – became common, and we have
discussed a number of possible reasons for this. However, it is important to
examine the works of later nineteenth-century composers. Which models did
composers draw on? Did they make use of transcendent and immanent
dominants? How effective were their solutions? Here, each model is individually
examined and summarised.
Models
 We have examined the traditional sonata-form model used by Beethoven
in his first and second periods, both paradigmatically in the Eroica Symphony,
and in countless other classical works. This type of work is front-weighted, with
a first movement in sonata form. This model perseveres well into the nineteenth
century, for example in some of Brahms’ output. Brahms did avail himself to this
model in certain works, though many others indicate he was aware of
alternatives; in fact, Brahms contributes musical examples to most tropes
discussed in this thesis. 318 Structural or immanent dominants seem to make
transcendent ones redundant, if we take the classical model of I – IV (vi) – I – I
across four movements as the norm.
 However, Beethoven’s late period sees the development of other models.
The String Quartet Op. 127 in E-flat places the clearest manifestation of the
structural dominant at the end of the work, in the finale, though there are hints of
Ibid., 638.
317
See, for example, the Quartet Op. 67, the Fourth Symphony, and the Clarinet Quintet Op.
318
115.
94
it within the third movement. Harmonic tension continues through the first two
movements by use of tonic and subdominant axial relatives, whilst the finale is a
straightforward sonata form; here, the dominant becomes reified in its immanent
form. No transcendent dominant is at work here.
 The String Quartet Op. 131 in C-sharp minor is similar, but instead
reserves the dominant until the penultimate movements, the fifth and sixth,
before the dramatic finale. The effect is more end-weighted, and thus in this work
the arrival of the structural dominant gains more force because of its previous,
conspicuous, absence. Here, the fifth and sixth movements create the
transcendent dominant (using Vβ and v respectively). These movements are
transcendent in that they support the notion of being part of a dominant of the
entire work, whilst there is an immanent dominant in the sonata form finale.
 We might describe the Hammerklavier Piano Sonata Op. 106 in B-flat as a
mono-axial model. Though there is a contrast between the keys of B-flat major
and minor of the first two movements and the F-sharp minor of the slow
movement, the keys share an axis and there is no transcendent dominant. Rather
than positing the immanent dominant within the final movement itself,
Beethoven places it in the finale’s slow introduction: a remarkably un-dramatic
solution to the finale problem.
 The Cello Sonata Op. 102 is the final example, and I describe this model as
proto-progressive. 319 A small opening movement in a weakly tonicised key is
overwhelmed by a disproportionate emphasis on the subdominant area, with the
initial tonality only “rescued” in the final movement. This technique imbues the
music with drama and gives it a sense of activity. Though it is not actually
progressive, in that it does not begin in one key and end in another, the
significance of the tonic displacement gives a similar, if less dramatic effect. I
Dika Newlin, Bruckner - Mahler - Schoenberg (London: Marion Boyars Publishers Ltd., 1978),
319
129. Newlin coined this term in relation to Mahler’s tonal procedures, particularly in his
Symphonies.
95
suggest that the weak statement of the tonic key at the opening sets up a
requirement for dominant establishment later in the work. A similar effect occurs
in Beethoven’s Op. 101 Piano Sonata, and Schumann’s String Quartet Op. 41 No.
1.
While some of the techniques discussed here may seem remarkably similar, the
effects are subtly different. As we shall see, there are multiple solutions to the
problem of the finale: the selection and crafting of different models and the
success of such depends on whether a composer chooses to reserve the dominant
then “strike with his best weapons” later in the work, or craft flat movements
that rely on contrapuntal complexity to sustain them. Beethoven drew on such
solutions himself. Any discussion of form in the nineteenth century is bound to
be circumpolar and thorny: nevertheless, it is a discussion worth delving into for
the plethora of compositional procedures that Franck was undoubtedly aware of;
procedures which he selected or modified to service his architectonic needs.
96
Chapter Four
Late Beethovenian Models
This chapter examines more closely some of the important works Beethoven
composed in his late period; those works which have motivated formal models
other composers have drawn on.
Beethoven’s Ninth Symphony
The Ninth Symphony is a paradigm that illustrates Beethoven’s preoccupation
with tonality and form in his late period. A. Peter Brown discusses this here:
Establishing a tonality was a problem Beethoven pursued throughout his creative
life. Tonality was not to be assumed, but proven. Before the Ninth, the beginning
of a movement gave strong indications of what was to be accomplished or at
least set forth a tonal orbit if only temporarily, but the Ninth begins further
removed from these premises.320
This may be linked to the avoidance of root position dominants at critical
junctures in the Ninth Symphony, such as bars 33 – 35 in which Iβ supplants i, or
bars 116 – 119, in which the harmony is dyadic rather than triadic, and hence
functionally ambiguous. Cosima Wagner writes about this feature in her diary:
“A work such as [the Ninth] remains a mystery; R. [Wagner] says how
remarkable in B. [Beethoven] is the hatred of trivialities, the avoidance of
dominants, for example, and the enormous artistic instinct.”321 There are many
6/4 chords, for example, but the music seems to “fail” instead of asserting itself
with the expected cadential event that this chord historically presages. Two
320A. Peter Brown, The Symphonic Repertoire: The First Golden Age of the Viennese Symphony:
Haydn, Mozart, Beethoven and Schubert vol. II (Bloomington & Indianapolis: Indiana University
Press, 2002), 533.
321 Harvey Sachs, The Ninth: Beethoven and the World in 1824 (New York: Random House,
2010), 189.
97
important examples include the first important cadence into D minor (bar 17) –
actually a D minor 6/4 chord – and the hammered cadence into B-flat at bar 150,
where only B-flats and Ds are present. There is a link here to the Schreckensfanfare
(mentioned previously), built from both a B-flat major and D minor chord.
Though the remainder of this chapter is concerned with chamber and piano
works, we know that Beethoven is preoccupied with tonality and functionality in
symphonic and chamber works – and the voicing of chords is one way in which
keys can be affirmed or weakened.
String Quartet in E-flat Major, Op. 127
The E-flat major Quartet illustrates Beethoven’s preoccupation with the
distribution of what Downes describes as “symphonic weight”. 322 Kerman
considers that the weight of the E-flat major Quartet is central, lying in the slow
movement. 323 The second movement, in expanded variation form, sees “the
hymn-variation in E major serve as a spiritual climax for the whole quartet.”324
Spiritual climax though it may be, the following analysis suggests that the weight
in terms of background tonicisation falls later. The third movement, the Scherzo,
contains an immanent dominant, but – crucially – Beethoven does not allow it to
function as such: instead, the finale contains the true immanent dominant. Hence
the I – V – I process has been drawn over the course of the Quartet. The principle
of contrast is also important, a principle which preoccupied Beethoven more and
more in his final period. Whilst the first movement of this Quartet attempts to
minimise contrast (somewhat, as Kerman states, a contradiction in sonata form)
the Scherzando vivace maximises contrast in both key and rhythmic structure. 325
Otago, 1994), 50.
324 Similarly, the emotional core of the B-flat Quartet Op. 130 lies in the slow movement. Ibid.,
241.
325 Ibid., 228.
98
This principle may also help in deducing which movements Beethoven intended
to be structurally dramatic or non-dramatic.
Figure 4.1: Beethoven, String Quartet (Op. 127), I.

I
Maestoso
1–6 E-flat I Declamatory, triad outlined in 1st Violin
Allegro
7 – 32 E-flat I First subject – sounds like an ending phrase
33 – 40 E-flat – G I – Iα Fragments of first subject
41 – 68 G minor iα Second subject with strong cadence 64-65
69 – 74 G major Iα
Maestoso
75 – 80 G major Iα Declamatory, as opening
81 – 97 G major Iα Development/use of 1st subject
98 – 116 C minor ivα Fragments of 1st subject
117 – 120 A-flat major IV Briefly more stable
121 – 124 A dim Outlines dim 7 in Violin 1
125 – 128 B-flat minor v Briefly more stable
129 – 132 B dim Similar to 121 – 124 but a tone higher
133 – 134 C major IVα Arrival: this is a climax point, and links with the
end of the finale
Maestoso
135 – 138 C major IVα Declamatory, triad outlined
Allegro
139 – 146 C major IVα First subject
147 – 158 Progression to A-flat In A-flat: vii/ii – ii – V 7; fragmentary figure
rising in First Violin
159 – 166 A-flat, but sequence continues and A-flat becomes IV of E-flat at bar 167
167 – 174 E-flat I Statement of first subject in tonic key
175 – 182 E-flat I Derivative of first subject
182 – 198 Progression / preparation for A-flat or E-flat – A-flat eventuates.
In A-flat: V7 – I96– V7c – I7 – vii7/vi – vi7 – V7 – I
198 – 206 A-flat IV Second subject melodically – plagal pre – I
99
207 – 221 E-flat & A-flat I & IV Tonic pedal
222 – 230 E-flat I More firmly – no D-flats. Cadence into E-flat:
231 – 238 B-flat as V V Ending cadential figure from bar 65
239 – 270 E-flat, but lots of emphasis on key of A-flat – almost ‘rocking’ plagally at times
271 – 282 E-flat – now more stable, but pianissimo
Of the first movement, Kerman writes: “The phrase structure of the exposition
seems unbelievably simple, hardly conducive to dramatic tension” 326 and further
that “no quartet fast movement in Beethoven moves so lyrically as this one.” 327
This is likely, I suggest, because Beethoven is specifically aiming for tranquility,
plotting to place dramatic emphasis elsewhere. Kerman also suggests that “this
movement lives not on contrast but on the inherent beauty of the consequent
doublet phrase.”328 The simplicity is brought about by an antecedent-consequent
phrase, which “passes from force to gentleness”: an arpeggio figure answered by
a lyric doublet phrase.329
Figure 4.2: Beethoven, String Quartet (Op. 127), I, bars 7 – 14
Also notable is that the first subject – the doublet theme itself – begins with a
harmony on the subdominant, which reinforces the emphasis on the
subdominant key in this movement, particularly toward the end. But the
figuration of the opening arpeggio figure is also significant: none of the dominant
chords appear in root position, instead being crafted with the fifth, then seventh,
326 Ibid., 205.

327 Ibid.
328 Ibid.
329 Ibid., 204.
100
in the bass (a link may be inferred here between Op. 127 and the first movement
of the Ninth Symphony – see the above discussion).
This arpeggio figure is employed to start the development in G, or Iα. Kerman
suggests the key is “chosen to contrast as little as possible” 330; telling, as E-flat
and G share the same axis. 331 Beethoven reinforces this lack of drama in the
development section by use of canon, achieving “a thoroughly novel quality of
flatness.”332 Likewise, the use of G major (Iα) and C major (IVα), avoid the drama
of the dominant axis. Given this purposeful avoidance, we might describe the
movement as “non-tonicising” sonata form. 333 Kerman also adds that “The
blustering passages go nowhere and develop nothing significantly”.334 There has
certainly been no immanent dominant, which is in reserve to appear in a later
movement (or movements).
Figure 4.3: Beethoven, String Quartet (Op. 127), II.

II
A-flat Major (IV overall)
Adagio, ma non troppo e molto cantabile
First Variation: 1 – 20 A-flat I
Very spacious and lyrical. The only accidentals present are the D natural (in V/V, 9 – 10), an
occasional G-flat (creating v, I7) and D natural (creating vii/V, 19).
Second Variation: 20 – 39 A-flat I
A few more chromatic nuances happening now. Local tonicisation of F minor (27).
Andante con moto
Third Variation: 39 – 59 A-flat I
330 Ibid., 209. Kerman seems to be hinting at an axial interpretation without fully expounding
it; his language often points to the neutrality that axial theory posits.
331 This is one of many occasions where Kerman’s turn of phrase seems to prefigure axial
theory.
333 Non-tonicising sonata form can be described as a form that has the traditional schematic
subdivisions but no structural dominant.

101
More lively, staccato feel, though still relaxed. Tonic minor appearing occasionally (i.e. 44).
Emphatic moment on D-flat minor (50) perhaps to signal an enharmonic pivot for what is to
come. (D-flat minor enharmonically C-sharp minor).
Adagio molto espressivo
Fourth Variation: 60 – 77 E major Iβ (in context of whole quartet,
IVβ)
Rhythmic change from 12/8 to 4/4. Similar transition to Op. 131 between 1 st and 2nd movements.
Forte on Cmaj7 in 67, 71, 75 (Iα.) Transition back simply by E natural, falls to E-flat in bar 78.
Tempo I
Fifth Variation: 78 – 100 A-flat major I
Back in 12/8 tempo. At 96 things start changing – touch on D-flat, B-flat minor is suggested, then
E-flat minor, before transition to C-sharp minor between bars 100 – 101.
Sixth Variation 101 – 107 C-sharp minor iv (enharmonically)
Similarly iambic as before, though not quite as hushed as the E major variation. Mode shifts to A
major at 107.
Continued . . . 108 – 118 A-flat major I
Mainly straightforward and diatonic. Ends with an imperfect cadence at bar 118.
Coda 119 – 127 A-flat major I
120: A-flat7 – D-flat second inversion – Preparation for I, but then moves to D-flat minor – E
major – V7b – E major (contraction) – then moderate scale I6/4 – V – I including the tonic minor.
Slow and tender ending, with B-flat falling to A-flat in the bass.
This Adagio ma non troppo e molto cantabile is the longest movement, and arguably
the centerpiece of the Quartet.335 Kerman emphasises its importance: “Both works
[Op. 127 and Op. 132] bank more heavily on their slow movements than any
other of the quartets, earlier or later . . . even in absolute terms the Adagio . . .
stands out as the most monumental of Beethoven’s variation movements.”336 The
movement is in A B A form. The melody is opulently lyrical, and Kerman
describes the climax of the melody as being the subdominant touch in bar 17 –
important, as the subdominant is the most structurally significant key in this
335 Ibid.
336 Ibid., 226.
102
work as a whole.337 Variations one and two constitute the first section, variation
three the second, and variations four five and six the last. The design is
symmetrical, setting it apart from the variation forms of other quartets. 338
Harmonic relationships utilise the tonic and subdominant as well as their axial
relatives: the A section is in A-flat; the B section is in E major (Iβ of this
movement) moved to by “a spasm”339 without preparation; variation four returns
“just as abruptly” to A-flat; variation five emphasises the subdominant, D-flat
minor; and variation six returns to the tonic. 340 Despite this movement being the
emotional core of the Quartet, the tonal scheme consists entirely of tonic and
subdominant axis members relative to A-flat.
Relative to the Quartet as a whole this movement represents a significant
prolongation of the subdominant axis. In this, it seems we have discovered a new
category: the transcendent subdominant, where the subdominant section of the
work as a whole is embodied within a movement, as it is here. This also possibly
links with Beethoven’s Op. 130 quartet, in which the Cavatina plays a similar role
emotionally and, in conjunction with the G major Alla danza tedesca before it,
represents a transcendent subdominant phase of the work. Regarding Op. 127, it
is unlikely that the size of this movement would have been possible had the first
movement been a traditional sonata allegro with structural dominant. Discussing
another monumental slow movement – that of the Hammerklavier – Downes
writes that:
The salient point here is whether or not a slow movement of this length would be
possible in a work where the tonic had been already outlined at the background
structural level . . . such a movement would be inappropriate because it would
have no tonal function to perform. . . . the tension created by the as yet unfulfilled
337 Ibid., 212.

338 Ibid., 214.
339 Ibid.
340 Ibid.
103
background tonicisation is conceivably what sustains the Adagio’s enormous
length.341
We can also apply this observation to the Quartet Op. 127. This work shows
Beethoven stretching his refusal to act tonally to extreme lengths and the
extended slow movements in each work show both the functional and lyrical
ends made possible by withholding the structural dominant.
It is also important to note the subtle thematic allusions between the end of the
second movement and the beginning of the third. Four bars before the end, the
second movement outlines a I – V – I progression in E. What seems to be
happening is a prefiguring of the point in the work where the music will become
tonally active; where the structural dominant will eventuate. Beethoven uses a
similar device in the String Quartet Op. 131: in that case, the semiquaver cello
figure in variation six thematically prefigures the Scherzo; similarly, the cello
arpeggio in the coda of the same movement rhythmically prefigures the finale.
From these allusions, it seems that Beethoven is at pains to communicate – at a
foreground thematic level – the shift from passive to active function in the work
as a whole.
Figure 4.4: Beethoven, String Quartet (Op. 127), III.

Scherzando vivace
1–2 E-flat – B-flat I&V Declamatory chords: I – V – I – V
2 – 17 E-flat I First subject (B-flat, E-flat)
17 – 25 E-flat I Continuation/repeat 1st subject
26 – 36 Transition to B-flat V Touch on C minor (31) but becomes ii of
ii-V-I in B-flat. 2 – 36 repeated.
36 – 40 Backward cycle of fifths Vicious dotted motive in unison
Downes, "An Axial System of Tonality Applied to Progressive Tonality in the Works of
341

104
suggesting backward cycle of 5ths: ends
on G (V/C minor)
41 – 46 C minor ivα Inverted first subject
47 – 50 F minor WT/I (v/V) Fragments first subject
51 – 55 D-flat WT/I (vβ/V) Fragments first subject
56 – 59 V/E-flat minor – E-flat minor – transition, somewhere between E-flat minor and
G-flat
60 – 69 Vicious dotting: through G-flat (Vβ) – C-flat (IV/Vβ) – G-flat – D-flat (V/Vβ) – F-
sharp minor (Vβ) – D (Vα): then suggesting V / I / V but ending on a unison E-
flat, which turns into SD6 of G minor
Allegro
70 – 74 G minor iα In 2/4 – definite disruption
Tempo I
75 – 77 vii7/G vii7/iα Fragments first subject
Allegro
78 – 80 As 70 – 74
Tempo I
81 – 84 V/F V/V/V Fragments first subject
85 – 89 F V/V Fragments and – re-transition to 1st sub.
90 – 112 E-flat I Full first subject – harmony v. similar
113 – 115 VI – ii – V in E-flat: altered harmony from first Ritmo di tre battute
116 – 118 Extension of 113 – 115
119 – 122 E-flat I Big cadence with downward scale figure
122 -141 E-flat I First subject becomes ending figure.
Tonic pedal from 129 – 137. C-flats in
texture (♭VI – V trills in first violin.)
Forecast of what is to come
142 – 143 E-flat minor i 2nd beat chord changes abruptly from
major to tonic minor
Presto (Key change: E-flat minor)
144 -167 E-flat minor i – V – i – IV (V/VII) – VII: legato and pianissimo
167 – 173 Suggests D-flat WT/1 C naturals in texture – unison
174 – 185 D-flat WT/1
186 – 189 V/B-flat V/V
105
190 – 200 B-flat minor v As bar 148, but in the tonic minor
201 – 207 Transition – around E-flat minor/G-flat major
208 – 219 G-flat major Vβ As bar 174
220 – 223 V/E-flat V
224 – 236 E-flat minor i As bar 148, but instead of going to IV
there is a transition back to the major
beginning with the dominant major (B-
flat) which appears from 235.
(Key change: E-flat major)
237 – 268 V/E-flat V Dominant preparation for tonic return,
but odd notes above dominant pedal
from 260 – 268: 9ths, 2nds: bar 269 is A-
C-E-flat over B-flat!
269 – 272 Transition to restatement of original material: inversion of first subject in the
dominant
273 – 414 As opening section – Presto
Presto
415 – 426 Truncation of earlier Presto – i – V
Tempo I
428 – 435 Little cadential statement – ending on I
So, how might Beethoven follow such a monumental slow movement; how to
sustain this feeling, this breadth? Though the slow movement was extended and
weighty, there are still two movements to follow, and we are awaiting the
structural dominant. Following the slow movement is the Scherzando Vivace, back
in the tonic E-flat. This opens with four emphatic chords, then a fugue, with the
subject in the dominant, answer in inversion, then short episodes. 342 Kerman
suggests that although the late quartets – including Op. 127 – are remarkable for
their integration, “contrasts – within movements or between movements – may
be more extraordinary than ever.” 343 This movement certainly contrasts the

343 Ibid., 229.
106
previous two, but more significantly, it contains a failed immanent dominant.
There are instances of the dominant, to be sure, such as hints at the end of the
opening statement and at bar 60, G-flat being Vβ. However neither instance is
prolonged (it collapses back into E-flat minor at bar 224): a vital condition for the
attainment of the immanent dominant is a degree of prolongation, which is here
denied. In this, we can consider the immanent dominant as synonymous with the
Schenkerian structural dominant. The Trio’s tonal scheme is I – V (double bar)
VII – v – III – V – I (returning to the tonic in the repeat).344 Bars 37 – 59 feature a
sequence traversing C minor, F minor, B-flat minor and D-flat major, before
arriving at G-flat major (Vβ) in bar 60. Four bars later this swings to F-sharp
minor, acting as a pivot v of G minor which arrives at bar 70. After a cycle of
fifths, we arrive back at E-flat major. The trio at bar 148 employs the tonic minor,
the antithesis of tonal contrast.
Yet it is important to realise that in this movement the dominant becomes
manifest in a stronger way, with root position chords (as opposed to the inverted
chord used frequently up till this point) and becomes a more significant
harmonic center. Yet there seems to be a correlation between Beethoven’s
introduction of the dominant (in a “failed” immanent fashion) and a lack of
emotional seriousness: for example, in Scherzo movements. The dominant is then
“absorbed” into a more formal context in the final movement, but its
introduction is usually off-hand. This is true of both Op. 127 and Op. 131.
Figure 4.5: Beethoven, String Quartet (Op. 127), IV.

1–4 Chromatic opening
5 – 36 E-flat I First subject group. Naturals in both
melody and texture – some foreground
tonicisation of B-flat (V).
344 Ibid., 233.

107
37 – 44 E-flat I First subject – fragmented
45 – 54 B-flat V Transition to dominant - first subject
55 – 66 B-flat V Second subject
67 – 72 Transition Emphasising unison G, D, E-flat, F
73 – 80 C7 chord – F7 V/V (in B-flat)
81 – 96 Transition suggesting B-flat – E-flat then E-flat augmented, before C minor
97 – 100 As Chromatic opening
101 – 105 E-flat to F minor (briefly)
106 – 110 C7 chord (suggesting more F minor) but then V/C7 in 110 becoming:
111 – 120 C minor ivα or vi Second subject
121 – 124 C minor – C major ivα – IVα First subject – chromatic alteration
125 – 134 On C7 – two brief glimpses of F
135 – 144 Transition: first suggesting F minor, Sequence through F minor, E-flat major, D-
flat major, before returning to E-flat to become V/A-flat
145 – 160 A-flat IV False recapitulation
161 – 176 A-flat IV Second aspect of first subject, but
ending on a diminished chord in
bar 176
177 – 186 Transition back to tonic – rising tones to some degree in all voices
187 – 218 E-flat I Entire first subject group
219 – 226 E-flat I Second subject
227 – 230 B-flat V Transitory
231 – 236 As at 67, but down a fifth
237 – 244 F7 chord (suggesting V/V) but this is only a quick transition to the tonic
245 – 250 E-flat I
251 – 255 B-flat7 – B dim (vii7/C) – C minor – C major
Allegro con moto (Key change: C major)
256 – 260 C IVα Variant on first subject – 260 with #4
261 – 263 A-flat IV Semitonal rise from G – A-flat, as above
264 – 266 F-flat (E) IVβ Completes sweep round axial cycle,
cyclically recalling the slow movement.
267 – 268 Short transition through E-flat7 – A-flat – Ddim9 – E-flat
269 – 272 E-flat I
108
273 – 276 Transition: all around E-flat, but pattern in bass moving through semitones then
falling a major third.
277 – 284 E-flat I Variant on first subject – ff arrival
285 – 288 Cadential approach – p
289 – 292 E-flat I ff iteration – scales
293 – 299 E-flat I 293: trill on tonic chord punctuated with
C-flats (♭VI). Also A naturals a few bars
before the end.
As the emotional weight of the Quartet lies in the slow movement, and the failed
immanent dominant has been introduced in the Scherzo, the finale is simple by
necessity.345 The structure of the phrasing is simple, and there is little transitional
material; a pair of statements in the tonic, followed by a pair in the dominant,
without transition or modulation. G is also emphasised, as in the first movement.
The sonata form is conventional and includes a false recapitulation in the
subdominant at bar 153, again reinforcing both convention and the subdominant
bias of the work as a whole.
Despite this simplicity, the finale is entrusted a crucial job – to flesh out the
background structural dominant; the immanent dominant. Beethoven places this
dominant at bar 183, and this gives us the crucial sense that the sonata form
achieves prolongation. The recapitulation proper at bar 256 is of particular
interest as it brings with it a glorious axial harmonic sequence: after restating the
main theme, the harmony swings from C major through A-flat, E, and E flat. 346
This is nothing less than a sounding of the entire subdominant axis in one large
sweep, through steps of a major third in the bass, and Kerman describes these
bars as a “sheer dream.” 347 Given the tonal regions touched on, we might
interpret this moment as a revisiting of tonal regions used in the Adagio, the
345 Ibid., 234.

346 Ibid., 237.
347 Ibid., 238.
109
heart of the Quartet, and a reemphasis of the importance of the subdominant
region in the Quartet as a whole. We might call this reemphasis the immanent
subdominant; an echo of the transcendent subdominant that is the second
movement. This subdominant emphasis is underpinned thematically as this
section also recalls the Adagio in its use of rising sevenths and octaves in the
melody in bars 259 – 260 and 262 – 263. This interval supplants the fourth used in
the finale theme.
As shown, the first half of the Quartet is passive in that the tonicisation process is
withheld. Beethoven seems to have calculated this lack of drama to allow him to
expand the slow movement to its significant expressive length: in producing a
non-tonicising agenda in the first movement, he could induce an acceptance of
what, by classical standards, amounted to a proportional distortion of movement
size. Kerman writes that the slow movement’s design is: “eloquent and weighty –
another ‘expansion’ upon the delicate formal outlines of the first movement.” 348
The second half of the Quartet completes the tonicisation process, with the
introduction of the dominant axis and the substantiation of the “failed”
immanent dominant in the Scherzo by a tonally active sonata form finale with a
“successful” immanent dominant. It is as though a single sonata movement has
been stretched – albeit loosely – across the entire Quartet.
String Quartet in C-sharp minor Op. 131
If the fourth movement provides the immanent dominant in Op. 127, it is the
seventh in Op. 131: the final movement. Yet, here there is a transcendent
dominant – one that acts as the dominant of the entire work – in movements five
and six. Hence the reservation of the structural dominant until late in the work,
but interest and intensity are maintained by making use of axially related tonally
348Ibid., 241.
110
“flat” or neutral key areas, including the subdominant, and the crafting of the
transcendent dominant in the penultimate and antepenultimate movements.349 In
this way Op. 127 and Op. 131 are similar, though as we will see in Op. 131
Beethoven presents the structural dominant with more dramatic force. However,
it is worthwhile surveying each movement individually to deduce Beethoven’s
tonal scheme.
This Quartet opens with an expansive, mournful fugue. Though Beethoven
experimented with withholding tonicisation in his later works (possibly but not
necessarily as alternatives to sonata form), his use of a fugue as the first
movement here is still unprecedented. If we consider that Beethoven’s tactic is
again the withholding of dramatic weight, however, his use of fugue can be
interpreted (and has been by Kerman) as a conscious effort to “eschew all the
tired characteristics of sonata style . . . the ‘deflection’ of sonata form and the use
of a fugue for an opening movement are different solutions to the same problem;
that of the imbalance in the distribution of symphonic weight.”350 Kerman writes
further:
An opening movement in sonata form could only have singled out one or two
keys, dramatically stressed and nervously highlighted: the Neapolitan, let us say,
plus the sixth degree. But a fugue can (and this one does superbly) map out with
dispassionate authority an entire terrain, the whole circumscribed tonal field of
the minor mode.351
However, this description of sonata style as “tired” is problematic, as when the
sonata allegro does arrive, it fulfils all the criteria of traditional sonata form with
an opening i – V – i progression, a second subject in a dominant axis key, and
349 Kerman also refers to movement of Op. 131 as possessing an “underlying quality” of
“flatness”. Ibid., 333.
Gustav Mahler and Nineteenth Century Antecedents", 50. Kerman, The Beethoven Quartets,
273.
111
dominant preparation prior to the recapitulation. Yet if we think in Schenkerian
terms, this “deflection” of sonata style by use of fugue also comes hand in hand
with abandonment (or rather a reserving) of the traditional tonic-dominant
polarity. Perhaps instead of searching for an alternative to sonata form,
Beethoven was simply keeping its inherent vitality and dramatic power in
reserve. The notion is reinforced by the dominant’s late recurrence in Op. 127,
Op. 131 and the Hammerklavier; it certainly seems to be a conscious architectural
tactic on Beethoven’s part.
As the fugue avoids dominant and dominant axis keys, it has a feeling of tonal
flatness. Kerman suggests v is tonicised at bars 34 – 35, but Downes strongly
refutes this, as even the weak suggestion of B features a dominant pedal and
weak perfect cadence, which the v “modulation” lacks. 352 Beethoven clearly
considers both keys to be of little importance, neither being firmly established
(notably, the modulations which instigate key signature changes are those to C-
sharp minor, E-flat minor, B, A and C-sharp minor). Reinforcing this tonal
neutrality is the emphasis placed on the subdominant. The fugal answer is in F-
sharp minor, and at bar 113 the tonic (C-sharp minor) turns to major, suggesting
the possibility of C-sharp being heard as a dominant of F-sharp. The transition to
the second movement heightens this ambiguity, where the bare C-sharp octaves
that end the movement are reinterpreted as the leading note of the D, the
Neapolitan and key of the second movement.353 The Neapolitan chords at the end
of the fugue also contribute to the ambiguity of C-sharp and its potential
reinterpretation of V of F-sharp. In a first movement, we would usually
experience tonicisation of a given key via sonata form. Instead, we remain in
352 Furthermore, Downes notes that “the wholetone scale is incapable of establishing a tonic,
except perhaps by emphatic reiteration.” Downes, "An Axial System of Tonality Applied to
Progressive Tonality in the Works of Gustav Mahler and Nineteenth Century Antecedents",
18.
353 Ibid., 53.
112
doubt as to what the tonic actually is. Certainly, the entire minor mode has been
mapped out in the fugue, but neither F-sharp nor C-sharp has been tonicised
unequivocally.
Our uncertainty about the tonic continues in the second movement. The key is D
(IVβ), and though sonata form is suggested with a move to the dominant, no
further modulation ensues and the music leads back to the tonic. 354 Kerman
describes this as “sonata form without development”;355 we might also describe it
as another “failed” immanent dominant. At bar 44 there is a C-sharp major
chord, the use of the major mode again suggesting C-sharp could be a dominant
rather than a tonic; this is not the only time Beethoven employs this technique in
this quartet.356 Though this is only momentary, it reinforces C-sharp’s ambiguity
and the potential of F-sharp to be the tonic. Downes states that “Beethoven is
making a point of dramatising the axial substitution of D for F-sharp”, which in
other words means a movement in F-sharp might be expected; however, by using
D, the Neapolitan, Beethoven can emphasise the semitonal relationship between
C-sharp and D. The use of the subdominant axis is formally logical given the
beginning of this quartet possesses a tonic – subdominant bias, and the latter half
a dominant – tonic bias.357 This it has in common with Op. 127 and Op. 106,
which I will examine shortly.
The third movement is a short bridge in B minor. Kerman suggests the bridge has
a functional use in that it avoids the slow movement (in A) sounding like the
354 Ibid., 52.

356 Beethoven uses a similar pairing in the Scherzo, where the tonic E is undermined by G-
sharp major chords, such as in bar 10 and 19 – 20, where a G-sharp chord begins the second
half of the main theme.

113
dominant of D, the key of the second movement.358 The bridge arguably ends on
E major, the dominant of A (Iβ), which is the key of the fourth movement, the
slow movement – a set of tonally “flat” variations. Reflecting on the movements
thus far, Downes states that the “first half of the work is set to establish the tonal
sequence I – IVβ – Iβ, successfully withholding any structural dominant either
within or between any of the four movements.”359 This is the same tactic at play
in Op. 127. However, the eighth variation of the slow movement brings some
interesting harmonic movement indeed. At bar 231, the music turns from A
major to C major or Vα. For Downes, the use of C major interrupts the
predictable flow of the variations and heralds the arrival of more dominant axis
sonorities which will soon become significant; a low level dominant that
prefigures the immanent dominant still to come. He observes that “the
expectation of the structural dominant is so strong now, as much on account of
its absence up until this point as Beethoven’s arousing our appetite for it at the
end of the Andante.”360
An E major (Vβ of C-sharp) Scherzo follows, with particular emphasis placed on
G-sharp (the dominant of C-sharp) by way of fermatas. To Downes, this shows
that the tonic – subdominant bias of the first half of the work has shifted to
become a dominant – tonic bias. 361 Although this dominant bias is becoming
evident, our desire for tonicisation is not yet satisfied, partly because of the
Scherzo’s lack of any dramatic content and the continued immanent
subdominant relationship between the Scherzo and its subdominant trio.
Furthermore, the I – V – I opening of the Scherzo is not composed out in the rest
of the movement: the opening antecedent-consequent phrasing hints that this

360 Ibid.
361 Ibid., 52.
114
may happen, but the move to III thwarts this expectation. Yet this movement also
looks forward to the finale in its harmonic and thematic material: an A chord
precedes the G-sharp that ends the first section, sounding like ♭VI – V in C-sharp
(though ♭VI is not spelled as such and hence open to contention), and this
prefigures the harmonic movement that bridges the sixth and seventh
movements.
The sixth movement is a brief Adagio in G-sharp minor that turns to G-sharp
major (the dominant of C-sharp) in its final bars, after which the finale arrives
and fulfills “our expectations for a movement that single-mindedly focuses on
the process of tonicisation.”362 This is where the immanent dominant arrives.
Kerman seems certain that this finale is a solution to the “problem”:

Beethoven has arranged things here so that he can strike frankly with his best
weapons – themes in strong contrast and in arresting juxtaposition, exciting
modulations, ‘expansive and argumentative development,’ triumphant returns,
and great summary codas. Sonata form is treated without rigidity and yet (what
is rare in finales) with full emphasis and amplitude. . .363
We must attribute the success of this movement to the pacing and tonal logic of
the entire Quartet. This is the heaviest movement, but does not seem unduly so
as the previous non-tonicising movements – with the exception of the fugue –
seem preparatory in comparison. One way which Beethoven achieves this is
through his use of axial relatives both within and between movements, which
effect modulation without creating preemptory tonal drama, but another is the
deep harmonic and thematic integration for which Beethoven’s late period works
are renowned. Additionally, the fugue remains crucial to the formal plan of this
quartet: its tonal disorientation and solemnity create a palpable depth of tension
362 Ibid., 54.

115
that infuses the work. It seems to me that it is this tension that commentators are
referring to when they describe Op. 131 in such ecstatic terms.
Piano Sonata in B-flat major Op. 106: Hammerklavier
In order to gain a broader perspective on late Beethoven handling of the finale
problem, it is worth examining a work outside the quartet genre. The
Hammerklavier Sonata Op. 106 is an excellent example and arguably one that
loomed large in the consciousness of succeeding composers. Here again we see
Beethoven shifting weight away from the first movement, and as a result the first
movement sonata allegro here focuses exclusively on the subdominant and its
axis members, with no structural dominant provided.364 However, the emphasis
on the subdominant indicates an immanent subdominant. Furthermore, we
might describe the Sonata as a whole as “mono-axial”: each movement is in a key
of the tonic axis meaning there is no functional tonal contrast between the keys of
the movements themselves. There is no transcendental dominant in this work
unless one counts the introduction to IV as a discrete formal unit, and the only
significant instance of dominant tonicisation is in the opening slow section to the
finale. Even the finale itself does not tonicise the dominant at the background
structural level.
In The Classical Style, Rosen suggests that these mediants and submediants create
long range dissonance and thus are “true” dominants. 365 Downes disagrees,
arguing that although Beethoven is creating “long range structural dissonance,
the resolution of which generates the form”, these dissonances are not true
dominants: 366 Beethoven knew the power of the dominant and simply reserved

116
it. This lack of a background dominant may be at odds with Schenker’s
definition of a sonata form movement, and the tonic may be “stumbled upon
rather than conclusively attained” in the recapitulation, but in a schematic sense
it is still sonata form. 367 This movement illustrates the extent to which axial
modulation can be form-generating, without relying on a structural dominant.368
The second movement, the Scherzo, does not provide a structural dominant
either. The movement is short; B-flat and B-flat minor being the only significant
key areas. Beethoven is reserving more weight (though not the dominant) for the
third movement, an extended Adagio slow movement. The absence of the
dominant is particularly conspicuous with the bare octaves in the opening of the
slow movement. We sense the possibility of tonal “activity”: yet this feeling is
almost immediately thwarted and replaced with the “passivity” which we have
been experiencing for the first two movements. Enhancing this is the foreground
manifestations of the subdominant in the first two movements with the emphasis
on B minor and B octaves. It is because we have a classical expectation of balance
between dominant and subdominant that this movement seems to withhold so
much: thus this “tragic reversal” possesses such emotional depth. The first two
bars of the slow movement are especially evocative: A octaves – C-sharp octaves
– F-sharp minor chord. Beethoven is acutely aware of our aural expectations and
at once surprises and fulfils them. After spending so long in B-flat major, Tovey
suggests we desire to hear A and C-sharp as a dominant (Vα) but instead “the F
sharp minor chord utterly transforms everything. These two notes constitute one
of the profoundest thoughts in all music.”369 It is also worth noting that these two
notes were one of the last things added to the score, in the final proof; 370 this
367 Ibid., 59.

368 For a fuller analysis of Beethoven’s Hammerklavier, refer to ibid., 56.
369 Donald Francis Tovey, A Companion to Beethoven’s Pianoforte Sonatas (London: Oxford
University Press, 1931), 238.

117
could be interpreted as a prefiguring of the opening octaves of the final
movement which composes out F – A.
Again, this movement uses the tonic axis, though internally the keys are almost
exclusively those of the subdominant axis: E-flat, B and G. Here lies another
immanent subdominant, representing something close to a retracing of the
development section of the first movement. As we have touched on in our
discussion of Op. 127, the only notion preventing Beethoven from writing such
an extended slow movement before this point was the possibility that it would
become dull: however, because the background tensions remain unresolved, the
length is “heavenly” (to borrow Schumann’s term).371 Downes writes:

If Beethoven had attained a structural dominant here, and Vα (A major) would
suffice, the pressure to complete the background tonicisation of the work would
not allow the work to dwell too long on a tonal dissonance before achieving
closure . . . an adagio of this length would be intolerable.372
This is an excellent example of axial relatives’ potential to generate
monumentality, but this is as much because of the suspended animation of the
tonicisation process as the axial keys themselves. Yet, crucially, the third
movement leaves our desire for a structural dominant unfulfilled.373
The slow introduction preceding the finale finally delivers the immanent
structural dominant. F (V) is established at the opening, G-sharp minor at the
preparation for the cadence into the Allegro (functioning as vα of Vα), and A
372 Ibid., 60.
373 This begs the question of when a dominant ceases to be necessary at all, which was to be
tackled by Wagner and his followers later in the nineteenth century. It might be argued that
Beethoven’s Ninth Symphony and Op. 133 String Quartet head in this direction - however,
the works examined in this thesis that use axial tonality continue to use dominants (and their
substitutes) as well as tonics. This may make an interesting study to another interested party,
however.
118
(Vα) is also utilised. According to Downes, this “fulfills, in axial terms, the
axiomatic requirements for the dominant axis to attain background level status,
which in turn ensures the completion of the Schenkerian ‘triangle’ for the entire
work.” 374 The finale emphasises the leading-note motions that have been
suppressed in the earlier movements, though only at a foreground level: the
finale does not contain the immanent dominant. Downes concludes that this
Sonata’s ‘“heavenly length” was only made possible by this delaying action,
which in effect gives the entire work responsibility for unfolding the tonal plot in
terms of background tonicisation, not the first movement only.375 Beethoven is
not merely experimenting with third-related keys; he is re-thinking classical
models, re-dramatising them, as it were, to sustain their duration and intensity.
Cello Sonata Op. 102 No. 1 in C Major
Some commentators consider this Sonata to be the first of Beethoven’s late
period, perhaps in tandem with the Piano Sonata Op. 101 which shares some of
its characteristics. Much of the discussion of the works in this chapter has centred
on avoidance of the dominant, but this Sonata deflects attention from the tonic by
tonal understatement and then rescues it in the final movement. In this sense the
Sonata cannot be considered to be an example of progressive tonality, but shares
the idea of making a key vulnerable to a(nother) competing key by tonicising
neither at a background level, meaning both keys battle to attain deeper level
tonicisation. As Schoenberg writes, “If the key is to fluctuate, it will have to be
established somewhere. But not too firmly; it should be loose enough to yield.” 376
Both Op. 101 and 102 No. 1 have short lyrical first movements in which cadences
in the tonic are undermined, followed by complete sonata form movements in a
375 Ibid., 62.
376 Arnold Schoenberg, Theory of Harmony, trans. Roy E. Carter (Berkeley, Los Angeles:

119
subdominant-axis key. The following analysis outlines these tonal events.
Though on paper they may seem unremarkable, the crafting of the Adagio and
instability of the tonic certainly gives the effect of what I am proposing to call a
proto-progressive tonal scheme in works such as this that admit a potent rival for
tonicity.377
The Sonata is considered in two parts: the first, the Andante and Allegro, and the
second the Adagio, Andante (which cyclically recalls the first Andante) and the
final Allegro.
Figure 4.6: Beethoven, Cello Sonata (Op. 102), Part I

Andante
1 – 27 C major I (of this movement and Sonata as a whole)
Allegro Vivace
28 – 46 A minor i of this movement
47 – 75 E minor v (not V) used.
76 – 88 Modulating sequence (development): C major (Vβ) with Vivace theme; 80 – 82 F-
sharp diminished (vii7/V/Vβ); 83 G diminished (vii7/Vα) ; 84 – 86 A 7 (V7/iv) ; 86
(end, briefly) F major (Iβ); 87 – 88 G-sharp diminished (vii7/i) to A minor (i,
second beat of 88.)
89 – 92 B-flat major IVβ
93 – 97 D minor iv A7 chord suddenly, then D minor
forte: at 97 there is a sudden shift to
vii7/i and we are thrust back into the
opening theme: recapitulation
98 – 106 A minor i At bar 106 there is a cadence into C
377Downes, "An Axial System of Tonality Applied to Progressive Tonality in the Works of
Gustav Mahler and Nineteenth Century Antecedents", 79. As Downes discusses, when
Mahlerian progressive tonality occurs it is often via similar means to proto-progressive
tonality, in which a weak tonic rivaled by a subdominant axis key; the progression from G to
E in Mahler’s Fourth Symphony is a clear example. In proto-progressive tonality the original
tonic is rescued; conversely, in true progressive tonality the subdominant axis key eventually
wins out and displaces the original tonic.
120
major
107 – 114 F major – D minor Iβ – iv Brief touch on tonic axis before
recapitulation continues
115 – 144 A minor i Regular recapitulation
145 – 151 D minor – B-flat major iv – IVβ Touch on subdominant axis
151 – 154 A minor i Ending cadence not particularly
emphatic
Figure 4.7: Beethoven, Cello Sonata (Op. 102), Part II

Adagio
1 On G/C V/I
2 C/F I/IV
3 A/(V/A) IVα / V/IVα
4 D minor – major – diminished V/V – then diminished chain
5 Dim to G minor – C minor vii7/v – v – i
6 C minor – A-flat major i – iβ
7 G major – C SOUNDS LIKE I – IV
8–9 G – now sounds like V
Tempo d’Andante
10 – 16 C I (14 brief touch on D minor, but
clearly in C): cyclically recalling first
Andante
Allegro Vivace
1 – 22 C major I Piano and cello rhythmically misaligned
23 – 30 C – A minor I – ivα Semiquavers begin
31 – 57 G major V
58 Rest
59 – 63 E-flat Vβ Cello and piano misaligned again
64 Rest
65 – 69 C I As 59 – 63. C sounds remote;
perhaps due to minor third modulation
70 Rest
71 – 78 A-flat Iβ
78 – 87 Sequence through F minor – C minor
121
88 – 93 C minor i
94 – 101 Sequence through A-flat, F minor, D-flat, C minor (tonic and subdominant
axes)
102 – 105 Dominant pedal V
106 – 131 C – A minor I – ivα As opening – still rhythmically
misaligned
132 – 139 F major – G major IV – V
140 – 166 C major I
167 Rest
168 – 172 A-flat Iβ
173 Rest
174 – 178 F major IV
179 Rest
180 – 196 D-flat IVβ
197 – 233 C major I
(At bar 221 the rhythmic misalignment is fixed. This only happens in the last few bars and
resolves the rhythmic tension of the work, much like the finale of the Franck’s Violin Sonata,
discussed in Chapter Nine. )
The first Andante section is clearly in C major, although there are only two full-
close cadences in the entire sonata – one at the end of the A minor section and
one at the end of the finale.378 Hence, C is hardly tonicised strongly, and indeed,
the A minor Allegro certainly seems the “apparent first movement”. 379 Though
the tonic is stated in the first movement, it is undermined by a transcendent
subdominant, which is reinforced by its own immanent dominant. Yet in the
second movement, when there are affirmative cadences into A minor, they are
also often undermined by being placed on the third beat rather than the first.
This is done to contrast with the end of the work, as Downes notes specifically in
the finale: “how emphatic the three tonic chords at the conclusion of the work
378 Lewis Lockwood, "Beethoven's Emergence from Crisis: The Cello Sonatas of Op. 102
(1815)," The Journal of Musicology 16, no. 3 (1998): 306.
379 Ibid.
122
are, having been given full metrical stress.”380 To summarise the movement, it
falls within Beethoven's tendency to inhibit sonata form in some way. The
difference between this and the Hammerklavier or the Ninth Symphony is that the
sonata mechanism is inhibited at a foreground rhythmic level (lacking
Lockwood’s “full metrical stress”), whereas in the Ninth the sonata mechanism
which is inhibited is at a foreground harmonic level.
However, the Adagio is of special significance. Lockwood notes that it “restores
C major, but not in a firmly anchored way.”381 Although the basic polarity is
between C and G (both major and minor are used) to our ears G is such a strong
force in bar seven that the move to a C major chord sounds like I – IV in G. The
return of the Andante theme after ten bars of Adagio seems almost as if the
music is “starting again”, and this time with a firm sense of tonicity. This G could
be described as a “miniature immanent dominant”, hinting at what will be fully
substantiated in the finale.
The Allegro Vivace is a sonata form in C (with coda), with a significant
misalignment between the piano accompaniment and the cello theme at several
junctures, such as those at bars 63 and 69. This is only corrected in the final 13
bars, when, at bar 221, the cello and piano finally “line up”: the sense of arrival
here is strong. This movement possesses the immanent dominant, but also the
immanent subdominant in bars 190 – 212, echoing the transcendent subdominant
in the first part.
The above models are representative of the most important formal types
Beethoven employed in his late period. His preoccupation with tonality,
381 Lockwood, "Beethoven's Emergence from Crisis: The Cello Sonatas of Op. 102 (1815)," 313.
123
functionality and form is evidenced in works discussed in this chapter, and how
later composers – particularly Franck – responded to such innovations is the
concern of the remaining chapters. There is crossover between the models,
indeed, but the effect of each is subtly different. These divergent paths
demonstrate the pervasiveness of Beethoven’s influence on mid-to-late
nineteenth-century composers and the difficulties in deciphering the inner
workings of Beethoven’s late forms. The analyses also reveal other compositional
tactics that Franck utilised. Chapter Five begins with a discussion of nineteenth-
century compositional issues, followed by an analysis of two of Franck’s
Symphonic Poems; Chapters Six to Nine analyse Franck’s absolute music.
Chapter Five
Issues of Nineteenth Century Composition and
124
Analysis of Le chasseur maudit and Psyché
Beethovenian forms posed an ongoing challenge to subsequent composers. We
must remember, however, that the expressive and technical landscape had also
changed. Consequently, this chapter begins with a brief overview of some
compositional issues pertinent to the nineteenth century: these must be examined
before beginning analysis. I then analyse two of César Franck’s Symphonic Poems:
Le chasseur maudit and Psyché. These Symphonic Poems are considered first because
of their clear programmatic goals and potential to provide insight into how Franck
might articulate active or passive tonal areas in absolute forms. Such
considerations are vital before examining Franck’s non-programmatic works in the
later chapters.
Compositional Techniques and Issues
Phrase Structure and Nineteenth-Century Expository Techniques
It is important to observe how the compositional and expository techniques of
composers writing in the second age of the Symphony differ from those of
composers writing in the first. This difference may also throw light on why
composers encountered problems whilst attempting to emulate earlier works: an
issue which is discussed at the beginning of Chapter Six in relation to Beethoven’s
Hammerklavier.
Expository techniques must be considered, as they are important demonstrators of
how composers handle musical material. Brahms’ string quartets are excellent
examples of a technique which ensured originality without padding: developing
variation. (The other technique we will examine is real sequence, which is
associated with Liszt and Wagner.) However, the enrichment of the fundamental
125
bass is necessary for the success of these techniques, which often (in the case of
these composers) necessitates the use of axial tonality. Dahlhaus notes that:
One small motive can become monotonous, however, no matter how
variously the material is presented, unless is it alleviated and balanced by
constant harmonic variety. The enrichment of the fundamental bass is the
correlative, both technically and aesthetically, of developing variation.382
Dahlhaus takes the ramifications for form and tonality no further, but his
comments beg the question of how the term “correlative” reinforces the connection
between the enrichment of the fundamental bass (and its co-requisite harmonic
enrichment) and these expository techniques, which were new for nineteenth-
century composers. It seems, paradoxically, that sonata form continued (and in
many cases grew enormously) despite musical ideas becoming shorter with these
new expository techniques.
Integration
In the symphonies of Haydn or Mozart, a work could be created from a number of
themes with material surrounding these themes of lesser value, which consists of
both transitional passages and what we might call “musical padding”. Dahlhaus
describes how “Both Wagner and Schoenberg complain about the mere padding to
be found in Mozart’s music, even in some of his principal themes, but this is
almost unavoidable when the musical form is made up of corresponding, well-
balanced parts which of themselves delineate the whole movement.”383
However, Beethoven represents a transition period where there are often motivic
connections between the main material and filler: “In Beethoven, formal ideas and
melodic detail come into being simultaneously: the single motive is relative to the
383 Ibid., 43.
126
whole.”384 This can be observed if we compare the first movement of Mozart’s 41st
Jupiter Symphony with the first movement of Beethoven’s Fifth Symphony. The
expositions alone are sufficiently illustrative: in the Mozart example, bars 17 and 18
are nothing more than a C major arpeggio, and bars 24 – 30, part of a bridge
passage, employ a downward C major scale. If we compare this to a corresponding
section in Beethoven’s Fifth, such as bars 44 – 56, we can see that although
Beethoven uses an arpeggiating figure, the famous opening motive saturates the
texture. This does not mean that classical compositions could not be motivically
coherent; rather, simply that in Mozart the filler material was not reliant on the
main thematic material. Furthermore, the scalic figure used at bars 24 – 30 in the
Jupiter symphony is the same as that used in the corresponding section in the
Prague Symphony. It is a useful fragment as it lends itself well to contrapuntal
development, and Mozart uses it in this way in the development sections of both
these symphonies. Beethoven does not conspicuously recycle themes in this
regard.
From these examples we can observe the effects of the Doctrine of Originality.
Ideally, musical ideas were to be self-sufficient, sustaining thematic importance
through every point. Where Beethoven saw each idea as a component, in romantic
composition a melodic unit would – ideally – be form-generating.385 This need for
originality deeply affected nineteenth-century composers: the quality of their
music was dependent on themes expounded at the start, as form was generated
from the opening ideas. Schoenberg states that: “segments or sections which fulfil
structural requirements should do so without being mere trash.”386 Each phrase
should be embedded with meaning, as superfluous padding was deemed
unacceptable. Readers may realise the irony in pairing “monumentality” with the
384 Ibid., 42.

385 Ibid., 41.
386 Arnold Schoenberg, Style and Idea (London: Lowe and Brydone, 1951), 64.
127
Doctrine of Originality – if every component of a musical composition needs to be
original, monumentality becomes even more difficult to achieve.
A comparison of Mozart and Beethoven succinctly shows the paradigm shift from
thematic material being relatively unimportant to its integration being paramount.
It will come as no surprise, then, that later nineteenth-century composers treated
their melodic ideas differently to their predecessors. For the task of maintaining
originality, or conversely eliminating padding, there emerged two expository
techniques which I shall now examine: real sequence, used predominantly by Liszt
and Wagner; and developing variation, used by Brahms.
Real ‘literal’ Sequence. The “literal” or “real” sequence used by Wagner and Liszt
differs from sequence used by Classical composers. In Haydn and Beethoven,
sequence is a developmental technique, part of the “working out” of a musical
idea. 387 However, real sequence is used to develop an idea which is musically
complete in itself, and “would not tolerate conventional ‘rounding-off’ in a closed
period.” 388 An excellent example of real sequence can be observed in Liszt’s B
minor Sonata. Three motives are stated at the opening, and through the splitting
and combining of the second two motives much of the melodic material for the
Sonata is generated.389 At bar 18, a syncopated downward flourish figure begins,
which occurs three times: each time a tone higher. There is no definite key centre as
the harmony consists of diminished sevenths. At bar 25, however we arrive
unambiguously at E-flat major and regular rhythm ensues. This is a significant
moment, highlighted by the ff marking and metric stability, especially given that E-
flat major is Iα. The effect of using real sequence before the moment in E-flat major
is dramatic, as the listener is jolted from a period of tonal flux to the E-flat major
387 Dahlhaus, Nineteenth Century Music, 46.

388 Ibid.
389 The first motive, a downward scalic G – G figure, is not utilised for combinational
purposes: rather, it occurs more frequently at important “hinge” points in the Sonata.
128
goal. Liszt seems intent on emphasising the neutral anchors of the home tonic –
that is, the tonic axial relatives – and because these relatives sound functionally
ambivalent, there remains the possibility of a genuine, oppositional second tonal
area. As we shall see when we return to the discussion of this work, functional
areas are preserved through axial modulation throughout the Sonata.
Though real sequence is employed more frequently in large scale symphonic
works or music drama, and developing variation more common to chamber music,
the two techniques are alternative responses to the same desire for originality.
Although Brahms is most often associated with the technique of developing
variation, his G minor Fantasy for piano, Op. 79, uses real sequence. The opening
appears to be in a state of tonal flux, with D minor, F major and G major all
touched on by way of real sequence. G minor is suggested in bars 11 – 12, but not
fully expounded until the development section at bars 61 – 85,390 where the second
subject group appears in the dominant, confirming the tonality. Dahlhaus asserts
that this tonal flux is achieved due to sonata form being second nature to listeners
in the nineteenth century, and by the employment of D minor as a second subject
area, G minor itself is confirmed. It is only retrospectively that G minor is
recognised as the “common denominator” of the first subject group, becoming the
tonic the ear demands. 391 We might suggest that axially related keys provide
opportunity for modulation without dramatising the opposition of tonic and
dominant too early – cadences occur in G major and B major, I and Iα respectively,
and the opening of the work also features axial relatives, with the first two
sequential occurrences beginning in E-flat major and G major, Iβ and I
respectively. These are but two works which suggest that both Liszt and Brahms
were aware of the potential of axial modulation in effecting functional neutrality,
allowing a truly oppositional tonicising second tonal area to emerge.

391 Ibid., 71.
129
Developing Variation. In Style and Idea, Schoenberg proclaims to be Brahms a
“great progressive”, adding that this may seem “contestable to an incarnate ‘old-
Wagnerian’ . . . [as they considered] themselves entitled to look with contempt at
Brahms, the classicist, the academician.”392 Schoenberg is drawing on the popularly
held idea that Brahms was part of a “conservative” symphonic school, with Liszt
and Wagner constituting the “new German” school. However, Schoenberg asserts
that this paints an inaccurate picture of Brahms, noting that: “there was as much
organisational order, if not pedantry in Wagner as there was daring courage, if not
even bizarre fantasy in Brahms.”393 In spite of the partisan differences that seemed
to exist between these composers it is easy to overlook that Brahms faced the same
problem as Liszt and Wagner: that of sustaining form and originality
simultaneously. 394
However, the solution Brahms most often sought was
developing variation. Like real sequence, developing variation is an expository
procedure, where elaboration of a thematic idea is the primary formal principle. 395
Schoenberg states that: “the most important capacity of a composer is to cast a
glance into the most remote future of his themes or motives.” 396 In other words, a
composer must know the potential and direction of his theme as he begins
composing, as this will affect the composition from the outset. For Brahms’
developing variation technique, this potential was both rhythmic and melodic.
However, any rhythmic potential must be distinguished from mere rhythmic
variation: Schoenberg observes that whilst Schubert and Schumann frequently
varied themes melodically but retained rhythmic similarity, Brahms “repeated
phrases, motives and other structural ingredients of themes only in varied
392 Schoenberg, Style and Idea, 56.

393 Ibid.
395 Ibid., 49.
396 Schoenberg, Style and Idea, 80.
130
forms.”397 Furthermore, developing variation in both a melodic and rhythmic sense
should not be confused with thematic transformation (discussed in Chapter One):
a longer-range unifying device more akin to re-birth than this variation technique
which often introduces its variant after a statement of the theme. Figure 5.1 is an
example of developing variation.
Figure 5.1: Brahms, Symphony No. 2/iii (Op. 73), bars 1 – 14
Developing variation is more closely related to a linkage technique, whereas
thematic transformation usually preserves rhythmic integrity in terms of
anacrustic or thetic events within the theme, whilst varying other parameters.
Dahlhaus examines Brahms’ developing variation closely in the G minor Piano
Quartet Op. 25. The first bar (D – B-flat – F-sharp – G) provides material for the
first eight bars through transposition and free inversion. The second part, built on
a descending second, is repeated sequentially and imitatively, and altered
rhythmically. Dahlhaus notes that: “Compositional economy, the building of
musical interest out of minimal capital, was taken to extremes by Brahms.” 398
However, when developmental techniques are employed in the exposition, sonata
form is altered by necessity. Dahlhaus observes how:
Sonata form takes on a different meaning from the one it originally had
when motivic development, the elaboration of thematic ideas, becomes the
397 Ibid., 185.

131
primary structural principle, in place of the pattern of key relationships
and the construction of symmetrical groups.399
As mentioned earlier, for Dahlhaus the enrichment of the fundamental bass is the
correlative of developing variation. As Brahms had expanded the number of
degrees in regular use, this endangered regular periodic structure at it existed in
1800.400 At this time, harmonic foundations were simple models such as I – V – V –
I, as we see in principal themes of Mozart’s Jupiter Symphony, for example.
Dahlhaus argues that periodic structure and harmonic enrichment can coexist, but
also notes “the procedure is self-defeating as long as the technical and aesthetic
criterion is the rule that development of all the elements of the composition should
be analogous.”401 Dahlhaus asserts that Brahms averts this by employing Bach’s
technique of composing tripartite groups which evolve from an initial phrase
through harmonically enriched developmental passages, rather than being tied
down to a cadential structure. However, Dahlhaus concludes by asserting: “this is
not the restoration of something that belongs to the past; it is a derivation or an
analogy made under fundamentally different historical conditions.” 402 We must
ensure, then, that when examining instances of developing variation, its effect on
structure is also considered.
As a brief example from another genre, Brahms’ second Piano Concerto illustrates
how harmonic enrichment can allow modulation in traditionally mono-tonal areas
whilst harnessing the axial systems’ functional neutrality to reinforce traditional
tonic/dominant opposition. 403 The problem concerto composers faced – and this
was as for Mozart as for Brahms – was how to provide harmonic interest in the
opening orchestral ritornello, as modulation is not an option before the soloist
399 Ibid., 61.

400 Ibid., 63.
401 Ibid., 62.
402 Ibid., 64.
403 Mozart uses the tonic minor briefly in his Piano Concerto K. 467, from bar 110.
132
enters at the second ritornello lest the music become tautological. Rosen articulates
this concept:
The first ritornello continued, as before, to begin and end in the tonic . . . but as
the first ritornello expanded to symphonic length, casting all of its themes and
motives in the tonic was not a simple matter . . . it was . . . difficult to handle a
succession of themes all in the same key without the danger of monotony.404
In this Concerto, Brahms maintains functional neutrality and averts this
“monotony” in the first ritornello by use of D minor, the iα relative to the tonic B
flat major, effectively expanding the tonic area. This ritornello occurs from bars 48
and continues until the piano re-entry.405 This might be considered a Schubertian
solution. Downes writes that: “For Schubert, the tonal interval of the major third
clearly had the capacity for what could almost be called passive modulation. This
property was exploited to increase the breadth and scale of some of his later
sonata-form works, without jeopardising the set functions of the various structural
sections.” 406 This major-third modulation allowed enrichment without wrecking
sonata opposition, and it is just as applicable to the concerto as it is to other genres.
Liszt and Brahms seem to be using axial systems for similar reasons. The works
discussed above show how composers can create harmonic goals which are non-
tonicising or passive chromatic incursions produced from expository techniques.
Such incursions can be used in conjunction with real sequence as well as
developing variation. However, these goals do not disrupt the fundamental tonal
opposition of sonata form. How such incursions might be adapted to
programmatic ends is considered later in this chapter.
Double-tonic complexes
404 Charles Rosen, Sonata Forms (New York: W.W. Norton & Company, 1980), 73.
405 The first piano concerto in D minor also incorporates tonic axial relatives.
Otago, 1994), 32.
133
As the notion of proto-progressive tonality is significant to this thesis, it is
pertinent to also examine a correlative: double-tonic complexes. Related to the idea
of progressive tonality (or proto-progressive tonality as discussed in the previous
chapter), Deborah Stein describes double-tonic complexes as possessing: “a
complex formal design in which the traditional common-practice polarity of two
closely related harmonies is replaced by a tension between two opposing and
remotely related tonalities.”407 The reader can see how this idea can be easily related
to the concept of progressive (or proto-progressive) tonality: two tonalities vie for
structural importance within a work. Yet double-tonic complexes are not
necessarily so teleological. Bribitzer-Stull states that:

While directional [similar concept to progressive] tonality comprises the
transformation of tonic function from one tonic chord (key) to another across the
span of a piece, the two tonic keys of a double-tonic complex are not simply its
opening and closing tonal centres: rather, they are the dual harmonic poles
between which the music oscillates, at one point suggesting one key, and on
another occasion the other.408
This may well be a relevant concept when looking at some of Franck’s works:
particularly Psyché, with its evocations of a dream-like state and remote
modulations. In this case, a double-tonic complex also possesses potential to
articulate a programme – for instance, a shift from sleeping to waking.

Having examined some important compositional issues, let us now consider the
implications they may have on Franck’s works, beginning with analysis of two Symphonic
Poems.
Franck’s Symphonic Poems
Franck’s Symphonic Poems may seem an unusual analytical choice given that most
of this thesis is concerned with absolute music. However, Franck – like Brahms and
407 Matthew Bribitzer-Stull, "The End of Die Feen and Wagner's Beginnings: Multiple
Approaches to an Early Example of Double-Tonic Complex, Associative Theme and
Wagnerian Form," Music Analysis 25, no. 3 (2006): 324.
408 Ibid.
134
others – was clearly interested in more than one compositional method, an interest
demonstrated by his use of different formal models in works analyzed later. Whilst
some of Franck’s absolute forms relate strongly to those of late Beethoven, these
are novel; his approach to form in these Symphonic Poems is mediated by the
dramatic demands of the chosen programmes. How Franck manipulates axial
tonality to these ends may inform our analysis of his absolute works.
Franck’s Le chasseur maudit shares characteristics with Liszt’s B minor sonata: not the
double function aspect, but the division of the work into sections which perform
different functions. (The same may be said of a number of Symphonic Poems of
Liszt, Franck and others.) Le chasseur maudit is particularly interesting, however,
considering the relationships between subject matter and the notions of activity and
inactivity. This designation may be applicable to immanent and transcendent terms.
Psyché, written between 1887 and 1888, is quite a different specimen, instead looking
forward to the impressionist techniques of Debussy. 409 This too relates to the
programme put forward by the subject matter, and is one of Franck’s most
harmonically innovative works.
Le chasseur maudit
Le chasseur maudit was composed in 1882 and first performed March 1883. It is based
on a ballad by Bürger, and:

narrates the terrifying adventure that befell a certain Rhenish count, who, one
Sunday morning, defying the holiness of the day, dashed forth upon a fantastic
hunting expedition while all the bells and the chanting of the church choirs were
sounding on every side. A menacing voice sounded a warning: ‘Sacrilege will lead
409James R. Briscoe, "Debussy, Franck, and the 'Idea of Sacrifice'," Revue belge de Musicologie /
Belgisch Tijdschrift voor Muziekwetenschap 45, no. César Franck et son temps (1991): 28. It is also
possible that Debussy influenced Franck: the former was a pupil at the Paris Conservatoire
for 11 years, much of which overlapped with Franck’s Professorship.
135
eternally to the fires of hell.’ Frightened by its tones, he races on even more quickly,
pursued by a horde of devils.410
A horn call begins the Symphonic Poem, before a gentle theme that represents the
singing of the church choirs. The horn call then returns and is expanded to represent
the riding of the hunters. This theme occurs many times, which I have represented
as RoH in the analysis and thereafter. The analysis here demonstrates how axial
tonality can be manipulated for dramatic effect.
Figure 5.2: Franck, Le chasseur maudit

G major/minor
Andantino quasi allegretto
1 – 76 G I (Though sounds like IV): horn call, Church
theme
(Key change: G minor)
L’istesso Tempo
77 – 92 G minor i ‘Announcement’ in horn; riding of the hunters
Poco piu animato
93 – 128 G minor i Sequence in minor thirds. RoH at bar 114.
129 – 139 F minor - V/B minor WT related – V/iα
140 – 191 B minor iα RoH theme, sequential movement
192 – 202 G I Sequential movement by thirds
(Key change: E-flat)
203 – 206 E-flat Iβ Based on RoH theme
207 – 210 Sequence
211 – 214 F major WT related RoH theme
215 – 222 Sequence, preparing for G minor
Un Poco piu animato
223 – 245 G minor i RoH theme, extended, triple forte
Poco meno vivo
246 – 272 Half-diminished chords on C-sharp and F-sharp (262); RoH theme, new feel
410 Léon Vallas, César Franck, (George G. Harrap & Co. Ltd., 1951), 176.
136
(Key change: B minor)
Molto lento
273 – 282 B minor iα ‘Curse section’ begins
283 – 292 D minor v As 273 – 282, but a minor third
293 – 311 Semitonal sequential movement
Piu animato
312 – 327 More sequential movement
Un poco piu animato
328 – 335 Sequence, preparation for G minor
336 – 345 G minor i Climax on G diminished – A-flat
German 6th – A ½ diminished chords
346 – 368 G minor i Whole RoH theme, variants, more V
369 – 386 A minor WT related – v/V As above, then V/G at 387
387 – 392 On D minor v/i Triplet figure
393 – 397 On B minor 7 iα7
398 – 405 B-flat minor vβ
Poco piu Allegro
406 – 425 Sequential patterns from RoH in F-sharp minor heading to G
Quasi presto
426 – 449 G minor i
450 – 453 G minor – F minor i – WT related
454 – 510 Dominant preparation for i, fortissimo
510 – 542 G minor i RoH theme eventually extinguished.
Clearly, Franck is using axial relationships (among other techniques) to generate
drama. More specifically though, it seems that particular types of harmonic
movement, through giving an impression of action or inaction, help convey the
story: action and inaction are virtually personified through harmony.
The opening horn call suggests D, and hence G sounds like the subdominant when it
arrives. This generates tension, as from bar 13 there is opposition between G and D,
with D (V7) over a G pedal point alternating with an A minor7 chord. This seems
137
programmatically linked to inaction; a musical embodiment of contemplation that
should be Sunday worship. We know that the protagonist does not subscribe to this
contemplation, so how this will play out musically remains to be seen. The
following section is largely in the tonic axis. At bar 37: the church choirs theme
appears in the celli. E-flats and B-flats are used in the texture, over a G pedal, and
the music builds to a gentle climax at bar 61. The cadences are largely plagal.
Bar 77 (L’istesso Tempo) presents an “announcement” from the horn, and the tonality
shifts to G minor. This is the “riding of the hunters” theme (RoH). Though there is a
sense of foreboding from the mode and driving rhythm, voice-leading remains
plagal, and harmonies alternate between G major and minor. Programmatically, the
shift of tonality and the introduction of the RoH theme might be seen to represent
the Count defying the “holiness of the day”, heading into the woods to hunt instead.
Between bars 77 and 92 the music seems as though it is struggling to gain impetus,
as there are four fermatas between these bars. However, these fermatas cease at the
Poco piu animato at 93. Simultaneous use of A-flats and F-sharps resolving like
German sixths are rife: these chords alternate with G minor, eventually becoming a
“proper” German sixth of C minor at bar 100. Despite this, C minor remains
unconfirmed, and G minor reasserts itself in the next section. We thus feel a sense of
stasis in the music.
At bar 105, the first sequence begins on minor thirds: F-sharp – A – C then D, which
eventually suggests the dominant of G minor; unsuccessful in its sequential
modulation effort. There is a G minor arrival with tutti forces at bar 114, using the
RoH theme; this is restated in B-flat major in bar 118. This minor third relationship is
emphasised again between bars 125 – 127. Although the thematic material is
bombastic, with a driving rhythm, the music does not harmonically progress
anywhere. I suggest this is because the modulations have been based around minor
thirds: it may sound as though we have been moving – indeed, the music has – but
138
as minor thirds equally divide the octave, if the music continues to modulate in this
way it may simply end up back where it started, and the music actually returns to
the tonic. So, despite an attempt to establish a second tonal area, the music has
“failed”, and we can ascertain programmatic indications from this: Franck seems to
be indicating that effort in an inappropriate endeavour is futile. From bar 129 the
music remains in the tonic axis, by a move first to F minor (emphatically – a
wholetone relative to G minor), then B minor – iα – at bar 139.
Franck emphasises the Neapolitan of B minor (spelled as German sixth), which then
becomes an actual German sixth of E minor, resolving in bar 163. E minor’s
Neapolitan then appears, but this time it resolves to C major (B minor’s Neapolitan)
at bar 167. More sequential material follows using minor thirds, beginning on A-
sharp and terminating on the dominant of B minor, before another B minor arrival at
bar 177. The return of the RoH theme then brings a D major restatement after the B
minor, finally transitioning back at bar 190; this parallels the G minor / B-flat
relationship of the initial statement of the main theme.
It seems that Franck is using a type of double exposition, where the initial expository
material is repeated almost exactly in another key. Franck also uses this technique in
his Symphony, where the exposition plays the initial material in both D minor (i)
and F minor (Vβ). In Le chasseur maudit, however, Franck uses two tonic axis keys,
effectively making the repeat less disruptive and more inactive (the Symphony in D
minor is examined more closely in Chapter Eight). Though it may seem from an
initial glance that this technique is related to the problematising of the tonic as in
Beethoven’s Op. 102 No. 1, the effect here does not actually undermine the tonic:
rather, it reinforces the idea of activity without gain. Shakespeare’s Macbeth put it
well: “it is a tale / Told by an idiot, full of sound and fury, / Signifying nothing.”411
411 Shakespeare. MacBeth, Act V Sc. V.

139
Unsurprisingly, we end up back in G major via B as a pivot note at bar 191, where
there is a localised sequence through minor thirds, though this time the music
resolves into A minor at bar 199. At bar 203, however, there is a sudden shift
through A major and B-flat major and we are thrust into E-flat major, the only tonic
axis key to not be emphasized so far. The thematic material is still based around the
RoH theme, and as such seems immensely repetitive. So, from the material thus far,
we can ascertain that the main key relations are axial: G-B-E-flat. These key relations
(and the localised minor third sequences) reinforce the idea that the huntsman is
running around in circles – there is a sense of great exertion in the transitional
sections – and achieving nothing. One more step and he will be back at square one,
and sure enough, at bar 223 the music returns to G minor, the tonic. Overall, the
music seems stuck in the tonic axis.
From what we have observed so far, passive axial modulation is here stigmatised as
negative and pointless. There are other factors contributing to the perceived
“uselessness” – tempo, figuration, and the monotonous repetition of the RoH theme,
and these combine with a tonality unable to achieve itself in a sonata sense. On this
basis, it would be logical to assume that Franck would carry the technique over into
his absolute works, to assert inactivity of some sort in the way his contemporaries
and forebears did: Beethoven, Schubert, Liszt, Brahms and others.
At bar 246 we reach the “curse” section. There is indeed a new feel to the music,
with rustling strings and diminished chords. By bar 273, we begin to sense the
severity of the “curse”: the music begins in B minor, but aggressive minor third and
other non-wholetone modulations begin, through D minor, F minor, E minor and E-
flat major. Franck often instigates modulation by introducing the German sixth of a
key and then reinterpreting it. This is one of Franck’s favourite devices, both in this
work and others. The music eventually makes its way to F-sharp minor (Vα) at bar
310. This contrasts markedly with the stuck-in-the-tonic-axis activity up to now, but
140
this music represents the marshalling of the spirit forces that will doom the hunter to
hell. His actions (hunting) achieve nothing as he is stuck in the tonic axis of G-B-E-
flat, but the actions of the spirit forces will have profound consequences. It is a
marvellous and perfectly logical application of tonal forces to depict the impotent
hunter and the far from impotent demons.
Now a new section begins: Piu animato. It is a harmonically complex section with
significant wholetone modulation: sinister murmurings in the strings move through
F-sharp minor, D major, and G-sharp major. Then the pattern begins again in A
minor and shifts through F major, then B major. The music then moves through C
minor, D minor, and E minor (all of these with added sixths) then G (V of C). The
climax of the section (and arguably the work as a whole) is the fortissimo tutti at bar
336, hammering G diminished, A-flat German sixth type and A half diminished
chords with a driving dotted rhythm. At bar 340 this German sixth type chord
resolves onto E-flat with G in the bass, before another fortissimo chord on E-flat, G
and B natural: the augmented triad on tonic axis. This is a logical manifestation of
whole-tonality given the relationships that precede it. At bar 344 we have a
chromatic resolution into G minor, using the dotted part of the RoH theme. This
time, however, the dominant is more emphatic: it articulates G minor in the fourth
bar of the phrase for the first time in the life of this motive. Furthermore, Franck
extends the motive of a rising third to include a perfect fourth as well as its usual
minor and major thirds, such as bars 358 – 360. The snippet of the RoH theme recurs
in A minor at bar 369, and we get back to V/ G at bar 386, before an unexpected twist
in the form of a B minor7 chord jolts us in bar 393.
Then, at the Poco Piu allegro at bar 406, the music moves into F-sharp minor, vα. This
leads into a sequence beginning at bar 410, in which the bass moves downward in
wholetone steps then jumps a sixth to begin again. Hence the second instance begins
on A in bar 414, and the third on C in bar 418. The third time, however, the
141
chromatic steps between C – B – B-flat are emphasised and eventually become B-flat
– A – A-flat in bar 425. This bar ends with a diminished seventh on A-flat, which
resolves into G minor, becoming the first bar of the Quasi presto. The Quasi Presto
seems to be an elaboration on gains greater emphasis with E-flat-D and B-flat-A in
the bass.
At bar 454 we return to B – unsurprising – but it is spelled as C-flat major. Sequential
motion follows, before the critical culmination of the Symphonic Poem: at bar 478
there is alternation between V and German sixth chords, with different rhythmic
emphasis each time. We are reminded of the important role rhythm plays in any
type of resolution. This continues until we reach a D pedal at bar 506. This resolves
as expected into G at bar 510, although the flat sixth and sharpened seventh scale
degrees remain in the texture. There is then a gradual decrescendo toward the end
before ending with a fortissimo chord in G minor.
Le chasseur maudit uses many of Franck’s familiar harmonic twists. The “wrong”
direction around the cycle of fifths (V – ii) occurs several times before it is
“corrected” near the end of the Symphonic Poem. The first half of the work uses
mainly tonic and subdominant axis keys with largely plagal voice-leading
(particularly in the 9/8 RoH theme). Franck also uses a double exposition, as the RoH
material is repeated almost exactly in B minor. Toward the end of the work,
however, more dominant axis keys are employed as well as dominant voice-leading,
introduced in the “curse” section. There is a great deal of foreground modulation,
but this usually does not stray far from the tonic. Many of the transitory sections
seem repetitive, and, like many of Franck’s works, fermatas cause a great deal of
stopping and starting. (This aspect of the Symphony has been criticised: however,
this may work better in a Symphonic Poem in reference to a programme.) Significant
142
emphasis also occurs on ♭VI both as a key area, scale degree and chord, often as a
German sixth.
The most important aspect to note, however, is Franck’s use of harmonic areas to
illustrate the content of the Symphonic Poem. Axially related keys are used to
illustrate and dramatise goal-lessness, and their use emphasises the inevitability of
the hunter’s fate – he cannot escape as he is stuck in the tonic axis. Minor third and
non-wholetone modulations dramatise action, forcing the hunter to hell, and
reinforcing the “active” nature of such relationships and the inevitable triumph of
the spirit forces.
However, this only partially explains the musical picture. Though the potential
impotence of the hunter can be appreciated, it seems that Franck imposes his own
moral perspective on the work, by way of major-third modulation. Franck imposes
the major-third modulation, but the hunter does not think this way: hence, the
huntsman lacks agency, undermining the overall drama of Le chasseur maudit. We
might consider Franck an “intrusive narrator”, in that he not only illustrates the
narrative musically, but also “freely comments on his characters, evaluating their
actions and motives and expressing his views about human life in general.” 412 This is
not a criticism of moral character – rather an observation that by allowing his
judgement to affect his compositional technique, Franck extinguishes the tonal
drama; the huntsman’s agency is not humanised, the formal result being randomly
linear. The potential alternative would have been an omniscient narrator, such as
Schubert’s Erlking – Schubert does not judge, he only observes. Franck’s intrusion
causes some similar issues in his Symphonic Poem Psyché, which will now be
considered.
M. H. Abrams, A Glossary of Literary Terms, 3rd ed. (Holt, Rinehart and Winston, Inc., 1971),
412
134.
143
Psyché
Franck’s original Symphonic Poem was in three sections and included chorus. The
version I have taken for analysis is the Symphonic Suite Franck wrote based on the
original, as a survey of recordings shows that excerpts from this are more frequently
recorded and performed than the work in its entirety with chorus.
A short synopsis helps impart the moods Franck may have wished to convey in the
music.413 Psyché, is a beautiful girl; so beautiful that people forget about the goddess
Venus. Venus becomes angry and plots to ruin Psyché, so instructs Cupid, her son,
to pierce Psyché with his arrow and make her fall in love with the most hideous man
alive. However, when Cupid sees how beautiful Psyché is, he pierces himself with
the arrow instead. Cupid and Psyché marry without Psyché ever seeing Cupid.
When Psyché’s sisters visit her, they surmise that she has never seen her husband,
and convince her to sneak a look at him. Conflicted, Psyché does so, and sees the
beautiful Cupid lying beside her. She weeps for her lack of faith. Cupid awakes and
deserts her, as love cannot live where there is no trust. Cupid returns to Venus, who
again decides to enact revenge on Psyché. She goes to Venus for help, and Venus
sets her impossible tasks to complete if she wants to see Cupid again. Against all
odds, Psyché completes all the tasks, the last of which is to go to the underworld and
collect some of Persephone’s beauty in a box. Psyché does so, but on her journey to
return to Venus, she becomes curious, opens the box, and immediately falls asleep.
Cupid finds her sleeping, and takes her to Zeus to request her immortality. Zeus
grants this, and Cupid and Psyche are married.
The original three parts were as follows: one, “Sommeil de Psyché”, (“Psyche
Asleep”); “Psyché enlevée par les zephyrs”(“Psyché awakened by the breezes”);
two, “Les Jardins d’Eros” (“The Garden of Cupid”); [with chorus], “Psyché et Eros”
Edith Hamilton, "Mythology Summary and Analysis,"

413
www.gradesaver.com/mythology/study-guide/section6.
144
(“Cupid and Psyche”), a love duet; and three, “Le Châtiment (souffrances et plaints
de Psyché)” (“Punishment: Psyche’s sufferings and griefs”) a tragic episode, then
finally; “Apothéose” (“The Apotheosis”) representing the forgiveness of Eros. 414 The
suite is entirely instrumental: Sommeil de Psyché; Psyché enlevée par les Zéphirs;
Les jardins d’Eros; Psyché et Eros. Whilst it would be inappropriate to take the four
sketches as a teleological story, we might consider the individual moods evoked by
particular titles and what Franck would have wanted to convey.
This movement is titled “Psyché asleep”.
Figure 5.3: Franck, Sommeil De Psyché (Sketch 1)

B major
Lento
1–9 B major – minor I–i #1 in Clarinet
10 – 17 D major – minor Vβ - vβ #1 Oboe and Flute
18 – 21 F-sharp minor v [18: Reh. A]
22 – 31 B major/ G-sharp minor I & ivα
26 – 29: G-sharp 1st inversion – B / E 1st inversion – G# / A# ½ dim to G# min (x2)
32 – 39 B major I [36: Reh. B]
40 – 46 B major/ G-sharp minor I – ivα
47 – 51 G-sharp minor ivα
52 – 53 Preparation for A-flat – A-sharp ½ dim and C-sharp major
54 – 60 A-flat – G-sharp ivα [54: Reh. C; 59: Reh. D]
61 – 64 C-sharp major/ minor WT to tonic axis
65 – 66 Transition (as bars 8 – 9 and elsewhere, melody emphasising 7th)
67 – 70 E-flat major WT to tonic axis
71 – 72 Transition (as 65 – 66) [71: Reh. E]
73 – 80 F minor, then transition WT to tonic axis
Diminished 7th (1/2) built on G, almost sounds like German sixth of B
81 – 84 V/B, then ambiguous

145
85 – 90 C-sharp7 – B 1st inversion / E minor – G-sharp minor / A-sharp ½ dim to G-sharp (x2)
/ B / V/B
91 – 94 B major to V/G-sharp I to V/ivα
95 – 98 G-sharp minor [95: Reh. F]
99 – 101 B/ G-sharp minor/ G minor I – ivα - Iβ
102 – 112 B I
G-sharp makes a strong claim for tonicity, but B major is both the opening and
closing key. However, this sketch feels tonally flat, and given its title – “Psyché
asleep” – this seems entirely natural. In fact, Psyché is one of Franck’s stronger
orchestral works because Franck’s harmonic procedures suit the text entirely, and by
this point in his career (Psyché was composed between 1887 and 1888) he was
beginning to handle orchestration in a more sophisticated manner. Franck’s picture
of Psyché is a very naïve one, evidenced by the relative simplicity of the motif
between Reh. A+14 – 17. Here, Cohn might argue that tonality is on the verge of
being annihilated: a vestige of the tone-row (discussed in Chapter Two) appears in
Reh. F+5 – 6, with weak harmonic movement, which places new importance on the
foreground.
Whilst other works by Franck use axial tonality at a middle or background level, this
work uses it in the foreground. This is not radically new – Franck does use such
gestures in other works – but here the foreground axial gestures are used much
more frequently and saturate the texture. The opening accompaniment, for example,
uses axial relatives as foreground harmonies: B major to G major in the string figure,
bars 4 – 5. Although the largely triadic harmony provides a sense of key, there is
actually little functional harmony: chordal movement is by major and minor third
(such as bars 8 – 9, which move from B major to D major, and bars 12 – 13 which
move from D major – B-flat major) with sevenths and ninths often present; this
harmonic movement accompanies a melody rising through a tonic triad and its
major seventh. The gentle tempo and ambiguity of time signature (between 3/4 and
146
6/8) slow the harmonic rhythm, so that directional harmony would be difficult to
sense regardless. This movement through a major seventh chord occurs again
between Reh. A-2 – 1, this time from D major – F-sharp minor (a flat, axial shift). We
might also note that when voices are not moving triadically, they are usually only
moving by semitone or tone; this contributes to the feeling of contracting and
retracting. The movement’s first strong cadential point occurs at Reh. B-4: a perfect
cadence into B major.
Because of the fluidity of key and the closeness of minor keys to their relative major,
the emphasis on G-sharp is not surprising – though we should note that it might also
function as IVα. After Reh. B-4 the harmony becomes more functional: a circle of
fifths movement precedes a cadence into G-sharp at Reh. B+11, approached by an
augmented cadence on Iα encompassing all the pitches of the tonic axis. This arrival
coincides with tutti forces at Reh. C, and G-sharp turns enharmonically to A-flat.
However, the harmonic movement in the following bars (F minor and E at C+1 – 2
respectively) means that when we hear G-sharp only four bars after the tutti
announcement of Reh. C, it sounds foreign. G-sharp then acts as V/ C-sharp, paired
with the first theme. At Reh. D+6 the major seventh melodic idea returns, inciting a
harmonic shift through E-major to F and E-flat (enharmonically Iα), confirmed with
a perfect cadence. Another movement through the major seventh melody in that key
begins at Reh. E.
Between Reh. E+2 and E+9 there is a transitional period, largely using the
syncopated melodic material and keys wholetone related to the tonic. We sense a
strong arrival in the dominant at bar 81 (F-sharp), but by Reh. E+14we are back in G-
sharp minor with the major seventh theme. A foreground dominant at Reh. F-5
tonicises an arrival of B at Reh. F-4, but again, four bars later, we are back in G-sharp
minor, with a tutti arrival at Reh. F+4. However, this is but part of a shift back to B,
the harmony being G-sharp minor (second inversion, then root position) – G minor –
147
B major. We then remain in this key for the rest of the movement, which ends very
gently.
It is interesting to compare the harmonic movement here with that of Le chasseur
maudit. In Psyché, the overall feeling is a luxurious one; of space and of tonal flatness.
Modulation occurs, to be sure, but usually either to tonic and subdominant axis
members, also encompassing the relative minor. Melodic material is largely triadic;
other voices move predominantly by step. This can easily be linked to the idea of
sleep; the gentle musical stasis underlines the subject matter and feels entirely
appropriate. Psyché is not “active” – very much the opposite – and the musical
materials reflect this notion. By comparison, the protagonist in Le chasseur maudit is
highly active and very much in control; whether his decisions are moral or not, he
decides his destiny, and his activity is reinforced by the active nature of the harmony
in the Symphonic Poem. These works contrast very effectively.
We turn now to the second sketch: “Psyché Enlevée par les Zéphirs”. Though Franck
provides no key signature here, this work is analysed in G major, the key of the
opening section. G is equivocal, however, and the music only asserts itself with the
arrival of C major, and continues to modulate until the end of the movement.
Figure 5.4: Franck, Psyché Enlevée par les Zéphirs (Sketch 2)

Begins in G major
Allegro vivo
1 – 30 G major I [17: Reh. G]
31 – 38 A-flat7 (V/D-flat) V/WT1 – unsubstantiated [35: Reh. H]
39 – 54 C major IV
55 – 58 Transition V/E-flat [55: Reh. I]
59 – 68 E-flat Iβ
69 – 79 A-flat IVβ [79: Reh. K]
80 – 99 Transition [93: Reh. L]
148
100 – 108 V/D-flat V/WT1 [105: Reh. M]
109 – 116 D-flat major WT1
117 – 124 B-flat minor vβ (or ivα of D-flat)
125 – 135 D-flat major WT1
This movement is based on Les Eolides, a Symphonic Poem that Franck composed
between 1875 and 1876. However, this sketch is much shorter than the original,
aiming only to capture something of the atmosphere of “Psyché awakened by the
breezes”. 415 This version uses the augmented triad a great deal, which diverts
importance away from the tonality and onto other features, such as the melodic
material and orchestration. The music begins on open Bs, which become a G major
chord. Then a specific harmonic movement enters which deserves discussion, as it is
one of the distinguishing features of this movement.
Figure 5.5: Franck, Psyché Enlevée par les Zéphirs (Sketch 2) bars 9 – 10
This harmonic cell (arguably the most memorable in the
movement) uses the flat-sixth and sharp-fourth scale
degrees. This chord is related to the German sixth
because it contains these tones (though the third of the
initial triad is major and hence not an actual German
sixth); alternatively it might be described as an augmented triad with an additional
sharpened fourth. It possesses a similar function to that of the German sixth though;
both tones require resolution, and indeed the tones move in this way. Here, Franck
uses a German sixth type chord in a way that differs from the traditional use; this is
also discussed in Chapter Nine regarding the first movement of Franck’s Violin
Sonata.
Les Eolides was based on a poem by Leconte de Lisle: "Floating breezes of the skies . . .
415
implanting kisses with fickle tenderness on the mountains and plains." Laurence Davies,
César Franck and His Circle (Boston: Houghton Mifflin Co., 1970), 209.
149
Triplet flourishes in flute and clarinet – based around the augmented triad G – B –
D-sharp – canter back and forth. Moving through D9 – F9 – A♭9 (active minor third
as in Le chasseur maudit) our first strong sense of key is C major at bar 39. However,
A-flats are included here, as the above notated harmonic cell is utilised, but based
around a C major triad. Poco piu lento shifts the tempo suddenly at bar 59, and from
here to bar 79 the music is exactly reminiscent of bars 488 – 507 in Les Eolides,
transposed a semitone higher (this section of Les Eolides modulates from A – D).
Following this there is a frenzied transition with no real key center made up of
augmented, diminished and chromatic movement. Melodic snippets suggest D and
E minor, but not strongly, before an A-flat7 chord gently bursts out of the texture at
Reh. M-4. This sounds like a dominant, and resolves at Reh. M+4 to D-flat major. The
rising seventh melodic idea from the first movement returns at Reh. M+12 – molto piu
lento – and the slow Les Eolides theme ends the movement in D-flat.
One does not have to be familiar with Les Eolides to appreciate the breeze-like quality
of the melodic material here. Predominantly built around augmented triads, with
rising and falling triplet figuration ebbing and flowing, this sketch is one of Franck’s
most pictorial compositions. The timbral and rhythmic shifts, combined with the
integration of the theme from the first movement, suggest that Psyché is indeed
awakening. This is a genuinely progressively tonal work; possibly employed to
illustrate the shift from unconscious to conscious.
Figure 5.6: Franck, Les Jardins D’éros (Sketch 3)
Key interpreted as C major (common denominator and end goal)
Poco animato
1 – 10 D-flat major IVβ
11 – 20 On E-flat – A-flat – shift to A7 V/Iβ – iv/Iα
21 – 32 E – G-sharp minor (+6) Iα – Iβ 1st theme [21: Reh. A]
150
33 – 36 B7 – prep for E, then away V/Iα
37 – 44 G-sharp C-sharp (II – V in preparation for F-sharp)
II – V of WT 1 [37: Reh. B]
45 – 56 F-sharp major WT 1 2nd theme [49: Reh. C]
57 – 60 A major IVα
61 – 66 C major, then transition via B major I via Vα
67 – 70 E-flat major Vβ [67: Reh. D]
71 – 74 C major I
Un peu plus large
75 – 84 C major I
85 – 90 A minor ivα
91 – 100 Dominants of C axis – V/C, V/E (via WT prog in bass), V/A-flat [91: Reh. E]
101 – 107 Prep for C – Ger6th of A-flat; G with 7th I axis
108 – 116 C7 – F7 (sounds like prep for B-flat) V/IV – V/WT1 [108: Reh. F]
117 – 121 Trans: chromatic bass-line, dim7ths inferred in texture. C arrives but no V in
preparation
122 – 140 C major I
This movement moves more swiftly than the previous two, and begins on D-flat in
first inversion. Shifts from major to minor occur frequently, and added sixths used in
upper woodwind parts colour the key without inferring ambiguity. We reach the
main theme at Reh. A by a plagal cadence from A into E – however, immediately
prior to the A7 chord the harmony is closer to A-flat, so this is more of an axial shift
than it first seems. The main theme uses regular four-bar phrases and mirrors the
harmonic movement associated with the melodic idea that comprised a seventh
from the first movement, moving from E-major to G-sharp minor. The pattern then
repeats, with the first four bars in G-sharp minor, before moving to what seems to
spell B minor. However, because of the sevenths and ninths in the melody, this
tonality could be interpreted as B minor with added sevenths and ninths or as D
major with added sevenths and sixths. Regardless, the music then shifts to G-sharp
major and moves through the cycle of fifths to reach F-sharp major at bar 45. This is
paired with a new theme: a figure in falling fourths outlining a minor seventh triad
151
in the relative minor of the given key (for example, D-sharp minor in F-sharp major
at Reh. C-4). With the perfect cadence into F-sharp, the music feels very settled.
At Reh. C+4 a theme begins in the violins that is reminiscent of Les Eolides in that it
supports semitonal movement in the outer voices.
Figure 5.7: Franck, Le Jardins D’éros (Sketch 3), reh. C+4 – 7
It might be argued that this movement reverses the previous one – here, D-flat gives
way to C – and Franck’s progressive scheme represent a “composing out” of the
motive from Reh. C+4, together with its relative that began the movement with the
contracting motion. This is unsettling, though, and this new material begins
modulating through minor-third related keys: beginning in F-sharp, we move
through A and C to reach E-flat at Reh. D; then back in the other direction to C
before tutti forces re-articulate C’s arrival at Reh. D+8 using the material that opened
this movement. A final minor third modulation to A occurs at Reh. D+18. Then, from
Reh. E, an axial sweep begins through the dominants of C, E and A-flat, using an
altered version of the four-bar phrase theme. A-flat then becomes ♭VI of C, where
we end up at Reh. F-2. From this point on the movement becomes much flatter as the
main type of movement is axial. F is touched on briefly, before the music returns to
C major to end the movement: the final phrases use the four-bar phrase theme as
well as the rising seventh theme. The latter seems to be thematically involved in each
movement thus far. Though the middle of this movement sounds rather repetitive in
its minor third modulation (as Le chasseur maudit did) this might well be an attempt
152
at reviving harmonic drama after the luxuriance of both the opening and the
movements which have gone before. Other than this central movement, the themes
are luxurious and lyrical; fully suited to the subject matter. Of special note is the
final harmony, which is in a different key from the opening. In the next chapter,
discussing the Quintet, I discuss the possibility that Franck considered the idea of
progressive tonality but did not employ it in a chamber work because of convention.
It is debatable whether this movement of Psyché is “progressively tonal” in a
teleological sense, or whether it simply winds up in a different place to where it
started. Both are possible: the first might support a programmatic interpretation of
some sort of change or progress in Les jardins d’Eros, and the second might suggest
that Franck’s harmonic language was moving toward that which Debussy took
further: the non-functional – or perhaps here less functional – world of the
impressionistic.
Figure 5.8: Franck, Psyché et Eros (Sketch 4)

A major
Allegretto modéré
1 – 10 Transient. Oscillating on A-sharp (1/2 dim) from 1 – 6, then A7 (suggesting D) from 7
– 8, before back to A-sharp (1/2 dim) from 9 – 10.
11 – 18 Still Transient. 11 – 12: E7 2nd inversion, then as 7 – 10 with interjections from
woodwind (similar to first movement).
19 – 24 On V/A V/I B/E alternating in bass; at 23 altered to become augmented
chord (i.e. E – G-sharp – B-sharp.) [22: Reh. O]
25 – 32 A I Melody in Viola/Cello
33 – 40 Transition – suggests V/C (V/Vβ)
41 – 44 Diminished chords rising through E, F-sharp, G-sharp, A-sharp (WTs)
45 – 48 A minor i With 7th though, so functionally ambiguous
49 – 54 F-sharp – C-sharp IVα - Iα Plagal cadence into C-sharp
55 – 63 C-sharp Iα Tutti forces; theme very triadic
64 – 71 C-sharp - E Iα – V 2nd subject: with 7th. E via vii7/B [64: Reh. P]
72 – 79 E–G V – WT As 64 – 71
153
80 – 87 G – B-flat WT – IVβ Same material as 64 – 71, but in woodwind
[80: Reh. Q]
88 – 101 B-flat IVβ (V/WT) With G-flats in texture. Becomes V/E-flat at 101.
102 – 109 E-flat – F-sharp WT – IVα As 64 – 71
110 – 113 F-sharp – D-sharp IVα – WT [110:Reh. R]
114 – 121 Transition to A, via TT related diminished chords: V/A at 121
122 – 133 A major I C-sharp in bass
134 – 136 V/A V/I Briefly on E (V/I)
137 – 144 Around C: asserted plagally V/C at 144 [137: Reh. S]
145 – 148 Trans: at 41 – 44 with WTs rising in bass.
149 – 152 A minor i
153 – 160 Prep for A: through F-sharp major/minor and VII7/A
161 – 168 A major I Tutti forces – as 64 – 71
169 – 176 C major Vα Sequence, as 64 – 71
177 – 180 E-flat WT Sequence, as 64 – 71
181 – 184 G-flat Completes sequence [181:Reh. T]
185 – 191 F-sharp minor ivα Really close to A, just mixture
192 – 212 A I
Poco piu Lento
213 – 217 A I As Les Eolides and earlier in Psyché
218 – 223 D ½ dim7 chord
234 – 236 A I
The opening of this movement seems transient for an extended period, because of
the alternation between A-sharp ½ diminished and A7c chords. Even when we reach
an E7 chord at bar 11, it does not resolve to our most likely tonic, A. However, A is
indeed the most important tonal centre in this piece, with F-sharp also an important
modulatory area. Sequences are used employing the diminished chord, such as at
Reh. P – Q+7 and Reh. S+24 – T+3. However, surveying the keys in this work, almost
all are related by or reached by minor or major thirds. Some foreground dominant
activity is present, to be sure, but the main emphasis seems to be on the melodic
material, and indeed it is this that is worthy of our attention here. Between Reh. O+3
154
and O+15, the main theme in the cello is related to those of previous movements
with its triadic features; then the third movement theme appears at Reh. O+20. Then
at Reh. O+30 we encounter a new idea using the interval of a falling fourth; this
interval was part of the melodic material in the third movement.
The most significant melodic event occurs at Reh. O+29; a new theme, based on the
seventh theme used predominantly in the first movement (but which has also
appeared in the second and third movements) which now appears in C-sharp. This
theme then moves through all related minor third relatives – E, G – before different
material enters in the harmonic area of B-flat. There are then shifts through E-flat
before reaching F-sharp at Reh. R with the falling fourth theme. Harmonies continue
to shift (or perhaps a better verb would be “drift”): but at Reh. R+12 we reach A
major, lightly tonicised with its dominant. A, as a key, seems to be gathering
strength here, albeit merely through repetition. Another minor third modulation
propels us into C at Reh. S+3, and we seem to be moving toward C’s dominant: but
we are instead confronted with its dominant minor at Reh. S+8. Yet by Reh. S+12 we
are back in A minor (shifted to through an F-sharp7 chord – C – A minor) then F-
sharp at Reh. S+16. We are reaching the climax; the chords of C-sharp major, a
French sixth on G-sharp and then a dominant seventh on the same note lead us to
the new transformed seventh theme in A major, triumphant. The minor third
modulation does not cease though: we move through C, E-flat and G-flat with this
thematic material before returning to A at Reh. T+10. The music settles here, and the
falling fourth theme returns in A at Reh. T+18. A brief touch on the Les Eolides motif
at Reh. T+31 reminds us of how far the music has progressed, before the piece runs
out gently on A, using semitonal figures to confirm itself.
The important factors in this movement are more melodic than harmonic. There is
harmonic movement, to be sure, and it is almost exclusively major and minor third
relations with some foreground dominant functions. However, this movement has
155
one of Franck’s best uses of what might accurately be called thematic
transformation: the ecstatic theme that appears at Reh. S+24 in this movement is a
combination of the seventh theme (used predominantly in the first movement but
also appearing in the others) and an earlier falling theme used previously in the final
movement (related to that theme beginning at Reh. O+3).
Figure 5.9: Franck, Sommeil De Psyché (Sketch 1), bars 8 – 9 (strings)
Figure 5.10: Franck, Psyché et éros (Sketch 4), reh. O+3 – 10
In some of Franck’s works, his use of “cyclic form” is obvious, and he only recycles
themes rather than re-working them. In Psyché et Eros, however, this is clearly not
the case: the theme is transformed both melodically and in mood. What was
previously a mournful lyrical idea now becomes a joyful burst of love, with full
orchestral forces; a complete transformation of the theme’s nature.
Figure 5.11: Franck, Psyché et Eros (Sketch 4), reh. S+24 – 29
156
These Symphonic Poems demonstrate Franck’s awareness of harmonic relationships
and their effect. In Le chasseur maudit, Franck demonstrates the ability of axial
relatives and minor thirds to articulate structure and underscore activity and
underlying lack of attainment respectively. Axial relationships in Psyché are not
structurally functional, but they may be considered colouristic and underscore the
passivity: desirable given the programme. These Symphonic Poems demonstrate
Franck’s awareness of axial tonality’s potential to articulate a programme: how he
utilises such relationships in absolute music remains to be seen. With this in mind,
Chapter Six analyses Franck’s Quintet in F minor. The next chapter begins with an
analysis of Beethoven’s late formal alternatives by more closely examining the works
that use models discussed above.
Chapter Six
The Mono-Axial Model and the Quintet in F minor
157
At a glance, the Quintet seems to be broadly mono-axial, at least in terms of its
overall key scheme and significant keys within sections. However, this being one
of Franck’s first mature works, composed in 1879, we will also find in it the
enunciation of a pc set comprising the notes within the tonic axis. This, together
with aspects of axial tonality, combine to give the Quintet a uniform sonic world:
whether these elements combine coherently in a formal sense, however, is open to
debate.
The Hammerklavier’s Influence: the Mono-Axial Model
We begin with a brief survey of works which belie the Hammerklavier’s mono-axial
influence. A number of composers explored the mono-axial model, or at the very
least sensed the significance of Beethoven’s Hammerklavier. But emulating and
understanding the Hammerklavier was not necessarily straightforward, as Rosen
writes:
For [some composers] Beethoven’s achievement provoked an emulation which led,
and could have led, only to disaster. Both Mendelssohn and Brahms imitated the
Hammerklavier with singularly awkward results. The sonatas and symphonies of
Schumann are constantly embarrassed by the example of Beethoven: their
splendour breaks through his influence, but never starts from it. All that is most
interesting in the next generation is a reaction against Beethoven, or an attempt to
ignore him, a turning away into new directions: all that is weakest submits to his
power and pays him the emptiest and most sincere of homages.416
Rosen is largely correct in this statement; it is certainly true that some of the
compositions that the Hammerklavier inspired seem empty in their homage, at least
as much as their reception histories would suggest. Yet the later works of Brahms
(his Hammerklavier-inspired Sonata being his Op. 1) including his first and third
Symphonies, show a deeper engagement and more thorough understanding of
internal processes similar to those of the Hammerklavier: these we cannot describe
158
as stillborn. Brahms’ first Symphony is mono-axial in its overall key scheme
(between movements), whilst the first movement of his third relates to the
Hammerklavier’s first movement, by withholding the dominant.
Although Walter Frisch’s analysis of Brahms’ Op. 1 is primarily concerned with
thematic development (and developing variation), his analysis illustrates the
problems Brahms encountered when attempting to assimilate the Hammerklavier.
This is a reasonable assumption of influence given the uncanny likeness between
the opening themes of each, though an alternate argument might be that
Beethoven’s Waldstein Sonata, Op. 53 in C Major, was also influential to the young
Brahms. 417 Let us compare axial analyses and excerpts of the main themes of
Beethoven’s Hammerklavier with Brahms’ Op. 1 Piano Sonata.
Figure 6.1: Beethoven, Hammerklavier Sonata (Op. 106), bars 1 – 5
Figure 6.2: Beethoven, Hammerklavier Sonata (Op.106), adapted from Downes418
I
Exposition
1 – 44 B-flat I
45 – 123 G IVα
417 Walter Frisch, Brahms and the Principle of Developing Variation (Berkeley and Los Angeles:
Otago, 1994), 57.
159
Development
124 – 192 E-flat IV
193 – 200 D Vβ of IVβ
201 – 226 B IVβ
Recapitulation
227 – 238 B-flat I
239 – 266 G-flat Iβ
267 – 272 B minor ivβ
273 – 405 B-flat I
II
1 – 46 B-flat I
47 – 106 B-flat minor i
107 – 175 B-flat I
III Relative to F-sharp minor Relative to B-flat
1 – 46 F-sharp minor i Iβ
47 – 72 D iβ iα
73 – 75 F-sharp minor i iβ
76 – 84 E-flat IVα IV
125 – 155 F-sharp I Iβ
156 – 163 G IVβ IVα
164 – 165 B minor – major iv – IV ivβ – IVβ
IV Largo
1 F – D-flat – G-flat V – Vβ – Iβ
2 G-flat – B Iβ – IVβ
3–8 G-sharp minor vα of Vα
9 – 10 (A) – V of B-flat Vα – V
Allegro risoluto
11 – 52 B-flat I
53 – 83 A-flat ♭VII [WT 1]
84 – 110 G-flat Iβ
160
111 – 123 B-flat minor i
124 – 149 A-flat ♭VII [WT 1]
150 – 204 B minor ivβ
205 – 229 G IVα
230 – 249 E-flat IV
250 – 278 D Iα
279 – 409 B-flat I
Comparing this table to that of Brahms’ Op. 1, we can see that Brahms’ emulation
is predominantly in the foreground gesture, rather than the overarching key
scheme.
Figure 6.3: Brahms, Piano Sonata (Op. 1), bars 1 – 4
Figure 6.4: Brahms, Piano Sonata (Op. 1)

C major
Allegro
1 – 17 C major I
18 – 38 V/A minor V/ivα
39 – 62 A minor ivα
63 – 69 B dim – B♭7d – German Sixth on B-flat (D)

70 – 79 D minor WT 1
80 – 87 Dominant Preparation I6/4 – V
Development (Key change: C minor)
88 – 99 C minor i
100 – 117 Through E-flat (vβ) D-flat (ivβ) then sequence beginning at 111
161
(Key change: B minor)
118 – 123 Transition to B minor
124 – 131 B minor vα
132 – 135 Transitional
136 – 139 Around C minor i
140 – 148 Preparing V/V
149 – 152 V/G V/V
153 – 156 G V
157 – 160 E minor iα
161 – 166 D
167 – 172 C-sharp diminished chord
Recapitulation (Key change: C major)
173 – 180 C7 – then V/C
181 - C minor . . . then V/V i
195 - C minor . . . then V/V i
(Key change: C minor)
198 – 223 C minor i
223 – 228 Ddim7 – D♭7 - transitional
229 – 237 F minor iv
238 – Transition
250 - C V6/4 – V – I
II
C minor
Andante
1 – 57 C minor i
(Key change: C major)
58 – 86 C major I
III
E minor
Allegro molto e con fuoco – Scherzo
1 – 102 Broadly E minor iα
Più mosso – Trio (Key change, C major)
162
103 – 211 Broadly C major I
IV
C major
Allegro con fuoco
1 – 41 C I Tonicised by E opening section
42 – 76 G V
77 – 100 Sequential downward movement
101 – 106 On E V/ivα
107 – 172 A minor ivα Modulation here, but broadly A minor
173 – 186 Preparation for tonic return – V/V
187 – 193 C (7) I (7) Recapitulation, same cadence as first
movement
194 – 212 Sequential/modulatory
213 – 215 C I Strong arrival of I
216 – 227 On G7 V7
228 – 241 Around Am / F IV axis
242 – 292 C I
Though we perceive echoes of the Hammerklavier here, in the most significant sense
this work attempts to fulfill the classical requirements of sonata form; the
Hammerklavier was a re-thinking of that process. Although Brahms evidently
imitates some surface level characteristics, including the highly unusual use of the
subdominant axis as a second tonal area, it seems he has not grasped the deeper
formal procedures of the Hammerklavier. Brahms does not emulate Beethoven’s
flatness and avoidance of dominants; despite the initial “overshooting” of the tonic
in the recapitulation, this is “righted” six bars later.
One structural issue lies in the metamorphosis of the exposition’s A minor theme
in the development: a lengthy dominant preparation (V/V) precedes it. Frisch
observes that this sounds like a re-transition, and indeed it does; yet, the “real” re-
163
transition (bars 180 – 181) is comparatively weak, so we arrive at the recapitulation
“unprepared”, as Frisch explains:

The harmonies actually regress along the circle of fifths, G moving to D (bar 161),
and D to a diminished-seventh chord built on C-sharp (bar 167, sounding like vii7
of D). The C-sharp then suddenly drops to C natural, and Brahms bursts into the
recapitulation through V7/IV, having bypassed the tonic altogether.419
But why would Brahms choose to do this? We have extensively discussed the
avoidance of the dominant previously in this thesis, but if there is a place in which
it deserves at least a little attention, surely it is just prior to the recapitulation.
Frisch surmises:
The formal process and harmonic syntax of this passage reveal that, as in op. 5,
Brahms is attempting (though not particularly succeeding at) an unusual kind of
sonata structure – one that undermines the force of a recapitulation: (1) by
preceding it directly with a stable thematic episode, (2) by withholding genuine
dominant preparation, and (3) by then passing immediately through the tonic to
the subdominant.420
Briefly considering the other movements, it becomes clearer that Brahms’
emulation is in foreground detail only; the slow movement follows the first (rather
than Beethoven’s reverse) and neither it nor the Scherzo is monumentally long.
The finale movement is lengthy, and uses a varied form of the first movement’s
theme, largely in 9/8. This might be one of Brahms’ first experiments with
developing variation, or possibly a type of cyclic form; nevertheless, the
reminiscence is clear.
Brahms’ third Symphony also shares characteristics with his Op. 1. It seems he has
a habit of destabilising the sense of arrival at the recapitulation, reserving the
stronger recapitulatory event for the finale. This Symphony, however, was
composed at Brahms’ musical maturity, and arguably demonstrates a deeper
419 Frisch, Brahms and the Principle of Developing Variation, 55.

420 Ibid.
164
understanding of late Beethovenian form. Nevertheless, compared to the
corresponding section in Beethoven’s Sonata, Brahms’ comes across as “concerned
more with melodies than with formal exigencies”.421 This may explain why Brahms
chose to compose this way when it seems that he could, perhaps, have made
different compositional choices. We must remember that this was written when
Brahms was a tender 20-year-old, surely under the weight of a myriad of
influences, Beethoven being one of the most significant. One might reasonably
conclude that Brahms did not fully understand, as a young composer, all of the
internal processes at work in the Hammerklavier, though he had clearly grasped
some of them.
Another composer influenced by Beethoven’s Op. 106 was Mendelssohn, whose
Op. 106 (oddly enough) has many gestural similarities with Beethoven’s.
Figure 6.5: Mendelssohn, Piano Sonata (Op. 106) bars 1 – 5
The key, opus number, register of first subject, keys used in movements (B-flat, B-
flat minor, E and B-flat) and some sections within movements all correspond to the
Hammerklavier. Mendelssohn also uses a fugal development in G in the first
movement. Yet there are significant differences that likely influenced Rosen to
describe this as an “empty” homage: Mendelssohn does not attempt to withhold
the dominant, and so the first movement, though short, is tonally complete in itself
in a sonata form sense. This Sonata is also significantly shorter than Beethoven’s
421 Ibid.
165
Op. 106, with a recording of the Hammerklavier lasting around 45 – 50 minutes,
whilst Mendelssohn’s Op. 106 only lasting about 18. Given the disparities, it seems
unlikely Mendelssohn projected a work of Hammerklavier proportions, instead
choosing to emulate smaller foreground details.
Furthermore, Franck himself emulated the Hammerklavier. The first section of his
organ work Final (from his Six Pieces) is similar in gesture, both in terms of melodic
shape, key of the first theme, and the keys used in the work as a whole. Franck’s
composition is in one movement, but does include a central slow section, adhering
to his preference for a tripartite layout.
Figure 6.6: Franck, Final (Op. 21), bars 28 – 32
Figure 6.7: Franck, Final (Op. 21)

B-flat major
Allegro Maestoso
Section A
1 – 42 B-flat I
42 – 80 G minor vi / ivα
81 – 94 D Iα
95 – 122 G (B, E-flat, G) IVα (IVβ, IV)
Section B (Key change: F-sharp major)
123 – 134 C-sharp (V/F-sharp) Vβ
135 – 146 F-sharp Iβ
147 – 154 A-sharp #VII (enharmonically I)
166
(Key change: No key signature – A minor/C major but this is not the key until bar 202)
170 – 181 flux
182 – 194 E IIα or V/Vα
202 – 214 flux, A minor, G
215 – 234 B minor IVβ
235 – 244 Am – D-sharp dim – F-sharp – D7
Section C (Key change at bar 407: B-flat major)
245 – 294 B-flat (F) I (V)
295 – 309 D Iα
317 – 352 B-flat I
353 – 369 E-flat IV
370 – 387 B-flat I
Final was one of Franck’s first large scale organ works. It seems to meander
somewhat, however, and has not enjoyed the same popularity in the organ
repertoire as, say, the Trois Chorals. If the mono-axial model is to be successful, as
in Beethoven’s Hammerklavier, we must consider other factors: shape and length of
movements, type of development, mixture, and shaping and placement of
immanent and/or transcendent dominants. Whether Franck realises this more
successfully later in his career is the subject of the remainder of this thesis.
Franck’s Piano Quintet in F minor
It is appropriate to begin the formal analysis of Franck’s absolute works with the
Quintet, as it is one of his first mature works; this strain of chromaticism is not
present in his earlier compositions. Furthermore, this piece has afforded a great
deal of analytical attention by musicologists who have otherwise written little or
nothing else on Franck. One reason for this may be the prevalence of what Cohn
refers to as “tonal poles” in the work (triads which share the same axis but no
167
common tones, such as C minor and E major). Cohn argues for special
consideration of “tonal poles” because:

First, contrary motion is involved in the voice-leading between juxtaposed polar
triads in closed position. (This is true of no other types of closed-position triadic
juxtapositions where common tones are absent.) The contrary motion allows the
second special feature of the polar relation to emerge: each triad contains the
other’s two most piquant tendency tones, the raised seventh and the flattened sixth
degree (or some enharmonic versions(s) thereof).422
According to Cohn, these special qualities mean that when “tonal poles” are
juxtaposed they have the ability to “annihilate tonality” due to the uncertainty
created by the presence of multiple voice-leading functions. Though the voice-
leading between tonal poles is logical, Cohn describes the effect of tonal pole
juxtaposition as “uncanny.”423 However, this is but one element to consider here.
The following table outlines the important keys of the movement.
Figure 6.8: Franck, Piano Quintet, I

F minor
Molto moderato quasi lento
1 – 19 F minor i Introductory double-dotted scalic
figure
19 – 25 A-flat minor – E-flat vβ – V/vβ
Maestoso
26 – 33 E-flat, then to A minor V/vβ – iα E-flat – C major – C minor – A
minor
34 – 37 D-flat – A minor Iβ – iα ‘Tonal pole’ oscillation
38 – 39 Unison on E-flat
40 – 44 Chain of diminished 7ths ending on D-flat which forms an E dim. 7 (vii7/i)
Allegro
45 – 49 F minor / D-flat area i & VI
422 Richard Cohn, "Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of Late-
Romantic Triadic Progressions," Music Analysis 15, no. 1 (1996): 20 - 21.
423 Ibid., 21.
168
50 – 82 F minor I First subject
82 – 89 D-flat major Iβ
90 – 103 C-sharp major Iβ – Vα – V/V – IV – Iβ
Second subject, ‘tonal pole’ oscillation
104 – 111 V/D-flat – V/ E-flat V/Iα – V/Vβ Sequence
112 – 123 E-flat V/Vβ Transition
124 – 149 A-flat & A-flat minor Vβ & vβ Second subject, sounds like
closing theme
150 – 158 C-sharp minor ivα Development
159 – 162 E-flat V/Vβ Development of 1st subject
163 – 184 Key unclear: many augmented/diminished sonorities
185 – 192 Around F minor – altered V/E i – V/vα Based on answer of first subject
193 – 206 E minor – E major vα – Vα Introductory motif (false
recapitulation)
207 – 215 C-sharp major V/ivβ Second subject bar 203
216 – 225 F-sharp minor – toward A ivβ – Iα Introductory motif
226 – 241 A major - D-flat Iα – Iβ Second subject, tonic axis
242 – 250 D-flat and E-flat Iβ – V/Vβ WT approach to tonic
259 – 270 Sequence, fermatas, approach to tonic; first subject emerging. Diminished
sonorities
271 – 300 F minor i Recapitulation: first subject
301 – 309 V/G-flat major V/IVβ
309 – 311 G-flat major IVβ From second subject
Molto moderato quasi lento / Allegro
312 – 319 D-flat major IVβ Piano/strings alternating, as
earlier in movement
320 – 322 D-flat – German sixth of v
Allegro
323 – 330 C major V
331 – 357 F minor - major i–I Second subject
358 – 365 F minor – A major i – Iα
366 – 383 Chromatic preparation for i
384 – 399 Complete I axis I – Iα – Iβ Using second subject; C+ = V/A
Animato
169
400 – 411 F minor i Fragment of second subject;
chromaticism
Più Presto
412 – 419 F minor i Uses C7♭5 as dominant
420 – 426 Chromatic sequence: French sixths used
427 – 440 F minor i First subject; end ‘dissolves’
The opening movement gives an impression of the music contracting and
retracting, with a chromatically moving bass-line. Emotionally, it feels intense, in
the way of Franck’s Quartet (discussed in the next chapter). The opening slow
section can be considered a complete miniature ternary form in itself, with the
lyrical piano theme of bars 6 – 13 recurring in the dominant axis key of A-flat; this
is a similar opening gambit to the D minor Symphony with its minor third
opposition. The cadence into this lyrical theme is the same in F minor and A-flat (at
bar 19), as well as later in the Quintet (bars 40 – 41 of the second movement). The
brief touch on G-flat at bar 12 foreshadows the emphasis on the subdominant axis,
which substantiates later. Despite a chromatically rising bass-line from bars 14 –
16, Franck overshoots C (V) in bar 17. Oddly though, a perfect cadence into E-flat
(V/Vβ) occurs in bar 22, the strongest we have experienced so far, which perseveres
into the Maestoso. Unusually though, the “active” E-flat does not bring any new
thematic material, instead using the double dotted theme of the opening. Then we
move through C major, C minor and A minor to get to the tonic axis, all whilst
growing in dynamic intensity. The introduction reaches its climax at bar 34, where
D-flat and A minor chords alternate; this is significant, as they are “tonal poles” (in
Cohn’s words) with no common tones, but tonally neutral to one another and the
tonic.
Let us consider these “tonal poles”: how we interpret them will have significant
implications for our interpretation of this movement overall. We know that D-flat
170
and A minor are tonally neutral, and Cohn suggests this relationship “annihilates”
tonality, in that D-flat defines A minor by two subdominant voice-leadings and
one dominant voice-leading, whilst A minor defines D-flat by two dominant and
one subdominant voice-leadings. These sound more or less equivalent (Cohn’s
point), but this also means that phrase structure becomes very important: here, it is
D-flat that occurs on the downbeat of the phrase, and hence has the stronger claim
for tonicity. In technical terms, we might describe D-flat as thetic and A minor as
anacrusic, in accordance with David Greene’s theory on temporality in Mahler’s
music.424 However, if we remove phrase structure from the equation, it is only the
hierarchy between two chords (here, D-flat and A minor) that is attenuated – not
tonality itself. As mentioned in Chapter Two, the tonal pole represents the non plus
ultra of ambivalence within diatonic axial relationships but does not annihilate
tonality outright. If we consider the notes here as a pitch-class set comprising g-
sharp – a – c – c-sharp – e – f, then the tonal poles play a special role of articulating
all the pitches of the given set. Franck’s use of this pc set continues throughout the
Quintet, and he uses it to bring about unity from a micro to a macro level.
The rough harmonic outline of the opening Moderato and Maestoso indicates a tonic
– dominant – tonic layout. The double-dotted theme provides the main melodic
material, together with the move to the axial dominant, which uses this material as
well. However, even though all the tonic axis members are present here, Franck
does not define the move away from and back to the tonic thematically.
A chain of diminished sevenths leads us to the Allegro, which begins around D-
flat, sounding as VI in F minor. Franck’s technique here is similar to the Lisztian
technique of using wholetone related keys in expository sections, which, as here,
are circular and do not achieve anything, finding their way back to the tonic. After
David B. Greene, Mahler, consciousness and temporality (New York: Gordon and Breach,
424
1984), 48.
171
a statement of the first subject – a descending scale figure beginning on the sixth
scale degree – there is a modulation rising through wholetones from bars 56 – 63.
We sense this modulation strongly because of the arpeggio-based thematic
material; again, it is Liszt-like in nature. At bar 64 a sequence begins which takes us
back to F minor, with falling semitones in the violin melody; analyst Jorgen Jersild
pays special attention to these.425 Then, Franck alters the A-flat minor chord to an
augmented one by semitonal downward movement of the root-tone, which effects
a change of axis through A-flat minor (Vβ), C-flat augmented (enharmonically
iv/ivβ), G minor (iv/ivα), B-flat augmented (IVaug), F-sharp minor (ivβ), F
augmented (Iaug) to F major (I.) This effectively emasculates each chord as it
becomes the subdominant of the following chord. We might describe this as
“extreme subdominant voice-leading”, a composing out of the D-flat – A minor
relationship. However, the harmony III+ – V7 – i also occurs twice between bars 71
– 73, which stabilises the key. The second subject begins at bar 90: a passionate
pianissimo theme in the piano that modulates through a diminished seventh of
minor thirds: C-sharp major, E major, G major and B-flat major before returning to
C-sharp. Each harmonic area also touches on its hexatonic pole, such as A minor in
C-sharp major. Cohn describes this passage as showing:

[There is] a strict pattern of pitch-class transposition through chromatic space.
Furthermore . . . the transposed material consists of a major triad alternating with
its (minor) hexatonic pole. Each alternation engages all six pcs of the hexatonic
system, so that the systems engaged serially by each passage are defined with
maximum efficiency.426
Cohn’s analysis is correct, but we must also consider the effect of the modulations
that give the second subject its distinctive sound. Minor thirds have a dominant
function due to the tones belonging to different wholetone scales, and this section
425 Jörgen Jersild, The Harmonics of Romanticism: Functional Principles Based on the Harmonic
Style of César Franck trans. Liz Lammers (London: J. & W. Chester Ltd., 1970), 22.

172
seems remarkably animated in a tonal sense; yet, it also possesses an attendant lack
of tonal control.427 The minor thirds seem to drive the music forward, but without a
clear sense of directionality, so although the systems may be “defined with
maximum efficiency”, the passage seems directionless nonetheless.
A transition using a long E-flat pedal point begins as bar 115, before a version of
the second subject appears in A-flat major and minor (Vβ and vβ). Again, the
harmony features a juxtaposition of tonal poles: between A-flat major and E minor
(spelled as F-flat minor) at bars 126 and 134. What seems particularly odd about
this is that A-flat – a significant key in a formal sense as it is part of the dominant
axis – coincides with a “closing” type theme (see Figure 6.9). This effectively brings
the exposition to a halt, and is problematic if Franck wishes to articulate tonality in
a sonata form context. Rosen explains these sonata processes: “The first section, or
exposition, has two events, a movement or modulation to the dominant, and a final
cadence on the dominant. Each of these events is characterised by an increase
[emphasis mine] in rhythmic animation.” 428 Franck’s A-flat theme is surely far
away from what we might describe as “rhythmic animation”, particularly given
that we are close to the development section. A sudden hush at bar 144 begins a
piano solo, using fragments of the first subject and modulating to D-flat. The
development then begins at bar 150 with a reference to the area of D minor
followed by an arrival into E-flat major at bar 159, developing the first subject.
Then, transitions and sequential periods begin. The interesting aspect to note,
however, is that almost every sequential period moves in either semitones or
minor thirds: a similar effect to that of Le chasseur maudit discussed in the previous
chapter.
Figure 6.9: Franck, Piano Quintet, I, bars 135 – 144
427 The Quintet shares this with Le chasseur maudit, discussed in the previous chapter.
173
At bar 193, however, we encounter a strong false recapitulation placed early in the
development – indeed, a good 77 bars before the recapitulation proper. One reason
the false recapitulation sounds convincing may be because there is so much
chromaticism surrounding it. By way of Classical precedent, Mozart illustrates
how a degree of wit is required for a false recapitulation to succeed: his Symphony
No. 38 in D has a false recapitulation on the dominant at bar 189, which only lasts
two bars before it turns to the minor (the recapitulation proper, however, is less
faithful to the original harmony of the main theme, this being the “joke”); and
Symphony No. 41 Jupiter, which recapitulates not the main theme but its
restatement from bar 24, before dissolving after a mere six bars. These instances are
short, witty devices; Franck’s false recapitulation seems prolix in comparison.
174
A restatement of the second subject after this is significant as it reverses the order
of keys we heard the second subject in earlier. Initially, the order was C-sharp
major – E major: this time it is E major – C-sharp major. The next significant key
area is F-sharp minor, emphasising the subdominant axis. Franck places further
emphasis on the second subject between bars 226 – 241, where it is stated in A
major and D-flat major – both members of the tonic axis – and transitioned
between by way of a German sixth. After the period in D-flat, the melody moves
up a minor third to begin the melodic recapitulation in F minor, though the
underlying harmony suggests D-flat until bar 246. We reach F minor at bar 250.
Following this, a regular recapitulation of both the first and second subjects occurs,
albeit with some chromatic alteration, and harmonically we are in the realm of the
tonic and subdominant axes. This demonstrates the mono-axial emphasis in this
work, even in the transition to the recapitulation. Consequently, however, this
thwarts any sense of tonic re-attainment and resolution. Rosen explains:

When the tension between tonic and dominant has been so weakened with no
substitute offered, resolution loses its meaning. What the ‘recapitulation’ resolves
is not the harmonic tensions of the opening, but the tensions set up by all the
different tonalities in the course of the piece.429
Though Rosen is writing about the Classical period, it seems reasonable to apply
this criterion to a later Romantic work, provided we consider Romantic
compositional concerns. Hereby, the keys focused on in the recapitulation do not
seem to fulfill this process: instead it is the pc set and its related keys. Yet, although
the recapitulation and coda are both based around F minor/major and the pc set,
the foreground chromatic material sounds as though it belongs in the
development, not the end of the movement. Furthermore, the music requires
sustained absence from the tonic for the recapitulation to sound like a re-
attainment. The final bars are particularly problematic: although Franck cadences
with a strong V♭9 – i6 in F minor between bars 432 – 433, and he reiterates the
429 Ibid., 92.

175
cadence using a German sixth on the flat second scale degree. With the
combination of the octave F in the bass, this progression sounds like a German
sixth in B-flat that does not resolve. Consequently, neither the F minor nor the final
four bars sound convincing; yet, this is not the impression given by the entire form.
One might make an argument to consider the dramatic minor tensions of the piece
and the possibility of non-resolution being desirable. In the chamber works of
Brahms, for example, the music basks in the attainment of the tonic minor, even if
it gives the impression of darkness or tragedy: in his F minor Piano Quintet, for
example, closure in the minor mode is not only attained, but emphasised and even
grimly celebrated. In this movement of Franck’s, I suggest that the pc set succeeds
in unifying the “sound”: if we infer sonata form processes, however, this
movement does not fulfill them fully.
At this point it is worth examining another analyst’s views. In the article
“Parsimony and Extravagance”, Cook examines passages of Franck’s Quintet
which feature tonal poles and the surrounding harmonic events. The two instances
he discusses are from bars 26 – 37 (the chromatic sequence followed by a D-flat – A
minor juxtaposition) – which he describes as “parsimonious” aside from the D-
flat/A minor “extravagance” – and the eight bar sequence between bars 124 – 131
(beginning in C-sharp major and ascending in minor thirds, including each new
tonic’s tonal pole) – which he describes as “extravagant”.430 In the second instance,
despite the possibility of smoothing the voice-leading between the modulations of
a minor third using their common tone, Franck emphasises the distance by
obscuring the modulation with a seventh. This intrigues Cook: he writes, “One
might reasonably expect Franck at least to make use of this [common tone] if not
emphasise it. Instead, the mediating seventh chords obscure the common tone.
430 I define these terms in Chapter Two.

176
One notices instead the extravagance of the voice-leading.”431 I suggest that Franck
may have crafted the passage Cook describes as predominantly parsimonious as
such because of its earliness in the movement. The tonal plan is still emerging, and
it is indeed a touch on the key of the dominant (C minor in bar 29), but the
parsimonious voice-leading smoothes the modulation and it is hardly dramatised.
The extravagant passage, in contrast, is harmonically and thematically dramatic:
the tonal poles are juxtaposed (e.g. 91 – 92: C-sharp major – A minor) and the
music suddenly modulates by minor third (e.g. 93 – 94 C-sharp7 – E major) moving
axes and wholetone scales. Modulation by a minor third is by its nature “active”
and creates harmonic drama, but at this juncture it is more appropriate: we are in
the exposition and this is the second subject.
Cook’s interpretations of these effects differ slightly to my analysis. He states that
the first (parsimonious) passage “stresses a diatonic relation – relative major and
minor – and a chromatic relationship easily integrated into diatonic interpretation
– parallel major and minor.”432 This passage is relatively straightforward and his
analysis seems reasonable. However, he also argues that the second sequence
“stresses chromatic relationships; the hexatonic poles are difficult to interpret
tonally, as are the seventh-chord mediations between sequential iterations.”433 Yet
if we interpret these relationships in light of an axial-tonal system, their functions
become much easier to understand diatonically, despite their undeniably rich
chromatic context. Harmonic poles represent the greatest distance within a single
axis in which there are no common tones between chords, yet because of their
proximity and voice-leading qualities such chords remain closely related and do
not annihilate tonality. The minor-third sequence represents a movement through
alternating wholetone scales; a technique Franck uses a great deal in the works of
431 Robert Cameron Cook, "Parsimony and Extravagance," Journal of Music Theory 49, no. 1
(2008): 113.
432 Ibid., 116.
433 Ibid.
177
the 1870s. This gives the impression of harmonic action without stressing the
dominant. The only problem, which one also finds in Le chasseur maudit, is that the
music can begin to feel repetitive, particularly if this chromatic/tonal contrast does
not coincide with thematic contrast.
Let us consider what Franck’s structural intentions may have been. In some of the
harmonic tactics of this movement, there seem to be tautological or problematic
elements: the tonic arrives before the recapitulation proper; entire structural
sections (such as the opening) repeat themselves almost verbatim in un-neutral
keys; and the tonic is undermined so severely at the end of the movement that
progressive tonality seems almost viable. Though it is quite possible that Franck
did not intentionally use Beethovenian models, what we do find in such exemplars
– particularly in Beethoven’s Op. 127 – is a turning away from the “heroic” type of
sonata form. In its place is an emphasis on something less dramatic, perhaps
comparable to the finale of Mahler’s Sixth Symphony in which the music is made
to “fail”,434 or the first movement of Beethoven’s Ninth Symphony, discussed in
Chapter Five, in which cadences at crucial junctures are undermined by way of
weak inversions. Therefore, we might describe this movement of Franck’s Quintet
as a de-powered sonata form, or perhaps a non-dramatic one. Of Beethoven’s Op.
127, Kerman writes: “a process of continuous free variation seems to supplant
traditional developmental energy in [Op. 127’s first] movement, in interest at least,
even perhaps in function”435 and later:

This movement is a burgeoning, not a dramatic statement. . . . Sensibility, not
structure, is the heart of this piece. Obviously form as such is not the major
expressive element. Yet the art required to mold unobtrusive form that will
434 Seth Monahan, "Inescapable" Coherence and the Failure of the Novel-Symphony in the
Finale of Mahler's Sixth," 19th-Century Music 31, no. 1 (2007): 63.
178
support the repetitive leisure of such a piece is very considerable, as many
nineteenth-century composers (or their listeners) learned to their sorrow.436
It seems viable that in the moving away from “heroic” sonata form to a more
lyrical structure, Franck’s efforts became somewhat confused. The result seems like
a half de-powered form: some moments are beautifully lyrical, others repetitive
and ill conceived. Though on the one hand the mono-axiality of the Hammerklavier
is absorbed, both Franck and Brahms – in his Op. 1 – struggle to break from
Beethoven’s heroic style, which the mono-axial model was (arguably) a reaction
against. It is also possible that Franck considered the idea of progressive tonality,
given the extent to which he undermines the tonic and emphasises other key
centers (often the subdominant, as in this movement). Franck’s employment of
progressive tonality in Psyché, discussed in Chapter Five, is further evidence that
he may have come close to using it in a more conventional work such as the
Quintet. It is also possible that Franck is making some sorts of preparation for the
second movement, which we will now consider.
Figure 6.10: Franck, Piano Quintet, II

A minor (iα overall)
Lento, con molto sentimento
1–4 A minor i
5–8 F minor iβ Melody inverted. Axial shift.
9 – 12 F minor – A minor iβ – i
13 – 19 A minor i Each part moving semitonally
II
20 – 35 Harmonically, as 1 – 16. Fuller figuration in all parts.
36 – 37 A major I Similar to 17, but in major
38 – 40 Transition: F-sharp major (IVα) – E-flat (V/Vβ) then French sixth on A-flat
III (Key change: D-flat )
41 – 46 D-flat Iα
436 Ibid., 209 - 10.

179
47 – 57 Chromatic transition and emotional climax
58 – 65 D-flat Iα Second subject from first movement. 59:
hexatonic pole (i) 66 – 68. Sequence in
whole-tones: V/D-flat; V/E-flat; V/F
IV
69 – 72 On C9 On Vβ9 C pedal in the bass; alternating with B
diminished 9 as a sort of dominant; C-flats
in melody; original violin theme but
interval flattened.
73 – 76 On D9 On IV9
77 – 83 In the area of the dominant; re-transition
V (Key change, A minor)
84 – 87 F – A major-minor Iβ – I – i Melody as opening: piano
88 – 91 F minor iβ Textural variation
92 – 94 E9 – F minor – i (V9 – ♭vi) i As bars 9 – 11
95 – 99 A minor i As 12 – 15
100 Unison A rises to unison B-flat in piano bass
101 – 102 B-flat IVβ Theme linked to Variation III
103 – 104 F11 Iβ11
105 – 106 Alternation of B ½ diminished and F7
107 F7 chord Iβ7 Acts as V of A minor like German sixth
108 – 109 A minor i
This movement is a set of variations. The violin’s first utterance emphasises the
subdominant as it falls through D – B – A, and there is a ivβ – i movement in bar
two.
Figure 6.11: Franck, Piano Quintet, II, bars 9 - 12
180
A feature of this movement, which makes itself known early on, is minor-major
alteration, such as bar three, where there is a sudden shift from A minor to A
major, which reverts in bar four.
This contributes to the feeling of semitonal contraction and retraction that was so
overt in the first movement, and indeed is a subset of the pc set. Passages such as
bars 17 – 19, in which every voice in the piano part moves semitonally, reinforce
this sensation. Early in this movement, Franck makes his awareness of axial
foreground function known. There are instances where the chord of E functions as
the dominant of both A minor and F minor (bars 9 – 12, illustrated as Figure 6.11).
This indicates an awareness of shared dominant function between the axes at a
foreground level, and if this is the case on the surface, it is unlikely to have escaped
Franck’s attention at a deeper level.
The second variation is harmonically similar to the first, until bar 36 when there is
a shift to A major then F-sharp major: more subdominant emphasis. Also worth
noting is the prevalence of augmented sixths, used both “regularly” (in terms of
resolution) and as a type of enhanced dominant, such as bars 45 – 46 in which a
German sixth in A-flat alternates with an A-flat major chord.
Also apparent in this movement is the emphasis on the C-sharp – A minor pc set
that was so prominent in the first movement (particularly with its emphasis on
tonal poles). Axial-tonal procedures are still at play to be sure, but at the same
time, there is a sense that this pc set – comprising all the tones of the tonic axis – is
where Franck intends us to focus aurally. For example, he emphasises members of
the pc set at cadence points, such as the D-flat on the tenth quaver beat of bar six,
and the F on the seventh quaver beat of bar eleven.
181
A deeper manifestation of this pc set is the modulation from A minor to D-flat at
bar 41. This is the “composing out” of a micro event from the first movement at a
macro level. As the chords are V♭5♭9 (second inversion) – I in D-flat, apart from
the C-natural to D-flat movement, all the voices at this cadence use subdominant
or “falling” voice leading. This weakens the cadence, as Franck intends: using a pc
set as a unification device means that the movement can be organised in a tonally
non-combative way. The manifestation of the pc set continues into the third
movement.
Figure 6.12: Franck, Piano Quintet, III

Interpreted with F as tonic
Allegro non troppo, ma con fuoco
1 – 16 Suggests V of D minor V/IVβ
17 – 32 Suggests V of G V/V/V
33 – 48 Suggests G V/V
49 – 52 Suggests G V/V Piano, first subject motifs
53 – 56 Suggests F-sharp minor ivβ (ii)
57 – 72 Suggests F major I
73 – 80 F major/minor I/i First subject
81 – 88 Toward D minor ivα
89 – 96 F-sharp minor - A minor ivβ/iα
97 – 112 F major/minor I First subject, intensifying
113 – 118 Dominant axis sweep Vα, Vβ, V
119 – 134 F – A-flat I – Vβ
135 – 146 C-flat/E-flat7 (acting as V) and enhanced G diminished chord
Key change to B minor
147 – 174 B major IIα Second subject: G minor 149, E-flat 153
175 – 179 B minor, then dissolves iiα Second inversion V – I in B; G minor
180 – 188 Suggests G minor ii B-flats interpreted as SD3 of G minor, #7 of
189 – 198 B minor iiα Transition
(Key change: F major)
182
199 – 210 C major/minor v First subject
211 – 218 A Iα From first subject
219 – 230 Tonic axis sweep x2 I axis From first subject
231 – 246 F-sharp minor ♭II Interlude; F-sharp minor/D
247 – 254 Chromatic sequence
255 – 258 B-flat minor iv Rhythmic emphasis on beats 1 and 2
259 – 266 G minor V/V or ii
267 – 270 G major V/V
271 – 274 Around E minor vα
275 – 278 A minor iα
279 – 296 Real sequence to F major (but on V, I, 291 – 296) uses cyclic theme from second
movement
297 – 301 Transition: C in bass but harmony contracting and retracting around it, no clear
key.
301 – 305 C+ chord (enhanced dominant, all notes of V wholetone scale)
306 – 313 F major i First subject
314 – 321 F – V of D minor i – V/ivα Moving up a major third
322 – 325 F-sharp minor ivβ Plagal-type cadence
326 – 329 A minor iα As above – moves up a minor third to sit
on C; becomes V7 of F
330 – 345 F minor i First subject in minor
346 – 351 Partial-axial sweep through E7, G-sharp7 and C-sharp7
352 – 403 F-sharp minor ivβ Second subject; subdominant
404 – 409 Transition, through F minor, E minor, to E flat minor
410 – 417 E-flat minor v/Vβ Five bar phrase
417 – 421 D minor ivα As 410 – 414, but D minor
422 – 427 Chromatic transition: oblique arrival
428 – 443 D-flat major Iβ Cyclic theme from movements 1 and 2
444 – 481 F minor – major i Cyclic theme
482 – 491 D-flat major – transition Iβ D-flat, first inversion
502 – 509 (Four bars which repeat) Through A Ger6, F7, A maj, C7
510 – 516 D-flat on 510 downbeat, then F on downbeat of 512 – this repeats itself; D-flat/B-flat
sounds like strong contender for tonality
516 – 520 Through descending F scale with♭2 and ♭6 (no 7th)
183
521 – 528 Hammers Fs in octaves – sounds like V!
The key is uncertain from the opening of the finale, which although interpreted in
F minor strongly suggests other keys, such as A minor, and also uses F-sharp
major in bar 14, the tonal pole of D minor. The first 32 bars emphasise notes within
the hexatonic pc set, though notes outside it are also present: A is on the downbeat,
and C and C-sharp also occur. The dotted theme, which begins in the piano at bar
13, foreshadows the first subject, which will not arrive for some time yet. However,
this theme also includes a harmonic reminiscence of previous tonal events in the
form of a V7♭5 in second inversion in bar 16. This is the same chord used to move
from A minor to D-flat in the second movement. However, the rustling violin
figuration continues whilst the key is still undetermined.
G major (wholetone related) becomes a significant player between bars 33 and 48:
it is by no means clear-cut, as there is little definition of key, but D7 sonorities
begin to sound like V7. Between bars 49 – 52 and 53 – 56 the suggested key area
begins to drop by semitone, beginning around G in the first four bar instance then
F-sharp in the second. It is only in bar 57 that the harmony begins to move toward
F major: the piano had suggested an outline of the first subject at bar 49, but it
takes until bar 73 for the first subject proper to arrive in the strings. Prior to its
arrival, F had been tainted with D-flats and E-flats, which made it sound like a
dominant (and indeed, this is one interpretation of the end of the work: that F
sounds like the dominant of B-flat). The D-flats are sometimes part of the German
sixth that Franck is so fond of, which can reinforce F as the tonic if they resolve
traditionally. However, the E-flats have the opposite effect, and the German sixths
do not always resolve to V6/4 . This is one of the features of Franck’s foreground
harmony that undermines his foreground tonicity. Sometimes the German sixths
appear and traditionally, as ♭VI – V, but the addition of #6 to chords – such as in
184
the Violin Sonata – can have an emasculating effect. If #6 is added to a tonic chord
(as occurs the Violin Sonata, discussed in Chapter Nine) and then resolves in the
“proper” way, the music will end up in the key of the major mediant. Hence, when
Franck adds #6 to an F major chord at bar 69 in this movement, the effect is
unusual and undermines rather than confirms the tonicity, even though the
German sixth is left unresolved.
Consistent use of the pc set, however, can achieve tonal unity, and this seems to be
one of Franck’s intentions in this work. The first subject, which has taken 73 bars to
arrive, uses pitches exclusively from the row (aside from a singular B-flat in bar 86,
which acts as an upper neighbour to A). Bars 79 – 80 even use one of the
augmented chords from the row to tonicise F, the key of the movement overall or,
at least, the strongest contender for tonicity at this juncture.
Once it has arrived, the register of the first subject rises, and development ensues.
At bar 113 we get a sweep of F’s dominant axis. This coincides with a definite
change in texture: instead of the jagged, register-shifting piano chords we have had
until now, this passage utilises flowing arpeggios. In this case, the shift signals a
move to a new key and thematic area – tonally around the supertonic axis and
thematically introducing fragments of the second subject. In Chapter Two I
discussed the importance of considering how chords interpreted in an axial way
function, rather than rudimentarily assigning axial names to them unreflectively.
The title “supertonic axis” is often a misnomer: more frequently, composers
achieve musical stasis by using areas that we might describe as supertonic axis
members in a wholetone related manner, rather than preparation for a dominant
axis harmony. Here, however, the supertonic axis does prepare the dominant axis,
a rare case. It also encompasses the key of the second subject. This transition begins
a bar 127, where we are beginning to enter the realm of the supertonic axis,
beginning with V/C-flat and then C-flat itself. At bar 147, the music changes key to
185
B major and the piano plays the second subject, in B major (iiα) though the
harmony also references G minor and E-flat major in bars 149 and 153 respectively;
touching on each key in the supertonic axis at a foreground level. It is also relevant
to note that the melody of the second subject encompasses a transposed version of
the tonic pc set, this new version being [2, 3, 6, 7, 10, 11.] At bar 163 the second
subject turns to B minor. These keys, at first glance, are not problematic: they
contrast with the tonic and incite contrast within their own foreground activity by
using their own axial relatives. These relationships, such as B minor – G minor and
B minor – E-flat major, are similar to the tonal poles in the second subject of the
first movement, which contrasts C-sharp major and A minor. It is also worth
noting that the relationship between F minor and B minor, a tritone, is the same
one that Franck uses in his Symphony, where D and A-flat also share a special
relationship. Though we cannot be sure exactly what Franck’s opinion of such axial
relationships was, we can be sure that he was aware of their existence and used
them for these sorts of expressive effects. Furthermore, his persistent use of them
throughout his late period, across genres, suggests an interest and preoccupation
that supersedes mere curiosity.
The music remains around B minor and its associated motivic area until bar 199,
when there is another key change to D minor/F major, and the development
begins. Here we have a large mixture of keys, as we might expect. There are some
particularly dense harmonic moments, several of which are worthy of analytical
attention. From bars 207 – 227, the foreground chords are all rooted in the pc set,
moving through A-flat, C, E, A, E7, A, C-sharp, E, F, A and C-sharp and so forth. It
seems that Franck is experimenting with the potential of this pc set, in terms of
both its melodic scope and its harmonic relationships, the second of which is the
case here. This chordal movement accompanies development of the first subject,
which is moving sequentially upward in register. Another significant moment is
the modulation to F-sharp minor, as how Franck attains F-sharp minor further
186
belies his awareness of the possibility of axial relatives to tonicise at a foreground
level: an F7 chord acts as Vα/F-sharp minor allowing the music to move simply
from one to the next between bars 230 – 231. This is similar to the foreground
activity in bar nine of the second movement.
The second subject reappears at bar 279. This time Franck presents the theme in
real sequence, transposed up a tone each time from its beginnings in C-sharp major
(Iβ), through E-flat major (iiβ, or a wholetone relative) to F major (the tonic, but
here used in second inversion). The reader may well realise that two of these keys
strongly utilise the pc set tones, contrasting with the earlier instance of the second
subject which used the transposed version of the set. A dominant arrives in the
guise of an augmented chord; significant as this encompasses all tones of the
dominant axis in a single triad. The pc set hence has the ability to affirm its own
tonicity as this augmented triad exists within it. F major arrives at bar 306, but
modulation continues; from here, the significant emphasis is on F-sharp minor, as
in the first movement.
Perhaps the most significant formal moment of the recapitulation is the
reappearance of the cyclic theme, first heard at bar 90 of the first movement. At bar
429 this theme reappears in D-flat (Iβ), one of Franck’s many instances of cyclicism
in a multi-movement work. This theme is stated again at bar 444, in F minor, this
time combining with the first subject of the finale and emphasising A-flat (Vβ).
Here, we come to an interesting juncture. Though Franck alters the time signature
from 4/4 in the first movement to 3/4 in the third, and the second time we hear the
cyclic theme it is combined with the finale first subject, I would argue that the
cyclic theme has not essentially changed in nature. It is not a reminiscence of a past
musical-tonal world – it is a reminder of the musical-tonal world we are still very
much present in. Perhaps this relates to the unified sonic world of the pc set, but
regardless, the inference remains. Arguably, thematic development was not what
187
Franck intended, and the return is only meant to be regarded as a simple thematic
reminiscence, not a grand thematic transformation. However, I suggest that when
a composer is deepening layers of complexity within a composition – as Franck
undeniably is in a tonal (read chromatic) and formal sense – other parameters such
as thematic and rhythmic relationships (which I will discuss later in the thesis)
must be revised to fit any new conditions. On the surface this statement may seem
to suggest that if one element of a composition is complex, then the others must be
also – yet, I am sure any adept musician or analyst can think of a number of pieces
that are complex in some ways, but not in others. Instead, I suggest that if new
complex compositional conditions are present, a composer must consider the
impact of these conditions on all musical elements: if the relationship of such
elements is an uncomfortable one, then revisions are essential if the work is to be
successful. I explore this idea further in later chapters: the point here is that form,
tonality, theme and rhythm are not necessarily “correlative” in Franck’s Quintet, to
say the least.
From bar 470, the music remains in the tonic axis, largely in F minor and D-flat – or
at least it seems to from the score. There is a great deal of chromatic alteration,
particularly using German sixths and diminished and half-diminished chords. The
aural effect of this, however, is an emasculation of F minor – F begins to sound like
a dominant, with B-flat sounding as a potential tonic when it arrives on the
downbeat of bar 520 – even D-flat begins to rival F for tonicity at bar 510. This,
again, is the problem with the continuous use of the German sixth – our ears want
to interpret what Franck spells as ♭II – I (such as bars 520 – 521) as ♭VI – V. The
inclusion of E-flats in the final page also harmonically problematises this. The
combination of this with the B-flat minor and D-flat minor chords on the strong
beats of the phrases means that any sense of F (minor or major) being the tonic is
essentially destroyed: a B-flat minor chord following the unison Fs at the end of
188
this movement would certainly not sound amiss.437 That said, it is possible that
Franck only ended the work in F for the sake of convention. We cannot know if he
really desired to experiment with progressive tonality, only surmise that such a
progressive ending might have worked here.
There are two main conclusions that can be drawn from this analysis, which we
may or may not consider mutually exclusive:
1) Franck was experimenting with (some sort of) progressive tonality, but for
whatever reason did not see it fit to actually compose a fully progressive tonal
work. Certainly, the sense of establishing one key and problematising it is at work
here. Only the final step – actually ending the work in a different key to that which
began it – is missing here.
2) Franck had problems integrating large-scale form with his dense chromatic
harmonic language, and hence some of his mature works seem loosely organised
as a result. Arguably, Franck’s use of the pc set sonically organises this work.
However, the pc set does not always contribute to the functionality of the work as
we might expect.
Considering pc sets for a moment, we might compare Franck’s work with a work
using a similar organisational principle: Liszt’s La Lugubre Gondola No. 1.
Composed only three years after Franck’s Quintet, it shares with the Quintet’s
finale the idea of using tones outside the pc set to evoke functionality. (Notes
outside the pc set indicated in brackets.)
Figure 6.13: Liszt, La Lugubre Gondola No. 1, bars 12 – 15
It might also be argued that this work could finish in D-flat given the emphasis that key
437
had been given just bars before. If so this could be linked to a gesture in Beethoven’s Ninth
Symphony: the closing of the exposition in B-flat (the sub-mediant) ‘shot through’ with the
tonic. Here, however, that gesture is reversed.
189
Like the Quintet as a whole, the piece imparts a singular sonority: it and the others
in the set “sound the same” – the sound-scape enabled by the pc world is specific,
but flat – perfect for Liszt’s evocations of the lugubrious gondola. However, Liszt’s
work is only a short one; and therefore arguably more suited to employing a pc set
than a work such as the Quintet.
Whether Franck was experimenting with progressive tonality or having integrative
issues with large-scale form we cannot determine from one piece alone. The next
work we will consider is the String Quartet. Though it does not follow the Quintet
chronologically, the two works share many features: hence, the close proximity of
their analyses desirable.
Chapter Seven
String Quartet
190
After Beethoven a generation of composers arose who approached the String
Quartet genre with “awe and superstition”.438 Few clues survive regarding stylistic
influences on many of Franck’s compositions – that is, those that Franck made
direct reference to – but the String Quartet is an exception: d’Indy recounts
Franck’s workbench as being “littered with the scores of Bach, Schubert and
Beethoven”.439
The following table illustrates the keys and main themes of the first movement.
Figure 7.1: Franck, String Quartet I

D Major
Poco Lento
1 – 26 D I ‘Lied’ theme: ‘A’ [15: Reh. A]
27 – 36 D minor i [35: Reh. B]
37 – 40 ‘On’ A7d V Reinforced by F7 (Ger6/A)
41 – 50 D I ‘A’
51 – 53 D I B-flat (Iβ) prolongs D, acts as Ger6/D
[51: Reh. C]
54 – 61 D – D minor I–i
62 – 71 Transition: largely an expansion of B-flat (Iβ)
72 – 80 D I
Allegro
81 – 112 D minor i ‘B’ [96: Reh. D]
113 – 120 E-flat V/A-flat ‘C’ [115: Reh. E]
121 – 126 A-flat minor WT 1 axis ‘C’
127 – 130 A-flat major/minor WT 1 axis
131 – 137 B major ivα [131: Reh. F]
135 – 137 D major I
138 – 172 F Vβ From 127 – 138: ‘active’ minor thirds
[146: Reh. G; 163: Reh. H]
438 Laurence Davies, César Franck and His Circle (Boston: Houghton Mifflin Co., 1970), 233.
439 Ibid.
191
Poco lento
173 – 200 F minor vβ ‘A1’ [190: Reh. I]
201 – 212 B-flat minor iβ
213 – 217 V/G minor (around F-sharp dim) [213: Reh. K]
Allegro
218 – 226 G minor iv
227 – 236 B-flat minor iβ [233: Reh. L]
237 – 241 F minor vβ
242 – 251 C – C minor V – v/Vβ [246: Reh. M]
252 – 258 E-flat – E-flat minor ♭II – ♭ii The same material as 248 – 251
256 – 258 F-sharp minor Iα
259 – 264 A major V Second subject
265 – 270 Transition – around V/D, but also prolonging ♭VI7(#6) [265: Reh. N]
271 – 285 D minor i Recapitulation – as 81 – 99
286 – 296 D minor-major i–I [292: Reh. O]
297 – 304 F-sharp minor Iα ‘C’
305 – 308 G major IV ‘C’ [307: Reh. P]
309 – 314 Transition – around V of B using diminished chords
315 – 318 B major IVα Same as 138
319 – 326 D major I As above
327 – 330 G major IV Parallel to bar 130
331 – 339 Transition to ‘Lied’:
Poco Lento
340 – 353 D I ‘Lied’ [346: Reh. R]
354 – 359 D/d I/i
360 – 374 D I ‘Tristan’ cadence at end
The form of the movement is novel; d’Indy describes it as a “sonata form inscribed
within a Lied.” 440 In some ways, this is similar to the slow movement-scherzo
amalgam in Franck’s Symphony. Most arresting, though, is the movement’s intense
chromaticism; chords with added sevenths and ninths occur frequently, as does the
440 Ibid.
192
German sixth chord. Such added notes have the potential to make chords sound
like dominants, which at times undermine the tonicity of the movement. Within
the Lied section the ambiguity becomes a structural impetus, the crux being
whether at a given moment the D9 chord is I or V/IV. This combined with the thick
chromaticism and retracting semitones, makes the movement seem emotionally
charged and harmonically rich. Add to this equation unequal phrase lengths –
usually three bars – and the music feels unsettled, in need of development and
resolution. Perhaps this opening is less an emotional slow introduction than a
motivic-harmonic roadmap, pointing the way toward the following sections.
It seems clear the Franck wanted to strike some kind of balance between
problematising tonicity and “annihilating” it altogether. Cook discusses this in
relation to Franck’s use of the Neapolitan chord, which arrives very early in the
movement – yet earlier drafts of the Lied theme show that he experimented with
the Neapolitan’s arrival as early as bar four. 441 He notes that “usage of the
Neapolitan is so common in the nineteenth century that it ceases to be ‘chromatic’
(read, by many, ‘complicated’) in anything but technical terms.”442 Cook goes on to
state that the eventual placement of the E-flat in the seventh bar is a middle-way,
which is strong –his opinion, however, is that it does not undermine D’s tonicity.
Yet Cook does not discuss here the sixths and sevenths, which are added to the D
chords that affect (and potentially undermine) the tonicity so strongly. Obviously
we must also consider their effect, and I discuss this factor below.
The first Lied section is in mini-sonata form. Franck paints his harmonic strokes
broadly, and this first section stays fairly firmly around D, despite the various
problematisations discussed above. Excursions to the dominant and subdominant
axes occur, and Franck places a great deal of emphasis on the Neapolitan; hence,
Works of César Franck" (PhD Diss., University of Chicago, 2001), 121.
442 Ibid., 120.
193
this section might be considered a large scale ♭II – V – I. Any written analysis,
however, does not accurately describe the effect of this section: it is emotionally
intense and deeply chromatic. Chords seem to merge into one another, submerging
rather than actively progressing, though this is in part due to the slow tempo. A
liberal use of seventh and ninth chords is partially responsible for this submerging,
as is Franck’s well-used German sixth. A great deal of the part movement is by
semitone, which gives the effect (particularly at the opening) of chords contracting
and retracting, not moving far from home. When sevenths invade the texture so
liberally, both V and I have potential to sound like the dominant, and I7, on
occasion, can sound like V of G as early as bar three. Second inversion chords are
used to similar effect, as the 6/4 chord is ingrained as a dominant approach chord
and is difficult to aurally interpret otherwise. I+7+9 is “problematised” – is it I or
V/G? ♭VI – II (and ♭II) is often not allowed to resolve as we might expect (followed
by V – I). Instead, Franck uses it expressively, and the tonic often simply “arrives”.
Further to this, unequal phrase lengths add to the intensity. Each time the Lied
theme recurs its phrasing is in three bar groupings, making it feel slightly
unfinished or unsettled; conversely, the rest of the section is largely quadratic. As if
to fulfill these mini-sonata form requirements, the second theme which we first
heard in A major/minor is retonicised at Reh. C+6.
From the Allegro section – which begins with an unsettled theme in D minor – there
is an emphasis on A-flat (a tritone away from D), which seems odd here. However,
A-flat shares the tonic’s wholetone scale, and hence the overall impression is one of
flatness.443 At Reh. F+7 we modulate to F (Vβ) and the second subject arrives. This
does not seem unusual and fits with sonata procedures we are familiar with, yet
Dahlhaus observes that we only recognise retrospectively which of the musical
443 This is similar to the coda of the Symphony: WT 1 axis emphasis.

194
thoughts constitutes the second theme, and considers the implications that this has,
formally:
This is another indication of the formal problem that a string quartet movement
had to solve, given the historical situation around 1880: how to integrate ‘late-
romantic’ harmony, the yardstick of musical progress, without abandoning the key
schemes of sonata-allegro form, to all appearances a defining feature of the genre.444
Dahlhaus also notes that the music only modulates to F major (the opposing key,
Vβ) at the end of the exposition, and thematically consists of a “closing group”. 445
Yet:
by melodic criteria, on the other hand, the actual second theme seems to be an idea
that modulates from D minor through E-flat major to A-flat minor. . . thus, however
much Franck wished to preserve the simple key scheme dictated by the classicist
theory of sonata form, he was no less unwilling than Liszt (the source of his
approach to form) to dispense with the effect of ‘wandering tonality,’ even when
presenting his themes.446
Though Dahlhaus touches on the problematic nature of integration between form
and harmony in the String Quartet, there is little evidence to suggest that Franck
wished to “preserve [a] simple key scheme”: why would Franck desire this, given
that chromaticism holds much harmonic potential? The link between Liszt and
Franck that Dahlhaus suggests, however, is a likely one: Franck is indebted to Liszt
in his use of fugue for a development section, and this further links to the
chromatic idiom: “it can be shown that the fugato, as in Liszt’s . . . Prometheus,
offers a way out of the difficulty of how to write a development section when
developmental techniques (chromatic modulation and sequential patterns) already
predominate in the exposition.”447 Beethoven’s Op. 106 or 127 may also be likely
influences here, as they additionally share with Franck’s Quartet the designation as
445 Ibid., 293.
446 Ibid.
447 Ibid., 292.
195
absolute music. I now examine other reasoning behind – and the effect of – this
fugato.
The Poco Lento fugato arrives unannounced at Reh. H+10, and lasts for 44 bars.
Though Beethoven [and perhaps others] sometimes inserted changes of tempo and
meter into a quartet movement in sonata form (see for example, Beethoven’s Op.
127) these are usually only a few bars long. Both the extended slow introduction
and the central Poco Lento suggest a reinterpretation of the slow sections’
proportions within a first movement. (It is worth noting that this use of fugue may
be another link to late Beethovenian form, in that Beethoven employs fugue in both
the Hammerklavier and Op. 131; I will discuss this further shortly.)
Of the Poco Lento itself, it is fugal and uses the Lied theme. It is in F minor (vβ) and
the fugal answer in its dominant minor, C minor. There is then a modulation to B-
flat minor, followed by its relative D-flat at Reh. I+11. Importantly, though, the Poco
Lento disrupts the flow of the piece considerably. The re-ensuing Allegro in G minor
(iv) sees a false recapitulation at Reh. K+10; the true recapitulation does not
actually arrive until Reh. N+6 (see Figure 7.2).
This is preceded by a short period in the “true” dominant (A major), yet the actual
arrival uses very weak voice-leading. Although the false recapitulation is less
tautological than that of the Quintet’s first movement, the arrival of the
recapitulation proper is not the decisive recapitulation we might desire in such a
sonata form movement.
196
Figure 7.2: Franck, String Quartet, I, reh. N+4 – 7
It seems that Franck undermines the tonic with a weak arrival here so that he can
enunciate it later in the Quartet. However, whether this is actually the effect is
another question, and will be considered below, as there is then an emphasis on
tonic and subdominant axis keys until the return of the Lied theme at Reh. P+33,
which ends the movement.
This movement brings up two questions:
1: Despite the Allegro sections’ adherence to the usual processes of sonata form, to
what extent can we consider the movement as such, given the inclusion of the Lied
and its effect? Considering d’Indy’s analysis of the Quartet, Cook states the
following:
The deeper issue would be one of unity, of integrity, questions which were raised
in criticism of Franck’s oevre, and to which d’Indy was sensitive. On one hand, if
d’Indy asserts formal unity for the movement, he faces the task of explaining how
the Lied material engages the principles of sonata form: the theme carries a great
deal of weight as the piece’s cyclic motto, and yet it appears not to participate in the
tonal and thematic processes proper to a sonata. Thus the theme risks violating the
principles of order among elements of a composition that d’Indy cites from Ruskin.
197
On the other hand, his chosen course of formal separation between Lied and sonata
elements skews the one possible option for asserting unity in the movement,
namely a symmetrical interleaving of parts, because his theory of form requires a
ternary structure for both designs.448
Despite the problems this causes with d’Indy’s analytical approach, I surmise
Franck was aiming for something like the “symmetrical interleaving of parts”,
despite my opinion that this disrupts the flow of the movement overall. Though
this process arguably blurs sonata functions, it is undeniable that the Allegro
begins in F minor and arrives back in F minor at Reh. L+4. This is a significant
prolongation of the dominant axis key before A arrives at Reh. M+13. Although the
foreground at the point of recapitulation is ♭VI7#6, this dominant prolongation is
still present: whether we sense its strength is the eventual crux of whether the
movement fulfills sonata processes, and that question may not be answerable until
we have considered the other movements of the Quartet.
2. We have seen in the Quintet – and again see here – that Franck has a number of
chromatic devices that he is fond of using. These include liberal use of added
sixths, sevenths and ninths to chords; German sixths that act as dominants rather
than resolve in one of their regular ways (see Figure 7.3); and harmonic regions
which he likes to emphasise, such as the Neapolitan in this movement. To a greater
or lesser extent, all these gestures have the ability to affect the central tonicity. This
is not a problem if a composer manages these techniques, balancing them with
more tonal counterparts, or if such gestures contribute to a form that works with
such a language (I am thinking specifically of Wagner here). It is not the
chromaticism itself that is the problem – many of Franck’s more successful
contemporaries used a deeply chromatic musical language – it is how and where
the chromaticism is managed that affects our overall impression. My impression is
Cook, "Transformational Approaches to Romantic Harmony and the Late Works of César
448
Franck", 115.
198
that this movement does not flow, and I suggest that although the form is novel, it
might be that it is the gesture rather than the form that is confused; stylistically,
aspects of this movement are disjointed, but it is still coherent enough to articulate
its tonal plan and harmonic sonata processes.
Figure 7.3: Franck, String Quartet, I, reh. C+20 – 21; a German sixth acting as a
dominant
(Though he does not discuss the German sixth – or the augmented sixth – acting
directly in this regard, Schoenberg touches on the possibility of cadences to omit
chords in “abbreviation of set patterns through omission of intermediate steps.” 449)
Relating this movement to late Beethoven models, I suggest that this movement
contains gestural elements of two – Op. 127 and the Hammerlavier: the first because
it encapsulates tonic and subdominant emphasis, and the second because of the use
of fugue. (In this it may also relate to the first movement of Op. 131.) Yet both of
these elements are gestural rather than being part of a larger formal plan, as is the
case in the Beethovenian forms. The emphasis on the tonic and subdominant axes
in Beethoven’s Op. 127 was to avoid the dominant so that it might be posited in a
Arnold Schoenberg, Theory of Harmony, trans. Roy E. Carter (Berkeley, Los Angeles:
449

199
later movement: in Franck’s Quartet, there is a section in the dominant axis; in
Beethoven’s Hammerklavier the fugue was employed for tonal flatness: here Franck
places it in a dominant axis key. Whether Franck employs these features for their
Beethovenian legacy or not is beside the point; regardless of their heritage,
tonic/subdominant prolongation and fugue have defined effects, and Franck knew
this. The form may be novel, but it is also problematic: an active background
(axially speaking) is difficult (perhaps even impossible) to reconcile with flat
foreground gestural features. Because of this, the first movement of the Quartet
may have to be included in the list of “stillborn” works emulating late
Beethovenian processes, however novel it may also be.
Figure 7.4: Franck, String Quartet, II

Vivace
1 – 77 F-sharp minor i ‘A’ scherzo theme
[30: Reh. A; 55: Reh. B]
78 – 86 F-sharp – C minor [78: Reh. C]
87 – 94 C minor – F-sharp minor
95 – 110 As above – instrumentation altered [110: Reh. D]
111 – 139 Around F-sharp minor
140 – 151 F-sharp minor
152 – 170 Around V of D major V/Iβ Rocking idea
171 – 194 D major Iβ Rocking idea [171: Reh. E]
195 – 219 D major
220 – 233 D Iβ Lied theme in cello (225)
234 – 250 D Iβ Repeat of 152 – 159
251 – 270 Transition
271 – 387 Recapitulation i [311: Reh. F]
388 – 415 F-sharp major I
This Scherzo lacks regular phrasing, making it feel wild and unsettled. The
harmony sounds manic, but is an exemplar of the potential axial modulation gives
200
composers; though it feels like we are constantly modulating, we end in the same
place we began. Bars 15 and 18 touch on D minor and B flat, but we return to F-
sharp, the tonic, shortly after; a Schubertian chromatic prolongation of the tonic.
Franck achieves a similar effect by tritone modulation, such as the modulation to C
minor at Reh. C+8: because tritones split the octave exactly in half, when the
process repeats the music ends where it began. Because of this, the movement gives
the impression of being tonally flat. From Reh. D, Franck plunges us into a frenzied
chromatic section. Meter is indistinguishable because of syncopation, and the
music modulates to B-flat (Iα) in first inversion. B-flat is the flattened sixth tone of
D – the next modulation – and this moment may serve to prepare us for that
modulation: yet, the music has still not gone anywhere. Two-imperfect cadences
then lead us back into F-sharp minor, before the music settles on F-sharp at Reh. E-
20, signaling a change of mood.
From E-19 to E-1, Franck establishes V/D before we reach D at Reh. E. A rocking
type theme begins, relaxed and flowing, which then develops into triple time.
There are brief touches on B minor and D minor, but the music mostly remains in
D major. Unlike the first movement of his Symphony, here Franck utilises the
sonata principle: tonal contrast highlighted by thematic contrast. However, because
the tonal contrast is within the same axis (and not as contrasting as a more remote
modulation would be) the thematic contrast must be significant enough to make
this impression. In order for this to be effective, thematic contrasts in this
movement are acute.
At Reh. E+34 the music shifts suddenly to F-sharp minor, leading us back to D
major with the Lied theme in the cello at Reh. E+54. B-flats appear in the texture, but
when they turn to A-sharps in Reh. E+66 it signals the return to F-sharp minor, and
the recapitulation is a carbon copy of the opening save for some pizzicato
arpeggios. At Reh. F+73 the music turns to F-sharp major, recasting the rocking
201
theme in the new key. The movement runs out inconclusively, with Violin 1 sitting
on scale degree three.
Incomplete in itself, this movement requires context for it to function the way it
seems Franck intended. As outlined, he neglects the dominant axis at a background
level, giving an overall impression of flatness despite the delirious energy of the
themes. This continues the feeling of flatness we have sensed in the first
movement, meaning the drama of the immanent and/or transcendent dominants
remains unfulfilled halfway through the piece.
Figure 7.5: Franck, String Quartet, III

Considered in B major (IVα)
Larghetto
1 – 33 Tonal fluidity; touches on G-sharp minor, F-sharp minor, C-sharp minor and B
(albeit briefly). ivα, v, ii (v/v), I [24: Reh. A; 33: Reh. B]
34 – 44 F-sharp minor v
45 – 52 A minor Wholetone relative tonicised by its own ♭VI#6
53 – 65 F-sharp minor v [63: Reh. C]
65 – 67 B minor i
67 – 96 B major I Opening of this movement recurs
[79: Reh. D]
97 – 102 Transition via diminished chord [101: Reh. E]
103 – 106 C major IVβ New theme – ‘S’
107 – 114 A-flat IVα
115 – 118 D-flat – C7 – E – E-flat [115: Reh. F]
119 – 126 V/A-flat V/IVα
127 – 130 Sequence to A via B-flat minor (here acting as Neapolitan) – A-flat minor – F-sharp
minor – E (V/A, WT relative and I of piece overall) – note the falling WT sequence
[127: Reh. G]
131 – 138 A WT 1 ‘S’
139 – 142 Around F – weakly WT 1 ‘S’ [139: Reh. H]
143 – 150 Sequence 146 – 149, rising in minor thirds: 148 V of B, then V of G-sharp
202
(Key change: B major)
151 – 160 Similarly fluid, as the opening [159: Reh. I]
161 – 164 Transition to D: WTs Vβ
165 - 172 Around V/B V ‘T’ as we have heard around V/A-
flat
173 – 181 B I Fragments of first theme
182 – 187 B I Theme from bar 34. This time it
appears in G-sharp minor,
allowing it to resolve into the
‘right’ key (the tonic).
The first suggestion of key in this movement is G-sharp minor. However, G-sharp
resolves on beat two, weakly: furthermore, the G natural in the same bar sounds
like ♭6 in B. There is a stronger cadence on to I6 (B major) in bar four, followed by a
cadence onto the dominant in bar eight before a confirming V – I. Though there
exists potential to set up a rivalry between G-sharp and B (as tonal rivalry is a
favourite technique of Franck’s) this is not the movement’s nature: instead it begins
with a sense of tonal fluidity between B and its relative minor. There is then a
cadence into C-sharp minor at bars 11 – 12. A new theme with iv – i harmony in C-
sharp minor appears, repeated sequentially in B minor at bar 17. Following this,
Franck alludes to D major and F-sharp minor from bars 21 – 23, B occurring with
certainty for the first time in the movement at Reh. A+2 – 4. This first theme is
reiterated, and there follows a period like a second subject in F-sharp minor.
From our analysis of the Quintet and the earlier movement of this Quartet, we
know that Franck has a number of techniques he draws on for expressive
emphasis: of those, what seems to predominate here is the unattainable nature of
the subdominant, both in terms of B major (being IVα of the Quartet as a whole)
and the subdominant axis of B major (E, G-sharp and C in this movement). Though
there is an extended period in C major (IVβ), in which Franck introduces a new
203
theme and emphasises B’s other subdominant axis keys, the music soon transitions
into A, a wholetone related key to B and V of the Quartet as a whole. Franck also
emphasises the interrupted cadence at Reh. I+21.450 The prolongation of G-sharp
minor that follows (via the iv-I progression) takes on a tragic demeanor, reinforcing
the true subdominant’s unattainable nature.
Furthermore, B major is only attained by a tierce de Picardie in the final bar – the
only previous substantiated instances of B major being II – V – I cadences into Reh.
A+5 and Reh. D_14 - 16. The rarity of the seeming “tonic” within the piece
reinforces the fragile and unsustainable quality of the subdominant, and it would
be difficult indeed to find a movement in the standard repertoire where the
nominal tonic is present for such a short period. Perhaps this has a programmatic
purpose.
Another interpretation might be that the movement makes sense as a sonata
without development (up to a point) because of the exposition’s repeat and the
second theme’s lack of retonicisation in B minor. The intrusion of the Lied material
delays this action and the second theme only arrives at bar 181 in the wrong key,
G-sharp minor, before righting itself to B-minor/B. This upward shift of a minor
third is in fact a repeat of the F-sharp minor – A minor material from Reh. F+5 (this
type of tonal handling is redolent of Schubert).451 If we consider the movement a
sonata form without development, the recapitulation conflates the events of Reh.
B+8 – 12 tonally with what happens at Reh. C-7 – 4. As such it sounds very short-
winded, as though it has tacked on a concluding gesture to one that implied, from
the precedent of Reh. B+15, continuation. A further instance of Beethovenian
450 This interrupted cadence is loaded with symbolism from Beethoven’s Lebewohl sonata, Op.
81a, and it seems unlikely that contemporary audiences would have failed to make this aural
connection.
451 This is similar to the way in which Schubert might have handled the material: Schubert is
known for such “lazy” modulations, evidenced in works like his Fourth Symphony. See fn. 19
of Chapter Eight for more detailed information.
204
influence is present in the final bar of this movement. The final bar’s melody – SD
2-3-1 – is an echo of the phrasal closing gesture in bar 30 and 39 in the Cavatina of
Beethoven’s Op. 130 B-flat major Quartet. As with the Lebewohl inference, this is
unlikely to have been lost on contemporary audiences.
Figure 7.6: Beethoven, String Quartet Op. 130 (Cavatina) bars 38 – 39
Figure 7.7: Franck, String Quartet, III, bars 186 – 187
Considering that the tonic of the work overall is D, bars 165 – 169 provide a
significant foretaste of this key. Though Franck emphasises F-sharp, D remains
significant, regardless of whether we feel it sounds like Vβ or III of B minor.
Figure 7.8: Franck, String Quartet, IV

D major
Allegro molto – and other tempi
1–6 F-sharp minor (also suggests F) iα (Vα) ‘Finale’ theme
7 – 11 Around D I Slow movement theme
12 – 24 F-sharp minor iα Finale theme
25 – 35 F-sharp minor iα Scherzo theme
36 – 52 Transition (as 12 – 24 ending on V/D)
53 – 58 D (with B-flat & C) I Lied theme
59 – 128 D I Lied theme, reworked (1S, b) [83: Reh. A]
129 –193 A minor v Scherzo theme etc (F/D hint) [129: Reh. B]
205
193 – 208 A major V Second subject [165: Reh. C]
209 – 223 C-sharp major Vα Second subject [209: Reh. D]
224 – 237 Transition V/A – ♭II/D – V/A
238 – 261 A minor v Using plagal voice-leading [254: Reh. E]
262 – 269 Moving down in WTs – sequence
270 – 280 A minor v Runs out onto G diminished
282 – 337 Key change – D-flat. Transition – D-flat, E-flat minor, E-flat, F minor (WT
movement)
338 – 359 Extended transition, chains of diminished chords [338: Reh. F]
360 – 375 C-sharp major - minor Vα – vα Second subject [367: Reh. G]
376 – 387 Transition using B-flat7, C-flat7, C7, B7 and C-flat7 again
388 – 392 E-flat major – minor ♭II – ♭ii Reworking of Lied
393 – 400 Transition through E – C-sharp – B-flat, a reversal of the progression which
occurred in
the first movement, through a tritone.
400 – 439 B-flat major – B-flat minor IVβ - IVβ Based on Lied [400: Reh. H]
440 – 451 Transition, mainly from diminished chords [440: Reh. I]
452 – 505 D major key signature. Transition through A (452); F-sharp major (456); E-flat
major
(460), then through this entire axis (G and B, comprising the subdominant axis)
[476: Reh. K]
470 – 476 ‘Finale’ theme with tremolo, until we arrive at an A-sharp in bar 476
477 – 496 On V/E-flat+9 V/IVβ
497 – 505 E minor – E-flat – D
506 – 513 D major I Lied theme
514 – 529 A-flat major – V/D-flat WT 1
530 – 546 ‘Finale’ theme in A-flat WT 1 [530: Reh. L]
547 – 559 D major I
560 – 587 D minor i Second subject ‘fifth’
588 – 615 D minor – G minor i – iv [588: Reh. M]
616 – 647 D major – F-sharp majorI – Iα Second subject
648 – 660 Transition: V/D minor V/i Transitional theme [648: Reh. N]
661 – 700 D minor i Second subject [685: Reh. O]
206
701 – 756 Transition to Coda – touching on many keys
705 – 712 F-sharp major Iα ‘E’ theme
713 – 730 F-sharp minor – F minor iα – Vβ – vii/Vα Scherzo theme in triplets [713: Reh. P]
731 – 756 A major – A minor V–v ‘E’ then Scherzo in triplets [749: Reh. Q]
757 – 766 A minor v Lied theme variant
767 – 784 D minor – D i–I Lied theme reworked [777: Reh. R]
785 – 808 B minor – D major ivα – I Slow movement theme [806: Reh. S]
809 – 829 D minor – F major i – Vβ Slow movement theme
830 – 847 Prep for D on V
848 – 859 Transition – sits on V/F-sharp initially; E-flat major7 approach to 860
860 – 881 D major (with E-flats, B-flats, C-sharps.) I
In this movement, Franck uses the technique Beethoven employed in his Ninth
Symphony: he recalls the themes from previous movements and then rejects them
in favor of a new theme (though the influence of thematic reminiscence in
Beethoven’s Fifth Symphony may also be important). Though cyclic thematicism
and thematic transformation (as discussed in Chapter One) can play immensely
significant roles in unifying a multi-movement work, the problem with
Beethoven’s technique is that it appears obvious and contrived: comparison to
works which use the technique (such as Beethoven’s Fifth and Ninth Symphonies)
is almost inevitable. Consideration of scale is also important: though Franck’s work
is large by Quartet standards, it cannot compare to Beethoven’s monumental
Symphonies and even in the case of the Ninth Beethoven quotes the most recently-
heard movement (the slow third movement) very briefly.
One oddity is that whist a new finale idea emerges from the rubble of the previous
themes; the Lied of the first movement actually comes back a second time and plays
a significant role in the final movement, as does the slow movement theme. So the
Beethovenian idea of “trying” ideas from each movement and rejecting them
before another emerges by necessity does not seem to be the case; instead it seems a
207
convenient way to begin the finale dramatically, despite the return of other themes
regardless.
However, once the movement proper begins at bar 59, the initial premise seems to
be one of Franck’s more straightforward and diatonic movements. The first subject
derives from the Lied theme. Reh. A+23 – 24 features one of Franck’s beloved
gestures – a German sixth acting as a dominant to the tonic, D. It seems that now,
with Franck using this so much, the dominant almost becomes unnecessary – we
supply it mentally, as the other components of the progression are present. At Reh.
B+8 we reach the dominant minor, and this constitutes one of the few occasions
Franck nominates to use the actual dominant rather than an axial or some other
substitute. Furthermore, A minor is actually firmly established by the length of its
presence and the use of its own dominant. Reh. D-16 brings the second subject,
occurring in both A major and C-sharp major (Vα) at lReh. D; again, we are in a
very diatonic framework of applied dominants and suchlike. In many ways, this is
more straightforward than other Franckian sonata form movements.
Yet despite this seemingly more straightforward nature, there are still instances
where Franck’s obsession with semitonal and Neapolitan relationships shines
through. One such instance is between Reh. E-19 – 27, where an E-flat chord,
approached by a B-flat half-diminished in first inversion, simply rises to an E7
chord to act as applied dominant to A minor. Even though such foreground events
are rarer in this movement than, say, the Quintet finale, they are still present.
A closing theme, loosely related to the second subject, signals the exposition’s end
in A minor. Here the development begins with a hushed key signature change to
D-flat major (Vα). This is the transitional material first associated with A minor,
and we are still in the dominant axis. Franck states the theme in E-flat before more
modulation occurs, and by Reh. F+8 we are back around A minor. This incites
208
further development, before the next significant tonal event: the modulation to B-
flat major (Iβ) at Reh. H. This is not necessarily odd in itself, other than in the sense
that the tonic axis is usually avoided in the development section, and B-flat gains
quite some stability through its prolonged presence. Next, not long before the
recapitulation, a very curious progression begins: at Reh. I+12 (which is also the
return to a D major key signature) the music moves through a minor third
progression downward, through A, F-sharp and E-flat, before reversing in the
other direction by major thirds, through E-flat, G and B – the subdominant axis.
Another transition then takes us to a B-flat7 chord, which is emphasised not only
dynamically but also in register between Reh. K and K+17. With a sudden hush,
however, B-flat turns to diminished seventh built on B, in first inversion. This in
turn shifts to a half-diminished seventh on C-sharp between Reh. K+21 – 23. Yet,
instead of using this sonority to resolve to D major, we move down another
semitone to a C half-diminished chord, which then slips to B half-diminished at
Reh. K+26, a B-flat7 at K+27, a D minor second inversion at K+28, followed by a
first inversion in K+29, before the first subject is back in D major – with no sense of
triumphal attainment – at Reh. K+30.
So, although the dominant axis establishes itself in this movement, there is no
extended period actually in it before the return to the tonic – and Franck returns to
the tonic itself in a most non-dramatic way. From here, the recapitulation continues
in a very regular way, with both transitional material and the second subject
recurring in the tonic, D. Further material, previously stated in the dominant,
recapitulates in D minor, and we reach a natural end of these ideas at Reh. P-13,
where the music begins to expire. Now, the themes from earlier in the Quartet are
reintroduced: the scherzo theme at Reh. P+2 around F-sharp minor, before sinking
to F minor (at Reh. P+10) in the ♭II – i gesture we are now so familiar with; the
music then sitting with a fermata on a D-sharp diminished seventh at Reh P+17.
209
This then resolves to A major in the next bar, before the theme from the scherzo
turns the music to A minor again. Throughout this time, there are multiple rest
bars and points at which only one instrument holds a note. This creates a sense of
unrest: the music’s direction becomes difficult to pinpoint.
However, a firmer section begins at Reh. Q, where the Scherzo theme becomes an
accompaniment figure to the first subject in the minor mode. Initially around A
minor, the 10-bar phrase repeats in D minor, before a sudden shift to somewhere
between D major and a D augmented chord between Reh. R and R+7, which is
followed by an anguished return of the slow movement theme at Reh. R+8. Though
around B minor, the key is not certain, and becomes even more distraught on its
repeat in D minor at Reh. S+2. This runs out onto an emphasised dominant seventh
chord of A (on most previous occasions departed by semitone – for example, Reh.
N+10, E-18), which, this time, shifts to a C-sharp major chord (V/Iα) instead. With
the Lied theme being developed in the upper parts, this subtly changes to a E-sharp
diminished chord at Reh. S+50; this simply “falls” to an E-minor half-diminished,
then E-flat7, which leads us back to D major with the Lied theme ending the piece
(with added G-sharps and B-flats, but the tonicity not in doubt).
I find this movement rather convincing up until the coda, which is problematic.
Here Franck has generally used axial tonality in its useful, regular way (stabilising
chromaticism in areas where we might expect it) but in the coda, where we would
expect some stability, he continues to modulate in a non-axial (tonally active) way;
he has effectively continued modulating without cause, which results in structural
failure. At P-13 the music seems to be drawing to a close, yet Franck tries to start it
up again, seemingly unsuccessfully. I suggest that the inadequacy one feels here is
linked to the structure of the Quartet as a whole, and the problem of the finale.
210
The success of the finales in late Beethoven (particularly the chamber works and
Op. 106) seems to lie in the way Beethoven leaves some vital task for the finale to
enact. 452 In Op. 127, 106 and 131, this critical task is the more vital sonata
tonicisation which only occurs in the final movement. Drama in the finale can be
brought about in other ways, however, if something in procedure or content is
new: this is why Lewis states that the finales of Beethoven’s Third (double
variation form) and Fifth Symphonies (cyclicism and progressive modality) and
Brahms’ Fourth (Passacaglia in which “duality is implicit”) are so successful; the
use of Fugue (Beethoven’s Große Fuge or the Hammerklavier finale) and the
emergence of melody which changes everything (Beethoven’s Ninth Symphony)
may arguably also subscribe to this theory. 453 However, it is no surprise that
Franck’s Quartet suffers from the problem of the finale: despite its formal novelty,
the first movement is a tonicising sonata form. Though it is necessary to retonicise
D after the key scheme of the central movements, using a regular sonata form for
the finale still sounds tautological: the first movement confirms D beyond doubt.
Franck’s finale lacks conditions of “implicit duality” or new procedure or content,
which is why from Reh. P-13 in the finale it seems so repetitive. The first
movement’s conventional tonal layout effectively works against the novelty of its
form; Franck has imitated Beethoven in gesture, but not emulated the background
that necessitates those same gestures. Though I surmise Franck recalls these themes
to promote thematic unity, the overall effect actually undermines the flow of the
coda to my ears. Perhaps it would have been possible to use these themes less
ostentatiously and more fluidly; had this been the case, perhaps then the coda
would not seem so problematic.
452It might be argued that Brahms’ Third Symphony and Piano Quintet do indeed leave this
vital task until the final movement.
453
Christopher Orlo Lewis, Tonal Coherence in Mahler's Ninth Symphony (Ann Arbor,
Michigan: UMI Research Press, 1984), 101.
211
The other issue, of course, is one of proportion. The first movement of Franck’s
Quartet is 373 bars long – lengthy considering the Lento tempo of much of the
material. To illustrate this, compare the timings of recordings by the Oritz Fine Arts
Quartet (a), and the Academia String Quartet (b) – both recordings available on
Naxos Music Library.
Figure 7.8: Comparison of recordings
Movement (a) (b)
I: Lento: Allegro 13.46 16.30
II: Vivace 5.27 5.00
III: Larghetto 10.19 13.10
IV: Allegro molto 14.02 13.10
This means that Franck will want to balance the proportions of the first movement
with the others, which he does; but this only becomes a problem in the finale,
where – by Reh. P-13 – the music seems to be repeating itself. As the first
movement’s length necessitates the length of the finale, this is similar to the
situation Kerman writes about regarding Beethoven’s String Quartet Op. 59 No. 1.
To re-quote Kerman, Beethoven struggled specifically with the problem of the
finale in some movements, including the finale of this quartet: “so exhaustive an
experience has been conveyed [in the first movement] and in so many facets, as to
make any further communication superfluous . . . the later movements are set
awkwardly for the total impression.”454 Arguably, this is what has happened in
Franck’s Quartet – the first movement’s size and tonal articulation cannot be
matched in and by the finale. Dahlhaus’ assessment of the String Quartet is
particularly apt here:
212
It was no easy matter to balance simplicity and lucidity of form . . . with the
harmonic complexities that even composers with conservative instincts felt
compelled by the zeitgeist to adopt . . . Just how difficult this could be is
illustrated by Franck’s D-major String Quartet (1889) with a charity characteristic
of major works of art, part of whose nature it is to make problems manifest rather
than concealing them in artifices, a sure sign of mediocrity.455
Though beautiful moments abound, the problems here are certainly manifest. The
same might be said of Franck’s Symphony in D minor, which I analyse next.

213
Chapter Eight
Symphony in D Minor
Much is written about the Symphony and its first performances, and it is worth
surveying these views prior to the analysis to deduce the criticisms leveled at the
music. Franck wanted the conductor Lamoureux to give the premiere, but,
disliking Franck’s music, Lamoureux suggested taking the Symphony to the Société
des Concerts du Conservatoire instead. It was accepted, but – if d’Indy is a reliable
source – the orchestra hated it, and it was only at the insistence of conductor Jules
Garcin that the premiere took place on February 17, 1889. Contemporary reaction
was mixed: “only four or five of us [Franck’s pupils] were able to slip in”, writes
Pierre de Bréville. “[We] were almost alone in applauding, and even that greatly
scandalized our neighbours.”456 Romain Rolland was present at the premiere, and
his diary recounts:

An organ style. A regular development, powerful, stiff. Phrases chopped up harshly,
shouted out by the brass. Dryness at times. Brusque passages without transitions,
from fff to ppp (as in the Béatitudes.) But there is grandeur, emotion, thoughts that
recall the Béatitudes. A personality. In the hall, three audiences: frenzied applause –
very few; more numerous – “Hush! Hush!” (They are normally rare at the
Conservatoire.) These calls came mostly from the front boxes. During the
performance, I saw some listeners pretending to stop up their ears. Finally, the mass
of the public, indifferent.457
The aspects Rolland has noted are relevant; some critics took exception to the
repetitive four-bar phrases, cyclic thematicism, and the thick instrumentation.
456 A. Peter Brown and Brian Hart, The Symphonic Repertoire Volume III Part B: The European
Symphony from ca. 1800 to ca. 1930: Great Britain, Russia, and France (Bloomington and
Indianapolis Indiana University Press, 2008), 605.
457 Ibid., 608.
214
Contemporary commentators also criticised the use of the Cor Anglais in the
middle movement, despite Berlioz having incorporated it in the Symphonie
Fantastique years earlier, as well as d’Indy and Haydn employing it in
Symphonies.458 Boutarel reviewed the Lamoureux performance and found Franck’s
work problematic: as Cook summarises, “the piece ponderously bludgeons the
listener with very little, and uninteresting material. We can imagine that the return
of the slow, unison opening in F minor bores Boutarel, seeming to him like a
repetitive sermon or catechism lesson. ‘Why not fresh music for the new key’? he
wonders.”459
Furthermore, Franck’s Symphony has attracted criticism for its tonal
discursiveness. Yet we must examine the work to see whether the tonal
discursiveness is oppositional – and therefore of structural importance – or merely
fanciful modulation. Axial analysis can help to reveal this. We have already seen
how Franck has used axial tonality in his earlier works, and the Symphony is no
exception; in the first movement, it seems that Franck establishes D minor as the
initial tonality; threatens it with F minor, the Beta axial dominant; then “rescues” D
minor. D’Indy states that Franck conceived of the Symphony this way: “a struggle
between the opposing ‘tonal poles’ of D and F, culminating in the victory of the
former; all the important harmonic areas relate to one of these ‘poles’”. 460 Whether
this technique is effective and how axial tonality might be involved depends
largely on the crafting of large scale form, which we will examine shortly.
Before I examine the Symphony’s tonal relationships, it is pertinent to mention
several features of the first movement. The first of these is the multiple fermatas
458 Ibid., 608n.

Works of César Franck" (PhD Diss., University of Chicago, 2001), 138.

215
and changes in tempo between Lento and Allegro. This aggravates many critics, and
is likely one of the problems Boutarel is alluding to in the above quote. Tovey
complained that Franck’s pace was (by this point, late in his life) “becoming
Wagnerian.”461 Yet pauses do draw attention to structural peculiarities, the most
important being the complete and literal repetition of the first part of the
exposition in F minor (vβ), which also sees the return to the opening slow tempo of
the movement. Norman Demuth argues that the success of such pauses in Franck’s
Symphony depends on how a conductor shapes the work, suggesting that:

[the Symphony’s] discursive moments are rarely free from interruption and
argument [and] lead very often to nothing. There is a lot of questioning and
answering . . . These points are emphasised by nearly every conductor who essays
the work. It is the happy hunting ground of the ‘tempo-rubato’ conductor who
cannot by any means be persuaded to play the music as the composer wrote it. He
exaggerates everything and the work is never allowed to get a move on.462
Such pauses are a feature of Franck’s late style, and some may interpret them as
the organist-composer changing manuals or shifting stops, but such labels veil the
problem: such pauses are rarely as effective in orchestral works as they might be
between the organ’s Swell and Great manuals in St. Sulpice. Fermatas in the Trois
Chorals give the music breath; in the Symphony, they seem to stifle the music’s
progress. We must not discount, however, the possibility that this was Franck’s
intention, and if so it is most likely intended to draw attention to specific features.
The opening motif – Mus es stein? – was used by Beethoven in the F Major String
Quartet Op. 135, Liszt in Les Préludes, and Wagner in Die Walküre.463 It is not known
whether Franck intended any extra-musical reference, but to conjure Beethoven,
Liszt and Wagner in one musical breath might be considered reverential given not
461 Donald Francis Tovey, Essays in Musical Analysis Volume II: Symphonies (London: Oxford
University Press, 1966), 62.
462 Norman Demuth, César Franck (London: Dennis Dobson Ltd., 1949), 84 - 86.
463 Demuth, César Franck, 80.
216
only that this is Franck’s only Symphony but also the importance placed on the
symphonic genre in the late nineteenth century. As Dahlhaus writes, the
Symphony developed new significance after Beethoven:

The task that composers faced in assimilating the Beethoven legacy had to do with
a will to large-scale form. Beethoven had transformed the symphony into a
monumental genre . . . thereafter, a symphony manifested compositional ambitions
of the highest order, the audience it addressed being no smaller than the whole of
humanity.464
It is extremely likely that this “anxiety of influence” had an effect on Franck and
his contemporaries. The following table shows the keys used in the first
movement.
Figure 8.1: Franck, Symphony in D minor, I

D minor
Lento
1 – 12 D minor/G minor i - iv
13 – 20 D minor/G minor i – iv [17: Reh. A]
21 – 24 F-sharp minor – B minor iα – ivα
25 – 28 Bass movement: through B-flat and C to D to V/D Wholetone movement
Allegro non troppo
29 – 38 D minor i
39 – 46 D minor/G minor i – iv
47 – 48 V/C – C (as V/F) F is Vβ [47: Reh. B]
Lento
49 – 76 F minor vβ [61: Reh. C]
Allegro non troppo
77 – 94 F minor vβ [91: Reh. D]
95 – 110 F major Vβ
111 – 116 D-flat major Vα [111: Reh. E]
117 – 120 E-flat major IVβ & WT to D-flat/F
121 – 124 Transition to F (as 117 – 118)
217
125 – 145 F Vβ
145 – 160 F – A – C-sharp Vα – V – Vβ – whole axis [145: Reh. F]
171 – 178 F Vβ [171: Reh. G]
The above box represents the only significant instances of V in this movement
179 – 186 D minor – major i–I WT 1
187 – 190 B major IVα WT 2
Development
191 – 205 A-flat minor WT WT 1 [195: Reh. H]
206 – 209 C-flat/B (spelled as both) ivα WT 2
210 – 212 B minor ivα
213 – 220 D I WT 1 [213: Reh. I]
221 – 226 Transition to A-flat, resolving like a German sixth
227 – 234 A-flat major WT WT 1
235 – 248 B-flat minor – A-flat iα WT 1 [239: Reh. K]
249 – 259 A-flat – F-sharp minor – E All WT 1
260 – 266 E – D – C minor All WT 1
267 – 274 To C minor in 272 v/Vβ [267: Reh. L]
275 – 280 To E minor in 279 v/V
281 – 293 G minor - B-flat7 – C-sharp7 iv – V/♭II – V/Iα
293 – 300 E V/V WT 1 [296: Reh. M]
301 – 306 A-flat – C Vα WT 1
307 – 322 Through C – E-flat – G-flat, C-sharp fully diminished at 320 then E diminished;
suggests ii – V [317: Reh. N]
323 – 330 Diminished – V/D
Lento
331 – 345 D minor – B minor i – ivα
345 – 348 WT transition: B minor – C-sharp minor – E-flat minor – then V/E-flat minor
Allegro
349 – 360 E-flat minor [349: Reh. O]
361 – 374 G minor, with D pedal iv [370: Reh. P]
375 – 380 E-flat (briefly) then to V/D ♭VI/iv – V/I

381 – 384 D I
218
385 – 400 D, then V/B-flat I – V/Iβ [389: Reh. Q]
401 – 406 B-flat major Iβ
407 – 418 Transition through B-flat – C WT transition; see also 473 – 484
[409: Reh. R]
419 – 434 D I
435 – 450 Each key in tonic axis, before returning to V/I [435: Reh. S]
451 – 464 Transition – back to D I
465 – 472 D, then moving away I
473 – 484 Through B-flat – C minor Iβ – WT [473: Reh. T]
485 – 512 D minor i [485: Reh. U; 501: Reh. V]
Lento
513 – 521 ♭VI – ii in G minor – D major I
The harmony of the opening phrase could be interpreted as either G minor or D
minor, creating a subdominant bias right from the beginning. This is significant
because – if Franck intends to create a rival tonicity before rescuing the “true” tonic
– he must establish tonal ambiguity from the outset. (Though F minor becomes the
true rival tonicity, G minor casts the initial shadow; hence, D minor is not
established unequivocally.) Franck also uses the German sixth a great deal, as we
have already seen: in many of his late works, he uses German sixths as remote
modulatory devices. Bar eight is such an instance: a dominant seventh on A
(implying V/D) is resolved as though it were a German sixth into a D-flat major
chord in first inversion. D-flat and C-flat (both of which follow in bar nine) belong
to the same wholetone scale as G. This may reinforce G’s claim of tonicity, or at
least underline the oppositional nature of its wholetone scale. Reh. A+9 – 11
illustrates a similar tactic, but this time within the wholetone scale of D through G-
flat – A-flat – B-flat (in first inversion then root position). Semitonal movement
breaks this wholetone movement to the dominant before the tempo change to
Allegro non troppo. As it is difficult to define a key in whole tones alone, it is little
wonder that the impression of this opening section is one of inaction.
219
Conversely, the rhythmic and harmonic action begins with an increase in tempo at
the Allegro non troppo. At Reh. A+17, Franck employs a new double dotted motif, an
“intervallic compression” of the main theme. This theme also makes use of the [0,
1, 4, 5, 8, 9] row of the Quintet, with the theme comprising 0-1-8-9, which might
build a case for a D/B-flat polarity; whether this eventuates remains to be seen. No
sooner has this music begun, however, when E-flats appear (Reh. B-8). Though
fleeting, this haven of E-flat – another subdominant axis member, harking back to
the introduction – undermines D’s claim to tonicity, and the C-flats and A-flats in
the texture reinforce plagal voice leading.465 The music reflects back on D before
quickly modulating to C major (V/F minor or Vβ) and the section runs out with a
molto rall.
At Reh. B+2 we are presented with another Lento section, repeating the opening
material in F minor (vβ). Critics debate the purpose and effect of this stark break
fiercely. Tovey, having compared Franck’s stop-start tempo tactics to Beethoven’s
String Quartet Op. 130, feels such a repeat is an anticlimax: “No wonder critics are
disappointed who expect anything like the athletic movement of Beethoven’s Op.
130. Franck’s drastic simplicity belongs to quite another view of the universe.” 466
Demuth,467 however, is more defensive, quoting Guy Ropartz’s assertion that “the
movement proper is in the nature of a ‘double’,” and later pronouncing the
movement as “very satisfying.” 468 Earlier criticism aside, this technique is
unarguably conspicuous and assumes a lack of sensibility on the audience’s part.
What could be less subtle than a complete repetition of the opening material in a
new key? Furthermore, this section presents potential tonal problems: the work’s
opening phrases indicated a subdominant bias, but Franck has set this section in
465 This plagal voice leading is similar to the flat-6 flat-7 major mode that is found in Brahms,
Tchaikovsky and Rachmaninov.
466 Tovey, Essays in Musical Analysis Volume II: Symphonies, 63.
467 Demuth, César Franck.
468 Ibid., 81.
220
the mediant minor, vβ. This is problematic because of the lack of sonata principle:
tonal contrast highlighted by thematic contrast. The movement only attains
thematic contrast at Reh. D+9, when the second subject coincides with a change of
mode to F major. The lack of such contrast before this point only makes the
opening seem tautological retrospectively.
Nevertheless, the second subject does fulfill traditional expectations. It is in F
major, an axially related key to the dominant, and has a lyrical nature. At Reh. E
there is a modulation to D-flat major (Vα), from which wholetone movement
ensues and we modulate through E-flat to reach F at Reh. E+14. At Reh. E+18 there
is a new melodic idea in F – with the nature of a closing theme – and includes a
reference to V/D-flat in Reh. E+23. This new idea is then repeated in each dominant
axis key (F, A and C-sharp) at letter F through F+15; a similar way to how Schubert
presents such material.469 The modulation presents an interesting paradox: though
this theme is four-square and shaped as a closing theme (as we might expect at this
point in a classical sonata form work), Franck continues to modulate within the
dominant axis. In other words, despite signaling the close of the exposition
melodically, Franck continues to “enrich the fundamental bass” harmonically by
maintaining axial fluctuation.470 Modulation within a first or second subject area is
not unusual, but the important aspect is where Franck is modulating in terms of the
overall sonata scheme. Where other composers might choose tonal stasis (such as
this point in the exposition of Liszt’s B minor Sonata) Franck chooses to forego
such stasis, despite the theme being a closing one in nature. If we compare this
with Schubert’s String Quartet in G major (D887), it is notable that although all
dominant axis members are used in the exposition, there is a definite stable period
469 For a discussion of how Schubert utilises the entire dominant axis in a development, see
Cohn, “As Wonderful As Star Clusters: Instruments For Gazing At Tonality In Schubert”, 225.
470 Carl Dahlhaus, Between Romanticism and Modernism: Four Studies in the Music of the Later
Nineteenth Century (Berkeley: University of California Press, 1980), 63.

221
in D (V) from bars 142 – 169.471 This is coupled with a rhythmically stable theme. In
contrast, one might surmise that Franck either has another scheme in mind, or does
not know what he is doing.
As the section closes, there is another harmonic surprise at Reh. G+8, with the
sudden reappearance of D. Considering the development section is imminent, D is
an unusual choice.472 Nevertheless, the closing-type motif in the first movement of
Franck’s Symphony recurs in D and B major before the development proper
begins, with the same motif occurring in A-flat minor at Reh. G+20. In the
development, members of the subdominant axis provide tonal contrast, as well as
those keys that share the tonic’s wholetone scale; Franck avoids the dominant axis.
Hence, the next key area is C-flat – enharmonically B and IVα – but the real
surprise is the return of D major at Reh. I. This emphasis on the tonic major seems
tautological and is quite literally unprecedented: why would Franck choose to
restate the tonic in the development section? Perhaps he considered it would
sound like a mediant given the prior B minor context, but would this not
undermine the impact of the tonic return at the recapitulation? The effect,
however, is not what one might expect; this period in D major sounds less like a
premature return and more like an exotic modulation. Franck has achieved this
effect through his double exposition of D and F minor. By casting doubt over the
“true” tonic, the return of D makes F minor seem like a tonal usurpation, aurally
comparable to IVα of F minor rather than I. In this we sense the late Beethoven
tactic of making the tonic vulnerable in order to rescue it, as discussed in the
“proto-progressive model” in Chapter Four. (Further discussion follows in the next
472 However, this is not without precedent – Brahms used a similar technique in his fourth
Symphony. In Brahms’ work, however, the return of the tonic implies a literal repeat of the
exposition, thwarted by the onset of the development section instead. Brahms’ fourth
Symphony pre-dates Franck’s in both performance and publication, and thus could have
influenced Franck’s Symphony.
222
Chapter.) The success of such a scheme will depend on whether the remainder of
the Symphony logically follows the path of this agent and how Franck “composes
it out”. The contrast between background and foreground key areas is also notable.
So far, the larger tonal scheme has traversed D, F and A-flat minors, whilst
foreground modulations have included D, B, and A-flat. These keys are mirror
images of one another. This emphasis on A-flat (IIα) is extended in the finale,
where the supertonic axis becomes a significant key area.
Sequential material follows this excursion to D major, emphasising F’s dominant
axis. At Reh. K-10 a D-flat7 resolves like a German sixth into A-flat major (Vβ in F).
A wholetone sequence then moves through A-flat and F-sharp minor, before
cadencing onto E at Reh. K+20 (again, possibly prefiguring the emphasis on the
supertonic axis in the finale). This sequence repeats, moving down from E to D and
cadencing in C at Reh. K+24. Following this, an extension of C minor occurs; a
major third sequence through A-flat, E and C, before a shift to E minor at Reh.
K+34. There is a brief return to G minor, moving to B-flat, before an arrival in E
major at Reh. M-7.
Though the keys of A-flat-C-E form an axis of their own, to describe them as the
“supertonic” axis here is to misconstrue their function (unlike in the String Quartet
– see Chapter Seven). Combined with the emphasis on F-sharp and D as key areas,
the vast majority of the development centers on the tonic wholetone scale, as
illustrated by Figure 8.1. With all this wholetone activity a key lacks definition,
making the interpretation of this section as development of D problematic. A more
accurate interpretation might be that this modulation is preparing a tonic return of
F minor rather than D.473 After all, A-flat C and E – predominant tonal areas within
It is tempting to draw a parallel here with Mahler’s first Symphony, as it shares with
473
Franck’s the opposition of F and D minors. However, though I argue that Franck is on the
cusp of progressive tonality (indeed, this is why I have chosen the term “proto-progressive
223
the development – constitute F’s dominant axis. Franck’s emphasising on keys
related to D’s wholetone scale points to deliberate planning on his part, though
whether this is able to balance with the rest of the Symphony is yet to emerge.
From Reh. M-7, the harmony implies even more axial relationships: E acts as a
springboard for the reiteration of F’s dominant axis. A thick tutti passage follows,
harmonically situated around V/D (though diminished in nature). This leads to a
fully diminished C-sharp dim7 at Reh. N+3, before D minor returns at Reh. N+14 in
the opening Lento tempo. Franck gives this return full tutti force, but by now we
are so well prepared for F minor that D minor is not entirely persuasive, and the C-
sharp dim7 chord at Reh. N+3 could have resolved to either D or F minor. This
chromaticism destroys any sense of clear tonal direction, aided by the ambivalence
of D as the true tonic and F’s role as tonal hijacker. There is a single glance at D
major and B minor (ivα) before reaching E-flat minor (ivβ). E-flat minor takes the
form of the opening Allegro, on the flat side, before a dynamic withdrawal of
forces and axial modulation to G minor in which Franck cunningly uses a
dominant pedal of D to reiterate the tonic whilst in the key of the subdominant.
The re-casting in E-flat minor is pivotal for the recapitulation to function
assertively. By placing the Allegro theme in a subdominant relative key (ivβ),
Franck forces the listener to hear it as a counterweight to the original key of F
minor: vβ effectively neutralised by ivβ.474
Reh. P+5 sees a brief flirtation with E-flat major before dominant preparation
begins and D reappears at Reh. P+11; however, B-flat (Iβ) is the chosen key for the
restatement of the second subject at Reh. Q+12. It is arguably here that the tonic is
tonality”, discussed previously and also more explicitly in Chapter Nine) the tonal opposition
in Franck’s Symphony is righted by sonata convention (the re-tonicisation of the second tonal
area) rather than the re-thinking of symphonic form (Progressive Tonality) we find in Mahler.
474 Schubert is known to use a similar scheme to modulate in some works, including his
Fourth Symphony, the first movement of which (in C) modulates through C to A-flat in the
exposition and then G to E-flat in the recapitulation (sequences notwithstanding).
224
“rescued”: tonicity becomes stable, with the second subject reassuringly
unaffected. There is some wholetone modulation in the tonic axis before a
significant arrival at Reh. T-12, where the music from the beginning of the
development section is restated, echoing the earlier register and orchestration, but
properly enunciated as the tonic. This arrival builds on the work of the second
subject in happily confirming D.
The music remains around D/B-flat for the remainder of the movement, but the
final Lento reemphasises the subdominant area by returning to G minor one final
time. The final cadence also uses a type of German sixth chord instead of V: a
romantic touch, but passive in a tonal sense. The infiltration of E-flats suggests a
dominant function for D – possibly a dominant minor ninth – before D is
“rescued” one final time. From this gesture we can hear that the Symphony’s
opening ruse of problematising D minor with its subdominant continues, even in
the movement’s final bars. Despite the dramatisation, though, the deeper
tonicisation processes mean that F sounds suitably rejected and D appropriately
affirmed.
Examining Cook’s analysis reinforces many of the ideas we have explored in this
movement. He explores this movement in detail, and the crux of his discussion
revolves around the role of the double exposition.475 He questions:

Does the F-minor repetition of the opening theme ask more of the material than it
can deliver, forestalling desired thematic development and hanging a numb,
useless limb on an otherwise vigorous torso? [For the critics], the piece makes an
incoherent thematic argument, rendering the formal structure disjointed. In
contrast, [Franck’s apologists] take the logic of thematic relationships for granted:
the demands of form justify the repetition of the first theme.
475
Franck", 134.
225
Here Cook’s analysis aligns with my identification regarding the lack of sonata
principle. Without delving too deeply into his techniques, it is possible to state that
Cook shares with our analysis a discomfort in form here, both agreeing (in some
ways) with the original critics, whilst seeking to understand the deeper procedures
at play in this movement. Predominantly, it is the severing of thematic and tonal
relationships – previously so intrinsically bound – that Cook identifies as
perturbing for the Symphony’s first critics:

they expected a strict tonal hierarchy – it is not the chromaticism itself that
offended, but its apparent incoherence – that would organise the themes and their
transformations into a coherent narrative. Instead, Franck’s disconcerting, rapid
chromatic progressions left even sympathetic commentators searching for some
principle with which to order the thematic material and justify its formal
disposition.476
Cook’s final assessment of the movement (he does not consider the other
movements of the Symphony) is that “the dialectical disposition of thematic
material is a map, a grid, through which we interpret the chromatic tonal process
as the developing, affect-bearing layer of the listening experience.”477 In some ways
this assessment is unproblematic; it largely seems to fit with my analysis. With the
nature of nineteenth-century form, though, it seems likely to me that Franck
intended this work to be heard in its three movements, and furthermore, that the
movements constitute – on some level – a larger structure that informs our
perceptions of the Symphony. Therefore, whilst each movement should be
considered separately, they should also be considered as one part of a larger
whole. Though Cook’s experiences through the “tonal map” are indeed a
legitimate experience of the processes of this movement, axial tonality invites us to
examine harmonic unity at a larger level, both within and between movements.
Given this, let us now consider the second movement.
476 Ibid., 159.

477 Ibid., 168 - 69.
226
Second Movement
In contemporary reviews of Franck’s Symphony, this movement is the least
criticised of the three: “the oasis in this desert” as Camille Bellaigue called it.478 In a
novel twist of form, here Franck telescopes the Scherzo inside an otherwise regular
slow movement. The outer sections feature a Cor Anglais solo, a point of
contention for Franck’s critics; some of them thought its inclusion put the work
outside the genre of the Symphony.479 The Scherzo section, from Reh F – O+3, is
similar in nature to Berlioz’ “Queen Mab” Scherzo from Roméo et Juliette, a possible
influence. Franz Berwald also uses the overall A–B–A / Slow–Scherzo–Slow form
in his Symphonie singulière, composed in 1845. 480 Regardless of other possible
precedents, this Allegretto remains one of the most innovative experiments in
telescoping two movements for inclusion within a tripartite Symphonic
framework.
Figure 8.2: Franck, Symphony in D minor, II

B-flat minor
Allegretto: ‘Slow movement’
1 – 40 B-flat minor; D-flat (Vβ) 13 – 14 i [16: Reh. A; 39: Reh. B]
41 – 48 Suggests D-flat, then B-flat major Vβ – I [48: Reh. C]
48 – 56 B-flat major I
57 – 61 V/C-flat – C-flat major WT2
62 – 69 Toward B-flat minor - E♭#6 at 68 to G minor i – ivα [64: Reh. D]
69 – 86 Through D-flat minor (vβ, 70 – 71); then a chain of diminished chords built on G
and A-flat. V76 – i (G minor) in 78 – 79, before a re-transition to B-flat by bar 86.
[78: Reh. E]
87 – 90 B-flat minor i Reprise of opening
91 – 94 B-flat – D major i – iα
94 – 96 D major V/ivα
479 Ibid., 608.
480 Ibid., 600.
227
‘Scherzo’
96 – 108 Transition – G minor cadence at 104 ivα [96: Reh. F]
108 – 134 G minor: plagal cadence at 117 – 118 ivα [113: Reh. G; 124: H; 133:
I]
135 – 142 E-flat major IV
143 – 150 G-flat to V of E-flat Iβ – V/IV [149: Reh. K]
151 – 175 E-flat area IV [159: Reh. L; 174: Reh. M]
176 – 183 G minor ivα
184 – 199 G minor – C minor ivα – WT 1 [194: Reh. N]
200 – 221 B-flat minor [218: Reh. O]
Almost all in the subdominant axis
‘Slow movement’
222 – 233 B major ♭II (or IVβ) [229: Reh. P]
234 – 248 Transition: sounds like a D major arrival at 245 – 248; this could possibly be aurally
interpreted as a flirtation with V/G minor – more subdominant bias. [236: Reh. Q]
249 – 262 B-flat I
The first sixteen bars show a tendency toward flat keys; V7 and vii of IV are both
implied in bar five. As well as flat bias there is a mirroring of the first movement
key scheme in the use of vα (and Vα) or iii and III. This begins at Reh. A-2,
reinforced by vi (G-flat minor), which acts as a plagal toniciser of D-flat in bars A-3
– 2 and Reh. B-2 – 1. This tonal emphasis might be linked to the dual slow
movement/scherzo function of this movement, as elements of parody frequently
occur in Scherzo movements, and the B-flat minor – D-flat minor relationship
could be interpreted as a mirror of the D minor – F minor relationship so prevalent
in the first movement.481 Franck reinforces this mirroring of tonal processes further
by modulating to B-flat major with a new theme at Reh. C-1; this same shift
occurred when we heard the second subject in the first movement.
481It seems that the f-A a-F opposition in the Quintet (that reinforced a relational coherence
across movements in a tonal sense) was a preoccupation of Franck’s.
228
Fragmented themes are developed, working through C-flat (IVβ) and G minor
(ivα) before a digression to G-flat at Reh. D+5. This might be a composing out of
the G-flat heard briefly in the opening harmony, but it is momentary as the music
moves through G minor at Reh. E+2 before returning to B-flat. We hear the first
four bars of the Cor Anglais melody before we reach a familiar Franck-style
fermata. The next four bars belong to the horn, generating an axial shift to D
minor. The upper woodwinds then induce the semitonal shift from F to F-sharp,
and the music rests on D major with another fermata at Reh. F. Therefore, although
Franck employs a representative of the dominant axis, the ongoing emphasis is on
the tonic and subdominant axis keys.
The Scherzo begins scurrilously, with urgent triplets in the violins outlining G
minor. This opening sounds like accompaniment with the solo missing, as it uses
the harmonic outline of the Cor Anglais theme without the melody itself. The
music cadences onto G minor at Reh. F+8, while other harmonic inflections are
similarly parodistic: a touch on E-flat (iβ in G minor); a sudden shift from V/G
minor to B-flat (III or Vβ/G minor) then D minor (v); and a plagal reinforcement of
B-flat major between Reh. H+4 – 5. A shift to E-flat major occurs at Reh. I+2
alongside the first instance of dotted rhythms in this movement. Franck
emphasises the chord of C-flat (♭VI in E-flat) and in Reh. I+8 reinterpreted as B
natural in the progression III – V – I in E-flat. (This may be an echo of the earlier
“miraculous” modulation at Reh. C-1 with its tierce de Picardie effect: III – V – I in
B-flat minor/major.) The music remains around E-flat and G minor until an odd
progression: from Reh. N-2 we progress through C minor, E-flat6/4, B76/4/2, A (with G-
sharp min7 acting as dominant), then C-flat alternating with B-flat. The music
sounds as though it is building to E-flat through German sixth and dominant
preparation, so when the Cor Anglais tune begins in B-flat minor at Reh. N+6 it
seems rather odd, particularly as there is a iv – i inflection in B-flat major in the
229
previous bar. Like the early emphasis on the dominant in the first movement,
subdominant axis keys permeate the Scherzo so strongly that B-flat begins to
sound like V of E-flat; hence, when B-flat returns with the main theme at Reh. N+6
it virtually sounds as though it is in the dominant. This is similar to one of the
harmonic idiosyncrasies of the Quintet: the Neapolitan is used as an approach
chord to I, which has the effect of sounding like ♭VI of IV.
From Reh. N+6 the Cor Anglais melody reappears but the triplets in the violins
remain; melodically, we are in slow movement territory while retaining a Scherzo
articulation. This remains a literal reprise until a shift to B major (IVα) at Reh. O+4.
It is possible that the unusual harmonic events of Reh. F-3 were a foreshadowing of
this modulation, and also possible that this section is parodying the E-flat minor –
D minor relationship of the first movement, a hypothesis which is strengthened by
musical geography, as in both movements the modulation to IVα occurs near the
end.
The slow movement returns, texturally speaking, at Reh. Q-2, with a
reconfirmation of the tonic after all the subdominant emphasis. From the B major
discussed above (IVβ), there is a shift to its dominant, F-sharp major (Iβ of this
movement). There follows a short transition (where B is respelled as C-flat) until a
cadence into D major at Reh. Q+11, confirming the tonic axis. We might interpret
this as further emphasis on III. From D it is only a short distance (tonally speaking)
to B-flat, confirmed by a cadence at Reh. Q+23. It is also significant that the final
cadence is iv7 – I; only one semitone different from a German sixth built on E-flat,
but with a much more passive effect; the voice-leading of IV7 or iv7 is almost
identical to a German Sixth in that the notated ♭7 actually resolves upward like an
augmented sixth.
230
Subdominant emphasis saturates this movement; it permeates foreground
middleground and background. Such an effect gives a relaxed “inactive” feel to the
movement, and sufficiently avoids emphasis of the dominant and dominant axis.
We might also consider the movement parodistic at some level – certainly the
scurrying triplets and the imitation of earlier harmonic events lend this feeling –
but a better description might be “reflective”, as this movement gently reiterates
relationships from the first movement without being overtly repetitive. It also
balances the first movement tonally, however, with its emphasis on subdominant
key areas rather than dominant ones. We might nevertheless consider this
reiterative; as the subdominant has “collapsed” into Iβ (B-flat) – we might consider
whether this signals the finale’s course so strongly that it undermines it. Indeed,
given all that has gone before, one wonders how the finale might feel tonally
conclusive when the first two movements have already affirmed the tonic, or at
least the tonic axis, so decisively.
Third movement
Franck’s critics castigate this movement. Like many of his chamber works, the
third movement of this Symphony uses the late-Beethovenian technique of reusing
themes from earlier movements in the finale. However, we will examine the tonal
plan first: major third relatives and wholetone related keys extend the tonic axis,
whilst the subdominant axis provides the main opposing force.
Figure 8.3: Franck, Symphony in D minor, III

D major
Allegro non troppo
1 – 36 D major I Touch on WT 1; V – ii progression
[25: Reh. A]
37 – 44 D major I
45 – 50 F-sharp major Iα [49: Reh. B]
51 – 71 Transition – B-flat; bar 70 V/I altered to become vii/IVα [68: Reh. C]
231
72 – 79 B major IVα [84: Reh. D]
80 – 87 E-flat major IVβ
88 – 93 G major IV
94 – 97 V/B major V/ivα
98 – 124 B minor – flux – B minor ivα (ii7–V7 103 – 104; V7 – i 109 –110)
[98: Reh. E; 121: Reh. F]
125 – 139 B minor iv 2nd movement theme [138: Reh. G]
140 – 170 B – G – E-flat IV axis – IVβ – IV – IVα 1st subject
[161: Reh. H]
All based on the subdominant axis
171 – 194 A-flat WT 1 [187: Reh. I]
195 – 202 C WT 1
203 – 211 E WT 1 [203: Reh. K]
212 – 227 G minor iv [224: Reh. L]
228 – 245 Around G minor iv Sounds like V/C minor
246 – 267 V of D V–I [246: Reh. M; 262: Reh. N]
268 – 299 D major I [296: Reh. O]
300 – 317 D minor – major i–I 2nd movement theme
[316: Reh. P]
318 – 329 Axis sweep through every ‘9’ chord from E9 – F9. E9 rises chromatically up to C9 at
326, then becoming II9 – V9 in B-flat
330 – 345 B-flat Iβ 1st movement ‘C’ theme
346 – 365 Short transition – D I 1st movement ‘A’ theme
[346: Reh. Q]
366 – 381 E-flat ♭II
382 – 389 E WT 1 [382: Reh. R]
390 – 393 A-flat WT 1 IIα WT related to both E and B-flat, the chords
of transition
394 – 397 Transition to tonic via II9 – ♭VI#6 – I6
398 – 440 D major I [398: Reh. S]
One of the most important tonal irritants makes itself known in the first 36 bars: a
manipulation of the cycle of fifths. Though the music begins securely in D, at bar
232
18 there is an alternation of a B7 chord with one of F-sharp minor in the pattern V –
ii – V – ii of E. This gives the effect of cycle of fifths movement, but backwards, so
instead of the music moving forwards (as it would in regular cycle of fifths
movement) it seems stuck, going nowhere. This occurs again between Reh. A+4 – 9
with the progression V – ii – V – ii – V♭9 in D. The technique is effective, as it can
either suggest a key area without necessarily modulating there or weaken an
arrival by its inverse harmonic pattern. (In this case, the arrival of the tonic is
further weakened by the use of ♭VI between V♭9 and I at Reh. A+10 - 11.)
Furthermore, as in the Violin Sonata (discussed in the next Chapter), backwards
cycle of fifths manipulation can give an impression of drifting; however, in this
movement it suggests modulation to E major that will be substantiated later. In the
classical period a modulation to V/V would be common as preparation for a
modulation to the dominant; bar 22 – 28 in the Eroica is such an example. Yet we
know retrospectively that the dominant is not a main player in this movement,
suggesting there is an ulterior motive. As Franck is using passive wholetone
modulation to effect tonal flatness, it is possible that E is utilised as a wholetone
relative rather than V/V.
From Reh. B+2 through Reh. D+9 there is a transitory section in which Franck uses
German sixth chords to make his way around the entire subdominant axis:
beginning in B and using a new theme, the initial tonic (B) is reinterpreted as a
German sixth in the new key (E-flat) and resolves there as such. The procedure
repeats itself in E-flat, whereby E-flat7 becomes a German sixth of G. At Reh. E
we are back in B, having completed a sweep of the entire subdominant axis, but
this time in the minor mode in preparation for the next section. This is a clear
example of Schubertian structural stability, as well as Franck’s axial thinking.
233
The first instance of Franckian cyclicism occurs at Reh. F+4, which uses the Cor
Anglais theme of the second movement before swiftly turning to the major for
another subdominant axial sweep similar to B+2 through Reh. D+9. It seems that
this modulation is moving “around” rather than “forward”; this stretch of
subdominant material seems designed to offset the dominant axis bias of the first
movement. However, it is arguable that this bias has already been sufficiently
offset by the subdominant bias later in the first movement and further in the
second. Is the finale attempting to perform a job that is not required? Only at Reh.
H+8 does “forward” movement begin again, with E-flat moving to A-flat, this time
instigating another axial sweep related in whole-tones that might seem to be
dominant preparation: A-flat, C and E. Though we might refer to these keys as the
supertonic axis (as they were functioning in the Quintet) in this case the keys are
expanding the tonic wholetone scale rather than preparing the dominant, which
only occurs here at a foreground level. Additionally, we might interpret this axis as
IV of IV. But either way, the principal effect is of passive modulation. The
movement reaches E at Reh. K, which could possibly be the fulfillment of the cycle
of fifths movement begun in the first section. It is worth noting, though, that
although we have reached ii – a traditional dominant approach chord – V does not
materialise. What is stranger still is that thematic arrival does not align with tonal
arrival: the arrival of one of the finale’s main themes at Reh. I is not reinforced
cadentially – in fact, its arrival engenders axial modulation through A-flat, C and
E. As such this technique lacks agency, and is explicitly un-Beethovenian.
More cyclicism appears at Reh. K+9, with a reminiscence of themes from the first
and second movement in G minor. This is not a direct quote: in the antecedent
phrase, however, the rhythm in the cello combines the first theme from the first
movement within the tie across the bar that permeates the themes of the final
movement, whilst the consequent phrase quotes the main theme of the second
234
movement and is merged with the rhythm of the second subject from the first
movement.
Figure 8.4: Franck, Symphony in D minor, III, Reh. K+9 – 12
So, despite its apparent newness within this movement, Franck builds this theme
on motivic ideas from this final movement, and reinforces the lack of thematic
contrast without which sonata principle cannot operate. Though he may be using
cyclicism both to provide contrast from the third movement’s themes and to imbue
a sense of overall coherence between movements, the integration of cyclic themes
in this movement is problematic: the quotations sound awkward as Franck
transplants them verbatim rather than crafting them as integrated motifs. The
worst offender is the Cor Anglais theme from the second movement, and Tovey
describes the outcome:

The wisdom of the serpent is foreign to the harmlessness of the dove; and the
combination has an exotic glamour all its own. The artistic danger of the
combination is that innocence may break through in a disconcerting form of bad
taste: the saint does not really know what the world understands of its formulas.
And so the exquisite main theme of this middle movement will eventually find
itself striding grandly, in its white confirmation dress, over a large area of the
finale; and the finale has a mildly sentimental second theme of its own, which of all
235
types of phrase is most vulgarised by being given to red-hot brass, however softly
played.482
We return to Reh. K+9. Although this subdominant minor has been a “tonal
irritant” from the outset, here it can appear without threatening the true tonic. It is
particularly unnerving though, because Franck crafts it in such a way that it
sounds like v/C minor. Extensive and chromatic dominant preparation follows
this, until the arrival of D major at Reh. N+6, featuring this movement’s main
theme with tutti forces. This arrival too, however, is undermined by the placement
of the seventh in the bass and scoring emphasising the leading note and
augmented fifth. Further to this, a German sixth chord at Reh. O+3 replaces (and
hence undermines) the dominant (as in the second movement). These examples
suggest that Franck is deliberately avoiding direct movement from dominant to
tonic. The music then turns to D minor, repeating the Cor Anglais theme that
Tovey so maligns above. Though it is unsurprising that Franck reprises this theme
in the tonic minor, he gives its return enormous weight with thick orchestration
and loud dynamics. We might consider this a transcendent retonicisation, as the
(nominal) main theme for this movement does not occur in the tonic. One recalls
Tovey’s comment about the theme “striding grandly”483: this recall simply sounds
out of place and character. In defence of Franck’s orchestration, he confided in his
pupils that if he were to write another Symphony “he would never provide the
brass with parts like those he had written for them in the finale”. 484 However, to
accept this is to attribute the problem to scoring and nothing else.
Beethoven, the nineteenth-century archetype, employed daring thematic recall in
his Fifth Symphony, recalling part of the Scherzo theme in the finale. We can
interpret Beethoven’s recall as a threat to the sonata mechanism, undermining its
482 Tovey, Essays in Musical Analysis Volume II: Symphonies, 60.

483 Ibid.
484 Léon Vallas, César Franck, (George G. Harrap & Co. Ltd., 1951), 216.
236
teleology and “darkness to light” programme with its use of the tonic minor and
placement at the end of the development. One might say that it is implacable,
representing a fate that will eventually be victorious (death), but which the music
can transcend. Tovey claims that it is: “a stroke of genius” and “a memory which
we know for a fact but can no longer understand.”485 This is because Beethoven’s
recall is not purely thematic but also reminiscent of the gesture and timbre of the
oboe’s recitative-like moment in the first movement as well as the first movement’s
irregular phrase structure at the close of the development. Hence, in the finale,
Beethoven recalls the three crisis moments at once: the first movement
development’s irregular phrase structure, which obliterates forward movement;
the oboe recitative, which does likewise by intrusion; and the funeral march, which
is a dispirited version of the initially confident Scherzo theme. If we read the Fifth
Symphony as autobiographical (as many commentators have done), it might
represent survival: as much of a victory as possible within the confines of the
human condition, acknowledged by the funeral march. Beethoven’s cyclicism in
the Fifth Symphony, therefore, is cumulative. There is evidence to suggest that
Brahms appreciated this depth of cyclicism (I discuss this further in Chapter Nine)
but it seems that Franck’s cyclicism is merely singular, rather than cumulative. 486
This is why we do not make similar claims of Franck’s cyclicism in the Symphony:
something more perfunctory is going on here.
The arrival of a new section refreshes the music: an axis sweep emphasising
the tonic wholetone scale related E-A-flat-C and ninth chords before a
modulation to B-flat at bar 330, with antecedent-consequent phrasing between
485 Donald Francis Tovey, Essays in Musical Analysis. Volume I: Symphonies (London: Oxford
University Press 1968), 44.
486 Incidentally, Franck does emulate Beethoven in one aspect of his cyclicism: the order of his
recall. In his fifth Symphony, Beethoven recalls each gesture in backwards chronological
order: first the Scherzo, then the oboe recitative, followed by the development’s conclusion;
Franck similarly recalls the second movement theme before that of the first movement.
However, Franck’s cyclicism remains purely thematic, whilst Beethoven’s is thematic,
gestural and rhythmic (in terms of phase structure).
237
the flute and clarinet. As well as simplifying the texture from the immediate
past melody, it “corrects” the V7-ii9 backward cycle-of-fifths movement of
bars 18 – 21 to ii – V. At Reh. P+14 we arrive at B-flat, where the closing theme
of the first movement appears. This recurrence seems much less hackneyed
than the reappearance of the Cor Anglais theme because of the altered
harmony in the third bar: it was V7 in the first movement, and here IVmaj9,
lending a more relaxed feel than the decisive V – I emphasis it previously
iterated, though squareness is still present. The aural distance between the
first and final movement is also a factor here, and hence this theme sounds
like a more distant memory than the Cor Anglais theme, which, due to its
temporal closeness, sounds like a simple repeat. We might suggest that in
many cases, temporal distance seems to govern the success of this type of
cyclicism, which we see in the recalled themes of Beethoven’s Ninth and
Bruckner’s Fifth and Eighth Symphonies. Beethoven’s Fifth may be the
exception, but we must remember that his approach is multi-faceted,
compressing several significant gestures into one important passage, and that
the Scherzo reprised in the finale had not enjoyed an emphatic reprise within
its own movement. Hence, temporal distance in Beethoven’s Fifth does not
seem so problematic. Though Tovey is not explicit, his criticism of Franck is
almost certainly informed by his admiration of Beethoven’s monumental
achievement.
Returning to Franck, the third movement of the Symphony reminds us of the ninth
chord’s importance in the transition from Reh. Q to Q+19. Rocking between D/F-
sharp and D/G (no third, and thus more like G9 than D), the second chord becomes
functionally ambivalent. It now seems that D has been confirmed beyond doubt,
and V – I movement seems redundant, hence the emphasis here on less purely
functional harmony and ninth chords. The G is persistent enough for a shift to E-
flat at Reh. Q+20 to sound like an interrupted cadence, followed by another

238
semitone shift up to E at Reh. R. We then hear an axial shift: the G-sharp of E (in
first inversion) is reinterpreted as I/A-flat – axial relative of II, which we might
expect to reinforce the approach to the dominant as this the final approach to the
tonic within the symphony as a whole. However, the transition to the tonic is made
via a II9 – ♭VI#6 – I6 progression: Franck seems utterly determined to avoid ii – V –
I! D major arrives at Reh. S, via a foreground E and G minor; D is settled on at Reh.
S+28 with deliberate V to I movement in contrast to the “Tristan” cadences that
ended the other movements.
Further thoughts
Prior to its premiere, Franck discussed aspects of the Symphony with his pupil
Pierre de Bréville. It is apt to quote him at length here:

At the end of the first movement there is a recapitulation, exactly as in other
symphonies, but here it is in an alien key. Then follow an andante and a scherzo. It
was my great ambition to construct them in such a way that each beat of the
andante movement should be exactly equal in length to one bar of the scherzo,
with the intention that after the complete development of each section one could be
superimposed on the other. I succeeded in solving that problem. The finale, just as
in Beethoven’s Ninth Symphony, recalls all the themes, but in my work they do not
make their appearance as mere quotations. I have adopted another plan, and made
each of them play an entirely new part in the music. It seems to me successful in
practice, and I fancy you will be pleased!487
Clearly, Franck makes several points that reinforce our observations. Firstly, he
describes the recapitulation (almost certainly referring to the E-flat minor section in
the Allegro at Reh. O) as being in an alien key. This confirms the consideration of E-
flat – in conjunction with its axial relative, G minor – as being a tonal irritant,
clearly outside the realm of the tonic. Secondly, Franck’s idea of creating a scherzo
equal in length to the slow movement suggests that he is concerned with structure

239
and pace issues, attempting to avoid one movement being significantly longer than
another. Furthermore, describing the second movement as “solving that problem”
adds weight to the importance of compositional anxieties in late nineteenth-
century symphonists. Thirdly, Franck’s assertion that the cyclic themes in the finale
play “an entirely new part” implies an attempt at formal originality, reinforcing
the idea that existing symphonic forms were becoming problematic and required
revision.
So, is Franck successful? It seems to be a difficult question to answer. A number of
peculiarities in Franck’s Symphony imply an awareness of structural issues, and
his handling of large-scale form and large-scale tonality looks, on paper,
successful. Aurally, however, his techniques become overbearing and overstated.
The double exposition of the first movement could have been effective but for the
emphasis placed on it by jagged juxtaposition. One surmises that it would have
been sufficient contrast to hear the fairly remote modulation, but the complete stop
and exact repeat of the opening material seems not only caricatured in its
obviousness, but also neglects the sonata principle. The late-Beethovenian
technique of problematising a tonic to rescue it later is more effective when the
tonic quietly drifts into the distance; Franck’s D minor tonic demands we sit up
and notice it has gone! We must concede, however, that despite the lack of subtlety
in this technique it does suitably undermine D’s tonicity, proven by the appearance
of D in the development section where it sounds more like an exotic modulation
than a return to the tonic. Yet to view the tonal scheme as two rival tonics – D and
F minor – makes the first movement seem more logical. Axial analysis is helpful in
identifying areas where a key is prolonged, but there are instances where this type
of analysis falls short and passages are better interpreted as wholetone relatives of
the tonic (and therefore tonally flat) than as supertonic axis members. Axial
analysis reveals a relative lack of tonal tension that corresponds with how the
music sounds, and to label the A-flat-C-E axis as the supertonic implies that the
240
keys function as preparation for the dominant, but this is rarely the case: therefore
the interpretation of such keys are wholetone relatives to the tonic is more
functionally accurate. Following on from the early emphasis of the dominant axis,
it is logical to emphasise subdominant keys in the recapitulation and later
movements to balance the rivalry. Classical Sonata form movements were often
followed by slow movements in the subdominant for much the same reason. To
summarise, we might say that Franck confirms D logically, but at the expense of
rhythmic fluidity, so the Symphony as a whole makes tonal sense but frustrates
with its rhythmic plainness.
In terms of form, Franck was ahead of his time in telescoping a Scherzo into a slow
movement, with its shape as a large-scale variation form with subtle parodistic
aspects. While fermatas add unnecessary weight, as in the first movement, overall
the movement is formally innovative and melodically lyrical. It makes use of axial
modulations that reinforce tonic and subdominant axis areas, with particular
emphasis on IVβ (the Neapolitan) as a tonal area. The movement avoids the
dominant and feels “inactive” as a result – that said, as a slow movement in
variation form, “flatness” is to be expected.
What does seem to work in the final movement is the manipulation of the cycle of
fifths. (I discuss this technique further in relation to Franck’s Violin Sonata in the
next chapter.) However, such instances occur only at a foreground level and can
only promote action (or inaction) for so long. More successful – in terms of creating
integration of the Symphony as a whole – is the reuse of G minor with the
recurrence of ‘a’ from the first movement. Both the theme and the key remind us of
the first movement without creating additional tonal work for the finale to do (as G
minor has already been refuted). As in the first movement, wholetone-related key
areas play an important role in expanding the tonic area without inciting tonal
241
drama. This keeps the movement tonally flat, with the aural emphasis being on
quotations borrowed from other movements.
Franck’s techniques in terms of cyclicism may be summarised by René de Récy,
who, when comparing Franck’s Symphony to Saint-Saëns’, stated: “[I]n a word,
Saint-Saëns develops; César Franck arranges and combines.” 488 However, the same
cannot be said of the Violin Sonata, one of Franck’s most well integrated cyclic
works, which is the subject of the next chapter.
488
Franck", 139.
242
Chapter Nine
“Proto-Progressive” Tonality and Franck’s Violin
Sonata
Franck’s Violin Sonata is an example of what I am arguing is proto-
progressive tonality. This is not “progressive tonality” in that it does not
begin in one key and end in another: rather, the Sonata’s opening
movement undermines the assumed tonic, and the rest of the Sonata
makes work of righting the tonality. There are some similarities here to the
process at play in the opening movement of the Symphony. However,
perhaps due to the more lyrical nature of the solo Sonata and a more
assured handling of tonal and rhythmic processes – as well as a more
subtle working of cyclic techniques – this Sonata is an exemplar of all that
a proto-progressive work can be and do. Moments in the first and third
movements arguably look forward to the impressionist techniques of
Debussy, which is further evidence of both Franck’s ingenuity as a
composer and his influence on the succeeding generation of French
composers.
Violin Sonata in A
This 1886 work is one of Franck’s best known and loved.489 The movement
employs a single “motto” theme throughout, which outlines a major ninth
built from intervals of rising and falling thirds. Davies describes how this
interval “posed the possibility of a rocking melodic figure – a kind of
cybernetic oscillation – deriving its components from the notes of the
489 Laurence Davies, César Franck and His Circle (Boston: Houghton Mifflin Co., 1970), 229.
243
common chord. The composer’s favourite sequential devices were
facilitated by its use”.490 Despite its derivation from the “common chord”,
Franck’s ninth is something else. Indeed, the terms “rocking” and
“oscillating” are apt terms to describe the harmony of the first movement.
Whilst many works of the central European tradition tend to “compose
out” foreground details into the background (which can often stimulate
axial relationships), this work does not seem to. This sets the Violin Sonata
apart from Franck’s Quintet and Quartet, and marks it as all the more
unusual given the clear organic relationships between melodic material in
and between movements.
First movement
The following table outlines the important keys used in the first
movement.
Figure 9.1: Franck, Violin Sonata, I

A Major
Allegretto ben moderato
1 – 10 A I 1st subject (violin) ‘A’. 9 – 10:
German sixth
11 – 14 to C-sharp Iα 13 – 14: German sixth
15 – 30 to E V Sequence, 17 – 24
31 – 34 E V 2nd subject (piano) ‘B’
35 – 36 Flux around B flat and D flat
37 – 38 to F-sharp minor ivα or vi
39 – 44 F-sharp minor ivα or vi ‘B’ (piano)
45 – 46 to C-sharp minor iα
47 – 50 C-sharp minor iα ‘A’ (violin)
51 – 56 C-sharp minor – B minor – A minor Sequence: 51 – 55
57 – 59 flux fragments of opening
490 Ibid., 231.

244
60 – 62 to A I fragments of opening
63 – 72 A I ‘A’ – recapitulation
71 – 74 around F-sharp ivα or vi based on ‘A’ (violin)
75 – 82 Sequences, on C-sharp then F-sharp: combines rhythm of 1st and melody of
2nd
83 – 88 to A I
89 – 117 A I ‘B’ (piano) recap
The opening seems virtually impressionist due to the added notes in the
first chord (whose function we do not yet know) and the lack of an
assertive cadence in any key. The violin enters in bar five with the “motto”
theme. However, this chord cannot automatically be interpreted as tonic
or dominant (or anything else) as neither has been clearly defined – in fact,
the use of V9 and ii chords, at least initially, impart a clear lack of
functionality – though, their function becomes clearer as the movement
unfolds. This is largely due to Franck’s deceptive use of quadratic
structure. The violin’s entry in bar five establishes quadratic structure,
confirmed by a new melodic idea in bar nine alongside a change of key
and harmony in bar thirteen. With this quadratic structure comes a
rhythmic hierarchy between bars of strong-weak-medium-weak. At first
juncture, B minor (ii) sounds subsidiary to E9 (V9), but with the lack of
any other harmonic movement it might well be interpreted as I9 – v – I9 –
IV in E mixolydian, meaning (because of the rhythmic emphasis placed
upon it) E9 could easily be a tonic or a dominant. To add to this
ambiguity, E9 – B minor is a clockwise movement around the cycle of
fifths, the opposite of functional tonality.491
491If V9 – ii is reversed and repeated, the effect is altogether different, as the harmony
pulls towards resolving onto I in the consequent bar because of the emphasis placed on V.
Placing ii – V9 in this order gives a much more assertive progression, in which case A
would, unequivocally, sound like the tonic.
245
Here, Franck’s use of the “chord pair” is significant both rhythmically and
harmonically. Often paired with iambic rhythm, this device was a
favourite of Franck’s and occurs when “the second chord carries with it
the impression of a sforzando.”492 Trevitt describes the application of this
technique in the first movement (bar 9) as “classical”, and indeed this
instance of I – ♭VI#6 would not be out of place in a late classical
symphony. However, Franck also “applied his method of thematic
development to a harmonic context; a chord pair, for instance, may be
repeated with slight alteration to the second chord, resulting in a stronger
implied sforzando.”493 An important instance of this also occurs in the third
movement of this sonata.
When A arrives in the second beat of bar 8, it is in the weakest possible
place – harmonically correct, but metrically wrong. The clockwise cycle of
fifths movement is partially fulfilled, as the first chord arrived on in bar 8
could be interpreted as F-sharp minor7 in first inversion, or A with added
sixth, or a suspended F-sharp resolving on the second beat. Whether or
not we consider the notion of double-tonic complexes at play here – the
idea of F-sharp minor (IVα) rivalling A for tonicity is perhaps unlikely – it
does suggest, at least at this point, that this opening possesses a strong
subdominant bias. Bar nine – A again – does not strengthen A’s case for
tonicity, but is found lacking arrival and harmonic traction as it has
already occurred in the previous bar! The I – ♭VI#6 – I is a step toward
confirming A, and this might be read as an undermining of A itself as the
chord becomes a German sixth in bar 11, resolving to the new tonic, C-
sharp. The cadence into C-sharp is also an emphatic one: German sixth –
492 John Trevitt and Joël Marie Fauquet, "César Franck," in Grove Music Online. Oxford
Music Online.
493 Ibid.
246
V7 – I, and furthermore, C-sharp arrives at a structurally strong point,
suggesting it has as much claim to tonicity as A at that point. It could even
suggest the possibility of F-sharp, given the subdominant bias in this
movement.
Though the use of axial relatives in this sonata is immediately evident, a
strong subdominant-axis bias also permeates the first movement, and
arguably the entire sonata.494 This, as well as ample use of the German
sixth chord oscillating with the tonic, serves to undermine the tonic in the
work overall by preventing it from fulfilling a strong I – V – I progression;
effectively, the German sixth “emasculates” the tonic. Certainly, the first
movement is not devoid of the dominant-axis altogether (there is an
arrival on E at bar 25, which might act as the dominant of A or the axial
dominant of C-sharp, depending on one’s perspective) nor inter-
dominants, but such sonorities often turn in unexpected directions and do
not function as we might expect. Similar progressions to the V9 – ii
discussed above occur at bars 25 – 27 (in B major, V/V) and bar 32 (as part
of the second subject). There is also a sense of harmonic relaxation in the
#vi7 – vi – E/g-sharp progression which begins the sequence in C-sharp
minor at bar 51. Between here and bar 59, each phrase is repeated a
wholetone lower each time. Of particular note is the section between bars
55 and 60, which suggests C major by that chord and its German sixth.
However, C (Vβ) then shifts to E (V/A) at bar 60, as we might expect at the
end of the development section – yet this E does not perform a dominant
function, “resolving” into V9 before the recapitulation (the harmony
which does not include an emphatic cadence to the tonic). This passivity
The second movement is in the subdominant minor, D minor, and the third movement
494
modulates from around D minor to F-sharp minor, another member of the subdominant
axis.
247
reinforces the idea that “flatness” is Franck’s desired quality, as it is
movement between, rather than within, wholetone scales which generates
harmonic momentum (bass movement in the strong cadential progression
vi – ii – V – I oscillates between the two whole tone scales). This is in
keeping with the subdominant bias of the movement; as well as belonging
to the same wholetone scale, these tones move away from, rather than
toward, the dominant. It is apt to remember Bribitzer-Stull’s discussion of
directional tonality and double-tonic complexes from Chapter Four, and
consider that this might apply to this Sonata:

While directional [similar concept to progressive] tonality comprises the
transformation of tonic function from one tonic chord (key) to another across
the span of a piece, the two tonic keys of a double-tonic complex are not
simply its opening and closing tonal centres: rather, they are the dual
harmonic poles between which the music oscillates, at one point suggesting
one key, and on another occasion the other.495
This seems to be the case here: rather than a sense of teleology throughout
the Sonata, there is a sense of two centers, which, though linked, function
independently. How Franck achieves this is discussed below.
Rhythmically, in this work Franck often undermines harmonic arrivals
such as V – I by placing them in weak places within a quadratic phrase, as
discussed above. At the transition to the recapitulation (bar 62 – 63), E is
both the final beat of bar 62 and the first beat of bar 63, giving it no sense
of arrival. Furthermore, bars 83 – 84 use the V9 – ii – V9 – I progression –
in E – and, just as the first time we heard it, this order weakens E’s arrival.
Conversely, modulations into F-sharp and C-sharp (bars 38 – 39 and bars
46 – 47 respectively) are rhythmically and harmonically correct, affording
Matthew Bribitzer-Stull, "The End of Die Feen and Wagner's Beginnings: Multiple
495
Approaches to an Early Example of Double-Tonic Complex, Associative Theme and

Wagnerian Form," Music Analysis 25, no. 3 (2006): 324.
248
them a much greater sense of arrival.496 Most significantly, it would be
difficult to argue that the first movement fulfils a I – V – I progression at a
background level, as there is no prolongation of V due to the strong
subdominant bias. This viewpoint might become problematic if we simply
list the keys Franck modulates to in this movement – indeed; he does
modulate to the dominant. The issue, though, is that the second subject in
E is not tonally secure, and the only “confirmational” cadence comes at
bars 38 – 39 in F-sharp minor – a subdominant axis key, IVα. Hence, there
is a distinct lack of correlation between thematic arrival and tonal solidity,
making the establishment of tonal hierarchy difficult. Furthermore, it is
unclear where the development begins – by the time we sense any
development we are back in the first subject group, and the recapitulation
seems to passively occur rather than articulate its achievement. All these
issues undermine any sense of dominant prolongation we might
otherwise notice within the course of the first movement.
The subdominant axis encompasses D, F sharp and B-flat. A glance over
the first movement will confirm that the music spends ample time in these
areas, and often at significant structural points. From bars 34 – 37,
following the occurrence of the second subject (given by the piano in bar
31, outlining a descending scale and using both G-sharp and G natural),
there is a “leaning” toward B-flat. Though V♭9 of B-flat occurs twice, we
do not get a cadence into B-flat, being “interrupted” by D-flat each time,
and we might consider this further subdominant bias by the function of
the (supposed) tonic axis being blurred – are F and D-flat Iβ and Iα, or
V/IVβ and V/IVα respectively? Reinforcing the V/IV potentiality, there
follows a cadence into F-sharp minor, in which the second subject recurs.
Interestingly, this harmonic order can be reversed, and is in 45 – 46; V/C-sharp - ii

496
occurs in C-sharp, and is then reversed to become a cadence into C-sharp minor.
249
We stay in this key until a modulation to C-sharp minor in bar 45, and
this, too, is significant; though C-sharp is a member of the tonic axis, it
relates closely to F-sharp, being its dominant, and at bar 47 we hear the
first subject in this key. Furthermore, in the recapitulation, bars 71 – 73
feature the now familiar I – ♭VI#6 – I “circling” in D; bars 75 – 77 feature
V9 – ii – V9 movement in F-sharp; and finally, bars 110 – 113 have a
subdominant bias: I – ♭VI#6 – I occurs in D, and B-flats occur melodically
in the violin. This strong subdominant emphasis in the recapitulation
seems unusual if Franck’s desire is to affirm the tonic. If we consider that
Franck is deliberately avoiding the dominant to emphasise the
subdominant, however, these emphases seem more logical.
Another important aspect is the length of this movement – by classical
standards it is disproportionately short. A survey of two recordings shows
the timings of the movements to be 6.39/8.45/7.36/6.34 497

and
5.56/8.08/7.04/6.19498. Whilst classical finales are often shorter than their
preceding movements, it is highly unusual for a first movement to be so
short; calling into question Franck’s intended function for it. Tonally, we
might describe this as “passive”: a background tonicisation of A is lacking,
and there is a subdominant bias (discussed above). Davies also notes that
this movement lacks development: “The structure of his [Franck’s] music
is strangely inorganic. His material does not develop.” 499 Veritably, no
sooner have we become familiar with the second subject and undergone a
short sequence using the first subject (bars 51 – 55 discussed above), than
the recapitulation begins, with block chordal accompaniment, at bar 63.
The precedent for this type of small non-tonicising first movement might
497 César Franck. Franck: String Quartet and Violin Sonata. Decca. 4768463.
498 Cesar Franck. Franck: Sonata for violin and piano. Decca. 421 154-2 DM.
250
be Beethoven’s Op. 101 or 102 No. 1, discussed previously with the idea of
proto-progressive tonality. Strengthening this possible precedent is the
placement of the second movement in the subdominant minor, which I
examine below.
Second movement
This movement parodies the first, in that the main theme relates closely to
bars 9 – 12 of the first movement. However, unlike the most well known
examples of parodistic scherzo movements, the key is D minor: iv, rather
than the obvious I or i.500 The function of a parodistic scherzo is literally
that: to parody, and this device is most effective if the following
movement is in the same key, i.e. the tonic major or minor. Though it is
not unusual for parodistic scherzos to be in sonata form, it is usually
sonata form without development, as the composer may wish to withhold
the tonal drama of the work overall.501 As such the effect is usually “static”
tonal opposition by choice of a neutral key for the second subject, such as
the tonic major or minor, relative major or minor, or a tonic or
subdominant axis key. Beethoven’s tendency is often to use a type of
modified sonata form but with a non-tonicising key relationship, such as C
major in the Ninth Symphony’s D minor Scherzo and the Ghost Trio’s
slow movement (not a Scherzo, but still parodistic). In this movement,
however, Franck makes ample use of the dominant axis, making this
movement far more dramatic than a traditional parodistic scherzo.
Instances of tonal opposition occur as they would in a tonally dramatic
movement, such as bars 88 – 93, some time before the recapitulation at bar
138. Also note that the E9 chord which did not function as expected in the
500 Examples in this category include Beethoven’s Hammerklavier Sonata and Ninth
Symphony, Brahms’ Op. 7, and Mozart’s G minor Quintet.
501 Examples of this include Beethoven’s Ninth Symphony and Hammerklavier Sonata, as
well as Brahms’ B major Piano Trio.

251
first movement here behaves as a dominant approach chord in the final
cadence, moving II9(♭9) – V7(♭6-5) – i.
So, then, if the role of the first movement is to act as a “passive” prelude,
then why put the “active” scherzo in the subdominant minor, particularly
given its parodistic elements? One explanation is that this is part of the
tonal ploy of proto-progressive tonality; by utilising a key other than the
tonic for an active movement (in that its own tonic is confirmed), then
creates further work for the later movements, which must – if the proto-
progressive plan is to hold – reaffirm the initial “true” tonality.
Figure 9.2: Franck, Violin Sonata, II

D Minor
Allegro
1 – 13 D minor i 1st subject piano ‘C’
14 – 23 D minor i ‘C’ violin
24 – 29 Transition D-flat reinterpreted as C-sharp
30 – 31 F-sharp minor iα Violin descending with ‘C’ extracts
34 - 43 D minor i ‘C’ altered
44 – 47 D minor i ‘A1’; based on ‘A’.
48 – 51 Around F Vβ 2nd subject ‘D’ – no cadence in F
though
52 – 55 Around A V ‘D’; no cadence in A either
56 – 59 V of B minor ivα Spanning diminished 7th
60 – 64 V of D minor i
65 – 66 V of F minor Vβ
67 – 78 F minor Vβ Confirmed by subdominant minor
in bars 69 – 70.
Quasi Lento
80 – 83 Transition F minor – E diminished
252
84 – 87 As 80 – 83, up a minor 3rd, ending on Aug 6th on B-flat, suggesting D
or A
88 – 93 Similar to 67 - 78. F – A – B-flat minor, before A minor (v)
Tempo 1 – Allegro
94 – 97 on Am v Similar to bars 80 – 83; violin G-
sharp diminished
98 – 102 Transition – C Aug 7 /F7d/D7d
(Key change – C-sharp minor)
103 – 108 C-sharp major/ minor – F-sharp minor (Vα – vα – iα) Related to ‘A’
(augmented)
109 – 111 C-sharp minor vα ‘C’ (Piano only)
112 – 114 C-sharp transition – rhythmically related to ‘D’
115 – 121 G-sharp minor vα/V As bars 109 - 111
122 – 125 F-sharp major Iα ‘D’ in piano
(Key change – E-flat major)
126 – 128 E-flat IVβ ‘D’ in piano
129 – 136 Combining ‘C’ in piano and ‘D’ in violin
137 – 150 D minor i Recapitulation
151 – 157 Related to 27 – 32 up a perfect fourth.
168 – 171 B minor ivα ‘A1’
172 – 175 D major I ‘D’
176 – 179 F-sharp major Iα More ‘D’ fragments
180 – 190 Spanning diminished 7th as bars 56 – 59
Poco Piu Lento
191 – 201 D minor i Confirmed by G minor subdominant
in bar 193
202 – 229 D minor/major i/I Coda
As the analysis above shows, the development section (which is certainly
present) utilises dominant axis keys, and there is little emphasis on the
subdominant in this movement, save for the key of the movement itself
being iv of A. The short instance of real sequence occurs in bars 88 – 93,
where the violin melody moves up (though down in register) a major
253
third each time: A – C-sharp – F, oscillating between wholetone scales. As
I discussed regarding the first movement – where the sequence stayed
within the same wholetone scale – this oscillation stimulates harmonic
intensity.
Furthermore, note the backward cycle of fifths movement between bars 17
– 18. Here, there is a two-bar pattern with V on the first beat and ii-dim7
on the third. We might consider this a unifying device as we have already
experienced backward cycle of fifths movement in the first movement,
and this will happen again in the finale. In this movement, the Scherzo,
Franck straightens out this backwards movement within the theme itself
in bars 22 – 23, and a similar process takes place in the finale.
Franck’s rhythmic devices are also at work: from the outset, our sense of
phrase structure is disturbed, as there is a three-bar introduction before
the piano begins the main theme. Accents are placed on the third beat
every second bar in the melody – the highest note melodically in each
phrase, to be sure, but still unsettling; the relaxed iambic nature of the first
movement’s melody is literally reversed (we might consider it
rhythmically parodied), becoming quaver-crotchet rather than crotchet-
quaver and emphasising syncopations. Though the rhythm of the second
subject is less syncopated, the piano provides polyrhythmic triplets in the
accompaniment, creating a dramatic effect. Furthermore, bars 44 – 46
could be interpreted as a type of chord pair, throwing emphasis on the
second beat and reinterpreting the nature of the first subject in the melody
to put the emphasis on the A; this occurs similarly at bars 168 – 171, but
with more decisive emphasis on the second beat. Finally, just prior to the
recapitulation, there is an unexpected 2/4 bar which does not seem to
provide any function other than to disturb the rhythmic flow. Also of note
254
is the four-note figure at bar 202; this figure is later inverted and used as
thematic material in the finale.
Franck’s rhythmic innovations here are similar to those of Brahms. Take
Brahms’ Op. 101 C minor trio, for example: this is a rhythmically
sophisticated work, emphasising units of three and nine, often doubled.
The first movement opens with paired phrases in 9/8; the second theme of
the scherzo features pair of 3/2; and the third movement opens with pairs
of threes (3/4, 2/4, 2/4 but there is a sense of a three-bar phrase in
antecedent and consequent). 502 This 3/4 2/4 2/4 pattern in the slow
movement actually amounts to 7/4, which Brahms then expands on at the
lyrical climax before the trio by essentially doubling his pattern, 3-3, 2-2, 2-
2. The finale is particularly effective: there is an enlarging of tension
between bars 240 – 246 by use of a 9/4 polymeter across 6/8. This
simultaneously heightens the tension and release in the harmony, which
can be summarised as statements of vii7 and I.
I have drawn this analysis from Peter Latham. As he discusses Brahms’
Trio explicitly, he notes that the above verbal description has “all the
dullness of an exercise in arithmetic.”503 However:

translated into music as Brahms translates it, it is manifestly inspired. During
the two bars of three the melody soars as it has never soared before; during
the four bars of two it sinks to rest in a cadence that is final. One has the
experience, so rare in art, of a thought expressed completely once and for
all.504
Though we might easily contest dated musicological description like this,
Latham’s point is that the rhythmic pattern allows the music to reach its
502 Peter Latham, Brahms, Master Musicians Series (London: Dent, 1975), 104.
503 Ibid.
504 Ibid.
255
full shape and gain freedom, and that Brahms’ rhythmic innovations are
as complex as his melodic development. It is through rhythmic
innovation, coupled with a sophisticated musical integration across all
other parameters, that Franck is able to gain the same sense of freedom in
this Sonata.
Returning to the Violin Sonata analysis, one can see that D minor’s
dominant axis plays a significant role in this movement. We might
describe this as an instance of the transcendent subdominant with its own
tonicisation processes, similar to Beethoven’s Op. 127, Op. 101 and 102 No.
1. However, rhythm becomes particularly important in the final two
movements, as we shall see.
Third movement
In The Classical Style, Rosen remarks: “placing the minuet second instead
of third in the order of the movement throws the expressive weight of a
quartet or quintet towards its latter half.”505 It does not seem unreasonable
that Rosen would extend this observation to solo sonatas also, particularly
given the Hammerklavier and its deeply expressive third movement.
Whether Franck’s decision to opt for this order was due to the parodistic
scherzo or expressive weight we cannot be sure, but in expressive terms,
this third movement is substantial in its emotional content. Davies
describes this movement as recitative-like: “the cadenzas and flourishes
proclaim a debt to the baroque composers, but the harmony remains
modern.”506 Furthermore, the movement begins “around” one key (loosely
256
D minor) and ends in another: a further recitative-like attribute. 507 The
key(s) of the movement are an important component of the Sonata overall;
the keys of movements, respectively, are A – D minor – D minor (loosely)
/F-sharp minor – A. On a smaller scale, Franck also links this movement
with the earlier two by way of backward cycle of fifths movement in bars
12 – 13 and 15 – 16.
Furthermore, the opening theme comprises oscillating thirds (as in Figure
9.3); with this musical material, Franck evokes the main theme of the first
movement as if he is starting the Sonata over again. This, in particular,
captures the essence of Beethoven’s Op. 102 No. 1.
Figure 9.3: Bars 12 – 13 of Franck’s Violin Sonata, iii.
Figure 9.4: Franck, Violin Sonata, III
Opening interpreted as if it were D minor

Ben Moderato
1–4 Ambiguous tonally, ends on D7 - I
This suggests that Franck was on the cusp of Progressive Tonality: willing to employ it
507
within a single movement, if not over the course of a larger, multi-movement work.
257
5 – 10 E-flat – G minor ♭II or IVβ – iv Violin solo, related to ‘A’ – ‘A2’
11 – 13 G minor – E minor iv – v/V Piano reminiscent of sonata opening
14 – 16 E minor – C-sharp minor v/V – vα As bar 11
17 – 21 C-sharp minor – dim vα
22 – 25 As bars 1 – 4, but A V
26 – 31 B-flat – D minor Iβ - i Violin solo, ‘A2’
32 – 33 D minor i Violin triplets
34 – 44 Transition – Violin (semiquavers) Bars 41 – 44 Piano reminiscent of
sonata opening, around G-sharp –
V/Vα)
45 – 47 Arrival on A V Related to ‘A2’
48 – 50 Arrival on B-flat Iβ Related to ‘A2’
51 – 52 Arrival on B IVα Related to ‘A2’
(Key change: F-sharp minor)
53 – 58 F-sharp minor New i
59 – 70 F-sharp minor I Violin new theme ‘E’ Bars 63 – 64
touches on Gm (Neapolitan); bar 65
theme minor third higher, but still F-
sharp minor based.
71 – 80 F-sharp minor – D minor – F minor i – iβ – vα ‘X’ theme
81 – 92 F-sharp minor i ‘E’
93 – 98 C-sharp major V Dominant pedal: Violin very similar
to ‘A bars 97 - 99
98 – 100 Piano transition
101 – 108 Transition ‘X’
109 - 117 F-sharp minor i Bar 111 related to 17 – G-sharp/G
natural
Here, Franck’s rhythmic manipulation is again evident, and the following
table illustrates this: six bar phrases are present in this movement in the
falling three-note phrase (Molto Lento from bars 17 – 21) and the extension
of the cyclic theme (bars 59 – 62, which recurs in the fourth movement). I
have indicated all irregular phrases in bold.
258
Figure 9.5: Franck, Violin Sonata, III, rhythmic analysis
Bars Phrasing (bars) Thematic/Harmonic Material (if significant)
1 – 16 Irregular (recit.) Violin largamente con fantasia
17 – 21 5 Molto Lento: Three-note falling theme.
22 – 25 4 Piano with opening material in new key
26 – 31 Irregular (recit.) Violin largamente con fantasia (‘A 2’)
32 – 40 3x3 Rising & falling triplets – semiquavers
41 – 44 4 Semiquavers, extended
45 – 50 2x3 Fantasia idea in violin, accompanied (‘A 2’)
51 – 52 2 Condensation of 45 – 47 phrase (‘A 2’)
53 – 58 6 Lead into tranquillo
59 – 64 4+2 Tranquillo theme (reused in finale) (‘E’)
65 – 70 4+2 As 59 – 64
71 – 80 8+2
81 – 92 Regular Tranquillo theme, then piano transition
93 – 100 Regular Based on 1st movement theme
101 – 8+2 As 71 – 80 (‘X’)
110 6+1 Molto lento theme
111 –
117
In addition to the phrase lengths noted above, in bar 11, the piano part is
related to the opening of the Sonata, but due to the different time
signature (4/4 instead of 9/8) there is a sense of stillness rather than flow.
The poignant molto lento section at bars 17 – 18 uses two chord pairs, and,
as mentioned previously, the alteration of the second chord of the second
pair gives the impression of a stronger sforzando. 508

Here, Franck
emphasises D-natural, the Neapolitan of C-sharp minor: this is the
approximate key at this point. Further Neapolitan emphasis in F-sharp
minor occurs around bars 63 – 64, where reach G minor (the Neapolitan,

259
which is also used between the Piano and the Violin in the opening); at bar
65 this theme occurs a minor third higher, but is still F-sharp minor based.
Figure 9.6: Franck, Violin Sonata, III, bars 65 – 68
Though the melody becomes more quadratic and relaxed from the
modulation into F-sharp minor, the accompaniment continues to fight this
rhythm with triplet accompaniment. Though this section feels relaxed, this
underpinning is still rhythmically unsettling. Franck has used the
Neapolitan in this way before, in both the Symphony (second movement
before the recapitulation) and the Quintet, and it can have the effect of
sounding like ♭VI of IV. Such harmonic ambiguity – albeit relaxed – is,
however, not a feature of the fourth movement.
Fourth movement
Davies describes this movement as a “canonical rondo” of “entirely new
design.”509 More specifically, its design is a type of sonata form, close to
Beethoven Op. 102/1 finale in its canonical aspects, with what we might
consider a written-out repeat of the exposition in Iα, C-sharp major. Also,
note the similarity between this theme and Beethoven’s Op. 59 No. 1/i.
Figure 9.7: Beethoven, String Quartet Op. 59 No. 1/1, bars 1 – 4

260
Figure 9.8: Franck, Violin Sonata, IV, bars 1 – 4
Davies explains that, “What serves to distinguish this music from so much
else that Franck wrote is its freedom from spiritual cant. Unburdened by
any kind of programme, its melodies ripple like the water from a
mountain spring.” 510 I suggest that the freedom Davies mentions is
synonymous with a freedom from quadratic structure. This is especially
evident in the falling four-note phrase (first occurring at bars 224 – 229)
which is phrased in six bar groups: unlike so much of Franck’s music, we
cannot sense where the resolution will be from the outset, yet this hardly
concerns us as the lyricism of the phrase is so stunningly beautiful.
Figure 9.9: Franck, Violin Sonata, IV

A Major
Allegretto poco mosso
1 – 37 A major I Main theme: ‘F’
38 – 45 A Major I ‘E’ in piano: touch on F-sharp minor (ivα) in
bar 42
46 – 51 Transition to C-sharp minor (iα) which arrives at bar 49: C major bar 47 (Vβ)
51 – 64 C-sharp minor iα ‘F’, violin leading
65 – F-sharp minor IVα ‘E’ in violin
73 - C major Vβ
80 – 98 E major V ‘F’, piano leading, developing falling motif
99 – 108 C-sharp minor iα ‘A1’: E pedal throughout harmony changes
109 – 116 Transition – chains of diminished chords in piano
510 Ibid., 232.

261
117 – 133 B-flat minor – major / E-flat minor – major ivβ - IVβ / v/Vα – V/Vα,
Developing ‘F’
133 – 142 on A-flat minor vα Piano develops part of ‘F’
(Key change: D-sharp minor)
143 – 148 D-sharp minor vα of V ‘X’
(Key change to B-flat minor)
149 – 161 B-flat minor ivβ Piano develops part of ‘F’
162 – 169 F minor iβ ‘X’
170 – 173 C major Vβ ‘E’ in violin
174 – 185 A minor i ‘E’ in violin: 184 – 185 shift to A major
186 – 222 A major I ‘F’
223 – 243 A major I Coda: falling four notes developed (as bars
80 – 98)
The approach to the recapitulation (using i) is particularly unusual, with
none of the dramatic vigour we might expect at this point. This oblique
arrival also has links to late Beethoven: examples include bars 159 – 160 in
the C-sharp major finale of Op. 131, and the recapitulation of the
Hammerklavier’s first movement. Here, Franck simply “falls” into the major
without using the dominant. The reprise of the main theme at bar 186 does
not constitute an arrival at all, as we have been hearing A major for the
previous two bars. However, we do get a significant arrival at bar 222, due
to Franck reversing the V9 – ii progression (perhaps more aptly described
as regression) set up in the first movement. Bars 211 – 212 use V and ii
respectively, but B minor then turns up abruptly on the second beat of bar
212. This allows Franck to place B minor in the next bar at a strong
metrical point – the third of a four bar phrase, which is then followed by
V6 in the fourth bar. After a little more rhythmic alteration, the tonic itself
finally arrives at a strong rhythmic point in bar 222, brought about by an
emphatic vi – ii – V7 – I progression. I illustrate the significance of Franck’s
262
rhythmic structure in the following table, with irregular phrase lengths in
bold type.
Figure 9.9: Franck, Violin Sonata, IV, rhythmic analysis
Bars Phrasing (bars) Thematic/Harmonic Material (if significant)

1 – 29 Regular (4) Main theme (canonic) (‘F’)
30 – 37 4 + 3 in Violin Cadence between 36 – 37
38 – 45 Regular B theme (‘E’)
46 – 51 4+2 Heading to C-sharp major
52 – 59 Regular Main theme (‘F’)
60 – 64 4+1 Cadence between 63 – 64 into F-sharp minor
65 – 72 Regular B theme (‘E’)
73 – 78 4+2 To E major
79 – 86 Regular Canonic (‘F’)
87 – 98 2x6 Falling theme, derived from main (‘F’)
99 – 102 Regular Theme in 3rds, related to 3rd movement (‘A1’)
103 – 108 4+2 As 99 – 102
109 – 116 Regular Chromatic transition – piano alone
117 – 132 Regular Based on main theme (‘F’)
133 – 138 2x3 Piano – main theme derivation in LH, A-flat minor (‘F’)
139 – 142 Regular Piano – as 133 – 138
143 – 150 Regular 3rd movement theme in Violin (‘X’)
151 – 156 2x3 As 133 – 138 but in B-flat minor (‘F’)
157 – 160 Regular As 139 – 142, but in F minor
161 – 168 Regular 3rd movement theme in Violin (‘X’)
169 – 184 Regular - - Some rhythmic crossover with Violin
185 – 212 Regular Main theme back in tonic major (‘F’)
213 – 220 Regular Cycle of fifths ‘righted’, emphatic V – I cadence
221 – 227 6 Falling theme, derived from main
228 – 235 7 Falling theme, derived from main
236 – 239 Regular Ending 4 bar phrase, cadence
240 – 242 Final Flourish
263
Of particular importance is the correlation between the falling theme (a
derivative of the main theme) and six-bar phrasing. This is a connection
with the six-bar phrasing evident in the third movement, which it shares
with Beethoven’s Op. 102 No. 1: both works feature a falling four-note
figure that coincides with six-bar phrasing, and in Franck’s Sonata this is a
connection to the upward four-note figure that ends the second
movement. One cannot attest to Franck’s rhythmic regularity here; in this
Sonata the music begins to shake free from Franck’s quadratic tendencies.
Franck and cyclic thematicism
Lawrence Davies asserts that the “habit of building entire works around a
‘motto’ theme gradually took possession of the composer, as it is alleged
to have done with Beethoven and Tschaikovsky, leading him to view the
Violin Sonata and the D minor Symphony in the same light.”511 It is easy to
hear the connections between themes in the Violin Sonata; Franck builds
the cyclic materials with the same bricks of recurring melodic intervals.
The intervals of a major and minor third occur frequently, for example,
often juxtaposed.
We have referred to the first and main theme (and first subject of the first
movement) as ‘A’. It is made of rising and falling thirds, outlining a
dominant major ninth, and feels rhythmically relaxed, being in 9/8 time.
The second subject (‘B’) falls by step then sighs. Continuing to the second
movement, the first theme here (‘C’) moves chromatically then up by
minor third. ‘C’ is partially derived from a melodic fragment of the first
movement (first movement, bars 9 – 12). Due to this link, the scherzo has a
feeling of parodying the first movement. Furthermore, the dotted crotchets
511 Ibid., 227.

264
followed by quavers – such as those in bar 20 – seem like an intensification
or parody of the flowing crotchet quaver rhythm of ‘A’. Also closely
related is the theme at bar 44 ‘A1’, outlining a D minor chord by first rising
a major third, similar to ‘A’s first occurrence in the first movement.
Moreover, the shape of the second movement second subject (‘D’) is
similar to the shape of ‘B.’ Figure 9.11 is provided for reference.
Figure 9.11: Franck, Violin Sonata, main themes
265
Franck builds the third movement from both new and recycled material.
The violin solo from bar four draws on minor and major third movement,
which I refer to as ‘A2’. Bar 11, featuring only the piano, is reminiscent of
the opening of the Sonata, but does not feel as relaxed: this is partially due
to the 4/4 rather than 9/8 rhythm. Two new themes occur in this
movement: ‘E’ at bar 59, a tranquil melodic idea, and ‘X’ at bar 71,
featuring a swooping fall of the major sixth. The final movement recycles
the most material. The main theme (‘F’) has a similar melodic contour to
‘A’ in its arch shape, and we can hear a relationship between it and ‘C’
with some filled-in passing notes. The fourth movement also features
recurrences of ‘E’, ‘A1’ and ‘X’. There are also cyclic connections that are
rhythmic rather than harmonic: Franck uses six bar phrases in both the
third and fourth movements for important melodic material, and the
significance of these uneven phrase lengths cannot be underestimated; the
rhythmic expansion and lack of predictability seems to be one of the
driving forces contributing to the success of this Sonata.
As I discussed in relation to the Quartet, in Chapter Seven, Franck was
particularly fond of the Beethovenian technique of directly recalling
previous movements at the opening of the fourth movement. Franck uses
this technique in the String Quartet, and elements of cyclic thematicism in
the Quintet and the Symphony. Although cyclic thematicism has much
compositional potential, Franck’s attempts at this borrowed technique can
often sound contrived, and this is one of the consistent faults I have found
with the works examined previously. In the Violin Sonata, however,
thematic recall is crafted more subtly: although the fourth movement
recalls themes from all movements, Franck integrates each theme into the
form. The sense of satisfying inevitability perhaps comes from the recall of
more than one event simultaneously; the second subject reappears in the
266
tonic key, but the second movement’s phrase also appears. The references
occur in reverse order: ‘E’ from the third movement is heard first, then A1
from the second movement (and based on the first subject of the first
movement), and the four note falling motive, which dominates the coda, is
very similar to the second subject from the first movement. More
importantly though, these are not mere quotes that remind us of the
previous themes; they are integrated into the fabric of the sonata, crafted
in new ways. The fourth movement is particularly effective in this way, as
Franck reminds us of what has been overcome tonally – the darkness of
the second movement’s D minor, and the third’s intense yearning is re-
framed in the “sunny” fourth movement, affirming A beyond doubt and
in an immensely lyrical manner.
This Sonata stands alone within Franck’s oeuvre. Harmonically, it is well
structured, and Franck’s gift of melody is at its best. More importantly,
though, one perceives a more mature handling of form; a lyric freedom
that is, as Davies asserts, free from Franck’s usual “spiritual cant” 512. It
provides very enjoyable listening indeed, and I believe that this analysis,
at least in part, validates the Sonata’s popularity.
512 Ibid., 232.

267
Chapter Ten
Conclusions
The previous analyses have demonstrated how axial tonality might inform ideas
about Franck’s compositional techniques. Franck was not a composer who found
one compositional method and continued with it for his career; he tried different
models – some of which were arguably more successful than others – without an
apparent clear chronological development. He shares this experimentation with
models with Brahms: despite varying compositional results, Brahms and Franck
also share an admiration of (potentially extending to a preoccupation with) late
Beethovenian form. What this lack of chronology in Franck’s use of form suggests
is: a) Franck did not remain content with any formal model he tried, possibly
because, b) Franck did not comprehend all the complexities of late Beethovenian
form as, indeed, few did.
Because of this mixed compositional method, different types of analyses have been
appropriate for different works, as we have discovered. Both axial-hexatonic and
pc set analyses have proved relevant as they reveal, respectively, different
elements in different compositions. Through this, both background and
foreground procedures can be illuminated.
All the analyses provide strong evidence that Franck was aware of the potential of
axial relationships and the way they might be used, specifically by incorporating
aspects of late-Beethovenian forms. This is evidenced in details as small as voice-
leading and as large as an over-arching key-scheme. In the discussion below I
reiterate the late-Beethovenian models, grouped with the Franck works that
268
identify with them, and reinforce the significant details in support of my
conclusions.
Beethovenian Models
1) The Eroica model: a “classical” tonicizing Sonata Form movement with an
immanent dominant and transcendent subdominant or axis equivalent (such as vi
in the Eroica). Though Franck does not seem to use the Eroica model in any of the
works examined here, it is important to recall the importance of this model as a
point of departure. Schubert aside, composers after Beethoven realised the
difficulties that ensued later in multi-movement works if such tonicizing
movements were used at the opening. They sought other models, and within these,
late Beethoven forms loom large.
2) The Op. 127 model: a “middle-heavy” work with a transcendent subdominant
used to reserve the dominant until later in the work (this differs from the proto-
progressive model in that the tonic is not initially undermined). Franck does not
seem to use this model either, though Franck’s String Quartet uses aspects of this
paradigm and that of Beethoven’s Hammerklavier in its use of a transcendent
subdominant in the third movement (as the second movement of Op. 127) and a
central fugal section in the first (Hammerklavier).
3) The mono-axial model: where one axis forms the basis for the entire work, such
as in Franck’s Quintet, Symphony and Symphonic Poem Le chasseur maudit. Some
of the smallest gestures here reveal Franck’s awareness of axial and related
harmonic relationships to evoke neutrality. Foreground examples include the
cadences in bar nine and eleven in the second movement of the Piano Quintet,
where E major is used as the dominant of both F and A minor, and the end of the
Quintet’s final movement with D-flat chords on the downbeat disturbing the effect
of tonicity F is (supposedly) meant to give.

269
Harmonic movement within the foreground of a work also belies Franck’s
awareness of the potential of the mono-axial model and the effect of different types
of harmonic movement. Le chasseur maudit demonstrates this in the use of minor
thirds to represent the hunter’s exertions within a larger framework of passive
major third axial modulation. This underlines the outcome Franck seems to infer is
inevitable.
Furthermore, the Symphony’s double exposition is one of the clearest examples in
the repertoire of an axially related key being set up to rival another for tonicity;
given the i – vβ relationship between D minor and F minor, we might describe this
as a case of the Eroica model in the first movement, perhaps redolent of
Beethoven’s fifth in its melancholic introduction. Though arguably simplistic in its
conception and effect, the exposition of the first movement of the Symphony
certainly succeeds in promoting opposition between the two axially-opposed keys.
The same might be said of processes within the exposition of Le chasseur maudit,
though this effect is actually different as Franck uses tonic axially related keys: G,
E-flat and B minor are I, Iβ and Iα respectively, rather than the Symphony’s
dominant-related D minor and F minor (i and vβ).
The background structure of the multi-movement works is also significant. The
Quintet’s overall key scheme (between movements) is F minor – A minor – F
minor. This in itself is an axially “flat” scheme, potentially suggesting that drama
will be created in other ways. However, the final movement emphasises other keys
(particularly D-flat and B-flat minor) to the extent that F’s tonicity is threatened,
suggesting that Franck may have been experimenting with harmonic effect on a
large scale, which we have discussed as proto-progressive tonality. Notably,
Franck’s only Symphony is also mono-axial: its movements are in D, B-flat and D.
270
4) The Op. 131 model: here the dominant is reserved for late in the work through
axial modulation prior to the final movement(s). This fits relatively well with
Franck’s String Quartet, though it is formally less clear-cut than many of the other
works discussed here. However, Franck’s use of F minor (Vβ) as the key of the
fugue in the middle of the Quartet’s first movement also echoes the D minor – F
minor relationship of the Symphony, and additionally may be linked to
Beethoven’s use of Fugue in the Hammerklavier. The employment of F-sharp minor
(iα) for the second movement belies further axial awareness, and though the use of
the relative minor is common for a slow movement, the key of B minor remains
almost unattainable in this movement, which itself predominantly uses
subdominant axial relatives, such as the Neapolitan (IVβ) at bar 103.
5) The proto-progressive model: here, the initial tonic is undermined from the
outset whilst another key is tonicised, before the original tonic is “rescued” late in
the work. Beethoven employs this model in Op. 101 and Op. 102 No. 1, wherein a
small opening movement in a weakly tonicised key is overwhelmed by a
disproportionate emphasis on the subdominant area, with the initial tonality only
“rescued” in the final movement. Franck’s Violin Sonata and the third sketch of
Psyché use this form, as does the Symphony (as a possible interpretation). The
Violin Sonata’s first movement, in A major, is preludial in effect and is the shortest
movement of the four, whilst the second movement is a dramatic scherzo strongly
tonicizing D minor (iv). The third movement begins with its key unclear (though it
suggests D minor) and then ends in F-sharp minor, whilst the finale tonicises A
major. The tonic is indeed “rescued” in the final movement, but not before the
subdominant axis has made its presence felt strongly in the second movement. The
lack of unequivocally functional tonality in the first movement is important, as it
allows a rival key to assert itself in a later movement.
271
Chromaticism
What makes Franck different from his contemporaries, specifically those
composing in the same genres? I suggest one difference is his rich chromaticism
and characteristic application of it. His chromatic idiom includes (but is not limited
to) the addition of sevenths and ninths to chords, which may or not act as
dominants and may or may not resolve; the use of German sixths and diminished
sevenths to modulate remotely in both the regular sense and also irregularly (such
as the way the German sixth is manipulated in the Piano Quintet); the use of
augmented triads as enhanced dominants; the use of the axial-hexatonic system as
both a foreground and background organizational principle; and organizational
principles that may not be strictly diatonic or even diatonically chromatic. This last
feature we see particularly in the Piano Quintet, which we might consider a
breakthrough in Franck’s use of a pc set: it is a “new way” compositionally, and
results in a uniform sonic world that prefigures the late works of Liszt.
Franck was unarguably chromatically innovative, being much more chromatic
than many of his contemporaries composing in similar genres, including Brahms,
Saint-Saëns, d’Indy and other pupils. At some moments, Franck’s harmony looks
forward to Impressionism and the dissolution of functional harmony, though this
is only ever momentary. However, Franck’s chromaticism is nevertheless linked to
many of the complaints and criticisms of his work. What my analysis has
demonstrated is that axial tonality is an organizational element in Franck’s work
and can help us understand many of the difficult elements in the works that have
been examined. Such axial analysis may help us to perceive stability where we
were previously unsure of it, but does not – and does not aim to – alleviate other
criticisms levelled at Franck’s chromaticism. For, whatever the organizational
principle, the most important element must always be the overall effect of a
composition, and if this is poor or has problematic elements then the initial
272
organizational principle has either been ill-applied or requires other parameters to
be manipulated for its success.
Take, for example, the Piano Quintet. The analysis using pc sets is appropriate – it
shows the organization of the work is coherent from background to foreground
levels. Franck’s use of a pc set cannot necessarily be considered new: Beethoven’s
Schrekensfanfare is paradigmatic; Franck’s Quintet and such compositions only
extend the concept of annihilating tonality. Via this analysis, Downes’ and Cohn’s
respective theories are proved compatible: Downes’ “passive modulation” is
Cohn’s “annihilation”, both theorists arguing for a consideration of a tonal
opposition which is not achieved.
Yet the coherence that pc sets lend is not necessarily enough to create an engaging
composition, and the result of the pc set use here – and its saturation – is the
creation of the aforementioned “sound” that we associate with Franck and his
harmonic procedures mentioned above. Liszt uses a pc set in La Lugubre Gondola:
yet this is a very short and programmatic work, about five minutes long, and is
formally incomparable to the extended sonata form – variations – modified sonata
form that moulds the Quintet. Thinking from an organist’s perspective, the same
might be said of Bach’s Chorale Preludes; though there are some that stand out for
their craftsmanship, they share a sonic world that incites coherence but not
compositional drama. Clearly, though, this comparison belongs to a different
compositional world.
However, if we can compare Franck to Reger – a fellow organist whose
compositions have motivated much debate and analysis about function and
harmony – it seems clear why Franck’s organ works fill more concert programmes
than those of Reger. Keeping Reger’s quip that “any chord can follow another
chord” in mind, we might surmise that any work crafted with such a mindset is
273
going to have few organizational principles to adhere to, particularly harmonic
ones. Comparing this to Franck, who has a solid soundworld even if it may
become tiresome after three movements, the elements that contribute to coherence
become clear: harmony does work in this way.
Tackling late Beethoven and the Hammerklavier
This thesis references the Hammerklavier Sonata a great deal. That this work proved
influential to composers in the Second Age of the Symphony is evidenced in the
Hammerklavier quotations detailed in the Piano Quintet chapter. This affect is
reinforced by other Beethoven quotations: those of the Lebewohl Sonata and the Op.
130 String Quartet (referenced in the String Quartet chapter). Though not
Symphonic, the Hammerklavier seems to represent the problem of the finale, which
may be one explanation for the ongoing curiosity composers have with it.
However, whether Franck, Mendelssohn, the young Brahms or anyone else
actually perceived all of the compositional processes inherent in the Hammerklavier
is an issue for debate. I suggest that though many composers may have sensed that
there was something “different” or even transcendent about the Hammerklavier,
few grasped the compositional processes sufficiently to emulate it successfully.
One might argue that Brahms did – indeed, by the time Brahms had reached
compositional maturity, his large-scale works were able to articulate a dominant
withheld and reserved for a final movement, which might account for much of the
success of, say, his third Symphony. A mono-axial end-weighted shape is by no
means the only way to handle large scale form, but it is an effective device if a
teleological feel is desired throughout a large-scale work. Comparing Brahms’
third to Franck’s Symphony, we are reminded of Dahlhaus’ assessment of the
String Quartet: the problem of making problems “manifest” rather than
“concealing them in artifices”. 513 Problematization may be a genuine form
274
generator, evidenced in the double exposition of Franck’s Symphony: however,
one cannot help but perceive the problem within the double exposition, and so
perhaps Franck’s difficulty indeed lay in the crafting of compositional artifice, as
Dahlhaus suggests.
Franck’s Chromaticism in Form
When Liszt developed the Symphonic Poem, he quipped that “new wine demands
new bottles”.514 Though he was referring to the title “Symphonic Poem”, I believe
the quote can also be interpreted in this context, relating chromaticism to form. In
the classical style, tonality articulated form (read sonata form). If that tonality is
challenged by the density and richness of chromaticism, then the form will likely
need to be altered, even if diatonic principles at large are adhered to. I suggest that
this is one of the problems we sense in Franck’s work but find difficult to
articulate: he is classical in genre but Wagnerian in harmony. How can the two co-
exist? Recalling the Dahlhaus contention alluded to above, the assessment may
seem harsh, but it succinctly expresses the problem we perceive in much of
Franck’s music; a seeming incompatibility of form and harmony. However, this
may also be linked to large-scale axial organization: many seminal nineteenth
century works use axial relations to articulate a broader structural coherence, such
as Liszt’s B minor Sonata or Brahms’ Third Symphony. However, whilst this is true
of some of Franck’s works, others do not share such tonal fissures and therefore do
not cohere to a wider organisational principle. Such fissures might include the role
of E-flat at the start of the Quartet, and similarly the E-flat in the introduction of
the Quintet; though E-flat is expounded as a key area at the beginning of the
respective works – creating a tonal fissure in both cases – the fissure does not then
play out functionally.
514 Alan Walker, et al, "Liszt, Franz," (Grove Music Online. Oxford Music Online.).
275
I do not suggest, however, that Franck was unaware of such issues. I suggest that
part of Franck’s use of cyclicism (as in the recalling of early themes late in a work,
not the subtler integration closer to thematic transformation) was an attempt to
add an additional element of integration to a composition, the coherence of which
could be compromised by the density of its chromaticism and the formal-harmonic
relationship. The Piano Quintet, String Quartet and Symphony all display elements
of cyclicism; they also possess some of the richest chromaticism in the canon. It
seems likely that Franck desired a deeper sense of coherence, and cyclicism in a
Franckian sense is an easy enough device to apply. However, as Dahlhaus states,
“to prevent the recurrence of a theme from sounding like an interpolated quotation
it must be integrated into the musical context.”515 Here, it is apt to return to the
discussion from the first chapter concerning the differences between cyclicism and
thematic transformation. “Cyclic form” might be the better term applied here,
which we earlier described as applying to: “works where thematic links bind more
than one movement” rather than “applied to mere thematic resemblances.”516 This
description is very similar to Taylor’s designation of “cyclicism” as “close thematic
affinity between movements or, more properly, explicit recall of music from one in
another.”517 If we look for aspects of “thematic recall” in Franck’s works, then, the
term “cyclic form” can certainly apply to the Piano Quintet, the Symphony and the
String Quartet. The last example here is particularly overt in its Beethovenian
thematic recall, the problematic use of which I discussed in Chapter One, further
evidencing both the influence Beethoven and his late forms had on nineteenth
century composers and the difficulties those composers had in fully understanding
Beethoven’s formal methods.

516 Walter Frisch, "The Snake Bites Its Tail: Cyclic Processes in Brahms's Third String Quartet,
Op. 67," The Journal of Musicology 22, no. 1 (2005): 155.
517 Benedict Taylor, "Cyclic Forms in the Instrumental Music of Felix Mendelssohn: Time,
Memory and Musical History" (PhD Diss., University of Cambridge, 2006), 9.

276
The “Thematic Transformation” definition we used earlier describes it as: “a term
used to define the process of modifying a theme so that in a new context it is
different but yet manifestly made of the same elements.” 518 Those Franckian works
which use thematic transformation, are, I suggest, more successful: these are the
Violin Sonata and Psyché. The effects of such transformations are discussed in their
respective chapters; however it is worth recapping why such effects might be
considered more successful. The understanding of music as an experience of a
teleological process became prominent in the nineteenth century, specifically, with
the works of Beethoven. It is apt to quote Scott Burnham here:

[There was an] important shift in early nineteenth-century music criticism to the
standpoint of the whole, a shift arguably brought about as a response to
[Beethoven’s Eroica and other heroic works]. In the case of Beethoven, repeated
hearings were now deemed necessary for a complete understanding of any of his
works. Making sense of the local complexities of this music depended on knowing
the entire process, for an inherent destiny is projected back onto the idiosyncrasies
of Beethoven’s themes by the way they come to be consummated in the coda. The
music is heard to be about thematic process and development; the full
understanding of a theme waits upon a knowledge of its eventual outcome.519
Though Burnham later challenges the notion of perceiving music as a process (or
in terms of progress), this mode of thought is very much present today, and
certainly would have been within Franck’s lifetime. Therefore, I suggest that it is
more satisfying to recognise the reminiscence or relative of a theme, which we can
recognise as having progressed and changed, than identifying a simple recall.
Transformation denotes progress, having reached a time or place different from
where the music started, and notions of progress are positively associated with
large-scale form. In Psyché, this transformation is associated with the
transformation of Psyche’s being into a goddess, and is wholly appropriate in the
Hugh Macdonald, "Transformation, Thematic," in Grove Music Online. Oxford Music Online.
518
Scott Burnham, Beethoven Hero (Princeton, New Jersey: Princeton University Press, 1995),
519
122.
277
programmatic context. Examining the context of the initial theme and that of the
fourth sketch, it is also linked to Psyché’s sleep and awakening, both in a physical
and mythological sense. In terms of progress, cyclic form simply returns to or
reminds us of a theme’s initial musical state, without a sense of progression.
However, the most important aspect to consider in the success of the Violin Sonata
(and, conversely, many of the formal-harmonic problems with Franck’s other
works examined here) is the consideration of other altered parameters. Franck’s
chromaticism seems incompatible with the form of many works, which – as I have
suggested – may be part of his reason for using cyclic form. When examining the
Quintet, I suggested that if such rich chromaticism is to be successful, then other
parameters such as form, rhythm and phrase structure must be altered to give a
sense of formal unity as appropriate. The Violin Sonata is one of the few Franck
works examined here which pairs the attainment of the structural dominant with
an altered rhythmic emphasis in the final movement. This throws musical weight
onto a significant structural moment in a way that rarely occurs in Franck’s work,
and I suggest that this may contribute to the Violin Sonata’s ongoing popularity
and success. This rhythmic manipulation is particularly significant given that
criticisms levelled at much of Franck’s output pertain to his “four-square”
approach to rhythm. To the uninitiated, it might be difficult to believe that the
same composer who wrote these quadratic themes in the Symphony, Quintet and
Psyché –
278
Figure 10.1: Franck, thematic examples
– also composed the lilting 9/8 of the Violin Sonata’s opening movement, the
recitative-like yearning in its third, and this manipulation of emphasis in its final
movement. This is not to say that other composers did not write quadratic phrases
when it was appropriate to do so; of course, such phrases are the building blocks of
classical music, in particular the Viennese School. Beethoven’s late period Quartets
provide examples where visionary quadratic phrases are used, but with a sense
that they represent a utopian view that cannot – will not – be attained; hence such
phrases usually disintegrate, such as the end of the Heiliger Dankgesang of the A
Minor Quartet Op. 132. 520 In Franck, however, such phrases are the utopia: the
melodic goals, the main material. As Kerman writes, regarding the Op. 127 Quartet
in E-flat:
The folk-like tone [including quadratic phrasing] is so magical and true, so lively
and calm, that one feels solemn to talk about subtleties of construction . . . But folk
279
accents can sound banal just as easily as enchanting. What sets and assures the tone
is the way the musical elements are put together.521
Dahlhaus realises the problematic relationship between periodic structure and
harmonic enrichment, and states: “the procedure is self-defeating as long as the
technical and aesthetic criterion is the rule that development of all the elements of
the composition should be analogous.” 522 Franck’s rhythm and phrase structure
does not seem as developed as harmony and cyclic form in the Symphony and
String Quartet. If this is the case, we must examine works of other composers to
deduce more successful methods.
If we were to compare Franck to a Teutonic composer who composed in similar
genres to Franck but with a different sense of rhythmic organization – Brahms – we
would observe that although Brahms’ harmonic palate is rarely as chromatic as
Franck’s (though chromatic nonetheless), Brahms possesses a strong sense of the
importance of rhythmic organization, particularly polymeter. Conversely, in
Franck, because of the consistent quadratic structure, we sense when the music is
going to resolve, even if we do not know how. Therefore, despite Franck’s
innovative chromatic idiom, his grasp of innovating phrase structure is usually
weak. Dahlhaus writes that Franck “emphasises rhythmic foursquareness”, and
this is particularly evident in the Symphony and Piano Quintet. I support
Dahlhaus’ view that, ideally, there should be a correlation between rhythmic
structure and chromaticism. However, Franck’s success in this area is illustrated by
the complementary harmonic and rhythmic analyses in the chapter on the Violin
Sonata. It is also important to note that different genres require different handling
of musical material. What might be appropriate for a Symphonic Poem may not be
521 Ibid., 234.

280
for a chamber work or even a symphony. This, I believe, can account for some of
Franck’s rhythmic foursquareness, or dense chromaticism, but not all of it.
The analysis now complete, I wish to address the semantics of Cohn and Downes’
“annihilation” and “tonal passivity” respectively. Evidently, both commentators
are addressing the same phenomenon: the use of axial relatives to prolong a
harmonic area, whilst retaining tonal flatness. Yet, we must note that
“annihilation” possesses destructive, negative overtones, whilst “tonal passivity”
is a more neutral phrase. I suggest that both terms are useful in future axial
analysis. “Annihilation” may be appropriately used to describe instances like the
Schrekensfanfare, where – through superimposed chords – tonality (and all that
has gone before it in the Symphony) is indeed “annihilated”. In contrast, “tonal
passivity” may describe instances where tonic or subdominant axial relatives
prolong a key area for an expressive purpose, such as Beethoven’s use of the tonic
axis in Op. 127 or many of Schubert’s slow movements. A lack of tonal action can
be expressive; negative overtones are not necessary in such cases.
Franck did compose some works of great beauty, but other works are dogged with
problems that he did not, and I would argue could not, resolve. Axial tonality
offered additional means of coherence whilst maintaining a chromatic idiom, but
the compositional environment was a difficult one and, reception history would
prove, not enough. Examining the problems and utopias of this era is a theoretical
minefield, one which I predict to be active for some time to come.
281
Bibliography
Abrams, M. H. A Glossary of Literary Terms. 3rd ed: Holt, Rinehart and
Winston, Inc., 1971.
Bribitzer-Stull, Matthew. "The A♭-C-E Complex: The Origin and Function of
Chromatic Major Third Collections in Nineteenth-Century Music." Music
Theory Spectrum 28, no. 2 (2006): 167 - 90.
———. "The End of Die Feen and Wagner's Beginnings: Multiple Approaches
to an Early Example of Double-Tonic Complex, Associative Theme and
Wagnerian Form." Music Analysis 25, no. 3 (2006): 315 - 40.
Briscoe, James R. "Debussy, Franck, and the 'Idea of Sacrifice'." Revue belge de
Musicologie / Belgisch Tijdschrift voor Muziekwetenschap 45, César Franck et son
temps (1991): 27 - 39.
Brown, A. Peter. The Symphonic Repertoire: The First Golden Age of the Viennese
Symphony: Haydn, Mozart, Beethoven and Schubert Vol. II. Bloomington &
Indianapolis: Indiana University Press, 2002.
Brown, A. Peter, and Brian Hart. The Symphonic Repertoire Volume III Part B:
The European Symphony from ca. 1800 to ca. 1930: Great Britain, Russia, and
France. Bloomington & Indianapolis: Indiana University Press, 2008.
Burnham, Scott. Beethoven Hero. Princeton, New Jersey: Princeton University
Press, 1995.
282
Caldwell, John, et al. "Keyboard Music." In Oxford Music Online. [Accessed
June 2009.]
Childs, Adrian P. "Moving Beyond Neo-Riemannian Triads: Exploring a
Transformational Model for Seventh Chords." Journal of Music Theory 42, no. 2
(1998): 181 - 93.
Cohn, Richard. "As Wonderful as Star Clusters: Instruments for Gazing at
Tonality in Schubert." Nineteenth Century Music 22, no. 3 (1999): 213 - 32.
———. "Introduction to Neo-Riemannian Theory: A Survey and a Historical
Perspective." Journal of Music Theory 42, no. 2 (1998): 167 - 80.
———. "Maximally Smooth Cycles, Hexatonic Systems, and the Analysis of
Late-Romantic Triadic Progressions." Music Analysis 15, no. 1 (1996): 9 - 40.
———. "Weitzmann's Regions, My Cycles, and Douthett's Dancing Cubes."
Music Theory Spectrum 22, no. 1 (2000): 89 - 103.
Cook, Robert Cameron. "Parsimony and Extravagance." Journal of Music
Theory 49, no. 1 (2008): 109 - 40.
———. "Transformational Approaches to Romantic Harmony and the Late
Works of César Franck." Diss., University of Chicago, 2001.
Cooper, Martin. French Music: From the Death of Berlioz to the Death of Fauré.
Oxford: University Press, 1951.
283
Cortot, Alfred. French Piano Music. Translated by Hilda Andrews. London:
Oxford University Press, 1932.
D'Indy, Vincent. César Franck: A Translation from the French of Vincent d'Indy
with an Introduction by Rosa Newmarch. London: John Lane the Bodley Head,
1909.
———. "La Première Manière De César Franck." Revue belge de Musicologie /
Belgisch Tijdschrift voor Muziekwetenschap 4, no. 5 (1923): 2 - 7.
Dahlhaus, Carl. Between Romanticism and Modernism: Four Studies in the Music
of the Later Nineteenth Century. Berkeley: University of California Press, 1980.
———. Nineteenth Century Music. Berkley: University of California, 1989.
Davies, Laurence. César Franck and His Circle. Boston: Houghton Mifflin Co.,
1970.
———. Paths to Modern Music. London: Barrie and Jenkins, 1971.
Demuth, Norman. César Franck. London: Dennis Dobson Ltd., 1949.
Douglass, Fenner. Cavaillé-Coll and the French Romantic Tradition. New Haven
and London: Yale University Press, 1999.
Downes, Graeme. "An Axial System of Tonality Applied to Progressive
Tonality in the Works of Gustav Mahler and Nineteenth Century
Antecedents." Diss., University of Otago, 1994.
Eisen, Cliff et al. "Mozart." In Oxford Music Online. [Accessed May 2010.]
284
Fanning, E.M. "The 19th Century French Organ of Cavaillé-Coll and Organ
Works of César Franck." Diss., Boston University, 1964.
Fauquet, Joël Marie. César Franck. Paris: Fayard, 1999.
Forte, Allen. "Pitch-Class Set Analysis Today." Music Analysis 4, no. 1/2 (1984):
29 - 58.
Frisch, Walter. Brahms and the Principle of Developing Variation. Berkeley and
Los Angeles: University of California Press, 1984.
———. "The Snake Bites Its Tail: Cyclic Processes in Brahms's Third String
Quartet, Op. 67." The Journal of Musicology 22, no. 1 (2005): 154 - 72.
Gallois, Jean. "Chausson, Ernest." In Oxford Music Online. [Accessed July
2009.]
Grace, Harvey. "Franck, César Auguste." In Grove's Dictionary of Music and
Musicians, edited by H. C. Colles. New York: The MacMillan Company, 1927.
———. The Organ Works of César Franck. London: Novello and Company Ltd.,
1948.
Greene, David B. Mahler, Consciousness and Temporality. New York: Gordon
and Breach, 1984.
Hamilton, Edith. "Mythology Summary and Analysis."
www.gradesaver.com/mythology/study-guide/section6. [Accessed May
2010.]
285
Harrison, Daniel. Harmonic Function in Chromatic Music. Chicago: University
of Chicago Press, 1994.
———. "Max Reger Introduces Atonal Expressionism." The Musical Quarterly
87, no. 4 (2005): 660 - 80.
———. "Supplement to the Theory of Augmented-Sixth Chords." Music
Theory Spectrum 17, no. 2 (1995): 170 - 95.
Hedley, Arthur. "Liszt the Pianist and Teacher." In Franz Liszt: The Man and
His Music, edited by Alan Walker. London: Barrie & Jenkins 1970.
Hoeckner, Berthold. Programming the Absolute: Nineteenth-Century German
Music and the Hermeneutics of the Moment. Princeton: Princeton University
Press, 2002.
Hyer, Brian. "Reimag(in)Ing Riemann." Journal of Music Theory 39, no. 1
(1995): 101 – 38.
Hyer, Brian and Alexander Rehding. "Riemann, Hugo." In Oxford Music
Online. [Accessed July 2010.]
Jersild, Jörgen. The Harmonics of Romanticism: Functional Principles Based on the
Harmonic Style of César Franck. Translated by Liz Lammers. London: J. & W.
Chester Ltd., 1970.
Kerman, Joseph. The Beethoven Quartets. Oxford: Oxford University Press,
1966.
286
Kopp, David. Chromatic Transformations in Nineteenth-Century Music.
Cambridge: University Press, 2002.
Latham, Peter. Brahms, Master Musicians Series. London: Dent, 1975.
Lewin, David. "Behind the Beyond: A Response to Edward T. Cone."
Perspectives of New Music 7, no. 2 (1969): 59 - 69.
Lewis, Christopher Orlo. Tonal Coherence in Mahler's Ninth Symphony. Ann
Arbor, Michigan: UMI Research Press, 1984.
Lockwood, Lewis. "Beethoven's Emergence from Crisis: The Cello Sonatas of
Op. 102 (1815)." The Journal of Musicology 16, no. 3 (1998): 301 - 22.
Macdonald, Hugh. "Idée Fixe" In Oxford Music Online. [Accessed February
2011.]
———. "Transformation, Thematic." In Oxford Music Online. [Accessed
February 2011.]
McCreless, Patrick Phillip. "Wagner's Siegfried: Its Drama, History, and
Music." Ann Abor, Michigan: UMI Research Press, an imprint of University
Microfilms International, 1982.
Monahan, Seth. "Inescapable" Coherence and the Failure of the Novel-
Symphony in the Finale of Mahler's Sixth." 19th-Century Music 31, no. 1 (2007):
53 – 95.
287
Newlin, Dika. Bruckner - Mahler – Schoenberg. London: Marion Boyars
Publishers Ltd., 1978.
The New Oxford Dictionary, edited by Judy Perasall. Oxford: Clarendon Press,
1998.
Platinga, Leon. Romantic Music. A History of Musical Style in Nineteenth-Century
Europe. New York: W.W. Norton and Company, 1984.
Rosen, Charles. The Classical Style. London: Faber and Faber Ltd., 2005.
Rothfarb, Lee A. Ernst Kurth: Selected Writings. Cambridge: Studies in Music
Theory and Analysis. Cambridge: Cambridge University Press, 1991.
Sachs, Harvey. The Ninth: Beethoven and the World in 1824. New York: Random
House, 2010.
Salzer, Felix. Structural Hearing: Tonal Coherence in Music. New York: Dover
Publications, 1952.
Schoenberg, Arnold. Style and Idea. London: Lowe and Brydone, 1951.
———. Theory of Harmony. Translated by Roy E. Carter. Berkeley, Los
Angeles: University of California Press, 1978.
Smith, Rollin. Playing the Organ Works of César Franck. New York: Pendragon
Press, 1997.
288
———. Toward an Authentic Interpretation of the Organ Works of César Franck.
New York: Pendragon Press, 1983.
Snarrenberg, Robert. "Schenker, Heinrich." In Oxford Music Online. [Accessed
November 2008.]
Stewart-MacDonald, Rohan H. New Perspectives on the Keyboard Sonatas of
Muzio Clementi. Bologna, Italy: Ut Orpheus Edizioni, 2006.
Strasser, Michael. "Art as Teaching: Vincent d’Indy and Programming at the
Société Nationale after 1886 " In Vincent d'Indy et Son Temps. 135 – 148
Sprimont: Mardaga, 2006.
———. "The Société Nationale and Its Adversaries: The Musical Politics of
'L'invasion Germanique' in the 1870's." 19th-Century Music 24, no. 3 (2001):
225 - 51.
Talbot, Michael. The Finale in Western Instrumental Music. New York: Oxford
University Press, 2001.
Taylor, Benedict. "Cyclic Forms in the Instrumental Music of Felix
Mendelssohn: Time, Memory and Musical History." Diss., University of
Cambridge, 2006.
Tovey, Donald Francis. A Companion to Beethoven’s Pianoforte Sonatas. London:
Oxford University Press, 1931.
———. Essays in Musical Analysis Volume II: Symphonies. London: Oxford
University Press, 1966.

289
———. Essays in Musical Analysis. Volume I: Symphonies. London: Oxford
University Press 1968.
Trevitt, John, and Joël Marie Fauquet. "César Franck." In Oxford Music Online.
[Accessed March 2008.]
Vallas, Léon. César Franck. George G. Harrap & Co. Ltd., 1951.
Walker, Alan, et al. "Liszt, Franz." In Oxford Music Online. [Accessed March
2008.]
Williams, Adrian, ed. Franz Liszt: Selected Letters. Oxford: Oxford University
Press, 1998.
Winter, Robert et al. "Schubert, Franz." In Oxford Music Online. [Accessed
March 2008.]
290

Tesis Sobre Franck

Uploaded by

Copyright:

Available Formats

Tesis Sobre Franck

Uploaded by

Document Information

Original Description:

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

Tesis Sobre Franck

Uploaded by

Copyright:

Available Formats

TONALITY, FUNCTIONALITY AND

BEETHOVENIAN FORM IN THE LATE

INSTRUMENTAL WORKS OF CÉSAR FRANCK

Rachel Mary Swindells

A thesis submitted for the degree of Doctor of Philosophy at

the University of Otago, Dunedin, New Zealand

After Beethoven’s compositional achievements, particularly those of his late

period, nineteenth-century composers drew on his ideas in many different ways.

The harmonic language of music darkened considerably in the nineteenth

century, with strong chromaticism becoming prevalent and Wagner’s influence

century compositions were harmonically organised was with axial tonality, a

system which privileges major-third relationships due to their functional

equivalence. Composers from Beethoven onward were familiar with such

organisational principles, and the instrumental works of César Franck make an

illuminating case study in how axial tonality operates in a deeply chromatic

functional principles of tonality despite an enriched harmonic language, and what

changes were necessary to formal models as a result. Seminal late-Beethovenian

relation to immanent and transcendent dominants which operate differently in

harmonic language and formal procedures. Though some of Franck’s works

importance of rhythm, and cyclic form, a technique closely associated with

support and provision of an excellent work environment by way of the Elman

Poole Fellowship, which I was fortunate to receive during 2009 – 2010.

translations, and Margaret Finney for proof-reading.

staff and students, particularly Dr. Alan Davison.

my mother) taught me to love music.

2 Historical Documentation and Theory of Third-Relatives in Music 34

3 The Eroica Symphony and the Problem of the Finale 73

4 Late Beethovenian Models 97

5 Issues of Nineteenth Century Composition, Le chasseur maudit and Psyché 125

6 The Mono-Axial model and the Quintet in F Minor 158

7 String Quartet 191

8 Symphony in D Minor 214

9 Proto-Progressive Tonality and the Violin Sonata 243

Figure 2.1 Hauptmann’s diatonic key-schema

Figure 2.2 Riemann’s early Tonnetz

Figure 2.3 Downes’ Major Third Axis System

Figure 2.4 Downes’ dominant/subdominant functions

Figure 2.5 Dominant voice-leading efficiency (adapted from Downes)

Figure 2.6 Subdominant voice-leading efficiency (adapted from Downes)

Figure 2.7 Supertonic voice-leading efficiency (adapted from Downes)

Figure 2.8 Cohn’s “Table of Tonal Relations”

Figure 2.9 Cohn’s hyper-hexatonic system

Riemann: root-interval system (from Kopp)

Figure 2.11 Harrison – disassembly of a mediant

Figure 2.12 Harrison – disassembly of a diminished seventh chord

Figure 2.13 Pitch-class set [4, 5, 8, 9, 12, 13]

Figure 3.1 Beethoven, Symphony No. 3/iii (Op. 55), bars 1 – 7

Figure 3.3 Schubert. Symphony No. 9 (D 944). Adapted from Downes

Figure 4.1 Beethoven, String Quartet (Op. 127), I.

Figure 4.2 Beethoven, String Quartet (Op. 127), I, bars 7 – 14.

Figure 4.4 Beethoven, String Quartet (Op. 127), III.

Figure 4.5 Beethoven, String Quartet (Op. 127), IV.

Figure 4.6 Beethoven, Cello Sonata (Op. 102), Part I

Figure 4.7 Beethoven, Cello Sonata (Op. 102), Part II

Figure 5.1 Brahms, Symphony No. 2/iii (Op. 73), bars 1 – 14

Figure 5.2 Franck, Le chasseur maudit

Figure 5.3 Franck, Sommeil De Psyché (Sketch 1)

Figure 5.4 Franck, Psyché Enlevée par les Zéphirs (Sketch 2)