b. 5

Double bar-line in GC

Single bar-line in FE (→EE1)

Repeat sign in GE & EE2

For many years, the issue whether to include Grave at the repetition of the exposition has been a contro­ver­sial subject among musi­col­ogists, pianists and editors. The issue was raised twice by Charles Rosen, who firmly believed the introduction should be re­peat­ed1. Anatole Leikin tried to justify a different view in an article dedicated to this issue2. Franco Luigi Viero published polemics with Leikin's argu­ments, partially also with the Rosen's ones, on the Internet3.

Source arguments

  1. A repeat sign in bar 5 appears only in GE and – probably on its basis – in EE2. Therefore, it is absent both in GC and in FE1, which proves that it was also absent in [A], as both sources were based directly on it. The sign is also absent in FE3, FE4 and EE1, in which preparation Chopin must have participated. Finally, it is not to be found in any of the three preserved pupil's copies.
    Although a possible influence of Chopin on the text of GE1 cannot be entirely excluded, yet it is highly unlikely (see the characterization of GE1). It means that Chopin did not write a repeat sign in this place.
  2. One can ponder whether Chopin could have forgotten to insert a repeat sign or even considered it to be obvious. It is the main line of Leikin's source argumentation, who claims that excluding introduction from the exposition's repetition was a common norm at that time if the introduction was maintained in another (slow) tempo. As a proof that such a sign "was generally considered to be superfluous," he quotes examples of four sonatas by Dussek, one by Hummel and two works by Beethoven4, in which there is no sign starting a repeat after a slow introduction. The argumentation surprises with its inappropriateness – in all examples the introduction is extensive, comprising at least a dozen bars, hence their possible inclusion at the repetition of the exposition would have distorted the proportions of particular fragments to an extent excluding the classification of the whole as a sonata form (by the way, in five out of seven examples the end of the exposition does not even combine with the beginning of the introduction due to rhythmic or textural incompatibilities). Moreover, in each of the examples the beginning of the following allegro is clearly graphically separated, so that there are no doubts that it is a beginning of a new section, hence not requiring a sign opening the repetition5. All of the above is by no means similar to the context of the four-bar-long motto of Chopin's Sonata – the only certain element separating the introduction from the following section is a new tempo indication (the authenticity of the double bar line is uncertain), the beginning of Grave perfectly combines with the end of the exposition, and the inclusion of the introduction at the repetition does not violate the proportions of a sonata form at all. Therefore, there is no basis to consider the omission of Grave at the repetition of the exposition to be comprehensible in itself and not requiring an indication, which de facto is not there.
  3. All that remains is to determine whether in Chopin's autograph bars 4-5 were separated with a single or double bar line. According to us, an arbitrary addition of a double bar line by the copyist is more likely than ignoring such a bar by the engraver of FE. It also cannot be entirely excluded that Chopin himself added the second line in GC (Cf. II mvt, bars 36-37). As the issue is practically of no great significance – a double bar line does not influence the performance; it only graphically divides the introduction, which is already separated rhythmically and texturally – in the main text we adopt a single bar line as the version with the highest probability of authenticity.
  4. Another confirmation of the correctness of the text of FE – a single bar line – is a slur of the L.H., added at the transition between bars 4-5 in the proofreading of FE3. It means that Chopin, when looking at this place, did not feel the need of a clearer separation of the introduction with a double bar line, just the opposite, he pointed to a necessity of its closer relation to the beginning of the quaver movement in bar 5, suggesting that the figuration results from the preceding Grave. In turn, if Chopin had wanted to omit Grave at the repetition, he would have most probably noticed the absence of a repeat sign, necessary in this situation6.

Stylistic arguments

  1. Harmony.
    While analysing the fixed editorial and performance tradition, copying the repeat sign after GE, Rosen writes: "the erroneous sign [of the repetition] is musically impossible, as it breaks the triumphal cadenza in D major with an entrance of an accompanying figure in B minor [...]." Moreover, Viero indicates a harmonic ambiguity, asking "which degree is the B minor, with which bar 5 starts, [if we exclude Grave], tonic, i.e. the 1st degree, or the 6th?". Both the incomplete cadenza in the ending of the exposition and the aforementioned ambiguity concerning the function of the chord opening its return are something absolutely uncommon at the transition between the ending and beginning of an exposition in a sonata form. Generally, in such transitions tonal and function relations are very clear – the exposition always ends with a distinct cadenza (these are frequently a few cadenzas), after which, however, a short episode may follow, leading to the beginning of the repeated section. It is proved by, e.g., all piano sonatas by Beethoven, Schubert, Weber, Kalkbrenner, Schumann and others. Therefore, the omission of Grave at the repetition of the exposition leads to a significant disfunction of the harmonic course, which is inconceivable in the case of Chopin.
  2. Form.
    When discussing different types of introductions preceding the first theme of a sonata form, Leikin divides them into two general types:
    – introductions maintained in the same tempo and time signature as the main section of the allegro;
    – introductions with a different tempo and independent time signature.
    He then proves that the first are generally included at the repetition of the exposition, whereas the second are not, which, according to him, excludes the possibility of repeating Grave. Although a different manner of approaching both aforementioned types of introductions seems to be well-documented, the conclusion concerning Chopin's Sonata is erroneous, as Grave belongs to the first and not to the second type. The doppio movimento indication does not change the actual tempo of the music course; it only signalises a change of the notation – from this place on the minims equal the previous crotchets ( = ). What is more, imposing such a close time signature and rhythmic relation of the introduction and the following quaver movement makes audibly perceptible sense only in the context of a planned return of Grave after the final chords of the exposition – at the beginning of the piece, the relation, even when observed (which, after all, almost does not happen in actual performances), is practically imperceptible7.

The analysis of both the harmonic course and formal aspects leads to a conclusion that without repeating Grave the exposition is structurally incomplete, as only the four-bar-long phrase creates both its ending and a transition to the initial theme in B minor. Therefore, Grave plays a double role: it is a dramatic introduction and an episode closing the exposition, which could be entirely written as its ending prima volta8Rosen sees "an excellent concept" here, whereas Leikin does not want to see it at all, considering Grave to be only an introduction.


1 Charles Rosen, The First Movement of Chopin's Sonata in B-flat Minor, Op. 35, "Nineteenth-Century Music" 14 (1990), pp. 60-66 and The Romantic Generation (Cambridge, Harvard University Press, 1995), pp. 279-283.
2 Anatole Leikin, Repeat with Caution: A Dilemma of the First Movement of Chopin's Sonata op. 35, "The Musical Quarterly" 85(3), Fall 2001, pp. 568-582.
3 Franco Luigi Viero, Can a Historical Context Make up for an Ear Problem?, XI 2011 – accessed 27-11-2018.
4 Ladislav Dussek, Op. 13 No. 3, Op. 25 No. 2, Op. 45 No. 2, Op. 61; Johann Nepomuk Hummel, Op. 38; Ludwig van Beethoven, Op. 74 and 81a.
5 In six cases in which it was possible to check the first edition on-line (except for Hummel's Sonata), there are at least four out of the following elements separating the beginning of an allegro: time signature (changed or repeated), new tempo indication, double bar line, key signature (changed or repeated), clefs and notation from a new line.
6 Viero also noticed this slur, although his interpretation of this sign seems to be exaggerated.
7 Rosen pays attention to the arbitrariness with which the pianists interpret this relation at the beginning of the Sonata. He also underlines a perfect rhythmic (as well as harmonic and textural) combination of the ending of the exposition with the beginning of Grave.
8 In the Sonata in F major, Op. 28 by Friedrich Kalkbrenner, the 1st movement opens with a two-bar-long introduction, which is then written out again at the end of the exposition, creating a link leading back to bar 3 (as well as to the development). As far as the expressiveness and significance are concerned, the simple motif can be hardly compared to Chopin's genius phrase, however, it plays a similar double structural function.

category imprint: Differences between sources

issues: GE revisions, Changed phrase length

notation: Shorthand & other

Back to note