¹ The original 1842 edition of this opus, which was published in both Copenhagen and Leipzig in a bilingual edition with Danish underlay beneath Chamisso's translation, is available online in its entirety at Anne ørbæk Jensen, Jens Egeberg and Bruno Svindborg, eds, ‘Hans Christian Andersen and Music – A Web Presentation’, English version trans. David Hohnen, http://www2.kb.dk/elib/noder/hcamusik//martsviolerne/index_en.htm (accessed 19 July 2009). Other published commentaries on these unjustly overlooked songs include Sams, Eric, The Songs of Robert Schumann, 3rd ed. (Bloomington and Indianapolis: Indiana University Press, 1993): 143–8;Google ScholarDaverio, John, Robert Schumann: Herald of a ‘New Poetic Age’ (Oxford: Oxford University Press, 1997): 207–8;Google Scholar Frančois-Gildas Tual, ‘Hans Christian Andersen et Robert Schumann, “Der Soldat”, opus 40 n^o 3: La mise en musique d'un rècit elliptique’, Musurgia 8 (2001): 31–5;Google ScholarFinson, Jon W., ‘Between Lied and Ballade – Schumann's Op. 40 and the Tradition of Genre’, in Schumanniana nova: Festschrift Gerd Neuhaus zum 60. Geburtstag, ed. Appel, Bernhard R., Bahr, Ute and Wendt, Matthias (Sinzig: Studio Verlag, 2002): 250–65; and,Google Scholar from a more literary perspective, Massengale, James, ‘Ut Poesis (Picturae) Musica’, in H.C. Andersen: Old Problems and New Readings, ed. Sondrup, Steven P., University of Southern Denmark Studies in Scandinavian Language and Literatures 68 (Odense and Provo, UT: The University Press of Southern Denmark and Brigham Young University, 2004): 33–74.Google ScholarFinson's, Jon W.Robert Schumann: The Book of Songs (Cambridge, MA: Harvard University Press, 2007),Google Scholar which discusses op. 40 on pp. 98–102, appeared only after this article was accepted for publication.

² There is no evidence of any communication between Schumann and Andersen prior to the composition of op. 40 in July 1840, nor is there evidence that the composer was at that time aware of any of Andersen's poems other than the four translated by Chamisso and included in his Gedichte, 2nd ed. (Leipzig: Weidmann, 1834)Google Scholar (the Andersen poems do not appear in the first edition from 1831, but they do appear in later editions). Clara was in Copenhagen in March and April 1842, at which time she informed Andersen about Schumann's settings and asked him on her husband's behalf if Robert could dedicate the songs to him; she then returned home to Leipzig with a letter for Robert from Andersen. The two men did not meet until July 1844, when Andersen was travelling through Leipzig and spent an evening at the Schumanns’ household, where he heard a performance of three of the op. 40 songs. For more information on the relationship between Schumann and Andersen, see Boetticher, Wolfgang, ed., Briefe und Gedichte aus dem Album Robert und Clara Schumanns (Leipzig: VEB, 1979): 19 and 211–13;Google ScholarAppel, Bernhard R. and Hermstrüwer, Inge, eds, Robert Schumann und die Dichter: Ein Musiker als Leser (Düsseldorf: Droste, 1991): 187–9;Google ScholarMylius, Johan de, ‘Hans Christian Andersen and the Music World’, in Hans Christian Andersen: Danish Writer and Citizen of the World, ed. Rossel, Sven Hakon (Amsterdam and Atlanta: Rodopi, 1996): 202–3; andGoogle ScholarCelenza, Anna Harwell, Hans Christian Andersen and Music: The Nightingale Revealed (Aldershot: Ashgate, 2005): 101–6Google Scholar.

³ ‘Was müssen Sie von mir denken, daß ich Ihnen auf Ihre liebenswürdigen Zeilen, die mich so sehr erfreuten, so lange die Antwort schuldig geblieben bin? Aber – ich wollte nicht mit ganz leeren Händen vor Ihnen erscheinen, obwohl ich recht gut weiß, daß ich Ihnen eigentlich nur etwas zurückgebe, was ich erst von Ihnen empfangen. Nehmen Sie denn meine Musik zu Ihren Gedichten freundlich auf. Sie wird Ihnen vielleicht im ersten Augenblick sonderbar verkommen. Ging es mir doch selbst mit Ihren Gedichten so! Wie ich mich aber mehr hineinlebte, nahm auch meine Musik einen immer fremdartigeren Charakter an. Also, an Ihnen liegt die Schuld allein. Andersen'sche Gedichte muß man anders componiren, als “Blühe, liebes Veilchen” usw. usw.’ The translation of this letter, as well as all of the other German texts in this article, is my own. A facsimile of Schumann's original letter is available online at Jensen, Egeberg and Svindborg, ‘Hans Christian Andersen’. German transcriptions of portions of the letter are available in Storck, Karl, ed., Schumanns Briefe (Stuttgart: Franz Stassen, n.d.): 179–80, andGoogle ScholarAppel, and Hermstrüwer, , Robert Schumann und die Dichter, 189.Google Scholar Various English translations (both complete and partial) are available in Storck, Karl, ed., Letters of Robert Schumann, trans. Bryant, Hannah (London, 1907; reprint ed., New York: Benjamin Blom, 1971): 237–8;Google ScholarSams, , Songs of Robert Schumann, 143Google Scholar; Massengale, , ‘Ut Poesis’, 50Google Scholar; Celenza, , Hans Christian Andersen, 104; andGoogle ScholarFinson, , Robert Schumann, 98–9Google Scholar.

⁴ For a detailed discussion of Schumann's activities during the Liederjahr, see Daverio, , Robert Schumann, 182–202.Google Scholar In addition to Frauenliebe und Leben op. 42 (composed 11–12 July), Schumann had by this time written his most famous song cycles, including the Heine Liederkreis op. 24 (February), Myrthen op. 25 (March and April), the Eichendorff Liederkreis op. 39 (May), and Dichterliebe op. 48 (also composed in May).

⁵ For an overview of these songs, see Finson, , Robert Schumann, 175–82;Google ScholarDaverio, , Robert Schumann, 427–33; andGoogle ScholarStein, Jack, ‘Musical Settings of Songs from Wilhelm Meister’, Comparative Literature 22 (1970): 139–42Google Scholar.

⁶ In examining just these three songs, I am clearly sidestepping the issue of cyclicity among the five songs of op. 40. The question as to whether Schumann conceived these songs as a coherent entity that must be performed as a whole is open to debate. Certainly the tonal structure of the five adds musical coherence to the opus: the outer songs are in G major, the second and third songs are in D minor (with the third song featuring a substantial G-minor section and ending on a half cadence that is ‘answered’ by the start of the next song), and the fourth song fluctuates between D minor and G minor, ending in G major (see the discussion of the song below). Poetically, however, there is nothing that clearly connects the poems. Although Chamisso did group them together in Schumann's order in his Gedichte, Andersen published them in an entirely different order; they do not appear adjacent to each other in his publications and, in fact, the fourth song appeared in a different volume from the others (see n. 7 below). Schumann's reason for including the fifth song also remains open for interpretation; the poem does not appear near the Andersen poems in Chamisso's edition, and its subject matter, a flippant love song, has nothing in common with the rather dark Andersen poems (but that may well have been Schumann's point: to counterbalance the seriousness of the preceding songs with a frivolous conclusion). Furthermore, Schumann himself at least once authorized a performance of just a portion of the opus, when Andersen visited the composer's home in 1844 and heard a performance (with Clara at the piano) of just the first, third and fourth songs. David Ferris, in his discussion of Schumann's compositional process for his song cycles and song sets as evidenced by the drafts in the composer's ‘Berlin Notebooks’, includes op. 40 as an example of a set in which ‘the drafts appear either in the exact order of publication or very close to it, but they do not have a title page and are not set off as a separate unit’ (unlike the Heine Liederkreis, Dichterliebe, and Frauenliebe und Leben); see his Schumann's Eichendorff Liederkreis and the Genre of the Romantic Cycle (Oxford: Oxford University Press, 2000): 174.Google Scholar Tual, however, presents arguments for a cyclic conception of the cycle in ‘Hans Christian Andersen et Robert Schumann’, 34–5, as does Finson in ‘Between Lied and Ballade’, 252–6, 258–9 and Robert Schumann, 98–9. For more studies dealing with the controversial question of unity in other Schumann cycles, see Komar, Arthur, ‘The Music of Dichterliebe: The Whole and its Parts’, in Robert Schumann, Dichterliebe: An Authoritative Score, ed. Komar, Arthur (New York and London: W.W. Norton, 1971): 63–94;Google ScholarNeumeyer, David, ‘Organic Structure and the Song Cycle: Another Look at Schumann's Dichterliebe’, Music Theory Spectrum 4 (1982): 92–105; andGoogle ScholarMcCreless, Patrick, ‘Song Order in the Song Cycle: Schumann's Liederkreis, Op. 39’, Music Analysis 5 (1986): 5–28.Google Scholar See also Ferris's critique of these studies in Schumann's Eichendorff Liederkreis, 25–58, 185–9, 209–10 and 222–3.

⁷ The poem of the second song was the first of the four to be published, in Nyt Repertorium for Moerskabslæsning 1/1 (4 Nov. 1829): 15.Google Scholar It was reprinted, together with the first and third songs, in Andersen's first book of poetry, Digte (Copenhagen: E.H. Robert, 1830).Google Scholar The fourth song appeared in Andersen's second book of poetry, Phantasier og Skizzer (Copenhagen: Paa Forfatteren, 1831).Google Scholar The information on Andersen's life and works in this section is drawn from Bredsdorff, Elias, Hans Christian Andersen: The Story of his Life and Works 1805–75 (London: Phaidon, 1975)Google Scholar; Rossel, Sven Hakon, ‘Hans Christian Andersen: The Great European Writer’, in Hans Christian Andersen: Danish Writer and Citizen of the World, 1–125;Google ScholarWullschlager, Jackie, Hans Christian Andersen: The Life of a Storyteller (New York: A.A. Knopf, 2001)Google Scholar; Andersen, Jens, Hans Christian Andersen: A New Life, trans. Nunnally, Tiina (Woodstock, NY: Overlook Press, 2005); andGoogle ScholarZipes, Jack, Hans Christian Andersen: The Misunderstood Storyteller (New York: Routledge, 2005)Google Scholar.

⁸ Andersen, Hans Christian, Mit eget Eventyr uden Digtning, ed. Topsøe-Jensen, Helge Gottlieb (Copenhagen: Nyt Nordisk Forlag, 1942): 65Google Scholar; translation from Kofoed, Niels, ‘Hans Christian Andersen and the European Literary Tradition’, in Hans Christian Andersen: Danish Writer and Citizen of the World, 222Google Scholar.

⁹ For an overview of Heine's aesthetics, see Perrey, Beate Julia, Schumann's Dichterliebe and Early Romantic Poetics: Fragmentation of Desire, Cambridge Studies in Music Theory and Analysis 18 (Cambridge: Cambridge University Press, 2002): 17–18 and 71–107. See alsoGoogle ScholarYouens, Susan, ‘“Pure” Song: “Du bist wie eine Blume” and the Heine Juggernaut’, Nineteenth-Century Music Review 3/2 (2006): 3–32Google Scholar.

¹⁰ See, for example, Bøggild, Jacob, ‘Ruinous Reflections: On H.C. Andersen's Ambiguous Position Between Romanticism and Modernism’, in H.C. Andersen: Old Problems and New Readings, 75–96, andGoogle ScholarKofoed, Niels, ‘The Arabesque and the Grotesque – Hans Christian Andersen Decomposing the World of Poetry’, in Hans Christian Andersen: A Poet in Time, ed. Mylius, Johan de, Jørgensen, Aage and Pedersen, Viggo Hjørnager (Odense: Odense University Press, 1999): 461–9,Google Scholar as well as Kofoed's more general overview in ‘Hans Christian Andersen and the European Literary Tradition’, 209–56.

¹¹ Throughout the article, the German poems are given as they appear in Chamisso's edition, not in Schumann's music. The texts in the music examples, however, follow those given in the 1842 musical edition.

¹² Massengale, , ‘Ut Poesis’, in which he builds upon the work of Jørgen Bonde Jensen, H.C. Andersen og genrebilledet (Copenhagen: Babette, 1993).Google Scholar

¹³ Modern readers may be a bit confused by the final line of the poem; the implication (made more explicit in Andersen's original) is that the son will be hanged for robbery, at which time the ravens will feed upon his corpse.

¹⁴ The original Danish versions of the op. 40 poems are presented side by side with Chamisso's translations in Massengale, ‘Ut Poesis’, 71–3. For a discussion of other differences between Andersen's and Chamisso's versions of the poem, see ibid., 55–6.Finson also discusses the effect of Chamisso's changes in Robert Schumann, 100.

¹⁵ Finson also notes the similarity to ‘Zwielicht’ in ‘Between Lied and Ballade’, 262–3, and Robert Schumann, 100.

¹⁶ Brinkmann, Reinhold, ‘Lied als individuelle Struktur: Ausgewählte Kommentare zu Schumanns “Zwielicht“’, in Analysen: Beiträge zu einer Problemgeschichte des Komponierens: Festschrift für Hans Heinrich Eggebrecht zum 65. Geburtstag, ed. Breig, Werner, Brinkmann, Reinhold and Budde, Elmar (Stuttgart: F. Steiner, 1984): 257–75.Google Scholar

¹⁷ The pitch nomenclature used in this article is the so-called ‘Hemholtz system’, in which middle C is written as c', lower pitches as C, C', C”, and so on, and higher pitches as c”, c”’, and so on.

¹⁸ The text of Heine's poem is: ‘Ich wandelte unter den Bäumen / Mit meinem Gram allein; / Da kam das alte Träumen / Und schlich mir ins Herz hinein. // Wer hat euch dies Wörtlein gelehret, / Ihr Vöglein in luftiger Höh’? / Schweigt still! Wenn mein Herz es höret, / Dann tut es noch einmal so weh. // “Es kam ein Jungfräulein gegangen, / Die sang es immerfort, / Da haben wir Vöglein gefangen / Das hübsche, gold’ne Wort.”// Das sollt ihr mir nicht erzählen, / Ihr Vöglein wunderschlau; / Ihr wollt meinen Kummer mir stehlen, / Ich aber niemandem trau’ (‘I wandered under the trees alone with my grief; the old dreams came along and snuck into my heart. // Who taught you that little word, you birds in the airy heights? Be silent! When my heart hears it, then once more it hurts so much. // “A young woman came along, who sang it again and again; From there we birds captured the pretty, golden word.”// You should not have told me that, you wondrously sly birds; You want to steal my sorrow from me, but I trust nobody).

¹⁹ Massengale makes a case for the poem as a genre painting in ‘Ut Poesis’, 59–64.

²⁰ Genette, Gèrard, Narrative Discourse: An Essay in Method, trans. Lewin, Jane E. (Ithaca, NY: Cornell University Press, 1980).Google Scholar For another study that deals with the hybrid generic classification of the op. 40 poems, see Finson, , ‘Between Lied and Ballade’, esp. 253–7Google Scholar.

²¹ There is no full-length scholarly study of the nineteenth-century German ballad; my understanding of the genre is indebted to Johnson, Graham with Sams, Eric, ‘Ballad, §II: The 19th- and 20th-Century Art Form, 1. German Song’, in The New Grove Dictionary of Music and Musicians, 2nd ed., vol. 2 (London: Macmillan, 2001): 548–51, andGoogle ScholarSauer, Florian, ‘Ballade, IV: Vokalballade nach 1700’, in Die Musik in Geschichte und Gegenwart, 2nd ed., Part 1, Sachteil, vol.1 (Kassel: Bärenreiter, 1994): cols 1134–52.Google ScholarHolden's, Robert B. ‘The German Narrative Dramatic Ballad from Loewe to Mahler: The Development of the Genre and its Use as a Teaching Tool for Communication’ (DMA diss., Temple University, 2000)Google Scholar is intended primarily for performers and educators, and is not a detailed, comprehensive examination of the ballad as a genre.

²² As Tual (‘Hans Christian Andersen et Robert Schumann’, 31–4) and Massengale (‘Ut Poesis’, 60–3) have pointed out, the poem is not entirely in the present tense, for the final stanza shifts suddenly to the past tense (in Andersen's version, only the final two lines are in past tense). According to Tual, the tense change signifies the poet's refusal to admit the reality of the event, while Massengale interprets the entire poem as a memory. In Schumann's setting, a lengthy interlude before the final stanza creates the effect that the final verse is a commentary on events that just occurred during the interlude.

²³ Some may consider ‘Waldgespräch’, the third song of the Eichendorff Liederkreis, to be a ballad on account of its portrayal of a conversation between two individuals (Holden includes a discussion of this lied in ‘The German Narrative Dramatic Ballad’, 52–5). I do not, however, consider this a ballad, primarily because it presents only the direct speech of the characters and lacks a discernible plot, a narrator and any scenic description. Likewise, some may consider Chamisso's ‘Die Kartenlegerin’ op. 31, no. 2 (also composed in July 1840 with the other two Chamisso ballads) a first-person hybrid ballad similar to ‘Der Soldat’; Finson describes both of the poems as ballads with a ‘reflexive’ narrative (Robert Schumann, 97 and 100). However, because of its lack of a strong plot (beyond presenting the situation of a girl reading her future in a deck of cards) as well as its heavy emphasis on the protagonist's thoughts and emotional reactions (captured in delightfully dramatic fashion by Schumann's music), I consider this poem more a dramatic monologue than a true ballad.

²⁴ Finson discusses Schumann's entire body of ballads, in Robert Schumann, 93– 119. However, he defines the genre much more freely than I do (by which he is able to characterize all five of the op. 40 songs as ballads) and he does not discuss the songs in chronological order of composition.

²⁵ Sams, , The Songs of Robert Schumann, 35.Google Scholar See also Finson's discussion of the song in Robert Schumann, 116–19.

²⁶ Finson points out that this new melody ‘uses the same basic melodic shape and even the same pitches as the other strophes. Schumann has smoothed the line and its uneven rhythms a bit, but he has maintained enough resemblance to create a distant but still recognizable variation’ (ibid., 107–8).

²⁷ Schumann continued to show an interest in hybrid ballads that contain the narrator's personal reflections; his settings of Eichendorff's ‘Frühlingsfahrt’ op. 45, no. 2, and Kerner's ‘“Stirb, Lieb’ und Freud”’ op. 35, no. 2 (composed respectively in October and November 1840), are on poems that set dramatic narratives told by an impartial third-person narrator, but in which the final stanza becomes a personal, first-person reflection on the story.

²⁸ The strophes are in fact paired in their melodic resemblances, with the closest similarities existing between the first and fourth stanzas on the one hand, and the second and third on the other.

²⁹ Sams, , The Songs of Robert Schumann, 145–6;Google ScholarDaverio, , Robert Schumann, 207Google Scholar; Tual, , ‘Hans Christian Andersen et Robert Schumann’, 32–4; and Massengale, ‘Ut Poesis’, 60.Google Scholar Finson goes further by describing the entire lied as consisting of ‘graphic writing’ and ‘vivid musical imagery’ (Robert Schumann, 100–101).

³⁰ Tual (‘Hans Christian Andersen et Robert Schumann’, 32) also discusses the significance of the repeated ‘dazu’, but not from a metrical perspective.

³¹ Finson also points out this similarity in Robert Schumann, 101.

³² Daverio, , Robert Schumann, 208.Google Scholar

³³ Massengale (‘Ut Poesis’, 65–6) also offers insightful commentary as to the possible interpretations of the poem. As he points out, any reading of the poem depends on a ‘remarkably “expressionistic” idea’, and, in addition to the two interpretations I offered above, he also posits that ‘the whole scene may be regarded as a memory in the mind of a madman, who only arrives at a partial realization that the story he relates for us is actually his own’.

³⁴ Finson also notes the ‘tonal instability’ of the song, which he interprets as a manifestation of ‘the bereft fiddler's madness’ (Robert Schumann, 101).

³⁵ On directional tonality and tonal pairing in music of the earlier nineteenth century, see Krebs, Harald, ‘Alternatives to Monotonality in Early Nineteenth-Century Music’, Journal of Music Theory 25 (1981): 1–16;Google ScholarKinderman, William, ‘Directional Tonality in Chopin’, in Chopin Studies, ed. Samson, Jim (Cambridge: Cambridge University Press, 1988): 59–76;Google ScholarSchachter, Carl, ‘Chopin's Fantasy, Op. 49: The Two-Key Scheme’, in Chopin Studies, 221–53;Google ScholarDenny, Thomas A., ‘Directional Tonality in Schubert's Lieder’, in Franz Schubert: Der Fortschrittliche? Analysen – Perspektiven – Fakten, ed. Partsch, Erich Wolfgang, Veroffentlichungen des Internationalen Franz Schubert Instituts 4 (Tutzing: Schneider, 1989): 37–53; andGoogle Scholar Part 1, ‘The Origins and General Principles of Tonal Pairing and Directional Tonality’, of The Second Practice of Nineteenth-Century Tonality, ed. Kinderman, William and Krebs, Harald (Lincoln: University of Nebraska Press, 1996),Google Scholar which includes the following essays: Krebs, Harald, ‘Some Early Examples of Tonal Pairing: Schubert's “Meeres Stille” and “Der Wanderer”’ (17–33);Google ScholarSamson, Jim, ‘Chopin's Alternatives to Monotonality: A Historical Perspective’ (34–44); andGoogle ScholarKorsyn, Kevin, ‘Directional Tonality and Intertextuality: Brahms's Quintet Op. 88 and Chopin's Ballade Op. 38’ (45–83)Google Scholar.

³⁶ Krebs, , ‘Some Early Examples of Tonal Pairing’, 31.Google Scholar

³⁷ See Sams, , The Songs of Robert Schumann, 146–7;Google ScholarDaverio, , Robert Schumann, 208; andGoogle ScholarFinson, , Robert Schumann, 101–2Google Scholar.

³⁸ Sams, , The Songs of Robert Schumann, 147.Google Scholar

³⁹ This concept of open-endedness and the role of the audience in the construction of meaning as an important part of Romantic aesthetics is described in Ferris, , Schumann's Eichendorff Liederkreis, 59–88Google Scholar.