Music Theory Online
The Online Journal of the Society for Music Theory
Volume 7, Number 5, October 2001
Copyright � 2001 Society for Music Theory
Poet's Love and Composer's Love
Get new DjVu plug-in for musical examples (see the Editor's Comments for more information)
KEYWORDS: romantic lied, hermeneutics, analysis, Edward T. Cohn, Robert Schumann, Heine Liederkreis op. 24, multiple persona
ABSTRACT: This article proposes a modification of Edward T. Cone's notion of a composer's persona, which he developed in two significant variants in order to analyze and interpret the romantic lied. Specifically, I respond to Cone's most recent article about Schumann's Dichterliebe, "Poet's Love or Composer's Love," which had assumed a "a unitary vocal-instrumental protagonist that is coextensive with the persona of the actual composer of the song." Drawing on Novalis's Romantic hermeneutics, my modification proposes a "multiple persona" with an independent poetic voice, exemplified in a reading of Schumann's Heine-Liederkreis, Op. 24.
Received 13 July 2001
[1.1] The opening song of Schumann's Heine Liederkreis, Op. 24 (Example 1)(1) presents a man waiting for the woman he loves. In this state of expectation, his life has turned upside down: hopeful day-dreaming and nightly laments alternate. The chiasmus of "morgens/abends" versus "Nacht/Tag" links the poem's ending with its beginning and the simple present tense keeps the speaker's emotional ups and downs in a state of perpetual circularity. The music seems simple and straightforward. A walking bass in the accompaniment alternates with chords doubling the melody. Only in the second line from the end, "tr�umend wie im halben Schlummer," does the vocal line depart from the upper voice of the chordal accompaniment to rise twice to a syncopated E5 (m. 30).(2) The top notes of the three preceding phrases, B4, C#5, and D5--the circled notes in Example 1--have led stepwise to this melodic highpoint, whose chromatic inflection to Eb5 in m. 32 ushers in a quick descent to the concluding vocal cadence on the tonic. In the postlude, however, the piano quickly regains the register of the vocal climax. In the third bar from the end, a group of eighth-notes encapsulates the top pitches of the first three vocal phrases. But the syncopated E5, now on the dominant, is surpassed by an F#5 appoggiatura in the penultimate measure. The postlude then closes an octave above the voice.
[1.2] Given the guitar-like accompaniment, the protagonist of the song could be a singer creating his own text and music, whose individual phrase endings are punctuated with a half cadence and cadences to iii, IV, and I. While there is no real progress in the text's cycle of hope and disappointment, the gradually rising vocal line appears directed towards the upper tessitura, without, however, reaching the high tonic third, and falling instead back onto a cadence in the opening register. When the accompaniment is finally freed from vocal constraints in the postlude, it quickly regains, almost without effort, the register of the vocal line and finally reaches F#5, which, though only fleetingly touched upon and thus lacking full harmonic support, appears nevertheless as the melodic and expressive goal of the song. The overall melodic contour may be graphed analogously (though in an unorthodox way) to a Schenkerian Zug extending from F#4 to F#5 and offering an alternative close an octave above the voice (see Example 2). Thus the instrumental postlude attains, however briefly, what the human voice had been unable to reach.
[1.3] Instead of projecting a single unified utterance, then, a subtle split occurs between the protagonist's three creative selves, for what he says differs from what he sings, and what he sings from what he plays. Indeed, hearing the song as a multi-voiced utterance, which includes the poetry as an independent voice, captures very well the protagonist's mixed feelings and unfulfilled desire. In isolation, his voices send contradictory signals, but taken together their incongruence carries the song's message: while the poetic voice shows the lover in limbo, the linear drive of the music seeks to break this cycle. The vocal self, however, fails to do so, returning to the point from which it began, so that the music could enter into an endless loop, analogous to the situation in the poem. It is the instrumental self that takes up the unspoken cause of the voice by closing an octave above in the upper tessitura--in a different sphere, as it were. Despite the structural downbeat on the tonic in m. 36 (which moments ago had been the dominant of the preceding G major), the protagonist seems to continue wandering, and proceeds to an alternative ending in m. 45, suggesting that the upper register provides a qualitatively different closure than the cadence an octave below. In other words, the linear progression in pitch space is as important as the cyclical return of pitch class. As a result, the song both suggests closure and remains open--which would seem quite appropriate for the beginning of a song cycle.
[1.4] This hearing of the opening song from Schumann's Op. 24 serves a twofold purpose. It serves first as the impetus for revisiting, reexamining, and refining the concept of a musical persona in the well-known and influential approach to the analysis and interpretation of romantic art song by Edward T. Cone. I will do this in the second part of the paper by exploring further the above analysis and by drawing on romantic hermeneutics, specifically the notion of a plural voice, which I will call multiple persona. This concept will then, in the third part of this paper, open up a venue from the analytical-hermeneutic model to a broader critical reading, which seeks to gain a deeper understanding of the interaction between poetry and music and cyclical structure in Schumann's largely neglected first song cycle.
[2.1] To speak of "voices" in the analysis of song goes back to Edward T. Cone's 1974 book, which has had a significant impact on music criticism in the last two decades.(3) By using "voice" not merely in its narrower technical sense, Cone opened the door--at the time only a crack--to literary criticism, where the term can comprise the entire performative dimension of an artistic utterance. Meanwhile, that door has swung wide open,(4) with the Lied becoming the locus of increasingly esoteric experiments in the application of literary and critical theory analysis.(5) Yet Cone's initial approach remains a truly elegant hermeneutic device and valuable pedagogical tool for approaching the romantic art song, and perhaps other hybrid genres as well.(6) An important modification of Cone's model may be in order, however. In fact, it is inspired by Cone's own attempt at refinement in an article on Robert Schumann's Dichterliebe, entitled evocatively, "Poet's Love or Composer's Love?"(7)--a question which Cone settled in favor of the composer. Hence our focus on what is sometimes nicknamed "little Dichterliebe"; and hence the title of the present essay, which suggests that we move beyond Cone's either/or logic and give the poetry a voice of its own, one that stands on an equal footing with the music.
[2.2] To recall: Cone's concept of a composer's "voice" is equivalent to the idea of a musical "persona," a surrogate for the empirical composer, who projects "the illusion of the existence of a personal subject" behind or within a work of art.(8) As a "role" assumed by the artist, a persona (originally a "mask" in ancient theater) suggests for Cone "that all music, like all literature, is dramatic; that every composition is an utterance depending on an act of impersonation which it is the duty of the performer or performers to make clear."(9) Since there are two performers in accompanied song, Cone heard the composer speak with a "double voice, through a musical persona that assumes a double guise"(10): an instrumental persona and a vocal-poetic persona. He called the latter simply "vocal persona," because it expresses itself "at least as much by melody as by speech, and as much by tone-color as by phonetic sound."(11) In other words, "the vocal persona adopts the original simulation of the poetic persona and adds another of his own: he 'composes,' not the words alone, but the vocal line as well."(12) In the romantic art song, Cone argued, "the composer's persona governs words as well as music. The words, that is, have become a part of the composer's message, utterances of his own voice. In a sense, he composes his own text."(13)
[2.3] The idea that the composer exercises complete creative control over the entire song, including the poetry, was the central problem of early nineteenth-century Lied aesthetics. Goethe's preference for Friedrich Zelter's simple strophic settings of his poems over Schubert's elaborate through-composed Ges�nge is the most well-known example.(14) Since Cone preferred the romantic Lied, he has sided with Schubert on this issue, which explains his ruling in favor of the composer in "Poet's Love or Composer's Love?"(15) Cone does so after revisiting an old problem that had preoccupied him in his book: whether or not protagonists of romantic art songs are conscious of their singing. Originally, Cone had assumed that they were not and that this would be the normal condition. Since many characters portrayed in poetic texts appear to be speaking, he proposed that "the musical components, vocal and instrumental, [stem] from the subconscious levels." Although the protagonist is not conscious of singing or hearing the accompaniment, "his subconscious both knows and hears."(16)
[2.4] Returning to the issue, Cone still finds this division useful when approaching the complexity of, say, Wagnerian opera, but no longer for cases like Schumann's Dichterliebe. Since the protagonist of Heine's poems is already portrayed as a singer who creates his own songs, Cone assumed that in Schumann's cycle, "[t]he instrumental accompaniment directly conveys certain aspects of the musical consciousness of the vocal protagonist." Quoting from his book, he finds it no longer necessary to posit
"a triad of personas, or persona-like figures, involved in the accompanied song: the vocal, the instrumental, and the (complete) musical." According to that analysis, "the complete musical persona is to be inferred from the interaction of the other two"; I called it an implicit persona, or, "as the vehicle of the composer's complete message . . . the composer's persona." . . . But when the accompaniment proceeds directly from the imagination of the protagonist, a separate instrumental persona becomes superfluous, and therefore no complete musical persona is to be "inferred from the interaction of the other two." Instead, my three original figures have collapsed into one: a unitary vocal-instrumental protagonist that is coextensive with the persona of the actual composer of the song. . . . My hope is that an interpretation based on the foregoing principles might result in a closer and more sensitive relationship between voice and accompaniment than would otherwise be possible. Singer and accompanist, instead of taking for granted that each of them represents a unitary agent (to adopt the jargon of my book), would try to hear the song-texture as composite; and they would determine to what extent their parts could be made to coalesce in order to project a single persona.(17)
[2.5] Through this radical simplification of his original conception, Cone hoped to account for the "essential identity" of the two musical components in the majority of songs. Created by the same consciousness, the musical personae are, in fact, unified--one voice.(18) But with his refined model Cone seems to have lost as much as he gained. True, if the protagonist of a song is, as in Dichterliebe, the singer of his own songs, it makes sense to locate words and music in a single mind. Yet by collapsing the vocal and instrumental personae into a "unitary vocal-instrumental protagonist," Cone gives up a powerful tool for hearing the voice and the accompaniment as distinctly different utterances nevertheless, even when they are "inseparable components of a single invention."(19)
[2.6] My proposal, then, is to keep the basic conception of Cone's earlier model, while accommodating his later modification: to adopt the notion of a single creative mind, while still hearing independent voices. What is more, where Cone heard a complete musical persona constituted by instrumental and vocal personae, I hear a triple voice, which includes a poetic persona that remains on a par with the musical ones. Even when a poem has been molded into a through-composed song; even when its words have lost the rhythm of their original meter; and even when its text has been altered by the composer: the poetic text still remains an independent component of a song. Even sung, the words assume their own dramatic agency within what may be called a composer-poet's multiple voice.
[2.7] One can find a precursor for the concept of a multiple persona as an interpretive tool in early Romantic hermeneutics. Novalis formulated it in this entry on "Personenlehre" from his poetic encyclopedia, Das allgemeine Brouillon, from the late 1790s:
A truly synthetic person is a person who is several persons at once--a genius. Every person is the seed of an infinite genius. Though divided into various persons it is nevertheless also capable of being a single person. True analysis of a person as such produces several persons--a person can only break up, divide, and dissolve into several persons. . . .
Every personal utterance belongs to a certain person. All utterances of the person belong to the indeterminate (universal) personality and to one or several personalities at the same time. I.e. an utterance as human being, citizen, father, and writer at the same time.(20)
[2.8] The context for this passage is the early Romantic theory of the novel which was developed in Friedrich Schlegel's famous essay on Goethe's Wilhelm Meister.(21) Novalis's approach to the constellation of characters in a novel literally pre-"figures" the hermeneutic circle. A reader understands the plurality of characters that constitutes the novel as a product of a single organizing mind.(22) Thus the creative genius and the congenial reader engage in reciprocal activities: while the former is synthetic, the latter is analytical.(23) Novalis works out the performative and interpretive dialectic of genius in another entry on "Menschenbildungslehre" (a term that resonates with the concept of education-as-formation in the eighteenth-century Bildungsroman). Here the musical dimension of the multitude of characters born out by a single narrating voice is most explicit:
In order to develop the voice, a human being has to develop several voices--through this his organ becomes more substantial. Similarly, in order to develop his individuality, he must assume several individualities and know how to assimilate--through this he becomes a substantial individual. Genius. Everything m a d e by man is human--or quod idem est a part of man--a human being. (Science, Art, etc.) (24)
Just as a human being divides only into viable entities that have a human face, so does a voice divide only into voices. The result is the paradox of several-persons-in-one, comparable to the Holy Trinity, which Cone had originally invoked in his book in order to illustrate the idea of a composite musical persona.(25)
[2.9] The often numerous characters of a novel sometimes appear in song, say when a singer and accompaniment represent different (separate) characters--lover and beloved, poet and brook. We should nevertheless distinguish between the complex plurality of many characters in a novel and the more limited number of voices dwelling in the creative consciousness of a poet-composer. And there is, of course, a difference between a "person" and a "persona." The former does not imply the distinction, common today, between author and what Wayne Booth called "implied" author.(26) Still, even if we take into account the premises of idealist aesthetics, the religious implications behind Novalis's secularized exegetic model of the Holy Trinity, and the specifics of a particular genre: the idea of a multiple voice still offers a powerful heuristic tool that helps us to conceptualize the often paradoxical plurality of voices emerging from a single source. Thus, if the protagonist of a song appears as a poet-composer who creates both words and music, we may conceive of his poetic, vocal, and instrumental voices as separate entities: as different artistic "selves" inhabiting the same creative consciousness.
[3.1] Let us return, then, from the hermeneutic model to interpretive practice and explore further how the idea of a multiple persona might deepen our understanding of Schumann's Heine-Liederkreis. The idea of this exploration, however, is not to produce more analyses in the manner of the above close reading of the first song, but to follow up on how the special condition of the protagonist as projected through the multiple personae affects the remainder of the cycle. Our view of other songs, therefore, is necessarily limited to looking at a few motives which tell, like Dichterliebe, the story of unrequited love. In both works, hopeful sentiments are dashed early on in the cycle, followed by various attempts to cope with the loss. Perhaps the most striking parallel between the two cycles is the ending: the disappointed lover's wish to bury his songs, which nevertheless seem to come to life again in the last number with the recapitulation of earlier music. For Dichterliebe, Schumann selected sixteen poems from the more than sixty that Heine had grouped together in his Buch der Lieder under the title "Lyrisches Intermezzo." The nine songs of the Heine-Liederkreis, on the other hand, are identical with a collection of nine poems simply called "Lieder," which thus retain the narrative sequence of Heine's original.(27)
[3.2] As suggested above, a subtle incongruence between vocal melody, accompaniment, and poetry, of the various voices in the first song of the Heine-Liederkreis initiates an open-ended structure. This has ramifications for the rest of the cycle. Recall Schumann's repetition of the syncopated E5 in the vocal line, which highlights the verse "tr�umend wie im halben Schlummer" (mm. 29-32). This melodic repetition seems to spill over into the following line of text, where one finds a corresponding reiteration of "tr�umend," lacking in the original poem. This double repetition occurs over a transitional prolongation of G major--the local tonic before measure 28 and the global subdominant again after measure 33. It is partly responsible for a less than convincing return to D major and thus helps to motivate the piano prelude to complete the job. It embodies the in-between state of the dreamer, who hopes for the fulfillment of his desire for love. This "dream motive" is, of course, not a leitmotiv in the Wagnerian sense, but its variants permeate the cycle and support its narrative structure (see Example 3a below). Indeed, the postlude can only partly fulfill the harmonic and melodic implications of the passage, since it involves neither the voice nor the text. This lack of agreement between the musical and poetic personae is analogous to the unfulfilled narrative desire for the denouement of the story, that is, the protagonist's imagined union with the distant beloved. The resulting tension has reverberations throughout the cycle that I want to examine now.
[3.3] After a stormy second song in B minor, in which the protagonist impatiently awaits a rendezvous with his beloved, the third song in B major begins with a transformation of the dream motif in the prelude. It thus returns to the premise of the yet unanswered declaration of love that was behind the protagonist's fluctuating utterance in the first first song. Now a variant of the characteristic figure of the E lowered to Eb (now D#) wanders through a circle-of-fifths progression (see Example 4 and cf. Example 3a and c). But it is unclear what this is meant to say. Typical for many romantic song cycles, the intermittent narrative precludes knowledge of how much time elapsed after the second song and what happened in between. Because of the mode change from B minor to B major and the peaceful pace and atmosphere, the protagonist's meandering thoughts could be happy ones; but the text of the first stanza suggests that the migrating motive is the return of old dreams:
I was walking under the trees
alone with my grief,
when the old dreaming came
and crept back into my heart.
In Cone's most recent model, these dreams would be integral to, and unified in, the protagonist's creative consciousness: music that he can hear. Yet the prelude suggests that the complexity of the creative act is well explained by the individual agency of the utterance. Here, composition is not necessarily a voluntary act; the protagonist's reminiscence seems to enter his heart like an involuntary memory ("the old dreaming came"). The dreams lead a life of their own, just as melodies with which we wake up are often impossible to get out of our ear during the day. Indeed, the return of previous music is conspicuous: apart from the dream motif in the prelude, the vocal line of the first stanza also draws upon the initial gesture of the previous song, its former haste now shuffling along (see Example 5). Moreover, when the protagonist refers to his dreams at the end of the stanza, the chordal accompaniment breaks open and the "dream motif" literally "creeps into" his singing as a painful exclamation on a non-syncopated high G#5; the melodic contour and rhythm (not the exact pitch content) of the gesture reverberates twice in the piano (III:11-13).(28) Thus, as if his creative powers had been weakened, the protagonist lives on memories that are part of himself and part not. In this respect, too, the song's eerie G-major middle section invokes the birds to whom he lost his poetic craft (and, by implication, his voice), symbolized by the "golden word" that they caught from his beloved:
A young girl came this way
and she sang it all day long.
It was then that we birds caught
the pretty, golden word.
Thus the potential disintegration that lurks behind the incongruence of multiple voices at the beginning of the cycle develops into a pathological condition of hallucinations, which are typical of the romantic individual, whose love-sickness is--for Heine at least--also a cultural malaise.
[3.4] Indeed, if the loss of love threatens the protagonist with the loss of creative control and mental sanity, the fourth song, "Lieb' Liebchen, leg's H�ndchen auf's Herze mein," captures quite graphically how the dreams that "crept" back into his heart now wreak physical havoc on it. The off-beat accompaniment in the right hand lacks the bass notes of the first song (a disturbed heartbeat, as it were) and pounds against the dream motif that is no longer syncopated (see Example 3d). As a result, the rhythmic disjunction between vocal and instrumental personae at the end of both strophes is much more pronounced than in the first song (IV:15-17 and 35-37). What is remarkable, therefore, is how much the first song functions as both a material and a metaphorical matrix on which a central narrative strand of the whole cycle is constructed. Thus the fifth song, "Schöne Wiege meiner Leiden," restores the simple figure of alternating eighth notes with the bass articulating the meter that we heard in the accompaniment from the first song. But now its duple time is transformed into the soothing rocking of triple time. This lends a feeling of calm and consolation to the theme of farewell in the fifth song, very well-suited for what was originally planned to be the last number of the first book of the cycle. The publisher eventually followed Schumann's wish to print it in one volume, but the internal division remains visible in the tonal plateau reached between the fifth and sixth songs. But as we will see below, the sense of closure and repose is illusory; it is merely a stage in the protagonist's ongoing journey, his restless "wandeln" and wandering. Schumann structured the key sequence of the cycle around a double trajectory of falling fifths partly alternating in mode: D-b-B-e-E (first part); E-A-d-D (second part). If the motion by fifths suggests a downward motion through tonal space, however, hearing the trajectory as the result of stepwise voice leading reveals that Schumann exclusively chose triadic connections that produce an exclusively upward motion, something that David Lewin has recently called "upshift" voice-leading.(29) According to Richard Cohn, "the intermediate resting point, E major, is exactly half-way between the two D-majors in terms of semitonal motion." Thus D+ up to E+, via b, B, and e, requires a total of six semitones upward: A-B (2 semitones), D-D# (1 semitone), D#-E/F#-G (1 + 1 semitones); G-G# (1 semitone). Then, for the next group of songs E+ to D+, via A and d, requires a total of six further upward semitones: G#-A/B-C# (1 + 2 semitones), E-F/C#-D (1 + 1 semitones), F-F# (1 semitone).(30) As a result, the final D major is "higher" than the initial one--a fact that will be of significance when we consider the return to material from the opening song in the last song of the cycle.
[3.5] On the tonal plateau in the middle of the cycle, then, two songs of farewell and departure are juxtaposed, but their emotional and gestural character could not be more different. The fifth song articulates the point in the emotional journey at which the protagonist seems ready to leave the site of his sad memories behind, not without a sense of sweet nostalgia, which emerges from the strophic stanzas 1, 2, 4, and 7 (the last being a recapitulation of the first). These stanzas give a sense of closure: the long tonic pedal with which they begin is rounded off by the emphatic tonic cadences on the "Lebewohl" salutes, whose repetition is lacking in the original poem, but supports here the musical refrain. At the same time, feelings of anger and pain disrupt the reconciliatory retrospective, suggesting restlessness and failure to leave the past behind. Given the link to the accompaniment of the opening song, the melodic profile of the strophic stanzas is conspicuous in its gradual ascent to E5, by now surely a referential pitch in the cycle. A liminal remainder and reminder of the dream motif, this E5 returns almost obsessively, piercing the ear at the emotional highpoint of the song, which occurs toward the end of the section in which the fifth and the sixth stanzas are elided. Here the voice dwells on E5 for nearly six measures, briefly touching on F5 as if in utmost strain (see Example 6). At no point, however, is the E5 supported by more than a fleeting tonic sixth chord, let alone prepared by a root-position dominant. While the many cadences on E4 emphasize closure with the "Lebewohl" refrain in the lower register, the frequent, but unstable, E5 remains a reminder of a melodic pitch, which, in the overall trajectory of the cycle, was to go somewhere else: to F#5. The protagonist may leave behind the place, but not the past.
[3.6] After this ambiguous sense of closure, we will finally examine the second half of the cycle. Here our newly-gained sense for the protagonist's multi-voicedness will help to elucidate the disturbing interplay of musical and verbal irony, itself the trope of incongruence between literal and figurative meaning. Following the false farewell at the end of the first part of the cycle, the sixth song, "Warte, warte wilder Schiffsmann," stages the actual departure in a sarcastic address to the once revered woman. The poem is all hyperbole, which culminates in characterizing her as the union of biblical Eva and mythological Eris (the goddess of discord):
Apples brought all our calamity,
with them Eva brought death.
Eris brought the flames of Troy,
you brought both, flames and death.
The overstatement in the poetry correlates with a variety of musical figures: disjunct octave leaps and excessive runs in fast tempo, abrupt registral shifts (mm. 52-53 and 99-103), repetition to the point of ridicule (mm. 24-31 and 78-90); tritones in the melody set to alternating dominant seventh chords a semitone apart (mm. 36-40 and 44-48), vocal ellipsis (mm. 55-65), and more. Since verbal irony often manifests itself as a deliberate deformation of a "normal" way of speaking, these musical devices, too, deviate sharply from the previous "tone" of the cycle. Thus the tritone leaps function as a caricature of the "dream motif," which is blown out of proportion in the final verse ("Du bracht'st beides, Flamm und Tod"). Here the motif's characteristic repetition ends climactically on a rising unresolved seventh, whose A5 is even the highest sung pitch of the entire cycle (see Example 3e). In sharp contrast to the extreme intensity of the vocal part, the piano then tapers off in the postlude with a sense of understatement, suggesting that the ironic interplay between voice and accompaniment are far from presenting an integrated persona, but one on the verge of complete disintegration. Perhaps the ultimate irony emerges when the grave matter of the song is sharply thrown in relief against its diminutive attribute "Liedchen."
[3.7] In the seventh song, as a striking contrast, the protagonist, now traveling on the Rhine, peacefully contemplates the play of water. The vocal line, which follows the strophic arrangement of the poem, blends seamlessly into a self-sufficient lyrical piano piece, reminiscent of a suave barcarole. As the speaker paints a romantic idyll, poetry and music seem to be in perfect harmony. But the apparent integration of the poetry and music is illusory--just like the surface of the water. It does not at all reflect the gradual change of mood that leads in the last stanza to an analogy between the beloved and the river's bright surface but deadly depth:
Delightful on the surface, treacherous at heart,--
river, you are the image of my beloved.
She also has a friendly nod,
and an innocent and gentle smile.
At the end of the poem, however, Heine's notorious ironic reversal is not made explicit: the sinister part of the analogy--the beloved's inner wickedness--has to be inferred. And there is no hint of it in the music either. Since Schumann had sometimes been criticized for such apparent lack of a musical response to Heine's poetic irony, Cone felt the need to stand up for the composer:
Schumann's Dichter inhabits a different world from that of his poetic original: a world in which words give way to music as the primary vehicle of expression, in which to speak is to sing. . . . We should therefore not expect the personality of Schumann's Dichter to be the same as that of his purely poetic original.(31)
This defense, however, loses its raison d'etre when we allow for poetry to retain its own voice. Because the poem's irony remains in effect, the composer may choose not to match it musically. On the contrary, in "Berg und Burgen schaun herunter," the irony unfolds precisely between the sarcasm of the poem and the enchanting lyricism of the music. As with the river, the harmful implications of the words lurk below the bright surface. Had E5 not been such an important pitch in the cycle, one might not even notice how much the voice dwells on it here, moving so smoothly to F#4 that the music appears oblivious to the ironic meaning behind the text, except, perhaps, when leading first to a cadence in the supertonic before reaching the tonic itself (cf. VII. mm. 13-20 and Example 7 mm. 37-44). That this is the "image" of the beloved, caught between dream and reality, is made clear in that the fourth stanza goes over the same passage drawing the conclusions from the first three stanzas. Singers sometimes color the last stanza with a "darker" or somewhat threatening tone, in order to let us sense the song's falseness.(32) But this added effect would spoil the ironic disjunction between musical sweetness and textual bitterness. We need neither blame nor defend Schumann when his protagonist does not put his music where his words are. Double talk may require a double voice.
[3.8] The music thus amplifies the verbal irony with its own means (as in the sixth song); or irony emerges only in contradiction with the text (as in the seventh song). The eighth song of the Heine Liederkreis resorts to yet another device, that of parodia sacra. Moving furthest beyond the expressive realm of the cycle, it produces the ironic effect by a shift in style. Withdrawn from the world and looking back on the burden of unfulfilled love, the protagonist seeks refuge in religion, which is not born out by the poem, but insinuated by the music, which sets it to the melody of the chorale "Wer nur den lieben Gott lässt walten" and thus gathers, like a congregation, all his different voices into a unified utterance.
At first I almost wanted to give up
and I believed I could never bear it.
Yet I have born it;
but don't ask me: how?
The tone of the hymn would seem to answer the question "Nicht wie?" (set to a half-cadence), which is repeated at the end of the quatrain and calls, cleverly, the sincerity of the parody into question. And indeed, when the half-cadence resolves into the D major of the last song, a different answer emerges. At the point of utmost distance and alienation, the protagonist's instrumental identity suddenly breaks out into the emphatically pianistic idiom that brushes the questionable pietism aside. (The beginning of last song is reminiscent of Schumann's first Novelette, Op. 21.) Leaving behind the different ironic variants of aggressive attack, subtle sarcasm, and religious regression, the protagonist's original emotions "erupt" to re-collect his scattered creative selves into one genuine voice:
With myrtles and roses, lovely and fair,
with fragrant cypresses and spangled gilding
I would adorn this book like a shrine
wherein to bury my songs.
[3.9] The passionate call for the burial of his songs together with his love parallels the final song of Dichterliebe, where the songs are buried at the bottom of the ocean, only to rise from their grave in the famous postlude (itself a recapitulation of the postlude from the twelfth song, where flowers gave voice to the unspeakable sorrow of the silent protagonist). As has often been noted, Dichterliebe closes in Db (or C#)--one possible dominant of the tonally ambiguous first song (which invokes both A major and F# minor). Thus the protagonist is not only conscious of having composed a corpus of songs which are capable of leading an afterlife of their own, but tonal cyclicity supports narrative cyclicity. In this respect, the Heine-Liederkreis is even more explicit, and its interpretive and critical ramifications are suggestive. Here the songs are not buried in the ocean, but in a book. The tome becomes a tomb, past passion frozen into print. Just as Romanticism newly negotiated the dichotomy between speech and writing, the corpus of songs becomes a corpse, foreshadowing, avant la lettre, the notorious association in deconstruction between death and writing. Still in the spirit of romantic utopia, however, the "spirit of love" can revive the body, like the divine pneuma that once breathed life into a clump of clay. The ultimate test, however, is whether the "book of songs," brought alive, will be able to do better than the composer-poet himself; whether the story will be more moving than the teller; whether the letters imploringly looking at the beloved will speak more eloquently than the eyes themselves.
[3.10] Nowhere, perhaps, is this look more poignant than in vision of pastness becoming presence, or distance becoming nearness, during the fourth stanza, when the music returns to the first song. Only now, in a final transformation of the dream motif, both musical personae and the voice of poetry come together, suggesting--unequivocally--how the dream of requited love might be fulfilled. Only now, the non-syncopated double gesture B4-E5 leads directly to an F#5 on the downbeat that is supported by the tonic harmony (see Example 3f and Example 8). Only now the desire for love, first articulated in the incongruence between poetry and music in the first occurrence of the dream motif, which initiated and sustained the narrative desire throughout the cycle, is--momentarily--resolved in the emphatic declaration to the "sweet love in distant land." Taking back all the bitterness for a brief instant, this declaration is a paradigmatic example of Romantic transcendence: no less a dream than during the first song, but it is a dream that articulates its attainment--one that the beloved should read, would hear, and might understand.
[3.11] True, the protagonist's hope is that his sad story, by being told, will ultimately be undone; that by closing the gap between the first and the last song, it will overcome the emotional and physical distance between him and his beloved. True also, that his wish remains the very dream that it was from the very beginning. As in Dichterliebe, the illusion is only fleeting, the conceptual and corporeal unison of the three voices breaks apart. The final stanza wants to retreat from the achieved F#5 and falls back onto a cadence in the lower register; and the postlude closes like an exhausted reflex of the prelude with a diminished seventh, whose Bb harks back to the D-minor chorale (and perhaps the upward leading A# in the postlude of the first song), as if to stall forever the upward progression of the cycle. But once at least we were to believe that love was both the composer's and the poet's. This, however, we can hear only if we lend an ear both voices: those of music and poetry.
Return to beginning of article
Copyright � 2001 by the Society for Music Theory.
All rights reserved.
 Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.
 Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:
This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.
 Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.
This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.
Brent Yorgason, Editorial Assistant
Updated 18 November 2002