Amsterdam International Electronic Journal for Cultural Narratology (AJCN)

Amsterdam International Electronic Journal for Cultural Narratology (AJCN)

What is autobiographical authenticity in music? The question of the “secret vocal part” in Alban Berg’s Lyric Suite revisited from a narratological perspective

Karl Katschthaler

University of Debrecen, Hungary

To state that music is a medium and narrative is not a medium, but a transmedial cognitive frame, will hardly trigger controversies. But if transmediality is a key feature of narrative in the sense that narrative has the ability to occur in different media, narrative may be defined in two ways. The first is to construct a media-independent concept of narrative on a level of abstraction that makes it possible to apply this concept to different media. The other way is to define a prototype of narrative with a series of typical features, some, but not all of which, must be found in the medium in question. An example of the first approach is the definition of narrative in the article about “Music and narrative” in the Routledge Encyclopaedia of Narrative Theory as “a cognitive frame that can inform a plurality of signifying practices in order to meaningfully represent, and make sense of, temporal experience” (Wolf 2005: 324).

Fotis Jannidis criticizes highly abstract media-independent concepts of narrative such as this as “nothing more than a marginally useful hypostatized abstraction” (Jannidis 2003: 51). As an alternative, he proposes the following prototype: “a narrator tells an audience of listeners something that happened,” to which he attributes a series of typical features: “The histoire is a self-contained meaningful structure whose most important components are chronology, causality, teleology, and intentionality” (51). The most important feature of narrative, however, is representationality, because a “story is not a narrative, but the representation of a story is” (50). The logical consequence of this representational conceptualization of narrative is that it must not be defined media-independently but “should always be treated as something anchored in a medium” (50). For the purposes of the present discussion, the question is thus: What features of narrative are anchored in the medium of music?

If, on the one hand, the prototype of narrative is that somebody tells a story, and on the other hand, that it is widely recognized that music cannot tell stories, the question of narrative in music necessarily starts from a position of distance from the prototype. I think the dilemma can only be solved when we don’t think of narrative in music as a formal structure but as a representation of meaning or, as Lawrence Kramer has put it: “narrative elements in music represent, not forces of structure, but forces of meaning” (Kramer 1991: 161). Kramer’s point of departure is the observation that music has been used to accompany stories since the Renaissance in a whole range of different genres from songs and programme music to symphonies which, although without specified programs, nevertheless compel audiences to find stories in them (cf. 154). Stating that “[i]n relation to narrative, music is a supplement, in the deconstructive sense of the term” (144), he seeks to make Jacques Derrida’s philosophical concept of the supplement productive in the musicological context of the relationship of music and narrative. The term “supplement” is used ambiguously, for it is simultaneously an excess and a remedy. In Kramer’s words:

By taking on a supplement, a presumed whole puts its wholeness into question. The act of addition exposes an unacknowledged lack which the supplement is needed to counter. And in countering that lack the supplement exceeds its mandate and comes to replace the whole it was meant (not even) to repair. (Kramer 1991: 155)

As a result, the original relationship of narrative as primary and music as secondary, an accompaniment which adds something extra to the narrative, is reversed: “the music becomes the primary term and the story its mere accompaniment” (155). [1]

Susan McClary shares with Kramer the conviction that forces of meaning and not of structure are generated by narrative elements in music when she, like him, reads musical compositions as cultural texts (cf. McClary 1997: 21). However, she adds a historical dimension to her argument. She restricts narrative in instrumental music to the period from around 1700 to around 1900, a period of 200 years “in European history most focused on notions of the centered Self” during which instrumental music “traces narratives of subjective becoming or Bildung” (24). With the crisis of the concept of Bildung[2] and the crisis of the centred self around 1900, avant-garde music of the twentieth century developed an anti-narrative stance: “The radical compositional devices associated with primitivism, expressionism, and chance emerged as attempts at breaking the hegemony of narrativizing musical processes, so engrained by 1900 that extreme solutions such as these seemed the only recourse” (22). McClary identifies Schönberg’s design of twelve-tone music in this context as an attempt to prevent unintended returns of tonality’s demand for narrative closure (cf. 32, n. 9). Schönberg’s pupil and devoted follower, Alban Berg, was surely familiar with his master’s anti-narrative stance.

However it may be, Berg’s string quartet entitled Lyric Suite is a piece of instrumental music which, despite the fact that it is a piece of twelve-tone music, cannot simply be located in the realm of anti-narrative music. Instead, it should be regarded as a paradigmatic example for anti-narrative music in crisis or, from another perspective, as narrative music in crisis. In defence of this thesis, I will examine first the supplementary relation of music and narrative in the case of the Lyric Suite, then elaborate on a number of cultural texts referred to by narrative elements in the piece and, finally, deliver a narratological critique of the notion of “music as autobiography” introduced by Constantin Floros (1992) in his monograph on Alban Berg, where the Lyric Suite plays a prominent role.

Since Floros and George Perle, independently of one another, have “discovered” a suppressed narrative of infidelity and impossible love, culminating in the despair expressed in Baudelaire’s poem De profundis clamavi (translated by Stefan George), the so-called secret vocal part in the last movement (Largo desolato) of the Lyric Suite, audiences are routinely provided with this narrative in the programme notes or in similar texts accompanying a performance of the piece. The simple private narrative of Berg’s extramarital affair with Franz Werfel’s sister, Hanna Fuchs-Robettin, is presented as the suppressed programme of the quartet and the key to understanding the music. This persistent foregrounding of the narrative might give us reason to doubt Kramer’s concept of music as the supplement of narrative (which he illustrates with Wagner) when he asks: “Why bother to follow all that stuff Wotan is saying to Erda when we can just listen to the doom-laden procession of the leitmotives?” (Kramer 1991: 155). Why then do we bother so much with the love story of Alban and Hanna? Nonetheless, Kramer is right, I think. I would even go as far as to argue that nowhere else does the supplementary character of music-narrative relation become more apparent than in case of the Lyric Suite

Berg was not the first to deny the programmatic nature of his composition or to have suppressed an existing programme. [3] Mahler, for one, had done it before in the case of his Symphony nr. 3, the programme notes of which he explicitly withdrew. Consider also the case of his Symphony nr. 6, known as “The Tragic.” Here, the “secret” programme was not detected by musicologists but was revealed by Mahler’s widow, Alma, although musicologists did take a part in propagating the secret.[4] A similar fixation of critics and audiences has since set in with regard to the three hammer blows in the finale, supposedly symbolising the anticipated three strokes of fate in Mahler’s life. But there is a difference: Mahler’s symphony stands in the tradition of the Beethovenian symphony, which in Kramer’s words “compels audiences to find originary stories where the composer has left them unspecified” (1991: 154). Berg, however, didn’t write a symphony of this type, predicated on the elaboration of an implicit narrative, but a composition in a technique, which, if we follow McClary, was invented partly with the purpose of putting an end to the nineteenth-century narrativisation of music.

Berg consciously suppressed the narrative, not only by doing away with the programme and removing the text of Baudelaire’s poem, the text of the so called the hidden vocal part, but also with his decision, already taken, to write a twelve-tone composition. In the reception of the piece, however, a reversal of the supplementary relation of music and narrative seems to have occurred: the suppressed narrative as a “secret” makes itself the centre of attention, pushing the music into the background. With the revelation of this “secret,” the autobiographical narrative itself starts to act as a supplement, squeezing out the cultural narratives the music refers to.

What cultural narratives can we associate with the Lyric Suite? One of them is hiding behind the “concealed vocality” of the Largo desolato, as discovered in the 1950s by Hans Ferdinand Redlich (cf. 1957: 142). Behind this concealment is more than just a “secret vocal part” and the suppression of the text of Baudelaire/George’s poem De profundis clamavi. It is true that Berg wrote to Hanna Fuchs-Robettin of the Largo desolato as “this song without words (for no one but you is to know that these notes of the last movement are underlain by Baudelaire’s words)” (Berg quoted in Perle 2005: XVII); but to conclude, as George Perle does in his critical edition of the Lyric Suite, “including the secret vocal part,” that the composer “suppress[ed] the authentic version and le[ft] to the world in its stead what amounts to an ‘arrangement’ with no more title to acceptance as an ultimately authentic representation than is the once familiar orchestral version of Isoldens Liebestod from Tristan und Isolde” (XVI) is mistaken.

It is mistaken, first, because as Hermann Danuser explains, through Baudelaire’s poem the aesthetic subject is transposed into a sphere of aesthetic modernity. Therefore, the work should not be reduced to an autobiographical confession (cf. 2001: 30–31). The suppressed poem gives the work a radical meaning which exceeds the frame set by the author for his addressee. The concept of immortalization of love in the Liebestod, to which Berg refers with his quotations from Wagner’s Tristan (one of them in measure 26 you can hear at the beginning of the first audio example below), gets linked with Baudelaire’s modernist concept of the impossibility of the love of the poète maudit, cast away into the abyss (cf. 32).

Perle’s claim is further mistaken to the extent that Berg did not simply suppress the text of the poem, but turned it into music. Calvin Scott has shown that the melodic line of the “secret” vocal part perfectly mirrors the intonation of the verses recited in the manner of contemporary actors reading poetry. This transformation of words into music, known as “intermedial contouring of the declamation,” is evident, for example, in the arch-shaped contour of the melodic lines in measures 22–27, the acceleration in measure 30, and the descending melodic line beginning in measure 40 (cf. Scott 2007: 139).

Scott also points out that George’s translation of the Baudelaire text is already an intermedial transformation because its starting point is the musical quality of language. He notes that George tried to transform the “mélodie rythmique” and the musical sounds of the French poem into a “flow of carefully chosen sonorous words” (139). This shift away from meaning and towards sound in modern poetry is described by Jacques Le Rider following his interpretation of Julia Kristeva’s Revolution of Poetic Language as the “feminizing of writing,” one of the two types of modernity he distinguishes (1990: 151). This type of modernity follows the dream of a cosmogonic eros and is, for Le Rider, represented by Gustav Klimt. The other type he associates with Arnold Schönberg and Ludwig Wittgenstein, whom he calls the ascetic moderns. Their vision is one of recovery from the crisis of culture and artistic creation not by erotising culture and art, but on the contrary, through the genius of the male, who hardens himself, deadening sensuality and erecting barriers against the feminization of art (cf. 162). Male protest, the search for aristocratic beauty, the soul, on the one hand, and the feminizing of writing, writing as a promise of happiness, of the body, on the other hand – these are the binary oppositions of male and female in this cultural context (cf. 152).

While the first phase of Arnold Schönberg’s works can be associated with Jugendstil, in the following phases, with the liberation of dissonance and the elaboration of the twelve-tone system, he turned away from sensuality and mysticism in favour of logic and rationalism (cf. Le Rider 1990: 160; Gerlach 1985 and Rosen 1996: 26 ff.). It is no coincidence that Schönberg, in the preface to his Theory of Harmony (1911), refers to Otto Weininger, who considered the female principle as uncompromisingly opposed to the male. Karl Kraus agreed, but in contrast to Weininger, who regarded this antagonism as deadly, he concluded that this conflict between the two principles is the creative force of man (cf. Le Rider 1990: 159). But at this point a dilemma arises. On the one hand, male genius needs the feminine to be creative; on the other hand, this need pulls him down. The only possible solution, as Nike Wagner concludes, is sexual abstinence, as illustrated by the figure of Gustav Mahler, described by his wife in her autobiography as living in celibacy because he feared being pulled down by the feminine (cf. Wagner 1982: 149). When Kraus staged Wedekind’s Büchse der Pandora as a private performance in Vienna, he idealized the figure of the writer Alwa. A decadent product of the Kaffeehaus world in the Wedekind’s play, Kraus characterized him as the only one who, intellectually, is above the incidents taking place around Lulu. In his version, Alwa is a masochist-heroic prisoner of hopeless love who does not perish but turns his suffering into creativity (cf. Wagner 1982: 182–185). Alban Berg, who was one of the guests invited to attend Kraus’s private performance, made Alwa a composer in his Lulu, identifying with this figure in a similar way. Leon Botstein summarises Berg’s idealizing of the feminine inspired by Kraus:

Like Kraus he idealized a premodern and nostalgic notion of nature and the feminine, and therefore an idealized characterization of love. The spiritual in love emerged from the carnal desire [sic] but had to lead to creativity. (Botstein 2010: 321)

Now, how do these cultural narratives relate to the Lyric Suite? We have seen that Berg, when he wrote his “song without words” in the Largo desolato, transcribed the words of the poem into music by turning away from the discursive meaning of the words and towards their sound and rhythm. The consequence of this process is that sound and rhythm are highlighted but that the meaning of the words disappears. This intermedial transposition can be regarded as the final consequence of the process of the feminization of writing. This suppression of the words of the poem and their sublimation in the music must be correlated with the question as to why Berg chose the genre of the string quartet in order to declare his love to Hanna Fuchs-Robbetin. In his long letter to her in July 1925, he wrote that he would love to write songs but that he could not because the words of these songs would reveal his secret. He concluded that he was forced to write songs without words and then laconically declared: “Maybe I will write a string quartet!” (cf. Floros 2001: 32). He could hardly have chosen a better genre to declare and conceal his love at the same time!

According to Melanie Unseld, the post-Beethovenian genre of the string quartet is characterized, on the one hand, by the highest level of rationality, and for Berg is thus a way of both gaining distance and concealing his inner self (cf. Unseld 2001: 203). On the other hand, the string quartet is an intimate genre because it is intended for performance in a non-public space. For these reasons, it can be used to express intense subjectivity (cf. 331, n. 47). Unseld concludes that Berg redirected his intense emotions into extreme aestheticization (203).

It should also be pointed out that a similarly high level of aestheticization is already present in Stefan George’s translation of Baudelaire’s poem, the translation Berg used for his “song without words” in the Largo desolato. We have already seen that George’s aestheticism can also be located in the context of the feminization of writing. Baudelaire, however, was situated by Le Rider in a cultural context where the cult of mundus muliebris and misogyny coexist. This coexistence of antagonistic notions of the feminine follows a pattern similar to that of the coexistence of “spleen” and “ideal.” Karin Westerwelle (2007) has argued that Baudelaire did not seek to transcend “spleen” in favour of “ideal.” The production of art (the idéal) is guided by the perspective of temperament, “spleen.” On this basis, De profundis clamavi can be read as a descent (katabasis) into the hell of the “spleen.” Its first title was La Béatrix, and was later entitled Spleen. A dark inner vision and representing the ideal figure of the feminine seem to be connected, the two poles seem to be invertible (cf. Westerwelle 2007: 33–36).  

All this places Berg in an intermediate space – not between romanticism and modernism, which remains a popular characterization of Berg even today – but between different modernisms: the ascetic modernism of Schönberg’s twelve-tone system as opposed to the feminization of writing. In no way, then, is it a coincidence that we can find this ambivalence inscribed in the music of the Lyric Suite, precisely at the point where the poem exclaims “und dieser nacht: ein chaos riesengroß!” [And of this night, a gigantic chaos!]. In measures 31 and 32 (see example 1), we find a surface of sound which is itself in motion [in sich bewegte Klangfläche], a technique used in atonal music as a tonal area of resolution [Auflösungsfeld]. Schönberg uses one in the last movements of his second string quartet op. 10 (measures 93–99) (see example 2 and play the second audio example). Above this Klangfläche, the vocal part turns melismatic to the words “ich fühle wie ich über letzter wolke in einem meer kristallnen glanzes schwimme” [I feel as if above the last cloud swimming in a sea of crystal radiance] (T. 95–98) from George’s poem, Entrückung [Rapture].


Example 1: Alban Berg: Lyric Suite, Largo desolato, measures 31–32

LaSalle Quartet: Schönberg, Berg, Webern: String Quartets, CD 1; Track 8

Courtesy of Universal Music Hungary Ltd.



Example 2: Arnold Schönberg: String Quartet Nr. 2

(for soprano and string quartet), Op. 10, 4th movement, measures 93–98

© Copyright 1912 by Universal Edition A.G., Wien/UE 35543

LaSalle Quartet, Margaret Price: Schönberg, Berg, Webern: String Quartets, CD 4; Track 6

Courtesy of Universal Music Hungary Ltd.

As it is generally the case in Schönberg’s “Weltanschauungsmusik,” the combination of sound surface and an ethereal vocal part represents the idea of the supernatural in music (cf. Sichardt 1990: 7–27). In the preceding measures you can hear at the beginning of the second audio example above, the subject of the poem looks at a “sonnerfüllte klare freie” [sun-filled, open expanse] and climbs “über schluchten ungeheuer” [over enormous canyons]. The subject of De profundis clamavi, in contrast, is trapped in the “deepest abyss,” in darkness and chaos. Again, the antagonistic poles seem to be interchangeable, as they were in the title of Baudelaire’s poem. Using such a surface of sound not in atonal but in twelve-tone context, Berg places himself between Schönbergian male heroism and the feminization of music Le Rider associates with his Lulu and his Violin concerto: To the memory of an angel (1990: 162).

But should we, after all, call the Lyric Suite “music as autobiography” or “experienced music,” as Constantin Floros (1992: 99) suggests? I think we should not. If music cannot be narrative, it cannot be autobiography either. But if music can contain narrative elements as forces of meaning, it can contain autobiographical narrative elements, too. There can be autobiographical forces of meaning in music, and surely there are such forces in Berg’s music.

Floros uses the term “autobiography” in a naive way. He does not even ask the question what kind of narrative autobiography is. Thus he fails to take into account the role of both intertextuality and intermediality in this context. Like other narratives, the autobiographical narrative also circulates in a field of intertextuality on a textual level. Dealing with autobiographical narratives in music further complicates the case, because at least two different media, words and music, and various kinds of relations between them are involved. If we thus look only for an apparently simple and closed autobiographical narrative behind a musical composition and use it as the key to understanding that music, this will force us to entirely overlook the intertextual nature of cultural narratives connected to the narrative elements in music and their specific intermedial transformations. [5] Paradoxically, then, “music as autobiography” continues the practice of ruling out the narrative reading of music in terms of cultural narratives. With the aphoristic phrase “Better NO meaning at all than THOSE meanings!” (McClary 1997: 31), McClary reveals the subliminal motivation lying behind this practice. In relation to the Lyric Suite and its interpretation as autobiographical music, we may modify McClary’s aphorism: Better only AUTOBIOGRAPHICAL meaning than THOSE CULTURAL meanings! I am convinced, however, that we should read musical works as cultural texts, considering their whole range of intertextuality and intermediality.


Botstein, Leon (2010). “Alban Berg and the Memory of Modernism.” Alban Berg and His World. Ed. Christopher Hailey. Princeton: Princeton University Press, 299–343.

Danuser, Hermann (2001). “Nahe Ferne. Aufgehobene Dichtung in moderner Musik.” Stimme und Wort in der Musik des 20. Jahrhunderts. Ed. Hartmut Krones. Wien: Böhlau, 27–44.

Floros, Constantin (1992). Alban Berg: Musik als Autobiographie. Wiesbaden: Breitkopf & Härtel.

— (2001). Alban Berg und Hanna Fuchs: die Geschichte einer Liebe in Briefen. Zürich: Arche.

Gerlach, Reinhard (1985). Musik und Jugendstil der Wiener Schule, 1900–1908. Laaber: Laaber-Verlag.

Jannidis, Fotis (2003). “Narratology and Narrative.” What Is Narratology? Questions and Answers Regarding the Status of a Theory. Ed. Hans-Harald Müller and Tom Kindt. Berlin: Walter de Gruyter, 35–51.

Kramer, Lawrence (1991). “Musical Narratology: A Theoretical Outline.” Indiana Theory Review 12: 141–162.

Le Rider, Jacques (1990). Das Ende der Illusion: Die Wiener Moderne und die Krisen der Identität. Wien: Österreichischer Bundesverlag.

McClary, Susan (1997). “The Impromptu That Trod on a Loaf: Or How Music Tells Stories.” Narrative 5.1: 20–35.

Perle, George (2005). “Preface.” Lyrische Suite Für Streichquartett (1926). Neuausgabe von George Perle (inkl. der geheimen Gesangsstimme“). Wien: Universal Edition, XIV–XVII.

Redlich, Hans Ferdinand (1957). Alban Berg: The Man and His Music. London: J. Calder.

Rosen, Charles (1996). Arnold Schönberg. Chicago: University of Chicago Press.

Scott, Calvin (2007). “Ich löse mich in tönen ...” zur Intermedialität bei Stefan George und der Zweiten Wiener Schule. Berlin: Frank & Timme.

Sichardt, Martina (1990). Die Entstehung der Zwölftonmethode Arnold Schönbergs. Mainz and New York: Schott.

Unseld, Melanie (2001). “Man töte dieses Weib!” Weiblichkeit und Tod in der Musik der Jahrhundertwende. Stuttgart: Metzler.

Wagner, Nike (1982). Geist und Geschlecht: Karl Kraus und die Erotik der Wiener Moderne. Frankfurt am Main: Suhrkamp.

Wellmer, Albrecht (2009). Versuch über Musik und Sprache. München: Hanser.

Westerwelle, Karin (2007). “Baudelaires Rezeption der Antike. Zur Deutung von Spleen und Idéal.” Charles Baudelaire: Dichter und Kunstkritiker. Ed. Karin Westerwelle. Würzburg: Königshausen & Neumann, 27–75.

Wolf, Werner (2005). “Music and Narrative.” Routledge Encyclopedia of Narrative Theory. Ed. David Herman, Manfred Jahn and Marie-Laure Ryan. London and New York: Routledge, 324–329.


[1] This is an elegant argument but it certainly cannot be the last word about the relationship between music and narrative. Kramer seems to be thinking of nineteenth-century opera and musical theatre, especially of Wagner, when he states that we do not need to listen to the words but only to the music. Kramer mentions Wagner’s leitmotivs in this context (cf. 1991: 155), but he does not refer to their narrative function of transferring a good part of the narrative from the text to the music. This shows that his argument refers to a special case of intermediality and also to a historically limited epoch. Even considering this period, we may ask whether the way of reception suggested by Kramer, i.e. listening to the music and not to the words, is universal or whether it is one historically and socially determined listening mode among others. If we take for example Romeo Castellucci’s production of Christoph Willibald Gluck’s opera Orfeo ed Euridice in Vienna in 2014, it is hard to imagine that the audience would be able to listen only to the music. Castellucci paralleled the narration of the opera with that of the biography of a young woman in Minimal Conscious State (MCC): pictures of the woman listening to a live broadcast of the music on headphones in her hospital room are projected onto a big screen in the back of the stage. The audience gets to know the basic biographical information about the former ballet dancer and her family through short text inserts, while the camera moves through avenues alongside a grid and through narrow corridors until it arrives at her bed. A narratological analysis of this production would have to take several levels of intermediality into account.

[2] The notion of Bildung refers to a process of education and self-cultivation aiming at the development of a special kind of bourgeois subjectivity in the nineteenth century. This process is modelled in German bildungsroman as the growth of a young man to a character integrating aspects of rationality, emotion, aesthetics and social responsibility. The paradigmatic education novel in this sense is Johann Wolfgang von Goethe’s Wilhelm Meister’s Apprenticeship (1795–96). Several modern novels written at the turn of the century and in early twentieth century are based on this paradigm but show the crisis of this concept of Bildung. A famous example is Robert Musil’s novel The Confusions of Young Törless (1906).

[3] Berg’s composition contains several references to his love affair, his longing for fulfilment of this love and its impossibility. These references are coded with the help of number symbolism and German pitch names (for example A B for Alban Berg and H F for Hannah Fuchs) without any explanations and therefore invisible in the published score. Berg, however, annotated a printed score revealing and explaining these hidden references in different colours and writing the words of Stefan George’s translation of Baudelaire’s poem between the staffs of the last movement. George Perle, who discovered this annotated score, constructed what he calls a “secret vocal part” by extracting and, where indicated, transposing the notes assigned to the words (2005: XVI).

[4] In contrast to the case of Berg’s Lyric Suite there is no evidence that Mahler had such a “secret” programme in mind when he was composing his symphony. The programme “revealed” by Alma is possibly her own fabrication serving the purpose of popularizing the symphony.

[5] To model narrative as a transmedial cognitive frame allows us to talk about narrative elements in music. Music, however, does not signify in the same way as language and therefore does not narrate in the same way either. Albrecht Wellmer (2009) has pointed out that musical events are not signs like phonetic events in verbal discourse. While discourse establishes semiotic interrelations between signs, musical meaning is not based on signifying signs but on a field of significance (cf. 111). Because music, however, is also addressed to talking and reflecting animals (cf. 103), a discursive and reflective dimension is not something attached to it, but intrinsically belongs to it (cf. 105). Wellmer calls this phenomenon the “latent indermediality of music” (24). Thus, whenever music becomes narrative, inevitably intermediality is involved.