Recurring Hauntings and Trauma in Schubert's ‘Unfinished’ Symphony

Abstract

Schubert's interest in Gothicism is explored in numerous songs written between the 1810s and early 1820s and, in recent years, has served as an aesthetic agenda that some scholars have applied to his instrumental music. One notable exception is the ‘Unfinished’ Symphony (D. 759, 1822), a work whose thematic presentation and form have been frequently related to states of terror and horror, but rarely correlated further to Gothicism and never consistently so across the two completed movements. In light of this relative neglect, this article offers a Gothic reading of the symphony, finding correspondence with Gothic signifiers of ghostly hauntings and the ‘problem of closure’, and draws upon relevant literary criticism and psychoanalytic theory. As I show, the concept of psychoanalytic trauma – a concept widely deployed in current literary criticism to scrutinize repetitive patterns such as hauntings and circular temporality in Gothic literature – is especially instructive in terms of helping construct a richer understanding of the symphony.

Introduction: Ghostly Hauntings and Gothicism

The past is never dead. It's not even past.¹

Franz Schubert's interest in presenting terror and horror in music is well-known. This is demonstrated not only in his early Schauerballaden, lieder of classical mythological subjects, and other vocal and operatic works that contain scenes of sudden fantastical disturbance,² but, as commentators regularly emphasize, also in a number of instrumental works. Of these, the ‘Unfinished’ Symphony (D. 759, 1822) has received much attention for its perceived display of musical terror and horror. This is said by many to be palpable right in the symphony's opening phrase (Ex. 1, bars 1–8), whose ombra character as well as ill-defined, liminal metrical and tonal features immediately evoke an ominous and mysterious atmosphere.³ Still more pronounced is the way this phrase and motivic materials related to its return – almost inescapably – throughout the symphony's first movement, sometimes even sounding menacingly at a loud volume, in tutti orchestration, and with shuddering tremolo. For Xavier Hascher, this opening phrase, or ‘the “O” theme’ as he calls it, is ‘obsessive’ and it repetitively begets some intense ‘moments of crisis’, while Charles Fisk notes that ‘the tragic first movement … never break[s] the hold of its haunting and melancholy opening phrase’.⁴

Ex. 1 Schubert, ‘Unfinished’ Symphony, D. 759, mvt. i, bars 1–8

Notwithstanding the extensive reception surrounding the symphony's representation of terror and horror, surprisingly this work has rarely been correlated further to the Gothic, or so-called ‘dark-Romantic’, aesthetic with which it shares unmistakable stylistic resemblances.⁵ The Gothic in the late-eighteenth and early-nineteenth centuries, as Marjorie Hirsch perceives, is characterized by ‘the arousal of extreme emotions’, ‘disequilibrium and irregularity’, ‘rapid, frenzied movement’, ‘excess and exaggeration’, ‘interruption, fragmentation, disjointedness’, and ‘a sense of having been conceived in the heat of the moment’.⁶ Recent studies have also explored the way in which such generic Gothic characteristics are expressed musically in Schubert's ‘late’ instrumental works through a combination of the ombra topic, formal irregularities, repetitions, and allusion to ‘archaic’ styles (Fig. 1).⁷ Certainly, many of these stylized Gothic characteristics and musical traits correspond well to the musical terror and horror commentators have noted in the symphony's first movement. In line with this correspondence, moreover, the opening phrase or ‘the “O” theme’ (to borrow Hascher's phrase), given its ombra and recursive features, may even bring to mind a familiar Gothic trope – that is, ghostly hauntings. ⁸ In Gothic texts, for example, the ghostly figures, as though a past that resists passing, always obsessively repeat the same actions and haunt the same locations, exemplifying a looping, or ‘anti-clockwise’,⁹ temporality critics identify as quintessential to this genre.¹⁰

Fig. 1 ‘Gothic’ musical traits in Schubert's works

An enthusiast of Gothic horror,¹¹ Schubert was well acquainted with the theme of ghostly hauntings. Composed seven years prior to the ‘Unfinished’ Symphony, his setting of Ludwig Hölty's ‘Die Nonne’ (D. 208, 1815) offers an early example of responding to literary explorations of ‘anti-clockwise’ temporality and the supernatural. Hölty's ballad narrates the story of a nun who, heartbroken by the betrayal of her beloved knight, revenged herself by hiring assassins to murder him. She thereafter dragged his corpse out from the vault within the church where he was buried, tore out his heart and stamped on it so forcefully that the sound reverberated throughout the whole church. The ballad's closing stanzas relate that since her own death, the nun's ghost has been haunting that very same church, and every midnight repeating the same heart-breaking (both literally and figuratively) act:

In Schubert's setting, the recursive hauntings of these stanzas are underlined by a switch to a strophic structure (see the ‘Mässig, mit Grauen’ section in Ex. 2) from what was before through-composed (bars 1–155). In Schubert's song, as in Hölty's ballad, the ‘anti-clockwise’ temporality is further dramatized through the lack of a definite, final resolution, which insinuates a possible continuation of hauntings beyond the ostensible ‘end’ of the work – this is referred to in literary criticism as the ‘problem of closure’, a phenomenon not uncommon in Gothic texts.¹³ In the ‘Mässig, mit Grauen’ section of the song, for instance, in representations of the steadfast fixation of the nun's ghost, the pounding chords that return in the penultimate vocal phrase's accompaniment (bars 168–169) – a recall of the originary scene of violent stamping in bars 149–154 – eventually even overspill into the postlude following the final PAC (bar 171) of each verse. The way a dramatically charged gesture lingers on ‘after the end’ (in the post-cadential space), together with the continual weakening of the tonic arrival by the accented dissonance in these closing bars, suggest a never-ending sense of haunting.

Ex. 2 Schubert, ‘Die Nonne’, D. 208, bars 149ff.

In this article, I apply the ‘anti-clockwise’ temporality characteristic of Gothic ghost stories to the ‘Unfinished’ Symphony, to disclose its musical horror and terror in a new light. My discussion explores the possibility to interpret the symphony as ‘isomorphic’ to subjective states,¹⁴ understanding the work in terms of a virtual musical protagonist around whom the haunting story unfolds. While the capacity of an artwork to provoke suspense from analogy to a ghostly haunting invites – or commands – particular interpretative attention, more specific exploration will evoke the psychological concept of trauma. This human dimension to a supernatural tale is a staple of literary critiques of ghosts and hauntings. Applying such connection to ‘Die Nonne’, for example, suggests the obsessional repetitive behaviour of the nun's ghost as indicative of what psychoanalysts diagnose, from traumatized subjects, as an unconscious compulsion to relive a disturbance. More specifically, from the perspective of psychology in literature, this ghostly reoccurrence might also be recognized as symbolizing the traumatic past per se, especially in light of Cathy Caruth's perceptive conceptualization of trauma as conditioned upon an unrepresentable, absent-present experience that ‘returns to haunt the survivor later on’.¹⁵ This psychoanalytic literary paradigm of ghostly hauntings serves as a conduit for this article's focused examination of the highly loaded first movement of the ‘Unfinished’ Symphony. But this article also explores beyond the symphony's first movement to discuss what a Gothic-traumatic reading might reveal about the second movement and, furthermore, the symphony's contestable ‘unfinished’ status.

Recurring Hauntings: The ‘O’ Theme

The ‘O’ theme that is pertinent to many discussions of the symphony's horror and terror forms the first focus of my exploration. Hascher notes that ‘the symphony seems to start ex nihilo without our being able to say whether there is any link between the opening bass-line theme and [the] new beginning’ from bar 9.¹⁶ A closer look reveals that within the first eight bars there is some suggestion of an introduction (the slow rhythm and, especially, the long-held $\hat{5}$ in the phrase ending (bars 6–8), which echoes the dominant-function tonal goal of conventional introductions),¹⁷ but also a departure from normative formal models, one entails a different type of large-scale agenda. Despite its unmistakable introduction-like features, the opening phrase is included in the expositional repeat; indeed, this return seems to be a deliberate attempt to counter the non-appearance that is typically consigned to thematic introductions.¹⁸ The decision to return to this phrase, which initially conveys the impression of an introduction, after the first-time bar cannot be understood as merely a formal repeat, however, but as indicative of something that resists passing. The ‘O’ phrase, to borrow the descriptors of Hascher and Fisk, is ‘obsessive’ and ‘haunting’,¹⁹ and as I hear it, it embodies an ‘anti-clockwise’ temporality, not dissimilar to that of a ghostly haunting.

The ‘O’ theme creeps in again after the second expositional ending to launch the development section (Ex. 3, bar 114). Cast in the subdominant key (E minor), the theme departs from its previous ending (on a long-held $\hat{5}$) to descend instead by step to C♮, which is then sustained for seven bars (bars 122–128). For Fisk, this juncture marks that the development section has entered a distinct ‘realm’ or aesthetic level: ‘in its depth and near-stillness, and in the arrest of its motion on C, the unstable sixth degree, this passage evokes a descent into a mysterious, threatening realm – or, perhaps, a venturing forth to an outer limit’.²⁰ To apply a dualism widely regarded to be crucial to Schubert's music, this shift in aesthetic levels might be seen as one across a waking state and dream – that is, nightmare.²¹ In particular, the static, motion-arresting flat-submediant sonority underscoring this shift is both sonically and aesthetically closely comparable to Schubert's trademark use of the flat-submediant harmony, which numerous scholars elsewhere have associated with dreams and altered states of consciousness.²² If the development section is to be heard as a musical nightmare, as we shall see, this nightmare is determined by the haunting of ‘O’.

Ex. 3 Schubert, ‘Unfinished’ Symphony, mvt i, development section

Right from the development section's opening (bars 122–133), the tremolando bassline immediately delivers a sense of fear and shudder.²³ Here, moreover, the head motif of ‘O’ (‘motif x’) joins with a descending-semitone appoggiatura (‘motif y’, a two-note truncation of the original y in ‘O’ that now falls semitonally) to form a four-bar unit (Ex. 4b) which, as though some alienating ghostly whisper, sounds softly but incessantly between the violins, dovetailing with the violas and bassoons. This haunting music soon intensifies as x – now inverted and encompassing G–F♯–E in the high register – relentlessly blares against the F♯ bassline and the sforzando syncopations in the inner voices to declaim a dominant ninth harmony (bars 134–145). This passage, lasting over a total span of 12 bars, leads forcefully to the first climactic moment of the development section (bar 146) and, thereafter, ushers in the next episode of the nightmare.

Ex. 4 Schubert, ‘Unfinished’ Symphony, motivic derivation of the development section

In the new episode, a tremolando arpeggio featuring the rhythm of x (Ex. 4c), after its first appearance in C♯ minor (bars 146–150), returns twice in diminished harmonies to quash an expected cadence that a secondary group-based unit²⁴ twice sets up (bars 154–158 and 160–165). As the secondary group-based unit eventually arrives at E minor (bar 170), furthermore, any sense of cadential resolution is shattered by its elision with the fortissimo and tutti reprise of the entire ‘O’ theme (bars 170–176), which culminates with the development section's second climax.²⁵ The constant threatening returns of ‘O’-based music, together with the vividness of bold harmonic progressions, cadential evasions, and intense shudder (tremolos) typically associated with the ombra, stimulate a heightened sense of Gothic horror and hauntings throughout this episode.

As the second climax unfolds, the hauntings of ‘O’ continue through two contrasting passages featuring respectively an extended motif x and motif y (Ex. 4d), both played canonically between the low and high registers and in loud volume throughout (bars 176–183 and 184–194). In addition, the way this episode is brought back to E minor – the key with which the development section starts – which it leaves (bars 176–183) and once again circles back to (bars 184–194), highlights a particular sense of stasis and confinement. In contrast with a Bildungsroman trajectory, in which a return to the origin often indicates the attainment of a higher existential level,²⁶ the fact that the development section is still infused with ‘O’ would also imply a continued haunting and lack of transcendence. The contorted returns of thematic materials analysed so far in the development section conform to archetypal Gothic plots, where paths and stairs to escape always twist back on themselves, trapping the protagonist in the endless repetition of the same horror.²⁷

The nightmarish atmosphere, pervasive in the development section, is suddenly broken through with a fortissimo fanfare eruption in D major (bar 202), the relative major that is never truly attained in the exposition. Such ‘triumph is illusory’, however (as Fisk points out),²⁸ since there is no modulation established and only a slip from E minor to D major, both in first inversion, by way of preparation. The fanfare topic is also twice undercut by a whispering F♯-major harmony (bars 205 and 209), with the latter elongated to form dominant preparation for the recapitulation, where the primary group and B minor tonic return, and a darkened realm once again takes hold. In reflecting on the journey of the development section to draw extensively on motifs x and y from the ‘O’ theme, its final gesture of dominant prolongation (bars 209–217) cast around the same $\hat{5}$ that informed the original end of ‘O’ (‘motif z’, bars 6–8) can be seen as indicative of another musical haunting (Ex. 4e). The similar way in which both the F♯ harmony here and the prolonged F♯ in the opening phrase act as a long-held dominant that then resolves to the accompaniment of the primary group also lends corroboration to this correspondence (compare bars 208–218 with 6–9).

The haunting of the first movement with ‘O’, however, does not end with leaving the development section behind and its nightmarish vision. Like a déjà vu, in uncanny resemblance to its earlier returns following the first and second expositional endings, the all-too-familiar ‘O’ theme reappears (bars 328ff.), precisely after the primary, secondary and closing groups are retraced in the recapitulation. The theme's continuation to form the coda suggests a more sustained vision of a nightmare, one that not only resembles the expressive intensification of the development section, but also, just like that section, concentrates on little else other than what is derived from ‘O’. While a recursive, anti-clockwise quality is never described as such by Hascher, he notes that the continuous return of ‘O’ does not suggest any ‘logical continuity and ‘formal teleology’. In particular, ‘we realize that instead of advancing, as we thought, we are in fact in the same place. The “O” theme, we have seen, does not progress. It is obsessive, and all elements emerging from it are devoid of a sense of “becoming”’.²⁹ Taken further, the ‘O’ theme and other ‘O’-based materials, which jettison ‘progress’ and ‘becoming’ in favour of repetition and once and again arouse heightened musical tension, are also structurally and expressively analogous to the ghostly hauntings of Gothic texts.

So far, while I have focused on the multiple dysphoric and haunting returns of ‘O’, the movement's ‘anti-clockwise’ structure can still be illuminated through comparing Schubert's individual treatment of sonata form with archetypal models, and, in doing so, can further sustain a Gothic reading. As pointed out by Hepokoski and Darcy, between the mid-eighteenth and nineteenth centuries, it was common for composers to write sonata-form development sections with reference to the thematic ordering of the exposition, what the two theorists call ‘rotational principle’:

P is usually elaborated upon first, and the music may then move forward – though often the cycle may not proceed any further at all – perhaps to TR and thence to a selection from the exposition's part 2: either C-material alone (the most common choice) or something from S preceding the possibility of C-material.³⁰

Following Hepokoski and Darcy, any sonata-form movements in which a rotational development section is featured would be at least ‘tri-rotational’ in formal layout, with the exposition and recapitulation constituting two outer rotations and the development section enacting one or more inner rotation(s). For them, therefore, even the formal processes of the archetypal classical sonata form may already be considered cyclical. As we know, however, most rotational development sections only offer partial return of themes as presented in expositional and recapitulatory rotations. As a result, they might sound seem to be causing a breach with, rather than sustaining, the cycles of the outer sections.

In the development section of the first movement of Schubert's ‘Unfinished’ Symphony, while primary- and secondary-group derivatives make an appearance, they cast a shadow on what is absent: a fuller representation of the themes from these groups (Ex. 3, bars 150–184). The formal procedure here, as I have shown, instead foregrounds thematic-motivic materials drawn from the introductory ‘O’ that precedes the primary group. Thus rather than producing inner rotations, the development section can be more readily understood as featuring a non-rotational procedure, one based on ‘O’.³¹ With the outer sections, Schubert provides an example of Hepokoski and Darcy's Type 3 sonata, whereby there is a double-rotation of two grand cycles: ‘introduction or P⁰ + exposition’ and ‘the development + recapitulation’.³² Importantly, this meta-structure softens any sense of the development section breaking off from the exposition and recapitulation in not following their rotations exactly, and its commitment to using ‘O’ throughout helps to elevate the opening phrase to a level of formal significance. In this reading, Schubert's sonata form can be seen as containing two cycles (the opening phrase-exposition complex and the development section-recapitulation complex) that feature not only the same rotational principle (‘O’, P, TR, S, C) but also a strong level of parallelism between the exposition and recapitulation, where thematic material and their ordering are practically identical. Taking the repetition of the opening phrase (‘O’)-exposition (P, TR, S, C) complex prompted by the first-time bar repeat, and retaining this thematic ordering across the development-recapitulation, provides three appearances of essentially the same rotation (Fig. 2). Containing the movement within this cyclical structure brings further attention, albeit on a grander scale, to a sense of ‘anti-clockwise’ quality.

Fig. 2 A cyclical reading of the formal structure of the first movement of the ‘Unfinished’ Symphony

In keeping with the rotational pattern so far in the movement, the coda starts with the ‘O’ theme (Fig. 2). This repetition and placement give further emphasis to a sense of fixedness, to the constant looping of the same events or cycle of events. From a hermeneutical standpoint, this movement could be comprehended as featuring an indefinite looping process that begins with and is always re-directed to ‘O’. This structural circularity is emphasized even more strongly in Schubert's original piano sketch for the symphony, in which the use of a much briefer coda, whereby the reprise of ‘O’ immediately leads to an assertive ending (Ex. 5), also promotes a greater affinity between the movement's beginning and ending. However, in the final version of the symphony, with the coda treating the ‘O’ theme in similar fashion as the development section, there is opportunity to claim a return to the haunted nightmare as embodied in that section. Drawing on the notion of musical subjectivity, the first movement of the symphony could be heard as representing a virtual protagonist who is constantly haunted by the same ghostly presence in nightmares – that is, recurring nightmares.³³

Ex. 5 A reproduction of Schubert's original sketch for the ‘Unfinished’ Symphony, mvt i, coda

The process of recurring nightmares suggested for the first movement of the ‘Unfinished’ Symphony elicits further connection with psychoanalytic conditions, specifically ‘traumatic neuroses’ identified by Freud, in which dreams have the peculiarity of ‘repeatedly bringing the patient back into the situation of his [sic] accident’.³⁴ These experiences, which marked for Freud ‘the only genuine exceptions’ to his grand theory of dreams as wish-fulfilment advanced in The Interpretation of Dreams (1899),³⁵ offer an important interpretative lens for further contextualizing hauntings and recurring nightmares in this movement. Indeed, the repetitive pattern across this movement, not dissimilar to one in traditional Gothic stories or ‘Die Nonne’, might be construed as symptomatic of an originary traumatic occurrence – a process vividly re-enacted in the movement's secondary group.

Reliving the Trauma in P, TR, S, and C Groups

In preparing for the secondary group, Schubert instantly arrests the preceding momentum of the primary group with a brief transition that sounds like the gesture of a horn call (Ex. 6, bars 38–42) – a common symbol of distance and halcyon memory in German Romanticism.³⁶ As if called into existence as an idyllic memory from this gesture,³⁷ the secondary group enters (bars 42ff.) pianissimo in a straightforward texture of melody–accompaniment in the flat-submediant key (G major), the ‘inward’ key par excellence. ³⁸ Yet just as in Schubert's songs, where pleasant dreams and memories are liable to be shattered,³⁹ it soon becomes clear that what is ‘remembered’ in the secondary group is later to be crudely destructed. For Susan McClary, this offers a retrospective understanding of the secondary group as a ‘vulnerable lyrical subject, which is doomed to be quashed’.⁴⁰

Ex. 6 Schubert, ‘Unfinished’ Symphony, mvt i, bars 38–71

Such annihilation is initiated through cadential evasion and breakdown of phrase: following the elided PAC (Ex. 6, bar 53) brought by the lyrical theme's first statement,⁴¹ this theme's restatement raises the expectations of a ‘one more time’ PAC,⁴² only for a bar of silence to intervene suddenly (bar 62), ‘as if in the face of a horror’.⁴³ Without any anticipation, three chords explode throughout most of the orchestra with sforzando and tremolando articulation, violently blaring out C minor (bars 63–64), G minor (bars 65–66), and E flat major (bars 67–68), followed by an augmented sixth harmony (bars 68–70) and a culmination on a diminished seventh (bar 71).⁴⁴ As if responding to this shocking series of events, the previously unassuming secondary theme returns with an agitated character (bars 71–93), torn between the high and low registers to encapsulate ‘the “victimized” remainder of a multimodular S that, with much strain, pieces together the shards of a shattered S¹ to secure the EEC in bar 93’.⁴⁵ Following the immediate subsidence from the diminished seventh peak (bars 71–72), Glenn Stanley applies the metaphor of ‘a shocked person's response to a terrible human action or natural event’, suggesting that ‘one can … imagine an individual who, through overwhelming horror and fear, can no longer speak and is reduced to the most primitive, perhaps even non-verbal articulations, an inchoate moaning’.⁴⁶

To return to the chordal explosions preceding the ‘aftermath’, these moments (Ex. 6, bars 63–71) elicit highly descriptive responses from different scholars, including Stanley, who finds here ‘a completely unforeseen intrusion … a catastrophe’.⁴⁷ Regarding this music, also, Hepokoski and Darcy state that the way it starts ‘could not be more negative in its implications’; for Hascher, it denotes a ‘moment of crisis’ and ‘can be described as hysterical in nature’; while for Davies, its ‘sonic force and expansiveness, in conjunction with its sudden entrance, serve to overwhelm the senses, defying logical explanation’.⁴⁸ To my ear, it is not just that the extreme musical force here (minor harmony, loud dynamics, shuddering tremolos) strongly evokes certain characteristic features of Gothic horror; also, and more broadly, the level of intensity gained from the lack of preparation and suddenness of these explosive chords constitutes a close musical approximation to what Freud called ‘fright’. His diagnosis of this is particularly apt for Schubert's passage and gives a more focused interpretation from what other scholarship only marginally hints at: ‘the state a person gets into when he [sic] has run into danger without being prepared for it; it emphasizes the factor of surprise’.⁴⁹ This condition, as Freud explained in his foundational text on psychoanalytic trauma, Beyond the Pleasure Principle (1920), is the fundamental psychological cause of traumatic neuroses. In a nutshell, fright causes the raw, unprocessed stimuli of an accident to breach through the mind's ‘protective shield’ and strike it directly.⁵⁰

Exploring Freud's diagnosis further from a cognitive perspective, Caruth perceptively argues that the state of fright in traumatic neuroses is in fact a disruption in temporal experience:

[Fright] is not simply, that is, the literal threatening of bodily life, but the fact that the threat is recognized as such by the mind one moment too late. The shock of the mind's relation to the threat of death is thus not the direct experience of the threat, but precisely the missing of this experience, the fact that, not being experienced in time, it has not yet been fully known.⁵¹

According to Caruth, this ‘unclaimed experience’ or lack in trauma, while lurking beneath consciousness in a state that eludes either representation or comprehension, nonetheless continuously demands a proper response from the subject, resulting in different types of traumatic repetitions, which tell ‘the story of a wound that cries out, that addresses us in the attempt to tell us of a reality or truth that is not otherwise available’.⁵² As Caruth states, ‘trauma is not locatable in the simple violent or original event in an individual's past, but rather in the way that its very unassimilated nature – the way it was precisely not known in the first instance – returns to haunt the survivor later on’.⁵³ This Caruthian understanding of trauma as an absent–present experience that ‘cries out’ and ‘returns to haunt the survivor later on’ – a concept employed widely in literary criticism in discussions of ghosts and ghostly hauntings – also provides a fitting model for understanding the musical hauntings in the symphony.

Taking a closer look, analysis reveals certain motivic resemblances shared between these explosive chords that constitute the traumatic encounter and the ‘O’ theme that I have associated with ghostly hauntings. In the former music example, the underlying bass line (C–D–E♭) articulates a tone higher the same ascending stepwise motion as the head motif (‘motif x’) of the latter example (B–C♯–D) (cf. Ex. 6, bars 63–68 and Ex. 4a, bars 1–2).⁵⁴ An even more revealing connection between later encounters of the two can be observed in the episode that follows right after the first climax of the development section. In this episode (Ex. 3, bars 146–170), the way the cadential expectations raised by the secondary group-based figure being repeatedly disrupted by the rhythm of x (bars 154–157 and 162–165) and motif x (the ‘O’-reprise at bar 170) distinctly re-enacts, as though in a manner of compulsive repetition, the originary traumatic moment, where these explosive chords violently quash the cadential progression of the secondary group. Viewed from a psychoanalytic standpoint, it might be said that it is precisely the horror in these explosive chords – one that became clear to the consciousness only one moment too late – that suggests a return of a ghostly presence, as if haunting a virtual protagonist repeatedly in nightmares.

As Freud reported, ‘dreams occurring in traumatic neuroses have the characteristic of repeatedly bringing the patient back into the situation of his accident, a situation from which he wakes up in another fright’.⁵⁵ Due to the unassimilated nature of the traumatic event, in repetitions, the subject does not experience it as something in the past (like memories of non-traumatic events) but relives it as though the same event is encountered in the present over and over again. Elaborating upon Freud, Caruth elucidates that

If fright is the term by which Freud defines the traumatic effect of not having been prepared in tim[e, then the trauma of the nightmare does not simply consist in the experience within the dream, but in the experience of waking from it. It is the experience of waking into consciousness that, peculiarly, is identified with the reliving of the trauma.⁵⁶

According to Caruth, the trauma of traumatic nightmares is, therefore, constituted not simply by the nightmare per se but also by the moment of awakening, the very ‘enigma’ of ‘having survived, precisely, without knowing it’, thereby missing, once more, the chance for addressing the threat properly.⁵⁷ If the recurring nightmares of ghostly hauntings in this movement (the ‘O-based’ sections) represent the symbolic re-enactment of the traumatic encountering of threat, then, we might say, the fright of reliving of the trauma that concerns Caruth would always occur in the music that follows these nightmares: that is, the primary group.

The unsettling psychological state upon traumatic awakening is perceptible right from the first appearance of the primary group (bars 9–21), following the introductory ‘O’ (bars 1–8). The shivering semiquavers in the violins and pizzicato primus paeon (long–short–short–short) rhythm in the low strings directly strike Spitzer as resembling the ways the human conditions of ‘trembling’ and ‘pounding heartbeats’ are usually represented in modern film scores, projecting ‘a frightened response to the frightening introduction’.⁵⁸ In a similar light to my reading of nightmares and awakening, Spitzer hears the opening phrase as ‘an object of Fear, whilst the first group is the fearing subject’.⁵⁹

The intensity of fright is, in addition, eloquently expressed in the way the tonal progress, as though within a Gothic loop – or as Spitzer puts it, ‘running on the spot’⁶⁰ – repeatedly points towards yet fails to secure the relative major. In bar 17, the restless, dreadful feeling is significantly subdued with the continuation of the primary theme there initiating a modulatory progression that ultimately leads to the bright tonality of D major through a briefly reassuring PAC (bar 20). However, as if being gripped by a sudden fear upon a flashback of the nightmare, a dominant ninth chord of the tonic – the very same harmony that culminates with the first climax of the development section – erupts sforzando right after the arrival of this cadence, forcefully reinstating the dark realm of B minor. The rest of the primary group features two further instances where a musical process starting in D major is violently interrupted by a dominant harmony soon after (bars 28–29 and 35–38), the latter instance then contributing to an IAC in B minor (bar 38), emphatically to round off the primary group. The futility of breaking through a distressing minor mode is similarly emphasized in the recapitulation. While in the exposition the primary theme in its first statement (bars 13–20) at least attained a PAC in D major, in the recapitulation other potential junctures for establishing a relative major – bar 229 for B minor → D major and bar 238 for E minor → G major – are both abandoned, with F♯ minor eventually being forcibly established via a full cadence (bar 252). Hepokoski and Darcy remark that ‘the desire to be emancipated from minor into major constitutes the basic narrative paradigm … of minor-mode sonata form’.⁶¹ In constantly failing to break free from the shackles of the minor mode, the primary group nonetheless represents vividly the devastating impact of being swamped by fright every time trauma is relived.

A comparable musical expression of traumatic fright can also be heard in the closing group (bars 104–110 and 322–328), following the secondary group that I read earlier as another re-enactment of the trauma. The closing group starts with a tutti outburst of an unharmonized B, which instantly shatters the PAC attained at the end of the secondary group (bars 104 and 303). This gesture, combined with the pizzicato descending line that follows (bars 105–110 and 323–328), convey the image of an intense shock and aftershock that also effectively set up the frightened psychological response of reliving the trauma.

Taking into account the projection of the circular temporal trajectory in the overall formal structure of this movement, it might be said that our virtual protagonist is imprisoned by his or her trauma, compulsively repeating the cycle of 1) the ‘O’-based sections: being haunted in nightmares by the unaddressed trauma that takes the shape of a ghostly presence; 2) the primary group: overwhelmed by fright upon awakening; 3) the transition and secondary group: recalling the disturbing memory of the originary traumatic experience; and 4) the closing group: another frightened response to the reliving of the trauma. A model of this cyclic procedure is produced in Figure 3 (see below). From the perspective of psychoanalysis, especially the Freudian tradition, the compulsion to repeat is not only a symptom of trauma but also a means for sufferers to build towards attaining a recovery: the goal of repetition is to work through the traumatic impact by retrospectively developing the preparedness for the originary traumatic event, in order to bring closure.⁶² However, as we shall see in the next movement of Schubert's ‘Unfinished’ Symphony – where a literary Gothic agenda may again serve to illuminate – such closure may not always be possible.

Fig. 3 Model of traumatic subjectivity in the first movement of the ‘Unfinished’ Symphony

Unfinished Hauntings and the Second Movement

The literary critic J. Hillis Miller declares that ‘no narrative can show either its beginning or its ending. It always begins and ends still in medias res, presupposing as a future anterior some part of itself outside itself’.⁶³ This ‘problem of closure’, while applicable to narratives in different literary genres, is especially salient to Gothic literature. The intensity of recursive hauntings in the Gothic genres is often so strong that it becomes impossible to resolve them completely by the end.⁶⁴ Thus, instead of offering a definite closure, many Gothic narratives end in ways that draw attention to the difficulties of doing so. This is often the case with German ghost stories of the nineteenth century, of which, as Muriel Watkins Stiffler points out, ‘a sense of incompleteness’ or something left unaccounted for by the end, is a trademark feature.⁶⁵ Importantly, as Rebecca Martin argues, the problematic closure in Gothic literature, in conveying a sense of surplus anxiety, could potentially also imply that repetition may continue even beyond the ostensible ending of the text and that ‘such spectacular scenes of suffering’ carry on unabated (as in my analysis of ‘Die Nonne’).⁶⁶ The problem of closure in Gothic narratives, therefore, fundamentally calls into question Freud's therapeutic model for traumatic repetition, pointing instead to a darker truth that repetition itself gives no guarantee of restoration to a pre-traumatic stage. Sometimes, once traumatized, the subject can do nothing but undergo repetition that knows no definite end.⁶⁷

One common way to formulate a problematic ending is through the process of subversion, whereby some later moment of contentment initially promises a resolution, or ‘happy ending’, is suddenly withheld by an insinuation that haunting or suffering might not have been really over.⁶⁸ Although not a Gothic narrative, Schubert's literary tale ‘Mein Traum’ (1822) written just three months prior to the ‘Unfinished’ Symphony presents a comparable subversive closure following a highly repetitive trajectory. In the first part of ‘Mein Traum’, the protagonist, having been in conflict with his father, undergoes a repetitious process of exile and homecoming. While familial tensions are mostly resolved in a heavenly scene in the second part of the tale, a sense of redemption is subverted, and a lingering feeling of residual conflict remains: ‘I felt as though eternal bliss were compressed into a single moment. My father, too, I saw, reconciled and loving. He clasped me in his arms and wept. But not so much as I.’⁶⁹ This subversive ending, like one characteristic of Gothic literature, could imply to the reader the potential occurrence of an ‘after-the-end’ repetition. Regarding this ending in ‘Mein Traum’, Maynard Solomon simply argues that ‘there can be no reasonable expectation that the fabulous reunion with which it closes is other than a temporary prelude to a renewed separation’.⁷⁰ A few commentators have noted certain programmatic correspondences between ‘Mein Traum’ and the ‘Unfinished’ Symphony, in particular relating the tale's final scene of redemption or reconciliation to the symphony's second movement.⁷¹ As I shall show, although from a slightly different perspective, the Gothic stance taken in this article, by raising the possibility of exploring the second movement in light of a problematic closure following the repetitious first movement, may also prompt a link between the two.

Cast in ABA′B'A″ form, the Andante con moto second movement opens in the bright key of E major with a small-ternary pastorale-like first group (A). The construction of this initial refrain, with pianissimo outer sections (bars 1–32 and 45–60) and the forte contrasting middle (bars 33–44), suggests a positive existential condition of, respectively, serenity and exuberant fanfare, sounding immediately comforting especially after the intense nightmarish loop of the first movement. Such reconciliation is soon subverted, however, by the recollection of traumatic passages of the first movement. With the elision of the concluding PAC of the first group with a C♯-minor arpeggio (Ex. 7, bars 60ff.), Schubert, as Fisk remarks, ‘explicitly returns to [the] traumatic moment’ of the first movement's development section (cf. Ex. 3, bars 146–150).⁷² Through this arpeggiated line, the arrival of the second group (B) brings a more prolonged reminiscence of the traumatic music of the first movement. Not only does the syncopated accompaniment unmistakably recall that of the first-movement secondary group, but its thematic melody, in outlining the interval of an ascending minor third (bars 66–69), offers yet another variant of the haunting x. Even more pronounced is the way in which in bar 96, after C♯ minor is brightened up in the enharmonic major (bar 84), this ‘x-based’ theme suddenly breaks in fortissimo with tutti orchestration to wrench the second group back to C♯ minor. This strikingly recalls the originary traumatic moment, as represented in the chordal outburst within the first-movement secondary group (cf. Ex. 6, bars 63–71).

Ex. 7 Schubert, ‘Unfinished’ Symphony, mvt ii, bars 60–99

The dysphoric character of the C♯-minor passage, which also returns in the recapitulation of the second group (B′), is twice dispelled by the returns of the assuaging refrain and the E-major tonic (A′, bars 142ff. initially, and A″, bars 257ff.). Later in the last refrain (A″), however, the idyllic evocation of the section is twice threatened by reprises of the arpeggiated line (bars 280–286 and 290–296). Even though the arpeggiated line is, in each instance, instantly suppressed without bringing back more traumatic reminiscence, its late reappearance sullies what would be otherwise a ‘happy’ and untroubled ending. Thus in this movement, while the hauntings of the ghostly, ombra-inflected ‘O’ theme appear to be finally over, the fact that the traumatic music of the first movement still lurks – in the arpeggiated line as well as the second group – clearly suggests that the trauma of the musical protagonist has not yet been properly resolved. Like the subversive ending in Gothic narratives and ‘Mein Traum’, this movement's closure, too, highlights a sense of surplus anxiety that leaves open the possibility for the listener to imagine the sinister loop of traumatic suffering to be restored after the symphony's first two movements.

Apart from the subversive ending of the second movement, perhaps what also appears to be apparently ‘problematic’ about the ‘Unfinished’ Symphony is that this work is literally left open-ended without providing a complete third movement (scherzo and trio) and a finale. And, significantly, unlike the various fragments and incomplete multi-movement instrumental works Schubert had left behind between 1818 and 1823 (the so-called ‘years of crisis’),⁷³ he seemed confident enough in his vision of the symphony to leave it unfinished: indeed, it was the only incomplete work for which Schubert had sought a public performance, from no less than the highly reputable Styrian Music Society of Graz, whose honorary members included Beethoven and Salieri. Such plans have raised questions amongst musicologists to whether Schubert himself considered the two movements as providing a complete symphony.⁷⁴ For those, on the other hand, who seek to tease more out of the ‘unfinished’ status of the symphony, the problem of closure I have applied to the work may support and offer further perspectives. To integrate the lack of remaining movements with a Gothic agenda, those found in equivalent literary examples are relevant:

In contrast to a work … that features the conscious attempt to close … any number of gothics produce endings which foreground the difficulty of closing, call attention to the weakness of their use of closural gestures, or display their incompletion by setting up a structure, a manuscript, for example, or a correspondence, and then not closing it.⁷⁵

Different from the widely held view that the two complete movements of the symphony produce a self-contained ‘finished’ form,⁷⁶ an alternative interpretation might claim the opposite: the work is deliberately meant to sound open-ended and ‘unfinished’ by the composer choosing not to complete anything beyond the two-movement structure. The latter reading, whereby full closure is not attained, therefore takes even further the already ‘problematic’ closure imposed by the traumatic reminiscences in the second movement: with the ‘non-closure’ of the symphony, the recurring pattern of suffering is indeed unfinished.

Relieving the Trauma in the ‘Great’ Symphony

In this article, I have read the ‘Unfinished’ Symphony in light of the Gothic trope of ghostly hauntings. I have demonstrated that Schubert's peculiar musical expressions and treatment of sonata-form procedure in the first movement of the symphony afford to the music a sinister looping mechanism, through which a series of haunted nightmares is represented. Furthermore, as I have endeavoured to show, interpretive approaches borrowed from Gothic literary theories and related psychoanalytic theories might also be utilized to illuminate this, so to speak, ‘Gothic’ symphony. If the musical nightmares, scrutinized through a psychoanalytic lens, are understood as symptomatic of trauma, then the ‘problem of closure’ provoked by the subversive ending of the second movement – and, arguably, by the open-ended, ‘unfinished’ status – insinuates to the listener that the trauma in this music might, in fact, never be properly resolved.

This outcome and sense of surplus anxiety the symphony enacts may entice a listener to explore beyond the work for resolution, a reflection of the desire engendered in readers of Gothic literature to plough through Gothic novels one after another in order to see conflicts being satisfactorily reconciled or ‘getting it right’ the next time.⁷⁷ To achieve potential closure, I shall briefly draw upon Schubert's next symphony, the ‘Great’ Symphony in C Major (D. 944, 1826).

The beginning of this work faintly recollects the ghostly opening phrase of the ‘Unfinished’ Symphony, with an eight-bar opening solo and unharmonized texture, slow pace, soft dynamics, and, particularly, stepwise ascending-third head motif. As this phrase is played by the horn (an established symbol for memory and distance), any connection between the two works becomes more intriguing. On closer inspection, however, contrast revealed is more telling: the opening phrase of the ‘Great’ Symphony not only starts more brightly with an ascending major-third ($\hat{1}\ndash \hat{2}\ndash \hat{3}$) in the middle register, but also produces a less ambiguous and more substantial introduction, to signify a more positive connotation than that of the ‘Unfinished’ Symphony. Further difference from the ghostly opening phrase of the ‘Unfinished’ Symphony, whose ambivalent, haunting qualities demonstrate the applicability of a Gothic narrative or ‘anti-bildungsroman’,⁷⁸ revolves around the very different structural use of the opening phrase of the ‘Great’, not lease as forming a full-blown introduction that returns heroically in the coda (bars 622ff.). John M. Gingerich accurately characterizes this work as a ‘bildungsroman’ in which ‘the chief protagonist … undergoes trials, adventures, and tribulations to emerge tempered, matured, and triumphant’.⁷⁹

In final comparison, if the ‘Unfinished’ Symphony evokes the unending repetition of suffering indicative of Gothic literature, the ‘Great’ Symphony, as ‘a true stream of strength and health’,⁸⁰ presents an optimistic sequel to or counterpart of that story, in which traumatic conflicts may finally be relieved.