3.1. Introduction

In a detailed discussion of sonata form, James Hepokoski (2002, pp. 91-154) has raised the issue of explanatory deficiency in our normative descriptions of first – and other – movements in sonata form. They lack, he contends, sufficient finesse and nuance to accommodate important works that either omit or depart from the usual formal procedures of sonata form. Those departures are often thematic ideas which turn up once and then never return in the movement, like the new idea at the beginning of the development in the first movement of Mozart’s ‘Hunt’ quartet in B flat major, K. 458 or his piano sonata in F major, K.332. Another kind of non-normative procedure is when a thematic idea comes back in the wrong key – or the wrong place – like the second subject in the finale of Beethoven’s E minor ‘Razoumovsky’ string quartet, op.59 no.2, which comes back in the right key – E minor – but in the ‘wrong’ place, in the development, not the recapitulation. In the first movement of the ‘Eroica’ symphony, let alone the first horn entering too early at the recapitulation, the most famous example of all aberrant procedures is the new E minor theme in the development.[1] None of these kinds of instances in important works, according to Hepokoski, are satisfactorily accounted for by existing descriptions of sonata form.

Perhaps some sense of that lack led Hepokoski and Warren Darcy (2006) to undertake their extensive re-examination of sonata form in Elements of Sonata Theory. The book contains many interesting observations in an extensive discussion of frequently used as well as non-normative procedures. But exhaustiveness does not necessarily render clarification. In two areas – multiplying sonata form types to five and using complex terminology such as primary-theme zone (P) and transitional zone (TR) in place of usual terms first subject and transition – the book tends to obscure rather than elucidate the fundamental principles of how sonata form movements operate.

It might be more accurate to say how we understand them to operate, because we conceptualize musical forms in a similar way to how we understand grammatical rules or species of animal, in terms of an underlying schema with many adherents and some notable exceptions. By knowing the characteristic traits of a particular group or category – the ‘equineness’ of horses, for example – we can determine which are close members, like racing stallions, and which more distant members, like zebras. It seems that our hard wiring is organized into these background conceptual grids, against which, as George Lakoff discusses (1987), we gauge the fit of individual cases and assess areas of departure. Making music theory – as an active endeavor rather than a reactive description – has, in some way, to parallel the perceptual process. Just as perception clusters and organizes undifferentiated incoming data into coherent units in space according to schemas of people, places and buildings, so music theory similarly clusters and organizes notes according to relevant criteria to make shapes and patterns in time. To construct meaningful reality, theory, in an abstract way, mirrors the processes of perception. If we try to construct theory as a description underlying reality, then each category is predicated on a basic underlying schema that proposes essential courses of action and against which individual cases can be evaluated: reduction not proliferation. This means that rather than a comprehensive encyclopedia to encompass all exceptions, such a schema provides a roadmap of probabilities against which we can assess the specific events of an individual work. As Thomas Kuhn (1962, pp. 17-18) notes: “To be accepted as a paradigm, a theory must seem better than its competitors, but it need not, and in fact never does, explain all the facts with which it can be confronted.” Just as we continually match perceptual events against expected implications at many levels[2] in order to make sense of our environment, so understanding unusual events in sonata movements is similarly an evaluative process against the background schema of sonata form.

Identifying the landmarks and the route – what an element is and how it functions in the structural design – requires a clear differentiation between the precise, explanatory criteria and temporal direction of scientific theory and the paradigms of music theory. In terms of temporal direction, a scientific theory has to be predictive. Based on a convincing, powerful explanation of the current behavior of events in a specified environment, it can extrapolate forward to explain related phenomena in the future. Scientific theory is expressed as symbolic logic and has to be testable, or, in Popper’s stronger version of this, falsifiable.[3] Music theory, by contrast, is not inductive and for the most part relates to past repertories. It is laid out as a set of mathematical, graphic or verbal statements governing structural components and relationships. Music theories that have proclaimed the direction of art works of the future have been, at best, polemical, and most of the time, wrong.

Descriptive rather than prescriptive, music theory – and this is particularly germane to sonata form – accounts for the normative procedures in a certain repertory or genre. Making those accounts as perceptive as possible also has to take into account that great composers, in the exploratory, inventive business of writing music, do unusual and unpredictable things. Rethinking a more focused view of sonata form, as Hepokoski demands, does not entail multiplying the types of sonata forms in order to provide a catch-all of exceptions. Rather the task seems to be to reconfigure the background description so that it better explains two different demands: the ground rules – the form’s organizational principles and normative procedures; and the elbowroom – compositional choice in realizing the implications of elements of the material, and flexible space for their imaginative solutions. These two constituents make up what is described here as scenarios of sonata form. Constructing such scenarios may enable us to gain new insight into the particular patterns in Schubert’s sonata form movements.

3.2. A Distinctive (Non-Beethovenian) Path

We have to recognize that Schubert’s principal individuality in the handling of this kind of form was the transference of the lyrical and its forms into the province of sonata form. (Salzer 1928, p.125, author’s translation).

Scenarios are essentially the scene sets of form. No idle play on words, the shared etymological root between scene, scene set and scenario reveals two related meanings: one is that scenarios frame and contextualize action, just as scene sets do; the second is that, just as scene sets provide identifiable markers of period and place, so scenarios of sonata form also reference markers of compositional practice. Unlike reductive abstractions, which strip down local and mid-level events to their underlying lines, scenarios build up descriptions of sonata form from the bottom upwards. In the process of realization from the abstract to the actual, such descriptions are impacted by a complex interaction of factors, including the increasing individualization of genres during the early nineteenth century;[4] and new ways of characterizing expressive content, as subjective reflections of the poet’s or composer’s temperament.[5]

History, taking the long-range perspective of events, tends to flatten details in the interest of tracing trends, particularly with the knowledge of hindsight. But suppose that if Google Earth were transported in a time machine back to Vienna during the first two decades of the nineteenth century, it would show all kinds of details – buildings, institutions and neighborhoods. Not just details of place: in trying to trace the strands of everyday musical life in the city, a complex temporal simultaneity of different composers’ ‘presents’ emerges within that spatial juxtaposition. Some of those ‘presents’ overlap through friends in common or performing musicians; others intersect through patrons, publishers or performances of works. On the other hand, some ‘presents’ may co-exist in nearby neighborhoods but, for social or temperamental reasons, have no point of intersection.[6] Despite proximity in the same city, there was no personal interaction between Schubert and Beethoven as there had been between Beethoven with Mozart and Haydn. Beethoven’s formidable ability and irascible moods, exacerbated by deafness, were a virtually unbridgeable gulf for Schubert.[7]

Affected by training, social environment and musical temperament, Beethoven’s and Schubert’s movements play out the sonata form paradigm with different temporal implications and in distinctive stylistic domains.[8] Beethoven’s middle-period sonata movements, in piano sonatas, chamber music and symphonies, are most characteristically grounded in tautly-written prime material, motivically-based development and the dramatization of return. Schubert’s sonata movements also emphasize the saliency of return, but quite different tonal paths inflect the compositional route. Rather than causative goals, Schubert’s use of keys is often digressive. Disjunct keys a tone or semitone apart turn the direction, and material, into unexpected tonal pathways. In Schubert’s Lieder, disjunct keys a tone or semitone apart are often used for memory or dreams to recapture lost time, shifting between tonal locations proximate in place but remote in relationships. ‘Die Winterreise’ contains many examples of such semitonal shifts which convey the effect of the physical and emotional ground slipping from under one’s feet, like the eerie side-slipping from E minor to D sharp minor in the first verse of ‘Auf dem Flusse’. The vocal line’s falling semitone B to A sharp on “liegst kalt und unbeweglich” (lying cold and unmoving), with a drop of dynamics from double to triple pianissimo, depicts a coldness that cuts to the heart in the symbiotic imagery of the frozen river and the narrator’s frozen emotions.[9] [Example 1] Towards the end of ‘Im Dorfe’ the falling semitone in the vocal line from F sharp to F natural on the line “ich bin am Ende alle Träumen” (I have come to the end of all dreams) effects the change from D major to B flat major, again underscored by a drop in dynamics to pianissimo.[10] This change seems to open up the abyss of desolation, because if there are no longer any dreams, there is no longer any hope.

Whereas poetry often contains ‘lacunae’– spaces of meaning in the text which the poet leaves open to the reader’s imagination – in Schubert’s Lieder music fills these spaces in one of three ways – through the delineation of time, place and action. In the first of these, time is not directed like an arrow pointing to the future. Instead, the present has permeable boundaries. Prompted by an encounter, event or image that evokes the past as a mémoire – “die liebe Farbe/die böse Farbe” – whose boundaries dissolve as the present opens onto the past through imagination and remembrance. Deflected by memory, the present becomes a ‘reconstructed’ time, filtered through subjective experience.

In the second kind of delineation, place, music fills the open envelopes of meaning by harmonically intensified images to convey both physical and experiential location. In ‘Gefror’ne Thränen’, the third song of ‘Die Winterreise’, at the climactic line “des ganzen Winters Eis” burning tears are transformed into hard, crystal shards of ice.[11] As well as tears/ice being one of the primary images of the cycle, crystals were regarded in late eighteenth and early nineteenth century philosophy and ‘Naturphilosophie’, as a system of order in a microcosm of nature (Cassirer 1981: 315). The process of crystallization, as the moment when fluid suddenly acquires definite shape, was a metaphor for the sharp clarity of understanding when the shape of a thought suddenly becomes crystal clear – the lightning bolt of knowing, in this case, the knowledge of despair.

Action, as the third route of expressive delineation, is the musical depiction of intensified exchange between characters, like the three contrasted protagonists in ‘Erlkönig’ or, as Christoph Wolff (1982, pp.43-71) discusses, the distraught girl and the calming figure of death in ‘Der Tod und das Mädchen.’ Analogous to their characterization in Lieder, in Schubert’s sonata movements the shift to different strands or ‘voices’ of the narrative relates to these three kinds of delineation.[12] Transferred to instrumental music, their characteristics can be described as ‘poeticizing sound.’

Expressive deflections to remote keys, as means of evoking memory in the Lieder or casting a shadow over a poetic image, have their counterparts in the instrumental music. Disjunct keys, like the flat submediant or keys a semitone apart, similarly open up expressive domains outside the existing narrative of the movement. In the impromptu in A flat major, op.90 no.4, with its mixed modal coloring of A flat minor and major, the middle section in C sharp minor (enharmonically the minor subdominant),[13] while formally the center of a ternary movement, functions as an interpolation, displacing the opening arpeggiated material (see Gülke 1991: 204). The C sharp minor section, with its remote key and intensely characterized melodic line over insistent left hand repeated chords, is like an impassioned recitative, seemingly imported from outside the existing narrative and emotional domain of the piece and thrust into its center. In the C major ‘Wanderer’ fantasy, the C sharp minor variations are similarly a large-scale digression set at the center of the work,[14] a digression so extensive that it has its own internal form, a set of variations, but its expressive delineation within the context of the fantasy is the reverse of the impromptu. After the incisive rhythmic flourish of the fantasy’s C major first section, the variations are a withdrawal into the isolation and melancholy of the ‘Wanderer’ emotional landscape. The incessant dactylic rhythm – long, short, short – which characterized the extrovert opening C major section, is now pulled back into almost slow motion. The C sharp minor variations are ‘outside’ in two distinct but interrelated senses: one is poetic, the isolated ‘Wanderer’ as ‘outsider’ to society and cut off from connections to it, now transferred to the instrumental location of the fantasy; the other is musical, as material brought in from ‘outside’ the movement’s tonal frames of reference and connectivity.

If such gestures of ‘poeticizing sound’ inflect the tonal design and expressive action of the piano fantasies and impromptus, they also operate on both small- and large-scale in Schubert’s sonata movements. The ‘modus operandi’ of disjunct keys by side-slipping to a remote tonal area in the fantasies and impromptus is not only replayed in the sonata movements, but is recontextualized within the paradigm of sonata form. It accordingly appears as a tonal deflection which delays the second subject key on the large-scale; and as a harmonic deflection within the second subject key on the small-scale. An example of the latter can be seen of the first movement of the C minor sonata, D. 958, where the unmediated shift from E flat major to D flat major is underscored by a drop in dynamics from piano followed by crescendo to pianissimo. Although D flat has been subtly hinted at in the preceding passing tonicization of A flat over an E flat eighth-note pedal, the downward shift to D flat eclipses the expected move to the subdominant but is subsequently restituted by the succeeding dominant resolution in E flat. [Example 2] Although there are precedents for such unmediated shifts up or down a tone or semitone in sonata movements, they are usually deployed for a repeated statement of the first subject – the first movement of Haydn’s string quartet in C major, op. 33 no. 3 ‘The Bird’ or the first movements of Beethoven’s ‘Waldstein’ and ‘Appassionata’ sonatas come to mind – but these tonal shifts do not occur within a thematic area so much as between thematic statements. Silence throws the repetition of the movement’s opening declarative statement on the succeeding pitch level into sharp relief. Haydn’s and Beethoven’s use of proximate unrelated keys serves to dramatize repetition as structural function, while Schubert’s use, similar to related contexts in Lieder, brings out their expressive and coloristic qualities. The parallel between Lied and Lied-inflected sonata processes can be extended even further, since the second subject of the C minor sonata first movement in turn initiates variations which resemble an instrumental strophic song in the context of a sonata movement.

Schubert’s large-scale inflective gestures effect digression in the course of tonal action over a more extensive span, not at the service of dramatic discourse but by using pivots that initiate a section of inward reflection and lyricism. As in the small-scale example, these deflective tonal planes often occur at the second subject, re-interpreting the characterization of the second subject as a lyrical remote ‘otherness’ rather than in terms of opposition or mere contrast. In the exposition of the first movement of the C major string quintet in C, D. 956, the preparation of G major for the second subject is deflected to E flat major in one of Schubert’s most magical moves: the first cello retains the pitch G while the second cello descends chromatically beneath it to E flat in a move of maximum economy and expressive impact. The unexpected tonal direction opened up by this connecting hinge seems to come from ‘outside’ the expected – and previously prepared – tonal contour of the movement, emerging instead into a different, imaginary landscape.

Together with his more expansive melodic style, it is these bends in temporal direction into lyrical reflection as much as digressions in tonal design that differentiate Schubert’s sonata writing from Beethoven’s.[15] When Beethoven refers back to an earlier part of the music, such as the skeletal scherzo quotation in the finale of the Fifth symphony, it is musical memory used as a dramatic tactic, to cast a shadow over the C major finale and temporarily eclipse its brilliance by the spooky return of the scherzo before its triumphant re-emergence. In the finale of the Ninth symphony the scena-like references to earlier movements are similarly dramatic, even operatic in purview, to review, and then summarily dismiss the music’s past in order to prepare for its future. Both the Fifth and the Ninth symphony references are temporary deflective dramatic devices in music that is essentially teleological.

More complex, and inward, are interpolative sections in movements from the private music of Beethoven’s late period – private as in piano sonata and string quartet by contrast with the public gestures in the symphonies. The ‘L’istesso tempo di Arioso’ section embedded in the fugal finale of the sonata in A flat major, op. 110, and the Cavatina from the string quartet in B flat major, op 130, whose intense ‘beklemmt’ middle section displaces the richly lyrical opening, and in turn vanishes when the Cavatina returns, like Prospero’s spirits, “leave not a wrack behind”. Both of these embedded sections are recitatives and are the more impassioned for their restraint. Both are in minor keys remote from their framing major key tonal contexts, and both are operatic in provenance, which allows a greater degree of flexibility (“phantasie”) in their respective movements. Neither of the movements is in sonata form.

In Schubert’s instrumental music, the loops into remote tonal and temporal regions created by hinge points and side-slipping are is not so much reactive< to an existing musical narrative as deflective from it, frequently digressing into pathways that seem to recall an earlier time, even if the musical events have not been actually heard before, such as the D flat major section in the A flat major impromptu, op.142 no. 2. This ‘out-of-timeness’ occurs when a different event-stream is interpolated into the predominant course of action and later merges with it. Characterized not so much by juxtaposition as by displacement, such temporal slippage in the Lieder is depicted when dreams of action and heroic self-image in ‘Am Feierabend’ in ‘Die schöne Müllerin’ are displaced by the miserable weakness of reality.[16] A comparable case of displacement in the trajectory of time in Schubert’s instrumental music occurs in the first movement development of the sonata in B flat major, D.960. The second time bar at the end of the exposition unfurls an extraordinary switch point from F major to C sharp minor. [Example 3] This in turn initiates the development in C sharp minor, opening with the first subject, which has a poignant strangeness in this remote key. After thirteen bars the tonal path shifts unexpectedly, via an interrupted cadence, into A major for an extended triplet segment which traverses G sharp minor, B major, B flat minor, ending with a decisive four octave descent to D flat major, as tonic major of the opening key of the development. The semitone shift up from G sharp to A that launches the music onto this new tonal plane parallels comparable places in both the ‘Wanderer’ and the F minor fantasy, except for differences in their respective formal contexts. Since the triplets and their underlying left hand trochaic figure in the B flat sonata are the movement’s second subject, the expressive effect of ‘remoteness’ and ‘out-of-timeness’ is effectively grounded in sonata form procedures.

Semi-tonal side slippage or tonal hinge points between remote keys can be seen to serve essentially the same expressive purpose in Schubert’s sonata movements as in his impromptus and fantasies – to deflect temporal continuity by contrasted interpolated material. The nature of those interpolations often resemble the characteristics of strophic Lieder – self-contained, lyrical melody and repetition of verses – similarly incorporated into the context of Schubert’s sonata form movements, with melodic sections repeated either exactly or with minimal elaboration in different keys. They may occur in any part of the movement but are most striking in development, as a highly individual treatment of the material that is meditative rather than discursive and additive instead of causative. The first part of the development of the first movement of the A major piano sonata, D.959, for example, replaces the idea of development as ‘working out’ by ‘reflecting on’. It is a static oscillation of four-bar units between C major and B major (bars 131-4), the C major segment ‘piano’, the B major, more remote in affect ‘pianissimo’. Each is a self-contained melodic unit with complementary phrases, and shifts, unchanging, between one pitch level and the other.

While Beethoven’s middle period works frequently use development techniques of contraction – motivic combinations in conflicted discourse, intensifying texture and propelling momentum towards the movement’s structural dominant, Schubert’s sonata movements characteristically release the center of action for a kind of meditative reflection on the movement’s melodic resources. There are exceptions to such reflective qualities in development, mostly in minor key movements, in the use of stretto and motivic intensification in the first movements of the A minor piano sonata, op. 42, and the ‘Death and the Maiden’ quartet. But even the first movement development of the C minor sonata, D. 958, is characterized more by exploration than argumentation, while its finale, like the first movement development of the A major sonata, D. 959, repeats melodic sections on different pitch levels, a technique transferred and transformed from Lieder.

The opening of this discussion considered how conceptual understanding of sonata form is based on a working paradigm with diverse foreground realizations. It is also consistent, with reference to Schubert’s works, with Hepokowski’s and Darcy’s view that: “A central premise of this method is the conviction that we must seek to understand the backdrop of normative procedures within the different zones or action spaces of the […] sonata.”[17] Predicated on such a background ‘game plan’, Schubert’s sonata movements import characteristics borrowed from other kinds of works in his output, particularly Lied and fantasy, rather than adhering to Beethoven’s model of sonata governance. These individual foreground realizations can be viewed from two perspectives, as scenarios of sonata form. One of those perspectives is structural – sonata form as reciprocity. The other is an expressive viewpoint – sonata form as narrative.

3.3. Scenario 1: Sonata Form as Reciprocity

By contrast, aesthetic comprehension seems to specify the category of reciprocity by reshaping time itself. This suggests that the imagination can directly intuit coexistence by stopping the forward flow of time. (Makkreel 1990:, p. 72, emphasis added)

We have seen how both scientific theory and music theory are abstract mirrors of reality. In Conjectures and Refutations, Popper (1989) discusses how a scientific theory is a powerful description, which, as closely as possible, approximates to truth – where ‘approximates’ indicates the degree and match of the theoretical framework to the phenomenon under discussion. The formula or theoretical description is accordingly the paradigm in terms of which scientific events are evaluated. Despite the different domains of inquiry, music theory is similarly paradigmatic and evaluative, to propose a perspective that may not have been explored before.

The present discussion of sonata form as reciprocity offers such a ‘way of seeing’ as John Berger (1973) would say, such as Rosen proposed for long-range dissonance in the exposition requiring extensive tonic weighting in the recapitulation as resolution, or Hepokowski and Darcy for the rotation of event sequences.

If music theory mirrors, in abstract, the reality of musical works, and in this case sonata form movements, then the description of sonata form as reciprocity also draws on the everyday sense of reciprocal exchange as the give and take of like kind. A formal description of sonata form as reciprocity needs to consider the definition of elements in the arena of exchange; and the identity of salient components, specifically the difficult issue of second subject.

Looking at definition first, the fundamental premise is that salient elements in the exposition of a sonata form movement reciprocally return later in the movement; and, as will be argued later, if they do not return, then they are not salient. It is important to bear in mind that a scenario of sonata form is like reconstructing underlying compositional principles. As a working paradigm, its foreground realizations can, and do, manifest different solutions.

In the formal description: Prime tonic (first subject) material and important non-tonic (second subject) material is initially presented in the exposition of a sonata form movement. Both prime tonic and non-tonic material – and the concomitant tonal relationship between them – will be brought back later in the movement in one of two ways. Either all the prime material is restated in the tonic, and usually, although not invariably, returns in the recapitulation. Or, a significant part of the prime material is brought back in the tonic, as part of a reciprocal tonal layout between its initial presentation and its subsequent appearance. This layout retains the intervallic relationship of the subject areas in the exposition although it may be inverted in the reciprocal return.

The “significant part of the prime material brought back in the tonic” later in the movement is usually the first subject, but not always. This description of reciprocity would accordingly encompass subdominant recapitulations, like the first movement of the ‘Trout’ quintet, where the exposition’s intervallic tonal layout of A major-E major is precisely reciprocated in the recapitulation by D major-A major.[18]

“Identifying important non-tonic material (second subject or second subject group): however compact the motivic shape, it needs to be long enough to delineate a key and have a melodic/motivic and rhythmic profile distinctive enough that it can be recognized on its return, even if rescored or with other secondary elements changed, such as dynamics.”

More succinctly, what goes out in the exposition, as the movement’s salient material, comes back in equivalent ways later in the movement as part of its structural and tonal articulation. The return, most of the time, is in the recapitulation.

If definition, in the description of sonata form as reciprocity, depends on the articulated profile of prime material reinforced in memory, then the identity of such material to play a strategic role in the movement – and surely this is what we mean by saliency – depends in turn on its return later in the movement as part of its essential ‘dramatis personae’, rather than a transitory player who does not impact the large-scale action. Even if a musical idea has a recognizable profile and key, it cannot be considered salient if appears just once, because the single appearance, like the idea opening the first movement development of Mozart’s sonata in F, K.332, is almost invariably local in function, rather than elucidating long-range relationships or setting up network connections. Reinforcement in memory is critical to the continued identity of character – both musical and dramatic – after an intervening absence. To be a main player in a drama, you have to come back, which is why the re-appearance of the Commendatore in the in the act II finale of ‘Don Giovanni’ is so powerful, even if he has been off the stage for the past two hours. Sonata movements also have their own plot, even if they are wordless, and their own logic. What defines the coherence of this logic is the differentiation of function between principal and secondary ideas – between the musical protagonists and the secondary cast, whose roles are extension or temporary moments of interest. A passing idea – such as the connective theme in G minor between the first and second subjects areas in the first movement exposition of Mozart’s piano concerto in C, K. 467 – does not impact the long-range design or the tonal plan that underpins that design, whereas a salient idea has, literally, a part to play. When a salient idea plays its role as a main protagonist in the movement, it defines a specific tonal location, and has a recognizable identity when it first appears. Coming back later in a reciprocal section of the movement in equivalent tonal territory, it reclaims that position in the compositional matrix – which is why the beginning of the recapitulation, whether the first movement of Beethoven’s E minor ‘Razoumovsky’ string quartet, op. 59 no.2, or Mahler’s ‘Resurrection’ symphony, is such a powerful point of structural return.[19]

Out of two strategies of return in the scenario of reciprocity, the first and most frequent brings back important exposition non-tonic material – the second subject or second subject group – in the tonic key in the recapitulation. The second option, of partial tonic return, occurs less often but in major works, where the second subject in the exposition is in a key other than the dominant in a major key movement, or the relative major in a minor key work. In the first movement of Beethoven’s ‘Waldstein’ sonata, for example, the chorale-like second subject is in E major in the exposition (a third up from the tonic) as is its repeat in the left hand overlaid by triplets.[20] [Example 4] The chorale-like theme returns in A major (the equivalent third down) in the recapitulation, but its repeat in A minor moves smoothly into C major for the repeat with triplets. A related case of reciprocal keys for the second subject in the recapitulation can be found in the first movement of Brahms’ Third symphony, where the second subject is in A major in the exposition (like the ‘Waldstein’ a third above the tonic). The recapitulation shows a similar kind of double re-presentation, the second subject presented in D major but the two ‘capping bars’ (bars 156-7) of the second subject are in the tonic, F major. As Gisilher Schubert (1994, p. 17) notes, the change of orchestral sonority between clarinets and violins for the two statements underscores the structural function of the ‘double’ recapitulation. [Example 5]

In these two examples, return of prime tonic material at the beginning of the recapitulation marks a place of salient structural articulation in the movement. Important non-tonic material is reconfigured in the recapitulation to reciprocate the original intervallic distance of the second subject key from the tonic. But a special case of reciprocal tonal return is Schubert’s subdominant recapitulations, where the first subject, not the second, returns in a key other than the tonic. In the recapitulation of the first movement of the ‘Trout’ quintet, after a seven bar dominant lead-up, the first subject in the subdominant, D major, enunciates the beginning of the recapitulation, its scoring, articulation and dynamics precisely replicating the exposition. Tonal layout in the exposition of a perfect fifth between the main tonal areas – the first subject in A major and second subject group in E major (E minor, D major, E major) – is reciprocated in the recapitulation, with the first subject in D major, the second subject group in A major (A minor, A major, G major, A major). Even beyond identity and tonal relationships between exposition and recapitulation, the curving tonal trajectory is also reciprocated along the arrow of time. In the exposition, the direction is away from the tonic as point of departure, whereas in the recapitulation it is towards the tonic as the goal of arrival.

A similar case reciprocity in the tonal planning between exposition and recapitulation can be seen in the first movement of the String Quintet in C, D. 956. The tonal ‘staging posts’ of the exposition are C major (1st subject), E flat major (2nd subject), G major (modified 2nd subject),[21] delineating a triad of tonal centers moving away from C major. In their reciprocal appearance in the recapitulation, prime thematic material returns in the keys of F major (1st subject), A flat major (2nd subject), C major (modified 2nd subject), forming a triad that moves towards C.

But in addition to reciprocal replay, another technique of return is deployed in the recapitulation. Parallel to the beginning of the movement, the recapitulation starts in C major with the movement’s opening material, its sustained pianissimo chords and turn figure, now rescored in the first cello, [Example 6a] [Example 6b] and, as at the beginning of the movement, the whole group is repeated on the supertonic. In the exposition, the extension figure had lead back to C major for the first subject’s rhythmically articulated prime material, whereas in the recapitulation this figure is rewritten so that the bass, impelled down by step, makes a strong arrival on F to initiate the subdominant recapitulation that is nested inside the tonic return.

In a striking example, the first movement of the string quintet demonstrates two strategies of return, which exemplify the two kinds of reciprocity. For the first of these, tonic reprise in the recapitulation, the C major pianissimo opening section returns, rescored, at the beginning of the recapitulation as tonic restatement. In the second kind of reciprocity, the strongly articulated fortissimo prime material that was presented in C major in the exposition, now returns in F major in the reciprocal subdominant layout.[22]

As if to underscore the importance of the movement’s structural elements, they are all re-presented in the coda: the sustained introduction with its turn figure, as the first element of the first subject, now fortissimo; the articulated prime material over the German sixth of C major, also fortissimo, leading to V; and the second subject in C major, to close the movement. Not only do all structural elements appear in the tonic, but new relationships are also revealed. The first element of sustained chords and turn figure, formerly pianissimo, now appears fortissimo and strongly articulated, showing a striking relationship between the two previously contrasted elements of the first subject, and bonding them as one group. Even more interesting is the slight rewriting of the melodic line of the second subject. This subtle change in the contour of the line (compare bars 323 and 430) now enables us to hear the complementary relationship between the sustained first element of the first subject and the second subject. The sustained opening contains F# which rises to G in its inner voice (2nd violin) to open the movement; the second subject in the coda has F natural, which falls to E in its inner melody (2nd violin) to resolve it. As melodic variants, they frame the opening and closing of the movement.[23]

The principle of reciprocity in sonata form accordingly enables us to describe unusual features in particular movements by differentiating between salient musical ideas, which play a formative role in the movement by their subsequent recurrence, and transitory ones, which do not impact the large-scale design.

Schubert’s sonata forms are shaped by distinctive compositional choice – the elbow room in the scenario of reciprocity – that can occur at any place in the movement, but found most frequently in development. As the opposite of intensification and causative goals, it is characterized by reflective melodic writing with long-breathed parallel phrases; and by deflected time, effected through extraordinary moments of tonal slippage between two different temporal realms. But the distinctive character of Schubert’s sonata forms cannot be entirely explained in terms of reciprocal tonal planning alone. Nested inside that reciprocal delineation of form is an inner narrative that draws on the landscape of the imagination. In order to access the contours of that inner world, its affective depiction needs to be traced back to Romantic poetic imagery in Schubert’s Lieder. The departure point for that inquiry is Goethe’s Mignonlied ‘Kennst du das Land’, as the longing for an idealized home. Like the search for Camelot, the journey home is both the dream of acceptance and the archetypal journey of wandering, a journey imprinted with image and illusion, rhyme and rhythm, yearning and despair.

3.5. Coda

Schubert’s realizations of sonata form show highly individual choices of tonal planning, melodic repetition, digression and reminiscence. Such techniques enabled him to pursue compositional strategies quite different from the predominant model of Beethoven’s middle period works, where taut construction, sharp polarities of subjects, and motivically based development are prime characteristics of sonata design. But the predominant model was not the only model. Beethoven’s late works reveal new paradigms of sonata writing. The first movements of the string quartets op.130 and op.132 show interpolation, mediant and submediant relationships and temporal bifurcation, in the alternation between contrasted strands of musical events streams.[39] The slow movements of op.127 and op.131, like the slow movements of the piano sonatas op. 109 and 111, are variation sets, at the heart of which is a profound inner stillness that we also hear in the slow movement of Schubert’s string quintet in C major. Whether the result of subconscious influence or independent discoveries in parallel compositional universes, both composers arrived in their later works at different, and distinctive, ways of reinterpreting form, and especially the compositional schemas of sonata design.

In his reconfiguration of sonata form, Schubert draws into its inner narrative the formal and expressive ‘Gestalten’ of Lieder. Transferred into the structural parameters of sonata form, these ‘Gestalten’ are recollections of the sonic worlds of his poetic imagery –- the desolate figure of ‘der Leiermann’ on the horizon, the longed-for land where “die Gold-Orangen glüh’n”, and the broken ecstasy of “Sein Kuss.” Just as Schubert brings back salient material in order to reinforce its structure, so affective elements, as ‘poeticizing sound’, make another kind of return. Transformed into the inner narrative of poetic form and expression, they infuse his sonata forms with interpolations, looped digressions on the journey and the wistfulness of remembering.