The most important principle of musical form, or formal type, from the Classical period well into the 20th century. This form is that of a single movement, not a ‘sonata’ as a whole; such a movement is most often part of a multi-movement instrumental cycle such as a sonata, piano trio or quartet, string quartet or quintet, symphony etc., or an independent movement like an overture or tone poem. Sonata form as such is less common in fantasies and the like, small movements, concertos and vocal music, but its principles may influence other features of form in such works. Though most characteristic of first movements in fast tempo, it often appears in middle movements and finales, and in moderate and slow tempo; hence the synonyms ‘sonata-allegro form’ and ‘first-movement form’ are best avoided.
A typical sonata-form movement consists of three main sections, embedded in a two-part tonal structure. The first part of the structure coincides with the first section and is called the ‘exposition’. The second part of the structure comprises the remaining two sections, the ‘development’ and the ‘recapitulation’. The exposition divides into a ‘first group’ in the tonic and a ‘second group’ in another key, most often the dominant. Both first and second group may include numerous different ideas; the first or most prominent theme may be called the ‘main theme’, ‘first subject’, ‘primary material’ etc., while the most prominent theme in the second group is often called the ‘second theme’ (or ‘subject’), whether or not it actually is the second important musical idea. The development (the misleading term ‘free fantasia’ is now obsolete) usually develops material from the exposition, as it modulates among one or more new keys. The last part of the development prepares the recapitulation. The recapitulation (or ‘reprise’; but see §3 (iii)) begins with a simultaneous ‘double return’, to the main theme and to the tonic. It then restates most or all of the significant material from the exposition, whereby the second group is transposed to the tonic. The movement concludes either with a cadence in the tonic paralleling the end of the exposition, or with a coda following the recapitulation.
JAMES WEBSTER
Like any form in tonal music, a sonata-form movement creates its designs in time. The form is a synthesis of the tonal structure, the sectional and cadential organization and the ordering and development of the musical ideas. In addition, most sonata-form movements depend on ‘sonata style’ (Tovey), i.e. the articulation of events in ‘dramatic’ or ‘psychological’ fashion. Sonata form is not a mould into which the composer has poured the contents, but at most an ‘ideal type’ (Dahlhaus); each movement grows bar by bar and phrase by phrase, with the meaning of each event depending both on its function in the structure and its dramatic context; its true form becomes clear only on close analysis in terms of its effect in performance.
The old dispute, whether sonata form is binary or ternary, is idle and superficial; the form is a synthesis of binary and ternary principles. It is bipartite, in that the exposition has the same tonal structure as a half-cadence or the first half of a binary form: it is open, poised on the dominant, tonally incomplete (ex.1). Hence, notwithstanding the effect of closure at the end, the exposition requires resolution by a balancing second part that closes in the tonic. Most sonata-form movements articulate the tonal polarity of the exposition by contrasting material, or contrasting treatment of the material, in the second group. The modulation out of the tonic usually occurs in a dramatic fashion; the establishment of the new key is an event of aesthetic as well as tonal significance. The paragraphs of about eight to 32 bars that create these sections vary in phrase rhythm, level of activity, harmonic structure and cadential strength; this sense of varied pace is essential to the style. (If a first movement begins with a beautiful, self-sufficient melody in square rhythm which closes in the tonic, for example Haydn’s String Quartet op.76 no.5, it may not be in sonata form at all; a slow movement or finale can afford a more relaxed beginning.)
The second part of a sonata-form movement is longer than the first; it comprises two sections, the development and the recapitulation. The central structural event, distinguishing sonata form from all others that begin with an exposition, is the simultaneous return of the main theme and the tonic key in the middle of the second part. Neither a simple restatement of the main theme alone, nor a simple return to the tonic alone, has the intense impact of this simultaneous return. It creates a parallelism between the beginning of the movement and the beginning of the recapitulation; there is no such relationship in binary form.
In order to give the simultaneous return its maximum effect, the development delays and prepares it. Structurally, the development is a (gigantic) transition from the end of the exposition to the beginning of the recapitulation, analogous to the first section of the second half of ‘rounded binary’ form (see §1(ii)). But the development is also a middle section in its own right, with its own aesthetic: it modulates widely, develops the material and increases the complexity of texture. Hence, when the return finally arrives, it functions as a relaxation of tension or as a triumph over difficulties. The development and reprise are thus dialectically related: without the reprise, the development has little point; the larger and more complicated the development, the more satisfying is the reprise.
The second group in the exposition presents important material and closes with a sense of finality, but it is not in the tonic. This dichotomy creates a ‘large-scale dissonance’ (Rosen) that must be resolved. The ‘sonata principle’ (Cone; the term is misleading, insofar as this is only one of several relevant principles) requires that the most important ideas and the strongest cadential passages from the second group reappear in the recapitulation, transposed to the tonic. The subtle tension of stating important material in another key is thus grounded, and the movement can end. But the recapitulation has now also become a complete section, whose material parallels that of the exposition. The sections thus have the pattern ABA. (In a true ternary form, by contrast, the first A is complete in itself, closing in the tonic, B is merely a contrasting section and the last A merely a restatement.) The power and sophistication of sonata form lie in this synthesis of a three-part design and a two-part tonal structure.
To illustrate these points, exx.2 and 3 show an analysis of the first movement of Mozart’s Eine kleine Nachtmusik K525. The form as a whole is summarized in ex.2; ex.3 shows the material (occasionally simplified), the harmonic content, and the phrase structure. Each new staff corresponds to a new sentence in the music.
Sonata form belongs to the larger class of ‘binary’ (Tovey) or ‘key-area’ forms (Ratner), i.e. forms in two structural parts whose first part ends out of the tonic (see ex.4, where many of these formal types are vertically aligned with respect to their structurally equivalent elements). Within this class, sonata form alone exhibits all three of the following features: a distinct development section, including a retransition; the simultaneous return of the initial theme and the tonic; and a more or less full recapitulation of the second group.
In binary form proper the second part is, at most, modestly longer than the first, adds no new material and exhibits neither a tonic reprise nor a distinct development section. The ‘quatrain’ and A|B-A forms are especially characteristic of minuets and of main themes in variation movements and rondos; their second parts are more highly organized than in binary form. The important but little-studied ‘quatrain’ (Bartha) approximates (on a small scale) Ratner’s ‘key-area’ form. Each of the two main parts is symmetrically subdivided, producing four phrases or sections, a b|x {ab}. The third phrase combines development and retransitional elements, while the last, although beginning as a reprise, often synthesizes both phrases of the first part (Haydn, String Quartet op.33 no.3, trio: bars 47 to 49 (first beat) correspond with bars 35 to 37 (first beat), but the cadential bars, 49–50, correspond with the cadential 41–2); the dialectical relation between the last two phrases is a source of real power. On the other hand it includes neither a true development nor a full recapitulation, and the second part remains roughly the same length as the first. A|B-A form (here with a close on the dominant in the first part) exhibits the same run-on, dialectical relation between B and A. Although often called ‘(two-part) song form’, it has no particular connection with vocal music. Even when its initial A closes on the tonic (a non-binary form not otherwise relevant in this context), it must be distinguished from the three-part A|B|A or ‘ternary’ form, in which B is independent and there is no dialectical B-A relation. When executed on a large scale and with dense musical argumentation, A|B-A movements are indistinguishable from small-scale sonata forms (Mozart, String Quartet k387, minuet, trio) or indeed fully fledged ones (Beethoven, Symphony no.9, scherzo).
Among
the larger forms, in ‘expanded binary’ form the middle section or development
leads, not to a retransition and reprise of the opening, but directly to the
second group in the tonic (often prepared by the original transition, also
transposed down a 5th); there is no ‘simultaneous return’ and hence no full
recapitulation. This form is especially characteristic of Domenico Scarlatti.
In ‘rounded binary’ form there is a complete middle section, usually including
discussion of important ideas from the exposition in transposition, and
sometimes genuine development as well; this section ends with full closure in a
related key (most often the relative minor). The final section either is a full
recapitulation beginning in the tonic (J.S. Bach, Two-part Invention in E;
Haydn, Sonata in B
hXVI:18) or, as in expanded binary, begins in some other key
(most often the subdominant) and moves to the tonic during its course (this is
especially characteristic of J.S. Bach, but also occurs later, e.g. Haydn,
Sonata in G hXVI:6). Such movements must be distinguished
from those with an off-tonic reprise following a true development and
retransition (see §3(iii)). In the ‘sonata without development’, finally, there
is no middle section (see §7(i)).
These distinctions, admittedly, are often difficult to sustain in analytical practice and are far from universally acknowledged (although many apparent disagreements are largely about terminology). Many authorities emphasize the common reliance of all these forms on an Ursatz structure (Schenker) or key-area form (Ratner). Nevertheless, sonata form proper was central compositionally throughout the period under discussion and remains so in our reception of music. In this article it is assumed that only movements exhibiting all three criteria noted above (development and retransition, simultaneous return, full recapitulation) exhibit this form.
The rise of sonata form must be understood in the context of the broad stylistic changes during the 18th century. A movement from before 1750 is usually based on a single main idea, and governed by a single Affekt (psychological state, rhythmic profile, texture etc.). It develops this idea motivically, in uniform texture with a linear bass line (basso continuo) and in metrically orientated rhythms which change only at the cadences. Contrast appears only between opposing planes (solo and tutti, loud and soft etc.). A movement from the late 18th century, on the other hand, usually exhibits several contrasting ideas (often including contrast within a single idea), which develop dramatically in passages of tension and resolution. Its rhythm is that of the bar, the phrase and the antecedent–consequent period. It is essentially melodic in conception (when not downright popular or sentimental), clear in structure, with relatively thin textures, subordinate accompaniments and harmonic basses.
Numerous Baroque stylistic features affected sonata form. Formally, its roots lay in binary form, which had arisen in various French dance movements. This form has already been described (see ex.4); it was important also for its rhythmic organization in phrases and periods based on two-bar units. By the time of Bach, most movements of this type were in expanded or rounded binary form. The second half of these forms resembles sonata form, in that it is longer than the first, and divided into two sub-parts; indeed this expansion was one origin of the later development section.
The origins of the ‘simultaneous return’ are more complex. A return of the opening music in the tonic was common following a ‘trio’, and in the da capo aria, the Italian opera overture, the concerto and the simple aria (final ritornello in the tonic). But none of these is equivalent to a true simultaneous return following a development; rounded binary form lacks both this feature and a full recapitulation in the tonic. Sonata form transformed the division within the second part of rounded binary form into a return to the original theme in the tonic. Once this integration of tonality and material had been achieved, the other novel elements of sonata form – the focus on the simultaneous return as an event, the role of the development as preparation for that return, and the repetition of the entire essential contents of the exposition in the recapitulation – inevitably followed.
The simple, phrase-orientated melody of the pre-Classical period developed in song and opera buffa. The principle of contrast developed in the concerto, the French overture, the da capo aria and pairs of dance movements. Contrapuntal elaboration, which during the pre-Classical period survived in fugues and French overtures, eventually revitalized the melodic style: it returned both in entire fugues (Haydn, String Quartets op.20 nos.2, 5, 6, finales) and, more characteristically, in the synthesis of contrapuntal texture and sonata style (Mozart, String Quartet k387, ‘Jupiter’ Symphony,k551, finales). It also led to ‘thematische Arbeit’ or, in Beethoven’s preferable expression, ‘obbligato accompaniment’. This essential new technique gave the inner parts and the bass rhythmic and motivic independence, while maintaining the aesthetic subordination of the melody. It thus allowed a synthesis of the melodic style and phase-orientated rhythm with sophisticated part-writing and complex textures. It was first fully achieved in Haydn’s quartets op.33 of 1781 (ex.5: op.33 no.1, opening).
Sonata form developed in instrumental music. Indeed its rise was part of the unprecedented triumph of instrumental music – especially the new genres of the keyboard sonata, string quartet and symphony – as the leading type of Western music in the later 18th and the 19th centuries. For sonata form, the most important of these genres seems to have been the symphony. Even the earliest Italian symphonists (before 1740) exploited the driving quaver rhythms, the homophonic texture and the phrase-orientated rhythms of the early Classical style. They often wrote a clear second group in the dominant including a contrasting lyrical theme and closing group, and a clear recapitulation. Where contrast in the exposition was lacking, however, the middle section was often little more than a ‘trio’ in reduced scoring or in the minor (Sammartini, Symphony in A, j–c16; here and in all succeeding citations, reference is to the first movement of the cited work unless otherwise indicated). The more elegant symphonies of J.C. Bach, whose cantabile second themes influenced Mozart, are a later refinement of this style.
The Mannheim symphonists, famous for their orchestral technique, developed a more dramatic style with higher pretensions. But they often omitted the simultaneous return, preferring that variant of rounded binary called ‘mirror’ form (occasionally described as sonata form with ‘reversed recapitulation’), in which the main theme returns at the end of the movement: A (tonic),B (dominant), development, B (tonic), A (tonic). Mozart also occasionally used this form in the 1770s (Violin Sonata k306/300l. The Mannheimers’ brilliant orchestration often masked a certain poverty of invention and incoherence of structure. In Vienna, Monn and Wagenseil often included all the features of sonata form. They transferred the piano episode in the minor from the development to the second group, freeing the section following the double bar for true development. However, this mid-century repertory was short-winded and unpretentious.
In the solo sonata, which 18th-century theorists often described as more expressive or ‘rhetorical’ than the brilliant, festive symphony, large and complex rounded binary forms remained common well into the second half of the 18th century. Most of Domenico Scarlatti’s sonatas are in this form. C.P.E. Bach’s multifarious sonatas often adopt a quasi-improvisatory style with abrupt contrasts. His development section often does not differ markedly from his exposition; a full recapitulation appears more often than not. The string quartet and allied genres of chamber music developed later than the symphony and sonata; their formal designs adopted characteristic features of both those other genres, along with a tendency towards contrapuntal elaboration.
In sum, although all the elements of sonata form, and occasional movements in complete sonata form, can be traced from the 1730s on, neither the form itself nor its combination of dramatic style, structural rigour and thematic logic appeared consistently before Haydn (who mastered it from his earliest works). Its unchallenged reign in instrumental art music on a large scale did not begin until the last quarter of the 18th century.
Sonata form, §3: The Classical period
In
18th-century works, the second group of a movement in a major key almost
invariably stands in the dominant. If the movement is in a minor key, the
second group usually stands in the relative major, less often in the dominant
(minor). Beethoven occasionally used ‘third-relationships’ for the second
group: including closely related keys like the submediant or relative minor
(String Quintet in C op.29), the mediant minor (Sonata in G op.31 no.1,
coloured by the mediant major) and the submediant (String Quartet in F minor
op.95); and remote keys including the flat submediant (String Quartet in B op.130), the mediant major
(Waldstein Sonata in C op.53) and the submediant major (Archduke Trio in B op.97, Hammerklavier
Sonata in B op.106).
The second group usually establishes one single key. The apparent exceptions
usually reveal themselves as expanded transitions (Beethoven, Symphony no.8,
bars 34–45; finale, bars 48–59). Two real exceptions, both in the minor, are
Haydn’s ‘Farewell’ Symphony, no.45, and Beethoven’s Coriolan overture
op.62. Much more common are temporary contrasting modulations within the second
group, often to a remote key such as the flat submediant (Mozart, String
Quartet k499 in D, bars 57, 65).
Depending on the scale and the style, the first and second group may each have one idea or many, organized in a single paragraph or in several. The first group in Haydn’s String Quartet op.1 no.1 consists of a single eight-bar antecedent–consequent period; the second group comprises two eight-bar periods. In Mozart’s Nachtmusik (ex.3) the first group and transition, and the second group, each comprise a single large paragraph with several different themes; but the first group is tightly bound by elisions and question–answer relationships, while the second group is more relaxed and periodic. In the ‘Eroica’ Symphony, the first group comprises three paragraphs, each beginning with the main theme (bars 3, 15, 37); the last of these moves to the dominant of the dominant, where the important theme beginning at bar 45 appears. The second group proper contains seven paragraphs, beginning respectively at bars 57, 65, 83, 99, 109, 132 and 144 (the italicized bar numbers indicate the strongest, section-defining cadences). At the same time the rhythm varies enormously, from the square main theme, through the offbeat accents and syncopations of bars 25, 28, 45, 83, 109, 113 and 119 to the climax in bars 123–31. Most large-scale expositions include a cadential closing group or ‘codetta’ following the largest paragraph of the second group (Mozart, ‘Jupiter’ Symphony k551, bars 101, 111; Beethoven, Sonata op.27 no.2 (‘Moonlight’), finale, bars 43, 57).
Especially before about 1780, the relationship between the first group and the second can be simply that of antecedent and consequent on a large scale, without an independent transition. This boundary is marked by the (poorly named) ‘bifocal close’: the first group ends with a half-cadence on (but not in) the dominant, followed immediately by the second group actually in the dominant (Haydn, String Quartet op.33 no.3, bars 26–7; Mozart, Paris Symphony k297/300a, bars 51–2; Beethoven, String Quartet op.18 no.5, bars 24–5). More often, especially after 1780, a clear transition appears. The transition often develops out of a restatement of the main theme (Mozart, Symphony in G minor k550, bar 22; Beethoven, Waldstein Sonata, bar 14). In any case it modulates beyond the dominant to the dominant of the dominant (ex.3) in order to establish the dominant itself more strongly and to make the home tonic ‘sink below the horizon’ (Tovey). In these cases a caesura on the dominant of the dominant may still separate the exposition into two parts (Mozart, ‘Jupiter’ Symphony, bar 55). Not all expositions are divided into two parts. An important exposition form in Haydn comprises three parts: first group in the tonic; transition elided to an active second group avoiding firm cadences; contrasting closing group (Symphony no.99, bars 19–34, 34–70, 71–89).
The
second group often begins with or includes a contrasting lyrical theme (Haydn,
String Quartet op.33 no.5, bar 49; Mozart, Symphony in G minor k550,
bar 44; Beethoven, Waldstein Sonata, bar 35). Haydn often began the second
group with an adaptation of the main theme, usually varied in harmonization,
texture, contrapuntal accompaniment, scoring or phrase rhythm (Sonata no.49 in
E, bar 25). Such
movements are often called ‘monothematic’; but since they almost always bring
new material later in the second group (ibid., bars 28, 42, 53, 60), that term
is better restricted to those very rare movements that are based entirely on
only one theme. A new theme may appear towards the end of the first group, still
in the tonic, dominating not only the ensuing transition but other sections as
well (Mozart, String Quintet k516, bar 30). The closing
group often restates the main theme in varied form (Mozart, String Quartet k465, bar 91; Beethoven, Symphony no.2, bar 112). Many
expositions have little or no thematic contrast (Haydn, ‘Farewell’ Symphony;
Beethoven, Sonata op.101).
In 18th-century music the exposition is almost always directed to be repeated, with or without a transition back to the opening. This repetition lends the material greater solidity and familiarity; and it allows the exposition, whose tonal structure motivates the entire form, a chance to make its full effect.
Sonata form, §3: The Classical period
It is important to distinguish between the process of development (Ger. Entwicklung) and the part of a sonata-form movement called the development section (Durchführung). This section is by no means devoted exclusively to the development of the material; conversely, this process is not restricted to development sections. Still, thematic development characteristically reaches its culmination here. Typical techniques include fragmentation of a theme into shorter motifs, often combined with rapid modulations based on sequences (Mozart, ‘Jupiter’ Symphony, bars 133, 171), combination of a theme with a counterpoint (Mozart, Symphony in G minor k550, bar 115), contrapuntal combination of originally separate themes (Beethoven, ‘Eroica’ Symphony, bar 186), juxtaposition of contrasting themes originally stated separately (Haydn, Symphony no.102, bars 116–17), increased complexity of texture (‘Eroica’, bar 220; cf 45), extension by sequence (ibid., bar 178), alteration of the rhythmic structure (Haydn, ‘Surprise’ Symphony, no.94, bar 107), and so forth indefinitely. When combined with excursions to remote tonal areas or passages of tonal instability, these techniques can make the development section a passage of great tension. In a psychological sense, this tension is the climax of the movement. At the same time, it prepares the structural climax, the simultaneous return which begins the recapitulation.
Hardly
any rules can be laid down regarding the choice or treatment of material. The
development may use only one theme from the exposition (Mozart, Symphony in G
minor k550) or several (Haydn, Symphony no.102). In many
works before 1780, the development still begins with a statement of the main
theme in the dominant (Haydn, Symphonies nos.6 and 8), sometimes proceeding to
the main theme in the tonic before the development proper (Haydn, Symphony
no.36). Beethoven even occasionally began with the main theme in the tonic,
soon breaking off in new directions (Sonata op.31 no.1; String Quartet op.59 no.1).
After 1780, if the main theme opens the development, it is usually transformed
in key, motivic content, phrase structure etc. (Haydn, Symphony no.94; Mozart,
Symphony in G minor k550; Beethoven, Symphony no.9). The
development may also begin with a theme from the second group (Mozart, Symphony
in Ek543,
bar 145; cf 110) or the closing group (Mozart, ‘Jupiter’ Symphony; Beethoven,
Sonata op.10 no.2).
Especially before about 1780, restatements of material without substantial change save for the key are common (Haydn, Symphony no.47, bars 90–101; cf 36–47). Fairly often, especially in Mozart’s works from the late 1770s and early 1780s, the development begins with a new theme; this theme usually provides a point of repose following an exposition of unceasing activity (String Quartet k458). A new theme elsewhere is always a special effect (Haydn, ‘Farewell’ Symphony, bar 108; Beethoven, the famous E minor theme in the ‘Eroica’, bar 283). In a finale, however, a new theme often functions analogously to a rondo episode (Mozart, String Quartet in A k464, finale, bar 114).
After the opening sentence, the development usually avoids repetition of material in the same key in which it originally appeared. An exception is Haydn’s ‘false recapitulation’, i.e. a seemingly misleading statement of the main theme in the tonic as if the return were at hand, followed by further development and, eventually, the true return (Haydn, String Quartet op.17 no.1, bars 62, 76); in later years, the false recapitulation may appear in a foreign key (Haydn, Symphony no.102, bar 185).
Before 1780, many developments reflect their origins in rounded binary form by centring on one closely related key, most often the relative minor (Haydn, String Quartet op.17 no.5), less often the supertonic or mediant. Towards 1800, it became increasingly common to include more keys, and more remote ones (Mozart, Symphony in G minor k550, both outer movements), or to create a sense of instability by modulating rapidly or by virtually being in no key at all (Beethoven, Symphony no.4, bars 257ff). The tonal plan of a development is often a bridge prolonging the dominant by a neighbouring key such as the relative minor or subdominant (ex.2), or by a transformation of the dominant triad into the dominant 7th, impinging on the home tonic. If the home tonic is minor and the second group is in the relative major, the development will complete the large-scale arpeggiation of the tonic triad by proceeding further from the mediant to the dominant.
Mozart’s developments are fairly short, perhaps 50–60% of the length of the exposition. Most centre on a single process in the middle, introduced by a contrasting passage and followed by the retransition (Symphony in G minor k550, bars 101–14, 115–38, 139–65; Nachtmusik,ex.3). Haydn’s typical developments are perhaps 75% of the length of the exposition, Beethoven’s perhaps 90%; occasionally, they wrote developments longer than the exposition (Haydn, Symphony no.102; Beethoven, ‘Eroica’ Symphony). Both Haydn and Beethoven divided the development into a number of distinct sections; Beethoven further integrated these into a single psychological progression which, belying the outward diversity and adventure, prepares the recapitulation with unparalleled power and excitement.
The
development almost always arouses expectations of the simultaneous return. Most
characteristic is a passage of dominant preparation, with or without references
to the main theme (Haydn, Symphony no.102, bars 217–26; Mozart, Symphony in G
minor k550, bars 153–65; Beethoven, Waldstein Sonata,
bars 136–55). Often this dominant is coloured by the tonic minor (Haydn and
Beethoven, ibid.; Mozart, Nachtmusik). The tonic may be approached
indirectly by a sequence (Mozart, Symphony in E k543, finale, bars 181–4) or
following a half-cadence in a related key, usually the dominant of the relative
minor (Haydn, String Quartet op.64 no.6, bars 97–8). When Haydn and Beethoven
opened with a theme lying off the tonic (Mozart avoided this), they usually
aimed the return at the other sonority (Haydn, Symphony no.94; Beethoven,
Sonata op.81a).
Sonata form, §3: The Classical period
It is useful to distinguish between ‘recapitulation’, in the sense of the entire third section of a sonata-form movement or any large part thereof (e.g. the second group), and the return of a given idea or passage, for which ‘reprise’ can be employed.
The recapitulation almost always enters unambiguously with the ‘simultaneous return’ of the opening theme in the tonic. When two or more themes occur in the first group (not the transition), the return to the tonic may coincide with the second of these; the opening theme then appears earlier in a foreign key (Beethoven, Sonata op.10 no.2), immediately afterwards in the tonic (Beethoven, String Quartet op.59 no.1), or as a coda following the second group (Haydn, String Quartet op.50 no.3). The main theme may return transposed in mode (Haydn, Symphony no.47). But if the main theme never returns, or if the return to the tonic is delayed until the second group, the movement is in one or another version of rounded binary form. In the pure type, the first group never returns (Haydn, Sonata in G hXVI:6); or it may follow the second group, producing ‘mirror’ form (Mozart, Sonata k311/284c). Still closer to sonata form is a full recapitulation in which the main theme returns in the subdominant (Mozart, Sonata k545; Beethoven, Coriolan overture; Schubert, Symphony no.5). In masterworks, the recapitulation does not then simply repeat the exposition mechanically, transposed down a 5th; thus ink545 the transition is expanded and leads, as did the exposition, to a half-cadence on the dominant.
In
recapitulating the remaining material of the exposition, the sonata principle
always applies: material first presented outside the tonic is repeated in the
tonic. But in Haydn, Mozart and Beethoven these repetitions are never
mechanical. Their regular procedure was to repeat the main theme, to refer to
the other first-group themes and the transition with an appropriate harmonic
orientation, and to repeat the chief ideas from the second group in the same
order (Mozart, Symphony in E k543; Beethoven, Symphony
no.5).
Within this framework, substantial alterations in the latter part of the first group and the transition are common. Material may be omitted (Mozart, Symphony in D k297/300a, bars 40–47), expanded (Mozart, Symphony in G minor k550, bars 191–216, cf 28–33; Beethoven, Waldstein Sonata, bars 167–74, cf 12–14), or wholly recomposed (Beethoven, ‘Eroica’ Symphony, bars 402–39; cf 7–36). These expansions often take on a similar character to the development (Rosen: ‘secondary development’), as if the issues raised there also require recapitulation (Mozart, ibid.). If there is a transition, it must be altered so as to prepare the tonic, not the dominant (Mozart, ‘Jupiter’ Symphony, bars 225–43; cf 36–55; see also ex.2). More remote keys, usually on the flat side (Beethoven, String Quartet op.59 no.1, bar 279; cf 30), and changes of mode (Mozart, ‘Jupiter’ Symphony, bar 212, cf 24; Beethoven, ‘Appassionata’ Sonata op.57, bar 152, cf 17) may also occur in these passages; such flat-side or minor-mode emphasis often creates a quasi-subdominant feeling.
In
the second group proper, Mozart almost always introduced subtle alterations,
such as expansion of a sensitive harmonic area (Symphony in G minork550, bars 245–53; cf 62–5), additional repetitions (Symphony
in Dk504, bars 259–64; cf 112–15), or enrichment through
contrapuntal elaboration, fuller scoring, obbligato inner parts etc. (Symphony
in E k543,
bars 254–62, horns and second violins; cf 97–105). Beethoven is much more
likely to recapitulate the second group literally (even in the gigantic
‘Eroica’); presumably his larger developments and codas require this symmetry.
In later Haydn, the recapitulation does not necessarily follow either the course of the exposition or any other definable pattern. Even where more or less the same events occur in more or less the same order, every sentence may be rewritten (String Quartet op.33 no.3). In other cases paragraphs appear in reverse order (Symphony no.47, bars 127–49; cf 36–46, 13–21), subsidiary material is expanded (Symphony no.99, bars 157–90; cf 71–87 and the development), or the whole simply rewritten (‘London’ Symphony, no.104). But the most important material from the second group is still recapitulated in the tonic (ibid., bars 247 and 267; cf 65 and 100). And the material following the simultaneous return usually approximates in length to the second group, thus preserving the proportions and the tonal balance of the whole.
Important
new tonal relations within the second group are usually prohibited by the
necessity of grounding the material in the tonic. Temporary excursions to other
keys usually recur in appropriate transposition (Mozart, String Quartetk499), often expanded (Mozart, String Quartet k465,
finale). An apparently arbitrary modulation may refer to the development, as in
Haydn, Symphony no.100 in G, where in bar 239 E resolves the juxtaposition of the dominant and B at the beginning of the
development. A special situation arises when a movement in the minor has
originally placed the second group in the relative major. Mozart, and Haydn in
his early and middle-period music, normally recapitulated these second groups
in the tonic minor, i.e. altered in mode. In his late music Haydn transposed
the second group (not the first group) to the tonic major, thus ending
cheerfully with a change of mode within the recapitulation. Beethoven combined
both procedures: the second group is transformed (at least in part) to the
tonic major, with an air of release which reveals itself as ‘tragic irony’
(Tovey) when ‘catastrophe’ strikes in the coda (Symphony no.5). In finales and overtures,
the second group may remain in the minor, so that the coda can ‘triumph’ in the
major (String Quartet op.95, finale).
Before 1780, the entire second part (development and recapitulation) was usually directed to be repeated – another indication of the binary structure of the whole. After 1780, this repetition became increasingly rare, even when the exposition was repeated. After the finale of Beethoven’s ‘Appassionata’ Sonata, it had become obsolete.
Sonata form, §3: The Classical period
Many
first movements and an occasional slow movement and finale are preceded by a
slow introduction. Its primary function is to strike a more serious or grander
tone, and to establish a larger scale of motion, than would be possible by the
Allegro alone. Tonally, almost every introduction first establishes the tonic,
and then cadences on the dominant to prepare the Allegro. Occasionally, the
goal is a half-cadence in another key (Haydn, Symphony no.103, ‘Drumroll’) or a
chord other than the dominant (Haydn, Symphony no.92, ‘Oxford’; Beethoven,
Sonata op.81a). The Allegro which follows an introduction can begin with
a squarer melody than is otherwise possible (Mozart, Symphony in Ek543;
Haydn, no.104 in D). Before 1790, there was little thematic or psychological
connection between introductions and allegros. From then on such connections
became increasingly common (Haydn, Symphony no.103; Beethoven, ‘Pathétique’
Sonata op.13). Beethoven eventually integrated introductory material into the
movement so completely that a separate section can no longer be distinguished
(Symphony no.9; String Quartet op.130).
The conclusion of a sonata-form movement follows either of two main principles. The first is to end with the same music as concluded the exposition (Mozart, String Quartet k428/421b; Beethoven, Sonata op.22). Mozart and Haydn often combined this formal symmetry with an expansion late in the second group (Mozart, String Quartet k464, bars 234–62, cf 73–83; String Quintet k515, where bars 320–52 replace the single bar 130). Often this expansion involves a new statement of the main theme, the end of the exposition still returning to round off the whole (Mozart, Symphony in G minor k550, bars 283ff; Haydn, Symphony no.102, bars 282ff).
These expansions give something of the effect of a coda while maintaining the symmetry of binary form. A true coda, by contrast, follows the recapitulation of the closing group (or breaks off from the pattern of the exposition, the omitted material not returning). Save for the functions of expansion (equivalent to the examples just cited) and ‘peroration’ (Tovey), and notwithstanding several recent studies, the structural significance of codas is not well understood. The German expression ‘second development’ is applicable only to certain parts of certain Beethoven codas; and in any case the structural function of the development, to prepare the return, cannot be repeated. In rhythmic terms, the coda has been called a gigantic ‘afterbeat’ to the form as a whole; in Schenker’s theory, it consists of the music following the background descent to the tonic.
Almost every coda restates the main theme; this restatement is often transformed into a climax. In some Beethoven works, every previous statement of the theme will have been incomplete or deprived of strong root-position tonic support (String Quartet op.59 no.1, bars 1, 242, 250–55). In these cases, the coda provides ‘thematic completion’ (Kerman) by presenting the definitive or climactic version (op.59 no.1, bar 348). As befits the end of a large tonal structure, most codas include some emphasis on the subdominant, especially if none has occurred in the recapitulation.
In
his late music Haydn often rewrote the recapitulation so thoroughly that the
concept ‘coda’ makes little sense, unless one accepts Tovey’s interpretation
that these sections synthesize the functions of recapitulation and coda. Mozart
wrote a coda about as often as not; it is usually a single short paragraph
centred on a restatement of the main theme, often in altered or contrapuntally
enriched form (Nachtmusik; String Quintet k516). A
transitional passage often prepares this climax (String Quartet k465;
‘Jupiter’ Symphony, finale); cadential passages, whether on the main theme
itself (Quartet k465) or on less highly charged material
(Quintet k516; ‘Jupiter’ finale), then conclude the
movement. Beethoven almost always wrote a large coda, often as long as the
development (Symphony no.5), occasionally as long as the rest of the movement
(Symphony no.8, finale). A typical Beethoven coda might begin by turning away
to new questionings, often in new keys (‘Eroica’ Symphony, bars 557–80).
Eventually a climax is reached on the main theme in the tonic (Symphony no.5,
bar 478) or alternating tonics and dominants (‘Eroica’, bars 631–73). The final
cadences, perhaps preceded by the lyrical theme (Waldstein Sonata, bar 284),
comprise a separate paragraph following the climax proper. The coda may also
recapitulate events from the development. In the ‘Eroica’ development, the E
minor theme establishes the minor Neapolitan (flattened supertonic), and
appears later in the tonic minor (bars 284, 322). In the coda it returns in the
ordinary supertonic, F minor, and, again, in E minor (bars 581, 589); the ensuing crescendo on V is
related to the retransition (bars 338–61, 603–20). Thus the remote key is ‘grounded’
by its diatonic equivalent, at the same time as the whole section gains an
appropriate subdominant emphasis.
Two broad strains may be identified in 19th-century music: a ‘Romantic’ one, focussing on vocal music, programme music and the characteristic piece for piano; and a ‘classicizing’ one, focussing on the traditional genres of absolute music. Only the latter tradition gave sonata form much prominence. The relatively uncommon sonata-form movements in the former tradition (and some in the latter) often treat the form in an academic manner, as a mould, not a process, or as a ‘vehicle for the sublime’ (Rosen). Many large instrumental works in this repertory, while referring to sonata form, seem also to be searching for different forms altogether (Berlioz, Symphonie fantastique; Liszt, Sonata in B minor).
Changes of style in the 19th century lent new meaning to many aspects of sonata form. The most important of these was the Romantics’ attitude towards musical material. Their programmatic and self-expressive tendencies focussed on the explicit content of music in unprecedented fashion. The primary focus of 19th-century compositional ‘inspiration’ comprised striking and original themes, often harmonized with chromatic or apparently free harmonic progressions. This concentration on themes for their own sake was related to the rise of the lied and the characteristic piano piece, where the quality of the theme was the chief raison d’être of the composition. A related phenomenon was the tendency to favour square phrasing in four-bar phrases – perhaps a side-effect of the concentration on themes and harmonic progressions – in place of the supple Classical phrase rhythms.
All
these features led to the central importance of the second theme in Romantic
sonata form. Following a noble, stormy or in some way difficult first
paragraph, and an agitated transition, composer and listener alike welcomed the
chance to indulge in a beautiful melody in the new key. This became legitimized
in the 19th-century doctrine that sonata form was based on the duality of two
contrasting themes (often characterized as ‘masculine’ and ‘feminine’) rather
than on the tonal duality of the exposition. Even the classicizing tradition
almost always included a contrasting second theme (Schubert, String Quintet in
C d956; Mendelssohn, Overture, Fingal’s Cave;
Brahms, Symphony no.3). Indeed, many opening themes in this style are complete
paragraphs in themselves, preceding the ‘drama’ (Schubert, Sonata in B d960;
Brahms, Sextet in B
op.18). These self-sufficient themes alternate with impassioned climactic
passages; despite the presence of many features drawn from the sonata-form
tradition, many such movements do not exhibit sonata form as a whole (Chopin,
Ballade in G minor).
Related to these tendencies was the Romantic bias against literal repetition. The second half of a sonata-form movement (development and recapitulation) is never repeated, the first half but rarely. The main theme is often varied on each return; new themes are derived by ‘thematic transformation’ (Berlioz, Liszt as above; Schumann, Symphony no.4). The continual thematic development, combined with the bias against repetition, diminished the importance of the recapitulation. The main theme may not be recapitulated (Weber, Sonata no.1 in C; Schumann, Symphony no.4, finale; Chopin, Sonata in B minor); the second group may be omitted, the coda following directly on the return of the main theme (Schumann, Symphony no.4, first movement); or the main theme may crown the work in an ‘apotheosis’ (Cone) or ‘transfigured’ recapitulation (Chopin, Polonaise-Fantasy). Conversely, the coda increased in importance; indeed the climax often comes not at the simultaneous return but in the coda (Schumann, Symphony no.2, finale). This is one aspect of the 19th-century tendency to displace towards the end the weight of every form, single movements and whole cycles alike; but few 19th-century composers other than Beethoven achieved such climaxes within an overall sonata-form aesthetic.
The
other principal difference in 19th-century sonata form is the greatly expanded
system of tonal relations. The basis of this expansion is the acceptance of
major and minor as equally valid representations of the tonic (Schubert,
Quartet in G d887, opening, beginning of the
recapitulation, final bars; also the finale). Schubert commonly used remote
keys in the second group: flat submediant (Grand Duo in C for piano, four
hands); flat submediant minor (Sonata in B d960); flat mediant (String
Quintet in C); and even the leading-note minor (‘Reliquie’ Sonata in C). As in
these cases, such second groups often fall into two distinct sections: the
first, devoted to the obligatory lyrical theme, stands in the remote key; the
second appears in the dominant, often following a developmental transition, and
contains business-like and cadential paragraphs. In Brahms – the only
19th-century composer whose mastery of form was comparable to Haydn’s, Mozart’s
and Beethoven’s – the special key in the second group may lead to the dominant
(Symphony no.2, D–f
–A)
or stay in its own orbit (Piano Quintet, f–c–D; Symphony no.3, F–A–a), but in either case these
contrasts are always integrated into the whole. Schubert’s developments often
consist of little more than a gigantic sequence (String Quintet); his
preparations for the return are always masterly (Sonata in B d960).
Mendelssohn’s and Brahms’s developments, though lacking Beethoven’s illusion of
teleological ‘necessity’, always seem logically related to what has gone
before. All three composers give full recapitulations which tonally resolve the
second group, generally in the tonic or in the keys a 5th below those in the
exposition (Schubert, Symphony no.9, C–e–a–G becomes C–c–c–C; Brahms, Piano Quintet, f–c–D becomes f-f-F). They also resemble Classical sonata form in
usually requiring only a final paragraph or two as coda, keeping the weight of
the form centred on the beginning of the recapitulation. This divergence in
treatment of recapitulation and coda is the chief distinction between
classicizing and Romantic sonata forms.
Sonata form still appeared in tonal music on conventional models by composers of all nationalities, persuasions and styles: Richard Strauss and Hindemith, Elgar and Britten, Copland and Piston, Roussel and Milhaud, Prokofiev and Shostakovich. Except for the increasingly dissonant harmonic style and widely ranging tonality, the outward features of form in these repertories do not differ significantly from 19th-century ones.
More
interesting are the reinterpretations of sonata form by the great 20th-century
innovators who, after World War I, re-established contact with traditional
styles: Stravinsky and Bartók. In his ‘neo-classical’ period (c1920–50)
Stravinsky often adhered to the outward conventions of traditional forms, but –
as in all his music – re-created them anew. Thus the Symphony in C articulates
the leading-note B as a stronger ‘dominant’ than the orthodox G; the resulting
implications of E minor create Stravinsky’s characteristic multiple tonality in
interlocking planes (cf the Symphony of Psalms and the Symphonies of
Wind Instruments). The internal relations among the parts of the form are
not dynamic, but more nearly circular, symmetrical and static – in keeping with
other aspects of Stravinsky’s style. Bartók’s wild, dissonant style based on
unusual diatonic scales renounces the perfect triad as the basic sonority in
favour of (often dissonant) primary intervals (for example C–E and C–F in the String Quartet
no.4). But since these intervals can still imply directional motion and
articulate tonal areas, they can create a sense of tonal potential
(exposition), conflict (development) and resolution (recapitulation).
More surprising, and more problematic, is the use of sonata form in atonal repertories. With the development of the 12-note method in the 1920s, Schoenberg, Berg and Webern returned to large instrumental movements based on traditional formal plans. Sonata and related forms appear, for example, in Schoenberg’s Piano Piece op.33a and String Quartets nos.3 and 4; Berg’s Wozzeck (Act 2 scene i), Chamber Concerto, Lyric Suite for quartet, and Lulu (Act 1 scenes ii and iii); and perhaps in Webern’s Trio op.20, Symphony op.21, Quartets opp.22 and 28 and Concerto op.24.
In these cases the form can be articulated only by the sectional structure and the development of the musical ideas. Such techniques as inversion or complementation of the set (Schoenberg’s Quartet no.4, slow movement), emphasis on the perfect 5th and varying segmentations of the set (his op.33a), or the repetition of characteristic intervals or specific pitches in different transpositions of the set (Webern’s Symphony op.21) can clearly articulate pitch groupings and intervallic complexes. But these distinctions hardly function analogously to that between tonic and dominant in Mozart or Brahms; insofar as tonality is the essential force governing sonata form, then 12-note ‘sonata form’ is necessarily different in practice. Even if the material and its development articulate sections which mimic exposition, development etc., the unity of (developmental) process and (tonal) structure that had characterized tonal sonata form is exploded; the form is an abstract norm of coherence, independent of the 12-note procedures. But this dissociation is a typical 20th-century solution to an artistic problem, comparable in its own way to Stravinsky’s dissociation of textures and tonalities. Webern’s unique forms based on canon, variation and retrogrades seem more organically related to the 12-note material; here too, however, if sonata form is implied, it is hardly audible in any traditional sense.
With the decline of heroic musical modernism following World War II, and especially in the heterogeneous musical world of the last quarter of the 20th century, sonata form ceased to be a major aspect of the structuring of significant music. It made perhaps its last meaningful appearance in, or rather with respect to, a work of Boulez (the last great modernist figure), in his programmatic comment that the first movement of his second piano sonata was conceived as a project in its ‘destruction’.
No adequate description of sonata form appeared before the 1790s. Most 18th-century speculative and literary writing on music focusses on traditional subjects (opera, continuo, counterpoint) or on aesthetic rather than formal matters. Binary form is described in Mattheson’s Das neu-eröffnete Orchestre (1713) and F.E. Niedt’s Musikalische Handleitung, ii (2/1721). Scheibe’s Der critische Musikus (1745) describes ‘rounded binary’ form with modulations within the (longer) second part. Quantz’s Versuch on the flute (1752) clearly describes the second group in the dominant. The best early account of musical form appears in Joseph Riepel’s Anfangsgründe zur musicalischen Setzkunst, ii: Grundregeln zur Tonordnung insgemein (1755). Riepel’s essay gives a detailed account of phrases and cadences; it describes the development section, complete with modulatory plans; and it analyses a complete movement in terms of his new criteria. On the other hand, Riepel did not describe the simultaneous return to the main theme and the tonic as a constituent of the form.
The fullest 18th-century description of sonata form occurs in vol.iii (1793) of Heinrich Christoph Koch’s Versuch einer Anleitung zur Composition. Koch expanded and refined Riepel’s phrase-rhythmic analysis; he applied these principles to ever larger segments of music, culminating in an analysis of an entire sonata-form movement as an expansion of the form of a minuet. Koch also gave the first adequate description of the twofold division of the second half into development and recapitulation and also described various types of development section. Comparable but less detailed accounts appear in other works from this decade: Francesco Galeazzi’s Elementi teorico-pratici di musica, ii (1796), A.F.C. Kollmann’s An Essay on Practical Musical Composition (1799) and Carlo Gervasoni’s La scuola della musica (1800).
With only scattered exceptions, these 18th-century writers described sonata form as binary, not ternary (even when the recapitulation was clearly distinguished from the development). They understood its organization primarily in terms of the tonal structure of the exposition, as well as the phrase rhythm and cadence plan. Similarly, they described the form in terms of a single main theme (which may, to be sure, lead to derived subsidiary themes). Only Abbé Vogler, in Betrachtungen der Mannheimer Tonschule, ii (1779), and Burney in his description of J.C. Bach in the General History of Music, iv (1789), referred to a cantabile, contrasting second theme in a sonata-form exposition. Neither is an important theorist; neither goes beyond simple description of a particular case. The notion of a second theme as a vital constituent of the form did not arise until the 19th century.
The prescriptive or textbook doctrine of sonata form arose simultaneously with many composers’ acceptance of an academic or abstract version of the form. This doctrine described the form primarily in terms of the material and its development, that is as an ABA: the exposition, consisting of a (‘masculine’) main theme in the tonic, a transition to the new key, a contrasting lyrical (‘feminine’) second theme, and perhaps a closing group; the development, whose function was to attain a climax by developing the material in remote keys; a full recapitulation; and a coda.
This model arose in part as an attempt to explain the difficult works of Beethoven, in part as a recipe for use in teaching composition and in popular analysis. It was influenced by analyses in the Leipzig Allgemeine musikalische Zeitung, by the writings of E.T.A. Hoffmann, and by Heinrich Birnbach’s article ‘Über die verschiedenen Formen grösserer Instrumentalstücke’ (Berliner Allgemeine musikalische Zeitung, 1827–8). An early influence in France was J.-J. de Momigny’s treatise Cours complet d’harmonie et de composition (1806). The theory appeared fully developed in Antoine Reicha’s Traité de haute composition musicale, ii (1826), in A.B. Marx’s Die Lehre von der musikalischen Komposition (1837–47) and in Carl Czerny’s School of Practical Composition (1848–9).
In the 20th century, new methods of analysis and fresh historical investigation have brought about numerous reinterpretations of sonata form. There have been six main currents. (1) Tovey persuasively ridiculed the textbook model, meanwhile producing an immense body of penetrating individual analyses. He also reinterpreted the whole theory in terms of ‘sonata style’, i.e. the dramatic and tonal effects obtainable in the form. (2) Schenker’s structural theory of tonal music, while not overtly concerned with form as such, demonstrated in a long series of profound analyses that every exposition is a single ‘half-cadence’, tonic-dominant, every development a prolongation of the dominant (or the equivalent). Thus the binary structure and the primary role of tonal forces in sonata form were laid bare. (3) A renewed interest in rhythmic analysis (Ratz, Georgiades, Cooper and Meyer, Cone, Morgan, Schachter, Rothstein and others) has fostered appreciation of the central role of phrase and period in sonata style. (4) Revived appreciation of the rhetorical and referential dimensions of 18th-century music, in both theory and practice, has led to rhetorical and semiotic analyses of large instrumental forms by Ratner, Agawu, Bonds and others; interest in literary theory has fostered narratological and Bakhtinian analyses, primarily of 19th-century works, by Newcomb, McCreless, Maus, Edwards and others. An analogous trend is the recognition of the lack of closure in certain sonata-form movements and, consequently, their larger-scale function within entire works (Webster, Kinderman, Haimo and others). (5) Recent interest in music ‘as cultural practice’ (L. Kramer: Music as Cultural Practice, 1800–1900, Berkeley, 1990, for example) has led to interpretations of large-scale instrumental movements in terms of their composers’ real or supposed social, cultural and sexual orientations; the relation of such interpretations to more traditional analytical results is not yet clear. (6) Finally, with the postmodernist suspicion of any kind of formalism (perhaps also related to the decline of sonata form in compositional practice), some theorists and analysts have begun to contest the importance of sonata form in genres (notably vocal music) to which it had been uncritically applied.
Since World War II, Newman, Ratner, Feil, Ritzel, Churgin and others have made the first systematic studies of 18th-century writings on rhythm and form, as, more recently, have I. Bent and others of 19th-century writings. Meanwhile Fischer, Tobel, Larsen, Newman, Kamien and others have begun to draw careful distinctions among various types of Classical sonata form. These perspectives have, for the first time, laid the foundations for a historically and analytically differentiated history of sonata form – which, admittedly, remains to be written.
Closely
related to sonata form is the common form comprising an exposition and
recapitulation but no development (and usually no repeats). As it often occurs
in the first movements of sonatinas, and in slow movements, it is often called
‘sonatina form’ or ‘slow-movement form’; but it also appears in other contexts,
so the more neutral term ‘sonata without development’ is preferable. (In binary
form, the second part begins in the dominant, not in the tonic.) The form is
common in slow movements (Haydn, String Quartet op.50 no.5; Mozart, Symphony in
E k543;
Beethoven, Sonata op.31 no.2; Brahms, Symphony no.4), overtures (Haydn, Orlando
paladino; Mozart, Le nozze di Figaro; Beethoven, Die Geschöpfe
von Prometheus) and finales (Mozart, String Quartet in E k428/421b;
Brahms, Symphony no.1). The return may enter immediately on the final cadence
of the exposition (Mozart, String Quintet k516, Adagio),
after a brief transition on a dominant pedal (Mozart, String Quartet k465, Andante), or following a brief modulating passage (Haydn,
String Quartet op.33 no.3, Andante, bars 59–64). In the last case the
dividing-line between transition and development is not always clear
(Beethoven, Sonata op.10 no.1, finale, bars 46–57). Often one or more
paragraphs will be considerably expanded in the recapitulation, giving the
satisfaction of a ‘secondary’ development in an appropriate context (Mozart,
String Quartet k465, Andante, bars 57–74, 85–96; Brahms,
Symphony no.1, finale, bars 204–19, 232–300, including material from the
introduction). Some movements of this sort resemble Mozart’s favourite
sonata-rondo form, A–B–A–C/(development) B–A.
Classical rondos often place the first episode in the dominant, like a second group; it then is almost always recapitulated in the tonic as the third episode (the middle episode begins in a different key, usually of the opposite mode to the tonic, and functions like a development): A (tonic) –B (dominant) – A (tonic) – C (various keys) [?= development] – A (tonic) – B (tonic) – A (tonic) – coda (tonic). This is sonata rondo form par excellence; the only essential difference from sonata form is the return to the main theme in the tonic immediately following the second group (or first episode). But most sonata rondos also exhibit the lighter style characteristic of finales, with squarer phrasing and complete rounding off of the main theme in the tonic. (SeeRondo.)
Most 18th-century concerto movements are based on the ‘ritornello principle’; that is, an alternation of tutti (T) sections with solo (S) ones (the latter often modulating), most commonly:
Though often on a larger scale, the initial tutti was structurally equivalent to the opening ritornello of an aria. The ‘little returns’ to the main theme in the subsequent tuttis, and especially the rhyme between the cadence of the initial tutti and the final cadence, define the principle. During the second half of the century, the three solo sections increasingly took on aspects of exposition, development and recapitulation: Mozart synthesized the ritornello principle with sonata form, producing a new form altogether. (It appears only sporadically in other composers.)
In Mozart’s concertos, although the opening orchestral tutti presents much of the material that will later be stated during the solo exposition, it is not a ‘first exposition’ (as it is often called), for it remains in the tonic throughout, and it retains the aesthetic function of preparing the entry of the soloist, who enters with either the main theme (k488), a transition (k467) or a new theme (k466). Soloist and orchestra then execute a complete exposition together, with first group, transition, and second and closing groups in the dominant. The second group often includes one or more new themes (k503, bar 170) and omits one or more themes from the ritornello (ibid., bars 51, 59 omitted). An orchestral passage in the dominant, based on the opening ritornello, concludes the exposition. The development is structurally equivalent to that in sonata form, but it is often more sequential or episodic in style. The recapitulation, all in the tonic, is a synthesis of the exposition and the opening ritornello, combining soloist and orchestra; ritornello themes omitted from the exposition are usually restated here (k503, bar 365). At the close of the recapitulation, an orchestral passage leads to a tonic 6-4 chord over a fermata; the soloist then executes a cadenza, which leads through the dominant to a cadence on the tonic. The orchestra enters on this tonic and closes the movement with material from the ritornello, most often its closing paragraph. The form of the soloist’s portions are thus comparable to sonata form, but the form of the whole is still governed by the ritornello.
Form in vocal music was until recently little studied. In larger and through-composed works the text and considerations of rhetoric often seem to determine the form. Although full sonata form is found (Mozart, Don Giovanni, ‘Ah taci, ingiusto core’; Haydn, The Creation, ‘With verdure clad’; Beethoven, Missa solemnis, Benedictus; Brahms, German Requiem, ‘Ihr habt nun Traurigkeit’), it is neither as common nor as characteristic as has traditionally been supposed. In particular, the majority of vocal movements that begin with a clear exposition closing in the dominant proceed more freely and less ‘formally’ in their later stages, with less emphasis on the return to the tonic, than instrumental movements. Even in the Classical period, simpler and more flexible forms are commoner: binary form (Mozart, Die Zauberflöte, no.10, ‘O Isis und Osiris’), sonata without development (Mozart, Le nozze di Figaro, Act 3, sextet; Haydn, partsong An die Frauen), and a richly varied repertory of ternary, rondo and ritornello forms.
In sonata style, even these forms and genres often incorporate sonata-form procedures. Variation movements often return to the main theme at the end, following an Adagio variation or a climax, with the clear import of a reprise (Mozart, String Quartet k464, Andante, bar 164; Beethoven, Diabelli Variations, nos.33–4); in the 19th century, analogous effects may follow contrasting sections or even entire movements (Beethoven, Symphony no.9, finale). Haydn’s ‘double variations’ and ABA movements often break off in favour of developmental passages which return to the original theme as if to a recapitulation (Piano Trio hXV:27, Andante). Fugues often betray the influence of sonata style, whether in relaxation of the contrapuntal texture towards the end (Haydn, String Quartet op.20 no.2, finale), quasi-recapitulation of the subject (Mozart, String Quartet k173, finale, bars 52, 61–2), or division into large sections reminiscent of four-movement cycles (Beethoven, Grosse Fuge). Many pieces entitled ‘fantasy’ and the like are in sonata form (Mozart, k396/385f – unfinished, but the intention is clear), and many others include reprises: Mozart’s k475 has a simultaneous return at bar 161, and recapitulates bars 6ff and 10ff at bars 165ff and 171ff; Haydn’s FantasyhXVII:4 (a rondo) recapitulates bars 29–69 at bars 357ff and 423ff.
D.F. Tovey: ‘Sonata Forms’, Encyclopedia Britannica (London, 11/1911; repr. in Musical Articles from the Encyclopedia Britannica (London, 1944; 2/1956 as The Forms of Music)
H. Schenker: ‘Vom organischen der Sonatenform’, Das Meisterwerk in der Musik, ii (Munich, Vienna and Berlin, 1926/R), 45–54; Eng. trans., 1996
H. Schenker: Neue musikalische Theorien und Phantasien, iii: Der freie Satz (Vienna,1935; Eng. trans., 1979, as Free Composition), pt iii, chap.5
R. Tobel: Die Formenwelt der klassischen Instrumentalmusik (Berne, 1935)
K. Westphal: Der Begriff der musikalischen Form in der Wiener Klassik (Leipzig, 1935)
D.F. Tovey: Essays in Musical Analysis (London, 1935–9/R, abridged 2/1981)
E. Ratz: Einführung in die musikalische Formenlehre (Vienna, 1951, 3/1973)
J.P. Larsen: ‘Sonatenform-Probleme’, Festschrift Friedrich Blume zum 70. Geburtstag, ed. A.A. Abert and W. Pfannkuch (Kassel, 1963), 221–30; Eng. trans., 1988, in Handel, Haydn and the Viennese Classical Style, 269–79
E.T. Cone: Musical Form and Musical Performance (New York, 1968), chap.3
C. Dahlhaus: Analyse und Werturteil (Mainz, 1970); Eng. trans., 1983, as Analysis and Value Judgment, 45–84
C. Rosen: Sonata Forms (New York and London, 1980, 2/1988)
H. Macdonald: ‘To Repeat or Not to Repeat’, PRMA, cxi (1984–5), 121–37
M.E. Bonds: Wordless Rhetoric: Musical Form and the Metaphor of the Oration (Cambridge, MA, 1991)
Indiana Theory Review, xii (1991) [whole issue on narrative in music]
J. Schmalfeldt: ‘Towards a Reconciliation of Schenkerian Concepts with Traditional and Recent Theories of Form’, MAn, x (1991), 233–87
C.J. Smith: ‘Musical Form and Fundamental Structure: an Investigation of Schenker’s Formenlehre’, MAn, xv (1996), 191–297
W.S. Newman: ‘The Recognition of Sonata Form by Theorists of the 18th and 19th Centuries’, PAMS 1941, 21–9
L. Ratner: ‘Harmonic Aspects of Classic Form’, JAMS, ii (1949), 159–68
A. Feil: Satztechnische Fragen in den Kompositionslehren von F.E. Niedt, J. Riepel, und H. Chr. Koch (Heidelberg, 1955)
L. Ratner: ‘Eighteenth-Century Theories of Musical Period Structure’, MQ, xliii (1956), 439–54
B. Churgin: ‘Francesco Galeazzi’s Description (1796) of Sonata Form’, JAMS, xxi (1968), 181–99
F. Ritzel: Die Entwicklung der ‘Sonatenform’ im musiktheoretischen Schrifttum des 18. und 19. Jahrhunderts (Wiesbaden, 1968)
C. Dahlhaus: ‘Der rhetorische Formbegriff H. Chr. Kochs und die Theorie der Sonatenform’, AMw, xxxv (1978), 155–77
J. Budday: Grundlagen musikalischer Formen der Wiener Klassik: an Hand der zeitgenössischen Theorie von Joseph Riepel und Heinrich Christoph Koch dargestellt an Menuetten und Sonatensätzen (1750–90) (Kassel, 1983)
J. Stevens: ‘Georg Joseph Vogler and the “Second Theme” in Sonata Form’, JM, ii (1983), 278–304
C. Dahlhaus: ‘Ästhetische Prämissen der “Sonatenform” bei Adolf Bernhard Marx’, AMw, xli (1984), 73–85
R. Gwilt: ‘Sonata-Allegro Revisited’, In Theory Only, vii/5–6 (1984), 3–33
G. Wagner: ‘Anmerkungen zur Formtheorie Heinrich Christoph Kochs’, AMw, xli (1984), 86–112
S. Schmalzriedt: ‘Charakter und Drama: zur historischen Analyse von Haydnschen und Beethovenschen Sonatensätzen’, AMw, xlii (1985), 37–66
S. Burnham: ‘The Role of Sonata Form in A.B. Marx’s Theory of Form’, JMT, xxxiii (1989), 247–72
R. Graybill: ‘Sonata Form and Reicha’s Grande coupe binaire of 1814’, Theoria, iv (1989), 89–105
I.D. Bent: Musical Analysis in the Nineteenth Century (Cambridge, 1994)
W. Fischer: ‘Zur Entwicklungsgeschichte des Wiener klassischen Stils’, SMw, iii (1915), 24–84
V. Helfert: ‘Zur Entwicklungsgeschichte der Sonatenform’, AMw, vii (1925), 117–46
F. Tutenberg: Die Sinfonik Johann Christian Bachs (Berlin, 1928)
R. Benton: ‘Form in the Sonatas of Domenico Scarlatti’, MR, xiii (1952), 264–75
R. Kirkpatrick: Domenico Scarlatti (Princeton, NJ, 1953), chap.11
H.-R. Dürrenmatt: Die Durchführung bei Johann Stamitz (1717–1757) (Berne, 1959)
J. Vinton: ‘The Development Section in Early Viennese Symphonies: a Re-evaluation’, MR, xxiv (1963), 13–22
P. Barford: The Keyboard Music of C.P.E. Bach Considered in Relation to his Musical Aesthetic and the Rise of the Sonata Principle (London, 1965)
R. Kamien: ‘Style Change in the Mid-18th-Century Keyboard Sonata’, JAMS, xix (1966), 37–58
R.M. Longyear: ‘Binary Variants of Early Classic Sonata Form’, JMT, xiii (1969), 162–85
K.F. Heimes: ‘The Ternary Sonata Principle before 1742’, AcM, xlv (1973), 222–48
E.K. Wolf: The Symphonies of Johann Stamitz: a Study in the Formation of the Classic Style (Utrecht, The Hague and Boston, 1981), chaps.9–12
M.M. Fillion: The Accompanied Keyboard Divertimenti of Haydn and his Viennese Contemporaries (c.1750–1780) (diss., Cornell U., 1982), chap.4
B. Churgin: ‘The Recapitulation in Sonata-Form Movements of Sammartini and Early Haydn Symphonies’, Joseph Haydn: Vienna 1982, 135–40
J. Webster: ‘Binary Variants of Sonata Form in Early Haydn Instrumental Music’, Joseph Haydn: Vienna 1982, 127–35
R. Winter: ‘The Bifocal Close and the Evolution of the Viennese Classical Style’, JAMS, xlii (1989), 275–337
NewmanSCE
H. Abert: ‘Joseph Haydns Klavierwerke’, ZMw, ii (1919–20), 553–73
H. Abert: ‘Joseph Haydns Klaviersonaten’, ZMw, iii (1920–21), 535–52
W. Fischer: ‘Instrumentalmusik von 1750–1828’, AdlerHM
D.F. Tovey: ‘Some Aspects of Beethoven’s Art-Forms’, ML, iii (1927), 131–55; repr. in Essays and Lectures on Music, ed. H.J. Foss (London, 1949), 271–97
D.F. Tovey: ‘Haydn’s Chamber Music’, Cobbett’s Cyclopedic Survey of Chamber Music (London, 1929); repr. in Essays and Lectures on Music, ed. H.J. Foss (London, 1949), 1–64
D.F. Tovey: A Companion to Beethoven’s Pianoforte Sonatas (London, 1935/R)
D.F. Tovey: Beethoven (London, 1944)
P. Barford: ‘Formalism in Clementi’s Pianoforte Sonatas’, MMR, lxxxii (1952), 205, 238
F. Neumann: ‘Der Typus des Stufenganges der Mozart’schen Sonatendurchführung’, MJb 1959, 247–61
E.K. Wolf: ‘The Recapitulations in Haydn’s London Symphonies’, MQ, lii (1966), 71–89
H. Andrews: Tonality and Structure in the First Movements of Haydn’s Solo Keyboard Sonatas (diss., U. of North Carolina, 1967)
C. Rosen: The Classical Style: Haydn, Mozart, Beethoven (London, 1971, 2/1972)
R.M. Longyear: ‘The Minor Mode in Eighteenth-Century Sonata Form’, JMT, xv (1971), 182–229
A.P. Brown: ‘The Structure of the Exposition in Haydn’s Keyboard Sonatas’, MR, xxxvi (1975), 102–29
L. Ratner: Classic Music: Expression, Form, and Style (New York, 1980), chaps.12–13
D. Bartha: ‘Song Form and the Concept of “Quatrain”’, Haydn Studies (New York and London, 1981), 353–5
B. Shamgar: ‘On Locating the Retransition in Classic Sonata Form’, MR, xlii (1981), 130–43
P. Cahn: ‘Aspekte der Schlussgestaltung in Beethovens Instrumentalwerken’, AMw, xxxix (1982), 19–31
C. Dahlhaus: ‘“Dritte Themen” in Clementi’s Sonaten? zur Theorie der Sonatenform im 18. Jahrhundert’, AnMc, no.21 (1982), 444–61
J. Kerman: ‘Notes on Beethoven’s Codas’,Beethoven Studies, iii (Cambridge, 1982), 141–59
D. Beach: ‘A Recurring Pattern in Mozart’s Music’, JMT, xxvii (1983), 1–30
M. Broyles: ‘The Two Instrumental Styles of Classicism’, JAMS, xxxvi (1983), 210–42
K. Heimes: ‘Tonal and Thematic Determinants of Form in the Eighteenth-Century Sonata: an Evaluation of Two Modes of Synoptic Comprehension’, South African Journal of Musicology, iii (1983), 1–19
M. Hunter: ‘Haydn’s Sonata Form Arias’, CMc, nos.37–8 (1984), 19–32
P. Dinslage: Studien zum Verhältnis von Harmonik, Metrik und Form in den Klaviersonaten Ludwig van Beethovens (Munich, 1987)
M. Bandur: Form und Gehalt in den Streichquartetten Joseph Haydns: Studien zur Theorie der Sonatenform (Pfaffenweiler, 1988)
M.E. Bonds: Haydn’s False Recapitulations and the Perception of Sonata Form in the Eighteenth Century (diss., Harvard U.,1988)
E. Cavett-Dunsby: ‘On Mozart’s Codas’, MAn, vii (1988), 31–51
L. Finscher: ‘Zur Coda bei Mozart’, Florilegium musicologicum: Hellmut Federhofer zum 75. Geburtstag, ed. C.-H. Mahling (Tutzing, 1988), 79–94
E. Haimo: ‘Haydn’s Altered Reprise’, JMT, xxxii (1988), 335–51
R.G. Hopkins: ‘When a Coda is More than a Coda: Reflections on Beethoven’, Explorations in Music, the Arts, and Ideas: Essays in Honor of Leonard B. Meyer ed. E. Narmour and R.A. Solie (Stuyvesant, NY, 1988), 393–410
V.K. Agawu: Playing with Signs: a Semiotic Interpretation of Classic Music (Princeton, NJ, 1991)
G. Edwards: ‘The Nonsense of an Ending: Closure in Haydn’s String Quartets’, MQ, lxxv (1991), 227–54
W. Drabkin: ‘Beethoven’s Understanding of “Sonata Form”: the Evidence of the Sketchbooks’, Beethoven’s Compositional Process, ed. W. Kinderman (Lincoln, NE, 1991), 14–19
J. Webster: ‘The Analysis of Mozart’s Arias’, Mozart Studies (Oxford, 1991), 101–99
J. Webster: Haydn’s ‘Farewell’ Symphony and the Idea of Classical Style: Through-Composition and Cyclic Integration in his Instrumental Music (Cambridge, 1991)
B. Churgin: ‘Harmonic and Tonal Instability in the Second Key Area of Classic Sonata Form’, Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, ed. W.J. Allanbrook and others (Stuyvesant, NY, 1992), 23–57
W. Kinderman: ‘Integration and Narrative Design in Beethoven’s Piano Sonata in A Major, Opus
110’, Beethoven Forum, i (1992), 111–45
J. Webster: ‘The Form of the Finale of Beethoven’s Ninth Symphony’, Beethoven Forum, i (1992), 25–62
J. Galand: ‘Form, Genre, and Style in the Eighteenth-Century Rondo’, Music Theory Spectrum, xvii (1995), 27–52
E. Haimo: Haydn’s Symphonic Forms: Essays in Compositional Logic (Oxford, 1995)
J. Schmalfeldt: ‘Form as the Process of Becoming: the Beethoven-Hegelian Tradition and the “Tempest” Sonata’,Beethoven Forum, iv (1995), 37–71
J. Hepokoski and W. Darcy: ‘The Medial Caesure and its Role in the Eighteenth Century Sonata Exposition’, Music Theory Spectrum, x (1997), 115–54
B. Churgin: ‘Beethoven and the New Development-Theme in Sonata-Form Movements’, JM, xvi (1998), 323–41
NewmanSSB
D.F. Tovey: ‘Franz Schubert’, The Heritage of Music, i (London, 1927); repr. in Essays and Lectures on Music, ed. H.J. Foss (London, 1949), 103–33
V. Urbantschitsch: ‘Die Entwicklung der Sonatenform bei Brahms’, SMw, xiv (1927), 264–85
F. Salzer: ‘Die Sonatenform bei Franz Schubert’, SMw, xv (1928), 86–125
D.F. Tovey: ‘Tonality in Schubert’, ML, ix (1928), 341–63; repr. in Essays and Lectures on Music, ed. H.J. Foss (London, 1949), 134–59
D.F. Tovey: ‘Brahms’s Chamber Music’, Cobbett’s Cyclopedic Survey of Chamber Music (London, 1929); repr. in ibid., 220–70
H. Mersmann: ‘Sonatenformen in der romantischen Kammermusik’, Festschrift für Johannes Wolf, ed. W. Lott, H. Osthoff and W. Wolffheim (Berlin, 1929), 112–17
H.J. Therstappen: Die Entwicklung der Form bei Schubert, dargestellt an den ersten Sätzen seiner Sinfonien (Leipzig, 1931)
K. Westphal: ‘Die romantische Sonate als Formproblem’, SMz, lxxiv (1934), 45–9, 117–22, 189–92
A. Mitschka: Der Sonatensatz in den Werken von Johannes Brahms (Gütersloh, 1961)
W. Korte: Bruckner und Brahms: die spätromantische Lösung der autonomen Konzeption (Tutzing, 1963)
M. Boyd: ‘Schubert’s Short Cuts’, MR, xxix (1968), 12–21
R. Morgan: The Delayed Structural Downbeat and its Effect on the Tonal and Rhythmic Structure of Sonata Form Recapitulations (diss., Princeton U., 1969)
U. Dammeier-Kirpal: Der Sonatensatz bei Frédéric Chopin (Wiesbaden, 1973)
D. Coren: ‘Ambiguity in Schubert’s Recapitulations’, MQ, lx (1974), 568–82
R. Pascall: ‘Some Special Uses of Sonata Form by Brahms’, Soundings, iv (1974), 58–63
H. Kohlhause: ‘Studien zur Form in den Streichquartetten von Felix Mendelssohn Bartholdy’, HJbMw, ii (1977), 75–104
C. Dahlhaus: ‘Die Sonatenform bei Schubert’, Musica, xxxii (1978), 125–30; Eng. trans. as ‘Sonata Form in Schubert: the First Movement of the G-major String Quartet, op.161 (D.887)’, Schubert: Critical and Analytical Studies, ed. W. Frisch (Lincoln, NE, 1986), 1–12
F. Krummacher: Mendelssohn – Der Komponist: Studien zur Kammermusik für Streicher (Munich, 1978), pt iii, chaps.1–2
J. Webster: ‘Schubert’s Sonata Form and Brahms’s First Maturity’, 19CM, ii (1978–9), 18–35, iii (1979–80), 52–71
D.R. Beveridge: Romantic Ideas in a Classical Frame: the Sonata Forms of Dvořák (diss., U. of California, Berkeley, 1980)
R.F. Jones: Thematic Development and Form in the First and Fourth Movements of Mahler’s First Symphony (diss., Brandeis U., 1980)
S. Winklhofer: Liszt’s Sonata in B minor: a Study of Autograph Sources and Documents (Ann Arbor, 1980), 41–9, 115–68
R. Kaplan: ‘Sonata Form in the Orchestral Works of Franz Liszt: the Revolutionary Reconsidered’, 19CM, viii (1984–5), 142–52
R.M. Longyear and K. Covington: ‘Tonic Major, Mediant Major: a Variant Tonal Relationship in 19th-Century Sonata Form’, SMC, x (1985), 106–39
A. Newcomb: ‘Schumann and Late Eighteenth-Century Narrative Strategies’, 19CM, xi (1987–8), 164–74
T. Denny: ‘Articulation, Elision, and Ambiguity in Schubert’s Mature Sonata Forms: the op.99 Trio Finale in its Context’, JM, vi (1988), 340–66
R. Graybill: ‘Brahms’ Integration of Traditional and Progressive Tendencies: a Look at Three Sonata Expositions’, JMR, viii (1988), 141–68
R.M. Longyear: ‘Sources of the Three-Key Exposition’, JM, vi (1988), 448–70
G.J. Vitercik: ‘Mendelssohn the Progressive’, JMR, viii (1989), 333–74
J. Webster: ‘The General and the Particular in Brahams’s Later Sonata Forms’, Brahms Studies: Washington DC 1983, 49–78
J. Samson: Chopin: the Four Ballades (Cambridge,1992)
M. Heinemann: Liszt: Klaviersonate h-Moll (Munich,1993)
J. Hepokoski: ‘Masculine-Feminine’, MT, cxxxv (1994), 494–9
J. Daverio: ‘From “Concertante Rondo” to “Lyric Sonata”: a Commentary on Brahms’s Reception of Mozart’, Brahms Studies, i (1994), 111–36
H.J. Hinrichsen: Untersuchungen zur Entwicklung der Sonatenform in der Instrumentalmusik Franz Schuberts (Tutzing, 1994)
P. Smith: ‘Brahms and Schenker: a Mutual Response to Sonata Form’, Music Theory Spectrum, xvi (1994), 77–103
J. Lester: ‘Robert Schumann and Sonata Forms’, 19CM, xviii (1994–5), 189–210
K. Berger: ‘The Form of Chopin’s Ballade, Op.23’, 19CM, xx (1996–7), 46–71
K. Westphal: ‘Die Sonate als Formproblem der modernen Musik,’ Monatschrift für moderne Musik, xi (1929), 160–63
W.W. Austin: ‘Traditional Forms in New Musical Idioms’, IMSCR VIII New York 1961, 100–08
G. Perle: Serial Composition and Atonality (Berkeley, 1962, 6/1991), chap.6
E.T. Cone: ‘Stravinsky: the Progress of a Method’, PNM, i/2 (1962–3), 18ff; repr. in E.T. Cone and B. Boretz, eds.: Perspectives on Schoenberg and Stravinsky (Princeton, 1968, 2/1972), 155–64
G. Perle: ‘The Musical Language of Wozzeck’, Music Forum, i (1967), 204–59
S. Borris: ‘Die Krise der Sonata im 20. Jahrhundert’, Musa–mens–musici: im Gedenken an Walther Vetter (Leipzig, 1969), 361–78
C. Dahlhaus: ‘Die Neue Musik und das Problem der musikalischen Gattungen’, Gestaltungsgeschichte und Gesellschaftsgeschichte (Stuttgart, 1969); repr. in Schönberg und andere: Gesammelte Aufsätze zur Neuen Musik (Mainz, 1978); Eng. trans., 1987, in Schoenberg and the New Music, 32–44
K. Schweizer: Die Sonatensatzform im Schaffen Alban Bergs (Stuttgart, 1970)
R. Travis: ‘Tonal Coherence in the First Movement of Bartók’s Fourth String Quartet’, Music Forum, ii (1970), 298–372
H. Keller: ‘Schoenberg: the Future of Symphonic Thought’, PNM, xiii/1 (1973–4), 3–20
H. Fladt: Zur Problematik traditioneller Formtypen in der Musik des fruhen 20. Jahrhunderts, dargestellt an Sonatensätzen in den Streichquartetten Bela Bartóks (Munich, 1974)
A. Forchert: ‘Zur Auflösung traditioneller Formkategorien in der Musik um 1900’, AMw, xxxii (1975), 85–98
J. Forner: ‘Sonatenform-Probleme bei Béla Bartók: eine vergleichende Studie’, JbMP 1981–2, 62–75
P. Evans: ‘Sonata Structures in Early Britten’, Tempo, no.82 (1987), 2–13
J. Straus: ‘Sonata Form in Stravinsky’, Stravinsky Retrospectives, ed. E. Haimo and P. Johnson (Lincoln, NE, 1987), 141–61
T.M. Davidian: Debussy’s Sonata Forms (diss., U. of Chicago, 1988)