B eyond ‘N orms and D eformations

Although published as recently as 2006, James Hepokoski's and Warren Darcy's much heralded and substantially delayed Elements of Sonata Theory has been a significant presence in the field of music theory and analysis for over a decade.1 Indeed, one reads this monumental work with an unavoidable sense of déjà-lu. Draft copies were frequently quoted in conference papers and articles for several years prior to publication. Also, many of the book's key precepts are aired by both authors in a variety of publications dating from 1992 onwards dealing with works by Beethoven, Bruckner, Sibelius and Richard Strauss amongst others, as well as historical issues such as the reception of Beethoven's symphonies and theoretical concepts including the so-called ‘sonata principle’, ‘rotation’, ‘deformation’ and the ‘medial caesura’.2 Given the intense level of advance exposure for the authors’ ideas and the wealth of insightful precursor texts about sonata form in the High Classical Era, it is not self-evident that there is actually a gap in the existing literature for Elements of Sonata Theory to fill, all the more so since the book was pre-empted by its main market competitor, William Caplin's commanding and elegantly concise Classical Form.3 Nevertheless, closer inspection reveals that the publication of Elements of Sonata Theory is justified at least by its encyclopaedic aspect, by the incorporation of just enough new material (the concluding three chapters on Mozart's concertos in particular) and by the elaboration and refinement of some fundamental premises. Chapters 1 to 4 situate Sonata Theory within the field and introduce its core precepts. Hepokoski and Darcy broadly identify three main strands in existing thought about sonata form: the ‘sonata principle’ elaborated first by Edward Cone and then by Charles Rosen, which stresses the notion that sonata movements dramatise fundamental properties of the Classical Style, especially polarisation and resolution, and which thus requires that non-tonic material in the secondary and closing areas of the exposition be recapitulated in the tonic or else ‘brought into a closer relation’ with it (p. 242); what Mark Evan Bonds terms the ‘conformational view’, propounded initially in nineteenth-century Formenlehren beginning with those of Carl Czerny and A. B. Marx, which essentially sees sonata form as an architectural or tectonic blueprint; and what Bonds labels the ‘generative view’, first expounded in detail by Schoenberg, which regards sonata forms as products of material process.4 In contrast, Sonata Theory claims to transcend all these viewpoints, positing that classical composers are in ‘dialogue’ with a constellation of ‘generic defaults’, which are hierarchically organised according to frequency of usage. When classical composers override ‘standard options’, they ‘deform’ generic conventions. Naturally, the ‘genre sonata form’ is subject to ‘diachronic transformation’, with the result that constellations of norms undergo incremental change: a deformation in, for example, Beethoven can ‘become a lower-level default in Schumann, Liszt or Wagner’ (p. 11). Owing to the progressive reification of sonata ‘defaults’ through theoretical abstraction, departures from the norm in works by nineteenth-century composers are increasingly in dialogue with textbook models. Nevertheless, the authors contend that most of the ‘sonata norms remained in place as regulative ideas throughout the nineteenth century’ (p. vii). In order to construct a convincing analysis of any classical movement, Sonata Theory maintains that one has to identify ‘the essential generic markers’ of sonata works in the High Classical Style. These markers form a hierarchically organised ‘generic layout’, shaped on the highest level by an ‘essential sonata trajectory’ (EST) that comprises the three ‘action zones’ or ‘rotations’ traditionally labelled ‘exposition’, ‘development’ and ‘recapitulation’. The exposition unfolds an ‘essential expositional trajectory’ (EET) and culminates in an ‘essential expositional closure’ (EEC), usually in the form of a perfect authentic cadence (PAC) in a non-tonic key. The EEC is paralleled in the recapitulation by an ‘essential structural closure’ (ESC) that affirms the tonic. The EET breaks down into four principal ‘spaces’– primary, secondary, transitional and closing (P, S, TR and C) – with TR and S being demarcated by ‘a mid-expositional break or medial caesura (MC)’. The EET is ‘launched’ by P, supplied with ‘energy gain’ in TR, ‘relaunched’ by S and closed by C. On a lower level, the four ‘spaces’ within the expositional and recapitulatory action zones are made up of ‘spans’ punctuated by clear PACs, and the spans are themselves normally further broken down into ‘modules’. Since the EET concludes in a non-tonic key, it is a ‘structure of promise’, whereas the full, tonic-directed ESC constitutes a ‘structure of accomplishment’. Many sonata-form movements of course have slow introductions and/or codas, which are classified as ‘parageneric’ areas lying outside sonata space. Hepokoski's and Darcy's view of sonata form is thus ostensibly a mixture of temporal and spatial concerns. It is also predominantly goal-directed – concerned more with endings than beginnings – and tonally orientated, despite the distinction the authors draw between what they term ‘tonal form’ and ‘rhetorical form’, which ‘includes personalized factors of design and ad hoc expression’ (p. 23). Indeed, the EST is basically a reformulation of an interrupted Schenkerian Ursatz.5 Chapters 5 to 13 are devoted to fleshing out the details of the ‘generic layout’. Chapter 14 deals with issues specific to the minor mode and Chapter 15 examines various properties of the ‘three- and four-movement sonata cycle’. The final seven chapters (16 to 23) propose and extensively elaborate a taxonomy of five sonata types. Type 1 (common in slow movements and overtures) comprises only an exposition and a recapitulation with no or minimal link and is often called ‘sonata without development’, ‘exposition-recapitulation’, ‘slow-movement sonata’ or ‘sonatina’ form. Type 2, often labelled ‘binary’ or ‘polythematic binary’ sonata form, lacks a ‘full’ recapitulation and has instead what Hepokoski and Darcy prefer to call a ‘tonal resolution’ occurring in conjunction with secondary or, less commonly, transitional material. Type 3 is the standard sonata model with a development and a recapitulation usually, but not invariably, beginning with the opening theme in the tonic. Type 4 is what is generally known as the ‘sonata rondo’. Finally, Type 5 is the hybrid of ritornello (tutti-solo) principles and other sonata types (usually Type 3) employed in concertos. There are two concluding appendices further elucidating some of Sonata Theory's ‘grounding principles’ and terminology. It is of course impossible to deal with all the intricacies of the authors’ arguments without writing another book, so this essay is confined to pursuing some salient issues: Sonata Theory's generic affiliations; its three ‘fundamental axioms’ of the ‘genre sonata’, ‘rotation’ and ‘deformation’; the taxonomy of five sonata types; and what I have termed the book's ‘neologising impulse’. The main aim of my concluding remarks is to sketch an alternative approach with particular reference to sonata-form works of the first half of the nineteenth century. As its title implies, the model for Elements of Sonata Theory is ostensibly the scientific textbook, a genre that essentially requires in excess of 500 double-column pages (which is what we get – 661 pages including appendices and indices, to be precise).6 The book's generic allegiance is confirmed on the first page of the Preface: ‘From one perspective the Elements is a research report, the product of our analyses of hundreds of individual movements by Haydn, Mozart, Beethoven, and many surrounding composers of the time (as well as later composers)’ (p. v). Hence both vocabulary and symbology rely heavily on scientific conventions. As far as the former is concerned, instead of sections and themes or thematic groups, there are ‘actions zones’ and ‘spaces’, and the general discourse is throughout liberally peppered with ‘trajectories’, ‘vectors’, ‘rotations’ and the like. Even the problematic and much-debated term ‘deformation’ is justified by analogy to usage in the physical sciences: ‘“deformation” is descriptive of a certain state of a solid object – a change of shape, a departure from its original, normal, or customary state resulting from the application of a force’ (p. 619). As a general rule, sentences seem constructed to maximise the number of abbreviations and quasi-scientific buzzwords. The description of expositional strategy in Chapter 2 is typical: ‘The large dotted-line arrow in figure 2.1a suggests a broadly vectored trajectory from the start of the exposition to the EEC; the smaller dotted-line arrow below it suggests a subordinate trajectory from the beginning of S to its own point of PAC-closure at the EEC’ (p. 18). As regards symbols, the diagrams themselves of course draw on mathematical graphing conventions (see in particular figures 2.1a and 2.1b on p. 17); all that is really missing is the use of Greek letters. The different categories of ‘medial caesura’ are allocated elaborate designations such as ‘V: PAC MC’; the different ‘spans’ of P, S, TR, C and the rest are designated by superscript integers (‘P1’, ‘P2’ and so on); and within these spans, any smaller ‘modules’ are identified by decimalised superscript integers (‘P1.1’, ‘P1.2’ etc.). A variety of further symbols is employed for different types of large- and small-scale function: ‘’ denotes ‘mergers’ or elisions, subscript letters are added to functional chord symbols (for example, ‘VT’ distinguishes a tonicised dominant from one that is sounded but not tonicised, which is designated ‘VA’), and so forth. The notation becomes particularly involved where concerto (Type 5) movements are concerned, as the additional ritornello-solo aspect of the structure spawns extra colons and backwards slashes: an individual module within ‘P-space’ in the opening orchestral ritornello is, for instance, identified as ‘R1:∖P1.1’. When entire ‘action-spaces’ are summarised, convoluted quasi-mathematical formulae result. For example, the ‘recapitulatory rotation’ of the evidently deceptively approachable Finale of Mozart's Piano Sonata in C major, K. 309 (1777), merits the following near-impenetrable sequence: ‘Prf[S1.4—(’) Episode S1.2—] (’) S1.1 S1.3 S1.4 RT! (’) [Prf!! S1.2!!](’) C [S1.4!!]’ (p. 412). As the Preface leads the reader to expect, scientific metaphor also governs the presentation of key concepts and the evaluation of major ‘generic markers’ within movements. In fact, the initial elucidation of the central idea of a hierarchy of generic defaults and deformations is defined in terms likely to appeal to the most hard-core of computer enthusiasts: For novice-composers, one might wittily fantasize . . . something on the order of an aggressively complex ‘wizard’ help feature within a late-eighteenth-century musical computer application, prompting the still-puzzled apprentice with a welter of numerous, successive dialog boxes of general information, tips, pre-selected weighted options, and strong, generically normative suggestions as the act of composition proceeded. (p. 10) As a result, the chapter on the medial caesura (pp. 23–50), for instance, exudes statistical propriety, establishing a four-tier hierarchy of defaults on the basis of frequency of occurrence in the sample of ‘hundreds of individual movements’ and then defining the structural role of each type of medial caesura partly in percentage terms: ‘Our research suggests that the deployment of the I: HC MC is flexible, occurring typically within the 15–45 percent range’ (p. 37); ‘When selected, the V: HC MC option is typically placed from about 25 to 50 percent (more rarely, 60 percent)’ (p. 39); and so on. This scientific orientation of Elements of Sonata Theory worries me, for it promises rather more than it delivers. To begin with, in a scientific ‘research report’ one would expect a full account of the sample, complete descriptive statistics and an explanation of sampling methodology. In this particular case, the reader could derive reassurance from confirmation that careful consideration had been given to the chronological, geographical and generic distribution in the selection of movements. Given Sonata Theory's emphasis on hierarchies of defaults, one would also expect at least some basic statistical analysis. An examination of modal frequency, standard deviation and regression, for example, would clearly add much valuable definition to the bare percentages quoted with regard to the deployment of different types of medial caesura. Unfortunately, readers are obliged to do the spadework for themselves. The sample can of course be reconstructed from the Index of Works (pp. 639–48). Altogether, 665 sonata movements are cited in the book.7 The sample is heavily weighted towards the period c. 1750–90. That is not necessarily a problem given the book's subtitle, but the small number of movements (59, or 8.8%) written by composers born after 1800 does not obviously imbue with authority the authors’ claim in the Preface that their theory provides a ‘foundation for considering works from the decades to come’ (p. vii); all the more so, since nearly all such pieces referred to are overtures or the first movements of symphonies and more than a quarter of them are by a single composer, specifically Brahms. Even amongst composers born before 1800, Mozart seems unduly prominent; in fact, his 228 movements constitute 34% of the overall total. When one turns to the 87 actual musical examples drawn from sonata movements, the skewed nature of the sample becomes yet more troublesome. Fig. 1 breaks these down by composer and genre. Apart from a single overture, only four genres are represented (concerto, keyboard sonata, symphony and string quartet). Not one movement after Beethoven is actually accorded a musical example and the latest piece to be included is the second movement of Beethoven's Symphony No. 6, Op. 68 (1808). A colossal 76% of the examples are taken from Mozart's works, 42% of which are concertos. Tellingly, more than a quarter (26%) come from just six Mozart pieces: Piano Concertos Nos. 9, K. 271 (1777), and 21, K. 467 (1785); the Piano Sonata in F major, K. 280 (1775); Symphonies Nos. 39, K. 543 (1788), and 40, K. 550 (1788); and the String Quartet in C major, K. 465 (1785). Hepokoski's and Darcy's musical examples classified by composer and genre The impression given by all of this is that Sonata Theory has been constructed mainly on the basis of a relatively restricted Mozartian corpus, an impression that is reinforced when one scrutinises pieces cited in the text but not dealt with in any detail. It is not, for instance, evident that the authors conducted independent analyses of any of the seventeen Clementi piano-sonata movements to which they refer. All the analytical information supplied can be found in Leon Plantinga's 1977 monograph, which contains some implausible analytical interpretations.8 By way of an example, Fig. 2 summarises the structure of the Finale of Clementi's Piano Sonata in B minor, Op. 40 No. 2 (published 1802). This has a lengthy slow introduction adumbrating the core components of the primary material, shown in Ex. 1a , Ex. 1b . There follows a relatively uncontroversial ‘Allegro’ exposition with clear relative-major secondary and closing areas (bars 452 and 63 respectively) prepared by a minor-inflected medial caesura. The retention of minor colouring for the first bar of the secondary zone creates a The development the main theme in of secondary and transitional material, concluding at bar with the medial caesura to the dominant of bar the of the second theme is then in the concluding with an interrupted which at bar a transitional the this point there is an of the slow introduction by a lengthy that is by a of the main A convincing analysis of this individual movement would have to account for the that what Hepokoski and Darcy would term a 2 sonata with into dialogue with the categories of the and a of the all the as the many core of Sonata Theory – particularly in its to the (the of the of the expositional after from p. – yet the authors seem to that Plantinga's analysis is and Clementi's movement in as an example of a piece in which a slow introduction before the In fact, Clementi's strategy seems much to Hepokoski's and Darcy's the of the slow introduction the tonic of the primary material, to which it is of Piano Sonata in B minor, Op. 40 No. 2, Piano Sonata in B minor, Op. 40 No. Piano Sonata in B minor, Op. 40 No. the on Mozart's music would not this broadly both a of the classical and the central point of reference for later a of the suggests that in many key Mozart was areas given in Elements of Sonata Theory are the concerto and the minor In terms of and nineteenth-century of Mozart's concerto are For example, in the first movements of Mozart's concertos his default in the opening ritornello is to state the secondary material in the only a single concerto K. of has an the non-tonic secondary material, and the is within The in the period the of the nineteenth and to a tonally a found in five of seven five of six of two of two of four by two by three by Beethoven, No. 2, and many Mozart's concertos not to have as for major nineteenth-century Piano No. 1 an dialogue with Piano and in F minor Piano much of its sonata and thematic material from Piano No. The the minor mode is Mozart's is for a exposition by a recapitulation in which the relative-major secondary and closing material is in the tonic This of approach is, was much more in his frequently the major mode for various of secondary and closing material in his – the Finale of the Piano No. in minor, a exposition with a recapitulation in the major Beethoven's are also more has identified two principal ‘the of C minor for its major and the of other minor their minor Beethoven's in C minor, Op. has a as a type of or in Elements of Sonata and a non-tonic recapitulation (pp. and and his in F minor, Op. what Hepokoski and Darcy as a (p. Beethoven's according to the norms of the but Beethoven actually his for with a of minor movements to his sonata than Haydn, Mozart or Beethoven, and frequently as well as non-tonic and In fact, in Clementi's keyboard his most to a exposition is a non-tonic In Elements of Sonata Theory is all a book about Mozart particularly his a which the authors’ claims of historical and geographical Given their sample, it is impossible for the reader to what to of such as the the or – of less and works – is from the perspective of the five sonata are likely to be (p. a more and title might have been more Caplin's Classical is more A Theory of for the of Haydn, Mozart, and Beethoven and to its By to the for that can be as the basis for and of Hepokoski and Darcy actually seem to have the of the valuable they have to about a specific The quasi-scientific vocabulary is in an by physical and I the ‘rotation’ and in a variety of on a A scientific definition of ‘rotation’ is in a through a given a description that in any a is a as well as by a drawn from its to its final It is of course impossible to of these with Sonata Theory's In that ‘rotation’ would to of a thematic and the trajectory from the start of the exposition to the EEC’ seems to constitute more than a from an initial tonic to an cadence in a secondary key. The of readers is further when they are to impossible such as a ‘generic when through Elements of Sonata Theory one is to the and (pp. and the only to that much and more alternative is might at least expect the of symbols and abbreviations to be but there are The for the initial integers to labels to be not on to only after a first PAC has been (p. Unfortunately, In for example, with music that to a more that music is designated as there is no PAC between and The is yet more when an exposition has an medial In such the standard labels are to be by and the of a in most the only a single (p. the is not the use of two of that a standard exposition with one medial caesura is different from an exposition a Fig. 3 summarises Hepokoski's and Darcy's analysis of the exposition of Beethoven's Piano Sonata in C major, Op. 2 No. 3 of (pp. 25 to are as a HC medial with The theme to and at bar into transitional This leads to a V: at bar and to a in the at bar In contrast, as the and the of without to new There is no of between the two both the theme at bar to but to seems more to the In the medial are not restricted to sonata forms and in within movements. Beethoven was that two later an theoretical of the sonata-form of his would major on the basis of the number of medial in a does it seem likely that would have thought his strategy in the of the first movements of Op. 2 No. 3 and his published sonata, Op. in to be essentially the former has two and the only of Hepokoski's and Darcy's analysis of Beethoven, Piano Sonata in C major, Op. 2 No. the smaller level, are yet more The for from ‘P1.2’ and are thematic and to Caplin's a of by a the A variety of other The reader is actually that ‘the of is no but a to be by the individual as or sees (p. there is a between the and the of Sonata the Hepokoski and Darcy have an with ‘the of analytical which they identify in the work of and amongst others, and from which they ostensibly themselves. The sense of by the book's of scientific is by the in some of its towards To begin with, there are the of all to are by the despite the that has that of have no with regard to the a of Sonata Theory seems to view as Mozart has to override a default to Beethoven is to be only in The of to of the is more especially to those first is not the musical and sonata movements are the of not The exposition in the first movement of Mozart's No. 4 in K. is particularly with the and up a with to on the terms that have to we The with the into the with a and opening the to the more TR that a (p. to all this is the authors’ of of the quasi-mathematical the Finale of Mozart's Sonata K. 309 quoted really to be several that in Mozart's concertos it is the norm for to begin in the tonic in the opening one can that it is for a book to a of the in Elements of Sonata Theory a sense of about generic and the book's The surrounding the generic and much of the of Elements of Sonata Theory have a on the of central concepts enough to further elucidation in the appendices (pp. the ‘genre and The notion of sonata form as a genre rather than a form’ is not it at

Art · Art history · Concerto · Literature · MOZART · Piano · Symphony · Diverse Musicological Studies · Musicians’ Health and Performance · Musicology and Musical Analysis · Philosophy

• Liszt and the Symphonic Poem

  Joanne Cormac•Liszt and the Symphonic Poem•2017

  Add to list

• Bantock and Southey

  Michael Allis•Music and Letters•2014

  Add to list

• Memesatz contra Ursatz

  Open Access•Steven Jan•Musicae Scientiae•2010

  Add to list

• Musical Form

  Open Access•Edward Klorman•Music Theory Online•2014

  Add to list

• The Evolution of Chopin’s Sonata Forms

  Open Access•Andrew Aziz, Andrew I Aziz•Music Theory Online•2015

  Add to list

• Scriabin’s Lisztian-Stravinskian Sonata Forms

  Open Access•Rajan Lal•Journal of the Royal Musical…•2025

  Add to list

• R hetorical C losure in the F irst M ovement of S chubert ' s Q uartet in C major , D. 46

  Open Access•Anne M Hyland•Music Analysis•2009

  Add to list

• Mutual Deformity

  Open Access•Benedict Taylor•Music Analysis•2016

  Add to list

• The Development‐Ending Dominant and the Late Cadence in vi

  Open Access•BENJAMIN DUANE•Music Analysis•2019

  Add to list

• Structural Acceleration in Nielsen'sSinfonia espansiva

  Open Access•Christopher Tarrant•Music Analysis•2019

  Add to list

• The Formal Ramifications of Bruckner's Bipartite Sonata Form

  Open Access•Eric Lai•Music Analysis•2018

  Add to list

• In Search of Romantic Form

  Open Access•Steven Vande Moortele•Music Analysis•2013

  Add to list

• Apparent Type 2 Sonatas and Reversed Recapitulations in the Nineteenth Century

  Open Access•Steven Vande Moortele•Music Analysis•2021

  Add to list

• Parallel Binary or Tripartite? Formal Hybridisation of Sonata Types in the Nineteenth Century

  Open Access•Peter H Smith•Music Analysis•2021

  Add to list

• Process and Pendulum in Schubert's Expanded Type 1 Sonatas

  Open Access•Judith Guez•Music Analysis•2021

  Add to list

• Compound Aaba Form and Style Distinction in Heavy Metal

  Open Access•Stephen S Hudson, Hudson•Music Theory Online•2021

  Add to list

• A Myriad of Exceptions? Challenging the Sonata-Form Typology for the Classical Repertoire

  Open Access•Markus Neuwirth•Music Analysis•2021

  Add to list

• Form and Orbital Tonality in the Finale of Bruckner's Seventh Symphony

  Open Access•Julian Horton•Music Analysis•2018

  Add to list

• The 'Continuous Exposition' and the Concept of Subordinate Theme

  Open Access•William E Caplin, Nancy J Martins et al.•Music Analysis•2016

  Add to list

• Schubert's Bi-rotational Sonata Forms

  Open Access•Anne M Hyland•Music Analysis•2021

  Add to list