Sonata Principles

Textbooks on sonata form have long been out of fashion. The phrase 'sonata form' was coined by nineteenth-century pedagogues and today it is widely held that their work poorly models eighteenth-century practice and deviates from the concerns of eighteenth-century theory. Much modern scholarship on the period has been devoted to overcoming the nineteenth-century legacy. Donald Francis Tovey and Leonard Ratner led the way,¹ and in their wake there developed a pervasive scepticism towards all definite rules for late eighteenth-century sonata composition: there is always an exception to the rule, and it is often a masterpiece. So James Hepokoski andWarren Darcy's Elements of Sonata Theory is a bold statement.² It is palpably a textbook, at times even resembling a technical manual (text printed in double columns, 'troubleshooting' sections for the student). Its dimensions are monumental (661 pages), its treatment a model of rigour, its coverage of the standard repertory extensive. The book presents a novel method for the analysis of late eighteenth-century sonatas, albeit one that has been trailed in earlier journal articles. Its arguments will take years, if not decades, for scholarship to absorb and digest, and it opens countless possibilities for future research that music theorists are unlikely to ignore. It seems that nineteenth-century Formenlehre has made a surprise comeback at the start of the twenty-first century.

But there is a vital difference: although the authors seek to relegitimize the use of norms, types, and rules for the analysis of late eighteenth-century sonata movements, they do so by way of twentieth-century literary theory. The sonata norms and types they propose are not conceived as transcendent ideals, and the existence of exceptions is to be expected. Norms are inferred inductively from common practice and are understood to regulate generic contracts between composer and listener; actual sonatas are in dialogue with these norms. At each stage of a sonata movement the composer can select from various default options (first-level, second-level, and so on), and the analyst must examine and interpret these choices. The deliberate breaking of a contract—a more decisive step than the selection of a low-level default—amounts to a sonata 'deformation', which is all the more significant for interpretation. The authors dispute what they see as the excessively tonal orientation of mid-twentieth-century accounts of sonata form. They are interested in the rhetoric of thematic 'rotations' (the presentation and re-presentation of thematic material in a fixed order in the exposition, the recapitulation, and also, in their view, the development) and in the generic significance of caesuras and cadences. Thus Sonata Theory, as the authors call it, has to be built from first principles. After two introductory chapters, seven chapters are devoted to the exposition, one to the development, two to the recapitulation, and one to 'parageneric spaces' (introduction and coda). There follow chapters on sonata forms in minor keys and on the multi-movement sonata cycle. The last seven chapters deal with various sonata 'types'—the forms that in earlier accounts might have been called 'slow-movement sonata form', 'sonata-rondo form', and 'concerto form'. The last four of these chapters amount to nothing less than a substantial volume of its own on Mozart's concertos.

For a book that is ostensibly concerned with communication, Elements of Sonata Theory makes few compromises with its own readers. Like a technical manual, the book has not been written with a view to cover-to-cover linear reading. It will test the patience of even the most sympathetic reader who turns the pages in chronological order, or even the student who studies it as part of a course. Some chapters contain lengthy classificatory sections that outline formal typologies or list compositional options in descriptive fashion. Arguments are justified in the main text by the quick-fire citation of examples drawn from across the sonata literature; these are piled upon one another in a way that is annoying to a reader who is not surrounded by the relevant scores open at the correct pages (or who does not possesses encyclopaedic knowledge and a photographic memory). All this...