But there are more sophisticated reasons to sketch: most of classical music arranges the themes of each movement into a substantial architecture, for instance involving sonata form.
The eighteenth century theorists H. C. Koch and J. G. Sulzer, in their written advice to composers, suggested that they should prepare sketches that would lay out how the various themes of the work would be arranged to create the overall structure.
[2] Marston adds that Koch and Sulzer's recommendations "do in fact accord well with what scholars, borrowing from the terminology developed in relation to Beethoven's sketches, call a 'continuity draft', a notational form in which [for example] 'Beethoven can be seen fitting together the more fragmentary ideas made earlier into a coherent whole' (Cooper, 1990, p. 105).
The watermark studies of Alan Tyson yielded among other results the conclusion that Mozart would sometimes leave a work only partially complete (in sketch form) for a number of years, then finish it when an opportunity for performance arose.
[9] This in turn has been taken to support the view that Mozart carefully retained his sketches simply as a good business practice, keeping open the possibility of future performances and publication for works not immediately promising in this respect.
[12][13][14][15][16][17][18][19][20][21][22][23][24][25] Schubert's unfinished piano music ranges from rejected sketches of a few bars (e.g. D 309A) to fairly elaborate drafts with several complete movements (e.g. Reliquie sonata).