« Timbre Analysis in the Mirror of Compositional Process: A Case Study of Chopin’s Berceuse Op. 57 »

     Timbre is one of the most fascinating dimensions when considering the creative process in music: Potentially inscribed in the written object constituted by the score, however only actualized through sound objects achieved by instrumental execution – to use Delalande’s terminology –, timbre may be considered not only as the colour of sound, but as a metadimension integrating all musical parameters (Dufourt). Timbre-oriented analysis has strongly developed in recent years, but generally regardless of the consideration of the compositional processes themselves. However, to which extent can a timbral formal structure be considered as the result of a compositional strategy – whether implicit or explicit? How can analytical representations of a piece’s timbral organisation shed light on some aspects of the composer’s representations and processes in the elaboration of musical forms?

     These questions will be discussed through a case study  of  Chopin’s  Berceuse  Op. 57, composed in 1843–44 and published in 1845. The sources and data will include, besides the Berceuse’s score (ed. Ekier), several pre-texts (avant-textes) such as an autograph clean fair copy and, above all, an autograph draft (see Kobilańska, 1977) showing, by means of rewriting and erasure marks, Chopin’s reorganisation of the piece’s overall form – originally conceived as a series of four bars “variantes” – through a succession of modifications affecting the different sections’ order. On the basis of the reconstruction of the piece’s genesis, scores for the different stages of the piece’s composition will be elaborated, as well as recordings of these preliminary and final versions, specifically produced for the present research in collaboration with pianists from the Conservatoire of  Strasbourg.

     This written and sound material will serve as a basis for a timbre-oriented analysis  of the Berceuse’s formal elaboration process. It will be particularly interesting to examine the successive reconstructed versions of the piece by means of a timbre- oriented analytical method called Gestaltist analysis, developed following Roy (2003) and Thoresen (2015). Based on comparative sonogram representations and transcriptions elaborated using software like Acousmographe, this method reveals  in particular a striking and unusual timbral overall structure of the Berceuse’s final version organised around a central axis of symmetry. The analysis of the progression, throughout the piece, of timbral  factors like horizontal density, harmonicity, or the number of simultaneous polyphonic parts, in conjunction with their pianistic realisation examined by means of audio descriptors, will also underline several formal processes oriented towards timbral climaxes at local and global levels, as well as their compositional elaboration from the preliminary to the final versions of the piece.