Francisco Mignone e os Manuscritos de Buenos Aires: contexto histórico, interpretação e edição de obras para duo de violões recém-descobertas.

Abstract

The research object of this thesis is a collection of autograph manuscripts containing guitar duets written in August of 1970 by Brazilian composer Francisco Mignone. These works (1ª Valsa Brasileira, 2ª Valsa Brasileira, Canção, Lundu and Quatro Peças Brasileiras), which we called the Buenos Aires Manuscripts, belong to Martínez Zárate collection housed at the National Institute of Musicology "Carlos Vega" in the Argentine capital. Most of these works have remained unknown to the musical and academic communities until its recent discovery. Our approach to these works followed three main research lines: musicological, interpretative and editorial. From the musicological point of view, we were primarily interested in the circumstances surrounding their composition, especially because they afforded us a more refined and detailed look at a crucial moment for Mignone's guitar work, which was his (re)discovery of the instrument in July of 1970. Our approach to the interpretation of the works contained in the Buenos Aires Manuscripts was based on analyses of recorded performances of several of them by the composer himself at the piano. This methodological approach situated this part of our research in the field of performance practices studies, which prompted a reflection on the justifications and validity of this type of study in view of the current debate about interpretation and the role of the performer. Finally, we presented our bases for the preparation of performance editions of the pieces, which was anchored in analyses of the procedures that Mignone himself used when elaborating his versions for guitar duo of these works, almost all of them composed originally for the piano. The conceptual framework for our editing work was built on the examination of the ideas of musical and textual philologists, in an attempt to establish the limits of the editors interventions, which, we concluded, are conditioned by the realization that editing is, ultimately, an interpretative, and thus subjective enterprise.