The Creative Process Behind Dialogismos I: Theoretical and Technical Considerations Gilberto Bernardes 1 , Nuno Peixoto de Pinho 2 , Sofia Lourenço 2, 3 , Carlos Guedes 1, 3 , and Bruce Pennycook 4 , Erik Oña 5 1 University of Porto, Faculty of Engineering; 2 CITAR – Center of Research in Science and Technology of the Arts, School of the Arts, Portuguese Catholic University; 3 Polytechnic of Porto, ESMAE; 4 University of Texas at Austin; 5 Basel Musik-Akademie Abstract — This paper examines the aesthetic dimension and the technical realization of Dialogismos I, a piece for saxophone alto and electronics by the composer Nuno Peixoto de Pinho. The conceptual basis of the work relies on the notion of ‘intertextuality’ coined by the Bulgarian-French philosopher and literary critic Julia Kristeva, which was somehow transposed to the music domain by J. Peter Burkholder under the concept ‘musical borrowing’. The compositional problems raised by applying an intertextual musical thinking as a key driver of the composition were solved using two different approaches. The first approach was the manual selection of elements from several music works with different granularities to devise the overall structure of the work and to create the saxophone score. The second approach was applied to the realization of the electronic part and relied on concatenative sound synthesis as an algorithmic computer assisted composition method and a real-time synthesis technique. Index Terms — Music, intertextuality, concatenative sound synthesis. I. INTRODUCTION This paper details the creative process behind Dialogismos I, a piece for alto saxophone and live electronics, which resulted from a collaborative work between the composer Nuno Peixoto de Pinho and the saxophonist and programmer Gilberto Bernardes. Our aim was to create a new composition entirely based on decontextualized dialogic speech between multiple works from various composers. Metaphorically speaking, Dialogismos I can be seen as an offspring that share a congenital relation to a collection of parental works. Yet, even if the structural elements of the piece, such as the sound events, the phrases, the sections, and the form derive from works by various composers, the granularity of the samples and the assemblage process ensures an identity to the new work that if far beyond a simple imitation of the “parental” works, or any aesthetic affiliation. The artistic notion of creativity as an activity that embodies in itself a continual dialogue with existing artworks from other artists, or even the use of creative processes that involve direct or implicit allusions, absorptions and/or transformations of creative work is an important historical concern in creative arts, which the music field entirely embraces [1]. The conceptual cornerstones of Dialogismos I are the work of Julia Kristeva, a Bulgarian-French philosopher and literary critic, particularly her concept of ‘intertextuality’ [2] coined in late-1960s, and the work of the music theorist J. Peter Burkholder, whose essays on Charles Ives introduce the seminal concept of ‘musical borrowing’. Dialogismos I builds upon a major idea stated by Burkholder that can be synthesized in one of his ‘borrowing’ techniques, i.e. modeling. According to Burkholder, modeling follows a compositional paradigm in which a work or section is drawn from “an existing piece assuming its structure, incorporating part of its melodic material, imitating its form or procedures, or using it as a model in some other way.” [3]. Our work expands previous approaches towards a formalization of intertextual music strategies through the use of computer software to automate most tasks of the process. This paper is organized as follows: in the incoming section, we discuss previous works that follow compositional strategies similar to Dialogismos I. Section 3 clarifies the theoretical assumptions that support the work, i.e. the idea of ‘dialogic’ processes between different authors and works, and their technical realization. Section 4 examines the compositional methods of writing the instrumental (saxophone) score and enlightens the contributions of each ‘borrowed’ composer to our work. Section 5 details the electronic music realization. In detail, section 5 A introduces earGram, an application for concatenative sound synthesis (CSS) that is at the core of the production and performance of the electronic part. Section 5 B reports the first phase of the work that comprises the generation of raw audio material by algorithmic computer-assisted composition strategies in earGram and the later edition and processing of the resulting media, which was done in a commercial audio workstation. Section 5 C details the real-time audio processing techniques employed in the live performance.