Tracking the creative process in Trevor Wishart’s Imago

Often it may seem that electroacoustic music provides fewer resources and less evidence for musicologists to work on when investigating creative processes. Electroacoustic works are produced in the studio,often composed directly onto disc and there may not be sketchbooks with musical notation of drafts of passages, or diagrams or notes. Often digital sketches may be overwritten or discarded as the creative process moves on. Software and hardware used in a composition may become obsolete over time and its method of operation may not have been recorded in detail. One of the goals of the our 30-month project TaCEM, finishing in March 2015, is to try and trace such compositional processes in electroacoustic works,and to emulate the techniques used by composers and to study how they were deployed in creating particular works. In some respects, the project is related to work analysing the technical and musical aspects of electroacoustic music using software in the form of Interactive Aural Analysis (Clarke 2012). Other related work includes: Battier (2003) on a constructivist approach to analysis, and Baudouin (2007)and Dahan (2007)on the preservation and reconstruction of technology related to creative practice. The overall goal of the project is to examine the relationship between technological innovation and creative potential in electroacoustic music. We are doing this by examining eight case studies, ranging across the repertoire, in which new technological innovation has played a key part in enabling composers to bring to fruition new creative approaches. We are studying the context of each work, investigating the technology employed and analyzing the music.
One of our first completed case studies is Trevor Wishart’s Imago (2002). The way that Wishart worked and the software he used (his own Sound Loom software package) have resulted in an unusually rich resource for tracking the composer’s creative process over time and examining the relationship between the technical resources he used and the final musical outcome. The Sound Loom software works by taking an input sound file and applying a transformational process to this file (e.g.transposition, time stretching or brassage), using parameter data entered either directly into the program or as a data file, so as to create a new output sound file. The program does not operate in real time but generates sounds one at a time with clearly defined processes and parameters. This way of working leaves behind a trace, a set of date-stamped audio files which represent the evolution of the work over time.
The development of sound materials, from an original source sound through a sequence of processes, can be traced potentially, and the gradual assembly of components for the final piece can be tracked over time. Furthermore, the data files used to control the various processes can provide information about the details of the different transformations that were deployed. If, as is the case with Imago, these materials have been to a large extent preserved an extremely rich resource is available for tracing the creative process as it evolved.
How might such a digital resource best be used by musicologists for researching the creative process and for analyzing the musical work? Central to the methodology employed is the use of software both in undertaking our research and in presenting our facilitates engagement with musical works as sound and the investigation of the techniques used through software emulations. So our primary means of researching the archive of sound and data files relating to Imago has been through software. One resource we have created is a calendar of activity. Set out like a traditional calendar this shows which files were created at what time and on which day. This calendar is not simply visual, the calendar being in software, each file is represented by an onscreen button and this can be used to play the sound or open the text file. It is therefore possible to trace in detail the process by which the work evolved day-by-day and hour-by-hour.
Another software resource we have produced shows the relationships between the sounds. A network of inter-related sounds, leading from the original source sound (Imago is built entirely from one single short recorded sound) through sequences of transformations branching out in different directions, to the completed work, is represented on screen. Each node is an on-screen button that can play the sound concerned. A slider control also allows the user to move through the evolution of the work in time, so that nodes on the diagram (i.e. sounds) gradually appear on screen in the order they were created. It is possible to see the temporal evolution of the work and to hear it by playing the files. We can discover whether sections were composed in the order they appear in the work (mostly they weren’t), whether materials for a particular section were all developed in close temporal proximity or not (sometimes they were)and how the overall form of the work took shape.
This leads to a related chart that also shows the relationships between sound files but this time focusing on the genealogy of the sounds –how different sounds are related through similar branches of processing. Again the nodes of the tree can be played. Furthermore, in this case it is possible to learn in detail about how sounds are interrelated. The processes that link sounds can be opened in an associated window on the screen. It is then possible to recreate that transformation using our own emulation of the Sound Loom software. The data used by the composer can be entered to produce an exact or very close replication of what Wishart himself did, or alternative settings can be tried to learn more about the processes and their potential and about the choices the composer was faced with. Imago is a long work, over 25 minutes, and the archive comprises well over 1000 files and employing many different processes. We have not therefore been able to analyse the whole piece at this level of detail but have chosen representative passages for in depth examination.
In summary, we have used software to help us understand more about the creative process involved in the composition of Imago. We have been able to track its temporal emergence moment by moment. We can follow each step in the creative process over time and see what that involved in terms of technical manipulations of the sounds. This technical and analytical work has been carried out alongside discussions and interviews with the composer so that we can cross-reference the composer’s thoughts on the creative process with our own interpretations. The outputs resulting from this research, which include interactive software, text and recordings of interviews with the composer, provides the reader with a multi-dimensional resource for exploring the structure of Imago and for developing an understanding of the creative process that led to its creation. Our presentation will include a demonstration of the software we have developed as well as discussion of Wishart’s creative process in composing Imago.