It has been well argued that ‘the score’ underdetermines performance in significant ways, leaving room for unique performance outcomes (Cook 2004, Clarke 2012), the products of which underpin the current classical music economy. The attribution of creative authorship in performance is critically important where livelihoods, careers and reputations are at stake as the parade of glossy CD and DVD jackets and concert publicity materials vigorously brandishing performers illustrates. Research investigating the collaborative relationships between composers and performers as well as the creative process involved in improvisation or semi-improvised scores has drawn out the interwoven nature of composition and performance, but as yet little attention has been given to the creative process in large ensemble performance, and in particular that of orchestras.

There are some practical concerns that make studying orchestras difficult, but an additional factor is the presence of an explicit leadership hierarchy that masks a more complex web of interactions. For example, Stacy DeZutter and Keith Sawyer acknowledge the potential for distributed creativity in symphony orchestras but maintain that ‘the creativity of an orchestra performance resides, in large part, in the creativity of the composer and of the conductor’ and that ‘such individuals, and their creative processes, can be successfully studied using individualistic methods’ (DeZutter & Sawyer 2009, pg. 91). In contrast, a recent study of professional orchestral musicians by Melissa Dobson & Helena Gaunt, found that orchestral musicians’ actions in performance depended significantly on adapting to and synchronizing with their colleagues in rapid exchanges within an intricate web of allegiances and hierarchies, of which the conductor only played a small part. Such complex levels of communication and negotiation cannot be reduced to a uni-directional flow of communication, and their participants devoted greater attention to describing the ‘skills required for communicating and interacting with colleagues than to those required for following the directions of the conductor’ (Dobson & Gaunt 2013).

Furthermore, previous studies of leadership in orchestras show a disparity between the reports of orchestral musicians who cite the dictatorship of the conductor as one of the overwhelming components that together contribute to a lack of personal artistic satisfaction; and those reports that indicate that orchestras are quite often dealing with incompetent conductors who are best ignored. Leadership by the conductor is just one component in a complex network of relationships between orchestral musicians, in which explicit authority structures do not always map onto musical dependencies, which may change from moment to moment based on musical context and immediate need.

I have sought to track the creative process in orchestral performance through a series of case studies of individual orchestras rehearsing and performing in specific contexts. The goal of the case studies is to reveal the often hidden processes at play during the accomplishment of orchestral music-making. The approach is broadly ethnographic with an emphasis on detailed description, focusing on short segments of music in order to capture the nuances in player experiences. Unlike traditional ethnographic methods which rely primarily on interviews and observation, I have worked with a team of computer scientists at Cornell University to develop an online computer program that enables the collection of qualitative data form a large group of musicians simultaneously, soliciting phenomenological information from all of the players in a single orchestra about very specific parts of their rehearsal or performance.

MERID (Media Enabled Research Interface and Database) allows the researcher to post video clips of a recent rehearsal or performance and invite all members of the orchestral to leave time-stamped comments about their experience tagged to the video footage. Participants are asked to allow the video to take them back to that part of the rehearsal and comment on how they, the conductor, or their colleagues may have contributed to the creative shaping of the music. Did they see, hear or sense something that made them respond in a particular way? While the participants themselves are unable to see each other’s commentaries, the researcher is able to triangulate perspectives to draw a richer picture of what may be happening as a particular passage unfolds. Although this method of data collection has some limitations, notably the lack of researcher-participant dialog, it is extremely successful in obtaining detailed information from many musicians simultaneously and relatively immediately, and facilitates the teasing apart the complex web of interactions, enabling a more global perspective to be acquired.

The data is exported from MERID and imported into NVivo (qualitative analysis software program) in order to facilitate thematic analysis. As comments coalesce to form traceable exchanges, vignettes are extracted and form part of the descriptive analysis.
Theoretically, my research draws on Mihaly Csikszentmihalyi’s ‘systems model’ (Csikszentmihalyi 1988) of creativity combined with a sociocultural approach to creativity articulated by Vlad-Petre Glaveanu in which ‘creativity is considered social in nature and located in the space “in between” self and others’ (Glaveanu 2011, pg. 480). Such a perspective rests on the premise that individual agency is dialogically constructed, and that language, culture and agency are located ‘in the interstices between people, rather than within individuals themselves’ (Ahearn 2001, pg. 129). From these theoretical stances, the unique profile an orchestral performance is an inherently distributed and collaborative product that obtains creative status through selection in the ‘field’. The question is not whether orchestral performances can be judged to be unique or creative, but rather: How can the authorship of the artistic profile of an orchestral performance be described?

The analysis to-date extends the findings of Dobson & Gaunt by detailing the multiple and changing locations to which the instrumentalists direct their attention in the course of rehearsal and performance. Amongst other things it reveals how much players rely on previous experience and of ‘knowing how it goes’ in order to successfully accomplish group performance from notated scores, suggesting a reliance an aural tradition frequently overlooked in classical music research. My paper will present findings from the case studies, the strengths and limitations of the MERID system, and avenues for future research.