Shared musical creativity: teaching composer-performer collaboration

This paper focuses on teaching composerperformer collaboration as a path to new paradigms of 21st century art-making. First, we discuss the interplay between theory and practice in creative musical processes, with special attention to the performative modality of composition. Then, we approach collaboration itself, considering how the term has been both overused and misunderstood for decades. Finally, we present a case study of teaching composerperformer collaboration at the graduate level, in the “Atelier for Composition and Performance” (Ateliê de Composição e Performance Contemporânea), at the Federal University of Bahia.


Introduction
When writing about the pedagogy of Music Composition and the inseparability of theory and practice, Paulo Costa Lima (2012) states that composition might be something "unteachable (but learnable)": In the pedagogy of Music Composition, this inseparability gets close to what is usually considered 'unteachable (but learnable)'. It is possible to offer horizons of theoretical and analytical building. It is possible to offer models, poetic environments, to analyse and try cases, desires and opportunities. But at some point, the student-composer will have to make their own decisions about the connections they intend to establish, and that will be built upon the skills they have (LIMA, 2012, p. 24).
On the other hand, Nagy (2017) proposes compositional creativity as an interplay between the plasticity of cognitive modality and the physicality of performative modality, considering the embodiment of musical meaning. We compose and listen to music while creating meaning through our bodies and the music's cultural significance.
When placed in the context of compositional creativity, however, the creative cognition model implies a particular type of exchange between a conceptualization and a contextualization of creative processes, embodying a sense of connection between the generative act of mentally envisioning musical sound, for instance, and the explanatory act of physically producing the sound (NAGY, 2017, p. 17).
Performativity, as it is part of embodied musical creativity, must be considered an essential issue for teaching composition. Cox argues that we construct meaning in music via imitation, according to his mimetic hypothesis: One may notice that the gestures of performers seem somehow relevant to musical meaning [...]. The motor imagery evidence of the mimetic hypothesis suggests that we understand these gestures via mimetic participation and that these gestures are relevant as a normal part of music perception and cognition (COX, 2001, p. 204).
Considering performativity of compositional creativity and the performative role of performers' gestures, collaboration between composer and performer may contribute to an embodied approach to teaching composition. In this sense, we argue performance as a fundamental part of the creative processes, rather than just an interpretation of some ideal masterpiece. Collaboration makes it possible to establish an environment that improves the creative processes through its performativity, taking the interplay of theory and practice as a given.
Previous concepts of teaching composition must also be adjusted to a more complex set of distributed and interconnected artistic relationships, in order to better represent what young artists aspire to today.
Reinaldo Ladagga (2012, p. 9) writes that there are increasing numbers of artists developing works that require complex, intermingled strategies, involving different realms, including a broader community of potential participants. He goes further when affirming that these artists are less interested in making specific works of art, but instead, in participating in the creation of "cultural ecologies" (LADAGGA, 2012, p. 11).
We argue that artistic collaborations between composers and performers in the 21st century -in the context of Post-secondary Music Schools -may be a path to new paradigms of art-making, one that encourages sharing instead of individuality. This is corroborated by Ladagga's thesis of new cultural ecologies sought by 21st century citizens. Indeed, a quick survey of students entering Post-secondary schools -at least in Brazil, where our current research is based -shows that they are raised in an interconnected world, with far fewer barriers of style and musical elitism: as artists they have become much more interested in dynamic and participative creative processes. They arrive with a diverse set of skills and backgrounds, completely different compared to the many years of Eurocentric Conservatory training that students from only a few decades ago had. In other words, 21st century student-composers and student-performers are There is a school of thought, not new to this research, of collaboration as a ʻthird handʼ; the idea that an artwork created by several individuals has the potential to be different and greater than the work that any of those individuals may have achieved on their own. For Myles Mumford, this means "that if we only do what I want to do then we just made what I wanted to make, and I might as well have just been making it by myself." Elaine Miles described the experience of making work with artists from different backgrounds as "producing work you couldn't produce on your own… everyone brings skills and you get a much bigger and broader outcome than you could ever do on your own" (GRANT, 2010, p. 15-16).
Paraphrasing Lima, collaboration might well be unteachable -but learnable. We can offer tools and horizons of understanding for the collaborator-students to develop their own paths into the foundation of unique collaborative processes. This is what we have been pursuing with the "Atelier for Composition and Performance" (Ateliê de Composição e Performance Contemporânea), a Graduate Seminar that we have been offering since March 2018 in the Music Graduate Studies Department at the Federal University of Bahia, Brazil.

Collaboration
When discussing Contemporary Classical Music, the term 'collaboration' has been both overused and misunderstood for decades. There are many ways to develop artistic collaborations, and the term itself has many connotations. Focusing on interactions between composers and performers, it may even embrace contradictory meanings, varying from a quick consultation about the feasibility of specific gestures by a performer -or the performer giving the composer access to his or her "box of tricks" (FITCH; HEYDE, 2007, p. 93) -to a full-on immersion in a collaborative project. Surprisingly, the most common use of the term collaboration in recent Contemporary Classical Music involves very little, or no interaction whatsoever, being closer to a commission of a new work, a process in which the individual roles are kept completely separate. Hayden and Windsor categorize interactions between composers and performers into three groups: DIRECTIVE: here the notation has the traditional function as instructions for the musicians provided by the composer. The traditional hierarchy of composer and performer(s) is maintained, and the composer aims to completely determine the performance through the score. (…) The collaboration in such situations is limited to pragmatic issues in realisation.
INTERACTIVE: here the composer is involved more directly in negotiation with musicians and/or technicians. The process is more interactive, discursive and reflective, with more input from collaborators than in the directive category, but ultimately, the composer is still the author.
COLLABORATIVE: here the development of the music is achieved by a group through a collective decision-making process. There is no singular author or hierarchy of roles. The resulting pieces either (1) have no traditional notation at all, or (2) use notation which does not define the formal macro-structure. In (2), decisions regarding large scale structure are not determined by a single composer (HAYDEN; WINDSOR, 2007, p. 33).
The need to reflect upon what composer-performer collaboration means comes from accepting that composition and performance are activities done by different individuals, and that their roles are kept separate. Historically, this has not been the case. One could actually argue that this is a recent development in Music History. Until the 19th century the most celebrated musicians "were almost always performers and composers, practical musicians as well as creative artists" (JAMASON, 2012, p. 106). However, even in cases where they were not the same people, their interaction was highly collaborative: Prior to the nineteenth century 'fidelity to the score' meant that performers were expected to 'complete' the notation through a variety of means. Performers sought not only to express an individual composer's particular style but were also very much working within a larger framework of shared performance practices within their time period and geographical area (JAMASON, 2012, p. 107).
During the nineteenth century, composition assumed a new, more individualistic approach, valuing the uniqueness of each artist. One could say, in a quite simplistic way, that composers began then to be revered as heroes, and composition as a higher art in itself. 2 From then on composers were placed hierarchically above the -much less heroic -performers. Indeed, despite the highly technical needs required to perform the new music being created, performers have been seen as "executors" of a score, expected to reproduce it in the most transparent way possible, blocking any intention of creating on, or -in the composers' minds, altering -the composer's creation. Therefore, besides being lower in the hierarchy, performers were not valued for their own creativity but for their loyalty to somebody else's. 3 This has been the paradigm of Composition for almost 200 years, having possibly plateaued in the second half of the twentieth century. As put by Hayden and Windsor (2007, p. 29-30), "the roles that composers and performers inhabit have become so strongly demarcated over the last two hundred years that the distribution of working practices and associated hierarchies passes largely without comment". It is in this environment that Composition and Performance exist today at music schools and universities. While co-existing, there seem to be few institutional opportunities to interact, and when opportunities do exist, they are often due to individual efforts.
In this context, performers interested in expanding their repertoires to the music of today seek composers to write for their instruments, and from these often unplanned, rarely sponsored, interactions, arise commissions and premieres of new works. Making use of Hayden and Windsor's categories, these interactions tend to be mainly directive and interactive. Very few are collaborative. Nevertheless, they are almost always referred to, vaguely, as collaborations. Fitch and Heyde reinforce this, when they discuss the common understanding of collaboration in the Contemporary Classical Music world and how it often reinforces the boundaries between roles: Collaboration is frequently a matter of the performer giving the composer access to his 'box of tricks', or of the composer presenting notated sketches to be tried out, adopted, discarded, or refined. Such pragmatic approaches may well be beneficial to both parties, but they come at the cost of reinforcing the boundaries inherent in their respective roles. We have felt it worthwhile to present an account of our work, in so far as it represents a more dynamic model of the collaborative process, in order to articulate some of the ways in which creative practice may be understood as research (FITCH; HEYDE, 2007, p. 93).
A "more dynamic model" of collaboration begins with accepting the flexible nature of those processes, with a mixture of levels of relationship depending on the stage of a collaboration. When composer Luke Styles describes his relationships with performers, he explains that: During the early stages of creating a work an interactive and collaborative relationship with performers (particularly musicians) takes place, where ideas are being suggested and I am making decisions about these ideas in a creative, often experimental environment. There are also occasions when I leave initial decisions about how to develop aspects of the music (such as dynamics, repetitions or tempi) up to the performers and then together with them (and other collaborators) we decide what should be set in the score. A score may therefore include the multiple musical decisions that have been reached as a process of collaboration and composing alone, but not every aspect of a final performance involving multiple non-musical elements. During rehearsals the relationship with performers moves towards a directive and interactive one. There will still be a dialogue with performers about how to interpret and perform the work, as interpretation needs to come from the performer, but as a collaborator who can view the work outside of performing it, I am in a position to see and hear aspects of a performance that may be missed by the performers in the act of performance. It is therefore important to take on a directive relationship with performers at this stage in order to continue to shape the work that has been made collaboratively, ensuring that overall cohesion of the work remains, whilst performers are occupied with the act of performing (STYLES, 2016, p. 10).
Indeed, a more dynamic model of collaboration begins with the understanding that ideas suggested by one individual will be reflected upon by the other artists, and shared (or not), reacted to, molded, developed, challenged, and in this way a path of creative decisions is slowly built, throughout which the artistic vision is shared and enriched among the collaborators.
Social interaction involves two or more people talking or in exchange, cooperation adds the constraint of shared purpose, and working together often provides coordination of effort. But collaboration involves an intricate blending of skills, temperaments, effort and sometimes personalities to realize a shared vision of something new and useful (MORAN; JOHN-STEINER, 2004, p. 11). This is a very important consideration -the sharing of a vision through an intricate blending of skills, It makes sense, therefore, that the most productive collaborations take place between long-time collaborators, often among people with shared personal history. Catarina Domenici points to the fact that composers, when asked about performers who better represent their music, immediately bring up names of those with whom they have personal relationships, the ones they are friends with (DOMENICI, 2013, p. 10). The personal connection that is often generated by, or leads to, a productive collaborative process, contributes to the uniqueness of each collaborative process, and to the consequent writing about these processes.
When researching recent publications on artistic collaboration, it becomes apparent that most academic writing has indeed been personal reports of an autoethnographic nature. Accepting the uniqueness of each creative process and each partnership, such individual reports are expected and welcome.
Thus, we believe that it is time to deepen our research in collaborative processes. This is of great relevance concerning the teaching of creative aspects of music, such as Composition and Performance, for graduate and undergraduate programs. Methodologies to direct these creative processes are lacking. Hence, the collaborations have been happening empirically and intuitively. It is therefore imperative that we find common ground and begin building theoretical horizons for teaching artistic collaboration in the context of University students of Music Composition and Performance.
Without any presumption to suggest that this is the only solution for the wide division between the areas of Composition and Performance, nor the only way to teach (or learn) composition, we propose that teaching music collaboration can be a path for a deeper integration between the areas, and among individuals who are at once deepening their skills and learning to participate in a more democratic creative process, one that values both theory and practice equally in this new creative ecology of the 21st century.

Atelier
The Graduate Seminar "Atelier for Composition and Performance" (Ateliê de Composição e Performance Contemporânea) was added to the curriculum of Master's and Doctoral Students at the Federal University of Bahia as an elective course in 2015. Composer and professor Guilherme Bertissolo was responsible for this implementation. The first class was offered in 2018 when pianist Luciane Cardassi joined the Music faculty as a Visiting Professor at UFBA. A collaboration in itself, this Graduate Seminar has been a shared module between Bertissolo and Cardassi, with the intention of providing an opportunity for students and faculty members to reflect on collaboration through an academic programme of study that values practice and theory equally.

Organization and Rationale
In some respects, the structure of this module follows the regular teaching schedule of other Graduate Seminars: weekly 3-hour meetings. Each meeting has a theoretical component followed by a practical component. Time division between these components is variable.
The theoretical component consists of discussion of articles previously agreed to. The faculty proposes questions to encourage students' critical thinking about collaboration and subjacent areas, as well as sharing of individual experiences and interests. All are expected to participate in the discussion. We also have an online shared group via a smartphone application through which we continue discussion after class (programme and bibliography addressed in this component will be discussed subsequently in this paper).
The second part of the class is the hands-on collaboration component. In this, each duo of composer and performer (or even a trio or quartet) meets separately to develop their own collaboration. Beyond this weekly meeting with faculty guidance, each group is invited to continue their processes at alternative places and times of their choosing. Our intention with the once a week guided collaboration meeting is to provide regular access to support from Bertissolo and Cardassi. The role of the faculty here is to assess, observe, inspire, and encourage the collaboration processes. It is also, and perhaps more importantly, "being sensitive to bringing all voices to the foreground, minimizing differences in experience, attitude and personality that could potentially unbalance the horizontal character of the collaborative processes we are hoping to achieve" (CARDASSI; BERTISSOLO, 2019, p. 5).
We argue that regular supervision of the students' collaboration meetings is essential for positive results. Without guidance by more experienced practitioners and scholars, we believe that the module could lead to reinforcing the division of roles and power between performers and composers. In fact, Doffman and Calvin seem to have observed this in their graduate programme at the Royal College of Music in London: More than once, performers expressed a sense of relative powerlessness in waiting for materials to be passed on to them for rehearsal and private practice, and some cited this as a major collaborative problem during the module. Composers were sometimes unable to get material to performers in sufficient time for them to prepare for the recital, and this had other consequences. (...) One performer expressed reservations about their work being collaborative at all. (DOFFMAN; CALVIN, 2017, p. 11-12).
We strongly believe that if student composers and performers are led to organize their own schedule of workshops, meetings and rehearsals, with a deadline for finished composition and performance at the end of the semester, we would be reproducing in class the paradigm of the Contemporary Music professional world. In that paradigm, performers often receive the score with insufficient notice before the premiere, resulting in the discomfort of both parties. Performers are unhappy with the brief time they have to prepare their work properly, and composers are unhappy with the premiere performance, thus further broadening the gap between them. The challenge is to provide the necessary support to create a new paradigm, one in which collaboration takes place as a joint effort by two individuals encouraged to deepen their level of expertise, without a power game that only leads to frustration.
Another essential point for us is to always prioritize the collaborative process itself, even if it means not having a complete score or piece ready to be performed at the end of the semester. Instead, we intentionally set a date in the last week of classes for a public presentation, when we present to the audience our goals and intentions with this module, and each group plays snippets, or first versions of the pieces being developed. As well, halfway through the semester each group has a short presentation in which they update the class and faculty about their process.
We strongly believe that the fact that there is no formal recital, and the students are not required to submit a score or polished performance at the end of the semester, is our only way to escape the stress and power games that often lead performers to distance themselves from the Contemporary Music world.
Besides the public presentation, each group prepares a final oral and written presentation for the class, in which they reflect upon their own creative process as connected with the issues addressed in the bibliography studied during the semester.
We found that it is important to dedicate at least three initial meetings for participants to get to know each other. We encourage them to present themselves, have conversations about a broad range of issuesmusic and non-music-related -while we begin discussing the first articles on collaboration from the syllabus. Throughout these initial meetings the faculty may identify similarities between individuals and commonalities in interests, but the groups are ultimately formed by the students themselves. Once more the faculty role is to assist and encourage, overseeing group formation, without deciding for them.

Programme and Bibliography
Although relatively flexible, adapting to each group's specific interests, the programme for the Atelier has several stable references, especially with regard to definitions and the specific bibliography on artistic collaboration, documentation and analysis of the processes, autoethnographic studies, art and formativity, as well as topics on authorship among others, with a focus on the artistic processes themselves.

Artistic Collaboration and Distributed Creativity
An important discussion, which takes place early in the semester, has to do with concepts, what collaboration is, and how the bibliography on collaboration can help us better understand it. We focus on distinguishing commissions from collaborations, reading the bibliography in search of clear definition. We adopt the Hayden and Windsor definition (2007, p. 33), already detailed in this paper, which defines collaboration -or a collaborative interaction -as when the participants in the process are open to more flexibility in their individual roles, seeking ways to negotiate the decisions in a shared manner. This takes place without in any way ignoring their individual expertise. On the contrary, we agree with John-Steiner when she writes that the more the collaborators share with each other what they know, the more they deepen their knowledge of their own métier, discovering, thus, the benefits of reciprocity (JOHN-STEINER, 2000, p. 204).
In addition to Vera John-Steiner (2000) and Margaret Barrett (2014) -prominent scholars on collaborative creative practice today in the Northern Hemisphere -, discussion includes the works of Brazilian researchers Catarina Domenici (2010) and Sonia Ray (2010), among others. We have also been reflecting upon the concept of distributed creativity, recently developed by Clarke and Doffman (2017) which expands the concept of creativity, stretching it to all participants of the creative process, including performers and listeners. Clarke and Doffman go beyond, affirming that contemporary music provides a unique environment for the study of creativity in its distributed character.
Framed by conceptual developments in musicology, psychology, anthropology, sociology, computing and neuroscience, there is increasing recognition (long overdue, one might think) of the extended and distributed character of music's creative processes. While such recognition applies in principle to the music of any culture and period, contemporary music offers a particularly fruitful target domain-not least because of its very 'presentness' and the opportunities it therefore offers for detailed and direct investigation (CLARKE; DOFFMAN, 2017, p. 2).
Our Atelier for Composition and Performance aligns with Clarke and Doffman's thesis, providing an ideal environment for investigating collaborative processes and developing possible theories and methodologies to guide us in teaching collaboration at graduate and undergraduate levels.

Auto Ethnographic studies
The sharing of personal stories among students and faculty at our Atelier is part of our activity.
It is not a surprise that most authors when talking about collaboration, make use of a confessional tone. The uniqueness of each partnership directs us onto that path. It is also worth noting that in several of these reports, the authors highlight the changes they have seen in the level of participation of performers of our times. For the contrabass player Fausto Borém (1999, p. 21): "in order to minimize the disparities between what is composed and what is heard, it is imperative that bassists give up the passive role (of just reproducing the music) and go after the composers". As pointed out by Silva (2014, p. 24), this switch in today's performers' behaviour -abandoning a passive role for a proactive one -is supported by scholars such as Ray (2010), who discusses the figure of a "performer-leader" in our century, with artistic and pedagogical goals; and Domenici (2010) who introduces the 21st century performer as someone not only active but with a voice and rights in the partnership.
As stated previously, one of the objectives of our Atelier is to promote deeper reflection upon collaborative processes. We find that personal reports are an effective starting point in order to identify connections, similarities, or, in other words, elements that perhaps could be generalized, even if partially, guiding us towards a methodology of collaboration.

Creative Process: importance of documentation
Since the Atelier focuses on the creative process, instead of the product such as a score, recording, and so on, we propose special attention be paid to documentation throughout the process. This is a substantial difference between our approach and other similar activities (DOFFMAN; CALVIN, 2017;TOKESHI, s.d.).
Concerning this issue, we encourage discussions on creative processes from at least two perspectives: Genetic Criticism and Formativity. The first is an interesting area of study for the creative processes. Salles (2014), in approaching Genetic Criticism, points out the importance of sketches, notes, drafts, letters and other preparatory documents as objects of analysis. Documentation helps keep track of the problem-solving and decision-making processes.
Pareyson argues that the artistic process "performs and produces the works at the same time it invents the way of doing them" (PAREYSON, 2005, p. 23). In this sense, each process is unique, and we must pay special attention to documentation in order to further understand and discuss decision-making and distributed creativity. Chiurazzi (2018, p. 418) points out that "Pareyson's is not, actually, an object oriented aesthetic, since it does not focus on the object, on its qualities, on its properties, does not assume, above all, the object as the determining factor for deciding what is art". Considering this, discussion about collaboration must consider creative process, instead of addressing the works per se. We focus on "how", rather than "what".

Cognition, Gesture and Performativity
Since we consider the importance of performativity in compositional creativity (NAGY, 2017) and the mimetic hypothesis in the construction of meaning (COX, 2001;, performers' gestures become a key-concept for addressing collaborative creative processes, as it is part of the construction of meaning. Gritten and King (2006) propose an interesting overview of gesture, which may help to clarify issues on gesture in creative collaboration, contributing to an embodied approach to teaching composition. Ideas such as musical forces (LARSON, 2012), memory mechanisms (SNYDER, 2000), metaphor (NOGUEIRA, 2011;SPITZER, 2004), image schema (BROWER, 2000;NOGUEIRA, 2009) and expectation (HURON, 2006) also may contribute to the discussion of gesture in collaborative creative processes from a cognitive point of view.

Theory and Practice in Creative Processes
As we have previously discussed, our Atelier considers the inextricable interplay between theory and practice in composition (LIMA, 2012) and creative processes in music, in general (as something unteachable, but learnable). In this sense, we believe it is important to discuss creative processes through theory and practice. We propose addressing issues about creative processes based on the ideas formulated by Laske (1991), Reynolds (2002), Reche (2007), and Amorim (2014).

Final considerations
We believe that collaboration improves the creative processes, as it unfolds interesting pathways for research and creative activity in music composition and performance. We often collaborate, ourselves, with composers, performers and other media artists. Among her many collaborations, Cardassi has been the music director for a large multimedia project entitled Piano Destructions, with video artist Andrea Buettner, and 9 pianists performing live, at the Banff Centre for Arts and Creativity, in 2014. Meanwhile, Bertissolo has developed a number of interarts collaborations in recent years, notably with Lia Sfoggia (Dance and Video-Dance), from which partnership a web of pieces has been developed, some becoming the core of Sfoggia's doctoral thesis (2019).
In the context of our Atelier at UFBA, we have been welcoming students from Dance and other arts disciplines, besides the initial student-composers and student-performers. This interaction has proven fruitful for all involved. Current faculty leading this activity -the authors of this paper -have themselves developed a collaborative project concomitant to the Atelier activity. Converse, for piano, dancer and realtime electronics/video, is a collaboration between Bertissolo, Sfoggia and Cardassi, developed in 2018, during the first two semesters of the Atelier at UFBA. This reinforces our belief in the necessity of establishing a web of creative processes, and learning from one another, applying these processes among artists from different realms, backgrounds and levels of experience.
The collaboration took place in biweekly meetings throughout the first semester of 2018. Initial sessions included brainstorming of ideas among the three collaborators. The starting point for the creative process was the idea of "state of readiness" (SFOGGIA, 2019). This notion was inferred in "Brazilian capoeira", observing the way through which a player must be constantly prepared for immediate response to change within the context and constraints of capoeira. Once the "state of readiness"' was chosen, performer and composer worked towards investing the music with this concept. Meanwhile the dancer (who is also responsible for the photos and video) interacted with the others via improvisatory movement sessions. The result of this work is a performance embracing the unpredictable at its core, within constraints created and practice during rehearsal. The collaborators create on stage a unique performance that dialogues with notions inferred in this Afro-Brazilian cultural staple.
In Converse the performative modality is a key for the creative process. For instance, at the very beginning of the piece, a gesture is written in a way to keep itself open to the logic of the pianist's fingers.
The accents are notated in a separated staff, in order to point notes, according to the "state of readiness" of the musician during the performance.
For future work, we believe music cognition may offer interesting methodologies and contexts for research. On the basis of subjects in music cognition such as memory, expectation, metaphor and image schema, it is possible to propose an approach to the study of musical meaning in a collaborative, creative environment. Clarke and Doffman (2017, p. 2), for instance, propose collaborative creative processes as target-domains, highlighting a metaphorical perspective on collaboration.

ACKNOWLEDGMENT
We would like to acknowledge and thank all the UFBA student-collaborators who have engaged in our Atelier program to date.