Complicating leadership: choral conducting training through movement theatre practice

Conductors are typically presumed to possess the physical, interpretative control in choral performance. Questioning that presumption, this article explores how student conductors might be encouraged to engage physically with the musical sound – and sounding bodies – of a choir. It argues that singers’ vocal performance directly and fruitfully impacts on a conductor’s gestural leadership. Borrowing techniques from established physical/movement-based performance and theatre, it explores how conductors might act as the embodied nexus of the poietic and esthesic dimensions of interpretation (Nattiez, 1990), thus collaboratively constructing a performance. To frame the discussion, a conceptualisation of the overlap between body and voice is set out. This conceptualisation emerged during the development of vocal-physical performance projects (2015-16) and was subsequently developed into a broader philosophical orientation. Focusing on issues of embodiment and empathy, this orientation is enlisted to re-examine choral conducting training practices. The influence of these explorations on Daniel Galbreath’s choral conducting teaching is outlined. Additional action-research with theatre practitioner and teacher Gavin Thatcher is then detailed to demonstrate further developments and disruptions to Galbreath’s practice. As a result, a conducting training practice emerges from these practical enquiries that exploits performers’ mutual, direct physical contact via sound.


INTRODUCTION
To conduct is to lead, and a conductor's gestures are a form of physical leadership. Yet the conductor is not the sole creative force in ensemble performance. Every ensemble member self-evidently has creative agency: after all, those individuals create the actual musical sound. Each performer has an individual, subjectively experienced body, and a unique reading of the music which they understand and transmit with that body, just as a conductor realizes and communicates their own interpretation through gesture. Physical, embodied creativity, interpretation, and leadership is clearly distributed across the ensemble. This article complicates the statement that "to conduct is to lead". It does so by asking how, in a field where the conductor's agency and control is often the primary focus of pedagogical and scholarly attention, a conductor might be trained to receive and negotiate the sonic, creative material of singers. This article seeks to offer answers to this question in the form of practical exercises for students and teachers of choral conducting.
We have explored this complication with conducting students at the Royal Birmingham Conservatoire (RBC) and the English Choral Experience Conducting Course (ECE) in classes and workshops . Our aim has been to teach gesture in such a way as more closely unites it with the voices it influences and is influenced by, drawing on elements of dance and physical theatre practice to encourage student conductors to embrace the creative soundings of their singers. We initially complicate the concept of leadership by expanding upon a framework, borrowed from Jean-Jacques Nattiez by way of Nicholas Cook (2001), which posits how conducting gesture can both inform and "receive" vocal sound. We also articulate the inherent relationship between leadership and discipline, drawing on circumscribed ideas espoused by Michel Foucault to consider how the "disciplined" role of the body might be reconsidered while training conductors to reconceptualize how they gesturally lead an ensemble. This attitude has informed how Galbreath approached his teaching work at RBC and other workshops; as a starting point for further explorations, we outline several relevant exercises used in the classroom. We then discuss how Thatcher's disruptions and developments of Galbreath's teaching practice offer additional possibilities whereby student conductors might "feel" the musical dually: both externalising their artistic and technical interpretations, and internalising shared, embodied vocal interpretations via singers' sounds.

Complicating leadership
Before reconsidering the practice of training of choral conductors, it is necessary to interrogate the central notion of leadership along two philosophical trajectories. We borrow initially from Jean-Jacques Nattiez's tripartite structure for how meaning is read into music (1990), before deploying Michel Foucault's concepts of discipline and docility to examine the treatment of singers' bodies as a topic within conductor training.
In Nattiez's model (Figure 1), a "trace" -a point at which a work's ontology is provisionally fixed, such as a score or performance -has both "poietic" processes that precede and feed into its creation, and "esthesic" processes through which the trace is understood (1990, pp. 11-17). To understand how this model might apply in conductor training, we must specify the trace with which we are concerned here, and lay out its poietic and esthesic dimensions. The first of these objectives, specifying a trace, is the most difficult. Cook has observed that musical works do not exist as single, replicable objects, but are instead "allographic, instanced equally by scores, performances, or sound recordings" (2001, p. 179). The situation of conducting a piece in rehearsal and performance is self-evidently allographic: the musical work is instanced by all of Cook's examples and, to complicate matters, each example is refracted into further performative traces by the various different people involved in performing. Cook's examples seem to focus on interpretations and performances experienced by audiences and analysts, and bring most readily to mind the audible performance of the choir as the performative trace most worth considering. However, when Cook clarifies that his concern is with the "material traces" of music -any element of a work's allographic existence which is read as being bound up with its poietic and esthesic dimensions (pp. 179, 181) -a further option is opened: performers might "read" the enactment of musical interpretation in each other's performances, even as they themselves perform. Within the enactment of a piece of music (including in rehearsal), we might focus on a conductor's gestural "performance" as having readable meaning: it is also a trace. A conductor's body and movement are therefore at the nexus of poietic and esthesic processes (as represented in Figure 2, below). The conductor's gestural "performance" is informed by their interpretation of the piece, and is read and itself interpreted by the singers (note the left-facing arrow between Singers and Conductor in Figure 2). Nattiez writes of the esthesic dimension of a trace that "'receivers', when confronted by a symbolic form, assign one or many meanings to the form […]" (1990, p. 12). But, as Nattiez clarifies, singers would not "'receive' a 'message's' meaning […] but rather construct meaning, in the course of an active perceptual process" (1990, p. 12). Esthesic processes act upon the trace (Nattiez, 1990, p. 17): recall the leftward-facing arrow in Figure 2, above. Beyond their agency in interpreting a conductor's gesture, however, singers also inevitably have subjective experiences of pieces, which it would be absurd to presume could -or should -be entirely subserved to the interpretative will of the conductor. Singing therefore can be seen to assign meaning to a conductor's gesture while also acting poietically upon that gesture.
The poietic dimension of a trace comprises the "process of creation " (1990, p. 12, emphasis in original) that preceded it; for a conductor, this process obviously includes score preparation and interpretation. We argue that it also includes the sound received by the conductor's body from singers (represented in Figure 3, below). This contention rests upon our extension of Konstantinos Thomaidis' idea of "physiovocality", used to describe the experience and training of voice and body as unified (2014, p. 243). Conceptualised along physiovocal lines, the voice is not separate from the body. Instead, following on the work of philosopher Adriana Cavarero, Thomaidis posits a post-dualist voice: rather than placing the corporeal workings of the body at the service of the mind to transmit language, using the voice might express the embodied self (Thomaidis 2015, p. 10). By extension, the voice therefore is a physical transmission of embodied creativity.
Crucially, physiovocality in an ensemble is realized relationally: a "performer's body", now regarded as incorporating the voice, "exists only in coexistence" among performers (2014, p. 251). Vocal sound connects, and impacts upon, the bodies of all participants in the choral situation, including the conductor. This proposition suggests that vocally-realized musical ideas, and therefore interpretative initiative, can flow physically in multiple directions, including from the singers towards the conductor. The voice is not only a part of the body; the phenomenon of vocal sound also creates a physical connection between performers (Thatcher & Galbreath, 2019, p. 354). In this relational vein, Arnie Cox hypothesises that a listener kinaesthetically responds to their perceptions of the efforts behind musical sounds (2016, p. 143). Cox writes that as the listener infers that those efforts "must be those of some other agent 'in' the music", what had been "one's own desires and satisfactions are attributed to the musical agent" (2016, p. 143). In essence, as we experience sound, we (in an empathetic, embodied way) infer the interpretation of the creator of that sound. Among a conductor's many duties is, of course, to be a keen listener; and in listening, they experience the embodied interpretations of singers. Cox's "mimetic hypothesis" (2016, p. 45-53) offers a powerful argument that a conductor not only influences sound, but also physically receives it. We would expand upon his argument, however: in addition to being interpreted by the body in a physical way, we contend that sound is received by the body as a physical immanence which creates real "physical contact between bodies" (Thatcher & Galbreath, 2019, p. 353); there is, in effect, the presence of the body in the voice that it produces.
The corollary of this notion of relational physiovocality is that conductors' bodies receive sound even as they inspire it. Much choral conducting literature disregards this dynamic. Didactic works, like Abraham Kaplan's well-known manual Choral Conducting (1985), parse the gestural craft of conducting into components to be mastered, never venturing into murkier questions of somatic responsiveness within an ensemble. Colin Durrant adopts a different approach in his more recent Choral Conducting: Philosophy and Practice (2003). He is orientated more towards the embodied experience of choral conductors leading singers. Yet influence on the performed musical interpretation flows exclusively from the conductor to the singers in Durrant's model of conducting philosophy and, notably, teaching conducting (2003, pp. 166-178). In one instance, he recalls an experience of singers influencing his gesture, but only by making a verbal request (2003, p. 151). Even a more explicitly physical study of choral conducting by Jeremy Manternach treats musical leadership as flowing in one direction, from conductor to singers (2011). In a recent approach bearing comparison to ours, Caron Daley applies Dalcroze techniques to choral conducting: unsurprisingly, these techniques are offered as a means of developing the traditional conductorial role as an embodiment of authority and information (2013).
Liz Garnett's Choral Conducting and the Construction of Meaning: Gesture, Voice, Identity (2009) provides a more probing philosophical and practical examination of the physicality of conducting, investigating how gesture impacts on the formation of an ensemble's vocal identity, while maintaining a focus on how the conductor transmits musical ideas to singers. Garnett delves into how conductors and singers create and share a space in which all might participate in musical creation, using a model based on David McNeill's study of conversational cooperation (2009, pp. 159-164). The sharing of ideas occurs through mimicry, in which one participant comes to share a thought initiated by another (2009, p. 162), or through appropriation, where a thought, once begun by an individual, is taken over by another (2009, p. 162). In reframing these principles as musical events, she maintains the conductor's position as the originator of shared thoughts, yet opens the situation to greater mutuality: The choir participates in the conductor's growth point [the moment when ideas emerge both intellectually and gesturally] and completes their gestural image by giving voice to the structural-musical half of the thought. The conductor, meanwhile, responds to the choir; the shape of their gestures emerges in the context of the choir's sound emerging over time. (2003, p. 163) Within Garnett's conceptualisation of shared influence between conductor and singers, in which even the shape of gesture "emerges in the context of" vocal sound rather than in response to it, the "structural-musical" half of the thought remains the domain of the conductor. This notion raises two questions: what is the nature of the singers' "half of the thought", and do those singers not also offer their own "structural-musical" interpretative initiative? We argue that an answer to the first question must allow for a view of singers not only completing conductors' thoughts, but also initiating them; this response addresses the second question.
Central to our reconsideration of the conductor's embodied experience of vocal sound is a reconsideration of how that body is trained. Our view of bodily training is influenced by several of Michel Foucault's central proposals about discipline, surveillance and docility. Foucault traces the genealogy of disciplinary mechanisms, arguing that modes of disciplinary power and surveillance produce docile bodies (1977, p. 138), bodies that are trained to act and behave in certain ways. If the expectation of conductors is for them to lead, and that sense of leadership is unidirectional, then the examination of discipline in conductor training has two purposes: firstly, by disrupting existing disciplinary structures of conductor training, there is an opportunity to train conductors to respond dynamically and creatively with singers, rather than being taught to impose an interpretation through "correct" gestural technique; secondly, the disruption of these disciplinary structures may lead to a conductor's individual reconceptualization of their own leadership and disciplinary acts when working with singers.
Notions from dance and physical theatre help us in pursuing a revised training of conductor's bodies by offering modalities of physical training which encourage creative, spontaneous responsiveness in gesture and which complicate gestural leadership to allow for the poietic input of singers' vocal sound. Dick McCaw offers a heuristic starting point to bodily training, treating gesture as being both receptive and in keeping with a monistic view of mind and body. In his recent book Training the Actor's Body: A Guide, McCaw provides a view of sensory preparation for a performer as a "bringing into consciousness of one's physical self [...] sharpening and opening our senses as well as preparing our motor apparatus (as if they could be separated)" (2018, p. 102). McCaw further argues for embodied subjectivity: "each of us has a unique structure with its own history of use and development -my [own] body" (2018, p. 101), uniting corporeal exploration and sensory perception to free gesture to creatively respond to creative input. Similarly, Israeli choreographer Ohad Naharin's "Gaga" technique suggests practical approaches to inviting greater receptivity in conducting students' physical/acoustic perception: writer Deborah Galili observes that Gaga participants are "[e]ncouraged to listen to their bodies throughout" (2015, p. 368) and hierarchical orthodoxies of the teacher-student relationship are rejected as the teacher participates in movement exploration with performers (Inception VR, 2017). The application of Gaga principles in the conducting classroom rehearses the creative negotiation these conductors will eventually undertake with their choirs.

Starting points for (re)training choral leadership
These philosophical orientations, and collaborative performance projects directed by Thatcher and Galbreath (Thatcher & Galbreath, 2017), moved Galbreath to experiment in his choral conducting classes at RBC, expanding upon his own training received in the USA and the UK. We focus here on three exercises which have been most consistently used by Galbreath to orientate conductors towards an integration of gesture and voice from the outset of a class or workshop, and which most explicitly draw conductors' gestures towards responding to voices. These exercises are outlined below, accompanied by a brief explanation of how they might begin to promote a view among students of gesture as being informed by the corporeal and acoustic realities of the voice.
Exercise 1: Conductors are invited to feel the quality of motion in their abdomens that results first from breathing, then from sustaining voiced and unvoiced fricative sounds. First while lying down, then while seated or crouched and bent over slightly (Image 1), they are asked to place their hands on their abdomens and on the latissimus dorsi muscles of the lower back. Image 1. Conducting student crouching while feeling lower back movement during breath This contact enables them to feel the quality of muscular motion occurring while breathing and sustaining sounds. This exercise aims to begin to train their hands and arms to show a sustained use of the breath during legato singing, and an easy inhalation that involves the released expansion of the muscles of the lower abdomen. By encouraging them to understand these gestures via the corporeal workings of their own voice, conductors' gestures move towards greater integration into the vocal workings of their singers.
Exercise 2: Gestures layer onto a basic vocal warm-up. Sliding up and down the span of a perfect fifth in different areas of their vocal ranges, on a sustained voiced sound, conductors are asked to accompany their singing with three different gestures: an upward "flip"; a horizontal, sustained motion; and a "grounding" lowering of the body. Whereas in Exercise 1, vocalising drives gesture, Exercise 2 allows gesture to reciprocally impact on vocalisation. Each gesture is used, as it might be in a singing lesson, to encourage a different sensation of vocal motion.
Exercise 3: To clarify to conducting students how gesture can impact and unify vowel sounds across multiple singers, and to highlight the importance of vowel unity to tuning and timbral consistency, the class once again combines gesture and singing. Galbreath asks the class to sustain each of the five cardinal vowels (usually to an open fifth interval, to make issues of timbre and tuning more apparent); while singing the vowels, they are asked to move their hands in a prescribed way for each phoneme. For instance, the vowel [i] is accompanied by an upward "blade" gesture with the hand in front of the face.
Galbreath selected these gestures to reflect how these vowels are produced orallyheight within the mouth, or roundness of the lips, for instance.
The evolution of these exercises over several years of use prompted us to undertake more clearly-planned research into conductor training along these lines. Galbreath invited Thatcher to engage in a practice-research sequence in which the latter assumed the roles of conducting student, class observer, and workshop co-facilitator. Thatcher was therefore able to disrupt and develop Galbreath's conducting teaching practice from the "inside", extending Amanda Bayley's musicologist-performer unity to further incorporate "teacher" (2011,392). The research involved classes at RBC (detailed below). Smaller, exploratory sessions involved two singers from Birmingham, referred to here only by their initials: CJ, a freelance singer who completed postgraduate work at RBC, participated in sessions early in the process; GB, a finalyear undergraduate student in the RBC vocal and operatic department, contributed later in the process. The research process unfolded as outlined below. These phases of research will be referred to throughout the remainder of this article by the five numbered stages presented here.
Stage 1. (Autumn 2017) Galbreath gave Thatcher a one-to-one lesson, using the exercises outlined above, along with several others used frequently in class. These included a discussion and exploration of the moveable anatomy of the shoulder joint and upper torso, several range-of-motion exercises exploring how freely arms might move from the shoulder/torso, and movements highlighting the mobility of individual joints. The lesson also included an introduction to giving an upbeat that also cues a healthy breath, and providing a basic legato four-beat pattern. Thatcher, who has some musical background, had never conducted before this lesson.
Stage 2. (Autumn 2017 and spring 2018) CJ was invited to join Thatcher for two more lessons, which expanded Stage 1 with further explorations. These developments consisted of the contact work described below, as well as attempts to create a greater flow between Exercises 1 and 2 as outlined above.
Stage 3. (Spring 2018) Thatcher observed Galbreath's classes for first-year undergraduates, observing group dynamics and teaching methods. These classes have run for several years, and formed the basis for the lesson Thatcher received in Stage 1 above. He was therefore able to witness the process with some critical distance while still understanding its application.
Stage 4. (Spring 2018) Thatcher led a session of "Conducting Voices", an elective for postgraduates and undergraduates at RBC. The elective covers conducting in a broad range of contexts where voices are involved. Thatcher's session focused specifically on the new physical-theatre-derived ideas that we explore further in below in the section "Conductor training: new approaches" Stage 5. (Spring 2018) Thatcher, Galbreath, and colleague GB workshopped several ideas from the above steps in a more focused way, also photographing exercises for an upcoming presentation.
Group reflection followed each phase; these impressions inevitably informed our own accounts and reactions but a large-scale inductive enquiry into participant experience lies beyond the scope of this article. Discoveries made at each stage of the research process iteratively influenced subsequent events (Smith & Dean, 2009, p. 5), allowing for a reasonably focused refinement of the exercises discussed in the next section. This approach sought to engage in what Robin Nelson terms "intelligent practice or material thinking" (2013, p. 5), using our own, and conducting students', embodied knowledge to understand musical undertakings (Doğantan-Dack, 2011, p. 244).
Our reflections adopted a narrative form espoused by dance theorist Maxine Sheets-Johnstone (2009, pp. 30-31). There is an autoethnographic dimension to this research; we embrace the philosophy of acknowledging our complicity and positionality (Adams, Ellis, & Jones, 2014, p. 22), and, especially, the value of using our "personal experience to describe and critique […] practices and experiences" (Adams, Ellis, & Jones, 2014, p. 1). The conductor experiences their ensemble, and their ensemble's acoustic "product", subjectively, as would an autoethnographer (Adams, Ellis, & Jones, 2014, p. 22), creating an important parallel between methodology and the subject of study.

Conductor training: new approaches
Exercises 1-3 were developed throughout this research sequence in various ways, and with varying levels of success. While our discussion here describes several experiments and activities, indicating an order of progression towards receptiveness, they do not necessarily reflect the order in which they were undertaken or devised. We anticipate that conducting teachers might draw on these approaches selectively, and in any order, in future implementations. We begin by discussing how a shift in room-arrangement in class teaching might broaden students' awareness of themselves and others (i.); we then narrow our focus to expanding on how students use their bodies within their own personal space (ii.). Following these discussions, we outline how students' use of their bodies might become more relational, first through hands-on physical contact (iii.), and then through a vowel exercise that they, and we, experienced as entailing direct physical contact with sound (iv.).

i. Spatial arrangements in class teaching
During the first conducting class of Stage 3, students and teacher naturally arranged themselves into the shape of a typical classroom, with the instructor front and centre, faced by a semi-circle of students. This spatial dynamic establishes and maintains the hierarchical structure of leader and follower, and is paralleled in Western traditions of choral and orchestral configurations. 1 The power relations evident in both educational and professional contexts "have an immediate hold upon [the body]" (Foucault, 1997, p. 25), with acts of discipline and surveillance "[producing] subjected and practiced bodies, 'docile' bodies" (Foucault, 1997, p. 138). Foucault's concept of docility goes some way to explaining the seemingly spontaneous positioning witnessed in the first conducting class, by exposing and articulating the habits that students have developed during their previous education and experiences in musical ensembles.
However, unsurprisingly, these power structures are not exclusive to choral and orchestral traditions. They are also present in theatrical and dance practices. In some Western dance practices, the body is not only rendered docile through leader/follower spatial arrangements, with choreographer/teacher at the front of the space, but also through the use of mirrors which impose a type of self-surveillance during rehearsals. This forward-facing bias redirects performers' attention away from the feeling of movement, or kinesthesia, to the two-dimensional appearance of movement. Naharin's Gaga training challenges this twodimensionality by encouraging the performer to discover the position and form of their body in space without the use of mirrors (Dance Consortium, 2012). Naharin avoids leading from the front, instead moving through the space with dancers, while offering suggestions to provoke further bodily exploration (Inception VR, 2017). By positioning himself with the dancers in the space, Naharin attempts to resolve the hierarchical dynamic of the studentteacher relationship, 2 which acknowledges and utilizes the embodied knowledge(s) of the participants in their explorations in class. Through the development of his Gaga technique, which is described as "a process of growth" (Katan, 2016, p. 26), Naharin encourages threedimensional physical exploration through a series of instructions (input) which respond to the emerging physical dynamic (output) of the performers, and therefore is different from class to class. In order to train choral conductors to be more receptive to singers' vocal input, we predicted that the inherent discipline "that coerces by means of observation" (Foucault, 1977 p. 170-171) could be destabilised by dismantling the hierarchical orthodoxies of space and redirecting students' attention to the three-dimensionality of their moving bodies in the space.
In leading the "Conducting Voices" elective (Stage 4), Thatcher involved the class in a disruption of these hierarchies. Rather than offering students a full Gaga technique-based class, Thatcher worked with two Gaga principles and explored how they could be applied to conductor training. These principles were: (a) that instructors participate in the class among and with participants, and (b) that participants remain moving for the duration of the session. During the class, Thatcher did not lead from the front; he was positioned among the students, performing activities with them, guiding them with a series of instructions. While it was not possible at this point to maintain rigorously the principle of constant movement, there was a clear departure from the classroom orthodoxy of standing and "receiving" information or technique. An active discovery of the body through theatrical ensemble performance-training practice resulted in an observable difference in the level of physical engagement in comparison to previous classes. Students demonstrated an increased awareness of their own bodies, and the bodies of others, establishing the idea of the conductor being "in relation" to multiple bodies in space. By extension, conductors might also learn to operate "in relation" to the physical voices of their ensembles, pointing towards how their awareness while conducting might encompass a broader range of poietic influences.

ii. General physical awareness
As discussed, one aspect of Galbreath's conducting teaching involved exploring conductors' ranges of motion, focusing primarily on the mobility of their arms and the flexibility of individual joints. An ongoing obstacle Galbreath observed in many of his students has been a discomfort with arm movement outside of a relatively limited space; arguably, this reluctance also suggests a limited awareness of the students' own movement capacity. Thatcher has likewise observed that some first-year theatre students exhibit these inhibitions, which may reflect cultural perceptions and social factors acting upon the body. Foucault's notion of "docile bodies" posits that social power structures exercise an "immediate hold upon" the body; they "mark it, train it, […] force it to carry out tasks [and] to emit signs" (Foucault, 1977, p. 25). A conducting lesson or, especially, class can be seen to reinforce this docility among students, yet our (likely uncontroversial) contention is that we should alleviate the social burden placed on students' bodies so that they might become more gesturally expressive. To this end, we worked to expand upon Galbreath's previous range-of-motion exercises in several ways.
This pursuit exposes a complication, however. Beyond broadening conductors' capacity to express themselves gesturally, we maintain the importance of developing a physical receptiveness to sound. That receptiveness runs the risk of simply transferring Foucault's external authority from general social expectations and norms to the choir itself. However, this conflation misconstrues the nature of the leadership that our proposition affords to singers. Receptiveness to sound, as an influence on a conductor's gesture, entails an opening up of a conductor to mutuality. By providing a conductor with more material to inform how they guide an interpretation, certain entrenched social strictures of the choral situation as a whole might be lifted.
We therefore sought to deal with the implications of docility, and to encourage students to become more aware of their bodies -to "misbehave" and explore their range of movement more imaginatively and with greater openness. In Stages 4 and 5, Thatcher introduced Rudolf Laban's idea of the "kinesphere": the spherical space that surrounds the body, and which that body can dynamically occupy and move within (Laban, 2011, p. 184). In Stage 5, Thatcher guided GB to discover the boundaries of GB's reach in all directions (Image 2), initially resulting in a somewhat mechanistic and methodical physical investigation. With further encouragement to find transitions between positions within the kinesphere which were more "interesting" or "exciting", GB began to demonstrate different qualities of movement.

Image 2. Explorations of the kinesphere
Thatcher expressed surprise at the amount of space one can potentially occupy. Following this process, Galbreath and Thatcher asked GB to conduct a familiar, simple tune; while acting as her "choir", we observed GB's more grounded stance and sense of alignment and an increased variety in GB's expressive gestures. Reflection between the researchers and GB revealed that our perceptions were matched by her impression of how she had conducted.

iii. Relation and contact
As students became increasingly comfortable with discovering their range of threedimensional movement in space, we introduced relational training and contact work. In his conducting classes (observed in Stage 3), Galbreath used an exercise to improve students' physical articulation and form in the four-beat pattern. Galbreath asked participants to hold their own hand, then arm, at points just above each joint, conducting a legato four-beatpattern with each joint in isolation. This was one of the first instances of touch being used in the training process, attempting to direct students' attention to the mechanisms of each joint in conducting.
Having had some brief informal encounters with Moshe Feldenkrais's work, and further motivated by the core and relational training for actors encountered in workshops with Alison Hodge at Royal Central School of Speech and Drama in 2015, Thatcher sought to outsource the role of holding the hand or arm above the joint to a partner, in an attempt to reduce both the effort and tension that sometimes emerged in Galbreath's original exercise and the excess sensory stimulation of both holding and moving one's own hand or arm. In Stages 4 and 5, working in pairs, one partner took responsibility for supporting the weight of the conductor's arm, allowing the conductor to direct their attention exclusively to their own motion. This activity aimed to provide space for the student to focus on economy of motion and become aware of their body's physical structure through movement. Students perceived changes in how their arms felt, reporting a sensation of lightness and ease, and a feeling that their arms were longer. Galbreath observed that the legato movements had a greater sense of extension and flow. This contact exercise also introduced the idea of a conducting student's gesture being sensitively responsive to the input of other performers, an idea that we developed further in our encounters, moving beyond touch as a point of contact, and exploring the relationship between gesture, vowels and sound.

iv. Vowel and gesture
In his teaching, Galbreath has frequently borrowed a suggestion recited by innumerable conductors and conducting teachers, to feel contact between the sound and their hands. 3 We argue that this admonition should be taken literally, not metaphorically: the hands experience a physical sensation of the sounds singers create, in addition to being in empathy with the physicality of the singers being conducted. This axiom had been explored largely through Exercise 3, detailed above. During Stage 2, Thatcher, Galbreath, and CJ rotated through conducting, and singing in, this exercise. The activity struck Thatcher as worth further attention: while the idea that gesture can impact on voice is not unprecedented (see, for example, Durrant's discussion of how gesture can impact on vocal technique, cited above), this exercise presented a good opportunity to draw on physical theatre practice to more deeply interrogate the capacity of voice to affect gesture.
Despite feeling a palpable, corporeal change as a singer during Exercise 3, Thatcher questioned whether the tangible feeling of this timbre change was also experienced by the conductor. The exercise had relied on a set gestural vocabulary, from which Thatcher now proposed we deviate. Disregarding established hand motions, the conductor was encouraged to be more gesturally creative, experimenting with shapes and movement types while the singers sustained a vowel sound. We also removed the element of shared gesturing, as used in Exercise 3; here, only the conductor "shaped" the singers' vowels. Upon assuming the role of conductor, Thatcher noted greater reciprocity than expected, and felt that the connection between sound and gesture did not flow exclusively from conductor to singer -rather, there was an experience of sound's tangibility through movement. This exercise serves as the culmination of those described before it: it most obviously opens the door to allow sound to impact directly upon gesture. The results of this exploration are difficult to pin downexpressed as they are here as initial impressions and reflections -but also powerfully direct this practice-research most clearly towards further avenues of enquiry. There is ample scope to develop further explorations of the relationship between vocal timbre, vocal physiology, and external physicality.

CONCLUSION
This article has shown how a complication of ideas of musical leadership can point to new means of training choral conductors. New or altered approaches to gestural training can better prepare students for the shared musical imperative of choral rehearsals, providing an alternative to older notions of authority by embracing the poietic influence of singers' sound. Destabilising spatial hierarchies and promoting continuous movement can open up space for mutual engagement among conducting students. Students' learning can become more active and physical, and might benefit from the complex web of influences which emerges through a freer sharing of space. Students' awareness can be opened further still, encompassing influences that are different from socially-asserted limitations. Encouraging motion throughout the kinesphere can create a more receptive physicality, and holds potential to powerfully broaden gestural freedom. We focused more closely on how that gesture might be seen as relational by making relationality obvious: contact exercises offer an awareness of range of motion and, consequently, gestural variety, which facilitated responsiveness to external physical realities. That variety and relational sensibility can be used to engage directly with sound, through exploration of the relationships between vowel and gesture.
These exercises only begin to point towards how conductor training might incite awareness of the many directions in which interpretative initiative flows in choral performance, via relational physiovocality. They do not amount to a pedagogical system as such, and we do not propose to devise a codified pedagogy now or in the future. Instead, we seek to offer not only a new perspective on the choral conductor's role, but also a glimpse of how we have effected that shift through practice-research. This enquiry has afforded us a further understanding of how the physical act of conducting might entail the negotiation of a shared, emergent interpretation that is simultaneously created by all participants. Various interpretative material can be seen as being received by the conductor. This poietic, vocal material is mediated and consolidated by that conductor, who aims to produce a coherent esthesic interpretation, manifested through gesture and the audible product presented to audiences by singers. Amid this circuitous web of influences and outcomes, the physicality of sound is key.
This enquiry is presented as a starting point. It would benefit from further collaborations with singers and conductors, along with teachers of both practices. In particular, as a result of our findings to date we argue that there is need for a more systematic interrogation of (i) conducting students' embodied experience of the physicality of sound, (ii) how that experience impacts on their gestural leadership, and (iii) whether that impact can be increased. Additionally, we suggest it is worth investigating how similar issues might apply in instrumental or mixed settings.
Any enquiry such as this is necessarily underscored by a complication, even destabilisation, of conductorial authority. Yet this does not negate the leadership of the conductor. In our view, this undertaking in fact offers a more realistic view of conductorial leadership, involving the careful management of numerous influences which act on the shared pursuit of a common goal. During the process detailed and evaluated in this article, conductors gain information from their singers. This article has expanded upon how students of conducting can take their learning beyond our lessons together, into a profession where they can constantly learn from their singers.