In the groove and in the moment: epistemology and ethics in ethnography with Sudanese musician revolutionaries

This study discusses the methodological component of a research project with Sudanese musicians and their associated activist groups. The methodology included song-writing and, as such, is an example of collaborative creative research practice. Proponents of collaborative creative practice argue that the combination of aesthetic methods with ethnographic and participatory research methods brings both epistemological and ethical dividends. This paper considers whether these alleged epistemological and ethical advantages bore out in this research project. While confirming some benefits, my study also shows evidence of underlying tensions between aesthetic ‘micro-methods’ and ethnographic and participatory traditions of knowledge production. In relation to the alleged epistemological dividends, I argue that autoethnographic embedding in collaborative creative practice is alone insufficient. It requires a theoretical framework which theorises the relationship between one player’s musical experience and another’s. Only with this, can the sensory experiences of the researcher be used to inform analysis of participant observations and interviews. The autoethnographic experiences of the researcher are not findings in themselves. In relation to the ethical dividends, unlike other arts-based research, I found that the aesthetic micro-methods in this study did not naturally lend themselves to participant empowerment. The pursuit of aesthetic goals has its own division of labour which can lead to the deprioritisation of self-expression and co-learning which constitute the primary aims of classical participatory research. Overall, collaborative creative practice did enhance this research project but there are important caveats. To reflect these, I aruge that creative collaboration should not be considered as a simple sub-set of either ethnographic or participatory research but as a method in its own right.


Introduction
Inspired by feminist and postcolonial critiques of knowledge production, many sub-fields within the social and political sciences have recently heralded an 'aesthetic turn' (Bleiker, 2017). Artistic products and practices are being incorporated into socio-political investigations in the form of data corpuses, analytical frameworks and dissemination tools (Bauer and Gaskell, 2000;Carless and Douglas, 2011;Kara et al., 2015;Knowles and Cole, 2007;Leavy, 2020). As such, methodological conventions from disciplines previously alien to one another are now being brought into productive conversations. Within this trans-disciplinarity spirit, artistic collaboration with participants at the point of data collection is still the exception (Möller, 2016: 7). In the context of socio-political research, collaborative creative practice is a research methodology which incorporates aspects of autoethnographic and participatory knowledge production in combination with methods of aesthetic production such as song-writing or film-making (Grabska, 2020). This article contributes to the methodological debates about the potential and pitfalls of this method.
Several epistemological and ethical dividends are allegedly brought about by collaborative creative practice. According to proponents, they offer access to experiential and emotional knowledges which conventional approaches cannot supply (Degarrod, 2013;Loaiza, 2016;Mullett, 2008;Vacchelli, 2018). Others argue that they level power imbalances between the researcher and research populations and can even actively empower participants (Aure et al., 2020;Ugolotti, 2020;Vacchelli, 2018;Furman et al., 2019;Mullett, 2008). Critics, on the other hand, note that collaborative creative methods have a reputation for producing 'unethical, sloppy and self-indulgent' research (Kara, 2015: 33). Certainly, collaborative creative practice as a socio-political research method is still in its infancy and lacks methodological reflection (Coemans and Hannes, 2017: 41;van der Vaart et al., 2018: 1). It is important for the growth of this methodological approach in this field that the researchers using it can defend it to sceptics. This involves establishing realistic standards of what collaborative creative practice can and cannot bring to sociopolitical enquiries.
This paper contributes to these conversations by reflecting on a project which involved collaborations with Sudanese refugee musicians in the Netherlands as they sought to support an uprising in their homeland. In mid-December 2018, the towns Damazine and Atbara erupted into mass protest. Only a few months later, the entire country was mobilised. An occupation of the military headquarters in the capital Khartoum eventually unseated President Omar Al Bashir from his thirty-year rule. Throughout this time, a resplendent wall of anti-regime murals wound for miles through Khartoum. A powerful sound system at the sit-in pumped live music through the city, day and night. The uprising has been dubbed a 'creative revolution' in acknowledgement of the important role played by artists of all kinds (Hashim, 2019;Satti, 2019). During this transformative moment, Sudanese communities around the world found ways to support their homeland as it underwent drastic change (Etienne and Franck, 2019). Just as the revolution in Sudan itself was notable for its artworks, activists in exile also embraced creative performances and communications.
The purpose of my research was to understand the effects of these artistic expressions, specifically music, on the overall trajectory of transnational political activism. My research is not a piece of ethnomusicology but is situated in specific socio-political debates in migration studies which relate to transnational political participation and political community formation. Music has been largely absent from such conversations as argued by Martiniello (2019), Martiniello and La Fleur (2008) and Ni Mhuruchú (2016) and thus, the research was motivated by the need to bring music into these debates. While conventional qualitative methods such as interviews and participant observation could have been used alone to address this research question, I chose to also deploy collaborative creative practice. The substantive findings will appear elsewhere in a forthcoming paper.while the present article presents the methodological findings of this research. Here, I consider whether the epistemological and ethical claims made by advocates of collaborative creative practice bore out in this project. This paper is structured as follows: Firstly, I explain the context of the research and the necessary methodological processes and materials needed for replication. The next sections consider each alleged epistemological and ethical dividend in turn starting with two epistemic claims related to sensory and emotional knowledges and continuing to consider three ethical claims related to power, self-expression and co-learning.

Methods: Musicking with Twin Niles
This study borrowed methods from performance-based ethnomusicology (Baily, 2008;Feld, 1984;Fuhr, 2011;Small, 1998;Turino, 2008) and sound-based participatory research (Aure et al., 2020;Cambria, 2018;Ugolotti, 2020). Research materials included music-making hardware; a guitar, two microphones and stands, a mixing desk, monitor speakers and headphones. As well as a digital audio workstation (Abelton Live) with additional sound design plug-ins. The research took place over 2 years and included active participation in over 30 rehearsals, recording sessions and performances events in the Netherlands, United Kingdom and Khartoum.
As the main part of the fieldwork, I joined the Netherlands-based Sudanese band Twin Niles and its associated activist groups Sudan United and Above the Noise 1 . The study is thereby situated in the 'collective social world' (Delamont and Atkinson, 2018: 2) of refugee musician activists during the time of the Sudanese revolution. The lead singer, percussionist and manager is Mosib Aziz 2 , claiming asylum from Sudan and living in the Netherlands since January 2018. He manages the band very successfully; they are regularly hired to play at events, festivals and functions. They also perform at Sudanese demonstrations and as part of site-specific performances and installations.
The band has a fluid membership with many Dutch, international and Sudanese players. The Sudanese contingent includes one of the sons of Sharhabil Ahmed who was a pioneering musician in Sudan in the 1950s and 60s and globally notorious as the King of Sudanese Jazz. The collective also includes contributions from Mo Ali and Mohammed Al Tayeb who were both prominent figures in the Khartoum music scene before they claimed asylum in the Netherlands. Twin Niles is self-described as 'a collective of musicians who play Sudanese and reggae music'. The band mixes several traditional Sudanese styles with globally popular genres such as reggae, samba, rock and pop. Mosib is self-trained to perform in several of these musical styles, and there is a strong emphasis on live improvisation in the band's way of working.
Like others (Baily, 2008;Fuhr, 2011;Reily and Brucher, 2018), I followed Small's (1998: 9) definition of 'musicking' as collaborative creative practice. Musicking is 'to take part, in any capacity, in a musical performance, whether by performing, by listening, by rehearsing or practicing, by providing material for performance (what is called composing) or by dancing'. Within Twin Niles, my activities involved playing guitar and singing in live settings, learning songs, writing songs, recording and producing songs, listening to and sharing music. Like Mosib, I am not classically trained, I have self-taught myself guitar, vocals and electronics in the Western popular music style. Working in a post-folk band and an electronic duo, I reached semi-professional status as a singersongwriter and have experience of playing on large festival stages, releasing records and appearing on national radio. Sudanese styles and even reggae, samba and chart pop were unfamiliar to me as a performer and writer, just as my skills in electronic production were not part of Twin Niles's normal way of working. As will be elaborated on below, our collaborations resulted in some interesting combinations and also some rather awkward fusions.
An important aspect of this collaborative creative research was the co-writing of songs. Song-writing can be defined as the creation of original melody and lyrics (known as the topline), as well as chord structure and instrumentation. Co-writing can involve participation in any of those processes, and contributions can be equal or unequal to other cowriters. Both traditional ethnomusicological studies (Street et al., 2008;Turino, 2008) and sonic ethnographic research in other disciplines (Cambria, 2018;Fuhr, 2011) offer examples of joint participation in music-making but not many involve the specific practice of collaborative song-writing itself. For this part of the fieldwork, pop music and production studies were combined with ethnomusicological methods to garner practical guidance (Collins, 2007;Bennett, 2011Bennett, , 2013.
In line with musicking ethnography, I also collaborated in the organisation of performance events as part of the activists groups Sudan United and Above the Noiseboth are grassroots collectives of Sudanese in the Netherlands voluntarily supported by a small and dedicated group of Dutch community development and media specialists. In these groups, I attended planning meetings as a person who was on hand to help with small tasks such as editing short documentary films, making music videos and gig posters. In addition to this, some of the most enjoyable times were spent playing music on YouTube while being taught to cook Sudanese staples such as ful, bosh and molokhiya.
As is broadly true for all ethnography, this project also draws on material outside of the actual musical experiences in which I was embedded. This fieldwork was happening at a critical time in Sudanese politics when events moved rapidly and culminated in a revolutionary moment. Every day there was something new to talk about, and sometimes music took a back seat in our conversations. Occasionally, the Sudanese activists in our group would seem puzzled about why I was so interested in music when there seemed to be more important things to investigate. I had to acknowledge that there were clearly times when music drove activism and times when it seemed peripheral, even to the musicians themselves.
Far from being irrelevant or merely background information, I would suggest that these non-musical discussions and extra-musical experiences were important 'listening events'. Especially because this musicking ethnography placed song-writing at its centre, it was important to also be part of the absorption of non-musical information. Creative acts like song-writing are always responses to, and in dialogue with, what has been listened to beforein both musical and non-musical senses. The writing processes I shared with the members of Twin Niles were shaped by our experiences of listening to both musical and non-musical provocations. Thus, in coming to know about the social and political world of music-making, 'listening with' musicians also formed an important part of this ethnography. This was a useful way to distinguish between the types of reflections on everyday events which occur in song-writing and music-making compared to nonmusical political actions.

Epistemology I: Sense and experience
One of the alleged epistemological dividends of creative collaboration is that it has the potential to elicit insights which other methods cannot bring about (Degarrod, 2013;Loaiza, 2016;Mullett, 2008;Vacchelli, 2018). As summarised by Pink (2009) 'A sensory ethnography invites new forms of ethnographic knowing and routes into other people's experiences. It provides us with ways of responding to research questions that involve focusing on forms of intimacy, sociality and emplacement' (Pink, 2009: 153). Sensory and embodied knowledge fall under the category of 'experiential knowledge', as outlined by Heron and Reason (2008), and contrast with both 'propositional knowing' (knowing about something) and 'presentational knowing' (knowing through reflection on experiences). Experiential knowing is 'being present with' or 'being face to face' during 'direct encounters' and is rooted in the immediacy of perception and their resonances on the body (Vacchelli, 2018). In musicking ethnography, therefore, 'the field' is 'reconfigured as an experience rather than a place' (Hellier-Tinoco, 2003: 26) in which the researcher falls into a 'no-place' (Rice, 2008: 51) between an insider (one who is able to access experiential knowledge) and an outsider (one who relies on presentational knowledge).
To report something epiphanic about creative collaborative practice would be an overromanticisation of an experience during which my outsider status was always evident to me and my collaborators. Learning Sudanese songs and singing them over a beat did indeed produce experiential knowledge in the sense that I grasped the feeling of the melodies as I used my own voice to replicate them. My hands formed muscle memories of chord patterns. I felt basslines rumble through sound systems and was part of the jostle of bodies dancing in front of stages. This opened up my understanding of how music brings about political belonging through 'shared sensing'. But as for experiential knowledge of how music constitutes the transnational political field of globally connected revolutionaries, I was inevitably at one remove.
To answer my research questions, autoethnographic experiences of collaborative creative practice were on their own insufficient. They needed to be brought to bear, in a theoretically informed way, on more traditional participant observations. Specifically, I needed a layer of theory to overlay onto my own sensory experiences in order for them to play a part in my analysis of participant observations. Relational musicology combined with theories of sensory intimacy allowed me to derive an epistemological value from my own sensory embeddedness.
As theorised by Schütz (1951), Small (1998) and advanced by others (Born, 2010;Crossley, 2018), music-making is an experience of a set of relationships; firstly, between organised sounds but also between people, between individuals and society and even between humans and non-humans. Importantly, participation as a musician-researcher is an intervention into these relationships. Even adding just one note into a piece of music creates harmony or discord with every other note and, in the process, changes their perceived, if not original, character. When you add your own voice, melodies or rhythms into a mix of sounds, your contribution ripples through the whole ensemble. You are brought to an understanding of the relationships between organised noise, between people and between societies through your own disruption of them.
Experiencing this web of sonic and social relations from the positionality of my collaborators is indeed impossible. However, being part of the relational web rather than outside of it looking in meant that the imaginative leap required in all attempts to comprehend another's experience was made shorter. Pink (2009) refers to sensory ethnography producing 'intimacy', and I would agree that becoming part of the relational web of musickingrather than viewing it from the outsidedid make possible a closer 'intimation' of musical sensing as it is experienced from any other part of the network. My sensory experiences of musicking allowed me to take my own sensing as an anchor point from which I could comprehend the data I received during participant observations. In this research, therefore, the field was indeed 'an experience not a place' (Hellier-Tinoco, 2003: 23), but emplacement was not irrelevant; my experience was not of place but in place among a weave of sonic and social relationships. From there, I was able to invoke networked sensory intimacy in my analysis of participant observations and glean some situated experiential insights into the significance of musicking on transnational activism.

Epistemology II: Emotion and affect
Proponents of collaborative creative methods also claim that they garner emotional knowledgea sub-type of experiential knowledge which is implicitly contrasted with cognitive or rational knowledge. This has brought about a tension between those who choose and those who choose not to complement musicking ethnography with interviews. As argued by DeNora and Ansdell (2017: 232), being interviewed about musical experiences involves a 'proxying' of 'experience' into 'narrative' which may not be commensurate with what happens during actual engagement in musical experience. Music-making can give a sense of the 'feel' or the 'spirit' of something which you cannot really put into words. One could argue this goes some way towards explaining the very existence of music in the first place. Because of this, DeNora (2013) suggests, immersion in musicking garners more ecological validity than interviewing participants about their own musicking. Furthermore, it has been argued that insisting on the conversion of emotional experience into discourse echoes the Western urge to notate (Visse et al., 2019). Thus, there is a strong decolonial critique of the interview as a method for understanding musical experience.
By contrast, several other proponents of collaborative creative practice propose that dichotomous distinctions need not be made between reason/emotion and language/body. Fuhr (2011: 238 argues it is only through the integration of both discourse and musicking experiencesand through analysing both in 'constant interrelation'that she could gain emotional knowledge about the meanings music held for her participants. Ellingson's (2017) work on embodiment also shows that '[i]t is not that the mind makes (cognitive) knowledge out of the raw perceptive (bodily) information. Instead, the mind-body system feels, thinks, experiences, and makes sense of sensory information holistically, which then may be rendered under a rational/cognitive guise via language' (Ellingson, 2017: 18).
Concurring with Fuhr's and Ellingson's holistic approach, in this research, I found a complex intersection of emotional and cognitive knowledges in both sonic and lexical media. It is true that the sung voice, compared to the spoken voice, has a far wider palette of expressive elements which can enable a 'tapping' of emotional energy. I can certainly remember some quite arresting moments during musicking with Twin Niles. But, on closer reflection, these were not experiences of raw emotion but of emotions which had been treated and controlled to become performances of their Platonic ideals. Art is always, to some extent, artifice and, like discourse, music-making also requires the conversion of emotion into a communicative medium. Furthermore, many of the musical exchanges in this creative collaboration were highly cerebral; learning a melody accurately or figuring out the ideal tempo of a song was mentally taxing for all involved.
At the same time, informal interviews were ongoing throughout the project and took place in the form of conversations at rehearsals, in houses, at gigs, backstage, during protests and demonstrations. Formal interviewsof which there were 12 in this projectwere not much more formal than this except that they occurred at specific times which had been set aside for the purpose. Unlike informal interviews, there would be only two of us present, sitting together with a recording device between us. The conversation was more one-sided than informal interviews but maybe only marginally so. Most of the time, formal interviews took around 2 hours within a whole day or half day spent together, meeting in a café for the interview and then going on to protests, to explore the city, visit record stores, jam or attend concerts.
Like Bennett (2013) and Fuhr (2011), I found that interviews added to my comprehension of emotion. Interviews allowed me to garner more specific insights into my participants' relationships with music, activism and the paths which led them towards both music and activism. It was an opportunity to ask direct questions and probe if necessary, something for which informal settings do not allow. Certainly, different forms of knowledge came out of the interviews compared to musickingprimarily, useful propositional knowledge which provided important background information. In knowing something about a musician-activist's life story, the significances of musicking for sociopolitical activism could be more richly contextualised and historicised. As well as this, questions which asked about participants' experience of music and their significance for their activism provided important presentational knowledge which was not at all short of emotion. Throughout my research with Sudanese participantsmusicians or otherwise -I have been moved by my interlocutors, almost all of whom have a talent for poetic speech.
While creative collaborations and interviews resulted in both emotional and cognitive knowledges, it was only in their combination that I was able to address my research questions. Having experienced emotionand its complex intersection with cognitionas a participating musician, I was able to understand the reflections made in interview discourse as a researcher better than if I had not participated in collaborative musicking. Again, relational musicology in combination with sensory intimacy helped me to theoretically link my own emotional knowledge with that of my participants. I could not assume my own emotional response to musicking was equivalent or comparable to my collaborators. For this, I needed to ask them about this during interviews and use my own emotional knowledge to enhance my analysis of their responses. When analysing these responses, the imaginative leap required to comprehend their experiences was significantly shortenedor to elaborate on Pink's (2009) termmade more 'intimate' by my own correlating emotional responses to musicking. This indicates there is a distinction to be made between gaining 'knowledge of emotion' (propositional knowledge) and 'emotional knowledge' (experiential knowledge) in and of itself. In making this distinction, I can consider that I received emotional knowledge as a result of musicking. This combined with reflections to create presentational knowledge which, when used to inform analysis of interviews, brought me towards a presentational 'knowledge of emotion' among my participants. It was in the combination of interviews with collaborative creative practiceand their theorised relationalitythat I garnered insights for my research.

Ethics I: Power and status
The inherent coloniality of the ethnographic method remains a long standing critique with which its advocates still grapple (Mackinlay, 2015). Combining ethnomusicology with arts-based participatory methods has been muted as one way to counter-act the colonial complicity which ethnographers face (O'Neill, 2018). Arising out of concerns with power imbalances between researcher and researched populations (Gaventa and Cornwall, 2008), participatory research has found its niche in studies which include vulnerable populations and/or in cases where the identities of researcher and researched are historically embedded in patterns of exploitation (Rahman, 2008). The activity of creating something together, rather than the researcher extracting information from the participant, allegedly mitigates the native-imposter dynamic.
These purported ethical dividends made collaborative creative practice an appropriate choice for my research project. Many of my research participants were seeking asylum and therefore belonged to a vulnerable status group. As a young ethnically white female researcher, a native English speaker with middle-class professional status, I recognise that recruiting predominantly male subjects from the Global South creates an inevitable power imbalance. Added to this, the history of Sudan as a protectorate of colonial Britain cannot be ignored. All these factors increase the risk of exploitation, epistemic extractivism and colonial complicity. In my previous research with non-musician Sudanese activists, power imbalances arising from my visible privileges have affected our interactions. For instance: After dinner Ali [member of Sudanese youth activism organisation] shows me the location of his father's farm on google maps. He asks me straight-faced 'Do you know that Sudan is a really big country?' 'Well, not as big as it used to be!' I say smiling, referring to the secession of the South in 2012. Not taking the joke, Ali then informs me of some of the basic geographical facts of Sudan. Strange because he has known me for years as someone who researches Sudan and I have just got back from seeing his family in Khartoum (extract from fieldnotes, Netherlands, 29 June 2019).
This small anecdote is indicative of how most of my non-musician research participants prefer to speak with me about Sudan. They would rather I listen and learn as though I was hearing about Sudan for the first time from them. This dynamic serves to put me in a lower status than my visible identity indicators might dictate. I have not asked my participants why this is the case, and I would not want to assert my privilege by raising the issue. However, these dynamics make me question whether my status as a young white British female academic with knowledge about Sudan makes my participants feel as though their space is being intruded upon. With this in mind, conducting research with musician-activists in as ethical a way as possible seemed to have been served by the choice of collaborative creative practice as a method.
It was true that my relationships with musician activists were totally different to those I have with non-musicians. To illustrate with a contrasting anecdote: In the car on the way to the studio, Kam, in a completely relaxed fashion asks me why I do my research on Sudan. I gave some explanations about how it's an interesting place which combines North African with Sub-Saharan political cultures, its history has elements of East and West. He takes his eyes off the road momentarily to look at me and with a wry smile says 'So, white guilt then?' 'Yeahhh' I say with a sigh. We both laugh out loud (extract form fieldnotes, Khartoum, 07 September 2018).
An unguarded conversation like this in which our respective identities are explicitly noted and made into the butt of a joke was common among my musician participants in Sudan, the Netherlands, and the UK. Nothing like this has ever happened with nonmusician participants.
My experience suggests that this is due to the ways that status is configured in musical collaboration. It is important for musicians to understand one another's skill level and type, and to figure out ways of playing together within that constellation of abilities.
Meeting the other musicians in Twin Niles therefore involved a ritual of 'sizing up' where these levels are evaluated and a shared practice established (or not, if alignment cannot be reached). These encounters involved instances of hierarchical flipping since it is often the players with the lowest level of technical ability (in this case, myself and Mosib) who end up leading improvisation sessions; more technically adept musicians can work around them.
In addition, musicians naturally find themselves in peer circles or 'scenes' whereby people are brought together due to having shared career experiences and skill ranges. In this study, the Sudanese musicians I ended up collaborating with were my musical peers as far as possible. We all had reached a similar level of semi-professional status in music while being accustomed to writing and communicating without classical training. We all knew about hustling with bookers, agents, publishers and what it means to compete for space at the over-crowded lower echelons of the music industry. These were the topics of so many of our conversations and became a good basis of our friendships. I was always going to be recognised as a peer by those who had similar experiences to me but as an interloper by those who did not. Unlike with my non-musician participants, we had this common ground on which to build a relationship.
Certainly, it would not be right to say that musical skill level and type are all that matter in musicking. Of course, gender, racial and class dynamics play out, not least because they also determine how these skills and experiences are perceived and assessed by other members of music scenes. These identity indicators were relevant to how we related as musicians, for better or for worse. But specifically when it comes to addressing the inherently colonial researcher-researched dynamic identified by critics of conventional ethnographic methods, musicking went some way towards mitigating this imbalance. In a mundane sense therefore, musicking lends itself towards fostering peer-level relationships between researcher and researched populations even in enquiries in which political identities threaten to interfere.

Ethics II: Empowerment and expression
Participatory research is said to empower participants due to the opportunity it gives for self-expression. The vast majority of participatory arts-based studies focus on collaborative practices as an avenue towards self-expression among those for whom formal interviews are inappropriate or intimidating (Aure et al., 2020;Ugolotti, 2020;Vacchelli, 2018;Furman et al., 2019;Mullett, 2008). During participatory research, overstepping is a deep concern and researchers are encouraged to relinquish as much control as possible and to allow participants to drive processes in order to maximise participants' opportunities for self-expression (Grant et al., 2008: 592-593). For the participants in this project, however, music was not an alternative mode of expression but a life-long passion and a semi-professional undertaking. My participants indicated to me that interviews were not an intimidating or inappropriate medium for self-expression. As already stated, interviews were married with, rather than replaced by, participatory elements in this study. The research therefore required a methodological supplement from the field of songwriting. Bennett (2011: 7) has outlined seven main forms of co-writing. They range from full 'demarcation' modelswhere, for example, a lyricist will work separately to a musicianto the 'Nashville' model where two writers work together usually in the same room and with no pre-assigned roles. When preparing songs for professional release and registration with the Performing Rights Society, all co-writers are required to agree on their percentage contributions. The relevance of this practice for this study is to demonstrate that partnership in co-writing is not focused on expressive equity but on the proper dovetailing of expertise (Barrett, 2014). In short, in the event that there are two co-writers, they are not striving for a 50-50 contribution (although this might indeed arise) but are instead focused on what proficiencies and ideas do they have between them and how can these be best combined in service of good creative activity.
Far from combining easily, the participatory requirement to prioritise empowerment through self-expression clashed with the aesthetic commitment required in the songwriting method in this project. According to the participatory method, Mosib's opportunities for self-expression should have been maximised, and given priority through the creative process. Expressive equity would have required something along the lines of the Nashville method (Bennett, 2011) in which the creative process was shared from the outset. However, in practice, the first song we worked on required my creative input in order to kick off. We began with my chord structure and topline. Mosib added log drum and a second vocal line with new lyrics. The rest of the chord sequence was worked out with the band during jam sessions. Both Mosib and I continued to work on recording the song with me leading this part of the process.
As the process continued, when making decisions in song-writing, myself and my collaborators were taking courses of action which would serve the best aesthetic outcome and not those which would maximise opportunities for self-expression. If a player's contribution was not fitting well in a song, we would replace it with something else. This happened with many of my parts, and many of Mosib's. If we disagreed about what sounded good, I was sometimes in a bizarre situation where I was making the case to delete sections of my collaborators' work and replace themperhaps even with my own ideas. Those committed to participatory practice could view this as an instance of 'silencing' but it would have been patronisingand indeed disempoweringfor me to act otherwise. We were, after all, creative peers.
Unlike most arts-based research, therefore, the invocation of an aesthetic method did not naturally serve participatory goals. In this study, therefore, it was more important to take on board the spirit, rather than the letter of the law when it came to participatory practice. This is because, as song-writers, both myself and my collaborators were driven towards creating something with aesthetic value (something that sounded good to us and that we could be proud of making). This motivation had its own logical division of labour which commitments to text-book participatory practice only disrupted. Respecting collaborators and avoiding overstepping are indeed highly relevant practices. Keeping that in mind, this research process went more in service of those aims when I inhabited my role as songwriter and embraced its associated aesthetic method. Primarily, this involved an honest appraisal of my own strengths and weaknesses and how they could tesselate with my co-writers for the best aesthetic outcome.

Ethics III: Transformation and training
Arts-based participatory methods have strong links with critical pedagogy through their roots in Freire (Coghlan and Brydon-Miller, 2014) and via their complementarity with action learning theory (Wood, 2019: 42-44). Many typical participatory studies involve peer-training ( Thomas Bernard, 2000: 182). Indeed, this is one way of ensuring that participants receive some concrete and sustainable benefit from being part of research projects (van der Vaart et al., 2018: 26). In my research project with political activists, it was important to leave some lasting positive benefit rather than merely swooping in and extracting information.
Taking this on board, I envisaged I could train Mosib in recording and productionthis would add a string to his bow as a musician and also open up opportunities for expanding his use of music in activism. At the start, we were both enthusiastic for this, but as the project went on, I realised that the recording of music was almost anathema to the passion that he had for it. The recording process I am used to, and what I was offering to train Mosib in, involves recording each individual component (e.g. percussion, bass line and lead vocal) to a metronome known as a click track and sometimes also a guide vocal or guide piano. Each player is alone when they record their piece, and they listen to the backing track through headphones. These fragments are then layered on the top of one another piece by piece in production software where they are treated with various production effects and their relative volumes and placement in stereo worked out. The raw and unproduced layers played together are known as a mixdown. Normally, I avoid showing this to anybody, including participating musicians, because it can be a rather chaotic and lifeless draft which does not give a good account of what the end product will be. After making these individual recordings (known as stems) over many weeks, Mosib asked to listen to the mixdown, eager to hear the results of our recordings. Against my instincts, sent him the file.
The mixdown sounds pretty terrible, but it is not unsalvageable. Can't get Leno to play those nice licks on GT2 like in the jam […] And Mosib's vocal needs EQ scoop. Double bass sounds great with the two mics. Mosib hates the whole thing! He says the Iphone demo is better and wants to re-run the session with us all playing live. We have to be 'in the groove and in the moment'. He 's right, but I think we can pull it around, I just wish we'd played to the jam demo instead of a click track. That way it could have recreated the feel (extract from fieldnotes, Netherlands, 5 May 2019).
Both of us were missing live energy in the mixdown; I wanted to supply this after the fact through production and polish, but Mosib wanted to achieve this by going back into the live moment and reliving it. This can partially be explained through Mosib's love for Sudanese traditional percussion which, at its core, is live and collective. In an interview, he explained that the rhythms were originally used by villages to communicate with one another. Rhythmic patterns were learned off by heart and played in unison by whole communities to send messages to those living close by. 'It simply would not make sense for one person to play a rhythm alone' he said. Similarly his love of reggae which, although recording is customary in this genre, Mosib describes as 'coming to life when we jam it'.
For me, however, being an electronic writer and producermany of my songs are only played live by musicians after they have already been written and recorded. They are composed of layers upon layers of wave forms which have only ever been melded together in the audio interface. Even when recording acoustic instruments, I prefer to conceptualise them as waves and get them acting as such in the production software as soon as possible in the writing process. The live aspect of music, for me, is normally not part of the construction of the song.
Our different relationships with music were made evident in this episode, and our subsequent co-writing relationship was driven by it. As the months went on, both of us changed and adopted aspects of one another's approach. One might call this a ritual of 'tuning-in' as outlined in Schütz' (1951) seminal paper. I had my eyes opened to the creative potential of the collective jam and similarly Mosib began to see the aesthetic benefits of production. By the end, we had developed a shared practice which combined both of our preferences. I would record Mosib and others improvising with percussion and vocals and then add in electronic elements after the fact.
Coming to a natural understanding of one another's relationship with music proved the lynchpin of successful co-learning in this study. A sustainable co-learning component, and not a tokenistic one, must include sufficient time to come to know one another as creators. It does not make sense to speak of 'training' in this scenario even though one party might be more adept in certain methods than the other. A creative collaboration will involve mutual tuning-in as a matter of course, and the study design should be set up for this from the outset.
But did this collaborative creative practice achieve participatory transformation? Mata-Codestal et al. (2020: 203) describe the 'main aim' of participatory research as the 'equal distribution of power among all those involved in the research process'. In participatory studies, new knowledge issues from research after the fact of empowerment. In contrast, more conventional social science methods seek to obtain knowledge as their main aim, and they recognise that this should be done in such a way that ideally empowers, but at the very least does not disempower, participants. Because we were focused on creating something with aesthetic value, empowerment through co-learning was sometimes deprioritised to the level of a secondary or tertiary goal. In the interests of making something sound as good as possible, the novice's contribution would give way to the expert's. Many of my improvised guitar parts have been confined to the bin, just as some of Mosib's early vocal takes have been. Our learning might have suffered from this in the immediate term and even in the lifetime of the research.

Conclusions
Having reflected on the epistemological and ethical issues arising during this musicking ethnography, I can come to some tentative conclusions which promote collaborative creative practice for socio-political research. In relation to the alleged epistemological dividends, I found that autoethnographic embedding in creative collaboration can play a part in the production of embodied and emotional knowledges. Importantly, these insights were only gained in this research when autoethnographic sensory experiences were combined with both participant observations and interviews. Furthermore, the relationship which links autoethnographic experience to the experiences of others required theorisation. I used relational musicology and sensory intimacy to provide the groundwork on which my own sensory embedding could be utilised to enhance my analysis of participant observations and interview responses. Without this, my autoethnographic embedding would have been confined to self-knowledge and perhaps found guilty of the self-indulgence sometimes identified by critics of creative methods. To sum up; my own ethnographic experiences did not constitute findings but rather constituted an important part of my analysis of participant observations and interviews.
On the ethical points, I concur with the claim that creative collaboration can mitigate some imbalanced power relations between researcher and researched populations. This happened in this study because determination of skill level and type are the primary basis for interactivity during musicking. On the question of whether or not collaborative creative practice can actively empower participants, I am more uneasy about the claims made by its promoters. I found that maximising self-expression and sustainable benefits through co-learning did not necessarily fit naturally with the aesthetic goals inherent in the song-writing method. Rather, I found that the pursuit of aesthetic value had the potential to actively derail direct opportunities for self-expression and co-learning.
Overall, reflections on this research project indicate that collaborative creative practice should not be considered as a straightforward sub-set of either ethnographic or participatory arts-based research. Autoethnographic embedding is insufficient for socio-political research questions which aim to contribute to debates outside of ethnomusicology. Similarly, in creative collaborations, aesthetic value is an end in itself and not, as in participatory practice, a means to the primary goal of empowerment. This can result in occasions where participatory values are deprioritised. These distinctions are important and should be recognised in study designs. Overall, in this research, aesthetic micro-methods dovetailed productively with ethnographic and participatory methods to bring about some epistemological and ethical enhancements, albeit with some important qualifications.