On liveness: using arts workshops as a research method

Drawing on a research project using arts workshops to explore pain communication, we develop a methodological reflection on the significance of the liveness of arts-based methods. We discuss how liveness informed the design of workshops to provoke novel forms of communication; how it produced uncontrollable and unpredictable workshops, whose unfolding we theorize as ‘imprography’. It also constituted affective and collective experiences of ‘being there’ as important but difficult-to-record parts of the data, which raises challenges to current understandings of what constitutes data, particularly in the context of team research and in light of directives for archiving and reuse. We explore the implications of liveness for methodological practice.


Introduction
Social researchers use arts-based methods to address particular kinds of methodological problems, such as how to access knowledge which is not easily expressed in words or with participants for whom language is difficult (Bagnoli, 2009;Barone and Eisner, 2012;Leavy, 2009Leavy, , 2015Stuckey and Nobel, 2010). Such methods can also be a way of presenting material to different audiences and increasing accessibility (Bagley and Cancienne, 2001;Bartlett, 2015;Denzin, 2003;Foster, 2012). Within arts-based research there are two broad traditions: participatory approaches, which emphasize the agency of participants and the possibilities of engaging in new modes of enquiry with them (Wang and Burris, 1997); and performative approaches, which emphasize that methods themselves create or perform their social realities, and which thus pay careful attention to the forms these methods take (Law and Urry, 2004). These traditions differ in their conceptions of agency and materiality, but share a broad commitment to producing particular social realities from the research process, and to the view that research makes things rather than simply documenting them.
Performative approaches have recently called for the development of 'live' (Back, 2007;Back and Puwar, 2012); and 'inventive' (Lury and Wakeford, 2012) methods. Back has argued for the need for methods which 'document and understand social life without assassinating it ' (2007: 164), on the basis that 'the lacklustre prose of methodological textbooks often turns the life of the research encounter into a corpse fit only for autopsy ' (p.163). Similarly, taking research as an active engagement with an everchanging social world rather than an investigation of a static reality, Lury and Wakeford (2011: 2) argue for the development of 'inventive methods' as methods which 'enable the happening of the social world -its ongoingness, relationality, contingency and sensuousness -to be investigated'.
This article takes the notion of 'live' methods further, exploring the significance of liveness across the research cycle by reflecting on our recent research using arts workshops to investigate new ways of communicating about chronic pain. We explore the importance of liveness in the workshops' design, unfolding, analysis and re-use. We argue for the importance of thinking about method beyond individual aspects of data collection, recording, or analysis, to consider the broader research assemblage. The meanings of the workshops were not contained in any of the individual outputs (the 'data' in a traditional sense). Rather their research significance lay, in part, in the experience of participation, in difficult-to-record phenomena of affective engagement, ambiguity, or discomfort, whose traces were recorded in experience, memory, or skill development. These meanings were also constituted through their participation in a broader assemblage of epistemologies, researchers' disciplines, participants' histories, and workshop leaders' professions, among other influences. The article therefore focuses on the workshop as research process rather than arts-based methods as research product.
Our approach here contrasts with more participatory approaches to arts-based research which focus on 'giving voice' to underrepresented populations (Conrad, 2002) and presenting data in new ways (e.g. Bagley and Cancienne, 2001;Cox et al., 2009;Saldaña, 1999). It also moves away from evocation and empathic experience as primary reasons for undertaking arts research. For Barone and Eisner (2012) arts-based methods aim 'to create an expressive form that will enable an individual to secure an empathic participation in the lives of others and in the situations studied' (p.8-9). Leavy (2009: 17) suggests that arts-based research should meet criteria such as 'how does the work make one feel? What does it evoke or provoke? What does it reveal?' While these are valuable in many contexts, evoking empathy was not a central aim of our use of arts workshops. Participants did develop images and metaphors that evoked their pain-and at times provoked it, as in the case of sounds, such as a spoon rubbing against a grater or the screech of windshield wipers, that were brought in to represent pain aurally. However participants emphasized their desire to protect others from their pain; to not share it. Rather than better representing the experience of chronic pain, our workshops aimed to rework chronic pain communication itself, by changing the frame in which it occurred: from a series of isolated (and isolating) interactions between a clinician and patient or a sufferer and non-sufferer, to a collective space for discussing, sharing and reinterpreting the experience of pain. In keeping with the aims of performative approaches, we wanted to reframe pain as constituted in and through the social interactions between people. The art produced in the workshops is therefore not the whole story or even the primary expressive aim. Rather, it was the process of participation which was central, and the improvisational space that unfolded within the workshops was of primary importance, as we document below.
Our approach also differs in important respects from existing work in performance ethnography (e.g. Denzin, 2003;Saldaña, 1999). Performance ethnography brings liveness to the research process through theatrical techniques, by writing and disseminating ethnographic work as performance. This allows participants to explore and adopt other perspectives by taking on new roles, and illustrates those perspectives to audiences in ways that are lively and engaging. Performance ethnography can thus be both a way of engaging with other viewpoints, particularly those of underrepresented populations (Denzin, 2003) and new audiences. Such work evokes liveness in its presentation and development, but differs from our workshops, which were not about inhabiting new positions or giving voice to underrepresented people so much as challenging and bringing uncertainty to one's own position. We did not aim to produce performance as an outcome, but were interested in what happened improvisationally within the workshop itself.
We begin by situating chronic pain communication in relation to the literature on pain, health and art, to articulate the rationale for live methods as means of generating new versions of pain communication. We then discuss the design of the workshops, and their unfolding in practice, to illustrate the challenges of the improvisational nature of live methods. This is followed by a consideration of the question of 'what is data?' in the context of live methods, and the implications of this issue for the questions of data reuse. We conclude by addressing the methodological challenges raised by liveness, and suggest ways forward for future research.

Art, pain and health
Chronic pain is a significant biopsychosocial phenomenon which is still not fully understood. It is difficult to communicate and cannot be measured objectively. Standardized diagnostic tools such as the McGill Pain Questionnaire (Melzack, 1975) measure pain by translating it into words describing its sensory, spatial and evaluative properties. More commonly, pain is measured on some form of Numerical Rating Scale or Visual Analogue Scale (Price et al., 1983). However, qualitative studies such as those by De Souza and Frank (2000) have shown that patients' own spontaneous linguistic descriptors do not necessarily match those of the MPQ. Moreover, in studies on clinician-patient communication about pain, people with pain repeatedly report feeling misunderstood or disbelieved by clinicians (Kugelmann, 1999;Werner and Malterud, 2003). Communication about pain often repeats binaries of real-imagined, or mind-body, in ways which delegitimize and stigmatize the experience of people who have pain (Kenny, 2004).
Against this background, our research sought to explore alternative ways of communicating about pain which did not rely on standardized descriptors. We were interested in non-verbal forms of communication, and what methods from the arts might offer as expressive resources for people with pain (Padfield, 2003;Vick and Sexton-Radek, 2005). As well as using unfamiliar materials for communication, we also sought to create an unfamiliar environment: one not marked by instrumental interests in producing a diagnosis, treatment or medical outcome. Thus, we aimed to create a live situation, in which communication about pain could unfold in new, unfamiliar ways.
Arts-based qualitative health research is often focused on enabling aspects of health experience to be heard which might otherwise be silenced. Stuckey and Nobel (2010) review current literature on art and health and note a wide range of studies emerging in music engagement, visual arts, movement-based creative expression, and creative writing, observing that 'art helps people express experiences that are too difficult to put into words' (p.256). Fraser and al Sayah's review shows how arts-based methods have been used for uncovering or producing knowledge in health research, and for translating and disseminating research to new audiences (2011). They note that visual arts methods such as drawing and photography are most commonly used for producing new knowledge, for example about how participants' perceive the world or their experiences of surgery (Radley, 2010) or how they experience chronic conditions such as heart disease or menopause (Guillemin, 2004). Performing arts methods are more frequently used for making research findings more vivid, as in Bagley and Cancienne's (2001) work on 'dancing the data', or for engaging new publics, as in Cox et al.'s (2009) use of theatre performance and audience evaluation to make the issue of preimplantation genetic diagnosis accessible and engage citizens and stakeholders in developing public health policy.
Arts workshops themselves have been used elsewhere as a research method (see Butterwick, 2002;Conrad, 2002;Donmoyer and Yennie-Donmoyer, 1995;Lambert et al., 2013 among others). For Hewson (2007), theatre workshops provided a place for education students to learn to tackle challenging situations and conflicts, an 'intermediate step' between the textbook and the classroom. Collie and Kante (2011) describe art therapy workshops with marginalized women with breast cancer, predominantly from ethnic minority groups. They argue that the workshops enabled new forms of communication for participants who experienced major challenges which they had often not articulated. In Rooke's (2010) work exploring the science of sex and gender through art with transgender youth, workshops produced a creative and inclusive space in which the young people felt safe and were able to choose how they wished to be identified, questioned medical authority, and built a sense of solidarity, producing new forms of relating and communicating. Rooke reflects on the uniqueness of the space created by the workshops and the sense of loss experienced by many participants when the workshops concluded, but says less about the workshops themselves as a methodological tool. In this paper we build on the existing literature on arts workshops by aiming to unpack and conceptualise the methodological affordances of workshops as 'live' improvisational spaces.

Arts workshops as live methods
Four workshops were held fortnightly on Saturdays over a period of two months. A total of 22 participants attended the four workshops, with 17 of these attending only one while five attended multiple workshops. The majority were attendees with pain, while two were carers and five were pain clinicians. Workshops were held out of hours at a health treatment centre with spare rooms with beds for participants to lie down if they needed to rest, as well as mats, balls, pillows, and other equipment.
We hired professional artists to run the workshops, each in conjunction with another practitioner with complementary skills. The first workshop was run by a visual artist with expertise in medical work together with a neuroimaging specialist who could talk about fMRI scanning; the second, on 'body mapping', involved a photographer skilled at working with people with pain and a somatic education teacher who worked with bodies in pain; the third was led by a sound artist and a music therapist; and the fourth, on spatial mapping of pain, by a physical theatre practitioner and a cultural geographer.
Each workshop began with an overview of the project by the research team and participant introductions. The workshop leaders then introduced the topic of the day and an initial exercise. There was a lunch break, after which the second half of the workshop usually revolved around some transformation of the discussion or materials from the first part: re-imagining the process of neuroimaging through body scan meditation represented with art materials; considering how a pain-related object photographed earlier might be transformed in some way; working together to produce something to 'treat' painful sounds; and dramatizing relations to spaces of inclusion and exclusion.
Our introduction to the workshops was carefully choreographed. Pain specialists had advised us that in a workshop scenario, people would likely introduce themselves and their experience with pain by telling the story of their condition. Stories of misdiagnosis and incomprehension would be time consuming and risked derailing the workshop's intent by reinforcing divisions and feelings of miscommunication between people with pain and clinicians or carers. With this in mind, we asked participants to introduce themselves through other experiences, such as their background in art-making, the objects they had brought to represent their pain, or what they had been thinking on the way to the workshop. These introductions enabled people to learn something about each other without immediately focusing on their role as clinician, patient, or carer, or indeed as researcher or artist/workshop leader. Stories of pain conditions were still told, but in relation to the workshop activities and in dialogue around them rather than as individual narratives.
This narrative displacement had profound effects in terms of enabling communication around pain. What was most effective-and affective-in the workshops was the improvisational element itself. Each person was slightly off balance in terms of their usual role. As researchers we had to step back and allow things to unfold in ways that were unusual and sometimes uncomfortable; equally, we had to draw on our own experiences and become participants in the workshops in ways we would not have done had we been leading them. The workshop leaders were often unfamiliar with working with people with pain, and with one another. People with pain were not there simply to convey their individual stories; clinicians were not there to diagnose or even merely to listen to patients, but had to bring in their own experiences and work from their own bodies. The carers who attended did not work with the person they were accustomed to caring for during small group work. This role unfamiliarity meant each person needed to read the other participants for cues as to how to proceed. By stepping outside these roles, a more egalitarian space was created in which we could think through together what and where pain might be, and how it could be translated from one form to another: from body to image, to object, to sound, and back again.
We had aimed to open new ways of communicating about pain and to understand pain as relational, in contrast to traditional views of pain which see it as stuck within an individual trying more or less successfully to get out (Kugelmann, 1999). We anticipated that arts-based methods would open new channels of communication, but did not anticipate the extent to which the live encounters of the workshops themselves would be critical to the communication that resulted. What was most interesting and productive was not necessarily the art itself, but the experience of producing art together. The workshops opened up what we are calling an imprographic space, drawing on the language of performance as well as performative social science (Law and Urry, 2004). Imprography is an integral part of the 'liveness' of the workshops as a method and what they accomplished.

Unfolding of the workshops: imprography
Imprography can be understood as a combination of improvisation and choreography. 1 The choreographic element may initially seem unnecessary, since improvisational practice in the performing arts always occurs within certain set structures. Theatrical improvisation has spoken and unspoken rules about the degree of involvement of the team of actors and/or the audience, and about engagement with the theatrical space which frames the performance. Jazz musicians may improvise, but they do so within the structures of the music itself and the instrument they play (see also Sudnow, 1978). Dance improvisation too contains these structured elements: steps, movement vocabularies, the boundaries of the studio or theatre where it occurs. What sets 'imprography' apart is the quantity and specificity of structured elements.
Most qualitative research could be conceptualized as falling on a continuum somewhere between fully improvisational (unstructured interviewing resembling this most closely) and entirely choreographed (as in the standardized questionnaire). What distinguishes our work here is that once the research process began, we had little ability to step in and change the course or direction of the events, which were directed by workshop leaders and participants. While initially challenging and discomfiting, it is precisely this displacement that was central to the method's success.
Our inability to intervene in the unfolding of the research process is the main contrast with more traditional 'people-based' qualitative methods. Even a relatively unstructured interview is directed, through prompts and probes, by the interviewer. While our participatory role in the workshops could be considered ethnographic, is important to preserve the difference between ethnography, participant observation and other qualitative methodologies (Hockey and Forsey, 2012;Ingold, 2011). Our research process resembled ethnography insofar as it was through that process as a whole that we learnt to frame pain collectively with participants, displacing language frameworks in the process of working with materials. Imprographic methodologies however differ from most ethnographic designs that emphasize long-term engagement, documentation and interpretation. Our learning derived from engaging material, processual and gestural dimensions as ways of learning, focusing on opening what Ingold (2013) termed 'possibilities of correspondence', ways of knowing 'with' rather than 'about' the body.
Significant 'choreographic' planning had gone into producing the workshop space. In particular, Elena held frequent discussions and meetings with the workshop leaders in order to mentor them through their preparations and ensure the workshops addressed the topics we were interested in and followed the same basic structure. We also undertook consultation with our advisory board of pain specialists; reconsidered the physical space of the workshops and found a location that enabled greater accessibility; planned workshop meals on the basis of significant dietary restrictions around gluten, wheat, and dairy; set up rooms and recording equipment with an eye to the record of the events we planned to produce for ourselves; and developed detailed consent forms which enabled multiple configurations of consent and Creative Commons licensing for workshop outputs.
Yet once the initial round of introductions was complete, we left the workshops in the hands of the artists we had appointed to lead them and became participants ourselves. The workshop then became like a performance unfolding improvisationally before us: we could act as 'stagehands' in setting up recording equipment, tea and lunch breaks, but we could not halt or pause the action, or change its direction.
At times, such as when workshop leaders said things we did not ourselves agree with or which positioned participants in uncomfortable ways, this was difficult to endure. However our non-intervention paid dividends when participants openly challenged views they disagreed with and in doing so, came to a clearer shared understanding of pain. In one case, a workshop leader and bodywork practitioner suggested that participants could separate themselves from the pain, because 'you is you', a core 'you' waiting to be reclaimed. One participant responded: I'm finding it difficult because to me there is the pain and then working against it, and the rejection just makes it worse, so then, the, embracing it, seems to feel to me that it's ok, it is part of me and its alright. But when I'm listening to you, I'm imagining that there is an identification with the pain, so I am getting a bit confused.
Another replied that like it or not, the pain had shaped her because 'I'm not 21, pre-the pain'. Through this exchange, participants negotiated the meanings of pain and the role it played in their self-image and identity formation. While suggesting that pain could/ should be let go and separated from the individual was not something we ourselves would have advocated, the themes of 'going into the pain' versus 'blocking it out', and of whether pain was inherent to one's experience and identity, became a central theme in the workshops.

A data assemblage
One of the most striking impressions left by the workshops was a sense of elation at a collaborative creative process. Participants were overwhelmingly positive, often effusive in their evaluations: the power of producing a photograph and the release it gives you is limitless.

[Evaluation form, Body Mapping workshop]
Sound is a useful way to communicate -importantly because it teaches you to listen! Pain is not only an internal private experience but a shared dynamic of the space around you.

(Evaluation form, Sound workshop)
Following the sound workshop, the following discussion took place: Gordon: Do you think your sense of hearing has changed, having played as a group? Katarina: Yes, definitely. I am now into a heightened state. Can't believe this is over, now I am into it. Naomi: I can't believe I've just participated in this and loved it, I feel like, where did that come from? Before we were in like a performance. Luis: The process of creation is very exciting. The process of creation that's manifesting. … it's interesting that what we are used to hearing isn't any of this. What we created here was us, not me, not in relation to something else, and what we are used to hearing is part of an us, if it's constructed within spaces. Naomi: We should have a label. It is collaborative, you create something outside of ourselves that others could receive. Sophie: It would be fun actually to ask people not to listen to us but to that. 2 Yet the 'buzz' of improvisation itself was not recordable. While our notes remarked upon 'the transformative sense of being part of a collectivity, and being surprised by the collective outcomes' (Flora, Sound workshop fieldnotes) some of this was inevitably lost in the things we traditionally identify as data. The workshops produced three types of outputs: the art works themselves; text-based data about the day, which included our fieldnotes and the participants' evaluation forms; and about five hours of video data per workshop. Fieldnotes and evaluation forms gave us details of affective relations, interpersonal engagements and off-camera interactions, yet remained predominantly text-based, and the evaluation forms rarely gave critical feedback. Video recordings provided a sense of timings, detailed interactions and whole group discussions, but missed interactions that occurred between small groups or outside the room and provided a view from above rather than amidst the action. Arts outputs were evocative and documented process, but were also difficult to interpret without the wider context. Their ambiguity was a strength, as we saw in the final workshop when participants provided differing interpretations of the images and drawings others had contributed, showing how different spaces could be interpreted in different ways. Any full interpretation however needed to put them in the context of the discussion that surrounded them.
There is also another body of data, partially but not entirely recorded in our fieldnotes: things said to us during the breaks or at the end of the workshops; encounters we had observed or participated in during small group work; affective relations developed out of workshop participation; comments and discussions we had with participants outside the workshop context. These 'headnotes' (Emerson et al., 1995) provide a further, fourth layer of data.
Each of these bodies of data is obviously partial, and partially connected. They do not triangulate in a traditional sense. For instance, not all making of the art works is well documented on the video recording, and moments of tension and debate between participants appear on the video but are not reflected in the evaluation forms. Nonetheless, taking them together provides a fuller picture of the workshops. What binds them, significantly, is the experiential layer. This is at least partially absent in our recorded data. Following Law's (2004) work on methods assemblages, we argue that this absence is productive, rather than problematic. Law refers to his own work on alcoholic liver disease, where 'there was something important about the scene [an Alcohol Advice Centre] that could not be put into words and escaped the possibilities of language ' (2004: 87). This resistance to explicit symbolisation is part of the methods assemblage: the process of crafting and enacting boundaries in research between what is present, what is manifestly absent, and what is 'Other' in the sense of being necessary to presence, yet hidden, repressed or unrepresentable for some reason (2004: 161). Below, we use an example of a key theme from the workshops, that of transformation, to show this process at work in our 'data assemblage'.
In the first workshop, an artist led the group through a 'body scan meditation' (Kabat-Zinn, 2005) which participants used to created images of their own embodied experiences (these were not necessarily of pain, as not all participants experienced pain, or felt it during the body scan). A transformation occurred between what was experienced within the body during the 20 minute guided meditation-influenced by the sound of the recorded voice leading the exercise, the space, and the presence of others-and the art materials through which an image of that experience was made. Unlike other workshops, where transformations from one form to another often happened in smaller groups and/ or off camera, most members of the group were present in the room when this was occurring. There was some interaction between people in terms of sharing art materials, but for the most part the transformation took place individually. One participant produced a drawing of a red eye, filling the page. Here is a reconstruction of her drawing process on the basis of the video: The participant begins by selecting a red chalk pastel, putting it to the paper and drawing repeated clockwise circles which get gradually wider. She then selects a darker pastel and goes over the circles in the opposite direction. A third colour is chosen. Beginning with the tip of the pastel on the edge of the circle, more anti-clockwise circles are drawn out from the edge. She continues to go over the circles, changing pastels. She then uses her left hand to smudge the pastel in larger clockwise circles. The circles grow more blurred. She switches hands, smudging with her right hand around the edges of the circle. She returns to draw a darker circle overtop of the central circle, then trades for a more orange pastel and does the same, deepening the colour she had smudged. She puts her left hand in the centre of the drawing, lifting it up to examine it. It leaves a handprint in the centre of the image. She experiments with this, putting both hands on the sides of the circle she has drawn, leaving more handprints. She returns to the process of alternately smudging and darkening the image. She pauses to speak to another participant about the shape he is sculpting, then returns to her picture and smudges the edges, smoothing the joins between colours.
Thoughtfully, she draws two lines down from the centre of the image with her fingers. She begins to sweep outward at the edges of the circle, toward the right edge of the paper. Her fingers pause and return to the centre of the circle, tracing lines down to the bottom once more. She stands back, watching others again, then returns to tracing downward and smudging the edges. She watches the group, then looks down at her hands, seemingly uncertain whether she is finished. She returns to tracing lines from the middle of the circle several times. 'Is that actually part of the art?' another participant jokes, seeing her chalky hands. 'Yeah,' she says, 'I'll just stand with it'. Present in this data is a sense of time and process: we see how the participant creates the shape of an eye, but that it did not become an eye until the end of the drawing. The creation of the drawing takes about 15 minutes, and involves a good deal of standing back and thinking. We understand that she has been considering the drawing carefully; that she is not entirely certain how to proceed. We see that her experience of the drawing is tactile as well as visual; there is also a brief discussion with another participant about the sound of the materials. However, there are also telling absences. In the discussion that follows, the participant does not volunteer any information or interpretation of her drawing. In a later written reflection, she said she had felt threatened when a workshop leader asked a general question about interpreting the pieces that had been made. She wrote that the choice not to explain the work was deliberate, and she also felt that it should be able to stand on its own. The process of creating it was an intimate one for her, and a space in which further research intrusion was unwelcome. Her refusal to interpret tells us a great deal about the role she ascribes to the artwork, and what she understands it to communicate.
Yet what is effaced in this description is equally important in making our data assemblage coherent: that is the liveness and the imprographic element we described in the previous sections, which helps to make sense of the experience of making this work as transformation. Liveness escapes documentation because it must: the improvisational elements of uncertainty and displacement experienced at the time of the workshop are resolved in the recorded data, reviewed at a later time.
This Othering is both necessary, and more than incidental. The workshop space was productive because it set to one side the identities with which all participants entered the workshop: these were not irrelevant, but they were displaced. In a discussion at the end of the final workshop, one participant explained I feel like I've finished up a completely different person from the work, because I wouldn't [write words on my drawing] now, I'm sure. I wouldn't be able to say the words or be able to write the words now. I feel like that part of my brain's been taken out, which is, I need a rest from that word, brain thing.
On evaluation forms, one participant commented that 'It was the most amazing day I've had for such a long time. Being with other people that have chronic pain, but also able to be creative. I completed [a hospital pain management course]. Great team, but I found the group's dynamics in a hospital setting very difficult' (Body Mapping workshop); while another wrote that it was valuable 'to share knowledge between such a varied group of people coming from different backgrounds' (Imaging and Imagining workshop). As researchers, too, our roles shifted over the course of the workshops, as we went from discomfort and uncertainty to feeling comfortable with our involvement as participant-facilitators. These shifts were central to what the workshops accomplished in exploring new, communal ways of communicating about pain. This 'Other' data, which is not documented in the video recordings and only glimpsed in fieldnotes and evaluations, is nonetheless critical for understanding the workshops. For instance, a central defining interaction in the Body Mapping workshop occurred off-camera between a participant with pain and a carer whose partner had pain. After a series of images of a pain 'object' (a hard glass ball surrounded by spikes) being unreachable or blocked, they considered how to transform the image. 'I want to take the pain from you,' offered the carer, while the person with pain replied, 'I don't want to give it to you'. They enacted a series of images of struggles over the pain, finally coming to an image of holding the object together, the carer's hands supporting the hands of the person with pain. 3 While we have traces of this interaction through the photographs, fieldnotes and a brief group discussion, the interaction itself is merely indicated rather than captured by traditional 'data points'.

Liveness, thereness and reuse
The liveness of the data and the importance of context, process and 'being there' to what the workshops produced, creates challenges for data interpretation, particularly for documentation and reuse. These problems are not unique to working with arts-based methods. Ethnography is also process oriented and has long wrestled with the issue of 'being there' (Borneman and Hammoudi, 2009;Davis and Konner, 2011;Watson, 1999). Yet most ethnographic texts say curiously little about the 'thereness' of being there: what is it that makes it different than a good quality video recording of 'there'? In what way do ethnographic fieldnotes and other documents from the field reproduce or fail to reproduce 'thereness'?
The field of performance studies contains more insights via this issue of live and recorded performance. Phelan's statement that 'performance cannot be saved, recorded, documented, or otherwise participate in the circulation of representations of representations: once it does so, it becomes something other than performance '(1993: 146) is widely quoted. To 'become itself through disappearance' is in effect what we are discussing here with liveness. Yet Auslander (2008) has challenged the view that live performance is ontologically different: rather, he argues, the value of liveness itself has come about only recently and only in relation to mediatisation: 'the very concept of live performance presupposes that of reproduction-[…] the live can only exist within an economy of reproduction ' (p.57). Any distinction between the phenomenological experience of live performance and recordings needs therefore to carefully consider how the relationship between the two is articulated. Auslander also acknowledges that the experienced differences between live and recorded performances may be greater in improvised and smaller scale works, criteria which both apply to our arts workshops, and other live methods.
Secondary qualitative analysis has also attended to the differences between presence and recorded or transcribed data. Arguing against reanalysis of qualitative data, Mauthner et al. (1998) write 'the differentiation of […] "background" data from interview and observation data is a false distinction. … data cannot be treated as discrete entities'. (p.742). However, Moore (2007) counters that the reflexive and contextual data gathered by 'being there' are useful for one kind of analysis, while transcribed data are useful for another. She questions the very notion of 'pre-existing' data, suggesting 'that the data in fact are not "out there" at all, that the data are here and now, being constructed in the process of a new research project' (3.5). 'Data' in her formulation are flexible, construed by the researcher for particular purposes in particular contexts. They are part of what Law (2004) would call the 'methods assemblage'.
Similarly, James concludes that secondary analysis is 'eminently do-able, with the "problem" of context being more-or less-troublesome depending on for what purposes the secondary data analysis is taking place ' (2013: 567). For her, the distinguishing element is what she calls the analytic imagination, which can be used productively to make interpretations of data after the fact. However, the 'problem of context' can also be more or less troublesome depending on the purposes of the primary data collection and analysis. Our analysis here suggests that it may indeed be more problematic in arts-based research for some members of the research team to be absent from data collection or production. This is not to say that portions of our data could not be used for another analysis, but that non-textual data of this kind are more difficult to interpret without written context (Banks, 2001) and that this new analysis would need to be treated as another performative intervention, a re-slicing of the data for another purpose or aim.

Conclusion: live methods, live data
Live methods are not simply new data collection tools, which produce objects for analysis akin to interviews or focus groups. Rather, an ontology and epistemology of liveness (i.e. theorising the social world as one in flux, constituted in our methods) informs the implementation of a set of methods (e.g. arts workshops), which in turn produce live (imprographic/improvisational) data. Live methods thus culminate in live data.
Performative approaches to arts-based research require us to rethink what constitutes our data: to acknowledge that method, as a process, as well as output-what is produced through the method-are both part of the data assemblage. This is consistent with Moore's (2007) suggestion of seeing data as constructed through the research process, but goes further, in that treating method as performative means we must understand it as an inseparable part of data during and through our analysis; not merely as a condition of data production but as a central and defining element in the data assemblage.
As we have argued, liveness and its unrepeatability are central to the work our workshops did and the productive and communal space they produced. This unrepeatability means we must take into account a wider range of things that are present, absent, and Othered, and bring these into the analytic process. It also means that the experience of 'being there' is critical to the process of interpretation. Encounters and interpretations are not arbitrary, but are real instances of social life conceptualized as ambiguous, contingent and unfolding (e.g. Lury and Wakeford, 2012). Live methods thus produce live, or 'lively' data (Savage, 2013). These data are 'lively' in the sense that they are contextdependent, improvisational and one-off. If we are lucky, they are also 'lively' in the sense that they evoke a new version of the world, are provocative and stimulating.
The concern with liveness and experiential dimensions of work connects with Flyvberg's (2001) argument that social scientists develop through exposure to multiple complex cases, a fluid expertise that is not reducible to specific inputs. It also reflects Ingold's (2013) position that the valuable conclusions of a piece of research result from the researcher's engagement in a process of learning, drawing on multiple sources of data and a learning process that results from engagement, but which is not fully predictable or 'there' in the data before the research process. There are also overlaps with other arts-based approaches, in particular performance ethnography, which strive to make work 'live' and promote social change through enacting new roles and making work accessible to non-academic audiences.
Live methods open space for further methodological development, since performative approaches often reject overly formalized methodological procedures. Alternative approaches exist for making methodologies concrete in ways that are consonant with an interest in preserving the liveness of data, for example Maclure's search for moments that 'glow' (2013), Gillespie and Cornish (2014)'s 'sensitising questions' or Revsbaek and Tanggaard's 'analyzing in the present' (2015). Further developments along such lines would help advance the teaching, learning and critical discussion of live methods.
Finally, there are implications for the forms taken by research using live methods, which should communicate aspects of process, uncertainty and indeterminacy. If the significance of the research arises through the whole research assemblage, it becomes especially important to document that full assemblage in the writeup. Powerful live methods articles would seek to bring their phenomenon to life: to convey the immediacy of being there, the contingency of the researchers' interpretations, and allow the reader to engage in a live interpretative interaction with the material. Much qualitative research, especially in arts-based approaches, aspires to such characteristics already. Leavy (2015: 285) suggests that we do our best work when 'we accept and indeed embrace […] messiness.' Further methodological development under the rubric of 'live methods' has much to contribute to the role of qualitative research in provoking novel, lively versions of social reality.
She is currently writing a book on Sensory Methods in Social Research and co-editing a book on chronic pain communication with Elena Gonzalez-Polledo, to be published by Palgrave McMillan in 2017.
Elena Gonzalez-Polledo is a Lecturer in the Faculty of Social Sciences at the University of Sheffield.
Flora Cornish is an Associate Professor in the Department of Methodology at the London School of Economics and Political Science, where she teaches qualitative research methods and researches community mobilisation in international health and the uses of evidence in informing health policy and practice. She was a Co-Investigator of the NCRM/ESRC funded Methodological Innovations project Communicating Chronic Pain: Interdisciplinary Strategies for Non-Textual Data (www. communicatingchronicpain.org).