The world café is an unmethod within co-produced research

ABSTRACT

The world café (WC) has gained popularity as a participatory method for collecting qualitative data. In this article, we present an instrumental case study -the Moving Social Work cafés- to illuminate why and how the WC might be applied within co-produced research. Our principal argument is that the WC constitutes a coherent and effective means for living up to the principles and values of co-production, as long as it is not treated as a method. We also contend that the WC should be approached as an unmethod, and we explain this approach through the metaphors of jazz and contact improvisation. Broadly, the article encourages qualitative researchers to resist the temptation to methodologize processes that have more potential when they remain constitutionally immature.

KEYWORDS:

• Anti-methodological rhetoric
• co-production
• conversational knowledge
• online methods
• participatory methods
• physical activity

Introduction

June 1995, Homestead Boulevard, Mill Valley, California. Juanita Brown looks out the window at the large patio, worried. In no time, David Isaacs and herself will be hosting a strategic dialogue among twenty-four participants. The plan was to conduct the strategic dialogue on the patio, but a problem arises. It pours and no one can go outside. David looks at Juanita’s worried eyes.

Why don’t we set up our TV tables in the living room and just have people get their coffee and visit around the tables while we’re waiting for everyone to arrive? – he suggests. We’ll then put away the tables and begin with our normal dialogue circle (Brown, Isaacs, and Brown, Isaacs, and The World Café Community 2005, 14)

With little time to ponder, they set out the TV tables in the living room. Juanita breathes deeply. Then the first participant arrives, and proclaims: These look like café tables, and café tables need some tablecloths! She puts white sheets of easel paper over each of the paired TV tables, and adds crayons ‘just like those in many neighbourhood cafés’ to each table (p. 14). Juanita brings flowers to make the atmosphere more hospitable; and the icing of the cake: a sign for the front door – WELCOME TO THE HOMESTEAD CAFÉ.

Newcomers find the setting amusing. As they get their coffee and croissants, they gather in informal groups around the tables and begin to talk around a general question they were asked in a previous strategic dialogue. Participants talk and talk and talk and scribble on the tablecloths. The room is alive; it wouldn’t feel right to interrupt the conversations to start the strategic dialogue circle, Juanita and David think; so they forget the intended circle and encourage people to keep conversing. After 45 minutes, a participant has an idea:

I’d love to have a feel for what’s happening in the other conversations in the room. Why don’t we leave one host at the table and have our other members travel to different tables, carrying the seed ideas from our conversation and connecting and linking with the threads that are being woven at other tables?

(p. 15)

In agreement, people start moving around the room, leaving one host at each table. It works; everyone is fully involved in conversations. Another hour goes by. Then another participant proposes: ‘Why don’t we experiment by leaving a new host at the table, with the others traveling, continuing to share and link what we’re discovering?’ (p. 15). With people carrying their ideas across different tables, threads of topical insight become connected and woven. Time flies: David and Juanita realise it’s almost lunchtime. David calls a close to the conversations and asks the participants to gather around a large piece of rolled-out mural paper that lays on the living room floor. Each table group are invited to place their own tablecloth on the edges of the paper and then tour the tablecloths to notice patterns and insights.

As the collective discoveries unfold, Juanita and David form a question in their heads: ‘What happened here that enabled such great conversation and breakthrough thinking around critical strategic issues?’ (The World Café Community Foundation, 2023). The very next day, they meet with one participant to dissect the event and try to answer that question. Out of this discussion, a term is proposed to capture the possibilities that developed in the improvised ‘Homestead Café’. That term is the World Café (WC).

In this coming-into-existence story (originally told by Isaacs, in Brown, Isaacs, and The World Café Community 2005), the WC was casted as a serendipitous achievement or, as Brinkmann (2020a) might name it, a ‘gift of chance’. Later, when Brown, Isaacs and The World Café Community begun disassembling the story and detaching the WC from it, the WC turned into a ‘designed conversational process’ (2005, p. 5). Then, the basic WC process encompassed five elements:

• An informal environment similar to a café

• A warm welcome and an event overview from the host

• Three or more 20 min rounds of conversation between small groups of four or five participants

• A key question underpinning each round, which may build upon one another in subsequent rounds

• A final round to share the results of conversations with the wider group.

Beyond this structure, and seven general principles¹, Brown, Isaacs and The World Café Community put emphasis on creativity: ‘by using your imagination, you can improvise and design a World Café dialogue process, with or without café tables, that reflects your unique situation’ (p. 172). Given the flexible nature and apparent simplicity of the format, the WC was soon embraced by a multitude of groups in diverse non-academic contexts including multinational corporations, small non-profits, community-based organizations, government offices, and educational institutions (The World Café Community Foundation, 2023).

According to Löhr, Weinhardt, and Sieber (2020), the WC arrived in the field of qualitative inquiry at the hands of social work scholars Fouché and Light (2010). These authors themselves argued that ‘no direct application of the World Café in the field of qualitative research has been discovered’ prior to their contribution (p. 36). Their contribution was twofold: they documented the design, implementation, and perceived value of a WC approach; and they initiated a dialogue for the implementation of the WC in social work research and the broader field of qualitative research. Soon thereafter, academics working in fields such as nursing and business started to incorporate the WC into their projects (e.g., Broom et al. 2013; Chang and Chen 2015), transforming its very nature. Once the WC entered the field of qualitative research, it evolved into an academic data collection method. Subsequently, academics begun to examine its strengths and weaknesses as a method, providing some comparisons with other research methods. Löhr, Weinhardt, and Sieber (2020), for example, compared it to semi-structured interviews and focus groups, arguing that the WC is more conversational, can help increase the size of the reference sample, the range and scope of individual views on a certain topic, and the level of participation. Further, Schiele et al. (2022) developed an extensive ‘method comparison’ between WC, expert interviews, Delphi, consortium benchmarking, and focus groups. According to these authors, advantages of the WC over other well-established methods in qualitative research include speed (they argued that the WC is an ‘accelerating form of data collection method’), iteration in a one-time session (it avoids the shortcoming of requiring multiple sessions), and attractiveness (opportunities for learning and exchange ideas with peers makes this method attractive for participants).

A small group of publications focussing on the WC have also paid attention to methodological coherence. As Poucher et al. (2020, 165) explained, ‘for a study to be methodologically coherent, researchers must be aware of the philosophical assumptions underpinning their research and select the most appropriate methods to achieve their intended research aims’. While there is some theoretical flexibility in the application/use of the WC, this is often aligned with participatory theories and, hence, is generally known as a participatory method (Löhr, Weinhardt, and Sieber 2020). Pettican et al. (2021), to name an example, argued that the WC ‘is a method concerned with enabling the collaborative construction of knowledge and therefore aligned with PAR [participatory action research]’ (p. 163). Closely related but different to PAR is co-production: a participatory research approach that, although not new, has gained much popularity over the last years (Goodwin 2019). Herbison et al. (2023) used the WC within a type of co-production called Integrated Knowledge Translation (IKT), whereas Pettican et al. (2023) put the WC to work in a project underpinned by Equitable and Experientially-informed co-production (EEiC). Smith et al. (2023) established the core differences between IKT and EEiC and sustained that EEiC is the co-production type ‘most centrally concerned with answering the calls for a participatory turn in research’ (p. 167). These authors then suggested that designing, testing, and leading WCs is a way of putting EEiC principles in practice. However, they did not go further to describe the ways in which the WC and EEiC are arguably the most methodologically coherent fit, or how this fit might look like.

The intention of this article is not to determine if the WC works better within EEiC than within IKT -although readers could indeed form an opinion after reading the article and Smith’s and McGannon (2023). Instead, the paper focuses on exploring several questions revolving around the possibilities and challenges of the combination of EEiC and the WC in action. These include: Why co-produce through using the WC? How can EEiC help us host good WCs? How can the recently developed methodological literature on the WC help us do genuine EEiC? Is such literature placing too much emphasis on the methodological aspects of the WC? What are the consequences of framing the WC as a data collection tool or method? Should we strive to master the tool, or should we back up to see things otherwise? Should we perhaps take back the WC to 1995?

Those who consider using the WC within co-production should not proceed without contemplating the above questions. Ignoring them would mean that we are operating on the automatic, that we have no interest in understanding our practice or critically thinking through past practices, especially in light of new research ideas. That is not how researchers should operate. Accordingly, the purpose of the article is to examine the meaning and significance of the questions, to answer them, and to connect them with the specialised literature on the use and abuse of qualitative methods.

To a great extent, this is a theoretical paper. The focus is placed on reflecting about the nature of the WC, and how this nature depends on the placing of WC within EEiC. Still, the questions that interest us stemmed from our empirical experience using online WCs as part of a co-produced project called Moving Social Work (MSW). We thus need here to share information about MSW and report on our WC experience before introducing our conceptual contribution and the practical implications that follow. Having clarified this, the article will be organised as follows. We start with a brief conceptualisation of EEiC. Next, some essential information about the project and the role of the WC within it is presented. The reasons why we decided to choose the WC will be subsequently listed. To provide further important contextual information, we then provide some detail about the design of the MSW WCs. We contextualise our decision to bring the WCs to an online setting, and clarify how the stage for the WC events was set. The following section will account how these events developed, and what we learned about the WC and EEiC throughout the process. Against this backdrop, we close by advancing our theoretical reflections and how these might apply in practice. We hope these reflections will guide both thinking and use of the WC in participatory research, including but not limited to EEiC.

Equitable and experientially-informed co-production

EEiC is a collaborative process in which diverse academics and non-academics work together from the start to the end to generate impactful research. Underpinning EEiC is democratic egalitarianism, a political philosophy aiming to favour different forms of equality. EEiC is especially concerned with ‘relational equality’: the idea that people should relate to each other as equals and enjoy the same fundamental social status (Arneson, 2013). Likewise, EEiC subscribes to the principle of equality of intelligence, which implies that the different forms of knowledge of all those collaborating on the research is equally valuable and important (Dierckx et al. 2021). People collaborating on the research might include academics, funders, practitioners, and policymakers. More importantly, collaborators must include people and communities whose knowledge results from lived experience rather than from formal education or professional training. As Smith and McGannon (2023) stressed,

The point is that this type of co-production attempts to address issues of equality, diversity, and inclusion partly through ensuring that those who have traditionally been excluded and/or marginalised are essential partners in the co-production process and their relevant lived experience meaningfully influences the research.

(p.165)

To ensure that collaborators – researchers and people with lived experience – are equal partners with respect to influence and decisions, EEiC emphasises on acknowledging and addressing power relations between them (Farr et al. 2021). Of course, a society without relations of power is a utopian fantasy. Addressing power relations, therefore, does not mean dissolving them, but rather allowing them ‘to be played with a minimum of domination’ (Foucault, 1991, p. 18). The work against power asymmetry is twofold. It is necessary to empower marginalised actors and, simultaneously, the ‘egos’ of those who benefit from conventional structures of power-knowledge has to be managed (Turnhout et al., 2020; Vincent et al. 2020). In our resource to guide EEiC (Smith and McGannon 2023), we suggest some strategies for how this might be done in action. Practical tips to help with other dimensions of co-production (e.g., accessibility) along with criteria for judging or evaluating the quality of it are also proposed.

While the driver for EEiC is to address issues of equality, diversity, and power, a parallel motivation is as well to create technocratic value, that is, to produce useful knowledge that can be applied into policy and practice (Smith and McGannon 2023). Indeed, EEiC projects are often born out of a desire for changing or benefiting culture, policy, the environment, health, wellbeing, or services beyond academia (Smith and McGannon, 2023). That was the case of Moving Social Work (MSW), the project we introduce next. Alongside Pettican et al.’s (2023), this is one of the first projects within our field to fully commit to the principles of EEiC.

Moving social work and the role of the WC approach within the project

MSW was designed to help social workers gain the knowledge, confidence, and skills to promote physical activity (PA) among disabled people. Overall information about the project context, rationale and structure can be consulted in. Here, suffice to say that the WC was utilised late but extensively in the first stage of the project, the goal of which was to generate a training programme prototype aimed at social workers. In this stage, two prototype versions were developed. The earliest one was informed by academic and expert knowledge, which was gathered in several studies (Monforte, Smith, and Smith 2022; Monforte et al. 2022). The MSW co-production collective played an active role in developing these studies, with some members being co-authors. The initial prototype version was co-produced but not fully co-produced yet; the evidence used to produce it was developed together with people with lived experience, but it did not contain the views and insights of people with lived experience, such as social work students and disabled people. To address this gap, and build the second version of the prototype, we considered options and decided on using the WC.

Why the WC? A co-produced rationale

The MSW co-production collective preferred the WC over other approaches on the basis of three reasons. First, the WC had been used before to address the kind of issues that traverse MSW, including social work (Fouché and Light, 2011), higher education (Estacio and Karic 2016), health promotion (Recchia et al. 2022), PA (Pettican et al. 2021), and disability issues (Bumble and Carter 2020, 2021). Generally, this literature casts the WC as a useful and stimulating method in projects involving people with lived experience. Second, the WC principles were similar to, and compatible with, the principles of EEiC (Smith and McGannon 2023). In short, EEiC is about encouraging everyone’s contributions as well as valuing and blending diverse knowledges, which is what the WC aspires to achieve. The last key reason for choosing the WC over other strategies was the possibility to bring together large, independent groups of participants over time. Time is central in the process of iterative prototyping through which various layers of experiential knowledge are turned into practical, tangible resources. The idea was to organise different cafés in which multiple individuals with lived experience would be brought together in order to know what they like and dislike about the prototype, think should be retained and excluded, and be added and modified. As such, the WC itself could be framed as an iterative cycle of training design, moving between people’s responses and changes to the drafts.

The online WC

As we began to schedule the cafés, a new wave of COVID-19 was afoot. Given that disabled people are at greater risk of contracting this virus, but also of developing severe health conditions and dying from it, we moved the world cafés online. At first glance, the prospect of having to go digital seemed to compromise one of the key principles of the WC: The creation of a hospitable space for conversation, similar to a café. However, as Estacio and Karic (2016) reminded, generating a welcoming, hospitable space extends beyond the physical environment to the socio-relational dimensions of the environment. Since no comparative analysis of in-person versus online WCs exists, the latter should not be seen as certainly inferior. Thus far, only three online WC based studies are published (Albrecht et al. 2022; Banfield, Gulliver, and Morse 2022; McKimm et al. 2020). These sustain that online WCs are suitable for participatory engagement as long as some actions are taken, for example: ensuring that future participants can access the internet; offering instruction and training for using digital devices and virtual communication platforms; preparing for fluctuating numbers of participants due to possible technical interruptions that may arise during the event; and giving the choice of using audio, video or chats, based on participants’ internet connectivity and comfort level. Consistent with the available studies, Zoom™ was the videocall conferencing platform of choice.

Setting the stage for the WC events

Two questions mattered for us: how could we improve the training content? And what could be done to overcome potential barriers to the programme success? To stimulate discussion around these questions, but also to enable a direct link between the participants’ ideas and the programme prototype, we used prompts. Prompts allow a high level of participant-led involvement (Kwasnicka et al. 2015), but almost no studies using the WC to explore disability issues employed prompts (Bumble and Carter 2021). Together with the MSW collective, we produced two: an easy-read summary of the training programme prototype, and an infographic showing the potential barriers to programme implementation that were identified in our previous research (reference redacted for peer review). In the event of having in-person cafés, we would print the prompts; for online cafés, we would use the sharing screen function on Zoom to display them. In either case, the prompts would be sent a few days in advance via email, so that participants could familiarise themselves with them. Other documents would be attached in the same email: an information sheet; an invitation to contact the host in case of need of clarifications, specific adaptations, or Zoom training; and an informed consent. All these documents were part of the ethics proposal approved by Durham University (SPORT-2020-0218T17_18_37-dmgf98).

In the ethics proposal we have just mentioned, we provided prescriptive information on how the data collection process would unfold. We attended this (essentially bureaucratic requirement) by developing a protocol based on the WC processes that have been reported in the qualitative literature since 2010. Although every process is unique, a standardized version of the WC is being promoted via the introduction of big-tent quality standards (e.g., Bumble and Carter 2020) and efforts to have it recognised as a proper data collection tool for scientific research (Schiele et al. 2022). Of course, this phenomenon is not limited to the WC; as Brinkmann (2015) noted, qualitative psychology as a whole is becoming increasingly standardized.

Genuine EEiC is incompatible with standardization as collaborative work is often complex and unpredictable. For example, people with lived experience set the research agenda, establish priorities, and direct what and how conversations unfold. When we sent the protocol to the MSW collective members and partners from Disability Rights UK, they suggested several changes. For example, they added another key question (What would it take to ensure every disabled person has the opportunity to be physically active?), to guide an additional (i.e., fourth) conversation round. It was also resolved that the protocol would be better tried out in practice, through a pilot WC. In this pilot café, the members of the MSW collective could experience the process as participants and give feedback from that perspective. Although the main goal of the pilot was to deconstruct the protocol, or to turn it into a co-produced one, the MSW collective members shared helpful suggestions to improve the programme prototype. As they were people with lived experience themselves, it was important to include their suggestions in the process. This meant that their insights were used to create a newer version of the prototype, that is, of the prompts that would be used in the upcoming café.

The pilot experience had further consequences. First, the MSW collective members registered themselves as co-hosts of the succeeding cafés (this role was deemed more attractive than hosting, as it allows people to interact with peers and experience the conversations, while the host is more concerned about functional issues). Second, they recruited a few people with lived experience that, under their judgement, would enjoy participating in the cafés and would make important contributions from underrepresented perspectives. At that time, a number of participants had been already recruited via the organisations’ networks (e.g., Disability Rights UK, Sport For Confidence) and individual partners (e.g., a social work lecturer). Having recruited 87 people, and having considered their availability, it was decided 6 WC events should be hosted. To close the circle and evaluate the resulting changes in the training programme prototype, as well as changes in the way the WC was hosted throughout, it was decided that the sixth café would involve, once again, the members of the MSW co-production collective.

The WC events in action

Counting the pilot, the WC events took place between 02.12.2021 and 01.02.2022. The 87 people that participated in the cafés received a £50 voucher for their contribution. Nine different co-hosts were involved throughout, and Jake adopted the hosting role. Each café lasted around 2.5 hours. Table 1 displays the WC events in further detail.

Table 1. The Moving Social Work Café events.

Café Event	Space	No. of participants	Participating member roles and lived experiences	Host	Co-hosts
Pilot (02.12.2021) 2 hours	Virtual	5 (MSW co-production collective)	Active Partnership staff member Disabled people staff memberStaff from disabled people’s organisation (DPO)	Jake	Javier Staff member from DPO
1 (14.12.2021) 2.5 hours	In-Person	9	Student social workers Social work educator	Jake	Javier Active Partnership staff member Social work lecturer
2 (10.02.2022) 2.5 hours	Virtual	14	Social workers Disabled people* Social worker/carer Staff from charity organisation * BSL interpreters hired to support participating member to attend virtual café. Interpreters not included in total number of participating members	Jake	Disabled person Javier Staff member from DPO/disabled person
3 (10.01.2022) 2.5 hours	Virtual	8	Social workers Social care worker Disabled people Disabled person/Volunteer from charity organisation	Jake	Disabled person Javier Staff member from DPO
4 (20.01.2022) 2.5 hours	Virtual	23	Social workers Disabled people Carers* Social care workers Social work student Social work educator Staff from DPO * Carer of Disabled person participating in this café was present throughout. They have been included in the number of participating members due to their own insights provided.	Jake	Disabled person Staff of social enterprise organisation Staff member from DPO/Disabled person Staff member from DPO Javier
5 (25.01.2022) 2.5 hours	Virtual	20	Social workers Disabled people* Social work educator Social care workers Staff from DPO Staff member from Sport England Staff from DPO Carer *Partner of participating member joined the café towards the end but did not interact. They were not included in the total number of participating members	Jake and Javier	Disabled person Jake Staff member from DPO/Disabled person Staff member from DPO
6 (01.02.2022) 2 hours	Virtual	8 5 (MSW co-production collective)	Disabled people Staff from DPO Social care workers Social work educator Active Partnership staff member	Jake Staff member from DPO	Javier

Only the first café occurred in-person. We tried to make a social work university classroom look like a café, but the masking, social distancing, and hand sanitisers on the tables did create an uncanny atmosphere. Participants, though, reassured us: it was a stimulating atmosphere, nevertheless. Paper tablecloths were used for annotation, and the participants preferred the co-hosts to be the ones taking notes. Many, many notes were taken.

The insights generated in this café were blended with the insights from the pilot café, bringing about a newer iteration of the prompts that were then used in the next, virtual café. As we expected, shifting to a virtual setting presented logistic challenges, although these were solved rapidly and thus did not compromise anyone’s participation. British Sign Language (BSL) interpreters were arranged for people with hearing impairments whenever required, and accessible versions of the prompts, information sheet, and informed consent were made available. Importantly, time was afforded for individuals who wished to participate who were unaccustomed with using videoconferencing software like Zoom. Days prior to each WC, Jake spent time with individuals to offer guidance, ensuring that each participating member had the opportunity to fully participate in the cafés.

Throughout the cafés, the prompts did several things for us. For example, they acted as reminders of the table questions and enabled the formation of very concrete ideas: ‘what appears in the screen now should be said earlier’ or ‘I would take this out’ or ‘when you talk about this, make sure to mention that’ or ‘what if we rephrase that sentence?’ Moreover, the prompts made it easier for the co-hosts to bring up the points made by former WC participants who instigated changes in the content showed to new participants. The MSW programme summary became a materialisation of multiple blended perspectives, of conversational patterns and unique contributions of people with lived experience. These will be reported elsewhere [reference redacted for peer review]. Without going into the detail of content, it is worth highlighting the quantity and quality of key messages that were collected throughout. At the halfway of the café series, a co-host commented: ‘Everybody has said something unique; we have at least one original point per participant. Being 87 participants involved, the final impact of the process on the programme will be incredible!’. One thing that let us to ensure this impact was achieved was giving enough time between events to refine things carefully. Time between events also gave us time to reflect on what happened, and how to improve practice.

For us, practice improved as we problematised protocol-driven actions. If given too much weight, the protocol could become a red herring, a distraction that ‘disturb[s] people who are at work’ (Mills 1959, p. 27). Our work was to create the conditions for participants to be listened to. In our head there was a goal forged in fire: equal participation. As Smith and McGannon (2023) suggested, we bore in mind that some people will not necessarily consider what they have to contribute to be particularly valuable, while others may intentionally or unintentionally dominate conversations. We then intervened on a few occasions to ensure all partners had the chance to fully express their thoughts and feelings in their way. Otherwise, we became modest witnesses of how the participants formed the cafés according to their own specifications. The so-called move from ‘ordinary conversations’ towards ‘conversations that matter’ was on their hands. One thing happened when this shift was produced in a (virtual) table: conversation would lead the table group to a state of social flow whereby people are absorbed together in a conversation to the point of losing the sense of time (Walker 2010). All the co-hosts reported being immersed in the flow at some point as well, and even to momentarily forget their role.

In time, we moved on from the protocol. The protocol became a ‘burden’ for doing effective, organic WCs. It was also inconsistent with the logic and practice of genuine EEiC. As Monforte and Úbeda-Colomer (2021) put it, ‘it is not about bringing life to a previewed process, but to invent the process in the doing; to improvise it’. The improvisers, however, were not the researchers only; in our increasingly horizontal process, they were the participants as well. To illustrate, during a coffee break we had between a second and the third round of conversation, the participants talked to each other and decided (without asking for permission to the host or co-hosts) to make the groups smaller by adding a new table (online breakout room). This was EEiC in action: participants are allowed to ‘be as disobedient as possible to the protocol, and to be as capable to raise their own questions in their own terms’ (Latour 2000, 116). Allowing this requires from hosts and co-hosts to suspend the expectations of what ought to be, and to be receptive, contemplative; to let things happen. In other words, hosts have to adopt a non-instrumental attitude that Brinkmann (2021) called patiency. We believe that commitment to EEiC helps patiency blossom, and that patiency is a helpful attitude in equitable and experientially-informed cafés.

The last WC with MSW collective participants was very different from the pilot café. The MSW programme summary was visible changed, and the new insights from the MSW members were affected by the conversations they got involved into as co-hosts. A newcomer to the MSW collective gave her insights, but also asked curiosity-driven questions about the previous cafés. It was the non-researcher members who answered her questions, showing a feeling of ownership about the work done. This sense of ownership was further shown in other contexts. For instance, some members took part in an interview as part of an independent evaluation process by an external organization experienced in co-production, to determine if MSW was adopting tokenistic practices or worthy of being recognised for good co-production practice (which we achieved and were awarded the ‘The Dialogue and Change Award https://arc-nenc.nihr.ac.uk/pice/the-dialogue-and-change-award/). During the interview, we were told, members spoke about the WC as way of practicing co-production; for them, the cafés epitomised the values and principles of EEiC, and helped them better understand what it really means to co-produce. It helped us (the research team) too, but we were not done thinking about the WC.

Reflecting about the nature of the WC: the unmethod

‘Think. Reconsider. Undo’ (Morse 1999, 717)

At the beginning of the MSW project, some of us set a ‘research sanctuary’ devoted to discussing our theoretical concerns (Spiller et al. 2015). After the WC, we cleared our diaries and met regularly to have exciting conversations over coffee. In such conversations, we ‘went meta’, as Jorgenson and Steier (2013) put it, and thought about the WC as a whole instead of limiting reflection about the MSW café events. One question we kept asking was: Is ‘method’ the optimal term to describe the WC, or is it counterproductive when the WC is located in EEiC? We sensed where these questions were coming from, but there was something blurry, something we wanted to understand with more depth. Driven by curiosity, we read and discussed and were left with a clear conclusion: the WC should not be considered a method, and especially when it is used within co-production approaches like EEiC. But what does this mean? How did we arrive to that ontological statement? If not a method, what then the WC is or can be? Or, as Brinkmann (2020b) already asked: ‘are there other ways of thinking about interviews and other qualitative methods than in the idiom of methods?’ (p. 452). Let us start from answering the last question and build from there. Our answer is: the WC is, or can be, an ‘unmethod’.

Un- as a prefix means non- or not-a-method, but it can also mean challenging or undoing ‘certain versions or accounts of method, of the normativity of method, that currently prevails within the social sciences’ (Law 2004, 5). What prevails is an image of the method as a specific reproducible procedure for data collection that researchers implement to develop ‘rigor’ and ‘validate’ their process (Hovey et al. 2022). The work of Schiele et al. (2022) is a paradigmatic example of how researchers, in the name of ‘rigor’, reform the WC to make it ‘fully applicable as academic data collection method’. This kind of intervention shows that, indeed, ‘participatory methods are in danger of being seen as a “fool-proof” technology that – when applied carefully and conscientiously – will enable research … to achieve ethical and epistemological validity’ (Gallacher and Gallagher, 2008, p.513). It is also a symptom of what Chamberlain (2000) called methodolatry: a concern with a ‘correct’ method and the privileging of method over theoretical matters and research relationships. Along this line, St. Pierre (2021) and others concur: there is now a ‘rage to methodologize anything and everything’. The unmethod calms such rage; it frees researchers from the will to method: the will to domesticate research into repeatable equations, which is contrary to the spirit of creativity underlying the WC.

Admittedly, this sounds too abstract. To enhance readability and advance our argument, it will be useful to use metaphors. The use of metaphors to facilitate the familiarisation and clarification of complex ideas goes back to the times of ancient Greece, and it has been granted epistemological legitimacy as a viable means for guiding qualitative reflection (Jensen 2006). As Morgan (1986) and Inkson (2002) noted, different metaphors can be used as complements to each other, allowing to shed light on a single but complicated concept. We hope also the metaphors provide a way of thinking with the practicalities of doing WCs and how we might go about doing WCs - as well as a way of showing the influence of Medical Humanities on our thinking. Accordingly, we introduce two different but related metaphors for the WC.

The first metaphor is the WC as Jazz. It is a borrowed metaphor; Oldfather and West (1994) used it to illuminate key qualities embedded in processes of qualitative inquiry, while Humphreys, Brown, and Hatch (2003) focused on the analogies between ethnography and jazz. The metaphor of the WC as jazz applies on multiple levels. From a surface level, we might say that jazz is at home in cafés where jam sessions take place. In these informal gatherings, relationships form organically and are non-hierarchical. A host can facilitate the event, but the idea is to make the improvisation collective. Very often, participants experience flow in improvisation (Forbes 2021), with scholars suggesting that ‘jazz musicians improvising in a jam session should report more social flow than the less interdependent members of a marching band’ (Walker 2010, 4). Each participant in the jam session adds their own voice to the mix and layer their interpretations with those of the others. As Humphreys, Brown, and Hatch (2003) highlighted, ‘the best jazz is a conversation involving the crossfertilization of ideas’. Such crossfertilization of ideas is, as Löhr, Weinhardt, and Sieber (2020) put it, the key strength of the WC. From a profounder level, the jazz metaphor enables exploration of both the tacit, deep structures that guide the WC process and the improvisatory qualities that allow co-researchers to make their way ‘without prescriptions of fully orchestrated scores’ (Oldfather and West 1994, 23). As jazz is guided by a deep structure of chord progressions and themes, WCs used within EEiC research are guided by onto-epistemological principles (e.g., democratic egalitarianism), socially constructed values (e.g., mutual respect) and inquiry focuses (e.g., privileging the interests of people with lived experience). The point is to improvise the practice of the WC while having the deep structure of EEiC in mind. We will repeat this important point later when it makes more sense. For now, let us introduce the second metaphor. That is the WC as contact improvisation. This metaphor might connect to Janesick (1994) and her idea of qualitative research as dance, but it is a more concrete one.

Contact improvisation is a form of dance aiming to ‘let the dance happen’ between two (or more) equal partners (Behnke 2003, 41). Partners are equal because, unlike other types of partner dance, nobody acts as ‘lead’ or ‘follow’. As Torrents et al. (2010) put it, ‘Dance relations are egalitarian’ and contact improvisation ‘is based on a social interaction founded on reciprocity’ (p. 55). In practice, contact improvisation involves ‘moves’ such as shoulder and hip lifts, head-to-head improvisation, rolling on the floor, or being surfed by the partner. These moves, however, do not arise from a preconceived model, but rather from the open-ended communication between partners in the midst of an improvisatory act. While improvised, body movement is guided by a deep structure, called ‘improvisation structure’, which creates a recognizable style of movements. This way, contact improvisation is often talked about as a ‘resource for movement generation’. We may think of the WC, analogously, as a resource for what Brinkmann (2020a) called ‘participant conversation’. Participant conversationalists, Brinkmann argued, do not follow procedural steps of research from designing to publishing as recommended by textbook. Instead, they allow themselves to become involved in the conversations, deciding together with them from one conversation to the next what ‘we’ should talk about and bring forth. In contrast to standard interviews ‘that are initiated and to a large extent controlled by the researcher’, wrote Brinkmann, these conversations in which each individual inventively participates in concert with the other are ‘more egalitarian and participatory’ (no page).

One key idea that both the jazz and the contact improvisation metaphors convey is that the conversations that happen in the WC cannot be planned in advance (e.g. via protocol) or the process of doing a WC controlled by the researcher. ‘When I start off, I don’t know what the punch line is going to be’, said Jazz musician Buster Williams. In the same way a contact improvisation dance cannot be planned in advance, in the same way a jam session cannot be pre-determined and carefully applied, the WC cannot be used by following a precise methodological repetition – often conveyed through a protocol. That is why the WC protocol we developed was irrelevant for our process. The music, the dance, the conversation, is not in the ability to follow a protocol. Rather, it is in the ability to fall into conversations, an ability that ‘comes through both understanding the deep structures and giving oneself the freedom to let go and apply those deep structures in improvisatory ways’ (Oldfather and West 1994, 23).

Fall into conversations? In a sense, the verb feels ill-chosen. After all, WCs are staged events with the explicit goal of hosting purposive conversations. Yet in another sense, falling is an appropriate word choice, in the sense that people, as conversational creatures, ‘stumble’ with things in conversations and allow themselves ‘to stay unbalance’ (Brinkmann 2020a). For Gadamer (1989), in fact, falling into conversations is a descriptive but also a normative concept: for a conversation to be genuine, we should fall into it:

We say that we ‘conduct’ a conversation, but the more genuine a conversation is the less its conduct lies within the will of either partner. Thus, a genuine conversation is never the one we want to conduct. Rather, it is generally more correct to say that we fall into conversation, or even that we become involved in it. The way one word follows another, with the conversation taking its own twists and reacting its own conclusion, may well be conducted in some, but the partners conversing are far less the leaders of it than the led. No one knows in advance what will ‘come out’ of a conversation.

(Gadamer 1989, 383)

If the host or the co-host of a WC has an awareness or disposition for method as a rigorous procedure to be followed, they might not fall into conversations. In this case, we can say that methods drive people away from the inventive, from acting as a knowledgeable researcher. That is why we should go beyond method, or perhaps use it only in small amounts. That is why we should consider the unmethod, which is ‘the opposite of acting on the basis of scripts and protocols; those are for beginners, and continuing reliance on them can doom actors to remain beginners’ (Frank, 2004, p. 221).

Cognizant of the problem of research amnesia – intentionally or unintentionally forgetting what has gone on in the past to make novel claims or advance one’s career – (Maines 2001), we want to make clear that we are not the first ones to put the anti-methodological argument forward in qualitative research (see Hammersley, 2009, for a review). We are also aware that giving up the notion of method and emphasising inventiveness raises questions about the status of the work as research. So, anything goes? Don’t we need rigorous methods in qualitative research in order to be objective? The usual answer to this question is that we do not need to be objective as qualitative researchers, but there is another possibility that goes by rethinking objectivity. Understanding objectivity as reflecting the nature of the object researched or as being adequate to a subject matter, Brinkmann (2014) contented that ‘the most objective forms of qualitative research are often the ones with the loosest designs. The more one decides to “collect data” in a methodological way … the less objectivity … can be attained’ (p. 724). That is precisely what the unmethod can help us achieve: a more objective form of qualitative research: a loose design that prevents method developing methodologically. The unmethod, in other words, is about method remaining ‘constitutionally unfinished’ or ‘methodologically immature’ (Gallacher and Gallagher, 2008). As an unmethod, therefore, the WC would remain open-ended, like it was in the rainy day it was conceived.

Conclusion

Whilst the methods literature on the WC has been helpful for us, at the same time its emphasis on method can have negative consequences for doing WC and good co-produced research. The WC, we propose, should not be conceptualized as a method, or practically engaged with as a method. It is better understood as a unmethod and practiced like jazz or as contact improvisation. It is about making space and time to support people to fall into conversations, and in the process enable their voices to cross pollinate. As researchers we should then resist the temptation to standardize and protocolize the WC. Perhaps we back up to 1995 and use the story we started with as a companion to guide the use of WC in research and helping us to understand this way of living up to good co-production as another unmethod.

Acknowledgements

We thank the members of the Moving Social Work Collective and all the people with lived experience who participated in the cafés for all their contributions. Our thanks also go to the anonymous reviewers for their useful comments and Brendan Gough for his editorial assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Institute for Health and Care Research (NIHR) [Applied Research Collaboration North East and North Cumbria (NIHR200173)] and Sport England.

Notes on contributors

Javier Monforte

Javier Monforte is an Assistant Professor at the University of Valencia, Spain. His qualitative research on the promotion of leisure time physical activity among disabled people has been awarded by several organisations, including the International Society of Qualitative Research in Sport and Exercise and the European Federation of Adapted Physical Activity. His work is published in journals such as Qualitative Inquiry and Qualitative Health Research, and in books such as the APA Handbook of Research Methods in Psychology (2^nd ed) and the The Sage Handbook of Health Psychology (2^nd ed, forthcoming). ORCID: 0000-0002-0909-0271 Twitter: @Monforte92

Jake Netherway

Jake Netherway is a PhD student at Durham University, UK. His research interests are physical activity, health behaviour and promotion, disability, social work, and participatory research. In particular, Jake is interested in how qualitative methodologies and co-production can help tackling inequalities and promoting social justice in the context of physical activity. ORCID: 0000-0001-8953-1622 Twitter: @JakeNetherway

Brett Smith

Brett Smith is a Professor of physical activity and disability, and Principal Investigator of Moving Social Work. He is the President of the International Society of Qualitative Research in Sport and Exercise and former Editor of Qualitative Research in Sport, Exercise and Health. Brett has co-authored/edited seven books, including the Routledge Handbook Qualitative Research in Sport and Exercise. His research has been published in journals such as Health Psychology, Sociology of Health and Illness, Social Science and Medicine, and the Lancet. Twitter: @BrettSmithProf

Notes

¹ The principles are: clarify the context; create a hospitable space; explore questions that matter; encourage everyone’s contribution; connect diverse perspectives; listen together for patterns and insights; and share collective discoveries. See https://theworldcafe.com/key-concepts-resources/design-principles/ for a description of each principle.