Power,position and social relations: Is the espoused absence of hierarchy in Open Dialogue naïve?

Abstract

Open Dialogue practitioners aim to reduce social hierarchies by not privileging any one voice in social network conversations, and thus creating space for a polyphony of voices. This sits in contrast to the traditional privileging of those voices credited with more knowledge or power because of social position or professional expertise. Using qualitative interviews, the aim of this current study was to explore Open Dialogue practitioners’ descriptions of challenges in implementing Open Dialogue at a women’s health clinic in Australia. Findings revealed how attempts to rhetorically flatten hierarchies among practitioners created challenges and a lack of clarity regarding roles and responsibilities. As the practitioners tried to adjust to new ways of working, they reverted to taking up engrained positions and power aligned with more conventional social and professional roles for leading therapy and decision-making. The findings raise questions about equity-oriented ways of working, such as Open Dialogue, where intentions of creating a flattened hierarchy may allow power structures and their effects to be minimised or ignored, rather than actively acknowledged and addressed. Further research is needed to consider the implications that shifting power relations might have on the roles and responsibilities of practitioners in the move to equity-oriented services.

Introduction

Open Dialogue is a resource-oriented and needs-adapted model of mental health care (Buus et al., 2017; Gromer, 2012; Haarakangas et al., 2007; Seikkula and Arnkil, 2006). It is both a therapeutic intervention and a service delivery model (Freeman et al., 2019) that has been shown to be a promising alternative approach for persons experiencing psychosis (Gromer, 2012). Originating in Finland in the 1990s, Open Dialogue is rooted in traditions of family therapy with treatment being organised via a series of network meetings. Network meetings include at least two Open Dialogue practitioners, the person ‘at the centre of concern’ and anyone they choose to invite from their social networks, including family members, friends, and other clinicians involved in their care (Seikkula and Arnkil, 2006). In the network meetings there is a focus on facilitating dialogue among all members of the network to elicit various points of view on the situation (Olson et al., 2014).

Open Dialogue practitioners aim to support persons seeking care, and their network, to take control over managing their situation in collaboration with care providers (Olson et al., 2014). This approach deviates from a conventional therapeutic encounter, where drawing on biomedical and psychological frameworks, an expert therapist takes responsibility for assessing, diagnosing, planning care and prescribing treatment, while the person seeking care is assigned responsibility for answering questions, learning about their diagnosis, and adhering to recommended treatments. Such a conventional model of care is structured with prescribed roles and responsibilities of therapists and persons seeking care and unequal distribution of power between them (Guilfoyle, 2003; Ong et al., 2023).

In contrast to conventional approaches, Open Dialogue practitioners take a particular ‘dialogical’ therapeutic stance to attenuate traditional power relations and create space for interaction amongst network meeting participants (Olson et al., 2014; Rossen et al., 2020). Key strategies of Open Dialogue include: following the person’s lead rather than the practitioner’s (Anderson, 2002; Seikkula, 2008, 2011); ‘being with’ rather than ‘doing to’ (Freeman et al., 2019); developing understandings by listening appreciatively without imposing derailing questions or interpretations (Rossen et al., 2020; Schriver et al., 2019); being transparent about the therapeutic process (Seikkula et al., 2006; Seikkula and Trimble, 2005); sitting with uncertainty rather than rapidly finding standardised solutions (Seikkula and Arnkil, 2006); promoting equity in decision-making (Seikkula and Arnkil, 2006); and avoiding subjugating any particular understandings, including those of the family or other dominant voices in the network (Rober and Seltzer, 2010). Practitioners orient to the person’s language use and own narrative and understandings of their situation (Olson et al., 2014). They are encouraged to disposition themselves as experts and instead take a ‘not-knowing’ stance (Anderson and Goolishian, 1992), participating in the meeting by engaging with the person’s story through their own personal experiences and vulnerabilities (Schubert et al., 2021). As each network member engages with the person’s story from their unique perspective, a multiplicity of voices is elicited, co-existing in responsive but unfinalized ways (Holquist, 2002). Bakhtin’s (1984) metaphor of ‘polyphony’, where multiple autonomous co-existing voices come together, is used by Seikkula and colleagues to describe the process of inviting and clarifying perspectives within network meetings. Polyphony within dialogical therapeutic conversations contrasts with traditional monological or totalising therapeutic conversations. Facilitating the opportunities for polyphony in network meetings is central to the work of Open Dialogue practitioners.

The dialogical strategies central to Open Dialogue are designed to flatten traditional hierarchies that exist in healthcare and to promote more humanising and recovery-oriented responses (Ong et al., 2023). As an alternative approach to therapeutic practice, many have enthusiastically received Open Dialogue as a hopeful and promising way of working that might mitigate power differentials experienced in other ways of working (Buus et al., 2022). For instance, an introduction of Open Dialogue in a women’s refuge found that both this disadvantaged population of women and the practitioners experienced polyphony within the network meetings as a helpful therapeutic resource, which had the potential to ‘reduce feelings of powerlessness for service users and disrupt dominant relations of power within the therapeutic encounter’ (Dawson et al., 2021a, 2021b: 137). Healthcare professionals in this study identified Open Dialogue as aligning more closely with their values as caregivers, when compared to more traditional or medicalised approaches (Dawson et al., 2021a, 2021b). Similarly, clients and peer workers have considered Open Dialogue as a more respectful way of receiving and giving care (Bellingham et al., 2018), and a need for ideological change towards more person and recovery oriented care in mental health services has been identified (Dawson et al., 2021b). At present, however, there remains limited research regarding whether hopes for Open Dialogue as a more equitable approach to providing care, are justified.

While mitigation of traditional power relations is a key component of Open Dialogue, the role of power has generally been under-theorised (Ong et al., 2023), and is a vexed issue in family therapy more generally (Flaskas and Humphreys, 1993). Some recent research has examined the experience of using dialogical processes in therapy sessions with findings suggesting that the dialogical stance can mitigate against the inherent power that therapists hold in the room (Ong et al., 2021a, 2021b), and that clients can experience network meetings as more equalising and humanising than other interactions with health services (Dawson et al., 2021a, 2021b). However, there has been scant research into the implications of shifting power relations from a broader organisation context, when dialogical approaches are adopted.

Implementation and evaluation of Open Dialogue for people with a variety of mental health concerns is underway in several countries around the world (Buus et al., 2017, 2021). However, very little peer-reviewed research considers the perspectives and experiences of Open Dialogue practitioners or service users (Tribe et al., 2019). More specifically, while Open Dialogue literature and training focus on conceptual approaches that resist hierarchical relations to encourage equal voices within dialogical processes, there is little attention to practitioners’ actual experiences of hierarchies within the service. Questions remain as to how a service is impacted when such an approach is introduced in an organisation (Buus et al., 2021). Therefore, the aim of the current paper is to explore Open Dialogue practitioners’ descriptions of challenges in the processes of implementing Open Dialogue.

Method

Semi-structured in-depth interviews were conducted with practitioners who had participated in network meetings at an Australian women’s health clinic over a period of 12 months. All practitioners at the clinic were invited to participate in an interview. Participants gave their informed consent to participate, based on written and oral information about the study. Interview responses were managed confidentially. Details that could identify individual informants have been altered and interview transcripts are not publicly available. The study was approved by The University of Sydney’s Human Research Ethics Committee.

Study Site

The women’s health clinic operates through two sites in a low socioeconomic suburban area of a major Australian city. The clinic team consists of a small group of administrative staff members and approximately 20 female clinicians, including nurses, a clinical psychologist, drug and alcohol counsellor, and general practitioner. The clinic offers drop-in services during weekdays for women and attends to a wide variety of health issues using holistic approaches. The women who access the clinic come from diverse backgrounds and experience physical and mental health issues. The clinic envisioned that an Open Dialogue approach, later named ‘The Open Dialogue Initiative’, could support their clients who present with complex bio-psycho-social issues. The clinic has policies underpinned by values of respect, dignity, empowerment, community engagement and accountability.

Participants

Of the nine practitioners engaged in the Open Dialogue Initiative at the clinic, seven agreed to participate in the study. Participants had different levels of engagement with the clinic and with the Open Dialogue Initiative. Three of the participants were employed by the clinic (internal practitioners), one in a leadership role (clinic lead) and two were primary health care nurses, all identified as women. The three internal practitioners would only facilitate meetings when they did not have direct clinical responsibility for the person at the centre of concern. When they had such responsibilities, they would take a supportive, network position during the meetings. The remaining four participants held roles external to the clinic and facilitated network meetings with the Open Dialogue Initiative on a voluntary basis. These external practitioners included an academic director of a research centre initiating Open Dialogue training and research; a mental health nurse academic; a clinical psychologist; and a mental health nurse clinician. Two of the external practitioners (research centre director and clinical psychologist) identified as men. There were varying levels of expertise working in mental health among the seven participants. All practitioners involved in the Initiative had completed a minimum of 1-week intensive introductory training in Open Dialogue. Two of the external practitioners were part way through an intensive 3-year Open Dialogue training course. Others were substantively engaged in Open Dialogue practice, training, and research. All participants were engaged in supervision and had opportunity to reflect on and discuss any tensions between Open Dialogue and traditional models of care.

There are currently no fidelity criteria that can be used to determine whether an organisation has fully adopted and implemented an Open Dialogue approach (Waters et al., 2021). The health clinic aimed at a smaller implementation of Open Dialogue practices in their work with selected women and did not envision using Open Dialogue to fundamentally transform their organisational work processes and position in the general service delivery system. Also, given the lack of fidelity criteria, the practitioners had to develop a unique version of Open Dialogue that had a good fit with the clinic’s organisational values and practices. However, while this local version may have differed from other versions of Open Dialogue, the practitioners understood it to be genuine and fully in line with Open Dialogue philosophy.

Interviews

Semi-structured individual interviews were conducted and were approximately 1-hour in duration. Questions focussed on participants’ experience of network meetings and the perceived impacts of working in this way at the health clinic, including questions such as, ‘What has the experience in the network meetings been like for you?’ and ‘Is there anything specific that you think Open Dialogue brings to the clinic?’; ‘What do you think is helpful/unhelpful about network meetings in this context?’. Participants were also asked about their experience of the introduction of network meetings to the clinic and the processes around implementation. Interviews were audio-recorded and transcribed verbatim. All interviews were conducted by LD, a doctoral prepared qualitative researcher, clinical psychologist and experienced family therapist. At the time of the study, LD was working in the research centre and undertaking an intensive 2-year Open Dialogue training course but did not have a role within the clinic or its Open Dialogue Initiative. Three out of four authors were enrolled in an extensive Open Dialogue training under supervision.

Analysis

This paper offers an analysis of case study data (Yin, 2009). RE and LD independently performed an open coding (Coffey and Atkinson, 1996) of the interview texts. RE is a doctoral prepared qualitative researcher and a registered nurse working at the research centre. She has undertaken 1-week Open Dialogue introductory training but did not have a role within the clinic or its Open Dialogue Initiative. She applies expertise in feminist and critical social methodologies in analyses of issues related to health, social inequity and violence.

Participants spoke about navigating the dispositioning of expertise inside and outside of the network meetings. Following their emphasis in discussing shifting relations between practitioners, we were particularly curious about these experiences. Interpretations of their described shifts were guided by understandings of dominant hierarchical structures, social positioning and relations of power. Specifically, conceptual understandings of how dominant relations of power can be interrupted and reformed to maintain the social order (Foucault, 1991); and how those who support emancipatory aims run considerable risk of dominating, precluding, or ignoring the voices of those with the least power (Spivak, 1988) together provided a theoretical guide for the analysis. Through extensive discussions, comparative analyses, and memo writing we gradually and collaboratively developed an interpretation regarding shifting power relations in the implementation process, which is reported here.

Results

During the implementation of the Open Dialogue Initiative in the health clinic, the participants described significant relational shifts. These shifts were organised into three stages that form the overarching themes of this analysis. In stage one, ‘Alliances, sharing values and vulnerability’, participants described forming alliances within the context of shared values, vulnerability and excitement about learning together. In stage two, ‘Who is in charge? Confusion of roles and leadership’, they described a lack of clarity around roles and responsibilities, where in the absence of dialogue or transparency around decision-making relations of power operated. In stage three, ‘Re-establishing control and clarity regarding accountability and responsibilities’, dominant relations of power were re-established and played out in the development of rules and expectations, gender of practitioners became a contested space, and partnerships were tested.

Stage One: Alliances, sharing values and vulnerability

Fit with values and ‘being in it together’

The Initiative was started after a small group of clinicians from the health clinic completed a 1-week training course in the Open Dialogue approach. They understood Open Dialogue to be aligned with holistic ways of working already used in the clinic. The clinic lead explained, ‘Open Dialogue fit with our values [. . .] that women have expertise in their own lives, and that we should be led by the women we’re working with’ (P5). Another clinician contrasted traditional approaches with Open Dialogue,

‘organisations and medical systems seem to be so siloed. There’s no human connection and you have to go over your story time and time again. And it’s not what they want to talk about, it’s what the expert wants them to talk about. So, for me Open Dialogue takes all of that away and they only talk about what’s important for them. So, it’s all given back to them. They have control so they’re not pointed at or judged or told what’s wrong with them’. (P2).

While these clinicians wanted to initiate use of Open Dialogue within the clinic, they felt they needed support. At the same time, a group of mental health practitioners were looking for opportunities to practise Open Dialogue as a compulsory part of their training and volunteered to support an Open Dialogue Initiative at the clinic. At the outset of the Initiative, there was a high level of engagement and excitement from both the health clinicians (internal practitioners) and the volunteers (external practitioners), and the arrangement was described by several study participants as very collaborative. Both groups described how Open Dialogue offered a sense of alignment between personal and professional values. Most described how practicing dialogically felt good and how Open Dialogue carried a sense of ‘interpersonal magic’ (P4) that extended beyond the work, into personal relationships. One participant reflected on how it fostered a sense of idealism amongst the practitioners, who ‘felt very inspired’ (P5) and ‘fell in love’ (P5) with the process. Amidst their shared values, relations among the practitioners deepened professionally and personally.

Learning together while at different stages

Establishing Open Dialogue was a new enterprise for all the practitioners, who described being nervous and excited at the outset. The practitioners had different levels of experience in psychotherapy practice, and while the external practitioners had more intensive training in Open Dialogue none had led facilitation of network meetings. The clinic lead recalled, ‘I don’t think I was entirely clear about how much experience the [external practitioners] had in Open Dialogue when they first came in because ultimately it was less than I thought it was. But that was okay, because we were all learning together’ (P5). All were relatively new to Open Dialogue, hence there was a feeling that they were ‘learning together’.

Group supervision was provided monthly to support the Initiative by an experienced family therapist not affiliated with the health clinic, who had received training in Open Dialogue. However, it was difficult to find a time when all practitioners could attend, and while practitioners noted how helpful supervision was, they also observed it was not frequent or consistent enough. As one participant reflected, supervision offered ‘a space to come together and think more. [. . .] I would have kind of liked more of that during that time, more of the supervision’ (P1).

Practitioners progressed in their practice of Open Dialogue at different rates, according to their opportunities for training and practice. Some described feeling more confident and skilled in facilitating network meetings compared to others, which led to the formation of new hierarchies between practitioners. One staff member recounted, ‘I was really mindful during these initial sessions and have been since, how moving at these different paces or through these different phases [. . .] created some sort of divide’ (P1). Experience in Open Dialogue and level of affiliation with the clinic were factors in relations of power that began to emerge.

Stage Two: Who is in charge? Confusion of roles and leadership

Dis-positioning expertise: Lack of clarity of roles

Within the equity-oriented approach of Open Dialogue, practitioners aimed to work collaboratively. For example, in utilising an Open Dialogue approach, there was an expectation that all practitioners at some point would take the lead in facilitating network meetings regardless of their discipline, or professional status. Some practitioners stepped back from facilitating meetings, a role that may have been seen as held by those with more authority, and invited others, with less relative power, to step up. The hierarchical nature of pre-existing organisational, professional, and social roles and positions began to shift. The research centre director described,

‘I’ve seen this in other situations that when you’re trying to sort of change the existing order of things [. . .] something happens when you start learning about Open Dialogue and you’re sort of at the bottom of the hierarchy and you get these new skills that are about flattening the hierarchy and suddenly you end up trying to find a new position in your organisation’ (P4).

Repositioning of practitioners to create greater professional equity was not a smooth process. An internal practitioner recounted jostling between some of the practitioners in what she called, ‘office politics’: ‘I feel like there’s just a lot of office politics. And I just don’t understand because it [Open Dialogue] is such a great thing [. . .] but it [office politics] still seems to be there and I’m not sure how to navigate it’ (P7). Some participants expressed uncertainty regarding how to step up and take a lead in facilitating network meetings and questioned who had the legitimacy to step into such roles. Another practitioner similarly described a loss of direction, ‘it’s a very bumpy sea and I’m in a boat and I’m not sure where my compass is going’ (P6).

Reverting to traditional hierarchies – professional status and gender

Despite that none of the practitioners had previous experience facilitating Open Dialogue network meetings, when deciding who would take a lead role in facilitating the first meeting, the research centre director felt the others deferred to him. He described that it felt as if his colleagues said, ‘‘Please go ahead and take the riskiest position to begin with’. So, I ended up leading a lot of the groups [network meetings] at the beginning’ (P4). The uncertainty regarding roles and leadership was experienced by the participants within network meetings but also in relation to organisational aspects of the Open Dialogue Initiative as a whole. For example, the notion of who was organisationally ‘in charge’ was initially unclear. The clinic lead described how, despite her leadership role at the clinic, she initially deferred a sense of ownership of the Open Dialogue Initiative to the research centre director, ‘I was probably still thinking about the Initiative as [his] Initiative that is running from [the clinic]’ (P5). Rather than being interpreted as related to gender, these deferrals and constitution of the research centre director as the leader of the Initiative might be interpreted as related to his senior position in a centre running Open Dialogue training and research. Interestingly, the other male practitioner, an experienced clinical psychologist who was no more experienced in Open Dialogue than others, described how he was consistently positioned as lead facilitator within the network meetings. He described feeling, ‘somewhat responsible for the success of the Initiative [on my days. . . .] It’s not that I feel that I was “the” person who ran it, and it would happen without me, I’ve got no doubt, but I was aware that my consistency there was valued [. . .] it felt like I was kind of in an important position’ (P1). Thus, in addition to professional expertise dominant social gender roles may have played some part in the organisation of leadership.

The hierarchies of no-hierarchy

Subtle moves began to shape a new hierarchy around legitimacy to work within the Initiative, in terms of insiders (internal practitioners) and outsiders (external practitioners) and whether practitioners believed they had more or less expertise than others. In reflecting on the implementation of the Open Dialogue Initiative, the clinic lead and the research centre director both noted a sense of jockeying for power among practitioners, which rather than flattening the hierarchy, created a new one. The clinic lead noted,

‘even though you can aim to be less of an expert, how do other people perceive you? How do you step in and out of different roles at different times because everybody in the Initiative is carrying multiple roles with them? How do people, um, build power through relationships and alliances with each other? [. . .On the surface it was] like, “Oh no, there’s no alliances and there’s no power in our circles” .I’m not buying that. Because I feel like I’ve seen examples of that’ (P5).

Echoing this idea, the research centre director noted, ‘there’s this weird thing about hierarchy but no hierarchy at [the clinic]. We had a supervision session the other day [. . .] without any question as to why is the clinic lead in the room? Because that would be a thing I would be questioning as odd’ (P4). Another participant agreed but extended this questioning to his presence in supervision as well, ‘my leader is in the supervision, so that’s a challenge. [. . .] Same as [the director] it’s difficult. Cause they’re in charge’ (P2).

Some practitioners described feeling that their practice of Open Dialogue during network meetings was criticised by other practitioners. Aiming for equity amongst themselves, practitioners were self-regulating with no one assigned to give overt feedback about dialogical practice to others with less training or experience. Some practitioners gave feedback that was experienced by others as criticism. A practitioner described, ‘I was really scolded [. . .], I felt attacked’ (P6). At the same time, this participant expressed relief when this behaviour ‘had been noticed by the seniors. And then they changed how we reflected at the end and [. . .] [Since it was addressed] I feel so much more confident to say something [in a network meeting], even if I get it wrong’ (P6). This practitioner questioned the legitimacy of the initial critical feedback she received based on its delivery, and by positioning the practitioner who gave the feedback as junior. In this way, this response reverted to traditional hierarchies. Similarly, the clinic lead commented on how,

‘when people are learning new skills as well as practicing new skills, [. . .] thinking about how we’re sensitive to each other and keeping each other safe is also really important. [. . .] there are definitely practitioners [. . . who have] felt hurt by feedback other practitioners have given them at the end of an Open Dialogue session, and that they haven’t felt able to say that’. (P5)

Overt misuse of power was perceived by some participants in organisational aspects of the Initiative as well. The research centre director recounted how, in planning the network meetings ‘[an internal practitioner] would favour some people’ (P4). For example, some practitioners would be left on or off the schedule for network meetings, or not informed about schedule changes. An external practitioner noted that when a session was cancelled, while others were notified, for example, ‘I wasn’t informed. My first thoughts were value. [Another practitioner] has value, I don’t’ (P6).

A lack of preparation for implementation was an issue for another external participant, who contextualised relational issues among practitioners, saying: ‘I don’t know how much thought was put into the process in advance. [. . .] What do we need to run this effectively to keep people safe? Not just the people who are accessing support, but the people who are providing the service’ (P1). While supervision was generally described as helpful, an external practitioner reflected how it was not able to effectively address these issues, which left them feeling ‘unsafe’ (P6). Another participant agreed and reported that supervision became another space for these interactions to play out, ‘the supervision that we did have, I think it was helpful. But again, like for me that really played out all of the dynamics’ (P7).

Despite emerging challenges between the practitioners, they reported that they were able to create a dialogical space during network meetings. However, a flattened hierarchy between the practitioners outside meetings was more challenging to achieve. A health clinic staff member argued that, ‘the network meetings [. . .] They’re fine, from my experience. I think, for me, it’s the after and the before’ (P7). The clinic lead reflected,

‘I’m [. . .the] clinic lead, [the research centre director. . .] holds the purse strings. He holds power, I hold power, [another internal practitioner] holds power over the schedule. [. . .] I’m not sure that we thought about the power we hold enough and I’m not sure that we thought enough about how we were exercising that power at different times as we were rolling it out’ (P5).

Stage three: Re-establishing clarity regarding accountability and responsibilities

Approaches to organisational decisions were not planned as part of the implementation of the Initiative. Participants aimed to work equitably and collaboratively, however the unforeseen impacts of power relations eventually emerged as a challenge. With growing concerns and accountabilities related to organisational aspects of the Initiative and dynamics between the practitioners, the health clinic lead re-established control of decisions related to the Open Dialogue Initiative. She described,

‘roles weren’t clear enough in the beginning [. . .] I just ended up basically saying that it’s our [the health clinic’s] Initiative and we’re in charge of it. And you know with the utmost respect and gratitude for [the external practitioners] the generosity of receiving the training. And [. . .] for the role everyone had in us establishing the Initiative. Because without the support of all these different, amazing, kind, well-intended, experienced practitioners we wouldn’t necessarily have an Initiative like that’. (P5)

Over time, internal practitioners had developed clarity around how they wanted the Initiative to run and felt that independence from the voluntary external practitioners was important for the sustainability of the Initiative. They developed new guidelines for network meetings in consultation with one external practitioner who had the most consistent presence. These guidelines were communicated unilaterally from the clinic lead to external practitioners in an email. The guidelines included: 1. internal practitioners would lead facilitation of the network meetings; 2. only one male practitioner could be in a network meeting at a time; and 3. external practitioners would be required to commit to consistency, timing, and regularity of participation.

Independence of the Initiative underpinned the rationale for the first guideline. The clinic lead described: ‘staff had developed enough skills and experience to lead more of the sessions. For the Initiative to be sustainable into the future we want to make sure that there’s [internal] leadership in the Initiative, working of course in partnership with everybody’ (P5). Yet, not all clinicians were comfortable putting these guidelines into practice. One described how, ‘it’s difficult, I still have that, even though I work with it all the time now, it’s still, I’m not an expert in it, so it’s still that bit intimidating with the other practitioners’ (P2). Further, the research centre director described concern with this decision, ‘I understand their ambition to have an autonomous provision of Open Dialogue where they are, but at the same time, I would be cautious about thinking that they are able to do the Open Dialogue with the quality that would be needed. So, it’s both a symbolic thing, but it’s also a safety/quality thing’ (P4). A discourse of overriding expertise sits in tension with the principles and aim of Open Dialogue to avoid privileging certain voices and to disposition dominant understandings of what constitutes expertise. Yet at the same time, the internal practitioners were perceived as having less formal training/experience in mental health practice, were not included in further Open Dialogue training, and had limited opportunities for supervision. While the external practitioners were also relatively new to the therapeutic approach of Open Dialogue, all had expertise in mental health practice, were participating in intensive Open Dialogue training, and had extensive ongoing supervision as a requirement of their training or separate clinical practice.

The second guideline limited the two male external practitioners’ access to network meetings. Due to the sensitive nature of care, the Women’s health clinic employed a high proportion of female staff. The internal practitioners described tensions with having male practitioners in network meetings. The inclusion of male practitioners had been ‘really controversial. And something that I really have stuck my neck out, on the line for. [. . .] Because [the research centre director] was the head of the Initiative, to get it happening we had to have men here. I’ve had lots of different questions about that from lots of different people’ (P5). Additionally, another internal practitioner identified her own caution with working with men in this context, ‘It’s different working with men than it is with women. There’s patriarchy and power involved [. . .] it’s not a comfortable place when you’re a women’s service because it makes you vulnerable, it makes you question your own practice’ (P2). However, the imposition of this new limit was complicated by an external practitioner advocating for a women-only Initiative, which would have excluded the director. Another internal practitioner concurred, suggesting that this guideline might be complicated by other aspects of relations, ‘I’m not sure if it’s necessarily gender though, or if it’s dynamics generally in terms of personality or professions’ (P7).

The third guideline required an increased commitment on behalf of the volunteer external practitioners. Up to this time, they had been supporting network meetings but were not engaged in administration or follow up of issues raised by the women. While the external participants had constituted this as beyond their responsibility as volunteers, the internal practitioners described how the Initiative had increased their workload. As one internal practitioner described,

‘it’s a very wonderful thing to have people donate their time. But it’s also very difficult because they’re not obliged. [. . .] for us there’s a lot of feeling obliged [. . .to support] continuity. [. . .] staff members are going to be utilised a lot more to kind of help fill the gaps [. . .]. On one hand, I’m really grateful because I want to practice, I want to keep learning, I want to go to more training and get better at it. But on the other hand, it’s just kind of like, it’s just a long day [working at the health clinic in addition to participating in network meetings]. It’s a long day. And we technically become volunteers [. . .] as well because we don’t get paid extra’ (P7).

Furthermore, external and internal practitioners had different priorities. External practitioners required specific practice hours for their Open Dialogue training, thus working at the Initiative contributed to their training, which could be applied to other paid roles. However, internal practitioners who were not specifically mental health practitioners nor researchers, did not share similar future opportunities. Instead, they described an obligation, ‘if we don’t get it up and running and consistent, it kind of feels like another thing we can add to the basket of things that have sort of let them [the women we serve] down’ (P7).

Discussion

During the implementation of the Open Dialogue Initiative in the health clinic, attempts to attenuate relations of power in the interest of providing an inclusive service occurred within the context of existing hierarchical relations. After moving into a collective vulnerability to engage in a dialogical approach, the ethos of equity and emphasis on dialogue did not translate beyond the network meetings. In the absence of these conversations, a sense of partnership between the internal and external practitioners declined, and hierarchies seemed to reform around insider and outsider status at the clinic, gender, and knowledge, training, and expertise in Open Dialogue. Limited resources also played a role in some of the relational challenges between practitioners. However, in the context of these challenges, our analysis highlighted a need for more careful considerations of the power relations at play within Open Dialogue implementation. It seemed that the interpretations of Open Dialogue and implementation in the clinic was somewhat naïve to relations of power. Below, we discuss this on the basis of a review of key Open Dialogue texts, where we noted a lack of guidance to navigate structural, hierarchical, and cultural factors relevant to local implementations.

Social ordering is achieved through engagement between individuals and social structures, practices and discourses that incite self-management or governance of behaviour to avoid punishment (Foucault, 1991). Davies and Harré (1990: 52) illustrate how ordering occurs through the subtleties of discursive practices, ‘Any narrative that we collaboratively unfold with other people thus draws on a knowledge of social structures and the roles that are recognisably allocated to people within those structures’. Our analysis suggests that participants’ expectations of non-hierarchical experiences of Open Dialogue may operate paradoxically, in the sense that instead of being attenuated, relations of power are made covert and thus more difficult to address. Thus, attempts to work in equity-oriented ways require careful attention to the context and power relations in which they take place, particularly when the espoused aim is to flatten hierarchies where power structures and their effects might be less noticeable, minimised, or ignored.

While health services have increased efforts to include patient and public involvement, a lack of attention to hierarchies of power has meant that social positioning continues to impact influence (O’Shea et al., 2019). Similarly, within the literature and practical training for Open Dialogue, there is little attention yet paid to power relations, the implications of social positions, or the challenges and complexities of navigating hierarchical relations within or outside the network meetings (Ong et al., 2023). An exception is Seikkula and Arnkil’s (2006) Epilogue: On power and empowerment. In this epilogue, Seikkula and Arnkil (2006: 187) focus on power relations between therapists and clients, drawing on a discursive tradition of ‘empowerment’ to describe power as something that experts hold and may choose to bestow upon clients, by ‘helping clients to help themselves’.

Conceptualisations of power within empowerment discourses have roots in radical social liberation movements (Friere, 1970; Solomon, 1976), however by the 1990s empowerment discourses were transformed in a widespread uptake in social policy (McLaughlin, 2015). In this transformation, the focus shifted from the collective to the individual, supporting neoliberal health promotion policy targeted at addressing individual behaviours rather than social determinants of health (McLaughlin, 2015). Seikkula and Arnkil’s conceptualisation of sharing power with clients to promote certain types of individual behaviours whereby clients help themselves, echoes the later discourse of empowerment. Rather than being emancipatory, this sharing of power has the condition of self-regulation.

Foucault (1991) describes how power relations within post-monarchical governance are characterised by self-regulation, with a shift away from overt physical force to much more subtle forms. Specifically, relations of power are designed and built into the structures of the social in ways that enrol active subjects in their own governance. To illustrate the central role of self-regulation in contemporary society, Foucault used the metaphor of The Panopticon, a famous prison design created by Jeremy Bentham (Foucault, 1991). The Panopticon was designed with a central watchtower allowing unobstructed views into each prison cell and thus affording the possibility or threat of constant surveillance of every prisoner. This threat meant guards were no longer needed to maintain order in the prison by force, rather, prisoners were enrolled in self-regulation. Yet, drawing on the work of Cruikshank, 1999; Seikkula and Arnkil, 2006: 189) argue ‘that empowerment is a form of power that promotes rather than represses subjectivity [. . .and] produces and relies upon active subjects rather than subjugation’. Positioning subjugation in opposition with agency (‘rather than’) demonstrates a lack of appreciation for the nuances of relations of power. Subjugation requires the active (agential) subject for self-regulation (Foucault, 1991).

Seikkula and Arnkil (2006: 191) also argue that: ‘Multilateral situations do not lend themselves easily to unilateral control. Therefore, they afford both the need and ample opportunities for experiencing and joining polyphony’. Seikkula adapted Bakhtin’s (1984) metaphor of ‘polyphony’ to Open Dialogue. Polyphony was used by Bakhtin to describe how the characters in Dostoevsky’s novels govern the plotlines through a plurality of voices, decentring the usual authority of the author’s voice. Bakhtin (1984: 6) explains, ‘What unfolds in his works is . . . a plurality of consciousness, with equal rights and each with its own world, combining but are not merged in the unity of the event’. This analysis of Dostoevsky’s dialogical prose offers a significant departure from the monologic truths of expert opinions, familiar to the patriarchy of psychiatry. Open Dialogue sits on the premise that dialogical approaches can disrupt these power relations, and rests on an assumption that all voices can be shared and can be heard in polyphony. However, this assumption fails to fully appreciate the political aspects that constitute opportunities for voice. Over the past half a century, feminist and critical race scholarship has convincingly established how social ordering and relations of power constitute not only the content of what can be said, when and by whom, but these relations also constitute eligibility to speak (Spivak, 1988). We believe that throughout this chapter, Seikkula and Arnkil oversimplify relations and operations of power, naïvely overlooking self-regulation as central to governmentality, and social positioning and context constituting opportunities for voice.

Ong et al.’s (2021b) conversation analysis offers another exception to the lack of discussion of relations of power in the Open Dialogue literature, however they also focus on the relations that occur within network meetings. They analysed how Australian Open Dialogue practitioners directed network meetings while managing hierarchical relations and adhering to Open Dialogue’s aim to avoid privileging certain voices. They differentiate theoretically between ‘deontic stance’, which refers to the level of authority that a person displays in a situation, and ‘deontic status’, which refers to the level of situated authority of a person’s social position. Ong et al. (2021b) demonstrate that clinicians systematically downgrade their deontic stance, such as using the phrase ‘I’m wondering if. . .’ rather than, for instance, ‘I think that . . .’. A key point here is, that while the rhetorical downgrading of stance can create a sense of more symmetrical relationships and can make new actions possible, it does not change participants’ deontic status. Privilege, responsibility, and disadvantage associated with, for instance, age, gender, social class, role, and ability, continue to inform power relations and influence dialogue.

The existing literature that considers power in dialogical spaces has been dominated by exploring the need for flattening hierarchies between professionals and the people they serve (Seikkula et al., 2006; Seikkula and Arnkil, 2006) with little or no attention paid to what happens between professionals themselves in those processes (Buus et al., 2021). A rare exception was Søndergaard’s (2009) study that argued that the introduction of Open Dialogue in a small Danish outreach mental health team significantly intensified the relationships between team members. This, for instance, happened through the changes in bodily interactions (e.g. clinicians sitting on chairs in circles) that make ‘everything possible’. Søndergaard (2009: 252) concluded that: ‘. . . innovations do not start with a decision nor end with a result. They start with a dream and end with practical negotiations’. In Søndergaard’s study, intense negotiations ultimately led to the demise of the team and its dialogical work. In our current study, we have offered a similar example of how identifying and addressing relations of power between professionals may be especially difficult amidst emancipatory ideals. Not only do these ideals set expectations for providers, but they invite vulnerability. More discerning research is needed in this area that appreciates the importance of attending to how relations of power operate in the context of equity-oriented initiatives. In addition, a critical awareness of the various possible implications that might unfold within a disruption of relations of power and the usual social order is needed. Possible implications include impacts on roles and responsibilities of professionals and others engaged in Open Dialogue, and their relationships with each other.

In our study, the reestablishment of centralised decision-making increased clarity of expectations but compromised a sense of partnership between practitioners. Awareness of how positions of deontic authority also carry responsibility and organise possible responses, points to the need for a structured mechanism for dialogue regarding organisational aspects throughout an implementation process. Another potential strategy that could support organisations with implementation of an Open Dialogue service might include explicit attention to how relations of power can shift among practitioners, an expectation that this is likely to occur, and considerations regarding how this situation might be addressed. Finally, increased resources to support supervision with this focus might be beneficial for future implementations.

We acknowledge that the analysis is only one possible interpretation of the implementation processes at the women’s health clinic. This interpretation is influenced by our interest in the operations of hierarchical structures, social positioning, and relations of power, and it would be possible to emphasise other elements of the implementation process. Further, we appreciate there were two practitioners who did not participate in the study, thus their voices are missing from both the data and our analysis. At the same time, a collaborative approach to this analysis inspired a reflective and ‘complicating’ (Coffey and Atkinson, 1996) interpretative process that prevented premature closure of interpretations. We believe that this complicating process combined with the presentation of highly contextualised data are key contributors to the overall credibility of the analysis. We make no claims about the generalisability of the findings but invite readers to consider the contextualised findings in a case-by-case manner, cf. (Brinkmann and Kvale, 2014), and the extent to which the findings might have relevance in their own implementation contexts, as well as in other programmes of equity-oriented service reform.

Conclusion

Open Dialogue has admirable aims regarding flattening dominant hierarchies that are associated with inequity and its potential for harm. While there has been some consideration of how to achieve this within network meetings for clients, families and other network members, less attention has been given to how professionals negotiate, acknowledge, and attend to power between themselves. Individuals and organisations who aim to work in more equitable ways need support. The findings of this study confirm that, regardless of the ambitions of everyone involved, power is ubiquitous. The challenge to identify, consider and attenuate organisational hierarchies in a sustained way within the context of an Open Dialogue Initiative in a women’s health clinic, highlights how resisting hierarchy that permeates our social world is complicated. How to manage relations of power in ways that allow for helpful, meaningful, and sustainable engagement in Open Dialogue and other equity-oriented interventions is needed if to fulfil the ambitions of improving equitable approaches.