The Research Relationship: Negotiating Multiple Selves and Boundaries in Exploring Sensitive Topics

Abstract

This article considers responsibilities and challenges inherent in the research relationship, from the position of a researcher who is also a counselling practitioner. It draws on my experience of undertaking a qualitative interview-based doctoral research study with adult survivors of childhood sexual abuse, engaging critically with the debates in the research literature concerning researcher–practitioner role boundaries and comparable (and distinct) areas of practice between research and counselling. I suggest that within well-held, monitored boundaries, practitioner identities and contextual knowledge are invaluable to the research relationship and that a collaborative fluidity can operate between researcher and professional (in this case, counsellor) identities rather than them being in conflict. Though the issues addressed here arise from the researcher as counselling practitioner, I believe they have a wider relevance for all qualitative researchers. What happens in the research relationship is complex, involving the various identities (personal and professional selves), emotions, and subjectivities of both researchers and research contributors. Our personhood in research can help to generate rich sources of understanding and at the same time demands our critical reflexivity to interrogate our subjectivities and their influence. In undertaking research which asks individuals to reflect in detail and depth on intimate areas of their lives, researchers need to be prepared for the potential emergence of distress and feel equipped, through training, support, and contextual-based knowledge, to be able to respond appropriately. It calls for reflexive relational competence at the heart of qualitative research.

Introduction

Undertaking research into “sensitive” topics poses many challenges for researchers and research contributors alike. Often, such research involves in-depth semi-structured or unstructured interviews (conversations) which focus on personal and intimate aspects of people’s lives. These conversations create the possibility of entering the lifeworld of the contributors and fostering rich, detailed knowledge and understanding of the research topic (Kvale, 2008). They can also evoke painful memories and emotional distress. This is particularly relevant in research with survivors of childhood sexual abuse (CSA), where the impact of abuse is often complex and difficult to assimilate. In speaking about their experiences, survivors may reconnect with forgotten memories, become distressed, or experience heightened anxiety, possibly triggering trauma reactions (e.g., dissociation and flashbacks) in the interviews themselves or later as a result (Burke-Draucker, 1999; Carlson et al., 2003). Researchers too may experience powerful emotional responses and can be at risk of secondary traumatization (Dickson-Swift et al., 2007; Stoler, 2002; Williamson et al., 2020). The reactions in researchers and contributors cannot always be predicted in advance. Lee and Renzetti (1993) suggest sensitive research that “delv[es] into deeply personal experience” or “intru[des] into the private sphere” constitutes a “substantial threat” to those involved, both researchers and research contributors (p. 6). The solution is not to avoid such research on account of these threats but to “confront seriously” the problems they present (p. 11). It is incumbent on the researcher(s) to ensure that ethical issues are anticipated (as far as possible), recognized, and addressed, not only in the preparatory stages of the project but considered and negotiated throughout the research process.

This paper examines aspects of relational ethics which arose in my doctoral studies, concerning role boundaries and the responsibilities as a researcher to be able to respond appropriately to the dynamics and emotional impact of taking part in research interviews in sensitive topics. Central to this discussion, alongside my role as a researcher, is my background as a professional counsellor, working for many years previously with CSA survivors in this capacity.

In the study, I met with nine CSA survivors, who had been active in some form of creative practice (e.g., visual art, writing, music, and performance), for several in-depth individual conversations as part of a narrative inquiry methodology. The aim was to explore the impact of creative activities upon their recovery process. The study was approved by the University of the West of England (UWE, Bristol) Research Ethics Committee (approval no. HAS.21.06.161), and all participants (co-researchers) provided written informed consent prior to enrolment in the study. A preliminary online meeting took place, which was followed by sending written information, so that potential contributors could understand what was involved and make informed decisions about participation. A screening procedure was integral to this process to identify and ensure, as far as possible, that individuals who might be at risk of harm were not included in the study. We discussed the possible impact of being involved alongside the individual’s current circumstances and protective factors, such as support networks. Survivors also completed a self-report risk form based on the (debarring) criteria used in a study by Meston et al. (2013). I was aware that these subjective assessments might still result in individuals at risk of harm being included and further marginalize others who were excluded with the loss to the study of their experiences and insights. The protocols (in the information sheets) for responding to differing levels of co-researcher distress and the clarification that changes in an excluded individual’s circumstances might enable their participation at a later stage helped mitigate some of the discomfort I felt. It is worth noting that the term recovery was contentious for several individuals in the study who acknowledged the ongoing presence of effects of abuse in certain aspects of their lives and preferred the terms rebuilding lives or living with the impact.

Survivors were invited to tell their stories with and about specific “artworks” that they had created, which had been significant to them in their processes of rebuilding their lives/living with the effects of CSA. The arrangements for the frequency, location (in person or online), and interval between interviews were agreed with and varied for each individual. The number of conversations varied between 2 and 4 and each lasted between 1 and 1.5 hours. Though exploring creative practice was the focus of the research conversations, inevitably individuals talked about aspects of their personal histories which evoked distressing memories. It is relevant that everyone who took part had had previous experiences of counselling/psychotherapy, and some were in therapy at the time of our conversations. Many were survivor-activists, who had chosen to be involved not only because of the personal significance of the research topic but because they wanted to “make a difference.” Several individuals had previously shared their stories with others in various public events to raise awareness and promote change. I use the term co-researcher for the individuals who have taken part in the study to reflect their status as experts-by-experience. The choice of term also denotes the collaborative intention of the research, the co-constructed nature of the data, and the aim of minimizing power differentials while acknowledging the impossibility of creating an equal relationship. Given the silencing that many CSA survivors experience, considerations about voice and power have been fundamental to the research journey. One of the main reasons for choosing narrative inquiry as the methodology for the study was that it upholds the uniqueness of individuals’ experiences and foregrounds their voices and stories (Bochner & Riggs, 2014; Clandinin & Connelly, 2000). The collaborative intentions continued throughout the study beyond the meetings, in the dialogical construction of written narratives where differences were transparently negotiated or made explicit, in seeking co-researchers’ responses to the common threads across stories, and in thinking together through ethical issues regarding dissemination and what would be reported (including discussions about anonymity, risks of identification, and redacting material).

The Research Relationship

The basis of all research is a relationship, this necessarily involves the presence of the researcher as a person ….. [which] must be made full use of. (Stanley & Wise, 1993, p. 161)

The traditional view of the researcher is of a neutral, objective, detached observer within a value-free “scientific inquiry,” having negligible impact on the research process. However, this perspective has largely been reconceptualized in the aftermath of post-modernism. Linda Finlay (2012) aptly comments that even if the researcher attempts to be “objective and non-directive, this very effort will have an impact, resulting in particular kinds of answers” (p. 324). There is a broad consensus in qualitative research that who we are as individuals, our personal backgrounds, characteristics, and social contexts will have an influence on different elements of the research process (methodological choices, research–participant relationship, interpretations, and presented work) and that research knowledge is therefore situated and contextual (see Letterby, 2013). Holstein and Gubrium (2016) have described research interviews as social encounters where information is actively shaped and formed between the researcher and the research contributors. We may try to bracket off our assumptions and preconceptions to get as close as possible to our contributors’ experience, but this is not something we can manage completely. The question then arises of how to conceptualize and respond to the influence of our subjectivities (and intersubjectivities) as we strive to produce research that is of value, trustworthy, accountable, and ethical. Our subjectivities can be viewed as “biases” that threaten the inquiry process and from which it must be protected (Hammersley & Gomm, 1997, Hammersley, 2011) or alternatively as something more complex where reflecting on their inevitable influence can produce valuable material and paradoxically make the research more objective through theorizing subjectivity and making this transparent (Letherby, 2003). Reflexivity then becomes an essential part of the research process (Bishop & Shepherd, 2011; Etherington, 2004; Finlay, 1998, 2002, 2012; Letherby, 2013). Most qualitative researchers accept reflexivity’s relevance to the research, even if it is only at the level of considering intersectional standpoint(s) and power differentials.

Reflexivity is also central to counselling practice, bending back our awareness on ourselves as active agents in the process. The client’s needs and beneficence are the core purpose of our reflections, as we attempt to gain a clearer understanding of what has been happening between us, what we may have missed, become caught-up in, and how issues from our own lives might be impacting the work together (Etherington, 2004). We learn about ourselves too in this process. Regular supervision from another counselling practitioner is an ethical requirement and provides opportunities to reflect in depth on all aspects of our practice in order to work as safely, effectively, and ethically as possible (BACP, 2018).

I am aware that foregrounding the relationship in research is, for me, a natural extension of the fundamental position it holds within the practice of counselling (see McLeod, 2019). Of course, research is not therapy and the goals for an effective counselling relationship (therapeutic benefit for the client) and an effective research relationship (deeper understanding and knowledge for the researcher) are different. However, there are useful transferable concepts, processes, and parallels. Just as across research traditions, therapy modalities have in general moved away from a detached, objective expert-driven position to upholding a more “authentic” and collaborative approach. I would describe my own counselling practice as an integration of person-centered and relational psychodynamic approaches, where both counsellor and client are understood to contribute subjectivities and templates (conscious and unconscious) from their own personal histories to what happens between them, as well as “realistic” responses. Michael Kahn (1997) powerfully describes how the unconscious aspects lay the foundations for “two hidden dramas [to be] played out in complex interaction” (p. 127), which makes the therapists’ reflexive discipline, ethical responsibilities, and supervisory discussions all the more important.

The relational turn in qualitative research considers the researcher similarly as part of the “data,” and the research process as involving the emotionality and intersubjectivity of the researcher and research contributors (Bocher & Riggs, 2014; Harris & Huntington, 2001; Hollway & Jefferson, 2013; Holmes, 2014). Research is construed as dialogical and co-constructed, with the researcher’s reflexivity an essential requirement, and their positionality (insider–outsider, intersectional identities, values, and personal qualities) as having influence (Reinharz & Chase, 2002). Reflexivity in research is not, as some critics claim, a self-indulgent or narcissistic process (assuming we do not privilege our subjective voices and overshadow those of our contributors). It is essential, I believe, in both research and counselling to help deepen our understanding of what is “going on” and as ethical practice, to ensure the safety and welfare of the client/research contributor and the therapist/researcher (Guillemin & Gillam, 2004).

Narrative inquiry is fundamentally a relational process, in which the researcher tries to build a relationship where co-researchers can recount and explore personal experiences in rich detail and depth. Doing so depends in part on the levels of trust and rapport that the researcher is able to create. The subtle interpersonal cues, registered by the co-researchers implicitly, of the researcher’s capacity to provide a safe, empathic, non-judgmental, emotionally responsive relationship and space, will have a bearing on what they disclose (Josselson, 2007). Of course, such cues on their own do not determine what happens in the research encounters, which are essentially shaped through what both individuals (contributor and researcher) bring to the relationship. Research conversations involve some similar tasks to those of counselling (Coyle & Wright, 1996; Dickson-Swift et al., 2006; Kvale, 2008): to create a safe, boundaried relationship, enabling self-disclosure and reflection upon experiences in the presence of a listener. Counselling skills (active listening) and ways of being (empathic, non-judgmental, and genuine) (Rogers, 1992) facilitate both research and counselling relationships (Coyle & Wright, 1996; Kvale, 2008). Though, clearly some types of interventions that are part of a therapy discourse (e.g., psychological interpretations) are inappropriate in a research context. Coyle and Wright (1996, p. 431) advocate “foster[ing] counselling attributes and use of counselling skills” in research interviews and question the ethics of engaging in sensitive research topics without being equipped in such ways to deal with the emotional content and potential distress of what may be shared.

I approached my first conversations with the co-researchers from a position of appreciating the similarities between research and counselling conversations and consequently the benefits my counsellor-self could bring to the research relationship. However, these similarities also have the potential to give rise to ethical tensions and boundary management issues (Ashton, 2014; Davison, 2004; Dickson-Swift et al., 2006; Gabriel, 2009; Hiller & Vears, 2016; Kidd & Finlayson, 2006). Kvale (2008) notes boundaries can become blurred, and draw us into quasi-therapeutic relationships, where the research focus is lost, and contributors and researchers may feel emotionally overwhelmed by the material that emerges.

While there are significant differences between research and therapy, in the reasons for initiating the contact, the purposes of the conversations, and whose needs primarily are being met (Birch & Miller, 2000; Dickson-Swift et al., 2006), participants in many research studies experience taking part as therapeutic (Dickson-Swift et al., 2009; Hewitt, 2007). This does not seem problematic to me, where therapeutic benefits are an unintentional biproduct for a co-researcher and the purpose of the conversations as research is clearly understood by both parties. In a debrief at the end of our first meeting, one of the co-researchers, Jeanie, made a direct link between her experience of taking part in the research and therapy:

So, it feels a bit like the, um, therapy process, which is good. You know that kind of like, “Oh yeah, that’s ……” [realization] Like, you reflect back to me what I said, “Oh yeah. Okay!” (shared laughter) I don’t normally speak about it that much. I’ve done, like, one talk, I think, to people, um, but yeah, so that’s …. that’s what’s quite exciting and interesting for me. The questions are making me think.

Holly too recognized some parallels, saying this in a check-in at the beginning of our second meeting:

It’s easier to do it this time. I think, you know, we met last time and now there’s more familiarity between us. I do …. now that we’re doing this process, I do quite want to come back and show you some more, because it feels like now we are in a process, which is different to therapy but feels like it’s in a process, like it has a form and a direction—a progress. So, I would quite like to continue with it as we have been. I mean ….. it’s always good to keep looking at it, because, because it’s not a circle, is it, it’s a spiral. And it’s helpful for me, ‘cos I need to keep healing from it, um, so it’s helpful and useful and I might contribute to something, you know, that’s useful and impactful.

Though in this statement Holly explicitly identifies therapeutic motivations for continuing to review her artworks in our meetings, both of us were clear about the research aims and focus. However, the expectations of contributors may not always be the same as ours or conscious, especially when it is known that the researcher is a therapeutic practitioner (Hay-Smith et al., 2016; Long & Eagle, 2009). Both therapy and research practices are vulnerable to misuse of power (persuasion and coercion). The power differentials signified by a practitioner role and therapeutic misconceptions may leave a research contributor less able to protect themselves from harm in the research context (Hay-Smith et al., 2016; Hiller & Vears, 2016). Duncombe and Jessop (2014) critique the lack of attention to power inequalities in research interviews in the use of counselling “techniques” for “doing [my emphasis] rapport,” potentially leading a research contributor to reveal more than they intended to the researcher or to themselves, which they might later regret. They describe this as “breaching the interviewee’s right not to know” (p. 112). These risks may be heightened in CSA research, where the betrayal of trust experienced through the abuse can create longstanding difficulties in managing closeness and distance and setting appropriate boundaries, and where memories may be unclear or fragmented (Sanderson, 2013). It is dangerous to assume that each contributor has the same level of awareness, agency, and capacity for ensuring their safety in the research interview. At the same time, we must not assume a standard power ratio in the researcher/contributor relationship; power balances can shift, and researchers too are vulnerable. Similarly, we must not dismiss an individual’s capacity to exert control and protect themselves. Fundamentally though it is the ethical responsibility of the researcher to continually consider and monitor the impact of inviting intimacy in research projects (Birch & Miller, 2000).

Ethics of Care

Attention and adherence to ethical principles of non-maleficence, beneficence, consent (autonomy), confidentiality (within clear parameters), and fidelity (integrity, respect, honesty, and trust) underpin both research and counselling relationships. Obtaining informed consent and compiling signposting resources before embarking on the research is only the first stage of ethical safeguarding. It is highly likely that unforeseen situations will emerge as the study progresses. Understanding consent as a continual process allows the impact of taking part to be reviewed at each phase of the research and as issues surface. Ethical dilemmas in narrative inquiry often arise from the tensions between the dual responsibilities of the researcher to protect those participating from harm, while seeking depth of material to produce research of “good” quality and standards (Josselson, 2007). Researchers who are also practitioners in the helping professions may experience a pull to act as practitioners in such situations and have to find ways of managing these role conflicts and boundaries (Ashton, 2014; Etherington, 1996; Gabriel, 2009; Kidd & Finlayson, 2006). Difficult ethical decisions must be made in situ, requiring “experience-based situational judgement, clear perception and proper attention to the particularities of the situation” (Brinkmann & Kvale, 2005, p. 170).

In counselling and therapy, client distress is understood to be part of the process of change and healing rather than being perceived as inevitably harmful. Within the containment of the counselling relationship and space, painful emotions can be safely acknowledged and understood. However, for all clients, and particularly for those where there has been complex trauma, the work of counselling needs to develop in stages, firstly establishing safety and taking into account the pacing of sessions, issues of power and control, boundaries, and individuals’ capacities to regulate their emotions in order to avoid harm (re-traumatization) (Rothschild, 2000; Sanderson, 2013). As ethical researchers, we must similarly create a safe space for our research conversations, where risks of distress are minimized, but I believe that we should not assume it is inherently “harmful” if individuals become distressed when reflecting on their experiences (Hollway & Jefferson, 2013). Some distress may be a very normal contextualized response to the topic (Morse et al., 2008), and affective expression can be an indication that the individual feels secure enough in the research relationship to go to a deeper level of exploration. A number of research studies assessing the impact on survivors of taking part in sexual trauma research (Burke-Draucker, 1999; Campbell et al., 2010; Carlson et al., 2003; Kennedy-Bergen, 1993) found that the overwhelming majority of individuals described their experiences positively, using words such as helpful, therapeutic, cathartic, validating, empowering, and insightful, even though at the same time they might have become distressed or found the conversations painful. Our ethical and moral responsibilities demand that we respond as sensitively and empathically as we can, within the parameters of the research relationship, when we notice a research contributor becoming upset in a research interview. We must offer an opportunity to pause, to negotiate with them if they need to stop, stay alongside them until they regain composure, refer to external support if the level of distress requires this, and check in with them later. As Morse et al. (2008) note, emotional expression in itself (depending obviously on the severity) is not the cue to stop an interview—which could also be experienced as harmful (rejecting/invalidating)—but a cue to consider the option(s) together. The onus is upon the researcher to ensure an ethic of care (Gilligan, 1982) alert to the potential of harm, comprising attention and responsibility to their research contributors and the competence to respond appropriately to their distress.

Compared to counselling, one of the challenges in qualitative research interviewing in sensitive topics concerns the time frame, which does not allow for the foundations of a trusting relationship to develop before focusing on the subject matter, and in some cases exiting shortly afterward. Etherington (1996) describes having to put aside a part of her self-as-counsellor in order to cross those thresholds earlier than she would have in counselling, when undertaking research interviews with male survivors of CSA. I too have experienced some discomfort in relation to this. Invasion of boundaries is obviously a central element of sexual abuse. “Trespassing the person” (Fog, 1984, as cited in Brinkmann & Kvale, 2005, p. 169) carries the potential of replicating that sense of intrusion for survivors and adversely affecting the research relationship. My counsellor-researcher selves aligned in thinking how best to begin our conversations to create a safe-enough base for the research work to take place. Foundational elements from counselling practice such as contracting for the work together and not rushing in to tell trauma stories were my guide for the first meeting. Re-stating the research focus (creativity and rebuilding lives), clarifying that co-researchers could stop at any time, could indicate anything they didn’t want to talk about, and didn’t have to go deeper than they felt comfortable to do, was the grounding for our discussions. I tried to ease into our conversations by starting with a “safer” topic, only to appreciate what might be safer for one person could be threatening for another. For instance, asking Susannah to tell me about her creative activities in general plunged her directly into childhood memories:

Okay. Yeah …… immediately “Boom!”, that’s how unfortunately I think of it, like that ….. right like lots of things at once ……

I was also aware that in my concern to provide a trauma-informed, safe space, I might be overly protective or cautious, and that I shouldn’t underestimate my co-researchers’ capacities to decide how they might want to start, and what or how much they wanted to say. For co-researchers Lily and Mary, it was important to tell the stories of what happened to them (the abuse) right from the outset. With Lily in particular, I felt she needed me to hear and be a witness to her story before we could look at how writing had been helpful to her. With Mary, I began by asking her to tell me about her creative practice, to which she responded:

Mary: I can’t really talk about the artistic process without also going into the abusive experiences. So, I’m happy to roll them together.

Alison: Yeah. If you’re happy to do that, that’s fine. I’m just aware that obviously this is the first time we are meeting and, you know, delving into that with a relative stranger can feel quite difficult. But if you are okay .. then, yes, that’s fine.

Mary: Yeah, I’m quite well versed in doing that because I’ve worked as an expert-by-experience and I’ve done this kind of thing before, basically, so it’s … I’m fine … yeah.

My experience as a practitioner facilitated making judgments to steer away from deeper explorations in the research conversations when I intuitively felt to continue would have been damaging. The cues I picked up in my first meeting with Alba alerted me not to go into their history or ask questions about a specific piece of artwork which they had brought along while still working on it. In the debrief at the end of this meeting, they said:

It was a nice way to introduce it. I think to have started with the abuse would have been like, urghh … argh … yeah, too much. It was nice that you were sensitive to that piece [of artwork] just feeling like this is quite fresh, you know … and because of that I feel more able to do this.

The line between when and what is okay to explore and when to steer away isn’t always clear for the researcher or research contributors. Abuse dynamics and power differentials can make it harder for some survivors to keep within tolerable bounds of what they share in research conversations and to show or indicate when they might need to stop (Castor-Lewis, 1988). Several co-researchers commented in the debriefs that they had experienced some level of dissociation in our conversations, or that they found themselves more emotional than they had expected, but this had been manageable and through telling their stories in the past tense they had been able to acknowledge how far they had come and how their lives had changed.

Alison: It’s hard talking about this; do you want us to move on to something else?

Susannah: No ….. I don’t mind. I’m sorry I’m upset.

Alison: There’s no …. you don’t have to apologize … it’s just … I want to make sure you feel okay with this.

Susannah: No, no, no. It’s not like that. I am sad about it …..

Alison: Yeah. It’s really painful.

Susannah: … yeah, but you know … things have changed ………….

[She talks about the ways in which her life is different now. Then after a few minutes she comes back to being upset.]

I knew …. I’m surprised at how emotional I got. I didn’t …. I thought I’d be quite shut off from it, almost ….… um, but that’s not necessarily a bad thing, I think. Um ….. Yeah. I wasn’t robotic about it …

Alison: No.

Susannah: ….. and that’s okay actually. I think things aren’t going to impact me and it does and I’m okay ……. I’m really okay with that. And sometimes I’m a bit like, “Oh, that’s not a problem!” um …. but I’m learning that … well, it’s more like, I don’t want it to be. And it’s not a problem ……. it’s okay …… it really is okay.

At other times, my sense of balancing along the explore/steer away divide felt more precarious. With a co-researcher, Nancy, I worried that I had allowed my “researcher-self” to take over when being shown some powerful artwork which I immediately identified as rich material for the study. When Nancy didn’t respond to my email sending the transcript of that meeting, the concern that I had lost the moderating observations of (what at that point was attributing to) my counsellor-self started to niggle away. Had my researcher-self probed too far? Had our discussions about these images, which Nancy had mainly kept private, been triggering, drawing out more than she had been ready to share or think about? Were power dynamics being replicated in the research relationship and disenfranchising her sense of control? I felt I had sidelined the counsellor part of myself and went into a place of disquiet, shame, and self-blame, scrutinizing my practice, reproaching myself for not being sensitive enough to recognize “the signs.” Holding on to a sense of not knowing in this interim absence of contact was challenging. When we had a discussion, she was able to tell me that while she had wanted to share her drawings at the time, she had felt shame and embarrassment afterward, which had in part led to the break. I was able to bring my counsellor skills back to the fore, to acknowledge my appreciation that she could tell me how she had felt, to say I was sorry that showing her drawings had impacted her in this way and remind her that the shame was not hers. Something was restored through this conversation, and in fact it led Nancy to take the drawings to her therapist for them to work through together. In discussing her written narrative (story-ing) for the research, Nancy said:

I found it really helpful to read the writing you had done. It helps me feel a lot more compassion towards myself, which isn’t something that comes easily to me. It’s nice to see my contributions and I feel hopeful that my art and what we talked about will help people realize how important creativity can be for overcoming and processing child sexual abuse.

Just as in therapy, it is important to be proactive in attempting to repair any ruptures that occur and re-establish a working relationship (Safran & Muran, 2000). It would be unethical not to. In retrospect, I can see there was another rupture that needed reparation. I polarized my researcher-self as “bad” (appropriating and intrusive) and my counsellor-self as “good” (if negligently absent initially) in this situation, creating a false dichotomy. The skills I attributed to my counsellor-self are also integral to being a “good” researcher. Both counsellors and researchers are fallible and will get things wrong at times. The most important consideration as a counsellor and as a researcher is to do no harm, and for that we need to keep our work under continual scrutiny, developing both our anticipatory awareness and attunement to what may be happening in the here-and-now of the relationship.

And though in many ways, as Gabriel (2009, p. 149), I might “aim to be researcher first” in our conversations, it is clear that my counsellor and researcher selves have attributes and skills in common and are alongside each other. One of the recurring metaphors in the literature is that of changing hats or glasses between researcher–practitioner roles as needed. Like Hay-Smith et al. (2016), I believe that the roles cannot be so easily separated and shed and that their duality might be more accurately described as “a coherent moral identity that recognises both sets of obligations, rather than oscillating between the two roles” (p. 12). I see the necessity of staying uncomfortable as a researcher. It serves as an ethical compass for navigating the research journey. The relationships with our research contributors and clients in therapy are messier and more entangled than the research articles or case studies which we read often convey. I believe it is necessary to be reflexive and alert to what can be evoked for both participants and ourselves as researchers in order to uphold ethical responsibility, judgment, and practice (Etherington, 2007; Josselson, 2007).

Multiple Selves—Shifting Positions

The emotional work inherent in researching sensitive topics inevitably affects the ways in which the research process and the evolving researcher–co-researcher relationships are experienced and interpreted by both parties (Dickson-Swift et al., 2006). Carol Warren (2002) suggests that the positions from which researchers and research contributors speak to each other shift and are not based in “stable and coherent standpoints,” or single aspects of identity (p. 84). Other selves are created or activated in the research through the perceptions and interactions with the research contributors (Etherington, 2001; Finlay, 2012; King & Horrocks, 2010; Reinharz, 1997). Finlay (2012, p. 324) cites DeYoung’s (2003) description of the therapy relationship as a “thickly populated” encounter and suggests that this is also the case with research interviews. Different aspects of our intersectional identities, personal histories and characteristics, both overt and hidden (implicit or unconscious) can come to the fore and dominate the perceptions we have of each other and our interactions at different times just as in therapy (Holmes, 2014; Safran & Muran, 2000). For many survivors, a common impact of trauma has been to fragment and split off aspects of abuse experiences and memory to adapt and survive (Fisher, 2017; Sanderson, 2013; Van der Kolk, 2014). I was aware, for instance, with co-researcher, Elizabeth, that sometimes her pronouns, verb tenses, and language style would shift and her “child-self” would be more present in our conversations.

An aspect of my own self that troubled me periodically in the research has come from not being a CSA survivor. Being an “outsider-researcher” heightened concerns about being exploitative: an anxiety that I might replicate appropriative abuse (power) dynamics within the research process, in our conversations, in my interpretations, and in the research output. Earlier I described how, in attempting to assuage the uncomfortable feeling of being appropriating and intrusive, I split my counsellor–researcher dual role into “good” and “bad,” identifying with one and disowning the other. Researching emotive subjects such as CSA can challenge our capacities to stay rooted in a grounded stable position.

Reminding myself of countertransference dynamics that can arise in therapy, where the counsellor can experience themselves as, for instance, the wished-for caregiver who protects; the neglectful caregiver who doesn’t, who abandons or doesn’t see; and the hostile perpetrator who abuses and exploits (Sanderson, 2013), helped to make sense of some of the more difficult emotional responses and anxieties I experienced as the researcher. With Nancy, I felt as if I had been both the neglectful-researcher and the exploitative-appropriating-researcher. It would be easy to categorize countertransference as projective identification—where feelings are “put into the researcher” that relate to the co-researcher’s relational templates and history alone (Holmes, 2014, p. 170). Such interpretations may conveniently dispel our responsibilities as researchers for our mistakes, neglect our own subjective contributions/projections, and reinforce unequal power relations (Parker, 2010). Taking an intersubjective stance (Holmes, 2014; Kahn, 1997) allows for a deeper and more nuanced understanding of what is co-created dynamically between both individuals in the research dyad. In reflecting upon the sense of shame I experienced, I could see how this mirrored Nancy’s feelings of shame after our conversation and the introjected toxic shame that is so self-attacking in CSA trauma. But I also recognized a parallel “Critical-self” of my own had risen up, familiar from childhood, who held little compassion for “getting things wrong.” In my concern to “get it right” and avoid identification with the neglectful or exploitative researcher positions, I could also overcompensate, be overly protective, undermine co-researchers’ capabilities and autonomy, and infantilize them (and a part of myself). I needed the reflexive functioning of my internal supervisor (Casement, 1985), standing slightly apart to review, contain, and process the discomfort and stay with the uncertainty. Through this and the support of my supervisory team, I could shift back into a more grounded and constructively engaged position. Without making opportunities to reflect in these ways, we can find ourselves pulled into unhelpful enactments, which may be damaging for our research relationships.

At points, I wondered if I was hiding behind my counsellor training as some sort of self-protection. For Holly, repeating the phrase that I was a counsellor seemed in itself to offer her some reassurance like a soothing rallying cry. When she was about to show me some artwork which she anticipated might be disturbing to us both, she would say:

But I know that you’re a therapist, so I know you can cope with that.

Perhaps it offered a similar bolster to me. I was aware in some of our meetings co-researchers would be concerned about the impact upon me of telling their stories:

Sophie: … I worry about you, now. I worry that it’s so dark that it can go into you, and latch on to you, and that you’re carrying that. And I don’t want you to carry that. [She talks more about the worries she had in the past of damaging her therapist by talking about the abuse.] That’s how great the fear was.

Alison: The terror of how dangerous it felt putting that into words?

Sophie: Yeah.

[I feel the need to reassure her I am okay, that I can hear and am not endangered by what she has shared.]

Alison: I need to let you know, I’m fine. I hear these stories in my role, horrible traumas, and it’s awful that these things happen to people and it touches me, because it is awful …….. but I’m fine, I have my own support mechanisms, my training, supervision, for looking after myself ……

Sophie: Good!

In undertaking research in sensitive topics, as well as the emotions in our research contributors, we need to manage, contain, and make sense of our own responses as researchers to what is evoked for us, our self-as-person, in order to remain fully engaged and listen (Dickson-Swift et al., 2009). Mostly my counselling experience has enabled me to do this. But earlier in this second meeting with Sophie, I lost the sense of time and had become disorientated about how long we had been talking. I thought we were reaching the end of the 90 minutes that we had scheduled when in fact we had only met for 1 hour. For a while, this resulted in some confusion in our meeting as I drew attention to our time (as I thought) being about to come to an end, before realizing my error (which Sophie noticed) and getting back on track. Perhaps this was also a catalyst for Sophie’s concern (above). On reflection, I think I had needed to “step away” and protect myself momentarily from being overwhelmed by her harrowing story and my awe at what she had survived, her courage, resilience, and determination despite all the fears. How could I not have been affected and touched by what she had so openly shared with me. Through my experience as a counsellor, knowing how hyper-alert many survivors are to incongruities in another’s actions, non-verbal cues, and words, I understood how important it was to be genuine and human in acknowledging what had happened in this situation. Discussion of what happened became part of our meeting and later part of the written narrative of her story-ing, shared and co-constructed with her. During the writing of this article, I also became aware of the parallels with the “therapeutic hour,” a connection I hadn’t made until this point. In that moment of overwhelm, it seems I may have lost clarity about the research focus of our conversation and intuitively fallen back on the containment of that familiar practitioner framework.

Lynne Gabriel (2009), herself a therapist, describes aiming to hold a position of compassionate distance to manage the researcher–practitioner role conflict, the term encompassing both a boundary and an empathic connection. This makes sense to me as a representation of the ethical researcher’s positioning in the relationship, but at the same time I wonder if it misses something of the complexity of the intersubjective relationships. Given the power of what can be evoked, the sense of impingement and disturbance which we can experience when talking about sexual abuse, the co-produced relational dynamics can be hard to untangle. The researcher’s reflexivity, in this case “with a [counselling] accent” (Hollway & Jefferson, 2013, p. 157), and the constant monitoring of the “process, content and form” of interactions (Gabriel, 2009, p. 155) are essential to hold on to, or to re-attain the position of being both with and separate; appropriately boundaried rather than “caught-up” or enmeshed.

Conclusion

The potential threats of engaging in researching sensitive topics such as CSA are real and must, as Lee and Renzetti (1993) assert, be confronted seriously. Issues of boundary management are central to our ethical responsibilities for the care of both research contributors and researchers. As researchers, we bring multiple selves to the research conversations. In undertaking this project as a counselling practitioner-researcher, I have viewed my counsellor and researcher identities as offering a collaborative partnership, rather than being in conflict or necessarily blurring boundaries. The contextualized knowledge and experience from my counselling practice has facilitated more nuanced and containing responses (in general) to the challenges and complexities of the research process and to the researcher–co-researcher relationships. This does not dismiss the discomfort that can arise in holding these two identities alongside each other, or the internal (and external) negotiations that need to take place when their priorities conflict, or the potential to make mistakes. As researchers, we need our critical reflexivity to continually monitor and understand what is happening in the research process, to help us to respond appropriately and to take reparative action when we might have got things wrong. Above all, an ethics of care must be at the forefront and must take precedence over other research priorities. Engaging in researching any sensitive topic brings with it implications for researcher training and support, to be able to respond appropriately and sensitively to what may emerge. I advocate detailed preparation including developing topic-based awareness and knowledge, reflexive work exploring motivations and possible personal impact in undertaking the work, interview training including using recordings to reflect upon/develop skills and identifying strengths and support needs. We do not stand outside the research process. Debriefs for researchers are just as necessary as for research contributors. The presence of a knowledgeable, supportive supervisory team (and potentially additional clinical supervision) where the emotional impact of the work and our vulnerabilities can be openly explored is essential. The anchors of journaling and the presence of my own internal supervisor in the here-and-now of the research encounters have also been indispensable reflexive tools in navigating these challenges. I believe the depth of our research knowledge and our ethical care for our participants both benefit from our capacities for intersubjective reflexivity.

Just as in a counselling relationship, it has been important to me to create an ethic of being alongside my co-researchers in our conversations, a “wit(h)ness.” I never felt I was doing rapport or using techniques, not “don[ning] a therapist [or researcher] mask … voice … posture … or vocabulary” (Kahn, 1997, p. 163) or “hat” or “glasses” (Hay-Smith et al., 2016), but rather responding authentically within a containing (mostly) and boundaried relationship and, as Stanley and Wise advocate, making full use of the researcher-as-person who is touched and moved by what has been shared.