Ethical Autoethnography: Is it Possible?

Abstract

Autoethnography is a widely applied qualitative research method to examine self-experience in relation to life events, and also situated experiences in cultural and institutional contexts. In this paper the ethical challenges arising in conducting and presenting autoethnographic research are presented and explored, first through reflection on personal experience of being described and identified in an autoethnographic presentation without my permission, then through the challenges of my own experiences undertaking autoethnographic work. Following Ellis’ relational ethic as a third dimension along with procedural and situation ethics, a fourth dimension of the ethic of the self is presented. Ways we can enhance the ethic of respect in autoethnography is further elaborated.

Autoethnography is a highly regarded and widely used research methodology and practice whereby the researcher is deeply immersed in self-experience while observing, writing, journaling and reflecting. “A researcher uses tenets of autobiography and ethnography to do and write autoethnography. Thus, as a method, autoethnography is both process and product” (Ellis et al., 2011, p. 273). The development of autoethnography emerged from ethnography, with the work of Carolyn Ellis and her collaborator Art Bochner catalytic in its burgeoning and impact. There are examples of early ethnographic work in which researchers focused on self-experience rather than primarily on descriptions of their observations of others. This was not a common practice but the small number of examples align with the contemporary practice of autoethnography (Reed-Danahay, 2002).

Autoethnography recognizes self-experience as a social phenomenon valuable and worthy of examination. Autoethnographic research seeks to deepen understanding of multiple complex dimensions of culture and interpersonal dynamics in such locations as, for example, in a community (Schmid, 2019), an organisation (Murphy, 2008), or even within family dynamics (Lahman, 2020). Troubling experience is often the focus of autoethnographic study, such as the personal experience of loss through bereavement (Furman, 2006; McKenzie, 2015), or workplace bullying (Pheko, 2018).

Autoethnography can be undertaken with temporal congruence or retrospectively, whereby the decision of the researcher to engage autoethnography occurs at a later date than the described events. There is an emerging practice of collaborative autoethnography in which a group of people experiencing a similar phenomenon undertake research and report collectively (for example Reyes et al., 2020). A further descriptor is organisational autoethnography, sometimes termed workplace autoethnography (Lee, 2018) in which the researcher focuses on their experiences of an organisation or institution to reflect on and expose culture and practices (Herrmann, 2017).

Writing autoethnographic accounts of self-experience necessarily involves others. In his (self-described) manifesto on autoethnography Bochner identified that “[h]uman beings are relational beings, and thus every story of the self is a story of relations with others” (Bochner, 2017, p. 76). As Tolich (2010) described in relation to problematic ethics in autoethnography, “…the self is porous, leaking to the other without due ethical consideration” (p. 1608).

Claiming autoethnography as a queer methodology Adams and Holman Jones (2008) also reflected on the relational dimensions of autoethnography and concluded that “[s]uch work views identities as relational accomplishments: manifestations of selves that shift and change, that must be negotiated and cared for, and for which we are held personally, institutionally, and ethically responsible.” (p. 374). This relational aspect raises dimensions of care needed to conduct autoethnography when others, by the nature of our relational lives, are likely to be referred to in the final published document.

Ongoing reflection is needed to consider how including others within an ethic of research practice is provided appropriate attention within autoethnography, to ensure that accountability and care are situated at the centre in responsible research. The concern guiding this paper is the extent to which autoethnography requires negotiation of consent for inclusion from people described in the published work.

An Ethic of Autoethnography

This reflection on issues in the ethical conduct of autoethnography is prompted from multiple experiences. Currently I lead the research office in a large Faculty in a university. About a year before submitting this paper my team and I organized, ran, and attended a research conference held over a week to provide a showcase for research. One paper focused on the presenters’ disappointment as to current neoliberal management within universities. The audience were advised that the identities of all people mentioned in the paper were disguised because the presenter had worked in multiple units within the university where the presentation was being given, and had worked elsewhere. A short time later one of the presenters described a situation in which she felt demeaned and mistrusted due to the requirement that she seek a signature on a form. She named me as the signatory while pointing at me. I did not challenge her contradiction between her intention not to identify anyone, and her identification of me. However, sometime later she and her colleagues published an autoethnographic paper about the problems of the neoliberal university. It was claimed that the requirement to seek my endorsement was evidence of the institution having demeaning, irrelevant, and officious procedures which impacted the autonomy of academic staff. My team and I were easily identifiable in the account. I would have liked the chance to offer a corrective to some of the claims made, not because I am somehow a vulnerable person hurt by the accusations made, but because it would have been quite easy to show the authors that some of the claims were exaggerated, and some untrue. I felt silenced and judged. I had no way to offer a counter-narrative or further context.

As an example of the challenges when including others in such accounts, I recall my discomfort when trying to publish an autoethnography about my work at a pediatric teaching hospital. I started my PhD while working there, and others have since told me that I was the first doctoral student in the department to undertake research using qualitative methods. I often felt out of place as a young female, focused on psychosocial rather than medical care, and although consistently treated with kindness I often felt distanced from the work of others, for example researchers in genetics, or animal-based research. My consistent over-riding experience was that I did not fit in as a researcher focused on evaluation of relational therapeutic practice with children and families.

Some years later I reflected on experiences from this period through a retrospective autoethnography I wrote while on sabbatical leave at one of the world’s oldest universities. My use of autoethnography was prompted by encouragement from Professor Carolyn Kenny (Kenny et al., 2004), a leading Indigenous scholar who later became a dear friend. I was visiting as a research fellow in a predominantly male, science-oriented area, and while people were welcoming and kind, somehow this intellectual environment elicited memories of formerly experiencing myself as an out of place scholar, trying to find ways to describe my work to others who had never heard of my discipline or the research approaches I use.

I wrote and presented the outcome of this autoethnographic memoir several times, including at a seminar at my sabbatical home, but I never submitted it for publication. Presenting in other countries felt relatively safe to hide identities—especially as it was a time before social media was so prevalent—but I could not find a way to adequately anonymize the people I was describing in order to present the work in a permanent form. I did not think it possible to attain consent from my co-workers from so long ago, and definitely not from the patients to whom I referred.

I was recommended to develop a fictionalized account of my work at the hospital by the editor of a qualitative methods journal I met at a conference. I tried but it was not a comfortable process, and I choose not to publish. I cannot claim the experiences are unable to be fictionalized as I have published case material which is based on my clinical practice but has no reference point to an actual person (e.g., Edwards & Parson, 2017), known in healthcare as composite cases, and usually not requiring ethical clearance for scholarly publication. However, fictionalizing my retrospective autoethnographic research may be more difficult for me than for some others. As one of only two music therapists in Australia working with hospitalized children and their families at that time, and one of only a few qualified music therapists work working in a burns unit worldwide, attempting to locate the work elsewhere, to completely re-work experiences of interactions with staff perhaps by combining characters, or somehow changing children’s identities through composite cases might not have worked as I wanted to report what actually happened. Composite cases usually combine events to make an exemplar case, my recall of actual events and persons in context could not be treated in this way. It might be suggested that families would be unlikely to remember their therapist from that time decades prior, or my name, but recent personal experiences contradict this.

Our Responsibilities When Referring to Others in Autoethnographic Research: Ellis’ Relational Ethic

My concerns about the autoethnographic report by my colleagues which mentioned the need to gain my signature did not negatively impact my own work identity but a professional staff member in my team was upset. A link to a report of the material included in the presentation was circulated in the university newsletter, which the team member opened and read. She recognized herself in some of the descriptions, and she reported her apprehension about their account of events to me. I am concerned that she had to bear a burden of upset for descriptions of her role in procedures she has little power to influence or change.

How then might an ethic of autoethnography be located in relation to reference to others in accounts of self-experience? Ellis (2007) identified two well-known ethical dimensions—procedural ethics and situated ethics—and described a third as relational ethics. She proposed that process consent, checking in with the person that they have understood how their interactions will be represented, is necessary (Ellis, 2007). However, Tolich (2010) noted with concern that Ellis (2007) had not followed this advice in relation to publishing an autoethnography about her mother which she did not show her, and read excerpts of a further autoethnography to her mother but skipped over parts she thought might upset her. At the same time it must be acknowledged that Ellis (2007) described her unease, along with the complexity of issues that arose for her, as she grappled with her decisions about what needed to be done to ensure adherence to a relational ethic.

How consent is negotiated and further processes of good practice are contextual and cultural thereby challenging the guidance process for ethical conduct in autoethnography as reported by Ellis (2007) and Tolich (2010). In describing Indigenous Autoethnography Michele Bishop, a Gamilaroi woman, described that the auto in autoethnography has a different origin and ethic for her and other Indigenous scholars,

…my knowledge is not just coming from me, or from books and articles accessed because of my exclusive university library membership. My knowledge primarily comes from my family, my communities, my connections. My “self” belongs to them. Therefore, I must constantly be reflecting on “Who do I speak for?” “Whose stories and knowledges am I able to share?” alongside, “What am I speaking for?” and “Who am I speaking to?” (Bishop, 2020, p. 6)

Autoethnography cannot be one fixed and unassailable method but instead requires a potentially messy, fluid, and highly contextual approach to exploring and understanding self-experience in context. The constant questioning of the right to speak about or even for others as described by Bishop (2020) requires self-interrogation, deep reflection, and a responsibility of integrity. Research involving human participants is also bound by institutional or national research ethics requirements, and this might not be over-ridden by autoethnography’s intent to focus on self-experience.

Lee (2018) proposed that without guidance for new researchers, the ethics of autoethnography can expose a minefield; what Dilger (2017) more agreeably described as necessitating muddling through. There is always the possibility in writing autoethnography that “people in narratives become fixed…which can have serious and far reaching consequences for the author and their characters” (Lee, 2018, p. 311). Some have described this portrayal of others in autoethnographic narratives as betrayal (Tolich, 2010). When writing about others “all others have rights over how they are represented regardless of any apparent consents they may have given at the outset” (Andrew & Le Rossignol, 2017, p. 245).

I am aware of how it feels to be included without permission in someone else’s work after finding myself and other team members described in an autoethnography referred to in the introduction of this paper. I perceive I was instrumentalized in the description of my needing to sign a form as a proxy for actual evidence of a. managerialism and a. removal of agency from colleagues, resulting in their victimisation. In the othering and dehumanization of me as university management, I had no opportunity to clarify, or to assist the concerns of these colleagues. My attempt to help procedures in the academic unit run more smoothly, and to communicate these was perceived quite differently than my intent or responsibility. I wish the authors had come to me to explain what they found problematic and upsetting about the signature I was required to provide. They may have been surprised to find there was no new and additional procedure as they assumed and reported. The signature of a senior academic leader was always required, the new procedure simply transferred the requirement for a signature to an alternate senior staff role.

It might be perceived that the primary responsibility for the ethical conduct of the research—whether a retrospective or contemporaneous autoethnography—lies with the autoethnographer. However, even with consent of those described the autoethnographer might experience the similar challenge as in ethnography, whereby the researcher expects that participants know what they are getting themselves in for, proposed as one of Fine’s (1993) 10 lies of ethnography.

It is almost impossible to know how one’s autoethnographic work will be received, and/or how, if people recognize themselves in the narratives, they will feel and, in turn, respond to us. This is an issue of concern in autoethnographic practice but is especially troubling with regards the retrospective autoethnography, published with no consent or knowledge of those described, and in turn publishing takes control out of the author’s hands (Lapadat, 2017). However, accounting for challenging unworkable situations and the actors within them a means by which organisational autoethnography can be a powerhouse of revelation, especially with regards problematic workplace practices requiring remediation.

An Ethic of the Self

A further dimension is the ethic of the self, foundational to autoethnography. The researcher has an obligation to describe and investigate their own experience authentically. However, it can be painful to recall difficult past events. With reference to an ethic of the self, opportunities to consider whether harm can be caused by in-depth personal revelation is prescient. Rambo (2016) described multiple points of career and reputational risk in revealing via autoethnographic inquiry her employment prior to becoming a university professor, including work as an exotic dancer.

This risk to reputation might also be considered alongside the potential mental health risk through unexpected challenges arising from in-depth introspection about experiences which might lead to rumination. Lee (2018) referred to the introspection in autoethnography as unending. How these aspects are navigated in autoethnographic writing is potentially challenging, negotiated anew each time we approach topics with this method.

The events reported through autoethnography happened in our daily ongoing lives, they did not happen in some separately located space of research such is the case in interview-based research where we inquire about the experiences of others and listen deeply. These experiences related by others in interview may move us, but they are not our own experiences in the moment of real time, experiences in our bodies and minds, and re-experienced in memory through retrospective autoethnography.

Until recently I was a leading trainer in the therapeutic approach I described in the unpublished autoethnographic account of my experiences as a healthcare practitioner. I have since wondered whether publishing about my vulnerability in relation to feelings of uncertainty in some aspects of work with children and families might have unintentionally caused reputational harm to me, or a loss of confidence in my skills by students in the learning environments I am employed to facilitate, or for families with whom I might work in the future.

Through autoethnography multiple relational experiences between myself and patients, and tensions between my role and other professional roles, exposed aspects of a power dynamic of which I was not aware when mired in the demands of professional work. When describing and reflecting on experiences alongside colleagues or in interactions with patients and their families where I was exposed as making a miss-step, and my colleagues were described as unfeeling or unhelpful in some way, there is potential risk for my professional reputation. Mutual trust, crucial to team-based healthcare, might be potentially impacted. What if I join another therapeutic team in future and they perceive I am observing them and will write about our interactions critically?

Balancing Pitfalls, Opportunities and Risk in Autoethnographic Research

Key responsibilities and objectives within an ethic of autoethnography include allowing the voice of the author to be heard. To silence this voice because descriptions of others’ culpability in causing the author discomfort or distress might expose these others to a type of harm is a difficult balancing act with regards the competing claims of the right to be heard and oppressive silencing (Lee, 2018). Techniques such as masking may be considered useful to ensure the context or colleagues are difficult to identify (Jerolmack & Murphy, 2019) but in some instances it might be impossible to achieve this.

Rambo (2016) recounted how an autoethnographic presentation that won an award was not able to be published because of potential risk to those she described. She was advised to create a fictionalized account as a story or play, rather than an academic peer reviewed paper. In a related experience of publishing a highly personal account, Anonymous (2019) could not use her name as author as her autoethnography referred to experiences of her parents, who would lose their anonymity by publication of her name. It is unlikely these are isolated experiences.

A further pitfall is regards the proposition that the autoethnography is an artful construction designed to “represent the author in a particular way” (Walford, 2020, p. 5). Armstrong-Gibb (2019) mused as to whether this expectation of self-revelation might promote a culture of inflated self-importance of the researcher, similar to Walford’s claim of autoethnography as pure self-indulgence (2020), perhaps where the researcher’s self-reflections are not particularly interesting, or at worst reveal a shallow narrative whereby a person uses the process to whinge about their circumstances. As Hackley (2020) described with regards organizational autoethnography, when the writer’s expression of consciousness of events reveal endless self-absorption, it is not able to be hidden from the reader.

How Do Ethnographers Negotiate Issues That Can Arise When Writing About Others?

There are well-documented experiences of ethnographers disguising participants where the anonymity of their communities was not able to be maintained post publication. Arlene Stein’s research was reported in The Stranger Next Door (Stein, 2010), Nancy Sheper-Hughes’ research focused on mental illness in a rural Irish community (Sheper-Hughes, 1979), and Carolyn Ellis’ book reported her research over many years in a fishing community in Chesapeake Bay (Ellis, 1986). Stein (2010) suggested the resultant flap in response to Ellis’ book drove her to focus on local and personal experiences in subsequent research studies, eventually resulting in the promotion and expansion of autoethnographic method.

The three authors published later reflections on what happened, and what they might have done differently (Ellis, 1995; Sheper-Hughes, 2000; Stein, 2010). All described the betrayal that communities and their members experienced when reading unflattering accounts of themselves and their conduct in minute ethnographic detail. Sheper-Hughes’ (2000) participants seemed particularly exercised by her account, claiming they had not been given credit for the strengths, beauty and resiliency of the community, but rather perceived she had vilified them as backward, out of touch and insane.

Ethical Challenges for Autoethnographers

The organisational autoethnographer does not have the ethnographer’s challenge of accessing communities. They write and research from the point of access, usually their workplace or their social context, where they hold multiple roles, never only as a researcher. However, retrospective organisational autoethnography has an additional ethical challenge as it is usually not possible to achieve retrospective ethical clearance for such work.

An ethnography represents multiple voices of those with whom the researcher observes and interacts (Lester & Anders, 2018). In writing the autoethnography the author/researcher, is at the centre. Rather than representing experiences of others, we write about ourselves, and our experiences in interaction with those who were present when the experiences occurred. It is here that I recognize Lee’s (2018) claim that “[t]he autoethnographer strives to achieve a version of the self and an account of events that is consistent and acceptable to their own conscience” (p. 313). Wood and Liebenberg (2019) emphasized the value of respectful relationships, and the careful use of words to reflect this respect.

In recalling events to create a retrospective organisational autoethnography (Herrmann, 2017), Lee (2018) identified multiple sources accessed in verification of her account. She used her partner as a sounding board. Her partner was involved in aspects of the problem outlined in the autoethnography and/or had listened to the author’s account soon after events occurred. Lee also accessed medical and court/police records to verify details. She indicated the assistance of her doctoral supervisor. These points of verification are vital to supporting the integrity of the research.

We are responsible to communities, of practice, culture and influence, in which our self-experience is lived and richly recounted through autoethnography. In all research we need to be vigilant to ensure the rigour, accountability and integrity of our work.

We need to find a range of acceptable ways to ensure that persons referred to in autoethnography should be advised, and where possible their recall as to what happened should be sought for clarification. If this is not possible a range of alternate options might find satisfactory ethical bounds, such as writing the account as fiction, or disguising the others extensively. There is a delicate balancing act involved in getting this right, and what may seem to be advisable in one situation might not be appropriate in another. The challenges arising from the tension between the relational ethic and the ethic of the self in autoethnography are real and ongoing.

Coda

In order to submit this paper for publication I applied for a waiver from the Chair of the relevant Human Research Ethics Committee. This was granted on the basis that the events described were recorded in the public domain, at a conference which was open to the public along with a recording later available on the university website, and in a published paper.

With regard my description at the opening where the presenter promised anonymity but then pointed to me and named me, I sent her the exact wording of my description. I acknowledged it was my recall only but there was also a recording we could consult if our shared recall did not align. I had not listened to the presentation recording at that stage. She first replied that she did not use my name or point to me, she may have looked at me but she did not do so to identify me. I then checked the audio recording and found she had used my name twice. I sent her the recording and advised her the time point at which she had said my name. She then justified identifying me.

I also wrote to my colleague who expressed upset after reading the published paper. She approved I could refer to her and agreed the wording of the entry which described her reaction. She requested to read the final paper. I provided a copy and she advised it reflected her experiences.