The ethical issues around autoethnography are not easily dealt with by University Ethics committees. The fundamental issue is that no-one lives in a vacuum and writing about the self, no matter how tightly controlled the context, includes writing about others.

You can't do 'good' autoethnographic work without constantly questioning the ethics of your pursuit. As soon as you put that 'I' on the page, you can't avoid asking if your revelations might be harmful to you or anyone else (Jago, 2002, p. 753).

Delamont puts this dilemma just as strongly, stating that autoethnography is almost impossible to publish ethically (Delamont, 2007, pp. 2), in part because the reader demands that the story to be "true". It is an issue that authors wrestle with in all forms of personal narrative, whether telling their own stories or those of others. Do no harm is the nebulous and inadequate concept in the place of guidelines, and the requirement to judge each instance on its merits leaves the door wide open to ethical errors, concerns and criticism.

Anonymity and the protection of participants and significant people are central issues here. It is important to defend individuals from the critical gaze of others. In an ethnographic narrative about some extraordinary circumstance or experience, it can be difficult to cloak the individual in anonymity, given the pervasive nature of the media and the public thirst for so-called human interest stories. We expect journalists to respect privacy and decry their morals when they do not, so the responsibility for researchers weighs heavily. It is possible to entirely fictionalize names and places if the narrative is sufficiently obscure for the public, but we are still left with the issue of individuals recognizing themselves and their neighbours if they read the research.

If the narrative's lead character is identified as the author, the problem just gets worse. How does one protect one's significant others from identification by association? Is it necessary anyway? If an author is writing about a condition that has genetic aspects, as in this case, what will the story imply about familial generations in both directions? Will a relative consider themselves exposed, now or in ten years time. The ethics of informed consent will necessarily influence what we tell and how we tell it (Adams, 2008, p. 12), whether we choose to reveal or withhold and this may also have implications for the conclusions we draw.

In this study, I have based my ethical decisions on several over-arching factors. Firstly, it is my own identity, reputation and public image that is most at stake - I should be true to my own protection, and carefully edit data that is not relevant to the study. My loved ones are under my protection but there is no choice but for them to be identifiable by association. Solutions are really two-fold. If something places an identifiable person in a negative position, the story is excluded. It is also possible to alter the narrative so that the relationship is changed and the person anonymised without affecting the fundamental truth and meaning.

The second aspect of defence concerns the revealing of things hitherto unknown within existing relationships. Carolyn Ellis discusses "relational ethics" in this context (Ellis, 2007), bringing friends both old and new, colleagues and family under the umbrella of significant others. There are moments reading her earlier work that I think the ethic is misplaced, or perhaps simply that her understanding has developed and changed over time. Indeed this is recognised in her later work (ibid.) and her own criteria have become more stringent. However, as I study the effects of my autistic traits on my musical practice, I am increasingly finding that the real learning about these things is coming from other aspects of my life in both past and present. In some cases there are things that may have enormous impact but I feel may be inappropriate to reveal, for fear of hurting others (Ellis, 2007, p. 17) because they are things I did not know about relationships, or have not previously shared. Do the conclusions then become less valid because the means of reaching them is undisclosed? Do I need to go back and look for other, more benign, evidence elsewhere, and what do I do if I can't find it? In the past five years there are relationships that have changed as a result of harsh discovery and the vulnerability that goes with the territory.

Ellis uses the term "appropriate" in her writings about ethics, not least in the Telling Secrets article (Ellis, 2007) so I will appropriate the term here. Definitions of that word include "suitable", "fitting", "proper", "correct", "right" or "OK". All of these have value judgments attached - they are not absolute, they require a defence, not least to those involved, but ultimately I have to rely upon my own ethical sense and the advice of supervisors to maintain the clarity of my own conscience.

Related Posts

1. Ethics - overview: A brief introduction to the ethical considerations of the research.
2. Evoking Anxiety: Are evocative autoethnography and an impairment in empathy an impossible combination?
3. ToM and Narrative: Individuals with poor or altered Theory of Mind construct narrative in different ways to "typically developed" people.
4. Informed Consent: Discussion on the process of defining and obtaining Informed Consent from participants.
5. Autoethnographic Approaches: Autoethnography is not without its detractors and controversies.
6. Beginnings ...: A little background. How did all this start?
7. Recruiting: Apart from finding suitable participants for surveys, interviews and the like, recruiting participants takes on new meaning when writing about one's self, as there are those who are implicated by association.