pondering on her blog if Collaborative Autoethnography (CAE) is an appropriate method for a dissertation. Rebecca, as far as I know, is currently ABD and looking at wrapping up her degree - I have no doubt that soon she will be Dr. Rebecca :-).
I think that there are many reasons why CAE is a good approach to researching certain things. I am introducing my own bias here when I say that I prefer working with others on research projects and on publishing. I've written some things myself, and there is a benefit to the lone researcher with his readings, literature reviews, data crunching, and data analysis. It's sort of like going to the gym on your own and working out on your own. It has its place. That said, despite the fact that I do not see myself as a super social person, I do prefer the company of others. I had some of my best workouts when I trained with fellow taekwondo students. I've had some great "aha" moments when studying with fellow classmates for my comprehensive exams, and I've had some great new insights when I worked with others on collaborative research projects (both papers and presentations).
When it comes to dissertations, however, I think that we have a bit of a problem. On the one hand a dissertation is meant to augment the amount of knowledge that we have in our world. This is great, however I think that dissertations are under-valued if this is one of their main goals. Dissertations are never (or super rarely) published as books as they exist in dissertation form, and they are often cannibalized and taken apart to create publishable journal articles out of them. While there is nothing wrong with journal articles (I enjoy working on them, and providing peer review for them), I do think that there are positions that are at odds with one another when we simultaneously think of dissertations as a monolithic magnum opus, and at the same time we are encouraged to write them in a way so that they are easily taken apart for standard research articles.
The other issue about doing a CAE in a dissertation is the collaborative part. From my own explorations in PhD-land (having spent 5 years looking into programs before committing to Athabasca University) is that the dissertation is seen as a final exercise of your research abilities, a culmination of the domain specific knowledge, and research methods knowledge, that you have accumulated and are now putting to the test. This, by necessity, is a lonely path, one which the researcher/student/candidate, needs to do on their own.
If a CAE, or any other collaborative work, is done properly it's hard to distinguish who did what part. I saw that with the work that I've done with the rhizofolks. We produced some great work (if I do say so myself), but we've swarmed those docs so thoroughly that it's hard to say what parts I contributed. I think that in order for any collaborative work is to be acceptable as a method for a dissertation there need to be at least two things in place: First, the person who is defending the piece, the candidate, needs to be fully aware of every nook of the research piece. They need to live, eat, breathe, that document. Otherwise it's plain to see what parts they did and what parts they ignored. Secondly, the other researchers who took part in the document need to write testimonials and go on the record to indicate that the candidate for the PhD did actually contribute, and they contributed to a satisfactory degree as defined by the rules and regulations of the university (which I suppose is the 3rd part, those rules and regs need to be set). Once these three things are done, then the dissertation committee needs to approve the dissertation.
In short, I don't see CAE as a methodology that would be approved by a dissertation committee. I think that an Autoethnography could be acceptable, however the method needs to fit within the established acceptable methodologies of your examination panel. If your panel recognizes it as an important method and sees value in it, then you're all set! If you have a couple of strict empiricists on your panel, then I don't think that it will be accepted.