Fanfic Forensics

All right. I'd been meaning to do a write-up of all the ethics questions relevant to my research project for quite a while, but the Surveyfail trainwreck suddenly makes the need for a write-up feel a lot more pressing. Fortunately, others have written quite extensively on how to study fandom while not being offensive and using a sound methodology. See, to name just a couple, theorynut, Kristina Busse's post on research ethics, and of course the AOIR guide. I try to follow these good examples, and the rest of this text explains how, in long and excruciating detail. I'll be tacking a link to this at the top of my main research site and its LiveJournal and InsaneJournal mirrors. Did I mention this is really, really long? A couple of warnings to start out with:

• First and foremost, TRIGGER WARNING. Some of the fanworks referred to in my journals and site, pieces of content written by me, or references listed in the bibliography may contain triggering wording or images. Please proceed with caution.
• The journals/site contain and/or link to material that is not appropriate for minors.
• Much content on the journals/site deals with yaoi or slash, which (very broadly speaking) means male-male relationships in (fan) media. If this offends you, kindly turn away now.

Some potentially relevant personal and fannish basics. I'm a white, bisexual woman in my late twenties. Before I went to university, I was actively involved in Elfquest and Zetsuai 1989/BRONZE fandom, writing fic and drawing art. After that, I lurked in several manga/anime fandoms for many years, until increasing involvement in Harry Potter fandom with the start of my research in 2008 convinced me to give active fannishness another go. I missed squeeing and drawing, and I didn't want to do research on the fandom without giving back a little. I'm still not very loud or publish much art, but I'm slowly getting back in shape. I read left and right in other fandoms besides HP, but usually don't get actively involved beyond the occasional LJ comment. The purpose of my research is to compare and contrast the narratives and visuals of fanworks created by English-speaking and Japanese-speaking fans. Since I'm conducting the research in order to earn a PhD in Japanese Studies, the fanworks created by Japanese-speaking fans are the main focus of the project. My end goal is to provide accurate data about the content of dojinshi (Japanese amateur manga), data which is unavailable at the moment in studies of Japanese fans, and to create debate about the similarities and differences between these two kinds of fanwork from different cultural settings. My methodology involves taking large numbers (hundreds) of fanworks, mainly fanfics and dojinshi, describing a range of easily identifiable elements contained in them (such as, who is the narrator, what is the main pairing), and gathering these data in large sets so I can compare and contrast the results for dojinshi with those for fics. I will try to point out interesting similarities and differences, examine those in more detail, offer a range of possible interpretations, and solicit interpretations from others during the process as in this pilot study conducted last year. (If you caught me talking about my research early on in the project, you may have heard me say that I wanted to explain the differences between fanfic and dojinshi narratives. That was indeed the idea I started out with, but obviously, both the "explaining" and "differences" parts were very problematic. I've corrected that outlook now.) My research involves only content analysis of fanworks (mainly fanfic and dojinshi, will include fanart and Japanese-language text fics later on), not research on human subjects through questionnaires, interviews, or other methods. I try to use as samples only works created by authors who are not minors. Aside from looking for suggestions that they might be minors, I don't search for or keep data about authors. The data about the fanworks I use as samples are/will be stored in this dojinshi data set and this fanfic data set. Note that there isn't much to see there now, I'm just getting started on these, but this short data set created for an early pilot test of the methodology gives a good idea of what I want to make. Any story summaries of the fanfics and dojinshi included in the data sets are written by me for the purpose of contrasting the narratives, and they contain SPOILERS for the stories. The ratings/warnings are the original author's; please pay attention to them before clicking any links, since the stories may contain adult material. If you think I've misrepresented or misinterpreted something, please let me know and I'll correct the information right away. I don't link to or use as samples works on sites that use some form of password protection. (Note: if something on LiveJournal or another journaling service gets f-locked after I first bookmarked it, it may take a long time before I notice this and remove the sample.) I try to practice open notebook science, meaning that I publish anything research-related that I do -brainstorming, data adding, etc- as it happens, as much as possible, under a Creative Commons licence, to invite immediate criticism/feedback and hopefully allow others to re-use my material in newer and better projects. See Wikipedia for more information on the concept. The data sets published on this site are therefore a work in progress; they are definitely incomplete and may contain factual errors at this moment because they haven't been revised thoroughly -I'm extremely busy with other work and am just recording the basics of the samples I have. (I'm not interpreting these flawed data sets yet, of course.) I identify fics and dojinshi by title and author and, wherever possible, link directly to the stories or to the relevant warnings page if the author has requested that works not be linked to directly. This is done because I want other people to be able to verify my findings easily and quickly instead of having to take my word for it that X percentage of dojinshi depict French kissing, or whatnot. More than a few papers discussing fic that I've read make remarkable claims that I dearly wish I could verify, but can't because the samples are porely defined or locked up in paper fanzines. I want to avoid that. However, I realize that it can be creepy to see one's name or work showing up in this kind of research out of the blue. While I don't believe anybody has the god-given right to control every way their work is referenced anywhere, I haven't been granted the absolute right to publicly analyze a piece of fanwork in whatever way pleases me, either. If you're the author of a fanfic/dojinshi included in the data sets and you don't want me to include your work or point people towards it, drop me a note and I'll immediately remove identifying information from the entry -author name, title, link, summary- or remove the entry from the data set entirely, as per your specifications. I will also link to stories via a different URL or make other changes that you request. (If you feel like including the exact reason why you want me to remove or change information, please do so, but there's no need. It's none of my business.) Am I making my sample sets less representative by allowing people to withdraw their work? In theory, yes. In practice, I'm quite confident it will make very little difference. Am I going against my own stated goal of offering data that others can verify by removing identifying information from data about a fic or dojinshi? Definitely, but no methodology is sacred enough that it's worth ignoring people's express wishes. This research is not about open heart surgery or nuclear safety. Nobody is going to die and no outrageously wrong conclusions will suddenly be drawn as a consequence of allowing the people who made the samples I'm playing with to control their own work. Important note to other researchers not affiliated with fandom: divulging information about fans and/or fanworks to parties with no understanding of the concept of transformative/derivative works can be very harmful for fans. Please, whatever you do with the information on particular fanfics and dojinshi contained in my data sets, don't put it in the hands of those for whose eyes most fannish works are definitely not meant -authors and other rights holders of the original works on which the fanfics or dojinshi are based, people in the content industry who consider fanworks illegal or evil, or minors. If it ever turns out that the data sets I publicize are being put to some kind of use that is risky for fans, I will take the data down immediately, open notebook science or not. It took me a long time to decide that I wasn't going to contact authors and ask for permission to use their works as samples before starting work on the data sets. While getting informed consent is not generally considered necessary when studying pieces of literature (as opposed to the behaviour and interactions of authors), contacting fic and dojinshi authors just seemed like the right thing to do. But it turned out that the logistics involved would make for hundreds of very ambiguous situations. What if the contact information I have is outdated (a relevant concern since a lot of my samples are from as early as 2001), but I either don't realize this or can't find the new, correct information? Can I use the work or not? If people don't react, does that mean I can use the work or not (and will the author(s) share my interpretation of their lack of reaction)? In the end, I decided this approach would probably lead to a lot of misunderstandings and muddle up the sample set to such a degree that it wouldn't be even vaguely "representative" anymore. Also, I didn't want to come across as asking authors to do my homework by looking at my data and giving their interpretation. Note that while I don't want to spam individuals about my research, I absolutely do want people to look at the data and let me know their ideas. That's one of the reasons for publishing everything online as-it-happens instead of not asking for feedback until the final draft of the thesis is finished. While analyzing data, I'm helped but also hindered by my own limited point of view, personal experiences, intellectual ability, amount of knowledge, conscious and unconscious convictions, and general humanity. This was never more obvious to me than when I threw my first results out on LJ and got a volley of new interpretations thrown back at me through various channels. I do not want to engage in the kind of science that requires one to read material in total solitude, process it, and emerge after several months or years bearing pearls of wisdom. That won't make me a very good academic and would make my research of far less use to the rest of the world than it could be. If you see anything in the data sets or on the site as a whole that interests you, that I misinterpreted, or that I seem to have overlooked, I will be eternally grateful if you decide to send me a message or post a comment about it. (If you want I'll do a Snape sketch in return for your help, just shout for it.) Any comments you post on the main site or in its LJ or IJ mirrors are and will remain public, but I will not quote or mention you without express permission in the final thesis or mention your name anywhere if you send me a private message. I will never, ever, under no circumstances use the real name of anyone who addresses me using a fannish name. While it would take some real work to screw up as spectacularly as the Surveyfail "scientists" did, I may still say or do dodgy things at some point. Should that happen, or if something in the research ethics outlined above strikes you as questionable, please, please whack me with something. Additionally, I try my hardest to avoid racist, ableist, sexist or otherwise harmful language and ideas, but with mixed success because I've lived a very privileged life and am still in the process of gathering clues that should have been acquired a long time ago. Any corrective whacking you might care to do in this area would be tremendously appreciated as well. Thank you.