For those us interested in the culture of play in MMOGs, ethnographic participant-observation is a mainstay method. We game, we think about games, we read and write on games, typically in that order. But it raises a lot of practical, ethical issues regarding how it should be done. What exactly is 'informed consent' and when is it necessary? What constitutes legitimate 'participation' in an multiplayer world? And how do folks negotiate in-game research vs. play? Thought it might be useful to start a public conversation on some of these issues, with an eye toward developing heuristics we (both gamers & researchers) might agree on...
What exactly is 'informed consent'? There has always been a rather wide range of views on what constitutes 'ethical' when it comes to collecting observation data on/about other people in the field, and I don't think we are going to suddenly resolve the debate (or ought to even try) through the issues raised specifically in relation to fieldwork in virtual worlds. There are two main poles in the conversation, which I will shamelessly parody as:
1) The 'Protect the People' Pole, arguing that every single person who is a potential participant in your research deserves full disclosure of what you are doing and how, and even after formal (e.g. signatures on human subjects forms) is given, such consent must be constantly negotiated and re-negotiated on an ongoing basis since any initial agreement cannot adequately insure that they truly understand the nature of what it is they are consenting to. Folks like Liz Reid (Thanks to Richard Bartle for this lead) would say that you can do real damage to a community and therefore you are morally bound to protect them in all ways, including getting informed consent from all people involved, even marginally, in your research work.
2) The 'Don't Butt In' Pole, arguing that soliciting formal informed consent from every possible participant in an ethnography is disruptive and largely unnecessary (within, of course, certain constraints). Giving a long description of your research to every person you cross virtual paths with, let alone contacting them outside the game in order to snailmail them human subjects forms, would be disruptive to actual participation in the space/culture/game (both your and theirs), which is unnecessary and unwanted. The idea here is that you do a disservice to other gamers and your own legitimate observations if you constantly make what you do an issue with everyone.
Most ethnographers fall somewhere in the middle and consider it part of their job to negotiate what constitutes 'ethical' in practice depending on the context of their work, the attitudes of those they research, and the cultural norms of the community under observation. But negotiating this out in actual practice is hard work, and there really aren't any clear, uncontested guidelines to build on. Ethical decision-making and Internet research: Recommendations from the aoir ethics working committee is one useful resource. Others?
What constitutes 'participation'? Is gaming alone in a MMOG participation? Is gaming only with other researchers actual 'participation'? What about setting up a clan for the sole purposes of researching them, unbeknownst to the members? What if they do know – is that viable research? Remaining anonymous to everyone in game? Titling your avatar 'academic_researcher' so anyone who talks to you knows what you do in RL? Most of us would agree that it's contingent on your research questions (for a start), but its a sticky issue because it also depends on your playstyle as well...
How should one negotiate research vs. play when in the field? Am I allowed to PK in a PvP game? Can I tell someone to STFU when I want and am 'in character'? Or am I expected to somehow serve as a 'non-partisan role-model' when out in the field? Would doing so compromise my own legitimate participation? Personally, I am a PvP player, which makes these issues even more complex.
Example from the field. I'll use my own work as an example, since I'm selfish and want to discuss these issues as a way to refine my own practice.
I play Lineage (first I, now II) and run my own clan, despite my research needs, not because of them. Both are labor-intensive activities that compete for my limited time, often with my research coming in second to my obligations to running the clan. But, then, I have strong affiliations with my clan members, I love running the group (webwork and all), and I consider my obligation to my clan to take precedence *gasp* over my need to do research. (NB: I'm very lucky here in that I work in an academic setting that considers gameplay an absolute prerequisite to theorizing games. I realize other institutions aren't the same.)
My general MO to date has been to keep the lines between my professional identity and my Lineage identity transparent to whoever is interested, treating in-game disclosure of information about my 'academic' life the same way I treat academic disclosure of details about my 'game' life, based on the notion that I am bound to both communities to be generally forthright about what I do. To date, people I interact with inside either magic circle respect the boundaries between my professional life and my game life. I have never, for example, had another researcher judge my in-game activities by academic professional standards or my gamer friends judge my academic activities by gamer standards (except to say that my writing is really flipping dull). However, I have occasionally had what I see as an uncomfortable bad boundary mix-up. For example, I've had academic colleagues I've never met before log onto the game and asking to join my clan for research purposes. I've also had (though far rarer) a gamer mistake my in-game role as clan leader for some sort of beyond-game role as counselor (I'm trained as a cognitive psychologist, not counseling psychologist) calling me at home at 2:00 am in the throes of a personal domestic crisis (my personal information is as easy to find on the web as anyone else's). Clearly the way I navigate the research vs. play issue is idiosyncratic, built to fit my own needs and what I'm ethically and personally comfortable with. And both instances had a happy enough ending. But surely there are alternative ways to negotiate the sticky issues involved with being both a gamer and a games researcher that might, in the long run, avoid the occasional SNAFU in the first place. If so, I'd love to hear them. Learning the hard way is er... the hard way.
Now that you make me think about it, I guess I have been lucky. When I am in a virtual world, I am actually going there for the same reason as everyone else, about 100 percent of the time. When I do research, it usually involves grabbing data from eBay or fansites, or from surveys. I sprinkle my papers with anecdotes about my in-game experiences, but the in-game experiences are not the research, per se.
If I spend a year in New York and then write a paper about real estate prices there, where I get the data from the New York Times, I think it is OK to then write up anecdotes about how the landlords were. In fact, it is fairly incumbent to *not* think about research when doing the 'living' part. So, in my MMORPG time, I have always tried to follow my nose for personal satisfaction. Oddly enough, that started out as clericing because I like being called into groups, but now it has evolved to solo exploration play - that's all I did in SWG, for example - and damage-dealing wizards, a role that also has high autonomy. The exact opposite of healing! But the point is, I never choose on the basis of research needs. Makes you wonder where the line between research and play really falls.
Posted by: Edward Castronova | Oct 17, 2004 at 15:58
This is a topic that certainly lingers a lot in my mind ever since I started grief play - the more so grief play is a topic that is heavily emotional for a lot of MMORPG participants.
IMO, there is a lack of specification on how one should go about ethical research in an MMORPG, but from long discussions I had with my supervisors and the ethics review people in my institution, one sense I get when I'm stuck with what ethical procedure to follow (or not) is to ask myself:
1) Are the participants inconvenienced in any way as I research/observe?
2) Will the participants be identifiable when I publish my research?
Curiously, at least for grief play, I've found in-game observation to be less useful compared to other methods of data collection. With that in mind, so far, I've kept my in-game participant observation not different from as a normal player who isn't a research. i.e. I play, I game, I interact like I always do. I see events unfold like I normally do and refrain from influencing the course of events. When there is something that seriously strikes me significantly in that event, that's when I follow up with individuals through querying, and that's when I properly identify my interest and see if the participant is willing to share his story more intently.
Posted by: Chek Yang | Oct 17, 2004 at 18:50
I think you can question whether being an aloof participant-observer in the anthropological sense is completely ethical. (See Tuhami for more on that.) Researchers shouldn't beam down from the Enterprise and obey the "Prime Directive." I'm sure the MMORPG natives have the wisdom to handle our technologies.
"Ethics" has a few meanings. If it means "rules or standards governing the conduct of a person or the members of a profession," then who knows, maybe you should send snailmail consent forms to be consistent with the established practices within your cognitive psychology profession.
If ethics means your personal "system of moral values," and you're considering sending snailmail consent forms to study your clan -- that really seems like a lot of hassle. Probably the only thing you need to do is protect their privacy to the extent they want it protected -- and privacy concerns would seem most likely to arise when you get past the virtual avatar and start talking about the controller IRL, right? (?)
Both your examples of boundary mix-ups seem pretty funny (fellow academic researcher shows up ingame / gamer thinks you're a counselor) -- but they don't seem particularly troubling from a moral standpoint.
Posted by: greglas | Oct 17, 2004 at 22:29
It’s worth noting that the AoIR (the Association of Internet Researchers) ethics working committee has done some work on these general issues. This work resulted in a report titled: Ethical decision-making and Internet research.
The report can be downloaded as a pdf, here are the meta-data and the Table of Contents:
Title: Ethical decision-making and Internet research: Recommendations working committee
Authors: Charles Ess and the AoIR ethics working committee
Approved by AoIR, November 27, 2002
Available online: www.aoir.org/reports/ethics.pdf
Contents:
I. Audience, Purpose, Rationale and Approach
II. Questions to ask when undertaking Internet research
A. Venue/environment - expectations -authors/subjects - informed consent
Where does the inter/action, communication, etc. under study take place?
What ethical expectations are established by the venue?
Who are the subjects posters / authors / creators of the material and/or inter/actions under study?
Informed consent: specific considerations
B. Initial ethical and legal considerations
How far do extant legal requirements and ethical guidelines in your discipline “cover” the research?
How far do extant legal requirements and ethical guidelines in the countries implicated in the research apply?
What are the initial ethical expectations/assumptions of the authors/subjects being studied?
What ethically significant risks does the research entail for the subject(s)?
What benefits might be gained from the research?
What are the ethical traditions of researchers and subjects’ culture and country?
III. Case Studies
IV. References, Resources
V. Addendum 1: “Ethical Protocols” - Questions and decision-making guides for Internet research ethics
VI. Addendum 2: Discussion of contrast between utilitarian and deontological approaches as reflected in contrasts between the U.S. and Europe (Scandinavia and the EU) in laws regarding privacy and consumer protection
VII. Addendum 3: Sample consent forms (courtesy, Leslie Regan Shade) for parents and children involved in Internet research.
Posted by: ren | Oct 18, 2004 at 07:47
My view on this is connected to my overall critique of the codification of ethical norms in ethnographic work in anthropology and other disciplines. My feeling in general is that the ethical norm of rigorous disclosure of purpose, informed consent from informants (though most cultural and social anthropologists don't go so far as to have people sign forms and so on), and the general casting of the researcher as an intrusive, potentially destructive, force, is actually a huge problem. I think that rather than addressing disparities in power between the researcher and his/her subjects, many of these norms actually enshrine those disparities, they underscore the idea that the "natives" somehow cannot be trusted to understand and look after their own best interests.
These issues vanish even more, at any rate, when you're talking about the relationship between a 1st world academic and the players of MMOGs. I don't believe there's anything that needs "protecting" in the usual sense, and to try and be "protective" is a guarantee that what you find out about cultural practice within a gameworld is not going to look anything like how players actually behave. This is a problem that some academic work on virtual communities and online culture in general has: its formality and distance from its subject makes it in some important respect simply wrong in its representation of its subject matter.
I think the ethnographic norm in studying virtual worlds ought to be much closer to good first-person reportage, albeit resting on a foundation of scholarly practice and scholarly canon. Julian Dibbell's work is what I'd like to see many ethnographers interested in gameworlds undertake. Does that mean, for example, that someone like Chek Yang should commit grief play in order to study grief play? Well, yes, at least once, I'd say that's worth doing. But of course you shouldn't set out to create practices that wouldn't exist but for you creating them and then represent them as common.
And that's where Chek Yang's comments are especially on-target. I also find that if you want to understand cultural and social practice in a given gameworld, that direct observation and reportage within the gameworld is actually not all that useful past a certain point. You can understand how the game functions, what the limits of its possibilities are, what the ebb and flow of play is, where the nodes of player interest and activity are. You can observe what players do. But to really know what's going on beyond that, I think you have to talk to players in the game and even more importantly, to read gamer discourse across a wide spread of forums--the official forum for a given game, forums at Warcry and IGN, forums at Grimwell's and Corpnews and f13, forums maintained by guilds or groups, and so on. It's true that if you do so, you'll still not know much about the consciousness of players who are part of the "silent majority", people who just play and don't ever do anything else, but it's the discourse outside the frame of the game that will tell you a great deal of what people do and know inside the gameworld.
And that, I think, is simply like reading texts in any other context: it comes with no ethical burdens save to read honestly and well.
Posted by: Timothy Burke | Oct 18, 2004 at 07:53
Ditto to Tim's first three paragraphs, with one small point:
Tim> Julian Dibbell's work is what I'd like to see many ethnographers interested in gameworlds undertake.
Me too, but I got the sense, in LambdaMOO, that the community had some mixed reactions to Julian's work for various reasons-- I think he's part of the reason for the current notice on the front door. Some people think the reporting was too accurate, while others seem bothered by the fact he used psuedonyms. (Browse these.) In MTL, he admits the whole P/O question was a problem. See ix-x.
Posted by: greglas | Oct 18, 2004 at 08:14
I spent a fair amount of time on LambdaMOO in Julian's wake, not as a researcher but just as a participant, and it's absolutely true that people had very mixed and divided reactions to the Bungle piece. Some said that it was because they did not want to be "studied".
I think the reactions themselves were interesting and important and profound--but in the end, I essentially think that the people who object simply to being written about by others are wrong, even when they are articulate and careful about how they voice that objection.
If a student of mine writes about me twenty years from now, either by fictionalizing me into a novel or otherwise describing me, I do not have the right to prevent him. I have the right to sue him if his portrayal is falsely defamatory; I have the human right to object in ethical terms if his portrayal is arguably true in some respects but designed with malice in some fashion to be hurtful. I certainly might feel hurt, depressed, chagrined, by that portrayal. Who among us wants to be described by others except in laudatory terms? But to insist that such a writing should not take place, that there is an ethics which ought to prevent anyone from doing such a thing, is more or less to crush both realistic fiction and journalism/travel writing/vividly individualized ethnography. Such writing--and the revelations it may deliver--depends on a kind of cruelty, on the willingness to describe what one person sees. We hope they describe what they see faithfully, clearly, fairly. We cannot guarantee it. I'd rather have an ethics that instructs people to be fair and observant than an ethics which instructs them to swaddle interesting subjects in twenty layers of cotton and so prevents any sharply observant picture from emerging.
In one virtual community I've been part of, the sacred commandment was: YOYOW. You Own Your Own Words. I think there's another: YOYOE. You Own Your Own Experiences. What you see, what you do, and what you see others doing, is yours to write about, describe, pass on, narrativize. If you're a scholar, you have a set of analytic and canonical tools to process your own experience through that produces some additional insights (possibly at the cost of losing some potently individualistic observations).
Posted by: Timothy Burke | Oct 18, 2004 at 08:59
Suddenly it occurs to me to ask whether EULAs are languaged in such a way that the MMOG is a private club, in which case even external reportage might be a no-no. I don't think there's any prohibition in current EULAs on speaking publicly about events in the Secondary World. But I could envision clauses like "Licensee shall not use events, conversations, or depictions of events or conversations, in any form (video, audio, written, etc.) for personal benefit without the express written consent of [the owners] and all those so depicted or overheard." That would restrict quite a lot of the Ludlow-type stuff right away, and probably all academic exploration as well. At least, they could force us to get approval from their PR department before publishing anything.
Not saying they should do that, but it's kind of surprising that there's no policy, what wth all the media attention.
Posted by: Edward Castronova | Oct 18, 2004 at 09:09
Is that even legally possible? Can a TV company censor what I say about their programme, for example?
Posted by: always_black | Oct 18, 2004 at 13:57
It's a EULA, a contract. My lawyer friends tell me that (almost) anything goes in contract law.
Posted by: Edward Castronova | Oct 18, 2004 at 14:34
Well, I know that some of these EULAs regard everything that happens in game as their intellectual property, from chat logs to screen shots. Their position is weakened, however, by encouraging fan sites to post screen shots and describe their character's lives in world. Game companies seem to be in a difficult spot on this one: how do you encourage the proliferation of good news fansites, while at the same time trying to prevent sites like the Herald that are apt to report on sketchier aspects of the gamespace.
Second Life has a clause prohibiting the posting of chat logs -- even public ones -- which the Herald tends to flout. My imperfect solution is to let other people post that stuff.
On the general point about ethnography and the fetish for rules that people seem to have about it, my thought is this: if we can't experiment with new ways of studying groups in an environment like this -- a gamespace -- then we truly are lost. We will have permanently handcuffed ourselves with a code of ethics that is stifling and anal retentive in equal measure and utterly inappropriate for studying game spaces. People may chose to follow such rulez on their own, but then they are apt to learn exactly what they deserve to learn: nothing.
All that having been said, I'm not in these games to study anything. I'm there to have fun. If I get a bright idea and want to write it up I will. If I don't, I won't. Even if I write something up it probably doesn't end up on my vita. Now try to tell me that I need all my friends in these games to sign permssion forms before I hang out with them. It ain't goin' to happen.
Posted by: Peter Ludlow | Oct 18, 2004 at 14:48
There are some chapters in existing ethnographies of virtual worlds that concern the ethics of studying them. Two that people who are interested in this subject ought to check out are:
Lori Kendall, Hanging out in the Virtual Pub.
Lynn Cherny, Conversations and Community.
I have copies of both these books on my shelf. Interestingly, if you don't then you can try out your own ethics. The amazon.com links above take you to pages that have a "search inside" facility available (click on the book picture to invoke it). In Kendall's book, the relevant part is Appendix B; in Cherny's it's Chapter 7.
Richard
Posted by: Richard Bartle | Oct 19, 2004 at 11:16
Richard> I have copies of both these books on my shelf. Interestingly, if you don't then you can try out your own ethics.
But I suppose if I walked down the street to the library and looked it up, I wouldn't need to try out my own ethics? :-)
Posted by: greglas | Oct 19, 2004 at 14:30
greglas>But I suppose if I walked down the street to the library and looked it up, I wouldn't need to try out my own ethics?
That depends on your ethics. If you believe that it's OK to sell one copy of a book that multiple people will read instead of buying their own copies, that belief must come from somewhere, right?
It also depends on the ethics of the library. If they have an extensive list of banned publications, or if they favour writings expressing particular opinions, or if they favour books from particular groups of people, you may find your own ethics kicking in.
A quick trip to Google shows there are several articles on ethics and librarians/librarianship; unfortunately, they all seem to be in journals and books that I don't have on my shelf...
Richard
Posted by: Richard Bartle | Oct 20, 2004 at 02:55
Some reactions and comments here:
http://www.richardparent.net/fs0003/2004/10/ethics-metaverses.html
Posted by: greglas | Oct 20, 2004 at 12:10
Several of the previous comments suggest that it's condescending to think a researcher must 'protect' their participants, as thought they are the 'simple innocents' that cannot act with actual 'informed consent.' I have never thought of it in quite those terms before, but I do think it's a great point.
One solution might be to treat them as colleagues. Given the metagame theorizing that is often a natural part of gameplay, I often find quotes and comments about the game that I'm more comfortable treating as citations to other scholarly work on games rather than data to be theorized per se. Our conventions for this sort of thing are awkward, which says something about the changing nature of publishing, I think.
Posted by: Constance Steinkuehler | Oct 23, 2004 at 09:48
Clarification: Second Life has rules against posting conversations and similar without permission... on Second Life Forums and in Second Life, specifically. There is no mention of restricting postings on foreign sites.
The only rules that would apply to such external postings are the actual laws involved, plus any regulations on the part of the one posting. Linden Labs has no responsibility to stop them.
I have not looked into other massively multiplayer game's rules on these matters, but as Second Life is a much more intensely personal multiplayer game than, say, EverQuest, it is SL that will likely be the forerunner in protecting or failing to protect personal rights.
Posted by: Craig | Oct 28, 2004 at 13:08
I am currently teaching a class on (and in) MMOGs and starting tomorrow, my class will be examining Second Life as their second virtual world (Star Wars Galaxies being the first).
I found it interesting that Linden Labs sees SL more as a virtual environment for education, than as a site to be studied.
They have a "Campus Life" program which is "a program to allow college level classes to use the powerful tools and realistic environment of Second Life as a venue for learning." But they are quite explicit about both identifying classroom participants and divorcing the program from precisely the kinds of ethnographic work being discussed here.
Posted by: Douglas Thomas | Nov 01, 2004 at 08:02
First I wholeheartedly agree with Mr. Lulow's point about being free enough to construct new rules for studying new groups/societies. We should understand and appreciate the value of bringing rigor to our research without feeling compelled to be bound by rule sets designed for the olden days of "the white man studying the brown man."
Second, with full awareness of where I am virtually, I wonder about any differences in this question of ethics which would arise in researching non-persistent online multiplayer games. Clans, guilds, teams, etc. all exist for games other than EQ, UO, Leg. I&II, etc.,in fact - there exist other communities as well - I'm thinking of the mod communities - that center around these other experiences that are not really mirrored in persistent worlds. My question (poorly worded I'm sure) being, were I to undertake a ethnographic study of this environment, would there be any lessening of the ethical dimension because the game session is not persistent - although the participants' online personalities may well be?
Posted by: mark oehlert | Nov 01, 2004 at 10:24
well done
Posted by: fladreck mapasure | Dec 09, 2004 at 02:36
I'm posting here with a great lag, but I figure it's worth a try.
I have a different kind of research program in mind than the participant-observer ethnographic study mentioned here. Coming from the field of economics, I want hard data. Coming from the subfield of experimental economics, I want to be able to implement changes in the VW and collect data on the effects.
For me, many of the ethnographic concerns aren't likely to be important. With a few exceptions, player anonymity in my research reporting would be the norm, for instance. Likewise I don't think I'd have to worry about whether my personal interactions with other players are deceitful, since I would not need any personal interaction at all. And in terms of informing the players, I believe a fairly blanket statement early on to all players that research may occur should be satisfactory.
However, making changes to the world will obviously impact the players, and we all know that many changes are considered negative. I'm not sure how to deal with this, and it's stymying my research goals. Part of me wishes to say, "These players face changes to the world all the time in most games, and they live with that (although they may complain) and they can always quit if they dislike it enough." However, we also know that these are more than games to many people, and someting which impacts gameplay significantly could potentially impact players' well-being overall. So it can't be trivialized.
On a more practical note, I have to keep in mind that it's not just up to me to determine by myself what is or is not ethical, since any research I do would have to go one way or another through our IRB (Institution Review Board). Currently, they aren't any more prepared than I am to decide on these matters.
Any further advice or discussion would be very welcome.
Posted by: Timothy Dang | Feb 09, 2005 at 17:24
If it is presented as a lab experiment and you don't deliberately make changes that cause the players to blame each other or creates turmoil between groups of players or individuals then it sounds quite ok? Or is it different because it is running over a long period of time?
Posted by: Ola Fosheim Grøstad | Feb 09, 2005 at 18:04
Ola Fosheim Grøstad> If it is presented as a lab experiment and you don't deliberately make changes that cause the players to blame each other or creates turmoil between groups of players or individuals then it sounds quite ok? Or is it different because it is running over a long period of time?
The goal would be to have the experimental VW running over a significant (possibly indefinite) length of time, since one big potential benefit of VW's over the lab for research is their ongoing society.
Given that, and people's attachment to their characters and the activities of their characters, it's reasonable to think that any given change could upset players not merely in terms of "nerfing", but maybe even their relationships and identities. Suppose for instance a price ceiling is implemented, and a player-merchant decides it's no longer worth their while to conduct business. That player may not only be upset about that, but also lose part of their in-game identity and framework for interacting with others.
I can make lots of arguments that it's not worth being overly concerned about this. Players would be warned that things could change; players of ORPG's should generally be used to the idea that things do change; while it's not "just a game", there's still a certain moral discounting which seems appropriate compared to similar effects on RL..., but I'm trying to see things from the worry point of view to see what solutins there are, as well as consider what is likely to satisfy an IRB that a research project is cool.
There is certainly one way to do research projects, which is to have initially different worlds side-by-side from the outset, and then not change anything thereafter. This method should be OK for some projects, actually, although I still have to make the argument that I'm not putting people's emotional and social well-being too much at risk.
Posted by: Timothy Dang | Feb 09, 2005 at 21:10
Just a thought. Couldn't you make it reset based? You reset the game world with new mechanics every 2 months or so.
Posted by: Ola Fosheim Grøstad | Feb 10, 2005 at 07:04
Ola Fosheim Grøstad> Couldn't you make it reset based? You reset the game world with new mechanics every 2 months or so.
I actually hadn't thought of that. I'll give it some more thought. I'm personally not well acquainted with how reset-based MUDs tend to work out, whether players tend to remain through resets and such. In any case, it wouldn't allow for some of what I'd like to study (how people respond to changes, particularly changes in economic institutions), but could be useful in much the way that running side-by-side VW's.
Thanks for the thought.
Again, any more thoughts on how to set things up, explain things to subjects, explain things to IRB's would be very welcome.
Posted by: Timothy Dang | Feb 15, 2005 at 14:24
You might want to look at some of the web-based games. The ones that are tick-based. They seem to be reset-based and have monthly-or-so-resets. I know that Planetarion worked that way and it had lots of players (100000?) when it was free.
Some players will probably leave between rounds. Not sure if paying participants who complete at least 2 rounds is an option...
Posted by: Ola Fosheim Grøstad | Feb 15, 2005 at 14:44