Speaker: Amy Bruckman (Media Lab, MIT)
Good morning. Welcome everybody. My name is Amy Bruckman and I do research on technology and education. You are here for my first and last rendition of this talk on more political issues; it is not my field of expertise. But when Jennifer [Mnookin] told me about this conference, I thought it would be an interesting opportunity to reflect back on some complex and, on the whole, painful events which took place a couple years ago. I want you all, just for starters, to think about the "D" word. What is democracy? Think about it a little bit. See if you can come up with a definition in your mind. I'll come back to this again at the end of the talk because I think this question is central to the events that took place.
I am going to talk about an experiment in the democratic control of a virtual community called MediaMOO--an experiment that failed--and why it failed. Democratic control of communities on-line is very different from using the on-line environment to connect with real world politics. I will be talking about the former, not the latter. I think they are very different and should not be confused.
So what is MediaMOO? MediaMOO is a text-based virtual reality environment or MUD on the Internet designed to be a professional community for media researchers. We have about 1,000 members from 30 countries and we have been open since January 1993. In most of these environments, most MUDs on the net, people are anonymous. On MediaMOO, people are identified by their real name and a description of their research interests. You can find out what kind of work someone is doing and strike up a conversation saying, "Oh, really? You do work on cyberspace and law? Fabulous! Tell me about it." While most MUDs are populated by undergraduates who should be doing their homework, MediaMOO is, hopefully--we try at least--populated by a group of people with a shared intellectual interest--in this case, the study of media. It is a bit like an endless reception for a conference like this one, where you can meet people like the people in this room and talk with them about issues of common interest.
We tried an experiment on MediaMOO from November 1993 to September 1994 with democratic control of the community--whatever "democratic" means, which we will come back to again. There are some incidental factors as to why this went wrong and there are some broader lessons. Obviously, for the purposes of today, I will try to talk about the broader lessons and not the incidental factors.
Why did we try a democratic experiment anyway? Well, that is my fault; it was probably my idea. I was particularly interested by what I saw as a fundamental failure of direct democracy on LambdaMOO. Now LambdaMOO is the largest MOO, which is a particular kind of MUD on the Internet, filled with 200 (or more) people logged on at any given time. There was, as some of you might have read in Julian Dibbell's wonderful article in the Village Voice, a kind of virtual rape on LambdaMOO, which led to one of the managers of the community (or wizard), to summarily cancel the account of the offender. This led the community to say, "Can the wizards just get rid of people? Is this wizard judge and jury? What's going on here?" This, in turn, led the founder of the community and author of the MOO software, Pavel Curtis from Xerox Park, to say, "You know what? I think we need to get the wizards out of the business of having any kind of social control here. Let's have a direct democratic process to manage that from now on. The wizards will just be technical people who work in the background and stay out of these social issues." And one of the ongoing violently debated issues on LambdaMOO is to what extent are the wizards living up to that ethic of staying out of social affairs and to what extent is that possible.
On LambdaMOO, there is a system of petitions. If you get enough signatures on your petition, it becomes a ballot. If the ballot gets a two-thirds vote of the population, then it becomes a law. There is also an arbitration process, which was set up by a ballot, to handle disputes between people. There is a whole separate process involving an architecture review board which decides who gets more quota and what goes into public spaces. There are separate elections for those people [the architecture review board]. Quite an elaborate system has evolved. It is an elaborate system which, in my opinion, doesn't work particularly well.
Basically, it is a flame-fest; it is a horrendous flame-fest. I wanted to try to give you a feeling for what a horrendous flame-fest it was so what I did was pick a random message. I generated a random number between one and 1,000 and took one of the last 1,000 messages on the Social Issues mailing list on LambdaMOO. I picked this totally randomly--this is completely representative. I want to read you that message (I'll cut off some preface here):
The larger issue here is about what type of society we shall have--elite rule by the political class or individual rights [This is starting off sounding kind of interesting. People are doing this in their spare time for fun]. It is sad and depressing to see someone respond to my post calling Idagi a fascist bitch by declaring that since she might soon become a wiz we should be nice to her. For those who hold such short-sighted views, when will you stand up for yourself? When LambdaMOO Takes a New Direction is taken away? When arbitrary wiz rule is imposed? Why do you want to suck up now? Such fools! In the future you will find yourself either slaves or recycled. Now is the time to demand a society based on the individual and not the self-appointed judges of those who are good or bad. All of which reminds me, the wizzes involved in the theft of Roy Cohn's program or bit -- you have about twenty-four hours to return it and issue an apology to Starsosh or I will be forced to file disputes for your violations with LambdaMOO Takes a New Direction.
As it turns out, this message is totally representative. People get called fascist bitches on Social Issues every day. Actually, that is relatively tame. So there is a problem. Here are people who are ostensibly working toward a shared goal and yet, in practice, it is just this horrendous flame-fest. The conversation is completely dominated by a few extremely vocal individuals who post at great length and if you wanted to respond to everything they said, it would take you all day. Sensible people stay clear of this process. I mean, who has time to mess with this? By definition, you have to be an idiot to mess with this because, if you weren't an idiot, you wouldn't be wasting your time reading stuff like this. So the whole political process is left to people who have nothing better to do than spend all of their free time in a flame-fest.
An additional problem, on another level, is that there is no broader vision. Because anyone can propose a petition or ballot, there is no one saying, "This is the direction we want to go. Let's make our policy changes move with a vision--with a coherence." Everything just kind of wanders. One ballot undoes another ballot and Lord knows what it all means. So direct democracy on LambdaMOO, in my opinion, really doesn't work.
I thought it would be interesting to experiment with representative democracy [on MediaMOO]. In particular, network computer systems have the potential to try something new--a kind of immediate accountability to constituents where the system is representative. If your constituents don't like what you have done, you can be voted out of office in an hour. The voting system on MediaMOO had an update rate of once an hour.
This didn't work either. I did quite a bit of reading though the old mail archives before preparing this talk. I came across what I think to be quite a prescient post very early on in the archive. One of the members of the community was interested in participating when we first started this whole representative council business. He posted a position paper describing how he wanted to be a member of the council. Then, two days into this whole process, he posted back:
Hi, again. After observing what has been happening on this MOO in the past few days--the campaigning, the power things, the joking and half-serious behavior toward candidature for the council, I have decided not even to attempt to run for it anymore. This has nothing to do with trying to sort out practicalities to make this MOO a more functional place anymore. The introductory post to this list by the candidates sound mostly like Clintonian or Bushian big politics [Remember, this was in 1993]. Somehow, I don't think I have a place in this and I have a suspicion that, in some cases, neither does the MOO.
Well, it took the rest of us a year to figure out he was absolutely right. I wish we had realized how prescient he was at the time.
I want to read you another post. We have the pleasure of actually having two of the MediaMOO counselors who participated in this experiment and devoted their sweat to it here today. Maybe they will have something to say during the question session. One of the counselors, Michael Day, did send me a letter which, with his permission, I am going to read. This is his view of what happened:
As I think I said in Berkeley last month, the hard lesson I learned from the failed experiment was that in our attempt to be inclusive and democratic about the process, we opened ourselves up to the possibility of being dominated by people who had more time to devote to posting lots of words to the news groups. Since I work full time and have plenty of other projects besides the MediaMOO Council, I never seem to have the time to answer the volley of messages that came my way. I felt that answering the objections once, and answering them succinctly, should be enough. But there was a sense that--like dealing with a child who keeps asking, ³Why?² every time you explain something--my responses would never be adequate because they did not measure up to the frequency and volume of words in the objecting posts. And, since many of us--my constituents and many of the fellow council members--agreed, I felt that we had strength in numbers. The problem was that we did not have the time to be as vocal as the opposition nor did we believe that the advisory role of the council should be a time-consuming endeavor. In a way, it felt like we were being out-shouted or bullied into submission. In my heart, I want to believe that consensus is possible and that all voices should be heard, but I found that consensus does not work unless members of the community hold a majority of values in common. I thought that MediaMOO had been founded on a principle of scholarly sharing and that one of its most unique qualities is that we all know each other's identities and addresses. I felt, and still feel, that we have the right to be different from anonymous MUDs and MOOs if we so choose, and that those not comfortable with this policy have the right to leave the community. Another observation: we have been without a council for quite awhile now and we seem to be doing okay. Granted, active membership is down a bit, but perhaps we never needed a council in the first place.
What kinds of issues did the council deal with? Unfortunately, one of the very first issues it had to deal with was a problem regarding what to do about an accusation by a member of the community that another member of the community was sexually harassing her. What made it even more horrendously complicated was that the victim was a member of the council and the victim didn't particularly feel like recusing herself from the process of debating what should happen . Although she did [recuse herself] to some extent, she was quite publicly vocal and had a degree of authority about the whole thing. So here is this fledgling group of people trying to work out a process for solving group decisions and the first thing they have to deal with is this horrendously complicated issue.
Another issue that was particularly divisive was the issue of admissions policy--who should we allow to be members of the community. My position has always been that MIT would not be MIT if everybody who wanted to come could come. And that the same goes for MediaMOO--that our having an admissions requirement is part of what makes the community special. But with the egalitarian flavor of the net, this does not go over easily. This is an incredibly controversial policy and people ended up arguing passionately about their views about equality and exclusivity. The discussion was tremendously productive if your goal was to reflect on these issues, but if your goal was to move the community forward in some fashion, it was tremendously time-consuming and divisive.
One of the big problems, first of all, was that I set up a voting system and said to the members of the council, "You guys can decide to change the voting system and I am not going to tell you how to decide things amongst yourselves. You are the council; you figure it out." That was a horrible mistake. There was one member of the council who insisted that it work on a consensus basis. If someone insists on consensus, then I guess it has to be consensus, because you would need a consensus to decide to use something other than consensus. You know what? Consensus works for Quakers. Quakers have shared values. Quakers have patience. They like to sit in silence. Quakers are nice people. They have goodwill towards one another. All of those things were missing on MediaMOO. Consensus works for the Media Lab faculty, interestingly enough, for a completely different reason. It works for the Media Lab faculty because, boy, if you piss these other people off, they can make your life miserable. There is a very high level of accountability amongst faculty members. Shared values, patience, goodwill, mutual accountability--all of these things are what makes consensus process work. All of these things were one-hundred percent lacking in this experiment. So that was one reason. Another issue was simply the time constraints. As Michael Day's message said, these people have real lives. Democracy takes time; the consensus process takes exponentially more time. Working out even insignificant issues seemed to take forever.
Now, I have to say here, I would be remiss in this presentation if I didn't mention that a big part of the problem--a lot of the fault--lies with me. I didn't know what I was doing. I just thought it would be kind of cool. I'm not a politician and I'm never giving this talk again. This is the one and only time. This is not my field of research.
Another one of the issues relates to control. I never entirely gave up control to the council. And, I don't think it would have been possible for me to do so because they weren't willing to accept it. But then there was this back-and-forth about who's really in control . People were throwing around accusations and my participation in the process was problematic. Now, all I did was, if people were talking about issue X, I might have posted a public message saying, "Well, gee, this is just my opinion, and the council is really in control, but, you know, I think about X that maybe this..." But just my voicing any opinion whatsoever was tremendously damaging because, as the founder of the community, I needed to just get out of it entirely. Even voicing a "this-is-just-my-opinion" kind of an opinion had a tremendous weight.
Also, I had a much greater commitment to the community than to the political experiment. The point of this community wasn't to pursue political issues. So, I wasn't willing to let the whole thing go down the tubes in the name of the political experiment because, to me, the political experiment was a kind-of-interesting little feature of the community, but not the core of it. I needed to maintain the basic integrity of MediaMOO as an experiment on different issues which I think are much more interesting and much more important.
Finally, you can't really have a democratic system unless the members of the community really own the machine. Unless the participants actually owned it, there were just some power factors that couldn't be gotten around. Anyway, those are incidental factors. I don't think they are of profound relevance, but I think I would be remiss in not mentioning them.
Maybe one of the most fascinating issues in all of this was why was there such lack of goodwill. Everybody on the council was absolutely at each other's throats. Members of the community were at each other's throats. We were debating totally trivial issues and tearing each other's eyes out about it. Why was there such a tremendous lack of goodwill? I think there are a number of issues.
One of them is that some people just have a problem with authority. They were transferring their general feelings of hostility towards the council members who were now new symbols of authority. So, if you have a teenage down-with-the-establishment kind of mentality, down-with-the-council is a pretty good outlet for that.
Second, as I mentioned before, this issue of exclusivity ties into a lot of people¹s most deeply held, and in my opinion, most naive values about pluralism and when it is and when it is not appropriate.
I think there was a degree of transference, not just about people's feelings toward authority, but about people's feelings about their general lives. I think someone's dog died or they hated their job or they hated their spouse and it was just a lot easier to yell at people through this political process and express their personal feelings of anger through that channel than to deal with the fact that the truth is that they hate their lives. The political process on MediaMOO became a safe place to have an outlet for a lot of anxiety with very few consequences.
And, of course, also relevant was the classic thing everybody talks about in on-line discourse: the lack of body language and expression and tone of voice played a tremendous role in it. It is really easy to forget that there is human being who means well on the other side of that computer terminal and to assume, instead, that it is the devil incarnate.
Bill Moyer (a sixties community organizer type, not Bill Moyers, the journalist) makes the interesting point that pathologically divisive people are indistinguishable from agents provocateurs. He has found this from his experience working in the Civil Rights movement and the Anti-Nuclear movement. I think we had our fair share of problems with people who fit that description.
That probably relates directly to the issues of transference that I was discussing. The content of a lot of the posts was not the issue, it was the tone and also the sheer volume of some people's contributions. If one person has all day to post at tremendous length about a particular issue, as Michael Day wrote in his message, you can't sit there and respond to every point because you have a real job and you have a real life. And yet how can you leave it unanswered when all of these horrible things have just been said?
After this whole thing was over, I found that that a professor at Brandeis, Carmen Sirianni, had mailed me a paper called "Learning Pluralism." It sat on my floor in a big pile with all of my other unread stuff. I read it about a year later on the beach and my jaw just dropped because I think, in a lot of ways, we were reinventing the wheel. Sirianni's paper is about people's attempt to use a kind of consensus-based process in the Civil Rights and Women's movements and about the kinds of social problems that emerge in groups. For instance, one of the things he focuses on is the issue that there is no such thing as a leaderless group. Very often assuming that you have no leader means that you end up with a de facto leader and, in some ways, it is more pernicious to have leaders that no one can name. Very often, the people who end up in control are the people who just have more time to commit. I think these issues have cropped up in a variety of movements and not just in this particular instance.
We eventually did dismantle the experiment. It had ended up becoming "all pain, no gain" and I think fundamentally there was too little at stake. We started all of this because we were interested in experimenting with political processes (because I was interested--maybe I inflicted it on everyone else). The point is that the issues we were addressing were fundamentally trivial. Not that many people really cared that much. Maybe if the issues were of greater importance people would have contributed more; people would have invested more. There was just too little at stake. They say there are two things you don't want to see the making of--laws and sausages. Democratic process is tremendously messy in how it proceeds and even messier when the people doing it don't quite know what they are doing. We do have one remnant of it--a membership advisory committee persists on MediaMOO which advises me on who to admit to the community and who not to. I did say that it was all pain and no gain and, at the time that we dismantled the experiment--about a year after it started, in 1994--I certainly felt that way. In retrospect, I think it was a tremendous learning experience for a lot of people who participated, particularly a learning experience (and this brings me back to where I started) about politics and democracy.
Let me return to the "D" word. What is democracy? Well, I'll tell you one thing: it is not one thing. It is a broad term for a collection of strategies, and just because you are one or aren't one--whatever that means--doesn't mean that your society is good or just. A lot of the passionate arguments that went forth in this whole political process were arguing that this isn't a democracy--implying somehow that if it were, everything would be wonderful. I mean, come on! Look around the real world, folks. I think people's concept of what democracy is and what it means to be one is tremendously vague. I think we all stopped in third grade and that is a failure of our educational system. And I think through this process, we all--I, certainly, and, I hope, other people--learned a tremendous amount about what democracy really is and how group processes really work. But I definitely would not do it again.
I just want to thank my advisor, Mitch Resnick, Pavel Curtis who wrote the MOO software, the former councilors who put up with so much nonsense, and the people of MediaMOO who put up with equal amounts of nonsense.
© 1996 by Amy Bruckman
Amy Bruckman's Abstract and Bio
