I am sure most readers of this essay know well the “community cycle” – the way on-line communities get started, they thrive with peaceful, civil conversations, helpful strangers and kind atmosphere, and later how suddenly some seemingly innocent misunderstandings grow into flamewars, people stop listening to each other and only want to win the fight. And then some other community is started. Or an Open Source project gets started, people get excited, contributor add new features, volunteers write documentation, everybody answers questions at mailing lists and suddenly a difference in opinions about some technicalities grow into a fight and community is split with a fork. It is sad to observe that pattern repeating over and over again and it is also intriguing how it can happen so repetitively when everyone involved knows how it ends. Why does that happen?
According to René Girard a similar cycle of peace and conflict was the basic dynamic of archaic human communities. I am not an anthropologist, but it strikes me how his descriptions of ‘mimetic conflict’ fit the reality of on-line flamewars (or ForestFire or many other modes of on-line strife).
Whether the violence is physical or verbal, an interval of time
passes between each blow. And each blow is delivered in the hope
that it will bring the duel or dialogue to an end, constitute the coup
de grace or final word. The recipient of the blow is thrown
momentarily off balance and needs time to pull himself together, to
prepare a suitable reply. During this interval his adversary may
well believe that the decisive blow has indeed been struck. Victory -
or rather, the act of violence that permits no response – thus
oscillates between the combatants, without either managing to lay
final claim to it. Only an act of collective expulsion can bring this
oscillation to a halt and cast violence outside the community.
While Girard above states that only the collective act can halt the exchange of violence between the two antagonistic sides, I think that the blows exchanged are not without importance, eventually it is the public that chooses the final victim, but they will usually choose the side that fails. This would explain the importance of the public in on-line conflicts, people know that he violence have started and they feel being watched for any sign of weakness.
In a way flamewars could be viewed as a modern stichomythia, a literary technique that is:
exchange of insults and accusations that corresponds to the
exchange of blows between warriors locked in single combat
In our contemporary society we are well guarded against the escalation of violence linked to mimetic conflict and we have lost many intuitions about it, that is why on-line conflicts take us by surprise. We are not aware about the many mechanisms that protect us and we have not copied them on-line making our on-line assemblies much alike the archaic societies that are described in Rene Girard works. We need to relearn these intuitions and rediscover the social mechanisms if we want to consciously build more durable on-line communities. This does not mean that our situation on-line is identical to that of a human in the ancient world, there is one crucial difference – we don’t kill each other via the Internet, it all works only in the sphere of symbols but it seems that there is a similar group dynamic at work. This dynamic is not deadly (as the off-line can be), but it is annoying, wasteful and often cause community splits and in effect the fragmentation so well known in the Open Source world. If we learn how it works we could prevent much of it.
We can use the mimetic theory to: explain the nature of on-line conflicts, understand how humanity copes with conflicts off-line and try to copy the mechanisms on-line, interpret currently used on-line counter measures and inspire the development of new techniques specially suited for the on-line world.
According to the theory the first counter-measure discovered was finding a scape goat, a common enemy that unifies the community, this might explain the popularity of the vi-emacs wars. This solution worked for most of the history of our species, but I hope I don’t need to explain that it is somehow unacceptable for our civilized point of view. It is also not very effective online, because the solution is never as ultimate as in off-line circumstances, it is not possible to silence people on the internet even if they are banned from the community spaces they will easily find another outlet for their voice, and everything is one click away on the internet.
Another counter-measure is reject comparison, refuse to enter the competition for the best text editor, solution for a problem, programming code. This is my interpretation of TIMTOWTDI. Comparing code, argumenting about the best solution etc. is practically never done in void, it is always about why my solution is better than your solution, and that can so easily start the mimetic circle of attacks and counter-attacks. Once we understand where the power of TIMTOWTDI comes from we can formulate similar advice like trying not to link ideas to people so that they can be debated without the social balast, but that can be difficult.
We can also draw lessons from science and law, which have provided effective strategies to deal with conflict. That means agreeing on some minimal set of basic rules/believes and accepting that there are some objective ways to evaluate claims against that basic set of rules. This can be for example a set of objective measures (like speed benchmarks) and rules about evaluating software according to those measures. Or constitutionalizing the communities (like the Debian project). Effectiveness of these measures depends how much authority can be transferred into the initial basic set of rules, but functioning of our own society relies very much upon similar structures so we can be pretty confident about them.
A recurring characteristic of flame wars is how much offence they cause: in other words, how seriously participants seem to take them, no matter how trivial the object of contention is we seem to fall into the pit of boiling drama. I am not talking here about other people, I know it so well on my own example how seriously it all seems when you are in the diabolic circle. This also well resonates with the mimetic theory which says that the object of the mimetic rivalry loses it’s significance over the course of the conflict, often it is destroyed, but this does not end the rivalry which becomes self-sustaining at some point.
As the sacrificial conflict increases in intensity, so too does the
violence. It is no longer the intrinsic value of the object that inspires
the struggle; rather, it is the violence itself that bestows value on
the objects, which are only pretexts for a conflict. From this point on
it is violence that calls the tune.
This characteristic seriousness suggests another way to cope with the conflict, use humour for it can destroy the layers of drama and let us see our positions more objectively. I am convinced by my experience that this is a very effective way of restating the problem in a less personal way, but I have not yet encountered any theoretical analysis of humour and mimetism.
Another characteristic of the mimetic conflict is the increasing speed of alternations, and this also can be observed in the case of flamewars. It is quite reasonable to expect that slowing the discussion down can help the participants to break the vicious circle of conflict. The good thing about this measure is that it could be done automatically, the list server software could detect the increasing speed of discussion and automatically start delaying the emails.
At the very height of the crisis violence becomes simultaneously the
instrument, object, and all-inclusive subject of desire.
Finally perhaps the most important thing is not to make violence the object of desire. This can easily become the case when administrators of common resources are abusing their powers to show off or vent their anger by ‘baning’ people off them. Sometimes this can work in attracting new followers, who imagine themselves being the violent ruler, but it ultimately leads to a double bind situation and threatens the future evolution of the community.
All quotes above are from “Violence and the sacred” by Rene Girard (it’s google books page).
I’d like to thank Gabriella Coleman for numerous comments and editorial advice and Nicolas Messina whos initial remarks inspired me to rewrite my short blog post into this longer essay.