MMO’s and Conflict Dynamics


It is often said that arguing on the internet is a pointless endeavour. One needn’t delve deep to understand where the sentiment stems from. Take any comment section, whether it is Facebook, YouTube, twitter or a news website and you’ll often find the same dynamic. Not only are discussions rarely about insightful topics, those that are quickly become derailed. Most responses are short and vile. Having ran a facebook group I can tell you that one gets away with the feeling that you have to choose between a group which is highly unstable and eventually inactive  and one that is more active and has potential to grow in numbers but has a demography which tends to lend itself to vile comments where the goal is closer to procrastination than education.

In order not to lock yourself in a dynamic where the conversation always returns to the same stale topics and comments you’d have to introduce rules. These rules however tend to induce hostility by some of the most active, but least thoughtful, members. They also tend to burden those who mean well with the idea that their comment would not be good enough. As such it is hard to reach critical mass and retain growth. So far the problem that has been sketched is largely about demographics: how do you form a structure where the right people find your group and are motivated to write and share ideas.

But as any game theorist will tell you behaviour is in part defined by the options an actor has at it’s disposal. When faced with behaviour which is strongly deviant from real life experience we tend to blame anonymity or the bigger gap between those talking. We tend to blame the lack of facial expressions, body language and the feedback that this brings. I agree that these all play a (vital) role but I believe the discrepancy can not be fully reduced to these differences.

To understand why I believe it is informative to look at different social networks and even games. Because it turns out every MMO (Massive Multiplayer Online) has it’s own demographics, social structure, culture and there are big differences in behaviour between these games. There are also big differences within a game in behaviour. The evidence I have is anecdotal, though I believe that the experience from playing Guild Wars & Travian for several years and running a guild does help in assessing the important factors here.

I believe that every game, every forum is it’s own microcosm. And that we should in theory be able to correlate the behaviour of the players with the demographics and the actions the players can perform. A few examples of how I see this working:


  • In Guild Wars and Travian you have Guilds and alliances respectively. These allow people to group together and usually also include a certain hierarchy where officers are placed above members and are able to kick and invite members. This allows for a social control. If any of my members would have acted inappropriately I’d have reprimanded them and kicked them. People are free to leave and join guilds (important!) and this allows for an interesting dynamic. Choosing officers which are unpopular can cause people to leave (or get kicked for no reason). Officers also tend to be older, more experienced players who can work together with younger, more unexperienced players. Given that guild members are much more likely to help each other out and you’re often friends with many of your guild you tend to value membership. Obviously there are a lot of permutations (some people just hop from guild to guild, some are barely active) but the net effect it seems to me is that allowing for people to organize in this (hierarchical) structure induces a civility that is otherwise not present. I’d compare this to experiences of friend who play League of Legends where guilds don’t seem to exist or are less important and the atmosphere tends to be a lot more toxic. A significant portion of the community tends to quit, be it for a short while or permanently, due to these vile remarks.


  • In Guild Wars you had PvP (Player versus Player, competition between groups) in different formats. Here there is a clear difference in behaviour between RA (random arenas) where you are randomly assigned groups, HA & AB (Heroes Ascent and Alliance Battles) where you chose to join other people beforehand and can choose to work with the same people again, and GvG where you have to compete with your guild against other guilds. My conclusion is that randomness does matter. Anonymity matters somewhat. But the effect is smaller than you’d think. In a sense everyone can remain anonymous should he wish to do so. All you need for repeated games is the ability to contact each other, which occurs under a pseudonym.


  • If you look at YouTube or Facebook there tends to be less need for cooperation. Cooperation happens outside, IRL and there is no common goal on Facebook. So the people who do post regularly often have a bone to pick or time to kill. This difference in priorities combined with the lack of hierarchy and therefor social control it seems to me is just as important as anonymity. One can see this simply by looking at comment section on newspaper articles. Most people just post with their real name and picture.

So what seems to be important at first sight? I propose the following:

  • Possibility for cooperation
  • Possibility for repeat cooperation. Reruns with the same people.
  • Social structures and social control.
  • Proportional escalation. The ability to escalate in a way that is proportional if offense is given. If the only option is to mute/block people, this tends to exacerbate offensive behaviour.

What doesn’t seem to work ? I’d say the following:


  • Monetary (virtual or IRL) incentives. It seems this confirms Samuel Bowles his work in that with league of legends you get rewards for good behaviour. However, I suspect people just roll into a cost-benefit analysis then and value the ability to let of steam higher than the rewards you could realistically give. Therefor perhaps even worsening the situation.


I’d also remark that an additional problem is that the pool of potential people to cooperate with is so big that it devalues cooperation in a certain way. It doesn’t matter if you burn bridges with someone, you can always find others. Especially if those people tend to land on another side of the fence (different interests, different (political) opinion, or different guilds. For now, it seems that the civilising influence of guilds is stronger than the possible hostility between guilds. This seems to mimic the outside world where nationalism and religious activities seem to have much bigger gains in parochial altruism than the losses caused by potential hostilities between different groups (on average).

I therefor conclude that the possibilities a player or person has at it’s possibility, whether it is downvoting, commenting, friend-requests, trade or anything may play an important role in the way we behave. Expanding or limiting these interactions may allow us to steer social behaviour towards more favourable realities.

Hope you found this interesting!


Awaiting your feedback!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s