How's this for a method of reaching consensus?

Suppose a group wants to reach consensus on an issue, and wants individual interests to have as little as possible to do with the decision, and instead wants the group’s interests as a whole taken into consideration as much as possible.

So they do the following:

They meet anonymously online. A computer program knows who each individual is, but does not reveal this to anyone.

The participants discuss the issue, couching the argument in terms of group goals and interests as much as possible.

However, if any of the group can correctly identify any others, the one identified is kicked out of the discussion. (If an attempt is made to identify, and fails, then the one who made the attempt is kicked out instead.)

If too many people are kicked out, then the quorum is lost and the group has to reconvene after a set interval, with all participants re-invited to the new anonymous discussion.

What kind of group would this be for, if any? (I’m imagining probably an ideological one, but maybe my imagination is failing me.)

What would be some ways to game this system, if any? (I figured out one: If losing one more participant would result in loss of quorum, and you’re finding you’re not on the “winning” side of the discussion, you can throw out an identification, and right or wrong someone will get kicked and quorum will be lost.)

Does this even slightly resemble anything that’s actually been tried whether in reality or fiction?

What would be a snazzy name for this method of discovering group consensus?

I don’t think concealing individual identities is helpful in avoiding individual interests; it just makes it harder to see them.

I am sure his has been discussed to no end in the academic literature surrounding game theory. that is probably a good place to look to see what the variables are - maybe not what you think - and what variations there might be on your plan.

The identifying part slightly resembles the Mafia game.

Perfect concealment is the same as nonexistence. :wink:

Anyway, the idea isn’t to pretend to avoid individual interests altogether, the idea is to cause them to have as little effect as possible on the final decision. (Again, as per the OP, I’m not advocating that this is often or even ever a good idea… I’m just musing as to whether there could be a way to actually pull it off.) If you let your own interests drive your line of argument, you’re running the risk of being identified and so kicked out of the discussion. Best try to hide your interests as well as possible–and in this context that basically means letting them have as little effect as possible on the conversation.

Just musing here…

I think in practice the tendency is going to be for people to portray their own interests as if they were group interests. I suspect this will happen whether participants are known or unknown.

The problem I see is that this only works if people’s identities and interests can be matched. For (a completely ludicrous) example, the group is setting up a mission to colonize a distant planet. Everyone knows Alice is a lumber industry lobbyist, so she quickly gets identified by her proposals to build the spaceship out of mahogany. No one knows that Bob secretly spends his spare time watching fart fetish videos on Youtube, and while they think it a little odd that someone wants the colonizers to subsist on a bean-based diet, they don’t know who’s coming up with the suggestion, and might think it’s based on sound nutritional principles.

I think what they’re saying is that it’s not biased in favour of people who can discuss impartially; it’s biased in favour of people who can appear to do so.

Also, sometimes issues to which a person has an attachment can be legitimately important. What if only one member of the group was minority X, and someone made a minority-X-hating proposal? They wouldn’t want to speak up, and if the problem is one only they are aware of, the other members won’t be able to speak up.

Besides, what’s to stop a bloc from colluding using another communications method (e.g. SMS)?

The fundamental flaw, which **waterj2 **touched on, is that the proposed method confuses hiding identities with hiding agendas.

A better approach might be one which encouraged or forced all agendas into the open, rather than trying to hide them. Then an intelligent debate can be had on the merits of mahogany spaceships with the clear understanding that Alice’s views can’t be trusted.