I've been part of countless discussions in multiple forums and email threads. They are in general futile, since the type of forked discussion these media tend to foster suffers from a few problems:
- They tend to fork indefinitely. People will address specific parts of each post and derive a new discussion for each. Threads are never closed and even if at one point one thread ends the conclusion is never rolled back to the main discussion.
- People can choose to ignore parts of posts that are inconvenient to their claims. It is just too simple.
- People can repeat claims that have been refuted indefinitely.
- It is impossible to get a clear view of the discussion and where it is at, other than at a thread by thread level.
- It is impossible to keep a discussion on topic (yes, I'm the worst offender here with this post).
- Et cetera.
Both email, web forums and other mechanisms have the same weaknesses.
In fact, free flowing language discussions tend to follow a chaotic structure. Sometimes a response addresses the claims made in a previous post, sometimes it seems to do but it isn't evident and sometimes it´s just babble. Sometimes a response brings new debatable points which fork out and sometimes get addressed, sometimes they do not. Most of the time discussions get extended until people get bored and it is not very frequent that a conclusion is reached.
In verbal discussions, forking is contained by the ability to interrupt and our limited memory to track nested conversations. In written discussions they tend to be limited by lack of structure (inline responses at some point get unreadable, and sequential accumulative responses get meaningless to the reader) among other things.
I would almost claim that no productive conclusion was ever reached in this sort of activity. OK, maybe someone, at some point in the past reached one conclusion. But over a billion such discussions that doesn't seem like a high hit rate.
So I've been thinking on the possibility of building a more structured discussion mechanism that allows for constructively building a conclusion by different parties each contributing their own facts and reasoning. I haven't yet formed the perfect system (if I had, we would all be posting there, not here) but I've modeled many alternatives and the best I've found is as follows:
- Someone begins a discussion by making a statement. In this case the statement would be something like "Evolution has been falsified by Lenski's experiment with e.coli, so ID is the true explanation for life on earth".
- Other parties cannot edit that statement, but they can annotate it with comments addressing different parts of the statement. The different comments can in turn be annotated by other parties. For example someone could annotate the first claim as "I do not agree with this claim. Lenski's experiment was done over too few generations to show evolution at a large scale", which could be in turn annotated by someone else with "lenski did 40.000 generations" and so on until that particular fork is resolved.
- Additionally, statements can be flagged with things like "non sequitur", "citation needed", or other useful tags (especially for different types of well defined fallacies) that would make it possible for the poster to edit the statement to make it accepted by the others.
- When a note about a statement in the original claim or in any annotation becomes addressed and agreed on, it can be closed by the originator of that statement or comment by editing the original statement to a version that satisfies all parties. A statement can be closed when all parties that participated in it agree to close it.
- Annotations get thus rolled up to the original statement.
- In the end, a statement that is agreed by all parties is reached.
I know it sounds a little bit optimistic, but I've done simulations and it seems to work as long as the parties are few (building "teams" might be a way to scale the process to larger discussions) and all the parties have good will and the intention to reach a valid logical conclusion (and vandalism or lack of good will are extremely obvious in this format, so it is unlikely that someone will use it to "win" a discussion.
Such a process would obviously be much more complex than as described, but I'm convinced that such a discussion mechanism should be possible and that the benefits would be huge. I've seen some things that are somewhat aligned to what I'm proposing (some types of wikis and Google's Wave, for example) but they aim more at collaboratively building something than at resolving differences.
Having a practical means to have controlled discussions would be incredibly important, as most of our human conflicts seem to hinge on our inability to reach agreements through ordered conversations. Would it be the end of all wars? Maybe not, but it definitely would help finishing many endless arguments were parties have no vested interests and are exclusively interested in reaching a valid conclusion. And I think that would be worth something.
What do you think? Can anyone suggest an improvement to the proposed framework? Do you think something like this could work?
Discussions in related communities
|
|||||||||||||||||||||||||||||

