Viliam_Bur comments on Firewalling the Optimal from the Rational - Less Wrong

86 Post author: Eliezer_Yudkowsky 08 October 2012 08:01AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (339)

You are viewing a single comment's thread. Show more comments above.

Comment author: hankx7787 08 October 2012 08:56:52PM *  4 points [-]

Coming from a hard-core Objectivist, the Objectivist community is unfortunately rife with all sorts of so-called "schisms". I think this is intrinsic in any community of thinkers who are focused on objectivity/optimality/rationality/etc in general, because inevitably people will feel differently on a given issue, and then everyone goes around blaming the other group that they aren't really objective or rational or optimal, etc.

This leads to me having to qualify a statement about some issue X with something like this:

As a result of pretty much universal confusion, here is a list of things I am not saying in this post:

I am not saying everyone who does not agree with X should or should not be "purged from Objectivism".
I am not saying people with varying views on different issues should or should not be called "objectivists".
I am not saying this group should or should not be limited to only "real objectivists".

Now it should be said of course that one group is actually right - but schisms are very unhealthy for any community, or any social group in general. The success of a social group per se is based very much on all-inclusiveness. That being said, identifying optimal, mainstream positions of a given philosophy is absolutely good for the philosophy per se.

So I would add something like: "firewall optimal philosophy from optimal community"

Comment author: Viliam_Bur 09 October 2012 07:22:52AM *  6 points [-]

Now it should be said of course that one group is actually right

I think this ignores the whole concept of probability.

If one group says tomorrow it will rain, and another group says it will not, of course tomorrow one group will be right and one group will be wrong, but that would be not enough to mark one of those groups irrational today. Even according to best knowledge available, the probabilities of raining and not raining could possibly be 50:50. Then if tomorrow one group is proved right, and another is proved wrong, it would not mean one of them was more rational than the other.

Even if we are not talking about a future event, but about a present or past event, we still have imperfect information, so we are still within the realm of probability. It is still sometimes possible to rationally derive different conclusions.

The problem is that to get perfect opinion about something, one would need not only perfect reasoning, but also perfect information about pretty much everything (or at least a perfect knowledge that those parts of information you don't have are guaranteed to have no influence over the topic you are thinking about). Even if for the sake of discussion we assume that Ayn Rand (or anyone trying to model her) had perfect reasoning, she still could not have perfect information, which is why all her conclusions were necessarily probabilistic. So unless the probability is like over 99%, it is pretty legitimate to disagree rationally.

Comment author: buybuydandavis 10 October 2012 09:20:58AM 3 points [-]

I think this ignores the whole concept of probability.

I thought it was ignoring the possibility that everyone involved could be wrong.

Worse, they could all be not even wrong.

Comment author: hankx7787 10 October 2012 06:51:01PM *  1 point [-]

You entirely missed the point of my including that statement.

My intention was merely to stress that I'm not merely trying to say something like, "nobody can every really know what the right answer is, so we should all just get along," or any such related overly "open-minded" or "tolerationist" nonsense like that.

My point was to say that such differences are perfectly fine and meaningful to fight about philosophically, but that you shouldn't use one's position on whatever derivative philosophical issues as the basis for community membership.

Comment author: TheOtherDave 09 October 2012 03:20:00PM 1 point [-]

Even if for the sake of discussion we assume that Ayn Rand (or anyone trying to model her) had perfect reasoning, she still could not have perfect information, which is why all her conclusions were necessarily probabilistic. So unless the probability is like over 99%, it is pretty legitimate to disagree rationally.

Hm. There's an implicit "...iff the disagreeer has access to better information than she had" here, right?

Comment author: Viliam_Bur 10 October 2012 07:10:03AM *  1 point [-]

There's an implicit "...iff the disagreeer has access to better information than she had" here, right?

If the disagreer has access to different information. Or just has different priors.

(I want to avoid the connotation "better information" = "strict superset of information".)

Comment author: TheOtherDave 10 October 2012 01:59:46PM 1 point [-]

Point.