Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

# MrMind comments on Double Crux — A Strategy for Resolving Disagreement - Less Wrong

61 29 November 2016 09:23PM

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

Sort By: Best

Comment author: 01 December 2016 04:07:54PM *  0 points [-]

Let me rephrase: does the double crux method contains any improvement that is not already covered by tabooing terms? Or simply saying "why do you think this is the case?"

"Steel is better than aluminum because aluminum is worse than steel" is also equivalent, adheres to the letter of the prescription but does not move the discussion forward.

What I'm trying to prove wrong is that the logical prescriptions, given as explanation of what a crux is, do not really capture anything substantial.

Comment author: 02 December 2016 08:00:03AM 0 points [-]

Let me rephrase: does the double crux method contains any improvement that is not already covered by tabooing terms? Or simply saying "why do you think this is the case?"

In this particular argument, no. (In fact, if both participants are willing to examine their own chain of reasoning and consider that they might be wrong, then asking "why do you think this is the case?" sounds like a perfect first step in the double crux method to me)

In cases where the disagreement is due to (say) Bob making a mathematical error, tabooing terms is unlikely to reveal the error, while double crux seems likely to do so. So, as a general disagreement-solving technique, it seems powerful as it can be applied to a wide variety of causes of disagreement, even without knowing what the cause of the disagreement actually is.

Comment author: 01 December 2016 06:44:01PM *  0 points [-]

If I'm understanding correctly, I think you've made a mistake in your formal logic above—you equated "If B, then A" with "If A, then B" which is not at all the same.

The search for a double crux encourages each side to adopt the causal model of the other (or, in other words, to search through the other's causal models until they find one they can agree is true). I believe "If B, then A," which is meaningfully different from your belief "If ¬B then ¬A." If each of us comes around to saying, "Yeah, I buy your similar-but-different causal model, too," then we've converged in an often-significant way, and have almost always CLARIFIED the underlying belief structure.

Comment author: 02 December 2016 07:56:32AM 0 points [-]

If I'm understanding correctly, I think you've made a mistake in your formal logic above—you equated "If B, then A" with "If A, then B" which is not at all the same.

No, he only inferred "If A, then B" from "If not B, then not A" which is a valid inference.

Comment author: 02 December 2016 10:06:50PM *  0 points [-]

1) if B then A

2) if not B, then not A. Which implies if A then B.

... but then he went on to say "How can an equivalent argument have explanatory power?" which seemed, to me, to assume that "if B then A" and "if A then B" are equivalent (which they are not).

Comment author: 05 December 2016 08:48:55AM 0 points [-]

I read that statement as implying that argument A is equivalent to argument B. (Not (1) and (2), which are statements about arguments A and B)

And, if A implies B and B implies A, then it seems to me that A and B have to be equivalent to each other.