[SEQ RERUN] The Parable of the Dagger
Today’s post, The Parable of the Dagger was originally published on 01 February 2008. A summary (taken from the LW wiki):
A parable adapted from Raymond Smullyan, showing a failure of logic to reach a desired goal.
Discuss the post here (rather than in the comments to the original post).
This post is part of the Rerunning the Sequences series, where we’ll be going through Eliezer Yudkowsky’s old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Newcomb’s Problem and Regret of Rationality, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.
Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day’s sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.
I don’t understand the point of the parable. In the jester problem, the inscriptions didn’t limit the possibilities because the logical possibilities allowed the gold and frog to be anywhere. In the king problem, the king strategically omits asserting that the inscriptions have any relationship to what is on the boxes.
What’s the lesson here? (1) It’s possible to use formal logic to not say anything about the world. (2) Know your audience (i.e the king does not necessarily enjoy when someone yanks his chain). (3) Read the fine print, including what it doesn’t say. (4) The powerful can be powerful (cf. Meddle not in the affairs of wizards . . .)
As a lawyer, I can say that (3) is good advice . . . if you are dealing with a jerkwad. Lesson (4) is potentially useful to tax protestesters, but you don’t see many of them here, and that’s good for the average rationality of the community.
But it’s hard for me to conceive of any community like this one needing lessons (1) or (2). Adjusting upward the probability that I mis-solved the puzzles.
I found the parable amusing, but have to agree that the point is not particularly visible. I supposed it was some variation of “the map is not the territory”, here in the specific form of “don’t fully rely on provability of a proposition in your formal model (here, logic + axioms written on the boxes) as the model itself can be incorrect”. Which is mostly (1) in your classification.
A typo or a snarky name for those folks? (Reminds me of cdesign proponentsists.)
(I think that the jester problem only has one solution, the one that the jester intended.)
The important bit is this bit:
We’re just about to enter the “words” sequence and Eliezer wants to remind us that words carry no information unless we know why they were written.
Restatement: The meaning of words is entirely mediated by their entanglement with reality; beware of drawing new inferences about reality from arguments about words alone.
I have always interpreted it in a similar way, and I believe that this post is included on the wiki with the words sequence. However, when I posted it yesterday, a different interpretation occurred to me, given that it immediately followed Newcomb’s Problem and Regret of Rationality. In this interpretation, there is a “reasonable” reason to pick the second box, but that doesn’t matter. The point of logic is to avoid finding the dagger, and if it fails, use a different logic.
Sometimes the relevant logic is the one with this axiom: “The King is a fink!”
(I believe this may fairly be called Idiotic logic.)