= 27d567bc2d48d9f40e9ccc20ea370b15)
see comments on Harry Potter and the Methods of Rationality discussion thread, part 15, chapter 84 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (1221)
Hmm? We have no good evidence to distinguish between the following two hypotheses:
All we know is Quirrell has let hints drop that he was the hero who disappeared. There is no reason to expect that any of his hints are anything other than deliberate lies. If a competent investigation would discover that Qurrell's not really Qurrell, then the deception absolutely requires a second layer to last the year, so people like Bones can feel satisfied that they've discovered "the truth" about Quirrell without suspecting he's Voldemort. The existence of this second-layer deception now does not provide any evidence that the same deception existed eighteen years earlier.
Quirrell certainly talks about the need to act exactly as the person you're impersonating would act. His speech to Hermione would be no evidence at all if it were delivered by someone who practised what Quirrell preaches.
But that isn't Quirrell. Far from putting up a perfect facade, Quirrell's mask is constantly slipping. He "makes a game of lying with truths, playing with words to conceal his meanings in plain sight." His dialogue is peppered with hints to his identity, his past, and his intentions. Almost everything he says about himself is a clue.
His love of the killing curse and his intent to kill. His childhood ambition to become a Dark Lord. The Muggle dojo. The Pioneer plaque. His intention to crush Rita Skeeter. Repeated use of the word 'Riddle'. His willingness to be identified as having eaten 'death'. His wish for Britain to grow strong under a strong leader. The story of Merope's enslavement of Tom Riddle Sr. His theft of Quirrell's body using incredibly dark magic.
I think you've confused the actual character of Quirrell with the master of deception that he claims to be. When he tells Hermione about the time he spent as a hero, that is evidence that what he's relating is a twisted version of the truth. Because it usually is.
Incidentally, while I was collecting links, I discovered that Quirrell foreshadowed this after all.
I read this as his saying Voldemort has previously played the part of a hero. And, as above, I think it's probably true. What's your take?
Yeah we do. When EY writes that the heroic Scion of X vanished while traveling Ablania in 45 he is telling the readers that Voldemort took him by making a shout out to what happened to Quirrell in canon.
The Ablanian Shuffle is good evidence.
I guess it depends on your definition of "good". Care to quantify yours?
I guess you should quantify your own definition of the word, perhaps in the same post in which you ask someone else to quantify theirs, since you used it first.
I'd say p>0.95 that "Went on a graduation tour abroad and disappeared while visiting Albania." is meant to communicate something to the readers that it does not communicate to the characters.
I'd say p>0.75 that the thing it is meant to communicate is that the hero was compromised by Riddle, like Quirrell was in canon.
I don't expect it to be the same. Voldemort's shade in canon may have had possession capacity that young Tom Riddle did not.
I'd say p>0.5 that the hero was replaced, that Tom Riddle physically played both roles in his own flesh.
Your turn.
(I asked you to quantify what you meant by "good" because I was suspecting you were treating a probability of, say, 30% as "good", and we were getting our terms crossed. Obviously not.)
Whereas I'd put that at roughly p=0.25.
I mean, sure, it might be trying to communicate that, but, I've got:
"Reader! She's about to undercover the Defense Professor is Voldemort!" as a message intended to be sent to the reader but not the characters at about p=0.25.
"The heroic Slytherin discovered something about Riddle in Albania in 1945, and spent his time trying to follow up on it. When Voldemort came back openly to Britain, so did he. What the hero learned in 1945, or in the years between 1945-1970, is going to be important to Harry's defeat of Voldemort, and here's the hint that keeps it from coming entirely out of the blue" (or variations of the theme) as about p=0.15
"The 'heroic' Slytherin died in 1945 in a confrontation with Riddle/Voldemort. In 1970, an ambitious person unconnected to Voldemort then tried to exploit the Voldemort's rise as a chance to make himself leader of Britain under the dead man's name, and died or quit in 1973. Voldemort then found it useful to try the same con as a backup for Quirrel." at roughly p=0.15
And, "Eilizer is planning to do something else with it, that I haven't thought of" at about p=0.2
Fantastic.
I dismiss the bait and switch because the passage does not seem to lay down that tease; p<0.01. It starts right out with a date that we now know is meant to assure us this is not Riddle. If EY originally intended the bait and switch, then regretted it, p>0.8 he would clean out other things that only exist to support his ill conceived tease. There isn't a WHAM paragraph with few words surrounded by white space. It's just not built like a bait and switch shocker.
While reading, I thought that Scion of X did fight Riddle and did as Hermione suggested:
And after Voldemort killed him he kept the identity close because things like that can be useful. But I know that I am gullible and literal (p>0.2 that I under value literal interpretations after an alternative is available), so I dismissed that as soon as I thought up an explanation that worked on a more in character plot. p<0.01
I dismiss the unknown, unrelated, unremarked third party because of Conservation of Detail. p<0.01
I don't have any other speculation worth mentioning, so "something else" gets p<0.25.
If 1970-1973 was a con by Voldemort, why was it given up in 1973? Surely he expected it to take longer than a couple of years to begin with, didn't he?
I don't know how long he thought it would take, but it sounds like he had no idea how hard it would suck.
What other things?
That is, if the bait-and-switch was intended, he would've had to come up with an actual character that fit all those facts as well, and it seems like "he spent seven years sleeping in the same room as Voldemort" is a non-trivial detail to change.
The Albanian Shuffle. See says there is a real chance that it is mentioned just to string the reader along and make us think Bones is about to say that Quirrell is Riddle.
I dismiss this because EY changed the date, which comes at the top of the passage, just so readers wouldn't jump to think Bones is talking about Riddle. If EY took such a step to prevent the tease that Bones was about to name Riddle, then I would expect EY would not leave things in that were only there to build up that tease.
So the Albanian Shuffle is dismissively unlikely to be referenced for the sake of making the reader think Bones was about to name Riddle. I really don't know how you could think that in the first place unless you first read that paragraph after already thinking that Bones was going to name Riddle.
Before the date change, there was a legitimate chance that the reader would come away from the discussion thinking that the person Bones was describing actually was Riddle, and that both Bones and Quirrell understood her to have been talking about Riddle. Which if unintended is a far greater problem than "thinking Bones was about to name Riddle, then it turns out no". This was, in fact, my reading when I was actually going through the chapter.
(tl;dr: It's not a "tease" that Bones was about to name Riddle that's the problem, the problem is that it wasn't resolved with a clear indication that they're not talking about Riddle)
Changing the date fixes this because the reader can go look it up and realize that it can't be Riddle after all.
"OhmygodohmygodOHMYGOD! Bones is going to figure out Quirrell is Voldemort! OHMYGOD! What's he going to do?!?! He's surrounded by aurors, he's in DMLE headquarters!... Oh my GOD! Those aurors are so screwed!!"
looks up Tom Riddle online because that's totally what all readers would do
"Oh, hm. That's not Riddle then. I wonder who it is?"
...
Are you really suggesting that EY means the reader to do this? He said he wasn't going to lie to us anymore. See's low-probability theory of tease and WHAM involves EY lying to his readers, but your take on it that they were supposed to be totally tricked until the look it up online (?!?!) is turns that up to ridiculous levels.
That would only have changed if the year he started Hogwarts changed, which it did not. The birth date didn't change by a whole year, just from late enough in 1926 to enter Hogwarts in 1938 to early enough in 1927 to enter Hogwarts in that same year.
Yes. Exactly. That's my point.
(Not sure why you said this.)
(I lost track of what you were trying to argue, and the comment in isolation seemed to suggest that the non-trivial change had happened. A clause like "so the fact that this was carefully kept constant is evidence in favor of ..." would have helped. )
On the contrary, the reference to Albania is almost certainly a clue to the reader that the hero was replaced.
Almost certainly?
It's suggestive, sure, that he showed up from a long disappearance at about the same time as the "First Wizarding War" began. But the year could just be the one he returned at for some other reason, like, say, his Muggle Albanian wife dying.
In the latter case, why 1970? Because, like the original coincidental 1926 birth year, Eliezer was trying to make people go, "'Oh no, is she about to identify Voldemort?' . . . to be contradicted soon after by the Gaunts not exactly being on the Wizengamot or having a patroness grandmother."
Anyway, if you really think it's "almost certain", I'd like to arrange a bet on that. Say, my $5 against your $100?
You'd lose. Canon!Quirrell was possessed by Voldemort in Albania.
Canon!Riddle got Ravenclaw's diadem out of a hiding place in Albania circa 1945. Thus the inclusion of all four details — 1926, 1945, Albania, and 1970 — can all be explained as part of the same "Oh no, is she about to identify Riddle/Voldemort?" fake-out that was then deliberately blown up by the inconsistency with the Gaunts.
None of them need a separate cause to explain why Eliezer included them; all of them are explained sufficiently by the author-confirmed fake-out without adding the hypothesis "The hero who showed up in 1970 was a trick by Voldemort."
But if you're so sure I'm wrong, how about I put up my $5 against, say, your $500?
It would be more like your 1¢ against my empty instant ramen packaging, so no.
If you're that eager to lose money, loserthree can have it.
Thank you, Pedant One.
I will make three excuses for not taking the gracious offer of $5.00 (pre-tax) from the Holy See of Whereever, then I will give you a real answer.
I kind of lied: the status matter is totally part of the issue. But the real reason is that I have been conditioned to overvalue losses to gambling. I can only play poker by convincing myself it isn't gambling.
Really, though, isn't that gauche? Does it feel right to taunt a stranger with two-thirds of a fast food meal?
Even Urban Dictionary is no help with this one. What?
It was totally non sequitur. Also very old. Maybe obscure.
http://www.bash.org/?10739
I think it's still in the top 200 quotes on the site.
Oh, first google result if I hadn't used quotes. Duh.
Is that really it? Would you have been happier with, say, $20 against $400? I generally think of "almost certain" as indicating p≥0.95; the $5 was driven by the math of keeping your potential loss down.
And, not that I'm asking for you to actually answer, but ask yourself - is it really that you overvalue losses from gambling? If I were offering to put $2,000 up against your $100, would you still refuse because of the exact same chance you'd lose that exact same $100?
(I raised the odds to reflect p≥0.99 for pedanterrific because he implied that I was being a sucker for offering 1-for-20 odds.)
If I understand you correctly, you're saying that unless I am willing to tie significant amounts of money up for the sake of winning tiny amounts of money, I lack the confidence I claim.
It that didn't have the sound, the feel of a scam with it's a-sure-thing-isn't vibe it'd be a passable bullying tactic for the mathematically adept sort with something to prove. I've pushed people around for a living, so I like to think can appreciate a good push.
Or maybe that scamishness is part of the push?
The mark thinks himself a rational fellow, so he's unlikely to bolt. Bringing money into it makes the mark nervous. The mark mistakes his nervousness about money for doubt about his claim. You capitalize when the mark starts backing down, and claim some petty victory.
Seems like an awful lot of work for a small show, but thanks for the trick. Maybe I'll make something of it.
No, not at all. Not being willing to tie up the money is a perfectly sensible reason to refuse the bet. Opportunity cost isn't remotely connected to confidence levels; that I quite confidently expect a Treasury bond to pay me the promised interest doesn't mean I'd rather spend the money on something else that I value more.
And you certainly don't owe me an explanation of any kind as to why you refuse a bet. You merely owe yourself a good one instead of a bad one.
Then it would seem improbable that I'm putting the work in for the small show you identified, no? Not impossible, of course, but maybe you need a better theory of what I was trying to do.
I'm not saying it can't be used as a status attack the way you're suggesting, but this is a thing people do here. Something about calibrating confidence levels.
It's better than that - the classic con is to make the mark feel like he's putting one over on you.
The author has suggested we pay attention to the Conservation of Detail. With that in mind, the involvement of Albania is enough for almost certainty.
The detail is already conserved by its known use to try to make the reader suspect Bones is going to discover Quirrell is Riddle/Voldemort. Now the question left is whether Albania is a Chekhov's Boomerang, or whether theories based on it are an example of Epileptic Trees. I know I'm not "almost certain" either way.