By the way, I will note that I am both quite surprised and, separately, something like dismayed, at how devastatingly effective has been what I will characterize as “Said’s privileging-the-hypothesis gambit.”
Like, Said proposed, essentially, “Duncan holds a position which basically no sane person would advocate, and he has somehow held this position for years without anyone noticing, and he conspicuously left this position out of his very-in-depth statement of his beliefs about discourse norms just a couple of months ago”
and if I had realized that I actually needed to seriously counter this claim, I might have started with “bro do you even Bayes?”
(Surely a reasonable prior on someone holding such a position is very very very low even before taking into account the latter parts of the conjunction.)
Like, that Vaniver would go so far as to take the hypothesis
Duncan has, I think, made it very clear that that a comment that just says ‘what are some examples of this claim?’ is, in his view, unacceptable”
and then go sifting through the past few comments with an eye toward using them to distinguish between “true” and “false” is startling to me.
“Foolishness,” Severus said softly. “Utter foolishness. The Dark Mark has not faded, nor has its master.”
“See, that’s what I mean by formally insufficient Bayesian evidence. Sure, it sounds all grim and foreboding and stuff, but is it that unlikely for a magical mark to stay around after the maker dies? Suppose the mark is certain to continue while the Dark Lord’s sentience lives on, but a priori we’d only have guessed a twenty percent chance of the Dark Mark continuing to exist after the Dark Lord dies. Then the observation, ‘The Dark Mark has not faded’ is five times as likely to occur in worlds where the Dark Lord is alive as in worlds where the Dark Lord is dead. Is that really commensurate with the prior improbability of immortality? Let’s say the prior odds were a hundred-to-one against the Dark Lord surviving. If a hypothesis is a hundred times as likely to be false versus true, and then you see evidence five times more likely if the hypothesis is true versus false, you should update to believing the hypothesis is twenty times as likely to be false as true. Odds of a hundred to one, times a likelihood ratio of one to five, equals odds of twenty to one that the Dark Lord is dead—”
“Where are you getting all these numbers, Potter?”
“That is the admitted weakness of the method,” Harry said readily. “But what I’m qualitatively getting at is why the observation, ‘The Dark Mark has not faded’, is not adequate support for the hypothesis, ‘The Dark Lord is immortal.’ The evidence isn’t as extraordinary as the claim.”
The observation “Duncan groused at Said for doing too little interpretive and intellectual labor relative to that which he solicited from others” is not adequate support for “Duncan generally thinks that asking for examples is unacceptable.” This is what I meant by the strength of the phrase “blatant falsehood.” I suppose if you are starting from “either Mortimer Snodgrass did it, or not,” rather than from “I wonder who did the murder,” then you can squint at my previous comments—
(including the one that was satirical, which satire, I infer from Vaniver pinging me about my beliefs on that particular phrase offline, was missed)
—and see in them that the murderer has dark hair, and conclude from Mortimer’s dark hair that there should be a large update toward his guilt.
But I rather thought we didn’t do that around here, and did not expect anyone besides Said to seriously entertain the hypothesis, which is ludicrous.
(I get that Said probably genuinely believed it, but the devout genuinely believe in their gods and we don’t give them points for that around here.)
Again, just chiming in, leaving the actual decision up to Ray:
My current take here is indeed that Said’s hypothesis, taking fully literal and within your frame was quite confused and bad.
But also, like, people’s frames, especially in the domain of adversarial actions, hugely differ, and I’ve in the past been surprised by the degree to which some people’s frames, despite seeming insane and gaslighty to me at first turned out to be quite valuable. Most concretely I have in my internal monologue indeed basically fully shifted towards using “lying” and “deception” the way Zack, Benquo and Jessica are using it, because their concept seems to carve reality at its joints much better than my previous concept of lying and deception. This despite me telling them many times that their usage of those terms is quite adversarial and gaslighty.
My current model is that when Said was talking about the preference he ascribes to you, there is a bunch of miscommunication going on, and I probably also have deep disagreements with his underlying model, but I have updated against trying to stamp down on that kind of stuff super hard, even if it sounds quite adversarial to me on first glance.
This might be crazy, and maybe making this a moderation policy would give rise to all kinds of accusations thrown around and a ton of goodwill being destroyed, but I currently generally feel more excited about exploring different people’s accusations of adversarialness in a bunch of depth, even if they seem unlikely on the face of it. This is definitely also partially driven by my thoughts on FTX, and trying to somehow create a space where more uncharitable/adversarial accusations could have been brought up somehow.
But this is really all very off-the-cuff and I have thought about this specific situation and the relevant thread much less than Ray and Ruby have, so I am currently leaving the detailed decisions up to them. But seemed potentially useful to give some of my models here.
I think you are mistaken about the process that generated my previous comment; I would have preferred a response that engaged more with what I wrote.
In particular, it looks to me like you think the core questions are “is the hypothesis I quote correct? Is it backed up by the four examples?”, and the parent comment looks to me like you wrote it thinking I thought the hypothesis you quote is correct and backed up by the examples. I think my grandparent comment makes clear that I think the hypothesis you quote is not correct and is not backed up by the four examples.
Why does the comment not just say “Duncan is straightforwardly right”? Well, I think we disagree about what the core questions are. If you are interested in engaging with that disagreement, so am I; I don’t think it looks like your previous comment.
(I intended to convey with “by the way” that I did not think I had (yet) responded to the full substance of your comment/that I was doing something of an aside.)
By the way, I will note that I am both quite surprised and, separately, something like dismayed, at how devastatingly effective has been what I will characterize as “Said’s privileging-the-hypothesis gambit.”
Like, Said proposed, essentially, “Duncan holds a position which basically no sane person would advocate, and he has somehow held this position for years without anyone noticing, and he conspicuously left this position out of his very-in-depth statement of his beliefs about discourse norms just a couple of months ago”
and if I had realized that I actually needed to seriously counter this claim, I might have started with “bro do you even Bayes?”
(Surely a reasonable prior on someone holding such a position is very very very low even before taking into account the latter parts of the conjunction.)
Like, that Vaniver would go so far as to take the hypothesis
and then go sifting through the past few comments with an eye toward using them to distinguish between “true” and “false” is startling to me.
The observation “Duncan groused at Said for doing too little interpretive and intellectual labor relative to that which he solicited from others” is not adequate support for “Duncan generally thinks that asking for examples is unacceptable.” This is what I meant by the strength of the phrase “blatant falsehood.” I suppose if you are starting from “either Mortimer Snodgrass did it, or not,” rather than from “I wonder who did the murder,” then you can squint at my previous comments—
(including the one that was satirical, which satire, I infer from Vaniver pinging me about my beliefs on that particular phrase offline, was missed)
—and see in them that the murderer has dark hair, and conclude from Mortimer’s dark hair that there should be a large update toward his guilt.
But I rather thought we didn’t do that around here, and did not expect anyone besides Said to seriously entertain the hypothesis, which is ludicrous.
(I get that Said probably genuinely believed it, but the devout genuinely believe in their gods and we don’t give them points for that around here.)
Again, just chiming in, leaving the actual decision up to Ray:
My current take here is indeed that Said’s hypothesis, taking fully literal and within your frame was quite confused and bad.
But also, like, people’s frames, especially in the domain of adversarial actions, hugely differ, and I’ve in the past been surprised by the degree to which some people’s frames, despite seeming insane and gaslighty to me at first turned out to be quite valuable. Most concretely I have in my internal monologue indeed basically fully shifted towards using “lying” and “deception” the way Zack, Benquo and Jessica are using it, because their concept seems to carve reality at its joints much better than my previous concept of lying and deception. This despite me telling them many times that their usage of those terms is quite adversarial and gaslighty.
My current model is that when Said was talking about the preference he ascribes to you, there is a bunch of miscommunication going on, and I probably also have deep disagreements with his underlying model, but I have updated against trying to stamp down on that kind of stuff super hard, even if it sounds quite adversarial to me on first glance.
This might be crazy, and maybe making this a moderation policy would give rise to all kinds of accusations thrown around and a ton of goodwill being destroyed, but I currently generally feel more excited about exploring different people’s accusations of adversarialness in a bunch of depth, even if they seem unlikely on the face of it. This is definitely also partially driven by my thoughts on FTX, and trying to somehow create a space where more uncharitable/adversarial accusations could have been brought up somehow.
But this is really all very off-the-cuff and I have thought about this specific situation and the relevant thread much less than Ray and Ruby have, so I am currently leaving the detailed decisions up to them. But seemed potentially useful to give some of my models here.
I think you are mistaken about the process that generated my previous comment; I would have preferred a response that engaged more with what I wrote.
In particular, it looks to me like you think the core questions are “is the hypothesis I quote correct? Is it backed up by the four examples?”, and the parent comment looks to me like you wrote it thinking I thought the hypothesis you quote is correct and backed up by the examples. I think my grandparent comment makes clear that I think the hypothesis you quote is not correct and is not backed up by the four examples.
Why does the comment not just say “Duncan is straightforwardly right”? Well, I think we disagree about what the core questions are. If you are interested in engaging with that disagreement, so am I; I don’t think it looks like your previous comment.
(I intended to convey with “by the way” that I did not think I had (yet) responded to the full substance of your comment/that I was doing something of an aside.)