Did you even read my goddamn footnote? Did you even read the comment that I am replying to? This is not my dichotomy, it’s the dichotomy of the comment I am replying to. Here is the original text in case you somehow can’t be bothered to read Benquo’s comment.
If members of a language community predominantly experience communication as part of a conflict, then someone trying to speak an alternate, descriptive dialect
Like, man, sure, if you want to claim this whole “class of frameworks is fake” then have a conversation with Benquo about it. No need to throw this much snark at me.
Benquo was referring to your own views in his comment, so this foisting-off of responsibility seems a bit odd. Insofar as he agrees with you, I disagree with him as well. I don’t see what’s weird about that.
I was replying to views expressed by you. If you think that I have mis-apprehended what views you in fact hold, by all means tell me; but that’s not what you seem to be saying?
That is why physical touch requires special signals in order to be permissible
Also, come on, on the object level it’s obviously the case that concern around violence absolutely shapes behavior around things like dating, casual touch, and an enormous number of other things involving touch.
This is… such a vague statement that I can’t disagree with it, as such… but it’s also not really responsive to what I wrote. (But if you did mean this as a contradiction of what I wrote, then—I disagree and think that you are mistaken.)
Also, the aggregate statement just seems false. The vast majority of touch in our society gets initiated without consent. When people flirt with each other they do not generally ask each other explicitly before holding hands, friends do not ask each other before leaning their head on other people’s shoulder, in many contexts people do not ask before shaking your hand, and family members routine hug each other without asking first. Touch is absolutely not governed by a universal environment of explicit consent, and people do not by default “dislike all non-consensual touch”. Sure, if you define “consent” as “ultimately accepted positively by the receiver” then sure, but that’s just begging the question.
I didn’t say anything about explicit consent. It’s true that “friends do not ask each other before leaning their head on other people’s shoulder”, but that doesn’t make this non-consensual! Of course it’s consensual. (Usually. But then, there are many cases when someone initiates this sort of “implied-consensual” touching, but then it turns out that consent wasn’t actually implied. Whoops! In such cases, the distinction between implied consent, and lack of consent, suddenly becomes quite salient.)
And it’s true that “family members routine hug each other without asking first”. It’s also true that many, many people experience this as a violation of their autonomy and their boundaries, and greatly dislike it. (Stuff like this is about as novel and surprising as “what’s the deal with airline food” at this point.)
People do not just “have a preference against being non-consensually touched”, many preferences about touch are present because touch involves physical vulnerability. Completely denying that just seems insane to me.
Perhaps if you understand “physical vulnerability” in a broad way. But otherwise, no, I do not agree. My point is that the possibility of violence usually has little to do with it.
But somehow trying to claim that costly differential signaling does not apply at all to touch, because none of the negotiation of physical touch is about literal physical vulnerability seems really very confused to me.
I wouldn’t say “none”, perhaps. But almost none.
I agree that norms around physical touch are different from norms around speech. I think it’s a decent but not perfect analogy, intended primarily to convey the structure of my argument, not as a social precedent in support of it. Feel free to choose from any other domain with asymmetricly large costs where much value is lost and costly differential signaling is required, even without the domain consisting “predominantly” of the costs.
Er, you’re the one making the analogy, so surely it’s up to you to choose a better example, not your interlocutor…
I agree that norms around physical touch are different from norms around speech. I think it’s a decent but not perfect analogy, intended primarily to convey the structure of my argument, not as a social precedent in support of it. Feel free to choose from any other domain with asymmetricly large costs where much value is lost and costly differential signaling is required, even without the domain consisting “predominantly” of the costs.
Er, you’re the one making the analogy, so surely it’s up to you to choose a better example, not your interlocutor…
I agree he could have chosen a better example. But like, are you trying to understand Habryka? Or are you just trying to litigate proper use of analogies? Your comment reads to me as 20% responding to his point, and 80% litigation. This sort of thing feels like an advanced version of arguing about definitions instead of just tabooing the word and talking about the underlying point each person is trying to make.
I think that habryka’s point, as described in the top-level comment of this thread and as explicated by the analogy he used, is wrong. If he wants to amend his claims to make them less wrong (possibly by choosing a different analogy and then reasoning on that basis, or possibly in some other way), that’s fine, he can do that, and I’ll be glad to read any such corrective commentary. But I can’t do that for him.
The idea that disagreeing with someone just means that the disagreer doesn’t understand and needs to work harder to try to understand needs to die a flaming death. Yes, sometimes misunderstanding happens, but sometimes people just think that you’re wrong. They understand what you’re saying and they disagree. Your task then is to convince, not to explain—and certainly not to whine about your interlocutors not working hard enough to convince themselves of your claims on your behalf.
Hmm, seems like I didn’t communicate well enough. Trying again.
I believe you understand and disagree with his point. I also believe you think his analogy is bad. When I first read your reply I thought you were disagreeing because of/due to the disanalogies, and for no other reason. I no longer think this.
Your disagreement of his point, and your critique of his analogy felt very mixed together. Which like, he’s using the analogy to explicate his point for a reason, so fair. But there’s something like a difference between using an analogy as supporting argument, and using an analogy just to point at the thing.
If you desire another analogy, most computer traffic is not malware or exploits, nevertheless it sure really matters a lot whether your specific message is malware or some kind of exploit.
As far as I can tell your comment doesn’t address this point directly? I’m asking for something like a clearer distinction between disagreeing with his point, and critiquing the analogy. Especially in this case where I don’t think the particulars of the analogy where central to his point.
I mean, the analogy is “bad” insofar as the point that it supports is wrong. Like, the two things which are claimed to be analogous in a certain important way, are in fact not analogous in that important way. (Or so I claim!)
(It’s like if I said “the sky is like a glass dome; if you fly high enough, you’ll crash into it; and since the sky is indestructible, much like a glass dome is, you also can’t break through it”. Well, no, you in fact will not crash into the sky; in this way, it is precisely not like a glass dome. And of course glass is totally destructible. The analogy successfully communicates my beliefs about the sky—that it’s a solid barrier which can be crashed into but not broken through. Those beliefs happen to be totally wrong. The glass dome analogy is “bad” in that sense.)
If you desire another analogy, most computer traffic is not malware or exploits, nevertheless it sure really matters a lot whether your specific message is malware or some kind of exploit.
As far as I can tell your comment doesn’t address this point directly?
True, I did not address that. I’ll do so now.
So, let’s recall what “it sure really matters a lot” means, specifically, in this context. The key claim from the earlier comment is this:
Communication doesn’t need to be “predominantly conflict” in order for it to be important to differentially signal that you are trying to have a more conflict focused or more descriptive-focused language
In the malware/exploit case, the analogous claim would be something like:
“Computer traffic doesn’t need to be ‘predominantly malware or exploits’ for it to be important to differentially signal that you are trying to send innocent, non-malicious data.”
Well… you can probably see the problem here. There’s basically two scenarios:
There exists a totally unambiguous, formally (which usually means: cryptographically) verifiable signal of a data packet or message being non-malicious. That signal gets sent, we check it, if it doesn’t check out we reject the data, the end. (If the signal can be faked after all, then we’re just fucked.)
There is no such verifiable signal. In this case, malicious traffic is going to be sending all the signals of non-maliciousness that “good” traffic sends. “A differential signal of innocence is being intentionally sent” is almost completely worthless as a basis for concluding that the data is non-malicious. Instead, we have to use complicated Bayesian methods to sort good from bad (as in email), or we have to enter into an arms race of requiring, and checking for, increasingly convoluted and esoteric micro-signals of validity (as in CAPTCHAs, UA sniffing, and all the other myriad tricks that websites use these days to protect themselves from abuse). (And any client that deliberately sends the signals we’re checking for is actually more likely to be a bad actor!)
This situation… is also not analogous to “posting on a public discussion forum”, which looks nothing like either of the above cases.
Hmm, seems like I didn’t communicate well enough. Trying again.
Aside: I do want to try to help you here, and warn you that it is my opinion Said Achmiz regularly cannot understand basic points and ideas and criticizes his interlocutors with extremely harsh language – regular unnecessary incredulity with extra question marks and exclamation points and italics, describing your positions as laughable, obviously wrong, deeply corrosive, etc – and I would encourage you to not continue trying to explain something to his satisfaction unless you would also find it a similarly worthwhile activity if your interlocutor was an LLM in whose system prompt it was written that it should not be able to either agree with or understand your point.
(You’re of course totally welcome to keep trying, I just felt I’d be irresponsible not to warn you.)
criticizes his interlocutors with extremely harsh language – lots of italics and harsh language like “needs to die a flaming death”
… implies that I was referring to some person or people with the phrase “needs to die a flaming death”. I hope you can see how totally unacceptable that implication is. (I won’t belabor the point that it’s false; you know that. But please correct the phrasing; falsely implying that I expressed a desire for a person to die violently is absolutely not ok, even if that implication was accidental, as I assume that it was.)
I would encourage you to not continue trying to explain something to his satisfaction unless you would also find it a similarly worthwhile activity if your interlocutor was an LLM in whose system prompt it was written that it should not be able to either agree with or understand your point.
Your position is laughable, obviously wrong, and deeply corrosive. I routinely succeed at explaining things to Achmiz’s satisfaction.
To illustrate, I’ll quickly pick two arbitrary examples from the top of my head. One, on this website in February 2023, I explained the existence of situations in which trying to minimize the damage from errors is preferable to trying to not commit errors. Two, in private correspondence earlier this month about an unpublished draft of mine, Achmiz wrote that an analogy I had used was “Confusing—because I don’t really get what sort of situation it can be analogous to [...] [C]onsider this objection to constitute a request for such examples!” I wrote up an explanation of a real-world situation of the class I was thinking of when I wrote the analogy, to which Achmiz responded, “Hmm … ok, I see what you mean [...] Anyhow, you’ve answered my question, yeah.”
To empirically test your claim about Achmiz vis-à-vis large language models, I supplied the explanation about minimizing damage from errors to Claude Opus 4 with the custom style prompt, “Emulate an obstinate interlocutor that should not be able to either agree with or understand the user’s point.” The difference between Achmiz’s actual response from February 2023 (“I see, thanks” and a relevant followup question which I also had no trouble answering) and the LLM’s response (“I don’t see how this example makes any sense at all” followed by another 200 words of blatant misreadings and non sequiturs) is quite apparent. Your claim that Morrison would find interacting with Achmiz and the LLM-prompted-for-obstinancy “similarly worthwhile” is clearly absurd. (And as it happens, Achmiz’s reply to Morrison seemed entirely cogent to me.)
I’m presenting this counterevidence to make a point, but really, I don’t think you believe what you wrote. The existence of examples of Achmiz being satisfied with explanations and the outcome of the LLM prompt were both easily predictable; you just didn’t think anyone would call your bluff on a hyperbolic insult. You were wrong, but more importantly: do you really think this is a good look for you or the website?
Good to know it’s ever happened! It is extremely uncommon in my experience of reading threads with Said in them.
Added: To be fair I wrote my comment upthread feeling my peak frustration about (what I read as) Said’s unproductive commenting style. I wrote to someone else at the time it would probably be my worst contribution to this ongoing conversation of hundreds of comments and hundreds of hours of conflict, so I’ll admit it is probably my peak sloppily-stated-times-aggressive comment and not to my usual standards.
Still, I think it’s more reasonable than you’re giving it credit for. I think I think it’s a fairly standard human phenomena to stick one’s heels in and be unwilling to budge on a position regardless of reason or argument, and I suspect Said of doing that from time to time (or alternatively of being quite dense), and you’ve told me before you also have felt that he seems sometimes to be kind of dense about very simple things, and I think people often do things they shouldn’t when they feel threatened, so I think you’re overstating that you think my idea is laughable and such.
But note the switch you’ve performed: you’ve now substituted “I suspect Said of doing that [‘stick[ing] one’s heels in and be unwilling to budge on a position regardless of reason or argument’] from time to time” in place of your initial “Said Achmiz regularly cannot understand basic points and ideas”.
It should be obvious that these are two extremely different thing.
Being unable to “understand basic points and ideas” is clearly bad. But, as Zack points out, it’s also just obviously untrue of me.
Being “unwilling to budge on a position regardless of reason or argument” is… well, as I note in the link above, it’s often entirely sensible even if the reasons are good and the argument convincing. But why assume that? Surely it’s not always the case? If your reasons are dumb and your arguments bad, then it would obviously be wrong for me to “budge”, yes? (I am reminded, here, of the discussion about “frame control”.)
But then the complaint is just that I sometimes (often?) find people’s arguments to be bad. Well, yeah. Is this… surprising? Surely you don’t think that most people’s arguments are good…?
Heck, this even applies to “understand[ing] basic points and ideas” as well! We have this implication:
“If a point or idea is basic, coherent, not nonsense, etc., then a reasonable, non-dumb person should be able to understand it.”
Well, you know what they say: one man’s modus ponens is another man’s modus tollens.
So, in the end, all you’re really saying is “in various cases, I disagreed with Said about whether some argument was good, whether some idea was coherent and straightforward, whether some reasons for some belief were good ones, etc.” Yes, it’s true! I sometimes disagree with people about such things. Guilty as charged! I do indeed think that sometimes (maybe even… often?) people claim that some idea is “basic”, but actually it’s incoherent nonsense. And sometimes (maybe even… often?), people think that some argument is good, but actually it’s bad.
But, as Zack points out, it’s also just obviously untrue of me.
By the by, I just want to note another instance of yousayingthingsareobvious. I perceive it to often be an attempt to equivocate between you just describing the world as you see it, and also attempting to quietly imply that your interlocutor’s perspective should be entirely dismissed (on at least this issue), and even worse in conflict scenarios to imply that your interlocutor is a fool. Insofar as this is true, it seems especially egregious to me as sometimes it seems to me that you are wrong (e.g. saying it’s ‘obvious’ that there’s no distinction in how hard it is to write comments or posts).
It’s not easy to have a conversation while someone takes every opportunity to not only say that they think you’re mistaken, but to say your perspective should be dismissed and imply you’re a fool for having it. Again, as Zack would be quick to say, perhaps you believe they’re a fool! But if that’s what’s intended, it’s not what’s happening here, as it’s being equivocated with or carefully masked.
(I am sure Said will point out he has never out-and-out called someone a fool, and so it is ‘obviously wrong’ of me to think that this reading is desired.)
It’s not easy to have a conversation while someone takes every opportunity to not only say that they think you’re mistaken, but to say your perspective should be dismissed and imply you’re a fool for having it.
False. It’s very easy. In fact, not only are you mistaken about this, but your perspective should be entirely dismissed. (But you’re not a fool—at least, no more than I am, on the whole, or than anyone else is. Indeed, that’s what makes this all so frustrating.)
Now, will you punish me for saying this? Will you try to use this comment as evidence to accuse me of… whatever it is that you’re claiming is bad about my comments? Or will you say “ah, you see, now that’s better, since it’s honest and straightforward instead of being masked; this is good and proper commenting, no complaints”?
I predict that you’re going to do the former, and not the latter. I make this prediction because I don’t believe that you’re sincere in your argument that it’s the non-explicitness that’s the problem. I think that you’re making that argument in bad faith.
I stand ready to apologize if I turn out to have been wrong about this.
I don’t believe my perspective on communication should be entirely dismissed. I think I have managed to navigate a lot of social situations unusually well – herding spiky and oddball rationalists in-person and online for nearly a decade into one of the few highly active and alive intellectual web-forums, helping build Lightcone Infrastructure to run many very successful and difficult events (e.g. LessOnline), writing good LessWrong posts about social dynamics, etc, and I think this is intertwined with my perspective on social interactions, what is being communicated, and how. Many smart and wise people have things to learn from my perspective, as do I theirs.
I agree that your perspective on communication should not be dismissed, and that your successful navigation of social situations at Lightcone and elsewhere are evidence for this. Achmiz is definitely obviously wrong about this. Shame on him!
But that’s not the interesting part of the grandparent. The interesting part is where Achmiz accuses you of bad faith (pretending to entertain one set of feelings while acting as if influenced by another) for criticizing his comments (above and elsewhere) for being allegedly passive-aggressive. I find it surprising that you would reply to the grandparent but not address the bad faith accusation (as contrasted to just not replying because you’re busy).
Maybe you think you shouldn’t be subjected to such accusations and don’t want to dignify them with a response. But I’m really interested in what your response would be, particularly because, separately from the bad faith accusation, the object-level complaint seems cogent: if passive-aggressive comments are bad, then saying the same thing in an overtly aggressive manner must be better, right? (If not, then the passive-aggressiveness wasn’t the problem, and presumably previous criticisms saying that were in error.)
For myself, when I get accused of bad faith, I usually do think it’s worth responding, because given human nature, I don’t think it’s crazy for someone to suspect that I might have some hidden motive for my speech that hadn’t already been made clear, and I’m eager to allay such concerns by trying to dump more context for why I said what I said. I don’t think I’m entitled to an assumption of good faith from my interlocutor; I think I can earn it. I think you can earn it, too.
I am busy, so not responding to all parts of these comments, especially not those with longer inferential distance, and prolly won’t keep replying here for at least a few days. But to be clear I said that Said was attempting to call people fools[1] with plausible deniability, and that he should do so without the subterfuge, then he denied that he thought people were fools, so there’s not a natural next step here where I praise him for saying it directly. I’m not sure I believe him but this does not seem like the margin to prosecute that criticism.
As an aside, I have a long draft blogpost from a year ago explaining why plausible-deniable aggression and plausible-deniable rule-breaking is really bad, which I hope to finish and publish one day. This isn’t some new position I have found in the course of this political conflict, it is something I have believed for a long time (and have been slightly surprised that you find it so improbable that I would hold this position, I would’ve thought it relatively clear why it is at least compelling once the hypothesis was raised).
“Calling them fools” here is a stand-in for “is intentionally attempting to lower their status, and not incidentally doing so as a side effect of criticizing their writing”. The specific word “fool” is not the load-bearing part.
I have a long draft blogpost from a year ago explaining why plausible-deniable aggression [...] I would’ve thought it relatively clear why it is at least compelling once the hypothesis was raised
I look forward to the post. (I’d love to read the draft, if you’re comfortable sharing.)
The reason I have the opposite intuition so strongly is because I think regulating social status emotions is a distraction from the intellectually substantive work we’re here to do. Norms that encourage people to overtly do more of that seem super toxic, and norms that encourage inquisitorial scrutiny of subjective minutiae of subtext to make certain it’s not happening covertly seem super toxic in a different way. I’d rather just let people’s emotional regulation happen in the background without having to talk about it. If that means some people get away with slipping in social “attacks” through the subtext of their writing without getting punished except by subtextual counterattacks from their interlocutors, that seems like a pretty normal part of talking naturally as a human, and I’d rather just let it happen than try to heavy-handedly control it.
“Calling them fools” here is a stand-in for “is intentionally attempting to lower their status, and not incidentally doing so as a side effect of criticizing their writing”. The specific word “fool” is not the load-bearing part. [...] then he denied that he thought people were fools, so there’s not a natural next step here where I praise him for saying it directly
I agree that your perspective on communication should not be dismissed, and that your successful navigation of social situations at Lightcone and elsewhere are evidence for this. Achmiz is definitely obviously wrong about this. Shame on him!
It’s not Ben’s perspective on “communication”, in general, that should be dismissed, but his perspective on the specific thing I quoted. I stand by that view.
It’s not easy to have a conversation while someone takes every opportunity to [...]
Isn’t it, though? You can just keep talking! Why would them implying you’re a fool end the conversation?
Okay, maybe if your only goal in the conversation is to persuade the other person, them appearing “closed-minded” (by implying you’re a fool) implies that you won’t succeed in persuading them, so you shouldn’t bother? But it can still be worth talking if you have other goals, like persuading third parties or (this is an important one!) being persuaded yourself if you’re wrong.
It’s a common failure mode on the Internet (and, by extension, on LW) that people believe a public response to/a criticism of an author’s writing or ideas is meant to convince the author that they’re wrong. It’s not. as a general matter, people believe what they believe and it takes a tremendous amount of time and effort to get them to change their minds, regardless of how smart or LW-”rational” they are.[1] It’s very rarely worth it to try to convince any single person of anything, unless they are super high-status or have a ton of decision-making power. And in any case, if I want to personally convince someone of something, I just PM them; humans more readily admit to mistakes when they’re not blasted in public from the outset.
On the contrary, explaining in public why someone is wrong is emphatically for the benefit of the audience. It is inherently performative. The author is just one guy,[2] but the audience is a usually a lot more guys. “This author’s argument here is wrong and you shouldn’t believe them” and “this author’s argumentative flaws illustrate why they’re fundamentally confused/epistemically broken/dumb and you shouldn’t listen to them again in the future” are both significantly higher-impact and therefore more worthwhile than “you made a mistake here, please change your mind.”
Of course, one man’s bug is another man’s feature. I called it a “failure mode” above, but that’s only relative to a specific set of end goals.[3] Another very common set of goals contains desires like “increase subjective hedonic enjoyment of online social interactions, as an end goal in and of itself.” If one subscribes to this, it’s not hard to figure out why others calling you a fool is something to be avoided and proactively guarded against.
Such as promoting epistemic hygiene, increasing map-territory correspondence, rewarding proper reasoning and disincentivizing sloppy thinking… you know, all the good stuff LW pretends it’s about
It’s a common failure mode on the Internet (and, by extension, on LW) that people believe a public response to/a criticism of an author’s writing or ideas is meant to convince the author that they’re wrong. It’s not. as a general matter, people believe what they believe and it takes a tremendous amount of time and effort to get them to change their minds, regardless of how smart or LW-”rational” they are.[1] It’s very rarely worth it to try to convince any single person of anything, unless they are super high-status or have a ton of decision-making power. And in any case, if I want to personally convince someone of something, I just PM them; humans more readily admit to mistakes when they’re not blasted in public from the outset.
That’s a fair point, but if someone is fully on the side of “I’m not actually trying to have a conversation with the author, I’m writing for the rest of the readership” then it is bad faith to write as though you are having a conversation with them e.g. addressing them with ‘you’ and writing to them questions and so forth, which I also think is a common bad discourse pattern.
… if someone is fully on the side of “I’m not actually trying to have a conversation with the author, I’m writing for the rest of the readership” then it is bad faith to write as though you are having a conversation with them …
This is a strawman of @sunwillrise’s point. He did not say anything about “not trying to have a conversation with” someone. What he said was:
… people believe a public response to/a criticism of an author’s writing or ideas is meant to convince the author that they’re wrong. It’s not.
It is entirely possible to have a conversation with someone without any intent or expectation of convincing that person that they’re wrong. (Indeed, it’s by far the more common scenario.)
Your response, on the other hand, implies that the only reason to have a conversation with someone who you think is wrong, is to try to convince them that they’re wrong. I hope you can see how silly that is. (I don’t think that you actually think this—but what you wrote implies it.)
I am not strawmanning sunwillrise’s position, I am making an additional related point.
Your response, on the other hand, implies that the only reason to have a conversation with someone who you think is wrong, is to try to convince them that they’re wrong.
I do not believe it is the only reason, but it is a common reason, and it is costly for people to repeatedly come to believe (based on the common social cues) that it is what is happening and engage on those terms, only to find out later that it is not and that they were engaged in a different social game they would rather not be playing where the goal is to make them look bad and for them to defend themselves.
I am not strawmanning sunwillrise’s position, I am making an additional related point.
You wrote:
if someone is fully on the side of “I’m not actually trying to have a conversation with the author, I’m writing for the rest of the readership”
Whom does this describe? Who has expressed any such sentiment?
The implicature of your comment was that the quoted bit was a restatement of the point to which you were responding. You were absolutely strawmanning sunwillrise’s position.
I do not believe it is the only reason, but it is a common reason, and it is costly for people to repeatedly come to believe (based on the common social cues) that it is what is happening and engage on those terms
If someone mistakenly believes that their interlocutor in a public conversation on a public discussion forum is just trying to convince them, personally, to change their minds, then this is almost certainly an error on their part. As a moderator of said forum, it would behoove you to spread awareness of the fact that such is not the default or the usual motivation for people to have public conversations on said public forum.
only to find out later that it is not and the person is substantially motivated by wanting to make them look bad.
This, too, is a strawman: “substantially motivated by wanting to make them look bad” is a tendentious description, and is certainly not one which most people would endorse, as applied to their own contributions to such conversations.
Insofar as conversations are 100% about communicating information on the object level about a topic, of course social information about status and personal hostilities are irrelevant to the point of a conversation. Insofar as one is not blinding oneself to the other dynamics but also living in them, then calling someone a fool a lot is quite relevant to whether to continue interacting with someone, as I’m pretty sure you’re aware.
Our comment threads are going in circles, and these perspectives are not being bridged. I expect more comments like this are not going to change much.
It would be a great mistake to suppose that it is sufficient not to become personal yourself. For by showing a man quite quietly that he is wrong, and that what he says and thinks is incorrect—a process which occurs in every dialectical victory—you embitter him more than if you used some rude or insulting expression. Why is this? Because, as Hobbes observes,17 all mental pleasure consists in being able to compare oneself with others to one’s own advantage. Nothing is of greater moment to a man than the gratification of his vanity, and no wound is more painful than that which is inflicted on it. Hence such phrases as “Death before dishonour,” and so on. The gratification of vanity arises mainly by comparison of oneself with others, in every respect, but chiefly in respect of one’s intellectual powers; and so the most effective and the strongest gratification of it is to be found in controversy. Hence the embitterment of defeat, apart from any question of injustice; and hence recourse to that last weapon, that last trick, which you cannot evade by mere politeness. A cool demeanour may, however, help you here, if, as soon as your opponent becomes personal, you quietly reply, “That has no bearing on the point in dispute,” and immediately bring the conversation back to it, and continue to show him that he is wrong, without taking any notice of his insults. Say, as Themistocles said to Eurybiades—Strike, but hear me. But such demeanour is not given to every one.
I have much work to do other than replying to these long threads with you and your allies under the clouds of this political conflict around whether to ban you from LessWrong. Do not expect many more replies any time soon.
Benquo was referring to your own views in his comment, so this foisting-off of responsibility seems a bit odd. Insofar as he agrees with you, I disagree with him as well. I don’t see what’s weird about that.
I was replying to views expressed by you. If you think that I have mis-apprehended what views you in fact hold, by all means tell me; but that’s not what you seem to be saying?
This is… such a vague statement that I can’t disagree with it, as such… but it’s also not really responsive to what I wrote. (But if you did mean this as a contradiction of what I wrote, then—I disagree and think that you are mistaken.)
I didn’t say anything about explicit consent. It’s true that “friends do not ask each other before leaning their head on other people’s shoulder”, but that doesn’t make this non-consensual! Of course it’s consensual. (Usually. But then, there are many cases when someone initiates this sort of “implied-consensual” touching, but then it turns out that consent wasn’t actually implied. Whoops! In such cases, the distinction between implied consent, and lack of consent, suddenly becomes quite salient.)
And it’s true that “family members routine hug each other without asking first”. It’s also true that many, many people experience this as a violation of their autonomy and their boundaries, and greatly dislike it. (Stuff like this is about as novel and surprising as “what’s the deal with airline food” at this point.)
Perhaps if you understand “physical vulnerability” in a broad way. But otherwise, no, I do not agree. My point is that the possibility of violence usually has little to do with it.
I wouldn’t say “none”, perhaps. But almost none.
Er, you’re the one making the analogy, so surely it’s up to you to choose a better example, not your interlocutor…
I agree he could have chosen a better example. But like, are you trying to understand Habryka? Or are you just trying to litigate proper use of analogies? Your comment reads to me as 20% responding to his point, and 80% litigation. This sort of thing feels like an advanced version of arguing about definitions instead of just tabooing the word and talking about the underlying point each person is trying to make.
What you’re doing here is just this.
I think that habryka’s point, as described in the top-level comment of this thread and as explicated by the analogy he used, is wrong. If he wants to amend his claims to make them less wrong (possibly by choosing a different analogy and then reasoning on that basis, or possibly in some other way), that’s fine, he can do that, and I’ll be glad to read any such corrective commentary. But I can’t do that for him.
The idea that disagreeing with someone just means that the disagreer doesn’t understand and needs to work harder to try to understand needs to die a flaming death. Yes, sometimes misunderstanding happens, but sometimes people just think that you’re wrong. They understand what you’re saying and they disagree. Your task then is to convince, not to explain—and certainly not to whine about your interlocutors not working hard enough to convince themselves of your claims on your behalf.
Hmm, seems like I didn’t communicate well enough. Trying again.
I believe you understand and disagree with his point. I also believe you think his analogy is bad. When I first read your reply I thought you were disagreeing because of/due to the disanalogies, and for no other reason. I no longer think this.
Your disagreement of his point, and your critique of his analogy felt very mixed together. Which like, he’s using the analogy to explicate his point for a reason, so fair. But there’s something like a difference between using an analogy as supporting argument, and using an analogy just to point at the thing.
As far as I can tell your comment doesn’t address this point directly? I’m asking for something like a clearer distinction between disagreeing with his point, and critiquing the analogy. Especially in this case where I don’t think the particulars of the analogy where central to his point.
I mean, the analogy is “bad” insofar as the point that it supports is wrong. Like, the two things which are claimed to be analogous in a certain important way, are in fact not analogous in that important way. (Or so I claim!)
(It’s like if I said “the sky is like a glass dome; if you fly high enough, you’ll crash into it; and since the sky is indestructible, much like a glass dome is, you also can’t break through it”. Well, no, you in fact will not crash into the sky; in this way, it is precisely not like a glass dome. And of course glass is totally destructible. The analogy successfully communicates my beliefs about the sky—that it’s a solid barrier which can be crashed into but not broken through. Those beliefs happen to be totally wrong. The glass dome analogy is “bad” in that sense.)
True, I did not address that. I’ll do so now.
So, let’s recall what “it sure really matters a lot” means, specifically, in this context. The key claim from the earlier comment is this:
In the malware/exploit case, the analogous claim would be something like:
“Computer traffic doesn’t need to be ‘predominantly malware or exploits’ for it to be important to differentially signal that you are trying to send innocent, non-malicious data.”
Well… you can probably see the problem here. There’s basically two scenarios:
There exists a totally unambiguous, formally (which usually means: cryptographically) verifiable signal of a data packet or message being non-malicious. That signal gets sent, we check it, if it doesn’t check out we reject the data, the end. (If the signal can be faked after all, then we’re just fucked.)
There is no such verifiable signal. In this case, malicious traffic is going to be sending all the signals of non-maliciousness that “good” traffic sends. “A differential signal of innocence is being intentionally sent” is almost completely worthless as a basis for concluding that the data is non-malicious. Instead, we have to use complicated Bayesian methods to sort good from bad (as in email), or we have to enter into an arms race of requiring, and checking for, increasingly convoluted and esoteric micro-signals of validity (as in CAPTCHAs, UA sniffing, and all the other myriad tricks that websites use these days to protect themselves from abuse). (And any client that deliberately sends the signals we’re checking for is actually more likely to be a bad actor!)
This situation… is also not analogous to “posting on a public discussion forum”, which looks nothing like either of the above cases.
Aside: I do want to try to help you here, and warn you that it is my opinion Said Achmiz regularly cannot understand basic points and ideas and criticizes his interlocutors with extremely harsh language – regular unnecessary incredulity with extra question marks and exclamation points and italics, describing your positions as laughable, obviously wrong, deeply corrosive, etc – and I would encourage you to not continue trying to explain something to his satisfaction unless you would also find it a similarly worthwhile activity if your interlocutor was an LLM in whose system prompt it was written that it should not be able to either agree with or understand your point.
(You’re of course totally welcome to keep trying, I just felt I’d be irresponsible not to warn you.)
Excuse me, what?
Italics is “harsh language” now??
Separately, your phrasing:
… implies that I was referring to some person or people with the phrase “needs to die a flaming death”. I hope you can see how totally unacceptable that implication is. (I won’t belabor the point that it’s false; you know that. But please correct the phrasing; falsely implying that I expressed a desire for a person to die violently is absolutely not ok, even if that implication was accidental, as I assume that it was.)
Oh yeah that’s fair, edited to other examples.
Thank you. (I’ve updated my vote on your comment accordingly.)
Your position is laughable, obviously wrong, and deeply corrosive. I routinely succeed at explaining things to Achmiz’s satisfaction.
To illustrate, I’ll quickly pick two arbitrary examples from the top of my head. One, on this website in February 2023, I explained the existence of situations in which trying to minimize the damage from errors is preferable to trying to not commit errors. Two, in private correspondence earlier this month about an unpublished draft of mine, Achmiz wrote that an analogy I had used was “Confusing—because I don’t really get what sort of situation it can be analogous to [...] [C]onsider this objection to constitute a request for such examples!” I wrote up an explanation of a real-world situation of the class I was thinking of when I wrote the analogy, to which Achmiz responded, “Hmm … ok, I see what you mean [...] Anyhow, you’ve answered my question, yeah.”
To empirically test your claim about Achmiz vis-à-vis large language models, I supplied the explanation about minimizing damage from errors to Claude Opus 4 with the custom style prompt, “Emulate an obstinate interlocutor that should not be able to either agree with or understand the user’s point.” The difference between Achmiz’s actual response from February 2023 (“I see, thanks” and a relevant followup question which I also had no trouble answering) and the LLM’s response (“I don’t see how this example makes any sense at all” followed by another 200 words of blatant misreadings and non sequiturs) is quite apparent. Your claim that Morrison would find interacting with Achmiz and the LLM-prompted-for-obstinancy “similarly worthwhile” is clearly absurd. (And as it happens, Achmiz’s reply to Morrison seemed entirely cogent to me.)
I’m presenting this counterevidence to make a point, but really, I don’t think you believe what you wrote. The existence of examples of Achmiz being satisfied with explanations and the outcome of the LLM prompt were both easily predictable; you just didn’t think anyone would call your bluff on a hyperbolic insult. You were wrong, but more importantly: do you really think this is a good look for you or the website?
Good to know it’s ever happened! It is extremely uncommon in my experience of reading threads with Said in them.
Added: To be fair I wrote my comment upthread feeling my peak frustration about (what I read as) Said’s unproductive commenting style. I wrote to someone else at the time it would probably be my worst contribution to this ongoing conversation of hundreds of comments and hundreds of hours of conflict, so I’ll admit it is probably my peak sloppily-stated-times-aggressive comment and not to my usual standards.
Still, I think it’s more reasonable than you’re giving it credit for. I think I think it’s a fairly standard human phenomena to stick one’s heels in and be unwilling to budge on a position regardless of reason or argument, and I suspect Said of doing that from time to time (or alternatively of being quite dense), and you’ve told me before you also have felt that he seems sometimes to be kind of dense about very simple things, and I think people often do things they shouldn’t when they feel threatened, so I think you’re overstating that you think my idea is laughable and such.
And also normatively correct.
But note the switch you’ve performed: you’ve now substituted “I suspect Said of doing that [‘stick[ing] one’s heels in and be unwilling to budge on a position regardless of reason or argument’] from time to time” in place of your initial “Said Achmiz regularly cannot understand basic points and ideas”.
It should be obvious that these are two extremely different thing.
Being unable to “understand basic points and ideas” is clearly bad. But, as Zack points out, it’s also just obviously untrue of me.
Being “unwilling to budge on a position regardless of reason or argument” is… well, as I note in the link above, it’s often entirely sensible even if the reasons are good and the argument convincing. But why assume that? Surely it’s not always the case? If your reasons are dumb and your arguments bad, then it would obviously be wrong for me to “budge”, yes? (I am reminded, here, of the discussion about “frame control”.)
But then the complaint is just that I sometimes (often?) find people’s arguments to be bad. Well, yeah. Is this… surprising? Surely you don’t think that most people’s arguments are good…?
Heck, this even applies to “understand[ing] basic points and ideas” as well! We have this implication:
“If a point or idea is basic, coherent, not nonsense, etc., then a reasonable, non-dumb person should be able to understand it.”
Well, you know what they say: one man’s modus ponens is another man’s modus tollens.
So, in the end, all you’re really saying is “in various cases, I disagreed with Said about whether some argument was good, whether some idea was coherent and straightforward, whether some reasons for some belief were good ones, etc.” Yes, it’s true! I sometimes disagree with people about such things. Guilty as charged! I do indeed think that sometimes (maybe even… often?) people claim that some idea is “basic”, but actually it’s incoherent nonsense. And sometimes (maybe even… often?), people think that some argument is good, but actually it’s bad.
But I’m always open to being convinced otherwise.
By the by, I just want to note another instance of you saying things are obvious. I perceive it to often be an attempt to equivocate between you just describing the world as you see it, and also attempting to quietly imply that your interlocutor’s perspective should be entirely dismissed (on at least this issue), and even worse in conflict scenarios to imply that your interlocutor is a fool. Insofar as this is true, it seems especially egregious to me as sometimes it seems to me that you are wrong (e.g. saying it’s ‘obvious’ that there’s no distinction in how hard it is to write comments or posts).
It’s not easy to have a conversation while someone takes every opportunity to not only say that they think you’re mistaken, but to say your perspective should be dismissed and imply you’re a fool for having it. Again, as Zack would be quick to say, perhaps you believe they’re a fool! But if that’s what’s intended, it’s not what’s happening here, as it’s being equivocated with or carefully masked.
(I am sure Said will point out he has never out-and-out called someone a fool, and so it is ‘obviously wrong’ of me to think that this reading is desired.)
False. It’s very easy. In fact, not only are you mistaken about this, but your perspective should be entirely dismissed. (But you’re not a fool—at least, no more than I am, on the whole, or than anyone else is. Indeed, that’s what makes this all so frustrating.)
Now, will you punish me for saying this? Will you try to use this comment as evidence to accuse me of… whatever it is that you’re claiming is bad about my comments? Or will you say “ah, you see, now that’s better, since it’s honest and straightforward instead of being masked; this is good and proper commenting, no complaints”?
I predict that you’re going to do the former, and not the latter. I make this prediction because I don’t believe that you’re sincere in your argument that it’s the non-explicitness that’s the problem. I think that you’re making that argument in bad faith.
I stand ready to apologize if I turn out to have been wrong about this.
I don’t believe my perspective on communication should be entirely dismissed. I think I have managed to navigate a lot of social situations unusually well – herding spiky and oddball rationalists in-person and online for nearly a decade into one of the few highly active and alive intellectual web-forums, helping build Lightcone Infrastructure to run many very successful and difficult events (e.g. LessOnline), writing good LessWrong posts about social dynamics, etc, and I think this is intertwined with my perspective on social interactions, what is being communicated, and how. Many smart and wise people have things to learn from my perspective, as do I theirs.
I agree that your perspective on communication should not be dismissed, and that your successful navigation of social situations at Lightcone and elsewhere are evidence for this. Achmiz is definitely obviously wrong about this. Shame on him!
But that’s not the interesting part of the grandparent. The interesting part is where Achmiz accuses you of bad faith (pretending to entertain one set of feelings while acting as if influenced by another) for criticizing his comments (above and elsewhere) for being allegedly passive-aggressive. I find it surprising that you would reply to the grandparent but not address the bad faith accusation (as contrasted to just not replying because you’re busy).
Maybe you think you shouldn’t be subjected to such accusations and don’t want to dignify them with a response. But I’m really interested in what your response would be, particularly because, separately from the bad faith accusation, the object-level complaint seems cogent: if passive-aggressive comments are bad, then saying the same thing in an overtly aggressive manner must be better, right? (If not, then the passive-aggressiveness wasn’t the problem, and presumably previous criticisms saying that were in error.)
For myself, when I get accused of bad faith, I usually do think it’s worth responding, because given human nature, I don’t think it’s crazy for someone to suspect that I might have some hidden motive for my speech that hadn’t already been made clear, and I’m eager to allay such concerns by trying to dump more context for why I said what I said. I don’t think I’m entitled to an assumption of good faith from my interlocutor; I think I can earn it. I think you can earn it, too.
I am busy, so not responding to all parts of these comments, especially not those with longer inferential distance, and prolly won’t keep replying here for at least a few days. But to be clear I said that Said was attempting to call people fools[1] with plausible deniability, and that he should do so without the subterfuge, then he denied that he thought people were fools, so there’s not a natural next step here where I praise him for saying it directly. I’m not sure I believe him but this does not seem like the margin to prosecute that criticism.
As an aside, I have a long draft blogpost from a year ago explaining why plausible-deniable aggression and plausible-deniable rule-breaking is really bad, which I hope to finish and publish one day. This isn’t some new position I have found in the course of this political conflict, it is something I have believed for a long time (and have been slightly surprised that you find it so improbable that I would hold this position, I would’ve thought it relatively clear why it is at least compelling once the hypothesis was raised).
“Calling them fools” here is a stand-in for “is intentionally attempting to lower their status, and not incidentally doing so as a side effect of criticizing their writing”. The specific word “fool” is not the load-bearing part.
Hope whatever tasks you’re busy with go well!
I look forward to the post. (I’d love to read the draft, if you’re comfortable sharing.)
The reason I have the opposite intuition so strongly is because I think regulating social status emotions is a distraction from the intellectually substantive work we’re here to do. Norms that encourage people to overtly do more of that seem super toxic, and norms that encourage inquisitorial scrutiny of subjective minutiae of subtext to make certain it’s not happening covertly seem super toxic in a different way. I’d rather just let people’s emotional regulation happen in the background without having to talk about it. If that means some people get away with slipping in social “attacks” through the subtext of their writing without getting punished except by subtextual counterattacks from their interlocutors, that seems like a pretty normal part of talking naturally as a human, and I’d rather just let it happen than try to heavy-handedly control it.
The world “fool” may not be load-bearing in your comment, but I’m pretty sure it is load-bearing in Achmiz’s denial (“But you’re not a fool [...] that’s what makes this all so frustrating”). He’s trying to lower your status by calling you dishonest, not stupid.
It’s not Ben’s perspective on “communication”, in general, that should be dismissed, but his perspective on the specific thing I quoted. I stand by that view.
Isn’t it, though? You can just keep talking! Why would them implying you’re a fool end the conversation?
Okay, maybe if your only goal in the conversation is to persuade the other person, them appearing “closed-minded” (by implying you’re a fool) implies that you won’t succeed in persuading them, so you shouldn’t bother? But it can still be worth talking if you have other goals, like persuading third parties or (this is an important one!) being persuaded yourself if you’re wrong.
It’s a common failure mode on the Internet (and, by extension, on LW) that people believe a public response to/a criticism of an author’s writing or ideas is meant to convince the author that they’re wrong. It’s not. as a general matter, people believe what they believe and it takes a tremendous amount of time and effort to get them to change their minds, regardless of how smart or LW-”rational” they are.[1] It’s very rarely worth it to try to convince any single person of anything, unless they are super high-status or have a ton of decision-making power. And in any case, if I want to personally convince someone of something, I just PM them; humans more readily admit to mistakes when they’re not blasted in public from the outset.
On the contrary, explaining in public why someone is wrong is emphatically for the benefit of the audience. It is inherently performative. The author is just one guy,[2] but the audience is a usually a lot more guys. “This author’s argument here is wrong and you shouldn’t believe them” and “this author’s argumentative flaws illustrate why they’re fundamentally confused/epistemically broken/dumb and you shouldn’t listen to them again in the future” are both significantly higher-impact and therefore more worthwhile than “you made a mistake here, please change your mind.”
Of course, one man’s bug is another man’s feature. I called it a “failure mode” above, but that’s only relative to a specific set of end goals.[3] Another very common set of goals contains desires like “increase subjective hedonic enjoyment of online social interactions, as an end goal in and of itself.” If one subscribes to this, it’s not hard to figure out why others calling you a fool is something to be avoided and proactively guarded against.
And even when they do change their minds, it’s usually in private over days or weeks of mulling over the problem in the comfort of their own abode
Not meant to imply any specific gender
Such as promoting epistemic hygiene, increasing map-territory correspondence, rewarding proper reasoning and disincentivizing sloppy thinking… you know, all the good stuff LW pretends it’s about
That’s a fair point, but if someone is fully on the side of “I’m not actually trying to have a conversation with the author, I’m writing for the rest of the readership” then it is bad faith to write as though you are having a conversation with them e.g. addressing them with ‘you’ and writing to them questions and so forth, which I also think is a common bad discourse pattern.
This is a strawman of @sunwillrise’s point. He did not say anything about “not trying to have a conversation with” someone. What he said was:
It is entirely possible to have a conversation with someone without any intent or expectation of convincing that person that they’re wrong. (Indeed, it’s by far the more common scenario.)
Your response, on the other hand, implies that the only reason to have a conversation with someone who you think is wrong, is to try to convince them that they’re wrong. I hope you can see how silly that is. (I don’t think that you actually think this—but what you wrote implies it.)
I am not strawmanning sunwillrise’s position, I am making an additional related point.
I do not believe it is the only reason, but it is a common reason, and it is costly for people to repeatedly come to believe (based on the common social cues) that it is what is happening and engage on those terms, only to find out later that it is not and that they were engaged in a different social game they would rather not be playing where the goal is to make them look bad and for them to defend themselves.
You wrote:
Whom does this describe? Who has expressed any such sentiment?
The implicature of your comment was that the quoted bit was a restatement of the point to which you were responding. You were absolutely strawmanning sunwillrise’s position.
If someone mistakenly believes that their interlocutor in a public conversation on a public discussion forum is just trying to convince them, personally, to change their minds, then this is almost certainly an error on their part. As a moderator of said forum, it would behoove you to spread awareness of the fact that such is not the default or the usual motivation for people to have public conversations on said public forum.
This, too, is a strawman: “substantially motivated by wanting to make them look bad” is a tendentious description, and is certainly not one which most people would endorse, as applied to their own contributions to such conversations.
Insofar as conversations are 100% about communicating information on the object level about a topic, of course social information about status and personal hostilities are irrelevant to the point of a conversation. Insofar as one is not blinding oneself to the other dynamics but also living in them, then calling someone a fool a lot is quite relevant to whether to continue interacting with someone, as I’m pretty sure you’re aware.
Our comment threads are going in circles, and these perspectives are not being bridged. I expect more comments like this are not going to change much.
I have much work to do other than replying to these long threads with you and your allies under the clouds of this political conflict around whether to ban you from LessWrong. Do not expect many more replies any time soon.