Author's note: This is a contemporary post about an ongoing current event. It's not a timeless essay in the way that most of my essays are, though it does contain insight and thoughts on timeless topics.
David Sabatini is a molecular biologist, previously employed as a tenured professor at MIT and a lead researcher at the Whitehead Institute. He was fired in August of 2021, after which most of his professional connections quickly dried up. He was almost restored to good standing in April of 2022 via a new position at NYU, but protests and external pressure caused NYU to withdraw their offer.
I shared a one-sided piece about Sabatini on Facebook, asking for people to falsify it. I was given various links and documents in response. I spent about five hours following up on sources, looking for original information, and trying to piece together a coherent take.
It proved to be impossible, and it proved to be impossible in a way that I think is interesting, and relevant to a lot of questions about how our culture functions (or doesn't). This essay is my attempt to digest and debrief, essentially writing to myself.
Why was Sabatini fired and blacklisted?
Option A: Because he engaged in romantic/sexual misconduct in conflict with the policies of his workplace, and created a hostile and sexualized environment that made work difficult or impossible for many of his subordinates.
Option B: Because a vindictive former lover enacted a revenge plot, partially enabled by an ideologue in the org's power structure who was looking for any pretext to shake things up.
Option C (for 'cynical'): Because scandal is costly regardless of whether it's grounded in fact, and there are a large number of highly-motivated people who have concentration of force against groups like MIT or NYU when it comes to highly charged questions like putative sexual misconduct.
If you buy reports like that of Suzy Weiss, the timeline looks something like this:
If you buy reports like that of Leonid Schneider, the timeline is more like:
There is a mental technique taught by the Center for Applied Rationality called "Murphyjitsu," in which people check their plans against their established intuitions about how the world works in a general sense.
Essentially, the core insight of Murphyjitsu is that merely thinking about your plan, and asking yourself "can I see this working? Is success consistent with all of my past experiences?" is not sufficient. You must also ask yourself "can I see this not working? Is failure consistent with all of my past experiences?"
If one finds that both success and failure are conceivable/coherent, then one does not, according to the standards of Murphyjitsu, have a real plan, and one must do more work.
Analogously: when I read the first account, and I ask myself "does this account feel consistent and coherent? Does this seem like the sort of thing that happens?" I get back the answer "yes, clearly." I've known people to be vindictive and dishonest; I've known people to be uncharitable in their interpretations and to willfully misrepresent events; I've known broader groups to act from cowardice and go along with the mob rather than take a costly stand in defense of the innocent.
And also, when I read the second account, and I ask myself "does this feel consistent and coherent? Does this seem like the sort of thing that happens?" I again get back the answer "yes, clearly."
I have also known people to be systematically abused. To be forced by circumstance to swallow their needs and endure tremendous mistreatment. To not-speak-up for years at a time, because they knew in their bones that doing so would result in their ruin.
Which means, à la Murphyjitsu, that I have to do more work. I find both accounts plausible—not necessarily equally plausible, but they both very much match my intuitions about How The World Works. Neither account is more than 90% likely, relative to the other; they are both in the same order of magnitude of believability.
My usual prescription in a situation like this is split and commit.
"Okay," I say to myself. "If it turns out that Knouse's interpretation of events is largely correct, and Sabatini is a pseudorapey asshole, then I know what sorts of things I would do. I know what kinds of lessons to take from this story, I know what kinds of comments to inject in places where people are discussing it, I know what updates to make about the competence and integrity of institutions and society at large."
"Also, if it turns out that Sabatini's interpretation of events is more accurate, and Knouse is a liar taking advantage of available MeToo momentum to execute a vendetta, then again I know what sorts of things I would do."
(I'm not being super concrete about these things here, but to gesture in the general direction: in the former case, I want to put marginal energy toward keeping up the societal momentum we've had over the past decade, since we still have not reached a point at which people like Knouse are safe from people like Sabatini. In the latter, I want to put marginal energy toward dragging the pendulum back, since in our zeal to protect the marginalized we have created a whole new class of victims. Obviously in either case I want to maintain flexibility, and not devolve to blindly black-or-white policies, but the Sabatini scandal could be an interesting hint as to whether Scylla or Charybdis is currently a more salient threat. It's a bit of useful information, to be combined with all of my other data.)
Alternately, it could (I supposed, at the outset) be that both of the accounts are substantively true, and the update is more just "people are shitty; be on guard." It's not necessarily the case that there is one villain in this story; a priori, there could quite easily be at least two.
What usually happens after you split and commit is that you find out which of the two possible worlds you're in.
(Or rather, which of the two possible worlds you conceived of seems closer to the real world, which is usually more complicated than what you managed to think of.)
Either that, or you move on without having actually found the answer, because you can't in practice chase down the answer to every possible question and in many cases the value is more in clarifying your own policies and values than in actually applying that clarification in the moment.
In fact, I'd like to amend the "usually happens" statement—what actually usually happens is indeed the latter thing, where you stay split and do not collapse the waveform (but you know how you will respond, if conclusive information happens to come across your path).
What is interesting to me is that, in the Sabatini case, I went actively searching for conclusive information, in part at the prompting of some of my friends, and was stymied.
Taking the image above as a rough representation of the situation (there's a messy reality, and a cleaner narrative that each of Team Sabatini and Team Knouse superimpose on that messy reality), the method I employed was pretty simple: look for verifiable spots in each narrative that make that narrative falsifiable, and go check.
If e.g. Sabatini claims that he and Knouse had a friendly and wholly consensual back-and-forth, and that text exchanges between them will unambiguously represent this, and meanwhile Knouse claims that she repeatedly rebuffed his advances and sent many messages expressing distress and hesitation, then voilà—simple, right? Both accounts substantially stick their necks out, so it shouldn't be hard to end up in a place where it's clear who's cleaving closer to the truth.
Unfortunately for my own, selfish desire-to-know-the-truth, this is not the case.
There are (without exaggeration) scores of articles and op-eds and thinkpieces about this whole scandal. And almost without exception, every single one of them seems to ground out in one of two documents:
These are the two most thorough public accounts, wherein each of the principles lay out their grievances against the other. Notably, they are each a set of claims and allegations, which (presumably) the courts of Massachusetts will evaluate for truth.
They are not, in other words, facts, in the sense that I am used to using the term.
This critical meta-fact is lost upon the vast majority of the people writing about the scandal, most of whom are not well-versed in mental moves like those found in Murphyjitsu, and who are not fond of asking what they think they know and why they think they know it. Blue tribe accounts take Knouse's complaint as gospel and repeat its claims as fact; red tribe ones take Sabatini's. A few, such as the Weiss piece linked above, seem to accurately represent other primary sources (such as text exchanges between Sabatini and Knouse), but it's not clear how far to trust them, just as it's not clear how far to trust the original investigations into Sabatini's lab.
Here is a small sampling of some of the close-to-crucial claims made by Team Sabatini—items which, if demonstrated true, are more likely in worlds where Sabatini is essentially innocent, than in worlds where he is essentially guilty:
None of these (of course) mean that Knouse's account cannot be true, but each, if borne out, is a non-negligible Bayesian update away from it, and toward the viewpoint expressed by Sabatini. There are other checkable facts as well; the above is just a sampling.
Symmetrically, evidence that would make the Knouse perspective much more likely:
Again, each of these is vastly more likely in the world where Sabatini is essentially guilty of all charges—none of them is by itself impossible in a world where he is innocent (and instead socially inept, for instance) but each one is a substantial Bayesian update in favor of the viewpoint expressed by Knouse.
These two worlds are insanely disjoint. It boggles the mind to consider that two such different accounts could have been circulating in the public eye for months, and neither obviously debunked. To quote Scott Aaronson:
Briefly: the lawyers' complaint and the Weiss article paint such diametrically, almost comically opposite pictures of Sabatini, Knouse, and their relationship, that at least one of the two MUST BE LYING. Yesterday, I'd concluded that the lawyers must've been, because their version of reality was impossible to reconcile with the actual texts of Knouse's emails. The trouble is, Weiss's version of reality is ALSO impossible to reconcile with the stuff Sabatini is reported to have said!In summary, Sabatini suffered a grave injustice in a version of reality that appeared to have been amply demonstrated by direct quotes. In a different version of reality, however, which also appears to be amply demonstrated by direct quotes, he was justly fired. In conclusion, then, 0=1, the universe is contradictory, and I give up, to return to easier questions like the computational complexity of quantum gravity.
Briefly: the lawyers' complaint and the Weiss article paint such diametrically, almost comically opposite pictures of Sabatini, Knouse, and their relationship, that at least one of the two MUST BE LYING. Yesterday, I'd concluded that the lawyers must've been, because their version of reality was impossible to reconcile with the actual texts of Knouse's emails. The trouble is, Weiss's version of reality is ALSO impossible to reconcile with the stuff Sabatini is reported to have said!
In summary, Sabatini suffered a grave injustice in a version of reality that appeared to have been amply demonstrated by direct quotes. In a different version of reality, however, which also appears to be amply demonstrated by direct quotes, he was justly fired. In conclusion, then, 0=1, the universe is contradictory, and I give up, to return to easier questions like the computational complexity of quantum gravity.
Presumably, someone has access to the texts, and to the testimony of the lab members, but that person is not me, and it is not you, and it is not any of the dozens of pundits desperately trying to demonstrate how this incident is a perfect microcosm of the brokenness of our present system and the plight of helpless victims in the face of heartless systems &cet.
(I'm one of those pundits, clearly, but I locate the problem one level up.)
It's also clear that, just as we-the-public did not get access to the direct text of the Whitehead Report, so that we could make our own assessment, so to will we-the-public not be given access to the pages and pages of direct source material that will be assessed by the Commonwealth of Massachusetts:
... followed by a section C, a section D, and a section E, then:
... followed by a section D and a section E, then:
... followed by a section D and then two more pages of original sources.
If I could get clear, unambiguous information on any (say) five of the above eighteen points, I think I could update to a stance like "70% likely Knouse, 30% likely Sabatini."
If I could get clear, unambiguous proxies on any of those points, that would at least be something. But every attempt to find proxies has failed. In my own Facebook threads, two friends of mine (whose reports I trust) had the following exchange:
I was curious after reading this and spoke directly to a Hughes scientist I know about this case; that scientist had heard of the case in some detail prior to our conversation.Their impression, which you can take or leave, was that Sabatini’s behavior was a very clear policy violation and not ambiguous, that potential harassment issues in science often involve situations that are much more “grey” and difficult to evaluate, and that it was quite understandable that institutions would not want to be involved with someone with such a reputation.That scientist thought there was a good chance Sabatini would be hired by a private institution or similar -- perhaps one that doesn't have students -- and continue working in the field, however.
I was curious after reading this and spoke directly to a Hughes scientist I know about this case; that scientist had heard of the case in some detail prior to our conversation.
Their impression, which you can take or leave, was that Sabatini’s behavior was a very clear policy violation and not ambiguous, that potential harassment issues in science often involve situations that are much more “grey” and difficult to evaluate, and that it was quite understandable that institutions would not want to be involved with someone with such a reputation.
That scientist thought there was a good chance Sabatini would be hired by a private institution or similar -- perhaps one that doesn't have students -- and continue working in the field, however.
What a previous (female) student in his lab told me was that she could not image him violating policy.
So ... someone kinda close to the situation (much closer than me, anyway) thinks that it goes obviously one way, and someone even closer thinks that it goes obviously the other.
(Insert the typical warning about how Person A not being abused by Person B doesn't update us all that much on whether Person B did or did not abuse Person C, but still.)
I would like to be able to find a source that seems to me to be exhibiting any unusual degree of epistemic hygiene, but alas, everything I look at seems to be biased in one direction or another. e.g. many sources find the following excerpts damning:
As one woman who worked in Sabatini's lab would put it in a text to a colleague, "the only way to get him to like you as a woman is to sexually appeal to him ... if you act in a way that suggests that you find him in some way attractive ..."
At least one male member of Sabatini's lab wrote that he believed that Sabatini orchestrated social events so that he could "drool" over attractive women
When the young woman started in the lab, she was greeted with advice as to how to get Sabatini's attention and approval—she was told to "play hard-to-get" and to "entertain him a little, then push him away."
... which—yes, unambiguously awful! Except that I have literally seen, with my own two eyes, people make claims exactly this level of bad about me with zero grounding in fact (and sometimes with evidence directly contradicting them, which did not deter them in the slightest).
It seems fairly reasonable to believe that the three people quoted above believe the things they are saying, i.e. they do indeed believe that Sabatini is a dog.
But I am ... let's say "not impressed" with the median human's ability to distinguish [what they think they know] from [what they actually know], or their conscientiousness in keeping their observations separate from the interpretations they layer atop those observations. The fact that these people think that you need to do X, Y, and Z to please Sabatini does not convince me that Sabatini wants or is in fact pleased by X, Y, and Z.
(I once had a colleague assert that he could literally think of no other explanation for my stances in various disagreements except that I was so depressed that I was delusional; said colleague helpfully purchased me unsolicited illicit antidepressants in the spirit of friendliness. Later, that same colleague went on to say that the only explanation they could think of for my fiancé's stance in their disagreement was that I had abused, gaslit, and mind-bent my fiancé. People who know Logan Strohl personally will get to enjoy the ridiculousness of this anecdote much more than people who do not, but suffice it to say that I find it quite plausible that the three individuals quoted above are Just Plain Wrong about Sabatini, or that they have some kind of axe to grind and are not particularly scrupulous about how they grind it. Unfortunately, I also find it plausible that they're exactly on point. 😕)
I struggled to take away actionable lessons from all of this. Most of what came to mind were prescriptions on the level of culture, and so the frame I'm going with is "things people in my culture take for granted."
1. You don't have to care about this.
This one reads as somewhat hypocritical, but in my defense, my deep dive was prompted by a specific request from some friends. For most people, the Sabatini scandal should not rise to the level of more than passing attention, and even for me it is mostly fascinating on the meta level, as a lens into the forces shaping the zeitgeist. It's not that the plight of a mistreated scientist deserves zero attention so much as that there are so many other things that are more worthy of attention—in a culture where people were a little less myopic and hypersensitized, perhaps hundreds of protestors would show up outside NYU feeling genuine panic and distress about one of the dozens of issues which directly impact their lives, rather than things which might marginally contribute to a culture of et cetera, whatever.
2. You probably shouldn't have a settled opinion.
In Duncan-culture, it would be considered somewhat gauche to confidently claim to know what had happened here, and how we ought respond on the object level; one would typically lose points for such a claim, and subsequently be taken less seriously, unless one could adequately demonstrate one's concrete reasons for confidence.
(I mean that part straightforwardly and seriously—in Duncan-culture, you would be listened to less after so demonstrably exhibiting such poor epistemic hygiene. "Okay, so there's a person whose confidence is Not A Proxy For Truth," thinks everyone else in the room.)
In our present culture, which has failed to have this particular social more, people are almost obligated to have a take, and in the more deranged corners of our society are treated with suspicion if that take is anything less than full-throated orthodoxy. This is bad; the correct answer to "are you on the side of abusers or are you on the side of justice" is "you keep using these words—I do not think they mean what you think they mean."
Or, to put it another way, even if the people wielding torches and pitchforks are correct that two plus two equals four, you should still feel a little unsettled at how insistent they are that you profess it, and should probably start thinking of ways to avoid or disperse such mobs in the future.
3. If you're not against the update cascades, then you're with the update cascades.
It's interesting to consider something like "where does morality live, in a society?" I have an as-yet-unwritten essay about how it basically lives in the bystanders, and in what will cause bystanders to take action. Someone shouting racist slurs in the grocery store will often be confronted by randos, because (at least in most places in America) we've reached the point where someone minding their business in the grocery store feels like it's their problem if someone is shouting racist slurs. Something they directly care about is at stake; the slurs are rending at a part of the social fabric that they consider to be their concern.
The social justice side of this disagreement understands this principle; the people showing up in protest at NYU are showing up in protest because they feel that sexual misconduct tolerated anywhere is a threat to good people everywhere.
But the other pole is woefully underrepresented; people generally do not seem to consider that they have something of a moral obligation to pump against [infinite punishment via each individual and group making the locally sensible decision to turn their backs].
According to Duncan-culture sensibilities, the darkest chapter of this saga is the one where NYU, which was clearly on the verge of forming a real, independent opinion instead of parroting the party line of either side, caved to public pressure. It is in fact the case that we do not know what happened with sufficient confidence to have an opinion that strong! From all the way over here (and I consider the vast majority of the protestors at NYU to be "all the way over here" in the relevant sense) we cannot tell whether MIT was right to fire Sabatini, and we cannot tell whether NYU was wrong to try to hire him, and the responsible thing to do is a) say nothing, except to b) shout down the people who are shouting anyway, despite having nothing justified to say.
99.9999% of us lack the moral prerequisites to confidently mete out punishments on the meta level, i.e. to punish non-punishers like NYU, or to punish punishers like MIT. It's certainly valid to hypothesize that either or both institutions did something morally culpable here (indeed, it's logically necessary in MIT's case, since they either harbored an abuser for decades or scapegoated an innocent man). But to move from hypothesis to conclusion, you need some kind of actual discerning information, and it can't shouldn't just be "well, this side felt more real to me."
It's particularly distressing that all of this seems to be downstream of the findings of the Whitehead Report, which really does seem to have accrued criticism far in excess of the usual "every report has its detractors" baseline. If it is in fact the case that multiple Whitehead employees complained about the lawyers' behavior—if it is in fact the case that NYU took a look and genuinely thought that Sabatini had been mistreated—in those worlds, it is really really bad that the report ended up being sufficient evidence, in practice, for every institution of higher learning to turn its back on one of the foremost researchers of the generation. I don't exactly expect courage and moral fiber from blankface institutions like MIT and NYU, but in Duncan-culture they do sometimes exhibit it, and it would've been nice to have a Welch stand up and say to all the people confidently declaring Sabatini a monster "Have you no sense of decency?"
I have other takeaways, but they're less coherent. There's something in here about paths-to-redemption, and how systems which kowtow to the mob don't really have them. There's something in here about monopolies on legitimate violence, and how those monopolies are threatened by modern social media, and we need norms and institutions to catch up. There's something in here about people allowing themselves to draw conclusions based on nothing, and then other people updating on those first people's confidence, which brings us right back around to the update cascades. There's something in here about the erosion of faith in institutions, such that basically no one is willing to trust in an institution that reaches the opposite conclusion of what one thinks is correct.
(i.e. the red tribe currently thinks that NYU made a grievous error; if NYU had stood its ground, the blue tribe would have concluded that NYU had made a grievous error. Neither side, upon hearing the surprising conclusion, would pause to think "huh, maybe we were wrong about this?" because let's face it, there aren't many people out there who are actually doing their due diligence and making a genuine attempt to mete out impartial justice.)
But mostly, the thing I want to keep shouting is YOU DO NOT KNOW WHAT YOU ARE PRETENDING TO KNOW. I feel mad about this—like the person in the grocery store squaring off against the raving racist, all of the people writing supremely confident opinion pieces that treat mere allegations as unassailable fact are threatening something that is precious and important to me. It's not that I particularly identify with Sabatini, and it's not that I fail to identify with Knouse. It's that I want to live in a world where, if I live through what either of them claims to have lived through, it's actually possible to get a fair hearing, and justice, and closure.
Right now, that seems to be literally impossible. These stories don't end, they just fall out of the spotlight, until ten years have passed and suddenly someone drags it all out into the open again.
That seems really bad. You should be able to settle these things, once and for all, and that only happens if people are sometimes willing to admit that they simply do not know.
It's not what it looks like.
I fear that the true legacy of this whole scandal, only ever uttered in the privacy of a Principal Investigator's mind, is that they should avoid mentoring women as much as possible, and attractive young women in particular. Or if they want to mentor a woman, they should significantly limit the closeness of the relationship, no breakfasts, no closed-door talks, no conferences together.
Not having sex with their mentees seems like a place to start.
Sure, but that might be too hard, in the same way that a recovering alcoholic wouldn't want to keep whisky close-by. Or there is a view that this risk is present even if nothing sexual happened at all.
It may be too late for some, but the idea is not to become an alcoholic in the first place.
But I agree with your second objection; innocence is not a reliable defense against false accusation. :(
It's interesting to consider something like "where does morality live, in a society?" I have an as-yet-unwritten essay about how it basically lives in the bystanders, and in what will cause bystanders to take action.
It's interesting to consider something like "where does morality live, in a society?" I have an as-yet-unwritten essay about how it basically lives in the bystanders, and in what will cause bystanders to take action.
This essay sounds like it shares a bunch with thoughts I’ve been thinking. I’d be interested to read this!
You don't have to care about this.
You probably shouldn't have a settled opinion.
These are indeed my positions :)
These two worlds are insanely disjoint.
In my experience, it is not that difficult to come up with an underlying reality where both are true - modulo some self-interested interpretation by both parties.
As long as the participants are aligned interpretations will mostly overlap but as people start to fight they tend to move diametrically apart.
And the participants will feel that their interpretation is "correct" - because - as Robin Hanson would say - that makes them more convincing to others.
I think there's a simpler explanation which is not mentioned in the post: the set of object-level facts mentioned in both narratives can be literally correct, and the narratives have approximately zero added epistemic value on top of that.
This works because of selection bias. If you have a robot toss a fair coin 1000 times and you care about whether there are more heads than tails or vice versa, and there are two people who can reveal to you 100 coins each with the goal of convincing you of one result or the other, it's obvious what will happen: the heads guy will reveal 100 heads, the tails guy will reveal 100 tails, and you'll be no closer to the truth than you were before.
This is almost always the case in the real world because the number of yes/no questions (something like Kolmogorov complexity in this setting) that you need to characterize a situation is always vastly larger than the number that people can actually process on a case-by-case basis, especially if the parties are pushing disinformation to pollute all the relevant information channels (which they will be doing). The result is that anyone can find many factoids to support their position and none of them need to be lying.
In this case a lot of what are claimed as object-level facts are directly contradictory, but I like this point (and the way you made it) overall. If I quote you in the future elsewhere (e.g. FB or other essays) would you prefer that I paraphrase or copy-paste, and would you prefer attribution or anonymity?
Yeah, I agree in this case there are some contradictory claims but I don't think they matter that much overall. That's because I think the line between lying and deception is not that sharp and there's an intuitively compelling sense in which the people revealing to you 100 heads or 100 tails out of 510 heads in 1000 tosses are "lying", or at least "being deceptive". That's the bigger problem, and the fact that they might mix this up with conflicting claims about e.g. whether the 758th coin came up heads or tails is in my opinion not too important.
As for quoting me, you can do that whenever you want and however you want. You can attribute it to me or not, you can paraphrase it or not, et cetera. Totally up to you.
And nice summary. Do you know if there is a writeup of an explanation like that somewhere that I can refer to (except of course your comment)?
Not contradicting what you say: This is sometimes (or even quite often) true, but it's still worth emphasizing that, if one side is in fact lying, while the other side is trying hard to be truthful, then your assumption would assign roughly equal blame, which incentivizes/rewards the lying. (It can be easy for bad actors to obfuscate the facts in a situation by lying about everything they think they can get away with.) So if you don't want to be a force that makes the world worse, you have an obligation to (at least) strongly consider and investigate the possibility that one side is almost completely responsible for all discrepancies.
your assumption would assign roughly equal blame
I wouldn't assign "blame" at all. Given their past and their cognitive structure, they acted as made the most sense in the moment.
But yes, I would give them the same benefit of the doubt. But over time, you would learn more about who behaves trustworthy. You rarely learn that about public persons, but that mostly doesn't matter.
Two people fighting is, in any case, an indication to not trust both.
But in public, you will only see fights because the good non-fighters are invisible. Thus: Do not watch the news.
>I wouldn't assign "blame" at all. Given their past and their cognitive structure, they acted as made the most sense in the moment. I concede that "blame" is a bad framing. What I'm trying to say is that, if we want to move towards a world where people are happier on average, it's important to treat liars and self-deceivers differently than we'd treat people who don't do these things. For instance, people who got caught doing these things repeatedly would no longer get the benefit of the doubt. (I think I'm saying something very trivial here, so we probably agree!)>Two people fighting is, in any case, an indication to not trust both. Kind of, when we think of loose correlations. But sometimes the connection is unwarranted for one of the participants of the fight, in which case it's important to hold space for that possibility (that one side was unfairly accused or otherwise caught up in something, or [edit:] that the side was correctly accused but somehow managed to deny, attack and reverse victim and offender in the social environment's perception of the incident).If you get dragged into a fight for unfair reasons, it's arguably kind of suboptimal to just give in to the attacker and avoid making a scene. "Be the bigger person" makes sense in a lot of circumstances, and even when it seems suboptimal, it can be very understandable. Still, there are situations where you definitely can't fault people for fighting back. Sometimes people who fight are justly defending themselves against an attacker. Sometimes that's brave, and other times that may even be their only option because the attack is an existential threat to their social standing, and they no longer have anything to lose. I think it's the case somewhat frequently that attackers make things up or bizarrely misrepresent stuff. Dark triad personality traits (and vulnerable dark triad) tend to be involved, and while it's possible for people with such traits to adhere to "Social Contract" principles, many don't. (It's often part of the symptom criteria that they don't, but I definitely don't want to demonize groups of people based on a cluster of criteria that doesn't always present in the same way.)Overall, I very much second Duncan's norm of "withholding judgment" rather than a stance like "two people fighting is an indication to not trust both." And I also think that it's good to cultivate a desire to get to the bottom of things, rather than have an attitude of "oh, another fight, looks like the truth is somewhere in between." That said, the original post by Duncan exemplifies that it's often not practically possible to get to the bottom of it, and that's arguably one of the most unfortunate things about human civilization.
I think we are in violent agreement.
if we want to move towards a world where people are happier on average, it's important to treat liars and self-deceivers differently
Yes, we should design incentives such that the better outcome is the more likely outcome.
If you get dragged into a fight for unfair reasons, it's arguably kind of suboptimal to just give in to the attacker and avoid making a scene.
Agree. My point was not to treat both people fighting equally forever. But initially, if all you know is that two people are fighting, that is some evidence for both of them being people you might want to avoid. I agree with "withholding judgment" - there is no reason to commit to one side or no side unless forced by sub-optimal mechanisms.
"trust" is too big a word to use here. Two people fighting is a reason not to invest much of your future happiness in either, unless you're part of the fight. Fighting harms all participants, EVEN the righteous. Your best EV, absent any other connections to the fight or fighters, is not to join.
You MAY want to join the meta-fight, arguing for procedures or framing that make this fight less harmful or easier to decide correctly. This is what I think Duncan is trying to do with this post - to show that most of us can't update very much in any direction based on this, so it should overall be given less weight than it is (in some circles, at least).
To avoid attracting unwanted internet debate, I have hidden this post from logged out users and newly created accounts.
Haven't discussed with other mods though, so might reverse if I reconsider the risk.
Hm. It seems entirely fine to me to have it be unpossible for logged-out users and newly created accounts to comment, but I think I'd prefer to have it readable by e.g. my FB audience who aren't LWers?
Else if you prefer I could shift this over to Medium.
Chatted with Ray and seems okay to make it generally readable again. I will do that.
Restricting comments does seem like a good option, though we'd have to create the option for that.
I think this is a really good post and considered even "frontpage but hidden from new users" since I think this does have timeless Rationality content.
My thought was: If this post anonymized the people-in-question, I'd feel pretty fine about it being frontpaged (without any special hidden-ness). I do think that'd also make it a worse post in some ways – the concrete examples are helpful, and the evidence you've specifically cited (pretty hard to anonymize), helps show what the process of evaluation is like.
Avoid attracting unwanted internet debate... reconsider[ing] the risk.
It seems entirely fine to me... if you prefer I could shift this over to Medium.
Seems like a good opportunity to practice murphyjutsu, which I just learned about just now and I really like it so I'm going to practice it to keep it cognitively available.
I've been taking what I think is a similar stance to the Johnny Depp / Amber Heard trial that's been all over the news recently: there's basically no way I'll ever find out the real truth, I have nothing personal at stake here and can't affect the outcome anyway, and I don't find much entertainment value in learning the details, so I'll continue to ignore it and have no opinion.
I went to tag this post, and... couldn't easily find a more specific tag than "rationality". I feel like this post is focused on a particular concept that's, like, an established sequence concept, but I'm not sure what it is. "Virtue of Lightness" / "What do you think you know and how do you think you know it?" are the two catchphrases that feel relevant, but those don't feel like they have quite the right type-signature for a tag.
There are 2 posts that are somewhat related, but shifted back 13 years, and they don't have super helpful tags either. Maybe there are lots of these laying around?
You Be the Jury: Survey on a Current Event
The Amanda Knox Test: How an Hour on the Internet Beats a Year in the Courtroom
Thanks for digging these up!
(I tried thinking for a while and couldn't come up with anything either. I think it maybe falls under the heading "split and commit," which I think really is one of the more fundamental motions or concepts, but it doesn't have a long history in LW culture and also doesn't feel like it's quite the right type signature.)
I think if we got to, like, 5 split-and-commit posts it'd make sense as a tag, but it still feels like that should just be a subset of a broader "lightness" concept.
"Epistemic Lightness", "Epistemic Hygiene" feel close to the right type signature.
Turns out we already had an epistemic hygiene wiki post (from Old LW), which I just converted into a tag and added this post to. (further edit: not quite sure it was the right thing but seems better to fail-forwards)
You probably mean "quasirapey asshole".