Scott Alexander


Introduction to Game Theory
The Blue-Minimizing Robot
Hypotheses and Hunches
Probability and Predictions
Parables and Prayers
Futurism and Forecasting
Community and Cooperation
Economics and Efficiency
Research and Reviews
Load More (9/13)

Wiki Contributions


My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

Thanks for this.

I'm interested in figuring out more what's going on here - how do you feel about emailing me, hashing out the privacy issues, and, if we can get them hashed out, you telling me the four people you're thinking of who had psychotic episodes?

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

I agree I'm being somewhat inconsistent, I'd rather do that than prematurely force consistency and end up being wrong or missing some subtlety. I'm trying to figure out what went on in these cases in more details and will probably want to ask you a lot of questions by email if you're open to that.

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

If this information isn't too private, can you send it to me?

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

Yes, I agree with you that all of this is very awkward.

I think the basic liberal model where everyone uses Reason a lot and we basically trust their judgments is a good first approximation and we should generally use it.

But we have to admit at least small violations of it even to get the concept of "cult". Not just the sort of weak cults we're discussing here, but even the really strong cults like Heaven's Gate or Jamestown. In the liberal model, someone should be able to use Reason to conclude that being in Heaven's Gate is bad for them, and leave. When we use the word "cult", we're implicitly agreeing that this doesn't always work, and we're bringing in creepier and less comprehensible ideas like "charisma" and "brainwashing" and "cognitive dissonance".

(and the same thing with the concept of "emotionally abusive relationship")

I don't want to call the Vassarites a cult because I'm sure someone will confront me with a Cult Checklist that they don't meet, but I think that it's not too crazy to argue that some of these same creepy ideas like charisma and so on were at work there. And everyone knows cults can get weird and end in mental illness. I agree it's weird that you can get that far without robes and chanting or anything, but I never claimed to really understand exactly how cults work, plus I'm sure the drugs helped.

I think believing cults are possible is different in degree if not in kind from Leverage "doing call on demonic energies and use their power to affect the practitioners' social standing". I'm claiming, though I can't prove it, that what I'm saying is more towards the "believing cults are possible" side.

I'm actually very worried about this! I hate admitting cults are possible! If you admit cults are possible, you have to acknowledge that the basic liberal model has gaps, and then you get things like if an evangelical deconverts to atheism, the other evangelicals can say "Oh, he's in a cult, we need to kidnap and deprogram him since his best self wouldn't agree with the deconversion." I want to be extremely careful in when we do things like that, which is why I'm not actually "calling for isolating Michael Vassar from his friends". I think in the Outside View we should almost never do this!

But you were the one to mention this cluster of psychotic breaks, and I am trying to provide what I think is a more accurate perspective on them. Maybe in the future we learn that this was because of some weird neuroinflammatory virus that got passed around at a Vassarite meeting and we laugh that we were ever dumb enough to think a person/group could transmit psychotic breaks. But until then, I think the data point that all of this was associated with Vassar and the Vassarites is one we shouldn't just ignore.

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

I'm having trouble figuring out how to respond to this hostile framing. I mean, it's true that I've talked with Michael many times about ways in which (in his view, and separately in mine) MIRI, CfAR, and "the community" have failed to live up to their stated purposes. Separately, it's also true that, on occasion, Michael has recommended I take drugs. (The specific recommendations I recall were weed and psilocybin. I always said No; drug use seems like a very bad idea given my history of psych problems.)


Michael is a charismatic guy who has strong views and argues forcefully for them. That's not the same thing as having mysterious mind powers to "make people paranoid" or cause psychotic breaks! (To the extent that there is a correlation between talking to Michael and having psych issues, I suspect a lot of it is a selection effect rather than causal: Michael told me once that he specifically seeks out people who are high in Eysenckian psychoticism.) If someone thinks Michael is wrong about something, great: I'm sure he'd be happy to argue about it, time permitting. But under-evidenced aspersions that someone is somehow dangerous just to talk to are not an argument.

I more or less Outside View agree with you on this, which is why I don't go around making call-out threads or demanding people ban Michael from the community or anything like that (I'm only talking about it now because I feel like it's fair for the community to try to defend itself after Jessica attributed all of this to the wider community instead of Vassar specifically) "This guy makes people psychotic by talking to them" is a silly accusation to go around making, and I hate that I have to do it!

But also, I do kind of notice the skulls and they are really consistent, and I would feel bad if my desire not to say this ridiculous thing resulted in more people getting hurt.

I think the minimum viable narrative here is, as you say, something like "Michael is very good at spotting people right on the verge of psychosis, and then he suggests they take drugs." Maybe a slightly more complicated narrative involves bringing them to a state of total epistemic doubt where they can't trust any institutions or any of the people they formerly thought were their friends, although now this is getting back into the "he's just having normal truth-seeking conversation" objection. He also seems really good at pushing trans people's buttons in terms of their underlying anxiety around gender dysphoria (see the Ziz post) , so maybe that contributes somehow. I don't know how it happens, I'm sufficiently embarrassed to be upset about something which looks like "having a nice interesting conversation" from the outside, and I don't want to violate liberal norms that you're allowed to have conversations - but I think those norms also make it okay to point out the very high rate at which those conversations end in mental breakdowns.

Maybe one analogy would be people with serial emotional abusive relationships - should we be okay with people dating Brent? Like yes, he had a string of horrible relationships that left the other person feeling violated and abused and self-hating and trapped. On the other, most of this, from the outside, looked like talking. He explained why it would be hurtful for the other person to leave the relationship or not do what he wanted, and he was convincing and forceful enough about it that it worked (I understand he also sometimes used violence, but I think the narrative still makes sense without it). Even so, the community tried to make sure people knew if they started a relationship with him they would get hurt, and eventually got really insistent about that. I do feel like this was a sort of boundary crossing of important liberal norms, but I think you've got to at least leave that possibility open for when things get really weird.

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

I don't want to reveal any more specific private information than this without your consent, but let it be registered that I disagree with your assessment that your joining the Vassarites wasn't harmful to you. I was not around for the 2017 issues (though if you reread our email exchanges from April you will understand why I'm suspicious), but when you had some more minor issues in 2019 I was more in the loop and I ended out emailing the Vassarites (deliberately excluding you from the email, a decision I will defend in private if you ask me) accusing them of making your situation worse and asking them to maybe lay off you until you were maybe feeling slightly better, and obviously they just responded with their "it's correct to be freaking about learning your entire society is corrupt and gaslighting" shtick. 

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

It was on the Register of Bans, which unfortunately went down after I deleted the blog. I admit I didn't publicize it very well because this was a kind of sensitive situation and I was trying to do it without destroying his reputation.

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

Thanks, if you meant that, when someone is at a very early stage of thinking strange things, you should talk to them about it and try to come to a mutual agreement on how worrying this is and what the criteria would be for psych treatment, instead of immediately dehumanizing them and demanding the treatment right away, then I 100% agree.

My experience at and around MIRI and CFAR (inspired by Zoe Curzi's writeup of experiences at Leverage)

I don't remember the exact words in our last conversation. If I said that, I was wrong and I apologize.

My position is that in schizophrenia (which is a specific condition and not just the same thing as psychosis), lifetime antipsychotics might be appropriate. EG this paper suggests continuing for twelve months after a first schizophrenic episode and then stopping and seeing how things go, which seems reasonable to me. It also says that if every time you take someone off antipsychotics they become fully and dangerous psychotic again, then lifetime antipsychotics are probably their best bet. In a case like that, I would want the patient's buy-in, ie if they were medicated after a psychotic episode I would advise them of the reasons why continued antipsychotic use was recommended in their case, if they said they didn't want it we would explore why given the very high risk level, and if they still said they didn't want it then I would follow their direction.

I didn't get a chance to talk to you during your episode, so I don't know exactly what was going on. I do think that psychosis should be thought of differently than just "weird thoughts that might be true", as more of a whole-body nerve-and-brain dysregulation of which weird thoughts are just one symptom. I think in mild psychosis it's possible to snap someone back to reality where they agree their weird thoughts aren't true, but in severe psychosis it isn't (I remember when I was a student I tried so hard to convince someone that they weren't royalty, hours of passionate debate, and it just did nothing). I think if someone has mild psychosis and you can guide them back to reality-based thoughts for a second, that is compassionate and a good thing to do in the sense that it will make them feel better, but also kind of useless because the psychosis still has the same chance of progressing into severe psychosis anyway - you're treating a symptom. Analogy to eg someone having chest pain from a heart attack, and you give them painkillers for the pain but don't treat the heart attack.

(although there's a separate point where it would be wrong and objectifying to falsely claim someone who's just thinking differently is psychotic or pre-psychotic, given that you did end up psychotic it doesn't sound like the people involved were making that mistake)

My impression is that some medium percent of psychotic episodes end in permanent reduced functioning, and some other medium percent end in suicide or jail or some other really negative consequence, and this is scary enough that treating it is always an emergency, and just treating the symptom but leaving the underlying condition is really risky.

I agree many psychiatrists are terrible and that wanting to avoid them is a really sympathetic desire, but when it's something really serious like psychosis I think of this as like wanting to avoid surgeons (another medical profession with more than its share of jerks!) when you need an emergency surgery.

Load More