Charlotte

humans: 

*a generally normal object but very, very small*

@onthecareandfeedingofcatgirls​ is typically the one I tag for more in-depth explanations.

is this about real people?? you mean human beings (mostly male???) are walking around with cat-ears tiaras, thinking themselves very rational, and calling themselves “catgirls”???

You take a vow of poverty but find a loophole by having your followers pay for the palace you live in so you can sit on Twitter and tell the rich that they’re evil for buying their own houses.

“Any system in which social relationships are determined entirely by economic factors is contrary to the nature of the human person and his acts…A system that “subordinates the basic rights of individuals and of groups to the collective organization of production” is contrary to human dignity…The Church has rejected the totalitarian and atheistic ideologies associated in modern times with “communism” or “socialism”“

pure and chaste

entirely abstaining from the pleasures of the flesh”

“That can be arranged”

1. Not sure if @bendini1 already knows this, but the Bay community wasn’t planned, in the sense of someone saying “let’s have a Bay community”. Lots of rationalists are involved in tech, and they all ended up in the world tech hub, and then some people founded organizations to take advantage of it being the world tech hub, and then other people moved in to work for those organizations, and then other people moved in to be with their friends, and then other people moved there because it seemed to have critical mass. I moved there originally because I was low on money and @luminousalicorn agreed to let me live with her. Then I met lots of people there and liked it and decided to stick around.

Why does this matter? Because anyone trying to start an alternative hub isn’t just replicating the original process that made the Bay Area a hub. They’re trying to do something that’s never been done before - push against entropy and try to bring people to a place they wouldn’t otherwise be. Some seasteaders or secessionists or micronationalists trying to start a new country might think “Oh, well, France is a country and they seem to be doing fine, how hard could it be?”. But French people didn’t “build” France, in the sense of wrest a group of people from all over the world to form a country despite all obstacles. France just happened. If you’re trying to start your own country, your task is completely different from France’s and any attempt to compare them is going to make you think it’s much easier than it is.

To create a community anywhere other than the Bay (or Seattle, or other major tech hubs/universities in exact proportion to the degree to which they’re a major tech hub/university) is going to take social technology and institutions and pushing-against-entropy-ability that have never been tested before.

2. I don’t live in a group house. But my girlfriend’s old group house had seven people. One was the head of MIRI. One was the head of AI Impacts. One worked for OpenAI researching AI alignment. One worked for CFAR working on rationality techniques. One worked for OpenPhil researching effective charities. One ran an anti-aging website and newsletter. And the last one was Ben Hoffman of Compass Rose. The group house threw great parties and a bunch of them were in polyamorous relationships with each other.

So when people say “Oh, people just live in group houses and throw parties and have polyamorous relationships with each other, which CLEARLY means they’re not doing any real work”, I start wondering where exactly they’re getting their information from.

Any claim that going to Berkeley has stopped people from doing anything interesting has to deal with MIRI, CFAR, OpenPhil, Center For Effective Altruism, GiveWell, Leverage Research, the Center for Human Compatible AI, the Berkeley Existential Risk Project, SPARC, OpenAI, all being based on the Bay and mostly or at least partly rationalists. Other people are working remotely for someone or other, like Ozy’s work for Wild Animal Suffering Research, or Ben and Oliver’s work on LW 2.0. Some people I know who aren’t involved in any of those things are doing big data startups, pharmaceutical startups, anti-aging startups, cryogenics startups, Bitcoin startups, AI startups, secret-Peter-Thiel-sinister-agenda-related startups, or just earning to give really effectively. Some people who aren’t involved in any of *those* things have their own personal projects but aren’t prepared to go public yet.

I’m really curious if the "people in the Bay Area seem unproductive” crowd have added up all of those things, taken a census of how many people are in the Bay Area, and decided the number of things going on is lower than they would expect given the population. Or if it’s more like “well, nobody’s cured death yet, guess they’re not really trying”.

3. The first step in claiming that rationalists have neglected “The Craft” is a very clear explanation of what exactly “The Craft” is.

I can think of one interpretation in which it’s sort of like Superforecasting on steroids. I agree there aren’t enough people doing this.

I can think of another interpretation in which it’s something more like what I do with Slate Star Codex. I go off and pursue things I’m curious about, come across difficult questions I don’t yet know how to answer, work to try to understand them better in some non-systematized and non-systematizeable way - then report back to others.

There’s some stuff in the Sequences like http://lesswrong.com/lw/9c/mandatory_secret_identities/ which suggests that pursuing “The Craft” separately from other projects is an error, that nobody should claim to be working on “The Craft” until they’ve already got other projects in traditional object-level fields going. In this model, the people who work boring day jobs and then report back on what they’ve learned about rationality are doing it exactly right.

There’s an even stronger version in which rationality is an attitude you bring to everything else. A person who does nothing but go to parties all day, but who watches and improves their mental processes as they decide which parties to go to, is a better rationalist than someone who writes books about biases and heuristics haphazardly and poorly. In this model, having a “rationalist culture” or “rationalist society” is a big part of what you need - you want to be interacting with people who will encourage those tendencies in you and keep you honest.

O, thou art like a summer’s day But fairer and more temperate Rough winds do shake the buds of May And summer hath too short a date.

Sometimes too hot the sunlight shines And sometimes is its color dimmed And every fair from fair declines By chance, or nature’s course, untrimmed.

But thy own summer shall not fade Nor shalt thou lose the fair thou ow'st Nor shalt Death have thou in his shade In timeless lines to Time thou growst.

A summer’s day? No, you’re more fair! There’s storms in May And sunshine’s rare.

The sky’s too bright Or else too dim; All charms take flight By nature’s whim.

But you won’t fade, Nor lose your fair, Nor reach death’s shade; These lines will spare!

The heat: a cage; None cheat Old age 

But thou Willt stay Ask how? If they 

Privileged people rarely take the voices of marginalized people seriously. Social justices spaces attempt to fix this with rules about how to respond to when marginalized people tell you that you’ve done something wrong. Like most formal descriptions of social skills, the rules don’t quite match reality. This is causing some problems that I think we could fix with a more honest conversation about how to respond to criticism.

The formal social justice rules say something like this:

You should listen to marginalized people.

When a marginalized person calls you out, don’t argue.

Believe them, apologize, and don’t do it again.

When you see others doing what you were called out for doing, call them out.

Those rules are a good approximation of some things, but they don’t actually work. It is impossible to follow them literally, in part because:

Marginalized people are not a monolith. 

Marginalized people have the same range of opinions as privileged people.

When two marginalized people tell you logically incompatible things, it is impossible to act on both sets of instructions.

For instance, some women believe that abortion is a human right foundational human right for women. Some women believe that abortion is murder and an attack on women and girls.

“Listen to women” doesn’t tell you who to believe, what policy to support, or how to talk about abortion. 

For instance, some women believe that religious rules about clothing liberate women from sexual objectification, other women believe that religious rules about clothing sexually objectify women. 

“Listen to women” doesn’t tell you what to believe about modesty rules. 

Narrowing it to “listen to women of minority faiths” doesn’t help, because women disagree about this within every faith.

When “listen to marginalized people” means “adopt a particular position”, marginalized people are treated as rhetorical props rather than real people.

Objectifying marginalized people does not create justice.

Since the rule is literally impossible to follow, no one is actually succeeding at following it. What usually ends up happening when people try is that:

One opinion gets lifted up as “the position of marginalized people” 

Agreeing with that opinion is called “listen to marginalized people”

Disagreeing with that opinion is called “talking over marginalized people”

Marginalized people who disagree with that opinion are called out by privileged people for “talking over marginalized people”.

This results in a lot of fights over who is the true voice of the marginalized people.

We need an approach that is more conducive to real listening and learning.

This version of the rule also leaves us open to sabotage:

There are a lot of people who don’t want us to be able to talk to each other and build effective coalitions.

Some of them are using the language of call-outs to undermine everyone who emerges as an effective progressive leader. 

They say that they are marginalized people, and make up lies about leaders.

Or they say things that are technically true, but taken out of context in deliberately misleading ways.

The rules about shutting up and listening to marginalized people make it very difficult to contradict these lies and distortions. 

(Sometimes they really are members of the marginalized groups they claim to speak for. Sometimes they’re outright lying about who they are).

(For instance, Russian intelligence agents have used social media to pretend to be marginalized Americans and spread lies about Hillary Clinton.)

The formal rule is also easily exploited by abusive people, along these lines:

An abusive person convinces their victim that they are the voice of marginalized people.

The abuser uses the rules about “when people tell you that you’re being oppressive, don’t argue” to control the victim.

Whenever the victim tries to stand up for themself, the abuser tells the victim that they’re being oppressive.

That can be a powerfully effective way to make victims in our communities feel that they have no right to resist abuse. 

This can also prevent victims from getting support in basic ways.

Abusers can send victims into depression spirals by convincing them that everything that brings them pleasure is oppressive and immoral. 

The abuser may also isolate the victim by telling them that it would be oppressive for them to spend time with their friends and family, try to access victim services, or call the police. 

The abuser may also separate the victim from their community and natural allies by spreading baseless rumors about their supposed oppressive behavior. (Or threatening to do so).

When there are rules against questioning call outs, there are also implicit rules against taking the side of a victim when the abuser uses the language of calling out.

Rules that say some people should unconditionally defer to others are always dangerous.

The rule also lacks intersectionality:

No one experiences every form of oppression or every form of privilege.

Call-outs often involve people who are marginalized in different ways. 

Often, both sides in the conflict have a point.

For instance, black men have male privilege and white women have white privilege.

If a white woman calls a black man out for sexism and he responds by calling her out for racism (or vice versa), “listened to marginalized people” isn’t a very helpful rule because they’re both marginalized.

These conversations tend to degenerate into an argument about which form of marginalization is most significant.

This prevents people involved from actually listening to each other.

In conflicts like this, it’s often the case that both sides have a legitimate point. (In ways that are often not immediately obvious.)

We need to be able to work through these conflicts without expecting simplistic rules to resolve them in advance.

This rule also tends to prevent groups centered around one form of marginalized from coming to engage with other forms of marginalization:

For instance, in some spaces, racism and sexism are known to be issues, but ableism is not.

(This can occur in any combination. Eg: There are also spaces that get ableism and sexism but not racism, and spaces that get economic justice and racism but not antisemitism, or any number of other things.)

When disabled people raise the issue of ableism in any context (social justice or otherwise), they’re likely to be shouted down and told that it’s not important.

In social justice spaces, this shouting down is often done in the name of “listening to marginalized people”.

For instance, disabled people may be told ‘you need to listen to marginalized people and de-center your issues’, carrying the implication that ableism is less important than other forms of oppression.

(This happens to *every* marginalized group in some context or other.)

If we want real intersectional solidarity, we need to have space for ongoing conflicts that are not simple to resolve.

https://www.aeaweb.org/articles?id=10.1257/aer.20131578 The classic one is that there’s some evidence that trade with China did cost the US a significant number of jobs, and this harmed a lot of people in the US. I am uncertain this claim is right, but until very recently you weren’t even allowed to conjecture this without being shut down as some sort of anti-intellectual racist hick. And then, years later, oops! I guess all those racist hicks might have noticed a real thing! Whoopsie!

A more recent one, from SSC, was this http://www.businessinsider.com/check-cashing-stores-good-deal-upenn-professor-2017-2 where it turns out that poor people keep going to check-cashing stores for…rational reasons. Because it works better for them given what they value as a demographic. Yet we’ve built our policy models around the idea that evil dirty check-cashing stores are tricking poor dumb yokels into using them, and that we need to educate those poor people right.

I apologize if I’m getting heated, but it really does seem like a recurring theme where we wonks create a model, it matches well enough, and then when it starts encountering contradictions with reality we ignore (or worse, punish) the humans that are contradicting our policy models. Chris Arnade is going out and gathering the results of our policies, and he’s trying to do it with compassion and giving dignity to the people involved. That is a valuable service, especially to those of us who want to dismiss it.

Because I agree with you that this kind of investigation is valuable and important, and not done often enough, especially in parts of academia (I think there ought to be far more “field work” done in economics, and that check cashing case is a great example of why this matters).  But everything I have seen from Arnade suggests that he is not simply doing his part among journalists (there are others!), academics, etc. who talk to people on the ground and thus provide this vital kind of knowledge.  My impression of Arnade is that he thinks (to exaggerate only mildly) he is the only person that ever talks to people on the ground, and – what’s more – that makes his own general models more plausible than anyone else’s, even in cases where he hasn’t talked to people, just because he is in this unique state of talking-to-people-related enlightenment.

Or, as OP said, he acts “like the only person on twitter who has ever met a poor person,” and uses this quality of himself as a person as a discursive weapon.

Take for instance this passage from the article I linked above:

Recently I came home from a day in the Bronx, exhausted by the cumulative effect of a thousand little dramas. I opened my computer and looked at my twitter stream. My old world reappeared. I read smart commentary on places like Brazil. I scrolled through tweet after tweet, all shouting the story of the day.

It was all logical and rational and clever. Everyone was throwing different numbers around to show they were right.

Few were telling the messy and complex stories of the struggles to navigate the illogical and absurd reality of life, about the consequences of the news on, you know, people.

One smart columnist, Matt Yglesias, was arguing why the death of 1,129 people in a clothing factory in Bangladesh was understandable and “OK.” Poor countries need lax labor laws before they can be rich … It went something like that.

The author had fallen so far down the wormhole of numbers and clever arguments that he had forgotten humans were involved. Like Wall Street and I had forgotten that it was humans we were loaning money to and that it was humans who we were foreclosing on and it was humans whose governments were defaulting.

I wanted to tweet back. “Go to Bangladesh. Talk to one of the children of the dead. Hell, don’t just talk to one. Spend two weeks listening.”

He hadn’t gone to Bangladesh. He had just read a few headlines from a competing columnist and decided to argue based on that.

After much thought I tweeted to him, “you are an idiot.”

Yglesias’ post was tonedeaf, and he apologized for it.  But let’s note a few things.  First, Arnade didn’t go to Bangladesh either.  He imagined the misery he would witness there, from the armchair, and imagined that Yglelsias would change his mind upon witnessing this misery.  The argument isn’t “you’re wrong, and I know because I actually talked to the affected people”; the argument, as far as I can understand, is “you’re wrong, and I know because I am the sort of person who talks to affected people.”  And the end result – perfectly and poetically – was yet more standard twitter discourse between armchair jackoffs: “you are an idiot.”

Second, there is a real policy question here (“do poor countries need lax labor laws before they can be rich?”), and like many policy questions, it involves accepting one bad thing, X, to avoid another, Y.  This is open to multiple kinds of attack: that X is in fact worse than Y; that X is immediate and assured while Y is only the prediction of a fallible model.  What does not constitute a good attack is to point out that X is in fact bad.  We know that already.  Probably some people do need to “go to Bangladesh” (or the equivalent, for whatever policy question) to have the true awfulness of X rubbed in their unwilling face.  Sometimes this will change minds that ought to be changed.  But the armchair posture of “the only one here who feels the badness of bad things” is not conducive to helping people.  There are always people suffering, no matter what policy is implemented, and in policy discourse, “you don’t get it, suffering is really bad” is an argument that can serve any master, prop up any side of any debate ad hoc.  Its application is always ad hoc, because nothing important is without downsides.

(This all carries over in p. obvious ways to Arnade’s writing/tweeting about Trump and Trump voters.)

Finally, on a more emotional level, there is something that just … squicks me out about the idea of individual people getting used as discourse one-upmanship tools by the journalists who’ve talked to them, who’ve “spent two weeks” listening and self-consciously emphathizing, who feel their pain.  Thankfully there is nothing in my life like the Bangladesh factory collapse, but there are cases where I can imagine becoming such a token.  I’ve talked a lot on here about how I spent 5 years in my adolescence on Risperdal, and how it messed me up, and how I think there needs to be more awareness about the dangers of antipsychotics.  But I know very little about the ugly procedural guts of the system that ended up producing my 5-year Risperdal prescription.  And my insides squirm when I think about some journalist coming to talk to me, listening to my own story, and then using “I’ve talked to people hurt by Risperdal” as a trump card in some complicated policy argument.  Would their favored policies actually help people like younger me?  Would they help them, but at the cost of hurting other people worse?  Those are among the questions I would ask if I were part of the debate, but I wouldn’t be part of the debate; having been Talked To, my usefulness would have ended.