"Everyone has a plan until they get punched in the face." 

- Mike Tyson

(The exact phrasing of that quote changes, this is my favourite.)

I think there is an open, important weakness in many people. We assume those we communicate with are basically trustworthy. Further, I think there is an important flaw in the current rationality community. We spend a lot of time focusing on subtle epistemic mistakes, teasing apart flaws in methodology and practicing the principle of charity. This creates a vulnerability to someone willing to just say outright false things. We’re kinda slow about reacting to that.

Suggested reading: Might People on the Internet Sometimes Lie, People Will Sometimes Just Lie About You. Epistemic status: My Best Guess.

I.

Getting punched in the face is an odd experience. I'm not sure I recommend it, but people have done weirder things in the name of experiencing novel psychological states. If it happens in a somewhat safety-negligent sparring ring, or if you and a buddy go out in the back yard tomorrow night to try it, I expect the punch gets pulled and it's still weird. There's a jerk of motion your eyes try to catch up with, a sort of proprioceptive sliding effect as your brain wonders who moved your head if it wasn't you. 

If it happens by surprise out on the sidewalk and the punch had real strength behind it, so much the worse. The world changes colour, it feels like time stops being a steady linear progression and becomes unbelievably detailed beads on a string at irregular moments, internal narration becomes disassociated and tries to come up with an explanation for why your body is moving the way it is, whether staring up at the sky with a gormless expression on your face or shaking and shoving your hands forward.

And two seconds before the hit, you were thinking "He's not actually going to hit me."

Anyway, your emotional reaction might surprise you. Getting hit shakes you up a bit. Other people often don't react intelligently either. "Are you okay?" Yeah, right as rain, I’m just holding my hand to my bleeding nose because I think I look good doing it. “I can’t believe he did that!” Do you not believe your eyes? “What just happened?” I just got punched in the face, that’s what happened, what’s up with you?

Punching people in the face is generally a bad idea. Not only is it likely to get you in trouble, but the face is one big crumple zone. Jaw, teeth, cheekbones, those are all hard and pointy. Your handbones are fragile. If you manage to put power behind the hit, they are going to have a very visible injury, which can put the observers on their side. And yet people still hit each other in the face. Sometimes that’s part of what stuns you. “Surely,” you think, “nobody would have just told a lie that obvious. Something else must have happened.” You can become so confused that you ask out loud “what just happened?” and a bystander has to say “you just got punched in the face, that’s what happened.”

Getting lied to is an odd experience. It's not the same experience to be sure, but I noticed enough things in common between the two that I kept drawing useful comparisons.

II.

I've noticed this flaw in my own mind where I'm either skeptical of everybody, or basically trust everybody. If I’m skeptical of everybody I tend to say more false things, and if I basically trust everybody then I’m a lot more open and honest. Importantly, once I’m skeptical of everybody I start checking more and more, trying to verify things I once took on faith or contemplating exactly what I think I know and how I think I know that.

This flaw is not unlike the long, slow march I made out of having hair-trigger reflexes around getting hit. For reasons I’m not going to go into at the moment, when I arrived at college I was pretty quick to pattern match fast or unexpected movements as incoming strikes. I didn’t actually wind up misfiring and hurting anyone, but there were a few close calls where a friend or partner did something startling and I started to react before having to abort.

In college I contemplated working in information security, and ultimately decided not to. I didn’t like the competitive nature of it, and suspected the kind of thing in this tweet would not be good for my head.

(I edited the screenshot of this Twitter thread. Did you notice?)

One archetypical example I can think of where someone points out someone else might be lying is Concerns About Intentional Insights. It’s very careful, thorough, and organized. For my money though, Alison Gu’s fraud defenses are more illuminating. Mrs. Gu was accused of bank fraud, identity theft, and lying on a passport application. Here’s some excerpts from the Manchester Journal:

Gu, 49, now from Cheshire, Conn., is especially unhappy with defense lawyer Lisa Shelkrot, who refused to use the seven Chinese-speaking actors that the defendant recruited and were provided scripts about what to say during the jury trial.

…

Shelkrot has said she determined the bogus witnesses were actors after they arrived in Burlington midway through the trial in November 2017. Shelkrot, in a court affidavit, now says as she prepared three of them to testify at trial, one happened to remark, “It’s for the movie.”

Shelkrot said she asked about what movie. The witness responded, “Are you a real lawyer?”

Shelkrot wrote, “I answered that I was a real lawyer with a real case and a real client, and that we were going to real court on Monday, where the witness would be expected to take a real oath.”

What level of CONSTANT VIGILANCE do you have to be operating on where when a court case witness comes in, you check that the witness thinks they’re here for a real court case instead of a movie? 

If I were a lawyer or judge, and I had to deal with people pulling that level of epistemic sabotage on me, I would become paranoid. After a couple months of that when I walked into a store and the cashier said “Welcome to Burger King” I would start reflexively checking nearby signs to make sure it wasn’t actually a Taco Bell, or maybe a Goodyear tire shop. I would start prodding my burger with a toothpick to make sure it didn’t secretly have live beetles in it. I would be so suspicious. 

(I uh. Did wind up in a role where people keep trying shenanigans on me, and while the above is a bit of an exaggeration for humorous effect the experience of the last couple years has made me a less trusting human being.)

III.

In The Dark Arts Are A Scaffolding Skill For Rationality, I talk about how manipulation and lies aren’t skills we want a polished rationalist to have, but that they might be skills useful for training rationalists.

The minimum viable version of that training might be poker. Not because poker teaches you to practice some basic deceit and bluffing. This would be a special version of poker. In advance of the poker game, the person bringing the cards would get one of those explanation cards with all the poker hands, only it’s custom printed to be wrong. The explanation card would then insist that you can also make a straight out of every other card, so for instance 5 7 9 J K can be a straight. They would have a pair of audience plants who would agree that’s always been how poker works. They would have rigged the router for the local wifi to present a man-in-the-middle attack on the wikipedia page for poker. 

As the magician Teller of Penn & Teller once said, “sometimes magic is just someone spending more time on something than anyone else might reasonably expect.”

Just so with deception, only there’s surprisingly low-hanging fruit in lies among rationalists.

I’ve run a few meetups where I wore a sign that said “Might Be Lying” on it in big letters. It took several tries before a group called me on the subterfuge I got up to during those meetups. I wasn’t even doing anything particularly weird or high effort! On the first occasion I spent fifteen bucks on a deck of marked cards and lied twice about the difference between a chi squared test and a t-test.

(I maintain that when I’m not wearing a Might Be Lying hat I’m at least if not more honest than the average person, though probably not the high watermark for honesty and forthrightness among rationalists.)

If you have never gotten straight up lied to - not a misunderstanding, not a subtle Non-Violent Communication use of words, not someone being mistaken, an obvious incorrect thing that there’s no way someone doesn’t know, yes I know this kind of description gets used as merely an intensifier an awful lot - you may not know the slippery feeling it can give and how it throws your previous plans out the window. 

Best get that emotional reaction out in a controlled environment if you can. One of the best things about my favourite martial arts instructor was that he once took the time to listen to my disjointed rants about getting hit, and helped ground me out of that headspace.

Everyone has a plan until they get lied to the face.

IV.

Enough complaining. What should your plan be?

In On Stance, I talk about mental stances. There's a useful reflex in physical activities where, when you're confused or startled, you drop into a practiced stance. My current best mental stance when I realize I think I just got lied to is to start paying close attention to what I've directly observed, and what I've been told by who. My thoughts and notes start using evidentials, a part of speech present in other languages but not present in English which tracks how you came to hold some piece of information.^[1]

(Just as in physical martial arts, false positive errors can be a problem just as false negative errors can be a problem, if a different kind of problem.)

First, slow down big decisions. If you were about to transfer money, or make a public statement, don't. 

Try to de-escalate. Offer a few ways you might have misunderstood, and see it they go for one of them. There's some paths they might take that recover from an honest mistake. Yep, you're giving them the opportunity to lie to you again, but this time you're braced for it and don't need to put trust in that statement. 

When you can, run back through what you think you know about the situation, and how you know what you think you know. When you're counting additional people's claims, think about whether those claims are direct or secondhand and possibly from the same source. Try to untangle what the world looks like if they're telling the truth and what the world looks like if they aren't, including why they might be saying the false thing.

Then move forward. If it turns out they said a false thing, make your new moves in the world where you're not going to be able to trust what they say. Sometimes it's going to make sense to recover that relationship. Sometimes it's not. Try to react proportionately where you can.

I don't think it makes sense to have too much of a plan though, especially the first couple times it happens.

Everyone has a plan until they get lied to the face. It's about knowing you're going to be confused and hurting, and having good habits that will kick in while everything is spinning up again. And I think it might help to say out loud that people can act weird for a bit after getting hit or lied to, a bit disoriented or oddly obsessed with some bit of sense data. You have to get your head back together, and get back into what needs to happen next.

My ideal rationalist community members have this as a practiced skill. They've been lied to, and they're not taken so flatfooted.

1. ^{^}

   It owes some of its origin to to ymeskhout's Miasma Clearing Protocol.