If your anxiety really is disproportionate, then I'm happy that I could help. But I always try to keep an eye out for people misinterpreting what I'm trying to say to harmful effect, and I want to say that you should really make sure that your anxiety is unwarranted. Sometimes you really should be anxious. My best example is whether or not you should hide that you're afraid when riding in a car with someone. You can make someone a worse driver by distracting them with fearful reactions when your fear is unwarranted, but if they're really a bad driver, and they really do need to know that their driving is reckless, and they're the sort of person who's willing to improve their driving, then you should signal your fear.

Another thing that I'm considering writing on later is that you probably shouldn't think of the relationship between Types 1 and 2 as adversarial, or think of Type 2 as the normatively superior Type. Sometimes Type 2 is the silly one. I find myself considering it more and more plausible that many people receive an understanding of dual-process theories that can actually be harmful.

Thanks. IAWYC, and would even make some additional claims.

I think that you should listen because it's the only way for your brains to become entangled. I think one problem with people trying to play therapist is an inability to consider that an outburst is warranted. If you started out trying to give advice, you might never find out that you never need to give any, that they really are self-regulating affect precisely as they should be.

And to elaborate on what you say about giving them the impression that you're listening, I think this fits well into my more general view of how rational persuasion works. It is true that you probably will have done something that looks like listening if you succeed at persuasion, but too often this is interpreted as, "If I do whatever I think 'listening' is, then I will succeed at persuasion." Ultimately, like other forms of persuasion, rational persuasion is a series of stimuli that you produce in order to sequentially alter someone's attentional biases in precise ways. To describe the effect of what you describe as 'listening', when my friend entered the room, he felt justified in his anger and considered it doubtful that others could provide arguments that would lead him to believe otherwise. You could call this low 'trust of others' and high 'self-trust'. Eventually, I had a lot more information about his mental and emotional state, and initially demonstrated, on my part, low 'self-trust', but after some time, I still wasn't explicitly endorsing his reaction. Because of this, his self-trust lowered and his trust of others rose, and my self-trust rose and my trust of others lowered. Another example is pointing out contradictions caused by fast thinking, like saying that there were two different factors that were responsible for a majority of his negative affect, or that he didn't want to reflect on himself, or that he didn't doubt himself at all. I took him from being focused on a comb, and a low-fidelity model of how other people do inconsiderate things, and how angry he was, and eventually took him to a place where he focused on how people actually think, and what the actual consequences of not having a comb were, and the cost of replacing it, and how other people judged the proportionality of his reaction from the outside.

So, I think 'not listening' is sort of the equivalent of trying to secretly draw someone's attention towards a certain direction, and trying to do this by snapping your fingers out of their hearing range. If you don't 'listen' then you will never be close enough to their current attentional framing to gradually put them in the frame that you want them to be in, where you can just explain things as you understand them in your terms and be done with the whole thing. The concept of inferential distance is also somewhat related to this.

Yes, I agree that this can be a big problem. Right now, I expect to have made a change, but I can also imagine having made no change. I think there are actually observable properties that you can look at and use to predict how effective any non-trivial attempt to change someone's behavior with rational argument is going to be. Framing matters of course: if they aren't an intellectual, then you can't relate to them like one. But for example, high g and high metacognitive ability are indicators in my experience, and my friend has both, even if it's not on the level of many LWers, for example.

But an important thing to note is that I've sort of reinterpreted events to avoid flak from the commentariat. We're both blue-collar workers, so my everyday conversations don't look entirely like this; in fact, there's a lot more aggression involved. So, because my friend prides himself on his intelligence, and knows a little bit about rationality, and believes in the virtue of Hard Work, and I'm the only person in my social circle with a reputation for the ability to consistently generate deep responses and persuade anyone given enough time, I can say things like, "I'm going to keep breaking you down," and it's actually productive. And I really drive it home when he contradicts himself in non-obvious ways, especially when it's something like "You can be as rational as you want and all it will do is frustrate me more. I'm not trying to use my higher-order brain functions right now. I'm trying to vent not to get over it." And he's kind of amused by how irrational he knows he sounds, but keeps going anyway. And I'll say, "Bullshit. You laugh as you speak and you know how ridiculous you sound. You just sat there and talked about how you came up with reasons to get more angry. You're already using your higher-order brain functions. And I bet you're frustrated, but it's not the only thing that rationality is doing." And part of this is indicating that if he doesn't try harder to control his impulses, I won't just throw my hands up and mourn the fact that all was for naught; I'll get angry eventually, and he'll have to deal with that. In the end, the aggression does little harm, and lots of good. I find in lower socioeconomic classes that learning and teaching are more often of the form "Make X feel stupid for believing Y so that they believe Z instead, and do it in a lighthearted but slightly aggressive way." And there are better ways, but it works, and it's what many people know, so it's easier to use it than to change it all at once. I believe that this can be hard for some LWers to accept, because they're not used to being outside of high IQ, rationalist bubbles. In those bubbles, people are typically underconfident about their ability to self-regulate and underconfidence in that ability grants you status. But in my bubble, overconfidence in your ability to self-regulate is the norm, and overconfidence grants you status more often than underconfidence. So you have to act way more confident than you might in your usual epistemic mode, because this is the only way you can relate to someone who's extremely confident about their ability to self-regulate and who looks for confidence rather than a lack of it to evaluate the accuracy of someone's judgment in situations like these. And the overconfidence really isn't all that dangerous for a rationalist, because it's usually really simple stuff that can't result in astronomical failure. And likewise, you have to be willing to get angry, because you're playing chicken in a bubble where most people remove their steering wheels before they even put the key in the ignition.

It's also worth saying that I'm particularly sensitive to affect-laden situations like this because of people I lived with previously, so I also related that, and he is motivated not only by an understanding that he'll make me angry if he doesn't control his impulses enough, but that he'll cause me rather substantial emotional discomfort beyond that. Anyway, ultimately, I think stories of ineffectiveness are examples of people failing the Art, rather than the Art failing people. And obviously you have to pick worthwhile people to use your Art with. Some people really are extremely hard to reach or unreachable.

The article didn't reflect any of this because I was trying to make the content more relatable to my expected audience without offending their sensibilities or including a bunch of caveats that would make the article read in a contrived way.