At least on the Left! I think the Right is a lot more obfuscated.


Let's take a good faith, charitable stance here. Let's steelman playing devil's advocate, and then discuss where it falls short, and how one can advocate for the infernal lord better.

The Phenomenon

I'm having a conversation with Bob about police violence. I say it's bad. He retorts:

Sure, it's definitely bad. But just to play Devil's Advocate, aren't most police pretty good?

A hunger for nuance

My argument is that a lot of people reject what feel like un-nuanced claims that violate their worldview. There are two parts here:

Part 1: Emotional rejection

It's hard for anyone to change their views in such a way that shakes the foundations of their beliefs. They want acknowledgement that they weren't completely wrong, that their ignorance is forgivable, that they're still grounded in a reality that makes sense to them. They want to be seen, respected, and acknowledged as opinion-havers and thinkers. This often comes out as an intense need to be partially disagreeable, so you are forced to concede that they're in part correct too.

How to defuse

If someone seems to be coming from this standpoint, there are two things to do. One, give them space and time. Acknowledge that you hear their point of view then leave them alone for a while. They'll often come around.

Part 2: Nonemotional rejection

This comes from a more meta-level concern: that the argument they're hearing seems to be ignoring a lot of edge cases. ACAB is illegible to a lot of smart people because they don't understand exactly what it means beyond the acronym itself. They don't have the cultural experience to parse it as a nuanced symbol, rather than a blanket statement about people they might know who are cops. This can happen for tons of different causes. Think about all of those (men and women) rallying behind “not all men”. I don't think these people are evil. I think they're just missing the point because they're getting worked up about arguments nobody is making. You could even say they're getting worked up on principle.

How to defuse

Acknowledge the nuance kindly! Show them that there's thought put into these things. That they have depth, that they allow for corner-cases. That complicity can vary, that all cops are no equally bad.

Disclaimer: EXTREMELY rough draft

Disclaimer: this is entirely from personal experience, mostly on Facebook and Twitter. I think the left does a ton of important work, mostly by existing and pushing leftwards. I think this works to fight back against general evil bullshit and right-wing pushing by corporate interests.

Mixed Messaging on Racism

  • “Racist” is used as a perjorative to mark enemies of the left. When someone is being utterly condemned, it is often the word “racist” that is used to drive that point home.


  • The left acknowledges that everyone is racist, in that they implicitly agree to racist systems.

To someone on the left, this obviously involves two different meanings of the word “racist”. But can you imagine untangling this as someone who doesn't have constant exposure to both meanings? There are only a few interpretations that someone could reasonably make a priori:

  1. The left hates everyone, including themselves, and considers nobody redeemable or good. The face value interpretation of “anti-racists” backs this up. Everyone and everything is cancelled.

  2. There's another one but I forget it


  • Lots of leftists abandon anyone who disagrees with them. The most important activism you can do isn't self-flagellation. And I think “education” is a bad way of thinking about it. I'd frankly like to call it mutual aid, in the spirit of leftism: The act of talking to people you disagree with to arrive together at the truth. The goal and hope is that the truth is one espoused by leftist ideology, but one has to be open to that not being the case.


  • This is where I'm gonna put all the normal complaints: weaponization if identity politics, catering to crazies (people with outrageous demands, asking people to dramatically change their life for seemingly tiny reasons, etc). I think the reason this happens is because it's hard to tell what is reasonable.

  • There's a level of catering to marginalized identities that makes a lot of sense to me. And it's a pretty damn high level. I think an interesting thought experiment is what would white people feel comfortable with doing, and how would they feel comfortable being acting when they become a minority. What would they be okay with putting up with? What wouldn't they? Would they have a problem with their culture being appropriated? (lol what culture etc). What about being unfairly targeted by police?

Lack of introspection w/r/t policy

  • Lots of leftist policies fail. I don't think this is a death sentence for leftism, I think it's something that can be actively improved. But it takes acknowledgement. There's not a ton of work done at a systems level to understand why and to make it better. At least not work that's being done in the public sphere, or work that enters public discourse.

Getting off-topic

  • It's hard to stay on-topic! The right is certainly equally bad, if not worse at it. But by debating specific points (“there was an attempt by the wealthy to divide working-class white and black indentured servants and slaves”) you can arrive at factual conclusions because the questions are answerable. Here's a list of debate topics you can bring up with people, for which there are solid answers to:

  • “There was an attempt by the wealthy to divide working-class white and black indentured servants and slaves” (true)

  • “Confederate statues were not built to perpetuate race separatism” (false)

  • “Confederate statues were seen as good by contemporary CSA figures” (false)

By isolating the conversation to a single issue, you can approach truth, which is, in all likelihood, on your side. And even when it isn't, it's a chance to push for improvement in your side.

Mental model: think of your brain activity as being a pie chart. Maybe 25% is devoted to work, 20% to relationships (romantic and platonic), maybe 20% is taken up by depression. Then you have, in this example, 35% left over. So what do you do with it? Here are some examples of how people fill their surplus brainhours:

  • Religion
  • Conspiracy theories
  • Hanging out in the lesswrong-o-sphere
  • Video games
  • Books
  • Partying

I think this model is helpful because it helps me understand why some people get really into religion or conspiracy theories. It's easy to fall down rabbit-holes, alt-right or marxist or flat-earther or anything else, because the brain wants to be doing something, and falling down a rabbit-hole requires almost nothing but a lack of resistance.

I think the cognitive mechanisms that drive people to read a lot, or to participate in virtuous online debate, are truly the same mechanisms that drive people to watch flat-earth youtube videos, or get drunk at parties all the time, or start drug habits. The beginning of addiction can be viewed in this way too.

There's that stat about how revolutions tend to happen in countries with high youth unemployment, and I think surplus brainhours is a useful way to think about why that happens.

This mostly is advice that applies to just me, as someone who has bad ADHD and (maybe unrelated) a hard time making long-term commitments to non-perfect software, but:

I find that when I use the default apps that come with a device, especially default apple apps, I have a much better experience than constantly switching between other apps. There are some notable exceptions, but usually I will accept some minor inconvenience in exchange for knowing I get a free, well-maintained tool with a huge userbase.

Knowing that I just use the default app (unless there's some breaking flaw in it) for a task takes a huge burden off my mind. It means I don't have to anguish over whether any particular trade-off or consequence is worth it or not. It keeps me productive and not-anxious in a world where decision fatigue could destroy me.

It's hard to know how to tackle a new subject you know nothing about. Cooking, gardening, knitting, hiking, drawing, a new programming language, a new spoken language, weight lifting, I could go on forever.

It's especially hard because many of these subjects require a large amount of practice. Indeed, successfully becoming adept at many of these require large amounts of failure and repetition, but it's hard to convince yourself to put in the time and exposure yourself to the amount of failure needed for improvement.

By writing a curriculum for yourself, you acknowledge a minimum amount of time and repetition that you think will be necessary for you to grow in an area. You can set out steps for yourself that guide your learning, steps that are all SMART goals. The important thing is that your curriculum is a dependency graph. You have to finish A in its entirety before starting B, and you have to finish B and C in their entireties before beginning D, and so on. This gives you a few benefits:

  • A concrete feeling of progression (a la achievements in video games)
  • A strong foundation for each new component/skill you learn
  • Enough practice to expose yourself to the nuances and unspeakable truths of a subject (i.e. learned intuition)

  • Use your middle and first initial
  • Use real names that start with the same letters as your real name
  • Use a random word generator
  • Use a quasi-anagram of your real name or parts thereof

Graham argues that by refusing to associate yourself, on a nominal/tribal level, with ideas, politics, and religions, you prevent yourself from suffering when those ideas, politics, and religions are attacked. You can think more clear-headed about them.

This is great armor, but a bad weapon.

Graham talks about ideas, politics, and religions, but doesn't make any statements about less inflammatory categories. Here's a list of examples that don't really fall into Graham's described pattern:

Poet, powerlifter, dancer, scientist.

The thing that differentiates these categories is that they are rarely, if ever attacked in credible ways. Any attacks are frivolous and can be easily dismissed. They wouldn't hit to the core of the people who consider these things part of their identities.

Labels also have power.

Scott Alexander describes the Perceptual Control model of behavior:

If one consider themselves a poet, instead of someone who has an interest in poetry and has written a few, they may, according to this model, find it easier to write more poems in the future, with less friction/required willpower.

I think one of the main components of habituation is what you expect of yourself. By doing something over and over, you expect yourself to do it. You begin to expect yourself to regularly do that task, and accommodate that in many small ways. This seems to heavily overlap with the benefits of considering yourself to be a , whether or not that label is “deserved”.

There's no way for me to conduct a rigorous experiment with myself, and no good way I can think of to double-blind myself. So all I have to go on is the following anecdata:

  • When I stopped considering myself a poet, I stopped writing poetry.
  • When I started thinking of myself as someone who went to the gym a lot, and someone who prioritizes going to the gym over other parts of his life, it made it easier to do just that.

Obviously there's no established causal direction here. It could easily be the case that I felt comfortable picking up these labels because I was embodying them, and dropped them when I felt like I was no longer embodying them. But my intuition seems to think otherwise: that there's a benefit to picking up a label aspirationally.

Unrelated to the main premise of this article, there's another benefit to picking up a label: access to resources.

By establishing yourself as a member of a community, you gain access to their social and material resources. You gain a feeling of acceptance. Note that these benefits are more closely tied to the pitfalls Graham warns about. But I think they're valuable and worth pursuing nonetheless. Just because there are reasons to keep your identity small doesn't mean that there aren't other reasons to add labels onto it.

epistemic status: 70%

I don't think Facebook killed our trust in media. I think it was Google. We have far less faith in institutional knowledge and the knowledge of our parents and other authority figures in our lives. Because why trust when you can find definitive answers online? Why have community when you can have answers? I think the previously high value of institutional knowledge bound groups of people together, and that now institutional knowledge is less valuable, so are communities. I think this is why hyper-specific niches continue to thrive, as well, since it's harder to find information on the respective subjects online.

The more people post in public, search-indexed forums, the faster communities die off.

I. (it's) nice to have you (here) (it's) nice to see you, too (one had) better be going

II. (Can I) give you a lift? (Do you) care if I join you? (Do you) know what I'm saying? (do) (you) want to make something of it?

III. (if you) sing before breakfast, (you'll) cry before night (someone) is not going to thank you for (something) (you) can't take it with you (when you go) (you) do the math Am I glad to see you! an atmosphere that you could cut with a knife and the horse you came in on and what have you Are you a man or a mouse? are you deaf? Are you feeling better? as fast as (one's) legs can carry (one) as far as (one) knows arouse (one) from as as you please as you do as best you can Are you threatening me? Are you religious? allow me an offer (one) can't refuse #poetry