[content warning: will be boring for non-LWers]
Last week was the fifth birthday of Less Wrong. I thought I remembered it was started March 11, but it seems to have been more like March 5. I was going to use that time to talk about it, but now I will just have to talk about it awkwardly five years and one week after it was started.
I wrote a post a while ago called Read History Of Philosophy Backwards. I theorized that as old ways of thinking got replaced by newer ways, eventually people forgot the old ways even existed or were even coherent positions people could hold. So instead of reading Hobbes to tell you that people can form governments for their common advantage – which you already know – read him to tell you that there was a time when no one believed this was true and governments were natural structures ordained by God.
It makes sense that over five hundred years, with births and deaths and so on, people would forget they ever held strange and incomprehensible positions. It’s more surprising that it would happen within the course of a single person’s philosophical development. But this is what I keep hearing from people in the Less Wrong community.
“I re-read the Sequences”, they tell me, “and everything in them seems so obvious. But I have this intense memory of considering them revelatory at the time.”
This is my memory as well. They look like extremely well-written, cleverly presented version of Philosophy 101. And yet I distinctly remember reading them after I had gotten a bachelor’s degree magna cum laude in Philosophy and being shocked and excited by them.
So I thought it would be an interesting project, suitable for the lofty milestone of five years plus one week, to go back and try to figure out how far we have progressed without noticing that we were progressing.
A while ago I wrote about how the idea that beliefs are probabilistic is very not intuitive and something most people never grasp. I don’t mean complicated controversial ideas about how you should be willing to bet on every belief or anything like that. I mean drop-dead basic “You are not 100% certain of everything you believe”.
Were we ever this stupid? Certainly I got in fights about “can you still be an atheist rather than an agnostic if you’re not sure that God doesn’t exist,” and although I took the correct side (yes, you can), it didn’t seem like oh my god you are such an idiot for even considering this an open question HOW DO YOU BELIEVE ANYTHING AT ALL WITH THAT MINDSET. I remember being very impressed by Robert Anton Wilson’s universal doubt, and not as disgusted as I should have been by people making arguments like “If there’s any chance at all a criminal might re-offend, we shouldn’t let them out of jail”. In all of these cases I was sort of groping at the right idea, but I didn’t have a framework for it, couldn’t put exactly what I meant into obviously-correct words.
But that’s Overcoming Bias stuff, Sequence stuff. What have we done on Less Wrong, in the past five years and one week?
It was around the switch to Less Wrong that someone first brought up the word “akrasia” (I think it was me, but I’m not sure). I remember there being a time when I was very confused and scandalized by the idea that people might engage in actions other than those rationally entailed by their beliefs. This seems really silly now, but at the time I remember the response was mostly positive and people upvoted me a lot and said things like “Huh, yeah, I guess people might engage in actions other than those rationally entailed by their beliefs! Weird! We should worry about this more!” For a while, we were really confused about this, and a really popular solution (WHICH I ALWAYS HATED) was to try to imagine the mind as being made up of multiple agents trying to strike a bargain. Like, your conscious mind was an agent, your unconscious mind was an agent, your sex drive was an agent, and so on. Ciphergoth was the first person to help us get out of this by bringing up hyperbolic discounting (there was a time Less Wrong didn’t know about hyperbolic discounting!) After that we started thinking more in terms of non-goal-directed systems. I remember when someone (Richard Kennaway?) first posted a lot of oversold stuff about Perceptual Control Theory, one of which was that it was a system that acted purposefully without modeling a goal, and I – and a lot of other people – commented that that was poppycock and not even possible. Later Anna posted Humans Are Not Automatically Strategic and I posted The Blue-Minimizing Robot, where we both started realizing the importance of non-goal-directed behavior in human affairs. This freed us from having to talk about “bargaining with unconscious agents” and allowed us to become more interested in things like behaviorism, even though I think later we became too willing to accept behaviorist just-so stories.
I remember it was around the beginning of Less Wrong when I first realized decision theory was a thing that existed. This was mostly my personal ignorance, but given that I read nearly every comment on Overcoming Bias and Less Wrong it must have been pretty obscure to the rest of the community as well to escape my attention for so long. I remember making a post that was very much about decision theory even though I didn’t know it, and then getting in a long argument with Eliezer that either of us would now be able to resolve in three seconds by saying “Scott, you’re rounding this off to Counterfactual Mugging; Eliezer, you’re rounding this off to Smoking Lesion, agree on where you’re sticking the locus of decision and the apparent different between your viewpoints will disappear.” But Counterfactual Mugging didn’t even appear as a concept until after this time! Eliezer didn’t discuss Timeless Decision Theory until six months after Less Wrong started, and I’m trying to imagine how our moral discussions must have taken place without it (“imagine Kant, only rigorous”). We didn’t get a really good explanation of different decision theories presented to everyone until three years after LW started.
There was a time before we had Eliezer Yudkowsky Facts, and that time must have been very sad to live in.
There are a couple of terms I coined – not because I’m especially smart, but because I talk to a lot of Less Wrongers and am especially sensitive to when they are using interesting ideas but haven’t yet named them or discussed them with anyone else – that I now can’t imagine living without. The idea of meta-contrarianism – or at least of signaling hierarchies – is one of them, and I find it especially useful in understanding some aspects of social class – and more awareness of the nature of class seems to be something else the community has gotten better at.
Likewise, this idea of trivial inconveniences having outsized effects has been very helpful to me, and got expanded into discussion of trivial fears, the possibly-grammatical-or-not trivial impetuses, the sort of related concept of an ugh field, and most important, this idea of fearing the twinge of starting. All of these make the internal experience of motivation make a lot more sense.
It wasn’t until well into the Less Wrong era that our community started to become aware of the problems with the scientific process. This wasn’t because we were behind the times but because the field was quite new; Ioannides didn’t publish his landmark paper until 2005, and it languished in specialized circles until the Atlantic picked it up in 2010. But as early as December 2009, Allan Crossman working off a comment of Eliezer’s wrote Parapsychology: The Control Group For Science. This pointed out that the frequent ability of parapsychology experiments to discover “telepathy” or “psionics” may not just be about them being especially dishonest, but also demonstrate a fundamental bias of the scientific system towards spurious positive results. The first I heard about Ioannides and medical replicability was a Less Wrong article from Nancy Lebovitz which then got further explored by gwern and Luke. I have since come out against the strongest interpretations of their claims, but it’s hard to imagine that as recently as five to ten years ago no one was really talking about the problems with medical studies or had good evidence anything was wrong with them.
I don’t know if there’s a good history of the efficient charity movement, but it seems to be pretty recent. GiveWell was founded in 2007. Giving What We Can was founded in 2009 by Toby Ord, a frequent commenter on the Less Wrong Sequences. 80000 Hours was founded in 2011 by William MacAskill and Benjamin Todd, both of whom are frequent Less Wrong commenters. I don’t want to overestimate the effect we’ve had on the efficient charity community – all of these people got into LW from EC rather than vice versa and we were not super important influences on any of them. And I don’t think we’ve ever been the super-cutting-edge. But I think we helped spread some ideas and give better philosophical grounding to others. I remember Money – The Unit of Caring and Purchase Fuzzies and Utilons Separately as having a big impact on me. A lot of people have told me that my 2010 LW post explaining efficient charity was important to them, and I’ve gotten requests to republish it in efficient charity books and manifestos. And in terms of just really neat philosophical framework-grounding, my most recent discovery is Katja Grace’s In Praise Of Pretending To Really Try, which neatly solves what was previously an irritating conceptual gap. On a purely practice level, after the past five years of exploring the concept, 30% of Less Wrongers now consider themselves effective altruists and among the lot of them donate over a million dollars to charity per year.
It continues to puzzle me that there was a time when I didn’t know what a Schelling point was. I imagine myself just sort of wandering through life, not having any idea what was going on or why. I’m pretty sure I corrected this well before I joined Less Wrong, but it wasn’t until I pieced together some information from a couple of Vladimir_M comments that I coined the term “Schelling fence”, which was in itself a miniature version of this kind of revelation. I thought for a while people were insufficiently impressed with how important this was, which makes me happy that people are starting to use it more. In terms of barrier-related metaphors, there’s also the Chesterton Fence, which has been around for eighty years but which I almost never hear mentioned outside the LWosphere. I feel like knowing about these two things has improved my ability to intelligently discuss politics dramatically.
A while ago I got an unusually weird complaint about my blog – someone didn’t like that I used the word “meme” a lot when memetics was an “unproven theory”. I objected that it was less of a theory than an extended metaphor. Some of the work done at Less Wrong has either tried to make it more of a theory, or at least extend and flesh out the metaphor. My favorite of these posts is Phil Goetz’ Reason As Memetic Immune Disorder. This seems to have permeated the culture enough that when I refer to my tendency to get freaked out and angry about ideas I myself support as “a memetic autoimmune disorder” everyone seems to understand what I’m talking about.
We’ve also moved in some interesting directions on friendships and relationships. My mind boggles to remember that for several years Less Wrong was not associated with polyamory, and that when I first met Alicorn she was mildly against it. Aside from a few very early adopters like Patri and Tilia, I don’t think it was really talked about in the community until Alicorn wrote Polyhacking in 2011. Two and a half years later, 15% of Less Wrongers consider themselves poly – a small minority, but still way more than the general population – and I think we’ve become pretty good at developing social norms of dealing with that. For example, just two months ago, Brienne discussed her idea of Tell Culture, and I have serious reservations with it that I’ve been meaning to get around to discussing, I agree that this is the sort of direction we should be thinking in and that this is the sort of “invent new and better ways of interacting with people” that makes me excited to be part of a community trying them.
I’ll end with something that recently encouraged me a lot. Sometimes I talk to Will Newsome, or Steve Rayhawk, or Jennifer RM, or people like that in the general category of “we all know they are very smart but they have no ability to communicate their insights to others”. They say inscrutable things, and I nod and pretend to understand because it’s less painful than asking them to explain and sitting through an equally inscrutable explanation. And recently, the things that Will and Steve and Jennifer were saying a couple of years ago have started making perfect sense to me. The things they’re saying now still sound like nonsense, but now I can be optimistic that in a few years I’ll pick up those too.
I find this really exciting. It suggests there’s this path to be progressed down, that intellectual change isn’t just a random walk. Some people are further down the path than I am, and report there are actual places to get to that sound very exciting. And other people are around the same place I am, and still other people are lagging behind me. But when I look back at where we were five years ago, it’s so far back that none of us can even see it anymore, so far back that it’s not until I trawl the archives that realize how many things there used to be that we didn’t know.
I don’t think Less Wrong ever reached the insight-a-minute pace of the Sequences. But it’s been pretty enlightening. And over the course of five years and one week, that has really, really added up.