When I talk to people about motte-and-bailey arguments, however, I start to feel worried about the entire culture the idea tends to incentivize. There seems to be a background assumption that you should “stay in the motte”, i.e., stay in the space of ideas you feel “safe” arguing.

This isn’t intrinsic to the idea. In fact, the main point of the idea is just about conflation: there are ideas that are harder and easier to defend and people to launder the harder to defend ideas by associating them with the easier to defend ideas. This doesn’t make either of the ideas “bad” or “wrong” it just means that they’re not the same.

There’s a similar sort of thing going on in Tal Yarkoni’s The Generalizability Crisis, which suggests that the secret sauce of Psychology’s magic (read: fake) ability to prove really complex hypotheses relies on seductive linguistic extensions of what has been show by the evidence. In Psychology it is very easy to tell “just so” stories that are very convincing because of the way we are constantly peddling in narrativizations of the human mind to understand each other in everyday life. But this problem of what experiments actually mean is really just all over the place.

Statistics won’t save you from it. Statistics only work when you’re underlying conceptual model of the situation isn’t already so fundamentally confused that you can catch yourself making a mistake. If you think storks deliver babies, you usually have to rely on simpler ideas like “have I ever seen a single stork do this?” to check your theories. If you already have the wrong framework, you will misunderstand what statistical tests are telling you or apply the wrong one.

The fundamental problem we have is that the world is really complicated, and we don’t actually have great tools to understand what’s going on at a fine enough level to know how to check if we’re wrong. That’s why, despite all the howling about “the scientific method” the real test of results in a given field is whether they get used by someone who’s not in that field. I am eternally suspicious that everything that has not been exported from a field yet is just a result of the culture of that field. You have to continually lose money by betting against a hypothesis for it to be real. And even then, it might be real because of memetics, the explanation is always suspect.

Then how can we stay in the motte?

That’s exactly the thing: the linguistic tool “the motte-and-bailey fallacy” is meant to safeguard you from over-extending, not stop you from extending at all. There are always claims we don’t feel comfortable defending in adversarial circumstances, but we need to rely on them as we’re thinking about how to gather more evidence. And the frontiers of knowledge are mostly made out of these claims, scaffolded by the extensions of claims we have strong evidence for in more well-understood circumstances.

Which is just to say something very simple: if you stay in the motte you will starve to death.

The bailey is where all the food comes from. It’s where you import aspects of the great unknown and then attempt to process them so they can become parts of your stronghold. If you’re under attack, then you can retreat the motte and say “all we’re claiming definitively is this bit, the rest is provisional” and you would be right and honest. This isn’t the motte-and-bailey fallacy, it’s just a normal part of the scientific method.

If you stay in the defendable zone you end-up fighting for scraps with everyone else who was too scared to be called out as wrong. But that’s the funny thing, when there’s enough competition (and the internet means there is for most knowledge pursuits), there is almost no alpha in trying to discover something in the defensible zone—because someone was going to discover it a moment after you anyway.

The overwhelming majority of value in contemporary knowledge pursuits isn’t in meticulously combing for missing gems in the land everyone wants to mine, it’s in coming up with left field hypotheses that are easy enough to test that it’s worth the risk.

If you stay in the castle looking for the treasures that haven’t already been looted, you will starve to death, because the economic activity producing food for thought is all happening outside—and you have nothing left to trade.

Even though a great deal of it had one or two interesting ideas, the books didn’t seem to take themselves seriously. It was as if the authors had said “I want to write a book where laser guns make the Government lose lol” and then proceeded to open up WordStar.

So I complained to my dad and he told me about “hard” scifi. I tried it out. Some of it was better, but mostly it was the same: the increased realism usually focused on why certain elements of the premise and technology make sense, rationalized one or two inferences about the dynamics being presented, and then largely slumped into the classic scifi mashup of “this is what I want out of tech” and “this is what people have always and will always be like”.

I’ve specifically asked people for recommendations of scifi that showcases how people would feel and act different in these imagined worlds. Most of the time I’ve ended up feeling like these recommendations, while explicitly focusing on key differences are usually:

(a) so focused on showing differences that they portray human action incredibly unrealistically (1984),

(b) so naïve in the way they setup new dynamics like alien cultures that they basically feel like toddler toys for hypotheticals (The Mote in God’s Eye),

and

(c) so ridiculously unable to jump-out from the zeitgeist that they tend to try to defend certain aspects of society, especially morality, as “universal” (Ancillary Justice).

The good ones only have two of these problems at a time.

In fact, I’m comfortable saying that as a means of exploring the possible, scifi is a failed experiment. It came from a certain variety of futurism that wanted to imagine, and is now held together by style and theme like all genreficton. It can barely look down on steampunk. Jorge Luis Borges beat scifi out of the water.

Yeah, yeah, yeah—I know that a few people were special. Asimov had some good points, I admit it. But I honestly couldn’t get through Foundations, because (again) I can’t help feeling that Asimov couldn’t take his own idea of “psychohistory” seriously, and basically just kept writing ideas in that ended-up being something like “yeah, but what if we can compensate for emergent effects lol”.

I think the failure of scifi leaves a small but deep market niche open: the niche of Tangible Fiction.

Today, people are constantly asking “What if?” questions on social media, but the follow-up discussions are often social. The idea of “what if X could be different? what if could design it to be different? but what consequences would that have?” is so normal to think about that it forces us to consider: why aren’t we writing more fiction like that today?

I think one of the big reasons is because it’s more dangerous than it used to be. Calling something normal or abnormal, suggesting ways that people would organize that frames certain kinds of political organizations well or badly, or really just fantasizing about anything that doesn’t explicitly evoke “a moral and beautiful future for all” or “the inevitable dystopia” are easy to attack in the increasingly flat public communication landscape.

But that only makes the mouth of this market niche wider:

Right now, fiction tends to avoid dynamics, especially if it’s not trying to get you to accept the world, as my friend Natural Hazard pointed out earlier today when writing about The Magicians:

Now, the show hasn't really given enough details of the world or the magic system for us, the viewers, to make reasonable claims about whether or not saving the ghost-kids is possible in-universe. There have been other moments in the show were a similarly unexplained-to-the-viewers problem was presented, declared impossible, and then resolved. But there have also been many points where there seemed to be agreement about certain fundamental limits of magic. Point being, we don't have any of the information we'd actually need to make a judgment about whether Alice or Elliot is right here.

Most contemporary fiction I see today tends to pit ideological or perspective differences against each other, where the actual dynamics of the world are too murky for us to make pragmatic distinctions about their appropriateness. This could be viewed as “idealized circumstances” but I just view it as lame. Real situations are full of details about the dynamics that become load-bearing as soon as they can be used for coordination—what Thomas Schelling calls “incidental details” in The Strategy of Conflict, a foundation text for the birth of a new genre that takes the underlying dynamics of imagined worlds seriously.

I propose we call it “Tangible Fiction”.

What I’m describing could easily be called “thought experiments”, but thought experiments have a deservedly bad rap: they tend to try to evoke the impossible by avoiding the question of construction, steady-states, etc.

What I want, and what I’m guessing a lot of other people would enjoy, is the exact opposite: I want fiction that takes the “What if?” seriously and fleshes out the phenomenology and social ecology of the situations it constructs. Tangible Fiction gets a free pass for a bunch of stuff in the premise, because you don’t have to know how you got there necessarily to take the rest seriously. It is usually the exact reverse in Hard SciFi: lots of explanation of why FTL is possible and then flat characters that use it in melodramatic ways, instead of getting rich by becoming the Intergalactic Information Suez Canal.

Tangible Fiction is a call for us to collectively start thinking and socializing around the actual ways that history feels different when you’re living in it, and the fact that these feelings actually result in different behavior. COVID-19 showed everyone how wonky our notion of reality could get, and how weak our narrativization skills had gotten by not experiencing such disruptions, especially when it comes to meaningfully communicating about narrative conflict.

What if we used the current cultural moment we’ve been handed to transform fiction into the ultimate, socially-distributed “What If” engine? What would that be like?

We do not all feel the same level of hotness—some of us would rather it was hotter and some would rather it was colder. But it is hotter than the air-conditioned building we left and it is one of the hottest nights in our little town.

Indeed, we have no really good way of talking about “how hot” one is feeling. We talk about these things in terms of function: I exclaim “It’s so hot you don’t need a jacket!”

Anne and Beth are wearing jackets.

Anne corrects me: “I still need a jacket, Crispy.”

“Well, of course you do, Anne. You’re always cold.” I reply.

I have factored-in Anne’s eternal coldness—the fact that it doesn’t change under these circumstances doesn’t really mean much to me. There are no jacketless nights for Anne in our town, so I don’t really think about her as playing the “Is it a jacketless night?” game.

Beth, on the other hand, just smiles. She doesn’t feel a need to correct me, because she knows how I would respond, and doesn’t feel like it would lead to a conversation of any importance.

Duke says, “Why do you always say it’s a jacketless night, when Anne’s always going to wear a jacket?” not because he doesn’t know why I said it, but because he’s indicating that I might as well play a game where everyone present is a participant. He’s indicating this without having to put me on the spot quite so hard, but enough that I feel a bit silly. This is “social pressure” and it’s a tool people use to show the kinds of games they want to play without insisting on them.

When I exclaim that you don’t need a jacket, I’m really making a point about how hot it is—one that degrees don’t really capture. This is why weather reports often have a “feels like” estimate, because the precise temperature is known to be a poor predictor of the actual experience of heat. Funnily, the “feels like” is also expressed in degrees. The idea, supposedly, is that this is being expressed in ideal conditions, e.g., without wind.

Even if the “feels like” estimate is accurate, people adjust differently to the same number. But our internal experience of the heat is mediated by how we react to it, which is why the first thing that jumped to my mind was the jacketlessness of the night. This observation was more than just personal—everyone felt the heat, because it was a relative change, but I wanted to express exactly what it meant to me and then see if I could socialize around the finer points of it, to see what kind of overlap with other sharing my experience of the world would lead to.

The resulting conversation ended-up being less about heat, and more about the meta-game of the claims I make and what social dynamics they create. I find this is often the case. That arguments about how to cut a cake, often have less to do with the cake, except as a lead-in. “I’ll cut the cake, but you can choose which piece.”

Relative semantics, like the increase in heat, are a vehicle for agreement in socialization. There are lots of relative differences we can agree on, where it’s difficult or impossible to characterize an absolute, e.g., how angry someone who left the party was.

Absolute semantics are only really possible once we reduce something down to observables. This is possible with lots of non-social stuff, because the context can be pinned down much easier. But even very simple social effects, like how spicy something is, are basically impossible to measure because the internal experience of it mediates the outward effect so much. We can measure someone’s biological reaction to capsaicin, but we can’t characterize what feelings they’re interpreting it as, we can’t use that to predict how spicy they like their food, and we can’t know how much they suffer spiciness out of machismo.

An absolute semantics of social effects is impossible with current tools, because most social effects are really about taking one path from a garden of forking paths, and what tipped things towards one option or another. Social games are reflexive, and what the participants think about each other matters a lot. For any fairly realistic social game, this information can’t be elicited. There is an absolute social calculus that could allow us to predict a person’s every actions, but short of simulation, our ability to elicit it will be very coarse for a very long time.

But we build up our models of other people by the information we get from relative semantics, and by using the “typical mind fallacy” judiciously. The typical mind fallacy is only a fallacy because a lot of suckers took it too far, and now we have to constantly argue about how far to take it instead of all the good stuff.

The good stuff is stuff like this: I can predict that any meme that’s used authentically when it’s getting popular will flip into being mostly sarcastic, and that the precise opposite will happen to initially sarcastic memes. It’s just a matter of pricing: when there are two natural interpretations and you dilute the value of one, the other is implicitly subsidized.

More on memetic subsidies soon.

Consider, this excerpt from the top conversation on letter.wiki: “Are The Methods Used By Science The Only Ways Of Knowing?”:

Which brings us to science. Some of the knowledge above isn’t seen as scientific knowledge because it’s not determined by scientists. But my definition of “the methods of science” is broad. I don’t mean institutionalized science of the sort I practiced, involving funded research, laboratories, and published papers. Instead, I see science as a set of tools, forged by experience, that help us find truth.

While there’s no one “scientific method”, those tools are familiar: observing reality, checking observations with other people, testing what your ideas predict, resisting your biases, and so on.  Observations generally segue into knowledge when repeatedly confirmed, adding more and more credibility to a hypothesis. Finally, empirical knowledge is always provisional, though of course some knowledge, like the formula for water being H2O, is unlikely to ever change. Still, nothing, including science, can tell us when we’ve attained the absolute and unchangeable truth.

The methods of science, then, simply involve applying the philosophy of naturalism, whose tenets, according to physicist Sean Carroll, are these: “There is only one world, the natural world; the world evolves according to unbroken patterns, the laws of nature; and the only reliable way of learning about the world is by observing it.”

This is not a process or a method, it is a description of an idea’s place in culture. This is the way that the word “science” has become a conceptual tool for understanding the “ground” of knowledge legitimization: if you can make the case that what you’re doing is science, then it can be said to produce real knowledge.

Interestingly, scientists themselves have gotten into a pickle of their own: in the current climate it is very difficult for anyone except an extremely famous and rhetorically clever scientist (i.e. a politician in a lab coat) to propose an idea. Instead scientists have to propose hypotheses about specific literal objects we currently have names for, or else they’re not being rigorous.

This is unfortunate, when one considers the fact that many of the ideas we consider really important in science were either vague, impossible to specify, or wrong in the given context when they were first proposed. Darwin’s On the Origin of Species does not refer to any experimental evidence (and the inquisitive reader is invited to consider whether experiments in evolutionary science are possible in the same way they are in physics), but rather refers to certain well-known (at the time) facts about the world and suggests an underlying idea that would allow us to thread them together in time. Mendel may have lied about his experiments (or not—I really couldn’t care less at this point), but surely thinking about Mendelian genetics really has nothing to do with that?

What we get from these scientists are metaphors—ways of thinking about a process that really exists as if it was a process one actually has the words to describe. It is not an analogy, because we treat these models as if they were the truth, until they are falsified and we make a new metaphor with which to see them. I use the word “metaphor” instead of “model” to emphasize that when we talk about the changing of species over time, it may be the case that we can’t even really say that species are discrete, but our perception of reality is extruded by the idea that a “species” evolves due to natural selection and that is the reality we are capable of witnessing in the abstract. For most people on most subjects, we cannot actually go and look at primary evidence, so these metaphors are literally the only way we are capable of having an acquaintance with the idea of a “species over time”. For investigative spirits who try to figure out more about “what’s really going on” we cannot keep all the evidence in mind at once, so we rely on a grab bag of metaphors to perceive a specific bit of evidence until we come up with an alternative, usually inspired by local observations that allow us to change the old metaphor or make a new analogy with something else we think we know about.

These metaphors are not hypotheses—they cannot be proven wrong in any meaningful sense because they float at too high a level of abstraction. Rather, they are a shared narrative that lead to mutually intelligible hypotheses that can then be tested. It is these shared narratives that underpin scientific communities, because without them we could not agree even on what evidence might mean. We would be stuck saying, “Sure, the light from that laser was reflected off of a massive object, but I don’t know what you mean precisely by ‘moon’.”

Currently, though, most scientists cannot propose new metaphors in scientific writing. So the famous ones do so in talks and blogs, and others find the closest thing they can that someone else wrote and attempt to tweak it into what they came up with. Of course, this disincentivizes spending too much time thinking about these conceptual tools in the first place. This is a very sad situation, but it is downright dangerous when it comes to inherently complex and fluid concepts like culture and language. We may have made GPT-3, but how do we plan to study it, if only Dawkins can coin the term “meme”? Imagine how many linguistic patterns that GPT-3 is picking up on that people are having trouble publishing about because their descriptions can’t be checked by another machine.

To move forward, we need to give-up on statistically verifiable hypotheses as the only means of knowledge legitimation—which means finding a way to make discourse where we can try to cooperate around what’s useful instead of “defending” our theses.

I am an academic in Computer Science, and I hope it will be a surprise to no one when I say that the vast majority of knowledge in papers depends on knowledge nowhere to be found in papers. I would say that across computer science this is less true than in many other fields, because many different communities require direct reproducibility of results (within some tolerance for different conditions) via code released by the authors of a paper. This is not nearly as much of a guarantee as it would sound like for three reasons:

(1) Even if you can run the code and get similar results, you don’t know what the code is actually doing, and often neither do the authors.

(2) What the evaluation metrics / scenario which you are evaluating X thing on actually mean in terms of, say, the efficiency difference between two methods is often not obvious unless you’re in the field. Benchmarks have to make simplifying assumptions to serve as barometers of some inherent property people care about which is probably not just one property and even more probably not well defined. When benchmarks are introduced they make some of these clear, completely refuse to mention most of them, and the authors are often unaware of the most important factors that inherently make a benchmark narrow.

(3) Even if the code works and you understand the benchmark, you usually don’t understand how the authors came up with the specific settings with which they used the code given the specific things they tested on. A lot of the time it turns out they tried a lot of things before they tried the thing they showed you, and they don’t like to talk about how hard it was to find them, how badly it worked in so many cases, etc.

So, when you come to get a PhD what do you spend most of your time doing? You spend it getting acculturated. Everyone talks about this, but they keep it vague because if you look at it too hard it’s easy to get really cynical. The place it comes out the nicest (and most presentably) is in group meetings when people discuss papers: a first year will ask a question that makes perfect sense if you believe papers at face value, but no one would even think to ask if they knew how the sausage is mad. For example, a first year might ask “Why didn’t they try method X on problem Y?” but everyone knows that all the benchmarks for Y are defunct and don’t really test Y, so the authors had to write their way around this issue.

The process of acculturation to academia is the process of learning how to see and use the intellectual fascia that hold a community together. What are all the different properties that are assumed about all the different objects? And what about their different relationships? What do people mean when they say certain kinds of things about them? What are places where people have a different “tint” to their statements about different sets of things and how can you compensate for that “tint”?

The process of acculturation never stops, either. One of the reasons it is impossible to write good textbooks that are up to date with modern practice, is that modern practice is just a bunch of fashionable methods and problems without hindsight to tell you what stuck. Most textbooks are mostly justifying what stuck under a narrativization that feels satisfying. If you tried actually thinking about a method of doing history in any kind of verifiable way for these fields, you would look at the papers published around the times that the textbooks are talking about and feel very, very uncomfortable. People who do look at the history of disciplines are aware of how messy it is, but for most of them they are rewarded—both intrinsically and extrinsically—to come up with their own pet theory, one that can be compressed down to an elevator pitch.

Even though people stop seeing some of the basic levels of this intellectual fascia, all serious researchers are aware of the layers that are still gooey enough to be subject to change. Most of us are gossiping about it, griping out this paper’s influence, or wondering whether X will take off to our colleagues. This is not idle chit-chat—this is the internal network of citations and relevance. In my subfield, papers are published extremely often. There are 2 top tier conferences a year in my subfield, plenty of highly regarded related conferences and second-tier conferences to go to in a pinch, but even this cannot keep up with our output. The main place people put their work first is Arxiv. When I first came to grad school, I would try to read the abstract of every paper that was posted on Arxiv in my subfield that day—giving a quick skim and putting aside anything that looked interesting. It would generally take me around an hour, but it was a great help in understanding the space. Today this would easily fill up my work day, and probably push me to find time in the evening for more.

So how do people manage understanding the space of the field in practice? There are specially designed tools, like Arxiv Sanity Preserver, which do get a decent amount of traffic. But in practice the answer is social—the first line of reconnaissance is Academic Twitter, with the hope that really important things will bubble up as your colleagues discover them and post about them.

But all the truly public spaces like these are just lettuce on your hamburger—they don’t give you even one of the five to nine servings of papers and github repositories per day that the USDA (United States Department of Academia) recommends. Instead you have your lab meetings, where people present papers that maybe are only a few weeks behind, but perhaps the senior students and professors will name drop the good stuff that’s just coming out. Then you have your slack channels—hopefully you’re part of a few good ones where people will link drop Arxiv papers they just happened to see mentioned in some search result or by a friend of theirs. This might get you up to three servings of contemporary knowledge, but the real line is direct people.

If you are trying to “keep in the know” then you talk to people and you talk to people a lot. Talking to people who know people who know people who know a paper just came out is how people know about papers. But knowing about papers isn’t enough: it’s the changing configuration of the fascia that matter. When you have a good relationship with someone in academia, you begin to talk about what does and doesn’t work “in practice”—i.e. who was really lying about what, and even more important who knows they’re lying and how much they know they’re lying, because not everyone is equally good at understanding what’s really going on. A lot of the laundering of false information is just a little bit of motivated reasoning by someone who would have to accept some crucial facts about the current situation of the field to even begin doing the kind of verification required to get more reliable results.

The most important thing, by far, is discussing what property of the current situation allowed X to make Y claim. Sometimes the answer is technical: they took advantage of some condition—whether it be a new technique, a specific aspect of a problem, or a defect of a benchmark. Sometimes the answer is social: X was able to claim Y with less evidence today than would have been required yesterday, because result Z makes Y seem almost like a forgone conclusion. But all of these little bits of info and developments add up to something more: the way the field as a whole is thinking. It does not think in one direction, but instead is lots of little subnetworks of the papers and results and code and people that people “see” as grouping together, and thus coordinate around. It is verbal citations that actually constitute motor force in moving the field forward, as they are the ones that rewire how people in the conversation think about their situation and thus their next experiment.

These are the citations and interpretations that matter most, the ones that happen behind closed doors. They are subject to just as much strategic interaction as the ones that happen in public, but their audience is not diffuse—it is another individual, who usually has some actual research interest regardless of what kind of politicking they may also be involved in. If you are lucky one or more of your colleagues could be a “write ’em up” kind of person—they will write brief PDFs that clean-up some of these thoughts and these PDFs will be handed around like gold, because they are the most reusable currency of the backchannel. The PDFs, however, are a very dusty mirror, because they must be presentable to the public when inevitably they are passed around more than the original author intended.

Yet it is exactly because of the methods of this backchannel that it is ephemeral—and my suspicion is that rereading old research papers is so alienating exactly because the flesh has decayed away and all we have left is the fossil.

If you didn’t, stop reading this.

There’s not a lot to explain—I’ve been a bit busy at work, a bit overwhelmed by people who needed my help, and then suddenly gripped by ennui and the need to do nothing. After fighting it initially, I remembered what had helped me deal with it every other time: giving in and letting my mind be. So I did, and I already feel a lot better, but I still need some time.

I may start writing again soon, but I’m not making any promises. I’ve been breaking a lot of those lately and I need to pace myself. Definitely feel free to unsubscribe if having your name on this list bothers you; I do plan to get back going with Attack on Titan, essays, and maybe even a fiction piece I’m working on but, again, no promises about timing.

In the meantime, if you need to reach me you can hit me up at whyamicrispy@gmail.com

Be well, my friends.

In some sense, the title of this episode “To You, 2,000 Years in the Future - The Fall of Zhiganshina (1)” explains the lion’s share of the plot. This will be a history of what happened, and what is happening now is the return of some ungodly monsters.

This is the “getting acquainted” episode, so let’s list what we got acquainted with:

Titans—big muscley babies

that apparently like eating people.

3D Maneuver Gear—

a very fancy way of getting around when you’re fighting big babies.

Given how complex these things are, we expect them to be important strategic tools.

Eren Jaeger—A heard headed curious young man, who won’t be told what to do!

He wants to join the Survey Corps, which explores the outside, but has apparently met with little but defeat.

He’s the son of Karla Jaeger (who was giving him a concerned look two images above) and Doctor Jaeger

who’s apparently a pretty big deal. Saved the entire city from a pandemic, and maybe has a cool sex dungeon?

Who knows!

Don’t worry, Eren’s not always that 2D, but you’ve got to get the basics down, right?

Mikasa—Sibling (step-sibling? something is off) of Eren. I mean it with 100% sincerity that thing I most empathize with in the 1st episode is the dead look in Mikasa’s eyes when she has to deal with Eren’s shit.

She does all the work and keeps things in line.

Okay so, the walls haven’t been breached for a hundred years—but what happened to them? Were the Titan’s organizing? Why did they stop if so? Something is weird.

The walls that the kingdom, and more locally Zhiganshina safe are really big and sturdy. They’ve even driven a religious kind of worship:

But apparently humanity does know squat about Titans. Nor are the guards involved in frequent enough conflict to actually be ready.

Uh-oh.

How did the walls get built, then? Are the Titans a recent phenomena? Did technology used to be better? There’s a mismatch in technologies, and the government has declared interest in the outside is taboo, but is funding a very expensive branch of the military to go out and look for stuff, so something must be going on.

For me the biggest disconnect is Eren’s non-stop comparison of humanity to “cattle”. Sure, he’s an adventurous young man who wants to explore, but humanity’s situation isn’t really that cattle like? Can you imagine how scary it would be if cattle (or maybe more accurately hamsters) built up fortresses that were significantly taller than most humans, couldn’t be destroyed, and created a civilization within?

Humanity might have seen better days, but there’s clearly some sense in which they have the upper-hand, and yet the Survey Corps at least claim to know nothing. How did this state of affairs come to pass? And why is there suddenly one really, really big Titan that can knock down the walls? And why now?

We don’t see any answers, and the questions are still nebulous, but AoT begins to show what it’s really all about with the final scene of the first episode, Karla is stuck under the Jaegers’ now collapsed house and she’s no fool. She knows the Titans are here, her legs are crushed, and there’s only one chance for her children’s escape: for her to be left behind.

That’s when our boy Hannes, the unprepared soldier, is going to take down the monster and save everybody:

Then he sees it:

And then he knows:

See, Hannes isn’t an idiot. He admits right from the start that they’re not prepared for a real breach. The only reason he tried in the first place is because of the debt he owes, but he’s aware of the scale of things.

Organization is all about the scales of things. The scale of conflict, as determined by supply and demand of resources needed for people to achieve power and manipulate their environment. There are two true heroes in this episode, and Hannes is the second one, because when faced with a problem he can’t solve, he decides to find a different problem, that overlaps with the first. It’s easy to write about never giving up, and it’s very hard to give up a little bit for the sake of everybody when you’ve been telling yourself stories like that. In the back of his mind, Hannes knew this was a possibility, because all of his training was based around events he had no direct acquaintance with. The spirit of humanity is to solve the problems you can till you get to solve the problems you want, and keep a pocket in your heart for the unrecoverable casualties that occur along the way.

The first hero? That would be Karla.

There’s nothing sexy about laying down and dying, but Karla is so on top of her shit that she knows she can’t stop herself from screaming out and covers her own mouth:

That is, perhaps, the best metaphor for the resource allocation game humans are automatically entered into: we can’t control ourselves completely, but we must use the parts of ourselves we can control to deal with our environment and ourselves.

Things only get stickier when that environment has other people in it who are trying to do the same thing.

I am writing this series because my passion in life is to try to describe the patterns and principles of human communication, to show how it is an ecosystem of sublimely efficient and subtle microeconomic organization. I call it Coalition Companion because it is a meditation on how coalitions are born and evolve. Is that too many words? I really just mean to say that people organize in really interesting ways, and while lots of fields of study take bites out of the whole—anthropology, sociology, managerial studies, psychology, cognitive science, and linguistics—I can’t help but feel that the divisions between these fields and their foundational assumptions have become more distracting than the simple question: When I look at some humans, what is going on?

These essays will not be complete play-by-plays. The intention is for you to watch an episode, then read the accompanying essay. I will link to the crunchyroll page for each episode to streamline this, while also including pertinent screenshots as anchor points.

There’s a lot in AoT, and I won’t be able to cover all the themes. One strain of things I want to think about more, but will likely write little about, is how Attack on Titan reflects impressions of war in the modern era deeply influenced by WWII, the atomic bomb, and modern political structure. Exploring this aspect of the series deserves a book of its own, and I’ll mention it at times, but the focus here will be the micropolitical way people form groups in dynamic situations and how that influences and is influenced by everything else.

Indeed, I think of AoT as the best visual primer of micropolitical organization I have yet found. I haven’t read the manga yet, but I’d like to take a moment of thanks for 諫山創 (Hajime Isayama), its creator, because it has profoundly enriched my ideas and given me source material to discuss things with examples. That is the only way to discuss things, really.

Updates will be approximately weekly, with one essay for each episode. At some point this will speed-up a little bit, because I would like to cover all the current episodes, before the final episodes arrive in the U.S. in early-mid 2022.

(If you’d rather not read these, but still want to see my other essays, simply filter the term “Attack on Titan” from emails sent with this sender, I’ll be careful to not mention it or use a euphemism in other posts.)

For those of you who will be reading Coalition Companion, welcome. I’m brimming with excitement about all there is to dig into.

This is the first installation of my podcast Words with the Bird where we casually talk about what the hell communication is, how it happens, and anything else since it’s all communication.

This is my first discussion (ever!) with Mycelium Mage.

We talk vibing, formalization cages, the space between people, what it means to be “stuck” in a communicative space, and how being overwhelmed by plans is often just the desire to be alive right now.

These are the essays being passed around in the underground, and they can be passed around in public, because you have the comfort of being able to read them in your private inbox. These are the essays by which we coordinate around, because our friends will have read them, and those who have read them are people we’ll consider as potential friends.

I think that’s the secret sauce about the Lipstick Effect: when times are hard, we put more stake in ourselves. We want to invest in showing our potential and in creating a community of mutually committed people, because in some sense that’s the only way to make anything matter and the only way to feel safe. In good times, we want this, but we’re a little more lax about making sure it happens now, about seeking out opportunities for it at every corner.

Why are you here? I’d bet that a decade ago the level of confidence you had that the way you express yourself today would still be as generally acceptable a year later was at least three times higher than today. The strategy to deal with this cultural uncertainty is to build a new base of mutually legitimizing cultural institutions. As an isolated individual you might not know how to do that, but you want to be part of the solution, so you lend your patronage to people who are becoming beacons of legitimization to cultural regimes you can believe in. You do this for three reasons:

1. You want to help out the people who you think are in a unique position to do something you can’t, even if you’re helping in a different way.

2. You serve as the grounding for legitimization, because ultimately legitimacy comes from someone’s ability to associate and speak how they want and still feed themselves.

3. You want to be in on the fun part of this legitimization enterprise, and ultimately shared media is how we have fun together: we need the mutual experiences to talk to each other about.

You are also likely at home, bored, and lonely, while a helluva a lot seems to be happening to the world, to policy, and to the ideas in the public discourse. Maybe you want to talk about it or maybe you don’t, but you want to talk. You feel depressed and scared and you’re not even trying to feel less depressed or scared but trying to do something to make a system go. It’s not even clear what the exact problem is or what the system you’re helping run, but it’s a people problem, and the solution to people problems is generally getting a bunch of people together, talking in a goodnatured and fun way, and then facing the world together.

Woohoo! Let’s go kill a god with the power of friendship.

What are you doing on Substack in particular though? You’re reading your favorite writers!

But why are they here?

Substack isn’t anything you can’t more-or-less setup on Patreon or by yourself. The point is that it’s:

1. Easy to setup.

2. Free.

3. Is streamlined to disseminate information directly to people, but make it publicly linkable.

4. Is an ecosystem that is attempting to become a bastion against the fall of liberalism on the internet.

(4) is why a lot of the big shot writers are here, and those are why other big shots came here even if they weren’t worried about being thrown out of where they already were.

(1), (2), and (3) are quite important too—they’re why a certain kind of person is here. We don’t see too many people commissioning hentai through Substack, because why bother forcing that into a platform like this? Anyway, Substack probably wouldn’t be super interested in defending your right to do that if you ended up stepping on a mine.

Ultimately, these writers have to make money. Paid Substack is essentially a freemium model: pay more, get more from this thing you already do for free…except in the case of Substack more means “more writing” which is quite a tricky proposition, for at least two reasons:

1. As reader, if a piece is really that good, everyone is going to be forwarding it to everyone and you’ll end up bumping into it (perhaps guiltily) before you even hit “subscribe”.

2. As a writer, it would be crazy to hide your best work behind a paywall. It used to be that you could be coaxed into doing so for an important enough publication, because the exposure and credentials mattered That Much. There’s a bit of prestige to being good enough that people will pay for more, but that’s peanuts compared to being better known.

When faced with the paradox of why people pay money for Substack I start making arguments about convenience, loyalty, etc. But then I start thinking about what Substack actually is.

Substack is really just a way for you to subsidize the new information ecosystem, and that’s why a lot of us here—but the freemium metaphor is still dissonant. The conclusion is simple: premium content shouldn’t have anything to do with disseminating information, premium content should be about being part of the club that legitimizes a certain direction, a lens, a set of ideas, a way of doing something, a person who you believe can do great things. Premium is about being part of the club, because the old narratives have collapsed and it’s time to cast your vote for what the narrative is—and this time we’re using real money.

When I think about the Patreon creators I’ve supported over the years, the main things that I received for my patronage were little trinkets that I could use for some added style or to brag to my friends about how cool the people I supported were. There was some custom content, though I wasn’t too interested in it and it was often not as good as the base-level or publicly available stuff. I think a lot of people like getting to hear or see or read a given creator’s take on X, so why not? It seems like it was pretty low-friction for most artists. And then, of course, there were sometimes calls or zooms with the creators themselves, because that’s the ultimate club: association through interaction.

My prediction is that all paid content on Substack will asymptotically approach exactly this space of things. This is good, because the purpose of Substack is to create the essays that will be passed around in a million different undergrounds, a public ledger for private coordination. A marketplace for the lipstick economy of ideas. Substack’s main sell is that people can figure out how to coordinate around content, instead of by haphazardly put together social. Coordinating in public is becoming harder, and people are becoming more hesitant to reveal their true beliefs, so the beautiful mating dance of like-minds can’t result in good pairings. Substack is a kind of get together of ways of thinking that let’s people use public documents to form cliques through semi-private means.

This is also why it makes more sense to start a Substack than a blog today. Substack is a bit easier to use than WordPress, which is nice, but the main point is that Substack is a thing. It’s a thing that’s happening. And it’s happening now because the demand is being built up through scarcity on other platforms. The writers you like are being fired, aren’t writing anymore, or you can’t find them because the internet is now a cesspool of Search Engine Optimized junk that google can’t properly sift through, and anyway the real stuff is going on in an encrypted Telegram chat.

Substack is the notice board in the tavern, you can read while you’re there anyway. With a little bit of conversational suggestion, you can figure out who else was looking at the same corner as you were.

Substack: don’t screw this up. You’re getting extra community capital for free because you made the right initial investments and the right writers are coming here. People want to discuss writing after they read it, which is a big, big advantage. Threads are a good start, and there’s no reason to make things too complicated; people will find a way to coordinate on their own. The important thing is that the public ledger remain intact. As long as there’s enough of a commons to scrawl a poem own, people will find a way.

A euphemism is not really meta-level, it’s a reference to one thing, understood to be another in context. An implied subject (“SOMEONE wouldn’t like that.”) is the same game.

These, however, are not very effective methods at truly deflecting blame or getting to talk out in the open. This is the internet, if you drop words that trivially correlate with your intended meaning, people will search them and understand. You may even become a target for using such futile measures, because they are so obvious once found. You can write l*k* th*s all you want, but anything mechanical, is mechanically undoable, and personally punishable by exactly the people you were trying to avoid.

This isn’t a new problem, nor is it a problem that has ever gone away in the first place. When we discuss something sensitive but necessary-to-address at work, like resolving who to hire, we talk in such a way that our intention is clear to the people who will help us, and at least defensible to those who don’t. Usually “defensible” means that you can say “Oh, of course, I’m not saying X is bad, just relatively bad when comparing on this axis…” but much more nicely of course.

“Say, Crispy, since you imply that people some how ‘soften the blow’ in order to be defensible, why can’t you show us how people say things ‘nicely’. in the above example?” I’m glad you asked!

It’s because being nice has to do with the specifics of what you’re dealing with. We use the minutiae that surround us, to pin deciding factors on things that don’t matter but others will have trouble arguing with. This is one reason why people make arguments about politicians that have so much to do with personality: because it’s easy to argue about, and people naturally have opinions of personality. This is why when your teacher didn’t like you in high school they could always find random crap in your essay to give you a worse grade than you “deserved”. This is why when you ask for a promotion your boss says that you’re not “ready” and they “need” you in your current position. To have no reason is so shameful that some superior figure might take action, but your reason is also not defensible to the public eye. To have hard-to-argue-back reasons makes it worth nobody’s time to argue.

When we discuss things, we often do not want to completely steamroll our interlocutor, but we tend to want to control the bounds within which they have freedom to speak. We really do want to have some back and forth, but we don’t want to be open to “attack”. People who are a lot more optimistic than me say that this fear is unfounded or at the very least not useful. People like me who don’t know how to vibe, instead see the millions of little handshakes that people are constantly doing—and rejecting—and feel like we’re trying to stay in a very dynamic safety zone, like a hummingbird in a bubble.

The internet is different, but not that different. The point of being oblique on the public internet isn’t usually to be understood by just a single person, unlike when you and your friend were trying to hide behind flimsy metaphors when talking about Forbidden Subjects with mom one room over. Instead, what we put on the internet is meant to serve as a beacon as much as a directed message, and that means we have to make ourselves understood on some level, while defensible to the general view.

Since we tend not to know in advance everybody we’re trying to reach, or at least not know them in depth, the result is a way of talking that’s more like song lyrics, more like poetry. People in my Twitter circles call it “illegibility” but it’s certainly not meant to be universally illegible. It’s meant to be encrypted, with experience as the key—so that when the right person reads it something clicks into place and they say “exactly” to their computer screen.

As the public facing internet begins to evaporate into an “internet of tunnels”—where one node can lead you to other nodes (e.g. group chats, discord servers, urbit planets) but there’s no “front page” the way google or reddit used to be—the need to write poetry comes back. I often thought it ridiculous and fundamentally annoying that so many document from previous eras were written so as to hide behind implications in a way that even the audience of the day often had to search for, and couldn’t directly accuse the writer of. But communication as a beacon isn’t meant to reach everyone, and poetry is how we fill in the “recipient” field without knowing proper names.

The delightful John Nerst points out that we have more of a problem organizing and moving around knowledge than we do creating it. I think we need a Conceptual Logistics that tells us what kind of patterns people have observed over the centuries, how they relate to each other, and where we can find evidence for them.

The simple way to say it is: we need ways of organizing information about the patterns observed about the world and what we believe the scope of this pattern is. What makes patterns complicated to "scope" is not so much about figuring out some perimeter where this knowledge applies and where it does not, because the shape of that perimeter is fractal-like and unknowable. Since this is infeasible, we are left to reason by "positive spaces" we believe concepts can be usefully applied. Most important is the kind of metonymies we will accept for a given thesis: what parts of a statement can be replaced without distorting it too deeply?

If I say "Birds fly." to a child it is not long before they discover the Penguin and the Ostrich and look at me with scorn. Yet, most children quickly learn that this is the precision they can expect.

If I say, "Too much individualism is killing us!" then who would you believe "us" refers to? Probably I'm talking about some subset of what might be called the "West", despite the fact that many "Eastern" countries are experiencing increasing individualism as we live through the 21st century. That's because my choice of English as a language and my choice of words recall similar statements that I am assumed to have some relationship, statements about the decadence of Western society and how it is falling apart.

Again, I would like to return to the problem of teleology. We tend to view things as being for uses, even though things just exist. In this same way, we tend to view the history of a field as a stepping-stone path to the place we have come to today. It may be the path that was taken, but for most fields the knowledge that exists today should be verifiable by other means that the fact that an important person wrote about it.

Looking at historical documents is important for many reasons. The three main ones I feel strongly about are:

1. Summaries of other people's ideas are distorted and censored, consciously and unconsciously. The real thing is whatever the real thing was allowed to be at the time, which is as good as you're going to get.

2. Documents that have survived for a long time and are important to some group usually tell you interesting things about that group and often reveals changes in historical circumstances.

3. Facing the mistakes of past thinkers is the most important exercise for being a truly self-critical thinker.

However, I believe that none of these necessitate introducing ideas as products of history. Don't get me wrong: ideas are products of history. And I believe the reason we teach them as having "arisen" in history is (a) because that's kind-of-sort-of how it happened and (b) because it makes it clear that they come from somewhere, they're not God-given. And yet, teaching and storing and writing about ideas history-first has led to three things I think are really awful:

(1) Cargo Culting ideas of a given thinker, because the thinker is often smarter in some certain way than anyone else seems to be. Many of these historical figures really are that insightful, but I really don't care for anything that you can't give me grounding I can go and find in the world, even if that "world" is the archives of historical documents. Marx comes to mind.

(2) Believing that different "parts" of the idea have to work together in a certain way, because literature that describes them describes the historical idea's anatomy very closely. Ideas are anything you can conceive of and we can change them in subtle ways that are hard to come-up with if we're taught in terms of boundaries. For instance, natural selection occurs in lots of phenomena, like rocks, not just things with genetics. However, I have trouble explaining natural selection of anything non-genetic or non-meme related to most audiences because people can't "see" the natural selection there. They were taught the boundary more than the mechanism.

(3) The most painful issue, for me, is that people seem not to believe they have the right to believe things from the ground-up. "Didn't X big thinker say Y, which is basically the same?" Newsflash: "basically" isn't good enough. We need conceptual tools that work for us today and usually we want to take apart ideas into components and see how robust these parts are. I call this their metonymic robustness, because it generally describes how much different parts of a statement stand-in for more general categories.

Conceptual Logistics

Teaching, describing, studying, and adding to fields as if they are historical scrolls you have caught the bottom of has many uses, but I think there is a space for a new kind of conceptual logistics, one that tries to isolate the usable tools from history. We will leave a paper trail to the original thinkers in our footnotes, but we need modern thinkers that see how the assumption of Chekhov's Gun operates in people's readings of GPT-3 outputs, convincing them of the importance of every quirk.

There will be no single "conceptual logistics", instead it will be a style, defined by the description of theses as tools that have dynamics that are well-understood in certain environments and still need to be studied in others. This is the problem of learning how to use guns in space; what components of gun use on earth and physical laws can we extract predictive power from that will minimize astronaut casualties?

The Inexact Sciences has a unique need for such a style, because too much of human behavioral studies are clouded by assumptions and styles of the time. The pipeline was never filtered because this murkiness became the culture of these fields, even as it mutated. It is time we became serious about concepts as tools, not artifacts.

Sure enough it was, and this lead to a lot of progress on the question of "Where the hell do birds go when they're not here, anyway?"

All in all it was a great day.

One thing that I love about this event, is that it's pretty hard to explain the reasoning that gets you from "African arrow in a stork" to "migration" as scientific in the overly-narrow definition given to "the Scientific Method". Of course, a lot of other things made migration make sense, but fundamentally the reasoning process that went into this wasn't statistical inference.

When I tell this to my fine colleague Suspended, he tells he "idgi" and "can't we just describe that as Bayesian reasoning?"

Foolishness!

Am I'm just fetishizing the idea that we can wrest some ground back from "falsification radicalists" who want science to be about t-tests and nothing else? No.

The devil, as usual, is hiding in the details. I really have no objection to the statement "we could invent a Bayesian system to describe this inference" or "maybe the brain is all just Bayesian inference." Maybe it is! The problem is that I believe theories because of their instrumental value, and this kind of Bayesian description always seems to happen after the fact. The kinds of models Bayesians I know make are so general I wouldn't know how to begin applying them, or else so specific it's not clear how thinking can really be reduced to them.

And I think I know why.

Science, statistics, and the quantitative crowd as a whole has very little respect for the fact that the representation and organization of knowledge is often the key to the entire thing. It doesn't matter if all you do is memorize, as long as your internal archive is structured in such a way that you can recall the precisely needed fact in the right context. Indeed, this seems to be much of what learning actually is; consider the last time you tried to learn a language.

But current attempts to explain reasoning processes as statistical invariably assume a representation to do statistics over. To me, this is like explaining how a car works by assuming that you already have a working engine and you just need to connect it to the axels. Something is happening at the heart of this system to create and maintain a map of the world while doing a million other things, so assuming "P(birds go to Africa sometimes | arrow in stork) is close to 1" is just something that the brain can pull out if it wants to is ridiculous. It's downright ridonculous.

This gets down to the fact that quants usually believe the brain is capable of a kind of hyper-symbolic calculus that will discretely represent anything that's useful. The brain is eventually capable of such symbolic reasoning—but it must become acculturated to it. When people learn calculus they struggle a lot with infinitesimals because they need to get used to this new subset of symbols and how they're manipulated. This tells us our brain is good at reasoning symbolically in domains, and the history of innovation tells us that at least some people are really good at transferring over patterns from one domain to another.

But how does this transferring over work? It's not statistical inference, but a kind of "reasoning by models". Consider this classic XKCD comic:

This is another kind of reasoning, it's one where you choose to think about two phenomena that share certain properties as possibly sharing more properties than the ones you've already observed: analogical reasoning.

But there's no really semantic difference between reasoning, inference, and significance testing, only a few details about application. Statistical significance tells us when a result is likely not caused by the kind of randomness we're assuming would be the default in a given scenario. But it's just as reasonable to consider "axiomatic significance"—where you start with a number of axioms and, upon observing something that breaks your system, start to look for which axioms would need to be removed to fit your theory.

What we see in Pfeilstorch—wherein people stopped believing the birds turned into mice when they vanished from the trees—is yet a more complicated kind of reasoning we have studied very little...until Cognitive Science and Machine Learning have simultaneously begun to force us to acknowledge that intelligence rests on inherently shaky (and therefore flexible!) ground: distributional simulation. Migration is a theory that was put together under the understanding that there are certain things that animals are simply more likely to do (like move) than others (like turn into mice, or die only to appear next year). The space of "things that could happen" was limited to the kind of stories people were willing to construct, and turning into mice was something people had thought about a lot in fairy tales before. But without the creation of these kinds of intuitive theories statistical inference has nothing to falsify.

But it's more than just a foundation for statistics. Much of the time you don't need any explicit statistics at all, because you're not working with enough explicit data that you could ever make a real argument one way or the other. Instead, you need to reason under the dynamics you can agree on with your community. It sounds ridiculous, like a lesson in management but it's the original kind of inference. If you wanted to avoid lions, you had to agree with your fellow hunters where to go under some understanding of lion dynamics and how they refer to lion observations.

If you think we've gotten past that I challenge you to find another basis for the World Happiness Report. On what other basis than agreed dynamics could we ever discuss hypotheses about human behavior? Most sciences are not predictive, it is time we started explaining on what foundations they stand, instead of being mildly embarrassed whenever the topic is brought-up.

What if you designed a robot to mow your lawn, by making sure it gets orgasms from well-cleaned lawns? Except you botch up the entire thing by making the robot so smart it just rewires its own programming to get 10,000 orgasms a second.

Surely this can happen—I don't mean to deny that this sequence of events can occur. Instead, I would like to say that the view of the robot as "cheating" is a mistake, the same kind of mistake that people make when they think that evolution has "designed" the spleen. Evolution has cutaway and added to what was once not a spleen, till our left abdomens look as they do today; we have named the spleen. In this same sense, the above is not the robot "cheating" but rather the designer not being able to do a good job specifying the reward structure for the expected perspective of the agent.

And when we start looking at it that way, we see that this has been happening non-stop for all of recorded history. Consider the College Board, which creates and administers standardized tests. Is it College Board's job to make sure high test scores correlate with:

a) intelligence

b) competency at school work

c) good financial outcomes

I'm sure readers will realize it's d) none of the above. College Board's job is to get students to buy tests. They might even have secondary customers, e.g. book deals, tutoring deals, agreements with colleges. I have no idea and frankly I don't care, the point is that however teleologically you want to view its organizational structure, College Board does not "see" your viewpoint, even though it's well aware of the general feelings people impose on it. College Board's instrumental narrative is that more students should be taking more tests, because it is the fattening of the middle class educational apparatus that has allowed it to prosper.

I am not against this, and I think anyone who is morally against it is misguided. A moral attempt at creating standardized tests would be a disaster, and you don't have to take my word for it: you will see exactly this attempt to happen in the next five years. It will be much like a professor: extremely lopsided and with a tenuous connection to evidence due to its addiction to leading questions.

The problem is that names are generally teleogical: a "can opener" is meant to open cans. For entities that don't do much when you're not using them, this is perfectly fine. For entities with their own thing going on, this is not. Anteaters do not just eat ants.

This problem is in some sense the opposite of the concept my colleague has been developing an extended theory of: surrogation, wherein a metric or indirect percept of a phenomenon comes to stand-in for the "real thing". Think IQ vs. situated problem-solving ability: there's no Mensa club for situated problem-solving ability, how would you decide who gets in? If you just thought "What about the Nobel prize?" please close this window.

I struggle with surrogation, because I think in reality most of the time there is no "real thing" from the get go. We were never acquainted and coordinated enough around intersubjective consistency to claim that the deviation is the fault of measurement, so much as our inability to have any idea what we're talking about. Consider the tired argument around what entities are "alive". I can hardly disagree that what counts as alive will have big implications, but I'm unconvinced that there was ever enough agreement at the edges over what is "really" alive that we have any right to worry that such definitions have betrayed our intuitions. Rather, our use of "that pig is still alive" didn't shove these edges in our face.

Our lack of respect for the limitations of our definitions has betrayed us. There are many reasons why definitions are slippery, but wireheading reveals the most important one: teleology. We generally use names for a purpose and these names change very easily if they are untethered from that purpose. But we never really had a good understanding of the purpose, that was purpose of the name. Names trick you into "bottoming out" your level of inquiry. They give grounding out of the assumption that the other people you share the name with find the same things salient. When this shared attention vanishes, so does the ground beneath us.

Perhaps keeping my initials in my pseudonym was foolish, and perhaps I’m foolish for revealing it now, at the beginning of the end of the American defense of radical openness on the internet. But I want to tell you why it was important for me to keep these initials. In the YA Anime Code Geass there’s a character whose name is CC. Cursed to live forever by someone who simply passed-on their own curse, she wanders the earth attempting to do the same, but doesn’t when she gets the chance.

All the writers who love making living forever a curse and other such lovely inversions of human desires have never lived for a very long time themselves—rarely even a century. I don’t know what living forever would be like, but it seems difficult to believe it has a finite synopsis. Instead, we recognize in the inability to die a loss of control that leads to the world running through our fingers, as if the dying was just another action in the game. And it is. “There is only one really serious philosophical problem,” Camus says, “and that is suicide.”

Wrong. This is the classic analytic mistake: the belief that if you can partition a space with words then you’ve made a meaningful distinction. I either commit suicide or I don’t, that much is true. These arguments are not what make people kill themselves. People are what make people kill themselves. They build themselves up to it because of what suicide might mean to them in their specific time and place. When I was 16 I didn’t step in front of a train even though I tried to, because the train was big and scary and I couldn’t fight my own instincts. I couldn’t get the action to mean the thing I wanted it to in my head—and I kept thinking about how misinterpreted my actions would be.

I have suffered from depression for most of my life. On and off. Anxiety. Bipolar. Words, words, words—I am not very convinced by the people who tell me that these clusters mean very much. They are pharmacological categories, not mental ones. But in the midst of my deepest depression I awoke from a dream with a vague story in my head. The mind of an animal, forced into different environments, transferred to different bodies, tortured by the universe in a way the book of Job should have studied up on first—but always simply trying to manipulate its space not merely to survive, but simply to entertain itself, because these were not really so different. I’m that animal. I’m CC.

—

When I was very young I wanted to be a Scientist. I wanted to understand how things worked and for there to be a way to prove that’s really how things were, not just the way I was choosing to think about it. But the more I pried away at the philosophical onion, the more I realized I wasn’t ever going to prove much of anything. Even though all the evidence points to there being some kind of objective universe, proving anything about it turns out to be tough and everybody who continues down that route either ends-up believing in things that require a level of faith I lack or require one to tone-down one’s hopes to the kinds of things people felt were meaningfully provable. Despite the incredible extensions of the scientific apparatus into every facet of life, the kinds of “proof” we are willing to countenance have continually shrunk as we’ve moved to the idea that knowledge-production is so complex it can only be handled by experts behind closed doors.

In most sciences this comes down to some kind of test of statistical significance or power. This kind of hypothesis-falsification theory of knowledge is so painfully incomplete that I will not bother with a comprehensive study of why. Instead consider why you are able to operate in society: you reason by counterfactuals without ever being able to repeat your experiments. You compare experiences to each other and see which variables you can alter much like a scientist without control of their laboratory. But, then again, where do these discrete “variables” even come from and are you sure that your notion of “good mood” as in “my boss is in a good mood when I sell more Generic Product” is well-defined? You have an intuitive theory and more importantly you can use your intuitive theory to communicate with others, so it sticks because you want tell Katie that the boss is in a good mood today.

If you see a stork with an arrow in its throat, you know it has been in contact with the place the arrow came from:

—

When I was 14 my uncle took me on a roadtrip. I tried to convince him that the universe was meaningless: we could not reach objective truth, even through intersubjective agreement so it was all a big loss. He told me that when he saw his children born it was meaningful and it didn’t mean much if someone else agreed or not. I thought about that, and I realized I couldn’t take the meaning away from him, how could I deny his claim to have caught meaning by its corner?

For a while I tried to force my own mind to find meaningful whatever I consciously decided to find meaningful. If I had continued for long I might have been a rationalist, but alas I was not so strong-willed. Instead, I noticed that my feelings were telling me what was meaningful and it was actually my job to describe the whole.

Was there a whole?

I could do things many times and sometimes they became meaningful. Was the meaning just latent? If it was, I still had the choice to activate it in certain ways, isn’t that making meaning?

—

I have been trying to become an academic for a few years and might well continue for years to come. The situation is dismal, people do the things that get them papers and people let in the papers that look like papers. All of this for a theory of what?

No big theory is coming out of the Academy, not one that explains the kinds of things I want to know about like: How is it that I’m able to establish shared terminology on the internet with others, purely linguistically?

—

My friend sent me this quote:

When I arrived, in 1992, at Warwick University—a dour, concrete campus set in the UK’s grey and drizzling Midlands—I was a callow and nervous teenager, also filled with the hope that philosophy would afford me access to some kind of ‘outside’—or at the very least, some intellectual adventure. Almost entirely overcome with disappointment and horror at the reality of academic life within weeks, it was a relief to meet one lecturer who would, at last, say things that really made sense: Think of life as an open wound, which you poke with a stick to amuse yourself. Or: Philosophy is only about one thing: making trouble. Land was tolerant of my hanging out in his office smoking and drinking coffee, as he (habitually hyperexcited and quivering with stimulants) worked on his comically antiquated green-screen Amstrad computer, and eagerly relayed the latest insights he had garnered from molecular biology, nanotechnology or neuroscience. One could not help but be impressed by the sense of a man whose entire being was invested in his work; for whom philosophy was neither a nine-to-five affair nor a straightforwardly life-affirming labor; and who took seriously the ridiculously megalomaniacal aspiration of philosophy to synopsize everything that is known into a grand speculative framework. He was uniquely able to open up students’ minds to the conceptual resources of the history of philosophy in a way that made philosophical thinking seem urgent and concrete: a cache of weapons for ‘making trouble,’ a toolkit for escaping from everything dismal, inhibiting, and tedious.

from this article. And there it is. That’s really all there is to it.

We live in a certain kind of a society: it views knowledge production certain ways, it delegates certain kinds of powers, it imposes certain kinds of definitions. Forget all this for a moment—the point is that making trouble is good because that is the only thing there is to make. Suicide isn’t the only question, but it is one possible answer to the only important one: What now?

When I came to Academia I was hoping to find allies to pursue in earnest the goal of systematizing our understanding of language. Even then I knew it was a naïve hope and now I can only chuckle. Academics were not having it, but that’s not even the main problem. The main problem is that it’s not clear what systems should look like. Not just this one, but any of them. If I told you to invent language from scratch what would you make it look like? I was so used to the notion that systems can be described and built from the bottom-up because computers are somehow possible that I convinced myself that natural phenomena were as isolatable as those systems we have made tractable by construction.

And yet people talk to each other. Words exist and, unlike so many other things, are repeatable. A “the” is recognizable as a “the”, a pointer to something that nowhere exists held-up by a system we have no really good description of the mechanics of. What point would there be in building-up such a stamp collection for describing these interactions? The Academics would scorn it and the layperson would never have incentive to even learn it existed. Forget it.

Our philosophy must pay rent, and it must do so by letting us create trouble. Of course, most successful people know this as they ride on the backs of ideas they don’t believe because those ideas are incomparably more powerful than any individual. Meanwhile those who flatter themselves as loving wisdom or searching for truth like to complain that people are either stupid or disingenuous. It is time that we allowed more of both in the search for truth, because the real question is why does philosophy care so much more about belief than action?

We say that people do not believe their espoused ideas when they appear to demonstrate behavior we would consider against their interests if they believed such an idea. Are we not, instead, creating a very synthetic notion of “believe” that just tries to impose a kind of order that’s not there?

The most seductive idea in the history of thought is that words can say what they mean, and it is wrong. Words say what they do

—

After the epidemic of construction and deconstruction it has become difficult to talk about pragmatism. Pragmatic about what? We must be even more radically pragmatic and accept that we cannot change ourselves fully. I would like to understand how language works or at least grind my ego into dust until it points in a different direction. I would like others to join me, but they are trapped in a system just as much as I am and I should not convince them with arguments alone—but actions that shape the system.

When I say, words say what they do, I am declaring myself a Functional Pragmatist. When I declare myself a Functional Pragmatist, I am saying something very simple: I want to explain all the things that make me curious, but only in ways that let me do something. Of course, Academia has long been like this. That “something” is getting papers accepted by the community, but this point is not widely accepted. Instead, most insist that the results that get accepted by the community are statistically robust if you squint at the word “robust”.

There are many kinds of knowledge we cannot capture statistically, nor with falsifiability. I cannot falsify the process that created the word “doomscrolling”, but I can make a theory that allows me to popularize new words or convince people to accept a frame within which to discuss this process. This is blatant snake oil salesmanship—I can clearly sell you things like this that will later be falsified. Science is already like that—remember “There’s no evidence masks work on COVID-19.” There wasn’t statistical evidence of the variety that we have to come view as the only island to stand on.

The professionalization of Science is a tragedy, because we are not very good at standardizing things that are supposed to change. So we have decided the kinds of results that can exist in order to make it practical for a profession to be built out of the search for what’s new.

We can have more truth than statistical truth: we can describe the systems that allow our descriptions to keep holding up. Words are a weak substance, and we can keep remolding them forever. Perhaps we will. But what patterns are consistent when we remold them? And how can we use them to cause enough trouble to keep asking fun questions?

—

There is no community that would eat a theory of everything enough to make it interesting or useful, except as a political coordination mechanism. But there are ideas that coordinate people into eating communities—the way commas and periods have changed the global linguistic landscape forever, the way money has redefined the relationship between two people. What is the idea that eats the current global order and poops out something more playful?

I want you to understand Elizabeth Harmon because so far she is the best explanation of the kind of painfully alienated pragmatism I feel unable to communicate about my innerworld. The walls are always closing in as your opportunity to shine feels like you might not make it: you can’t seem to control yourself, so you rely on unhealthy means to slow yourself down into a tractable state. There are people on your journey who helped you. They mean something to you, but you seem only to be grateful to them when they’re not around, because when they are around they can’t stop themselves from tinkering with your psyche like you were an almost-perfect clock.

Forgive the beginning for being a little too shrouded, perhaps, a little too stereotypical. Quickly we see the kind of person Harmon is: someone who is angry, but has been broken down enough that she knows the routines to run with. When she finds a way to slip through the cracks she does and grabs what she can.

Harmon is a genius. She learns how to manipulate the world a certain way and pursues it furiously, but rebuffs every attempt other people make at teaching her better manipulation strategies. Most of them can’t keep up with her in her forte, but many of them have something to teach regardless. The problem is they want to apply for a position that doesn’t exist, they don’t know how to interact with her and they’re constantly trying to frame things so that Harmon has less and less space to maintain the structure that keeps her mind together the way she needs to be. She does find a real person she can lay her shoulder on, but the only reason that person can be there is because she doesn’t really understand Harmon.

To find space to think Harmon is not just always looking for exits, she is creating them, imposing her game on the current social dynamics awkwardly and often at a cost. And yet, if one thing can be said about this, it is that it allows her to think. She goes around dealing with people like a talented amateur gymnast, making fantastic leaps and bounds, and sometimes injuring herself and retreating into the corner. Yet few can deny that from this flexibility is a kind of exploration, one she leverages as she continually digs herself deeper because she has trouble controlling herself.

The acting and the shots here perfectly capture the result: a constant ambiguity about whether the current moment is enjoyable. The shots and acting perfectly capture how she can never seem to decide if this situation is working for her, if this is what she wants. In a few moments there is a burst of joy and you feel it, because you realized the tension was pushing you back and forth across the margin between pain and pleasure.

And that’s the real key, the real reason why this mini-series matters so much to me. The scenes are structured right and there is enough silence that you begin to understand Harmon’s biggest issue is traversing her own mental landscape. Everyone else has stood in her way at times, but dealing with them only required a little pain, a little growth, and few goodbyes to previous standards. But dealing with herself is intractable, because she uses the way her mind functions as an engine. She wants to keep it hot, but it is painful to maintain and she does not know how to tune it up and down smoothly. It is a painful and damaging process that others assume should look different and so there is the struggle to present.

Harmon does love a few people, but she is a selfish person. She has to be. The world has made handles that are the wrong shape and wonders why she is complaining that her hands ache. She kicks the door open and does not apologize anymore, she has only one goal now and the door was in the way.

I don’t know if I can say she succeeds. Watch the show, and I’ll write a follow-up about exactly that.

—

The idea of meaning started as arguments over a ledger—the ledger is what people agree has been spoken. Previously, to win arguments people were fond of arguing over the symbols on the ledger—what was said—and of course this still happens even with airtight recordings. But in general the conversation has shifted to what things mean and this argument has created a massive demand for technologies of equivocation. Of course, they’re not called equivocation, because they’re coordinated around group identity, so within the group things are not only clear, they’re more precise than they used to be because they are honed to pass specific messages to certain people and take advantage of intersubjective agreement. But in truth, this creation is one of getting around the definiteness of the ledger, to allow people room to breathe, for human beings were never meant to be bound to an irreversible ledger.

—

What should we try to describe, when we talk about language? A naive view would be say we should be able to predict what people say. This is clearly impossible, simply due to information access that no one will have in the near future. Of course, our confidence about what someone will say should certainly go up, but probably most of that is being done pretty well by GPT-3 and the correct calibration of the word “the” will probably be more important than anything we write down for the purpose of increasing density….

I would argue that our goal is to get the mechanics down in a way we can do something with, and the question should be what do we want to do?

Well, one thing I’d like to do is argue about memetic strategy—why do people make explicit cultural references in order to garner certain kinds of attention, support, and backlash? How do they know what they are going to get? Of course people don’t know but there are plenty of good examples of people who are incredibly effective at memeing their part of the internet into a certain shape. How can we formalize their knowledge?

This is one of the easiest things to do, because of data access, but when it comes down to fundamental questions these kinds of analyses become even more important in face-to-face relationships. People shrug, purposefully misunderstand, react in degrees, choose to follow certain paths, talk quickly, move around, bring-up parts of the physical environment, use body language to adjust their implied attention, and a whole host of other moves in order to structure conversation. What are they accomplishing? It might be tempting to try a bottom-up description of this: show how facing someone shows attention and then describe what that means, but the top-down compositionality is too strong for me to take this seriously. In fact, communication has so much top-down compositionality I’m convinced one of the major draw backs is that we can’t think of reasonable examples to share in our powerpoints that people will think are polite, be able to understand, and be willing to agree with in public.

We might not know exactly what to explain, but whatever it is, it should explain why a definitive choice is taken. You’ll know a definitive choice when you observe it, by the fact that it deviates from a counterfactual version that wouldn’t have drawn your attention. We don’t and can’t access these counterfactuals and even finding many copies that are similar enough that we can use lots of data to control for the differences is implausible. That’s why this is the Inexact Sciences—we need to rely on intuitions, assume they are true and try to explain the cracks inbetween. But we will not specifically “accept” these intuitions, only confirm them with an extrinsic evaluation: our ability to explain other emergent behavior that would result, e.g. how the nascent ceremony of one generation will interact with another.

—

Where should we begin? We should begin by thinking on what intuitions we do trust or else face the need fortabula rasa science—a prospect I think is not just bad, but truly impossible. Hypotheses come from somewhere. In a space where we cannot execute reproducible experiments, we must especially rely on intuitions, because we can only test thoroughly at higher levels of abstraction (aggregate data of social experiments) and lower levels of abstraction (lab experiments that deal with cognitive effects).

Some intuition:

• Ceremonies generally manipulate underlying objects that serve social roles.

• Complaints are requests for sympathy and sometimes call for systematic change—they serve to establish what is a truth and what is personal by the reaction of the community.

• Their decay indicates these objects decaying or a new game that manipulates these objects, a key signal to causality.

• Compliments make people happy, but their strength in a given context indicates their sincerity.

• People seek status.

• Status is naturally produced by scarcity hubs embodied in well-connected and highly-appraised people.

• In the modern age, defining boundaries of in-groups at multiple different levels is a key purpose of communication.

—

Let us finally begin in earnest.

The point of science and theory is to sculpt out useful idealized machinery to make thinking about very complex systems tractable. People always point to physics for this because they’ve gotten the best ratio of reduction to insight of everyone else. Idealized conditions, though very expensive to approach, reveal fundamental laws that are largely unchangeable. “Largely” because, of course, things like relativity come to mess up the system every once in a while.

When we think about language, this is tricky. The tradition from linguistics has largely been to study language as a mechanism in itself, divorced from as much context as possible, and see how much we can explain with such patterns. While sometimes driven by philosophy, it seems like one of the biggest drivers behind this setup has been the fact that linguistic contexts are very difficult to share (a) because they involve so many different parts of perception, including layers of reality we don’t have any encoding for (e.g. social context) and (b) because it’s pretty clear we don’t even have a really good working theory for what all those components are, or what it would mean to “explain” language in light of them. In contrast, explaining syntactic preferences, presents a space of possibilities that can be studied in idealized lab conditions.

What would it mean to try to explain language? Most sentences have something close to a “literal” meaning—the explicit relationship it draws out among different references. Yet, often what these explicit relationships imply is more accurately described as the meaning. Consider:

A: How’s life lately?

B: Life is…life.

As far as I am aware, “Life is life.” is not idiomatic. Yet, the tautology is understood here to give a lack of positive confirmation—confirmation that is expected if B was simply feeling neutral, implying B has been having a negative experience. Most semantic representations attempt to encode this “literal” or “explicit” meaning under the (correct) intuition that other meanings are largely parasitic on the literal. “Largely” because in every private community there are linguistic cues that come to be understood as having connotations that are prior to the literal meaning of the sentence, usually in reference to a key event or common signal that comes up often in the given community.

The question I’m interested in is: How do we understand what is “understood” from a sentence? Social contexts are varied, so most representations I’m aware of would be incredibly insufficient. Where do we start with our idealizations? Representing the internals of a person’s head is even more of a mess than representing outcomes. Yet, I think if we approach things through the classic attempt to represent the literal I think indexical distinctions that come up become murkier and murkier, until we end-up in the shadow of Chomsky trying to explicate meaning from syntax.

Perhaps the biggest obstacle is just data—language spoken as part of a formal experiment simply isn’t representative of most of the factors that go into private language use. And explaining communication at least as hard as explaining human intelligence—likely more complicated because of network effects. Again, it depends on our definition of “explain”

Data seems like an insurmountable barrier…except for large social media companies that wish to study their users. Data on a single platform is still usually not rich enough to understand real coordination, but it is certainly more informative than laboratory conditions. Needless to say such use of data would lead to a publicly outcry now. That is why it is happening behind closed doors and stringently guarded, with plenty of descriptions of side-projects that justify it, e.g. Fake News Detection. I would be willing to wager that 30 years from now using private data on a platform to study the human psyche will be viewed as at most a “necessary evil” as long as “proper privacy protocols are followed.”

What will we do when all the real understanding of language has been monopolized by individual companies with no desire to create and share a general theory? The time to act was 70 years ago at the dawn of AI, to understand that communication would be a tricky and sensitive notion, because it is the coordination of hidden representations in the brain, representations that are essentially our thoughts and identities. Current ethical discourse suggests we should shutdown anything too sensitive as privacy infringement and largely revolved around what is “too dangerous to study”. I cannot, in good faith, accept this line. If we want a universe where we have an understanding of the technologies of manipulation, the way to do it is to make a public science of the social games people play.

In 2020 this has become an especially sensitive topic, as people’s words are increasingly weaponized against them in culture wars. There has likely not been another time since the rise of the internet that people were less willing to have their words used for public study. I don’t claim to know what should be done, but it is precisely these kinds of Twitter dynamics where real social power is being brokered that needs neutral scientific study.

Perhaps this is the rise of the anonymous research group. Cryptocurrency fluctuates in value, but reproducible findings on public data can lay claim to some kind of fundamental value. Or perhaps the fragmentation and hoarding of power by groups that are well coordinated is now inevitable.

Regardless, I intend to attempt a public study of communication and most especially power dynamics. Status and power are essential elements of every conversation, and they inevitably determine the stances of interlocutors—yet they are completely irrelevant to literal meaning and are thus pushed to the secondary place of “flavoring” text rather than showing its interpretation. This couldn’t be more wrong. It is status that determines whether a joke is funny because it should be taken sarcastically or literally. It is power that determines what a person is willing to say and therefore the information entropy of their response, because if there is only one possible response no information has been transferred at all.

In the Information Age, power is becoming easier to quantify, because we can study quantifiable properties (e.g. followers) and effects (e.g. repeated themes in new messages of other users). If we do not have the data or the conceptual vocabulary to understand the dynamics of two friends chatting, then I suppose we will have to wait. But the time is ripe to understand the new sociality of the web, where the majority of socialization in the USA is taking place under lockdown. The game is to cultivate the in-group, but how do people know where the lines are and enforce them so efficiently?

The point is simple though—we’re always communicating for effect. The idea of a singular model is very strong, but it obviously doesn’t hold. We communicate a totally different stable, continuous stream of information to our friends vs. our parents vs. our partners. The point is to be stable, not to be true, and people use this bandwidth to get things done, which is really the main evolutionary pressure on language.

—

The Paradox of Confidence — Big Dog Mindset. People who are sure of themselves are comfortable admitting not knowing and not being up to a task. This can be very advantageous and even help one achieve higher status. Yet, if it is employed by someone too lowly or in the wrong direction (e.g. moral ambiguity among puritans) then it will duly punished. In other words, it is means of a class warfare—to be used by social climbers who understand the class they are aiming at well.

—

Fallacy of the commons — if I’m nice I can expect others to be nice the same way.

What actually happens — if I’m nice, I attract people who like that kind of niceness.

And sometimes those people will try to snuff out every other part of me as inauthentic.

—

Growing up without knowledge of the war was the hardest thing—constantly defending yourself against heroes because you were bad. Because there was no war—why were you starting a fight? It was a battle not to start one, against tinder and matches.

—

The trouble with signaling—people aren’t really that Machiavellian, but if you try to understand signaling you end-up thinking they are, because you assume all meaning is consciously intentional in your explanation. Signaling, like everything else, is created by a process of natural selection. And natural selection takes advantage of every level of behavior, because it works automatically.

—

Scheduling as a lifestyle can be good. Scheduling as an activity is bad.