It’s not necessary to introduce Slate Star Codex (SSC, www.slatestarcodex.com) on Less Wrong. I would say that it is probably one of the best blogs on the internet. By chance, I forwarded one of its articles just today.

Yet, some posts have been bugging me (some for months). I hope Less Wrong is the right place to address these and, crucially, to enable a discussion. Please note, even when I am using polemics, I do so in good faith: directed at bad ideas only.

I’ll start with the worst post on SSC, Archipelago and Atomic Communitarianism (http://slatestarcodex.com/2014/06/07/archipelago-and-atomic-communitarianism/). I’m calling it the worst post because it is both wrong and a defining (top) post. Also, it is still referenced from time to time (the latest I remember is from November: http://slatestarcodex.com/2017/11/13/book-review-legal-systems-very-different-from-ours/).

To sketch the idea in a few words: in a different world, a magician shows humanity an archipelago where everyone can start a community with whatever rules they can think of. If you don’t like a community, you just leave. Plus, that magician introduces a government that solves all problems that arise.

I.

The first part of my argument is very simple: every utopia needs to apply to our world. This could be in a negative way: no poverty, hunger, or suffering. Or it could be expressed positively: let’s say swarms of nanobots that combat all illnesses. But it has to apply to us to be relevant.

In this case, Scott Alexander (purposely) uses a magician to create a blank slate, a canvas, for his ideas. The Archipelago’s different rules allow him to make his point. But they also block the way back to our world and society.

Just showing that certain elements are like today is not enough. Only if the whole would work for us could we draw conclusions from it. Due to the way it is constructed, the Archipelago is inherently not transferable to our world. So, nothing can be deduced from this “exercise in political science”, especially concerning politics.

II.

This could be enough, but my argument is that the flaws of the text go deeper. So, we have to have a closer look at it. Importantly, the text touts “communities” as the solution of the problems of both liberalism and conservatism. This is not new: Communitarianism (https://en.wikipedia.org/wiki/Communitarianism) has been around for some time. The new idea is (mainly) that everyone can just leave their community and start a new one. Thus, the title Atomic Communitarianism.

So, we’ll have a look at these communities and how their environment influences them. I want to start with the forces that pull at the communities: if it’s everyone’s highest right to leave and start a new community, it’s easy to see that there will be not much left of them. Even the strictest people don’t include themselves under their rules, and even the most principled monarchist (or the like) imagines himself on the throne or very nearby. Any conflict or looming punishment would drive people to start a new island. Most communities would fray until they are composed of one or two persons – truly atomic communitarianism. Just think of all the people that violate their own rules, apparently designed for others only; I don’t think anyone wants to argue that they would start to drink the hemlock in the Archipelago.

Then, there are the forces that push: while the elementary problems of our society, food and work, are only hinted at, the author mentions that the communities should be self-sustaining. This means that large communities must form, especially when the luxuries we are used to are considered. Not only is that a horribly inefficient way to organize work, but it also leads to oppression: people stay at communities if these produce the goods they need or crave. A simple answer to this point would be to give this competence to the government, but this leads to more issues.

Before we talk about the government, I want to discuss something the author cites as reasons for communities, namely that they make people more “trusting, generous, and cooperative”. In a truly uniform community, these words change their meaning. The positive connotation implies the relation to someone else, from an out-group. If they only apply to the in-group, they are not positive any more, but expected. Or would you use those words to describe a co-worker to your boss?

Let’s follow the argument of the text for a while. A first constraint is introduced to the otherwise free communities: they all pay taxes for a military. That military prevents wars, global warming and unwanted influence on others. Also, it enforces the right to an inland, or the right to leave an island. The most pressing question would be here who would voluntarily form the military, and who would control it?

I don’t know an answer to the former question, but the latter is probably the government. The government is supposed to be as small as possible, but again, there is no explanation who will be part of it or what checks and balances are used. Presumably, its legitimacy comes from controlling the military.

The second constraint to the communities is central schooling by the government to insure against brain-washing. At this point, I might repeat myself, but who are the teachers? Who decides the curriculum?

Government power might be envisioned as even greater: once in paragraph V, the author mentions redistributive taxation. I’m unsure whether this is meant as a possibility or as a given, but it would obviously and greatly increase the power of the government.

The government also has a capital, a “neutral option”. Judging (a little bit unfairly) from the examples, we could also say a not-extreme option, be it religious or political. The only rule for this capital are strictly open borders.

There is a third restraint that you are not allowed to switch communities for your benefit, only for your beliefs. Finally and most importantly, the fourth constraint is that those people love paying taxes to a government they don’t have a say in. To get rid of oppression through laws or customs, the proposition is taxation without representation. Fittingly, the sentence starts with “In my fantasy”.

Later, this is declared a “leftist utopia” because it “solves differences of opinion”. I would argue it makes opinions meaningless instead. Furthermore, the only meaningful opposite of oppression is freedom. And I don’t mean surrendering to a community of your choice. Is it non-oppressive if all decisions are made for you and you don’t have a say in it? Surely only if you don’t object.

And yes, you could go to the capital (and hope it is at least as good as our society). But this brings me to my final point: in the last parts of his text, the author backpedals a lot from his utopia. He mostly tries to show that elements of it are here already, and the capital seems to be closest (we don’t know, he doesn’t talk much about it).

To me, this diminishes the – already low – payoff: being left alone. Can this even be called a utopia? It’s not really great, and it’s more or less possible now regardless.

III.

But what does this mean? I will try to make some sense of this text, explicitly against the author and his intentions. These are my interpretations. The text was meant as political, and close to our society. Let’s treat it so.

If we look at the Archipelago again, it seems that the islands are largely irrelevant, and everything is decided in the government (or its capital). More to the point, the islands have to be self-sustaining, yet pay taxes to the government. The government, in turn, steps in when the islanders disturb someone else.

Back in our society, who could populate these islands? Probably the superfluous of today: low-skilled and uneducated, possibly homeless or drug-addicted. Not only not employable, but not necessary to employ – the work will be done without them. Their hopes and dreams are negligible, but they can become a thread to society if they start to riot or vote wrongly. Wouldn’t it be better if they had islands where they could live happily without disturbing the grown-ups? And, mind you, that is a nice proposition, especially compared to those you might also find out there on the internet.

Needless to say, don’t expect me in that Archipelago.

New Comment
8 comments, sorted by Click to highlight new comments since: Today at 12:04 AM

I also have a bit more of a meta comment: It seems needlessly adversarial to have your username be "EverythingWrongWithSSC". I don't want hate/sneer clubs on LessWrong, and especially want to avoid people defining their identities in opposition to other authors on LessWrong. In my model of communities this quickly leads down the road to internal splintering and internal conflict while wasting a ton of resources and generally making it hard for anyone to feel safe or think clearly.

I currently downvoted this post, but would probably remove my downvote if you change your username to something that will less likely introduce needless tribal politics into this whole discussion. Sent you a PM to which you can respond with whatever username you want to have, and I am happy to change it for you.

Only if the whole would work for us could we draw conclusions from it. Due to the way it is constructed, the Archipelago is inherently not transferable to our world. So, nothing can be deduced from this “exercise in political science”, especially concerning politics.

This seems basically wrong to me. No model of anything can be fully transferred to the real world. The quality of a model comes from the model highlighting or clarifying important dimensions of reality that help us make better decision.

I think you are mostly constructing a strawman of Scott here, that doesn't really criticize the model on the dimensions on which it tried to be useful, and where it hoped to be applied, and instead attacks the model on a bunch of irrelevant extrapolitions where the model never claimed to have much accuracy or meaning. (Example: Critizing the atomic theory of the nucleus for clearly being bad at describing the way electricity flows through wires. The model is primarily optimized to help you understand phase transitions, collisions and chemical reactions, and will fail in many cases. This does not make it a bad model.)

Actually, the last paragraph is exactly my point. If I am not completely wrong, the Archipelago suggests a utopia, a political state to strife to achieve (at least in part). But you can't draw conclusions about something else, in this case politics. You would have to show that it has meaning, accuracy or transferability for that, however you want to phrase it.

That's why I don't think the word model is right here. For a describtion, a model would be fine, and might help. I don't think the Archipelago is a model in that sense.

EDIT: Let me give you a stupid and arbitrary example. Imagine a world where all dogs are at least 7 feet high, and people still loved dogs. Obviously, with their height, they could potentially be dangerous. Thus, in that world, it would make sense to say that all dogs should undergo mandatory training. But that doesn't follow for our world.

I myself wouldn't count the post as on of the better one's of Scott but I don't feel like you are engaging with it in a way that's insightful. The point of his article isn't that we should create the Archipelago society but Scott gives it as a model so we can think about decisions whether we move towards it or away from it.

If you think about a decision like California or Texas seceeding from the US that would be a move to become more Archipelago like and the case that's made is that moves like that would be good.

[-]Elo6y30

I feel like the title here could be, "I think this ssc post is wrong here's why".

You're right, but I still think it's the worst. So, I apologize for the click-baitness and I'll try to change it from "The worst post on SSC".