Against Cultural Separatism

There’s an idea that goes around among people who are unhappy with the status quo that the right thing to do is cultural separatism. Build our own communities, our own schools, our own law, our own media, our own fraternal networks, and drop out from mainstream culture.

The problem with this is that it’s ghettoization. It alienates us from the very people we need to be strong for. If we become like Scientology, we lose touch and lose legitimacy.

We should not be trying to make a cult that escapes from the mainstream. We should be becoming the mainstream.


Optical Illusions


The above is an optical illusion. You will notice, abstractly considering the scene, that the animation represents a rotating mask of a nordic-looking man such that you see it alternately from the front and outside, and the back and inside. The mask is of zero thickness, so that the inside has precisely the same shape and details as the outside.

The “illusion” is that if you pay attention to the way your mind’s eye parses the rotating mask, you will notice that you always “see” the face as if it were facing forwards. Your brain refuses to parse the face as inside out. Your low level face-detection skills are automatically interpreting the face as if it were “proper”. The knowledge that, considering the whole scene, it’s actually inside-out, is not making its way down to the low level parsing, unlike other optical novelties like the cube and the rabbit/duck where the same scene can be parsed multiple ways. In this one, we are finding a parse which we know to be right at a high level, but which we are unable to perform at a low level.

That said, I’m sure some readers will be able to parse it differently, due to differences in parsing and deliberate concept-application skills.

Thus this is a clear and strong case of a true illusion. Our mind is unable to do what we think it ought to in this case. Why?

The straightforward explanation is obvious: We never see zero-thickness inside-out faces in everyday life, so the concept is totally foreign to the really low level face-parsing stuff that we were either born with, or learned when we were young. The low level stuff isn’t going to change how you parse reality at a basic level just like that, even if it can in principle. Why not?

Low level foundational perceptual parsing has to quickly, accurately, and reliably give you info about the world. And generally, what you see is unambiguous; bottom-up concept-matching, from simple things like edges, up to the interpretation of the scene as a whole, just works. Bottom up matching is way faster, for basic computational reasons, than doing a two-way match, where higher-level constraints can fully override lower level high-confidence concept matches. We would need a two way match to parse the above inverted face properly. But bottom-up matching is more computationally practical, so that’s what we use.

But we could in principle bottom-up parse the inverted face, by having an optimized inverted face concept ready to go. Like learning to play an F on guitar, it would just take some practice to get the use of the concept fast and fluent enough for it to be able to be looped into visual parsing. Even if some way of seeing is correct and we know it, we might not actually have internalized the skill of seeing it that way.

Some would say that there is an additional problem: bottom-up non-inverted facial parsing is so ingrained in fundamental hardware, because of the above reasons, that it’s actually cognitively impossible or always at least some fixed difficulty for normal people to see the inverted face as an inverted face; the face-detection hardware will always override. This is an interesting claim with which I disagree; I think brain hardware is general, enforcing at most weak bottom-up parsing, without specific “face hardware”, and it’s very possible to learn to see an inverted face. I think the only reason we can’t see the inverted face is because we’re so un-used to it; we just don’t have the skill yet.

The reason for why skills can take longer to acquire than it takes to know what to do in principle, is the same reason it takes longer to write fast code than it does to know in principle what you want to calculate. You have to come up with an algorithm, implement it well, strip out extraneous abstractions and cruft, and do some testing and tweaking before it’s ready for prime-time in an inner loop. Likewise, the high level “inverted face” concept, useful for abstract cognition, is too symbolic, too interconnected to other high level concepts, and too full of extraneous content to be used in the equivalent of an inner loop in your visual processing. Hooking up high level symbolic concepts to low level perceptual parsing is like hooking up a callout to a symbolic math solver in the inner loop of your graphics code – it’s just not going to work. Concepts in your mind aren’t magic, they’re technology like anything else, and have to obey computational constraints.

So for purely computational reasons, we wouldn’t necessarily expect to be immediately able to see the inverted face because we don’t have the concept at the appropriate level of abstraction and performance, and it takes actual work to build such a concept.

There may be the additional problem that low-level visual parsing stuff is like firmware, and there’s an additional step of actually physically installing the relevant concept in the visual equivalent of muscle memory. I suspect this is not a useful way to model it; I expect that building and installing the low-level version of the concept is like building and installing any other concept, except for the inherent differences of building a low level concept. “Muscle memory” and low level perceptual parsing memory, if they are physical areas in the brain with different properties, are just space that is particularly suitable for low-level IO concepts, but accessed like normal brain space.

So I think it’s possible to see the inverted face, even though I can’t at first. If I’m to put my money where my mouth is, I should train myself to see the inverted face. There are two ways we could approach this:

  1. Build the low-level-optimized inverted face concept and install it, so that normal bottom-up parsing can match it as an inverted face.
  2. Build the full two-way parsing skill, such that higher level concepts fluently propagate down to low-level perception.

Even if our brain hardware is general enough to do something like the latter, I think it’s so computationally impractical to do it fluently that it’s not going to happen. But if we insisted, what would it look like?

First of all, the fast and easy bottom-up perceptual parsing would be your basis. Then when you detect a mis-parse and decide that it’s a problem, you would build the lower-level version of the relevant concepts, and reparse. Other algorithms are possible, but anything else would be silly. The problem basically boils down to how good it is possible to get at building low-level IO-grade concepts from high-level reasoning-grade concepts.

Am I right that I should be able to train myself to parse the thing as an inverted face? Right now I can’t. It’s the nose that gets me. I can’t see the nose as anything but forward. This will take a while.


“We live in a good neighbourhood” -> “We live in an upper middle class white neighbourhood”

“Good schools” -> “I only want my kids to associate with the higher races”

“Literacy is high status” -> “Racially superior heritage is high status. Literacy was a good proxy in the 18th century”

“He’s uneducated” -> “He’s a nigger, or white trash”

“Educated people” -> “High quality specimens of the higher races”

“Low socio-economic status” -> “Racially inferior”

“Some sketchy looking gentlemen” -> “Some blacks”

“The New York crowd” -> “The Jews”

“Greedy global capitalists” -> “Jews”

“Communism” -> “Race-traitors, Jews, and Untermensch”

“Hippy-foksy organic” -> “White”

“Racism sucks” -> “The other tribe of whites, which are both racially inferior, and our political enemies, talk more openly about race. We are very afraid that they might wake up to the racial nature of our hatred for them, and our use of the lower races as a weapon against them.”

“Disadvantaged peoples” -> “Allies of the lower races in the race war against most of the white race”

High socioeconomic-status, literate, educated people who live in good neighbourhoods and eat organic have a hard time talking directly and openly about race. So we don’t. We use proxies. These proxies occasionally confuse us, tricking us to think that race isn’t the issue. But most of the time, it’s actually about race.

Note that the racial categories educated people actually use are more fine grained and nuanced than the simplistic “white, black, brown, yellow”. They distinguish quality of blood, not just broad racial type, and the implicit politics is messier than a naive black vs white vs brown vs yellow interpretation. But they are still racial categories.

Once we realize how much of the supposedly non-racial language we use is actually about race, we have to start taking race a lot more seriously in politics and society. Most of the time these days, politics isn’t about values, principles, moralities. Those things are important, sure, but they are like law in that they come after victory. At the most basic level, politics, at least today, is about us vs them. And us and them are racial categories very much more often than openly acknowledged.


Modern Marriage Isn’t Marriage; Sex is Marriage

I called my Social Matter article “How to Catch a Wife” rather than “How to Get Married” because modern marriage is what gays do. It’s about big parties, tax advantages, and cruising for hookups with your sexual life-partner of unspecified gender. Call something “marriage” and everything gets focused on the degenerated modern life script around marriage, call it “catching a wife”, and we’re a bit clearer about how this thing actually works.

If White Sharia (I dislike the term but it’s the best we’ve got right now for an important concept) is going to work, we need to rebuild our understanding of marriage from the ground up, without the cultural baggage of the modern world.

One of the most obvious failures of conservative people talking about marriage is the claim that you shouldn’t have sex before marriage. I get the idea, but it relies way too heavily for its truth on a concept of marriage that doesn’t exist anymore.

When the “no sex before marriage” rule was formulated, here’s how it worked:

  1. You see a woman who catches your fancy. You want her. You also happen to be in want of a wife.
  2. You go to her father and negotiate a transfer of ownership of her sexuality from father to husband.
  3. If he agrees and she agrees, you get married the next day.
  4. Time to first fuck: about a week. Time to marriage: about a week.
  5. Bam! Done! Problem solved. Let’s move on to the next problem, because real cultures have more important things to do than to waste time in years-long engagements and extravagant weddings.

In this scenario, where marriage is a fast low-bullshit formalization of a transfer of ownership, then yeah, no sex before marriage.

But in the modern world, we have no property rights in women’s sexuality, no patriarchy, “marriage” is slow and high-bullshit, and “marriage” doesn’t even confer the relevent rights and duties. We are not talking about the same thing, so “no sex before marriage” just becomes a kind of kooky and obsolete cultural leftover, which is why no one does it.

So let’s throw out everything we thought we knew about marriage and sexual ethics, and rebuild it from ground.

My solution is simple: Sex implies marriage. If you are having sex with someone, you are married. You ought to formalize that as quickly as possible with your social community with a wedding or equivalent, but the important factor is not whether it’s formalized, but whether the sexual substance of the relationship exists. And once you’re having sex, the rights and duties of marriage apply: no divorce, no adultery, no independent existence.

The problem is not people having sex without having done the proper rituals beforehand. That’s fine. It’s great. I’m glad for them. They would have done the rituals if the social setup was such that they were important. The problem is when they stop having sex with each other, and start having sex with other people. That’s when the drama starts, and that’s the thing that needs to be controlled.

The rule isn’t “Don’t have sex before marriage”, it’s “Don’t have sex with a girl you aren’t intending to make your wife”, and “Once you’re having sex, don’t break up, don’t cheat, and do become one entity together.”

But why?

I laid this out in my Mannerbund article: to get cooperation working among real men, you need to control sexual overlap. If a woman has two men, they fight. If a man has two women, he’s depriving his brothers. And either way, accepted adultery complicates the social script and creates uncertainty and drama.

These concerns don’t start with “marriage”, they start with sex. Once you’re having sex, you’re married for the purposes of mannerbund, because if you break the rules of old-school marriage, it causes problems.

And what about the other side? Is it a problem to have sex without getting formally “married”, if you follow the rules? Not really. Your bros all know who you’re fucking, and as long as you act like you’re married, there’s no trouble.

The formality of the marriage ritual is an additional social technology that builds on top of the basic mannerbund sexual morality. The ritual reinforces the rules, commits you to them, broadcasts intent, and makes everything work better, but it is secondary to core sexual morality. It is entirely useless to get “married” if you’re not going to follow the core sexual morality.

Once your community can establish basic sexual morality, then you can formalize it with fast and low-bullshit, but binding, rituals.

So don’t worry about whether you’re formally married or not. Worry about whether you intend to be married in the true sense of the word. Formalization follows substance.



Definitions vs Distinctions

We use the term “Elite” a lot. Along with other terms like “Natural”, “Power”, “Wealth”, “Capitalism”, “Darwinism”, “Empiricism”, etc, it’s a term that we use in many different and often contradictory ways. Out of this confusion comes annoying semantic disagreements and confusions. There are two ways to handle these:

  1. Define Terms. Grab a dictionary. Offer a final definition. Make it clear exactly what is meant by the term in question, so that each usage of it can be checked to ensure it makes sense. This tends to eliminate sloppy and confused arguments, and clarify thought, so everyone generally understands it to be a useful thing to do when encountering word-confusion.
  2. Distinguish Different Senses. Sometimes, when the use of a word is confused, it’s because there are actually multiple related but legitimately different concepts in play. In this case, the important thing is making the distinction.

People get into all kinds of meta-confusion by trying to define their way out of a distinction problem. If you actually mean two different things by “Capitalism”, then you’re not going to be able to find a satisfactory definition. If you’re in an argument with someone in this situation, it’s easy to get into conflicts about what the words “should” mean. This is a mistake.

The most important thing is making sure each different concept is clearly distinguished, and making sure each concept is itself crisply defined. What concept is attached to what word is important for efficient communication, but distinctions and clarity are important for correct thought.

Wiki Excerpt on Totalitarianism vs Authoritarianism

You can find the most interesting things on Wikipedia. This is a segment of the “Authoritarianism” article, on the relationship between Authoritarianism and Totalitarianism:

Totalitarianism is an extreme version of authoritarianism. Authoritarianism primarily differs from totalitarianism in that social and economic institutions exist that are not under governmental control. Building on the work of Yale political scientist Juan Linz, Paul C. Sondrol of the University of Colorado at Colorado Springs has examined the characteristics of authoritarian and totalitarian dictators and organized them in a chart:[13]

Totalitarianism Authoritarianism
Charisma High Low
Role conception Leader as function Leader as individual
Ends of power Public Private
Corruption Low High
Official ideology Yes No
Limited pluralism No Yes
Legitimacy Yes No

Sondrol argues that while both authoritarianism and totalitarianism are forms of autocracy, they differ in “key dichotomies“:

(1) Unlike their bland and generally unpopular authoritarian brethren, totalitarian dictators develop a charismaticmystique‘ and a mass-based, pseudo-democratic interdependence with their followers via the conscious manipulation of a prophetic image.

(2) Concomitant role conceptions differentiate totalitarians from authoritarians. Authoritarians view themselves as individual beings largely content to control, and often maintain, the status quo. Totalitarian self-conceptions are largely teleological. The tyrant is less a person than an indispensable ‘function’ to guide and reshape the universe.

(3) Consequently, the utilisation of power for personal aggrandizement is more evident among authoritarians than totalitarians. Lacking the binding appeal of ideology, authoritarians support their rule by a mixture of instilling fear and granting rewards to loyal collaborators, engendering a kleptocracy.[13]

You could almost read it as sympathetic to Totalitarianism. I’ve said before and I’ll say again that Totalitarianism is a useful and sometimes even positive concept badly distorted by 20th century memetic warfare. The word itself is probably a lost cause, but there are important debates to be had about the underlying concepts.

For example, do we want a high-charisma high-legitimacy publicly involved ideological government, or something more impersonal, detached, private and apathetic? There are good points on either side, and this conceptual space should be mapped out and understood so we can choose well or know how to transcend the question.

Refining Moldbug on Authoritarianism vs Orwellianism


An authoritarian state has no need to tell its subjects what to think, because it has no reason to care what they think. In a truly authoritarian government, the ruling authority relies on force, not popularity. It cares what its subjects do, not what they think. It may encourage a healthy, optimistic attitude and temperate lifestyle proclivities, but only because this is good for business. Therefore, any authoritarian state that needs an official religion must have something wrong with it. (Perhaps, for example, its military authority is not as absolute as it thinks.)

There are two types of power being discussed here: Force, and Persuasion. Persuasion can of course verge into Orwellian levels of mind control.

The claim in the passage is that if a regime has absolute force available to it, it will not need to use persuasion. In a trivial sense, this is true. A militarily invincible regime cannot be overturned, and thus does not need to use persuasion to prevent itself from being overturned.

But the regime itself is composed of men who are motivated and coordinated by their beliefs, and can be demoralized or turned against the regime by a persuasive adversary. The regime can shoot any men who don’t carry out orders, and crypto-lock the weapons, but at some point, an unopposed sufficiently persuasive adversary can make a mess of the warrior class and overturn the regime. The levels of perfection of force control required to totally disregard mind control are staggeringly unrealistic.

Even given perfected force-based control sufficient to make the regime invincible, there is still a use for persuasion. Moldbug acknowledges this: “It may encourage a healthy, optimistic attitude and temperate lifestyle proclivities, but only because this is good for business.”

In other words, even given political security using force, there are problems of the proper organization of society that are best solved with persuasion. Moldbug minimizes these, but they are actually quite extensive. You want people to be internally lawful. Most lawfulness comes from belief in the law, not fear of force. You want people to live good traditional lives, you want people to be virtuous, you want people to think particular fields of work are high status, you want people to socially enforce the desirable behaviours in a distributed way, you want people’s worldviews and cultures to be locally compatible, you want people to know which movements and subcultures and fashions are good or bad for the community, etc. These “business needs” are all better served by persuasion than force.

A possible objection to my framing is that these things are not “telling the people what to think”, because the subject is not the political legitimacy of the regime and related ideas. To be clear, we can further divide applications of power by their purpose:

  1. Security of Power. We have discussed how an infinitely forceful regime would not need persuasion. The underlying problem is how to secure political dominance of society.
  2. Organization of Society. The above discussion of the “business needs” for persuasion is about the organization of society. The problem is, given secure political dominance, how do you (re)organize society to achieve the best results for your ultimate purpose (presumably something like the glorification of God through the excellence and flourishing of your society).

So 2009 Moldbug seems to think that the application of persuasion in pursuit of the security of power is some combination of unnecessary, because an “infinitely” forceful regime doesn’t need persuasion, and bad, because political persuasion is often harmful. But he also seems to think that some level of persuasion that’s “good for business” is OK.

Let’s look closer at the structure of the problem. We can cross the types of power presented at the beginning of the post with the purposes of power listed above, to make a 2×2 matrix of ways and ends to apply power:

Security of Power Organization of Society
Force Firing into a destructive mob, putting down rebellions, arresting people who criticize replacement immigration on facebook. Forcing economic reorganization, punishing immorality, putting natural slaves to work.
Persuasion Legitimacy propaganda, divine right, democracy education, fake news. Behavioral propaganda, astroturf or supported activist campaigns, moral education.

Moldbug objects to the third quadrant, the use of persuasion for the end of political security. Or at least he criticizes the way our current regime abuses that quadrant. But we see that there are both legitimate and destructive things in each quadrant. None is bad or good in itself. When we critique an application of power, there are two things we might be taking issue with:

  1. Desirability of the ends. We may dislike regime propaganda or force applied against rebellions because we don’t support the regime. It is an axiom that everything the enemy does is bad, because they are the enemy. We may dislike some moral enforcement or propaganda campaign because we think the targeted reorganization is stupid or harmful. For example, many modern liberal fads backed up by both force and persuasion are actively harmful to the social fabric.
  2. Appropriateness of the means. We may think that the means applied to achieve the result are the wrong thing, or that it’s not worth the cost. For example, some regimes corrupt the scientific and epistemological processes of society to shut down dissenting ideas, which is an enormous cost. Likewise, forcing people to live a certain way against their will is usually quite harmful and not worth it. In these cases, there may be other ways to accomplish the same end which are less objectionable: simply arresting active traitors is usually better than trashing society’s ability to think to shut them down, and persuading people to live a socially better way is usually much less harmful than force.

The point Moldbug is making here is about the second objection: appropriateness of the means. In his argument, the reason Orwellian regimes do abusive things to the societal capacity for thought is because they have to, because they don’t have enough force, or can’t use force for some reason. The use of persuasion as a means is often inappropriate, but if a regime doesn’t have enough force, they have to do it anyways. The argument is relatively simple:

  1. There are two classes of power: force and persuasion.
  2. Sometimes force is the best way to achieve an end, and sometimes persuasion is the best way.
  3. The optimal strategy is to use each where it is appropriate. Force when it is the best way, and persuasion when it is the best way.
  4. If a regime has insufficient capacity for force, or is constrained to not use force, it will sometimes not be able to use force even though it is the best way. Instead it will have to use persuasion.
  5. Inappropriate use of persuasion can be more destructive than necessary, and is a departure from optimal.
  6. The more force power available to a regime, the less it has to rely on persuasion in circumstances where persuasion is inappropriate.
  7. The more force available to a regime, the closer it is able to get to the optimal governance strategy.
  8. An authoritarian regime is unconstrained in its use of force. It can arrest the opposition for example. A democratic regime cannot.
  9. Thus a democratic regime is less able to achieve the optimal strategy than an authoritarian regime, and more reliant on Orwellian levels of “persuasion”.
  10. (Unargued hidden premise). A democratic regime is not likely to have a stronger tendency to optimal governance than an authoritarian regime that is sufficient to outweigh its lower capability.

So Moldbug concludes that democracy is bad, and more capacity for force is good. Great. We agree. However, he also writes off persuasion and state churches and all that. But note that the argument is symmetrical: it works just as well to show that more persuasion is good, and an authoritarian regime that is ideologically or structurally constrained in its use of persuasion will be forcefully abusive and not as good as one that is strong in each capability, and able to use each where it is appropriate.

In other words, assuming all else is equal with respect to the desirability of the ends of the state, we want a regime that is both very good at using force, and very good at using persuasion. The scholarly consensus is somewhat confused here, but I think the technical term for this is “totalitarian“, that is, an authoritarian state that additionally has high charisma, an emphasis on a public conception of the good, official ideology, low corruption, and high legitimacy. Totalitarianism has of course been tarnished by the 20th century regimes that were brutal by necessity as a result of their growth out of internal and external total war.

So I think Moldbug is wrong about authoritarian regimes not telling people what to think. The smart ones will definitely develop political and social persuasion to the highest economical level, even if they do have very high force capability.

So that leaves two more problems with Moldbug’s claim:

  1. Real regimes don’t and won’t have an absolute force monopoly that is immune to all challenge. As you approach infinite force, the persuasion needed goes down. Likewise as we break the assumption of infinite force, we realize that practical regimes need a lot of persuasive capability to help make up the power budget, not just for when persuasion is ideally the best approach.
  2. An official religion, or equivalent centralized persuasive apparatus, is useful, even necessary. It is not just a hack to deal with insufficient force authority, but an efficient solution to real problems in its own right.

But as usual, Moldbug’s rhetoric is a bit more over the top than what he is actually saying with his underlying model. In this case, the underlying model is the above considerations about the destructive Orwellian trade-offs you have to make if your regime is not forceful enough, or alternately trade-offs you don’t have to make if you can just arrest the opposition. About this core point he is right.