“Who are they and what do they want?” Reconceptualizing the importance of coordination

The recent Times article on coordination between anti-Trump leftists as a significant factor behind the 2020 election shenanigans has caused a lot of thought. Many “emergentists” are reassessing their prior assumptions. But they’re not the only ones. This article and another event, the half-forgotten “Coomercaust,” has made me reassess the practical difference between emergence and coordination.

Emergentists generally deny that powerful people explicitly coordinate to achieve goals. Rather, they believe that such people are merely following rational incentives. Previously I have represented their views in a way that makes them seem quite ridiculous: I have said that they deny that these people influence each other at all. But, in my recent experience, when pressed, emergentists readily admit the importance of elite inter-influence with regards to elite behavior. In other words, they recognize the thoughts of other elites as a significant factor in an individual decision maker’s incentive-environment.

In my essay “Centralization and Coordination,” I conceptualized coordination in a simple way, as the extent to which the elite influence each other’s decisions. In that paper, I used the term “formal” to describe the extent to which the ruling class coordinates explicitly. Evidently what emergentists mean by “coordination” is, per my parlance, “formal coordination”; the same goes for most coordinationists, myself included, at least before I wrote the aforementioned essay. And what the Times article demonstrates is the reality of formal coordination, not coordination in general.

So, does it matter whether coordination is formal? Are the outcomes different? It depends which outcomes we’re imagining. But two major questions are informed by the extent to which the elite is found to be formally coordinated. These are, upon reflection, the questions I’ve really been asking. They are questions people seem to care quite a lot about, in contrast to the extent of informal coordination, something emergentists discuss without hassle. The questions are: “who has power and what motivates their use of it?”

Centralization in my parlance is perfectly analogous to the first question, but coordination is only one line of evidence that can inform the answer to the second. This is unfortunate, since I realize now that I care little for coordination on its own but greatly for the question I was trying to answer through it, because the latter is a key to predicting the movement of history. The difference between a little or a lot of (formal/informal) coordination matters little except insofar as it informs us of the degree of centralization (i.e. who is in charge) and what motivates those who possess power. This is why emergentists will accept significant degrees of informal coordination without a fuss. Their issue is really with who has power and what motivates them, not with the extent to which they influence each other per se. Emergentists are generally bent on asserting that the powerful are motivated by relatively innocent, lizard-brain incentives such as status, greed, and sex. There is this obsession with painting the elite as these clownishly unconscious agents that don’t have any higher motivations. Accepting the idea of informal coordination does not disrupt this picture, because it can be painted as unconscious; in fact, some of the smarter emergentists rely on the idea of informal coordination to explain away events like coordinated social media censorship. The idea of formal coordination, however, is too much for the emergentist. The existence of formal coordination lends plausibility to naughty thoughts regarding what really motivates the elite. If the ruling class consciously comes together and plans, it becomes difficult to paint them as unconscious agents who are probably only motivated by basic desires. They might have complex dreams, like human instrumentality or destroying white people. Let me demonstrate by considering the Time article.

            Here’s the chain of events:

A weird thing happened right after the Nov. 3 election: nothing.

The nation was braced for chaos. Liberal groups had vowed to take to the streets, planning hundreds of protests across the country. Right-wing militias were girding for battle. In a poll before Election Day, 75% of Americans voiced concern about violence.

Instead, an eerie quiet descended. As President Trump refused to concede, the response was not mass action but crickets. When media organizations called the race for Joe Biden on Nov. 7, jubilation broke out instead, as people thronged cities across the U.S. to celebrate the democratic process that resulted in Trump’s ouster.

A second odd thing happened amid Trump’s attempts to reverse the result: corporate America turned on him. Hundreds of major business leaders, many of whom had backed Trump’s candidacy and supported his policies, called on him to concede. To the President, something felt amiss. “It was all very, very strange,” Trump said on Dec. 2. “Within days after the election, we witnessed an orchestrated effort to anoint the winner, even while many key states were still being counted.”

In a way, Trump was right.

What is the go-to emergentist explanation, particularly with regards to the actions of corporate America? It is precisely that those actions must have been motivated by basic incentives. Perhaps the business leaders wanted status (why anti-whiteness and wokeism is statusful is never addressed by these people). Perhaps there was informal coordination, and as a few true progressives turned against Trump, all felt the need to fall in line, without formal networks behind the scenes. Perhaps, for some unknown reason, all of these corporate leaders are true progressives. But it is inconceivable to the emergentist that there was an explicit plan to turn on Trump, because this implies the existence of a discrete class of rulers, who consciously plot amongst themselves and whose motivations could, consequently, plausibly transcend basic hind-brain impulses for power and wealth.Yet we find here that the corporate rulers who own the means of propaganda know it and use it consciously to their advantage:

 But behind the scenes, the business community was engaged in its own anxious discussions about how the election and its aftermath might unfold. The summer’s racial-justice protests had sent a signal to business owners too: the potential for economy-disrupting civil disorder. “With tensions running high, there was a lot of concern about unrest around the election, or a breakdown in our normal way we handle contentious elections,” says Neil Bradley, the Chamber’s executive vice president and chief policy officer. These worries had led the Chamber to release a pre-election statement with the Business Roundtable, a Washington-based CEOs’ group, as well as associations of manufacturers, wholesalers and retailers, calling for patience and confidence as votes were counted.

The Voting Rights Lab and IntoAction created state-specific memes and graphics, spread by email, text, Twitter, Facebook, Instagram and TikTok, urging that every vote be counted. Together, they were viewed more than 1 billion times. Protect Democracy’s election task force issued reports and held media briefings with high-profile experts across the political spectrum, resulting in widespread coverage of potential election issues and fact-checking of Trump’s false claims. The organization’s tracking polls found the message was being heard: the percentage of the public that didn’t expect to know the winner on election night gradually rose until by late October, it was over 70%. A majority also believed that a prolonged count wasn’t a sign of problems. “We knew exactly what Trump was going to do: he was going to try to use the fact that Democrats voted by mail and Republicans voted in person to make it look like he was ahead, claim victory, say the mail-in votes were fraudulent and try to get them thrown out,” says Protect Democracy’s Bassin. Setting public expectations ahead of time helped undercut those lies.

The emergentist delusion that the ruling class does not explicitly coordinate, using their forebrains to advance their agenda, is efficiently dispelled. Other questions are raised. We only know of this widespread, integral formal coordination by the grace of Times. What are they not willing to report? Furthermore, what are they lying about? The article is filled with blatant spin against evil Nazi demon Donalf Blumpfler. For instance:

Trump addressed the crowd that afternoon, peddling the lie that lawmakers or Vice President Mike Pence could reject states’ electoral votes. He told them to go to the Capitol and “fight like hell.” Then he returned to the White House as they sacked the building. As lawmakers fled for their lives and his own supporters were shot and trampled, Trump praised the rioters as “very special.”

The reality, of course, is that Trump denounced his supporters in an embarrassing Twitter video wherein he told his supporters to go home. The context of him calling his supporters special was roughly “You guys are special b-but please go h-home, o-ok-k??” No more words on the dishonesty, people either know this already or are incapable of understanding. The question remains, if they are willing to write shit like this, what else are they lying about? It is amazing that they have allowed us to see, so much more clearly than before yet still so obscured, the true extent of formal coordination behind the scenes of important, historical chains of events. We should not assume that what was not reported did not happen. What we should recognize is that, if formal coordination among the ruling class is so commonplace yet hidden that a shoddy Time article is a revelation, the potential for big secrets is much higher than what emergentists can mentally. It is no leap now to hypothesize that the ruling class may have explicitly coordinated to rig the election by manufacturing and counting fake ballots in certain areas.

We are even told by the article that conspirators knew how many Biden votes would come in during fraud-events (the massive, homogenous, 4 AM spikes in Biden ballots). In fact, they had been hard at work modelling how many votes Biden would need to beat Trump, a massive waste of energy if the goal was simply to net Biden as many real votes as possible. As a result of this knowledge, the conspirators even called off anti-Trump protests.

Election night began with many Democrats despairing. Trump was running ahead of pre-election polling, winning Florida, Ohio and Texas easily and keeping Michigan, Wisconsin and Pennsylvania too close to call. But Podhorzer was unperturbed when I spoke to him that night: the returns were exactly in line with his modeling. He had been warning for weeks that Trump voters’ turnout was surging. As the numbers dribbled out, he could tell that as long as all the votes were counted, Trump would lose.

So the word went out: stand down. Protect the Results announced that it would “not be activating the entire national mobilization network today, but remains ready to activate if necessary.” On Twitter, outraged progressives wondered what was going on. Why wasn’t anyone trying to stop Trump’s coup? Where were all the protests?

Podhorzer, meanwhile, was warning everyone he knew that polls were underestimating Trump’s support. The data he shared with media organizations who would be calling the election was “tremendously useful” to understand what was happening as the votes rolled in, according to a member of a major network’s political unit who spoke with Podhorzer before Election Day. Most analysts had recognized there would be a “blue shift” in key battlegrounds– the surge of votes breaking toward Democrats, driven by tallies of mail-in ballots– but they hadn’t comprehended how much better Trump was likely to do on Election Day. “Being able to document how big the absentee wave would be and the variance by state was essential,” the analyst says.

When confronted with formal coordination of this magnitude, my immediate thought is … “Why?!” From here we can see how the answer must ultimately be “emergent” from gene-environment interactions. But the proximal answer to “why” is informed by the extent of formal coordination. If the ruling class are so comfortable with explicit coordination, why would their goals be any less conscious and complex than their plans to fulfill them? There is no reason. The emergentist hypothesis that, for instance, the media pushing the idea that ballots would take a long time to count was simply due to the status-drives of journalists is a million miles from conscious, secretive, malicious plotting. But the reality, that careful, thoughtful, formal coordination among elites was the proximal cause of the media narrative, is only an inch away from malice. In their drive to defeat Trump the elite formally coordinated. But why did they want to defeat Trump? Do you think they don’t discuss this at their secret “economic” summits? Do you think they don’t explicitly coordinate their attitude towards things like Trump at their dinner parties? At some point desires are ultimately emergent, but there’s no reason to think that these people don’t explicitly discuss, oh, I don’t know, their plan for national demographics. In fact, they almost certainly do discuss these things. The elite are smarter than neoserfs, the adjacents ones I know love this type of discussion. The only question is if they’re honest with each other about their fundamental motivations. Is the Jewish inclination towards open borders emergent of Jewish ethnic hatred of white gentiles or something else, and are they, among themselves, honest about what that something is? Are they consciously aware? All are interesting questions, but one thing is certain: they all know they love open borders and they will plot explicitly to get there. They know they’re the impetus behind open borders. The Time article conspirers knew that the election would have gone to Trump without them. It was the thesis of the article:

As I was reporting this article in November and December, I heard different claims about who should get the credit for thwarting Trump’s plot. Liberals argued the role of bottom-up people power shouldn’t be overlooked, particularly the contributions of people of color and local grassroots activists. Others stressed the heroism of GOP officials like Van Langevelde and Georgia secretary of state Brad Raffensperger, who stood up to Trump at considerable cost. The truth is that neither likely could have succeeded without the other. “It’s astounding how close we came, how fragile all this really is,” says Timmer, the former Michigan GOP executive director. “It’s like when Wile E. Coyote runs off the cliff–if you don’t look down, you don’t fall. Our democracy only survives if we all believe and don’t look down.”

In light of there being high levels of explicit, formal coordination, there’s no reason to think that they aren’t aware of each other’s true motivations (Culture of Critique is very relevant here). If you listen they will even tell you.

Just, you know, only occasionally and with the right tone. Emergentists should appreciate such prudence as it is most plausibly emergent; using myself as a case study, my closest would know my real motivations while my media would generally obscure them, totally or with tone, were I a billionaire, knowing that not everyone agrees with me and it is my job to persuade them. Whereas the idea that if I had class and power I would not ever explicitly coordinate with my peers is absurd. Of course I would, of course we would plot to overthrow Trump, of course I would talk about how much I can’t wait for white people to be gone at dinner, of course we would discuss good ways to spin things in the media that my friends and I help control. Of course it would slip out that I really do hate “whiteness” and that this isn’t just for status, and no, in Nazi Germany I would not be a Nazi because I hate white people because I am not a white person and Nazi Germany happened or something.

What of the white elite? Perhaps they are simply weak-willed. That seems to be MacDonald’s thesis, anyway. I cannot do this question justice here, because it is a topic for serious investigation.

The point is that motivations allow for prediction. If it’s all just a status shell-game, anti-whiteness might be able to be replaced. But if the elite are motivated by white hatred at their core, then anti-whiteness is a property of our elite. They cannot be changed.

In fact, I advance that “who are they and what do they want?” is THE predictive question. Let me illustrate with the Coomercaust.

“The Coomercaust” refers to the chain of events that occurred in late 2020 that ultimately ended with Pornhub removing the all “amateur” content, the vast majority of their videos. On the surface, it went like this: Nicholas Kristof wrote an article published in the New York Times that was deceivingly titled “The Children of Pornhub.” A day or two later, Mastercard and Visa deplatformed Pornhub. Shortly after, despite rumblings that Pornhub may shift to Bitcoin or create their own payment processor, Pornhub launched the Coomercaust.

This situation lends itself to differential analysis along the lines of emergentism vs. coordinationism. The emergentist narrative is something along the lines of, “Kristof, seeking clout, decided to Think of the Children. Mastercard and Visa, feeling pressured and seeking accolades, deplatformed Pornhub. Pornhub folded rationally. The end.” If this were the case, we might infer that Thinking of the “Children” is a human constant, and wherever it happens, similar patterns as observed here will emerge. It furthermore may be implied that the attitude of the mass is consequential to some degree. But if we consider the Time article, and postulate that there was formal coordination that preceded this seemingly-natural chain of events, we should make different predictions. Thinking of the Children may have merely been a cover to obscure the true motivations of Mastercard and Visa. For instance, the people who run those companies could own shares in a “service” like Onlyfans. Purging “free” amateur pornography via sex crime hysteria drives coomers towards such paid websites. Knowing formal coordination to be the case here can imply secret, intelligent motivations, which lead to different predictions than those made under emergence. But suppose there was coordination, but on an off chance they only coordinated because they really do Think of the Children. In this case, our predictions are no different than with emergence. Therefore, “who are they and what do they want” is THE predictive question. Coordination only informs our answer to this question.

Are women immature relative to men? Assessing female mental age with regards to that of men and boys.

It’s been said in the PUA reddit community TheRedPill that women are the “oldest 18 year old” in the house, that is, that they have the mind of an 18 year old man. What does this mean? Others, such as Youtuber boogie2988 have claimed they have the “mind of a 14 year old.” How can this sentiment be quantified and tested? Consider a 12 year old: it’s commonly thought that people this age are, generally, deficient in cognitive abilities that are relevant to social standing and day to day life. These can be broken down into three (not necessarily completely distinct) categories that seem to cover the vast majority of cognitive differences: emotional maturity, or control over and adaptive value of emotions; judgment, or the ability to avoid bad decisions; and rationality, or the complexity of thought. Here, theRedPill’s claim about the minds of women will be investigated: how women compare to men in terms of emotional maturity, judgment ability, and rationality, and in turn how they compare to boys of different ages.

Let judgement be operationalized as “the ability to avoid harmful decisions.” There are then a number of subcomponents that can be identified that make it easier to recognize what kind of data might indicate differences in judgment. One is risk aversion, the extent to which risk taking is optimized. Too much risk, and you’re destroyed. Too little, and you’re dominated by competitors. The risk-taking component of judgment intuitively seems to be important in regards to financial decisions and investments. Further, the ability to make healthy long term decisions, such as whether or not to take a drug, or what field to start a career in, or how many children to have, is also probably influenced by risk taking. One task that neatly measures judgment, and risk taking in particular, is the Iowa Gambling Task, in which participants have to choose from a number of decks of varying risks and in which they are punished or rewarded accordingly. Women consistently underperform in comparison to men to the tune of d =~ 0.5 (Singh 2016)(Reavis & Overman 2001)(Bolla et al 2004)(Van den Bos et al 2013). In comparison, teenagers are disparaged because they perform the same as men, with the performance gap on the task being statistically insignificant and d < 0.1. However, 10-11 year old boys underperform men to the same degree as women, as d = 0.48 (Cauffman et al 2010).

Another measure that is adjacent to judgment and risk taking is known as selective attention or impulse control.  “Examples of psychological measures of inhibitory control include the Stroop task, Simon task, Flanker task, antisaccade tasks, delay-of-gratification tasks, go/no-go tasks, and stop-signal tasks”. These are designed to measure the ability to “control one’s attention, behavior, thoughts, and/or emotions to override a strong internal predisposition or external lure, and instead do what’s more appropriate or needed” (Diamond 2013). However, not all tasks are created equal. The go/no-go task, for example, does not correlate with the Simon task, and it’s easy to see why. The go/go-no involves split-second motor accuracy in regards to a go or no-go signal. The Simon task, on the other hand, involves the ability to executively depress the tendency to get distracted by irrelevant stimuli when completing a task. The former might be said to measure the impulse-accuracy of the underbrain while the latter probably has more to do with executive function. Women score maturely on the go/no-go task and the similar stop-signal task. However, women underperform on the Simon task at d = 0.39 (Stoet 2017)(Merritt et al 2005) as they are more easily distracted and suffer greater performance losses after distraction. This is most similar to the gap between men and 13-15 year old boys (Couperus 2011). Narrowing it down, women can be estimated to perform equivalently to males of about 14 years on selective attention loaded tasks. The fact that women underperform on more complex tasks may reflect equal processing speed with men, as there is little to no IQ gap between the sexes. It would seem, however, that processing quality does have some differences.

Hedonism and emotional maturity also intersect with judgment and are further linked. The former is defined as susceptibility to momentary whims and may interact with influenceability in women. The latter is defined as general rigor of and control over emotions. Less control or less rigorous emotions may lead to more emotional decisions (states not involved in most lab tests), and higher rates of traumatization per incident. Hedonism and emotional maturity, likely not completely distinct concepts, plausibly both interact in susceptibility and severity to drug addiction, for instance. Of all the domains of cognition, children, at least older children, are in particular thought to be deficient in emotional maturity and to act more hedonistically and materialistically. They are thought to have less control over their feelings and to have more fragile and easily influenced emotional states compared to men. How do women hold up?

Women impulse buy much more than men per self report (Segal & Podoshen 2012), and roughly match the self reports of 14 year old boys on the same scale (Lin & Chen 2012). This is striking for two reasons: social acceptability of impulse between the two groups and differential financial pressures. If I had to bet, women are underreporting impulse buying more so than boys because it is seen as more irresponsible and negligent at older ages, in part because a lot of women are responsible for children and necessary costs, while most 14 year olds have mostly, if not totally, expendable income. 14 year old boys are probably in a better position to impulse buy and suffer from less embarrassment when reporting their behavior than women. The fact that these two groups report roughly the same rates of impulse buying is significant. Furthemore, less impulse control might lead to being more influenceable or vice versa. While they don’t report  more dependence on material goods for happiness, women do report being more invested in vanity than men at d = .3 and in being more impacted by advertisements and friend’s consumption expectation at d = .4 (Keech et al 2019). Women similarly report being more self conscious at d = .3 (Keech et al 2019). Boys are similarly reported to be elevated in relation to men on scales of self consciousness and influenceability (Lin & Chen 2012) (Pechmann et al 2005) (Chaplin & John 2017), but the research in the advertising field is too poor and too averse to straight forward discussion of early vs. mid vs. late adolescent vs. adult effect sizes (because straightforward discussion reveals what has been shown here) for a male-age to be clearly assigned to women based on this data in particular. However, it is almost certainly between 12 and 15. Moving on, however, social anxiety and conformity data seems closely related to financial influenceability and self consciousness data. Women seem to differ from men in social anxiety at d = .3, suggesting it overlaps with self consciousness significantly. Furthermore, 20 year old women report in one study as much social anxiety as 10th grade boys (Puklek & Vidmar 2000). In another study, 15 or 16 year old women reported as much social anxiety as boys at their peak anxiety-age: 13 or 14 (Ranta et al 2007). The data is not perfect but it consistently suggests that grown women experience self consciousness at similar levels as young teen boys. Conformity data backs this up. A metaanalysis found women conform more than men at d = .3 (Hyde 1990). The data on conformity and age is not as nice as it could be, but in light of what has already been found it essentially replicates by elevated conformity scores of pubertal boys. In one study, 20 year old women conformed as much as 15 year old boys (Costanzo & Shaw 1966). Another study found elevations in conformity for males around 13-15 and reported that “girls have less behavioral independence from parents than boys” (Berndt 1979). Overall, women seem to impulse buy, conform, and experience self consciousness at rates elevated to men and most similar to boys in puberty who are 12-15 years old. These behaviors indicate underlying susceptibility to momentary whims (hedonism) and increased emotional dependency (meaning less independent control over emotions, i.e. emotional maturity). They also reflect and are reflected by deficiencies in selective attention and overall judgment. Furthermore, they may also reflect an overall lower complexity of thought (rationality) in women, compared to men, where it is hypothesized that women think more in black and white and are less inclined to think critically/independently.

Moving on, addiction, traumatization, and mental illness data also reflect differences in susceptibility to momentary whims and control and rigor of emotions. Children are often thought to be more mentally fragile, unable to handle trauma and suffering as well as adults. Furthermore, “adolescents” are frequently said to be at an elevated risk of addiction, leading to calls for high age restrictions (Pechmann et al 2005). Women, when compared to men, also show elevated susceptibility to addiction as well: “Women exhibit more rapid escalation from casual drug taking to addiction, exhibit a greater withdrawal response with abstinence, and tend to exhibit greater vulnerability than men in terms of treatment outcome” (Becker 2016). Another study reported that recovering female addicts crave their drug more than recovering men (Kennedy et al 2013). This is interesting because for some drugs, such as alcohol, there are more male abusers. The difference is possibly caused by differences in social pressures — men are more likely to have hard times and face less shame for bouts with drugs. Despite this difference in overall use, however, the female brain shows a similar susceptibility to drugs as found in kids, with addiction coming more easily and being harder to kick. It is difficult to quantify this tendency as to compare women directly to boys since most of the data is qualitative and there are serious social factors that influence overall drug usage rates in a population (but not addiction susceptibility after exposure). Furthermore it is considered unethical to administer addictive drugs in a lab to test addiction susceptibility. One study, however, replicated previous discussions by reporting “Unlike men, the initial use of heroin by women was highly influenced by a man [in women]” (Hser & McGlothlin 1987). Differences in addiction susceptibility between the sexes are almost certainly biological, relating to structural brain differences and hormones (Bobzean et al 2014). Suffice it to say that, in regards to addiction susceptibility, women most likely show similar susceptibility as pubescent males, owing to previous discussions.

Women are also more susceptible to traumatization (per incident) than men (Breslau & Anthony 2017). One study reported “Women were found to be at significantly increased risk for PTSD following exposure to serious trauma (odds ratio approximately 5), even when sexual trauma–which predominates in women–was excluded (odds ratio approximately 3)” (Stein et al 2000). This difference is extreme: 70 year old women (the least likely age group of women to have PTSD) are as likely to have PTSD as 13 year old boys, the latter being far more likely to be the victim of serious violent crime (Ditlevsen & Elklit 2010) (Perkins 1997). This indicates a serious gap in female and male emotional rigor and independent emotional control, consistent with previously reviewed data. Other mental illness data replicates this. Teenagers and women have about a 10% chance of being diagnosed with Major Depression in any given year, while men only have a 4% chance (Lieber 2020) (RAMH 2020). Women are also twice as likely as men to have an anxiety disorder (RAMH 2020). Depression and anxiety are doubtlessly intertwined with social factors, but nonetheless, the idea replicates. This all indicates that women are more emotionally fragile and likely to break down from stress than men, similarly (or worse) to boys of about 13 and under.

Judgment and emotional maturity have been thoroughly examined, leaving rationality, the complexity of thought. A subset of complexity of thought can be indicated by a person’s susceptibility to propaganda, as that indicates the extent to which they think critically, deeply, and independently. Women have already been shown to be roughly as susceptible to propaganda and advertising as 14 year old boys. Another well known difference between men and women relates to complexity of thought. Men are much more likely to be “systemizers”, systemizing being “the drive to analyse and explore a system, to extract underlying rules that govern the behaviour of a system; and the drive to construct systems” (The Guardian). Systemization increases with age in both sexes, making this deficit a neotenous trait (Auyeung et al 2012). If systemizing is the drive to analyse and think deeply, then other data will probably show that women as a group tend to think more in black and white. Borderline Personality Disorder data shows this: a symptom of it is black and white thinking, and 90% of patients are women (Daros et al 2013). Both this and systemizing data replicate the finding that women have lower complexity of thought, a trait thought to be common in children of quite young ages.

Overall, the data indicates that, relative to men, women roughly have the “mind of a 13 or 14 year old.” More generally, women tend to score on tasks related to cognitive functioning as pubescent boys do, whether that be children who have just started puberty (10-12) or boys in the midst of it (12-14). From about 15 onward, men outperform women, having reached their maturity after a longer time of development than women. Indeed, this result was anticipated by sex differences in puberty. Women reach adult physiology around 13 or 14, while men do the same around 15 or 16. Hence men have about 2 more years of development than women. 15 or 16 year old males consistently demonstrate equivalent cognitive capacities to older men. It is intuitive that, their development being “cut short”, women would forever demonstrate cognitive capacities similar to those of boys the age at which women stop maturing, 13 or 14. Furthermore, bodily neoteny anticipated the results: women look and sound more childish. Women are smaller, have less hair, more childish faces, have higher pitched voices, etc. Their physical strength is roughly equal to that of 12 or 13 year old boys. It is intuitive that behavior might mirror these traits, and here it’s been found that it does.

However, there are still uncertainties. For instance, it could be debated as to what causes the above-evidenced performance gaps. Tasks on which women outperform men could be discussed. Women tend to have an aversion to risk taking, despite it seeming that they are more impulsive. This could lead to greater judgment in some situations in which risk is intolerable. Furthermore, even if women have “younger” minds than men, experience and learning might play a role in behavior and make women more “mature” than pubescent boys in many ways. The policy implications of the findings here are open to these contention points and ethical considerations.

Sources

  1. Singh, V. (2016). Sex-differences, handedness, and lateralization in the Iowa Gambling Task. Frontiers in psychology, 7, 708. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4885871/
  2. Reavis, R., & Overman, W. H. (2001). Adult sex differences on a decision-making task previously shown to depend on the orbital prefrontal cortex. Behavioral Neuroscience, 115(1), 196–206. doi:10.1037/0735-7044.115.1.196 https://psycnet.apa.org/doiLanding?doi=10.1037%2F0735-7044.115.1.196
  3. K.I. Bolla, D.A. Eldreth, J.A. Matochik, J.L. Cadet, Sex-related Differences in a Gambling Task and Its Neurological Correlates, Cerebral Cortex, Volume 14, Issue 11, November 2004, Pages 1226–1232, https://doi.org/10.1093/cercor/bhh083
  4. Van den Bos, R., Homberg, J., & de Visser, L. (2013). A critical review of sex differences in decision-making tasks: Focus on the Iowa Gambling Task. Behavioural Brain Research, 238, 95–108. doi:10.1016/j.bbr.2012.10.002 https://www.sciencedirect.com/science/article/pii/S0166432812006419?via%3Dihub
  5. Cauffman, E., Shulman, E. P., Steinberg, L., Claus, E., Banich, M. T., Graham, S., & Woolard, J. (2010). Age differences in affective decision making as indexed by performance on the Iowa Gambling Task. Developmental psychology, 46(1), 193. http://psych.colorado.edu/~mbanich/p/Cauffman2010_age_differences.pdf
  6. Stoet, G. (2017). Sex differences in the Simon task help to interpret sex differences in selective attention. Psychological Research, 81, 571 – 581. https://www.semanticscholar.org/paper/Sex-differences-in-the-Simon-task-help-to-interpret-Stoet/31711e4006429777390d6fe5c79b67f6d6048991
  7. Diamond, A. (2013). Executive functions. Annual review of psychology, 64, 135-168. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4084861/
  8. Merritt, P., Hirshman, E., Wharton, W., Devlin, J., Stangl, B., Bennett, S., & Hawkins, L. (2005). Gender differences in selective attention: Evidence from a spatial orienting task. Journal of Vision, 5(8), 1000-1000. https://jov.arvojournals.org/article.aspx?articleid=2131748
  9. Couperus, J. W. (2011). Perceptual load influences selective attention across development. Developmental Psychology, 47(5), 1431. https://www.researchgate.net/publication/51234267_Perceptual_Load_Influences_Selective_Attention_Across_Development
  10. Segal, B., & Podoshen, J. S. (2012). An examination of materialism, conspicuous consumption and gender differences. International Journal of Consumer Studies, 37(2), 189–198. doi:10.1111/j.1470-6431.2012.01099.x  https://doi.org/10.1111/j.1470-6431.2012.01099.x
  11. Lin, Y.-H., & Chen, C.-Y. (2012). Adolescents’ Impulse Buying: Susceptibility to Interpersonal Influence and Fear of Negative Evaluation. Social Behavior and Personality: An International Journal, 40(3), 353–358. doi:10.2224/sbp.2012.40.3.353 https://www.ingentaconnect.com/content/sbp/sbp/2012/00000040/00000003/art00001;jsessionid=49dldks9g37e.x-ic-live-03
  12. Keech, J., Papakroni, J., & Podoshen, J. S. (2019). Gender and differences in materialism, power, risk aversion, self-consciousness, and social comparison. Journal of International Consumer Marketing, 1–11. doi:10.1080/08961530.2019.1647125 https://www.tandfonline.com/doi/abs/10.1080/08961530.2019.1647125?journalCode=wicm20
  13. Keech, J., Papakroni, J., & Podoshen, J. S. (2019). Gender and differences in materialism, power, risk aversion, self-consciousness, and social comparison. Journal of International Consumer Marketing, 1–11. doi:10.1080/08961530.2019.1647125 https://www.tandfonline.com/doi/abs/10.1080/08961530.2019.1647125?journalCode=wicm20
  14. Pechmann, C., Levine, L., Loughlin, S., & Leslie, F. (2005). Impulsive and self-conscious: Adolescents’ vulnerability to advertising and promotion. Journal of Public Policy & Marketing, 24(2), 202-221. https://www.jstor.org/stable/30000660?seq=1
  15. Puklek, M., & Vidmar, G. (2000). Social anxiety in Slovene adolescents: Psychometric properties of a new measure, age differences and relations with self-consciousness and perceived incompetence. European Review of Applied Psychology, 50(2), 249-258.
  16. Ranta, K., Kaltiala-Heino, R., Koivisto, A.-M., Tuomisto, M. T., Pelkonen, M., & Marttunen, M. (2007). Age and gender differences in social anxiety symptoms during adolescence: The Social Phobia Inventory (SPIN) as a measure. Psychiatry Research, 153(3), 261–270. doi:10.1016/j.psychres.2006.12.006 https://www.sciencedirect.com/science/article/abs/pii/S0165178106003969
  17. Chaplin, L. N., & John, D. R. (2007). Growing up in a material world: Age differences in materialism in children and adolescents. Journal of consumer research, 34(4), 480-493.http://assets.csom.umn.edu/assets/93681.pdf
  18. Hyde, J. S. (1990). Meta-analysis and the psychology of gender differences. Signs: Journal of women in culture and society, 16(1), 55-73. https://www.jstor.org/stable/3174607?seq=1#page_scan_tab_contents
  19. Costanzo, P. R., & Shaw, M. E. (1966). Conformity as a Function of Age Level. Child Development, 37(4), 967. doi:10.2307/1126618 https://www.jstor.org/stable/1126618?origin=crossref&seq=1
  20. Costanzo, P. R., & Shaw, M. E. (1966). Conformity as a Function of Age Level. Child Development, 37(4), 967. doi:10.2307/1126618 https://www.jstor.org/stable/1126618?origin=crossref&seq=1
  21. Berndt, T. J. (1979). Developmental changes in conformity to peers and parents. Developmental Psychology, 15(6), 608–616. doi:10.1037/0012-1649.15.6.608 https://psycnet.apa.org/doiLanding?doi=10.1037%2F0012-1649.15.6.608
  22. Becker, J. B. (2016). Sex differences in addiction. Dialogues in clinical neuroscience, 18(4), 395.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5286725/
  23. Hser, Y. L., Anglin, M. D., & McGlothlin, W. (1987). Sex differences in addict careers. 1. Initiation of use. The American journal of drug and alcohol abuse, 13(1-2), 33-57.https://www.tandfonline.com/doi/abs/10.3109/00952998709001499
  24. Bobzean, S. A., DeNobrega, A. K., & Perrotti, L. I. (2014). Sex differences in the neurobiology of drug addiction. Experimental neurology, 259, 64-74. https://www.sciencedirect.com/science/article/abs/pii/S0014488614000351?via%3Dihub
  25. Kennedy, A. P., Epstein, D. H., Phillips, K. A., & Preston, K. L. (2013). Sex differences in cocaine/heroin users: drug-use triggers and craving in daily life. Drug and alcohol dependence, 132(1-2), 29-37. https://www.sciencedirect.com/science/article/abs/pii/S0376871612005017
  26. Breslau, N., & Anthony, J. C. (2007). Gender differences in the sensitivity to posttraumatic stress disorder: An epidemiological study of urban young adults. Journal of abnormal psychology, 116(3), 607. http://www.epi.msu.edu/janthony/documents%20up%20to%202010/K05/Breslau_Gender%20differences%20in%20the%20sensitivity%20ptsd.pdf
  27. Stein, M. B., Walker, J. R., & Forde, D. R. (2000). Gender differences in susceptibility to posttraumatic stress disorder. Behaviour research and therapy, 38(6), 619-628. https://www.ncbi.nlm.nih.gov/pubmed/10846810
  28. Ditlevsen, D. N., & Elklit, A. (2010). The combined effect of gender and age on post traumatic stress disorder: do men and women show differences in the lifespan distribution of the disorder?. Annals of general psychiatry, 9(1), 32. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917414/
  29. Perkins, C. A. (1997). Age patterns of victims of serious violent crime. US Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.https://www.bjs.gov/content/pub/pdf/apvsvc.pdf
  30. Lieber A.(2020). https://www.psycom.net/depression.central.major.html
  31. RAMH (2020). https://ramh.org/guide/gender-differences-in-mental-health/
  32. Baron-Cohen, S. (2009). The essential difference: Male and female brains and the truth about autism. Basic Books.
  33. Auyeung, B., Allison, C., Wheelwright, S., & Baron-Cohen, S. (2012). Brief report: development of the adolescent empathy and systemizing quotients. Journal of autism and developmental disorders, 42(10), 2225-2235.https://link.springer.com/article/10.1007/s10803-012-1454-7
  34. The Guardian. https://www.theguardian.com/life/table/0,,937441,00.html
  35. Daros, A. R., Zakzanis, K. K., & Ruocco, A. C. (2013). Facial emotion recognition in borderline personality disorder. Psychological Medicine, 43(9), 1953-1963. https://www.researchgate.net/figure/Demographic-and-clinical-characteristics-of-borderline-personality-disorder-patients-in_tbl1_265008346

Lessons from the election

First, let me gloat that my prediction was mostly correct. Biden “won.” I accurately predicted the will of the elite. Whether or not Trump was in on 2016 or if he’s just a useful idiot remains to be seen: if he concedes, he’s in on it. If he hangs on to the end, he’s probably not.

What I got wrong was in thinking that people would actually vote for Biden. They didn’t. That settles it: “the masses” is like a dog, not a rock. If you poke it with a stick it gets mad. But it has no memory and forgives if you give it a treat. If Biden assumes office, no one will ever hold the fraudsters accountable for their actions.

We have also learned that the elite are idiots. More specifically, ~120-130 IQ and loosely coordinated, probably not strictly coordinated. Only by the grace that the masses are practically retarded is this fraud business going according to plan. It’s hyperobvious and if smarter people had power against the perpetrators, there’d probably already be people in jail right now.

It was a sloppy, emergency job. Statistics alone prove Trump is the rightful winner of the election. In particular that is Benford’s law and the discrepencies between legislative and presidential ballots. The suspicious timing of ballot counting is icing on the cake.

I’ve also been graced with a new stupidity detecting heuristic (the others are HBD and Jewish influence). Fraud deniers are some of the cringiest bunch I’ve seen. The exact epitome of pseud. So many rationalist and wignat griftors are saying garbage like “well technically it says in the lugenpresse that Benford’s law doesn’t 100% prove fraud and nobody has paid me shekels to learn about the law and the current evidence so it’s still a mystery as to whether or not fraud happened.” No. It happened. LMAO.

Biden will win.

95% chance, and if he doesn’t, there’ll most likely be an economic crisis during Trump’s second term. Either way, the right looks terrible despite Trump actually being a liberal.

My main evidence is the media’s behavior and that of the Democrats. Hillary was a terrible candidate. Biden and Diversity Token are much better. Trump has also cooled himself down too much, he’s very low energy. And he handles the coronavirus thing as terribly as possible.

His polls are historically terrible but the media isn’t hyping that up like in 2016. Thus no equal and opposite reaction.

He’s failed white people. We’re screwed either way, and he deserves to lose. Depending on a few things he may be devastated or he may not be. And either way, I predict the right will look terrible.

We just can’t win, can we? We need more money. The only guy willing to stand up for us is a total fraud.

Does academia discriminate against geniuses, and could this explain the decline of science?

[This was inspired by a recent Sean Last article.]

Last makes a compelling case that scientific progress is slowing in the West. To summarize, evidence shows that the rate of new discoveries is falling.

The productivity of researchers is falling while the number of researchers across fields balloons.

Last also makes a compelling case that this isn’t simply due to scientific problems getting harder. He points out that the rate of progress actually increased for quite a long time before beginning to decline, and that it appears to have begun declining across all fields starting around the same time. If this were happening due to low hanging fruit being all picked up, then fields should begin decline at different times. Instead, progress seems to have started to slow universally in the mid to late 20th century.

Next, he shows that most progress is accomplished by a small group of geniuses. If the problems haven’t gotten harder, then the pool of geniuses must have gotten smaller.

There are a number of reasons why this may be. Hypotheses range from the idea that average IQ has declined, and therefore there are less geniuses born, to blaming a myriad of broad social factors. I wish to examine a different hypothesis: that there is less scientific progress not strictly because there are less geniuses born, but because, at least in part, the modern Western academy discriminates against potential genius, selecting against the relevant traits and thereby denying geniuses entry into the ivory tower and access to modern research equipment, condemning them to a life of wage-slaving.

To know if geniuses are discriminated against, we have to know what the markers of genius are. A genius is defined simply as a person with extra-ordinary scientific output, whether that be selective output of high theoretic quality or rapid output of high fact-content quantity.

Last shows that being a gentile white male is an extremely strong predictor of being a genius.

Disregarding theories of “systemic oppression,” the reason why being a white male is such a strong predictor of being genius is probably that white males are more likely to possess genius behavioral traits. The primary trait everyone thinks of is IQ, the threshold of which to be a genius seems to be 125 at a bare minimum and 145 on average (some will claim higher, but I’m extremely skeptical that said IQs can be reasonably assigned by the tests used)( The Making of a Scientist. (Roe 1952) Women have a narrower intelligence distribution, resulting in less geniuses (https://heterodoxacademy.org/the-greater-male-variability-hypothesis/). Nonwhite, non-Asian races have less geniuses due in part to have lower average intelligence.

Intelligence alone is not sufficient for genius. Ivy league schools are sufficient proof of this: they’re filled with the brightest of the bright, but few of these pursue even simple original ideas. Many scientists posses sufficient intelligent, if not most; the average IQ of one is about 130.

Last sums this up:

This is empirically attested to by Lewis Terman’s famous study of gifted children. Terman collected data on 1528 Californian children who had a mean IQ of 151. The lowest IQ of the bunch was 135. IQs of 170 or higher were found for 77 of the subjects. Relative to the general population, the rate of extreme success among this group was extraordinary. In a 35 year follow up, 77 of the participants had been included in American Men of Science, a list of America’s top scientists, and 33 were included in America’s Who’s Who (Eysenck, 1995). Eventually, two went on to win noble prizes (Feist, 2014). However, the vast majority of the 1,528 subjects studied did not go on to become geniuses, and so these results also make clear that most people with extremely high IQs do not go on to achieve a notable degree of eminence.

What differentiates good from bad scientists then? Last says:

* Van Zelst and Kerr (1954) reported significant partial correlations between productivity and describing oneself as “argumentative,” “assertive,” and “self-confident.”

* Bachtold and Werner (1972)  administered Cattell’s 16 Personality Factor to 146 women scientists and found that they were significantly different from women in general on nine of the 16 scales, including dominance (Factor E) and self-confidence (Factor O).

*  Feist (1993) reported a structural equation model of scientific eminence in which the path between observer-rated hostility and eminence was direct, and the path between arrogant working style and eminence was indirect but significant.

* The scientific elite also tend to be more aloof, asocial, and introverted than their less creative peers. In a classic study concerning the creative person in science, Roe (1952, 1953) found that creative scientists were more achievement oriented and less affiliative than less creative scientists.

* Chambers (1964) reported that creative psychologists and chemists were markedly more dominant, ambitious, and self-sufficient, and had more initiative than their less creative peers.

*  Helson (1971) compared creative female Mathematicians with less creative female mathematicians, matched on IQ. Observers blindly rated the former as having more “unconventional thought processes,” as being more “rebellious and nonconforming,” and as being less likely to judge “self and others in conventional terms.”

* Wilson and Jackson (1994) reported that both male and female physicists were more introverted and conscientious than nonscientist controls.

That ends my summary of Last’s post. Further evidence that H. J. Eysenck described in his 1995 book Genius is that a higher than normal rate of psychotic illness, and a higher than normal rate of psychopathic traits (which include low Agreeableness and low Conscientiousness but also more subtle characteristics) is found among geniuses.

It would appear in Big 5 terms  that geniuses are more disagreeable, introverted, and open than the typical person or scientist. This lines up perfectly with Dutton and Charlton’s Endogenous Personality theory (see The Genius Famine):

What is the Endogenous Personality? And why is he so important? In a nutshell, we argue that the Endogenous personality is the type of a potential genius – a compound of abilities and attitudes, of intelligence and innerness. As a strong generalization: the true geniuses are Endogenous personalities; and it is from Endogenous personalities that geniuses arise. The Endogenous personality is the ‘inner’ Man; a person whose outlook on life is ‘inward.’ He is inner-directed, inner-driven, inner-motivated; one who uses inner modes of thinking, inner evaluations, in-tuition; one who is to a high degree autonomous, self-sufficient; one who is relatively indifferent to social pressures, influences and inducements. He stands in stark contrast to the Exogenous personality; that is, to most people. The Exogenous Personality is orientated toward the environment, particularly the social environment. These are people who want more than anything else social (including sexual) status, worldly success; people whose perceptions are directed outwards and who try to align their behaviour with group norms. When described in such terms, the Endogenous personality might appear anti-social, uncooperative, a dreamer, not the kind of person we might wish to have to deal with on a regular basis. We would probably be accurate in perceiving the Endogenous Personality in this negative way. We probably wouldn’t want to go for a drink with him, let alone be friends with him. But he is important; he is very important. Because the Endogenous personality is the archetypal ‘genius.’

Asians, then, produce less geniuses,  despite their average intelligence, because they have a higher average for temperamental obedience or “Exogeny.” They are stereotypically more agreeable and most of all less open than the average white person, who still falls far short on these traits compared to a genius.  The evidence is producing strong convergence.

Now that we have a theory of what genius is, we can test whether or not it is discriminated against by the academy. Charlton has extensively argued that it is (Why are modern scientists so dull?, 2009).

Question: why are so many leading modern scientists so dull and lacking in scientific ambition? Answer: because the science selection process ruthlessly weeds-out interesting and imaginative people. At each level in education, training and career progression there is a tendency to exclude smart and creative people by preferring Conscientious and Agreeable people. The progressive lengthening of scientific training and the reduced independence of career scientists have tended to deter vocational ‘revolutionary’ scientists in favour of industrious and socially adept individuals better suited to incremental ‘normal’ science.

A quick look at admissions practices indicates this argument holds water. White males are severely discriminated against for their sex and race in favour of people who have virtually no chance of being a genius.

This is, of course, relatively new. The academy may have not ever been perfect, and for our hypothesis to be supported we must show that there has been a change in the last 100 years or so in how much potential geniuses are selected against. The racial and sexual discrimination alone is a major piece of evidence in favour of our hypothesis, as it simply didn’t occur in the past when Western researchers were almost entirely white males. Now potential-genius white men are told there isn’t any room for them in the Ivory tower — a quick peak in reveals that the reason why is often that it’s filled with women and non-genius-producing races instead!

^ That’s Harvard! Virtually 70% of their hires are guaranteed to not be geniuses.

Nationwide, more than half of the hires into the ivory tower are almost certain to not be geniuses. It would be insane to think potential geniuses aren’t getting cut out of the equation (most people with STEM PhDs don’t get into the academy) given the presented evidence alone. But when we consider what kinds of white males are favored, denial of discrimination against potential geniuses becomes even more bizarre.

In the past, entrance exams were a much bigger deal. This can be gleamed from sufficient historical reading. Now, universities are obsessed with things that might be antithetical to genius – volunteering, social extracurriculars, and non-genius personality traits in general that have nothing positive to do with academic abilities. The personal experience of anyone born in the last 40 years should be sufficient to attest to this, but more evidence exists.

In 2018 Harvard was sued, and in the   “PLAINTIFF’S MEMORANDUM OF REASONS IN SUPPORT OF ITS MOTION FOR SUMMARY JUDGMENT,” the college was exposed for discriminating against the genius-phenotype directly:

Each applicant is given four component ratings and an overall score by the Admissions Office: (1) academic; (2) extracurricular; (3) athletic; (4) personal; and (5) overall.  …  The last, “overall” score, is not a formulaic compilation of the scores in the other ratings. Instead, Harvard instructs readers to assign the score by “stepping back and taking all the factors into account and then assigning that overall rating.” … [Personal is] a “subjective” assessment of such traits as whether the student has a “positive personality” and “others like to be around him or her,” has “character traits” such as “likability … helpfulness, courage, [and] kindness,” is an “attractive person to be with,” is “widely respected,” is a “good person,” and has good “human qualities.”

Potential geniuses are bound to fail on this abomination of an admissions standard independent of race or sex. Such types are introverted and disagreeable, the opposite of the feminine type they select for in the personal rating; potential geniuses are not going to be kind, agreeable, “positive,” “attractive” to the exogenous judges, and so on. They are also less likely to engage in athletics or extracurriculars, especially to a significant degree. Such things will simply not interest them as much as reading a textbook on quantitative genetics. Being disagreeable, potential geniuses will be less susceptible to pressures to spend more time than they already do pleasing their “superiors.” The type who has stellar, resume fit extracurricular and sportsball achievements is the type that does things because they’re the “in” thing. This is the opposite to the type that revolutionizes science, yet this is what is getting picked while potential geniuses are left out to dry. It’s no wonder that scientific progress is slowing.

Sadly in the literature there is a daunting lack where bountiful research exposing the phenotype of the average elite admittee or tenure-track hiree of 2020 should be. No matter, because we have sufficient evidence as it is that this phenotype is very domesticated, the opposite of potential-genius.

The question as to what caused all of this shall be left for another time. Suffice it to say that the hypothesis is supported: in being more concerned with diversity, “kindness,” “likability,” “diversity,” token-economy extracurriculars, and athletics than creativity, originality, autonomy, and dignity, academia does discriminate against potential geniuses, contributing to the downfall of Western science.

An examination of the causes behind the black-white IQ gap

Psychology’s biggest success story is perhaps that branch of applied statistics known as intelligence research. An IQ test is a capable predictive tool, unlike whatever is put out by “social psychologists” and the like. For example, a meta-analysis with a combined sample size of over 90,000 people found that IQ is the most powerful known predictor of subjects’ levels of occupation, education, and income (Strenze 2007).

There is a gap between the average scores of white test takers and black test takers of about one standard deviation, or 15 points. This black-white IQ gap has been detectable since the 1960’s. Occasionally some claim it has shrunk, generally through the use of tests with poor predictive validity and non-random samples. Regardless, a 2012 meta-analysis conclusively found no such evidence of shrinkage (Rushton 2012), meaning the black-white IQ gap has remained stable for more than two generations even though society has become much more supportive of blacks through desegregation and affirmative action.

When controlling for IQ, most content of most racial performance gaps vanishes (Herrnstein & Murray 1994, 320-340). Under IQ-controlled conditions, blacks are more likely than whites to graduate from college and to attain a high IQ occupation. Blacks, after controlling for IQ, make just as much money as whites and are only 5% more likely to be in poverty than whites. On the other hand, without controlling for IQ blacks are 20% more likely to be in poverty than whites. Low IQ is an immediate and primary reason for poor black performance. By syllogism, knowing the cause of the black-white IQ gap can reveal a significant portion of the causation of the racial performance gaps.

Heritability is an important concept when it comes to understanding the causation behind variance in a metric like IQ; it is defined as the proportion of variance in a population’s trait that is explained by variance in the genetics of that population. With older methods (known as quantitative genetics) heritability can only be directly measured within a population. Since molecular genetics have no yet progressed to the point that those techniques can be used to estimate the proportion of genetic causation behind the black-white IQ gap, the heritability of the between-groups gap has to be inferred indirectly. Luckily, decades of quantitative genetics research has provided more than enough data to allow for such an inference.

It fits to start with the within-group heritability of IQ. For first world whites this figure has potential to be one of the best replicated findings of psychology and behavioral genetics. Let it suffice to cite a study with a sample size of 11,000 twin pairs from white countries that found that the heritability of IQ is about 66% at age 17 (Haworth et al. 2010).

Before the age of 17, the heritability of IQ is generally measured to be lower (this is called the Wilson Effect). This doesn’t mean that IQ is more malleable when a subject is a child. For example, intervention programs and other tactics designed to raise the IQs of children tend to produce an effect in childhood that fades by adulthood (Jensen 1969)(Protzko 2015). This “Wilson Effect” could be a result of poorer g-loading and reliability of child IQ tests. Child IQ data is in fact less reliable (Jensen 1973, 82). Since variance due to measurement error is included in the “nonshared environmental variance” umbrella in the standard model, the fact of lower reliability of child IQ tests does explain at least some portion of the Wilson Effect. The possible reasons for poorer reliability are intuitive: the test may be inferior in its increased simplicity, developmental differences between children of the same age may cause noise, and so on. What cannot explain the Wilson effect is increased long-term malleability of child IQ, because variance due to permanent malleableization in childhood is variance still present in adults.

Other studies have found that what is not explained by genetics is mostly explained by unshared environment (McGue et al. 1993)(Bouchard & McGue 2003), which is defined as the environment that two children raised together don’t share. Unshared factors might vary from family to family but on the aggregate they include peer groups and stochastic effects and exclude family pressures and school quality. But in particular research has pointed to the idea that differential treatment by parents is a major unshared environmental factor (Plomin 2013) in general. But it’s easy to see how differences in parental behavior may not be as stochastic or environmental as they are a result of the child’s genetics. In typical methods that calculate “narrow” heritability, treating factors like this as wholly environmental can lead to underestimates of the true heritability of a trait. Methods involving fraternal twins or lower grades of relation also discount nonadditive genetic effects since siblings are only expected to share about a fourth of these effects, half siblings even less and so on, while identical twins have identical genomes. IQ heritability estimates using monozygotic twins reared apart tend to return values of 75% to 85%, for instance. All of this is to say that there are good reasons to think 66% might be an underestimate of the heritability of IQ. What might be the “best” method returns the result that the heritability of IQ within the WEIRD population is probably about 80%.

The point of discussing these details about the within-group heritability of IQ is not to attempt to claim that the heritability of the black-white IQ gap is the same. Rather, it’s to simply establish that genes have a large say in determining IQ for both individuals and populations. With that in mind, the next question is: how do the black and white IQ-relevant environments differ on the average? If the races have the same IQ-relevant environment on the aggregate, the heritability of the black-white IQ gap is 100% because the totality of the difference must be due to difference in the two gene pools. The lesser the heritability of the black-white IQ gap, the worse the black IQ-relevant environment must be when compared to the white IQ-relevant environment. The exact heritability of the gap can be written as 1 – (Between-groups variance in IQ-relevant environmental factors)/(Total between-groups variance).

It is highly unlikely that the heritability of the black-white IQ gap is insignificant given both the knowledge on within-groups heritability and the simple truth that race is real. Black people and white people look significantly different. Did evolution stop at the skin? No. For instance, blacks and whites have different norms for healthy kidney functioning. If evolution was significant enough that we can see it, and it didn’t stop at the skin, it’s doubtful it stopped at the neck, especially when the obviously divergent aggregate behavioral patterns of the races are noticed. But suffice it to say that computers can classify ancestry groups by DNA and match self-reported race with 99% accuracy (Bamshad et al. 2003).

Another study (see image) graphed racial clusters as identified by a computer, found they matched self reported race, and that the clusters group semi-discreetly (Guo et al. 2014). Africans were the most divergent from other racial groups, including whites. Both the data and evolutionary theory predict that there is some significant degree of heritability in regards to the black-white IQ gap. It would be improbable for the IQ-relevant gene pools to be the same or to produce the same IQ-phenotypes in the same environment. Is there evidence, then, showing that US blacks have extremely deprived IQ-relevant environments such that this data-trend can be overcome?

There are roughly three types of environmental factors that environmentalists like to point to: test bias, physical factors, and psychological factors. The first one involves the claim that IQ tests are not measuring a real difference in intelligence between blacks and whites because IQ tests are alleged to be culturally be biased. Black people are thought to do only superficially poorly. This hypothesis has been debunked since the late 20th century – in 1987, it was found that the difference between black and white scores on questions independently deemed to be culturally biased as actually less than the differences on less “biased” test questions (Jensen & McGurk 1987), leading the authors to conclude that racial testing differences reflect a real difference in intelligence. Jensen also wrote a whole book on this in 1980 titled Bias in Mental Testing which came to the same conclusion as the 1987 study.

Next, environmentalists, having accepted the reality of the intelligence difference, like to claim that blacks have a disadvantaged IQ-relevant physical environment on the aggregate (with the implication that if true, this is white people’s fault). Specifically, it’s claimed that blacks live in poverty due to white people, which impacts black nutrition and school quality, which depresses their IQ scores. The simplest way to show that this doesn’t matter is that when these factors are controlled for, the black-white IQ gap doesn’t dissipate.

 The Minnesota Transracial Adoption Study (MTRAS) tracked blacks, 50/50 black-white mixes, and whites who were adopted out to white middle class parents at very young ages. At the least, variance in age at adoption in the study didn’t matter – the correlation between IQ at 17 and age at adoption was only 0.13. Unless there exists racism so pervasive as to make adopters of black orphans give those orphans a significantly impoverished environment due to their skin color, this study effectively controls for socio-economic status, nutrition, school quality, and a whole host of other factors that could be hypothesized to be depressing black IQ. To the affirmation of the theory of evolution, the MTRAS found a black-white IQ gap of one standard deviation between black and white adoptees in young adulthood, the same gap that’s found in the normal population (Weinberg et al. 1992). It also found that the gap between the mixed race adoptees and the other groups was about half a standard deviation, which is good evidence against the racism narrative (Drake and Obama are “black”) and excellent evidence for the genetic hypothesis.

More convincing is that what the MTRAS data is corrected for the Flynn effect, the racial gaps slightly widen (Sternberg 2000, 185). This is probably because the Flynn effect doesn’t act on general intelligence (g), as the correlation between Flynn increases and g loadings of different tests ranges from -0.40 to -1.00 (te Nijenhuis, & Van Der Flier 2013). As an aside, some people who know little about IQ like to claim the Flynn effect means IQ can’t be “genetic” or that it’s an invalid concept. This isn’t the case because the effect does not cause a larger increase in black IQ and renorming the test works.

The MTRAS is a sublime piece of evidence, but it is corroborated by others. For example, the black-white IQ gap remains roughly the same at all income levels (Herrnstein & Murray 1994, 288), meaning lower group income couldn’t be the cause. It’s also likely that income-matched families are also nutrition-matched, school-matched, and so on. If this is the case, then neither poor schools nor poor nutrition nor whatever else are causes of the black-white IQ gap. In regards to schools in particular, voucher studies reveal that school quality does not cause or correlate with changes in student performance (Chrisman et al. 2012) (Wolf et al. 2010), something which is much less heritable than IQ yet highly correlated with the latter.

After accepting that depressed black IQ is almost certainly not caused by “poor schools” or bad nutrition or low income, the environmentalist typical moves on to claims about the specter of slavery and segregation. But black IQs around the globe tend to be depressed relative to white IQs. Unlike the Irish, whose IQs were depressed but rose when they urbanized, urbanized blacks still feature depressed IQs no matter the country (although US blacks tend to have higher IQs than African nationals). The same IQ gaps exist in the US, the UK, and South Africa, (Lynn 2015, 63) (Rushton & Skuy 2000) (Lynn & Meisenberg 2010) despite their divergent histories on the treatment of black people.

Next they might speak of stereotype threat or black culture. Not only is stereotype threat a priming effect, there is no evidence that it explains racial gaps on actual IQ tests (Stricker & Ward 2004) (Wei 2008).As for black culture, not only is culture a function of genetics, but motivation measuring items are the only test items blacks tend to do better on than whites (Jensen 1973, 111-113). This means that blacks aren’t scoring poorly because their culture just makes them think “dis test shit wack yo.” And finally let it be remembered that the mixed race adoptee data from the MTRAS essentially destroys the systemic racism narrative.

That narrative is further discredited by the previously shown data that reveals that blacks and whites with equal wealth experience the same IQ gap. And, furthermore, when IQ is controlled for, most racial performance gaps disappear. What systemic racism can there be, then, that would cause a one standard deviation IQ gap? Blacks would have to be treated so poorly by whites after controlling for income, nutrition, and school quality that it causes there to be essentially no black geniuses. And in every country on Earth! If systemic racism really were this severe in a country where being black is worth 230 free SAT points (Epenshade et al. 2004), then it would be a waste of time combatting it at all. The races would be better off going their separate ways.

Environmentalism has been falsified. But there is still more positive evidence for a genetic difference being responsible for at least most of the black-white IQ gap. Sibling regression toward the mean is one line: a black sibling of someone with an IQ of 120 will have an IQ of about 100, while a white sibling would have an IQ of about 110 (Jensen 1973, 118). This implies differential nonadditive genetic effects, especially when combined with the dearth of evidence for an IQ-relevant environment gap. Another line of evidence is that the blacks have a smaller standard deviation for IQ than whites (Jensen 1973, 212). This means that if blacks and whites have a similar IQ-relevant gene pool, black heritability will be higher than white heritability because there must be less environmental variance overall for black people. But this is not the case, suggesting that the black IQ gene pool is dissimilar to the white IQ gene pool (if the variance in environment is similar).

All the evidence seems to corroborate the idea that the black-white IQ gap is mostly explained by differences in black and white IQ-relevant gene pools and not differences in IQ-relevant environments. It is virtually certain that the heritability of the gap is at least 50%, but the current state of the evidence warrants giving an estimation of 100%. At this point, the only evidence that can cast doubt on this position is that which demonstrates that an environmental reduces the black-white IQ to an extent that is statistically significant. But since environmentalists are heavily funded and have had decades to find their factors, it’s probably safe to say they’re wrong.

REFERENCES         

Bamshad, M. J., Wooding, S., Watkins, W. S., Ostler, C. T., Batzer, M. A., & Jorde, L. B. (2003). Human population genetic structure and inference of group membership. The American Journal of Human Genetics, 72(3), 578-589. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1180234/

Bouchard Jr, T. J., & McGue, M. (2003). Genetic and environmental influences on human psychological differences. Journal of neurobiology, 54(1), 4-45. https://web.archive.org/web/20161114181852/http:/isites.harvard.edu/fs/docs/icb.topic1169974.files/Bouchard%20McGue%202003.pdf

Chrisman, J., Swenson, D., la Tarte, A., & Natzke, N. (2012) Test Score Data for Pupils in the Milwaukee Parental Choice Program (Report 5 of 5). http://legis.wisconsin.gov/lab/reports/12-14full.pdf

Espenshade, T. J., Chung, C. Y., & Walling, J. L. (2004). Admission preferences for minority students, athletes, and legacies at elite universities. Social Science Quarterly, 85(5), 1422-1446. https://web.archive.org/web/20180907052552/https://www.princeton.edu/~tje/files/files/webAdmission%20Preferences%20Espenshade%20Chung%20Walling%20Dec%202004.pdf

Guo, G., Fu, Y., Lee, H., Cai, T., Harris, K. M., & Li, Y. (2014). Genetic bio-ancestry and social construction of racial classification in social surveys in the contemporary United States. Demography, 51(1), 141-172. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3951706/

Haworth, C. M., Wright, M. J., Luciano, M., Martin, N. G., de Geus, E. J., van Beijsterveldt, C. E., … & Kovas, Y. (2010). The heritability of general cognitive ability increases linearly from childhood to young adulthood. Molecular psychiatry, 15(11), 1112-1120. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2889158/

Herrnstein, R. & Murray, C. (1994). The Bell Curve.

Jensen, A. (1969) How Much Can We Boost IQ and Scholastic Achievement. Harvard Educational Review: April 1969, Vol. 39, No. 1, pp. 1-123. https://hepgjournals.org/doi/10.17763/haer.39.1.l3u15956627424k7

Jensen, A. (1973) Educability and Group Differences. Harper & Row. https://libgen.is/book/index.php?md5=501F355B6E474BDC0B7F3130DC3BF9C0

Jensen, A. R., & McGurk, F. C. (1987). Black-white bias in ‘cultural’and ‘noncultural’test items. Personality and individual differences, 8(3), 295-301. 11.       http://arthurjensen.net/wp-content/uploads/2014/06/Black-White-Bias-in-Cultural-and-Noncultural-Test-Items-1987-by-Arthur-Robert-Jensen-Frank-C.-J.-McGurk.pdf

Lynn, R., & Meisenberg, G. (2010). The average IQ of sub-Saharan Africans: Comments on Wicherts, Dolan, and van der Maas. Intelligence, 38(1), 21-29. https://www.researchgate.net/publication/222062631_The_Average_IQ_of_Sub-Saharan_Africans_Comments_on_Wicherts_Dolan_and_Van_der_Maas

Lynn, R. (2015) Race Differences in Intelligence: An Evolutionary Analysis, Second Revised Edition. https://www.intelligence-humaine.com/wp-content/uploads/2019/03/Race-Differences-in-Intelligence-second-edition-2015-1.pdf

McGue, M., Bouchard Jr, T. J., Iacono, W. G., & Lykken, D. T. (1993). Behavioral genetics of cognitive ability: A life-span perspective. https://psycnet.apa.org/record/1993-98367-003\

te Nijenhuis, J., & Van Der Flier, H. (2013). Is the Flynn effect on g?: A meta-analysis. Intelligence, 41(6), 802-807. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.694.8353&rep=rep1&type=pdf

Plomin, R. et al. (2013) Behavioral Genetics, 6th Edition. https://libgen.is/book/index.php?md5=9303B823B96B155DDC249D9267EAD8DC

Protzko, J. (2015). The environment in raising early intelligence: A meta-analysis of the fadeout effect. Intelligence, 53, 202-210. https://www.sciencedirect.com/science/article/abs/pii/S016028961500135X

Rushton, J. P., & Skuy, M. (2000). Performance on Raven’s Matrices by African and White university students in South Africa. Intelligence, 28(4), 251-265. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.503.9118&rep=rep1&type=pdf

Rushton, J. P. (2012). No narrowing in mean Black–White IQ differences—Predicted by heritable g. American Psychologist, 67(6), 500–501. https://doi.org/10.1037/a0029614

Sternberg, R. (2000) Handbook of Intelligence. https://libgen.is/book/index.php?md5=7173C0103012053229366179E9546CD3

Strenze, T. (2007). Intelligence and socioeconomic success: A meta-analytic review of longitudinal research. Intelligence, 35(5), 401-426. 1.            https://emilkirkegaard.dk/en/wp-content/uploads/Intelligence-and-socioeconomic-success-A-meta-analytic-review-of-longitudinal-research.pdf

Stricker, L. J., & Ward, W. C. (2004). Stereotype Threat, Inquiring About Test Takers’ Ethnicity and Gender, and Standardized Test Performance 1. Journal of Applied Social Psychology, 34(4), 665-693. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1559-1816.2004.tb02564.x

Wei, T. E. (2008). Stereotype threat, gender, and math performance: Evidence from the national assessment of educational progress. Working Paper, Harvard University. https://web.archive.org/web/20161114181944/http:/isites.harvard.edu/fs/docs/icb.topic630262.files/sthreat_naep.pdf

Weinberg, R. A., Scarr, S., & Waldman, I. D. (1992). The Minnesota Transracial Adoption Study: A follow-up of IQ test performance at adolescence. Intelligence, 16(1), 117-135. 10. http://faktasiden.no/dokumenter/minnesota-transracial-adoption-study.pdf

Wolf, P., Gutmann, B., Puma, M., Kisida, B., Rizzo, L, Eissa, N., Carr, M., & Silverberg, M. (2010) Evaluation of the DC Opportunity Scholarship Program Final Report. https://ies.ed.gov/ncee/pubs/20104018/pdf/20104018.pdf

The social liberal’s fallacy

“{1} Children [age group defined by the speaker] shouldn’t be allowed to do [thing the speaker thinks adults should be allowed to do]” without “{2} because it’s net harmful as evidenced by [firm evidence]” is the Social Liberal’s Fallacy. “Harmful” is open to interpretation because this is not about ethics and as long as the speaker believes there to be harmfulness based on good evidence, there ceases to be the Social Liberal’s Fallacy among that speaker’s thought. In the wild, things are usually very implicit and the fallacy is expressed simply as “Children shouldn’t be allowed to do [thing]” by a speaker who thinks adults should be allowed to do [thing]. But {2} may only be absent for brevity’s sake. Therefore let us call the Hard version any time when {2} does not accompany {1} and the Soft version any time where {2} does not exist for the speaker, i.e., the speaker is unaware of any {2} for their {1} and cannot give it on demand.

This tool allows us to detect inconsistencies in normative beliefs when they are applied to different groups. It does not demand, “You must let children do that now.” It can alternatively imply that “adults shouldn’t be allowed to do that thing.” I firmly believe all types of fornication are wrong, so I do not commit the Social Liberal’s Fallacy in my opposition to pedophilia. Of course, I could be wrong about fornication, but elsewhere I show what I believe to be firm evidence that it is not the case that I am wrong. And in addition for any {1} there can be a {2}, and while I think sodomy is net harmful in general, child-rape is certainly worse.

Contemporary parenting norms and most age restrictions are prime examples of this fallacy in action. (This does not mean those actions-as-conclusions are wrong, that would be the fallacy fallacy.) Ask the average person, a social liberal, why children shouldn’t be allowed to curse or why 17 year olds shouldn’t be allowed to drink, and they’ll either short-circuit or repeat mindless platitudes that don’t qualify as {2}.

People do this when their beliefs unreasonable, i.e. fallacious. The conclusion is for some reason so important to them that they don’t care that they literally believe mutually exclusive things or believe complicated empirical things for no reason. With the average person, I find in life that their beliefs are just post-hoc rationalizations of their actions to give when asked, and their actions in domains like this one are mostly a function of obeying power.

Why, then, does power “promote” the Social Liberal’s Fallacy? Possibly because Social Liberalism is untenable in the long run and things would have already collapsed if we didn’t have a weird dualism where children live under extreme Social Conservatism while adults totally abandon all morals. This is implied if most views nonjustified by the Social Liberal’s Fallacy have the correct solution that [thing] universally should NOT be allowed, because if this is the case Social Liberalism is generally harmful.

I do believe that social liberalism is generally flawed and used to destroy our civilization. Tradition is a high-powered longitudinal experiment as the sample size is in the billions or more and the time range is literal millennia. And up until about 200 or 100 years ago surviving was much harder due to vastly inferior energy harvesting methods. People who believe in “progress” are therefore naïve; technological progress implies forgiveness towards negative mutation away from a delicate equilibrium because technology makes it appear like we can afford all the harms that come with, say, the destruction of traditional gender roles. It’s obvious and confirmed in many of my essays that social liberalism is a waste of energy and therefore is morally wrong.

For example, women objectively make a worse choice then men on the aggregate when it comes time for election, as white women vote way more for the party that openly aims for white genocide as opposed to the other party that at least pretends to be against it. Women being in the work force en masse is also wasteful because they’re worse workers on the aggregate (people are reading Sexy and Smart by Roderick Kaine), most jobs are BS and there’s enough men to do the real ones (read: Bullshit Jobs), and smart/white women aren’t having enough kids, causing a dysgenic trend/white endangerment.

So why have feminism? There’s no truly ethical reason. Sadly, however, most people today are hedonists and the answer is because “feminism makes women feel good.” Silly social liberal! Childism would also make children feel good. Why then do children still have bedtimes? Let them eat cookies for dinner if they want. YOU DO YOO! The Social Liberal short circuits and shrinks back; his control scheme has been found out. Childism would not help the Social Liberal puppet masters who own the Means of Propaganda control the world. For this same reason, I can’t just do whatever I want. How can there be laws when they make me feel bad? Oh? It’s about collective pleasure? Why are we all equal then? I suppose farming has to stop. It’s absurd to say all humans are morally equal but not all mammals or animals or life. Wait no, maybe it’s intelligent egoism. Laws are ultimately better for you because YOU don’t want to be victimized. This is the closest to the truth we’ve gotten. Why then should there be feminism if you are a man? He shrinks back to imaginary veils in the sky whereupon preexistent souls vote on what is ethical (unironically the absolute state of modern ethics).

I lied when I said this wouldn’t be an ethics post. Are you mad? The answer is: your feelings don’t matter. What does matter is the existence of our consciousness. Good is an idea that exists within the observer, and so what is good is the existence of that quality observation. Anything but the goal of maximizing intelligent life’s continuance is a distraction and a road to death. Those who preach those things in a way do not even exist, for the end of their philosophy is destruction.

Which is better for continuance? Hyper-trad racist sexist misogynist national IQism or the garbage that is wokeism? Under the former Star Wars and anime would be real already, while under the latter white people are transitioning to Endangered Species status in our own countries, we’re distracting ourselves to death with fake divide and conquer astroturfed “issues,” and the average IQ is going back down to gorilla level at a rate of about 4 points per century.

Against epistemic aestheticism and towards an empirical and statistical understanding of society

There is a spectre over the dissident right: the spectre of epistemic aestheticism. This spectre threatens to lead us down a false path backward into the ages of subhuman squalor. It is the true culmination of reaction for reaction’s sake: a complete return to prehistorical intellectual subhumanity via the rejection of truth and the restoration of will on the throne of epistemology.

While its adherents swear this idea is adverse to postmodernity, they make a crucial error: while this project is reactionary indeed, the postmodern project discriminates not between “reaction” and “progression” in themselves, despite the transformation of the meaning of “progress” into the totality of postmodern regress. In actuality, the clear trend since Nietzsche, the father of epistemic postmodernism, has been towards epistemic aestheticism, the rejection of reason in thinking and its replacement with pure, imperfect human will. How else does detritus such as Critical Theory justify itself? How do the race deniers think if not by the aesthetic of equality? The right has long been special in its masculine orientation towards the truth and its opposition to “human will as reality.” But the epistemic aestheticists, the “Postmodern” or “Art” Right threaten to subvert humanity’s last hope of setting upon the path to perfection. If they are successful, the raw, based, and unshakeable reality of the world will remain forever unknown to us, as we exterminate ourselves slowly through the worship of an imperfect flesh’s will.

The main disseminators of epistemic aestheticism are Bronze Age Pervert, Zero HP Lovecraft, and Mencius Moldbug. Each of these figures is associated with the NRx strand of the dissident right, and for each of them there are signs of astroturfing. An article in the Atlantic attempts to expose some of these signs. One: Peter Thiel is NRx or adjacent and at the least funds Moldbug’s business ventures. And Two: Moldbug “joked” that BAP is very close to Steve Bannon. At any rate, his opinionated book received a suspicious level of attention. As for Zero HP Lovecraft, his twitter is suspiciously well followed and he also oddly is allowed to post on the Claremont Institute’s, a “conservative think-tank,” magazine called The American Mind where he writes replies to Moldbug (who replies to BAP; all three are on this site!) on occasion. He’s also active in the BAP twittersphere. Furthermore, despite the relative popularity of these figures, none of them have proven themselves capable of producing work that transcends epistemic mediocrity. While at times interesting, all of their work indeed boils down to mere aesthetic, always lacking epistemic rigor. Proselytizing explicitly for epistemic aestheticism is only fitting for these three. Last but not least, all of them fail a fundamental thinker-quality litmus test: the acceptance of the idea of massive Jewish influence on society and the idea that said influence is integral to the ideological status quo (this is not to say it’s the only factor). Moldbug is confirmed to be Jewish and fails to rise to the level of men like Ron Unz, admitting he’s never read MacDonald but lazily rejecting the two ideas nonetheless. Zero HP Lovecraft lacks a fundamental understanding of the nature of zeitgeists and spreads Dawkin’s liberal meme theory, regarding “the Jews” as just a boogeyman and not an essential piece of the puzzle of postmodernity. And finally, BAP is BAP.

Knowing enough to be weary of intentions and the rigor of ideas, let us examine the doctrine of epistemic aestheticism (my naming) in their own words. I am not well read enough in BAP (nor will I ever be, unless perhaps I am sent to prison for hate speech, and I am allowed to read, and BAP is all they have. Not even then honestly) to know if, when, or where he might have exhorted epistemic aestheticism, but if I were it would be irrelevant anyway since BAP does not respect truth in any of his utterings, meaning his literal words cannot be reasoned about. Instead, BAP’s whole, singular book encapsulates epistemic aestheticism. BAP tells us on the first page that “this is not a book of philosophy. It is exhortation.” A book of science or truth is never thought of. The doctrines of wokeism and neoliberalism are furthermore said to be “filthy” and “a great ugliness” but not untrue. In these lines are the entirety of epistemic aestheticism: the truth doesn’t matter. In its place, beauty, which is ultimately just an emotional reaction of an imperfect creation, at least in the way it is meant. For I believe there is true beauty in the world, and it is tied inextricably with the true objective good, but that man, being flawed, is not entitled to feel it beyond that which reaches his hindbrain through a glass darkly. What is good and beautiful is furthermore, also true. BAP does not understand this and nor do other epistemic aestheticists, because if they did they would not have strayed by exalting their unconscious sin-drive over their God-given reason. Only dedication to truth is dedication to beauty; though BAP may not be capable of feeling it, being imperfect (nor am I necessarily any more capable of feeling it, to be clear), the one true beautiful aesthetic is the one arrived at when man and all his dirty, lying institutions is purified by truth. In essence epistemic aestheticists truly understand neither epistemology nor aesthetics nor anything at all but the ability for the low to violate the aristocratic principle and bring down the high, via the gross manipulation of lowly emotions.

Moldbug basically states this explicitly in his essay “The Deep State vs. The Deep Right:” “Art [is] a weapon. … Man invented art for one reason: to mog.” And Moldbug is happy to use this weapon wantonly. Truly, in light of the low level of epistemic rigor exhibited throughout the rest of his work, Moldbug has often be excused by followers as being “an aesthetic.” But if art is a weapon, a dangerous one at that, one that Moldbug spuriously (per usual) claims “All revolutions begin as a fundamentally aesthetic break,” who should wield such a weapon? Should it be, as Moldbug writes about Marx, Jews “whose ideas are drivel and whose writing is divine?” Obviously not. If indeed art is a great weapon, it should only be wielded by those who are firmly based. That is, those who are based in truth and reality. Not, dare I say, those who still believe in liberal theories of the zeitgeist like Dawkin’s memetics. Not those unwilling to acknowledge the influence of the Jewish tribe upon postmodernity even in pseudonymous writing. And certainly not those who openly eschew any reading at all (especially if it’s scientific) like BAP. Moldbug even “agrees” with me: “The easiest path to aesthetic dominance is mere truth. Above all, one feature makes any story ugly: lies.” But upon a deeper inspection it is revealed in his second Clear Pill essay that truth for Moldbug is just a thing, and “other qualities might also be beautiful. If they could overpower [truth], the result would be a beautiful lie.” This sentiment goes back far for Moldbug, as in an early 2007 writing titled “A Formalist Manifesto,” he plainly states “I am not a big fan of ‘empirical evidence.’” Thus it is revealed why such an intelligent man can seem so careless with his ideas: truth matters only insofar as imperfect people think it’s beautiful, so Moldbug optimizes his writings for persuasiveness or faux-beauty and not for real-beauty or truth. This is dangerous and actually ugly. Insofar as Moldbug serves as a gatekeeper of the truth and not an imperfect gateway to it, he and his ilk might be a net-negative subversionary force among the Truth-Right and the world at large.

Among the three, Zero HP Lovecraft has the best chance of being a net-positive, despite esoteric-fedposting. Sadly his head seems all twisted up in speculation (“philosophy”) and perhaps in justifying this he sings the praises of epistemic aestheticism. Responding to Moldbug’s essay on the same topic in the American Mind, Zero HP Lovecraft references a Borges story called “Tlön, Uqbar, Orbis Tertius.” The story has themes of epistemic relativism and subjectivism, which ZHL explains: “It begins with a small brotherhood who conspire to fabricate a history for a non-existent place …

In time, scholars and laymen alike become fascinated with the imaginary country of Uqbar in the region of Tlön, and they participate in the creation of its lore, its histories, its economies, and its fashions. The encyclopedia of Tlön is so detailed, the narrator surmises it to be the work of “astronomers, biologists, engineers, metaphysicians, poets, chemists, algebraists, moralists, painters, geometers…directed by an obscure man of genius.” As its knowledge spreads, Tlön becomes more real than what is real, and the culture of earth is supplanted by this new simulacrum, as people increasingly adopt the customs and languages of the world they have invented.

Note that this is a perfect example of what a meme theorist might consider to be a viral meme that is also a lie that is more “beautiful” than a truth. He continues expressing his goal to create such a lie and have it rule the world: “Your schools will be invaded by the language of Tlön; the teaching of its harmonious history (filled with moving episodes) will erase the one which governed in your childhood.” This is all no better than progressivism and its “beautiful” lie of equality. In fact, this is all quite weak. Those who must cope live by lies but those with strength can bask in the full penetrating glory of the truth. The only reason to create and “reside in” Tlön is precisely that the truth is not tolerable for the imperfect subject in question and the current pretty lies are not enough of a cope, for whatever reason.

Nevertheless the theme of epistemic nihilism or relativism should be addressed. We know the data but not the words published in science are safe from zeitgeist-lies because (a) scientific inventions work and (b) social science data independently converges on taboo truths like that of Jewish influence, genetic inequality, and female-femininity. History is another question. It indeed is in a much sorrier state, but there is a good reason why it’s incorrect to use encyclopedias as an historical source. In the case of Tlön being inserted into encyclopedias, it would be trivial to show that no primary sources exist that can verify the supposed history of the fake city. And I see no evidence, even anecdotal, that humans are in such a sorry state that it would be possible to systematically fabricate never-before-seen primary sources in such a way that they become universally accepted and thought to be real. Although I suppose that as sovereignty becomes more efficient a la 1984 this could change. For now though, my own research into history seems to indicate that the evidence converges onto taboo truths in many ways. All of this is to say that epistemic nihilism or relativism is just a cope.

What is to be thought of epistemic aestheticism, then? Moldbug in particular executes a delightful motte & bailey, noting (spuriously of course) that art is a weapon. I’m inclined to agree with this, where art is propaganda. The great mass lacks the capacity to fully appreciate the truth and should be guided by those who can until the time when that great mass can be improved. This is done primarily by propaganda and force. It is therefore not wrong in itself to craft an aesthetic, but it is wrong if such an aesthetic is used to turn people away from the truth. Ideally, the one who crafts the mere aesthetic for the masses is inferior to the one who discovers the truth the aesthetic uses for its scaffolding. The aestheticist must be licensed by good understanding of what is real or else he should take orders from someone with such an understanding. But what we have with the lying artist is a rebellion against natural hierarchy, something many instantly identify to be a core trait of “leftism.” It is true that since the truth is still largely unknown, the artist must lie some of the time. But a lie can point towards the truth more so than another lie. The question is: where does NRx lie along this spectrum?

Attacking Moldbug: A brief critique of his style and epistemic methods

After reading an analysis of Moldbug by “The Worthy House,” a Catholic blogger, and after skimming some of his early writings myself, I’ve come to believe that Moldbug’s writings have serious flaws, despite their popularity among the dissident right. His epistemological method and style both have deep many deep errors, and accordingly the actual substance of his popular ideas are at best epistemologically lucky and at worst nonsensical.

Moldbug’s writing can be summed up as often being invalid epistemologically and dishonest rhetorically. In terms of the former, he almost always uncritically asserts certain assumptions and never seriously justifies them. Instead, he commits what I call the “therefore fallacy”: he spends almost all of his time discussing the implications of the high level claims that he presumes to be true. In terms of the former, Moldbug is often vague, sometimes only implying his assumption, and often veers off into smalltalk as opposed to analysis. Often this smalltalk is in the form of imagery, painting the world as according to Moldbug’s assumptions in the reader’s mind. This is a particularly vapid form of wasting his time discussing the implications of his assumptions as opposed to actually empirically analyzing their merits in the first place. Such vagueness, imagery, and smalltalk serve to make the reader drop his guard. When it is considered that much of his writing is composed of such chaff, it is implied that Moldbug writes not to tell or discover the truth, but to “persuade,” that is, to propagandize and trick certain unaware readers into uncritically accepting Moldbug’s dictations. Another rhetorical trick that Moldbug frequently uses is the non-central fallacy. This essentially involves the redefinition of a connotatively loaded word, like Orwellian, church, or leftist, into another definition that allows something else to be labeled with the word. This therefore transfers the previous connotation to the newly labeled thing undeservingly. One example highlights all of these issues. In the beginning of his “Gentle Introduction,” Moldbug asserts that liberal societies are “Orwellian Mind Control States.” This is totally unjustified, practically unverifiable, and asserted uncritically. Instead of carefully defining the term and testing whether or not certain societies fit the bill, Moldbug displays a low-info 5th grade understanding of history by asserting that “Nazi Germany and the Soviet Union” are examples of Orwellian Mind Control States. In the midst he babbles about brain eating worms and speaks directly to the reader in an obnoxious tone instead of just getting on with the “argument”. The necessarily intelligent man seriously portrays himself as incapable of critical thinking and rigor. Only AFTER saying that every society involved in WWII is an “Orwellian Mind Control State” (despite vaguely asserting that National Socialism is “categorically different” from liberalism) does Moldbug vaguely define his term as “a society that maintains their legitimacy by shaping public opinion [in a way which is inaccurate].” So what he has done here is he coined a vague term, asserted that the US, the USSR, and Nazi Germany all fit the term for no particular reason in between inane smalltalk and imagery babble, and then he defines the term, retroactively and uncritically designating all three societies as “Orwellian.” This is not only an example of an epistemologically invalid thesis interspersed with obnoxious smalltalk, it’s also an non-central fallacy tactic designed to cheaply and uncritically throw the connotation of “Orwellian” onto real societies. Afterwards it’s more smalltalk about Jedi and Sith and then an equally as stupid and sloppy assertion that Harvard is really a church, another example of anticoncept. The epistemologically proper way to argue this would be ask the question as to whether the US and other societies rely on shaping public opinion to believe lies in order to function. Then it should have been empirically examined as to whether or not they do this, and if so, what instrumental purpose it has. Depending on the results, which are to be based on detailed historical fact and perhaps some contemporary data collection, the question can be answered in the affirmative or in the negative. In the pursuit of truth, the rhetorical tactics should also be cut out: no more interspersed small talk about the wacky implications of the unproven assertion, imagery having to do with brain worms, or anticoncepts. Indeed, when it is imagined that those things are cut out of Moldbug’s writing and only the substance remains, it becomes clear that Moldbug didn’t test his assumption at all. As opposed to asking a question, researching it, and concluding, Moldbug literally asserted that the answer to the question is yes and moved on to the next topic. The rhetoric serves in part to obscure this empirical deficiency.

This one example is characteristic of all of Moldbug’s writing. Moldbug himself has admitted this briefly in a variety of places. In an early 2007 writing titled “A Formalist Manifesto,” he plainly states “I am not a big fan of ‘empirical evidence.’” Much later his thoughts seem the same. In “The Clear Pill Part I,” published in 2019, Moldbug wrote “Stage magic works by presenting true facts in a pattern that suggests a false story, and obscures a true story … Political stage magic is the psychological engineering of the population. … Maybe I’m the magician myself! Be careful.” Finally, in a draft of the first chapter of a book posted in June 2020 titled “#1: a general theory of collaboration,” Moldbug demonstrates a wanton disregard of the value of the truth, saying “My answers are long, hard, and very likely wrong. Maybe someone else can do better with the same questions.” His methodology is exactly the same in the 2020 writing as it was in 2007. Absolutely no care has gone into finding the truth; instead, Moldbug seems to be satisfied with thinking of himself as a most likely wrong, unempirical “stage magician,” This is what an analysis of his writings indicate that he is, and these flippant comments are evidence that he knows it.

A final point about the man: He portrays himself as if he writes much more than he reads. His writings are too vague. For instance, his treatment of history is almost always generalized, consisting of large, vague brushstrokes like the understanding of a child. Yet with this basis in facts he theorizes about power and memetics regardless. This demonstrates a reckless intellectual negligence characteristic of somebody with little understanding and who probably skims Wikipedia more than he actually reads the books he claims to read. His only form of citation, which is pasting the Amazon link to a whole book after claiming something that he perceives to be vaguely related, also seems like the sign of a man who reads more book summaries than he does whole books. The Worthy House concurs and finds additional evidence of this hypothesis: “for example, he repeatedly ascribes to Machiavelli the phrase ‘if you strike at a king, you must kill him,’ though it really comes from Emerson. And it was not Edmund Burke, but Adam Smith, who said ‘there is a lot of ruin in a nation.’ Such errors, rarely fatal but always irritating and undermining Yarvin’s claim to have a macroscopic view, crop up with metronomic regularity.” This is important because Moldbug’s style is highly demeaning and pedagogical; he dictates what is supposedly the truth to the reader, with little to no outside justification. It’s as if Moldbug just knows his assumptions are true from a lifetime of scholarly research. Except his knowledge level comes off as embarrassingly undetailed and underdeveloped, enhancing the issues with the way he writes.

Below I have attached the dissection of “The Clear Pill Part II” I performed. Though he toned down his use of the non-central fallacy, his method was largely the same. The annotation revealed that extremely little justification for his views is to be found in the first large chunk of the essay, while therefore fallacy passages, unsubstantiated assertions, and small talk permeate the piece. After briefly justifying some trivial points about the role of propaganda in Nazi Germany and the USSR, Moldbug asserts his belief in a decentralized theory of social control involving ideas as real objects and justifies absolutely none of it while not sparing the reader at all from the therefore fallacy, filling line after line with the minute implications of his faith, bordering on asserting new unjustified assumptions much of the time. In contrast, another source was dissected as that was on a similar topic, revealing that the therefore fallacy and the smalltalk was completely absent and that the source was absolutely loaded with justification for every line. That this source is considerably less heard of, alongside its particular disagreements with Moldbug, is telling.

Moldbug’s computer project Urbit was funded by Peter Thiel, who also funded Hoan Ton That, a transsexual Vietnamese developer who created Clearview AI, a dystopian nightmare software that can match any security camera clip of a person to identifying government photos. Hoan Ton That has gone on record to say that they are a neoreactionary and a fan of Moldbug. Peter Thiel is similarly amiable. A true understanding of money and memetics might lead to the hypothesis that part of Moldbug’s comparative popularity on the right is due to him being shilled by moneyed interests. He doesn’t even have to be aware. Quite literally, all they have to do is pay to get his Google ranking bumped or even pay shills to spread his ideas in seemingly organic ways. That Moldbug advocates for “disengagement,” (i.e., doing nothing) as the solution to our problems alongside his decentralized power theory and denial of the JQ only adds to any suspicion that he might invoke. But to true Moldbuggians, who have been convinced, based on what seems to be almost nothing of substance, that ideas and power are decentralized, this theory could never be entertained. What might be deeply flawed by many measures is really perfect. The same might be said for contemporary pop music and its many fans.

https://imgur.com/HnMdYlb