Richard Hanania is Posting Bad Coal

Hanania has been posting a lot of coal lately, and I suppose I’ll be the first to formally call him out on it.

His coal has some clear, discernible features that sets it apart from the other coal on twitter. First of all, these posts are intended to infuriate his right wing audience, appear superficially non-partisan, and, most importantly, the coal he posts isn’t even accurate. This is similar to what Richard Spencer does (edit: he’s gotten better as of 11/2023), with the difference being that Spencer is less intelligent and more repetitive than Hanania, making him an even bigger foot hemorrhoid.

I would probably turn the other cheek and post a criticism on twitter if he had only done this once or twice, but Hanania has already posted 6 unreasonably bad takes:

Exhibit 1: Conservatives are stupid.

Anybody who is familiar with this man knows how much he likes to emphasize that conservatives are less intelligent than liberals.

If you tunnel on the difference in intelligence between the left and the right, the difference is not large, and the magnitude and direction of the effect depend on how the question is asked and on the specific time period (Kirkegaard, 2023).

Across all time periods, the difference between liberals and conservatives is about 4 IQ points, while the difference between Republicans and Democrats is not even statistically significant. The difference in intelligence between the left and right seems to have accelerated recently, the cause of which I suspect is Donald Trump being the central figure of American conservatism. However, the differences are generally small, especially in the years before 2015.

In general, asking whether individuals are conservative or liberal generates larger differences than asking for political affiliation. If you use a better intelligence test than the wordsum, you see similar results with a slightly larger effect size (Kanazawa, 2010). If you look at specific political views, you find that libertarian-like political views are the ones most associated with intelligence (Carl, 2014). This finding has been replicated several times, but with varying effect sizes (Jedinger, 2022; Kirkegaard, 2016; Kemmelmeier, 2008). Naturally, this argument rests on the assumption that I think libertarians a right wing, which they clearly are. First, Republicans are two times more likely to identify as libertarian (PR, 2014). Second, progressives tend to view libertarians with disdain, while most Conservatives support them.

The perception that true beliefs will always correlate with intelligence should be treated with skepticism, as individuals who believe that ‘Blacks are less economically successful due to lower levels of inherent learning ability’ are less intelligent than those who don’t (d = .83) (Jensen, 2022), despite the fact that the belief is mathematically correct (Jensen & Nyborg, 2001; (Hu, 2022).

A commenter noted that Hanania is critical of the intelligence of the online, populist right, and not necessarily of the right as a whole. If so, then I guess I can’t blame him since a lot of these types of people seem rather unintelligent. Regardless, if he is talking about a specific type of right winger, then he is not doing a good job of communicating it. In the original twitter quote he was complaining about the ‘online right’, which could reasonably apply to any group, from populists to dissident right twitter users. In this tweet he also applies his criticizes Republicans as a whole for being unintelligent, as he does in the article. I do agree that the right gravitate towards spoken media moreso than the left, though I don’t think this is indicative of lower intelligence, as Hanania himself admits:

Finally, “watching TV and reading” are not simply euphemisms for “low IQ and high IQ,” although there is certainly a relationship to intelligence. To have a consistent ideology requires one to be smarter, but reading about politics is more high effort than watching TV or listening to the radio, and many intelligent people are too lazy or distracted by other things to get very into the world of ideas.

Exhibit 2: Conservatives win all of the time.

“Cthulhu may swim slowly. But he only swims left. Isn't that interesting?”

-Curtis Yarvin

One of the most characteristic frames of the new right is that liberal “democracies” always tend towards left wing policies in the long run. Hanania contests this idea in his article Conservatives Win All the Time, arguing that the right wing has achieved many victories in recent years in the domains of gun control, abortion, homeschooling, and taxes. For instance, in 1988 only one state had unrestricted right to carry, compared to the 25 states that allow it in the current year.

Contrary to him, Sean Last notes that a lot of the victories the conservatives achieved are merely regaining lost ground. The overturning of Roe v Wade is self-evidently retreading old ground, as the ruling had only been established in the 70s. He also comments that Hanania commits the marginal tax rate fallacy, as many times individuals did not pay the high taxes that were enforced in the 50s to 70s due to loopholes. In fact, it is hard to dispute that taxes are an issue that the right has lost on when you consider the historical trend.

Homeschooling is indisputably a right wing victory - Sean believes this to be a reaction to the left wing subversion and capture of public education systems. I don’t think this is the whole story - as Bryan Caplan and Joseph Bronski note, the public education system serves primarily as a credentialing service which ceases to teach useful skills to most children past the 8th grade. I am sure that a lot of the motivation behind homeschooling is religious, but I doubt that the efficiency of the public education system is helping either.

Even considering that he may be right about guns and homeschooling, we can’t forget the left’s overwhelming victories on:

• Homosexuality

• Gender roles

• Race

• Regulation

• Taxes

• Civil rights

• Religion

Exhibit 3: Women are disgusted by pro-life men, and these men are low status and arrogant

In a recent article, Hanania argues that women are uniquely disgusted by men who know nothing of the struggles of pregnancy and only think about regulating women’s behaviour. Like Hanania, I am also a fan of ad-Hominem arguments because I think that personality is a factor that causally influences the beliefs of individual. However, I would probably try to find more evidence for my claim than one anecdote of a feminist writer finding pro-life men disgusting. That would be like concluding that women don’t care about height because my 5’1’’ sisters say they don’t.

Instead of resorting to anecdotal data, I would rather examine what the empirical evidence suggests. Using the Add health dataset, I found that, within men, thinking abortion is morally acceptable doesn’t significantly correlate with any big 5 personality trait (wave 4), relative weight (aka BMI) (wave 4), height (wave 4), attractiveness (wave 2+4), drug use (wave 3), or self-reported criminal behaviour (wave 3). In fact, it is rather remarkable how unpredictable men’s thoughts on abortion are. Within women, thinking abortion is morally acceptable correlated with openness (r = .072, p < .01), drug use (r = .098, p < .001), and self-reported criminal behaviour (r = 0.1, p < .001). Apparently, other researchers find that supporting abortion correlates with opennness (Osborne, 2022). If you were convinced that pro-life men are particularly unempathetic or arrogant, you would expect them to be more disagreeable on average, but they aren’t.

Out of curiosity, I ran the same analysis only within non-religious people (n=494), and found supporting abortion correlated with openness (r = .19, p < .001), drug use (r = .16, p < .001), self-reported criminality (r = .14, p < .01), and attractiveness (r = .096, p = .038). I must say though, p = .038 is nothingburger territory and I would be very unsurprised to see it not replicate.

In addition, women were more likely to think abortion was morally unacceptable across the board (d = -0.09, p < .01). From other national samples, it appears that there minor differences in how men and women think about abortion. For supporting overturning Roe v Wade, the difference in support is small (30% in men, 26% in women), when asked whether they support a 6 week abortion ban, the difference is moderate (41% in men, 32% in women), and the difference in wanting most abortions to be legal is also small (58% in men, 63% in women).

There does seem to be an increasing trend in women identifying as pro-choice, however. I honestly have no idea why he didn’t use something like this as evidence instead of the wordwall of an article he posted.

Given regression to the mean is a thing, I expect stabilization at 50s vs 40s in favor of pro life.

What Hanania misses in his article is that it seems that pro-life positions are simply not that popular, in men or in women. Interestingly, this group of “low status” men and “overly empathetic” women seem to have successfully gotten their way in recent years. Indeed, liberal “democracy” sure has worked out well for us! What a fantastic system.

Exhibit 4: The media is honest and good

Last but far from least, this man has even argued that the media is honest and good. This article may have been prompted by Scott Alexander’s controversial piece arguing that the media very rarely lies, and merely communicates a poorly reasoned or deceptive interpretation of information. Both of them are correct in claiming that the media is not lying - what Hanania misses is that large centralized media organizations are not useful. Like him, I don’t blame the press for falling for Iraq WMDs and Russiagate, since (if I am not mistaken) these lies were promoted by officials in the American government. What I do blame them for, is poorly covering politically related topics and pushing the country to the left.

The poor coverage of politically sensitive topics is something others have touched on - they fell for a study that claims that perceptions of brilliance of fields negatively affect the amount of women who enter the fields, neglecting that the average GRE math score by field was an even more robust predictor of female representation (r = -0.82) than percieved brilliance (r = -.6) (Alexander, 2015). Or when the media will touch on racial inequalities in police shootings, without mentioning that the use of force by police towards minorities is usually in line with how likely the police are to encounter them:

chart from Cremieux

Let us not forget that the media reported that only 0.01% of welfare recipients said they were on drugs, when the test involved the applicants self-reporting their drug use (Alexander, 2022).

I think that this dead horse has been beaten enough - the other issue with centralized media is that they are, overall, left leaning. This is evidenced by the fact journalists tend to support left wing parties (Kirkegaard et al., 2021), and that about half of journalists are left wing in the UK, while the other half are either moderates or conservatives (Reuters, 2016).

political parties grouped by support among journalists vs people and by ideology (Kirkegaard et al., 2021)

In addition - there is a very clear difference in how presidents Obama and Trump were received:

from Pew Research

Hanania is very aware of this:

But I don’t have high standards for humanity. “Be intelligent, don’t explicitly lie to me, don’t see yourself as on a team trying to ‘own’ the other side, and have some kind of professional standards where you at least care a little bit about truth” is about the best that I think we have the right to expect. And institutions like the NYT, the Washington Post, and the Atlantic generally meet that standard, at least to a much greater extent than most of their critics. I would argue that much of academia is broken in the way that a lot of media critics think the press is.]

I reject this framing of the debate as originating in standards of conduct. What I am interested in is the utilitarian approach - would burning these institutions to the ground do humanity a service or a disservice? I find it hard to believe it isn’t for the best. It would break up the left’s patronage and propaganda networks, and the competent journalists in those institutions would find ways to make it on their own.

I assume Hanania’s response would be that the MSM promotes objective information independent of their interpretation of it. What he doesn’t consider is that few people reading this work have the right combination of skills to evaluate this information objectively if it collides with the sacred. For starters, if they aren’t intelligent enough, they just wont understand the information well enough. They also can’t be highly biased, or else they will interpret information in a way that always concords with their ideology. Put it this way: if many journalists routinely communicate poorly reasoned interpretations of evidence, how much better should be expect the American public to be?

Exhibit 5: countersignaling anonymity

I’ll refrain from making a list of suggestions here, but one thing I hope that Substack will consider is creating the option of not having to interact with anonymous accounts through a real verification system. A lot of right-wingers associate anonymity with freedom and the ability to avoid being punished by censorious employers. But much more often, it’s used to hide behavior that people would be cancelled or ostracized for in real life for good reason.

Not even 3 days after starting this draft, it seems that Hanania has posted another piece of coal, which is suggesting an option to blanket block anonymous accounts.

Not all of us have the privilege of having no heterodox opinions or outside funding to support us, so if we want to say the emperor has no clothes, we must do so behind an anime avatar or some other mask. Sure, anonymity does allow people to engage in otherwise objectionable behaviour, but guess what? I think it’s great. Whether it is edgy biden deepfakes, petty insults, horny posting, race science, or schizoposting, anonymity has made twitter a better place.

If blanket bans were on option, that would incentivize people to become facelords. While there are some pretty based ones, most facelords are boring and feel the need to comply with sterile norms to keep their jobs. In addition, those based ones who are not as sterile typically have lots of cushy funding that keeps them afloat in the midst of all of the zombies and status regulators. That isn’t to say that I don’t appreciate what these based facelords do, they give some legitimacy and consistency to the chaotic mess that is the internet.

Exhibit 6: supporting immigration

Richard Hanania supports immigration. My objection to immigration is simple:

socioeconomic development and national IQ from Kirkegaard

To be clear, National IQ causes socioeconomic development (Francis and Kirkegaard, 2022), and is not a mere correlate of it. I also assume that socioeconomic development causes national IQ to an extent, as IQ scores in Africa (65-70) are lower than what you observe in American Blacks (85), even when you take into account that American Blacks are about 20% European.

It is technically possible to create a political system where the productive and noble members of the society do not have to deal with the issues of the not so noble members of it. The problem is that all political systems and societies inevitably decay, making this point moot.

Conclusion

I think Hanania has been trying too hard to “challenge his audience”, or as us younger folk put it, post coal. It is not clear why this is the case, whether he is subconsciously optimizing for controversial beliefs, deliberately posting inflamatory content for self-interest, and/or just happened to come to these conclusions naturally.

Normally, I am a pretty agreeable guy. I could understand believing in a few crazy things, Hanania has gone off in the deep end to a point where I feel compelled to call him out.

edit - explained more clearly that I think that left wingers are more intelligent

edit 2 - edited article after receiving dms

edit 3 - note that I no longer agree with the anonymity take, I think few people would choose to block anon accounts to the point to which it would change incentives meaningfully. Regardless, the ‘anonymous troll demons’ take is one of the worst things to come out of intellectual circles recently.

Edit - it has come to my attention that coal means the same as ‘cringe’ and it means what it means because coal is ‘Black’. I had assumed people didn’t like coal because it is depressing and is only made for fire, but I guess people think more simply than I give them credit for. A different message has told me that coal is ‘coal’ because it is the opposite of a ‘gem’. I guess people have different reasons for thinking coal sucks, and I suppose it isn’t important.

here is the code used for the abortion analysis. im too lazy to post the whole thing + its the code im using for a formal study which will come out in several months or so:

hanania <- merge(wave2quest, wave4quest, by="AID")
hanania <- hanania[!duplicated(hanania$AID), ]

cor.test(hanania$H2MO15, hanania$extr)
cor.test(hanania$H2MO15, hanania$open)
cor.test(hanania$H2MO15, hanania$agree)
cor.test(hanania$H2MO15, hanania$consc)
cor.test(hanania$H2MO15, hanania$neur)

men <- subset(hanania, hanania$Female==0)
women <- subset(hanania, hanania$Female==1)

men$attu = men$att2 + men$att4
women$attu = women$att2 + women$att4

cor.test(men$H2MO15, men$extr)
cor.test(men$H2MO15, men$open)
cor.test(men$H2MO15, men$agree)
cor.test(men$H2MO15, men$consc)
cor.test(men$H2MO15, men$neur)
cor.test(men$H2MO15, men$attu)
cor.test(men$H2MO15, men$height4)

cor.test(women$H2MO15, women$extr)
cor.test(women$H2MO15, women$open)
cor.test(women$H2MO15, women$agree)
cor.test(women$H2MO15, women$consc)
cor.test(women$H2MO15, women$neur)
cor.test(women$H2MO15, women$attu)
cor.test(women$H2MO15, women$height4)

hanania2 <- merge(wave2quest, wave3quest, by="AID")
hanania2 <- hanania2[!duplicated(hanania2$AID), ]

men2 <- subset(hanania2, hanania2$Female==0)
women2 <- subset(hanania2, hanania2$Female==1)

mean(men2$H2MO15, na.rm=T)
mean(women2$H2MO15, na.rm=T)

cor.test(men2$H2MO15, men2$drug3)
cor.test(men2$H2MO15, men2$problem3)

cor.test(women2$H2MO15, women2$drug3)
cor.test(women2$H3TO111, women2$drug3)
cor.test(women2$H2MO15, women2$problem3)
cor.test(women2$H3TO49, women2$problem3)

d <- cohen.d(wave2quest$H2MO15, wave2quest$BIO_SEX2)
d$p
cor.test(wave2quest$H2MO15, wave2quest$BIO_SEX2)
wave2quest$BIO_SEX2


unrel <- subset(hanania, hanania$H2RE1==29)
unrel$attu = unrel$att2 + unrel$att4

cor.test(unrel$H2MO15, unrel$extr)
cor.test(unrel$H2MO15, unrel$open)
cor.test(unrel$H2MO15, unrel$agree)
cor.test(unrel$H2MO15, unrel$consc)
cor.test(unrel$H2MO15, unrel$neur)
cor.test(unrel$H2MO15, unrel$attu)

unrel2 <- subset(hanania2, hanania2$H2RE1==29)

cor.test(unrel2$H2MO15, unrel2$drug3)
cor.test(unrel2$H2MO15, unrel2$problem3)