Rationalism's Lamprey

If you are familiar with Rationalism, you probably know the name Steve Sailer. For the uninitiated, Sailer was once a mainstream paleoconservative who fell out of favor with the conservative establishment^[0] and has since developed a monomaniacal obsession with race. He is a leading figure in the "race science" movement centered on the idea that African racial groups are genetically intellectually inferior to Europeans, and popularized the term "human biodiversity" (HBD) as a polite euphemism for it. Over the past decade, he has been an active and increasingly central presence in Rationalist and Rationalist-adjacent spaces. Long-time readers of Slate Star Codex, the most influential Rationalist-adjacent blog of the 2010s, will remember his name from almost every comment thread. He was also active on other major sites such as LessWrong.

Sailer's goal was to convince readers that:

1. There is an enormous, genetically-driven difference in average IQs between racial groups
2. Public policy should be designed around this

His modus operandi was simple, but effective:

1. Join communities that emphasize freedom of thought and allowing unpopular ideas
2. Never throw insults or use slurs
3. Find a racial angle to any story or comment thread
4. Cite obscure studies or statistics supporting your position
5. If a particular study or statistic is refuted, pivot to another one
6. Hope nobody is as well read in obscure race studies as he is

These communities continued to welcome him on the grounds of freedom of thought, and on the principle that more information is always better than less. And perhaps there could have been a world where he was countered by equally committed people who had read the same literature but with the opposite priors. But in this world he had no such opponents, and so any time a commenter pointed out the flaws in his arguments/studies he could always pivot to another one. Over time, he scored enough "wins" by exhaustion that regular readers started to take note.

Sailer was not entirely successful. Many commenters successfully pushed back on his first point, pointing out the methodological flaws in his studies. Perhaps the gap exists, they would grant, but it is not plausible that the average IQ in Ghana is 62^[1]. But in haggling over the details, they would implicitly grant his framing—that non-white immigration and intermarriage are making America primitive. If you grant this, does the degree matter so much? They simply must be banned.

Of course, this is nonsense. We do not need to make nearly any public policy decision on the basis of race. Immigrants to the US are very far from a random sample of a population, and there are many levers we could use if we want to make our system even more meritocratic. And individually we are under even less obligation to do so. You do not hire an employee or begin a relationship with someone based on—much less only on—their race.

The typical counter-argument is that a racial difference in IQ would show that anti-discrimination efforts like affirmative action are misguided. But giving a fixed amount of advantage to a group with a broad distribution of ability in an otherwise meritocratic selection system will lower the average ability of those selected regardless of whether that group is better or worse on average. Such programs can be justified if they correct anti-meritocratic biases or serve some separate social purpose, but the average ability level of the advantaged group does not enter the calculation. The only programs for which this would make a difference are those which adjust the advantage to target a percentage outcome. These are rare^[2] and obviously foolish. We do not need to go searching for a way to discredit the likes of Ibram X. Kendi; they tend to do this themselves.

Sailer's biggest coup was Scott Alexander, the author of Slate Star Codex—where much of this discussion took place. Alexander never publicly endorsed Sailer, but based on his hands-off moderation and reading between the lines of his writing^[3] it was clear that he was at least intrigued, and the tenor of the comment section changed accordingly. This was later confirmed by statements like "HBD is probably partially correct or at least very non-provably not-correct" in his leaked emails. Over the years, Slate Star Codex and its successor developed a feedback cycle, attracting and converting HBD believers while boiling off those most averse to it^[4]. By 2024, 31% of his readers reported a favorable view of HBD on his own surveys.

What is particularly ironic is that Scott was aware of this dynamic:

The moral of the story is: if you’re against witch-hunts, and you promise to found your own little utopian community where witch-hunts will never happen, your new society will end up consisting of approximately three principled civil libertarians and seven zillion witches. It will be a terrible place to live even if witch-hunts are genuinely wrong.

I do not know if this awareness came too late, or if their shared battles against wokeness endeared Sailer to him. But through Scott—who was and is an influential figure in the Silicon Valley literati—Sailer managed to smuggle his views into its halls of power.

0. ^

   The National Review stopped publishing him around 2002.

1. ^

   This is an actual number he's trotted out from the "View on IQ" database. It would imply that e.g. the two Ghanaian STEM PhD students I've known are overwhelmingly likely to be the smartest Ghanaians of their generation. What a coincidence!

2. ^

   The only example I can think of is the FAA hiring scandal, which was overturned by Congress in 2016 for very obvious, non-HBD reasons.

3. ^

   Many of you probably remember the NYT "hit piece" on Scott Alexander that unmasked his identity, and was widely criticized in Rationalist-adjacent spaces for imputing that Scott was sympathetic to Sailer et al. because of his tolerance for their comments and a few uncharitably read sentences. I criticized the journalist's practices at the time and am still uncomfortable with such inferences, but in this case he was right and I was wrong.