Race and Technology

HumAnIze · May 28, 2025, 11:12pm

Carleas:

HumAnIze:

[Emil Kirkegaard’s blog/Mankind Quarterly]

I won’t just dismiss this outright, but I think it’s important to note that it shouldn’t be given the same weight as a peer-reviewed journal article. Mankind Quarterly was literally founded by a Nazi who worked with Mengele to experiment on prisoners in concentration camps. It’s currently published by the Human Diversity Foundation, which was founded by … Emil Kirkegaard, author of this article. The Human Diversity Foundation is a continuation of the Pioneer Fund, an explicitly segregationist and eugenicist organization, which distributed Nazi propaganda in pre-WWII US (HDF was created to transition the Pioneer Fund into an LLC so that it had to disclose less information about its activities).

So we have an organization created to advance a particular view, publishing an article advancing that view, written by the founder of that organization. This should significantly reduce our credence in its claims to the extent they rely on a trustworthy source. Do you agree that throughout history there have been people who have hated other races for unscientific reasons? That people have organized to label others as ‘inferior’ (including Jews and Asians, who the modern incarnation of race science says are actually genetically special)? That given that it happened in the past, it is possible that papers put out by this organization are not objective science?

That doesn’t necessarily mean dishonesty (though it wouldn’t be the first time lies were used to spread hate), but we have scientific bureaucracy in part because motivated science is often poorly done. And this is nothing if not motivated science, and it has not passed through the normal filters used to catch bad science. Those filters aren’t perfect, and have their own biases, but if nothing else, circumventing them to avoid their biases also deprives a paper of that indicia of reliability.

That would be a problem even if we could actually see the article, but it’s behind a paywall (and not in sci-hub). So we’re using an author’s press release about their questionably-reliable article. We also have good reason to distrust this source and diminished ability to evaluate the claims. That’s not great.

But, correct me if I’m wrong, but it looks like the finding here is that a machine learning model trained on 50,000 variables computed from MRI scans of kids brains and their IQ was able to predict IQ from brain scans with a correlation coefficient of .51, is that right? In other words, a fairly detailed map of someone’s brain is associated with intelligence? And the author acknowledges the dataset oversamples poor black kids? And the framing is that this “confirms race differences in brain size and functioning”?

I dunno, I’m not a stats guy, but seems like not what we’re looking for.

HumAnIze:

[Philippe Rushton]

Another leading light of scientific racism, who led the Pioneer Fund until his death. Perhaps to Kirkegaard’s credit, he begins his blog post by noting that, “Rushton’s own claims about his life revealed some inconsistencies between historical records and his claims, which some might call lies.”

On the other hand, Rushton was an actual academic, and this paper was published in an actual journal. He’s certainly got his academic detractors, but it’s hard to tell how motivated they are in their conclusions (although one critic pointed out that he once cited to Penthouse magazine, which is egregiously unscientific but also based).

Kirkegaard finds his methods “crude” and his results “overly strong”, and his paper was intended to fix those problems.

But again, I don’t think either of these gets at our disagreement. We know lots of brain features are shaped by environment, e.g. wealth, parental investment, nutrition, and stress, all of which differ significantly between races. So at most we’ve got a somewhat predictive collection of physical characteristics of the brain that weakly correlate with both intelligence and social race.

HumAnIze:

Russell T. Warne

This is easily the best of the three (here’s a PDF of the full paper): not written by a race scientist or published in a journal founded by a literal Nazi, and actually makes a claim that gets at our disagreement. And the claim is several lines of evidence, when taken together, show that average differences in intelligence between groups are more than zero percent genetic (among English speakers in the United States).

So first note that this is a modest claim, and intentionally so – part of what make the article credible is that it approaches a complex subject with due care. It’s nothing like the claims being made in this thread.

And assuming the argument is perfectly sound, it does not support claims amount evolutionary divergence between human populations or neanderthal admixture. Groups in America can’t be assumed to be a representative sample of the groups from which they evolved, because they pass through the filter of ending up in the US.

Also note that genetic causation is complex. Genetic analyses have a hard time separating 1) genes that cause low IQ because of how they encode brain structures, from 2) genes that cause low IQ because of how they encode superficial features that change how the individual is treated, from 3) genes that cause low IQ because they make people worse parents and kids get their genes from their parents.

I’m willing to get more into this paper if this is actually a basis for your belief. Is it? For now, I’ll leave it at that, and also point out that a number of papers citing this paper challenge some of its claims, e.g. this one challenging whether within-group heritability is generalizable to differences between groups. Citations are not uniformly negative, but if you’re just showing me one article that supports your beliefs, consider that there are others that contradict it.

HumAnIze:

Certain allele groups closely associate with higher intelligence.

They are not, they are weakly correlated with small effect sizes. At most, they are about as predictive of intelligence as parents’ socieconomic status, and they only work on people with European ancestry (i.e. they’re associated with within-group variation in intelligence).

HumAnIze:

Natural selection produced different subgroups of the human species, which subgroups we call races. Each race then produced its own culture, society language tendencies values etc.

Kallikantzaros:

No…that’s an ethnicity.

Race+culture = ethnicity
A single race can have multiple ethnicities.

Genes/Memes

Races are whatever the speaker needs them to be.

Seems like we should be more specific and use different words to refer to the various concepts we might be referring to with the word ‘race’.

Wow, that’s a lot of straw man, character assassination, and distraction.

“That person was bad”
“Race scientist”
“Nazi”
“Not a real journal”
“Papers saying the opposite also exist”

Sounds like excuses to me. Why can’t you address the core ideas here? Someone being a nazi has nothing to do with whether or not their ideas are wrong. You are merely attempting to character assassinate rather than address the ideas.

Good point about differentiating how genes can affect intelligence, either directly or through the indirect routes you mention; and guess what, all of that matters. If your genes code you to be a shitty parent, and being a shitty parent correlates with lower intelligence, then your genes are still contributing to the causality of why you are lower intelligence. Because in all likelihood your own parents were shitty to you, as you said it’s in the genes bro, and genes affect environment. Which was my point from the beginning: even (some of) environmental influence can be returned back to genes.

Genes causing both direct and indirect effects on intelligence, it all matters. I realize the liberal view is grandly utopian, everyone gets a perfect life with perfect parenting perfect education perfect nutrition perfect experiences perfect life never any bad experience no negativity no stress no worries no challenges no negative emotions no harms. I get that. In such a world, most of those indirect genetic effects would be partly or entirely negated.

Unfortunately, we live in the actual world, not a libtard utopia. Also unfortunately, for your position I suppose, that even if we lived in such a fantasy the direct genetic effects would still be there, and would still matter… that is, until the scientists of Libtardia begin using genetic engineering to alter babies in the womb or pre-conception so as to counteract these bad genes so everyone can be peak intelligent, tacitly admitting that intelligence DOES have a meaningful genetic component but I am sure that will go above the heads of the wonderfully progressive citizenry of this hypothetical future utopia.

Yes, because if race cannot be perfectly defined with absolute clarity and zero ambiguity, or if race has imperfect boundaries, then this must mean race doesn’t exist. Of course. Why didn’t I think of that before.

A person’s genes can be used to determine their race, just as they can be used to determine their intelligence. How do you think 23andme works to tell you your racial heritage? How do you think they do twin studies to confirm the heritability coefficient of general intelligence? The entire human genome has been mapped. Even if we don’t know what every gene or combination of genes does it is quite easy to take DNA samples from across different groups and then analyze them compared to other variables, like intelligence. Which has already been done. Gene groups that are identified with high intelligence are more prevalent in white European populations than in African populations (not sure about Asians, would need to go back and check that). Just like how brain volume differences between races are present right away in the womb.

You would jump through any and every hoop possible just to avoid admitting there are meaningful genetic differences between races, wouldn’t you? Oh yes, of course you would. That is what the religion of modern neoliberal progressivism requires.

Um, no. That would be you. I am the only one among the two of us who cares about the truth. You are motivated by your ideology. And other than calling people “libtards”, what offensive things have I said? Oh right, you think that stating FACTUAL CLAIMS is “offensive”. Like if I say white Europeans have superior genes than black Africans, when it comes to the genetics of intelligence, you think that is “offensive”, ah yes. And you say that and also somehow still claim to care about the truth. Yikes. “Unable to defend my beliefs”, classic. I’ve made dozens of arguments already, given you rational explanations, evidence and other sources, and clear logical arguments supporting my position. You have responded with… “meh, those dudes were bad, also some other papers that don’t agree, and look at this thing over here”.

Honestly, the most simplified form of my argument could be stated like this: How in the hell could you honestly believe that genetics have nothing to do with measurable between-race differences? I thought you are a scientific, atheist-materialist sort of person. I am sure you believe in Evolution and genetics. I am sure you understand how natural selection works to cause differences between groups based on changing their genes. So… how in god’s name can you say “naw bru, none of that matters AT ALL EVEN ONE LITTLE BIT when it comes to differences between HUMAN POPULATIONS. Because humans are UNIQUELY SPECIAL IN THE ANIMAL KINGDOM and have ZERO GENETIC CAUSES for between-population differences.”

I mean… just.

Sorry to say, but the only dishonest person in this discussion is you. And yes that is a factual statement, and not meant to be “offensive”. But I suppose you will take it that way anyway.