Reply to "Genes, culture, and scientific racism"

Could an article be any more wrong?

In my Quantitative Sociobiology Manuscript, I wrote

Memetics and “cultural evolution” are essentially two names for the same concept: studying how the human informatic environment influences and causes variations in human behavior. This chapter argues that the development and focus on these discourses stem from political blank-slatism, with “memetics” being non-mathematical and unscientific, and “cultural evolution” relying on mathematical models that lack empirical grounding and involve flawed data interpretation to support dubious blank-slatist assumptions. These assumptions, broken into low genetic bias and time-consuming informatic spread, are challenged through a review of existing literature and newly collected data, leading to the conclusion that they have been falsified. The chapter then introduces two variance component models, one static and one dynamic, to analyze the interplay of memetics and genetics in behavior shifts, arguing that the significance of memetics is best understood through the model parameters m^2 and mt^2​, representing the proportion of behavioral variance explained by informatics at a given time and over time. Ultimately, it is demonstrated that these parameters are close to zero for modern political behavior, such as the divide between right-wing and left-wing ideologies.

I specifically comment on a book by Cavalli-Sforza & Feldman called Cultural Transmission and Evolution.

The book is essentially a big, overly sophisticated exercise in committing the sociologist’s fallacy, written by some usual suspects who make wild claims like “race doesn’t exist.” Now, the 82 year old co-author Feldman is in PNAS claiming race doesn’t exist, admitting that his book was written as a form of activism against “racism.”

Below we describe how part of the original motivation for this work was to challenge strong hereditarian positions and crude genetic determinism. This 50th-anniversary special edition is an appropriate setting to attempt an authoritative statement on genes, culture, and race. Below we describe the historical context in which CE/ GCC emerged, and how the field’s findings can illuminate race-related controversies and are incompatible with racism.

….

While Jensen’s thesis [about black IQ being genetic] appeared in a respected academic journal, the contemporaneous racist speeches by Stanford physicist William Shockley, who openly advocated eugenics, received much wider media attention. Both Jensen and Shockley were roundly criticized by most geneticists. The controversy was swiftly followed by the human sociobiology debate of the 1970s and 80s, and the publication in 1994 of Herrnstein and Murray’s The Bell Curve, which reignited the race and IQ debate.

…

This was the backdrop to the emergence of CE and GCC. Cavalli-Sforza and Feldman were disturbed by the publicity garnered by the writings and speeches of Shockley and Jensen…

Cavalli-Sforza and Feldman’s model essentially amounts to claiming heritability estimates are biased because of vertical cultural transmission, and a lot of mathematical fluff on top of that, which never leads to anything scientists can verify or estimate. For what it’s worth, vertical transmission is well-defined, being the causal effect of parental phenotype on child phenotype. The problem is that it turns out that vertical transmission effects are basically null, especially for IQ and political phenotype.

The rest of the PNAS article is one falsehood after another, all piled on top of each other to form a gigantic tower of crankery. I will review the highlights, but the reader’s homework is to open the article, read part of it on your own, and find a falsehood I didn’t cover — don’t worry, it won’t be difficult. You won’t have to look far.

The tone is set well in the beginning:

Scholars interested in race and ethnicity concur that human races are social constructs, and that there are no meaningful genetic differences between socially defined races (1– 5).

5 citations to something which is not true. How does this happen? If you look through the citations, you’ll find they’re each entirely verbal opinion pieces. Maybe we’ll find that this is a pattern.

Eugenics, either “positive” (i.e., increasing the fertility of people with “desirable” traits), or “negative” (reducing the fertility of those with “undesirable” traits) was accepted by mainstream science until the atrocities of Nazi Germany became widely known (10, 12)

Richard Lynn wrote on this, saying that Nazi Germany’s eugenics program was neither the first nor the most extensive. The Nazis consequently had nothing to do with the decline of eugenics. Rather, it was probably biological changes, including mutational load accumulation, which lead to increased leftist sentiments in the population, thereby leading to the decline in the popularity of eugenics.

Lumsden and Wilson’s book is not “fringe” — this is just a mindless smear tactic invoking a common fnord. Looking at their citations:

This is the only article — they also cite two of their own books, the original and the mass audiences version, and a couple other books for mass audiences. There is no “extensive data” for Lumsden and Wilson’s model to be inconsistent with. Again, the citations and mostly verbal and mostly opinion. The hard statistical analyses are absent.

They call The 10,000 Year Explosion “racist” and claim that Ashkenazi Jews have a 10 point advantage in IQ because of “culturally transmitted norms.” This is obviously untrue, given the heritability of IQ and its insensitivity to education. Further, it’s unclear how Jews could have systematically different information about education than gentiles in the 21st century. How can there be significant “cultural” (informational ) differences between races in, for instance, the US, in the 21st century, under the presence of the internet? Jews also have higher polygenic scores for IQ.

We’ll end off with this:

What are we doing here? I already said earlier in the article that vertical transmission didn’t pan out — what is Feldman citing here, to act as if vertical transmission is legitimate? All three articles are by him, and they are just math. Not words, but math without data, after the data already debunked his model in the 1980s (one of the articles is from the 1970s and the others are from 2013 and 2021 respectively).

When you’re this wrong about the shape of the Earth, they call you a Flat Earther and alienate you. When you’re this wrong about human sociobiology, you get a professorship at Stanford and publications in PNAS, for decades on end. Isn’t it foolish to take academic status seriously after something as bad as this? It’s academic Gell-Mann amnesia.