Steve Sailer's Greatest Hits

His collected essays

The book is called Noticing. It collects some of Sailer’s essays over several decades.

Sailer is known for being a “race realist.” That means that he notices differences in average IQ in statistics collected by race. In “It’s all relative,” published in 2002, he defines race this way:

A racial group is an extended family that is inbred to some degree.

All right. That is the definition of something, and you could probably make it scientific.

My problem is that I do not know how to connect the dots between that definition and the statistics about race in America. Those are based on the government’s definition, which is pretty much “We have these crude classifications that have no scientific basis. But using your folk understanding of race, pick one of these classifications to describe yourself.”

To me, if you’re going to be a “race realist,” you probably have to go with what the government says, even though as Sailer puts it (and I agree)

we Americans should be wary of using the vast power of government to exacerbate the natural divisiveness of race by officially classifying people by race.

Glenn Loury has objected to “race realism.”

Can we sensibly aspire to a more complete social integration than has yet been achieved of those who now languish at the bottom of American society? A political movement that answers no to this question must fail, and richly deserves to.

In fact, I would be willing to argue that Sailer and other race realists have better ideas than progressives for making our social arrangements less unfavorable to people with low IQ. And Loury would be the first to agree that the progressives’ go-to remedy of affirmative action is counterproductive. But in the mainstream media, the terms of the debate have been set by the self-described anti-racists, and there is no way for anyone with a different perspective to participate.

One of my favorite essays in Noticing is “Cousin Marriage Conundrum.” It appeared in January of 2003, as America was getting ready to attempt nation-building in Iraq.

In Iraq, as in much of the region, nearly half of all married couples are first or second cousins. A 1986 study of 4.500 married hospital patients and staff in Baghdad found that 46% were wed to a first or second cousin, while a smaller 1989 survey found 53% were “cosanguineously” married…

By fostering intense family loyalties and strong nepotistic urges, inbreeding makes the development of civil society more difficult. …

…In the U.S., where individualism is so strong, many assume that “family values” and civic virtues such as sacrificing for the good of society always go together. But, in Islamic countries, family loyalty is often at war with national loyalty. Civic virtues, military effectiveness, and economic performance all suffer.

…Unlike the Middle East…Europeans became increasingly sympathetic toward the right of a young woman to marry the man she loves. Setting the stage for this was the Catholic Church’s long war against cousin marriage, even out to fourth cousins or higher.

Seventeen years after this essay, Joseph Henrich published The WEIRDest People in the World, a book that argued that the distinctive culture of the West can be traced to the marriage patterns dictated by the Church, including monogamy and the ban on cousin marriage. As I wrote in my review.

The opposite of WEIRD PC is kinship or clan PC. In kinship PC, what knits a society together are familial interconnections, including many marriages among close relatives.

I find it useful to think of society as consisting of games played at three levels. There is the individual, the small group, and the large society. What you do as an individual may or may not be in the interest of the small group, so the small group has to use norms and gossip to police your behavior. What a small group does may or may not be in the interest of the larger society, so the society needs formal rules and impersonal enforcement to put boundaries around nepotism and clan favoritism.

Within this model, an inbred family is a small group that is too cohesive for a larger society to be able to control. That is why cousin marriage is incompatible with Westernization.

That is an example of Sailer “noticing.” For some reason, Henrich notwithstanding, American intellectuals have been unable to absorb this insight. Darn if we still aren’t trying nation-building in the Middle East (“Let’s put a reformed Palestinian authority in charge of Gaza.”)

Another provocative chapter is “America’s Black Male Problem.” It first appeared early in 2023. It discusses the work of Raj Chetty, an economist who in 2013 earned the Clark Medal, which is often a precursor to, and arguably even more prestigious than, a Nobel Prize.

Chetty has strenuously positioned himself as an anti-racist good guy,

And in fact a theme of what Bryan Caplan would call social desirability bias runs through a lot of Chetty’s research. That is, Chetty’s papers show that policy makers can “fix” under-achievement and poverty, whether by using high-quality kindergarten teachers or getting poor people into better neighborhoods.

But when Chetty published his first map of high and low social mobility in The New York Times in 2013, I immediately pointed out that his findings of regions with strong upward mobility were simply white places, like the upper Great Plains and Utah.

Sailer writes,

Chetty’s methodological brainstorm was to forge relationships with federal agencies like the Internal Revenue Service and the Census Bureau so that they would provide him with individual data, such as your tax returns, but in “anonymized” form.

…he knows what your income is from your 1040 and what your race is from your Census form. He knows who your kids are…what college they went to…

There is indeed a gold mine of “administrative data” available to researchers. Records of recipients of federal aid, such as food stamps, can be had in a format that keeps the individuals anonymous, but allows linkage to other government data about them.

Sailer’s essay focuses on a particular analysis that Chetty produced, comparing incarceration rates of black and white males, after controlling for the income of the household when they were adolescents (parental income is very predictive for both black and white males). Sailer writes,

All else being equal in terms of household income during adolescence, black men are four times as likely to find themselves behind bars as white men

…Is the cause racism?

Well, if it is, racism doesn’t much hinder black women. they appear to be incarcerated only about 30% more often than white women raised with the same family income, not 300 % more often as with black men.

…I’ve been arguing since the 1990s that blacks average more masculinity than other races.

Well. As far as I know, there is no “masculinity” scale on which psychologists measure men. You would have to define masculinity and then develop an index of observable characteristics.

Imagine a psychometrician coming up with a questionnaire. Would you be more likely to stop and stare at earth-moving equipment or kittens? Would you rather go to a wrestling match or the ballet? It seems like an interesting project, although good luck getting funding for it.

Instead, I would prefer to speculate on data that we could already obtain, at least in principle. My intuition is that there is a “marshmallow test factor” (MT) that combines IQ with conscientiousness (a trait measured by psychometricians). My speculation is that very low MT in males is predictive of incarceration, and that blacks are somewhat more likely to have very low MT.

(As with most characteristics, much of the variation is within racial groups. There will be many blacks with higher MT than some whites.

It may be that, on average black men can sprint faster than white men. Blacks certainly dominate at the highest levels of competition. But that doesn’t mean that there aren’t lots of white guys who can out-sprint many black guys.)

Regarding MT, the IQ portion is mostly genetic. The conscientiousness portion is partly genetic and partly environmental. The environment could include parenting style, culture (perhaps growing up in an honor culture leads to less impulse control) and experiences (perhaps being born into poverty and/or mistreated on account of race reinforces low impulse control).

Higher MT leads to better outcomes. So as you move up the income scale, you find parents of both blacks and whites who pass along higher MT.

Low parental income is an indicator of low MT. For males, really low MT is predictive of incarceration. So that is the story for how low parental income predicts high incarceration rates in both black and white males.

What about females? Either having really low MT is not as likely to lead a female to behave in a way that results in incarceration, or females are less likely to have a really low MT.

To explain the black/white ratio in racial incarceration rates using MT, one would have to show that, controlling for parental income, blacks have lower MT. You would need a statistical story for that.¹

A challenge with any attempt to use statistics to discuss group differences is that most people do not understand statistics. As Sailer writes in a chapter “Why Lesbians Aren’t Gay” (from 1994),

Many journalists today write as if they are unable to distinguish between perceptive observations about the average traits of a group and blanket assertions about each and every group member. . .

Conspicuously missing from current debates is that most useful of all conceptual tools for thinking about both the similarity and the diversity of human beings: the probability distribution (more roughly known as the bell-shaped curve).

I said before that with most characteristics, much of the variation is within racial groups. Sailer would say the same thing, but it means nothing to people who cannot process statistical concepts.

For that and other reasons, I do not see “race realism” as the way to go. I am more in the Coleman Hughes camp (read that essay’s conclusions to see my views on the futility of race debates). Treat people as individuals, and stop keeping score by race.

substacks referenced above:

@

The Spiritual Blind Spot of the "The Bell Curve"

1. If the MT distribution of W is to the right of the MT distribution of B, then high values of MT in B parents are likely to be outliers, so that regression to the mean would lead to lower MT in the children of high-income B’s than of high-income W’s.