How significant is racial discrimination in Brazil?

Here is one recent paper by Leticia J. Marteleto and Molly Dondero:

Racial disparities in education in Brazil (and elsewhere) are well documented. Because this research typically examines educational variation between individuals in different families, however, it cannot disentangle whether racial differences in education are due to racial discrimination or to structural differences in unobserved neighborhood and family characteristics. To address this common data limitation, we use an innovative within-family twin approach that takes advantage of the large sample of Brazilian adolescent twins classified as different races in the 1982 and 1987–2009 Pesquisa Nacional por Amostra de Domicílios. We first examine the contexts within which adolescent twins in the same family are labeled as different races to determine the characteristics of families crossing racial boundaries. Then, as a way to hold constant shared unobserved and observed neighborhood and family characteristics, we use twins fixed-effects models to assess whether racial disparities in education exist between twins and whether such disparities vary by gender. We find that even under this stringent test of racial inequality, the nonwhite educational disadvantage persists and is especially pronounced for nonwhite adolescent boys.

The pointer is from the highly rated but still underrated Ben Southwood.


That's pretty much the explanation I came up with 16 years ago for why, 500 years after Cortez, white people still more or less run Latin America. I looked at Mexico rather than Brazil, but the process is similar:

In Mexico, white conquistadors interbred with Indian women to produce mestizos. Let's assume that in 1519 the Spaniards and the Mexican Indians were equal in IQ and other significantly heritable traits that aid economic success. I'll follow Jared Diamond (Guns, Germs & Steel: The Fates of Human Societies) in stipulating that the conquistadors won solely because by luck they had the guns, germs, and steel on their side, and that the only reason they had superior technology was because Spain was less isolated than Mexico.

Now, imagine a conquistador and his Indian woman have two sons in the 1520s. These two mestizo brothers grow up and go out in the world to seek their fortunes. One is smarter, and he strikes it rich. The other wasn't so lucky in the genetic lottery, and he becomes poor. The rich son has a wide variety of potential wives to choose from. Like most men, and like almost all Mexican men, he is more attracted to blonde women, and thus marries one. (If you aren't familiar with the depths of Mexico's blonde obsession, try watching Spanish-language TV shows. Almost all the women on Mexican TV look like Finns.) His impoverished brother, in contrast, cannot attract a blonde wife. So he marries an Indian girl.

Then, the brothers have children. On average, the smarter, richer brother's kids, who are 3/4 white, are smarter than their underprivileged 1/4 white cousins. They're smarter not because they are whiter, but because their father had more smartness genes than their uncle. This trend continues: in both families, the smartest, most energetic, and most ruthless sons marry the blondest wives, while the blondest daughters marry the husbands with the most Right Stuff. Repeat for another dozen and a half generations. By 2000, this pattern could lead to the most European-looking people being the most naturally formidable, even if they weren't when they arrived in 1519.

Now, in Mexico every century or so, there is a massive upheaval like the Revolution of the 1910s. The white monopoly is fractured. Up through the cracks come the most talented mestizos and Indians. They start dynasties that persist to this day … but their grandsons and great-grandsons are notably whiter than they were, since the men of the family have been exploiting their social ascendancy to marry white women.

Ok, but at what point does this change with society?

I don't think blonde women are particularly prized anymore. (Also speaking for myself here).

Statistically, blondes die younger than non-blondes, so that cuts against Sailer's "DNA fitness" thesis, though if the difference is small probably people would not notice; however, I think close to the equator being too white is a skin cancer health hazard. Also you notice the trend, as meets implies,, of "brown being sexy" thus even Miss Sweden, last I checked (or at least once) had a deep tan. But Sailer is right: blondes and white skin do seem to have a premium, even in brown countries like the Philippines (they change their skin color using chemicals to look a sickly white and also dye their hair peroxide blonde; I btw don't like that).

What colour hair does your half-your-age gf have?

Are you suggesting that IQ correlates with skin colour even within the same family? That seems implausible.

If the study is correct, then Mexican penchant for blonds makes sense. They want their kids to avoid racial discrimination!

@Nick or whoever: explain more simply this sentence: "We first examine the contexts within which adolescent twins in the same family are labeled as different races to determine the characteristics of families crossing racial boundaries". Is it saying that twins in the same family have different skin tone shades, so that one is classified as "black" and the other as "white"?

He didn't suggest that, but I am a little unclear on how his theory relates to the post

By the way, can't twins have different fathers? Wouldn't such twins be much more likely to be classified as different races?

Thus Jews who were almost all poor in Europe were genetically stupid so killing them to improve Germany's gene pool makes perfect sense?

Thus the Holocaust never happened, just eugenics rationally implemented on a large scale?

And because natives are obviously stupid, Machu Picchu was obviously built by aliens because white people todat can barely do the engineering to build similar structures that can withstand the elements for a century, much less five centuries.

Another interesting way that Latin American countries differ from North American societies is that due to the lack of a One Drop rule, people who look pretty white in Brazil often have more black ancestry than Americans would guess.

Consider two sisters in Brazil. There father is 100% European and their mother is 100% African, so the two sisters' genetic admixture is 50-50. But one sister looks like she's 75% white and the other sister looks like she's 75% African. The sister who looks whiter is more likely to land a husband who is higher up the social scale (i.e., whiter).

But the whiter looking sister is just as black as her blacker-looking sister in the rest of her genome on average.

Over the generations, this pattern continues so you can end up with very white looking people with considerable African ancestry -- they just lack the African genes for looks.

In contrast, in the U.S., we've learned from 21st Century genome testing that people who identify as white tend to be really, really white genetically. Among Americans participating in 23andMe's genetics test who identified as white, the percentage of black ancestry was under 1%. The average white American appears to be about 1 / 512th black.

That actually shouldn't have been surprising because that's the natural result of the One Drop Rule and the difficulties of Passing.

Nice talk about percentages but the knowledge of alleles is a bit different.

Trans-racialism is a big thing in Brazil.

For example, when the current top Brazilian soccer star, Neymar Jr., first became famous at age 16 around 2009, he was a black youth. But Neymar's official picture for the 2014 World Cup shows a guy who appears to be mostly white:

The one-drop rule in the U.S. was very good for black racial solidarity. In contrast, in Brazil, white people have always been able to practiced divide and conquer on the offspring of the lower orders, accepting the whiter-looking girls and rejecting the blacker-looking ones.

Black activism in Brazil is generations behind black activism in the U.S., which was almost always led by people who were significantly white (Frederick Douglass, Booker T. Washington, W.E.B. Dubois, etc.) but couldn't qualify as white due to the one-drop rule.

In Brazil, talented mixed race men are often co-opted into the ruling class and their grandchildren are whiter than they are.

The word "IQ" does not occur in the article. Is there anything left to discuss?

Yep, the whole thing was bollocks.

Why would it matter? I wouldn't expect IQ to correlate with outward racial characteristics within twin pairs.

What does "expect" have to do with it? Measure it! We used to call that science.

IQ differences are largely due to polygenic effects. Given Mendel's laws of segregation and independent assortment, the expectation is that IQ won't covary with skin color and other outward racial characteristics within twin pairs unless there's pleiotropy. Therefore, if co-twins are categorized into different races and race correlates with outcomes like IQ and educational achievement within such twin pairs, then there's good evidence that racial discrimination of some sort is present. Mendel's laws are what I'd consider science, but if you disagree, please give your argument.

My argument could scarcely be simpler. Stop guessing; measure it.

"if co-twins are categorized into different races "

Probably there is a reason why they are categorized into different races.

In the US, there appears to be no correlation between skin color and IQ or educational attainment in black siblings.

Luke, yes, the reason is that they look different.

I'm fascinated by IQ supremacists. Why do they want to live in a world where IQ is even more determinant? Why people is so sure to be in that group and not belonging to the 99/100? Why people on the 90-99 percentile IQ think they deserve more than other humans? The problem of playing the IQ game is that unless you're a rarity, you're always below thousands and thousands of people. I think people is overconfident because they just compare with family, friends or co-workers, but if they realize there's a lot of intelligent people around the planet.....they'd think twice before advocating IQ importance/relative status.

If you're commenting here, then your criteria for a livable society will require a mean IQ of around 100, unless you don't actually mind squalor and paying kidnapping insurance premiums.

IQ is not everything. China has an average above 100 and it's not a place I'd like to move to.

There is a tsunami building out there, and the tabula rasa theory of human development is about to go the way of phlogiston and the crystal spheres.

"the nonwhite educational disadvantage persists and is especially pronounced for nonwhite adolescent boys."

If you assume this proves discrimination then you will fail to fix the problem. If on the other hand you recognize that the "nonwhite adolescent boys" are their own problem then maybe you can fix it or at least fix some of it. But that is a very unattractive narrative for the left, i.e. that poor people and people of color actually cause their own problems and it is part of their culture. Part of their culture to bully smart children of color as acting "to white" if they do well in school. To much emphasis in their culture on drugs, alcohol, violence and casual sex. Can it be fixed?? Sadly I don't think it can, at least not completely. The underlying problems are real, lower average IQ, lack of impulse control, a culture that idolizes crime, poor to non-existent parenting and excessive family size. I think that in non-white populations that crime rates 4-10 times higher than white populations will continue to be the norm.

A certain commentator seems missing here.

Isn't race endogenous in Brazil? Two people with identical "racial" features may be identified as belonging to different "races" if they have different socioeconomic statuses. This is very different from the situation in the United States, in which race is mainly exogenous.

Damn I'm feeling really WHITE today!

Thus, the twin study...all the IQ-and-race trolls in this comments section can't seem to even come close to reading the damn abstract, much less the paper.

The twin study is a straw man ("skin color").

My point is that an American reader will see "twin study" as meaning "same race" (as it would in the United States), but a Brazilian reader will not necessarily see that connection, because "race" in Brazil is not entirely physical. Identical twins in Brazil can, in Brazilian eyes, be of different races. So race is endogenized in Brazil, and thus not held constant, as it would be in a twin study in the United States.

Comments for this post are closed