There is a long, dubious history of people separating humanity into groups based on appearance but it has consistently been arbitrary and illogical. Mostly racial groups are based on phenotypic characteristics (skin, hair, and eye color and head size and shape). The application of this simple classification is pretty obvious, especially when mapped out where we’re to believe that the entire continent of Africa consists of a homogeneous group and people that straddle arbitrary political borders are intrinsically different:

Johann Blumenbach’s 1775 racial typologies mapped onto the modern world. Modern “races” are derived from Blumenbach’s typologies.

When you tack on a politically correct filter, it gets more insane, as seen in the “Check Your Race” form list. The options often are Caucasian, African-American, White Hispanic, Non-White Hispanic, Asian, Native American/Pacific Islander, or my favorite race, “Other.” In this PC parody, you must identify as a mountain range, an American inextricably linked to another continent, a pale speaker of a specific language, a speaker of a specific language with more melanin, a really, really big continent, or an arbitrary collection of everyone else. It’s absolutely absurd, but that’s because race is absurd. It’s trying to whittle 7.5 billion people down into a few tidy categories. And when it’s done is a PC culture, it’s even more absurd, making the designations laughably arbitrary.

I knew a Hebrew South African girl who spoke Spanish. She could legitimately check off all those boxes and perhaps the last one for good measure.

Not only is this classification system laughably arbitrary, it also appears to be genetically insignificant. As Population geneticist Luigi Luca Cavalli-Sforza explains, these stereotypes, “reflect superficial differences that are not confirmed by deeper analysis with more reliable genetic traits.”

A closer approximation of race theory would look something like the map based on Y-Chromosome haplogroups:

There clearly are thousands of genetic markers that reveal ancestry from various genetic clusters in different parts of the world, but the boundaries are problematically ambiguous and there are exceptions throughout.

More important, though, is the fact that race (even the highly segmented constructs) is a poor indicator of human variation.

In 1972, Richard Lewontin performed a fixation index (FST) statistical analysis using 17 markers, including blood group proteins, from individuals across classically defined races (Caucasian, African, Mongoloid, South Asian Aborigines, Amerinds, Oceanians, and Australian Aborigines). His findings, which have been repeatedly verified, showed that of the roughly 0.5% of DNA variation between humans, the vast majority is found within races and a relatively insignificant amount of variation is found between races.

Surprisingly, he found most of the genetic variation (85.4%) within specific population groups (eg Austrians, Nigerians, Navajo), then another 8.3% between populations within a race. Only the remaining 6.3% of genetic variation between individual humans was accounted for by race.

Another way to look at it is relative to the percentage of genes that are similar across humanity:

Legal scholar Dorothy Roberts says, “Race [however defined] collapses infinite diversity into a few discrete categories that in reality cannot be demarcated genetically.” In other words, race has very little if any legitimate application.

Visualizing this is difficult, but a rare pair of fraternal twins named Lucy and Maria Aylmer help. They are biological twins with the same biological parents born at the same time—as close as you can get genetically apart from identical twins—and yet they look completely different. Many even claim that they are different races:

The point here being that the phenotypes that make up the traditional race categories aren’t as meaningful as people seem to think and that genetic variation even with offspring of the same sexual partners accounts for a vast majority of the variation across humanity.


There are some diseases that are linked with race. For instance, sickle cell anemia is associated with Africans (1 in 13 are at risk). It may be a useful rule of thumb in a medical emergency but simplified associations like these are not scientifically sound and may lead to increased risk. There are other populations that have a high risk (Middle Eastern, Indian, Mediterranean) and there is a high variance within the African population. According to one study, “those with the haplotype designated Senegal have decreased severity, those with the Benin haplotype have intermediate severity, and those with the Central African Republic (CAR) haplotype have the most severe clinical expression.” Race is not nuanced enough to be medically accurate, even for such diseases like sickle cell.