Race: The Social Construction of a Biological Reality, pt 2

Note: This post still contains a lot of oversimplification for the sake of explaining a few things.

Welcome back to our discussion of the geographic dispersion of humanity. On Tuesday, we discussed how two great barriers–the Sahara desert and the Himalayas + central Asian desert–have impeded human travelers over the millennia, resulting in three large, fairly well-defined groups of humans, the major races: Sub-Saharan Africans (SSA), Caucasians, and east Asians.

Of course, any astute motorist, having come to a halt at the Asian end of our highway, might observe that there is, in fact, a great deal of land in the world that we have not yet explored. So we head to the local shop and pick up a better map:


Our new map shows us navigational directions for getting to Melanesia and Australia–in ice age times, it instructs us, we can drive most of the way. If there isn’t an ice age, we’ll have to take a boat.

900px-oceania_un_geoscheme_-_map_of_melanesia-svgThe people of Melanesia and Australia are related, the descendants of one of the first groups of humans to split off from the greater tribe that left Africa some 70k ago.

As the name “Melanesian” implies, they are quite dark-skinned–a result of never having ventured far from the equatorial zone.

Today, they live in eastern Indonesia, Papua New Guinea, Australia, and a smattering of smaller islands. (Notably, the Maori of New Zealand are Polynesians like the Hawaiians, not Melanesians, descendants of a different migration wave that originated in Taiwan.)

Fijian mountain warrior
Fijian mountain warrior with curly, “African” style hair

There is some speculation that they might have once been wider-spread than they currently are, or that various south-Asian tribes might be related to them, (eg, “A 2009 genetic study in India found similarities among Indian archaic populations and Aboriginal people, indicating a Southern migration route, with expanding populations from Southeast Asia migrating to Indonesia and Australia,”) but I don’t think any mainland group would today be classed as majority Melanesian by DNA.

They may also be related to the scattered tribes of similarly dark-skinned, diminutive people known as the Negritos:

Males from the Aeta people (or Agta) people of The Philippines, are of great interest to genetic, anthropological and historical researchers, as at least 83% of them belong to haplogroup K2b, in the form of its rare primary clades K2b1* and P* (a.k.a. K2b2* or P-P295*).[7] Most Aeta males (60%) carry K-P397 (K2b1), which is otherwise uncommon in the Philippines and is strongly associated with the indigenous peoples of Melanesia and Micronesia. Basal P* is rare outside the Aeta and some other groups within Maritime South East Asia. …

Naural blond hair
Two Melanesian girls from Vanatu (blond hair is common in Melanesian children.)

A 2010 study by the Anthropological Survey of India and the Texas-based Southwest Foundation for Biomedical Research identified seven genomes from 26 isolated “relic tribes” from the Indian mainland, such as the Baiga, which share “two synonymous polymorphisms with the M42 haplogroup, which is specific to Australian Aborigines“. These were specific mtDNA mutations that are shared exclusively by Australian aborigines and these Indian tribes, and no other known human groupings.[12]

A study of blood groups and proteins in the 1950s suggested that the Andamanese were more closely related to Oceanic peoples than African Pygmies. Genetic studies on Philippine Negritos, based on polymorphic blood enzymes and antigens, showed they were similar to surrounding Asian populations.[13]

Negrito peoples may descend from Australoid Melanesian settlers of Southeast Asia. Despite being isolated, the different peoples do share genetic similarities with their neighboring populations.[13][14] They also show relevant phenotypic (anatomic) variations which require explanation.[14]

In contrast, a recent genetic study found that unlike other early groups in Malesia, Andamanese Negritos lack the Denisovan hominin admixture in their DNA. Denisovan ancestry is found among indigenous Melanesian and Australian populations between 4–6%.[15][16]

Australian Aboriginal man
Australian Aboriginal man

However, the Negritos are a very small set of tribes, and I am not confident that they are even significantly related to each other, rather than just some short folks living on a few scattered islands. We must leave them for another day.

The vast majority of Aborigines and Melanesians live in Australia, Papua New Guinea, and nearby islands. They resemble Africans, because they split off from the rest of the out-of-Africa crew long before the traits we now associate with “whites” and “Asians” evolved, and have since stayed near the equator, but they are most closely related to–sharing DNA with–south Asians (and Indians.)

So we have, here, on the genetic level, a funny situation. Melanesians are–relatively speaking–a small group. According to Wikipedia, thee are about 12 million Melanesians and 606,000 Aborigines. By contrast, Tokyo prefecture has 13 million people and the total Tokyo metro area has nearly 38 million. Meanwhile, the Han Chinese–not a race but a single, fairly homogenous ethnic group–number around 1.3 billion.

Of all the world’s peoples, Melanesians/Aborigines are most closely related to other Asians–but this is a distant relationship, and those same Asians are more closely related to Caucasians than to Aborigines.

As I mentioned on Tuesday, the diagram, because it is 1-dimensional, can only show the distance between two groups at a time, not all groups. The genetic distance between Caucasians and Aborigines is about 60 or 50k, while the distance between Asians and Caucasians is around 40k, but the distance between Sub-Saharan Africans and ALL non-SSAs is about 70k, whether they’re in Australia, Patagonia, or France. Our map is not designed to show this distance, only the distances between individual pairs.

Some anthropologists refer to Bushmen as "gracile," which means they are a little shorter than average Europeans and not stockily built
Some anthropologists refer to Bushmen as “gracile,” which means they are a little shorter than average Europeans and not stockily built

Now if we hopped back in our car and zoomed back to the beginning of our trip, pausing to refuel in Lagos, we’d note another small group that has been added to the other end of the map: the Bushmen, aka the Khoi-San people. Wikipedia estimates 90,000 San and doesn’t give an estimate for the Khoi people, but their largest group, the Nama, has about 200,000 people. We’ll estimate the total, therefore, around 500,000 people, just to be safe.

The Bushmen are famous for being among the world’s last hunter-gatherers; their cousins the Khoi people are pastoralists. There were undoubtedly more of them in the past, before both Europeans and Bantus arrived in southern Africa. Some people think Bushmen look a little Asian, due to their lighter complexions than their more equatorial African cousins.


Various Y chromosome studies show that the San carry some of the most divergent (oldest) human Y-chromosome haplogroups. These haplogroups are specific sub-groups of haplogroups A and B, the two earliest branches on the human Y-chromosome tree.[48][49][50]

Mitochondrial DNA studies also provide evidence that the San carry high frequencies of the earliest haplogroup branches in the human mitochondrial DNA tree. This DNA is inherited only from one’s mother. The most divergent (oldest) mitochondrial haplogroup, L0d, has been identified at its highest frequencies in the southern African San groups.[48][51][52][53]

I loved that movie
The late Nǃxau ǂToma, (aka Gcao Tekene Coma,) Bushman star of “The Gods Must be Crazy,” roughly 1944-2003

In a study published in March 2011, Brenna Henn and colleagues found that the ǂKhomani San, as well as the Sandawe and Hadza peoples of Tanzania, were the most genetically diverse of any living humans studied. This high degree of genetic diversity hints at the origin of anatomically modern humans.[54][55]

Recent analysis suggests that the San may have been isolated from other original ancestral groups for as much as 100,000 years and later rejoined, re-integrating the human gene pool.[56]

A DNA study of fully sequenced genomes, published in September 2016, showed that the ancestors of today’s San hunter-gatherers began to diverge from other human populations in Africa about 200,000 years ago and were fully isolated by 100,000 years ago … [57]

So the total distance between Nigerians and Australian Aborogines is 70k years; the distance between Nigerians and Bushmen is at least 100k years.

When we zoom in on the big three–Sub-Saharan Africans, Caucasians, and Asians–they clade quite easily and obviously into three races. But when we add Aborigines and Bushmen, things complicate. Should we have a “race” smaller than the average American city? Or should we just lump them in with their nearest neighbors–Bushmen with Bantus and Aborigines with Asians?

I am fine with doing both, actually–but wait, I’m not done complicating matters! Tune in on Monday for more.

Is Race a Social Construct?

People mean a lot of things when they say “social construct.” Mostly they mean “made up.”

Luckily for us, Google is very helpful:

I may be abusing the word "luckily"

Dear Google and the NY Times: Not only is that not the biological definition of race, it’s not even the biological definition of SPECIES. This is not what laymen mean when they speak of race, not what racists mean when they speak of race, not what blacks or whites or Hindus mean, and definitely not anyone who actually studies human biology and genetics.

The simple folk definition of “race” is “a group of people who look kinda similar and come from the same large area of the world.” This, of course, absolutely exists. Most people in the world look a lot like their neighbors, especially if they live in their ancestral homeland and their country hasn’t been invaded lately.

Now, the exact details of how you racially classify people–are Hindus Caucasian? How about North Africans and Iranians? What about mixed-race people?–are socially constructed. This mean that a word like “black” may mean something different in Russian than it does in the Dominican Republic than it does in the US.

This does not change the underlying reality–the humans referred to as “black” still possess the quality of looking similar to other people from their ancestral part of the world. Reality does not disappear just because people sometimes disagree on exactly how to use words to define it.

The scientific, biological definition of race gets a little more complicated, since it matters whether we are talking abut chromosomal races, fungal races, or humans. A couple of definitions:

Geographical race
A distinct population that is isolated in a particular area from other populations of a species,[9] and consistently distinguishable from the others,[9] e.g. morphology (or even only genetically[4]). Geographic races are allopatric.[7]
Physiological race
A group of individuals that do not necessarily differ in morphology from other members of the species, but have identifiably different physiology or behaviour.[10] A physiological race may be an ecotype, part of a species that is adapted to a different local habitat, defined even by a specific food source.[11]

Notice that neither of these include, “possessing a gene or cluster of genes common to everyone in the race but to no one outside of it.”

But if you don’t like the Wikipedia, here’s what Biology Reference has to say:

The biological definition of race is a geographically isolated breeding population that shares certain characteristics in higher frequencies than other populations of that species, but has not become reproductively isolated from other populations of the same species. (A population is a group of organisms that inhabit the same region and interbreed.) Human racial groups compose a number of breeding units that in the past remained geographically and perhaps temporally isolated, yet could interbreed and produce viable offspring within the species Homo sapiens sapiens.

The Biology-Online Dictionary has some more definitions.
These races are real things, even if biologists disagree about exactly which race a mushroom should belong to.
The reality on the ground:
There are few truly isolated groups in the world, though the Onge (and most likely the Sentinelese) actually fit the NY Times’s wacky definition of a “race” due to thousands of years of isolation on tiny islands in the middle of nowhere:
Click for full size
From Haak et al.
The Onge are the peach stripe between the olive brown section and the purple section.
Major groups in this dataset, running from left to right (excluding the ancient skeletons at the far left):
Light Green: Brazilian rainforest dwellers who may be most closely related to Melanesians
Light Pink: Aztecs and their relatives
Brown: Canadian Indians
Rose: North-East Africans
Dusty Blue: Bantus
Light Blue: Pygmies
Magenta: Tanzanian hunter-gatherers
Orange/Blue/Teal: Europeans
Orange/Purple/Teal: Middle Easterners
Olive Brown: Inuit (Eskimo)
Peach: Onge
Purple: PNG/Australia (Melanesians and Aborigines)
Light Green/Teal: India
Yellow/Red: East Asia
Yellow: Taiwan
Red: Siberia
Some of these groups have very mixed ancestries; people from eastern Canada or the middle of Eurasia, for example. Others are quite distinct–there is no doubt that the Eskimo and Pygmies are genetically distinct, physically distinct, geographically distinct, behave differently, and do not generally marry each other.
We may argue about whether Turks should be considered “Europeans” or “Middle Easterners,” or perhaps say that all orange people should be grouped together, or all teal or blue, but here geography does its job: Europeans look genetically like other Europeans; Indians look genetically like other Indians; Middle Easterners look like each other (except for Bedouins,) etc.
We may also argue about how many races we want to distinguish–people usually determine races based on whichever people are around, but obviously the world is more complicated than this. Americans generally think of “African Americans” as part of a broader race that includes all Africans, but we have distinguished here between 5 different groups, some of which are quite distinct–the ancestors of today’s Pygmies and Bantus, for example, split apart about 100,000 years ago, whereas the ancestors of today’s Bantu’s and Koreans split about 70,000 years ago (as far as we know.) Most African Americans are genetically Bantu (with a bit of European admixture,) not Pygmy. We might in a folk-sense refer to both of these groups as “Africans” or “black,” but genetically (and behaviorally) they are distinct.
Of course, you do not have to call them “races.” Most people studying human genetics use terms like “population,” “ethnic group,” “ethny,” or “clade” instead, but the practical meaning is the same.
But the idea that groups that are genetically, physically, behaviorally, or geographically distinct or distinguishable from each other do not exist is pure nonsense.