Race Is a Human Invention - But That Doesn't Make It Useless

Race remains one of the most controversial concepts in modern society, often caught between outdated prejudices and claims that it has no scientific basis at all. In this thought-provoking essay, the author challenges both extremes, arguing that the debate has become clouded by politics, ideology and misunderstanding. Drawing on developments in genetics, evolutionary biology and medical research, the piece explores whether race can still serve as a useful scientific classification—and where its limits lie. The result is a nuanced examination of a deeply sensitive subject that raises difficult questions about identity, human diversity and the pursuit of knowledge..By Garth Zietsman*.A lot of social science – beginning with the Victorians – uses the concept of race to analyse humanity. Much of that analysis is uncomplimentary toward non-white groups. It appears self-serving toward the British Empire and is frankly just racist.A common woke response is that race does not exist and that any science that uses the concept is pseudo-science. Both approaches are wrong and oversimplify things in unhelpful ways. In this essay I hope to introduce some clarity.The concept of race is an attempt to classify humans into categories in order to assist scientists in thinking about humanity. In the same way, the concept of species helped us understand life forms like animals, insects, plants, fungi or bacteria. Classifying matter – for example, into types of sub-atomic particles – is helpful to physics.It is difficult for humans to think without using classifications and is especially so in Western abstract reductionist analytical thinking. These categories help in the formation of scientific theories and predictions. For example, set theory is foundational to mathematics and knowledge, and what is set theory but a classification scheme? The important question is whether the classification scheme is helpful to scientists or is it just arbitrary?The way scholars classified humans in the past was not useful, but it could be.Let me begin by specifying what race is not. Races are not essences passed down by nature or God that are immutable for all time. No racial classification scheme can neatly classify every individual. Race as a classification scheme is a human invention, but one meant to model reality. It is a fuzzy set scheme with some noise at the borders, much like the concept of species.Zoologists tried to make the concept of species hard by claiming that the inability to interbreed (produce fertile offspring) defined species boundaries. That did not work. Blue and fin whales, beluga and narwals, wolves and coyotes, brown and polar bears, various birds and bacteria all interbreed in the wild.We can crossbreed plants quite easily. Many species that do not interbreed in the wild can readily produce fertile offspring in captivity, such as, for example, most small cats, or dogs and some jackals. Some can produce partial fertility – female ligers (hybrid offspring of a male lion and a female tiger) are fertile, but males are not. Even humans have interbred with different hominid species – Neanderthals and Denisovans. There is evidence that one ‘race’ of Denisovans interbred with a more archaic species like Homo Erectus.Race and speciesObviously race and species are not the same concept.Before the genetics revolution, scientists used physical characteristics to classify creatures into species (Linnean), and people into races. These systems are quite consistent, and genome data confirms them.There are however some surprises. For example, hippos turned out not to be a sort of pig but a sort of whale instead. Scholars assigned people to races based upon superficial features like eye, hair and skin colour or morphology, but also the shape and proportion of the skeleton, eyes or noses.Critics made a mistake in claiming that these criteria are purely ‘skin deep’ and that humans are identical on the ‘inside’. AI can use X-ray and scan images of every organ to accurately classify people into self-identified racial categories, even when radiologists have no idea what features the decisions are based upon and when the images are fuzzy. We also know that races differ on disease susceptibility, and on responses to drugs, so differences extend beyond anatomy to physiology.One important critique is that, with few exceptions, every gene exhibits more variation within each race than there is between them. The critique is correct. On average only about 15% of the gene variation is between races. What the critics miss is that a variety of one gene tends to correlate with varieties in other genes, so that differences in big patterns become predominantly between, rather than within, races.Furthermore, it is possible that those rare gene varieties that do differ more across than within races, like skin colour, could be quite important. A good example is the gene for sickle cell anaemia. Another important claim by critics is that genetic variation does not show sharp breaks but is a cline – that there is a steady gradual change from one extreme of races to the other. Try to visualise a bar, solid red at one end changing imperceptibly through pink to pure white at the other. That situation would make colour or racial classification at any point a complete farce.The genetic pictureThe critics are wrong on this claim. Principal component analysis is a statistical method of using correlations between elements to reduce the overall variance into a few broad interpretable factors. If one looks at a picture of the first two principal components of genome variation, which account for 60% of the total variation, you see that the genetic patterns of ancestral groups, ethnicities or tribes are quite lumpy. There is no smooth transition all the way through but clear separation between clusters. The genetic picture fits the pre-genetic global racial classification very closely.On the other hand, the principal component picture points to a clear deficiency in the Victorian three-race classification of Negroid, Caucasian and Mongoloid. One can see that some African groups are closer to the European cluster than they are to other African groups. The same applies to the Asian groups. It is clearly more accurate to use a more fine-grained scheme to classify African and Asian populations. In addition, the three-race scheme leaves out Australian/New Guinean and new world populations altogether. A twelve-race scheme accounts for all the clusters across the whole globe much more reliably.A good way to think of races is that they are slightly inbred extended families.Earlier I said that race is not essential or immutable. The ability to extract and analyse ancient DNA tells us that almost every population today is the result of a mixture of several distinct ancient populations.In the case of Europe, these were Western Hunter Gatherers (the Ice Age cave painters), Early European Farmers from Anatolia (builders of stone monuments like Stonehenge) and Indo-Europeans (the Yamnaya who domesticated the horse) from the Pontic Steppes. DNA samples from ten thousand years ago also produce a principal component picture but one which is completely different from the current picture. The ingredients of race shift and recombine all the time. Race is not fixed. Your distant ancestors belonged to a different race (or races) than you do, and so will your distant descendants.Why bother?If that is the case, you may well ask why bother classifying humans into races at all?The most contentious of reasons are political. In South Africa racial classifications were, and are once again, used to exclude or otherwise oppress some ancestral groups. This kind of thing happens everywhere, and we rightly call this racism and condemn it. Racial classifications also encourage the pernicious practice of identity politics, although that would happen without an official racial classification.A better reason is that a race classification that captures most of the genetic variance in as few sets as possible makes studies of human adaption to various contexts more efficient.Reducing the number of studies, while minimising the genetic differences within groups, provides much more bang for buck. Examples of good target studies include how immune systems have adapted to particular disease contexts, how metabolisms have adapted to regional diet, which groups are more prone to high cholesterol or insulin resistance, or how they respond to drugs.The first would include phenomena like sickle cell anaemia and other malaria adaptions, and smallpox or plague resistance. The second would include the development of different paths to the ability to digest milk as an adult, adaptions to high (Peru, East Asia) and low (Eskimos) carbohydrate consumption, and the short stature of pygmies as an adaption to exceptionally low iodine levels. This knowledge helps deliver medical care or disease prevention in a more targeted and effective way.It makes a trivial difference how researchers classify groups in order to investigate the causes of group socio-economic status differences. Researchers need to justify the use of race as a classification scheme on a case-by-case basis. If race is the basis of identity politics in a particular country, then using the accepted racial classification is the appropriate option.Where the political outlook is individualist, it would not be an appropriate option because there would be no point in classifying people in ways that make identitarian social issues salient. Rather than banning the use of race in research, the choice to use it would have to be subject to open critique – for example, a group lumping together diverse populations such as native-Americans and Pacific Islanders or South and East Asians.Sometimes authorities try to change a label of something they think has become negative, such as using the ‘disabled’ instead of ‘crippled’. It does not help. People know it applies to the same thing and simply transfer all the meanings across. The same applies to race. A different way of classifying people would result in similar controversies, so we may as well continue to use the concept, hopefully in an informed and wiser manner..*Garth Zietsman is a professional statistician who initially focused on psychological and social research at the Human Sciences Research Council, followed by banking and economics, and then medical research. .This article was first published by The Daily Friend and is republished with permission..Sign up for your early morning brew of the BizNews Insider to keep you up to speed with the content that matters. The newsletter will land in your inbox every morning on weekdays. Register here.Support South Africa's bastion of independent journalism, offering balanced insights on investments, business, and the political economy, by joining BizNews Premium. Register here.If you prefer WhatsApp for updates, sign up to the BizNews channel here.

Race Is a Human Invention - But That Doesn't Make It Useless

Related Stories

Ivo Vegter: Stop fearing Artificial Intelligence. Start fearing its would-be masters

Is the AI arms race about to outrun human judgment? The Economist thinks so

Mark Oppenheimer: Joburg Bar's undeniable race-based admissions

South Africans lean non-racial despite political rhetoric – IRR poll