I have decided to start a new series that I am calling (R)anthropology Class. I draw on my anthropological training in so many different ways, and I know that this training is what has helped me view the world in a critical and objective way. Because I know that most people have never taken an anthropology class – or if they have, it was long ago – I have decided to focus some of my posts on some of the most basic, but important, anthropological concepts. These usually occur to me when I am in the midst of trying to make sense of some new idea or trying to explain my point of view to others. That was the case recently when I was trying to explain to someone why DNA ancestry tests can be extremely misleading for people who don’t know much, if anything, about population genetics. So with this inaugural (R)anthropology Class rant, I will sketch in the basics of why a commercial DNA ancestry test cannot tell you your race.

There is no question that race is a real thing. All we have to do is look at the people around us to know that this is true: different skin colors, hair textures, facial characteristics, even height and body type show us that people are different from one another. Yet, from a genetic standpoint, race is not a real thing at all. Instead, what we call race is a cultural construct that reflects the human need to seek and identify patterns in our surroundings that help us to understand and categorize our world. Race, in other words, is cultural rather than biological.

How can I say such a thing? All you have to do is look at someone to know what race they are – right? Dark skin = African. Epicanthic folds in the eyelids = Asian. Blue eyes = European.** And if a person happens to have parents of two different races, then that person will show a blend of different racial characteristics from his or her parents. Or if a person is descended from several different races, they will still show some traits that help you identify those ancestral races – or so we like to think. But ask any person who identifies as mixed race and you will find that their lives are full of mistaken assumptions about what their race is – and concomitantly full of different types of treatment depending on what people might unconsciously assume their race to be. Again, this is all entirely based on cultural categories, and is part of what anthropologists call ascribed status. An ascribed status is a status that a person can’t do anything to change – such as age, gender, or in this example, race. But people make mistakes in the statuses they ascribe to others all the time. When I teach my students about this in my lecture on race, I ask them if they have ever been mistaken for a race other than the one they assign to themselves. The hands of my students of color always, without fail, shoot into the air. And, sometimes, my students who look white raise their hands, too – and they surprise their classmates by identifying as having African-American, non-white Hispanic or Latino, Asian, or some other non-white racial ancestry.*** I have Filipino students tell me they are mistaken for Latino or Middle Eastern; I have Asian students describe how they are always ascribed to the Japanese or Chinese category when they are actually Korean or Vietnamese or Thai; I have students with roots in countries throughout South America tell me that nobody seems to know that there are countries other than Mexico south of the US border.

Here’s the deal, biologically: humans are 99.999% genetically identical. That means that only one out of every 1000 DNA nucleotides is different between any two individual humans. But that tiny .001% difference is reflected in some very recent, visible physical differences between human populations. Human physical traits – called phenotypes – have evolved based on adaptation to specific geographic regions and the pressures of natural selection within those regions. So, natural selection results in phenotypic variation in traits like skin color. Skin color has evolved in response to sun exposure and vitamin D metabolism – the further north a population lives, the less sun they get, which means the less essential vitamin D they are able to metabolize. So by virtue of natural selection, lighter skin color that allows for more efficient vitamin D absorption has evolved in human populations that live in low-sunlight areas, whereas the melanin that causes darker skin has remained abundant in populations closer to the equator. Hence, populations in equatorial Africa are very dark, whereas populations in far northern Europe are very light. This same sort of natural selection has operated on other genes as well, resulting in a wide variety of phenotypes throughout the world. And naturally, those phenotypes remain clustered within the populations where they evolved, which makes it simple for pattern-seeking humans to use those phenotypes to categorize people into the physical types that we have labelled “races.”

Another important point about phenotypic variation is that it is continuous. In other words, there is no sharp, clear dividing line between different types. If you were to line up every person in the world in order from palest skin to darkest, where would you draw the line between dark and light? Or even if you came up with more categories – pale white, medium white, light tan, dark tan, light brown, etc. – where would you put those lines? It’s like trying to pinpoint the exact spot in a rainbow where the color turns from red to orange – it can’t be done! And yet the rainbow continuously shifts in colors until you go all the way from red at the beginning to violet at the end. Continuous human phenotypes operate in exactly the same fashion. Consider, also, that if you were to draw skin color lines in the human rainbow, you would find individuals from several different races or ethnicities within a single skin color category – Australian aborigines, east Indians, and sub-Sarahan Africans could all be found within one dark-skinned group! A bottom-line way of putting it is this: there is no single trait that can be found in one so-called racial group that does not also exist in some other so-called racial group. You can find dark skin in several groups, epicanthic folds in several groups, and blue eyes in several groups. Race as biology is a cultural fiction.

So, what does this have to do with DNA ancestry tests? I have serious misgivings about the way these tests are marketed because they trade on people’s lack of knowledge about the biological fiction of race and give them the impression that they are finding out about their own supposed racial ancestry. In fact, if not strictly unethical, I think that the companies who peddle these tests are at best taking advantage of people’s forgivable ignorance about the complexity of genetics. Now, I’m not saying that people can’t or even shouldn’t research their ancestry if it interests them; it would be fascinating to find out that what you thought was your completely European ancestry actually had, say, a branch from a part of Asia. But when I say that, I’m talking about genealogical, not genetic, ancestry research. Genetic ancestry research cannot tell you that you have an Asian ancestor; it can only tell you if you have genetic markers that are associated with particular broadly-defined genetic populations.

DNA ancestry tests use what are called haplogroups to assess genetic ancestry. A haplogroup is a group of similar genes – called haplotypes – that reflect single nucleotide polymorphism (SNP) mutations. What is important about SNP haplogroups is that they can be used to broadly delineate genetic populations. This goes back to the discussion above about phenotypic traits that arise in particular geographic regions in response to specific selective pressures. These haplogroups can be traced in two ways: either on the Y chromosome, which is only present in males; or in mitochondrial DNA (mtDNA), which is a separate set of DNA from the nuclear DNA that codes for our particular physical traits. Mitochondrial DNA is only found in our cells’ mitochondria, and we get it only from our mothers. Both Y haplogroups and mtDNA haplogroups are very stable and have a very slow rate of mutation, so they remain relatively unchanged for long periods of time. This means that we can compare these haplogroups in people today to ancient haplogroups associated with particular regions and populations. Commercial DNA ancestry tests look at an individual’s haplogroups and compare the results to known population haplogroups.* These results are used to complete a statistical analysis of a person’s possible ancestry. So, if you have a haplogroup associated with Asia, your DNA test results will say so.

Here’s where things get problematic. Most people don’t know all the things about DNA and populations genetics that I am writing about in this post, so when they see a result of, say, 12% African, they think it means they are “part Black.” I can’t stress enough that this is not what these results mean. What it means is that the person has a haplogroup that is associated with known ancestral African genetic populations. It’s a statistical correlation, not an absolute. And things get even trickier when you realize that Y and mtDNA haplogroups can be incredibly diverse even within a seemingly homogenous regional population. In fact, population geneticists know that there is more variation within the groups we call races than there is between the groups we call races. I am just as likely to share identical mtDNA ancestry with someone from Asia – where I have no ancestors that I know of – as I am to share it with someone from Sweden, where I know my immediate ancestors came from. Don’t get me wrong – I’m not saying we can’t learn anything from comparing haplotypes – we can. But it is completely wrong to say that you are “part Black” or “part Native American” or “part Asian” based on DNA ancestry testing. All it tells you is that you have a haplogroup that could have entered your genetic lineage thousands of years ago that derived from that part of the world.

This has been a long post, and it is a complicated subject. I have no doubt that some people will read this and misunderstand. Let me part with this: if you want to get a DNA ancestry test, feel free. Just don’t make the mistake of thinking it is telling you anything about your race – it’s not. And finally, never forget that even though what we think of as race is not biologically real, it is still a complex, vital, and unmistakable social reality. We continue to treat people differently on the basis of it; and some people even still insist that the behaviors we associate with race and ethnicity have a genetic basis. We are not that far from the days when the civil rights of African Americans, Native Americans, and other people of color were denied because of the belief that they were genetically inferior to white people. In that sense, race is historically, culturally, and painfully real.

*Since I wrote this post in February 2015, the technology used by commercial testing companies has expanded to include autosomal DNA haplogroups. Autosomal DNA is the DNA found in the nucleus of all your body’s cells. You get 50% of your autosomal DNA from your mother and 50% from your father. Using autosomal DNA allows a DNA test to see your results from both your maternal and paternal lineages, but it does not automatically mean you will get a more accurate picture; because of the process of meiosis, which is how sperm and eggs are made, each sperm or egg only has half of a person’s DNA. That means that every sperm and egg is essentially unique, and does not contain every possible haplogroup that is part of a person’s autosomal DNA. So, the sperm and egg that made YOU does not have every one of your parents’ haplogroups; and if you have kids, they won’t have all of your haplogroups, either. This is why even full siblings often will not have the exact same results, because each person carries a unique combination of DNA. (Edited January 27, 2018)

**A comment from a reader pointed out that there are problems with the use of the term Caucasian – problems, embarrassingly, that I had never considered, but which seem obvious to me after a little bit of reading and reflection. I have edited the post to replace the word Caucasian with European or white. This article by Yolanda Moses provides a compelling and succinct explanation for why we need to stop using the term. (Edited February 13, 2018)

***A comment from a reader pointed out a lack of clarity here, given that there is a subset of Latino and Hispanic that falls within the white European racial category. (Edited January 8, 2019)