More Recent Comments

Friday, November 22, 2024

Kostas Kampourakis says that human races do not exist

Populous, geographically disperse, species are often subdivided into subspecies, races, demes, or geographically differentiated populations. Homo sapiens is no exception; there are many subpopulations whose overall genetic compositions are significantly different—so different, in fact, that we have no trouble identifying members of those subpopulations and the sequence of their genomes can assign them to the different groups.1

We could use the word "race" to distinguish the largest of these subpopulations if the word wasn't so loaded with non-scientific meaning. I believe that, from a scientific perspective, humans races exist. [Do Human Races Exist?]. Jerry Coyne is much better (braver?) than I at defending the biological and evolutionary reality of human races and attacking the well-meaning, but mistaken, attempt to deny the existence of human races. [Genetic ignorance in the service of ideology] It's part of a larger effort to combat something he calls The Ideological Subversion of Biology. The point he's making is that we are teaching our children a number of misconceptions that conflict with science and this contributes to a mistrust of science.

There are two sides to this story. Kostas Kampourakis does not believe that human races exist. He has been awarded a "Friend of Darwin" award by the National Center for Science Education (NCSE) so I assume that NSCE—a strong promoter of science education in the United States—endorses his views on human races.

Here's a seminar that Kampourakis gave to the Teacher Institute for Evolutionary Science (TIES) in February 2023. According to their website, "TIES provides middle school and elementary teachers the tools they need to effectively teach evolution and answer its critics based on new Next Generation Science Standards." I assume this means that they want their science teachers to teach that human biological races do not exist.

Watch the video to see how Kampourakis presents his version of the scientific case against human races. It will give you a good idea of why this controversy is so difficult to resolve.

"Examines the differences between race, nationality, and ethnicity and the extent to which these concepts are socially constructed, not genetically determined."



1. This is not to deny that interbreeding between the groups can take place, they are, after all, part of the same species. That means that many individuals carry DNA markers from one or more groups and if this mixing continues in the future the distinctiveness of the groups will disappear.

1 comment :

Zach Hancock said...

As a population geneticist whose career is focused on distinguishing discrete from continuous genetic structure, human races don't exist. But we need to be clear about what that means.

To say "human races don't exist" is not the same as saying there are no genetic differences in humans. Of course there are. There are genetic differences between you and your parents, between you and your cousins, etc. It also doesn't mean that you can't cluster humans into groups - you could, for example, if you trained a large enough dataset on human sequences tell someone with a high degree of confidence where their ancestors lived on the planet. This is due to the very simple fact that dispersal is limited and so there are correlations between ancestry and geography.

Next, we need to be clear in that what we're talking about is whether our social concepts of race correspond to biological races as defined taxonomically. Importantly, taxonomically, we create groups to represent discrete biological entities, and these discrete entities need to display a few characteristics to be defined as such. For example, they should have more diversity between than within the race. Traits used to define "race" should be monophyletic (e.g., if I say "black skin" defines the "African race," then all individuals with the trait should be more related to each other than they are to any other skin color). And, importantly, ancestry should show discrete breaks - it should not be continuous across space. For biological populations with continuous population structure, any cluster is necessarily arbitrary (and hence not biologically real).

Humans do not satisfy any of these criteria. Most of the diversity among humans is shared across populations, with out-of-Africa populations being merely a subset of the diversity within Africa. Skin color is not a trait that reliably reflects ancestry - there are African populations that are more related to Europeans than they are to other Africans, and Melanesians are more related to Asian populations than to Africans, despite having the same skin color. And, most importantly, human ancestry is completely continuous. The probability you're related to someone is a simple function of geographic distance. Taxonomically, we do not break-up continuous populations into discrete units; if we do, taxonomy loses any objectivity it might have.

So we're faced with three options: 1) arbitrarily define biological races to only *kinda* correspond to our social concepts, but in a way that we wouldn't define any other living thing with (i.e., it wouldn't pass peer-review in a taxonomic journal). 2) define "race" as a town-by-town, family-by-family kind of thing in which the bounds of the genetic cluster is arbitrary. or 3) admit that social races don't correspond to biological reality - human ancestry is continuous, not discrete, with no traits that identify a race socially being reliable to do so biologically.

I don't think this is a "woke" or "politically correct" take - I think it's empirical that there are no human races, and the majority of population geneticists working in human genetics agrees.