Using “genomic thinking” to ‘re-think’ human races is a mistake: Part 2 - praxis

Using “genomic thinking” to ‘re-think’ human races is a mistake: Part 2 - praxis

Emeritus Prof. Tim Crowe

So why resurrect/reify human races?

Does genomic racial ‘diagnosability’ demonstrate that human races have ontological, evolutionary, or any other, status?  Does it detect evolutionarily significant units?
No.
Like species, races/subspecies in taxonomy are discovered.  They are species in limbo; self-defined by consilient, geographically congruently varying characters (taxonomic features that are locally invariant) within a core geographical area.  They are NOT species because they are still genetically ‘permeable’ metapopulations and not evolutionary lineages, and are circumscribed geographically by character ‘suture zones’ within which character congruence breaks down because of unimpaired interbreeding.   These taxa should be detected objectively through the study of representative, uniformly distributed samples from throughout a species’ range and delineated by characters.  Samples MUST NOT be predefined geographically, with sampled areas only retrospectively discriminated statistically using correlations between traits (measures that can vary within a particular locality). 
For an example of how subspecific taxonomy ‘works’, in a ‘racial’ practical at University of Cape Town, several groups of undergrad students are each presented with a pile of 450 cards of unknown provenance illustrating, actual, continent-wide, local and geographical external anatomical and DNA variation in specimens of Helmeted Guineafowl Numida meleagris, Africa’s most widespread and geographically variable bird.  The students invariably sort them into nine major piles that confirm the existence of the geographically definable, widespread subspecies recognized in a 1978 monograph and illustrated in the attached figure.  This because these well-marked ‘races’ are delineated by a range of independent anatomical and molecular genetic features that are highly uniform within populations and vary geographically congruently between them.
But, regardless of these differences, where these ‘self-defining’ guineafowl taxa come into contact, because members of their populations have an identical avian ‘language’ and courtship behaviour and are genetically compatible post-mating, they interbreed freely without hybrid disadvantage. 
This results in another 20, much smaller ‘practical piles’ comprising ‘different’ individuals that are morpho-genetic ‘hybrids’ from 100-300 km-wide ‘suture’ zones where races come into contact and within which local within-population variation is high.  Many of these intergrades were, at one time or the other, erroneously ‘recognized’ as ‘waste-basket’ subspecies.

If one were to use the same, uniform sampling strategy for African humans (anatomically corrected for height), one would get two geographically delineated piles; golden-skinned individuals with women exhibiting steatopygia (KhoiSan from southwestern Africa), and a highly variable, widespread one comprising the rest. A microcosm of latter would include South Africa’s ‘Coloured People’, arguably the most morphologically and genetically complex and diverse ethnic group on Earth, with genetic ties to Khoisan (32–43%), Bantu African (20–36%), Northern European (21–28%), East Asian (9–11%), Indian and even Yemeni peoples.

Indeed, if one sorted the same specimens (indeed human specimens from throughout the world) by skeletal anatomy and/or DNA, there would be only one mega-pile, with all individuals outside of Africa being subsets of those found there.  Everything is just ‘mixed up’.  This is because our species is evolutionarily young and highly mobile and genetically admixed, especially during the last 3000 years.
Then ‘why races’?
So why do genetic racialists (like Edwards, Rosenberg and Dawkins), Ancestry.Com/DNA, ‘scientific’ racists and decolonists, in the face of this evidence, persist in supporting racially partitioning modern humans?  From a genomic perspective, this is because, multivariate statistical analyses of multilocus genetic information and anatomical measurements employing specific methods (e.g. cluster analysis) and selected algorithms (there are more than 100) can discover ‘overall’ differences between populations, and, in some instances, ‘correctly’ assign individuals to continental areas or putative human races. 
They’re wrong for two reasons.

First, their samples are often ‘cherry-picked’ to reflect core geographic areas of putative races or by continental limits.   Populations considered to be “admixed,” as well as individuals of “mixed ancestry,” are often excluded from sampling (e.g. as in Dawkins’ ‘photos’).  In their 2004 paper Evidence for Gradients of Human Genetic Diversity Within and Among Continents, David Serre and Svante Pääbo showed, contrary to Rosenberg et al., that when humans are sampled homogeneously and uniformly across the globe, the pattern seen is one of clinal (geographically gradual without sharp breaks) gradients of allele frequencies that extend over the entire world, rather than discrete clusters that reflect putative races or continental areas. In fact, the distance between two populations is by far the best predictor, explaining more than 80% of the variation of human genetic differentiation.

Rosenberg et al. 2005 conceded (at last in part) on this point: “In several populations [identified as meaningful “clusters” by their approach], individuals had partial membership in multiple clusters, with similar membership coefficients for most individuals. These populations might reflect continuous gradations across regions or admixture of neighboring groups.” “For population pairs from the same cluster, as geographic distance increases, genetic distance increases in a linear manner, consistent with a clinal population structure.” This means that there is clinal variation WITHI their ‘clusters’.  “Our evidence for clustering should not be taken as evidence of our support of any particular concept of ‘biological race’.”

However: “At the same time, we find that human genetic diversity consists not only of clines, but also of clusters that STRUCTURE (their cluster analysis program) observes (give certain assumptions) to be repeatable and robust.” since “for pairs from different clusters, genetic distance is generally larger than that between intracluster pairs that have the same geographic distance.”

So: “Loosely speaking, it is these small discontinuous jumps in genetic distance—across oceans, the Himalayas, and the Sahara—that provide the basis for the ability of STRUCTURE to identify clusters that correspond to geographic regions.” that may be able “to facilitate further research into such topics as human evolutionary history and the identification of medically important genotypes that vary in frequency across populations.”

In short, Edwards/Rosenberg ‘clusters’ are statistically significantly different (not evolutionarily cohesively similar) entities within which, for gene-genealogical purposes,  at least some of your ancestors existed at one time or other. They may also, in some instances, be useful proxy study groups in tackling problems related to improving medical health care. 

But, they have no utility in investigations relating to the evolutionary structuring of human genetic diversity, and NO ontological status that racially motivated people can employ in support of their ‘causes’ or ‘being’.

Second, the genetic differentiation, F_st, between putative human races (0.04-0.08) falls well below that found between the generally recognized non-human subspecies (0.15-0.20) and species (0.35).

What pro-race geneticists and biometricians don’t admit is that, pushed to the limit, their Procrustean approaches could also statistically ‘discriminate’ between many of the putative 20000 ethnic groups in the world, the hundreds said to occur in Africa and at least 14 in South Africa.  Their demographic ‘tools’ would have been ‘manna from heaven’ for Hendrik Verwoerd and his separate-development, social anthropologist mentor Werner Eiselen.  Of course, the ‘downside’ for these Broederbonders is that such ‘tests’ could place many of their cultural ‘kindred’ on the ‘wrong side’ of the “one-drop” black spectrum.  Inversely, not long after Afro-American Harvard Prof. Henry Louis Gates Jr. was arrested for ‘breaking into’ his own home by Irish-American police officer James Crowley, he announced on The Oprah Winfrey Show that one of these tests indicated that they shared a common Irish ancestor!

Where do pro-racialists draw the line?  What is the role of these genomic “Dasein” [a “primal nature of being”, a self-identity based on a “shared history and destiny”], especially when the isolation of such entities is ‘reinforced’ by linguistic, religious, other ethnic and political/ideological ‘differences’? What if these Dasein also correspond to hyper-taxonomized, polytypic entities within humanity recognized when employing a Phylogenetic Species Concept?

Rather than enriching our understanding of human evolution and human relations, these ‘races’, ‘clusters’ and ‘phylogenetic’ subspecies only encourage racists to ignore, misunderstand, hate and make war/genocide the ‘others’, and ‘cleanse’ themselves of the ‘odd-ones’ within?”

This sounds like racism and eugenics to me.


Robert Sobukwe got it right.  There is only one, crazy mixed up human race.