Thoughts on race

A concept too thin to be worth the fuss.

Regino of Prum (d.915) offered a definition of nation, riffing off Classical discussions of the same, that became the standard fourfold definition during the medieval period:

Diversae nationes popularum interse discrepant genere moribus lingua legibus.

(The peoples of various nations differ by origin, customs, languages and laws.)

Origin (genere) is a matter of lineages, but the rest are cultural phenomena. This fits in with Homo sapiens being the most cultural of species. In the words of biologists Heather Heying and Bret Weinstein:

Humans are not blank slates, but of all organisms on Earth, we are the blankest (p.146).

One of the notorious problems with the concept of race is how its usage has changed over time.

Source

The medieval European concept was basically “people who speak the same language” (cuius linguae gens, the race of that language). John of Fordun, the first chronicler of Scotland, wrote:

The manners of the Scots vary according to their language, for they employ two languages, Scottish [Gaelic] and Teutonic [Scots/English]. The race of Teutonic language has the sea coasts and lowlands, that of Scottish language inhabits the mountainous areas and the outer isles. The race of the sea coasts is domesticated, civilized, faithful, patient, cultivated, decently dressed, refined and peaceable, devout in church worship, yet always ready to withstand any harm done by its enemies. The island or mountain race, however, is wild, untamed, primitive, intractable, inclined to plunder, leisure-loving, quick to learn, skilful, handsome in appearance but vilely dressed, and continually fiercely opposed to the English people and language, but also to their own nation, on account of the difference of language. Nevertheless they are loyal and obedient to the king and the kingdom, and also easily subdued to the laws, if they are ruled properly.

Thus, Scotland could be a nation of two races. Over time, race shifted from ethnicity, to groups of ethnicities, to recognisable patterns in physical appearance on a continental scale. As a study of global genetic variation noted:

We also found that in an unsupervised cluster analysis, individuals grouped into geographical clusters largely corresponding to sub-Saharan Africa, Europe and the part of Asia west of the Himalayas, the part of Asia east of the Himalayas, Oceania, and the Americas. These observations are compatible with serial sampling, assuming that major geographic barriers such as oceans, the Sahara desert, and the Himalayas were not frequently crossed during human migrations.

Given differences in selection pressures and ancestral bottlenecks — despite considerable admixtures over time — we are generally quite good at picking people’s continental or sub-continental origin simply by physical markers of appearance. Such is an assessment of lineage origin: including, of course, mixed origin.

Biology is a science with various fuzzy boundary sets. Taxonomy is a matter of assembling lineages in appropriate sets (taxon). These taxa go from the most general to the most specific. In the words of Wikipedia:

domain, kingdom, phylum (division is sometimes used in botany in place of phylum), class, order, family, genus, and species.

The three domain system divides cellular organisms into two domains of those with no nucleus (prokaryotes) and one of those with membrane-bordered nucleus (eukaryotes).

Then we have kingdoms of animals, plants, fungi …. Followed by phyla based either on genetic relation or body plan. Some biologists argue phylum should be replaced with clade: i.e. common ancestor.

All these levels of taxonomy are still subject to lively debate as to whether various classifications are structurally robust — so stable — or not. These arguments are focused on structure and ancestry.

Arguing for race as a natural biological kind, yet not to be understood taxonomically, is somewhat contradictory. If one is not distinguishing sets of lineages — which is what a race is — by structure and ancestry, what is the analytical point of the concept?

If one argues for the existence of sub-species/races in other species as an indicator of the reasonableness of the concept of race for humans — as Ernst Mayr does, for instance — then one is arguing for a taxonomic concept. If one is arguing that race is somehow uniquely human, then we are in “special creation” territory.

Where humans are distinctive is in being so markedly the cultural species. A dynamic that tends to reduce, rather than increase, the analytical salience of geographic, continental-scale, race.

Your lineage extends back through many species that no longer exist and back through all the above levels of taxa. The taxonomic question is: when do structural differences between sets of lineages become sufficiently large to be robustly distinguishable?

To note that variation within human groups is greater than that between groups is to accept that we are a single species. It does not, in itself, disqualify the existence of races. The question is: how distinctive is the variation between groups?  Given that we can pick ancestry as well as we do, not as much as one might expect:

When we divide the populations into seven geographical regions and estimate three variance components, the within-population component is 93.9%, the among-population-within region component is 2.4%, and the among-region component is 3.8%.

Remembering that those genetic variations include the genes for the physical markers of ancestry.

The degree to which there is overlap between human groups matters for taxonomy, for analysis and for expectations. Many of those who are happy to parrot the point about genetic variation within human groups being much greater than between them — as some point against the existence of races — are also happy to take the opposite view of variance-within being greater than variance-between when making moral and analytical mountains out of molehills of differences in outcomes between groups. Suddenly, they are happy to ignore the typically much larger level of overlap in such outcomes.

Variation in average IQ among human populations is often cited in race-realist literature. IQ is highly heritable and correlates strongly with a lot of social outcomes, so it matters, including for variation in social outcomes among human groups. It is also affected by factors such as infant and child nutrition, parasite load and stimulation. Do the genes affecting human cognition show distinctive variation between groups? If they do, how much? What is the population structure of those differences?

In their 1953 paper, E. O. Wilson and W. L. Brown discuss the concept of subspecies or race for general biological taxonomy and find it to be more misleading than useful. One of the problems being that focusing on some obvious pattern of variation often obscured whether that accorded with other patterns of variation within that species.

Human groups interbreed easily and people can shift between groups. One of the ways that human groups compete is by seeking to attract new members and to hold onto the ones they have. Hence, a limited form of multi-level selection that includes selection for being groupish — and mobilising our groupishness — makes more sense than does group selection.¹ Notice that such competition is going to generally be way more intense across time and space within — rather than between — continental-scale races.

The capacity for religion, and religion analogues, to induce us to being devoted agents — rather than instrumentally rational ones — so mobilising our groupishness, has selection advantages for lineages precisely because we are an ultra-social (i.e. groupish) species. We are a species that evolved highly cooperative subsistence and reproduction strategies able to support raising biologically expensive children.

Hence human societies are structured around transferring risks from child-rearing and resources to child-rearing. It is a pattern that extends beyond kin among foragers.² A fundamental pattern that is why human societies have typically put considerable effort into creating and maintaining male teams, but within social units much smaller than continental-scale race.³

The fuzzy porousness of human groups is precisely why attempts to generate legal definitions of race — to legally separate people into racial groups — have proved to be so fraught. While being a fuzzy-boundary set is not in itself disabling — consider the electromagnetic spectrum and colour — for race to be a biologically useful concept, there needs to be some robust structural feature or features that race identifies. It is perfectly possible for sets of lineages reasonably identifiable as races to exist and yet for it to analytically not matter very much.

The contemporary argument about the usage of race for sets of humans comes down to whether it picks up biological structures resulting from lineage that are robustly variant. For if we want to include any cultural features, then a continental or sub-continental scale origin notion of race will not do, as such regions are collections of cultures and folk can, and do, shift between cultures. Given the relative clarity of ethnicity, and language, the concept of race is either biological or it is pointless.

The obvious argument for race is that we can identify continental or sub-continental origin with quite a high degree of accuracy. The strongest argument against race is that the genetic structure of humans is a continuum, without the “troughs” of difference that make sub-species a useful concept among other organisms. The obvious physical markings accord poorly with other variations in human populations. In large part because we are such a highly cultural species.

A response is that colour is also a spectrum, but we still find it useful to divide the colour spectrum into different colours. Against that, the electro-magnetic spectrum is a single dimension along which we can get colours to do different things for us. Human differentiation is much more complex (and allocating usages by race is not optimal).

If you want to explain different distributions of outcomes among human groups, race is not very useful, precisely because we are so much the cultural species. Some phenomena are overwhelmingly cultural, while others are a result of gene-culture interactions, and culture does not match race, unless one wants to incorporate ethnicity with race. Given that ethnicity is a perfectly good term for the conjunction of breeding population and culture, using race for ethnicity is an analytically retrograde step, given that race also applies to continental-scale differences.

Even within continental or sub-continental regions, the selection pressures among human groups could and did vary significantly. A more promising differentiation at such a level is how much the selection pressures came from the wider environment and how much from other humans, though such a variation only weakly correlates with the normal conceptions of continental-level race.

Source: Karmin et al. Intense selection pressure from other humans, much stronger outside Africa..

In his mega-paper on the subject, John Fuerst refers to race as:

…ancestrally informative correlated variation.

And as:

…the primary sense, in which biological races represent cluster class discrete sets and a secondary sense in which biological races represent cluster class discrete-plus-fuzzy sets.

He accepts that not everyone belongs to a specific race, just as not every member of a species belongs to a sub-species. He wants to include ethnicity as a level of race, which strikes me as trying to buff up race rather than improving understanding. The value of ethnicity as a concept is precisely because it includes the cultural, identifying a (fuzzy) conjunction of culture and breeding population.

Once you admit ethnicity as a useful level of analysis, and it is both biological and cultural, it is hard to see what benefit the continental-scale differentiation we are left with has in analytical worth. Even very general expectations of tendency are likely to be dominated by local cultural context. So one is left with very thin analytical gruel.⁴ Race is too poor an indicator — one is much better off going to smaller groups, notably ethnicity.

There is something that race picks up — that’s why we can pick continental-level ancestry quite well — but little that is robustly variant enough to be analytically useful as the genetic differentiation is not very high, is so much a continuum and we are so much the cultural species.

Hereditarianism is true, and blank slate conceptions are false, because the social is emergent from the biological. We do not need to buff up race to argue that case.

References

Books

Bryan Hayden, The Power of Ritual in Prehistory: Secret Societies and the Origins of Social Complexity, Cambridge University Press,  [2018] 2020.

Heather Heying and Bret Weinstein, A Hunter-Gatherer’s Guide to the 21st Century: Evolution and the Challenges of Modern Life, Swift, 2021.

James C. Scott, The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia, Yale University Press, 2009.

Articles, etc

Robert Bartlett, ‘Medieval and Modern Concepts of Race and Ethnicity,’ Journal of Medieval and Early Modern Studies, 31 (2001): 39 - 56.

John Fuerst, ‘The Nature of Race: the Genealogy of the Concept and the Biological Construct's Contemporaneous Utility,’ Open Behavioral Genetics, June 20, 2015.

Jeremy Ginges, Scott Atran, Douglas Medin, and Khalil Shikaki, ‘Sacred bounds on rational resolution of violent political conflict,’ PNAS, May 1, 2007, vol. 104, no. 18, 7357–7360.

Herbert Gintis, Carel van Schaik, and Christopher Boehm, ‘Zoon Politikon: The Evolutionary Origins of Human Political Systems,’ Current Anthropology, Volume 56, Number 3, June 2015, 327-353.

G. Hofstede, ‘Dimensionalizing Cultures: The Hofstede Model in Context,’ Online Readings in Psychology and Culture, (2011), 2(1).

Monika Karmin, et al., ‘A recent bottleneck of Y chromosome diversity coincides with a global change in culture,’ Genome Resources, 2015 Apr;25(4):459-66.

Thomas McEvilley, ‘An Archaeology of Yoga,’ Anthropology and Aesthetics, 1981 1:, 44-77.

Ernst Mayr, ‘The Biology of Race and the Concept of Equality,’ Daedalus, Vol. 131, No. 1, On Inequality (Winter, 2002), 89-94.

Joseph K. Pickrell and David Reich, ‘Toward a new history and geography of human genes informed by ancient DNA,’ Trends Genet, 2014 September ; 30(9): 377–389.

Noah A. Rosenberg, ‘A Population-Genetic Perspective on the Similarities and Differences among Worldwide Human Populations,’ Human Biology, 2011 December; 83(6): 659–684.

Hammad Sheikh, Jeremy Ginges, Alin Coman, Scott Atran, ‘Religion, group threat and sacred values,’ Judgement and Decision Making, (2012) 7(2), 110-118.

Hammad Sheikh, Jeremy Ginges, and Scott Atran, ‘Sacred values in the Israeli–Palestinian conflict: resistance to social influence, temporal discounting, and exit strategies,’ Annals of the New York Academy of Sciences, September 2013, 1299, 11–24.

Hobart M. Smith, David Chiszar and Richard R. Montanucci, ‘Subspecies and Classification,’ Herpetological Review, 28(1), 1997, 13-16.

Jordan E. Theriault, Liane Young, Lisa Feldman Barrett, ‘The sense of should: A biologically-based framework for modeling social pressure’, Physics of Life Reviews, Volume 36, March 2021, 100-136.

Michael Tomasello, ‘The ultra-social animal,’ European Journal of Social Psychology, (2014) 44, 187–194.

E. O. Wilson, W. L. Brown, ‘The Subspecies Concept and Its Taxonomic Application,’ Systematic Biology, Volume 2, Issue 3, September 1953, 97–111.

Richard W. Wrangham, ‘Two types of aggression in human evolution,’ PNAS January 9, 2018, Vol.115, No.2, 245–253.

1. Cancer cells are cells that defect from cooperation and operate at an earlier evolutionary stage, including relying on glucose but not ketones for energy.

Folk with Cluster B personality disorders, sociopathy and psychopathy, defect from normal human social cooperation, and operate at an earlier, primate, level of non-empathetic, non-internalised norms but with the human capacity for learning and imitation.

2. Marx locating the human Fall into alienation and oppression as arising from the division of labour outside kin groups is utter nonsense. Division of labour by sex is a feature of all human societies until very recently, with foraging societies not limiting transfers of resources and risks to among kin. That became a dominant pattern only in the most intensely kin-group dominated agro-pastoral societies. Moreover, the oldest profession is not prostitution, it is being a shaman and its priestly derivatives. A highly self-conscious species developed specialists in rituals, shared narratives, and emotional management and catharsis (“meaning”). So even division of labour within or across sexes predates farming.

3. It is highly likely that what made us Homo sapiens were teams of unrelated beta males systematically killing off the alpha males. Male teams were how we murdered our way to niceness.

4. A lot of race realism is dominated by the US experience, or inferences therefrom. Running together descendants of American slaves, Afro-Caribbeans and recent African immigrants generally degrades, rather than enhances, analysis. Similarly, Afro-Caribbeans and recent African immigrants in the UK have quite different social patterns and outcomes.