The previous essay argued that the 1,000 Genomes Project and modern population genomics help dissolve the biological idea of race. That is broadly true — most human genetic variation is shared — but the story is more complicated. A number of solid, peer-reviewed studies identify loci where allele frequencies differ sharply between regions, sometimes to the point of near-fixation. Those differences matter for phenotype and health. Taken out of context, they can be (and historically have been) used to justify Morton-style claims of innate, hierarchical human difference.
Below I collect a handful of well-cited studies, include short quotes from the papers themselves, and then draw cautious, evidence-based inferences — and limits.
1) Population structure exists — but it’s a matter of degrees
“Within-population differences among individuals account for 93 to 95% of genetic variation; differences among major groups constitute only 3 to 5%.” Science
Rosenberg et al. (2002) — a landmark study using hundreds of microsatellite markers across global samples — showed that most genetic variation is within populations, not between them. Still, the remaining 3–5% is enough for multilocus methods to cluster individuals by continent of origin. This is the factual basis for later claims that genetics can “recreate” population groups.
Why Morton-style readers seize on this: small but consistent allele-frequency differences across many loci let statistical methods assign an individual to a geographic cluster with high accuracy. That looks, superficially, like “biological races.”
Why that inference is weak: clustering arises from correlated allele-frequencies produced by population history (migration, drift, bottlenecks, isolation), not from discrete, immutable biological types.
2) Multilocus classification works — Edwards’s critique of Lewontin
A.W.F. Edwards (2003) summarized the statistical point plainly: although most variation is within populations, “the correlation structure can be used to classify individuals into populations.” PubMed
Edwards’ point (often called “Lewontin’s fallacy”) is technical and important: looking locus-by-locus misses the information contained in correlations among loci. Put bluntly, many weak differences across many loci combined give strong predictive power.
What this enables: accurate ancestry inference and forensic/genetic clustering.
What this does not prove: any hierarchy of worth, intelligence, or moral capacity — those are extra-scientific claims.
3) Single loci with large effect: pigmentation, lactase persistence, EDAR
Genetics gives us clear examples where a single allele explains a big, visible phenotype and has very different frequencies across regions.
-
Skin pigmentation — SLC24A5: As Lamason et al. (and follow-ups) show, a single nonsynonymous SNP in SLC24A5 accounts for a large fraction of the European vs. West African skin-pigmentation difference. “A single nucleotide difference … accounts for 25–38% [of] European–African pigmentation differences.” ResearchGate+1
-
Lactase persistence — LCT regulatory variants: Enattah et al. (2002) identified C/T-13910 upstream of LCT, which “completely associates” with lactase persistence in some European samples. This allele has high frequency in pastoralist populations and low frequency elsewhere. PubMed
-
EDAR V370A — East Asian trait: The derived EDAR allele V370A shows a very high frequency in East Asians and affects hair thickness, tooth morphology, and sweat glands; it’s described as “one of the strongest candidates of recent positive selection.” PMC
Mortonish temptation: these are clear, functionally meaningful genetic differences that map onto continental regions — to a Morton-style reader, that looks like the kind of “fixed” biological difference he used to assert.
Reality check: these are specific, local adaptations tied to environment, diet, or demographic history. They do not imply global hierarchies of ability or worth.
4) Near-fixations and sweeps: Duffy and malaria adaptations
Some alleles have essentially swept to fixation in particular regions because of strong local selection:
-
The Duffy (FY*O) null allele rose to (near-)fixation in sub-Saharan Africa because it confers resistance to certain malaria parasites. Modern analyses reconstruct a strong selective sweep at this locus. PMC
-
The HBB sickle-cell allele is geographically restricted and confers malaria resistance in heterozygotes — the classic example of balancing selection described by Allison. PubMed
Inference: local, high-impact selection can produce alleles at (or near) fixation in certain regions — so yes, at the level of single loci, you can get very strong, geographically stratified genetic differences.
5) Putting it together: what the facts allow — and what they forbid
Facts
-
Human populations differ in allele frequencies; multilocus patterns permit accurate geographic ancestry inference. Science+1
-
Some single variants (e.g., SLC24A5, LCT regulatory variants, EDAR V370A, Duffy) show large frequency differences and clear phenotypic or physiological effects. PMC+3ResearchGate+3PubMed+3
What these facts do justify
-
Scientific claims about population history, migration, and adaptation. (Why did EDAR rise in frequency? Why did lactase persistence evolve in pastoralists?) PMC+1
-
Medical and public-health applications that use ancestry or variant frequency to predict risk or tailor treatment (pharmacogenomics, sickle-cell screening, G6PD deficiency considerations, etc.). Science
What these facts do not justify
-
Any claim that allelic differences imply ranked moral, intellectual, or social worth. Genetics gives mechanistic explanations for phenotype, not blueprints for social hierarchy. Edwards’s statistical point about classification is not a biological justification for racism. PubMed
-
The notion of discrete, bounded “races” with immutable essences — population structure is clinal, continuous, and shaped by history. Science
6) The Mortonish danger: cherry-picking loci, reifying clusters
Morton’s error was not simply bad data; it was interpretive misuse: he treated a small set of measurements as if they proved a global, hierarchical natural order. Modern Mortonism would look the same: cherry-pick a few high-difference loci (skin pigmentation, EDAR, lactase, Duffy), and generalize from those to claims about intelligence, culture, or worth.
Two technical facts fuel that misuse:
-
Loci differ in effect size. A few loci explain big phenotypic differences (e.g., pigmentation, lactase). Generalizing those to global traits is unsound. ResearchGate+1
-
Multilocus clustering is predictive but not normative. You can predict ancestry; you cannot infer superiority from that prediction alone. Science+1
7) Final caution: science’s power and limits
Yes — genomics gives us concrete examples of geographically stratified, functionally meaningful genetic differences. Those facts complicate a simplistic claim that “race is only a social construct” in the sense that some traits do track ancestry strongly. But acknowledging those facts does not bring us back to 19th-century biological hierarchy. Instead it imposes moral and scientific responsibilities:
-
Do not conflate population-specific adaptation with global hierarchies of worth.
-
Do use allele-frequency knowledge to improve medicine and understand human history.
-
Do make social policy based on ethics, not on simplistic biological readings.
In short: the reality is not all hunky-dory — genetic structure and strong local adaptations exist and matter — but neither are these facts any justification for Morton-style racial hierarchies. The scientific task is to understand the mechanisms (selection, drift, migration), and the ethical task is to refuse any reduction of human dignity to allele counts.
Key sources quoted or used above
-
Rosenberg NA et al., Genetic structure of human populations (2002). Science
-
Edwards AWF, Human genetic diversity: Lewontin's fallacy (2003). PubMed
-
Lamason RL et al./Mallick et al., work on SLC24A5 and pigmentation. ResearchGate+1
-
Kamberov YG et al., EDAR V370A selection and phenotypes (2013). PMC
-
Enattah NS et al., LCT regulatory variant and lactase persistence (2002). PubMed
-
McManus et al./others on Duffy (FY*O) sweep in Africa. PMC
-
Allison AC, classic work on sickle-cell and malaria. PubMed
No comments:
Post a Comment