A recently published study, Race, Population Affinity, and Mortality Risk during the Second Plague Pandemic in Fourteenth-Century London, England states in the abstract:

We investigate whether hazards of death from plague and physiological stress at a fourteenth-century plague cemetery (Royal Mint, London) differed between populations using N = 49 adults whose affiliation was established using macromorphoscopic traits. Compared to a nonplague cemetery (N = 96), there was a greater proportion of people of estimated African affiliation in the plague burials.

[...] These findings may reflect premodern structural racism’s devastating effects.

The study appears to build on an earlier study by the same authors, “Officially absent but actually present”: Bioarchaeological evidence for population diversity in London during the Black Death, AD 1348–50 which gives the basis for the idea that were significant numbers of black people in medieval London.

The previous study explains that DNA evidence is essentially not recoverable in most cases, so they rely on craniometry:

The aDNA analysis [...] were only able to sequence the mtDNA from 5902. The haplotype of this female with Black African/Asian ancestry was U5b2a5, which is commonly found in England, Finland, and Tajikistan. Given their light stable isotope results, it is most likely that their maternal ancestry was from the British Isles. This result suggests that although these individuals are matching people with Black African/Asian ancestry in the macromorphoscopic database, in fact they may have had diverse heritage.

Earlier in the same study it is provided sample 5902 has strong macromorphoscopic Asian traits, while no craniometric traits could be assigned, while the DNA evidence suggests this is a person of British ancestry.

In the conclusion to that study they say

Our study ... provides circumstantial evidence for the presence of males and females with Black African ancestry and dual heritage. These data are primary evidence for these people being “actually present” in Medieval London. However, the data provided a very unexpected result as it designated several individuals as having full or dual Asian ancestry. We propose that these results are a false positive, foremost because comparable temporally and geographically appropriate craniometric and macromorphoscopic reference data are not available.

They continue by saying that people's skulls have grown narrower since the medieval period. It seems that they are saying their algorithm, which assigns skulls to races based on separate craniometric and macromorphoscopic data, may at least in respect of macromorphoscopic data be rather unreliable, since it's comparing unknown skulls, which they say could be black or white, but they don't think can be Asian, with their dataset of skulls from the years 1965 and onwards. Despite acknowledging changes to skull features in the last six hundred years they are still happy to assign medieval London skulls as African or white, even while saying the algorithm is likely incorrectly determining them to be Asian.

This rather draws into question the subsequent study by the same author which is using "affiliation was established using macromorphoscopic traits" to determine that supposed black women were at outsized risk of plague, given the earlier study seems to give cause to doubt the reliability of such traits for 600-year-old skulls compared to more modern ones.

When the study was provided to media outlets, the UK Equalities Minister, Kemi Badenoch, condemned it, describing it as "woke archaelogy".

Obviously regardless of whether the sample size is sufficient, if the study's methods are not accurately assigning race, the study's conclusions become worthless.

Can the ethnicity of medieval skulls be determined by shape?

  • 1
    I'm not sure this question is answerable as currently framed. The key phrase in the first quote is "estimated African affiliation". An estimate by definition is not known with exact certainty. If you want a general review of how reliable such estimates might be, this looks promising: ncbi.nlm.nih.gov/pmc/articles/PMC6474543. You may or may not also find more dissenting or supporting opinions about the particular paper you're asking about, but none of this really resolves the question.
    – Brian Z
    Commented Dec 14, 2023 at 2:41
  • @BrianZ that paper has the same co-author as the two above, although the first listed author for the two I quote is "Rebecca Redfern", who is absent in your link, as is all the rhetoric about "structural racism". There seems to be very substantial difference in tone aside as your link goes with measured "this is likely an African male" type statements, which aren't exciting, but when used to make sensational "medieval London was substantially African and they all died of plague because racism" clearly becomes so.
    – thelawnet
    Commented Dec 14, 2023 at 3:16
  • Why do you consider that the references provided in the paper - describing how this is done - are insufficient? Commented Dec 14, 2023 at 11:31
  • 3
    It does help notability, though the media outlets would have been enough. But it's not really relevant to the question; to go to the article, "I agree with my honourable friend that this type of research is damaging to trust, to social cohesion and even to trust in health services." makes it clear the author isn't speaking about scientific issues with the research.
    – prosfilaes
    Commented Dec 14, 2023 at 16:13
  • 1
    @JackAidley There are two separate questions: 1) how reliable are these methods for modern skulls, 2) is there any study which attempts to apply these technique for historic skulls, or discusses the validity of them when applied to historic skulls - i.e. if the method is reliable for modern skulls, does it follow that it should be reliable for historic skulls. If it does not follow then clearly the study is of no value, because it purports to show that historic black Londoners were more likely to die of the plague, when (it might be) that they were in fact white.
    – thelawnet
    Commented Dec 14, 2023 at 16:15

1 Answer 1


Can the ethnicity of medieval skulls be determined by shape?

What we can answer is that the ethnicity of modern skulls can be determined by shape, with a certain error rate.

There is a database of skull with associated ethnicity information, the Macromorphoscopic Databank (MaMD, see Hefner, 2018). In the study you mention the particular method was validated on 1537 individuals from MaMD and the categorisation results were:

Group African Asian European
African 153 83 69
Asian 45 544 90
European 22 85 446
MIN86 5 13 23

Classification matrix from a canonical analysis of the principal coordinates using macromorphoscopic data

Here the rows are the results recorded in the database (MIN86 is the medieval burial) and the columns are the results of the analysis. As can be seen there are certainly errors, but the method categorises most individuals correctly.

  • Note that the book chapter cited is mentioned in the question.
    – Oddthinking
    Commented Dec 14, 2023 at 13:09
  • @Oddthinking I had somehow not noticed that. I have edited the answer but it does make the whole question a little superfluous.
    – User65535
    Commented Dec 14, 2023 at 13:25
  • 1
    "there are certainly errors, but the method categorises most individuals correctly." Your answer shows that a population comprised entirely of individuals of European heritage will be classified as 5% African, and 15% Asian. In addition your final sentence "As can be seen there are certainly errors, but the method categorises most individuals correctly." doesn't really answer the question in that it's not clear why we should believe that medieval London was 44% non-European as per MIN86 in your table
    – thelawnet
    Commented Dec 14, 2023 at 14:59
  • 1
    nor is it clear why the 'method categorises most individuals correctly' when the question is about the ability of this method to identify specifically medieval skulls, whereas in my question I note that the study author acknowledges skull shapes have changed in 600 years, so we would really need a source that addresses the validity of this method to identify the ethnicity of non-modern skulls. Also the question's study addresses both 'Macromorphoscopic ' and 'craniometric' traits, whereas I think (?) you've only addressed Macromorphoscopic
    – thelawnet
    Commented Dec 14, 2023 at 15:04
  • 1
    @thelawnet, a goodness-of-fit test solidly rejects the hypothesis that medieval London's population consisted entirely of modern Europeans.
    – Mark
    Commented Dec 14, 2023 at 22:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .