A Plague Grave Could Only Be Dated to a Century. Now It Can Be Connected to an Outbreak
A medieval grave with plague DNA in it is easy to date and hard to place. Radiocarbon will usually say the person died somewhere inside a window of a century or more, which is precise enough for archaeology and close to useless for history. Across those same decades, the chroniclers of a single town might have written up several separate visitations of the disease, each with a year against it. The bones and the written record describe the same catastrophe and cannot be brought into the same sentence.
Closing that gap is the point of a study published Sept. 10 in the Proceedings of the National Academy of Sciences, led by Marcel Keller of the Institute of Genomics at the University of Tartu. Ancient DNA has by now produced over 100 genomes of Yersinia pestis, the bacterium that causes plague, from the second pandemic, the long run of European outbreaks that opens with the Black Death and continues for four more centuries. For most of those genomes, the authors write, a radiocarbon date that regularly spans more than 100 years is the only chronological information there is, and where the outbreaks started and how they spread remains poorly understood.
Keller framed the problem against a recent pandemic. "With COVID-19, scientists could reconstruct the spread of individual strains extremely well because the genomes came with precise timestamps. For historical pandemics, those timestamps are often missing or may cover more than 100 years, which limits our ability to interpret the genetic data," he said in a statement issued by the University of Stirling.
Their proposed fix, which they call Phylogenetically Informed Radiocarbon Modeling, uses something a family tree knows that a carbon measurement does not: order. A strain that descends from another has to come after it, so once the genomes are placed on a tree, each one's possible dates are hemmed in by its relatives as well as by its own radiocarbon range. Folding the two together, the team reports, gives dating intervals that are both narrower and more accurate. It is a proposal, not a settled standard, and it inherits an open question: how fast the plague bacterium's genome accumulates mutations is still contested in the field.
Alongside the method comes new material: 26 genomes from 11 sites in Europe, 11 of them read thoroughly enough to count as complete and 15 of them patchier, from burials dated between 1349 and 1710. The university's release places those sites in Estonia, Russia, England, the Netherlands and Switzerland.
With those intervals in hand, and a fine-grained pass through the outbreaks recorded at the time, the team tentatively associates 75 second-pandemic genomes with epidemics that historians had already named. The hedge is theirs, and it carries weight: each match is a proposal, open to a historian or another geneticist who reads the evidence differently.
"We were able to improve dating intervals for many samples, which allowed us to connect them to specific plague waves and outbreaks that were recorded in the respective towns or regions by chroniclers," said Philip Slavin, a historian in the Division of History, Heritage and Politics at the University of Stirling and one of the paper's authors.
The Stirling release illustrates the kind of connection at stake with burials from Domat/Ems in Switzerland, which it ties to the plague that came with the Thirty Years' War. It also singles out one finding above the rest: evidence of a major expansion of plague lineages, the branches of the bacterium's family tree, around 1450 to 1500, long after the Black Death itself had passed.
What that changes is the question a plague burial can be asked. The old answer stopped at a century. A genome placed inside a named outbreak arrives carrying whatever the written record knows about that outbreak: the season it struck, the road or the ship it is thought to have come in on, how long it lasted, how many it killed. Chroniclers wrote down where plague came from and who it took; genomes carry how the bacterium itself was moving. Since the first plague genome was sequenced, the two accounts have run in parallel, often describing the same disasters without being able to point to the same one. The dates are what would let them meet, and this paper is an argument that the dates can be sharpened.