Infectious disease modeling study casts doubt on impact of Justinianic plague

Infectious disease modeling study casts doubt on impact of Justinianic plague

Work shows value of new examinations of old narratives of this pandemic

Justinianic Plague mathematical modeling
Costumes of All Nations (1882), by Albert Kretschmer, painters and costumer to the Royal Court Theatre, Berin, and Dr. Carl Rohrbach. Picture in the public domain

ANNAPOLIS, Md. - Many have claimed the Justinianic Plague (c. 541-750 CE) killed half of the population of Roman Empire. Now, historical research and mathematical modeling challenge the death rate and severity of this first plague pandemic.

Researchers Lauren White, PhD and Lee Mordechai, PhD, of the University of Maryland's National Socio-Environmental Synthesis Center (SESYNC), examined the impacts of the Justinianic Plague with mathematical modeling. Using modern plague research as their basis, the two developed novel mathematical models to re-examine primary sources from the time of the Justinianic Plague outbreak. From the modeling, they found that it was unlikely that any transmission route of the plague would have had both the mortality rate and duration described in the primary sources. Their findings appear in a paper titled "Modeling the Justinianic Plague: Comparing hypothesized transmission routes" in PLOS ONE.

"This is the first time, to our knowledge, that a robust mathematical modeling approach has been used to investigate the Justinianic Plague," said lead author Lauren White, PhD, a quantitative disease ecologist and postdoctoral fellow at SESYNC. "Given that there is very little quantitative information in the primary sources for the Justinianic Plague, this was an exciting opportunity to think creatively about how we could combine present-day knowledge of plague's etiology with descriptions from the historical texts."

White and Mordechai focused their efforts on the city of Constantinople, capital of the Roman Empire, which had a comparatively well-described outbreak in 542 CE. Some primary sources claim plague killed up to 300,000 people in the city, which had a population of some 500,000 people at the time. Other sources suggest the plague killed half the empire's population. Until recently, many scholars accepted this image of mass death. By comparing bubonic, pneumonic, and combined transmission routes, the authors showed that no single transmission route precisely mimicked the outbreak dynamics described in these primary sources.

Existing literature often assumes that the Justinianic Plague affected all areas of the Mediterranean in the same way. The new findings from this paper suggest that given the variation in ecological and social patterns across the region (e.g., climate, population density), it is unlikely that a plague outbreak would have impacted all corners of the diverse empire equally.

Xenopsylla cheopis, photo by Katja ZSM, CC BY-SA 3.0

"Our results strongly suggest that the effects of the Justinianic Plague varied considerably between different urban areas in late antiquity," said co-author Lee Mordechai, an environmental historian and a postdoctoral fellow at SESYNC when he wrote the paper. He is now a senior lecturer at the Hebrew University of Jerusalem, and co-lead of Princeton's Climate Change and History Research Initiative (CCHRI). He said, "This paper is part of a series of publications in recent years that casts doubt on the traditional interpretation of plague using new methodologies. It's an exciting time to do this kind of interdisciplinary research!"

Using an approach called global sensitivity analysis, White and Mordechai were able to explore the importance of any given model parameter in dictating simulated disease outcomes. They found that several understudied parameters are also very important in determining model results. White explained, "One example was the transmission rate from fleas to humans. Although the analysis described this as an important parameter, there hasn't been enough research to validate a plausible range for that parameter."

These high importance variables with minimal information also point to future directions for empirical data collection. "Working with mathematical models of disease was an insightful process for me as a historian," reflected Mordechai. "It allowed us to examine traditional historical arguments with a powerful new lens."

Together, with other recent work from Mordechai, this study is another call to examine the primary sources and narratives surrounding the Justinianic Plague more critically.


White, L.A. & Mordechai, L. (2020). Modeling the Justinianic Plague: Comparing hypothesized transmission routes. PLOS ONE. doi: 10.1371/journal.pone.0231256

About SESYNC: The University of Maryland's National Socio-Environmental Synthesis Center (SESYNC) in Annapolis brings together the science of the natural world with the science of human behavior and decision making to find solutions to complex environmental problems. SESYNC is funded by an award to the University of Maryland from the National Science Foundation. For more information on SESYNC and its activities, please visit


Press release from the SESYNC, University of Mariland.

Details of first historically recorded plague pandemic revealed by ancient genomes

Details of first historically recorded plague pandemic revealed by ancient genomes

Analysis of 8 new plague genomes from the first plague pandemic reveals previously unknown levels of plague diversity, and provides the first genetic evidence of the Justinianic Plague in the British Isles

Justinianic Plague Yersinia pestis
Map and phylogenetic tree showing the newly published (yellow) and previously published (turquoise) genomes. Shaded areas and dots represent historically recorded outbreaks of the First Pandemic. Credit: Marcel Keller

An international team of researchers has analyzed human remains from 21 archaeological sites to learn more about the impact and evolution of the plague-causing bacterium Yersinia pestis during the first plague pandemic (541-750 AD). In a study published in PNAS, the researchers reconstructed 8 plague genomes from Britain, Germany, France and Spain and uncovered a previously unknown level of diversity in Y. pestis strains. Additionally, they found the first direct genetic evidence of the Justinianic Plague in the British Isles.

The Justinianic Plague began in 541 in the Eastern Roman Empire, ruled at the time by the Emperor Justinian I, and recurrent outbreaks ravaged Europe and the Mediterranean basin for approximately 200 years. Contemporaneous records describe the extent of the pandemic, estimated to have wiped out up to 25% of the population of the Roman world at the time. Recent genetic studies revealed that the bacterium Yersinia pestis was the cause of the disease, but how it had spread and how the strains that appeared over the course of the pandemic were related to each other was previously unknown.

In the current study, an international team of researchers led by the Max Planck Institute for the Science of Human History analyzed human remains from 21 sites with multiple burials in Austria, Britain, Germany, France and Spain. They were able to reconstruct 8 new Y. pestis genomes, allowing them to compare these strains to previously published ancient and modern genomes. Additionally, the team found the earliest genetic evidence of plague in Britain, from the Anglo-Saxon site of Edix Hill. By using a combination of archaeological dating and the position of this strain of Y. pestis in its evolutionary tree, the researchers concluded that the genome is likely related to an ambiguously described pestilence in the British Isles in 544 AD.

High diversity of Y. pestis strains during the First Pandemic

The researchers found that there was a previously unknown diversity of strains of Y. pestis circulating in Europe between the 6th and 8th centuries AD. The 8 new genomes came from Britain, France, Germany and Spain. "The retrieval of genomes that span a wide geographic and temporal scope gives us the opportunity to assess Y. pestis' microdiversity present in Europe during the First Pandemic," explains co-first author Marcel Keller, PhD student at the Max Planck Institute for the Science of Human History, now working at the University of Tartu. The newly discovered genomes revealed that there were multiple, closely related strains of Y. pestis circulating during the 200 years of the First Pandemic, some possibly at the same times and in the same regions.

Despite the greatly increased number of genomes now available, the researchers were not able to clarify the onset of the Justinianic Plague. "The lineage likely emerged in Central Asia several hundred years before the First Pandemic, but we interpret the current data as insufficient to resolve the origin of the Justinianic Plague as a human epidemic, before it was first reported in Egypt in 541 AD. However, the fact that all genomes belong to the same lineage is indicative of a persistence of plague in Europe or the Mediterranean basin over this time period, instead of multiple reintroductions."

Sampling of a tooth from a suspected plague burial. Credit: Evelyn Guevara

Possible evidence of convergent evolution in strains from two independent historical pandemics

Another interesting finding of the study was that plague genomes appearing towards the end of the First Pandemic showed a big deletion in their genetic code that included two virulence factors. Plague genomes from the late stages of the Second Pandemic some 800-1000 years later show a similar deletion covering the same region of the genomes. "This is a possible example of convergent evolution, meaning that these Y. pestis strains independently evolved similar characteristics. Such changes may reflect an adaptation to a distinct ecological niche in Western Eurasia where the plague was circulating during both pandemics," explains co-first author Maria Spyrou of the Max Planck Institute for the Science of Human History.

The current study offers new insights into the first historically documented plague pandemic, and provides additional clues alongside historical, archaeological, and palaeoepidemiological evidence, helping to answer outstanding questions. "This study shows the potential of palaeogenomic research for understanding historical and modern pandemics by comparing genomes across millennia," explains senior author Johannes Krause of the Max Planck Institute for the Science of Human History. "With more extensive sampling of possible plague burials, we hope to contribute to the understanding of Y. pestis' microevolution and its impact on humans during the course of past and present pandemics."

Lunel-Viel (Languedoc-Southern France). Victim of the plague thrown into a demolition trench of a Gallo-Roman house; end of the 6th-early 7th century. Credit: 1990; CNRS - Claude Raynaud

Press release from the Max Planck Institute for the Science of Human History / Max-Planck-Instituts für Menschheitsgeschichte