Did Nature Bring Down the Western Roman Empire? — What Lead in Ice and Bacteria in Bones Can and Cannot Tell Us

In 476, When the Western Half Ended

The Roman Empire was an ancient state that united the lands surrounding the Mediterranean Sea, and from the late fourth century in 395, it was ruled, split into eastern and western halves. The western half lost land throughout the fifth century, plagued by various groups pushing across its borders and internal civil wars, and in 410, the Visigoths, a Germanic group, sacked the city of Rome. In 476, the Germanic general Odoacer forced Romulus Augustulus, the last emperor of the Western Roman Empire, to step down, and textbooks record this year as the end of the Western Roman Empire. This was roughly 1,500 years ago. The eastern half, the Eastern Roman Empire with its capital in Constantinople (today's Istanbul), continued for nearly another thousand years after that. In the mid-sixth century, the Eastern Roman Emperor Justinian I even managed to win back some of the old lands in Italy and North Africa.

Why did the West collapse? Historians have long answered this question through the lenses of politics, military affairs, and the economy. A one-sentence answer circulated more widely among the general public: the ruling class became corrupt, groups from outside the borders pushed in, and the lead in water pipes and vessels slowly poisoned the Roman people.

Over the past few decades, evidence from natural sciences has entered this question. It is lead trapped in Greenland ice, ancient weather left in trees and mud at the bottom of the sea, and genes of bacteria rescued from bones inside tombs. American historian Kyle Harper brought climate and plagues into the main cast of the Roman story in his book The Fate of Rome, and other researchers published a three-part rebuttal (DOI: 10.1111/hic3.12507).

This article separates what these three pieces of evidence added to and failed to add to the "fall of Rome" story. The plagues that appear are three, and the times they occurred are far apart from one another. The Antonine Plague struck the empire around 165 in the late second century, and the Plague of Cyprian in the mid-third century. The Plague of Justinian was an event that began in the Eastern Roman Empire in 541, which is 65 years after the end of the Western Roman Empire in 476.

Lead Left in Ice — A Record of the Years When Mines Were Busy

If you measure the layer of Greenland ice corresponding to the period from 2,500 to 1,700 years ago, from 500 BCE to 300 CE, it contains roughly four times as much lead as it does in its natural state. If you add up all the lead that settled onto the Greenland ice sheet during those eight centuries, it reaches 15% of the amount deposited by leaded gasoline since the 1930s. This means that the lead and silver mining and smelting of Greece and Rome already polluted the Northern Hemisphere's skies on a hemispheric scale 2,000 years ago (PMID 17797222).

An ice core is an ice cylinder pulled out by driving a long tube into a place like Greenland where snow does not melt and piles up year after year to become ice, trapping the dust and chemicals that were floating in the air of that year in each layer. The path that lead took to that distant north is as follows. In ancient times, silver was mainly obtained by melting ores containing both lead and silver, and the lead that flew into the air at that time rode the wind and settled on the Arctic ice (PMID 29760088). Years with a lot of lead in the ice were years when mines and smelting furnaces were busy.

In 2018, the period from 1100 BCE to 800 CE was measured more finely than by single years, and even the paths by which the air carried the lead were calculated with models to estimate Europe's lead emissions year by year. Emissions began to rise when the Phoenicians expanded their influence into the Mediterranean, accelerated when Carthage and Rome expanded their mines mainly in the Iberian Peninsula, and were highest under the Roman Empire. During the Republic, they fluctuated in tune with wars and political turmoil, dropped sharply around the same time as two major plagues in the second and third centuries, and stayed low for more than 500 years. The amount of silver contained in silver coins also decreased in the same pattern. The authors conclude that this record shows the economy grew steadily during the first two centuries of the empire, and that the Antonine Plague of the second century ended that growth (PMID 29760088).

The interpretation of reading lead as the pulse of the economy has been a subject of debate ever since. In 2026, a statistical analysis matched the period of the "Pax Romana" (27 BCE–180 CE), when the empire was relatively stable, year by year. Roughly one-quarter of the year-to-year fluctuations in lead emissions could be explained by three things: summer temperatures, the volume of silver coins issued, and wars. Years with warm summers tended to be those with active economic activity as read through lead, while in years when many silver coins were minted or wars were fought, lead emissions were actually lower. The authors view this as a result that supports with statistics the contentious claim that "the lead record contains information on ancient European economic activity" (DOI: 10.1016/j.eeh.2026.101765). The remaining three-quarters of the fluctuations were not explained by these three factors.

The same peak is repeated in multiple records. When collecting 89 ice, lake-bottom, and peat-bog cores from Europe and Greenland, 81 of them show a distinct Roman-era lead peak, of which 44 were interpreted as hemispheric pollution and 22 as local pollution such as nearby mines (DOI: 10.1080/14614103.2023.2181295). Lead in the ice is a record of how busy mines and smelting furnaces were, and reading that record as a record of the economy is building up statistical support.

Lead in the Air and Lead in the Body

Romans used lead in water pipes and vessels, cosmetics and medicines, and paints. Boiled grape juice to sweeten taste was purposely boiled in lead vessels. Roman writers also mentioned the poison of lead, but the typical symptoms of long-accumulated lead poisoning were first described in records in the 7th century. Lead in bone clearly increased in the Roman period, but even at its highest, it was at the level of 41–47% of today's Europeans. The authors of this review paper therefore consider that chronic lead poisoning did not contribute significantly to the fall of the Western Roman Empire (DOI: 10.4314/actat.v26i2.52570).

The pathways for lead entering the body were many, including glazed vessels, paints, cosmetics, and intentional consumption, but for the non-elite rural majority, the largest pathway may have been air pollution spewed by mines and smelting furnaces. A 2025 study translated this pathway into numbers. It is a model estimation that superimposed airflow models onto lead records in Arctic ice, and then applied the 「airborne lead → blood lead → cognitive decline」 relationship found by modern epidemiological studies. At the peak of the Pax Romana, European airborne lead increased by an average of over 1.0 ng/m³, and near smelters, it exceeded 150 ng/m³. A young child's blood lead is estimated to have risen by about 2.4 µg/dl higher than the 1.0 µg/dl estimated from the Neolithic baseline, and as a result, the intelligence quotient is estimated to have dropped by 2.5 to 3 points across the empire (DOI: 10.1073/pnas.2419630121). What this estimation speaks of is children's cognitive decline, and the abstract does not connect it to the fall of the empire.

There are also data that measured bones directly. When bones spanning 700 years were measured at a site in the northwestern Iberian Peninsula, the westernmost end of the empire, rural people of the Roman period possessed twice as much mercury and lead as post-Roman people who lived in the same place. During the Roman period, an average of 57% of lead in bones (85% at the highest) came from air pollution, and after Rome declined, that share decreased to 24%. That change mirrored the lead records accumulated in nearby wetlands (PMID 31926414). That airborne lead entered even the bodies of rural people is thus confirmed in the bones of a single village.

Lead records in ice are records of economic activity noting how active smelting was, and by themselves do not tell who was poisoned and how much. Lead in the body had to be measured separately. That airborne lead spread across all of Europe is an estimation of a model, and that that lead actually entered the body was confirmed in the bones of a single site. Lead that entered through water pipes, vessels, and boiled grape juice is another pathway not captured in the ice records.

The Weather Remembered by Trees and Mud at the Bottom of the Sea

In the mud at the bottom of the sea off southern Italy, temperatures and rainfall from around 200 BCE to around 600 CE are recorded at intervals of just over 3 years. This is the first high-resolution climate record from the heart of the Roman Empire (DOI: 10.1126/sciadv.adk1033). A paleoclimate proxy is data used to read whether a past time was dry or wet by measuring instead the traces left in natural objects that changed depending on the weather, when direct records of past weather are missing. Tree rings, cave stalagmites, and layers accumulated at the bottoms of lakes and seas all fall into this category.

The Roman Climate Optimum is a name given to a period when the climate around the Mediterranean was relatively stable and favorable for farming around the time the Roman Empire was growing. In the first synthesis bringing together written records and various natural records from Western Eurasia from 100 BCE to 800 CE, the Roman Empire rose amid a stable and favorable climate, those conditions worsened during the Crisis of the Third Century, and improved briefly again as the empire recovered in the fourth century (DOI: 10.1162/jinh_a_00379).

The mud record from southern Italy continues from the Roman Climate Optimum into the sixth century, showing instability and cooling from around 100 CE, which becomes more pronounced after around 130 CE. There were three particularly cold periods: around 160–180 CE, around 245–275 CE, and after around 530 CE. All three coincide around the times of major pandemics. The authors read this as a suggestion that climate stress was intertwined with social and biological factors (DOI: 10.1126/sciadv.adk1033). The abstract does not mention the names of the diseases and writes only "pandemics." When matched by year, the first overlaps with the period of the Antonine Plague, the second with the Plague of Cyprian, and the third with the Plague of Justinian.

In a record tracing summer rainfall and temperatures in Central Europe over the past 2,500 years using tree rings, summers were wet and warm during the prosperous periods of Rome and the Middle Ages. The period from 250 to around 600 CE with increased climate variability overlapped with the decline of the Western Roman Empire and the so-called Migration Period, when various groups moved back and forth across Europe (DOI: 10.1126/science.1197175). The earlier synthesis goes a step further, noting that regional differences in climate went hand in hand with the differing fates of the Eastern and Western empires in the following centuries, and suggests that climate outside the empire spurred two major migrations of Central Asian nomads (DOI: 10.1162/jinh_a_00379).

And in the sixth century, a sharp cold snap arrives. The Late Antique Little Ice Age is a cold period in the mid-sixth century that began when the Northern Hemisphere cooled by about 1 degree in less than a decade starting around 536 CE. When the relationship between this cold and social turmoil is measured across various societies, the relationship is statistically clear but uneven. There were societies that cooled significantly and changed very little, and societies that cooled slightly and changed greatly. The magnitude of cooling and the magnitude of social change did not match up, and the authors read this as a sign that vulnerabilities differed from society to society (DOI: 10.1177/0959683620941079).

What climate records add is the overlap of periods, and turning that overlap into the cause of the fall of the Western Roman Empire remains a matter of interpretation.

Bacteria Found Inside Bones

Human bones from a 6th-century grave contained DNA of the same bacterium, independently confirmed by two different laboratories. Ancient DNA is genetic material extracted and read from bone fragments and teeth that are thousands of years old. Yersinia pestis is a bacterium that causes plague, the very same bacterium already confirmed as the cause of the 14th-century Black Death. What the two laboratories found was the DNA of this bacterium. Until then, a heated debate had continued over whether this bacterium was the cause of the 6th-century plague, and the authors wrote that this result should end the debate over the cause (PMID 23658525).

The First Plague Pandemic is a name grouping together the entire series of plague outbreaks that began with the Plague of Justinian in 541 and repeated for a little over two centuries until around 750. Since then, genomes have accumulated. From one person in a 6th-century grave near Munich, a precise Yersinia pestis genome emerged by reading the same spot an average of 17.9 times over, which corrected 19 mutations that were not actually present in a previously published, less precise genome (PMID 27578768). This is an example of a more precisely read genome fixing the errors of an earlier one.

In 2019, human remains from 21 archaeological sites in Austria, Britain, Germany, France, and Spain were scanned to reconstruct 8 Yersinia pestis genomes. For the first time, the plague in the British Isles—previously guessed only from ambiguous historical texts—was confirmed by genes, revealing that several different strains of the bacterium circulated together in Europe (PMID 31164419). In 2025, a mass burial site from the mid-6th to early 7th century in Jerash, present-day Jordan, was investigated, yielding nearly identical Yersinia pestis genomes from five individuals. This is the first genomic evidence from the Eastern Mediterranean, where records first noted the plague, and points to an epidemic driven by a single strain of the bacterium at that time (PMID 40869973).

The situation is different for the Antonine Plague. This plague is a major pandemic that began around 165 during the reign of Roman Emperor Marcus Aurelius. In 2015, Harper wrote that based on symptom descriptions left by the contemporary Roman physician Galen, viewing this plague as smallpox is the most plausible interpretation (DOI: 10.1017/s1047759415002470). However, combining recent ancient genomes and evolutionary studies, it is now heavily doubted whether a virus capable of causing a disease recognizable to us as smallpox even existed in ancient times, and the authors conclude that it is time for archaeologists and historians to erase smallpox from ancient history (DOI: 10.1017/s1047759422000290). As of the 2025 paper, the pathogen of the Antonine Plague has not been identified by molecular evidence, including ancient DNA (PMID 39946392). What ancient DNA has definitively added to the history of Rome is the identity of the 6th-century plague and the extent of its spread, while the identity of the 2nd-century plague remains a blank.

Identity and Scale Are Different Questions

In a 2015 article, Harper could write that neither the population scale nor the long aftermath of the Plague of Justinian was doubted. Writing that the severe demographic effect of the Antonine Plague also seemed widely accepted, he added that its long-term political and social consequences were under active debate (DOI: 10.1017/s1047759415002470). Both sentences meet counterarguments afterward.

A 2019 research team directly challenged the conventional estimate that the Plague of Justinian killed tens of millions in the Mediterranean and Europe and played a part in ending antiquity. Looking at eight strands of data one by one and together—written records, laws, coins, papyri, inscriptions, pollen, ancient DNA, and cemetery archaeology—their conclusion was that none of them supported a clear connection to the plague or the assumption of a 「great catastrophe」. Placing the data together instead reveals continuity crossing the plague period, and they argue that while changes in the sixth to eighth centuries continued, the common claim that the plague was their main cause lacks evidentiary support (PMID 31792176). Another paper from the same year argued that the interpretation that half the Mediterranean population died should be discarded. The conclusion is that while the plague may have caused high mortality in some places, it neither triggered a wide population decline nor kept populations low, and any direct medium- to long-term effects, though present, were small (DOI: 10.1093/pastj/gtz009). A 2020 paper viewed this plague as gaining its immense power less from what it actually did than from a terrifying myth of 「what a plague should be」, and named rats, climate, and ancient genomes as the three pillars propping up that myth (DOI: 10.1093/ahr/rhaa510).

The other side is clearly present as well. One rebuttal points out that this 「negligible pandemic」 argument has considerable methodological weaknesses, and holds that the plague must earn its place in the transition from late antiquity to the early Middle Ages. It adds, however, that that place is on the side of cultural, religious, and social change rather than demographics (DOI: 10.25162/mhj-2020-0006). A 2021 perspective piece views some historians as trying to downplay the plague's importance right when genetics made major strides, and argues that geneticists, environmental scientists, archaeologists, and historians should together reassess the plague's impact across Afro-Eurasia beyond the Mediterranean (DOI: 10.1093/pastj/gtab024).

A city's trash heap also sits beside this debate. At Elusa, an Eastern Roman city in today's Negev Desert of Israel, the systematic hauling of garbage to the city's edge stopped in the mid-sixth century. That timing, determined by applying radiocarbon dating—which measures age using carbon left in once-living things—to burned seeds and charcoal and aligning it with pottery, closely overlaps with the start of the Late Antique Little Ice Age and the onset of the plague in 541. The authors read this record less as the plague's death toll than as an indication that the empire's frontier lost its resilience to rapid climate change (DOI: 10.1073/pnas.1900233116). Because this is the record of a single frontier city, it is difficult to broaden it to the situation of the entire empire.

The plagues of the second and third centuries face the same question. Modeling the spread of disease in the single city of Rome from 165 to 189 with multiple pathogens showed that overall mortality increased by at most 7%, and none of the compared pathogens emerged as the likely culprit. This model result conflicts with the mortality rates of historians who view the plague as having greatly reduced the empire's population. The daily death toll generated by the model for the final few years also did not match ancient records, for which the authors list three candidate reasons: exaggeration by ancient writers, other diseases, and a seasonal effect worsening in winter (PMID 39946392). A reassessment of the Plague of Cyprian concludes that this plague, named after the third-century bishop Cyprian of Carthage who recorded it, arrived alongside the invasion of the Goths who crossed the Danube, and while it clearly fueled the political and military crisis of the mid-third century, it is probably better not regarded as the root of the crisis itself. This is a point where it diverges from Harper's argument (DOI: 10.1017/s1047759421000349).

Ancient DNA confirmed what the Plague of Justinian was, but how many people died and what it changed remain questions on which both sides still clash.

Layering Three Pieces of Evidence on the Timeline

In the first and second centuries, when the empire was at its largest and most peaceful, lead emissions in the ice peaked (PMID 29760088). In years with warmer summers, economic activity as read through lead tended to be more active as well (DOI: 10.1016/j.eeh.2026.101765), and lead from the air accumulated in the bones of rural people on the Iberian Peninsula (PMID 31926414). The weather in southern Italy began to fluctuate around the year 100 (DOI: 10.1126/sciadv.adk1033).

Around 165, the Antonine Plague arrived. Around the same time, southern Italy was in a cold period (160–180), and lead in the ice dropped sharply. Three records converge at a single point in time. The authors of the lead record suggest this plague brought an end to growth, but the pathogen of this plague is unknown, and a model of Rome suggests mortality increased by at most 7% (PMID 39946392).

In the mid-third century, the Plague of Cyprian overlapped with a second cold period (245–275). Climate syntheses note that the climate deteriorated when the empire experienced the "Crisis of the Third Century," and the empire's recovery in the fourth century overlapped with a shorter period of favorable climate (DOI: 10.1162/jinh_a_00379).

And in the fifth century, the western half came to an end. Two climate studies speak directly about this period. One observation is that a long stretch of high climate variability from 250 to 600 overlapped with the decline of the Western Roman Empire (DOI: 10.1126/science.1197175), and a synthesis shows that region-specific climates ran parallel to the differing fates of the Eastern and Western empires (DOI: 10.1162/jinh_a_00379). Both directly target the decline of the Western Roman Empire in the fifth century, but what they describe is the overlap and parallelism of the timing. Around this time, lead in the ice, as seen through smelting emissions, sat in a low stretch that followed two sharp drops in the second and third centuries, and genomes confirming the plague in the West in the fifth century are not present in the data reviewed by this article.

In 536, the Late Antique Little Ice Age began. In 541, the Plague of Justinian broke out in the Eastern Roman Empire. In the mid-sixth century, garbage collection stopped at Elusa, and for the following two centuries, Yersinia pestis left traces in bones from Britain to Jordan. This was two generations after the Western Roman Empire had ended.

Event evidence such as plagues and sharp drops in lead is concentrated in the second, third, and sixth centuries, and what speaks directly to the fifth century—when the Western Roman Empire ended—is the observation that climate variability overlapped with that decline. What that observation shows is an overlap in timing.

Added and Remaining

What the three pieces of evidence added to the Roman story are three things. Lead in ice became a record showing year by year when the imperial economy was busy and when it cooled. Paleoclimate proxies told us on a scale of several years when good weather and bad weather arrived. Ancient DNA confirmed the identity of the sixth-century plague and showed that it had spread all the way to the island of Britain.

What they failed to add is also clear. The lead records do not tell us the degree of poisoning, climate records do not turn correlation into causation, and genomes do not count the number of the dead. The observation that a greater cold did not necessarily mean a greater societal shock points, in the authors' words, to differences in vulnerability (DOI: 10.1177/0959683620941079). What made those differences is a question outside climate records. The last of the three and a half critiques of Harper's book aimed at the same point. Critics noted that in that book environment is everywhere, but the actions of people, societies, and cultures are largely missing, and concluded that complex historical processes require more subtle causal explanations (DOI: 10.1111/hic3.12507).

The tools used in this article were unpacked separately in the primer series. How to date using tree rings and charcoal is in "Dating with Carbon — Radiocarbon Dates, Read with an Hourglass and Tree Rings", how to read lead and traces of life accumulated in bones is in "What Bones Tell — One Person's Diary, Split Between Teeth and Bone", and how to fish bacteria and human genes out of bones is in "What Genes Can Tell Us — An Old Book Copied Down the Generations". How much to trust the plague descriptions of Galen and ancient authors is the exact question dealt with in "Reading Historical Records — Grandmother's Diary and That Day's Newspaper".

Nature has now become a cast member in the story of Rome, but why the Western Roman Empire ended in the fifth century remains a question of interpretation that weighs political, military, and economic circumstances together.

Source Papers

• "Greenland ice evidence of hemispheric lead pollution two millennia ago by greek and roman civilizations." (1994) — PMID 17797222 — Lead in Greenland ice from 500 BCE to 300 CE was about 4 times natural levels, with cumulative fallouts at 15% of the leaded gasoline era. Greek and Roman mining and smelting caused hemisphere-scale pollution.
• "Lead pollution recorded in Greenland ice indicates European emissions tracked plagues, wars, and imperial expansion during antiquity." (2018) — PMID 29760088 — Estimated annual lead emissions from 1100 BCE to 800 CE. Peaked under the empire, plummeted around the plagues of the 2nd and 3rd centuries, and stayed low for over 500 years. Authors' conclusion: the Antonine Plague ended the growth of the first two centuries.
• "Economic dynamics of the Early Roman Empire: Insights from lead pollution, coinage, weather, and war" (2026) — DOI: 10.1016/j.eeh.2026.101765 — Summer temperatures, silver coin minting, and wars explain about a quarter of annual lead fluctuations during the Pax Romana. Statistically supports the controversial claim that lead equals economic information.
• "Environmental Impact of Roman Mining and Metallurgy and Its Correlation with the Archaeological Evidence: A European Perspective" (2023) — DOI: 10.1080/14614103.2023.2181295 — Roman-era lead peaks in 81 out of 89 columns, showing 44 hemisphere-scale and 22 regional pollution events.
• "Lead poisoning in ancient Rome" (2010) — DOI: 10.4314/actat.v26i2.52570 — Bone lead in the Roman era peaked at 41 to 47% of modern Europeans. The authors consider that chronic lead poisoning did not significantly contribute to the fall of the Western Roman Empire.
• "Pan-European atmospheric lead pollution, enhanced blood lead levels, and cognitive decline from Roman-era mining and smelting" (2025) — DOI: 10.1073/pnas.2419630121 — Estimated using ice records, atmospheric models, and modern epidemiological relationships. Children's blood lead +2.4 µg/dl, intelligence quotient reduced by 2.5 to 3 points.
• "Human bones tell the story of atmospheric mercury and lead exposure at the edge of Roman World." (2020) — PMID 31926414 — Mercury and lead in the bones of Roman-era rural people at a northwestern Iberian site were double those of post-Roman times. Atmospheric-origin lead dropped from 57% (peak 85%) to 24%.
• "Climate change, society, and pandemic disease in Roman Italy between 200 BCE and 600 CE" (2024) — DOI: 10.1126/sciadv.adk1033 — Southern Italian marine sediment with about a 3-year resolution. Cold periods after 160–180, 245–275, and 530 are linked to pandemics.
• "Climate Change during and after the Roman Empire: Reconstructing the Past from Scientific and Historical Evidence" (2012) — DOI: 10.1162/jinh_a_00379 — The empire rose amid stable, favorable climates and worsened during the third-century crisis. Regional climate differences align with the differing fates of the Eastern and Western empires.
• "2500 Years of European Climate Variability and Human Susceptibility" (2011) — DOI: 10.1126/science.1197175 — Central European tree rings. Wet and warm summers during Roman prosperity, with increased variability from 250 to 600 overlapping with the decline of the Western Roman Empire and the Migration Period.
• "Climate and social change at the start of the Late Antique Little Ice Age" (2020) — DOI: 10.1177/0959683620941079 — The Northern Hemisphere dropped about 1 degree within a decade around 536–560. The link between cooling and social upheaval is plausible but varies greatly across societies.
• "Yersinia pestis DNA from skeletal remains from the 6(th) century AD reveals insights into Justinianic Plague." (2013) — PMID 23658525 — Two independent laboratories identified plague bacterium DNA in 6th-century remains. The authors consider that the debate on the cause should end.
• "A High-Coverage Yersinia pestis Genome from a Sixth-Century Justinianic Plague Victim." (2016) — PMID 27578768 — A 17.9-fold genome from 6th-century remains near Munich. Corrected 19 false mutations in previous genomes.
• "Ancient Yersinia pestis genomes from across Western Europe reveal early diversification during the First Pandemic (541-750)." (2019) — PMID 31164419 — 21 sites in 5 countries, 8 genomes. First genetic evidence of plague in the British Isles, with multiple lineages of strains.
• "Genetic Evidence of Yersinia pestis from the First Pandemic." (2025) — PMID 40869973 — 5 individuals in a mass grave at Jerash, Jordan, with nearly identical genomes. First genomic evidence from the Eastern Mediterranean.
• "Pandemics and passages to late antiquity: rethinking the plague of c.249–270 described by Cyprian" (2015) — DOI: 10.1017/s1047759415002470 — The 2015 view that the Antonine Plague is most plausibly smallpox based on Galen's description, with widely accepted severe demographic impacts. States that the scale and aftermath of the Justinianic Plague are unquestioned.
• "Smallpox's antiquity in doubt" (2022) — DOI: 10.1017/s1047759422000290 — Synthesis of paleogenomic and evolutionary studies. Casts major doubt on whether a virus recognizable as smallpox existed in antiquity.
• "Comparative SIR/SEIR modeling of the Antonine Plague in Rome." (2025) — PMID 39946392 — No molecular evidence of pathogens. Models for the city of Rome in 165–189 show total mortality increased by at most 7%.
• "The Justinianic Plague: An inconsequential pandemic?" (2019) — PMID 31792176 — None of the eight strands of data support estimates of a "great catastrophe." Continuity, and the argument that plague was the primary driver, are unsupported by evidence.
• "Rejecting Catastrophe: The Case of the Justinianic Plague" (2019) — DOI: 10.1093/pastj/gtz009 — The interpretation of half the population dying should be discarded. Concludes there was a possibility of localized high mortality, but no widespread population decline.
• "The Justinianic Plague and Global Pandemics: The Making of the Plague Concept" (2020) — DOI: 10.1093/ahr/rhaa510 — Argues that the historiographical power of the plague stems from myth rather than actual evidence. Points to rats, climate, and paleogenomics as three pillars.
• "The 'Justinianic Plague': An "Inconsequential Pandemic"? A Reply" (2020) — DOI: 10.25162/mhj-2020-0006 — Points out methodological weaknesses in minimalist arguments. The plague has a place in the transition period, leaning more toward culture, religion, and society than population.
• "Viewpoint New Approaches to the 'Plague of Justinian'" (2021) — DOI: 10.1093/pastj/gtab024 — A response to attempts to downplay the impact of the plague. A perspective that the impact across Afro-Eurasia should be reassessed through interdisciplinary collaboration.
• "Ancient trash mounds unravel urban collapse a century before the end of Byzantine hegemony in the southern Levant" (2019) — DOI: 10.1073/pnas.1900233116 — Systematic garbage collection in Elusa ended in the mid-6th century, overlapping with the Late Antique Little Ice Age and the plague of 541. Suggests reduced resilience in the periphery.
• "The "Plague of Cyprian": A revised view of the origin and spread of a 3rd-c. CE pandemic" (2021) — DOI: 10.1017/s1047759421000349 — Introduced by the Gothic invasion of the Danube. Concludes that while it intensified the 3rd-century crisis, it is difficult to view it as the root cause.
• "Plagues, climate change, and the end of an empire: A response to Kyle Harper's The Fate of Rome (3): Disease, agency, and collapse" (2018) — DOI: 10.1111/hic3.12507 — The final part of a three-part rebuttal to Harper's book. Criticizes the deterministic approach for omitting the agency of people, societies, and cultures.