Invisible Environmental History: Infectious Disease in Late Antiquity

In: Environment and Society in the Long Late Antiquity
Author: Kyle Harper
Full Access


This study argues that the biological environment is properly a part of environmental history. The microorganisms— bacteria, viruses, protozoa—that cause infectious disease were the principal cause of mortality in ancient societies, but the particular array of pathogens was both locally specific and unstable over time. Pathogenic microbes are ecologically sensitive, so the background of local climate, and the influence of climate variability and climate change, determined patterns of disease and mortality. The connections between climate variability and climate change, on the one hand, and the disease profile of a population, on the other, are complex, and this paper traces some of the main pathways of influence, with specific reference to a few of the best known diseases and epidemic events in the later Roman period.

Introduction: Disease and Environment

The Chronicle of Pseudo-Joshua the Stylite is an account of the history of Edessa and environs in the years AD 494–506.1 While its author is unknown, the text is the work of an eyewitness who had experienced the dramatic convulsions that shook the eastern edges of the Roman empire in these years. The first half of the text describes a sequence of natural disasters, while the second narrates the war between the Persians and Romans that occurred during the reign of Anastasius I. The author’s viewpoint is coloured by his religious perspective; he sees the events of his age through the lens of the people’s sin and God’s chastisement. While the text has often been mined for details about the early 6th c. conflict with Persia, its description of the natural calamities in Edessa are an unusually vivid and detailed record of sudden environmental turbulence in one corner of the late antique world.

The environmental crisis started in AD 494/95, when a plague befell ‘all’ the people. God sent a pestilence that would make the body ‘like a mirror’, revealing the state of the soul on the exterior of the body. ‘Swellings and tumours appeared on all our citizens, and the faces of many became puffed up and filled with pus, making a fearful sight. Some had sores or pustules over their whole body, even to the palms of their hands and the soles of their feet’.2 The epidemic receded, but the ‘marks of the afflictions’ remained visible upon the victims. Retrospective diagnosis is a notoriously hazardous exercise, particularly in the case of texts written by non-medical personnel some 1500 years in the past. We will not delve into the methodological issues here, but suffice it to say that the Chronicle’s brief report of this pestilence is, despite the inherent uncertainties, exceptionally important. It is hard to imagine that a pustular rash covering the entire body, including the face, palms, and soles, could be anything other than epidemic smallpox, caused by the virus Variola major. The details of the pathology described—especially the incidence of lesions on the palms and soles—are among the most important diagnostic symptoms that differentiate smallpox from other diseases.3

Even worse lay in store for the eastern city. The Chronicle reports that in March of AD 500, there was a plague of locusts upon the fields. In April, scarcity set in, and price of grain rose. In June and July, millet (a panic crop) was sown, but it faltered. Now there was misery and true famine; subsistence migrants poured into the city. The poor turned to eating inedible plants and scraps of refuse. In the winter, mortality spiked, as disease carried off the exposed and starving refugees. The limited capacities of the municipal authorities and the Church were overwhelmed by the sick and the dead. The toll continued to rise for months, as pestilence—perhaps caused by the same pathogen as before, although we cannot say—erupted again across the region, carrying off rich and poor indiscriminately. The crisis had overwhelmed the buffers—better nutrition, preferable housing—that offered the wealthier classes modest protection from epidemic disease.4

The Chronicle of Pseudo-Joshua the Stylite is remarkable by any measure. It was featured in Peter Garnsey’s classic study of famines in antiquity, precisely because the kind of true food crisis it portrays was relatively uncommon.5 Normally, ancient societies had layers of redundancy and batteries of stored energy to make them resilient in the face of environmental fluctuation. But in this case, one by one, these strategies—storage, safety crops, political patronage, connectivity—all failed. The threshold of endurance was crossed, and the social order experienced a truly catastrophic sequence of disasters. The Chronicle reports these events with an eye for detail, for example the price of grain at various stages of the crisis, the daily death tolls, and so on. This kind of social meltdown was the exception rather than the rule in antiquity, but it was always a real possibility in an agrarian society exposed to the vagaries of the environment.

The Chronicle of Pseudo-Joshua the Stylite offers another lesson, though, for the student of ancient environmental history. As reflected in its pages, the phenomenon of infectious disease was both environmental and historical. Infectious diseases are caused by microbial agents, such as bacteria, viruses, and protozoa, that are pathogenic to humans. These germs are ecologically specific and environmentally sensitive. They are the product of evolution, and the force of natural selection constantly operates on them. Thus, the disease environment changes over time, as humans transform ecologies, as climate variability and climate change reshape local ecologies, and as pathogens evolve. If environmental history is the study of humanity’s changing relationship to nature over time, including both the physical and biological contexts of life, then disease history is properly a part of environmental history.

The purpose of this chapter is to lay out some of the relationships between the physical environment and the disease environment, with a focus on the late antique world. Throughout, we will emphasise three dynamics: the importance of local ecologies, the role of the physical climate as a cause of biological instability, and the predominance of evolution as a source of dramatic change. It is inevitably challenging to reconstruct ancient disease history, but, increasingly, traditional written and documentary sources are complemented by bioarchaeological and paleo-molecular evidence.6 The history of disease is being revolutionised by the study of skeletal remains and DNA, and in the process, a new paradigm is beginning to take shape, one which puts disease evolution in the foreground.7 Disease evolution has been a powerful and volatile force in human history. The human story has been intertwined with the rise of new and evolving pathogens, and as we have come to realise the importance of ‘emerging infectious diseases’ in our own world, we are sensitised to appreciate their importance in the distant past.

The ecological, environmental, and evolutionary background of disease history matters enormously for the study of Late Antiquity, not least because the arrival of Yersinia pestis in AD 541 has a claim to be the most consequential event in the entire period. The ongoing study of this bacterium’s genome and microbiology has helped us to understand that its advent was the result of exquisitely complex environmental and historical contingencies. The sheer destructiveness of this single pathogen is an example of the overwhelming power of nature. But the story of the first plague pandemic is a paradigmatic example of how human societies have shaped nature, preparing the way for nature in turn to redirect the course of human societies.

Two Big Ideas: Malthus and McNeill

Thomas Malthus might be considered the father of environmental studies, and his basic model emphasising the ecological constraints of human society remains (with qualification) a fruitful way to approach environmental history. Starting with the first edition of his Essay on the Principle of Population, published anonymously in 1798, Malthus developed a theory of development with two variables: human population and the food supply. Since human populations could grow faster than output in agriculture, which was strongly constrained by the availability and quality of land, population growth would eventually be slowed or reversed by one of two mechanisms. This was either by the “preventative check”, such as cultural brakes on fertility, like late marriage, or more violently by the “positive check”, by which is meant spikes in mortality via war, famine, or pestilence. In neo-classical Malthusian theory, the mortality rate is endogenously determined by the wage level, and population levels should fluctuate homeostatically around the subsistence wage.8

The theories of Malthus, ironically, proved to be the most spectacularly wrong in the case of his own country. England was in the vanguard of the Industrial Revolution, which was first and foremost a transformation in the energy basis of society. In the formulation of John McNeill, the transition from plant energy to fossil energy made us modern.9 But Malthus’ model works much better in the conditions of pre-modern economies, where the hard constraints of plant energy held. In the ancient world, energy was forbiddingly scarce, and in the long run, as his theory would predict, economic development, in the form of technical advance, was turned into population growth rather than higher incomes.10 Crudely, in the millennia between the invention of agriculture and the Industrial Revolution, populations grew, but wages did not. Life near subsistence was the common lot in the age of agriculture. When Malthus published the Essay, average global income was not significantly higher than it had been for the first farmers,11 but by 1800, there were a billion humans alive.12 At large scales, Malthusian predictions seem to hold.

However, as Keynes said, in the long run we are all dead. Even when considerable allowances are made for the complexity of the “positive check,” the study of historical demography has shown that mortality rates do not neatly obey the predictions of Malthusian theory.13 In other words, wage levels are not a very good predictor of mortality rates, which are often (though not always) found to be weakly endogenous, or strictly exogenous, to the economic status of a population. Mortality has been a wild force throughout human history, rising and falling without a clear relationship to wage levels. The main reason for this flaw in Malthusian theory is closely related to the biology of infectious diseases. In pre- modern societies, infectious diseases were the main cause of death, and the mechanisms that determined the impact of infectious disease were only partially mediated by factors under the control of the wage level. It is true that lower wages meant poorer nutrition, and that poor nutrition inhibited the body’s immune response to invasion, thereby exacerbating morbidity and mortality. But, the range of factors at play in the determination of mortality rates by infectious disease was much wider. Hence, mortality rates moved in ways that were not synchronous with changes in income levels.

In short, mortality has not been endogenous in the way Malthus’ theory would predict, because germs are not a generic and interchangeable array that act mechanistically to regulate mortality in response to wage levels. The specific biology of microbial pathogens matters, and so the history of disease must take into account a broader set of ecological factors. The first history of disease to do so was William McNeill’s Plagues and Peoples, which, in 1976, offered what might be considered the classic model of disease history.14 With little more than some core principles of epidemiology (how diseases behave at the population level) and remarkable intuition, McNeill offered a global history of disease that foregrounded change over time in response to the altered circumstances of human societies. In the classic model, humans were afflicted by a set of very old infectious pathogens from our primate days, and we picked up others during our dispersal around the globe. But the ‘Big Bang’ of disease history was the Neolithic Revolution. Sedentary habitation and close proximity to domesticated animals were decisive, allowing zoonotic pathogens to leap to humans. Diseases evolved in the various centres of civilisation, and over the course of time, trade and greater connectivity brought the disease pools of the Old World together with explosive effect, as germs found virgin populations. The colonisation of the New World that followed was an even more devastating microbial migration.15

Given that it was formed in an age without molecular evidence, it is impressive that the essentials of the ‘classic model’ have held up so well, but the intervening decades have brought new evidence and new perspectives. The 1970s were the peak of a triumphal phase in the history of medicine, as one by one the old scourges of the past were subdued by greater sanitation, vaccines, and antibiotics. The eradication of smallpox was the pinnacle of achievement for global public health; optimism reigned.16 But the victory over infectious disease has now stalled, and in some cases reversed. The terrifying roster of emerging infectious diseases—Zika, Ebola, AIDS, avian influenza, SARS, MERS, and so on—has reminded us that the ‘creative destruction’ of nature is not spent. Evolution remains a volatile source of new enemies and new tools for old enemies. In all, the importance of Old World ‘crowd diseases’, supposedly contracted from domesticated animals, looms smaller than when the classic model of disease history was constructed. The sheer density of city life is only one factor that can sustain the threat of pathogenic microbes, as the worst infectious diseases leapt to humans not from farm animals, but from wild animals. Wild animal reservoirs now seem of greater importance than the classic model ever envisioned.17

The Neolithic Revolution, then, may have been less than the ‘Big Bang’ in the history of disease. It was not a greater proximity to farm animals that has been decisive, but rather the combination of growing human populations and greater connectivity, interacting with a background of ongoing disease evolution, very often in the wild. Here, genetic evidence is rapidly filling in the picture. ‘Molecular clocks’ are a way of mathematically estimating the date of evolutionary events based on the rate at which a given microbe is thought to evolve. In recent years, it has become clear that some of the most notorious pathogens in human history—such as tuberculosis, malaria, smallpox and bubonic plague—are astonishingly young.18 A basically stable set of diseases has not existed since time immemorial; rather, the last few millennia have witnessed dramatic changes in the overall landscape of pathogenic enemies faced by humanity.

The history of disease, then, is a complex part of the human story. Sometimes the disease regime is shaped by endogenous factors: the impulse-response between population pressure and higher mortality. More often, the patterns of disease history are shaped by the specific biology—the tools as well as the constraints—of individual pathogens. These microbes are ecologically and environmentally sensitive, and they have a volatile evolutionary history. Ecological transformation, climate change, and random genetic mutation are all sources of instability and change, on various timescales, in the long and tumultuous history of our relationship with the invisible companions who share the earth with us.

Ecology, Environment, Evolution

We can appreciate the particularity of pathogenic microbes by reflecting on how rare they are. Microbial life is roughly three and a half trillion years old, and Earth is home to maybe a trillion different microbial species.19 The individual human is home to an average of 40 trillion bacterial cells alone,20 and yet most of this microbial life is indifferent to us. There are only around 1400 microbes known to be pathogenic to humans, even fewer are obligate human parasites, specific to humans; fewer still are mortally dangerous.21 These parasites (pathogenic bacteria, viruses, protozoa, fungi) have evolved particular molecular tools, known as virulence factors, that allow them to cause disease in humans. Pathogens are, in fact, highly exceptional beings.

A pathogen’s biology determines how it operates: how it is transmitted, how it invades the body, how it evades the innate and adaptive immune response, and so on. These particularities shape the ecological needs and preferences of individual germs. Consider just a few examples. The smallpox virus, for instance, was a directly-transmitted obligate human parasite and required a chain of continuous infections to survive.22 Plague is a rodent disease transmitted principally by flea bite; it is permanently present in colonies of burrowing rodents in various parts of the globe, and only crosses to humans under peculiar circumstances.23 Yellow Fever is spread by mosquito bite, and generally confined to tropical and sub-tropical climates where its vector thrives.24 Some parasites take advantage of dense populations to spread directly between humans, and thus flourish in overcrowded urban environments. Others are transmitted via the faecal-oral route, and find opportunity in unsanitary conditions; and so on.

Weather and climate exert strong controls over the ecology, and therefore the prevalence, of infectious diseases of humans at particular times and places (fig. 1). The mechanisms by which weather and climate control infectious diseases are many and complex. The physical environment can influence the pathogen itself, as well as intermediate species such as hosts or vectors, and/or the human victim. Ambient temperature, for instance, might alter microbial physiology. Disease vectors and hosts, like mosquitos, rodents, or bats, have breeding cycles that are controlled by heat and humidity. Human physiology and behaviour are influenced by temperature and precipitation, and the effects of these parameters on agricultural subsistence. The physical environment can affect human physiology, nutritional status (critically important for immune function), social behaviour (such as crowding, migration, or violence), and the effectiveness of environmental controls (e.g. sanitation). All of these mechanisms can operate on multiple timescales, from short-term ecological changes, such as a rainy year leading to a malaria outbreak, to long-term evolutionary ones. The latter could include the strong geographic patterning of many diseases, even today, or the prevalence throughout historical times of the most deadly form of malaria (P. falciparum) in much of southern Europe, as Newfield describes in his chapter of this volume.

Figure 1
Figure 1

How climate controls infectious diseases (author).

As has been understood since at least the time of Hippocrates, cyclical changes in temperature and precipitation throughout the year have complex effects on the incidence of disease. Ancient historians lack the ‘cause of death’ records that, while often crude, allow historians of the later Middle Ages and Early Modern period to trace patterns of mortality in past societies.25 But an empirical and methodological breakthrough was accomplished by Brent Shaw and Walter Scheidel, who realised that the abundant evidence for the seasonality of death offered indirect clues about disease ecologies in the Roman world.26 Tens of thousands of Christian tombstones from the later Roman empire record the date when the deceased passed into the afterlife. In the modern world, where infectious disease has receded as a cause of death, seasonal variation in mortality has been strongly (though not completely) suppressed: death comes more or less equally in all seasons. However, in environments where infectious disease is a predominant factor, mortality rates fluctuate across the year in patterns that reflect the specific habits of the pathogenic microbes that cause death. Maybe the most notable fact to emerge from these studies was the reality of strong regional differences in patterns of death within the Roman provinces. Egypt looked different from Rome, northern Italy different from southern Italy; local ecological factors clearly exerted influence on the seasonal rhythms of dying in the Roman empire.

The seasonal mortality patterns from ancient Rome have revealed a strong amplitude of variation, with a huge wave of deaths in the late summer to early autumn. When the data are controlled by age group, it is notable that the elderly (and only the elderly) experienced a small, secondary wave in the winter, likely due to respiratory infections that commonly prey on those with weaker immune systems in the cold months. Even by the standards of pre-modern societies, the Roman population seems to have experienced an unusual degree of seasonal variation in mortality. The pattern is even more striking given that the Roman sample is biased toward young adults, who generally comprise the hardiest sub-population. The sheer amplitude of variation makes a prima facie case that infectious disease was an exceptionally strong mortality factor in ancient Rome.27

The dominance of high summer and early autumn in the mortality regime at Rome suggests two kinds of culprit at work. First, it has been reasonably hypothesised that the ‘dog days’ of summer exacerbated mortality from gastro-enteric diseases, caused by bacteria such as Salmonella and Shigella. In a city with an extraordinarily large and densely packed population, the challenge of environmental sanitation overwhelmed even the Romans’ engineering prowess. These diseases are spread via the faecal-oral route, ingested in contaminated food and water. They cause fever and diarrhoea and, particularly in a world without antibiotic pharmaceuticals, can cause severe morbidity and mortality, especially among frail sub-populations. The Roman habit of early weaning meant that infants were perilously exposed to these pathogens, although infant mortality remains something of a ‘black box’ in the case of the ancient Romans.28

Second, the surge of deaths in the autumn has been taken as a signature of malaria. The seasonal mortality profile is only one indication among many that malaria was an especially heavy burden on the Romans. Thanks above all to the work of Robert Sallares, we have a detailed and reliable understanding of the role of malaria in Roman antiquity.29 Malaria is a disease caused by Plasmodium protozoa. These are single-cell parasites with complex life-cycles, transmitted between humans via the Anopheles mosquito. There are different species of malaria-causing Plasmodium protozoa, including P. malariae and P. vivax. While these are not to be underestimated, they caused a milder, feverish disease with lower rates of mortality than the great killer, P. falciparum. As Newfield shows in his wide analysis of the source material, P. malariae and P. vivax were dangers even in more northern parts of Europe in Late Antiquity and the Early Middle Ages, while P. falciparum seems only to have been endemic in southern Europe.30 Both the protozoa and the mosquito vectors are ecologically sensitive, and, in Newfield’s words, “the less severe the malaria the more tolerant its parasite and its vectors are of cold weather”.

Plasmodium diseases are old. They originated in the African tropics; but now, genomic evidence is sharpening the picture. The deadliest branch of Plasmodium, P. falciparum, is an evolutionary novelty. It is a descendant of a gorilla pathogen, and may be less than 10,000 years old.31 It moved into the Mediterranean in the millennia before the Romans built their empire, but a variety of factors aligned in the Roman period to make it an especially formidable nemesis.

Roman medical authors, including the 2nd c. AD physician Galen, reveal an intimate familiarity with malaria, which they knew, in its most virulent form, as semi- tertian fever. They associated it closely with life in the capital itself:

We no longer need the word of Hippocrates or anyone else as witness that there is such a [semi-tertian] fever, since it is right in our sight every day, and especially in Rome. Just as other diseases are typical in other places, this evil abounds in this city.32

Now there is also firm evidence from archaeological genomes that malaria was a presence in the Roman empire.33

The name malaria comes from the Italian for ‘bad air’, and it is the quintessential ecological disease. It was not uniformly distributed across the empire, but rather concentrated in areas that supported the breeding of the mosquito vector. Malaria is above all a wetland problem; it flourishes in marshy and wet places. The specific configurations of Roman economy and society may have unleashed the malarial potential of the landscape. The Romans felled forests in great swaths, accelerating run-off and making it easier for mosquitos to breed. They drained swamps and built roads through pestilential landscapes, and their civil engineering brought water into close proximity to human habitations everywhere. Wherever it takes hold, malaria is not just one disease among others; it is a burdensome presence, exacerbating malnutrition and weakening immunity against other pathogens. It seemed to hasten the corruption of all life: “Why do men grow old slowly in places with fresh and pure air, while those in hollow and marshy places grow old rapidly?”34

If human transformations of the landscape shaped the history of malaria, so too did natural climate change. The breeding habits of the Anopheles mosquito are sensitive to climate parameters, on multiple time-scales. In the long term, the climatic conditions of the early imperial period may have been propitious for the mosquitos. The Roman Climate Optimum (RCO) (ca. 200 BCAD 150) was a phase of Late Holocene history when the circum-Mediterranean was not only consistently warm but also anomalously wet.35 A range of climate proxies suggest that the conditions of the RCO mirrored the millennia of the Mid Holocene, which were much warmer and wetter in the regions controlled by the Roman empire. It is altogether likely that these climate parameters exacerbated the impact of malaria in core regions of the empire.

On shorter timescales, inter-annual variability in the climate drove wild oscillations in the incidence of malaria. Much of the Roman empire sat along the border between the temperate mid-latitudes and the sub-tropics, and the prevalence of malaria here could be exquisitely sensitive to the fluctuation of the climate, which might send the disease escalating toward epidemic heights. Galen reported the conventional wisdom, while remaining unaware of the actual mechanisms: “… when the entire year becomes wet or hot, there necessarily occurs a very great plague.”36 In early modern Italy, epidemic outbreaks of malaria erupted on average every five-eight years.37 The ancients, lacking germ theory, did not imagine, much less specify, the pathogenic agents of epidemic events; their terminology for ‘plagues’ or ‘pestilences’ was stubbornly generic (lues, loimoi). It is evident that short-term climate anomalies (hot or wet years) were the proximate cause of most disease outbreaks recorded in the sources. It is also likely, but not provable, that malaria was a principal cause of many of these ‘plagues’.

The incidence of epidemic events, malarial or otherwise, points toward one of the most important features of the demographic regime in Late Antiquity: its instability. The endemic disease pool is the ‘background’ set of diseases present in a population. Epidemic disease refers to the sudden upsurge in death due to one or more pathogens, and just as the cycle of the year was marked by strong variations in the timing of mortality, so there was enormous volatility between years. Death came in waves, not as a steady drip. The causes of an epidemic can be extraordinarily complex. Climate variability can affect the behaviour of pathogen, vector, or host; food shortage may lower the nutritional status, and therefore the immune resistance, of a population. External factors, like siege warfare or army movements, can stir epidemics; and other more purely biological factors, like changes in the balance of people with acquired immunity in a population, or changes in the virulence of a pathogen, can also trigger disease outbreaks.38

In the deep background of changes to the disease environment lies the constant evolution of the microscopic organisms themselves, as the interplay of mutation and natural selection creates novel forms of old pathogens and also new species. It should also be acknowledged that horizontal gene transfer—in which genes are transmitted between organisms, instead of vertically from parent to offspring—has been important in the evolution of microbial organisms. In a sense, pathogens can bypass some of the ‘trial and error’ of random mutation by absorbing already functional genetic tools from other bacteria.39 Because the particularity of individual germs is such an important factor in determining mortality, the arrival of new pathogens has continually reshaped the demographic regime and, in some cases, intervened forcefully in the course of events at the highest level.

One notable instance in which the Roman disease pool witnessed change was the diffusion of tuberculosis in the centuries of Roman rule.40 TB is, still today, a devastating respiratory illness caused by the bacterium Mycobacterium tuberculosis. The bacterium causes a chronic infection, grinding down its victims over the course of weeks or often years with coughing and consumption. In the past, it was a lethal presence, rather like malaria, casting a spell over entire societies where it took hold. The history of TB has been rewritten in the last few years, thanks to genomic evidence. Here is a paradigmatic case where the classic model of disease history has been overturned: while it was long believed that humans contracted TB from cattle, in fact human TB is ancestral to bovine TB:41 we made the cows sick, not vice versa. TB is a relatively young disease, maybe only 5,000 years old, but even within its short time as a species, Mycobacterium tuberculosis has experienced an eventful history, with evolutionary developments that have altered its virulence. Genomic analysis has revealed that a major evolutionary event occurred between 1800 and 3400 years ago, leading to the most virulent modern lineage. Future work will narrow this event window, but sometime in the centuries of classical antiquity, TB took a turn towards a deadlier career.

The example of TB is interesting for another reason, too. Whereas most infectious diseases do not leave characteristic marks on skeletons, the pathology of TB leaves visible traces of its presence in bones. Thus, bioarchaeologists can follow its diffusion in the skeletal record. TB is hard to find in pre-Roman skeletons in Europe, but it becomes far more common in the period of the Roman empire. These centuries have been called “… a watershed moment for the spread of tuberculosis in Europe”.42 The empire created the ecology where TB could thrive; the interconnectedness of the empire, and its dense urban habitats, were conducive to the spread of a truly devastating killer. The genomic and skeletal evidence alike are pointing to the Roman empire as an important part of the story of TB.

Tuberculosis was a chronic disease. It moved slowly, claiming its victims in an unhurried way and diffused across the Roman empire at a snail’s pace. In this, it was limited by its own mechanisms of dispersal. We should pause to appreciate how historically contingent the Roman disease pool was; it was shaped by the biology of the pathogens that existed in this time and place. The movement of these pathogens was constrained by their mechanisms of transmission. Parasites that depend on mosquito vectors or the faecal-oral route can only spread so far, so fast. The spread of slow-moving, chronic infections like tuberculosis highlights the potential inherent in Roman systems of communication, but, in the early empire, acute infectious diseases seem strongly limited by their inherent constraints. On close inspection, there were no large-scale, inter-regional epidemics over a long span of time between the late republic and early empire. Disease outbreaks were local or regional in scale, spatially limited by the agents of diseases like malaria or paratyphoid fever. The arrival of highly communicable acute infectious diseases in the auspicious conditions of the Roman empire would prove to be explosive.

In Late Antiquity, the Roman empire intersected with the evolutionary history of some of the worst pathogens humanity has ever encountered. The entire period between the 2nd and 7th c. might be considered an age of pandemics. By giving attention to the spatial and temporal dynamics of disease events, we can put into relief how exceptional in scope and magnitude a series of outbreaks in Late Antiquity truly were. Whereas previous epidemics were regionally confined, starting with the pestilence in the reign of Marcus Aurelius (known as the Antonine Plague), new germs—what we would today call ‘emerging infectious diseases’—began to arrive to rattle the foundations of Roman civilisation. For the first time, there were disease events truly deserving the label of pandemic, an excess mortality event on an inter-regional or continental scale.

The reasons for the advent of pandemic mortality events are ecological, environmental, and evolutionary. The Roman empire fostered an ecology that was conducive to the transmission of acute infections: its thick web of connections and dense cities were advantageous for microbes. At the same time, repeatedly, sharp environmental changes seem to lie in the background of the major pandemic events. Given the nature of the ancient evidence, the causal links between climate change and pandemic events remain obscure, tantalizingly out of reach. But in each case, we can note the circumstantial evidence and consider reasonable hypotheses. The coincidence between global-scale climate change and the eruption of Roman pandemics is not disputable; the causal links are necessarily speculative. Along with this, finally, there is evolution. The picture that is coming together, from a combination of molecular and historical evidence, points to the role of evolutionary novelties in causing pandemic events.

In AD 165, a pestilence arrived in the Roman empire,43 and contemporary observers were shocked by its scale and its toll. The Antonine Plague was associated in ancient sources with the invasion of Parthia under Marcus Aurelius and Lucius Verus, and specifically the sack of Seleucia on the Tigris. It was said that a Roman soldier opened a chest in a temple of the “unshorn Apollo” and released a miasma that spread over the whole world. While it is plausible that one spur of the pandemic ran through Parthia, and was accelerated by the return of Roman legions to the empire, we should not be too credulous. The disease was inside the empire at least a year before the return of the troops, and it is probable that evidence for a devastating pestilence in Arabia recorded in the AD 150s—noted, somewhat unusually, in both Roman sources and Arabian inscriptions—reflects the march of the same disease towards Roman borders.44

There had simply never been anything like this disease event before; within a few years, it was taking victims from one end of the empire to the other. It has left a large amount of evidence, of various kinds. The pandemic motivated a sudden outburst of religious activity aimed at the god Apollo, the debris from which are remarkably preserved in inscriptions across the empire.45 The written sources testify to a disease event that unsettled contemporaries, laid waste whole regions, and decimated the army. By sheer luck, the Antonine Plague overlapped the career of the most prolific ancient medical writer, Galen, whose observations of the “greatest plague” are vitally important for understanding the disease. He describes a disease that caused oesophageal lesions, bloody diarrhoea, a black extrusive rash covering the whole body, with scabbing and scarring in the aftermath, and a crisis around days nine-twelve.46 His descriptions of the pathology have led modern historians to identify the agent of the Antonine Plague as the smallpox virus. Retrospective diagnosis is full of uncertainty, and it should be pointed out that Galen’s description does not unequivocally point to the differentiating symptoms of smallpox. The consensus that has formed among modern historians needs to remain a hypothesis, until there is positive paleomolecular identification from an archaeological victim. But smallpox remains the best candidate, and it is likely that genomic evidence will eventually confirm the hypothesis.

Certainly, the genomic evidence is already starting to fill in the broader history of smallpox. Two recent studies are of particular relevance for understanding the history of the virus. One reconstructs the phylogenetic tree of the orthopoxviruses, which include Variola major, the smallpox virus.47 It shows that the closest relatives of smallpox are camelpox virus and Taterapox virus. The latter infects only the naked sole gerbil, a small rodent that inhabits the dry forests of Africa. All three diverged from an ancestral rodent poxvirus, only 2,000–4,000 years ago, somewhere in Africa. A 2016 study reconstructed the genome of Variola from a 17th c. mummy from Lithuania; the startling discovery was that this virus sat on a lineage ancestral to all known 20th c. smallpox specimens. What this implies is an extremely volatile and extremely recent evolutionary history, closely connected with modern globalisation.48 So, at present, the early history of smallpox is a story full of gaps, to be filled in with more molecular evidence.

The case of the Antonine Plague is important, since it is conceivably the historical debut of the smallpox virus in Eurasia. It could have evolved in Africa, very shortly before the Roman pestilence, and travelled along the trading networks that the empire had intensified along the Red Sea and Indian Ocean. The middle of the 2nd c. was also a period of climate reorganisation on a global scale; it marked the end of the Roman Climate Optimum and the beginning of a period of much sharper climate variability.49 The global nature of this change is reflected in the fact that even the Nile flood, dependent on the monsoons, suddenly becomes more erratic. Perhaps these vibrations in the climate system helped to stir viral evolution, or somehow encouraged the spill-over events that allowed the smallpox race onto the conveyor belt of global trade; but these are speculative hypotheses. However, it does not seem to be the case that the Antonine Plague was brought on by the greater vulnerability of the population due to Malthusian pressures; real wages were rising right down to the advent of the pandemic.50 The background of the pandemic is to be found in the ecology of the empire, global climate turbulence, and the evolutionary history of the smallpox virus.51

Mortality estimates for the Antonine Plague have varied from as little as 2% of the population to as much as 25–30%, although I have argued for something in the realm of 10% of the empire’s population as a whole. That would represent an astonishing 7–8 million victims, probably the largest single mortality event in absolute terms up to that point. Nonetheless, the pestilence did not topple the empire; even if staggered, the empire proved resilient and enjoyed a period of renewed demographic and economic expansion under the Severan dynasty. The next pandemic event, however, proved more fundamentally disruptive.

In recent work, I have tried to call attention to the importance of the episode known (somewhat misleadingly) as the ‘Plague of Cyprian’.52 This pestilence struck the empire in the obscure and tumultuous years ca. AD 249–62. Despite the relative paucity of source material for this period, the attestation of a massive disease event is abundant. Some 24 testimonies, including seven independent eyewitness sources, describe a disease that struck, again within the space of a couple of years, from one end of the empire to the other. Contemporaries were horrified by the grisly course of the disease and its huge death tolls. Although this event has been neglected in recent historiography, it in fact belongs among the few pandemic events of the ancient world.

Once again, we sense the role of global climate events in the background of the pestilence. The empire was afflicted by an epochal inter-regional drought from ca. 244. It included Egypt, which points to the role of global-scale climate perturbations. The drought is attested by contemporaries, one of whom claimed that the bed of the Nile River was dry.53 It is also evident in the papyri, which reflect what has been called the worst food shortage in Egypt throughout the entire span of Roman history.54 Yet, as with the case of the Antonine Plague, we cannot offer firm hypotheses about the mechanisms of the relationship between climate turbulence and the disease event. In this case, it is more likely that the widespread food crisis rendered the population susceptible to infectious disease, but this does not exclude other possibilities. We do not know the causative agent of the Plague of Cyprian, although I have outlined the hypothetical case for a viral haemorrhagic fever.55 It was an exotic pathogen of exceptional virulence, and this time the empire was not able to endure the environmental and demographic shocks. The empire experienced a multi-faceted social, economic, political, and military crisis, of which environmental factors were a major cause. Whereas the empire emerged on the other side of the Antonine Plague in relatively unchanged form, the Plague of Cyprian helped to precipitate more radical transformations in the basic structures of empire. This crisis cut deep, although the new empire that took shape would endure for centuries.

After the dual blow of these two pandemics, there was nothing on a similar scale for a long time. The disease environment of the later empire remained unfavourable and volatile, but disease outbreaks were regionally contained. This lull would end, however, in dramatic fashion, when the single most violent sequence of climate change and biological catastrophe in all of antiquity unravelled in the 530s and 540s.

The Great Event: The Plague of Justinian

From his accession as sole ruler in AD 527 down until 541, the career of Justinian was a resounding success. He had made peace with Persia, reconquered Africa, and subdued most of Italy. He survived a coup, reformed the imperial administration and codified all of Roman law. He went on a building binge as expansive as any emperor in Roman history, crowned by the construction of the Hagia Sophia in Constantinople. Behind two centuries of demographic and economic growth in the eastern provinces, the eastern Roman empire was resurgent. Then the environment intervened to turn the course of events. The reversals of the age shocked contemporaries: “I cannot understand why it should be the will of God to exalt the fortunes of a man or place, and then to cast them down and destroy them for no cause that is apparent to us.”56 In contrast to the famous achievements in the earlier part of Justinian’s rule, Mischa Meier has evocatively called the rest of the emperor’s reign “the other age of Justinian”.57

The environmental crisis in the age of Justinian was marked by changes in the physical as well as the biological climate.58 In both cases, natural archives play a crucial role in complementing the written evidence. Historians have long known that the year 536 was a ‘year without a summer’,59 but for various reasons, including a lack of calibration in the proxy records, the precise sequence of natural events remained uncertain. Recently, however, a satisfying alignment in the physical proxy record—especially ice cores and tree rings—has been achieved, as Newfield describes in this volume.60 In early 536, there was a huge volcanic eruption in the northern hemisphere; megatons of sulphate aerosols were ejected into the stratosphere, reflecting solar radiation back into space. Then, in 539 or 540, a second and even more powerful tropical explosion occurred. These events are registered in proxy records around the world, and they constitute one of the strongest climate anomalies in the last several millennia. The decade 536–45 was the coldest decade in the last 2000 years.61

The effects of climate turbulence were felt principally in the long run.62 The Roman empire seemed able to withstand a year of poor harvests, but the instantaneous cooling was not transient. Independently, the levels of solar energy emitted by the sun declined in the 6th and 7th c., and the entire period has started to be characterised as the ‘Late Antique Little Ice Age’. Yet, the most consequential effect of this raucous climate instability was registered in changes within the biological environment. Right in the aftermath of the spasm of volcanic activity came the worst mortality event of the ancient world: the Plague of Justinian.

We cannot say with certainty how the climate catastrophe sparked the disease outbreak, but we must believe that the two were causally linked. To understand why, we must appreciate that an epidemic of bubonic plague, caused by the bacterium Yersinia pestis, requires an exquisitely complex ecological platform. Yersinia pestis has been the agent of three historical pandemics.63 The first of these erupted under Justinian,64 the second was the medieval pandemic that started with the Black Death in 1346. A third erupted in 1894 in Yunnan Province, China and diffused globally. The first two pandemics were probably the most devastating mortality events, in proportionate terms, in all of human history.

Humans, though, have been merely incidental victims in what is really an animal disease; Y. pestis is a bacterium of rodents spread principally by flea bite, and colonies of burrowing rodents are its natural reservoir.65 The bacterium probably evolved in central Asia, whose highlands are home to great colonies of marmots and gerbils. Most of the time, the plague remains enzootic, quietly lurking in its maintenance hosts.66 Under the right conditions, plague will become epizootic, spilling beyond its host species into other animals, including rodents like the black rat, Rattus rattus. In recent years, the versatility of Y. pestis has become more broadly appreciated; it is capable of infecting a wide range of species.67 But, its most notorious historical spill-over events seem closely connected to Rattus rattus, a commensal rodent that lives in close quarters with humans.68 A human epidemic is actually the collateral damage of an underlying epizootic event. Thus, there are at least five species involved in a plague outbreak: the bacterium, the maintenance host, the commensal rodent, the flea vector, and humans. The sheer complexity of this arrangement, and the intricate dependence on the black rat epizootic, long engendered doubt that Y. pestis could have been the agent of the historical pandemics. But in recent years, genomic confirmation has laid to rest any doubts that Y. pestis was the guilty party, in what were in fact disease events of stunning contingency.69

The plague came to the Roman empire from beyond. Genomic evidence indicates that the lineage of Y. pestis that caused the Plague of Justinian is most closely related to specimens in present-day grey marmots and long-tailed ground squirrels in the Xinjiang region of western China. In the words of Monica Green, “All narratives of plague’s history must be connected to that place of origin.”70 The arrival in 541 of Y. pestis at Pelusium, on the southern shore of the Mediterranean, was a conspiracy of evolutionary, environmental and ecological circumstances. In evolutionary terms, we are learning that Y. pestis is not an ancient pathogen that has existed unchanged, since time immemorial. It has been found in a range of Bronze Age burials from Eurasia, and these archaeological samples of the bacterium have revealed important genetic changes in the millennium leading up to the Plague of Justinian.71 These genetic adaptations allowed Y. pestis to become the deadly germ behind the pandemic event.

The abrupt climatic changes preceding the outbreak triggered the spill-over event, even if the mechanisms remain uncertain. Still today, climate perturbations regulate the incidence of plague. In China, El Niño years are correlated with plague outbreaks, and there is a strong relationship between volcanism and El Niño, so the volcanic eruptions of 536 and 539/40 could have sparked the sequence of events leading to the pandemic.72 In truth, though, the climate could have acted on any of the five organisms involved in a plague event. Rain might have fostered vegetation growth, that instigated runaway population explosions among the burrowing rodents. Indeed, in a recent book, I argue that both long-term and short-term climate mechanisms played a role, with several decades of heavier precipitation in central Asia leading up to the plague probably resulting in rapid growth in rodent populations. Sudden cooling may have helped the temperature-sensitive rat flea to continue to reproduce in the sweltering Indian Ocean, and survive the southern passage into the Roman empire. Grain shortages could have stirred human movements, or rendered hungry bodies vulnerable to infection, though I am in agreement with Newfield’s arguments, in this volume, that famine is so far a weak candidate for a trigger of the plague. In all likelihood, several of these climate mechanisms were at play simultaneously.

Perhaps the most intricate requirements for the plague pandemic, though, were ecological. If the hand of nature was at work in genetic evolution and the climatic background, the human role was supreme in constructing the ecological platform for the plague. In at least two senses, human civilisation was responsible for establishing the conditions under which the plague pandemic could occur. The first was the web of connections binding east and west.73 A robust trade network across the Red Sea and Indian Ocean carried silk, spices, slaves, ivory, aromatics, and other commodities: as well as germs. Tellingly, the Plague of Justinian first appeared at Pelusium, right at the hinge of the Mediterranean and the Indian Ocean worlds. The second human factor was the diffusion of the black rat. Native to south-east Asia, Rattus rattus is an invasive species, and McCormick has shown that its colonisation of the West was part and parcel of Romanisation.74 The storage and transport of grain was conducive to the expansion of rat colonies all over the empire. The massive rat colonies of the Roman world were the fuel to the fire of the pandemic. In 541, Y. pestis met the rats of the Roman empire, and the consequences were explosive.

The transmission of the disease by flea bite is the key to its insidious power.75 In the most common course of infection, the dermis is inoculated with Y. pestis, and the lymphatic system drains the bacterium to the nearest lymph node. There Y. pestis replicates, causing the visible swelling of the lymph node, called a bubo in Greek. Fevers, chills, headache, malaise, and delirium move quickly, and the lymph node grows like a tumid orange or grapefruit.76 The victim succumbs to sepsis, with case fatality rates around 80%. Less often, the bacterium can enter directly into the bloodstream and cause sepsis with exceptional speed, a course of infection called primary septicaemic plague. The bacterium can also spread from the lymphatic system into the lungs, causing secondary pneumonic plague. This form of the disease was uniformly fatal. All of these courses derive from the flea bite. It seems increasingly likely that plague could travel not only by the rat flea but via other ectoparasites, including human fleas. Furthermore, the plague could also spread directly between humans via aerosol droplet expelled by coughing. When plague was inhaled into the lungs and induced respiratory infection, it caused primary pneumonic plague. But the significance of primary pneumonic plague was probably limited in the pandemic, and fleas remained integral to the dynamics of the outbreak.

We have two star witnesses, as well as a number of minor ones, to the first outbreak of the Plague of Justinian.77 The eastern churchman John of Ephesus, and the classicising historian Procopius, each provided vivid testimonies to the plague’s destruction. Despite their radically different cultural outlooks, their accounts are remarkably consistent, and allow us to reconstruct in outline the course of the event. It erupted in Egypt in 541 and diffused over land and sea across the eastern Mediterranean. By 542 it was in Constantinople, and the next year in the West.78 Its destruction was vast in town and country, and in the capital it was horrific:

At first only a few people died above the usual death rate but then the mortality rose higher until the toll in deaths reached five thousand a day, and after that it reached ten thousand, and then even more.79

According to John, the dead were counted until the number reached 230,000, but after that the authorities lost count. Both authors claim that basic order broke down, in the food supply, retail commerce, and municipal governance. At first the dead were buried, but when room ran out, piles of corpses were ferried across the inlet to Sykai, where the dead were thrown en masse into giant military towers. The events at Constantinople are described in the greatest detail, but the disease reached the Danube, Italy, North Africa, Gaul, Spain, and the British Isles. According to Procopius, half the population was lost.

Historians of Late Antiquity have struggled to know what to make of the Plague of Justinian; was it a truly destructive event?80 The preponderance of the evidence has always leaned towards the conclusion that it was an episode on a par with the medieval Black Death. The literary sources insist on the plague’s massive toll, and these accounts deserve more sympathetic reading than they have sometimes been given. They are independent, vivid, and credible. Attestations of pandemic events in the ancient record are truly limited, and deserve cautious consideration, not brash dismissal. Moreover, the continued accumulation of evidence underscores the credibility of the literary texts. McCormick has just published a database of mass graves from Late Antiquity and found a major rise in the 6th and 7th c.; he cautiously and plausibly associates this phenomenon with the arrival of pandemic plague.81

The maximalist case for the plague’s impact, though, has received stunning confirmation in the last few years. Two cemeteries, from Aschheim and Altenerding outside Munich, have yielded ancient DNA evidence for Yersinia pestis from the 6th c.82 The significance of these discoveries could not be greater. First, the positive identification of Y. pestis vindicates the description of the disease in the written accounts. Second, the bacterial genomes recovered show that the agent of the plague was a form of Y. pestis similar to the lineage that caused the medieval pandemic. Third, the place of the discovery matters: the bacterium was found in some rather modest multiple burials in tiny villages in the German countryside. By far the most uncertain hypothesis in the maximalist case for the Plague of Justinian was that the disease spread far and wide, including into rural settlements. In short, the discovery of the bubonic plague at Aschheim and Altenerding confirms the worst: if the monster was here, it was everywhere.

What made the plague so devastating was its mode of transmission, which relied on the underlying rat epizootic. It was thus able to penetrate more easily into the rural sphere, where most of the population lived. The medieval Black Death carried off 50–60% of the European population. Our 6th c. evidence does not allow for the same analysis or confidence that supports such conclusions, but we can say that everything we know is consistent with a death toll in the same range for the first episode in 541–43.83 And even with this violence, the pandemic was not finished. After the initial outbreak, the bubonic plague recurred for over two centuries in the circum-Mediterranean and Near East. Every 10–20 years, the disease erupted again, and was able to do so for two reasons. First, the immunity conferred on survivors of infection by Y. pestis is weak and transient. Second, and more importantly, I argue in a recent book that the bacterium became established in host rodent populations in the West.84 We should imagine the relapse events not as reintroductions from without, but as amplifications from interior plague reservoirs. Hence, the metaphor of epidemic ‘waves’ is misleading. The recurrent plague events were explosions from within, spatially complex events shaped by physical and human geography. As Newfield observes in this volume, the ‘Late Antique Little Ice Age’ is the backdrop for the broader First Pandemic, and recurring climate anomalies could have played a role in each of the amplification events.

We do not know where, exactly, the plague reservoir was, and there could have been more than one. From the first outbreak down to around 620, it is clear that Constantinople was a major node in amplification events. Probably, the disease found its way to the capital of the empire, which then acted as a nexus of metastatic dispersal. Even the western amplifications depended on an integration with eastern networks. Then, from 620 onwards, the importance of Constantinople declined, while the focus of the disease in Syria and Iraq remained intense.85 Then, in the middle of the 8th c., the disease disappeared as mysteriously as it had arrived. The DNA evidence shows that the Justinianic Plague was caused by an extinct branch of Y. pestis. The bacterium retreated to its haunts in central Asia, and the Black Death was a new introduction event. The role of climate change in the end of the First Pandemic deserves more attention, and a comparative analysis with the medieval pandemic, too. Certainly, public health responses centred on ‘quarantine’ played no role in the end of the First Pandemic, so answers must be sought in some combination of the physical climate, rodent ecology, and epidemiology.


The impact of the Plague of Justinian was registered in the long term, in two centuries of demographic history burdened by the presence of this single, devastating germ. Moreover, the longer term effects of a cooler, less hospitable climate acted in synergy with biological catastrophe. The ultimate result, I argue in a recent book, was a prolonged sequence of stagnation and state failure, stretching from the 540s to the 630s. The particular history experienced by each of the societies under the umbrella of Roman rule was determined by a range of local factors, and some regions suffered worse than others. But the general fact of demographic volatility and decline, and the fiscal disequilibria of the central state, proved insuperable.

We should not discount the resilience of these ancient societies, nor the circumstantial and human element in the final events of the later Roman period. But it would be distorting not to attribute strong agency to nature as well. We have learned, thanks to new physical evidence, that the worst biological enemy in the history of humanity arrived on Roman shores at this time, and that the most pronounced episode of climate change in the Late Holocene struck almost simultaneously. The history of Late Antiquity needs to be written, in the provocative words of a recent book on modern environmental history, “as if nature existed.”86 Late Antiquity witnessed violent turbulence in the physical and biological environment. We do justice to the men and women who worked to make their world out of such conditions by acknowledging the powerful role of nature in their story.


Primary Sources

  • Euseb., Hist. eccl. = G. Bardy ed., Eusèbe de Césarée. Histoire ecclésiastique. Livres V–VII. Texte grec, traduction et notes (SC 41) (Paris 1955).

    • Search Google Scholar
    • Export Citation
  • Gal., De Temperamentis = G. Helmreich ed., Galeni De temperamentis libri III (Leipzig 1904); P. N. Singer transl., Galen: Selected Works (Oxford 1997).

    • Search Google Scholar
    • Export Citation
  • Gal., De morborum temporibus = I. Wille ed., Die Schrift Galens Peri tõn en tais nósois kairõn und ihre Überlieferung (Kiel 1960).

    • Search Google Scholar
    • Export Citation
  • Procop., Pers. = G. Wirth ed., Procopii Caesariensis opera omnia, vols. 1–2 (Leipzig 1962–1963); A. Kaldellis transl., The Wars of Justinian (Indianapolis 2014).

    • Search Google Scholar
    • Export Citation
  • Pseudo-Aristotle, Problemata = I. Bekker ed., 1960, Aristotelis opera, vol. 2 (Berlin 1960).

  • Pseudo-Joshua, Chron. = J.-B. Chabot ed., Incerti auctoris chronicon anonymum Pseudo-Dionysianum vulgo dictum 1 (Corpus scriptorum Christianorum orientalium 91) (Paris 1927); F. Trombley and J. Watt edd. and transl., The Chronicle of Pseudo-Joshua the Stylite (Translated Texts for Historians 32) (Liverpool 2000).

    • Search Google Scholar
    • Export Citation

Secondary Sources

  • Achtman M. (2016) “How old are bacterial pathogens?”, Proceedings of the Royal Society B: Biological Sciences 283 (2016) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Allen P. (1979) “The ‘Justinianic plague’”, Byzantion 49 (1979) 520.

  • Ari T. B. et al. (2011) “Plague and climate: scales matter”, PLoS Pathogens 7 (2011) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Arjava A. (2005) “The mystery cloud of 536 CE in the Mediterranean sources”, DOP 59 (2005) 7394.

  • Babkin I. V. and Babkina I. N. (2015) “The origin of the variola virus”, Viruses 7 (2015) 11001112.

  • Barrett R. and Armelagos G. (2013) An Unnatural History of Emerging Infections (Oxford 2013).

  • Benedictow O. (2004) The Black Death, 1346–1353: the Complete History (Woodbridge 2004).

  • Biraben J.-N. (1975) Les hommes et la peste en France et dans les pays européens et méditerranéens, vol. 1: La peste dans l’histoire (Paris 1975).

    • Search Google Scholar
    • Export Citation
  • Biraben J.-N and Le Goff J. (1969) “La peste dans le haut moyen âge”, AnnÉconSocCiv 24 (1969) 14841510.

  • Bos K. I. et al. (2014) “Pre-Columbian mycobacterial genomes reveal seals as a source of New World human tuberculosis”, Nature 514 (2014) 494497.

    • Search Google Scholar
    • Export Citation
  • Brock S. (1979–1980) “Syriac historical writing: a survey of the main sources”, Journal of the Iraqi Academy (Syriac Corporation) 5 (1979–1980) 296326.

    • Search Google Scholar
    • Export Citation
  • Brooke J. (2014) Climate Change and the Course of Global History: a Rough Journey (New York 2014).

  • Bruun C. (2012) “La mancanza di prove di un effetto catastrofico della “peste antonina” (dal 166 d.C. in poi)”, in L’impatto della “peste antonina”, ed. E. Lo Cascio (Bari 2012) 123165.

    • Search Google Scholar
    • Export Citation
  • Bruun C. (2007) “The Antonine plague and the ‘third-century crisis’”, in Crises and the Roman Empire: Proceedings of the Seventh Workshop of the International Network Impact of Empire, Nijmegen, June 20–24, 2006, edd. O. Hekster, G. de Kleijn and D. Slootjes (Leiden 2007) 201–19.

  • Bruun C. (2003) “The Antonine plague in Rome and Ostia”, JRA 16 (2003) 426434.

  • Büntgen U. et al. (2016) “Cooling and societal change during the Late Antique Little Ice Age from 536 to around 660 AD”, Nature Geoscience 9 (2016) 231236.

    • Search Google Scholar
    • Export Citation
  • Campbell B. (2016) The Great Transition: Climate, Disease and Society in the Late-Medieval World (Cambridge 2016).

  • Carmichael A. (1986) Plague and the Poor in Renaissance Florence (Cambridge 1986).

  • Chabot J.-B. (1927) Incerti auctoris chronicon anonymum Pseudo-Dionysianum vulgo dictum 1 (Corpus scriptorum Christianorum orientalium 91) (Paris 1927).

    • Search Google Scholar
    • Export Citation
  • Clark G. (2007) A Farewell to Alms: a Brief Economic History of the World (Princeton Economic History of the Western World) (Princeton 2007).

    • Search Google Scholar
    • Export Citation
  • Comas I. et al. (2013) “Out-of-Africa migration and Neolithic coexpansion of mycobacterium tuberculosis with modern humans”, Nature Genetics 45 (2013) 11761182.

    • Search Google Scholar
    • Export Citation
  • Conrad L. I. (1981) The Plague in the Early Medieval Near East (Ph.D. diss., Univ. of Princeton 1981).

  • Crosby A. W. (1986) Ecological Imperialism: the Biological Expansion of Europe, 900–1900 (Cambridge 1986).

  • Cui Y. et al. (2013) “Historical variations in mutation rate in an epidemic pathogen, Yersinia pestis”, Proceedings of the National Academy of Sciences 110 (2013) 577–82.

  • Debié M. (2015) L’écriture de l’histoire en syriaque: transmissions interculturelles et constructions identitaires entre hellénisme et islam (Leuven 2015).

    • Search Google Scholar
    • Export Citation
  • Diamond J. (1997) Guns, Germs, and Steel: the Fates of Human Societies (New York 1997).

  • Duggan A. et al. (2016) “17th century variola virus reveals the recent history of smallpox”, Current Biology 26 (2016) 34073412.

    • Search Google Scholar
    • Export Citation
  • Duncan-Jones R. P. (1996) “The impact of the Antonine plague”, JRA 9 (1996) 108193.

  • Durliat J. (1989) “La peste du VIe siècle. Pour un nouvel examen des sources byzantines”, in Hommes et richesses dans l’empire byzantin, I: IVe–VIIe siècle, edd. J. Lefort and J. Morrisson (Paris 1989) 107119.

    • Search Google Scholar
    • Export Citation
  • Eddy J. J. (2015) “The ancient city of Rome, its empire, and the spread of tuberculosis in Europe”, Tuberculosis 95 (2015) 2328.

    • Search Google Scholar
    • Export Citation
  • Eisen R. and Gage K. (2009) “Adaptive strategies of Yersinia pestis to persist during inter-epizootic and epizootic periods”, Veterinary Research 40 (2009) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Elliott, C. (2016) “The Antonine plague, climate change and local violence in Roman Egypt”, PastPres 231 (2016) 331.

  • Enscore R. E. et al. (2002) “Modeling relationships between climate and the frequency of human plague cases in the southwestern United States, 1960–1997”, American Journal of Tropical Medicine and Hygiene 2 (2002) 186196.

    • Search Google Scholar
    • Export Citation
  • Feldman M. et al. (2016) “A high-coverage Yersinia pestis genome from a 6th-century Justinianic plague victim”, Molecular Biology and Evolution 33 (2016) 29112923.

    • Search Google Scholar
    • Export Citation
  • Fenner F. (1988) Smallpox and its Eradication (Geneva 1988).

  • Fouquet R. and Broadberry S. (2015) “Seven centuries of european economic growth and decline”, Journal of Economic Perspectives 29 (2015) 227244.

    • Search Google Scholar
    • Export Citation
  • Gage K. and Kosoy M. (2005) “Natural history of plague: perspectives from more than a century of research”, Annual Review of Entomology 50 (2005) 505528.

    • Search Google Scholar
    • Export Citation
  • Gage K. et al. (2008) “Climate and vectorborne diseases,” American Journal of Preventive Medicine 35 (2008) 436450.

  • Garnsey P. (1988) Famine and Food Supply in the Graeco-Roman World: Responses to Risk and Crisis (Cambridge 1988).

  • Garrett L. (1994) The Coming Plague: Newly Emerging Diseases in a World Out of Balance (New York 1994).

  • Gourevitch D. (2005) “The Galenic plague: a breakdown of the imperial pathocoenosis and longue durée”, History and Philosophy of the Life Sciences 27 (2005) 5769.

    • Search Google Scholar
    • Export Citation
  • Green M. (2014) “Taking ‘pandemic’ seriously: making the Black Death global”, The Medieval Globe 1 (2014) 2761.

  • Harbeck M. et al. (2013) “Yersinia pestis DNA from skeletal remains from the 6th century AD reveals insights into Justinianic plague”, PLoS Pathogens 9 (2013) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Harkins K. M. and Stone A. C. (2015) “Ancient pathogen genomics: insights into timing and adaptation”, Journal of Human Evolution 79 (2015) 137149.

    • Search Google Scholar
    • Export Citation
  • Harper K. (2017) The Fate of Rome: Climate, Disease, and the End of an Empire (Princeton 2017).

  • Harper K. (2016a) “People, plagues, and prices in the Roman world: the evidence from Egypt”, Journal of Economic History 76 (2016) 803839.

    • Search Google Scholar
    • Export Citation
  • Harper K. (2016b) “Another eye-witness to the plague described by Cyprian and notes on the ‘persecution of Decius’”, JRA 29 (2016) 473476.

    • Search Google Scholar
    • Export Citation
  • Harper K. (2015a) “A time to die: preliminary notes on seasonal mortality in late antique Rome”, in Children and Family in Late Antiquity. Life, Death and Interaction, edd. C. Laes, K. Mustakallio and V. Vuolanto (Leuven 2015) 1534.

    • Search Google Scholar
    • Export Citation
  • Harper K. (2015b) “Pandemics and passages to Late Antiquity: rethinking the plague of c. 249–270 described by Cyprian”, JRA 28 (2015) 223260.

    • Search Google Scholar
    • Export Citation
  • Hatcher J. (2003) “Understanding the population history of England 1450–1750”, PastPres 180 (2003) 83130.

  • Hin S. (2013) The Demography of Roman Italy: Population Dynamics in an Ancient Conquest Society, 201 BCE–14 CE (Cambridge 2013).

  • Horden P. (2005) “Mediterranean plague in the age of Justinian”, in The Cambridge Companion to the Age of Justinian, ed. M. Maas (Cambridge 2005) 134160.

    • Search Google Scholar
    • Export Citation
  • Jones C. P. (2006) “Addendum to JRA 18 (2005): Cosa and the Antonine plague?”, JRA 19 (2006) 368369.

  • Jones C. P. (2005) “Ten dedications ‘to the gods and goddesses’ and the Antonine plague”, JRA 18 (2005) 293301.

  • Kausrud K. et al. (2010) “Modeling the epidemiological history of plague in central Asia: palaeoclimatic forcing on a disease system over the past millennium”, BioMed Central Biology 8 (2010) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Klein Goldewijk K., Beusen A. and Janssen P. (2010) “Long-term dynamic modeling of global population and built-up area in a spatially explicit way: HYDE 3.1”, The Holocene 20 (2010) 565573.

    • Search Google Scholar
    • Export Citation
  • Kostick C. and Ludlow F. (2015) “The dating of volcanic events and their impact upon European society, 400–800 CE”, Post-Classical Archaeologies 5 (2015) 730.

    • Search Google Scholar
    • Export Citation
  • Landers J. (1993) Death and the Metropolis: Studies in the Demographic History of London, 1670–1830 (Cambridge 1993).

  • Le Roy Ladurie E. (1973) “Un concept: l’unification microbienne du monde (XIVe–XVIIe siècles)”, ZSchwG 23 (1973) 627696.

  • Little L. (2007) ed. Plague and the End of Antiquity: the Pandemic of 541–750 (New York 2007).

  • Littman R. J. and Littman M. L. (1973) “Galen and the Antonine plague”, AJP 94 (1973) 243255.

  • Livi-Bacci M. (2012) A Concise History of World Population (Oxford, 5th edn. 2012).

  • Lo Cascio E. (2012) ed. L’impatto della “peste antonina” (Bari 2012).

  • Locey K. and Lennon J. (2016) “Scaling laws predict global microbial diversity”, Proceedings of the National Academy of Sciences 113 (2016) 5970–75.

  • Loy D. E. et al. (2017) “Out of Africa: origins and evolution of the human malaria parasites Plasmodium falciparum and Plasmodium vivax”, International Journal for Parasitology 47 (2017) 8797.

    • Search Google Scholar
    • Export Citation
  • Maddison A. (2001) The World Economy: a Millennial Perspective (Paris 2001).

  • Manning S. W. (2013) “The Roman world and climate: context, relevance of climate change, and some issues”, in The Ancient Mediterranean Environment between Science and History, ed. W. V. Harris (Leiden 2013) 103170.

    • Search Google Scholar
    • Export Citation
  • Marciniak S. et al. (2016) “Plasmodium falciparum malaria in 1st–2nd century CE southern Italy”, Current Biology 26 (2016) R1205–1225. Doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Marcone A. (2002) “La peste antonina: testimonianze e interpretazioni”, RivStorIt 114 (2002) 803819.

  • McCormick M. (2016) “Tracking mass death during the fall of Rome’s empire (II)”, JRA 29 (2016) 10081046.

  • McCormick M. (2015) “Tracking mass death during the fall of Rome’s empire (I)”, JRA 28 (2015) 325357.

  • McCormick M. (2003) “Rats, communications, and plague: toward an ecological history”, Journal of Interdisciplinary History 34 (2003) 125.

    • Search Google Scholar
    • Export Citation
  • McCormick M. (1998) “Bateaux de vie, bateaux de mort. Maladie, commerce, transports annonaires et le passage économique du Bas-Empire au moyen âge”, in Morfologie sociali e culturali in Europa fra tarda Antichità e alto Medioevo. 3–9 aprile 1997, vol. 1 (Settimane di studio del Centro italiano di studi sull’alto Medioevo 45) (Spoleto 1998) 35118.

    • Search Google Scholar
    • Export Citation
  • McCormick M. et al. (2012) “Climate change during and after the Roman empire: reconstructing the past from scientific and historical evidence”, Journal of Interdisciplinary History 43 (2012) 169220.

    • Search Google Scholar
    • Export Citation
  • McNally A. et al. (2016) “‘Add, stir and reduce’: Yersinia spp. as model bacteria for pathogen evolution”, Nature Reviews Microbiology 14 (2016) 177190.

    • Search Google Scholar
    • Export Citation
  • McNeill J. R. (2015) “Energy, population, and environmental change since 1750: entering the Anthropocene”, in The Cambridge World History, vol. 7, Production, Destruction, and Connection, 1750-Present, Part 1: Structures, Spaces, and Boundary Making, edd. J. R. McNeill, K. Pomeranz (Cambridge 2015) 5182.

    • Search Google Scholar
    • Export Citation
  • McNeill J. R. (2010) Mosquito Empires: Ecology and War in the Greater Caribbean, 1620–1914 (New York 2010).

  • McNeill J., Pádua J. and Rangarajan M. (2015) edd. Environmental History: As if Nature Existed (New Delhi 2015).

  • McNeill W. H. (1976) Plagues and Peoples (New York 1976).

  • Meier M. (2016) “The ‘Justinianic plague’: the economic consequences of the pandemic in the eastern Roman empire and its cultural and religious effects”, Early Medieval Europe 24 (2016) 267292.

    • Search Google Scholar
    • Export Citation
  • Meier M. (2003) Das andere Zeitalter Justinians: Kontingenzerfahrung und Kontingenzbewältigung im 6. Jahrhundert n. Chr. (Göttingen 2003).

    • Search Google Scholar
    • Export Citation
  • Mitchell S. (2015) A History of the Later Roman Empire, AD 284–641 (Malden, 2nd edn. 2015).

  • Morris I. (2013) The Measure of Civilization: How Social Development Decides the Fate of Nations (Princeton 2013).

  • Morris I. (2010) Why the West Rules—For Now: the Patterns of History, and What They Reveal About the Future (London 2010).

  • Müller R. et al. (2014) “Genotyping of ancient mycobacterium tuberculosis strains reveals historic genetic diversity”, Proceedings of the Royal Society B: Biological Sciences 281 (2014) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Newfield T. (2017) “Malaria and malaria-like disease in the Early Middle Ages”, Early Medieval Europe 25 (2017) 251300.

  • Prowse T. et al. (2008) “Isotopic and dental evidence for infant and young child feeding practices in an imperial Roman skeletal sample”, American Journal of Physical Anthropology 137 (2008) 294308.

    • Search Google Scholar
    • Export Citation
  • Rasmussen S. et al. (2015) “Early divergent strains of Yersinia pestis in Eurasia 5,000 years ago”, Cell 163 (2015) 571582.

  • Rathbone D. W. and von Reden S. (2015) “Mediterranean grain prices in classical antiquity”, in A History of Market Performance: From Ancient Babylonia to the Modern World, edd. R. J. Van der Spek, J. Luiten and B. van Zanden (London 2015) 149235.

    • Search Google Scholar
    • Export Citation
  • Roberts C. A. (2015) “Old World Tuberculosis: evidence from human remains with a review of current research and future prospects”, Tuberculosis 95 (2015) 117121.

    • Search Google Scholar
    • Export Citation
  • Roberts C. A. and Buikstra J. (2003) The Bioarchaeology of Tuberculosis: a Global View on a Reemerging Disease (Gainesville, Florida 2003).

    • Search Google Scholar
    • Export Citation
  • Robin C. (1992) “Guerre et épidémie dans les royaumes d’Arabie du Sud, d’après une inscription datée (IIe s. de l’ère chrétienne)”, CRAI 136 (1992) 215234.

    • Search Google Scholar
    • Export Citation
  • Sallares R. (2002) Malaria and Rome: a History of Malaria in Ancient Italy (Oxford 2002).

  • Sarris P. (2002) “The Justinianic plague: origins and effects”, Continuity and Change 17 (2002) 169182.

  • Scheidel W. (2010) “Real wages in early economies: evidence for living standards from 1800 BCE to 1300 CE”, JESHO 53 (2010) 425462.

    • Search Google Scholar
    • Export Citation
  • Scheidel W. (2001) Death on the Nile: Disease and the Demography of Roman Egypt (Leiden 2001).

  • Scheidel W. (1996) Measuring Sex, Age and Death in the Roman Empire (Ann Arbor 1996).

  • Sender R., Fuchs S. and Milo R. (2016) “Revised estimates for the number of human and bacteria cells in the body”, PLOS Biology 14.8 (2016) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Shaw B. (1996) “Seasons of death: aspects of mortality in imperial Rome”, JRS 86 (1996) 100138.

  • Sigl M. et al. (2015) “Timing and climate forcing of volcanic eruptions for the past 2,500 years”, Nature 523 (2015) 543562.

  • Slack P. (2012) Plague: a Very Short Introduction (Oxford 2012).

  • Stathakopoulos D. (2004) Famine and Pestilence in the Late Roman and Early Byzantine Empire: a Systematic Survey of Subsistence Crises and Epidemics (Aldershot-Burlington, Vermont 2004).

    • Search Google Scholar
    • Export Citation
  • Stathakopoulos D. (2000) “The Justinianic plague revisited”, Byzantine and Modern Greek Studies 24 (2000) 256276.

  • Stone A. C. et al. (2009) “Tuberculosis and leprosy in perspective”, Yearbook of Physical Anthropology 52 (2009) 6694.

  • Toohey M. et al. (2016) “Climatic and societal impacts of a volcanic double event at the dawn of the Middle Ages”, Climatic Change 136 (2016) 401412.

    • Search Google Scholar
    • Export Citation
  • Trombley F. and Watt J. (2000) edd. and transl. The Chronicle of Pseudo-Joshua the Stylite (Translated Texts for Historians 32) (Liverpool 2000).

    • Search Google Scholar
    • Export Citation
  • Wagner D. M. et al. (2014) “Yersinia pestis and the plague of Justinian 541–543 AD: a genomic analysis”, Lancet Infectious Diseases 14 (2014) 319326.

    • Search Google Scholar
    • Export Citation
  • Wickham C. (2016) Medieval Europe (New Haven, Connecticut 2016).

  • Wickham C. (2005) Framing the Early Middle Ages: Europe and the Mediterranean 400–800 (Oxford 2005).

  • Wiechmann I. and Grupe G. (2005) “Detection of Yersinia pestis DNA in two early medieval skeletal finds from Aschheim (Upper Bavaria, 6th century AD)”, American Journal of Physical Anthropology 126 (2005) 4855.

    • Search Google Scholar
    • Export Citation
  • Wiemer H.-U. (2006) “Kaiser und Katastrophe. Zur Bewältigung von Versorgungskrisen im spätrömischen Reich”, in Staatlichkeit und politisches Handeln in der römischen Kaiserzeit, ed. H.-U. Wiemer (Berlin-New York 2006) 249281.

    • Search Google Scholar
    • Export Citation
  • Woolhouse M. and Gaunt E. (2007) “Ecological origins of novel human pathogens”, Critical Reviews in Microbiology 33 (2007) 231242.

    • Search Google Scholar
    • Export Citation
  • Xu L. et al. (2015) “The trophic responses of two different rodent—vector—plague systems to climate change”, Proceedings of the Royal Society of London B: Biological Sciences 282 (2015) doi: (last accessed March 2017).

    • Search Google Scholar
    • Export Citation
  • Xu L. et al. (2014) “Wet climate and transportation routes accelerate spread of human plague”, Proceedings of the Royal Society of London B: Biological Sciences 281 (2014) doi: (last accessed March 2017).

  • Zelener Y. (2003) Smallpox and the Disintegration of the Roman Economy after 165 AD (Ph.D. diss., Univ. of Columbia 2003).

  • Zhang Z. et al. (2007) “Relationship between increase rate of human plague in China and global climate index as revealed by cross-spectral and cross-wavelet analyses”, Integrative Zoology 2 (2007) 144153.

    • Search Google Scholar
    • Export Citation

The text is edited by Chabot (1927) 235–317. On the text, see Debié (2015) 10–14; Wiemer (2006); Brock (1979–80) 11. A modern English translation is available: Trombley and Watt (2000).


Pseudo-Joshua, Chron. 26 (edd. and transl. Trombley and Watt (2000) 23).


Stathakouplos (2004) 248–49, on the outbreak, identifying it as smallpox. I have offered further reflections on this source and others for smallpox at: (last accessed March 2017).


Pseudo-Joshua, Chron. 38–44.


Garnsey (1988) 3–6.


See Green (2014), for an example of consilience in practice.


See e.g. Harkins and Stone (2015).


See the helpful discussion of Landers (1993).


McNeill (2015).


Clark (2007) 32: “in the preindustrial world sporadic technological advance produced people, not wealth.”


Maddison (2001). Few studies venture earlier than ca. AD 1000, but see Scheidel (2010) and more generally Morris (2013) and (2010). For a recent conspectus of wage divergence in the long run, see Fouquet and Broadberry (2015).


McNeill (2015); Livi-Bacci (2012); Klein Goldewijk, Beusen, and Janssen (2010).


See e.g. Brooke (2014); Hatcher (2003); Landers (1993).


McNeill (1976). Anticipated by some of the work of the Annales school, e.g. Le Roy Ladurie (1973).


McNeill (1976); Crosby (1986); Diamond (1997).


Garrett (1994) 30–52, is a brilliant overview of this triumphalist moment in global public health.


Harkins and Stone (2015); Barrett and Armelagos (2013).


TB: Bos et al. (2014); Comas et al. (2013). Malaria: Loy et al. (2017). Smallpox: Duggan et al. (2016). Plague: Rasmussen et al. (2015); Cui et al. (2013).


Locey and Lennon (2016).


Sender, Fuchs, and Milo (2016).


Woolhouse and Gaunt (2007).


The classic study of smallpox is Fenner (1988).


Benedictow (2004).


McNeill (2010).


E.g. Carmichael (1986).


Harper (2015a); Scheidel (2001) and (1996); Shaw (1996).


Harper (2015a).


See e.g. Prowse et al. (2008) for important bioarchaeological evidence.


Sallares (2002).


Newfield (2017); Newfield, in this volume.


Loy et al. (2017).


Gal., De morborum temporibus 7.435K.


Marciniak (2016).


Pseudo-Aristotle, Problemata 14.7.909 (transl. in Sallares (2002) 2820).


Harper (2017) chapter 1; Hin (2013); Manning (2013); McCormick et al. (2012).


Gal., De Temperamentis 1.4.531 (transl. Singer (1997) 212).


Sallares (2002) 229.


Landers (1993).


See McNally et al. (2016), for the role of horizontal gene transfer in the specific case of Yersinia.


Overviews of TB: Roberts (2015); Müller et al. (2014); Roberts and Buikstra (2003).


TB genome: Achtman (2016); Bos et al. (2014); Comas et al. (2013); Stone et al. (2009).


Eddy (2015).


A select bibliography would include Elliott (2016); Lo Cascio (2012); Bruun (2012), (2007) and (2003); Jones (2006) and (2005); Gourevitch (2005); Zelener (2003); Marcone (2002); Duncan-Jones (1996); Littman and Littman (1973). The overview presented in brief here is covered more extensively in Harper (2017).


SHA, Ant. Pius 9.4. Arabic sources: Robin (1992).


Collected in Harper (2017).


Gourevitch (2005); Littman and Littman (1973).


Babkin and Babkina (2015).


Duggan et al. (2016).


Harper (2017); Elliott (2016).


Harper (2016a).


The pestilence in Edessa is also of particular importance potentially, since it provides strong observational evidence for epidemic smallpox in the 5th c. Medical sources from the 7th to 9th c.—from Alexandria to Iraq, Persia and India—demonstrate familiarity with an endemic form of smallpox. So, what can be said securely, is that going forward, the combination of traditional and novel forms of evidence will be necessary to fill out the history of smallpox, a virus whose evolutionary history is recent and eventful.


Harper (2017); (2016b); (2015b).


Euseb., Hist. eccl. 7.21.


Rathbone and von Reden (2015) 184.


Harper (2017) chapter 3.


Procop., Pers. 2.10.4 (transl. Kaldellis (2014) 93).


Meier (2003).


All of what follows is treated more extensively in Harper (2017) chapters 5 and 6.


Arjava (2005) on the written evidence.


Sigl et al. (2015).


Toohey et al. (2016); Kostick and Ludlow (2015); Büntgen et al. (2016).


Harper (2017) chapter 6.


Slack (2012).


The essays in Little (2007) represent the state of the field. Major treatments include Meier (2016); Mitchell (2015) 409–13, 479–91; Horden (2005); Meier (2003); Stathakopoulos (2004); Sarris (2002); Stathakopoulos (2000); Conrad (1981); Durliat (1989); Allen (1979); Biraben (1975) 22–48; Biraben and Le Goff (1969).


Cui et al. (2013).


Eisen and Gage (2009); Gage and Kosoy (2005).


Of both mammals and fleas. See Campbell (2016) esp. 232–33 for the medieval Black Death.


McCormick (2003).


Feldman et al. (2016); Wagner et al. (2014); Harbeck et al. (2013); Wiechmann and Grupe (2005).


Green (2014) 37.


Specifically, the late development of the ability to make Ymt, a protein that lets the bacterium survive in the mid-gut of the flea, as well as an amino acid substitution in the pla protein that heightens the virulence of the pathogen.


Plague and climate generally: Ari et al. (2011); Kausrud et al. (2010); Gage et al. (2008). El Niño: Zhang et al. (2007). Xu et al. (2015) and (2014); Enscore et al. (2002).


A point I emphasise in Harper (2017) chapter 5.


McCormick (2003).


Harper (2017). See Campbell (2016) and Benedictow (2004) for the Black Death.


Benedictow (2004) is clear, reliable, and helpful.


Harper (2017) for a fuller account.


Meier (2003) 92–93 offers a slightly different dating, with the outbreak arriving in the capital by AD 541, but I find McCormick (1998) more convincing, and think it is far preferable to follow Procopius.


Procop., Pers. 2.23.2 (transl. Kaldellis (2014) 123).


For expressions of minimalist views, see Wickham (2016) 43–44 and (2005), building on Durliat (1989), which is still the most important version of the sceptical case, although I think it is deeply flawed. For a recent presentation of the maximalist case, Meier (2016) is convincing.


McCormick (2016) and (2015).


Aschheim: Wagner et al. (2014); Harbeck et al. (2013); Wiechmann and Grupe (2005). Altenerding: Feldman et al. (2016).


See Harper (2017).


Harper (2017).


See Conrad (1981).


McNeill, Pádua, and Rangarajan (2015).


All Time Past Year Past 30 Days
Abstract Views 67 0 0
Full Text Views 1253 764 57
PDF Views & Downloads 1099 731 64