Plague and Famine in Late Antiquity and Byzantium

Tomorrow I’m presenting in Prof. Sercan Yandim Aydin and Prof. Luca Zavagno’s Byzantium at Ankara seminar series in a session titled “Famine and Plagues in Byzantium: archaeology, documentary and hagiography in a comparative perspective.”

I have to admit to feeling more than a little nervous about the topic which is pretty far from my core area of expertise (however narrow that might be). I’ve spend the last few days reviewing some of the key works on the topic of plague and famine, and I have to admit that it’s been a nice break from my other simmering projects.

While I’m not planning to present a formal paper, it’s useful for me to get my ideas together. For the following observations, I’m indebted to the crowd-sourced “Archaeology of Epidemics” syllabus.

It seems to me that the archaeology of plagues and famines recognizes the long-standing ties between disease and sedentary agriculture. The latter tends to be a precondition for  historical understandings of famine (although famines are, of course, possible among hunter-gatherers, many of the preconditions for famine appear to be more prominent among settled agrarians than more mobile hunter-gatherers). More than that, settled agriculture increased human proximity to animals, to one another, and to human waste and distinct environmental conditions which undoubtedly contributed to an increase rates of infectious diseases. Settled agriculture is obviously case in the Later Roman and Byzantine Empires as were a series of epidemics (that verge on being pandemics) starting in the late 2nd century and continuing through the middle-8th century.

The second basic idea for any consideration of plagues and famines in the past is that both phenomena are incredibly complex and generally the result of multiple variable. Diseases, for example, can vary not only on the biological level. Yeresinia Pestis, for example, is a bacteria transmitted primarily by rats and humans; Malaria, as another example, is an amoeba transmitted by a limited number of types of mosquitos. Cholera and Typhoid are primarily transmitted through contaminated water (and food). The range of different kinds of diseases, therefore, impacts the ways in which diseases spread, take hold, and impact mortality.

Famines likewise represent a series of conditions that range from weather and climate to economic and political decisions. Famine often results in malnutrition and compromised immune systems that can produce not only a greater susceptibility to  The relationship between famine and disease then requires that we consider not only politics, ecology, economics, and environmental conditions, but also the distinctive character of the diseases moving through the human population.

The archaeology of plague and famine in the Late Roman and Early Byzantine Mediterranean has its own unique characteristics as well. There are four major trends, I think, in how archaeologists have approached these phenomena.

1. Bioarchaeology. Certainly the most sensational efforts to understand disease in the ancient world have come from bioarchaeologists. The publication of evidence for Yersinia Pestis (the Bubonic Plague) in dental pulp DNA samples from two 6th century cemeteries in Bavaria has added weight to hypothesis that the Justinianic plague was, indeed, the Bubonic plague, although as a few commentators have noted, 10 individuals with the plague in a rather remote region hardly represents a meaningful sample of the Mediterranean population at the time.

The challenges facing bioarchaeology involve not only the still-developing technologies necessary to analyze human skeletal remains at the scale necessary to produce a sufficiently significant body of evidence to allow for large scale conclusions. More than that, only certain kinds of diseases leave recognizable traces in human remains. Tuberculosis, for example, leaves tell-tale lesions on bones, but other illnesses like malaria are more elusive meaning that our view of epidemics illnesses in the ancient world will likely remain uneven for the time.

2. Climate and Environmental Archaeology. The second major trend in understanding plagues and famines in Late Antiquity is the growing interest in ancient climate change and the role of climate and the environment in creating conditions favorable with the development of epidemic and pandemic outbreaks of disease. The most thorough version of this approach is Kyle Harper’s 2017 book  The Fate of Rome: Climate, Disease and the End of an Empire (Princeton 2017)  or the recent volume of the Late Antique Archaeology series dedicated to Environment and Society in the Long Late Antiquity (Brill 2018).

Harper and others have sought to understand the appearance of pandemic scale diseases in the late 2nd century, starting with the Antonine plague, and continuing through to the Justinian’s plague. Harper argues that end of the Roman Climate Optimum and the start of the Late Antique Ice Age (and the Late Antique Little Ice Age) may have created environmental conditions suitable for spread of new diseases in the Mediterranean basin. Harper and other scholars who have emphasized that large scale climate change is not an explanation in and of itself, but the impact of these changes must be understood in the context of both local environmental conditions and the larger political and social world of Late Antiquity. There is little doubt however that changes in global climate triggered changes in the agricultural regimes that contributed to “years without summer” and subsequent famines that are known from literary sources. The wide ranging debate around the “The Mystery Cloud of 536 CE” and its connection to the Justinianic plague is only one such example.

As any number of scholars have shown in recent years, for example the work done by Haldon, Elton and Newhart have done in Anatolia, regional variation in the environment plays a significant role in understand the impact of climate change at the local level. Sturt Manning has recently made clear the significant inter-annual variations in rain fall on Cyprus, which may have had as much of an impact on the health and prosperity of communities than large scale climate variation on the global level. Roman efforts to drain swamps, build roads, and deforest hills would have introduced new hydrological patterns that would have supported, for example, the mosquitos that carried malaria, as Robert Sallares has shown in his work. Bret Shaw has argued on the basis of epigraphic data from across the Mediterranean that there are clear seasonal patterns of mortality. The wide range of variables that contribute to these patterns include local weather during the hotter and more humid late summer months whose impact on the mortality of diseases like malaria and tuberculosis remains difficult to assess.

Environmental archaeology has also shed valuable light on global conditions that could have disrupted plague foci in, for example, rodent colonies in Central Asia or East Africa which would have dispatched plague bearing fleas to the Mediterranean through established trade routes in the Indian Ocean and Red Sea. Understanding these phenomena, of course, is extremely complex and we’re only at the very beginning of understand the existence of disease foci and their relationship to climate conditions. In fact, both of these things as independent variables remain difficult to understand much less in relation to one another.

That being said, a growing body of environmental proxies from Greenland ice cores to dendrochronology is beginning to allow us to map global climate change in antiquity. Unfortunately the chronological resolution and distribution of this data does not always coincide with the kinds of historical questions that Mediterranean archaeologists and historians are asking. Moreover, they do not always make the cause of climate variation clear and are often difficult to correlate with local conditions. This isn’t to say that this work is not producing meaningful results, but that the impact of climate science on our understanding of specific historical events like the Justinianic plaque continues to develop.

3. Archaeological Evidence. Michael McCormick, one of the leading scholars in the study of plagues in antiquity, has pulled together the evidence for mass burials in the Late Roman and Early Byzantine period and published it in the Journal of Roman Archaeology. McCormick recognizes that these kinds of direct evidence for episodes of mass mortality may well provide indications of plague events and coincide with evidence from literary sources for mass burials.

The challenge, of course, is that the situation and dates of many of these mass burials remain unclear. In some cases, such as Kopetra on Cyprus, where a cistern becomes the tomb for at least 9 individuals, there is reason to at least suspect that these are people who died from disease. In other cases, like the Andritsa cave in Argolid, it’s more challenging to see the bodies of over 50 individuals as victims of the disease alone and might be better recognized as the result of a complex series of events that range from diseases and famine to regional political or military disruptions. As with efforts to understand the relationship between climate and disease, it seems likely that many of these burials represent the intersection of a series of conditions including the availability of a cave or a cistern for a mass burial. The analysis of the demographics of the individuals, when such data is available, and any chronological clustering of the incidents of mass burial likewise allow for a more refined interpretation of these events. McCormick shows that of the 48 mass burials datable to between 300 and 800, 36 of them occurred during the 6th and 7th century. That being said, the ongoing efforts to refine the dating of Late Roman ceramics, for example, may well complicate the chronology of these burials or at least complicate any clear correlation between events datable in the literary sources to a particular year or span and shifting archaeological chronologies.

4. Landscape, Settlement, and Demography. Finally, there has been a massive amount of attention on Late Roman settlement and demography over the last twenty years prompted in part by the rise in landscape and regional survey projects that have shed valuable light not only on rural settlement, but also on the relationship between city and country and between various regions in the Mediterranean basin.

On the most basic level questions of demography and settlement patterns provide a backdrop for the economic conditions present in the long Late Antiquity. As our understanding of the countryside, particularly in the Eastern Mediterranean, improves and reveals a “busy countryside” in the 5th, 6th, and 7th centuries, traditional arguments for large scale demographic and economic decline have become less compelling. The neighborhood of the mass burial at Kopetra and the Andritsa cave in the Western Argolid appear to be active and even moderately prosperous with imported table wares, storage and transport amphora, and purpose made cooking pots appearing in urban and rural areas alike. Of course, this is not meant to suggest that plagues cannot strike economically prosperous communities or that they necessarily would have a negative impact on local economies, but to challenge views of the Late Antique landscape which saw conditions of decline and deprivation as particularly suitable for the outbreak of disease.

Instead, recent work has emphasized the ongoing connectivity of the Mediterranean in Late Antiquity and this connectivity goes well beyond well-known links between major urban centers or regions and now involves the myriad of small links that constituent the dense networks of small-scale and local exchange between microregions. These connections provided the Late Antique world with its remarkable resilience in times of political or environmental stress, and, at the same time, established the context for the spread of disease throughout the Later Roman world.

At present, our efforts to understand the character and extent of this connectivity in a nuanced way remains in its infancy. For example, we’re only now making strides to understand the countryside in the neighborhood of Adritsa Cave which in many ways shows remarkable continuity with earlier centuries. This makes it difficult to present even a local or regional context for this unusual site much less ascertain its place within larger regional trends.


If my contribution tomorrow does anything, I hope that I’m able to present a view of the archaeology of plagues and famines that provides some disciplinary context. This involves not only understanding the larger methodological trends that characterize research on these topics including bioarchaeology, environmental archaeology, excavations, and survey projects, but also emphasizes the epistemological and practical limits that shape archaeological discourse.

It’s cliche to note that the goal of archaeology is not necessarily to supplement the narratives established by historians and scholars of texts. In fact, archaeological evidence often remains oddly incompatible with these disciplines. Often the incommensurability stems from the different scales at which we operate. Even with the precisions of dendrochronology, most archaeological dates established through C14 dating or conventional ceramic typologies are more broad and imprecise than dates provided in texts.

As significantly, archaeologists sometimes work at the highly focused scale of a single site, part of a site, or trench, where evidence for phenomena like plagues may not be immediately visible. This situation means that even when archaeologists work at the scale of the region, we’re often forced to recognize that our inconsistent sample is better suited to understanding intraregional variation especially when considered at finer chronological resolutions.

Advances in bioarchaeological sampling are unlikely to resolve these issues quickly because of variability in discovery and preservation of human remains, the need to secure samples through particularly rigorous and time consuming excavation techniques that are not always possible, and the cost of analysis. The same holds true for many of the techniques that are producing important insights into regional environmental conditions that require specific situations (e.g. anaerobic lake beds), expertise, and funding that not all projects can access in the same way. The resulting patchworks, like the 10 individuals with the plague in Germany, will result in datasets that seemingly beg to be over generalized or dismissed as outliers especially in their relationship to particular environmental or epidemiological events.

Where archaeology shines is in providing evidence for long-term trends at the regional level and recognizing the ebb and flow of populations, prosperity, connectivity, and settlement. Plagues and famines while seemingly more common between the 3rd and 8th century often emerge as mere blips on the archaeological radar and soon disappear again the backdrop of persistent activity and the resilience of long-standing communities.

While this will likely disappoint historians who persist in their hope that archaeology or archaeological science can unlock the relationship between climate, disease, populations, economies, and politics, maybe it’ll be heartening in our contemporary situation where the COVIDs seem destined the fundamentally transform our everyday life. Viewed at the archaeological scale, it may be that The COVID pandemic will appear as little more than some discarded latex gloves and empty bottles of hand sanitizer.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s