plague of Justinian
plague of Justinian
- Peregrine Horden
The Nature of the Pandemic
“An evil destiny of bubo and armpit” (CIG 8628), the plague of Justinian is the name given—unfairly, since the emperor did not cause it, himself contracted it, and was long outlasted by it—to the pandemic of “bubonic plague,” infection by Yersinia pestis, that struck western Eurasia and North Africa towards the middle of the 6th century ce and that recurred in phases until at least the middle of the 8th century.1 In geographical extent, demographic and social impact, and chronology, it probably far surpassed the major epidemics of the Roman imperial and late antique period, the 2nd-century Antonine plague and the mid-3rd-century plague described by Cyprian of Carthage. Study of the Justinianic pandemic has been transformed in the 21st century not only by ever more sophisticated exploitation of the usual range of historical sources, literary and material, but by the recovery of the DNA and the reconstruction of several of the genomes of the pathogen from the skeletons of some of its victims. And yet almost every aspect of the pandemic remains debated. There is no consensus on its route into the world of Justinian, its subsequent epidemiology, virulence, or macro-historical consequences.
The pandemic first visibly irrupted in the summer of 541 ce in the Egyptian port of Pelusium. It spread rapidly both east and west to Alexandria and other parts of Egypt as well as to Palestine. By the spring of 542, it had reached Constantinople, where Justinian contracted it but recovered. In the same year, it spread to Syria, the western Balkans, Tunisia, Sicily, and Spain. By 543, it had touched Armenia (where, reportedly, the invading Persians first encountered it) as well as Italy and Gaul. Soon it would reach England and, probably, Ireland. It has also been detected in (modern) Germany. Claims that it extended as far north as Scandinavia rest, however, on very ambiguous archaeological evidence. On the other hand, and in the other direction, a late antique presence in Nubia and Ethiopia as well as in sub-Saharan East Africa cannot be ruled out. In the Arabian Peninsula, Yemen was afflicted but not the Hejaz: this was a disease more of settled populations, both urban and rural, than, in the Middle East, of nomadic societies.
Plague was thus in evidence for about two centuries, disappearing mysteriously and suddenly around 750. But it was not in evidence everywhere at any one time, and several contemporary observers remarked on the patchiness of its severity. Nor was it in evidence continuously. Using written sources, historians have come to see it as recurring in about seventeen “amplifications” (from some relatively nearby reservoir) after the initial onslaught, with varying intervals between them. Some of the component epidemics of these are mentioned in laconic texts hard to interpret or to date. Localized outbreaks of other diseases may have been wrongly recruited as examples of a much bigger plague. Western written sources are often too meagre to allow anything like a full picture. Moreover, it now appears that plague was present in areas for which no epidemics are reported. Still, the model of distinct phases seems broadly indicative of the character of the whole pandemic. It suggests an overall waning of Constantinople as the epicentre of infection after the early 7th century and a considerable lull in the west from 600 to about 660; these are both plausible in wider political and economic terms: a much diminished political centrality for Constantinople; fewer people in that period to sustain an epidemic, fewer connections between major settlements.
What was the disease? The widespread reporting of buboes, inflamed lymph nodes, as in the inscription quoted in the section on “The Nature of the Pandemic,” right across the geography of the plague and chronologically across its two centuries, had long been taken as diagnostic of bubonic plague, infection by the bacterium Yersinia pestis, a disease of rodents transmitted to human beings by their fleas. This made the plague of Justinian (including all its recurrences) the first plague pandemic, to be followed later by the Black Death and the modern pandemic that began in the later 19th century. Doubts about the identity of the first two pandemics were raised because the mortality and speed of diffusion of the modern pandemic were respectively minuscule and slow by comparison with those of the Black Death. Now that the DNA of multiple strains of Y. pestis has been recovered from the skeletons of presumed victims of the plague of Justinian in France, Germany, Spain, and Britain, its presence in the period is secure, and it is generally taken to have been the principal, if not the exclusive, cause of the first pandemic. The DNA evidence does not, however, solve the problem of epidemiology. How did the Justinianic plague traverse the Mediterranean and the Middle East so rapidly? The sequencing of the genome of Y. pestis and the elaboration of phylogenetic trees of its different strains have not as yet revealed any differences in virulence between the first two pandemics and the third. Were other vectors than the “classic” black rat and rat’s flea therefore extensively involved? On this, there is no clear evidence and no agreement. Nor is there much agreement on the plague’s prehistory, before it arrived at Pelusium.
There is a long backstory here. Y. pestis is at least known to have infected some individuals in late Neolithic and Bronze Age Eurasia as far west as Sweden, but was not then transmissible by fleas. It seems to have acquired the necessary virulence factors by the time inhabitants of the region of Samara in south-west Russia contracted the disease around 3,800 years ago. The bacterium’s history between then and its arrival in Byzantine Pelusium remains unclear. The lineage of the pathogen that led ultimately to the Justinianic plague did not derive from those Bronze Age strains but seems to have emerged on the border of modern Kyrgyzstan with China several centuries before Justinian’s time. How it moved thence to the Mediterranean is much debated. Some surviving fragments of medical writings of around the 1st century ce suggest that plague was present in Egypt, and perhaps Ethiopia and farther south too, without giving rise to a major epidemic. That could reflect some reservoir of the disease that had earlier established itself in East Africa. Yet the apparent oddity of the pandemic’s irruption at Pelusium rather than Alexandria or Constantinople would be better explained by a route from the Red Sea via the port of Clysma and Trajan’s canal, and thus an approach to the Mediterranean via the Indian Ocean trade and some link between Indian Ocean ports and central Asia such as the Hindu Kush. Suppose, by contrast, the disease’s passage to the Mediterranean had been entirely overland. The discovery of a Y. pestis strain in a burial from the Tian Shan mountains that dates to about 180 ce and that is directly ancestral to the Y. pestis strains found in late antique Europeans might suggest this. It is unclear, however, whether that unfortunate “Hun” or others in the region later were westward bound. If they were, we might expect the disease to have struck Persia first, rather than Byzantium.
Some major disturbance was probably needed to facilitate the mid-6th-century outbreak of the pandemic, whether the disease was then in central Asia or much closer to the Mediterranean. Many historians think they have found one such disturbance in the quite exceptional volcanic activity of 535/6 and 539/40, and the sudden summer cooling it ushered in, heralding the general late antique little ice age of cooler Eurasian summers that followed until about 660. How changes in temperature and precipitation might have affected the plague or its likely hosts or vectors remains, however, to be determined.
Estimates of mortality must underlie any account of the effects of the pandemic. Yet mortality, let alone morbidity, is where our ignorance is at its greatest. The palaeogenetic evidence is as yet of no help, except through its wide geographical distribution, which might suggest high aggregate mortality. Only a handful of inscriptions have been taken to commemorate death from plague. The vivid and horrifying accounts of the main chroniclers, writing in Greek or Syriac, such as Procopius, John of Ephesus, and Evagrius, again suggest heavy losses but offer no reliable statistics. The assumption that the death rate in the first wave of 541–543/4 was of the order of 40–60 percent of the population of much of Europe and of the Middle East is based mainly on the analogy with the first wave of the Black Death of the 14th century. The same pathogen has been detected in both, with no genetic evidence of any difference in virulence; both penetrated rural areas; and mortality from the Black Death tends to be revised ever upwards as scholarship takes in more evidence and weighs it more subtly. On other hand, Europe, and perhaps the Middle East too, was far more densely populated and probably better connected in the 14th century than in the 6th, so the comparison, in terms of plague epidemiology, is inexact. Estimates of cumulative mortality as the early medieval pandemic struck repeatedly over two centuries must be even more speculative. Specific plague mortality is hidden within a longer-term decline of perhaps as much as 50 percent in numbers of settlements and population across much of north-western Europe and the western Mediterranean, and this decline will have had multiple causes.
The effects of numerous and rapid deaths on society are most readily detectible in the Byzantine Empire. In the short to medium term, labourers’ wages rose despite governmental attempts to restrain them. Survivors leasing land came to enjoy greater security of tenure. The gold coinage was debased; some taxes had—grudgingly—to be remitted. There were fewer new secular buildings. It was harder to field armies for service in Italy or against the Persians in the Caucasus, although campaigns were still fought, against enemies whose own belligerence was probably undermined by plague. Overall, but especially in the more densely populated and more connective Byzantine east, the pandemic was a sudden disaster, probably of greater magnitude than the Antonine or Cyprianic plagues. The picture over the longer term in the east is more varied and harder to ascribe to plague, whether solely or in tandem with climatic deterioration. If many cities witnessed a significant decline in population and economic productivity after about 550, there were still settlements, both rural and urban, such as in the southern Levant and Egypt, whose prosperity lasted into the 7th century and indeed the period of Islamic domination.
Yet more difficult to determine are the cultural effects of the Justinianic plague and its subsequent returns. The pandemic has been associated with changes in funerary commemoration, an upturn in church building, intensified apocalyptic speculation, the elevation of Mary the Mother of God to a primary position as state protector, the veneration of icons, and the sacralization of the imperial office. In the controversy that such claims have generated, which can be reduced to weighing up long-term cultural change as against the direct result of the shock of plague, it is worth noting that comparable debates about the effects of the Black Death on piety and its imagery have generally favoured the long-term perspective, even though plague may have accelerated some pre-existing developments.
It is in that light that the grandest claims of all should be assessed: those which associate plague, in synergy with the late antique little ice age, and the ending of empire. Just as the Black Death defined the threshold between the medieval and the modern, it is argued, the Justinianic plague ended antiquity and ushered in the Middle Ages: a fragmented impoverished west in which the centre shifted away from the Mediterranean northwards, and an east dramatically reduced by the forces of Islam, which had not suffered from plague until they invaded Egypt and Syria. Yet Byzantium—reduced, retrenched, reconfigured—proved to be “the empire that would not die.”2 Empire had, of course, already died in the west before plague arrived. There is no obvious caesura in the 6th-century west, as some historians have posited for the east, except perhaps in Anglo-Saxon England. And in any case monocausal or indeed causal explanations tout court of the “end of antiquity” are not discussed in other contexts. What caused the end of antiquity is not generally seen as a question worth asking. Full weight needs to be given to the cumulative evidence for the plague of Justinian as a disaster in the short to medium term without letting it skew our whole characterization of the period of two centuries during which plague recurred.
- Benovitz, Nancy. “The Justinianic Plague: Evidence from the Dated Greek Epitaphs of Byzantine Palestine and Arabia.” Journal of Roman Archaeology 27 (2014): 487–498.
- Feldman, Michal, et al. “A High-Coverage Yersinia pestis Genome from a Sixth-Century Justinianic Plague Victim.” Molecular Biology and Evolution 33, no. 11 (2016): 2911–2923.
- Gebre Selassie, Yohannes. “Plague as a Possible Factor for the Decline and Collapse of the Aksumite Empire: A New Interpretation.” ITYOP̣IS 1 (2011): 36–61.
- Green, Monica. “Putting Africa on the Black Death Map: Narratives from Genetics and History.” Afriques 9 (2018).
- Haldon, John. The Empire That Would Not Die: The Paradox of Eastern Roman Survival, 640–740. Cambridge, MA, and London: Harvard University Press, 2016.
- Harper, Kyle. The Fate of Rome: Climate, Disease, and the End of Empire. Princeton, NJ, and Oxford: Princeton University Press, 2017.
- Keller, Marcel, et al. “Ancient Yersinia pestis Genomes from across Western Europe Reveal Early Diversification during the First Pandemic (541–750).” Proceedings of the National Academy of Sciences 116, no. 25 (2019): 12363–12372.
- Little, Lester K., ed. Plague and the End of Antiquity: The Pandemic of 541–750. Cambridge, UK: Cambridge University Press, 2007.
- Meier, Mischa. “The ‘Justinianic Plague’: The Economic Consequences of the Pandemic in the Eastern Roman Empire and Its Cultural and Religious Effects.” Early Medieval Europe 24, no. 3 (2016): 267–292.
- Mordechai, Lee, and Merle Eisenberg. “Rejecting Catastrophe: The Case of the Justinianic Plague.” Past and Present 244, no. 1 (2019): 3–50.
- Mordechai, Lee, et al. “The Justinianic Plague: An Inconsequential Pandemic?” Proceedings of the National Academy of Sciences 116, no. 51 (2019): 25546–25554.
- Mulhall, John. “Plague before the Pandemics: The Greek Medical Evidence for Bubonic Plague before the Seventh Century.” Bulletin of the History of Medicine 93 (2019): 151–179.
- Newfield, Timothy P. “Mysterious and Mortiferous Clouds: The Climatic Cooling and Disease Burden of Late Antiquity.” In Environment and Society in the Long Late Antiquity. Edited by Adam Izdebski and Michael Mulryan, 89–115. Leiden, The Netherlands: Brill, 2018.
- Sessa, Kristina. “The New Environmental Fall of Rome: A Methodological Consideration.” Journal of Late Antiquity 12, no. 1 (2019): 211–255.
1. Denis Feissel, Chroniques d’épigraphie byzantine: 1987–2004 (Paris: Association des amis du Centre d’histoire et civilisation de Byzance, 2006), 267 (no. 847).