Death on the Nile: Disease and the Demography of Roman Egypt (review)

The demography of the ancient world has become an extremely controversial topic over the past decade, as scholars have attempted to interpret several centuries' worth of complicated evidence drawn from all over the Mediterranean basin. Numbers in Greek and Roman authors are notoriously unreliable, and although attempts have been made to use Model Life Tables as a guide, not all have concurred. The most likely source of reliable evidence has been supposed to be the papyri of Roman Egypt, where detailed census records and other administrative minutiae appear to permit a more accurate reconstruction than for, say, the city of Rome; historians have then tended to use conclusions based on Egypt to refer to the whole of the Roman Empire.

Walter Scheidel here combines the evidence from a variety of Egyptian sources with modern demographic studies of later periods in the Nile Valley, in order to warn of the sheer fragility of almost all hypotheses. He is ready to point out the biases of the evidence that lead to somewhat implausible conclusions: tax registers, for instance, say far more about men than women, and, for obvious reasons, more about the living than the dead, thus leading to an overrepresentation of healthy males aged thirty to forty. City populations are less well represented than those of small towns and villages, but a reliance on the data from one or two small towns may in turn lead to historical distortion.

A few positive conclusions emerge from the statistics and tables—most notably, the discovery (or confirmation) of two seasonal peaks of disease, one in November-December, the other in summer, varying in time within the Nile Valley. There are some parallels with nineteenth-century Egypt, with perhaps over 60 percent of the population dying before the age of thirty. The effects of the great Antonine plague of 166-170 C.E. are not easy to pick out immediately, but reveal themselves over the next two generations, with substantial population losses (perhaps due to recurrent localized outbreaks?).

But the greatest value of this study is in its insistence on the need to balance the data of population against other factors, such as the possible patterns of disease and the ability of the agricultural base to support a certain number of mouths. The result may be impressionistic, but at the same time it also lays down the limits of the possible. Within these limits, however, there may be great fluctuations. Like his friend and erstwhile collaborator Robert Sallares in his Malaria and Rome: A History of Malaria in Ancient Italy (2002), Scheidel stresses the possible microenvironments of disease, and warns against a facile extrapolation of conclusions from one region to another. If there are discernible different patterns between Upper and Lower Egypt, or between Alexandria and Tebtunis, one should be very cautious about using Egyptian evidence to describe Roman Britain or Asia Minor.

One problem seems to be dismissed too hurriedly. Rufus of Ephesus, around 100 C.E., reported an epidemic of buboes in North Africa, Egypt, and Syria, a century or so earlier. Was this bubonic plague? Scheidel thinks it was, although conceding that there is no other evidence for this disease until the sixth century, and that medical texts and demographic data show no signs of a plague epidemic. He is reduced to positing regular minor local outbreaks, and perhaps a major mutation of the bacillus, in order to link the buboes of Rufus with the plague pandemic of 541/42. Others have been more skeptical about the evidence of Rufus.

Like Sallares, and like Peregrine Horden and Nicholas Purcell in The Corrupting Sea: A Study of Mediterranean History (2000), a book that may have appeared too late to be considered here, Scheidel...