by David Keys
“The means by which one contracted this disease were diverse and beyond telling. For some perished just through association and living together, others by physical contact, or by being in the same house, or even [through contact in] the market-place. Some people had escaped infected cities and themselves remained well, but passed on the disease to those who were not sick. There were those who remained entirely unaffected, even though they lived with many of the afflicted, in fact coming into contact not only with many sufferers but also with the dead. Others positively embraced death on account of the total loss of their children and family, and for this reason went cheek-by-jowl with the sick, but still were not struck down, as if the disease resisted their will.”
Many historians have tended to see the plague pandemic that devastated so much of the world in the sixth and seventh centuries as consisting of a series of distinct outbreaks. Some church historians and others who were alive at the time even saw it that way, but they were often looking at the catastrophe from the vantage point of the large cities where they lived—places such as Constantinople, Antioch, and Alexandria.
In reality, both the major plague epidemics and the less extensive outbreaks should be regarded as one integrated event, albeit a very long one, which lasted for between 180 and 210 years. Rather than looking at the records of simply the most prolific contemporary historians, it is vital to trawl through a wider number of sources to find even the smallest reference to plague.
Historians have found that there were dozens of outbreaks over the years from 541 to 717, and perhaps even as late as 745.4 And those are just the epidemics and outbreaks that are recorded. From c. A.D. 600 onward, there appears to be a reduction in the frequency of plague outbreaks in the Roman Empire, but this may simply be a function of the paucity of sources from the seventh century. Indeed, probably only a small percentage of the outbreaks were ever recorded, and of those records that were made, only an even smaller percentage have survived to the present day. These records are best for the eastern Mediterranean region and for western Europe; the pandemic also affected, though not initially, China and Persia. Yemen was almost certainly hit by some time in the 540s. And then there are vast areas—such as Africa or central Europe—that no doubt were affected but for which virtually no written records exist.
The plague passed from rat to human, sometimes from human to human, hardly pausing on its unpredictable journeys. Everywhere it rampaged, it must have substantially reduced population levels, thus creating vast tracts of abandoned agricultural land.5 Sometimes it would spread to myriad towns6 and villages in a single year, while on other occasions it would bide its time, skulking in a few quiet or remote localities, only to burst forth from these nameless havens of death a few years later. Indeed, it is likely that at no time between 541 and c. 750 was the plague ever entirely absent from the Mediterranean region and its various hinterlands.
2
T H E O R I G I N S O F
T H E P L A G U E
Twenty-five days’ sailing time down the east coast of Africa, one arrives at a “metropolis” called Rhapta.1 This information, recorded by the second-century A.D. Greek geographer Ptolemy, is the last-known contemporary reference to a now long-lost African city that once flourished somewhere along the coastline of what is now Kenya and Tanzania.2
The only other reference—in a first-century A.D. pilots’ manual called The Periplus of the Erythraean Sea—says Rhapta was a source for “great quantities of ivory and tortoiseshell” and was inhabited by “very big bodied men.”3 The metropolis was located on a river “not far from the sea”4 and was also involved in the export of “rhinoceros horn, and a little nautilus shell” and the importation of glass beads and iron goods, especially “axes, knives and small awls.”5
According to The Periplus, written around A.D. 40, the place was at least nominally under the control of Arab merchants from Yemen. It appears that these merchants intermarried with local women, gave gifts of wine and grain to the local chiefs, and had royal Yemeni approval to exact tribute from the area.
From The Periplus and Ptolemy, it is clear that Rhapta was the most remote—and the largest—of four ancient East African trading ports, from north to south: Opone (now known as Ras Hafun, in Somalia), Essina and Toniki (both near modern Barawa in Somalia), and Rhapta itself.
Opone—spectacularly sited on what is essentially an island linked to the coast by a thirty-mile-long sandbar—may have had several hundred inhabitants, covered up to five acres, and appears to have been abandoned some time in the mid–sixth century A.D. The latest pottery found by archaeologists on the site dates from the fifth or early sixth century. Up till that time, it seems to have acted as a transshipment point for Mediterranean, African, and Indian trade goods.
The other three ports, Essina, Toniki, and Rhapta, have never been archaeologically detected—probably because, like Opone, they never made it into the medieval period. Certainly an examination of the twenty-two pre-eleventh-century settlement and trading sites on the East African coast that have been archaeologically investigated shows that nineteen started functioning only between the seventh and ninth centuries A.D., two may possibly have started up before the sixth century, and only one definitely came into existence before the sixth. That strongly suggests severe settlement discontinuity in the sixth century. What is more, throughout East Africa, the pottery type abruptly changes at exactly the same time. Before the sixth century it is all early Iron Age (Kwale ware), while after the sixth century it is all late Iron Age (Tana ware). There was also a move in some areas at the same time from a concentration on agriculture to a more pastoral economy.6
The sixth century was a great watershed in East African history—a period of very rapid change and probably decline, in which the key ports simply ceased to exist and the agricultural economy shrank. The culprit was almost certainly plague—the same epidemic that devastated Europe and the Near East in that same fateful century.7 Indeed, it was most likely from an ancient East African wild-animal reservoir of plague that the disease broke out to infect so much of the late antique world.
Historically, there have been several major natural plague reservoirs in which the disease circulated harmlessly among specific high-immunity wild animals. These areas—the Himalayas, Central/East Africa, and the central Asian steppes—have been the ultimate sources for the plague epidemics that have hit Europe and elsewhere over the centuries.8
Evidence that the sixth-century pandemic originated in Africa rather than Asia is very clear.9 First of all, the major Asian high-population region, China, did not become infected until half a century after the Mediterranean region had suffered its first visitation. Indeed, China was infected from the Mediterranean region via the Middle East. Certainly the major Middle Eastern power, Persia, was infected only after the disease had hit the Roman Empire. The Persians apparently contracted it from Roman soldiers. Second, there is no evidence of plague being endemic on the central Asian steppes prior to the later medieval period. Third, a major contemporary source, the Syrian-born historian Evagrius, actually recorded that the epidemic came from Africa (“Aethiopea”).
As already described, the first town in the Mediterranean world to be hit was the port of Pelusium, where, after transiting the Roman world’s equivalent of the Suez Canal, cargo originating in the Red Sea area and in Africa was unloaded for transshipment to the rest of the Roman world. What’s more, Yemen—halfway between East Africa and the Mediterranean—seems to have been an early victim of the plague, being hit by the disease sometime in the 540s.10
But why did the plague break out of its animal reservoir in East Africa at that particular time?11 The answer is prosaic in the extreme—the weather.
Modern research on surviving wild-animal reservoirs of plague—monitored by the U.S. Centers for Disease Control—has concluded that most plague outbreaks are caused by sudden and severe climatic change. Massively excessive rainfall is the most likely cause of plague spread, especially if it follo
ws a drought, although a severe drought followed by normal weather could theoretically also spark an outbreak.
When there is excessive rainfall, vegetation growth increases. Thus there is more food available for herbivorous animals and insects, and rodents—including those that are carriers of the plague bacterium but are themselves immune to it—therefore breed more. Their larger numbers enable a greater survival rate vis-à-vis the slower-breeding predators who feed off the rodents, and a rodent breeding explosion occurs. In order to find their own foraging territory, the cumulative range of the rodents has to increase, and a virtual bow wave of these plague-carrying wild animals spreads inexorably outward over a period of months. Soon the creatures come into contact with other normally plague-free rodents, which then spread the disease to humans.
In the slightly less likely, though theoretically feasible, drought scenario, lack of rainfall and food kills huge numbers of plague-carrying wild rodents and the larger predators that normally eat them. However, the minute the drought is over, the fast-breeding rodents recover their numbers quickly compared to the slower-breeding predators. There is then, for a few years, a great imbalance between hunter and hunted in favor of the hunted. A breeding explosion takes place and the plague-infested rodents spread like wildfire.
However, the most dramatic scenario of all is one in which a severe drought is followed by significantly increased rainfall. That, or something very much like it, is almost certainly what took place in East Africa during the worldwide climatic chaos of the 530s.
While weather was without doubt the motor that drove the spread of plague in East Africa, the key vector was the humble flea. Although the rodents were immune to plague, the fleas that lived on them were not. Fleas die of plague—but it’s actually the process of dying itself that helps them spread the disease.
As a flea becomes ill, and under specific climatic conditions, part of its gut becomes blocked by a mixture of multiplying plague bacteria and clotted blood.12 The flea then begins to starve, and becomes so ravenous that it will jump onto virtually anything that moves, irrespective of whether it is its normal host species or not. Of course, the flea’s hunger will not be satisfied, because its gut is blocked. So it will move rapidly from host to host, biting each one—and consequently spreading plague—in an impossible mission to quell its hunger.13 The disease itself thus produces the very mechanism for its own spread.
The species in East Africa that were probably the reservoirs for the disease were gerbils and multimammate mice. The sandy-colored gerbil normally has two litters (totaling ten offspring) per year. Gerbils are very territorial, and an individual will travel two to three miles per season in search of an area it can control as its own exclusive territory. Thus in optimal food conditions, when gerbil numbers increased, the need of each gerbil to find its own territory would have resulted in a wave of plague-carrying individuals spreading outward at substantial speed.
The multimammate mouse—a dark brown rodent about the size of a golden hamster—lives in colonies consisting of up to fifty individuals. Their gestation time is twenty-three days, and they have two litters per year. Normally they have only five offspring per litter, but when there is optimal food availability, the number can treble to fifteen. A pair can produce over a thousand descendants in a year. Today they are still a principal wild host of plague in Africa.
It is likely that the gerbils and multimammate mice then passed the disease to a ratlike creature in the genus Arvicanthus. The latter would not have been immune to plague, but in appropriate climatic conditions would have outbred even the multimammate mouse: In wet weather it can achieve densities of up to a hundred per acre, and it and its offspring can produce thousands of new individuals per year. Neither the multimammate mouse nor Arvicanthus is averse to invading human settlements, and would therefore have come into direct contact with nonimmune Rattus rattus—the black rat, a species that specializes in infesting human environments, including farms, storehouses, houses, villages, towns, markets, ports, and ships.
In good climatic conditions, one pair of black rats (also known as house rats and ship rats) can produce thousands of descendants each year, especially if their slower-breeding predators are rarer than normal (due to, say, a recent drought). The species is aggressive, highly adaptive, and able to eat virtually anything—insects, seeds, meat, bones, fruit, even each other!
Once the starving fleas had jumped in their billions from gerbil and multimammate to Arvicanthus and on to the black rat, it would have been only a matter of days, even hours, before the first humans started contracting the plague.14
Transported by ships from port to port, Rattus rattus carried the plague bacterium from community to community. The archaeological evidence suggests that as the disease rolled northward up the Red Sea to Egypt, in its wake a whole way of life collapsed in East Africa and probably in southern Africa as well. The metropolis of Rhapta was inhabited by early Iron Age Bantu people, and the other ports of Opone, Essina, and Toniki were probably inhabited by late Neolithic Cushites, or possibly early Iron Age Bantu people.15 At the time of the plague, as already noted, these ports seem to have virtually disappeared. Apart from Opone, their precise locations are not even known.
Inland, Bantu agriculture seems to have declined, and the Bantu appear to have rapidly and increasingly adopted from the Cushites both the latter’s cattle-based pastoral economy and their particular style of pottery. In the seventh century (i.e., after the plague had started), this cattle-based pastoral tradition began to spread south and supplant cereal growing all over southern Africa.
Two questions remain, however. How did the plague give an advantage to pastoralism (a livestock-based economy) over agriculture (a crop-based economy)? The answer lies in the number of rats and other plague-carrying rodents attracted to the two different economic systems. Food crops—whether in fields or in storage—attract rats. Food sources on four legs—in this case, cattle—do not. It was this difference that appears to have given pastoralism an advantage over agriculture at this critical time.
The second question is, what were ships carrying between East Africa and the Roman Empire? Ivory was one of the most valuable commodities needed by the empire. Demand for magnificent ivory chairs, exquisite ivory children’s toys for the rich, ivory writing tablets, religious relic boxes, and countless other ivory works of art had generated a trading system that no doubt stretched deep into Africa. Well before the sixth century, the elephants of Eritrea, on the Red Sea—used in antiquity as beasts of war—had all been hunted to extinction. So East Africa (modern Kenya and Tanzania) became virtually the only source for the vast quantities of ivory the Roman Empire desired.
Up until the plague and its destruction of the East African ports, the Roman Empire imported up to 50 tons of ivory every year from East Africa. This level of ivory trade necessitated the killing of up to five thousand elephants a year. In terms of cash, the merchandise was worth up to 220,000 gold solidi (equivalent to around $400 million today) to the Arab and Greek merchants who controlled the trade.16
In East Africa, the trade sustained not only a series of ports but also a series of coastal chieftaincies, which must have exercised disproportionate amounts of local power through the trade goods and imported weaponry at their disposal. After the plague had substantially reduced the population and destroyed the ports, the ivory trade virtually ceased.
Between the year 400 and the eve of the plague (c. 540), of the estimated 400,000 major ivory artworks made, some 120 survive; from the period 540 to 700, only 6 survive. The surviving figures are so strikingly different that they show, without doubt, that after the mid–sixth century very little ivory was coming into the empire. The golden age of ivory artistry had been terminated by the plague.
A century later, Mediterranean and European population levels had declined significantly, with Constantinople shrinking from a city of half a million to one of fewer than a hundred thousand inhabitants. Meanwhile, the mid-sixth-century climati
c crisis and its consequences had been generating other mechanisms through which parts of Europe and Asia were to be transformed. The remote steppe of Mongolia was to become the unlikely source of change.
PART TWO
THE BARBARIAN
TIDE
3
D I S A S T E R O N
T H E S T E P P E S
In A.D. 557 or 558 a fierce Asiatic people called the Avars arrived on the eastern fringes of Europe from Mongolia. Twenty years later, they had conquered significant parts of the eastern half of the European continent and had humbled the Roman Empire by invading the Balkans, including Greece itself, either directly or by proxy through their vassals. They became a major element in the drawn-out process in which the empire gradually lost so much of its territory and its strength, and through which European and Middle Eastern history was fundamentally changed. But what caused the Avar migration has always been something of a mystery.
All that history records is that by A.D. 545, after 150 years of being the ruling ethnic group in Mongolia, Avar power was challenged by another Mongolian people—the early Turks. It is quite clear from the historical sources that the Avars had inexplicably become weaker in relation to their vassals, the Turks, and by 552 the Turks had turned the tables on their Avar overlords and taken over Mongolia.1 Many of those Avars who had not been slaughtered by the victorious Turks trekked west into exile, toward Europe.