The Third Horseman

by William Rosen

  Every famine, therefore, is to a greater or lesser degree a creature of changes in climate, or at least weather. One of the best known of those changes occurs most Decembers when an area in the tropical Pacific Ocean experiences a shift in temperature of about half a degree Celsius. When the temperature increases, the result is an El Niño (so-called because it’s generally noticed around the same time that the west coast of South America prepares to celebrate the birth of Jesus).* This temperature change creates a huge turning gyre that displaces surface water largely east to west, while an equatorial countercurrent moves west to east along a line several degrees north of the equator.

  Typically, an El Niño surges annually. But one or two times each decade, it is reinforced by the so-called southern oscillation—the temperature around Tahiti goes up when the temperature around Darwin, in Australia, goes down, and vice versa—and earns the acronym ENSO, for El Niño–Southern Oscillation.

  ENSOs have their largest impact on nations and people that surround the Pacific, from China to Peru, but it’s such a huge driver of worldwide climate that its effects are everywhere. The ENSO of 1876–77, for example, produced huge rains over Southwest Asia and the American tropics, with corresponding droughts in Brazil and southern Africa. Another ENSO preceded the Ethiopian famine of 1984–85. The “great drought” or grande seca of 1877–79 in Brazil killed at least half a million, and has been called “the most costly natural disaster in the history of the western hemisphere.”

  One artifact of researching famines is that almost everyone discovers fewer of them in antiquity; the fact that Ethiopia suffered “only” four famines between 100 CE and 1400 CE, and twenty-three between 1900 and today, is almost certainly evidence of better documentation rather than greater hunger. Even for famines that can be found in the historical record, their effects are hard to quantify; claims that the Chinese famine of 209–203 BCE killed up to 90 percent of the population, or that the Bengal famine of 1770 resulted in a loss of one-third of the population, or even that the famine that accompanied World War I killed 40 percent of the population of Persia, are both impossible to verify and highly improbable. Nonetheless, there is some consensus around the numbers associated with the deadliest famines in history.

  The first multiyear famine probably occurred in Egypt during the third millennium BCE, during which the Nile failed to break its banks for seven years in a row (the memory of which may be an inspiration for the biblical story of Joseph). The two worst collapses, in terms of the percentage of deaths among the pre-famine population, are both Irish: the frost-driven famine of 1740–41, in which 13 percent of the population died, and the “potato famine” that ran from 1846–52, with 12 percent mortality. But they were, of course, restricted to a single island. The excess mortality during the seven years of the Great Famine was somewhere between 5 and 12 percent for all of northern Europe.

  In its geographic extent, in its duration, in the number of lives it touched and erased, the Great Famine was unprecedented. It wasn’t merely a crisis of production, population, weather, or war. David Arnold, a historian at the University of London, has found that almost all famines tend to follow one of four different scenarios: (a) a brief but traumatic shock to agricultural productivity that pushes far beyond its best-case-in-good-weather output; (b) a sustained failure of weather, particularly rainfall; (c) a sustained problem of distribution, usually caused by war (including the special case of famines caused by siege, where food is not only scarce but can’t be acquired elsewhere, at any price); or (d) conservatism in agricultural practice, which turns a modest problem into a cascade. The Great Famine followed all four. And it followed them year after year after year.

  This matters even more than it seems. One season of poor weather could destroy an entire harvest. But a truly serious famine isn’t caused by a single lost harvest; it is almost always the result of back-to-back losses, which means that its likelihood is the same as the probability of successive seasons of bad weather. Nearly four centuries of monthly mean temperature data in Europe reveal what a statistician would call positive serial correlation: good years tend to be followed by other good years. However, the weather that matters for famine isn’t the average but the extremes. In Europe, for millennia, an extremely cold or rainy year is one that deviated from the average by more than 10 percent—an extra month of frost with five or even six times the normal rainfall—and such deviation has been recorded in successive years only twice since 1659, which makes the years 1315–1317 about a once-every-two-hundred-years event.

  Nature wasn’t the only enemy facing the rural peasantry. Even in non-famine years, European farmers learned to fear disettes—the time after the grain from one harvest had been exhausted and before that from the new one arrived. Any decade from the middle of the twelfth century on would see variations in harvest size of 10 to 20 percent just based on fluctuations in rainfall and sunshine, and a shortfall of 20 percent meant starvation for at least some people. The shortfall in 1315 was as much as 60 percent, and it would be more than 20 percent for at least two more years throughout Europe. Death records for the peasant population living on the manors of Hampshire, Berkshire, and Somerset show deaths in the range of 10 percent between 1316 and 1318. In Essex and Worcestershire, the same data show a mortality rate of 15 percent. Both were at least three times normal.*

  Those who avoided starvation faced another threat: lawlessness, already rife in medieval Europe, but dramatically worsened after the failed harvest of 1315. In Kent, during 1316–17, a third of all thefts were of grain and grain products like bread and ale; 40 percent were livestock. Rioters took over the French town of Douai. Landless knights and men-at-arms, who had been well trained in looting as a necessity and tactic of war, were well positioned, in between battles, to put their skills to use in a more entrepreneurial fashion. Extortion became chronic, as armed men demanded the output of the peasantry, only occasionally using the cover of a (frequently fraudulent) royal warrant. One minor noble, William de Cotes, used a false commission to demand ten oxen, two horses, four cows, and £20 of assorted agricultural produce from the village of Saintbury; another, Richard de Richmond, blackmailed a parson for a ransom of £40. With thousands of men like de Cotes and de Richmond, few could feel free of the threat of extortion.

  Crimes of all kinds are always one of famine’s most reliable companions. Perhaps because, among all natural (or mostly natural) disasters, famines are by far the slowest moving, they are particularly able to undermine the more elevated human feelings, one hungry day at a time: Honesty and generosity don’t disappear, but they become harder to find when people go without food. The same people who show enormous courage in the face of earthquakes and fires find their bravery exhausted by months with too little to eat. Hopelessness replaces hope. And hopeless people commit acts they would otherwise find unbelievable, even unthinkable. Such as cannibalism.

  There are always stories of cannibalism during famines. Some are better documented than others; during a famine in third-century BCE China, the emperor officially granted parents the right to eat their own children; or, if too squeamish, to sell them for food. The frequency of cannibalism during the Great Famine isn’t easy to document, but there is no doubt that it was widely believed at the time to be widespread: In Estonia, in “Anno Domini 1315, tanta fames in Lyvonia et Esconia orta est, quod matres filiis vescebantur” (the mothers were fed the children). An Irish chronicle recorded that between 1315 and 1318, “people used to eat one another, without doubt, throughout Erin” and, at the height of the famine, they “were so destroyed by hunger that they extracted bodies of the dead from cemeteries and dug out the flesh from the skulls and ate it; and women ate their children out of hunger.”

  The monk and chronicler Johannes de Trokelowe wrote, in his Annales, “Men and women furtively ate their children and even strangers in many places” while “jailed thieves . . . devoured themselves at the moment when they were half-alive.” The fift
eenth-century annals of the priory of Bermondsey—probably written around 1433—claimed (from earlier records) that “pauperes enim pueros suos manducabant, canes, murelegos, stercus, columbarum.”* Another contemporary chronicle, from Poland and Silesia, describes the same horrific scene: “in many places parents devoured their children and children their parents; also many ate the flesh from cadavers” hanging from gibbets.

  • • •

  The horrors of the famine that began with the lost harvest of 1315 were not a direct result of climate change. Weather is a nonlinear system, one in which a small change in initial conditions can have giant consequences in subsequent ones. Climate isn’t, at least not in the same way. While some tantalizing research suggests that this phenomenon replicates itself in long-term climate change—that a graph displaying six months of weather can resemble a graph of six centuries of climate; a self-similar “fractal” like a coastline that looks the same for a hundred feet as it does for a hundred miles—it is a long way from widespread acceptance.

  For one thing, long-term climate seems less sensitive to even dramatic changes in initial conditions. In 1257, a huge volcanic eruption, probably in Indonesia—the biggest in millennia—may have spread a particulate veil of sulfur around the planet and initiated a series of reinforcing feedback loops between Northern Hemisphere sea ice and the water surrounding it, with the result that ice caps in Arctic Canada and Iceland began advancing twenty years later. There is even a good argument to be made that this same volcanic eruption was responsible for a famine of its own: “In that year [1257],” according to the Chronicles of the Mayors and Sheriffs of London, “there was a failure of the crops; upon which failure, a famine ensued, to such a degree that the people from the villages resorted to the City for food; and there, upon the famine waxing still greater, many thousands perished.” However, and despite an abnormally hot summer in 1262 (also related to the volcanic eruption, which messed up atmospheric circulation for at least five years), things reverted to normal. There were tremendously cold winters regularly between 1308 and 1312—dogs could hunt rabbits in the middle of the frozen-over Thames during the winter of 1309–10—but by 1312, the NAO index had shifted again, and the winter of 1312–13 was again mild.

  Moreover, the weather systems that caused the Great Famine, and carried decades of instability along with a generally declining average temperature, were, like the Medieval Warm Period itself, largely confined to Europe, particularly northern Europe. Farther afield, the Indian monsoons, Nile floods, and other annual weather events that determined agricultural productivity continued to operate within normal boundaries. The rains of 1315–1316 were distinctively western European.

  This doesn’t mean that worldwide climate didn’t change at all. Much of the world did experience a climatic downturn throughout the fourteenth century (really beginning in 1275). The Little Ice Age, as it has come to be known—the term dates to a 1939 paper by the geologist François Mathes on the growth of glaciers in California’s Sierra Nevada Mountains—was real, even if its causes are even less well understood than those of the Medieval Warm Period that preceded it. There are respectable theories ranging from increased albedo (the amount of light reflected by the surface of the Earth) because of greater cultivated land to a drop in atmospheric CO2 because of deforestation to variations in solar radiation. Sunspot activity started an astonishingly regular increase around the year 800, peaked just before the beginning of the fourteenth century, and then fell like a stone for the next century, reaching a minimum not seen since the third century BCE. One thing about long-term climate change not in doubt, however, particularly when the change is from a period of anomalous stability to something else, is that it brings with it far greater volatility.*

  If the normal destructiveness of weather is amplified by underlying climate change, the great rains of 1315 were also magnified by human activity. Some of that activity was bureaucratic: By 1316, Edward’s tax collectors were collecting grain and fodder from England’s peasantry, as part of their “prises”—property confiscated at the prerogative of the king. The abuse was exacerbated through 1317, as both prises and taxes on “moveable” property, needed to support ongoing English garrisons and to repay the money borrowed from Antonio Pessagno to pay for the 1314 campaign that foundered at Bannockburn, were collected simultaneously despite the famine. It was a losing game: peasants cut back on consumption, which left even more grain and other movables to be taxed. But as much as the tax policy harmed the south of England, it was disastrous in the northern counties of Cumberland, Westmorland, and Northumberland, where residents begged the Crown for a three-year holiday from purveyors and other of the king’s middlemen.

  Nor were the Crown’s attempts to control prices effective. The Ordinance of 1315, passed at Westminster in order to fix prices for commodities during the famine, failed so utterly that Edward ordered it repealed on January 14, 1316. He wasn’t the only one to try price controls. The magistrates of London, on September 21, 1316, ordered a cap on the price of ale: no one was permitted to sell a gallon of the worst ale for more than three farthings, or the best for three halfpence, under threat of expulsion from the city for a third offense. Edward liked this one so much that in January 1317, he ordered that the price of the best ale in country towns couldn’t exceed a farthing, which meant, of course, that brewers were obliged to pay profiteering prices for grain but to sell their product at a heavy, and unprofitable, discount. Contemporary chroniclers called the price regulations “beyond reason,” stating “it is better to buy dear than to find in case of need that there is nothing to be had.”

  And there was, frequently, nothing to be had. Throughout 1315, the Scots continued to press their advantage by raids into the Tyne valley and other parts of northern England, burning everything they couldn’t carry, and returning with enormous herds of cattle. So fearsome were James Douglas’s raids that Northumbrian mothers sang their children to sleep with the lullaby, “Hush ye, hush ye/Do not fret ye/The Black Douglas/Shall not get ye.”

  The raids were destructive in themselves; in combination with the famine, they were devastating. An anonymous chronicler from Meaux Abbey in Yorkshire claimed that Northumberland was left a wasteland for fifteen years, “deserted by men and wild and domestic beasts.” The Chronicle of Lanercost documents that the raids by Bruce and his lieutenants between 1314 and 1319 were tactics in service of deliberate starvation, “trampling down the crops by themselves and their beasts,” burning the harvest “when the crop had been stored in barns . . . both the corn upon which the people depended for sustenance during that year and the houses where they had been able to take refuge.”

  By the end of 1315, moreover, the combination of Edward’s taxes, the first of two lost harvests, and constant destruction of infrastructure—barns, farm equipment, and mills—by the troops of both Bruce and Edward was devastating not only the agricultural productivity but the entire economy of both northern England and southern Scotland. Literally dozens of lords in the Anglo-Scottish borderlands complained—loudly—to both Bruce and Edward of their near bankruptcies. As 1316 began, an inquiry into the value of lands held by lay lords in the borderlands described estate after estate as “waste.” And not just cropland: the value of a typical fishery on the River Tweed, which might have been worth hundreds of pounds in 1300, fell by anywhere from 30 to 100 percent by 1315.

  Bruce could ravage northern England, but he couldn’t, by raids alone, force Edward to recognize an independent Scotland. He attempted to get the attention of the English king by authorizing privateers to act in the Irish Sea. It worked well enough to force Edward to pull back on his support for the campaign in Flanders being waged at the time by his brother-in-law, Louis X, but as long as most of the wealth of the nation was in the south, beyond the reach of Scottish hobelars, Edward could afford to lick his wounds, think about invading Scotland yet again, and contemplate the collapse of England’s ability to feed itself.

 
; The reliably melancholic writers of the Vita Edwardi Secundi tell us that “the floods of rain have rotted almost all the seed, to such an extent that the prophecy of Isaiah might seem now to be fulfilled, for he says that ten acres of vineyard shall yield one little measure, and thirty bushels of seed shall yield three bushels . . . in many places the hay lay so long under water that it could neither be mown or gathered.” By the end of 1316, the rains abated—some—but the winter of 1317–18 was the harshest of all, lasting (in the record of a French chronicle) “from the feast of Saint Andrew or thereabouts until Easter” . . . which is to say, from November 30 to April 23, or just over twenty weeks. An English chronicle reads: “A thusent winter ther bifore com nevere non so strong . . . com never wrecche into Engelond that made men more agaste.”

  They were “agaste” outside of England as well. One consequence of the frighteningly volatile weather was sea erosion. As the Arctic Ocean becomes colder, it increases the temperature differential between it and the Atlantic Gulf Stream; the bigger the gradient in the oceans between about 50o and 65oN, the more storms in the North Sea basin. The more storms, the more frequent the storm surges; and geography dictated that Atlantic storm surges were narrowed and compressed like a funnel between Britain and the Continent—the Channel is only twenty-two miles wide between Dover and Calais—making the northern European coastline uniquely vulnerable.

  Which is what the record shows: floods increased from fewer than five annually in the twelfth century and before to more than twelve by the end of the thirteenth and the beginning of the fourteenth, a peak in storm activity “unsurpassed in the last 2,000 years.” The low-lying wetlands that had been reclaimed over the four centuries of the MWP were inundated, in northern France, the Netherlands, and southeast England; more than two thousand acres of onetime marshland that had been converted to cropland in eight villages in Sussex alone were submerged. Dunwich, which was one of the five wealthiest ports in England in the late thirteenth century, was so devastated by one flood after another that 269 houses (including crofts and barns), 10 other buildings, and 2 shops had disappeared—fully a quarter of the entire port town. By the 1330s, most of five parishes of the town, and up to 600 houses, had been “reclaimed” by the newly unstable sea.