Yale School of the Environment
Admissions Alumni Giving

When Civilizations Collapse

Tell Leilan site
For the past 30 years, midway through the drive from his expedition living quarters in Qahtaniyah to his archeological excavation at Tell Leilan in northeastern Syria, Harvey Weiss has contemplated the enigma of a railroad station. French colonial authorities built the modest two-story structure, along with a railroad to the Mediterranean Sea, shortly after World War I, and everything about it belongs in rural France—the hipped roof, the multipaned casement windows, the interior post-and-beam timbers. It is the architecture of empire, writ small. 
Tell Leilan, ca. 1920 railroad station
Photo by Harvey Weiss (Yale University)

Early French mandate, ca. 1920, railroad station, post-and-beam construction, at Tell Leilan/Kubur al-Bid, northeast Syria, for cereal grain transport from Habur Plains west to Aleppo for capitalized grain market and redistribution. Former Ottoman lands lost to cultivation during Little Ice Age were seized and developed by the French state for retrieval of new cereal agriculture harvests.

Top photo: Tell Leilan Acropolis temple, ca. 1800 BC, with semi-engaged mud-brick spiral columns, was one of the first monumental buildings excavated by the Yale University Tell Leilan Project. When recovered in 1982, the temple and artifacts upon its floors were key for identifying the site as the “lost” city of Shubat Enlil, and attracted much attention among Near Eastern archaeologists. Now, after decades of paleoenvironmental and regional survey work, the temple is iconic of the post-2200 to 1900 BC drought, when conditions favorable for high-yield, dry-farming cereal agriculture returned to the region and encouraged resettlement of the Khabur Plains by former pastoral populations.

For Weiss, though, the train station has come to tell a larger story about climate change and human adaptation to it (or failure to adapt)over thousands of years. The French built it, he explained recently, to pull out the cereal harvest of the Khabur Plains, the most fertile rain-fed agricultural area in northern Mesopotamia, and it still serves that function. At harvest time, the 100-kilo sacks of wheat are piled high there, awaiting shipment to market. 
But why was the Khabur region barren and largely abandoned for hundreds of years before the French arrived? “Why do all the 18th- and 19th-century travel accounts indicate that the region was empty?” asked Weiss. “And what is the meaning of that abandonment for the Ottoman economy and the historical problem of Ottoman decline?” Other scholars have argued for the past 50 or more years that provincial administrators in the declining centuries of the empire were corrupt, inefficient and unable to maintain law and order, allowing the region to be abandoned despite its high productivity and its tax value for the central government. “The argument being,” Weiss added, “that there were no environmental and certainly no climate reasons for the abandonment.”

The alternative answer, in the title of a paper just published by Weiss and his French co-authors in Proceedings of the National Academy of Sciences, may seem unsurprising: “Drought is a recurring challenge in the Middle East.” But it has been the subject of bloody academic combat ever since Weiss first proposed 20 years ago that long-term shifts in climate are a key factor in understanding the tumultuous history of the Middle East. Before then, archaeologists working in the region took it for granted that the climate had been essentially stable and benign over the long-term, and that social, political, military and economic forces were the key determinants in the rise and fall of civilizations.
The new study suggests merely that climate change caused the late Ottoman abandonment of the Khabur River valley, not collapse of the entire Ottoman Empire. But it also makes the case that archaeologists and historians ignore climate change at their peril. “There is an environment in which history occurs,” said Weiss. “There are reasons for regional abandonments that are definable, observable and testable,” even when ancient peoples have left no written record of climate changes.
For the new study, Weiss and co-authors David Kaniewski and Elise Van Campo from the Université de Toulouse, France, used pollen to reconstruct 10,000 years of climate history in the region. Their technique was to take a column of stratified sediment from the side of a dry river channel near Tell Leilan and identify the mix of plant types in different layers. The percentage of pollen in a sample from dry climate plants (like dryland wormwood and tamarisk) or humid climate plants (like flowering sedges and buttercups) provided a measure for estimating rainfall and agricultural productivity at that period. To construct a chronology, the researchers determined radiocarbon ages on plant remains in different layers, using mass spectrometry.

Related Article

The pollen record showed a pattern of climate fluctuations, with periods of relatively moist climate vegetation alternating with periods of arid climate vegetation. One such dry spell began suddenly around 1400 and lasted until the beginning of the 20th century, the same bleak era when Weiss’s regional archaeological surveys showed villages on the Khabur Plains being emptied. Because this part of Syria is semi-arid to start with and most farmers depend on a single crop of wheat or barley grown in the moist winter months, people then, as now, were highly vulnerable to climate fluctuations. In the absence of irrigation or other technological means of adapting, they practiced what Weiss characterizes as “habitat-tracking,” or moving to areas that could still sustain agriculture.
The evidence suggests, in other words, that incompetent Ottoman bureaucrats were not solely to blame for the demise of agriculture on the Khabur Plains, nor do ingenious French bureaucrats deserve much credit for its 20th-century revival. An intervening force was climate change. And those two words, together with the idea of collapsing civilizations, may explain the intensity of the reaction and the abundance of research Weiss’s work has provoked. 
The Khabur River rises in Turkey and then flows south through eastern Syria, parallel to its border with Iraq, before joining the Euphrates. Weiss first arrived there in 1978 as a young archaeologist at Yale. He recalls being impressed by the harvested wheat stacked in “huge mounds” at the train station—“You don’t see that sort of thing growing up in Queens”—and it immediately struck him how productive unirrigated, rain-fed “dry farming” could be.
His focus was on the site now called Tell Leilan. Even before the height of the Akkadian empire almost 5,000 years ago, it grew from a rural village into one of the most important cities in northern Mesopotamia. The city walls from that era still rise above the Khabur Plains, enclosing almost a square kilometer of the ancient metropolis. The excavations Weiss directed revealed construction during the same period of grain storage and administrative facilities for collecting and shipping barley and wheat. Agriculture was being “commodified” and “imperialized” to support a central government or a distant imperial force. It was the ancient equivalent of the French train station.
But around 2200 B.C., both the major Khabur Plains settlements and the Akkadian Empire suddenly collapsed. The next 300 years have left their mark on Tell Leilan in the form of a thick deposit of wind-blown sand, with no architecture and hardly any trace of human habitation. Those centuries also survive in a desolate contemporary poem long thought to be a fictional account of the divine wrath that ended the empire:

For the first time since cities were built and founded,
The great agricultural tracts produced no grain,
The inundated tracts produced no fish,
The irrigated orchards produced neither wine nor syrup,
The gathered clouds did not rain…

In a 1993 article in Science, Weiss proposed, in effect, that the poem was nonfiction. The gods were, of course, no more to blame than were Ottoman bureaucrats in more recent centuries. But the ancient agricultural collapse was real, and the cause was an abrupt climate change. Co-author Marie-Agnes Courty, a soil scientist at the National Center for Scientific Research in Paris, documented the process of sudden drying that producedwind-blown pellets and dust.  Other researchers later confirmed an abrupt region-wide dust-spike. Weiss also linked what happened in Mesopotamia to the simultaneous failure of agricultural civilizations from the Aegean to India. This suggested that abrupt climate change had reduced rainfall and agricultural production across the region, reduced imperial revenues, and thereby caused the collapse of the Akkadian Empire. More recently, cores taken by other researchers from ocean and lake floors, cave stalagmites, and glaciers have indicated that this abrupt climate change was probably global.   
Why the climate changed abruptly then remains a mystery. The researchers at Tell Leilan found evidence of a distant volcanic eruption, in the form of rare and microscopic volcanic tephra, but regarded that as insufficient to explain a drought that lasted 300 years. Abrupt climate changes of the past were different from modern climate change, says Weiss, in at least two regards: The cause was natural, not the result of human behavior. And where we now have technological means to track and model climate change, society then “had no prior knowledge and no understanding of the alteration in environmental conditions.” 
“There are reasons for regional abandonments that are definable, observable and testable,” even when ancient peoples have left no written record of climate changes.
The 1993 Science paper was one of the first to link an abrupt climate change to the collapse of a thriving ancient civilization. But in the 20 years since then, other researchers have followed with studies implicating abrupt droughts lasting decades or centuries in a variety of other collapses, among them the ancient Cambodian Khmer civilization at Angkor, the pre-Inca Tiwanaku around Lake Titicaca, the great urban center of Tenochtitlan in ancient central Mexico,and the Anasazi in the American Southwest. Early this year, yet another article in Science reported that the collapse of the Mayan civilization coincided with prolonged episodes of reduced rainfall. Researchers have been careful not to say that climate change was the proximate cause of the collapse, but that drought slashed agricultural productivity and likely aggravated social and political unrest, ultimately leading people to abandon the area.  
But many archaeologistshave  been skeptical ofany connection between climate change and the collapse of civilizations.And,at times,the response has displayed all the loopy vehemence of the modern climatechange debate, taking denial back 5,000 years. Critics have characterized such research as “environmental determinism” and dismissed the proliferation of evidence as merely a “bandwagon” product of an intellectual zeitgeist that is, as one 2005 article in a scholarly journal put it, “suddenly sympathetic to the idea of environmentally triggered catastrophes.” That article even seemed to categorize the idea of climate-induced collapse with “Nazi-tainted eugenic theories” about Darwinian determinism. The authors acknowledged that the Akkadian and some other collapses “were of a scale and a rapidity that seemed impossible to explain by purely cultural means.”But they complained that paleoclimate researchers fail to take account of “the inseparable nature of environmental and cultural influences.”In place of farms drying up and people going hungry, the authors preferred to explain it all in terms of cascading collapses in self-organizing systems, “as easily triggered by a pebble as by a boulder.” But not, apparently, by a loaf of bread. 

“There’s a book a year,” Weiss said, incredulously, “that claims to point out the errors, lapses, gaps in data and misinterpretations” in the relationship between abrupt climate change and regional collapse. “It’s almost like, ‘Do you believe in evolution?’ They don’t ‘believe in’ the paleoclimate data and they don’t understand that the settlements that remained on the Khabur Plains after the abrupt climate change were few, tiny and short-lived.”
But paleoclimatologists have perhaps been too quick “to couple climatic and human events,” said co-author David Kaniewski. This encouraged traditional archeologists to treat the climate data “as simplistic, just because it failed, in their minds, to adequately consider and make enough room for the social and political context.” Paleoclimatologists and archaeologists could work together more closely “to study coupled natural and social systems.”
The PNAS paper notes that drought periods have become more intense and disruptive in the Middle East just in the past decade and are likely to become more common in the near future. “Interacting with other social, economic and political variables, they act as a ‘threat multiplier’” in a region that has plenty of threats to start with and that has also long exceeded the water needs of its population. One drought that lasted from 2007 to 2010 displaced hundreds of thousands of people in the Tigris and Euphrates basin, suggesting that Syria faces “the same environmental vulnerability as in antiquity.”
Nationwide, that drought drove 1.5 million people from the countryside to the cities, with no jobs or other means of support. Such underlying environmental causes rarely get much attention in reporting on the protests and violent crackdown in Syria. But they are liable to be a recurring challenge even if political and human rights issues get resolved. (Weiss has suspended his research at Tell Leilan because of the continuing crisis. But he has a research permit to drill a pollen core in a swamp alongside the nearby Iraqi border. Asked if he will be able to do the work before the permit expires, he shrugs and says, “I always go back. Let us hope the present tragedy ends quickly.”)
“In spite of technological changes,” Weiss has written elsewhere, “most of the world's people will continue to be subsistence or small-scale market agriculturalists,” who are just as vulnerable to climate fluctuations as they were in past societies. In the past, people could go elsewhere.
“Collapse is adaptive,” says Weiss. “You don’t have to stay in place and suffer through the famine effects of drought. You can leave. And that’s what the population of the Khabur Plains did. They left for refugia, that is, places where agriculture was still sustainable. In Mesopotamia, they moved to riverine communities.” But climate change is now global, not regional, and with a world population projected to exceed 9 billion people by midcentury, habitat-tracking will inevitably bring future environmental refugees into conflict with neighbors who are also struggling to get by. 
One of the most important differences between modern climate change and what he is describing at Tell Leilan, said Weiss, is that we can now anticipate and plan for climate change. Or we can do nothing. The danger is that, when there’s no longer any grain to stack up at the train station, the strategy of collapse-and-abandon may be streamlined to a simpler form: Collapse.