Four Lost Cities

by Annalee Newitz

In the early 900s, roughly a century after Jayavarman II declared himself the divine king, King Yasovarman moved the capital slightly northeast and directed Angkorians to dig an enormous reservoir known as the East Baray. Emperors generally built a baray to celebrate their ascension to the throne,1 but this one was different. For one thing, it was enormous. Measuring 7.5 kilometers by 1.8 kilometers, the East Baray was a long rectangle that held roughly 50 million cubic meters of water—the equivalent of 20,000 Olympic swimming pools. To fill it, workers built a canal that redirected the Siem Reap River into the center of Angkor.

  The East Baray would have dominated the city center’s low-density mix of temple mounds, wooden houses on stilts, and rice fields spread along the western banks of the Tonle Sap. These were early days for the metropolis; it would be two centuries before workers cut the sandstone to build the dramatic towers of Angkor Wat. Yasovarman had to order entire neighborhoods of people to abandon their homes to make way for his monumental project. He also would have had to muster the equivalent of an army to build the thing. And that was another way the East Baray was different from previous reservoirs. It was one of the first Angkorian infrastructure projects to require vast amounts of human labor drawn from all over the growing empire.

  When I visited, the East Baray had melted back into the jungle, and time had smoothed its earthen retaining walls into a gently sloping landscape thick with trees and scattered farms. It’s hard to believe this place was once a luxurious ceremonial center where Yasovarman led his retinue through lush pageants. Maybe that’s the point. Only a huge labor force could have turned this wild land into a symmetrical pool, and now that they’re gone, so is the baray. The true marvel of Angkor was its workers. And yet we rarely hear about them in historical accounts or inscriptions from the walls of Angkorian temples. They are the anonymous masses executing the will of Yasovarman.

  Every time I interviewed an archaeologist about Angkor’s city plan, I always asked who built the barays. Thinking of Rome, I imagined that it had to be slaves. But their answers were complicated because there isn’t an easy equivalence between the way labor was organized in the Roman and Khmer Empires. Angkorian inscriptions suggest that kings and other elites kept workers, but a common Old Khmer word used for this group, khñum, can refer to a wide range of roles.2 Khñum could mean temple workers who were lifetime slaves, often taken from ethnic minorities (Zhou calls them “savages”) or imprisoned during war.3 They could also be indentured workers, sometimes referred to as debt slaves,4 who endured temporary bondage as a form of taxation. Sometimes these workers are listed in temple inscriptions as property, alongside other valuables like fabrics, precious metals, and animals. Khñum, like Roman slaves, ran the gamut from manual laborers to learned scholars. These workers are also identified by many titles, including gho, gval, tai, lap, and si, which are used to mean everything from “worker” and “servant,” to “slave” and “commoner.” For simplicity’s sake, I’ll refer to them as khñum.

  The khñum debt slavery scenario sounds brutal until you consider that most capitalist cultures in the West use a similar system. In the United States, it’s not unusual for people to graduate from college with so much debt that they have to work their whole lives to pay it off. Others take on debt to pay for a house or buy a car. Though technically all of us can choose what kind of work we do to pay off these debts, it’s rare to find anyone who is doing the exact kind of work they’d like to do. Many of us feel like we’re being told to dig ditches by some distant corporate authority, or risk losing everything. Still, we keep working instead of rising up against the banks, for complicated reasons. Maybe we don’t want to rock the boat because our lives are relatively comfortable, or maybe we need health insurance to pay for a child’s hospitalization, or maybe the corporations seem too powerful to defeat. Those feelings might have kept khñum in line, too.

  Angkorian society was built on debt slavery, but the idea of indebtedness permeated every layer of public life. Inscriptions on temple and palace walls reveal that everyone in Khmer society owed something to someone.5 Even Khmer kings owed their subjects clean water, roads, and other amenities. Debt also cemented political connections between the rulers of outlying kingdoms through the Khmer patronage system. Distant elites paid Yasovarman in precious metals, fine fabrics, and more ineffable tributes like favorable trade relationships and supplies of human labor. In return, Yasovarman gave them huge amounts of land to farm. If farming didn’t strike a royal’s fancy, Yasovarman could bring him to court to enjoy the pleasures of the city. There are records of some kings bestowing weird ceremonial positions at the Angkorian court, including fan-bearer, barber, and wardrobe keeper.6 Presumably these were well-compensated sinecures that gave his allies an excuse to hang out in the Angkorian court and goof off.

  Yasovarman did more than share his spoils with his fellow aristocrats, though. He and other kings frequently left Angkor, making perilous journeys to visit the courts of their subjects. Ostensibly they did this to be worshipped, but it was also a way of acknowledging the importance of their people. The king may have been powerful, but he was nothing without all the labor power that turned Angkor into a shining cosmopolis.

  Labor was Angkor’s most valuable asset. That’s not because the Khmer had an unsophisticated economy; indeed, slave-owning societies all the way up into the 19th century often depended largely on their workers to generate wealth. Sociologist Matthew Desmond, writing about slave labor in the US South, has noted that by the time the Civil War started, “the combined value of enslaved people exceeded that of all the railroads and factories in the nation.”7 The Khmer Empire was held together by a system that normalized servitude by casting it as something people owed to their leaders, and by incorporating it into public rituals. As Miriam Stark puts it, “Leaders cajoled more than they coerced, and used display as much as military might to legitimize their rule.”8 Still, Angkor’s many attractions only existed because its people felt obliged to build them. A king who gave them nothing would eventually get nothing in return.

  Stark emphasized how unstable this whole arrangement was, partly because it depended on loyalty at every level of the social ladder. At the top was the king with his family. Directly beneath them were other noble families who lived in Angkor alongside ministers, officials, and a hereditary priest class who served as advisers to the king. And then, out in the provinces and countryside, there were semiautonomous local systems of government. Ruling alongside the king’s inspectors and local officials were usually village chiefs and a council of village elders. And beneath them all was the largest group: the khñum, made up of slaves, commoners, and servants. To expand the empire, the king needed all these groups. And because there were no rules about succession, even people at the top could fall—and those beneath them could rise. This led to wars over succession, local uprisings, and recurring cycles of chaos. Pondering the city’s eventual demise, Stark mused, “Suppose what collapsed was as much social as it was environmental or physical?”9

  The urban population explosion

  Over a hundred years after the construction of the East Baray, a new Angkorian king won a long battle over succession. In the early 11th century, King Suryavarman became Angkor’s first expansionist king, growing the Khmer Empire’s borders north into Laos and Thailand, and south to the Mekong River Delta in Vietnam. Partly he accomplished this thanks to his strong relationship with the Chola Kingdom in what is now southern India. The Chola were a source of war allies and trade throughout his reign. But Suryavarman’s success as a king was also due to his relentless focus on city-building. During Suryavarman’s reign, new cities rose along the Tonle Sap River, as well as the Mekong, Sen, and Mun, all of which were natural rivers that radiated outward from the Angkor area. The king was growing an urban network connected by water, which could be used for the transit of dignitaries, as well as trade. According to Ball State historian Kenneth Hall, the number of places in the Khmer Empire using the Sanskrit word “pura” as a suf
fix, meaning “city,” jumped to 47 during Suryavarman’s reign. Fifty years earlier, only 12 cities were recorded.10 Suryavarman’s khñum constructed roads and temples in far-flung regions, sometimes leaving a linga shrine behind as a sign of his sovereignty.

  At Angkor, Suryavarman’s passion for urbanism expressed itself in perhaps his best-known monument: the West Baray, still recognized today as one of the largest reservoirs ever built without the aid of industrial equipment. Located a few kilometers from the East Baray, it would turn the king’s palace into a jewel set between two great, long artificial seas. Looking at Evans’ lidar maps of the city, it’s easy to see that the long rectangles of West and East Baray line up neatly along an east-west axis, though the West Baray is taller. An even greater showcase of labor power than the East Baray, the West Baray measures approximately 8 kilometers by 2.1 kilometers. A person ambling along its banks at a leisurely pace could spend an hour walking from one end to the other, and circumnavigating both reservoirs might take a whole afternoon. To keep the West Baray full, workers shoveled out another canal to divert water from the Siem Reap River, which was already feeding the East Baray; they also built canals connecting it to the Tonle Sap. River water was supplemented by rain during the monsoon seasons. At its most flooded, the West Baray probably held about 57 million cubic meters of water,11 putting its size at roughly 23,000 Olympic pools. Construction took so long that it wasn’t finished until after Suryavarman’s death in the mid-tenth century. The reservoir is partly full again today, thanks to 20th-century reconstructions.

  Even when the West Baray was completed, the work of maintaining it and the rest of the city’s water infrastructure would have been ongoing. We have to imagine that Suryavarman used his patronage of many distant kingdoms to relocate thousands of people from the country to do this work. Some would have been sent by their local rulers as tribute to the king, while others did it to pay their taxes. Building the West Baray was one way to make sure that the precarious political hierarchy was stable.

  It was also Suryavarman’s way of using urban design for revisionist history. The khñum digging the reservoir had to raze all the neighborhoods, roads, and farms built around Yasovarman’s old palace enclosure, probably evicting residents in the process. And, as they reached the deepest levels of the reservoir, Suryavarman’s construction workers started to eradicate what Çatalhöyük archaeologist Ian Hodder calls “history within history.” Directly beneath the West Baray floor lie the remains of a 3,000 year-old settlement.12 We know this because the West Baray dried out in May 2004, and École française d’Extrême-Orient (EFEO) director Christophe Pottier took the opportunity to excavate there. Right below the surface, he and his team discovered telltale human burials, crushed pottery, and even bits of fabric and bronze that hinted at people living there in the early first millennium BCE. Suryavarman’s 11th-century workers wouldn’t have seen these specific remains, but it’s very likely they uncovered other evidence that Suryavarman’s reservoir was built on top of extremely ancient proto-cities. Building the West Baray meant unburying thousands of years of historical settlements, and then reburying them under millions of gallons of water.

  Suryavarman had come from outside the royal family to take the throne, and he was also the Khmer Empire’s first Buddhist king. Perhaps he wanted to signal that a new era had begun, and he did it by erasing history beneath his new artificial seas of creation.

  Debates rage among archaeologists about whether the West Baray was useful or cosmetic. Obviously this water would have been important for drinking and farming, but we can’t be sure how much of it actually reached people’s homes and farms. The city was already riddled with other canals, and each city block had its own water collection pools that look like masses of tiny pinpricks on lidar maps. So it’s not impossible that the West Baray was largely ceremonial. This interpretation would fit with evidence suggesting the West Baray was probably something of a boondoggle. To keep it perfectly in line with the east-west orientation established by the East Baray, it was built on a slope that kept the west end underwater while the east dried out. It rarely looked full. This pattern continues today, leaving the reservoir looking like a half-eaten rectangle whose waters almost never reach the ceremonial road to the king’s palace.13

  To get an idea of the size of Angkor’s reservoirs during the city’s heyday, Evans and I took a small boat into a medium-sized reservoir from an elevated walkway near the temple Angkor Thom. At one time, the now-empty East Baray would have been directly south of the waters where we floated, a light rain disturbing the lily pads nearby. Even though this reservoir was half the size of Yasovarman’s gigantic East Baray, it was sizable enough that I felt like I was boating on a natural lake. Evans peered around, pointing out the fine construction of the reservoir’s retaining walls. It brought us back to discussing the poor engineering at the West Baray.

  I wondered aloud if there were engineers beating their heads against the wall a thousand years ago when their king told them the West Baray had to be oriented east-west.

  “That would never be in an inscription,” Evans laughed.

  His offhand joke foregrounds one of the problems of studying urban life at Angkor. The roughly 1,200–1,400 temple inscriptions we have from this civilization only capture a tiny part of this city’s story. We know the spiritual traditions of Hinduism and Buddhism influenced the arrangement of the city’s east-west orientation, its layout following the track of heavenly bodies across the sky. But we don’t have any writing left by engineers or construction workers telling us what they thought about building a reservoir that was obviously uneven. More poignantly, we don’t know what khñum thought when they had to demolish their neighbors’ homes to make way for a reservoir that might have been purely decorative. Based on the work they did, however, it’s obvious that many people were willing to buy into the Angkorian system of debts and rewards. What made all that free labor worth it?

  UCLA archaeologist Monica Smith has excavated a number of ancient cities, and believes that their lure is social. Like Stark, she argues that cities grew out of ritual centers that villagers traveled to once or twice in a lifetime to meet strangers and experience novelty. But as cities grew, people settled there because they wanted that kind of excitement year-round. It wasn’t about spiritual pageantry anymore; it was about day-to-day interactions with thousands of other people. “Only cities could make that opportunity for intense interaction permanent and for a much greater range of purposes—social, economic, and political—than could ever have been envisioned for a ritual space,” Smith explains.14 Villages were enclaves of familiarity and sameness. But, she writes, “In urban settlements, unfamiliarity became the measure of human relations.” Villagers who moved to work in Angkor were their own form of attraction for other would-be immigrants. Saskia Sassen, a sociologist who studies modern cities, echoes this sentiment, arguing that cities are places for delightful chance meetings and life-changing random encounters.15

  It’s worth considering that Suryavarman’s mania for building city infrastructure may have served a purpose that he couldn’t have understood. The more he induced khñum to expand Angkor’s infrastructure, the more the city became a haven for the laboring classes. Santa Fe Institute network theorist Geoffrey West explores this idea in his book Scale,16 based on his research into today’s fast-growing cities. He’s discovered that urban populations grow faster than their own infrastructure. West has found that doubling the size of, say, a city’s water canals would more than double its population. Due to the benefits of sharing resources at high density, urbanites need about 15 percent less infrastructure than you’d expect based on population size. Put simply, urbanites multiply faster than their own urban spaces.

  At Angkor, Suryavarman’s focus on infrastructure would have enabled an urban population explosion. The barays represented merely the most ostentatious parts of a canal infrastructure that redirected rivers from the Kulen Mountains, and Zhou reported that it allowed the Angkor
ites to reap the benefits of three to four harvests every year. The water infrastructure was in excellent shape, the farms were expanding, and so were the Khmer Empire’s river-linked cities. If West is right about urban growth, we have to assume that Angkor’s population was growing even faster than the city’s footprint.

  Wealth without money

  What did all those urbanites do when they weren’t farming or digging canals? The only contemporary description of life in ancient Angkor comes from Zhou Daguan’s late 13th-century account, which he wrote in part as a travel guide to Khmer life for his Chinese audience. As a result, we learn little from him about what ordinary city dwellers thought, and a lot about how awkward it was for Zhou to use Angkorian toilets (no toilet paper!) and how hot the king’s thousands of concubines were (he claims to have examined them pretty thoroughly from a balcony overlooking their quarters).

  The Khmer themselves left behind over a thousand inscriptions on temple walls, giving us a tantalizing glimpse of how some Angkorians viewed their world. Unfortunately these writings are mostly temple bureaucrats praising their great leaders, or receipts of donations to the temples. But recently, data archaeology has given us a way to pluck the strands of people’s everyday lives from these seemingly threadbare records.

  Our ability to explore Khmer life through inscriptions in its people’s native language is relatively recent. For over a century, Westerners exploring Angkor focused on the more easily translated Sanskrit inscriptions, which consisted of poetic evocations of gods and praise for kings. Because Sanskrit was from Indic regions, these inscriptions became evidence for scholars who mistakenly believed that Khmer culture was “Indianized,” basically a carbon copy of South Asian society. It wasn’t until the Khmer linguist Saveros Pou translated the Old Khmer inscriptions that we gained full access to Angkorian history.