Four Lost Cities
Page 16
The Old Khmer language was unique to the region, and the only examples of it that we have are from Angkorian writing. Pou was fascinated by these ciphers growing up in Cambodia, and in the mid-20th century, she moved to France to study with the linguist George Cœdès. After living in Southeast Asia for several decades, Cœdès had translated most of the Angkorian Sanskrit inscriptions, and published an influential book that popularized the idea that Angkor had been “Indianized” or “Hindouisé.” Because Pou was more rooted in modern-day Khmer culture, she struck out on a different path. In the 1960s and ’70s, she focused on the idea of a specifically Khmer linguistic tradition. She immersed herself in the Khmer version of the Ramayana, and eventually compiled the only dictionary we have of Old Khmer. In the process, she had to invent a transliteration system for the ancient language, since Old Khmer has its own alphabet. Then she painstakingly translated words from Angkorian-era inscriptions into French and English for modern scholars. Without Pou’s work, and her corrective to the “Indianization” hypothesis, we would still be struggling to understand how Angkorians organized their labor force.
The Old Khmer inscriptions contain a lot of nitty-gritty details about Angkorian life that the Sanskrit verses never touch. Written in prose, they tell us about economic life, as well as the occasionally boring details of who owes what to whom. It’s revealing that temple workers wrote about elevated topics such as religion in Sanskrit, while they described everyday transactions in Old Khmer. Economic dealings were linguistically separate from the elite pursuits of kings and gurus. Taxes may have been paid to temples, but people didn’t write about spiritual matters and financial ones in the same way.
In 1900, the French explorer Étienne Aymonier dismissed inscriptions about khñum as “those interminable lists of slave names.” His attitude is reflected in a lot of 20th-century scholarship about Angkor, which focuses almost exclusively on elite life. But recently, University of Sydney archaeologist Eileen Lustig has used data archaeology techniques to study those “interminable lists” in depth, creating cross-referenced databases of every word from every inscription to look for intriguing patterns.17 One of the first patterns to leap out at her was that temple servant names were 60 percent female. Given that we see gendered divisions of labor going all the way back to Çatalhöyük, she believes that farming and other temple work was handled by women. There’s evidence to suggest Khmer women were in charge of farming outside temples as well. So when we imagine life in the Angkor Thom enclosure, for example, we have to assume it would have been dominated by women.
At Angkor, it appears that people didn’t have work weeks—they had work fortnights. An inscription from Suryavarman’s reign lists a group of temple workers, organized in two-week increments:
Slaves to provide what is due: tai Kanso; another tai Kanso; tai Kaṃvṛk; tai Thkon; tai Kañcan; si Vṛddhipura—these for the fortnight of the waxing moon. [For] the fortnight of the waning moon: tai Kandhan; tai Kaṃbh; si Kaṃvit; tai Samākula; si Saṃ’ap; si Kaṃvai.18
Each person, such as Kanso or Saṃap, has a title like tai (female slave/servant) or si (male commoner), and a work shift during either the two weeks when the moon is waxing, or the two when it is waning. Temples also organized festivals and rituals around the moon’s phases. At one temple in the 11th century, we find these instructions on what kinds of offerings the temple workers should give to the gods:
At changes in the moon’s phase: two pāda of melted butter; two pāda of curdled milk; two pāda of honey; two ’var of fruit juice; at the saṅkrānta, one thlvaṅ of milled rice; at changes in the moon’s phase, only one je of milled rice …19
Here you can see amounts are measured in things like “pāda” and “thlvaṅ,” whose sizes probably varied over time. There doesn’t seem to have been a standardized set of weights and measures throughout the Khmer Empire. In addition, scholars have debated when the holy day of saṅkrānta happened; it may have been fortnightly or annually, depending on the region.
Paychecks, such as they were, arrived fortnightly as well. Some inscriptions show temples gave out rice and other food to their khñum every two weeks. Even political cycles were measured by fortnight, with some inscriptions suggesting that heads of state expected major economic transactions to coincide with the lunar cycles, including tax payments and land grants. We have to imagine Angkor running on a schedule where the work fortnight was deeply connected to festival days and statecraft. Temple staffs often included astronomers who plotted the course of the moon, keeping track of work shifts and festival days. They also decided which days during the fortnight were most lucky, and that no doubt had some bearing on when people made major purchases or donations to their local temples.
The fortnightly pay schedule raises a question about how laborers survived. If khñum were fed rice based on two-week shifts each month, what did they do for food during the other two weeks? We know that higher-ranking temple workers were allowed to eat some of those tasty-sounding fortnightly offerings to the gods, and local elites also got leftovers on feast days when people poured gifts into the temple coffers. Possibly khñum could skim extra rice off these offerings, too. But more likely, they simply went back home and lived with their families during their weeks off. At Angkor, that would have been particularly easy because the walled enclosures around temples were residential neighborhoods.
We know this based on evidence written into the earth. Evans’ lidar scans showed that the earthen mounds of house foundations can be found in neat grids around temples. Curious to know more, Alison Carter dug up one of these mounds inside Angkor Wat’s walls in 2015. She discovered what appears to be the remains of a brick stove, complete with ceramic vessels for cooking.20 Chemical analysis revealed remains of pomelo fruit rind, seeds from a relative of the ginger plant, and grains of rice. This is what archaeologists call “ground-truthing,” and it’s further confirmation that temples were at the center of neighborhoods full of people engaging in trade, farming, textile manufacture, and other domestic tasks. The people who lived there were paying their taxes in labor, at least part of the time. But they were also part of secular communities. In these temple neighborhoods, women farmed right alongside monks who composed Sanskrit poetry about their king.
These neighborhoods may not have been exactly like their counterparts beyond the temple walls, but they give us a sense of what life was like for people like Kanso and Kaṃvit, who reported for duty every other fortnight. As Stark warned, these places were what made the empire great, but they were also its vulnerability. Keeping people in line is a lot harder than keeping water contained in a baray.
There was another strong signal that emerged from Lustig’s inscription data, though it might be more accurate to call it a lack of a signal. None of the economic records in Old Khmer mention money. At the same time, inscriptions show temples selling big-ticket items: roughly 75 percent of exchanges recorded are for large plots of land, 18 percent are for khñum, and 7 percent are for services related to marking land boundaries, sort of the Angkorian version of land value assessment. The few remaining items would have been temple supplies.21 It’s not as if Angkorians didn’t know about money yet. They traded with other kingdoms that used coins, and had plenty of metal if they wanted to mint their own. There’s also evidence that earlier Khmer cultures may have used money. Lustig identifies a couple of pre-Angkorian inscriptions22 in which the scribes used specific units of silver as a way to place value on a rice field and an enslaved woman. The Khmer also had advanced math (including the revolutionary concept of zero), and sophisticated ways of measuring debts, interest rates, and exchanges.
After Jayavarman II founded Angkor, however, we never see goods or people valued in silver or any other unit of exchange. So what was the equivalent of Angkorian cash? Maybe, say some historians, there was a widely agreed-on list of valuable items that could be used instead of money.23 In some early 12th-century exchanges, we see that a piece of land was sold for “2 gold rings,
1 silver bowl, various units of silver … 1 vessel, 2 water vases, 5 plates, 3 utensils, 1 candle holder, 20 cubits of fine fabric, 2 fast oxen, 2 lengths of new fabric 10 cubits long, and 3 goats.” A khñum and her four children were sold for “60 garments.” Generally we see worth measured with lists of items like these, a combination of animals, people, metals, and finely made household goods. Money may not have been needed, because each transaction could be cobbled together on a case-by-case basis from standard luxury items. What this suggests is that elite wealth was measured in land and the tools (including people) required to work it.
Everyday financial dealings in the khñum class were different. When Zhou visited Angkor in the late 13th century, he described how the city streets were lined with women selling food and other goods from blankets spread on the ground. Customers used coins from China and other places, as well as rice, grain, and fabrics as forms of money. It appears that inexpensive items might be had for cash, especially in these more informal marketplaces. Of course, there’s always the chance that wealthy Angkorians did use money, and the people writing inscriptions thought monetary values were so obvious that they didn’t need to record them. Another possibility, suggested by Zhou’s observation that women conducted all trade in the Angkor marketplace, is that dealing with money was part of women’s work and therefore not worth noting.
Taken together, the temple inscriptions and Zhou’s observations reveal something profound about the Angkorian state. There doesn’t seem to be a centralized form of control over economic exchanges. Local kingdoms could set their own variable prices for land and khñum at temples, while ordinary people got by on a combination of barter and foreign coins. The system seems unwieldy for those of us brought up in fully monetized societies, but it makes sense for a civilization where land and labor power were the most valuable items a person could own. Certainly Angkorian leaders loved their gold and silver—and probably they traded precious metals with foreigners who coveted the stuff—but they were not hoarders of money. Instead, they were hoarders of taxpayers who could be used to transform land. Most Angkorian kings got rich on endless supplies of free labor from people like tai Kaṃvṛk and tai Thkon.
The fragility of stone
Nearly 200 years after Suryavarman’s labor force broke ground on the West Baray, King Suryavarman II took the throne. Like his namesake, he did not inherit his position but instead fought for it in a bloody battle of succession. A prince from an outlying kingdom in what is now Thailand, he took to Angkorian life and left behind one of the city’s most famous monuments: Angkor Wat, a mountainous temple compound that lies to the south of the two great barays. Suryavarman II was not an expansionist king, nor a particularly great warlord, but he’s remembered because he did an especially good job of maintaining the canals and roads that connected Angkor to other parts of the empire and lands beyond. It probably didn’t hurt that Suryavarman II also made sure to include a lot of very flattering pictures of himself in Angkor Wat’s many dramatic reliefs. He’s the first Angkorian king to depict himself in art, and it’s hard to forget the image of him sitting in his palace on soft rugs, while a bunch of servants hold several multi-tiered parasols over his head.
Suryavarman II’s glamorous self-representation was entertaining, but I wanted to know more about the people holding parasols. That’s why I spent a quiet morning with Damian Evans visiting the place where Angkor Wat was made. The walled complex at Beng Mealea, located about 50 kilometers northeast of Angkor, was another one of Suryavarman II’s many construction projects. Today it is rarely visited, and restoration efforts have just begun on its nested square galleries, libraries, and canals that flowed through the entire temple and the palace at its heart, forming a floor of gleaming water between ornate walkways. Long ago, it was flanked by a baray that was twice the size of the complex. But today the reservoir and deep moat around Beng Mealea are piled with the gargantuan stones that once formed its walls. Access to the palace is difficult, and Evans led me through the rear western gate, careful to stay on the path because we’d been warned about land mines in the jungle.
Around us the landscape was bumpy with regularly spaced mounds. “You can see there’s an unnatural topography here,” Evans remarked, referring again to the concept of anthropogenic geomorphology, or ways that humans shaped the land. We were seeing all that remained of the commoners’ neighborhood of wooden houses that once surrounded Beng Mealea. When we reached the eroded walls of the complex, Evans paused. “We are in the middle of a dense downtown, surrounded by temple staff housing,” he said simply. I imagined the towering trees vanishing around us, replaced by roads lined by tidy, thatch-roof houses on stilts, smoke coming off the stoves located underneath their elevated living quarters. Children shouting and farm animals grumbling.
Then we mounted sturdy wooden stairs that led into the temple complex over mossy, collapsing corridors. Arched ceilings sheltered a gallery whose long windows were fitted with stone balusters that acted as blinds. Each baluster had been lathed into a rippling, fluted shape that cast complex shadows. At the center of the complex, stone floors had been floated on columns so that water could run beneath them.
Evans joined me as we climbed past rocks that gushed from doorways, the wooden stairs creaking slightly beneath our feet. When we reached the top of an outer retaining wall, I looked down into a deep stone canal that was once part of Beng Mealea’s outer moat. Today it appears to flow with blocks the size of elephants, as if time had somehow washed the structure into its own dried-out waterways. It was still early morning, and the air was cool under trees that grew from piles of stately rubble. All we could hear were the songs of birds and insects. As Evans unrolled a lidar map of the compound, I rebuilt the ruins in my mind. Thousands of khñum bustled past us, entering or leaving via long promenades that faced the four cardinal directions. They shuttled between their neighborhood farms beyond this moat and the elites who lounged at the center of Beng Mealea’s many causeways. They were doing all the usual tasks of temple maintenance, tending farms, and pruning the gardens whose scented flowers were a key part of many fortnightly rituals. But this area was so densely populated because it was no ordinary provincial outpost.
Beng Mealea specialized in an industry that was of particular interest to Suryavarman II. It was strategically located in a spot at the nexus of two major roads and several waterways. One road connected to sandstone quarries in the Kulen Mountains to the north, and the other to an iron processing center at the Preah Khan temple complex in Kompong Svay to the west.24 Both Beng Mealea and Angkor Wat are built from Kulen sandstone, which I could see piled around me now as if it had just arrived from the mountains. Here at Beng Mealea, khñum worked in shipping, receiving, and sometimes goods processing, too. As sandstone came into the complex along the canals, they cut it down into blocks and kept it for use at their own temple complex or sent it along to Angkor. When iron came from Preah Khan, they put it on barges for the long journey down the Khmer Empire’s artificial waterways to Angkor. It’s likely they did the same for rice and other goods heading into Angkor from the productive hinterlands. Suryavarman II, himself a child of one of these provincial cities, would have been keenly aware of the key role Beng Mealea played in his empire.
Some researchers believe that Beng Mealea was a beta version of Angkor Wat. It’s the first place where engineers experimented with new kinds of archways and high walls that are ubiquitous at Angkor Wat. There are also some similarities in urban patterning. Evans showed me how the lidar reveals symmetrical city blocks around Beng Mealea that look extremely similar to those at Angkor. Here in the hinterlands, though, the neighborhoods show a greater variety of shapes and layouts. Still, Angkorians would no doubt have felt at home on the streets here, with their row houses separated by fish ponds. It was also thanks to lidar that we know what Beng Mealea’s chief business was. After the technique revealed deep sandstone quarries in the Kulen Mountains, it was easy for researchers to connect the canal routes and see
where all the sandstone to build Angkor Wat and Beng Mealea had come from.
There’s still a lot we don’t know. The lidar surveys revealed two previously unseen structures that nobody has been able to explain so far. The first is a complicated rectangular maze pattern dubbed the “coils,” “spirals,” or “geoglyphs.” These were first spotted outside the moat at Angkor Wat during the 2012 survey, but the 2015 survey revealed similar coils outside the enclosures at Beng Mealea and Preah Khan. At first glance they appear to be waterworks, but Evans and his colleagues dismissed that idea because they are too shallow and are cut off from the city’s general waterworks. The reigning hypothesis is that these rectilinear coils were specialized gardens for growing plants used in temple rituals. The often-flooded channels might have contained lotus, while Evans and his colleagues write that the raised areas could have supported “aromatics such as sandalwood trees.”25
More mysterious are the so-called mound fields found near some of Angkor’s largest reservoirs and canals. Unlike the residential mounds excavated by Carter and her colleagues, these mounds aren’t packed with ceramics and food remains. They are just mounds, clearly the foundations for an elevated structure or structures. Their locations suggest that they may have been related to the city’s waterworks, but of course correlation does not equal causation. The coils and mound fields are reminders of how much we still don’t understand about how the ancient Khmer built their cities.
Suryavarman II and his predecessors were nothing without all those commoners and khñum cutting sandstone, smelting iron, harvesting rice, and shipping it back to the capital. When I visited Angkor Wat, it was hard to see the temple’s ornate towers as the legendary Mount Meru, center of the cosmos, towering over the glistening waters of creation. Instead, I kept seeing piles of stone created in the quarries and workshops of thousands of unpaid laborers. I went inside its pale walls with a swarm of tourists, and paid to leave an incense offering for the spirit of the city, who inhabits a gold-draped Buddha sitting on an ancient pagoda in the temple complex. As I perused the famous reliefs showing Suryavarman II going to war with the Cham in today’s Thailand, I looked mostly at the bodies of the men who bore his litter. The more infrastructure Angkor’s khñum built, the more responsibility their king bore to maintain it. And as Stark warned, the patronage and debt systems were always on the verge of toppling.