(1/3)
Big disclaimer: I'm not an expert in Angkor so I am not familiar with all the latest research. Angkor is very understudied, far more understudied than Rome, and we are learning so many new things about it literally right at this very moment. Our existing theories are constantly being both confirmed and challenged, so if anyone knows of more updated theories I would be glad to hear them. I also know even less about Rome than I do about Angkor so I can’t do a comparison, I can only give a description of Angkor.
Angkor developed over several centuries and it did not stay the same all that time. In fact, much of what we associate with Angkor, such as the Angkor Wat temple complex, only appears later in the settlement’s existence. We can divide Angkor’s history into roughly two periods: the first period is from the 9th to the 10th/11th century. The 11th century sees some significant changes in Angkor, including in the realms of social, political and economic organisation, and we see some significant changes in Angkor’s landscape as well.
Angkor was a very different kind of ‘city’ from what is familiar to us today. Nowadays, when we think of a city, we think of a densely populated, specialised settlement whose inhabitants hold non-agricultural jobs. However, Angkor was more accurately described as ‘low-density agrarian urbanism’ where farming took place within the settlement itself.
In an 2016 Guardian article, archaeologist Damien Evans described his team's LiDAR findings thus:
“I’m reluctant to use the word ‘city’,” says Evans. “Angkor doesn’t follow the usual pattern of an ancient walled city with a clearly defined edge. Instead, we discovered a very densely populated downtown urban core, covering an area of 35-40 sq km, which gradually gives way to a kind of agro-urban hinterland. It slowly dissolves into a world of neighbourhood shrines, mixed up with rice fields, market gardens and ponds.”
That ‘world of neighbourhood shrines, mixed up with rice fields, market gardens and ponds’ was what passed as 'urban sprawl' in Angkor.
As Angkor does not really conform to our usual definition of a city, defining its borders and land area is also difficult. Evans et al. suggested in 2007 that rather than use population density or occupation as a means to define Angkor’s extent, we should instead use its infrastructure network.
Angkor is often described as a 'planned city', not so much because everything was nice and neat, but because it was the infrastructure network that led development. In some other cities, a community will develop around, say, a mine, and then infrastructure is built to link that community to other communities. In Angkor it was the opposite - if the state wanted to develop a plot of land, it would extend the state-run infrastructure to that location to make it suitable for habitation and farming, and that would in turn attract farmers and other would-be residents.
All in all, the Greater Angkor Region covers about 3,000 square kilometres, within which some 1,000 to 1,500 square km of temples, rice paddies and housing were bound together by the infrastructure network (this figure keeps getting bigger as more of Angkor gets uncovered!). To understand this infrastructure network, it is easiest to see it as a water management system, with the disclaimer that most parts of the system served multiple purposes, being used for water management, transport, religion, housing etc.
9 TO 11C ANGKOR WATER MANAGEMENT SYSTEM
Angkor was challenged by extreme climate conditions. Like much of SEA, the Angkor area saw 2 seasons - there was a dry season that lasted approximately from December to April, during which there would be little to no rain, and a rainy season that lasted approximately from May to November, during which it rained buckets and the area got flooded.
Angkor seems to have been built in such a way to ameliorate these extremes. During the wet season, water was rerouted so as not to cause devastating floods, and also stored in reservoirs, great pools and temple moats. During the dry season, this stored water was doled out until the wet season arrived to replenish supplies.
Angkor is built between 2 natural features. To the north and northeast are the Kulen and Khror hills. These hills are the point of origin of several rivers that flow south and southwest. Their end point is the huge Tonle Sap lake that lies to the south and southwest of Angkor.
Angkor’s water management system seems to have begun in the southeast, in the area known as Roluos, home to the first royal palace of Angkor. Today, to its north is the Roluos river that, like most other rivers in the area, flows from northeast to southwest, but which abruptly changes course to flow south to the Roluos area. This implies that a canal was cut to divert the river into the Roluos network, which consists of a baray (reservoir) called the Indratataka, various channels and moats around the Bakong ancient city, a canal that runs south to the Tonle Sap lake and a canal that runs to the west.
Subsequently, Angkor embarked on more major building projects that extended the water management system to the northwest. This was done primarily by capturing the flow of more and more rivers further and further to the north and west. Eventually, this led to a water management system composed of 3 zones.
The Kulen and Khror hills functioned as a catchment area that collected rainfall, especially during the wet season. This water flowed down several large rivers down the hills and into the Angkor north/northeast zone. This was a water collection and flow management zone. The zone consisted of a complex grid of long linear channels and embankments with numerous right-angle turns and cross channels. North-south channels carried the water southwards, while the east-west embankments served to slow the water to a manageable rate while also reducing erosion. East-west embankments carried the water to the west, eventually, more channels and embankments carried it back to the east before finally turning it south to the main reservoirs of the central zone.
These channels and embankments were enormous. Some of the channels were 40 to 60m wide, about the length of an Olympic swimming pool, and were lined with stone. There were embankments on either side, built with clayey sand, which was easily available on the Angkor plain. These embankments, together with the east-west embankments, were not only tall enough to rise above the flood waters during the wet season but were very wide as well (unfortunately I don’t have exact numbers). It has been theorised that they were also used as roads, and the very wide ones even had houses built on them. During the wet season, there would likely be large pools of water on the northern side of the east-west embankments.
From this zone the water was channelled into the central zone for storage. The centre of this zone is dominated by 2 huge barays. The East Baray was built in the late 9th/early 10th century. It is 7km by 1.8km and could hold over 50m cubic metres of water, enough to fill 20,000 Olympic swimming pools. This was dwarfed by the 11th century West Baray, a reservoir 8km by 2km, ringed with embankments over 10m high and 120m wide. This held over 155m cubic metres of water, enough to fill over 60,000 Olympic swimming pools. In its centre was a temple, the West Mebon, which contained a 6m-long reclining bronze statue of Vishnu.
Finally, the water was channelled to the southern disposal zone on the shore of the Tonle Sap. Here, a great embankment ran from northwest to southeast, collecting water on its northern side for irrigation. Excess water was channelled even further south into the Tonle Sap.
These are just the broad outlines of the water management system and its impact on the landscape. The actual system was far more complex with features such as spillways (devices made of stone blocks that could release excess water from reservoirs into canals) ensuring its smooth functioning. The system also included many minor canals to distribute the water from the main features mentioned above, which we will explore in the next section.