More than a temple
When people say “Angkor”, they often mean the temple of Angkor Wat. Historically, Angkor was a capital region: a spread of temples, palaces, houses, rice fields, reservoirs and roads that grew over several centuries. The word is thought to derive from the Sanskrit term for city, and the Khmer kings’ capitals here had Sanskrit names such as Hariharalaya and Yasodharapura. This article describes the Angkor city plan as a whole, and leaves the temple itself to other articles in the cluster.
A city that moved
Angkor was not one city built once. The royal centre moved several times within the region. Jayavarman II’s successors based themselves at Hariharalaya, near modern Roluos, in the ninth century. Around 889, Yasovarman I established Yasodharapura, the first city at the Angkor site proper, with Phnom Bakheng at its centre. A brief interlude saw the capital at Koh Ker in the tenth century, after which it returned. Later kings built new temples and modified the city, and Jayavarman VII’s walled Angkor Thom is the last major phase.
Because the centre moved and each king built a new state temple, Angkor looks like a series of overlapping capitals, not a single planned city. Angkor Wat itself, built in the early twelfth century, lies outside the walled enclosure of Angkor Thom and was constructed before it.
Moats and enclosures
Angkor’s temples were set within rectangular enclosures bounded by walls and moats, and many of them were arranged on an axis. Angkor Thom is the most famous enclosure, roughly three kilometres on each side, but many other temples had their own moated precincts. The enclosures were mainly sacred and administrative spaces. Most ordinary people lived outside them.
Reservoirs, canals and the water system
The most distinctive feature of Angkor’s plan is the water system. The Khmer built huge rectangular reservoirs, called barays, with embankments of earth. The Indratataka at Roluos and the East Baray, created in the late ninth and early tenth centuries, were followed by the West Baray in the eleventh century, which is roughly 8 km long and over 2 km wide. Smaller reservoirs and temple moats added to the system, and canals and dykes diverted the Siem Reap river and other streams, linking the reservoirs and carrying water through the landscape.
What was it all for? The French scholar Bernard-Philippe Groslier argued in the mid-twentieth century that Angkor was a hydraulic city, with rice irrigation as the economic basis of royal power. Later work questioned this. Some researchers pointed out that the barays may have had symbolic meaning, as the ocean around a temple mountain, or ritual roles, and that not all fields would have been watered directly. Current thinking often combines the ideas: the system may have helped to control floods, hold water for dry periods, supply people and fields, and demonstrate royal power. The exact balance is still discussed.
Roads and connections
Angkor was linked to the wider Khmer realm by roads. Raised causeways and embankments connected temples within the region, and longer roads ran out to provincial centres. A road to the east is associated with the route towards Phimai, in what is now Thailand, and other routes extended in different directions. Inscriptions from the reign of Jayavarman VII mention rest houses along roads and a network of hospitals, which suggests an effort to connect and care for the realm, though the numbers recorded in inscriptions should not be taken as exact.
Where did people live?
Most Angkorians lived in simple wooden houses, most likely raised on posts and with thatch roofs, which have not survived. The sites of settlements are known mainly from small mounds, ponds and the scatter of pottery. The pattern was low-density: clusters of houses, ponds and gardens separated by rice fields, in contrast to the closely built cities of the ancient Mediterranean.
Mapping work in the early twenty-first century, led by teams from Australia, Cambodia and France as part of the Greater Angkor Project, showed that the urban area of Angkor extended over many hundreds of square kilometres, with one estimate placing the extent of the urban complex at around 1,000 square kilometres. The sheer size makes Angkor one of the largest pre-industrial settlement complexes known.
What LiDAR changed
Airborne laser scanning, known as LiDAR, can see through forest cover and map surface features. A LiDAR survey in 2012 covered parts of the Kulen plateau and the Angkor area, and later surveys extended coverage. The results revealed a regular pattern of roads, canals, house-platforms and temple sites, including previously unknown features and the remains of a planned city on Phnom Kulen, associated with early Khmer rulers. Radiocarbon dating of excavated materials has helped to place some of these sites in sequence. These results reinforced the view that Angkor was a dense network of infrastructure and settlement, though they do not by themselves date every feature or say who lived there.
How many people?
There is no ancient census, so estimates of the population come from the area, the number of house-mounds and other proxies. Figures proposed by scholars vary widely, from the low hundreds of thousands to numbers around a million, and the upper figures are disputed. The honest answer is that Angkor was probably very large for its time, but its population at any given period is uncertain.
A city built for gods and kings
The Angkor city plan was shaped by religion as much as by practicality. Temples, reservoirs and roads were royal and religious acts, and each ruler added his own works. This explains the layering of the landscape, and why some sites look like part of an ordered city while others seem to be scattered. For anyone trying to understand Angkor, it helps to think of it as an extended, managed landscape in which temples were the stone parts of a much larger and mostly wooden living city.