angkor_city.md

world/regions/angkor_khmer/page/angkor_city.md

Angkor as the Hydraulic City — Deep Research (region page 4, angkor_khmer / angkor_city zone)

Research target for CH_05 SPREAD excursion (sites kra_traverse, angkor_wat, bayon_taprohm,

west_baray, myson_champa). Frame year for the game = 1200 CE. This file covers the Angkor city

core (hydraulic system, monuments, the devaraja institution, the 1200 era-state, and forward

context to the 1431 Ayutthaya sack). The Champa/war surfaces are the elevated-care lane — handled in

detail here only where they touch the Angkor core (the 1177 Cham war); the sanctuary/modern-Cham

material is a separate research target.

BINDING FRAMING FOR THE BUILD (from the task brief, not re-derived here):

• Anti-fringe hard line: Angkor was built by the Khmer. No lost-race / ancient-aliens / "too advanced

for its builders" framing anywhere. The hydraulic city is real, dated, inscribed archaeology — cite it.

• §17.1: communities are never combat targets; antagonists route through apparatus/brokers.
• Reveal ceiling: CH_05 is pre-Ch-38 — no House-of-Velheim name/pattern/agent-noun who-frames.
• Do-not-invent: chapter/numerical/naming canon is Josh's. This file DERIVES and CITES; where a claim

is contested I flag it "CONTESTED"; where a claim is pseudo-archaeology I flag it "FRINGE — REFUSED".

Local-corpus note: the repo's RCM Ch 5 section (_source/01_.../T1_Research_Corpus_Master) is

Sumatra-Java-focused (Miksic Borobudur, Coedes Indianized States, Batak). Angkor is confirmed as the

Ch 5 mainland-Southeast-Asia SPREAD excursion in T1_World_Geography_Master v2.3 §Coverage-architecture

and the traversal line (§ around L981). No deep Angkor substrate exists in-repo yet — this file is that

substrate. Coedes, "The Indianized States of Southeast Asia," is already an owned Tier-1 source for Ch 5

and is the standard scholarly frame for the Khmer state; recommend cross-listing it to this region page.

---

1. THE HYDRAULIC CITY — the baray system and Greater Angkor archaeology

1.1 The core claim (real archaeology, not myth)

• Angkor was the most extensive urban complex of the preindustrial world — an engineered low-density

city integrated by a vast water-management network of canals, embankments, reservoirs (barays),

ponds, moats, rice fields and radial roads. The monuments sat inside a dispersed settlement, not a

walled compact city. SOURCE: Evans, Pottier, Fletcher et al., "A comprehensive archaeological map of

the world's largest preindustrial settlement complex at Angkor, Cambodia," PNAS 104 (2007),

https://www.pnas.org/doi/10.1073/pnas.0702525104

• Greater Angkor covered roughly 1,000–1,200 km² — "the size of New York City's five boroughs" —

with a circuit of great monuments ~20–25 km out from the center integrated into a single operational

system. SOURCE: PNAS 2007 (Evans et al.); summarized in IFLScience,

https://www.iflscience.com/at-its-peak-the-ancient-khmer-city-of-angkor-dwarfed-the-great-cities-of-europe-59603

1.2 The barays (the hand-cut reservoirs — the "hydraulic" in hydraulic city)

• West Baray: ~7.8 × 2.1 km, the largest baray at Angkor and one of the largest hand-cut

reservoirs on Earth; current max capacity ~53 million m³, retained by earthen dikes ~12 m high. At

its center is the West Mebon, a Hindu island-temple. Construction begun 11th c. under

Suryavarman I, completed under Udayadityavarman II. SOURCE: West Baray / West Mebon, Wikipedia,

https://en.wikipedia.org/wiki/West_Baray ; https://en.wikipedia.org/wiki/West_Mebon

• West Mebon bronze Vishnu: in 1936 the West Mebon yielded the largest known bronze in Khmer art —

a fragment (head, upper torso, two right arms) of a reclining Vishnu that complete would have run

~6 m long; now in the National Museum, Phnom Penh. SOURCE: West Mebon, Wikipedia (above); Cambodianess,

"Where Vishnu Once Floated," https://cambodianess.com/article/where-vishnu-once-floated-the-revival-of-west-mebon-temple

• East Baray (built by Yashovarman I, late 9th c.): ~7.8 km × ~1.8 km, covering ~12.5 km²,

capacity ~37 million m³. Its island-temple is the East Mebon (mid-10th c.). SOURCE: search synthesis

(multiple), Wikipedia East Baray/East Mebon.

• Indratataka: the earliest great baray (late 9th c., Indravarman I, at the Roluos/Hariharalaya

group); the East Baray was ~5× larger than the Indratataka. SOURCE: search synthesis; renown-travel /

angkor-travels reservoir pages.

• Srah Srang ("the royal bath"): a smaller reservoir built under Rajendravarman II (mid-10th c.),

~0.3 km², capacity ~1 million m³; still holds water and has a laterite landing terrace. SOURCE: search

synthesis (Wikipedia Srah Srang).

• The moat of Angkor Wat (~1.5 × 1.3 km rectangle, ~190 m wide) is itself part of the water system

and helps stabilize the temple's foundations by keeping the water table constant — a real engineering

function, not only symbolic. SOURCE: History Skills, "Angkor Wat,"

https://www.historyskills.com/classroom/year-8/angkor-wat/

1.3 The Greater Angkor Project + LiDAR (how we know)

• The Greater Angkor Project (GAP) — a University of Sydney / APSARA / EFEO collaboration led by

Roland Fletcher and Damian Evans, building on Christophe Pottier's (EFEO) southern-Angkor

mapping — produced the comprehensive archaeological map (radar + ground survey) published in PNAS

2007. SOURCE: Evans et al., PNAS 2007 (above); Fletcher & Pottier work summarized at Virtual Angkor,

https://www.virtualangkor.com/waterclimate

• Airborne LiDAR, April 2012: a consortium organized by Damian Evans imaged ~370 km² in ~20 hours

of helicopter flight, covering Angkor, Phnom Kulen and Koh Ker. It revealed the buried street grid,

house-mounds, ponds and field systems beneath the forest, and laid bare a 9th-c. city on Phnom Kulen

known from inscriptions as Mahendraparvata. SOURCE: Evans et al., "Uncovering archaeological

landscapes at Angkor using lidar," PNAS 110 (2013), https://www.pnas.org/doi/10.1073/pnas.1306539110 ;

Science, "The Hidden City of Angkor Wat," https://www.science.org/content/article/hidden-city-angkor-wat

• A second larger LiDAR campaign (2015, Cambodian Archaeological Lidar Initiative) extended coverage;

machine-learning + LiDAR population modelling later put the Greater Angkor peak population near

~700,000–900,000. SOURCE: Klassen et al. population study, reported ArtNet News,

https://news.artnet.com/art-world/angkor-wat-ancient-population-1966958

1.4 The hydraulic-city THESIS and its debate (flag as CONTESTED — historiography)

• Groslier's thesis (1979): French archaeologist Bernard-Philippe Groslier argued Angkor was a

"cité hydraulique" — a city defined and sustained by a vast managed water network of canals and

barays that irrigated rice for a huge labor population, and that was ultimately overwhelmed by

over-exploitation of the soil (deforestation, erosion, siltation). SOURCE: Groslier, "La cité

hydraulique angkorienne: exploitation ou surexploitation du sol?" BEFEO 66 (1979), 161–202; English

translation on ResearchGate,

https://www.researchgate.net/publication/282332249 (Groslier hydraulic-city translation)

• CONTESTED — the irrigation-vs-ritual debate: From the 1980s, critics (notably W. J. van Liere and

Robert Acker) argued the barays were primarily symbolic/religious (cosmic oceans around temple-

mountains) and/or for urban supply, NOT large-scale irrigation, partly because some barays appeared to

lack outlets. This was a genuine, decades-long scholarly dispute. SOURCE: Virtual Angkor, Water &

Climate, https://www.virtualangkor.com/waterclimate ; overview in The Past, "The secrets of Angkor

Wat," https://the-past.com/feature/the-secrets-of-angkor-wat-how-archaeology-is-rewriting-history/

• Resolution trend: Christophe Pottier and the GAP mapping showed the great barays DID have inlets

and outlets and were tied into a channel-and-embankment network — supporting at least the functional

(water-distribution) half of Groslier's thesis. Modern consensus: the network was real and

multi-functional (supply, flood-buffering, some irrigation, and symbolic), and its brittleness under

climate stress contributed to decline. SOURCE: Fletcher, Pottier, Evans, Kummu, "The development of

the water management system of Angkor: a provisional model," Antiquity/World Archaeology (2008),

https://www.researchgate.net/publication/228667792 ; ScienceDaily 2007,

https://www.sciencedaily.com/releases/2007/09/070905145001.htm

• HONEST FRAMING FOR THE BUILD: present the hydraulic city as REAL and Khmer-engineered (that is not

contested); present the *purpose-of-the-barays* question (irrigation vs. ritual vs. supply) as a live

historiographic debate that modern survey has largely tilted toward "functional + symbolic, both."

Do NOT stage the water system as mysterious or beyond its builders.

---

2. ANGKOR WAT — construction and the orientation debates

2.1 Construction facts

• Built by Suryavarman II (reigned c. 1113 – c. 1150), begun early in his reign (~1113) and largely

finished by ~1150 — roughly 30–37 years of sustained work. Dedicated to Vishnu (Angkor Wat is

Vaishnavite, unusual among Angkorian state temples, most of which were Shaivite). SOURCE: Angkor Wat,

Wikipedia / New World Encyclopedia, https://www.newworldencyclopedia.org/entry/Angkor_Wat ;

History Skills (above).

• Stone and transport: ~90% of the sandstone came from quarries at the foot of Phnom Kulen,

~40–50 km NE (the main quarry cluster around O Thma Dab; 50+ quarries identified). Blocks were

floated down purpose-dug canals to the Siem Reap River and up to the site — a 2012 study by Etsuo Uchida

& Ichita Shimoda re-routed the accepted transport path via these canals. SOURCE: Uchida & Shimoda,

"Quarries and transportation routes of Angkor monument sandstone blocks," Journal of Archaeological

Science (2013), https://www.sciencedirect.com/science/article/abs/pii/S0305440312004384 ; Live Science,

https://www.livescience.com/24440-angkor-wat-canals.html

• Labor: estimates of 5,000–50,000 workers over decades, organized by corvée and fed by rice

surpluses — i.e., the hydraulic/agricultural base and the monument-building are linked. Stone was

fitted without mortar (precise joinery; iron cramps and dowels at joints). SOURCE: History Skills

(above); Missed History, https://www.missedhistory.com/article/how-angkor-wat-was-built

• Scale: Angkor Wat is the largest religious monument in the world by land area (~162–200 ha within the

moat). SOURCE: standard (Wikipedia Angkor Wat).

2.2 The west-facing orientation debate (flag CONTESTED)

• Angkor Wat faces WEST, while nearly every other major Khmer temple faces east. West is the

direction of death/the setting sun in the tradition, which is the crux of the debate. SOURCE: Angkor

Wat Equinox, Wikipedia, https://en.wikipedia.org/wiki/Angkor_Wat_Equinox ; Freeman & Jacques cited

therein.

• Funerary/mausoleum reading: the dominant scholarly interpretation is that the western orientation

marks Angkor Wat as a funerary temple — a mausoleum for Suryavarman II, apotheosized as Vishnu after

death. Supporting details: the bas-reliefs of the third gallery are read counter-clockwise

(prasavya), the direction of Hindu funerary ritual. SOURCE: Angkor Wat Equinox (Wikipedia, above);

New World Encyclopedia (above).

• Vishnu-association reading: Freeman & Jacques counter that several Angkor temples depart from

eastern orientation and that Angkor Wat faces west simply because it is dedicated to Vishnu, who

is associated with the west — no funerary explanation strictly required. SOURCE: Freeman & Jacques,

"Ancient Angkor," cited in Angkor Wat Equinox (Wikipedia, above).

• Archaeoastronomy reading (CONTESTED, partly popular-fringe-adjacent): Robert Stencel, Fred

Gifford & Eleanor Morón (1976, "Astronomy and Cosmology at Angkor Wat," Science) and later **Eleanor

Mannikka** ("Angkor Wat: Time, Space, and Kingship," 1996) argued the temple encodes solar/lunar

alignments and a calendrical numerology — e.g., at the spring equinox the sun rises directly over the

central tower seen from the western causeway, and the Churning-of-the-Sea-of-Milk figure counts encode

solstice-to-equinox day counts. SOURCE: Stencel/Gifford/Morón, Science 193 (1976); Mannikka (1996);

summarized at Angkor Wat Equinox, Wikipedia (above), and IU ScholarWorks archaeoastronomy paper,

https://scholarworks.iu.edu/journals/index.php/sdh/article/download/22846/29089

• CONTESTED caveat: the equinox-alignment and numerology claims are debated — some are robust (the

equinox sunrise-over-tower is widely accepted and observable), others (fine numerological encodings)

are viewed skeptically by archaeologists and are heavily amplified by "sacred geometry"/mystery-site

blogs. FRAMING: the equinox alignment is fair to stage as a real, observed feature; treat the deep

numerology as an interesting-but-contested scholarly hypothesis, and keep it away from any

ancient-mystery / lost-knowledge register (that tips into the refused fringe lane).

2.3 Later religious history of the temple (useful texture)

• Angkor Wat was gradually converted from a Vaishnavite to a Theravada Buddhist temple over later

centuries and — unlike most Angkor sites — was never fully abandoned; it remained a living Buddhist

pilgrimage site into the modern era, which is why it is comparatively well preserved. SOURCE: Angkor

Wat, Wikipedia / New World Encyclopedia (above).

---

3. THE BAYON, JAYAVARMAN VII, and TA PROHM (the Buddhist Angkor Thom program)

3.1 Jayavarman VII — the 1200 frame-year king

• Reign c. 1181 – c. 1218. Came to power in his mid-50s after expelling the Cham occupiers (see §5),

and made Mahayana Buddhism the state religion (a break from the Shaivite/Vaishnavite norm). He is

the great builder-king: Angkor Thom (the last walled city), the Bayon (its state temple),

Ta Prohm, Preah Khan, Banteay Kdei, Neak Pean, plus roads, bridges, hospitals and

rest-houses across the empire. SOURCE: Jayavarman VII, Wikipedia,

https://en.wikipedia.org/wiki/Jayavarman_VII ; Britannica, Cambodia — Jayavarman VII,

https://www.britannica.com/place/Cambodia/Jayavarman-VII

• IMPORTANT FOR THE 1200 FRAME: in 1200 CE, Jayavarman VII is reigning at/near his peak; **Angkor

Thom and the Bayon are actively under construction**, the Cham wars are recently concluded/ongoing

(Khmer invasions of Champa continued to ~1203), and Angkor is a Mahayana-Buddhist imperial capital.

This is the era-state the excursion walks into. Angkor Wat (~1150) already stands, ~50 years old.

• Public-works program (real, inscribed — strong anti-fringe material): inscriptions record ~102

arogyasala (hospital chapels) and ~121 dharmasala ("houses of fire"/rest-houses) spaced along

the royal roads (rest-houses ~15 km apart), distributed across a territory spanning modern Cambodia,

Thailand, Laos and parts of Vietnam — evidence of a centralized administration. SOURCE: Arogayasala,

Wikipedia, https://en.wikipedia.org/wiki/Arogayasala ; History Today, "The Angkor Empire's National

Health Service," https://www.historytoday.com/miscellanies/angkor-empire%E2%80%99s-national-health-service ;

Cambodianess, "Hospitals Under King Jayavarman VII,"

https://cambodianess.com/article/hospitals-under-king-jayavarman-vii

NOTE: hospital and dharmasala counts vary slightly across inscriptions/sources (102 vs 104 hospitals;

121 vs 104 dharmasala) — treat as "roughly a hundred of each" rather than a hard number.

3.2 The Bayon

• State temple of Jayavarman VII, at the exact center of Angkor Thom; late-12th / early-13th c.; a

Mahayana Buddhist temple-mountain (later modified toward Hindu and then Theravada use). Main entrance

faces east (traditional orientation, unlike Angkor Wat). SOURCE: Bayon, Wikipedia,

https://en.wikipedia.org/wiki/Bayon ; Britannica, "the Bayon," https://www.britannica.com/topic/the-Bayon

• The face-towers (CONTESTED identity): ~37 surviving towers (originally ~49+) bear serene stone

faces (popularly ~200 faces). Scholarly candidates for whom they depict: the bodhisattva

Avalokiteshvara / Lokeshvara (bodhisattva of compassion) — the leading view once the temple was

recognized as Buddhist; Jayavarman VII himself (the faces resemble his portrait statues); or

Vajrasattva / Brahma / a fusion of king-and-bodhisattva. The debate persists because there is no

identifying iconography and Jayavarman VII's particular Buddhism is imperfectly understood; many

scholars now read them as an intentional king-as-bodhisattva fusion. SOURCE: Bayon, Wikipedia

(above); Smarthistory, "The Bayon: A temple with many faces,"

https://smarthistory.org/bayon-temple-angkor-thom/

• Construction (CONTESTED/complex): unlike the pre-planned Angkor Wat, the Bayon was built and

repeatedly modified in situ via "responsive field techniques" — three major building phases, its plan

evolving as religion shifted; this is one reason it feels labyrinthine. SOURCE: Templemountains.org,

"The three construction phases of the Bayon,"

https://templemountains.org/the-three-construction-phases-of-fhe-bayon-angkor-s-last-temple-mountain.html ;

Bayon, Wikipedia (above).

• Bas-reliefs: the outer galleries carry rare scenes of everyday Khmer life (markets, cockfights,

childbirth, fishing on the Tonle Sap) AND the naval war against the Cham (see §5) — an unusually

documentary program compared with the mythological reliefs of Angkor Wat. SOURCE: Bayon, Wikipedia

(above); Smarthistory (above).

3.3 Ta Prohm (site bayon_taprohm)

• Originally Rajavihara ("the king's monastery"); begun ~1186 under Jayavarman VII as a Mahayana

Buddhist monastery and university dedicated to his mother (its companion Preah Khan was dedicated to

his father). SOURCE: Ta Prohm, Wikipedia, https://en.wikipedia.org/wiki/Ta_Prohm

• The foundation inscription (a real, quantified population snapshot — excellent anti-fringe texture):

a stele records that the temple's domain supported ~12,640 people resident/serving it (incl. 18

high priests, 2,740 officials, 2,232 assistants, 615 dancers), that it owned ~3,140 villages, and

that ~79,365 people across the realm provided goods and services to it — including quantities of

gold, pearls and silks. This is direct inscriptional evidence of the scale and administrative reach of

the Angkorian state. SOURCE: Ta Prohm, Wikipedia (above); Ta Prohm inscription discovered 1913.

NOTE: figures are as the temple recorded its own endowment — read them as a state's self-account of

scale, not a modern census.

• The trees / unrestored condition: Ta Prohm was deliberately left largely unrestored by the EFEO

(and now the APSARA Authority) as a "concession to the picturesque" — the giant Tetrameles nudiflora

(silk-cotton) and Ficus (strangler-fig) roots grown through the galleries are conserved in place.

(This is a 20th–21st-c. conservation choice, not an ancient feature — worth keeping straight so the

"jungle-swallowed lost temple" image is framed as modern presentation, not mystery.) SOURCE: Ta Prohm,

Wikipedia (above).

---

4. THE DEVARAJA INSTITUTION (flag: heavily CONTESTED — do not overstate)

4.1 The traditional account

• The devaraja ("god-king") cult is traditionally said to have been established c. 802 CE by

Jayavarman II on Phnom Kulen (Mahendraparvata), founding the Angkorian state and its ideology of

the king as a universal sovereign linked to Shiva, whose divine essence was embodied in a royal

linga housed on a temple-mountain. For centuries the cult is said to have supplied the religious

basis of Khmer royal authority. SOURCE: Britannica, "Devaraja,"

https://www.britannica.com/topic/devaraja

• The term's key attestation is the Sdok Kak Thom inscription (K. 235), dated 8 Feb 1053 CE (a

temple in modern Thailand), which narrates the founding rite and gives the Khmer gloss

kamrateng jagat ta raja ("the Lord of the Universe who is King"). SOURCE: Devaraja, Wikipedia,

https://en.wikipedia.org/wiki/Devaraja

• The temple-mountain form (Bakong, Phnom Bakheng, Baphuon, Angkor Wat) is read as the architectural

expression of this cult — the king's linga/deity at the summit = Mount Meru, axis of the world.

4.2 The revisionist correction (CONTESTED — this is the current scholarly caution)

• Hermann Kulke ("The Devaraja Cult," 1978) and later scholars argued the older "each king made

himself a god" reading is not tenable. The devaraja (kamrateng jagat ta raja) is better understood

as a specific protective deity of the realm worshipped BY the kings — "god OF the kings" / a

distinctly Khmer tutelary god — rather than the apotheosis of the living king himself. The cult was

neither a simple worship rite nor a literal self-deification; its exact nature is still debated.

SOURCE: Kulke, "The Devaraja Cult" (Cornell SEAP, 1978),

https://books.google.com/books/about/The_Devar%C4%81ja_Cult.html?id=jGhDwQEACAAJ ; Kyoto Review,

"The Idea of Kingship in Buddhist Cambodia," https://kyotoreview.org/issue-11/the-idea-of-kingship-in-buddhist-cambodia/ ;

Devaraja, Wikipedia (above).

• Additional caution: the term devaraja appears in only ONE inscription (Sdok Kak Thom, and only its

Sanskrit portion), 250 years after the events it describes — so building a whole "god-king system" on

it is a thin evidentiary base. FRAMING: the divine-kingship IDEOLOGY (king ↔ Shiva/Vishnu ↔ Meru

temple-mountain) is real and well-attested in the monuments; the specific "devaraja = every king is a

living god" formula is the contested part. Present divine kingship as real; present "devaraja" itself

as a term scholars now read carefully.

• For the 1200 frame specifically: Jayavarman VII reframed royal divinity in Buddhist terms — the

king as a bodhisattva/Buddharaja (the Bayon face-towers, king-as-Lokeshvara) rather than a

Shaivite devaraja. The ideology of the sacred king persisted; its deity changed. SOURCE: Smarthistory

(above); Kyoto Review (above).

---

5. THE 1177 CHAM WAR (Angkor-core half of the elevated-care war lane)

NOTE: This is the Angkor-side context only. The Cham/Champa/Vijaya/My Son material and the modern-Cham

communities block are the ELEVATED-care surfaces per the Champa care ruling and are a separate research

target. The COMBAT CEILING is unchanged — the two-peer 1177 river-war set-piece stands at full

intensity; care = weight and authenticity, not omission. Route any story antagonist through

apparatus/brokers, never through the Cham people as a collective.

• 1177: a Cham (Champa) fleet under king Jaya Indravarman IV sailed up the Mekong into the

Tonle Sap and stormed the Khmer capital Yasodharapura, killing the Khmer king

Tribhuvanadityavarman; the lightly-defended (wooden-palisade) capital fell and the Cham occupied parts

of Cambodia for ~4 years. SOURCE: Khmer–Cham wars, Wikipedia,

https://en.wikipedia.org/wiki/Khmer%E2%80%93Cham_wars ; Battle of Tonlé Sap, Wikipedia,

https://en.wikipedia.org/wiki/Battle_of_Tonl%C3%A9_Sap

• 1181: the future Jayavarman VII led the Khmer navy to defeat the Cham and retake the capital,

then took the throne. This victory is the founding event of his reign and of the Bayon/Angkor-Thom

program. Khmer counter-invasions of Champa continued; organized Cham resistance collapsed ~1203, and

Champa was briefly annexed as a Khmer province. SOURCE: Jayavarman VII, Wikipedia (above); Khmer–Cham

wars, Wikipedia (above).

• The Bayon bas-reliefs depict this war — Khmer and Cham war-galleys (the Cham identifiable by their

distinctive lotus-petal helmets), the Tonle Sap battle, and land engagements, alongside scenes of

everyday Khmer life. Banteay Chhmar carries related reliefs. This is the in-world visual primary source

for staging the river war. SOURCE: Bayon, Wikipedia (above); Smarthistory (above).

• The two peers are genuinely comparable maritime/riverine powers (mature two-peer war, appropriate to

the full-intensity ceiling) — the Cham are a sophisticated Austronesian-speaking, Hindu-then-Muslim

maritime state, not a barbarian foil. (Cham-source grounding obligation attaches to the sanctuary/war

surfaces — flagged for the separate elevated-care research pass.)

---

6. ZHOU DAGUAN — the era-adjacent eyewitness (honest date handling)

• Zhou Daguan (Chou Ta-kuan), a Chinese envoy of the Yuan (Temür Khan) mission, reached Angkor in

August 1296 and stayed until July 1297, at the court of King Indravarman III. His

"Customs of Cambodia" (Zhenla Fengtu Ji / A Record of Cambodia: The Land and Its People) is the

only surviving first-person eyewitness account of daily life at Angkor at/near its height —

40 short thematic chapters on the court, royal processions, houses, clothing, slavery, women, trade,

agriculture, fish, festivals, justice, and the temples. SOURCE: The Customs of Cambodia, Wikipedia,

https://en.wikipedia.org/wiki/The_Customs_of_Cambodia ; Zhou Daguan, Wikipedia,

https://en.wikipedia.org/wiki/Zhou_Daguan

• The best modern English edition is Peter Harris's translation ("A Record of Cambodia: The Land and

Its People," Silkworm Books, 2007), from the Chinese (earlier English versions came second-hand via

Pelliot's French). A Kent Davis / Solang & Beling Uk edition also exists. SOURCE: Wikipedia (above);

DatAsia, https://www.datasia.us/publishing/customs-of-cambodia-zhou-daguan/

• HONEST DATE FLAG (do not paper over): Zhou Daguan visited in 1296, roughly 96 years AFTER

the game's 1200 CE frame. Between the two: Jayavarman VII died (~1218), Jayavarman VIII's Hindu

reaction and iconoclasm intervened (see §7), and Theravada Buddhism rose. So Zhou describes an Angkor

slightly past its Jayavarman-VII zenith and already religiously shifted (he notes Theravada monks,

"Chinese"-style features, ongoing temple life). USE HIM as the richest window into Angkorian daily

life/material culture, but do NOT present his 1296 details as a literal snapshot of 1200 — note where a

detail is late-13th-c. specifically (e.g., the religious mix he describes). The city's physical fabric

(barays, Bayon, Angkor Thom's walls and gates, the causeways) is largely continuous 1200→1296, so his

topography and infrastructure descriptions transfer well; the religious/political particulars are 1296.

• Vivid transferable details from Zhou (texture, not load-bearing canon): the gilded central tower and

golden bridge/lions of the royal palace; the king's processions; the "official" who reads the water;

the Tonle Sap fishery abundance; women traders dominating the market; houses ranked by rung/roof

material; the practice of the king appearing at a golden window. SOURCE: Customs of Cambodia summaries

(Wikipedia; Wonders of Cambodia, https://wondersofcambodia.com/zhou-daguan-chronicler-of-angkors-golden-era/).

CAUTION: some of Zhou's more sensational passages (e.g., ritual customs, the "naked" descriptions) are

a foreign observer's read — frame authorially, don't reproduce uncritically.

---

7. FORWARD CONTEXT — decline and the 1431 Ayutthaya sack (past-1200, design-layer only)

Everything here is AFTER the 1200 frame — L3/design-layer forward context, NOT player-visible in Ch 5.

• Jayavarman VIII (r. 1243–1295) — Hindu reaction / iconoclasm (CONTESTED attribution): a Shaivite

reversal of Jayavarman VII's Buddhism. Buddhist images across the Jayavarman-VII temples were

systematically defaced/chiselled out; the great seated Buddha of the Bayon's central sanctuary was

smashed and (later) recovered in pieces from a pit; ~274 Buddha images were found buried at Banteay

Kdei. NOTE — the attribution of this iconoclasm specifically to Jayavarman VIII is a widely-repeated

hypothesis, not proven. SOURCE: Jayavarman VIII, Wikipedia,

https://en.wikipedia.org/wiki/Jayavarman_VIII ; Southeast Asian Archaeology, "Decoding Hindu-Buddhist

rivalry at Angkor," https://www.southeastasianarchaeology.com/2007/11/27/decoding-hindu-buddhist-rivalry-at-angkor/

• 1295 →: Indravarman III (a Theravada Buddhist) took power; Theravada became dominant — this is the

religious world Zhou Daguan reports in 1296.

• Climate stress (real, dendrochronology-grounded): the Buckley/Penny tree-ring record (from

*Fokienia hodginsii* cypress in Vietnam's Bidoup Nui Ba highlands, ~759 years reconstructed) shows a

multi-decadal megadrought from the ~1330s–1360s, then a shorter, more severe drought in the

~1400s–1420s, interspersed with unusually intense monsoons. The problem was variability — the

swing between drought and flood — which damaged/silted the aging hydraulic network that was already

under-maintained. SOURCE: Buckley et al., "Climate as a contributing factor in the demise of Angkor,

Cambodia," PNAS 107 (2010), https://www.pnas.org/doi/10.1073/pnas.0910827107 ; National Geographic,

"Tree rings reveal two droughts that sealed the fate of Angkor,"

https://www.nationalgeographic.com/science/article/tree-rings-reveal-two-droughts-that-sealed-the-fate-of-angkor

• Infrastructure fragility: Penny, Fletcher et al. (2018) showed the water network's failure was a

systemic vulnerability — a tightly-coupled system that could cascade — rather than a single knockout

blow; elites appear to have been leaving Angkor Thom already in the 14th c., before any sack. SOURCE:

Penny et al., "The demise of Angkor: Systemic vulnerability of urban infrastructure to climatic

variations," Science Advances 4 (2018), https://www.science.org/doi/10.1126/sciadv.aau4029

• 1431 Ayutthaya sack (CONTESTED as the cause of collapse): the traditional account has the Thai

kingdom of Ayutthaya sacking Angkor in 1431, after which the Khmer court moved its capital

south (toward Phnom Penh / Longvek). MODERN CONSENSUS: 1431 was a punctuation mark, not the cause —

decline was gradual and multi-causal (climate variability + hydraulic-network failure + a shift of

economic gravity to maritime trade on the Mekong/Gulf + religious change from state-temple Hinduism to

congregational Theravada Buddhism, which no longer needed god-king temple-mountains). Angkor was never

fully abandoned; Angkor Wat remained a living Buddhist site. SOURCE: Fall of Angkor, Wikipedia,

https://en.wikipedia.org/wiki/Fall_of_Angkor ; Penny et al. 2018 (above); NC State Applied Ecology,

https://cals.ncsu.edu/applied-ecology/news/lessons-from-climate-change-induced-collapose-angkor-wat/

• FRAMING: if any forward-echo is used (an era-rhyme, an L3 seed), stage decline as the

climate+infrastructure+economic story, NOT as a mysterious/sudden "vanished civilization" — the

vanishing-city trope is exactly the register the anti-fringe hard line rejects.

---

8. ANTI-FRINGE — the hard line, sourced

REFUSED (do not use, in any register, ever):

• FRINGE — REFUSED: "Angkor was built by a lost race / by refugees of a prior super-civilization /

with lost or non-human technology / is too sophisticated for the Khmer." This is false and is the

lost-race / ancient-aliens pseudo-archaeology the brief bars. Angkor is exhaustively dated by

inscriptions, architectural seriation, and AMS radiocarbon dating (e.g., direct dating of

the Baphuon), and its builders, kings, quarries, canals, and labor system are documented. SOURCE:

Angkor, Wikipedia, https://en.wikipedia.org/wiki/Angkor ; UNESCO World Heritage — Angkor,

https://whc.unesco.org/en/list/668/ ; Baphuon direct-dating study, PMC,

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4633138/

• FRINGE — REFUSED: treating the barays/hydraulic system as evidence of "impossible" engineering or

hidden knowledge. It is impressive but entirely explicable earthwork-and-hydrology engineering, mapped

and modelled by the GAP. (§1 above.)

• FRINGE — REFUSED: reading Angkor Wat's alignments as proof of a lost astronomical super-science or

as an "Orion/precession" monument. Real archaeoastronomy (equinox sunrise over the tower) is fine and

cited (§2.2); the super-science overlay is not.

POSITIVE anti-fringe assets to foreground instead (all real, all Khmer): the inscriptional record (Sdok

Kak Thom, Ta Prohm, the hospital steles), the quarry-and-canal transport archaeology, the LiDAR-mapped

lived city, and the dated dynastic sequence. The most powerful counter to "mysterious lost city" is the

sheer administrative concreteness — 12,640 named servants of one temple, ~100 hospitals, a corvée-fed

workforce of tens of thousands. Use the receipts.

---

9. BUILD-RELEVANT SYNTHESIS (for the region-page author)

• Frame the city at 1200 CE precisely: Jayavarman VII reigning near peak; Angkor Thom + Bayon under

active construction; Angkor Wat ~50 years old and Vaishnavite; the Cham war freshly won (Khmer

campaigns into Champa ongoing to ~1203); Mahayana Buddhism ascendant as state religion; the hydraulic

city at its greatest extent (~1,000 km², ~700k–900k people in Greater Angkor). This is a living

imperial capital at its zenith, not ruins.

• The five CH_05 sites map cleanly: kra_traverse (the overland/portage approach — the Kra Isthmus

land-bridge between the Bay of Bengal and the Gulf, the historic east-west traverse); angkor_wat (the

Vishnu funerary-temple, west-facing, ~1150, the orientation debate is stage-able texture); bayon_taprohm

(Jayavarman VII's Buddhist Angkor Thom — face-towers, the Cham-war reliefs, Ta Prohm's inscription and

conservation-jungle); west_baray (the hydraulic-city set-piece + the reclining Vishnu of the West

Mebon + the combat-staging elevated surface); myson_champa (the Cham sanctuary — separate elevated-care

research target).

• Contested things to stage AS debates (not as settled facts): baray purpose (irrigation vs ritual);

Angkor Wat's west orientation (funerary vs Vishnu-association vs astronomy); the Bayon faces

(Lokeshvara vs the king vs fusion); the devaraja's meaning (god-king vs god-of-kings); and the cause

of the 1431/decline (multi-causal, not a single sack). Presenting these honestly is itself the mature,

anti-slop register the project wants.

• Do-not-invent watch: the site slugs, the CH_05 beat structure, and chapter assignment are Josh's

canon — this file supplies real-world grounding under those anchors, not new canon. All numbers above

carry their sources; where sources disagree (hospital/dharmasala counts, population range, labor

estimate) I have flagged the spread rather than picking one.

Generated by harness/site/structure_site.py — the URL path is the repo path. review root