Vietnam · who controls the story

A riverbed full of timbers, and two stories about them

In 2019 Vietnamese press tied wooden stakes at Cao Quy to the battle of 1288. A 2026 paper dates eight of them to 566-352 BCE. The argument says a lot about how Vietnam's early past gets told, and by whom.

vizmaya desk · 7 October 2026

3Cao Quy and Dam Thuong

Today a former riverbed and reclaimed paddy fields in Thủy Nguyên, Hải Phòng, with timbers buried under fine estuarine mud. In 2019 press linked the stakes to 1288 and gave dates of 1270-1430 CE, with no laboratory named.

The 2026 team reads them as Dong Son stilt-house foundations, built when the sea stood 1-2 m above today's and drowned after about 2,200 years ago.

contested

The 2026 dates sit about 1,640 to 1,850 years before the battle

Sources: The Holocene 36(9), 2026 (abstract read; full text paywalled); Dan Tri and Tuoi Tre, December 2019 (laboratory not named).

The gap

1,640

years, roughly, from the youngest timber date (352 BCE) to the battle of 1288. The oldest date, 566 BCE, makes it about 1,850.

What it does not say

The abstract does not claim the 1288 battle never happened, and it does not test other stake fields. A national museum labels stakes from the Chanh River, found in 1976, as 1288 relics. The fight is over which fields, not whether the battle took place.

Five of the seven places sit around the Red River Delta

Coordinates for Luy Lâu and Óc Eo come from a fetched source. The other five are approximate to one decimal and need a GPS check before any survey use.

1Tràng An karst, Ninh Bình

Limestone towers on the southern edge of the Red River Delta, now a World Heritage site. Cliff notches record long stable sea levels during the mid-Holocene rise, around 8,000-4,000 years ago. The terrain flipped between inland valley and archipelago, and human activity shows up in both states.

documented

2Quan Lan Island, Hạ Long Bay

A core from an island wetland shows low-level subsistence wet rice and burning between 1150 BCE and 950 CE, under a stronger monsoon. After 950 CE a port appears, nearby population drops to undetectable, rice declines and the climate turns drier. The authors name both causes and do not rank them.

documented

Three sea-level figures that do not line up

The Red River delta grew outward as sea-level rise slowed. Rapid rise ran from 11,000 to 8,500 years ago, and outward growth has dominated for the last 6,500.

Timing and datums differ across the three and are not reconciled here. Sources: Tanabe et al. record; The Holocene 2026; Taylor, A History of the Vietnamese, Cambridge 2013 (excerpt).

4Cổ Loa, Đông Anh, Hà Nội

Three concentric earthen ramparts 15-17 km north of central Hanoi, a special national relic since 2012. The outer enclosure covers about 600 ha. The middle rampart went up before the Han arrived, an older Dong Son rampart lies under the inner wall, and the 2004-2007 dig found stone moulds for bronze arrowheads.

Legend gives it to An Dương Vương, capital of Âu Lạc.

site documented, founder contested

Cổ Loa has three official founding dates, and wall levels reaching back to about 300 BCE

The 2nd-century range runs past the 111 BCE annexation. Sources: Kim, Lai and Trịnh, Antiquity 2010; National Museum of History; Southeast Asian Archaeology news items.

5Luy Lâu (Dâu), Bắc Ninh

A citadel site centered on Diên Ứng Pagoda, excavated eight times. After the Han took the region in 111 BCE it became the first capital of the Jiaozhi commandery. The 2014 season recovered over 1,200 objects dated Dong Son through Sui-Tang.

The excavators reject the older idea that Viet lived outside the walls and Han inside.

documented

Where the ground sides with whom

  • Pre-Han Cổ Loa: the national account wins. A joint Hanoi-Wisconsin team showed the middle rampart was built before the Han arrived.
  • Cổ Loa's date: the national account disagrees with itself. Three official-leaning sources give 257 BCE, 214-208 BCE and the 2nd century BCE.
  • Luy Lâu undercuts the tidy colonial split. Local people were inside the Han city, not shut out of it.
  • The Bạch Đằng timbers fit neither the court chronicle nor the Chinese record. They point to a drowned Dong Son settlement nobody was looking for.
  • Óc Eo sits inside Vietnam's borders, but its polity is told elsewhere. The state the Chinese called Funan is usually tied to Cambodian history , and the name is an outsider's label.

6Trà Kiệu and Mỹ Sơn, Quảng Nam

Bửu Châu hill and citadel remains on the Bà Rén river, about 18 km inland of Hội An, with the Mỹ Sơn temples further up the valley. The earliest levels, from the last centuries BCE, mix local, Indian and Han material and have little in common with Sa Huỳnh pottery.

How long Trà Kiệu was a capital, and how it ties to Chinese Lin Yi, is disputed. No Cham-authored source was used.

contested

7Óc Eo and Ba Thê, An Giang

Flat delta ground at the foot of Ba Thê mountain. Vietnamese remote-sensing work finds ancient canals joining Châu Đốc and Óc Eo to the Rạch Giá bay, plus a seaport at Nền Chùa. Chinese chroniclers named the polity Funan. A secondary summary says the name never appears in local inscriptions and that Hà Văn Tấn argued it should go.

canals documented, label contested

The version the country tells itself

Cổ Loa went up in 214-208 BCE as the capital of Âu Lạc and is the first capital of Vietnamese civilization, says the National Museum of History. Three naval victories on the Bạch Đằng, in 938, 981 and 1288, fix a line of national heroes. The Hanoi museum labels Chanh River timbers as vestiges of 1288.

The version abroad

Taylor (Cambridge, 2013) argues the Vietnamese past has no internal logic leading to the present. He says the "thousand years of Chinese domination" cliché hides how language and culture formed in contact, and that court historians of the 13th and 15th centuries wrote a parallel southern history.

The two sides differ on whether Dong Son runs into Vietnamese identity at all.

37 claims logged, 4 kept off the page

A figure, date or causal claim needs two independent sources or one primary source. Anything weaker stays out of the route and the headline.

Contested claims on this board are labeled as such.

What we could not read

The 2026 paper is paywalled, so only the abstract and an excerpt of its introduction were read. No lab report exists for the 1270-1430 dating. No Cham-authored source covers Trà Kiệu and Mỹ Sơn. The oldest density figure is the colonial one from 1931.

Left out on purpose

Whether the drought of about 5-4 thousand years ago shaped settlement in the Red River Delta. A stalagmite record shows the drop in northern Laos, but no source ties it to the delta.

People today

101,112,656 people on 1 April 2024. In the 2019 census, 85.3% were Kinh (82,085,826 of 96,208,984) and six minority groups topped 1 million each, the Tay largest at 1.85 million. The Mekong Delta grew 0.05% a year from 2009 to 2019, the slowest of any region.

the claim the places the north, argued over the ground two versions the checks

  1. Who gets to say what the ground means

    Seven places in Vietnam, from a drowned estuary to a canal port in the Mekong Delta. Each has a physical record and at least one story laid over it. Follow the tour or drag the board yourself.

  2. Timbers older than the battle

    In 2019 Vietnamese press tied stakes at Cao Quy to the 1288 battle. A 2026 paper dates eight timbers to 566-352 BCE, roughly 1,640 to 1,850 years earlier.

  3. What the paper does and does not say

    The authors read the timbers as Dong Son stilt-house foundations. The abstract does not claim the battle never happened, and a national museum still labels other stakes as 1288 relics.

  4. The delta was still moving

    Sea level, rice and burning shaped where people could live before anyone wrote anything down. The three sea-level figures here do not line up, and we have not forced them to.

  5. Seven places, five in the north

    Filled circles have coordinates from a source. Open circles are approximate and still need a GPS check.

  6. Cổ Loa: older than the Han, unsure of its birthday

    The middle rampart predates the Han arrival and some wall levels reach back to about 300 BCE. Official sources still give three different founding dates.

  7. Luy Lâu and the colonial split

    After 111 BCE the record turns administrative and Chinese. Excavation finds local people inside the Han city, not shut outside it.

  8. The south has other owners of its past

    Trà Kiệu's capital years are disputed, and no Cham-authored source was used. At Óc Eo the canals and port are documented, while the Chinese label Funan is contested.

  9. Two versions of one millennium

    The country's account runs from Cổ Loa through three river victories. Taylor argues culture and language mostly formed in contact during Chinese rule.

  10. Where the ground sides with whom

    The national account wins on pre-Han Cổ Loa and contradicts itself on the date. Luy Lâu and the Bạch Đằng timbers fit neither tidy story.

  11. What we checked and what we could not read

    Of 37 logged claims, 24 are verified and 9 contested. The 2026 paper's full text, a Cham voice and any pre-1000 CE population figures are missing.

  12. Sources and method

    Every link behind the board, in one place.