Record · Ancient world · East Asia · 1700 BC

Is Erlitou China's first dynasty, or a city no one can name?

By the Our Earth's History editorial team · Researched and written with AI assistance, checked against the sources below · How we make it

Published 2026-09-19

TL;DR — Erlitou has palaces, a bronze foundry and grid roads, but no writing. Chinese archaeologists call it a Xia capital; several Western specialists say the name cannot be tested.

No text found at Erlitou names its builders, and whether the city in Henan province was a capital of the Xia dynasty has been argued since 1959. The ruins sit on the plain between the Yi and Luo rivers, in the Yanshi district of Luoyang, and cover about three square kilometers. Radiocarbon ranges published for the site differ by source and fall between about 1900 and 1500 BC. What has come out of the ground is a walled palace precinct, a bronze foundry, a grid of wide roads and the earliest known Chinese ritual vessels cast in bronze. What has not come out of the ground is a single readable character.

Quick facts

When: Site dated about 1900–1500 BC by competing radiocarbon ranges · found in 1959 · turquoise dragon excavated in 2002 · museum opened October 19, 2019
Where: Erlitou village, Yanshi district of Luoyang, Henan province, on the plain between the Yi and Luo rivers in what is now central China
Who: Xu Xusheng (1959 survey); Zhao Haitao and Wang Wei (Chinese Academy of Social Sciences); Sarah Allan, Robert Bagley and Robert Thorp (skeptics); Li Liu and Hong Xu (third position)
Outcome: Palaces, a walled bronze foundry, grid roads and the earliest known Chinese cast ritual vessels, with no writing that names the rulers; the Xia identification remains argued

A survey that went looking for the Xia

The site was found by a man who had already decided what he wanted to find. Xu Xusheng, a historian at the Institute of Archaeology in Beijing, worked from the received texts. Chen Chun and Gong Xin, writing in Social Evolution & History in 2018, recorded that Xu identified the Luoyang plain and the lower Fen River valley as the two likely areas and went to Henan to survey them. His report of the 1959 season carried the phrase “ruins of Xia” in its title.

Xinhua and People’s Daily Online both dated the discovery of Erlitou to that survey; Sarah Allan of Dartmouth College wrote that the site was first found in 1953. The Institute’s Luoyang field team began excavating in 1959 and reported three layers of stratigraphy running from the late Longshan period to the early Shang.

What the excavators uncovered

The center of the site was a walled precinct of about 10.8 hectares holding rammed earth platforms that carried large timber buildings. Allan recorded at least eight such platforms identified or dug, the largest of them 100 by 108 meters and the second 58 by 73 meters, all laid out on a north-south axis. The buildings are called palaces because of their size and their courtyard plan; Robert Thorp of Washington University, whose objections Chen and Gong summarized, wrote that rammed earth foundations do not by themselves prove a palace, and that the compounds may have been ceremonial.

Roads 10 to 20 meters across enclosed the precinct on all four sides, according to Allan, who recorded traces of carriage wheels in them and dated their construction to the second of the site’s four periods. Crossing roads cut the core into walled blocks. Zhao Haitao of the Chinese Academy of Social Sciences, who has led the dig, told China Daily in 2022 that this grid was one of the larger finds of the preceding years. Excavation has reached 59,500 square meters, under two percent of the site.

The foundry, the wine cups and the dragon

A bronze workshop stood inside its own walled enclosure south of the palaces, and casting waste showed that the vessels were made on site. Allan counted more than a dozen bronze jue from Erlitou, along with at least four jia, one gu and one he. The jue was a footed cup with a spout and a pointed tail, used for fermented drink; the largest stood 22 centimeters high. All were cast by the piece-mold method, in which the vessel is built from several fitted clay sections.

In 2002 the team opened tomb M3 in the courtyard of Palace 3. The occupant was a man of about 30 to 35, laid out with his head to the north, a bronze bell with a jade clapper at his waist and, across his chest from shoulder to thigh, a dragon figure some 65 centimeters long made of more than 2,000 pieces of turquoise glued to a backing that had rotted away. China Daily gave the length as nearly 70 centimeters.

Why the name Xia is contested

Chinese archaeology has treated the identification as settled in the main. Chen and Gong set out four arguments used to support it: Zhou texts place the Xia in western Henan and southern Shanxi, the palace foundations imply a state, the radiocarbon dates fall inside the traditional Xia span, and Sima Qian’s account of the Shang proved reliable at Anyang, so his account of the Xia may be reliable too. The Xia-Shang-Zhou Chronology Project, a state program under Li Xueqin that began in 1995 and reported in 2000, published a king list and a chronology for the Xia and dated Erlitou by calibrated radiocarbon to about 1860 to 1545 BC.

The objections turn on the same missing evidence. The earliest written mention of the Xia appears in Western Zhou material in the Book of Documents, and Sima Qian wrote more than a thousand years after the dynasty he described. Robert Bagley of Princeton University argued that the Zhou invented the Xia to justify their overthrow of the Shang, and that the Shang oracle bones give no sign that their kings saw themselves as heirs to Xia legitimacy. Allan, who in 1984 read the Xia as a Shang myth later turned into history by the Zhou, wrote in 2007 that unless writing is found there is no way to determine whether the people at Erlitou called themselves Xia, or were called that by anyone else. She proposed cultural hegemony in place of dynastic succession as the frame for the site. Li Liu and Hong Xu took a third route in Antiquity in 2007, separating the textual histories from the excavated sequence and arguing the archaeology on its own. The Metropolitan Museum of Art’s chronology labels the dynasty “Xia dynasty (unconfirmed).”

What would settle it

Both sides named the same test: contemporary writing from the site. Marks scratched on Erlitou pottery have been read by some as ancestral to later characters, and Yuan Guangkuo of Capital Normal University told China Daily that their shapes recalled oracle bone inscriptions and that written characters were probably still buried at the site. Nothing at Erlitou yet records a name, a king or a year.

The Erlitou Site Museum of the Xia Capital opened in Luoyang on October 19, 2019, with more than 2,000 objects on 32,000 square meters, built for 630 million yuan. Its name settles in signage a question its exhibits do not. Excavation continued: in 2024 the team traced moats at the Guchengcun site across the river, one of them more than 1,700 meters long, which Zhao described as a possible outer boundary of the city.

Key facts

Erlitou lies in the Yanshi district of Luoyang, Henan province, between the Yi and Luo rivers, and covers about three square kilometers.
Radiocarbon ranges differ by source: about 1750 to 1520 BC (China Daily, 2022), about 1860 to 1545 BC (Xia-Shang-Zhou Chronology Project, 2000).
Xu Xusheng surveyed western Henan in 1959 while looking for the “ruins of Xia”; excavation by the Institute of Archaeology began that year.
The walled palace precinct covers about 10.8 hectares and held at least eight rammed earth platforms, the largest 100 by 108 meters (Allan, 2007).
Roads 10 to 20 meters across, with traces of carriage wheels, enclosed the precinct on all four sides (Allan, 2007).
The bronzes include more than a dozen jue, the tallest 22 centimeters, cast on site by the piece-mold method; the turquoise dragon came from tomb M3 in 2002.
No writing has been excavated at Erlitou, and about 59,500 square meters, under two percent of the site, has been dug.

FAQ

Was Erlitou the capital of the Xia dynasty?

Chinese archaeologists and the Xia-Shang-Zhou Chronology Project identify it as a late Xia capital. Sarah Allan, Robert Bagley and others hold that without writing from the site the identification cannot be tested. Nothing excavated at Erlitou names its rulers.

How old is the Erlitou site?

Published radiocarbon ranges differ. China Daily reported about 1750 to 1520 BC in 2022, the Xia-Shang-Zhou Chronology Project gave about 1860 to 1545 BC, and china.org.cn reported dates of 3,750 to 3,520 years before 2025.

What is the turquoise dragon of Erlitou?

A figure some 65 to 70 centimeters long made of more than 2,000 turquoise pieces, excavated in 2002 from tomb M3 in the courtyard of Palace 3, lying across the chest of a man buried with a bronze bell at his waist.

Sources

Cite this article

APA: Our Earth's History. (2026, September 19). Is Erlitou China's first dynasty, or a city no one can name? Our Earth's History. https://ourearthshistory.com/east-asia/ancient/erlitou-first-dynasty-or-nameless-city/
Chicago: Our Earth's History. “Is Erlitou China's first dynasty, or a city no one can name?” Our Earth's History, September 19, 2026. https://ourearthshistory.com/east-asia/ancient/erlitou-first-dynasty-or-nameless-city/.
MLA: “Is Erlitou China's first dynasty, or a city no one can name?” Our Earth's History, 19 Sep. 2026, https://ourearthshistory.com/east-asia/ancient/erlitou-first-dynasty-or-nameless-city/.

Text licensed CC BY-NC-ND 4.0. Found an error? Tell us.