An original AI-assisted historical narrative based on archaeological publications actually consulted.

After a spring rain, the soil in one excavation area at Taosi began to look different. The archaeologists scraped the surface again. Rammed earth, natural deposits, and pits dug in later periods had been difficult to disentangle. Now another arrangement emerged: thirteen blocks of rammed earth following an arc. This was Xiangfen in Shanxi, with Mount Ta’er in the distance. Only the foundations of a building more than four thousand years old remained. Nobody could immediately explain what had once stood on those blocks.

This was not the kind of discovery in which an intact object could be lifted, measured, and examined for decoration. The evidence lay across the ground, cut through by subsequent activity. While Gao Jiangtao worked in this area, the excavation repeatedly required a little more earth to be removed, followed by another look at what was already exposed. Several field seasons, from spring 2003 to spring 2005, passed before an unusual structure gradually took shape.

Wu Jiabi, an archaeologist with training in the history of astronomy, suggested observing sunrise through the gaps. The team erected a temporary model and watched and photographed the eastern horizon. The rising sun changed position along the mountain ridge with the seasons, and certain openings seemed capable of marking that movement. The possibility was exciting, but it made the questions harder. If this was an astronomical structure, where had observers stood? How tall were its pillars? What reconstruction did the surviving foundations actually permit?

To understand the interest in this discovery, we first need to step away from the arc and consider the city around it. Taosi was not an isolated observation platform beside a mountain. Archaeological work extended from burials to settlement remains and then to walls, revealing a substantial prehistoric center. Taosi culture is dated broadly to about 2300–1900 BCE. The large city of its middle phase covered approximately 2.8 square kilometers. A picture of a few dozen households enclosed within a village could no longer contain what was emerging here.

Finding the walls had also required considerable effort. Fields and modern settlement did not mark the boundaries of the buried city. Rammed earth had to be distinguished from natural deposits, one excavation trench at a time. As sections of wall and ditch were identified, the outline gained substance. Smaller enclosed areas existed inside the city. Palatial buildings, craft activity, and burial grounds had not simply been crowded together at random. Space was being allocated: some for living, some for making things, and some for a restricted part of society.

The cemetery made those differences particularly visible. Early excavations cleared more than thirteen hundred graves. Some were small; others contained sets of special objects. In spring 1980, a painted pottery plate emerged from a large burial designated M3016. A coiled creature occupied its surface, with a patterned body, an unusual head, and something branching beside its mouth. It became known as a dragon plate. What that branching form meant cannot be settled by projecting familiar modern dragon imagery backward.

These were not plates found in every household. In this group of burials they belonged to a few large graves, accompanied by other distinctive objects: drums, stone chimes, and equipment for preparing and serving food. Similar arrangements and combinations in different graves suggest ritual practices that may have been repeated according to established expectations. We do not know how a particular ceremony began or who struck its first drumbeat. The objects nevertheless show that some people controlled resources and ceremonial equipment unavailable to others.

The placement of the plates offered another clue. Examples came from large graves including M3072, M3073, and M2001 as well as M3016. Comparing burial plans, researchers found that a plate generally lay toward the upper right of the deceased, one to a grave. An object displayed alone today had once belonged to an entire arrangement for burying a person. Its significance depended not only on its image but also on the objects beside it and on who could receive such an assemblage.

Differences in living conditions also survived in buildings. Colored plaster recovered from the palatial area included red, white, and blue. The buildings had not been as uniformly earth-colored as their remaining foundations appear today. Their upper parts are gone, and a few pieces of plaster cannot reconstruct a complete palace. Yet these finished and decorated structures, considered alongside other residential remains, remind us that people living within the same city did not necessarily share the same homes, possessions, or social positions.

Now return to the thirteen blocks. Further excavation showed that some apparently separate blocks were connected underneath. The arc, the platforms, and the proposed observation point had to be explained together. One conveniently aligned opening was not enough. Archaeologists and astronomers tried to bring excavation, measurement, and repeated observation into a single investigation. Questions did not diminish the discovery. They determined what additional evidence was needed.

In March 2009, researchers conducted another experiment, using brick pillars reconstructed at the original site. On March 17 and 18, the sun appeared against the distant ridge through a particular opening. The seasonal pattern produced by the model supported an astronomical interpretation. Seeing the sun that day, however, could not prove that every ancient pillar had matched the modern reconstruction. Nor should the observation date simply be equated with the spring equinox in a modern calendar. What survived archaeologically was a foundation; researchers supplied reconstructed pillars and a proposed method of observation.

If this interpretation approaches the structure’s ancient purpose, it concerns something practical: making time recognizable in common. The sun rises every day, but not at the same place on the ridge. Repeated observations and remembered positions could provide relatively stable markers within the seasonal cycle. Sowing, harvesting, and ceremonies all required arrangements. We cannot, however, assign a specific agricultural task to an individual opening and thereby invent an ancient timetable.

Walls, palatial buildings, exceptional burials, and this unusual structure together make Taosi important to the study of early complex societies. Their prominence also creates a temptation to supply familiar names. Could this have been the capital of Yao? Later transmitted texts associate Yao with ordering time and governing the world, and Taosi’s location and period have encouraged serious discussion of that possibility. But a large city sharing some features with later accounts is not the same as a contemporary document naming its ruler.

Pottery bearing cinnabar-written signs has been found at Taosi. Some readings command more agreement than others, while interpretations of other signs differ widely. They deserve close study. A disputed sign cannot simply become a seal that establishes the identity of an entire city. Between “Taosi may be connected with traditions about Yao” and “Taosi has been proved to be Yao’s capital” lies an evidentiary distance. Preserving that distance makes neither the walls smaller nor the painted plate less vivid.

The city’s ending was unsettled as well. Late remains include damaged graves, disturbed building areas, and human bones subjected to unusual treatment. The earlier order evidently underwent serious change. Researchers discuss violence, struggles over power, and social conflict, but traces of destruction cannot alone produce a complete military history explaining who attacked whom and over what interval. Nor can the familiar tradition of Yao peacefully yielding power to Shun be pasted, episode by episode, onto the excavated remains.

Taosi therefore offers something other than a royal list with every name filled in. It preserves the labor of building walls, dividing urban space, making ceremonial objects, decorating houses, and perhaps watching the sky over long periods. It also preserves traces of disrupted order. We cannot see who stood between the pillars on a particular morning or hear what that person said afterward. We can approach the ways in which a community organized land, objects, and time—and the questions it left underground.

Researchers returning to the arc today face the same east but inhabit a different world of knowledge. The ancient structure survives as foundations, and modern explanations remain subject to testing. The thirteen blocks merit repeated measurement because they have preserved a question with unusual precision: how could a city enable its people to recognize together that the same season was arriving?

Sources and editorial notes 1. Gao Jiangtao, oral account compiled by Du Jiabing, “Gao Jiangtao: Exploring ‘China’ at Taosi More Than Four Thousand Years Ago,” published March 14, 2025. The passages on the 2003–2005 excavation, soil after rain, thirteen rammed-earth blocks, and experimental observation were consulted. This is a later participant interview, not a contemporary field diary. https://www.cssn.cn/kgxc/kgxc_kgdz/202503/t20250314_5857922.shtml 2. National Astronomical Observatories, Chinese Academy of Sciences, report on experimental observation at Taosi, originally published July 1, 2009. The current migrated URL contains 2022. The account of observations on March 17–18, 2009, and the reconstruction description were consulted. A possible astronomical function remains an interpretation; the model is not an excavated original, and March 17 is not presented as the modern spring equinox. https://nao.cas.cn/news/zh/202203/t20220330_6414386.html 3. He Nu, study of Taosi and ideas of the world in traditions about Yao and Shun, 2015. City scale, spatial divisions, objects, and location were used. Its broader identification of Yao’s capital and interpretation of a prehistoric worldview were not adopted as established facts. Its reference location, 35°53′ N, 111°30′ E, is suitable only as an approximate site-area point. https://www.cass.cn/xueshuchengguo/wenzhexuebulishixuebu/201506/t20150605_2023339.shtml 4. Liangzhu Museum, discussion of the dragon plates excavated at Taosi, October 3, 2022. Burial numbers, forms, associated objects, and limits of interpretation were consulted. https://www.lzmuseum.cn/LiangBoXinWen/20225386654.html 5. “Details of Civilization: A Report from Taosi,” March 31, 2025. Interviews concerning the search for walls, buildings, and late destruction were consulted. The nature of violence and social change remains open to interpretation; no specific invader or legendary war is identified here. https://news.cctv.cn/2025/03/31/ARTIa5S1hrzIsQMRMzhZfmWk250331.shtml 6. Dates are approximate, drawing on the established series research range and the archaeological material above. Early, middle, and late phases are archaeological divisions, not verified royal successions. No ancient individuals, dialogue, ritual sequence, weather, or private thoughts have been invented. The opening rain belongs to the modern excavation account. This is Clio Walker’s original AI-assisted editorial narrative. Modern studies supply factual checks and limited paraphrase; no protected text or image is reproduced in full.