Searching for the Production Site of Terracotta Warriors

Xu Weihong

Xu Weihong graduated from the Department of Archaeology of Jilin University and is a researcher at the Shaanxi Provincial Institute of Archaeology. He has participated in the excavation of the Terracotta Warriors and Horses of the Mausoleum of Qin Shi Huang for more than 20 years， served as the leader of the third excavation of the No. 1 pit， and is currently the archaeological leader of the Xianyang City Ruins Project in Qindu. He has been engaged in archaeological research of the Qin and Han dynasties for more than 30 years， has deeply realized the importance of telling Chinese stories well， and strives to popularize archaeological achievements to the public with his own meager efforts.

Shen Shen

Shen Shen is a Master of Cultural Relics and Museums and the 9th National Top Ten “Outstanding Disseminators of Chinese Cultural Relics Stories.”

Xu Weihong， the leader of the third excavation of the No.1 pit of the Terracotta Warriors and Horses and the excavation of the ruins of Qin Xianyang City， has been unremittingly “l(fā)ooking for relatives” and studying the history of Qin for 30 years. In this book， she presents us with many subtle discoveries from the archaeological excavation site of the Terracotta Warriors， which together reproduce a vivid picture of Qin Dynasty society: Along the vein of the Qin Warriors family， we can see how the Qin people innovated and inherited， the reality and the expectations.

The vast array of warriors required production spaces for three key activities: clay extraction， kiln facilities， and paint application areas. Where were all these located? This question has puzzled archaeologists for decades and remains unresolved to this day.

The production of pottery certainly requires clay， usually sourced locally. Conventional thinking suggests the clay for the Terracotta Warriors likely came from nearby areas， prompting archaeologists to exhaust every method at their disposal， including archaeological surveys to find clay pits， neutron activation analysis to identify the elemental composition of clays， and experimental pottery-making to determine suitable clay sources -- truly a Herculean effort.

The survey results were straightforward: No large-scale signs of clay extraction were found.

Neutron activation analysis has revealed rich details， predominantly indicating that the clay was sourced near the Mausoleum of the First Qin Emperor， yet the samples from the Terracotta Warriors exhibited some distinctiveness.

Why not just mix up some clay and fire it up? Only by pre-mixing loess and brown-red soil from the Qin Dynasty strata with about 20% sand can one achieve the standard necessary for molding clay figurines， producing terracotta warriors that most closely resemble the actual Qin warriors.

China boasts a rich and diverse range of soil resources. Classifications may follow 12 types based on genesis， or three types based on soil texture. Loess falls under genesis types， while clay is categorized by texture. Although categorized differently， both the testing team and the reconstruction group seem to agree， according to reports.

Black loess is hard in texture， becoming slightly more pliable as the ground warms up during the spring bloom， perfect for immediate tilling. A geologist friend mentioned that this type of soil is dark， rich in lime， has a viscous essence， and is primarily found in areas of mild soil erosion and flat terrain on the Loess Plateau. Worried that I might not grasp the cross-disciplinary jargon， my friend broke it down thoroughly.

I understood. Materials for producing the Terracotta Warriors are widely distributed outside the Mausoleum of the First Qin Emperor. Circling the mausoleum， I didn’t give much thought to whether our research approach was the right one.

In the summer of 2019， an archaeological dig in Xianyang City led to what could only be described as a miraculous discovery. This discovery doesn’t directly relate to the Terracotta Warriors production sites， yet it holds significant indirect implications.

“Come here， come here， come here， you come here. Look， what is this?” The field survey team leader， with a secretive and hurried， almost sneaky demeanor， urgently called out multiple times， signaling urgent matters. Opening his tightly clasped hand， he revealed two small stone fragments.

“Ah， armor tiles.” These types of tiles， from the burial pits accompanying the Mausoleum of the First Qin Emperor， have been unearthed in large quantities before. I am familiar with their shape， color， size， and manufacturing techniques. Suddenly， my heart raced， and I quickly asked， “Where?”

Over the following two months， through drilling， trial excavations， and comparisons， we uncovered an area of several thousand square meters under tall wild grasses and in the sweltering heat， just over 500 meters from the Qin capital Xianyang City palace ruins， where stone armor was produced. The raw materials， specifications， shapes， manufacturing processes， and uses of the copper bars for threading the armor tiles were identical to those found at the Mausoleum of the First Qin Emperor.

It’s prudent to say that the stone armor produced at the site near the ancient city of Xianyang likely ended up at the Mausoleum of the First Qin Emperor. The stone armor burial pit was discovered in 1998， and the production site was first confirmed in 2001， about 4.5 kilometers north of the mausoleum. Twenty years later， similar sites were discovered 40 kilometers away， mere hundreds of meters from the Qin palace.

The news immediately hit the headlines. The fact that stone armor was produced in multiple locations suggests that the Terracotta Warriors might have been produced under similar conditions. Leaving the familiar Terracotta Warrior archaeology， I had regrets. Little did I expect that those dead-end issues would suddenly spark a glimmer of hope at the Xianyang City site.

Must the production of terracotta figurines necessarily involve local materials and nearby firing? The discovery of a new stone armor processing site challenged this conventional thinking. The loess in the area of the Mausoleum of the First Qin Emperor is a subtype of black loess. This type of soil is discontinuously distributed across approximately 1，500 kilometers in China， with significant concentrations in the Dongzhi Plateau of Qingyang， Gansu， and the Luochuan Plateau of Yan’an， Shaanxi， covering a broad area.

The Qin Dynasty’s Guanzhong area was governed by an internal history officer. North of the jurisdiction of the internal history officer， centered around today’s Qingyang， was the Northern Commandery， and centered around today’s Yan’an， was the Upper Commandery. The saying “l(fā)ocal conditions nurture local people” holds true: the robust people living on black loess soil prove it. The robust body types of the Terracotta Warriors， modeled after the potters themselves， are accurately described in ancient texts. Holding the newly discovered armor tiles， I got carried away and overly excited.

The second issue concerning the production sites of the Terracotta Warriors: the kiln sites.

The concept of a kiln site is a broad spatial one. Before firing terracotta warriors， the processes of fuel storage， clay mixing， clay resting， molding， and air-drying each require ample space， not to mention the need for living quarters nearby， even if just simple， large communal sheds.

This， too， is conventional thinking. If clay sources can be broadened geographically， why can’t kiln sites be dispersed? While neutron activation analysts believe the materials for firing the Terracotta Warriors came from the vicinity， they also concluded that the three samples of Terracotta Warriors had relative independence， with the clays used not being completely identical. They specifically noted that the clay source for Pit No. 3 was more concentrated， while the clay sources for Pits No. 1 and No. 2 were more dispersed. Given the vast number of warriors in Pits No. 1 and No. 2， it would be impossible to fire them all in the same place， at the same kiln site.

Soil samples for composition testing were all taken from around the Mausoleum of the First Qin Emperor， with the sole exception of a porcelain shard from the Yaozhou Kiln area in Tongchuan. The figurines and porcelain inherently differ in material， and I wonder if comparing their compositions is meaningful. I’m curious how much the soil quality varies between other regions under the jurisdiction of the internal history officer， the Upper Commandery， the Northern Commandery， and the area around the Mausoleum of the First Qin Emperor.

We should not dismiss the possibility that some terracotta warriors were indeed fired locally， especially since at least three kiln sites from the Qin and Han periods are known to exist around the mausoleum and its vicinity. We should also not be constrained by conventional thinking. Just like another processing site for stone armor was discovered， who knows， one day in some place， a few pieces of terracotta warrior clay molds and deformed terracotta warrior torsos might just unexpectedly surface.

In the core area of the Xianyang City site， now known as Yaozhuang Town， there indeed are many pottery kilns from the Qin and Han periods. Dozens of kilns west of the palace area， aligned in rows， mainly provided bricks and tiles for palace construction and seemed unrelated to the production of terracotta warriors. Near the Wei River floodplain is another cultural heritage site known as the “Pottery Workshop Area” due to the dense distribution of kilns there， primarily producing everyday pottery. In 2017， some deformed pottery funerary objects were unearthed there， also seemingly unrelated to the production of terracotta warriors.

“Seemingly!” Archaeological discoveries are always unpredictable. We can only speak based on what is found， not hastily declare what isn’t. Over 2，000 years， soil erosion and the northward shift of the Wei River have submerged historical secrets， which is precisely what makes archaeological exploration so enticing.

How likely is it that Terracotta Warriors were produced within Xianyang City of Qin? Perhaps the archaeological discoveries in Western Han Chang’an City serve as a guiding light.

The Qin palace was renovated during the Han period. During the Western Han period， Emperor Liu Bang made Chang’an the capital， involving renovations and expansions of the palace district in Xianyang City. The northwestern corner of Chang’an City featured the Eastern and Western Markets， areas for artisan workshops and markets. Archaeologists have discovered both imperial burial terracotta warriors and pottery kilns for firing terracotta warriors at the Western Market site.

In 1990， 21 pottery kilns were discovered there， marking that year as significant. The timeframe of these kiln sites would not extend beyond the end of Emperor Jing’s reign or the beginning of Emperor Wu’s reign， and would not go later than the end of the Western Han period. Managed as a unified operation， they were officially sponsored kilns， with two of them still filled with bare terracotta warrior molds. Estimations show that small kilns could hold over 350 warrior molds， and large kilns around 450， with a single batch across the 21 kilns capable of producing approximately 8，638 bare terracotta warriors.

The terracotta warrior molds seen and the bare terracotta warriors unearthed are strikingly similar to those from Emperor Xuan’s burial pit in Duling， located northeast of Chang’an City. Duling is about 24 kilometers straight-line distance from the kiln discovery site.

During the Western Han period， the production and supply of imperial burial items were managed by the “East Garden Craftsmen” under the Lesser Treasurer’s jurisdiction. This system can be traced back to the Qin period， as evidenced by pottery from the Mausoleum of the First Qin Emperor， which was marked with the characters “East Garden.”

Qin laws stipulated that without an imperial edict， any unauthorized production of other items during the annual manufacturing cycle would result in a penalty of two ranks for both the craftsman and his supervisor. Unauthorized production without an edict required a penalty， and conversely， local authorities were bound by duty to execute any edicts for producing terracotta warriors.

“It’s very likely that the process involved centralized clay collection and storage， then distribution to the imperial craftsmen for uniform production. Each craftsman and their team would then choose different additives and manufacturing techniques based on the specific terracotta warriors they needed to produce，” speculated an expert.

Centralized collection， storage， and redistribution seem a bit redundant. Tasks that could be accomplished by craftsmen at Lishan could equally be done by their counterparts in other regions.

The selection and acquisition of clay materials were generally managed by the mining department. Twice it was assessed as substandard. Mining rated as inferior twice incurs a penalty of one rank to the miserly officer and an assistant shield. Rated as inferior for three consecutive years results in a penalty of two ranks to the miserly officer， with dismissal and no chance of future employment. Under the strict labor evaluation system and the “imperial edict” model for production， the making of terracotta warriors involved local clay collection and nearby firing， “near” concerning the kiln sites. Once the finished products were assembled in groups and moved to burial pits， the transportation pressures and various unforeseeable risks during transit must have been a concern， right?

To construct the Mausoleum of the First Qin Emperor， a large amount of stone from the northern mountains was transported down the Wei River over a hundred kilometers， at one point even blocking the river’s flow. In 220 BCE， the First Qin Emperor initiated a major project to construct a national network of thoroughfares， including express and post roads， centered around Xianyang. This formed a vast fan-shaped network radiating north， east， and south， securely incorporating various prefectures and counties. Today， there’s no need to doubt the transportation capabilities of the Qin Dynasty.