"A Magic Army for the Emperor"
Lothar Ledderose, Ten Thousand Things. Module and Mass Production in Chinese Art, Princeton, Princeton University Press, 1998, pp. 51 - 73.

Què s'hi va fer
Per què es va fer
Com es va fer

Of the many sensational archaeological excavations made since the founding of the People's Republic of China in 1949, the most sensational of all was the discovery of the terra-cotta army near the tomb of China's First Emperor. The First Emperor was one of the most powerful men in Chinese history and, indeed, in world history Originally the king of the state of Qin, he ruthlessly obliterated the other states of his day and unified the realm in 221 B.C., thereupon calling himself Qin Shihuangdi, the First August Emperor of Qin. The pattern of empire he established lasted for more than two millennia into the present century, and the name of his dynasty, Qin, is said to have given its name to China.

In March 1974, subterranean pits containing the emperor's terra-cotta army were discovered in Lintong County, Shaanxi Province, some thirty-five kilometers east of Xi'an, the provincial capital. The excavated figures have been restored and now stand in situ in their original battle formation. Since the site opened to the public on October 1, 1979, a never-ending line of visitors have made their way there. Some two million arrive every year, of whom about 17 percent are foreigners. To accommodate this traffic an international airport was built, opening in 1991. Having come to light again, the terra-cotta army has played a part in changing the economic fabric of Shaanxi Province.

Millions of visitors have also seen selected terra-cotta figures in exhibitions outside China. In 1976 two warriors and one horse were shown in Japan. For an exhibition originating in New York in 1980, eight figures came to the West for the first time. (1) Since then terra-cotta figures have traveled to one foreign country or another almost every year.

The most extraordinary fact about this archaeological find is a quite simple one: buried were not a few, or even a few dozen, lifelike soldiers but several thousand. About two thousand figures have been unearthed so far, and it is estimated that there are more than seven thousand altogether.

The figures are not only lifelike but life-size, of various types, including armored and unarmored infantrymen, standing and kneeling archers, cavalrymen with horses, charioteers, petty officers, and commanders. Armored infantrymen appear most often. Each held a spear or halberd in his right hand, and some possibly held a sword in their left hands. The charioteers wear caps that indicate their rank as officers and extend both arms forward to grasp the reins. The standing archers, with turned bodies, are dressed in simple and light uniforms that allowed for speed and maneuverability. The kneeling archers wear waist-length suits of scaled armor that simulates leather; their arms are flexed for cradling the crossbow. All details of the clothing, the armor, and the faces are modeled with great care, down to the stippled tread of the sole of the archer's sandal. (2)

The lifelike quality of these warriors must have been even more striking when they were still painted with their original colors, which indicated precisely the different parts of their dress. When these figures are unearthed, their colors are still visible. Upon excavation, however, most of the pigment adheres to the surrounding earth and not to the figures. Moreover, once exposed to air, the lacquer in which the pigments are embedded tends to crumble rapidly, reducing the colors to powder. Only traces last thereafter. Only recently has a method been found to stabilize the polychrome coating. (3) Because the problems of conservation remain unsolved, excavation work now proceeds at a slow pace.

The following discussion of this army, which the emperor commissioned for his tomb, will consider three questions: What was done? Why was it done? And how was it done?

What Was Done?

The Necropolis

The Grand Historian Sima Qian (born 145 B.C.) gives a detailed account of the tomb in his Records of the Historian (Shiji), which he completed in 91 B.C. (4) Work began as soon as the future emperor ascended the throne as king of Qin in 247 B.C., when he was thirteen years old. It is believed that the king, as was customary, put his chancellor in charge of designing the royal tomb and supervising construction. This was Lü Buwei (died 235 B.C.), who had been a trusted adviser to the king's father.

In 237 B.C., Lü Buwei fell from grace, and Li Si (ca. 280-208 B.C.), one of the ablest men of his time, followed as chancellor. He was instrumental in implementing the unification of the empire and in forging its administrative structures. He is personally credited with creating the stately, tectonic Small Seal Script for the imperial steles that were set up to glorify the emperor's unification of the realm. As mentioned in chapter 1, these steles set a precedent for the millions of stone steles that followed in succeeding centuries. It seems that Li Si also took extraordinary measures to prepare a worthy tomb for his ruler. It was probably he who, in 231 B.C., turned the area around the tomb into a government district with its own administrative center, named Liyi (District of Li) after the nearby Mount Li. The people in this district were responsible for the construction and later for the maintenance of the imperial necropolis. (5)

After the king became emperor in 221 B.C., the design for his tomb seems to have been expanded to a much larger scale. As the series of military campaigns had come to an end, large numbers of conscripts became available, and more than seven hundred thousand men from all parts of the realm were recruited to build the emperor's palace and his tomb. Most of them were forced laborers, slaves, and prisoners, "men punished by castration or sentenced to penal servitude," in the words of the Grand Historian. (6). Work on the terra-cotta army probably started at this time. (7)

In 212 B.C., Li Si had thirty thousand families resettled to the district, (8) but when the emperor died two years later construction stopped at once, even though his tomb compound was not yet finished. The laborers at the palace also ceased work to join the men at the tomb. All seven hundred thousand of them heaped earth on it during the following year. (9)

Today a large tumulus still occupies the center of the compound. (10) This artificial hill has the shape of a truncated pyramid, with a base of approximately 350 meters. The original height is said to have been about 115 meters. Erosion has taken its toll and reduced the tumulus to its present height of 76 meters, or less. The exact location of the emperor's tomb was thus known throughout history, even after the walls and halls above ground had decayed, since nobody could overlook the tumulus. The stepped profile of the tumulus is still clearly visible. (11)

Bushes and trees planted since adorn and protect the hill because Chinese archaeologists have decided not to excavate the tomb in our time. Knowing what wonders wait for them once they open the ground, they also know that they would be unable to preserve properly what they might find.

In a famous passage, Sima Qian tells us of the tomb's content:

As soon as the First Emperor became King of Qin, excavations and building had been started at Mount Li, while after he won the empire more than seven hundred thousand conscripts from all parts of the country worked there. They dug through three subterranean streams and poured molten copper for the outer coffin, and the tomb was filled with models of palaces, pavilions and offices, as well as fine vessels, precious stones and rarities. Artisans were ordered to fix up crossbows so that any thief breaking in would be shot. All the country's streams, the Yellow River and the Yangzi were reproduced in quicksilver and by some mechanical means made to flow into a miniature ocean. The heavenly constellations were shown above and the regions of the earth below. The candles were made of whale oil to ensure their burning for the longest possible time.

The Second Emperor decreed, "It is not right to send away those of my father's ladies who had no sons." Accordingly all these were ordered to follow the First Emperor to the grave. After the interment someone pointed out that the artisans who had made the mechanical contrivances might disclose all the treasure that was in the tomb; therefore after the burial and sealing up of the treasures, the middle gate was shut and the outer gate closed to imprison all the artisans and laborers, so that not one came out. Trees and grass were planted over the mausoleum to make it seem like a hill. (12)

The tomb thus contains a microcosm, an ideal model of the realm over which the emperor had ruled and intended to continue to rule after his death. Doubtless it will be difficult for any future excavator to preserve quicksilver streams and heavenly constellations. Indeed, the tomb may have been looted long ago. The Grand Historian talks about the destruction wrought upon the Qin empire and its capital Xianyang in 206 B.C. by General Xiang Yu: "Xiang Yu led his troops west, massacred the citizens of Xianyang, killed Ziying, the last king of Qin, who had surrendered, and set fire to the Qin palaces. The conflagration raged for three whole months. Having looted the city and seized the women there, he started east." (13)

General Xiang Yu is also said to have dug up the emperor's tomb. (14) At the same time he may have destroyed the underground pits housing the terra-cotta army Excavations have revealed that they have been burned.

Above ground the rectangular layout of the imperial tomb resembled that of a palace with an outer and an inner wall. Except in a very few places, these two walls made of pounded earth have completely vanished. The walls measured 8 meters in width, and they are believed to have had an original height of 8 to 10 meters. The inner wall was 1,355 meters long and 580 meters broad. The outer wall was 2,165 meters long and 940 meters broad. Watchtowers guarded the four corners, and gates opened at the four sides.

The compound contained buildings of many sorts, making it a necropolis, a city for the dead. Several foundations of mighty architectural structures have been located and excavated. About 53 meters north of the tumulus lay a large square hall, enclosed by a covered corridor 57 meters wide and 62 meters deep [núm. 1 al plànol]. It was the retiring hall (qindian) and contained the emperor's garments, headgear, armrests, and walking staffs. Once every month these relics were taken in a procession to an Ancestor Temple of the Absolute (jimiao). After having received due sacrifices and veneration, the emperor's personal objects were carried back. (15) The Ancestor Temple of the Absolute was dedicated to the First Emperor only It is believed that it was situated to the south of the necropolis and connected to it by a road. (16)

According to literary sources, in earlier times a retiring hall and a temple hall (miao) belonged to an ancestor-temple compound in the capital, where all the rulers of one lineage received sacrifices. Transferring the retiring hall to the necropolis and making it the ceremonial center there was a means to augment the importance of the ruler's posthumous presence.
North of the retiring hall, but still within the inner wall, foundations of more buildings have been identified [núm. 2 al plànol]. These may have been side halls (biandian), where the visiting members of the imperial family put on mourning dress and prepared for making sacrifices. (17)

Apart from these special occasions, routine sacrifices of food and drink were made three or four times a day These and other activities were administered by an office organized like the imperial household agency at the capital. Cooks prepared the sacrificial food. At the west side of the necropolis, between the inner and outer walls, the remains of three buildings have been located, and inscriptions on ceramic shards found there identify them as the provisions office (siguan) [núm. 3 al plànol]. Ceramic drainage channels, stone pillar bases, iron tools, and objects of daily use have been found here. (18)

About 20 meters west of the tumulus and 7.8 meters below ground, two bronze chariots, each drawn by four bronze horses, have been unearthed [núm. 4 al plànol]. They had been waiting there for the imperial Lord who, at some point in time, may have wanted to leave through the nearby gate and be driven around. Large quantities of hay were deposited for the horses. Although the chariots were badly damaged by earth that had fallen into the pit, it has been possible to restore them to their original splendor. They are about half the size of real chariots. The exquisitely made models exemplify the techniques of chariot building and horse harnessing in great detail. One of the chariots consists of 3,462 individual parts. (19)

Beyond and slightly to the south of the western gate of the inner wall, burials of thirty-one rare birds and animals in clay coffins lay in rows [núm. 5 al plànol]. In addition to the skeletons, the coffins contained clay dishes for food and collars attached to the animals. Clay wardens guarded them. (20). In life, the precious animals probably inhabited an imperial pleasure garden or the animal enclosures in the Supreme Forest (shanglin), the extravagant hunting park where the emperor kept rare specimens of flora and fauna from all over the world. (21)

An area about fifteen hundred meters long and fifty meters wide, to the east of the outer wall near the present village of Shangjiaocun, holds three hundred to four hundred pits, close to one hundred of which have been surveyed. Each pit either contains the skeleton of a horse or the lifesize terra-cotta figure of a kneeling groom, or both. The grooms are of exquisite quality and resemble the figures of the army The horses lie facing the center of the tomb. Most of them seem to have been buried alive. Some of the horses, however, have slashed limbs, indicating that they were first slaughtered before they were placed in wooden coffins. Inscriptions on ceramic shards prove that the animals came from palace stables. (22)

Test excavations in the southwestern corner of the necropolis revealed an L-shaped pit, more than one hundred meters long and nine meters wide. Another some three hundred skeletons of horses were stacked in it and accompanied by terra-cotta grooms [núm. 6 al plànol]. This pit must also represent one of the imperial stables.

Human skeletons have also been found in the necropolis of the First Emperor. A few meters to the west of the pits with horses and clay grooms near Shangjiaocun, a row of seventeen pit tombs with ramps and wooden coffins has been located, each of which bears the remains of one person, either male or female. Objects of gold, silver, jade, and lacquer and fragments of silk indicate their high social rank. It seems that they had all been put to death and their limbs severed. (23) It is not clear yet whether these people asked to follow their lord to the grave, whether they were sacrificed in religious rituals, or whether they were the victims of political intrigue. The Grand Historian tells us that the son of the First Emperor, after he usurped the throne, ordered the executions of numerous princes and princesses as well as ministers loyal to his father. Others were pressured into asking to be allowed to follow their former lord in death and to be given a resting place in his tomb "at the foot of Mount Li." (24)

More than one hundred human skeletons lie in mass burials about 1.5 kilometers west of the tumulus near the village of Zhaobeihucun [núm. 7 al plànol]. Except for three women and two children, most of them were men aged between twenty and thirty They were workers at the necropolis who may have been sentenced to death. At least twenty-six men were accompanied by small individual clay tablets with a written record. Eight of them show stamped-seal impressions of the government office to which the dead workers had belonged. The other eighteen shards have short inscriptions noting each man's name and his native place. Ten men were forced laborers working off debts (juzi). Sometimes their rank, which had an influence on their sentence, is specified. [This] inscription identifies a man who came from a faraway village in present Shandong Province. It reads: "Forced laborer Sui, rank bugeng, from Dongjian Village, Dongwu City" (25) Bugeng was stage four in an ascending series of twenty ranks. (26) Hence laborer Sui was a petty officer, probably working off the redemption of a punishment. These are the earliest tomb inscriptions found in China so far and reflect the bureaucratic control over the workforce.

The Army in Its Pits

The most spectacular burial outside the tomb proper is, of course, the terracotta army [núm. 8 al plànol]. (27) The Grand Historian does not mention it, nor does any other historical source. Its discovery in 1974 came as a complete surprise. There is a cluster of four separate pits 1,225 meters east of the outer wall. Pit no. 1, which is 230 meters long and 62 meters wide, contains the main army in battle formation with more than 6,000 figures of warriors and horses. Pit no. 2, with various cavalry and infantry units as well as war chariots, has been explained as representing a military guard. The small pit no. 3 is the command post, with high-ranking officers and subordinates and a war chariot drawn by four horses. Several of the figures lack heads, which are believed to have been stolen by grave robbers as early as the Qin or the Han period. (28) Pit no. 4 did not contain any figures and was probably left unfinished by its builders. Together, the four pits seem to represent a complete garrison: pit no. 1, the right army; pit no. 2, the left army; pit no. 4, the middle army; with the headquarters in pit. no. 3.(29)

The pits' architectural structures were devised for solidity and permanence. The outer walls and the walls between the eleven parallel corridors in pit no. 1 consist of pounded earth. The earthen walls were originally held in place by wooden frames that also supported the roof beams. The roof, in turn, carried a layer of reddish mortar and a layer of earth three meters thick. The floor was also made of pounded earth as hard as cement, and altogether covered by some 256,000 tiles. It has been calculated that about 126,940 cubic meters of earth were moved to excavate the pits, and that 8,000 cubic meters of timber were needed. (30)

The solid wooden construction must have been finished before the figures were put into place; otherwise their installation would have been too dangerous. At the front side of the pit, ramps have been identified down which the figures were hauled into the long, probably torch-lit corridors. This prompts an intriguing thought: nobody, not even the First Emperor, ever saw the terra-cotta army in its entirety The breathtaking view of the now world-famous columns of soldiers only became possible after excavation in 1974. Obviously, the army did not need to be seen to serve its purpose. It was enough that it was there, like inner organs concealed in a human body.

Beams and tiles provide clues for dating the terra-cotta army The Grand Historian mentions that timber for the emperor's palace and his tomb was shipped from the region of present-day Sichuan and from the state of Chu. (31) The sturdy pine wood in the subterranean pits is believed to have come from these southern parts of the empire, and this probably happened only after unification, when the emperor had the finest material from all parts of the country at his disposal. Chu was subjugated only in 223 B. C.

Stamped characters in many of the floor tiles identify their makers. The character meaning "metropolitan" (du) is an abbreviation for "Metropolitan Boats" (duchuan), a factory under the commander of the capital (zhongwei). The character for "palace" (gong) is an abbreviation for "Palace Water" (gongshui), a factory that was part of the imperial manufactures (shaofu) and responsible for all kinds of waterworks. Both factories seem to have made tiles only after the unification. (32)

Of the mighty timber construction only traces of ash remain. As mentioned, it is believed that general Xiang Yu, when looting the First Emperor's capital Xianyang in 206 B. C., also set his necropolis on fire. The burning beams collapsed and subsequently the earth slid down and smashed all the terra-cotta figures. None has come to light intact.

The army in pit no. 1 was set within the corridors in rows of four soldiers, many of them clad in heavy armor, all modeled in clay In six of the corridors, wooden war chariots, each drawn by four terra-cotta horses, are spaced in regular intervals among the infantry The army is facing east. Three rows of warriors form the vanguard, two rows protect the army at its sides, facing north and south respectively, and a rear guard of three rows faces backward.

The movements of this awesome fighting machine would have been controlled by audible signals from drums and bells, which have been found at the site. When the drum sounded once, the magic army moved forward, on the second drum roll, it attacked. On the sound of the bell, the troop stopped, and when the bell rang again, it retreated. (33)

The figures were kiln fired between 900 and 1050 degrees centigrade. (34) At this relatively low temperature, the unglazed clay remains porous and is called "terra-cotta," literally baked earth. Although several kilns for making tiles and other utensils have been explored in the area of the necropolis, no kiln for firing the figures has yet been excavated. (35) About two hundred meters to the southeast of pit no. 1, fragments of terra-cotta figures indicate a kiln site that is estimated to have been large enough for firing two horses or six warrior figures at a time. (36).

After firing, painters applied a lacquer coating colored by brilliant pigments. Lacquer is a natural resin obtained in small quantities by preparing the sap acquired by tapping various lacquer trees (Toxicodendron verniciflua). (37) The trees grow in tropical and subtropical areas, and were native to southern China. As lacquer is highly resistant to water, heat, and acids, it is well suited for protecting objects and is also amenable to surface decoration. Carefully made war gear from many areas in East Asia often bore a lacquer coating to protect it from rot and rust. Chinese examples from the centuries preceding the unification of the empire include armor made principally of lacquered leather. (38) Most of the armor molded on the terra-cotta soldiers represents leather armor. With their lacquer coating, the terra-cotta figures became more durable and seemed even more "real." (39)

To completely coat thousands of large clay figures in lacquer is an astounding achievement in itself, but certainly not beyond the imagination of an emperor, whose son once considered lacquering all the city walls of his capital. (40) In some areas of the figures, such as the faces, two separate coatings were applied. The mineral pigments were also mixed with lacquer and painted with a brush on top of the coating. The geometric patterns are very intricate, like those of the textiles of the period. (41)

The terra-cotta figures carried real weapons, such as spears, halberds, dagger axes, swords, crossbows, and arrows. Almost five hundred weapons and more than ten thousand scattered arrowheads have been found in pit no. 1. Save for four iron pieces, all weapons were cast of bronze. Some of the blades are still razor sharp. (42)

The lock mechanism of the crossbows is ingeniously devised. After the archer discharged his arrow he brought the lock back into the original position by quickly jerking the bow backward. The four mechanical parts are cast with such precision that they fit together perfectly The tolerance for error lies within fractions of a millimeter. It was this precision that helped the Qin state overpower the rival feudal states.

Yet precision alone was not enough. It had to be matched by the ability to produce large quantities of weapons. The Qin could do this, too. Indeed, their weapons industry was another early example of mass production in China. It is known from inscriptions on the weapons that many came from state factories capable of churning out huge numbers of technically perfect products. (43)

Mass production of weapons started in Qin long before the time of the First Emperor. Inscriptions began to appear in the middle of the fourth century B.C., the law decreeing that each weapon had to be inscribed. (44) For more than a century these inscriptions identified the chancellor as supervisor of weapons production, testimony to the importance attached to the weapons industry. The weapons carried by the terra-cotta army date from the early years of the king's reign to 228 B.C. As later weapons have yet to be found, they may all have been used by real warriors before being entrusted to the hands of the clay soldiers.

As late as 239 B.C., the chancellor was still named as the supervisor, but at this date the names of individuals from lower ranks in the production hierarchy began to appear on the weapons as well. The inscription on one such blade reads: "17th year [230 B.C.]. Government Workshops, produced by master Yu, worker Diao." The characters for "Government Workshop" (sigong) occur once ore on the other side of the blade and on the guard as well. Engraved on the shaft is the serial number: "Series zi, five-nine." (45)

The Government Workshops produced not only weapons but also carriages and utensils for everyday use. Their products are among the first in China on which an individual inscribed his name. It would certainly be naive to try to detect budding individualism here, nor do the names indicate any personal pride on the part of the maker. These inscriptions had only one purpose: quality control. (46)

This visit to the First Emperor's necropolis has shown that it was laid out like the palace of a living emperor, with the private quarters under the tumulus and a main hall above with various palace offices, a pleasure garden, horse stables and carriages, tombs of family members and loyal retainers, and with a mighty army, ready to attack any intruder.

Why Was It Done?

This question calls for another, closer look at the historical situation. Unification of the empire doubtless was the single most outstanding event of the period, if not of Chinese history. (47) Many other efforts preceded, accompanied, legitimized, and glorified this event. Building the imperial tomb was one of them. This political context explains the grand scale of the project and the unusual efforts lavished on the terra-cotta army. Moreover, and more specifically, the builders of the necropolis were continuing a time-honored tradition of royal tomb building. The First Emperor or his chancellors may have wished to reinforce this tradition, or to give it a new direction, but they could not abandon it, and did not want to do so. Two separate sets of factors thus determined the design of the necropolis, the historical situation and a tradition of furnishing rulers' tombs.

Standardizing Society

Of all the emperors that ever considered themselves to be the mightiest man in the world at a particular time in history, the First August Emperor of Qin was perhaps the only one who was right. His power was rivaled but not matched by the ruler of the Maurya dynasty in India-which was just past its prime under Emperor Ashoka (about 274-232 B.C.)-and perhaps by the nascent Roman Empire. The Romans, however, had yet to face the battle at Cannae in 216 B.C., a severe setback on their way to achieving eventual supremacy over the Mediterranean Sea. When the king of Qin unified China in 221 B.C., he could hardly have had any knowledge of those empires in other parts of the ancient world. Yet he brought under his control all the civilized world known to him, and thereby-unknown to him-created the mighties empire on the globe, the largest in both area and population.

During the Period of the Spring and Autumn Annals (Chunqiu; 722-481 B.C.), the area of China, although in name still under the control o the rulers of the Zhou dynasty, was split up into rivaling feudal states engaged in constant warfare. Indeed, the following period (453-221 B.C.) is called the Period of the Warring States (Zhanguo). The Qin established their power base in the eighth century B.C. in what had been the original heart land of the Western Zhou. This allowed them to view themselves as the Zhou's legitimate heirs. The First Emperor even claimed to succeed the Yellow Emperor and other mythical rulers of remote antiquity In the middle o the fourth century, the feudal state of Qin was still rather small, but then i expanded forcefully in all directions. Only seven states were left when the First Emperor became king of the Qin state in 247 B.C. The young king soon pursued a policy of the iron fist. In 230 B.C. he embarked on a decade-long series of military campaigns, conquering the remaining feudal states one after the other. Within two years after his principal rival, the culturally superior southern state of Chu, succumbed in 223 B.C., Yan and Qi, the lass northern states, had also fallen.

Many factors came together to allow for Qin's success. Among the technological advances made during the previous centuries, smelting, forging and casting iron had developed apace. (In the West, iron was not cast until the fourteenth century A.D.) Farmers began to use iron ploughs, which allowed them to break the earth deeper and faster, thereby boosting agricultural pro. duction. Blades for spades and hammers were also fabricated of iron. (48)

Metal technology was equally vital for producing weapons, the very tools by which the Qin destroyed their neighbors. Crossbows and other deadly weapons in the hands of the terra-cotta soldiers show how successful the Qin were in this domain. Yet other feudal states had made advances in iron and weapon technology as well. The crossbow was actually invented in Chu, and the swords of the states of Wu and Yue in the lower Yangzi region were praised throughout the realm for their supreme quality Qin, however, surpassed al other states when it came to organizing masses of people and coordinating their efforts toward ambitious goals, in the civil as well as the military sphere

In the civil sphere, law and order were Qin's major values. Households were organized in units of five and ten and were held jointly responsible for misdoings committed by any member. A law code promised draconian punishments but also equal justice to everyone. The code, fragments of which archaeologists have discovered written on bamboo slips, has been ranked among the most influential legal systems in world history. (49)

The Qin administrators knew how to foster efficiency through standardization. They developed commerce by regulating weights and measures and the axle lengths of carts; they forged a monetary union by standardizing coins; and, as seen in chapter 1, they created a uniform system of script.

Extensive building activity also testified to the Qin's capability in successfully organizing huge labor forces. After ascending the throne, the First Emperor embarked on a number of large construction projects that served to consolidate political, commercial, and cultural unity, and to glorify his achievements. He connected existing portions of walls built by some feudal states thereby anticipating in effect what has become the most famous of all Chinese monuments: the Great Wall. A network of highways radiating from the capital of Xianyang linked together faraway places in the new empire. One such road that has been archaeologically identified led from Xianyang straight north over a distance of eight hundred kilometers into present-day Inner Mongolia. Canals were renovated and newly dug, completing a system of waterways from the Yellow River in the north to the Huai River in the south. The emperor had a giant palace, called Apanggong, built for himself, which, although destroyed soon after construction, has been remembered since as the epitome of ostentatious architecture. Last, but not least, the emperor could look forward to moving into the greatest necropolis China had seen to date.

The Qin were also more successful (and probably more ruthless) than their competitors in forging efficient military organizations. They pushed to the limits a process that changed warfare fundamentally: the deployment of mass infantry This momentous development has its parallels in the West. In his epic poems of the eighth century B.C., Homer extols the virtues of the aristocratic warrior whose strength, skill, and courage decide the outcome of the battle. Achilles slaying Hector with his spear and then dragging the corpse behind his chariot through the dust is one of the most memorable scenes from the dawn of European history

In the fifth century, however, the Greeks prevailed over the Persians thanks to their development of the phalanx, which was composed of infantry citizen-soldiers trained to act together. (50) Large bodies of infantry, in addition to cavalry, also formed the core of Alexander's fighting forces when he set out to win the Orient in the fourth century B.C. With their infantry legions, the Romans conquered the Mediterranean world. The flexible, disciplined, and obedient cohorts of anonymous mercenaries marching under a unifying command proved invincible.

In China, infantry existed as early as the Shang period. Oracle bone inscriptions mention forces of up to 13,000 men. (51) Fighters on horse-drawn chariots accompanied the foot soldiers and directed their movements. The precious chariots were status symbols to the aristocrats and consequently they were buried next to their owners' tombs. After the Shang, this type of warfare continued and the number of chariots increased. The state of Jin mustered 700 of them in a battle of 632 B.C. (52) It has been estimated that an army with chariot warriors would not have far exceeded 10,000 men in the seventh century, and that it may have reached a maximum of about 50,000 men in the sixth century

Then, after the mid-sixth century B.C., the size of armies increased dramatically and they consisted mostly of infantry formations. In the Period of the Warring States, infantry armies could deploy several hundred thousand men in the field, because soldiers were now recruited from lower social levels, especially from the rural population and from larger territories. (53)

Conscripted men in those armies could not possibly possess the martial skills that the aristocratic fighter acquired through years of training and constant testing in hunts. Nor could the lower-class warriors boast expensive equipment such as a technically sophisticated chariots, custom-made armor, and ornamented weapons. Nevertheless, the infantry armies won out through sheer mass, standardization, organization, and discipline.

These soldiers' uniforms were not fancy but they were efficient. The same can be said for the standardized, mass-produced weapons. An army was organized into a composite body of units, each with its different weapons and tasks: vanguard and rear guard, hand-to-hand fighters, crossbow archers, spearmen, and swordsmen fought simultaneously in a force marked by its "division of labor." Single fighters and groups of fighters became interchangeable, modular units.

Discipline and absolute obedience were the supreme virtues of the soldier. Outstanding individual courage and unchecked bravery were not called for. One story tells of a bellicose soldier who, before battle began, charged over to the enemy lines and returned after having killed two of the foe. His commander had him executed on the spot for disobedience. (54)

The commander directed all movements and actions of the mass army He took all the credit for victory, and all the blame for defeat. His authority was absolute, mirroring that of the ruler of the state. The commander's standing was based, in part, on his knowledge of military treatises. He was a master of texts and did not even need to excel as a fighter himself. Above all he was an integrator of men and an organizer. (55) His role, in some respects, resembled that of a manager or even a bureaucrat.

The shift from warrior elite to mass army occurred in all the feudal states in China, but most radically in Qin. Qin forged a social order in which civil administration and military structures converged. The rise of an individual in a system of aristocratic ranks, the amount of land allotted to him by the state, or his salary as an official depended on the number of enemies he had killed as a soldier. (56) The organization of the civilian population into units of five for mutual surveillance and liability also applied to the military sphere. Members of one unit were held mutually responsible for one another's performance in battle. (57)

Seen in this context, the terra-cotta army is a monument glorifying Qin's superiority in military organization. At the same time, this advanced war machine, which proved so irresistible to its neighbors, is a model of Qin society itself.

None of this, however, yet explains why the figures were buried underground, and why they were made of terra-cotta. As far as we know, this was the first terra-cotta army To appreciate what that meant, one must take a look at the tradition of burials installed alongside and around Chinese royal tombs.

Fitting out Burials for Rulers

From the earliest dynasties down to the mausoleum for Mao Zedong built in 1976, tombs for their rulers have been a major concern of the Chinese. The Han dynasty rulers, for example, are said to have spent one third of the state revenue on imperial tombs. (58) As seen in the preceding chapter, the Shang constructed gigantic tombs, and their belief in an afterlife necessitated the burial of dozens of humans and horses with the lord of the tomb. The Zhou continued this practice, and the number of horses and humans interred increased with time. The pits around a sixth-century B.C. tomb in the capital of the northern state Qi in present Shandong Province have a combined length of 215 meters. They contained the skeletons of probably more than 600 horses, 228 of which have been excavated. (59)

The early rulers of the Qin State, like their rivals in other parts of China, indulged in the construction of large tombs. Thirteen tombs have been found in a cemetery in Fengxiang County near the ancient city of Yong, the capital of the Qin State from 677 to 384 B.C. Tomb no. 1, which is believed to belong to Duke Jing (577-537 B.C.), is the largest tomb in pre-imperial China known so far. The pit of this gigantic structure, including the ramps, is 287 meters long. Skeletons of 166 men and women buried together with the duke have been identified. Another 20 skeletons were found in the earth used to fill the pit. The excavators had to dig 24 meters down to reach the bottom, after which they were so exhausted that they were not able to publish a report. (60)

By this time a trend was under way in China that eventually led to fundamental changes in tomb contents and methods of construction. Ultimately, the very concept of what a tomb should be began to change. Vessels made specially for burial replaced bronze vessels that had actually been used in ritual, and objects of lesser quality or cheap copies in clay became acceptable as burial items. These substitutes were called spirit utensils (mingqi). The first such items occur in the late Shang dynasty, and their numbers increase after the eighth century. (61)

Similarly, real humans and horses were sometimes replaced by figures of humans and horses. Early examples made of ceramic or wood occur in the sixth century B.C., and the custom gained acceptance during the Period of the Warring States. (62) The largest numbers of tomb figures that have been preserved are made of wood and found in burials in the state of Chu. (63)

These changes in the character of the objects that went into a tomb were followed by changes in tomb construction. Early tombs constitute a building type in its own right that had little in common with above-ground architecture. They are basically pits, the larger and deeper ones being accessible by ramps. In the Period of the Warring States, however, tombs began to acquire features of dwellings used by the living. By the Han dynasty a distinct tomb architecture existed, making use of bricks and hollow tiles and featuring doorways, a chambered layout, and vaults. (64)

An early example of a chambered tomb is that of Marquis Yi of Zeng (after 433 B.c.). Found intact and excavated in 1977-78, it yielded fifteen thousand artifacts, including as mentioned earlier, an enormous quantity of bronzes. The tomb has four chambers built of wooden logs and connected through small doors. Each chamber has a different content and function. The central chamber with the famous set of sixty-five bells and ritual vessels is modeled on a royal ceremonial hall. Then there is an armory with various weapons and chariot trappings, a harem with the coffins of thirteen female attendants, and the private or retiring quarters of the lord of the tomb, containing the coffins of the marquis himself and of eight more of his female companions in death. Painted on the lord's inner coffin were two doors and a window. (65)

As tombs began to resemble dwellings of the living, they also came to contain more utensils of daily use, such as dishes. In addition, miniature models, usually made of clay, substituted for actual items that the buried person had needed or cherished in life, including stoves, granaries, and animals.

A shared pattern of thought is evident in burying effigies instead of living beings, clay copies instead of bronze vessels, and in laying out a tomb like a house. All this turns a tomb into a replica, a model of reality The tomb represents, idealizes, and perpetuates the reality of life on earth. By making ideal versions of the world, the tomb designers of the first millenium B.C. created the first paradises in Chinese art.

At the time of the First Emperor, the shift in the function of the tomb was not yet complete. His necropolis represented his palace, and the main chamber under the tumulus contained a replica of the universe. Pits represented imperial stables, with models of imperial chariots and an army in replica. Yet not everything was a substitute: the pit stables contained the skeletons of real horses and real nobles and forced laborers were buried in
the necropolis. (66)

Apparently, man-made replicas were not seen as being less effective than real things or living beings, nor did material or size matter. Terra-cotta grooms guarded the burials of real birds and horses. Half-size bronze chariots and bronze horses were as useful as life-size clay soldiers with real weapons and clay horses drawing wooden chariots. All served their emperor equally Why then did the builders of the necropolis choose terra-cotta figures for the army instead of real soldiers?

Pragmatic considerations may well have influenced this decision. Rationalists surely pointed out that dead men and horses, chariots, bronze utensils, and weapons buried underground simply could not be used anymore, and that burying them was wasteful rather than pious. Humanitarian feelings probably also came into play The great humanist Confucius (551-479 B.C.) lamented the practice of burying people and professed ignorance about the workings of the netherworld.

Yet the Qin rulers had traditionally liked large companies in their graves: 166 human skeletons were in the tomb of Duke Jing, and 177 persons accompanied in death Duke Mu (659-621 B.C.) (67) Human sacrifice was widely practiced in the Warring States Period, and although a king of Qin attempted to ban the practice in 383 B.C., princes and workers were still slaughtered at the First Emperor's necropolis. (68) As the Grand Historian and others tell us, the First Emperor was a cruel megalomaniac, ordering all artisans and laborers to be imprisoned in his grave and buried alive after they had finished their jobs. (69) Had he been convinced that this was the best way to protect himself in his tomb, he might conceivably have sacrificed seven thousand men.

There must have been other reasons why the emperor preferred figures to real soldiers, and why they were not made of wood but of clay, which may never have been used before for life-size sculptures. As no written sources are extant, one must deduce these reasons by examining the army itself and asking what its makers set out to achieve. Apparently they pursued three goals: the army had to be durable, it had to be finished within a reasonable time, and it had to look "real."

The emperor was obsessed with immortality. Unlike his palace in life, his posthumous residence was to last eternally What he needed, therefore, was everlasting protection. He and his advisers may already have been sufficiently accomplished archaeologists to know that human bodies decay fast, even in sturdy tombs. Certainly, figures made of clay would be more durable than humans of flesh and blood, and also more durable than wooden figures. Subsequent history has proven this reasoning right. The army is still extant-at least so far. The clay figures were smashed but did not burn during the conflagration of the necropolis. Nor did they decay in the water that seeped into their subterranean hideout. Only after men who no longer believed in the figures' magic power began to unearth them did their final, perhaps irrevocable, decay begin.

The workers also met their deadline-almost. Although the emperor fervently tried to acquire the elixir of immortality in his lifetime, the makers of his army did not take any chances. Working on a tight schedule, they filled three of the four pits in the subterranean garrison with figures, which shows that they had given themselves only a few years to complete everything. Clay was the most expedient material to do this.

Clay is found in loess, and large areas of China's northern plains are covered with layers of loess, three hundred meters deep in some places. Where loess has been washed by water in the course of the millennia it may become enriched with clay and suitable to be formed and fired. Clay-bearing loess was and has always been available in China in almost limitless supply and could be formed into bricks or tiles or pots. (70) At the time of the First Emperor, wood was more abundant in China than today; nevertheless, it was probably cheaper and logistically easier to use wood for firing rather than for carving. Clay can be brought into a certain form more quickly; it does not require metal tools to do so, and it also lends itself better to division of labor. If the emperor's artisans had ventured to carve thousands of life-size soldiers in wood instead, their task would have been more laborious, and they might not have finished in time.

A lifelike appearance of the army must have been an urgent concern of the emperor. Probably only insofar as the army looked "real" could it successfully fulfill its magical function. No written evidence explicitly confirms this assumption, but the history of religious sculpture after the Han dynasty makes it evident that realism and the efficacy of magic went hand in hand. (71)

The wish to equip the clay soldiers with real weapons that had proven their usefulness in earlier battles must have been the main reason why the figures were made life-size. Yet for the entire army to look real, two more features were essential: numerical significance and credible variety. In order to be convincing as a fighting force, the number of soldiers had to be considerable, approaching or equalling that of a real army On the other hand, enough variety among figures was needed if they were to be more than lifeless mannequins. But the makers of the army worked hard and ingeniously to achieve these goals, and clay was the material best suited for the task.

The emperor's order thus might have read, "Make me a magic army It must never decay but protect my residence for eternity it must look like a real army in all respects. Only then will the magic work!"

How Was It Done?

The logistical problems in procuring material and men must have been considerable. A sufficient supply of clay, firewood, lacquer, and pigments had to be constantly available at the necropolis. Although none of the workshops and kilns have been located so far, the figures were presumably formed, fired, and painted near the pits to avoid costly and hazardous transportation of the semifinished or finished products. Only a well-observed sequence of steps in the production process and a tight schedule could guarantee that work went smoothly.

No less demanding was the task of organizing the labor force. Once the workers were recruited, they had to be fed and given shelter. Their efforts had to be coordinated and the results supervised. Moreover, the makers of the terra-cotta army were but a small group among several hundred thousand busy men at the site. There is no doubt that all this activity necessitated detailed planning and an organizational framework.

The Workforce

What kind of workers could possibly have had the experience necessary to fire such large clay figures in such quantities? The answer: the makers of drainage pipes! An impressive drainage system made of clay was found under the provisions office in the necropolis and under the Qin palace. [There are] two pieces excavated from the pounded-earth terrace that formed the substructure of Qin palace no. 1. Both pipes have round cross sections of different sizes at each end. The diameter of the knee-shaped tube is 26-27 centimeters at one end and 21-22 centimeters at the other. The larger tube is 56 centimeters long. In size and proportion these tubes resemble the legs of the terra-cotta warriors. The technique used to make the drainage pipes must have been similar to the technique for making circular torsos, legs, and arms from slabs of clay rolled into tubes.

Inscriptions confirm this observation. The foremen of the palace workshops used to stamp their names on floor tiles and roof tiles. Some of the same names have been discovered on the terra-cotta figures. (72) This indicates that the makers of the terra-cotta army, rather than being sculptors, were potters who knew how to manufacture ceramic architectural parts.

The inscriptions on the terra-cotta figures also enhance our understanding of the organization of the workforce. When the first 1,383 warriors and 132 horses were excavated, archaeologists gathered preliminary statistics. They counted 477 inscriptions. The characters had either been incised with a stylus or stamped before firing, mostly on inconspicuous parts of the body Two hundred thirty inscriptions consist of serial numbers only. Most frequently, one finds the numbers five (40 times) and ten (26 times). This indicates that the figures were counted in groups of five and multiples of five, as was customary with real men in civil and military administration. (73)

Altogether, on 249 figures the names of 85 different foremen appear. Eleven of them preface their names with the character meaning "palace" (gong). Once again, this is an abbreviation for the Palace Water factory and also is seen on floor tiles of the pit. The foremen in this factory usually stamped the inscription into the soft clay with a seal, for example: "[Foreman] Jiang of the Palace" (gong Jiang).

A second group of 23 foremen put a place-name in front of their names. In all but three cases it is the capital Xianyang. One such inscription reads "[Foreman] Ge of Xianyang" (Xianyang Ke). This type of inscription is always incised and never stamped, which makes it somewhat less official than a seal imprint. It is believed that these masters came from local factories.

The third group is the largest, comprising 48 men who neither added the name of a factory nor a place-name to their own names. Most of these names occur only once. Those men probably also belonged to either one of the first two groups.

The inscriptions reveal that staff members from state factories and workers from local workshops pooled their efforts in one big project. This is common in later periods, too, for example in Han-dynasty lacquer workshops or Qing-dynasty porcelain factories. Typically, workers from state factories set standards of quality and enjoyed a somewhat better position than private workers. They were allowed to use seals and were given a higher proportion of the work. The names of the eleven men from the palace workshops are found on 87 figures; two of them made 21 figures each. By contrast, together all 23 local foremen produced only 56 figures.

Yuan Zhongyi, the chief excavator of the terra-cotta army, has described stylistic differences between the figures made by the two groups. Not surprisingly, the figures from the palace factories are more static, their bodies more sturdy, and their expressions stern and heroic. They also show a more consistent level of workmanship and greater stylistic uniformity Figures from local workshops are more varied in their postures and faces, and tend to be more realistic. Yuan Zhongyi even talks about the styles of individual masters. (74)

It was hardly artistic pride, however, that caused the foremen to put signatures on the figures they had made, nor did those who commissioned the figures look at signatures in this way As in the case of the weapons, quality control was the sole purpose of the inscriptions. The numbers served to verify how many figures had been completed; the names guaranteed the quality of craftsmanship. If the overseers found a figure to be faulty, they were able to track down a man whom they could hold responsible. Precise laws specified the fines in such cases. (75)

The foremen did not work alone, of course. They probably had the status of masters (gongshi), and each of them controlled a team of perhaps ten workers. (76) The 85 foremen identified so far would then have directed 850 men. Since many terra-cotta figures have not yet been excavated, presumably more names of foremen will come to light. The total workforce may have comprised a thousand men. Assuming they set up shop only after the unification of the empire in 221 B.C. and continued until the emperor's death in 210 B.C., they would have completed more than 7,000 figures in eleven years, or close to 700 figures in one year, which is quite conceivable for a workforce of about a thousand men. Yet in the first years the workers certainly produced less than the average, due to their lack of experience. Firing figures of such dimensions was tricky, especially as the thickness of the clay walls varied considerably, and in the kiln the figures shrank in size by about 10 percent. Presumably there were frequent misfirings.

Assembling the Figures

Valuable as they are, the inscriptions do not say what the workers actually did, which can only be inferred by closely examining the figures. The foremen and their many subordinates worked within the framework of a well thought out production system. Each of the figures of the standing warriors weighs between 150 and 200 kilograms, and normally consists of seven major parts: a plinth, the feet, the legs below the garment, the torso, the arms, the hands, and the head. (77) The workers first modeled each part separately and then fitted them together. They luted the arms to the body with wet clay then inserted the prefabricated head into the opening at the neck where they secured it with wet clay, too. Because the figures are assembled from set parts, they have the appearance of mannequins.

When fabricating cylindrical parts, such as torsos and arms, the workers first kneaded clay into thick slabs that they then rolled into tubes. In other cases they built up the circular forms from coils of clay. They also formed the wet clay in molds composed of dried or fired clay Although not employed for all parts of the body, molds were an important means by which to standardize and speed up production. The molds for the plinths may have been made of wood in the shape of a flat open box, a simple device the workers would have known from making tiles. The basic form of a head was put together from two halves formed in hemispherical molds. Usually the seam runs vertically over the head in front of or behind the ears. [There had been found] two halves of a head that broke at the seam. Still visible is the imprint of the hand of the worker who pressed the wet clay into the negative mold. Heads of horses were similarly formed with paired molds.

Once the basic form of the figures was completed, the workers took additional clay to shape details such as those on the shoes and the armor. The greatest care was lavished on the heads. The workers attached or reworked by hand the headgear, hair, ears, eyebrows, eyes, mustaches, and lips. Some parts, like the ears, were formed in molds first. Similarly, the ears and forelocks of the horses were preformed in standardized shapes.

A division of labor certainly existed. The signatures indicate that one foreman and his team were responsible for making an entire figure up to the point when it was handed over for firing. Beginners in the team or laborers of little skill could perform simple jobs like preparing and mixing the clay or forming the plinth. Experienced team members may have concentrated on more delicate tasks such as modeling the faces. The firing was undoubtedly a job for specialists, as was painting the fired figures.

Correct timing must have been vital. The different parts of the body had to be joined when the clay was neither too hard nor too soft. The heavy torsos, for example, could only be hauled onto the legs after they had dried to such a degree as to be sufficiently firm. On the other hand, reworking had to be done while the clay was still somewhat damp. The members of the team had to set up their schedule accordingly Probably they could solve these problems more easily when they worked on several figures simultaneously One explanation for the frequency of serial numbers that are multiples of five could be that five figures were made at a time.

Thus, there were no long assembly lines with many workers, each performing one small operation. Rather, the team under each foreman formed its own small, self-contained assembly line that saw a figure through all the production stages. Nevertheless, the teams worked according to similar blueprints. All workers assembled their figures from the same small number of basic parts. There were variations, to be sure. The shoes could be fixed to the plinth, for example, and the lower legs could be made from rolled slabs or coiled clay Yet the basic structure remained the same.

Uniformity is also evinced in the measurements. The respective parts of each clay body have similar dimensions, although oversize limbs or heads appear from time to time. The overall height of the standing soldiers, including the plinth, varies between 180 and 195 centimeters, the length of the feet from 25 to 29 centimeters, the circumference of the torsos from 85 to 107 centimeters, the breadth of the head from about 19 to 23 centimeters, and the length of the face, normally, from 19 to 20 centimeters. (78) The soldiers thus are large, but life-size. This is confirmed by the fact that they carry real weapons, and it is also apparent in comparison to the size of the clay horses. The variations in all measurements remain within a consistent range.

Even more important, only a small and quite limited repertoire exists of different types for all parts of the figures. The authors of the archaeological report have, for example, identified three types of plinth, two types of feet, three types of shoes and four types of boots, two types of legs, eight types of torso, and two types of armor, each category having three subtypes. The most elaborate repertoire exists for the heads, for which the report identifies eight different types. (79) The components of the face are also standardized and the number of types is limited. (80)

In spite of this uniformity of the body parts, the army still conveys an overall impression of extraordinary variety, for two reasons: first, although standardized, the parts are joined together in a multitude of combinations. This allows for large numbers of units that differ from one another. Following the authors of the report in distinguishing eight types of heads, and further, identifying various types of eyebrows, lips, mustaches, and so forth, one soon arrives at a huge number of possible combinations. Second, the workers went over the figures and their parts by hand. They had to do this anyway in places where they luted parts of the body with wet clay Again, the possibilities are exploited most fully in the faces. The makers reworked the physiognomical features when attaching small parts such as eyebrows or mustache, and they added further structure through the use of small incisions. In this way they achieved the truly endless variety that was necessary to make the army appear real.

We are not, however, dealing with portraits, as the English text of the report wants us to believe. (81) By definition, a portrait must attempt resemblance of an individual in a comprehensive creative process. A production method that divides figures and faces into standardized parts does not operate within the holistic concept of a personality that cannot be dividedthe "individual."

Yet it was precisely the creative achievement of those who designed and made the terra-cotta army that they developed a system that allowed them to assemble the figures from a limited number of clearly defined parts. These parts of the body are modules according to the definition presented before, as a final look at the hands will confirm.

The workers formed the hands separately and then inserted them into the open sleeve, sometimes affixing them with a plug. [This figure] shows the perforations for such a plug in the cuff and the wrist. Among the approximately fifteen thousand hands of the terra-cotta soldiers, there are only two main types: hands with extended fingers, and hands with bent fingers. The standing warrior shows these two types. (82)

Hands with a flat palm were made in two molds. A break can reveal the original joints. The joint often runs horizontally through the hand and the fingers. (83) Both parts were fitted together while the clay was still soft, and probably bonded by liquid clay or slip. One mold could also be used to form the back of the hand and the fingers and another mold was employed for the inner palm and the lower part of the thumb.

Hands with bent fingers are half open in order for them to hold a weapon or another object. Palm and fingers could first be made separately in molds and then joined together. (84) Or two molds were used, one for the palm with the thumb, and one for the back side of the hand and its bent fingers up to the second joint. The fingertips were then modeled separately and joined to the rest of the hand. (85)

Some thumbs were formed separately in a bivalve mold. In [this figure], the joint is visible on the backside of the thumb at the left. In the example on the right, the joint runs parallel to the thumbnail. Still visible in both cases are the fingerprints of the worker who pressed the soft clay into the negative mold.

The angle at which the finished thumb was affixed to the hand varied. Thus it was possible to make different types of hands for different functions, as the hand of the kneeling archer demonstrates. Moreover, varying with the angle of the arm, the same hand type could be used for different functions. The standing archer also has the familiar hand with stretched fingers. A chariot driver holds the reins in a hand with bent fingers. Although there are only two basic types of hand, all hands look different because they have been worked over before firing. [This figure] allows us to gauge the diversity.

The observations on the hands can be summed up thus: the hands are prefabricated parts; they are composed of several elements; their measurements are standardized; the number of types is very limited; and the same type of hand can assume a different function in a different context. Hence the hands are modules. The same applies to the other parts of the body of the warrior figures.

Only the use of modules made possible the most extraordinary feat of the terra-cotta army: the enormous quantity of diverse figures. Only by devising a module system could their makers rationalize production to such a degree that, with the material and time available to them, they were able to meet the expectations of the emperor-the creation of a magic army that would protect his tomb for eternity.


1. Chûka jinmin kyôwakoku kodai seidôki ten, cat. nos. 127-29, and discussion of the terra-cotta army by Maxwell K. Hearn in Fong, ed., Great Bronze Age, 334-73.
2. Hearn, ibid., 370, fig. 127. See also Ledderose and Schlombs, eds., Jenseits der Grossen Mauer, 282, fig. 229.
3. Thieme et al., Zur Farbfassung der Terrakottaarmee.
4. Sima, Shiji, 6:265. Translation in Yang and Yang, trans., Selections, 186.
5. Sima, Shiji, 6:232. Yang and Yang, trans., Selections, 163. Cf. Yuan, "Qindai de shi, ting taowen," 96 (reprint p. 77). Yuan, "Qin ling bingmayong de zuozhe," 60-61 (reprint pp. 201-202).
6. Sima, Shiji, 6:256. Yang and Yang, trans., Selections, 179. For the legal status of the workers, see Liu Yunhui, "Lishan tukao."
7. For archaeological evidence, see page TK of this book.
8. Sima, Shiji, 6:256; Yang and Yang, trans., Selections, 179.
9. Sima, Shiji, 6:268 f.; Yang and Yang, trans., Selections, 189. Cf. Yuan, "Qin Shihuangling kaogu jiyao," 143 (reprint pp. 19-20).
10. The literature on the necropolis of the First Emperor is considerable and redundant. Yuan and Zhong, Qin yongyanjiu wenji, 677-86, includes a bibliography of articles published between 1962 and 1985. Among the many comprehensive descriptions of the tomb compound, the most detailed is Yuan, Qin Shihuangling bingmayong yanjiu. Strong on all military aspects is Wang Xueli, Qin yong zhuantiyanjiu. In English, see Thorp, "Archaeological Reconstruction"; Li, Xueqin, Eastern Zhou and Qin, 251-62. Schlombs, "Grabanlage und Beigaben," has made full use of studies by Chinese archaeologists. I follow her and the concise account by Yuan, "Qin Shihuangling kaogu jiyao," 133-46.
11. Segalen, Voisins, and Lartigue, Mission archéologique, vol. 1, pl. 1.
12. Sima, Shiji, 6:265. Yang and Yang, trans., Selections, 186.
13. Sima, Shiji, 7:315. Yang and Yang, trans., Selections, 221.
14. Ban, Hanshu, 1:44. Translation by Dubs, History of the Former Han, vol. 1, 90.
15. While the custom is only documented for the Western Han dynasty, it is believed to have originated in the Qin. Croissant, "Funktion and Wanddekor," 95-97. Yuan, Qin Shihuangling bingmayong yanjiu, 56-59.
16. Wu, "From Temple to Tomb," 95 f.; Wu, Monumentality, 115-17.
17. Zhao, "Qin Shihuangling."
18. Qin Shihuangling kaogudui, "Qin Shihuangling xice."
19. Qin ling erhao tong chema, 65.
20. Qin yong kaogudui, "Qin Shihuang lingyuan peizang."
21. For the Supreme Forest, see Schafer, "Hunting Parks."
22. Qin yong kaogudui, "Qin Shihuangling dongce majiukeng."
23. Qin yong kaogudui, "Lintong Shangjiaocun."
24. Sima, Shiji, 87:2553. Translation in Bodde, China's First Unifier, 36.
25. Shihuangling, "Qin Shihuangling xice," 6, no. 5 (reprint p. 569). Yuan, Qindai taowen, 27-37, 105, no. 234. This mass burial of forced laborers is not unique; 522 such tombs of the second century A.D. have been found near Luoyang. Inscriptions on tiles identify the tombs' occupants as criminals and frequently record their penalty and the date of their death. See Zhongguo kexueyuan kaogu yanjiusuo, "Dong Han Luoyangcheng."
26. Petty officers of the fourth rank stood on chariots to the right of the driver. Wang Xueli, Qin yong zhuanti yanjiu, 201, 630 f. For the ranking system, see Loewe, "Orders of Aristocratic Rank," 99; and Li, Xueqin, Eastern Zhou and Qin, 464-66.
27. For the following, see the archaeological excavation report: Qin Shihuangling bingmayong heng, vol. 1, 4-45; and Schlombs in Ledderose and Schlombs, eds., Jenseits der Groffen Mauer, 272-78.
28. Personal communication by professor Yuan Zhongyi, October 1995.
29. Yuan, Qin Shihuangling bingmayongyanjiu, 69.
30. Ibid.
31. Sima, Shiji, 6:256. Yang and Yang, trans., Selections, 179.
32. Yuan, Qin Shihuangling bingmayong yanjiu, 69 f.; and Yuan, Qindai taowen, 43-45.
33. According to the text Wei Liao zi, which may date from the third century B.C. See Weigand, Staat und Militär, 115. Quoted by Chun-mei Tschiersch in Ledderose and Schlombs, eds., Jenseits der Grossen Mauer, 83.
34. Yuan, Qin Shihuangling bingmayong yanjiu, 351.
35. Qin yong kaogudui, "Qindai taoyao yizhi."
36. Yuan, Qin Shihuangling bingmayong yanjiu, 352.
37. According to Brandt, Chinesiche Lackarbeiten, 7, lacquer was a product of species other than Rhus verniciflua, which is often cited as the exclusive source of lacquer. Therefore it is more correct botanically to use the generic term Toxicodendron verniciflua.

38. Yang Hong, Zhongguo gu bingqi luncong, 4-18. Leather armor for men and horses from the tomb of Marquis Yi of Zeng (after 433 B.c.) has come to light. See Hubeisheng bowuguan, ed., Zeng hou Yi mu, vol. 1, 332-52, and vol. 2, pls. 112-20. For Tang-dynasty fragments of scale armor made of lacquered leather, now in the British Museum, see Whitfield, Art of Central Asia, vol. 3: Textiles, Sculpture, and Other Arts, pl. 49.
39. Very small scales represent special armor made of iron. A standing general with crossed hands wears such an iron apron. See Ledderose and Schlombs, eds., Jenseits der Grossen Mauer, 240.
40. Sima, Shiji, 126:3203. Yang and Yang, trans., Selections, 408.
41. Wang Xueli, Qin yong zhuanti yanjiu, 508-21. Detailed report on the pigments by Thieme et al., Zur Farbfassung der Terrakottaarmee, and Lin, "Lack and Lackverwendung."
42. Qin Shihuangling bingmayong keng, vol. 1, 249.
43. Yuan, "Qin zhongyang duzao de bingqi." Wang Xueli, Qin yong zhuanti yanjiu, 374-419. Li, Xueqin, Eastern Zhou and Qin, 234 f.
44. Hulsewé, Remnants, 59.
45. The inscribed blade of the type pi is the first one on the upper left in Fig. 3.16, and the center one in Fig. 3.17. See Qin Shihuangling bingmayong keng, vol. 1, 265, 268.
46. Hounshell, From the American System, has shown that developing technologies for the manufacture of firearms in nineteenth-century America was essential for the emergence of modern mass production. I should like to thank Dr. Edward Eigen for this reference. See also Ledderose, " Qualitatskontrolle."
47. Bodde, "State and Empire of Ch'in," 20.
48. For iron utensils found in the vicinity of the necropolis of the First Emperor, see Wagner, Iron and Steel, 206-207; and Chun-mei Tschiersch in Ledderose and Schlombs, eds., Jenseits der Grossen Mauer, 210-16.
49. Translation by Hulsewé, Remnants. See also Heuser, "Verwaltung and Recht."
50. The Greeks began to employ infantry formations after 700 B.c. As in China, the new type of warfare harbored considerable social and moral implications. See Murray, Early Greece, 124-36. It is my pleasure to thank Dr. Fernande Holscher for this reference.
51. According to Kolb, Infanterie, 64, 142.
52. For a detailed account of this battle, see ibid., 240-52.
53. Lewis, Sanctioned Violence, 60 f.
54. Quoted by Lewis, ibid., 113.
55. Ibid., 97-135.
56. Loewe, "Orders of Aristocratic Rank," 104.
57. Lewis, Sanctioned Violence, 62.
58. Fang, Jinshu, 60:1651. Memorial submitted by Yu Shinan (A.D. 558-638) in Wang Pu, comp., Tang huiyao, 20:393. Quoted by Wechsler, Offerings of Jade and Silk, 266n.15.
59. Illustrated in Shandong sheng, "Qi gucheng," 17.
60. Preliminary reports by Han, "Fengxiang," and Han and Jiao, "Qindu Yongcheng," 120 f. Brief discussion in Yuan, Qin Shihuangling bingmayong yanjiu, 51.
61. Cai, "Suizang mingqi." References to mingqi in classical literature cites von Falkenhausen, "Ahnenkult and Grabkult," 48, nos. 35, 39.
62. Wang Renbo, "General Comments," 39. Rawson, ed., British Museum Book, 138 f.
63. Cf. Cox, Exhibition of Chinese Antiquities. For Chu culture, see Lawton, ed., New Perspectives on Chu Culture.
64. Thorp, "Qin and Han Imperial Tombs."
65. Archaeological report: Hubei sheng bowuguan, ed., Zeng hou Yi mu. Thote, "Double Coffin."
66. Kesner, "Likeness of No One," 126, emphasizes the "fictive reality through various figurative methods" in the necropolis.
67. Sima, Shiji, 5:194. Quoted by von Falkenhausen, "Ahnenkult and Grabkult," 41.
68. Lewis, Sanctioned Violence, 27.
69. Sima, Shiji, 6:265. Translation quoted from Yang and Yang, trans., Selections, 186.
70. For loess as a raw material in ceramic technology, see Freestone, Wood, and Rawson, "Shang Dynasty Casting Moulds," 257 f. See also Rawson, ed., British Museum Book, 214 f.
71. Physiognomical variety as a condition for magic efficacy has been suggested by von Erdberg, "Die Soldaten," 228. For magic portraits in later Chinese art, see Croissant, "Der unsterbliche Leib," 235-68.
72. Some 170 tiles with seals have been found. For lists of the names, see Yuan, "Qin ling pingmayong de zuozhe," 53, 56 (reprint pp. 186, 192); Yuan, Qindai taowen, 19; and Yuan, Qin Shihuangling bingmayong yanjiu, 354, 361.
73. Yuan Zhongyi first discussed the inscriptions in "Qin ling pingmayong de zuozhe." The archaeological report of the excavation, Qin Shihuangling bingmayong keng, vol. 1, 194-207, 433-43, comprehensively treats all inscriptions found in pit no. 1 up to 1984. Yuan's treatment in Qindai taowen, 13-26, is briefer but takes into account the inscriptions from pits nos. 2 and 3 as well. He basically repeats the same numbers in his Qin Shihuangling bingmayong yanjiu, 352-65. The following numbers are based on Yuan's Qindai taowen.
74. Yuan, "Qin ling pingmayong de zuozhe," 57-60.
75. Hulsewé, Remnants, 110 f.
76. For the legal position of the gongshi, see Hulsewé, Remnants, 62. For his responsibilities, see Yang Jianhong, "Cong Yunmeng," 89.
77. Qin yonggeng kaogudui, "Qin Shihuangling pingmayong keng chutu de taoyong." Qin Shihuangling bingmayong keng, vol. 1, 163-83.
78. Detailed lists with exact measurements in Qin Shihuangling bingmayong keng, vol. 1, 349-75.
79. Ibid., 110-14, 127-38, 142-56, 163-70.
80. Von Erdberg, "Die Soldaten," 228, defines four types of mustache. The excavators now count twenty-four types of beard. Personal communication by Professor Yuan Zhongyi, October 1995.
81. Qin Shihuangling bingmayong keng, vol. 1 (English text p. 498). The assertion that the faces of the soldiers are portraits has already been refuted by von Erdberg, "Die Soldaten." Kesner, "Likeness of No One," discusses theoretical aspects of the issue. For a systematic treatment of portraiture, see Seckel, "Rise of Portraiture," and his comprehensive study, Porträt.
82. The report Qin Shihuangling bingmayong heng, vol. 1, 181, identifies four types of hands: (1) a hand with stretched fingers made in two molds; (2) a hand with bent fingers made from molds; (3) a hand that is mostly covered by the sleeve; and (4) a hand with bent fingers in which the fingertips are modelled and attached to a palm made from molds.
83. Qin Shihuangling bingmayong keng, vol. 2, 161, fig.5.
84. Corresponds to type 2 in note 82, above.
85. Corresponds to type 4 in note 82, above.