Li, Xiuzhen. 2020. Bronze weapons of the Qin terracotta warriors: Standardisation, craft specialisation and labour organization (BAR international series, 2992; Archaeology of East Asia, vol. 3). Oxford: BAR Publishing. ISBN: 978 1 4073 1690 1 (paperback); 978 1 4073 5602 0 (e-format). 221 pp., numerous colour illustrations.

Review for Asian perspectives
Donald B. Wagner
29 March 2020

Among the wealth of amazing finds at the tomb of the First Emperor of China in Xi’an, Shaanxi, some 40,000 bronze weapons have been found, including crossbow triggers, arrows, shafted weapons, and swords. This book is a massively detailed study of this material with the specific goal of learning how weapons production was organized and the extent to which products were standardized. Thousands of artefacts were measured very precisely with ‘digital callipers’, and statistical studies of the measurements, together with studies of the inscriptions on many of the artefacts, give important insights into both questions. The same author has also participated in studies of the specific bronze alloys used, but this aspect is barely touched on here (pp. 94–97; Martinón-Torres et al. 2014).

An introductory chapter gives a broad account of the history of the state and empire of Qin and the archaeology of the First Emperor’s tomb complex. Chapters 2 and 3 lay out the theoretical and practical bases for the study of standardization and labour organization from artefacts alone, without the aid of any archaeology of the workshops themselves.

Chapter 4 considers the inscriptions found on many of the artefacts. These can include dates, names of offices, names of persons, and some marks of unclear function. Information from the inscriptions, studied comprehensively, allows the construction of organizational diagrams of the hierarchies in some production units. These fit well with the conclusion of the statistical studies that production was organized ‘in semi-autonomous units – or a cellular production system – rather than as a flow-line production’ (p. 137).

Each of chapters 5–7, on crossbow triggers, arrowheads, and ferrules and long weapons, begins with an extensive review of the literature on the artefact types under consideration. Though this study has the particular aim of investigating aspects of production organization, it incidentally provides much more information on the individual artefacts than we are accustomed to seeing in Chinese archaeological reports.

In Chapter 5, 229 crossbow triggers (out of 262 excavated) are classified, measured, and subjected to statistical analysis. Each is composed of five parts, three moving parts and two bolts. Each part is classified into three or four types; the differences between types appear to be stylistic rather than functional. The parts were cast to exacting dimensional tolerances, generally less than a millimetre, so that they could be assembled without much adjustment. Microscopic scratches indicate that some small amount of filing was necessary to make the fit.

Dealing with 37,348 arrowheads (ch. 6) necessitated developing a robust sampling strategy. After a preliminary pilot study the strategy chosen was to concentrate on those arrows that were found in groups of 90 or more, clearly corresponding to ‘bundles’ of arrowheads in crossbowmen’s quivers. Six arrowheads were randomly selected from each of 262 bundles. This permitted measuring variability within bundles and between bundles.

The arrowheads consist of a tip and a tang. The tang was cast first and the tip was then cast onto it. Traces were found in a few cases of a bamboo shaft into which the tang was inserted. The tips showed an ‘extraordinary degree of standardization’ in their dimensions across the entire sample (p. 94). This suggests immediately that the proper functioning of the crossbow demanded precise standardization, though extensive experimentation would be needed to explore this aspect.

Alloy analysis of the tips shows very little variation within bundles, but a somewhat larger variation between bundles. The tangs provided weight and a firm attachment to the shaft, but standardization was clearly less important here: they show much greater dimensional variation, but again a smaller variation within bundles. All this suggests again cellular production, though the author declares herself less certain of this particular conclusion (p. 97).

The arrowheads provide a clue to the vexed question of iron in the First Emperor’s tomb. Iron tools were found at a stone-dressing site associated with the building of the tomb (Wagner 1993: 206–209), but the only iron artefacts found in the entire tomb complex itself were two arrowheads with bronze tip and iron tang (p. 86) and an iron spearhead (p. 108). Two different explanations are most often suggested for this almost total lack of iron: either that iron was more costly than bronze, or that it was common, but was avoided in this tomb because it is not permanent; it rusts. I have always favoured the second explanation: the First Emperor was clearly obsessed with permanence, for example placing stone inscriptions on mountaintops describing his accomplishments and (according to later legend) sending explorers out to find an elixir of immortality.

Thousands of bronze–iron arrowheads, very similar to these bronze–bronze arrowheads, have been found in Chinese sites from the Warring States period and perhaps even earlier (e.g. Barnard and Satō 1975: 119 and ‘Table of sites and remains’, pp. 164–295, column 3), and I believe it is correct to say that very few arrowheads entirely of bronze, or perhaps even none at all, are known in Chinese archaeology outside of this tomb complex. The bronze–iron arrowheads have the bronze tip cast onto the iron tang, just as the tips of the arrowheads in this study were cast onto bronze tangs. It is difficult to see a practical reason for the division of an arrowhead into two bronze parts (but see pp. 94–95), but it could be explained if we suppose that, within an ongoing, very large-scale, production of bronze–iron arrowheads, bronze tangs were substituted for iron tangs to meet the demands of an obsessed Emperor for his tomb.

Measurement of 126 ferrules, classified in three types, indicates, for each type, close uniformity of outer dimensions but less uniformity of inner dimensions. This is no doubt because the outer mould for the ferrule would have been used repeatedly, while the core would have been made separately for each casting. Foundrymen prefer a ‘crushable’ mould-core, because castings shrink as they solidify. At the other end of the long weapons, spearheads and dagger-axe heads were too few for statistical study, but the studies here of these artefacts do give more information on this type of artefact than we have otherwise had. One entire halberd was found, with ferrule, wooden shaft, spearhead, and dagger-axe head. Its length is 2.87 metres.

Appendices (pp. 151–221) give the details of dimensions, inscriptions, and exact find location of each of the hundreds of artefacts treated in the study.

The conclusions to this study concerning standardization and production organization are very general as well as rather tentative, but they will no doubt have a place in future wider studies of Qin production based on epigraphy, historical sources, and excavations of workshops. The great contribution of the book is its study and extended demonstration of a complex of methods to approach these questions. And its intensive study of the artefacts themselves will be widely useful in Chinese archaeology.


Barnard, Noel and Satō Tamotsu. 1975. Metallurgical remains of ancient China. Tokyo: Nichiōsha.

Martinón-Torres, Marcos, Xiuzhen Janice Li, Andrew Bevan, Yin Xia, Kun Zhao, and Thilo Rehren. 2014. ‘Forty Thousand Arms for a Single Emperor: From Chemical Data to the Labor Organization Behind the Bronze Arrows of the Terracotta Army’. Journal of Archaeological Method and Theory 21.3: 534–562.

Wagner, Donald B. 1993. Iron and steel in ancient China (Handbuch der Orientalistik, vierte Abteilung: China, no. 9). Leiden: Brill.