Shamanism & the Origin of the Chinese State - Part Three
The arrival of bronze, Erlitou, Erligang and Shang cultures, human sacrifice, divination and kingship
You can read part one and part two of this series on the origin of the Chinese state by clicking on the links below
Over the last two essays we have covered around 7,000 years of Chinese prehistory, starting with some of the earliest agricultural communities and finishing with the extraordinary rise of multiple proto-cities such as Shimao and Taosi. These seven millennia or so saw the economic and technological foundations being established which would lead to the rise of genuine civilisations, including millet and rice agriculture, traditions of ceramics, the domestication of fruits, soybeans, silkworms and animals such as pigs and chickens. Some of these were local innovations, and some came from the outside world. Bronze metallurgy also likely came from the northern steppe, which linked China to the wider Eurasian sphere, but very quickly this technology passed through the cultural crucible of early Chinese societies and became something distinctive. Similarly developments in the religious sphere of life appeared to be a blend of local and outside influences. The preoccupation with ancestors is identifiable very early in Neolithic societies, clear examples being the cemeteries at Shuiquan and Jiahu. Other elements including divination, astrology, magical animals such as dragons, the use of jade and later human sacrifices, these swirl and coalesce around the Peiligang, Yangshao, Longshan and Hongshan cultures, and solidify in important settlements such as Taosi and Shimao.
If K.C Chang is correct, then the first Dynasties such as the Xia and Shang should inherit a ‘shamanic substrate’ from these early societies, one which underpinned political power in the early states. We have arrived at the point where can examine this claim now, by looking at some of the first bronze age cultures in China: Erlitou, Erligang and the rise of the Shang.
From whence bronze?
To open the question of when, where and how bronze metallurgy arrived in the Central Plains of China we need to jump ahead slightly. We’ve already noted that several Longshan era settlements possessed metal objects - bracelets, knives, ‘gear wheel’ ornaments - which match northern Eurasian styles of metalworking. However, by 1600 BC, we see a discontinuity as the only major forms of bronze working are large cast vessels, used for communal feasting.
What happened?
It is entirely possible that between the end of the Neolithic and the beginning of the Bronze Age there was a ‘collapse’, and metallurgical knowledge and traditions which had accumulated were not passed on. One piece of evidence for this is the deurbanisation of major settlements around 2000 BC. Ten Longshan-era urban sites (~300 ha in size) which demonstrate this are:
In the middle Yellow River basin: Taosi, Zhoujiazhuang
In Shandong: Tonglin, Liangchengzhen, Hetou and Yaowangcheng
In the northern marginal zone: Shimao, and Lushanmao to the south
In the Sichuan region: Baodun
In the middle Yangzi River basin: Shijiahe
All of these settlements underwent substantial deurbanisation in the transitional period from the Longshan to Erlitou periods (2000-1900 BC).
Consequently, the lowlands, which had a considerable number of walled megasettlements of around 300 ha, as well as smaller proto-urban centers, ended up with no sites greater than 50 ha and with a fundamental reduction in labor-intensive constructions toward the end of the Longshan period. What made this phenomenon even more distinctive was that neither the previous megacenters nor their related regions in the lowlands recovered their former glory; the lowland deurbanization marked a fundamental divergence from the natural fluctuations of individual centers.
Taking into consideration the difference in the time span of the Longshan period (c. 500 years) and the Erlitou period (c. 400 years), the drop in the number and size of settlements suggests that the lowlands may have lost around 75 percent of the area occupied during the Longshan era and over 85 percent of its settlements. On this basis, it has been argued that a general population decline occurred in the lowlands of the Yellow River basin and adjacent regions at the Longshan–Erlitou transition
-Erlitou: The Making of a Secondary State and a New Sociopolitical Order in Early Bronze Age China (2022) Li Jaang
The rise of Erlitou from this dissolution of major networks of Longshan settlements fundamentally altered the trajectory of central Chinese civilisation, insofar as how bronze was to be incorporated into the body politic. In place of metal personal ornaments and tools/weapons came the cast bronze drinking vessels which intentionally mimicked the high-status Longshan ceramic vessels. This new tradition stood in marked contrast to basically every other metal-using culture on earth at this time.
The introduction of bronze metallurgy from the Steppe and the north-western borders of China was an essential ingredient of this stage. A conjunction of two completely different traditions cemented a fundamental contrast in the material culture of Eastern Eurasia. People along the borderlands shared the bronze production of weapons and body and dress ornaments that formed part of wider Northern Eurasian cultural practices with steppe groups. Conversely, the elites of the Central Plains concentrated on the use of cast bronze vessels for rituals that reinforced not the standing of an individual, but of an individual in a lineage, and represented an essential element in the socio-political network. They also perpetuated the Neolithic custom of using jade for personal ornaments and symbolic weapons. Hence, the societies on the Central Plains took a very different direction from that of the heroic warriors recognised in Western Eurasia.
China’s major Late Neolithic centres and the rise of Erlitou (2019) Zhang et al
The entry of bronze into the Chinese Neolithic sphere clearly came from the northern steppe zone, where cultures such as the Okunev and Seima-Turbino could interact with borderland farmers and agro-pastoralists. Copper and tin alloys were employed to strengthen the bronzes, but the specialisation and refinement of bronze casting at Erlitou and the wider regions would produce lead-alloyed bronze. Again, the Central Plains of China were unique in this development, charting their own course of bronze vessel metallurgy for their own purposes.
Erlitou: A Xia State?
From about 1900 - 1500 BC the Yellow River valley was dominated by the Erlitou complex, an early example of a mostly indigenous state arising without external invasion. Erlitou has become synonymous with the Xia Dynasty for many Chinese archaeologists, working within the tri-framework of Xia, Shang and Zhou mapping onto archaeological sites. The concept of the ‘sky full of stars’ by archaeologist Su Bingqi, referring to the multiple Neolithic origins of the eventual dynastic states, seems to capture something of the multiple cultural tributaries feeding into places like Erlitou, which crystallised and controlled those forces.
The site itself was first occupied during the Yangshao, then the Longshan cultures (3,500 - 2,500 BC). It was abandoned for around 600 years before being fashioned into a 100 hectare proto-urban settlement of around 2-3,000 people. Erlitou lasted roughly 350 years, going through four archaeologically recognisable stages wherein the site was expanded and rammed-earth palatial constructions were built and refashioned. A bronze foundry, a turquoise workshop, jade and ceramic working areas and other craft centres seemed to have dominated the economic activity of the settlement. It was here that bronze metallurgy was fundamentally reimagined and repurposed into a unique cultural phenomenon - the production of jue and ding vessels, rather than tools, ornaments and weapons.
Given the marked difference in wealth and status between the palatial and ordinary areas of the Erlitou site, plus the extraordinary efforts and expenditure in producing cast bronze vessels, it is clear that - like at Taosi - social hierarchies were entrenched and becoming wider with the introduction of bronze. As it grew in power, Erlitou itself grew to around 300 hectares and maybe 25,000 people. The Longshan collapse likely prompted a reorganisation of politics and with it the ‘metropolitanisation’ of the entire region, with Erlitou at the centre.
It is possible that the part of the dramatic population resettlements and the network built around Erlitou was a coercive act. Forced settlements of large numbers of people is well documented during state formation, including under the Assyrians and the Incas. By breaking up geographical ties of ethnicity and kinship and constructing an economic model of centre-and-periphery, the Erlitou elites could have re-created their world for themselves.
Chinese archaeologists often link Erlitou with the Xia Dynasty, the first in Chinese historiography. The government sponsored ‘Xia-Shang-Zhou Chronology Project’ which began in 1996 classified Erlitou as the Xia capital and presented radiocarbon dates which conform with a revised historical dating system. Written evidence of the Xia is slim, coming from Sima Qian’s Shiji, the Book of Documents and the Bamboo Annals. Most records agree that the dynastic founding arose after Yu the Great controlled the great flood. The flood myth is not completely without proof - the Jishi Gorge outburst flood is a well documented event. It occurred around 1920 BC and destroyed multiple settlements, some of which have been excavated, most famously at Lajia where the oldest noodles have been found. However, many scholars are sceptical of the Xia story and see little firm evidence to link Erlitou to the dynasty or even that the dynasty ever actually existed.
Proto and early Shang
Skipping ahead slightly, the traditional accounts of the early Shang describe how they went to war with the Xia, and ultimately won. Archaeologically we have a real muddle, with no consensus. Between 1600 and 1400 BC a new powerful set of polities suddenly appeared in the Yellow River valley. At Zhengzhou, Yanshi and Panlongcheng, rammed earth wall compounds with bronze casting foundries were established, breaking with Erlitou tradition by standardising bronze vessel production and decorating them with curious motifs. This new culture has been called the Erligang. Some have argued this culture is one of many small aggressively territorial groups, others believe the Erligang to represent the earliest Shang people. For example, this description from a 2022 publication:
The Shang Dynasty in the Erligang period gradually entered a prosperous stage. As a political, economic and cultural center, the highest‑ranking social unit represented by Zhengzhou Shang City had the strength to conduct large‑scale conquests in surrounding areas. Therefore, it was especially necessary for the early Shang rulers to open up a strategic route to the north and maintain the link between the central government and the north. Before the Erligang period of the Shang Dynasty, Anyang was the distribution region of the Zhanghe‑type Xiaqiyuan Culture (漳河型下七垣 ).
In the late Erlitou Culture (generally considered to be the late Xia Dynasty period), the elites from the Xiaqiyuan Culture moved south and took control of the eastern part of the Central Plains, seizing control of the Erlitou capital in the Luoyang basin and beginning to build Zhengzhou Shang City. as the capital of a widespread territorial state. During this process, there might have been a small number of people left in the Anyang, or residents of other tribes may have moved in. By the time of the upper Erligang period, with the weakening of the Shang’s power in the Anyang region, this area was likely no longer under the control of the Shang Dynasty.
Regardless, given that the Shang people moved their capital to Anyang there‑ after, the central forces of the early Shang Dynasty launched military campaigns against the Anyang area to clear the way or to seize resources. The sawed‑off skulls in the HSD of Zhengzhou Shang City site were likely those of prisoners of war acquired in the process. Moreover, the Shang Dynasty was able to reconquer this area due to a series of armed conflicts, allowing later Shang people to have unhindered access to Anyang and establishing their capital there for centuries.
-The Truth of Unusual Deaths under Military Expansion: Evidence from the Stable Isotopes of a Human Skull Ditch in the Capital City of the Early Shang Dynasty (2022) Fang Fang et al
Other researchers refer to the Erligang civilisation or culture as a separate phenomenon. Either way it was highly expansive and obsessed with obtaining bronze, salt from Hedong Salt Lake, marine foods and products, proto-porcelain from the Yangzi region and jade.
The next century saw the abandonment of Zhengzhou (with bronze vessels hastily buried and never recovered) and the complete destabilisation of the valley and surrounding regions. New centres of political power were established at Huanbei, Xiaoshuangqiao and finally at Anyang/Yinxu - the political capital of the Late Shang period. By 1300 BC everything had changed, and the Shang Dynasty ruled.
Religious and political power
We can now take a step back and begin to assess this maelstrom of change through the lens of ritual, of religion and of political power. The first and most striking point, which bears repeating, is the totally unique way these early bronze age states used bronze metallurgy. The Erlitou, Erligang and Shang cultures were brutal in their acquisition of raw copper and tin and utterly obsessed with transforming them into communal vessels. Bells, knives and latterly military weapons were produced using bronze, but the devotion these elites poured into making ritual bronzes characterised this period more than anything else.
So what did they do with these bronzes?
The Erlitou period bronzes seem to continue the tradition from the Longshan of storing, offering, pouring and warming alcohol. This was expanded upon by the Erligang who began adding motifs and images to the bronzes, as well as standardising their production and widening the scope of their uses. Finally the Shang made these ritual bronzes famous all across China - from Inner Mongolia to Sichuan, Shandong to Guangxi - these hugely elaborate and richly ornate vessels are still a jewel in the crown of Chinese prehistory. In all three cultural settings the use of the bronzes was to create a ritual ensemble where food and drink could be offered to the ancestors, and high-ranking elites could be buried with them. Alcohol and food would be served in numerous types of bronze vessel during large feasts, at which the spirits of the clans and families would be present to consume with the living. As we saw in the previous descriptions of the Yangshao and Peiligang cultures, being connected to a line of ancestors was one of the pillars of early Chinese civilisation. Where bronze was used to enhance the status of a warrior class elsewhere, here in the Yellow River valley and beyond - bronze was used to enhance the status of an elite who could afford to feast in great opulence with their forebears.
The actual motifs on these vessels is where things become even more interesting. The primary and ubiquitous motif is called the taotie, or ‘two-eyed’ motif, sometimes the ‘animal mask’. Either way, this is a recurring inscription characterised by two staring eyes surrounded by decorative swirls, ribbons and often animal imagery.
The earliest appearance of this motif in the Bronze Age was at Erlitou (see the dragon image above) inlaid into turquoise/jade plaques and associated with bronze bells. It has also been found on ceramics and ‘ritual’ baton-like objects placed into tombs. Contextually it is part of the religious set-up of drinking, ancestral feasts, burials and high-status goods, invoking the ideas of altered states of consciousness and/or of being watched, seeing and visions. Some have described it as a shamanic motif. The transference of the motif to bronze during the Erligang and then Shang cultures suggests that it had a deeper significance that transcended each specific elite group. Erligang and Shang decorations expanded to included crouching human figures, humans inside the mouths of tigers and tiger fangs on the edges of sacrificial bronze axes. The taotie retained its importance as a symbol of power and then violence.
The taotie of Erlitou culture is found in a mortuary context that suggests an association with shamanism; that is, the dead wore bronze inlaid plaques with a two-eyed motif on their chests that was the forerunner of the taotie motif later found on bronze vessels. The dead were also accompanied by bronze wine vessels, bells, and jade instruments. The eyes on the plaques tell us that they were “seers,” people with a special power of vision, who performed ceremonies using wine. Secondary evidence suggests that a form of this motif was also found on perishable artifacts such as lacquer vessels. Thus, when bronze technology developed in the early Shang period, the motif was transferred to bronze vessels. In the early Shang period, the eyes within an undefined face on bronze vessels suggest not only seers that transcend the boundary of this world and that of the dead, but an unknown power that sees but cannot be seen, thus producing a sense of fear or unease in the viewers.
The lack of definition and separation between image and ground serves to increase this sense of the unknowable. The pottery shard of a bifurcated human flanked by snakes found at Zhengzhou suggests that a more complicated iconography existed in other media, such as lacquered wood. By the late Shang, we have a more fertile aesthetic vocabulary and range of techniques, which prevent a single reading of the taotie and other motifs. The taotie face includes horns and ears of animals, including humans, that were hunted and/or used in sacrificial rites. The use of bird and dragon-like creatures in the overall composition further suggests a tiered cosmos, with sky above and water below.
- The Taotie Motif on Early Chinese Ritual Bronzes (2016) Sarah Allan
Human sacrifice was not unknown during the Neolithic, at Taosi and Shimao (see earlier articles), but the nucleation of power at Erlitou and its subsequent expansion in both space and time was accompanied by larger and more elaborate animal and human sacrifices. Two types become increasingly evident - people killed and buried in mass graves, sometimes decapitated and placed under houses, gates or walls - and people killed and placed on secondary ledges surrounding the grave of an important person (rensheng 人牲 or renji 人祭 (human sacrifice) and renxun 人殉 (companions in death or martyrs)). At the Shang capital of Yinxu/Anyang both are plentifully evident: not only are numerous sacrificial victims identifiable within royal graves, such as the ‘priestess’ Fu Hao’s 16 accompanying bodies, but the field around these graves has turned up over 2,000 pits of sacrificed people. Maybe 10-15,000 victims were killed during the last 200 years of Shang rule at Yinxu alone. Skulls were also manipulated and turned into vessels, ornaments and tools, such as those found at Zhengzhou buried under a rammed-earth wall, which may have been slaves or captured prisoners-of-war. It is possible that many of these people belonged to the Qiang culture, identified as pastoralists at war with the Shang and often killed alongside animals and placed into these pits.
This has been inferred from the vast number of oracle bones which have been uncovered from this period. 5,000 years earlier, at Jiahu, turtle shell oracles were discovered during excavations. These bear an uncanny resemblance to the Shang turtle shell and cattle scapula oracle bones, indicating (yet again) a deep continuation of the role of divination from the Neolithic to the Bronze Age. The importance of Shang-era oracle bones cannot be overstated. Many characters inscribed on them are legible today to Chinese speakers, and they provide an unprecedented glimpse into the life of the Late Shang people. Depressingly unknown numbers of these bones were ground up as medicinal powders prior to the 19th century, but what has been retained has been sufficient to reconstruct royal genealogies and the nature of the divinatory process.
Altogether these elements indicate that Erlitou, Erligang and Shang religious and political power were fused together, combining the production of ritual bronzes, jades and special objects with ancestral offerings, royal burials and increasingly large human/animal sacrifices. The use of oracular divination appears to have shifted from a heterogenous, even outsider practice, to the very centre of power during the Shang. The preparation and pre-treatment of the bones was very different during the Late Neolithic and Erlitou periods and then became standardised under the Shang. Some have argued that the Shang were a proto-bureaucratic society, conducting magico-religious divinations in a sober and calm manner - but the evidence for brutal human sacrifice in the context of massive amounts of alcohol would suggest otherwise. Was the Shang king a shaman himself? Probably not, since we have numerous ritualistic roles to fill here, including conducting feasts, reading the oracle-bones and executing captives. The term wu (巫) is ever present in literature and inscriptions from this period, denoting the religious specialist often identified as a shaman. How accurate this is remains a point of huge contention and the debates are not worth recounting here. Suffice to say that something like a ritual specialist involved in rain-making, spirit-journeys and likely the most important -exorcisms and spirit mediumship - existed within the Erlitou-Erligang-Shang cultural sphere.
Evaluating the shaman-king hypothesis
Over three articles we’ve covered an immense span of time and cultural change. From the Neolithic Cishan communities of 10,000 years ago to the end of the Shang Dynasty around 3,000 years ago, we’ve seen the slow drumbeat of ancestral worship, jade and ceramic production, mortuary feasts, human and animal sacrifice, magico-divinatory objects and inscriptions steadily increase to its apogee in the Shang state. The great temptation and practice of Chinese archaeology is to read the later literature back into the excavations and graves, trying to cordon off the first Dynasties, the early Emperors, the great myths and magical beasts which we recognise today. Ultimately I think this is a mistake, and although there is undeniable longevity in the Central Plains cultural developments, we should view these peoples as independent of this weight of history. The actual evidence for shamanism proper is very limited, and in almost every case an alternative ritualistic explanation could be offered. Ancestral cults, sacrificial violence, oracles and divination, glimpses of a cosmological system which united heaven and earth, ritualistic bronze vessels overflowing with alcohol and adorned with watching eyes, fearsome tigers consuming human life, astronomical platforms, decapitated heads placed under fortified gates - as fascinating as these facets are, none are convincingly shamanic in character.
If shamanism is the use of alternative states of consciousness to communicate and visit the spirit world for healing and contact with the dead and the animal worlds, then we have little to go on here. A shamanic-like figure may have existed, but it would be a stretch to call this the substrate of all political power during the Chinese Bronze Age. This period is bewilderingly unique in human history, a strange crucible of political power concerned with wielding the right to feast with the ancestors and utilise the power of bronze, sacrificial blood and divination for its own ends. For K.C. Chang the taotie and swirling animal motifs on Shang vessels was the evidence for a shamanic worldview where beast-human hybrids mediated between the worlds. Personally I think the evidence points towards a cosmology of distinct realms which are not traversed by an ecstatic person or during soul-flights, but are placated and communicated with using a ideological system of divination and worship - a system with deep roots in the Neolithic past.
Bibliography (not in text)
In and Outside the Square: The Sky and the Power of Belief in Ancient China and the World, c. 4500 BC – AD 200. Sino-Platonic Papers (2009) John Didier
The Zoomorphic Imagination in Chinese Art and Culture (2016) Jerome Silbergeld and Eugene Y. Wang
The Earliest China (2022) Hong Xu
The Taotie Reconsidered: Meanings and Functions of the Shang Theriomorphic Imagery (1991) Ladislav Kesner
The Animal in Shang and Chou Bronze Art (1981) K.C. Chang
Ornament, Representation, and Imaginary Animals in Bronze Age China (2006) Robert Bagely
Many thanks for these essays. In line (I think) with your conclusions, I recall that the 'Cambridge History of Ancient China' refers to a marked change in religious practices in the late Shang. This seems to reflected in the following account of from the 'Book of Poetry', representing, we may suppose, Zhou practice:
"For offerings and sacrifice;
We seat the representatives of the dead, and urge them to eat:
Thus seeking to increase our bright happiness [福 fú] [. (P-209.1)
ctext.org/book-of-poetry/chu-ci
It strikes me that there's a distinctly transactional aspect to this work: the living entertain the dead in expectation that the latter will intervene with the Heavenly powers on behalf of the former (their descendants). The happiness in question [福 fú] refers to, 'good fortune, favour, blessing' (Kroll: 120). A splendid view of human relations with the Divine, I think: rather less grovelling and more mutual relationship.
Moreover, the 'representatives of the dead' are personators [尸 shī] 'persons serving as a surrogate for the deceased' (Kroll: 410); the spirits of the dead supposedly occupying the living person, pro tem. These were usually younger members of the family and verse 5 records them withdrawing 'having drunk to the full'. One imagines that this was to permit the adults to get on with some serious drinking. In a comment, James Legge, somewhat disapprovingly, suggests that the Zhou practice seems to have been an excuse for a drinking party.
This distinctly down-to-earth view of ritual and religion seems, to me at least, to point to the later pragmatic, not to say deistic, views of, in particular, Confucius and Xunzi.
Thanks again.
amazing set of essays. thank you!