The Myth of the Xia Dynasty

Author(s): Sarah Allan

Source: The Journal of the Royal Asiatic Society of Great Britain and Ireland , 1984, No.
2 (1984), pp. 242-256
Published by: Cambridge University Press

Stable URL: https://www.jstor.org/stable/25211710

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access
to The Journal of the Royal Asiatic Society of Great Britain and Ireland

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY

By Sarah Allan

Was there a Xia Dynasty? By the mid-nineteen thirties, the works of Henri
Maspero and other scholars in the West and of Gu Jiegang ?% SS ffl'J and his
compatriots in China had clearly established the originally mythological charac
ter of Yu 4), the founder ofthe Xia Dynasty (traditionally ca. 2200-1760 B.C.)
and of the rulers who preceded him in traditional Chinese historiography.1 The
excavations near Anyang of late Shang palaces, tombs and inscribed oracle bones
had also established the authenticity of the Shang Dynasty which followed the
Xia, or at least of the latter part of it. In 1936, Chen Mengjia f& $ % published
an article in which he related the Xia king list to the Shang and argued that the
two periods were the same.2 For the next forty years, the question of the
authenticity of the Xia was left largely in abeyance although most scholars did
continue to assume that the Xia Dynasty, which was hereditary like the Shang,
would some day be authenticated by archaeological excavation.3
Recently, however, Chinese scholars have begun to classify archaeological
finds as 'Xia'. What has brought about this change? The most important exca
vations of the major 'Xia' site, Erlitou in Henan Province, were done before
1964 but at that time they were classified as 'early Shang'.4 No new inscriptions
which could confirm the existence of an earlier dynasty have been found. Indeed,
although Tang :Jb, traditionally the founder of the Shang Dynasty, is revered in
the oracle bone inscriptions as a high ancestor, the genealogy goes back to Wang
Hai .? S , the first mythical ancestor of the Shang, without any indication that
this was the critical juncture at which the dynasty was founded or that other
rulers were defeated. The primary cause of this change appears to be the intro
duction of radiocarbon dating: we now know that finds at Erlitou and other
sites date to within the traditional Xia period.5
In the following paper, I shall reconstruct the role ofthe Xia in early Chinese
mythology. I shall argue that the Shang had a myth of the Xia as a previous
people who were their inverse, a dark and watery people who had been overcome
by the Shang sun-kings. This myth was transformed into a previous dynasty
when the Zhou conquerors of the Shang proposed the theory of a changing
mandate of Heaven. To demonstrate that there was a Shang myth about the
Xia is not to prove that the Xia were a myth. Nevertheless, an historical recon
struction should not be based on materials which are part of a mythological
system.
Myths in the traditional sense are stories of the supernatural set in a time
'long ago'. From the Zhou Dynasty on, the past in China was interpreted in
terms of repeating themes and a dynastic cycle. As I have discussed in The heir
and the sage, 'history' in Chinese texts from the 5th to the 1st centuries B.C. had

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 243

a mythical function in the sense defined by Claude Levi-Strauss, i.e. it served to

mediate inherent social conflict, and the repeating themes obscured the differ
ence between the past and the present, the cycles connecting that time 'long
ago' with the present.6 Within the structure of the legends recorded in these
texts, the Xia Dynasty is a critical turning point. Yu, the founder of the Xia,
was given rule by Shun who also received rule non-hereditarily and without
violence; but the Xia were overthrown by the Shang and all subsequent dyn
asties were founded by overthrow and continued hereditarily.
Yet even in these texts, there are the remnants of an earlier mythology about
a time in which men were not like those of today, when heaven and earth were
not separate, and time was undifferentiated. These take such forms as: 1) stories
of cosmogonic events, 2) supernatural births, 3) the transformation of human
beings into animals or inanimate objects, 4) unreasonably long lives, and 5)
confusion of time sequence in the historical chronology indicating that stories
which took place in the mythic time of long ago have been systematized by
different hands. The mythology of the Xia which I shall reconstruct herein
is not dynastic legend, but this earlier layer of 'true' mythology.
In 'Sons of suns: myth and totemism in early China', I traced a mythological
system based on the Fu Sang or 'Mulberry Tree' to the Shang and demonstrated
that the ritual calendar of the Shang oracle bone inscriptions is based on a
totemic relationship between the Shang ancestors and the ten sun-birds which
rose from that tree on the ten days of the Shang week. I identified Jun ^(some
times transcribed as Nao$l) in the oracle bone inscriptions and the ten-sun myth
with Di Ku ^ % in the literary tradition (in which the Shang were born of the
egg of a black bird) and with Shun in the historical tradition. I also argued that
a Shang origin myth in which Shang Di -^ '^ , the Lord on High appointed the
progenitor of the Shang was transformed into the Zhou legend of Yao's abdi
cation to Shun.7
Yet, according to the Zhou legend, Shun did not continue to rule but abdi
cated in favour of Yu who founded the Xia. Only when the last Xia ruler Jie -&
was defeated by Tang was the Shang Dynasty established. Why was the Shang
rule not continued if their progenitor was appointed by the High Lord? The Xia
are ubiquitous in Zhou texts, including those chapters from the Shang shu which
may be dated to the early Western Zhou.8 In these the new Zhou kings sought
to legitimize their overthrow of the Shang by citing the precedent of a Xia
Dynasty overthrown by the Shang and arguing that the mandate of Heaven is
not constant. For this propaganda which was addressed to the Yin (i.e. Shang)
people to have been effective, the Shang must have believed that a Xia people
had preceded them and lost the favour of the High Lord. The problem is to
determine the nature and meaning of this belief.
In the following, I shall demonstrate that the mythical themes associated with
the Xia are an inversion of those associated with the Shang in Shang mythology.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 244 THE MYTH OF THE XIA DYNASTY

The relationship is similar to that of later yin-yang dualism of which it was the
harbinger. The Shang were identified with the ten suns, birds, the Mulberry
Tree, the East, the sky and life. The Xia, on the other hand, were identified
with water creatures such as dragons and turtles, the Ruo Tree, the West, the
Yellow Springs and death. This dualism was part of a single mythological system
which preceded and laid the foundation for the Zhou theory of a dynastic cycle
and the changing mandate of Heaven.9
The genealogy of the Xia rulers can be divided into three eras: The earliest,
a clearly mythological period, is from the 'Yellow Lord' Huang Di to Yii the
flood hero's son Qi. Next there is the period from Tai Kang to Shao Kang in
which rule was "usurped" by the descendants of Archer Yi. Finally, there is the
period from Yu to the last king Jie when the Xia are supposed to have reigned
over "all-under-Heaven". This paper will be primarily concerned with the first,
mythological period, but I will conclude with some remarks about the two
following periods and the historicity ofthe Xia.

The cosmology
In 'Sons of suns', I traced an association between the ten-sun myth and the
Shang. In the following, I shall argue that opposing elements of the same mytho
logical system are consistently associated with the Xia. Those elements of the
ten-sun myth relevant to the present discussion are summarized as follows: In the
far east, there was a Mulberry Tree (Fu sang $l j& ) on the branches of which
the ten suns perched. These suns were also birds and they rose one by one on the
ten days of the Shang week to fly across the sky and roost on the Ruo Mu % ^ ,
the Western counterpart of the Fu Sang. The Ruo Mu forms a pair with the Fu
Sang but there is no trace of a distinct cult in the textual tradition.
The Valley ofthe Sun, Yang Gu 1% & (i.e. **,*$, or^) lay at the foot of
the Mulberry Tree and in it was a pool of water, sometimes called the Xian Chi
A /<?j in which the sun-birds bathed before their morning flight. There was
also a gorge at the foot of the Ruo Tree, called the Yu Yuan jt #?|. This has
been identified by Chen Bingliang with the Feather Abyss $ '$ where the
sun-ravens shed their feathers when they were shot by the Archer Yi and where
Gun was transformed into a yellow dragon or three-legged turtle, as I shall dis
cuss below.10 The water in this gorge is sometimes called the Meng ^.('Hidden')
Si>?i ? si means a stream which returns to its source.
The motifs of the Fu Sang and Ruo Mu are sometimes transformed into the
Kong Sang ? & , or 'Hollow Mulberry', and the Kong Tong ? ?\, 'Hollow
Paulownia'. The Kong Sang which is an axis mundi and is identified with rule
in the mythology of the Xia discussed below was the birthplace of Yi Yin,
traditionally the minister of the Shang founder Tang. The Kong Tong was a
place of death and disgrace, the death place of Tang and the place to which
Yi Yin banished his son Tai Jia.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 245

The 'Yellow Lord' Huang Di and the Yellow Springs

In the Xia annals of the Shi ji, the Xia ancestry is traced from Yii back to
Huang Di, the 'Yellow Lord'.11 The annals of the five emperors, the first chapter
of the Shi ji, also begins with Huang Di. Thus, by the end of the 2nd century
B.C., Huang Di was regarded as the first ruler of China. This tradition, as Yang
Kuan observed in the Gu shi bian, did not begin until the late Warring States
period.12 But even in the texts of the late Warring States and early Han, history
conventionally begins with the time of Yao.13 Yao, as I noted above, was a
transformation of Shang Di, the High Lord of Shang religion, and he is called
simply Di in the Shang shu Yao dian.
Huang Di was credited with the ancestry of many ancient tribes, but he is
particularly associated with the Xia, so much so that Chen Mengjia attempted to
identify him with Yu.14 His surname, for example, was You Nai Shi7!" Ifc, &
or, alternatively, Han Yuan Shi M~ -?&- &. As I shall discuss below, both
Sun and Yu turned into yellow nai, a three-legged turtle or dragon. Han Yuan
is also the Heavenly or Black Turtle (tian ? or xuan i? vL ) and in some
versions of the myth the name of the mountain through which Yu passed
when he turned into a nai.15
Yang Kuan, on the other hand, argued that Huang Di was Shang Di.16 This
is based on a series: Shang Di, Huang Shang Di Jl X $*, Huang Di ^ T,
Huang Di -^ ^ - in which Huang Di, the 'Yellow Lord' is supposed to have
originated with a taboo for the character huang, 'august' referring to the High
Lord. Huang, 'yellow', and huang, 'august' were homophones in classical Chinese
(*g'wang),17 but their meaning is quite distinct and it is unlikely that the cult
of the Yellow Emperor, which was very popular in the late Warring States and
Han times, could have derived from a taboo character. Huang, 'august' (origi
nally a shining sun over earth) refers to the sky (as in da huang & %-) and it was
used as an adjective to describe Shang Di as early as the Shijingi^f ,*1.18 (Only
after the 'First Emperor' of Qin styled himself Shi Huangdi, did huangdi come
to refer to an earthly ruler rather than the August Lord.) Huang, 'yellow', on
the other hand, is the colour of the earth as well as the springs which ran under
the earth, of dusk, and of the centre in five-element theory.19
Another possibility is that the Yellow Lord was originally the Lord of the
underworld, the counterpart of Shang Di, the Lord on High. The netherworld
in ancient China was called the Yellow Springs (Huang Quan-? &). This watery
region beneath the earth is depicted in Han tomb art with turtles, dragons, or
large fishlike creatures, water animals such as those implied by the Yellow Lord's
surname.20 The earliest explicit reference to the Yellow Springs as the land of
the dead which I have found is a passage in the Zuo zhuan Js~ fy in which
Duke Zhuang of Qing who had been wronged by his mother swore to his mother
that: "We shall not meet one another until we reach the Yellow Springs". Later
on he regretted his oath, so he accepted advice that "if you scoop out the earth

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 246 THE MYTH OF THE XIA DYNASTY

until you reach springs (quan) and then, having dug a tunnel, meet one another,
who could say that it was contrary [to your oath] ."21
This device suggests that all underground springs were branches of the Yellow
Springs. The Mencius, Xunzi, and Huainanzi also record the belief that worms
"eat soil and drink from the Yellow Springs".22 Similarly, Wang Chong observed
in the Lun heng that people do not like to work in mines because they are
"next to the Yellow Springs".23 The Yellow Springs, then, ran under the earth
just as the sky surmounted it. This dualism is sometimes made explicit. The
Zhuangzi, for example, speaks of "treading the Yellow Springs and climbing
to the great sky (da huang).2* The great flood, as I shall discuss below, was a
problem of controlling these waters which were rising up and threatening heaven.
In five element theory, Huang Di was associated with the earth, yellow,
dragons, and the centre. According to the Lushi chanqiu which was written in
the late 3rd century B.C. and includes one of the earliest formulations of five
element theory, "In the time of the Yellow Lord, Heaven first caused large
earthworms and mole-crickets to appear. The Yellow Lord said, "The spirit of
earth is in ascendancy'. The spirit of earth was in ascendancy, therefore he
esteemed yellow as his colour and took soil as his concern" (13/4a). Earth
worms, as mentioned above, both eat soil and drink from the Yellow Springs.
Mole-crickets bore in the earth and sing sadly in the evening.25 The Huainanzi
(3/3a), about a century later, stated, "As for the central land, its lord is the
Yellow Lord,... his animal, the yellow dragon." Elsewhere in the Huainanzi
(4/11 a), we are told that the Yellow Dragon born ofthe ether of the central
earth (after a number of transformations) hides in the Yellow Springs. The
yellow earth and the yellow springs which run beneath it are thus connected.
However, Huang Di is not only the Lord of the central region. He is fre
quently identified with a cult of immortality associated with the Kun Lun
Mountains in the far west of China. At the foot of the Kun Lun Mountains,
there was a Ruo (*niok) Ik River identifiable with the % (*riiak) River which
had its source at the Ruo Tree.26 This was the birthplace of Huang Di's des
cendants, as I shall discuss below. It may also have been an entry to the land
of the dead. (Ultimately all waters must have derived from the Yellow Springs).
The Kun Lun mountains were also the home of Xi Wang Mu 3h ^ ^ , the
Queen Mother ofthe West who gave the Archer Yi the elixir of immortality.
In ancient China, there were two souls, the hun ^and thepo *fe ,the ethereal
and the corporeal, the "heavenly" and the "earthly".27 The hun ascended
whereas the po settled. Thus, the Yellow Springs may not have been simply
the land of the dead, but more precisely, the land of the po souls. The origin
of this belief in two souls is difficult to determine because only the hun soul
was the object of the ancestral cult. There was also a reluctance to discuss what
happened to a person after death before the late Warring States period when
the two souls began to be discussed in connection with the developing yin-yang

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 247

theory. The earliest reference which I have found is a discussion of ghosts

without tombs in the Zuo zhuan (Shao Gong 7): "When men are born and
begin to move, it is called po. When the po is already born, the yang (element)
is called hun. As the form and spirit become greater, the hun and po become
stronger.... When ordinary men and women are killed by violence, the hun
and po are still able to attach themselves to people . . .28
In the late Warring States Period, men also began to attempt to preserve
themselves not simply as ancestors, i.e. hun souls receiving offerings, but cor
poreally, by prolonging life and by preservation of the corpse, as at Mawangdui.
This may explain the sudden prominence of the Yellow Lord. In Shang times,
the Lord on High, Shang Di (usually called simply Di) was the lord of the spirit
world. In Zhou times Shang Di was identified with Tianf^, Heaven.29 If a Lord
of the Yellow Springs, i.e. a Lord Below, existed, he had no cult. This only
developed with the Warring States cults of immortality.
The Lord of the netherworld was thus transformed into the Lord of the
western paradise. In five-element theory, however, he became the Lord of the
centre. In this guise, he fought the Red Lord Yan Di k f (i.e. Chi You# ?)
whose emblem was fire for possession of the Hollow Mulberry ? perhaps a new
transformation of the earlier fire and water dualism.30 In the historiography,
there also became five emperors: Huang Di, Zhuan Xu (his descendant, as dis
cussed below), Yao, Shun, and Yu.

Chang Yi to Zhuan Xu
According to the Shi ji, Huang Di had a son by the 'Woman of the Western
Mound', Lei Zu ,*& & called Chang Yi ? *.31 In the Shanhai jing, Lei Zu is
written as 'I? & , "thunder ancestress".32 Chang Yi and his descendants are
identified with the West, water, and death or immortality. In the Xia Annals,
Chang Yi was the father of Zhuan Xu 8jk *&, but in the Shanhai jing, Han Liu
Jtff y$u intervenes between the two generations:

Huang Di's wife Lei Zu gave birth to Chang Yi. Chang Yi des
cended and made his home in the Ruo River. He begot Han Liu.
Han Liu had a long throat and small ears, a human face with a
pig's snout, a scaly body, thighs like wheel rims and pettitoed
feet. He took Zhuozi i$ 4- who was called A Nil Vi & as his
wife; she bore Di Zhuan Xu.33

The Guben zhushu jinian also records that Chang Yi descended and made his
home in the Ruo River although in that confusion of sequence which is charac
teristic of this mythological era, Chang Yi and his son ? called Han Huang
$t fej in this text ? are placed before Huang Di.34 The Ruo River, as I noted
above, had its source at the Ruo Tree, the western counterpart of the Fu Sang.
These texts might have been interpreted as "made his home at the Ruo River,

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 248 THE MYTH OF THE XIA DYNASTY

but since Han Liu's appearance is that of a dragon, I assume he lived in the
river. Di Zhuan Xu is also described as a water creature in the Shanhai jing:
"There is a fish which is withered on one side. Its name is Yu Fu & -k^ ('Fish
Lady'). When Zhuan Xu had died, he thus came to life again. The wind blew
from the North, and then the sky became a great spring of water. A snake was
transformed into a fish. This is what is called Yu Fu: Zhuan Xu who had died
come to life again.35 The term 'withered on one side' (bian guifa ii) is also
used for Zhuan Xu's descendant, the flood hero Yu.36
Earlier in the same section ofthe Shanhai jing, it also states that ".. . there is
a mountain called the Great Wasteland Mountain which is where the sun and
moon set. There are people there who have three faces. These are the sons of
Zhuan Xu. They have three faces and one arm. Three-faced people don't die."37
These three-faced descendants of Zhuan Xu appear to be an inversion of the
three-bodies sons of Xihe, the mother of the ten suns. The Xia are further
identified with the far West ? where the sun and moon set ? and with life
beyond death.
The Lushi chunqiu (5/9a) also states that Di Zhuan Xu was "born from the
Ruo River", and adds, "he actually made his home in the Hollow Mulberry.
Thereupon, he rose up and became Lord (di). When heaven was harmonized,
the principal wind blew . . . . Di Zhuan Xu liked its sound and so he ordered the
Flying Dragons to make sounds which imitated the eight winds and bestowed
the name 'Containing Clouds' on (the music) for making sacrifice to the High
Lord." The Hollow Mulberry in this passage is an axis mundi, allowing access
to the heavens. Zhuan Xu's retinue were dragons, those which fly on clouds ?
clouds rose from springs as water mist and fell again as rain, thus although
dragons were water creatures, they sometimes ascended to the skies.38
As the second of the five lords, however, Zhuan Xu was opposed by Gong
Gong. According to the Huainanzi (3/la?b), "Long ago, Gong Gong contested
with Zhuan Xu to become Di. He became angry and butted Bu Zhou Mountain
(in the northwest corner of the earth), breaking the pillar of Heaven and sever
ing earth's cord. Heaven inclined in the northwest, so the sun and moon, stars
and constellations move in that direction; earth did not fill up the southeast, so
the water and dust turn towards there." Elsewhere in the Huainanzi (15/lb),
we are told that when Zhuan Xu contested with Gong Gong, Gong Gong caused
a "water catastrophe".39
The earliest reference to Gong Gong's rampage is in the Chu ci, Tian wen
(ca. 5th century B.C.), "When Kang Hui was greatly angered, why did the earth
incline in the southeast? How were the nine states divided and why were the
river valleys made deep?"40 The first of these two questions refers to Gong
Gong's butting of Bu Zhou Mountain; the second to Yu's division ofthe central
kingdom into nine states and dredging of the riverbeds to drain the flooding
waters. Thus the tilting of the earth and the flooding are once again linked. The

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 249

Tian wen do not, however, tell us when these events occurred except for refer
ence to Yu's solution.
As I mentioned above, the historical sequence of five lords which is related to
five-element theory is later than that of Yao, Shun and Yu. Most frequently, the
great flood is placed during this period and this tradition is also recorded in the
Huainanzi (8/5b): "During the time of Shun, Gong Gong stirred up and made
swell the flooding waters (zhen tao hong shui ^ >-?? ^ ^) until they pressed the
Hollow Mulberry". According to the Lushi chunqiu (14/3b?4a), the mother of
Yi Yin, traditionally the minister of the Shang Dynasty founder Tang, turned
into a hollow mulberry when she turned around to look back upon the flood
which was emerging from her kitchen mortar. This flood was undoubtedly the
great flood ? other records indicate a drought rather than a flood at the begin
ning of the dynasty41 ? and the tree from which Yi Yin was supernaturally born
was the axis mundi, but because these stories were originally myths, placed in
that time "long ago", there is confusion when they are placed in historical order.
These events are also transformations of the cosmogonic myth in the Shang
shu Yao dian. Therein, Di (i.e. Shang Di)42 having first arranged the heavens
and the calendar, then turned to the lower world where "voluminously the
great waters everywhere are injurious, extensively they embrace the mountains
and rise above the hills, vastly they swell up to Heaven (tao tian)".43 Having
first rejected his son Dan Zhu and then Gong Gong who also "swelled up to
Heaven" (tao tian), he then turned to Zhuan Xu's son Gun to allay the flood
ing, as I shall discuss below. In all of these stories, there is a symmetry of sky
and water, a battle between the rising waters and the Heaven represented by
the High Lord or by the Hollow Mulberry which allows access to the skies.
In the Tian wen, the flooding waters are called hong quan -Jk %~ "flooding
springs" and it seems that those waters which surged to heaven and pressed
the Hollow Mulberry in cosmic battle of sky and water were those which were
later confined to the netherworld from which they watered the riverbeds.44
In the Mencius (5/1 lb, 3A.4), we are told, "In the time of Yao, the world was
not yet level, and the flooding waters flowed laterally (i.e. not in riverbeds),
inundating the world." Although this passage does not mention the tilting of
the earth caused by the attack on Bu Zhou Mountain, the sense is similar. In
no text is the flood ever attributed to rainfall, nor is there any suggestion of
divine punishment. William Boltz has identified the etymonic root of Gong
Gong as 'bellicose' or 'wanton'45 and there is a battle between order and dis
order, and between the high and the low in the story of the flood, but the
biblical concepts of sin, guilt and retribution are completely absent.

Gun
Gun $& whose name may be divided into xuan yu ~? P., 'dark fish' tried,
according to the Chu ci, Tian wen, to allay the flood by following a pattern

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 250 THE MYTH OF THE XIA DYNASTY

made by owls and turtles, symbols of night and water: "the owls and turtles
linked together, tail in mouth; why did Gun follow them."46 According to the
Shanhai jing, "Gun stole Di's swelling mould (xixiangis *H) and thus dammed
up the flooding waters,"47 but his attempts failed and he was executed by
Di.48
Although Gun was executed, he did not die but was transformed into a
yellow nai%, a three-legged turtle or possibly a dragon,49 and thus he gave birth
to Yu. The earliest reference to this story is once again that in the Tian wen,
"Long he lay cast off on Yu Shan. Why did he not rot for three years? Lord
Gun brought forth Yu from his belly. How was he transformed?" and again,
"When Gun came to the end of his westward journey, how did he cross the
heights? He turned into a yellow turtle, how did the shamans bring him back
to life?"50 The answer to the questions may be found in the Gui cang, "When
Gun had been dead for three years and did not putrefy, they cut him open
with a knife of Wu. He was transformed into a yellow turtle (nai). They tore
him open with a knife of Wu ?and thus he gave birth to Yu."51 And in two
almost identical passages in the Zuo zhuan (Zhao Gong 1) and the Guo yu
(Jin yu 8, 14/4b): "Long ago, when Yao executed Gun on Yu $ ('Feather')
Mountain, his spirit was transformed into a yellow turtle (nai) and thus he
entered the Yu ('Feather') Abyss."
This story has many of the motifs discussed above in association with the
Xia ancestors. The Yellow Lord Huang Di had the surname nai. According to
the Shanhai jing, Zhuan Xu's descendants were three-bodied, matching the
three-faced descendants of Shun. Gun was a three-legged turtle rather than
three-bodied, but as such he is a counterpart for the three-legged sun-bird of
the Shang. The Feather #1 (*giwo) Abyss is the & (*ngiwo) Abyss, the gorge
where the sun set at the foot of the Ruo Tree52 ? the land of death and entry
to the Yellow Springs.

Yu and Qi
Yii (*giwo) was born when Gun was transformed and his name is phoneti
cally identical with that of the Feather Abyss. It means chong A, a class of
animals usually translated as 'insects' but which includes both dragons and
tortoises. The story of how Yu dredged the riverbed and built up the high
land so that the water flowed peacefully in channels to the sea is the best
known of early Chinese myths and need not be recounted in detail here. Just
as Gun followed a pattern made by owls and turtles, Yu followed yellow
dragons.53 Yu's wife was the Lady of Tu Shan^ ^. She is sometimes identi
fied as Nu Gua -k Jfo who in still another version of the cosmogonic myth
recorded in the Huainanzi (3/6b) cut off the four legs of a turtle to prop up
the sky when all four poles were broken causing fire and flood. This story
may give a clue to the origin of the three-legged tortoise for if only Bu Zhou

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 251

Mountain were broken as in other versions of the myth, only one leg would
have been needed to prop up the sky.
Yu also had a miraculous transformation resulting in the supernatural
birth of his son Qi &. The earliest reference to this story is probably a line
from the Tian wen which refers to a diligent son who slew his mother.55 A late
account explains that when controlling the flood, Yu passed through Han Yuan
Mountain (Han Yuan was one of Huang Di's surnames) and was transformed
into a nai. The Lady of Tu Mountain saw him and fled in fright. When she
reached Song Gao Shan % % X9 she turned into a stone. Yii said, "Return
my son to me" and the stone broke open, giving birth to Qi.56
Thus far we have been within a mythical era, a time long ago' when the
world was first taking shape before there was a separation between the super
natural and human worlds. Yii gave the world its physical order when he con
trolled the flooding. He also gave it its political order for he harmonized the
nine states and had cast the nine sacrificial ding vessels. Qi whose name
means 'Beginning' was the last of the Xia ancestors to be born miraculously. He
was also the first hereditary ruler in the historiography of ancient China. His
reign was that in which heaven and earth were separated and his role was tran
sitional.
Qi still had access to Heaven from which he took music, the Jiu ge Ju %L,
the Jiu bian A, #f , and, according to some texts, the Jiu shao Ji> 4S . This
tradition goes back to the Tian wen and Li sao of the Chu ci51 and is made
explicit in the Shanhai jing: West of the Floating Sands (i.e. in the extreme
west), there is someone with green snakes in his ears who rides a mount of two
dragons. His name is Xia Hou Qi %_ & &. Qi ascended and played host to
(bin 4% ) Heaven three times. He obtained the Jiu ge and Jiu bian and des
cended . . . ."58 Yii had cast the ding; now, with sacred music, sacrifice could
be made by those below.
The Chu ci includes texts entitled Jiu ge which includes eleven rather than
the nine songs indicated by the title) and Jiu bian. The meaning of the 'nine'
(jiu) in these texts has been much disputed, but the meaning may be understood
in the light of the myth of Yu and Qi. Yu laid out the earth which was square
in nine States. Ideally, these must have been arranged like the fields in the
fabled well-field system-*. Nine, then, symbolically represented all-under
heaven and it was the number of the imperial sacrificial vessels. The nine songs,
etc., were thus the sacred music of the nine states and the man who ruled over
them.
The stories of the era from Huang Di to Qi are clearly myths in the tradi
tional sense of stories of the supernatural set in a time 'long ago'. From the time
of Qi, Heaven and earth were separated and the Xia "Dynasty" begins officially.
But it begins strangely ? with the loss of the state.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 252 THE MYTH OF THE XIA DYNASTY

Tai Kang to Shao Kang: an interregnum

According to the Shi ji, "When the Xia ruler Di Qi died, his son Tai Kang
k H was established (on the throne). Di Tai Kang lost his state and the five
brothers all fled to Luo Nei >&- :A; they made the Song of Five Brothers. When
Tai Kang died, his younger brother Zhong Kang t A was established (on the
throne); this was Di Zhong Kang. In the time of Di Zhong Kang, the Xi ^and
He fa behaved lasciviously, disrupting the seasons and throwing the days into
confusion. Yin JtL went to correct them and made the Yin Zheng Jib &-. When
Zhong Kang died, his son Di Xiang % # was established. When Di Xiang died,
his son Shao Kang was established."59
The Xia Dynasty thus began with the loss of the state. Although the Shi ji
does not record to whom Tai Kang lost rule, other texts record that it was
usurped by the archer Yi If, he who shot down the ten suns, and there is an
echo of this in the Shi ji reference to Xi and He (mythologically Xihe, the
mother of the suns) creating confusion. Yi was cuckolded and then murdered
by Han Zhuo who fed him to his sons who refused to eat him and died. Han
Zhuo continued Yi's rule and passed it on to his son Jiao >A (or ^), but Shao
Kang whose pregnant mother had escaped, returned and killed Jiao, restoring
the dynasty.60
This interregnum from Tai Kang to Shao Kang, as Lu Simian observed in
the Gu shi bian, appears to be an insertion.61 The story forms an integral whole
which can be separated from the following king list. It is distinguished by the
naming style of the kings. This style is that of the Shang rulers: Kang is prob
ably a substitution for the cyclical character Geng m based upon a taboo.62
This naming style is used only by Kong Jia -3L f and, in alternative genealogies,
Yin Jia HL f who, as I shall discuss below, are also distinguished from the rest
of the king list. Logically, a dynasty cannot begin with usurpation. Indeed, if
there were a "dynasty" in this period, it is that ofthe Archer Yi.

The king list: Yii to Jie

According to the Xia Annals of the Shi ji, Di Shao Kang was succeeded by
his son Di Yu -f. The Guo yu also refers to Yii's ability to follow the lead of
Yu, a further indication that this is in some sense a new beginning.63 The Shiji
lists nine kings between Yu and the last Xia king Jie, but the only ruler about
whom other than genealogical information is included is Kong Jia. Kong Jia is
also the only ruler in the Shi ji king list identified in the Shang naming style,
i.e. with one of the ten cyclical characters which also represented the ten suns
and the ten days of the week. Significantly, the story of Kong Jia is mytho
logical and involves dragons, the water creatures with whom, as I have shown
above, the Xia were consistently associated.
The Shi ji Annals tell us that Kong Jia was attracted to magic and the super
natural and was lascivious and disorderly. The virtue of the Xia declined and

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 253

many lords rebelled. Heaven sent down two dragons, a male and female, but
Kong Jia could not feed them as he had no descendant of the Dragon-feeding
clan in his service. He appointed a descendant of Yao who had studied training
dragons from the Dragon-feeding clan, but the female died and he fed it to the
Xia ruler. When Kong Jia requested the dragon, he fled in fright.
The Guben zhushu jinian genealogy is slightly different from that in the
Shi ji and it does not include Kong Jia. It does, however, include a Yin Jia who
is not in the Shi ji genealogy but may be identifiable with Di Jin M. Yin Jia
"dwelt at the Western River. Heaven had an ominous disaster: the ten suns came
out together. In that year, Yin Jia died."64
Both of these stories are about the decline of the Xia. The supernatural
motifs which signify this decline are consistent with the mythological pattern
established above. In the first story, the Xia declined after the Xia ruler ate a
dragon ? a creature of his own kind for his ancestors were such creatures. In the
second, the ruler who lived on the Western River (once again the Xia are assoc
iated with the West and water) is cursed with an omen of ten suns, the symbol
of the Shang kings.

Conclusion
In the above, I have argued that the Xia were regularly associated with a
series of motifs ? the water and water creatures, the Ruo River which was at
the foot of the Ruo Tree where the ten sun-birds set, the West, the netherworld,
the colour yellow and death. These motifs were part of a single mythological
system which can be traced back to the Shang in which the Shang kings repre
sented the opposite values, light as opposed to darkness, the birds which were
suns and rose each morning from the Mulberry Tree in the East, the world
above, the sky, life itself. We also know from the Western Zhou chapters of the
Shang shu in which the founders of the Zhou Dynasty argued to the Shang
people that their ancestors had overthrown a Xia people before them that the
myth of the Xia preceded the Zhou. In these same chapters, however, this
dualism of Xia and Shang, this myth of a dark and watery people from the land
of death overcome by the Shang ancestor Tang, he who emerged from the pool
of water in the Valley of the Sun, has been transformed into a theory of dyn
astic cycles in which the Xia and Shang were two dynasties succeeded by the
Zhou because of the changing mandate of Heaven.
But was there a Xia Dynasty or even a Xia people defeated by the Shang
who provided the background for this myth? An analysis of the myth cannot
answer this question. Nevertheless, it can assist in understanding the materials
upon which any historical reconstruction would have to be based. Those mater
ials which are mythological in the true sense, i.e. the stories until the time of
Qi when the human world was ordered and separated from the heavens, cannot
be considered as historical in any sense. Thus, the identification of the neolithic

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 254 THE MYTH OF THE XIA DYNASTY

site at Deng Feng -gr ?$ in Henan Province with the flood hero Yu's capital
must be treated with suspicion.65 Similarly, since the Xia were symbolically
associated with the West in this mythological dualism, we cannot take an identi
fication with the West as factual. Some scholars have noted that the region of
the Xia tradition was essentially that of the Shang, but if the Xia were a Shang
myth, this is as we should expect.
I have concentrated in this paper on the mythological aspects of the Xia
and dealt only very cursorily with the later king list. There is very little infor
mation about these kings in early texts other than the mythological material
which I have discussed and the story of Jie's evil deeds which will be dealt with
in a later paper. However, there is no reason to doubt that this king list is an
authentic genealogy. The problem remains: whose list was it? Where and when
did they rule, if they did so, and what was the nature of their power?

Postscript: Since completing this article, a brief report of a new find of the
remains of an ancient city at Yanshi Huxianggou in Henan Province has appeared
in Kaogu 1984.4 (p. 384). Huxiang f ftp was identified in the Han shu as the
site of Tang's capital Xi Bo and the authors of this report believe this city to
be the first Shang capital. It is only five or six kilometres from Yanshi Erlitou
and they further call into question the identification of those remains with the
Xia.
NOTES

1 Henri Maspero, 'Legendes mythologiques dans le Chou king', Journal Asiatique

CCIV (1924), 1-100; Gu Jiegang ti t* fl-J, ed. Gu shi bian-t A *f , 7 vols., Peking and
Shanghai, 1926-41; see also Wolfram Eberhard, Lokalkulturen im Alten China I (supple
ment to T'oung Pao, v. 37, Leiden, 1942), II (Monumenta Serica, monograph III, Peking,
1942).
2 Chen Mengjia .& 'f % , 'Shangdaide shenhua yu wushu' $ -K fi ft Ifc $? & ^,
Yanjing Xuebao .? t, f ?1 XX, Dec, 1936, p. 291.
3 See, for example, K.C.Chang, Archaeology in China, 3rd ed., New Haven, Yale
Univ. Press, 1977, p. 216.
4 Xu Xunsheng i? te i., '1959 nian yu xi diaocha "Xia Xu" de chubu baogao' t 9 s <?
f ?fc s t$ & ( 1 *? ) #J ^ ? #1 ? A^c^w 1959, no. 11, pp. 592-600; 'Henan Erlitou
yizhi fajue baogao' :?\ % ^ *f> - ? $ it & # ^ & ? A:<zo#w 1965, no. 5, pp. 215
224;'Henan Yanshi Erlitou zao Shang gongdian yizhi fajue jianbao'>7 .*> lit #F - ? M -f
$ f t it ^ f t p| i tftfo^w 1974, no. 4, 234-48; Henan Yanshi Erlitou yizhi
san ba qu fajue jian bao' >T ft -'& ^f - 1 f if tt ^ a ^ j? 4& ^ #. Kaogu 1975,
no. 5, pp. 302-9.
5 For a sympathetic summary of the new attributions see K. C. Chang, Shang Gviliza
tion, New Haven, Yale Univ. Press, 1980, pp. 335-55.
6 CMC, San Francisco, 1981.
7 Bulletin of the School of Oriental and African Studies XLIV, pt. 2,1981, pp. 290
326.
8 E.g., the Shao Gao, Duo Shi and Duo Fang. See The heir and the sage, p. 4.
9 For discussion of the development of the theory of a changing mandate of Heaven, see
also S.Allan, "Drought, Human Sacrifice and the mandate of Heaven in a lost text from
the Shang shu", B.S.O.A.S., forthcoming.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 THE MYTH OF THE XIA DYNASTY 255

10 (Chan Ping-leung) & ^ t "Zhongguo gudai shenhua xinshi liang ze" f -*5) -? X#
& *r # ^ 40 , Qinghua xuebao # 4 f *-VII, 2,1969, p. 212.
11 Peking, Zhonghua shuju, 1959, juan 2, p. 49.
12 /Wd.,v.7,pt. l,p. 189ff.
13 See The heir and the sage, p. 6.
14 "Shangdaide shenhua yu wushu",op. cit., p. 323.
15 The character % may be read as xiong, 'bear', or nai, a three-legged tortoise or
dragon, the gloss given to the creature into which Gun and Yii turned - see note 38 below.
As Chen Mengjia, ibid., p. 523, observed, Han Yuan is the name of the heavenly tortoise
and in some versions of the Yu myth, the mountain through which he passed before turning
into a nai. Thus, nai should also be the reading for Huang Di's clan name.
16 Gushibian,v.7,ip. 196.
17 All reconstructions given in this form are from Bernhard Karlgren, Grammata Serica
Recensa, Bulletin ofthe Museum of Far Eastern Antiquities XXIX, 1957.
18 Shi jing, 12/4b (song 192, Xiao Yu, Zheng Yue), 16/2b (241, Da Ya, Huang Yi).
19 I have translated xuan, the colour of the bird from whom the Shang descended as
black but, more precisely, it is the colour of mystery, the darkness of the unknown, not a
colour on the spectrum. Similarly huang describes a murky dimness difficult to penetrate,
not simply the colour yellow. In the Yi jing h & (l/6b, hexagram T),xuan huang is used to
describe the blood of a dragon.
20 Changsha Mawangdui yihao Han mu -k if $&*$- & % t> Peking, Wenwu
Press, I, p. 39; II, pi. 72; "Shandong Linyi Jinjueshan jiu hao Han mu chutu baihua"^ &
i% >t & & ^ ^ & # t A ? $ ? Wenwu 1979, no. 11, inside cover; "Changsha
Mawangdui er, san hao Han mu fajuejian bao"-& > }' .% 3- *% ~ ^ $L ;? & Jfc 4& rh $k
Wenwu 1974, no. 7, pi. 5.
21 Yin Gong 1 (Li Zongtong 4h % ^, ed., Chunqiu Zuozhuan jinshu jinyi & 4k jl
4 ^ it $ if.,Taibei, Commercial Press, 1970, p. 4.)
22 Mengzi 6/15a (3B.10); Xunzi l/10a; Huainanzi 3/5a (All references given in this
form refer to the /watt/page number of the Sibu congkan ^ *f H -H editions published in
Shanghai in the 1930*s. This belief may extend to snakes which may explain the prominence
of snakes in both oracle bone characters for curses and disasters and in the bronze vessels
for making ritual offerings to the ancestors of the Shang period.
23 Lun hengjiaoshi ifc #j & Q, Shanghai, Commercial Press, 1964, juan 13, p. 207.
24 Zhuangzi 6/26b (pian 17). See also Huainanzi 19/6a, "if it is not the top of the nine
layered sky, it is the bottom of the Yellow Springs".
25 See the Nineteen Old Songs, no. 16 (Gushiyuan -? # ;#,Taiping shuju, Hong Kong,
1966, p. 91).
26 Shanhai jing jiao zhu J* ? B. & vi, Yuan Ke J. *'T ed., Shanghai, Guji chubanshe,
1980, 446 (Hainei jing). These identifications are also made by Chen Bingliang, op. cit.,
p.211.
27 See, for example, Huainanzi 9/1 b.
28 Jinshu jinyi, p. 1112.
29 See H. G. Creel, The origins of statecraft in China, v. 1, Chicago, Univ. of Chicago
Press, pp. 493-506.
30 Cf. Shanhai jing 17/6a, Huainanzi 15/la-b, Lushi chunqiu 1/3&.
31 juan l,p. 10;juan 2, p. 49.
32 Shanhai jing jiaozhu (Hainei dong jing), p. 443.
33 Ibid.
34 Guben zhushu jinian jijiao dingbu-& & tr $ *l \ # & tj i% Shanghai, Xin
zhishi chubanshe, 1956, pp. 5-6. The identification Han Liu and Han Huang was first made
by Guo Pu, see ibid.
35 Shanhai jing jiaoshu (Dahuang xijing), p. 416.
36 Zhuangzi 9/38b.
37 p. 413.
38 See Huainanzi 4/1 la for the yellow clouds which rise from the Yellow Springs; see
also Zuo zhuan jinzhu jinyi (Shao Gong 17, p. 1192) for Huang Di's emblem as clouds.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms
 256 THE MYTH OF THE XIA DYNASTY

39 See Willam G. Boltz, "Kung Kung and the flood: Reverse euhemerism in the Yao
Tien", T'oung Pao, 67, 3-5 (1981), pp. 141.
40 Chu ci 3/7b (lines 35-6, David Hawkes' translation, Chu Tz'u: Songs of the south,
New York, Beacon, 1962).
41 For discussion of Yi Yin as a mythical figure, see "Sons of suns", 306-7, 313; for
the drought at the beginning of the Shang, see my article, "Drought and human sacrifice
in a lost text from the Shang shu," B.S.O.A.S., forthcoming.
42 For this identification, see "Sons of suns", 322-3.
43 B. Karlgren, The book of documents, B.M.F.E.A. XXII (1950), pp. 2-3,11.
44 Chu c/3/6b-7a (1.31-2).
45 W. Boltz, 150-3. I disagree, however, with Boltz's interpretation of Gong Gong as a
personification of the flood. There are water spirits in ancient Chinese texts (and rivers were
worshipped in the oracle bone inscriptions), but there is no pattern of personifying pheno
mena such as flooding as opposed to natural objects such as rivers and mountains, though
there appears to be in Indian mythology. Although the Yao Dian states that Gong Gong
"swelled up to Heaven" (tao tian), the text may be corrupt - elsewhere he is described as
causing the waters to swell up to Heaven and the Yao Dian appears to have many layers of
different periods. In any case, this one line is insufficient evidence upon which to establish
a mythical system.
46 Chu ci 3/6a (1. 25-6). In a painting found in tomb no. 1 at Mawangdui (seenote 20),
owls and turtles are depicted in the underworld, flanking large fishlike creatures in the centre.
47 Hainei jing, op. cit., juan 13, p. 492; see also Li ji 14/4a for the damming up of the
waters by Gun. The meaning of xi xiang is uncertain. The Huainanzi (5/15a) refers to the
states in Kunlun's eastern range (in the far west of the central region) where the "swelling
mould dams up the flooding waters. Cf. Huainanzi 4/2a which refers to Yu rather than
Gun damming up the waters with the swelling mould.
48 For the story of Gun's punishment by Yao, see the Heir and the Sage,pp. 35-6,6206.
49 This character is glossed in the Er Ya as a three-legged turtle and may also be written
as tt, as in the Guo Yu 14/14b (Jin yu 8). According to some accounts, it was once written
with three dots, but I have not found any extant examples. Some commentators have
alleged that it is an error for long il 'dragon', but the consistent use of the character nai
in various contexts makes this interpretation less likely. The three legs of the turtle recall
those of the sun-bird.
50 Chu ci 3/6a-b (Tian wen, 27-8, 73-4, based on Hawkes' translation, p. 48).
51 Gui cang % &., Changsha, Yu han shan fang jiyi shu, 1884,12b (Qi shi bian).
52 See p. 8 above.
53 Huainanzi 7/7a.
54 See the commentaries of the Shi ji, juan 2, 51, and the Diwang shiji jicun % 5.
& ?-L $ fr, Xu Zongyuan -it t %*, ed., Zhonghua shuju, Peking, 1964, p. 52.
55 Chu ci 3/15a (Tian wen, 66).
56 From a citation of the Sui chaozi ftl f: ~f as cited in the Yi shi, see Shanhai jing
jiaozhu, p. 210.
57 Chu ci 3/15a (Tian wen, 1. 65), 1/1 lb (Li sao, 1.14).
58 Shanhai jing jiaozhu, juan 16 (Dahuang xijing), p. 414. See also juan 1 (Hainei
xijing) and the Gui cang, 9a (Zheng Mu) for this story. For the Jiu shao, see the Guben
zhushu jinian, p. 9 (in which Qi danced the Jiu shao in the ninth year), and Shanhai jing
jiaozhu, j. 16, 414. The ShijiJ. 1, p. 43, describes Yu as the creator of this music.
59 Shijij'. 2,85-6.
60 The earliest references to this story are in the Chu ci, 3/15b-16b, 18b-19a, lines
67-72,85-90.
61 v. 7,282-92.
b2 Yang Junshi ik $ "f , "Kang Keng yu Xia hui" & k -& t i$ Dalu zazhi X ?&
# it 20, no. 3(1960), 83-88.
63 Guo yu, Lu yu, shang, 4/9a.
64 jijiao dingbu, p. 14.
65 See, for example, Zou Heng *$ #j, Xia, Shang, Zhou kaogu lunwenjit fi jS\ %
* & C & , Beijing, Wenwu Press, 1980.

This content downloaded from

195.220.106.16 on Wed, 14 Dec 2022 21:10:03 UTC
All use subject to https://about.jstor.org/terms