ChineseWuShamansManuscript

Abstract

The relationship of wu (巫) to shamanism is problematic, with virtually all mentions of historical and contemporary Chinese wu ritualists translated into English as shaman. Ethnological research is presented to illustrate cross-cultural patterns of shamans and other ritualists, providing an etic framework for empirical assessments of resemblances of Chinese ritualists to shamans. This etic framework is further validated with assessments of the relationship of the features with biogenetic bases of ritual, altered states of consciousness, innate intelligences and endogenous healing processes. Key characteristics of the various types of wu and other Chinese ritualists are reviewed and compared with ethnological models of the patterns of ritualists found cross-culturally to illustrate their similarities and contrasts. These comparisons illustrate the resemblances of prehistoric and commoner wu to shamans but additionally illustrate the resemblances of most types of wu to other ritualist types, not shamans. Across Chinese history, wu underwent transformative changes into different types of ritualists, including priests, healers, mediums and sorcerers/witches. A review of contemporary reports on alleged shamans in China also illustrates that only some correspond to the characteristics of shamans found in cross-cultural research and foraging societies. The similarities of most types of wu ritualists to other types of ritualists found cross-culturally illustrate the greater accuracy of translating wu as "ritualist" or "religious ritualist."

Full text

an Open Access Journal by MDPI
Chinese Chinese
WuWu
, Ritualists and Shamans: An Ethnological, Ritualists and Shamans: An Ethnological
AnalysisAnalysis
Michael James Winkelman
Religions
20232023, Volume 14, Issue 7, 852

Religions 2023, 14, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/religions
For the final article of record please see https://www.mdpi.com/2077-1444/14/7/852
Article
Chinese Wu, Ritualists and Shamans: An Ethnological Analysis
Michael James Winkelman
Aliation; michaeljwinkelman@gmail.com
School of Human Evolution and Social Change, Arizona State University
Abstract: The relationship of wu (巫) to shamanism is problematic, with virtually all mentions of
historical and contemporary Chinese wu ritualists translated into English as shaman. Ethnological
research is presented to illustrate cross-cultural paerns of shamans and other ritualists, providing
an etic framework for empirical assessments of resemblances of Chinese ritualists to shamans. This
etic framework is further validated with assessments of the relationship of the features with
biogenetic bases of ritual, altered states of consciousness, innate intelligences and endogenous
healing processes. Key characteristics of the various types of wu and other Chinese ritualists are
reviewed and compared with ethnological models of the paerns of ritualists found cross-culturally
to illustrate their similarities and contrasts. These comparisons illustrate the resemblances of pre-
historic and commoner wu to shamans but additionally illustrate the resemblances of most types of
wu to other ritualist types, not shamans. Across Chinese history, wu underwent transformative
changes into dierent types of ritualists, including priests, healers, mediums and sorcerers/witches.
A review of contemporary reports on alleged shamans in China also illustrates that only some
correspond to the characteristics of shamans found in cross-cultural research and foraging societies.
The similarities of most types of wu ritualists to other types of ritualists found cross-culturally
illustrate the greater accuracy of translating wu as “ritualist” or “religious ritualist.”
Keywords: wu; shaman; ethnological analogy; priests; mediums; healers; witch; China; evolution of
Chinese religion; sociocultural evolution of religion
1. Introduction: What Is the Wu?
The term wu (巫) has been widely applied to ritualists of China’s past and present
(Boileau 2002; Cai 2014; Hopkins 1945; Lin 2009; Michael 2015; Qu 2018; Schafer 1951;
Sukhu 2012; Xing and Murray 2018; Fu 2022). The spectrum of wu ritualists ranges from
presumed archaic practices that persisted as Chinese society transformed from matriarchy
to patriarchy, then to tribal chiefs and ancient kings, and eventually a wide range of
historical and contemporary ritualists, including mediums and ancestor worship priests.
Whatever the original manifestations and meanings of wu were, by the Warring States
period (fth to third centuries BCE), wu was widely applied to very dierent forms of
ritualists and virtually all subsequent forms of Chinese religious activity (Michael 2015;
Williams 2020).
Following Eliade’s (1951/1964) seminal book Shamanism, the term shaman began to
be applied to the translation of wu into English (Michael 2015). Even earlier, Hopkins
(1945) and Schafer (1951) translated wu into English as shaman, but they also used the
terms wizard and witch for wu. The increasing practice of Chinese scholars translating wu
as shaman following Eliade’s (1951/1964) seminal book was without critical assessments
of whether it was appropriate (see von Falkenhausen 1995; Keightley 1998; Boileau 2002;
Williams 2020 for critiques). Boileau addresses the consequences of such problematic uses
of the term shaman that lack clear references to established features of shamans and used
vague denitions that fail to dierentiate shamans from virtually any religious ritualist.
Citation: Winkelman, Michael
James. 2023. Chinese Wu, Ritualists
and Shamans: An Ethnological
Analysis. Religions 14: x.
hps://doi.org/10.3390/xxxxx
Academic Editor(s): Thomas
Michael, Feng Qu
Received: date
Revised: date
Accepted: date
Published: date
Copyright: © 2023 by the authors.
Submied for possible open access
publication under the terms and
conditions of the Creative Commons
Aribution (CC BY) license
(hps://creativecommons.org/license
s/by/4.0/).

Religions 2023, 14, x FOR PEER REVIEW 2 of 34
This widespread practice of translation of wu as shaman is surprising considering
Eliade’s explicit rejection of such equivalence (Eliade 1964, pp. 450–54). Eliade discussed
vestiges of China’s archaic shamanism in male practitioners (wu xi 覡 and physicians wu
yi 巫醫) but rejected the association of shaman with wu, whom he characterized as
mediums, noting their possession states as an aberrant form and reective of shamanic
decadence. Michael (2015) notes this disavowal by Eliade of any equivalence of
shamanism and wu: “Although Eliade discussed many examples of ecstatic ight, vestiges
of China’s archaic shamanism, he did not associate them with the wu . . . By wu, he
specically referred to ‘the exorcists, mediums, and ‘possessed’ persons . . . [who]
represent the aberrant shamanic tradition’ (Eliade 1964, p. 450). Shamanism was a heroic
venture, and possession was a decisive manifestation of its decadence” (Michael 2015, pp.
677–78). To Eliade, the loss of shamanism was evident in the features of possession
characteristic of the wu.
Nonetheless, scholars have ignored Eliade’s perspective and tended to use shaman
in a very loose way to refer to virtually all Chinese ritualists, a practice that undermines
the usefulness of the term and points to the need for a dierent translation of wu. Williams
(2020, p. 154) points out that by the Han dynasty (206 BC–220 AD), virtually all forms of
supernatural practitioners across China were called wu: “wu . . . comprises almost
everybody who has to do with supernatural phenomena.” Mair (1990) also noted that the
practice of translating wu as shaman has problems stemming from a lack of
correspondence with the characteristics of the Siberian shamans, beginning with the
centrality of the ecstatic ight and the shaman’s ritual with the whole community where
the healing rituals emphasized soul recovery. “This is in contrast to the wu who were
closely associated with the courts of various rulers and who were primarily responsible
for divination, astrology, prayer, and healing with medicines” (Mair 1990, p. 35). Allan
notes that many of the diverse ritualists called wu—also translated as diviners, conjurers
and healers—do not even engage in alterations of consciousness, a dening feature of
shamanism. Translations of wu as spirit medium (von Falkenhausen 1995, pp. 279–80),
magician (Mair 1990, p. 35) and diviner (Boileau 2002) illustrate reasonable alternatives to
the blanket use of shaman for all types of wu.
These descriptions illustrate that from the beginning of Chinese religion, there were
several distinctive types of wu ritualists. This wide range of wu ritualists suggests that the
term has a general meaning as [religious] ritualist rather than as shaman in particular and
emphasizes the need to dierentiate among types of wu rather than translating all of them
as shaman. The critical question involves the correspondence of any of the various types
of wu with empirically established cross-cultural features of shamans. Numerous scholars
of wu have lamented the lack of objective criteria for determining what is a shaman and
its features.
This paper presents such objective criteria from ethnological research to illustrate the
cross-cultural features of shamans and their dierences from other types of religious
ritualists. This research provides an objective etic framework for empirical assessments of
resemblances of the various types of Chinese wu with other religious ritualists, including
a cross-cultural foraging shaman. The usefulness of this approach is illustrated below by
showing the correspondence of some types of historical wu and Chinese shamans with the
etic features of shamans, as well as the correspondence of most wu types with other types
of religious ritualists found cross-culturally. These comparisons are used to show that
most ritualists called wu or Chinese shamans do not correspond to cross-cultural features
of shamans, but rather correspond to other types of religious practitioners.
2. Cross-Cultural Methods: A Derived Etic Model of Religious Practitioners
Rather than using theoretical or ideological principles to arbitrarily dene the types
of religious practitioners, Winkelman’s (1986, 1990, 1992, 2010a, 2022) cross-cultural
research project used a grounded method of deriving variables from descriptive
ethnographic data and performing quantitative analysis of this information to identify

Religions 2023, 14, x FOR PEER REVIEW 3 of 34
cross-culturally valid religious practitioner or ritualist types. This study is based on a
subsample of the Standard Cross-Cultural Sample (SCCS) (Murdock and White 2006)
which examined 47 societies worldwide (see Winkelman 1992 for details). This subsample
provided 115 culturally recognized types of religious practitioners, professional
occupations thought to have a special capacity for interacting with supernatural beings or
power. These practitioners were characterized by variables based on ethnographic
descriptions of their practices (for data see footnote1). This descriptive data represented as
variables (nominal, ordinal) was used to generate a matrix of similarities which was
entered into cluster analysis to determine the dierent types of religious ritualists based
on their empirically shared features (see Winkelman 1986, 1990, 1992 for details).
These statistical groupings were used to derive an etic model of types of ritualists
that are represented in the SCCS/CosSci2 variables as Shaman3 (879), Shaman/Healer4
(880), Healer (881), Medium (882), Sorcerer/Witch (883) and Priest (884). These variables
are referred to with initial capital leers in the text to distinguish these etic ritualist types
from common concepts represented in the same words. The common and distinctive
features of these etic religious practitioner types were determined by frequency analyses
of the variables5 used in determining the types of ritualists; these features are reported in
Winkelman (1986, 1990, 1992, 2010a, 2021a, 2022) and presented in Table 1 and others
below.
Table 1. Characteristics of Religious Practitioner Types (adapted from Winkelman 2022).
Ritualist Type
Principal Magico-
Religious Activity
Selection and
Training
Social and Political Power
Professional
Characteristics
Motive and
Context
Shaman
(Forager
Shaman)
Healing and
divination.
Protection from spirits
and malevolent
magic.
Hunting magic.
Cause illness and
death.
Dreams, illness, and
signs of spirit’s
request.
ASC induction,
normally vision quest
by individual
practitioner alone in
wilderness.
High social status.
Charismatic leader,
communal and war leader.
Makes sorcery accusations.
Ambiguous moral status.
Predominantly male,
female secondary.
Part time.
No group—
individual practice
with community.
Status recognized by
clients.
Acts at client
request for client,
local community.
Community-wide
ceremony at night.
Shaman/Healer
(Agricultural
Shaman)
Healing and
divination.
Protection from spirits
and malevolent
magic.
Hunting magic and
agricultural rites.
Minor malevolent
activity.
Vision quest, dreams,
illness and spirit
requests.
Training by group.
Ceremony recognizes
status.
Moderate social status.
Informal political power.
Moderate judiciary
decisions.
Predominantly moral
status.
Predominantly male.
Part-time.
Collective group
practice, ceremonies.
Specialized role.
Acts at client
request.
Performance in
client group.
Healer
Healing and
divination.
Agricultural and
socioeconomic rites.
Propitiation.
Voluntary selection,
large payments to
trainer.
Learn rituals and
techniques.
Ceremony recognizes
status.
High socioeconomic
status.
Judicial, legislative, and
economic power.
Denounce sorcerers.
Life-cycle rituals.
Predominantly moral
status.
Predominantly male,
female rare.
Full-time.
Collective training,
practice and
ceremony.
Highly specialized
role.
Acts at client
request in client
group.
Treatment in client
group.
Participates in
collective rituals
with Priests
Medium
Healing and
divination.
Protection from spirits
and malevolent
magic.
Agricultural rituals.
Propitiation.
Spontaneous
possession by spirit.
Training in
practitioner group.
Ceremony recognizes
status.
Low socioeconomic status.
Informal political power.
May designate who are
sorcerers and witches.
Exclusively moral.
Predominantly
female; male
secondary/rare.
Part-time.
Collective group
practice.
Acts primarily for
clients at client
residence.
Also participates in
public ceremonies.
Priest
Propitiation and
worship.
Protection and
purication.
Social inheritance or
succession.
Political action.
Incidental training
High social and economic
status.
Political, legislative,
judicial, economic, and
Exclusively male.
Full-time.
Hierarchically
organized
Acts to fulll social
functions,
calendrical rites.
Public rituals.

Religions 2023, 14, x FOR PEER REVIEW 4 of 34
Agricultural planting
and harvest rites.
Socioeconomic rites.
and/or by group.
Ceremony recognizes
status.
military power.
Exclusively moral.
practitioner group.
Sorcerer/Witch
Malevolent acts.
Kill friends, enemies,
neighbors, even kin.
Cause illness, death,
and economic
destruction.
Social
labeling/accusation.
Aribution of
biological inheritance.
Innate abilities, self-
taught or learned.
Low social and economic
status.
Exclusively immoral.
May be killed.
Male and female.
Part-time.
Lile or no
professional
organization.
Acts at client’s
request or for
personal reasons
such as envy, anger,
jealousy, greed or
revenge.
Practices in secrecy.
Ritualist Type
Supernatural
Power/Control of
Power
ASC Conditions
ASC Techniques and
Characteristics
Healing Concepts
and Practices
Shaman
(Forager
Shaman)
Animal spirits, spirit
allies.
Spirit power usually
controlled.
ASC in training and
practice.
Soul ight/journey,
death-and rebirth,
animal
transformation.
Isolation, austerities,
fasting, chanting, singing,
drumming and dancing.
Collapse/Unconsciousness.
Soul loss, spirit
aggression, sorcery.
Physical
manipulations,
sucking, blowing
massaging and
extraction.
Plant medicines.
Shaman/Healer
(Agricultural
Shaman)
Animal spirit allies
and impersonal
power (mana).
Spirit control, spells,
charms, exuvial and
imitative.
Power controlled.
ASC in training and
practice.
Shamanic and
mystical ASC.
Some have soul ight,
animal
transformation.
Isolation, austerities,
fasting, chanting, singing,
drumming and dancing.
Collapse and
unconsciousness.
Extraction and
exorcism, countering
spirit aggression.
Physical
manipulations,
massage.
Plant medicines.
Healer
Superior gods and
impersonal power
(mana).
Charms, spells,
rituals, formulas and
sacrice.
Propitiate &
command spirits.
ASC induction
limited.
No apparent ASC.
Social isolation; fasting;
minor austerities; limited
singing, chanting or
drumming.
Exorcism and
prevent illness.
Physical
manipulation of
body, empirical
medicine, imitative
and exuvial
techniques.
Medium
Possessing spirits
dominate.
Propitiation and
sacrices.
Power dominates, out
of control,
unconscious.
ASC in training and
practice.
Possession ASC.
ASC induced through
singing, drumming, and
dancing.
Tremors, convulsions,
seizures, compulsive
behavior, amnesia,
dissociation.
Possession and
exorcisms.
Control of
possessing spirits.
Priest
Power from ancestors,
superior spirits or
gods.
Impersonal power
and ritual knowledge.
Propitiation and
sacrices.
No control over spirit
power.
Generally no ASC
apparent or very
limited.
Occasionally alcohol,
sexual abstinence,
isolation, sleep
deprivation.
Purication and
protection.
Public rituals and
sacrices.
Sorcerer/Witch
Power from spirits
and ritual knowledge.
Contagious, exuvial
and imitative magic,
spells.
Power can be
unconscious, out of
control.
Indirect evidence of
ASC in reported ight
and animal.
transformation.
Night-time activities.
Illness by consuming
victim’s soul, spirit
aggression, magical
darts that enter
victim, unconscious
emotional eects of
envy, anger, etc.
Table 1. Characteristics of Religious Practitioner Types (adapted from Winkelman 2022).

Religions 2023, 14, x FOR PEER REVIEW 5 of 34
2.1. Social Predictors of Ritualist Types
Winkelman’s (2022) analyses6 of subsistence and social variables associated with each
ritualist type illustrate the source of some of the signicant dierences among them and a
model of sociocultural religious evolution. This involves transformation of primordial
foraging ritualists (Shamans) found worldwide in the premodern foraging and
horticultural societies through eects of intensive agriculture, warfare and political
integration. These specic social eects on religious evolution are illustrated in the
following distinctive ecological, productive and social relations associated with each type
of ritualist:
• Shamans (Foraging Shamans), the only ritualists in societies with a principal reliance
on foraging and without intensive agriculture, supra-community political integration
or warfare;
• Shaman/Healers (Agricultural Shamans) found in societies with intensive agriculture
but lacking supra-community political integration, and generally with the presence
of another ritualist, the Priests;
• Priests are found in intensive agricultural societies with supra-community political
integration and are always found in societies with the following types of ritualists:
• Healers, who are generally found in agricultural societies, but signicantly predicted
by supra-community political integration and the practice of warfare for resources;
• Mediums, characteristic of complex societies and signicantly predicted by supra-
community political integration and the presence of warfare for captives; and
• Sorcerers/Witches in societies with intensive agriculture and supra-community
political integration, but lack community integration (see Winkelman 2022 for
analyses).
Since these ecological and social variables were not used to determine the practitioner
types, the relationships provide an independent conrmation of the validity of the ritualist
types presented (see Winkelman 1986, 1992 for conrmatory analyses). These ritualist
types are found in specic congurations related to subsistence and political conditions,
and as illustrated in Figure 1, present a model of sociocultural religious evolution (also
see Winkelman 2022).

Religions 2023, 14, x FOR PEER REVIEW 6 of 34
Figure 1. Ritualist Types, Congurations and Sociocultural Evolution.
2.2. Cross-Cultural Features of Foraging Shamans (SCCS Variable 879)
The ritualists in the group (cluster) labeled as (Foraging) Shamans shared the
following characteristics, constituting an empirically derived and cross-culturally valid
set of features of Shamans:
• Pre-eminent group leader who performs a dramatic night-time communal ritual
involving enactments, drumming, dancing and singing;
• Principal ritual functions of spirit communication for healing, divination, hunting
and sorcery;
• Selection for the role through spirit encounters interpreted as an illness and
experienced in visions and dreams;
• Training with vision quests in the wilderness with fasting, austerities and often
psychoactive plants;
• An initiatory experience of death by animals which killed and dismembered the
initiate, followed by a rebirth and a reconstruction by animals incorporated as a
principal power;
• Ritual preparations of fasting and sexual abstinence;
• Altered states of consciousness (ASC) conceptualized as magical or soul ight (out-
of-body experience) and an experience of personal transformation into an animal, but
notably the absence of possession in the ASCs;
• Healing practices focused on recovery of patient’s lost soul, combating evil spirits
and the extraction of magical darts causing illness;
• Causing illness and death through darts, sorcery and soul theft; and
• Directing hunters and calling animals.
2.3. Biogenetic Bases of Shamanism
Support for using this empirically derived cross-cultural paern to determine what
is a shaman is further supported by the correspondences of these features with
biologically-based functions that produce these common features. These correspondences
have biological bases in the phylogenetic origins and evolution of hominin ritual as a
community integration process; mimetic and other ritual eects producing the

Religions 2023, 14, x FOR PEER REVIEW 7 of 34
physiological dynamics of ASC and healing; and central features of spirits and animal
powers as personal and social identity, features directly related to innate operators
(modules) of the evolved psychology of hominin adaptation (Winkelman 2009, 2010a,
2010b, 2021b). These biological bases producing worldwide uniformities in foraging
ritualists (Shamans) derive from the following biogenetic structures:
• A collective night-time/overnight conspicuous display with community drumming,
dancing and singing which have deep evolutionary antecedents illustrated in the
homologous sociality-enhancing maximal displays of chimpanzees (Winkelman
2009, 2021b);
• Selection for the shamanic role based on spontaneous visions, dreams and sickness
involving natural tendencies for ASC that enhance access to and integration of
unconscious processes (Winkelman 2010a, 2011, 2021c);
• Training in the alteration of consciousness (i.e., ASC) induced by practices of isolation
in the wilderness, fasting, abstinence and austerities that stimulate the
neuromodulatory neurotransmier systems (Winkelman 2017);
• ASC induced by engaging the mimetic operator (dancing, singing, drumming) which
produce an activation of the endogenous opioid system (Winkelman 2017, 2021a);
• Ritual activities leading to exhaustion and collapse, producing experiences of
communication with spirits and out-of-body experiences reecting innate modules
(Winkelman 2015, 2021c);
• Initiatory experiences of death/dismemberment from aacks by animals and a rebirth
that produces experiences of personal transformation and of incorporating animal
powers into identity and basic structures of self-consciousness (Winkelman 2010a);
• Spiritual experiences produced by stimulation, integration and dissociation of innate
modular cognitive structures operators (Winkelman 2021d); and
• Healing by recovery of lost soul, extraction of objects and removal of sorcery by ritual
elicitation of endogenous healing mechanisms (Winkelman 2010a)
The congruences of these shamanic features with features and functions of humans’
evolved psychology illustrate they are not arbitrary cultural features but have biogenetic
bases; consequently, they show these variables are the most objective criteria for
determining an etic transcultural characterization of shamans in comparative perspective.
These provide ve major biogenetic aspects for the bases of shamans:
• Mimetic community ritual: mimetic enactments in dramatic performances, collective
drumming, dancing and singing;
• Traumatic selection: a psychological dynamic manifested in spontaneous visions,
special dreams and psychosis-like sickness;
• ASC: out-of-body (soul ight) experiences and initiatory experiences of death and
dismemberment by animals followed by rebirth;
• Animal powers: Central roles of animals in formation of personal powers and
experiences of personal transformation into animals;
• Healing: involving recovery of soul loss or theft, extraction of sorcery objects and
removal of eects of aacking spirits
These biogenetic bases indicate the presence of a shaman is a justied assumption for
ancient foraging and simple agricultural societies worldwide. But agricultural
intensication produces global changes in the dynamics of societies and religion, with a
new form of ritualists in the role of the Priest who predominates with the emergence of
supra-community political integration. Nonetheless, the core features of Foraging
Shamans are initially maintained in Agricultural Shamans such as healing rituals with
drumming and singing; selection by spirits in visions, dreams and illness; training and
practices with ASC and experiences of soul ight, animal transformation and death and
rebirth; powers derived from animals; and healing practices related to soul loss and spirit
aggression.

Religions 2023, 14, x FOR PEER REVIEW 8 of 34
But intensive agriculture subsistence produces new paerns of ritual behavior
manifested in Priests who now lead community-wide rituals, relegating the Agricultural
Shamans to private rituals with clients and their families. Agricultural Shamans also
typically have professional organizations that provide training and recognition, features
reecting the social complexity and population concentrations enabled by large sedentary
communities. Greater complexity of society also supported a role specialization among
Agricultural Shamans (i.e., just diagnosis, treating specic kinds of illness or agricultural
rites) that would reduce intra-group competition among the practitioners.
Eventually, other ritualists—Healers, Mediums and Sorcerers/Witches—also emerge
in societies with intensive agriculture and hierarchical political integration and are
associated with war for resources and war for slaves and captives. These represent
dynamics that may have led to the transformation and eventual demise of an ancient
Chinese Foraging Shaman and the emergence of new types of ritualists illustrated in the
distinctive features of various historical wu.
3. Comparative Analyses: Dierent Types of Wu in Cross-Cultural Perspective
From the earliest recorded periods, literary sources identify distinctly dierent types
of wu rather than a homogenous prole for all practitioners called wu. Tong (2002) reviews
the evidence for both pre-historic (Neolithic) and historic wu and references
anthropological theories of religious evolution describing three subsequent types of
ancient Chinese wu ritualists: magicians (including shamans); priests; and sorcerers and
witches. Literary sources from the Shang period reveal a further early dierentiation
between female religious practitioners called wu and male ones called xi (覡wu xi and 巫
醫wu yi) (see Cai 2014; Lin 2009). This followed a division of labor between men’s
involvement in government, exemplied in the King’s (wáng 王) role; and women’s role
in religion as diviners on the King’s behalf, a function that was partially usurped by the
male ritualists who controlled them. Lin further distinguished wu yi from other ancient
wu called “commoner shamans” (民巫 Min wu (Lin 2016)), a distinction paralleled in
Michael’s (2015) contrast of two dierent forms of early wu, noting “two separate
traditions of early Chinese shamanism that I later call bureaucratic shamanism and
independent shamanism . . . [the laer a] tradition of folk shamanism that only
tangentially relates the wu to ocial positions of rulership and bureaucratic institution”
[pp. 652–53]). While independent shamanism resisted or even rejected central authority
and operated independently of the priests and rulers, the bureaucratic shaman was
subordinated to centralized authority and “all other functionaries of the bureaucratic
structures of state religion, including priests, temple ocers, sacricers, diviners, and
scribes” (p. 671). But eventually, wu became depreciated and even criminalized, as their
practices came to be characterized as sorcery and practitioners were subjected to
execution, exemplied in the wū gǔ (巫蛊) described by Lin (2009) and Cai (2014).
These overlapping distinctions illustrate a broad consensus that from the beginning
of Chinese religion, there were several distinctive types of wu ritualists. My analysis
focuses on these distinctive types of wu ritualists across the eras of ancient China, oriented
by information from Lin (2009); Tong (2002); Cai (2014) and Sukhu (2012) who provide
similar distinctions about these types of early wu:
(1). Pre-historic Neolithic wu revealed in archaeological (Tong 2002) and linguistic
(Hopkins 1945) evidence;
(2). Ancient wu (focused on the men wy xi, rather than women wu) from the Shang and
Zhou periods;
(3). Commoner wu (Lin 2009), called community wu by Tong (2002), is a professional class
recorded in the late Zhou period, particularly in the south, and strongly contrasting
with the bureaucratic practices of the state wu; and

Religions 2023, 14, x FOR PEER REVIEW 9 of 34
(4). State religious ocials, the ocial wu (siwu (司巫), nanwu (男巫), nuwu (女巫) and
others of the Zhou dynasty (1046–256 BC) described by Lin (2009); Michael (2015)
and Tong (2002), who labels them as priests.
There are also other forms of wu described for the Han and Qin periods:
(5). Female wu (巫尪 Wu Wang [Du and Kong (2000)] and 巫嫗 Wu Yu [Takigawa (2015)])
of the Warring States period (from Cai (2014) and Sukhu (2012)); and
(6). Wū gǔ (巫蛊) (Lin 2009; Cai 2014) of the Han period.
Two contemporary examples of ritualists alleged to be Chinese shamans are also
reviewed to illustrate the continued use of the shaman concepts for Chinese ritualists and
their variable correspondences with the etic model:
(7). the Chinese Reindeer-Evenki, whose indigenous ritualists, the šaman, are described
by Heyne (1999); and
(8). the bo of the Tu ethnic group of Qinghai Province in Northwest China, who is called
a shaman by Xing and Murray (2018).
The characters representing prehistorical wu (Figure 2), healing wu (Table 2 3) and
ocial wu (Table 4) illustrate a range of representations. The features of these dierent
types of wu are presented in Tables 3 and 5 and in the following material, where they are
compared with the etic ritualist types identied by Winkelman (1986, 1990, 1992, 2010a;
also see Table 1). In the Tables *##, the initial leer of the etic ritualist types (i.e., S, AS, H,
M, P, S/W) is assigned to those characteristics of Chinese ritualists who exhibit
characteristics unique to an etic type (i.e., S for Shamans’ animal powers, soul journey; M
for Mediums’ possession; and H, P for Healers’ and Priests’ formal political and judicial
powers).

Religions 2023, 14, x FOR PEER REVIEW 10 of 34
3.1. Pre-Historical Wu
There is linguistic, archaeological and mythic evidence of the presence of an ancient
Chinese wu (Tong 2002) that has central characteristics of Shamans. These ancient wu
practices, referenced in representations of wu characters in Shang and Early Chu dynasty
records, are interpreted by Hopkins (1945) as depicting “the unmistakable shape of the
dancing Shaman” (p. 3; see Figure 2). Hopkins (p. 4) asserts the religious centrality of these
actions of a shaman dancer represented in the ku wén character ( ), concluding that ku wén
and wu represent the same sound and word. Furthermore, they resemble representations
depicted in “early Bronzes and the inscribed Bones of the Honan Find”. Hopkins suggests
these Lesser Seal sources present a recognizable portrait of a dancing or posturing gure,
a direct mimetic representation using straight and curving lines to symbolically represent
an unmistakable dancing gure (see Figure 2). These dancing activities have a direct
anity to the core biogenetic mimetic bases of the practices of shamans.
Figure 2. Characters Representing Prehistoric Wu
In Shang and Early Chu dynasty records, the wu character with meanings of “to
posture” and “to dance” is found in both the Kangxi Dictionary and the Shuo Wen of Hsu
Shen. Hopkins identied the earlier form of representation for the wu in the Shou Wen
which he says denes the meaning of wu (wu chu yeh 巫 祝也) as “Invoker” or “Imprecator”
(p. 3). Shamans’ activities are central to the meanings assigned to this character:
“invoker”—with meanings of cause, conjure and incantation—as well as those of
“imprecator”—meaning to evoke evil and curse, as well as to pray and ask for and entreat.
These vocal activities reect the shamans’ chanting and singing challenges to the spirits
and seeking their removal—or sending them to enemies.
Hopkins proposes that the lower component was not originally kung (early form ;
Lesser Seal form ), referring to two hands, but rather the character ch’uan ( ), referring
to two feet (p. 4). Either interpretation conrms the mimetic base in the clapping or
dancing gure of the shaman, activities that can induce ASC and are a core feature of
shamanism. Hopkins reviews evidence showing that the meanings of the character wu are
in agreement with the notion of dancing represented in ts’ung wu chih i, ( )
referring to the ritual dancing of the typical yü ( ) ceremony customary in the Yin
Dynasty.
Early Chou texts represent wu chung ( or wu tung ) graphically as
meaning “to dance” but functionally, its representation as wu has the meaning of “the
negative verb”. Hopkins proposes that a disentanglement of the three modern characters
that are pronounced as wu can be achieved by reference to their primitive contours that
reveal their pristine meanings. Hopkins proposes the three modern characters
represented in English as wu have interrelated meanings of “shaman”, “the negative verb”

Religions 2023, 14, x FOR PEER REVIEW 11 of 34
and “to posture,” and that “all be traced back to one primitive gure of a man displaying
by the gestures of his arms and legs the thaumaturgic powers of his inspired personality”
(p. 4).
Tong (2002) integrates myth, legend and history with archaeological ndings to
illuminate the nature of these ancient Chinese ritualists. Tong calls aention to the
shamanic signicance of drums excavated from a cemetery in Shanxi Province in tombs
associated with the Longshan (Lung-shan) period (3000–1900 BC). These poery drums,
several covered with alligator skin and found in association with musical stones and other
high-status items, illustrate the ancient association of the wu with classic Shaman features
of music, dancing and drumming. Tong describes a nd of Neolithic poery bowls that
represent a group of dancing gures with depictions of tails which he suggests represents
the intimation of animals in a ceremonial dance, another core feature of the Shaman’s
relationships to animals. A compilation of documents, “The Tongdian,” provided similar
evidence of this association of pre-historic wu with animals illustrated by groups of
dancers wearing wooden masks depicting animals (dog and pig). Tong suggests that wu
activities were represented in many of “the drums, chime stones, whistles, and utes
discovered at various Neolithic sites in China” (p. 51). Tong (2002, p. 52) notably
emphasizes a feature core to shamanism: “In all these ceremonies, the drum was the most
signicant instrument.”
These analyses of linguistic and archaeological data provide evidence of ancient
Chinese traditions representative of a Foraging Shaman, practices which existed before
the dierentiation of the functions of the tribal wu, before wrien records and the
development of “shaman-king” and a state wu. The principal biogenetic aspects of
Shamans outlined above are present in the limited evidence about their central societal
role in collective rituals; the mimetic complex of dancing, drumming and singing; and the
signicance of animals, including animal transformation. The available data does not
provide a full prole of the etic Shaman but provides conrmatory evidence of the
presence of features of the Shamans found in the ethnological research, conrming that
wu shamans existed in Chinese pre-historic antiquity.
Chinese Character
Pinyin
English
巫
Wu
巫 祝也
Wu zhu ye
Invoker, imprecator (Hopkins 1945); Sorcerer (Shuowen Jiezi 2014)
覡
Wu Xi
Male Shamans (Cai 2014); Male Sorcerer (Xu 2002)
巫醫
Wu Yi
Physicians who treat medical and surgical conditions (Lin 2009)
民巫
Min Wu
Commoner Wu (Lin 2009, 2016)
官巫
Guan Wu
Generic term for officially appointed Wu (Lin 2016)
Female Wu
巫尪
Wu Wang
Female witch (Du and Kong 2000)
巫嫗
Wu Yu
Female witch/sorcerer (Takigawa 2015)
巫蛊
Wu Gu
Witchcraft activities (Lin 2009; Cai 2014)
Table 2.. Characters Representing Types of Healing Wu.
3.2. Ancient Wu Xi
A variety of forms of healing wu are reported (see Table 2, Table 3 for characteristics).
Literary sources from the Shang period reveal an early distinction of male religious
practitioners called wu xi (Cai 2014; Lin 2009). Lin characterizes the male wu xi of the pre-
Qin and Han periods as “ancient shamans.” These ancient wu xi (paraphrased from Lin
(2009)) had central functions of healing and divining, as well as performing sacrices to
the gods and spirits and rites of ancestor worship, assisting the sovereign in mourning
rites. Xi were people of high status and superior intelligence who held knowledge, respect

Religions 2023, 14, x FOR PEER REVIEW 12 of 34
and high regard. They were hereditary aristocrats who exhibited correct demeanor,
loyalty and trustworthiness and brought glory to the Kings. Their societal power was
illustrated in overseeing the religious aairs of the ruling class and the entire society
(state) and they enjoyed relatively great inuence. Their rituals involved special sacricial
vessels, vestments and the use of statues, animal gures and representations of gods and
ghosts, using sacrices, incantations and prayers to gods to obtain blessings. The ancient
xi likely engaged in ASC as suggested by their use of visions to illuminate maers and
their knowledge of how to “ascend and descend,” suggesting the shamans’ ights to lower
and upper worlds. Their rituals involved preparation with special baths and fasting and
ceremonies involving beating the drum, striking the bell and hollering to excite the heart,
with vigorous steps and dance to stir up the energies. The healing functions of the Wu xi
were considered their most important role, involving the use of drugs and plants to drive
away pestilence, incantations for the removal of illness and exorcisms and blessings to
avert misfortune.
Table 3 Comparisons of Shamans, Ancient Xi, Commoner Wu, Healers and Evenki šaman.
Leers indicate contrastive features of practitioner type: S = Shaman feature (FS & AS), AS =
additional Agriculture Shaman feature, H = Healer feature, P = Priest feature; S/W = Sorcerer/Witch
feature; ? = questioned by authorities; * = often aributed to ritualist type but not part of etic model.
Ritualist
Type
Principal Magico-
Religious Activity
Selection and
Training
Social and Political
Power
Professional
Characteristics
Motive and Context
of Ritual
Shaman
(Forager
Shaman = S;
(AS for added
Agriculture
Shaman
feature)
Healing and divination
Protection from spirits
and malevolent magic.
Hunting magic (S)
Cause illness, death (S,
S/W)
Assist Priests in
agricultural rituals (AS)
Dreams, illness, and
signs of spirit’s
request (S)
Vision quest by
individual alone in
wilderness (S)
Group training (AS).
Ceremony recognizes
status (AS)
High social status.
Charismatic (S)
Communal and war
leader (S)
Makes sorcery
accusations.
Informal power
Moderate judiciary
decisions (AS)
Predominantly male,
female secondary (S)
Part time.
Individual practice (FS)
Group ceremonies (AS)
Specialized role (AS)
Ambiguous moral
status (S)
Acts at client
request in
community-wide
ceremony
Ceremony over-
night (S)
Performance in
client group at client
request (AS)
Xi-“Ancient
Shamans”
(Shang and
Zhou periods
Healing & divining
Divine people’s fate,
regarding illness
Sacrices for deceased,
spirits and gods (M, H)
Ancestor worship (P)
Harvest, livestock (H, P)
Hereditary aristocrats
(P)
Ruling class (H, P)
Religious aairs of
state (H, P)
Very important and
great inuence
Males (H, P)
Correct demeanor, the
value of loyalty and
trustworthiness
People of superior
quality, with high
intelligence and
respected (*S)
Religious functions
of government (H,
P)
Commoner
Domestic
Professional
wu
Shang and
Zhou periods
Healing and protection
Arts of divination
Determine causes of
misfortune
Rituals to harm people
and gain advantage (S)
Worship ghosts,
animals and natural
phenomena
Hereditary family
trade
Innate selection or
from disease (S, M)
No formal group (S)
Commoner class (M)
Familial/tribal function
(S)
Prestige but lile
power (S, M)
Inuence communal
decisions, warfare and
hunting (S)
May be sanctioned,
executed (SW)
Male and Female (S)
Made a living but part-
time (S, M)
Often defective in body
or handicapped (S*)
Both heal and harm
(ambiguous moral
status) (S)
Services for clients
Covet goods and
heat people
Profession for gain
(H)
Healer (H)
Etic Ritualist
Type
Healing and divination.
Agricultural and
socioeconomic rites (H,
P)
Propitiation (H, P)
Life-cycle rituals.
Voluntary selection,
large payments to
trainer.
Learn rituals and
techniques
Ceremony recognizes
status.
High socioeconomic
status.
Judicial, legislative,
and economic power
(H, P)
Denounce sorcerers.
Predominantly male (H,
P)
Full-time (H, P)
Collective training,
practice and ceremony.
Highly specialized role
Predominantly moral
Acts at client
request and
treatment in client
group.
Collective public
rituals with Priests
Chinese
Reindeer-
Evenki šaman
Treatment of illnesses
Ward o misfortune
Seasonal celebrations
Dreams and visions
Illness/hysteria (S, M)
Could be inherited
Great signicance for
community =
charismatic (S)
Male & female (S)
Part-time
More mentally
capable(S)
Altruism for benet
of community (S)

Religions 2023, 14, x FOR PEER REVIEW 13 of 34
Life-cycle (marriage,
funerals and memorials)
Divining and predicting
Malevolence (S, S/W)
Solitude and fasting
in wilderness (S)
Self/Spirit taught (S)
Death-and-rebirth (S)
Unocial leader (S)
Not ocial clan head
or political leader
No material advantage
Courageous & strong
Ambiguous morality—
required altruism but
might abuse power (S)
Acted when person
or community was
disturbed
Community obliged
the ritualist to
perform (S)
Ritualist
Type
Supernatural
Power/Ritual
Techniques
ASC Conditions
ASC Characteristics
and Techniques
Healing Concepts and
Practices
Shaman
(Forager
Shaman = S)
(AS
Additional
Agriculture
Shaman
features)
Animal spirits & allies
(S)
Spirit power usually
controlled (S)
Impersonal mana (AS)
Spells, charms, exuvial
and imitative
techniques (AS).
ASC in training and
practice.
Soul ight/journey
(S)
Death-and rebirth (S)
Animal transform (S)
Mystical ASC (AS)
Isolation, austerities,
fasting, chanting,
singing, drumming
and dancing.
Collapse during ritual
Unconsciousness (S)
Soul loss (S)
Spirit aggression,
sorcery.
Sucking, blowing,
massaging and object
extraction (S)
Plant medicines.
Incantations for
removal
Xi-“Ancient
Shamans”
(Shang and
Zhou period)
Animals (S?)
Sacrices (H, P)
Pray for blessing (H, P)
Use statues and icons
(H)
Incantations
Bone Oracles (H)
Knew how to ascend
and descend (S?)
Vision illuminated
maers and hearing
penetrated them (S?)
Bathe and fast
Beat drum, strike bell
and holler
Excite the heart, body
Music, dancing, and
drumming
Knowledge of plants
Removal of illness
Exorcism (H)
Pray to gods for
blessing to averting
misfortune (H, P)
Dream interpretation
Commoner
Domestic
Professional
wu
Shang and
Zhou Periods
Spirits and ghosts of
dead, not “orthodox
gods”
Incantations, blessings,
praying (H)
Curses, charms (AS)
Sacrice (H, P)
Animals (?) (S)
Communicate with
spirits
Imaginary travel
around heaven &
earth (S)
Drumming and music
performed by
practitioners
Purgative drugs (S*)
Expelling baleful
inuences, Exorcism (H,
M)
Pray to invoke the
ghosts and gods for
blessings (H, P)
Sacrice (H, P)
Healer
Etic Ritualist
Type
Superior gods and
impersonal power
(mana).
Charms, spells, rituals,
formulas and sacrice.
Propitiate & command
spirits.
Material divination
system
ASC induction
limited
No apparent ASC (H,
P)
Social isolation;
fasting; minor
austerities; limited
singing, chanting or
drumming.
Exorcism and prevent
illness.
Manipulate body
Empirical medicine
Imitative and exuvial
techniques.
Chinese
Reindeer-
Evenki šaman
Supernatural qualities
of animals (S)
Changed themselves
into animals (S*+)
Master of the spirits
who were subservient
(S)
Techniques of ecstasy
Soul travel to spirit
world (S)
Death-and-rebirth (S)
Change into animal
(S)
Drumming and
singing central
elements for trance
Unconscious trance (S)
May use alcohol (P*)
Soul loss/recovery (S)
Remove malevolent
inuences of spirits
Drum and singing had
hypnotic inuence on
patient
Medicinal herbs and
plants
Table 3 Comparisons of Shamans, Ancient Xi, Commoner Wu, Healers and Evenki šaman.
3.2.1. Comparisons of Ancient Wu Xi and Shamans (SCCS Variable 879)
Table 3 provides the characteristics of the Wu xi and other healing wu. The xi did have
healing as a principal function, including the removal of illness characteristic of Shamans,
but their healing practices of exorcisms and blessings reect Healers. The ancient wu
acquired their power through formally learned prayers and rituals rather than selection
by the spirits and spontaneous and deliberate ASC characteristic of Shamans. The

Religions 2023, 14, x FOR PEER REVIEW 14 of 34
activities of ancient wu xi were based in ritual sacrices and manipulation of ritual objects
dedicated to the ancestors and gods of rain rather than the animal powers of Shamans. Xi
notably lack central qualities of Shamans such as malevolent acts and hunting rituals;
selection through illness and training alone in the wilderness; the focus on patient’s needs
rather than state functions; extensive ASC experiences involving soul ight/journey,
death-and rebirth and animal transformation; and healing of sorcery and soul loss
through a soul journey.
The xi did maintain some of the Shaman’s core ASC activities in their engagement
with practices of fasting, rituals involving beating the drum, striking bells and hollering,
as well as activities that excite the heart and exercise the body, such as dancing. But despite
statements about ight into the heavens and the mingling of divine and human entities
suggested in some texts about wu, these accounts do not have contexts where they aest
to experiences of ecstatic ights or out-of-body experiences characteristic of shamans
(Keightley 1998). Keightley proposes instead that the spirits did not descend into or enter
the wu (xi), but that instead they were descending to receive the sacrices that were
oered. In an analysis of ancient texts on the xi, Liu (2022, P 10 of 14) notes that “If we
emphasize that ecstasy is a judgment criterion for a shaman, Wu and Xi cannot be
categorised as shamans because this chapter did not mention ecstasy at all.”
As Boileau (2002) noted, while ancient (archaic) wu and Siberian hunting shamans
shared concerns with the well-being of the community, the xi’s involvement in the
religious functions of sacrices for the state and ancestors sharply contrast with Shamans’
primary role in communal healing. Similarly, the xi’s use of sacrices, praying to gods for
blessings and uses of statues and representations resemble Healers and Priests, not
Shamans (See Table 3). While xi shared Shamans’ healing features with the use of plant
drugs for the treatment of illness, they diered in an emphasis on exorcism and praying
to gods for blessings to avert misfortune. While some Shaman healing features related to
soul loss that are typical of Siberia and the etic model are aributed to xi, closer
examinations reveal that it is only a supercial resemblance. Williams (2020, p. 191)
reviews a poem called “Summons to the Soul” from the Elegies of Chu reecting activities
involving shamanic concepts of soul recovery. The poem describes the ritual activity of
summoning a soul back to save the patient’s life, with chants designed to impede the
soul’s travel to heaven, the underworld or four directions. Williams notes, however, that
this concept of a soul that can leave the body and the practices of soul loss recovery
expressed do not involve the personal soul journey recovery involving ecstatic ight to
the spirit world which is characteristic of Siberian practices and Shamans’ soul recovery.
Michael Pue (2002, cited in Michael 2015) illustrates central features of shamanism were
absent even in early China, which was lacking the beliefs in three spiritual realms and the
connecting axis mundi which were central to Eliade’s characterizations of the central
features of Siberian shamanism. Notably lacking in the ancient wu are Shamans’ trauma
and illness-related selection procedures and training involving ASCs (i.e., soul ight,
death and rebirth and animal transformation). Xi also lack special relations with power
animals, healing of soul loss and sorcery.
3.2.2. Comparisons of the Ancient Xi Wu with the Healer (SCCS Variable 881)
Healers provide healing and divination as principal functions, oering individual
healing services rather than collective healing ceremonies. The Healers also participate in
activities involving collaboration with Priests in collective agricultural rites and rituals of
propitiation of gods. Healers also have an important function in ociating life-cycle
rituals (i.e., naming, marriage and funerals). Healers acquire their roles through payments
for training by other Healers, and when completed, a professional group certies their
status with a ceremony recognizing their professional status. Their activities as full-time
specialists earn them a lucrative income. Healers are normally only males of high social
and economic status and participate in political, legislative and judicial processes.

Religions 2023, 14, x FOR PEER REVIEW 15 of 34
Healers’ supernatural power comes from ritual knowledge, particularly the creation
of charms and knowledge of spells and ritual techniques such as incantations and
sacrices used to propitiate the spirits. Their rituals do not seem intended to produce
ASCs in patients. Nonetheless, clients may experience minor ASCs because of placebo and
other consciousness-altering eects of the elaborate rituals, recited spells and incantations.
In the divination of information, the Healers do not rely on ASC experiences but instead
manipulate material systems (i.e., like I Ching or Tarot cards) that they randomize to
produce paerns that they interpret for diagnoses. Exorcism, removing spirits that aict
or possess the patient, is the central healing ideology of Healers and Mediums. Healing
rituals involved the repeated recitation of spells that appear to have hypnotic eects and
powerful encouragements apparently intended to have suggestive and placebo eects.
Although the Healers did not emphasize obvious ritual ASC, these are nonetheless
suggested by their training and pre-ritual periods of social isolation and fasting and
limited singing, chanting and percussion.
Xi features also characteristics of Shamans involving ASC induction (beating the
drum and vocalizations) and healing through plant medicines and expelling bad
inuences—these are not unique to Shamans but are shared by Healers as well. Xi more
closely resemble the Healers of Winkelman’s etic model, sharing characteristics such as
the following: religious functions such as propitiation of deities (King’s ancestors); social
qualities such as high socioeconomic status and governmental powers; being only males
who engage in functions with priests for collective rituals; providing sacrices to appease
superior gods; seeking blessings/propitiation; and healing through exorcism and the
prevention of misfortune. The Ancient xi use of oracle bones, a material system for
divinatory purposes, reects typical divinatory practices of Healers (use of mechanical
systems rather than intuition), not Shamans.
3.3. Commoner Wu
Commoner wu (Min wu 民巫 Lin 2016) were members of a professional class of the
late Zhou period. Tong (2002, pp. 62–65) summarizes the characteristics and functions of
what he referred to as “community” practices corresponding to the commoner wu. The
primary functions of these Commoner wu were providing healing services and prayers,
blessings and invocations to protect from disasters. Commoner wu were also skilled in
divination to determine causes of illness and other misfortune but notably lacked the
formal ancestral worship system characteristic of state rituals. Instead, these “community”
practices provided ceremonies for individuals and families with rituals involving the
worship of ghosts, animals and natural phenomena and provided spiritual support,
particularly to determine the client’s fortune or misfortune. Commoner wu also performed
rituals to harm people and gain advantage; consequently, the commoner wu were also
seen as potentially dangerous gures that needed to be controlled, or even banned.
The Commoner wu frequently practiced an inherited family trade but were selected
on the basis of innate abilities or as a consequence of disease, or were even bodily
handicapped. They could be male or female and acted on a part-time basis in providing
services at low expense to clients. Their functions were based on their abilities to
communicate with the gods and other spirits whom they queried with divinatory rituals
and addressed with sacrices to obtain their favor. They used charms and spells, as well
as prayers, exorcisms and sacrices.
While their role included services as tribal functionaries, they were without political
powers, formal organization or political privileges; nonetheless, they had prestige and
inuenced communal decisions regarding agricultural activities, as well as warfare and
hunting. Their performances were conducted without wrien scriptures but with singing,
drumming and dancing performed by the practitioners themselves. A signicant service
was exorcism rituals performed with incantations and sacrices. Beyond singing,
drumming and dancing, there is evidence of their ASC in purgative drugs used and

Religions 2023, 14, x FOR PEER REVIEW 16 of 34
reported ecstatic states of “imaginary travel around heaven and earth in chariots drawn
by dragons and phoenixes” (Tong 2002, p. 64).
Tong concludes that, in contrast to the state wu religious ocials of the Xia, Shang
and Zhou dynasties, the southern areas persisted with the ancient traditions and that these
“practitioners of southern folk religion still can be called wu” (p. 65). Tong also proposed
other translations for these earlier ritualists as magicians, sorcerers and even physicians.
3.3.1 Comparing the Commoner Wu and Shamans (SCCS Variable 879)
A comparison of the characteristics of Commoner wu with Foraging and Agricultural
Shamans (referred to here as Shamans) shows both similarities as well as some
divergences. Sucient detailed historical information is lacking for all relevant features
of the Commoner wu, but they reect Shamans in their principal functions such as healing
services, divination and engaging in practices of sorcery. Commoner wu also exhibit
Shamans’ features in selection on the basis of innate abilities or as consequences of disease,
as well as family tradition. The tribal and familial focus of Commoner wu is consistent
with the Foraging Shamans’ communal activities, as is their role as familial or tribal
functionaries without political power, formal organization or special privileges but
instead having prestige and inuence in communal decisions.
But community-wide rituals are not reported for Commoner wu; instead, client-based
practices characteristic of Agricultural Shamans are reported, as illustrated in their
emphasis on economic gain from their clients. Commoner wu also perform agricultural
rituals characteristic of Agricultural Shamans, as well as the worship of ghosts, animals
and natural phenomena, rituals more typical of Priests, Healers and Mediums. Both
Foraging Shamans and Agricultural Shamans’ features are also present in the commoner
wu’s part-time profession and reported involvement with activities of warfare and
hunting and malevolent rituals. But it is Agricultural Shamans that aempt to harm
through their use of curses, witchcraft (black magic) and charms, as do Commoner wu.
Commoner wu performances conducted without wrien scriptures or temples but with
drumming and music performed by practitioners themselves are consistent with, but not
exclusive to, Foraging and Agricultural Shamans. The Commoner wu’s engagement with
mimetic activities (singing, drumming and dancing performed by practitioners) are
features which are widely shared by other etic types (Healers, Mediums). But there is a
suggestion of uniquely Shamans’ ASC in ecstatic states of imaginary travel around heaven
and earth and even animal relations in the ecstatic travel via dragons and phoenixes and
the worship of animals.
Like Foraging and Agricultural Shamans, commoner wu provided healing services;
but while the healing functions of Commoner wu involves Shamanic concepts in expelling
baleful inuences and presumably countering sorcerers and witches, the Commoner wu
also focus on exorcism, prayers, blessings and invocations to protect from disasters and
sacrices; these are not like Shamans’ healing features, but instead reect activities typical
of Healers and Mediums. Notably lacking in Commoner wu are the Shamans’ practices of
the healing of soul loss. Furthermore, commoner wu’s healing functions dier from
Foraging and Agricultural Shamans’ healing in the focus on spiritual support of the
client’s fortune or misfortune and their emphasis on economic gain from their clients.
While some aspects of Shamans’ ASC are present, the typical experiences are not
directly reported for Commoner wu, and notably absent are the death-and-rebirth
experiences and the central role of animals as powers, identities and a form of personal
transformation, characteristic of Foraging and Agricultural Shamans. Features more
characteristic of Agricultural Shamans and Healers are seen in the Commoner wu’s
involvement with agricultural rituals and charging for services to clients. Features of the
Commoner wu such as agricultural rituals resulted from the eects of intensive
agriculture, features also found in Agricultural Shamans and Healers who also assist the
Priests in agricultural rituals. Similarly, the Commoner wu’s lower status likely reects the
higher status wu priests, a feature also exhibited by Agricultural Shamans. Techniques

Religions 2023, 14, x FOR PEER REVIEW 17 of 34
reported for Commoner wu—incantations, blessings, praying, curses and charms—
appear to reect the increasing commodication of healing practices manifested in the
Commoner wu’s focus on earning from clients. These more resemble features of
Agricultural Shamans and Healers—ritual techniques involving spells and charms with
exuvial and imitative rituals.
The Commoner wu far more closely resemble Foraging Shamans than did the Ancient
wu. These similarities include the evidence of trauma and illness in selection; the presence
of ecstatic ASC reective of soul ight; the relationships to animals; and healing involving
extraction of illness and addressing sorcerers. But notably, the Commoner wu lack a
communal audience and instead perform agricultural rites, use ritual techniques of
charms and spells, have lower social status and perform client-focused rituals, as do the
Agricultural Shamans with which they most directly correspond.
Lin (2009) characterizes the appearance of commoner or domestic professional
shamans in the literature as representing a major change. Lin aributes the appearance in
the literature of information about Commoner wu as a result of the eects of the collapse
of the Zhou feudal system (7th–6th centuries BC). But commoner wu practices are unlikely
to have been a consequence of the decline of the feudal system, although it may have
elevated their relative societal importance in the ensuing vacuum. If the Commoner wu
did not exist in the broader society during the Shang and Zhou dynasties, from where did
it originate? It was not a devolution of the bureaucratic wu or ancient wu yi, who had
already long departed from the Foraging Shaman paern. Rather, this appearance of the
Commoner wu in the literature must reect the emergence of record-keeping related to
these ongoing activities that undoubtedly had continuities with the pre-historic wu.
3.4. Bureaucratic Wu Priests
While archaeological evidence and texts lack sucient detail to infer all of the social
functions and characteristics of the wu before the emergence of writing and the state
system, evidence indicates an ancient dierentiation of functions of the tribal wu in the
development of “shaman-king” and a state wu (Tong 2002). Tong points out ancient
changes in the functions of the wu in the emergence of a new polity formed through their
symbiosis with the tribal elite, leading to a “religious elite of wu [who] reconstituted
themselves and became priests” (p. 53). Tong (2002) reviews ancient texts on the
consequences of these religious and political transformations during the emergence of
chiefdoms (“Period of the Five Legendary Emperors”, third millennium BCE) when
functions of the new elite ritualist’s centralized religious activity by organizing the
worship of the ancestors of the ruling clan as gods of the whole polity. These bureaucratic
functionaries of the government during the Xia, Shang and Zhou periods nonetheless had
wu in their titles, along with other formal designations that revealed their distinctive roles
as ocials of the state (see Table 4 for characters).
Chinese Character
Pinyin
English terms used by von Falkenhausen (1995) and Lin (2009)
男巫
nanwu
Male shamans
女巫
nüwu
Female shamans
司巫
siwu
Manager of the Spirit Mediums
師巫
shiwu
Officers
巫師
wushi
Instructors of Spirit Mediums
巫恆
wuheng
Spirit Mediums
無數
wushu
Male and Female Spirit Mediums
宗
zong
Temple Official
祝
zhu
Invocators
大祝
dazhu
Great Invocator
Table 4. Characters and Names of Wu Ocials.

Religions 2023, 14, x FOR PEER REVIEW 18 of 34
Michael (2015, p. 685) reviews texts that illustrate the distinctive characteristics of
these wu bureaucratic functionaries that he refers to as priests (zhu 祝) and temple ocers
(zong 宗). Michael (p. 685) characterizes the primary functions of these ritualists as
involving the performance of rituals for the ancestors. And rather than seances or face-to-
face communication with these spirits, the ritual performances involved sacrices and
other oerings, devotional prayers and songs and liturgical performances. Their religious
functions are assisted by temple ocers who manage ritual objects such as vessels and
jade and administer the sacricial oerings of animals and fruits of the harvest. Michael
notes that with the institutionalization of these roles in activities of the ancient Chinese
state religion within the spirit temples and ancestral halls that “the wu are no longer
recognized: shamanic authority has given way to centralized authority” (Michael 2015, p.
685).
Tong characterizes the activities of these wu state religious ocials as involving
primary responsibilities for state aairs and the needs of the royal family. These full-time
religious practitioners were formally appointed as government ocials to exercise
political powers and direct economic activities. Their positions depended on skills and
knowledge from study and long-term training, not supernatural gifts. They held high
prestige from their positions and were aliated with important political groups. The state
wu engaged in political and economic activity for the state, with each ritualist type holding
specic and delimited responsibilities. Some controlled groups of mediums (also called
wu) accompanied religious ocials in dancing during the rainmaking rituals and
addressed great calamities in rituals of crying, praying and singing. Their rituals followed
clearly established practices dictated in wrien texts and performed in special temples on
scheduled days. Their rituals emphasized sacrices, with some divination (dream
interpretation) and exorcisms.
Table 5. Comparisons of Priests, Mediums and Sorcerer/Witches with Tu Bo, State and Female Wu
and Wugu.
Leers indicate contrastive features of practitioner type: S = Shaman feature, AS = Agricultural
Shaman feature, M = Medium feature, H = Healer feature, P = Priest feature, SW = Sorcerer/Witch
feature; ? = questioned by authorities; * = often aributed to ritualist type but not part of etic model..
Ritualist Type
Principal Magico-
Religious Activity
Selection and
Training
Social and Political
Power
Professional
Characteristics
Motive and
Context of Ritual
Priest (P)
Etic Ritualist
Type
Propitiation and
worship.
Protection and
purication.
Agricultural planting
and harvest rites.
Socioeconomic rites.
Social inheritance
or succession.
Political action.
Incidental training
and/or by group.
Ceremony
recognizes status.
High social and
economic status.
Political, legislative,
judicial, economic, and
military power.
Exclusively moral.
Exclusively male.
Full-time.
Hierarchically organized
practitioner group.
High social and
economic status.
Acts to fulll social
functions,
calendrical rites.
Public rituals.
Ocial wu
(siwu, nanwu,
nuwu)—Zhou
dynasty
Sacrices to gods in
ancestral temple (P)
Protect from disaster(P)
Rituals for rain-making,
driving away
pestilence, protect
harvest (H, P)
Funerary services (H)
Selected/appointed
by government
personnel (P)
State ocials in
charge of training
(H, P)
Part of ocial structure
and ruling circles (H, P)
Regular members of the
bureaucracy (H, P)
Political and economic
activity (H, P)
Male , full-time (P, H)
Specialized hierarchy (H,
P)
Skills and knowledge
from study
High prestige from their
positions (H, P)
Social functions-
Formal rituals at
temples (P)
Calendrical rituals
at temples at
specied times (P)
Bo of the Tu
(Xing and
Murray)
Ensure good weather
and agriculture (P, H)
Ancestors worship (P)
Life cycle events
Community well-being
Resolve conicts
Not a source of healing
Hereditary, passed
down from one
generation to the
next within the
same family (P)
Leadership in collective
rituals (P)
Intervene to resolve
conicts (P)
Only males (P, H)
Part-time specialist
Annual festival (P)
Organized by
village association
Public festival in
temple (P)
Entire community
participates (S)
Medium (M)
Healing and divination.
Protection from spirits
Spontaneous
possession by spirit.
Low socioeconomic
status.
Predominantly female;
male secondary/rare.
Acts primarily for
clients at client

Religions 2023, 14, x FOR PEER REVIEW 19 of 34
Etic Ritualist
Type)
and malevolent magic.
Agricultural rituals.
Propitiation.
Training in
practitioner group.
Ceremony
recognizes status.
Informal political
power.
May designate who are
sorcerers and witches.
Exclusively moral.
Part-time.
Collective group
practice.
residence.
Also participates in
public ceremonies
Female Wu—
Qin and Han
dynasties,
especially
Warring States
Communicate with
gods
Assure well-being of
king and state
Control rains for
agriculture (P, M)
Healing rituals
Funeral rites
Appointed to
positions in the
royal courts
Group training (M)
Male spirit or god
descends on person
(M)
Generally low status
(M)
Contributed to
bureaucratic and
political decisions with
divination (H, P)
Female (M)
Organization served the
royal court (M)
Assure well-being
of king and state
(P)
Regular times
throughout year (P)
Sorcerer/ Witch
(SW) Etic
Ritualist Type)
Malevolent acts.
Kill friends, enemies,
neighbors, even kin.
Cause illness, death,
and economic
destruction.
Social
labeling/accusation.
Aribution of
biological
inheritance.
Innate abilities, self-
taught or learned.
Low social and
economic status.
Exclusively immoral.
May be killed.
Male and female.
Part-time.
Lile or no professional
organization.
Acts at client’s
request or for
personal reasons
such as envy, anger,
jealousy, greed or
revenge.
Practices in secrecy.
Wugu
Cause illness and death
(S, SW)
Denounced by
government
ocials (SW)
None
May be killed, executed
(SW)
Condemned as
immoral (SW)
Males and females (S,
SW)
Lower class (SW)
Personal gain,
payment (SW)
Revenge (SW)
Ritualist Type
Supernatural
Power/Control of
Power
ASC Conditions
ASC Techniques and
Characteristics
Healing Concepts and
Practices
Priest (P)
(Etic Ritualist
Type)
Power from ancestors,
superior spirits or gods.
Impersonal power and
ritual knowledge.
Propitiation and
sacrices.
No control over spirits.
Generally no ASC
or very limited
Occasionally alcohol (P)
Sexual abstinence,
isolation, sleep
deprivation.
Purication and
protection.
Public rituals and
sacrices.
Ocial wu
(siwu, nanwu,
nuwu)—Zhou
dynasty (1046-
256 BC)
Relate to royal
ancestors, high gods
(H, P)
Knowledge of texts and
rituals (H, P)
Appease ancestors (P)
Not
reported/apparent
(P, H)
Music, dancing, and
drumming
Exorcisms (M, H)
Sacrices (M, H, P)
Protection (P)
Prayer (H) *
Bo of the Tu
(Xing and
Murray)
Protective, ancestor and
tutelary spirits of
village (P)
Spirits not controlled--
may refuse to help (P,
H, M)
Procedures in hopes of
coercing spirits
Knowledge & skills
Possessed by spirits
(M?)
Become dierent
spirits
Spirits answer
through mouth of
bo (M)
Shivering with
presence of a spirit
(M?)
Uses a drum and
special chants and
dances
Not source of help in
illness (P)
Medium (M)
(Etic Ritualist
Type)
Possessing spirits
dominate.
Propitiation and
sacrices.
Power dominates
Unconscious.
ASC in training and
practice.
Possession ASC
ASC induced through
singing, drumming,
and dancing.
Tremors, convulsions,
seizures, compulsive
behavior, amnesia,
dissociation.
Possession and
exorcisms.
Control of possessing
spirits.
Female Wu-
Warring States
period
Appealed to deities
with sacrices &
prayers (P, M)
Decision maker the
possessing male spirit
(M)
Gods merged with
personality of wu
(M)
Possessing male
spirit or god
Induction with
dancing, incantations,
singing and wailing
Spirit responsible for
disease
Perform exorcisms (M,
H)
Herbal healing

Religions 2023, 14, x FOR PEER REVIEW 20 of 34
Spirit or god descends
on person (M)
descends on person
(M)
Sacrices, prayers, spells,
incantations
Anointing and ablutions
Sorcerer/ Witch
(SW)
(Winkelman
Etic Ritualist
Type)
Power from spirits and
ritual knowledge.
Contagious, exuvial
and imitative magic,
spells.
Power can be
unconscious, out of
control.
Indirect evidence of
ASC in reported
ight and animal.
transformation.
Night-time activities.
Illness by consuming
victim’s soul, spirit
Aggression, magical
darts that enter victim
Unconscious emotional
eects of envy, anger,
etc.
Wugu
Incantations and
Poisons
Imitative magic (S,
S/W)
Cursing (SW)
Burial of puppets
Nighime activities (S,
SW)
Ambition, greed (S/W)
Table 5. Comparisons of Priests, Mediums and Sorcerer/Witches with Tu Bo, State and Female Wu
and Wugu.
* = often aributed to ritualist type but not part of etic model.
3.4.1 Comparing Bureaucratic Wu and Priests (SCCS Variable 884)
The features of the bureaucratic wu and Priests are presented in Table 5. There are
direct correspondences of the features of the Priests discovered in Winkelman’s cross-
cultural research with the characteristics of these wu state functionaries and bureaucratic
ritualists central to Chinese religion and the clan institutions of ancestor worship. The
Priests hold dual political–religious roles, with supreme secular and religious power
generally invested in the group’s highest-level Priest. The central societal roles of Priests
give them high social status. The principal activities of Priests involve collective
agriculture (or pastoral) rituals that propitiate group deities with public sacrices and
feasts. This worship of the gods is to secure well-being in agricultural or other economic
activity through the performance of the most signicant calendrical rituals at the crucial
annual planting and harvesting cycles. The rituals associated with the agricultural cycle
are intended to aid the fertility of animals and crops, and the harvest rituals focus on
oering thanks to the deities for the abundance provided.
The supreme leader of the Priests is typically selected by patrilineal succession,
although political action and armed conict with competing relatives may be necessary to
secure the top position. Priests are normally limited to males, with females generally
serving only as assistants or servants unless it is the empress or queen who assumes the
position of high Priest by virtue of her royal position or widowhood. The head Priest is
assisted by diverse Priests who may have specialized training, often of a technical nature.
ASCs typically were not considered signicant in the training of Priests, but the training
did involve periods of social isolation, sexual abstinence and prolonged wakefulness.
Priests are typied by ancestor cults led by the senior lineage member who assumes
the role of the Priest for the group. Lower priests are senior lineage members and typically
have power over extensive economic resources and often hold important political
positions and exercise judiciary functions. Priesthoods are hierarchically organized in a
system of administrative control over society. Priests may exercise legislative functions
establishing the moral order of society. Priests obtain signicant economic resources from
their functions and their full-time profession.
Priests’ power comes from their close ancestral relations with spirits of collective
importance, especially ancestor spirits, village spirits and high gods. Priests are ritual
intermediaries with the spiritual world, petitioning deities on behalf of the community,

Religions 2023, 14, x FOR PEER REVIEW 21 of 34
but do not normally enter into the spirit realms nor claim to control the gods. Priests’ high
prestige and personal power related to their divine descent may also give impersonal
power such as mana (spiritual life-force energy). The Priest’s ritual knowledge guides the
timing of ocial public ceremonies and typically includes the sacrice of domestic
animals (cows, pigs, chickens, etc.) to the gods, which are consumed by participants in
community-wide festivals. The collective rituals often involved collective alcohol
consumption and feasting in prominent public places in collective ceremonies.
Healing is not a focal activity of Priests except in so far as to secure protection through
purication and well-being from the worship of the gods through sacrices. Although
individual healing services were not Priestly functions, healing eects were provided
through the psychological eects of purication and removing contaminated conditions
such as oending spirits or taboo violations. Propitiating and worshiping the gods was
also focused on well-being in seeking protection for the whole group, assuring abundance
and preventing future illness.
The close correspondences of these bureaucratic functionaries and the Priests
described in Winkelman’s cross-cultural research illustrate furthermore that they do not
remotely resemble Foraging Shamans. Michael asserts that the bureaucratic ritualists have
“direct contact or face-to-face communication between humans and non-bodily beings in
séance events” (Michael 2015, p. 671), suggesting ASC. This interpretation of the direct
communication of these wu with spirits is questioned by Keightley, who notes that the
spirits silently enter (descend) and leave the ritual after partaking in the sacrices they
have been oered. “There is nothing particularly shamanistic about the quality of the
religious experience described” (Keightley 1989, p. 10).
The Priests and bureaucratic wu share virtually every characteristic:
• primary religious activities (ancestor worship, agricultural well-being);
• selection (for the highest level, social succession/inheritance and political action);
• socio-political power (highest political, economic and judicial power);
• professional characteristics (male, full-time, specialized hierarchy, high status);
• motive and context of ritual (social functions, public calendrical rituals);
• supernatural power (ancestors, superior gods, ritual knowledge);
• ASC (very limited); and
• Healing (sacrices).
On the other hand, the bureaucratic wu and Shamans dier on virtually all features
besides high status. There is no empirical or rational justication to refer to the
bureaucratic wu as Shamans. As Tong (2002, pp. 53, 60) notes “From then on, it is no longer
proper to use the term wu to cover all the religious practitioners indiscriminately.
Henceforth, part-time folk magicians may still be referred to as wu, but the religious elite
of the community had become something closer to priests. . . . In their social status,
activities, functions, and beliefs, they were in fact priests.”
3.5. Female Wu
Literary sources from the Shang period emphasized the ancient distinctions of female
wu from male practitioners, but there was lile description of their activities in comparison
to the male wu xi and wu yi. They were represented with a variety of characters (i.e., 巫尪
Wu Wang and 巫嫗 Wu Yu; also see Table 2). But by the fth to third centuries BCE, only
accounts of wu who were females were well documented in literary sources and
government records because they held positions in the royal courts as advisors to supreme
rulers (Sukhu 2012) and performed rituals for the state religion, including funeral rites
(Michael 2015). The female wu were members of a coven who served the royal court
through the ministry of ritual and accompanied rituals in the temples where they
appealed to the ancestors and gods through sacrices.
These wu were appointed to positions in the royal courts as advisors to supreme
rulers and engaged in rituals involving dances and spirit communication to assure the

Religions 2023, 14, x FOR PEER REVIEW 22 of 34
well-being of the king and state, especially through rituals for rain which were seen as
important for agricultural abundance. Cai (2014, p. 145) characterizes the various
functions of these wu based on their abilities to communicate with the gods and other
spirits whom they queried with divinatory rituals and petitioned with sacrices and spells
to obtain their favor. Female wu played important roles in political decisions with
divination practices based on dream interpretation. But despite their employment by the
king during the Qin and Han dynasties to make sacrices to numerous spirits, especially
ancestors, the social status of the female wu was generally low, a pey position
comparable to court musicians and entertainers (Cai 2014, p. 145).
The ASC of female wu involved possession, a spirit communication in which a spirit
descends from heaven upon the person who is possessed, with the personality of the wu
entirely merged with that of the possessing male spirit. These experiences were also vital
for receiving divinatory revelations by the wu which were important in political decisions,
particularly decisions regarding war. Sukhu emphasizes their healing role as well, with
interconnected skills of communicating with spirits for divination to determine the spirit
responsible for a disease and treatments with exorcisms, sacrices and herbal healing.
Their healing rituals involved exorcisms carried out with incantations and sacrices for
appeasing the ancestors.
3.5.1 Comparing Female Wu with Mediums (SCCS Variable 882) and Shamans (SCCS
Variable 879)
The principal functions of Mediums involving healing, divination and spirit
communication are the same as Shamans, but their features are generally dierent. While
Mediums are similar to Shamans in dramatic experiences of ASC and an engagement with
spirits, their ASC are distinct in experiences of possession by a powerful spirit or god,
rather than the soul ight, death-and-rebirth and animal transformation of Shamans. This
selection through possession experience characteristic of Mediums presents psycho-
biological factors involving dissociation and trauma responses, including convulsions,
tremors and amnesia (Winkelman 2018). Mediums’ initial episode of spontaneous
possession typically occurs in late adolescence or early adulthood and is manifested with
tremors, convulsions, seizures, compulsive behavior and post-ASC amnesia. This episode
is considered an illness caused by an aiction of a possessing spirit, and the required
treatment initiates their training as a Medium. This training involves the ritual induction
of ASCs through singing, drumming and dancing during which the patient learns how to
control the possessing spirits and vocalize their demands. Once they achieve this control
they are cured and begin to serve in the capacity of a healer for the community. These
features of possession are not typical of Shamans.
Mediums lack the hunting ritual and sorcery features of Shamans. Instead, Mediums
are involved in agricultural rituals and the propitiation of spirits and deities, especially in
rituals to worship and propitiate their possessing spirits and make sacrices to them to
assure personal and collective well-being. Mediums are generally of low social and
economic status but have respect among women and beyond since they express the
personas of dominant male spirits. Mediums are subordinate to Healers and Priests but
may exert social inuence in the community because they manifest powerful spirits and
convey their messages. Mediums have supernatural power by virtue of their possession;
their bodies and voices are taken over by powerful spirits who communicate divine orders
for others to follow. Mediums are not thought to control these entities, but to express their
will. The Mediums’ primary ritual involves entry into a possession state to allow the gods
to work their inuence. Mediums also aempt to inuence these superior spirits through
sacrices.
Mediums’ diagnosis of illness and healing involves exorcism to remove the
inuences of possessing spirits and protection from malevolent spirits and witches.
Mediums produce their healing rst through control of their own possession episodes, as
well as through the general eects of ritual and ASC in eliciting endogenous healing

Religions 2023, 14, x FOR PEER REVIEW 23 of 34
responses (i.e., relaxation response, dissociation, placebo eects, hypnotic responses).
These possession experiences produce personal transformations by allowing women to
assume roles and express socially prohibited emotions. The widespread manifestations of
similar possession phenomena illustrate it is not to be understood in cultural particulars,
but instead as a manifestation of an evolved mechanism. This adaptation involves the
compartmentalization of consciousness to accommodate to accepting long-lasting
relationships and situations that are oppressive or even abusive (Winkelman 2018).
The features of female wu that are also typical of Mediums (but not Shamans) include
the following: agricultural rituals, their control by possessing spirits, lower social status,
predominantly female, a professional group, propitiation and sacrices and healing
involving exorcism. Lacking in these female wu are the central features of Foraging
Shamans, such as powers derived from animals and a belief in their transformation into
an animal; death and rebirth experiences during formation and an ASC of soul ight or
journey; and healing practices involving the recovery of soul loss and the extraction of
illness-causing sorcery objects. Furthermore, in those cases where the spirits descended
into the wu and spoke “through the mouths and bodies of the wu” (Michael 2015, p. 684),
we see evidence of possession rather than the shaman’s ecstatic or out-of-body experience.
Instead of the Shaman’s control over spirits, the spirits have control over the possessed
Mediums. Mediums are not typically associated primarily with animal spirits, but are
instead important to the gods, reinforcing Mediums’ reputation as exclusively moral
agents in contrast to the Shamans, who have moral duality balancing a reputation for both
healing and sorcery.
In spite of some literary mentions of the female wus’ ights to heaven (Williams 2020),
characteristics typical of shamanism of the traditions of north Asia, this soul ight feature
of Shamans’ ASCs is notably lacking in female wu. Rather than supporting Michael’s
argument for shamanism, the presence of possession is a clear indication of the female wu
being Mediums rather than Shamans. Michael reviews evidence in Chinese commentarial
precedents for this interpretation of possession in the classic texts in the word jiang zhi
which means ‘to descend and arrive’, with the spirit descending into the person to possess
the wu (also see Lin (2009, pp. 397–99), who argues that the meaning was clearly
possession). Their identity as Mediums is further reinforced by the distinguishing
ecological and social features of societies typied by Shamans (foraging) versus Mediums
(agricultural, with warfare and political integration) exemplied by these female wu in a
state-level agricultural society with rampant warfare between states (Warring States
Period). Other principal characteristics of these wu during the Warring States era
(paraphrased from Sukhu 2012) that correspond to features of Mediums but not Shamans
included being only women, their focus on the ancestors and gods, and their experience
of possession in training and professional practice, which they addressed with exorcisms.
3.6. Wugu as Malevolent Wu
Both earlier independent and bureaucratic wu underwent a further change at the end
of the Han dynasty, a “radical transformation, in which they were systematically
identied with the popular religion of the masses and became the targets of the active
suppression by the functionaries both of the Confucians and Daoists and, a bit later, the
Buddhists” (Michael 2015, p. 673). While Lin (2009) considers the criticism and doubts of
social leaders to have damaged the social image of the wu, it was the political interdictions
and aacks that led to their loss of political and social status. Although complete
repression and interdiction did not occur in the pre-Qin periods, it was well-developed by
the Warring States Period. After the transition from the feudal states to a unied imperial
empire, this tendency to forbid practices of the wu accelerated, leading to a dramatic
reduction in their status by the Han Period. Schafer also notes these dramatic changes in
the wu, astutely observing that “After the Chou dynasty, the female shaman . . . was forced
into sub rosa channels for the practicing of her magic arts, analogously to the witch of

Religions 2023, 14, x FOR PEER REVIEW 24 of 34
medieval Europe” (Schafer 1951, p. 134). Once the imperial system was in force, the wu
fell into the lower classes and were not able to recover their former glory (Lin).
Although wu were generally considered to be benevolent, performing white magic,
some texts indicate they also might engage in witchcraft. While this may have been a
confusion with practitioners referred to in some texts as wugu, some texts have aributed
malevolent practices to wu as well, who were employed by the powerful to cast evil spells
on the victim (Sukhu 2012, p. 75). In contrast to the generally benevolent wu, “Wugu was
the art of directing malevolent spirits to harm people” (Cai 2014, p. 146). The wugu
performed malevolent rituals with the use of poisons and the invocation of evil spirits for
personal reasons or to assist others to obtain power and wealth or take revenge against
enemies (Cai 2014, p. 146). Inscriptions of oracle bones indicate practices of manipulating
various insects and poisonous snakes to produce gu poison and the performance of
nighime rituals involving the manipulation of wooden dolls used to represent the
intended victims (Cai 2014, p. 146). Accounts of these practices of wugu aest to ritual
incantations to cause an evil spirit to invade a victim to cause illness and even death. Such
practices were prohibited, and people accused were publicly executed.
The importance of witchcraft and the execution of witches came to the forefront when
the aged and inrmed Emperor Wu accused a royal family of using a (foreign) wu to
perform a curse and manipulation of sorcery dolls to kill him. They were arrested and
tried and the entire family was executed. The notion that these practices came from
outside of the culture was indicated by the prosecutor Jiang who likewise hired wu from
central Asia to nd the sorcery items and exorcise the ghosts, leading to the arrest of
suspects who were tortured and put to death (Cai 2014, p. 145).
3.6.1 Wugu as a Sorcerer/Witch (SCCS Variable 883)
The features of the wugu correspond to the type of ritualist Winkelman labeled as
Sorcerer/Witch. The Sorcerer/Witch is devalued: an immoral aspect of the supernatural.
These ritualists generally deny the accusations of being a Sorcerer/Witch. They are
thought to be exclusively evil, to violate the moral order and even engage in acts of incest
and cannibalism. The Sorcerer/Witch is found in complex agricultural societies and is
signicantly predicted by political integration and warfare. Sorcerers/Witches are
considered to be exclusively malevolent, even causing illness and death to their own kin.
A Sorcerer/Witch is often tortured and may be publicly executed. They are denounced for
the destruction of economic resources, especially agriculture and livestock. The
Sorcerer/Witch may perform their evil for clients but are generally thought to act out of
revenge and for their own personal benet or being motivated by negative emotions
(anger, jealousy, envy, greed). Because of these emotions, the eects of the Sorcerer/Witch
may operate unconsciously. They may, however, have learned techniques or acquired
their power directly from their parents. Sorcerers/Witches exercise control over spirits and
use techniques generally involving curses and spells; contagious, exuvial or imitative
rituals; and discharges of power or darts that enter the victim. The Sorcerer/Witch engages
in nighime rituals and is believed to be able to y or transform into an animal.
The wugu do exhibit many of these features—devalued, immoral, being publicly
executed, acting out of envy and greed and using curses, spells and imitative magic. And
like the Priests and Healers who hold government positions, governmental ocials were
central in designating who is this antithesis of morality and accusing, judging and
executing those whom they considered to be guilty. In Winkelman’s (1992) research,
Sorcerers/Witches appear to generally be innocent victims, or local shamanistic healing
traditions, rather than evil ritualists primarily engaged in malevolence. These competing
ritualists are persecuted in a conict between local culture and hierarchical power,
reected in the signicant correlations of the Sorcerer/Witch with political integration
beyond the local community—such as a kingdom or empire. Winkelman’s model of the
formation of the Sorcerer/Witch is therefore consistent with the actual nature of the
ambiguous wu practices, being used for good (healing, worship) as well as for evil (curses,

Religions 2023, 14, x FOR PEER REVIEW 25 of 34
spells). Chinese history in the Han period provides clear evidence of this malevolent
activity by certain parties, as well as the persecution of the guilty and innocent alike by
government authorities.
3.7. The Chinese Reindeer-Evenki Shaman
If there is and was a “Chinese” shaman within the contemporary era, it clearly would
include the ritualist of the Chinese Reindeer-Evenki, who appropriately use the
indigenous term šaman to refer to their practitioners. Heyne (1999) integrated his own and
others’ research on their characteristics and activities; this is paraphrased and
summarized below.
The principal religious activity of the šaman was the treatment of illnesses, divination
and predicting the future and the performance of various sacrices to ward o misfortune.
The šaman also served as guides for the departed to the land of the dead and led seasonal
and social rituals (e.g., annual celebrations, weddings and memorials for deceased
persons). While the šaman was expected to act in an altruistic manner, they were also
regarded with feelings of fear for their ability to perform malevolent rituals. Rival
shamans engaged in magical bales that might end in the death of one of them.
The call to be a šaman came in mysterious illnesses, dreams and visions received from
the tutelary spirit of deceased ancestors, reecting the inheritance of the ability to become
a shaman. Once they fell victim to the call, the initiate often became depressed, distracted
and unresponsive and suered aacks of hysteria accompanied by convulsions and
periods of insensibility. When the clan system was still intact, the neophyte received
orientation from an experienced shaman of another clan, but now the shamanic initiation
always occurred as self-initiation, as a purely individual endeavor. The šaman sought long
periods of solitude and fasting in the wilderness where the person fell into a state of
ecstasy and lost consciousness. In this state, they received the call and teachings of the
spirit powers that gave the šaman the strength to conduct a new life through conversing
with the spirits. During their long meditation, the candidate was killed and cut into pieces
by the spirits who consumed the esh of the initiate and afterwards reassembled the bones
and brought the initiate back to life. When the initiate returned to the community the
person was already a shaman.
The šaman was a person of great signicance for the community but was never the
head of the clan or a political leader. The šaman could be male or female and their part-
time service did not generate any material advantage. The šaman had greater mental
abilities and courage than other clan members: a capable and inspired person. But despite
their altruistic service, the šaman had an ambiguous moral status based on the possibility
that he or she might abuse their power someday. Nonetheless, their rituals were
performed for the benet of the whole community when the normal life of a group
member was disturbed, or the group as a whole was falling into disorder. The
community’s feelings of condence obliged the shaman, who could not avoid this duty
once the call was received. The shaman’s task required an entirely altruistic behavior and
did not generate any material advantage.
The supernatural power of the šaman was derived from relations with animals,
especially the red deer, whose natural and supernatural qualities were transferred to the
costumed šamans as they rapidly danced around the re. The šaman were thought to be
capable of changing themselves into animals, i.e., into the animal-shaped bearers of their
souls, such as bear, elk or reindeer. The šaman was a master of the spirits who were
subservient and pliable to the needs of their shaman master. The šaman had command of
techniques of ecstasy. Singing and drumming were the most important techniques for
inducing the šaman to fall into a trance. The beat of the drum carried the Evenki shaman
to mount the magic elk or deer to travel to the otherworld. The soul of the šaman traveled
into the spirit world, changing into their alter ego in an animal shape world while lying
unconscious on the oor. Šamans might also use alcohol to facilitate their spirit journey
and increase their powers over spirits.

Religions 2023, 14, x FOR PEER REVIEW 26 of 34
The healing concepts of the šaman involved belief in the malevolent inuences of
spirits which could cause many forms of misfortune. The šaman had to overcome and
chase out the spirits. Soul recovery was used to recover the lost soul of a patient on its way
to the realm of the dead and bring it back safely. The shamanic ritual, particularly the
singing and drumming, had a hypnotic inuence on the patient. The šaman provided
protection for all members of the group with spiritual safeguards erected to help prevent
abnormal states of mind and misfortunes, illness, accidents and suicide. The šaman also
knew how to apply medicinal herbs and plants.
3.7.1 Comparisons of Šaman with Foraging Shamans (SCCS Variable 879)
Virtually all the exclusive features of Foraging Shamans are exhibited by the šaman.
The healing and divination, as well as malevolent acts, are shared, as is the selection based
on illness, dreams and visions from spirits. The formative period alone in the wilderness,
with fasting and visions leading to death-and-rebirth experiences are shared features; as
are the high status, lack of economic gain, both male and female part-time functionaries
and the altruistic action on behalf of the community they were obliged to serve. Šamans
controlled the spirits and their powers came from animals into which they were believed
to transform for journeys to the spirit world. These and other classic features of
shamanism and Shamans such as healing through soul recovery, removal of malignant
spirits and use of medicinal plants all illustrate features of the šaman which correspond to
Foraging Shamans.
The fact that the Chinese Reindeer-Evenki were hunters and pastoralists who
immigrated to China around 200 years ago and constitute an ethnic group outside of
mainstream Chinese culture is instructive. It is only outside of the core of Chinese culture,
in the tribal groups at the margins and periphery, that the Shaman is found in recent
history.
3.8. Is the Contemporary Chinese Tu Bo a Shaman?
The information that Xing and Murray (2018) present on the religious system of the
Tu ethnic group of Qinghai Province in Northwest China focuses on the bo , whom they
translate as shaman, but the actual descriptions they provide make it clear that it has
nothing to do with Foraging Shamans or Agricultural Shamans. Xing and Murray (2018)
note the ambiguous meaning of shaman and the problems of being “exported somewhat
willy-nilly to other religious specialists around the world . . . [but] nonetheless use the
term shaman” (p. 7 of 22). “What remains ambiguous, however, are the features that
dene shamans as a group, that justify a special label, and that distinguish them from
other religious specialists” (p. 6 of 22), but they arbitrarily characterize the shaman as
simply involving principally diagnosing and healing illness (also weather control and
contacting spirits).
But in discussing the Chinese ethnic Tu and their ritualist bo, whom they label a
shaman, they note the contrary, “that among the Tu there has been a radically diminished
involvement by shamans in the diagnosis and healing of illness” (p. 14 of 22), and instead
document the importance of the bo in agricultural (rain-making) rituals. The only
additional features they aributed to justify the label of shaman is that the bo are part-time
specialists who let spirits enter the community and are possessed by spirits to nd
solutions to individual, familial and collective problems through rituals of public
drumming and special chants and dances. All the core features of Foraging Shamans are
notably lacking.
While Xing and Murray suggest that the bo “ts neatly into the category of shaman,”
there is nothing about the bo that resembles the characteristics of the Foraging and
Agricultural Shamans, as seen in the characteristics of the bo discussed below (also see
Table 5). The authors explicitly recognize the inappropriate application of the term
shaman to the bo in stating “This public climatological function of shamanic performance

Religions 2023, 14, x FOR PEER REVIEW 27 of 34
is compatible with, but diers in ritual emphasis from, the cross-culturally more common
role of the shaman as a healer in private diagnostic and therapeutic sessions.” So why call
the bo a shaman? The features of bo enumerated in Table 5 illustrate no resemblance to
Shamans but rather closely resemble Priests.
Xing and Murray are paraphrased below to characterize the principal religious
activity of the bo. The bo ociates a festival to help ensure good weather, making
agriculture possible by sending rain and protecting against excessive rain. The bo also
performs rituals for life cycle events, such as weddings and for female fertility and
childbearing. The authors emphasize that the bo no longer provides a principal source of
help in illness.
The bo’s political power is illustrated in their role in providing solutions to collective,
familial and individual problems, and their leadership in collective rituals. Only males
may become a bo, but while not a full-time specialist, it is a hereditary position passed
down from one generation to the next within the same family (patrilineal succession). Bo
function principally in an annual festival coordinated with the planting season and
perform a public village festival in the temple where the entire community participates in
rituals to protect from natural disasters and to secure blessings for the community.
The rituals solicit the protective spirits longwang (dragon kings) and the niangniang
(queen mothers), benevolent gures associated with water and rainfall, as well as tutelary
and ancestor spirits for the care of the village and agrarian plots. The ancestors are
worshiped with rituals at ancestral altar gravesites with gifts. Notably, while the bo is
ascribed special knowledge and skills that bring the spirits to the community, he does not
control these spirits who may even refuse to keep their part of the ritual bargain. The bo is
ascribed an ASC, being possessed by spirits during ritual dancing in which he becomes
the various spirits who provide information through the mouth of the bo. The ASC is
induced by a drum and special chants and dances but is only notable from shivering
which indicates the presence of a spirit. While bo provided healing as a principal activity
in the past, they are no longer a principal source of help in illness.
The authors note that in the past, females typically functioned as bo. “Tradition has it
that in the past, not only could females function as bo, but that the bo were typically
females. That is now emphatically no longer the case. The contemporary absence of
females among the bo may be the result of a transition that came as an adaptation to Han
Daoist inuence” (p. 12 of 22). This suggests that the mediumistic functions of the bo were
usurped by men, who brought more of a priestly role to the position, reecting “syncretic
incorporation of Han Daoist practices into the inventory and repertoire of the bo.” (p. 12
of 22). The role of bo as Priests is illustrated in the most important contemporary role of
the bo involving performance in a public festival whose principal objective is the
recruitment of spirits to help in controlling the weather for assuring the success of
agriculture.
4. Results: A Summary of Shared Features and Divergences
The analyses above have provided direct comparisons of the wu ritualists with
Shamans and other etic types of ritualists. This section provides a summary assessment of
the similarities of wu7 based on the characteristics they share with Shamans or the other
ritualists they most closely resemble. The comparisons focus on dierentiating features,
those characteristics which distinguish Shamans (i.e., soul ight and soul loss), as opposed
to general characteristics shared by many types of ritualists (i.e., spirit interactions or
healing). Thus, the comparisons are based on the following 19 unique or diagnostic
features of Shamans:
• Ritual: (two variables) Community-wide ritual held over-night
• Distinctive Functions: (two variables) Malevolence/Sorcery, warfare
• Selection & training: (four variables) Dreams and illness as signs of spirit’s request,
vision quest alone in the wilderness, death-and-rebirth experience, self/spirit taught

Religions 2023, 14, x FOR PEER REVIEW 28 of 34
• Social Characteristics: (three variables) Charismatic leader, informal power, male and
female
• ASC: (two variables) Unconscious period, out-of-body (soul ight),
• Animal powers: (three variables) Animals as a personal and supernatural power,
experiences of animal transformation, hunting magic
• Healing: (three variables) Soul loss and recovery, sucking/object extraction, the
expulsion of aacking spirits (vs exorcism of possession).
Ancient wu (focused on the men = xi or wu yi, rather than women wu) have at most 2
of the 19 contrastive distinctive features of Shaman (perhaps out-of-body [soul ight] and
animal powers). On the other hand, the Ancient wu exhibit 11 distinctive features
characteristic of Healers and/or Priests. These include Priest—ancestor worship and
hereditary aristocrat status; Healer and Priest—agricultural and livestock rituals,
members of the ruling class, aendance of religious aairs of state, only males, the
performance of sacrices and prayer to gods for blessings to avert misfortune; Healer—
use of statues and icons, mechanical system of divination i.e., bone oracles and healing
through exorcism.
Commoner wu of the late Zhou period had 9 of the 19 contrastive characteristics of
Shaman (Foraging and Agricultural), including malevolence/sorcery, warfare, dreams
and illness as signs of spirit’s request, charismatic leader, informal power, both male and
female, out-of-body experience (soul ight), animals as personal and supernatural power
and hunting magic (as well as many other Shaman features which were not unique to
Shamans, i.e., part-time, ASC, communicate with spirits, no formal group, etc.). On the
other hand, the Commoner wu had only four contrastive features characteristic of Healers
or Priests.
State religious ocials did not have any features unique to Shamans, but instead had
virtually all of their features diagnostic of both Priests and Healers (12) or just Priests (six)
or Healers (two).
Female wu of the Warring States period had none of the unique contrastive features
of Shamans. Instead, the female wu had six8 variable areas exclusive to Mediums (group
training, male spirit or god descends on person, possessing spirit, gods merged with the
personality of wu, low social status, principally or exclusively female); three that Mediums
share with Priests or Healers (control rains for agriculture, perform exorcisms and appeal
to deities with sacrices and prayers) and three characteristics typical of Priests (assure
well-being of state, rituals held at regular times throughout year and contributed to
bureaucratic and political decisions with divination).
Guwu of the Han period had only four characteristics shared by Shamans and the
Sorcerer/Witch type (nighime rituals, males and females, imitative magic [AS] and cause
illness and death) and seven which were uniquely characteristic of the Sorcerer/Witch
type.
The Chinese Reindeer-Evenki šaman had 15 of the 19 contrastive features
characteristic of Shamans (community-wide ritual, malevolence/sorcery, dreams and
illness as signs of spirit’s request, vision quest alone in the wilderness, death and rebirth
experience, self/spirit taught, charismatic leader, informal power, both male and female,
out-of-body (soul ight) experiences, animals as a personal and supernatural power,
experiences of animal transformation, healing of soul loss and recovery and expulsion of
aacking spirits (vs exorcism of possession).
The bo of the Tu ethnic group of Qinghai Province Northwest China only has one
feature diagnostic of Shamans (community-wide ritual activity), but eight characteristic
features of Priests, three shared by Priest and Healers and three possibly indicative of
Mediums. I say possibly because the assertion of possession for the bo, while a
characteristic of Mediums, does not have the features of Mediums cross-culturally
(convulsions, amnesia, erratic behaviors, etc; see Winkelman 2018).
These assessments indicate that the etic categories for the Chinese ritualists are as
follows:

Religions 2023, 14, x FOR PEER REVIEW 29 of 34
• Ancient wu xi or wu yi: Healer
• Commoner wu: Agricultural Shaman
• State religious ocials, bureaucratic wu: Priest
• Female wu of the Warring States period: Medium
• Guwu of the Han period: Sorcerer/Witch
• Chinese Reindeer-Evenki šaman: Foraging Shaman
• Bo of the Tu ethnic group of Qinghai Province: Priest
5. Discussion: Critical Assessment of Translating Wu as Shaman?
Determining whether a wu is a shaman or other type of ritualist should depend on
their correspondence to cross-cultural paerns of shamanism and other religious ritualists
described above (Tables 1, 3 and 5). Keightley (1998) made astute criticisms of the
hypotheses that early Chinese ritualists were shamans in any way related to Eliade’s
conceptual framework. Williams (2020) further reviews studies (von Falkenhausen 1995;
Keightley 1998; Boileau 2002) that have illustrated the error in labeling as shamans the
various early Chinese religious practitioners referred to as wu. Yet many publications have
promoted confusion by asserting a false equivalence of wu and shamans.
Michael (2015) notes “Confusion in using the category of shamanism arises in part
because of a lack of consensus on which features to include in it, and that consensus can
only be informed by adherence to culture-specic representations that then can be utilized
in wider, cross-cultural studies” (p. 664). But it is the culturally specic allegations
(denitions) of what is a shaman independent of some objective criterion that undermines
eorts after consensus. We need cross-culturally valid concepts of shamanism such as
those discovered in Winkelman’s research, not some concept that loses comparative utility
by changing from culture to culture.
But Michael seems to absolutely reject a cross-cultural denition of shamanism in
asserting shamanism “does not exist as a natural piece of human behavior demonstrating
essential qualities to be discovered and cataloged” (p. 665). The cross-cultural research of
Winkelman and the biogenetic features shows that Michael is wrong. Cross-cultural
studies can create a consensus about just what the concept of shaman represents, and
whether it can be appropriately applied to culture-specic phenomena. A problem with
Michael’s approach is that he denes a shaman in a manner inconsistent with his own
criteria. He alleges that “The presence of shamanism in any society is recognized by their
representations of a séance event of direct contact (possession) or face-to-face
communication (spirit journey) between human beings and bodiless beings for the benet
of the community” (p. 665). But such characterizations are so broad that it frustrates other
aspects he emphasizes in his eort after a denition, which “also has to dierentiate the
shaman from all other religious and political functionaries; any denition that does not
aend to this will inevitably be so open-ended that it will lose all tractability” (p. 659). The
resolution of this problem of excessive open-endedness demands an ethnological
perspective such as presented above, but which has not previously been applied to uses
of the term wu.
Such cross-cultural characteristics of the empirically derived etic typology of
ritualists determined by Winkelman’s research provide an empirical framework to assess
what is a shaman in a cross-cultural perspective and across Chinese history. The
comparisons of this framework with Chinese ritualists presented above show that most
types of Chinese ritualists called wu do not correspond to Foraging or Agricultural
Shamans, but rather to other types of ritualists. This illustrates the necessity to distinguish
the diverse types of ritualists called wu and relate them to cross-cultural paerns of
ritualists besides shamans. Since the contrastive features of wu ritualists clearly map onto
the ethnological cross-cultural model, it shows that such perspectives help articulate
Chinese ritualists in comparative perspective and remove the erroneous perception that
all wu are shamans. As illustrated above, while principal aspects of the features of the pre-
historic and Commoner wu do correspond to the features of Foraging Shamans, the other

Religions 2023, 14, x FOR PEER REVIEW 30 of 34
types of wu (Ancient, State, Female and Wugu) do not correspond to the features of the
Foraging & Agricultural Shamans, but rather other types of ritualists (Healers, Priests,
Mediums and Sorcerers/Witches, respectively).
Boileau (2002, pp. 354–55) illustrates these diverse meanings of the wu character in its
reference to many dierent types of ritualists and supernatural roles, including diviner, a
possessed medium, a ritual scapegoat, a sorcerer, healer, priest and even ancestor worship
practices and eventually activities suppressed by government ocials who prosecuted
and even killed wu during the imperial period for their practices, which were called
sorcery and witchcraft. What these associations of wu with such diverse functionaries
reect is change, a socioeconomic transformation of ancient shamanic ritual practices into
other kinds of ritualists.
Michael (2015, p. 685) proposes the experiences at the basis of Chinese state religion
and centralized authority were “instituted, not by ancient, all-too human shamans or
sages, but by the will of the spirits (speaking through the possessed wu) at a time before
there were institutionalized priests and temple ocers.” This suggests that by the time of
the early Chinese civilization with its state religion of ancestor worship, whatever may
have been the early remnants of an aboriginal Chinese shamanism had already been
transformed into Healers and Mediums and the Priests who controlled them. Since the
Shang dynasty is considered the origins of Chinese civilization and was characterized by
diverse forms of wu ritualists (male wu yi, female wu and ocial wu), Shamans were absent
from the core of society at the beginning of Chinese civilization and only persisted at the
peripheries of culture and power. This is exemplied in the two clearest cases of Chinese
shamans presented here: (1) the Commoner wu, an Agricultural Shaman and (2) the
Chinese Reindeer Evenki šaman, who reects the classic features of Foraging Shamans.
Winkelman’s (2022) cross-cultural analyses show that these processes involved
agricultural intensication, supra-community political organization and eventually
warfare. These processes lead to the elimination of Shamans and the assumption of
religious and ritual functions by new types of ritualists that are not Shamans but rather
Priests, Mediums and Healers. This was the dynamic present at the beginning of the Shang
Dynasty— male xi Healers, the poorly documented female wu of that epoch and the
bureaucratic wu Priests. This dynamic expanded in the Warring years with the
development of the Female wu Mediums and the Wugu Sorcerer/Witch who was
persecuted by the Chinese state. These dynamics of evolutionary change in Chinese
ritualists are clearly reected in the sociocultural evolutionary mode of religion presented
in Figure 1.
What all the wu share is being a ritualist, not sharing the features of a Foraging or
Agricultural Shaman. The use of the root wu in denominating diverse types of Chinese
practitioners indicates that the best translation of wu is a religious ritualist.
5.1 So What Is a Shaman?
While cultural relativism might suggest that it makes sense to let each society dene
shamanism, there exists a problem that becomes clearer when asked in reference to other
concepts. By analogy, should each culture or society decide what a democracy is? Does
Putin’s claim to run a democratic country receive obligatory acceptance by scholars of
political systems? Or alternatively, should there be objective academic and scientic
standards of what a democracy is?
By analogy, if we are to have cross-culturally relevant characterizations of the
denition of a shaman, it must be based on empirical comparative data rather than
arbitrary or culturally specic denitions. Determination of whether a ritualist is a
shaman is not an issue of where the term originated or the culture where a particular
ritualist is found, but whether the ritualist resembles a well-delineated cross-cultural
phenomenon which justies the etic use of a word as a transculturally relevant concept.
Concepts such as bands, tribes and chiefdoms are commonly used in comparative
political assessments of the political complexity of societies. While each category shows

Religions 2023, 14, x FOR PEER REVIEW 31 of 34
variation, the dierences among them are clear and useful. No informed anthropologist
would confuse a band for a chiefdom. And even if people in chiefdoms band together for
some reason, they are still a chiefdom, not a band. By analogy, the dierences between
Shamans and Priests are clear, not arbitrary. Winkelman’s (1990, 1992, 2022) cross-cultural
research found a consistent paern of characteristics associated with the ritualists in
foraging societies, and these features correspond closely to the core concepts identied by
diverse scholars researching the nature of shamanism. This cross-cultural paern of
Foraging and Agricultural Shamans is the most objective criteria to use as a framework
for characterizing and evaluating what a shaman was. And even if a priest heals, divines,
talks to spirits, exorcises a patient and takes care of animals, his core features are that of a
Priest, not a Shaman.
6. Conclusions: The Wu Is a Religious Ritualist, Not a Shaman
As illustrated in the ethnological studies presented above, there are consistent
objective criteria about what the concept of shaman represents in foraging societies cross-
culturally, and the consistent beliefs and behaviors associated with these ritualists are the
criteria that should be used to determine whether the label shaman can be appropriately
applied to culture-specic phenomena.
But despite decades of publications documenting the inappropriateness of the term
shaman as a translation for wu, the use persists even among those who note that it is an
inappropriate term. Fu’s (2022) book has “shamanic” in the title, and in a chapter (6) on
“The northern shaman”, he writes about the well-preserved shamanic culture in myths
and legends, and in spite of appearing to claim otherwise, clearly rejects the notion of any
northern Chinese shamans: “on the basis of years of eld surveys on the peoples of
northern China, studies of shamans’ biographies and historical documents, and
interviews with the elderly and with old and new shamans, we can safely say that no
northern Chinese shaman really performs practices that accord with Eliade’s theories . . .
While performing sacricial rites for clan ancestors and nature deities, northern shamans
remain conscious and sober-minded. . . . northern shamans do not actually lose
consciousness, and their spirit certainly does not leave their body. Rather, they remain in
control of the emotional vortex of the rite. In a large-scale ritual, their seeming trance state
is actually a well-designed performance. In entering that state, their own human
subjective activity, not divine power, is what is mainly at work” (Fu 2022, p. 158).
To appreciate the nature of Chinese wu ritualists in a cross-cultural perspective,
particularly with respect to the concept of the shaman, we need what Feng Qu (2018) calls
for in assessing Chinese Mongolian ritualists, a two-way dialogue to resolve the
problematic aspects of both Western and Chinese perspectives on what has been called
shamanism. What is important is not just any Western perspective, but one informed by
ethnology rather than the romantic traditions Qu criticizes. Clarity in academic discourse
requires the unambiguous meaning of technical words. Translating wu as shaman
obfuscates, confuses and misleads. A wu is a ritualist and may be a Shaman, but more
likely is a Medium, Healer, Priest or even Sorcerer/Witch. Consequently, wu should be
translated into English as “ritualist, or as “religious ritualist” if a distinction from other
bureaucratic functionaries is needed.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Magico-religious-Practitioners data available at the Mendeley data
repository at hps://data.mendeley.com/datasets/34pjbr4kg4/2 Published November 18, 2020
Acknowledgments: Many thanks to Liang Liu for assistance with Chinese characters, pinyin and
English words used in Tables 2 and 4. Also thanks to the anonymous reviewers who requested
revisions that contributed to the robustness of the methods employed and ndings presented here.

Religions 2023, 14, x FOR PEER REVIEW 32 of 34
Conicts of Interest:

Religions 2023, 14, x FOR PEER REVIEW 33 of 34
Notes
1. For updated variables, values, variable descriptions, coding instructions and data see Winkelman and White (1987) or the
Mendeley data repository at hps://data.mendeley.com/datasets/34pjbr4kg4/2.
2. This access was made available by Doug White (RIP) but no longer appears available to the public.
3. Also referred to here as a Foraging Shaman.
4. Also referred to here as an Agricultural Shaman.
5. Winkelman (1992, p. 29) reports that the variables were aributed to a type if it was reported for 67% of the practitioners of the
type or if the incidence of the variable for the practitioner type was at least 50% of all cases reported for the variable. In the case
of Shamans, all cases of variables of less than 100% were correlated with data quality sources, with the consistently positive
correlations indicating missing data (under reporting) rather than true absence.
6. Analyses were performed in the CosSci program housed at the University of California, Irvine at
hp://socscicompute.ss.uci.edu/. This system is no longer publicly available.
7. I have not provided a comparison with the Pre-historic Neolithic wu because of the lack of adequate data for a meaningful
assessment.
8. Table 5 shows more than six matches with Mediums, but they are overlapping features.
References
(Allan 2015) Allan, Sarah. 2015. When Red Pigeons Gathered on Tang’s House: A Warring States Period Tale of Shamanic Possession
and Building Construction set at the turn of the Xia and Shang Dynasties. Journal of the Royal Asiatic Society 25: 419–38.
(Boileau 2002) Boileau, Gilles. 2002. Wu and Shaman. Bulletin of the School of Oriental and African Studies 65: 350–78.
(Cai 2014) Cai, Liang. 2014. Witchcraft and the Rise of the First Confucian Empire. Albany: State University of New York Press.
(Du and Kong 2000) Du, Yu, and Yingda Kong. 2000. Chunqiu Zuozhuan Zhengyi [A Commentary on Chunqiu Zuozhuan]. In Shisan
Jing Zhushu [A Commentary on the Thirteen Classic Texts]. Edited by Weizhong Pu, Kangyun Gong, Zhenbo Yu, Yongming
Chen and Xiangkui Yang. Beijing: Peking University Press. pp. 457ff
(Eliade 1964) Eliade, Mircae. 1964. Shamanism: Archaic Techniques of Ecstasy. New York, NY: Pantheon Books.
(Fu 2022) Fu, Yuguang. 2022. The Northern Shaman. In: Shamanic and Mythic Cultures of Ethnic People of Northern China II: Shamanic
Divination, Myths, and Idols. New York, NY: Routledge, pp. 148–64.
(Heyne 1999) Heyne, F. Georg. 1999. The Social Significance of the Shaman among the Chinese Reindeer-Evenki. Asian Folklore Studies
58: 377–95.
(Hopkins 1945) Hopkins, L. C. 1945. The Shaman or Chinese Wu: His Inspired Dancing and Versatile Character. Journal of the Royal
Asiatic Society of Great Britain and Ireland 1945: 3–16.
(Keightley 1998) Keightley, David. 1998. Shamanism, Death, and the Ancestors: Religious Mediation in Neolithic and Shang China
(ca. 5000–1000 B.C.). Asiatische Studien 52: 763–832.
(Lin 2016) Lin, Fushi. 2016. Wuzhe de Shijie [The world of Wu]. Guangzhou: Guangdong Renmin.
(Lin 2009) Lin, Fu-Shih. 2009. The Image and Status of Shamans in Ancient China. In Early Chinese Religion, Vol. 1: Part One: Shang
through Han (1250 BC–220 AD). Edited by John Lagerwey and Marc Kalinowski. Handbook of Oriental Studies. Leiden: Brill,
vol. 21-1, pp. 397–458.
(Liu 2022) Liu, Liang. 2022. Can Wu and Xi in Guoyu Be Categorised as Shamans? Religions 13: 741.
https://doi.org/10.3390/rel13080741.
(Mair 1990) Mair, Victor. 1990. Old Sinitic *MYag, Old Persian Magus, and English ’Magician’. Early China 15: 27–47.
(Michael 2015) Michael, Thomas. 2015. Shamanism Theory and the Early Chinese Wu. Journal of the American Academy of Religion 83:
649–96.
(Murdock and White 2006) Murdock, George, and Doug White. 2006. Standard Cross-cultural Sample. On-line edition. UC Irvine:
Social Dynamics and Complexity. Available online: https://escholarship.org/uc/item/62c5c02n (accessed on).
(Puett 2002) Puett, Michael. 2002. To Become a God: Cosmology, Sacrifice, and Self- Divinization in Early China. Cambridge, MA: Harvard
University Press.
(Qu 2018) Qu, Feng. 2018. Two Faces of the Manchu Shaman: ‘Participatory Observation’ in Western and Chinese Contexts. Religions
9: 388. https://doi.org/10.3390/rel9120388.
(Schafer 1951) Schafer, Edward. 1951. Ritual Exposure in Ancient China. Harvard Journal of Asiatic Studies 14: 130–84.
(Shuowen Jiezi 2014) Shuowen Jiezi, Gulin. 2014. Commentaries of Shuowen Jiezi. Edited by Fubao Ding. Beijing: Zhonghua, pp. 5020–
24.
(Sukhu 2012) Sukhu, Gopal. 2012. The Shaman and the Heresiarch: A New Interpretation of the Li Sao. Albany, NY: State University of
New York Press.
(Takigawa 2015) Takigawa, Kametaro. 2015. Shiji Huizhu Kaozheng [Commentaries on the grand scribe’s records]. Edited by Haizheng
Yang. Shanghai: Shanghai Guji, pp. 4206–7.
(Tong 2002) Tong, Enzheng. 2002. Magicians, Magic and Shamanism in Ancient China. Journal of East Asian Archaeology 4: 27–73.
(von Falkenhausen 1995) von Falkenhausen, Lothar. 1995. Reflections on the Political Role of Spirit Mediums in Early China: The Wu
Officials in the Zhou Li. Early China 20: 279–300.

Religions 2023, 14, x FOR PEER REVIEW 34 of 34
(Williams 2020) Williams, Nicholas. 2020. Shamans, Souls, and Soma: Comparative Religion and Early China. Journal of Chinese
Religions 48: 147–73.
(Winkelman and White 1987) Winkelman, Michael, and Doug White. 1987. A Cross-cultural Study of Magico-religious Practitioners
and Trance states: Data base. In Human Relations Area Files Research Series in Quantitative Cross-Cultural Data. Edited by David
Levinson and Roy Wagner. New Haven, CN: HRAF Press, vol. 3, pp. i–106.
(Winkelman 1986) Winkelman, Michael. 1986. Magico-religious Practitioner Types and Socioeconomic Conditions. Behavior Science
Research 20: 17–46.
(Winkelman 1990) Winkelman, Michael. 1990. Shaman and Other ‘Magico-religious Healers’: A Cross-cultural Study of their Origins,
Nature, and Social Transformation. Ethos 18: 308–52.
(Winkelman 1992) Winkelman, Michael. 1992. Shamans, Priests, and Witches. Tempe, Az.: Arizona State University Anthropological
Research Papers #44.
(Winkelman 2009) Winkelman, Michael. 2009. Shamanism and the Origins of Spirituality and Ritual Healing. Journal for the Study of
Religion, Nature and Culture 34: 458–89.
(Winkelman 2010a) Winkelman, Michael. 2010a. Shamanism: A Biopsychosocial Paradigm of Consciousness and Healing, 2nd ed. Santa
Barbara, CA: ABC-CLIO.
(Winkelman 2010b) Winkelman, Michael. 2010b. The Shamanic Paradigm: Evidence from Ethnology, Neuropsychology and
Ethology. Time and Mind 3: 159–82.
(Winkelman 2011) Winkelman, Michael. 2011. Shamanism and the Alteration of Consciousness. In Altering Consciousness
Multidisciplinary Perspectives. Volume 1: History, Culture and the Humanities. Edited by Etzel Cardena and Michael J. Winkelman.
Santa Barbara: Preager ABC-CLIO, pp. 159–80.
(Winkelman 2015) Winkelman, Michael. 2015. Shamanism as a Biogenetic Structural Paradigm for Humans’ Evolved Social
Psychology. Psychology of Religion and Spirituality 7: 267–77.
(Winkelman 2017) Winkelman, Michael. 2017. Shamanism and the Brain. In Religion: Mental Religion. Edited by Niki Clements. New
York, NY: MacMillan, pp. 355–72.
(Winkelman 2018) Winkelman, Michael. 2018. Shamanism and Possession. In The International Encyclopedia of Anthropology. Edited by
Hilary Callan. Wiley Online Library. https://doi.org/10.1002/9781118924396.wbiea1651.
(Winkelman 2021a) Winkelman, Michael. 2021a. A Cross-cultural Study of the Elementary Forms of Religious Life: Shamanistic
Healers, Priests, and Witches. Religion, Brain & Behavior 11: 27–45.
(Winkelman 2021b) Winkelman, Michael. 2021b. The Evolutionary Origins of the Supernatural in Ritual behaviors. In The
Supernatural After the Neuro-turn. Edited by Pieter F. Craffert, John R. Baker and Michael J. Winkelman. London: Routledge, pp.
48–68.
(Winkelman 2021c) Winkelman, Michael. 2021c. The Supernatural as Innate Cognitive Operators. In The Supernatural After the Neuro-
turn. Edited by Pieter F. Craffert, John R. Baker and Michael J. Winkelman. London: Routledge, pp. 89–106.
(Winkelman 2021d) Winkelman, Michael. 2021d. Shamanic alterations of consciousness as sources of supernatural experiences. In
The Supernatural after the Neuro-Turn. Edited by Pieter F. Craffert, John R. Baker and Michael J. Winkelman. London: Routledge,
pp. 127–47.
(Winkelman 2022) Winkelman, Michael. 2022. An Ethnological Analogy and Biogenetic Model for Interpretation of Religion and
Ritual in the Past. Journal of Archaeological Method & Theory 29: 335–89. https://doi.org/10.1007/s10816-021-09523-9.
(Xing and Murray 2018) Xing, Haiyan, and Gerald Murray. 2018. The Evolution of Chinese Shamanism: A Case Study from
Northwest China. Religions 9: 397. https://doi.org/10.3390/rel9120397.
(Xu 2002) Xu, Yuanhao. 2002. Guoyu Jijie [A Commentary on Guoyu]. Edited by Shumin Wang and Changyun Shen. Beijing:
Zhonghua Shuju.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury
to people or property resulting from any ideas, methods, instructions or products referred to in the content.