Shamanism and the Sacred Rhythms of Candomblé


As Ezra Pound would say, rhythm is the most fundamental aspect of music. As that which brings meaning and order into otherwise disconnected elements, rhythm is the necessary ground for a composition's unity. From a broader perspective, rhythm transcends the realm of music: its aesthetic necessity makes it is the indispensable foundation of all artworks. These are contingent upon the consolidation of their rhythm, a vital pulse that is always absolute and that conforms the great bass over which sounds, images and ideas are juxtaposed to produce a new metaphor, be it musical, literary, or visual. As rhythm brings order into chaos, it establishes aesthetic completion and meaning, allowing artistic communication and communion.

Rhythm is also the ground of ritual. It orchestrates processes and procedures, conducting successive moments of actions into desired outcomes. From rhythmic repetitions and reiterations emerges hierarchization. Time is made sacred in the performative duration of ritual, with rhythm framing a sacralized time. 

The sacred rhythms of Candomblé establish the unity of a performative séance. They underscore the ceremonial dramatization common to shamanic rituals, shaping the primeval rites that conform humanity's original form of religion. As an ancient system of communication with gods and spirits, shamanism is a timeless attempt at religare, the Latin word meaning to reconnect, that is, to restore the link between humans and the realm of deities that exists in parallel to earthly existence. Shamanic practices are, as Mircea Eliade called them, "archaic techniques of ecstasy." They involve trance and spirit possession, which are enacted through song and dance, expressive acts from which poetry also emerges as prayer and call.

All across the globe, shamanism continues to persist. It endures as long as humans are confronted with mystery. Throughout the ages, it withstood the onslaught of colonization, religious oppression, and demonization. Shamanism's is thus a history of resistance; one of continuous shock with politically stronger systems of thought, with resulting absorptions and adaptations. Its locus classicus is the Siberian Tungus, a topology that emerged from the works of Mircea Eliade and his Polish and Russian predecessors, Maria Czaplicka's and Sergey Shirokogoroff, whose Aboriginal Siberia (1914) and Psychomental complex of the Tungus (1935) preceded Eliade's Shamanism: Archaic Techniques of Ecstasy (1951) in pointing to the Siberian Tungus as the birthplace of the beliefs and practices that we came to understand as shamanism.

But what about shamanism's other faces? Are the new world indigenous populations' shamanisms equally Tungusic? Are those peoples ethnically Asian? Did they carry shamanism across the Bering land bridge some 12,000 years ago? Are their rituals similar to those found in Asia? Does their art and music show continuities?

And what about shamanism's African face? Is there a link between Africa and the Tungus? Are the shamanic manifestations found in Africa essentially different from those of East Asia? How is resistance enacted in both cases? Does music play a role? 

We shall consider East Asian shamanism, with its manifestations in countries such as China, Mongolia, Korea and  Japan. Below, I am including part of my current research on the topic of ancient shamanism's intersection with the arts. I am especially interested in the role of the female shaman in the transformation of music and dance from primarily ritual elements into artistic practices that transcended the religious sphere. I am considering also the relation between rhythm and trance as functions of shamanism's cultural resistance, as well as the broader political implications of that resistance. I believe that the confrontation between shamanism and more powerful systems of thought had a much larger role in the various processes of state-formation in East Asia that has been commonly assumed. That confrontation, I suggest below, had enormous consequences on the emergence of social stratification in the area, as well as on how gender relations were established according to the ensuing distribution of power. 

From a broader perspective, I am interested in how shamanic music compares in its East Asian, African, and Latin American versions. The persistence of rhythm and drumming as forms of inducing trance, aspects of vocal music and poetry brought out in song, as well as the role of female shamans as carriers of tradition are also elements that I am planning consider. In any case, as a caveat what appears below is not the product of a finished study and it may included digressions. If you decide to continue reading please be open minded. 

China, Mongolia, Korea and Japan comprise areas into which Tungusic peoples migrated at a time that preceded recorded history, carrying along with them their shamanic beliefs and rituals. These survive in various places across the region, including mainland China and Southeast Asia. Among the Han people, Wuism (Ch. 巫教 wūjiào) and the Tongji tradition (Ch. 童乩 tóngjī) remain alive as pre-Confucian religious practices. Among the non-Han, the Nuo folk religion (Ch. 儺), practiced by the Tujia (Ch. 土家族 tŭjiāzú), with its rituals of exorcism that seem to have identical roots with the Japanese Shinto tsuina ceremony (Jp. 追儺), as well as the shamanic healing practices involving the use of herbs and animal sacrifice performed by the Hmong people  (Ch. 赫蒙族 hèméngzú) are examples of shamanism persisting in mainland China among minority groups. In Japan, traces of Yayoi shamanism are found in various Shinto rituals, a fact that points to the Tungusic origin of what today is understood as the Japanese ethnic group. Shamanism appears to have prevailed in the archipelago in times prior to the adoption of writing and the emergence of a local recorded history in the 8th century CE. There shamanism served as the basis for a society marked by the power of the female shaman, embodied above all in the figure of Empress Himiko (c. 4th century CE). Shamanism was then repressed and transfigured to conform to the metanarrative of a powerful imperial ideology based on a mythological male lineage, an idea likely to have been introduced by horse-riding tribes of conquerors that invaded the archipelago from the Korean peninsula (following Egami Namio's theory). These brought in new religious views and practices, such as Confucian ancestor worship and a Buddhist vertical cosmology, elements found in present-day Japanese customs such as the Buddhist-Confucian Bon Festival (盆). Meanwhile, the ancient miko-shamans with their sensual dances marked by trance and ecstasy were assimilated into the "way of the kami"'s mythological construct. At some point they became descendants of Ame-no-Uzume, the goddess of mirth and revelry, even as they continued to sell their bodies to men seeking direct contact with the gods through sexual exchange with a shaman in trance. Scholars such as William Fairchild, Carmen Blacker, and Hori Ichirō have pointed to the power struggles that led Japanese shamanism to be constrained under what came to be called the Shinto religion.

In Korea, shamanism was also repressed. Closer geographic proximity to the Siberian Tungus and distinct political conditions, however, seems to have made the Korean version more resilient than the Japanese one. Several of the ancient peoples and tribes that inhabited the peninsula and what is today Manchuria practiced shamanism, such as the proto-Mongol Xianbei (鲜卑), the Xiongnu (匈奴), the Yamaek (穢貊), and later the Jurchen (女真), ancestors of the Manchus. From old, these peoples entered into confrontation with the Chinese, leading Han dynasty historian Sima Qian to refer to them as barbarians, or yí (夷), as opposed to the civilized Chinese, called Huaxia (華夏). This established the so-called Hua–Yi distinction (華夷), also known as the Sino-barbarian dichotomy. The opposition between the powerful Chinese and the nomadic groups found outside China's borders emerged before the Han times of Sima Qian, probably during the Eastern Zhou period (770–256 BCE), when nomadic tribes such as the Xianyun (獫狁), the Xirong (西戎), the Dongyi (東夷), the Beidi (北狄), and the southern Nanman (南蠻) started being identified by character names that included graphic pejoratives, such as bei (卑 low), nu (奴 slave), yun (狁 a type of dog), and man (蠻 a type of insect). The most remarkable of this denominations is perhaps the one given to the Yemaek (Hanja 穢貊, Hangul 예맥), or Huìmò, in the Chinese pronunciation, which includes the character huì 穢 (dirt, filth) and mò 貊 (wild, savage). The boundaries separating these groups is unclear and at times the ethnic slurs were used interchangeably to signify the same people. In general terms, the Yemaek are thought to be the direct ancestors of the Korean ethnic group, as well as the ones responsible for the formation of Kojosŏn, which lasted from its mythical founding by Tan'gun in 2333 BCE until its fall to the forces of the Han in 108 BCE. The same pejorative character used to name the Yemaek would appear centuries later in Japan in the word eta (穢多), a term that means "much filth" and was employed in Japan to refer to the country's outcastes. The Japanese pronunciation e (穢) appears to derive from the Korean ye (예). The original Chinese character huì (穢) was employed by the Japanese in rendering also the word kegare (穢れ), the abominable uncleanness or defilement issued from the violation of Shinto taboos which is the possible source of the eta's stigmatization. The presence of the character huì (穢) in both the words Yemaek (穢貊) and eta (穢多) has been pointed out as indication of the Korean origin of the Japanese outcaste group. This idea ramifies into that of the Korean ancestry of other marginal groups that existed in Japan, such as the sanka (山窩), a group of mountains dwellers who could be found in parts of Honshu until the second-half of the twentieth-century. These groups might have originated from the Korean slave stratum, or from the Korean outcaste group called paekchong, the rationale being that in fleeing slavery and persecution in Korea these groups would have migrated to the archipelago. 

The emergence of outcaste groups in pre-modern Korea and Japan has been hypothesized as resulting from warfare among different tribes or proto-national states in the region, with prisoners being enslaved and their descendants stigmatized. Notwithstanding, the origin of these groups may also be related to important religious differences existing between those in power and the lower classes. More broadly, those religious differences may also be at the root of the Sino-barbarian dichotomy mentioned above. The Yemaek's stigmatization through the employment of a Sino-centric ethnic slur appears to be connected to religious practices that made the former seem primitive in the eyes of the latter, namely the increasingly Confucian Chinese aristocracy. The emergence of the Hua–Yi dichotomy during the Eastern Zhou coincides with the onset of Confucianism and the adoption of Confucian rites as official Zhou rituals. These became symbolically powerful and politically determining, turning into a meter of sophistication and cultural refinement in the social arena. The Sino-centric ethnic slurs mentioned above were employed in regard to nomads and outsiders who did not take part in Confucian rituals, did not follow Confucian ethics, and practiced shamanic rituals.

As Confucianism was transmitted to Korea and Japan, the barbarian pejorative moved from being used by the Chinese towards the Yemaek to being employed by the Yamaek rulers who eventually adopted Confucianism towards the common people who remained linked to shamanic practices and rituals. In Korea this would become more evident after the rise of the Yi Dynasty, which embraced Confucianism more thoroughly. In any case, a clash between Confucianism and shamanism can be identified in both the peninsula and the archipelago, with major consequences for the arts in both places. Considering the transmission of Confucianism to Korea and Japan, we note that Confucius's ideas must have reached the northern part of the Korean peninsula in at least as early as the time of the Han commanderies, which started being implemented in 108 BCE (on the history of Korean Confucianism see Kang Jae-eun, The Land of Scholars: Two Thousand Years of Korean Confucianism (Homa & Sekey Books, 2005). In Japan, although the Kojiki and the Nihonshoki relate the transmission of the Confucian classics to the archipelago by the legendary Paekche scholar Wani (王仁)  early in the 5th century (the same scholar being construed also as having introduced the Chinese writing system), the onset of the Kofun period one century earlier, with its characteristic of having replaced the political power of female shamans, as mentioned above, with that of a male imperial lineage, suggests the possible transmission of Confucianism at the beginning of the 4th century CE, that is, carried in by those who established the male-centered dynastic system. An important factor to be considered in this regard is that the start of the emergence of the Kofun period in the archipelago coincides with the fall of the Han commanderies in the peninsula, a fact that suggests the possibility of horse-riding invaders issued from disbanded Chinese armies having entered the archipelago and implanted a Confucian mindset, if not Confucianism as such, in the less technologically developed Yamato Japan. This still an unexplored subject of research and speculation. In any case, the crucible of the early Kofun, whose major political and religious turn Robert Ellwood called a "patriarchal revolution," may suggest the beginning of shamanism's acculturation into Shinto. In this sense, Shinto may be seen as the very result of the clash between Confucianism and shamanism, where specific rituals pertaining to the latter, above all animal sacrifice, became interdicted and stigmatized, giving way to the emergence of an outcaste stratum locally.  

The clash between Confucianism and shamanism in East Asia had an enormous effect on the social and artistic practices conducted in the region. Shamanic rituals, as indicated above, are techniques of ecstasy involving music and dance. Issuing from this specific form of performative practice, the very body movements of the female shaman in trance led her to become an object of desire, opening an important dimension of sexuality to the rituals. The sensual nature of the female shaman's ecstatic dances, furthermore, seems to have corroborated with the stigmatization of shamanism as barbaric and uncivilized. A stigmatized class of female performers emerged in the male-dominated Confucian societies of China, Korea, and Japan, producing a courtesan culture of enslaved women artists. Soon the Chinese yìjì (艺妓), the Korean kisaeng (妓生), and the Japanese yūjo (遊女) appeared as female performers responsible for the production and transmission of the classical arts in their societies. In China, the emergence of a system of courtesans employed at the palace is  concurs with the dawn of Confucianism during the Spring and Autumn period (c. 771 to 476 BCE). That period saw the establishment of the guānjìzhìdù (官妓制度), an official system of female entertainers called nü yue (女樂), who are often seen as the predecessors of the aforementioned yìjì (艺妓), female performers who were forced into a life of bondage slavery and prostitution in the brothels of China's major cities (see《中国妇女生活史》,第60页,商务印书馆1937年版跳转). Together with the phenomenon of urbanization, an entertainment industry was set in motion, mirroring the official courtesan culture proper to the nobility. Some yìjì (艺妓) became esteemed for their skills in dance and literary composition. As the centuries progressed, different forms of recruitment were devised for these female performers both at court and at the entertainment districts of the emerging urban centers. During the Ming Dynasty (1368–1644), for instance, they were frequently chosen among foreigners, notably the wives of Mongol captives, who were forced into courtesan life. In any case, they tended to come from the social strata regarded as "barbaric," and their performing arts developed from shamanic dances and rituals.  

In Korea, it is at the beginning of the Koryŏ Dynasty (918–1392) that the female entertainers emerged as a specific social group employed at the service of the court. The Court Entertainment Bureau (kyobang) was established around the time of King T’aejo of Koryŏ’s (877–943) first year (918AD), and to the kyobang young women selected from the lower classes would be assigned and trained, working henceforth as state property under the official title of kwan’gi (官妓). As time passed, these women began to be called kisaeng (妓生). One of the first recorded inquiries on the kwan’gi-kisaeng social type was produced by Sirhak scholar Chŏng Yagyong (丁若鏞 1762–1836, pen name Tasan). In the Aŏn’gakpi (雅言覺非 1819), a social treatise from the late Chosŏn period (1392–1897), Tasan asserts that the kwan’gi are one and the same with the such’ŏkcha (水尺者), a nomadic outcaste group that engaged in activities such as hunting, wickerwork, and water scooping. From the production of wickerwork (柳器匠 Kr. yugijang), and water scooping (水尺 Kr. such’ŏk), this group acquired the designation yangsuch’ŏk (楊水尺). While yangsuch’ŏk males were generally conscripted by the state as regular slaves, females were selected as performer/court attendants, or ki (妓).

If in China the nü yue (女樂) and yìjì (艺妓) were frequently selected by the Confucian aristocracy from the barbaric peoples that circulated around the empire, or who entered into war with it, such as the Mongols, in Korea this was reproduced in relation to the tribes that entered into conflict with a now similarly Confucian Korean state, such as the Jurchens. In the Aŏn’gakpi, Tasan further refers to the yangsuch’ŏk as sŏbŏn (西蕃), an ethnic slur used to signify “western barbarians.” The term sŏbŏn, as Yi Yun-sŏp explained, was employed since the Koryŏ period to refer to Jurchen tribes that inhabited the area equivalent to today’s China’s Heilongjiang Province and Russia’s Maritime Province (Primorsky Krai) and whose members were in constant conflict with the Korean kingdoms (Yi Yun-sŏp. Yŏktongjŏk koryŏsa [A Dynamic History of Koryŏ]. Sŏul: P’ilmaek Ch’ulp’an, 2004, p. 137. In this regard see also Ch'u Myŏngyŏp, “Koryŏ chŏn'gi 'Pŏn' (蕃) insikkwa 'Tong-Sŏbŏn ŭihyŏngsŏng” [The consciousness of 'Pŏn' in early Koryŏ and the formation of the eastern and western barbarians], Yŏk'sa wa hyŏnsil 43 (1999): 14–46.). Although the Koryŏ classified the Jurchen as tongyŏjin (東女眞), or eastern Jurchen, and sŏyŏjin (西女眞), western Jurchen, both were subsumed under the general term sŏbŏn. In this regard, Pak Ŭn-bong noted that the Jurchens were also referred to as pŏnjŏk (蕃賊), or savage people, when some Jurchen tribes were conquered in 1107, the second year of King Yejong’s (睿宗 1079–1122) rein.

Strained relations between the Koreans and the Jurchen led to the Korean-Jurchen wars, which extended from the 10th to the 17th century. Known by several variant names, the Jurchen were a Tungusic people who inhabited the region of Manchuria until around 1630, when they combined with some of their neighboring tribes to form the Manchu people. The Jurchen ancestors, the Mohe, or Magal (靺鞨) tribes were subjugated by the Korean state of Balhae. After the fall of Balhae in 926 (conquered, in turn, by the Khitan Liao dynasty), the Magal became the primary ethnic group from which the Jurchen descended. As Evelyn Rawski pointed out, late-Koryŏ official documents refer to the Jurchen as “savages,” or yain (野人), that is, as animals lacking the capacity to be civilized. As cultural assimilation proved impossible within the Koryŏ dynasty, the conquered Jurchens appear to have sunk into a pariah class wondering at the margins of society and performing the tasks Tasan described as proper to the yangsuch’ŏk.

How much the shamanic practices of the Mongols and the Jurchens was a determining factor in their being perceived as barbarian, as well in their women being conscripted as courtesans in China and Korea, remains to be assessed. What is clear is that in both cases a Confucian aristocracy set in motion a class structure in which outcastes provided the human resources for a courtesan system where female performers produced artworks derived from shamanic rituals. The class structure in question is the one based on the so-called "four occupations," or, in Chinese, shì-nóng-gōng-shāng (士農工商). This social arrangement emerged in ancient China and was reproduced in Tokugawa Japan, being adopted also in Yi Korea, albeit in a modified version. In China, the four occupations/classes were, as shown in the concept mentioned above, those of the gentry/scholars (士 shì), the peasants/farmers (農 nóng), the artisans (工 gōng), and the merchants (商 shāng). Several social groups were excluded from these four classes, such as soldiers and guards, the Buddhist clergy, the eunuchs, as well as entertainers, domestic servants, prostitutes, and slaves, which conformed the outcaste groups. In Japan, the system was officially adopted during the Edo period (1603-1868), substituting the two-class structure enforced by the Ritsuryō system, which had come into effect much earlier, during the late Asuka period (538–710) as a product of the Taika Reforms (645 CE). Under the Ritsuryō, the Japanese class system was divided into ryōmin (良民) and senmin (賤民), that is, “good citizens” and “low citizens,” with several subdivisions within each group. The low senmin class included two categories of slaves, the kunuhi (公奴婢), slaves at the service of the court, and the shinuhi (私奴婢), slaves at the service of families.

In Yi Korea, the four classes were adapted into a more rigidly vertical structure than ever was the case in China and Japan. The four classes were called yangbanchunginsangmin and cheonmin, and the high degree of stratification among them reflected the Yi rulers' version of Confucianism. With the rise of the Yi Dynasty and the beginning of the Chosŏn period in 1392, the higher social import of Buddhism verified during the preceding Koryŏ period gave way to a strict form of Confucian hierarchy in which peasants and women suffered a substantial decline in social standing and a marked loss of freedom. In terms of the national arts, the court's assimilation of artworks rooted in shamanism can be verified in the music and symbolic dances favored by the ruling class. This is especially noticeable in the T’aep’yŏngmu (Hanja 太平舞), or Great Peace Dance, which functioned as a court ritual asking the spirits to grant peace for the nation. The T’aep’yŏngmu is a clear example of the Confucianization of shamanic rituals and dances taken into effect to serve the tastes of the Yi rulers. Following the fall of the Koryŏ, the first king of the Chosŏn Dynasty, King T’aejo of Chosŏn (1335–1408), established an aak (雅樂), or “elegant music,” department. The kwan’gi-kisaeng system embodied in the aforementioned The Court Entertainment Bureau (kyobang) was reinforced, and young girls were selected from the such’ŏkcha nomadic outcaste groups based on their dance skills to become government slaves employed at the court. As kwan’gi (官妓), they were given status of cheonmin, which in spite of being the lowest stratum of society was still a recognized social standing within the four class system. This was enough to rescue these girls from the outcaste condition, that is, the ones who were lucky, pretty, or skilled enough to be selected by the court. As the centuries progressed, a great discrepancy emerged between the kisaeng who worked primarily as prostitutes in country taverns and those trained from childhood to work at the palace. In both cases, however, the dances they performed were rooted in the shamanic folk tradition they learned as members of nomadic outcaste groups. The ritual dances marked by trance, ecstasy, and spirit possession that were performed by the mudangs (hangul 무당, hanja 巫堂), the shamans who enacted the kut (굿) rituals, were part of the kisaengs' lives even before they became a kisaeng. Once at court, these rituals were transformed and aestheticized to achieve the solemnity and gravity proper to the official Confucian rites and tastes. The following video brings a performance of the T’aep’yŏngmu in the version established by Han Seongjun (1875–1941), a well-known early twentieth-century dancer and drummer. As a disclaimer please note that the videos uploaded here have the aim of illustrating the artistic practices and developments being discussed, serving also the purpose of cultural diffusion; anyone who feels that his or her rights have been violated by the presence of any of the materials on this website please let me know me through the contact form. Regarding the T’aep’yŏngmu, note in the video below the usage of drumming and dancing patterns that, as will become clear later, are characteristic of shamanic rituals. The following video brings the Mudanchum, a dance that represents the rituals performed by the mudang. The third video is a fragment of the Confucian ritual Munmyo Jeryeak; a comparison between these materials should facilitate the understanding of the aesthetics of Confucianism in connection with that of shamanism:



The T’aep’yŏngmu is designated as one of Korea's national Intangible Cultural Properties (Muhyeong Munhwajae 무형문화재). Korean court dances are called jeongje and pertain to the aak (雅樂) repertory, being part of a system similar to that of Japanese gagaku (雅楽). Just as with gagaku, part of the Korean repertory is divided into pieces imported from the Tang Dynasty (Dangak jeongjae 당악정재) and pieces of native origin (Hyangak jeongjae 향악정재). 



Another video from KBS (Korean Broadcasting System). The Mudangchum is probably the clearest example of the intersection between the world of the female shaman (mudang) and that of the female performer (kisaeng). Note the leading role given to percussion instruments in the music (drums, bells, and chimes). Wind instruments, such as flutes and the piri (hanja: 觱篥), a double-reed oboe made of bamboo are also present. 


Munmyo Jeryeak

Compare the above shamanic-derived dances with the strictly Confucian rite Munmyo Jerye (hanja 文廟祭禮). The ritual music and dance is of ancient Chinese origin and is called Munmyo Jeryeak (hanja 文廟祭禮樂), being also part of the aak repertory.  The word Munmyo is the general Korean term for a temple of Confucius. Korea's principal Confucian temple is the Seonggyungwan Munmyo, located in Seoul.  


The role of women in the strictly Confucian Chosŏn society was one of almost thorough subordination. Similarly to the yìjì (艺妓) of the Tang and Ming China and the Shirabyōshi (白拍子) of Heian Japan, however, the kisaeng had a relatively privileged social position in their country. In spite of being de facto government slaves, they interacted with men from the yangban aristocracy on a daily basis and had more freedom and access to cultural artifacts then the regular wives of the nobles. As was mentioned above, the decline in Buddhist values on the part of the ruling class after the fall of the Koryŏ in the late 14th century led to growing social hierarchization in Korea. Court and folk arts became more and more distanced, and the shamanic practices of the groups perceived as barbaric were more strongly repressed. During the first years of the Chosŏn period, in which the chaos produced by the Mongol invasions of 1231–1259 was still present, relations with the barbaric-shamanist Jurchens became increasingly violent. When the first Yi ruler, Yi T’aejo (1335–1408), incorporated the territory up to the Tumen River frontier into his kingdom, the local Jurchen natives started launching repeated raids against the Koreans, promoting pillage and murder. Jurchen prisoners would add to the slave and outcast strata. Major Yain (Jurchen) rebellions broke out within the boundaries of Chosŏn in 1583 and 1605. With the conversion of the Jurchens into the Manchu people, now Manchu invasions of Chosŏn ensued, with outbreaks in 1627 and 1636. Back in the 15th century, King Sejong (1397–1450), the fourth king of the Yi Dynasty, implemented policies aimed at settling the paekchong in fixed communities. These came in the fifth year of Sejong's rule under the kaech’ing (改稱),  a law ordering that the outcastes, which included the group Tasan referred to as yangsuch’ŏk, be called paekchŏng, be settled in fixed communities, and be registered under the kingdom’s official system. The ghettoization of the paekchŏng resulted in outright failure; spatial segregation meant further stigmatization in comparison to the previous period, turning the paekchŏng into a group of despised outcastes. This could provoke nothing but resentment and social unrest. The paekchŏng became a target for the national laws known as the Kyongguk-taejon, according to which they were limited to reside in only certain areas of the capital and the provinces. Their status in Korea became very similar to that of the burakumin in Japan, and a common origin between the two groups has been suggested by several scholars, most notably Japanese ones such as Imamura Tomo (今村鞆 1870–1943), Imanishi Ryū (今西龍 1875–1932), and Ayukai Fusanoshin (鮎貝房之進 1864–1946), all of whom worked for the Chosŏn Ch’ongdok (朝鮮総督), the seat of the Japanese colonial government in Korea from 1910 to 1945. These official posts granted them a privileged position to carry out research in the peninsula. To them we should add also Kita Sadakichi (喜田貞吉1871–1939), who taught archeology and ethnology at Kyoto Imperial University from 1913 to 1924.

Kita and Imamura were the first scholars to note the existence of the paekchŏng in Korea, perceiving them as the outcaste class to which leather workers and entertainers belonged. Two of their works published between 1918-1919 set the tone for future studies on the subject (for a historical analysis of the development of Korean studies by the Japanese during the occupation period see Nomura Shin’ichi, “Senmin no bunkashi josetsu,” in Akasaka Norio (ed.) Haijo no jikū wo koete (Iwanami Shoten, 2003, p. 161-190). Imamura noted the assimilation policies of the early the Chosŏn dynasty through which the yangsuch’ŏk had been renamed and divided into two classes, the hwabaekchŏng (禾白丁), that is, those who worked in butchering and tanning, and chaebaekchŏng (才白丁), those who engaged in craftwork and entertainment. Imamura also noted the presence of ghettoes inhabited by the paekchŏng, which he termed “paekchŏng buraku” (白丁部落), employing the same toponymical designation pertaining to the Japanese outcastes (See Imamura Tomo, Chosen Fūzoku-shū [A Collection of Chosŏn Customs], Utsuboya Shōsekiten, 1919, p. 47-51). Kita, in an article of 1918 published in Shirin (史林), the journal of the Japanese Society of Historical Research, entitled “Chosŏn no paekchong to waga kugutsushi” [The Chosŏn paekchong and our kugutsushi],  claimed that the Japanese puppeteers kugutsushi (傀儡子) were members of Korea’s paekchŏng class who had emigrated to the Japanese archipelago (Shirin, 3:3, July 1918. For an account of Sadakichi’s argument see Yamauchi Tamihiro, “19 Seiki Chosŏn ni okeru tohan・paekchong shūdan no yaku to sōshiki” [The role of the tohan・paekchong in nineteenth century Chosŏn], Kan Nihonkai Kenkyū Nepō, 22 (2016): 17-29 and Tabata Hiroko, “Komojinja to Kairaishi” [The Komo Shrine and the karaishi], Kumamoto Daigaku Shakai Bunka Kenkyū, 14 (2016): 181-195, p. 192).

The kugutsushi were the puppeteers who worked in the Japanese bunraku theater (文楽, also known as Ningyō jōruri 人形浄瑠璃), living in nomadic groups of itinerant performers called tabigeinin (旅芸人). These groups included female performers who engaged in various activities. They sang and danced pieces derived from shamanist rituals, manipulated puppets as kugutsume (傀儡女), the kugutsushi’s female counterpart, assumed the role of miko (巫女) to perform sacred rituals as female shamans in Shinto ceremonies, and engaged in prostitution, being referred to by the general term of yūjo (遊女). The link between shamanism, the music and dance produced and transmitted by female performers, and social segregation in East Asia becomes clear through an analysis of the tabigeinin and the yūjo social types. The origin and nature of wandering performers has been the subject of extensive debate among Japanese scholars, as well as a riddle for the Japanese people in general. Regarded by many as a mysterious "other" whose origin is unclear in what was otherwise construed as unique and homogenous society, these types became the subject of much speculation and stigmatization, being rendered in literary works as a paradoxical source of attraction and repulsion. The most remarkable of these works is perhaps Kawabata Yasunari's Izu no Odoriko (1926), a short story that offers a lyrical rendering of the tabigeinin in their everyday plight. (For Japanese sources on the theme of the tabigeinin see the works by Okiura Kazuteru (沖浦和光 1927–2015), Nihon Minshū Bunka no Genkyō: Hisabetsuburaku no Minzoku to Geinō [The Homeland of Japanese Folk Culture: The Customs and Arts of the Segregated Buraku] (Hokaishuppansha, 1984) and Tabigeinin no Ita Fūkei: HenrekiRurōTosei [The Tabigeinin’s Lived Landscape: Prilgramage・Vagrancy・Livelihood] (Bunshun Shinsho, 2007).) 

The inquiry on the tabigeinin and the yūjo goes back to the works of Heian-period scholar Ōe no Masafusa (大江匡房 1041–1111), who left two important social treatises on the nomadic troupes that drifted around the outskirts of Kyoto during Heian times, the Kairaishiki (傀儡子記) and the Yūjoki (遊女記). The word Kairaishi in Ōe’s title is an alternative reading for kugutsushi (傀儡子). In the Kairaishiki, Ōe remarked the similarity between the tabigeinin and the group called hokuteki (北狄), that is, “northern barbarians.” The word hokuteki is a Japanese reading of the aforementioned Chinese word Běidí, on of the ethnic slurs used during the Zhou dynasty (c. 1046–256 BCE) to refer to nomadic groups that inhabited areas north of China. In the context of the Japanese archipelago, Ōe’s employment of the word suggests a possible reference to the Emishi people (蝦夷), which were also considered barbaric. In any case, in the Yūjoki, Ōe remarked the kugutsushi and the yūjo’s folk religious beliefs centered around the figure of Hyakudayū (百太夫), a male god worshiped as the travelers’ guardian deity. He also remarked some the of yūjo’s customs, such as sharing meat, which he relates to ancient Chinese laws established by the Han official Chen Ping (陳平d. 178 BC). Ōe further explained that the tabigeinin lived in tents, the men carried bows and arrows, hunted, and worked as jugglers and puppeteers, while the women wore facial makeup, sang and danced in gatherings by river bends, and engaged in prostitution, being frequently taken as mistresses (saishō 妻妾) by noblemen.

The works of Ōe no Masafusa served as the basis for important developments in the field of Japanese Ethnology. The marginal condition of the kugutsushi/yūjo groups generated considerable anxiety among early twentieth-century Japanese ethnologists, who found themselves at loss when confronted with an off-centered element among what they perceived as the singular Japanese people. With the intensification of Japanese imperialism, the question of the tabigeinin became intermingled with nationalist theories of Japanese racial uniqueness and superiority. As nihonjinron theoreticians rummaged around essentialist perceptions and aims, the tabigeinin, together with other Japanese outcaste groups, began being construed as foreign, that is, as being of Korean origin. This is seen in the works of ethnologists such as Yanagita Kunio (柳田 國男 1875–1962) and Takigawa Masajirō (滝川政次郎 1897–1992). Yanagita, who is considered to be the father of Japanese folklore studies, reinforced Kita’s claim of a direct link existing between the Korean yangsuch’ŏk and the Japanese kugutsushi, now suggesting that both descended from Gypsies who would have migrated from India to East Asia. (Yanagita’s Gypsy theory is found in his letters to Minakata Kumagusu (南方熊楠 1867–1941), another important early Japanese ethnologist (see Yanagita Kunio・Minakata Kumagusu Ōfuku Shokanshū  [The Letters of Yanagita Kunio and Minakata Kumagusu, vol. 1] (Heibonsha, 1994), p. 43-47). 

Yanagita's construal of the tabigeinin as foreigners was particularly important in a time when Korea was being colonized and a pristine image of the Japanese people and its imperial lineage had to be guaranteed. Although nihonjinron theories can be regarded as starting as far back as the times of the writers of the Kojiki (8th century CE), passing through Nichiren’s (日蓮1222–1282) thirteenth century claims on the superiority and originality of Japanese Buddhism, as well as through Motoori Norinaga’s (本居宣長1730 –1801) Edo period kokugaku (国学 National study), Yanagita's twentieth-century works, such as Tōno Monogatari (遠野物語, 1910), a volume centered on the folk culture of Iwate prefecture, set the tone for a new type of nihonjinron, one that claimed the existence of a continuity between the family, the village, and the imperial nation state. The construal of the foreignness of all those who did not revere the emperor, a notion that followed from the mythological standpoint of all Japanese being descendants of the same gods, ensued in regard to the tabigeinin, whose aforementioned worship of Hyakudayū (百太夫), a god that does not figure in the Kojiki and is associated with Korean sanshin (Kr. 산신), or mountain worship, made them a target for discrimination and segregation.  

The question of prostitution further complicated the problem of the tabigeinin's citizenship status. The employment of female performers from tabigeinin groups, that is, of yūjo, in Shinto shrines as miko maidens (巫女), or, more literally, as "shaman women," was a problem that produced considerable anxiety among the Japanese rulers since the first years of the Meiji Period (1868-1912), when the new government attempted to reinforce imperial ideology by reinstating Shinto as the national religion. This new national/imperial creed should comply with the aesthetic requirements of solemnity and mythical veracity. The scandal involving Kume Kunitake (久米邦武 1839–1931), a historian who was forced to resign from his professorial position in 1892 after publishing an article suggesting that Shinto was an ancient set of customs based on heaven worship and that the Japanese imperial house's origins were not divine, but human and Korean, illustrates the Meiji political climate. Japan was a country that had been under military dictatorship for more than 700 years, that is, since the end of the aristocratic Heian period in 1192, and although the reform of 1868 meant a political break with the shogunate system it was far from meaning a break with militarism. Kume would soon redress his ways and support the theory called nikkandōikikō (日韓同域考), which expressed the idea that Japan and Korea should share a common territory, which in fact meant that Japan should proceed with the annexation of Korea (regarding Kume's somewhat odd claims that together "Japan and Korea Constitute the Divine Country of Japan" see the volume edited by Ōkubo Toshiaki (大久保利謙 1900-1995) Kume Kunitake no Kenkyū (Tokyo: Yoshikawa Kōbunkan, 1991). 

The supporters of nikkandōikikō used various arguments to justify Korea's annexation. The works of comparative linguist Kanazawa Shōzaburō (金沢庄三郎 1872-1967) Nikkan Ryōkokugo Dōkeiron (日韓両国語同系論 1910) and Nissen Dōsoron (日鮮同祖論 1929), for instance, which asserted a common linguistic root and a common ethnic ancestry existing between the Japanese and the Korean languages and peoples, were used as rationale for the occupation; the idea of a single ethnicity seemed to validate the attempt at cultural assimilation and political subjugation. Note that although its use in support of Japanese expansionism is obveiously far from warranted, recent studies in ethnolinguistics by Western scholars have confirmed Kanazawa's early twentieth century analysis; the Japanese and Korean languages' have a common Tungusic root, from which a Proto-Korean-Japanese language is seen as having existed in the area of the ancient Koguryŏ kingdom. The proto-language has been traced primarily by means of toponymic analysis (see Christopher Beckwith, Koguryo: The Language of Japan's Continental Relatives: An Introduction to the Historical-Comparative Study of the Japanese-Koguryoic Languages (Brill, 2007); J. Marshall Unger, The Role of Contact in the Origins of the Japanese and Korean Languages (University of Hawaii Press, 2008); and Alexander Takenobu Francis-Ratte, Proto-Korean-Japanese: A New Reconstruction of the Common Origin of the Japanese and Korean Languages (PhD Dissertation, The Ohio State University, 2016).    

A second strategy to validate the annexation was to attempt to demonstrate the inferiority of Koreans. This would justify the latter's submission to the Japanese, who would be entitled to guide them into progress and development within the aims of the Greater East Asia Co-Prosperity Sphere. The works of ethnologists at the service of the Japanese colonial government would provide ideological support to this second strategy. Scholars such as Shiratori Kurakichi (白鳥庫吉 1865–1942), Ikeuchi Hiroshi (池内宏 1878–1952), and Ayukai Fusanoshin (鮎貝房之進 1864–1946)  drew on the ancient Chinese views that regarded the nomadic inhabitants of the north as barbaric to assert the equally primitive nature of the peninsula's contemporary inhabitants. Ayukai focused on the paekchŏng and on the Korean slave class to identify Korean society as decadent (see Ayukai Fusanoshin, Hwarang-kōPaekchŏng-kōNuhi-kō [Considerations on the hwarang, the paekchong and the nuhi] (Tokyo: Kokushōkankōkai, 1938), especially pages 222-25 and 350-365) More broadly, Koreans were identified as descendants of the aforementioned Tongye (Ch. Dōngyí 東夷), that is, the nomadic practitioners of shamanism that the Chinese distinguished through the use of ethnic slurs. The emphasis was on the Tongye being the ancestors of the peoples of Puyŏ, Koguryŏ, and Paekche, and consequently of the Jurchens and Koreans, including their slaves and outcastes. Ayukai was especially keen in asserting the single identity of these groups by noting that they were all Tungusic peoples (通古斯). All the while, the Tungusic origin of the Japanese themselves (via the Yayoi) was intentionally ignored and the construal of Japanese minorities (namely, eta, yūjo, sanka, burakumin, etc.) at home as foreign (that is, Korean) guaranteed the maintenance of the idea of Japanese ethnic singularity. 

The Chōsen Sōtokufu (朝鮮総督府) scholars' construal of Koreans as barbaric conflicted with the understanding of the Korean origin of the imperial lineage, making any assertions on the subject intolerable for the increasingly powerful members of the Japanese far-right. Repeating the early-Meiji incident involving Kume Kunitake, now Tsuda Sōkichi (津田左右吉 1873-1961), a professor of ancient history at Waseda University, became the target of censorship, being convicted in 1940 and imprisoned in 1942. In this context, Yanagita Kunio's studies in folklore contributed to deflect the more pressing inquiry on ancient history that involved the question of the early human migrations into the archipelago and the power struggles that led to the end of the Yamato period. This inquiry would reveal the peninsular or Siberian origin of the Yayoi and the conquest of these by a more technologically advanced tribe that brought material culture into the islands and allowed the inflow of various ethnic groups, which were subsumed under the concept of Toraijin (渡来人). This inquiry had to be inhibited, for it would reveal the fallacy involved in the myth of Japanese uniqueness derived from the imperial narratives of the Kojiki and the Nihonshoki, thus unveiling the untenability of the belief in the divine nature of the emperor.  

Such historical inquiry became possible only after the end of the war. In 1948, a group of scholars formed by Egami Namio (江上波夫 1906–2002), Oka Masao (岡正雄 1898–1982), and Ishida Ei’ichirō (石田英一郎 1903–1968) began the study of the early Kofun period and one year later Egami's kibaminzokuseifukuōchōsetsu (騎馬民族征服王朝説), know in the West as the "horse-rider theory," appeared in the article “Nihon minzoku = bunka no genryū to Nihon kokka no keisei” (日本民族=文化の源流と日本国家の形成), published in the journal Minzokugaku Kenkyū (民族学研究). Egami's inquiry became know in the West through the work of Gari Ledyard (see “Galloping along with the Horseriders: Looking for the Founders of Japan,” The Journal of Japanese Studies 1.2 (1975): 217-254) and Walter Edwards (see "Event and Process in the Founding of Japan: The Horserider Theory in Archeological Perspective," The Journal of Japanese Studies 9.2 (1983): 265-295). Another work that confirms the peninsular origin of those who established the earliest Japanese polity is Wontack Hong's Paekche of Korea and the Origin of Yamato Japan (Kudara International, 1994). The subject of these works is still very politicized and Egami's work is widely discredited in Japan. In the same vein, Wontack Hong's work is mostly (and intentionally) ignored in Japan and not a few Western scholars who side with the Japanese political agenda have criticized and attempted to discredit it.   

Notwithstanding, the theme of Shinto continues to intersect with that of shamanism. Most scholars, however, tend to overlook that intersection. Among Western analysts, John Breen and Mark Teeuwen, for instance, recognize that “Shinto, as the distinct, autonomous and independent religion we know today is an invention of nineteenth century Japanese ideologues” (see Shinto in History: Ways of the Kami (Routledge, 2000, p. 4), but rely on Kuroda Toshio's views to emphasize the Taoist element of Shinto, overlooking almost completely the Japanese religion's actual roots in shamanism as well as its deep connection with Korean folk religion (in “Shinto in the History of Japanese Religion” (The Journal of Japanese Studies 7.1 (1981): 1-21), Kuroda practically dismisses Korea in his analysis of Shinto, mentioning the country only once). In A New History of Shinto (Wiley-Blackwell, 2010), Breen and Teeuwen overlook the shamanic nature of the miko shrine maiden, perceiving it merely as "type of shaman" without noting the fact that mikos and mudangs are one and the same religious figure and, what is more, Shinto is primarily a Korean religion. 

The original identity between the Japanese miko (巫女) and the Korean mudang (巫堂) becomes clear when two important aspects are considered: the religious meanings they embody and the artistic practices they reproduce. The miko-mai (巫舞) shamanic dances performed in the context of Shinto rituals mirror the mudang’s mudangchum (巫堂舞)  in the context of musok (巫俗). The spirits and gods embodied and worshiped in both cases were originally alike, the difference having emerged only when the figure of the miko was adapted to serve the political aims of Japanese nation-building, with her gods and spirits being renamed accordingly. The miko’s absorption into the Shinto construct was achieved through a process of acculturation initiated at the beginning of the male-centered Kofun period and concluded by the Meiji nationalists, one through which the miko and her gods were transfigured to accommodate, both in the State and folk versions, a set of deities organized around the central figure of the emperor. 

The fundamental link between the miko and the mudang is becomes clear when the social stratum from which both types emerge is considered. In the landmark work Nihon Miko-shi (日本巫女史 History of the Japanese Miko, 1930) ethnologist Nakayama Tarō (中山太郎 1876–1947) demonstrated that the shamanic dances performed by the miko are one and the same with those of the yūjo, the wandering performer/prostitute who travelled the country together with the tabigeinin nomadic troupes. In chapter six, “Kabu ongaku no hozonsha to site no miko” (p. 574-84), Nakayama further develops his argument by showing how the words miko, yūjo, and kugutsume are nothing but different names for the same object (異称一体 Ishōittai). In chapter nine, “Seiteki Shokugyōfu to kashita miko no matsuro” (p. 729-739), he continues to show how prostitution became the miko's inevitable faith, further emphasizing the identity between the miko and the yūjo. More than three decades after the appearance of Nakayama's work, Takigawa Masajirō's (瀧川 政次郎 1897-1992) Yūjo no Rekishi (遊女の歴史 History of the Yūjo, 1965) proposed a reexamination of the subject. Takigawa argued that the yūjo type (or, as he called it, the asobi 遊 tradition) originated among female entertainers of slave status in China and was transmitted to Japan through Korea. Takigawa’s thesis drew on Yanagita Kunio’s previous analysis, which, as we have seen, reinforced Kita Sadamichi’s claim of a direct link existing between the Korean yangsuch’ŏk/paekchŏng and the Japanese kugutsushi/tabigeinin, as well as on Ōe no Masafusa’s remarks, seen above, regarding the yūjo’s worship of Hyakudayū (百大夫). Hyakudayū, according to Takigawa, is a god originated in Daoist lore, being later syncretized into Japanese Shinto as the traveler’s guardian deity (道祖神), appearing then in several subordinate shrines around Western Japan. Ōe would have overlooked such syncretism, and the Hyakudayū would be, in fact, a foreign god. From this, Takigawa argued that since Hyakudayū does not figure in the pantheon of Japanese deities, but is related to, as noted above, Korean sanshin, the yūjo type must be of Korean origin.

Takigawa also adduced as proof of the yūjo’s Korean origin the fact that the very word yūjo (遊女) did not exist in ancient Japanese language, being imported from China and Korea (Yūjo no Rekishi, p. 18). The assertion is unsustainable because words such as the saburuko (左夫流児) and ukareme (遊行女婦) existed of old to signify the same female type. A second point raised by Takigawa was that, differently from the case of gagaku court music, where Japanese musicians travelled to China and Korea, learned music pieces, and then brought them back to Japan, the same path was unthinkable in the case of the folk arts performed by the tabigeinin, these being arts that are not transmitted in the same manner, having thus being brought into the archipelago by the original performers themselves (Yūjo no Rekishi, p. 118). Takigawa’s claim on gagaku’s transmission is also unsubstantiated, the modes of diffusion of court continental and peninsular music into Japan are not as fully elucidated as assumes Takigawa (see, for instance, Kikkawa Eishi, Nihon Ongaku no Rekishi [History of Japanese Music] (Tokyo: Sogensha, 1965), p. 32-37). Furthermore, musical exchange between the peninsula and the archipelago does not presupposes the sort of migration suggested.

Scholars such as Amino Yoshihiko (網野善彦 1928–2004) and Fukutō Sanae (服藤早苗 b. 1947) criticized Takigawa’s theory as ethnocentric, biased against women, and expressive of an ingrained prejudice against Korea. (See, for instance, Amino’s Chūsei no hinin to yūjo [Medieval outcastes and prostitutes] (Akashishoten, 1994) Futukō's edited volume Kegare no bunkashi: monogatari, jendā, girei [The Cultural History of Kegare: Tales, Gender, and Rituals] (Shinwasha, 2005).) Takigawa’s analysis, together with those of Yanagita and Kita, is problematic because it fails to distinguish between the notions of ethnicity and nationality. More precisely, it fails to consider that even if the yūjo/tabigeinin/outcaste group crossed into the archipelago from the peninsula, so did originally the Yayoi, and consequently so did the vast majority of the Japanese people’s ancestors. This alone makes it virtually impossible to demarcate clearly around national frontiers ancient practices, beliefs, and peoples whose presence in the Japanese archipelago seem to pre-date the advent of a local state. To put it more bluntly, as long as analyses such as Takigawa’s continue to be based on the questionable premise of Koreans being an ethnic group distinct from the Japanese and proceed to identify populations that have inhabited the archipelago from old as Korean, they will remain susceptible to the charge of ethnocentrism. What is more, as long as the misguided habit of  subsuming ethnicity and nationality under a single idea, together with the lack of the concept of a multi-ethnic society, prevails in East Asia, fruitless analyses, social alienation, and injustice will continue to prevail in the area. In sum, both Takigawa’s and his critics fail to perceive that yūjo, paekchŏng, Tongye, Yemaek, Yayoi, Korean, and Japanese are different names used to define what is originally a single ethnic group, a group that originated in the Siberian Tungus, that shared a common language in the past, and that practiced common religious rituals, namely those defined by the term shamanism. 

The problem of social segregation and the emergence of outcaste systems in East Asia must thus be seen as resulting from a clash, or a series of clashes, between that original ethnic group and some external force that gradually took hold of it, creating a fissure within the group. This clash was concomitant with the process of state formation in the area. In the case of Japan, it was connected to the rise of an imperial lineage that took territorial control over part of the archipelago, bringing, I argue, Confucianism into contact with shamanism. The pragmatic nature of the former was incompatible with the other-worldly character of the latter, making the clash inevitable. In the process, boundaries between the acculturated and the non-acculturated established a fierce opposition between insiders and outsiders. The  outcaste system thus emerged from the consolidation of a peculiar pre-modern state supported by an ideology of racial purity in which the other was to be rejected at all costs.

The symbiotic relation between the imperial and the outcaste system in Japan was noted by Okiura Kazuteru (沖浦和光 1927–2015) in Tennō no kuni, senmin no kuni: ryōkyoku no tabū [Country of the Emperor, Country of the Outcaste: The Double-Edged Taboo] Tōkyō: Kōbundō, 1990). Okiura sees the Japanese imperial ideology as emerging from the Kojiki (c. 712 AD) and the Nihon Shoki (c. 720 AD), which he contrasts with the Chinese notion of zenjō hōbotsu (禅譲放伐), the practice of the abdication by an emperor in favor of a more virtuous successor. The imperial system and the existence of an outcaste group in Japan would have emerged as two faces of the same coin: the notion of the eternity and purity of the imperial bloodline would define itself negatively in relation to the impure outcaste, which was construed as the other who does not venerate the national god (namely, the emperor). The Japanese traditional arts, still in Okiura’s analysis, sprang from the social dynamic of the class system imposed by the imperial ideology, and those existing at the margins of that class system became responsible for upholding and transmitting the nation’s traditional artistic practices.

Okiura’s analysis is particularly effective in noting the imperial system’s materialization of the outcaste as the other who conflicts with the state narrative and becomes the carrier and transmitter of artistic forms. It fails, however, to contextualize this development within a given historical process and to inquire on the actual reasons for the outcastes’ becoming the keepers of the arts. In other words, Okiura overlooks the motives underlying the process that he correctly identifies. These motives are part and parcel of the conflicting worldviews informing the power struggles that resulted in the described political configuration. Above all, religious factors must be considered for a deeper understanding of the dialectic involved. One fundamental point to be considered is that by the time of the writing of the Kojiki and the Nihonshoki early in the 8th century, Confucianism had already existed as a powerful system of thought in East Asia for more than a thousand years. The Han Commanderies, even if not capable of establishing Confucianism as an official system throughout the entire peninsula, must have became an emanating source of Confucian philosophy all the way to the archipelago during their more than 300 years of existence. The Japanese imperial ideology must thus be closely analysed and its roots must be sought. Considering the dominant systems of thought identifiable as being available in the area at the time of the emergence of a powerful male-centered imperial lineage and speculating on which of these systems would have provided ideological support for that emergence, Confucianism appears as the strongest candidate. By the time Buddhism became favored by the aristocracy during the Asuka period, that male lineage had already been established. From this we can aduce two points that may serve to expand Okiura’s analysis: the first is that originally the imperial ideology appears as the product of a transfiguration of Confucian values enacted for the benefit of male-centered dynastic interests; the second is that the arts Okiura mentions as having become the preserve of the outcaste group are nothing but shamanic rituals adapted to performance. The question of the yūjo outcastes being Korean, then, becomes superseded by that of their being shamans.

More than being dismissed as simply wrong, Takigawa’s analysis thus deserves the criticism because it construes a population that is likely to have inhabited in the area since before the emergence of the Japanese State as foreign. His perception of the yūjo and her outcaste group’s worship of Hakudayū is, in any case, warranted by simple observation. The yūjo’s worship of the traveller’s god became especially noticeable during the Edo period, when the term dayū (太夫) came to signify a ranking for high level prostitutes working in the brothel districts of major urban centers, such as Edo’s Yoshiwara and Kyoto’s Shimabara. On this subject see Saeki Junko, Yūjo no bunkashi [Cultural History of the Yūjo] (Chūō Kōronsha, 1987), p. 87-94; Nagai Oshio, Yoshiwara Jiten [Yoshiwara Lexicon] (Asahi Shingun Shuppan, 2015), p. 331-334; and Horie Hiroki, Sandaiyūkaku: Edo YoshiwaraKyoto ShimabaraOsaka Shinmachi [The three major brothel districts: Edo Yoshiwara, Kyoto Shimabara, Osaka Shinmachi] (Gentosha, 2016).

The construal of the yūjo as a foreigner on the basis of Hakudayū worship, however, is especially undesirable for what it entails in terms of citizenship alienation, spatial segregation, and social exclusion. This connects with the yūjo's burakumin origin (as well with that of her Edo period counterpart, the geisha 芸者), which is well-documented (see for instance, Okiura Kazuteru's Nihon Minshū Bunka no Genkyō: Hisabetsuburaku no Minzoku to Geinō [The Homeland of Japanese Folk Culture: The Customs and Arts of the Segregated Buraku] (Hokaishuppansha, 1984) and Tabigeinin no Ita Fūkei: Henreki・Rurō・Tosei [The Tabigeinin’s Lived Landscape: Prilgramage・Vagrancy・Livelihood] (Bunshun Shinsho, 2007). Notwithstanding, Hyakudayū does emerge as a special case among deities worshipped in Japan due to the absence of clear records regarding its emergence in the archipelago or elsewhere. In this sense, Hyakudayū differs from deities such as Shiramyōjin (新羅明神) and Matarajin (摩多羅神), for instance, which were brought in from China and Korea and were adopted in Japanese shrines through a process of religious syncretism, instantiating the phenomenon of shinbutsu-shūgō (神仏習合), that is, the merging of Buddhism and Shintoism. (In this regardsee Yamamoto Hiroko, Igami: Chūsei Nihon no Hikkyōteki Seikai [Foreign God: The World of Medieval Japan’s Esoterism] (Heibonsha, 1998) and Kawamura Minato, Yami no Matarajin: Hengen suru igami no nazo wo ou [Matarajin’s Darkness: An Inquiry on the Mystery of a Foreign God] (Kawadeshobō, 2008). On syncretism and the religious beliefs of the tabigeinin see Hattori Yukio (服部幸雄 1932–2007), Shukushinron: Nihon Geinōmin Shinkō no Kenkyū [A Theory on the Shukushin God: Studies on Japanese Traditional Performers’ Beliefs] (Iwanami Shoten, 2009).)

The mystery of Hyakudayū has been considered at length by Maeda Hayao (前田速夫 b. 1944), who suggested that the god, who is referred to both as Hakudayū or Shiradayū (白太夫), is the same deity venerated in Hakusanshinkō (白山信仰), a folk religion based on mountain worship and centered around the cult of Hakusan (白山), one of the three sacred mountains of Japan. Among Maeda’s works on the subject the most relevant are Haku no Minzokugaku e: Hakusanshinkō no Nazo wo Otte [Towards an Ethnology of the Haku: An Inquiry on the Mystery of the Hakusan Faith] (Kawade Shobō Shinsha, 2006); Hakusanshinkō no Nazo to Hisabetsuburaku [The Mystery of the Hakusan Faith and the Discriminated Buraku] (Kawade Shobō Shinsha, 2013); and Umi wo Watatta Hakusanshinkō [The Hakusan Faith’s Ocean-crossing] (Gendai Shokan, 2013). The three sacred mountains, or Sanreizan (三霊山), are Mount Fuji (富士山), Mount Haku (白山), and Mount Tate (立山). Hakusan is located on the borders of Gifu, Fukui, and Ishikawa prefectures. Note that the sacred mountains are part of a polytheist system, housing various Shinto gods.

Maeda asserts that the Hakusanshinkō cult can be traced back to the Eurasian continent and to the abovementioned Tongye (東濊) people, the Tungusic group that occupied the northeastern part of the Korean peninsula from roughly the 300 BCE to 600 CE. Together with the Puyŏ (夫餘), Koguryŏ (高句麗), Okchŏ (沃沮), and Yemaek (濊貊), the Tongye are considered to be in direct ancestral line to the Korean people. In Umi wo Watatta Hakusanshinkō [The Hakusan Faith’s Sea-crossing] (2013), the author suggests that Tongye’s migrated from the area of today’s North Korea’s Kangwŏn province towards the south of the peninsula and subsequently crossed into the archipelago. The migration would have resulted from a military defeat at the hands of the Koguryŏ clan. The Hakusanshinkō cult would thus have made its way to Japan in the hands of a foreign population. 

(On the Yemaek’s establishment of the confederated kingdoms of Kojsŏn, Puyŏ, and Koguryŏ, as well as the Tongye’s assimilation by Koguryŏ during the first century of the common era see Jinwung Kim, A History of Korea: From “Land of the Morning Calm” to States in Conflict (Indiana University Press, 2012), p. 23-27. See also Robert E. Buswell Jr., Tracing Back the Radiance: Chinul’s Korean Way of Zen (University of Hawaii Press, 1991, p. 3), where the author describes the Koguryŏ clan’s conquest of the tribal leagues of Okchŏ, Tongye, and finally the remaining Puyŏ tribes in central Manchuria. For a critical examination of the idea that ethnic group responsible for the formation of Tan’gun’s kingdom of Kojosŏn (2333 BC–108 BC) the Yemaek are the first “Koreans” see Pai Hyung-Il, Constructing “Korean” Origins: A Critical Review of Archaeology, Historiography, and Racial Myth in Korean State-formation Theories (Harvard University Press, 2000).)

Maeda’s linking the Tongye to the Japanese marginalized groups of nomadic performers seems to advance through a pathway similar to the one seen in Takigawa’s analysis. Here, however, the connection is drawn between the Hakusan cult and the mountain dwellers of central Japan, the same groups that had already been the subject of Yanagita Kunio’s research under the concept of yamabito (山人). A comparison between Maeda and Yanagita’s views evinces the multiplicity of perceptions available to the Japanese in regard to the off-centered populations that inhabited the country’s mountainous rural areas. Yanagita’s conceptualization of the yamabito contrasts sharply with Maeda’s in that it overlooks important religious components surfacing in the group’s engagement with the performing arts. By downplaying ritual in favor of what he believed was a discernible oral-literary tradition, Yanagita produced an idealized version of the rural dewellers as a repository of tradition and as a fundamental part of the national culture.

Yanagita divided the yamabito in hyōhakumin (漂泊民), or wandering people, and jōmin (常民), or settled people. While the latter lived in fixed communities and possessed land, the former followed the courses of rivers, migrating according to the seasons and avoiding contact with mainstream Japanese society. Being of unknown origin, these mountain dwellers were referred to as sanka (山窩), a term whose etymology is unclear, but whose rendering in Chinese characters suggests the meaning of “mountain-cave people.” The true nature of Japan would be found in the sanka’s oral folk tales, a repository of lost tradition that would evince the true and unique Nihonjin character. As Hashimoto Mitsuru pointed out, “Yanagita believed that modernity had destroyed the true spirit of the common Japanese people, and that only those who had taken refuge deep in the mountains could keep this spirit alive” (Hashimoto Mitsuru, “Chihō: Yanagita Kunio’s ‘Japan’,” in Stephen Vlastos (ed.) Mirror of Modernity: Invented Traditions of Modern Japan (University of California Press, 1998), p. 135.). Drawing on his own nativist sentiments, Yanagita sought in the sanka the original values and sensibilities that modern Japan had forgotten in the course of its cultural engagement with the West.

(For other Japanese sources on the sanka see, for instance, Okiura Kazuteru (沖浦和光 1927–2015), Maboroshi no Hyōhakumim: Sanka [The Wanderers of Illusion: Sanka] (Bungei Shunjū, 2004) and Misumi Hiroshi (三角寛 1903–1971) Aiyoku no Seburi: Sanka no Kidan [The Tent of Passion: The Sanka Tales] (Tokyo Shobō, 1956). The sanka figured in several works of literature and cinema; see for instance the film Sebori Monogatari (瀬降り物語 Sanka Tales; Tōei 1985). Note that without clear lines demarcating the outcaste groups, these marginalized minorities were subsumed under the general term burakumin (部落民), or literally “village dwellers.” With the emergence of policies in the Edo period aimed at restricting these populations in fixed communities, attempts at differentiation surfaced through designations such as etakei (穢多系), hininkei (非人系), or sankakei (山窩系) or hisabetsuburaku (被差別部落 “discriminated villages”). In this regard see Tanigawa Ken’ichi, Gensen no Shikō: Tanigawa Ken’ichi Taidanshū [The Thinking of Sources: Tanigawa Ken’ichi Dialogues] (Tomiyamabō International, 2008), p. 378-379.)

In looking for stories and folk tales among the sanka, Yanagita overlooked the fundamentally shamanist nature of their culture, which revealed itself in rituals of mountain worship and the performing arts these gave birth to. He failed to observe also that the Hakusan cult and Hyakudayū worship they practiced were incompatible with the rituals of national Shinto. Yanagita’s view contrasts sharply with the one advanced by Maeda, who preferred to draw a direct line between the foreign Tongye and the Japanese nomadic mountain dwellers. This line, however, is argued primarly on the basis of the coincidence of mountain worship in pre-modern Korea and Japan, as well as through recourse to etymologies of Chinese words. Once again, the shamanic, or Tungusic, roots of the group’s religious practices are not taken into consideration.

Much more than a transmission occurring between distinct ethnic groups inhabiting fully formed polities, the shamanic link between the religions of the peninsula and the archipelago calls for an inquiry on the common origin between the groups in question. In this sense, both Maeda’s and Yanagita’s concoctions of foreignness and nationalness in regard to the segregated mountain dwellers of Japan fail by overlooking the Tungusic common cultural and ethnic basis shared by the inhabitants of the coutries that were formed on the peninsula and the archipelago.

Another important work on the subject of the sanka is Tsutsui Isao’s (筒井功b. 1944) Sanka no kigen: Kugutsu no hassei kara chōsen hantō e [The Origin of the Sanka: From the Emergence of the Kugutsu to the Korean Peninsula] (Kawadeshobō Shinsha, 2012), proceeds to assert boldly the identity between the sanka and the tabigeinin outcastes. Tsutsui expounds his main thesis in chapter 10, under the heading “Sanka, Kugutsu, Yangsuch’ŏk,” asserting that the sanka’s origin is to be found is Korea’s yangsuch’ŏk/paekchŏng, and that these are nothing other than the kugutsushi.

One point in common between the sanka and the yangsuch’ŏk/paekchŏng seems to lie in the habit of eating dog meat, a practice that might be rooted in shamanic rituals of animal sacrifice and that could be one of reasons for the outcastes’ original exclusion from general society. (On this subject see, in regard to the sanka, Jun’ichi Saga, Memories of Silk and Straw: A Self-portrait of Small Town Japan (Kodansha International, 1987, transl. Gary O. Evans), p. 51, an in regard to the paekchŏng Herbert Passin, The Paekchong of Korea. A Brief Social History,” Monumenta Nipponica 12: 3/4 (1956-1957):195-240, p. 209.)

Still a distinct theory regarding the sanka’s origin relies on Shinto lore and on the legend of Kuririhime no Kami (菊理媛神), also called Shirayamahime no Kami (白山比咩神), a god that is also worshiped in Hakusan shrines. The theory sustains that as a mountain goddess, Kuririhime would, in metamorphosed version, be the actual deity worshiped the persecuted tribes referred to as tsuchigumo (土蜘蛛), a derogatory term used to designate renegade clans in ancient Japan that did not come under the rule of the imperial line. Yoshie Akiko has argued that these persecuted clans appear to have followed matrilineal lines, with great power conceded to women within each group, hence the notion of tsuchigumoyasome (土蜘蛛八十女), or “the many female tsuchigumo.” From this perspective, the figure of Empress Himiko could be interpreted as that of one of the many female rulers that controlled the mutually opposing tribes of Japan’s proto-history. In this regard, see Yoshie Akiko (義江明子b. 1948), Kodaijosei-shi e no Shotai: “Imo no Chirakara” wo Koete [Introduction to the History of Ancient Women: Transcending the “Power of the Beloved”] (Yoshikawa Kōbunkan, 2004), p. 76. Note that research on the sanka started with Yanagita Kunio’s Itaka oyobi Sanka [Itaka and Sanka] (found in Teihon Yanagita Kunio Shū 4, Chikuma Shobō, 1963). For Yanagita’s analysis of the relation between the tsuchigumo and hakusanshinkō see Hitotsume-kozō and Others (Kadokawa Shoten, 1954). Hitotsume-kozō are fallen mountain gods, which include Shirayama-hime (白山比咩神), the goddess of Mount Hakusan. On the tsuchigumo and Hyakudayū see Kita Sadakichi, Tenaga to Tarinaga: Tsuchigumo no Kenkyū [Tenaga and Ashinaga: The Study of Tsuchigumo], (Heibonsha, 1982). The possibility of the practice of shamanism being at the core of these clans’ disjunctive position within the overall political structure of ancient Japan is congruent with my claim that a clash between shamanism and a more powerful ideology, which I suggest is strongly connected with Confucianism, can in fact be discerned and that this clash is the starting point of the outcaste groups’ emergence.

Be it at it may, going back to Tsutui’s analysis, we find not only the previously mentioned assumption that in order to escape slavery in Korea the yangsuch’ŏk emigrated to Japan, but also a link between, on one side, Tasan’s perception of the yangsuch’ŏk as the original group from which of the kisaeng (or kwangi) emerged, and, on the other, the notion that of sanka (originally Korean yangsuch’ŏk) as the source of the yūjo, on the other. In other words, the Korean underclass would be the foundation of both the Korean kisaeng and the Japanese yūjo.

Tsutsui claims to have found the decisive answer to the sanka’s mystery, and by extension to that of the origin of the yūjo/tabigeinin outcastes. Alternative views, however, subsist under the aforementioned possibility of the sanka’s origin in clans that would have resisted the control of the Yamato rulers during the formative stages of the Japanese polity. These clans, as we suggested above, would have been militarily defeated and subsequently persecuted, thus forming nomadic groups that inhabited the mountains.

In any case, it is worth reiterating that the problem with interpreting groups such as the sanka as foreign is that it may serve as justification for the its social exclusion and alienation from citizenship rights. This was precisely the case with the sanka, a group that existed in isolation from mainstream society until at least the decade of 1950 and that in doing so was prevented from having a koseki (戸籍), that is, a Japanese family registry that serves as certificate of citizenship. A similar problem ocurred in regard to Koreans in Japan after the end of World War II, when Korean residents who had gained Japanese citizenship during the colonial period had that citizenship revoked and were registered under the concocted nationality called Chōsen-seki (朝鮮籍), which in practice made them stateless.

During the colonial period, Koreans became part of the Empire of Japan as a result of the Japan–Korea Annexation Treaty of 1910. This was a time when the female performers of both countries would converge in Seoul, the seat of the Japanese authority in Korea. The arts rooted in shamanism transmitted by those women had become associated with prostitution in East Asia since ancient times, with, as mentioned above, yìjì (艺妓), kisaeng (妓生), and geishas (芸者) crystalizing respectively in China, Korea, and Japan as the national versions of a similar type. Now with Korea under Japanese rule, geishas would find work in Seoul and sometimes work side by side with kisaeng.

Much has been written about the origin of the Korean female entertainer. As we have seen, it was at the beginning of the Koryŏ Dynasty (918–1392) that these women appeared with a recognizable public function, that is, as dancers and entertainers employed at the service of the court and assigned to the Court Entertainment Bureau (kyobang). The tradition of female entertainers, however, existed prior to the establishement of the kyobang system during the Koryŏ, and the actual process through which this social type emerged gave room to much theorization.

Yi Nŭnghwa, for instance, in Chosŏnhaeŏhwasa (朝鮮解語花史, 1926), suggested two possibilities for the origin of the female court attendant type. The first was they could have emerged from the wŏnhwa (源花), a legendary class of cadets led by women and established during the Silla dynasty (57 BC–935 AD) in the reign of King Chinhŭng (眞興王526–576). After being disbanded as a militia group, the wŏnhwa would have been absorbed by the court as female attendants. Contemporary scholars such as Hwang Wŏn-gap, Kawamura Minato, and Pak Sun-I, however, have rejected this possibility as a romantic rendering of an otherwise much harsher reality. Hwang follows Tasan’s previous analysis and recalls the establishment of the kijŏk (鼓籍), a government record of all the kisaeng working in a particular district, as a means to control the yangsuch’ŏk population. Established during the reign of King Myŏngjong of Koryŏ  (明宗1131–1202) by Kyŏngju chieftain Yi Jiyŏng (李至榮, d. 1196), the kijŏk aimed at enhancing government control over the marginal population. The noble origin of the wŏnhwa would be incongruent with the kisaeng’s low social standing (see Hwang Wŏn-gap. Han’guksarŭl pakkun yŏindŭl [The women who changed Korean history] (Sŏul: Ch’aegiinnŭnmaŭl), 1997, p. 449; Kawamura Minato. Kīsen: “Mono o iu hana” no bunkashi [Kisaeng: A cultural history of “the flower that speaks”] (Tokyo: Sakuhinsha, 2001), p. 25-28; and Pak Sun-I, Ajia bunka to bungakushisō―nikkan hikaku no kanten kara [Asian Culture and Literary Thought: A Japanese-Korean Comparative Perspective] (Tokyo: Bushindo), 2010, p. 73.)

Yi’s second theory places the origin of the kisaeng class in the former Paekche population that was subjugated under the kingdom of Silla after the latter’s annihilation of Paekche in 660 AD.  This theory would concur with Tasan’s perception of the kisaeng’s origin in the yangsuch’ŏk, supposing, of course, that the latter is comprised of Paekche captives. Be it as it may, what interests us here is to note that if, as Nakayama Tarō suggested, the miko and the yūjo were one and the same social type, the same could be said of the mudang and the kisaeng, the link between the latter two types being the yangsuch’ŏk outcaste group from which both female shamans and female courtesans emerge. In both the Korean and Japanese cases, the link between female shamanism and courtesan life appears to be verifiable.

The kisaeng’s employment at the Chosŏn court was interrupted with the fall of the Yi Dynasty in 1897. The female performers who theretofore had served at the court suddenly faced a new predicament, and brothel life appeared as the most immediate option for survival. In 1902 Korea’s first red-light district was established in Pusan by a Japanese businessman, and on September 25, 1908, the Japanese administration issued the Kisaeng Regulation Order (Jp. Kishōdansokurei, Kr. Kisaeng Tansongnyŏng 妓生団束令), through which kisaeng were registered with the protectorate’s administration and obliged to form a union (see Heo Yeonhee, “Kiseng Education at Kwonbon in Korea from 1908 to 1942,” in Choreologia, Journal of the Japanese Society for Dance Research, Vol. 2, No. 31, pp. 48-59, p. 49. (許娼姫, 「韓国券番 (1908-1942) に おける妓生教育–妓 生教育の内容と舞踊教育–」舞踊學). As Hong Seong-cheol pointed out, starting on that date individuals wanting to become kisaeng or to work in prostitution houses were obliged to obtain their parents’ or relatives' permission and then submit a written authorization to the district police superintendent (Hong, Seong-cheol. Yukake yoksa (Peiporodu, 2007), p. 75). A system of licensed prostitution emerged with the establishment of kwŏnbŏn, the Korean version of the Japanese kenban (券番), which functioned as the geisha registry and school (Soh, Sarah. The Comfort Women: Sexual Violence and Postcolonial Memory in Korea and Japan (University of Chicago Press, 2009), p. 8).

Heo Yeonhee lists seven different kwŏnbŏn operating in Korea: Chosŏn Kwŏnbŏn, at the capital (朝鮮券番, 京城); Giseong Kwŏnbŏn, in Pyongyang (箕城券番, 平壌); Jinju Kwŏnbŏn, in Jinju (晋州券番, 晋州); Dalseon Kwŏnbŏn, in Taegu (達城券番, 大邸); Gwangju Kwŏnbŏn, in Gwangju (光州券番, 光州); Syowa Kwŏnbŏn, in Gunsan (昭和券番, 郡山); and Dongnae Kwŏnbŏn, in Busan (東莱券番, 釜山). At the Chosŏn Kwŏnbŏn, kisaeng were trained in both Korean court and folk dance, pansori (a genre of musical storytelling performed by a vocalist and a drummer), as well as in Japanese song and shamisen playing. Calligraphy and classical Chinese were also part of the curriculum. In the Giseong Kwŏnbŏn, the curriculum included Japanese dance and Japanese language (Heo Yeonhee, 52). According to Keith Howard, entering a Kwŏnbŏn to become a kisaeng became something many Korean young women strove for, as was the case of An Pich’wi (1926–1997), who entered the Chosŏn Kwŏnbŏn in the 1930s to become famous in both Korea and Japan through her radio broadcast performances in the following decade (Howard, Keith. Perspectives on Korean Music, Volume 1. (London: Ashgate, 2006), p. 93).

The reality of the kwŏnbŏn during Japanese colonial rule in Korea, however, was not limited to the idealistic goals of young women who dreamed of stardom. John Lie notes that the first kwŏnbŏn, established in Pyongyang in 1926, sought to train elite prostitutes (John Lie, “The Transformation of Sexual Work in 20th-Century Korea,” in Gender and Society 9.3 (1995): 313). As Ruth Barraclough pointed out, the 1920s became a decade of “complex reinvention for kisaeng as they negotiated the shift from courtesans (of varying grades) to new-style entertainers and sex workers in the kisaeng houses of Pyŏngyang and Seoul in Japan-occupied Korea” (Ruth Barraclough, “The Courtesan's Journal: Kisaeng and the Sex Labour Market in Colonial Korea,” in Intersections: Gender and Sexuality in Asia and the Pacific, Issue 29, May 2012.)  “Imperialism greatly expanded the market of Korean kisaeng” and “sexualised service work proliferated with industrialization and large numbers of people pouring into the cities looking for work (Ruth Barraclough, “A history of sex work in modern Korea,” in Mark McLelland and Vera Mackie (eds.), Routledge Handbook of Sexuality Studies in East Asia (Routledge, 2014), p. 296).” Lee Insuk remarks that since the promulgation of the Kisaeng Regulation Order in 1908, the Japanese police did not recognize the traditional hierarchy of kisaeng, but rather treated kisaeng, ch’anggi (娼妓), and chakpu (酌婦), all equally as sex workers, downplaying the kisaeng’s role as a skilled female professional entertainer (Lee Insuk, “Convention and Innovation: The Lives and Cultural Legacy of the Kisaeng in Colonial Korea (1910-1945),” Seoul Journal of Korean Studies 23.1 (2010): 71-93, p. 78).” In a similar vein, in his study of the kisaeng in Pusan, Kawamura Minato notes that during the period of Japanese occupation most prostitutes working in the Wanwol-dong (Hangul 완월동; Hanja 玩月洞) area were Japanese geishas; the kisaeng would be called from the kwŏnbŏn to work in private parties, with sexual exchange occurring under a façade of legal entertainment labor (Kawamura Minato. Kīsen: “Mono o iu hana” no bunkashi [Kisaeng: A cultural history of “the flower that speaks”] (Tokyo: Sakuhinsha, 2001), p. 9). Kawamura also notes that, in its modern form, kisaeng was the product of Japanese male demand for a Korean equivalent of geisha during the colonial occupation (See Caroline Norma’s further development of Kawamura’s analysis in “Demand from Abroad: Japanese Involvement in the 1970s’ Development of South Koreas Sex Industry,” The Journal of Korean Studies 19, p. 409).

As Japanese assimilation policies became formalized in 1912 under the Special Order for Koreans (Chosŏnmin saryŏng 朝鮮民事令), resistance groups emerged, such as the March 1 Movement of 1919, one of the earliest public displays of Korean resistance against Japanese rule, and the Hyŏngp’yŏngsa, launched in Jinju in 1923 as a paekchong liberation movement (see See Joong-Seop Kim, The Korean Paekjong Under Japanese Rule (Routledge, 2013). In regard to the kisaeng, the Japanese administration’s emphasis on promoting its prostitution aspects would result in a social stigma associated with the kisaeng’s identity in contemporary Korea.

Among the dances found in the kisaeng repertory, perhaps the Salpuri is the most well-known, as well as one of the most beautiful. The dance was originally performed as part of a shamanic exorcism ritual, during which the shaman removes the sal, or evil, and calms the spirit of the dead. The white garment implies mourning, and the entire piece expresses the sorrow of separation through death. The tragic sense of the Salpuri repeats that found in Pansori, a popular genre of musical storytelling performed by a singer and a drummer which originated among street performers during the 17th century.



Lee Mae-bang (李梅芳 이매방 1927-2015) was born in Mokpo, South Jeolla Province in 1927. He was designated as Korea’s National Human Treasure for Intangible Cultural Asset No. 27 and 97, namely the Seungmu and Salpuri dances. He started learning these dances directly from kisaeng in his hometown at the age of 7.


Shamanism had also an important role in the formation of the literary arts in East Asia. In Japan, for instance, scholars such as Nakayama Tarō have suggested that Japanese poetry emerged from the incantations vocalized by shamans. From the heightened role of female shamans, a feminine literature emerged, with women excelling since the earliest times. Nukata no Ōkimi (額田王 Nukata no Ōkimi, c. 630–690 CE), a noblewoman from the Asuka period, is an example of one of the major female poets of the Man'yōshū (万葉集 759 CE). In the following lines, Princess Nukata laments the transference of the court away from the site of Mount Miwa, asking the clouds not to cover it in her last view:

三輪山をしかも隠すか   miwayamawo shikamo kakusuka       Will you indeed hide Mount Miwa? 

雲だにも情あらなむ       kumodanimo kokoro aranan              Even the clouds are expected to have a heart

隠さふべしや                  kakusoubeshiya                                   Please don't hide it

The collection of poetry titled Nyōbō Sanjūrokkasen (The Thirty-Six Female Immortals of Poetry 女房三十六歌仙), compiled in the Kamakura period, attests to the success of female Japanese court poets. Poetry flourished also from the yūjo outcaste female performers who managed to make their way into the Japanese court during the Heian period (794-1185). Among the famous literati/courtesans of the Heian, Izumi Shikibu (和泉式部 born c. 976) is thought to be of yūjo origin (See Yanagita Kunio, Momen Izen no Koto [Before the advent of cotton] (Iwanami Shoten, 1979). Yanagita derives the notion of Izumi Shikubu’s yūjo origin from a reading of the Otogizōshi (御伽草子) a collection of prose narratives from the Muromachi period (1392–1573); see alsoWakita Haruko, Josei geinō no genryū: kugutsushikusemaishirabyōshi [The origin of female performers: kugutsushi, kusemai, shirabyōshi] (Kadokawa Gakugei Shuppan, 2014) and Saeki Junko, Yūjo no bunkashi (Chūō Kōronsha, 1987), p. 62-66. Shizuka Gozen (静御前 1165–1211) and Giō (祇王12th century) are also yūjo/shirabyōshi who achieved great success at the time.) 

As Chieko Irie Mulhern remarked, “thorn between worldly ties and physical desires, Izumi Shikibu left a wealth of passionate love poetry, fueling rumors that purported that she was a femme fatale with numerous lovers besides her two husbands and two princely lovers” (Japanese Women Writers: A Bio-critical Sourcebook (Greenwood Press, 1994), p. 155). According to Ōhashi Kiyohide, while still married to the provincial aristocrat Tachibana no Michisada (橘 道貞 d.1016), Shikibu began a love affair with Prince Tametaka (為尊親王 977 –1002), and after the latter’s death, a new affair with his younger brother, Prince Atsumichi (敦道親王 981–1007) (Ōhashi Kiyohide, Izumi Shikibu Den no Kenkyū (Izumi Shoin, 1994), p. 73. (大橋清秀『和泉式部伝の研究』、和泉書院、1994、p. 73). Consider Shikubu’s following verses:

黒髪の                    Kurokami no                My black hair

みだれもしらず     midare mo shirazu     Caring not about its disarray

うちふせば            uchifuseba                   And looking back towards the past

まづかきやりし    mazu kakiyarishi          The one who first disheveled it

人ぞ恋しき           hito zo koishiki              Is the one I love the most

Shikibu's poem can be regarded as a turning point in Japanese literary history, for it established an enduring symbol of Japanese female sensuality, namely a woman’s disheveled black hair after a night of passion. With these verses, the words kurogami (黒髪) and midare (みだれ) became a staple of Japanese erotic love.

A singular heritage of pre-modern female poetry is discernible in the three major literary traditions of East Asia. Connecting the literary arts of China, Korea and Japan, the poetry of the yìjì (艺妓), the kisaeng (妓生), and the yūjo (遊女) emerges as another remarkable artistic development issued from female shamanism. Considering the social status of women in those societies, and the underprivileged origin of the female performers and poets, one intuits why the theme of the abandoned woman became dominant in this tradition. Consider, for instance, the following poem by Hwang Jini (黃眞伊 c. 1522–1565), a kisaeng (妓生) who entered the history of Korean letters as one of the most prominent poets in the sijo style (transl. by Larry Gross):

동지달 기나긴 밤을 한 허리를 버혀 내여           Oh that I might capture the essence of this deep midwinter night

춘풍 이불 아래 서리서리 넣었다가                   and fold it softly into the waft of a spring-moon quilt

어론 님 오신 날 밤이여든 굽이굽이 펴리라       then uncoil it the night my beloved returns

Hwang Jini’s poem highlights the theme of female desire as longing for intimacy and suggests a tragic sense embedded in the lives of the courtesans. Commenting on Jini’s poem “First Moon” (初月), Lee Hai-soon notes that “the fact that Hwang Jini, who befriended the most brilliant men of the time such as Seo Hwadam, So Seyang, and others, looked upon the first moon and contemplated abandonment rather than renewed expectation reveals a somber aspect of her inner life” (Lee, Hai-soon. The Poetic World of Classic Korean Women Writers (Ewha Womans University Press, 2005), p. 17). A collection of poetry from mudangs can be found in the compilation entitled Han'gugŭi Muga (韓國의 巫歌) published in seven volumes (see 한국의 무가 , 홍태한, 민속원, 2004). 

Together with Hwang Jini, a list of Yi Dynasty kisaeng who excelled in poetry includes Songi (松伊), Maech’ang (梅窓), and Myeong-ok (明玉). To them we might add the name of Ho Chohui (許楚姬 1563–1589), a famous poet of upper class yangban origin. In the case of imperial China, a list of yìjì who entered history as major poets include Xue Tao (薛濤 768–831) and Yu Xuanji (魚玄機 844–869), from the Tang dynasty, and Xue Susu (薛素素 c. 1564–1637), Dong Xiaowan (董小宛 1624–1651) and Liu Rushi (柳如是 1618–1664), from the Ming dynasty. These women lived, performed and wrote in Nangjing’s legendary Quinhuai (秦淮) pleasure quarters established by the Ming emperor Zhu Yuanzhang (朱元璋 1368–1398) near the Yangzi river delta. They inherited the burden of living in a society built upon a male-centered ideology of ancient origin. Their lives conflicted with the exigencies of a time when works such as Liu Xiang’s (c. 77–18 BCE) Biographies of Exemplary Women (Liènǚ Zhuàn 列女傳) were a staple of female didactic literature, propagating tales of female martyrdom with the aim of strengthening patrilineal values. From that, abandonment became the privileged theme. Consider the following poem by Xue Susu (transl. Daria Berg):  

懷人詩                        Poem on Longing for Someone

良夜思君歸不歸,        This lovely night I think of you, wondering whether you will return;

孤燈照客影微微。     The lonely lamp shines on me, casting a faint shadow.

擕來獨枕誰相問,        I clutch one lone pillow, there is nobody to talk to;

明月空庭淚濕衣。     Moonlight floods the deserted courtyard, tears soak my dress

Among the exigencies included in the model imposed with the list of exemplary women was the avoidance of remarrying at the advent of widowhood. In this regard, Rafe de Crespigny records eleven widows who committed suicide in the face of remarriage during the Later Han (23-220 AD), together with several other who cut their noses in order to repel suitors (Rafe de Crespigny, A Biographical Dictionary of Later Han to the Three Kingdoms (23-220 AD) (Brill, 2007), p. 1305). Suicide became the fate of many yìjì, this being the case of Ming poet Liu Rushi, who took her life after the death of her husband, the official/scholar Qian Qianyi (銭謙益 1582–1664). The following is a fragment of a film on the life of the female performer/poet. Note how at an early age she becomes an apprentice in the dancing arts of the courtesan:  



Liu Rushi 柳如是 (1618–1664) was one of the female poets in the yìjì tradition who excelled during the Ming Dynasty. Later in the Qing Dynasty she was listed as one of the Eight Beauties of Qinhuai (秦淮八艳). Her outcaste origin points to the intersection between female performers and shamanism expounded above, that is, following the logic that in China ethnic groups that practiced shamanism were referred to by ethnic slurs and segregated since ancient times. As a young girl, Liu Rushi was sold by her family as a concubine to a high official, being then sold again to a brothel. She committed suicide in 1664 at the age of 36.


Considering the history of female shamans, a double layer of oppression is discernible. On one side, stigmatization springs from these women's role in a religion that involves magic and violence and is in itself subjected to social discrimination; on another, it emerges from gender subjugation. Regarding the former, the relation between shamanism and violence suggests a possible rationale for the practitioners of shamanism becoming regarded as barbaric. The use of animal sacrifice in shamanic rituals may be considered as an important factor in the process that led to social segregation. The coincidence in the habit of consuming dog meat among the Japanese sanka and the Korean paekchŏng may provide an instance for further speculation on the subject. If shamanism can be called a religion of resistance, then it is important to inquire on what are the concrete elements that made it conflict with the systems of thought that have attempted to supplant it. Animal sacrifice, I believe, is an important element of shamanic rituals that recurs across cultures and geographic areas and that have made shamanism unpalatable in a global scale. In any case, besides being unjustifiable by any standards of human morality, animal sacrifice is a theme that the rituals' practitioners tend to avoid, especially in places where violence against animals is outlawed. Below are three videos that show Korean shamanic rituals. Note how drumming and dancing are an indispensable element of shamanism and how rhythm plays a fundamental role in inducing trance:  



This is a kut ritual from the are of the pre-modern Korean province of Hwanghae, one of the eight provinces during the Chosŏn period. The regular beat played at the changgu (hourglass-shaped drum) is meant to induce trance. As we will see in more detail below, this form of drumming is common to most shamanic rituals. The changgu is probably one of the most ancient Korean instruments; representations of it were found in both the kingdoms of Silla (57 BCE–935 CE) and Koguryŏ (37 BCE–668 CE). 


Hwanghae-do Kut

Another Hwanghae-do kut ritual, this one presented at the Kaya Culture Festival in the city of Kimhae, South Kyŏngsang province. Note the graceful bearings of the mudang and the rhythmic patterns in the drumming characteristic of shamanic rituals together. Note also the signs of animal sacrifice towards the end of the video. There are three types of Hwanghae-do kut: Naerimgut (initiation rite), Jinogwigut (performed for the dead), and Baeyeonsingut (fishermen's rite wishing for abundant catch).  


Myŏngsŏng Hwanghu Kut

As a series of popular rituals, Korean shamanism has an important political dimension. On October, 8 1895, Empress Myeongseong, or Myŏngsŏng Hwanghu (1851–1895) (I am using the McCune–Reischauer romanization system throughout this text), was assassinated by a group of Japanese soldiers acting under orders from Miura Gorō (三浦梧楼 1847–1926), the Japanese Minister to Korea at the time. This became known in Korea as the Eulmi Incident (을미사변). Research on this kut can be found in the volume Myŏngsŏng Hwanghu haewŏn kut yŏn'gu (명성황후해원굿연구, 민속원, 2008).


The fundamental role of rhythm in shamanism appears in the former's function as a tool for inducing trance. Looking at the traditions of the Mongol, Buryat, and Tungus people (Evenks) we find drumming and music as an important part of ritual. The sound of repeated rhythmic appears to be capable of producing altered state of consciousness, that is, to have neurophysiological effects and the ability to elicit temporary changes in brain wave activity, from which imagery and ecstasy may ensue. The alleged connection with gods and spirits appears to be related to the trance produced. Please note that this phenomenon is unrelated to the repetitive beating of modern techno music played by youngsters in places like Ibiza, which seems to lead not to trance, but to brain numbness, imbecility, and mental retardation (or may actually presuppose it). Note that shamanic music has a primarily ritual meaning and is not necessarily related to our modern notions of art and performance.


Mongolian Hengereg Drum

The hengereg is a large ritual drum used in Mongolian and Siberian shamanic rituals. Bells sometimes are attached to the inside of the drum so that it can be shaken as well as struck. Note that the shaman covers his or her face with the drum. This appears to be partly related to the demand for the shaman's eyes being covered. Beholding a shaman's eyes while in a state of spirit possession is thought to be dangerous a capable of producing harm in the viewer. For an in-depth study of Mongolian shaman music see Carole Pegg's Mongolian Music, Dance, & Oral Narrative: Performing Diverse Identities  (University of Washington Press, 2001).


Mongolian Shaman - Trance

Another example of a Mongolian shaman in trance. The shaman's eyes being covered is a feature of the ritual that appears also in several other manifestations of shamanism across the globe, the most remarkable case being that of the Yoruba tradition in Africa. The drink given to the shaman in the video appears to be arshaan, a drink believed to be energized by the spirits and that may contain a mixture of vodka, milk, and tea (see Eva Jane N. Friedman, "Amidst Steppe and Taiga: Women Shamans in Hövsgöl Province, Mongolia," in Art Leete and R. Paul Firnhaber (eds.) Shamanism in the Interdisciplinary Context (Florida: Brown Walker Press, 2004), pp. 226-336, p. 231). 


The intersection between rhythm, music, and female shamanism can be observed in the video below, which brings a Candomblé ritual performed in João Pessoa, capital city of Paraíba State, Brazil. Please compare this footage with the above examples of Korean and Mongolian shamanism and consider the role of rhythm and drumming in the three cases. I will discuss the theme of the sacred nature ascribed to rhythm in the Afro-Brazilian religion in more detail in the next page.


Candomblé - Trance

This ritual shows a female shaman in an alleged state of spirit possession. The spirit in question is a "Pomba Gira," one of Candomblé's many "orixás" (deities). The Pomba Gira personifies female sexuality, beauty and sexual desire. She is worshiped with great and care because if unsatisfied she can become vicious and wrathful (note the semblance of dissatisfaction in the shaman's countenance as she moves and dances).  In many ways the Pomba Gira can be read as a very expressive type of Madame Bovary. The séance amounts to a real "fiesta," which is concocted to satisfy the Pomba Gira's whims and take her momentarily out of her boredom.


So far we have seen how shamanism's history is embedded in a dialectic of resistance and assimilation that resulted from the challenges it posed to politically more powerful religions and systems of thought. I suggested that a clash between shamanism and Confucianism exists at the root of formative cultural and political processes in East Asia. I pointed also to aspects of resistance found in the shamanic elements of the national dances of Korea, as well as of assimilation seen in the "shintoization" of the figure of the female shaman in the highly ideological Japanese religion. With this in perspective, we can now consider how shamanism emerged in the Americas, and how there Christianity took the place of Confucianism as shamanism's antagonist. This discussion will lead us directly into the theme of Afro-Brazilian Candomblé and its cultural manifestations. We must thus turn our attention to the African side of shamanism, where again we will encounter the previous themes of art, trance, violence, spirit possession and, above all, rhythm as elements that inform the above-mentioned process of resistance and assimilation.    

I started this text by mentioning Ezra Pound's view of rhythm as an underlying ground that is fundamental for the unity of an artwork. With the previous examples of shamanism's usages of rhythm in perspective, we can now reconsider Pound's theory under a new light. In the next page, we will see how rhythm can be read not only as an underlying principle of art, but also as a sacred element that connects art and ritual. 

Rhythm, ritual, and art are a response to fundamental human needs. They work by promising individuals and groups deliverance from complex fears and anxieties. Regardless of whether Korean mudangs, Japanese miko, and Brazilian mães de santo can be seen as embodiments of one and the same female archetype, what is clear is that they do share a common root in shamanism and that their arts hold a common import in tackling those universal human fears. With this in perspective, we can now turn our attention to how shamanism and its sacred, trance-inducing rhythms manifests itself at the other side of the globe.