Lost in participative obervation? - Music and gesture in the rituals of “possession” or shamanism

The article starts out by clearly distinguishing shamanism and possession. The shaman enters into trance voluntarily and is considered to be a ‘musicker’. On the other hand, the possessed passively experiences the trance and is said to be ‘musicked’. Music plays a considerable role among the elements of the system of the ritual. Indeed, the musical instrument becomes a tool which provides rhythm to rituals, due to its inherent scenography and the emotional and sensitive charact5eristics that derive from it. The set of these features encourages individuals to enter into trance. In a social context, music provokes different reactions according to the people’s culture. Music and dance enable to explore unexpected and diverse territories of the human “supernature”.

Music and gesture in the rituals of “possession” or shamanism

Reading this book calls for a reflection on gesture and sound in the relationship to spirits. In particular, it leads to giving serious consideration to the issue of “authenticity”, notably concerning the behavior called “trance”: researchers do not always interrogate the use of this word. I will here develop a commentary in four points. First, I will highlight the figures of the shaman and the possessed, and the terms “musicker” [musiquant in the French-language original] and “musicked” [musiqué in the original] a taxonomy borrowed from Gilbert Rouget, in the course of the study of a Siberian case. Then I will examine the respective roles of the gesture and the word as ways of representing the spirits, which will thirdly lead me on to discuss the issue of syncretism and the adaptation of the rituals to social and religious changes. Finally, I will question the need to “go into trance” and to seek “authenticity” in the behavior, on the one hand, and in the facts of shamanism and possession[1] on the other.

The shaman, the possessed, the “musicker”, and the “musicked”?

The terms of shamanism and possession have been placed by Western scholars in distinct categories, and each one crystallizes popular representations of two figures defined as the shaman traveling in the world of spirits and the possessed embodied by a spirit.

7The terms of shamanism and possession do not refer to the same etymological categories. The concept of shamanism was introduced into Western culture after being first mentioned by the archpriest Avvakum, who had been sent into exile in Siberia following the schism of the Orthodox Church at the dawn of the eighteenth century. He first coined this word in his travelogue in 1699. However, different hypotheses remain as to the origin of the word[2]. The most relevant one links it to a Tungus etymology and means “to move the lower body” (as in animals during rut) and “to shamanize”. In Mongolian and Tungus languages, the word means “jumping, leaping, head-butting” (Hamayon 1990: 493-498). The shaman has to find an agreement between the community and the spirits of the forest which provide game, so he holds more than one office: he is simultaneously a healer, a magician, a chieftain, an augur... This suggested to the intellectuals of the early twentieth century that there was no such thing as a shamanistic religion. Today, both in scientific literature and in popularized publications, the words shamanism and shaman are used beyond Siberia to describe religious phenomena in both North and South America, in Japan, Korea, Nepal, Australia, and also in Africa. It is difficult to oppose “shamanism”, whose etymology stems from an indigenous language, and the word “possession”, which refers in our language to a psycho-physiological state and is not linked to a specific cultural area. It refers to the incorporation[3] of a spirit by somebody during the ritual. Generally, as shown by researchers studying rituals in this category, the person embodying the spirit is different from the ritual officer. Thus, in Uttaranchal domestic cults in northern India, the Damai tailors-musicians cannot receive any initiation, but their presence is required to play music during rites (Bernède, p. 57).

For Bertrand Hell, shamanism and possession cults are “effective methods” (1999: 29) that give meaning to things at the moment when the inner self and the invisible meet; the symbolism of “these methods” is shared by the whole of society (ibid.: 83). The shaman and the possessed are for him the “masters of disorder”, capable of effectively relieving misfortunes (ibid. 29).

According to Gilbert Rouget (1990), shamanism may be said to be “passive”, while possession is said to be “active”. Through music, the shaman habitually enters into trance through his own will: he is the “musicker”. However possession is still deemed to be a passive process because the possessed “undergoes” it: he is “musicked”. To put it in another way, the music is played for him by other participants in the ritual (ibid. 441). Rouget notes that this duality comes in addition to three other types of dichotomy. The first opposition is that between a shamanistic trip within the possessed person's mind and a visit to the spirits by the possessed. The second is that between a control over the spirits by the shaman and a possessed person's submission to these same spirits. And finally the third is between the trance voluntarily induced by the shaman and the involuntary trance of the possessed (ibid.: 252-253). Actually, as we will see, some shamanistic rituals attempt to secure simultaneously the quest of the soul and healing, through an “exorcism” of the spirit responsible for the possessed person’s state of sickness.

The dichotomy continually pointed out by the dualistic categories proposed by scientists and accepted as such by authors and their readers, must be questioned because the shaman, for example, can travel but can also incorporate some spirits. Similarly, in some possession cults, the possessed also behaves as a shaman. In the Siberian area, according to Roberta Hamayon “the threshold is never clear between shamanism and possession, as they are usually defined” (1990: 451).

Additional figures to the “musicker” and the “musicked” - A Siberian example

6I propose to shed light here on the complementary characteristic between the functions of “musicker” and “musicked” by using a Siberian Yakut example where[4] the shaman is both “musicker” and “musicked”, possessor and possessed.

I will use both the metaphors which feature in the ritual: one of the “ridden” shaman (horse) and the other of the “riding” shaman (rider), where the shaman is both “musicker” and “musicked”, possessed and “shamanizing”.

The combination of the two images makes for the effectiveness of the ritual: i.e. the horse and the rider[5]. Indeed, the shaman calls his drum “his horse” and uses it to get into contact with the spirits. He is at the same time a horse and a rider: a strap (sometimes a chain), called in Yakut language “reins” (Khudiakov 1969: 324), is attached on the back of his costume. It is held by an assistant called kuturuksut (literally “He who holds the tail [kuturuk]”. [Samsonov, 1996: 64; see also Hamayon 1997: 199]).

Two associated figures also emerge: on the one hand, the assistant, kuturuksut, that is the rider holding the shaman horse, and on the other, the shaman rider encouraging his drum-horse by beating the wall. The strap which falls behind the shaman is called the “reins”, so the Western observer might think its role is to control the ritual officer. However, if the drum-horse metaphor is analyzed, it can be said that the shaman himself is called “rider” but he says that his horse is not trained[6]. The shaman and his assistant consider the drum as a means of transport that the shaman must use to cross the spaces peopled by spirits. In other words, the mount, whether it be the shaman himself or his drum, is always a non-trained horse. That is why neither the shaman nor his assistant is considered to be a true rider (the assistant is merely he who “holds the tail”). The idea of controlling the mount is absent, because it is precisely this kind of freedom which provides effectiveness to the ritual[7].

I propose the following table showing the way to get in relation with spirits, developed from the description of a Yakut ritual in the late nineteenth century (Khudiakov 1969: 303-416):

At first glance, some Yakut rituals should be called shamanism, and others possession. Indeed, the shaman is a human being among the spirits when he beats the drum, thereby symbolizing a sexual relationship with one or more spirits. In the process, the audience considers the shaman as both a spirit-possessed horse and a rider. However, when he embodies a spirit among humans, the shaman dances and jumps to mime the incorporation of the dead person's spirit. In the first type of relationship, the shaman leads the ritual without an assistant, while in the second, he needs to be attended. These are two modes of representations described by Roberta Hamayon (2006: 28-29) as “direct contact” with the spirits: simultaneously, the shaman has an impact on the spirit and he renders it present. The whole effectiveness of the ritual lies in the art of representing this link. The public obedience to the ritual depends on this (ibid.: 28).

Viktor Mikhailovsky[8] noted that “aS3n iron chain is attached to the back of the shaman's costume, symbolizing the shaman's power”. Actually, the analysis of the ritual shows that the assistant touches the chain during the incorporation of a spirit but not during the shamanic trip. For this reason, one can consider the chain more as a way of connecting with the shaman when he is possessed by a spirit, than a sign of power or strength. On the contrary, when the shaman incorporates a spirit, he symbolically takes a risk, which leads to a human necessarily controlling the ritual. Therefore the mount metaphor in the Yakut ritual possesses two modes of representing the spirits: shamanism and possession, but the presence of an assistant is only required in the part of the ritual implying possession.

Thus, as Henri Lecomte stated (p. 37), “the shaman [...] is not always his own musicker throughout the ritual”[9] and it can also happen that some of the possessed sing and become the “musickers” of their own trance (Xavier Vatin, p. 194).

Watching, Talking, it’s already doing it… Music semantics and ritual efficiency.

Musical instruments do not have a proper ritual function themselves[10], but people bestow this function on them because of the sound they make, their shape, the gestures musicians make according to their size, their weight and many other criteria. Rouget himself noticed (1990 : 183) that trance isn’t the result of a pre-determined rhythm. Rhythm is just a part of the ritualistic system, which takes into account many other features, such as movements, disciples or priests’ discourses, the environmental setting, the moment when the action takes place... However, in a very interesting article, Jean-Jacques Nattiez says that music has its “own symbolic capacity” (2004: 70). According to this author, “in music there is a strong emotional or affective nature that exists in any cultural codification. This cultural codification creates this emotional or affective nature and partly explains it” (ibid. 67)[11]. Laurent Aubert somewhat shares the same point of view :

“[…] Music can truly be efficient only within the social and factual context in which it takes place. This efficiency -when it’s demonstrated -is not only deduced from the sounds or their acoustic properties, but also from the function socially attributed to the music and to its sound codes in a specific situation. These are immediately perceived and used by the “musicked” disciples.”(p. 17).

I do not personally believe in an immanent property of music but rather in its subjective nature[12] and in what Roberta Hamayon (p. 21) called the “decoupling” of the physical registers (gesture), the psychic state (sound) and cultural ideas (symbolic values). It might be said, as Erwan Dianteill (p. 188) puts it, that music “has the sensitive function of being a mediator between interiority and exteriority.”

Everybody reacts differently to music according to their culture, to their representations of music, to their distant and close past, to their present and to their education. Reactions to a particular kind of music change according to different periods of life. It is possible that the same type of music creates similar behavior in different cultures, which does not mean that those behaviors are the result of an inherent capacity of the music to cause it. In different ways, two very different individuals will obtain the same results without giving them the same meaning.

I propose to illustrate my argument with a short comment about the Siberian concert following the conference that is at the origin of the book Shamanism and possession. German and Claudia Khatylaev, Yakut musicians, interpret an ethnic music[13], inspired by melodies and the singing techniques of their culture. They play shamanic drums and jalopies, considered to be ritual instruments in neighboring cultures[14]. During their concerts they invited people on stage and let them use their instruments to play a song together. This song is quite fast and powerful, and each musician simply hits the notes. However, in Geneva, the conference attracted some fans of shamanism among the general public and a man eventually got on stage and started playing the Jew's harp with his eyes closed. He sometimes opened his eyes to show only the whites, “visibly” manifesting a state of “trance”. He and his wife then followed the musicians to Annecy where they wanted to attend the next concert -although it was identical –and in order “to get high”, according to them, on this shamanistic music. The effects of the Yakut shamanistic drum beats is obviously identical on antique Yakuts and a modern French audience.

They consist in what literature and even science, often calls an “altered state of mind”. However, firstly, can we prove that “trance” exists and is not simulated? Secondly, is music the cause of trance[15]? Obviously, no one argues that “the foreigner does not experience [the trance] with the same emotions as autochthons, but it is quite possible that they understand the universe the music refers to with these associations” (Nattiez 2004: 68). In my opinion, I prefer to say there are different paths taken by auditors to reach the state called “trance”. Indeed, the Western listener has made many readings about shamanism, which gave him the key to get into contact with spirits on a “shamanistic trip”. So he might even give more value to the symbolism of the music than the autochthones. Whether he really enters into trance is a minor question, that will be dealt with later on.

Semantics presents music as an echo in the ritual gesture, whether it be the actual dance or the bodily expressions. Franck Bernède (p. 73) talks about a grammar of the body. According to him, “every detail of postures is a sign of value”. At the beginning of their initiation, mediums notice only a few differences between the moves of the gods and those of evil spirits. The range of postures gradually changes (id.).

As observed by Rouget (1990 : 221), it should also be noted that gesture –  one required to play an instrument or a dance without an instrument, by a priest or a disciple of the ritual - may be the source of a “trance” possession according to the disciples themselves. Dance[16], as well as music, is a way of representing spirits during rituals, a way of representing the incorporation of a body by a spirit, which is seen as a way of acknowledging the presence of spirits among men.

Finally, the gestures and ritual sounds have a functionalist role. As Nattiez noted, “the important thing is that the association with the supernatural exists for the autochthone”(2004: 63). The mediums' bodies are the god’s home on earth. The dialect terminology talks of intrusions, tremors, penetrations (Franck Bernède, p. 47). The ritual language is codified. Thus, the phases follow a logical cycle. For example, in Damai in Northern India, the time of the ritual is divided according to the following steps :

1. A musical prelude (+ bards assistants) that announces the evening ritual. Each instrument is a kind of god.

2 -A phase meaning “encouragement”, which stands as a prelude to possession dances.

3 -“Awakening” (longest phase), mediums embodying the gods by identifying dances.

4 -“Incarnation”, consulting gods.

5 -“Blessing”, the god has left and the assembly and musicians are blessed (Franck Bernède,p. 59).

Nattiez calls this progression the “syntactic succession order” (2004: 61). Music and dance have specific functions. They must represent the spirit or translate the message that should be delivered. Music and dance are identifiable or invocative[17]. In shamanism, these entities remain inaccessible to the community. The shaman is the messenger, acting as a mediator. However, in the case of possession spirits, they are believed to live on earth. The embodied subject and the community can meet and “chat” with them (Laurent Aubert, p. 12). Music and dance are “ways of reaching every corner of the supernatural”. (Hamayon 1990: 485) So they are efficient means, with a structure and a function.

To conclude this chapter, let’s chronologically identify the respective contributions of Shamanism and possession.

The adaptation of the rituals to change and the socialization of the possessed

S2The diversity of the contributions to Shamanism and possession shows that the current position of these cults is towards the outskirts of important religions. When a dominant religion, in general monotheistic, coexists with rituals that imply a direct contact with spirits, they will be transformed into exorcisms or theatrical performances only, that is to say, camouflage strategies for the survival of the cult. Henri Lecomte reminds us that shamanistic practices were once repressed under the Soviet atheist ideology, while the Evenk shaman “often hid behind the actor’s mask” (p.38), without this affecting the efficiency of the ritual in the eyes of the public. Because exorcism is a part of possession, it enabled to incorporate the saints or the gods of the dominant religion within the cult (Rouget 1990: 287; Hell 1999-79).

Dana Rappoport observes that trance and possession practices are all the more widespread that imported religions reach populations on the islands of Bavi, Java, Sumatra and Sulavesi. According to her, the arrival of Christianity led to the forgetting of songs and the reduction in the number of trances, thereby questioning the efficiency of the ritual (p.108). The ritual has practically disappeared, but today trances take place in Pentecostal churches to the beat of imported music (p.114). The reader would certainly like to get more information in order to draw parallels with the Evangelists’ success in particular. Erwan Dianteill shares a few key elements concerning Afro-American Pentecostalism: in Cuba, in Candomblé rituals, the supreme divinity is invoked through gospel songs inspired by Methodist and Baptist protestant anthems. Inversely, in Pentecostal churches, adepts are “submerged by the holy spirit, pictured as a delicate fluid that coats around the person” (p.187). Faiza Seddik-Arkam shows that the fact that some Touaregs have adopted a sedentary lifestyle in the Algerian South, has profoundly altered beliefs and culture in its socio-cultural aspect and on a musical level. Henceforth, only a few women play the monotonal “vièle” (imzad) that use to accompany rituals.

Some articles illustrate the perverse nature of the process that transforms initiation into exorcism, as well described by Bertrand Hell (1999: 65), with regards to the persecutions that the adepts of the cult had to face. This trickery of therapy is present in the tantric rituals in Italy’s Apulia. It’s a shame that Gino di Mitri has trimmed down his contribution to a historiography of the “psychology of trance” (p.128). Unfortunately, he doesn’t emphasize the disguising of a cult of initiation (that has to do with a carnal and symbolic relationship between the possessed women and Saint Nicolas) as a “musicked” exorcism. This exorcism aims at curing a sickness supposed to originate from the tarantula’s bite[18]. Faiza Seddik-Arkam, on the other hand, does stress this socializing aspect of possession in Algeria. There, the masters of ceremony, whom people fear because they are considered as being close to pre-Islamic entities, can allow themselves to go against the norm, in particular by reciting erotic and almost vulgar poems (p.154). The ritual enables to “symbolically transgress the dominant model” (p.152) and tear down the frontiers between slaves, officers, nobles and artisans, who do not normally mingle in times of profane life.

If rituals adapt to change, as the authors of the article demonstrate, Roberta Hamayon is far more categorical. In her contribution, she shows that the epic had a significant ritual importance in Central Asian societies. For an Epic to work, it had to be correctly recited, in other words it had to be “shamanized” (p.24). However, today, the bard (baqsy) has taken the bakshi (shaman’s) place and the Epic has lost its ritual value to become an object of identity (p.27).

I detect a similar process at work among the Turks of Northern Siberia, the Yakuts. Shamanism in its pre-soviet phase has disappeared. Fortunately, the Unesco proclaimed their epic World Heritage in 2005 after an enduring work of lobbying by members of the government. They faced two choices: either disappear or be recognized as an official religion. The latter seems to be a more likely possibility. These rituals could be elevated to the rank of “religions”, as has already been the case for voodoo in Benin in 1996. Yakut intellectuals had tried to theorize their beliefs during a conference which was held in Yakutsk, in 1992, “Shamanism as a religion: genesis, reconstruction, tradition”[19] . Without hope. Christianity remains the only official religion of the Republic to this day, despite the success of the national day of Yhyakh, for which a ritual of greetings of the sun was set up during the summer solstice[20] which gathered 140 000 people in 2009 according to local newspapers.

Should one “go into trance”?

9This volume on Shamanism and possession questions the authenticity of the state the adepts of the cult find themselves in and the belief in the efficiency of the ritual.

A number of contributors, at one point in their article, are deceived by discourse. They apply a value judgment or aert, from the beginning, their belief in “trance”. Jean During speaks of the authenticity of the false characteristic of certain dervishes’ ecstasy. Moreover, he wonders (p.91) what has become of the faculty of possession in a globalized society. Maybe the researcher could have extended the field of his reflection and wondered why possession has disappeared in our society. And Laurent Aubert might have added: “If it is followed by concrete and verifiable effects, this means that the reality of the powers it involved was sustained by tradition. As a matter of fact, the denial of such a possibility would lead to another dead-end: indeed it would not be so easy to explain why and how a ritual lasts in the situation where the possessed would systematically deceive his audience […] other ritually modified states of consciousness can rightly be associated with the kinds of possession, which may be softer but  which remain real.” Can one still talk of a scientific hindsight?

Here is how Dana Rappoport (p.104) describes women dancing around Sumatra during a ritual: “suddenly, they enter an alternative state, their eyes revolving, their eyes closed, their heads revolving to the back”. Jean-Jacques Nattiez himself (2004:64), elsewhere so rigorous evokes the “capacity of musical phenomena to reveal religious entities.” In Siberian societies, Henri Lecomte notes that “the modernism of the music does not systematically make it a powerless music” (p.40). Not powerless, without doubt, but it would be convenient to add: “in the eyes of the autochthones”. Indeed, the question of the point of view is essential knowing that the researcher can find proof outside of what the possessed says or shows. For want to reveal the spirits and the heart of their subjects, they get trapped in participative observation, which practically carries them away with the magic of the instant and which defeats their certitudes. This is what makes Bertrand Hell (p.176) himself wonder if anthropology is still able to comprehend the principles of the state of trance.

There still seems to be among the researchers a need to tell true from false as regards the ritual. Gilbert Rouget (p.211) speaks of “the authenticity of possession and trance”, challenged by Roberta Hamayon (1995), in his article “To bring the trance to an end”. He accuses her of reducing trance to its symbolic dimension and of forgetting its emotional power and consequences. He uses Laurent Aubert’s Les Feux de la Déesse (2004) to show the importance of this aspect.

Actually, to speak of Siberian societies, the autochtones grant little importance to the difference between truth and fiction. That is not the underlying question and a phone that rings during a ritual will not threaten it. The question of authenticity[21] seems to represent merely a Western concern. Indeed, what counts to them, as we saw above, is efficiency. Shamans and bards, for the Exirit-Bulagat of Cisbaïkalie, are “submitted to the obligation of result” (Hamayon, p.25).

For Gilbert Rouget (1990: 560), “possession cannot work if it does not become a play and the shamanistic session, in its dramatic phase, can be compared to a one man-show during which the most diverse musical episodes take place” (ibid.: 248). For example, Daniel Kister (1995), who calls the Korean shamanistic ritual a psychodrama, develops the idea that the spirit’s dramatized presence is used by the mudang (the shaman woman) in her client’s therapy. Bertrand Hell (1999) goes even further, rejecting any kind of hypothesis linked to hysteria, given the highly coded structure of the ritual. In his article for Shamanism and Possession, he shows that in Morocco, it is “the one that undergoes an initiation that manages his or her possession. He/she determines the time of the dance and controls its tempo. It is him/ her who solicits the drum lute to play the melody, allowing himself to cut his/her tongue and his/her members. Even if the possessed is exhausted, he/she will control the tempo with a quick gesture in order to resume the dance” (p.1639: Therefore, it cannot be said that one loses control and his/her condition cannot be seen as trance in a Western sense, meaning a modified state of consciousness where the subject could be said to lose control over their body and spirit and where one would undergo the experience. Bertrand Hell refuses to accept the psychopathology in trance, which prevents the analysis of the phenomenon in its social and cultural complexity (p.171).

The dramatization of the rituals cannot be retricted to the notion of play and the performers’ disguise. Way beyond this notion is the idea that the dramatization of the cult emphasizes the game or acting, acknowledged as an act or a process that is almost immediately efficient. In Siberian societies, where the etymology of the word “shaman” is often linked to the word “game”, the performers must plead the spirits for luck for the hunter, when the latter goes out looking for game. Yet the latter has to “win” (quote of Roberte Hamayon in Amiotte-Suchet &Plattet 2003). That luck is still submitted to a randomness that the shaman can only try to master without guaranteeing any result. But the obligation to produce results remains true. Society will substitute an inefficient shaman by another more promising one. The contributions of this volume which this introduction is dedicated to, here show that in shamanism and possession, the two modes of representation of the spirits, the essential element is the act of representing[22] in itself, which gives life to the spirits among humans during the time of the ritual. Hence the question is not to “go into trance”, to figure out whether the subject is effectively in trance, but to expose the reasons and the implications for the state they display.

8To conclude the reading of this volume, let it be recalled that practically only the word trance was used in this compilation. The word ecstasy only appears briefly in Jean During’s article. However Mircea Eliade (1968) hs already evoked the “archaic techniques of ecstasy”, Joan Lewis (1971) called shamanism and possession religions of ecstasy, and Gilbert Rouget (1900) subtly distinguished the use of the word from trance. Does not the present use of the word trance, which is almost exclusive, suggest a certain vacuum and limit to the current studies[23]? Finally, what would be the ideal point of view to analyze “trance”? A collaboration of the disciplines, or would a contribution by the autochthones themselves resolve the problem?

To go into trance, while taking it too seriously, would mean forgetting the dramatized and thetricl dimensions that exist in rituals of shamanism and possession. It would also entail believing in it. And faith in a modified state of consciousness would put a stop to any attempt to explain it.

The conference organized in 2005 by the Workshops of traditional music in Geneva chose a seductive title, which attracted an audience interested by shamanism no longer from an Ethno-musical or anthropological point of view, but rather from a practical standpoint. Let us not forget that neo-shamanistic communities are particularly successful in Switzerland, where the Michael Harner Foundation is established.

However, the publication of the works of this conference under the more neutral title of Shamanism and Possession opens fascinating lines of enquiry to the researcher. In parallel, he gains access to study cases which, even though they come from all over the world, demonstrate the surprising proximity of the ways of connection with the spirits in so-called traditional societies and fuel a diachronic reflection on the symbolic importance of music for societies in transition.

[1] I gathered the critical elements of this book during my seminar about “Shamanism and possession” as

part of my research contract at the Centre for Landscape and Culture (University of Tallinn, Estonian

Foundation for Science).

[2] According to a theory which dates back from the eighteenth century and which tends to prove the

existence of Buddhism traces in the Ural-Altaic languages, the word would come from the Indian

Sanskrit sramana which means Buddhist monk. According to other theories, the word would come from

the Tungus word (sa-) which refers to knowledge (cf. Van Gennep 1903; Laufer 1917 teenth 2008).

[3] Still qualified as « mystical absorption » by Franck Bernède (p. 56).

[4] The Yakuts are the most northern Turkish people. So far, in far-east Siberia, they raised horses and

cattle next to reindeer peoples. They also hunt, like other peoples of the Altai area, their cousins. They

practice two types of shamanism, one based on livestock farming, the other focused on hunting. In the

pastoral shamanism, the relationship with the invisible assumes the spirits' presence, considered as the

ancestors of humans and pets donors, while in hunting shamanism, the relationship to supernature

contemplates the existence of spirits in animal form.

[5] The ritual metaphor of the horse and the rider in possession is also present in Africa among the Hausa

and the Yoruba. In many rituals, the possessed is seen as the mount (often as the mare) of the mind (in

the culture of Hausa in Niger [Erlmann 1982]; in Nigeria [Matory 1994]). In the Yoruba ritual in Nigeria,

there is a superposition of two metaphors: the spirit's wife (that is why the possessed is often a woman

but he can also be a cross-dresser man) and the horse. The symbol of the spirit's wife is linked to the

idea of a sexual relationship with the spirit, while according to Matory (1994: 505) the mounted horse

metaphor comes from a historic past full of the impressive figure of a powerful politico-religious order.

The concept of “mount” is commented by Franck Bernède in his contribution (p. 62): mediums in the

jagar worship of North India are represented as both horses or gods' mounts, and as the gods


[6]I commented in detail the dramatization of the domestication and the relationship between man and

animal in this double metaphor of horse and rider, in an article published in Études mongoles, sibériennes,

centrasiatiques et tibétaines journal (Maj, 2010).

[7] This idea of the mount in shamanism is strongly related to horse farm methods for Yakuts, which are

closer to reindeer farm techniques on large spaces, than cattle one, near villages (see previous note).

[8] In Shamanism in Siberia and European Russia, Helsinki, Mémoires de la Société finno-ougrienne 1933, 46,

quoted by Eliade (1968 : 130).

[9] Having noticed the narrow border between shamanism and possession, Gilbert Rouget (1990: 501-505)

himself had proposed the term of communal trance to describe a phenomenon at the frontier between

the two: trance implied by the “musiquing” and trance led by the “musicated”.

[10] Gilbert Rouget (ibis. : 159-160) showed it for drums which can be considered as a rhythm and a

melody instrument in tone languages.

[11] Jean-Jacques Nattiez proposes a definition of music semantics: “Human beings can associate a musical phenomenon (pitches, intervals, rhythms, chords, instruments…) with their own experience of the world (affective, psychological, social, religious, metaphysical, philosophical…), according to their needs (religious, eating, ecological, economical, playful, affective…) and according to own  musical  symbolic ability.”

[12]  Rouget talks of “restarting (music) for himself”.

[13] I consider their music as “ethnic” and not “traditional” because Yahut songs are enriched by Central

Asian melodies and natural sounds usually not heard in popular tradition. I will not enter here into the

debate about the nature of traditional music.

[14]Ethnographic literature never describes (except Khudiakov 1969) the use of Jew's-harp in past Yakut


[15]  Indeed, Gilbert Rouget (1990 : 228) shows for example that sometimes possession trance happens

without any music and consequently without any dance.

[16] Separating the abstract dances (which could have caused trance) from figurative dances (which could

caused possession state), Gilbert Rouget (1990 : 222-223) notices how difficult it is to determine if a

dance belongs to a category and to which category it belongs do, because the community may forget the

signification of the dance.

[17] Gilbert Rouget proposes here to “separate common meaning of possession (spirit possession) of

medium possession (spirit mediumship). In the first situation, the possessed does not necessarily deliver

any message to others. But in the second situation, emphasis is given to communication. Responsible

spirit of medium possession has something to tell to an audience” (1990 : 255 ; see also Lewis 1971).

[18]Cf. De Martino (1966) for example.  

[19] On such issues, it is possible to look up Balzer, Mandelstam…

[20] Observation during fieldwork.

[21] The musical fact, in particular so-called traditional music.

[22]In other words, to “make (symbolically) present”…

[23] Roberta Hamayon (1993) has written a whole article on the issue.


EHESS Editions – L'Homme

2011/1 – n°197 (pages 93 to 110)

The 19th edition of “Cahiers de musiques traditionnelles”, edited by Laurent Aubert is concerns musicality and dance in possession and shamanistic rituals in the so-called “traditional” societies. This work brings together the communications of eight speakers at an interdisciplinary conference entitled “Enter the trance! Music, shamanism and possession”, which was held at the Museum of Ethnography in Geneva on 20 and 21 May 2005. This multidisciplinary conference, which gathered ethnomusicologists, anthropologists and historians, also made room for less academic interventions such as those of ethno-psychiatrists for example. Two concerts also punctuated the seminar: one featuring Moroccan women's songs, and the other, a group of ethnic music from Yakutia.

> Author's Biography

Born in 1978, Emilie Maj is a Franco-Polish anthropologist. An Ecole Pratique des Hautes Etudes (Paris) alumnus, she is specialized in Yakut culture, a northern Siberian people. She studies nature and music from an anthropological standpoint. Currently a post-doc at the Quai Branly Museum, she is also a research fellow at the Centre for Landscape and Culture at the University of Tallinn (Estonia). Emilie Maj simultaneously teaches at the INALCO in Paris, as she masters Russian, Polish and Yakut. Emilie Maj regularly invites traditional musicians from Siberia to perform in concerts she organizes throughout Europe.

>>See the WRW under construction (2): Spiritual Trance, Back to Stay?