Lecture Three: Shamanism 3.4

3.4 Poietic practice and trance


A crucial feature of a shaman’s work is constituted by the shaman session (or séance), a complex ritual in which the shaman enacts his or her powers in order to achieve a specific goal (such as healing illness, divination, guiding the souls of recently deceased towards the proper afterlife place, or propitiation of specific spirits or deities). The session is a public event, in which the whole community takes part as (active) audience. Three elements are typical of shaman sessions: (i) a narrative framework, (ii) the use of various means (music, singing, dancing, and often the consumption of specific psychotropic ritual substances) to enter, sustain, and leave a particular cognitive condition, and (iii) the actual achievement of this condition (usually referred to as ‘trance’).

Scholars have sometimes given priority to one or the other of these aspects, with a propensity to focus on trance-states as a distinctive feature of shamanism.[1] During a trance, the shaman experiences an encounter with the spirit world, which often takes the form of a spiritual travel. The sort of cognitive states associated with trance varies considerably even during the same session, spanning from catalectic states in which the shaman seems completely unconscious, to other states in which the shaman seems to retain some degree of awareness, including states in which the shaman may seem to be possessed by a certain spirit that speaks through him or her. Sometimes, the use of music (especially drums) or psychotropic substances is used as a means to achieve trance. However, all these aspects are always framed within a precise narrative framework, shaped and expressed through ritual songs, music and dance. This ritual framework embeds the hermeneutic coordinates of the cosmology that informs the whole shamanic worldview, and thus makes the shaman’s performance meaningful for the audience.

Gilbert Rouget, in his Music and Trance. A Theory of the Relations Between Music and Possession (first French edition 1980) has offered one of the most systematic maps for navigating through these phenomena. First, Rouget helpfully distinguishes between ‘trance’ and ‘ectasis’ as belonging to two opposite side of the spectrum of experience: trance is a dynamic state, often associated with socialized activities (including music) and with sensory overstimulation, while ectasis a more static and introvert condition, most often pursued in solitude and silence, associated with a reduction or even deprivation of sensory stimulation. As we are going to discuss in Lecture Four, this understanding of ecstasy is more typical of what are often labelled ‘mystical experiences,’ which are worth distinguishing (in content, goals, and practice) from shamanic performances.

‘Trance’ (from the Lain transire) literally means ‘going across’ and typically means ‘going beyond’ ordinary experience. Within the practices that rely on trance, Rouget further distinguishes between shamanic trance and possession trance, mostly based on the way in which they are performed and socialized. In shamanic trance, one individual is the main actor of their own trance, which is often expressed as a travelling through a different but connected dimension of experience, namely, the spiritual dimension. Shamanic trance has also a specific goal to achieve, like healing or divination. Possession trance, instead, is a more passive event, in which the body of an adept is taken up by the spirit of a certain divinity or other agent, which has been evoked by a ritual. The adept temporarily loses their own identity and becomes the divinity or spirit who possesses them. The adept then behaves and acts in a way that reveals the presence of the possessing spirit in them. In possession trance, it is often the very experience of possession that constitute a form of healing for the adept. In possession trance, a master of ceremony usually guides the possession without undergoing possession themselves, and the audience is significantly more involved in the performance, since they are responsible for singing and playing music. In shamanic trance the community assists and supports the shaman’s travelling by providing emotional and musical feedback to the shaman’s ritual, but it is the shaman him or herself who also actively performs the music.

Music is used as a stereotyped framework to induce trance, guide it (which is usually expressed through specific forms of dance or movement that the adept executes), and eventually bringing it to a close. In possession trance, for instance, fixed motives and gestures are associated with different gods and the fact that an adept begins to enact them is recognized as the manifestation of possession by this or that god. Shamans also learn complex repertoires of songs and motives that express various happenings in the unfolding of the narrative they engage with during a session. In general, music, dance and gestures are ways of expressing in a shared social space the sort of emotional and experiential transformation occurring during trance. At the same time, the social space also works as a frame that both receives and controls the unfolding of that experience. Once again, trance reveals a dialectic between individual and society, ruling the shareability of experience (or its lack thereof), and the way in which different members of the same group interpret it.[2]

This quick overview makes it possible to understand possession trance as a sort of broader socialization of shamanic trance. What in some cultures is a sort of exclusive right of the shaman, in other cultures is available for sharing among a number of adepts. DuBois (2009, 59-60, 165) also cites the existence of more ‘democratized shamanism,’ especially among North America Natives, which might constitute an intermediary form between more traditional shamanic trance and possession trance. As Rouget notices, possession trance, like shamanic trance, has a healing function and usually presupposes a similar initiatory process, articulated in crisis, training, and becoming adept. We shall say a bit more about possession in Lecture Seven, in relation to Ancient Greece. In general, more recent scholarship tends to acknowledge that both shamanic-trance and possession-trance may coexist, albeit perhaps at different degrees, within the same culture.[3]

Moreover, Rouget’s analysis convincingly shows that there is no necessary connection between trance and music. All sorts of music is associated with trance in a range of different cultures, and it is not uncommon that musical qualities connected with trance are also present in musical forms that do not induce trance, or that trance-states can be entered without any music at all. What seems to be more constant in terms of accessing trance are (i) the adept’s beliefs and dispositions to enter trance; and (ii) the appropriate social setting in which this state will occur and be supported. The entering into trance is not just a private affair, but it always has a social dimension. In the same way that initiation entails a transmission of knowledge from the senior to the junior practitioner, the enacting of trance entails a sharing of experience between the adept and the rest of the community.

These observations reveal that entering trance is directly linked with a preliminary decision, established by the individual, about the meaning of the experience itself. This meaning has been most often inherited and established through a preliminary initiation and passed from a generation to the next within the community. The experience of trance is viewed as a social experience, and it is because of this social aspect that the adept can enter the state of trance when the appropriate social conditions are given.

Considering shamanic trance as a social state provides a better understanding of it in the context of shamanic practice. Focusing narrowly on the shaman’s own mental state during trance, Western scholars have offered various psychological and physiological accounts to explain what is happening during trance. These explanations range from association of trance with certain pathological states (epilepsy, hysteria), to neurological explanations based for instance on the activation and deactivation of different brain areas (see discussion in DuBois 2009, 109-120). Direct pathologization of trance is today mostly rejected as both scientifically unwarranted and culturally dismissive. A core aspect of shamanic trance, which sets it apart from pathological conditions, is the degree of control that the shaman retains all along on the performance of trance. While a spontaneous or natural trance-like state might be entered by the neophyte around the time of their crisis, becoming a shaman is essentially linked with the ability to deliberately master trance-like states and become proficient in entering them at will in appropriate circumstances and under request. From Lecture One and Two we also know that while finding brain correlates for any mental or cognitive state can provide information on the bodily counterpart of first-person experience and indicate certain necessary enabling conditions for a given experience to occur, beyond the fuzziness of most of these associations, locating brain correlates cannot be effectively used to reduce the experience to ‘nothing but the activation of this brain area in this particular way.’ This dismissive reductionist attitude is unwarranted because (i) trance states are not private individual states, but socially embedded, and hence brain activation can be only part of what makes trance possible and meaningful; and (ii) the actual causal network that leads to a specific pattern of brain activation loops between brain activation and first-person experience, thus making it impossible to single out the brain as the main source or cause for the experience itself.

These reasons invite to consider shamanic trance in its proper social context. As DuBois observes:

From the performative perspective, it is often the shaman’s stirring narration of the spirit journey, and the hearing of the voices of spirit helpers or spirit enemies speaking through the body and movements of the shaman that create the most powerful impressions of the séance itself. With its oratory, song, dance, occasional ventriloquism, and starling feats of physical and manual dexterity and stamina, the séance can be a stirring sensory event, one which depends upon a shaman who is conscious and active for at least a large portion of its duration. Thus, altered states which permit the shaman to retain some active control of body and voice are important to the success of the shamanic role. (DuBois 2009, 126)

Perhaps it might be better to regard the whole shamanic session as a poietic practice, namely, as a way of enacting a certain worldview, and using it to foster the community’s wellbeing and harmony according to the way in which this wellbeing is conceptualized in the community’s own cosmology. Rouget notices that the performance of possession rituals is not qualitatively different from Western more secularized forms of musical playing, like opera. Accepting that even shamans are performers of a kind does not necessarily denigrate what they do, but rather acknowledges the creative dimension of their role (and how that dimension persists in Western society, even if unacknowledged as such). An actor is someone who endorses a role; and endorsing a social role is something that everybody does in all sorts of societal contexts. We are all social actors to some extent. In today’s Western world, following up on an old line of thought that goes back to Plato at least, theatre might be seen as a place in which something fictitious is represented, and the actor as someone who ultimately deceives the audience and perhaps themselves, by appearing as someone else. However, it is this particular Platonic view (profoundly shaped by a specific way of delineating reality from illusion) that could also be challenged. An actor might also be seen as someone who enacts a social role, by thus bringing to life a vital component of the actor’s social milieu, something that the community to which the actor belongs needs for its wellbeing. This seems to be the case of shamans and their performances. It is during this performance that the whole cosmology upon which the shamanic worldview rests becomes publicly available, is restated, shared, and it is used as a hermeneutic matrix to negotiate and foster the community’s harmony. In this way, shamanic performance supports the construction of a certain form of self: the self that knows how to domesticate uncertainty. The shaman, in their symbiotic relation with their community, offers a paradigm of this form of selfhood.

Poietic practice can be associated with the dream-like states discussed in Lecture Two. The common feature across these states is the activation of imagination. Visions, sounds, and a whole world are evoked and shaped in imagination, with a vividness and emotional charge that often overpowers more ordinary waking experience. Imagination and dreams offer a paradigmatic case in which experience seems to arise on its own and respond to genuinely independent sources of agency. Nonetheless, imagination is amenable to a form of control, as the potential for lucid dreams shows. Imagination and dream-like states thus offer the ideal middle ground for an encounter with agents and spirits. From a technical point of view, this encounter is induced, sustained and eventually debriefed through the regulated use of various means. Music, dance, and singing create a symbolic sensory shell around the shaman, in which the whole of experience progressively concentrates and converges on the main theme or task of the session. Psychotropic substances can be used to further foster the activation of imagination, but it is important to notice that sounds and songs also do this work. Language is powerfully associated with visual images and it can evoke emotionally charged visionary experiences. Music, dance and poetry naturally induce visionary experiences and support a degree of concentration and absorption into them. Strong emotional commitment and physical endeavour (the shaman’s performance is often physically very demanding both in terms of power and endurance) further contribute to a withdrawal of the cognitive processes and attention from more ordinary and scattered objects or concerns. This brings the imaginative experience to the absolute foreground, up to the point that the imaginative or poietic creation remains almost the only content of experience, as in a sort of deliberately induced and controlled dream. The presence of a participating audience, with its constant support and feedback, sustains and brings to completion the experiential closure of the whole session around the shaman’s visionary travel.

Comparing poietic practice with dreams might reinforce in the sceptic the idea that the shaman is just feigning his or her encounter with the spirits. How do we know that the shaman encounters actual spirits in their visions, or even just the spirit that the shaman is supposed to meet? To some extent, the possibility of cheating is contemplated in shamanic cultures and this is the reason why the audience often plays a judging, normative role. But, at a deeper level, the sceptical worry misses the point of the whole shamanic endeavour and view of reality. The sceptic assumes a realist stance, according to which cognition is the passive acknowledgment of a pregiven external world, and then applies this stance to the issue of contacting spirits. If spirits are real entities, then they should exist in their own right in the world. The shaman’s task is that of contacting these entities, and hence we need to ascertain whether this contact in his vision is a genuine contact with entities that would otherwise exist in their own right (if they exist at all). But if we take seriously the implications of the communitarian model of agency introduced earlier in this lecture, then this way of understanding the shaman’s work is misleading.

The shaman is not a sort of detective who tries to disclose a hidden reality. Spirits are real insofar as they are genuine centres of agency, irreducible to someone’s else agency. The space of imagination is a space of epiphany precisely because it allows for the manifestation of seemingly autonomous and free sources of agency, which can be controlled to some extent, but cannot be entirely reduced to own’s own endorsed agency (more on this point in Lecture Four). The spirit world that the shaman travels through is the same world inherited and shared by the shaman’s community and its cosmology. It provides a common hermeneutic background that determines what to expect and what can be found in the domain of imagination (understood broadly as we did in Lecture Two). This background also allows the audience to judge whether the performance is perhaps going out of track. The shaman’s spirit helper is the fundamental guide who made the shaman into a shaman. In a sense, the spirit helper is the shaman’s other self, a domesticated agent that allows the shaman to remain in tune with the objects or levels of experience within which spirits manifest. At the beginning of the session, the shaman is provided with a task or a goal to accomplish, which is received from the community. During the session, the shaman elaborates on this goal, creates an answer, which is shared and socialized during the development of the session. The session is dialogical in nature, and when successful the audience finds in the shaman’s creation a satisfying way of addressing the initial task or goal. The way of creating this answer is inherently experienced as a discovery of some relatively independent agent or spirit because the whole cosmology and underpinning assumptions of the shamanic session lead one to consider any source of agency that is recalcitrant to endorsement as something distinct in its own right. This sort of relational independence is not at odds, but rather compatible with its constructed and poietic nature. To construct and create does not mean to fake or to generate something illusionary.

This does not entail that the shaman cannot fail to satisfy their audience, or even deliberately fake their performance; quite the contrary. But this possibility is not due to a contrast between a good shaman who really gets in touch with genuinely existing spirits, versus charlatans who simply feign this contact with purely imaginary beings. A good shaman is one who is sensitive enough and has enough visionary power to genuinely see what the community is seeking for, and in sharing this vision is able to domesticate the alien forces that shape the community’s life, by steering them towards a greater degree of harmony. This is a difficult task, which can succeed (even if not always) precisely thanks to the creative, poietic power of imagination.

  1. Sometimes academic literature refers to these states as ‘altered states of consciousness.’ This expression has been introduced and used as a more value-neutral description, but it ultimately risks making what is described more obscure. ‘Consciousness’ is a debated notion and there is little agreement on how it should be best understood. Even more debatable is what would constitute a ‘non-altered’ state of consciousness. If the intuition is that ordinary waking experience is what provides the point of reference for a ‘non-altered’ state, this seems to miss the profound way that waking experience is shaped by endogenous imagination and it is ultimately inseparable from the continuum of other states, as discussed in Lecture Two.
  2. Rich Freeman, ‘The Teyyam Tradition of Kerala’ (2003), discusses the case of possession rituals (teyyam) in south India (Kerala), stressing their strong and complex social dimension, in which cast hierarchies, integration between Brahminic and Dravidic cultures, and performative rehearsals of (symbolic or historical) challenges to the socio-political status quo converge. In particular, he comments (p. 318): ‘there is no clear ontological break between the human and the divine in this cultural context, but rather a continuum of expressions of powers, always divine to the extent they manifest an awesomely heightened effectiveness, and always human, to the extent that they emerge from social relations in a narratively historical context. This narrative continuum of the human-divine power spectrum is fully consonant with, and perhaps even necessary, as the discursive support for a performative mode of worship whose whole rationale is the demonstrated transformation of low-caste human beings into the tangible embodiments of living gods. The worldview of teyyam clearly implies that the entirety of human life at its various levels – physical, social, and political – is suffused with unseen powers.’
  3. See further discussion in Brian Morris, Religion and Anthropology. A Critical Introduction (2006), 22-25 and 37-40. Morris’s Chapter 1, devoted to shamanism, is a good overview of many of the topics discussed in greater detail by DuBois. Morton Klass, Mind Over Mind: The Anthropology and Psychology of Spirit Possession (2003) offers an in-depth discussion of spirit possession. Chapter 7, in particular, outlines an account in which spirit possession can be envisaged as a form of identity dissociation that is not pathological: ‘dissociative disorders studied and treated by psychopathologists are illness variants of a normal (that is, a nondisorder) capacity of humans to dissociate, by either external or internal suggestion. But there is another, and very important, difference between [dissociative identity disorder] and spirit possession. In [dissociative identity disorder], the new identities that surface are essentially unpredictable and idiosyncratic: in spirit possession, on the other hand, the new identities are those of entities known to and accepted as part of the individual’s (and the individual’s community’s) belief system’ (Klass 2003, 115).


Icon for the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

The Tragedy of the Self Copyright © 2023 by Andrea Sangiacomo is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.