Lecture Three: Shamanism 3.2

3.2 A communitarian model of agency


There are recurrent aspects that, despite variations and exceptions, define a set of core common features of shamanism, both across ethnic groups worldwide and in historical records. But before detailing these features, it might be helpful to introduce a more general model, largely derived from the insights we have gathered from the previous lectures about the relational nature of selfhood, in order to better understand how shamanism provides a distinctive way of constructing the self as a device for mastering uncertainty.

Imagine a small-scale community that relies on its own resources alone to survive and thrive. Small-scale entails that all members of the community can be directly and personally acquainted with all the others (no member of the community can be regarded as a stranger). The fact that the community needs to rely mostly on its own forces makes it particularly important to establish and maintain a degree of discipline and order among its members. This self-regulating communal structure needs to allow for both diversification of roles and for those roles to be complementary to one another. A well-functioning community of this sort might be described as harmonious. That is, it exemplifies the successful blending of distinct and yet symbiotically related individuals.

For harmony to be maintained, each member of the community needs to strongly endorse its social role within that community. Each individual depends on the whole community for its own survival and thriving, and the community as a whole depends on each individual fulfilling their role.[1] This mutual dependence is best served through a strong endorsement of social roles assigned to each member, which might include elements of identification with one’s kin, endorsement of the community values, narratives, and cosmological views, and skilfulness in practicing those tasks assigned to each and necessary for the common good. If we consider endorsement a crucial component of the experience of being a self, then this emphasis on the endorsement of social roles leads members of this sort of community to first and foremost identify themselves with their own social role, which is the object of the strongest endorsement. In this scenario, what I am and how I define myself is profoundly shaped by my being a member of my community (I am a child of A and B, I belong to clan Z, I am the one who provide Y for the community, and so forth).

The self-organization of a harmonious community can be compared with the autopoietic genesis of a cell or a more complex organism (Lecture One). Like a cell or any other organism, a human community can survive and thrive only by constantly interacting with the environment within which the community lives. Just as with a cell or complex organisms, there is a symbiotic and mutual adaptation between environment and community. In the ideal case for the community, the environment naturally provides what is needed for survival and thriving, but most likely the community needs to act skilfully to get what it needs. The environment is not a general and amorphous entity, but it is itself a community of various beings, each one performing their own tasks, forming their groups, and behaving in different ways. And this is all subject to change over time. Finding the right way of fitting this ecosystem is the main task of the community. In the same way the community emerges as a coordination of different individual agencies, the environment itself can be seen as a larger community in which different agencies manifest and operate. The same process that tries to establish a form of harmony within the community must also be scaled up and applied to the relation of the community with the environmental agencies upon which it depends. Harmony must be achieved both within and outside the community, since the two dimensions are not really independent but mutually co-determined.

Notice how this model almost inevitably leads to a carving up of reality and experience not primarily in terms of entities and objects, but rather in terms of agents. The idea of entity or object has to do with something that is pregiven, self-standing, indifferent to its being engaged with (or not being engaged with) by a human being. Agents, instead, are realities capable of bringing about changes around them, potential interlocutors with whom one might interact. Agents can establish targets for their actions and treat certain aspects of reality as objects, but they are more fundamental and even more relevant (from the point of view of how to live in a certain reality) than objects themselves. Agents can be recognized because they appear as relatively autonomous and independent centres of change and activity, they start or steer processes, and they give birth to new events. Agents might sometimes operate in the most regular and predictable way, and yet their agency is distinctive insofar as it cannot be directly controlled or subjugated. In this sense, agency goes together with a degree of freedom, understood as the irreducibility to the full control of another agent.

Members of a community are agents in this sense. They are distinct persons and individuals because they can act in their own way, they are relatively independent, and this freedom is precisely what makes harmony not only necessary but also precarious. While it might be in the common interest of all members to act in a harmonious way, the possibility that this harmony might break apart is always present. The same applies to the environment. A community lives in a living environment, namely, a community of other agents, some of them might be other humans belonging to other human communities, but most of the agents in the environment are non-human. Animals are clearly agents in their own way, but plants are agents as well, and even what we would call ‘inanimate beings,’ like elements, rivers, stars, atmospheric phenomena and so on. Insofar as these beings do something, namely, perform certain actions, they are agents. The fact that different agents have different forms of agency is only to be expected if one understands agency as the essential irreducibility of any agent to the agency of others. Difference is constitutive of agency, and thus difference from human agency is but a further proof of the genuine nature of non-human agency. If the environment on which the community depends is itself a community of agents, then even the harmony established with this environment is desirable but inherently precarious.

Before reflecting on this precarity further, it is important to appreciate how this broad view of reality in terms of agency shapes selfhood and endorsement. Each member of the community operates under a strong pressure to play their role within the community itself. This role has to do with one’s specific form of agency within the community, defined by both what is required of the individual from the community, and what the individual can do within it and thanks to it. This endorsement is perhaps the most fundamental bond that keeps the community together. However, the notion of agency entails freedom, and the reverse of this entails that whatever cannot be fully reduced to one’s own (endorsed) agency, must be recognized as a distinct agent in its own right. Since endorsed agency is defined by the community structure itself, any form of agency that manifests but is not reducible to what counts as endorsed agency, must be considered as somehow foreign, as another form of agency. The natural environment is itself a community of foreign and diverse agents, but a single human individual can also host a variety of different and alien sources of agency.

In this model, identity and difference are not sharply distinguished but relationally defined. An agent can do things only because it interacts with other agents that are relatively distinct and independent, but also available as potential partners, if not enemies. In this view, the crucial aspect is not how sharply identity is segregated from difference (since they cannot be segregated), but rather how the two are mutually co-determined. Insofar as different agencies are amenable to be harmonized, they can form a certain unity. At the social level, this happens in the harmonization of the different members, while at the individual level this can happen in the harmonization of the different sources of agency that might spring from within a single individual, and at the environmental level harmony can equally be established between the community and the larger agential ecosystem in which the community is embedded. The fundamental threshold in this view is thus the difference between harmonic coordination or disharmonic segregation. Unity does not entail simplicity (the absence of internal differentiation) but harmonization of the differences. In this context, being one-self does not require having only one, absolutely simple source of agency and domain of endorsement, but rather being able to harmonize whatever sources of agency are present in such a way that the result can be endorsed as functional within the host community. The community as a whole must also reproduce the same scheme and find a form of internal harmony that will allow for a harmonic relationship within its host ecosystem.


This account of agency entails a relatively weak form of embodiment.[2] Weak embodiment means that there is no biunivocal relation between agents and physically individuated bodies. One single human body can host more than one agent, insofar as these agents are not harmonized and thus appear as genuinely different and relatively independent sources of action. More challenging for a naturalist view, though, is the fact that one single agency might not be tied to one particular physical body. As we saw in the previous two lectures, this possibility is precisely what naturalism denies. The sort of weak embodiment entailed by the view at stake here does not completely dismiss the notion of embodiment, but rather relaxes the idea that embodiment needs to occur with respect to an individual body. In this sense, weak embodiment takes issue with the very idea of strong physical individuality, according to which an individual body can be defined in its own right, as an entity or an object that is ontologically self-standing. The notion of agency discussed so far does not allow for this sort of rigid and substantial individuality, since it does not allow for rigid and substantial identity. Identity is always defined within a context of diversity and relationality. Moreover, since agency is more fundamental than entities and objects, agency cannot be grounded and depend on an individual physical body (an entity) because the latter can be identified as such only in virtue of the agency that it exercises.

Weak embodiment entails that agency is itself a relational locus of change, which arises from the whole environment in which it is embedded. On the one hand, weak embodiment sees the tie between an agent and a certain individual physical body as relatively contingent; it can be broken, replaced, transformed, without the agent being destroyed in this process. On the other hand, weak embodiment entails ecological embodiment, namely, agency is always dependent upon a whole ecosystem in which it acts and operates in its own distinctive ways. In this sense, agents are necessarily embedded, but not at the level of individual physical bodies, but rather at the global level of the whole system in which they operate. Weak embodiment does not deny any form of embodiment at all, but specifically rejects the individualistic model of embodiment that is taken for granted by more naturalistic views.

There are two sources of evidence that can corroborate weak embodiment. First, we already discussed in Lecture Two a number of instances in which one’s own sense of agency and subjectivity can appear divorced from embodiment or identification with one’s own physical body. Lucid dreams, out-of-body experiences, near-death experiences are all cases in which the various components that usually participate in the construction of selfhood are somehow disarticulated, and the resulting experience is that of an altered form of embodiment. This array of experiences can provide first-person evidence for the possibility that agency is only weakly embodied.[3]

Second, from the point of view of a certain community, all the roles that are instantiated in the community are currently embodied by different individuals who endorse them. And yet, those roles and their corresponding agency can be detached from any specific physical individual and be reproduced by others. Social roles are weakly embodied in the sense that the same role can be instantiated by different individuals, and all these different individuals can equally strongly endorse the same role as their own self. Being a homeowner or a child, a parent or a warrior, a healer or a hunter, a shaman or a farmer are all social roles, different patterns of agency that get instantiated at various times by various individuals, and provide to those who endorse them a core dimension of their own selfhood. For as long as the community keeps going, and despite inevitable evolutions and transformations in its constitutive roles, these roles are forms of trans-individual agency that are only weakly embodied, while remaining ecologically embedded in the community as a whole.

The naturalist might come up with counter-arguments, counter-evidence and new interpretations for dismissing this form of weak embodiment. However, for present purposes, let us take it as at least a conceivable option for now. Two major consequences follow. The first is that there can be more agents than individual physical bodies. One body can host more than one agent, and some agents might not be embodied in a particular individual body. The density of agents in the whole environment is greater than the discrete individual bodies within it. If each agent is such because it retains an inalienable and irreducible degree of freedom, this entails that the whole project of harmonization (at the three scales of community-environment, community-members, and within one single member of a community) is even more challenging. Even if a significant number of agents will remain neutral with respect to any given community, any time a certain agent acts on its own accord, despite the community’s needs, harmony is under threat. Weak embodiment thus reveals that the survival and thriving of a harmonic community is inherently and profoundly precarious. Or, more accurately, it does justice to its inherent precarity by offering an explanation for it based on the inextricable density of agency on which the harmonic life of the community rests.

The second consequence is that disharmonious agency cannot be fully controlled nor prevented. When an agent manifests itself as recalcitrant to cooperate for the sake of harmonious symbiotic living, that agent cannot be controlled by taking ownership of it, since its very disharmonic behaviour is a proof of its relative independence and freedom. Disharmonious agency can be completely disruptive of the established identity upon which it acts. In this case, the form of selfhood that has been endorsed is subject to an alien source of agency, with respect to which it has no possible control, no right to claim, and also seemingly no escape. This option makes apparent the existential and daunting dimension of uncertainty that characterizes selfhood. In the model presented so far, selfhood is first instantiated as a form of harmonic coordination, but this coordination is constantly exposed to its disruption by the very nature of the elements that it tries to coordinate, namely, a dense domain of different agents that always retain some degree of freedom to dissociate themselves or operate against the community’s own interest.

This predicament is evident at the three scales discussed so far. At the individual level, one individual can host multiple sources of agency, and some of them might act in ways that are contrary to the social role that the individual has otherwise endorsed. Various drives, but also just states of illness, are instances of how the normal functioning of an individual can be disrupted by seemingly alien forces that act within it. At the community level, some individuals can begin to act against the community itself, as free riders, contesters, or just by trying to pursue other goals. Social conflict within the community is thus the socialized equivalent of the effects of individual illness. And yet at the environmental level, other human agents from other communities, or just other natural agents, can begin to act in ways that are detrimental to the community at large. Aggressions, natural catastrophes, or simply bad luck are the most obvious instances of a disharmony between the community’s needs and the actions of other agents upon which the community depends. In these and similar cases, disharmony directly threats the community’s survival and that of its members. And given the shape of this model, disharmony ought to be understood as a conflict with other agents (the source of conflict cannot be itself endorsed, given that endorsement is directed by the community towards what fosters harmony within it).

Disharmony is the constant and immanent danger nestled in the ideal of harmonious living. It provides the main manifestation of the inherent uncertainty that characterizes the construction of selfhood in this model. If disharmony is not treated, it will simply overwhelm and eventually destroy the self. But to be treated, alien and disharmonic agency cannot be simply assimilated and overpowered, because in acknowledging the problem of disharmony the right of alien agency to exercise its own freedom of dissenting and challenging harmony is also acknowledged. The way disharmony can be defused and possibly dissipated is through domestication.

Domestication entails mutual recognition, communication, and negotiation. By mutual recognition, the alien disharmonic agent is recognized and acknowledged as an interlocutor in its own right who is expressing its freedom to act, even if this expression turns out to be detrimental for another party. Acknowledging entails a process of personification of the agent, which includes recognizing its identity, giving it a name and shape, making explicit the way in which the agent is embedded in the whole ecosystem. The alien agent is thus recognized as another with whom it is possible to establish direct contact and eventually induce its way of acting to change for the best. To achieve this change, there must be a way of communicating with the alien agent. Communication should not be seen here as a transfer of information, but rather as a way of sharing with and familiarizing oneself with others. Communication can involve language (including the more performative ways of speaking like praising, praying, supplying, and addressing), but it can also involve sharing of goods (sacrifices, offerings), or undertaking of deeds (exchange of actions). This whole complex process entails mediation and negotiation. Something needs to be done in order to convince the alien agent to support harmonic symbiotic living rather than disrupting it. As in any negotiation, one’s demands must be balanced by conceding something to the alien agent in order to include, attract, and pacify it. This process of domestication is not dissimilar to (in fact, it might be seen as a further refinement and extension of) the process through which young children are taught to become members of their community, or certain non-human animals are tamed and trained to cooperate with human beings.

Insofar as domestication is successful, it leads to harmony, and thus it sustains and supports the communitarian form of selfhood described so far. However, domestication is not just something that is enacted in special circumstances. Selfhood here relies on harmonization, but harmonization of agency is structurally exposed to the risk of disharmony, at all scales, at all times. Hence, domestication cannot be a punctual or occasional practice, it needs to be a constant process of prevention of, and reaction to, ongoing centrifugal drives towards disharmony. There cannot be any harmony without keeping disharmony at bay, and hence a harmonic self can be constructed only through constant exercise of domestication. The model presented here envisages the self as a specific way of mastering uncertainty through domestication. This can be seen as the core business of shamanism.

  1. The term ‘individual’ should be interpreted here as referring to any functional biological unit that is commonly identified as a certain living body. The term does not necessarily entail further overtones about singleness of identity and personhood that might be attached to it in another context. In other words, it might be helpful to reserve the term ‘individual’ to refer to the basic physical and biological ground upon which personhood (any role or persona that is enacted by an individual as a way of expressing a certain form of agency) and identity (any attitude of appropriation through which a certain personal role is interpreted as belonging to the one who enacts it) can be superimposed. This trichotomy shows that the three notions (individual, person, and identity) can converge or diverge in different ways and in different contexts. In the following, it will become clear that agents are persons, they can be hosted in individuals (an individual can host several agents) and they can become the basis for an identity, or else cause the disruption of it in case of conflict between different persons or agents within the same individual or between a certain individual and their community.
  2. In today’s Western philosophy, this sort of weak embodiment might fit accounts that see cognitive life as ‘extended,’ namely, as irreducible to the physical boundaries of the individual body. For an accessible outline of how such an account can be defended, see Alva Noë, Out of Our Heads. Why You Are Not Your Brain, and Other Lessons from the Biology of Consciousness (2009). The enactivist account presented in Lecture One might be articulated along the lines of an extended approach, insofar as it takes the individual-environment relation to be constitutive and more fundamental than each of these relata taken in its own right. However, extended approaches can be developed in various ways and the one presented here is just one possible alternative.
  3. In Lecture Two, we discussed how Thompson (rightly) points out that these sorts of experiences do not warrant the ontological claim about the alleged existence of a purely non-material and disembodied self or a non-material world. Notice that the point made here, though, does not concern ontology (what there is), but agency (who acts), insofar as it takes these possibilities of experience as evidence for the fact that there is no one-to-one correspondence between physical individuality and agency.


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.