Manvir Singh's target article outlines a possible account of widespread shamanistic beliefs. Singh must be commended for avoiding otiose terminological quibbles (there is a family resemblance here, so the challenge is to explain the recurrence of similar beliefs, rather than survey the denotation of “shamanism”) and for addressing the issue as one of cultural evolution informed by evolutionary theory. A focus on shamanism also serves as a reminder that, as far as we know, in the contexts in which humans evolved, the most widespread religious behaviors had no consistent doctrines, organizations of priests, fixed liturgy, communities of believers, or straightforward connection to moral prescriptions or prosocial attitudes.
To provide a serviceable hypothesis about the transmission of shamanism-related belief, Singh's model will require considerable additions and refinement with respect to the anthropological record, the psychological processes involved, and the mechanisms of cultural transmission, respectively.
As concerns the ethnographic database, there are many relevant variations within the family resemblance described by Singh (i.e., exceptional mental states and notion of superhuman agency associated with the management of misfortune and uncertainty). In particular, the ways in which people construe the special powers and activities of a shaman vary a lot (Vitebsky Reference Vitebsky1995b) among quasi-medical analogies (Gellner Reference Gellner1994); analogies to hunting and predator-prey interaction (Hamayon Reference Hamayon and Harvey2003); fights for the ownership of souls (Mallart Guimerà Reference Mallart Guimerà2003; Stepanoff Reference Stepanoff2014); attempts to reassemble dissociated souls (Crocker Reference Crocker1985); and many more. The point here is not that a model of shamanism should accumulate details and variants for the sake of documentation, but that observed variations in conceptions of shamans and their work may offer good terrain on which to test more precise hypotheses about the conditions of cultural transmission, as detailed below.
Regarding the underlying cognitive processes, Singh describes the situations using shamanism as those in which there is uncertainty about outcomes and no clear way to reduce that uncertainty. He also argues that people typically construe such outcomes in terms of agency (spirits, gods, souls, etc.) rather than general causal principles. The crucial connection between misfortune and superhuman agency, however, cries out for an explanation. In section 3.3.1., Singh mentions that “sociocognitive biases” are involved here; that, of course, is a label, not an explanation. One cannot blame Singh for not finding a solution to this difficult, unsolved problem, but it is crucial to recognize that this important part of the puzzle is missing.
To explain cultural transmission (explaining why these representations are recurrent in many places), Singh refers to “cultural evolution,” but there is no precise connection to models of cultural transmission. For example, in section 3.3.1., Singh alludes to prestige and conformity biases (Boyd & Richerson Reference Boyd and Richerson1985), which are indeed observed in many domains of transmission but may not be relevant here. The biases in question would require either (1) that shamanism actually provides good outcomes (so people keep practicing it, free-riding on the trial-and-error of previous generations, as Boyd and Richerson showed that they do for technology=conformity bias), or (2) that shamans are prestigious independently of their practice (so people imitate the shamanistic rituals as a way to achieve that social status=prestige bias). Neither is the case, however; shamanism is not efficacious, and shamans have social clout only to the extent that they are thought to be efficacious. Incidentally, note that, in many places, shamans are feared or despised (Hugh-Jones Reference Hugh-Jones, Thomas and Humphrey1996; Mallart Guimerà Reference Mallart Guimerà2003; Stepanoff Reference Stepanoff2014, p. 65).
Singh's own summary description of shamanism suggests that we are dealing with what cultural transmission theorists would call a cognitive attractor, that is, a particular point in conceptual space more likely than others to be represented in people's recurrent beliefs (Claidière et al. Reference Claidière, Scott-Phillips and Sperber2014; Claidière & Sperber Reference Claidière and Sperber2007). Specifically, what is observed here is a combination of (1) the assumption that concepts of superhuman agents are relevant to cases of uncertainty or misfortune, which is very common in many human societies; and (2) the assumption that demonstrations of publicly observable exceptional mental states indicate some special capacity in connection with those superhuman agents. This combination of assumptions was reinvented probably many times in human cultures. What makes it optimally relevant, such that people would tend to reconstruct it in such similar ways? As Singh points out, trance certainly suggests exceptional qualities. That is often interpreted in essentialist terms, suggesting that shamans are internally different from other people. We still do not know, however, why that difference would be associated with better access to superhuman agents or a better capacity to bridle them.
Perhaps ethnographic accounts may be of help here, for example, showing how people perceive the counterintuitive properties of trance and possession (an agent is both here and somewhere else, controlled and in control) as an index of mastery rather than passivity (Cohen Reference Cohen2007). More generally, it would seem that the model can be made much better if attention to cultural variants is systematically connected to a rich psychology, rather than the simple conjecture that people need reassurance against uncertainty and comfort in misfortune.
Manvir Singh's target article outlines a possible account of widespread shamanistic beliefs. Singh must be commended for avoiding otiose terminological quibbles (there is a family resemblance here, so the challenge is to explain the recurrence of similar beliefs, rather than survey the denotation of “shamanism”) and for addressing the issue as one of cultural evolution informed by evolutionary theory. A focus on shamanism also serves as a reminder that, as far as we know, in the contexts in which humans evolved, the most widespread religious behaviors had no consistent doctrines, organizations of priests, fixed liturgy, communities of believers, or straightforward connection to moral prescriptions or prosocial attitudes.
To provide a serviceable hypothesis about the transmission of shamanism-related belief, Singh's model will require considerable additions and refinement with respect to the anthropological record, the psychological processes involved, and the mechanisms of cultural transmission, respectively.
As concerns the ethnographic database, there are many relevant variations within the family resemblance described by Singh (i.e., exceptional mental states and notion of superhuman agency associated with the management of misfortune and uncertainty). In particular, the ways in which people construe the special powers and activities of a shaman vary a lot (Vitebsky Reference Vitebsky1995b) among quasi-medical analogies (Gellner Reference Gellner1994); analogies to hunting and predator-prey interaction (Hamayon Reference Hamayon and Harvey2003); fights for the ownership of souls (Mallart Guimerà Reference Mallart Guimerà2003; Stepanoff Reference Stepanoff2014); attempts to reassemble dissociated souls (Crocker Reference Crocker1985); and many more. The point here is not that a model of shamanism should accumulate details and variants for the sake of documentation, but that observed variations in conceptions of shamans and their work may offer good terrain on which to test more precise hypotheses about the conditions of cultural transmission, as detailed below.
Regarding the underlying cognitive processes, Singh describes the situations using shamanism as those in which there is uncertainty about outcomes and no clear way to reduce that uncertainty. He also argues that people typically construe such outcomes in terms of agency (spirits, gods, souls, etc.) rather than general causal principles. The crucial connection between misfortune and superhuman agency, however, cries out for an explanation. In section 3.3.1., Singh mentions that “sociocognitive biases” are involved here; that, of course, is a label, not an explanation. One cannot blame Singh for not finding a solution to this difficult, unsolved problem, but it is crucial to recognize that this important part of the puzzle is missing.
To explain cultural transmission (explaining why these representations are recurrent in many places), Singh refers to “cultural evolution,” but there is no precise connection to models of cultural transmission. For example, in section 3.3.1., Singh alludes to prestige and conformity biases (Boyd & Richerson Reference Boyd and Richerson1985), which are indeed observed in many domains of transmission but may not be relevant here. The biases in question would require either (1) that shamanism actually provides good outcomes (so people keep practicing it, free-riding on the trial-and-error of previous generations, as Boyd and Richerson showed that they do for technology=conformity bias), or (2) that shamans are prestigious independently of their practice (so people imitate the shamanistic rituals as a way to achieve that social status=prestige bias). Neither is the case, however; shamanism is not efficacious, and shamans have social clout only to the extent that they are thought to be efficacious. Incidentally, note that, in many places, shamans are feared or despised (Hugh-Jones Reference Hugh-Jones, Thomas and Humphrey1996; Mallart Guimerà Reference Mallart Guimerà2003; Stepanoff Reference Stepanoff2014, p. 65).
Singh's own summary description of shamanism suggests that we are dealing with what cultural transmission theorists would call a cognitive attractor, that is, a particular point in conceptual space more likely than others to be represented in people's recurrent beliefs (Claidière et al. Reference Claidière, Scott-Phillips and Sperber2014; Claidière & Sperber Reference Claidière and Sperber2007). Specifically, what is observed here is a combination of (1) the assumption that concepts of superhuman agents are relevant to cases of uncertainty or misfortune, which is very common in many human societies; and (2) the assumption that demonstrations of publicly observable exceptional mental states indicate some special capacity in connection with those superhuman agents. This combination of assumptions was reinvented probably many times in human cultures. What makes it optimally relevant, such that people would tend to reconstruct it in such similar ways? As Singh points out, trance certainly suggests exceptional qualities. That is often interpreted in essentialist terms, suggesting that shamans are internally different from other people. We still do not know, however, why that difference would be associated with better access to superhuman agents or a better capacity to bridle them.
Perhaps ethnographic accounts may be of help here, for example, showing how people perceive the counterintuitive properties of trance and possession (an agent is both here and somewhere else, controlled and in control) as an index of mastery rather than passivity (Cohen Reference Cohen2007). More generally, it would seem that the model can be made much better if attention to cultural variants is systematically connected to a rich psychology, rather than the simple conjecture that people need reassurance against uncertainty and comfort in misfortune.