Among many other insights into shamanism and supernatural belief, Singh has offered a useful decision tree for sorting different types of events and deciding when people are likely to develop superstitions. Superstition-prone events are those that are “uncontrollable, fitness relevant, and random” (sect. 3.1, last para.). I want to extend Singh's cultural evolutionary analysis to the nature of the superstitious beliefs themselves. It is one thing to explain when people tend to develop superstitions, and another to explain what form those superstitious beliefs will take. One dividend of such an analysis is that it helps to dispel the impression of charlatanism and insincerity in shamanistic traditions and, indeed, in religious traditions in general.
What sorts of supernatural powers do people attribute to shamans? Singh's classification of events provides a clue: People are unlikely to consult a shaman to bring about events that, though desirable and uncontrollable by natural means, will just never occur (e.g., preventing winter from coming). Why not? The reason is obvious. Belief in a magic ritual for halting the turn of the seasons would never be culturally stable, because it would lead invariably to disappointment. A similar point applies to beliefs about how to identify shamans. People might believe that a true shaman, when stabbed during a trance state, will not bleed, but they are unlikely to stab him in the heart or around the arteries.
This points to a problem that all supernatural beliefs – including those about the powers of shamans – have to confront in the real world: the potential destabilization of predictive failure. Now, unlike belief in an eternal summer, superstitions about events that are uncontrollable and random (from the limited epistemic perspective of believers) will at least result in occasional success. Even so, assuming that the interventions are causally innocuous (i.e., don't work), there will still be plenty of failures to account for. And though people are prone to confirmation bias, that does not mean they are immune to blatant refutations. How do beliefs in supernatural powers survive on a meager diet of confirmations and in the teeth of falsification?
Some misbeliefs are more resilient than others. Based on a cultural evolutionary framework, we can expect that overly fragile beliefs will be extinguished sooner or later, and that more resilient beliefs will survive. Beliefs can be resilient against falsification by providing more interpretive leeway to explain away failure. For example, rituals involving a relatively complex chain of steps are more resistant to falsification than straightforward ones. Given that rituals are causally opaque (Boyer & Bergstrom Reference Boyer and Bergstrom2008), their efficacy can be inferred only indirectly by observing the outcome. If the desired result fails to appear, that could mean the ritual doesn't work, but it could also mean that it was not carried out properly.
Because they contain more things that can go wrong, complex rituals provide more opportunities for retroactive ad hoc explanations (what Evans-Pritchard [Reference Evans-Pritchard1937] called “secondary elaborations”) to account for predictive failure, compared with more straightforward rituals. This line of reasoning may explain how rituals become more complex (Legare & Souza Reference Legare and Souza2012), as elements are added (or existing ones repeated) to explain why the ritual has failed on a particular occasion. Resilience is enhanced also by making the diagnosis more complex, as in the opening anecdote of Singh's article. A shaman kills a number of evil ghosts, but the patient dies anyway. Well, he laments, “in the end, the ghosts were too numerous.” Is this expressing fatalistic regret, as Singh put it, or is it better to say that the shaman is drawing a reasonable explanatory inference and might even change his procedure next time?
Beliefs also are better protected against refutation when they allow for subtle feedback loops between diagnosis and remedy (Boyer Reference Boyer1994, p. 144). For instance, shamans might try out different procedures until one appears to work and then use this observation to settle on a diagnosis. The same feedback loop can occur in the client's choice of shaman. If people believe that different shamans have different areas of expertise, they can shop around until they observe some improvement. In this way, the belief in shamanic powers itself is never threatened.
Finally, resilience is provided by the nature of the supernatural agents in question. In general, people are more likely to profess belief in supernatural agents that are capricious, moody, inscrutable, inattentive, and unpredictable, because such beliefs provide better resources for explaining failure. Gods who move in mysterious ways are culturally more successful than those who move in more reliable ways (Boudry & De Smedt Reference Boudry and De Smedt2011).
By appreciating how cultural evolution makes supernatural beliefs more resilient and less vulnerable to refutation, we can better understand the semblance of charlatanism. Skeptics have long asked rhetorically: Why don't people pray for an amputated limb to grow back if they really believe God is omnipotent? Why all the crutches and braces in the grotto of Lourdes and not a single wooden leg or glass eye (France Reference France1894)? When Meyer Fortes invited an informant to perform a rain dance for him, the man replied, “Don't be a fool, whoever makes a rain-making ceremony in the dry season?” (Tambiah Reference Tambiah1990, p. 54). A skeptic may ask, “Doesn't this show that it's all a sham?” Anthropologists and scholars of religion often express incredulity in the face of such incongruent behavior (Chaves Reference Chaves2010). When one looks at the evasive behavior of believers, the studious avoidance of potentially threatening observations, and the convenient resort to ad hoc explanations, one is left with an impression of insincerity (Boudry & Coyne Reference Boudry and Coyne2016; Humphrey Reference Humphrey1995).
But a cultural evolutionary framework helps to dispel this impression. Beliefs have evolved to become resilient and immune to refutation, which does not (necessarily) involve deliberate deceit or hypocrisy on the part of believers. The functional rationale of resilience is relocated on the level of the cultural representations themselves (Dennett Reference Dennett2006). If the cheesecake recipe has been fashioned by cultural evolutionary forces, there is no particular reason to doubt that the consumers genuinely enjoy it.
Among many other insights into shamanism and supernatural belief, Singh has offered a useful decision tree for sorting different types of events and deciding when people are likely to develop superstitions. Superstition-prone events are those that are “uncontrollable, fitness relevant, and random” (sect. 3.1, last para.). I want to extend Singh's cultural evolutionary analysis to the nature of the superstitious beliefs themselves. It is one thing to explain when people tend to develop superstitions, and another to explain what form those superstitious beliefs will take. One dividend of such an analysis is that it helps to dispel the impression of charlatanism and insincerity in shamanistic traditions and, indeed, in religious traditions in general.
What sorts of supernatural powers do people attribute to shamans? Singh's classification of events provides a clue: People are unlikely to consult a shaman to bring about events that, though desirable and uncontrollable by natural means, will just never occur (e.g., preventing winter from coming). Why not? The reason is obvious. Belief in a magic ritual for halting the turn of the seasons would never be culturally stable, because it would lead invariably to disappointment. A similar point applies to beliefs about how to identify shamans. People might believe that a true shaman, when stabbed during a trance state, will not bleed, but they are unlikely to stab him in the heart or around the arteries.
This points to a problem that all supernatural beliefs – including those about the powers of shamans – have to confront in the real world: the potential destabilization of predictive failure. Now, unlike belief in an eternal summer, superstitions about events that are uncontrollable and random (from the limited epistemic perspective of believers) will at least result in occasional success. Even so, assuming that the interventions are causally innocuous (i.e., don't work), there will still be plenty of failures to account for. And though people are prone to confirmation bias, that does not mean they are immune to blatant refutations. How do beliefs in supernatural powers survive on a meager diet of confirmations and in the teeth of falsification?
Some misbeliefs are more resilient than others. Based on a cultural evolutionary framework, we can expect that overly fragile beliefs will be extinguished sooner or later, and that more resilient beliefs will survive. Beliefs can be resilient against falsification by providing more interpretive leeway to explain away failure. For example, rituals involving a relatively complex chain of steps are more resistant to falsification than straightforward ones. Given that rituals are causally opaque (Boyer & Bergstrom Reference Boyer and Bergstrom2008), their efficacy can be inferred only indirectly by observing the outcome. If the desired result fails to appear, that could mean the ritual doesn't work, but it could also mean that it was not carried out properly.
Because they contain more things that can go wrong, complex rituals provide more opportunities for retroactive ad hoc explanations (what Evans-Pritchard [Reference Evans-Pritchard1937] called “secondary elaborations”) to account for predictive failure, compared with more straightforward rituals. This line of reasoning may explain how rituals become more complex (Legare & Souza Reference Legare and Souza2012), as elements are added (or existing ones repeated) to explain why the ritual has failed on a particular occasion. Resilience is enhanced also by making the diagnosis more complex, as in the opening anecdote of Singh's article. A shaman kills a number of evil ghosts, but the patient dies anyway. Well, he laments, “in the end, the ghosts were too numerous.” Is this expressing fatalistic regret, as Singh put it, or is it better to say that the shaman is drawing a reasonable explanatory inference and might even change his procedure next time?
Beliefs also are better protected against refutation when they allow for subtle feedback loops between diagnosis and remedy (Boyer Reference Boyer1994, p. 144). For instance, shamans might try out different procedures until one appears to work and then use this observation to settle on a diagnosis. The same feedback loop can occur in the client's choice of shaman. If people believe that different shamans have different areas of expertise, they can shop around until they observe some improvement. In this way, the belief in shamanic powers itself is never threatened.
Finally, resilience is provided by the nature of the supernatural agents in question. In general, people are more likely to profess belief in supernatural agents that are capricious, moody, inscrutable, inattentive, and unpredictable, because such beliefs provide better resources for explaining failure. Gods who move in mysterious ways are culturally more successful than those who move in more reliable ways (Boudry & De Smedt Reference Boudry and De Smedt2011).
By appreciating how cultural evolution makes supernatural beliefs more resilient and less vulnerable to refutation, we can better understand the semblance of charlatanism. Skeptics have long asked rhetorically: Why don't people pray for an amputated limb to grow back if they really believe God is omnipotent? Why all the crutches and braces in the grotto of Lourdes and not a single wooden leg or glass eye (France Reference France1894)? When Meyer Fortes invited an informant to perform a rain dance for him, the man replied, “Don't be a fool, whoever makes a rain-making ceremony in the dry season?” (Tambiah Reference Tambiah1990, p. 54). A skeptic may ask, “Doesn't this show that it's all a sham?” Anthropologists and scholars of religion often express incredulity in the face of such incongruent behavior (Chaves Reference Chaves2010). When one looks at the evasive behavior of believers, the studious avoidance of potentially threatening observations, and the convenient resort to ad hoc explanations, one is left with an impression of insincerity (Boudry & Coyne Reference Boudry and Coyne2016; Humphrey Reference Humphrey1995).
But a cultural evolutionary framework helps to dispel this impression. Beliefs have evolved to become resilient and immune to refutation, which does not (necessarily) involve deliberate deceit or hypocrisy on the part of believers. The functional rationale of resilience is relocated on the level of the cultural representations themselves (Dennett Reference Dennett2006). If the cheesecake recipe has been fashioned by cultural evolutionary forces, there is no particular reason to doubt that the consumers genuinely enjoy it.