Chapter 3 introduction

Posted on August 10, 2010


Adequate accounts of what magic and religion have in common as well as what the differences between them are have proved as elusive as the entities they refer to. Even so, it is in search of such accounts that this chapter sets out. Given awareness of the earlier failures, however, the focus will not be upon putting forward cut and dried definitions. Instead, the aim will be much more limited. Treating magic and religion as phenomena explainable in cognitive, cultural and, ultimately, evolutionary terms, the aim will be to present in outline the relevant mechanisms that underlie these particular kinds of beliefs and practices. As such, the proposals put forward here build upon the work of researchers such as Pascal Boyer but focus upon the epistemic aspects that have not been overtly present in much of the cognitive science of religion literature. The point, as through this book, will be to relate our understanding of religion and magic to our understanding of science but without relying upon the simplistic dichotomies of the past.

A common, though problematic way of thinking of magic and religion is as dealing with supernatural entities, where the supernatural is defined as that which cannot be investigated by science. This naive view exemplifies the very dichotomy between reason and superstition that is the focus of my investigations, yet it is ultimately worth examining as even refuted positions often contain insights that prove valuable in arriving at a richer understanding of the phenomenon. In the case of the category of the supernatural I focus on three issues. The first is that this way of defining the supernatural is typically combined with the attempt to define science as the investigation of natural phenomena – the circularity in the two definitions leading to a tautology. The second is that the definition fails to fit the phenomena that are normally classified as supernatural since many of them are subject to scientific investigation. Finally, the third issue is that the definition appears to assume that it is possible to indicate an unchanging set of limits beyond which science can never progress – a claim that does not stand up to scrutiny.

Rejecting the idea that religion and magic can be defined as that which science cannot grasp, however, does not entail that the issue of constraints upon empirical investigation is irrelevant to understanding religion and magic, as comes to be seen once discussion turns to the question of the factors that can stabilise particular beliefs within a culture. Working on the basis of Pascal Boyer’s account of religious beliefs as involving minimally counterintuitive concepts it is possible to generalise it to saying that religious and magical beliefs are mainly stabilised by cognitive and cultural factors. This way of putting the cognitive byproduct thesis makes it easy to compare such beliefs with those that are mainly stabilised by empirical considerations. Recognising that some beliefs are stabilised by some of these factors more than others leads to the question of what determines which factors are the most relevant to particular beliefs.

I go on to identify three general constraints that determine a belief’s openness to investigation. The most obvious one is the content of the belief in question since claims that are hard to verify are less open to empirical investigation. Just as important, however, are constraints arising due to the context in which the belief finds itself. The first of these is the social context, i.e. the various social attitudes that may make it more or less likely for the belief to undergo investigation. The second is the methodological context, i.e. the set of conceptual and technological tools available for investigating the belief. Religious/magical beliefs frequently have all three of these constraints on investigation acting to protect them and, thereby, ensuring that their content is determined by the cognitive and cultural factors.

It is this causal link between religion/magic and being effectively uninvestigable that is ultimately responsible for the attractiveness of the idea that the supernatural is, by definition, not open to scientific investigation. As I go on to show, it also lies behind Durkheim’s conception of religion/magic as sacred, since being treated as sacred gives the claims of religion and magic special social status that renders them less likely to be investigated. The resulting account of religious/magical beliefs is that they are: rendered stable within culture by cognitive and cultural considerations, while being protected against potential destabilising empirical factors. While this does not provide a definition of religion/magic it does serve to broadly distinguish it from other human beliefs that are maintained by different mechanisms.

Having thus characterised what it is that religion and magic have in common it comes time to examine the difference between them. The difference I will argue for is that while magical practices are intended to have mundane effects, the religious practices are meant to affect the world of the supernatural entities and forces. At the same time, both magical and religious practices share the property that their purported effectiveness is explained by a causal connection that is supernatural in nature. The difference is vital in that it puts religious and magical practices in very different epistemic positions. The efficacy of magical practices remains potentially investigable where the social and methodological contexts allow it. This means that magical beliefs are being maintained in the face of potential counterevidence. Religious practices, however, have purported effects whose very content renders them effectively uninvestigable. The price is that no experiences may be interpreted as showing the effectiveness of religious practices – unless the practices take on a magical character.

This way of understanding religious and magical practices bears immediate fruit in the way that it helps to understand why such practices have a strong tendency to become ritualised – the essential point that behaviour becomes stereotypical when cues to its effectiveness are unavailable.