Elsevier

Cognition

Volume 124, Issue 1, July 2012, Pages 1-15
Cognition

Evaluating ritual efficacy: Evidence from the supernatural

https://doi.org/10.1016/j.cognition.2012.03.004Get rights and content

Abstract

Rituals pose a cognitive paradox: although widely used to treat problems, rituals are causally opaque (i.e., they lack a causal explanation for their effects). How is the efficacy of ritual action evaluated in the absence of causal information? To examine this question using ecologically valid content, three studies (N = 162) were conducted in Brazil, a cultural context in which rituals called simpatias are used to treat a great variety of problems ranging from asthma to infidelity. Using content from existing simpatias, experimental simpatias were designed to manipulate the kinds of information that influences perceptions of efficacy. A fourth study (N = 68) with identical stimuli was conducted with a US sample to assess the generalizability of the findings across two different cultural contexts. The results provide evidence that information reflecting intuitive causal principles (i.e., repetition of procedures, number of procedural steps) and transcendental influence (i.e., presence of religious icons) affects how people evaluate ritual efficacy.

Highlights

► Rituals pose a cognitive paradox: although widely used they are causally opaque. ► How is ritual efficacy evaluated without causal information? ► Four studies examined reasoning about Brazilian rituals called simpatias. ► Simpatias were designed experimentally using ecologically-valid content. ► Intuitive causal principles affect how ritual efficacy is evaluated.

Introduction

“The problem of ritual is the familiar ‘rationality problem’ in a new guise—old wine in a new bottle” (Sax, 2010, p. 4).

Ritual is often interpreted in both popular scientific discourse and in ritual studies as action that is ineffective, irrational, or purely conventional (Sax, Quack, & Weinhold, 2010). Although some have argued that rituals are expressions of inner states of feeling and emotion, symbolize theological ideas or social relations, or represent psychophysical states, conceptualizing ritual exclusively in this way neglects the fact that the use of rituals for protective, restorative, and instrumental purposes is a pervasive feature of human culture (Sax et al., 2010, Sørensen, 2007). Using rituals to solve problems presupposes reasoning about their efficacy, a topic of longstanding interest and debate in anthropology (Csordas, 2002, Sax, 2004, Sax et al., 2010).

Rituals pose a cognitive paradox: although widely used to treat problems, they are cultural conventions and lack a causal explanation for their effects (Legare & Whitehouse, 2011). They are the result of “a positive act of acquiescence in a socially stipulated order”, and thus are not the product of individual innovation. “The peculiar fascination of ritual lies in the fact that here, as in few other human activities, the actors both are, and are not, the author of their acts” (Humphrey & Laidlaw, 1994, p. 5). Rituals, which we define as conventional, causally opaque procedures (Legare & Whitehouse, 2011), present a challenge to theoretical accounts of causal reasoning because they are both socially stipulated (Humphrey & Laidlaw, 1994) and not reducible to causal mechanisms (Bloch, 2004, Boyer and Liénard, 2006, Whitehouse, 2001). Even when rituals are explained in the context of a certain belief, there is often not an expectation of a direct causal connection between (ritual) actions and outcomes (Sørensen, 2007). We propose that rituals are irretrievably causally opaque because they (1) are not bound by the same kinds of intuitive physical–causal constraints that characterize non-ritualistic actions and (2) lack an intuitive causal connection between the specific action performed (e.g., rubbing a ceramic pot) and the desired outcome or effect (e.g., making it rain). For example, rituals intended to have particular effects (e.g., rituals promoting crop fertility or healing the sick) are not expected to do so by causal mechanisms that are transparent or even in principle knowable (Legare & Whitehouse, 2011). This raises a conceptual question: how do people evaluate the efficacy of ritual action in the absence of causal information?

Our objective is to examine the ‘hidden logic’ of ritual (Sax, 2010) experimentally, integrating and applying cognitive anthropological and cognitive psychological approaches to the study of ritual cognition. Rather than evaluate the efficacy of ritual by examining outcomes or experience (Csordas, 2002), we seek to examine the kinds of information that influence perceptions of the efficacy of ritual action.

We propose that the structure of ritual can be interpreted in light of intuitive causal beliefs about action efficacy or potency. In particular, rituals used for problem-solving purposes reflect intuitive beliefs about causal reasoning and the efficacy of goal-directed action sequences.

Consider Tambiah’s (1979) classic definition of ritual as practice: “Rituals are patterned and ordered sequences of words and acts, often expressed in multiple media whose content and arrangement are characterized in varying degrees by formality (conventionality), stereotypy (rigidity), condensation (fusion), and redundancy (repetition)”. We argue that the characteristics of ritual described by Tambiah (1979; i.e., rigidity, repetition) are the product of an evolved cognitive system (Atran and Norenzayan, 2004, Boyer and Liénard, 2006, Sørensen, 2007, Humphrey and Laidlaw, 1994, Whitehouse and McCauley, 2005) of intuitive causal principles. Rather than conceptualize ritual as a process of intensive symbolic communication (Tambiah, 1979), we suggest that the process of ritualization tends to evacuate actions of meaning through goal-demotion and redundancy (Humphrey & Laidlaw, 1994). Thus, we predict that intuitive causal reasoning, not content familiarity, is driving how ritual efficacy is evaluated.

Although there is written record of rituals used for problem-solving purposes dating from ancient Egypt (The papyrus ebers, 1931, The papyrus ebers, 1937) the use of rituals to treat problems as diverse in etiology as asthma and unemployment is widespread in contemporary cultural contexts such as the United Kingdom (Hutton, 1999), the United States (Crowley, 1989), Brazil (Cohen and Barrett, 2008, Souza and Legare, 2011), and South Africa (Ashforth, 2001, Legare and Gelman, 2008). Despite the seeming variability in the content, practices, and artifacts used in rituals around the world and over historical time, we propose that the way in which ritual efficacy is evaluated is predictable and constrained. For example, compare the following rituals used as remedies in Ancient Egypt and in present day Brazil. First consider this ritual, taken from the Papyrus Ebers, 1550 BCE, that was used to treat blindness: “Crush, powder, and make into one the two eyes of a pig [remove the water therefrom], true collyrium (i.e., mineral eye salve), red-lead (i.e., red oxide), and wild honey [in a clay bowl]. Inject [mixture] into the ear of the patient. When thou hast seen properly to this mixing repeat this formula: ‘I have brought this thing and put it in its place. The crocodile [god Sobek] is weak and powerless’. Repeat twice. Thereby he will at once recover” (The Papyrus Ebers, 1931, p. 104).

Now consider a ritual used to find a partner in Brazil: “Buy a new sharp knife and stick it four times into a banana tree on June 12th at midnight (i.e., Valentine’s day in Brazil, Saint Anthony’s day is on the 13th). Catch the liquid that will drip from the plant’s wound on a crisp, white paper that has been folded in two. The dripping liquid captured on the paper at night will form the first letter of the name of your future partner” (Scharf, 2010).

On the surface, there are many differences between these rituals. They involve different substances (e.g., red-lead vs. sap from a banana tree), different practices (i.e., mixing vs. paper folding), incorporate different artifacts (i.e., clay bowl vs. a knife), and treat different problems (i.e., blindness vs. attracting a partner). Yet, there are also many similarities. They involve information such as procedural repetition (i.e., repeat twice vs. twice a day for two weeks), a large number of procedural steps (i.e., seven vs. six), time specificity (i.e., early rising vs. June 12th at midnight), high levels of procedural detail (i.e., mixing wild honey vs. buying a new sharp knife and sticking it four times into a banana tree), and the presence of supernatural agents (i.e., Sobek, an ancient Egyptian deity vs. Saint Anthony, a Catholic marriage saint).

We hypothesize that information reflecting intuitive biases in causal reasoning (i.e., repetition, number of procedural steps, and the specificity of procedural detail) is used to evaluate the efficacy of ritual action. Although biases in causal reasoning are used to evaluate the efficacy of all action, their influence on action efficacy judgments may be especially salient or influential when information about causal mechanisms is unavailable. Whereas some of the intuitive causal principles hypothesized to influence perceptions of ritual efficacy examined in the present studies are likely to be related to previously documented biases in causal reasoning (i.e., repetition), others have not been well studied (i.e., number of procedural steps and specificity of procedural detail).

We propose that repetition of similar actions (e.g., pressing a button repeatedly to call an elevator) is perceived to be causally efficacious. A long-standing philosophical tradition supports the claim that beliefs about causal connections arise from impressions (projections of the mind) of repeated instances of similar relations (Hume, 1740). Converging psychological research has demonstrated that repetition may also influence reasoning about a variety of behaviors by making information more psychologically available (Oppenheimer, 2008), familiar (Scott & Dienes, 2008), and attractive (Zajonc, 1968).

The number of procedural steps and procedural specificity of the action sequence may also influence perceptions of causal efficacy. A larger number of procedural steps (e.g., seven steps) may increase the perception of causal efficacy over a smaller number of procedural steps (e.g., three steps) by giving the impression that multiple actions may have the capacity to produce a particular effect. The specificity of the action sequence (i.e., inclusion of specific, detailed information) might also influence perceptions of action efficacy. Given that human beings are expert intention-readers, seeing someone engaging in a detailed course of actions (e.g., catching the liquid that will drip from the plant’s wound on a crisp, white paper that has been folded in two) may give the impression that particular details of the action sequence (i.e., time specificity, item specificity) has the potential to produce the desired, intended outcome, even if the mechanism is unknown or unavailable (Legare & Whitehouse, 2011).

In addition to intuitive causal principles, supernatural agents may play a role in the evaluation of ritual efficacy (Sax et al., 2010). Indeed, images and icons of supernatural agents are frequently used in rituals (Cohen, 2007, Souza and Legare, 2011, Whitehouse, 2004) and thus intuitions about ritual efficacy may invoke the involvement of a supernatural agent at some level in the ritual sequence (Barrett and Lawson, 2001, Sørensen et al., 2006). If supernatural agency is involved in the ritual efficacy evaluation process, ritualistic actions may not be constrained by the same physical–causal expectations as non-ritualistic actions (Barrett and Malley, 2007, Boyer, 2001).

An additional objective of this research is to examine the extent to which evaluating ritual efficacy generalizes to believers and nonbelievers as well as to cultural groups unfamiliar with the content of particular ritualistic belief systems. There may be important differences in reasoning about ritual between those that endorse particular supernatural worldviews (believers) and those that do not (nonbelievers or members of other cultural contexts unfamiliar with particular ritualistic practices). For instance, social psychological research on differences in reasoning about existentially arousing phenomena indicates that the awareness of mortality reliably increases the tendency to believe in supernatural agency for those that espouse religious beliefs (Norenzayan & Hansen, 2006). When mortality and cultural worldview threats are primed, nonbelievers are more likely than believers to denigrate a culturally threatening message, possibly because believers respond to existential threats with a religious stance that transcends their secular cultural identity (Norenzayan, Dar-Nimrod, Hansen, & Proulx, 2009).

The objective of the current studies was to investigate the intuitive causal principles that are used to evaluate the efficacy of ritualistic action. To give empirical traction to this topic using ecologically valid content, data were collected in Brazil, a cultural context in which rituals or “recipes” – called simpatias – are available, endorsed, and used for everyday problem-solving purposes. Simpatias are ritualistic remedial procedures, and are not confined to any particular Brazilian religious group, although some of them do include religious information. They are used to solve a variety of everyday problems (e.g., sinusitis, asthma, depression, anxiety, lack of luck, and infidelity). Simpatias are available to the general population, are relatively low-cost, and do not require any specialized expertise to be performed.

We propose that studying how simpatias are evaluated from a cognitive perspective speaks directly to the general question of how people evaluate ritual efficacy, and thus provides a unique opportunity to use ecologically valid content to investigate reasoning about causally opaque events in particular and ritual cognition in general. An important aspect of cognitive science research involves weaving together different methods from different disciplines. An ecologically valid paradigm maximizes the best possible trade-off between internal and external validities (Markman, Beer, Grimm, Rein, & Maddox, 2010). The use of culturally meaningful content to create our experimental stimuli optimizes this trade-off for this research; our methods and materials thus closely approximated the real-life practices under investigation. We consider this innovative interdisciplinary methodology to be a core contribution of this research.

We hypothesized that information that reflects the defining characteristics of ritual such as frequency of repetition, number of procedural steps, and specificity of procedural detail will increase ritual efficacy evaluation because information of this kind reflects intuitive beliefs about causal potency. We also predicted that the presence of supernatural agents would impact the evaluation of ritual efficacy (Barrett and Lawson, 2001, Sørensen et al., 2006, Souza and Legare, 2011). Study 1 was designed to develop and assess the ecological validity of our experimental stimuli. Study 2 examined potential kinds of intuitive information or criteria that may influence how ritual efficacy is evaluated. Study 3 provided a more systematic investigation of the intuitive criteria that were found to influence the evaluation of ritual efficacy in Study 2 and explored the impact of these criteria on reasoning about ritual efficacy among believers. Study 4 examined the extent to which the findings from Studies 1–3 represent universal features of human cognition, using identical stimuli in a cultural context unfamiliar with these ritualistic practices (US sample).

Section snippets

Study 1

In order to explore the extent to which intuitive criteria influence evaluations of ritual efficacy, simpatias were designed experimentally (modeled after content and information available in real and widely available simpatias). The objective was to isolate and experimentally manipulate the kinds of information hypothesized to impact judgments of ritual efficacy. The objectives of Study 1 were (1) to identify the kinds of information that are widely available in simpatias in general in order

Study 2

The objective of Study 2 was to investigate the extent to which the frequently occurring kinds of information identified in Study 1 (i.e., the nine criteria) influence the evaluation of ritual efficacy. We hypothesized that information consistent with intuitive beliefs about causal efficacy (i.e., frequency of an action, specificity of the action sequence) would increase ritual efficacy evaluation to a greater extent than familiar information (i.e., information that is frequently found in

Study 3

The objective of Study 3 was to examine more systematically the criteria identified in Study 2 that were found to impact evaluations of ritual efficacy. We created multiple simpatias per criterion (six simpatias per criterion in total, three per condition) in addition to modifying the content of the simpatias to more carefully control for complexity, ecological validity, and word length (see Appendix B). We also sought to examine the influence of belief on the evaluation of ritual efficacy. We

Study 4

To investigate whether the findings from Studies 1–3 generalize to a different cultural context, we systematically investigated the extent to which information about specificity of time, repetition of procedures, number of procedural steps, and presence/absence of a religious icon impacts ritual efficacy ratings in a cultural group unfamiliar with these ritualistic practices (US adults). As in Study 3, we directly compared the mean efficacy ratings of simpatias that differed only in the content

General discussion

One of the most remarkable characteristics of human cognition is the capacity to reason about the causal relationships and mechanisms that explain the world around us (Ahn and Kalish, 2000, Carey, 2009, Gopnik and Schulz, 2007, Keil and Wilson, 2000). Understanding causal reasoning is of long-standing interest in both psychological and anthropological disciplines (Boyer, 1995, Sperber et al., 1995) due in large part to the fact that despite substantial cross-cultural variability in the content

Acknowledgements

We are grateful for funding from the James S. McDonnell Foundation Causal Learning Collaborative, the Fell Foundation at the University of Oxford, and a grant from the Cognition, Religion, and Theology Program funded by the John Templeton Foundation. We would like to thank Jacqueline Woolley, Patricia Herrmann, and an anonymous reviewer for helpful comments on previous versions of the manuscript and Brooke Wooley for assistance with transcription.

References (48)

  • D.M. Oppenheimer

    The secret life of fluency

    Trends in Cognitive Science

    (2008)
  • W. Ahn et al.

    The role of mechanism beliefs in causal reasoning

  • A. Ashforth

    An epidemic of witchcraft? The implications of AIDS for the post-apartheid state

  • S. Atran et al.

    Religion’s evolutionary landscape: Counterintuition, commitment, compassion, communion

    Behavioral and Brain Sciences

    (2004)
  • R. Baillargeon et al.

    How do infants reason about physical events?

  • J. Barrett

    Smart gods, dumb gods, and the role of social cognition in structuring ritual intuitions

    Journal of Cognition and Culture

    (2002)
  • J. Barrett et al.

    Ritual intuitions: Cognitive contributions to judgments of ritual efficacy

    Journal of Cognition and Culture

    (2001)
  • J. Barrett et al.

    A cognitive typology of religious actions

    Journal of Cognition and Culture

    (2007)
  • M. Bloch

    Bringing ritual to mind: Psychological foundations of cultural forms

    Journal of the Royal Anthropological Institute

    (2004)
  • P. Boyer

    Religion explained

    (2001)
  • P. Boyer et al.

    Why ritualized behavior? Precaution systems and action-parsing in developmental, pathological and cultural rituals

    Behavioral and Brain Sciences

    (2006)
  • P. Boyer et al.

    Ritual behavior in obsessive and normal individuals: Moderating anxiety and reorganizing the flow of behavior

    Current Directions in Psychological Science

    (2008)
  • P. Boyer

    Causal understandings in cultural representations: Cognitive constraints on inferences from cultural input

  • S. Carey

    The origin of concepts

    (2009)
  • E. Cohen

    The mind possessed: The cognition of spirit possession in Afro-Brazilian religious tradition

    (2007)
  • E. Cohen et al.

    When minds migrate: Conceptualizing spirit possession

    Journal of Cognition and Culture

    (2008)
  • V. Crowley

    Wicca: The old religion in the new age

    (1989)
  • T.J. Csordas

    Body/meaning/healing

    (2002)
  • Hume, D. (1740). A treatise of human nature (1967 ed.). Oxford, UK: Oxford University...
  • C. Humphrey et al.

    The archetypal actions of ritual

    (1994)
  • R. Hutton

    The triumph of the moon: A history of modern pagan witchcraft

    (1999)
  • F.C. Keil

    Science starts early

    Science

    (2011)
  • Cited by (0)

    View full text