Original Article
The cultural morphospace of ritual form: Examining modes of religiosity cross-culturally

https://doi.org/10.1016/j.evolhumbehav.2010.09.002Get rights and content

Abstract

Ethnographic, historical, archaeological and experimental work suggests the existence of two basic clusters of ritual dynamics or ‘modes of religiosity’ — a low-frequency, high-arousal cluster linked to the formation of small cohesive communities (imagistic mode) and high-frequency, low-arousal cluster associated with larger, more centralized social morphology (doctrinal mode). Currently, however, we lack a large-scale survey of ritual variation on which to test such predictions. Here, we compile data on 645 religious rituals from 74 cultures around the globe, extracted from the Human Relations Area Files, revealing that the cultural morphospace of ritual form favours rituals that are indeed either low-frequency and highly dysphorically arousing or high-frequency with lower arousal and that these ritual dynamics are linked to group size and structure. These data also suggest that low dysphoric arousal, high-frequency rituals may have been tied to the advent of agriculture and subsequent emergence of the first large-scale civilizations.

Introduction

Religious rituals show an enormous diversity of form and function across the globe, but this diversity is not unbounded. Various clusters of ritual features are hypothesized to co-occur cross-culturally as a result of both cognitive constraints on the range of possibilities and functional constraints on how features interact with each other and the broader social system. For example, different elements of ritual form have been associated with costly signalling (Irons, 1996, Sosis & Alcorta, 2003, Sosis et al., 2007), obsessive compulsive disorder and the human hazard precaution system (Boyer & Lienard, 2006), cognitive constraints on memory systems (Whitehouse, 1992), the role ascribed to supernatural agency (McCauley & Lawson, 2002), modes of codification and transmission (Barth, 1987, Turner, 1974, Whitehouse, 1995), and the scale and structure of religious communities (Gellner, 1969, Goody, 1986, Werbner, 1977, Werbner, 1989, Whitehouse, 2000). To develop an understanding of the cultural evolution of religious rituals, these theories need to be tested in the light of evidence from a range of disciplines, including developmental and social psychology, cognitive neuroscience, behavioural economics and anthropology.

Central to this project is the characterisation of cultural variation itself. With few exceptions, (e.g., Sosis et al. 2007), the above theories tend to be derived from ethnographic, archaeological or historical case studies and field work, and are therefore vulnerable to the charge of cherry-picking: that is, focussing on bodies of evidence and experimental designs likely to confirm the theory. To avoid this criticism, a large-scale global database of ritual variation is needed. By systematically cataloguing cross-cultural variation in the dimensions of interest, hypotheses can be tested statistically and the cherry-picking criticism overcome. Such an undertaking is also in line with the recent suggestion by Hauser (2009) that, just as evolutionary biologists can gain insight into the forces shaping biological form by documenting the distribution of observed variants within a theoretical morphospace, social scientists can benefit from considering the morphospace of cultural forms. The existence of ritual clusters along certain dimensions can highlight possible adaptive peaks, perhaps favoured by cognitive biases, whilst gaps in the theoretical morphospace may suggest developmental, physical, historical or ecological constraints on the range of possible forms. These patterns can then feed back into the development of theory and serve to inform experimental work to establish the limits and drivers of human cultural variation.

Collecting cross-cultural data is time consuming and constrained by the available ethnographic evidence. A comprehensive catalogue of all the potentially interesting aspects of ritual variation is therefore not feasible in a single study. Here, instead, we take a first step towards characterising the cultural morphospace of ritual form by focussing on those elements of ritual relevant to the theory of ‘modes of religiosity’ (Whitehouse, 2004) — a relatively well-developed approach that applies cognitive theories to the comparative study of religious traditions to make a series of testable predictions about the sort of ritual variation we should expect to see.

The modes theory proposes that a crucial component of ritual variation arises from cognitive constraints imposed by episodic and semantic memory systems favouring two different cultural attractors or modes of religiosity — a ‘doctrinal’ mode and an ‘imagistic’ mode. The principal predictions of the theory concern the relationship between the frequency and emotionality of ritual performances. In order to maintain and motivate a cultural tradition, ritual elements must be preserved in the memory of participants. Put simply, the proposed doctrinal mode is seen as favouring high-frequency, low-arousal rituals, allowing large bodies of religious teachings to be stored in semantic memory, reproduced stably and spread efficiently as oral tradition. Conversely, the imagistic mode is seen as favouring low-frequency, high-arousal rituals that must evince strong emotions to etch themselves into episodic memory if they are to have an enduring impact on the religious imagination and to be remembered accurately. Research into the storage and vivid recollection of traumatic events in episodic memory (Conway,1995) suggests that the relationship between frequency and arousal should be particularly strong for negatively valanced ‘dysphoric’ arousal, involving intensely unpleasant, perhaps traumatic, experiences.

The two modes are also thought to be associated with other features of the religious system and society in which they occur (Whitehouse, 2004). The doctrinal mode is based around frequently repeated teachings and rituals. High-frequency ritual performances allow complex networks of ideas to be transmitted and stored in semantic memory and give rise to generic identity markers ascribed to large-scale ‘imagined communities’ (Anderson, 1983). Much of the religious knowledge is codified in language and transmitted primarily via recognized leaders and authoritative texts. At the same time, routinization is thought to suppress certain kinds of creative thinking about the meanings of the rituals, which may be a necessary condition for the establishment of religious orthodoxies. It is also argued that the emphasis on verbal transmission facilitates highly efficient and rapid spread, through processes of evangelism and missionization. The emphasis on oratory and learning is taken to facilitate the emergence of venerable leaders and teachers: gurus, prophets and priests. Hence, the doctrinal mode is also thought to encourage the emergence of centralized ecclesiastic hierarchies, exerting influence over the content and organization of authoritative religious knowledge.

By contrast, the imagistic mode of religiosity is based on rare, climactic rituals — for instance, the hair-raising ordeals of initiation cults, millenarian sects, vision quests, and so on — typically involving extreme forms of deprivation, bodily mutilation and flagellation, and psychological trauma based around participation in shocking acts. Such practices are known to trigger enduring and vivid episodic memories for ritual ordeals (Whitehouse, 1996), encouraging long-term rumination on the mystical significance of the acts and artefacts involved (Whitehouse, 2002). Imagistic practices are likely to be much harder to spread than doctrinal traditions. A major reason for this is that the religious knowledge is created through collective participation in rituals rather than being summed up in speech or text. Traumatic rituals should create strong bonds among those who experience them together, establishing in people's episodic memories who was present when a particular cycle of rituals took place. The expected tendency, then, is towards localized cults, based on patterns of following by example, and so it is not expected to generate the kind of scale, uniformity, centralization or hierarchical structure that typifies the doctrinal mode.

Beyond the religious system itself, ritual frequency and arousal are also thought to be linked to other social factors related to resource acquisition and control. Endeavours involving high risk and temptation to defect (e.g., military operations, hunting of dangerous animals, protection of moveable stores of wealth) are likely to recruit low-frequency, high-arousal rituals (e.g., hazing, painful initiation rites), both as a mechanism of bonding and perhaps as a form of costly signalling to demonstrate group commitment (Irons, 1996, Sosis & Alcorta, 2003, Sosis & Bressler, 2003, Sosis et al., 2007). By contrast, endeavours requiring regular input of relatively small but cumulatively large resources over much larger territories (e.g., exchange networks, tribute extraction, routinized manual labour, farming) are expected to recruit high-frequency, low-arousal rituals (e.g., ritualized expressions of nationalist pride, religious commitment, corporate identity) (Beardsley & McQuinn, 2009). In line with this dichotomy, recent work relating the modes theory to archaeological evidence from the early Neolithic site at Catalhoyuk suggests that the emergence of a doctrinal mode may have been associated with the increasing reliance on agricultural production (Whitehouse & Hodder, in press).

Whilst the modes theory has received considerable attention in the last decade from historians, archaeologists, anthropologists and cognitive scientists (Martin & Whitehouse, 2005, McCauley, 2001, McCauley & Lawson, 2002, McCauley & Whitehouse, 2005, Whitehouse & Martin, 2004, Whitehouse & McCauley, 2005) and there is much to support it in the ethnographic record (Ketola, 2002, Naumescu, 2008, Whitehouse & Laidlaw, 2004, Xygalatas, 2007), the above predictions about variation in ritual form have not yet been tested using a statistical framework and global sampling of cultures. Here, we construct a database recording frequency, arousal and contextual information for 645 religious rituals from 74 cultures around the globe, as documented in the electronic Human Relations Area Files (eHRAF). We seek to test whether some ritual features do in fact cluster together predictably along frequency and arousal dimensions and to investigate the cultural correlates of variation in ritual form. A crucial point to make about the hypotheses tested is that they are intended to be probabilistic rather than law-like. The two modes are envisaged not as a typology (nor even a specification of Weberian ideal types) to which religious traditions should be assigned. Rather, they constitute ‘attractor positions’ (Sperber, 1996, Whitehouse, 2004) around which the variables identified would be expected to cluster, given the availability of a sufficiently large sample of cases.

We therefore examine the following modes of religiosity hypotheses (Whitehouse, 2004):

  • 1.

    Ritual frequency should be negatively correlated with levels of emotional arousal.

  • 2.

    Ritual frequency should be most strongly negatively correlated with dysphoric (painful or unpleasant), rather than euphoric arousal, because of the effects of pain and trauma on the activation of durable episodic recall (Xygalatas, 2008).

  • 3.

    The morphospace or ritual frequency vs. arousal should reveal two clusters — high frequency and low arousal vs. low frequency and high arousal.

  • 4.

    Major world religions (e.g., classical religions and large regional religions) should have higher frequency and lower arousal rituals than small-scale or tribal traditions.

  • 5.

    Highly arousing rituals should be more prevalent in societies that live in smaller groups.

  • 6.

    Highly arousing rituals should be more prevalent in less hierarchical societies.

  • 7.

    Highly arousing rituals should be less prevalent in societies that rely more on agriculture.

An alternative, though not necessarily incompatible, set of explanations for variation in dysphoric ritual arousal is based on signalling theory (Irons, 1996, Sosis & Alcorta, 2003), since enduring pain or trauma can be a credible costly signal. Sosis et al. (2007) used cross-cultural data to test a number of potential signalling functions for male initiation rites and found that the most traumatic and painful male initiations, and those most likely to involve permanent scarring, occur in warring societies. We take advantage of our dataset to follow up this work on initiation rites by looking at the costly signalling hypotheses with regard to dysphoric arousal in rituals in general. Excluding those predictions that are relevant only to males, we sought to test three additional hypotheses based on costly signalling predictions outlined in Sosis et al. (2007):

  • 8.

    Dysphorically arousing rituals should be more prevalent in cultures that live in larger groups (because of the increased need to guard against free-riders in all cooperative activities in the group).

  • 9.

    Dysphorically arousing rituals should be more prevalent in cultures that rely more on foraging and less on agriculture (because increased likelihood of food scarcity requires sharing resources and is thus susceptible to free-riding).

  • 10.

    Dysphorically arousing rituals should be more prevalent in cultures with higher rates of intergroup warfare (because there is a greater need for group members to signal commitment to the in-group).

Finally, we consider the effect of two other variables that recent work has shown may be important in the evolution of religion. Belief in moralizing high gods — defined as active in human affairs and specifically supportive of human morality (Swanson, 1960) — is associated with some indices of societal cooperation (Johnson, 2005, Johnson & Kruger, 2004, Norenzayan & Shariff, 2008, Snarey, 1996, Roes & Raymond, 2003), perhaps due to a supernatural policing effect. Moral high gods may be linked to the doctrinal mode of religion, and to the extent that moral high gods represent a proxy for larger, more hierarchical religious systems, the modes theory would predict that societies with such gods have rituals that are on average lower arousal/higher frequency. Likewise, we consider the adoption of writing, which has also recently been linked to stages of religious evolution (Boyer, 2005, Goody, 2004, Sanderson & Roberts, 2008).

Section snippets

Methods

Ritual data were compiled from eHRAF, a searchable database of ethnographies for over 180 cultures selected to provide a diverse sampling from around the globe. We focussed on a subset of 65 of the most well-attested cultures in eHRAF known as the “Probability Sample Files” (PSF). Each of the PSF cases was randomly selected from one of a set of major culture areas around the globe (Ember & Ember, 1998, Lagace, 1979). In addition to providing some of the most complete ethnographic information,

Results

As predicted, ritual frequency is negatively correlated with overall ratings of arousal (rho=−0.401, n=644, p<.001; Fig. 2A). With the separation of euphoric and dysphoric arousal dimensions, ritual frequency is most strongly related to dysphoric arousal (rho=−0.409, n=644, p<.001), with a weak, though still statistically significant, correlation between frequency and euphoric arousal (rho=−0.08, n=644, p<.05). Fig. 2B shows that whilst dysphoric arousal increases monotonically with decreasing

Variation within societies

Our findings support the existence and utility of ‘cultural attractors’ in the morphospace of ritual form (Hauser, 2009, Sperber, 1985, Sperber, 1996). Consistent with the modes of religiosity theory, ratings of emotional arousal experienced by ritual participants are inversely correlated with ritual frequency, such that less frequent rituals tend to elicit higher emotional arousal ratings. This is the first demonstration of a correlation between ritual frequency and arousal from a global

Conclusion

We have shown that the morphospace of ritual frequency and arousal supports key predictions of the modes of religiosity theory. If correct, the theory reveals some of the main causes and consequences of religious diversity, past and present. Of course, the results presented here are necessarily correlational and so do not allow causal relationships to be tested directly. Additional experimental work is required to determine whether the proposed proximate psychological mechanisms are able to

Acknowledgments

We wish to thank Will Lowe and two anonymous reviewers for useful advice and comments on the manuscript. We also thank the following students for their invaluable help with the project: Tim Wilkinson, Matthew Ball, Rachel Chew, Emma Sharp and Emmanuelle Degliesposti.

References (76)

  • BrownR. et al.

    Flashbulb memories

    Cognition

    (1977)
  • PillemerD.B. et al.

    Flashbulb memories of menarche and adult menstrual distress

    Journal of Adolescence

    (1987)
  • RoesF.L. et al.

    Belief in moralizing gods

    Evolution and Human Behavior

    (2003)
  • SosisR. et al.

    Scars for war: Evaluating alternative signaling explanations for cross-cultural variance in ritual costs

    Evolution and Human Behavior

    (2007)
  • AndersonB.

    Imagined Communities: Reflections on the Origin and Spread of Nationalism

    (1983)
  • BarthF.

    Cosmologies in the Making: A Generative Approach to Cultural Variation in Inner New Guinea

    (1987)
  • BaumeisterR.F. et al.

    Bad is stronger than good

    Review of General Psychology

    (2001)
  • BeardsleyK. et al.

    Rebel groups as predatory organizations: The political effects of the 2004 Tsunami in Indonesia and Sri Lanka

    Journal of Conflict Resolution

    (2009)
  • BellA.V. et al.

    Culture rather than genes provides greater scope for the evolution of large-scale human prosociality

    Proceedings of the National Academy of Sciences USA

    (2009)
  • BellC.

    Ritual Theory, Ritual Practice

    (1992)
  • BlochM.

    Symbols, song, dance and features of articulation

    Archives Europèennes de Sociologie

    (1974)
  • BoyerP.

    A reductionistic model of distinct modes of religious transmission

  • BoyerP. et al.

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

    Behavioral and Brain Sciences

    (2006)
  • ChristiansenS.A. et al.

    Hysterical amnesia: A case of aversively motivated isolation of memory

  • ConwayM.A.

    Flashbulb Memories. Essays in Cognitive Psychology

    (1995)
  • DunbarR.I.M.

    Mind the bonding gap: Constraints on the evolution of hominin societies

  • EmberC.R. et al.

    Cross-cultural research

  • GellnerE.

    A pendulum swing theory of Islam

  • GelmanA. et al.

    Data Analysis Using Regression and Multilevel/Hierarchical Models

    (2006)
  • GoldP.E.

    A proposed neurobiological basis for regulating memory storage for significant events

  • GoodyJ.

    The Logic of Writing and the Organization of Society

    (1986)
  • GoodyJ.

    Is image to doctrine as speech to writing? Modes of communication and the origins of religion

  • HauserM.

    The possibility of impossible cultures

    Nature

    (2009)
  • HodderI.

    Çatalhöyük: The Leopard's Tale

    (2006)
  • HumphreyC. et al.

    The Archetypal Actions of Ritual. A Theory of Ritual

    (1994)
  • IronsW.

    Morality, religion, and evolution

  • JohnsonD.D.P.

    God's punishment and public goods: A test of the supernatural punishment hypothesis in 186 world cultures

    Human Nature

    (2005)
  • JohnsonD.D.P. et al.

    The good of wrath: Supernatural punishment

    Political Theology

    (2004)
  • JohnsonK.

    Primary emergence of the doctrinal mode of religiosity in prehistoric southwestern Iran

  • Ketola K. (2002) An Indian guru and his Western disciples: Representation and communication of charisma in the Hare...
  • LagaceR.O.

    The HRAF probability sample: Retrospect and prospect

    Behavior Science Research

    (1979)
  • LeachE.R.

    Political Systems of Highland Burma

    (1954)
  • LeDouxJ.E.

    The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction

  • Lewis-WilliamsD. et al.

    Inside the Neolithic Mind: Consciousness, Cosmos, and the Realm of the Gods

    (2005)
  • McCauleyR.N.

    Ritual, memory, and emotion: Comparing two cognitive hypotheses

  • McCauleyR.N. et al.

    Bringing Ritual to Mind Psychological Foundations of Cultural Forms

    (2002)
  • Cited by (0)

    This work was supported by the EC-FP6 EXREL project grant 43225 and a grant to Q.A. from the John Templeton Foundation-funded Cognition, Religion and Theology Project at Oxford University.

    View full text