Speaker series
Evolution and the Social Mind
Speaker Series
How cultural representations spread:
Experimental evidence on religious concepts and evolved ontologies
A Presentation by Pascal Boyer
CNRS, Paris and University of Lyon
January 22, 1999 at 12:30 - 2:00pm
Humanities and Social Sciences Building 2001A
How can we come to understand and scientifically investigate culture, and relate it to minds and brains?  Ideas or representations spread from individual to individual, and when they become distributed widely enough in populations, we call them "cultural."   Whether an idea spreads easily and widely depends, in part, on the evolved architecture of the human mind, as well as on social factors.   Human universal cognitive mechanisms will cause some ideas to be easy to comprehend, interesting or memorable, while rendering others abstruse, dull, or difficult to acquire or recall.  There is growing evidence that very young children are designed to reliably develop a series of richly structured, neurally based reasoning circuits, including specializations for reasoning about objects, artifacts, animals, mental states, and contagion.  These impose cross-culturally recurrent conceptual orderings on certain domains of thought and discourse - orderings that are sometimes now called evolved ontologies.

Boyer has been applying research from cognitive development and cognitive neuroscience on domain-specific inference mechanisms to understanding the spread of religious ideas.  He has proposed that certain religious ideas spread as a side-effect of the design of these mechanisms. This explains why certain aspects of religious ideas are found in a strikingly similar form in so many different cultural environments. Religious ideas are a particularly good arena in which to see the projective effects of the human mind on cultural contents, because in the case of religious ideas, there are no actual external referents (e.g., ghosts, magic, and other supernatural actors) contributing to the nature of the representations.

Boyer has derived a series of testable predictions about the kinds of religious ideas that will spread widely, and about what aspects of religious ideas make them memorable.  He has most recently been testing the memory predictions both in the lab, and in fieldwork in Gabon and Nepal.

Pascal Boyer is an anthropologist and cognitive scientist at the C.N.R.S. and the University of Lyon, and an internationally known researcher into the anthropology of religion.  He is author of The Naturalness of Religious Ideas: A cognitive theory of religion. He spent last year at the Center for Advanced Study in the Behavioral Sciences, completing a new book.


Boyer, Pascal. The naturalness of religious ideas: A cognitive theory of religion. University of California Press; Berkeley, CA, US, 1994.

Abstract: (from the jacket) The main theme of Pascal Boyer's work is that important aspects of religious representations are constrained by universal properties of the human mind-brain.... The transmission of religious representations, as Boyer points out, does not occur in a cognitive vacuum. There is growing evidence that human minds are predisposed to acquire certain types of mental representations. In particular, experimental psychology shows that a number of universal, richly structured, early developed conceptual principles organize our understandings of particular aspects of natural and social environments. These representations in turn constrain the range of religious representations humans are likely to acquire, memorize, and transmit. This explains why certain aspects of religious ideas are found in a strikingly similar form in so many different cultural environments. The book will be widely discussed by cultural anthropologists and psychologists, as well as students of religion, history, and philosophy.

Contents: Preface. Part one: Religious ideas as conceptual structures. Recurrence, naturalness, and under-determination. The varieties of religious representations. Two aspects of conceptual structures. Part two: Four repertoires of religious representations. Natural ontologies and supernatural furniture. Causal judgments. Essentialism and social categories. Ritual episodes and religious assumptions. Part three: A sketch of cultural transmission. Cross-strengthening, religious truth, and stability. Cultural transmission and the biology in history. References. Index.

Select Articles and Book Chapters

Boyer, Pascal. Further distinctions between magic, reality, religion, and fiction. Child Development, 1997 Dec, v68 (n6):1012-1014.

Abstract: Comments on Jacqueline D. Woolley's "Thinking about fantasy: Are children fundamentally different thinkers and believers from adults?" in the same issue of Child Development, pp. 991-1011. It is suggested that children's representations of counterintuitive phenomena can be better understood if we take into account the following: (1) Children may develop a conceptual slot for "counterintuitive + real" phenomena. (2) Notions of "reality" in early childhood are linked to experience rather than ontological status. (3) We have no good description of children's handling of fiction. (4) Cultural systems of religious representations make particular demands on developmental processes. It is argued that Woolley's survey is very helpful as a detailed account of work done in the "reality-fantasy" frame.

Boyer, Pascal. Cognitive limits to conceptual relativity: The limiting-case of religious ontologies. IN: Rethinking linguistic relativity. Studies in the social and cultural foundations of language, No. 17.; John Joseph Gumperz, Stephen C. Levinson, Eds. Cambridge University Press, Cambridge, England. 1996. p. 203-231.

Abstract: This chapter questions some tenets which Boyer believes lie behind the following "Whorfian" argument: categorization requires grouping by similarity, this categorization is under-determined by objective reality, but is suggested by linguistic categories, which therefore play a crucial role in the transmission of culture across generations. He argues that categorization or conceptual grouping is really determined by an implicit theory about a particular domain: for example, animate objects are grouped together by their teleological behavior (for which there is a pan-human naive theory) and distinguished by properties of each natural kind (for which, again, there is a naive theory of essence and reproduction); such domain-specific implicit theories are not, he claims, learned, and therefore are biologically, not culturally, transmitted. Anthropologists, Boyer argues, are blinded by a few spectacular oddities to the enormous undercurrent of universal conceptual assumptions. He suggests that religious ideas have a reliable crosscultural structure: metaphysical beings are ascribed all the properties specific to the human domain (with its associated naive theory of intentional, purposeful behavior) with just a few outlandish properties--e.g. invisibility, omniscience, etc.; suggests that even the outlandish elements conform to limited types, which guarantee their memorability and hence their cultural transmission.

Boyer, Pascal. Causal understandings in cultural representations: Cognitive constraints on inferences from cultural input. IN: Causal cognition: A multidisciplinary debate. Symposia of the Fyssen Foundation. Edited by Dan Sperber, David Premack, and Ann James Premack. Clarendon Press/Oxford University Press, New York, NY, US. 1995. p. 615-649.

Abstract: The chapter focuses on a domain which has always been central to anthropological discussions of causation and causal concepts, that of religious 'magical' assumptions; religious concepts and assumptions seem to display obvious cross-cultural variation, perhaps in a more salient way than other types of cultural representations. Anthropological and psychological data are used to provide an answer to a series of 5 questions, with each answer being the starting point for the next; are concepts of causation culturally specific; what is the structure of early causal understandings; what is the structure of the causal understandings implied by religious categories; what is the role of early intuitive principles in the treatment of the cultural input; what does this tell us about causal understandings in culture. The chapter includes a discussion among F. Keil, A. Leslie, G. Lloyd, L. Talmy and P. Boyer.

Boyer, Pascal. Cognitive constraints on cultural representations: Natural ontologies and religious ideas. IN: Mapping the mind: Domain specificity in cognition and culture. Lawrence A. Hirschfeld, Susan A. Gelman, Eds. Cambridge University Press, New York, NY, US. 1994. p. 391-411.

Abstract: The point of a cognitive approach to cultural representations is to put forward a series of causal hypotheses in order to account for certain features of cultural phenomena; show to what extent this framework can help reformulate classical anthropological problems; dispel certain conceptual ambiguities that are pervasive in the anthropological literature, notably as concerns (1) the subjective "unnaturalness" of religious assumptions, (2) their cognitive diversity, and (3) the extent to which they depend on cultural transmission. Boyer argues that if cognitive hypotheses are relevant in the explanation of religious ideas, then other aspects of cultural representations will be a fortiori amenable to such a description.

 Top of page
Speaker series
Evolution and the Social Mind
Speaker Series