Religion is natural149
© 2007 The Author. Journal compilation © 2007 Blackwell Publishing Ltd.
U N C O R R E C T E D P R O O F
religious belief are cultural inventions, created by adults.A complete developmental account of the growth of religious belief, then, would be one of cultural learning.There is a growing body of work, however, that sug-gests that this is not entirely right. Instead, while cultureplainly plays some role, some of the universals of religion are unlearned (see Atran, 2004; Barrett, 2004;Bering, in press; Boyer, 2001; Bloom, 2004, 2005; Evans,2000, 2001; Guthrie, 1993; Kelemen, 2004; Pinker, 1997).There are two main threads of this argument.
1. Common-sense dualism
It is not controversial that young children naturallymake sense of physical entities in different terms thanpsychological entities: naïve physics is different fromnaïve psychology. The claim explored here is considera-bly stronger. It is the idea that we think of bodies andsouls as
; we implicitly endorse a strong sub-stance dualism of the sort defended by philosophers likePlato and Descartes. Under one variant of this account,our dualism is a natural by-product of the fact that wehave two distinct cognitive systems, one for dealing withmaterial objects, the other for social entities. Thesesystems have incommensurable outputs. Hence dualismemerges as an evolutionary accident (Bloom, 2004).Dualism has interesting consequences. If bodies andsouls are thought to be separate, you can have one with-out the other. Most things, such as chairs, cups, andtrees, are thought of as bodies without souls, not pos-sessing goals, beliefs, will, or consciousness. More signi-ﬁcant for religion, dualism makes it possible to imaginesouls without bodies. Christianity and Judaism, forinstance, involve a God who created the universe, per-forms miracles, and listens to prayers. He is omnipotentand omniscient, possessing inﬁnite kindness, justice, andmercy. But he does not, in any literal sense, have a body.Our dualism also opens the possibility that people cansurvive the death of their bodies. Religions provide dif-ferent accounts as to the fate of the soul: It might ascendto heaven, descend to hell, go off into some sort of parallelworld, or occupy some other body, human or animal.Indeed, a belief that the world teems with ancestorspirits, the souls of people who have been liberated fromtheir bodies through death, is common cross-culturally(Boyer, 2001).Dualism comes naturally to children. When asked, inimplicit and explicit ways, preschool children will saythat they believe the brain is only responsible for someaspects of mental life, typically those involving delibera-tive mental work, such as solving math problems. Butthe brain is not essential for activities such as pretendingto be a kangaroo, loving one’s brother, or brushing yourteeth (e.g. Gottfried, Gelman & Schultz, 1999; Johnson,1990; Lillard, 1996). This is done by people, not theirbrains.The most dramatic demonstration of childhood dualismconcerns the development of afterlife beliefs. Bering andBjorklund (2004) told children of different ages storiesabout a mouse that died, and asked about the persist-ence of certain properties. When asked about biologicalproperties of the mouse, the children appreciated theeffects of death, including that the brain no longer worked.But when asked about the psychological properties, mostthe children said that these would continue – the deadmouse can have feel hunger, think thoughts, and holddesires. The body was gone, but the soul survives. Andchildren believe this more than adults do, suggesting thatwhile we have to learn the speciﬁc sort of afterlife thatpeople in our culture believe in (heaven, reincarnation,spirit world, and so on), the notion that consciousness isseparable from the body is not learned at all; it comesfor free.One of the ﬁrst things an undergraduate learns in anintroduction to psychology class is that substance dualismis mistaken. It is assumed by virtually all scientiststhat mental life is the product of physical brains (thoughthere is little consensus as to
this all works). Here,as in other domains, common-sense – which is entrenchedin religion – clashes with science.
2. Over-attribution of agency and design
We have what Boyer (2001) has called a ‘hypertrophy of social cognition’: a willingness to attribute psychologicalstates, including agency and design, even when it isinappropriate to do so.The classic demonstration here is that of Heider andSimmel (1944), who made a simple movie in which geo-metrical ﬁgures – circles, squares, triangles – moved incertain systematic ways, designed, based on the psycho-logists’ intuitions, to tell a tale. When shown this movie,people instinctively describe the ﬁgures as if they werespeciﬁc people (bullies, victims, heroes) who have goalsand desires, and they repeat back pretty much the samestory that the psychologists had intended to tell. Furtherresearch ﬁnds that you do not even need bounded ﬁgures – you can get much the same effect with moving dots, aswell as in movies where the ‘characters’ are not singleobjects at all, but moving groups, such as swarms of tinysquares (Bloom & Veres, 1999).The anthropologist Stewart Guthrie (1993) was theﬁrst modern scholar to notice the importance of thistendency as an explanation for religious thought. In hisbook,
Faces in the Clouds
, Guthrie presents anecdotesand experiments showing that people attribute human