Forum: Science & Society

Would Tarzan believe in God? Conditions for the

emergence of religious belief
Konika Banerjee and Paul Bloom
Department of Psychology, Yale University, 2 Hillhouse Avenue, New Haven, CT 06520, USA

Would someone raised without exposure to religious

views nonetheless come to believe in the existence of
God, an afterlife, and the intentional creation of humans
and other animals? Many scholars would answer yes,
proposing that universal cognitive biases generate religious ideas anew within each individual mind. Drawing
on evidence from developmental psychology, we argue
here that the answer is no: children lack spontaneous
theistic views and the emergence of religion is crucially
dependent on culture.
Would an individual never exposed to religious ideas
such as Edgar Rice Burroughs character, Tarzan, who was
raised by apes after his first birthday nonetheless come to
believe in God, an afterlife, and the divine creation of
humans and other creatures? That is, do core religious
beliefs emerge spontaneously in the course of development,
even in the absence of cultural support? If one classifies
religion as yet another cultural invention, akin to agriculture or writing, the answer to these questions is plainly no.
However, many cognitive scientists see the universality
and pervasiveness of religious belief as suggesting that it is
a natural feature of evolved human psychology. Some
propose that certain aspects of religion are biological
adaptations perhaps belief in a punitive supernatural
entity makes individuals more moral and hence more
desirable as mates and social partners [1]; or perhaps
religion promotes social cohesion and cooperation, such
that groups with religion do better than those without [2].
Other researchers, including ourselves, see religion in
general, and religious belief in particular, as a biological
accident a natural byproduct of cognitive systems that
have evolved for other purposes [36].
If one is an adaptationist, then the answer to the Tarzan
question might be yes, because some adaptations emerge
in the absence of cultural support. How should a byproduct
theorist answer, however? Some byproduct theorists
would also reply affirmatively. It is sometimes proposed
that cognitive biases, such as an evolved hypersensitivity
to environmental cues to agency, produce religious concepts within individual minds. Barrett, for instance, suggests that it may even be that were children not provided
with ideas about gods, they would discover gods for themselves when combined with a tendency [. . .] of finding
design and purpose in the natural world ([4], p. 42). If
correct, this would constitute a simple explanation for why
Corresponding authors: Banerjee, K. (Konika.Banerjee@yale.edu);
Bloom, P. (Paul.Bloom@yale.edu)

belief in gods exists in every known human culture, past

and present.
In this article, we argue that the answer is no: cognitive
biases make humans receptive to religious ideas, but do
not themselves generate them. This means that an explanation for the universality of religion has to be found
elsewhere (P. Harris, unpublished; see also Gervais
et al. [7] and Rottman and Kelemen [8] for similar arguments). Our argument is based on research in child development. This might seem surprising, because findings
from developmental psychology are often interpreted as
providing support for the naturalness of religious ideas. We
think that they do only up to a point, however: they
support receptivity, but not generativity.
Consider belief in a divine creator. Young children are
prone to generate purpose-based explanations of the origins of natural objects and biological kinds. They believe,
for example, that clouds are for raining and animals are
to go in the zoo [9]. However, there is no evidence that
children spontaneously come to believe in one or more
divine creators. It is one thing, after all, to think about
natural entities as intentionally designed artifacts of a
sort; it is quite another to generate an enduring belief in
invisible agents who have created these artifacts. Indeed,
other studies find that young children are not committed
creationists; they are equally likely to provide explanations of species origins that involve spontaneous generation [10].
Older children, by contrast, do exclusively endorse creationist explanations. This shift to a robust creationist
preference arises in part because older children are more
adept at grasping the existential themes invoked by the
question of species origins (e.g., existence and final cause)
and also because the notion of a divine creator of nature
meshes well with their early-emerging teleological biases
[10]. However, these older children do not spontaneously
propose novel divine creators. Instead, they adopt the
particular creationist account that their culture supplies.
This might be a singular God or multiple gods; it might be
alien visitors or Mother Earth. If children are not exposed
to such cultural beliefs, the explicit notion of an intentional
creator might never arise.
Children also appear to readily acquire the concept of an
afterlife. When children are directly questioned, they tend
to assert that certain psychological traits, but not biological
traits, persist after bodily death [11]. These responses are
arguably rooted in the psychologically intuitive and
universal distinction between minds and bodies, which
makes it natural to consider mental life enduring without
7

Forum: Science & Society

the presence of a body [5]. Some, such as Barrett [4], take
childrens readiness to reason about life after death as
evidence that they are born believers in an afterlife.
This conclusion is probably too strong, however. There is
no evidence that belief in the afterlife arises spontaneously
in the absence of cultural support. For instance, research in
rural Madagascar, where there is widespread belief in
ancestral spirits, finds that the conception of an afterlife
emerges in the course of development [12]. Even if children
are natural-born dualists [5], this initial stance need not
directly give rise to the afterlife beliefs that are characteristic of many of the worlds religions.
More generally, if universal, early-emerging cognitive
biases generate religious ideas, we would expect to see
these ideas emerge spontaneously. This would be akin to
the process of creolization, such as when deaf children who
are exposed to non-linguistic communication systems create their own sign language [13]. However, such cases are,
as best we know, non-existent. There are many examples
where children are quick to endorse religious beliefs, often
surprising their atheist parents [4]. But this is consistent
with receptivity, not generativity, as these beliefs correspond to those endorsed within the social environment in
which children are raised.
Consider, as a test case, belief in multiple deities. This is
the historically foundational religious stance, with monotheism a more recent invention [14]. It would be striking
support for the generativity position if children raised in
monotheistic societies declared their belief in multiple
gods. However, to our knowledge, they never do. They
come to believe instead in the same singular omnipotent
deity that everyone else believes in.
The findings from developmental psychology support
the following theory of the emergence of religious belief:
humans possess a suite of sophisticated cognitive adaptations for social life, which make accessible certain concepts
that are associated with religion, including design, purpose, agency, and bodysoul dualism. However, more is
needed to generate fully-fledged, sustained, and conscious
religious beliefs, including a belief in gods, in divine creation of natural entities, and in life after death. Such beliefs
require cultural support.
If our analysis is correct, then certain conclusions
follow. First, this analysis suggests that any theory, adaptationist or otherwise, that posits that religious belief
will emerge without cultural support an affirmative
answer to the Tarzan question is mistaken. Second, this
perspective shifts the emphasis from representational
biases towards learning mechanisms (see also [7]). Any
adequate theory has to explain how such beliefs are acquired. Third, we are now left with the question of where
religious beliefs came from in the first place. Why do
people think them up?

Trends in Cognitive Sciences January 2013, Vol. 17, No. 1

Of course, the diversity of world religions is not the result

of a series of wholly independent moments of cultural invention. Rather, once religious ideas emerge, historical
processes of social movement and geopolitical conquest
spread them around the globe [14]. Nonetheless, an origin
story or a series of origin stories is still needed and here
the emphasis might shift to the types of analyses presented
by scholars such as Freud, Marx, and Durkheim, who
explored the social and motivational forces that lead people
to generate religious ideas. Although current developments
in the cognitive science of religion may explain why religion
sticks once it has got off the ground, an explanation for why
humans have it in the first place requires a very different
approach, one that moves away from biases in learning and
representation and explores, instead, considerations of
motivation, needs, fears, and creative insights.
Acknowledgments
We thank Paul Harris, Justin Barrett, and two anonymous reviewers for
their helpful comments on an earlier version of this article.

References
1 Johnson, D.D.P. and Bering, J.M. (2006) Hand of God, mind of man:
punishment and cognition in the evolution of cooperation. Evol.
Psychol. 4, 219233
2 Wilson, D.S. (2002) Darwins Cathedral: Evolution, Religion, and the
Nature of Society, University of Chicago Press
3 Boyer, P. (2001) Religion Explained, Basic Books
4 Barrett, J.L. (2012) Born Believers: The Science of Childrens Religious
Beliefs, Free Press
5 Bloom, P. (2004) Descartes Baby: How the Science of Child Development
Explains What Makes Us Human, Basic Books
6 Bloom, P. (2009) Religious belief as an evolutionary accident. In The
Believing Primate (Murray, M.J. and Schloss, J., eds), pp. 118127,
Oxford University Press
7 Gervais, W.M. et al. (2011) The cultural transmission of faith: why
innate intuitions are necessary, but insufficient, to explain religious
belief. Religion 41, 389410
8 Rottman, J. and Kelemen, D. (2012) Is there such a thing as a
Christian child? Evidence of religious beliefs in early childhood. In
Science and the Worlds Religions, Vol. 2: Persons and Groups
(McNamara, P. and Wildman, W.J., eds), pp. 205238, Praeger Press
9 Kelemen, D. (2004) Are children intuitive theists? Reasoning about
purpose and design in nature. Psychol. Sci. 15, 295301
10 Evans, E.M. (2001) Cognitive and contextual factors in the emergence
of diverse belief systems: creation versus evolution. Cogn. Psychol. 42,
217266
11 Bering, J.M. and Bjorklund, D.F. (2004) The natural emergence of
reasoning about the afterlife as a developmental regularity. Dev.
Psychol. 40, 217233
12 Astuti, R. and Harris, P.L. (2008) Understanding mortality and the life
of the ancestors in rural Madagascar. Cogn. Sci. 32, 713740
13 Senghas, A. et al. (2004) Children creating core properties of language:
Evidence from an emerging sign language in Nicaragua. Science 305,
17791782
14 Wright, R. (2009) The Evolution of God, Little, Brown and Company
1364-6613/$ see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tics.2012.11.005 Trends in Cognitive Sciences, January 2013,
Vol. 17, No. 1