The folk psychology of souls
Jesse M. Bering
Institute of Cognition and Culture, Queen’s University Belfast,
Belfast BT7 1NN, United Kingdom.
j.bering@qub.ac.uk qub.ac.uk/icc
http://www.qub.ac.uk/schools/InstituteofCognitionCulture/Staff/
JesseMBering/
Abstract: The present article examines how people’s belief in an afterlife, as well as closely related supernatural beliefs, may open an
empirical backdoor to our understanding of the evolution of human social cognition. Recent findings and logic from the cognitive
sciences contribute to a novel theory of existential psychology, one that is grounded in the tenets of Darwinian natural selection.
Many of the predominant questions of existential psychology strike at the heart of cognitive science. They involve: causal
attribution (why is mortal behavior represented as being causally related to one’s afterlife? how are dead agents envisaged as
communicating messages to the living?), moral judgment (why are certain social behaviors, i.e., transgressions, believed to have
ultimate repercussions after death or to reap the punishment of disgruntled ancestors?), theory of mind (how can we know what it
is “like” to be dead? what social-cognitive strategies do people use to reason about the minds of the dead?), concept acquisition
(how does a common-sense dualism interact with a formalized socio-religious indoctrination in childhood? how are supernatural
properties of the dead conceptualized by young minds?), and teleological reasoning (why do people so often see their lives as being
designed for a purpose that must be accomplished before they perish? how do various life events affect people’s interpretation of
this purpose?), among others. The central thesis of the present article is that an organized cognitive “system” dedicated to forming
illusory representations of (1) psychological immortality, (2) the intelligent design of the self, and (3) the symbolic meaning of
natural events evolved in response to the unique selective pressures of the human social environment.
Keywords: causal reasoning; death concept; evolutionary theory; existential psychology; folk biology; intelligent design; intentionality;
mental representation; teleological reasoning; theory of mind
Life is a jest, and all things show it; I thought so once, and now
I know it.
— John Gay, Epitaph
1. Introduction
By stating that psychological states survive death, one is
committing to a radical form of mind-body dualism. Yet
this radicalism is especially common. In the United
States alone, 95% of the population reportedly believes
in life after death (Greeley & Hout 1999; Lester et al.
2002). The majority of people from other societies, as
well, see death as a transitional event that unbuckles the
ethereal self from its body. The soul is typically represented
as the conscious personality of the decedent
and the once animating force of the now inert physical
form (Thalbourne 1996). Although there are many varieties
of afterlife beliefs, each – at least implicitly – shares
a dualistic view of the self as being initially contained in
bodily mass and as exiting or taking temporary leave of1
the body at some point after the body’s expiration
(Bloom 2004; Boyer 2001).
There is clear evidence showing that emotive factors can
be powerful contributors to people’s belief in life after
death (e.g., Alvarado et al. 1995; Dechesne et al. 2003;
Thalbourne 1996). In general, psychologists who study
this area have tended to focus on individual differences,
specifically the role of death anxiety, and have posited a
variety of “comfort hypotheses” involving the human
motivation to construct such supernatural beliefs. In contrast,
less is known about the basic components underlying
the strong cognitive bias to entertain belief in an immortal
soul (Astuti, forthcoming a). These more basic questions
concerning the cognitive architecture behind afterlife representations
are also important pieces of the puzzle and
will be explicitly addressed in the present article. Whatever
one’s personal motivations for rejecting or endorsing
the idea of an immaterial soul that can defy physical death,
the ability to form any opinion on the matter would be
absent if not for our species’ defining capacity to differentiate
unobservable minds from observable bodies
(Povinelli & Bering 2002; Suddendorf & Whiten 2001;
Tomasello & Call 1997).
Some researchers have already begun laboratory investigations
into the question of whether humans are
“common sense dualists,” work that seems to have implications
for our understanding of people’s intuitive conceptions
of souls and the afterlife (see Bloom 2004).
For example, in a modification of the classic expectancy
violation paradigm (which uses looking time as a measure
of nonverbal infants’ “surprise” at an event), Kuhlmeier
et al. (2004) positioned identical twin experimenters at
different points in the laboratory to test 5-month-olds’
BERING JESSE, Reader in the School History and
Anthropology Departments and Director of the Institute
of Cognition and Culture, Queen’s University
Belfast, has focused his work on the psychological foundations
of supernatural belief. An experimental cognitive
scientist, he is also a specialist in the evolutionary
bases of religion.
BEHAVIORAL AND BRAIN SCIENCES (2006) 29, 453–498
Printed in the United States of America
# 2006 Cambridge University Press 0140-525X/06 $12.50 453
ability to reason about the law of continuous motion as it
applies to human bodies. Like any material substance,
human bodies cannot go from A ! C without first
passing along the trajectory of B (a contiguous space
between the two points). For inanimate objects, infants
are surprised (i.e., look longer) when the object disappears
from behind one barrier and then seems to reemerge from
behind another nonadjacent barrier. In the case of a
human who violates the law of continuous motion,
however, 5-month-olds are not surprised (i.e., they do
not look longer at this event than the non-violation
event). The authors speculate that “infants do not readily
view humans as material objects” (Kuhlmeier et al. 2004,
p. 101) and that an “appreciation that people are just
objects may be a developmental accomplishment”
(p. 102; emphasis in original).
But how do we get from the common-sense dualism of
infants to beliefs of the afterlife so soberly entertained by
adults? Even a superficial pass over such beliefs strikes one
as involving many of the core problems in cognitive
science: causal attribution (how is mortal behavior causally
related to one’s afterlife? how are dead agents envisaged as
communicating messages to the living?), moral judgment
(why are certain social behaviors, i.e., transgressions,
believed to have ultimate repercussions after death or to
reap the punishment of disgruntled ancestors?), theory
of mind (how can we know what it is “like” to be dead?
what social-cognitive strategies do people use to reason
about the minds of the dead?), concept acquisition (how
does a common-sense dualism interact with a formalized
socio-religious indoctrination in childhood? how are
supernatural properties of the dead conceptualized by
young minds?), teleological reasoning (why do people so
often see their lives as being designed for a purpose
that must be accomplished before they perish? how do
various life events affect people’s interpretation of this
purpose?), and so on.
In what follows, I examine how this human folk psychology
of souls, as well as closely related supernatural beliefs,
may open an empirical backdoor to our understanding of
the evolution of human social cognition. Recent findings
and logic from the cognitive sciences contribute to a
novel theory of existential psychology, one that is
grounded in the tenets of Darwinian natural selection.
The central thesis of the present article is that an organized
cognitive “system” dedicated to forming illusory
representations of (1) psychological immortality, (2) the
intelligent design of the self, and (3) the symbolic meaning
of natural events evolved in response to the unique selective
pressures of the human social environment.
2. Psychological immortality as a cognitive
default
I’m a materialist, I swear it to you; I’m not going crazy. But
something’s the matter. I see my corpse; that’s not hard but
I’m the one who sees it, with my eyes. I’ve got to think . . .
think that I won’t see anything anymore and the world will
go on for the others. We aren’t made to think that.
— Jean-Paul Sartre (1937/1969), The Wall:
And Other Stories
From an evolutionary perspective, it is important to first
ask whether humans “naturally” reason about death as a
transitional state of consciousness or simply acquire such
ideas through cultural exposure (perhaps from adults
who “invent” such notions to ameliorate their own death
anxiety; see, e.g., Dechesne et al. 2003; Harris &
Gime´nez 2005). Although conventional wisdom tends to
favor a general learning hypothesis for the origins of afterlife
beliefs, recent findings suggest a more complicated
developmental picture.
For example, in a study by Bering and Bjorklund (2004),
children (as well as an adult comparison group) were
presented with a puppet show in which an anthropomorphized
mouse was killed and eaten by an alligator, and
then asked about the biological and psychological functioning
of the now-dead mouse. Kindergartners understood
that various biological imperatives (e.g., the
capacity to be sick, the need to eat, drink, and relieve
oneself) no longer applied to the dead mouse. The
majority of these children even said that the brain of the
dead mouse no longer worked, which is especially telling
given that children at this age also understand that the
brain is “for thinking” (Bloom 2004; Gottfried & Jow
2003; Johnson & Wellman 1982; Slaughter & Lyons 2003).
Yet when asked whether the dead mouse was hungry or
thirsty, or whether it was thinking or had knowledge,
most kindergartners said yes. In other words, young
children were cognizant of the fact that the body stops
working at death but they viewed the mind as still active.
Furthermore, both the children and adults were particularly
likely to attribute to the dead mouse the capacity
for certain psychological states (i.e., emotions, desires,
and epistemic states) over others (i.e., psychobiological
and perceptual states), a significant trend that will be
addressed in the following section.
In general, however, kindergartners were more apt to
make psychological attributions to the dead mouse than
were older children, who were not different from adults
in this regard. This is precisely the opposite pattern that
one would expect to find if the origins of such beliefs
could be traced exclusively to cultural indoctrination. In
fact, religious or eschatological-type answers (e.g.,
Heaven, God, spirits, etc.) among the youngest children
were extraordinarily rare. Thus, a general belief in the
continuity of mental states in dead agents seems not something
that children acquire as a product of their social–
religious upbringing, because increasing exposure to
cultural norms would increase rather than attenuate afterlife
beliefs in young minds. Instead, a natural disposition
toward afterlife beliefs is more likely the default cognitive
stance and interacts with various learning channels (for an
alternative interpretation, see Astuti, forthcoming a).
Moreover, in a follow-up study that included Catholic
schoolchildren, this incongruous pattern of biological
and psychological attributions to the dead mouse appeared
even after controlling for differences in religious education
(Bering et al. 2005).
Unlike intuitive reasoning about dead agents’ bodies,
which may help to motivate physical avoidance of these
dangerous objects in the environment (via the emotion
of disgust or agency detection mechanisms which err
on the side of caution for ambiguously dead/sleeping
agents; Barrett & Behne 2005; Rozin et al. 1993), intuitive
reasoning about dead agents’ minds would seem to leave
open the possibility for continued social relationships
with the dead.
Bering: The folk psychology of souls
454 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
2.1. The simulation constraint hypothesis and the
afterlife
Our own death is indeed unimaginable and whenever we make
the attempt to imagine it we can perceive that we really survive
as spectators.
— Sigmund Freud, Thoughts for the Times on War and Death
Try to fill your consciousness with the representation of
no-consciousness, and you will see the impossibility of it.
The effort to comprehend it causes the most tormenting
dizziness. We cannot conceive of ourselves as not existing.
— Miguel de Unamuno (1912/1954), Tragic Sense of Life
The causal mechanisms that lead young children to
represent dead agents’ minds as being psychologically
active have yet to be precisely identified. Nevertheless,
there is evidence that simulation constraints (i.e., the
inability to know what it is “like” to be dead) may comprise
an important set of factors. Like reasoning about one’s past
mental states during dreamless sleep or while in other
somnambulistic states, consciously representing a final
state of unconsciousness poses formidable, if not impassable,
cognitive constraints (Barrett 2004; Bering 2002a;
Bering & Bjorklund 2004; Bering et al. 2005; Clark 1994;
Gilbert 2001; Nichols, in press). By relying on simulation
strategies to derive information about the minds of dead
agents, one would be compelled to put themselves “into
the shoes” of such organisms, which is an impossible
feat. These constraints may lead to a number of telltale
errors, namely “Type I” errors (inferring mental states
when in fact there are none), regarding the psychological
status of dead agents. Koocher (1973, p. 374) described,
for instance, how a group of children tested on death comprehension
reflected on what it might be like to be dead
“with references to sleeping, feeling ‘peaceful,’ or simply
‘being very dizzy’.”
Attempts to simulate dead agents’ minds may even
result in Type I errors made by adults who profess not
to believe in the afterlife. Bering (2002a) found that
when undergraduate students were asked to reason
about the psychological abilities of a protagonist who had
just abruptly died in an automobile accident, even some
participants who later classified themselves as “extinctivists”
(i.e., those who endorsed the statement “what we
think of as the ‘soul,’ or conscious personality of a
person, ceases permanently when the body dies”; after
Thalbourne 1996), nevertheless stated that the dead
person knew that he was dead.
In addition, there is reason to believe that certain types
of mental states are more difficult to imagine being permanently
without than are others. In the study by Bering and
Bjorklund (2004), for example, participants at every age
were more likely to attribute emotions, knowledge, and
desires to the dead mouse than that they were psychobiological
and perceptual states (see also Bering et al. 2005).
This may be understood in relation to children’s growing
scientific knowledge. With regard to psychobiological
states, such as hunger or thirst, Slaughter and her colleagues
have shown that once children display an understanding
of the vitalistic purpose of the behaviors tied to
these states (i.e., that eating and drinking function to
sustain life), this knowledge facilitates scientific reasoning
about death (Slaughter & Lyons 2003; Slaughter et al.
1999). Indeed, children who appeal to a vitalistic biological
framework when reasoning about human bodies are more
precocious in their understanding of death (Slaughter &
Lyons 2003).
Similarly, because perceptual states are closely tied to
obvious bodily structures, children who possess teleofunctional
biological knowledge about these structures
(e.g., that ears are “for hearing”) may begin to reason
that, so long as the body has stopped functioning, the
capacity for such states must also become defunct at
death (O’Neill & Chong 2001).
In addition, because individuals are aware from their
own previous or current experiences what it is like, say,
not to be sleepy, not to hear, or not to be hungry, they
may draw from the phenomenal negation of such states
and apply these experiences to the minds of dead agents.
Thus, in some cases, simulation may actually corroborate
scientific knowledge and further reduce Type I errors.
In contrast to these categories of psychological states,
however, the nature of the body’s role in producing the
subjective experiences of emotions, desires, and beliefs
seems not as amenable to children’s scientific theories of
dead minds (or, indeed, even to adults’ formulation of
scientific theories regarding phenomenal consciousness
and the brain, e.g., qualia; see McGinn 1991). These
aspects of consciousness are not obviously related to the
body’s survival, nor are they linked to external bodily
accoutrements (i.e., sense organs) that become “broken”
by death. In the absence of scientific theory concerning
the isomorphic relationship between the brain and the
mind, individuals may defer to a simulation strategy in
reasoning about dead others, a strategy that inevitably
leads to Type I errors for these particular mental capacities
(Bering 2002a; Clark 1994; Gilbert 2001; Nichols, in
press). Firsthand experiences with the phenomenal negation
of mental functions such as desires, emotions, and
thought can never be had because these states are constantly
“turned on” during conscious periods (e.g., it is
epistemologically impossible to know what it is like not
to think), making people inclined to impute these
capacities to dead agents. Indeed, in looking at participants’
response latencies to state that a dead protagonist
lacked the capacity for various mental states, Bering
(2002a) reported that it took people longer to answer
that this was the case for “difficult-to-imagine-theabsence-of”
states (e.g., desire: “Now that he’s dead,
does he want to be alive?”) than for “easy-to-imaginethe-absence-of”
states (e.g., psychobiological: “Now that
he’s dead, is he still sleepy?”).2
2.2. Offline social reasoning: Why the afterlife
is a place
I forced myself to stop thinking of her as someone still somewhere,
if only in memory, still obscurely alive, breathing,
doing, moving, but as a shovelful of ashes; as a broken link, a
biological dead end, an eternal withdrawal from reality.
— John Fowles (1978), The Magus
In addition to simulation constraints, there are other
aspects of human social cognition that may encourage
attributions of continued psychological functioning to
dead agents. When investigating peoples’ intuitive
conceptions of dead agents’ minds, we are wise to remember,
for instance, that human relationships are largely
characterized by offline social events; those with whom
we have relationships are only periodically directly
Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 455
observable (e.g., Dunbar 1993; 2004). An offline social
system leads us to tacitly assume that individuals with
whom we have relationships are engaged in actions even
when we cannot observe them doing so. The fact that
your mother is not in the room at the moment does not
compromise your capacity to reason about her mind,
though obviously the accuracy of your social judgments
will be limited. When conjuring up her offline image you
are likely to imagine her as somewhere and as doing something
– in the kitchen washing dishes, in bed sleeping,
playing squash with the neighbor, and so on. Similarly,
the dead are envisaged not as inanimate objects slowly
decomposing in situ under the earth, but instead as
having relocated to some unobservable locale where they
are very much “living” their dead lives.
When it comes to death, human cognition apparently is
not well equipped to update the list of players in our
complex social rosters by accommodating the recent nonexistence
of any one of them. This is especially the case, of
course, for individuals who have played primary roles in
our social lives, who did so for a long time, and who
were never presumed to be continuously stationary
when they were out of our sight. Because our minds are
designed for offline as well as online social processing,
we expect the periodic physical absence of social partners.
Casual observation reveals that individuals will often, for
example, pick up the phone with the intention of calling
the decedent or fleetingly imagine how the decedent will
react when told about some good news, only to remember
that the person is not where they usually are – they have
“passed on” to someplace else.
Although these automatic cognitions are probably the
residue of habitual social behaviors, they also reveal something
about the challenges faced by the human cognitive
system when it attempts to process information concerning
the truth about dead agents’ physical whereabouts. A
person who has recently died and whose body has
already been disposed of may continue to be processed
by an offline social system for an undetermined period
of time. This place error is seemingly compounded by nonnegotiable
simulation constraints that tempt us into
reasoning about these dead agents’ continued psychological
functioning (as discussed in the previous section).
2.3. By-product versus functional analyses of belief
in immortal souls
THE CHILD: I’m frightened.
THE WOMAN: And so you should be, darling. Terribly frightened.
That’s how one grows up into a decent, god-fearing man.
— Jean-Paul Sartre (1937/1969), The Flies
There may be good reason to argue that natural selection
operated on the foregoing psychological biases. Representations
of the afterlife are culturally recurrent, proximally
driven by emotions, frequently implicated in social and
reproductive matters, and superficially fitted to the
ecological niche in which the human organism develops
(Bering & Bjorklund 2004; Dechesne et al. 2003;
Reynolds & Tanner 1995). These features are consistent
with what we know about the nature of psychological
adaptations (Bjorklund & Pellegrini 2002; Tooby &
Cosmides 1992). This is not to say that specific afterlife
beliefs – the culturally variable vicissitudes of the
hereafter – are direct products of natural selection. As
will soon be discussed, investigators such as Boyer
(2001) and Atran (2002) have shown that the “selection”
of explicit religious ideas occurs at the cultural level,
with the “survival” of such ideas being a feature of their
ability to become ensconced in the evolved architecture
of exposed human minds (Sperber & Hirschfeld 2004).
Instead, it is to argue that the subtle contours of a uniquely
human adapted design may stand out when closely examining
the folk psychology of souls, an intuitive pattern of
reasoning that does not appear to hinge on the presence
of explicit religious concepts per se (Bering 2002a).
But how might a representational bias for envisioning
personal immortality have impacted the net genetic
fitness of individual humans in ancestral environments?
Unfortunately, among cognitive scientists, scant attention
has been paid to the evolutionary significance of the
human capacity to represent the self sub specie aeternitatis
(“under the aspect of eternity”). Instead, many scholars
categorize afterlife concepts in the same way they do
other types of religious concepts, as especially virulent
strains of culturally transmitted ideas that are highly effective
at pirating core cognitive architecture (Atran 2002;
Boyer 2001; for an exception, see Baron-Cohen 1999).
According to this perspective, only the cognitive architecture
itself can be the product of natural selection; religious
ideas are seen as simply being parasitic on this evolved
architecture – as nothing more than noise that shares a
general frequency between cultures (e.g., Pyysia¨inen
2001; Sperber & Hirschfeld 2004). For example, in his
book Religion Explained, Boyer (2001, p. 40) writes:
“People have religious notions and beliefs because they
acquired them from other people. On the whole, people
get their religion from other members of their social
group.”
Boyer and other cultural epidemiologists’ view afterlife
concepts, as well as other types of supernatural concepts,
as unavoidable carryovers of cultural selection. Specifically,
Boyer (2000; 2001) has argued that religious ideas
exploit information-processing mechanisms into paying
attention to them because they violate ontological regularities
by hybridizing or transgressing natural categories
(see also Mithen 1996). Thus, religious ideas are especially
likely to attach to evolved cognitive templates that are
designed for reasoning about exemplars from natural categories
– such as PERSON or ANIMAL – because these
templates act as flypaper for salient, “counterintuitive”
cases (Atran & Norenzayan 2004; Barrett 2000; 2004;
Pyysia¨inen 2001; Slone 2004; Sperber & Hirschfeld 2004).
According to Boyer (2003a), then, a ghost is a person
who is without a physical body and as such is a conceptually
seductive idea. The concept of an afterlife therefore
is easily generated and transmitted between minds. Like
all other religious concepts, however, it is otherwise
treated as a biologically sterile by-product.
Similarly, Sperber and Hirschfeld (2004, p. 44) write
that, “explaining religion by a religious disposition lacks
insight and plausibility.” Instead, these scholars argue
that religion is a non-adaptive by-product (i.e., a spandrel),
one that arises through adapted human cognition acting
in concert with culturally migrating counterintuitive concepts
that change shape both within and between minds
(for critical reviews, see Alcorta & Sosis 2005; Bulbulia
2004; 2005; Sosis & Alcorta 2003).
Bering: The folk psychology of souls
456 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
2.4. Inhibition and the preservation of reputation
I believe that I am in hell, therefore I am.
—Arthur Rimbaud (1873/1999), A Season in Hell
Once the ability to entertain supernatural agent concepts
evolved, such ideas might have led our ancestors to
inhibit socially proscribed actions out of the fear that
gods or dead agents, now “full access strategic agents,”
were watching them (Boyer 2001). Some empirical
support for this general argument was found in a recent
study by Bering et al. (2005). In this study, undergraduate
students who were casually told that a ghost was recently
spotted in the laboratory were less willing to cheat on a
competitive computer task – as measured by latency of
response to delete the “accidentally” revealed answer –
when they were tested alone in the room than were
control participants who heard nothing of the fictitious
ghost (see also Burnham & Hare, in press; Haley &
Fessler 2005). In the case of supernatural beliefs, therefore,
it is helpful to highlight Haselton and Buss’s (2003,
pp. 29–30) general adaptationist point that, “the human
mind is designed to reason adaptively, not truthfully or
even necessarily rationally.” The genetic fitness effects of
such behavioral inhibition have real currency in natural
selection theory.
The relationship between supernatural morality and
behavioral inhibition is potentially a very important point
for evolutionary biologists. Because natural selection is
pragmatic, the illusion of a supernatural morality, if it
served to curb selfish behaviors and thus preserved
social reputation in the ancestral past, may be an illusion
by design (Alcorta & Sosis 2005; Bering 2005; Bering &
Johnson 2005; Bering et al. 2005; Boyer 2001; Bulbulia
2004; Dunbar 2004; Hinde 1999; Johnson & Kru¨ger
2004). Many writers have argued that, at some point in
the recent evolutionary past, hominid sociality underwent
a relatively abrupt shift that was characterized by strong
selective forces operating on reputation-related behaviors
(Alexander 1987, p. 110; Bering & Bjorklund, in press;
Bering & Shackelford 2004; Daly & Wilson 1994; Emler
1994; Frank 1988; Goffman 1959, 1963; Hilton et al.
1993; Schelling 1960; Wright 1994). Because of the risks
associated with social detection of selfish acts, and the
peculiar “stickiness” of bad reputations (e.g., Baumeister
et al. 2001; Goffman 1963), psychological traits that facilitated
the inhibition of selfish acts were likely subjected to
natural selection. The costs of underestimating the risk of
social detection would have been disproportionately
greater than the costs of prosocial decisions that were
contextually maladaptive. Even if altruism was costly
every time, if it avoided a lethal cost once, those other
costs would become negligible (Nettle 2004).
Experimental findings of prosocial behavioral change in
light of supernatural primes (e.g., Bering et al. 2005) also
link up with the ethnographic database concerning afterlife
beliefs. In some religious ideologies, the fate of the
soul after death is determined by the social behaviors of
the individual during life. Reflections on the ultimate consequences
of (im)moral actions (e.g., whether the soul is
expelled to Hell or dissipated in nirvana) should be
capable of exerting a causal influence on today’s overt
behavior, which would have the effect of preserving reputation
by encouraging the inhibition of selfish acts or facilitating
self-control. In the United States, for example, the
majority (79%) of people believe that there will come a day
when God judges them and decides whether they will go
to Heaven or Hell (Gallup Organization 1999; see also
Lester et al. 2002). For current purposes, such poll data
may actually be misleadingly low. What people say they
believe about the supernatural and how they implicitly
reason are quite different things (Subbotsky 1997; 2001).
Scientific knowledge about causal relations between
behaviors and consequences may therefore destabilize
this adapted system, but more by overriding supernatural
beliefs than by replacing them (McCauley 2000;
Subbotsky 2001).
In many traditionalist religious societies, the emphasis is
on worldly punishment for moral transgressions, whereby
norm violators are visited by sickness, poverty, or other
types of misfortune (see Bering & Johnson 2005).
Mostly, punishment is seen as being imposed by
disgruntled ancestors (Hinde 1999; Reynolds & Tanner
1995). In some cases, belief in the vicarious punishment
of dead agents achieves similar prosocial effects. In medieval
Europe, where people’s social behaviors were thought
to determine the fate of dead loved ones whose souls were
at limbo in purgatory, thoughts of the dead were so
prevalent in the daily affairs of the living that at least one
historian has even referred to the dead as constituting a
separate “age group” (Davis 1977; as cited in Luria 2001).
Although critical developmental studies have yet to
specifically address the etiology of full-access strategic
agent concepts and their consequences for behavioral
inhibition, some related findings with young children do
point to a human cognitive system prepared to reason
about “omniscient” supernatural agents (Bering 2005;
Bering & Johnson 2005). In a recent study, Barrett et al.
(2001) report that because, theologically, God is
all-knowing and therefore cannot hold false beliefs (and
therefore cannot be deceived), the social cognitive
systems of young children may be better suited to reasoning
about the culturally postulated mind of God than about
the epistemologically limited minds of humans and other
animals. For example, whilst 3-year-olds incorrectly
reason that a naı¨ve person knows the true contents of an
inaccurately labeled box, they correctly reason (at least,
in a theological sense) that God knows the true contents
as well. Thus, according to the authors, because of
egocentric biases in early childhood, there may be cognitive
precursors for full access strategic agent concepts
that developmentally precede even natural mental agent
concepts.
3. Souls and intelligent design
The concept of man in the mind of God is comparable to the
concept of paper-cutter in the mind of the manufacturer, and,
following certain techniques and a conception, God produces
man, just as the artisan, following a definition and a technique,
makes paper-cutter. Thus, the individual man is the realization
of a certain concept in the divine intelligence.
— Jean-Paul Sartre, Existentialism and Human Emotions
However, to understand the relationship between belief in
gods or other supernatural agents who are interested in
our social behaviors and belief in immortal souls requires
that we disentangle several related strands of causal
reasoning. Consider that if God does not exist, then the
Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 457
unique self (i.e., the individual “soul” of any given person)
cannot be the product of intelligent design; rather, it is
simply the end product of standard machinations of
genetic and environmental recombination. If the soul is
not the product of intelligent design, then there is no
teleological function that it is designed to fulfill, no
raison d’eˆtre to explain its existence beyond human attributions
of purpose. The task remains for cognitive scientists
to determine why the teleological position is so
frequently adopted, and prospers so vehemently, over
the mechanistic alternative. The human mind cannot
seem to easily accommodate itself to a godless, evolutionary
canon when it comes to the self’s existence.
In fact, resistance to the mechanistic theory of natural
selection may have as much to do with a cognitive bias
toward intentionality as it does with an emotionally laden
or moralistic bias. Recent findings converge to show that
humans have a strong teleological bent when it comes to
reasoning about the origins of artifacts, animals, and
natural objects (e.g., Evans 2001; German & Barrett
2005; Kelemen 2004; Kelemen & DiYanni 2005).
Kelemen (2004) has even gone so far as to dub children
“intuitive theists” because of their so-called “promiscuous
teleology.” According to Kelemen, most young children
would prefer the teleo-functional explanation that a
cloud is “for raining” rather than assent to the experimenter’s
suggestion that perhaps raining is just something
that a cloud does. This cognitive bias shows that young
children are “endorsing the view that natural entities are
‘made for something’ and that is why they are here”
(Kelemen 2004, p. 295). In a similar vein, Evans and her
colleagues have found evidence that most young children
prefer creationist arguments over evolutionary ones when
reasoning about the origins of species (e.g., Evans 2001).
Teleological reasoning is often applied to the origins of
the self, as well, such as talk about what one was “born to
do” or that one is leading a life that he or she was not
“meant for.” Indeed, the term conceive (from the Latin
concipere, “to take in and hold”), though originally used
to describe impregnation (“to take into the womb,
become pregnant”), was within that same century (c.
1280–1340) adopted to describe an intentional mental
process (“to take into the mind”).
The tendency for people to reason about the special
purpose of the unique self may differ from other forms
of teleo-functional reasoning in two important ways.
First, it appears to be much more resistant to scientific
knowledge. Although teleo-functional beliefs about
natural objects are found in Romanian Gypsy adults, a
group that does not possess scientific knowledge regarding
natural artifact origins (see Kelemen 2004), they generally
decline with age and are relatively rare among scientifically
educated adults. In contrast, ascriptions of intelligent
design when reasoning about the purpose of individual
lives appear to remain stable (and perhaps even increase)
over the life course, probably due to the accrual and retrospective
interpretation of autobiographical experience
(Bering 2003b; Bruner 2001; McAdams 2001).
Second, when it comes to lay beliefs about souls,
attributions of purpose occur frequently for individual
members of the same conceptual family. People tend
to ascribe special purpose more often to the specific case –
such as “what am I meant for?” – than they ascribe shared
purpose to members of the same natural category – such
as “what is the human species meant for?” For no other
natural categories do such special teleological ascriptions
seem to occur. (Imagine an evolutionary biologist hypothesizing
about the special purpose of a specific heart of a
specific organism of all the possible such organisms
within a given species.) Even for artifacts, teleo-functional
judgments for class categories (e.g., CHAIR) appear to
trump within-category exemplars (e.g., both a gothic
revival style and a Chinoisserie style chair are “for
sitting” although they may differ in design for posturing
the body), and rarely occur within the exemplar class
itself (e.g., the special purpose of an individual Chinoisserie
style chair) (Defeyter & German 2003).
The categorical question “Why am I here?” is important
for evolutionary analysis because it may set the stage for an
obligatory social relationship between the self and its
presumed supernatural creator. If this cognitive illusion,
enriched with social affect, plays a causal role in generating
genetic fitness-enhancing responses (e.g., through the
individual’s behavioral compliance with moral norms
which the creator is believed to have authored), then an
adaptationist hypothesis for the folk psychology of souls
gains support.
The tendency to endow human lives with an a priori
meaning is particularly obvious in the wake of recent
loss. Despite differences in religiosity, individuals who
are in mourning commonly report feeling a sense of meaninglessness
(Davis & Nolen-Hoeksema 2001; Golsworthy &
Coyle 1999; Smith et al. 1992; Yalom 1980). Such existential
despair, characteristic of the early stages of the grieving
process, betrays people’s implicit belief that they are part
of a privileged social relationship with some abstract
agent who exerts a causal influence over their everyday
lives. Many types of “premature” death (e.g., accidents,
fatal illnesses, homicides) seem to force surviving individuals
to acknowledge that this privileged social relationship
is illusory: the existence of the self is abruptly surrendered
to a veridical belief in the fundamental and mindless laws
of natural probability. The resulting existential despair can
be attributed to the realization that the predictability and
controllability of one’s own death, like that of the decedent’s,
is in actuality very low.
In this light, there is no emotionally invested God who
favors or disfavors the continued survival of the self. Consequently,
whatever social contracts previously entered
into with this nonexistent agent that led the self to
expect a reasonable deferment of death until old age are
exposed as being spurious. Avenues by which individuals
may reenter into this illusory contract include “just
world” beliefs (e.g., by reasoning that the person must
have been somehow deserving of death), and judging
that the decedent was different from themselves (and
thus unlikely to have been in the same sort of privileged
social relationship with God) (e.g., Hafer & Be`gue 2005;
Lerner 1980; Lerner & Miller 1978; Pyszczynski et al.
1995).
The fact that most individuals do tend to reenter into
these illusory social contracts, even in the face of seemingly
egregious violations, suggests that the affective
push to do so is capable of overthrowing any rationalist
Weltanschauung. However, if this is the expression of an
evolved system, as the current model alleges, then any
explicit philosophical position that discards meaning is
naturally disadvantaged, because the self can no sooner
Bering: The folk psychology of souls
458 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
“choose” to be a subjective atheist than retinas can
“choose” not to convert light energy into signals that are
carried to the brain by the optic nerve (Bering 2005;
McCauley 2000). Adapted psychological systems, by
definition, determine the way that information can be
processed due to design solutions in the brain that were
engineered by natural selection. In the present case, just
as we can close our eyes to prevent light from being converted
into neural signals, science may provide a minority
(i.e., nonbelievers) with the armamentarium to close their
eyes to the supernatural.
3.1. Suicide as a violation of intelligent design
I condemn that nature which, with such impudent nerve,
brought me into being in order to suffer – I condemn it in
order to be annihilated with me.
—Fyodor Dostoyevsky (1877/1949), Diary of a Writer
Public opinions concerning suicide further serve to highlight
the role of the design stance in the existential
domain. Those who believe that one’s life is owned by
God are more likely to view suicide – as well as abortion,
capital punishment, and medical euthanasia –as being
morally wrong (Ross & Kaplan 1994; Worthen & Yeatts
2001). It is not suicide per se that sheds light on this
teleo-functional bias, but the moral repugnance for the
act. Religious rules against suicide reveal a more
complex cognitive stance than is immediately apparent.
It is a counterintuitive stance in which the self’s will is
seen as imposing itself over the will of the creator of the
self’s will. According to the premise that a person’s life
belongs to God, an individual does not have the right to
purposefully cause his or her own death, because this
right is seen as being God’s alone. This conception
suggests that suicide is viewed as a moral transgression
in which an individual “cheats” God by stealing the
latter’s power of intentionality in causing the self’s death.
Suicide therefore becomes a form of intellectual theft;
the self redesigns its end in an act of mutiny against its
creator.
Suicide must be distinguished from acts of martyrdom,
in which an individual engages in self-sacrifice as a political
or wartime strategy (e.g., “suicide bombers” or kamikaze
pilots) (Atran 2003). Even here, however, we see how
intentionality critically underlies the folk psychology of
souls. Although suicide is treated as a sin by many of the
world’s religions, including Islam, those who are martyrs
are seen by some religious adherents as having been
chosen by God to fulfill His wishes and as being rewarded
with special experiential luxuries in the afterlife. For
example, during World War II, one of the most intense
and successful military operations ever launched by
Japanese fighter pilots against an American fleet was
deemed “Operation Heaven” by the Japanese commander
(Blanchard n.d., p. 17), and “kamikaze” is literally
translated as “divine wind.”
The religious messages that are conveyed by charismatic
leaders may be especially seductive because they capitalize
on an innate teleological bias for ascribing a special
purpose to the unique self. For example, in a transcribed
television interview from CNN in 1997, Osama Bin
Laden commented that: “We believe that no one can
take out one breath of our written life as ordained by
Allah. We see that getting killed in the cause of Allah is
a great honor wished for by our Prophet.” When juxtaposed
with simulation constraints concerning what the
afterlife may be “like” for those who sacrifice themselves
for prosocial in-group reasons, this becomes a particularly
volatile social cognitive phenomenon since martyrs are
promised privileged states of consciousness after death.
As one member of the Palestinian group Hamas put it:
“By pressing the detonator, you can immediately open
the door to Paradise – it is the shortest path to Heaven”
(Hassan 2001).
4. Meaning, morality and the afterlife
Some say that we shall never know and that to the gods we are
like flies that the boys kill on a summer day, and some say, on
the contrary, that the very sparrows do not lose a feather that
has not been brushed away by the finger of God.
—Thornton Wilder, The Bridge of San Luis Rey
Because the adjudication of an afterlife of eternal reward
or damnation is seen as superseding “mere” human authority,
people’s understanding of the origins of moral deontology
– what one ought and ought not to do in
life – shows a strong cognitive bias toward belief in a
supernatural creator of human morality (rather than, for
instance, a bias toward belief in design by nature or
human whim). Reincarnation beliefs that rely on karmic
principles are no exception because such principles
require an intelligent designer of this morality-based
rebirthing cycle. This overall vein of reasoning helps to
explain why people expect divine retribution for moral
transgressions only, rather than, say, for breaches of
social etiquette (e.g., Roes & Raymond 2003). As Camus
(1943/1991) writes, “revolt against men is also directed
against God” (p. 94). From a genetic fitness perspective,
what is important is that it is moral transgression that
scars reputation most deeply and has the most costly
effect on future social relations (Goffman 1963) and therefore
behavioral compliance in this domain is critically
important.
Surprisingly, cognitive scientists who study religion have
given the topic of morality relatively short shrift. For
example, Atran and Norenzayan (2004) recently argued
that culturally acquired supernatural concepts (cf. Boyer
2001) receive emotional staying power because they are
lent support by an evolved hyperactive agency detection
device (see also Atran 2002; Barrett 2000; Guthrie 1993).
According to Atran and Norenzayan, this mechanism
serves the protective function of hyper-vigilance in potentially
dangerous environments, but as a consequence,
affectively primes individuals and causes them to overattribute
intentions to the natural world, such as might
happen when a branch falls in the forest. The authors
conclude that “supernatural agents are readily conjured
up because natural selection has trip-wired cognitive
schema for agency detection in the face of uncertainty”
(Atran & Norenzayan 2004, p. 720). In particular, supernatural
attributions occur because environmental stimuli
“achieve the minimal threshold for triggering hyperactive
facial-recognition and body-movement recognition
schemata that humans possess” (p. 720).
Atran draws on findings from developmental psychology
showing that agency overgeneralization is an innate
Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 459
feature of human cognition. For instance, in a variety of
controlled experiments using nonverbal measures,
Csibra and his colleagues have demonstrated that, if
causal cues indicating rational agency are present, even
infants see inanimate movement as purposive behavior
(e.g., 12-month-olds ascribed intentions to dots on a
computer screen moving about in a “rational” manner;
see Gergely & Csibra 2003).
Despite minor theoretical differences with Boyer’s evolutionary
model of religion, Atran (2002) is united with
Boyer and other cultural epidemiologists in denying that
religion is an adaptation. However, although the explanatory
utility of cultural epidemiology theory has been unrivaled
among recent attempts to explain the evolutionary
basis of religion, it has problems of its own. By focusing
on the role of concept acquisition, this work may be
overshadowing more fundamental questions about the
natural foundations of religion – questions raised in
section 3 (Souls and intelligent design). For example,
Atran and Norenzayan’s (2004) model fails to account
for people’s tendency to assume that supernatural agents
are responsible for traumatic life events (Deridder et al.
1999; McAdams 2001; Pepitone & Saffioti 1997; Weeks &
Lupfer 2000). How can reasoning about the supernatural
causes of, say, suffering a miscarriage, being felled by
disease, or losing a loved one in an accident be triggered
by facial-recognition and body-movement recognition
schemata? There are no such environmental cues
capable of breaking the “hair trigger” of the authors’
proposed sensory driven hyperactive agency detector, yet
supernatural attribution occurs (arguably even more so
than for the exemplar hair-trigger cases).
Another approach to solving the riddle of religion is to
address whether the self’s view that it is something more
than a material body subject to the mindless and amoral
laws of nature is a product of natural selection (Dennett
1991; 1995). Evolutionary scholars in this area might
then begin to shift the primary theoretical frame from
one that centers on concept acquisition and agency detection
to one that centers on models of self representation,
morality, and meaning (Bering 2002b; 2003b). Although
these approaches likely reflect complementary levels of
analysis rather than alternative theoretical models, the
cultural epidemiology approach has, to date, not successfully
bridged the representation of supernatural concepts
with the Darwinian currency of behavior. What is required
to bridge this gap is the self, a conspicuously absent entity
in the cognitive science of religion.
4.1. “Signs”: Ascribing meaning to natural events
The intentional stance is the strategy of interpreting the behavior
of an entity (person, animal, artifact, whatever) by treating
it as if it were a rational agent who governed its “choice” of
“action” by a “consideration” of its “beliefs” and “desires”
. . . the basic strategy of the intentional stance is to treat
the entity in question as an agent, in order to predict –and
thereby explain, in one sense – its actions or moves.
—Daniel Dennett (1996), Kinds of Minds
If people naturally endow the events of their lives with a
hidden purpose, the self may then hold expectations
about the “behaviors” of supernatural causal agents,
canonical expectations that conform to standard rules of
fairness and justice. Research on just-world beliefs
shows that people indeed operate under the assumption
that others “get what they deserve,” especially when they
have little control over negative outcomes and when
help cannot be meted out to unfortunate innocents (for
a recent review of this literature, see Hafer & Be`gue
2005). Although just-world researchers have not generally
sought to interpret related religious notions, often implicit
in this type of causal reasoning about fortune and misfortune
is the idea that some behavior in the moral domain is
connected to an unrelated, uncontrollable life event.
Therefore, a central question is “who” is represented as
tallying up our deeds and as meting out just deserts in
the form of positive and negative life events (in whatever
ontological domain these happen to be administered).
Bruner (1990) has argued that, in everyday social
psychology, individuals will search for meaning whenever
others’ behaviors violate their expectations, or otherwise
fail to adhere to sociocultural scripts. For example,
subtle breaches of conversational maxims, such as non
sequiturs or other types of “conversational implicatures,”
will often encourage a search of the partner’s intentions
(see also Baldwin & Moses 1996; Baron-Cohen et al.
1997). Whenever unexpected autobiographical events
occur, individuals may similarly seek to identify the intentions
of the supernatural agent who has caused these
events (or at least allowed them to happen), because this
is presumably a purposeful agent who adheres to unwritten
rules of social reciprocity (Bering 2003b). In Nazi
Germany, for instance, some Holocaust survivors reportedly
thought that God had gone insane, since clearly he
had breached the most basic of social agreements with
his followers (see Wiesel 1961).3
This belief in a just world is so strong, in fact, that among
many groups personal calamities and hardships are taken
as evidence that the individual must have done something
horribly wrong. Often the only suitable remedy for these
hardships is spiritual excision by way of public confession.
Among the Igbo of Nigeria, for example:
[a]dultery by a wife is regarded as bringing supernatural punishment
upon herself and her husband . . . thus if a woman
experiences difficult labor, it is assumed that she has committed
adultery and she is asked to give the name of her
lover in order that the child be born. If a man falls sick, his
wife may be questioned as to whether she has committed
adultery. (Ottenberg 1958/1980, p. 124)
There may also be “nonreligious” developmental precursors
to this moralistic interpretation of uncontrollable
negative events. Piaget (1932/1965) argued that young
children evidence a belief in immanent justice in which
“the child must affirm the existence of automatic punishments
which emanate from things themselves” (p. 251).
Thus, in a classic study, Piaget (1932/1965) presented
children aged 6–12 years with the story of a child who
steals or disobeys and then, upon crossing a bridge, falls
into the water when the bridge collapses. Nearly all
(86%) of the youngest children in the study reasoned
that the accident would never have happened were it
not for the character’s earlier misdeeds.4
Indeed, people who have violated some moral rule often
appear expectant of existential punishment, and those
guilty parties who find themselves untarnished by their
wrongdoing may feel as though their current happiness
is undeserved. This is another common theme in
Bering: The folk psychology of souls
460 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
literature, exemplified by the works of Victor Hugo (e.g.,
Les Miserables) and Fyodor Dostoyevsky (e.g., The Brothers
Karamazov, Crime and Punishment). Landman (2001),
a narrative psychologist, tells the true story of a fugitive
who drove the getaway car in a heist that left a security
guard dead. Decades later, this woman “obsessed with a
desire to be punished, to seek expiation” (Franks 1994,
p. 54) turned herself in to clueless authorities. Asked
why she confessed, the now model citizen told her
lawyer that “my strongest weapon against suicide is my
contract with God . . .” (Franks 1994, p. 42).
In many societies, not only is supernatural punishment
envisioned to fall directly upon the heads of the wicked,
but is also believed to be sanguineous. Some supernatural
agents are seen as unforgiving and merciless, inflicting
lasting and far-reaching punishments across generations
(Bering & Johnson 2005). Perhaps the worst punishment
of all would be for one’s biological relatives, especially
offspring, to be cursed for the self’s misdeeds. This is a
particularly recurrent theme and is illustrated very
clearly in the following brief passage on the Pagai from a
Dutch missionary publication:
A missionary once acted emphatically against various [superstitious]
prohibitions in order to demonstrate their inefficacy.
Actually this made a totally wrong impression on the people
because they said: “The man knows perfectly well that he
himself won’t be punished but that the punishment will fall
on his children.” (Anonymous 1939, p. 9)
Recent laboratory findings suggest that there may be
identifiable cognitive developmental milestones that
promote the pan-cultural human tendency to see “signs”
or hidden messages in natural events. In a study by
Bering and Parker (in press), 3- to 9-year-old children
were informed that an invisible agent (Princess Alice)
would help them play a forced-choice game by “telling
them, somehow, when they chose the wrong box,”
whereas a matched control group of children were not
given this supernatural prime. On two unexpected event
trials, as soon as the child chose a box, an experimenter
triggered a simulated unexpected event in the laboratory
(i.e., a light turning on/off; a picture falling), and children’s
response to these events (i.e., moving their hand
to the opposite box) was coded. Thus, the study sought
to determine the age at which children first begin to
view natural events as being about their behaviors and as
stemming from the mind of a communicative supernatural
agent.
Results showed a significant interaction of age group by
experimental condition. The only children to reliably move
their hand to the opposite box in response to the unexpected
events were the oldest children (M ¼ 7 years, 4
months) who were primed with the invisible agent
concept. Whereas 82% of these oldest children assigned
to the experimental condition (and therefore told about
Princess Alice) changed their response, only 18% of
same-aged children in the control group (who were not
told about Princess Alice) moved their hand to the opposite
box after the unexpected event. For children’s posttest
verbal explanations, also, only the oldest children
from the experimental group saw the unexpected events
as being referential and declarative (e.g., “Princess Alice
did it because I chose the wrong box”). In contrast,
younger children (M ¼ 5 years, 6 months) saw the event
as being intentionally caused by the invisible agent
(e.g., “Princess Alice did it because she wanted to”),
whereas the preschoolers (M ¼ 4 years, 1 month) did
not invoke the invisible agent at all, but only physical
causes for the event (e.g., “The picture fell because it
wasn’t sticking very well”).
Although the cause of these age differences is controversial,
these findings nevertheless demonstrate that the
tendency to over-attribute intentions to the natural
world is not simply a matter of hyperactive agency detection
(e.g., Atran 2002; Atran & Norenzayan 2004; Barrett
2000; Guthrie 1993), but rather it also involves, at least
in older children and adults, making inferences of
communicative meaning within a social context. In this
case, the specific supernatural agent concept (Princess
Alice), which may be a rough analogue of culturally
specific supernatural agents, appeared to map onto this
inferential capacity for seeing signs in natural events.
Furthermore, these subjective inferences gave way to
objective behaviors, which is the primary currency of
natural selection. An event such as a picture falling to
the ground is not, in itself, a communicative event; it can
become so only through the phenomenal properties of
the child’s mind (“what is the meaning of the picture
crashing to the floor, precisely at this moment in time?”).
5. Conclusion
In reviewing the available – though still very limited –
evidence, there are good conceptual grounds to argue
that natural selection may have set to work on specific
human cognitive errors. These include simulation constraints
leading to Type I errors in reasoning about the
afterlife, teleo-functional errors leading to belief in the
soul’s intelligent design, and theory of mind errors fostering
a belief that natural events were intentionally caused
by supernatural agents. The resultant cognitive system
created the functional illusion that the social behaviors
of the self “mattered” outside of human relations. As a consequence
it became morally tamed under the auspices of
this existential rubric and therefore was less likely to
engage in acts that, if publicly exposed and harmful to
one’s social reputation, seriously impaired genetic fitness.
The present article has also served to lay out some
general future directions for investigators to more
precisely explore the Darwinian mechanisms at the heart
of the existential system outlined here. Such work can
further reveal how the standard architecture of ancestral
human minds was co-opted by natural selection to create
the functional illusion of an intelligently designed, immortal
soul that was under nearly unbreakable moralistic
contract with the natural world.
ACKNOWLEDGMENTS
I am grateful to the following individuals for their careful reading
and comments on earlier versions of this article: Joseph Bulbulia,
Dominic Johnson, Shaun Nichols, Harvey Whitehouse, Nicholas
Humphrey, and Paul Bloom. Also, I would like to thank several
people who shared with me their thoughts and ideas on the
topics discussed herein, including David Bjorklund, Todd
Shackelford, and Juan Quiles. Jared Piazza pointed out to me
Gay’s epitaph quoted at the introduction of this article and
Dan Gilbert informed me of Freud’s remarks on the “afterlife”
in Thoughts for the Times on War and Death. Finally, I wish to
Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 461
acknowledge Jeff Gredlein, without whose assistance I would
have been unable to complete this project.
NOTES
1. This is as in various physical resurrectionist beliefs, such as
the Anabaptist doctrine of “soul-sleep,” in which the soul is said
to hibernate, or lie in wait, until it may reanimate the physically
reconstituted body.
2. The simulation constraint hypothesis is indirectly supported
by recent findings of egocentric social cognitive biases
in adults (Epley et al. 2004). Epley and his colleagues found
that participants’ eye gaze preferentially moved to privileged
visual space in response to an experimenter’s ambiguous referential
communication. For example, the command “move the
bunny” elicited automatic eye gaze toward a stuffed bunny that
could be seen by the participant, but which was occluded from
the experimenter’s perspective, over a chocolate Easter bunny
to which both the participant and experimenter had visual
access. The authors argue that these findings show that egocentrism
is just as prevalent in adults as it is in young children. Adults,
however, more rapidly correct their egocentrism to adjust for
others’ limited knowledge (e.g., by quickly shifting their gaze
and moving the chocolate Easter bunny). If, as Epley et al.
(2004) reason, individuals do become better with experience at
making adjustments to correct for their initial egocentric views,
but then rely on simulation to revise their social attributions,
then even the best perspective-taking skills should falter when
it comes to reasoning about dead agents’ “perspective-less”
minds. This is because any attempt at correcting for egocentrism
by using simulation would still run up against simulation
constraints (e.g., “does he know that he’s dead?”) and generate
attributions of continued psychological functioning. Indeed,
this is what is generally found.
3. The atrocities of the Holocaust forced many survivors to
question God’s “benevolent” intentions, apparently prompting
some Jews to revise their theological views to accommodate the
possibility that God is in fact morally corrupt. Nowhere is this
theme more salient than in the semi-autobiographical chronicles
of Holocaust survivor Elie Wiesel. In Gates of the Forest, Wiesel
(1966, p. 197) writes:
In a concentration camp, one evening after work, a rabbi called
together three of his colleagues and convoked a special court. Standing
with his head held high before them, he spoke as follows: “I intend to
convict God of murder, for he is destroying his people and the law he
gave to them . . . I have irrefutable proof in my hands. Judge without
fear or sorrow or prejudice. Whatever you have to lose has long since
been taken away.” The trial proceeded in due legal form, with witnesses
for both sides with pleas and deliberations. The unanimous
verdict: “Guilty.” . . . [But] after all, He had the last word. On the day
of the trial, He turned the sentence against his judges and accusers.
They, too, were taken off to the slaughter. And I tell you this: if their
death has no meaning, then it’s an insult, and if it does have a
meaning, it’s even more so.
4. In his Bridge of San Luis Rey (1927/1955), Thornton
Wilder fictionalizes the sad tale of a collapsed bridge in eighteenth
century Peru that brought five travelers to their deaths
in the abyss below. In two chapters, one titled “Perhaps an Accident”
and the other titled “Perhaps an Intention,” Wilder
describes how the resident monk, Brother Juniper, troubled by
the seeming arbitrariness of this horrific event, embarks on a
“scientific experiment” to reveal why God chose to end the
lives of these five people rather than some other five, by collecting
and analyzing the facts and details of each person’s value in terms
of goodness, piety, and usefulness. Alas, “the thing was more difficult
than he had foreseen” and his quest for spiritual understanding
went unresolved. In a case of life imitating art, 14
people lost their lives in 2001 when a runaway tugboat rammed
two barges into an interstate bridge and caused about a dozen
cars to collapse into the Arkansas River. One of the victims was
a young army captain and father of four from California on his
way home to Virginia. The Oklahoman newspaper reported
that his commanding officer, echoing the thoughts of Brother
Juniper, “pondered the odds of making a 2,929-mile drive and
landing on a 500-foot stretch of bridge that, in the most bizarre
of accidents, plummeted precisely as he crossed it. ‘If [he] just
stopped at a rest stop or stopped to get gas . . . There’s just so
many variables—and the timing.’” (Owen 2002).
Open Peer Commentary
Simulation constraints, afterlife beliefs,
and common-sense dualism
Michael V. Antony
Department of Philosophy, University of Haifa, Haifa 31905, Israel.
antony@research.haifa.ac.il
http://philo.haifa.ac.il/faculty_pages/antony.htm
Abstract: Simulation constraints cannot help in explaining afterlife
beliefs in general because belief in an afterlife is a precondition for
running a simulation. Instead, an explanation may be found by
examining more deeply our common-sense dualistic conception of the
mind or soul.
Early on in his stimulating target article, Bering notes that the
ability to conceive of an afterlife requires a dualistic conception
of the relation between the conscious mind or soul and the
body; and he is sympathetic (as I am also) to the idea that our
common-sense concept of the mind/soul is dualistic, and in all
likelihood innate. An important question for Bering is
“how . . . we get from the common-sense dualism of infants to
beliefs of the afterlife [ . . . ]” (target article, sect. 1, para. 4).
And a major part of his answer is given by his “simulation constraint
hypothesis,” the idea that afterlife beliefs are explained
by our attempts to mentally simulate “what it’s like to be dead”:
putting ourselves “into the shoes” of dead agents, we are
compelled to ascribe to them mental states.
While simulation constraints may help explain the specific
types of mental states we project into the afterlife (as Bering
argues), I do not think they can help explain why people
believe in an afterlife in the first place. The point of a mental
simulation, after all, is to generate conclusions about an agent’s
mental states or behaviors (with the type of simulation run
depending on the types of mental states or behaviors about
which one wishes to derive information). The cognitive mechanisms
involved in planning simulations, accordingly, must assume
the existence of a mind – namely, that mind into the nature of
which one aims to gain insight through simulation. But this
must hold for the afterlife case too: prior to simulating a dead
agent’s mind, it must be assumed there is a mind to simulate.
But that already is to assume an afterlife. This mind/soul may
be taken to be phenomenally rich, or relatively barren (experiencing
“darkness,” “nothingness,” or what have you), but it must be
taken to exist, at least implicitly. Notice that Bering seems to
grant this in referring to “simulation strategies to derive information
about the minds of dead agents” (sect. 2.1, para. 1,
emphasis mine). It follows that nothing about a simulation
itself can explain our belief in an afterlife, since some such
belief or assumption is a precondition for the planning and
running of any such simulation.
If that is right, how might afterlife beliefs be explained? I
believe that the route from our common-sense dualism to
Commentary/Bering: The folk psychology of souls
462 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
afterlife beliefs is considerably shorter than Bering supposes.
Afterlife beliefs may fall out quite directly from how our
common-sense dualism is conceived. It may follow from our
dualism that the destruction of a person’s body has no bearing
whatsoever on the existence of his or her mind/soul – much as
it is entailed by my common-sense conception of the apple and
orange in my refrigerator that eating the apple will leave the
orange intact. Most of the work in explaining afterlife beliefs
on this view, therefore, will be done by a detailed account of
our concepts of our body, mind/soul, and their interrelations
(and how the question of an afterlife arises).
Regardless of the extent of the gap between our common-sense
dualism and afterlife beliefs, discovering how one gets from the
former to the latter will require a detailed characterization of our
dualist conception, something we currently lack. We would thus
do well to examine the features of our conception of the mind/
soul that are implicated in our conceiving the mind and body as
distinct. It will not do simply to say that we conceive of the body
and soul as ontologically distinct, and leave it at that, because we
must understand the particular type of distinctness involved, and
how it is grounded in the concepts of body and soul. (Objects
and events are also ontologically distinct categories, but are interdependent
in ways that bodies and souls are not.) We should
examine our conceptions of ourselves as conscious beings, selves,
experiencers, and “witnesses”; of the mind/soul as being essentially
private, “internal,” subjective, or phenomenal. For it is something
about these conceptions, arguably, that makes the mind/soul seem
so utterly unlike anything physical, that destroying the body can
leave the mind/soul intact.
One way to tap children’s understanding of the privacy and
“innerness” of conscious phenomena is to explore children’s
understanding of dreams, imagery, and sensations, conceived
of as private and “internal.” I shall hint at some possible directions
for research, with a few anecdotes. (Since I am not a
psychologist, they should be taken with a grain of salt; with that
said, their purpose is merely to illustrate some questions for
investigation.) At age three, my daughter appeared to understand
the idea that dreams involve “pictures in her head,” and seemed
able to sing her favorite song “in her head” and report when she
had finished. She insisted that others could not see the pictures
or hear the sounds “because they were hers,” and found the suggestion
that others might see them or hear them silly. (Interestingly,
she also insisted that she did not see the pictures in her
head; they were just there.) This conception of privacy also
applied to sensations like pain. Also at age three, she went
through a brief stage of lying about having hurt herself (for
sympathy, hugs, etc.) when noticing her baby sister receiving
attention. That she confidently lied about feeling pain in the
presence of others suggests she believed her sensations were
accessible only to herself.
Another matter to explore is children’s capacity to conceive of
objects, properties, and events in their experience as merely
phenomenal. I have in mind the capacity to grasp that what
appears in dreams is “not real,” as well as grasping the concepts
of hallucination (including radical hallucination, as in “The
Matrix”), illusion, after-images, and the appearance–reality
distinction more generally. One way to get at some of these
issues may be to probe children’s understanding of “inverted
qualia,” the idea that what you visually experience when looking
at objects we both call “blue,” for example, might be qualitatively
very different from what I experience when looking at those
objects. This idea can be explored intrasubjectively by adapting
one of a child’s eyes to bright light, and then having the child
look at a uniformly colored object one eye at a time. The
object’s color will appear to alternate between two different
shades. Assuming the child does not infer that he or she is causally
affecting the object by blinking, some understanding of the
concept of phenomenal color might be expected to reveal itself.
Conceptual abilities of these sorts enable Descartes to doubt
awaythe physical world while his mind/soul (plus phenomenology)
remains. This is a first step in Descartes’ argument for dualism.
However, it is also already very close to the idea of an afterlife,
since it is the idea of a mind/soul existing without the physical
world. This too suggests that the distance from our commonsense
dualism to afterlife beliefs may be short, at least if our
common-sense dualism is Cartesian in relevant respects.
Social cognition of religion1
William Sims Bainbridge
Division of Information and Intelligent Systems, National Science Foundation,
Arlington, VA 22230.
wbainbri@nsf.gov http://mysite.verizon.net/wsbainbridge
Abstract: Research on religion can advance understanding of social
cognition by building connections to sociology, a field in which much
cognitively oriented work has been done. Among the schools of
sociological thought that address religious cognition are: structural
functionalism, symbolic interactionism, conflict theory, phenomenology,
and, most recently, exchange theory. The gulf between sociology and
cognitive science is an unfortunate historical accident.
Bering is entirely correct that religious beliefs can help us understand
the evolution of human social cognition, but I would go
further to say that research on religious cognition could
become the first span of a substantial bridge between the cognitive
and the social sciences. Broad territories in my own field,
sociology, are cognitive in nature, and I would venture to say
that the largest troves of systematic data relevant to religious
cognition have been collected by sociologists. Some of these
data are freely available, such as the General Social Survey
(sda.berkeley.edu) or the many questionnaire datasets at The
Association of Religion Data Archives (www.thearda.com).
One function of Bering’s article is to alert readers to the
impressive group of cognitive or developmental psychologists
and cultural anthropologists who have done so much good
work on religious cognition over the past decade. This group,
however, has ignored vast bodies of relevant social science literature,
probably for two reasons. First, any new school of thought
needs to mature in intellectual isolation, until its ideas are sufficiently
well developed to stand critical scrutiny. We can call
this the allopatric principle of cultural innovation, by analogy
with allopatric speciation in biology: New cultural movements
develop more readily under conditions of social isolation from
existing movements.
Second, sociology, political science, to some extent economics,
and even important portions of social psychology remained aloof
thirty years ago when the multidisciplinary field of cognitive
science was being formed. This tragedy was largely the result
of misunderstandings and prejudices, augmented by turf
defense and an unwillingness to do the hard work required to
bring the disciplines together. Major schools of thought in sociology
– structural functionalism (Parsons et al. 1951) and symbolic
interactionism (Blumer 1969) – were predominantly
cognitive, emphasizing concepts such as overarching values,
social roles, group identity, and definitions of the situation. But
these approaches made little use of rigorous statistical methodologies,
and thus may not have seemed “scientific” enough to
be included in cognitive science. For all its emphasis on ideology,
the Marxist movement that was so influential in sociology claimed
to be materialist, an example of false consciousness if ever there
was one. However, Marxism informed conflict theory, and a cognitive
scientist can draw from that broader tradition an awareness
that sometimes language and even cognition itself may be moves
in a game of social power (Habermas 1971).
A psychologist seeking cognitive research in sociology might
find it in unexpected places. Cognitive scientists tend to
dissociate themselves from behaviorism, which in psychology
disparaged speculations about internal mental states. However,
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 463
in sociology behaviorism was remarkably cognitive in nature, as
illustrated by the extended analysis of the exchange of advice
for approval in Social Behavior by George C. Homans (1974).
Influenced by Homans, later sociologists developed the exchange
theory or rational choice explanation of religion: Humans seek
many rewards that are not available, following cognitive explanations
that become progressively supernatural in nature as the
humans continually fail to attain the deeply desired reward. If
the recent cognitive theories of religion lack an essential ingredient,
it is the motivation that drives people to act upon religious
cognitions, and to build complex and costly religious institutions.
Sociological exchange theory often makes use of artificial intelligence
computer simulation. This methodology has been applied
profitably to religion, and one direct reinforcement neural net
program showed that deprivation can cause an agent to
develop minimally counterintuitive beliefs (Bainbridge 2006).
Phenomenological sociology and its cousin ethnomethodology
are among the least rigorous approaches, but they still may have
something to contribute. Bering’s reports about how people
conceptualize death are reminiscent of the insightful early work
by theorist Alfred Schutz about the phenomenology of time.
Schutz is especially famous for his work on multiple realities,
which can be distinguished because their subjective flow of time
is different, and religious experiences are a case in point (Schutz
1971). Less well known is his theory that humans conceptualize
the future as a kind of past, seen as if it had already occurred
(Schutz 1967), a contradiction not unlike that when people conceptualize
a dead person: Dead is to alive as future is to past.
Potentially relevant empirical research in sociology is of many
kinds, including historical accounts of the thoughts of religious
leaders, ethnographies of religious movements, and a very well
developed tradition of questionnaire research. Bering discusses
suicide, and official statistics have been analyzed in ways relevant
to cognition, suggesting that the power of faith to deter suicide is
declining in advanced societies (Bainbridge, in press). Given
Bering’s emphasis on death, it is worth noting that the General
Social Survey contains several questions about how people conceptualize
the afterlife, and that the same questions have been
administered to members of radical religious groups, allowing
comparisons of such beliefs as how erotic the afterlife is
(Bainbridge 2002).
Bering talks about morality, but does not introduce the
extensive quantitative research on how religious faith does or
does not shape behavior. Especially relevant is the research on
juvenile delinquency. Consider the phenomenon I call the
Stark effect, because Rodney Stark discovered it: “Religious individuals
will be less likely than those who are not religious to
commit delinquent acts, but only in communities where the
majority of people are actively religious” (Stark 1996, p. 164).
That is, in primarily secular communities, adolescents who
believe in supernatural sanctions for misbehavior are just as
likely as their irreligious peers to steal or vandalize property. In
communities where the majority of adolescents are religious,
the beliefs of the individual child are indeed predictive. Thus,
cognition alone may not deter antisocial behavior.
A further complication is that many studies show that the
Stark effect does not apply to hedonistic behaviors, and religious
adolescents are less likely to use drugs or engage in sexual
experimentation even in very secular areas. Perhaps religion
serves an advisory function, helping to guide the adolescent’s
cognitive deliberations away from danger (Bainbridge 1992).
This research area is still unsettled, and studies by cognitive
scientists would be especially welcome.
NOTE
1. The author of this commentary is employed by a government
agency and, as such, this commentary is considered a work of the U.S.
government and not subject to copyright within the United States.
However, any opinions, findings, and conclusions or recommendations
expressed in this commentary are those of the author and do not necessarily
reflect the views of the National Science Foundation.
Parenting, not religion, makes us into moral
agents
Benjamin Beit-Hallahmi
Department of Psychology, University of Haifa, Haifa 31905, Israel.
benny@psy.haifa.ac.il
Abstract: The universal early experience of all humans, which means
being totally dependent on caretakers who attempt to inculcate impulse
control, should be considered as the psychological framework for the
creation of significant supernatural agents. The same early experiences
put us at the center of a moral universe, but there is no necessary
connection between the two processes. We do not need disgruntled
ancestors to make us behave; disgruntled parents will do.
“What came into existence beside the dead body of the loved one
was not only the doctrine of the soul, the belief in immortality
and a powerful source of man’s sense of guilt, but also the earliest
ethical commandments” (Freud 1915, p. 295). Bering, like
Freud, ties religion to death and morality, and makes the startling
theoretical claim that we should regard religion as an evolutionary
adaptation, because it supports viewing the self as a moral
agent. The universal tendency to tie misfortune (and blessings)
to supernatural agents buttresses group cooperation and thus
has great evolutionary value. One problem with this notion is
theoretical, and has to do with cooperation and reputation
effects. According to Ohtsuki et al. (2006), cooperation is a fundamental
aspect of all biological systems, and among humans it
can evolve even in the absence of reputation effects, but
Henrich (2006), points out that the reputation effect may act to
stabilize maladaptive and immoral behaviors.
Accounting for the parallel development of morality and
religion should involve both panhuman experiences and innate
tendencies. Both innate architecture and panhuman socialization
processes lead to the universal perception of the self as moral
agent. Evolved architecture leads, indeed, to an innate readiness
to over-detect causality and intentionality. The three kinds of
behavior described here – supernatural agents, ghosts, and
magical thinking (“Princess Alice”) – can all be accounted for
by the general hyperactive agency detection mechanism, which
operates to detect not just biological processes, or activity, but
another consciousness or another mind. The survival value of
detecting, and negotiating with other minds is so great that it
accounts for this hyper-vigilance. Friend–foe identification
enables us to be cared for and then take care of others.
Our early experience of our own consciousness and that of
other conscious beings leads to our belief in the enormous
power of the mind, our eternal soul. It was William James who
already stated: “Religion, in fact, for the great majority of our
own race, means immortality and nothing else” (James 1902,
1961, p. 406). Souls are important because they give us more
information about promised immortality. The supernatural
premise is fleshed out, so to speak, by enumerating the entities
in the spirit world, most of whom must be human souls before
birth and after death.
Some dead agents are psychologically important because we
have known them and interacted with them; they informally
join the pantheon outside the official hierarchy. The author’s
analysis of the role of the souls of the ancestors ignores the fact
that dead ancestors were once live parents.
Our innate architecture also produces egocentrism, attachment-seeking,
and the panhuman process of socialization. The
human baby is hard-wired to seek a caretaker and find security
as soon as it comes out of the womb (Bowlby 1973). The baby’s
helplessness is matched by the caretaker’s readiness to create it
in her own image. Socialization of the young aims at impulse
control. They are asked to reduce their egocentrism and impulsivity
in return for parental love. Whatever we call morality is
tied to powerful bonds developed between children and caretakers.
In all cultures, love is finite and conditional, and punishments
and the withdrawal of love are frequent and swift.
Commentary/Bering: The folk psychology of souls
464 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Children are made into moral agents easily through
socialization and social control mechanisms, as they are assigned
blame and learn to blame others and especially themselves. The
panhuman experience is that parents are the carriers of morality,
as they convey to their children a fantasy of a world ordered into
right and wrong, reward and punishment. The moral universe we
all inhabit was developed in early childhood in our private
consciousness, and it may be projected on the universe.
Do we need religion to support behavioral inhibition, as
Bering claims? The basic pattern of socialization precedes the
use of religious ideation. We are afraid of mother and father
because they punish us, long before they become the souls of
the ancestors. References to divine authority are sometimes
used by parents to bolster their authority in disciplining children.
Thus, the parents become allied with divine authority. Examples
can be found in all cultures (Geertz 1960). Nunn (1964) found
that this “coalition” with divinity was prevalent among parents
who were ineffectual and powerless.
Bering correctly points out that the connection between
morality and religion is rarely addressed in the behavioral literature.
But this is true not only for cognitive science or psychology,
but also for history, anthropology, and sociology, and with good
reason. Showing that religion has any consequences in prosocial
behavior has not been easy.
The four-dimensional model of religiosity (ideological, ritualistic,
experiential, and intellectual) so often used, originally had a
fifth dimension – the consequential – designed to measure the
effects of religiosity on conduct in other spheres. It was
dropped because consequences in nonreligious behavior could
not be found (Beit-Hallahmi & Argyle 1997). Modern findings
indicate that religion does have a considerable effect on secular
behavior in two areas: sex and the use of illicit drugs. Generalizing
beyond these specific areas has been difficult (Beit-Hallahmi
& Argyle 1997). We should note that not only has the academic
study of religion ignored morality, but the academic study of
moral development has largely ignored religion, with no apparent
damage or deficits.
The global secularization process means that we no longer
interpret misfortune as caused by supernatural agents. Some
misfortunes (earthquakes) result from inanimate natural forces,
and are viewed as unrelated to any moral agents (Neiman
2004). What can be morally condemned is human action. We
can condemn human cruelty, and we do. Moreover, the Enlightenment
has led to a new, totally secular, public discourse about
morality, focusing on justice, rights, equality, and human
welfare, and marginalizing the impulse control discourse. This
sea change in our moral outlook is evident in the Universal
Declaration of Human Rights (1948) and in such secular
groups as Amnesty International and Medecins sans Frontie`res.
Lifelong atheists have been found to be well-socialized, lawabiding,
and nonviolent (Beit-Hallahmi 2006). They obviously
go through blaming themselves, and others, and feel guilty.
They may anthropomorphize their diseases and inclement
weather. They have have gotten all that from their evolved architecture
and early experience, without the help of any functional
religious illusions.
Religion and morality: An anthropological
comment
Maurice Bloch
Department of Anthropology, London School of Economics, London, WC2
2AE, United Kingdom; College de France, Chaire Europe´enne, 75724 Paris
Cedex 05, France.
M.E.Bloch@lse.ac.uk
http://www.lse.ac.uk/collections/anthropology/mauricebloch.htm
Abstract: This commentary criticises Bering on two counts. First,
because we do not know what he attributes to natural selection and
what he sees as derived representations. Second, Bering’s ethnography
of religion is inadequate. People who practise ancestor worship are not
concerned with their own survival but with that of others. Many
supernatural beings are not thought of as morally motivated.
Jesse Bering’s article is nothing if not ambitious. It starts with an
experiment showing that young children attributed psychological
functions to a recently dead mouse and ends up with a proposal
that amounts to saying that religion, especially religion of a kind
which is strikingly similar to what Harold Bloom has called
“American Religion” (Bloom 1992), is innate, since it is the
product of natural selection. That is a long way to go, especially
because, as far as I am aware, most religious systems are not
much concerned with the survival of the souls of rodents.
The connectinglinksinthe argument arethat humans are natural
dualists because they inevitably have a belief in the survival of some
elements of agency on the part of the dead, who consequently are
attributed with mental states; that this leads to similar beliefs
about other supernatural beings such as God; that the existence
of supernatural agents give meaning to the individual self in the
world; that these beliefs make people behave morally; and that
this is good for them and their inclusive fitness because a reputation
for morality leads others to treat you and your offspring well.
I react to this proposal in two ways: first, as a critic of the
general theoretical issues raised by Bering; second, as a
traditional anthropologist who wants to test the theory against
the ethnographic record.
Bering claims to give us an evolutionary account that, in certain
respects, is critical of theories such as those of Boyer, Sperber, and
Atran because these seem to make religion the product of an
unfortunate malfunction of our cognitive apparatus. This indeed
seems odd for such a central feature of human culture, and I
too have been critical of these writers for linked reasons (Bloch
2002). Nonetheless, Boyer and those others seem to me to have
an advantage over Bering in an important respect, that is, they
are very clear as to what they are claiming. I do not find such
clarity in Bering. Following such writers as Tooby and Cosmides,
he tells us that natural selection has led to the evolution of certain
necessarily innate characteristics, which, as dispositions, have
affected the history of culture. Such a proposal requires a strict
division between that which is claimed to be innate, the product
of natural selection, and the cultural phenomena that have been
affected by such innate dispositions – inevitably together with
many other factors. It is precisely this distinction between
exactly what it is claimed natural selection has bred in us and
the derived factors that are affected by these inherited mechanisms
which is missing. Is it the morality/religion complex that
has been selected for in the distant past of the species? Or is it
merely the belief in the survival of the dead? I am not sure. For
example, Bering tells us, inter alia, that “people naturally endow
their lives with a hidden purpose.” What does “naturally” mean
here? Does that mean that natural selection has made us into
beings that “naturally” think in this way? Are we to understand
that certain other dispositions inevitably, but indirectly, lead us
to see our lives in this way? I do not feel I know.
Now let us turn to more ethnographic matters. At least two key
elements in the argument are to me unacceptable. One concerns
the characterisation of ancestors, the other concerns the characterisation
of supernatural beings in general.
Even if we accept that it is frequent to believe in some aspect
of the continuing functionality of the psyche of dead people, this
does not mean that these are regularly represented as concerned
with morality. However, it is this involvement which, for Bering,
would explain the selective advantage of such a representation
and therefore its frequent recurrence. For example, the Trobrianders
famously described by Malinowski have elaborate
beliefs in various forms of life after death, but they do not
believe for all that, that the dead enforce the morality of the
living (Malinowski 1916). There are many other examples of
such moral indifference, as in the case of ancient Greece as
described in The Iliad. Even in Catholic Christianity dead souls
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 465
of ordinary people are not concerned with the morality of their
descendents, and the early fathers of the church actually went
out of their way to eradicate such “pagan” beliefs.
Those people who do believe in ancestors as some sort of moral
police are also very different from what Bering seems to assume.
Such people are not concerned about what happens after their
own death; what matters is what the souls of already dead
people might do to them if they are displeased. This makes
Bering’s argument about the importance of the belief in
intelligent design for one’s own behaviour irrelevant. And, even
then, ancestors are rarely concerned with maintaining a universal
morality; they are concerned with punishing or rewarding actions
which ensure their own selfish reproduction via their descendants.
This interest in their own inclusive fitness is not particularly
altruistic and often overrides the interests of their own
descendants (Fortes 1959). The ethnographic record of beliefs
in an afterlife therefore gives us a quite different picture to
that suggested in the target article.
This is equally true of Bering’s characterisation of god-like supernatural
beings. The author seems to assume that supernaturals are
invariably on the side of good and against evil. This is to forget that
such creatures as devils and witches are on the side of evil. Even
more commonly, supernaturals are represented as neither good
nor evil, but as simply unconcerned with moral issues, though
their very existence certainly is believed to cause trouble. This
is the case, for example, of the nature spirits common in Africa,
of the spirits of aborted foetuses in Japan, and of the ancestral
spirits of Amerindians. Similarly, there are many cases of
supreme gods, such as the famous African otiose gods, who also
are characterised by indifference and arbitrariness (Forde 1954).
Prosocial aspects of afterlife beliefs: Maybe
another by-product
Pascal Boyer
Departments of Psychology and Anthropology, Washington Unversity,
St. Louis, MO 63130.
pboyer@artsci.wustl.edu http://artsci.wustl.edu/pboyer
Abstract: Bering argues that belief in posthumous intentional agency
may confer added fitness via the inhibition of opportunistic behavior.
This is true only if these agents are interested parties in our moral
choices, a feature which does not result from Bering’s imaginative
constraint hypothesis and extends to supernatural agents other than
dead people’s souls. A by-product model might handle this better.
Bering’s brilliant unpacking and explanation of afterlife beliefs
includes the claim that a disposition to such cognitive errors
may confer greater fitness by motivating prosocial (and inhibiting
opportunistic) behaviors (sect. 2.4). Indeed, in most cultures,
beliefs in dead agents are associated with moral feelings.
However, the particular evolutionary argument offered here
may not be the most parsimonious account of the evidence,
because (a) people associate morality with their supernatural
beliefs in many different ways, some of which do not mention
afterlife beliefs; and (b) more important, there is massive
evidence for these very same prosocial attitudes and inhibitions
outside of supernatural beliefs.
In some cultures people construe morality in terms of a code
given by the gods or a single god or ancestors or a specific cultural
hero; in other cultural environments they express moral norms in
terms of similarity to the behavior of paragons such as heroes or
gods; in other places the norms derive from constant interaction
with spirits or gods or ancestors; and in many places people mix
all three modes (Boyer 2001). This is a problem for Bering’s
account. Such diversity suggests that the association between
morality and supernatural beliefs is rather ad hoc, perhaps best
seen as a relevant, attention-grabbing and inferentially powerful
combination of prior elements that evolved for different reasons.
Indeed, the evolution of prosocial behavior and moral feelings certainly
does not require supernatural beliefs. A whole suite of
prosocial cognitive mechanisms evolved in human beings. They
include for instance reputation-monitoring, whereby we construct
precise and dynamic databases about the reputational effects of
own and others’ actual behavior, as well as inferred dispositions
and character (Wojciszke et al. 1998); commitment signals that
evolved out of other forms of reliable, hard-to-fake signals and
provide information about likely future behavior (Nesse 2000); a
coalitional psychology that helps us maintain strong associations
of non-kin and manage interaction with rival coalitions (Harcourt
& de Waal 1992; Kurzban & Leary 2001); in-group strong reciprocity
whereby we suspend ordinary principles of exchange to create
a domain of valued and selfless interaction (Gintis 2000); ethnic
signals that help maintain the boundaries of this domain (Kuran
1998); commitment gadgets that help us tie our own hands to
force ourselves to behave non-opportunistically (Schelling 1960);
and moral feelings that provide immediate, negative emotional
rewards for opportunistic plans and thereby compensate the motivational
effects of the discount curve (Frank 1988). All these
dispositions and processes evolved independently of supernatural
and religious beliefs, operate in similar ways in people with or
without such beliefs and regardless of differences in these
beliefs, and recruit different neuro-cognitive machinery from the
supernatural imagination (Boyer 2003b).
So we seem to have plausible hypotheses for the independent
development, cognitive implementation, and evolutionary history
of (a) beliefs in supernatural agents (including dead people) and
(b) prosocial dispositions. This may help provide a parsimonious
“by-product” explanation of morally relevant dead agents.
If we accept the first part of Bering’s scenario, a set of cognitive
constraints lead us to construe dead people as intentional agents.
These constraints do not necessarily imply that the agents are
“interested parties” in our moral choices with “full-access” to
morally relevant information about us (Boyer 2001). But all
that is required to entertain concepts of such full-access agents
is an assumption that is already contained in many of our prosocial
cognitive mechanisms. The dispositions listed above all carry
the assumption that information about our own behavior is not
safely confined, that it may leak to other parties in unforeseen
ways, and that it is generally safe to assume in others more knowledge
of our decisions than can be observed. This assumption
itself is not terribly mysterious in origin. There is a cognitive
cost in computing the extent to which others do not share information
that is manifest to us, which is why understanding false
belief takes children more time than understanding belief, and
can be impaired by a variety of pathologies, as well as attentional
load or altered states. So the assumption that others know what is
manifest to us is a default value of our intuitive psychology more
than a special elaboration of it.
Given all these elements, it would seem that the notion of “fullaccess
supernatural dead agents with moral interest” develops
without much cognitive effort, as it only combines prior assumptions,
and has great inferential potential. In particular, it provides
an explanatory context in which one’s own moral feelings, the
outcome of implicit processes, may be readily explained. This
by-product scenario seems more parsimonious than the one
offered in the target article.
The principle of ontological commitment in
pre- and postmortem multiple agent tracking
Nicolas J. Bullot
Department of Philosophy, University of Toronto, Toronto, ON L5L 1C6,
Canada.
nicolas@objectcognition.net
http://www.nicolas-bullot.org/Publi/PubliFrame.html
Commentary/Bering: The folk psychology of souls
466 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Abstract: This commentary suggests that understanding the “Folk
Psychology of Souls” requires studying a problem articulating ontology
with psychology: How do human beings, both as perceivers and
thinkers, track and refer to (1) living and dead intentional agents and
(2) supernatural agents? The problem is discussed in the light of the
principle of the ontological commitment in agent tracking.
Jesse Bering’s article addresses fascinating questions that
certainly deserve to be studied in an interdisciplinary science
of the “Folk Psychology of Souls” (henceforth, FPS). Whereas
the author alludes to existentialist philosophy, he nonetheless
overlooks research in contemporary analytic philosophy about
two relevant themes: (1) the problem of reference (Campbell
2002; Evans 1982; Kripke 1980; Perry 2001; Quine 1960;
Strawson 1959) and (2) the problem of personal identity
(Locke 1689/1975; Merricks 2001; Olson 1997; Parfit 1984;
Rorty 1976; Shoemaker 1959). Understanding the FPS requires
studying this fundamental question: How do human beings,
both as perceivers and thinkers, track and refer to (1) living
and dead intentional agents and (2) supernatural agents such
as ghosts and gods? I name “tracking” the ability to trace,
follow up, or pursue over space and time a set of traceable
individuals; it is useful to distinguish perceptual tracking, in
which a target individual is directly tracked by a sensory-motor
system, from epistemic tracking, in which an individual is
spatio-temporally pursued by indirect epistemic means such as
communication and reasoning.
In several passages (e.g., about simulation, cognitive system),
the author seems to overlook the problem raised by the
multiplicity of skills and methods used by human beings to
track (1) actual living and dead agents and (2) fictional mortal
and immortal agents. An account of this multiplicity might
threaten the hypothesis that evolution has selected a unique
organized cognitive system dedicated to forming illusory
representations of psychological immortality and supernatural
agents. This multiplicity becomes apparent when one considers
how deeply the varied kinds of agent tracking depend upon the
multiple assumptions available about agents’ (purported)
ontology. By “ontology” I mean an implicit representation or an
explicit understanding of the birth, persistence, and survival
conditions of the tracked agent. Philosophers have distinguished
bodily (Thomson 1997; Williams 1970) and biological criteria
(Olson 1997) from psychological criteria (Parfit 1971; 1984;
Shoemaker 1959; 1999) capable of defining the survival of a
person, or intentional agent. As considered in the discussion of
sortal concepts (Carey & Xu 2001; Hirsch 1982; Pylyshyn 2003;
Wiggins 1997; 2001), subjects or cognitive systems performing
tracking must possess information about some uniquely
distinctive features of the tracked agent in order to direct their
agent-tracking attitudes and actions appropriately. This can be
expressed by this Principle of the Ontological Commitment in
Agent Tracking:
The skill or method that a human subject (or a perceptual,
cognitive system) s uses to track a unique target intentional
agent a are dependent upon the ontology that she (or it) ascribes
implicitly or explicitly to a. (Characters in italicized and bold
fonts are standing for proper names.)
The author’s hypothesis is that the ontological commitment
about the immortality of the soul of postmortem agents is the
“default cognitive stance” selected by evolution. I would like to
remark that even if the hypothesis were true, we would still
have to account for multiple ontological commitments in agent
tracking and multiple manners of referring to afterlife agency.
This problem is relevant to the target article because it is sometimes
difficult to determine which kinds of agent-tracking behaviors
are discussed by the author. Do they involve behaviors and
beliefs relating to interactions with the tracked immortal soul?
Do they involve beliefs about the possibility of localizing the
soul? What are the purported characteristics of individual souls
that guarantee their survival and traceability? What are the
relationships between visual tracking (Pylyshyn 1989; 2003)
and living/dead agent tracking (Bullot & Rysiew 2005)? Can
these relations be studied experimentally? Some of the previous
questions might have distinct answers in cultures that have
evolved differently (Richerson & Boyd 2005) and are upholding
different ontological commitments.
To focus on a precise case: the author mentions the “continued
social relationships with the dead” (sect. 2, para. 4). Such a phrase
is ambiguous with regard to ontological commitment and tracking.
If one accepts empirical realism, this continued social
reference can be of at least two different types (see Fig. 1): (1)
reference to, and physical interactions with, existing material
traces of a dead agent, or (2) reference to a fictional immortal
soul as in “common-sense dualism” (Bloom 2004). (This dichotomy
is reminiscent of the distinction between knowledge by
acquaintance and by description; see Russell [1912; 1918;
1956], Strawson [1959, pp. 18–20], Evans [1982, pp. 143–
203], Clark [2000, pp. 130–63] or Campbell [2002].) In type
(1), the acquaintance- or empirically grounded reference,
subjects are referring either to an actual agent a or to the material
traces left by him. In type (2), the description- or fictionally
grounded reference, subjects are referring to a nonexisting fictional
agent f such as Sherlock Holmes or a ghost. When facing
type (1), for example, if someone is heard having a discussion
about an individual named “a,” you can search for that particular
individual. In frequent cases, you may eventually find her and be
in a situation to perceive a’s organism and the surfaces/movements
that convey information about a’s mental states. Similarly
to the case of other kinds of individuals (Campbell 2002;
Pylyshyn 2003), perceptually tracking of a’s organism thus
opens a wide range of epistemic possibilities, such as verifying
propositions about a’s current properties via, for example,
demonstrative identification, prosthetic perception, and biometric
measures. Even after a’s death, it is usually still possible
to trace and reach a’s remains or possessions (think about archeological
investigations: a’s corpse is marked with perceivable
traits or scars that are historical vestiges, which act as evidence
of events in a’s life). These epistemic actions are not available
with fictional reference, for the characteristics of a fictional
agent can only be known by means of descriptions or imaginary
depictions. If f is a fictional character, any search of the referent
of the name “f” will end in a so-called “block” in the naming
network (Donnellan 1974; Perry 2001). The dichotomy is
Figure 1 (Bullot). Fundamental differences between tracking
actual and fictional agents.
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 467
essential for understanding the FPS, because each type implies
drastically different cognitive procedures: type (1) accesses a
realm of empirical and perceptual evidence that is ontologically
closed to type (2) and type (2) accesses a realm which rests on
descriptive resources and individual/collective imagination.
Ecological variability and religious beliefs
Adam B. Cohen, Douglas T. Kenrick, and Yexin Jessica Li
Department of Psychology, Arizona State University, Tempe, AZ 85287-1104.
adamcohen@asu.edu douglas.kenrick@asu.edu
yexin.li@asu.edu
Abstract: Religious beliefs, including those about an afterlife and
omniscient spiritual beings, vary across cultures. We theorize that such
variations may be predictably linked to ecological variations, just as
differences in mating strategies covary with resource distribution.
Perhaps beliefs in a soul or afterlife are more common when resources
are unpredictable, and life is brutal and short.
Religious beliefs, including those about an afterlife and omniscient
spiritual beings, vary across cultures (Cohen & Hall, submitted;
Cohen et al. 2003). This does not mean they are not
adaptations, because human behavior represents a continual
and dynamic interplay between flexible evolved mechanisms
and variable environmental inputs (Kenrick 2006; Kenrick et al.
2002). Rather, an evolutionary ecological perspective inspires
questions about whether variations in religious beliefs and practices
are adaptively keyed to variations in human physical and
social environments (ranging from food and shelter to social
structure: e.g., status hierarchies, access to mates, and geographical
distribution of kin relative to self). Cultural norms surrounding
sexual liaisons (often centrally incorporated into religious
beliefs) provide one illustrative case. Such norms vary widely,
with some societies and some religions sanctioning only monogamy,
many also accepting polygyny, and a small percentage
permitting polyandry. These variations correlate predictably
with physical and social ecology. For example, Tibetan families
in which one man marries one woman have fewer surviving children
than do families in which brothers pool their resources
(Crook & Crook 1988). By sharing one wife, brothers can preserve
the family estate, which would not even support one
family if it were subdivided each generation. Brothers in other
species also engage in polyandrous mating when resources are
scarce. Regarding polygyny, multiple women are particularly
likely to marry one man when several conditions converge: (1) a
steep social hierarchy, (2) a generally rich environment so one
family can accumulate vast wealth, (3) occasional famines so the
poor face occasional danger of starvation (Crook & Crook
1988). Under these circumstances, a woman who joins a large
wealthy family reaps benefits, even if she would have to share
her husband with other women. This pattern is also found in
other species. For example, indigo buntings vary between monogamy
and polygyny, but multiple females only pairup with the
same male when that male controls a resource-rich territory
and his neighbors have poorer territories (Orians 1969).
We wish to apply a similar analytic strategy to variations in
belief in souls and the afterlife. Different religions have very
different emphases on the importance of belief in an afterlife
(emphasized less by Jews, more by Fundamentalist Protestants,
for example; Cohen & Hall, submitted). And within a religion,
some individuals have much stronger beliefs in an afterlife than
others do (Cohen et al. 2005). Furthermore, there are vastly
different forms of belief in life after death, including reincarnation,
heaven and hell, ghosts, and so forth. Similarly, individuals
and cultures vary in views of God as vengeful and punishing
(Abramowitz et al. 2002). It is sometimes claimed that the Old
Testament God is more vengeful, whereas the New Testament
God is more forgiving (but see Cohen et al. 2006).
Certainly, such variations may be due to particular historical
factors affecting the development of a particular religion or the
learning history of a particular individual. However, taking a
cue from Bering, and Atran and Norenzayan (2004) and others,
we propose a novel direction for theorizing about belief in life
after death. It would be worth investigating whether variations
in beliefs in afterlife or observant spirits are linked to recurrent
variations in social or physical ecology. Bering has proposed
that belief in souls has a moral function, among others.
Perhaps beliefs in a soul or afterlife are more common when
resources are unpredictable, and life is brutal and short. If
most people have predictable and sufficient resources, there
may be less need to regulate cooperation. If resources are unpredictable
or scarce, however, supernatural agents may be more
necessary: As Durant and Durant (1968, p. 51) suggested, “as
long as there is poverty there will be gods.”
Similarly, a belief in an omniscient God (who also metes out
punishment, both during life and after) might be more
common in societies in which people spend more time around
non-relatives (who are more likely to punish your transgressions
severely, and to cheat on you). If true, one would expect not to
find such beliefs as commonly in small groups of closely related
hunter-gatherers. In social groups including unrelated
individuals, on the other hand, other people can’t be watching
you all the time to make sure you are not poaching others’
mates or stealing their food. But invisible, supernatural agents
can (or, at least, you don’t know when they are and when they
are not). According to this line of reasoning, one might suppose
that the variable and harsh desert culture in which the Old
Testament is rooted promoted a view of God as harsh and
vindictive, whereas the more stable societal structure of the
New Testament promoted a view of God as more forgiving.
Religions that exist in harsh or unpredictable environments (or
religions rooted in such environments) may be more prone to
belief in souls, or may view God as more punitive. Religions
that exist in stable or resource-rich environment (or religions
rooted in such environments) may be less prone to belief in
souls, or may view God as more forgiving.
This analysis suggests a need for a functionally based taxonomy
of religious beliefs and practices, which can be mapped onto a taxonomy
of ecological variations to which human groups need to
adjust. An ecological approach suggests that the traditional
beliefs of international religions originally emerged in interaction
with particular environmental factors. There are likely pressures
to maintain the belief systems intact as members migrate to new
physical and social environments. Our analysis implies that the
group-level beliefs will change (perhaps slowly) to match new habitats,
and that individual commitment to particular features of those
beliefs will change (perhaps more rapidly) to reflect operation of
context-triggered behavioral and cognitive mechanisms. It may
be, for example, that even Roman Catholics (who belong to a religion
with strongly institutionalized checks on heretical thinking)
have very different complexes of supernatural beliefs and imagined
offenses depending on whether they are from an Irish fishing
village, a Sicilian farming community, or a California suburb.
Production of supernatural beliefs during
Cotard’s syndrome, a rare psychotic
depression
David Cohen and Ange`le Consoli
Department of Child and Adolescent Psychiatry, CNRS FRE 2987 Cognition et
Comportement, Universite´ Pierre et Marie Curie, APHP, Hoˆpital Pitie´Salpe´trie`re,
75013 Paris, France.
david.cohen@psl.aphp.fr angele.consoli@psl.aphp.fr
Abstract: Cotard’s syndrome is a psychotic condition that includes
delusion of a supernatural nature. Based on insights from recovered
Commentary/Bering: The folk psychology of souls
468 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
patients who were convinced of being immortal, we can (1) distinguish
biographical experiences from cultural and evolutionary backgrounds;
(2) show that cultural significance dominates biographical experiences;
and (3) support Bering’s view of a cognitive system dedicated to
forming illusory representations of immortality.
Cotard’s syndrome (CS) is a rare condition in which the central
symptom is a delusion of negation. Patients suffering from the
syndrome may deny that they exist or that a part of their body
exists. They may also complain of damnation, possession, or
other delusional ideas, such as feeling enormous and immortal
or believing that nothing exists or that another person’s identity
(doctor, mother) is false. CS generally occurs in patients suffering
from major depression with psychotic features, but it can also
occur in patients suffering from schizophrenia or organic
mental conditions (e.g., general paralysis, epilepsy) (Berrios &
Luque 1995). In young people, it is often associated with bipolarity
(Consoli et al., in press; Soultanian et al. 2005). While the
descriptions of many psychiatric conditions have changed
during the last century (e.g., catatonia, hysteria), CS has been
shown to have very stable clinical characteristics since it was
first described in 1880 (Berrios & Luque 1995). Healthy
people’s beliefs in an afterlife or in other closely related supernatural
ideas are not expressed in a delusional way. In the case
of CS, subjects are temporarily and without self-questioning convinced
that both their soul and body are immortal, or, alternatively,
that they are already dead or damned. The very
existence of CS supports Bering’s view of a cognitive system
dedicated to forming illusory representations of immortality
and symbolic meaning of natural events. We can hypothesize
that, for unknown reasons, the system is productive during CS
without any activation of inhibitory or elaborative repression.
Because most patients recover, it gives us an opportunity to get
some insights from subjects themselves regarding their feelings
of being immortal, guilty, or damned. This allows us to distinguish
biographical experiences from cultural and evolutionary
backgrounds.
Table 1 summarizes the clinical characteristics of 8 patients
who we have treated during the last 20 years. The first striking
observation is that, for most of the subjects (6/8) with CS, their
delusional ideas could be related to their own life stories
(Table 1, column 5), despite their having a similar delusional
framework that included delusions of immortality in 5 cases
(Table 1, column 4). The last two subjects provided little information
about their CS state because of negativism and mental
retardation, respectively. Case 3 is particularly interesting as
she showed remarkable insight after treatment with electroconvulsive
therapy (ECT) (Cohen et al. 1997). The delusion
consisted of the patient’s absolute conviction she was already
dead and waiting to be buried, that she was immortal, that
she had no teeth or hair, and that her uterus was malformed.
When she recovered, we asked her to express the free associations
that came to mind when her delusional ideas were
evoked. Concerning having no teeth, she was surprised to
find herself thinking of her brother-in-law, a dentist. She
added that she would be ashamed to receive dental treatment
from him, and that she had cried every night since her
sister’s wedding and departure. Concerning the idea of a
genital malformation, she remembered guilty feelings associated
with masturbation, which she had practiced from childhood
until the onset of puberty.
The second striking observation relates on the fact that
despite a history of syphilis confirmed by immunology testing,
a 55-year-old man with CS (Case 5) had hypochondriacal concerns
about AIDS, showing that collective and cultural significance
dominates biographical experiences during CS. Because
of pressure from the human social environment, AIDS has
substituted for syphilis as God’s punishment for sins of the
flesh. In summary, over time, first syphilis then AIDS
symbolized the amalgam of flesh, punishment, sin, guilt, sexuality,
and the devil. Indeed, the last case of CS with hypochondriacal
fears of syphilis was published in the 1970s (Bourgeois
1969). Since the 1980s, and in the current series (N ¼ 3), AIDS
has the same symbolic significance that syphilis had until the
1970s.
Based mainly on cognitive and developmental psychology,
Bering has postulated that an organized system dedicated to
forming illusory representations of immortality evolved in
response to selective pressures by the human social environment.
If we consider CS as a psychopathological model to explore the
pathological production of supernatural beliefs, Bering’s hypothesis
implies that (1) the beliefs should associate personal
elements with stable superstructured collective schemas; and
(2), while stable in their significance, collective schemas should
integrate influences from the social environment. As highlighted
Table 1 (Cohen & Consoli). Clinical characteristics of eight subjects with Cotard’s syndrome focusing on delusional ideas
related to biography
Age Sex History Cotard’s symptoms Examples of delusions and links [$] with biography
62 M MDE Dementia Negation, immortality, enormity [Enormous medical knowledge] $ [Physician]
70 F Bipolar Negation, immortality,
enormity, damnation, guilt
[Nazi criminal] $ [Family died in deportation]
[Urines invaded the whole world killing all
children] $ [First child was stillborn]
15 F Bipolar Negation, immortality, damnation [No uterus] $ [Masturbation until age 9]
[No teeth] $ [Brother-in-law a dentist]
58 M Bipolar Negation, immortality,
damnation, hypochondria
[Father not dead] $ [Father died as a hero]
[AIDS] $ [?]
55 M Bipolar Negation, damnation, hypochondria [AIDS] $ [Guilt because of hypersexuality
during mania]
19 F Bipolar IQ ¼ 65 Negation [No blood] $ [?]
17 F MDE Negation, enormity, damnation,
hypochondria
[Diabetes] $ [Mother died from diabetes]
[AIDS] $ [?]
19 F MDE Negation, immortality [No body] $ [?]
MDE ¼ Major Depressive Episode; IQ ¼ Intellectual Quotient.
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 469
by the data from recovered patients with CS, both hypotheses are
supported.
Taking a psychodynamic view that is not part of Bering’s
report, we can also refer to the theory of archetypes (Jung
1934), where an archetype is the collective schema or
architecture of ancestral human mind (see Bering’s conclusion).
Interestingly, Jung’s hypothesis was based on comparative
studies of religious and supernatural beliefs. In particular,
following Levy-Bruhl’s studies of primitive societies (1910) and
mythology (1935), Darwin’s theory of selection, and his own
experience with the psychodynamic approach to psychotic
patients, Jung also postulated that humans have a natural
disposition to believe in an afterlife and religious concepts,
influenced by evolution.
Evidence for early dualism and a more direct
path to afterlife beliefs
David Estes
Department of Psychology, University of Wyoming, Laramie, WY 82071.
estes@uwyo.edu
Abstract: Ample evidence for dualism in early childhood already exists.
Young children have explicit knowledge of the distinction between
mental and physical phenomena, which provides the foundation for a
rapidly developing theory of mind. Belief in psychological immortality
might then follow naturally from this mentalistic conception of human
existence and thus require no organized cognitive system dedicated to
producing it.
Bering proposes an organized cognitive system dedicated to
forming what he terms illusory representations of an afterlife
and psychological immortality. Belief in the continuation of
psychological states after death is, as Bering notes, a radical
form of mind–body dualism, and he seeks evidence for this
dualism early in individual development. In this commentary I
briefly summarize existing empirical evidence, not mentioned
by Bering, demonstrating that very young children are already
and explicitly dualists. I conclude by questioning the claim that
belief in psychological immortality requires a cognitive system
dedicated to producing it. I suggest instead a more parsimonious
alternative in which this belief is just a natural extension of how
people, including young children, already think about human
existence.
Bering is astute to focus on the common-sense dualism
between mind and body because this conceptual distinction
and its close relatives play a key role in cognitive development
by providing the foundation for a mentalistic understanding of
human behavior. This “theory of mind,” as I will argue, might
then be extended to become one source of intuitive notions
about the soul and an afterlife. However, it is simply not the
case that research into “whether humans are common-sense
dualists” is just beginning, as Bering implies. In fact, abundant
research, some of it now two decades old, clearly demonstrates
that very early in the preschool years children already understand
and use a whole family of conceptual distinctions closely related
to mind–body dualism. These include the basic ontological
distinction between the mind and the external world, as well as
kindred distinctions between mental and physical phenomena,
between fantasy and reality, and between specific thoughts and
the things they represent (Estes 1994; Estes et al. 1989;
Wellman & Estes 1986).
Briefly, this research shows that by 3 years of age children
already recognize the defining criteria that distinguish the
internal-mental from the external-physical world. They know
that mental entities (thoughts, memories, dreams, mental
images) are not real in the way that physical entities are, and
that they have no permanent existence apart from the mind in
which they occur, are inherently private rather than public,
and cannot be seen, touched, used, or shared with others in
the way that corresponding physical objects can. It is important
to emphasize here that preschool children’s knowledge of this
fundamental dualism is not just implicit in their behavior and
not merely inferred from their responses in different experimental
conditions. Instead, they clearly have explicit knowledge of
how mental and physical phenomena differ, as demonstrated
by their capacity to articulate this understanding with convincing
verbal justifications for their responses. These explanations are
typically telegraphic but interpretable at 3 or 4 years of age and
become remarkably adult-like by the age of 5 or 6. These experimental
findings are supported and extended by naturalistic
research on language development showing that even before 3
years of age children spontaneously refer to the distinction
between mind and external reality in their conversations in
natural settings (Bartsch & Wellman 1995).
How do young children comprehend with such apparent ease
this fundamental distinction between the internal-mental-subjective
realm and the external-physical-objective realm, of which
mind–body dualism is one aspect? It may well be the case that
this distinction is of such crucial importance in human social
life that, like language or face recognition, we are prepared by
evolution to get it quickly and easily. But at another level of analysis,
that of everyday human experience, it is also the case that this
is just how the world is. Children, perceptive creatures that they
are, rapidly discern this fundamental distinction, which is
constantly manifesting itself in their experience, just as it is in
ours. We have minds and we have bodies; there is an internalmental
realm and an external-physical realm; and there is
ample evidence in the child’s ongoing experience from which
to abstract these natural categories.1
The boundary between
them may of course break down under rigorous philosophical
or scientific analysis, but regardless of our expertise or theoretical
allegiances, we all take this foundational distinction for granted
and constantly use it in everyday life. The research cited earlier
shows that preschool children do so, as well.
How do we get from this basic dualism to belief in souls, an
afterlife, and psychological immortality? Bering’s rather clever
solution involves a collection of cognitive biases and errors that
together produce these “functional illusions” and thereby
enhance genetic fitness by making it less likely that individuals
will engage in acts harmful to their reputations. Leaving aside
the perennial question of whether there might really be an afterlife
of some unknown variety,2
and granting that the specific
mechanisms in Bering’s account (e.g., simulation constraints,
offline social reasoning) may indeed be part of the story,
perhaps the path from the young child’s dualism to belief in an
afterlife for immortal souls might be fairly direct and require
no special evolutionary solution. As we have seen, very young
children already have the distinction between mental and physical
phenomena solidly in hand and thus recognize the existence
and nature of immaterial entities. This provides the essential
basis for a mentalistic folk theory of human behavior and the
framework on which our more general beliefs about human existence
are based. Beliefs about an afterlife can thus be seen as just
beliefs about this life, idealized and extended in our imagination
beyond the mystery of death. And who wouldn’t want more of
such a good thing?
NOTES
1. In addition to abundant naturalistic evidence documenting that
they spontaneously talk about their mental states (e.g., Bartsch &
Wellman 1995), there is also experimental evidence indicating that
preschool children have conscious access to some types of mental activity
(Estes 1998), which of course would be a prerequisite for abstracting the
mental–physical distinction from their own experience.
2. And who knows for sure? Not Bering. And not Francis Crick (1994),
Theodosius Dobzhansky (1967), Martin Gardner (1999), or me.
Commentary/Bering: The folk psychology of souls
470 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
A case of stunted development? Existential
reasoning is contingent on a developing
theory of mind
E. Margaret Evans and Henry M. Wellman
Center for Human Growth and Development, University of Michigan,
Ann Arbor, MI 48109-0406.
evansem@umich.edu
http://www.umich.edu/evansem/ hmw@umich.edu
http://www.umich.edu/chgdwww/faculty/wellman.html
Abstract: Missing from Bering’s account of the evolutionary origins of
existential reasoning is an explicit developmental framework, one that
takes into account community input. If Bering’s selectionist explanation
was on target then one might predict a unique and relatively robust
developmental trajectory, regardless of input. Evidence suggests
instead that children’s existential reasoning is contingent on their
developing theory of mind.
Bering’s focus is naive or intuitive religion in the sense of its
import and place in human thinking about one’s own soul,
values, and place – its existential focus. He highlights important
issues, and presents many intriguing ideas concerning the
evolutionary origins of these existential themes. But missing is
an explicitly developmental framework; in the absence of such
a framework, it is difficult to agree with his claims.
Modern evolutionary theory is itself undergoing a radical
reconceptualization with development playing a central role,
so-called Evo-Devo (e.g., Carroll 2006). The discovery of critical
regulatory genes that alter patterns of gene expression over
development was only made possible because of this focus.
Similarly, any attempt to offer a modern evolutionary account
of a psychological process should incorporate development.
Intuitive existential psychology is closely aligned with intuitive
psychology – our everyday understanding of self and others as
intentional, believing-desiring, communicating agents – and,
according to Bering, with intuitive conceptions of intelligent
design. Past research is clear: intuitive psychology (theory of
mind) develops – initial infant biases lead to early conceptions
that are considerably revised and expanded in the course of
childhood development – and intuitive understanding of the
origins and functions of human life also develops. If Bering is
correct and existential reasoning is a consequence of selection
pressures, and not a spandrel, then one might predict a unique
trajectory, one that is robust and relatively independent of
these other developments. Bering indeed talks of possible developments.
One clear point is that, unfortunately, little is known,
yet both the developmental and evolutionary stories to be told
must be tightly constrained by such details. Developmental
details are also needed to frame and evaluate the connections
between existential psychology, theory of mind, and intuitive
conceptions of origins.
Regarding existence and mind, Bering’s selectionist arguments
encompass claims of early appearing existential sensitivities
backed by some admittedly preliminary data. The data definitely
demonstrate development, but the details are not only unknown
and insufficiently established, they are currently contradictory.
For example, Bering suggests (following Barrett) that young
children’s conceptions might be specially commensurate with
supernatural conceptions (e.g., gods are not subject to false
beliefs). But Barrett et al.’s (2001) work is just as easily interpreted
to show that conceptions of God as omniscient are only
made possible when children are able to reason about false
beliefs. Prior to that point, children cannot make such judgments.
Moreover, in Bering’s Princess Alice studies, for example, the
youngest children (4-year-olds) interpret unexpected events as
physically caused and only older children see them as indicative
of supernatural acts. Such findings suggest that existential
reasoning is contingent on a developing theory of mind.
An important claim is that a naı¨ve dualism leads to beliefs in an
afterlife in which mind continues independent of body
(psychological immortality) after death. The developmental
unfolding of understandings of mind and of death thus
becomes intriguing indeed. Bering (also Bering & Bjorklund
2004) presents a scenario in which younger children (5-yearolds)
attribute ongoing mental functions after death and such
attributions decrease with development. Although 5-year-olds
attribute mental functions to dead individuals in Bering’s
research, they do not do so (and neither do 4-year-olds) in
other research (e.g., Barrett & Behne 2005; Poling & Evans
2004). Furthermore, in Flavell’s research 5-year-olds often fail
to attribute ongoing mental functions (thinking) to waking
persons; relative to older children, they systematically downplay
the amount of consciousness that everyday folk have in everyday
life (Flavell et al. 2000). These findings provide an unlikely
platform for the bold proposal that rampant attribution of
mental life to the dead provides a natural starting point for intuitive
existential questions. We agree that how children understand
these issues is important and can inform our theories of intuitive
understandings of mind, existence, and the divine. But, those
developments, while to-be-discovered, do not as yet conform to
Bering’s initial outlines. Understandings of death also figure
into children’s understandings of origins and design.
Regarding existence and origins, Bering argues that conceptions
of intelligent design are effortlessly aligned with
beliefs in immortal souls. Yet, Evans’ (2000; 2001) studies of concepts
of species origins tell a more extended developmental story,
and one that varies depending on the context. Not surprisingly,
children from Christian fundamentalist communities, whatever
their age, prefer creationist (God made “X”) ideas. Younger children
from non-fundamentalist communities, on the other hand,
endorse a mixture of spontaneous generationist (the very first
“X” came out of the ground) and creationist ideas. Not until 8
to 9 years of age were children consistently creationist, regardless
of community of origin. More recent work along these lines
suggests that the younger children were not in a position to
grasp origins concepts, because they had not yet fully confronted
existential questions (Evans 2005; Evans et al. 2001). To be able
to respond to questions about the origins of animal kinds, children
have to understand that animals are not eternal, in that
they were not always here on earth, nor will they continue to
be on earth. In the latter studies, the creationism of 4- to 10year-olds
was related to their ability to grapple with existential
concepts (death, eternity), and to their understanding that
humans (not God) create artifacts, independently of the effects
of age. Once such existential questions have been grasped, only
then can the “origins” question arise: How did the animals get
on earth in the first place? Evans’ claim (2001; 2005) is that children
transfer their understanding of the human as an intentional
manufacturer of new tools, and apply that to objects that have
arisen naturally, such as “new” species. For younger children,
the idea that “God did it,” appears to be loosely associated knowledge,
not yet integrated into a conceptual structure (Evans 2001),
suggesting that “testimony” (Harris & Koenig 2006) plays a
crucial role in early God concepts. In sum, God as intelligent
designer is a complex (albeit possibly naturally developing), not
an effortless, idea, which becomes firmly rooted only at the
point when children reliably confront existential questions and
fully understand the role of human artifice (Evans 2005; see
also Defeyter & German 2003).
Thus predictions that might follow from Bering’s thesis, such
that existential reasoning is effortless, early acquired, and relatively
independent of other developmental processes, are not
borne out. On the contrary, we suggest that children’s developing
understanding of the mind, in particular, their naı¨ve theories of
intention, undergird and make possible religious/existential
reasoning. Furthermore, this development seems to require an
interaction between these processes and community input.
Ideally, we need an evolutionary-developmental theory of
existential reasoning that takes into account cultural context.
We are grateful to Bering for initiating this process.
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 471
ACKNOWLEDGMENT
The preparation of this commentary was supported in part by NSF Grant
540152 to Evans and NICHD Grant 22149 to Wellman.
Culture and development matter to
understanding souls, no matter what our
evolutionary design
Michel Ferrari
Human Development and Applied Psychology, OISE/University of Toronto,
Toronto, ON M5S 1V6, Canada.
meferrari@oise.utoronto.ca
Abstract: For Bering, appreciating that people are objects is a
developmental accomplishment. Baldwin and Piaget agree. However,
for Piaget, an immanent conception of the divine is more
developed than a separate transcendent God. Culture also matters. In
Plato’s Phaedo, Socrates’ belief in immortality was a reasoned
conclusion – not “built in” – for reasons similar to those still held by
modern scientists.
Almost a century ago, J. R. Angell (1911) wrote, “The term soul
has generally been applied to the supposed spiritual essence of
human personality which persists after death. As such, it is
connected with problems not soluble by empirical methods.
Psychology as an empirical natural science has consequently
ceased to use it as a familiar part of its terminology” (p. 46).
He goes on to say, “the term consciousness itself is likewise in
danger of extinction or at least essential modification” (p. 47).
Prophetic words. But with the return of an “essentially modified”
science of consciousness, the soul is again a candidate for rehabilitation
– as long as it remains subject to Neo-Darwinian natural
selection within a distinctively human social environment, and as
long as it is “illusory” (or at least that its immortality and purpose
are illusory).
For Bering, asking “Why am I here?” suggests a social relationship
between the self and a presumed supernatural
creator – a “cognitive illusion” that can help produce “genetic
fitness-enhancing” behavior by promoting normative prosocial
behavior that that creator has mandated. Bering also suggests
that because human social interaction relies on believing that
absent agents continue to exist, we have a hard time imagining
anyone to be dead; that our minds/brains are not well equipped
to update complex social rosters. But why go so far? Without
invoking anything supernatural, Parker’s (1998) proposal that
self-conscious emotions, like shame, may have evolved to allow
parents to govern their children when not physically present to
enforce social norms – an influence that might persist beyond
death. If so, then the idea of a universal care-giver, God, is a
natural (but culturally bound) extension of this direct social
experience.
Piaget devoted his first lab at the Jean Jacques Rousseau Institute
to the study of religious experience, and lectured on his
results and their implications at Sainte Croix (1923; 1928;
1930). Vidal (1994a; 1994b) claims that Piaget’s early empirical
work on religious experience aimed to provide empirical evidence
for his own metaphysical framework, centered around
the idea of the “immanence” of the divine in human experience.
Indeed, these early studies by Piaget showed that unconscious
and affective attachment to different kinds of religious experiences
of God (transcendent or immanent) depends on the type
of parenting one receives and the general socio-political cultural
environment of one’s upbringing (see also Bemmer 2002).
Piaget’s (1932/1978) studies of morality grew directly out of his
work on religious belief.
God thus becomes a “super-parent” – an idea also advocated
by James Mark Baldwin at the turn of the last century.
Bering’s very interesting point that it is structurally simpler and
so developmentally easier to imagine an omniscient other, God,
than to imagine someone who holds false beliefs is directly in
line with these older theories of development. Likewise,
Bering’s claim that appreciating people to be “just objects” is a
developmental accomplishment is exactly Baldwin’s thesis – an
idea he leverages for a very creative resolution of the mind–
body problem (Baldwin 1903; see also Ferrari 2003). Similarly,
Piaget’s (1928; 1930) mature thoughts on religious experience
led him to believe that the tension between transcendent and
immanent conceptions of God could be resolved developmentally
– that an immanent conception of the divine (i.e., God as
intrinsic to our lived experience) was a more developed stage
of religious experience than experience of a separate, transcendent
God. Writing in a very different Zen tradition, Suzuki
(1962/1972) captures this view well when he writes that, the
“ultimate Self is above all forms of dichotomy, it is neither
inner nor outer, neither metaphysical nor psychological,
neither objective nor subjective. If the term ‘Self’ is misleading,
we may designate it as ‘God’ or ‘Being’ or ‘the Soul,’ ‘Nothing’
or ‘anything’” (p. 3).
Are these claims unscientific? I agree with James (1902/1961),
that a “rigorously impersonal view of science might one day
appear as having been a temporarily useful eccentricity rather
than the definitively triumphant position which the sectarian
scientist at present so confidently announces it to be (p. 395,
footnote 8).” Certainly, empirical studies support the claim that
immanent experience of the divine is indeed much rarer and
develops later than transcendent experiences, documented in
children as young as age six (Argyle 2000). Thus, Bering’s
suggestion that children understand God to be a separate and
higher being is only half of a more sophisticated developmental
argument proposed by developmental psychologists of the last
century.
In another line of reasoning, Bering also proposes that
because we find it impossible to imagine what it is like for ourselves
to be dead (what he calls a “simulation constraint”) people
– especially children – naturally tend to think that
psychological agents survive death. The “simulation constraint”
on imagining death is very plausible. However, although it
may be impossible to imagine our own nonexistence psychologically,
we need not reason about the afterlife by analogy to our
own spiritual life. As Bering himself says, we know and understand
forms of human existence in which we are unaware – a
dreamless sleep, for example – and can imagine not returning
from that state. Or, to take a classic example, in Plato’s
Phaedo (c. 350 bce/1977, subtitled, On the soul), Socrates
believes he will survive death but wants to debate this so as
not to die holding a false belief. One objection, made by
Cebes, is that most “men find it very hard to believe what you
said about the soul [i.e., that it survives death]. They think
that after it has left the body it no longer exists anywhere, but
that it is destroyed and dissolved on the day the man dies, as
soon as it leaves the body; [. . .] dispersed like breath or
smoke, has flown away and gone and is no longer anything
anywhere” (Phaedo, 70a). A little later (85e–86d), Simmias
proposes this analogy: the soul is a kind of harmony produced
by the body, like the music of a lyre; smash the instrument
and the harmony is lost. This analogy is essentially the
Darwinian analogy for mind, something generated by the body
through the course of human evolution to help it survive.
Socrates has an answer to these objections, although one that
may not convince a modern audience – perhaps not even
Aristotle, writing a few decades later (see Wilkes 1992) – but
this shows that the idea of immortality was a reasoned conclusion.
It was not “built in,” at least not for most adults of
that time, for reasons that resemble those still held by modern
scientists; that is, that the soul is nothing other than an
expression of the operation of the body, which itself is just a
biological material thing, having nothing immaterial about it
that can survive death.
Commentary/Bering: The folk psychology of souls
472 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Autism, language, and the folk psychology of
souls
Stephen Flusberga
and Helen Tager-Flusbergb
a
Department of Psychology, Stanford University, Stanford, CA 94305;
b
Department of Anatomy and Neurobiology, Boston University School of
Medicine, Boston, MA 02118.
sflus@stanford.edu htagerf@bu.edu
Abstract: Anecdotal evidence suggests that people with autism, with
known impairments in mechanisms supporting a folk psychology of
mind or souls, can hold a belief in an afterlife. We focus on the role
language plays, not just in acquiring the specific content of beliefs, but
more significantly, in the acquisition of the concept of life after death
for all people.
The main goal of Bering’s article is to sketch a Darwinian model
that accounts for the near-universal belief in an immortal soul
and an afterlife. He argues that human social cognition has
evolved to process information in specific ways that both allow
for and engender dualistic thinking about mind and body, as
well as related areas of religious or existential thought. It is this
underlying cognitive architecture that constitutes the “folk
psychology of souls.” Bering stresses the role of theory of mind
and related cognitive systems in promoting default representations
of mental states surviving death.
This is an interesting and important hypothesis that has many
ramifications for the study of human cognition and culture.
Our commentary focuses on the consequences of this view for
predicting how people with specific social-cognitive deficits
might conceive of and react to death. We then explore the
implications of social-cognitive deficits for Bering’s model, to
address the question of whether underlying cognitive architecture
is both necessary and sufficient for representing life after
death.
Can people with autism believe in life after death? Bering’s
model offers guidelines for who is most likely to entertain
beliefs in a soul and afterlife, namely, individuals with an intact
theory of mind. Indeed, Bering cites evidence that most people
claim that what endures after death is the person’s mental
states. What about populations with deficits in this domain of
human cognition? It is widely accepted that autism (ASD) is, in
part, characterized by atypical social-cognitive development
and domain-specific impairments in theory of mind (e.g.,
Baron-Cohen et al. 2000). People with ASD have difficulty
representing the mental states of themselves and others even
when high-functioning individuals with ASD have aboveaverage
IQ scores and relatively good language skills (BaronCohen
2000).
Bering’s model suggests that people with autism would be
much less likely to engage in “existential” thought or to consider
mental states surviving death, given that they generally fail to
consider a person’s mental states even when they are alive.
Although we know of no systematic research that has tested
this hypothesis, anecdotal evidence suggests a more complex
picture. On the one hand, although people with ASD do form
emotional attachments (Rutgers et al. 2004), in our experience,
it seems that they do not respond with the same degree of distress
to the death of a loved one as do non-autistic individuals. This
provides support for Bering’s view, as he argues that affective
responses may trigger the formation of afterlife representations
based on existing social-cognitive mechanisms. Because people
with ASD have deficits in these underlying mechanisms, they
may not react to death with the same kind of existential crisis,
and may therefore be less likely to represent life after death.
On the other hand, this picture is complicated by the fact that,
again based on anecdotal evidence, some people with ASD can
hold a belief in a soul and afterlife. When asked about what
happens to a person after they die, some people with autism
claim that they continue to exist in some form; for example,
that dead people ascend to heaven.
We hypothesize that a person with autism may acquire the
belief in an afterlife via language, in the same way as they can
learn to pass false belief tasks (Tager-Flusberg & Joseph 2005).
Numerous studies have demonstrated that for children with
autism, the single best predictor of passing false belief and
other theory of mind tasks is linguistic knowledge, especially
vocabulary and grammatical knowledge. However, even people
who pass theory of mind tasks seem not to engage the same
neurocognitive mechanisms when reasoning about beliefs (e.g.,
Castelli et al. 2002), suggesting that language may provide an
alternative way of bootstrapping mental state attribution in
people who have impairments to the mechanisms that are generally
engaged for processing theory of mind tasks.
Does language contribute to the folk psychology of
souls? While Bering acknowledges the role of socio-cultural
indoctrination in the formation of specific religious concepts,
his theory emphasizes the causal role of underlying cognitive
mechanisms in giving rise to generally dualistic concepts and
modes of thought. However, given that people with autism can
hold dualistic religious beliefs, might language play a more
significant role in the development of the folk psychology of
souls? That is, does the structure of our linguistic concepts
help shape the way we think about mind, body, and soul?
Again, we know of no empirical research addressing this
specific claim, but the behavior of people with autism suggests
that language may play a causal role in the development of the
folk psychology of souls. Consistent with this hypothesis, many
philosophers have proposed that it is conceptual and linguistic
confusion that encourages mind/body separation, rather than
any innate predisposition. Specifically, they highlight the
various metaphorical ways we talk about the mind and mental
activity and argue that it is these disparate conceptual
representations that propel dualistic thought (e.g., Lakoff &
Johnson 1999; Melser 2004; Papineau 2002; Ryle 1949;
Wittgenstein 1953). Language and cognition are intimately tied
together, and the experimental evidence cited by Bering
cannot distinguish between the cognitive and linguistic factors
that could be driving universal dualistic beliefs.
Human social-cognition may have evolved in such a way so as
to support belief in a soul and afterlife, but this underlying architecture
may be neither sufficient nor necessary for such beliefs.
In our view, the prevalence of these beliefs likely indicates a
complex and dynamic process consisting of multiple interdependent
cognitive, affective, linguistic, and cultural components. As
Bering’s own research demonstrates, most people probably do
not have a stable, rational set of beliefs in the afterlife. It may
therefore be premature to privilege specific social-cognitive
factors underlying the “folk psychology of souls.” There is an
important need for future research to disentangle the different
elements that motivate these beliefs, and to address the issues
raised in both Bering’s article and in these commentaries.
ACKNOWLEDGMENT
Preparation of this commentary was supported by grants from NIH (U19
DC 03610; U54 MH 66398) to Helen Tager-Flusberg.
The supernatural guilt trip does not take us far
enough
Nathalia L. Gjersoe and Bruce M. Hood
Bristol Cognitive Development Centre, Department of Experimental
Psychology, University of Bristol, Bristol, BS8 1TU, United Kingdom.
n.l.gjersoe@bristol.ac.uk http://bcdc.psy.bris.ac.uk/
bruce.hood@bristol.ac.uk http://bcdc.psy.bris.ac.uk/
Abstract: Belief in souls is only one component of supernatural thinking
in which individuals infer the presence of invisible mechanisms that
explain events as paranormal rather than natural. We believe it is
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 473
important to place greater emphasis on the prevalence of supernatural
beliefs across other domains, if only to counter simplistic divisions
between rationality and irrationality recently aligned with the
contentious science/religion debate.
We are in agreement with Bering’s general thesis that the folk
psychology of the soul can be traced to the development of intuitive
theories regarding the nature of the reality and intentionality,
as well as the difficulty of conceiving of the state of nonexistence.
However, we contend Bering’s claim that there exists an
“organized cognitive ‘system’ dedicated to forming illusory representations”
of an afterlife that has “evolved in response to the
unique selective pressures of the human social environment”
(target article, sect. 1, para. 5). Bering has proposed that a
belief in the afterlife has the effect of promoting prosocial behavior
because of the perceived connection between the moral
implications of our actions whilst alive and the possible recriminations
from the deceased and/or possible jeopardizing of our
immortal souls on death. The first problem we have with this
central thesis is that there are other social mechanisms that do
not have anything to do with the folk psychology of souls that
also act to constrain and control social behaviour. A brief consideration
of the vast research field on compliance and cognitive
dissonance proves that people conform to social conventions
through the effect of peer pressure and social evaluation. A
belief in retribution from beyond the grave may contribute to
this list of cognitive mechanisms for socialization but it does
seem a little ad hoc to make it a primary mechanism operating
under Darwinian selection. After all, many social animals also
show behavioural inhibition and prosocial behaviour without
necessitating a specialised cognitive mechanism for a belief in
souls.
Our second problem with this central thesis, and the alternative
theoretical standpoints addressed in the article, is that they
fail to appreciate the extent of supernatural thinking as a
general feature of human cognition. Bering offers a convincing
range of evidence for the universality of beliefs in an afterlife
to cast doubt over the “spandrel hypothesis” of supernatural
thought. We would add that a growing body of literature suggests
that belief in an afterlife has many positive cognitive effects, such
as perceptions of control and security, which may have adaptive
advantages. We also agree that previous models of supernatural
belief based only on agency-detectors may be sufficient for
deities and ghosts but fail to capture many aspects of human
experience that are perceived to be under supernatural control.
For example, compelling evidence for supernatural beliefs in
the domain of folk biology comes from Paul Rozin and colleagues
(e.g., Nemeroff & Rozin 1994) who have repeatedly shown that
moral contagion from items associated with “evil” people is extraordinarily
difficult to ignore and is supported by a belief in a physical
manifestation of a moral stance. Or consider the peculiar and
yet prevalent belief (found in around 90% of adults) that we can
detect the unseen gaze of others (Titchener 1898). In both these
examples, we expect that a sizeable number of individuals who
explicitly reject notions of the afterlife and souls would still
nevertheless follow the general position that garments can be
contaminated and that they can feel the unseen gaze of others.
There are similar examples of naı¨ve beliefs in supernatural
forces in the domain of folk physics. For instance, naı¨ve reasoning
about dynamics is predominantly in terms of the belief that
objects are kept moving by internal forces and not external
ones (e.g., McClosky et al. 1980). These supernatural internal
forces are in direct contradiction to Newtonian laws of physics,
but are characteristic of medieval impetus theories and are
widely spread throughout both naı¨ve populations and those
with formal physics training. Like supernatural beliefs in an afterlife,
these naı¨ve impetus theories can be very hard to overcome
and are often held simultaneously with formal theories of
Newtonian dynamics and used interchangeably (e.g., Viennot
1979). The “hyperactive agency detector” could not extend to
explain these diverse supernatural beliefs across domains of
thought. On the other hand, it has not been suggested that
dedicated and uniquely human cognitive systems have evolved
individually in each of these domains that account for these
pan-cultural, early developing, and intransient naı¨ve errors. So
while we agree that supernatural thinking about the soul could
serve to cement social cohesion, supernatural thinking in many
domains could operate as socializing mechanisms that enable
us to think of ourselves as connected to others by tangible
forces, even though much of that reasoning may be implicitly
held. We would argue that supernatural thinking, in the form
of positing invisible forces that defy scientific validation, is an
innate human tendency that goes far beyond the realm of religious
thought into all domains of knowledge. We see little
evidence in this article that proves that naı¨ve beliefs in an afterlife
are qualitatively different from naı¨ve theories in folk biology and
folk physics.
We feel that it is important to extend this work into other
realms of reasoning because recent commentary, figure-headed
by such prestigious names as Richard Dawkins and Daniel
Dennett, polarizes the debate by aligning religious belief with
irrational memes propagated by the church and aligning
atheism with rationality. If religious inclination instead proves
to be associated with a universal human tendency towards supernatural
beliefs, from which even atheists are not exempt, this
arbitrary divide could prove to be both dangerous and scientifically
untenable. Rather, we would prefer that the proposal for
future research, and the debate in general, recognized that we
all entertain supernatural belief systems which must be taken
into account when studying human cognition and behavior.
Souls do not live by cognitive inclinations
alone, but by the desire to exist beyond death
as well
Jeff Greenberg,a
Daniel Sullivan,a
Spee Kosloff,a
and
Sheldon Solomonb
a
Department of Psychology, University of Arizona, Tucson, AZ 85721;
b
Department of Psychology, Skidmore College, Saratoga Springs, NY 12866.
Jeff@email.arizona.edu swolf22@email.arizona.edu
kosloff@email.arizona.edu ssolomon@skidmore.edu
Abstract: Bering’s analysis is inadequate because it fails to consider past
and present adult soul beliefs and the psychological functions they serve.
We suggest that a valid folk psychology of souls must consider features of
adult soul beliefs, the unique problem engendered by awareness of death,
and terror management findings, in addition to cognitive inclinations
toward dualistic and teleological thinking.
Bering’s analysis provides an inadequate “folk psychology of
souls” because folks have motivational and affective concerns
and are heavily influenced by culture, and these factors must
be considered, along with cognitive propensities, to account for
soul beliefs.
Bering’s reliance on cognitive biases particularly pronounced
in children is insufficient for two reasons. First, people relinquish
many childish beliefs as they mature, as Bering’s research shows.
Adults generally do not believe dead mice get hungry, or that
taller glasses necessarily contain more milk. Why do soul
beliefs persist, when so many childhood ideas do not? How can
someone smart enough to elude security and commandeer and
steer an airliner precisely into a building believe he will enter a
paradise filled with 72 virgins on impact?
Second, adult spiritual beliefs seem quite different than mere
cognitive errors of imputing mind; they vary widely across
cultures and are often quite complex (e.g., Watson 2005). In
some cultures, there was no immortal soul, in others only the
Commentary/Bering: The folk psychology of souls
474 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
wealthy, or only men, or only women who died in childbirth had
immortal souls. In some, moral action affected one’s afterlife, in
others, not. The first Chinese emperor sent vessels in search of
the Islands of Immortality. In the Sumerian Epic of Gilgamesh,
Gilgamesh, disturbed by the prospect of death, embarked on a
search for immortality. The Christian and Islamic conceptions
of soul and afterlife are extraordinarily elaborate, buoyed by
many symbols and rituals. The Aztec conception involved
sixteen stages of existence and elaborate rituals, including the
excision of sacrificial human hearts while still beating. If
immortality beliefs were a simple default by-product of cognition,
why would these beliefs be so varied and so complex?
These fervently held belief systems, with their extensive time,
effort, and life-consuming rituals, are neither child’s play nor
simple elaborations of cognitive errors.
By-products of cognitive inclinations cannot account for the
elaborate nature of soul beliefs and the deep commitments to
them. Nor is it likely they are primarily products of selection
for avoidance of socially prohibited behavior. The prophet
Zarathustra replaced the Persian class-based notions of afterlife
reserved for the wealthy with the first modern beliefs in the
soul’s fate determined by moral behavior on earth (Kriwaczek
2002). Thus, belief in afterlife rewards for altruism was
culturally constructed (partly to enhance social control) and
memetically transmitted, rather than selected for as a cognitive
predisposition.
A propensity for altruism could result simply from feelings of
empathy and attachment, and sensitivity to contingencies for tangible
rewards and punishments. Wouldn’t irrational worrying
about invisible forces have been counterproductive? Wouldn’t
self-serving immoral behavior be most adaptive when one’s
deities would be the only witnesses? How could cognitive predispositions
that caused individuals to sacrifice their own offspring,
perform time-consuming rituals, or feel crushing guilt at acts that
violated the Golden Rule, be selected for, unless these beliefs
served some more pressing evolutionary function than protection
of reputation?
Bering hints at such a function, mitigating existential despair,
but gives it insufficient weight. As many have observed (e.g.,
Rank’s Psychology and the Soul, 1931/1961), the dawning realization
of the inevitability of death had to be monumentally problematic
for proto-humans. Many of our physiological systems
function to keep us alive in a perilous world, and yet, thanks to
our intellect, we know they will inevitably fail. This had to
arouse intense concerns with personal vulnerability, and the
resulting potential for anxiety would have been immobilizing
without comforting mythic illusions of deistic protections and
an everlasting soul. To be willing and able to hunt large game,
compete for resources, and so forth, such beliefs provided
necessary equanimity and confidence. Although these spiritual
beliefs would not always over-ride fight or flight responses to
imminent danger, they would allow individuals and groups to
function more effectively, with their anxieties largely in check.
Why would people fear no longer existing when they cannot
easily simulate it? According to Zilboorg (1943) and others, we
are predisposed to fear death because it is highly adaptive to
do so – it helps keep us alive. Our brains are designed to react
to things that threaten our continued existence with fear,
arousal, and defensive responses; from such reactions, it is a
simple cortical inference that what we fear is death. Unfortunately,
there is no simple fight or flight response to the knowledge
of the inevitability of death, leading to the elaborate
symbolic defenses provided by culture.
Second, we fear many things before we have experienced
them, things we can’t simulate – cancer, AIDS, a root canal,
bungee-jumping. We fear whatever might cause us pain or end
our existence. Third, the non-dreaming phases of sleep and
being anaesthetized are somewhat similar to death – in these
states, we are not, as far as we can tell, conscious or thinking.
Finally, we fear death primarily not because of what we
imagine it to be but for what we can easily imagine it takes
away: life. We can imagine not seeing, not hearing, not tasting,
not smelling, not feeling, not being able to touch or communicate
with loved ones, not being able to listen to music, watch movies,
take walks, and so on.
Terror management theory (TMT), based on Becker (1971;
1973), posits that spiritual beliefs serve the function of helping
humans deny the finality of death (Solomon et al. 1991). The
theory posits that over childhood, the security base provided by
care-taking adults is replaced by deities and cultural authorities.
Just as the young child sustains the love and protection of its caretakers
by meeting standards of worth, the adult typically sustains
security by adhering to the standards of worth of the spiritual and
secular authorities of the culture. From this perspective, deities
have so commonly been patriarchal or matriarchal because
they have been modeled after the childhood care-takers.
Deities are also judgers and punishers because in a world full
of tragic, scary events, a deity who does not dole out punishment
is not plausible.
Over 250 studies supporting TMT have documented that
reminders of death (mortality salience: MS) increase advocacy
of beliefs and behaviors that serve to convince people that they
are worthy members of a meaningful universe, rather than
mere animals fated only to obliteration. For example, MS
increases identification with death-transcending groups and
ideologies (e.g., I am more than an animal, I am an American!)
and bolsters efforts to believe the world is just; and threats to
these protective beliefs increase the accessibility of deathrelated
thought (Greenberg et al., in press).
Becker argued that worldviews with spiritual components
work best for managing terror. This may be why correlational
evidence consistently finds that religiosity is associated with
mental health and lower death anxiety (see Pargament 1997).
Importantly, experimental research provides converging evidence
of a protective terror management role of spiritual belief
(Greenberg et al., in press). MS increases bias toward
members of one’s religion and against proponents of another,
reluctance to use religious icons inappropriately, and, among
the religious, MS increases belief in an afterlife, religiosity, and
belief in prayer. Finally, increased belief in an afterlife and
making religiosity salient to religious people reduce the use of
secular terror management defenses such as worldview bolstering,
and also reduce death-related thinking (Dechesne et al.
2003; Jonas & Fischer, in press). Thus, spiritual beliefs protect
people from concerns about mortality.
In sum, although one could limit analysis to cognitive inclinations,
doing so provides a very impoverished folk psychology
of souls. To truly understand the psychology of souls, we
should build on extant knowledge regarding evolution, the
nature of soul beliefs, and psychological defenses, and acknowledge
the role of the unique selection pressures engendered by
human awareness of death in the evolution of supernatural
beliefs.
Learning that there is life after death
Paul L. Harrisa
and Rita Astutib
a
Graduate School of Education, Harvard University, Cambridge, MA 02138;
b
Department of Anthropology, London School of Economics and Political
Science, London WC2A 2AE, United Kingdom.
paul_harris@gse.harvard.edu r.astuti@lse.ac.uk
Abstract: Bering’s argument that human beings are endowed with a
cognitive system dedicated to forming illusory representations of
psychological immortality relies on the claim that children’s beliefs in
the afterlife are not the result of religious teaching. We suggest four
reasons why this claim is unsatisfactory.
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 475
Bering proposes that human beings evolved a cognitive system
dedicated to the belief in an afterlife. In support of that claim,
he refers to experiments showing that young children often
make continuity claims – they assert that mental processes,
notably thoughts, feelings, and desires continue after death –
whereas older children are more likely to deny their continuity.
Bering concludes that this developmental pattern is consistent
with the early functioning of the proposed cognitive system but
not with the alternative hypothesis of religious teaching. On
that hypothesis, claims that mental processes continue after life
would be more frequent among older as compared to younger
children.
We think the developmental pattern is more complex than
Bering allows and that, on close examination, it underlines a
crucial role for religious teaching. First, two recent studies
indicate that continuity claims increase rather than decrease
with age, both in Spain (Harris & Gime´nez 2005) and in
Madagascar (Astuti & Harris, submitted). The most plausible
explanation of this age change is that as they get older, children
are increasingly likely to encounter and assimilate afterlife
beliefs in their community. A likely explanation for the retrenchment
of such beliefs reported by Bering is that children
come to differentiate between the fate of human beings and
other animals, including mice (the focus of Bering’s research):
they learn that human beings enjoy an afterlife whereas mice
do not.
Still, Bering could reasonably insist that children start out
with a global and innate set of afterlife beliefs, even if religious
teaching reinforces or denies their application to particular
creatures. However, other evidence undermines this defence.
Astuti and Harris (submitted) report that 7-year-old Vezo
children in Madagascar generally assert that all processes that
sustain or are sustained by life, including cognitive and emotional
processes, cease at death. The most plausible explanation for
this finding is that Vezo children have considerable first-hand
experience of the biology of death because they observe and
actively participate in the slaughter and dismemberment of
animals, and they routinely attend funerals where they observe
the persistent immobility of the corpse and experience the
stench of decomposition. On the other hand, they are given no
explanation of the meaning of the various ancestral and burial
rites that they witness (Astuti, forthcoming a).
Third, we note that other developmental findings cast
doubt on Bering’s simulation-based proposal that children find
it difficult to conceive of the absence or cessation of mental
processes, including thinking, because they have never experienced
any such cessation. A series of experiments by Flavell
and his colleagues has shown that young children readily
conceive of an absence of thinking. Indeed, they do so in circumstances
where adults would typically assume that thinking is all
but inevitable. For example, when asked whether it is possible
to sit quietly and entertain no thoughts for a sustained period,
the majority of 5-year-olds assert that it is possible (Flavell
et al. 2000).
Finally, we note that whatever disposition children and adults
show toward afterlife beliefs, their assertion or denial of those
beliefs is quite context-sensitive. When asked about death in
the context of religious practices, beliefs in continuity are activated;
when asked about death in the context of medical or
secular practices, beliefs in discontinuity are activated (Astuti &
Harris, submitted; Harris & Gime´nez 2005). This contextsensitivity
is mirrored in everyday life: A dead corpse may be
prepared for burial with no expectation that it retains sentience;
the dead person, by contrast, may well be attributed thoughts and
feelings (see Astuti [forthcoming b] for an ethnographic illustration
from Madagascar).
In sum, although we do not dispute the claim that children and
adults are prone to think in a dualistic fashion, we doubt that
such a tendency reflects an evolved system dedicated to afterlife
beliefs.
Folk psychology meets folk Darwinism
Jay Hegde´a
and Norman A. Johnsonb
a
Department of Psychology, University of Minnesota, Minneapolis, MN 55455;
b
Department of Plant, Soil, and Insect Sciences, University of Massachusetts,
Amherst, MA 01003.
hegde@umn.edu http://www.hegde.us
njohnson@ent.umass.edu
http://www.umass.edu/psis/personnel/johnson.html
Abstract: The fact that beliefs in the supernatural are useful to people
who hold them does not necessarily mean that these beliefs confer
an evolutionary advantage to those who hold them. An evolutionary
explanation for any biological phenomenon must meet rigorous criteria,
but the facts in this case, even when taken at their face value, fall well
short of these criteria.
Evolutionary adaptation is a special and onerous concept that should
not be used unnecessarily, and an effect should not be called a function
unless it is clearly produced by design and not by chance.
—G. C. Williams (1966)
Bering’s proposal has two main facets. The first deals with evidence
that purports to show that beliefs in the supernatural are
advantageous to those who hold them. For the sake of argument,
we will take this evidence at face value so we can focus on the
second facet of Bering’s proposal, which consists of his hypothesis
that such beliefs arise from “an organized cognitive
‘system’ dedicated to forming illusory representations” that has
“evolved in response to the unique selective pressures of the
human social environment” (target article, sect. 1, para. 5).
Bering claims that his hypothesis is “grounded” in the theory of
natural selection, but provides no explanation whatsoever as to
how. This is unfortunate, because such an exercise would have
made it self-evident that an evolutionary hypothesis is neither
warranted nor justified in this case. Although Bering’s hypothesis
is intuitively appealing, it is invalid because it arises from a misapplication
of the theory of natural selection. Given the evidence
at hand, invoking the theory of natural selection to explain the
prevalence of beliefs is about as scientifically valid as invoking
the theory of gravitation to explain the attraction between two
people.
Briefly, for a given trait to evolve through natural selection,
individual organisms with the trait must have greater fitness,
that is, be more likely to survive and reproduce. Specifically,
this means not only that a given trait must be heritable, but
also that the trait must either increase the fitness of the organism
by itself or must be associated with (or, technically speaking, be
pleiotropic to) one or more of other heritable traits that do so (for
a more rigorous treatment of the subject, see Futuyma 1998).
In the present context, the aforementioned evolutionary principles
require not just that the belief in the supernatural be a
heritable trait, but also that those in the relevant previous generations
who held such beliefs were more likely to have survived
and reproduced. Bering does not even begin to establish either
the heritability or the increased fitness, much less both. The
closest he comes to doing this is to argue that the beliefs in the
supernatural are useful to people who hold them. But sociological
utility is a far cry from fitness. In other words, the fact that
those who hold such beliefs fit in better from the social
standpoint does not mean that they have greater fitness from
the evolutionary standpoint. Of course, our objection here is
not that Bering fails to use insider’s jargon when referring to
evolutionary concepts, but that he confuses non-evolutionary
concepts for evolutionary ones. In as much as he infers natural
selection based on sociological utility, Bering is indeed confusing
utility with fitness. As to heritability, the closest Bering comes to
addressing it is to argue that “Kindergartners understood that
various biological imperatives . . . no longer applied to the dead
mouse” (sect. 2, para. 2), which hardly lays the issue to rest.
Of course, natural selection is not the sole mechanism of evolution.
Mutation, the ultimate source of variation on which natural
Commentary/Bering: The folk psychology of souls
476 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
selection can act, can alone result in evolutionary change over
time. While natural selection is the driving force behind all
adaptive evolution, non-adaptive processes such as genetic
drift, meiotic drive, and a few other forces can also lead to evolutionary
changes (Futuyma 1998). However, Bering’s hypothesis
would not be any more valid if it were based on non-adaptive
evolutionary forces.
To be fair, Bering is hardly alone in misapplying the theory of
evolution to explain higher cognitive functions. Since all living
things are products of evolution, there is a widespread tendency
to treat evolution as a default explanation for all things biological.
Although this is understandable, it is also scientifically naı¨ve.
Of course, that is not to say either that cognitive phenomena
have no basis in evolution or that they inherently defy evolutionary
explanations. Rather, it is to emphasize that any evolutionary
explanation for a given biological phenomenon, cognitive or
otherwise, must at a minimum demonstrate that the relevant
trait is heritable and, in cases where natural selection is
invoked, that it increases fitness. The genuine difficulty of studying
the evolutionary basis of cognitive phenomena is that both
heritability and fitness effects are exceedingly hard to establish
for these phenomena. This does not mean that no evolutionary
explanations for such phenomena are to be ventured, but that
they are to be ventured with appropriate caution and adequate
groundwork. Clearly, Bering’s hypothesis is burdened with
neither.
In a sense, evolutionary biology of higher cognitive phenomena
is like astrophysics or paleontology, where direct measurements
are often all but impossible, and experimentation is
harder. In such cases, one has no choice but to substitute tests
and measurements with informed speculation, “informed”
being the operative word. But in such an event, the speculative
aspects must not only be acknowledged, but highlighted, and
the underlying risks and implications of the substitutions must
be carefully assessed. Bering does none of this. In light of all
these problems, it is surprising to us that Bering chooses to
couch his hypothesis in the onerous theory of natural selection
and not some less exacting and more suitably ambiguous
concept like cultural evolution (see, e.g., Mesoudi et al. 2006;
Richerson & Boyd 2005). Why must it be natural selection and
why won’t a less demanding theory do? Being does not say.
Ultimately, in order to establish that his hypothesis has any
relation to the theory of natural selection, Bering must, at a
minimum (1) demonstrate heritability and fitness effects for the
belief system in question, (2) prove that these parameters are
somehow irrelevant to his hypothesis, or (3) show that our formulation
of the minimum requirements of the theory of natural
selection is incorrect. Failing this, he must concede that his
hypothesis has no basis whatsoever in evolutionary theory.
Natural selection and religiosity: Validity
issues in the empirical examination of afterlife
cognitions
Brian M. Hughes
Department of Psychology, National University of Ireland, Galway, Galway
City, Ireland.
brian.hughes@nuigalway.ie
Abstract: Bering’s target article proposes that the tendency to believe in
an afterlife emerged (in evolutionary history) in response to selective
pressures unique to human societies. However, the empirical evidence
presented fails to account for the broader social context that impinges
upon researcher–participant interactions, and so fails to displace the
more parsimonious explanation that it is childhood credulity that
underlies the acquisition of afterlife beliefs through cultural exposure.
As part of a fascinating case for a folk psychology of souls, Bering
argues that believing in an afterlife is an evolutionarily inherited
human tendency. However, although he provides much illustrative
evidence, it is largely circumstantial in nature. Bering fails
to take account of threats to validity that inevitably arise when
researching such speculative and sensitive cognitions as
people’s beliefs in their own psychological immortality.
To support the claim that afterlife beliefs are innate, Bering
cites research where child participants are asked to describe
the ongoing thoughts of a recently killed (fictitious) mouse
(Bering & Bjorklund 2004). The assumption inherent in this
work is that as children have not yet developed explicit religiosity,
their quasi-religious views are more likely to be innate than
acquired. Thus, when the children respond that the animal
continues to have thoughts and wishes, the researchers conclude
that this indicates their belief in an afterlife. However, the external,
internal, and construct validity of such research is highly
questionable.
External validity is threatened because children’s views on
dead mice are not clearly generalizable to their beliefs about
the immortality of souls. For one thing, children’s well established
capacity to engage in counterfactual thinking (Riggs &
Peterson 2000), which underlies their ability to engage in
pretend play, may lead them to think differently about dead
mice in experimental vignettes compared to dead people in
real life. Internal validity is threatened by a failure to include a
control condition, wherein children’s beliefs about the agency
of inanimate objects in general might be probed. The attribution
of agency to inanimate objects has been observed in both children
and adults (Barrett & Johnson 2003). Thus, it is impossible
to determine whether children’s comments about the “thoughts”
of dead mice are any more profound than similar comments
about chairs, cars, or computers.
As is typically the case in research with children, construct
validity is threatened by the likelihood that responses to experimental
questions will be influenced by the experimenters’
seniority in age and status. The fact that children make what
for them are counter-intuitive inferences in order to accommodate
the assumptions implicit in (adult) researchers’ odd questions
is long documented in psychology (e.g., McGarrigle et al.
1978; cf. Hilton 1995). In this case, perceiving the adult to be
an authority figure, child participants may have inferred from
the questions asked that it is to be expected that the mouse’s
mind continue to function. As it cannot be guaranteed that
participants genuinely hold the beliefs attributed to them, the
question of whether such beliefs might be innate becomes moot.
Rather than postulating an innate propensity to believe in
souls, a more parsimonious theory might invoke the evolutionary
benefits of credulity among children. Given the need for guidance
to navigate the treacherous environments that characterize
early childhood, it is likely that children’s unquestioning faith in
whatever adults tell them is highly adaptive (Dawkins 2003). As
virtually all young children are presented (directly and indirectly)
with the idea of the immortality of souls, it should be unsurprising
if such a notion becomes widely believed. It is this propensity
for credulity that represents evolution’s legacy to spiritualism,
and not an innate propensity to intuit the existence of an afterlife
per se. By relying on fewer antenatal inputs, theories of innate
credulity are more parsimonious than ones of innate beliefs
about existence. Indeed, researchers who infer an innate belief
in afterlives in the absence of sufficient evidence could themselves
be accused of holding unsubstantiated beliefs in a
beforelife, namely, the sense in which an individual’s personhood
“exists” (such that it is endowed with fundamental beliefs) before
he or she is even born.
However, Bering may well be correct about the reasons
why many adults develop strong beliefs in afterlives (which are
then transmitted to credulous children). Nonetheless, gathering
empirical evidence here is also problematic, as problems
arising from experimenter–participant interactions are not
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 477
confined to research on children. For example, in another study
cited (Bering 2002a), adult participants are presented with vignettes
and asked questions like “Now that [the person] is dead,
does he want to be alive?” This research is mentioned in the
context of simulation constraints, and so participant hesitation
is taken to imply an incapacity (among adults) to imagine what
being dead is like. However, again, the participant’s judgment
of the researcher’s own mental state is being ignored. It could
simply be that participants hesitate because they are confused
by an apparently bizarre interrogation (asking themselves “Is
this a trick question?”), or are contemplating how best to be
polite in a socially awkward situation (“How do I respond
without offending the questioner’s apparent belief in an afterlife?”).
Adults may readily imagine death, as might be suggested
by research that examines the consequences of being invited
to do so (e.g., research into Terror Management Theory;
Goldenberg et al. 2000).
However, despite the precarious nature of self-report evidence
in studies of controversial, emotionally charged belief systems,
Bering’s argument is not necessarily empirically unsupportable.
Comparison of the views of children who are and are not
presented with afterlife concepts by their environments (e.g.,
by their parents) might elucidate to what extent children
develop such beliefs spontaneously. Objective (e.g., biological)
indices of behavior may also be revealing. Studies of phenomena
such as the placebo effect and its stimulation by social support
(Wall 1999) may corroborate claims that humans possess innate
characteristics that reinforce “moral” behavior (which, by providing
people with a stake in long-term outcomes of behavior, would
indirectly support folk assumptions regarding psychological
immortality), while also informing theories about the evolution
of moral judgment. Complementary evidence may emerge
from research into the genetics of altruism (e.g., Jansen & van
Baalen 2006).
In summary, it is clear that many people believe in an afterlife.
However, Bering’s case that such a belief is evolutionarily primed
(and therefore innate) is persuasive but not conclusive. It does
not displace the more parsimonious explanation that childhood
credulity underlies the acquisition of afterlife beliefs through
cultural exposure.
Transcendental self-organization
Carl N. Johnsona
and Melanie Nyhofb
a
Department of Psychology in Education, University of Pittsburgh, Pittsburgh,
PA 15260; b
Department of Psychology, University of Pittsburgh, Pittsburgh,
PA 15260.
johnson@pitt.edu men19@pitt.edu
Abstract: Bering makes a good case for turning attention to an organized
system that provides the self with transcendental meaning. In focusing
on the evolutionary basis of this system, however, he overlooks the selforganizing
properties of cognitive systems themselves. We propose that
the illusory system Bering describes can be more generally and
parsimoniously viewed as an emergent by-product of self-organization,
with no need for specialized “illusion by design.”
Bering seeks to direct the cognitive science of religion beyond its
recent focus on concept acquisition and agency detection toward
considering how supernatural inferences frame the meaning and
morality of the self. This shift potentially opens the door for links
with the emerging study of spiritual development, which has
otherwise been focused on issues of meaning, morality, and
identity (see Roehlkepartain et al. 2006). In his present article,
however, Bering speaks exclusively to evolutionary scholars,
encouraging them to explore the possibility that an illusory
cognitive system evolved as the result of selective pressures.
While worthy of exploration, Bering’s evolutionary proposal
is limited in two significant ways. First, the “Darwinian
mechanisms” are left completely unspecified. Second, the Darwinian
proposal is not weighed against a non-Darwinian alternative.
Bering leaves it for future investigators to explore the mechanisms
that generate the illusory existential system. It is not even
clear what the mechanisms are supposed to produce. The
system as a whole includes three components: ordinary cognitive
processes (simulation, teleology, and theory of mind), the specific
illusions, and their organization into a cognitive system. Presumably,
Bering is not looking to account for the basic cognitive
processes. The search, hence, must be for some added illusionproducing
and integrative mechanisms that generate a distinctive
metaphysical theory of self.
The alternative, more parsimonious possibility is that the
cognitive illusory system emerges from ordinary processes
through self-organization. In a Kantian sense, transcendental
illusions are the inevitable product of the operation of ordinary
cognitive processes as they extend beyond normal boundaries
of operation. Beside the illusions that Bering describes, there
are classic illusions that arise from reflective ideas, wherein the
order inherent in concepts is uncritically assumed to exist in
the world. In any case, once generated, these transcendental
ideas are powerfully relevant and pragmatically regulatory, precisely
because they reflect higher-order organization that is
intrinsically valuable to the self (see Johnson 2000).
Systems of transcendental belief are thus the result of selforganization,
whereby ideas generated by the self come to
organize and regulate the self. In this framework, religious
ideas are not the sterile by-product of cognitive relevance
(attention and memory). Nor are they specifically adaptive
illusions by design. Rather, they are emergent by-products that
have self-relevance.
Epidemiologically, religious ideas are spread, not simply
because of their cognitive relevance, but because of their vital
relevance. Religious ideas stick around because they are relevant
to the goals, status, and value of the self.
Transcendental illusions are the natural outgrowth of human
cognitive organization. The cognitive system primarily functions
to orient the organism to what is vitally important, not what is
strictly, objectively real. To this end, information is organized
in terms of prototypes, ideals, essences, narratives, and the like.
These organizational processes commonly give rise to ideas
regarding the existence of a higher, deeper order, beyond the
perceptible given.
Clearly we need to know a lot more about the origins and
adaptive function of transcendental ideas. Bering turns attention
to a particularly intriguing system of belief. Whether or not this
particular system was selected by design, we need to better
understand the wider human tendency to imagine transcendental
order that serves to regulate the self.
Six feet over: Out-of-body experiences and
their relevance to the folk psychology of souls
David Kemmerera
and Rupa Guptab
a
Department of Speech, Language, and Hearing Sciences, Purdue University,
West Lafayette, IN 47907; b
Department of Biological Sciences, Purdue
University, West Lafayette, IN 47907.
kemmerer@purdue.edu gupta10@purdue.edu
http://www.cla.purdue.edu/slhs/Pages/fac_staff/faculty/
kemmerer.html
Abstract: During an out-of-body experience (OBE), one sees the world
and one’s own body from an extracorporeal visuospatial perspective.
OBEs reflect disturbances in brain systems dedicated to multisensory
integration and self-processing. However, they have traditionally been
interpreted as providing evidence for a soul that can depart the body
after death. This mystical view is consistent with Bering’s proposal that
psychological immortality is the cognitive default.
Commentary/Bering: The folk psychology of souls
478 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Religious experience and behavior play important roles in all
human cultures and hence deserve to be treated as natural
phenomena worthy of careful scientific investigation (Dennett
2006). We commend Bering for his insightful and provocative
contribution to this new field of research.
Bering argues that “common-sense mind-body dualism” is a
cognitive adaptation that evolved through natural selection.
According to this view, human beings are designed to believe
that everyone has an immaterial, immortal soul that is linked to
the body during life but leaves it behind after death (see
Humphrey 2006, pp. 124–29, for a similar argument). In this commentary
we relate Bering’s proposal to one of the most bizarre and
emotionally powerful alterations of consciousness that people are
capable of undergoing, namely out-of-body experiences (OBEs),
in which the subjective sense of self appears to part company
with the physical body (e.g., Blackmore 1982; Blanke et al. 2004;
Brugger 2006; Green 1968; Metzinger 2003).
Blanke and Arzy (2005, p. 16) state that an OBE has three phenomenological
characteristics: “disembodiment (location of the
self outside one’s body), the impression of seeing the world
from a distant and elevated visuospatial perspective (extracorporeal
egocentric perspective), and the impression of seeing one’s
own body (autoscopy) from that elevated perspective.” This is
illustrated by the following example (Irwin 1985; case 1): “I was
in bed and about to fall asleep when I had the distinct impression
that ‘I’ was at the ceiling level looking down at my body in the
bed. I was very startled and frightened; immediately [afterward]
I felt that I was consciously back in the bed again.” OBEs have a
prevalence of approximately 10% in the general population
(Blackmore 1982; Irwin 1985, pp. 219–59). They occur in
many diverse cultures (Shiels 1978) and are frequently mentioned
in folklore, mythology, spiritual writings, and literature
(e.g., Arzy et al. 2005; McCulloch 1992). Indeed, they are so
widespread that Metzinger (2003, p. 502) calls them a “phenomenological
archetype” of humanity. Although OBEs can be
induced by hallucinogenic drugs such as ketamine (Hansen
et al. 1988) and phencyclidine (PCP; Rosse et al. 1994),
they happen spontaneously only once or twice in a lifetime
(Blackmore 1982; Green 1968), usually in dangerous, traumatic
situations such as rape (Sierra & Berrios 1998) and near-death
episodes (Greyson 2000). Remarkably, in such circumstances
subjects feel as if it is their bodies that are threatened, not their
selves. Taking all of these factors into consideration, it is not surprising
that OBEs have been widely regarded throughout history
as confirming the intuition that every human being has an
ethereal soul that can literally detach from the physical body,
most importantly when that body expires. Metzinger (2003,
p. 503) even goes so far as to formulate the “soul hypothesis,”
which maintains that OBEs are what “first led human beings to
believe in a soul” (see also Metzinger 2005).
As yet, however, psychological experiments have failed to
verify the supernatural interpretation of OBEs as involving
genuine mind–body separation (Alvarado 1992; 2000; Blackmore
1982, pp. 200–39; Irwin 1985). In addition, Olaf Blanke and his
colleagues have succeeded in demystifying OBEs even more by
marshalling several sources of neuroscientific evidence that
suggest that these strange experiences arise from abnormal
self-processing in the temporo-parietal junction (TPJ), predominantly
in the right hemisphere (for reviews see Blanke & Arzy
2005; Blanke & Mohr 2005; Mohr & Blanke 2005). During an
invasive cortical mapping procedure with an epileptic patient,
it was found that direct stimulation of the TPJ reliably elicited
OBEs and other types of visual body-part illusions (Blanke
et al. 2002). More recent studies with neurological patients
(Blanke et al. 2004) and healthy subjects (Blanke et al. 2005)
have corroborated the importance of the TPJ in generating
OBEs and have begun to reveal the specific neurophysiological
mechanisms that underlie them. Normally the TPJ helps create
a unified, central representation of the body – a physical
anchor for the mental self – by integrating visual, tactile,
proprioceptive, and vestibular signals. OBEs may therefore
arise when paroxysmal dysfunctions in the TPJ lead to strong discrepancies
between the felt and the seen position of one’s own
body. Blanke and Arzy (2005) suggest that otholithic vestibular
dysfunctions may be an especially important precipitating
factor for OBEs, because they have been independently linked
not only with feelings of elevation and floating, but also with
180
Ð
inversions of one’s visuospatial perspective. In particular,
such illusions have been experienced by astronauts during
space missions (Mittelstaedt & Glasauer 1993) and by pilots
during the microgravity phase of parabolic flights (Lackner
1992). Further research will undoubtedly continue to illuminate
the neural bases of OBEs and their role in religious activity
(Previc, in press). For example, there may be connections
between OBEs and new evidence that the TPJ is engaged
when a person imagines how the spatial relations between
two objects would appear from someone else’s point of view
(Aichhorn et al. 2006). Similarly, there may be connections
between OBEs and new evidence that partially distinct cortical
regions subserve the visual perception of one’s own and other
people’s body parts (Saxe et al. 2006).
Nevertheless, it seems likely that no matter how much progress
is made in explaining OBEs solely in terms of the structures
and operations of the brain, a substantial proportion of the
human population will still prefer to interpret OBEs as involving
a true liberation of the soul from the body, a liberation of the kind
that everyone ultimately undergoes when they die. After all, as
Bering points out, belief in psychological immortality seems to
be our cognitive default.
ACKNOWLEDGMENT
Our title is borrowed from Roach (2006).
Cultural adaptation and evolved, generalpurpose
cognitive mechanisms are sufficient
to explain belief in souls
Kenneth R. Livingston
Department of Psychology and Program in Cognitive Science, Vassar College,
Poughkeepsie, NY 12604-0479.
livingst@vassar.edu
Abstract: It is suggested that general-purpose cognitive modules are
the proper endophenotypes on which evolution has operated, not
special purpose belief modules. These general-purpose modules
operate to extract adaptive cultural patterns. Belief in souls may be
adaptive and based in evolved systems without requiring that a specific
cognitive system has evolved to support just such beliefs.
In its strong form Bering’s evolutionary adaptationist argument in
the target article proposes to explain how it is that so many hundreds
of millions of people are capable of believing the same six
impossible things before breakfast. In this same strong form it
leaves unanswered the question of why so many hundreds of
millions of people (estimates vary but see Barrett et al. 2006)
disavow such beliefs. Bering suggests that the evolved tendency
to believe in souls remains operative even in self-proclaimed
“extinctivists,” some of whom endorse the idea that dead
people know that they are dead (Bering 2002a), but this
finding hardly shows that this confusion reflects an evolved adaptation.
Thoughtful and clever as it is, Bering’s analysis presents us
with a false alternative between two explanations of the widespread
belief in souls. The cultural epidemiological alternative
views all religious ideas, including ideas about an afterlife, as
non-adaptive byproducts of general-purpose cognitive processes.
Bering’s alternative is that the belief in an afterlife is a specific,
evolved adaptation that extends the temporal boundaries of
the self in ways that minimize counter-reproductive behavior. I
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 479
would like to suggest a third option that combines the cultural
selection of the former view with a limited version of the adaptationism
of the latter approach.
This cultural adaptationist view has to show that the case for
specific biological adaptation is weak, while showing that belief
in souls has some adaptive value under at least some circumstances.
Space permits only a sketch in response to Bering’s evolutionary
argument, but the general strategy can be demonstrated
with a few examples. For example, Kuhlmeier et al. (2004) found
that 5-month-olds do not apply the same principle of continuous
motion to humans that they apply to inanimate objects, from
which Bering concludes that they are intuitive dualists. But
Kuhlmeier et al. themselves acknowledge that one cannot tell,
“whether the results of the present study are due to a distinction
between animates versus inanimates, intentional agents versus
non-intentional objects, or humans versus other entities”
(p. 101). Nor does the fact that kindergartners make more psychological
attributions to a dead mouse than do older children or
adults show that the origins of such beliefs cannot be exclusively
cultural. Children routinely produce conceptual overgeneralizations
in early and middle childhood (Bloom 2004). Furthermore,
the non-adult pattern of attributions to the dead made by young
children must be viewed in the light of considerable evidence
that the concept of death itself is poorly grasped until well into
middle childhood (Childers & Wimmer 1971) and varies as a function
of culture (Yang & Chen 2006) and religious upbringing
(Florian & Kpavetz 1985). In general, alternative accounts,
sketched below, are available for the evidence Bering reviews.
Bering carefully notes that specific beliefs about the afterlife
can vary across cultures, arguing that the tendency to have
beliefs of this kind is universal and thus best explained by
appeal to evolutionary processes. But general-purpose cognitive
mechanisms operating on varying cultural content may still find
common patterns or kinds because some of these mechanisms
are in the business of doing just that. For example, the fact
that people continue to behave on occasion as if a deceased
person is still alive has to be weighed against the same tendency
to continue habitual behaviors toward vanished inanimate
objects (walking around a chair that has been moved, reaching
for a light switch that has been replaced, etc.). Similarly, the
fact that there is a tendency in many cultures to reinforce authoritarian
proscriptions by appeal to unseen watchers (ancestors,
gods, or God) might be evidence for an evolved functional illusion
that the self transcends time and place, but an equally
sound argument can be made that a combination of a general
capacity for off-line prediction and the ability to think abstractly
about non-perceptual events and objects have been recruited by
many successful cultures to enhance the adaptation of their
members to life’s vagaries. Although the relevant theory in this
case is also speculative, Occam’s razor favors an appeal to the
evolutionarily adaptive advantages of these general-purpose cognitive
endophenotypes (e.g., see Kanazawa 2004) over the appeal
to a specific, error-based module for belief in souls. Belief in an
afterlife is far from the only common feature of religious systems
and one can’t help worrying that religion modules will begin to
proliferate uncontrollably. If we need a special purpose soul
system, can modules for belief in creation narratives, the power
of ritual and magical artifacts, angels, demons, incantations,
and so forth be far behind?
Pursuing the theoretical strategy of using general cognitive
mechanisms to explain religion does not, however, commit one
to the view that religion is maladaptive or non-adaptive. A substantial
literature indicates an association between religious
commitment and happiness, health, and well-being (Hiadt
2005; Koenig & Cohen 2002; Livingston 2002). Often overlooked,
however, are the individual differences in the adaptive
significance of religious belief, including belief in the afterlife.
For example, Ellison (1991) has shown that the benefits of
religious belief tend to accrue only to those who hold them
with strong conviction. Furthermore, this degree of existential
certainty interacts with level of education and frequency of life
trauma to determine life-satisfaction and happiness. It also
appears that strong atheistic convictions confer the same benefits
(Shaver et al. 1980). Content of belief may matter less than commitment,
a pattern difficult to reconcile with Bering’s account.
Bering has performed an invaluable service by attempting to
integrate a disparate set of findings in support of an evolutionary
account of why human beings seem so drawn to belief in souls.
Indeed, it is easy to get caught up in finding other patterns
consistent with his view (e.g., the transformation of Buddhism
over the centuries from an atheistic, non-agentistic religion in
its original form to a system heavily populated with souls and
spirits in many of its more modern forms; see Livingston 2005).
By making the case so forcefully, he compels us all to think
more carefully and in greater detail about how to account for
the phenomena he describes. Some reformulation of existing
theory is clearly needed, but for the present, cultural adaptationism
grounded in evolved general purpose cognitive systems
represents a viable alternative to his account. Among the
virtues of this alternative is that it more readily explains
widespread and increasing rates of disbelief, as well as the folk
psychology of souls.
Beliefs in afterlife as a by-product of
persistence judgments
George E. Newman,a
Sergey V. Blok,b
and Lance J. Ripsc
a
Department of Psychology, Yale University, New Haven, CT 06520;
b
Department of Psychology, University of Texas at Austin, Austin, TX 78712;
c
Department of Psychology, Northwestern University, Evanston, IL 60208.
george.newman@yale.edu blok@psy.utexas.edu
rips@northwestern.edu
Abstract: We agree that supernatural beliefs are pervasive. However, we
propose a more general account rooted in how people trace ordinary
objects over time. Tracking identity involves attending to the causal
history of an object, a process that may implicate hidden mechanisms.
We discuss experiments in which participants exhibit the same
“supernatural” beliefs when reasoning about the fates of cups and
automobiles as those exhibited by Bering’s participants when reasoning
about spirits.
The central claim of Bering’s thought-provoking target article is
that evolution has produced a dedicated cognitive system to
support illusory beliefs in a soul – a psychological self that
persists after the physical body has ceased to exist. Here we
suggest, instead, that a more general and mundane cognitive process
– one needed to track individuals over time – may account
for belief in the survival of these individuals after death.
To conceive of any individual requires the ability to identify it
as the same entity over time and place. Often, such tracking must
occur through interruptions (e.g., occlusion, lapses of attention,
or sleep), changes in appearance (e.g., a child growing into an
adult), or feature instabilities (e.g., a cloud changing shape).
Here we focus on how people reason about the persistence of
objects – more formally, how people decide that a description
of an object at one time t0 belongs to the same object as does a
description at another time t1 (for a review of the way the
visual system makes similar judgments, see Scholl 2001).
Most philosophers agree that causal factors play a role in
object persistence (Nozick 1981; Parfit 1984). Based on these
accounts, we have recently proposed a cognitive theory that
specifies the role of causality in judgments of identity over time
(Rips et al. 2006). According to this Causal Continuer account,
two descriptions belong to the same object if (a) the object at
t1 is among those that are causally close enough to be genuine
continuers of the original item, and (b) it is the closest of these
close-enough contenders.
One of the key features of the causal continuer model is
that changes in similarity, spatial-temporal continuity, or even
Commentary/Bering: The folk psychology of souls
480 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
basic-level category membership do not necessarily entail that an
object goes out of existence. For example, Blok et al. (2005)
report an experiment in which participants read stories about
an individual (e.g., Jim) who has a severe traffic accident and
must undergo radical surgery. Participants learned that Jim’s
brain was transplanted to a different body. On some trials, scientists
placed the brain in “a highly sophisticated cybernetic body,”
whereas on others they placed it in a human body that scientists
had grown for such emergencies. In each case, Jim’s old body was
destroyed. The stories described the operation as successful in
allowing the brain to control the new body, but participants
also learned either that Jim’s memories survived the operation
intact or did not survive. After reading the scenario, participants
rated their agreement with each of two statements: (a) the transplant
recipient is Jim after the operation, and (b) the transplant
recipient is a person after the operation.
Participants were more likely to agree that the post-op recipient
was still Jim if Jim’s memories were preserved. But whether
these memories were embodied in a human or in a robot body
had a much smaller effect. In contrast, agreement about
whether the end product was a person mainly depended on
whether the recipient object had a human rather than a robot
body, and relied less heavily on whether Jim’s memories
remained intact. This combination of effects produced the
finding that when Jim’s memories survived in a robotic body,
participants were much more likely to think that the transformed
individual is Jim than that the transformed individual is a person!
The belief that Jim persists despite a radical change in basic-level
category may be analogous to the belief that there is an intuitive
causal continuer that shares a person’s psychological characteristics
after death – the individual is the same, yet the category
has changed. However, such judgments fall out of predictions
made by the Causal Continuer model and need not derive
from a specialized cognitive system for theological beliefs, as
Bering posits.
Bering claims that the evolutionary rationale for such an innate
theological system was to tame the self so that it became “less
likely to engage in acts that, if publicly exposed and harmful to
one’s social reputation, seriously impaired genetic fitness”
(sect. 5, para. 1). This system should therefore apply with particular
force to people. Similarly, as Bering notes, other theories
hypothesize that the concepts of person and animal may
promote supernatural beliefs because these concepts “act as
flypaper for salient, ‘counterintuitive’ cases” (sect. 2.3, para. 3)
(e.g., Atran & Norenzayan 2004; Barrett 2000; Sperber &
Hirschfeld 2004). Beliefs about the persistence of individual
objects, however, are clearly not limited to persons or animals
and are not necessarily counterintuitive or supernatural. Thus,
to determine whether causal identity mechanisms provide a
better account than a special theological one, it is important to
consider cases involving nonpersons. For example, do we
observe similar patterns of judgments with artifacts as those we
found with people like Jim?
In a second study, Blok et al. (2005) told participants about a
sci-fi “transporter” that was capable of dividing an object into
its most basic particles (a device that surely would have stirred
the interest of any Star Trek fan). Once disassembled, the particles
were sent through a “particle pipeline” and then reassembled
on the other end. The one catch to the transporter
was that sometimes there was a glitch – occasionally the reassembled
product came out looking like a different type of
object. For example, a car might turn out to resemble a boat. Participants
read about transformations involving both living kinds
and artifacts (e.g., Jim’s cat “Nancy,” or Jim’s car “Rustbucket”).
After reading each scenario, participants rated their agreement
with two statements: (a) the object is a [car] after the transformation,
(b) the object is [Rustbucket] after the transformation.
Relevant to the present discussion, we observed a pattern of
judgments similar to those of the person experiment described
above. For both animals and artifacts, when the transformed
item had the appearance of a different (but neighboring) category,
ratings of category membership were reduced to a
greater extent than were ratings of individual persistence. For
instance, when participants were told that the car Rustbucket
was sent through the transporter and reconfigured to resemble
a boat, people lowered their agreement ratings more to the statement
that it was still a car than to the statement that it was still
Rustbucket. Thus, belief in the persistence of individuals
through radical changes in kind is not restricted to persons and
need not include the notion of a soul. At least for some participants,
it is more likely that Rustbucket is “reincarnated” as a
boat than that Rustbucket ceases to exist when it ceases to be a
car. As a perhaps more ecologically valid example, some cultures
buried their dead with treasured artifacts (e.g., the Egyptians
or Mayans). Though obviously these artifacts (eventually)
decompose along with the body, such practices are consistent
with the idea that artifacts “survive” death in the same manner
as persons.
In sum, we suggest that general cognitive processes –
processes dedicated to keeping track of individuals across time
and transformation – may account for beliefs in an afterlife.
Religious and supernatural dogmas no doubt serve to enhance
the richness of such beliefs. They may be responsible for the
idea that deceased individuals live in a heaven populated with
loved ones. But the basic process of inferring the existence of
individuals after death may be a natural consequence of everyday
strategies for tracking these individuals. The “illusory” aspect of
belief in an afterlife may result simply from the believers’ lack
of knowledge about what is, in fact, a fairly sophisticated
idea: that a person’s psychological characteristics depend on
bodily processes and therefore come to a halt when these
processes do.
If this is correct, there is no need to posit a special purpose
cognitive system to explain belief in a soul. Nor is there a need
to trace such beliefs to a mechanism that has evolved in order
to shape moral conduct. Bering could argue that object tracking
is just another cognitive feature that evolution has “set to work
on” to produce a system of hardwired theological beliefs. But if
object tracking is the cause of belief in a soul, then the theological
system seems to have little work to do, and, like the soul itself,
there is correspondingly little reason to think that it exists.
Do children think of the self as the soul?
Shaun Nichols
Department of Philosophy, University of Arizona, Tucson AZ 85721.
sbn@email.arizona.edu
Abstract: Bering’s work provides new insight into the child’s concept of
the self. For his results indicate that children don’t regard bodily identity
as required for identity of self across time. Bering’s methodology for
investigating afterlife beliefs might also be exploited to explore the
extent to which children think that psychological similarity is required
for sameness of self.
Jesse Bering’s delightful research indicates that the belief in an
afterlife is quite natural for children. The work also has important,
but largely unnoticed, lessons on the child’s concept of the
self. The results provide some evidence for, and a methodology
for exploring further, the hypothesis that children think of the
self as the soul.
One central tenet of the traditional view that the self is the soul
is that the self is not the body. Surprisingly, earlier developmental
work on the child’s concept of self has suggested that children
identify the self primarily with bodily features. For example,
when young children were asked “What will not change about
yourself when you grow up?”, 7-year-olds tended to refer to physical
characteristics (e.g., hair color) and only rarely referred to
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 481
psychological characteristics (Mohr 1978, p. 428). Indeed, one
prominent view has been that young children have only a
“physicalistic” conception of the self (e.g., Montemayor &
Eisen 1977; Selman 1980). Bering’s results provide the antidote
to this view. In Bering’s experiments, children tend to say
that Brown Mouse is still hungry and still thinking about
Mr. Alligator, despite the fact that Brown Mouse’s body has
been destroyed. This provides new evidence that children have
a concept of self that is not identified with physical, bodily features.
For it’s natural to interpret the children as claiming that
Brown Mouse – the same individual – persists after the destruction
of his body. This indicates that children don’t regard bodily
identity as required for personal identity. The children seem to
think that the same self persists across destruction of the body.
A second central tenet of the soul view is that the self can’t be
identified with a set of memories, thoughts, or other psychological
states. Thomas Reid expresses the point with characteristic
directness: “Whatever this self may be, it is something which
thinks, and deliberates, and resolves, and acts, and suffers. I
am not thought, I am not action, I am not feeling; I am something
that thinks, and acts, and suffers” (Reid 1785/1969, p. 341).
Psychological states change constantly, and so they seem too
fickle to be the basis for an enduring self. Rather, soul theorists
maintain that the self is the thing that has the capacity for
psychological states, regardless of the particular psychological
states it happens to have. A soul theorist maintains that it is
because the self endures that it is possible for the psychological
states to persevere.
Bering’s experiments do not tell us whether children’s views of
the self coincide with this second tenet of the soul view. For the
experiments do not explore whether children would maintain
that Brown Mouse can continue to exist even if he loses his
distinctive psychological states. However, Bering has given us
the most promising methodology to date for exploring this
question. We can use his methodology to examine the extent to
which children think that psychological similarity is required
for personal identity. This might be done with two changes to
Bering’s design. First, one would need to ask exclusively about
psychological capacities rather than psychological states. In
Bering and Bjorklund (2004), some questions are about the
specific psychological states that Brown Mouse had before he
got eaten. For instance, children are asked of Brown Mouse,
“Is he still thirsty?” and “Is he still thinking about Mr. Alligator?”
To show that children’s views cohere with the second tenet of the
soul view, we would have to ask only about capacities, like, “Will
Brown Mouse ever think again?” The second change is more
challenging – we would need to specify that after Brown
Mouse is eaten, there will no longer be anyone who has the particular
psychological states that Brown Mouse had. For instance,
we might explicitly state that no one will ever remember the
things that Brown Mouse did. If children still tend to maintain
that Brown Mouse will think and feel again, then this would
provide evidence that children’s notion of the self also follows
the second tenet of the soul view.
There is yet a third tenet of the traditional soul view – that the
soul is an immaterial substance. It is natural to think of this third
tenet as providing a story about the metaphysical ground of the
capacity to have psychological states. Bering’s data do not show
that this tenet is reflected in the child’s view of the self, nor
does he suggest otherwise. I think it unlikely that children
naturally have opinions on such rarefied issues in metaphysics.
Rather, the idea that there is an immaterial substance underlying
our psychological states is most likely an intellectual innovation
that has become part of the culture in major religious and philosophical
traditions. But if children find it intuitive that the self
can survive the death of the body and the radical disruption of
psychological states, this would go some distance to explaining
why the doctrine that the self is an immaterial substance has
achieved such cultural prominence.
The Godfather of soul
Jesse Preston,a
Kurt Gray,b
and Daniel M. Wegnerb
a
Department of Psychology, University of Western Ontario, London, Ontario
N6A 5C2, Canada; b
Department of Psychology, Harvard University,
Cambridge, MA 02138.
jpresto@uwo.ca kurtgray@fas.harvard.edu
http://www.wjh.harvard.edu/%7Ekurtgray/
wegner@wgh.harvard.edu http://www.wjh.harvard.edu/wegner/
Abstract: An important component of souls is the capacity for free will, as
the origin of agency within an individual. Belief in souls arises in part from
the experience of conscious will, a compelling feeling of personal
causation that accompanies almost every action we take, and suggests
that an immaterial self is in charge of the physical body.
Why is it useful to think we have souls? Jesse Bering would have
us believe that the concept of a soul is derived from evolutionary
pressures of several kinds. In his view, the soul may arise because
we are enamored of the idea of agents, and we overextend and
embellish this idea to the point of perceiving people as having
essences that persist after death – if not also existing prelife.
From Bering’s perspective, a soul is a construction we place on
others. But what of the soul we know the best – our own?
Bering’s approach offers no special status to our own souls
because it overlooks the experience of human agency. One key
use of the idea of a soul is to explain the unfathomable source
of our own ability to do things merely by wanting them. The
soul is a way of understanding the experience of conscious will.
Think for a minute of the magic of agency. You may not be able
to create world peace just by willing it, but there are a remarkable
number of things you can do nonetheless. You can open a soda
can, lift it up, and toast Bering – all through some wildly supernatural
procedure whereby the things you desire just go ahead
and happen! And how do you know that you did this? True,
there are a number of cues a person could use to sleuth out personal
responsibility for action. But for the most part, conscious
will is experienced as an authorship emotion rather than a
reasoned deduction. We think we possess free will because it
feels like we do. Feelings of personal causation are indicated
by certain cues present around the time of an action, like visual
feedback, and physical feedback from the body (Wegner &
Sparrow 2004).
Perhaps most essential to the experience of will is the connection
between thoughts and action. You find yourself thinking
about eating lunch, then lo and behold you find yourself down
at the diner ordering a cheeseburger. What else could have
caused this, if not your own free will? According to the theory
of apparent mental causation, the experience of will arises
when we are aware of some thought about the action just
before it happens, and the action cannot be readily connected
to some other cause (Wegner 2004). Ordering the burger feels
like a freely willed act if you were just thinking about cheeseburgers
and their deliciousness, but less so if the cheeseburger
option was just recommended by your pushy waiter.
Several studies conducted in our laboratory suggest that these
three principles of priority, consistency, and exclusivity create the
feeling of conscious will over actions, even if actions are not truly
self-caused. In a study modelled after the Ouija board, people felt
more control over the movements of the disk when given a
preview where it was about to go, and if this preview occurred
just before the action (Wegner & Wheatley 1999, see also Aarts
et al. 2005). Feelings of conscious will are attenuated, however,
when there are other competing forces or agents that might be
responsible for the action. For example, people feel more responsibility
if they are subliminally primed to think of the self just
before an action, but when primed with the identity of other
agents, feelings of responsibility decline (Dijksterhuis et al.,
under review). These cues are so persuasive that they can
create feelings of conscious will over actions clearly beyond
Commentary/Bering: The folk psychology of souls
482 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
personal control, including controlling the actions of others
(Wegner et al. 2004) and the ability to enact actions at a distance
(Pronin et al. 2006).
The feeling of conscious will is an inescapable part of life, and
the experience is so compelling it is hard to imagine that it may be
just an illusion. Conscious will is not only an essential component
to the concept of souls, but the experience can also help explain
why souls have their ethereal quality. As Bering points out, belief
in souls stems in part from the inability to reconcile how mind
arises from body (the hard problem of consciousness). Although
we have a pretty good understanding of how the heart, lungs,
stomach, and other organs function, when trying to comprehend
how the brain thinks we are still left scratching our heads. The
experience of will presents a particularly tricky question in this
puzzle. Agents are seen as first causes, or uncaused causes – the
origin and author of action. Discussions of the function of an
authorship emotion for physical action point to the need for a
controller of action, an internal operator who guides our
decisions (e.g., see Wegner 2005). But if the brain is the source
of cognition, then how does the thinker direct the thinking in
one direction or another? Even if we thought we could pin
down the location of the internal operator in the brain, we
would still need to explain the controller itself. The popular solution
to the search for free will is that the brain is not the ultimate
source of thinking, but there is some nonmaterial self – the
soul – that somehow operates to control the body.
The pervasive belief in an afterlife is possible because of this
mind–body division. And wherever the soul goes, it brings free
will with it. Indeed, the experience of free will is often included
in the folk understanding of the moment of death: Death occurs
when one is ready to die, loses the will to live, or chooses to go
into the light. Bering suggests that belief in an afterlife results
in part because we don’t know what it is like to be dead. We
may also be unable to fathom the absence of will. It is difficult
to imagine that such a potent force could just dissolve into
nothingness. After all its majesty and magic, surely conscious
will cannot be just snuffed out of existence, at least not without
its permission. James Brown, move over: Conscious will is the
Godfather of soul.
No evidence of a specific adaptation
Ilkka Pyysia¨inen
Academy of Finland/Helsinki Collegium for Advanced Studies, University of
Helsinki, FIN-00014 Helsinki, Finland.
ilkka.pyysiainen@helsinki.fi www.mv.helsinki.fi/home/ipyysiai/
Abstract: Bering’s findings about the mental representation of dead
agents are important, although his opposition between “endemic” and
“cultural” concepts is misleading. Endemic and cultural are
overlapping, not exclusive categories. It is also difficult to see why
reasoning about the dead would require a specific cognitive mechanism.
Bering presents no clear evidence for the claim that the postulated
mechanism is an adaptation.
Bering argues that his theoretical framework is in some respect
more plausible than the “standard model” (Boyer 2005) in the
cognitive science of religion (e.g., Bering 2002a; 2003a). He
claims that the “cultural epidemiologists” regard religious concepts
as entirely explicit in nature and as “socially acquired”
or being “generated by culture” (Bering 2002a, p. 293). Bering
(2002a; 2003a), however, thinks that ghost concepts are
“endemic” or “innate,” not something learned from others. The
“central thesis” of the target article is that an organized cognitive
system, dedicated to form illusory representations of psychological
immortality and symbolic meaning, evolved in response to the
unique selective pressures of the human social environment.
Bering raises an important question about religion and
adaptation, although his nativist claim is somewhat problematic.
He argues that specific afterlife beliefs are not “direct products”
of natural selection; what has been selected for is “an intuitive
pattern of reasoning” that does not hinge on the presence of
explicit religious concepts. Bering thus makes a distinction
between explicit religious concepts learned from culture and
innate patterns of reasoning. He has previously argued that
“the very concept [of a ghost] itself, has natural foundations in
the human mind,” asking whether it might be possible that
“the general idea of an afterlife is not so much implanted in
people’s heads by way of ‘exposure’ to counterintuitive tales,
as it is already present” in human cognitive structures (Bering
2002a, p. 269).
This line of reasoning is also present in the target article.
Bering argues that Boyer (2001), Pyysia¨inen (2001), Atran
(2002), and Sperber and Hirschfeld (2004) approach religion as
a (fuzzy) set of ideas that survive in cultural transmission
because they effectively parasitize evolved cognitive structures.
These “epidemiologists” are not interested in how “a representational
bias for envisioning personal immortality” has “impacted
the net genetic fitness of individual humans in ancestral
environments.”
There is, however, a conceptual confusion involved in Bering’s
way of distinguishing his approach from the standard model.
Barrett (2003), Boyer (2003a), and myself (Pyysia¨inen 2003)
pointed out three years ago that Bering’s dichotomy between
endemic and cultural concepts is misconceived; epidemiologists
do not regard culture as a set of “entirely explicit” concepts
that are simply learned, irrespective of implicit biases. Concepts
cannot be divided into two mutually exclusive classes: those we
are born with and those we learn from others. There are only
concepts represented in mind; some of them are communicated
to others and thus may become widespread. A representation is
cultural to the extent that it is represented by many persons
and is preserved in various versions over time, no matter what
its origins are.
According to Barrett (2003), nativists such as Bering focus on
how the cognitive machinery of individuals “produces” intuitive
ideas and behaviors that may then receive augmentation from
explicit ideas learned from culture. The epidemiologists, in
turn, are not necessarily interested in the origins of concepts
and beliefs; instead, they try to explain the distribution of representations
in populations. Widespread ideas are typically
such that they are easily adopted because it is possible to
enrich them by one’s intuitions. Thus, there need not be a contradiction
between nativism and epidemiology.
The actual difference between Bering and the standard model
is in that Bering presupposes a pattern of reasoning dedicated
specifically to making inferences about afterlife. Whereas
Boyer thinks that beliefs about dead agents are mediated by
psychological adaptations that enable us to reason about agents
in general, Bering holds that reasoning about dead agents is
based on a task-specific adaptation (Boyer & Barrett 2005;
cf. Stone & Gerrans 2006).
By the same token, Bering is committed to an adaptationist
view of religion, whereas in the standard model religion is an
evolutionary by-product (see Atran 2002; Bulbulia 2004). All
ideas of counterintuitive agents, such as ghosts, gods, and ancestors
are represented using the normal mechanisms of agent
representation, adding one minor modification or “tweak” to an
intuitive representation (Boyer 2003b). No specific mechanisms
are needed to mediate different types of agent representations.
Surprisingly, Bering also suggests that interactions with
imagined dead agents might be cases of “off-line” social cognition
in the sense that the dead are represented as absent but existing
persons. This, however, implies that there is no specific mechanism
for reasoning about dead agents. The specificity of the “intuitive
pattern” has to be only in its domain of application, not in the
mechanism itself.
If this is so, then we also have to look for adaptations in agent
detection and theory of mind in general, not in special concepts
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 483
of dead agents. The weakest part of Bering’s contribution is
precisely his evolutionary speculations. It is not much of an
argument to say that supernatural illusions may have had
fitness-enhancing effects, and that if this is so, then they might
represent a specific cognitive adaptation (see Sterelny 2003).
Bering hints at the problem of the evolution of cooperation but
does not really explore it (see Axelrod 1990; Nowak & Sigmund
1994; 2005; Nowak et al. 2004; Richerson & Boyd 2005).
Saying that the ability to entertain ideas about supernatural
agents “might have led” our ancestors to accept the moral
norms of their community raises more questions than it solves
(see Boyer 2004; MacIntyre 2004; Pyysia¨inen 2006). As I see it,
Bering’s ingenious experimental work is best interpreted within
the standard model. Developing evolutionary arguments about
morality and cooperation would require a more rigorous
methodology.
An unconstrained mind: Explaining belief
in the afterlife
Philip Robbinsa
and Anthony I. Jackb
a
Department of Philosophy, Philosophy-Neuroscience-Psychology Program,
Washington University, St. Louis, MO 63130; b
Department of Neurology,
School of Medicine, Washington University, St. Louis, MO 63110.
probbins@wustl.edu http://artsci.wustl.edu/probbins/home
ajack@npg.wustl.edu
http://www.nil.wstl.edu/labs/corbetta/personnel/ajack.html
Abstract: Bering contends that belief in the afterlife is explained by the
simulation constraint hypothesis: the claim that we cannot imagine what
it is like to be dead. This explanation suffers from some difficulties. First,
it implies the existence of a corresponding belief in the “beforelife.”
Second, a simpler explanation will suffice. Rather than appeal to
constraints on our thoughts about death, we suggest that belief in the
afterlife can be better explained by the lack of such constraints.
Why do so many people believe in life after death? According to
Bering’s simulation constraint hypothesis, the common-sense
belief in the afterlife originates in part from the difficulty of mentally
simulating a scenario in which one does not exist. Such an
act of simulation is difficult because it invites one to imagine
what it would be like (in the technical sense of “what it is like,”
from Nagel 1974) not to exist – an impossible feat. Because
one cannot imagine what it would be like to be dead by imagining
what it would be like not to exist, one imagines instead what it
would be like to exist apart from one’s body. This leads to the
belief in psychological persistence after physical death.
If the simulation constraint hypothesis is correct, then, we
believe in life after death because we cannot imagine ceasing
to exist after death. But it is no easier to imagine one’s nonexistence
after death than it is to imagine one’s nonexistence
before the start of bodily life (say, before conception). What
the simulation constraint hypothesis predicts, then, is belief
in a stronger form of psychological immortality than the one
Bering considers, namely, belief in the afterlife coupled with
belief in the “beforelife.” The question is: Do these two beliefs
have equal currency?
This is an empirical question, and answering it will require
careful experimental investigation, using methods similar to
those described in the target article. This investigation is
ongoing (Bering, personal communication). But there is reason
to suspect that the answer to the question is no. There appears
to be a far greater prevalence of belief in the persistence of the
soul after bodily death relative to belief in the preexistence of
the soul before bodily life.
In historical and contemporary Christianity, for example, the
dominant views of the soul are creationism and traducianism.
According to these views, the soul originates at the moment of
conception, either by a creative act of God (as in creationism)
or by generation from the souls of one’s parents (as in
traducianism). Thus, both creationists and traducianists deny
that the soul exists prior to conception. Though some Christian
sects, such as Mormonism, insist on the preexistence of the
soul relative to conception, this is very much a minority view in
Christian theology, both past and present. The same is true of
Judaism and Islam. By contrast, belief in the continuing existence
of the soul after death is shared by most major world religions,
including Christianity, Judaism, and Islam. While this discrepancy
might be explained in other ways, it suggests that belief
in the afterlife and belief in the beforelife are not equally at
home in common sense. If that turns out to be true, the simulation
constraint hypothesis is in trouble.
Fortunately, we can explain the common-sense belief in life
after death just as well without appealing to the idea of
constraints on simulation. Instead, we can appeal to the relative
absence of such constraints. In the target article, Bering
himself gestures towards the explanation we have in mind:
“[T]he nature of the body’s role in producing the subjective
experiences of emotions, desires, and beliefs seems not as
amenable to children’s scientific theories of dead minds
(or, indeed, even to adults’ formulation of scientific theories
regarding phenomenal consciousness and the brain, e.g.,
qualia)” (sect. 2.1, para. 6).
Here’s how the story goes in a bit more detail. Children and
adults alike tend to think of the mind in general – and phenomenal
consciousness in particular – in terms that are sharply distinct
from, and even cognitively opposed to, the terms in which
they think of physical-causal mechanisms such as the brain and
other parts of the body (Robbins & Jack 2006; see also Bloom
2004). Those links between the mental and physical that we do
understand to exist are few and far between, and they apply
only to a limited range of mental states (Bering & Bjorklund
2004). Hence, our thoughts about the physical do little to
constrain our thoughts about the mental.
In particular, the thought that our body will give out at some
point in the future does not lead us to think that our mind will
do the same. Indeed, our cognitive architecture may dispose us
to reach the opposite conclusion. The very fact that the mental
and the physical seem so starkly different may lead us to
believe in mind–body dualism. It’s a short step from belief in
dualism to the belief that at least some psychological states,
such as conscious thoughts and feelings (which seem terminally
resistant to causal-mechanical explanation; see Robbins & Jack
2006), will persist after death.
Evolution’s lost souls
Lloyd E. Sandelands
Department of Psychology and Ross School of Business, University
of Michigan, Ann Arbor, MI 48109.
lsandel@umich.edu http://www.bus.umich.edu/FacultyBios/
FacultyBio.asp?id¼000119726
Abstract: The target article speaks loudest about what it cannot
see – that man exists in God. Its claim that supernatural beliefs are
“evolved errors” rests on unwarranted assumption and mistaken
argument. Implications for evolutionary study are considered.
Either man is the child of God, or he is not. If he is divine, he is
soulful and immortal, he transcends nature (is supernatural) and
transcends psychology. If he is not divine, he is soulless and
mortal, he is part of nature and his beliefs about soul and afterlife
are errors and illusions of folk psychology.
The target article is clear about the facts. Infants are commonsense
dualists who do not see people in the same way they see
material objects. Children hold natural beliefs in an afterlife.
Children reason better about the omniscient mind of God than
about the limited minds of people and animals. Undergraduates
Commentary/Bering: The folk psychology of souls
484 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
are spooked into moral behavior by the suggestion of supernatural
presence. And, generally, people are “intuitive theists”
who find teleology in the natural world and who find intelligent
design in their personal lives.
The target article is clear also about the question raised by the
facts:
[I]f God does not exist, then the unique self (i.e., the individual “soul”
of any given person) cannot be the product of intelligent design; rather,
it is simply the end product of standard machinations of genetic and
environmental recombination. [. . .] The task remains for cognitive
scientists to determine why the teleological position is so frequently
adopted, and prospers so vehemently, over the mechanistic alternative.
The human mind cannot seem to easily accommodate itself to a
godless, evolutionary canon when it comes to the self’s existence.
(target article, sect. 3)
And the target article is clear about its reckoning of this
question. Human thinking suffers three evolved errors: Type I
errors in reasoning about the afterlife, teleo-functional errors
leading to belief in the soul’s intelligent design, and theory of
mind errors leading to belief in supernatural causes. These
errors culminate in an illusion that the self is related to a supernatural
creator and has a moral obligation to that creator. And
these evolved errors and illusions explain why the “teleological
position” prospers and thereby why the folk psychology of souls
takes the form it does.
Although clear, the target article presumes everything by its
question. On the premises that God does not exist and the soul
is not an intelligent design, the article asks how evolution produced
appearances to the contrary. These premises are not statements
of scientific fact, but are statements of secular faith. Why
these premises? What is the evidence and argument for them?
And what is the cost in accepting them? More to the point,
what if God does exist and the soul is an intelligent design?
How dramatically the question changes – from being about
error and illusion to being about Natural Law, from being
about how people prosper in spite of themselves to being about
how they prosper in the light and love of God. What the target
article describes as the naı¨ve misapprehension of the child
becomes the wisdom of the child.
From its unjustified premises, the target article comes to a
false conclusion. Evolutionary science cannot explain why man
thinks what he does. Evolutionary science is about matter, not
about meaning. It explains the origin, proliferation, and disappearance
of material forms (e.g., physical traits, abilities, dispositions),
but not of mental forms (e.g., ideas, beliefs, feelings) (see
Atran 2002). The latter are not material survivals, but are immaterial
creations. This is plain in the article’s own ideas about the
“errors” of religious thinking. Are these ideas also evolved? And if
so, how are they selected alongside the ideas they rule out?
Furthermore, evolutionary science is about natural objects, not
about man. It misunderstands man when it traces his every
feature, including those that are unique, to his physical being
(e.g., if man is the only animal that can think creatively using
symbols, it is because he has an enlarged brain with a prominent
cerebrum). By confining man to his physical capacities, evolutionary
science fails to account for his unique and creative use
of these capacities. It leaves to question why he uses his
capacities as he does and why animals that share many of these
capacities (e.g., the chimpanzee) do not use them the same
way. Why does man reason? Why does man make art? Why
does man seek life in God? In a word, evolutionary science
fails to explain man qua man.
This, finally and ironically, is the loud unspoken message of the
article:
[J]ust as we can close our eyes to prevent light from being converted
into neural signals, science may provide a minority (i.e., nonbelievers)
with the armamentarium to close their eyes to the supernatural.
(sect. 3, para. 9)
To see man with eyes open – to see him as he is – is to
conceive him differently. It is to see him, not in the blinkered
metaphysic of evolutionary science, but in the revealing
metaphysic of Divine creation (Sandelands 2005). It is to see
him, not as an object of the natural world, but as a creative
being in God. It is to see him as a supernatural being, a
genuine soul. Only when seen in this second way – a way welldescribed
by the world’s great faiths – can satisfactory answers
be found to the main questions about his life: What is a
person? What is society? What is freedom? What does man
want? And, what is the good? A full and true study of man
must begin in God.
Reasoning about dead agents: A
cross-cultural perspective
Harvey Whitehouse
School of Anthropology, University of Oxford, Oxford OX2 6PE,
United Kingdom.
http://www.isca.ox.ac.uk/ harvey.whitehouse@anthro.ox.ac.uk
Abstract: Jesse Bering’s theory of the “folk psychology of souls” is
brilliantly elucidated and warrants further empirical investigation.
While additional experimental research is certainly required, we also
need to interrogate the evidence on reasoning about dead agents in a
wide range of real-world settings.
Bering’s “folk psychology of souls” comprises six main hypotheses
that warrant further investigation in light of evidence from comparative
ethnography.
The Simulation Constraint Hypothesis proposes that an inclination
to attribute consciousness to the dead arises from an
inability to simulate the absence of consciousness (whereas
we have less difficulty simulating the absence of perceptual
capacities such as seeing and hearing). Although this hypothesis
may be correct, it is not obviously so. Loss of consciousness (e.g.,
when sleeping) is a routine feature of human experience and
apparently quite easy to conceptualize, if not to simulate. It is
noteworthy that adults, in our own society at least, sometimes
imagine that children will be comforted by the idea that
grandma is sleeping rather than dead, not because an absence
of consciousness is hard to convey (quite the opposite), but
because a permanent absence of consciousness is alarming.
Further, an inclination to attribute consciousness to the dead
would, in itself, be inconsequential in behavioural (and thus
evolutionary) terms. Dead agents must additionally be attributed
capacities that enable them to intervene in human affairs: at least
minimally, they must not only be conscious but also have access
to socially strategic information (e.g., being able to hear and see
what mortals are saying and doing). The ethnographic record
furnishes abundant examples of representations of dead agents
whose perceptual capacities are preserved along with conscious
beliefs and desires, whereas representations of dead agents as
deaf or blind are rare or nonexistent. Thus, even if the simulation
constraint hypothesis proves to be correct, its impact on behaviour
must in turn depend on the activation of additional cognitive
tools (e.g., as proposed in Barrett 2004 and Boyer 2001).
The Mental Continuity Hypothesis proposes that attributions
of emotions, desires, and epistemic states to dead agents
(“mental continuity judgements”) result from universal cognitive
biases emerging early in childhood. This hypothesis seems to be
supported by evidence that among American children such attributions
decrease with age, “the opposite pattern that one would
expect to find if the origins of such beliefs could be traced exclusively
to cultural indoctrination” (target article, sect. 2, para. 3). If
this finding proves to be robust then, to rule out “cultural indoctrination”
and other effects of variable environments, we would
need to test the hypothesis across a range of cultural settings
Commentary/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 485
(a valuable start has been made by Harris & Gime´nez 2005 and
Astuti, forthcoming a).
The Absent Presence Hypothesis proposes that our habit of
thinking about significant others in absentia reinforces the
illusion that they are still around after they have died. Such
reinforcement may vary cross-culturally, however. In modern
complex societies, close kin often live far apart, relying heavily
on disembodied forms of communication to maintain contact.
Bering’s example of bereaved persons abortively telephoning
their loved ones may best be understood in that context. In
small-scale societies, including those in which our ancestors
evolved, physical separation from significant others is/was more
limited in duration and the reality of their absence after death
perhaps less easy (momentarily) to forget. This is a potentially
testable hypothesis. Are accidental communication attempts
with the dead less frequent in small, face-to-face societies with
low rates of out-migration than in modern urban settings?
The Good Citizen Hypothesis proposes that the presumed
presence of supernatural agents encourages prosocial behaviour
and inhibits antisocial behaviour. The idea that supernatural
agents reward good behaviour and punish the wicked is a recurrent
theme in the so-called ethical religions, but generalizes only
with considerable difficulty. In many Melanesian societies, for
instance, the ancestors are thought to incite homicidal behaviour
by systematically undermining more natural tendencies towards
peaceful cooperation and prosociality (Harrison 1993). A
broader question might be whether prosocial behaviour
enhances reproductive success in human populations, past and
present, as Bering’s evolutionary scheme proposes. In many
societies, the cultivation of a reputation for acquisitive and
aggressive behaviour, at least among males, appears to be an
effective strategy in accessing resources (including women).
The Purposeful Life Hypothesis proposes that people are cognitively
predisposed to think that they were called into existence
to serve a divine plan. This hypothesis, and the related contention
that suicide is an affront to God’s scheme, may prove difficult to
generalize cross-culturally, perhaps reflecting attitudes found
mainly in highly individualistic cultural traditions of modern
origin.
The Meaningful World Hypothesis proposes that people are
cognitively predisposed to interpret happenings as communicatively
driven. A readiness to interpret unusual occurrences as
signs or portents initiated by supernatural agents, rather than
as merely random and meaningless events, has been widely
reported by ethnographers. It is not clear why such a predisposition
should be adaptive, although it could be a spandrel, a
product of sign-reading capacities that evolved as an aid to tracking
quarry, detecting predators, and grasping the communicative
intentions of other humans (among other useful applications).
From an anthropological viewpoint, Bering’s arguments are
highly significant and warrant careful attention in future crosscultural
research.
Author’s Response
The cognitive science of souls: Clarifications
and extensions of the evolutionary model
Jesse M. Bering
Institute of Cognition and Culture, Queen’s University Belfast,
Belfast BT7 1NN, United Kingdom.
j.bering@qub.ac.uk qub.ac.uk/icc
http://www.qub.ac.uk/schools/InstituteofCognitionCulture/Staff/
JesseMBering/
Abstract: The commentaries are a promising sign that a research
programme on the cognitive science of souls will continue to
move toward empirical and theoretical rigor. Most of the
commentators agree that beliefs in personal immortality, in the
intelligent design of souls, and in the symbolic meaning of
natural events can provide new insight into human social
evolution. In this response I clarify and extend the evolutionary
model, further emphasizing the adaptiveness of the cognitive
system that underlies these beliefs.
R1. Introduction
Only in the past few years has the cognitive substratum of
religious and supernatural experience been penetrated by
the precision tools of experimental science (for reviews,
see Atran & Norenzayan 2004; Barrett 2000; Boyer
2001; Whitehouse 2004). Shaky hands are to be expected
in these first attempts at scientific exploration, and some
trembling is also evident at places in the target article.
But, as Bainbridge so eloquently discusses in his commentary,
picking up these tools to begin with is the hard
part, and I am very grateful to the commentators for offering
their expert hands to steady my own.
Since the focus of most of the commentaries concerned
the evolutionary model and because this issue is central to
the theoretical framing of the target article, most of my
response will be devoted to that topic, highlighting theoretical
points in the other commentaries whenever they
have some bearing on the evolutionary history of a folk
psychology of souls.
R2. Clarifying the evolutionary model
By far the most frequent criticism in the commentaries
questioned my evolutionary model, seeking greater
clarity and further attention to deviant cross-cultural
examples of religious belief systems that seemingly
violate the model’s core assumptions. For example,
Johnson & Nyhof protest that “the Darwinian mechanisms
are left completely unspecified.” Pyysia¨inen finds
similar fault with the target article’s evolutionary cast
(“The weakest part of Bering’s contribution is precisely
his evolutionary speculations”), a concern echoed in the
vituperative commentary by Hegde´ & Johnson, among
others. According to the majority of commentators who
addressed the evolutionary issues, the explanatory framework
into which I have placed my research programme
is flawed on several counts.
To recount the original argument, I claimed that three
basic cognitive mechanisms – those that produce illusions
of personal immortality, of teleological authorship in
the design of individual souls, and of natural events
as having symbolic meaning – formed an organized
“system” at some point in recent human evolution as a
result of the unique selective pressures operating in our
social environment. I placed the word “system” in scare
quotes because I see these illusions as being connected
through a sort of abstract, conjunctive tissue that biases
reasoning about personal existence, not as a modular,
task-specific system yielding static behaviours independent
of cultural variation (Livingston; Pyysia¨inen). I
have never claimed that “religion is innate” (Bloch) or
even that there exists “an evolved system dedicated to
Response/Bering: The folk psychology of souls
486 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
afterlife beliefs” (Harris & Astuti; also Pyysia¨inen). These
are oversimplifications.
My evolutionary model identifies a suite of very basic
cognitive building blocks, often but not always associated
with religious beliefs. If this suite indeed comprises a
true psychological adaptation that motivated adaptive
responses under recurrent challenging conditions, then
it should be canalized in modern humans, present in
anyone with a normal cognitive profile (Flusberg &
Tager-Flusberg) who does not develop under extreme
and species-atypical conditions (Evans & Wellman).
Hegde´ & Johnson conflate adaptations, which have
a heritability of zero, with heritable individual
differences associated with adaptations (see Tooby &
Cosmides 1992, pp. 122–31), presumably those that
would contribute to varying degrees of religiosity in
modern humans.
R2.1. The fact that there may be reputation maintenance
mechanisms other than those entailed by the folk
psychology of souls presents no difficulties for my
evolutionary model
Several commentaries suggested that my evolutionary
model was flawed because there are other psychological
mechanisms, none of which involve souls or supernatural
agents or seeing signs in natural events, which serve
a reputation maintenance function (Beit-Hallahmi;
Boyer; Ferrari; Gjersoe & Hood; Greenberg,
Sullivan, Kosloff, & Solomon [Greenberg et al.]).
This is a point elaborated by Boyer, who, after
summarizing several alternative mechanisms that
promote prosocial behaviours in humans, tells us that,
“All these dispositions and processes evolved independently
of supernatural and religious beliefs, operate in
similar ways in people with or without such beliefs and
regardless of differences in these beliefs, and recruit
different neuro-cognitive machinery from the supernatural
imagination.”
I do not see a conflict here. As far as I am aware, there is
no law in natural selection theory stating that there cannot
be distinctly evolved mechanisms serving the same
adaptive purpose. On the contrary, if these mechanisms
delivered a cumulative, buffering effect in solving a
shared adaptive problem or – at the very least – did not
impede one another’s functioning, evolution should
favour the selection of multiple adaptive designs. Moreover,
contrary to the criticisms raised by Hegde´ &
Johnson, an evolved folk psychology of souls meets
the important criterion of Darwinian conservativeness
because the types of existential illusions generated would
have emerged through a set of biases produced by
pre-existing structures. The system that I have outlined
therefore would not have required any substantive
neuro-cognitive reorganization.
Several commentaries discussed the role of shame in
inhibiting normatively deviant or antisocial behaviours
(Beit-Hallahmi; Ferrari; Gjersoe & Hood), particularly
how parents instill these feelings, and implied that this
obviates the adaptive utility of belief in supernatural
observation or punishment. Shame, however, is usually
experienced after a social transgression has already
occurred; it is the emotional aftermath of transgression
(Tangney 2003). Although the negative affect associated
with this experience may serve to discourage similar
actions in the future (see Fessler & Haley 2003;
Gilbert & McGuire 1998), and may attenuate severity of
punishment for an offence (Gold & Weiner 2000),
shame may not be very effective at preventing the
occurrence of a proscribed behaviour in the first place.
Shame and observability, of course, are not mutually
exclusive – in fact they are sister constructs. But
belief that one is under surveillance by supernatural
agents, and that there are consequences for misdeeds
even when they occur in private, may effectively deter
socially proscribed behaviours even in the absence of
shame.
R2.2. The importance of avoiding solitary incidents of
serious transgressions must be emphasized
Although reputations are mostly cumulative and can
perhaps be formulated as an image score that people
use to guide their interactions with social others (e.g.,
Nowak & Sigmund 1998), a single black mark can erode
an otherwise unbroken record of altruistic tendencies.
“Words are wolves,” according to Jean Genet. Language
would have enabled our ancestors to essentialize others
into social category memberships through the heuristics
of emotionally loaded words. One need only consider
how socially powerful are terms such as “rapist,” “paedophile,”
“thief,” “murderer,” “slut,” “racist,” “child abuser,”
or, recently, “terrorist,” to see the hazards of a publicly
revealed, solitary moral breach from the gene’s point
of view.1
Nemoroff and Rozin’s (1994) findings of moral contagion
(i.e., emotional aversion to physical objects such as
clothing that have been in contact with representatives
of such derogated social categories) may be seen as evidence
of this type of negative essentialism (Gjersoe &
Hood). Any public distancing from socially repudiated
others would serve to advertise a personal commitment
to in-group norms – that is, that one is not like the derogated
individual. The sociologist Erving Goffman (1963)
noted that people who are wanted on criminal warrants
were once referred to as “having smallpox” and their criminal
disease was said to be catching. Merely being seen
with them could lead to arrest on suspicion.
The folk psychology of souls (which at its core constitutes
a social relationship between the self and supernatural
agents) would have helped our ancestors to censor selfish
decisions associated with others judging them as being
essentially bad and/or morally undesirable (and therefore
to avoid the negative reproductive consequences of this
labelling). This was particularly the case in situations
where people were strongly tempted by selfish desires
and underestimated the likelihood of detection by other
in-group members. Miscalculating the odds of social
exposure for certain behaviours would have had calamitous
effects on reputation and, therefore, on genetic
fitness. The folk psychology of souls provided adaptive illusions
of watchful supernatural agents that helped to counteract
these dangerous miscalculations. These illusions
involved seeing supernatural agent(s) as being emotionally
invested in the self’s existence, as sharing (or at least
understanding) the in-group’s moral values, and as communicating
their attitudes and opinions about the self
through the occurrence of natural events and biographical
Response/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 487
experiences (Bering 2002b; Bering & Johnson 2005;
Johnson & Bering 2006). Natural events and biographical
experiences were perceived as the “evidence” that such
supernatural agents were real (cf. Bullot) and were
capable of punishing and rewarding social behaviours,
either in this life or in the hereafter.
R2.3. Experimental findings demonstrate social
sensitivity to being observed
Beliefs in watchful supernatural agents appear to militate
against the psychological state of deindividuation, which
occurs whenever “individuals are not seen or paid attention
to as individuals” (Festinger et al. 1952, p. 382).
Festinger and his colleagues described how deindividuation
is strongly associated with social disinhibition and
loss of inner restraints. Numerous laboratory experiments
have in fact shown that participants who believe that they
are making decisions under anonymous conditions tend to
be less altruistic, more aggressive, and more punitive than
those who believe that their identities are known (e.g.,
Diener et al. 1976; Ellison et al. 1995; Rehm et al. 1987;
Zimbardo 1969).
Building on experimental economic games, a flurry of
recent studies have also provided evidence that ambient
gaze, even when artificial, unconsciously primes prosocial
behaviours in human participants (Bateson et al. 2006;
Burnham & Hare, in press; Haley & Fessler 2005; Milinski
et al. 2002; Wedekind & Braithwaite 2002). Burnham and
Hare (in press), for example, found that people made more
altruistic decisions in a task involving allocation of scarce
resources even when the “witness” was simply an image
of a robot with large human-like eyes. Similar results
were reported by Haley and Fessler (2005), from a study
in which participants behaved more generously on a computerized
task when stylized eyespots were present on the
screen, which the authors interpreted as evidence that
subtle cues concerning observability factor prominently
in reputation management. (Gjersoe & Hood’s discussion
of Titchener’s classic unseen gaze findings, where people
believe they can “feel” when others are looking at them
behind their backs, may be interpretable within this evolutionary
framework, as well; see also Colwell et al. 2000).
Finally, as discussed in the target article, Bering et al.
(2005b) found that, when left alone in a room, participants
who were led to believe that a ghost may be observing
them cheated less on a competitive task compared to
those who did not receive this supernatural prime.
Real-world findings provide complementary evidence
that perceptions of anonymity are positively correlated
with antisocial behaviours. In a cross-cultural analysis of
warfare practices, for example, Watson (1973) discovered
that warriors who hid their identities before going into
battle were more likely to kill, mutilate, and torture than
those who did not. More recently, Silke (2003) found
that, of all sectarian violence incidents reported in Northern
Ireland over a two-year period (1994–1996), paramilitary
members who wore masks during their offences
attacked more people, inflicted more serious injuries,
committed more acts of vandalism, and were more likely
to threaten their victims after attacking them than paramilitary
members who were implicated in sectarian violence
but who did not hide their faces.
R2.4. Cross-cultural variability, supernatural beliefs,
and evolutionary dynamics
Although belief in supernatural observability has not yet
been targeted as a key research question in evolutionary
models of religion, the ethnographic literature does
suggest that such beliefs feature prominently in most
religious systems. In Pettazzoni’s (1955) cross-cultural
analysis of the types of attributes that are most frequently
attributed to the gods, one recurrent and defining characteristic
is the gods’ deep knowing of people as unique
individuals (i.e., their “hearts and souls”). In Borneo,
the Iban believe that “anyone who successfully cheats
another, or escapes punishment for his crimes, even
though he may appear to profit temporarily, ultimately
suffers supernatural retribution” (Sandin & Sather 1980,
p. xxviii). And Malinowski (1935, p. viii) wrote that “from
the study of past religions, primitive and developed, we
shall gain the conviction . . . that every religion implies
some reward of virtue and the punishment of sin.” Implicit
here is the assumption that supernatural agents who dole
out moralistic consequences are believed also to survey
and observe private behaviours, keeping their thumbs on
individuals within the group.
Cohen, Kenrick, & Li [Cohen et al.] ask whether
“variations in beliefs in afterlife or observant spirits are
linked to recurrent variations in social or physical
ecology” (also Whitehouse; see Reynolds & Tanner
1995). Although we do not yet have the data to answer
this important question, structuring the present evolutionary
model under these (ecologically dynamic) terms may
put into context the striking cultural diversity associated
with the moral dimension of supernatural beliefs. For
example, answering this question would potentially be
capable of addressing the sceptical query posed by
Greenberg et al., who ask, “If immortality beliefs were
a simple default by-product of cognition, why would
these beliefs be so varied across cultures and so
complex?” At the moment, I agree that such variation is
difficult to understand, but this is due to the embryonic
stage of data gathering in this area, not to any serious limitations
of the simulation constraint hypothesis or the
general evolutionary theory I have offered.
The commentaries reveal a wide variety of religious
beliefs that appear uniquely tied to specific cultures, geographic
areas, and historical settings. Why these adaptive
climates give rise to particular beliefs and not others is a
question for evolutionary analysis, just as Cohen et al.
reason. For example, collective symbolic interpretations
of disease and misfortune may serve to enculturate children
into specific moral environments (environments that
are themselves products of specific ecological and social
factors). Such symbolic interpretations offer children a
very clear picture of what it is that their society does not
condone. In their fascinating description of Cotard’s
syndrome, Cohen & Consoli write that “collective and
cultural significance dominates biographical experiences
[. . .] first syphilis then AIDS symbolized the amalgam of
flesh, punishment, sin, guilt, sexuality, and the devil.”
We need not look at exotic cultures to see how collective
symbolic interpretations of natural events can influence
moral development. As an eight-year-old, I was panicstricken
that an upcoming doctor’s visit, which I knew
would involve a routine drawing of blood, would publicly
Response/Bering: The folk psychology of souls
488 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
identify me as a homosexual. I was naı¨ve to the medical
facts about how people contracted the HIV-virus, but I
knew that AIDS could be detected in blood. I also understood
that many saw AIDS as a moral condemnation of
gay men, specifically as God’s culling of homosexuals.
Whether I personally saw such a moralistic message in
AIDS was inconsequential for this cultural illusion to
impact on my decision not to divulge my sexual orientation,
a decision that can be understood within fitnessrelated
terms. Other peoples’ symbolic interpretation of
this disease was enough to teach me that something in
my blood would expose me as being essentially bad,
worthy of being shunned – and, in fact, I had such anticipatory
anxiety about the social consequences of being
labelled a “homosexual,” that I collapsed in the waiting
room.
Although I agree with Pyysia¨inen that we are not in
a position to advance a detailed evolutionary argument
until a “more rigorous methodology” is developed – a
task that will require massive interdisciplinary collaboration
– it is unclear to me how one could ever begin to
construct such a methodology without first having a
general evolutionary theory capable of generating hypotheses
and offering an interpretive lens through which to
view the findings. I have posited a general evolutionary
theory that can act as such, as a crucible for weighing competing,
non-adaptationist hypotheses, something recognized
by several commentators (Cohen et al.; Evans &
Wellman; Hughes; Whitehouse).
R2.5. Absent third-party punishment is a uniquely
human adaptive problem
Gjersoe & Hood comment that “many social animals also
show behavioural inhibition and prosocial behaviour
without necessitating a specialized cognitive mechanism
for a belief in souls.” This is not in debate. But what
these commentators overlook is the fact that theory of
mind, and the concomitant emergence of declarative
language, introduced a genuinely novel adaptive problem
in human sociality – that of absent third-party punishment.
In short, absent third-party punishment is any
punishment that is administered by a person (or persons)
who were not present at the time of the offence, but who
learned about the offence through a second-hand source
(Fehr & Fischbacher 2004). Human beings are able to
mentally represent an absent third-party’s state of ignorance
about the unobserved event and are strategically
motivated – and emotionally driven – to disclose their
victimization to these naı¨ve third parties through declarative
language. This is, in fact, the very basis of all criminal
justice systems, no matter how informal.
What is unique about human sociality is that anybody
who witnesses a social event is a carrier of strategic information
who can then transmit that information to other
minds, over great spans of time and geographical distances.
“Seeing” therefore took on new meaning for
human beings, the only species for which, given these
social cognitive verities, short-term selfish gains were
traded in for long-term reputation gains. According to
Johnson (2005, p. 414), “Information about person A
could propagate via person B to person C, D, E, and so
on . . . even if person B and C do not care, it may not be
until person Z hears the news, or until enough people
hear the news, or until some authority hears the news,
perhaps weeks later, that punishment will come.” Given
the calamitous effects a mired reputation could have on
the actor’s genetic fitness (through punitive tactics such
as castigation, ostracism, exclusion, group expulsion, or
even execution), the presence of nearly any watchful
agent, human or supernatural, became capable of influencing
behavioural decision-making.2
R2.6. Selective pressures for solutions to the adaptive
problem were intense
It is impossible to overstate how strongly the third-party
punishment problem would have influenced the course
of human social evolution. This is especially evident
when one considers the relatively low degree of privacy
afforded to our ancestors, who lived in small-scale
gossipy societies of only 120 to 150 individuals (Dunbar
& Spoors 1995). In the environment of evolutionary adaptedness,
individuals would have been unable to easily emigrate
to new social groups and to “start over” if they spoilt
their reputation in their natal group (a strategy of sociopaths
in modern societies; Mealey 1995).3
Notice that inclusive fitness is also likely to be negatively
impacted by a spoiled public identity because of a
sanguineous bias, stigma attached to the biological kin of
the individual whose reputation is impugned through
transgression (e.g., see May 2000 for stigma effects on
murderers’ relatives). This means that third-party punishment
does not necessarily end at death. These inclusive
fitness issues concerning the effects of reputation on biological
kin also mean, in principle, that effectively managing
reputation is a more pressing evolutionary problem
than mitigating existential anxieties through symbolic
immortality (Greenberg et al.). Indeed, many of the
extensive findings from the Terror Management Theory
literature can be understood in these terms. If one is
reminded of his own inevitable death, better for his
family members’ genes that he go out as a staunch, reliable
defender of his community’s values than as apathetic or as
a social dissident.
R3. Propositional beliefs about the supernatural
do not always cause behaviour (and sometimes
they are in opposition to behaviour)
It is important to understand that the three existential illusions
identified in the target article (immortality, teleological
authorship of the soul, and symbolic natural events)
may not be as salient in industrialized societies today as
they were in the environment of evolutionary adaptedness,
where they were unlikely to be punctured by scientific
knowledge or discouraged through cultural secularization.
Even in modern scientific nations, however, among welleducated
and scientifically literate people, the biases
identified in the target article are not recognized as illusions
and continue to have deep emotional resonance.
Sandelands, for example, concludes his theologically
inspired commentary by stating that, “a full and true
study of man must begin in God.” (In some sense this is
true: Our species like any other must be understood
within the parameters of the modern synthesis and God
is just another slave to human genes.) And, moreover,
Response/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 489
even in recognizing them as illusions we fail to sever their
emotional underlay, which may still pump-prime behaviour
– the level at which natural selection operates. I
do not believe in the afterlife, but as a potential homeowner
I certainly would feel uncomfortable living in a
house where a stranger has recently died. In this case, it
is my eerie feelings and not my belief or disbelief in the
afterlife that would be a better predictor of whether I
make an offer on the house. This is not to say that propositional
beliefs about religion and the supernatural are
frequently epiphenomenal, but rather they are more properly
viewed as rough indices of unconscious reasoning
(and perhaps phenomenal states) than as accurate predictors
of behaviour.
R3.1. Global secularization cannot extirpate a true
psychological adaptation
Beit-Hallahmi writes that, “The global secularization
process means that we no longer interpret misfortune
as caused by supernatural agents.” But this “god-of-thegaps”
hypothesis has now been disconfirmed in social
psychology experiments (see Weeks & Lupfer [2000]
for an account of distal-proximal attributions to God).
Moreover, the argument that scientific or secular explanations
“replace” more naı¨ve or irrational supernatural
explanations is intuitively unpersuasive; obviously they
can occur alongside one another (e.g., Subbotsky
2001). Theologians who saw the recent tsunamis of
East Asia as an angry, moralistic message from God
were probably not naı¨ve to the fact that they were
caused by earthquakes on the Indian Ocean floor. No
matter how culturally secularized we become, God
pokes through, whispering in the most godless of scientists’
ears. At the end of their commentary, Gjersoe &
Hood correctly point out that in order to understand
supernatural beliefs from a scientific perspective we
must first acknowledge and recognize our own supernatural
dispositions.
The best research designs in the cognitive science of
religion are those that are able to pry apart unconscious
reasoning from explicit or “theologically correct” religious
beliefs (see Barrett 2000). Socrates’ “idea of
immortality” as described by Ferrari is therefore of
questionable countenance, since this “reasoned conclusion”
would be heavily influenced by the same underlying
cognitive constraints that motivate others to think
in this fashion. Innate psychological biases with regard
to the supernatural (and the behaviours they generate)
reveal themselves most clearly when they directly contradict
stated beliefs. For example, those who believe
God can do everything at once actually reason as if
God were constrained by a human attention span
(Barrett & Keil 1996); some people who believe that
the mind stops at death nevertheless reason about a
dead person as if he still has thoughts (Bering 2002a);
scientific theorizers are wary of magical incantations
(Subbotsky 2001); others who consider themselves to
be materialists refuse to sign a contract relinquishing
their souls at death to an experimenter (Haidt et al.
2004); and children who say they don’t believe in monsters
shy away from a box they are told contains a
monster (Harris et al. 1991).
Similarly, if the folk psychology of souls is a true
psychological adaptation, then it should be empirically
detectable, even in atheists. For example, McAdam’s
(2001) findings from narrative psychology suggest that
people tend to fall into one of two categories: those
who view personal misfortunes as contaminative episodes
in their life stories (where the event permanently
disrupted an otherwise positive life course and cast a
dark shadow over their biographies), and those who
view such events as redemptive episodes (where the
event, although difficult at the time, was responsible
for a positive redirection of their life course). It may
be possible to detect intentionality themes in atheists’
self-narratives through the use of such paradigms
(e.g., “it was a ‘life lesson,’ ” “it wasn’t supposed to
happen,” and so on).
R4. Developmental considerations
Evans & Wellman argue that, “if Bering’s selectionist
explanation was on target then one might predict a
unique and relatively robust developmental trajectory,
regardless of input.” This is certainly true, and I believe
that this trajectory will be borne out. To test Evans &
Wellman’s prediction, we need, first, to have an accurate
developmental model that delineates the ages at which
the three existential illusions (immortality, teleological
authorship of the soul, and symbolic natural events)
appear in childhood. We do not yet have enough data to
construct such a model and therefore developmental
research in this area is urgently needed. Although
Ferrari and Estes are right to point out that cognitive
developmentalists have for decades been exploring
related questions about children’s distinction between
the mind and body, particularly in the area of theory of
mind, this “abundant research” (Estes) hardly constitutes
a targeted attempt at systematically revealing the social
cognitive factors that lend themselves so seamlessly to
the existential illusions highlighted in the article. On the
contrary, such a targeted research programme is strikingly
absent, not only in developmental psychology, but in all
the subdisciplines of experimental psychology.
R4.1. Contradictory findings on the development
of children’s afterlife beliefs
It appears that the little we do know about the development
of a folk psychology of souls is contradictory, as
discussed in the commentaries by Evans & Wellman
and Harris & Astuti. These commentators tell us that
recent findings on children’s afterlife beliefs have failed
to replicate the pattern reported by Bering and Bjorklund
(2004; also Bering et al. 2005a.) For example, Harris and
Gimenez’s (2005) findings suggest that afterlife beliefs
increase with age rather than decrease and are moderated
by the religious context of the experimenters’ questions
(with children being more likely to endorse psychological
functioning after death when information about the dead
character includes words like “priest” and “God”). Therefore,
Harris & Astuti question whether belief in the
afterlife is in fact a default cognitive stance.
Response/Bering: The folk psychology of souls
490 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
R4.2. Some “contradictory” findings may
not be contradictory
This conflicting pattern of developmental findings,
however, is difficult to interpret at present. To begin
with, the central research questions motivating these
other studies on children’s concepts of death are very
different from my own (as well as from each other) and
the methodologies vary accordingly. Evans & Wellman
cite work by Barrett and Behne (2005) as evidence that,
in contrast to my findings, four- and five-year-olds in this
study did not attribute psychological states to dead
agents. Barrett and Behne’s study, however, did not investigate
children’s afterlife beliefs, but instead concerned
children’s ability to differentiate between dead and sleeping
animals in the physical environment. The investigators
reasoned that this is an adaptive function in that it prevents
unnecessary vigilance toward the bodies of dead animals
through the cue-driven activation of an innate “living/
dead remapping mechanism.”
In the study by Barrett and Behne (2005), children were
asked five questions about the dead versus sleeping
animal: Can it move? Know you were there? Move if
touched? Can it be afraid? Can it hurt you? The fact
that the youngest children answered “no” in reference to
the dead animal, but “yes” in reference to the sleeping
animal, is hardly prima facie evidence against my argument
that belief in the afterlife is a cognitive default. In
fact, if belief in the afterlife is a cognitive default, then
we would actually predict the pattern of findings reported
by Barrett and Behne (2005). That is, preschoolers is
should answer “no” to questions about the bodies of
dead animals (notice the key word “it” in the questions
posed to children) if indeed they view the mind as being
liberated from the body at death.
R4.3. Methodological concerns presently limit
theoretical inferences
Like Evans & Wellman, Harris & Astuti state that their
own research programme on the development of afterlife
beliefs reveals a set of findings that in many ways contradicts
the developmental trajectory reported by Bering
and Bjorklund (2004), or at least tells a more complicated
story, with religious testimony and cultural exposure
encouraging such beliefs. Again, however, it is difficult to
compare findings across these studies. We deliberately
avoided eschatological language in our research design
because we were wary of biasing children’s answers
through the experimenters’ language and behaviours,
and in fact our empirical reports list many of the safeguards
we used to protect against such biases (Hughes).
In contrast, such language was an important manipulated
variable for both Harris and Gime´nez (2005) and Astuti
and Harris (submitted).
Furthermore, the coding procedures used to determine
whether children attributed continued psychological functioning
to a dead agent meaningfully differed between our
studies and those described by Harris & Astuti. Our data
were coded on the basis of children’s follow-up answers to
the questions rather than their initial yes or no response.
We reasoned that a “no” response is inherently ambiguous
and should not be seen as clear evidence for non-continuity
judgements after death. Young children in our study often
answered “no” to the initial questions about the dead
agent’s continued capacities (“Can Brown Mouse still
see?”), but upon further questioning it became clear that
they were nevertheless reasoning in terms of an afterlife
(e.g., “because it’s too dark in the alligator’s belly”).
Harris and Gime´nez (as well as Astuti & Harris [submitted]
and Barrett & Behne [2005]) failed to operationalize
children’s “no” answers in this way; instead they took
them at face value as evidence of an understanding of the
non-functionality of the capacity in question. It is therefore
impossible to know whether the findings these authors
report is a product of the religious context of the story, as
they argue, or is in fact an artefact of their coding procedure.
Finally, the youngest children in the Harris and
Gime´nez study were seven-year-olds, whereas our most
robust findings for afterlife beliefs came from the threeand
four-year-olds we tested, providing the basis of our
nativist claims.
R5. Cognitive processes underlying the folk
psychology of souls
Several of the commentaries focused on the precise mechanisms
by which existential illusions are generated. Bullot,
for example, provides a distinction between two types
of agent-tracking mechanisms that he believes weighs
heavily on the theoretical integrity of an evolved folk psychology
of souls. In perceptual tracking, Bullot reasons, “a
target individual is directly tracked by a sensory-motor
system.” In contrast, epistemic tracking occurs when an
agent “is spatio-temporally pursued by indirect epistemic
means such as communication and reasoning.”
According to Bullot, because both types of tracking
require empirical or material traces of the agents’ ontological
existence, such as behavioural residue (e.g., fingerprints),
the present case of souls, gods, and ghosts poses
an important problem for the evolutionary model. Bullot
reasons that ghosts and gods should therefore be characterized
as fictionally grounded referents, “which rest on
descriptive resources and individual/collective imagination,”
rather than empirically grounded referents. Presumably
this would be evidence of the cultural origins of
supernatural agents. I have argued, however, that our
species has an innate predisposition to see natural events
as actual empirical traces of real supernatural agents. It
follows from this that natural events serve the same function
as tractable social behaviours, activating similar epistemic
tracking mechanisms. An infinite array of life’s
vicissitudes strewn throughout personal biographies is
represented as God’s “behaviours.” It is unclear to me
how Sherlock Holmes, a fictional character that Bullot
compares to gods and ghosts, could leave similar empirical
and perceptible traces that are capable of confirming
peoples’ intuitive hypotheses of his ontological existence.
R5.1. Individuation equals ensoulment
In a related commentary, Newman, Blok, & Rips
[Newman et al.] describe their fascinating research programme
on identity tracking, concluding that my argument
for a specialized cognitive system dedicated to
reasoning about souls lacks parsimony. This conclusion
hinges on a series of studies revealing that participants
Response/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 491
believe in the continued psychological existence of not
only individual people after dramatic transformations
(such as someone who has died or whose memories are
placed in a robotic body), but also the continued identity
of individual objects that have undergone similarly dramatic
transformations (e.g., a specific car, “Rustbucket”
is still identified as “Rustbucket” after going through a
“particle pipeline” and reconfigured into a boat). “Thus,”
argue Newman et al., “belief in the persistence of individuals
through radical changes in kind is not restricted
to persons and need not include the notion of a soul”
(para 8).
I interpret these findings very differently from these
commentators, however, and see them as generally supportive
of my evolutionary model rather than as falsifying
my hypotheses about the folk psychology of souls
(see also Nichols). Newman et al. miss the critical
fact that by individuating the target objects in this
manner (e.g., through proper naming) they may be
doing something akin to ensouling inanimate objects.
Through the experimental individuation of objects, participants
may be reasoning about such objects through
an animistic lens. As a consequence of this, they are
likely to tacitly endow these objects with psychological
states, in effect viewing them as possessing souls.
Bloch’s sardonic comments about the ecological validity
of the mouse puppet show paradigm (i.e., Bering &
Bjorklund 2004; Bering et al. 2005a) similarly dismisses
the animating effects of individuating target “characters”
in an experimental context.
R5.2. The simulation constraint hypothesis
Several commentaries focused on the simulation constraint
hypothesis (Antony; Cohen & Consoli; Robbins
& Jack; Kemmerer & Gupta; Preston, Gray, &
Wegner [Preston et al.]). To revisit the central thesis
of this hypothesis, I claimed that a delimiting phenomenological
boundary prevents people from experiencing the
absence of certain categories of mental states, such as
emotions, desires, and various episteme (the most
“ethereal” qualia). Because we can never know what it
feels like to be without such states, these natural representational
borders encourage afterlife beliefs. When we
attempt to reason about what it will be “like” after death –
and what it is “like” for those who have already died – we
inevitably get ensnared by simulation constraints and
reason in terms of a continued consciousness.
Preston et al. reason that belief in the soul stems in
large part from the illusion of conscious will, the feeling
that the self is a sort of abstract homunculus that consciously
wills the body to act (when in fact this feeling of
authorship of our own actions is epiphenomenal). I
agree that this is an important component of the folk psychology
of souls, but I see it as a refinement to the present
model, rather than an alternative account. Cohen &
Consoli’s description of Cotard’s syndrome as being
characterized by the delusion that one is already dead,
as well as Kemmerer & Gupta’s discussion of the
neurobiological basis of out-of-body-experiences, do
seem to provide at least indirect support for the simulation
constraint hypothesis.
In his thoughtful commentary, Antony reasons that in
order to “run a simulation” of a dead person’s mind, one
must already have a belief in the afterlife, which runs contrary
to this nativist position on the illusion of immortality.
“Prior to simulating a dead agent’s mind,” he argues, “it
must be assumed there is a mind to simulate. But that
already is to assume an afterlife.” He then writes, “It
follows that nothing about a simulation itself can explain
our belief in an afterlife, since some such belief or assumption
is a precondition for the planning and running of any
such simulation.” There is probably some truth to Antony’s
chicken-and-egg reasoning, but it is unclear to me why he
sees this as a problem for my nativist arguments concerning
the origins of afterlife beliefs. If children are confronted
with someone who has died, they need not
“assume an afterlife” – at least in any meaningful conceptual
sense, and certainly not in terms of a propositional
belief about an afterlife – to attempt to reason about the
dead person’s current state of mind through appeal to
their own mental states.
R6. Concluding remarks
The commentaries in response to the target article are all
that a Behavioral and Brain Sciences author could wish
for. They are filled with incisive criticisms, counterarguments,
and references to important work of which I was
unaware. All of these undoubtedly will add to a more
informed cognitive analysis of the subtle strands that
bind together morality, souls, and meaning.
NOTES
1. This is the literary device behind Nathaniel Hawthorne’s
The Scarlet Letter. Hester Prynne, an otherwise virtuous
woman, is publicly labelled an “adulteress” (literally, with a
capital “A”) and shunned by her small Puritan community.
2. In Sartre’s famous play “No Exit” (1946/1989), in which
three strangers find themselves uncomfortably together in a
drawing-room of Hell, there are no mirrors or windows in the
room, sleep is not permitted, and the light is always on. The characters’
eyelids are paralyzed, disallowing them even the luxury of
blinking. One of the characters, Garcin, reacts with muted horror
to the prospect of being constantly observed by the others for all
eternity. He is also convinced that he is under surveillance by
demons, “all those eyes intent on me. Devouring me.” It is not
hard to see why this would be such an exquisite torture.
3. Interestingly, after a long historical period when people
may have been able to emigrate to new social groups and to
“start over” if they spoilt their reputation, the present media
age, in some ways, more accurately reflects the conditions
faced by our ancestors. With newspapers, telephones, cameras,
television, and the Internet at our disposal, personal details
about medical problems, spending activities, criminal and financial
history, and divorce records (to name just a few potentially
sundry tidbits), are not only permanently archived, but can be
distributed in microseconds to, literally, billions of other
people. The old adage “Wherever you go, there you are” takes
on new meaning in light of the evolution of information technology.
The Internet, in particular, is an active microcosm of human
sociality that has not yet been properly analyzed in Darwinian
terms. From background checks to match making services,
from anonymous web site browsing to piracy and identity theft,
from “Googling” ourselves and peers to flaming bad professors
(e.g., www.ratemyprofessor.com) and stingy customers (e.g.,
www.bitterwaitress.com), the Internet is ancient social psychology
meeting new information technology.
Response/Bering: The folk psychology of souls
492 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
ACKNOWLEDGMENTS
I am grateful to Harvey Whitehouse and Paulo Sousa for helpful
comments on an earlier version of this response.
References
Letters “a” and “r” appearing before authors’ initials refer to target article
and response, respectively.
Aarts, H., Custers, R. & Wegner, D. M. (2005) On the inference of personal
authorship: Enhancing experienced agency by priming effect information.
Consciousness and Cognition 14:439–58. [JP]
Abramowitz, J., Huppert, J., Cohen, A. B., Cahill, S. P. & Tolin, D. F. (2002)
Religious obsessions and compulsions in a non-clinical sample: The Penn
Inventory of Scrupulosity. Behaviour Research and Therapy
40:825–38. [ABC]
Aichhorn, M., Perner, J., Kronbichler, M., Staffen, W. & Ladurner, G.
(2006) Do visual perspective tasks need theory of mind? NeuroImage
30:1059–68. [DK]
Alcorta, C. & Sosis, R. (2005) Ritual, emotion, and sacred symbols: The evolution of
religion as an adaptive complex. Human Nature 16:323–59. [aJMB]
Alexander, R. (1987) The biology of moral systems. Aldine de Gruyter. [aJMB]
Alvarado, C. S. (1992) The psychological approach to out-of-body experiences: A
review of early and modern developments. Journal of Psychology
126:237–50. [DK]
(2000) Out-of-body experiences. In: Varieties of anomalous experiences, ed.
E. Cardena, S. J. Lynn & S. Krippner, pp. 183–218. American Psychological
Association. [DK]
Alvarado, K. A., Templer, D. I., Bresler, C. & Thomas-Dobson, S. (1995) The
relationship of religious variables to death depression and death anxiety.
Journal of Clinical Psychology 51:202–04. [aJMB]
Angell, J. R. (1911) Philosophical and psychological usages of the terms mind,
consciousness, and soul. Psychological Bulletin 8:846–47. [MF]
Anonymous (1939) Wolanda Hindia: Zendingstijdschrift voor jongeren, vol. 13.
Liebenswu¨rdige Wilden. [aJMB]
Argyle, M. (2000) Psychology and religion: An introduction. Routledge. [MF]
Arzy, S., Idel, M., Landis, T. & Blanke, O. (2005) Speaking with oneself: Autoscopic
phenomena in writings from the ecstatic Kabbalah. Journal of Consciousness
Studies 12:4–29. [DK]
Astuti, R. (forthcoming a) Ancestors and the afterlife. In: The salvaged mind: Social
anthropology, religion, and the cognitive sciences, ed. H. Whitehouse &
J. Laidlaw. Carolina Academic Press. [aJMB, PLH, HW]
(forthcoming b) What happens after death? In: Questions of Anthropology, eds.
R. Astuti, J. P. Parry & C. Stafford. Berg. [PLH]
Astuti, R. & Harris, P. L. (submitted) Understanding mortality and the life of the
ancestors in rural Madagascar. [rJMB, PLH]
Atran, S. (2002) In gods we trust: The evolutionary landscape of religion. Oxford
University Press. [aJMB, IP, LES]
(2003) Genesis of suicide terrorism. Science 299:1534–39. [aJMB]
Atran, S. & Norenzayan, A. (2004) Religion’s evolutionary landscape: Counterintuition,
commitment, compassion, communion. Behavioral and Brain Sciences
27(6):713–70. [arJMB, ABC, GEN]
Axelrod, R. (1984/1990) The evolution of co-operation. Penguin. [IP]
Bainbridge, W. S. (1992) Crime, delinquency, and religion. In: Religion and mental
health, ed. J. F. Schumaker, pp. 199–210. Oxford University Press. [WSB]
(2002) The endtime family: Children of God. State University of New York
Press. [WSB]
(2006) God from the machine: Artificial intelligence models of religious cognition.
AltaMira. [WSB]
(in press) The secular abyss. Lexington. [WSB]
Baldwin, D. A. & Moses, L. S. (1996) The ontogeny of social information gathering.
Child Development 67:1915–39. [aJMB]
Baldwin, J. M. (1903) Mind and body, from the genetic point of view. Psychological
Review 10:225–47. [MF]
Baron-Cohen, S. (1999) The evolution of a theory of mind. In: The descent of mind:
Psychological perspectives on hominid evolution, ed. M. C. Corballis &
S. E. G. Lea. Oxford University Press. [aJMB]
(2000) Theory of mind and autism: A fifteen-year review. In: Understanding other
minds: Perspectives from developmental cognitive neuroscience, ed. S. BaronCohen,
H. Tager-Flusberg & D. J. Cohen, pp. 3–20. Oxford University
Press. [SF]
Baron-Cohen, S., Tager-Flusberg, H. & Cohen, D. J., eds. (2000) Understanding
other minds: Perspectives from developmental cognitive neuroscience,
2nd edition. Oxford University Press. [SF]
Baron-Cohen, S., Wheelwright, S. & Joliffe, T. (1997) Is there a “language of the
eyes”? Evidence from normal adults and adults with autism or Asperger
syndrome. Visual Cognition 4:311–41. [aJMB]
Barrett, D. B., Johnson, T. M. & Crossing, P. F. (2006) In: Britannica book of the
year, 2006. Retrieved June 4, 2006, from Encyclopædia Britannica Online:
http://search.eb.com/eb/article-9432655 [KRL]
Barrett, H. C. & Behne, T. (2005) Children’s understanding of death as the
cessation of agency: A test using sleep versus death. Cognition 96:93–
108. [arJMB, EME]
Barrett, J. L. (2000) Exploring the natural foundations of religion. Trends in
Cognitive Sciences 4:29–34. [arJMB, GEN]
(2003) Epidemiological and nativist accounts in the cognitive study of culture: A
commentary on Pyysia¨inen’s innate fear of Bering’s ghosts. Journal of
Cognition and Culture 3(3):226–32. [IP]
(2004) Why would anyone believe in God? AltaMira Press. [aJMB, HW]
Barrett, J. L. & Johnson, A. H. (2003) The role of control in attributing
intentional agency to inanimate objects. Journal of Cognition and Culture
3:208–17. [BMH]
Barrett, J. L. & Keil, F. (1996) Conceptualizing a nonnatural entity: Anthropomorphism
in God concepts. Cognitive Psychology 31:219–47. [rJMB]
Barrett, J. L., Richert, R. & Driesenga, A. (2001) God’s beliefs versus mother’s:
The development of nonhuman agent concepts. Child Development 72:50–65.
[aJMB, EME]
Bartsch, K. & Wellman, H. M. (1995) Children talk about the mind. Oxford
University Press. [DE]
Bateson, M., Nettle, D. & Roberts, G. (2006) Cues of being watched enhance
cooperation in a real-world setting. Biology Letters (Online publication:
DOI:10.1098/rsbl.2006.0509). [rJMB]
Baumeister, R. F., Bratslavsky, E., Finkenhauer, C. & Vohs, K. D. (2001) Bad is
stronger than good. Review of General Psychology 5:323–70. [aJMB]
Becker, E. (1971) The birth and death of meaning. Free Press. [JG]
(1973) The denial of death. Free Press. [JG]
Beit-Hallahmi, B. (2006) Atheists: A psychological profile. In: Cambridge companion
to atheism, ed. M. Martin. Cambridge University Press. [BB-H]
Beit-Hallahmi, B. & Argyle, M. (1997) The psychology of religious behaviour, belief
and experience. Routledge. [BB-H]
Bemmer, J. M. (2002) The rise and fall of the afterlife. Routledge. [MF]
Bering, J. M. (2002a) Intuitive conceptions of dead agents’ minds: The natural
foundations of afterlife beliefs as phenomenological boundary. Journal of
Cognition and Culture 2(4):263–308. [arJMB, BMH, KRL, IP]
(2002b) The existential theory of mind. Review of General Psychology
6:3–24. [arJMB]
(2003a) Religious concepts are probably epiphenomena: A reply to Pyysia¨inen,
Boyer, and Barrett. Journal of Cognition and Culture 3(3):244–54. [IP]
(2003b) Towards a cognitive theory of existential meaning. New Ideas in
Psychology 21:101–20. [aJMB]
(2005) The evolutionary history of an illusion: Religious causal beliefs in children
and adults. In: Origins of the social mind: Evolutionary psychology and child
development, ed. B. Ellis & D. Bjorklund. Guilford Press. [aJMB]
Bering, J. M. & Bjorklund, D. F. (2004) The natural emergence of reasoning
about the afterlife as a developmental regularity. Developmental Psychology
40:217–33. [arJMB, EME, BMH, SN, PR]
(in press) The serpent’s gift: Evolutionary psychology and consciousness. In:
Cambridge handbook of consciousness, ed. P. D. Zelazo, M. Moscovitch, &
E. Thompson. Cambridge University Press. [aJMB]
Bering, J. M., Herna´ndez-Blasi, C. & Bjorklund, D. F. (2005a) The development of
‘afterlife’ beliefs in secularly and religiously schooled children. British Journal
of Developmental Psychology 23:587–607. [aJMB]
Bering, J. M., McLeod, K. A. & Shackelford, T. K. (2005b) Reasoning about dead
agents reveals possible adaptive trends. Human Nature 16:360–81. [arJMB]
Bering, J. M. & Johnson, D. D. P. (2005) “O Lord . . . you perceive my thoughts
from afar”: Recursiveness and the evolution of supernatural agency. Journal of
Cognition and Culture 5:118–42. [arJMB]
Bering, J. M. & Parker, B. D. (in press) Children’s attributions of intentions to an
invisible agent. Developmental Psychology. [aJMB]
Bering, J. M. & Shackelford, T. K. (2004) The causal role of consciousness: A
conceptual addendum to human evolutionary psychology. Review of General
Psychology 8:227–48. [aJMB]
Berrios, G. E. & Luque, R. (1995) Cotard’s syndrome: Analysis of 100 cases. Acta
Psychiatrica Scandinavica 91:185–88. [DC]
Bjorklund, D. F. & Pellegrini, A. D. (2002) The origins of human nature:
Evolutionary developmental psychology. American Psychological
Association. [aJMB]
Blackmore, S. J. (1982) Beyond the body: An investigation of out-of-body
experiences. Heinemann. [DK]
Blanchard, E. (n.d.) Thunder gods and kamikazes: The suicide air offenses of World
War II. The War Times Journal. Retrieved November 8, 2004, from http://
www.wtj.com/articles/kamikaze/. [aJMB]
References/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 493
Blanke, O. & Arzy, S. (2005) The out-of-body experience: Disturbed self-processing
at the temporo-parietal junction. Neuroscientist 11:16–24. [DK]
Blanke, O., Landis, T., Spinelli, L. & Seeck, M. (2004) Out-of-body experience and
autoscopy of neurological origin. Brain 127:243–58. [DK]
Blanke, O. & Mohr, C. (2005) Out-of-body experience, heautoscopy, and
autoscopic hallucination of neurological origin: Implications for neurocognitive
mechanisms of corporeal awareness and self-consciousness. Brain Research
Reviews 50:184–99. [DK]
Blanke, O., Mohr, C., Michel, C. M., Pascual-Leon, A., Brugger, P., Seeck, M.,
Landis, T. & Thut, G. (2005) Linking out-of-body experience and self processing
to mental own-body imagery at the temporoparietal junction. Journal
of Neuroscience 25:550–57. [DK]
Blanke, O., Ortigue, S., Landis, T. & Seeck, M. (2002) Stimulating illusory
own-body perceptions. Nature 419:269–70. [DK]
Bloch, M. (2002) Are religious beliefs counter intuitive? In: Radical interpretation
in religion, ed. N. Frankenberry. Cambridge University Press. [MB]
Blok, S., Newman, G. & Rips, L. J. (2005) Individuals and their concepts. In:
Categorization inside and outside the lab, ed. W.-K. Ahn, R. L. Goldstone,
B. C. Love, A. B. Markman & P. Wolff, pp. 127–49. American Psychological
Association. [GEN]
Bloom, P. (2004) Descartes’ baby: How the science of child development explains
what makes us human. Basic Books. [aJMB, NJB, KRL, PR]
Blumer, H. (1969) Symbolic interactionism. Prentice-Hall. [WSB]
Bourgeois, M. (1969) Le syndrome de Cotard aujourd’hui. Annales
Me´dico-Psychologiques 127:534–44. [DC]
Bowlby, J. (1973) Attachment and loss. Hogarth. [BB-H]
Boyer, P. (2000) Functional origins of religious concepts: Conceptual and strategic
selection in evolved minds. Journal of the Royal Anthropological Institute
6:195–214. [aJMB]
(2001) Religion explained: The evolutionary origins of religious thought. Basic
Books. [arJMB, PB, IP, HW]
(2003a) Are ghost concepts “intuitive,” “endemic,” and “innate?” Journal of
Cognition and Culture 3(3):233–43. [aJMB, IP]
(2003b) Religious thought and behaviour as by-products of brain function. Trends
in Cognitive Sciences 7(3):119–24. [PB, IP]
(2004) Religion, evolution, and cognition. A review of S. Atran’s In gods we
trust, and D. S. Wilson’s Darwin’s Cathedral. Current Anthropology
45(3):430–33. [IP]
(2005) A reductionistic model of distinct modes of religious transmission. In:
Mind and religion: Psychological and cognitive foundations of religiosity,
ed. H. Whitehouse & R. N. McCauley. AltaMira Press. [IP]
Boyer, P. & Barrett, H. C. (2005) Domain-specificity and intuitive ontology. In: The
handbook of evolutionary psychology, ed. J. Buss. Wiley. [IP]
Brugger, P. (2006) From phantom limb to phantom body: Varieties of
extracorporeal awareness. In: Human body perception from the inside out,
ed. G. Knoblich, I. M. Thornton, M. Grosjean & M. Shiffrar, pp. 171–210.
Oxford University Press. [DK]
Bruner, J. (1990) Acts of meaning. Harvard University Press. [aJMB]
(2001) Self-making and world-making. In: Narrative and identity: Studies in
autobiography, self, and culture, ed. J. Brockmeier & D. Carbaugh. John
Benjamins. [aJMB]
Bulbulia, J. (2004) Area review: The cognitive and evolutionary psychology of
religion. Biology and Philosophy 19:655–86. [aJMB, IP]
(2005) Are there any religions? Method and Theory in the Study of Religion
17:71–100. [aJMB]
Bullot, N. J. & Rysiew, P. (2005) Tracking objects, tracking agents. Interdisciplines,
“Referring to Objects.” Available at: http://www.interdisciplines.org/objects/
papers/5. [NJB]
Burnham, T. & Hare, B. (in press) Engineering human cooperation: Does
involuntary neural activation increase public goods contributions? Human
Nature. [arJMB]
Campbell, J. (2002) Reference and consciousness. Clarendon Press. [NJB]
Camus, A. (1943/1991) The myth of Sisyphus, and other essays. Vintage. [aJMB]
Carey, S. & Xu, F. (2001) Infant’s knowledge of objects: Beyond object files and
object tracking. Cognition 80:179–213. [NJB]
Carroll, S. B. (2006) Endless forms most beautiful: The new science of Evo Devo and
the making of the animal kingdom. Norton. [EME]
Castelli, F., Frith, C., Happzˇ, F. & Frith, U. (2002) Autism, Asperger syndrome and
brain mechanisms for the attribution of mental states to animated shapes.
Brain 125:1839–49. [SF]
Childers, P. & Wimmer, M. (1971) The concept of death in early childhood. Child
Development 42:1293–1301. [KRL]
Clark, A. (2000) A theory of sentience. Clarendon Press. [NJB]
Clark, T. W. (1994) Death, nothingness, and subjectivity. The Humanist
54:15–20. [aJMB]
CNN March 1997 interview with Osama bin Laden. Retrieved November 7, 2004
from http://news.findlaw.com/hdocs/docs/binladen/binladenintvw-cnn.
pdf. [aJMB]
Cohen, A. B. & Hall, D. E. (submitted) Social support, existential beliefs, and
well-being among Catholic, Jewish, and Protestant older adults. [ABC]
Cohen, A. B., Malka, A., Rozin, P. & Cherfas, L. (2006) Religion and unforgivable
offenses. Journal of Personality 74:85–118. [ABC]
Cohen, A. B., Pierce, J. D., Jr., Meade, R., Chambers, J., Gorvine, B. J. &
Koenig, H. G. (2005) Intrinsic and extrinsic religiosity, belief in the afterlife,
death anxiety, and life satisfaction in young Catholic and Protestant adults.
Journal of Research in Personality 39:307–24. [ABC]
Cohen, A. B., Siegel, J. I. & Rozin, P. (2003) Faith versus practice: Different bases
for religiosity judgments by Jews and Protestants. European Journal of Social
Psychology 33:287–95. [ABC]
Cohen, D., Cottias, C. & Basquin, M. (1997) Cotard’s syndrome in a 15-year-old
girl. Acta Psychiatrica Scandinavica 95:164–65. [DC]
Colwell, J., Schro¨der. S. & Sladen, D. (2000) The ability to detect unseen staring:
A literature review and empirical tests. British Journal of Psychology
91:71–85. [rJMB]
Consoli, A., Soultanian, C., Tanguy, M. L., Laurent, C., Perisse, D., Luque, R.,
Berrios, G. & Cohen, D. (in press) Cotard’s syndrome in adolescents and
young adults is associated with an increased risk of bipolar disorder. Bipolar
Disorders. [DC]
Crick, F. (1994) The astonishing hypothesis: The scientific search for the soul.
Charles Scribner’s. [DE]
Crook, J. H. & Crook, S. J. (1988) Tibetan polyandry: Problems of adaptation and
fitness. In: Human reproductive behavior: A Darwinian perspective,
ed. L. Betzig, M. Borgerhoff-Mulder & P. Turke, pp. 97–114. Cambridge
University Press. [ABC]
Daly, M. & Wilson, M. (1994). Evolutionary psychology of male violence. In: Male
violence, ed. J. Archer. Routledge. [aJMB]
Davis, D. G. & Nolen-Hoeksema, S. (2001) Loss and meaning: How do people
make sense of loss? American Behavioral Scientist 44:726–41. [aJMB]
Davis, N. Z. (1977) Ghosts, kin, and progeny: Some features of family life in early
modern France. Daedalus 106:87–114. [aJMB]
Dawkins, R. (2003) Viruses of the mind. In: A devil’s chaplain: Selected essays
by Richard Dawkins, ed. L. Menon, pp. 128–45. Weidenfeld and
Nicolson. [BMH]
de Unamuno, M. (1912/1954) Tragic sense of life. Dover. [aJMB]
Dechesne, M., Pyszczynski, T., Arndt, J., Ransom, S., Sheldon, K. M., van
Knippenberg, A., & Janssen, J. (2003) Literal and symbolic immortality:
The effect of evidence of literal immortality on self-esteem striving in response
to mortality salience. Journal of Personality and Social Psychology
84:722–37. [aJMB, JG]
Defeyter, M. A. & German, T. P. (2003) Acquiring an understanding of
design: Evidence from children’s insight problem solving. Cognition
89:133–55. [aJMB, EME]
Dennett, D. C. (1991) Consciousness explained. Little, Brown. [aJMB]
(1995) Darwin’s dangerous idea: Evolution and the meanings of life. Simon &
Schuster. [aJMB]
(1996) Kinds of minds. Weidenfeld and Nicolson. [aJMB]
(2006) Breaking the spell: Religion as a natural phenomenon. Viking. [DK]
Deridder, R., Hendriks, E., Zani, B., Pepitone, A. & Saffioti, L. (1999) Additional
cross-cultural evidence on the selective usage of nonmaterial beliefs in
explaining life events. European Journal of Social Psychology
29:435–42. [aJMB]
Diener, E., Fraser, S., Beaman, A. & Kelman, R. (1976) Effects of deindividuation
variables on stealing among Halloween trick-or-treaters. Journal of Personality
and Social Psychology 33:178–83. [rJMB]
Dijksterhuis, A., Preston, J., Wegner, D. M. & Aarts, H. (under review) Effects of
subliminal priming of natural and supernatural agents on judgments of
authorship. [JP]
Dobzhansky, T. (1967) The biology of ultimate concern. New American
Library. [DE]
Donnellan, K. S. (1974) Speaking of nothing. The Philosophical Review
83(1):3–31. [NJB]
Dostoyevsky, F. (1877/1949) Diary of a writer. Charles Scribner’s. [aJMB]
Dunbar, R. I. M. (1993) Coevolution of neocortical size, group size and language in
humans. Behavioral and Brain Sciences 16:681–735. [aJMB]
(2004) The human story: A new history of mankind’s evolution. Faber and
Faber. [aJMB]
Dunbar, R. I. M. & Spoors, M. (1995) Social networks, support cliques, and kinship.
Human Nature 6:273–90. [rJMB]
Durant, W. & Durant, A. (1968) Lessons of history. Simon & Schuster. [ABC]
Ellison, C. (1991) Religious involvement and subjective well-being. Journal of
Health and Social Behavior 32:80–99. [KRL]
Ellison, P., Govern, J., Petri, H. & Figler, M. (1995) Anonymity and aggressive
driving behaviour: A field study. Journal of Social Behavior and Personality
10:265–72. [rJMB]
Emler, N. (1994) Gossip, reputation, and social adaptation. In: Good gossip,
ed. A. Ben-Ze’ev. University of Kansas Press. [aJMB]
References/Bering: The folk psychology of souls
494 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Epley, N., Morewedge, C. K. & Keysar, B. (2004) Perspective taking in children
and adults: Equivalent egocentrism but differential correction. Journal of
Experimental Social Psychology 40:760–68. [aJMB]
Estes, D. (1994) Young children’s understanding of the mind: Imagery,
introspection, and some implications. Journal of Applied Developmental
Psychology 15:529–48. [DE]
(1998) Young children’s awareness of their mental activity. Child Development
69:1345–60. [DE]
Estes, D., Wellman, H. M. & Woolley, J. D. (1989) Children’s understanding of
mental phenomena. In: Advances in child development and behavior, vol. 22,
ed. H. W. Reese, pp. 41–87. Academic Press. [DE]
Evans, E. M. (2000) The emergence of beliefs about the origins of species in schoolage
children. Merrill-Palmer Quarterly 46:221–54. [EME]
(2001) Cognitive and contextual factors in the emergence of diverse belief
systems: Creation versus evolution. Cognitive Psychology 42:217–66.
[aJMB, EME]
(2005) Teaching and learning about evolution. In: The virus and the whale:
Explore evolution in creatures small and large, ed. J. Diamond. NSTA
Press. [EME]
Evans, E. M., Poling, D. & Mull, M. (2001) Confronting the existential questions:
Children’s understanding of death and origins. Biennial Meeting of the Society
for Research In Child Development, Minneapolis, MN, April 19–22,
2001. [EME]
Evans, G. (1982) The varieties of reference, pp. 143–203. Oxford University
Press. [NJB]
Fehr, E. & Fischbacher, U. (2004) Third-party punishment and social norms.
Evolution and Human Behavior 25:63–87. [rJMB]
Ferrari, M. (2003) Baldwin’s two developmental resolutions to the mind-body
problem. Developmental Review 23:79–108. [MF]
Fessler, D. M. T. & Haley, K. J. (2003) The strategy of affect: Emotions in human
cooperation. In: The genetic and cultural evolution of cooperation,
ed. P. Hammerstein. MIT Press. [rJMB]
Festinger, L., Pepitone, A. & Newcomb, T. (1952) Some consequences of
deindividuation in a group. Journal of Abnormal & Social Psychology
47:382–89. [rJMB]
Flavell, J. H., Green, F. L. & Flavell, E. R. (2000) Development of children’s
awareness of their own thoughts. Journal of Cognition and Development
1:97–112. [EME, PLH]
Florian, V. & Kpavetz, S. (1985) Children’s concept of death: A cross-cultural
comparison among Muslims, Druze, Christians, and Jews in Israel. Journal
of Cross-Cultural Psychology 16:174–89. [KRL]
Forde, D., ed. (1954) African worlds. Oxford University Press. [MB]
Fortes, M. (1959) Oedipus and Job in West African religion. Cambridge University
Press. [MB]
Fowles, J. (1978) The Magus. Dell. [aJMB]
Frank, R. (1988) Passions within reason: The strategic role of emotions.
Norton. [aJMB, PB]
Franks, L. (1994) The return of the fugitive. The New Yorker, June 13,
pp. 40–59. [aJMB]
Freud, S. (1915) Thoughts for the times on war and death. In: The standard edition
of the complete psychological works of Sigmund Freud, vol. 14, pp. 274–301.
Hogarth. [BB-H]
Futuyma, D. J. (1998) Evolutionary biology. Sinauer. [JH]
Gallup Organization (1999) Americans remain very religious, but not necessarily
in conventional ways. Retrieved from http://www.gallup.com/poll/content/
login.aspx?ci ¼ 3385. [aJMB]
Gardner, M. (1999) The whys of a philosophical scrivener, 2nd edition. St. Martin’s
Press. [DE]
Geertz, C. (1960) The religion of Java. Free Press. [BB-H]
Gergely, G. & Csibra, G. (2003) Teleological reasoning in infancy: The naı¨ve theory
of rational action. Trends in Cognitive Sciences 7:287–92. [aJMB]
German, T. P. & Barrett, H. C. (2005) Functional fixedness in a technologically
sparse culture. Psychological Science 16:1–5. [aJMB]
Gilbert, D. T. (2001) Why economists are not afraid to die. Paper presented at the
First International Conference on Experimental Existential Psychology,
Amsterdam, The Netherlands. August 2–4, 2001. [aJMB]
Gilbert, P. & McGuire, M. T. (1998) Shame, status, and social roles:
Psychobiology and evolution. In: Shame, interpersonal behavior,
psychopathology, and culture, ed. P. Gilbert & B. Andrews. Oxford
University Press. [rJMB]
Gintis, H. (2000) Strong reciprocity and human sociality. Journal of Theoretical
Biology 206(2):169–79. [PB]
Goffman, E. (1959) The presentation of self in everyday life. Penguin. [aJMB]
(1963) Stigma: Notes on the management of spoiled identity. Simon &
Schuster. [arJMB]
Gold, G. J. & Weiner, B. (2000) Remorse, confession, group identity, and
expectancies about repeating a transgression. Basic and Applied Social
Psychology 22:291–300. [rJMB]
Goldenberg, J. L., McCoy, S. K., Pyszczynski, T., Greenberg, J. & Solomon, S.
(2000) The body as a source of self-esteem: The effect of mortality salience on
identification with one’s body, interest in sex, and appearance monitoring.
Journal of Personality and Social Psychology 79:118–30. [BMH]
Golsworthy, R. & Coyle, A. (1999) Spiritual beliefs and the search for meaning
among older adults following partner loss. Mortality 4:21–40. [aJMB]
Gottfried, G. M. & Jow, E. E. (2003) “I just talk with my heart”: The mind-body
problem, linguistic input, and the acquisition of folk psychological beliefs.
Cognitive Development 18:79–90. [aJMB]
Greeley, A. M. & Hout, M. (1999) Americans’ increasing belief in life after death:
Religious competition and acculturation. American Sociological Review
64:813–35. [aJMB]
Green, C. E. (1968) Out-of-body experiences. Institute of Psychophysical
Research. [DK]
Greenberg, J., Landau, M. J., Solomon, S. & Pyszczynski, T. (in press) What is the
primary psychological function of religion? In: Handbook of the psychology of
religion, ed. D. Wurff. Oxford University Press. [JG]
Greyson, B. (2000) Dissociation in people who have near-death experiences: Out of
their bodies or out of their minds? Lancet 355:460–63. [DK]
Guthrie, S. (1993) Faces in the clouds. Oxford University Press. [aJMB]
Habermas, J. (1971) Knowledge and human interests. Beacon Press. [WSB]
Hafer, C. L. & Be`gue, L. (2005) Experimental research on just-world theory:
Problems, developments, and future challenges. Psychological Bulletin
131:128–67. [aJMB]
Haidt, J., Bjo¨rklund, F. & Murphy, S. (2004) Moral dumbfounding: When
intuition finds no reason. Unpublished manuscript, University of
Virginia. [rJMB]
Haley, K. J. & Fessler, D. M. T. (2005) Nobody’s watching? Subtle cues affect
generosity in an anonymous economic game. Evolution and Human Behavior
26:245–56. [arJMB]
Hansen, G., Jensen, S. B., Chandresh, L. & Hilden, T. (1988) The psychotropic
effect of ketamine. Journal of Psychoactive Drugs 20:419–25. [DK]
Harcourt, A. H. & de Waal, F. B., eds. (1992) Coalitions and alliances in humans
and other animals. Oxford University Press. [PB]
Harris, P. L., Brown, E., Marriott, C., Whittall, S. & Harmer, S. (1991) Monsters,
ghosts and witches: Testing the limits of the fantasy-reality distinction in young
children. British Journal of Developmental Psychology 9:105–23. [rJMB]
Harris, P. L. & Gime´nez, M. (2005) Children’s acceptance of conflicting testimony:
The case of death. Journal of Cognition and Culture 5:143–64.
[arJMB, PLH, HW]
Harris, P. L. & Koenig, M. A. (2006) Trust in testimony: How children learn about
science and religion. Child Development 77:505–24. [EME]
Harrison, S. (1993) The mask of war: Violence, ritual, and the self in Melanesia.
Manchester University Press. [HW]
Haselton, M. G. & Buss, D. M. (2003) Biases in social judgment: Design flaws or
design features? In: Responding to the social world: Implicit and explicit
processes in social judgments and decisions, ed. J. Forgas, K. Williams, &
B. von Hippel. Cambridge University Press. [aJMB]
Hassan, N. (2001) Talking to the “human bombs.” The New Yorker, November 19,
2001. Retrieved November 7, 2004, from http://www.newyorker.com/fact/
content/?011119fa_FACT1. [aJMB]
Henrich, J. (2006) Cooperation, punishment, and the evolution of human
institutions. Science 312:60–61. [BB-H]
Hiadt, J. (2005) The happiness hypothesis. Basic Books. [KRL]
Hilton, D. J. (1995) The social context of reasoning: Conversational inference and
rational judgment. Psychological Bulletin 118:248–71. [BMH]
Hilton, J. L., Fein, S. & Miller, D. T. (1993) Suspicion and dispositional inference.
Personality and Social Psychology Bulletin 19:501–12. [aJMB]
Hinde, R. A. (1999) Why gods persist: A scientific approach to religion.
Routledge. [aJMB]
Hirsch, E. (1982) The concept of identity. Oxford University Press. [NJB]
Homans, G. C. (1974) Social behavior: Its elementary forms. Harcourt, Brace,
Jovanovich. [WSB]
Humphrey, N. (2006) Seeing red: A study in consciousness, pp. 124–29.
Belknap. [DK]
Irwin, H. J. (1985) Flight of mind: A psychological study of the out-of-body
experience. Scarecrow Press. [DK]
James, W. (1902/1961) The varieties of religious experience: A study in human
nature. (The Gifford Lectures on Natural Religion delivered at Edinburgh in
1901–1902). Longman’s Green/Collier. [BB-H, MF]
Jansen, V. A. A. & van Baalen, M. (2006) Altruism through beard chromodynamics.
Nature 440:663–66. [BMH]
Johnson, C. N. (2000) Putting different things together: The development of
metaphysical thinking. In: Imagining the impossible: Magical, scientific, and
religious thinking in children, ed. K. Rosengren, C. N. Johnson & P. Harris.
Cambridge University Press. [CNJ]
Johnson, C. N. & Wellman, H. M. (1982) Children’s developing conceptions of the
mind and brain. Child Development 53:222–34. [aJMB]
References/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 495
Johnson, D. D. P. (2005) God’s punishment and public goods: A test of the
supernatural punishment hypothesis in 186 world cultures. Human Nature
16:410–46. [rJMB]
Johnson, D. D. P. & Bering, J. M. (2006) Hand of God, mind of man: Punishment
and cognition in the evolution of cooperation. Evolutionary Psychology
4:219–33. [rJMB]
Johnson, D. D. P. & Kru¨ger, O. (2004) The good of wrath: Supernatural punishment
and the evolution of cooperation. Political Theology 5:159–76.
[aJMB]
Jonas, E. & Fischer, P. (in press) Terror management and religion – Evidence that
intrinsic religiousness mitigates worldview defense following mortality
salience. Journal of Personality and Social Psychology. [JG]
Jung, C. G. (1934) Archetypes of the collective unconscious. In: The archetype and
the collective unconscious. Collected works, 9 I. Princeton University
Press. [DC]
Kanazawa, S. (2004) General intelligence as a domain-specific adaptation.
Psychological Review 111:512–23. [KRL]
Kelemen, D. (2004) Are children “intuitive theists”? Reasoning about purpose and
design in nature. Psychological Science 15:295–301. [aJMB]
Kelemen, D. & DiYanni, C. (2005) Intuitions about origins: Purpose and
intelligence in children’s reasoning about nature. Journal of Cognition and
Development 6:3–31. [aJMB]
Kenrick, D. T. (2006) Evolutionary psychology: Resistance is futile. Psychological
Inquiry 17:102–108. [ABC]
Kenrick, D. T., Maner, J. K., Butner, J., Li, N. P., Becker, D. V. & Schaller, M.
(2002) Dynamic evolutionary psychology: Mapping the domains of the
new interactionist paradigm. Personality and Social Psychology Review
6:347–56. [ABC]
Koenig, H. G. & Cohen, H. J., eds. (2002) The link between religion and
health: Psychoneuroimmunology and the faith factor. Oxford University
Press. [KRL]
Koocher, G. (1973) Childhood, death, and cognitive development. Developmental
Psychology 9:369–75. [aJMB]
Kripke, S. (1980) Naming and necessity. Harvard University Press. [NJB]
Kriwaczek, P. (2002) In search of Zarathustra. Weidenfeld and Nicolson. [JG]
Kuhlmeier, V. A., Bloom, P. & Wynn, K. (2004) Do 5-month-old infants see humans
as material objects? Cognition 94:95–103. [aJMB, KRL]
Kuran, T. (1998) Ethnic norms and their transformation through reputational
cascades. Journal of Legal Studies 27(2):623–59. [PB]
Kurzban, R. & Leary, M. R. (2001) Evolutionary origins of stigmatization: The
functions of social exclusion. Psychological Bulletin 127(2):187–208. [PB]
Lackner, J. R. (1992) Sense of body position in parabolic flight. Annals of the
New York Academy of Sciences 656:329–39. [DK]
Lakoff, G. & Johnson, M. (1999) Philosophy in the flesh: The embodied mind and its
challenge to Western thought. Basic Books. [SF]
Landman, J. (2001) The crime, punishment, and ethical transformation of two
radicals: Or how Katherine Power improves on Dostoyevsky. In: Turns in the
road: Narrative studies of lives in transition, ed. D. P. McAdams,
R. Josselson, & A. Lieblich. American Psychological Association. [aJMB]
Lerner, M. J. (1980) The belief in a just world: A fundamental delusion.
Plenum. [aJMB]
Lerner, M. J. & Miller, D. T. (1978) Just world research and the attribution
process: Looking back and ahead. Psychological Bulletin 85:1030–51.
[aJMB]
Lester, D., Aldridge, M., Aspenberg, C., Boyle, K., Radsniak, P. & Waldon, C.
(2002) What is the afterlife like? Undergraduate beliefs about the afterlife.
Omega: Journal of Death and Dying 44:113–26. [aJMB]
Levy-Bruhl, L. (1910) Les fonctions mentales dans les socie´te´s infe´rieures. Presses
Universitaires de France. [DC]
(1935) La mythologie primitive. Presses Universitaires de France. [DC]
Livingston, K. (2002) Reason, faith, and the good life. Free Inquiry
22:42–47. [KRL]
(2005) Religious practice, brain, and belief. Journal of Cognition and Culture
5:75–117. [KRL]
Locke, J. (1689/1975) An essay concerning human understanding. Oxford
University Press. [NJB]
Luria, L. K. (2001) Separated by death? Burials, cemeteries, and confessional
boundaries in seventeenth-century France. French Historical Studies
24:185–222. [aJMB]
MacIntyre, F. (2004) Was religion a kinship surrogate? Journal of the American
Academy of Religion 72(3):653–94. [IP]
Malinowski, B. (1916) Baloma: The spirits of the dead in the Trobriand Islands. The
Journal of the Royal Anthropological Institute of Great Britain and Ireland
46:353–430. [MB]
(1935) The foundations of faith and morals: An anthropological analysis of
primitive belief and conduct with special reference to the fundamental problem
of religion and ethics. Oxford University Press. [rJMB]
May, H. (2000) “Murderers’ relatives”: Managing stigma, negotiating identity.
Journal of Contemporary Ethnography 29:198–221. [rJMB]
McAdams, D. P. (2001) The psychology of life stories. Review of General
Psychology 5:100–22. [arJMB]
McCauley, R. N. (2000) The naturalness of religion and the unnaturalness of
science. In: Explanation and cognition, ed. F. C. Keil & R. W. Wilson. MIT
Press. [aJMB]
McClosky, M., Caramazza, A. & Green, B. (1980) Curvilinear motion in the absence
of external forces: Naı¨ve beliefs about the motion of objects. Science
210:1139–41. [NLG]
McCulloch, W. H. (1992) A certain archway: Autoscopy and its companions seen in
Western writing. History of Psychiatry 3:59–78. [DK]
McGarrigle, J., Grieve, R. & Hughes, M. (1978) Interpreting inclusion: A
contribution to the study of the child’s cognitive and linguistic development.
Journal of Experimental Child Psychology 26:528–50. [BMH]
McGinn, C. (1991) The problem of consciousness. Blackwell. [aJMB]
Mealey, L. (1995) The sociobiology of sociopathology: An integrated evolutionary
model. Behavioral and Brain Sciences 20:523–99. [rJMB]
Melser, D. (2004) The act of thinking. MIT Press. [SF]
Merricks, T. (2001) Objects and persons. Oxford University Press. [NJB]
Mesoudi, A., Whiten, A. & Laland, K. N. (in press) Towards a unified science of
cultural evolution. Behavioral and Brain Sciences 29(4):329–47. [JH]
Metzinger, T. (2003) Being no one: The self-model theory of subjectivity. MIT
Press. [DK]
(2005) Out-of-body experiences as the origin of the concept of a “soul.” Mind and
Matter 3:57–84. [DK]
Milinski, M., Semmann, D. & Krambeck, H. J. (2002) Donors to charity gain in both
indirect reciprocity and political reputation. Proceedings of the Royal Society
B: Biological Sciences 269:881–83. [rJMB]
Mithen, S. (1996) The prehistory of the mind. Thames and Hudson. [aJMB]
Mittelstaedt, H. & Glasauer, S. (1993) Illusions of verticality in weightlessness.
Clinical Investigations 71:732–39. [DK]
Mohr, C. & Blanke, O. (2005) The demystification of autoscopic phenomena:
Experimental propositions. Current Psychiatry Reports 7:189–95. [DK]
Mohr, D. (1978) Development of attributes of personal identity. Developmental
Psychology 14:427–28. [SN]
Montemayor, R. & Eisen, M. (1977) The development of self-conceptions from
childhood to adolescence. Developmental Psychology 13:314–19. [SN]
Nagel, T. (1974) What is it like to be a bat? Philosophical Review 82:435–50. [PR]
Neiman, S. (2004) Evil in modern thought. Princeton University Press. [BB-H]
Nemeroff, C. & Rozin, P. (1994) The contagion concept in adult thinking in the
United States: Transmission of germs and interpersonal influence. Ethos
22:158–86. [rJMB, NLG]
Nesse, R. M. (2000) Strategic subjective commitment. Journal of Consciousness
Studies 7(1-2):326–30. [PB]
Nettle, D. (2004) Adaptive illusions: Optimism, control and human rationality.
In: Emotion, evolution and rationality, ed. D. Evans & P. Cruse. Oxford
University Press. [aJMB]
Nichols, S. (in press) Imagination and immortality. Synthe`se. [aJMB]
Nowak, M. A., Sasaki, A., Taylor, C. & Fudenberg, D. (2004) Emergence of
cooperation and evolutionary stability in finite populations. Nature
428(6983):646–50. [IP]
Nowak, M. A. & Sigmund, K. (1994) The alternating prisoner’s dilemma. Journal
of Theoretical Biology 168:219–26. [IP]
(1998) Evolution of indirect reciprocity by image scoring. Nature
393:573–77. [rJMB]
(2005) Evolution of indirect reciprocity. Nature 437(7063):1291–98. [IP]
Nozick, R. (1981) Philosophical explanations. Harvard University Press. [GEN]
Nunn, C. Z. (1964) Child control through a “coalition with God.” Child
Development 35:417–32. [BB-H]
Ohtsuki, H., Hauert, C., Lieberman, E. & Nowak, M. A. (2006) A simple rule
for the evolution of cooperation on graphs and social networks. Nature
441:502–505. [BB-H]
Olson, E. T. (1997) The human animal: Personal identity without psychology.
Oxford University Press. [NJB]
O’Neill, D. K. & Chong, S. C. F. (2001) Preschool children’s difficulty understanding
the types of information obtained through the five senses. Child
Development 72:803–15. [aJMB]
Orians, G. H. (1969) On the evolution of mating systems in birds and mammals.
American Naturalist 103:589–603. [ABC]
Ottenberg, P. (1958/1980) Marriage relationships in the double descent system of
the Afikpo Igbo of southeastern Nigeria. University Microfilms. [aJMB]
Owen, P. (2002) Soldier leaves behind wife, 4 children. The Oklahoman, May 30,
2002, p. 3-A. [aJMB]
Papineau, D. (2002) Thinking about consciousness. Oxford University Press. [SF]
Parfit, D. (1971) Personal identity. The Philosophical Review 80(1):3–27. [NJB]
(1984) Reasons and persons. Oxford University Press. [NJB, GEN]
References/Bering: The folk psychology of souls
496 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5
Parker, S. T. (1998) A social selection model for the evolution and adaptive
significance of self-conscious emotions. In: Self-awareness: Its nature and
development, ed. M. Ferrari & R. J. Sternberg, pp. 108–34. Guilford
Press. [MF]
Pargament, K. I. (1997) The psychology of religion and coping: Theory, research,
practice. Guilford Press. [JG]
Parsons, T., Shils, E. A. & Tolman, E. C., eds. (1951) Toward a general theory of
action. Harvard University Press. [WSB]
Pepitone, A. & Saffioti, L. (1997) The selectivity of nonmaterial beliefs in interpreting
life events. European Journal of Social Psychology 27:23–35. [aJMB]
Perry, J. (2001) Reference and reflexivity. CSLI Publications. [NJB]
Pettazzoni, R. (1955) On the attributes of God. Numen 2:1–27. [rJMB]
Piaget, J. (1923) La psychologie et les valeurs religieuses. In: Sainte-Croix 1922,
ed. L’association chre´tienne d’e´tudiants de la Suisse Romande. La
Concorde. [MF]
(1928) L’Immanence. In: Deux types d’attitudes religieuses: Immanence et
transcendance, ed. J. Piaget & J. de la Harpe. Editions de L’Association
Chretienne d’Etudiants de la Suisse Romande. De´poˆt Central. [MF]
(1930) Immanentisme et foi religieuse. Librarie H. Robert. [MF]
(1932/1965/1978) The moral judgment of the child. Free Press. (1932/1978
French edition, Le jugement moral chez l’enfant. Presses Universitaires de
France.) [aJMB, MF]
Plato (c.350 bce/1997) Phaedo. In: Plato: Complete works, ed. J. M. Cooper &
D. S. Hutchinson. Hackett. [MF]
Poling, D. A. & Evans, E. M. (2004) Are dinosaurs the rule or the exception?
Developing concepts of death and extinction. Cognitive Development
19:363–83. [EME]
Povinelli, D. J. & Bering, J. M. (2002) The mentality of apes revisited. Current
Directions in Psychological Science 11:115–19. [aJMB]
Previc, F. H. (in press) The role of the extrapersonal brain systems in religious
activity. Consciousness and Cognition. [DK]
Pronin, E., Wegner, D. M., McCarthy, K. & Rodriguez, S. (2006) Everyday magical
powers: The role of apparent mental causation in the overestimation of
personal influence. Journal of Personality and Social Psychology
91:218–31. [JP]
Pylyshyn, Z. W. (1989) The role of location indexes in spatial perception: A sketch of
the FINST spatial-index model. Cognition 32(1):65–97. [NJB]
(2003) Seeing and visualizing: It’s not what you think. MIT Press. [NJB]
Pyszczynski, T., Greenberg, J., Solomon, S., Cather, C. & Gat, I. (1995) Defensive
distancing from victims of serious illness: The role of delay. Personality and
Social Psychology Bulletin 21:13–20. [aJMB]
Pyysia¨inen, I. (2001) How religion works: Towards a new cognitive science of
religion. Cognition and Culture Book Series, vol. 1. Brill. [aJMB, IP]
(2003) On the “innateness” of religion: A comment on Bering. Journal of
Cognition and Culture 3(3):218–25. [IP]
(2006) Amazing grace: Religion and the evolution of the human mind. In: Where
God and science meet: How brain and evolutionary studies alter our
understanding of religion, I–III. Vol. I: The evolutionary psychology of
religion: How evolution shaped the religious brain, ed. P. McNamara, Ch. 10.
Greenwood. [IP]
Quine, W. V. O. (1960) Word and object. MIT Press. [NJB]
Rank, O. (1931/1961) Psychology and the soul (Perpetua Edition). Ed.
A. S. Barnes. [JG]
Rehm, J., Steinleitner, M. & Lilli, W. (1987) Wearing uniforms and aggression: A
field experiment. European Journal of Social Psychology 17:357–60. [rJMB]
Reid, T. (1785/1969) Essays on the intellectual powers of man, ed. B. Brody. MIT
Press. [SN]
Reynolds, V. & Tanner, R. (1995) The social ecology of religion. Oxford University
Press. [arJMB]
Richerson, P. J. & Boyd, R. (2005) Not by genes alone, how culture transformed
human evolution. The University of Chicago Press. [NJB, JH, IP]
Riggs, K. J. & Peterson, D. M. (2000) Counterfactual thinking in pre-school
children: Mental state and causal inferences. In: Children’s reasoning and the
mind, ed. P. Mitchell & K. J. Riggs, pp. 87–99. Psychology Press. [BMH]
Rimbaud, A. (1873/1999) A Season in Hell and The Illuminations. Trafalgar Square
Publishing. [aJMB]
Rips, L. J., Blok, S. & Newman, G. (2006) Tracing the identity of objects.
Psychological Review 113:1–30. [GEN]
Roach, M. (2006) Spook: Science tackles the afterlife. Norton. [DK]
Robbins, P. & Jack, A. I. (2006) The phenomenal stance. Philosophical Studies
127:59–85. [PR]
Roehlkepartain, E. C., King, P. E., Wagener, L. & Benson, P. L., eds. (2006)
The handbook of spiritual development in childhood and adolescence.
Sage. [CNJ]
Roes, F. L. & Raymond, M. (2003) Belief in moralizing gods. Evolution and Human
Behavior 24:126–35. [aJMB]
Rorty, A. O., ed. (1976) The identities of persons. University of California
Press. [NJB]
Ross, L. T. & Kaplan, K. J. (1994) Life ownership orientation and attitudes toward
abortion, suicide, doctor-assisted suicide, and capital punishment. Omega:
Journal of Death and Dying 28:17–30. [aJMB]
Rosse, R. B., Collins, J. P., Jr., Fay-McCarthy, M., Alim, T. M., Wyatt, R. J. &
Deutsch, S. I. (1994) Phenomenologic comparison of the idiopathic
psychosis of schizophrenia and drug-induced cocaine and phencyclidine
psychoses: A retrospective study. Clinical Neuropharmacology
17:359–69. [DK]
Rozin, P., Haidt, J. & McCauley, C. R. (1993) Disgust. In: Handbook of emotions,
ed. M. Lewis & J. M. Haviland. Guilford Press. [aJMB]
Russell, B. (1912) Problems of philosophy. Williams and Norgate. [NJB]
(1918) Knowledge by acquaintance and knowledge by description, Mysticism and
logic and other essays. Longmans, Green. [NJB]
(1956) Logic and knowledge; essays 1901–1950, ed. R. C. Marsh. George Allen &
Unwin. [NJB]
Rutgers, A., Bakermans-Kranenburg, M., van Ijzendoorn, M. & van BerckelaerOnnes,
I. (2004) Autism and attachment: A meta-analytic review. Journal of
Child Psychology and Psychiatry 45:1123–34. [SF]
Ryle, G. (1949) The concept of mind. Hutchinson. [SF]
Sandelands, L. E. (2005) Man and nature in God. Transaction. [LES]
Sandin, B. & Sather, G. (1980) Iban adat and augury. Penerbit Universiti Sains
Malaysia for School of Comparative Social Sciences. [rJMB]
Sartre, J.-P. (1937/1969) The Wall: And other stories. New Directions. [aJMB]
(1946/1989) No Exit and other plays. Vintage. [rJMB]
(1957) Existentialism and human emotions. Philosophical Library. [aJMB]
Saxe, R., Jamal, N. & Powell, L. (2006) My body or yours? The effect of visual
perspective on cortical body representation. Cerebral Cortex
16:178–82. [DK]
Schelling, T. C. (1960) The strategy of conflict. Harvard University Press. [aJMB, PB]
Scholl, B. J. (2001) Objects and attention: The state of the art. Cognition
80:1–46. [GEN]
Schutz, A. (1967) The phenomenology of the social world. Northwestern University
Press. [WSB]
(1971) On multiple realities. In: Collected papers, vol. 1, pp. 207–59.
Nijhoff. [WSB]
Selman, R. (1980) The growth of interpersonal understanding: Developmental and
clinical analyses. Academic Press. [SN]
Shaver, P., Lenauer, M. & Sadd, S. (1980) Religiousness, conversion, and subjective
well-being: The “healthy-minded” religion of modern American women.
American Journal of Psychiatry 137:1563–68. [KRL]
Shiels, D. (1978) A cross-cultural study in out-of-body experiences, waking and
sleeping. Journal of Social Psychological Research 49:697–741. [DK]
Shoemaker, S. (1959) Personal identity and memory. The Journal of Philosophy
56(22):868–82. [NJB]
(1999) Self, body, and coincidence. Supplement to the Proceedings of the
Aristotelian Society 73(1):287–306. [NJB]
Sierra, M. & Berrios, G. E. (1998) Depersonalization: Neurobiological perspectives.
Biological Psychiatry 44:898–908. [DK]
Silke, A. (2003) Deindividuation, anonymity and violence: Findings from Northern
Ireland. Journal of Social Psychology 143:493–99. [rJMB]
Slaughter, V., Jaakkola, K. & Carey, S. (1999) Constructing a coherent theory:
Children’s biological understanding of life and death. In: Children’s
understanding of biology, health, and ethics, ed. M. Siegal & C. Peterson.
Cambridge University Press. [aJMB]
Slaughter, V. & Lyons, M. (2003) Learning about life and death in early childhood.
Cognitive Psychology 46:1–30. [aJMB]
Slone, D. J. (2004) Theological incorrectness: Why religious people believe what
they shouldn’t. Oxford University Press. [aJMB]
Smith, P. C., Range, L. M. & Ulmer, A. (1992) Belief in afterlife as a buffer in
suicidal and other bereavement. Omega: Journal of Death and Dying
24:217–25. [aJMB]
Solomon, S., Greenberg, J. & Pyszczynski, T. (1991) A terror management theory
of social behavior: The psychological functions of self-esteem and cultural
worldviews. In: Advances in experimental social psychology, vol. 24, ed.
M. Zanna, pp. 91–159. Academic Press. [JG]
Sosis, R. & Alcorta, C. (2003) Signaling, solidarity, and the sacred: The evolution of
religious behavior. Evolutionary Anthropology 12:264–74. [aJMB]
Soultanian, C., Perisse, D., Revah-Levy, A., Luque, R., Mazet, P. & Cohen, D.
(2005) Cotard’s syndrome in adolescents and young adults: A possible onset of
bipolar disorder requiring a mood stabilizer? Journal of Child and Adolescent
Psychopharmacology 15:706–11. [DC]
Sperber, D. & Hirschfeld, L. A. (2004) The cognitive foundations of cultural
stability and diversity. Trends in Cognitive Sciences 8(1):40–46. [aJMB,
GEN, IP]
Stark, R. (1996) Religion as context: Hellfire and delinquency one more time.
Sociology of Religion 57:163–73. [WSB]
Sterelny, K. (2003) Thought in a hostile world: The evolution of human cognition.
Blackwell. [IP]
References/Bering: The folk psychology of souls
BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5 497
Stone, V. E. & Gerrans, P. (2006) Does the normal brain have a theory of mind?
Trends in Cognitive Sciences 10(1):3–4. [IP]
Strawson, P. F. (1959) Individuals: An essay in descriptive metaphysics, pp. 18–20.
Methuen. [NJB]
Subbotsky, E. V. (1997) Explanations of unusual events: Phenomenalistic causal
judgments in children and adults. British Journal of Developmental Psychology
15:13–36. [aJMB]
(2001) Causal explanations of events by children and adults: Can alternative
causal modes coexist in one mind? British Journal of Developmental
Psychology 19:23–46. [arJMB]
Suddendorf, T. & Whiten, A. (2001) Mental evolution and development: Evidence
for secondary representation in children, great apes and other animals.
Psychological Bulletin 127:629–50. [aJMB]
Suzuki, D. T. (1962/1972) What is Zen? Harper & Row/Perennial Library. [MF]
Tager-Flusberg, H. & Joseph, R. M. (2005) How language facilitates the acquisition
of false belief understanding in children with autism. In: Why language matters
for theory of mind, ed. J. Astington & J. Baird. Oxford University Press. [SF]
Tangney, J. P. (2003) Self-relevant emotions. In: Handbook of self and identity, ed.
M. R. Leary & J. P. Tangney. Guilford Press. [rJMB]
Thalbourne, M. A. (1996) Belief in life after death: Psychological origins and
influences. Personality and Individual Differences 21:1043–45. [aJMB]
Thomson, J. J. (1997) People and their bodies. In: Reading Parfit, ed. J. Dancy.
Blackwell. [NJB]
Titchener, E. B. (1898) The feeling of being stared at. Science 8:895–97. [NLG]
Tomasello, M. & Call, J. (1997) Primate cognition. Oxford University Press. [aJMB]
Tooby, J. & Cosmides, L. (1992) The psychological foundations of culture. In:
The adapted mind, Evolutionary psychology and the generation of culture,
ed. J. H. Barkow, L. Cosmides, & J. Tooby. Oxford University Press. [arJMB]
Vidal, F. (1994a) Piaget before Piaget. Harvard University Press. [MF]
(1994b) Les myste`res de la douleur divine. Une prie`re du Jean Piaget pour
l’anne´e 1916. Revue De The´ologie et de Philosophie 126:97–118. [MF]
Viennot, L. (1979) Spontaneous reasoning in elementary dynamics. European
Journal of Science Education 1:205–21. [NLG]
Wall, P. D. (1999) The placebo and the placebo response. In: Textbook of pain,
4th edition, ed. P. D. Wall, pp. 1419–30. Churchill. [BMH]
Watson, P. (2005) Ideas: A history of thought and invention from fire to Freud.
Harper Collins. [JG]
Watson, R. I. (1973) Investigation into deindividuation using a cross-cultural
survey technique. Journal of Personality and Social Psychology
25:342–45. [rJMB]
Wedekind, C. & Braithwaite, V. A. (2002) The long-term benefits of human
generosity in indirect reciprocity. Current Biology 12:1012–15. [rJMB]
Weeks, M. & Lupfer, M. B. (2000). Religious attributions and proximity of
influence: An investigation of direct interventions and distal explanations.
Journal for the Scientific Study of Religion 39:348–62. [arJMB]
Wegner, D. M. (2005) Who is the controller of controlled processes? In: The new
unconscious, ed. R. Hassin, J. S. Uleman & J. A. Bargh, pp. 19–36. Oxford
University Press. [JP]
Wegner, D. M. & Sparrow, B. (2004) Authorship processing. In: The cognitive
neurosciences, 3rd edition, ed. M. Gazzaniga, pp. 1201–209. MIT Press. [JP]
Wegner, D. M., Sparrow, B. & Winerman, L. (2004) Vicarious agency: Experiencing
control over the movements of others. Journal of Personality and Social
Psychology 86:838–48. [JP]
Wegner, D. M. & Wheatley, T. P. (1999) Apparent mental causation: Sources of the
experience of will. American Psychologist 54:480–92. [JP]
Wellman, H. M. & Estes, D. (1986) Early understanding of mental entities: A
reexamination of childhood realism. Child Development 57:910–23. [DE]
Whitehouse, H. (2004) Modes of religiosity: A cognitive theory of religious
transmission. AltaMira Press. [rJMB]
Wiesel, E. (1961). Legends of our time. Holt, Rinehart, Winston. [aJMB]
(1966) The gates of the forest. Holt, Rinehart, Winston. [aJMB]
Wiggins, D. (1997) Sortal concepts: A reply to Xu. Mind and Language
12(3/4):413–21. [NJB]
(2001) Sameness and substance renewed. Cambridge University Press. [NJB]
Wilder, T. (1927/1955) The bridge of San Luis Rey. Washington Square
Press. [aJMB]
Wilkes, K. V. (1992) Psuche´ versus the mind. In: Essays on Aristotle’s De Anima, ed.
M. C. Nussbaum & A. O. Rorty, pp. 109–27. Oxford University Press. [MF]
Williams, B. (1970) The self and the future. The Philosophical Review
79(2):161–80. [NJB]
Williams, G. C. (1966) Adaptation and natural selection. Princeton University
Press. [JH]
Wittgenstein, L. (1953) Philosophical investigations. Blackwell. [SF]
Wojciszke, B., Bazinska, R. & Jaworski, M. (1998) On the dominance of moral
categories in impression formation. Personality and Social Psychology Bulletin
24(12):1251–63. [PB]
Worthen, L. T. & Yeatts, D. E. (2001) Assisted suicide: Factors assessing public
attitudes. Omega: Journal of Death and Dying 42:115–35. [aJMB]
Wright, R. (1994) The moral animal. Why we are the way we are: The new science of
evolutionary psychology. Pantheon. [aJMB]
Yalom, I. D. (1980) Existential psychotherapy. Basic Books. [aJMB]
Yang, S. C. & Chen, S.-F. (2006) Content analysis of free-response narratives to
personal meanings of death among Chinese children and adolescents. Death
Studies 30:217–41. [KRL]
Zilboorg, G. (1943) Fear of death. Psychoanalytic Quarterly 12:465–75.
[JG]
Zimbardo, P. G. (1969) The human choice: Individuation, reason and order, versus
deindividuation, impulse and chaos. In: Nebraska Symposium on Motivation,
vol. 17, ed. W. J. Arnold & D. Levine. University of Nebraska Press.
[rJMB]
References/Bering: The folk psychology of souls
498 BEHAVIORAL AND BRAIN SCIENCES (2006) 29:5