Religion as an Adaptive Complex 323
Received August 11, 2004; accepted October 28, 2004; final version received December 7, 2004.
Address all correspondence to Candace S. Alcorta, Department of Anthropology U-2176, University
of Connecticut, Storrs, CT 06269-2176. Email: candace.alcorta@uconn.edu
Human Nature, Winter 2005, Vol. 16, No. 4, pp. 323­359. 1045-6767/98/$6.00 = .15
Ritual, Emotion, and
Sacred Symbols
The Evolution of Religion as an Adaptive Complex
Candace S.Alcorta and Richard Sosis
University of Connecticut
This paper considers religion in relation to four recurrent traits: belief systems incorporating
supernatural agents and counterintuitive concepts, communal ritual,
separation of the sacred and the profane, and adolescence as a preferred developmental
period for religious transmission. These co-occurring traits are viewed as an
adaptive complex that offers clues to the evolution of religion from its nonhuman
ritual roots. We consider the critical element differentiating religious from nonhuman
ritual to be the conditioned association of emotion and abstract symbols. We
propose neurophysiological mechanisms underlying such associations and argue that
the brain plasticity of human adolescence constitutes an "experience expectant"
developmental period for ritual conditioning of sacred symbols. We suggest that
such symbols evolved to solve an ecological problem by extending communication
and coordination of social relations across time and space.
KEY WORDS Adolescence; Costly signals; Emotion; Neuropsychology; Religion;
Ritual; Symbolic thought
The evolution of religion and its possible adaptive function have been the subject
of considerable recent investigation by a wide array of researchers with
diverse theoretical and methodological approaches. Cognitive scientists and evolutionary
psychologists have been prominent among these researchers (Atran 2002;
Barrett 2000; Bering 2005; Boyer 2001; Bulbulia 2004a, 2004b; Guthrie 1993;
Kirkpatrick 1999; Mithen 1996, 1999).They have primarily studied religion in terms
of beliefs, uncovering the psychological mechanisms that produce supernatural
agents in all cultures. With the notable exceptions of Bering (2005) and Bulbulia
324 Human Nature / Winter 2005
(2004a), these researchers have concluded that religion constitutes a by-product of
cognitive adaptations selected for "more mundane" survival functions. Evolutionary
anthropologists have also revitalized studies of religion over the past two decades
(see Sosis and Alcorta 2003). In contrast to the cognitive scientists, however,
these researchers have tended to focus on religious behaviors rather than beliefs.
The primary debate among these investigators has centered on the relative importance
of group selection and individual selection in the evolution of religious systems
(Cronk 1994a, 1994b; Rappaport 1994; Sosis 2003a; Sosis and Alcorta 2003;
Wilson 2002). Drawing on both ethological studies and a rich theoretical legacy
beginning with Durkheim (1969), evolutionary anthropologists have proposed that
religious behaviors constitute costly signals that contribute to social cohesion (Cronk
1994a; Irons 1996a, 1996b, 2001; Sosis 2003b). These theorists situate religious
ritual within a broader, nonhuman evolutionary continuum related to socially adaptive
behaviors. Costly signaling theory has received empirical support from the
research of Sosis and colleagues (Sosis 2000; Sosis and Bressler 2003; Sosis and
Ruffle 2003, 2004), whose work has demonstrated a significant and positive association
between participation in religious ritual and enhanced cooperation. However,
these researchers have yet to examine how the high levels of cooperation
observed within religious communities (e.g., Sosis and Bressler 2003; Sosis and
Ruffle 2003) translate into individual fitness gains.
Although not guided by evolutionary analyses, the cumulative findings of a third
body of research that has emerged over the past two decades does provide evidence
of individual benefits for religious practitioners. This work has been conducted by
sociologists, epidemiologists, psychologists, and physicians, and has explored the
health impacts of religion on adherents (Hummer et al. 1999; Levin 1994, 1996;
Matthews et al. 1998; Murphy et al. 2000). Accumulating findings from this body
of research show significant positive associations between religious participation
and individual health. These studies demonstrate decreased mental and physical
health risks, faster recovery times for a wide variety of disorders, and greater longevity
for those who regularly attend weekly Western religious services, even when
social and lifestyle confounds are controlled (Hummer et al. 1999; Matthews et al.
1998; Murphy et al. 2000). In association with ongoing neurophysiological research
(Austin 1998; McNamara 2001, 2002; Newberg et al. 2001; Saver and Rabin 1997;
Winkelman 1986, 1992, 2000), these findings suggest proximate mechanisms by
which religious participation may impact psychoneuroimmunological systems and,
thus, individual fitness.
FOUR FEATURES OF RELIGION
These various approaches to religion have provided significant insights, but individually
each is insufficient for an evolutionary understanding of religion. A synthesis
that encompasses religion's cross-culturally recurrent features and captures
that which differentiates the religious from the secular is required. We propose that
Religion as an Adaptive Complex 325
religion may best be understood as an evolved complex of traits incorporating cognitive,
affective, behavioral, and developmental elements selected to solve an adaptive
problem. Here we focus on four cross-culturally recurrent features of all religions
that we consider to be integral components of this complex. These are:
* Belief in supernatural agents and counterintuitive concepts;
* Communal participation in costly ritual;
* Separation of the sacred and the profane; and
* Importance of adolescence as the life history phase most appropriate for the transmission
of religious beliefs and values.
These four elements emerge and reemerge throughout the anthropological and sociological
literature and encompass cognitive, behavioral, affective, and developmental
aspects of religious systems across a wide variety of cultures (Douglas 1966;
Durkheim 1969; Eliade 1958, 1959; Malinowski 1948; Rappaport 1999; Turner
1967, 1969; Tylor 1871). Although each trait may be variably expressed across different
socioecological systems, their recurrence in societies as diverse as totemic
Arunta hunter-gatherers and Protestant American industrialists suggests that they
constitute basic elements of religion.
In this paper we examine each of these traits in relation to an evolutionary theory
of religion as an evolved mechanism for social cooperation. We posit that the critical
element in the differentiation of religious from nonhuman ritual was the emergence
of emotionally charged symbols. Drawing on the seminal insights of Durkheim
(1969), Turner (1967, 1969), and Rappaport (1999), we propose proximate mechanisms
by which religious ritual serves to invest stimuli with motivational meaning.
The brain plasticity of extended human adolescence is examined as an "experience
expectant" developmental period for the emotional valencing of emergent symbolic
systems. Following Richerson and Boyd (1998) we conclude that the symbolic
systems of religious ritual in early human populations solved an ecological
problem by fostering cooperation and extending the communication and coordination
of social relations across time and space.
Supernatural Agents and Counterintuitive Concepts
Belief in supernatural agents may be the most commonly offered definition of
religion (see Sosis and Alcorta 2003). Durkheim (1969) was the first to propose
that supernatural agents represent the reification of society itself and function to
maintain social order.Although Durkheim's reification of society as a causal explanation
for religion has largely fallen into disfavor, his observation that the type of
agent represented in a society's religion reflects the social organization of that society
has been subsequently supported by the work of Wallace (1966) and the crosscultural
analyses of Swanson (1960).
More recently, Guthrie (1993), and other cognitive scientists (Atran 2002; Barrett
2000; Boyer 2001; Kirkpatrick 1999; Pinker 1997) have reexamined the supernatu-
326 Human Nature / Winter 2005
ral beliefs of religious systems and have concluded that such beliefs are merely a
"byproduct of numerous, domain-specific psychological mechanisms that evolved
to solve other (mundane) adaptive problems" (Kirkpatrick 1999:6). Rejecting any
adaptive function of religious beliefs per se, these researchers view the conceptual
foundations of religion as deriving from categories related to "folkmechanics,
folkbiology, (and) folkpsychology" (Atran and Norenzayan 2004). Supernatural
agents, similar to moving dots on computer screens or faces in the clouds, are simply
the result of innate releasing mechanisms of agency detection modules evolved
to respond to animate, and therefore potentially dangerous, entities (Atran and
Norenzayan 2004). Likewise, the attribution of intentionality to supernatural agents
is viewed as the application of folkpsychology mental modules evolved in response
to complex human social interactions. For many cognitive scientists, supernatural
agents, as well as religious beliefs in general, constitute little more than "mental
module misapplications."
Anthropological and psychological evidence, however, suggests that supernatural
agents of religious belief systems not only engage, but also modify, evolved
mental modules. Moreover, they do so in socioecologically specific and developmentally
patterned ways. Although agency detection modules probably do give rise
to the human ability to imagine a broad array of supernatural agents, those that
populate individual religions are neither random nor interchangeable. Whether supernatural
agents are envisioned as totemic spirits, ancestral ghosts, or hierarchical
gods is very much dependent upon the socioecological context in which they occur
(Durkheim 1969; Swanson 1960; Wallace 1966). The types of religious practitioners
present, as well as the nature of religious practices performed in a society, have
been shown to be significantly correlated with measures of social complexity and
integration (Bourguignon 1976;Winkelman 1986, 1992).The shamanic use of trance
to communicate with totemic ancestors found among the Athapaskan hunter-gatherers
of the Arctic would be very familiar to the desert-dwelling Arunta huntergatherers
ofAustralia. Likewise, the presence of priests and hierarchical gods typifies
religions of state-level agricultural societies from the Maya of Mexico to theAshanti
of Africa. Cross-cultural statistical research by Swanson (1960) and subsequent
analyses by Roes and Raymond (2003) have shown that the presence of moralizing
gods "who tell people what they should and should not do" is significantly and
positively related to group size, social stratification, environmental resource levels,
and extent of external conflict.
The supernatural beings of all these religious belief systems engage evolved
mental modules of agency and intentionality, as noted by cognitive scientists. This,
however, does not preclude the possibility that religion is an evolved adaptation.As
we have argued elsewhere (Sosis and Alcorta 2004), evolution is opportunistic and
necessarily co-opts existing traits to solve novel ecological problems. It is the modification
of these traits through natural selection that constitutes evolution.The question
to be posed, therefore, is not "Does religion incorporate preexistent mental
modules?" Instead, the relevant question is whether there exists evidence of adapta-
Religion as an Adaptive Complex 327
tion of those modules to solve ecological challenges. Recent experimental work by
developmental psychologists suggests that the answer to this question is "yes." The
supernatural agents of religious belief systems incorporate attributes of agency and
intentionality, but they also possess an additional attribute not shared with natural
category agents. In contrast to natural category agents, the supernatural agents of
religious belief systems are "full access strategic agents" (Boyer 2001). They are
"envisioned as possessing knowledge of socially strategic information, having unlimited
perceptual access to socially maligned behaviors that occur in private and
therefore outside the perceptual boundaries of everyday human agents" (Bering
2005:419). Moreover, accumulating research indicates that humans exhibit a developmental
predisposition to believe in such socially omniscient supernatural agents,
appearing in early childhood and diminishing in adulthood (Bering 2005; Bering
and Bjorklund 2004). Cross-cultural studies conducted with children between the
ages of 3 and 12 indicate that young children possess an "intuitive theism" (Kelemen
2004) that differentiates this social omniscience of supernatural agents from the
fallible knowledge of natural social agents (Bering 2005). As the child's theory of
mind develops, parents and other natural agents are increasingly viewed as limited
in their perceptual knowledge. Supernatural agents, however, not only remain socially
omniscient, but are viewed by children in late childhood as agents capable of
acting on such knowledge. This developmental predisposition to believe in socially
omniscient and declarative supernatural agents contrasts with evolved mental modules
of folkpsychology for natural categories. It also goes far beyond natural agencydetection
modules to encompass socially strategic agents with behaviorally
motivating characteristics.
Supernatural agents of religious belief systems also diverge from evolved mental
modules for natural ontological categories (e.g., animate/inanimate; people/animals)
in another significant way. Such agents do not uphold natural categories; they
violate them. Totemic animals that can talk, dead ancestors who demand sacrificial
offerings and visit the living, and incorporeal gods capable of being in all places at
all times violate basic premises of natural ontological categories.Yet, these exceedingly
unnatural constructs comprise powerful religious schema that elicit deep devotion
and belief across traditional and contemporary cultures alike. If religious
beliefs are merely by-products of mental modules evolved to deal with the "natural
world," why do such beliefs consistently violate the basic cognitive schema from
which they are presumed to derive?
In addressing this question, a number of cognitive scientists have noted that the
counterintuitive concepts that characterize religious beliefs are both attentionarresting
and memorable (Atran 2002; Boyer 2001; Kirkpatrick 1999). Experimental
tests validate these observations (Atran and Norenzyan 2004; Boyer and Ramble
2001). Counterintuitive concepts, such as bleeding statues and virgin births, do
grab attention. Atran and Norenzayan (2004) note, however, that the efficacy of
counterintuitive concepts in engaging attention, improving recall, and promoting
transmission is highly dependent upon the broader context within which these con-
328 Human Nature / Winter 2005
cepts are framed. Comparing belief sets with intuitive and counterintuitive concepts,
they found that the specific profile of the counterintuitive/intuitive concepts
most frequently encountered in religious belief systems achieved the "highest rate
of delayed recall and lowest rate of memory degradation over time" (Atran and
Norenzyan 2004:723). Thus, the counterintuitive beliefs of religious systems not
only violate natural ontological categories, they do so in a specifically patterned
way that renders them maximally memorable and maximally transmissible. This
suggests selection for such concepts.
Counterintuitive concepts have yet another important feature of significance for
social groups. In addition to their mnemonic efficacy, they comprise almost unbreakable
"codes" for the uninitiated. Most language distortions occur within ontological
categories (Bartlett [1932] as reported inAtran and Norenzayan 2004).When
distortions do cross ontological boundaries, they are most common from
counterintuitive to intuitive concepts; distortions occurring from intuitive to
counterintuitive concepts are extremely rare. For example, it would be much more
likely for a listener to modify "talking horse" to "walking horse" than the converse.
These findings indicate that counterintuitive concepts are not readily generated on
the basis of intuitive concepts, and they suggest that the chances of spontaneously
re-creating a preexistent counterintuitive concept are exceedingly low. This probability
is lowered even further by embedding multiple counterintuitive concepts
within religious belief sets. By incorporating counterintuitive concepts within belief
systems, religion creates reliable costly signals that are difficult to "fake." They
must be learned, and since such learning has been orally transmitted throughout the
vast majority of human evolution, this also implies participation in religious ritual.
As a result, religious belief systems serve as both costly and reliable signals of
group membership.
Finally, the irrationality of counterintuitive concepts contributes to their efficacy
as honest signals of commitment to a group who share that belief (Lee Cronk,
personal communication 2002). Within a pluralistic context, adherents who propound
counterintuitive beliefs risk censure. Early Christian belief in the resurrection
of Christ constituted a potent signal to Romans, Jews, and other Christians.
Only individuals knowledgeable about the religious tenets of Roman Catholicism
would conceive of the transmutation of wine to blood, and only those initiated into
the faith through the emotional conditioning of those tenets would truly believe that
such a transmutation occurs during the sacrament of Communion.Through the eyes
of nonadherents, such beliefs may be viewed as extraordinary and irrational. Such
perceptions contribute to both the costliness and the effectiveness of religious signals.
In summary, neither the content nor the structure of religious belief systems
supports the assertion that such beliefs constitute epiphenomenal "by-products."
Although supernatural agents engage mental modules of agency and intentionality
that evolved in response to "mundane" selection pressures, they modify these modules
in specific and developmentally patterned ways. Cross-culturally, supernatural
agents are integral elements of religious beliefs and they consistently reflect sig-
Religion as an Adaptive Complex 329
nificant socioecological relations within their respective cultures. The agents of
religious belief are not natural category agents, as would be predicted if they were
simply by-products of mental modules evolved to deal with such agents. They are
instead counterintuitive agents that not only modify natural agency module parameters,
but do so in consistently patterned and behaviorally significant ways. A developmental
propensity to believe that such agents are not only intentional, but also
socially omniscient, is indicated by accumulating experimental evidence (Bering
and Bjorklund 2004). Although the predisposition to believe in such supernatural
agents appears to be innate, the development of such beliefs is dependent upon
cultural transmission. Religious cognitive schema exhibit structural elements that
maximize transmission through the incorporation of minimally counterintuitive
concepts that engage attention, promote recall, and insure exclusivity.
These features of religious belief systems provide ontogenetic lability for the
construction of socially relevant moral systems across diverse ecologies, and they
do so within a structure that is maximally transmissible and minimally invasive.
Bering (2005:430) notes that "children are simultaneously immersed in unique cultural
environments where morality is chiefly determined by socioecological conditions.
Although there is likely a common `moral grammar'underlying all children's
development in this domain, the moral particulates of any given society are given
shape by the demands of local environments." Bulbulia, too, argues that religion,
like language, exhibits an innate grammar in which "development consists of fixing
labels to preexisting cognitive structures" (J. Bulbulia, personal communication
2004). For both Bulbulia (2004a, 2004b) and Bering (2005), the idea of socially
omniscient supernatural agency is a central component of this system. These researchers
view the adaptive value of such agents to be the maintenance of group
cooperation and cohesion across a broad spectrum of socioecologies. Atran, likewise,
acknowledges religion's use of supernatural agents in "maintaining the cooperative
trust of actors and the trustworthiness of communication by sanctifying the
actual order of mutual understandings and social relations" but asserts that "religion
has no evolutionary function per se"(2002:278­279). In contrast, we argue
that religion's ability to promote cooperation is its evolutionary function, and that
the costliness of religious ritual bears a direct relationship to the nature of the collective
action problems faced. When individual costs are high, but the potential
benefits of cooperation are great, costly religious ritual provides a reliable mechanism
for minimizing free-riding and maximizing cooperation. We consider the cognitive
schema of religious systems to be a fundamental evolved element in ensuring
such cooperation. Both the ontogenetic and structural features of religious belief
systems suggest evolved features. Yet, we also maintain that religious belief systems
in isolation are incapable of "sanctifying the actual order of mutual understandings
and social relations" (Atran 2002:278). It is certainly possible to be
cognizant of religious beliefs without subscribing to them, as any schoolchild who
has ever studied Greek mythology can attest. In order for religious beliefs to sanctify
socialrelations,theymustfirstthemselvesbesanctified.Thisisachievedthroughritual.
330 Human Nature / Winter 2005
Communal Participation in Costly Ritual
The pivotal role of communal ritual in religion has been noted by numerous
researchers (Bloch 1989; Bourguignon 1973, 1976; Durkheim 1969; Eliade 1958,
1959; McCauley 2001; Rappaport 1999; Turner 1967, 1969). The formality, patterning,
repetition, and rhythm of religious ritual have direct parallels in nonhuman
ritualized display (Laughlin and McManus 1979; Lorenz 1965; Rappaport 1999;
Rogers and Kaplan 2000; Smith 1979). In animal species such displays have evolved
to serve intra- and inter-specific communication functions (Dugatkin 1997; Lorenz
1965; Rogers and Kaplan 2000; Rowe 1999). On its most basic level, nonhuman
ritual constitutes "a process by which behavior specialized to be informative becomes
differentiated from behavior that is informative only incidentally to its other
functions" (Smith 1979:54). Ritualized displays represent but one end of a continuum
in animal signaling systems that also includes simple indexical signals. Signal
costs appear to be driven by both competition and receiver selection. Under
conditions of ambiguity, or when signals can readily be faked, costlier signals may
evolve to improve signal reliability. Ritualized displays are among the costliest of
animal signals in terms of time, energy, and somatic resources required of the signaler.
Zahavi (1975, 1981) has argued that such costly signals provide honest information
for receiver assessment since only those who are sufficiently fit can bear the
costs of such displays (Johnstone 2000; Zahavi and Zahavi 1997). Empirical research
supports this hypothesis (Johnstone 2000; Zahavai and Zahavi 1997). Laboratory
experiments indicate that the costliness of ritualized display is driven by
receiver selection for reliable signals (Rowe 1999).The formality, sequence, repetition,
and patterning that increase both time and energy costs of ritual also improve
the ability of the receiver to assess the reliability of the message transmitted. These
elements alert and focus attention, enhance memory, and promote associational
learning (Rowe 1999). They neurophysiologically "prime" both the sender and receiver
for action (Lewis and Gower 1980; Rogers and Kaplan 2000;Tinbergen 1965).
The type of action that results is dependent both upon the receiver's assessment of
the sender and upon the encoded "action releasers" embedded within the ritual
display (Lewis and Gower 1980; Tinbergen 1965).
Animal signals and signal responses show considerable ontogenetic and
socioecological malleability (Ball 1999; Lewis and Gower 1980; Marler 1999; Rogers
and Kaplan 2000; Wingfield et al. 1999). Although some species-specific signals,
such as the pecking response of herring-gull chicks to red dots, constitute relatively
fixed, environmentally stable action-response sequences (Lewis and Gower 1980;
Tinbergen 1965), others incorporate individually variant and ontogenetically learned
patterns, as seen in the male courtship songs of various bird species (Ball 1999;
Marler 1999). Signals of some species, including the esthetic nest constructions of
male bowerbirds (Dissanayake 1995) and the friendship greeting rituals of baboons
(Watanabe and Smuts 1999) show considerable malleability and high proportions
Religion as an Adaptive Complex 331
of environmentally variable behaviors. The continuum of simple to complex ritual
signals clearly encompasses a broad range of "fixed" and "learned" elements.
Ritual signals communicate important information regarding the condition, status,
and intent of the sender. The intensity of plumage coloration in birds, the pitch
of croaking in frogs, and the stotting height of springboks constitute indexical signals
that provide information regarding parasite load, size, and agility, respectively
(Krebs and Davies 1984; Rogers and Kaplan 2000). Such signals may also convey
information regarding intent. In many species, intent signals frequently involve the
transference of behaviors from their original context to a ritual context. The incorporation
of food begging displays in bird courtship rituals, and presentation of the
ano-genital area by subordinate primates to dominants, both represent signals that
have been emancipated from their original feeding and copulating behaviors and
transferred to new contexts of courtship and social hierarchies. In both instances,
affiliative responses eliciting approach behaviors are associated with the original
function of the signal and the transferred signal intent (Lewis and Gower 1980).
Transference of these signals from their original contexts to ritual communicates
the intent of the sender by evoking the autonomic and neurophysiological state
associated with the signal's origins. The incorporation of these intent signals in
ritual derives from their preexisting motivational characteristics (Laughlin and
McManus 1979).
Religious ritual, like nonhuman ritualized displays, is demarcated from ordinary
behaviors and is composed of the same structural elements (Rappaport 1999). Formality,
patterning, sequencing, and repetition are basic components of religious
ritual, and signals of condition, status, and intent constitute "action releasers" embedded
within that structure. Pan-human social signals of dominance and submission,
such as bowing and prostration, are prominent components of religious ritual
worldwide (Atran 2002; Bloch 1989; Boyer 2001; Leach 1966; Rappaport 1999).
As in nonhuman ritual, these signals convey information regarding status and intent.
Religious ritual also incorporates indexical and iconic signals. Masks, statues,
and other "agent" representations are prominent elements in religious ritual across
cultures. They engage innate mental modules evolved for mundane functions and
potentiate human predispositions to autonomically respond to specific classes of
stimuli, including animate agents and angry faces (LeDoux 2002). Incorporation of
evocative, grotesque, and dissonant features further intensify such responses. Like
the signals of nonhuman ritual, the signals of religious ritual clearly elicit neurophysiological
responses in participants and influence the nature of social interaction
(Lewis and Gower 1980; Reichert 2000; Rogers and Kaplan 2000; Sapolsky
1999). In contrast to nonhuman ritual, however, iconic, indexical, and ontogenetic
signals are not the primary encoded elements of human religious ritual. The fundamental
elements of human religious ritual are, instead, abstract symbols devoid of
inherent emotional or cognitive meaning. Words such as "Allah," the geometric
designs of Australian Dreamtime paintings, and religious beliefs do not, in and of
332 Human Nature / Winter 2005
themselves, elicit any innate or ontogenetically derived neurophysiological response.
Although, like language, religious systems across cultures appear to share a "deep
structural grammar" that has an ontogenetic basis, the specific symbols embedded
within that syntax are shaped by historical and socioecological parameters. In contrast
to the signals of animal ritual, the meaning of abstract religious symbols must
be created, both cognitively and emotionally. This important difference between
nonhuman ritual and human religious systems not only requires that the abstract
symbols of religious ritual be learned; it additionally requires that the emotional
and behavioral significance of these symbols be learned as well. Whereas animal
ritual elicits behavior through encoded signals, religious ritual elicits behavior to
encode symbols. The creation of these symbols provides ritual tools for the shaping
of social behaviors across space and time. Sosis (2003b) has argued that ritual participation
generates belief among performers. He examined the psychological mechanisms
underlying this process. Here we extend this argument to explain the
interrelationship between emotions, symbols, and the sacred, and describe the neurological
underpinnings of how ritual participation impacts belief.
Separation of the Sacred and the Profane
Religious ritual is universally used to define the sacred and to separate it from
the profane (Douglas 1966; Durkheim 1969; Eliade 1959; Rappaport 1999). As
noted by Rappaport (1999), ritual does not merely identify that which is sacred; it
creates the sacred. Holy water is not simply water that has been discovered to be
holy, or water that has been rationally demonstrated to have special qualities. It is,
rather, water that has been transformed through ritual. For adherents who have participated
in sanctifying rituals, the cognitive schema associated with that which has
been sanctified differs from that of the profane. For Christians, profane water conjures
associations of chemical structure and mundane uses; holy water, however,
evokes associations of baptismal ritual and spiritual cleansing. Of greater importance
from a behavioral perspective, the emotional significance of holy and profane
water is quite distinct. Not only is it inappropriate to treat holy water as one treats
profane water, it is emotionally repugnant. Although sacred and profane things are
cognitively distinguished by adherents, the critical distinction between the sacred
and the profane is the emotional charging associated with sacred things.
This distinction in emotional valence is created through participation in religious
ritual. Sacred symbols have distinct cognitive schema, but their sanctity derives
from their emotional meaning. It is the emotional significance of the sacred
that underlies "faith," and it is ritual participation that invests the sacred with emotional
meaning. The creation of religious symbols from abstract objects, and the
imbuing of these symbols with attributions of "awe," "purity," and "danger" (Douglas
1966), are consistent and critical features of religious ritual everywhere (Douglas
1966; Durkheim 1969; Rappaport 1999; Turner 1969). Why is this so?
Religion as an Adaptive Complex 333
Emotions Motivate Behavior. Accumulating research indicates that emotions constitute
evolved adaptations that weight decisions and influence actions. Emotions
"rapidly organize the responses of different biological systems including facial expression,
muscular tonus, voice, autonomic nervous system activity, and endocrine
activity" (Levenson 1994:123) in order to prepare the organism for appropriate
response to salient sensory stimuli. The ability of emotions to "alter attention, shift
certain behaviors upward in response hierarchies, and activate relevant associative
networks in memory" (Levenson 1994:123) directly impacts individual fitness. Since
emotions are generated from limbic cortices that are outside conscious control,
they are difficult to "fake" (Ekman et al. 1983). They, therefore, provide reliable
communication signals among conspecifics. EEG patterns for simulated and real
emotions are not the same, nor are the motor control areas for an emotion-related
movement sequence and a voluntary act (Damasio 1994, 1998; Ekman and Davidson
1993). The somatic markers of emotion, including such things as pulse rate, skin
conductance, pupil dilation, and facial expressions, differ from those under voluntary
control. Emotionally motivated smiles engage different muscles from "Duchenne
smiles," as do emotionally motivated frowns (Ekman 2003). As a result, emotions
constitute powerful and honest cues of state and intent (Ekman 2003; Ekman et al.
1983).
Emotions may be elicited by sensory stimuli both internal and external to the
organism. Predators, passing thoughts, and pulse rate are all capable of evoking
emotional response. The emotional processing and appraisal of these stimuli engage
widespread and complex cortical and subcortical systems within the brain.
Initial unconscious processing of stimuli occurs in subcortical structures of the brain,
including the basal ganglia, the amygdala, and the hypothalamus. This "first pass"
level of processing appears to incorporate a superordinate division based on positive/approach
and negative/withdrawal ratings of stimuli (Cacioppo et al. 2002).
Positive Stimuli Activate the Dopaminergic Reward System. The dopaminergic
reward system constitutes "an emotional system that has evolved to motivate forward
locomotion and search behavior as a means of approaching and acquiring
rewarding goals" (Depue et al. 2002:1071). This system originates in the ventral
tegmental area of the midbrain and projects to the nucleus accumbens of the ventral
striatum. Its activation triggers the release of dopamine (DA), a neuromodulator
which functions as a reward for the organism (Davidson and Irwin 2002). Stimuli
intrinsic to somatic and reproductive success, such as food and sex, activate dopamine
neurons within this system, and initiate goal-seeking behaviors.The potentiation of
dopaminergic neurons induces a positive motivational state in the organism and
simultaneously increases stimuli salience and locomotor activity (Pearson 1990).
"Activation of this system has been shown to function as a reward, and animals will
perform an arbitrary operant in order to self-administer stimulation of this pathway"
(Pearson 1990:503). Drugs of addiction potentiate this system, as do subjec-
334 Human Nature / Winter 2005
tively rated "pleasurable" activities (Cacioppo et al. 2002). Repeated potentiation
of this system transfers "the ability to phasically activate DA transmission from
incentive stimuli intrinsic to the goal to incentive stimuli extrinsic to it" (DiChiara
1995:95). This results in the assignment of a positive affective valence to stimuli
perceived under that state (DiChiara 1995). Such "incentive learning" creates associational
neural networks that link stimuli associated with rewarding experiences to
behavioral motivators, thereby investing previously neutral stimuli with positive
valence. For former drug addicts, the paraphernalia, settings, and even neighborhoods
associated with drug use constitute such incentive stimuli capable of activating
mesolimbic neural networks, as revealed through brain imaging studies (DiChiara
1995).
Negative StimuliActivate theAmygdala. The amygdala is a subcortical collection
of specialized nuclei located beneath the temporal cortex. A central function of the
amygdala is the rapid appraisal of potentially dangerous and threatening stimuli.
Activation of the amygdala initiates a cascade of specific neuroendocrine events
that prepare the organism to respond quickly to threats and danger.These responses
appear to be "hard wired" in the nervous system (LeDoux 2002). Animals with
lesioned or removed amygdala lack a fear response, even when placed in highly
dangerous situations (LeDoux 1996).
In humans, the amygdala is also pivotal in initiating fear responses. Humans
exhibit an innate predisposition to negatively valence potentially harmful and threatening
stimuli, including animate objects and angry or fearful faces. There is considerable
evidence that such stimuli elicit a greater response than positive stimuli,
particularly in relation to action tendencies (Ito et al. 2002). The amygdala also
processes human facial cues in relation to social judgments of trust. This processing
occurs both consciously and unconsciously by the left and right amygdala, respectively
(Adolphs 1999, 2002a, 2002b; Adolphs et al. 1998; Dolan 2000; Morris
et al. 1998; Oram and Richmond 1999).
While specific stimuli innately activate the amygdala, it is also possible for neutral
stimuli to acquire negative valence through classical and contextual conditioning.
Previously neutral stimuli that are present or otherwise associated with a
negatively valenced stimulus that activates the amygdala may subsequently initiate
such response themselves. Once such conditioning occurs, it is difficult to reverse.
Extinction of such conditioning "is not a process of memory erasure, [but rather]
involves cortical inhibition of indelible, amygdala-mediated memories" (LeDoux
2002:404). As a result of both the negativity bias in information processing and the
indelible nature of emotional memory, amygdala-conditioned stimuli constitute
powerful long-term elicitors of emotional response.
The amygdala is highly interconnected with sensory, motor, and autonomic output
systems. These interconnections "provide an anatomical basis for adaptive responses
to stimuli" (Dolan 2000:1117). Interconnections with the hypothalamus
ensure rapid somatic responses to stimuli through a cascade of neuroendocrine events.
Religion as an Adaptive Complex 335
These events prepare the organism for behavioral response and provide feedback
information regarding body state to the amygdala. Reciprocal interconnections with
the nuclear basalis ensure amygdalar participation in cortical arousal and selective
attention. Direct interconnections of the amygdala with the hippocampal formation
allow affective modifications of spatial behavior (Cacioppo et al. 2002; Cardinal et
al. 2002; Damasio 1994, 1998; LeDoux 2002). Specific reciprocal projections from
the amygdala to other emotional processing regions, including the ventral striatum
and brainstem nuclei, provide an important link between positive and negative affective
systems (Dolan 2000; LeDoux 2002; Rolls 1998). It is, however, the direct
interconnections between the amygdala and the prefrontal cortex that are of particular
significance for human social and symbolic systems (Deacon 1997;
Groenewegen and Uylings 2000; Rolls 1998).
The Prefrontal Cortex Plays a Critical Role in Decision Making. McNamara (2001,
2002) has convincingly argued that the self-responsibility, impulse control, and
morality which religions seek to instill in adherents are frontal lobe functions. Ongoing
research supports the pivotal role of the prefrontal cortex in social judgment
and impulse control, as well as symbolic thought (Deacon 1997; Dehaene and
Changuex 2000; McNamara 2001, 2002; Rolls 1998). The orbitofrontal (OFC) region
of the prefrontal cortex is the area of the brain activated in anticipation of
rewards and punishments. Injuries to this brain area affect the delicate calculus of
personal interest, environmental contingencies, and social judgments that motivate
and guide individual behaviors within a social group (Dehaene and Changeux 2000;
Rolls 1998). The valuation of behavioral alternatives, particularly in relation to social
behaviors, appears to be processed in the OFC. Impairments to this area correlate
highly with socially inappropriate or disinhibited behavior (Anderson et al.
2002; Damasio 1994; Kolb et al. 2004).
The behavioral deficits of OFC impaired patients are also seen in individuals
who have intact prefrontal cortices and intact amygdala but lack interconnections
between the two (Damasio 1994; LeDoux 1996, 2002). These individuals perform
well on abstract reasoning tasks but are unable to apply such reasoning to personal
decision making (Damasio 1994; LeDoux 1996, 2002).The loss of emotion typical
of OFC impaired patients is also a characteristic of these disconnect patients. For
these individuals, the affective cues required for valuation of predicted outcomes
are absent. In the absence of emotional input from the basolateral amygdala, the
OFC lacks valuation information necessary for the prediction of reward/punishment
outcomes. Recent laboratory experiments conducted by Schoenbaum and
colleagues demonstrate that both the orbitofrontal cortex and the basolateral
amygdala are "critical for integrating the incentive value of outcomes with predictive
cues to guide behavior" (Schoenbaum et al. 2003:855). It is through the emotional
inputs of the amgydala that "otherwise neutral cues acquire motivational
significance or value through association with biologically significant events"
(Schoenbaum et al. 2003:863).
336 Human Nature / Winter 2005
Ritual, Emotion, and Sanctification. Religious rituals are biologically significant
events. Ongoing research with ritual participants engaged in meditation and trance
demonstrate changes in brain wave patterns, heart and pulse rate, skin conductance,
and other autonomic functions (Austin 1998; Davidson 1976; Kasamatsu and Hirai
1966; MacLean et al. 1997; Mandel 1980; Newberg et al. 2001; Winkelman 2000).
Meditation also alters neuroendocrine levels, including testosterone, growth hormone,
and cortisol (MacLean et al. 1997). Although little research has been conducted
on the neurophysiological effects of less intense religious participation, there
is mounting evidence that participation in weekly Western religious services may
impact blood pressure (Brown 2000; Dressler and Bindon 2000), adolescent testosterone
levels (Halpern et al. 1994), and other neurophysiological systems (Levin
1994, 1996; Matthews et al. 1998; Murphy et al. 2000). Experiments suggest that
some of these neurophysiological changes may be associated with the "rhythmic
drivers" that characterize human religious ritual.
Music Is a Universal Feature of Religious Ritual. Human and nonhuman ritual
share basic structural components of formality, pattern, sequence, and repetition.
Human religious ritual further amplifies and intensifies these elements through the
incorporation of "rhythmic drivers." Described by Bloch as "distinguishing marks
of ritual" (1989:21), these elements, including music, chanting, and dance, constitute
recurrent and important components of religious ritual across cultures.Although
Bloch derived these features from ethnographies of traditional societies, the recent
survey of U.S. congregations conducted by Chaves et al. (1999) found music to be
a consistent feature of contemporary U.S. religious services, as well. Even in the
most ritually constrained religions, music remains a key consistent feature (Atran
2002). Not only is music an important component of religious ritual, across traditional
cultures it is inseparable from it (Becker 2001).
Music has important neurophysiological effects. As a "rhythmic driver," it impacts
autonomic functions and synchronizes "internal biophysiological oscillators
to external auditory rhythms" (Scherer and Zentner 2001:372). The coupling of
respiration and other body rhythms to these drivers affects a wide array of physiological
processes, including brain wave patterns, pulse rate, and diastolic blood
pressure (Gellhorn and Kiely 1972; Lex 1979; Mandel 1980; Neher 1962; Walter
and Walter 1949).This "coupling effect" has been shown to be present in humans at
a very early age (Scherer and Zentner 2001). Music amplifies and intensifies this
effect through the use of instruments, or "tools," thereby providing a means of synchronizing
individual body rhythms within a group. Recent work by Levenson (2003)
has shown that synchronized autonomic functions, including such things as pulse
rate, heart contractility, and skin conductance, are positively and significantly associated
with measures of empathy. The prominent role of music in religious ritual
promotes such empathy.
Music also has demonstrated effects on measures of stress and immunocompetence.A
significant negative correlation between exposure to "relaxing" music and
Religion as an Adaptive Complex 337
salivary cortisol levels was found in experiments conducted by Khalfa et al. (2003).
Other research has demonstrated significant positive correlations between music
and immunocompetence, as measured by salivary immunoglobin A (SIgA), with
active participation correlating most highly with immunocompetence and no music
exposure correlating the least (Hirokawa and Ohira 2003; Kuhn 2003). These associations
between music and measures of stress and health may be mediated by music's
ability to alter autonomic functions and evoke emotions. The capacity of music to
alter skin temperature, muscle tension, cardiovascular function, respiration, norepinephrine,
and brain wave patterns all have subjectively reported "emotion inducing
effects" (Hirokawa and Ohira 2003; Scherer and Zentner 2001).The contour, rhythm,
consonance/dissonance, and expectancy within a musical structure contribute to
both the intensity and valence of the experienced emotion (Hirokawa and Ohira
2003; Scherer and Zentner 2001; Sloboda and Juslin 2001). Studies of subliminal
facial expression demonstrate that musically induced physiological changes closely
correspond with both involuntary facial expressions of emotion and subjectively
described emotions evoked by particular types of music (Krumhansl 1997).
The capacity of music to entrain autonomic states and evoke congruent emotions
in listeners provides the basis for creating and synchronizing motivational
states in ritual participants. Although the communal songs and vocalizations of
nonhuman species, including birds, whales, and wolves, may also function in social
accommodation, only human music is capable of amplifying, intensifying, and
modifying these effects through the use of "tools." The externalization of auditory
signal production through the use of musical instruments fundamentally alters the
signal/signaler relationship. The signal produced through the use of musical instruments
is no longer indexical of either the signaler's state or condition. Two warriors
can sound like twenty through the use of drums. Moreover, discrete sounds produced
with musical instruments can be manipulated and juxtaposed to create emotionally
evocative signals independent of the musician's state. Like the phonemes,
words, and sentences of language, the use of musical instruments to produce sounds
permits the combining of such sounds to create emotionally meaningful signals.
These, in turn, can be arranged and rearranged within encompassing musical structures.
The formality, sequence, pattern, and repetition of such musical structures
themselves elicit emotional response through their instantiation of ritual. Music
thereby creates an emotive "proto-symbolic" system capable of abstracting both
the signals and structure of ritual.This abstraction and instantiation of ritual through
music may well have established the foundation for symbolic thought in human
evolution. It certainly provided a tool for the evocation of communal emotions across
time and space.
Religious Ritual Evokes Both Positive and Negative Emotions. Cross-culturally,
the emotion most frequently evoked by music in religious ritual is happiness (Becker
2001). In its most intense version, this may reach ecstasy. Such extreme joy "almost
by definition involves a sense of the sacred" (Becker 2001:145) and is not unlike
338 Human Nature / Winter 2005
that attained through use of various psychoactive drugs. Such drugs also constitute
prominent elements in many religious systems.These components of religious ritual
activate noradrenergic, serotonergic, and dopaminergic systems in the brain that
heighten attention, enhance mood, and increase sociability (Regan 2001). These
components of religious ritual elicit positive emotional responses in participants
and engage the brain's dopaminergic reward system.
There are also numerous elements of religious ritual that evoke fear and pain
rather than happiness and joy (Douglas 1966; Eliade 1959; Glucklich 2001; Turner
1967, 1969). Many ritual settings, including caves, caverns, and cathedrals, arouse
vigilance by altering sensory perception through unpredictable illumination. Grotesque
masks, bleeding statues, and fearsome icons engage innate "agency" modules
that initiate emotional responses to danger and threat. Physical and mental
ordeals inflict suffering and alter autonomic states.Vengeful gods and demons mete
out punishment and demand painful sacrifices. Such negative stimuli comprise central
elements of many religious systems and are particularly prevalent within the
context of rites of passage (Eliade 1958, 1959; Glucklich 2001; Turner 1969). In
contrast to the positive affect induced by ecstatic religious ritual, these components
of ritual initiate responses related to fear and danger and evoke intense negative
emotions in ritual participants.
The ability of religious ritual to elicit both positive and negative emotional responses
in participants provides the substrate for the creation of motivational communal
symbols. Through processes of incentive learning, as well as classical and
contextual conditioning, the objects, places, and beliefs of religious ritual are invested
with emotional significance. The rhythmic drivers of ritual contribute to
such conditioning through their "kindling effects." Research on temporal lobe syndrome
patients has shown that repeated neuronal firing of the amygdala can result
in the conditioned association of arbitrary stimuli with heightened emotional significance
(Bear 1979; Bear et al. 1981; Damasio 1994; Geschwind 1979). The increased
religiosity characteristic of some temporal lobe epileptics has been attributed
to this kindling effect (Bear 1979; Saver and Rabin 1997). Rhythmic environmental
stimuli, including both music (Peretz 2001) and rapid, flashing lights, contribute to
the rapid neuronal firing that results in such kindling (LeDoux 2002). Temporal
lobe patients have a low threshold for such firing. EEG recordings have shown that
the driving effects of ritual, such as music, drumming, and dancing, are capable of
altering neuronal firing patterns in nonclinical populations, as well (Lex 1979; Neher
1962; Walter and Walter 1949). Elements of religious ritual that increase neuronal
firing rates prime ritual participants for the conditioned association of symbols and
emotions, both positive and negative, and create communal conditions for investing
religious stimuli with these emotions (DiChiara 1995). The "ecstasy" achieved
through the music and movement of Sufi dancing is transferred to the religious
poetry with which it is associated. Likewise, ingestion of peyote by the Huichol
Indians with its potentiation of the dopaminergic reward system provides a neurophysiological
basis for investing the communal Peyote Hunt itself with sacred sig-
Religion as an Adaptive Complex 339
nificance (Myerhoff 1974). The negative emotional responses elicited by shadowed
cathedrals, fearsome masks, and painful ordeals are heightened by drumming, music,
and chants.The emotions thereby elicited and intensified become conditionally
associated with the gods, ghosts, and demons that populate religious belief systems.
Such symbols are not inherently pleasurable, but they are motivationally powerful
and emotionally indelible. The use of communal ritual to invest previously
neutral stimuli with deep emotional significance creates a shared symbolic system
that subsequently valences individual choices and motivates behavior (Dehaene and
Changeux 2000).
Most Religions Incorporate Both Positive and Negative Elements. The extent to
which positive and negative elements are emphasized varies considerably both across
the rituals within a given religion and among religions. Whether religious ritual
predominantly incorporates positively or negatively valenced symbols appears to
be correlated with both the political characteristics of the group and the risk-tobenefit
ratio of their cooperative endeavors. We anticipate that when collective action
issues are predominantly problems of coordination with few potential costs to
individuals, positively valenced rituals will serve to promote affiliative cooperation.
Such rituals engender empathy among participants and conditionally associate
religious symbols with internal reward systems through incentive learning.When
the predominant collective action issues faced by a group involve high individual
costs but potentially great collective benefits, however, we expect increases in the
costliness of religious ritual through the incorporation of negatively valenced stimuli
to deter free riders. Since negatively valenced components of ritual are motivationally
more powerful than positive stimuli, they provide a more reliable emotionally
anchored mechanism for the subordination of immediate individual interests to
cooperative group goals. In societies lacking a central political authority with police
powers capable of subordinating individual interests to those of the group, intense
and negatively valenced religious rituals address the inherent free-rider
problems of collective action. The prominence of negatively valenced elements in
religions associated with large, socially stratified, preindustrial societies (Roes and
Raymond 2003) underscores this "policing" role of religion in motivating cooperation
when a central secular authority is weak (Paige and Paige 1981). This is particularly
pronounced in adolescent rites of passage in such societies (Eliade 1958;
Glucklich 2001; Turner 1969). The incorporation of painful and dangerous elements
in such rites is positively and significantly correlated with the incidence of
warfare in preindustrial societies (Sosis et al., n.d.). These highly charged negative
ritual experiences not only bond initiates, they also motivate intense cooperation
and obedience under conditions of high individual risk and low central authority.
The less powerful in such societies bear a larger share of the fitness costs of such
subordination, but they may still gain greater benefits as members of a successful
cooperative group than they would otherwise realize.
Yet, even when religious systems emphasize negatively valenced symbols, the
340 Human Nature / Winter 2005
use of ritual to invest such symbols with emotional meaning necessarily incorporates
positively valenced components that benefit ritual participants both psychologically
and politically. Powerfully valenced symbols that motivate behavioral
choices reduce cognitive dissonance, particularly under conditions of socioecological
stress. Research by Bradshaw (2003) indicates that in contemporary Western societies,
weekly worship attendance results in relatively greater decreases in psychological
distress for the socioeconomically disadvantaged. The positive correlation
of music and immunocompetence, and its inverse correlation with stress, suggests
that ritual participation may differentially benefit group members facing the highest
stress loads.
At the same time, joint participation in costly ritual creates empowering conditions.
Ritual not only promotes more efficient and effective group functioning for
politically and socially sanctioned endeavors, it simultaneously creates motivationally
coordinated coalitions that can surmount existing in-group/out-group boundaries
and provide a mechanism for social and political change (Bourguignon 1973).
The Protestant Reformation of the sixteenth century, the role of African-American
churches in the U.S. Civil Rights Movement, the contemporary importance of
Pentecostalism in Latin America, and messianic movements in general, all illustrate
the important role of religion in creating cooperative coalitions that have been
instrumental in transforming existent social and political relationships.
Adolescence and Religion
Adolescent Rites of Passage. Adolescent rites of passage are one of the most consistent
features of religions across cultures (Bettleheim 1962; Brown 1975; Lutkehaus
and Roscoe 1995; Paige and Paige 1981; van Gennep 1960). In some societies,
such as the Yamana and Halakwulup of Tierra del Fuego, such rites traditionally
consisted of little more than oral transmission of sacred knowledge from elder to
youth (Eliade 1958). In other cultures, such as the Ndembu and the Elema, pubertal
initiation rites involved "kidnapping" of adolescents, months of sequestered seclusion,
and ritual ordeals that included dietary restrictions, sleep deprivation, physical
pain, and genital mutilation (Eliade 1958; Glucklich 2001; Paige and Paige
1981; Turner 1969; van Gennep 1960). In modern societies, adolescence also constitutes
an important developmental period for religious training (Atran 2002; Elkin
1999; Regnerus et al. 2003). Although the intensity and duration of adolescent rites
of passage vary from culture to culture, all share a common structure (Turner 1969;
van Gennep 1960), as well as a common emphasis on the evocation of emotion and
its association with symbols in the teaching of sacred things (Eliade 1958; Turner
1967, 1969).
The expressed purpose of rites of passage is to initiate particular categories of a
society's adolescents into "the sacred." Initiates not only learn the sacred, they live
it. The social and psychological death, transformation, and rebirth of the individual
achieved through these rites not only train initiates, but transform them as well
Religion as an Adaptive Complex 341
(Turner 1967, 1969). Initiates enter as children but leave as adults invested with
both social and reproductive rights, as well as the responsibilities entailed therein.
Through rites of passage initiates learn what things constitute the sacred. This
requires the development of new cognitive schema for previously mundane things,
whether words, images, or objects, involving the generation of new neural associative
networks. More importantly, however, initiates directly experience the sacred.
The separation; sleep and food deprivation; exposure to novel, dangerous, and terrifying
stimuli; and subjection to physical and mental ordeals that are frequently an
integral part of such rites evoke autonomic and emotional responses in initiates.
Rites of passage purposefully engage unconscious emotional processes, as well as
conscious cognitive mechanisms. The conditioned association of such emotions as
fear and awe with symbolic cognitive schema achieved through these rites results in
the sanctification of those symbols, whether places, artifacts, or beliefs. Because
such symbols are deeply associated with emotions engendered through ritual, they
take on motivational force.When such rites are simultaneously experienced by groups
of individuals, the conditioned association of evoked emotions with specific cognitive
schema creates a cultural community bound in motivation, as well as belief.
Adolescent Brain Development. Adolescence may constitute a neurophysiologically
sensitive developmental period for the learning of abstract concepts and the
conditioned association of such concepts with emotions (Kolb et al. 1998; Kwon
and Lawson 2000; Plant 2002; Spear 2000). The human brain demonstrates great
plasticity during development. Infancy, childhood, adolescence, and adulthood are
marked by differentiated growth patterns in various brain cortices and nuclei (Casey
et al. 2000; Giedd et al. 1999; Keshavan et al. 2002; Kolb and Whishaw 1998; Kolb
et al. 1998; Kwon and Lawson 2000; LeDoux 2002; Plant 2002; Sowell et al. 1999;
Spear 2000; Walker and Bollini 2002). The differential patterns of brain growth
across the life course create sensitive periods for particular types of learning
(Greenough 1986). Early childhood language acquisition is an example of such
"experience expectant" learning (Pinker 1997). We propose that adolescence constitutes
a second critical period of "experience expectancy" for the learning of emotionally
valenced symbolic systems.
The Adolescent Brain Does Not Mature Uniformly. Whereas the preadolescent
brain grows through an increase in cortical gray matter, during adolescence synaptic
pruning eliminates as much as one-half of the number of cortical synapses per
neuron (Spear 2000). Synapse elimination does not occur uniformly throughout the
human cortex, however. Frontal and parietal lobes follow a similar developmental
trajectory, with increases in gray matter up to a maximum occurring at 12.1 and
11.8 years, respectively, for males and 11.0 and 10.2 years, respectively, for females,
followed by a decline, resulting in a net decrease in volume across adolescence.The
growth of temporal lobe gray matter has also been found to be nonlinear,
with maximum size reached at 16.5 years for males and 16.7 years for females, and
342 Human Nature / Winter 2005
slight declines thereafter (Giedd et al. 1999; Sowell et al. 1999). Both frontal and
temporal lobe maturation occurs late in development and is completed in early adulthood
(Keshavan et al. 2002; Sowell et al. 1999). This heterochronous adolescent
loss of cortical gray matter is accompanied by increased volume in amygdalar and
hippocampal nuclei. Concurrent changes in white matter density facilitate the propagation
of electrical signals and increase the speed of neural transmission (Keshavan
et al. 2002; Walker and Bollini 2002).
These changes streamline brain function by eliminating irrelevant interconnections
and enhancing those that remain. This ontogenetic sculpting of the brain results
from differential activation of specific neurons on the basis of experience in
the accommodation of environmental needs (Greenough and Black 1991; Kolb and
Whishaw 1998; Kolb et al. 1998; LeDoux 2002). Kolb and colleagues note that
"experience can alter different parts of neurons differently (and) . . . changes in
synaptic organization are correlated with changes in behavior" (1998:156). As a
result, "the environment or activities of the teenager may guide selective synapse
elimination during adolescence" (Giedd et al. 1999:863). Emotionally evocative
experiences that occur during adolescence may, therefore, actually shape neural
networks in the maturing brain. This is particularly true for brain areas such as the
temporal lobes and prefrontal cortices undergoing maturation.
The maturation of the prefrontal cortex that occurs during adolescence has important
implications for abstract reasoning abilities and symbolic thought.The prefrontal
cortex (PFC) is "essential for such functions as response inhibition, emotional
regulation, planning and organization" (Sowell et al. 1999:860).The interconnectivity
of the PFC with nearly all other brain regions uniquely situates this cortical structure
in its ability to associate diverse stimuli (Groenewegen and Uylings 2000;
Robbins 2000; Rolls 1998). Maturation of the prefrontal cortex during adolescence
provides the neurophysiological substrate for social cognition, abstract reasoning,
and symbolic thought (Adolphs 2002a; Deacon 1997; Robbins 2000).
There Is a Shift in the Dopaminergic Reward System during Adolescence. Significant
changes in neurotransmitter systems occur during adolescence. Receptors
for dopamine, serotonin, acetycholine and GABA (-aminobutyric acid) are pruned
from their preadolescent over-production, and limbic areas, including the hippocampus,
also undergo pruning of excitatory receptors. Hippocampal receptors for
endogenous cannabinoids peak during adolescence at higher than adult levels (Spear
2000). Studies by Carlson et al. (2002) demonstrate increased long-term potentiation
as a result of endocannabinoid production, suggesting enhanced memory functions
during this period. Concurrent with the decline in excitatory neurotransmitter
receptors during adolescence, a shift in dopamine balance from mesolimbic to
mesocortical regions occurs. This shift impacts reward learning and has significant
behavioral implications (Schultz et al. 2002). Dopamine inhibitory input to the prefrontal
cortex is greatest during adolescence, whereas dopamine activity in the anterior
cingulate cortex and other subcortical regions, including the amygdala, is
Religion as an Adaptive Complex 343
lowest. While dopamine activity in the anterior cingulate cortex is under inhibitory
control of the amygdalar dopamine system, the amygdala is, in turn, tonically inhibited
by prefrontal cortex activity.According to Walker and Bollini, "the enhancement
of neuronal connection between the cortex and limbic regions may play a role
in the integration of emotional behaviors with cognitive processes" (2002:18) during
this time. The shifting dominance of amygdalar dopamine projections from
anterior cingulate cortex to the prefrontal cortex during adolescence impacts both
conditioned associations and the intrinsic reward system. In addition to cortical
maturation during adolescence, MRI studies have shown differences in the activity
of the amygdala in adolescents, as compared with adults. Human adolescents exhibited
"greater brain activity in the amygdala than in the frontal lobe when engaged
in a task requiring the subjects to identify emotional state from facial
expressions, while adults conversely exhibited greater activation in frontal lobe than
amygdala when engaged in the same task" (Spear 2000:440).
Adolescent Changes in Brain Function Have Important Implications for Learning
and Behavior. The concurrent maturation of the temporal lobe and amygdala are
relevant to facial recognition and social judgments (Adolphs et al. 1998). Studies
indicate that the amygdala mediates judgment of other people's social behavior,
particularly with regard to approachability and trustworthiness (Adolphs 2002a,
2002b; Cardinal et al. 2002). The shift in the dopaminergic reward system from
mesolimbic to mesocortical dominance that occurs during adolescence provides a
unique developmental window for the conditioned association of abstract symbols
with intensely experienced emotions and for the integration of these associations
with both social interactions and symbolic thought. Heightened adolescent sensitivity
to stressors amplifies this process (Spear 2000). The synaptogenesis and neurotransmitter
shifts occurring during adolescence intensify the impacts of
environmental stimuli experienced during this developmental phase. This is particularly
true for the late-maturing frontal and temporal cortices, and for such limbic
nuclei as the amygdala and the hippocampus.The specific changes occurring in
the adolescent brain render this a particularly sensitive developmental period in
relation to social, emotional, and symbolic stimuli. These are precisely the type of
stimuli of greatest importance in adolescent rites of passage.
Adolescent Rites of Passage Bombard Initiates with Environmental Stimuli that
Engage Prefrontal, Temporal, and Limbic Functions. The ritual components of
these rites optimize stimulus impacts while amplifying the kindling effects of the
stimuli through rhythmic drivers, including music, chanting, and dance, all of which
may be particularly salient for adolescents. Intensification of the stimuli through
sleep and food deprivation, fear, physical ordeals, and drugs can be expected to
increase the neurophysiological impacts in terms of memory, reward learning, and
emotional charging of stimuli. The "breaking down" of initiates during the liminal
phase of adolescent rites of passage engenders a common autonomic state among
344 Human Nature / Winter 2005
initiates. The empathy and shared emotional charging experienced in rites of passage
valence the cognitive schema associated with sacred things.
Not all schemata constitute equal candidates for sanctification. Accumulating
evidence suggests a developmental propensity for schema incorporating socially
omniscient and declarative supernatural agents (Bering 2005). Moreover, schema
of religious belief systems exhibit consistent structural features. Dichotomies, inversions,
and counterintuitive concepts are consistent elements of this structure
(Atran 2002; Atran and Norenzayan 2004; Boyer 2001; Boyer and Ramble 2001;
Levi-Strauss 1963). The cognitive schema of religious systems also consistently
incorporate the salient socioecological features of the society in which they occur,
yet they do so while transcending the momentary, individual, and specific attributes
of those features. Both the abstraction of social relations and their transformation
into eternal truths are the hallmarks of religious schema (Rappaport 1999). These
schema leave no outwardly visible signs but instead carve their indelible mark on
the very minds of initiates. Through adolescent rites of passage, "the abstract is
made alive and concrete by the living substance of men and women" (Rappaport
1999:148).
THE EVOLUTION OF RELIGION
Many recent evolutionary studies define religion in terms of cognition, focusing on
the beliefs rather than the behaviors of religious systems. From a cross-cultural
perspective, however, it is ritual that lies at the heart of all religions (Durkheim
1969; Eliade 1958, 1959; Rappaport 1999; Turner 1967, 1969), and it is participation
in ritual that creates believers (Sosis 2003b). In the absence of ritual indoctrination
and practice, religious beliefs lack both emotional salience and motivational
force.
Ritual in nonhuman species functions to communicate social information and to
coordinate social behaviors through the use of species-specific signals evolved to
elicit neurophysiological responses in participants (Dugatkin 1997; Rogers and
Kaplan 2000; Rowe 1999; Wingfield et al. 1999). Although ritual displays may be
costly in terms of time, energy, and somatic expenditures, they provide information
to participants that can impact individual fitness. By providing reliable signals,
ritual allows accurate assessment of conspecific condition and intent (Zahavi and
Zahavi 1997). It also "primes" participants for social interaction. Ritual winners
reap resource and mating advantages; losers, however, also benefit from reductions
in conflict achieved through ritual. Within the context of social groups, ritual further
functions to decrease individual stress through the stabilization of social organization
(Sapolsky 1999) and provides a means of facilitating both group fission/
fusion and the coordination of group activities (Dugatkin 1997; Goodall 1986;
Laughlin and McManus 1979; Rogers and Kaplan 2000). The pre-hunt ritual of
wolves represents such coordination, and the friendship rituals of chimps and baboons
have been observed to facilitate cooperative alliances that force changes in
Religion as an Adaptive Complex 345
troop hierarchies (Goodall 1986;Watanabe and Smuts 1999).Among human groups,
these same functions are apparent in the rituals of both sports and politics.
Religious ritual, too, functions to communicate and coordinate social behaviors
and does so through the elicitation of neurophysiological responses. Participation
in religious ritual results in empirically demonstrated effects on both cooperation
(Sosis and Bressler 2003; Sosis and Ruffle 2003, 2004) and individual health and
longevity (Hummer et al. 1999; Matthews et al. 1998; Murphy et al. 2000). Like the
ritualized displays of nonhuman species, religious ritual is positively associated
with decreased stress and improved immunological function (Murphy et al. 2000).
Bradshaw (2003) has further found that decreases in psychological distress associated
with participation in religious ritual may be particularly relevant for the relatively
deprived. Under conditions of inequality, religious ritual may, thus, confer
direct fitness benefits for participants while simultaneously providing a mechanism
for cooperative action for political change. The parallels between nonhuman
and religious ritual extend, as well, to the use of religious ritual in the reintegration
of social groups across cultures, and the coordination of group endeavors. Nonhuman
and religious ritual clearly share important structural and functional elements
selected for their adaptive value in social communication (Rowe 1999). The
two are, however, separated by a critical distinction. While nonhuman ritual encodes
signals as neurophysiological primes for behavior, religious ritual encodes
symbols created through the ritual process itself.
Although it is impossible to retrace ritual's evolution to a symbolic signaling
system, the "distinguishing marks of ritual"--chanting, music, and dance--may
provide important clues.As discussed above, all religions incorporate music in some
form, and in most it is a dominant element. Music is uniquely adapted to instantiate
the structure of ritual precisely because it incorporates the formality, sequencing,
patterning, and repetition that define ritual. As a result, it is able to elicit the neurophysiological
responses associated with such ritual in the absence of ritual behaviors.
Music's direct impacts on autonomic function, its ability to enhance
immunocompetence (Kuhn 2002), and its role in entraining ritual participants may
all have led to its selection as a fundamental component of early hominid ritual.
Ultimately, however, the most important evolutionary consequence of music may
well have been its "proto-symbolic" attributes. The ability of music to abstract and
codify ritual meaning over time and space may have been the critical first step
toward symbolic thought. The introduction of such a symbolic ritual system introduced
a new type of cognition in hominid evolution. The use of ritual to create
associational neural networks linking symbolic, social, and affective systems provided
social groups with a highly flexible tool for motivating individual behavior,
forging inter-group alliances, and discriminating between friends and enemies. Individuals
within such groups would have realized fitness benefits resulting from
inter-alliance sharing of patchily distributed resources, as well as enhanced cooperation
for in-group ventures, including hunting and warfare.
When symbolic behavior emerged in human evolution remains unknown. Some
346 Human Nature / Winter 2005
researchers argue for the emergence of symbolic culture in early archaic populations
(Bednarik 1995; Hayden 1993; Marshack 1990); others maintain that symbolic
thought appeared in early Homo sapiens sapiens prior to migration out of
Africa (Henshilwood et al. 2001; McBrearty and Brooks 2000; Watts 1999). Still
others argue for a "big bang" theory of symbolic culture first appearing approximately
50,000 years ago in western European populations (Mithen 1996).All, however,
associate the emergence of symbolic systems with ritual. Mithen notes that
"the very first art we possess appears to be intimately associated with religious
ideas by containing images of what are likely to be supernatural beings" (Mithen
1996:155). Watts (1999) also argues for a ritual origin of symbolic systems but
maintains that such systems emerged some 100,000 years earlier than posited by
Mithen. Watts (1999) argues that the ubiquitous presence of red ochre pigments at
numerous African MSA (Middle Stone Age) sites indicates ritual activity. Noting a
jump in ochre presence over time, he concludes that "the preoccupation with redness
clearly indicates that ochre was primarily used for signalling" (1999:128) and
argues that "the habitual nature of such behaviour from the MSA2b onwards strongly
suggests that the signalling was symbolic rather than solely indexical or iconic"
(1999:137). The MSA Blombos Cave excavations of Henshilwood and colleagues
lend further support for symbolic behaviors in African MSA populations. These
researchers recovered twenty-eight bone tools dated ca. 70,000 years ago exhibiting
"formal" techniques of bone tool manufacture, as well as ochre pencils and
objects bearing geometric designs. They note that "bone tools are . . . only one
element of a range of techniques used at BBC during the MSA to produce practical
and/or symbolic artefacts indicative of a complex technological society" (2001:668).
The occurrence of pigment processing at numerous MSA sites, as well as the notching
and incising of ochre, bone, and ostrich shell, are also interpreted by McBrearty
and Brooks as evidence of symbolic behavior. These researchers note that "Despite
the relatively small number of excavated MSA sites, the quantity and quality of
evidence for symbolic behavior . . . far exceeds that known for the European Middle
Paleolithic where the site sample is more than ten times greater" (2000:531).
The irregularly patterned and increasing use of red ochre pigment by African
MSA/LSA populations suggests that ritual was of variant but increasing importance
in human social groups throughout this period.The widespread occurrence of
red ochre pigments has been interpreted by Dunbar (1999) as evidence of "badging."
He argues that red ochre badging increased during the African MSA in order to
mark and identify group members when both the size and the number of groups
were increasing.Yet, Dunbar notes that "external badges encounter a common problem
. . . they are easy to fake" (1999:202).
If, however, red ochre badging is viewed within a broader context of ritual, as
Watts and others (Knight et al. 1995) have interpreted it to be, then both the costs
and the reliability of these badges increase, as well. Participation in ritual entails
time and energy costs which may deter free riders (Irons 2001; Sosis 2003b). More
importantly, participation in communal ritual provides the context for the creation
Religion as an Adaptive Complex 347
and internalization of communally shared motivators. The use of ritual to emotionally
charge badges and other selected artifacts would have added to the costs of
such badges, but would have significantly increased their reliability as signals of
motivational intent. This ritual transformation of signal badges to emotionally
charged and positively valenced symbols of social relationships may have served to
facilitate the creation of alliances under conditions of resource scarcity and conflict
(Hayden 1987). The red ochre, beadwork, bone incising and regional stone- and
bone-working styles evident in the archaeological record of the African MSA between
250,000 and 50,000 BP all indicate an increasing importance of ritual, an
intensification of costly signals, and the emergence of symbolic systems specific to
social groups. The emergence of dance, music, and even language may have their
roots in this intensification process. Why did these changes occur during the MSA?
There is evidence of increasing population, increasing use of a fission/fusion
social organization, and shared use of patchy resources within an environment of
overlapping group ranges throughout the MSA. McBrearty and Brooks (2000) report
that MSA sites in Africa are more numerous than those of the Acheulian and
are found in previously uninhabited zones, suggesting both the need and the ability
of MSA populations to exploit a wider range of habitats. Moreover, these sites
provide evidence of deliberate foresight and planning in cooperative hunting strategies
(Chase 1989), specialized tool use (McBrearty and Brooks 2000; Shea 1988),
and in the transport of both water and materials across long distances (Deacon
1989). The development of technologies such as ostrich eggshell containers that
permitted the transport of critical resources such as water opened up previously
uninhabitable areas (Watts 1999). The appearance of blades, as well as retouched
stone and bone points, indicates increasing technological sophistication, as well.
McBrearty and Brooks (2000) have interpreted the diversification of MSA toolkits
and the varying proportions of different artifact classes at different sites as evidence
of regional tradition differences, as well as differences in extractive activities.These
authors present compelling arguments for continuing intensification and
scheduling of resource use throughout the African MSA into the LSA. Evidence of
both selective, tactical hunting of large game and intensifying use of aquatic and
small-scale resources is cited, as well as proliferation and geographic extension of
trade networks. This intensification of extractive and hunting technologies, as well
as expansion into previously unexploited habitats and increasing territorial sizes
during the African MSA, have been viewed by McBrearty and Brooks (2000) as
evidence of both population growth and environmental degradation.
The picture that emerges from the accumulating archaeological record for the
African MSA is one of population growth, geographic dispersion, and technological
intensification and specialization. Tactical hunting strategies for large game
emerged. Simultaneously, the irregular distribution of critical items, such as water,
and the regional distribution of other prized resources, such as obsidian, introduced
increased inter-group interaction and competition for utilization of these patchy
resources. The ecological context of human groups in the African MSA suggests
348 Human Nature / Winter 2005
that the nature of hominid social groups underwent change during this period. Larger
group sizes punctuated by seasonal fission/fusion, and the creation and maintenance
of alliances in response to resource irregularity, are indicated. An increased
reliance on cooperative subsistence strategies, including large game hunting and
joint utilization of dispersed water sources, as well as increased competition between
groups for patchy resources, can be surmised from the archaeological record.
Red ochre pigments and decoratively incised stone and bonework suggest that these
changes were accompanied by increases in ritual and the emergence of an abstract
symbolic system.
It is likely that the incorporation of rhythmic drivers in human ritual preceded
these developments. The drumming and "proto-dances" of chimpanzees suggest
that precedents of music, chanting, and dance existed in common ancestral hominoids
(Goodall 1986). Such behaviors may have originated as communication signals.
The ability of these drivers to enhance positive affect would have rendered
rhythmic ritual a useful tool in the reintegration of fissioned groups and in the
creation of inter-group alliances. The use of rhythmic ritual to invest artifacts with
symbolic, emotionally valenced meaning would have provided dispersed groups
with a tangible and motivational symbol of the abstract social relationships codified
through the ritual process.With increasing resource competition, however, there
would also be increasing need to differentiate and cohesify groups in order to more
efficiently and effectively extract and defend resources. These conditions would
further promulgate in-group specialization and stratification. Under such conditions,
negatively valenced religious symbols would assume increasing importance
owing to both their greater motivational force and their signaling efficacy (Johnson
and Kruger 2004).
In contrast to the indexical signals of animal ritual, which elicit congruent motivational
states within an immediate time and space, the symbols of religious ritual
afforded early humans a means of engendering congruent motivational states across
space and time. And, although signals elicit neurophysiological responses that permit
social interaction in the here and now, symbols extend the horizon of those
responses to future activities, as well. Religious symbols, thus, provided tools for
creating cooperative coalitions across time. In doing so, they introduced a new level
of cognition and social organization in human evolution.
CONCLUSION
Religion is an important and unique human adaptation defined by four recurrent
traits: belief systems incorporating supernatural agents and counterintuitive concepts,
communal ritual, separation of the sacred and the profane, and adolescence
as a preferred developmental period for religious transmission. Although the specific
expression of each of these traits varies across cultures in socioecologically
patterned ways, the belief systems and communal rituals of all religions share common
structural elements that maximize retention, transmission, and affective en-
Religion as an Adaptive Complex 349
gagement. The roots of these structural elements can be found in nonhuman ritual
where they serve to neurophysiologically prime participants and ensure reliable
communication. Religion's incorporation of music, chanting, and dance intensifies
such priming and extends the impacts of ritual beyond dyadic interactions. Music
constitutes an abstract representation of ritual that can be recreated across time and
space to evoke the emotions elicited by ritual. Human use of ritual to conditionally
associate emotion and abstractions creates the sacred; it also lies at the heart of
symbolic thought.The brain plasticity of human adolescence offers a unique developmental
window for the creation of sacred symbols. Such symbols represent powerful
tools for motivating behaviors and promoting in-group cooperation.Although
religion evolved to solve an ecological problem by promoting group communication
and cooperation across space and time, the symbols it created laid the foundation
for a new adaptive niche in human evolution.
A number of empirically testable hypotheses emerge from this view of religion.
We have posited that the adaptive function of religion is to ensure cooperation when
individuals can achieve net benefits through collective action, and we have proposed
that ritual serves to engender such cooperation through the motivational
valencing of symbols. If so, religious ritual should be most pronounced within groups
of individuals who are not genetically related and are pursuing high-cost cooperative
endeavors, and least pronounced among kin groups pursuing individualistic
subsistence strategies. Significant associations between ritual intensity, positive and
negative symbolic valence, and age of initiation should also exist among these variables.
We expect to find the highest intensity of ritual in groups encompassing
unrelated individuals who must engage in intermittent, high-risk, cooperative endeavors,
such as external warfare or long-term sharing of scarce and patchy resources.
In contrast, the lowest levels of religious ritual should occur among
non-cooperating groups of kin. We would further expect to find permanent, highly
charged religious symbolic systems in non-kin groups engaged in high risk or widely
dispersed cooperative endeavors. Based on emotion theory, we expect ritual systems
to incorporate more negative affect in the emotional conditioning of symbols
under conditions of large group size and political inequality. We have argued that
adolescence constitutes an experience expectant period for the emotional valencing
of symbols. We, therefore, anticipate adolescent rites of passage to be most intense
and prolonged among unrelated adolescents in societies engaging in high-risk, cooperative
activities. In addition, there should be a positive association between the
duration/intensity of adolescent rites of passage and concomitant changes in both
brain response patterns to religious symbols and individual cooperative behaviors.
Music should be a particularly powerful elicitor of such responses.
Numerous research questions remain. If adolescence is an "experience expectant"
period for the emotional valencing of symbolic systems, is adolescent development
dependent on such valencing? In the absence of religious ritual, how is
such valencing achieved? Does ritual participation impact adolescent health and
behavior? Do adolescent rites of passage measurably alter neurotransmitter and
350 Human Nature / Winter 2005
endocrine levels?Are there gender and/or status differences in the neurophysiological
effects of ritual? Can we empirically demonstrate autonomic congruence in ritual
participants? If so, is such congruence significantly associated with perceived empathy
and increased cooperation? To what extent do the various components of
ritual impact emotional charging of symbolic stimuli? Can we define socioecological
parameters associated with the positive and negative emotional charging of religious
symbols? Does the developmental propensity to believe in socially omniscient
supernatural agents peak in adolescence? Are such agents a necessary
component of symbolically charged belief systems, or can such systems instead
achieve cooperation through the emotional charging of unfalsifiable non-agent
schema, such as "liberty" and "freedom"? Finally, if religion is an evolved adaptation
for cooperation, can humanity achieve such cooperation in its absence? This is
among the most salient questions facing the world today. The answer must begin
with a better understanding of religion as a specifically human adaptation.
Signals are necessarily bound to the moment; symbols, however, have existence
and meaning that extend beyond the immediate to link the past, present, and future.
They, thus, lay the foundation for creating and identifying groups, but also for motivating
cooperation among the individuals within these groups across both space
and time (Rappaport 1999). Far from being an evolutionary by-product, religion
represents a critical adaptive complex evolved in response to ecological challenges
faced by early human populations. Individual fitness benefits resulted both from
participation in ritual itself, and from the cooperative activities it enabled. The use
of music-based ritual to imbue group signals with emotional and motivational meaning
gave impetus to a new system of social communication and a new level of
human cognition.
The authors would like to thank Joseph Bulbulia, Marc Dichter, James Dow, Patrick McNamara, Eric
Smith, Robert Storey, and two anonymous reviewers for their helpful comments on an earlier draft of
this paper.
Candace Alcorta is currently a doctoral student in the Department of Anthropology at the University
of Connecticut. Her research interests include the behavioral ecology and evolution of religion, and
the interrelationship between cultural and neurophysiological systems. She is currently conducting
research on adolescent religious participation, stress, and health.
Richard Sosis is an associate professor in the Department of Anthropology at the University of Connecticut
and a senior lecturer in the Department of Sociology and Anthropology at The Hebrew University
of Jerusalem. His current research interests include the evolution of cooperation, utopian
societies, and the behavioral ecology of religion. He has conducted fieldwork on Ifaluk Atoll in the
Federated States of Micronesia and is currently pursuing various projects in Israel aimed at understanding
the benefits and costs associated with religious behavior.
REFERENCES CITED
Adolphs, R.
1999 Neural Systems for Recognizing Emotions in Humans. In The Design of Animal Communication,
M. Hauser and M. Konishi, eds. Pp. 187­212. Cambridge: MIT Press.
Religion as an Adaptive Complex 351
2002a Social Cognition and the Human Brain. In Foundations in Social Neuroscience, J. T.
Cacioppo, G. G. Berntson, R. Adolphs, et al., eds. Pp. 313­332. Cambridge: MIT Press.
2002b Trust in the Brain. Nature Neuroscience 5:192­193.
Adolphs, R., D. Tranel, and A. R. Damasio, eds.
1998 The Human Amygdala in Social Judgment. Nature 393:470­474.
Anderson, S. W., A. Bechara, H. Damasio, D. Tranel, and A. R. Damasio
2002 Impairment of Social and Moral Behavior Related to Early Damage in Human Prefrontal
Cortex. In Foundations of Social Neuroscience, J. T. Cacioppo, G. G. Berntson, R Adolphs, et al.,
eds. Pp. 333­343. Cambridge: MIT Press.
Atran, S.
2002 In Gods We Trust: The Evolutionary Landscape of Religion. Oxford: Oxford University
Press.
Atran, S., and A. Norenzayan
2004 Religion's Evolutionary Landscape: Counterintuition, Commitment, Compassion, Communion.
Behavioral and Brain Sciences 27:713­730.
Austin, J. H.
1998 Zen and the Brain. Cambridge: MIT Press.
Ball, G. F.
1999 The Neuroendocrine Basis of Seasonal Changes in Vocal Behavior among Songbirds. In
The Design of Animal Communication, M. D. Hauser and M. Konishi, eds. Pp. 213­254. Cambridge:
MIT Press.
Barrett, J. L.
2000 Exploring the Natural Foundation of Religion. Trends in Cognitive Science 4:29­34.
Bartlett, F.
1932 Remembering. Cambridge: Cambridge University Press.
Bear, D.
1979 Temporal Lobe Epilepsy: A Syndrome of Sensory-Limbic Hyperconnectionism. Cortex
15:357­384.
Bear, D., L. Schenk, and H. Benson
1981 Increased Autonomic Responses to Neutral and Emotional Stimuli in Patients with Temporal
Lobe Epilepsy. American Journal of Psychiatry 138:843­845.
Becker, J.
2001 Anthropological Perspectives on Music and Emotion. In Music and Emotion, P. Juslin and J.
Sloboda, eds. Pp. 135­160. Oxford: Oxford University Press.
Bednarik, R. G.
1995 Concept-Mediated Marking in the Lower Paleolithic. Current Anthropology 36:605­634.
Bering, J. M.
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, B. Ellis and D.
Bjorklund, eds. Pp. 411­437. New York: Guilford Press.
Bering, J. M., and D. F. Bjorklund
2004 The Natural Emergence of Reasoning about the Afterlife as a Developmental Regularity.
Developmental Psychology 40:217­233.
Bettleheim, B.
1962 Symbolic Wounds: Puberty Rites and the Envious Male. New York: Collier.
Bloch, M.
1989 Ritual, History, and Power. London: Athlone Press.
Bourguignon, E., ed.
1973 Religion, Altered States of Consciousness, and Social Change. Columbus: Ohio State University
Press.
1976 Possession. San Francisco: Chandler and Sharpe.
Boyer, P.
2001 Religion Explained:The Evolutionary Origins of ReligiousThought. NewYork: Basic Books.
Boyer, P., and C. Ramble
2001 Cognitive Templates for Religious Concepts: Cross-Cultural Evidence for Recall of
Counterintuitive Representations. Cognitive Science 25:535­564.
352 Human Nature / Winter 2005
Bradshaw, M.
2003 Religion as Compensation for Deprivation: The Interactive Influence of Religion and SES
on Psychological Distress. Paper presented at annual meeting of the Society for the Scientific
Study of Religion, October, Norfolk, Virginia.
Brown, J.
1975 Adolescent Initiation Rites: Recent Interpretations. In Studies inAdolescence, R. E. Grinder,
ed. Pp. 40­52. New York: MacMillan.
Brown, C. M.
2000 Exploring the Role of Religiosity in Hypertension Management amongAfricanAmericans.
Journal of Health Care for the Poor and Underserved 11:19­32.
Bulbulia, J.
2004a Religious Costs as Adaptations that Signal Altruistic Intention. Evolution and Cognition
10:19­42.
2004b The Cognitive and Evolutionary Psychology of Religion. Biology and Philosophy 19:655­
686.
Cacioppo, J. T., W. L. Gardner, and G. G. Berntson
2002 The Affect System Has Parallel and Integrative Processing Components: Form Follows
Function. In Foundations in Social Neuroscience, J. T. Caccioppo, G. G. Berntson, R. Adolphs, et
al., eds. Pp. 493­522. Cambridge: MIT Press.
Cardinal, R., J. Parkinson, J. Hall, and B. Everitt
2002 Emotion and Motivation: The Role of the Amygdala, Ventral Striatum and Prefrontal Cortex.
Neuroscience and Biobehavioral Reviews 26:321­352.
Carlson, G., Y. Wang, and B. E. Alger
2002 Endocannabinoids Facilitate the Induction of LTP in the Hippocampus. Nature Neuroscience
5:723­724.
Casey, G. J., J. N. Giedd, and K. N. Thomas
2000 Structural and Functional Brain Development and Its Relation to Cognitive Development.
Biological Psychology 54:241­257.
Chase, P. G.
1989 How Different Was Middle Paleolithic Subsistence? A Zooarchaeological Perspective on
the Middle to Upper Paleolithic Transition. In The Human Revolution: Behavioral and Biological
Perspectives on the Origins of Modern Humans, P. Mellars and C. P. Stringer, eds. Pp. 321­327.
Edinburgh: Edinburgh University Press.
Chaves, M., M. E. Konieczny, K. Beyerlein, and E. Barman
1999 The National Congregations Study: Background, Methods, and Selected Results. Journal
for the Scientific Study of Religion 38:458­476.
Cronk, L.
1994a Evolutionary Theories of Morality and the Manipulative Use of Signals. Zygon 29:32­58.
1994b The Use of Moralistic Statements in Social Manipulation: A Reply to Roy A. Rappaport.
Zygon 29:351­355.
Damasio, A. R.
1994 Descartes'Error: Emotion, Reason, and the Human Brain. New York: Avon Books.
1998 The Somatic Marker Hypothesis and the Possible Function of the Prefrontal Cortex. In The
Prefrontal Cortex, A. C. Roberts, T. W. Robbins, and J. Weiskrantz, eds. Pp. 36­50. New York:
Oxford University Press.
Davidson, J. R.
1976 The Physiology of Meditation and Mystical States of Consciousness. Perspectives in Biology
and Medicine (Spring):345­379.
Davidson, R.
1994 On Emotion, Mood, and Related Affective Constructs. In The Nature of Emotion, P. Ekman
and R. J. Davidson, eds. Pp. 51­55. New York: Oxford University Press.
Davidson, R., and W. Irwin
2002 The Functional Neuroanatomy of Emotion and Affective Style. In Foundations in Social
Neuroscience, J. T. Caccioppo, G. G. Berntson, R. Adolphs, et al., eds. Pp. 473­490. Cambridge:
MIT Press.
Religion as an Adaptive Complex 353
Deacon, H. J.
1989 Late Pleistocene Paleoecology andArchaeology in the Southern Cape. In The Human Revolution:
Behavioral and Biological Perspectives on the Origins of Modern Humans, P. Mellars and
C. P. Stringer, eds. Pp. 547­564. Edinburgh: Edinburgh University Press.
Deacon, T.
1997 The Symbolic Species. New York: W.W. Norton.
Dehaene, S., and J. P. Changeux
2000 Reward-Dependent Learning in Neuronal Networks for Planning and Decision-Making. In
Cognition, Emotion, andAutonomic Responses:The Integrative Role of the Prefrontal Cortex and
Limbic Structures, H. B. M. Uylings, C. G. van Eden, J. P. D. deBruin, et al., eds. Pp. 219­230.
Elsevier: New York.
Depue, R. A., M. Luciana, P. Arbisi, P. Collins, and A. Leon
2002 Dopamine and the Structure of Personality: Relation of Agonist-Induced Dopamine Activity
to Positive Emotionality. In Foundations in Social Neuroscience, J.T. Caccioppo, G. G. Berntson,
R. Adolphs, et al., eds. Pp. 1071­1092. Cambridge: MIT Press.
DiChiara, G.
1995 The Role of Dopamine in Drug Abuse Viewed from the Perspective of Its Role in Motivation.
Drug and Alcohol Dependence 38:95­137.
Dissinayake, E.
1995 Homo aestheticus: WhereArt Comes From andWhy. Seattle: University ofWashington Press.
Dolan, R.
2000 Emotional Processing in the Human Brain Revealed through Functional Neuroimaging. In
The New Cognitive Neurosciences, second ed., M. Gazzaniga, ed. Pp. 1115­1132. Cambridge:
MIT Press.
Douglas, M.
1966 Purity and Danger. Frederick A. Praeger: New York.
Dressler, W. W., and J. R. Bindon
2000 The Health Consequences of Cultural Consonance: Cultural Dimensions of Lifestyle, Social
Support and Arterial Blood Pressure in an African American Community. American Anthropologist
102:244­260.
Dugatkin, L. A.
1997 Cooperation among Animals: An Evolutionary Perspective. New York: Oxford University
Press.
Dunbar, R.
1999 Culture, Honesty and the Freerider Problem. In The Evolution of Culture, R. Dunbar, C.
Knight, and C. Power, eds. Pp. 194­213. New Brunswick: Rutgers University Press.
Durkheim, E.
1969 The Elementary Forms of the Religious Life. NewYork: Free Press. (Originally published in
1915)
Ekman, P.
2003 Darwin, Deception and Facial Expression. In Emotions Inside Out, P. Ekman, J. J. Campos,
R. J. Davidson, and F. B. M. de Waal, eds. Pp. 205­221. Annals of the New York Academy of
Sciences 1000.
Ekman, P., and R. Davidson
1993 Voluntary Smiling Changes Regional Brain Activity. Psychological Science 4:342­345.
Ekman, P., R. W. Levenson, and W. V. Friesen
1983 Autonomic Nervous System Activity Distinguishes among Emotions. Science 22:1208­
1210.
Eliade, M.
1958 Rites and Symbols of Initiation:The Mysteries of Birth and Rebirth. Dallas: Spring Publica-
tions.
1959 The Sacred and the Profane:The Nature of Religion. NewYork: Harcourt Brace Jovanovich.
Elkin, D.
1999 Religious Development in Adolescence. Journal of Adolescence 22:291­295.
354 Human Nature / Winter 2005
Gellhorn, E., and W. F. Kiely
1972 Mystical States of Consciousness: Neurophysiological and Clinical Aspects. Journal of
Nervous and Mental Disease 154:399­405.
Geschwind, N.
1979 Behavioral Changes in Temporal Lobe Epilepsy. Psychological Medicine 9:217­219.
Giedd, J. N., J. Blumenthal, N. O. Jeffries, F. X. Catellanos, H. Liu,A. Zijdenbos,T. Paus,A. C. Evans,
and J. L. Rapoport
1999 Brain Development during Childhood and Adolescence: A Longitudinal MRI Study. Nature
Neuroscience 2:861­863.
Glucklich, A.
2001 Sacred Pain. New York: Oxford University Press.
Goodall, J.
1986 The Chimpanzees of Gombe: Patterns of Behavior. Cambridge: Harvard University Press.
Greenough, W. T.
1986 What's Special about Development? Thoughts on the Bases of Experience Sensitive Synaptic
Plasticity. In Developmental Neuropsychobiology, W. T. Greenough and J. M. Juraska, eds. Pp.
387­408. New York: Academic Press.
Greenough, W. T., and J. Black
1991 Induction of Brain Structure by Experience: Substrate for Cognitive Development. In Minnesota
Symposia on Child Psychology,Vol. 24, M. R. Gunnar and C.A. Nelson, eds. Pp. 155­200.
Hillsdale, N. J.: Lawrence Erlbaum.
Groenewegen, H. J., and H. B. M. Uylings
2000 The Prefrontal Cortex and the Integration of Sensory, Limbic, and Autonomic Information.
In Cognition, Emotion, and Autonomic Responses: The Integrative Role of the Prefrontal Cortex
and Limbic Structures, H. B. M. Uylings, C. G. van Eden, J. P. D. de Bruin, et al., eds. Pp. 3­28.
New York: Elsevier.
Guthrie, S. E.
1993 Faces in the Clouds: A New Theory of Religion. New York: Oxford University Press.
Halpern, C. T., J. R. Udry, and B. Campbell
1994 Testosterone and Religiosity as Predictors of Sexual Attitudes and Activity among Adolescent
Males: A Biosocial Model. Journal of Biosocial Science 26:217­234.
Hayden, B.
1987 Alliances and Ritual Ecstasy: Human Responses to Resource Stress. Journal for the Scientific
Study of Religion 26:81­91.
1993 The Cultural Capacities of Neanderthals: A Review and Re-evaluation. Journal of Human
Evolution 24:113­146.
Henshilwood, C. S., F. d'Errico, C. W. Marean, R. G. Milo, and R. Yates
2001 An Early Bone Tool Industry from the Middle Stone Age at Blombos Cave, South Africa:
Implications for the Origins of Modern Human Behaviour, Symbolism, and Language. Journal of
Human Evolution 41:631­678.
Hirokawa, E. and H. Ohira
2003 The Effects of Music Listening After a Stressful Task on Immune Functions, Neuroendocrine
Responses, and Emotional States in College Students. Journal of Music Therapy 40:189­
211.
Hummer, R. A., R. G. Rogers, C. B. Narn, and C. G. Ellison
1999 Religious Involvement and U.S. Adult Mortality. Demography 36:273­285.
Irons, W.
1996a In Our Own Self-Image:The Evolution of Morality, Deception and Religion. Skeptic 4:50­
61.
1996b Morality as an Evolved Adaptation. In Investigating the Biological Foundations of Morality,
J. P. Hurd, ed. Pp. 1­34. Lewiston, Idaho: Edwin Mellon Press.
2001 Religion as a Hard-to-Fake Sign of Commitment. In Evolution and the Capacity for Commitment,
R. Nesse, ed. Pp. 292­309. New York: Russell Sage Foundation.
Ito, T. A., J. T. Larsen, N. K. Smith, and J. T. Cacioppo
2002 Negative Information Weighs More Heavily on the Brain:The Negativity Bias in Evaluative
Categorizations. In Foundations in Social Neuroscience, J.T. Cacioppo, G. G. Berntson, R.Adolphs,
et al., eds. Pp. 575­598. Cambridge: MIT Press.
Religion as an Adaptive Complex 355
Johnson, D. D. P., and O. Kruger
2004 The Good of Wrath: Supernatural Punishment and the Evolution of Cooperation. Political
Theology 5:159­176.
Johnstone, R. A.
2000 Game Theory and Communication. In Game Theory and Animal Behavior, L. A. Dugatkin
and H. K. Reeve, eds. Pp. 94­117. New York: Oxford University Press.
Kasamatsu, A., and T. Hirai
1966 An Electroencephalographic Study on the Zen Meditation. Folio Psychiatrica & Neurologica
Japonica 20:315­336.
Kelemen, D.
2004 Are Children "Intuitive Theists"? Reasoning about Purpose and Design in Nature. Psychological
Science 15:295­301.
Keshavan, M. S.,V.A. Diwadkar, M. DeBellis, E. Dick, R. Kotwal, D. R. Rosenberg, J.A. Sweeney, N.
Minshew, and J. W. Pettegrew
2002 Development of the Corpus Callosum in Childhood, Adolescence and Early Adulthood.
Life Science 70:1909­1922.
Khalfa, S., S. D. Bella, M. Roy, I. Peretz, and S. J. Lupien
2003 Effects of Relaxing Music on Salivary Cortisol Level after Psychological Stress. In The
Neurosciences and Music, G. Avanzini, C. Faienza, D. Minciacchi, L. Lopez, and M. Majno, eds.
Pp. 374­376. Annals of the New York Academy of Sciences 999.
Kirkpatrick, L. A.
1999 Toward an Evolutionary Psychology of Religion and Personality. Journal of Personality
67:921­951.
Knight, C., C. Power, and I. Watts
1995 The Human Symbolic Revolution: A Darwinian Account. CambridgeArchaeological Journal
5:75­114.
Kolb, B., and I. Q. Whishaw
1998 Brain Plasticity and Behavior. Annual Review of Psychology 49:43­64.
Kolb, B., M. Forgie, R. Gibb, G. Gorny, and S. Rontree
1998 Age, Experience and the Changing Brain. Neuroscience and Biobehavioral Reviews 22:143­
159.
Kolb, B., S. Pellis, and T. E. Robinson
2004 Plasticity and Functions of the Orbital Frontal Cortex. Brain and Cognition 55:104­115.
Krebs, J. R., and R. Dawkins
1984 Animal Signals: Mind-Reading and Manipulation. In Behavioural Ecology: An Evolutionary
Approach, second ed., J. R. Krebs and N. B. Davies, eds. Pp. 380­402. Oxford: Blackwell
Scientific.
Krumhansl, C. L.
1997 An Exploratory Study of Musical Emotions and Psychophysiology. Canadian Journal of
Experimental Psychology 51:336­353.
Kuhn, D.
2002 The Effects of Active and Passive Participation in Musical Activity on the Immune System
as Measured by Salivary Immunoglobulin A (SIgA). Journal of Music Therapy 39:30­39.
Kwon, Y. J., and A. E. Lawson
2000 Linking Brain Growth with the Development of Scientific Reasoning Ability and Conceptual
Change during Adolescence. Journal of Research in Science Teaching 37:44­62.
Laughlin, C. D., Jr., and J. McManus
1979 Mammalian Ritual. In The Spectrum of Ritual, E. G. d'Aquili, C. D. Laughlin, and J.
McManus, eds. Pp. 80­116. New York: Columbia University Press.
Leach, E. R.
1966 Ritualisation in Man in Relation to Conceptual and Social Development. In A Discussion on
Ritualisation of Behaviour in Animals and Man, J. Huxley, ed. Philosophical Transactions of the
Royal Society, B:403­408.
LeDoux, J. E.
1996 The Emotional Brain. New York: Simon and Schuster.
2002 Emotion: Clues from the Brain. In Foundations in Social Neuroscience, J. T. Cacioppo, G.
G. Berntson, R. Adolphs, et al., eds. Pp. 389­410. Cambridge: MIT Press.
356 Human Nature / Winter 2005
Levenson, R. W.
1994 Human Emotions:A FunctionalView. In The Nature of Emotion, P. Ekman and R. J. Davidson,
eds. Pp. 123­126. New York: Oxford University Press.
2003 Blood, Sweat and Fears: The Autonomic Architecture of Emotion. In Emotions Inside Out,
P. Ekman, J. J. Campos, R. J. Davidson, and F. B. M. de Waal, eds. Pp. 348­366. Annals of the
New York Academy of Sciences 1000.
Levi-Strauss, C.
1963 Structural Anthropology. New York: Basic Books.
Levin, J. S.
1994 Religion and Health: Is There an Association, Is It Valid, and Is It Causal? Social Science in
Medicine 38:1475­1482.
1996 How Religion Influences Morbidity and Health: Reflections on Natural History,
Salutogenesis, and Host Resistance. Social Science and Medicine 43:849­864.
Lewis, D. B., and D. M. Gower
1980 Biology of Communication. New York: John Wiley and Sons.
Lex, B. W.
1979 The Neurobiology of RitualTrance. In The Spectrum of Ritual, E. G. d'Aquili, C. D. Laughlin,
Jr., and J. McManus, eds. Pp. 117­151. New York: Columbia University Press.
Lorenz, K.
1965 Evolution and Modification of Behavior. Chicago: University of Chicago Press.
Lutkehaus, N. C., and P. B. Roscoe, eds.
1995 Gender Rituals: Female Initiation in Melanesia. New York: Routledge.
MacLean, C. R. K., K. G. Walton, S. R. Wenneberg, D. K. Levitsky, J. P. Mandarino, R. Wziri, S. L.
Hillis, and R. H. Schneider
1997 Effects of the Transcendental Meditation Program on Adaptive Mechanisms: Changes in
Hormone Levels and Responses to Stress after 4 Months of Practice. Psychoneuroendocrinology
22:277­295.
Malinowski, B.
1948 Magic, Science, Religion and Other Essays. Garden City, New Jersey: Doubleday.
Mandel, A.
1980 Toward a Psychobiology ofTranscendence: God in the Brain. In The Psychobiology of Consciousness,
J. Davidson and R. Davidson, eds. Pp. 379­464. New York: Plenum Press.
Marler, P.
1999 On Innateness: Are Sparrow Songs "Learned" or "Innate"? In The Design of Animal Communication,
M. D. Hauser and M. Konishi, eds. Pp. 293­318. Cambridge: MIT Press.
Marshack, A.
1990 Evolution of the Human Capacity: The Symbolic Evidence. Yearbook of Physical Anthropology
32:1­34.
Matthews, D. A., M. E. McCullough, D. B. Larson, H. G. Koenig, J. P. Swyers, and M. G. Milano
1998 Religious Commitment and Health Status. Archives of Family Medicine 7:118­124.
McBrearty, S., and A. Brooks
2000 The Revolution That Wasn't: A New Interpretation of the Origin of Modern Human Behavior.
Journal of Human Evolution 39:453­563.
McCauley, R. N.
2001 Ritual, Memory and Emotion: Comparing Two Cognitive Hypotheses. In Religion in Mind,
J. Andresen, ed. Pp. 115­140. Cambridge: Cambridge University Press.
McNamara, P.
2001 Religion and the Frontal Lobes. In Religion in Mind, J. Andresen, ed. Pp. 237­256. Cambridge:
Cambridge University Press.
2002 The Motivational Origins of Religious Practices. Zygon 37:143­160.
Meyerhoff, B.
1974 Peyote Hunt: The Sacred Journey of the Huichol Indians. Ithaca: Cornell University Press.
Mithen, S.
1996 The Prehistory of the Mind. London: Thames & Hudson.
1999 Symbolism and the Supernatural. In The Evolution of Culture, R. Dunbar, C. Knight, and C.
Power, eds. Pp. 147­172. New Brunswick: Rutgers University Press.
Religion as an Adaptive Complex 357
Morris, J. S., A. Ohman, and R. J. Dolan
1998 Conscious and Unconscious Emotional Learning in the HumanAmygdala. Nature 393:467­
470.
Murphy, P. E., J. W. Ciarrocchi, R. L. Piedmont, S. Cheston, and M. Peyrot
2000 The Relation of Religious Belief and Practices, Depression, and Hopelessness in Persons
with Clinical Depression. Journal of Consulting and Clinical Psychology 68:1102­1106.
Neher, A.
1962 A Physiological Explanation of Unusual Behavior in Ceremonies Involving Drums. Human
Biology 34:151­161.
Newberg, A. B., E. G. d'Aquili, and V. Rause
2001 Why God Won't Go Away. New York: Ballantine Books.
Oram, M. W., and B. J. Richmond
1999 I See a Face--A Happy Face. Nature Neuroscience 2:856­858.
Paige, K. E., and J. M. Paige
1981 The Politics of Reproductive Ritual. Los Angeles: University of California Press.
Pearson, J.
1990 Neurotransmission in Brain Regions Associated with Emotion and Autonomic Control. In
An Introduction to Neurotransmission in Health and Disease, P. Riederer, N. Koppe, and J. Pearson,
eds. Pp. 253­263. New York: Oxford University Press.
Peretz, I.
2001 Brain Specialization for Music: New Evidence from Congenital Amusia. In The Biological
Foundations of Music, R. Zatorre and I. Peretz, eds. Pp. 153­165. Annals of the NewYork Academy
of Sciences 930.
Pinker, S.
1997 How the Mind Works. New York: Norton.
Plant, T. M.
2002 Neurophysiology of Puberty. Journal of Adolescent Health 31:185­191.
Rappaport, R. A.
1994 On the Evolution of Morality and Religion: A Response to Lee Cronk. Zygon 29:331­349.
1999 Ritual and Religion in the Making of Humanity. London: Cambridge University Press.
Regan, C.
2001 Intoxicating Minds. New York: Columbia University Press.
Regnerus, M., C. Smith, and M. Fritsch
2003 Religion in the Lives of American Adolescents: A Review of the Literature. Research Report
of the National Study of Youth and Religion, No. 3. Chapel Hill: University of North Carolina.
Reichert, S. E.
2000 Game Theory and Animal Contests. In Game Theory and Animal Behavior, L. A. Dugatkin
and H. K. Reeve, eds. Pp. 64­93. New York: Oxford University Press.
Richerson, P., and R. Boyd
1998 The Evolution of Human Ultra-Sociality. In Indoctrinability, Ideology and Warfare: Evolutionary
Perspectives, I. Eibl-Eibisfeldt and F. Salter, eds. Pp. 71­95. NewYork: Berghahn Books.
Robbins, T. W.
2000 The Arousal to Cognition: The Integrative Position of the Prefrontal Cortex. In Cognition,
Emotion andAutonomic Responses:The Integrative Role of the Prefrontal Cortex, H. B. M. Uylings,
C. G. van Eden, J. P. C. de Bruin, M. G. P. Feenstra, and C. M.A. Pennartz, eds. Pp. 469­483. New
York: Elsevier.
Roes, F. L., and M. Raymond
2003 Belief in Moralizing Gods. Evolution and Human Behavior 24:126­135.
Rogers, L. J., and G. Kaplan
2000 Songs, Roars, and Rituals. Cambridge: Harvard University Press.
Rolls, E. T.
1998 The Orbitofrontal Cortex. In The Prefrontal Cortex, A. C. Roberts, T. W. Robbins, and L.
Weiskrantz, eds. Pp. 67­86. New York: Oxford University Press.
Rowe, C.
1999 Receiver Psychology and the Evolution of Multi-Component Signals. Animal Behaviour
58:921­931.
358 Human Nature / Winter 2005
Sapolsky, R.
1999 Hormonal Correlates of Personality and Social Contexts: From Non-human to Human Primates.
In Hormones, Health and Behavior, C. Panter-Brick and C. Worthman, eds. Pp. 18­46.
Cambridge: Cambridge University Press.
Saver, J. L., and J. Rabin
1997 The Neural Substrates of Religious Experience. Journal of Neuropsychiatry 9:498­510.
Scherer, K. R., and M. R. Zentner
2001 Emotional Effects of Music: Production Rules. In Music and Emotion, P. Juslin and J. Sloboda,
eds. Pp. 361­392. Oxford: Oxford University Press.
Schoenbaum, G., B. Setlow, M. P. Saddoris, and M. Gallagher
2003 Encoding Predicted Outcome andAcquired Value in Orbitofrontal Cortex during Cue Sampling
Depends upon Input from Basolateral Amygdala. Neuron 39:855­867.
Schultz, W., P. Dayan, and P. R. Montague
2002 A Neural Substrate of Prediction and Reward. In Foundations in Social Neuroscience, J. T.
Cacioppo, G. G. Berntson, R. Adolphs, et al., eds. Pp. 313­332. Cambridge: MIT Press.
Shea, J.
1988 Spear Points from the Middle Paleolithic of the Levant. Journal of FieldArchaeology 15:441­
450.
Sloboda, J., and P. Juslin
2001 Psychological Perspectives on Music and Emotion. In Music and Emotion, P. Juslin and J.
Sloboda, eds. Pp. 71­104. Oxford: Oxford University Press.
Smith, J. W.
1979 Ritual and the Ethology of Communicating. In The Spectrum of Ritual, E. G. d'Aquili, C. D.
Laughlin, Jr., and J. McManus, eds. Pp. 51­79. New York: Columbia University Press.
Sosis, R.
2000 Religion and Intragroup Cooperation: Preliminary Results of a Comparative Analysis of
Utopian Communities. Cross-Cultural Research 34:70­87.
2003a Review of "Darwin's Cathedral: Evolution, Religion and the Nature of Society" by David
Sloan Wilson. Evolution and Human Behavior 24:137­143.
2003b Why Aren't We All Hutterites? Costly Signaling Theory and Religious Behavior. Human
Nature 14:91­127.
Sosis, R., and C. S. Alcorta
2003 Signaling, Solidarity and the Sacred: The Evolution of Religious Behavior. Evolutionary
Anthropology 12:264­274.
2004 Is Religion Adaptive? Behavioral and Brain Sciences 27:749­750.
Sosis, R., and E. Bressler
2003 Cooperation and Commune Longevity: A Test of the Costly Signaling Theory of Religion.
Cross-Cultural Research 37:211­239.
Sosis, R., and B. Ruffle
2003 Religious Ritual and Cooperation: Testing for a Relationship on Israeli Religious and Secular
Kibbutzim. Current Anthropology 44:713­722.
2004 Ideology, Religion, and the Evolution of Cooperation: Field Experiments on Israeli Kibbutzim.
Research in Economic Anthropology 23:87­115.
Sosis, R., H. Kress, and J. Boster
n.d. Scars for War: Evaluating Alternative Signaling Explanations for Cross-Cultural Variance in
Ritual Costs. Ms. in the authors' possession.
Sowell, E. R., P. M. Thompson, C. J. Holmes, T. L. Jernigan, and A. W. Toga
1999 In Vivo Evidence for Post-Adolescent Brain Maturation in Frontal and Striatal Regions.
Nature Neuroscience 2:859­861.
Spear, L. P.
2000 The Adolescent Brain and Age-Related Behavioral Manifestations. Neuroscience and
Biobehavioral Reviews 24:417­463.
Swanson, G. E.
1960 The Birth of the Gods: The Origin of Primitive Beliefs. Ann Arbor: University of Michigan
Press.
Religion as an Adaptive Complex 359
Tinbergen, N.
1965 The Study of Instinct. Oxford: Clarendon Press.
Turner, V.
1967 The Forest of Symbols. New York: Cornell University Press.
1969 The Ritual Process. Chicago: Aldine.
Tylor, E. B.
1871 Primitive Culture. London: Murray.
van Gennep, A.
1960 The Rites of Passage. Chicago: University of Chicago Press. (Originally published in 1909)
Walker, E., and A. M. Bollini
2002 Pubertal Neurodevelopment and the Emergence of Psychotic Symptoms. Schizophrenia
Research 54:17­23.
Wallace, A. F. C.
1966 Religion: An Anthropological View. New York: Random House.
Walter, V. J., and W. G. Walter
1949 The Central Effects of Rhythmic Sensory Stimulation. Electroencephalography and Clinical
Neurophysiology 1:57­86.
Watanabe, J. M., and B. B. Smuts
1999 Explaining Religion without Explaining It Away: Trust, Truth, and the Evolution of Cooperation
in RoyA. Rappaport's "The ObviousAspects of Ritual." AmericanAnthropologist 101:98­
112.
Watts, I.
1999 The Origin of Symbolic Culture. In The Evolution of Culture, R. Dunbar, C. Knight, and C.
Power, eds. Pp. 113­146. New Brunswick: Rutgers University Press.
Wilson, D. S.
2002 Darwin's Cathedral. Chicago: Chicago University Press.
Wingfield, J. C., J. D. Jacobs, K. Soma, D. L. Maney, K. Hunt, D. Wisti-Peterson, S. Meddle, M.
Ramenofsky, and K. Sullivan
1999 Testosterone, Aggression, and Communication: Ecological Bases of Endocrine Phenomena.
In The Biology of Communication, M. Konishi and M. D. Hauser, eds. Pp. 255­284. Cambridge:
MIT Press.
Winkelman, M.
1986 Trance States: A Theoretical Model and Cross-Cultural Analysis. Ethos 14:174­203.
1992 Shamans, Priests and Witches: A Cross-Cultural Study of Magico-religious Practitioners.
Anthropological Research Papers No. 44. Tempe: Arizona State University.
2000 Shamanism: The Neural Ecology of Consciousness and Healing. Westport, Conn.: Bergin